Академический Документы
Профессиональный Документы
Культура Документы
Edited by
T.N. Lomova
G.E. Zaikov
LEIDEN • BOSTON
2008
This book is printed on acid-free paper.
All rights reserved. No part of this publication may be reproduced, translated, stored in
a retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV
provided that the appropriate fees are paid directly to The Copyright Clearance Center,
222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA.
Fees are subject to change.
Preface xi
Foreword 1
Chapter 1
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and
Applications 5
M.A. Grin and A.F. Mironov
Introduction 5
1 Synthetic bacteriochlorins 6
2 Natural bacteriochlorins and their chemical modifications 14
3 Amphiphilic and water-soluble derivatives of bacteriochlorins 30
4 Conclusion 41
References 41
Chapter 2
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins:
Synthesis and Modification 45
A.S. Semeykin, S.A. Syrbu and O.I. Koifman
Introduction 45
1 Synthesis of meso(5,10,15,20)-tetrasubstituted porphyrins 46
1.1 Synthesis of meso-tetra-β-octasubstituted porphyrins 55
1.2 Synthesis of nonsymmetric meso-tetrasubstituted porphyrins 58
2 meso(5,15)-Disubstituted porphyrins 66
3 Synthesis of meso(5)-Phenyl-b-octaalkylporphyrins 74
4 Reactions of introduction and modification of substituents in phenyl
rings of meso-phenyl-substituted porphyrins 77
4.1 Introduction of substituents into phenyl rings of meso-phenylporphyrins 77
4.2 Modification of functional groups in phenyl rings of
meso-phenylporphyrins 79
4.2.1 meso-Oxyphenylporphyrins and their modification 80
4.2.2 meso-Aminoporphyrins and their modification 82
References 84
vi CONTENTS
Chapter 3
The Mechanism of Catalytic Action of the Coordination Centres
of Catalase Synthetic Models 93
T.N. Lomova, M.E. Klyueva and M.V. Klyuev
Introduction 93
1 Substituted copper(II) porphyrins as catalysts of the hydrogen peroxide
disproportionation reaction 95
2 Kinetic regularities and mechanisms of peroxide decomposition reactions
in the presence of acido complexes of highly substituted manganese
porphyrins 96
2.1 Kinetics and reaction mechanism of oxidation of manganese(III)
porphyrins by hydrogen peroxide 98
2.2 Kinetics of peroxide disproportionation in the presence of
manganese(III) porphyrins 108
3 Conclusion 114
References 115
Chapter 4
Complexation of Porphyrins with Ions and Organic Molecules 117
N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili,
Y. Inoue and O.I. Koifman
Introduction 117
1 Complexation of porphyrins with ions 118
2 Complexation of porphyrins with organic molecules:
the thermodynamic aspect 125
3 Complexation of porphyrins with organic molecules:
the chirality aspect 144
3.1 Host–guest systems based on monomeric porphyrins 144
3.2 Host–guest systems based on dimeric and oligomeric porphyrins 151
4 Conclusion 162
References 163
Chapter 5
Chemical Activation of Porphyrins in Coordination Core Reactions 169
D.B. Berezin and B.D. Berezin
Introduction 170
1 Porphyrin ligands with localized and delocalized NH bonds 171
1.1 Factors causing the delocalization of NH bonds in porphyrin
molecules 172
1.1.1 Spatial structure and polarization of the porphyrin molecule 172
1.1.2 NH activation in the course of porphyrin–solution
component interaction and porphyrin–solid phase interaction 181
1.2 Interaction of organic solvents and porphyrins with delocalized-type
bonds 185
1.2.1 Acid-base interactions 185
1.2.2 Tautomeric processes 190
1.3 Quantitative assessment of the state of NH bonds in porphyrin
molecules 199
CONTENTS vii
1 Experimental 342
2 Results and discussion 343
3 Conclusion 347
4 Acknowledgement 347
References 347
Chapter 11
Transport of Water as a Structurally Sensitive Process
Characterizing the Morphology of Biodegradable Polymer Systems 349
A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and
G.E. Zaikov
Introduction 350
1 Experimental 350
2 Hydrophobization of poly-(3-hydroxybutyrate) 351
3 Hydrophilization of poly-(3-hydroxybutyrate) 354
References 359
Chapter 12
A Novel Technique for Measurement of Electrospun Nanofiber 361
M. Ziabari, V. Mottaghitalab and A.K. Haghi
Introduction 361
1 Methodology 363
1.1 Simulation of electrospun web 363
1.2 Fiber diameter measurement 363
1.2.1 Manual method 364
1.2.2 Distance transform 364
1.2.3 Direct tracking 365
1.3 Real webs treatment 366
2 Experimental 367
3 Results and discussion 367
4 Conclusion 373
References 374
Chapter 13
Image Analysis of Pore Size Distribution in Electrospun Nanofiber
Webs: New Trends and Developments 375
M. Ziabari, V. Mottaghitalab and A.K. Haghi
Introduction 375
1 Methodology 376
1.1 Sieving methods 378
1.2 Mercury porosimetry 378
1.3 Flow porosimetry (bubble point method) 379
1.4 Image analysis 379
1.4.1 Real webs 380
1.4.2 Simulated webs 381
2 Experimental 382
3 Results and discussion 383
4 Conclusion 388
References 389
x CONTENTS
Chapter 14
Electrospun Biodegradable and Biocompatible Natural Nanofibers:
A Detailed Review 391
A.K. Haghi and R.K. Haghi
1 Introduction 391
1.1 Electrospinning setup 392
2 Effect of systematic parameters on electrospun nanofibers 393
2.1 Solution properties 393
2.1.1 Viscosity 393
2.1.2 Solution concentration 393
2.1.3 Molecular weight 395
2.1.4 Surface tension 396
2.1.5 Number of entanglements 396
2.1.6 Solution conductivity 396
2.1.7 Effect of salt addition 397
2.1.8 Solvent 399
2.2 Processing condition 399
2.2.1 Applied voltage 399
2.2.2 Feed rate 402
2.2.3 Distance of needle tip to collector 403
3 Theory and modeling 405
4 Natural fibers 407
5 Electrospinning of silk fibers 408
5.1 Introduction 408
5.2 Crystal structure of silk (fibroin) at various stages of electrospinning 408
5.3 Spinning dope preparation for electrospinning 411
5.3.1 Degumming 411
5.3.2 Dissolving of fibroin 411
5.4 Electrospinning of silk fibroin 411
5.4.1 Effect of silk polymer concentration on fiber diameter 412
5.4.2 Effect of voltage and spinning distance on morphology and
diameter 414
5.5 Characterization 416
6 Electrospinning of cellulose and cellulose acetate 417
6.1 Electrospinning of CTA solution 418
7 Concluding remarks 419
References 420
Nomenclature 422
To learn anything without thinking is absolutely useless,
Thinking about something without analysing and
Studying the subject of thinking is dangerous
Confucius, 551–479 BC
Ancient China
Preface
Everything around us has been created by chemistry (glass, gas, medicines, food, metals,
polymers, etc.). There is no pathos of “chemical” enthusiasm in this definition. Chemistry
is the basis of everything, which life produces (microbes, plants, animals, human beings).
Any living body is a giant chemical reactor with millions of coordinated chemical reactions
proceeding in it. Molecular biology, molecular genetics and gene engineering, biotechnol-
ogy and intellect – all this is chemistry.
Nobelist Prof. N.N. Semenov and his pupils (Profs. M.N. Emanuel, V.N. Kondratyev,
V.I. Goldansky, A.L. Buchachenko, K.I. Zamoraev, Yu.B. Monakov) insisted on that. And
we fully agree with them.
There is a certain shift in the world of chemistry from studies of more or less simple
reactions to get, for example, sulfuric acid, ammonia, phenol and acetone to research into
complicated biochemical and biological processes. This is required for solving medical
problems, because medicine is our health and our good life. It is common knowledge that
it is better to be healthy and rich than to be poor and sick.
Chemistry is more than a science. It is a festivity. This volume presents the reviews of
chemists working in the field of biochemistry, biology and, in the final analysis, for medi-
cine and health.
We look forward to readers’ comments, will be grateful for them and will definitely
use them in our future research.
Prof. Gennady E. Zaikov
Moscow, Russia
Prof. Tatyana N. Lomova
Ivanovo, Russia
x INTRODUCTION
To propagate education is to extend prosperity.
I mean the general prosperity but not one’s private wealth.
With extension of prosperity, part of evil disappears.
Alfred Nobel, Sweden
Foreword
Research into the kinetics and thermodynamics of processes involving biologically active
substances and multistage mechanisms of their interactions in living systems is of primary
interest for modern science. Many problems arising in the course of the studies may be
solved by investigating the reactivity of BAS in model systems and synthetic analogs of nat-
urally occurring biomolecules with a complex and often unknown structure. The authors of
this book have been for many years engaged in studies of the reactivity of BAS in various
aggregate states and in relation to the molecular structures and supramolecular forms.
The book covers many aspects of the effects of relatively simple biomolecules as mod-
el enzymes, molecular receptors, photosensitizers, pharmacophores, and biopharmaceutical
agents. The quantitative characteristics of the transitions of cations, anions and small or-
ganic molecules, enzymic catalysis, and diffusion of molecules through biological mem-
branes are presented. The mechanisms of the processes are discussed. The biological
activity of the compounds studied is assessed.
Hydrated porphyrin forms and their unique properties are of great interest. Chapter 1
deals with naturally occurring bacteriochlorins, their properties, isolation and chemical
modifications. A separate part is concerned with synthetic bacteriochlorins; here, general
and particular methods of production of these compounds, their spectral and physicochem-
ical characteristics, are considered. Special consideration is given to the use of naturally
occurring and synthetic bacteriochlorins in the production of new-generation photosensi-
tizers for photomedicine.
Chapter 2 describes methods of synthesis of meso-mono-, di-, tri-, and tetraphenyl-
porphyrins along with schemes of addition and modification of substituents in them.
Approaches are shown to fine setting of the physicochemical properties of porphyrins
obtained.
Chapter 3 represents the results of research into the catalytic activity of manga-
nese(III)- and copper(II)-porphyrins alkyl- and phenyl-substituted in β- and meso-positions
in the reaction of decomposition of hydrogen peroxide in the DMFA–KOH–H 2O system.
The ion-molecular mechanism of the decomposition with kinetically significant stages of
2 Foreword
Introduction
In the recent decade, attention of scientists, working in the field of developing new photo-
sensitizers (PS) for photodynamic therapy (PDT) of cancer, has been focused on com-
pounds with intensive absorption in the range of 770 up to 850 nm. The use of PS with this
therapeutic window of absorption opens new possibilities for the diagnostics and treatment
of malignant neoplasms. The light with these wavelengths scatters weakly and, therefore,
can penetrate deeper into the tissue (Scheme 1) [1]. This is of special significance in pig-
mented tumors, for instance, melanoma. Besides, it is important that accessible and cheap
semiconductor lasers can be used for this range.
Such compounds include derivatives of synthetic and natural bacteriochlorins. It is
known that in porphyrin systems two peripheral double bonds in opposite pyrrole rings
(B and D) are cross-conjugated, and their presence is not required for aromaticity to be
preserved.
6 M.A. Grin and A.F. Mironov
nm
1 Synthetic Bacteriochlorins
There are two approaches to the reduction of porphyrins to chlorins and bacteriochlorins.
The first includes catalytic hydrogenation or treatment with metals in an alcohol medium
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 7
nm
nm
nm
Figure 1 Electronic spectra of porphyrin (1), chlorin (2) and bacteriochlorin (3).
8 M.A. Grin and A.F. Mironov
R R
NH N R R NH N R
R A
N HN N HN
R R
4
B
R R
R N N R N N R
A R
Zn Zn
N N N N
R R
5
A: p-MeC6H4SO2NHNH2, K2CO3/Py, t [N2H2]
B: Zn(OAc)2
(boiling with sodium in amyl alcohol). The other approach is related to the use of diimide
NH=NH, which is formed from p-tosylhydrazine, as a reducing agent. When porphyrins are
reduced as free bases, a mixture of respective chlorin and bacteriochlorin 4 is formed,
whereas the reduction products of the zinc complex of porphyrin are chlorin and isobacte-
riochlorin 5 (Scheme 2) [4].
This approach was realized by R. Bonnett et al. in the reduction of 5,10,15,20-
tetrakis(m-hydroxyphenyl)porphyrin (m-THPP) 1 to form respective chlorin (m-THPC) 2
and bacteriochlorin (m-THPBC) 3 [5].
Though the molar extinction coefficients and λmax of the long-wavelength absorption
band increase in the sequence porphyrin > chlorin > bacteriochlorin, the photophysical
properties of the reduced species, including the quantum yields of the triplet and singlet ox-
ygen, differ insignificantly [6]. The photodynamic efficiency in the given sequence increas-
es at each stage of reduction,. Thus, at a tumour photonecrosis depth of 5 mm the doses of
PS introduced decrease from 6.25 mmol/kg for porphyrin to 0.75 mmol/kg for chlorin and
0.39 mmol/kg for bacteriochlorin [5].
However, in in vivo experiments bacteriochlorin, possessing a high photodynamic ac-
tivity, proved to be much less stable as compared with the chlorin analogue (Foscan® ); its
considerable amount was oxidized in cells within 24 h.
An interesting approach to the synthesis of di- and tetrahydroporphyrin derivatives
was proposed by Callot et al. [7], who showed that, during the action of diazoacetic-acid
methyl ester on tetraphenylporphyrin (TPP) 6, carbene formed in the reaction attacks the
double bonds in the pyrrole rings B and D to form cycloaddition products chlorin 7 (λmax =
650 nm) and bacteriochlorin 8 (λmax = 720 nm) (Scheme 4).
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 9
OH
HO NH N
N HN OH
OH
OH
HO NH N 2
+ B C
N HN OH
OH
OH
1 HO NH N
N HN OH
OH
Ph CO2Me CO 2Me
H Ph
Ph H
H H
H H
NH N NH N
NH N
A
Ph Ph
Ph Ph
+ Ph Ph
N HN N HN
N HN
H
MeO2C
Ph H Ph
Ph H
6 7 8
A: N2CHCO2Me, CuI
NH N NH N N N
A B
Zn
N HN N HN N N
HO
HO 10
C A A
HO
OH HO
OH
N N
NH N NH N
Zn
N N
N HN N HN
HO
HO HO
HO 9 11
O
A C
OH O
OH
O
NH N NH N NH N
C
+
N HN N HN N HN
Et
O O 13
O 12
O
OH
OH
NH N NH N
NH N A B
N HN N HN
N HN H H
H
H H
H O
O
O
14 MeO2C 15 MeO2C 16
MeO2C
O
OH
OH
NH N NH N NH N
A B
H N HN H N HN H N HN
H H H
CO2Me CO2Me CO 2Me
CO2Me CO2Me CO2Me
MeO2C MeO2C MeO2C
17 18 19
CHO CHO OH
OH
NH N NH N
A
N HN N HN
PMe O O PMe O O
O O
20 B 21
O
CHO CHO
O
NH N NH N
N HN N HN
PMe O O PMe O O
O O
786 nm 777 nm
22 23
A: OsO4/Py, H2S; B: H2SO4
H H
B B
NH N O NH N O N
A D
B Me
P
N HN N HN O
D D
HO
PMe HO PMe PMe
PMe
26 27
O O O
A B B
NH N O NH N O NH N O
A
N HN N HN N HN
D C D C
HO OH
Me HO P Me
P Me Me OH
PMe P PMe P
28 29 30
PMe = CH2CH2CO2CH3
OH
OH
O
NH N NH N
NH N B
A
N HN N HN
N HN H H
H
H H
H
O O
O
CO2Me 31 CO2Me CO2Me 32
CHO OH
CHO
OH
Me O
N
NH N
NH N
A B
35
N HN
H
N HN H +
O
H
H
O
O N
CO2Me
CO2Me 33
34
OH O
OH N
NH N NH N
A B
+
H N HN N HN O
H
H H N
CO2Me CO2Me
CO2Me CO2Me
MeO2C MeO2C
36
CHO CHO
OH
O
OH N
NH N NH N
A B
H N HN
+
H N HN O
H H
CO2Me CO2Me N
CO2Me CO2Me
MeO2C MeO2C
37
Scheme 9 Effect of electron-acceptor substituents in chlorins on the progress of the pinacoline re-
grouping.
14 M.A. Grin and A.F. Mironov
CO2Me CO2Me
CO2Me
MeO 2C
NH N NH N
A
N HN N HN
CO2Me
MeO2C
38 CO2Me 39 CO2Me
MeO2C
CO2Me
NH N A Me NH N
N HN N HN
CO2Me
CO2Me
CO2Me CO2Me
40 41
A: dimethyl ester of acetylenedicarboxylic acid (DMAD), diazabicycloundecene (DBU)
CN
NC CN CN
OH
OH
NH N A NH N NH N
B
H N HN H N HN H N HN
H H H
O O O O O O O O O
MeO2C 42 MeO2C MeO2C
43 44
C
H+
MeO2C H
CO2Me CO2Me CO2Me
H MeO2C MeO2C
H
H
NH N B: NH N
NH N
E D
H N HN N HN H N HN
H
H H H
O O O O O O O O O
MeO2C MeO2C MeO2C
47 45 46
A: ɨ-dichlorobenzene, boiling
B: tetracyanoethylene (TCE)
ɋ: DMAD
D: Et3N
E: DBU
NH N NH N NH N H
A B
H N HN 49
48
H
CO2Me CO2Me C
MeO2C
CO2Me CO2Me
H CO2Me H
MeO2C MeO2C
MeO2C
E NH N H D NH N
NH N H
51 50
A: ɨ-dichlorobenzene, boiling
B: TCE
ɋ: DMAD
D: Et3N
E: DBU
N N
CHO A N N
Ni Ni
N N N
N
CHO
53
52
A: HCl
Pigments of the first group have intensive absorption bands in the near IR region. Of
special interest are bacteriochlorophylls a and b, whose maxima are not only shifted to the
red region, but also have the highest extinction coefficients. Of the two, Bchl a is usually
taken as the initial material to develop novel photosensitizers for photodynamic therapy
(PDT) of cancer and other possible photomedical applications.
Bacteriochlorophyll a is a porphin derivative, which contains the pentanone ring (exo-
cycle) condensed with the tetrahydroporphyrin macrocycle, various peripheral substituents
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 17
(CH3)2C=CH(CH2)2C(CH3)=CH(CH2)2C(CH3)=CH(CH2)2C(CH3)=CHCH2OH (a);
(ɋH3)2ɋH(ɋH2)3ɋ(ɋH3)=CH(CH2)2CH(CH3)(CH2)3-C(CH3)=CHCH2OH (b);
(CH3)2C=CH(CH2)2-C(CH3)=CH(CH2)2C(CH3)=CHCH2OH (c)
R1 H HO H R1
R2 R2
N N R3 N N
Mg R5 Mg
N N N N
R3
H H
H
H
H
O O
CO2Me
R4O2C R4O2C
and the central atom of magnesium. To enumerate hydrogen and nitrogen atoms forming
the molecule of Bchl a and its derivatives, in this review we use the IUPAC nomenclature.
Derivatives of natural Bchl a can be divided into two groups [20]. The former includes
compounds, which contain the exocyclic fragment:
O
R1 H
bacteriochlorophyll ɚ: Ɇ=Mg, R1=Me,
H R2=COOMe, R3= phytyl;
N N bacteriopheophetin a: Ɇ=2H, R1=Me,
R2=COOMe, R3= phytyl;
Mg
bacteriopheophorbide a: Ɇ=2H, R1=Me,
N N R2=COOMe, R3=H;
H
bacteriopyropheophorbide a: Ɇ=2H, R1=Me,
R2=R3=H;
phytyl – a residue of the alcohol phytol
H
H
O HO
R2
R3O2C
O
R1
H
H bacteriochlorin ɟ6: R1=Me, R2=R4=COOH,
R3= CH2COOH
NH N
bacteriochlorin p: R1=Me, R2=R3=R4=COOH
H HN
trimethyl ester of bacteriochlorin p:
R1=Me, R2=R3=R4=COOMe
H
R4 R3 R2
Sources for production of Bchl a are biomass of the purple bacteria Rh. sphaeroides,
Rh. roseapersiana and Rh. capsulata. At our laboratory, Bchl a is isolated from the biomass
of Rh. capsulata [21], which contains no other bacteriochlorophylls; this greatly facilitates
the isolation and purification of the main pigment [22, 23]. Based on Bchl a, various groups
made research, the aim of which was to obtain stable individual bacteriochlorins with spec-
tral characteristics not inferior to initial bacteriochlorophyll and in some cases even exceed-
ing them. New photosensitizers should be less hydrophobic as compared with initial
bacteriochlorophyll and be sufficiently well soluble in polar solvents. It is also desirable for
these compounds to have functional groups, which enable adding other bioactive molecules
to them to increase the tropicity to cancer cells and targeted intracellular transport.
The first step on the way to solving these problems was to include an additional anhy-
dride exocycle into the main macrocycle, which led to the increase of stability of the pig-
ment [24]. For this, bacteriochlorophyll a 56 was extracted from biomass of the purple
bacteria Rh. capsulata and then, without isolation and additional purification, was oxidized
in an alkaline medium by oxygen of the air. The subsequent treatment with hydrochloric
acid led to the formation of the anhydride ring and production of bacteriopurpurine 57
(Scheme 15).
O O
C C
H H
H H
N NH N
Biomass of A
N
Mg B
Rhodobacter
capsulatus
H N N H N HN
H H
HO2C O O
H39C20O2C H O O
H3CO2C
56 57
Transformation of the pentanone ring into the anhydride cycle, which includes oxida-
tion of carbon atom C-132 in the exocycle by oxygen of the air in alcoholic solutions of
chlorophyll a, was observed well back at the beginning of the 20th century by the founder
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 19
O
O O
A B
NH NH N
N N A N
Mg
N HN
N N N HN H
H D C H
H HR
H B
CO2Me O
CO2Me O CO2Me O
MeO2C
PhytylO2C PhytylO2C
58 61.R= OH
56
62.R= OH
O O
O
NH N NH N NH N
D C
N HN N HN N HN
H H
Me
H H
HO O O CO2Me O CO2Me O
HOOC
HOOC MeO2C MeO2C
63 E 59 60
O
O O
NH N
NH N NH N
N HN
N HN N HN
CO2Me
O CO2Me CO2Me
O O O CO2Me
ROOC
ROOC MeO2C 64 65
57a. R=H F
57b. R=Me
57c. R=Pr
O
modification of C H
A B
the acetyl group
NH N H
Bacteriochlorophyll ɚ D C
H N HN
E
H
HO2C O O
etherification O
modification of
59, a mixture of oxidation products. Analysis of the latter showed it to consist of chlorin 60
and a mixture of diastereomers of 132-(R/S)-hydroxyderivatives 61 and 62 (epimers). The
presence of the hydroxyl group in the pentanone ring was reliably proven by the spectra of
1
H NMR; in this case, the ratio between R and S isomers was 1:4. These detailed studies
have shown that the S epimer, in which the carbomethoxy group in position 132 and the
propionic acid residue in position 17 of the macrocycle are directed to different sides rela-
tive to the plane of the macrocycle, is thermodynamically more stable under acidic condi-
tions [30, 31].
However, work with derivatives of Bchl a in acidic media is strongly complicated due
to the rapid oxidation of the pigment. The authors also observed allomerization in the pres-
ence of alkali. Thus, bacteriopheophorbide 59 in an air-bubbled solution of KOH–propanol
is allomerized to form “unstable bacteriochlorin” 63. The mechanism of allomerization un-
der alkaline conditions is unknown, but an indirect proof of the formation of intermediate
63 can be production of bacteriochlorin with the glyoxalic acid residue in position 15 of the
macrocycle 64 at the action of diazomethane on bacteriopheophorbide 59. “Unstable bac-
teriochlorin” 63 is converted into BP in the evaporation of the solvent; both free acid 57a
and ester 57c can be formed. As BP 57a possesses a low solubility in organic solvents, it is
expedient to convert it into methyl ester 57b, which facilitates chromatographic purifica-
tion. To increase the yield of BP, our laboratory developed a method of its production with-
out isolating bacteriopheophorbide as an intermediate. In this case, allomerization of Bchl a
in an air-bubbled KOH–isopropanol solution takes longer (1.5–2 h), apparently, due to the
presence of carotenoids, which are radical traps [32].
For bacteriopurpurine, a number of chemical transformations are possible (Scheme
17). They include the modification of the acetyl group with its reduction to an α-hydroxy-
ethyl group and its conversion into a vinyl group; conversion of the anhydride cycle into
the imide cycle; etherification of the propionic-acid residue by alcohols and serine.
Bacteriopurpurine is stable only in neutral and acidic media; in the presence of bases,
there occurs a rapid opening of the anhydride cycle, which leads to the recovery of the main
spectral band in the electronic spectrum.
Stability was increased and the spectral characteristics were improved by producing
cyclic imides of chlorins and bacteriochlorins by substitution of oxygen in the exocycle for
the nitrogen atom. The reaction was adjusted on methyl ester of purpurine 18 66 (Scheme
18) [33, 34]. The latter reacted to hexylamine at room temperature, giving a mixture of
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 21
NH N NH N
A
H N HN H N HN
H H
R1 R2
O O O
CO2Me CO2Me 67
66 a. R1 = CO2H, R2 = CONH(CH2)5CH3
b. R1 = CONH(CH2)5CH3, R2 = CO2H
NH N NH N
C
H N HN H N HN
R
H H
O N O O
X Y
CO2Me CO2Me
C6H13 68
69 a. X = O, Y = N(CH2)5CH3
b. X = N(CH2)5CH3, Y = O
isomeric amides in positions 131 67a and 151 67b at a ratio of 6:1 with a yield of 95%. In
order to obtain cyclic isoimides, the free carboxyl group of amides was activated using two
techniques.
In the first technique, a mixture of amides was treated with DCC to form two isoimides
(λmax = 690 and 696 nm) at a ratio of 6:1 with the total yield of 96%. In one of them, the
nitrogen of hexylamine is in position 131 68a; in the other, in position 151 68b. The mixture
of isoimides at the action of diazabicycloundecene (DBU) in toluene in an alkaline medium
at 60° is converted into the required cycloimide 69 with a low yield. The second approach
includes etherification of intermediate amides by diazomethane, followed by the treatment
of formed esters by a KOH methanol solution. In this case, cycloimide 69 is formed with
the yield of more than 80%. The authors note that the substitution of DBU by a stronger
KOH or NaOH base leads to a further increase of the yield.
Optimized conditions were used for the synthesis of bacteriochlorin derivatives
(Scheme 19) [34]. The mixture of amides 70a and 70b, obtained as the result of opening
the anhydride cycle of bacteriopurpurine by hexylamine, yields unstable carbodiimide
22 M.A. Grin and A.F. Mironov
O O
H H
H H
NH N NH N
A
H N HN H N HN
H H
R1 R2
O O O
CO2C3H7 CO2C3H7 70
1 2
a. R = CO2CH3, R = CONH(CH2)5CH3
b. R2 = CONH(CH2)5CH3, R2 = CO2CH3
O O
H H
H H
NH N NH N
C
H N HN H N HN
H H
O N O X Y
O
R CO2C3H7
CO2C3H7 71
72. R = C6H13 a. X = O, Y = N(CH2)5CH3
b. X = N(CH2)5CH3, Y = O
derivatives, which are rapidly converted to the more stable cyclic isoimides 71a and 71b.
The latter were separated chromatographically into individual isomers (λmax = 804 and 796
nm) at a ratio of 6:1. The base-catalyzed intramolecular cyclization of isoimides led, with
a yield of 45%, to cycloimide 72 (λmax = 822 nm). Similar results were obtained with
amides, whose carboxyl groups were methylated prior to the treatment by the base.
In an alkaline medium, along with the synthesis of the required cycloimide 72, there
occurred the oxidation of 12-CH3 group to form minor products: 12-formyl (4–6%) 75 and
12-hydroxymethyl (2–4%) 76 derivatives (Scheme 20). The authors believe that this un-
usual oxidation is the consequence of enolization leading to tautomers 73 and 74, which are
apt to oxidation in air. In turn, keto-enolic tautomerism in cycloimide is possible owing to
the strong electron-acceptor effect of the imide exocycle conjugated with the main tetra-
pyrrole system [35].
An unusual bacteriochlorin 78, having a Qy band at 849 nm, was obtained from 12-
hydroxymethylcycloimide 76 under acidic conditions (Scheme 21). The 2D ROESY spec-
tra showed the interaction of 10-H meso-proton with the adjacent 12-CH2 group and the ab-
sence of 8-H proton. The authors assume cycloimide 78 to have the structure depleted of
8-H proton to shift the double bonds inside the macrocycle. As the mechanism, they propose
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 23
N HN N HN H N HN H
A
H H
B
N O N O N
O O O O
Hexyl PrO2 C Hexyl Hexyl
PrO2C PrO2C
72 73 74
N HN N HN
CHO CH2OH
and
O N O O N O
Hexyl PrO2C Hexyl
PrO2C
75 76
A: KOH; B: HCl
O O O
NH N NH N NH N
A
N HN N HN N HN H
CH2OH CH2
H
O N O O N O N
O O
PrO2C Hexyl Hexyl Hexyl
PrO 2C PrO2C
76 77 78
+
A: H
elimination of the hydroxyl group under acidic conditions to form a carbocation of the ben-
zyl type 77, which is stabilized due to the abstraction of the proton at C-8 [29].
In contrast with natural bacteriochlorophyll a, the isoimide and imide analogues are
more stable, and their spectral characteristics compare favourably to respective precursors.
The photodynamic activity of the derivatives of chlorin p6 and bacteriochlorin p can
be further increased by introducing trifluoromethyl groups into cycloimide molecules
(Scheme 22) [36].
Fluorine-containing substituents enhance the solubility of compounds in lipids, which
increases the transport rate of such molecules through lipid membranes. The authors note
that the activity of PS in vivo is affected not only by the nature of the substituents but also
by how they are arranged in the macrocycle. Thus, cycloimide of chlorin p6 79b, containing
a bis-(trifluoromethyl)benzyl grouping at the nitrogen atom of the macrocycle, evokes a
more efficient inhibition of tumour growth as compared with the isomer containing this
24 M.A. Grin and A.F. Mironov
O Butyl
NH N NH N NH N
A B
H N HN H N HN H N HN
H H H
O O O O N O R O N O
R
CO2Me CO2Me CO2Me
R=CH3 R 79 R
C R=CF3
a. R=CH3
b. R=CF3
OR
NH N NH N
D a. R = H2C
H N HN H N HN
H H
O N O O N O CF3
CO2Me Butyl CO2Me Butyl
b. R = H2 C
80
CF3
A: 3,5-dimethylbenzylamine or 3,5-bis(trifluoromethyl)benzylamine;
B: HBr/CH3COOH, C4H7OH; C: C4H9NH2; D: HBr/CH3COOH, 3,5-dimethylbenzyl alcohol or
3,5-bis(trifluoromethyl)benzyl alcohol
substituent in the upper part of the chlorin macrocycle 80b (respectively, 100 and 66% of
tumour regression in 90 days).
Studies of the relation between the structure and activity were continued on cycloimi-
des of bacteriochlorin p. For this, bacteriopurpurine methyl ester was treated with 3,5-bis-
(trifluoromethyl)benzylamine. Along with required cycloimide, a Schiff base was formed;
the base proved to be labile in in vivo experiments. However, during the reduction of the
C=N bond, stable cycloimide 81 with the Qy band in the region of 796 nm was formed [37].
CF3
NH CF3
H
NH N
H N HN
H
O N O CF3
CO2Me
81 CF3
81
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 25
O NOH NOH
H H H
H H H
NH N NH N NH N
A A
H N HN H N HN H N HN
H H H
HO2C O O HO2C O O O HO2C O N O
O
57 82 OH 83
B C or D
2
O NOH NOR
H H H
H H H
NH N NH N NH N
A
H N HN H N HN H N HN
H H H
NaO2C COONa NaO2C COONa
COONa R1O 2C O N O
COONa
OR3
84 85 86
a: R = H; R2 = R3 = COCH3
1
b: R1 = R3 = CH3; R2 = H
Our laboratory was the first to perform the synthesis of cycloimides of chlorins and
bacteriochlorins by acting with highly nucleophilic agents hydroxylamine and hydrazine
hydrate on bacteriopurpurine.
Initially, for modification of bacteriopurpurine we used our earlier proposed method
of converting purpurine 18 to N-hydroxycycloimide [38, 39]. Herewith, the presence of the
acetyl group in bacteriopurpurine instead of the vinyl substituent in purpurine 18 leads to
the emergence of an additional reaction centre. Conducting the reaction of bacteriopurpu-
rine with hydroxylamine, it was shown (Scheme 23) that oxime 82 was initially formed
[40]. To prove this, we performed chemical conversions, which included the opening of the
anhydride cycle in oxime 82 and the treatment of bacteriochlorin p 84 by hydroxylamine
in pyridine. The identity of the products obtained indicates that in the treatment of BP by
hydroxylamine the first to enter into the reaction is the acetyl group.
Subsequent studies of the interrelation between bacteriopurpurine and hydroxylamine
have shown that, when excess reagent is used and time is increased up to 10 h, the second
molecule of hydroxylamine enters into the reaction with the anhydride cycle to yield oxime
of N-hydroxycycloimide of bacteriochlorin p 83. The latter compound can be obtained both
from bacteriopurpurine 57 and from intermediate oxime 82. The course of the reaction was
followed chromatographically and spectrally by the change of position of the long-wave Qy
band. In the first three hours, the maximum at 818 nm shifted to 792 nm, which corresponds
to the formation of oxime 82, after which it gradually returned to the long-wave region (812
nm). We believe that at this stage oxime interacts with the second molecule of hydroxy-
lamine to form compound 83. Thus, a derivative of hydroxamic acid was first obtained in
26 M.A. Grin and A.F. Mironov
OH
N N
HO H H
NH N H NH N H
N HN N HN
H H
H H
O O O N O
N
CO2Me CO2Me
OMe OMe
O O OH
H H H
H H H
NH N NH N NH N
A B
H N HN H N HN H N HN
H H H
CO2H CO2H CO2H CO2H
O O O
88
57 CO2H 87 CO2H
CO2H E C or D
H H OH
H H
H
H
NH N NH N
F NH N
H N HN H N HN
H N HN
H H
H
O N O O O O
O O O
CO2R
CO2R OR1
CO2H 89
91 90
a. R=H, R1=H
b. R=CH3, R1=H a. R=H
G b. R=CH3
H c. R=C2H5
c. R1=R2=CH3
d. R=CH3, R=Ts
A: NaOH, CH3OH; B: NaBH4; C: HCl, dioxan; D: TsCl, C5H5N; E: TsOH, CHCl3; F: NH2OH HCl, Py;
G: CH2N2, Et2O; H: TsCl, Py
the bacteriochlorophyll a series. The derivative, having a mobile hydrogen atom in its com-
position, readily enters into the acylation and alkylation reactions [41]. Thus, the treatment
with acetic anhydride leads to the formation of diacetate 86a, which proved rather labile
and in storage was decomposed into monoacetate and the initial compound 83.
The treatment of oxime of hydroxamic acid 83 by diazomethane yields N-methoxy-
cycloimide 86b with a good yield. This compound proved much more stable than acetate
and showed a good photodynamic activity in in vitro and in vivo experiments. Two inter-
esting facts were discovered in the course of the studies. The acidity of the hydrogen atom
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 27
in the oxime function is decreased so much that the function is not involved in the reaction
with diazomethane. Besides, comprehensive analysis (TLC, mass spectra) of N-methoxy
derivative 86b showed the substance to be a mixture of two isomers with very close values
of Rf. The isolated isomers were characterized by the 1H NMR spectra, including 1D NOE
spectroscopy; based on the data obtained, it was concluded that oxime 86b existed in the
form of two stereoisomers (sin- and anti-) (Scheme 24). Such isomers are formed in the in-
teraction of ketones with hydroxylamine [42].
As the acetyl group in BP strongly complicates the progress of the reaction with hy-
droxylamine, it was decided to convert it into the vinyl group.
The vinyl group in compound 90 was obtained from the α-hydroxyethyl group in al-
cohol 88 under the action of p-toluenesulfo acid (Scheme 25) [43]. The acetyl group was
reduced not on bacteriopurpurine 57, but on bacteriochlorin 87, as the treatment of bacte-
riopurpurine with sodium boron hydride had been earlier shown to contribute to the con-
version of the anhydride exocycle into the δ-lactonic one [44].
Triacid 87 obtained from BP was reduced by sodium boron hydride to respective al-
cohol 88, which has an absorption maximum at 740 nm. The treatment of the latter by hy-
drochloric acid in dioxane led only to the closure of the anhydride cycle, but not to the
dehydration of the α-hydroxyethyl group. Compound 89 was also observed to be formed
at the action of tosylchloride on alcohol 88; if the reaction was continued for more than 2 h,
the predominant product formed was O-tosylate.
A more efficient method of producing the vinyl group was developed based on the
treatment of alcohol 88 by p-toluenesulfo acid. Herewith, it was found that, along with the
formation of the vinyl group, there occurs the closing of the anhydride exocycle in com-
pound 90, which leads to the recovery of the main spectral band Qy to the region of 783 nm.
Interestingly, if the reaction was performed in chloroform, the product was obtained as free
acid 90a, whereas addition of methyl or ethyl alcohols directly to the reaction medium led
to the rapid etherification of propionic acid residue to form respective methyl 90b or ethyl
90c esters. This activity of the carboxyl group is, evidently, due to the formation of the
mixed anhydride of the residue of propionic acid and p-toluenesulfo acid.
3-Vinyl-3-deacetylbacteriopurpurine 90a is a structural analogue of purpurine 18 and,
for this reason, it was of interest to perform the above-described reaction with hydroxy-
lamine to assess the effect due to the conversion of the acetyl group into the vinyl group.
The treatment of compounds 90a and 90b by hydroxylamine produced respective
N-hydroxycycloimides 91a and 91b. Although the absolute values of the wavelengths of
the Q band in the spectra of N-hydroxycycloimides 86b and 91c are very close to 812 and
809 nm, the relative change of the long-wave absorption in the reaction products strongly
differs as compared with the initial substances. In the case of bacteriopurpurine, the forma-
tion of oxime by the acetyl group leads to a virtually complete levelling down of the effect
from the introduction of the imide exocycle. As the result, the terminal cycloimide is only
6 nm worse in the long-wave absorption than the initial pigment. In the case of the vinyl
derivative of bacteriopurpurine, transformation of the anhydride cycle to the imide cycle
gives a significant increment of the wavelength of the Q band up to 27 nm.
Recently, Sasaki and Tamiaki [45] synthesized a series of derivatives of bacteriopy-
ropheophorbide 92, which contain various substituents in pyrrole A.
The vinyl group was obtained by a different method as compared with that described
above. The high yield of compound 94 was achieved at the expense of the outgoing mesyl
group. Further on, the vinyl group was oxidized to the formyl group 95 using OsO4 and the
subsequent splitting of diol NaIO4 (Scheme 26). As the reactivity of the formyl group in
28 M.A. Grin and A.F. Mironov
NH N
Rhodobacter A B C
sphaeroides BChl - a
(purple bacterium) N HN
O
CO2Me
92
OH H 0
Ph
NH N D NH N E NH N
N HN N HN N HN
O O O
CO2Me CO2Me CO2Me
93 94 95
the chlorin series has been well studied, the authors carried out a number of similar trans-
formations in the bacteriochlorin series, which ncluded the oxidation and reduction of the
formyl group to the carboxyl and hydroxymethyl groups to form compounds 96 and 97, as
well as the Knoevenagel reaction with dicyanomalonic ester, which led to compound 98
(Scheme 27).
Studies of N-hydroxycycloimides of bacteriochlorin have shown that the hydroxyl
group can be successfully used to produce alkyl-substituted derivatives. At the same time,
introduction of acyl radicals by this way is less promising due to the lability of such deriv-
atives.
In this connection, we developed a method of producing N-aminocycloimides of bac-
teriochlorin p [46]. As is known, hydrazine and its derivatives possess a high nucleophilic-
ity with respect to sp2-carbon atoms [47]. For this reason, anhydrides of acids are
convenient reagents for acylation of hydrazine.
In the case of BP 57, we have shown (Scheme 28) that its treatment by hydrazine hy-
drate in pyridine initially leads to monohydrazide, evidently, in the form of two isomers 99a
and 99b. The acetyl group in this case is converted to hydrazone [45]. In the subsequent
treatment of the reaction mass by HCl, there occurs the intramolecular cyclization to form
an additional six- or seven-membered cycle 100a or 100b. Hydrazone at this stage was con-
verted to initial ketone.
It is known that during the acylation of hydrazine after the addition of the first acyl
group the subsequent acylation, due to a partial deactivation of the first nitrogen atom, usu-
ally occurs by the second amino group [48]. Thus, in the interaction of hydrazine with an-
hydrides of aromatic dicarboxylic acids symmetric hydrazides are formed with the increase
of the initial five-membered cycle to the six-membered cycle (Scheme 29) [49].
In the case of bacteriopurpurine, this could lead to the formation of hydrazide with six-
100a or seven-membered 100b cycles. The use of 2D heteronuclear resonance and the study
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 29
HO 0 0
NH N NH N
N HN N HN
O O
CO2Me CO2Me
96
A C
0 OH
H
NH N NH N
B
N HN N HN
O O
CO2Me CO2Me
95 97
D E
CN
OH
NC
NH N N N
Zn
N HN N N
O O
CO2Me CO2Me
98
NH N NH N
A
H N HN H N HN
H H 2
O O COR1 COR
O
CO2H CO2H
57 99
a: R1=NHNH2; R2=OH
b: R1=OH; R2=NHNH2
B,ɋ
O O
H H
H H
NH N NH N
H N HN H N HN
H H
O O O N N O
N
CO2Me NH2 CO2Me H H
100a 100b
.
A: N2H4 H2O, pyridine; B: 1N HCl; C: CH2N2, Et2O
O O
NH
O + NH2 NH2
NH
R R
O O
Scheme 29 Formation of cyclic hydrazides in the interaction of hydrazine with anhydrides of aro-
matic dicarboxylic acids.
O O
H H
H H
N N N N
A
Mg Mg
H N H N N
H H
HO2C
HO2C O O O
MeO2C MeO2C
101 H2N O 102
O
O
H
H
H
H
NH N
C N N
Pd
H N HN H N N
HO2C H
HO2C H
O O O O
H2N MeO2C
O 103 H2N MeO2C 104
O
residue and re-etherification of the carbomethoxy group in the pentanone ring lead to stable
derivatives of [Pd]-Bchlfd a with a high photodynamic activity [55]. The quantum yield of
AOS for these compounds is sufficiently high – from 1 in nonpolar solvents down to 0.5 in
aqueous solvents. Bacteriochlorophyllide a 101 and its complex with serine 102 showed a
high anti-tumour activity on the M2R mouse melanoma cell line (LD50 0.2–0.5 µM)
(Scheme 30) [56]. Substitution of Mg by Pd 104 increased the photodynamic activity,
which led to a decrease of the value of LD50 to 0.01–0.03 µM. Besides, owing to the high
fluorescence of 102 in cancer tissue as compared with healthy tissue (the selectivity reaches
8–10), the latter, along with PDT, can be also used for fluorescent diagnostics (FD).
A series of new negatively charged water-soluble derivatives of Bchl a was obtained
at the Weizmann Institute of Science, Israel. These compounds proved promising for vas-
cular-targeted photodynamic therapy (VPT). The damage of vessels providing for the blood
supply of the tumour was found to be the essential factor of its necrosis. VPT is efficient in
the treatment of hard tumours, as well as non-tumour processes associated with increased
vascularization, e.g., age-related macular degeneration. Regression and necrosis of the tu-
mour occur as the result of both the death of cancer cells and due to the occlusion or perfo-
ration of tumour vessels. However, hemorrhagic necrosis of the tumour, caused by the
above-named water-soluble derivatives of bacteriochlorophyll a, can have undesirable con-
sequences, especially when tumours are localized in such vital organs as bronchi and lungs.
The method of VPT leads to necrosis of the central part of the tumour, which can spread to
95% of tumour’s volume. However, cancer cells on the periphery of the tumour can fail to
32 M.A. Grin and A.F. Mironov
O
O
N N
N N
M
Pd
N N
N N
O O NH
CO2CH3 CO2CH3
R O CO2-K+
109 SO3- K+
56:M=Mg, R=phytyl
105:M=2H, R=OH
106:M=Pd, R=OH (WST09, Tookad)
107:M=Pd, R= O-succinimide-SO3-Na+a+
108:M=Pd , R=NH-(CH2)3-SO3-Na+
SO3- K+
O N O
N N N N
N N
Pd Pd
Pd
N N N N
N N
O NH O NH
NH NH CO2CH3 O NH
O O
CO2CH3 CO2-K+ CO2CH3
SO3- K+ SO3- K+
SO3- K+
SO3- K+ SO3- K+
Scheme 31. Negatively charged PS based on bacteriopheophorbide and its metal complexes.
die and grow again. In this connection, the VPT method is not universal and can be consid-
ered only in combination with chemotherapy, radiotherapy and photodynamic therapy.
Water-soluble negatively charged PS were obtained by aminolysis of the pentanone
exocycle in bacteriopheophorbide and its metal complexes (Scheme 31). Their hydrophilic
properties and capability of aggregation in aqueous solutions was studied [57]. Amphiphi-
licity was assessed by the octanol/water distribution coefficient (P). The effect of periph-
eral substituents on the ability of obtained compounds to be solved in water and polar
organic solvents was shown. It was found that the opening of the exocycle in bacte-
riopheophorbide significantly increased the hydrophilicity of bacteriochlorins and, as a
consequence, enhanced their ability to be solved in polar organic solvents, including meth-
anol, ethanol, DMFA and DMSO. Besides, compounds with the open cycle 109–116 are
well soluble in aqueous solutions (PBS) – up to 40 mg/ml, which is much greater than the
solubility of compounds with the pentanone ring, but having polar substituents on the pe-
riphery of the macrocycle (compound 108, 4 mg/ml). The coefficient P strongly increases
(1:19) at the substitution of the central metal ion Pd(II) by Mn(III) 116. At the same time,
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 33
a decrement of the hydrogen chain in the alkyl substituent 109 as compared with 112 does
not in practice affect the distribution coefficient. All compounds with the open exocycle in
aqueous solutions form aggregates (2–8 molecules), which dissociate upon dilution, inclu-
sion into micelles, as well in solutions with physiological concentrations of serum albumin.
Among new PS of the bacteriochlorin series developed at the Weizmann Institute of
Science in cooperation with Steba-Biotech (France) and Negma-Lagards (France), we
should note the second-generation Tookad, which is a Pd complex of bacteriopheophorbide
106. Tookad exhibits a high activity to such tumours as rat glioma [58], prostate cancer [59],
HT29 human colon carcinoma in experiments on laboratory animals.
The main mechanism of action includes tumour vessel damage, which leads to hypoxy
and necrosis of the tumour. At present, the preparation is at the second stage of clinical tests,
which are carried out at medical centres of Canada, Europe and Israel. As Tookad is poorly
soluble in aqueous solutions (octanol:water = 24:1), it is introduced as a suspension with
Cremophor.
At the action on the Pd complex of bacteriopheophorbide by taurine (2-sulfo-
ethylamine) there occurs the opening of the pentanone cycle and the formation of dianion
(code name WST11) 112 [60]. The preparation possesses a good solubility in phosphate
buffer (up to 50 mg/ml), where it is present in the shape of small aggregates. In blood serum-
containing solutions, it is subjected to deaggregation and occurs as monomers in a complex
with serum albumin (BSA) and high-density lipids.
O
Absorption
N N
Pd
N N
O SO3Na
N
CO2CH3 H
CO2H
nm
112
O
C CF3
OH O
H H R1 H
H H H
NH N NH N NH N
A B
H N HN H N HN
H N HN
H H
H O O
O O N
O O N
N OMe
OMe CO2Me
CO2Me
CO2Me OMe
117 118 119 - 122
119 O 121 O O
O O OH
HO
120 O OH 122 O OH
O
O O O
H H H
H H H
NH N NH N NH N
H N HN H N HN H N HN
H H H
MeO2C O MeO2 C O O MeO2C O O
N O N N
NH2 HN O HN O
C C
N N I
is absolutely consistent with the data by Israeli investigators for the action mechanism of
bacteriochlorophyll a derivatives.
Much less is known of the introduction of positively charged substituents into bacte-
riochlorins. However, just these photosensitizers may prove to be the most efficient in pho-
todynamic antimicrobial therapy. The method is based on the inactivation of viruses,
bacteria, yeasts and protozoa by active oxygen species, which are generated by photosen-
sitizers in illumination [62].
Available literature data show that this method is considerably behind photodynamic
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 35
O O
H O
H H
H H H
NH N NH N NH N
C
H N HN HN
H N H N HN
H H H
O N O O N O O N O
CO2Me CO2Me
HN O N O CO2Me
H 3C NHMe
123 125 127
N N
A
B
O O
H O
H
H H
H
H
NH N NH N
NH N
H N HN H N HN
H N HN
H H
H
O N O O N O
O N O
CO2Me
CO2Me N O
HN OH CO2Me
CH3 NH2
100
N
N
123 a 126
therapy of cancer by the level of fundamental elaboration and practical application. There
are only separate data on the photosensitization of nonpathogenic yeasts in the presence of
porphyrins and phthalocyanines [63–65]. As is known, the outer surface of bacteria carries
a negative charge, in connection with which the most efficient coupling to bacterial cells
and the photodynamic action is expected from cationic photosensitizers.
Such photosensitizers based on N-cycloimides of bacteriochlorin have been synthe-
sized at our laboratory. Initially, it was planned to use for these purposes the N,N-dimeth-
ylamino derivative. However, the treatment of it by excess methyl iodide or dimethyl
sulfate failed to lead to the quaternization of the nitrogen atom [46].
A more successful effort was introduction into the N-amino derivative 100 of the isoni-
cotinic acid residue 123 followed by the quaternization of the nitrogen atom in the pyridine
ring 124 (Scheme 33) [66]. For this, cycloimide 100 was treated with isonicotinic acid chlo-
roanhydride in pyridine. We showed hydrazide obtained to exist in two isomeric forms 123
and 123a, which are formed owing to keto-enolic tautomerism or due to the absence of free
rotation around the bond C(O)–N.
36 M.A. Grin and A.F. Mironov
O O
H H
H H
NH N NH N
A B
100
H N HN H N HN
H H
O N O O N O
CO2Me CO2Me
HN O HN O
128 129
N
N
C C
O O
H H
H H
NH N NH N
H N HN N HN
H
H H
O N O O N O
CO2Me CO2Me
H2N O H2N O
N
I
130 N
131
O N O
N O
H
N
Me
To establish the structure of the isomers, chemical modification of cycloimide 123 was
carried out; it included the treatment of the latter by diazomethane (Scheme 34). As the
result, two substances with the same molecular mass but different chromatographic mobil-
ity were obtained, which enabled their isolation in the individual form. A more mobile iso-
mer was assigned the structure of N-methylhydrazide 125; and the other isomer, of
O-methylimidate 126.
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 37
NH N NH N
A B
100
H N HN H N HN
H H
O N O O N O
CO2Me CO2Me
HN O HN O
132 133
N
C C
O O
H H
H H
NH N NH N
H N HN H N HN
H H
O N O O N O
CO2Me CO2Me
HN O HN O
N
134 135
I
I
N
free acid in the synthesis (Scheme 38). For this purpose, N-aminocycloimide of bacterio-
chlorin with the residue of propionic acid 138 was taken as the initial compound. All stages
of the synthesis, including acylation by the chloroanhydride of isonicotinic acid 139 and
the subsequent quaternization of the nitrogen atom 136, were performed at the presence of
an unprotected carboxyl group in the molecule. This route proved to be less successful, and
the total yield did not exceed 18%.
The obtained betaine derivative of cycloimide of bacteriochlorin 137 possessed a
sufficiently high solubility in water, which made it possible to abandon Cremophor usually
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 39
O O O
H H H
H H H
NH N NH N NH N
A B
H N HN H N HN ɋ H N HN
H H H
O N O O N O O N O
CO 2Me H 2N CO 2H H 2N CO2 H 2N
O O O
O O
H H
H H
NH N NH N
A B
H N HN H N HN
H H
HO2C O HO2C O O
N O N
NH2 HN O
138 139
N
O O
H H
H H
NH N NH N
C
D
H N HN H N HN
H H
HO2C O O O O
N N
HN O O HN O
O
136 137
N x N
used for dissolution of hydrophobic compounds and possessing an intrinsic toxicity [67,
68].
40 M.A. Grin and A.F. Mironov
O
H
H
NH N
H N HN
H
O N O
CO 2Me HN
O
OH
N
O
HO O
X
OH
OH
140
Besides the development of PS themselves for PDT of cancer, of great interest in the
recent years is the development of means of delivering PS inside cancer cells to damage-
sensitive targets. Targeted intracellular transport can help to achieve a significant enhance-
ment of photodynamic action, which makes it possible to reduce the concentration of the
drug. Some of such endocyted conjugates are capable of accumulating in lysosomes or pen-
etrating into the nuclei. The group of Prof. A.S. Sobolev (Moscow) develops targeted de-
livery systems, which include the model ligand (insulin, alpha-melanocyte-stimulating
hormone), photosensitizer (chlorin e6, bacteriochlorin p), nuclear localization signal of
SV-40 large T-antigen, and, as a carrier, E. coli hemoglobin-like protein (Fig. 3) [72–75].
Earlier studies by this laboratory have shown that the conjugate of bacteriochlorin p
Synthetic and Natural Bacteriochlorins: Synthesis, Properties and Applications 41
His-tag – (His)6;
D-Tox – translocation domain of diphtheria toxin;
HMP – E.coli hemoglobin-like protein;
NLS – nuclear localization signal of SV-40 large T-antigen;
α-MSH – alpha-melanocyte-stimulating hormone.
The pocket for porphyrins is in HMP.
4 Conclusion
This review has a chemical trend, as it was written by organic chemists. It does not include
numerous works on the biological tests of bacteriochlorophyll a derivatives. It is beyond
doubt that these compounds are in extreme demand by oncologists and specialists of other
medical fields (dermatology, ophthalmology) in view of their prospective use as photosen-
sitizers for photodynamic therapy.
The first works on the synthesis of bacterio-derivatives, in which our group was also
involved, appeared 10–12 years ago; now some of these compounds are at various stages
of clinical tests. All this is indicative of rapid developments in this field of the chemistry of
tetrapyrrole compounds.
It is perfectly evident that for the successful use of PDT in clinical practice health pro-
cessionals should have a series of PS with various therapeutic absorption windows. Deriv-
atives of bacteriochlorophyll a, which absorb in the near IR region of the spectrum, have
their own field of use, where other PS prove little efficient.
References
1. D. Dolphin, Can. J. Chem., 72, 1005 (1994).
2. R. Bonnett, Chemical Aspects of Photodynamic Therapy, Gordon and Breach Science
Publishers, UK.
3. C. Musewald, G. Hartwich, F. Pollinger-Dammer, H. Lossau, H. Scheer and
M.E. Michel-Beyerle, J. Phys. Chem. B, 102, 8336–8342 (1998).
4. H.W. Whitlock (Jr.), R. Hanauer, M.Y. Oester and B.K. Bower, J. Am. Chem. Soc., 91, 7485
(1969).
5. R. Bonnett, R.D. White, U.-J. Winfield and M.C. Berenbaum, Biochem. J., 261, 277–280
(1989).
6. R. Bonnett, P. Charlesworth, B.D. Djelal, S. Foley, D.J. McGarvey and T.G. Truscott, J. Chem.
42 M.A. Grin and A.F. Mironov
43.A.F. Mironov, M.A. Grin, D.V. Dzardanov, K.V. Golovin and Y.K. Shim, Mendeleev Commun.,
205–206 (2001).
44. A.F. Mironov, A.V. Efremov, O.A. Efremova, R. Bonnett and G. Martinez, J. Chem. Soc. Perkin
Trans., 1, 3601 (1998).
45. S. Sasaki and H. Tamiaki. J. Org. Chem., 71 (7), 2648–2654 (2006).
46. A.F. Mironov, M.A. Grin, A.G. Tsiprovskiy et al., J. Porphyrins Phthalocyanines, 7, 707–712
(2003).
47. D.E. Remy, R.E. Whitfield and N.L. Needles, J. Chem. Soc. Chem. Commun., 681–695 (1967).
48. A.R. McCarthy, W.D. Ollis, A.N. Barnes et al., J. Chem. Soc. (B), 1185–1193 (1969).
49. K. Belnaik, E. Domagaliva and H. Hopkala, Roczniki Chem., 41, 831–843 (1967).
50. A.F. Mironov, M.A. Grin, A.G. Tsiprovskiy et al., Russian Federation Patent 2,223.274,
10 February (2004) (in Russian).
51. M. Hoebeke, H.J. Schuitmaker, L.E. Jannink et al., Photochem. Photobiol., 66, 502–508 (1997).
52. A. Scherz, J. Salomon and L. Fiedor, EP Appl., 584552 (1994).
53. L. Fiedor, V. Rosenbach-Belkin, M. Sai and A. Scherz, Plan. Physiol. Biochem., 34, 393–398
(1996).
54. H. Stiel, K. Teuchner, D. Leupold, H. Sheer, Y. Salomon and A. Scherz, Photochem. Photobiol.,
72, 204–209 (2000).
55. A. Scherz, Y. Salomon, A. Brandis and H. Scheer, PCT Patent WO00/33833 (2000).
56. I.G. Meerovich, I.Yu. Kubasova, N.A. Oborotova, G.A. Meerovich, S.A. Demura, A. Brandis,
V. Rosenbach-Belkin, A.Yu. Baryshnikov and A. Scherz, Proc. SPIE, 5973, 121–131 (2005).
57.
58. S. Schreiber, S. Gross, A. Brandis, A. Harmelin, V. Rosenbach-Belkin, A. Scherz and
Y. Salomon. Int. J. Cancer, 99, 279–285, (2002.
59. N.V. Koudinova, J.H. Pinthus, A. Brandis, O. Brenner, P. Bendel, J. Ramon, Z. Eshhar, A. Scherz
and Y. Salomon, Int. J. Cancer, 104, 782–789 (2003).
60. O. Mazor, A. Brandis, V. Plaks, E. Neumark, V. Rosenbach-Belkin, Y. Salomon and A. Scherz,
Photochem. Photobiol., 81, 983–993 (2005).
61. G.V. Sharonov, T.A. Karmakova, R. Kassies, A.D. Pljutinskaya, M. Refregiers,
R.I. Yakubovskaya, M.A. Grin, A.F. Mironov, J.-C. Maurizot, P. Vigny, C. Otto and
A.V. Feofanov, Free Radicals in Biology and Medicine, 40, 407–419 (2006).
62. M. Wainwright, J. Antimicrob. Chemother., 42, 13–28 (1998).
63. Z. Malik, I. Hanania and J. Nitzan, Photochem. Photobiol., 5, 281–293 (1990).
64. T. Ito, Photochem. Photobiol., 34, 521–524 (1991).
65. M.G. Strakhovskaya, A.F. Mironov and A.M. Seregin, Dokl. Akad. Nauk, 384, 263–266 (2002)
(in Russian).
66. A.F. Mironov, M.A. Grin, A.G. Tsiprovskiy, R. Titeev, E. Nizhnik and I. Lonin, Mendeleev
Commun., 5, 204–207 (2004).
67. A. Grichin, A. Feofanov, T. Karmakova, N. Kazachkina, E. Pecherskih, R. Yakubovskaya,
A. Mironov, M. Egret-Charlier ana P. Vigny, Photochem. Photobiol., 73, 267 (2001).
68. A. Feofanov, A. Grichin, T. Karmakova, V. Lebedeva, A. Filyasova, R. Yakubovskaya,
A. Mironov, M. Egret-Charlier ana P. Vigny, Photochem. Photobiol., 75, 633 (2002).
69. K.R. Adams, M.C. Berenbaum, R. Bonnett et al., J. Chem. Soc. Perkin Trans., 1, 1465–1470
(1992).
70. A.A. Aksenova, Yu.L. Sebyakin and A.F. Mironov, Bioorg. Khimiya, 26 (2), 126–129 (2000)
(in Russian).
71. G. Zheng, A. Graham, M. Shibata, M.R.J. Missert, A.R. Oseroff, T.J. Dougherty and
R.K. Pandey, Org. Chem., 66, 8709–8716 (2001).
72. T.V. Akhlynina, D.A. Jans, A.A. Rozenkranz et. al., J. Biol. Chem., 272, 20328–20331 (1997).
73. T.V. Akhlynina, D.A. Jans, N.V. Statsyuk et al., Int. J. Cancer, 81, 734–740 (1999).
74. A.A. Rozenkranz, D.A. Jans and A.S. Sobolev, Immunol. Cell. Biol., 78, 452–464 (2000).
75. A.S. Sobolev, D.A. Jans and A.A. Rozenkranz, Prog. Bioph. Mol. Biol., 73, 51–90 (2000).
76. A.A. Rozenkranz, V.G. Lunin, P.V. Gulak, O.V. Sergienko, M.A. Shumiantseva, O.L. Voronina,
D.G. Gilyazova, A.P. John, A.A. Kofner, A.F. Mironov, D.A. Jans and A.S. Sobolev, FASEB J.,
17, 1121–1123 (2003).
77. E.O. Artemenko, D.G. Gilyazova, A.A. Rozenkranz, V.G. Lunin, O.V. Sergienko,
K.N. Timofeyev, M.A. Grin, A.F. Mironov, A.B. Rubin and A.S. Sobolev, Mol. Meditsina, 4,
43–47 (2005) (in Russian).
44 M.A. Grin and A.F. Mironov
- после 25-ой схемы идет 27; две схемы 33
2 meso-Phenylporphyrins as
Synthetic Models of Natural
Porphyrins: Synthesis and
Modification
Introduction
Porphyrins and metalloporphyrins are widespread in nature and are of great biological sig-
nificance. Their most important representatives are chlorophyll, which, as part of the
protein– lipid complex, performs the initial stage of photosynthesis in green plants, and
blood heme, which, in combination with the protein globin, performs the reverse binding
and transportation of oxygen to living cells. Metalloporphyrins are also present in some
enzymes – catalases, peroxidases, etc. At present, interest in porphyrins and their metal
complexes is extremely great. Results obtained in studies of the chemistry and properties
of these compounds are discussed in detail in a number of monographs [1 – 10].
Real ways have been outlined for practical applications of porphyrins and their metal
complexes as efficient catalysts of oxygen electroreduction, oxidation of sulfur dioxide in
electrochemical synthesis of sulfuric acid and fuel hydrogen, polymerization regulators of
acrylates, as photooxidation sensitizers, medical preparations and analytical reagents, semi-
conductors and model compounds in studies of such biological processes as photosynthesis,
46 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
reversible binding of oxygen, enzyme catalysis [6]. The successful development of these
trends depends on reliable methods of synthesis and modification of porphyrins possessing
the required physicochemical properties and resistant to aggressive media and reagents.
The broad use of porphyrins in engineering, technology and medicine is restrained by
the low accessibility of most porphyrins, many of which are obtained with very low yields.
In this connection, of special interest and topicality are issues of the chemistry of synthetic
porphyrins. In this context, one’s attention is attracted by porphyrins containing in meso-
positions aryl substituents, which can be subjected to various chemical transformations. At
present, the properties of natural porphyrins are modelled mainly using their synthetic ana-
logues meso-tetraphenylporphyrins 1, readily obtained by condensation of commercially
available pyrrole with benzaldehydes. However, in some cases they are not very con-
venient, as, unlike natural porphyrins, they have no alkyl or pseudoalkyl substituents in
β-positions of the porphyrin cycle, whereas meso-positions are substituted. On the other
hand, sufficiently accessible octaalkylporphyrins 2 are not always convenient, either, as they
have no active groups, which can be changed to impart them with various physicochemical
properties. Therefore, of great interest are porphyrins, which combine the advantages of
these two classes, such, for instance, as 5,15-diaryloctaalkylporphyrins 3 and 5-aryl-
octaalkylporphyrins 4. These porphyrins are produced mainly by condensation of α-unsub-
stituted linear derivatives of pyrroles with benzaldehydes.
Ar R R R Ar R R Ar R
R R R R R R
NH N NH N NH N NH N
Ar Ar
N HN N HN N HN N HN
R R R R R R
Ar R R R R R Ar R
1 2 3 4
o NH N
140-240 C
+ RCHO R R
N N HN
H
R
5
Scheme 1
and 237 against 115°C, respectively), which makes it possible to perform the condensation
reaction at an atmospheric pressure and in the presence of oxygen of the air as an oxidant.
By the Rotmund’s method, the reaction is performed at high concentrations of the re-
acting substances (more than 1 M), but a comparatively low yield of meso-substituted por-
phyrins restricts its application. Methods of high-temperature synthesis have been modified
to date, which sometimes makes it possible to completely avoid the use of the solvent. Thus,
the reaction of a mixture of pyrrole and benzaldehyde in the presence of a metal salt (in the
absence of a solvent) in a sealed ampoule at 150– 250°C produces H2TFP with a yield ex-
ceeding 50% [18]. Treatment of a mixture of pyrrole, benzaldehyde and silica gel or ceolite
in a microwave oven leads after a chromatographic purification to H2TFP with a yield
reaching 9% [19–21]. Introduction of pyrrole into the ampoule containing aldehyde in the
gas phase (200–250°C) in a air atmosphere leads to the production of some meso-substi-
tuted porphyrins with yields of up to 23% [22].
Recently [23] tetraphenylporphins were reported to be obtained with yields of more
than 20% in a condensation reaction in 2,4,6-trichlorophenol during the boiling in the pres-
ence of air at reagent concentrations of about 0.6 mol/kg. It is difficult to assign this reaction
to the Adler’s method (see further) due to the small acidity of this phenol (pKa = 6.1). Be-
sides, it is of interest that manganese complexes of many even di-ortho-substituted tetraphe-
nylporphyrins are obtained with yields exceeding 50%, which implies the presence of a
template effect on the Mg(II) ion.
It should be noted that synthesis of meso-substituted porphyrins in basic media, unde-
servedly forgotten in the recent years due to the discovery of acid-catalysis methods, has
its specific areas of application, such, for instance, as synthesis of meso-phenylporphyrins
having in phenyl rings groups labile in an acidic medium or containing in meso-positions
some heterocyclic residues (e.g., furanic or pyrrolic).
Based on the works [13, 24] (their authors obtained with a small yield H2TPP at a pro-
longed boiling of a mixture of pyrrole and benzaldehyde in a methanol– pyridine medium),
Adler [25], when studying the effect of various solvents, found that the condensation reac-
tion is catalyzed by acids more efficiently than by bases. This conclusion was confirmed in
the work by Treibs [26]. Adler [25] determined that when performing the reaction in
aerated, boiling, acid-containing organic solvents the yield of H2TPP by the spectrophoto-
metric estimate reached 40%. At present, the condensation reaction of pyrrole with alde-
hydes (Scheme 2) in acid-containing medium in the presence of the air is one of the major
methods of producing meso-substituted porphyrins (the Adler’s method). The reagents
mainly used as acidic solvents are acetic acid [25, 27], propionic acid [28, 29]; mixed sol-
vents: pyridine-acetic acid [26], benzene-chloroacetic acid [25], toluene-p-toluene sulfo
48 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
NH N
air
+ RCHO R R
H+
N N HN
H
R
5
Scheme 2
acid [30], xylene-chloroacetic acid [31– 33] and some others. The reaction is usually per-
formed at the boiling temperature of the solvent, sometimes at the passage of air through
the reaction mixture [26]. It has been found [34] that the yield of porphyrins depends on the
temperature of the reaction medium, and the largest yield is obtained at temperatures close
to 140°C. At lower temperatures, the porphyrin formation rate is small, and at higher tem-
peratures the rate of oxidation of porphyrins produced is high. An optimal concentration of
the reagents in the reaction medium is of the order of 0.2– 0.4 mol/l.
Finally, there are new modifications of the Adler’s method for the synthesis of meso-
substituted porphyrins: condensation of pyrrole with aldehydes in a medium of dimethyl-
formamide (dimethylsulfoxide)-aluminium trichloride (yield of H2TFP, 30%) [35], or in
propionic acid with microwave irradiation (yield of meso-tetraarylporphyrins, 20–43%)
[36]. The reaction of anisic aldehyde with pyrrole at 120°C in propionic acid containing
30% nitrobenzene, which in this case is an oxidant, made it possible to obtain a respective
porphyrin with the yield of 45% [37]. The yields of 5–20% were observed for a number of
other arylaldehydes [37, 38].
R HH
H
H
NH N
R R
N HN
R
6
It should be noted that under conditions of the reaction of acidic condensation, respec-
tive chlorins 6 are also formed besides porphyrins, and in some cases chlorins become the
main reaction products [39]. However, they can easily be transformed into respective por-
phyrins by the treatment with benzoquinone derivatives: para-chloranil (p-CA) or 2,3-
dichloro-5,6-dicyanobenzoquinone-1,4 (DDQ) [40–43].
The use of a mixture of valeric acid and nitrobenzene at 160°C for the condensation
of pyrrole with benzaldehydes makes it possible to significantly increase the yield of por-
phyrins, which contain no chlorins [44].
Isolation of porphyrins causes no problems, when they are crystallized during the
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 49
H R R
CF3COOH
or NH HN NH N
BF3etherate R H DDQ R
+ R-CHO R
N CH Cl , 25oC H
2 2 NH HN
R
N HN
H
R H R
7 5
Scheme 3
cooling of the reaction mass in a sufficiently pure form [26, 29]; in other cases, the solvent
is driven off under vacuum [45] or with steam [31] and the residue is chromatographed. In
the case of using acids as solvents the reaction mixture can be neutralized by a solution of
ammonia, the residue be filtered and porphyrin also be isolated by chromatography on a
suitable sorbent [46].
The Adler’s method proved itself to be excellent for preparative-scale synthesis of por-
phyrins from aldehydes, which are relatively stable. The possibility to obtain quickly and
easily an approximately 20% yield of such porphyrins at reagent concentrations of up to
0.4 M makes the Adler’s method sufficiently convenient. The use of the Adler’s method is
limited in syntheses of meso-substituted porphyrins from aldehydes having substituents,
which do not withstand the action of acids at high temperatures, many 2,6-disubstituted ben-
zaldehydes and many aliphatic aldehydes.
A comparatively recently developed new method of synthesizing meso-substituted
porphyrins under mild conditions [47, 48] consists in the condensation of pyrrole with al-
dehydes in chloroform or methylene chloride in the presence of trifluoroacetic acid or boron
trifluoride etherate in an inert atmosphere at room temperature to respective porphyrinogen
7 with subsequent oxidation of the reaction mixture by a stoichiometric amount of DDQ or
p-CA (the Lindsey method) (Scheme 3). There is an example, where oxidation of porphy-
rinogen 7 is performed by hydrogen peroxide in acetic acid [49]. Besides the chemical
methods, meso-alkylsubstituted porphyrinogens, which are more resistant to oxidation, can
be converted to porphyrins photochemically [37].
The reaction, as found in [47], is sensitive to reagent concentrations. The highest yields
of H2TPP (35–40%) are obtained in the reaction of 10 mM benzaldehyde and pyrrole each,
and the yield decreases approximately two times at reagent concentrations of 100 mM each
and 1 mM each. The decrease of the yield at a higher concentration of the reagents can be
partially reduced by increasing the amounts of acid catalyst. For instance, at 100 mM ben-
zaldehyde and pyrrole each, the yield of H2TPP was 23% with 1 mM BF3 etherate; 30%,
with 3.2 mM and 29% with 10 mM BF3 etherate. These magnitudes are still lower than the
35– 40% yield obtained at reagent concentrations of 10 mM each in CH2Cl2 with BF3 ether-
ate or trifluoroacetic acid as a catalyst [50]. The reaction is also sensitive to the concentra-
tion of acid catalyst: for the reagent concentrations of 10 mM each, BF3 etherate is efficient
at a concentration of 1 mM, whereas trifluoroacetic acid is required at a much higher con-
centration of 20– 50 mM [47]. Addition of sodium or ammonium chlorides was found to
significantly increase the yield of porphyrins [51, 52].
By the example of tetra(mesityl)porphin (5 R = 2,4,6-trimethylphenyl; H2TMP), it has
been shown [53– 55] that, for synthesis of tetraphenylporphyrins containing electron-donor
50 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
substituents in both ortho-positions of the phenyl rings, at the stage of condensation it is re-
quired to use BF3 etherate as a catalyst in the presence of a cocatalyst, 0.75% ethanol. This
method was improved to obtain gram amounts of H2TMP [56, 57]. Cocatalysts in the syn-
thesis of H2TMP and similar porphyrins can be, besides ethanol, also ethylene glycol,
2-methoxyethanol [55] or methanol [58]. Syntheses of tetraphenylporphyrins containing
electron-acceptor substituents in both ortho-positions of the phenyl rings do not require the
use of cocatalyst. Application of 2,2-dimethoxypropane as a dehydrating agent also leads
to an increase of the yield of H2TMP [58].
A further improvement of the method was to increase the reagent concentrations and
to use oxygen of the air as an oxidant of porphyrinogen in the catalysis by iron phthalocy-
anine and p-CA [51]. This slightly decreases the yield of the end product, but makes it pos-
sible to use the method in large-scale syntheses [59].
Mild conditions of the reaction at the stages of condensation and oxidation enable the
use of the Lindsey method for a broad range of aldehydes. This method is of special signif-
icance for synthesis of spatially hindered porphyrins [54, 57, 60] and meso-alkylporphyrins
[61, 62], which are obtained by other methods with low yields. The yield of porphyrins can
reach 50% depending on the aldehyde. At present, this is the most widely used method of
synthesizing meso-substituted porphyrins.
There are several modifications of this method: use of pyrrole with aldehydes of clays,
such as monomorillonite and the like as condensation catalysts [20, 21, 63, 64]. The activity
of clays is ascribed to their porosity but not acidity, and their efficiency as condensation
catalysts is associated with stabilization of porphyrinogens in clay nanotubes [21, 65]. For
synthesis of porphyrins with polar groups, which are obtained by the Lindsey method with
low yield [47, 66], the method of micellar synthesis was proposed [67], which consists in
conducting the condensation reaction in hydrochloric-acid solution of dodecylsulfate used
as a surfactant, followed by oxidation of the reaction mixture by a solution of p-CA in tet-
rahydrofurane. This method makes it possible to obtain tetraphenylporphins having car-
boxy, hydroxy, acetamido, ether and ester groups in the phenyl rings.
R R
R
NH N NH N
N HN N HN
R R R
R R
D,E,E,E- 8 D,E,D,E- 8
R
R
R R R
NH N NH N
N HN N HN
R
R R
D,D,E,E- 8 D,D,D,D- 8
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 51
The Lindsey method has the mildest reaction conditions. The yields are in most cases
higher than those for the Adler’s method, though the reaction at reagent concentrations from
0.01 M up to 0.1 M requires large amounts of solvent to be removed. The method is appli-
cable in the case of aldehydes having substituents unstable to the action of acids, 2,6-dis-
ubstituted benzaldehydes, aliphatic aldehydes and little-accessible expensive aldehydes
(for production of porphyrins from which high yields are of great importance).
meso-Tetraphenylporphyrins, having in one of ortho-positions of the phenyl rings
bulky substituents, are formed as a statistical mixture of atropisomers 8 owing to the impos-
sibility of the turn of teh phenyl rings; in some cases, when substituents are polar, the mix-
ture can be separated chromatographically [68, 69].
The synthesis mechanism of meso-substituted porphyrins in an acidic medium
(Scheme 4) includes the interaction of pyrrole by the carbonyl group of aldehyde to form
pyrrylcarbinol 9. Further, it is protonated, yielding stabilized carbocation 10, which then
H+ H+ +
+ RCHO CHOH CH + CH
N N N N
R R R
H H H H
9 10
pyrrole RCHO
CHOH H CH OH
N H R N H+ N H R N N
R R n
H H H H H
11 12
H R H R H R H H
R R
N
H
NH HN NH HN NH HN
R H RCHOR H R H pyrrole NH HN
H R H R H R H H
NH HN NH HN NH NH HN
R R
H H NH
R OH R OH
1. cyclization
2. oxidation
R R R
R R
N
NH N H
NH HN NH N
N N
R R R R R R
N HN N HN N R R
NH NH HN
R R
13 5 14 15
Scheme 4
Using the Alder’s method, the linear polypyrrole chain can be also partially oxidized to the
closure into the cycle. Depending on the reaction conditions, relative concentrations of the
reagents and catalyst used the end result of the cyclization and subsequent oxidation can
differ. Thus, conditions were determined, at which, besides porphyrins 5, also corroles 13
can be formed (excess of pyrrole, yield up to 10%) [70], N-confused porphyrins 14 (yield
up to 10%) [71, 72] and tetraphenylsapphirins 15 (yield up to 1%) [73, 74] (Scheme 4).
The condensation reaction of pyrrole with aldehydes is also used for the synthesis of
compound structures containing porphyrin fragments. Thus, in condensation of dialde-
hydes, in which phenyl rings via ortho-positions are connected by a sufficiently long di-
oxymethylene chain [R = –O(CH2)nO–, n ≥ 4], with pyrrole, a mixture of three isomeric
belted porphyrins 16–18 is formed; herewith, the most strained porphyrin 16 is not formed
if the chain is too small (n = 4) [75–77]. At an even smaller chain length (n = 2, 3 and 4 for
meta-positions), under the condensation conditions, dimeric sandwiched porphyrins 19 are
formed [78, 79].
R R O R R O
O
O O O O O
NH N NH N NH N
N HN N HN N HN
O
O
R R
O O
16 17 18
O
R O R
O
O
N
HN N
HN
NH
N NH
N
O
O
R O R
O
19
R R R R
R R
R R
R R
R R
NH N NH N
NH N
N HN N HN
N HN
20 21 22
NH NH
O O
HO
O
O O
NH N
N HN
O
23
R
R
X
R
NH N NH N
X X X=
N HN N HN
R
X
R
R
24
The adduced exotic structures are usually obtained with yields not exceeding 10%;
however, synthesis of respective polyaldehydes presents no large problems in most cases,
which makes such porphyrins interesting in many biological and catalytic studies of por-
phyrin systems.
The first intermediate compounds in the condensation of pyrrole with aldehydes are
pyrrylcarbinols 9. Therefore, of great interest are studies of the possibility of using these
compounds for synthesis of meso-substituted porphyrins.
Three methods of synthesis of pyrrylcarbinols have been developed to date (Scheme
54 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
I RCOCl
N
MgX R LiAlH4 RLi H
(NaBH4) CHR III
N N N
RCON(CH3)2 O OH O
II H H H
N POCl3 25 9 26
H
Scheme 5
5). The first two methods are based on the initial production of 2-ketopyrroles: the reaction
of the pyrrole Grignard reagent with chloroanhydride of the acid yields 2-ketopyrrole 25
and 3-ketopyrrole with yields of ~60 and 15%, respectively (I) [89]; an alternative method
of producing 2-ketopyrroles consists in the interaction of N,N-dimethylamides with pyrrole
according to Wilsmeyer in the presence of phosphorus oxychrolide (II) [90– 92]. 2-Keto-
pyrroles 25 are then reduced by LiAlH4 or NaBH4 to pyrrylcarbinols 9. The direct one-stage
method of producing pyrrylcarbinols is performed by the reaction of pyrrole-2-carboxyal-
dehydes 26 with lithium or magnesium organic compounds (III) [90– 92]. The last two
methods have no drawback of the first method – nonselectivity of substitution in acylation
of pyrrole. Besides, it has been found [93, 94] that pyrrylcarbinols with perfluoroalkyl res-
idues are easily formed in condensation of hydrated perfluoroaldehydes with pyrrole in the
presence of alkali.
Pyrrylcarbinols 9 are self-condensed, yielding meso-substituted porphyrins under con-
ditions of Rotmund’s reaction (Scheme 6) [95]. The yields of porphyrins in this reaction are
comparable with those for condensation of pyrrole with aldehydes. Besides, coloured inter-
mediate compounds observed in condensation of pyrrole with aldehyde, are the same as in
self-condensation of respective pyrrylcarbinol. These results prove that pyrrylcarbinols 9
are indeed intermediate products in the condensation reaction in an acidic medium [95].
NH N
CHR H+
N R R
OH [O]
H N HN
R
9 5
Scheme 6
_ + + + +
0
_ _ _
+
NH N NH N
_ _ 0 0 _
+
0 _ 0 0 0
+ N HN
N HN _ _
+ _
0 _ + + +
+
saddled ruffled
The most known and basic method of distortion of the porphyrin cycle is introduction
of substituents to its periphery to adjacent meso- and β-positions; the distortion rises with
their number and size increasing.
Thus, in condensation of benzaldehyde with pyrrole and its 3,4-disubstituted analogue
under the action of boron trifluoride etherate, a mixture of porphyrinogens was obtained;
condensation of the mixture by DDQ led to six possible porphyrins 27–32 (Scheme 7)
[107– 109]. The X-ray diffraction analysis of specimens of these porphyrins showed [107]
the deformation of the macrocycle to grow from almost plane tetraphenylporphin 27 (A =
Ph) to strongly distorted dodeca-substituted porphyrin 32 (A = Ph, B = Et, Ph).
Condensation of 3,4-diphenylpyrrole with formaldehyde and benzaldehyde under the
56 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
A B A B A
B B
NH N NH N NH N
A A A A A A
N HN N HN N HN
B
A A A B
27 28 29
B A B B A B B A B
B B B B B B
NH N NH N NH N
A A A A A A
N HN N HN N HN
B B B
A B A B A B
30 31 32
Scheme 7
B B B A B B A B
B B B B B B
NH N NH N NH N
N HN N HN N HN
B B B B B B
B B B B B A B
33 34 35
B A B B A B B A B
B B B B B B
NH N NH N NH N
A A A A A
N HN N HN N HN
B B B B B B
B B B B B A B
36 37 32
Scheme 9
B A B
B B
B B
[O] NH N
+ ACHO A A
N H+
N HN
H
B B
B A B
32
It has been found [121–123] that the strong distortion of the porphyrin cycle also
results from the occurrence of bulky substituents, such, e.g., as tert-butyl or adamantyl in
meso-positions in the absence of substituents in β-positions of the cycle.
In production of distorted β-alkylsubstituted porphyrins 32 (B = Alk), it is better to use
the condensation in acid-containing alcoholic solvents followed by the isolation of porphy-
rinogen and its subsequent oxidation by DDQ in THF. For synthesis of β-phenylporphyrins
32 (B = Ar), the condensation is better to be conducted in boiling acetic acid followed by
oxidation of porphyrinogen or without its oxidation in boiling with DDQ, or else with
58 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
isolation and subsequent oxidation in boiling toluene also using DDQ. The use of the Lind-
sey method [120, 124] is justified in the synthesis of porphyrins 32 having o-disubstituted
phenyl rings in meso-positions.
It should be noted that condensation between benzaldehyde and pyrrole nonsymmet-
rically substituted in β-positions produces a statistical mixture of four randomeric porphy-
rins (39– 42).
B C A B C B B C A A C A
A B A A A B B B
NH N NH N NH N NH N
C C C C C C C C
N HN N HN N HN N HN
B A A A A A A A
A C B B C B B C B B C B
39 (type I) 40 (type II) 41 (type III) 42 (type IV)
43 44 45
porphyrins are obtained the most often in mixed-aldehyde condensation. The yield of A3B
porphyrin is maximal at a ratio of the reacting aldehydes 3:1: A4, 31.64%; A3B, 42.19%;
cis-A2B2, 14.06%; trans-A2B2, 7.03%; AB3, 4.69%; B4, 0.39%. A larger content of alde-
hyde A increases the yield of A4 porphyrin at a decrease of the absolute amount of A3B por-
Scheme 10
NH N NH N NH N
A A A A A B
N HN N HN N HN
A A A
A4 A3B cis-Ⱥ2ȼ2 (ȺȺȼȼ)
46 47 48
B B B
NH N NH N NH N
A A A B B B
N HN N HN N HN
B B B
trans-Ⱥ2ȼ2 (ȺȼȺȼ) Ⱥȼ3 ȼ4
49 50 51
phyrin. However, the ratio of the reacting aldehydes, which gives the highest isolated yield
of A3B porphyrin, depends on the real reactivities of these aldehydes, as well as on the ease
of separation of the porphyrin mixture [128]. When a mixture of strongly-polar hard-of-
access aldehyde and low-polar aldehyde is used to produce AB3 porphyrins, it is sometimes
more advantageous to use a large excess of low-polar aldehyde, so that the strongly-polar
one reacts more completely. Herewith, symmetrically substituted B4 porphyrin is mainly
produced, which is, however, easily separable chromatographically from the required AB3
porphyrin [129–131]. Using this method, a number of dimeric porphyrins linked via meso-
positions by various spacers was synthesized [131–136].
In practice, isolated yields of AB3 porphyrins in Adler’s method are mainly at the level
of 5%, with an approximate total yield of the porphyrin mixture 20% and a statistical dis-
tribution of the mixture.
Reactivities of aldehydes in mixed-aldehyde condensation sharply differ in most cas-
es; for this reason, namely the Lindsey method is the most suitable for such syntheses, or a
modified Adler method, in which the first stage of the reaction is performed in the absence
of oxygen of the air to avoid oxidation of porphyrinogen formed by more reactive aldehyde
and to achieve an equilibrium between porphyrinogens, which is set by the ratio of the ini-
tial aldehydes.
Sometimes, the reaction is conducted with two aldehydes of close reactivities or with
60 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
O N O
O + O
O H O
O O
O OHC R air O R
+ EtCOOH NH N
CHO OHC
N HN
CHO N O O
O O N
52 53 54
Scheme 12
H3C CH3
(CH2)nCO 2H CHO
N
(CH2)nCO 2H
3 + + 4 NH HN
N N
H H N
CH3
H3C CH3
55 56
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 61
Scheme 13
H R
NH R NH HN
TPAA or R
RCHO + NH +
BF3 etherate H H NH
excess NH
11 57
Ⱥ ɛ ɛ
made of trifluoroacetic acid or boron trifluoride etherate; the reaction is run at room tem-
perature in a solvent or without it. Removal of excess pyrrole under vacuum yields raw
dipyrrolylmethane, which is recrystallized [147], chromatographed [147] or distilled [146,
155] to obtain a pure product. The isolation of respective tripyrrane 57 from nonpurified
dipyrrolylmethane has been also reported [155].
In a similar way, using excess pyrrole, meso-substituted dipyrrolylmethane 11 was
synthesized from pyrrylcarbinol 9 with the yield of 95% (Scheme 14) at the catalysis by
trifluoroacetic acid or BF3 etherate [147].
Scheme 14
R NH
CHOH + NH PhCHO
N TPAA or
H R H NH
excess BF3 etherate
9 11
Scheme 15
B B
NH N NH NH N
BCHO A BCHO
A A A A
N HN CCHO H
NH N HN
C B
ABAC + ABAB+ACAC ABAB
59 11 58
ACHO
BCHO
B
NH N
A A
N HN
A
BA3 + ABAB+A4
60
substituents of dipyrrolylmethanes, ACBC porphyrin 61. However, in both cases other por-
phyrins are also formed (Scheme 16); their chromatographic separation is possible only in
the occurrence of polar groups in substituents.
Scheme 16
C A
NH N NH HN NH N
CCHO A B ACHO
A B + A B
H NH HN H N HN
N HN
C A
ACBC + ACAC+BCBC BA3 + ABAB + A4
61 60
It should be noted that for such kind of syntheses, the suitable method is only the
low-temperature Lindsey method, as acidolysis of meso-substituted dipyrrolylmethanes in
acid media at high temperatures and regroupings leads in the subsequent oxidation to the
formation of a compound mixture of porphyrins (Scheme 17).
However, in [165] it was shown that during the condensation in propionic acid in the
presence of zinc acetate only trans-porphyrin 58 without regrouping was formed, which is,
possibly, due to oxidation processes prior to the closure of the cycle. This reaction conduct-
ed by the two-stage Lindsey method forms a mixture of isomeric porphyrins, though with
a large yield [162, 165]
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 63
Scheme 17
NH pyrrole B NH
A H+ + BCHO +
CH + CH + H
H NH N N H+ N H NH
A B
H H H
A number of studies [157, 158] have shown dipyrrolylmethanes having spatially hin-
dered or perfluoroalkyl substituents in meso-positions to be the most resistant to acidolysis. .
The condensation reactions using meso-substituted dipyrrolylmethylcarbinols 62 in-
stead of a mixture of aldehyde and dipyrrolylmethane make it possible to avoid acidolysis
due to the impossibility of forming nonstabilized dicarbocation (Scheme 18).
Scheme 18
B B
+
CHOH CH
NH NH
A H+ A H+ + +
HC CH +
H H N N
NH NH A B
H H
62
A NH NH N
A A
H NH N HN
B
ABAB
62 58
B B
CHOH
NH HN NH N
A C
+ A C
H NH HN H N HN
HOHC
D D
ABCD + ABAB+CDCD
64
64 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
B B
CHOH
A NH HN NH N
C
+ A C
H NH H
HN N HN
CHOH
B B
ABCB
63 11 65
B B
CHOH
A NH HN NH N
C
+ A C
H NH H
HN N HN
CHOH
D D
ABCD
66 11 64
ɛ
To obtain carbinols 62 and 63, use can be made of the methods applied for synthesis
of pyrrylcarbinols (Scheme 5): acylation of the magnesium derivative of dipyrrolylmethane
by acylchlorides or according to Wilsmeyer to acyldipyrrolylmethane followed by its re-
duction to carbinol or interaction of respective carbaldehyde with aryllithium derivatives.
A new convenient method of synthesis of carbinols has been developed [169, 170]. The re-
action of o-mercaptophenol with carboxylic acid or its derivatives in the presence of BF3
etherate yields benzoxathiol tetrafluorate 67, a masked equivalent of aroyl. Benzoxathiol
Scheme 20
B S BO
B -
+ BF4 S
O S NH NHO NH
pyridine 66
+ A A A
CH3CN ɩpyridine
NH NH CH3CN NH
S
B O
67 11 68 69
tetrafluorate readily reacts in acetonitrile containing one equiv. pyridine at room tempera-
ture with dipyrrolylmethane 11 to yield 5- 68 or 5,5′-disubstituted dipyrrolylmethane 69
(the use of another benzoxathiazol tetrafluorate at the 2nd stage ultimately leads to nonsym-
metrically substituted dimethylcarbinol 66), whose oxidative hydrolysis to acyldipyrrolyl-
methanes and subsequent reduction yields required carbinols 62 or 63 [166] (Scheme 20).
There are methods of synthesis of nonsymmetric meso-tetrasubstituted porphyrins us-
ing tripyrranes ([3+1] condensation) [171] (Scheme 21) and bilanes [166] (Scheme 22). The
reaction of pyrrole with benzoxathiazole tetrafluoroborate 67 successively yields 2-substi-
Scheme 21
O
A A A
A O OH
- S
+ BF4 pyridine LiAlH4
O S NH HgO
+ NH NH
N CH3CN BF3 OEt2
H S
A O OH
A A
O
71
67
70
B
HO TFA, 25oC
excess
pyridine
HN
A B A
N N
HO
B H
NH HN NH
1. TFA H
N
N 2. DDQ
A B A
73 72
Scheme 22
B B B
O OH
NH NH NH NH HN
1. EtMgBr LiAlH4 BF3-Et2O
A A A A
2. BCOCl excess
ɩɢɪɪɨɥ
NH NH NH pyrrole
ɢɡɛɵɬɨɤ NH HN
O OH
B B B
11 63 74
B
1. CCHO, TFA
2. DDQ
NH N
A C
N HN
B
65
66 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
tuted pyrrole and then 2,5-disubstituted pyrrole 70 unlike the repeated substitution in posi-
tion 4 in the acylation reaction. The use of dicarbinol 71, obtained after the reduction, for
synthesis of tripyrrane 72 in excess pyrrole makes it possible to perform further the 3+1 con-
densation to obtain cis-A2B2 porphyrin 73.
The use of dipyrrolylmethane dicarbinol 63 makes it possible to obtain bilane 74 for
synthesis of ABCB-type meso-tetrasubstituted porphyrin 65 [166].
Thus, the use of linear meso-substituted oligopyrroles makes it possible to obtain in-
dividual nonsymmetric meso-tetrasubstituted porphyrins without impurities of isomeric
products. However, these methods require sufficiently complex syntheses of initial oli-
gopyrroles.
2 meso(5,15)-Disubstituted Porphyrins
5,15-Diphenylporphyrin and its β-alkylsubstituted analogues are of interest because these
compounds have some common features peculiar of 5,10,15,20-tetraphenylporphyrin,
namely the presence of two meso-phenyl groups, with some features of β-alkylporphyrins
owing to the occurrence of unsubstituted meso-positions, and porphin, in the absence of
β-substituents; herewith, such a porphyrin is not so inaccessible as the totally unsubstituted
initial porphin.
5,15-Diphenylporphyrins 4 of symmetric structure (trans-) have been known since
rather long in the form of octaalkylsubstituted derivatives, but in the form of β-unsubstitut-
ed analogues they appeared only recently, when unsubstituted dipyrrolylmethane 75 [26,
93, 172, 173] and meso-phenyldipyrrolylmethanes 11 [147, 172, 174] became relatively ac-
cessible.
Dipyrrolylmethane 75, which has neither meso- nor β-substituents, is obtained in three
stages from pyrrole and thiophosgene via thioketone 76 [175], which by the action of hy-
drogen peroxide in an alkaline medium is converted to ketone 77 [175], reduced to 75 under
the action of diborane with the total yield of about 40% [173, 176] (Scheme 23). Recently,
it was found that 75 could be obtained by direct desulfonation of thioketone 76 by Reney
nickel or sodium boron hydride [172]. Besides, it was possible to perform a small-scale con-
densation of polyformaldehyde (paraform) with excess pyrrole in the presence of boron tri-
fluoride etherate or trifluoroacetic acid, which yields dipyrrolylmethane 75 with the yield
of 40% in one stage [177, 178].
Proceeding from the features of structure of meso(5,15)-disubstituted porphyrins, sev-
eral possible routes for their synthesis can be presented (Scheme 24). All these methods
Scheme 23
S O
CSCl2 H2O2
OH- NH HN
N NH HN
H
76 77
(CH2O)n
BF3 Ni or NaBH4 B2H6
NH HN
75
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 67
were tested for synthesis of trans-substituted porphyrins, but the most applicable are routes
A and B, as routes C and D require an additional stage of producing formyldipyrrolyl-
methane. However, the use of route D makes it possible to obtain porphyrins nonsymmet-
rically substituted via phenyl rings without impurities of other porphyrins.
Scheme 24
R R
CHO
R R
NH H NH
+ ArCHO
A R Ar R C Ar
NH NH
R R
R R
R NH N R
78 80
R N HN R R
X R R CHO X
R R R
B D
H NH R Ar R H NH HN H
+ HC(OMe)3 4 +
Ar NH Ar NH HN Ar
R X = H, CO2H R R
X CHO X
R R R
79 81 79
Scheme 25
B B B
COOR COOH
A A A
COOR
B NH NH NH
CCHO C OH- R=Et C t C
NH
A H+ H H2(Pd/C) R=Bz H
NH H NH
NH
A A A
COOR COOH
B B B
83 84 85 86
Scheme 27
A B
+
RO2C CH2 B OH B B B
N
H2C
H
87 A A A A
+
NH HN -ɋH2O NH HN
H2O
RO2C CO2R RO2C CO2R
A B
88
RO2C CH2OH
N
H
Scheme 28
A B A B A B A B
+
+
R CH2X -X
- R .. CH2 R + CH2 R .. CH2
N N N N
H H H H
87 89
+ +
X = -Br, -Cl, -OCH3, -OH, -OCOCH3, -NR3, -NHR2 etc. R = CO2Et(Bz)
delocalization of charge on the pyrrole ring. Pyrrylcarbinyl cations, e.g., 90, can be incul-
cated via the pyrrole– methylene carbon link, followed by the release of the alternative pyr-
rylcarbinyl cation 91, which can lead to the removal of the aryl substituent from meso-
position and isomerization of β-alkyl substituents in porphyrins obtained in the subsequent
oxidation of porphyrinogens (Scheme 29).
Scheme 29
A H Ar
B
NH + A B B A
H H
H + H
H Ar CH2 + Ar
NH N N
B H H
90
A H Ar
B H H
A
NH + A B B A
H H
H + H
Ar H CH + H
NH N Ar N
A H H
91
B H H
B H H B H H B H H
A
+90 A B A + -91
A
NH NH H NH
H H
Ar Ar
Ar N NH H NH
NH H
A A B
B H H B H H A H Ar
Initially, the synthesis was conducted under conditions similar to the production of tet-
raphenylporphins by a one-stage method in solvents containing acid [26]; however, further
studies [190] showed the best results to be provided for by the use of a two-stage method
with intermediate formation of porphyrinogen 93 from 92 under the action of acid, and the
subsequent oxidation of 93 to porphyrin 94 by benzoquinone derivatives (Scheme 30).
Condensation of 5,5′-unsubstituted tetraalkyldipyrrolylmethanes 92 (A, B = Me, Et)
with benzaldehydes in methanol in the presence of strong inorganic or organic acid in an
Scheme 30
A A H R A A R A
B B B B B
NH + NH HN NH N
H [O]
+ RCHO
NH NH HN N HN
B B B B B
A A H R A A R A
92 93 94
70 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
B H H
B H H B NH
A
A A NH
NH HN B A
92a 92b
B B
A A
R NH N
B B O N HN O
A A
O A A O
NH N (CH2)n B B
(CH2)n
B B
N HN
A A A A
O NH N O
B B
95 N HN
A A
B B
96
dimeric porphyrins with four bridges 19; however, they exist in the form of three fixed at-
ropisomers with different distances between the reaction centres of the porphyrin nuclei
α,α,α,α-, α,α,α,β- and α,β,α,β- 97. Other possible isomers are not formed due to too large
strains in bridge groups [200, 201].
NH N NH N N HN
N HN N HN NH N
NH N NH N N HN
N HN N HN NH N
NH N
S
N HN
CH2 CH2 H2C CH2
S S
NH N
H2C CH2
H N HN H NH N
N HN
98 99
72 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
mixture of six porphyrins formed in condensation of two aldehydes and pyrrole (Scheme
10), so this method is also used sufficiently widely [207, 209–214]. It should be noted that
this method sometimes makes use of a large excess of one, more accessible aldehyde, so
that the other aldehyde is used more completely [215, 216]. Other examples include the syn-
thesis of rotoxanes based on porphyrins [217–219] and oligoporphyrins [207, 220].
Scheme 31
C C H A C C A C
D D D D D
NH A-CHO H + NH HN NH N
[O]
+ + A2 + B2
B-CHO
NH NH HN N HN
D D D D D
C C H B C C B C
92 100 (AB)
As it was pointed out earlier, the reaction (route A) is complicated by that it forms, to
a greater or smaller extent, impurities of monophenylporphyrins due to partial regrouping
in an acidic medium of dipyrrolylmethanes or intermediate tetrapyrroles [190] (Scheme
29). In some cases, monophenylporphyrins become the main or even sole reaction products,
especially at elevated temperature [183, 189, 190], however, their yield is lower than that
of diphenylporphyrins.
To rule out regrouping processes, of great interest is the use of a method similar to Rot-
mund’s synthesis of diphenylporphyrins 94. The method consists in the condensation of
dipyrrolylmethanes 101 with aldehydes in a medium of high-boiling heterocyclic solvents
(pyridine, quinoline, collidine) in the presence of coordination agents (zinc acetate)
(Scheme 32) [221]. In the case of using pyridine, the process is run at 180°C under pressure
using nitrobenzene as an oxidant, and the application of higher-boiling quinoline makes it
possible to conduct the reaction at boiling in the presence of air. The yields of phenyl-sub-
stituted porphyrins in this reaction are 10–20%; however, on the one hand, the method rules
out regroupings proceeding in an acidic medium, and on the other it enables using more ac-
cessible α,α′-dicarboxydipyrrolylmethanes 101 (X = COOH) stable to oxidation. Besides,
the method is of interest in condensation with aldehydes unstable in an acidic medium.
Another, the most known, method of synthesis of 5,15-diphenylporphyrins 94, first
used by Baldwin [222], consists in condensation of α,α′-unsubstituted or α,α′-dicarboxy-
meso-substituted dipyrrolylmethanes 102 with trimethoxy(ethoxy)methane in the presence
of trichloroacetic acid or trifluoroacetic acid in chloroform or methylene chloride [152, 172,
Scheme 32
A A R A
X
B B B
NH N N
Zn(OAc)2
+ RCHO Zn
pyridine,
NH nitrobenzene N N
B B B
X
A X=H, COOH A R A
101
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 73
Scheme 33
B B B B B
X
A A A A A
H NH H+ NH N NH N
H R [O]
+ HC(OMe)3 R R
R NH R H
N HN N HN
A A A A A
X
B X=H, COOH B B B B
102 103 94
ɛ / / ɮ
221, 223– 226] (Scheme 33) (route B). An oxidant of intermediate porphodimethene 103
can be oxygen of the air or 1,4-benzoquinone added at the end of the synthesis [223, 225,
226] or DDQ [152]. The reaction proceeds via formylation of 102, condensation of formyl-
dipyrrolylmethenes to porphodimethene 103 and its subsequent oxidation to porphyrin 94,
being, in this way, a simplified variety of routes C or D.
In [225], the yield of porphyrin 94 was shown to strongly depend on the extent of dry-
ing of initial α,α′-dicarboxylic acid of dipyrrolylmethane, whereas the use of α,α′-unsub-
stituted dipyrrolylmethane makes this reaction less dependent on its conditions. Application
of the benzoquinone oxidant at the last stage increases the yield of porphyrins as compared
with the use of air oxidation.
Method B is widely used for synthesis of belted porphyrins 95 [179, 222, 227]
(Scheme 33) and cyclophane dimeric porphyrins 88 [200] (Scheme 34). In this case, self-
condensation of bis-dipyrrolylmethane 104 or 105 is performed.
Scheme 33
R R
B B B B
COOH HOOC
O A A O O A A O
NH HN H+ NH N
+ HC(OMe)3
NH HN [O] N HN
A A A A
COOH HOOC
B B B B
104 95
Scheme 34
B B B
COOH
A A A
NH NH N
O NH O N HN O
A A
COOH
B H+ B B
(CH2)n B + HC(OMe)3 (CH2)n (CH2)n
[O] B B
COOH
A A A
O NH O NH N O
NH N HN
A A A
COOH
B B B
105 96
74 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
Scheme 35
B C B C
OHC
A D A D
NH HN NH N
R H H +
+ R R1
H NH R1 [O]
HN N HN
A D A D
OHC
B C B C
106 107 108
3 Synthesis of meso(5)-Phenyl-β-octaalkylporphyrins
There are several methods of synthesis of 5-phenyloctaalkylporphyrins 3. Earlier, it has
been shown [189, 190] that monophenylporphyrins are formed, owing to regroupings in an
acidic medium (Scheme 29), as an impurity in condensation of α-unsubstituted dipyrrolyl-
methanes with aldehydes; herewith, in some cases they are the main products of this reac-
tion, though their yield is lower than that of respective diphenylporphyrins.
The most currently widespread method of synthesis of 5-substituted porphyrins 113
consists in condensation of meso-substituted α-unsubstituted dipyrrolylmethanes 106 with
5,5′-diformyldipyrrolylmethanes 111 in alcohols or methylene chloride under the action of
strong acids (hydriodic, perchloric or p-toluenesulfo); the oxidant of intermediate por-
phodimethene 112 is oxygen of the air or benzoquinone derivatives (Scheme 37) [207, 225,
230–240].
5,5′-Diformyldipyrrolylmethanes 111 are produced by formylation of α-unsubstituted
dipyrrolylmethanes 92 by a mixture of POCl3 –DMFA according to Wilsmeyer [173] or a
mixture of ortho-formic ester and trifluoroacetic acid [241].
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 75
Scheme 37
A C A C A C
OHC
B D B D B D
NH HN + NH N NH N
H H H [O]
+ R
R NH HN R N HN
N HN
B D B D B D
OHC
A C A C A C
106 111 112 113
Scheme 38
A C A C
MeOH2C
B D B D
NH HN NH N
H
+ Br- R
R +
NH HN N HN
B D B D
MeOH2C
A C A C
106 114 113
Scheme 39
C C
CH3 CH2OMe
D D
C D
NH NH
HBr 1. Br2
Br- Br-
H3C CO2R HCOOH + 2. MeOH +
N NH NH
H
D D
R = H, Et, tBu CH3 CH2OMe
C C
115 116 114
76 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
Scheme 40
B B B R B
A A A A
+
NH HN + - NH N
H or OH
2Br- + RCHO
+
NH HN N HN
C C C C
D D D D
117 113
Scheme 41
C B B
X
D A A
B A +
NH + NH HN
H
2Br-
+ CHO +
NH N NH HN
D H C C
X X= H, COOH
C D D
101 109 117
Scheme 42
B B B B
A A A A
+
NH HN NH N
OH-
2Br-
+
NH HN NH HN
C C C C
D D D D
117 118
Scheme 43
HO3S SO3H
N N
H2SO4
NH HN NH HN
o
100 C
N N
HO3S SO3H
119
Isolation of compound 119 from the reaction mixture is associated with a number of
problems, however, is possible via its sodium or ammonium salts by way of multiple re-
sedimentation from a methanol– acetone solution [250], via the insoluble potassium salt
[251], as well as by dialysis [252].
To investigate the liquid-crystalline properties of H2TPP derivatives, the authors of
[253] conducted the sulfonation of covalently bound linear dimer of H2TPP to hexasul-
fophenyl derivative 120 with the yield of 90%. The reaction was carried out by heating por-
phyrin in concentrated sulfuric acid in a sealed ampoule on a boiling water bath for 4 h.
Similarly, sulfonation of mono-, di- and tetraphenyl-β-octamethylporphins was carried out
[254]. The results obtained show that the reaction is general irrespective of the number of
phenyl residues and the presence or absence of β-alkyl groups.
Electron-donor groups in phenyl rings facilitate sulfonation. Thus, in sulfonation of
β-octabromo-meso-tetramesitylporphin by oleum at 120°C all eight free positions of meso-
mesityl groups are sulfonated. This porphyrin was assumed to be used as a water-soluble
catalyst in oxygen transfer reactions [255, 256]. Production of sulfonated derivatives with
hydroxy- and methoxy- groups in ortho- and para-positions of phenyl rings [257– 259] and
meso-tetrakis(2-thienyl)porphin 121 was described [260]. These porphyrins were assumed
to be used in analytical purposes. It is interesting to note that sulfation of meso-tetrakis-
78 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
N N
NH HN NH HN
N N
The authors of [263], using the reaction of H2TPP to chlorosulfonic acid at room tem-
perature, obtained respective tetrachlorosulfonic derivatives. Chlorosulfonation of meso-
tetrakis(2,6-dichlorophenyl)porphin under the same conditions yields a mixture of mono-,
di- and tri-para-sulfonyl derivatives. Chlorosulfonation can be carried out with a high yield
at 100°C [263]; besides, tetrachlorosulfonic derivatives of H2TPP can be obtained by treat-
ing salts 119 with excess thionyl chloride in DMFA at 50°C for 3 h [264].
Chlorosulfonyl groups are easily converted to free sulfo acid, sulfamides or sulfo es-
ters at the nucleophilic attack with water, ammonia (amines) or alcohols, respectively [263,
264] (Scheme 44).
Scheme 44
ClO2S SO2Cl SO2OH
H2O
N
HNR2 SO2NR2
NH HN
N
HOR SO2OR
ClO2S SO2Cl
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 79
To obtain models of cytochrome P-450, porphyrin 123 was obtained by the treatment
of the sulfochloride derivative with respective amine [265, 266].
RO2S
Cl Cl
RO2S
Cl N Cl
Me
NH HN R = NH(CH2)3 Si OEt
OEt
Cl N Cl
SO2R
Cl Cl
SO2R
123
Scheme 45
NO2
N N
HNO3
NH HN 0 NH HN
CHCl 3, 0-5 C
N N
9
124
benzene rings of porphyrins obtained with high yields. Of great interest are porphyrins con-
taining in their phenyl rings active substituents, such as oxy or amino groups, as they are
capable of various chemical conversions. By direct synthesis, these porphyrins are obtained
with a low yield or are not formed at all.
Scheme 46
OCOR
-
OH
-
OH RHal
OSO2R OH OR
K2CO 3
BBr3
OCH3
Scheme 47
OH OCH2CO2Et OCH2CO2H
ClCH2CO2Et 1. OH-
K2CO3 2. H+
125 126
R A R A A R A
OH
B B B B aR=
NH N NH N NH N
Ar Ar NO2
N HN N HN bR =
N HN
C C B B
NH2
Ar D D R
A A ɫR=
127 128 129
A generalized scheme of the synthesis of porphyrins with active groups by the alkyla-
tion reaction is presented in Scheme 48.
As is seen in the scheme, the attachment of residues with active groups to porphyrin
82 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
Scheme 48
O(CH2)nX
Br(CH 2)nX
OH A
HX
B O(CH2)nBr
Br(CH 2)nBr
can proceed via two routes: alkylation of oxyphenylporphyrin by haloalkane with an active
group (route A), or preliminary alkylation of oxyporphyrin by excess α,ω-dibromoalkane
to form ω-bromoalkoxyphenylporphyrin with its subsequent interaction under similar con-
ditions with a compound having an active group (route B).
Route A is more preferable, as it is shorter and provides a higher yield of the end prod-
uct (with respect to initial oxyphenylporphyrin). However, it is not always possible to ob-
tain and purify haloalkane having an active functional group. In this connection, route B is
mainly used [279, 288].
only a small amount of groups prevent it, in contrast with reactions using Grignard reagents.
Application of the diazotization reaction enables production of an interesting and important
class of organic compounds – azo dyes.
Tetra(aminophenyl)porphins are readily diazotized by sodium nitrite in aqueous solu-
tions of mineral acids [296–298]. The obtained diazonium salts are rather stable. Their no-
ticeable decomposition with evolution of nitrogen is observed only at a temperature higher
than room temperature. The porphyrin cycle under conditions of the diazotization reaction
is stable; however, diazotized tetra(2-aminophenyl)porphin during the heating yields com-
pounds of nonporphyrin character. Owing to this, ortho-substituted tetraphenylporphins are
formed with low yields and a large amount of hard-to-separate impurities. However, the
azocoupling reaction of diazotized tetra(2-aminophenyl)porphin to phenol successfully
proceeds on the cold to form tetra(2-oxyphenylazophenyl)porphin. Diazotized tetra(3- and
4-aminophenyl)porphins during the heating yield tetra(oxyphenyl)porphins as well as enter
into the Sandmeyer reaction and the azocoupling reaction, which makes it possible to obtain
tetra(halogenophenyl)porphins and azo dyes based on porphyrins, which are impossible to
obtain in any other way (Scheme 49).
Scheme 49
F
I
HBF4
Cu KI
+
NO2 SnCl 2 NH2 HNO N2 Cu 2X2 X
2
HCl H+ X=Cl,Br,CN
t
HR
OH
N=NR
Scheme 50
N(CH3)2
CH3I
NH2 K2CO 3
CH3I +
N(CH3)3
2,6-lutidine
The chemistry of porphyrins and their analogues is rapidly developing, so this review
does not pretend to be a complete coverage of the literature on the subject. It can be recom-
mended as a strategy for the synthesis and modification of required meso-substituted por-
phyrins based on the methods developed at present the most.
References
1. B.D. Berezin, Coordination Compounds of Porphyrins and Phthalocyanine, Moscow: Nauka,
280 pp. (1978) (in Russian).
2. G.P. Gurinovich, A.N. Sevchenko and K.N. Solovyev, The Spectroscopy of Chlorophyll and
Related Compounds, Nauka i Technika: Minsk, 517 pp. (1968) (in Russian).
3. The Porphyrins and Metalloporphyrins, ed. by K.M. Smith, Elsevier: New York, etc., 590 pp.
(1975).
4. The Porphyrins, ed by D. Dolphyn, New York: Acad. Press, 7 v. (1978–1979).
5. Porphyrins: Structure, Properties, Synthesis, ed. by N.S. Enikolopyan, Nauka: Moscow, 333 pp.
(1985) (in Russian).
6. Porphyrins: Spectroscopy, Electrochemistry and Applications, ed. by N.S. Enikolopyan, Nauka:
Moscow, 384 pp. (1987) (in Russian).
7. M.R. Tarasevich and K.N. Radyushkina, Catalysis and Electrocatalysis by Porphyrins, Nauka:
Moscow, 168 pp. (1982) (in Russian).
8. B.D. Berezin and N.S. Enikolopyan, Metalloporphyrins, Nauka: Moscow, 160 pp. (1988)
(in Russian).
9. Advances of the Chemistry of Porphyrins, in four volumes, ed. by O.A. Golubchikov,
Chemistry Research Institute, St.-Petersburg University: St.-Petersburg (1997, 1999, 2001,
2004) (in Russian).
10. The Porphyrins Handbook, in 20 volumes, ed. by K.M. Kadish, K.M. Smith and R. Guilard,
Acad. Press: New York (2000–2005).
11. P. Rothemund, J. Amer. Chem. Soc., 57 (7), 2010 (1935).
12. P. Rothemund, J. Amer. Chem. Soc., 61 (9), 2912 (1939).
13. P. Rothemund, J. Amer. Chem. Soc., 63 (1), 267 (1941).
14. R.H. Ball, G.D. Dorough and M.A. Calvin, J. Amer. Chem. Soc., 68 (11), 2278 (1946).
15. J.H. Priesthoff and C.V. Banks, J. Amer. Chem. Soc., 76 (3), 937 (1954).
16. M.M. Williamson, C.M. Prosser-McCartha, S. Mukundan (Jr.) and C.L. Hill, Inorg. Chem., 27,
1061 (1988).
17. O. Bortolini, M. Ricci, B. Mennier, P. Frant, I. Ascone and J. Goulon, Nouv. J. Chem., 10 (1), 39
(1986).
18. D.B. Sharp, U.S. Patent 3.076.813 (1963).
19. A. Petit, A. Loupy, P. Mallard and M. Momenteau, Synth. Commun., 22 (8), 1137 (1992).
20. P. Laszlo and J. Luchetti, Chem. Lett., 3, 449 (1993).
21. M. Onaka, T. Shinoda, Y. Izimi and E. Nolon, Chem. Lett., 1, 117 (1993).
22. C.M. Drain and X. Gong, Chem. Commun., 2117 (1997).
23. G.A. Mirafzal, H.M. Bosse and J.M. Summer, Tetrahedron Lett., 40, 623 (1999).
24. D.V. Thomas and A.E. Martell, J. Amer. Chem. Soc., 78 (7), 1335 (1956).
25. A.D. Adler, E.R. Longo and W. Shergalis, J. Amer. Chem. Soc., 86 (15), 3145 (1964).
26. A. Treibs and H. Haberle, J. Liebigs Ann. Chem., 718, 183 (1968).
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 85
113. B. Evans, K.M. Smith and J.-H. Fuhrhop, Tetrahedron Lett., 5, 443 (1977).
114. K.M. Barkigia, M.D. Berber., J. Fajer, C.J. Medforth, M.W. Renner and K.M. Smith, J. Amer.
Chem. Soc., 112 (24), 8851 (1990).
115. O. Finikova, A. Cheprakov, I. Beletskaya and S. Vinogradov, Chem. Commun., 3, 261 (2001).
116. S. Ito, T. Murashima, H. Uno and N. Ono, Chem. Commun., 1661 (1998).
117. N.G. Kuzmin, A.S. Semeykin and O.I. Koifman, USSR Author’s Certificate 1574603, Byul.
Izobr., 24 (1990) (in Russian).
118. T.D. Lash and P. Chandrasekar, J. Am. Chem. Soc., 118, 8767 (1996).
119. C.J. Medforth, M.O. Senge, K.M. Smith, L.D. Sparks and J.A. Shelnutt, J. Amer. Chem. Soc.,
114 (25), 9859 (1992).
120. J. Takeda and M. Sato, Chem. Pharm. Bull., 42, 1005 (1994).
121. T. Ema, M.O. Senge, N.Y. Nelson, H. Ogoshi and K.M. Smith, Angew. Chem. Int. Ed. Engl.,
33 (18), 1879 (1994).
122. W. Jentzen, M.C. Simpson, J.D. Hobbs, X. Song, T. Ema, N.Y. Nelson, C.J. Medforth, K.M.
Smith, M. Veyrat, M. Mazzanti, R. Ramasseul, J.-C. Marchon, T. Takeuchi, W.A. Goddard III
and J.A. Shelnutt, J. Am. Chem. Soc., 117 (45), 11085 (1995).
123. M.O. Senge, L. Bischoff, N.Y. Nelson and K.M. Smith, J. Porphyrins Phthalocyanines, 3, 99
(1999).
124. J.A. Hodge, M.G. Hill and H.B. Gray, Inorg. Chem., 34 (4), 809–812 (1995).
125. T.D. Lash, K.A. Bladel, C.M. Shiner, D.L. Zajeski and R.P. Balasubramaniam, J. Org. Chem.,
57 (18), 4809 (1992).
126. J.A. Anton and P.A. Loach, J. Heterocycl. Chem., 12 (3), 573 (1975).
127. R.G. Little, J.A. Anton, P.A. Loach and J.A. Ibers, J. Heterocycl. Chem., 12 (2), 343 (1975).
128. J.S. Lindsey, S. Prathapan, T.E. Johnson and R.W. Wagner, Tetrahedron, 50, 8941 (1994).
129. S. Noblat, O. Dietrich-Buchecker and J.-P. Sauvage, Tetrahedron Lett., 28 (47), 5829 (1987).
130. I. Tabushi, K.-I. Sakai and K. Yamamura, Tetrahedron Lett., 21, 1821 (1978).
131. I. Tabushi, S.-I. Kugimiya, M.G. Kinnard and T. Sasaki, J. Amer. Chem. Soc., 107 (14), 4192
(1985).
132. J.P. Collman, D.A. Tyvoll, L.L. Ching and H.T. Fish, J. Org. Chem., 60 (7), 1926 (1995).
133. J.P. Collman, H.T. Fish, P.S. Wagenknecht, D.A. Tyvoll, L.-L. Ching, T.A. Eberspacher, J.I.
Brauman, J.W. Bacon and L.H. Pignolet, Inorg. Chem., 35, 6746 (1996).
134. K. Kohata, H. Higashio, Y. Yamaguchi, M. Koketsu and T. Odashima, Bull. Chem. Soc. Jpn.,
67, 668 (1994).
135. H. Meier, Y. Kebuks and S.-I. Kugimiga, J. Chem. Soc. Chem. Commun., 14, 923 (1989).
136. I. Tabushi and T. Sasaki, Tetrahedron Lett., 23 (18), 1913 (1982).
137. D. Hammel, C. Kautz and K. Müllen, Chem. Ber., 123 (6), 1353 (1990).
138. Z. Gross and I. Toledano, J. Org. Chem., 59, 8312 (1994).
139. E. Rose, M. Soleithavoup, L. Christ-Tommasino and G. Moreau, J. Org. Chem., 63, 2042
(1998).
140. U. Rempel, B. von Maltzan and C. von Borczyskowski, Chem. Phys. Lett., 245, 253 (1995).
141. U. Orth, H.-P. Pfeiffer and E. Bretimaier, Chem. Ber., 119 (11), 3507–3514 (1986).
142. H. Kamogawa, T. Nakata and S. Komatsu, Bull. Chem. Soc. Jpn., 64, 2300 (1991).
143. G. Casiraghi, M. Cornia, F. Zanardi, G. Rassu, E. Ragg and R. Bortolini, J. Org. Chem., 59,
1801 (1994).
144. M. Cornia, S. Binacchi, T. Del Soldato, F. Zanardi and G. Casiraghi, J. Org. Chem., 60 (16),
4964–4965.
145. G. Casiraghi, M. Cornia, G. Rassu, C. Del Sante and P. Spanu, Tetrahedron, 48, 5619 (1992).
146. D. Hammel, P. Erk, B. Schuler, J. Heinze and K. Müllen, Adv. Mater., 4, 737 (1992).
147. C.-H. Lee and J.S. Lindsey, Tetrahedron, 50 (39), 11427 (1994).
148. T. Mizutani, T. Ema, T. Tomita, Y. Kuroda and H. Ogoshi, J. Am. Chem. Soc., 116 (10), 4240
(1994).
149. N. Nishino, R.W. Wagner and J.S. Lindsey, J. Org. Chem., 61 (21), 7534 (1996).
150. J.P. Nagarkatti and K.R. Ashley, Synthesis, 186 (1974).
151. G. Shipps (Jr.) and J. Rebek (Jr.), Tetrahedron Lett., 35, 6823 (1994).
152. B. Vaz, R. Alvarez, M. Nieto, A.I. Paniello and A.R. DeLera, Tetrahedron Lett., 42 (42), 7409
(2001).
153. S.J. Vigmond, K.M.R. Kallury and M. Thompson, Anal. Chem., 64, 2763 (1992).
88 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
233. D. Heiler, G. Mc Lendon and P. Rogalskyj, J. Amer. Chem. Soc., 109 (2), 604 (1987).
234. A. Osuka and K. Maruyama, Chem. Lett., 825 (1987).
235. A. Osuka and K. Maruyama, J. Amer. Chem. Soc., 110 (13), 4454 (1988).
236. A. Osuka, K. Maruyama, I. Yamazaki and N. Tamai, J. Chem. Soc. Chem. Commun., 18, 1243
(1988).
237. A. Osuka, K. Ida and K. Maruyama, Chem. Lett., 5, 741 (1989).
238. A. Osuka, H. Tomita and K. Maruyama, Chem. Lett., 7, 1205 (1988).
239. J.L. Sessler and S. Pierind, Tetrahedron Lett., 28 (52), 6569 (1987).
240. J.L. Sessler and M.R. Johnson, Angew. Chem., 99 (7), 679 (1987).
241. P.S. Clezy, C.J.R. Fookes and A.J. Liepa, Austral. J. Chem., 25, 1979 (1972).
242. H. Fischer, P. Halbig and B. Walach, Justus Liebigs Ann. Chem., 452, 268 (1927).
243. G.M. Trofimenko, A.S. Semeykin, M.B. Berezin and B.D. Berezin, Zhurn. Koord. Khim.,
22 (6), 505 (1996) (in Russian).
244. D. Karris, A.W. Johnson and R. Caete-Holmes, Bioorg. Chem., 9, 63 (1980).
245. A.M. Shulga and G.P. Gurinovich, Dokl. Akad. Nauk BSSR, 25, 55 (1981) (in Russian).
246. D. Dolphin, A.W. Johnson, J. Zeng and P. van der Brock, J. Chem. Soc., 9, 880 (1966).
247. A.W. Johnson and I.T. Kay, J. Chem. Soc., 3, 1620 (1965).
248. J. Winkelman, Cancer Res., 22 (5), 589 (1962).
249. R.F. Pasternak, P.R. Huber, P. Boyd et al., J. Amer. Chem. Soc., 94 (13), 4511 (1972).
250. E.B. Fleisher, J.M. Palmer and A. Srivastava, J. Amer. Chem. Soc., 93 (13), 3162 (1971).
251. A. Srivastava and M. Tsutsui, J. Org. Chem., 38 (11), 2103 (1973).
252. C.A. Busby, R.K. Dinello and D. Dolphin, Can. J. Chem., 53 (11), 1554 (1975).
253. V.V. Bykova, N.V. Usoltseva, A.S. Semeykin et al., Zhurn. Org. Khim., 35 (4), 632 (1999)
(in Russian).
254. V.S. Radnyuk, T.V. Lyubimova and A.S. Semeykin, Abstracts of the XIX All-Russian
Chugaev Conf. on the Chemistry of Complex Compounds, Ivanovo, p. 171 (1999) (in
Russian).
255. P. Hoffman, G. Labat, A. Robert and B. Meunier, Tetrahedron Lett., 31, 1991 (1990).
256. D. Mandon, P. Ochsenbein, J. Fisher et al., Inorg. Chem., 31, 2044 (1992).
257. A.A. Fernandes, C.M. Stinson and A. Shamim, Pakistan J. Sci. Ind. Res., 30 (9), 643 (1987).
258. X. Chen, F. Tang and C. Wang, Phys. Test. Chem. Anal., 26 (2), 70 (1990).
259. F. Tang, X. Chen and C. Wang, Chem. Reagents, 9 (1), 29 (1987).
260. S. Tong and G. Sun, Chem. Reagents, 14 (1), 7 (1992).
261. W.A. Lee, M. Gratzel and K. Kalyanasundaram, Chem. Phys. Lett., 107 (3), 308 (1984).
262. W.J. Kruper, T.A. Chamberlin (Jr.) and M. Kochanny, J. Org. Chem., 54 (11), 2753 (1989).
263. A.M. d’A. Rocha Gonsalves, R.A.W. Johnstone, M.M. Pereira et al., Heterocycles, 43, 829
(1996).
264. V.V. Morozov, A.S. Semeykin, B.G. Gnedin and B.D. Berezin, Khim. Geterotsykl. Soed., 6,
770 (1988) (in Russian).
265. K.J. Ciuffi, H.C. Sacco, J.B. Valim et al., Non-Cryst. Solids, 247, 146 (1999).
266. M.S.M. Moreira, E.A. Vidato, O.R. Nascimento and Y. Iamamoto, ICPP-2, Kyoto, Japan, O-7,
p. 223 (2002).
267. L. Jaquinod, in: The Porphyrin Handbook, Acad. Press: New York, 1, Pt. 5, p. 202 (2000).
268. S.A. Syrbu and A.S. Semeykin, Abstracts of the I-st Int. Conf. on the Topical Problems of
Chemistry and Chemical Engineering, Ivanovo, p. 81 (1997).
269. V.I. Melnik, PhD (Chemistry) Thesis, Odessa State University: Odessa, 21 pp. (1979)
(in Russian).
270. A.S. Semeykin, O.I. Koifman, B.D. Berezin and S.A. Syrbu, Khim. Geterotsykl. Soed., 10,
1359 (1983) (in Russian).
271. L.R. Mildrom, J. Chem. Soc. Perkin Trans., 1 (10), 2535 (1983).
272. A.C. Chan, J. Dalton and L.R. Mildrom, J. Chem. Soc. Perkin Trans., 2 (6), 707 (1982).
273. J. Dalton and L.R. Mildrom, J. Chem. Soc. Chem. Commun., 14, 609 (1979).
274. E. Tsuchida, E. Hasegava, T. Komatsu et al., Chem. Lett., 3, 389 (1990).
275. S. Matile, T. Hansen, A. Storster and W.D. Wogjon, Helv. Chim Acta, 77 (4), 1087 (1994).
276. M. Momenteau, F. Le Bras and B. Looch, Tetrahedron Lett., 35 (20), 3289 (1994).
277. S.A. Syrbu, A.S. Semeykin and B.D. Berezin, USSR Author’s Certificate 1684284, Byul.
Izobr., 38 (1991) (in Russian).
meso-Phenylporphyrins as Synthetic Models of Natural Porphyrins 91
278. S.A. Syrbu and A.S. Semeykin, Zhurn. Org. Khim., 35 (8), 1262 (1999) (in Russian).
279. R.G. Little, J. Heterocycl. Chem., 15 (2), 203 (1978).
280. M. Momenteau and B. Loock, J. Mol. Catal., 7, 315 (1980).
281. A.S. Semeykin, O.I. Koifman, G.E. Nikitina and B.D. Berezin, Zhurn. Org. Khim., 54 (7), 1599
(1984) (in Russian).
282. B.D. Berezin, A.S. Semeykin, G.E. Nikitina et al., Zhurn. Fiz. Khim., 59 (9), 2226 (1985)
(in Russian).
283. S.A. Syrbu, A.S. Semeykin, B.D. Berezin and O.I. Koifman, Khim. Geterotsykl. Soed., 10,
1373 (1989) (in Russian).
284. Methods and Advances of Bioinorganic Chemistry, ed. by K. MacOliff, Mir: Moscow, 416 pp.
1978 (Russian translation).
285. L. Ding, G. Etemad-Moghadam, S. Cros et al., J. Med. Chem., 34 (3), 900 (1991).
286. P. Kus, G. Knerr and L. Gzuchajowski, Tetrahedron Lett., 31, 5133 (1990).
287. X. Jiang, P.K. Pandey and K.M. Smith, J. Chem. Soc. Perkin. Trans., 1, 1607 (1996).
288. S.A. Syrbu, A.S. Semeykin, B.D. Berezin and O.I. Koifman, Khim. Geterotsykl. Soed., 8, 781
(1987) (in Russian).
289. J.P. Collman, R.R.Gagne, T.R. Halbert et al., J. Amer. Chem. Soc., 95 (23), 7868 (1973).
290. A.S. Semeykin, O.I. Koifman and B.D. Berezin, Khim. Geterotsykl. Soed., 10, 1354 (1982)
(in Russian).
291. A.S. Semeykin, O.I. Koifman and B.D. Berezin, Izv. Vuz. Khim. Khim. Tekhnol., 28 (11), 47
(1985) (in Russian).
292. F. Tang, L. Wang and Z. Chai, Chem. Reagents, 15 (6), 324 (1993).
293. X. Wu, Z. Chen and Z. Ziang, J. Wunan Univ. Natur. Sci. Ed., 4, 30 (1993).
294. S.E. Gribkova, V.N. Luzgina and R.P. Evstigneeva, Zhurn. Org. Khim., 29 (4), 758 (1993)
(in Russian).
295. A. Palka and L. Czuchajowski, Chem. Lett., 3, 547 (1994).
296. A.S. Semeykin, O.I. Koifman and B.D. Berezin, Khim. Geterotsykl. Soed., 4, 486 (1986)
(in Russian).
297. A.S. Semeykin, O.I. Koifman and B.D. Berezin, Izv. Vuz. Khim. Khim. Tekhnol., 24 (5), 566
(1981) (in Russian).
298. S.A. Syrbu, A.S. Semeykin and B.D. Berezin, Khim. Geterotsykl. Soed., 11, 1507 (1990)
(in Russian).
299. S.A. Syrbu, A.S. Semeykin and T.V. Syrbu, Khim. Geterotsykl. Soed., 5, 668 (1996)
(in Russian).
92 A.S. Semeykin, S.A. Syrbu and O.I. Koifman
- ссылки 7 и 14 одинаковые
- упоминается табл. 6 (a таблиц всего 5)
This chapter presents the results of studying the catalytic activity of manganese(III)
and copper(II) porphyrins alkyl- and phenyl-substituted at β- and meso-macrocycle
positions in hydrogen peroxide decomposition in DMFA–KOH–H2O. The
ion-molecular mechanism of the process with kinetically significant stages of two-
electron metalloporphyrin oxidation and further partial reduction as well as acid-
base peroxide equilibrium reactions is stated. Effective catalysis of hydrogen per-
oxide decomposition is determined by the degree of porphyrin binding, mixed
coordination sphere structure and mutual influence of the ligands. The compounds
studied are shown to act similar to naturally occurring catalases.
Introduction
Studies of chelate complexes of copper as catalysts of the hydrogen peroxide decomposi-
tion reaction [1] showed the common factors for individual reaction mechanisms to be
94 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
coordination of peroxide particles on the metal atom and availability of two free coordina-
tion sites for the catalytic action of copper(II) compound to be realized. Copper(II) porphy-
rins (CuP), as polychelate macrocyclic complexes, are of interest in both aspects. First, they
contain a coordinatively unsaturated copper atom of the 3d 9 electron configuration in the
environment of the cyclic aromatic ligand; this copper atom forms direct N→Cu dative
π-bonds along with coordination σ-interaction. A similar π-interaction is also possible in
the case of addition of axial ligands containing unshared electron pairs. Second, CuP satisfy
the requirement of the two-side access of reagents to the coordination core. Prospectivity
of using metalloporphyrins with a two-charge metal cation (MP) for catalyzing the dispro-
portionation reaction of hydrogen peroxide is evident at present [2, 3]. However, the re-
quired coordinative unsaturation of metal in metalloporphyrins is the greater, the higher the
formal charge of the metal atom is [4, 5]. For this reason, studies of acidoporphyrin metal
complexes of the oxidation degree greater than two (X)n–2MP is also rather promising. A
suitable model in the given case are manganese(III) porphyrins. Thus, in [3], monomeric
and covalently bound dimeric manganese porphyrinates were found to exhibit a catalase ac-
tivity in the presence of the basic nitrogen of imidazole under conditions of phase-transfer
catalysis.
However, no rigorous theory has been developed to predict the reactivity of new co-
ordination compounds, porphyrin complexes of copper and manganese including, and to
explain the differences in their catalytic activity with respect to hydrogen peroxide decom-
position. Several attempts have only been made to study the effect of the ligand environ-
ment of the central ion on its catalytic activity and the H2O2 degradation mechanism [6, 7].
In this situation, of special significance is to obtain new information on the mechanism of
the catalytic action of complex compounds with regulated coordination saturation of the
central atom. The latter is realized in the case of metalloporphyrins by targeted functional
substitution of porphin in the complexes and rational choice of acidoligands X – in mixed
(X)n–2MP complexes.
The results obtained by the authors of this chapter in studies of copper(II) porphyrin
complexes with various degrees of substitution by β- and meso-positions of the macrocycle
(formulas 1–4) as catalysts of the hydrogen peroxide decomposition reaction in the
DMFA–KOH–H2O system (reaction 1) are considered in [8, 9]. This chapter presents the
major regularities and conclusions pertaining to the catalytic properties of copper porphy-
rins with respect to reaction (1). For manganese complexes, new data are presented on the
catalysis of the disproportionation reaction (1) of manganese(III) porphyrins with a regu-
larly changing structure (formulas 5–9) [10]. The catalytic reactions involving manga-
nese(III) tetraphenylporphyrins 10 and 11 were studied earlier [8, 9].
R R3 R
fast
fast
slow
H2O + O2
For the reaction catalyzed by Cu5,10DPOEP (compound 2), the spectral characteristics in
the course of conversion are similar to complexes 1 and 3. In reaction mixtures, where
CuTPP is present as a catalyst, attempts to record the electronic absorption spectra were un-
successful due to the vigorous evolution of oxygen. However, after CuTPP was isolated
from the reaction mixture to CHCl3 the electronic absorption spectra reflected the absorp-
I I
tion of initial CuTPP (λmax = 538.5 nm) and its π-cation-radical form (λmax = 700 nm and
II
λmax = 664 nm).
According to the spectral data, CuTPP and Cu5,10DPOEP (complexes 2 and 4) in the
absence of alkali do not enter into a chemical interaction with H2O2, unlike 1 and 3, as well
as in contrast with similar manganese(III) complexes considered below. If the reaction with
hydrogen peroxide is assumed to run to form axial complexes with the catalyst (Scheme 1),
as is the case in the functioning of natural catalase [1, p. 466], which is in complete agree-
ment with the experimental kinetic equations (2, 3), the above difference in reactivity can
be explained by different stabilities of the above axial complexes.
The catalytic activity of metalloporphyrins in the H2O2 decomposition reaction, the
rate of which, according to Scheme 1, is W = k4 ·K, is explained by the ability of copper
cation in the complex to coordinate ligands additionally to the fifth coordination site and
by low ionization potentials of the coordinated porphyrin macrocycle. Inertness of com-
plexes 2 and 4 in the reaction with H2O2 in the absence of KOH is due to their poor capa-
bility of axial coordination: a weak holding of OH· ligands in (·OH)CuP+· eliminated in the
course of reduction at the slow stage (Scheme 1) leads to a sharp decrease of the activation
energy in the case of CuTPP (27±3 kJ/mol) and CuDPOEP (17±2 kJ/mol) as compared
with other CuP, for which E changes within the limits of (39±3)–(73±7) kJ/mol [9].
the efficient charge of the central atom is not completely compensated for in coordination
and strong binding of the aromatic macrocyclic ligand. After elucidating the potential of
manganese complex compounds as catalysts of redox processes and understanding the role
of these compounds in medicine, agriculture and other fields, interest of researchers in them
permanently rises. Consideration of the role of manganese in enzyme as well as model sys-
tems simulating the functional properties of various enzymes [14] showed the significance
of compounds of this element in vital processes. Of especial importance are complexes of
manganese with porphyrins, as in their presence many reactions proceed under milder con-
ditions and in media, where simpler derivatives of manganese are inert or nonspecific [14].
Using complexes of manganese with various derivatives of porphin and various coordina-
tion centres, varying solvents and temperature, it is possible to achieve selective processes
of hydrogen peroxide decomposition, thus, modelling the action of catalases. The most sig-
nificant results in this direction were obtained in [15–20].
In the course of hydrogen peroxide decomposition catalyzed by metalloporphyrins and
metallophthalocyanines, the catalyst is observed to be partially (up to 30–50%) degraded
or its activity to be inhibited by an excess of bases added into the system, such as imidazole
[3, 21]. The data of the previous section suggest redox processes of the catalysis in the
course of the H2O2 decomposition reaction. For this reason, studies of the reactivity of por-
phyrin catalysts at the action of hydrogen peroxide on them are topical. In [22, 23] published
well back in 1960s, metallophthalocyanines have been shown to be subjected to oxidative
destruction under the action of H2O2 in an acidic medium. Pheophytin and its complexes
with metals are also oxidized but much slower. Since that time, conversion processes of
metalloporphyrins directly at the action of peroxide have not been studied systematically.
In studies of the kinetics of H2O2 decomposition in the presence of manganese(III) tet-
ra(2,6-dimethyl-3-sulfonatophenyl)porphin at pH 7.6–12.1, a conclusion was made of the
equilibrium coordination of the hydrogen peroxide molecule by the catalyst and the subse-
quent slow formation of the π-cation-radical form of oxomanganese(IV) porphyrin or ox-
omanganese(V) porphyrin [15]. Formation of oxo complexes (O)MnIVP, [(O)MnIVP]+·,
(O)MnVP, readily passing into one another, is also noted when using manganese
porphyrin–peroxide systems for oxidation and epoxidation of aliphatic and aromatic com-
pounds [16–20]. Thus, interaction of peroxyacetic acid with tetra(2,6-dichloro-4-R-phe-
nyl)porphyrin (R = CH3O, H, Br, Cl, NO2) complexes of manganese(III) in an
acetonitrile–acetic acid mixture leads to the formation of an intermediate
“catalyst–oxidant” complex [18]. In catalytic reactions of oxidation of cis-stilbene and
naphthalene, one observes an irreversible conversion of an intermediate complex
[(HOAc)MnRTDCPP](X) formed in the substitution of anion X by a molecule of HOAc in
the inner coordination sphere, into a mixture of two high-valent oxomanganese complexes.
The complexes are assumed to have the composition (X)(O)MnVRTDCPP and
[(O)MnIVRTDCPP]+· (X) [19]. The first complex is responsible for epoxidation of cis-stil-
bene; the second, for hydroxylation of naphthalene. In the medium of CH3CN, formation
of a stable complex is registered by the appearance of an absorption at 415 nm in the reac-
tion of (Cl)MnTDCPP with peroxyacetic acid, whereas in a medium of CH2Cl2 such a com-
plex proves to be short-lived and rapidly breaks down to destroy the macrocycle [20].
Conclusions on the conversions of catalysts in the cited works were mainly made
based on the analysis of the conversions of reagents or special compounds – traps – as in
[15], while the state of the catalyst itself – metalloporphyrin – was not studied. In this re-
view, we analyze the results of studies of the direct reaction of manganese porphyrins with
the oxidant H2O2. Using the spectrophotometric and kinetic methods, we studied the reac-
98 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
According to the spectrophotometry data [8, 9], the end products of the reaction of
complexes 10 and 11 with hydrogen peroxide are not the same depending on the peroxide
concentration ranges, which are 0.017–0.5, 0.67–1.0 and 1.26–3.98 mol/l. The products
are identified as an oxidized form with the localization of electron deficit at the aromatic
macrocycle – the π-cation-radical of manganese(III) tetraphenylporphin in the first interval
of H2O2 concentrations and the form oxidized at Mn atom – manganese(IV) tetraphe-
nylporphin – at higher concentrations of peroxide. In all cases, kinetic measurements re-
vealed the first order by initial metalloporphyrin. Experimentally found complete kinetic
equations for the H2O2 concentration range of 0.017–0.1 and 1.26–3.32 mol/l are written
down as equations (5) and (6):
−1/ 2 . (6)
− dC(Cl)Mn III TPP / dτ = k v3 ⋅ C(Cl)Mn IIITPP ⋅ CH O
2 2
Here kv1 and kv3 are the true rate constants of the reaction with hydrogen peroxide respec-
tively for the first and third concentration ranges of peroxide.
For the overall reaction described by kinetic equation (5), the following sequence of
elementary stages is proposed, with account for which the rate constant in equation (5) is
equal to kv1 = k2 ·K1:
K1
(Cl)MnIIITPP + H2O2 (Cl)(H2O2)MnIIITPP , (7)
(Cl)(H2O2)MnIIITPP ⎯⎯→
2 k
О=MnIII(TPP+ ?) + Н2О + Cl- .
. (8)
log I
a b
log CH2O2
CH2O2, mol/l
Figure 1 A titration curve (a) and a dependence of log I on log CH2O2 (b) (R2 = 0.98) at the titration
of the complex of (Cl)MnIIITPP in DMFA by hydrogen peroxide.
(Ap − A0 )/(A∞ − A0 ) 1
K = ⋅ , (9)
1 − (Ap − A0 )/( A∞ − A0 ) ( С H O − C (Cl)M nTPP ⋅ (Ap − A0 )/( A∞ − A0 ))
0
2 2
where A0, A∞, Ap are the optical densities of the solutions of initial (Cl)MnTPP, of the com-
plex (Cl)(H2O2)MnIIITPP and of an equilibrium mixture at the working wavelength of
468 nm.
The scheme of the reactions conforming to kinetic equation (6) is more complex due
to the appearance of perhydroxyl HO2– in kinetically significant amounts at the increase of
the concentration of hydrogen peroxide:
K1
−
H2O2 H+ + HO 2 , (10)
− k −
(Cl)MnIIITPP + HO 2 ⎯⎯
2
→ (Cl)(HO 2 )MnIIITPP fast , (11)
K3
−
(Cl)(HO 2 )MnIIITPP (Сl)O=MnIV(TPP
+•
(TPP+·) ) + ОН- , (12)
+• − k
(TPP+·) ) + HO 2 ⎯⎯→
(Сl)O=MnIV(TPP 4
O=MnIVTPP + HCl + O2 slow , (13)
K5
−
ОН- + H2O2 HO 2 + H2O . (14)
It has been experimentally found that, due to the low concentration of manganese por-
phyrin in the form of (Cl)O=MnIV(TPP+·), there is no gas evolution in the experiment, and
the O2 formed is, evidently, in solution. The rate constant in equation (6) includes the equi-
librium constants (10), (12) and (14):
100 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
k3 = k4 ⋅ K3 ⋅ K5 ⋅ K1−1/ 2 . (15)
The numerical value of K1 is known: 2.4·10 –2 at 298 K [25]. The values of kv1 and
kv3, found by optimizing the dependences of the efficient rate constant of the reaction of
(Cl)MnIIITPP with H2O2 on the concentration of the latter, are given in Table 1 [8].
Table 1 True rate constants, E and ∆S≠, of the reaction of (Cl)MnTPP with H2O2.
(Cl)MnTPP
0.017–0.10 288 1.41 70±1 –44±3
298 3.90
308 9.45
1.26–3.32 288 0.72 46±1 –133
298 1.34
308 2.51
(AcO)MnTPP
0.013–0.052 288 7 54±1 –87±3
298 15
308 31
(a)k
v1 and kv 3 for the CH2O2 ranges of 0.017–0.10 and 1.26–3.32, respectively.
The overall reaction of (Cl)MnTPP with H2O2 at all concentrations of the latter can be
presented as Scheme 2. At low concentrations of H2O2 the reaction is limited by the stage
of two-electron oxidation of manganese porphyrin and H2O2 reduction on metal in
(Cl)(H2O2)MnTPP (k2′); at high concentrations, by the stage of two-electron reduction of
π-cation radical (Cl)O=MnIV(TPP+·) up to O=MnIVTPP and Cl – and oxidation of HO–2
(k3′) with evolution of O2 and H+.
HO
Under conditions of H2O2 excess, the complex O=MnIVTPP oxidized by the central
atom of manganese is stable in solution in time.
It proved that the acetate complex (AcO)MnIIITPP (compound 11) reacts with H2O2
much faster than the chloride complex (Cl)MnIIITPP (Table 3). Already at CH2O2 equal to
1.0 mol/l the reaction is instantaneous, the reaction product in this case being O=MnIVTPP.
The order for [H2O2] for this reaction at low CH2O2 (0.013–0.052 mol/l) is close to unity
(0.72), which indicates the common character of its mechanism with that for (Cl)MnIIITPP
(Scheme 2, route K1′ → k2′). The rise of reactivity in the case of (AcO)MnIIITPP is explain-
able by a stronger binding of acidoligand AcO – as compared with Cl – . In the medium of
DMFA–H2O, where the reaction is run, in the initial complex at the coordination of H2O2
there occurs the substitution of the DMFA molecule coordinated to the sixth coordination
site. Owing to the effect of trans-influence of AcO – , substitution of DMFA is easier in the
(AcO)MnIIITPP complex than in (Cl)MnIIITPP. Besides, owing to the possibility of trans-
ferring the electronic effect of acidoligand via the transition metal with a partially filled
d-shell to the macrocycle, the macrocycle donates the electron easier to form π-cation rad-
ical in (AcO)MnIIITPP than in (Cl)MnIIITPP.
Manganese(III) porphyrins studied possess a catalase activity in an alkaline medium.
The results presented above show the ease of forming oxidized forms of complexes and the
absence of destruction of the macrocycle in the interaction with H2O2. This makes manga-
nese(III) porphyrins rather promising for their use as models of natural catalases.
New data on the spectral and kinetic studies of the reaction of (chloro)manganese(III)
octaethylporphin (Cl)MnOEP with hydrogen peroxide in water–DMFA media show that
what was said above regarding catalase models also pertains to the complex of manga-
nese(III) with another porphin derivative octaethylporphin.
The kinetics of the reaction of (Cl)MnOEP with H2O2 in an H2O–DMFA medium was
studied using a spectrophotometric method within the temperature range of 288–308 K.
The optical density of the solutions in the course of the reaction was measured at a wave-
length of 458 nm. The initial concentration of H2O2 in water was 17.4 ± 0.1 mol/l.
Interaction of (Cl)MnOEP with H2O2 at temperatures close to room temperature is ac-
companied with the change of its electronic absorption spectra. Herewith, the isobestic
points at 394, 446, 478, 523 and 582 nm are preserved (Fig. 2), which indicates the mutual
conversion of two stable stained compounds, one of which is the initial (Cl)MnOEP, the
form of existence of which differs at different concentrations of peroxide, as is seen from
the further discussion.
The intensity of the absorption bands of initial (Cl)MnOEP at 367, 458 and 541 nm
decreases, and at 434, 517 and 726 nm goes up. The end product was identified using liter-
ature data [26–28]. The spectrum of the end product corresponds to π-cation radical of man-
ganese(III) octaethylporphin (Fig. 2). The pattern of spectral changes presented in Fig. 2 is
characteristic of the entire range of H2O2 concentrations studied, which distinguishes the
octaethylporphin complex from above-considered (Cl)MnTPP, for which the similar reac-
tion was observed to yield various end products at different CH2O2. The first order of the
reaction of (Cl)MnOEP with H2O2 with respect to metalloporphyrin is demonstrated by the
data of Fig. 3. The efficient rate constants of the reaction are given in Table 2.
The order of the reaction with respect to [H2O2] is variable. Experimentally, we found
three ranges of peroxide concentrations: 0.01–0.07, 0.2–0.52 and 1.00–2.22 mol/l, each of
which was observed to have its own order for hydrogen peroxide (n) (Figs. 4 and 5): by the
T
slope of the lines in the coordinates log kef – log CH2O2, n is, respectively, 1/2, 0 and –1/2.
The zero order is evident due to the absence of the dependence of kef on CH2O2 (Table 2).
102 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
λ, nm
Figure 2 Electronic absorption spectra of (Cl)MnOEP in the course of the reaction with H2O2.
298 K, CH2O2 = 1.0 mol/l.
ln
τ, s
Figure 3 A dependence of ln(C0 /Cτ) – τ for the reaction of (Cl)MnOEP with H2O2. C(Cl)MnIIIOEP
= 2·10 –5 mol/l. CH2O2 = 0.33 (1), 0.49 (2), 0.05 (3), 1.35 (4), 0.03 (5), 0.01 (6) mol/l. T, K: 308 (1),
303 (2), 298 (3), 293 (4, 5, 6). R2 = 0.99.
The considered results make it possible to write down the experimentally found com-
plete kinetic equations for the first, second and third reacting sequences, respectively:
−1/2 . (18)
− dC(Cl)Mn IIIOEP / dτ = k v3 ⋅ C(Cl)Mn IIIOEP ⋅ CН О2 2
The Mechanism of Catalytic Action of the Coordination Centres of Catalase Synthetic Models 103
Table 2 Efficient kinetic parameters of the reaction of (Cl)MnOEP with H2O2 in DMFA.
log kTe
-log CH2O2
log kTe
log CH2O2
Figure 5 A dependence of log kefT on log CH2O2 for the reaction of (Cl)MnOEP with H2O2 within
the H2O2 concentration range of 1.00–2.22 mol/l. T, K: 308 (1), 303 (2), 298 (3), 293 (4). R2 = 0.98
(1, 3, 4), 0.92 (2).
104 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
The numerical values of the true rate constants kv1 and kv3, found by optimizing the
dependences presented in Figs. 4 and 5, and the reaction activation parameters are given in
Table 3.
Table 3 True rate constants, activation energy and entropy of the reaction of (Cl)MnOEP with
H2O2 in DMFA.
Considering the spectral data (Fig. 2) and taking into account equation (16), the fol-
lowing scheme of elementary stages can be proposed for the first range of H2O2 concentra-
tions. The scheme is supported by comparing experimental equation (16) with the
theoretical equation deduced for the system of equations (19–21).
K1
−
H2O2 Н+ + НО 2 , (19)
K2
− −
(Cl)MnIIIOEP + НО 2 (Cl)(НО 2 )MnIIIOEP , (20)
−
(Cl)(НО 2 )MnIIIOEP
k3
⎯⎯→
.
O=MnIII(OEP + • ) + ОН- + Cl- slow . (21)
CH2O2, mol/l
Figure 6 A curve of the titration of the complex (Cl)MnIIIOEP by hydrogen peroxide in DMFA.
acidoligand Cl remains to stay in the coordination sphere. This is confirmed by the example
of (X)MnTPP, where the rate of the process was found to strongly depend on the nature of
X (see above).
The rate equation for the limiting stage has the form:
After expressing C(Cl)HO–2)MnIIIOEP via the equilibrium constant K2 and then CHO–2 via
the equilibrium constant K1, we obtain equation (23):
The first equality in equation (23) is written with account for equilibrium (20). Thus,
from the comparison of equations (20) and (23), we have
k ν1 = k3 ⋅ K11/2 ⋅ K 2 , (24)
The third range of H2O2 concentrations with experimental kinetic equation (18) needs
the account for the following five reactions:
K1
−
H2O2 Н+ + НО 2 , (25)
− k2 −
(Cl)MnIIIOEP + НО 2 ⎯⎯→ (Cl)(НО 2 )MnIIIOEP fast , (26)
−
(Cl)(НО 2 )MnIIIOEP
K3 .
(Cl)O=MnIV(OEP + • ) + ОН- , (27)
K4 −
ОН- + Н2О2 НО 2 + Н2О , (28)
2(Cl)O=MnIV(OEP+ .) + HO2-⎯⎯→
k5 .
2O=MnIII(OEP + • ) + HCl + O2 + Cl- slow . (29)
Due to the low concentration of HO–2 , no gas evolution is observed in the experiment,
and O2 formed is, evidently, in solution.
The first stage involving manganese porphyrin (26) – coordination of hydrogen per-
oxide in the form of HO–2 to the sixth coordination position of the manganese atom – occurs
at higher concentrations of H2O2 rapidly and irreversibly. The transfer of two electrons
from the readily polarized porphyrin macrocycle and metal cation to the coordinated mol-
ecule of peroxide proceeds in the equilibrium stage (27). Probably, this reaction is made
equilibrium by the stability of π-cation radical of (Cl)O=MnIV(OEP+·) decreased as com-
pared with O=MnIII(OEP+·) (equation (21)), and the presence of higher concentrations of
negative ions HO–2 and OH –, which prevent the ionization of acidoligand Cl – . The latter
remains in the coordination sphere of π-cation radical (Scheme 3). Finally, the reaction with
the second H2O2 (in the form of HO–2) becomes possible; the reaction proceeds slower than
reaction (26) and limits the process. The reduction of π-cation radical of manganese(IV)
porphyrin does not proceed to the end, that is to the relatively stable O=MnIVOEP, but stops
at the one-electron reduced form O=MnIII(OEP+·), which is registered in the experiment as
a relatively stable end product. This is one more distinction from the (Cl)MnIIITPP–H2O2
system, where the end product is a two-electron reduced form.
We used the word “relatively” when assessing the stability. The cation-radical form
O=MnIII(OEP+·) upon isolation into chloroform after the reaction of (Cl)MnOEP with
H2O2 and washing off excess H2O2 still passes into the MnIV complex: bands at 394, 450
(arm), 514, 546 and 667 nm appeared in the spectrum, which is indicative of the formation
of O=MnIVOEP. In turn, this compound, if kept still in solution for 24 h, slowly passes into
manganese(III) porphyrin with a characteristic band in the region of 473 nm (the fact known
in the chemistry of manganese porphyrins [26]).
The rate equation for the limiting stage has the form:
−1
C(Cl)O= Mn IV (OEP +•. ) = K3 ⋅ C(Сl)(HO− )Mn IIIOEP ⋅ CОН −, (31)
2
2 −1
CОН − = СHO − ⋅ K4 , (32)
2
−2
C(Cl)O= Mn IV (OEP +•.) = K3 ⋅ C(Сl)(HO− )Mn IIIOEP ⋅ K4 ⋅ CHO − . (33)
2 2
Upon substitution of the expressions for concentrations, (33) and (34), into equation
(30), we obtain equation (35), which is consistent with the found experimental equation
(18):
Thus,
k ν3 = k5 ⋅ K3 ⋅ K 4 ⋅ K11/ 2 , (36)
tetraphenylporphin – the form oxidized at Mn atom. In the case of (Cl)MnOEP, the end
product of the reaction is π-cation radical of manganese(III) octaethylporphin within the en-
tire range of H2O2 concentrations. Hence, it follows that, by changing the electronic state
of the macrocycle and of Mn–N bonds, we can achieve optimal parameters for the coordi-
nation reaction of H2O2 molecules and HO2– ions at various stages of the reaction with per-
oxide and then use them in the development of synthetic catalases.
ml
τ, min
Figure 7 A dependence of the volume of evolving O2 on time in the course of the H2O2 decompo-
sition reaction. C(Cl)MnMPOEP = 1.08·10 –5 (1), C(Cl)MnOEP = 2.77·10 –5 (2), C(AcO)MnOEP =
2.25·10 –5 (3), C(Cl)Mn5,15DPOEP = 3.87·10 –5 mol/l (4), CH2O2 = 3.49 mol/l, CKOH = 1.5·10 –2 mol/l.
T = 343 K.
The Mechanism of Catalytic Action of the Coordination Centres of Catalase Synthetic Models 109
Table 4 Decomposition rates of H2O2 (W) and catalytic activity (A) of metalloporphyrins at 343 K.
Table 4 Decomposition
(continued) rates of H2O2 (W) and catalytic activity (A) of metalloporphyrins at 343 K.
– – 0.62±0.02 –
(Cl)MnOEP 0.095 0.76±0.01 47±1
0.19 0.83±0.02 26±1
0.38 0.84±0.06 13±0.6
2.77 1.58±0.06 3.3±0.1
5.11 2.10±0.06 2.4±0.1
(AcO)MnOEP 0.07 0.74±0.01 63±1
0.14 0.74±0.04 31±2
0.35 0.73±0.03 12±0.5
2.25 1.29±0.04 3.4±0.1
(SCN)MnOEP 0.09 0.73±0.03 48±2
0.18 0.73±0.02 24±0.6
0.32 0.75±0.03 14±0.5
0.45 0.78±0.05 10.3±0.6
2.71 1.48±0.05 3.2±0.1
(Cl)MnMPOEP 0.45 0.97±0.03 12.7±0.4
0.72 1.45±0.04 12.0±0.3
1.08 1.86±0.06 10.2±0.3
(Cl)Mn 5,15DPOEP 0.097 0.76±0.02 46±1
0.48 0.73±0.03 9.0±0.4
0.97 0.76±0.02 4.6±0.1
3.87 1.05±0.02 1.61±0.03
8.53 1.52±0.03 1.06±0.02
MnCl2 0.60 1.26±0.04 12.4±0.4
1.43 1.3±0.1 5.5±0.5
5.95 1.4±0.3 1.5±0.3
11.90 1.7±0.2 0.8±0.1
(a)C –2 –5
KOH = 1.5·10 mol/l, C(X)MnP = 0.2·10 mol/l,
(b)C –5
H2O2 = 3.49 mol/l, C(X)MnP = 0.2·10 mol/l,
(c)CH2O2 = 3.49 mol/l, CKOH = 1.5·10 –2 mol/l.
Table 5 Catalytic H2O2 decomposition reaction rates at various temperatures and activation
parameters. C(X)MnP = 0.2·10 –5 mol/l, CH2O2 = 3.49 mol/l, CKOH = 1.5·10 –2 mol/l.
E ± ∆E
∆S ≠ = 19.1 ⋅ log W + − 19.1 ⋅ log T − 205 . (37)
T
It is seen from Table 4 that the decomposition rate of hydrogen peroxide increases with
the rise of the concentrations of the manganese(III) complex, KOH and H2O2 in the system
on condition of a significant excess of the other reagents as compared with the complex.
There is a satisfactory correlation between W and the catalyst concentration, which
points to the first order of the reaction with respect to this component (Fig. 8). A similar
dependence for CH2O2 is not linear. Figure 9 demonstrates a correlation in the logarithmic
coordinates log W – log CH2O2 with the slopes of the curves close to 2 (n = 2.01 for
(Cl)MnOEP, n = 2.26 for (Cl)Mn5,15DPOEP). A similar dependence for COH – is also linear,
with the slope α close to unity (Fig. 10). Thus, according to the data of the experiment, we
can write down the equation for the reaction rate:
W, ml O2 / min
logW logW
logCKOH
logCH2O2
Fig. 9 Fig. 10
Figure 9 A dependence of log W on log CH2O2 for the reaction of H2O2 decomposition in the pres-
ence of (Cl)MnOEP (1) and (Cl)Mn5,15DPOEP (2). C(X)MnP = 0.2·10 –5 mol/l, CH2O2 = 3.49 mol/l,
CKOH = 1.5·10 –2 mol/l, T = 343 K. R2 = 0.99.
Figure 10 A dependence of the logarithms of the rates of H2O2 decomposition in the presence of
(Cl)MnOEP (1) and (Cl)Mn5,15DPOEP (2) on the logarithms of KOH concentration. C(X)MnP =
0.2·10 –5 mol/l, CH2O2 = 3.49 mol/l, CKOH = 1.5·10 –2 mol/l, T = 343 K. R2 = 0.99.
2
dCO2 / dτ = k ⋅ Ccatalyst ⋅ СН 2О 2
⋅ COH - ⋅ CO0 . (38)
2
The numerical values of the true rate constants k are calculated as arithmetic means of
three true constants obtained in optimizing the dependences in coordinates W–Ccatalyst,
log W–log CKOH and log W–log CH2O2 at integer values n, respectively, 1, 1 and 2. The con-
stants k for (Cl)MnOEP, (AcO)MnOEP and (Cl)Mn5,15DPOEP are, respectively,
(1.73±0.06), (1.64±0.17) and (1.49±0.07) l4 ·mol –3 ·s –1.
A special spectrophotometric study of the interaction of (Cl)MnOEP with H2O2 in a
DMFA–H2O medium in the absence of alkali, the results of which are considered in the pre-
vious section, made it possible to reveal in the entire range of H2O2 studied (0.01–3.74
mol/l) a manganese(III) porphyrin → π-cation radical of manganese(III) porphyrin transi-
tion (Fig. 2).
With account for the data considered, we can write down the assumed scheme of con-
versions in the course of the catalytic reaction of H2O2 decomposition (equations
(39)–(43)).
K1
(Х)MnIIIOEP + H2O2 (Х)(H2O2)MnIIIOEP , (39)
(Х)(H2O2)MnIIIOEP
k2
⎯⎯→
.
O=MnIII(OEP + • ) + H2O + Х- fast , (40)
K3 −
H2O2 + OH- HO 2 + H2O, (41)
.
O=MnIII(OEP + • ) + HO 2−
K4
(HO 2 )O=MnIII(OEP + • ),
− . (42)
(HO 2 )O=MnIII(OEP + • )
− . k5
⎯⎯→ [MnIIIOEP]+ + H2O + O2 slow. (43)
The Mechanism of Catalytic Action of the Coordination Centres of Catalase Synthetic Models 113
For the limiting stage (43), we can write down the kinetic equation:
−dC(HO- )O=Mn III (OEP+•.) / dτ = dCO2 / dτ = k5 ⋅ C(HO- )O=Mn III (OEP+•.) . (44)
2 2
We write down equalities (46) –(48) for reactions (41), (40) and (39), respectively:
Let us perform the substitution of the expressions for CHO-2 and CO=MnIII(OEP+·) also
into equation (45):
2
C(HO- )O=Mn III (OEP+•.) = K 4 ⋅ K1 ⋅ K3 ⋅ C(Х)Mn III OEP ⋅ CH 2 O2
⋅ COH - (49)
2
After substituting the expression for C(HO-2 )O =MnIII(OEP+·) (equation (49)) into (44), we
obtain equation (50), which is consistent with experimentally found equation (38):
2
dCO2 / dτ = k5 ⋅ K 4 ⋅ K1 ⋅ K3 ⋅ C(Х)Mn IIIOEP ⋅ CH 2 O2
⋅ COH - . (50)
From the comparison of equations (38) and (50), it follows that k = K1 · K3· K4· k5.
In accordance with equations (39)–(43), the catalytic reaction of H2O2 decomposition
proceeds with the participation of the catalyst, which changes its redox states, and two per-
oxide units (in the form of the H2O2 molecule and perhydroxyl anion HO2– ). At the first
stage, H2O2 is coordinated to the sixth coordination site (X)MnIIIOEP; as the result, the
O–O bond, which is not too strong as it is (200 kJ/mol), is activated [25]). The coordinated
molecule of H2O2 breaks down to evolve H2O and to eliminate two electrons from metal-
loporphyrin. The latter passes into a π-cation radical form, being oxidized via the macro-
cycle and metal, which remains triple-charged only formally (O=MnIIIOEP+·) R
[O=MnIVOEP]+), as, for instance, in the known stable oxidized forms of diphthalocyanines
of lanthanides LnIIIPc2· [29]. The second molecule of peroxide is added as a more nucleo-
philic particle HO–2. The formed π-cation radical of the complex, being in a highly coordi-
nated state, slowly reduces with abstraction of molecular O2. It should be emphasized that
no manifestations of the radical mechanism of H2O2 decomposition have been revealed in
support of the considered mechanism, in particular, no slow stages of reaction initiation.
Variation of the ligand, which is in the fifth coordination position, with the view to
confirm the provisions of the mechanism of the catalytic action of manganese(III) porphy-
rins, gave the following results. The catalytic activity of manganese(III) octaethylporphins
(Table 6) is essentially not dependent on the nature of acidoligand (Cl, AcO, SCN), which
114 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
3 Conclusion
The catalytic-action mechanism of manganese(III) porphyrins expounded above repeats in
many ways the mechanism of action of natural catalases. The general regularities of con-
versions involving the enzymes hemoproteins have been discussed already in the mono-
graph [25]. The enzyme catalase, which has a prosthetic group of iron protoporphyrin, is
bound to the protein by means of one coordination link (one coordination site of Fe) and
via propionic acid residues. H2O2 in the amount of several micromoles per litre is either
decomposed or reacts with other particles. With iron protoporphyrin, hydrogen peroxide
forms, apparently owing to coordination to the sixth coordination position, complexes of
three types: Fe·H2O2, Fe·OOH and Fe·OOH –, the last of which is active as catalase. This
complex further reacts with another molecule of H2O2 or other active particles. Thus, the
expounded conclusions regarding the catalytic-action mechanism of manganese(III) por-
phyrins obtained in our work basically correspond to the data for the mechanism of natural
catalases. Complexes of natural catalases and peroxidases have not been isolated individu-
ally due to their instability. The use of simpler models, as shown by our study, makes it pos-
sible to describe intermediate complexes spectrally. Besides, a possibility appears to
The Mechanism of Catalytic Action of the Coordination Centres of Catalase Synthetic Models 115
regulate and, which is more important, increase the catalytic activity of catalases by modi-
fying the aromatic moiety of their molecules.
The work was supported by grants from the Programme for Basic Research of the Rus-
sian Academy of Sciences “Elaboration of Methods for Producing Chemical Substances
and Development of Novel Materials” (No 8, 2006); from the Analytical Departmental Tar-
get-oriented Programme “Development of Higher-School Scientific Potential”
(2006–2008). Activity 2: Scientific Methodological Provision of the Development of High-
er-school Science Infrastructure”; from RNP 2.2.1.1.7181 “Development of the Mecha-
nisms of Integrating Ivanovo State University and Institute of Problems of Chemical
Physics RAS”; and from the Russian Foundation for Basic Research 06-03-96343-r.
References
1. G. L. Eichorn (ed.), Inorganic Biochemistry, Elsevier: New York, vol. 2 (1973).
2. I. Naruta, M. Sasayama and K. Ishichara, Zhurn. Org. Khim., 32 (2), 233 (1996) (in Russian).
3. O.V. Cheremenskaya, A.B. Solovyeva, G.V. Ponomarev and S.F. Timashev, Zhurn. Fiz. Khim.,
75 (10), 1787 (2001) (in Russian).
4. M.Yu. Tipugina, T.N. Lomova and T.A. Ageyeva, Zhurn. Obshch. Khim., 69 (3), 459 (1999)
(in Russian).
5. M.Yu. Tipugina and T.N. Lomova, Zhurn. Neorg. Khim., 47 (7), 1085 (2002) (in Russian).
6. M.R. Tarasevich and K.A. Radyushkina, Catalysis and Electrocatalysis by Metalloporphyrins,
Nauka: Moscow, p. 168 (1982) (in Russian).
7. A.Ya. Sychev and V.G. Isak, Coordination Compounds of Manganese in Catalysis, Shtiinca:
Kishinev, p. 321 (1990) (in Russian).
8. M.E. Klyueva, T.N. Lomova and M.V. Klyuev, in: Peroxides at the Beginning of the Third
Millennium. Synthesis, Properties and Application, ed. by V.L. Antonovsky, O.T. Kasaikina and
G.E. Zaikov, Nova Science: New York, pp. 143–166 (2004).
9. T.N. Lomova, M.V. Klyuev, M.E. Klyueva, E.N. Kiseleva and O.V. Kosareva, Ross. Khim.
Zhurn., XLVIII (4), 35–51 (2004) (in Russian).
10. E.N. Kiseleva, M.E. Klyueva and T.N. Lomova, Abstracts of the VIII-th Scientific School-
Conference on Organic Chemistry, p. 139, Kazan’, 22–26 June (2005) (in Russian).
11. A.D. Adler, F.R. Longo and J.D. Finarelli, J. Org. Chem., 32, 476 (1967).
12. N.S. Dudkina, P.A. Shatunov, E.M. Kuvshinova, S.G. Pukhovskaya, A.S. Semeykin and
O.A. Golubchikov, Zhurn. Obshch. Khim., 68 (12), 2042 (1998) (in Russian).
13. A.D. Adler, F.R. Longo, F. Kampas and J. Kim, J. Inorg. Nucl. Chem., 32, 2443 (1970).
14. A.Ya. Sychev and V.G. Isak, Coordination Compounds of Manganese in Catalysis, Shtiinca:
Kishinev, p. 321 (1990) (in Russian).
15. P.N. Balasubramanian, E.S. Schmidt and T.C. Bruce, J. Am. Chem. Soc., 109 (25), 7865 (1987).
16. R.D. Arasasingham, G.-X. He and T.C. Bruce, J. Am. Chem. Soc., 115, 7985 (1993).
17. J.T. Groves and M.K. Stern, J. Am. Chem. Soc., 109 (12), 3812 (1987).
18. S. Banfi, M. Cavazzini, G. Pozzi, S.V. Barkanova and O.L. Kaliya, J. Chem. Soc. Perkin Trans.,
II, 871 (2000).
19. S. Banfi, M. Cavazzini, G. Pozzi, S.V. Barkanova and O.L. Kaliya, J. Chem. Soc. Perkin Trans.,
II, 879 (2000).
20. S. Banfi, M. Cavazzini, G. Pozzi, S.V. Barkanova and O.L. Kaliya, J. Chem. Soc. Perkin Trans.,
II, 1577 (1997).
21. O.A. Golubchikov and B.D. Berezin, Usp. Khim., 55 (8), 1361 (1986) (in Russian).
22. B.D. Berezin and G.V. Sennikova, Kinet. Kataliz, 9 (3), 528 (1968) (in Russian).
23. B.D. Berezin and G.V. Sennikova, Zhurn. Fiz. Khim., 43 (10), 2499 (1969) (in Russian).
24. E.N. Kiseleva, M.E. Klyueva and T.N. Lomova, Abstracts of the IX-th Int. Conf. on the
Chemistry of Porphyrins and their Analogues, Ivanovo, p. 106 (2003).
25. W.C. Schumb, C.N. Satterfield and R.L. Wentworth, Hydrogen Peroxide, ACS Monograph,
Reinhold Publishing Corp.: New York, 757 pp. (1955).
26. H. Volz and W. Müller, Chem. Ber. Recueil, 130, 1099 (1997).
27. L. Kaustov, M.E. Tal, A.I. Shames and Z. Gross, Inorg. Chem., 36 (16), 3503 (1997).
116 T.N. Lomova, M.E. Klyueva and M.V. Klyuev
28. N. Carnieri, A. Harriman, G. Porter and K. Kalyanasundaram, J. Chem. Soc. Dalton Trans., 7,
1231 (1982).
29. P.I. Moskalev, USSR Author’s Certificate 525318, Byul. Izobr. 4 (1978) (in Russian).
30. O.A. Golubchikov, S.G. Pukhovskaya and E.M. Kuvshinova, Advances of Porphyrin Chemistry,
vol. 4, ed. by O.A. Golubchikov, Chemistry Research Institute, St.-Petersburg University:
St.-Petersburg, p. 45 (in Russian).
31. M.E. Klyueva, T.N. Lomova, E.E. Suslova, and A.S. Semeykin, Teor. Eksp. Khimiya, 39 (5), 299
(2003) (in Russian).
4 Complexation of Porphyrins
with Ions and Organic
Molecules
Introduction
At present, the complex-forming properties of porphyrins are understood to be both the abil-
ity of porphyrins to enter into complexation reactions with metal cations to form metal-
loporphyrins and the ability of porphyrins to form supramolecular complexes. While
metalloporphyrin formation processes have been fruitfully studied for many decades, works
on supramolecular complexes of porphyrins having no less than two binding points, ap-
peared in the literature only in the recent 10 to 15 years. Presented material generalizes and
118 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
analyzes the literature data and own results by the authors on supramolecular complexes of
porphyrins, showing the main trends of the development of the modern supramolecular
chemistry of porphyrins.
Table 1 Rate constants (k) and activation parameters (E and ∆S≠ ) of the complexation of
copper(II) cation with chlorophyllic acid in organic solvents at 298 K [1].
Solvent k, E, ∆S ≠,
l/(mol·s) kJ/mol J/(mol·K)
under the action of an electric field of the cation pass into solution and bind with solvent
molecules. Therefore, both the rate of formation of metalloporphyrin and activation param-
eters of the process depend on the extent of change of the N-H bond covalence and electron
density at atoms of nitrogen of the porphyrin coordination core under the influence of sub-
stituents in the molecule. Table 2 shows how great that influence is. The formation rates of
metal complexes, depending on the structure of porphyrin, change hundreds of times; and
activation energies, from 41.2 up to 115 kJ/mol.
Table 2 Rate constants and activation parameters of the complexation reaction of porphyrins with
copper(II) cation in ethanol at 298 K [1].
Porphyrin k, E, ∆S ≠,
l/(mol·s) kJ/mol J/(mol·K)
It should also be noted that, as compared with complex-formation processes with open
(not macrocyclic) ligands, the formation rates of metalloporphyrins are extremely low. This
is due to the rigid planar structure of the ligand, to the shielding of the coordination core by
the adjacent atoms and π-electronic cloud. This spatial shielding of the reaction site is typ-
ical of the formation processes of rigid aromatic macrocycles and forms the basis of the
macrocyclic effect. It is the rigidity of the ligand, determined by the aromaticity of porphy-
rin, that does not enable it to take on the conformation suitable for rapid coordination. In
the course of interaction, it is not the ligand that has to adjust to the coordination sphere of
the salt, but, the other way round, the salt to the ligand, and the reagent is primarily the salt.
Upon introduction of an alkyl group, the porphyrin molecule is strongly polarized,
which contributes to the solvation of the molecule as a whole. As the result, the transition
state of meso-alkyl-substituted porphyrin becomes more probable [1]. The coordination rate
of phylloporphyrin 1 by Cu(II) cations in ethanol is three times as large as that of pyrropor-
phyrin 2. meso-Ethyl substitution of octaethylporphyrin (compound 3) increases the rate of
the reaction with zinc cation in acetonitrile 6.6-fold (Table 3). Introduction of a hexyl group
into hexamethyldiethylporphyrin 4 increases the rate of coordination of the molecule by
copper(II) and zinc(II) cations in propanol by approximately an order [2]. That is, with the
length of alkyl in meso-position (Alk = CH 3, C2H5, C6H13) increasing, the rate of the pro-
cess rises 3- to 10-fold.
120 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Table 3 Kinetic parameters of the coordination reaction of octalkylporphyrins (1–6) with various
substituents in meso-position, copper and zinc cations in alcohols at 298 K [3].
(1) k
1.5 = kef / cCu (OAc)
2
(1)
R4 R5
R3
1 : R1 =CH2CH2COOH, R2= R3= R5 = R7 = R9 = Me, R4= R6=Et,
R6
R8=H,
NH N
2 : R1 =CH2CH2COOH, R2= R3= R5 = R7 = Me, R4= R6=Et,
R8=R9 =H
3 : R1 = R2= R3= R4 = R5= R6= R7 = R8= R9=Et
N HN
4 : R1 = R2= R3= R6 = R 7 =R 8 =Me, R4 = R5=Et, R9 =C 6H13
R2
R7
5 : R1 = R4 = R5 =R 8 = Me, R2= R3 =R 6=R 7 =Et, R9 =C6H13
1-6
Complexation of Porphyrins with Ions and Organic Molecules 121
R R 7 : R = Bu, R' = H
8 : R = Bu, R' = Ph
H3C
CH3
NH N 9 : R = Bu, R' = PhOMe-p
R R 13 : R = Me, R' = Et
Table 7 Kinetic parameters of the complexation reaction of 5,15-diarylporphyrins (7, 8, 10, 11)
with copper(II) cation in ethanol at 298 K: cCu(OAc)2 = 1.211·10–3 mol/l; cºH2P = 4.520·10–5 mol/l.
No meso-Substituent k1.5 ·102, E, ∆S ≠,
l0.5 ·mol–0.5 ·s–1 kJ/mol J/(mol·K)
The electron-acceptor substituent (para-NO2, 12) slightly slows down the reaction.
Against the background of pronounced effects of benzene fragments’ substituents in por-
phyrins 7--12, the role of solvation effects increases. Thus, in pyridine (Table 6), diphe-
nylporphyrin 8 and its ortho-methoxy derivative 11 have almost the same reactivity
(α,α-atropisomer was studied, in which both methoxy groups are on one side of the por-
phyrine macrocycle), whereas in acetonitrile (Table 5) the derivative 11 reacts with cop-
per(II) cation two times as slow as the initial compound 8.
In passing from meso-unsubstituted porphyrin 7 to dialkylporphyrins 13, 14, the reac-
tivity of the macrocycle increases (Tables 8, 9). The rate constant in ethanol changes from
219·10 –2 l/mol·s (compound 7) to 271·10 –2 l/mol·s (compound 13). Probably, alkyl sub-
stituents, possessing a positive induction effect, increase the electron density at tertiary at-
oms of nitrogen and facilitate the interaction of metal cation with porphyrin in transition
state. A decrease of the activation energy (~ by 7.0 kJ/(mol·K)) in meso-alkyl substitution
of porphyrin in ethanol is due to an increase of solvation of the salt--porphyrin system’s
transition state during the emergence of alkyl substituents in the molecule. It is known that
introduction of two ethyl groups in meso-position of octamethylporphyrin causes an in-
crease of solubility of the molecule in benzene by two orders of magnitude. The effect of
substituents is more pronounced in acetic acid (Tables 4, 9), which specifically solvates the
reaction centre of porphyrin and decreases its reactivity in the interaction with copper(II)
cation. The strength of the hydrogen bonds between nitrogen atoms of porphyrin and hy-
drogen atoms of acid is reduced at the introduction of acceptor substituents into 8. This
causes an increase of the reaction rate (Table 4). At 298 K, the rate constant increases from
5.1·10 –2 (8) to 11·10 –2 l/(mol·s) (para-nitroderivative 12). As compared with diphenylpor-
phyrin, the activation energy of the nitroderivative decreases by ~20 kJ/mol; and activation
entropy, by 67 J/(mol·K). Introduction of electron-donor alkyl groups leads to a decrease
of reactivity of the macrocycle (Table 9).
Complexation of Porphyrins with Ions and Organic Molecules 123
Table 8 Kinetic parameters of the complexation reaction of porphyrins (7,13, 14) with copper(II)
and zinc(II) cations in ethanol at 298 K.
The rate constant in the case of zinc complexes of porphyrins is of the second order (kv, l/mol · s).
Table 9 Kinetic parameters of the complexation reaction of 5,15-diarylporphyrins (7, 13, 14) with
copper(II) cation in acetic acid at 298 K: cCu(OAc)2 = 2.410·10–3 mol/l; cºH2P = 6.721·10–5 mol/l.
O
NH N
NH
NH
N NH
R
R anion
R
R
X
15, 16
O
NH N
15 : X = H , 16 : X=
NH
NH
N NH
P
P P
NH
P
NH P P
O NH
NH NH
O O
NH O NH
NH O
NH O
NH NH
OPrOPr NH
OPr OPr
OPr OPr OPr OPr
OPr OPr OPr OPr
17 18 19
N
P= NH
HN
N
porphyrin fragments possess a sufficient flexibility for placing ions I – and Br – , irrespective
of their location [cis- or trans-], on the upper rim of calix[4]arene. What is more, the pro-
poxy groups located between porphyrin fragments of 1,3-alternate do not prevent the com-
polexation with anion. At the same time, significant differences in the binding constants for
larger anions (I – and NO3– ) point to certain spatial restrictions for complexation. With the
diameter of anion increasing, the binding constant of anions decreases [for porphy-
rin-calix[4]arene 19 KCl (6.9·105 l/mol) > KBr (6.9·104 l/mol) > KI (2.4·103 l/mol)]. Here-
with, compound 19 exhibits a better complexing ability with respect to Cl – as compared
Complexation of Porphyrins with Ions and Organic Molecules 125
with the system having a similar structure of the calix[4]arene platform, but instead of tet-
rapyrrole fragments containing phenyl groups (KCl 4.6·103 l/mol) [20].
Anion 17 18 19
no more than fourfold. The largest values of Ka are observed in aromatic solvents of the
type of benzene and toluene.
The effect of alkyl and aryl substitution on the binding constants in porphyrins 20--31
is levelled out even more (no more than twofold), as the electronic influence of phenyl frag-
ments and alkyl groups on the state of the central zinc ion is insignificant. At the same time,
in the sequence of metalloporphyrins 32--49 with different deformation degrees of the tet-
rapyrrole macrocycle the binding constant of monodentate basic nitrogens changes, de-
pending on the nature of this macrocycle, by several orders (Table 12).
The main factors determining the axial coordination process in these porphyrins are:
a) schielding of the reaction centre of the tetrapyrrole macrocycle (on one or both sides); b)
change of electron density on nitrogen atoms and of charge on Zn(II) cation as the result of
R8 R1
R
R7 R2
N N
Zn
N N
N N R6 R3
Zn
N N
R R R5 R4
22, 23
22 : R1=R2=R3=R4=R5=R6=R7=R8=H
R 23 : R1=R3=R5=R7=CH3, R2=R6=C2H5,
20, 21 R4=R8=(CH2)2COOCH3
20 : R = H, 21 : R = CO2CH3-m
R1 R1 R
R
R2 R2 N N
N N
Zn
Zn
N N
N N
R1 R1 CH3OOCCH2CH2 CH2CH2COOCH3
24-28 29-31
deformation of the macrocycle; c) correspondence of the size of the cavity formed by ster-
ically hindered porphyrin macrocycle to the size of extra ligand; d) emergence of additional
interactions of the type of π−π, CH−π and others between metalloporphyrin (host molecule)
and ligand (guest molecule).
Coordination of axial ligands on picket-fence, strapped and dimeric porphyrins has its
own features. The term “picket-fence” is used for porphyrins with peripheral bulky substit-
uents. In most cases, these are derivatives of tetraphenylporphin containing an ortho-sub-
stituent in each benzene nucleus. Depending on the orientation of substituents relative to
the plane of the tetrapyrrole macrocycle (α, on one side; β, on the other side), four atropi-
somers are distinguished: α,α,α,α- (α4), α,β,α,α-, α,α,β,β- (cis-α2) and α,β,α,β
(trans-α2).
Works by Imai and Kyuno investigated the axial coordination on picket-fence metallo-
porphyrins (32, 33) of pyridine, piperidine and quinoline in toluene and other noncoordi-
Complexation of Porphyrins with Ions and Organic Molecules 129
O t-Bu t-Bu O
NH O t-Bu
t-Bu NH
NH
O N
O t-Bu NH N
N Zn
N
N N
NH N Zn O t-Bu NH
H
N
NH O
R
NH
R O
33
32
Et O R1
Et R1 HO HO
O
N N OH
Zn
N N OH
O OH
Et
OH Et O OH
O
O
O O O R1
34
R1 O O
OH
X
O
Bu Bu
O
N N O RO RO O
Zn Et Et
N N O O
N N
Zn
Bu N N
Bu
Et Et
35, 36
37,38
35 : X =
OCH3 37 : R = COCF3 38 : R = H
36 : X = CH2 CH2
OCH3
nating solvents [46, 49]. The results obtained showed trans-α2 atropisomers to have a high-
er affinity to pyridine and quinoline than α2 atropisomers.
As in the case of trans-α2 atropisomers both sides of the reaction centre of porphyrin
are equally accessible for coordination, an increase of affinity to pyridine and quinoline is
130 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
O O
N N
O
O Et O O Et
O O O
O N
N N N
O Zn Zn
N N N N
O
Et Et
39 40
C5H11
C5H11 C5H11
C5H11 C5H11
C5H11 C5H11
C5H11
O O
O O
O O
O O O
O R
O O R
R O O
O O
R N
R NR N Zn N
N Zn N N
N
43, 44
41, 42
O
41 : n = 1, 42 : n = 5
R= NH (CH2) n 43 : n = 1, 44 : n = 5
O
Et Et 45 : X= O
O
O
N N
Zn O
N N
46 : X= O
X
Et Et Et O
Et
O
X
N N
47 : X= O
Zn
N N
O
O O
Et Et O
O N N
48 : X=
45-48 O
O O O
determined by the emergence of weak CH-π interactions in the system considered, which
is the cause of higher binding constants. In α4 atropisomers 32, both sides of the reaction
centre of porphyrin are inadequate. Four ortho-substituents shield the reaction centre almost
completely, and the ligands are coordinated only by the “open” side of the porphyrin mac-
rocycle, without forming any additional bonds. .
Complexation of Porphyrins with Ions and Organic Molecules 131
R R .
N N
N N Zn
Zn N N
N N
RO
R OR
( )n R RO
R R ( )n OR
N N N N
Zn Zn
N N N N
R R
49 - 51 52, 53
O
49 : R= , n=2 52 : R = H, 53 : R = COCF3
O
50 : R = (CH2)2COOCH3 , n = 2
51 : R = (CH2)2COOCH3 , n = 4
Ligand 41 42 43 44 Zn-TPP
233 46 80 20 1.6
N
0.4 10 52 13 1.4
N
1077 140 290 79 1.5
N
O
0.7 0.3 62 19 0.3
NH 2
O
N H 2 0.4 13 15 1.6 0.3
0.2 16 20 41 1.4
N
outer side of the receptor with the binding constant equal to or even smaller than in Zn-TPP.
Binding of smaller-size molecules (N-methylimidazole, pyridine, picoline) strongly de-
pends on the shielding effect of the calix[4]arene cover and on the electron density at the
donor atom of nitrogen. The presence of long and flexible binding links in receptor 42 leads
Complexation of Porphyrins with Ions and Organic Molecules 133
to the formation of a large-size cavity, and most molecules of the host are bound on the inner
side of the receptor (excluding nicotine). Simultaneously, the shielding effect of the
calix[4]arene “cover” is attenuated. The decrease of the number of binding links in complex
43 to two leads to an increase of the molecule’s flexibility and, as a consequence, to an in-
crease of the distance between the porphyrin and calix[4]arene fragments. This enables the
molecule of 4-phenylpyridine to penetrate into the inner cavity of the receptor. Porphy-
rin-calix[4]arene 43 binds nicotinamide 200 times stronger as compared with Zn-TPP and
is the best receptor in the sequence of complexes 41--44 for binding nicotinamide, 4-phe-
nylpyridine and isonicotinamide.
An increase of the length of the binding links in compound 44 leads to a decrease of
the shielding effect of the calixarene cover as compared with complex 43. In compound 44,
the flexibility of the binding links enables the molecule to “fold” the binding links with a
shift of the porphyrin and calix[4]arene fragments relative to one another. This decrease of-
the inner cavity leads to a decrease of the complexing ability of receptor 44 with small mol-
ecules as compared with complexes 41--43. An exception is nicotine, whose efficient
binding can, probably, be explained by the additional formation of hydrogen bonds with
amide groups of the binding links.
In cyclophane dimers 45--53, each monomeric porphyrin fragment is also shielded on
one side owing to the other porphyrin fragment, as the result of which Ka of the ligands by
dimers 45, 47, 48 and 51 is lower than of their monomeric analogues. In coordination by
47, 49 and 50 of monodentate ligands to form complexes of composition 1:1, the binding
constants are observed to increase as compared with monomeric analogues (in the case of
the coordination of pyridine, Ka increases two- to threefold). Some authors associate this
with electronic effects [34, 54].
In their opinion, pyridine, along with the formation of a σ-bond Zn--N, interacts with
metal by the π-type with the transfer of electron density from dxz, dyz orbitals of Zn(II) to
vacant π*-molecular orbitals of pyridine. In passing from monomeric porphyrin to dimeric
one, the probability of this interaction increases due to the enhancement of the electron-do-
nor properties of the deformed porphyrin macrocycle, which ultimately is accompanied
with an increase of Ka of the pyridinate axial complex of cyclophane dimer.
According to the opinion of other authors [59], an increase of the binding constant of
one ligand by dimer as compared with monomer, as in the cases of spatially hindered por-
phyrins, is determined by the correspondence between the size of the inner cavity of dimeric
porphyrin and that of the ligand. The better this correspondence, the greater the binding con-
stant is. In the sequence of para-substituted pyridins, the best correspondence has para-tert-
butylpyridine (Ka increases 50-fold). The second extra ligand is attached only from the out-
er side and its binding constant is much weaker.
In coordination by cyclophane dimeric porphyrins of bidentate ligands (with two ni-
trogen atoms), if the geometric requirements of the interplane space correspond to the size
of the ligand, a dimeric porphyrin--bidentate ligand complex of composition 1:1 is formed,
in which the bidentate ligand is located in the inner cavity of dimeric porphyrin and had two
binding points.
Conditions of forming various-composition complexes, their stability constants and
other regularities are described in works by Sanders by example of such ligands as pyrazine,
4,4′-dipyridyl and DABCO (1,4-diazo-bicyclo[2,2,2]octane) [53, 56, 60]. Zn-porphyrin
with monodentate ligand forms a complex of composition 1:1 with the binding constant Ka
(Scheme 1). In the case of bidentate ligand, a 1:1 complex is formed with the binding con-
stant Ka,1 ≈ 2Ka (Scheme 2). Addition of the second ligand to form a 2:1 complex occurs
134 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Scheme
Scheme1 1
L
N Ka
N N
Zn + L Zn
N
N N N N
Scheme 2
Scheme 2
N N
Zn
L Ka,1 N N
N N
Zn
N N + L
L
L
Scheme 3
Scheme 3
N N
N N Zn
Zn N N
N N
N N Ka,2 L
N
Zn L
N
+
L
L
N N
Zn
N N
with Ka,2, which is related to Ka,1 as 4Ka,2 = aKa,1, where a is the “interaction” parameter,
being a measure of interactin between two reaction centres (Scheme 3).
In the case of bifunctional ligand and dimeric porphyrin, the situation becomes more
complicated. Complexes can be formed both on the inner (Ka (in)) and outer ((Ka (out)) side
of the dimer. The strongest complexes are formed in the case, when bifunctional ligand is
coordinated by the inner cavity of dimer. Formation energy of this complex in the general
Table 14 (continued)
Ka of Zn-porphyrins with bidentate ligands.
case is not the doubled energy of forming a complex with monomer, owing to a whole range
of factors: the chelate effect, interaction between reaction centres of bidentate ligand, con-
formation changes, as well as new interactions, which can emerge in the system. These
components of the total binding energy of the dimeric porphyrin--bidentate ligand complex
depend on the nature and conformations of linking bridges in dimeric porphyrin and can be
rather significant. Depending on the nature of a linking bridge, the constant of binding of
dimeric porphyrin with bidentate ligand by the two-site scheme can be higher, as compared
with the monomeric analogues, from 2 up to 5,000 times (Table 14).
Coordination of bidentate ligands by linear dimeric porphyrins depends on the flexi-
bility of the linking bridge. In the case of dimer 54 with the rigid aromatic bridge, porphyrin
fragments are fixed in space at a large distance one from another, and complexation with
DABCO in CH2Cl2 proceeds by the type of 1:2 interaction with Ka ≈ 105 l/mol (each
metalloporphyrin fragment of the dimer binds one molecule of bidentate ligand) [54, 61].
In dimer 55 with a more flexible linking bridge, complexation with DABCO proceeds to
form an “internal” complex of composition 1:1 with Ka ≈ 107 l/mol (two metalloporphyrin
fragments of the dimer bind one molecule of bidentate ligand). “Internal” complexes with
DABCO are also formed by dimer 56 (Ka = 6·105 l/mol) [59, 62]. The authors of [55] note
a high affinity to diamine in “flexible” dimer 57 with a crown-ether linking bridge; in the
presence of barium perchlorate, the affinity increases.
The recognizing ability of calixarene-porphyrin conjugates 58, 59 with respect to
DABCO was studied in [19, 56]. It is noted there that compound 59 forms an internal 1:1
136 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
OH
N N
N Zn N N Zn N
N N
54
O O
O O
N N
N Zn N N Zn N
N N
55
Ph
N N
Zn Ph
N N
N N
Zn
N N
O Ph
O
O
O
N N
O O
Zn
N N
Ph
N N
Zn
N N Ph
56
57
Ph
complex, whereas 58 coordinates DABCO only on the outer side. A significant difference
of the complexing ability of the calix[4]arene derivative 58 from thiacalix[4]arene deriva-
tive 59 is explained by the authors by differences in the size of respective calixarene cavi-
ties. A larger cavity in conjugate 59 corresponds better to the geometry of bidentate ligand
containing two diametrically located nitrogen atoms, which leads to the formation
Complexation of Porphyrins with Ions and Organic Molecules 137
O
t N
Bu NH N
O Zn
N N
XX
t
Bu O R
t
Bu o R
X
O
X N N
t O NH Zn
Bu N N
58-59
58 : X=CH2, 59 : X=S
R R
N N
Zn
.
N N
R R
SiMe3
R R n
N N
Zn
N N
Me3Si
60-62 R
R
60 : R = CH2CH3, n=3 63
61 : R = (CH2)2COOCH3, n=3
62 : R = (CH2)2COOCH3, n=4 n=4, R=CH2CH2COOCH3
of an intramolecular 1:1 complex with the binding constant Ka = (1.0±0.1) ·107 l/mol in
chloroform at 294 K. Conjugate 58 containing the usual calix[4]arene fragment forms a
complex by way of independent coordination of DABCO molecules by two tetrapyrrole
macrocycles (each metalloporphyrin fragment coordinates a ligand molecule) to form a 2:1
complex.
A detailed investigation of binding oligopyridyl ligands by di-, tri- and tetrameric por-
phyrins was carried out in [59--62]. A good geometrical correspondence of the size of por-
phyrin cavity and the size of ligand is observed in dimeric porphyrin 50 with BiPy (6 ·106
l/mol), trimeric porphyrins 60, 61 with Py3P (9 ·109 and 4·1010 l/mol, respectively), tet-
rameric porphyrin 62 with H2Py4P (2 ·1010 l/mol), which is slightly larger than in linear
tetramer 63 with this ligand (7·109 l/mol) [43]. Dimeric, trimeric and tetrameric porphyrins
synthesized according to an improved method have even more capacious intromolecular
cavities than cyclophane dimers considered above. The distance between porphyrin frag-
ments in dimer 51 is approximately 1.5 nm. This enables the zinc complex to be the host
138 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
R R
N N
Zn
N N
R R
n
64
n=3, R= CH2CH2COOCH3
N N N N
Py2C2
BiPy N
N
N N
N
N
N N
N
Py3P N Py3C12H3
NH N
N N
N HN
H2 Py4P
H2P H2P
H2P N Ru N H2P
H2P
H2P
65
+
N
N N
Zn
N N N N
+ +
N M N
R
N N
67-69
66 : M=H2, Zn, Co
plexes are formed based on two or three (maximum four) porphyrins; based on
Ru-porphyrin complexes, compounds based on five, seven and nine porphyrins were pro-
duced. An example of such a compound is Ru-complex 65 [64].
In processes of molecular recognition of amino acids, sugars, quinones and purine
bases by porphyrins, of great significance are hydrogen bonds, van-der-Waals and disper-
sion interactions. Recognition of amino acids in aqueous media was studied in [67, 68] by
the example of tetrakis-(4-methylpyridyl)-porphyrin 66 and its zinc and cobalt complexes.
In the case of metalloporphyrins, the main driving force of amino acid recognition is
extracoordination of the amino group of amino acid to metalloporphyrin. Besides this, a
weighty contribution is introduced by the interaction between the molecule of porphyrin
and amino acid (ligand--ligand interaction) and dispersion interactions between carbonyl
anion of the guest and pyridinium cation of the host. The value of dispersion interactions
depends on pH [68]. Ligand--ligand interaction takes place in the case of amino acids with
aromatic substituents phenylalanine and triptophane. This π--π interaction between the por-
phyrin molecule and aromatic moiety of amino acid stabilizes the complex [69], thus sig-
nificantly increasing the binding constant (from 60 l/mol in glycine up to 1200 l/mol in
triptophane).
140 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Complexation of porphyrinates 70–76 with amino acid esters was studied in [74]. For-
mation of an additional hydrogen bond between the hydroxy group of porphyrinate and oxy-
gen atom of amino acid (complex 77) can be judged by the ratio of the association constants
(Ka) of porphyrins with hydroxy and methoxy groups of analogous structure. Thus, for three
diphenylporphyrins 71, 73 and 75, the hydrogen bonds with all amino acids studied are
formed only when the hydroxy group is in para-position (Table 16). The association con-
stants (Ka′) of complexes with two binding points (donor-acceptor and hydrogen bonds) dif-
fer from the association constants (Ka′′) of complexes with one binding point (donor-
acceptor bond) two- to fourfold. As seen from the data of Table 17, the energy of axial co-
ordination for all systems studied is within the limits of 22–24 kJ/mol, whereas the largest
value of hydrogen bonds, which are formed in Zn-porphyrin–amino acid methyl ester
H R
H3C CH3
70 : X = H
H9C4 C4H9 71 : X = OH-p CO2CH3
N N 72 : X = OMe-p NH2
Zn 73 : X = OH-m HO OH
74 : X = OMe-m Zn
N N
H9C4 C4H9 75 : X = OH-o
76 : X = OMe-o
H3C CH3
77
X
70-76
Complexation of Porphyrins with Ions and Organic Molecules 141
complexes, are ~4 kJ/mol. The number of phenyl fragments with hydroxy groups also af-
fects the value of Ka, but to a lower degree. The largest differences are observed in cases
with the symmetric and nonsymmetric arrangement of phenyl rings.
Table 16 Association constants observed in binding of esters of amino acids and porphyrinates
(72--77) in toluene at 293 K.
Zn-porph. (1) Ka, l/mol Zn-porph. (2) Ka, l/mol Ka' /K''a
Gly-OMe
70 1750
71 6270 72 1420 Ka(71)/ Ka(72) 4.4
73 2730 74 1430 Ka(73)/ Ka(74) 1.9
75 1680 76 1400 Ka(75)/ Ka(76) 1.2
L-Ala-OMe
70 1020
71 2660 72 915 Ka(71)/ Ka(72) 2.9
73 1680 74 920 Ka(73)/ Ka(74) 1.8
75 1080 76 910 Ka(75)/ Ka(76) 1.2
L-Leu-OMe
70 1660
71 2850 72 1120 Ka(71)/ Ka(72) 2.5
73 2480 74 2040 Ka(73)/ Ka(74) 1.2
75 1390 76 1380 Ka(75)/ Ka(76) 1.0
Table 17 Energies of hydrogen bonds (∆Ghb), axial coordination (∆Gac) and the total free
association energy (∆Gtotal) of porphyrin–amino acid ME in toluene.
The effect of hydroxyl groups, which are at both sides of porphyrin receptor 78, on its
recognition properties to amino acides is considered in [75[. The largest binding constants
(6.9·104 l/mol) are observed in the case of aspartic acid dimethyl ester with two –COOCH3
groups, whereas in leucine methyl ester the binding constant is four times as small (1.8·104
l/mol).
142 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
R N
N N
N NH2
OH
OH OH
OH
N N
Zn
N N N N
Zn
OH N N
OH
NH2
N
78
N
N 79
N
R
The recognizing ability of porphyrin receptors with respect to purine bases was studied
by the example of Rh(III) and Zn(II) complexes. Rh(III) complexes exhibit a very strong
affinity to nucleobases, especially to adenine. For instance, meso-tetraphenylporphyrin
Rh(III)·Cl has Ka in CH2Cl2 with 9-ethyladenine of the order of 107 l/mol, which in practice
indicates the irreversibility of this process [76]. 9-Ethyladenine is coordinated on Rh(III)
of the metalloporphyrin fragment by N-1 atom, which is indicated by X-ray data [71]. Un-
like Rh(III)-porphyrins, the coordination process of nucleobases on Zn(II) porphyrins is a
reversible process. Ka of Zn(II)porphyrin--purine base complexes of composition 1:1 are
within the limits of 102 -104 l/mol [77].
Porphyrins with two hydroxyl groups on one side of the macrocycle form complexes
79 with adenine; in these complexes, one of the adenine molecules is coordinated by cation
of the tetrapyrrole fragment, and the other by the hydroxyl groups of the naphthalene frag-
ments. Such porphyrins are called “bidentate”, and the binding constants of such complexes
are several orders higher [77].
Complexes of sugars with Zn-porphyrins containing two nitrogen atoms at the periph-
ery of macrocycle 80 were studied in [78--83]. The hydroxyl groups of sugars were found
to be involved in the formation of bonds with: 1) Zn cation of the porphyrin’s reaction cen-
tre; 2) nitrogen atoms of the aryl substituent. The binding constants of these complexes are
presented in Table 18.
Besides Zn-porphyrins, some porphyrin ligands also possess a recognizing ability with
Complexation of Porphyrins with Ions and Organic Molecules 143
OR −
δO
δ− O H
+ O
δH O
δ− δ+H O
H δ−
N N
N N
δ+Zn
N N
80
NH N
N HN Br
Br N N
X X X
X Y X N
N Y X N N
OR OR N
NH HN OR OR
X N X X N X
N
OR OR OR OR NH N
Br N HN Br
O O
O O O O
N N S
S N N
M O
M
N N O
N N
O
O
83 : M= H2, Zn
respect to sugars [84-86]. Thus, Kral et al. made use for this purpose of such porphyrin
receptors as macrocycles with 1,1′-binaphthosubstitution 81 or cyclic cryptand-porphyrin
conjugates 82 [84--86]. The latter are more efficient in aqueous media.
Increased interest in triple acceptor--donor--acceptor systems for studies of particular
stages of photosynthesis led to a large number of works on quinone-substituted porphyrins.
Of special interest are model compounds with well-defined geometry. Thus, in [87] it was
found that the conformational arrangement of the quinone fragment in porphyrin 83 de-
pended on the presence or absence of Zn cation in the tetrapyrrole macrocycle. In the pres-
ence of Zn, the quinone fragment is immediately over the reaction centre of porphyrin and
enters into a donor-acceptor interaction with it.
Porphyrins with four hydroxyl groups possess a recognizing ability with respect to var-
ious quinone substituents,, irrespective of whether it is metalloporphyrin or a free base.
144 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
R1 R1 R1 R1
OH
OH
OH N
NH - + -
- N
N
HN SO3 SO3 - SO3
N 2+ SO
N3
OH Zn
N N N N+
+
R R
+
N
84 R
85 : R1= O
O ,
R=
Especially high values of binding constants are observed in the case of porphyrin 84 and
methoxy-substituted quinones. Ka for 2,3,5,6-tetramethoxy-4-benzoquinone is 3.5·104
l/mol [88].
As an example of supramolecular porphyrin-based complexes formed only at the ex-
pense of dispersion interactions, we can name complexes formed by tetracations of Zn-por-
phyrins and tetraanions of calix[4]arenes [89--90]. The high strength of complexes formed
is indicated by high binding constants -- Ka of complex 85 in CHCl3 is 1.4·107 l/mol [89].
R1
R3 Me
Me
Me Me
NH N
NH N
R R
R2 R
N N HN
HN
Me Me
Me R1 Me R1
86-88 89
86 : R = Ph(OMe)2 -o,o; R1=R2=R3=Ph
R=PhB(OH)2-o, R1 =Ph
87 : R = R2= Rh(OMe)2 -o,o; R1=R3=Ph
88 : R = R1=R2=Ph(OMe)2-o,o; R3=Ph
Et Et
Et Et
N N
N
Zn
N N
N
Et Et
Et Et
90
acids). These supramolecular complexes have a rather large signal of induced circular
dichroism (CD) in the region of the Soret band, whose sign correlates with the relative value
of substituents at the asymmetric centre and makes it possible to determine the absolute con-
figuration. Although the authors do not discuss the electronic nature of observed optical ac-
tivity, induced chirality exhibits the properties of memorizing the optical activity in the
replacement of chiral acid by achiral acid. Another interesting example is the porphyrin
sensor for D-lactulose, whose selectivity is determined by the correspondence in the ar-
rangement of the reaction centres of sugar considered and two fragments of boric acid in
cis-porphyrin 89, which leads to the formation of a 1:1 complex with a CD signal of one
sign in the Soret band region [119]. The works [79, 120, 121] developed and investigated
the complexation properties of porphyrin-containing sensors for saccharides 68, 78, 90, 91
and alkaloids 92.
We would especially like to mention host--guest systems, which contain amino acids
being natural building blocks. Thus, using trans-isomers of porphyrins 68 and 90, the au-
thors of [75] developed a receptor for amino acids with a high selectivity with respect to
aspartic acid methyl ester (the binding constant corresponds to 6.98·104 M –1). The confor-
mational correspondence in this case creates conditions for three-point binding (one coor-
dinational interaciton and two interactions to form hydrogen bonds). In all systems studied,
a negative-positive CD signal of small intensity was found in the region of the Soret band,
induced by electronic interaction through the space between porphyrin and carbonyl group
of amino acid, though the influence of other factors is not to be ruled out, either. To expand
146 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Ph
O(CH2)3OH
Me O(CH2)3OH
N N N
HO(H2C)3
N N Ph Zn PhOCH2COR-o
M
N N
N N
Et Me
Et (CH2)3OH Ph
92 : R = OH, NH2
91 : M = Lu(NO3), Gd(NO3)(Cl)
R1
XOOC(H2C)n
93 : M = Zn, R = R1 = COOX
N N
R1 M R XOOC(H2C)n
N N
X = Me or K, n=1,4,10
R1
93-95
O
X
O O
94 : M = Gd(acac), R = R1 = Y
O O
X O
the potentialities of using receptors for amino acids, the work [122] proposes a number of
porphyrins (93), which in nonpolar organic solvents can be used as esters, and in aqueous
media as products of their hydrolysis. It has been found that two factors contribute to the
complexation: electrostatic interactions in coordination of the amino group of amino acid
and Zn ion of the porphyrin cycle in organic solvents (enthalpic forces) and host--guest dis-
persion interactions (enthalpic force) in combination with the desolvation-controlled
Complexation of Porphyrins with Ions and Organic Molecules 147
PhSO3H-p Br Ph PhOMe-o
Br
Br
N N
NH N
Ph Zn Ph
p-HO3SPh PhSO3H-p
N N
N HN
Br Br
Br Br
PhSO3H-p Ph
96 97
process of complexation (entropic force). The latter predominate in water. In [123], recog-
nition of the optical activity of zwitterionic amino acids was studied in a two-phase organic
solvent--water system. For extraction from the aqueous phase, use was made of porphyrins
94, which formed 1:1 complexes with amino acids. Their CD spectra have two well-re-
solved opposite-sign signals, which correspond to a split Soret band and absorption bands
in the visible range of the spectrum. The value of CD molar amplitude (A) strongly depends
on the nature of solvent and the nature of substituents of phenyl rings. The largest sensitivity
was found for aromatic solvents and substituent X=t-Bu, and the sign of induced CD is de-
termined by chirality of amino acids. This method is proposed by the authors for determin-
ing the absolute configuration of α-amino acids. The sensitivity of this method was
subsequently increased by including a crown ester fragment into porphyrin molecule 95
[123]. Simultaneous coordination of the CO2– 2 group of amino acid by the lanthanide reac-
tion centre of porpyrin and the NH3+ group by the crown ester fragment in combination with
the respective choice of metal significantly increases the extraction capacity and value of
A. In [125], recognition of amino acids was done using the method of solid-phase optical
detection. It is based on hypsochrome shifts in the visible spectrum of porphyrin 96 immo-
bilized as a monolayer on a cellulose film, which occur owing to the interaction with amino
acids. The extent of spectral changes depends on the structure of amino acids, which makes
possible quantitative assays of amino acid in solutions.
In the literature, there is an large body of information on supramolecular systems, in
which both the guest molecule and the host molecule possess a chirality. On the whole, the
effect of chiral recognition is based on the principle of three-dimensional correspondence
according to the lock--key principle, when the complexing centres of the guest and host well
correspond to each other, and usually requires the presence of three sites for respective in-
teractions. Ideally, of two enantiomeric forms of the guest molecule, only one form would
be efficiently bound by the respective host molecule to form a stable complex, whereas the
other enantiomeric form woudl be bound weaker. This difference in binding can be estab-
lished by various spectral methods.
For instance, two diastereomeric complexes of porphyrin 97 and (S)-2-pyrro-
lidinemethanol were easily identified in [126] by the method of 1H NMR as a consequence
of the stabilization of the optimal complex due to coordination and formation of hydrogen
bonds. In another case, a direct correspondence was established between enantioselectivity
in complexation and catalytic oxidation [127]. Thus, at the substitution of Fe by Al, 1,1′-bi-
naphthyl derivative of porphyrin 98 was found to have clear distinctions in the electronic
148 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
O
R N
H
O
NH
N N
Cl MeO
Fe R=
MeO
N N
HN
O
H R
N
98
Ph
N N
O O
Ph Zn R
P
N N
O
100 : R =
Ph
99-101
SiMe3
O O
O O
P P
O SiMe3 O
99 : R = 101 : R =
absorption spectra in antipodal binding of epoxides and alcohols; these distinctions occur
against the background of the linear dependence between respective differences in free
energy for complexation and catalytic activity, which indicates the similar stereoselective
mechanism for both processes. Binaphthyl fragments were also used to obtain optically
active phosphite-containing porphyrins 99--101. They, in combination with various phos-
phorus-containing chiral and achiral ligands, were applied as supramolecular bidentate
Complexation of Porphyrins with Ions and Organic Molecules 149
R2CONH
N N 105 : M =Zn, R=R1=Ph,
+
R2=R3=CH2N Me3
M
N N
NHCOR
NO2
O
NH
O
HN
N N
Zn
N N
NH
O
O 2N
HN
O
104
tive Zn-porphyrin complex (103) can also efficiently recognize the [S]-antipode of bulky
1-(1-naphthyl)ethylamine with a 2:4 enantiomeric binding ratio [132]. The further improve-
ment of selectivity up to 7.5 (also for valine methyl ester) is achieved in the case of porphy-
rin 104 by way of respective coordination and hydrogen bonds between host and guest
molecules [132]. The thermodynamic analysis revealed a significant change in the guest
molecule owing to the coordination process, whereas the enthalpy factor determined mainly
by the formation of hydrogen bonds indicates the stabilization of the complex as a whole.
Two enantiomers of chiral water-soluble porphyrin 105 were also used for chiral recogni-
tion of a number of amino acids and dipeptides in an alkaline aqueous solution. The stere-
oselectivity changing from 1.2 up to 3.3 is based on a combined action of the effect of
coordination, Coulomb and spatial interations [134]. The work [135] describes one of the
most efficient ways of recognizing amino acid esters with an enantioselectivity of 21.54 (for
methyl ester of phenylalanine) using amino acid-modified porphyrins 106--108 [135].
Enantioselectivity was shown to depend exponentially on temperature, and it was found that
diastereoisomeric complexes exhibit various CD properties . In particular, the CD spectrum
of porphyrin 106 in the presence of weakly bound methyl ester of L-analine is similar to the
spectrum of individual 106 and represents a single negative Cotton effect in the Soret band
region. On the other hand, strongly bound methyl ester of D-alanine gives a positive-neg-
ative signal determined by exciton interactions between porphyrin and the carbonyl group
of amino acids. The importance of the model of three-site interactions for chiral recognition
was additionally confirmed by low stereoselectivity (15%) of the Ru complex of porphyrin
109 in the binding of racemate isocyaanides (it is capable of forming respective coordina-
tion complexes, but has no additional interactions sites related to the formation of hydrogen
bonds) [136]. In [137--141], by the example of chiral porphyrins 110, 111, it was clearly
shown using the methods of X-ray diffraction analysis and 1H NMR that a combinaiton of
one coordination interactin and two interactions by the type of forming hydrogen bonds pro-
vides for efficient chiral recognition of amino alcohols and amino acid esters as in the case
of the Co complex of porphyrin 110. Moreover, this porphyrin is suitable for the determi-
nation of the enantiomeric excess in primary amines and aziridines by 1H NMR, yielding
a linear dependence for a large number of enantiomeric compositions with a relative error
of several percent. Porphyrin 110 was also found to feature the dependence of the confor-
Me
O Me COOMe
O
R R
O O Me Me COOMe
N N N N
M M
N N O N N
O MeOOC Me Me
R R
O O
MeOOC Me
Me
Me Me
110 : M =Co, Fe
109 : R = O O
Complexation of Porphyrins with Ions and Organic Molecules 151
Me Me Me Et
Me
A
HO Et
OH O O
Me Me CONH(CH2 )2 NHCOCH2
O
Et Et
NH HN
Me Et
HO
Me N N Me
H
N
R R3 111 : R=R3=Et, R1=Me, R2=CH2A
mational state of the macrocycle on the nature of the metal’s central atom; the dependence
is exhibited in the conversion of α,α,α,α-conformation to α,β,α,β-conformation at the sub-
stitution of Zn by Ni.
Various porphyrin analogues possessing a chirality are also used for binding optically
active compounds. Thus, in [142] complex lasalocid-sapphirine conjugates 111--113 were
used to transfer aromatic amino acids from one water phase into another water phase via an
organic layer. The proposed reaction mechanism includes the formation of a stable su-
pramolecular ensemble between the conjugate and zwitterionic amino acid owing to the
electrostatic coordination of carboxylate anion to protonated sapphirine and the formation
of hydrogen bonds between the NH3+ group and polyester ionophore fragment of lasalocid.
The above-considered examples of host--guest systems based on monomeric porphy-
rins clearly show the significance and broad potential of their applications. However, in
many cases the sole monomeric porphyrin fragment can not efficiently funciutn without the
synergic support of another porphyrin fragment, or several porphyrins. This aspect promot-
ed the development of new chirogenic host--guest ensembles based on dimeric and oligo-
meric porphyrins, which shall be considered in the next section.
R1
R
R2 N.
R N .
N
N
Fe N.
N . R
R N
N
R1
R
Ce
O
R R
N R2
R N
N N
Fe N
N
N R R2 N
R1
R
114-115 116-121
The signicance of these two factors is easily understood from the following examples.
Thus, double-decker bis-porphyrins 114--121 linked centre-to-centre were developed
based on µ-oxo-dimers and Ce complexes to observe the chirality of various guest mole-
cules [143--148]. The association mechanism includes the interaction of the bidentate guest
with two complexing groups of adjacent porphyrins, which makes these porphyrins rotate
aroung the central axis and, respectively, to turn macrocycles left or right depending on the
stereochemistry of the guest molecule. This asymmetric travel leads to the emergence of a
significant optical activity (the consequence of interporphyrin exciton interaction) in the re-
gion where porphyrin absorbs. Though the electronic nature of induced CD has not been
completely explained, these systems were successfully used to observe chirality of saccha-
rides (for 114, 115, 119 and 120), dicarboxylic acids (for 116--118), dianions (for 121) and
memorize chirality (for 118). The generated optical activity was excellently stored for three
days at 0°C and even for one year at --37°C. Host--guest complexation occurs with a high
degree of cooperativity, exhibiting a positive allosteric effect, and CD properties strongly
depend on the guest [143, 144] and solvent [148] molecule structure and pH [147]. It is in-
teresting to note that the effect of a change of optical activity controlled by the number of
saccharide links was established for porphyrin 119 in the case of maltooligosaccharide
guest molecules. An entropically unfavourable face-to-face arrangement of bis-porphyrins
can be also preserved owing to the diametrically located covalent bridges as in the case of
cryptand-containing porphyrins 82, 122 [86]. These host molecules are also capable of
binding various saccharides predominantly by encapsulation. An observed very weak in-
duction of CD in the region of Soret band is a consequence of the inability of considered
bis-porphyrins to follow the stereochemistry of the guest molecules, probably, because of
the structural features of the linking bridge. More flexible covalent bridges of various struc-
tures and lengths were used to ensure the face-to-face orientation in porphyrins 123--125
[149, 150]. bis-Porphyrins 123, 124 with a longer bridge showed a high correspondence to
Complexation of Porphyrins with Ions and Organic Molecules 153
R R2 R1 A
R1 R
NH N
(CH2)6 N
N HN
CO(CH2)4CO (CH2)6
R1 R1 N
R R2 R
A
A R R2 R1
(H2C)6 (CH2)6
N
R1 R N
NH N (CH2)6
N HN
R1 R1
R R2 R
R R1 R
122 R R
N N
122 : R=Pr, R1 = Me, R2=H M
R R1 R
122 R R
N N
122 : R=Pr, R1 = Me, R2=H M
N N
R R
R R1 R
A R1
A
R R
R R
N N
M
N N
R R
R R1 R
bis-pyridyl substituted chiral guest molecules of respective length, which leads to the for-
mation of stable inclusion complexes of composition 1:1 with association constants of the
order of 2.3·106 M-1. It was found that the helicity of the guest molecule plays a key role
in transferring information of chirality and induction of a strong exciton interaction in the
reigon of Soret band. bis-Porphyrin 125 with a shorter bridge was used for complexation
with α-hydroxybenzeneacetic acid. Herewith, the activity of CD (A =260 cm – 1 M –1) was
observed to increase significantly (more than sevenfold) as compared with respective
monomer, owing to the conversion of chirality due to the nonplanar structure of the por-
phyrin ring to that due to the helicity of the ensemble as the whole. The stepwise increase
of the concentration of the guest molecule leads to complex changes of CD signals, which
reflect a multilevel equilibrium of the complexation process. Although the observed chiral
responses have not yet been totally explained, these chiroptical systems are of certain in-
terest as potential sensors of chirality in studies of the stereochemistry of interacting guest
molecules. Another structural type of bis-porphyrins is based on combining two chro-
mophore fragments via one covalent bridge. As the mutual arrangment of the porphyrin
154 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
R1 R R1 R
R R
R2
R2 R2 R2
N N N N
R1 M A M R1
N N N N
R2 R2 R2 R2
R R1 R R R1 R
126-133
126 : M =Yb(acac), Gd(acac), R=R2=H, R1=Ph 127 : M =Yb(acac), Gd(acac), R=R2=H, R1=Ph
A CH2O(CH2)2OCH2
A CH2OCH2
macrocycles becomes less predictable in this case due to the reduced rigidity of the system
as a whole, a number of additional factors, such as complexation with the bidentate guest
molecule or shortening of the bridge, can bring two porphyrin fragments closer to each oth-
er and fixe bis-porphyrin in face-to-face orientation or linear conformation. Thus, series of
lanthanide bis-porphyrins 126, 127 linked with ester bonds of various lengths were prepared
as potential host molecules for binding polyions of cystine [151]. These receptors exhibit
a significant selectivity to size in studies of the optical activity of guest molecules. Thus,
the Yb complex of porphyrin 126 with a short bridge efficiently extracts cystine from an
aqueous solution to organic phase, whereas cystathionine, homocystine and methionine are
extracted weakly. In contrast, porphyrin 127 preferably extracts a longer homocystine.
Complexation occurring via the formation of tweezer complexes of composition 1:1 leads
to a complex CD signal in the Soret band region, which is significantly increased as com-
pared with respective monomer. The authors of [152] report of a high selectivity to length
in binding of chiral diamines by porphyrins 128, 129 with rigid bridges. The dimensional
correspondence in the host--guest system is an important condition for efficient complex-
ation with the association constant of the order of 2.4·106 M –1 for the best pair. Induced
CD strongly depends on the volume and number of stereogenic sites, peripheral substituents
of the porphyrin macrocycle and temperature. For instance, the largest value of A (1340
cm –1 M –1) was increased 1.5-fold by decreasing the temperature to --45°C. The binding
link of the type of crown ester significantly enhances the potentialities of controlling the
conformation of bis-porphyrin via specific complexation with ions of alkaline metals
[153--155]. This effect is beautifully demonstrated by the example of an almost twofold in-
crease of the binding constant (from 2.6·105 M –1 to 4.5·105 M –1) upon addition of Na+
ions to porphyrin (130)/trans-1,2-diaminocyclohexane system of the tweezer type due to the
change of geometry of bis-porphyrin. Addition of Na+ ions significantly increases the value
of A of induced CD. A similar effect of binding and CD increase was found in [154] for
porphyrin 131 at the interaction with N-alkyl substituted 1,2-diaminocyclohexane. Howev-
Complexation of Porphyrins with Ions and Organic Molecules 155
A (OCH2CH2)4O
OMe MeO
A N O N
3
O O O
CONH NHCO
A CO2(CH2)5OCO
er, a larger-size K+ ion is required in this case for a larger crown ester cavity. It is noted that
porphyrin 131 is capable of recognizing the chirality of such potassium carboxylates as
camphorates and mandelates. What is more, induced chirality of porphyrin 57 in the twee-
zer conformation was efficiently saved in the substitution of chiral diamine for achiral one
in the presence of Ba ions. The authors of [155] note that in 24 h the intensity of CD in this
system decreases only insignificantly. Although the discussed systems of the tweezer type
exhibit a rather high optical activity due to the chiral spatial arrangement of two porphyrin
chromophores, no detailed explanation of electronic and structural factors involved in this
phenomenon has been given until now.
One of the first attempts to solve this issue in the case of bis-porphyrin systems was
made in [156--166] using a more flexible porphyrin 132, which is capable of taking on a
tweezer conformation in the complexation with bidentate ligands. It has been found that the
emergence of optical activity is based on the stereospecific differentiation of the relative
volume of substituents at the asymmetric carbon atom due to the bis-porphyrin molecule
taking on a spatially less hindered conformation. It should be also taken into account that
other factors, such as formation of hydrogen bonds; presence of heteroatoms, solvent etc.,
can also affect the geometry of the ensemble as a whole. All this leads to an oriented ar-
rangement of two tetrapyrrole fragments in porphyrin 132, which on the strength of exciton
interactions induces a CD signal, whose sign can be used as a tool for determining the ab-
solute configuration of various bidentate guests. In the cases when the relative size can not
be determined directly, the conformation analysis is made using calculations according to
the method of molecular mechanics with the view of explaining the observed contradiction
between the predicted chirality and the one obtained in some cases. Though it is well known
that metalloporphyrin chromophore contains two degenerated (or almost degenerated, due
to the asymmetry determined by meso-substitution) electronic transitions along the meso-
5-15 and 10-20 axes, only the exciton interaction of the pair of 5-15 oscillators was chosen
for explaining the induced optical activity in porphyrin 132. However, the contribution of
10-20 oscillators due to homo- and heteroconjugation can also play an important role in cer-
tain conformations. The works [156--166] also adduce some dependences of induced CD
on the nature of the guest molecule, the solvent and temperature.
156 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
N N N N N N
Zn Zn Zn
N N N N N N
R R R
n-2
134, 135
CO2H
O
134 : n = 2, R = B 135 : n = 2, 3, 4, 8, R = OC12H25
O CO2H
158 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
R
A
R R N N
136 : R = Zn B
N N
R R
A
A = B = Ph(t-Bu)2-m,m
R or
A = H, B = Ph(OC8H17)2-m,m
136-140
O2C X
O2C X
O2C X
137 : R = O2C
O2C
O2C X
138 : R = O2C
O2C
O2C X
O2C X
Further modification of the geometry of porphyrin was done in [186] with the view of
developing supermolecular systems for chiral recognition and enantioselective catalysis
based on heterometallic trimer 141. At the first stage, various chiral carboxylic acids were
totally linked to Sn porphyrins 141 by the type of hexacoordination to form a 1:4 ensemble
possessing a chiral cavity. However, subsequent attempts of optical separation of the race-
mate mixture of chirally substituted pyridines and catalytic asymmetrical epoxidation were
unsuccessful, apparently, due to the loss of rigidity inside the chiral cavity.
A broader application in various fields proved to be chirogenic host--guest systems, in
which both component parts possess a chiral activity due to a number of specific properties.
In particular, chiral recognition is the most appropriate application of various derivatives
of chiral bis-porphyrins. Thus, a series of bis-porphyrins 142--144 was prepared in [187]
for recognition of dicarboxylate anions via a combination of Coulomb attraction and for-
mation of hydrogen bonds. It was found that porphyrins 142 and 143 formed strong com-
plexes with N-carbobenzoxyaspartic acid (N-Cbz-Asp) and N-carbobenzoxyglutamic acid
(N-Cbz-Glu) with association constants within the limits of 104--105 M –1 and exhibit a
preference towards glutamic acid as compared with aspartic acid. Porphyrin 143 shows an
insignificant enantiomeric selectivity, while cyclic and more rigid porphyrin 144 exhibits
a low affinity to guest molecules considered, but shows a significant chiral recognition. Ap-
parently, the latter occurs from a good dimensional and geometric correspondence between
the host and guest as the result of fixation of two porphyrin fragments by two covalent
bridges, however, the detailed mechanism of recognition still requires additional studies.
Complexation of Porphyrins with Ions and Organic Molecules 159
O2 C X
O2 C X
O2C X
O2 C X
140 : R =
139 : R = O2C O2C X
O2 C
O2C X
O2 C
O2 C
O2 C X
O2 C X
O2 C X
N N OCH2 Ph(OMe)2-m,m
X = O2C Zn
N N
OCH2Ph(OMe)2-m,m
MeO2C(CH2)2 (CH2)2CO2Me
Me Me
N N
Ru(CO)(Py)
N N
Me Me
Me MeO2C(CH2)2 (CH2)2CO2Me
(CH2)2CO2Me
MeO2C(CH2)2 Me
Me
N Me
N
N
Sn(OH)2 (CH2)2CO2Me
MeO2C(CH2)2 N N (CH2)2CO2Me
Sn
MeO2C(CH2)2 N (OH)2
N
Me N
Me
MeO2C(CH2)2
Me Me (CH2)2CO2Me
141
Spatially fixed porphyrin 145 was used to produce 80--86% and 48% enantiomeric ex-
cess in binding of esters of histidine and lysine by means of two-site interaction between
two nitrogen atoms of the guest molecule and two zinc ions of the host molecule to form a
tweezer structure [188]. Similarly, for chiral recognition of amino acid derivatives the work
[152] used porphyrin 146 rigidly bound via 1,1′-binaphthyl fragment. This host molecule
exhibits an extremely high affinity to lysine derivative due to the formation of tweezers.
160 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Me
Et
Et Me
Et
N
Me Et
NH
Me N
HN Me
HN
R
Me NH
NH
N
Me HN
Me
Et N
Et
Me Et
Et
Me
142,143
142 : R = A, 143 : R = B
-(H2C)2OCHN -(H2C)2OCHN
A = B =
-(H2C)2OCHN -(H2C)2OCHN
Me
R
Et
Me
N
Me Et
NH
Et
HN N
Et HN Me
Me NH Et
N NH
Et
HN
Et Me
N
Me Et
R
Me
-(H2 C)2OCHN
144 : R =
-(H2 C)2OCHN
Optimization of the geometry of host--guest complexes showed that the selectivity mechan-
ims is based on spatial repulsion between two methoxy groups of porphyrin 146 and the
amide group of the amino acid derivative in the less preferable diastereomeric complex. Af-
ter inclusion of the chiral residue of leucine into the covalent bridge of achiral dimeric por-
phyrin 124 the obtained chiral compound 147 was used by the authors of [189] for optical
separation of a series of artificial bidentate ligands of different lengths, containing a leucine
group in the middle of the molecule (the source of optical activity) and two end pyridine
(providing for the two-site binding. A 80% value of enantiomeric excess with significant
Complexation of Porphyrins with Ions and Organic Molecules 161
R R R R
N N
N
N Zn N N Zn N
N
N N
R R R R
145 : R = Ph(t-Bu)-m,m
Me Me Me Me
Et Me
Me Et
N N N N
Zn A Zn
N N N N
Et Me Me Et
Me Me Me Me
146 : A MeO
MeO
N N
Zn
N N
R
A R A
N N
Zn
N N
147 : A = -CO(NHCMe2CO)3NHC(CHPh)CONHCH(i-Bu)CONH(CHPh)CO(NHCMe2CO)3NH-
enhancement of the CD signal in the Soret band region due to the formation of a preferable
tweezer complex. Although the detailed mechanism of chirality enhancement is not pre-
sented in the work, comparative studies showed the observed effect to be stipulated by the
stabilization of the twisted geometry of the host molecule due to the helicity of the guest
molecule. An opposite tendency is described in [190]. In particular, in the presence of cit-
162 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
Ph Ph
N Ph Ph N
N N
Co Co
N N
Ph N N Ph
O O
O O
S S
148
Et Et
Et Et
N N
Zn CH2 *
N N
Et Et
Et Et
149
4 Conclusion
Material presented indicates that supramolecular systems based on porphyrins are of un-
doubted interest for establishing the absolute configuration of organic molecules of various
nature and developing molecular receptors for a certain type of substrate. The occurrence
of tetrapyrrole chromophores in such systems enables application of spectral methods tra-
ditionally used in porphyrin chemistry for studies of molecular recognition processes. Re-
cent advances in this field are a promising prerequisite for the development of new
molecular technologies, such as chiral nanotechnology, intelligent molecular devices and
chiroptical memory.
Complexation of Porphyrins with Ions and Organic Molecules 163
References
1. B.D. Berezin and N.S. Enikolopian, Metalloporphyrins, Nauka: Moscow (1988) (in Russian).
2. B.D. Berezin, The Coordination Compounds of Porphyrins and Phthalocyanines, Nauka:
Moscow (1978) (in Russian).
3. G.M. Trophymenko, B.D. Berezin and A.S. Semeikin, Rus. J. Inorg. Chem., 39 (9), 1493–1496
(1994).
4. N.Zh. Mamardashvili, L.V. Klopova and O.A. Golubchikov, Rus. J. Coord. Chem., 18 (1),
70–79 (1992).
5. N.Zh. Mamardashvili, L.V. Klopova and A.S. Semeikin, Abstracts of the XXII-th International
Conference in Chemistry of Coordination Compounds, Minsk, 190 (1990).
6. N.Zh. Mamardashvili, L.V. Klopova, S.A. Zdanovich and O.A. Golubchikov, Abstracts of the
III-rd Russian Conference in Chemistry and Application of Nonaqueous Solutions, Ivanovo, 128
(1993).
7. N.Zh. Mamardashvili, S.A. Zdanovich, L.V. Klopova and O.A. Golubchikov, Abstracts of the
VIII-th International Conference in Spectroscopy and Chemistry of Porphyrins and Their
Analogs, Minsk, 129 (1998).
8. B.D. Berezin and N.I. Sosnikova, Kinetika i Kataliz, 7 (2), 248–253 (1996).
9. O.A. Golubchikov and B.D. Beresin, Rus. J. Phys. Chem., 60 (9), 2113–2128 (1986).
10. V. Gutman, Chemistry of Coordination Compounds in Nonaqueous Solutions, Mir: Moscow
(1971) (in Russian)
11. E. M. Kuvshinova, Rus. J. Phys. Chem., 57 (6), 1413–1422 (1987).
12. O.A. Golubchirov, E.M. Kuvshinova, A.S. Semeikin and C.U. Korovina, Rus. J. Phys. Chem.,
63 (4), 912–917 (1989).
13. N.Zh. Mamardashvili, A.S. Semeikin and O.A. Golubchikov, Abstracts of the XV-th
International Symposium in Macrocyclic Chemistry, Odessa, 94 (1990).
14. O.A. Golubchikov, E.M. Kuvshinova and A.S. Semeikin, Abstracts of the XV-th International
Symposium in Macrocyclic Chemistry, Odessa, 93 (1990).
15. E.M. Kuvshinova, O.A. Golubchokov and B.D. Berezin, Rus. J. General Chem., 61 (8),
1799–1804 (1991).
16. B.D. Berezin, T.I. Potapova, R.A. Petrova and L.I. Kalenkova, Rus. J. Phys. Chem., 55 (1),
116–119 (1981).
17. N.Zh. Mamardashvili and O.I. Koifman, Rus. J. Org. Chem., 41 (6), 807–826 (2005).
18. G.M. Mamardashvili, N.Zh. Mamardashvili and O.I. Koifman, Rus. Chem. Rev., 74 (8),
839–855 (2005).
19. P.D. Beer, Chem.Commun., 689–690 (1996).
20. M. Dudic, P. Lhotak, H. Petrickova, I. Stibor and P. Proskova, Org. Lett., 5, 2129–2139 (2003).
21. J.K.M. Sanders, in: The Porphyrin Handbook, ed by K.M. Kadish, K.M. Smith and R. Guilard,
Academic Press: San Diego–San Francisco–New York–Boston–London–Sydney–Tokyo, 3,
347– 381 (2000).
22. H. Ogoshi, T. Mizutani, T. Hayashi and Y. Kuroda, in: The Porphyrin Handbook, ed by K.M.
Kadish, K.M. Smith and R. Guilard, Academic Press: San Diego–San Francisco–New
York–Boston–London–Sydney–Tokyo, 6, 279–340 (2000).
23. E.V. Antina, V.P. Barannikov and A.I. Viugin, Rus. J. Inorg. Chem., 35 (2), 400–404 (1990).
24. B. Cheng and W.R. Scheidt, Inorg. Chim. Acta, 237, 5–14 (1995).
25. H. Imai, S. Nakagawa and E. Kyuno, J. Am. Chem. Soc., 114, 6719–6723 (1992).
26. M. Nappa and J.S. Valentine, J. Am. Chem. Soc., 100, 5075–5080 (1978).
27. B.D. Berezin and O.I. Koifman, Rus. Chem. Rev., 49, 2388–2417 (1980).
28. O.I. Koifman, T.A. Koroleva and B.D. Berezin, Rus. J. Coord. Chem., 4 (5), 1339–1342 (1978).
29. T.A. Koroleva, O.I. Koifman and B.D. Berezin, Rus. J. Phys. Chem., 7 (9), 2007–2012 (1981).
30. K. Belanger, M.H. Keefe, J.L.Welch and J.T. Hupp, Coord. Chem. Rev., 190, 29–35 (1999).
31. A.I. Viugin and G.A. Krestov, Solutions of Nonelectrolites in Liquids, Nauka: Moscow (1989)
(in Russian).
32. V.I. Smirnov, A.I. Viugin and G.A. Krestov, Rus. J. Coord. Chem., 16 (7), 896–901 (1990).
33. E.F. Caldin and J.P. Field, J. Chem. Soc. Faraday Trans. 1, 78, 1923–1927 (1982).
34. O. Middel, W. Verboom and D.N. Reinhoudt, J. Org. Chem., 66, 3998–4005 (2001).
164 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
35. S.G. Puhovskaya, L.Zh. Guseva, E.M. Kuvshinova, N.Zh. Mamardashvili and O.A. Golubchikov
Rus. J. Org. Chem., 11, 851–856 (1998).
36. S.J. Cole, G.C. Curthoys, E.A. Magnusson and J.N. Phillips, Inorg. Chem., 11, 1024–1025
(1972).
37. G.R. Miller and G.D. Dorough, J. Am. Chem. Soc., 74, 3977–3981 (1952).
38. N.Zh. Mamardashvili and O.A. Golubchikov, Abstracts of the XXII-th International Conference
in Chemistry of Coordination Compounds, Minsk, 167 (1990).
39. R.P. Bonar-Law and J.K.M. Sanders, J. Chem. Soc. Perkin Trans. 1, 3085–3096 (1996).
40. C.A. Hunter, M.N. Meah and J.K.M. Sanders, J. Am. Chem. Soc., 112, 5773–5780 (1990).
41. O.E. Storonkina, G.M. Mamardashvili and N.Zh. Mamardashvili, Abstacts of the IX-th
International Conference in Complexation of Macrocyclic Compounds, Ivanovo, 130, (2004).
42. R.S. Wylie, E.G. Levy and J.K.M. Sanders, Chem. Commun., 1611–1612 (1997).
43. H.L. Anderson, S. Anderson and J.K.M. Sanders, J. Chem. Soc. Perkin Trans. 1, 2231–2245
(1995).
44. I.Yu. Bazlova, Abstacts of the IX-th International Conference in Complexation of Macrocyclic
Compounds, Ivanovo, 116, (2004).
45. W. Zielenkiewicz, N.Sh. Lebedeva, M. Kaminski, E.V. Antina and A.I. Vyugin, J. Therm. Anal.
Cal., 58, 741–749 (1999).
46. H. Imai and E. Kyuno, Inorg. Chem., 29, 2416–2419 (1990).
47. H.L. Anderson, C.A. Hunter, M.N. Meah and J.K.M. Sanders, J. Am. Chem. Soc., 112,
5780–5789 (1990).
48. J. Froidevaux, P. Ochsenbein, M. Bonin, K. Scheck, P. Maltese, J.-P. Gisselbrecht and J. Weiss,
J. Am. Chem. Soc., 119, 12362–12363 (1997).
49. H. Imai and Y. Uemori, J. Chem. Soc. Perkin Trans. 2, 1793–1794 (1994).
50. I.P. Danks, I.O. Sutherland and C.H. Yap, J. Chem. Soc. Perkin Trans. 1, 421–423 (1990).
51. N. Zh. Mamardashvili and O.A. Golubchikov, Rus. Chem. Rev., 69, 337–356 (2000).
52. L.Zh. Guseva, Abstacts of the IX-th International Conference in Complexation of Macrocyclic
Compounds, Ivanovo, 97 (2004).
53. C.C. Mak, N. Bampos and J.K.M. Sanders, Chem. Commun., 1085–1086 (1999).
54. C.C. Mak, N. Bampos and J.K.M. Sanders, Angåw. Chem. Int. Ed., 37, 3020–3023 (1998).
55. Y. Kubo, Y. Murai, J. Yamanaka, S. Tokita and Y. Ishimavu, Tetrahedron Lett., 40, 6019–6021
(1999).
56. M. Dudic, P. Lhotak, H. Petrickova, I. Stibor, K. Lang and J. Sykora, Tetrahedron, 59,
2409–2415 (2000).
57. B.D. Berezin, O.I. Koifman and O.A. Golubchikov, Rus. J. Inorg. Chem., 18 (6), 1540–1544
(1986).
58. B.D. Berezin and A.N. Drobisheva, Rus. J. Phys. Chem., 41 (2), 402–408 (1967).
59. H.L. Anderson, S. Anderson and J.K.M. Sanders, J. Chem. Soc. Perkin Trans. 1, 2223–2229
(1995).
60. H.L. Anderson, C.L. Walter, A. Vidal-Ferran, R.A. Hay, P.A. Lowden and J.K.M. Sanders,
J. Chem. Soc. Perkin Trans. 1, 2275–2279 (1995).
61. Z. Clyde-Watson, N. Bampos and J.K.M. Sanders, New J. Chem., 1135–1138 (1998).
62. A. Vidal-Ferran, N. Bampos and J.K.M. Sanders, J. Am. Chem. Soc., 36, 6117–6126 (1997).
63. A. Vidal-Ferran, Z. Clyde-Watson, N. Bampos and J.K.M. Sanders, J. Org. Chem., 62, 240–241
(1997).
64. S.L. Darling, C.C. Mak, N. Bampos, N. Feeder, S.J. Teat and J.K.M. Sanders, New J. Chem., 23,
359–363 (1999).
65. V. Marvaud, A. Vidal-Ferran, S.J. Webb and J.K.M. Sanders, J. Chem. Soc. Dalton Trans.,
985–990 (1997).
66. H.-J. Kim, N. Bampos and J.K.M. Sanders, J. Am. Chem. Soc., 121, 8120–8121 (1999).
67. E. Mikros, A. Gaudemer and R. Pasternack, Inorg. Chim. Acta, 153, 199–210 (1988).
68. C. Verchere-Beaur, E. Mikros and M. Perree-Fauve, J. Inorg. Biochem., 40, 127–131 (1990).
69. C.A. Hunter and J.K.M. Sanders, J. Am. Chem. Soc., 112, 5525–5534 (1990).
70. Y. Kuroda and H. Ogoshi, Synlett., 319–324 (1994).
71. T. Mizutani and H. Ogoshi, Tetrahedron Lett., 30, 5369–5371 (1996).
72. T. Mizutani, T. Ema, T. Yoshida, Y. Kuroda and H. Ogoshi, Inorg. Chem., 32, 2072–2077
(1993).
Complexation of Porphyrins with Ions and Organic Molecules 165
73. Y. Aoyama, A.Yamagishi, M. Asagawa, H. Toi and H. Ogoshi, J. Am. Chem. Soc., 110,
4076–4077 (1988).
74. N.Zh. Mamardashvili, O.E. Storonkina and G.M. Mamardashvili, Rus. J. Coord. Chem., 30 (5),
416–420 (2004).
75. T. Mizutani, T. Murakami, T. Kurahashi and H. Ogoshi, J. Org. Chem., 61, 539–541 (1996).
76. H. Ogoshi, H. Hatakeyama, K. Yamamura and Y. Kuroda, Chem. Lett., 51–54 (1990).
77. H. Ogoshi, H. Hatakeyama, J. Kotani, A. Kawashima and Y. Kuroda, J. Am. Chem. Soc., 113,
8181–8183 (1991).
78. R.P. Bonar-Law and J.K.M. Sanders, J. Am. Chem. Soc., 117, 259–271 (1995).
79. T. Mizutani, T. Murakami, N. Matsumi,T. Kurahashi and H. Ogoshi, J. Chem. Soc. Chem.
Commun., 1257–1263 (1995).
80. T. Mizutani, T. Kurahashi, T. Murakami, N. Matsumi and H. Ogoshi, J. Am. Chem. Soc., 119,
8991–9001 (1997).
81. M. Tichy, Adv. Org. Chem., 5, 115–118 (1965).
82. O.A. Golubchikov, Abstacts of the IX-th International Conference in Complexation of
Macrocyclic Compounds, Ivanovo, 77 (2004).
83. F.A.J. Singelenberg and J.H. van der Maas, J. Mol. Struct., 243, 111–117 (1991).
84. V. Kral and J.L. Sessler, Tetrahedron, 51, 539–554 (1995).
85. O. Rusin and V. Kral, Chem. Commun., 2367–2368 (1999).
86. V. Kral, O. Rusin and F. P. Schmidtchen, Org. Lett., 3, 873–876 (2001).
87. G.M. Sanders, M. van Dijk, A. van Veldhuizen and H.C. van der Plas, J. Chem. Soc. Chem.
Commun., 1311–1313 (1986).
88. T. Hayashi, T. Miyahara, N. Koide, Y. Kato, H. Masuda and H. Ogoshi, J. Am. Chem. Soc., 119,
7281–7290 (1997).
89. R. Fiammengo, P. Timmerman, F. de Jong and D.N. Reinhoudt, Chem. Commun., 2313–2317
(2000).
90. R. Purrello, R. Geremia, L. Randaccio, F.G. Gulino and V. Ravone, Angew. Chem. Int. Ed., 40,
4245–4248 (2001).
91. D. Voet and J.D. Voet, Biochemistry, 2nd edn, Wiley: New York–Chichester–Brisbane–Toronto–
Singapore (1995).
92. J.W. Steed and J.L. Atwood, Supramolecular Chemistry, Wiley: Chichester–New York–
Weinheim–Brisbane–Singapore–Toronto (2000).
93. J. Kyte, Structure in Protein Chemistry, Garland Publishing: New York–London (1995).
94. M.-C. Hsu and R.W. Woody, J. Am. Chem. Soc., 93, 3515–3525 (1971).
95. G. Blauer, N. Sreerama and R.W. Woody, Biochem., 32, 6674–6677 (1993).
96. S.C. Björling, R.A. Goldbeck, S.J. Paquette, S.J. Milder and D.S. Kliger, Biochem., 35,
8619–8627 (1996).
97. A. Boffi, J.B. Wittenberg and E. Chiancone, FEBS Lett., 411, 335–338 (1997).
98. V.V. Borovkov, G.A. Hembury and Y. Inoue, Acc. Chem. Res., 37, 449–459 (2004).
99. H. Ogoshi and T. Mizutani, J. Synth. Org. Chem. Jpn., 54, 906–917 (1996).
100. H. Ogoshi and T. Mizutani, Acc. Chem. Res., 31, 81–89 (1998).
101. H. Ogoshi and T. Mizutani, Curr. Opin. Chem. Biol., 3, 736–739 (1999).
102. J.-C. Marchon and R. Ramasseul, in: The Porphyrin Handbook, ed. by K.M. Kadish, K.M.
Smith and R. Guilard, Academic Press, An Imprint of Elsevier Science: Amsterdam–Boston–
London–Oxford–Paris–San Diego–San Francisco–Singapore–Sydney–Tokyo, 11, 75–132
(2003).
103. X. Huang, K. Nakanishi and N. Berova, Chirality, 12, 237–255 (2000).
104. T.D. James, K.R.A.S. Sandanayake and S. Shinkai, Angew. Chem. Int. Ed. Engl., 35,
1911–1922 (1996).
105. A. Robertson and S. Shinkai, Coord. Chem. Rev., 205, 157–199 (2000).
106. S. Shinkai, M. Ikeda, A. Sugasaki and M. Takeuchi, Acc. Chem. Res., 34, 494–503 (2001).
107. S. Shinkai and M. Takeuchi, Biosens. Bioelectron., 20, 1250–1259 (2004).
108. J.H. Hartley, T.D. James and C.J. Ward, J. Chem. Soc. Perkin Trans. 1, 3155–3184 (2000).
109. H. Tsukube and S. Shinoda, Enantiomer, 5, 13–22 (2000).
110. H. Tsukube, S. Shinoda and H. Tamiaki, Coord. Chem. Rev., 226, 227–234 (2002).
111. H. Tsukube and S. Shinoda, Chem. Rev., 102, 2389–2403 (2002).
112. G. Simonneaux and P.L. Maux, Coord. Chem. Rev., 228, 43–60 (2002).
166 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
153. D. Monti, L.L. Monica, A. Scipioni and G. Mancini, New. J. Chem., 25, 780–782 (2001).
154. Y. Kubo, Y. Ishii, T. Yoshizawa and S. Tokita, Chem. Commun., 1394–1395 (2004).
155. Y. Kubo, T. Ohno, J.-I. Yamanaka, S. Tokita, T. Iida and Y. Ishimaru, J. Am. Chem. Soc., 123,
12700–12701 (1998).
156. X. Huang, B.H. Rickman, B. Borhan, N. Berova and K. Nakanishi, J. Am. Chem. Soc., 120,
6185–6186 (1998).
157. X. Huang, B. Borhan, B.H. Rickman, K. Nakanishi and N. Berova, Chem. Eur. J., 6, 216–224
(2000).
158. T. Kurtan, N. Nesnas, Y.-Q. Li, X. Huang, K. Nakanishi and N. Berova, J. Am. Chem. Soc., 123,
5962–5973 (2001).
159. T. Kurtan, N. Nesnas, F.E. Koehn, Y.-Q. Li., K. Nakanishi and N. Berova, J. Am. Chem. Soc.,
123, 5974–5982 (2001).
160. Q. Yang, C. Olmsted and B. Borhan, Org. Lett., 4, 3423–3426 (2002).
161. G. Proni, G. Pescitelli, X. Huang, N.Q. Quraishi, K. Nakanishi and N. Berova, Chem. Commun.,
1590–1591 (2002).
162. X. Huang, N. Fujioka, G. Pescitelli, F.E. Koehn, R.T. Williamson, K. Nakanishi and N. Berova,
J. Am. Chem. Soc., 124, 10320–10335 (2002).
163. G. Proni, G. Pescitelli, X. Huang, K. Nakanishi and N. Berova, J. Am. Chem. Soc., 125,
12914–12927 (2003).
164. H. Ishii, Y. Chen, R.A. Miller, S. Karady, K. Nakanishi and N. Berova, Chirality, 17, 305–315
(2005).
165. A. Solladie-Cavallo, C. Marsol, G. Pescitelli, L.D. Bari, P. Salvadori, X. Huang, N. Fujioka,
N. Berova, X. Cao, T.B. Freedman and R. Nafie, Eur. J. Org. Chem., 1788–1796 (2002).
166. H. Ishii, S. Krane, Y. Itagaki, N. Berova, K. Nakanishi and P.J. Weldon, J. Nat. Prod., 67,
1426–1430 (2004).
167. V.V. Borovkov, J.M. Lintuluoto and Y. Inoue, Org. Lett., 2, 1565–1568 (2000).
168. V.V. Borovkov, J.M. Lintuluoto and Y. Inoue, J. Am. Chem. Soc., 123, 2979–2989 (2001).
169. V.V. Borovkov, J.M. Lintuluoto and Y. Inoue, J. Phys. Chem. A, 104, 9213–9219 (2000).
170. V.V. Borovkov, N. Yamamoto, J.M. Lintuluoto, T. Tanaka and Y. Inoue, Chirality, 13, 329–335
(2001).
171. V.V. Borovkov, J.M. Lintuluoto, H. Sugeta, M. Fujiki, R. Arakawa and Y. Inoue, J. Am. Chem.
Soc., 124, 2993–3006 (2002).
172. V.V. Borovkov, J.M. Lintuluoto, M. Fujiki and Y. Inoue, J. Am. Chem. Soc., 122, 4403–4407
(2000).
173. V.V. Borovkov, G.A. Hembury and Y. Inoue, J. Porphyrins Phthalocyanines, 7, 337–341
(2003).
174. J.M. Lintuluoto, V.V. Borovkov and Y. Inoue, J. Am. Chem. Soc., 124, 13676–13677 (2002).
175. V.V. Borovkov, G.A. Hembury, N. Yamamoto and Y. Inoue, J. Phys. Chem. A, 107, 8677–8686
(2003).
176. V.V. Borovkov, G.A. Hembury and Y. Inoue, Angew. Chem. Int. Ed., 42, 5310–5314 (2003).
177. V.V. Borovkov, J.M. Lintuluoto and Y. Inoue, Org. Lett., 4, 169–171 (2002).
178. V.V. Borovkov, J.M. Lintuluoto, M. Sugiura, Y. Inoue and R. Kuroda, J. Am. Chem. Soc., 124,
11282–11283 (2002).
179. V.V. Borovkov, J.M. Lintuluoto, G.A. Hembury, M. Sugiura, R. Arakawa and Y. Inoue, J. Org.
Chem., 68, 7176–7192 (2003).
180. V.V. Borovkov, I. Fujii, A. Muranaka, G.A. Hembury, T. Tanaka, A. Ceulemans, N. Kobayashi
and Y. Inoue, Angew. Chem. Int. Ed., 43, 5481–5485 (2004).
181. V.V. Borovkov, T. Harada, Y. Inoue and R. Kuroda, Angew. Chem. Int. Ed., 41, 1378–1381
(2002).
182. M. Ikeda, S. Shinkai and A. Osuka, Chem. Commun., 1047–1048 (2000).
183. C. Ikeda, Z.S. Yoon, M. Park, H. Inoue, D. Kim and A. Osuka, J. Am. Chem. Soc., 127,
534–535 (2005).
184. M. Takase, R. Ismael, R. Murakami, M. Ikeda, D. Kim, H. Shinmori, H. Furuta and A. Osuka,
Tetrahedron Lett., 43, 5157–5159 (2002).
185. W.-S. Li, D.-L. Jiang, Y. Suna and T. Aida, J. Am. Chem. Soc., 127, 7700–7702 (2005).
186. S.J. Webb and J.K.M. Sanders, Inorg. Chem., 39, 5920–5929 (2000).
187. J.L. Sessler, A. Andrievsky, V. Kral and V. Lynch, J. Am. Chem. Soc., 119, 9385–9392 (1997).
168 N.Zh. Mamardashvili, V.V. Borovkov, G.M. Mamardashvili, Y. Inoue and O.I. Koifman
188. M.J. Crossley, L.G. Mackay and A.C. Try, J. Chem. Soc. Chem. Commun., 1925–1927 (1995).
189. Y.-M. Guo, H. Oike, N. Saeki and T. Aida, Angew. Chem. Int. Ed., 43, 4915–4918 (2004).
190. R. Paolesse, D. Monti, L.I. Monica, M. Venanzi, A. Froiio, S. Nardis, C.D. Natale,
E. Martinelli and A. D’Amico, Chem. Eur. J., 8, 2476–2483 (2002).
191. V.V. Borovkov, G.A. Hembury and Y. Inoue, J. Org. Chem., 70, 8743–8754 (2005).
5 Chemical Activation of
Porphyrins in Coordination
Core Reactions
Introduction
Coordination cores in molecules of porphyrin ligands (H2P) and their complexes with p, d
and f metals (MP) are some of their main reaction centres [1–3]. These are complex mul-
tifunctional centres, as a rule, N4H2 and MN4, respectively. Their unique reactivity is de-
termined by a combination of a number of factors. They include not only the chemical
affinity of atoms and atomic groupings constituting these cores, with respect to electro-
philic-nucleophilic reagents, their capability of synchronous donor-acceptor interactions,
the possibility of redistributing electronic density from one atomic grouping to another for
reasons of their involvement in the united system of π-conjugation [1, 4]. One of the most
effective factors is a relatively rigid shielding of these reaction centres by an atomic envi-
ronment and π-electron cloud of the aromatic macrocycle (Fig. 1). The extent of shielding,
in turn, can change under the influence of external effects. This phenomenon was given the
name of the macrocyclic effect (MCE) of porphyrins and metalloporphyrins [4–8]. MCE
affects the physicochemical properties and reactivity of H2P and MP by means of π-elec-
tronic and structural (steric) components. Their effects on various physicochemical pro-
cesses occurring in the molecule are, as a rule, different and nonsymbate one to another [8].
Figure 1 Atomic and π-electronic shielding of the reaction centre MN4 of metalloporphyrins MP
(a), X(MP) (b), (X)2MP (c) [4–5].
conformational conversions of H2P molecules can cause secondary changes due to the
asymmetric redistribution of π-electronic density in them. The emerging polarization of
molecules on the whole and separate chemical bonds in particular, e.g., the chemical acti-
vation of NH bonds, can drastically change the reactivity of compounds.
Of special significance is the role of the medium, in particular, the nature of solvent
(if processes are run in a liquid-phase system), which is capable of significantly intensifying
or inhibiting the polarization processes in H2P molecules.
Thus, another important factor, which along with atomic electronic shielding (MCE),
determines the reactivity of porphyrins in vivo, is polarization of the molecule and, as a rule,
chemical activation of NH bonds in its coordination core. This conclusion has been evident
earlier. It follows, e.g., from the reaction mechanism of complexation of porphyrins with
d metal salts [1, 5, 25], the limiting stage of which is namely polarization by the solvent and
dissociation of NH bonds of the H2P macrocyclic ligand.
An important problem, which needs to be solved, is to elucidate the relation between
the nonplanar structure of the porphyrin macrocycles and the chemical activity of NH bonds
in the coordination core of these molecules. As a particular case of a jumpwise change of
the properties under the influence of conformational regroupings, the phenomenon of NH
activation is very important for modelling and studying the processes involving H2P in vivo,
where in the course of acceptance of metal ions by bioporphyrins NH activation is per-
formed also under the influence of the donor-acceptor protein-lipid environment.
The first group includes many porphyrin bioligands, e.g., chlorophylls, blood porphy-
rins, cytochromes [5, 27], as well as their synthetic β- and meso-substituted analogues, e.g.,
β-octaethylporphin {I, H2(β-Et)8P}, meso-tetraphenylporphin {II, H2TPP} and meso-
tetrapropylporphin {III, H2T(n-Pr)P}. These compounds are also called classical H2P [28],
R1 R R1 R X
N
R1 R1 X
NH N meso-position NH N NH N
R R R R N N
N HN β−position N HN N HN
R1 R1 X
N
R R1 R X
R1
because the properties they exhibit totally correspond to those traditionally known in por-
phyrin chemistry. The second group are dodeca-substituted (peripherally highly substitut-
ed) H2P (compounds IV–VII) [28–33] and some other strongly nonplanar molecules. The
third group comprises rigid highly aromatic porphyrazines (compounds VIII–IX) [34–37],
as well as tetrabenzoporphyrins {e.g., H2TBP (X)}, having substituents in condensed rings
but not in meso-positions [38]. Meso-substituted tetrabenzoporphyrins (e.g., IV) belong to
the second group of compounds, dodeca-substituted porphyrins [31–32, 38–39]. H2P with
nonclassical properties constitute the second and third groups of compounds [26, 28].
• the structure of the macrocycle is changed to become rigid (aromatic) or, vice versa,
strongly nonplanar;
• electron-acceptor groups or groups of various electronic nature which lead to the
push-pull-type polarization of π-chromophore [40–43], are present in the molecule;
• the properties of the medium, contributing to the polarization of NH bonds of the H2P
molecule, are changed by means of solvation and chemical interactions, e.g., with
molecules of electron-donor solvents [4–5], or by adjusting the reaction NH centres of
porphyrins to the reaction centres of the medium in the solid phase during the sorption,
immobilization etc. [44–45].
At least two out of the three conditions, each of which is necessary but not sufficient,
are required to be satisfied to achieve the NH activity of H2P.
Figure 2 Main types of nonplanar conformations of H2P and their complexes ( and assume the
location of atoms over and under the initial plane of the macrocycle, respectively) [16, 19].
Chemical Activation of Porphyrins in Coordination Core Reactions 173
Property
Ì HOMO–LUMO energy gap
Ê H2P redox properties
Ì stability of MP complexes
Figure 3 Porphyrin structure–property dependences: group 1, planar classical H2P; group 2, non-
planar nonclassical H2P; group 3, planar nonclassical H2P.
The properties related to the reactivity of the coordination core of the H2P molecule
on the whole and the chemical activity of NH bonds in particular are especially dependent
on the peculiarities of the macrocycle geometric structure [58]. Herewith, both porphyrins
with the rigid π-macrocycle (compounds VIII–X) and strongly nonplanar compounds
IV–VII acquire more pronounced acidic properties [35, 61–62], and in a some cases be-
come capable of interaction with molecules of electron-donor solvents [30, 32, 55], which
is exhibited both in their solvation characteristics [32,63–64] and in the change of the re-
activity of these ligands in the complexation reaction (1) with salts of d metals in elec-
tron-donor media [36, 52, 55] as compared with predominantly planar porphyrins proper
(compounds I–III). As the result, there arises a nonlinear dependence of the change of some
properties of H2P on the change of structure of the molecules (Fig. 3).
–2Solv
H2P + MX2(Solv)n–2 [H2P … MX2(Solv)n–4]# MP + 2HX + (n–4)Solv. (1)
of studies up to the present. As a rule, inclusion of acid-base centres into a unified system
of conjugation leads to the symbateness in the change of the chemical activity of these cen-
tres [69]. It is observed as in the case of five-membered aromatic heterocycles, when an in-
crease of acidity of NH groupings occurs against the background of a decrease of the main
properties of nitrogen tertiary atoms (see the sequence pyrazole – triazole – tetrazole in
Table 2). A similar situation is also observed in the case of complex macroheterocyclic
Table 2 NH acidity (pKa) and N basicity (pKBH+) of five-membered nitrogen-containing hetero-
cycles in DMSO and H2O [66–67].
N N N N
Heterocycle N N N
N N N N N
H H H H H
Pyrrole Pyrazole Imidazole 1,2,3-Triazole Tetrazole
compounds, in particular, porphyrins with predominantly planar structure (I–III). Thus, in-
troduction of one electron-acceptor chlorine atom each into each of the meso-phenyl rings
of the H2TPP molecule (II), even into meta-positions, increases the strength of this NH acid
in the elimination of the first proton ((pKa1, equation (2)) almost 1.5-fold in a DMSO me-
dium. Simultaneously, the main properties (measured in acetonitrile) weaken in the total
process of elimination of two protons from porphyrin dication (pK3,4, H4P2+ ' 2H+ +
H2P) by two orders [61, 69]. Even the fact that the acidity and basicity of compounds was
assessed in different solvents does not disturb the general pattern of acid-base interaction
(equations (2)–(5)).
KK11
H2P(solv) H+(solv) + HP –(solv) , (2)
KK21
HP –(solv) H+(solv) + P2–(solv) , (3)
KK31
H4P 2+(solv) H+(solv) + H3P+(solv) , (4)
KK41
H3P +(solv) H+(solv) + H2P(solv) . (5)
Interference of spatial factors related to the change of the geometric structure of the
π-macrocycle very often results in the simultaneous increase of NH-acidic and N-basic
properties of nonplanar porphyrins. For instance, already in passing from unsubstituted por-
phin (H2P) to slightly more nonplanar H2TPP (II) both the rise of acidity and basicity of the
176 D.B. Berezin and B.D. Berezin
molecule are observed simultaneously (Table 3). This phenomenon acquires a much more
pronounced character in strongly nonplanar dodeca-substituted porphyrins IV–VII. Takeda
and Sato [29] were among the first to explain the simultaneous enhancement of both acidity
Cl
Porphyrin NH N NH N NH N
N HN N HN
N HN
Cl
Cl
and basicity of these molecules as the result of spatial deshielding and increased accessibil-
ity of atoms and chemical bonds of the coordination core of the conformationally mobile
macrocyclic ligand for chemical reagents. As it became clear later, this explanation is not
sufficient, as far from any disturbance of the planar structure of the H2P molecule chemi-
cally activates NH bonds in it (see Section 3). On the contrary, according to the classical
polarization views, a greater conformational mobility of the molecule should lead to a wors-
ening of the conditions of the distribution of π-electrons on the conjugated system and,
therefore, its localization on individual atoms and bonds, in particular, on coordination core
atoms. This conclusion is consistent with the increase of the basicity of tertiary atoms of
nitrogen in the molecule of strongly nonplanar H2P ligand; however, it does not explain but,
vice versa, contradicts the enhancement of NH acidity. What is more, 1H NMR studies of
a number of nonplanar porphyrins, whose conformations in the solid phase can be charac-
terized as ruffled [70], have shown that rather significant extraplanar distortions of these
molecules lead to a mere 5% decrease of the π-electron ring current. Herewith, the dou-
ble-dipole model was used, which proved itself to be good in the assessment of ring currents
of planar porphyrins, including chlorophyll ligand molecules [9, 10, 71–72]. Apparently,
an insignificant decrease of the ring current is a specific feature of nonplanar porphyrins.
Its presence makes it possible to assume that π-chromophores of such nonplanar molecules
shall be as readily polarized under the influence of the electronic effects of substituents,
light quanta and chemical reagents.
Indeed, a recent work [22] has found that dodeca-substituted porphyrins with an in-
creased chemical activity of NH bonds possess, despite the highly symmetric structure,
rather significant dipole moments (µ), while in planar H2P pertaining to a similar structural
group this value is equal to zero (Table 4).
Thus, practically planar tetraphenylporphin (II) has the
zero dipole moment, whereas the symmetric octa-
β-alkyl substitution of this molecule, leading to the non-
R planar structure of the π-system in β-octaethyltetraphe-
R nylporphin {V, H2(β-Et)8TPP}, increases this
NH N characteristic of H2P up to 1.2 D. In the case of porphy-
rins (XIV–XV), which are more planar as compared
N HN with compound V but are additionally polarized due to
R R1 the asymmetric substitution, the dipole moments are
R R1 even higher.
The most nonplanar from asymmetrically substituted
H2P (compound XV) possesses the maximal dipole mo-
XIV. R = Et, R1 = H; ment (1.89 D). An unusual fact is that the dipole mo-
XV. R = R1 = Et; ment of saddle-distorted ligands is directed
orthogonally or almost orthogonally to the plane of H2P
(see Fig. 4).
Ghosh et al. [23] acknowledge that the entire body of changes in the properties of H2P
molecules in passing from the planar structure of the macrocycle to a strongly nonplanar
conformation can not be explained by only the difference in the geometry of the molecule.
Based on the above said, it can be concluded that the cause of changes in the physicochem-
ical properties and reactivity of nonplanar porphyrin ligands, in particular, those caused by
an increased chemical activity of coordination core’s NH bonds, is internal polarization of
nonplanar molecules. It also determines many of their photophysical properties [22, 49–50]
and reactivity [29–30, 52].
Polarization is a consequence of conformational conversions of H2P nonplanar mole-
cules. Such structural changes are associated with the asymmetric change of bond lengths
and valence angles, and lead to the redistribution of π-electronic density in it. Thus, they
are the cause of secondary electronic changes in the molecule, which affect the reactivity
of its reaction centres (N4H2, MN4 etc.).
Polarization of H2P molecules under the influence of the electronic effects of substit-
uents has an additional positive effect on the activation of NH bonds. In the case of highly
aromatic molecules, such as tetrabenzoporphyrins (X) and porphyrazines (VIII–IX), the
Figure 4 Structural causes of the formation of the orthogonal dipole moment in saddle-distorted
porphyrin ligands by the example of H2(β-Et)8TPP (V) [22, 33].
178 D.B. Berezin and B.D. Berezin
planar structure creates the most favourable conditions for their polarization by the light
wave, which is exhibited in the change of the photophysical characteristics of the com-
pounds. Besides, this is favourable for internal polarization of H2P by means of substituents
of various electronic nature and for the emergence of the push-pull effect at the time when
simultaneously donor and acceptor atomic molecular functions (compound XVI) occur
[40–43, 71].
The influence of the
push-pull effect proper on
the reactivity of porphyrin
coordination cores and on
NH bonds practically has NH N
not been studied. Only the (M e)2N C C C C NO
optical properties were in- N HN
vestigated, including the
nonlinear optics properties,
as well as such parameters
of “push-pull” porphyrins XVI
as polarizability β and di-
pole moment µ [40–43]. The highest polarizabilities of the order of 5000·10 –30 esu and
dipole moments of more than 10 D were found in “push-pull” porphyrins carrying elec-
tron-donor (e.g., NMe2) and electron-acceptor (e.g., NO2) groups in phenyl rings, which
are linked with meso-positions of the macroring via an ethynyl spacer [41]. A later work
[43] showed the most efficient among π-linking spacers (–CSC–, –N=N–, –CH=CH–,
–CH=N–) to be exactly the ethynyl spacer.
The consequences from introducing substituents of any electronic nature – donors or
acceptors – into the H2P molecule (as a rule, with a chemically low active NH bond) have
been studied well [1–3, 9–11]. A change of reactivity of the coordination core of such mol-
ecules is observed in complexation reactions of H2P with solvatosalts of d metals
[MX2(Solv)n–2] (1) and acid-base interaction ((2)–(5)), as well as of dissociation (6) and
metal exchange (7) in MP porphyrin complexes. Peripheral substituents of the same elec-
tronic nature in H2P molecules affect different processes differently. The cause of this is
oooo
MP + nH·Solv+ + nX – H4P2+ 2X – + MX2(Solv)n + n–4HX, (6)
the difference of the mechanisms of the above listed reactions and the different degree of
involvement of NH bonds in these mechanisms. Thus, the complexation reaction (1) has a
strong dependence on the donor-acceptor properties of the solvent, in which it is run, and
for H2P with a chemically inactive NH bond (e.g., para-substituted derivatives
{H2(p-R)4TPP} of tetraphenylporphin II) is accelerated by electron-donor substituents in
phenyl rings, and is slowed down by electron-acceptor substituents [1–3, 73]. The rate of
reaction (1) in this case is determined by the strength of reacting solvatosalts of the com-
position [MX2(Solv)n–2] [74–78] and is, therefore, maximal in carboxylic acids (AcOH,
C2H5COOH) and other weakly coordinating solvents, such as esters, ketones, nitriles and
alcohols [5]. Reaction (1) of porphyrins with a chemically active NH bond is, in contrast,
accelerated by electron acceptors, which are simultaneously coordination-core activators,
Chemical Activation of Porphyrins in Coordination Core Reactions 179
and slowed down by electron donors [5, 31, 32, 35]. The highest catalytic effect for reaction
(1) is achieved in a medium of polar electron-donor solvents, also activating NH bonds in
the H2P molecule. A characteristic example are porphyrazines (VIII–IX, XVII–XVIII)
substituted in β-positions by electron-acceptor halogen atoms [35, 54]. From the data of Ta-
ble 5, it follows that the rate constants (kv) of reaction (1) of NH-active macrocycles with
ZnAc2 increase with the rise of electronegativity of β-substituents, as well as at an increase
Table 5 Kinetic parameters of the formation of zinc complexes with tetraazaporphin (VIII) and its
halogen derivatives (IX, XVII–XVIIII) in AcOH–Py mixtures, CH2P = 2·10 –6 mol/l, CZnAc2 =
3.3·10–4 mol/l [54].
N N
-ɨɛɪ
σ πinv .
-ɨɛɪ
σ
πinv .
M -ɨɛɪ.
πinv
σ
-ɨɛɪ.
πinv
σ
N N
the functional groups located in the phenyl rings of the macrocycles: –NO2, –NH3+, –Cl,
–Br, –COOH < H < –CH3, –OH, –OCH3 [79–80]. If a substituent enters into direct con-
tact with the π-macroring when being in β- or meso-positions of the MP molecule, reaction
(6) proceeds with account of the electronic effects of substituents ±I and ±C, as well as the
electronic effects of coordination – σ-donor–acceptor interaction (N → M), forward dative
π-bond (πfwd, N → M) and inverse dative π-bond (πinv, N ← M) (Fig. 5) [1, 4–5]. A number
of dissociation rates in such MP can change and even reverse depending on the combination
of the above listed effects [80].
Studies of the effect of basicity of substituted meso-tetraphenylporphin (II) (Table 6)
on the rates of metal exchange reaction (7) in the {CdP/Zn(ClO4)2/Py/0.5M H2O} system
within the temperature range of 288–310 K has shown that electron-donor groups in the
phenyl rings of H2TPP accelerate the metal substitution process [81–82].
Thus, the true reaction rate (kv298, l·mol–1 ·s–1) in passing from p-CN to p-OCH3 de-
rivative increases from 1.3 up to 10.2. The metal exchange rates in the considered cases are
600–6000 times higher than the rate of complexation of zinc perchlorate with respective
ligands (Table 6).
Table 6 Kinetic parameters for the metal exchange reaction of {CdP/Zn(ClO4)2 /Py/0.5M H2O} of
meso-substituted CdTPP [81].
The rates of acid-base interaction processes involving the coordination core of the H2P
molecule (equations (2)–(5)) react to the electronic nature of substituents in the porphyrin
macrocycle in a way traditional for this type of reactions, i.e., the basicity increases and the
acidity goes down at the drop of π-electronic density in the macrocycle [61].
Even the introduction of polarizing substituents into H2P molecules with inactive NH
bonds (compound II) strongly changes the physicochemical properties and reactivity of the
compounds. Thus, the substitution of one of β-positions in the H2TPP molecule (III) stabi-
lizes two trans-tautomers [33, 72], increases the dipole moment of the molecule up to
Chemical Activation of Porphyrins in Coordination Core Reactions 181
approximately 7 D [33], due to the charge-transfer (CT) interactions cardinally changes the
photophysical pattern of the process in its main S0 and excited (S1, T1) states [83].
Benzene Ethanol
interaction assumes a weak sensitivity of compounds to the nature of the solvent, e.g., in-
significant changes in their photophysical properties [48, 60, 71], high chemical and ther-
mal stability of most H2P in solutions [1–5, 64]. At the same time, the nature of solvent
plays a significant role in the change of reactivity of N4H2 coordination core, in particular,
in reactions of acid-base interaction [61] and complexation [75].
Owing to the polarization of molecules of NH-active porphyrins [22, 71], their reac-
tivity in solutions of electron-donor solvents significantly increases; this applies not only
to nonplanar H2P possessing the intracyclic reaction centres more accessible for reagents
[30, 32, 52], but also to planar porphyrins with rigid π-chromophore [53–55]. As we have
already mentioned, the rise of aromaticity of the H2P molecule assumes not only the shield-
ing of intracyclic reaction centres at the cost of the MCE structural component, but also the
redistribution of the π-electronic density by means of the MCE electronic component,
which entails the pronounced chemical activity of particular bonds [7–8]. In a number of
chemical processes, the latter of the two factors prevails. At the same time, the solvation
characteristics of planar and nonplanar H2P with a chemically active NH bond differ no-
ticeably. Nonplanar H2P are well soluble in organic solvents irrespective of the type of
bond, as, e.g., N-substituted porphyrins XIX–XXI with a chemically inert NH bond [85],
whereas planar ones are still little soluble, especially in weakly proton-donor media (see
Table 7) [86].
Thus, the solubility of predominantly planar compounds I–II with a chemically inac-
tive NH bond in benzene decreases by 1–3 orders with the rise in the macrocycle rigidity
(compounds X, IX, XXII). At the same time, for planar NH-active compounds the value of
S298 is always higher in electron-donor solvents (e.g., in pyridine) as compared with uni-
versally solvating ones (Table 7) owing to the additional specific interaction of the Solv
molecule with NH bonds of the coordination core of the H2P molecule. In a medium of
weakly proton-donor solvents (EtOH), porphyrins, in contrast, are very weakly soluble, in-
cluding at the expense of depolarization of NH bonds in them. Indeed, in some cases con-
sidered the solvent can regulate the extent of chemical activity of these bonds. Partially
delocalized NH bonds of these compounds experience an additional activation (polariza-
tion) in a medium of electron-donor solvents [32] and, on the contrary, are completely lo-
calized already in a weakly proton-donor medium (carboxylic acids) [31]. These structural
changes are reflected in a characteristic jumpwise change of the complexing properties of
NH-active porphyrins in solvents of the types considered, and are also seen in the EAS [30,
55] (see Sections 1.2.1 and 1.3).
Consider schematically the mechanism of action of electron-donor (Py) and elec-
tron-acceptor (AcOH) solvents on the state of the coordination core (N4H2) of H2P mole-
cules of various structural groups: porphyrins proper (classical) and their rigid, or,
conversely, strongly nonplanar nonclassical derivatives (Fig. 6) [31].
In molecules of classical H2P (compounds I–II) [28] the macrocycle is almost planar
[13, 65], so the access of solvent molecules to –N= and –NH– solvation sites is hindered.
Besides, NH bonds in such molecules are polarized very weakly. Both of these factors lead
to a situation, when the interaction of NH groupings with molecules of electron-donor Py
is almost absent in the main state. Thus, a change of enthalpy in the transfer of H2TPP (II)
from inert solvent benzene to electron-donor Py, which characterizes the change of relative
solvation, is very small (∆trH0 = –1.9 kJ/mol) [7, 64]. In this case the NH bond remains
totally localized (is not activated), and the reaction proceeds with difficulty.
Due to the low basicity, as well as owing to structural causes, the molecule of AcOH
enters into only a very weak acid-base interaction with tertiary –N= atoms of the classical
Chemical Activation of Porphyrins in Coordination Core Reactions 183
meso
meso meso
meso
Figure 6 State of the coordination centre of porphyrin molecules with a chemically inactive group
(a) and a chemically active NH bond (b) in proton-donor (AcOH) and electron-donor (Py) media; b1,
nonplanar H2P; b2, planar H2P; , transfer of π-electronic plane in the macrocycle [31, 39].
H2P molecule. The H associate formed (Fig. 6, a) is, apparently, characterized by a low de-
gree of proton transfer, because it is not registered in the EAS. Nevertheless, protons in this
associate act as electron acceptors, decreasing the electronic density in the macrocycle and,
thus, promoting the rather weak polarization of the NH bond, e.g., in reaction (1).
This mechanism works only for predominantly planar porphyrins with a moderate
rigidity of the macroring. In the case when the macrocycle acquires a nonplanar conforma-
tion under the influence of multiple substitution at the periphery of the molecule or in any
other way [8, 16–17, 19, 87] (see also Section 1.4), the basicity of its tertiary nitrogen atoms
sharply increases, and the action of AcOH leads to partial or total protonation of these N
atoms (equations (4) and (5)) and their blocking as coordination centres in reaction (1)
(Fig. 6, b1).
Recently, it has been found that
R1 R R1 strongly distorted dodeca-substi-
N
tuted H2P, e.g., compounds
R1 R1 V–VII, in their titration in toluene
N-X N NH N solutions by weak organic bases
N
R R N form H complexes (or proton-
N HN
N HN transfer complexes, PTC (equa-
R1 R1 N tions (8)–(9)), which are charac-
terized by a rather profound
R1 R R1
transfer of NH proton to the mol-
XIX. R = H; R1 = Et; X = Me XXI. ecule of electron-donor solvent
XX. R = Ph; R1 = H; X = Me and, for this reason, are clearly
XXI. R = X = Ph; R1 = H seen in the EAS [30, 55, 85]. This
fact makes it possible to assert
that the NH bond in them is noticeably polarized. The distortion of the macrocycle increases
the accessibility of this bond for the attack by the Py molecule, and the formation of the H
complex with proton transfer to pyridine additionally polarizes the bond, bringing it closer
to the state characteristic of the transition state of reaction (1). Indeed, reaction (1) with non-
planar H2P proceeds very slowly in AcOH and easily in Py (see Sections 1.3 and 1.4). The
184 D.B. Berezin and B.D. Berezin
difference in complexation reaction rates in these solvents is up to more than seven orders
of magnitude [8, 51, 57, 85].
Another group of nonclassical H2P – most often these are the aza derivatives of por-
phin [26, 28, 34–37], such as compounds VIII–IX, XVII, XVIII, XXII – is characterized
(Fig. 6, b1) by a high aromaticity of the macrocycles and an enhancement of the macrocyclic
effect (MCE) [6–8]. The MCE structural component [8] contributes to the shielding of –N=
and –NH= centres from the penetration of reagents. This is one of the causes of a decrease
of the main properties of tertiary nitrogen atoms of tetraazaporphin (H2TAP, VIII) and tet-
ra-β-bromotetraazaporphin {H2(β-Br)4TAP, IX} [61]. For this reason, there is almost no
interaction of them with AcOH. The bridge (meso-) aza atoms are involved in only a weak
associative interaction with AcOH molecules, because their unshared electron pairs are in-
volved to a significant degree in the conjugation with the π-system of the macroring [1, 35].
As for the state of the NH bonds, they are completely localized in a solution of AcOH,
which does not contribute to the exchange of NH protons of H2P with the medium.
Simultaneously, intramolecular effects, such as the enhancement of the MCE π-elec-
tronic component [8], evoke a significant polarization of NH bonds in molecules of rigid
H2P. This polarization follows from an increase of their acidity [61], which for azaporphy-
rins in a DMSO medium is 10–12 orders higher as compared with classical H2P. Protona-
tion of NH bonds facilitates the formation of the PTC of rigid macrocycles with molecules
of electron-donor solvents, where these bonds are already partially delocalized [55] (Fig. 6,
b2). Formation and stability of PTC are considered in more detail in Section 1.2.
Apparently, the considerable potential of activating NH bonds and modifying the prop-
erties is in the effect of the solid-phase environment on porphyrins. Unfortunately, these
issues are far from being understood at present, and there are only separate works on the
topic in the literature [44–45, 88–89]. All these publications note the catalytic effect of the
solid phase in H2P or MP formation reactions. Tsukahara and Suzuki [44] studied the com-
plexation ability of H2TPP (II) sorbed on Kieselgel 60HR. They showed that its reaction
with Zn2+ ions sorbed on SiO2 proceeds much faster in a sequence of weakly polar or non-
polar solvents (CCl4 > toluene > 1,2-dichloroethane > acetone) as compared with the reac-
tion in solution. At room temperature, reaction (1) is complete in only 5 min in all solvents
except weakly coordinating acetone at an H2TPP–salt concentration ratio of 1–18/260,
whereas in solution it runs slowly and at increased temperatures [75]. This indicates the de-
localization of NH bonds under the influence of sorbent’s active centres, as well as the spe-
cific adjustment of activated reaction centres of the solvatosalt to the reaction centres of
porphyrin. Another example of solid-phase activation of NH bonds was observed when the
chlorophyll ligand and its derivatives immobilized on water-soluble polymer material
(polyvinyl alcohol) reacted in a glacial acetic acid and its aqueous solutions with Cu(II),
Zn(II) and Co(II) acetates at rates not lower and even higher than in the absence of polymer
carrier [45]. The above mentioned facts may come as a surprise because at first glance the
solid-phase and, moreover, polymer environment strongly hinders sterically the attack of a
reagent, e.g., the solvatosalt in reaction (1) on the coordination core of the H2P molecule.
Letts and Mackay [88] studied the complexation reaction of H2TPP with Cu(II), Mg(II),
Mn(II), Zn(II) and Co(II) ions in a water-emulsion system in the presence of cationic or an-
ionic surfactants, noted both the facts of its inhibition and acceleration by various deter-
gents. In the literature, there are facts of heterogeneous catalysis of not only MP formation
reactions (1) but also of condensation processes to form H2P ligands. Thus, for instance,
Onaka et al. [89] observed a more than twofold increase of the yield of tetra-meso-alkylpor-
phyrins (35–45%) in a system containing porous montmorillonite K10.
Chemical Activation of Porphyrins in Coordination Core Reactions 185
Kk1
H 2P + B k –1
H2P·B, (8)
k21
K
H2P·B + B k –2
H2P·2B. (9)
λ, nm λ, nm
Figure 7 Change of electronic absorption spectra in the course of formation in the C6H5Cl–DMSO
system (a) and partial destruction in the C6H6 –tert-BuNH2 system (b) of the PTC of β-tetrabromo-
tetraazaporphin (IX) [90].
as compared with media, where the formation of PTC with porphyrins is not registered
spectrally, i.e., the extent of interaction of protons of NH groups with molecules of the sol-
vent is lower. No such cases were observed for strongly nonplanar H2P.
Interactions of planar derivatives of porphin, in particular, porphyrazines, have been
studied in the literature in greatest detail [55, 90–99]. Proton-transfer complexes were first
spectrally (EAS) observed in 1961 by Whalley in pyridine solutions of H2(β-Ph)8TAP
(XXIII) and called pyridine salts [91], which is not quite correct as the PTC is not a salt but
represents a product of incomplete acid-base interaction [92]. The scheme of complete
transfer of proton from acid HA to base B has the form:
K1 K1 K1 K1
HA + B AH…B Aδ– …H…Bδ+ A –…HB+
K1
A – ·HB+ A – + HB+, (10)
where AH…B is a molecular complex with the H bond, Aδ– …H…Bδ+ is an intermediate-
type complex with the delocalized H bond, A –…HB+ is an H-bound ion pair, A – ·HB+ is a
contact ion pair, A – + HB+ are complete proton-transfer products. The H2P with partial pro-
ton-transfer to the base molecule (B or Solv) represents, depending on the strength of acid
and base, permittivity and solvating power of the medium either an H associate (AH…B)
or an ion pair (A –…HB+) or else an intermediate complex with the delocalized position of
proton (Aδ– …H…Bδ+) (Fig. 8) [90]. Formation of such kinds of products is difficult even
in solution and is possible only in the case of weak acids and bases with spatially shielded
reaction centres [93].
The kinetics and mechanism of PTC formation with porphyrazines (VIII–IX,
XXII–XXV) have been studied in detail in a series of works [55, 90, 94–99].
Based on the spectral and kinetic data, a probable structure of the PTC of the type of
Chemical Activation of Porphyrins in Coordination Core Reactions 187
Figure 8 Assumed structures of the porphyrin PTC with a weak base molecule B: (a) ionized, (b)
H-bound [35, 55, 90].
H2P·2B (Fig. 8) has been proposed. It has also been shown that the formation of kinetically
stable proton-transfer complexes is observed only if NH-acidic properties of H2P are suf-
ficiently strongly pronounced, the electron-donor ability of the base is moderate and the per-
mittivity of the medium is high, as, e.g., in DMSO or inert solvent–DMSO mixtures as well
as in a DMSO–Py medium [90] (Table 8). Otherwise, e.g., in most media with low permit-
tivity based on aliphatic amines {Et2NH, n-BuNH2, tert-BuNH2PhCH2NH2, Et3N,
n-Bu3N, Pip}, PTC are subjected to gradual destruction to colourless products (Table 9, Fig.
7, b). Both the PTC formation and destruction rates are controlled by steric factors. The
more bulky the amine molecule and the more shielded its reaction centre, the more complex
the proceeding formation and destruction processes of PTC are (Fig. 9, Table 9). The ob-
served phenomena are well described within the framework of the PTC formation and dis-
sociation mechanisms [55, 90].
Table 8 Kinetic parameters of acid-base interaction of H2TAP derivatives with DMSO in C6H5Cl
(CH2TAP = 0.5 · 105 mol/l; C 0DMSO = 3.88 mol/l).
Table 9 Kinetic parameters of acid-base interaction of H2TAP derivatives with N bases in benzene
{CH2TAP = 0.4–0.5·10–5 mol/l}.
Figure 9 Dependence of log k of the destruction of H2(β-Br)4TAP (IX, filled dots) and
H2(β-Cl)4TAP (XVII, empty dots) in benzene in the presence of basic nitrogens at 298 K.
eff
DMSO
Figure 10 Dependence of log keff on CDMSO for reaction (1) of formation of Mg(β-,4- NO2Ph)8TAP
in benzene at CMgAc2 = 1.8·10 –4 mol/l; T = 308 K (1), 318 K (2), 328 K (3).
λ, nm
Figure 11 Change of electronic absorption spectra for H2(β-Br)8TPP (VII) in toluene in the
presence of piperidine, CH2P = 4.23·10 –4 mol/l; Cpip, mol/l: 0 (1), 0.008 (2), 0.016 (3), 0.025 (4),
0.033 (5), 0.041 (6), 0.058 (7), 0.083 (8), 0.083 (9), 0.166 (10).
lg CHP- / CH2P
Figure 12 Graphic determination of the number of base B molecules participating in the equilib-
rium with H2(β-Br)8TPP (VII) in toluene: 1, Pip; 2, DMF; 3, Py; 4, DMSO; n = tan α ≈ 1.
Indeed, in a number of cases H2P molecules activated in this way enter more readily
into complexation reactions (1).
At the same time, it should be noted that the complexing ability of PTC in reaction (1)
is determined by its kinetic stability and can be both strongly increased and decreased with
the change of the concentration of the base in the same reacting system (Fig. 10).
The ability of strongly nonplanar dodeca-substituted porphyrins to interact with elec-
tron-donor solvents was found comparatively recently [29–30]. It was revealed that the
substitution of a weakly solvating (inert) solvent (benzene, toluene) by an electron-donor
solvent completely changes the structure and number of bands in the electronic spectrum
of the molecule (Fig. 11) [30]. The fact that the number of bands decreases, and the EAS
starts to resemble more the spectrum of a metal complex, indicates the formation of struc-
tures with a higher symmetry (D2h → D4h). Similar spectral changes were also observed in
the interaction of porphyrazines with bases [55, 90]. In contrast with porphyrazines, PTC
in dodeca-substituted porphyrins are formed in an equilibrium manner. Karmanova et al.
[30] calculated the stability constants of forming delocalized acid-base forms from the re-
sults of titration of solutions of H2(β-Br)8TAP (VII). The stability of PTC proved to in-
190 D.B. Berezin and B.D. Berezin
crease with the rise of polarity of the medium, even if the electron-donor properties of the
base weaken: Py < DMF < DMSO < Pip. In contrast with solutions of porphyrazines, the
breakdown of PTC proceeded in DMSO without the destruction of ligands at the addition
of N bases. Besides, it was found that in the course of titration of compound VII by the base
only a PTC of the type of H2P·B is formed (equation (8)). In particular, this follows from
the value of the slope of the dependence log CSolv = f (log CHP– /CH2P) (see Fig. 12).
Table 10 Stability constants (Kst) of proton-transfer complexes (PTC) and thermodynamic
parameters of acid-base interaction of H2 (β-Br)8TPP) (VII) with organic solvents [30].
NH HN
N N
II N HN
NH N
I NH N
N HN
Figure 13 Synchronous (I) and asynchronous (II) routes of N–NH tautomerization of porphyrins.
naphthyl fragment decreases the rate of the process to 0.004·104 s –1, i.e., by about three
orders of magnitude. NH tautomerization does not change the properties of the H2P mole-
cule significantly, because the rearrangement of the π-conjugation contour in the conver-
sion of tautomers leads neither to the change of symmetry nor to the activation of the already
available reaction centres or the emergence of new reaction centres. .
Sophistication of the H2P structure can enrich the reactivity to a significant degree and
often provides a possibility for the existence of the molecule in various molecular forms
depending on the nature of the medium, on the presence of some or other reagents in solu-
tion. These manifestations determine the whole range of practically useful properties of the
molecules [110]. One of the examples of such compounds are H2P carrying active (one or
several) hydroxy or amine groups in meso-positions. In solutions, these compounds enter
into tautomeric conversions of, respectively, keto-enol or amine-imine type. As a rule, be-
ing in a nonionized medium, these compounds exist in keto or imine form, which is reliably
confirmed by spectral methods and follows from the data on their reactivity. Keto-enol
equilibria of oxoflorins have been studied the most [111–112]. Existing in nonionized me-
dia in a keto form, oxoflorins (XXVI) are subjected to tautomerization during the action of
strong acids (XXVIb), as well as in the course of coordination of these nonaromatic ligands
by double-charged metal ions (XXVIc), and form meso-hydroxyporphyrins (enols) (Fig.
14) possessing all the features of aromatic molecules. For this reason, oxoflorins are stron-
ger acids and bases as compared with H2P, they also enter more readily into complexation
reactions [111]. The conversion mechanisms of oxoflorons (Fig. 14) have not been studied;
however, it is evident that the process in this case involves chemically activated NH bonds.
Chlorophyll and its structural analogues are also subject to keto-enol tautomerization
involving the cyclo-pentane ring in accordance with scheme (11) [1]. As in the case of sim-
pler molecules, the tautomerization in chlorophyll a (H2Cl a, XXVII) can be shifted to-
wards ketone or enol, which is controlled by the nature of the peripheral substituents in the
moloooecule, the oo
OH
OH O O
NH H N NH HN
N N NH H N HX
MX2 HX
M -X N H HN
-2HX N HN -X N H HN
N N
H 3C C2H 5
NH N
H N HN
C H3
H 3C
H 2C H
H O
H 2C C O O C H3
C 20 H 39 O O C
XXVII
molecule, the nature of the medium, by other factors. Thus, enol (compound XXVII) is
formed not only in the medium of alcoholic alkaline [1, 113–114], but also in an
ethanol–glycerol (1:1) system and also in solutions containing a large amount of the second
component. The enol form of chlorophyll a (XXVI) is reactive. Thus, in an alkaline medium
in the presence of air oxygen it is unstable and reacts to break down the cycle (11).
NaOH, ROH O2
C C
(11)
- + - +
O O Na H3COOC O O O Na
H3COOC H H3COOC H
Inversion of one of the pyrrole rings in the formation of the tetrapyrrole macrocycle
of H2P leads to the formation of a new class of compounds – single-inverted analogues
of porphyrins [115–116] (compounds XXVIII–XXIX). They, in particular, compound
XXVIII, are characterized by an unusual type of tautomerization (12) associated with the
intramolecular transfer of one of the intramolecular NH protons (the form XXVIIIa) to the
external tertiary atom of the molecule (the form XXVIIIb:
K1
XXVIIIa + Solv XXVIIIb…Solv. (12)
A
3.0
ɚ
2.5
b
c A/6
2.0
1.5 a b
1.0
c
0.5
0.0
400 450 550 600 650 700 750 λ, nm
Figure 15 Electronic absorption spectra of compound XXVIII (c = 1.5·10 –5 mol/l) in organic sol-
vents: a, C6H6; b, DMSO; c, Py.
Table 11 Properties of organic solvents and forms of existence of compound XXVIII in solutions.
λ, nm
Figure 16 Steady state fluorescence spectra of compound XXVIII depending on the excitation
wavelength (λexc = 360–460 nm) in toluene (a), DMF (b) and Py (c). T = 298 K.
δ 21CH δ NH δ CH δ NH δ NH3
(internal) (β + Ph) (external)
Figure 17 Correlation of the value of Kt of compound XXVIII in reaction (12) with characteristics
of organic solvents: a) Kt = f(µ); b) Kt = f(DN); 1, DMF; 2, DMSO; 3, DMPU; 4, HMPTA.
atom 2N fixes in this molecule the conjugation contour, which corresponds to that of the
tautomeric form b of nonmethylated analogue of XXVIII. As the result, by the extent of
aromaticity and position of proton signals the macrocycle of XXIX is close to the tautomer-
ic form XXVIIIb (Table 12).
The increase of the acceptor number AN of the solvent relative to the donor number
DN indicates an increase of its proton-donor properties, so equilibria (13) and (14) can be
realized in a medium of solvents with DN < AN. As the result, a mixture of tautomer
XXVIIIa with single- (H2[2NH,21CH]TPP+) and double- ({H3[2NH,21CH]TPP}2+) proto-
nated forms of compound XXVIII is formed in solution (because the permittivities of these
media are, as a rule, not high), if the acceptor number of the solvent AN < 50 as in weakly
proton-donor propanol-1 (C3H7OH-1) and nitromethane (CH3NO2); or only the double-
protonated form is formed, when AN > 50, e.g., in a medium of acetic acid (AcOH) or tri-
fluoroacetic acid (TFA) (Table 11). According to the data of [115], the first proton in the
molecule of XXVII is added to the external, more sterically accessible nitrogen atom 2N,
and the second to the intracyclic atom 23N.
K1
H2[2N,21CH]TPP + HX {H2[2NH,21CH]TPP}+X – , (13)
K1
{H2[2NH,21CH]TPP}+X – + HX {H3[2NH,21CH]TPP}2+2X – (14)
The tautomers of compound XXVIII (a and b) are converted into each other in an equi-
librium manner during the change of the composition of the inert solvent–electron-donor
solvent system (12). Berezin et al. [124] obtained tautomerization equilibrium constants (Kt
= [b]/([a]·[Solv]2, where [a], [b] and [Solv] are the concentrations of tautomer a, tautomer
b and titrant, respectively, for process (12) in C6H6 –DMF, C6H6 –DMSO, C6H6 –HMPTA
and C6H6 –DMPU systems.
As seen from the data of Table 13, the stability of the tautomeric form b increases with
the increase of solvent’s polarity, which is characterized by the values of ε and µ, and its
donor strength expressed by the donor number DN. The values of Kt were found to correlate
the best with the values of µ and DN (Fig. 17, a and b). Apparently, the DN value of the
solvent determines the strength of the bond of the donor site of compound XXVIIIb with
the extracyclic NH proton, and the high dipole moment µ favours the shift of equilibrium
(12) to the right owing to the better stabilization of ionic particles, formed in the conversion,
Chemical Activation of Porphyrins in Coordination Core Reactions 197
by the solvent. At the same time, this correlation is observed only for solvents, whose donor
strength and polarity parameters are within a certain interval favourable for the formation
of tautomer b.
It should be noted that for bulky HMPTA the value of Kt is much lower (Fig. 17) than
it could have been expected based on the values of its µ and DN, which are maximal in the
sequence of solvents considered (Table 13). Evidently, this is explainable by the worsening
of the steric conditions for the solvation of the 2NH grouping of inverted porphyrin by the
bulky molecule of the solvent. Thus, the molar volumes (VM) for DMF, DMSO and DMPU
are within the limits of 100 cm3/mol, whereas VM for HMPTA is almost twice as high
(Table 13).
Table 13 Tautomerization constants of porphyrin XXVIII in a C6H6 –donor solvent medium and
characteristics of the solvents [124].
The value of Kt in the temperature range of 298–328 K within the limits of experimen-
tal error does not depend on temperature [124]. From this, it follows that tautomerization
process (12) proceeds with the activation energy E close to zero, which is consistent with
the data of quantum chemical calculations of tautomers a and b [109, 124, 127]. Thus, the
equilibrium state of (12) is largely controlled by entropy factors and, at an active participa-
tion of a solvent in the process, by solvation factors. Indeed, earlier studies showed tautomer
b to be stabilized by the molecule of an electron-donor solvent even in a crystal [117].
Several ways are possible to realize the tautomerization mechanism of inverted ana-
logues of porphyrins: within the involvement of a solvent by tunnelling or intermolecular
transfer of NH proton, as well as owing to its intramolecular transfer with participation of
a solvent [117, 128]. In the opinion of Furuta et al. [117], intramolecular proton transfer in
the course of the tautomeric conversion of compound XXVIII is performed owing to the
emergence of a strongly nonplanar conformation, in which the inverted pyrrole ring is in-
cluded towards the coordination core, and the inner and outer nitrogen atoms are brought
closer to each other. This treatment of the mechanism appears to be open to question, as the
reaction centres of a tetrapyrrole molecule involved in proton transfer are too far away one
from the other to enter into direct interaction, and the approximation of these centres to at
least a distance of an effective H bond (≤3 Å) should strongly destabilize the molecule.
According to the data of quantum chemical calculations and XDA data, both tautomers of
compound XXVIII are almost planar and, besides, close energy-wise molecules. The
difference in energy of the two tautomers is 3.4–5.7 kcal/mol according to the data of
[109, 127].
The strong dependence of the value of Kt on the nature of solvent, the evident impor-
tant role of the entropic contribution to the Gibbs energy of the tautomerization process, as
well as the fact of the involvement of two solvent molecules in the proton transfer found by
us (it follows from the values of tan α of the indicator dependence [log(Cb/Ca) = log (CSolv)
198 D.B. Berezin and B.D. Berezin
Figure 18 Dependence of the indicator relation (logCb/Ca) of the concentration of the tautomeric
forms of compound (II) in benzene on the logarithm of the concentration of the donor solvent
(log Csolv): 1, DMF; 2, DMSO; 3, DMPU; 4, HMPTA. tan α = 2.
N N
CH HN CH HN
Solv
- H Solv+ - iso
K ɢɡɨ
NH N N N
K1
a a1
-N Solv ...HN
CH N CH N
H Solv+
N HN K2 N HN
b1 b
(16)–(18). Within the framework of this approach, the participation of two molecules of
electron-donor solvent in the course of tautomeric process (12) is well explained.
K1 = [a1]/([a]·[Solv]), (16)
K2 = [b]/([b1]·[Solv]. (18)
Thus, inverted analogues of porphyrins, such are compound XXVIIIa, also pertain to
tetrapyrrole macrocycles with chemically active NH bonds (see the mechanism). Tautomer
XXVIIIb is less aromatic and, therefore, possesses no NH activity.
of δNH in porphyrins can be caused by two main causes – activation of the NH bond with
abstraction of protons by a donor solvent from the region of their shielding by the ring cur-
rent of the π-system and disturbance of the planar structure of the molecule [72]. The type
and depth of nonplanar conformation varies in various H2P [19], so it would not be correct
to compare the values of δNH of these compounds to make a judgement on the delocaliza-
tion of the NH bond. When substituting the solvent, the conformational composition of H2P
molecules does not in fact change [14], and the NH bond can only be activated, so to assess
the latter, one should use namely the value of the relative chemical shift of the signals of
NH protons, ∆δNH (19). It is positive, if porphyrin has chemically active NH bonds, and
negative in the cases when the bonds, as in classical H2P, can not be activated by elec-
tron-donor solvent (Table 14). According to the 1H NMR spectral criterion, the sequence
of chemical NH activity of H2P considered is as follows: IV << VIII < VI < VII. NH bonds
are minimally localized in weakly nonplanar porphyrin IV. Polarization of NH bonds in-
creases in transition to rigid nonclassical H2P (compound IX) and is maximal in strongly
nonplanar dodeca-substituted porphyrins (VI–VII).
200 D.B. Berezin and B.D. Berezin
Table 14 Effect of a solvent on the position of NH proton signals (δ NH, ppm) in 1H NMR spectra
of porphyrins.
The values of the constants given in Table 15 make it possible to determine compounds
I–II as classical H2P, and compounds IV–VII and VIII–X as nonclassical ones, and indi-
cate an increase of the chemical activity of NH bonds in the following sequence of porphy-
rins pertaining to groups (2) and (3) (Fig. 3): X < VIII < IX and IV < V < VI < VII. From
the data of Tables 14 and 15, it follows that polarization of the H2P molecule by peripheral
substituents, in particular, those of opposite electronic nature (the push-pull effect) enhanc-
es the chemical activity of NH bonds. This is well seen by the example of H2(β-Br)8TPP
(VII), carrying eight electron-acceptor functional groups in β-positions of the molecule, as
well as compounds VIII–IX (Table 15).
Chemical Activation of Porphyrins in Coordination Core Reactions 201
Table 15 Dependence of the rate of reaction (1) of porphyrin with Zn(II) and Cu(II) acetates (kv,
l ·mol–1 ·s–1) on the donor-acceptor nature of a solvent.
AcOH Py
(a)
Csalt = 0.005 mol/l.
Table 16 Effect of the nature of porphyrin on the entropic characteristics of complexation reaction
(1) with Zn(II) and Cu(II) salts in C6H6 and DMF.
Thus, for instance, during the change of the concentration of DEA in the
H2TPTBP(IV)–ZnAc2 –Py–DEA system from 0 to 1 mol/l, the true reaction rate constant
increases from 0.1 up to 3.93 l⋅mol –1 ⋅s –1, and the entropy change value increases from
–143 up to –48 J·mol –1 ⋅s –1. Herewith, the value of kv was found to directly depend on
the molar concentration of DEA (Fig. 20). Such changes can not be explained by a decrease
of additional solvation in the formation of the transition state of reaction (1), but are caused
by a decrease of the entropy of the initial state, as the result of the interaction of DEA mol-
ecules with chemically active NH bonds of H2P before entering into reaction (1).
kv, mol/(l·s)
CDEA, mol/l
Figure 20 Dependence of the rate constant of reaction (1) on the concentration of DEA for H2TBP
(X) in the Py–DEA system (a) and H2TPTBP (IV) in Py–DEA (b) and DMSO–DEA (c) systems.
Chemical Activation of Porphyrins in Coordination Core Reactions 203
However, electron spectroscopy data can not be a reliable criterion, as they make it possible
to characterize the chemical NH activity of H2P only in the cases when this property is pro-
nounced in compounds [32]. A large number of porphyrins, pertaining in accordance with
the above three criteria to H2P with a chemically active NH bond, e.g., compounds IV, VIII
and X, do not practically change their EAS in electron-donor solvents (Table 17) because
their PTC formation equilibria (21) are strongly shifted to the left. An example of an EAS
for a weakly NH-active porphyrin IV [32, 39] is given in Fig. 21, a. In contrast, compound
V, which possesses a high chemical activity of NH bonds, forms in DMSO a PTC, which
204 D.B. Berezin and B.D. Berezin
(a)
(b)
λ, nm λ, nm
Figure 21 Electronic absorption spectra of (a) H2TPTBP (IV) and (b) H2(β-Et)8TPP (V) in CHCl3
(1) and DMSO (2).
is accompanied by a decrease in the number and by a shift of EAS bands (Fig. 21, b). Here-
with, the shift of the most batochromic Qx band (Table 17) is the greater, the easier the PTC
is formed by porphyrin.
Some porphyrins can not be tested for the chemical activity of NH bonds by all three
criteria. It has been found [26, 58, 85] that H2P, for which at least two of the three assess-
ment criteria of the state of NH bonds are observed, can be considered to be NH-active. A
larger number of criteria of NH activity can be suggested using other methods.
As the result of extensive X-ray diffraction studies [16, 19], all known H2P and MP
with nonplanar structure were classified by the type of molecule distortion as saddled, ruf-
fled, domed, waved, stepped, twisted, gabled etc. [19].
The porphyrin nonplanarity criteria can be not only XDA data [12–13,19, 65], which
describe the structure of macrocycles in the solid phase, but also fluorescence data [22, 33,
49–50, 107, 135, 140, 142, 147], RR (resonance Raman) spectroscopy [134, 148–149], and
in some cases NMR spectral results [72, 132, 150–151].
A crucially important issue is that of the relation between the type of nonplanar struc-
ture of the porphyrin macrocycle and the chemical activity of NH bonds in its coordination
core. If such a relation does exist and, thus, a simple change of a nonplanar conformation
of the H2P molecule may cardinally change its properties, the mechanism of the processes
involving these molecules in vivo can be assumed to be similar [8, 17, 27].
The most pronounced distortion types of the planar structure of the H2P molecule in-
clude, besides dodeca substitution, the substitution of meso-positions by bulky, most often
alkyl, groups, which leads to the ruffled conformation, as well as the substitution of one or
a larger number of nitrogen atoms in the coordination core to form N-substituted porphyrin
analogues [8, 19, 87]. In the latter case, the saddled conformation of an asymmetric type
was found [142].
Section 1.3, which discussed the criteria of the chemical activity of NH bonds in H2P
molecules, did not consider other types of strongly nonplanar macrocycles, except dodeca-
substituted.
Figure 22 Absorption spectra (solid line) and fluorescence spectra (dashed line) of H(N-CH3)TPP
(XX) in toluene, T = 298 K [142].
At the same time, both meso-substituted H2P [19, 49–50, 70, 72, 136–140] and N-sub-
stituted porphyrin analogues [15, 24, 61, 71, 85–87, 141–144] by their spectral character-
istics, physicochemical properties and reactivity behave as compounds with a nonplanar
structure. Thus, all bands in the visible range of their EAS are shifted batochromically rel-
ative to the predominantly planar H2P [15, 139], which, however, is not a reliable criterion
of nonplanarity [23]. The photophysical characteristics describing the properties of excited
states are more reliable [107]. Significant Stokes shifts of adjacent bands in the fluorescence
spectra relative to absorption bands (22), low quantum yields of fluorescence and triplet
states’ yields are considered to be features of nonplanar porphyrins [33, 49]. Besides,
oooooooo
Table 18 Enthalpic characteristics of deprotonation of {∆H f(–1) and ∆H f(–2) (kcal/mol)} and
formation of mono- and dianionic forms {∆H (–1) and ∆H (–2) (kcal/mol)} of porphyrins with planar
and nonplanar structure (calculated by the AM1 method).
dination by d metal salts (Table 16). In contrast, the positive values of ∆S # = S # – S init in
NH-active H2P are associated with a significant solvation of the initial state on the coordi-
nate of reaction (1).
There are practically no data on the complexation kinetics of nonplanar porphyrins in
ruffled conformation. Our studies, using H2T(iso-Pr)P (XI) and H2T(cyclo-Hex)P (XII)
(∆Cmeso for their complexes with Ni(II) are equal to 0.74 and 0.77 Å [19], respectively),
have shown that the rates of this reaction with copper, zinc or cadmium acetates in electron-
donor solvents (DMF, DMSO), assuming NH activation, are very low, and respective com-
plexes within the temperature range of 298–338 K are not formed in practice. This fact is
a good proof of a localized NH bond in the molecules of meso-substituted porphyrins.
It can also be noted that H2P with bulky meso-substituents, as a rule, do not in practice
enter into complexation reactions with d metal salts in acetic acid and other solvents with
the pronounced proton-donor function. The disturbance of the planar structure of the mac-
rocycle leads to partial localization of the π-electronic density in pyrrole rings and an in-
crease of basicity of the molecule. As protonated forms of porphyrins produced in this case
in proton-donor media in reaction (1) are inactive [1, 7], this rule is also valid for other non-
planar H2P, such as N- or dodeca-substituted molecules. Some dodeca-substituted com-
pounds also enter into reaction (1) in a medium of AcOH, but at very low rates [51–52].
The best medium for MP synthesis in the case of nonplanar meso-substituted porphyrins is
a polar solvent with moderate donor-acceptor properties, e.g., CH3CN, in which reaction
(1) with more nonplanar H2T(cyclo-Hex)P (XII) proceeds at 298 K very rapidly, and with
H2T(iso-Pr)P (XI) with kv ~ ⋅10 l⋅mol –1 ⋅s –1.
Thus, the criteria of the chemical activity of NH bands in the coordination core are a
universal tool for the assessment of the reactivity of porphyrins and are applicable for both
H2P belonging to various structural groups and their analogues. The second important con-
clusion is that a disturbance of the planar structure of the H2P molecule does not automat-
ically assume the presence of chemically active NH bonds. The chemical activity of these
bonds depends in a specific way on the type of conformation, in which a nonplanar macro-
cycle is. Thus, only the occurrence of the symmetrically distorted pronounced saddled con-
formation, possessing a high orthogonal dipole moment [22], leads to the emergence of an
NH activity in nonplanar H2P. Nonplanar porphyrins in ruffled conformation are low-polar
and exhibit no chemical activity of NH bonds.
Porphyrin
ɉɨɪɮɢɪɢɧ 2 Intermediate
ɂɧɬɟɪɦɟɞɢɚɬ
×
solv .H......N
solv
N X. solv
Ɇ Metalloporphyrin
Ɇɟɬɚɥɥɨɩɨɪɮ
N
X. solv
solv . H ......N
solv
Transition state
Figure 23 A schematic mechanism of the complexation reaction of porphyrins with d metal salts
[1, 5, 11, 25].
reaction (1) is controlled by the intensity of the macrocyclic effect, i.e., is determined
by the rate of the electronic and structural MCE components affecting the process
[6–8].
1.1. The structural component always hinders reaction (1) to a greater or smaller extent
due to the spatial shielding of the coordination core by the surrounding atoms and
π-electronic cloud.
1.2. The π-electronic MCE component is determined by the extent of aromaticity of the
H2P macrocycle, and in the case of high delocalization of π-electrons in the macroring
induces the polarization of the molecule and some chemical bonds in it.
Thus, the character of the macrocyclic effect depends on the type of H2P structure –
planar or nonplanar [8].
2. The progress of reaction (1) (Fig. 23) for porphyrins with a different degree of rigidity
strongly depends on the nature of a solvent.
2.1. Reaction (1) of H2P with a chemically inactive NH bond proceeds the most readily
in a medium of solvents – moderate-strength ligands (carboxylic acids, alcohols),
because it is determined by the strength of the solvatosalt [MX2Solv(n-2)] [75]. The
strength of the solvatosalt depends on the coordinating ability of the metal ion and the
nature of ligands – molecules of a solvent (Solv) and anions of a salt (X – ). If the bonds
M–X and M–Solv in the solvatosalt are sufficiently strong, an intermediate is formed,
a complex with H2P of amine type (Fig. 20), which is easily broken by electron-donor
additives [1, 25].
2.2. Complexation reaction (1) for H2P with a chemically active NH bond is accelerated
in solutions of electron-donor solvents. In this case, the state of the NH bond is the
major factor determining the rate of reaction (1) [1, 32, 129].
Chemical activation of NH bonds not only leads to accelerate reaction (1), e.g., owing
to the interaction of these bonds with molecules of an electron-donor solvent, but also to a
possibility of controlling more complex and fine processes, e.g., in the course of tautomeric
conversions. Thus, it has been found that, depending on the state of tautomeric equilibrium
210 D.B. Berezin and B.D. Berezin
F F
F F F F
F F F F
F FN F F F F F F
HN
C N p-TolSO2NHNH2 C N
II
F Cu III F F Cu F
N N DDQ N N
F F F F F F F F
F F F F
F F F F
F F
Figure 24 Redox processes involving the fluorinated analogue of compound CuXXVIII [121].
of compound XXVIII, the oxidation of metal occurring in the ligand can be controlled; here-
with, the metal is in a bidentate state (XXVIIIb) or a tridentate state (XXVIIIa) [117,
122–124]. Thus, Maeda and Furuta [122] propose a simple method of oxidation of Cu(II)P
DDQ to Cu(II)P, which proceeds without the change of the geometry of the coordination
node (plane-square complexes) (Fig. 24). The inverse process is performed in the presence
of p-toluene sulfonyl hydrazide. Both processes proceed quantitatively and almost instan-
taneously. They are realized as easily electrochemically {Cu3+ \Cu2+ = 0.15V} [121].
regrouping
Figure 25 Schematic reaction of the nucleophilic substitution by H2P molecule coordination core
atoms in saturated carbon atom [15, 85].
Although the substitution process is nucleophilic, by one of the tertiary nitrogen atoms
of the coordination core, progress of the reaction at the regrouping stage depends to a great
extent on the chemical activity of the NH bond.
the group of planar classical porphyrins with a chemically low-active NH bond [28, 58],
selective activation of bonds in these molecules in vivo, nevertheless, can occur. It can pro-
ceed under the influence of the protein-lipid environment, leading to conformational chang-
es of a certain character [17, 19] and, probably, to the polarization of the macrocycle [22,
26]. As we showed above, depending on the type of H2P conformation emerging in solu-
tion, the polarization of the macrocycle and individual bonds in it can occur or not occur at
all [26, 85]. It can be expected that processes in biosystems proceed in accordance with sim-
ilar mechanisms.
Incorporation of metals (Fe, Mg, Co, Ni and others) into molecules of heme precur-
sors, chlorophylls or cytochromes is performed in vivo on the enzymic level [10–11, 14,
27, 258]. Nevertheless, the effect of conformationally initiated activation of NH bonds can
also accelerate metal inclusion processes by several orders, as it occurs in solution [26, 35,
52, 85].
Unfortunately, until now it has not been established in detail in conformations of which
types precursors of bioporphyrins exist and how they structurally change in vivo, as well as
which factors regulate the process of metal inclusion into H2P ligand [14]. It can only be
assumed that their main conformation in this process would be saddled [16, 19], as the result
of the formation of which the synthesis of MP proceeds more efficiently owing to the ac-
cessibility of the reaction centre on the whole and the activation of NH sites in particular
[26].
Thus, studies of the peculiar features of conformational conversions of the porphyrin
macrocycles and related changes in the physics, chemistry and reactivity of compounds has
wide prospects. Thus, for instance, knowledge of the requirements imposed on porphyrins
– molecular-recognition receptors or photosensitizers in photodynamic therapy [110] –
from the point of view of the geometry of conformations, from the positions of the best cor-
respondence of reaction centres of the participants in the process will undoubtedly make it
possible to intensify it significantly.
References
1. B.D. Berezin, Coordination Compounds of Porphyrins and Phthanocyanines, Nauka: Moscow,
280 pp. (1978) (in Russian); Coordination Compounds of Porphyrins and Phthanocyanines,
Wiley: New York–Toronto, 286 pp. (1981).
2. P. Hambright, Dynamic Coordination Chemistry of Porphyrins, in: Porphyrins and Metallo-
porphyrins, ed. by K.M. Smith, Elsevier: Amsterdam–Oxford–New York, pp. 233–278 (1975).
3. J.W. Buchler, Synthesis and Properties of Metalloporphyrins, in: The Porphyrins, ed. by D.
Dolphin, Acad. Press: New York, San Francisco, London, vol. 1, pp. 389–483 (1978).
4. B.D. Berezin, Application of Porphyrins for Studies of Electronic, Steric and Solvation Effects
of Coordination, in: Porphyrins: Spectroscopy, Electrochemistry, Applications, ed. by N.S.
Enikolopyan, Nauka: Moscow, pp. 182–213 (1987) (in Russian).
5. B.D. Berezin and N.S. Enikolopyan, Metalloporphyrins, Nauka: Moscow, 159 pp. (1988)
(in Russian).
6. B.D. Berezin and M.B. Berezin, Zhurn. Fiz. Khim., 63 (12), 3166–3181 (1989).
7. B.D. Berezin, M.B. Berezin and D.B. Berezin, Ross. Khim. Zhurn., 41 (3), 105–123 (1997).
8. D.B. Berezin and T.N. Lomova, The Macrocyclic Effect and Biological Activity of
Metalloporphyrins, in: Problems of Solution Chemistry: Structure, Thermodynamics,
Reactivity, ed. by A.M. Kutepov, Nauka: Moscow, 326–362 (2001).
9. Porphyrins and Metalloporphyrins, ed. by K.M. Smith, Elsevier: Amsterdam–Oxford–New
York, 890 pp. (1975).
10. The Porphyrins, ed. by D. Dolphin, Acad. Press: New York–San Francisco–London, vol. I–VII
(1978–1979).
Chemical Activation of Porphyrins in Coordination Core Reactions 213
11. Porphyrins: Structure, Properties, Synthesis, ed. by N.S. Enikolopyan, Nauka: Moscow, 333 pp.
(1985) (in Russian).
12. J.L. Hoard, Stereochemistry of Porphyrins and Metalloporphyrins, in: Porphyrins and
Metalloporphyrins, ed. by K.M. Smith, Elsevier: Amsterdam–Oxford–New York, pp. 317–380
(1975).
13. W.R. Scheidt, Porphyrin Stereochemistry, in: The Porphyrins, ed. by D. Dolphin, Acad. Press:
New York–San Francisco–London, vol. 3, pp. 463–512 (1978).
14. The Porphyrin Handbook, ed. by K.M. Kadish, K.M. Smith and R. Guliard, Academic Press:
New York, vols. I–X (2000).
15. D.K. Lavallee, The Chemistry and Biochemistry of N-substituted Porphyrins, VCH Publishers:
New York, 313 pp. (1987).
16. W.R. Scheidt and Y.-J. Lee, Struct. Bonding (Berlin), 64, 1 (1987).
17. M.O. Senge, J. Photochem. Photobiol. B: Biol., 16, 3 (1992).
18. K.M. Barkigia, L. Chantranpong, K.M. Smith and J. Fayer, JACS, 110, 7566–7567 (1988).
19. M.O. Senge, Highly Substituted Porphyrins, in: The Porphyrin Handbook, ed. by K.M. Kadish,
K.M. Smith and R. Guliard, Academic Press: New York, vol. 1, pp. 239–347 (2000).
20. S.G. Di Magno, A.K. Wertching and C.R. Ross, JACS, 117 (31), 8279–8280 (1995).
21. B.D. Berezin, Koord. Khim., 22 (4), 285–289 (1996).
22. V.S. Chirvony, I.V. Sazanovich, V.A. Galievsky, A. van Hoek, T.J. Schaafsma, V.L.
Malinovskii and D. Holten, J. Phys. Chem., 105 (32), 7818–7829 (2001).
23. A. Ghosh, I.H. Wasbotten and J. Conradie, J. Phys. Chem., 107 (15), 3613–3623 (2003).
24. D.B. Berezin, Zhurn. Obshch. Khim., 74 (3), 506–511 (2004).
25. B.D. Berezin and O.I. Koifman, Usp. Khim., 42 (11), 2007–2037 (1973) (in Russian).
26. D.B. Berezin, Zhurn. Fiz. Khim., 2006 (in press) (in Russian).
27. Inorganic Chemistry, ed. by G. Eichhorn, Mir: Moscow, 1437 pp. (1978) (Russian translation).
28. B.D. Berezin and D.B. Berezin, State of Art of the Chemistry of Nonclassical Porphyrins and
their Complexes, in: Advances of Porphyrin Chemistry, ed. by O.A. Golubchikov, Chemistry
Research Institute, St. Petersburg University: St. Petersburg, vol. 2, pp. 128–141 (1999) (in
Russian).
29. J. Takeda and M. Sato, Chem. Lett., 971–972 (1995).
30. T.V. Karmanova, T.V. Gromova, B.D. Berezin, A.S. Semeykin and S.A. Syrbu, Zhurn. Obshch.
Khim., 71 (5), 856–861 (2001) (in Russian).
31. D.B. Berezin and O.V. Toldina, Zhurn. Inorg. Khim., 47 (12), 2075–2081 (2002) (in Russian).
32. D.B. Berezin, O.V. Toldina and R.S. Kumeyev, Zhurn. Fiz. Khim., 78 (8), 1427–1432 (2004)
(in Russian).
33. V.S. Chirvonyi, Picosecond Photoinduced Processes in Porphyrin Molecules and their
Complexes with Nucleic Acids, DSc (Physics and Mathematics) Thesis, Inst. Mol. and Atomic
Physics, Nat. Acad. Sci. of Belarus: Minsk, 315 pp. (2005) (in Belorussian).
34. B.D. Berezin and O.G. Khelevina, Tetrasubstitution and Physico-chemical Properties of
Porphyrins, in: Porphyrins: Structure, Properties, Synthesis, ed. by N.S. Enikolopyan, Nauka:
Moscow, 83–113 (1985) (in Russian).
35. P.A. Stuzhin, G. Khelevina and B.D. Berezin, Azaporphyrins: Acid-base and Coordination
Properties, in: Phthalocyanines: Properties and Applications, ed. by C.C. Leznoff and A.B.P.
Lever, VCH Publ.: New York, vol. 4, pp. 19–77 (1996).
36. P.A. Stuzhin and O.G. Khelevina, Coord. Chem. Rev., 147, 41–86 (1996).
37. P.A. Stuzhin and O.G. Khelevina, Structure and Coordination Properties of Azaporphyrins, in:
Advances of Porphyrin Chemistry, ed. by O.A. Golubchikov, Chemistry Research Institute, St.
Petersburg University: St. Petersburg, vol. 1, pp. 150–202 (1997) (in Russian).
38. B.D. Berezin, E.B. Karavaeva and T.I. Potapova, Formation, Stability and Properties of
Tetrabenzoporphyrin and its Complex Compounds, in: Problems of Solvation and
Complexation, Ivanovo Inst. of Chemistry and Technology: Ivanovo, issue 6, pp. 25–31 (1979)
(in Russian).
39. O.V. Toldina, Complexation of meso-Substituted Tetrabenzoporphyrins with Metal Salts in
Organic Solvents, PhD (Chemistry) Thesis, Ivanovo University of Chemistry and Technology:
Ivanovo, 168 pp. (2004) (in Russian).
40. K.S. Suslick, Ch.-T. Chen, G.R. Meredith and L.-T. Cheng, JACS, 114 (17), 6928–6930 (1992).
41. S.M. LeCours, H.-W. Guan, S.G. DiMagno, C.H. Wang and M.J. Therien, JACS, 118 (6),
214 D.B. Berezin and B.D. Berezin
1497–1503 (1996).
42. M. Yeung, A.C.H. Ng, M.G.B. Drew, E. Vorpagel, E.M. Breitung, R.J. McMahon and D.K.P.
Ng, J. Org. Chem., 63 (21), 7143–7150 (1998).
43. T.E.O. Screen, I.M. Blake, L.H. Rees, W. Clegg, S.J. Borwick and H.L. Anderson, J. Chem. Soc.
Perkin Trans., 1, 320–329 (2002).
44. S. Tsukahara and N. Suzuki, Inorg. Chim. Acta, 245, 105–108 (1996).
45. A.G. Govorov, A.B. Korzhenevsky and O.I. Koifman, Russ. J. Phys. Chem., 70 (4), 582 (1996).
46. V.M. Mamayev, I.P. Gloriozov and V.V. Orlov, Izv. Vuz. Khim. Khim. Tekhnol., 25 (11),
1317–1332 (1982) (in Russian).
47. P.A. Stuzhin, J. Porphirins Phthalocyanins, 7 (12), 796-815 (2003).
48. V.A. Kuzmitsky, K.N. Solovyev and M.P. Tsvirko, Spectroscopy and Quantum Chemistry of
Porphyrins, in: Porphyrins: Spectroscopy, Electrochemistry, Applications, ed. by N.S.
Enikolopyan, Nauka: Moscow, 7–126 (1987) (in Russian).
49. S. Gentemann, C.J. Medforth, T.P. Forsyth, D.J. Nurco, K.M. Smith, J. Fajer and D. Holten,
JACS, 116 (16), 7363–7368 (1994).
50. S. Gentemann, N.J. Nelson, L. Jaquinod, D.J. Nurco, S.H. Leung, C.J. Medforth, K.M. Smith,
J. Fajer and D. Holten, J. Phys. Chem. B, 101 (7), 1247–1254 (1997).
51. N.S. Dudkina, P.A. Shatunov, E.M. Kuvshinova, A.S. Semeykin, S.G. Pukhovskaya, O.A.
Golubchikov, Zhurn. Obshch. Khim., 68 (1), 2042–2047 (1998(in Russian).
52. O.A. Golubchikov, S.G. Pukhovskaya and E.M. Kuvshinova, Spatially Distorted Porphyrins:
Structure and Properties, in: Advances of Porphyrin Chemistry, ed. by O.A. Golubchikov,
Chemistry Research Institute, St. Petersburg University: St. Petersburg, vol. 4, pp. 45–75 (2004)
(in Russian).
53. O.G. Khelevina, N.V. Chizhova and B.D. Berezin, Koord. Khim., 17 (3), 400–404 (1991).
54. O.G. Khelevina, S.V. Timofeeva, B.D. Berezin and S.I. Vagin, Zhurn. Fiz. Khim., 68 (8),
1423–1426 (1994) (in Russian).
55. O.A. Petrov, Koord. Khim., 27 (7), 483–492 (2001).
56. P. Bhyrappa, M. Nethaji and V. Krishnan, Chem. Lett., 869–872 (1993).
57. X.-Zh. Song, W. Jentzen, S.-L. Jia, L. Jaquinod, D.J. Nurco, C.J. Medforth, K.M. Smith and L.
Shelnutt, JACS, 118 (51), 12975–12988 (1996).
58. D.B. Berezin and B.D. Berezin, J. Porphyrins Phthalocyanines, 8 (4–6), 516 (2004).
59. A. Rosa, G. Ricciardi, E.J. Baerends and S.J.A. Gisbergen, J. Phys. Chem. A, 105 (13),
3311–3327 (2001).
60. G.P. Gurinovich, A.A. Sevchenko and K.N. Solovyev, Spectroscopy of Chlorophyll and Related
Compounds, Nauka i Tekhnika: Minsk, 520 pp. (1968) (in Russian).
61. V.G. Andrianov, O.V. Malkova and D.B. Berezin, Acid-basic Properties of Porphyrins, in:
Advances of Porphyrin Chemistry, ed. by O.A. Golubchikov, Chemistry Research Institute, St.
Petersburg University: St. Petersburg, vol. 3, pp. 107–129 (2001) (in Russian).
62. D.B. Berezin, Yu.B. Ivanova and V.B. Sheinin, Zhurn. Fiz. Khim., 2006 (in Russian) (in press).
63. A.I. Vyugin, E.V. Antina and G.A. Krestov, Dokl. Akad. Nauk SSSR, 315 (5), 1149–1151 (1991)
(in Russian).
64. A.I. Vyugin, E.V. Antina and M.B. Berezin, Solvation Effects and Coordination Properties of
Porphyrins and Metalloporphyrins in Solutions, in: Advances and Problems of Solvation
Theory: Structural and Thermodynamic Aspects, ed. by A.M. Kutepov, Nauka: Moscow,
pp. 208–241 (1998) (in Russian).
65. W.R. Scheidt, Systematics of the Stereochemistry of Porphyrins and Metalloporphyrins, in: The
Porphyrin Handbook, ed. by K.M. Kadish, K.M. Smith and R. Guliard, Academic Press: New
York, vol. 3, pp. 49–112 (2000).
66. M.I. Terekhova, E.S. Petrov and E.M. Rokhmina, Khim. Geterotsykl. Soed., 8, 1104–1108
(1979) (in Russian).
67. G.I. Koldobsky and V.A. Ostrovsky, Khim. Geterotsykl. Soed., 5, 579–592 (1988) (in Russian).
68. Yu.A. Fialkov, Solvent as a Means of Controlling a Chemical Process, Khimiya: Leningrad, 240
pp. (1990) (in Russian).
69. B.D. Berezin, V.G. Andrianov and O.V. Malkova, Dokl. Akad. Nauk SSSR, 314 (6), 1432–1435
(1990) (in Russian).
70. C.J. Medforth, C.M. Muzzi, K.M. Shea, K.M. Smith, R.J. Abraham, S. Jta and J.A. Shelnutt,
J. Chem. Soc. Perkin Trans., 2, 839–844 (1997).
Chemical Activation of Porphyrins in Coordination Core Reactions 215
71. D.B. Berezin, Effect of Electronic, Solvation and Steric Factors on the Chromophore Properties
of Porphyrins and their Protonated Forms, PhD (Chemistry) Thesis, Ivanovo State University of
Chemistry and Technology: Ivanovo, 187 pp. (1997) (in Russian).
72. C.J. Medforth, NMR Spectroscopy of Diamagnetic Porphyrins, in: The Porphyrin Handbook, ed.
by K.M. Kadish, K.M. Smith and R. Guliard, Academic Press: New York, vol. 5, pp. 1–80
(2000).
73. F. Longo, E. Brown, D. Quimby, A. Adler and M. Meot-Ner, Ann. New York Acad. Sci., 206,
420–424 (1973).
74. B.D. Berezin and O.A. Golubchikov, Zhurn. Fiz. Khim., 60, 2113–2128 (1986) (in Russian).
75. B.D. Berezin and O.A. Golubchikov, Coordination Chemistry of Solvation Complexes of
Transition Metal Salts, Nauka: Moscow, 240 pp. (1992) (in Russian).
76. B.D. Berezin, Koord. Khim., 17 (5), 597–606 (1991) (in Russian).
77. B.D. Berezin, Usp. Khim., 60 (9), 1946–1968 (1991) (in Russian).
78. O.A. Golubchikov, Izv. Vuz. Khim. Khim. Tekhnol., 47 (5), 10–26 (2004) (in Russian).
79. B.D. Berezin, G.A. Tsvetkov and L.P. Shormanova, Zhurn. Inorg. Khim., 25 (10), 1645–2652
(1980) (in Russian).
80. T.N. Lomova, L.P. Shormanova and M.E. Klyueva, Electronic and Steric Effects in the
Functional Substitution of Metalloporphyrins, in: Advances of Porphyrin Chemistry, ed. by O.A.
Golubchikov, Chemistry Research Institute, St. Petersburg University: St. Petersburg, vol. 1, pp.
129–149 (1997) (in Russian).
81. J. Reid and P. Hambright, Inorg. Chim. Acta, 33, L135–L136 (1979).
82. P. Hambright, Chemistry of Water Soluble Porphyrins, in: The Porphyrin Handbook, ed. by
K.M. Kadish, K.M. Smith and R. Guliard, Wiley: San Diego, vol. 3, pp. 129–210 (2000).
83. V.S. Chirvony, A. van Hoek, T.J. Schaafsma, P.P. Pershukevich, I.V. Filatov, I.V. Avilov, S.I.
Shishporenok, S.N. Terekhov and V.L. Malinovskii, J. Phys. Chem. B, 102 (48), 9714–9724
(1998).
84. B.D. Berezin, Solvation Processes in Porphyrin Chemistry, in: Problems of the Chemistry of
Solutions and Liquid-phase Materials, Ivanovo Institute of Solution Chemistry RAS: Ivanovo,
pp. 238–248 (2001) (in Russian).
85. E.N. Misko, Spectral, Kinetic, Thermochemical Characteristics of Solvation and the
Coordination Properties of Some N- and Dodeca-substituted Porphyrins, PhD (Chemistry)
Thesis, Ivanovo Institute of Solution Chemistry RAS: Ivanovo, 198 pp. (2005) (in Russian).
86. G.M. Mamardashvili and B.D. Berezin, Thermodynamics of Porphyrin Dissolution, in:
Advances of Porphyrin Chemistry, ed. by O.A. Golubchikov, Chemistry Research Institute, St.
Petersburg University: St. Petersburg, vol. 3, pp. 130–149 (2001) (in Russian).
87. A.M. Stolzenberg, S.W. Simerly, B.D. Steffey and G. Scott Hatmond, JACS, 119 (49),
11843–11854 (1997).
88. K. Letts and R.A. Mackay, Inorg. Chem., 14 (12), 2990–2993 (1975).
89. M. Onaka, T. Shinoda, Yu. Izumi and E. Nolen, Chem. Lett., 117–120 (1993).
90. O.A. Petrov, The Reactivity of Tetraazaporphyrins in Processes of Acid-base Interaction and
Formation of Molecular Complexes, DSc (Chemistry) Thesis, Ivanovo State University of
Chemistry and Technology: Ivanovo, 264 pp. (2004) (in Russian).
91. M. Whalley, J. Chem. Soc., 866–869 (1961).
92. B.D. Berezin, V.G. Andrianov and O.V. Malkova, Dokl. Akad. Nauk SSSR, 314 (6), 432–1435
(1990) (in Russian).
93. B.D. Berezin, O.G. Khelevina, N.D. Gerasimova and P.A. Stuzhin, Zhurn. Fiz. Khim., 56 (11),
2768–2772 (1982) (in Russian).
94. O.A. Petrov, O.G. Khelevina, and B.D. Berezin, Zhurn. Fiz. Khim., 69 (5), 811–815 (1995)
(in Russian).
95. O.A. Petrov, O.G. Khelevina and S.V. Timofeeva, Zhurn. Fiz. Khim., 69 (10), 1771–1775
(1995) (in Russian).
96. O.A. Petrov, N.V. Chizhova and B.D. Berezin, Koord. Khim., 25 (3), 235–240 (1999)
(in Russian).
97. O.A. Petrov and N.V. Chizhova, Koord. Khim., 25 (5), 393-400 (1999) (in Russian).
98. O.A. Petrov, Zhurn. Fiz. Khim., 76 (9), 1577–1582 (2002) (in Russian).
99. O.A. Petrov, G.V. Osipova, A.S. Semeykin and B.D. Berezin, Koord. Khim., 31 (12), 941–945
(2005) (in Russian).
216 D.B. Berezin and B.D. Berezin
100. V.A. Kuzmitsky and K.N. Solovjev, J. Molec. Struct., 65, 219–230 (1980).
101. V.M. Mamaev, S.Ya. Ishchenko and I.P. Gloriozov, Izv. Vuz. Khim. Khim. Tekhnol., 32 (1),
3–21 (1989) (in Russian).
102. M. Asakawa, H. Toi, Y. Aoyama and H. Ogoshi, J. Org. Chem., 57 (21), 5796–5798 (1992).
103. J. Braun, M. Schlabach, B. Wehrle, M. Köcher, E. Vogel and H.-H. Limbach, JACS, 116, 6593
(1994).
104. J.R. Reimers, T.X. Lü, M.J. Crossley and N.S. Hush, JACS, 117 (10), 2855–2861 (1995).
105. V.A. Kuzmitsky, Khim. Fiz., 15 (12), 61–74 (1996) (in Russian). (in Russian).
106. I.K. Shushkevich, V.N. Knyukshto, V.A. Kuzmitsky, V.N. Kopranenkov, A.M. Vorotnikov
and K.N. Solovyev, Zhurn. Prikl. Spektrosk., 57 (1–2), 92–99 (1992) (in Russian).
107. V.A. Kuzmitsky, V.I. Gael and K.N. Solovyev, Quantum-chemical Calculations of the
Electronic Structure and Spectroscopic Properties of Tetrapyrrole Molecular Systems, in:
Spectroscopy and Luminescence of Molecular Systems, Belorussian State University: Minsk,
pp. 96–120 (2002) (in Russian).
108. K.N. Solovyev, A.L. Gladkov, A.S. Starukhin and S.F. Shkirman, Spectroscopy of Porphyrins:
Vibrational States, Nauka i Tekhnika: Minsk, 415 pp. (1985) (in Russian).
109. A. Ghosh, Quantum Chemical Studies of Molecular Structures and Potential Energy Surfaces
of Porphyrins and Hemes, in: The Porphyrin Handbook, ed. by K.M. Kadish, K.M. Smith and
R. Guliard, Wiley: San Diego, vol. 7, pp. 1–38 (2000).
110. The Porphyrin Handbook, ed. by K.M. Kadish, K.M. Smith and R. Guliard, Wiley: San Diego,
vol. 6, 350 pp. (2000).
111. P.S. Clezy, Oxophlorins (Oxyporphyrins), in: The Porphyrins, ed. by D. Dolphin, Acad. Press:
New York, vol. 2, pp. 103–130 (1978).
112. M.G.H. Vicente, Reactivity and Functionalization of β-Substituted Porphyrins and Chlorins, in:
The Porphyrin Handbook, ed. by K.M. Kadish, K.M. Smith and R. Guliard, San Diego: Wiley,
vol. 1, pp. 149–200 (2000).
113. B.D. Berezin and O.B. Lapshina, Zhurn. Fiz. Khim., 48 (2), 303–307 (1974) (in Russian).
114. B.D. Berezin and O.B. Lapshina, Zhurn. Fiz. Khim., 50 (8), 2007–2011 (1976) (in Russian).
115. P.J. Chmielewski, L. Latos-Grazynski, K. Rachlewicz and T. Glowiak, Angew. Chem. Int. Ed.
Engl., 33 (7), 779–781 (1994).
116. H. Furuta, T. Asano and T. Ogawa, JАCS, 116 (2), 767–768 (1994).
117. H. Furuta, T. Ishizuka, A. Osuka, H. Dehima, H. Nakagawa and Y. Ishikawa, JACS, 123 (25),
6207–6208 (2001).
118. J.P. Belair, Ch.J. Ziegler, Ch.S. Rayesh and D.A. Modarelly, J. Phys. Chem. A, 106 (27),
6445–6451 (2002).
119. T. Morimoto, Sh. Tanaguchi, A. Osuka and H. Furuta, Eur. J. Org. Chem. 3887–3890 (2005).
120. J.L. Shaw, Sh.A. Garrison, E.A. Aleman and C.J. Ziegler, J. Org. Chem. 69 (22), 7423–7427
(2004).
121. H. Maeda, Y. Ishikawa, T. Matsuda, A. Osuka and H. Furuta, JACS, 125 (39), 11822–11823
(2003).
122. H. Maeda and H. Furuta, J. Porphyrins Phthalocyanines, 8 (1), 67 (2004).
123. A. Srinavasan and H. Furuta, Acc. Chem. Res., 38 (1), 10–20 (2005).
124. D.B. Berezin, I.A. Maltsev, A.S. Semeykin and V.L. Bolotin, Zhurn. Fiz. Khim., 79 (12),
2220–2226 (2005) (in Russian).
125. P.J. Chmielewski and L. Latos-Grazynski, J. Chem. Soc. Perkin Trans., 2, 503–505 (1995).
126. P. Smirnov, L. Weng and I. Persson, Phys. Chem. Chem. Phys., 3, 5248 (2001).
127. P.J. Chmielewski and L. Latos-Grazynski, Inorg. Chem., 36 (5), 840 (1997).
128. H. Furuta, T. Morimoto and A. Osuka, Org. Lett., 3 (9), 1427 (2003).
129. E.M. Kuvshinova, A.S. Semeikin and O.A. Golubchikov, Russ. J. Phys. Chem., 71 (3),
371–373 (1997).
130. B.D. Berezin, T.V. Karmanova and T.V. Gromova, Zhurn. Inorg. Khim., 47 (12), 890–895
(2002) (in Russian).
131. V. Bhyrappa and V. Krishnan, Inorg. Chem., 30 (2), 239–245 (1991).
132. C.J. Medforth, M.O. Senge, K.M. Smith, L.D. Sparks and J.A. Shelnutt, JACS, 114 (25),
9859–9869 (1992).
133. M.O. Senge and W.W. Kalish, Inorg. Chem., 36 (26), 6103–6116 (1997).
134. J.A. Shelnutt, X.-Z. Song, J.-G. Ma et al., Chem. Soc. Rev., 27, 31–41 (1998).
Chemical Activation of Porphyrins in Coordination Core Reactions 217
135. J.L. Retsek, S. Gentemann, C.J. Medforth, K.M. Smith, V.S. Chirvony, J. Fajer and D. Holten,
J. Phys. Chem., 104 (29), 6690–6693 (2000).
136. T. Ema, M.O. Senge, M.I. Nelson, H. Ogoshi, K.M. Smith, Angew. Chem. Int. Ed. Engl., 33,
1879–188 (1994).
137. M.O. Senge, T. Ema, K.M. Smith, J. Chem. Soc. Chem. Comm., 733 (1995).
138. M. Veyrat, R. Ramasseul, J.-C. Marchon, I. Turowska and W.R. Scheidt, New J. Chem., 19,
1199–1202 (1995).
139. M.O. Senge, I. Bischoff, N.T. Nelson and K.M. Smith, J. Porphyrins Phthalocyanines, 3,
99–116 (1999).
140. S. Gentemann, C.J. Medforth, T. Ema and N.J. Nelson, Chem. Phys. Lett., 245, 441–447
(1995).
141. S.-I. Aizawa, Y. Tsuda, Y. Ito, K. Hatano and S. Funahashi, Inorg. Chem., 32 (7), 1119–1123
(1993).
142. I.V. Sazanovich, A. Van Hoek, A.Yu. Panarin, V.L. Bolotin, A.S. Semeykin, D.B. Berezin and
V.S. Chirvony, J. Porphyrins Phthalocyanines, 9 (1), 59–67 (2005).
143. M.O. Senge, W.W. Kalisch and S. Runge, Liebigs Ann./Recueil, 1345–1352 (1997).
144. T.E. Clement, L.T. Nguyen, K.M. Smith et al., Heterocycles, 45 (4), 651–658 (1997).
145. M.O. Senge, T.P. Forsyth, L.M. Nguyem and K.M. Smith, Angew. Chem. Int. Ed. Engl.,
33 (23/24), 2485–2487 (1994).
146. B. Cheng, O.Q. Munro, H.M. Marques and W.R. Scheidt, JACS, 119 (44), 10732–10742
(1997).
147. V.S. Chirvony, A. Van Hoek, V.A. Galievsky, I.V. Sazanovich, T.J. Schaafsma and D. Holten,
J. Phys. Chem., 104 (42), 9909–9917 (2000).
148. C.J. Medforth, J.A. Shelnutt, M.D. Berber and K.M. Barkigia, JACS, 113, 4077–4087 (1991).
149. L.D. Sparks, C.J. Medforth, M.S. Park et al., JACS, 115 (2), 581–592 (1991).
150. J.W. Dirks, G. Underwood, J.C. Matheson and D. Gust, J. Org. Chem., 44 (14), 2551–2555
(1979).
151. C.J. Medforth, R.E. Haddad, C.M. Muzzi, N.R. Dooley, L. Jaqwinod, D.C. Shyr, D.J. Nurco,
M.M. Olmstead, K.M. Smith, J.-G. Ma and J.A. Shelnutt, Inorg. Chem., 42 (7), 2227–2241
(2003).
152. D.B. Berezin, Zhurn. Obshch. Khim., 75 (5), 807–810 (2005) (in Russian).
153. K.M. Kadish, Electrochemistry of Metalloporphyrins in Non-aqueous Media, in: The Porphyrin
Handbook, ed. by K.M. Kadish, K.M. Smith and R. Guliard, San Diego: Wiley, vol. 8, pp.
1–114 (2000).
154. D.B. Berezin, E.N. Misko, E.V. Antina and M.B. Berezin, Zhurn. Obshch. Khim., 2006 (in
Russian) (in press).
155. I.V. Tsvetkova, V.G. Andrianov and B.D. Berezin, Izv. Vuz. Khim. Khim. Tekhnol., 37 (1),
73–76 (1994) (in Russian).
156. V.B. Sheinin, V.G. Andrianov, B.D. Berezin and T.A. Koroleva, Zhurn. Org. Khim., 21 (7),
1564–1570 (1985) (in Russian).
157. V.B. Sheinin, Studies of the Acid-base Ionization of Some Natural and Synthetic Porphyrins,
PhD (Chemistry) Thesis, Ivanovo Inst. of Chemistry and Technology: Ivanovo, 157 pp. (1981)
(in Russian).
158. V.Ya. Bykhovsky, Tetrapyrrols: Diversity, Biosynthesis, Biotechnology, in: Advances of
Porphyrin Chemistry, ed. by O.A. Golubchikov, Chemistry Research Institute, St. Petersburg
University: St. Petersburg, vol. 1, pp. 27–51 (1997) (in Russian).
218 D.B. Berezin and B.D. Berezin
6 Synthesis, Structure
Peculiarities and
Biological Properties
of Macroheterocyclic
Compounds
Introduction
A conspicuous group of natural macrocyclic pigments among the great diversity of macro-
heterocyclic compounds are hemoglobins, chlorophylls, cytochromes and other species.
They play an exceptional role in vital processes of photosynthesis and respiration, are in-
volved in regulation of fine metabolic processes in living organisms. These substances are
tetrapyrrole macroheterocyclic compounds (porphyrins), the basis of which is porphin 1. It
has a planar structure, and their inner macroring consisting of 16 carbon and nitrogen atoms
contains 18 π-electrons and is aromatic.
Owing to their unique structure, porphyrins and their metal complexes exhibit quite a
number of interesting properties and find the most broad application in science, technology
and medicine. In particular, they are used as photosensitizers for photodynamic therapy of
cancer [1], as catalysts [2] and electrochemical sensors [3]; are promising materials for non-
linear optics, optoelectronics, liquid-crystal systems [4] etc.
X N
N HN N HN
X N N
X
NH N NH N
X N
1 X = CH; 1a X = N 2
N
N
R
R
N NH
N N N
N
NH N NH
R
N
N
3 4
structural modification of the macrocycle skeleton, which do not rule out the incorporation
of pharmacophore groupings, introduction of substituents by the periphery, as well as metal
atoms into the inner coordination cavity, make this class of compounds rather promising
and attractive for the search of substances with practically valuable properties, in particular,
with potential biological activity.
A
NH
2H N A
NH2
+ 2 B B B
2
NH
A
Scheme 1
The reaction of three-link BAB products with the functional derivatives of phthalod-
initrile – 1,3-diiminoisoindoline or 1,1-dialkoxy-3-iminoisoindolines – is used to produce
nonsymmetric Mc 3 (Scheme 2).
A A
NH
B B + B B B
NH HN NH
B
Scheme 2
As initial diamines, use is usually made of aromatic five- or six-membered carbo- and
heterocyclic compounds containing amino groups in 1,3-positions. The use of 1,4-, bi- and
polynuclear diamines leads to the formation of Mc with an increased internal cavity [7, 8].
Being included into the macrosystem, cyclic fragments of diamines can support or in-
terrupt macroring conjugation. In this connection, of special interest are five-membered
heterocyclic diamines, e.g., based on 1,2,4-triazole 5–8 and 1,3,4-thiadiazole 9, as they are
heteroanalogues of pyrrole and their presence in the macrosystem can lead to the formation
of an intracyclic conjugation system similar to the porphyrazine system.
Substituted diaminotriazoles – 1(H)-3,5-diamino-1,2,4-triazole 5 (guanazole), 3,5-di-
amino-1-phenyl-1,2,4-triazole 6, as well as 3,5-diamino-1-(α-naphthyl)-1,2,4-triazole 7 –
are obtained by condensation of hydrazine hydrate, phenyl hydrazine chloride or α-naph-
thyl hydrazine, respectively, with dicyandiamide [9, 10].
222 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
R 5 R = H; N N
N N 6 R = ɋ6ɇ5;
H2N NH2
H2N NH2 7 R= ɋ10ɇ7 S
N 8 R = C12H25
5-8 9
NH Alk
AlkS CN
NH C4H9 11,
AlkS CN C8H17 12,
NH
10 C12H25 13
Phthalodinitrile and its substituted are rarely used for Mc synthesis directly. Usually,
they are transferred into more reactive alkoxy- or diiminoisoindole derivatives.
The reaction of phthalodinitrile with alcoholates of alkaline metals in an alcoholic me-
dium (Alk = Me, Et, Pr) smoothly proceeds at room temperature [21] in accordance with
Scheme 3.
14 15 16 16a
Scheme 3
OAlk
CN OAlk
N N
CN Na
N NNa
14 14a 15
Scheme 4
It should be emphasized that the formation of bipolar ion 14a is postulated, and the
role of the “polarizing effect of the medium” is not disclosed. Herewith, no proofs support-
ing this mechanism are given in the literature.
Borodkin [22] studied the interaction of phthalodinitrile with sodium methanolate in
benzene and methanol. He notes that, in the former case, a product of sodium methanolate
addition to one of the nitrile groups is formed, whereas in the reaction in methanol the re-
action product is an isoindolenine derivative. Therefore, the mechanism given in Scheme 4
is, apparently, not realized. It is also not ruled out that the isoindolenine derivative is formed
as the result of the intramolecular cyclization of the monoaddition product. The results of
kinetic studies of the reaction of phthalodinitrile with phenol given in a review [23] are in
favour of such a mechanism.
As phthalodinitrile can be recrystallized from alcohol and other organic solvents,
while addition of alcoholate to its alcoholic solution already at room temperature leads to
the formation of alkoxy compounds, the activating role of alkaline metal becomes evident.
Interaction of phthalodinitrile and its substituted derivatives with nucleophilic re-
agents in the presence of alkaline metal cations was studied using the MNDO quantum
chemical method [24]. The action of cation was found to cause a polarization of the nitrile
group, as the result of which the carbon atom of this grouping acquires a significant positive
charge and becomes a strong electrophilic site. As the result of the nucleophilic attack, a
product of sodium methanolate molecule addition for this group is formed (Fig. 1); after an
intramolecular regrouping, this product turns into an isoindole derivative.
Thus, based on the generalization of the literature data and performed theoretical stud-
ies, the following mechanism of isoindolenine cycle formation appears to be the most prob-
able: formation of the product of alkaline metal alcoholate addition for one of the nitrile
224 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
O15 H16
C7
C1
N8
126.4
6A
2.7
C2 Li
C9
N10
OAlk OAlk
CN
AlkONa NNa N
CN CN
NNa
14 21 15
Scheme 5
The effect of substituents on the direction and rate of this conversion was explained
within the framework of this two-step mechanism [25].
1,3-Diiminoisoindoline and its substituted species are obtained by the interaction of
respective phthalodinitriles in methanol in the presence of sodium methanolate with ammo-
nia [21]. Apparently, this compound is formed via 1-methoxy-3-iminoisoindolenine, which
in the interaction with ammonia is converted into the end product. This reaction mechanism
was considered in [26] using the semiempirical AM1 method. Conformational and tauto-
meric conversions of 1,3-diiminoisoindoline, as well as the influence of substituents and
solvents on the energetics of isomeric transitions was studied in [27]. Analysis of the critical
points of potential energy surface showed the cis-form 22 to be the most stable (Scheme 6).
H H
H H
N N N
NH NH N
N N N
H H H
22 22a 23
∆Hf = 81.24 kcal mol-1 ∆Hf = 83.19 kcal mol-1 ∆Hf = 88.14 lcal mol-1
Scheme 6
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 225
Planar inversion (53 → 53a) proceeds with overcoming a relatively low energy barrier
(~20 kcal·mol –1). For realizing tautomeric conversions (53a–54) as the result of intramo-
lecular transfer, this value is ~70 kcal·mol –1. Calculations performed in “supramolecular”
approximation showed that the prototropic rearrangement with participation of hydroxyl-
containing solvents (MeOH, H2O) was made with lower energy expenses by about 17
kcal·mol –1. The data obtained are qualitatively consistent with ab initio calculations
(6-31G basis) for intermolecular proton transfer in formamidine hydrate [28], where the ac-
tivation energy is also observed to decrease by 20.8 kcal·mol –1 as compared with the in-
tramolecular transfer in formamidine.
Of special interest is the structure of products of condensation of 1,3-diiminoisoindo-
line with amines at one of the imino groups, as they are intermediate substances in Mc syn-
thesis.
The structure of 2-(1-aminoisoindolenine-3-ylideneamino)-5-thioxo-1,3,4-thiadiaz-
ole-4-in 24 was studied using X-ray diffraction analysis [29]. It has been shown that com-
pound 24 is crystallized as the solvate with one molecule of DMFA and water. In the crystal,
the molecule is planar, despite the much shorter intramolecular contact N(1)...S(1) 2.74 Å
(the sum of van der Waals radii 3.34 Å [30]) (Figs. 2, 3).
226 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
166 H
N N
B
S S N C
N
∆Hf, kcal/mol
164 NH2
162
Figure 4 Internal-rotation energy profiles for tautomer 24a. The structural formula is given for ϕ =
180°.
The energy profiles of inner rotation of the thiadiazoline (thiadiazole) fragment around
the ordinary bond N(3)-C(9) in molecule 24a were studied using the semiempirical quan-
tum chemical AM1 method. The results of the analysis are given in Fig. 4.
In the configuration space studied, the internal rotation proceeds with overcoming the
low energy barrier ∆∆G = 4.02 kcal·mol –1, which makes it possible to assign this molecule
to structurally nonrigid molecules.
The low activation barrier of internal rotation suggests that not internal but external
factors shall be of determining significance in the stabilization of this or that configuration,
which is what is observed in the crystal (Figs. 2, 3), where form C is stabilized owing to
intermolecular interactions.
bis(1-Imino-3-isoindolinylideneamino)arylenes and azoles (TZP) are important inter-
mediate products in the synthesis of both symmetric- and nonsymmetric-structure Mc
[31–33]. bis(1-Imino-3-isoindolinilydeneamino)arylenes 25, 26 are the most studied of
them.
N X N
NH HN
NH HN
25 X = CH; 26 X = N
Despite the similarity in structure, they exhibit different reactivities. Thus, in the in-
teraction with diiminoisoindoline, compound 25 forms a triisoindolebenzene macrocycle,
but from compound 26 a similar product can not be obtained under the same conditions.
A structural feature of three-link products 25 and 26 is the occurrence of two suffi-
ciently “heavy” isoindole fragments connected with the diamine residue by aza bridges,
which appears to determine the aptitude of the molecules to the formation of nonplanar
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 227
Figure 5 Model of complex 25 with DMFA. Figure 6 Model of complex 26 with AlOH.
NH N R N
H2N-R-NH2
NH NH HN 26, 27-29
NH NH HN
H3CO OCH3
NH MXn
NH L(X)
N R N
Cl Cl 26a,b; 27a-e;
28a,b,d; 29b
NH . HCl N M N
NH NH HN
H Ph
N N
N N N N
R=
N S
N N
26 27 28 29
Information on products of TZP interaction with metal salts is rather scarce and of con-
tradictory character. Thus, the work [35] reports the synthesis of complexes of
2,6-bis(1-imino-3-isoindolinilydeneamino)-pyridine with Cu, Ni, Co and Au of 1:1 com-
position. A later paper [36] presents data on the synthesis of copper complexes of TZP
based on 1,3-phenylene, 1,2,4-triazolidene-3,5 and 4-chloro-1,3,5-triazinylidene-2,6,
which represent 2:1 complexes.
The work [37] reports the synthesis of metal complexes 28a,b of 1:1 composition
(Scheme 7), which were further used for template synthesis of ABBB-type Mc.
L
L N N
N N
N Cu N
N M N NH
N N
N
N N
N N N N N N
NH M N N Au N
N N
N N N N
These compounds proved to be unstable to the light and to alkaline media, so further
studies in this direction were stopped.
230 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
3 Azolophthalocyanines
Of special interest among macroheterocyclic compounds of ABBB type are Mc with azole
fragments because, as compared with tetrapyrrole precursors, they have no centre of sym-
metry during the preservation of the internal macroring of the structure close to that of the
porphyrazine one.
Analysis of the literature [31] as well our own experimental studies have shown that
attempts to obtain Mc with three isoindole fragments and heteroaromatic diamine residue
by the method similar to the above method for the benzenetriisoindole macrocycle fail. Pos-
sible causes are the discrepancy of geometric sizes between the reaction centres of
bis(1-imino-3-isoindolinilydeneamino)heteroarylenes and 1,3-diiminoisoindoline, as well
as the low stability of TZP under reaction conditions [32]. As shown by quantum chemical
calculations presented above, TZP-based metal complexes can be of interest as templates
for the synthesis of nonsymmetric-structure Mc.
Synthesis of metal complexes of azole-containing macroheterocyclic compounds was
done by template condensation of respective TZP metal complexes with phthalodinitrile in
a phenol medium [37] (Scheme 8).
CN
N R N Ln N R N Ln
CN
N M N N M N
PhOH
NH HN N N N
The major methods of their production are statistical condensation and interaction of
TZP with respective nitrile or its functional derivatives in the presence of nickel acetate in
organic solvents. In the former case, a mixture of substituted 1,3-diiminoisoindoline, 3,5-di-
amino-1,2,4-triazole and nickel acetate in a molar ratio 3:1:1 is heated in organic solvents
(usually in 2-ethoxyethanol, butylonitrile, butanol) to form Mc containing the same substit-
uents in all three isoindole fragments. In the latter case, one succeeds in obtaining
ABB′B-type Mc with different substituents.
Nickel proved to be the most convenient
in the synthesis of this type of Mc; the
N N
successful use of copper salts is also re-
N N ported. Other metals do not lead to the
N
formation of required compounds.
X N Ni N X Recently, Torres et al. succeeded in pro-
ducing metal-free triazolephthalocya-
N nine [54], however the purification of
N N this substance proved very difficult.
There are reports in the literature [53] on
the production of nickel complexes with
substituted triazolenaphthalocyanines
Y of ABBB and ABB′B types.
Triazolephthalocyanines have a con-
37a, b stant dipole moment, are capable of
37a X = OC8H17, Y = NO2, forming ordered Langmuir–Blodgett
37b X = OC8H17, Y = tBu layers [55, 56] possessing semiconduc-
tor properties [57], exhibit interesting
nonlinear optical properties [58, 59], liquid-crystalline properties [60] and possess a high
thermal stability [61].
The use of substituted pyrrolines instead of substituted isoindoles led to the discovery
of a new class of noncentrosymmetric Mc of ABBB type – derivatives of porphyrazine,
called by the authors [18] triazoleporphyrazines. Thus, hexa(4-tert-butylphenyl)triazole-
porphyrazine 38 and its N-butyl-substituted analogue 39 were produced with good yields
by the interaction of 3,4-di(4-tert-butylphenyl)-2,5-diiminopyrroline with 3,5-diamino-
1,2,4-triazole or 1-dodecyl-3,5-diamino-1,2,4-triazole, taken in a molar ratio 3:1, in dried
butyl alcohol (Scheme 9).
R
N N
N N N
NH N N
R H
N N N
3 NH + N N
H 2N N2H
N
NH
38 R = H, 39 R = C12H25.
Scheme 9
232 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
H
N N N N N N
N N N N N N N N N N
N N
H H
N N N N N N N HH N
H H H
N N N N N N N N
N N N N
H 1.309
11 N N
1.328 1.0127.7
3
5
8.
1.38
N N 10
1.381 13
1.4
10
2
1.36
3
1.3 1 107.8
2.1
.3 107.4 128.3 N
11
1 N 58
1.369 113.8 108.1 .362 N 1.
1
5
N1 104.6 N 12
1.010
.2
1.32
20 1 .35 N 2
1.307
127 32
1.8
1.302
.6 5 . 9 .2 1 127.4
1 11 1.47 H
1.3
.7 126 110.7 2
127 .8 1.48
1.485
.
77
6 06
1.350
N N105.4 11
1.3
2
10 109.7 110.1 1
32
6. 6. 06.
1.347
95
1.347
3 0 1
1.
105.8N N 105.1 110 111.6
0
1.47
35
5.6 5 .7 2 H 127.4
10112.6
1 .3
1.010
1.
14
H 112.2
1.026
1.4 12
1.348
1.3
84 12 83 48 8 2.
5.0
25
.3 125.3 1. 4 N 3
1.366
7 N
11
9
1 N .3 111.0 128.1
1. 1
12
12
.5
1.3109.9 1
0.
7 106.4
N
8
130
1.45
1.307
10
106.7 106.9 1.313
8.
1.467
1
6
10
8 .2
67
1.360
8.2
1 .4
10
1.349
40a 40b
1.279 1.294
N N N N
3
5
0 5
7. 7.
1.39
1.42
10 10
1 .33 5
1.
111.6 1.326 1 111.0
N 34 102.9 129.0 N N .3 103.0 128.0 N
9 9 61 2
N 0. N 9.
5
9
12 11
1.31
1.31
1.860
9 127.3 4 127.5
1.44 12 1.45
12
1.3
1.3
106.4 0
4. 7. 110.0 9
3
. .
89
77
1.012 08 4 06
1.353
1.365
1.878
N H H N110.9 110 N Ni N105.6 110
8.0 6.9
106.7 110.7
9
7
1.869
1.45 1.45
36
36
0 125.9 4 127.4
1.
1.
12 12
1.328
1.326
2. 0.
64 N 4
76 N 4
N 1 .3 105.3 127.9 N N 1 .3 105.8 127.5 N
1.33 111.2 1.31 110.3
7 9
1.457
1.473
10
10
6.
6.
8
1
1.349 1.352
40c 41
Figure 7 Lengths of bonds (Å) and valence angles (°) in molecules of tautomeric species 40a–c and
Ni complex 41 optimized on the level of DFT B3LYP/6-31G**.
The difference in the values of energy in the sequence 40a > 40b > 40c (Table 1) is
relatively small. Tautomer 40c, the structure of whose internal ring is the most close to that
of the porphyrazine ring, is the most advantageous from the energy point of view.
In the sequence 40a, 40b, 40c the length of the N–N bond of the triazole cycle de-
creases and, in the case of structure 40c, approaches the length of the double bond N=N
(Fig. 7). At the same time, the N–C bonds linking the N–N grouping with the internal mac-
rocycle become longer to make 1.425 Å for 40c. This indicates that the N–N bond in this
structure is significantly isolated from the conjugation system of the macroring. In the case
of complex 41, the length of this bond is 1.294 Å and it occupies an intermediate position
between respective bonds in structures 40b and 40c.
Based on the geometric characteristics (Fig. 7) we [62] calculated the HOMA [63] and
234 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
NICS [64, 65] criteria and analyzed the aromaticity of 1a, 40a-c and 41 (Table 2, Fig. 8).
Table 2 EN, GEO and HOMA indices of tautomers 40a –c and Ni complex 41.
N NH N N N N N N
a a a a
N N N N N N N N N
N N N N
H H
N N d N N b d N N b d N H H N b d N Ni N b
H H H
N N N N N
N N N N N N N N
c c c c
As it follows from the data presented in Table 2, the total aromaticity increases in the
sequence of triazoleporphyrazines 40a < 40b < 40c < 41, which corresponds to the tendency
of bond length equalization in these molecules.
Transition from 40a to tautomers 40b and 40c is observed to be accompanied by an
increase of the total aromaticity against the background of a decrease of the local aromatic-
ity of the triazole cycle. The latter is even due to the fact that part of the triazole cycle in-
cluding atom N-4 and adjacent carbon atoms are involved in the conjugation system of the
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 235
H N N N N
-11.6 N N
N N -12.6 -5.60 -5.87
-13.2
N N N N N N N N
N N N N N N
H -3.84 N H
-15.5 -6.90 -14.1 -14.8
N N N N NH HN N Ni N
N N
H H
H
N N N N N N N N N N N N
N N N
Figure 8 Calculated NICS values (ppm) for structures 1a, 40a–c, 41.
internal macroring. This leads to the seclusion of the N–N bond, which in the case of struc-
ture 40c acquires a pronounced double character. As the result, the alternation of single and
double bonds in the triazole cycle increases, which is manifested in an increase of index
GEO from 0.034 up to 0.225 in passing from structure 40a to 40c. It should be noted that
the difference in the HOMA values of total and local aromaticity of the triazole nucleus de-
creases in the sequence 40a > 40b > 40c (0.508, 0.320 and 0.042, respectively). Aromaticity
of the internal cycle in this case increases (40c, HOMA = 0.948).
Formation of nickel complex 41 is accompanied with a greater equalization of bonds
and an increase of the HOMA value for the entire molecule up to 0.643. The bonds at the
bridge nitrogen atoms become shorter and equalize with the simultaneous elongation of
C–N bonds, nitrogen atoms of which directly interact with nickel. As the result of these
changes, the aromaticity of the internal macroring (HOMA = 0.913) decreases as compared
with similar characteristics of metal-free structures 40b and 40c.
Analysis of the aromaticity criteria for [18]heteroannulenes designated on respective
structural formulas 40b, 40c and 41 by bold lines (Table 2) shows that the largest value of
HOMA = 0.803 is in structure 40b, in which the conjugation contour passes through the
azo group of the triazole cycle. The respective contour of structure 40c is characterized by
a smaller value, HOMA = 0.753. In the case of metal complex 41 the contour including the
N–N group of triazole appears to be more aromatic (HOMA = 0.756) than the alternative
contour (HOMA = 0.634).
The magnetic criterion NICS (Fig. 8) confirms the tendencies of aromaticity change
in the sequence 1a, 40a–c, 41, revealed using the HOMA criterion.
Thus, 40a is the least aromatic, NICS = –6.90 ppm, whereas 40c appears to be the most
aromatic of the tautomers (NICS = –14.8 ppm). A rather interesting fact was also that the
triazole nucleus successively loses local aromaticity, which is also consistent with HOMA
criteria for this sequence of tautomers. Analysis of aromaticity using the NICS criterion
supports the conclusion that for the efficient participation in the formation of the aromatic
macrocycle, its constituent heterocycles should lose part of their aromaticity, which should
be compensated for by the gain in energy owing to the established conjugation in the macro-
system.
The spectrum of compound 38 given in Fig. 9 is characterized by four intensive ab-
sorption bands. Their intensity decreases with the advance to the long-wavelength region.
The band at 244 nm, which appears to be the result of electronic transitions in the phenyl
nuclei of substituents, is the most intensive. The band at 324 nm, by analogy with tetraaza-
porphin, can be interpreted as a Soret band. The bands in the visible spectrum are the result
of electronic transitions involving highest occupied (HUMO) and lowest unoccupied
(LUMO) molecular orbitals. An intensive band at 417 nm, exhibited in the spectrum of
236 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
D
A 1.2 391
1.5 374
1.0 325338
328
421 0.8
1.0 324 535 376
417 0.6 328 472 518 629
527 1 1
0.4 497 606 2
0.5 2
0.2 3
0.0 0.0
300 400 500 600 700 300 400 500 600 700
λ, nm λ, nm
Fig. 9 Fig. 10
Figure 9 Electronic absorption spectra: 1, 38; CHCl3, c = 2.16·10 –5; 2, 39, CHCl3, c = 1.71·10 –5
(mol·l –1).
Figure 10 Electronic absorption spectra: 1, M=Ni, 38a, CH2Cl2, c = 1.71·10 –5; 2, M=Cu, 38b,
CHCl3, c = 2.05·10 –5; 3, M=Co, 38c, CHCl3, c = 1.03·10 –5 (mol·l –1).
substituted porphyrazine [66] as a weak band at 450 nm, apparently, is due by its origin to
the disturbance of symmetry of the molecule owing to the substitution of the pyrrole frag-
ment by the triazole fragment.
The long-wavelength band at 527 nm in substituted triazoleporphyrazine 2 is shifted
hypsochromicly by 139 nm as compared with the band of respective porphyrazine [66].
The shape of absorption spectral bands of 38 and of dodecyl-substituted compound 39
is practically the same (Fig. 9), which indicates the identity of their chromophore systems.
At the same time, in compound 39 the 1H-triazole form is fixed by the presence of the alkyl
substituent in position 1. This fact supports the presence of a hydrogen atom in position 1
of the triazole cycle in the molecule of triazoleporphyrazine 38. The absorption bands in
the case of 39 are slightly shifted towards the long-wavelength side, which is a consequence
of the exhibition of a weak +I effect by the substituent.
Three groups of bands can be singled out in the EAS of complexes 38a–c (Fig. 10):
the long-wavelength absorption band at 602–629 nm and a broadened band at 570 nm,
which is exhibited in the case of compounds 38a and 38b as an inflection; a broad band in
the mid part of the spectrum at 472–518 nm; and two bands in the ultraviolet part of the
spectrum at 325–391 nm. The arrangement and sufficiently high intensity of the
long-wavelength absorption band indicate the aromatic character of metal complexes. The
nature of metal has an influence on the position of the absorption bands; the batochromic
shift of the long-wavelength absorption band increases in the sequence Ni < Co < Cu.
phyrazine donor sites involved in acid-base interaction, the character of the interaction, as
well as the values of stability constants of acidic species produced depend on the structure
of porphyrazine and the properties of the proton-donor medium. Porphyrazine ligands can
produce acidic species both by the intracyclic and extracyclic (meso-) atoms of nitrogen. In
metal complexes, ϕn orbitals of the intracyclic nitrogen atoms participate in the formation
of bonds with complex-forming metal, therefore, only meso-atoms of nitrogen are involved
in acid-based interaction [67].
Triazoleporphyrazines, which differ from porphyrazines by the substitution of one of
the pyrrole cycles by a triazole residue, have two additional nucleophilic centres in the tri-
azole cycle, which can be considered as possible protonation sites. Besides, the proximity
of these atoms to the internal macroring makes it possible to assume a significant effect of
protonation on the properties of triazoleporphyrazines.
The process of acid-base interaction with participation of porphyrazines is of a com-
plex character [68]. Proton transfer from acid HA to base B proceeds via the stages of form-
ing acidic associate, H associate, ion-ionic associate and totally ionized protonated species:
1
HA + B B…HA B…H…A BH+…A – BH+ + A– (1)
acidic iassot
acidic associate H associate H associate ion-ionic associate protonated
protonatedassociate associate species
The forming acidic species differ one from another by the degree of proton transfer
from the molecule of acid to the donor site and should be spectrally distinguishable. The
final result of acid-base interaction depends on the electronic and geometric structure of acid
or base, as well as on the solvation features of anion A– and cation BH+, which is determined
by the nature of solvent. In media with low ionizing ability, the electron-donor centres of
porphyrazine, by participating in weak acid-base interaction, form H associates and ion-
ionic associates. The total proton transfer is possible only in a strongly ionizing medium.
Interaction of porphyrazines with acids is accompanied with characteristic changes in
the visible range of the electronic absorption spectrum (EAS), which correspond to the for-
mation of various acidic species. A quantitative measure of basicity is the stability constant
of the acidic species (Ks). Porphyrazines are Hammett indicators [67], and the value of Ks
can be determined using the Hammett equation:
where H0 is the Hammett acidity function, Ii = Ci /Ci–1 is the ratio of an ith and (i–1)th acid-
base species being in equilibrium (indicator ratio), n is the number of donor sites involved
in the acid-base interaction at a given stage. The values of acidity functions are known from
the literature [69]. In mixtures of carboxylic acids with organic solvents, for which the H0
functions are not known, one can determine the concentration constants of stability of acidic
species, using a modified Hammett equation:
where x is the number of acid molecules involved in acid-base interaction at a given stage.
The values of n and x can be determined as the slope of lines log Ii = f(H0) or log Ii =
f(logCHA).
We investigated the stage and stability of 1H-hexakis(4-tert-butylphenyl)triazolepor-
phyrazine 38 (H2TrPz) in a medium of benzene and 100% acetic acid.
238 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
.
H
N N
N N N
N N
H
N N N
38
Introduction of a small amount of acetic acid (0.4 mol·l –1, H0 = 7.5) into a benzene
solution of triazoleporphyrazine leads to a hypsochromic shift of the Q band in the EAS
(Fig. 11). Formation of the acidic species is completed in an acetic acid–H2SO4 –antipyrin
medium (H0 = 3.95). Herewith, the absorption maximum of the Q band is shifted hypso-
chromicly to 473 cm –1.
The stability constant of acidic species 38, as determined by the Hammett equation, is
equal to 4.48±0.18. The number of donor sites involved in acid-base interaction is 0.97,
i.e., ~ 1 (Fig. 12).
Taking into account that acid-base interactions of triazoleporphyrazines occur in a me-
dium with low ionizing ability, it can be concluded that an ion-ionic associate is formed in
this case.
A
1.6
531
1.4 1
2
1.2 3
4
1.0
0.8
0.6
0.4
0.2
0.0
450 500 550 600 650 λ, nm
log I
0.5
0.0
4.0 4.5 5.0 5.5 6.5 6.0 H0
-0.5
-1.0
-1.5
-2.0
Α
531
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
450 500 550 600 λ, nm
313 K
0.07
0.06
0.03 333 K
0.02
0.01
0.00
0 10 20 30 40 50 τ, min
Figure 14 Dependence of log (C 0H2TrPz / CH2TrPz) on time τ for the destruction reaction of triazole-
porphyrin in a benzene–acetic acid medium at temperatures 313 K, 323 K, 333 K and H0 = 4.87.
log ke 333
-4.0
-4.1 323
-4.2
-4.3
313
-4.4
-4.5
-4.6
Figure 15 Dependence of log keff on log C CH3COOH for the destruction reaction of triazoleporphy-
rin 38.
character (Fig. 15) with the slope close to 2, i.e., the reaction is of second order with respect
to the concentration of acetic acid.
Destruction of porphyrazines can be protolytic and acidolytic under the action of acid
molecules, or a more complex solvoprotolytic process under the action of solvated proton.
Destruction of porphyrazines in aqueous solutions of H2SO4 is hydrolytic [71], i.e., pro-
ceeds under the action of hydroxonium ions H3O+. To split the macrocycle, the system must
in all cases be destabilized by two H3O+ ions. Destruction of triazoleporphyrazine in a ben-
zene-acetic acid medium is acidolytic, i.e., proceeds under the action of AcOH molecules.
The following scheme of destruction of the triazoleporphyrazine macrocycle can be
proposed. First there occurs the protonation of nitrogen atom, which was localized in posi-
tion 4 of the triazole cycle (centre N-3 in Table 4); herewith, the aromaticity of the macro-
cycle decreases. In the further successive interaction with two acetic acid molecules, there
occurs the protonation of nitrogen meso-atom. Further, acetate ion attacks the α-carbon
atom of the pyrrole fragment, and the bond C–N in the macrocycle breaks down.
Quantum chemical studies were carried out by the AM1 method to elucidate the most
reactive nucleophilic centres (nitrogen atoms) in the molecule of the monoprotonated spe-
cies of triazoleporphyrazine. Table 4 presents the main thermodynamic characteristics of
diprotonated species, which show that the most probable sites of attack by acetic acid mol-
ecules are nitrogen atoms N-8 and N-10 of the macrocycle (Table 4).
We considered the effect of metal on the basicity of the triazoleporphyrazine macro-
cycle by the example of acid-base interaction of Cu(II) and Ni(II) complexes with 1H-hexa-
kis(4-tert-butyl)triazoleporphyrazine 1 and 2, of a nickel(II) complex with 3,4-di(4-tert-bu-
tylphenyl)dibenzotriazoleporphyrazine 3 in benzene–AcOH and dichloromethane–AcOH
media, as well as of a Ni(II) complex with [4-(n-triphenylmethylphenoxy)] 7,8:12,13:17,
18-tribenzotriazoleporphyrazine 4 in a dichloromethane–AcOH medim.
242 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
Table 4 Major thermodynamic characteristics of diprotonated species (AM1 method, 298 K).
2 1 H
N N
10 a 4
N N
N
3
. 2H
d N 9 5 N b
H
N N7 N
8 c 6
N N N N
N N N N
N N
N Ni N N Cu N
N N N N N N
38a 38b
N N N N
N N N N
N N
N Ni N N Ni N
N N N N
N N
O CPh3
42 43
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 243
0.5
628
1
2
0.4 3
4
0.3 5
6
7
0.2
0.1
0.0
400 450 500 550 600 650 700 750 λ, nm
Based on the dependences logIi = f(H0) and logIi = f(logCAcOH), it has been found
that one proton is involved in acid-base interaction (the slope of these lines in a
benzene–AcOH medium for compounds 38a is 0.89; 38b, 1.14; 42, 1.07; and in a
dichloromethane–AcOH medium for compound 42, 0.80; for 43, 0.82) (Figs. 17, 18).
The high basicity of the macrocycle of triazoleporphyrazine complexes, a hypsochro-
mic shift of the Q band in the EAS suggest that the protonation centre is a nitrogen atom in
the pyrrole cycle in position 1. Quantum chemical studies of protonated species of triazole-
porphyrazine we recently carried out [70] support this suggestion.
244 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
log I
0.9
0.6
0.3
0.0
5.4 5.7 6.0 6.3 6.6 6.9 H0
-0.3
Figure 17 Dependence of log I on H0 for the acid-base interaction of compounds 38a,b and 42 in a
benzene–acetic acid medium: °, compound 38a; •, compound 38a; S , compound 42.
log I
0.9
0.6
0.3
0.0
-1.0 -0.5 ,0.0 0.5 1.0 log CCH3COOH
-0.3
-0.6
Figure 18 Dependence of log I on log C AcOH for the acid-base interaction of compounds 42 and 43
in a dichloromethane–acetic acid medium: S , compound 42; •, compound 43.
Experimental data show that nickel and copper complexes are more basic than the
ligand; and the copper complex is more basic than the nickel complex. The presence of the
triphenylmethylphenoxy substituent in the molecule of triazoleporphyrazine decreases the
basicity of the macrocycle.
N
N NH
N
N
NH
N
N
44
The structure of compound 44, as well as of its complexes with Ni 44a, Cu, Co, Mn,
Zn and Ge, was studied using the X-ray diffraction analysis [74–79].
It has been found that the molecule of symmetric macroheterocyclic compound 44,
constructed from isoindole and pyridine nuclei, is in a solid state capable of acquiring both
the planar and nonplanar configuration depending on the nature of complex-forming metal,
as well as on the presence or absence of solvated molecules of solvent.
Hecht and Luger [75] present information on the studies of symmetric benzene mac-
roheterocyclic compound 45 by the XDA method. It has been shown that in the crystalline
state the macrocycle exists as a solvate, which includes molecules of ethanol and water and
has a nonplanar structure.
N
NH
N N
NH
N
45
N(1s)
O(1s)
N(3)
N(5)
O(2s)
N(2s)
45a 45b
Figure 20 Structure of complexes 45a,b according to the X-ray diffraction analysis data. (Hydro-
gen atoms at carbon atoms are omitted for clarity.)
45PL, D2h, ∆Hf = 259.8 kcal·mol –1 45A, C2h, ∆Hf = 250.4 kcal·mol –1
45B, C2v, ∆Hf = 242.5 kcal·mol –1 45*DMF, C1, ∆Hf = 200.0 kcal·mol –1
Figure 21 Spatial structure of symmetric benzene Mc 45 according to the data of quantum chemical
calculations by the AM1 method.
to 0.6 kcal·mol –1, and for the inverse transition this characteristic corresponds to 8.5
kcal·mol –1. The so small conformation-transition energies commensurable with the energy
of the hydrogen bond suggest that similar conversions can occur in solvents. Herewith, de-
pending on the nature of the solvent, one or another conformation can be stabilized. This
conclusion drew confirmation from a recently published work [82], which, using tert-
butyl-substituted Mc as an example, considered the effect of solvents by means of the 1H
NMR method.
Thus, symmetric-structure Mc are structurally nonrigid molecules capable of existing
as various conformers separated by low activation barriers. The predominant content of this
or that species in crystalline state and in solution is determined by the character of intermo-
lecular interactions, the significant of which is specific solvation.
It should be noted that unsubstituted Mc are hard to dissolve in organic solvents. How-
ever, the solubility of Mc can be significantly increased by introduction of bulky substitu-
ents, e.g., tert-butyl groups.
Symmetric-structure tert-butyl-substituted Mc 46–50 were synthesized in accordance
with Scheme 19 by the interaction of 4-tert-butylphthalodinitrile or respective alkoxy- or
three-link compounds with aromatic diamines in ethylene glycol or butanol. It proved the
most expedient to synthesize compounds 46–50 from 4-tert-butylphthalodinitrile via re-
spective alkoxy compounds, without isolating the latter from the stock (Scheme 12).
OAlk N R N
CN AlkONa,
AlkOH R(NH2)2 R(NH2)2
N NH HN
CN
NH NH HN
R(NH2)2 R(NH2)2
N R N N R N Ln
M(CH3COO)2
NH HN N M N
N R N N R N
46 – 50 46a-c – 50a-c M = Cu (a);
Co (b);
N NH N N Ph N N (1-Nph ) Ni (c)
R=
N N N N
46 47 48 49 50
Scheme 12
At the introduction of tert-butyl groups, the character of Mc electronic spectra remains
almost the same. Usually, a small long-wavelength shift of the bands is observed, with ab-
sorption in the region of 300–450 nm preserved.
The experimental values obtained by means of NMR spectroscopy are consistent with
these data. Proton signals of imino groups are in the region of 10.2–12.6 ppm. The presence
of signals of intracyclic protons in a weak field is experimental proof of the absence of a
unified macrocyclic conjugation system in compounds 46–50.
The conclusion of the absence of a unified aromatic system in the macrocycles is con-
sistent with the results obtained in studies of a germanium complex of the symmetric pyri-
dine macroheterocycle by 1H NMR spectroscopy [84]. A later work [85] registered signals
of intracyclic protons in a weak field at 15.7 and 15.2 ppm (CDCl3) in an 1H NMR spectrum
of symmetric-structure tert-butyl-substituted Mc with fragments of 1-dodecyl-1,2,4-triaz-
ole containing two nitrile groups in one of the isoindole fragments.
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 249
CN
N N
3 PhOH N
3
CN H2N NH2 S
S N
N N N
N H
N
N H S
N
NH N H
N N N
EtOEtOH N
N N S
3 NH 3 N
H 2N NH2
S
NH
Scheme 13
250 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
N N N
NH S H S
N N
N N i N N
3 NH + 3 N N
H2N S NH2 H H
NH N N
N N
N S N
52
Scheme 14
The structure of Mc was established using the methods of electron, IR, 1H and 13C
NMR spectroscopy, as well as mass spectrometry data.
Thus, the PMR spectrum of compound 51 measured in deuterated chloroform, along
with signals in the region of 1.40–1.25 ppm, which characterize absorption of protons of
tert-butyl groups, and signals in the region of 7.94–7.61 ppm, which are evoked by absorp-
tion of protons of benzene nuclei of isoindole fragments, yielded a singlet in the region of
12.35 ppm, which vanishes upon addition of deuterated water and can be assigned to the
absorption of protons of imino groups. The location of the latter absorption in the weak field
indicates the nonaromatic character of the macrocycle.
It should be emphasized that under similar conditions, an ABAB-type macrocycle is
formed from 3,5-diamino-1,2,4-triazole. The cause of so significant differences in the struc-
ture of the condensation products is the nature of the thiadiazole fragment. As shown by the
XDA data of 3,5-diamino-1(H)-1,2,4-triazole [91] and 2,5-diamino-1,3,4-thiadiazole [92],
the lengths of S-C bonds in the thiadiazole cycle are much larger than the respective N-C
lengths in the triazole cycle. As the result, the angle between the C-NH2 bonds in the mol-
ecule of 2,5-diamino-1,3,4-thiadiazole proves to be 12.5° larger than that in the molecule
of 3,5-diamino-1(H)-1,2,4-triazole, which predetermines the possibility of forming macro-
heterocyclic compounds of larger size in the case of 2,5-diamino-1,3,4-thiadiazole.
The spectral curves obtained for solutions of Mc 51, 52 (Fig. 22) in chloroform have
an unusual appearance and are practically the same. Two most intensive absorption bands
are observed in the region of 392 and 413 nm for 51 and at 428 and 452 nm for 52, as well
as two bands of average intensity respectively at 463 and 501 nm and at 515 and 552 nm.
The presence of the main absorption in a so short-wavelength region (390–450 nm) indi-
cates the nonaromatic character of the macrocycle.
The features of the electronic and geometric structure of ABABAB-type macrohetero-
cyclic compounds were studied using semiempirical quantum chemical methods [90].
Analysis of the potential rotation energy surface of thiadiazole nuclei has shown that these
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 251
1.5
D
S
N 392 413
N
N N N
N H
N 428 452 S
N
N H S N
N N N N
N N H
H
1 N N
N
N
N H
N
S
S N
N
N H
N N N
N
S
N
0.5
0
250 350 450 550 650
Wavelength, nm
Figure 22 Electronic absorption spectra (solvent, CHCl3) of 51, c = 2.63 g-mol·l –1; of 52, c =
1.65·10 –5 g-mol·l –1.
molecules are structurally nonrigid. The most preferable configurations are nonplanar ones;
in them, sulfur atoms of thiadiazole fragments are oriented outside. Six bridge atoms of ni-
trogen lie in one plane, whereas thiadiazole nuclei as well as pyrrole or isoindole nuclei are
out of this plane. The result is an umbrella-like configuration (Fig. 23).
The so unusual structure of Mc, containing three thiadiazole fragments and three pyr-
role or isoindole fragments, assumes the occurrence of unusual coordination properties as
compared with porphyrazines and phthalocyanines. These Mc proved to be capable of
forming stable complexes of composition 3:1. What is more, these metal complexes could
be synthesized both by the interaction of metal-free compounds with salts of respective met-
als (stepwise synthesis) and by heating a mixture of initial substances in the presence of a
metal salt (template synthesis).
The complexes were synthesized by the interaction of metal-free Mc with metal salts,
252 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
taken in a molar ratio of 1:3–5, in DMFA at 100°C. As complex formers, metals were taken
which have relatively small covalent radii (Ni, Cu, Co) and are widely used for the synthesis
of metal complexes based on macroheterocyclic compounds.
3
R 3 R
S N R
N N N N
N
R M
N M N S 3X
N N R N
N N N M N
N N
S N M S S N
N 3X R
N N R
N N
M M
N N
N N
R N S N R
Metal complexes 51a–c and 52a–c are formed with good yields. Their structure is
confirmed using mass spectrometry, electron and IR spectroscopy, as well as flame ioniza-
tion spectroscopy. Mass spectra of compound 52a are observed to have peaks correspond-
ing to a complex with three metal atoms (Fig. 24). The isotopic composition of the signals
for the complex obtained completely coincides with that calculated theoretically.
Rather unexpected was also the fact that, besides clusters of signals corresponding to
the structure of compounds with three metal atoms [3+3+3M]+, the mass spectra of the
complexes obtained have intensive peaks corresponding to structures [3+3+3M+O]+, in
which one atom of oxygen is present instead of acetate or chloride anions.
OR H2N
CN N N
NC SAlk
H2N S NH2
CN
OR NC SAlk
Ni(OAs)2*4H2O BuOH EtOEtOH Ni(OAs)2*4H2O
RO
RS
S N OR SR
N S N
RO N N N N
N
X
Ni
N N N N
Ni N
N Ni X X S RS X N
N S
Ni N N Ni X X N
RO N N N N RS Ni
OR N N N N
S N N
RO S N SR
53 R = C5H11 55 R = C4H9
54 R = C10H21 56 R = C18H17
X = OAc(OH) 57 R = C12H25
X = OAc(OH)
Scheme 15
7 Coordination Properties
Macroheterocyclic compounds of ABABAB type – structural analogues of hexaphyrin
[93] – are interesting objects for studies by coordination chemistry methods. As per the
classification proposed in [94] for monomacrocyclic ligands, compound 58 in accordance
with the type of heteroatoms should be assigned to the group of mixed polythiaazamacro-
cycles with a set of donor atoms SnNm. However, the specific structure, namely the “extra-
cyclic” arrangement of atoms of S and the formation of the internal “cavity” by three pyrrole
and six thiadiazole atoms of N, provides macroheterocyclic compound 52 with properties
of an N-donor ligand (“polyazamacrocycles with donor atoms Nm”).
Lomova et al. [95] were the first to determine the acid-base characteristics and stability
of the ABABAB-type macrocycle to the action of acids by the example of macroheterocy-
clic compound 52 constructed from six successively alternating fragments of thiadiazole
and 3,4-bis(4-tert-butylphenyl)pyrrole linked by six aza bridges (McH3).
Owing to the presence of six 4-tert-butylphenyl substituents, compound (McH3) 52
dissolves in nonpolar organic solvents, thus forming true solutions with a concentration of
(0.1–10)·10 –5 mol·l –1. In transition from neutral organic solvents to individual and mixed
protolytic solvents, the EAS of compound 52 are observed to have noticeable changes,
ooooooo
254 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
N N N
N N N N
N H N N
S S
S H S
N N
N N
N N
H H N N
N N
H H
S N N
N N N N
N N N N
S
58 52
which depend on the concentration of acid (Fig. 25). Experimentally, by dilution of solu-
tions with respect to acid, these changes in a benzene–AcOH medium were shown to be
reversible. The protonation process of McH3 in a benzene–AcOH proton-donor medium at
different temperature and acidities (H0 = 4.44–6.61) was studied by the method of spectro-
photometric titration (Fig. 26). It was found experimentally that protonation is a two-stage
process (equations (5) and (6)). Respective protonation constants K1 and K2 were found by
the method of least squares using Microsoft Excel by the mass action law equation (3) made
up for a three-component system with account for the proportionality of concentrations and
optical densities (A0, A∞ and Ap), the neutral and protonated species of McH3 and their equi-
librium mixtures at the working wavelength of 449 nm.
A 1
2
A
0.08
0.06
0.04
0.02
4.2 5.2 6.2 7.2 H0
Figure 26 A curve of spectrophotometric titration of McH3 in benzene by acetic acid at 298 K, the
first stage; working wavelength, 449 nm.
log I
1.2
tan α = 1
0.6
0
5.5 6 6.5 7
-0.6 H0
-1.2
Figure 27 Dependence of log I on H0 for McH3 in a benzene–acetic acid medium, the first stage
(I = (Ap – A0)/(A∞ – Ap), correlation coefficient 0.99, tan α = 1.15.
Ap − A0
A∞ − A0 1
K= × . (7)
Ap − A0 ⎛ Ap − A0 ⎞
n
1− 0
A∞ − A0 ⎜⎜ H 0 − CMcH3 × A − A ⎟⎟
⎝ ∞ 0 ⎠
Here n is the number of protons attached in one stage. The numerical values of K (Ta-
ble 6) and n, equal to 1 at both stages, were found by optimizing the dependence in coordi-
nates of equation (7) reduced to a linear form: log(Ap – A0)/(A∞ – Ap) – H0 (Fig. 27). The
Table 6 Equilibrium constants K1 and K2 of the McH3 protonation
reaction in the benzene–acetic acid system at various temperatures.
298 0.14
308 0.23
321 0.36 1.3
256 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
-ln K1 3
0
3.1 3.2 3.3 3.4
1/T 10-3
521 nm
1
2
λ, nm
1/A 120
1
90
60
2
30
3
0 40 80 120 160
τ, min
Figure 30 Dependence of 1/A on time (τ) for McH3. C 0H2SO4, mol/l: 17.1 (2); 17.4 (1, 3). T, K: 343
(1); 333 (2); 324.2 (3).
protonation equilibrium and to determine how many protons are attached additionally in
changing from AcOH to H2SO4. However, it can be assumed that in a sulfurous solution
for (McH3)Hnn+ n = 3 (for a solution in AcOH, n = 2), because up to 100% H2SO4 the EAS
with the band of 521 nm does not undergo changes, i.e., no new protonated species are ob-
served.
The destruction kinetics of (McH3)Hnn+ was studied quantitatively in H2SO4 with a
concentration of 17.1–18.1 mol ·l–1 at temperatures of 324–343 K. The kinetic parameters
of the destruction reaction are presented in Table 7. The order of the reaction with respect
to the concentrations of McH3 (Fig. 30) and H3O+ were found experimentally (Fig. 31) and
Table 7 Efficient rate constants (kef), energy (E) and entropy (∆S #) of activation of the McH3
destruction reaction in concentrated H2SO4.
are equal to 2 and 3, respectively. Kinetic equation (8) is treated with account for the run
(in the reaction system) of equilibrium protonation reaction of (McH3)H22+ and water and
the irreversible destruction of the macrocycle (equations (9)–(11)).
log [H3O+]
0.7 0.9
-1
1
2
-3
3
-5
tan α = 3
log ke -7
Figure 31 Dependence of log keff of the destruction of McH3 on log[H3O+]. T, K: 343 (1); 333 (2);
324.2 (3).
KK31
(McH3)H22+ + H3O+ (McH3)H33+ + H2O, (9)
KK41
H2O + H2SO4 H3O+ + HSO4– , (10)
1 A0 − A∞
keff = ln . (12)
τ Aτ − A∞
Table 8 Efficient rate constants of the complexation reaction of nickel(II) with macrocyclic ligand
52 in DMFA(a).
The polydentate ligand with an extended coordination cavity and a large amount of
electron-donor heteroatoms – potential coordination cores – features a rather complex ki-
netics of complexation reactions. In the course of conversion, the character of the McH3
absorption spectral curve changes significantly and, according to the data of Fig. 32 and
[98], coincides in the end of the reaction with the spectrum of a complex compound of the
3:1 metal–macrocyclic ligand composition.
The logarithm log keff linearly correlates with the value of log[Ni(OAc)2] (equation
A
0.4 1
0.3
2
0.2
0.1
3
0.0
400 450 500 550 600 λ, nm
Figure 32 Electronic absorption spectra of McH3 in DMFA (1), of the complexation product of
McH3 with Ni(OAc)2 in DMFA (2) and, after isolation from the reaction mixture, in CHCl3 (3).
260 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
log ke
-4.0
1
-4.5
Figure 33 Dependence of log keff on logCNi(OAc)2 for the reaction of macrocyclic compound McH3
with Ni(OAc)2 in DMFA. T, K: 353 (1), 358 (2) (ρ equals 0.93 and 0.99, respectively).
(13), Fig. 33) only within a narrow range of concentration of the salt (1.58–1.87)·10 –4
mol·l –1, where the order close to two is observed with respect to [Ni(OAc)2]. Respectively,
only for this interval of reaction mixture concentrations with respect to the nickel salt we
can write down kinetic equation (14). At Csalt close to 1.47·10 –4 and 2.06·10 –4 mol·l –1
the order is, respectively much lower and higher than two.
Experimental kinetic equation (14) is treated with account for the reversible and irre-
versible reactions (equations (15)–(18)), where acetate ion is dropped in writing the formu-
la of the salt):
KK11
(McH3)2 2 McH3, (15)
kK2 1
McH3 + Ni2+ [NiMcH2]+…H+, slow, (16)
KK31
[NiMcH2]+ + Ni2+ [Ni2McH]2+…H+, (17)
kK41
[Ni2McH]2+ + Ni2+ [Ni3Mc]3+ + H+, slow. (18)
The rate of the limiting stage (14) is equal to k4 ·С[Ni2McH]2+…H+ ·СNi2+ or, after ex-
pressing the concentration [Ni2McH]2+…H+ with account of equilibrium (17): k4 ·K3 ·
С[NiMcH2]+…H+ ·(СNi2+)2. The latter coincides with the right-hand side of experimental
equation (18) owing to the equilibrium (19) and a fast irreversible transition of McH3 into
[NiMcH2]+…H+ (equation (16)). Thus, the experimental rate constant is k = k4 ·K3.
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 261
The value of K1 is first determined experimentally in this work and calculated by equa-
tion (19): K1 = (1.6±0.18)·105 mol –1 ·l. The equilibrium with constant K1 (equation (15))
proves kinetically significant, because it provides for the presence of McH3 molecules in
DMFA solutions during the complexation in predominant concentrations as compared with
the dimer (McH3)2.
Ap − A0
A∞ − A0 1
K= (19)
Ap − A0 0 Ap − A0
1− Cs − CMcH
A∞ − A0 3 A − A
∞ 0
(in the equation, C0McH3 and Cs are the initial concentrations of, respectively, McH3 and
DMFA; A0, Ap, A∞ are the optical densities at a working wavelength of 420 nm for solutions
of (McH3)2, equilibrium mixture at a definite concentration of DMFA and McH3).
It is characteristic that the second order with respect to the salt is disturbed at the broad-
ening of the salt concentration interval. The assumed scheme of stepwise reactions explains
this regularity. When the consentation of the salt is increased, stages (16) and (17) merge
into one equilibrium stage, which in the expression of the concentration [Ni2McH]2+…H+
leads to the third order with respect to the salt. In contrast, at low CNi2+ the coordination
reaction does not proceed to the end, and stage (16) becomes slow and limits the entire pro-
cess. Respectively, the order with respect to the salt becomes equal to unity.
The high sensitivity of the reaction to the concentration of the salt enables an assump-
tion on the detailed mechanism of limiting stage (18) as a dissociative one according to the
Langford–Gray classification [98]. The temperature dependence of the rate constants is
nonlinear, which does not make it possible to obtain the value of activation energy and, re-
spectively, entropy for the reaction. If the narrow temperature range (353–358 K) is used,
one can obtain the activation energy, regularly decreasing with the salt concentration rise
from 82 down to 25 kJ·mol –1. The value of ∆S # in this case changes from –109 down to
–266 J·(mol·K) –1. Large negative values of activation entropy (–109 to –266
J·(mol·K) –1) indirectly confirm the statement of the dissociative mechanism of reaction
(14), thus indicating a more ordered transitory state of the reacting system as compared with
the initial state. Indeed, the polar transitory state [[Ni2McH]2+ ·[Ni(OAc)2(DMFA)3]] #, as
the state formed from the pentacoordinated nickel complex (the dissociative mechanism of
reaction (14)) and from the macrocyclic ligand polarized by the N–H bond, is, probably,
more solvated in the medium of the main solvent, DMFA.
Thus, most probably, the nickel coordination reaction of the macrocyclic ligand with
an extended coordination cavity McH3 proceeds as follows: the macrocycle is readily co-
ordinated with one nickel ion, the second ion is coordinated with greater difficulty, and the
reaction does not proceed to the end (equilibrium (17) is established), the third nickel ion
is coordinated with even greater difficulty, also, apparently, reversibly, however, due to the
low rate of the direct reaction, no equilibrium is achieved under the conditions of the ex-
periment. The stepwise entry of metal cation into the coordinating space of McH3 opens
prospects for finding conditions of synthesizing heteronuclear complexes with ligands of
similar structure and for their practical applications.
262 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
Table 9 Prediction of the range of biological activity of Mc, their metal complexes and initial
substances for their synthesis.
N N N
Kconf 15.7 18.9 – – 17.6
NH HN
N N
NH HN
Keff 1.7 – – 9.2 3.1
NH HN IV
N NH
N N N
Kconf 14.8 28.9 – 15.4 6.6
N Cu N
Keff 2.0 8.5 – 5.5 43.7
N N N
V
N NH
N NH
N N N
Kconf 14.8 28.9 – 15.4 2.8
N Co N
Keff 2.0 8.5 – 5.5 3.3
N N N
VI
N NH
264 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
Table 9 Prediction
(continued)of the range of biological activity of Mc, their metal complexes and initial
substances for their synthesis.
N NH
N N N
Kconf 14.8 28.9 – 15.4 6.6
N Ni N
Keff 2.0 8.5 – 5.5 43.7
N N N
N NH
VII
N N
N S N
Kconf 19.7 21.1 – 25.3 11.8
N Zn N
VIII
Note: For reading convenience, in this section we use different numeration of compounds.
As it follows from the data presented in Table 9, for all tested compounds one should
expect the manifestation of an antitumour activity, which should decrease in the following
sequence: II > III > I = V = VI = VII > IV > VIII.
In this connection, chemical compounds presented in Table 10 were tested for the an-
titumour activity on a model of transplantable tumour L-1210.
Table 10 Indices of antitumour activity of compounds at repeated intraperitoneal injection of DBA
with L-1210 to female mice.
1 3 2 4 5 6
(continued)
Table 10 Indices of antitumour activity of compounds at repeated intraperitoneal injection of DBA
with L-1210 to female mice.
1 3 2 4 5 6
Preliminarily, maximum tolerated doses (MTD) were determined for each compound,
based on which daily therapeutic doses were chosen. The values of MTD made it possible
to assign all tested compounds to toxicity class IV (low-toxic compounds) [111].
Experimental studies of the antitumour activity were carried out under conditions of
intraperitoneal introduction of preparations. The results of this part of work are presented
in Table 10.
As it follows from the table, compounds 3,5-diamino-1,2,4-triazole (II) and the nickel
complex of symmetric-structure macroheterocyclic compound with fragments of 3,5-di-
amino-1,2,4-triazole (VII) exhibit a moderate antitumour activity with respect to lymphoid
leukemia L-1210.
In the sequence of triazole-containing compounds studied, two of them (3,5-diamino-
1,2,4-triazole and the nickel complex of Mc with fragments of 1,2,4-triazole) possess a
moderate antitumour activity. By the value of the increase of the average life expectancy,
266 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
these compounds can be arranged in the following sequence: I > VII > III > V > VI > IX.
Thus, the antitumour activity of triazole derivatives studied depends both on the struc-
ture of the organic moiety of the molecule and on the nature of metal complex former. In-
troduction of metal into the Mc molecule significantly increases the value of ALEI, thus
weakening the negative effect of descriptors present in the Mc molecule. The nickel com-
plex of Mc VII proved the most efficient in the sequence of all macrocycles and their com-
plexes studied, while the cobalt and copper complexes exhibited no activity with respect to
leukemia L-1210.
In the sequence of 1,3,4-thiadiazole derivatives, no manifestation of antitumour activ-
ity was found for any of the compounds studied, and the TZP with a fragment of 1,3,4-thi-
adiazole IV was even capable of significantly initiating the development of the disease
(Table 10). 2,5-Diamino-1,3,4-thiadiazole I itself exhibited a weak activity with respect to
leukemia L-1210, which is consistent with the literature data [112].
The results obtained served as the basis for continuation of works on the assessment
of the antitumour and antimetastatic activities of compounds II and VII on a solid LLC
(Lewis-lung carcinoma) tumour resistant to all currently known antitumour preparations.
The results of studies are presented in Table 11.
Table 11 Indices of antitumour and antimetastatic activity of compounds in female mice C57BL/6
in repeated injection.
As indicated by the data presented, the compounds studied exhibited no specific ac-
tivity with respect to LLC tumour.
Thus, of the compounds we studied, only 3,5-diamino-1,2,4-triazole and the nickel
complex of Mc with fragments of 3,5-diamino-1,2,4-triazole exhibited a moderate anti-
tumour activity on an L-1210 lymphoid leukemia model, which gives grounds for directed
synthesis and search for more active compounds among their analogues (1,2,4-triazole de-
rivatives) for subsequent works on the development of antitumour means.
In conclusion, it should be noted that accumulated material is too little to judge on con-
crete structure–property dependences. At the same time, the already available results
enable a statement to be made that Mc are promising compounds for the search of substanc-
es with interesting biological properties.
The work was supported by the grant 05-03-33003a from the Russian Foundation for Basic
Research.
References
1. R.K. Pandey and G. Zheng, Porphyrins as Photosensitizers in Photodynamic Therapy, in:
Porphyrin Handbook, vol. 6, ed. by K.M. Kadish, K.M. Smith and R. Guilard, Academic Press:
New York, Chapter 43, pp. 157–230 (2000).
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 267
32. M.E. Baguley and J.A. Elvidge, J. Chem. Soc., 709–719 (1957).
33. F. Fernández-Lazáro, T. Torres, B. Hauschel and M. Hanack, Chem. Rev., 98 (2), 563–576
(1998).
34. M.K. Islyaikin and A. Baranski, Khim. Geterotsykl. Soed., 8, 1047–1055 (2001).
35. P. Bamfield and P.A. Mack, J. Chem. Soc. C, 1961–1964 (1968).
36. V.A. Gnedina, R.P. Smirnov and G.A. Matyushin, Izv. Vuzov, Khim. Khim. Tekhnol., 16 (4),
589–592 (1973) (in Russian).
37. E.V. Kudrik, M.K. Islyaikin and R.P. Smirnov, Zhurn. Org. Khim., 33 (7), 1107–1110 (1997)
(in Russian).
38. L. Latos-Grazynski, Core-Modified Heteroanalogues of Porphyrins and Metalloporphyrins, in:
Porphyrin Handbook, vol. 2, ed. by K.M. Kadish, K.M. Smith and R. Guilard, Academic Press:
New York, Chapter 14, pp. 361–416 (2003).
39. M.S. Rodríguez-Morgade, Gema de la Torre and T. Torres, Design and Synthesis of Low-
Symmetry Phthalocyanines and Related Systems, in: Porphyrin Handbook, vol. 15, ed. by K.M.
Kadish, K.M. Smith and R. Guilard, Academic Press: New York, Chapter 99, pp. 125–160
(2003).
40. N. Kobayashi, Synthesis and Spectroscopic Properties of Phthalocyanine Analogs, in: Porphyrin
Handbook, vol. 15, ed. by K.M. Kadish, K.M. Smith and R. Guilard, Academic Press: New
York, Chapter 100, pp. 161–262 (2003).
41. R.P. Smirnov, V.F. Borodkin and G.I. Lukyanova, Izv. Vuzov, Khim. Khim. Tekhnol., 1,
118–121 (1964) (in Russian).
42. L.M. Fedorov, A.I. Mikhailov, R.P. Smirnov, N.A. Kolesnikov and M.I. Alyanov, Izv. Vuzov,
Khim. Khim. Tekhnol., 15 (12), 1817–1820 (1972) (in Russian).
43. L.M. Fedorov, R.P. Smirnov and G.P. Shaposhnikov, Chemistry and Technology of Dyeing,
Synthesis of Dyes and Polymer Materials, Interinstitute Collected Volume, No 1, pp. 57–61
(1973) (in Russian).
44. R.P. Smirnov, B.D. Berezin and L.A. Skvortsova, Zhurn. Fiz. Khim., 41 (7), 1630–1634 (1967)
(in Russian).
45. R.P. Smirnov and B.D. Berezin, Zhurn. Fiz. Khim., 43 (10), 2494–2498 (1969) (in Russian).
46. R.P. Smirnov, V.N. Kharitonov and L.N. Smirnov, Proc. Ivanovo Institute of Chemistry and
Technology, 4, 111–116 (1972) (in Russian).
47. Chemical Additives to Polymers (reference book), A.S. Baranova, G.S. Baryshnikova, N.S.
Glazunova, V.M. Delyustro, K.A. Zolotareva, I.P. Maslova, L.A. Pugacheva and L.A. Skripko,
Khimiya: Moscow, 264 pp. (1981) (in Russian).
48. N.A. Kolesnikov, L.M. Fedorov and E.E. Kolesnikova, Izv. Vuzov, Khim. Khim. Tekhnol.,
19 (9), 1352–1354 (1976) (in Russian).
49. V.F. Borodkin, T.I. Chesnokova and M.K. Islyaikin, Izv. Vuzov, Khim. Khim. Tekhnol., 23 (8),
1044–1046 (1980) (in Russian).
50. T.M. Ivanova, M.Yu. Bazanov, A.V. Petrov and E.S. Yurina, Koord. Khim., 32 (1), 75–78
(2006) (in Russian).
51. R.K. Bartlett, L.V. Renny and K.K. Chan, J. Chem. Soc. C, 1, 129–133 (1969).
52. F. Fernández-Lázaro, A. Sastre and T. Torres, J. Chem. Soc. Chem. Com., 1525–1526 (1994).
53. M. Nicolau, B. Cabezón and T. Torres, Coord. Chem. Rev., 190–192, 231–243 (1999).
54. S. Esperanza, M. Nicolau and T. Torres, J. Org. Chem., 67, 1392–1395 (2002).
55. F. Armand, M.V. Martínez-Díaz, B. Cabezón, P.-A. Albouy, A. Ruaudel-Teixier and T. Torres,
J. Chem. Soc. Chem. Commun., 1673–1674 (1995).
56. F. Armand, B. Cabezón, M.V. Martínez-Díaz, A. Ruaudel-Teixier and T. Torres, J. Mater. Chem.,
7 (9), 1741–1746 (1997).
57. F. Armand, B. Cabezón, O. Araspin, A. Barraud and T. Torres, Synth. Metals, 84, 879–880
(1997).
58. B. Cabezón, F. Fernández-Lázaro, M.V. Martínez-Díaz, S. Rodríguez-Morgade, A. Sastre and
T. Torres, Synth. Metals, 71, 2289–2290 (1995).
59. G. Rojo, F. Agulló-López, B. Cabezón, T. Torres, S. Brasselet, I. Ledoux and J. Zyss, J. Phys.
Chem. B, 104, 4295–4299 (2000).
60. B. Cabezón, M. Nicolau, J. Barberá and T. Torres, Chem. Mater., 12, 776–781 (2000).
61. V. Stefani, B. Cabezón, E.L.G. Denardin, D. Samios and T. Torres, J. Mater. Chem., 10,
2187–2192 (2000).
Synthesis, Structure Peculiarities and Biological Properties of Macroheterocyclic Compounds 269
62. M.K. Islyaikin, V.R. Ferro and J.M. García de la Vega, J. Chem. Soc. Perkin Trans., 2 (12),
2104–2109 (2002).
63. T.M. Krygovski and M.K. Cyranski, Chem. Rev., 101, 1385–1419 (2001).
64. P.v.R. Scheyer, C. Maerker, A. Dransfeld, H. Jiao and N.J.R.v.E.J. Hommes, Am. Chem. Soc.,
118 (26), 6317–6318 (1996).
65. M.K. Cyranski, T.M. Krygovski, M. Wiciorowski, N.J.R.v.E. Hommes and P.v.R. Shleyer,
Angew. Chem. Int. Ed., 37 (1/2), 177–180 (1998).
66. L.E. Marinina, S.A. Mikhalenko and E.A. Lukyanets, Zhurn. Org. Khim., 43 (9), 2025–2029
(1973) (in Russian).
67. P.A. Stuzhin, O.G. Khelevina and B.D. Berezin, Phthalocyanines: Properties and Applications,
ed. by C.C. Leznoff and A.B.P. Lever, VCH: New York, vol. 4, pp. 19–77 (1996).
68. P.A. Stuzhin and O.G. Khelevina, Koord. Khim., 24 (10), 783–793 (1998) (in Russian).
69. P.A. Stuzhin, A. Ul-Hak, N.V. Chizhova, A.S. Semeykin and O.G. Khelevina, Zhurn. Fiz.
Khim., 72 (9), 1585 (1998) (in Russian).
70. O.G. Khelevina, V.R. Ferro, M.K. Islyaikin, E.A. Veselkova, M.G. Stryapan, J.M. García de la
Vega, J. Phys. Org. Chem., 18 (4), 329–335 (2005).
71. B.D. Berezin and O.G. Khelevina, Porphyrins: Structure, Properties, Synthesis, ed. by N.S.
Enikolopyan, Nauka: Moscow, p. 83 (1985) (in Russian).
72. J.A. Evidge and R.P. Linstead, J. Chem. Soc., 5008–5012 (1952).
73. J.B. Campbel [E.I. du Pont de Nemours and Co.], US Patent 2765308, Applied for 15.08.1952,
Issued 02.10.1956.
74. J.C. Speacman, Acta Cryst., 6 (10), 784–791 (1953).
75. H.-J. Hecht and P. Luger, Acta Crystal., 30 (12), 2843–2848 (1974).
76. W. Hiller, J. Sträle, K. Mitulla and M. Hanack, Liebigs Ann. Chem., 1946–1951 (1980).
77. S.-M. Peng, Y. Wang, T.-F. Ho, I.-C. Chang, C.-P. Tang and C.-J. Wang, J. Chem. Soc. (Taipei),
33, 13–21 (1986).
78. S.-M. Peng, Y. Wang, C.-K. Chen, J.-Y. Lee and D.-S. Liaw, J. Chem. Soc. (Taipei), 33, 23–33
(1986).
79. S.P. Konovalov and M.K. Islyaikin, Izv. Vuzov, Khim. Khim. Tekhnol., 36 (4), 76–81 (1993)
(in Russian).
80. V.F. Borodkin, V.A. Burmistrov and M.K. Islyaikin, Khim. Geterotsykl. Soed., 1, 62–64 (1981)
(in Russian).
81. O.V. Shishkin, A.Yu. Kovalevsky, M.V. Shcherbakov, M.K. Islyaikin, E.V. Kudrik and A.
Baran- ski, Kristallografiya, 46 (3), 461–464 (2001) (in Russian).
82. V.V. Aleksandriiskii, M.K. Islyaikin, I.G. Shutov, G.A. Zhurko and V.A. Burmistrov, Russ. J.
Phys. Chem., 79 (1), S130–S134 (2005).
83. E.A. Danilova and M.K. Islyaikin, Synthesis and Properties of tert-Butyl-substituted Macro-
heterocyclic Compounds and their Complexes with Metals, in: Advances of Porphyrin Chemist-
ry, Chemistry Research Institute, St.-Petersburg University: St.-Petersburg, vol. 4, pp. 356–375
(2004) (in Russian).
84. J.N. Esposito, L.E. Sutton and M.E. Kenney, Inorg. Chem., 6 (6), 1116–1120 (1967).
85. G. de la Torre and T. Torres, J. Org.Chem., 61, 6446–6449 (1996).
86. E.A. Danilova, M.K. Islyaikin and R.P. Smirnov, Zhurn. Org. Khim., 65 (11), 1882–1884 (1995)
(in Russian).
87. V.F. Borodkin and N.A. Kolesnikov, Khim. Geterotsykl. Soed., 2, 194–195 (1971) (in Russian).
88. N.A. Kolesnikov and V.F. Borodkin, Izv. Vuzov, Khim. Khim. Tekhnol., 15 (6), 880–882 (1972)
(in Russian).
89. N. Kobayashi, S. Inagaki, V.N. Nemykin and T. Nonomura, Angew. Chem. Int. Ed., 40 (14),
2710–2717 (2001).
90. M.K. Islyaikin, E.A. Danilova and L.D. Yagodarova, Izv. Vuzov, Khim. Khim. Tekhnol., 46 (2),
3–7 (2003) (in Russian).
91. G.L. Starova, O.V. Frank-Kamenetskaya, V.V. Makarsky and V.A. Lopyrev, Kristallografiya,
25 (6), 1292–1294 (1980) (in Russian).
92. Hitoshi Senda and Juro Maruha, Acta Cryst., 43, 347–349 (1987).
93. J.L. Sessler and D. Seidel, Angew. Chem. Int. Ed., 42, 5134–5175 (2003).
94. K.B. Yatsimirsky and Ya. D. Lampeka, The Physical Chemistry of Complexes of Metals with
Macrocyclic Ligands, Naukova Dumka: Kiev, 256 pp. (1985) (in Russian).
270 M.K. Islyaikin, E.A. Danilova, Yu.V. Romanenko, O.G. Khelevina and T.N. Lomova
95. T.N. Lomova, E.E. Suslova, E.A. Danilova and M.K. Islyaikin, Zhurn. Fiz. Khim., 79 (2),
263–269 (2005) (in Russian); T.N. Lomova, E.E. Suslova, E.A. Danilova and M.K. Islyaikin,
Russ. J. Phys. Chem., 79 (2), 201–206 (2005).
96. R. Bell, The Proton in Chemistry, Mir: Moscow, 381 pp. (1977) (Russian translation).
97. B.D. Berezin, A Study of the Physico-chemical Properties of Phthalocyanine Complex
Compounds, DSc (Chemistry) Thesis, Inst. of General and Inorganic Chemistry: Kiev, 56 pp.
(1966) (in Russian).
98. C. Langford and H. Gray, Ligand Substitution Processes, Mir: Moscow, 157 pp. (1984)
(Russian translation).
99. B.I. Stepanov, Introduction to the Chemistry and Technology of Organic Dyes, 3rd edn,
Khimiya: Moscow, p. 541 (1984) (in Russian).
100. L.G. Krolik and B.D. Vitkina, Zhurn. VKhO im. D.I. Mendeleeva, 11 (1), 60–69 (1966)
(in Russian).
101. L.M. Fedorov, N.A. Kolesnikov, R.P. Smirnov and M.I. Alyanov, Izv. Vuzov, Khim. Khim.
Tekhnol., 15 (4), 537–540 (1972) (in Russian).
102. R.P. Smirnov, V.V. Andreyanov, Yu.G. Vorobyev, V.A. Shorin and L.M. Fedorov, Izv. Vuzov,
Khim. Khim. Tekhnol., 27 (10), 1239–1241 (1984) (in Russian).
103. C.G. Birch and R.T. Iwamoto, Inorg. Chem., 12 (1), 66–73 (1973).
104. C.G. Birch and R.T. Iwamoto, Inorg. Chem. Acta, 6 (4), 680–682 (1972).
105. M.I. Bazanov, L.V. Kokhova, V.A. Bogdanovskaya, M.R. Tarasevich, R.P. Smirnov and N.A.
Kolesnikov, Izv. Vuzov, Khim. Khim. Tekhnol., 29 (7), 43–46 (1986) (in Russian).
106. M.I. Bazanov, V.V. Kudrinsky, N.A. Kolesnikov, R.P. Smirnov,. Izv. Vuzov, Khim. Khim.
Tekhnol., 36 (10), 59–64 (1993) (in Russian).
107. R.P. Smirnov, A.L. Smirnov, D.G. Snegirev and G.A. Zdorikova, Izv. Vuzov, Khim. Khim.
Tekhnol., 35 (6), 66–71 (1992) (in Russian).
108. L.N. Smirnov, G.A. Zdorikova, A.L. Smirnov and D.G. Snegirev, Izv. Vuzov, Khim. Khim.
Tekhnol., 35 (6), 71–76 (1992) (in Russian).
109. G. Rojo, F. Agulló-López, B. Cabezón, T. Torres, S. Brasselet, I. Ledoux and J. Zyss, J. Phys.
Chem. B, 104, 4295–4299 (2000).
110. Yu.V. Burov, L.V. Korolchenko and V.V. Poroykov, Byul. All-Union Scientific Centre on
Safety of Biologically Active Compounds, Moscow, pp.4–25 (1990) (in Russian).
111. N.F. Izomerov, I.V. Sakotsky and N.N. Sidorov, Toxicity Parameters of Industrial Poisons in
Single Exposure, Meditsina: Moscow, 239 pp. (1977) (in Russian).
112. D.L. Hill, Cancer Rev., 4, 215–220 (1980).
7 The Photochemical Aspect of
Reactions of Flavonols with
Molecular Oxygen
E.A. Venedictov
Institute of Solution Chemistry, Russian Academy of Sciences,
1 Akademicheskaya Street, Ivanovo, 153045, Russia
email: eav@isc-ras.ru
This chapter presents the results of research into the kinetics of photochemical
reactions of biological pigments and related compounds with molecular oxygen.
Introduction
Flavonoids belong to the most widespread natural compounds [1–10]. Their sources are
mainly higher plants. They also occur in algae, lichens, mosses, fungi, ferns and some rep-
resentatives of insects and microorganisms. They are found in roots, wood, leaves, flowers,
fruits. These compounds are present in various cell organelles of plants. They are revealed
in a broad colour palette of the flora and are represented in a whole range of products of
plant raw material processing (juices, wine, oil and extracts).
Owing to the ability of flavonoids to pigment plant tissues, protection of plants from
the damaging action of solar UV radiation was historically considered to be their most sig-
nificant function [6–12]. Subsequently, this point of view found its experimental proof.
The rise of flavonoid biosynthesis in response to an increase of the level of UV radiation,
and the localization of these compounds mainly in the upper cell layer of plant tissues are
considered at present as an adaptation mechanism of plants to their new habitat conditions
[13–15].
In reality, the physiological functions of flavonoids are more diverse. They are be-
lieved to be involved in electron transfer processes, stimulate plant growth and protection
from harmful biological effects occurring with participation of pathogenic organisms and
insects [1–10, 16].
272 E.A. Venedictov
1 Structure of Flavonols
Flavonoids form a large family of natural O-heterocyclic colorants, the main structural unit
of which is 2-phenyl-4-benzopyrone (flavan). For this reason, flavonoids can be considered
as its derivatives. One of the classifications [10] divides these compounds into ten main
classes: catechins, leucoanthocyanidins, flavanols, dihydrochalcones, anthocyanidins, fla-
vanonols, flavones, flavonols and aurons. The structural features of some of them are shown
in Fig. 1.
The structural unit of flavonols is flavone. For this reason, these compounds are often
considered as its derivatives. The structural types of some of them are given in Table 1.
A feature of flavones is the presence of a C2C3 double bond conjugated with the
π-electronic system of cycle B. Emerging cross conjugation between the carbonyl group
and cycles A and B [7] determines the character of their electronic absorption spectra.
Flavonols can contain from one up to six free OH groups [1–4, 7, 9]. Compounds with
etherified OH groups (OR) are known to exist, as a rule, at C-3, C-5 and C-7 carbon atoms.
Methyl and acetyl radicals often occur as substituting R groups, as well as residues of car-
bohydrates (glucoses, rhamnoses, galactoses and others). Flavonols containing 3-, 7- and
4′-OH groups are weak acids and show three stages of titration within the range of pKa
6.7–11.6 [8, 32, 39, 40]. Compounds with the 7-OH group possess the maximum acidity,
whereas the acidity of other OH groups is on the level of phenol and is to some extent related
to the structure of molecules. For derivatives of flavonol with ortho-substitution of hydro-
gen atom in cycle B the acidity of the 3-OH group decreases. The cause of this is nonco-
planarity of cycles B and C, which emerges due to steric hindrances and prevents the
The Photochemical Aspect of Reactions of Flavonols with Molecular Oxygen 273
3'
2' 4' H
8 1 B
O 2 5' O
7 1'
A C 3 6' H
6
5 4 OH
H H
O
I II
O O
OH OH
O
III IV
Figure 1 Structural formulas of some representatives of the family of flavonoids: flavone (I), cat-
echine (II), flavonol (III) and anthocyan (IV).
delocalization of excess electron charge on the C2C3 double bond. The 5-OH group is the
most hard to ionize due to the strong intramolecular hydrogen bond formed with its parti-
cipation [8, 39].
Owing to the structural features of flavonols, their 3- and 5-OH groups and the carbo-
nyl atom of oxygen mutually affect one another, as the result of which cyclic unstressed
intramolecular hydrogen bonds can emerge between them (an H bond) [1, 40–43], as
shown schematically in Fig. 2. These bonds are formed in noncoordinating solvents and
Flavone Substituents
Galangine OH OH H OH H H H H H H
Fisetine OH H H OH H H H OH OH H
Chrysine H OH H OH H H H H H H
Apigenine H OH H OH H H H H OH H
Luteoline H OH H OH H H H OH OH H
Kampferol OH OH H OH H H H H OH H
Morine OH OH H OH H H OH H OH H
Quercetine OH OH H OH H H H OH OH H
Rhamnetine OH OH H M H H H OH OH H
Myricetine OH OH OH H H H H OH OH OH
Rutine P OH OH H H H H OH OH H
O O
O
O O O
H
H
I II
Figure 2 Structure of 3-hydroxyflavone (I) and 5-hydroxyflavone (II) with intramolecular H bonds.
O O
hν
O O
O OH
H
N T
OH OH
OH OH
HO O HO O
M+n
OH O
OH O OH O
M+[n-1]
I II
M+n
M
O
O
HO O
O
OH O
M+[n-1]
III
Figure 4 Structural images of the coordination compounds of quercetine (I): complexes of the com-
position 1 : 1 (II) and 1 : 2 (III).
conformation due to steric reasons is the strongly nonplanar conformation with all ensuing
consequences. According to theoretical studies [57], the close-to-coplanar conformation of
cycles B and C seems to be achieved only in excited state.
Cycle B plays the role of an electron-density donor with respect to γ-pyronic cycle C.
This is indicated by the results of theoretical studies, which show a high density of electrons
at C-2 carbon atom of flavonols [45]. This is also testified by the data of PMR studies show-
ing an increase of aromaticity of cycle C [57]. Intramolecular charge transfer is enhanced
in excited state, if electron donors are substituents in cycle B [68].
One of the consequences of the structural features of hydroxyflavones is their capabil-
ity of coordination interactions with metal ions [4, 70–72]. Complexation can occur with
participation of carbonyl oxygen and 3-OH or 5-OH group, as well as involving ortho-OH
groups of cycle B. This property is the most pronounced in natural flavonols, which can
have up to three coordination sites (Fig. 4), the filling of which depends on the nature of
metal and specific conditions of the medium.
Complexation of flavonols can occur not only in the main state but also in excited state
[68], both from normal and tautomeric forms, thus enhancing intramolecular charge trans-
fer from cycle B to cycle C, which, in turn, would lead to an increase of the stability of the
complex. The ability of flavonols to bind heavy metal ions is believed to be one of the key
mechanisms of their antioxidant action [16, 18–20, 34].
276 E.A. Venedictov
Flavonols are weakly-coloured compounds, which absorb light in the UV range of the solar
spectrum. Owing to the presence of two cross-conjugated systems, the electronic spectra of
these compounds consist of two groups of close-intensity bands lying in the range of
240–270 and 320–380 nm and related to the electronic transitions of ππ*type in the “ben-
zoyl” (cycle A) and “cinnamoyl” (cycles B and C) fragments of molecules, respectively
[2, 4, 7, 8, 17]. The first group of bands coincides with the mixed range of action of the le-
thal effect of UV radiation on living organisms, which is due to the absorption of proteins
and nucleic acids [23]. The second lies in the region of near UV light considered to have an
indirect damaging effect [23, 73] associated with the photochemical properties of natural
and artificial photosensitizers, which are either products of cell metabolism or get into the
organism with food or drugs.
Electronic spectra bear an imprint of the structural features of flavonols. Here we shall
give only the general ideas of the structure–property and medium–property relationships.
More profound views can be found in [2, 4, 7, 8, 17, 38, 39, 52, 66, 68].
Available data indicate that an increase of the number of OH groups in cycle B induces
a batochromic shift of the long-wavelength band and an increase of its absorption coeffi-
cient. This fact can be explained within the framework of an extended system of conjugated
bonds, including those due to superconjugation with participation of OH groups. As for the
short-wavelength band, the rise of the number of OH or OCH3 groups in this cycle usually
does not lead to a significant change of its position.
The successive increase of the number of these groups in cycle A leads to a similar
effect. However, it is less pronounced.
The spectra of many flavones change at high pH due to the ionization of “acidic” OH
groups. Herewith, a batochromic shift of the maxima of the bands and the redistribution of
their intensities are observed.
The strongest changes in the absorption spectra of flavonols are related to the 3-OH
group. Its elimination is accompanied by a hypsochromic shift of the absorption bands. A
similar change is observed at the substitution of hydrogen atom of this group by methyl rad-
ical. The effect of methylation of OH groups on the absorption spectrum of quercetine is
demonstrated by the data of Table 2.
The effect of a solvent on the spectra of flavonols is expressed to a lower degree [52].
No unequivocal relation between the parameters of the spectra and the nature of a solvent
Table 2 Parameters of the first band in the absorption spectra of quercetine and its derivatives in
different solvents [86].
has been revealed. Nevertheless, it is believed that the effect of the medium reflects effects
of general and specific solvation.
The strongest changes in the absorption spectra of hydroxyflavones are caused by
complexation with metal ions, which leads to a batochromic shift of both bands. Herewith,
the complexation involving the 3-OH group is characterized by a maximal effect (Fig. 6).
Additional complexation via the ortho-site of cycle B (Fig. 4) enhances this effect.
Thus, it is seen that the diversity of flavone structures provides for the maximum effect
of absorption and reflection of solar UV radiation at the most harmful directions of its
action.
In contrast with flavone [42, 67, 69, 74], flavonols possess fluorescence. This is ex-
plained by the fact that in substitution of hydrogen atoms of flavone at C-3 carbon atom by
an OH group the excited triplet nπ* level of the molecular proves much higher than the ex-
cited singlet level of ππ* type and renders no effect on its radiative ability [67, 74].
Fluorescence of flavonols is of dual character. It is considered as a radiation from two
excited singlet states with energies of 69–71 and 53–55 kcal/mol, respectively [39, 40, 50,
56]. Short-wavelength luminescence is determined by the properties of normal molecules,
and long-wavelength luminescence is due to the properties of their tautomers formed as the
result of reversible intramolecular phototransfer of proton from the 3-OH group to carbonyl
oxygen.
Intramolecular proton phototransfer is an essential feature of flavonols. It occurs dur-
ing a time shorter than 8×10–12 s and is controlled by structural factors and, in particular,
specific properties of the medium [40, 51, 52, 56, 59, 64, 66]. Proton phototransfer can be
due both to the properties of singlet and triplet stages depending on the structure of a fla-
vonol. As a rule, it takes place in compounds containing a 5-OH group [57, 63, 69].
Intramolecular proton phototransfer determines the short lifetime of an excited singlet
state of flavonols, thus limiting its chemical potentialities. However, it does not totally sup-
press the interconversion of excited singlet molecules into triplet ones. The latter are char-
acterized by a lifetime of 7.2–7.5 µs and energy of 61–53. 4 kcal/mol [60, 65].
Fluorescence of tautomers of flavonols is characterized by a relatively large lifetime,
which has some dependence on the nature of substituents, their number and substitution site
in cycle B and is from 1.3 up to 4.4 ns [39, 48, 50, 56, 65]. The lifetime of the triplet state
of tautomers depending on the structural features of flavonols is from 9 up to 28 µs [49, 56,
60, 65]. The energy level of this state is within the limits of 31–33 kcal/mol [65].
In contrast to fluorescence [49], oxygen efficiently quenches triplet states of flavonols
[57, 60, 62]. Its removal from solution is an obligatory condition for studies of their prop-
erties. Studies have shown that the rate constants of quenching of triplet molecules by ox-
ygen are close to diffusion rates. In n-heptane, for instance, for the main and tautomeric
forms of unsubstituted flavonols they are 2.1×109 M–1 s–1 and 2.7×109 –3.2×10 9 M–1 s–1
[60, 62], respectively.
Unlike flavonols containing no 5-OH group, quercetine does not luminesce in practice
and gives no long-lived triplet states in direct excitation [57]. One of the causes is a high
probability of internal radiativeless conversion of absorbed energy.
However, formation of a long-lived triplet state of quercetine can be caused by the
triplet–triplet transfer of energy [57]. In this case, sensitizers having a triplet level no less
than 57 kcal/mol transfer their energy to the triplet level of the main form of quercetine with
a rate constant exceeding 1.2×10 9 M–1 s–1. Then there occurs rapid tautomerization of
triplet molecules as the result of intramolecular proton phototransfer from the 5-OH group
to carbonyl oxygen. This is one more significant fact, which indicates that quercetine is
278 E.A. Venedictov
capable of participating in quenching of excited triplet states of proteins and nucleic acids,
whose energy exceeds 57 kcal/mol [75].
The work [57] obtained the characteristics of the triplet state of the tautomer of quer-
cetine. It has an energy within the range of 17.1–24 kcal/mol and lifetime from 12.9 up to
18 µs depending on the solvent. The triplet tautomer of quercetine is quenched by oxygen.
In isopropanol, the quenching rate constant is 1.7×109 M–1 s–1.
Finally, it is necessary to note the luminescent properties of coordination compounds
of flavonols. Compounds with Mg2+, Zn2+, Cd2+, Al3+ and a number of other elements pos-
sess a high yield of fluorescence [68, 70, 71]. These results are interpreted from positions
of limitations imposed by complexation on intramolecular proton phototransfer [68].
Electron-donor substituents in para-position of cycle B increase the yield and lifetime
of fluorescence [68]. The latter is associated with the increase of the probability of transition
from an excited singlet state into the basic one with illumination and a reduction of the prob-
ability of internal radiativeless conversion. Introduction of a heavy atom leads to a reverse
phenomenon of the decrease of the quantum yield and a reduction of the fluorescence life-
time. As the rates of these transitions are relatively small, one should take into account the
possibility of formation of a long-lived triplet state of coordination compounds of flavonols
in absorption of light by them. As will be shown below, these conditions are realized also
in the case of quercetine complexes.
Nevertheless, unlike other flavonols, quercetine attracts attention by its capability of
highly efficient utilization of high-energy solar radiation.
3
S + O2 (3Σg– ) ↔ 1(S … O2) → S + 1O2 (1Σg+ and/or 1∆g),
↔ 3(S … O2) → S + O2,
↔ 5(S … O2).
Here 3S is the triplet molecule of a sensitizer; 1, 3, 5{S … O2 } are the singlet, triplet
and quintet states of the exciplex. In this scheme, the processes of conversion of singlet and
triplet exciplexes are considered to be the most probable, as they are allowed by the spin
selection rules. The theoretical values of statistical balances of these processes are 1/9 and
1/3, respectively. Their real values can be assessed from the ratio kQ /k diff, where kQ and
k diff are the rate constant of quenching of triplet molecules by oxygen and the diffusion rate,
respectively. For triplet states of the normal form of unsubstituted flavonol and its tautomer,
these values in n-heptane are 1/7 and 1/5, respectively. Comparing them with theoretical
values, it can be concluded that the probability of 1O2 formation at the expense of the triplet
state of the normal form is expected to be higher.
For a triplet state of tautomeric quercetine, the ratio kQ /kdiff is equal to only ~1/2 [57].
Evidently, in this case all triplet molecules are quenched by oxygen according to the mech-
anism of internal interconversion. Therefore, this process can be considered within the
framework of the photoprotector action of quercetine as an important link for transforming
the energy of the tautomer’s excited state to thermal energy.
Thus, from the data presented it can be concluded that natural flavonols containing a
5-OH group provide for highly efficient radiativeless dissipation of absorbed solar energy
without forming 1O2.
1
4 O2 Reactions
The first steps in this direction were made in [80], which showed the selective character of
the reaction of flavonols with 1O2, aimed to open the C2C3 double bond of cycle C.
The physical character of quenching 1O2 by quercetine was first demonstrated in [57].
The same work postulated the interaction mechanism determined by the formation of exci-
plex intermediate with reversible partial charge transfer, as it is proposed for other phenols
[25, 81, 82, 93–96]. Herewith, the possibility of not only the above-considered process was
shown, but also of oxidation of quercetine and electron-vibration transfer of energy with
excitation of the vibratory levels of its molecules. Therefore, the total rate of quenching in
this case can be presented by a sum of the rates of three elementary processes:
k t = k q + k r + k E-V,
where k t is the overall quenching rate constant; k q is the physical rate constant determined
by the formation of the exciplex intermediate; k r is the oxidation rate constant of flavonoids;
k E-V, the physical rate constant determined by electron-vibration energy exchange. The re-
sults of measurements and assessments of these constants are given in Tables 3 and 4.
280 E.A. Venedictov
Though for quercetine the value of k E-V is relatively small, it is still comparable with k r in
acetone. Therefore, it is natural that the expediency and reality of such a scheme for other
representatives of this family requires additional consideration and experimental proof.
Electron-vibration energy exchange should be expected exactly for molecules contain-
ing C–H and O–H bonds, as they have the highest vibratory states [57, 90–92]. The rate
constant of this process could not be measured, but yields to numerical assessment on the
basis of the additivity principle [90–92] by the formula
(a)The values of k were obtained relative to the reactivity of tetracene and were refined using its
r
values of kr [89] equal to 5.2×107 M–1 ·s–1 (pyridine), 7.7×107 M–1 ·s–1(DMFA), 2.2×107 M–1 ·s–1
(acetone) and 1.4×107 M–1 ·s–1 (benzene).
The Photochemical Aspect of Reactions of Flavonols with Molecular Oxygen 281
Here k C–H, arom = 480 M –1 s–1, kC–H, aliph = 390 and kO–H = 2145 M –1 s–1 are rate constants
of 1O2 quenching by particular C–H (aromatic), C–H (aliphatic) and O–H bonds, respec-
tively [92]; m, n and p are the number of each of these bonds in the molecule. The values
of k E-V for some flavonoids are presented in Table 4. The same table gives the values of
k q. For compounds with a large number of OH groups, k E-V is only a small part of k q, not
exceeding 5%. At the same time, the values of k E-V and k q are commensurable for a whole
range of simplest hydroxyflavones. For this reason, the role of this process in 1O2 quench-
ing for them should not be underestimated.
The values of k q for some flavonoids are given in Table 4 together with their oxidation
potentials. The donor-acceptor nature of 1O2 quenching is confirmed by the dependence be-
tween k q and the oxidation potential, in which the rate constant increases when the oxida-
tion potential decreases. Therefore, this process can be believed to be controlled by charge
transfer and can be considered within the framework of exciplex intermediate formation.
Table 4 Parameters of the elementary reactions of 1O2 physical quenching by some flavonoids and
their oxidation potentials.
If we turn to the results of [84], it can be found that in the sequence of flavonols be-
tween the values of k q and k r a correlation takes place, which in methanol is described by
the equation
Based on this, the following conclusion can be made [97]: exciplex formation is the
common stage of physical and chemical quenching of 1O2.
A study performed within the structure–property framework [84] made it possible to
single out a number of structural features affecting the quenching of 1O2 by flavonols. It
has been found that compounds with the catechol structure of cycle B are stronger quench-
ers (Table 3). The effect of the 3-OH group on the process occurring in methanol-d6 proves
to be relatively weak. This conclusion is supported by the data of an earlier work, which
studied the quenching of 1O2 by unsubstituted flavonol and 3-methoxyflavone in relatively
inert solvents, such as benzene and CCl4 [83].
The works [77, 86, 88] have shown that complete substitution of hydrogen atoms of
OH groups of quercetine by methyl radicals leads to a rise of efficiency of 1O2 quenching.
The presence of a free 3- or 5-OH group leads to an insignificant decrease of the quenching
rate (Table 3). An amplification of 1O2 quenching in these cases as compared with quercet-
ine should be, apparently, ascribed to a decrease of energy expenses for exciplex formation
owing to an increase of the density of π-electrons in molecules of these compounds due to
282 E.A. Venedictov
20
ln kr
10 1
2
0
0 20 40
DN
Figure 5 Dependence of the logarithm of the rate constant of oxidation of quercetine (1) and
3-hydroxy-5,7,3′,4′-tetramethoxyflavone (2) by singlet oxygen on the donor strength of the solvent.
the electron-donor effect of the methoxy groups. Herewith, the greatest contribution is, ap-
parently, made by methylation of ortho-OH groups of cycle B.
According to the data of [83, 86, 87], the efficiency of 1O2 quenching by flavonols ris-
es in highly specific media. Due to the absence of data in the case of elementary reactions
with unsubstituted flavonol, we carried out a simplified consideration of this phenomenon.
In the order of increasing k t, the solvents are in the following sequences: CCl4 ≤ benzene
< acetonitrile < methanol ≤ heavy water (pD = 3) < pyridine < heavy water (1 M NaOH)
for unsubstituted flavonol and acetone < methanol-d4 < N,N-dimethylformamide < pyri-
dine for quercetine, which give no grounds to relate the quenching rate to the polarity of
solvent. At the same time, it is evident that the efficiency of the process increases with the
rise of the ability of the solvent to form a strong intermolecular H bond with the OH group.
This is illustrated by the dependence of k t on its donor ability of its molecules (DN). For
unsubstituted flavonol, this type of dependence has the form:
exciplex intermediate, the result of which is additional delocalization of the fractional pos-
itive charge emerging during its formation.
In contrast with k q, the value of k r in different media changes within broader limits.
The value of k r is related to the solvent in the same manner (Fig. 5). Unlike quercetine, ox-
idation of compounds without the 3-OH group experiences practically no effect of the me-
dium (Table 3). Therefore, the observed difference in the effect of the solvent is quite
naturally related to the solvation of the 3-OH group.
Undoubted is a symbateness in the change of k q and k r in these media, which is similar
to that considered above, though is less pronounced. This behaviour of k q and k r, evidently
corresponds to the case when the oxidation rate is controlled by the formation of exciplex
intermediate [98].
Another example is oxidation of 3-hydroxy-5,7,3′,4′-tetramethoxyflavone (Fig. 5).
The observed tendency for flavonols is apparently of general character. Still, it is not diffi-
cult to notice some difference. It is seen that the oxidation of 3-hydroxy-5,7,3′,4′-
tetramethoxyflavone possesses the least pronounced sensitivity to the solvent. To explain
this fact, it is necessary to admit that in this case the emerging cationic site in the formation
of exciplex intermediate is capable of more efficient intramolecular delocalization.
In [80], oxidation of flavonols by 1O2 molecules was demonstrated, unlike flavones,
to affect the C2C3 double bond. Herewith, the main product of oxidation is depside (Fig. 6,
VI). In the literature, two most possible mechanisms of this reaction are discussed (Fig. 6):
one assumes oxidation to proceed as an ene reaction (Fig. 6, III) with concerted addition of
1
O2 [80]; the other postulates the formation of a dioxetane intermediate (Fig. 6, IV) [84].
Based on the above analysis, it can be concluded that the primary intermediate of oxidation
O 1 O O
O2 O
O
O
OH OH O
O
O O H
O
I II III
H
OO
O O
O
I
O O
OH
O O
IV V
O
CO
O
COOH
VI
k, s-1
100000
50000
0
0 0.00002 0.00004
c, mol/l
Figure 7 Effect of the concentration of Mg-quercetine on the rate constant of 1O2 luminescence
decay in pyridine (pulse fluorimetry method [88]; 1O2 photosensitizer, anthracene, λexc = 337 nm).
of flavonols is exciplex (Fig. 6, II). What is more, the difference in the kinetic behaviour of
flavonols and flavone suggests that its formation is related to the involvement of π-electrons
of the C2C3 double bond of cycle C.
Assessing on the whole the ability of flavonols to quench 1O2, it can be concluded that
the nature of the molecular environment is one of the most important factors making it pos-
sible to achieve the high efficiency of the process.
ln (A0/At)
1.2
0.6 1
2
0
0 30 60
Time, s
Figure 8 Kinetics of photosensitized oxidation of Mg-quercetine (1) and tetracene (2) in pyridine
during their joint presence in solution (1O2 photosensitizer: Pd-tetra(t-butyl)phthanocyanin, λex > 640
nm, A0 and At are the initial and current optical densities of the solution at the measurement wave-
lengths, respectively).
quenching found for Mg-quercetine are by approximately two orders of magnitude higher
than for quercetine (Table 3). We have here a rather interesting example of activation of
quercetine with respect to the reaction of 1O2, similar, apparently, to the activation of un-
substituted flavonol due to the ionization of the 3-OH group [83].
To date, there is sufficient proof of accelerated destruction of coordination compounds
of quercetine in the light [44, 86]. Consider the photochemical properties of these com-
pounds in greater detail by the example of Mg-quercetine. Experiments yield the following
results [86]. The range of action of photodestruction coincides with the absorption spectrum
of this compound. For this reason, it is natural to consider the photodestruction as a self-sen-
sitized photoreaction. Its yield depends on oxygen. The involvement of 1O2 in it is indicated
by the fact that para-hydroquinone inhibits the photoreaction with the efficiency, with
which it quenches the luminescence of 1O2 (Fig. 9). Hence, an important conclusion can be
made that Mg-quercetine is a photosensitizer of 1O2. As the lifetime of its fluorescence in
pyridine does not exceed 2×10–9 s, and the fluorescence is not quenched by oxygen, it can
be concluded that 1O2 is formed as the result of quenching of excited triplet molecules of
Mg-quercetine by oxygen.
We can not but mention another significant fact, namely, that the photooxidation
mechanism of coordination compounds of quercetine depends on the nature of a metal. For
instance, inhibition of Cd-quercetine photooxidation by para-hydroquinone is much less ef-
ficient [86]. In this case, the contribution of the reaction with participation of 1O2 is reduced
to 60–70%. It is evident that in the case of coordination compounds with heavy elements
other oxidation routes are possible in principle; oxidation of excited triplet molecules by
oxygen can be assigned to these routes [25, 99].
The determination of 1O2 quantum yield is more complicated. A low spectral resolu-
tion of the setup used in [86] did not make possible the direct proof of luminescence photo-
sensitization from the 1∆g condition of O2 in solutions of these compounds. To assess the
286 E.A. Venedictov
k, rel. units
1
1
2
3
0.5
0
0 0.0006 0.0012
c, mol/l
quantum yield of 1O2, we used the kinetic data of the direct photooxidation of Mg-quercet-
ine and tetracene, obtained under conditions of monochromic excitation by the light with
λ = 447 nm (Fig. 10). Consider this issue in more detail.
As shown in Fig. 9, the photooxidation reactions of Mg-quercetine and tetracene are
inhibited by para-hydroquinone with the same efficiency. These data indicate that oxida-
tion is related exclusively to 1O2. Therefore, photoconversion of these compounds can be
presented as the following scheme [97–101]:
hν 1 3
S → S→ S
3 –
S + O2 (3 Σ g ) → S + 1O2
1 –
O2 → O2(3 Σ g ) + hνlum, (krad)
1 –
O2 + Solv → O2 (3 Σ g ) + Solv + Q (ks)
1 –
O2 + S → O2 (3 Σ g ) + S (kq)
where S, 1S and 3S is the photosensitizer in the basic and excited singlet and triplet states,
respectively; Q is the energy of 1O2 dissipated as heat; Solv is solvent; krad, ks, kq, kr are
rate constants of respective processes. Hence, we can write the following expression for the
oxidation rates for each of these compounds:
The Photochemical Aspect of Reactions of Flavonols with Molecular Oxygen 287
0.49
0.48 sequence 1
sequence 2
0.47
0.46
A
0.45
0.44
0.43
0 100 200 300 400
Time, min
Figure 10 Optical density changes in solutions of Mg2+-quercetine (1) and tetracene (2) at a wave-
length of 447 nm during the excitation with a monochromatic light (λ = 447 nm) in air-saturated pyri-
dine (initial concentrations of Mg2+-quercetine and tetracene are, respectively, 6.5·10 –5 and
1.1·10 –4 mol/l) [86].
k r γ I [ S ]τ
–d[S] ⁄ ( dt ) = ------------------------
-.
1 + k t [ S ]τ
Here γ is the quantum yield of 1O2; I, the intensity of the absorbed light flux; [S], concen-
tration of sensitizer; τ = 1/(krad + ks [Solv]), the lifetime of 1O2 in the solvent; kt = (kq + kr).
For tetracene, the term (1 + kt [S]t) in most common solvents can be accepted to be unity
[86, 89, 98].
If solutions of these compounds are illuminated under identical conditions, then for the
extent of their conversion not exceeding 15%, the relative value of γrel can be assessed from
the ratio
α = β γrel / (1 + kt [S]τ),
where α, β and γrel are the ratios of the observed rate constant, kr and γ for Mg-quercetine
to analogous values for tetracene; [S] is the initial concentration of Mg-quercetine.
It is seen from the data presented in Fig. 10 that in pyridine α = 1.8. The value of β,
according to the above-presented data, in the same solvent is equal to 12. Then, using τ =
18×10–6 s, for the concentration of Mg-quercetine equal to 6.5×10–5 mol/l, we obtain γrel
= 0.32. Taking for tetracene γ ~ 0.71 [103], we find that γ ~ 0.22. As is seen, this value is
sufficiently high and comparable with the value of γ for unsubstituted flavone, but is lower
that the value of γ for 3,5,7,3′,4′-pentamethoxyflavone. Therefore, complexation, by elim-
inating the effect of the hydrogen atom of the 3-OH group and restricting the effect of the
5-OH group on the photophysical properties of quercetine, contributes to the photoproduc-
tion of the triplet state of quercetine capable of exciting O2.
288 E.A. Venedictov
Summing up these results, it can be inferred that an unequivocal conclusion on the pho-
tochemical inertness of quercetine with respect to oxygen can not be correct without ac-
counting for the effect of complexation with a whole range of metal ions on this property.
6 Conclusion
At present, there is no doubt that flavonols, depending on conditions, can play the role of
both photoprotectors and photosensitizers. This confirms the reality of information men-
tioned at the beginning of the chapter on the photodynamic properties of some flavonoids.
Realization of photochemical reactions is determined, apparently, by whether conditions
and structural changes favour these processes.
References
1. Biochemistry of Phenolic Compounds, ed. by J. Harborne, Mir: Moscow, 451 pp. (1968)
(Russian translation).
2. T.Y. Mabry, K.R. Mackham and M.B. Thomas, The Systematic Identification of Flavonoid
Compounds, N.Y.–Heidelberg–Berlin, 353 pp. (1970).
3. V.L. Kretovich, Fundamentals of Plant Biochemistry, Vyssh. Shkola: Moscow, 464 pp. (1971)
(in Russian).
4. A. Blazhei and L. Shutyi, Phenolic Compounds of Plant Origin, Mir: Moscow, 239 pp. (1977)
(Russian translation).
5. V.A. Baraboi, Plant Phenols and Human Health, Nauka: Moscow, 160 pp. (1984) (in Russian).
6. T.S. Lebedeva and K.M. Sytnik, Pigments of the Plant World, Naukova Dumka, 86 pp. (1986)
(in Russian).
7. G. Britton, The Biochemistry of Natural Pigments, Mir: Moscow, 422 pp. (1986) (Russian
translation).
8. V.P. Georgievsky, A.I. Rybachenko and A.L. Kazakov, The Physico-chemical and Analytical
Characteristics of Flavonoid Compounds, Rostov State University: Rostov-on-Don, 143 pp.
(1988) (in Russian).
9. P. Atkins, Molecules, Mir: Moscow, 216 pp. (1991) (in Russian).
10. M.N. Zaprometov, Phenolic Compounds, Nauka: Moscow, 272 pp. (1993) (in Russian).
11. A.C. Waiss (Jr) and J. Corse, J. Am. Chem. Soc., 87, 2068 (1967).
12. A.C. Waiss (Jr), R.E. Lundin, A. Lee and J. Corse, J. Am. Chem. Soc., 89, 6213 (1967).
13. M. Tevini, Photochem. Photobiol., 2, 401 (1988).
14. M.A.K. Jansen, V. Gaba and B.M. Greenberg, Trends in Plant Sci., 3, 131 (1998).
15. P. Carletti, A. Masi, A. Wonisch, D. Grill, M. Tausz and M. Ferretti, Environ. Experim. Botany,
50, 149 (2003).
16. M.S. Simmonds, J. Phytochem., 64, 21 (2003).
17. B. Havsteen, Biochem. Pharm., 32, 1141 (1983).
18. C.A. Rice-Evans, N.J. Miller and G. Paganga, Free Radic. Biol. Med., 20, 933 (1996).
19. C.A. Rice-Evans, N.J. Miller and G. Paganga, Trends in Plant Sci., 2, 152 (1997).
20. G. Cao, E. Somc and R.L. Prior, Free Radic. Biol. Med., 21, 749 (1997).
21. B. Halliwell and O.I. Aruoma, FEBS Lett., 281, 9 (1991).
22. A.C. Bowling and M.F. Beal, Life Sci., 56, 1151 (1995).
23. J.G. Scandalios, Trends Biochem. Sci., 27, 483 (2002).
24. S.V. Konev and I.D. Volotovsky, Photobiology, Belorussian State University: Minsk, 383 pp.
(1979) (in Russian).
25. C. Foote, in: Free Radicals in Biology, Mir: Moscow, vol. 2, p. 96 (1979) (Russian translation).
26. Y. Sorata, U. Takajama and M. Kimura, Biochim. Biophys. Acta, 799, 313 (1984).
27. E.R. Blazek, J.G. Peak and M.J. Peak, Photochem. Photobiol., 49, 607 (1989).
28. K. Fukuzawa, K. Matsuura, A. Tokumura, A. Suzuki and J. Terao, Free Radic. Biol. Med., 22,
923 (1997).
29. M. Thompson, C.R. Williams, G.E.P. Elliot, Anal. Chim. Acta, 85, 375 (1976).
The Photochemical Aspect of Reactions of Flavonols with Molecular Oxygen 289
Photoprocesses in Biological Compounds, Nauka: Moscow, 304 pp. (1973) (in Russian).
76. T. Matsuura, T. Takemoto and R. Nakashima, Tetrahedron, 29, 3337 (1973).
77. E.A. Venedictov and O.G. Tokareva, Mendeleev Commun., 2, 84 (1997).
78. A.P. Darmanyan and C.S. Foote, J. Phys. Chem., 96, 3723 (1992).
79. F. Wilkinson, D.J. McGarvey and A.F. Olea, J. Phys. Chem., 98, 3762 (1994).
80. T. Matsuura, H. Matsushima and R. Nakashima, Tetrahedron, 26, 435 (1970).
81. V.Ya. Shlyapintokh and V.B. Ivanov, Usp. Khim., 45, 202 (1976) (in Russian).
82. E.A. Lissi, M.V. Encinas, E. Lemp and M.A. Rubio, Chem. Rev., 93, 699 (1993).
83. S.L. Studer, W.E. Brewer, M.L. Martinez and P.-T. Chou, J. Am. Chem. Soc., 111, 7643 (1989).
84. C. Tournaire, S. Croux, M.-T. Maurette, I. Beck, M. Hocquaux, A.M. Braun and E. Oliveros,
Photochem. Photobiol., 19, 205 (1993).
85. E.A. Venedictov, O.G. Tokareva and N.O. Konstantinova, Zhurn. Fiz. Khim., 70, 1284 (1996)
(in Russian).
86. O.G. Tokareva, PhD Thesis, ISC RAS: Ivanovo (1997) (in Russian).
87. E.A. Venedictov and O.G. Tokareva, Zhurn. Fiz. Khim., 72, 2185 (1998) (in Russian).
88. E.A. Venedictov and O.G. Tokareva, Kinet. Kataliz, 41, 186 (2000) (in Russian).
89. E.A. Venedictov and E.J. Tulikova, Tetrahedron Lett., 44, 3215 (2003).
90. M.A.J. Rodgers, J. Am. Chem. Soc., 105, 6201 (1983).
91. R. Schmidt and H.-D. Brauer, J. Am. Chem. Soc., 109, 6976 (1987).
92. R. Schmidt, J. Am. Chem. Soc., 111, 6983 (1989).
93. M.J. Thomas and C.S. Foote, Photochem. Photobiol., 27, 683 (1978).
94. A.A. Gorman, L.R. Gould, I. Hamblett and M.C. Standen, J. Am. Chem. Soc.,106, 6956 (1984).
95. A. Pajares, J. Gianotti, E. Haggi, G. Stettler, F. Amat-Guerri, S. Criado, S. Miskoski and N.A.
Garcia, Photochem. Photobiol., 119, 9 (1998).
96. M.I. Gutierrez, A.T. Soltermann, F. Amat-Guerri and N.A. Garcia, Photochem. Photobiol., 136,
67 (2000).
97. A.A. Krasnovsky (Jr), E.A. Venedictov and O.M. Chernenko, Biofizika, 27, 966 (1982)
(in Russian).
98. J.-M. Aubry, B. Mandard-Cazin, M. Rougee and R.V. Bensasson, J. Am. Chem. Soc., 117, 9159
(1995)
99. A.N. Terenin, The Photonics of Dye Molecules, Nauka: Leningrad, 616 pp. (1967) (in
Russian).
100. B. Stevens, S.R. Perez and J.A. Ors, J. Am. Chem. Soc., 96, 6846 (1974).
101. A.S. Cherkasov, in: Molecular Photonics, Nauka: Leningrad, 244 pp. (1970) (in Russian).
102. L. Gammet, Fundamentals of Physical Organic Chemistry, Mir: Moscow, 534 pp. (1972)
(in Russian).
103. A.P. Darmanyan, Khim. Fizika, 6, 1192 (1987) (in Russian).
8 Solvation of Drugs as a Key
for Understanding
Partitioning and Passive
Transport Exemplified by
NSAIDs
Passive transport properties of drug molecules are of utmost importance for their
pharmacological and biopharmaceutical effectiveness. Diffusion in different
media and through lipid bilayers are in many cases the rate-determining steps for
the distribution in the body. In the present review, an attempt is made to
demonstrate the importance of solvation of drug molecules for the diffusion and
partition/distribution in phases of different lipophilicity. Different approaches
known in the literature to describe solvation of compounds with flexible
conformation are discussed as well as the experimental methods to directly
measure the energy of solvation. NSAIDs are chosen as an example of a class of
drugs of different molecular structures that have already been studied thoroughly
in many aspects. Thermodynamic characteristics of solvation of the drug
molecules yielded by independent classical experimental methods (Gibbs energy,
enthalpic and entropic terms of Gibbs energy) are used in order to better
understand the diffusion and distribution properties. Correlations between in vitro
data (the partition coefficient, enthalpy of solvation) with biopharmaceutically
relevant characteristics (plasma half-life) are also discussed.
292 German L. Perlovich and Annette Bauer-Brandl
Introduction
Combinatorial chemistry, high throughput screening and in vitro receptor affinity studies
nowadays confront the developer of new drug formulations with a huge number of potential
new drug substances. Although receptor affinity is in many cases undoubtedly a key issue
for potent drugs, other factors may be equally important for the usefulness in vivo, such as
solubility, partitioning behavior, absorption properties, active and passive transport prop-
erties and biodegradation. In many cases, unfortunately, these important aspects are studied
only late in the drug discovery / drug development process. Therefore, as most new drug
candidates are in the first place barely tested in vitro, one faces this enormous amount of
promising new drug substances of high receptor affinity with horrible physicochemical ma-
terial properties. In many cases the substances are of extremely low solubility in any of the
physiological media resulting in tremendous absorption and distribution problems. This
should be a serious drawback for a drug candidate to become a useful drug preparation, be-
cause it is hard to compensate such bad properties even by using the most advanced drug
delivery systems. It would be much more efficient and economical to pick those candidates
with less difficult properties right in the beginning of the drug development process. There-
fore, adding studies of those physicochemical and biophysical characteristics of the com-
pounds, which may be biopharmaceutically relevant to the selection procedure and the
optimization process of new drug candidates at an early phase of development, is an impor-
tant issue today. It is the purpose of the present review to draw attention to physicochemical
properties of importance in this context, and to conclusions that may be drawn from them.
boundary of phases
phase 1 phase 2
a
partitioning
The partitioning between water and octanol is the phase transition of the solute molecule
across the partition boundary of the two separated phases (water and octanol) due to the
thermodynamic driving force, which is defined by the difference of the chemical potentials
of solute molecules dissolved, respectively, in water and in octanol. However, if one sticks
to the thermodynamic formalism, a lot of questions appear about a model to describe the
elementary act of the partitioning process. First of all, it is obvious that the solute molecules
in the solution interact with their nearest neighbor solvent molecules in a stronger way com-
pared to molecules further out in the bulk. Therefore, a so-called solvation shell is created
around the solute molecule. Strength and nature of the drug–solvent interactions within this
solvation shell are non-homogeneous (non-uniform): there are weak van-der-Waals inter-
actions as well as stronger donor–acceptor interactions, hydrogen bonds (as the extreme
case of a donor–acceptor interaction) and electrostatic interactions. Since most of the drugs
are compounds with the ability to create hydrogen bonds, the molecules are solvated in so-
lutions, particularly in aqueous solutions, by strong hydrogen bonds. During the partition-
ing/distribution process, the molecule solvated in water must rebuild this solvation shell
into the octanol solvation shell (resolvation). One can imagine several variants of the resol-
vation mechanism which differ by the height of the activation energy for the partitioning
(Fig. 1):
(a) Complete destruction of the water solvation shell and creation of the new shell in the
octanol phase. This case would be preferred with weak drug–water hydrogen bonds;
the fraction of the hydrogen bond energy in this case would be much less compared
294 German L. Perlovich and Annette Bauer-Brandl
The last two variants ((b) and (c)) will be realized if the energy fraction of the hydrogen
bonds is comparable with the van-der-Waals energy.
Using the proposed classification, the effective size of molecules taking part in the par-
titioning process varies in size of the aggregate (solvation shell). As noted above, the main
driving force of the partitioning process is the difference of the chemical potentials of the
solute molecules in the considered phases, whereas the height of the activation barrier is
determined both by thermodynamic and kinetic factors [9]. There is mutual solubility of
water and octanol (which is important in the considered water–octanol systems), which will
influence the height of the barrier of activation, and the size of the solvation shell as well.
It may therefore be assumed that the height of the activation barrier of partitioning widely
depends on the solvation characteristics of the solute (drug molecule).
Molecular weight, Da
Figure 2 Relationship between lateral diffusion coefficients and molecular weight of drugs. (Data
from Johnson et al. [5].)
defects of the medium (i.e. the lattice of the solid/liquid crystal). A better model concept
for the characterization of self-diffusion processes in the considered case could be the “void
volume” (which is often used for the description of diffusion processes in liquids) or va-
cancies (bi-vacancy etc., which is used for solid/liquid crystals) [17–19]. The elementary
diffusion step can be made if the following conditions are fulfilled: a) availability of a “void
volume” of the size being comparable to the size of the diffusing particle (molecule, aggre-
gate etc.); b) the activation energy of defects (“void volume”) should be enough in order to
remove it; c) the activation energy of the diffused molecule should be high enough in order
to partially destroy the interaction with the nearest neighbors (lipid matrix or octanol, for
example) and to remove the molecule from the region where the defect of the structure is
situated. Therefore, the activation energy of the diffusion process is a function of the Gibbs
energy of solvation:
Consider the lateral diffusion coefficients of drug molecules into the lipid bilayers of
the stratum corneum (SC), which have been calculated from permeability values [5]. The
lateral diffusion coefficients versus the respective molecular weights are presented in Fig. 2
(data are obtained from the work by Johnson et al. [5]).
These authors paid special attention to the bifunctional size dependence of the trans-
port characteristics (diffusion coefficient) within SC. There is an overall dependence of the
lateral diffusion coefficient on the molecular weight of the compound. However, if one
picks compounds with a molecular weight of ≈ 120 Da (the dashed line in Fig. 2) as an ex-
ample, the ratio of the maximum and minimum diffusion coefficients (points 1 and 2) may
be approximately a factor of 200. This huge variation may be explained as follows. Firstly,
probably a number of different diffusion mechanisms take part in the overall diffusion (i.e.
different defects of the structure are present in the elementary diffusion step). Secondly,
296 German L. Perlovich and Annette Bauer-Brandl
aggregates (solvated drug molecules) of different size may take part in the elementary dif-
fusion act. Thirdly, the studied drugs (at pH = 7.4) are ionized to a different degree and
therefore build hydrated aggregates (solvates) of different sizes. The noted reasons are con-
sidered to be closely interrelated. Nevertheless, the conception of solvated drugs appears
useful for explanation of the transport parameters.
The solvation of one mole of solute molecules in the solvent can be defined as the total
change of the standard thermodynamic functions (∆G0, ∆H0, ∆S0) of the compound when
transferring it from the gas phase (ideal gas, single molecules without interaction) into the
solvent. The thermodynamic cycle of solvation is illustrated by Fig. 3, from which it follows
that
0 0 0
∆Ysolv = ∆Ysol − ∆Ysub , (3)
where ∆Y 0 is the standard change of any of the thermodynamic functions of the solvation
(∆Y 0solv), dissolution (∆Y 0sol) and sublimation (∆Y 0sub) process. Therefore, the following
equations may be defined:
In order to study the solvation process, which is experimentally not accessible, we need to
investigate the other two processes: sublimation and dissolution.
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 297
Gas
Solution
Solid Dissolution
state
Accessible surface
........
........... .......
..........
............
............. ...........
.............
..............
............... ..............
...............
...............
.............. ...............
............... Solvent probe
.............. ..............
.............
...........
......... ...........
.........
Figure 4 Schematic picture illustrating an accessible surface (volume) and an excluded (molecular)
surface (volume) introduced by Lee and Richards [20, 21].
is reasonable. In some cases the noted approach is also applied to big solvent molecules
(benzene, n-hexane), the size of which is comparable to the size of the solute molecule. In
such cases various modifications to the model are being made: an ellipsoid (or cylinder, or
any other geometric body) is used to describe the solvent molecule. It should be kept in
mind that a lot of substances are topologically complicated and their conformation is flex-
ible. The calculation of excluded (molecular) and accessible volumes (surfaces) of such
molecules is very sensitive to the conformational state they are in. Therefore, different ways
to calculate the mean statistical molecular conformational states have been suggested. For
example, let us consider some modifications of the Lee’s and Richards’s approach:
is the free energy of hydration of the solute molecule in conformation i, expressed as the
sum of the free energies of hydration of all its constituent atoms or groups k. Vwa,k(i) is the
water accessible volume of group k in a given conformation i of the compound, and ∆gh,k
is the free energy density of hydration of group k.
In polyfunctional molecules, a correction term is added to account for the polarization
of the hydration shell around polar groups by the presence of neighboring polar groups. The
free energy of hydration of a compound is obtained as a Boltzmann weighted average over
all conformations. In order to estimate ∆gh,k, free hydration energies of various classes of
compounds have been used, which have been averaged for definite atoms (functional
groups).
-3.5
-4.0
-4.5
logPc [cm⋅c-1]
-5.0
-5.5
-6.0
-6.5
-7.0
Figure 5 Relationship between log Pc, the Caco-2 cell monolayer permeability coefficients and
PSAd,water, dynamic polar molecular surface area of β-adrenoreceptor antagonists. (Data from Palm
et al. [8].)
different low-energy conformations are compared. It appeared that the dispersion of these
values was from 6 to 54% of the dynamic average (the last value corresponds to conforma-
tionally flexible drugs). The correlation analysis shows that there are linear dependencies
between the logarithm of the Caco-2 cell monolayer permeability and PSAd values of water
(Fig. 5) for a number of β-adrenoreceptor-blocking agents that have been chosen as a sub-
ject of investigations.
The models presented above demonstrate attempts of estimating the thermodynamic
functions of solvation of solute compounds. There may be experimental data on the solva-
tion of some organic substances; however, for drug substances such data are practically ab-
sent. This is an essential obstacle for the development and improvement of models
describing the transport characteristics of drugs from the point of view of the solvation phe-
nomenon. In recent publications [26–31], in contrast to the equilibrium partitioning ap-
proach in water–octanol systems, we have tried to define an absolute lipophilicity scale for
drug molecules of the group of non-steroid anti-inflammatory drugs, NSAIDs. The work is
based on the analysis of drug solvation characteristics, where the experimental data were
obtained using classical thermoanalytical methods: sublimation, solution calorimetry and
the solubility method. This enables to us compare the solvation characteristics of a wide
range of compounds using quantitative parameters derived from experimental data. The
main results of the noted studies will be presented below.
of vapor loss through a small orifice under conditions of free molecular diffusion by
gravimetry.
However, during the last ten years, a transpiration (inert gas flow) method to investi-
gate the sublimation process of various molecular crystals [32, 33] and NSAIDs [27–31]
has been developed. The method has a number of advantages in comparison to effusion
method. Firstly, the transpiration method works over a wider temperature interval, which
increases the number of experimental data points and promotes a statistically better inter-
pretation of the results. Secondly, the transpiration method works at lower temperatures (in
comparison to the effusion method), which decreases the risk of chemical decomposition
of the substance. It should be noted that both sublimation methods have practically not been
used for drug substances except for the above mentioned and the following works: investi-
gation of (±)-ibuprofen sublimation by Ertel et al. [34] applying the effusion method; and
studies on the polymorphism of caffeine, theophylline and carbamazepine by Griesser et
al. [35] using the transpiration method.
In brief, the transpiration method can be described as follows. A stream of an inert gas
passes above the sample at a constant temperature and at a known slow constant flow rate
in order to achieve saturation of the carrier gas with the vapor of the substance under inves-
tigation. The vapor is condensed at some point downstream, and the mass of sublimate and
its purity determined.
The velocity of the carrier gas stream through the sublimation chamber should be cho-
sen very carefully in order to establish and maintain the thermodynamic equilibrium be-
tween the solid and the gaseous state of the substance. The apparatus needs to be tested
before starting the exact measurements by determining the relation between P and ν and
choosing the gas flow velocity value adequate to the appearance of a plateau on P = f(ν)
curve. The velocity of carrier gas flow for the considered compounds was 1.8 dm3/h.
The equipment is calibrated using the benzoic acid standard substance. The standard
value of sublimation enthalpy obtained should be in good agreement with the value recom-
mended by IUPAC of ∆H 0sub = 89.7±0.5 J·mol –1 [36]. In order to gain valid results, the
saturated vapor pressures should be measured 5 times at each temperature with the standard
deviation being within 3–5%. The experimentally determined vapor pressure data can be
described in (ln P; 1/T) coordinates by eq. (9):
ln P = A + B/T. (9)
whereas the entropy of sublimation at a given temperature T was calculated from the fol-
lowing relation:
298 = ∆H T + ∆H
∆Ηsub T 298
sub cor = ∆H sub + (0.75 + 0.15·Cpc )·(T – 298.15) (12)
In order to investigate solubility, usually the method of isothermal saturation is used. This
is a routine method and has extensively been described in the literature (for example [37]).
The Gibb’s energy of solution is derived from equilibrium conditions, i.e. solubility:
However, in this case the thermodynamic functions of ∆G 0sol and ∆H 0sol have self-
consistent experimental errors and their respective dependence is difficult to interpret. In
order to make discussion of the data more correct, one needs to measure ∆H 0sol values by a
direct method (for example, calorimetry).
COOH
F
Ibuprofen H CH 3
(IBP)
CH 3 C CH 2 C H
CH 3
COOH
COOH COOH
H3CO
R R
R H
C C
O O O O O O
H R
H H H
O H R
O O S
R R O O
C R
S(6) C(2) R D
R2 (8)
2
Scheme 1
(a)
O1
C6 C5
(b) C12
C11 C10 C7 C4 C2 C1
O2
b C8 C9 C3
C13
c
a PowderCell 2.0
Figure 6 A perspective view of the cyclic dimer of (+)-Ibuprofen (a) and (±)-Ibuprofen (b); frag-
mentation of the molecule for calculations.
304
Crystal data Shankland et al.(b) Perlovich et al.(c) Kim et al.(d) Perlovich et al.(e) Kim et al.(f) Briard et al.(g) Kim et al.(h)
Crystal system monoclinic monoclinic monoclinic triclinic monoclinic triclinic monoclinic
Space group P21/c P21 P21/c P1 C2/c P1 P21
a, Å 14.397 (8) 12.456 (4) 11.430 (1) 9.299 (3) 34.666 (6) 13.893 (8) 13.375 (5)
b, Å 7.818 (4) 8.0362 (11) 6.591 (1) 12.721(4) 3.743 (1) 7.741 (3) 5.793 (2)
c, Å 10.506 (6) 13.533 (3) 11.395 (2) 5.786 (5) 20.737 (4) 6.136 (2) 7.914 (3)
α, ° 90.00 90.00 90.00 82.99 (4) 90.00 89.61 (3) 90
β, ° 99.70 (3) 112.86 (2) 95.68 (1) 107.28 (4) 110.57 (2) 94.56 (4) 93.91 (3)
γ, ° 90.00 90.00 90.00 106.94 (3) 90.00 88.78 (4) 90
German L. Perlovich and Annette Bauer-Brandl
Volume, Å3 1165.6 (11) 1248.2 (5) 854.2 (4) 624.7 (6) 2519.4 (4) 657.639 611.7 (3)
Z 4 4 4 2 8 2 2
Dcalc, g⋅cm–3 1.175 1.098 1.400 1.293 1.324 1.28 1.25
Radiation pulsed neutron Mo Kα Cu Kα Mo Kα Mo Kα Mo Kα Mo Kα
T, °C –173 25 (2) 25 (2) 25 (2) 25 (2) 10–33 10–33
R[F2 >2σ (F 2)] 0.077 0.0385 0.046 0.0378 0.045 0.0661 0.042
Graph set assign. R 22 (8) R 22 (8) R 22 (8) R 22 (8) R 22 (8); S (5) R 22 (8) C (4)
(a) In brackets, standard deviations; (b)(CSD-IBPRAC02) Ref. [59]; (c)Ref. [31]; (d) Ref. [60]; (e) Ref. [50]; (f)Ref. [61]; (g)(CSD-KEMRUP) Ref. [62];
(h)
(CSD-COYRUD11) Ref. [63].
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 305
states, in contrast to (±)-IBP, where the dimer is formed by hydrogen bonds across the cen-
ter of inversion (space group P21/c), one molecule being R and the other S. Therefore, the
topology of the hydrogen bonds networks of both ibuprofens can be described by R22 (8)
graph set assignment.
Some special characteristics of the conformational states and geometry of hydrogen
bonds of both (+)- and (±)-IBP are presented in Table 3. As the two molecules of (+)-IBP
are situated in the asymmetric unit, the geometries of the two hydrogen bonds are not equiv-
alent: one of them is shorter than the other. Comparison of the data leads to the following
conclusions: a) the angle of the hydrogen bonding of the racemate is more planar in com-
parison to the enantiomer; b) the D…A distance of one of the hydrogen bonds of the (+)-IBP
is approximately equal (within experimental errors) to the analogous value of the enanti-
omer, whereas the second hydrogen bond is longer; c) the H…A distance of the (+)-IBP is
the average of the analogous values of the enantiomer. The conformational states of the
enantiomer molecules in the asymmetric unit cell are different. As it follows from Table 3,
one of the two molecules in the (+)-IBP asymmetric cell has approximately the same con-
formational state as in the racemic IBP crystal.
Table 3 Some parameters, which characterize the conformational states and hydrogen bond
geometry of (+)- and of (±)-ibuprofen molecules in the crystal lattice.
(+)-IBP (±)-IBP
A B
(+)-IBP (A) O1A-H1AO…O2B 0.94 (5) 1.73 (6) 2.651 (4) 169 (5)
(+)-IBP (B) O1B-H1BO…O2A 1.07 (5) 1.58 (5) 2.634 (4) 168 (4)
(±)-IBP O1-H1O…O2(a) 0.963 (13) 1.664 (10) 2.627 (7) 179.5 (7)
(a) Symmetry
code: 1-x,1-y,1-z
(a) (b)
(c)
Figure 7 The packing architecture of flurbiprofen (a), ketoprofen (b) and acetylsalicylic acid (c)
crystal structures.
(a) (b)
with graph set assignment C (4) along the two-fold screw axis. The packing architecture of
diflunisal is shown in Fig. 8b and may be characterized by motifs R22 (8); S (6).
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 307
with graph set assignment C (4) along the two-fold screw axis. The packing architecture of
diflunisal is shown in Fig. 8b and may be characterized by motifs R22 (8); S (6).
Based on the carried out analysis, it can be concluded that the topology of hydrogen
bond networks in the crystal lattices of the outlined class of compounds differs. For exam-
ple, ASA, (+)- and (±)-IBP, KETO and FBP molecules are packed with only dimer aggre-
gates with the graph set assignment R22 (8). The DIF molecules in the crystal lattice create
additional multicenter hydrogen bonds (combination of inter- and intramolecular hydrogen
bonds). Due to this fact, the topology of hydrogen bond networks has a higher level, includ-
ing several motifs: R22 (8); S (6). NAP molecules form only a C (4) motif. In order to find
out correlations between the crystal lattice parameters and the thermochemical and thermo-
dynamic characteristics, let us consider some experimental methods and approaches which
give an opportunity to analyze the outlined functions correctly.
Table 4 Coefficients and correlation parameters of the regression equation ln (P [Pa]) = A + B/T
for the studied NSAIDs.
The Gibbs energy of the sublimation process at room temperature can be separated into
relative fractions of the enthalpic and entropic terms by the following parameters:
Results of these calculations are also shown in Table 5. It is not difficult to see that in
all studied cases the sublimation process is an enthalpy controlled process because the en-
thalpy exceeds the entropy by a factor of approximately 1.8. However, the relative fraction
308
T (f) [°C] 122.3±0.5 141.0±0.5(e) 50.3±0.4(h) 74.0±0.4(f) 113.5±0.2(k) 212.8±0.2(j) 154.4(m) 93.9±1.3(o)
298 (e) (h) (f) (k) (j) (m)
∆ H fus [kJ·mol–1] 18.1±0.2 30.2±0.2 15.4±0.4 23.1±0.4 26.4±0.5 35.9±0.2 31.5±2.1 21.0±0.8(o)
298
∆ H fus [kJ·mol–1](q) 13.6 21.7 14.2 19.8 20.4 22.0 22.0 17.1
German L. Perlovich and Annette Bauer-Brandl
298
∆ S fus [J·mol–1 ·K–1](b) 45.8 72.9 48 67 68.3 73.9 74 57
298
∆ H vap [kJ·mol–1] 76.9 88.5 93.5 96.2 89.9 98.1 108.1 94.8
of the enthalpic term varies between 61.7% for benzoic acid and 67.1% for ketoprofen. For
a better comparison, benzene [39], biphenyl [39], benzophenone [40] and naphthalene [41]
as the non-substituted analogues of the studied NSAIDs (which in their crystal lattices ex-
clusively interact non-specifically by van-der-Waals forces) were investigated. The value
of solid state benzene ςH (Ben) = 53% differs significantly from the benzene derivatives of
the NSAIDs (ςH (BA) = ςH (ASA) = ςH ((+)-IBP) = ςH ((±)-IBP) ≈ 62%). The analogous val-
ue of biphenyl ςH (BiPh) = 60% differs also from the biphenyl substituted compounds:
ςH (FBP) = ςH (DIF) ≈ 66%. The same situation is observed for naphthalene ςH (Naph) =
59% and naproxen ςH (NAP) ≈ 65% and also for benzophenone ςH (BenzPhen) = 62% and
ketoprofen ςH (KETO) ≈ 67%. The causes of the non-systematic share between enthalpy
and entropy with respect to the skeletal structures are supposed to be a different distribution
of the total crystal lattice energy between van-der-Waals interactions and hydrogen bonds
depending on the functional groups in the molecules.
This value quantifies the difference between the entropies of the crystal lattices of the
racemate and the enantiomer and is only caused by particularities of the respective crystal
lattice structures.
For a deeper understanding of the nature of interaction of IBP molecules both in the
racemate and the enantiomer in the crystal lattice, the packing energies are calculated. For
this purpose X-ray data for (+)-IBP obtained by us [31] and the neutron diffraction data for
(±)-IBP from Shankland et al. [59] (refcode CSD – IBPRAC02) were used. Molecular crys-
tals consist of discrete molecules, which interact with each other by intermolecular
oooooooo
310 German L. Perlovich and Annette Bauer-Brandl
∆Ssub((+)-IBP)
∆Ssub(IBP)
∆∆S
1/2 (+)-IBP
1/2 (-)-IBP
chiral crystal racemic IBP
crystal
Scheme 2
non-bonded interactions. Therefore the crystal lattice energy, Elatt, may conditionally be di-
vided into three main terms: van der Waals, E vdw; electrostatic (Coulombic), E coul; and hy-
drogen bonds energy, E HB:
The results of calculations of the energetic terms of the (±)- and the (+)-IBP crystal
lattices for both the Mayo et al. [65] (M) and Gavezzotti [66] (G) force field are presented
in Table 6. As it follows from Table 6, the van der Waals terms of both (+)- and (±)-IBP
are approximately equal if calculated by the same force field. It should be noted that this
value is slightly higher (absolute value) for the G force field compared to the analogous val-
ue for the M force field. In contrast, the terms related to energies of hydrogen bonds show
an opposite trend: for the G force fields the term on approximately 3 kJ·mol –1 less than for
the M force field. The opposite trends sum up to approximately the same total values of the
Table 6 The calculation results of the various energetic terms of (+)- and (±)-ibuprofen crystal
lattices obtained by the two types of the force fields (Mayo et al. [65] and Gavezzotti et al. [66]).
Gavezzotti et al.
Evdw –78.0 (71.0) –78.8 (72.8) 0.8
Ecoul –2.8 (2.6) –2.5 (2.3) –0.3
EHB –29.0 (26.4) –26.9 (24.9) –2.1
Elatt –109.8 –108.2 –1.6
Mayo et al.
Evdw –71.7 (65.2) –73.1 (68.2) 1.4
Ecoul –2.8 (2.5) –2.5 (2.1) –0.3
EHB –35.5 (32.3) –31.9 (29.7) –3.6
Elatt –110.0 –107.5 –2.5
298
H sub 116.0 ±0.6 107.7±0.5 8.3
(a)[kJ⋅mol–1];
in brackets E term/Elatt in % is presented.
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 311
- (+)-IBP
- (±)-IBP
40
30
[%]
vdw
EX-X / E
20
10
0
H-H C-H H-O C-C C-O O-O
Types of interactions
Figure 9 The energetic terms of different types of nonbonded van der Waals interactions of the (+)-
and (±)-ibuprofen crystal lattices (Mayo et al. force field).
crystal lattice energies obtained by the two force fields considered. It should be mentioned
that for (±)-IBP the ratio between hydrogen bonding energy and common crystal lattice en-
ergy is sensitive to the choice of the force field: 26.4% for the G force field and 32.3% for
the M. The same tendency is observed for (+)-IBP: 24.9% for G and 29.7% for M. It should
also be noted that the van der Waals term of (+)-IBP is higher (absolute value) compared
to the (±)-IBP irrespective of the force field used. Therefore, it may be assumed that for
(±)-IBP the loss of van der Waals energy (in comparison with (+)-IBP) is compensated by
hydrogen bonding energy. Probably, these two energies are competing when enantiomer
and/or racemate crystals are growing.
Comparative analysis of the energetic terms of different types of nonbonded van der
Waals interactions for the crystal lattices considered (Mayo et al. force field) was carried
out. The results thereof are presented in Fig. 9. As it follows from Fig. 9, both for (+)- and
(±)-IBP, the dominating contributions are interactions between: C–C > C–H > C–O. More-
over, a transition from (+)- to (±)-IBP makes the relative contributions of the following
terms decrease: C–H, C–C and H–O, while a slight increase of the H–H, C–O and O–O
interaction is noted.
Because the contribution of the C–H interactions for both compounds contribute by
more than 25% to the common energy of the nonbonded van der Waals interactions, it
should be expected that if the positions of the hydrogen atoms in the unit cells could be re-
solved more accurately by neutron diffraction experiments, the accuracy of the final result
should be essentially increased. Lacking these data, we tried to estimate the influence of the
C–H distance on the van der Waals term of the crystal lattice energy. For this purpose, in
the calculation procedure, only the C–H distance is changed (from 0.95 to 1.20 Å), and the
same coordinates of the other atoms in the unit cell retained. The results of the calculation
are presented in Fig. 10. In Fig. 10 the filled symbols mark the E vdw values corresponding
to the C–H bonds obtained from diffraction experiments. From Fig. 10, it follows that with-
in the noted variation interval of the C–H bond lengths, the E vdw values for (+)-IBP are
changed within 4.5 kJ·mol –1, whereas for (±)-IBP – within 17.6 kJ·mol –1. This fact con-
firms once more that the van der Waals energy of the enantiomer is approximately two times
less sensitive to C–H bond variations (6.2%) in comparison to the racemate (24.4%). This
312 German L. Perlovich and Annette Bauer-Brandl
-54 (±)-IBP
-56
-58
-60
Evdw [kJ⋅mol-1] -62
-64
-66
(+)-IBP
-68
-70
-72
-74
0.95 1.00 1.05 1.10 1.15 1.20
RC-H [Å]
Figure 10 The dependence of the van der Waals term of the crystal lattice energy on the length of
the C–H bond (the filled symbols mark the Evdw values corresponding to the C–H bonds obtained
from the diffraction experiments).
fact also stresses that the estimation of crystal lattice energy for (+)-IBP does not too much
depend on the accuracy of hydrogen atoms coordinates (by X-ray or neutron diffraction).
Figure 11 Dependence of fusion entropy on fusion enthalpy of drugs (the graph set assignment is
shown under the drug name). DIF: Diflunisal; NAP: Naproxen; IBP: Ibuprofen; ASA: Acetylsalicylic
acid; FBP: Flurbiporfen; Keto: Ketoprofen; SA: Salicylamide; i-OH-BA: i-hydroxybenzoic acids;
Me-, Et-, Pr-, BuPB: Parabens.
Figure 12 Dependences of melting points on molecular free volumes in crystal lattices (symbols
correspond to Fig 11).
increase of V free values leads to a decrease of the melting temperature, with an essential
exception for naproxen due to structural disordering of the molecules in respective crystal
lattices.
HD(gas)
∆Hsolv
∆Hsub ∆Hsolv
HD(buffer pH 2.0)
∆Hsol
∆Hdep
∆Hsol
HD(solid) (D- + H+)
(buffer pH 7.4
Scheme 3
Table 7 Coefficients and correlation parameters of the regression equation ln (X2) = A + B/T for
the studied NSAIDs in buffers with pH 2.0 and pH 7.4.
[t1°C – t2°C](a) 20.0 − 42.0 20.0 − 42.0 20.0 − 42.0 20.0 − 42.0 20.0 − 42.0 20.0 − 42.0
pH 2.0
–A 2.3±0.2 1.0±0.3 –2.9±0.2 6.0±0.4 2.33±0.02 –1.8±0.2
–B 2874±76 3518±101 5195±65 2582 ±118 3367±6 4123±48
r(b) 0.9990 0.9988 0.9998 0.9969 0.9999 0.9998
σ(c) 1.45·10–2 1.24·10–2 1.24·10–2 2.27·10–2 1.09·10–3 9.26·10–3
N(d) 5 5 5 5 5 5
pH 7.4
–A 6.0±0.2 4.1±0.2 8.8±0.2 8.6±0.2 9.08±0.04 4.4±0.1
–B 1670±55 2345±48 811±54 952±62 628±13 1929±32
r(b) 0.9984 0.9994 0.9934 0.9938 0.9994 0.9996
σ(c) 1.06·10–2 9.22·10–3 1.04·10–2 1.19·10–2 2.41·10–2 6.22·10–3
n(e) 5 5 5 5 5 5
(a) experimental temperature interval; (b) pair correlation coefficient; (c) standard deviation;
(d) number of the experimental points
Table 8 Thermodynamic characteristics of solubility and solvation processes of some NSAIDs in aqueous buffers at pH 2.0 and 7.4 at 25 °C.
(+)-IBP (±)-IBP DIF FBP KETO NAP
pH 2.0
0
∆G solv [kJ · mol–1](a) 29.7 31.7 (34.85)(e) 36.3 36.0 (35.7)(e) 29.8 (28.5)(e) 33.8 (33.72)(e)
0
∆H solv [kJ· mol–1] 23.9±0.6 29.3±0.8 (9.6)(e) 21.5±1.0 43.2±0.5 (12.5)(e) 34.3±0.4 (26.4)(e) 28.0±0.1 (21.3)(e)
0
T∆S solv [kJ·mol–1] –5.8 –2.4 –14.8 7.2 4.5 –5.8
0
∆S solv [J·K–1 ·mol–1] –19.5±2.1 –8.0±2.7 –49.6±3.4 24.1±1.7 15.1±1.3 –19.5±0.4
0
– ∆G solv [kJ · mol–1](b) 11.9 12.5 21.3 17.3 27.2 24.7
0
– ∆H solv [kJ · mol–1] 83.5±1.1 86.5±1.2 97.8±1.6 65.2±1.0 75.8±0.9 100.3±0.6
0
– T∆S solv [kJ · mol–1] 71.6 74.0 76.5 47.9 48.6 75.6
0
– ∆S solv [J·K–1 ·mol–1] 240.1±3.7 248.2±4.0 256.6±5.4 160.7±3.4 163.0±3.0 253.6±2.0
ςHsolv [%](c) 53.8 53.9 56.1 57.6 60.9 57.0
ςTSsolv [%](d) 46.2 46.1 43.9 42.4 39.1 43.0
pH 7.4
0
∆G solv [kJ· mol–1](a) 28.8 29.8 29.2 28.7 27.0 27.7
0
∆H solv [kJ·mol–1] 13.9 ± 0.5 19.5±0.4 7.9±0.5 6.7±0.5 16.0± 0.3 5.2± 0.1
0
T∆S solv [kJ·mol–1] –14.9 –10.3 –21.3 –22.0 –11.0 –22.5
0
∆S solv [J· K–1 ·mol–1] –50.0±1.0 –34.5±1.5 –71.4±1.7 –73.8±1.7 –36.9±1.0 –75.5±0.4
0
– ∆G solv [kJ·mol–1](b) 12.8 14.4 28.4 24.6 30.0 30.8
0
– ∆H solv [kJ·mol–1] 93.5±1.0 96.3±1.0 111.4±1.1 101.7±1.0 94.1±0.8 123.1±0.6
0
– T∆S solv [kJ·mol–1] 80.7 81.9 83.0 77.1 64.1 92.3
0
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport
– ∆S solv [J⋅K–1 ·mol–1] 270.7 ±3.3 274.7 ±3.3 278.4±3.7 258.6±3.4 215.0±2.7 309.6 ±2.0
ςHsolv [%](c) 53.7 54.0 57.3 56.9 59.5 57.1
ςTSsolv [%](d) 46.3 46.0 42.7 43.1 40.5 42.9
0 0 0 0 0 0
315
(a)accuracy,
2 %; (b)Ref. [77]; (c)ςH = (⏐ ∆H solv ⏐/(⏐ ∆H solv ⏐+⏐ T∆S solv ⏐)) ·100%; (d)ςTS = (⏐ T∆S solv ⏐/(⏐ ∆H solv ⏐+⏐ T∆S solv ⏐)) ·100%; (e)Ref. [53].
316 German L. Perlovich and Annette Bauer-Brandl
The respective solution enthalpies, ∆H 0sol , were calculated using the van’t Hoff relation-
ship, which – in contrast to other works [69] – was found to be satisfactorily linear. Further
it was found that the dissolution of ibuprofen in the buffers, both the racemate and the pure
enantiomer, is endothermic (Table 8). Moreover, the values of the entropy of the dissolution
process, ∆S 0sol , (calculated from solubility and enthalpy) are negative. Probably, while a
molecule is transferred from the solid state into the solution, some structure is built in the
solvation shell and in the surrounding solvent, overcompensating the increase in entropy
caused by the dissolution due to a “hydrophobic effect” [70–73] of solvation.
Analysing the solvation process in more detail (Table 8), in both buffers the solvation
is found to be exergonic, and Gibbs energy of solvation, ∆G 0solv , to comprise negative val-
ues for both its enthalpic and entropic terms, ∆H 0solv , and ∆S 0solv . In the current case, the
main driving force of solvation is enthalpy, which is regarded as a “classical” hydrophobic
interaction as the mechanism of solvation [70–73]. The significance of entropy for the sol-
vation process, which is decreasing and working in the opposite direction by probably cre-
ating solvent cages around the solute molecules, is not much smaller than enthalpy at room
temperature.
When further comparing the racemate with the pure enantiomer, all the (absolute) val-
ues of the thermodynamic solvation functions for (+)-IBP are slightly smaller (taking into
account the experimental errors) compared to (±)-IBP, for both the dissociated and non-dis-
sociated form. This means that the solvation of racemic IBP molecules is slightly stronger
compared to the pure enantiomer. This behaviour is probably connected with a difference
in the molecular association states for the racemate and the pure enantiomer in solutions.
The molecules may be exposed to interaction of neighboring molecules /solvation shells
present in the buffers, like dimers (or multimers). These would probably be of different
symmetry in the case of the racemate compared to the pure enantiomer (similarly to the
symmetry of dimers in crystals). Different symmetry determines small variations of the
structure of the solvation shells and their thermodynamic characteristics: the more symmet-
rical (±)-IBP dimer has a stronger ability of solvation. However, the solubility is consider-
ably higher (by approx. a factor of 2) for the pure enantiomer than for the racemate, which
means that the distance to neighboring molecules is smaller. It is difficult to decide whether
this effect accounts for the difference in solvation energy, considering the generally very
low solubility.
Comparison of the enthalpy values in the respective solutions of different pH, as pre-
sented in Scheme 3 and Table 8, enables one to calculate the enthalpy of deprotonation of
the molecules. Deprotonation is exothermic, the absolute value of enthalpy of protonation
/deprotonation is 10 kJ·mol –1, coinciding within experimental error both for (+)- and
(±)-IBP.
It should be noted that the ionic state of molecules in general is more important for the
solvation thermodynamics compared to non-ionic molecules (Table 8). This difference is
higher than the differences between the racemate and the pure enantiomer at the same pH.
The total solvation abilities of (+)- and (±)-IBP for the different states of protonation over-
0 0 0 0
lap: | ∆Ysolv ((+)-IBP)| < | ∆Ysolv ((±)-IBP)| < | ∆Ysolv ((+)-IBP-)| < | ∆Ysolv ((±)- IBP-)|. Partic-
ularly for the entropy, the protonated / deprotonated state of the molecule is of significance.
Therefore, it may be speculated that also the interaction of the solvated IBP-molecules with
membranes and receptors is widely dependent on the properties of the surrounding medium
in terms of pH, possibly also on ion strength and composition.
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 317
- pH 2.0
- pH 7.4
60
50
40
ζΗ [ %]
30
20
10
0
(+)-IBP IBP DIF FBP KETO NAP
Figure 13 Comparative analysis of the relative enthalpic parameter, ςH, of the studied drugs for
buffers with pH 2.0 and 7.4.
5.5
(+)-IBP IBP
5.0
KETO FBP
4.5
NAP
pKa
4.0
3.5
DIF
3.0
0 1 2 3 4 5 6 7 8
Figure 14 Dependence of pKa value on the transfer of Gibbs energy of solvation from the buffer at
pH 7.4 to the buffer at pH 2.0; ∆Gtr(pH 7.4 → pH 2.0) = ∆G0solv
pH2.0 – ∆G0 pH7.4.
sol
ubility with pH that is found for DIF and FBP (Table 7) compared to the other compounds
studied.
It is not difficult to see a regularity between transfer energy ∆Gtr (pH 7.4 → pH 2.0)
values and pKa (Fig. 14): the weaker the acid, the lower the value of the driving force for
the transfer process (and the easier it is to protonate the respective base). It should also be
noted that a compensation effect is observed between the thermodynamic functions of
transfer, which can be described by the following equation:
In other words, the entropic term of the Gibbs energy is 0.8 times less compared to the
enthalpic term.
It should also be kept in mind that determination of pKa values of poorly soluble drugs
is a delicate experiment, and the values may differ considerably according to the method
used. In the case of IBP, for example, using apparent pKa values in different solvent/water
mixtures and extrapolating to 0% solvent content, the pKa value varies between 5.2 and 4.3
[75]. The pKa values reported for other NSAIDs vary within the similar ranges [76]. In the
present study, this was not taken into consideration, neither for correlation analysis nor in
Fig. 14, because it would not affect the key messages anyway.
namic relations between the parameters are illustrated in Scheme 4. Knowledge about
hydration and solvation characteristics of drug molecules exclusively enables the use of an
absolute energetic scale. The discussed thermodynamic parameters together with related lit-
erature data are presented in Table 9.
Buffer pH 2.0
Buffer pH 7.4
T⋅∆Str
∆ Gt
T⋅∆Str ∆ Gt
∆Ht
n- Octanol
∆Ht
Scheme 4
From Table 9 and Scheme 4 it follows that the relationship between the outlined func-
tions can be described as follows:
Buffer pH 2.0 → octanol: ∆Htr < 0; ∆Str > 0; |∆Htr |<|T∆Str |
Buffer pH 7.4 → octanol: ∆Htr > 0; ∆Str > 0; |∆Htr |<<|T∆Str |.
Thus, the two types of the transfer processes (non-dissociated molecule to octanol
phase, and dissociated molecule to octanol) are basically different regarding their respective
driving force. Here, partitioning is a typically enthalpy-driven process, whereas the second
case, i.e. distribution of the charged form of the molecules, in contrast, is entropy driven
[70–73].
It has extensively been discussed in the literature how far water–octanol systems can
be used to predict properties of transport through biological membranes. One of the forcible
arguments of opponents of this approach was the different nature of the driving forces of
the processes: the octanol–water system was classified as enthalpy-driven, whereas the lip-
id phase–water system should have entropic driving forces [78]. Unfortunately, the works
devoted to studies of the partitioning/distribution processes analysed only the change in
Gibbs energy (log P, log D). This approach does not provide the opportunity to understand
the mechanism of the process. However, from the present results it follows that the basic
differences claimed to exist between the thermodynamics of the transfer in octanol–water
and lipid–water systems do not exist. The nature of the driving forces of the processes as
well as the ratio between enthalpic and entropic terms is determined by an eventual charge
of the drug molecule, and the energetic state of this molecule within the respective phases.
In buffer at pH 7.4, the charged drug molecule interacts stronger with the solvation shell
(by additional electrostatic interactions) compared to the uncharged molecule. As a conse-
quence, more energy costs are needed for resolvation of a charged molecule in comparison
with uncharged molecules at partitioning/distribution. Moreover, the costs in enthalpy for
resolvation of a charged molecule are not completely compensated by the solvation en-
thalpy in the octanol phase. This fact may be an essential argument for the assumption that
drug molecules may transfer (during partitioning/distribution processes) with partly
Table 9 Thermodynamic characteristics of the transfer process from n-octanol to buffer (pH 2.0 / pH 7.4) of some NSAIDs at 25°C.
320
n-Octanol
0
– ∆G solv [kJ·mol–1] 40.2(f) 49.3 47.1 50.4 48.0
0
– ∆H solv [kJ·mol–1] 90.9(g) 108.6 86.5 82.5 107.1
0
– T∆S solv [kJ·mol–1] 50.7(h) 59.3 39.4 32.1 59.1
octanol pH2.0
∆Y tr = ∆Y solv – ∆Y solv
∆Gtr [kJ·mol–1] –27.7 (24.35)(d) (–25.7)(e) –28.0 –29.8 (–28.0)(d) (–23.8)(e) –23.2 (–21.96)(d) (–17.8)(e) –23.3 (–23.8)(d) (–20.0)(e)
∆Htr [kJ·mol–1] –4.4 (6.7)(d) (–6.1)(e) –10.8 –21.3 (–4.6)(d) (–15.6)(e) –6.7 (–1.7)(d) (–5.2)(e) –6.8 (0.0)(d) (–13.3)(e)
T ∆Str [kJ·mol–1] 23.3 17.2 8.5 16.5 16.5
ςH [%](a) 15.9 38.6 71.5 28.9 29.2
ςTS [%](b) 84.1 61.4 28.5 71.1 70.8
German L. Perlovich and Annette Bauer-Brandl
octanol pH7.4
∆Y tr = ∆Y solv – ∆Y solv
∆Gtr [kJ·mol–1] –25.8 –20.9 –22.5 –20.4 –17.2
∆Htr [kJ·mol–1 5.4 2.8 15.2 11.6 16.0
T∆Str [kJ·mol–1] 31.2 23.7 37.7 32.0 33.2
ςH [%](a) 14.8 10.6 28.7 26.6 32.5
ςTS [%](b) 85.2 89.4 71.3 73.4 67.5
log P2.0 (c) 3.50 4.44 4.16 3.12 3.34
log D7.4 (c) 1.07 0.76 0.85 –0.25 0.33
pKa(c) 5.2 3.3 4.6 4.6 4.15
(a)ς (b) (c) (d) (e) (f) –1
H = (|∆Htr|/(|∆Htr| + |T∆Str|))·100%; ςTS = (|T∆Str|/(|∆Htr| + |T∆Str|))·100%; Ref. [54]; Ref. [53]; Ref. [55]; 42.6 kJ⋅mol , Ref. [56];
(g)
84.2 kJ⋅mol–1, Ref. [56]; (h)41.4 kJ·mol–1, Ref. [56]
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 321
1.0
FBP
0.8 DIF
0.6
log(D7.4)
0.4 NAP
0.2
0.0
KETO
-0.2
-0.4
90 95 100 105 110 115 120 125
-1
-∆Hsolv [kJ⋅mol ]
Figure 15 Dependence of the buffer/octanol distribution coefficient in the form of log D7.4 on sol-
vation enthalpy in buffer pH 7.4, ∆Hsolv .
retained solvation shells. The volume and structure of the “removing/accompanying” shell
is determined by the ratio of all the thermodynamic parameters.
It is interesting to compare the solvation characteristics with the experimentally deter-
mined partitioning properties of the drugs taken from the literature (Table 9). The depend-
0
ence of log D7.4 plotted versus H solv (pH 7.4) is shown in Fig. 15.
As the absolute value increases, the log D7.4 value decreases (with an exception for
KETO). Probably, this regularity is connected with considerable energetic costs for molec-
ular resolvation at the transfer step from the buffer to the octanol phase during the distribu-
tion. Because for buffer pH 7.4 a compensation effect between the solvation functions is
observed, the character of the correlation dependencies between log D7.4 values and entro-
pic term does not change. It should be noted that the value for KETO does not deviate from
the log D7.4 versus ∆G0solv correlation as the ∆H0solv does (Fig. 16). This fact confirms that
the enthalpic and entropic terms are more sensitive to the nature of the occurring processes
compared with Gibbs energy [57].
As a consequence of this fact, also the widely studied (and in many cases relatively
poor) correlations between Gibbs energy of drug – cyclodextrin complexation and log P are
of limited value as a measure of hydrophobicity, because comparison of Gibbs energies of
two different processes (complexation and partitioning/distribution) does not consider their
driving forces [70]. It is obvious that a good correlation can only be expected to be observed
in cases where the values and signs of the enthalpic and entropic terms of both processes
are identical.
Finally, let us consider the distribution/partitioning process from the point of view of
solvation. In order to transfer a molecule from one phase (buffer) to another (octanol) it is
necessary to overcome a potential barrier, which is “hypothetical” and equals the solvation
enthalpy in the buffer (Scheme 5). The height of this barrier determines the kinetic param-
eters of the partitioning/distribution process. Obviously, the outlined process does not in-
tend to desolvate the molecule completely: the resolvation process presents itself as a
322 German L. Perlovich and Annette Bauer-Brandl
1.0
FBP
DIF
0.8
0.6
log(D7.4)
0.4 NAP
0.2
0.0
KETO
-0.2
-0.4
10 15 20 25 30 35
Figure 16 Dependence of log D7.4 on Gibbs energy of solvation in buffer pH 7.4, ∆Gsolv (pH 7.4).
complicated process including the simultaneous destruction of the old shell and the creation
of a new one. As a consequence of this competition, the height of the activation barrier de-
creases considerably. The value of the activation barrier may be estimated from kinetic pa-
rameters of the partitioning/distribution process. This may in the future be helpful for
further characterisation of biopharmaceutical properties of drug molecules.
Scheme 5
14
DIF
NAP
12
10
PHLm [h] 8
4 FBP
IBP KETO
2
ASA
0
Figure 17 Dependence of mean plasma half-life values, PHLm, vs. solvation enthalpy in octanol,
oct .
∆Hsolv
bution into CSF. This relation prompted us to carry out a correlation analysis between the
mean plasma half-life values, PHLm, [44] for the NSAIDs under investigation and the ob-
tained thermodynamic functions in octanol as a model for a lipophilic compartment. As it
follows from Fig. 17, a common trend line is derived for the solvation enthalpies of the not-
ed drugs in octanol, ∆H oct solv: good solvation in octanol corresponds to a long plasma
half-life. All drugs studied are carboxylic acids – and, therefore, all of them have a high
plasma-protein-binding affinity; correlations between the plasma-binding capacity of drugs
and their lipophilic–hydrophilic properties (expressed as log P) have already been ac-
knowledged [45]. The specific interaction between plasma proteins and the (acidic groups
in the) drug molecules is complex as there are quite a number of binding sites of different
affinity and hydrophobicity [46]. However, the plasma half-life may from a general point
of view be looked at as the equilibrium between the binding and the re-distribution of mol-
ecules from the (hydrophobic) plasma binding sites into the surrounding aqueous phase,
where similar thermodynamic prerequisites apply as for partitioning.
From Fig. 17, it can be seen that both ibuprofen and acetylsalicylic acid have a lower
plasma half-life compared to flurbiprofen and ketoprofen. It may be supposed that this dif-
ference in behavior is related to their being close derivatives of benzene, whereas all the
other substances have two benzene rings. This would in particular indicate that the interac-
tions of major importance in the redistribution of the drug molecules from the plasma pro-
tein binding into the surrounding aqueous phase are van-der-Waals forces. On the other
hand, Fig. 17 can be expressed for flurbiprofen and ketoprofen in terms of a somewhat long-
er plasma half-life.
Anyhow, the present approach may be useful in estimating plasma half-life values
from thermochemical measurements of solvation. A connection between the quality of sol-
vation in octanol and the plasma half-life seems reasonable as there may be some basic sim-
ilarity in the nature of the drug–octanol and drug–plasma protein interactions.
324 German L. Perlovich and Annette Bauer-Brandl
Acknowledgments
This work was supported by Norges Forskningsråd, project number HS 58101, and a
personal grant for GP from the Russian Science Support Foundation.
References
1. S.H. Yalkowsky and S. Banerjee, Aqueous Solubility (Methods of Estimation for Organic
Compounds), Marcel Dekker: New York (1992).
2. A.-L. Ungell, S. Nylander, S. Bergstrand, Å. Sjöberg, H. Lennernäs, J. Pharm. Sci., 87 (3),
360–366 (1998).
3. M. Sznitowska, S. Janicki and A.C. Williams, J. Pharm. Sci., 87 (9), 1109–11014 (1998).
4. P.S. Burton, R.A. Conradi, N.F.H. Ho, A.R. Hilgers and R.T. Borchard, J. Pharm. Sci., 85 (12),
1336–1340 (1998).
5. M.E. Johnson, D. Blankschtein and R. Langer, J. Pharm. Sci., 86 (10), 1162–1172 (1997).
6. H. Lennernäs, L. Knutson, T. Knutson, L. Lesko, T. Salomonson and G. Amidon, Pharm. Res.,
12, 396 (1995).
7. K. Palm, K. Luthman, A.-L. Ungell, G. Strandlund and P. Artursson, J. Pharm. Sci., 85, 32–39
(1996).
8. K. Palm, K. Luthman, A.-L. Ungell, G. Strandlund, F. Beigi, P. Lundahl and P. Artursson,
J. Med. Chem., 41, 5382–5392 (1998).
9. J.H. Lin and A.Y.H. Lu, Pharmacol. Rev., 49 (4), 403–449 (1997).
10. H. Kubinyi, in: Progress in Drug Research, ed. E. Jucker, Birkhäuser Verlag: Basel, vol. 23,
p. 97 (1979).
11. Y.C. Martin, in: Quantitative Drug Design, Marcell Dekker Inc.: New York (1978).
12. F. Barbato, G. Caliendo, M.I. La Rotonda, C. Silipo, G. Toraldo and A. Vittoria, Quant.
Struct.-Act. Relat., 5, 88–95 (1986).
13. R.E. Jacobs and S.H. White, Biochemistry, 28, 3421–3437 (1989).
14. S.A. Margolis and M. Levenson, Fresenius J. Anal. Chem., 367, 1–7 (2000).
15. M. Kodama, H. Kato and H. Aoki, J. Therm. Anal. Cal., 64, 219–230 (2001).
16. M. Kodama, H. Aoki, H. Takahashi and I. Hatta, Biochim. Biophys. Acta, 1329, 61–73 (1997).
17. B.E. Cohen and A.D. Bangham, Nature, 236, 173–174 (1972).
18. M.H. Cohen and Turnbull, D. J. Chem. Phys., 31, 1164–1169 (1969).
19. W.R. Lieb and W.D. Stein, Nature, 224, 240–243 (1969).
20. B. Lee and F.M. Richards, J. Mol. Biol., 379 (1971).
21. F.M. Richards, Ann. Rev. Biophys. Bioeng., 6, 151 (1977).
22. Y.K. Kang, G. Nemethy and H.A. Scheraga, J. Phys. Chem., 91, 4105–9 (1987).
23. Y.K. Kang, G. Nemethy and H.A. Scheraga, J. Phys. Chem., 91, 4109–4117 (1987).
24. Y.K. Kang, G. Nemethy and H.A. Scheraga, J. Phys. Chem., 91, 4118–4120 (1987).
25. Y.K. Kang, K.D. Gibson, G. Nemethy and H.A. Scheraga, J. Phys. Chem., 92, 4739–4742
(1988).
26. G.L. Perlovich and A. Bauer-Brandl, Pharm. Res., 20, 471–478 (2003).
27. G.L. Perlovich, S.V. Kurkov and A. Bauer-Brandl, Eur. J. Pharm. Sci., 19 (5), 423–432 (2003).
28. G.L. Perlovich, S.V. Kurkov, A.N. Kinchin and A. Bauer-Brandl, J. Pharm. Sci., 92 (12),
2511–2520 (2003).
29. G.L. Perlovich, S.V. Kurkov, A.N. Kinchin and A. Bauer-Brandl, Eur. J. Pharm. Biopharm.,
57 (2), 411–420 (2004).
30. G.L. Perlovich, S.V. Kurkov, A.N. Kinchin and A. Bauer-Brandl, AAPS Pharm. Sci. 6 (1), article
3, http/www.aapspharmsci.org (2004).
31. G.L. Perlovich, S.V. Kurkov and A. Bauer-Brandl, J. Pharm. Sci., 93 (3), 654–666 (2004).
32. W. Zielenkiewicz, G. Perlovich and M. Wszelaka-Rylik, J. Therm. Anal. Cal., 57, 225–234
(1999).
33. G.L. Perlovich, O.A. Golubchikov and M.E. Klueva, J. Porph. Phthal., 4, 699–706 (2000).
34. K.D. Ertel, R.A. Heasley, C. Koegel, A. Chakrabarti and J.T. Carstensen, J. Pharm. Sci., 79 (6),
552 (1990).
35. U.J. Griesser, M. Szelagiewicz, U.Ch. Hofmeier, C. Pitt and S. Cianferani, J. Therm. Anal. Cal.,
Solvation of Drugs as a Key for Understanding Partitioning and Passive Transport 325
9 Biodamage of Materials:
Adhesion of
Microorganisms on the
Surface of Materials
Introduction
Problems set by Professor N.M. Emanuel in the field of applied works were characterized
by a fundamental approach. In particular, at the beginning of 1980s the issues of polymer
materials’ biodamage traditionally related to microbiology were included by N.M. Emanuel
into the chemical destruction section of polymer materials science. Subsequent investiga-
tions in this field showed active destructing agents of materials in interaction with micro-
organisms to be products of their vital functions – metabolites representing famous
chemical agents: water, salts, acids, alkali, enzymes, toxins, i.e., biodamage proceeds by
laws of chemical destruction [1–10].
Investigation of the results of biodestruction generally concerns two stages: adhesion
or attachment of microorganisms to the surface of material and growth of microbial biomass
as the result of substrate-support utilization. These two stages should predetermine further
processes of material degradation.
Given the current ecologically fraught environment, fast adaptation of various micro-
bial species to the changing environmental conditions and accumulation of various wastes
of synthetic origin, there is a requirement in research and development of both rapidly de-
grading materials and materials resistant to biodeterioration [11–12].
Studies of microbial cell adhesion are of great interest now, but in many cases these
investigations are “exotic” as far as the selection of microbial species is concerned. The au-
thors of [13] studied selective adhesion of extremophile cultures to micaceous plates coated
with polyethyleneimine. Using the method of scanning electron microscopy, adhesion of
extremophiles to a polyethyleneimine coating was found to be preferred in comparison to
that to a polylysine coating. The main physicochemical factors of material surface such as
hydrophobicity and roughness and the morphology of microbial cells are discussed in
[13–16]. These works and the current state of the problem confirm the necessity of inves-
tigating the nature of adhesion interaction on the material–microbial cell wall interface with
the view of a quantitative estimation of this process.
100
5
Adhesion number, % 4
80 3
2
60 1
40
20
0
0 5 10 15 20 25 30 35 40 45 50
Time, h
Figure 1 Kinetic curves of adhesion of A. niger conidia to various polymer materials at a tempera-
ture of 22°C and relative moisture of 30%. 1, polyethylene; 2, epoxide resin; 3, polymethyl methacry-
late; 4, acetyl cellulose; 5, cellophane.
relation to the time of preliminary ageing under given thermo-moist conditions (T = 22°C,
ϕ = 30%) is presented in Fig. 1. Adhesion interaction of the polymer material–micro-
organism system is of kinetic character. Each polymer material–microorganism system is
characterized by the time of completion of adhesion forces’ formation between conidia and
the surface (approaching the plateau in the kinetic curve), which depends on external
conditions.
The dependences presented in Fig. 1 are well described by eq. (1):
ln γ/γ ∞ = – K ⋅ t , (1)
where γ and γ∞ are, respectively, the number of particles remaining on the surface of poly-
mer material after preliminary ageing for time t, and of those irreversibly adherent under
given conditions; K is the rate formation constant for adhesion forces between conidia and
the surface of material. The parameter values determined from the equation for polymers
studied and also the value of force per cell are given in Table 1. At ω = 15000 min –1 the
force influencing each conidium was equal to F = 1.2·10 –4 dyn/cell. As it is known [18]
the value of force under adhesion corresponds to γ = 50%. If we know F50, the time of the
kinetic curve approaching the plateau and the exponential course of kinetic curves of adhe-
sion, we may determine the values of adhesion forces for each polymer under given thermo-
moist conditions.
Thus, under fixed thermo-moist conditions each material may be characterized by a
proper set of values of K and F. With respect to the strength of A. niger conidium adhesion
Table 1 Parameters of adhesion of A. niger conidia to the surface of polymer materials.
100
5
4
3
Adhesion number, %
80
60 2
1
40
20
0
0 10 20 30 40 50
Time, h
Figure 2 Kinetic curves of adhesion of A. niger conidia to polyethylene and cellophane at a relative
moisture of 30% and various temperatures: 1, polyethylene, T = 38°C; 2, polyethylene, T = 22°C; 3,
polyethylene, T = 10°C; 4, cellophane, T = 22°C; 5, cellophane, T = 10°C.
the investigated polymer materials can be arranged in the following sequence: polyethyl-
ene, epoxide resin, polymethyl methacrylate, cellophane, lavsan, acetyl cellulose.
As we mentioned above, polymer materials with various degrees of hydrophilicity
were used in experiments. It is obvious from Fig. 1 and Table 1 that hydrophilic polymers
have high values of adhesion force and formation rate constants, whereas for hydrophobic
polymer materials these values are significantly lower. Obviously, the values of K and F
are determined by the ability of material to strong adhesion of microorganisms in relation
to the moisture capacity of material.
For a more detailed understanding of the process, we investigated the influence of ex-
ternal operational factors on adhesion. The kinetic curves of adhesion on polyethylene and
cellophane at various temperatures of conidial ageing on the polymer surface are presented
in Fig. 2. It is obvious that with a temperature rise, adhesion of A. niger conidia to polymer
materials and the rate of adhesion bond formation are decreased, which is characteristic of
physical phenomena.
The values of the characteristics of adhesion interaction for polyethylene and cello-
phane at various temperatures and relative moisture are presented in Table 2.
As is obvious from Fig. 2 and Table 2, the adhesion force (F) and the constant of ad-
hesion force formation rate are observed at a 98% relative air moisture. Obviously, the
Table 2 Adhesion parameters of interaction for polyethylene and cellophane at various thermo-
moist conditions.
T, °C φ, % Polyethylene Cellophane
0 70 2.20·10–5 3.0·10–4 – – –
10 30 85 1.66·10–5 5.2·10–4 90 1.0·10–4 2.5·10–3
98 100 9.70·10–5 1.9·10–3 – – –
22 30 55 1.66·10–5 3.3·10–4 85 1.0·10–4 1.6·10–3
38 30 50 1.66·10–5 1.3·10–4 – – –
Biodamage of Materials: Adhesion of Microorganisms on the Surface of Materials 331
120
40
20
0
0 10 20 30 40 50 60
Time, h
Figure 3 Kinetic curves of adhesion of A. niger conidia to polyethylene at 10°C and a relative mois-
ture of 0 (1), 30 (2) or 100% (3).
conidia of macroscopic fungi in an air medium interact with the surface of polymer material
at the expense of molecular forces and at the action of capillary forces of liquid condensed
in the gap between the conidium and polymer surface under the action of forces of Coulomb
interaction and other causes. In the presence of air moisture, the condensation of vapor oc-
curs between the conidium and polymer surface. Capillary forces are the greater, the higher
the surface tension of liquid whose vapor surrounds the area of interaction of conidia with
polymer surface, where the moistening of the polymer surface is better. The liquid interlay-
er between interacting objects (in this case between conidia and polymer surface) excludes
or to a great extent weakens the action of forces of electric nature. That is why the results
of experiments carried out with polyethylene at a fixed temperature 10°C but at various air
moistures (see Fig. 3 and Table 2) confirm the physical character of adhesion interaction of
the system. To confirm the assumption, adhesion interaction of A. niger on the surface of
polyethylene at 22°C and relative air moisture of 30% was studied during days on a raster
electron microscope. An increase of interaction time leads to a change of the geometric
shape of A. niger conidia and an increase of the contact area by almost 100%. Probably, this
fact may explain the kinetic character of conidial adhesion to polymer surfaces that allows
a quantitative description of microbial adhesion to various surfaces and its characterization
by kinetic parameters.
The process of microbial adhesion is also determinative in biofouling of metals in wa-
ter and air media [19–22]. Those experiments made use of conidia of Tr. viride fungi, which
occur in water media and possess optimal sizes for quantitative microscopic analysis. They
determined the force, causing the detachment of 50% of conidia from the general number
of cells attached to the metal surface, from the integral curves for the distribution of conidia
by adhesion forces characterizing the dependence of part of detached particles on the force
of detachment.
The force of detachment of 50% of conidia was calculated to be
where ρav is the density of the medium (water); ρc is the density of conidia equal to 1.15
g/cm3, R = 4 cm is the radius of the centrifuge rotor, ω is the number of revolutions, r is the
332 K.Z. Gumargalieva, I.G. Kalinina, S.A. Semenov and G.E. Zaikov
100 1
2
Adhesion number, %
80 3
4
5
60 6
40
7
20
0
0 5 10 15 20 25
Time, h
Figure 4 Curves of the increase of adhesion of Tr. viride conidia to metals: 1, zinc; 2, copper; 3, alu-
minium; 4, nickel; 5, titanium; 6, tantalum; 7, molybdenum.
conidium radius equal to (3 ±0.3)·10 –4 cm. The adhesion of conidia to a metal surface was
characterized by two parameters γ∞ and F50. The curves of adhesion of Tr. viride conidia
to various metals are presented in Figure 4, the values of γ∞ and F50 are given in Table 3.
By adhesion parameters, all metals may be divided into two groups: the first is char-
acterized by F50 ≥ 10 –4 dyn/cell; the second, by F50 from 10 –5 up to 10 –7 dyn/cell.
Prehistory of surface plays a significant role; e.g., for nickel treated by cold plastic de-
formation γ∞ = 79±8%, and F50 = 2.0·10 –4 dyn/cell, and for nickel aged for 1 h at a tem-
perature of 700°C and pressure 10 –4 mm Hg, γ∞ = 50±2%, and F50 = 8.2·10 –4 dyn/cell.
Adhesion is observed to be minimal on gold, tungsten and molybdenum; the maximum
is on zinc, lead and copper. Aluminium and titanium occupy an intermediate position.
On the whole, the sequence of a limited number of adhesion is the following: lead,
zinc, copper, aluminum, nickel, silver, titanium, tantalum, platinum, molybdenum, gold,
Table 3 Adhesion parameters of Trichoderma viride conidia at T = 22°C and ω = 15000
revolutions in water medium to metals treated in solvent by a − boiling, b − at 22°C.
No. Metals a b
100
Adhesion number, % 80
60
4
40
3
20
2 5
1
0
10 15 20 25 30 35 40
Angle rate, min_1
Figure 5 Distribution of adhesion number (γ) of A. niger conidia in relation to the angle rate (ω) on
polyethylene (1, 2) and cellophane (3, 4) at 22°C and ϕ = 98%. 1 and 3, ageing time 4 h; 2 and 4, age-
ing time 24 h; 5, distribution of polystyrene particles in relation to ω in an integral form under the
same adhesion conditions.
tungsten; and by a decrease of values of force of 50% detachment of conidia: lead, zinc,
copper, silver, aluminium, tantalum, platinum, nickel, titanium, molybdenum, gold, tung-
sten.
Thus, Tr. viride conidia reveal a low level of adhesion to metals, which are not oxi-
dized under experimental conditions and do not form oxides at all, e.g., gold.
The results obtained testify that the main characteristic of adhesion – the force of ad-
hesion – increases in time reaching an equilibrium value and is determined by the nature of
material surface and type of microscopic fungi. The experimental fact of the distribution of
adhesion forces in relation to applied stress was also noticed. This effect was studied earlier
and probably is caused by both the heterogeneity of the surface of material and heteroge-
neity of conidia by size. The dependences of the number of A. niger conidia remaining on
the surface of a material after centrifugation (adhesion number γ ) on the angle rate of rota-
tion for polyethylene and cellophane at a temperature T = 22°C and ambient moisture ϕ =
98% are presented in Fig. 5. From the data obtained, using eq. (2) we calculated forces of
50% detachment of conidia, F50. The experimental results show that at time of ageing
conidia on a polymer surface for 4 and 24 h the adhesion force Fa (force of 50% detachment
of conidia F50) is increased for polyethylene 2 times, and for cellophane 1.5 times. This fact
obviously testifies to a significant difference in the constants of adhesion force formation
rates for the same system (Tables 1 and 2).
For a more detailed investigation of the dependences of the adhesion number on the
number of revolutions γ = γ (ω) (in implicit form it is the dependence of adhesion number
on applied force γ = γ (F) that represents the distribution by adhesion forces) the derivative
of this function was determined for polyethylene γ (F) = dγ /dF(F) (Fig. 6). The dependence
presented in Fig. 6 is described by an equation of the type corresponding to the Gaussian
distribution:
19 2
17
15
Y, units
13
11
9 1
7
5
3
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
F, dyn/cell
40
30
N
20
10
0
1 2 3 4 5 6 7
r, µm
on a TESLA BS 300 raster electron microscope. The measured values of conidial diameters
on electron micrographs demonstrate a significant scattering by size (Fig. 7) that should in-
fluence the scattering of adhesion force values.
Let us show that the distribution of conidia by size contributes (i.e. is described by the
same function) into the distribution of adhesive conidia by force.
We present the derivatives of adhesion number by radius and force:
We should prove that fromγ = γ (r) it follows that γ = γ (F). Let us write the distribution
by conidial size in the following form:
100 2
80
1
60
40
20
0
0 10 20 30 40 50 60 70 80
Time, h
Figure 8 Kinetic curves of adhesion of polystyrene particles to polyethylene (1) and cellophane (2)
at T = 22°C, ϕ = 30%, ω = 15 000 min –1.
It follows from the experimental data (the dependences F = F(r) and γ = γ (r)) that for
all values of radiuses (r) of conidia at the functioning force of detachment from F1 =
5.5·10 –3 dynes per cell up to F2 = 2.3·10 –2 dynes per cell (Fig. 6) the values in the
right-hand side of eq. (5) will be
dγ /dr = 4.7±0.3.
Within the limits of experimental error these values are equal, i.e., the distribution of
conidia by size is one of the causes of the real distribution by adhesion force. The hetero-
geneity of support should also contribute to the distribution of particles by adhesion force.
To elucidate this, we carried out experiments with spherical polystyrene particles (di-
ameter d = 2.8 µm; Serva). The kinetic curves of adhesion and distribution of particles by
adhesion force obtained in the experiments are given in Fig. 8.
By their character, the kinetic curves and curves of distribution of conidia A. niger and
polystyrene particles are similar, although there are some differences in parameters of ad-
hesion interaction (Figs. 4 and 6). The constants of adhesion force formation for polystyrene
particles have high values in comparison with A. niger conidia. Moreover, the scattering of
adhesion force values γ in the case of polystyrene particles is significantly lower.
The results of the experiment suggest that adhesion interaction of system polystyrene
particles–polymer changes in time; the distribution of particles by adhesion forces is de-
scribed by the Gaussian equation. Consequently, the picture of the adhesion of polystyrene
particles and A. niger conidia to the polymer surface is analogous, which may be explained
for polystyrene particles by the following:
• an increase of contact area of polystyrene particles with polymer surface causes the
rise of adhesion force in time at the expense of deformation of the surface;
• capillary concentration of moisture in the contact zone also promotes an adhesion
force;
336 K.Z. Gumargalieva, I.G. Kalinina, S.A. Semenov and G.E. Zaikov
C = D/l·E,
D C, % D C, % D C, %
c
1.0
0.7
a b
1.0
0.5 0.5
0.5
0.5
Figure 9 IR spectra of the surface layer of conidia: a, A. niger, 3 days; b, A. niger, 15 days, 30 days;
c, A. terreus, 3 days; d, A. terreus, 15 days, 30 days; e, Tr. viride, 3 days; f, Tr. viride, 15 days, 30 days;
j, P. funiculosum, 3 days; h, P. funiculosum, 15 days, 30 days.
conidia in the course of growth noticeable quantitative changes in the structure of cell walls
occur, and for other types significant changes were not noticed.
IR spectra of the surface layer of A. niger conidia at various development times are
presented in Fig. 9 as an example. The IR spectra of 15- and 30-day-old conidia coincide
almost completely, which is explained by the known factor of the decay of the metabolic
processes and by the cell quiescent state already after 13–14 days. So, for conidia of
Tr. viride type the concentration of amides in the surface layer of the cell decreased from
45% (age, 3 days) down to 26–28% (age, 15 days). A weak rise of the concentration of
338 K.Z. Gumargalieva, I.G. Kalinina, S.A. Semenov and G.E. Zaikov
80
60
40
20 2
1 3
0
0 5 10 15 20 25 30
Angle rate, min_1
Figure 10 Distribution of the adhesion number (γ) of A. niger conidia in relation to the angle rate
(ω) on polyethylene at T = 22°C and ϕ = 98%. The ageing time of conidia on the surface of polyeth-
ylene, 24 h. 1, 3-day-old conidia, 2, 15-day-old conidia, 3, 30-day-old conidia.
where F5015 and F5030 are the forces of 50% detachment (γ = 50%) of 15- and 30-day-old
conidia correspondingly, that is 3.4 times higher for adhesion force in the case of 3-day-old
conidia.
For 3-day-old conidia, the adhesion force is:
Comparing the data on the adhesion of 15- and 30-day-old conidia with the IR spectra
of their surface layer, we can see that after 14– 15 days of cultivation of the fungi all pa-
rameters are stabilized, metabolic processes decay and cells pass into a quiescent state.
Biodamage of Materials: Adhesion of Microorganisms on the Surface of Materials 339
For small-size conidia (A. terreus, P. funiculosum) the concentrations of the main com-
ponents of the cell wall practically do not change, whereas for conidia of large size (T. vir-
ide, A. niger) such changes are observed (Table 4). Obviously, for conidia of small size the
surface layer of the cell is formed in the time not exceeding 3 days. For conidia of larger
size (for example, A. niger) the process of surface layer formation does not cease in 3 days
and continues for a longer time. At the first moment of adhesion, albumin macromolecules
evolve intensively [11]. For the case of A. niger conidia, the albumin component is in-
creased in the course of cell growth and their adhesive interaction also increases.
Thus, experimental material suggests that the adhesion force of conidia of microscopic
fungi in the course of their growth changes differently. It depends on the change of the con-
centration of albumin components in the surface layer of the cell having a mosaic structure.
In the cases when these changes are significant (predominantly for large-size conidia) the
force of adhesion interaction with surfaces increases and if there are no clear changes (pre-
dominantly for small-size conidia) the adhesion force does not change.
References
1. N.M. Emanuel and A.L. Buchachenko, Chemical Physics of Ageing and Stabilization of
Polymers, Moscow: Nauka (1982) (in Russian).
2. E.A. Ermilova, Theoretical and Practical Basis of Microbiological Destruction of Chemical
Fibers, Moscow: Nauka (1991) (in Russian).
3. P.S. Hocking, J. Macromol. Sci., Rev. Chem. Phys., 1, 35 (1992).
4. M.R. Timmens and R.W. Lenz, Trends in Polym. Sci., 1, 15 (1994).
5. K.Z. Gumargalieva, G.E. Zaikov and Yu.V. Moiseev, Usp. Khim., 63 (10), 905 (1994) (in
Russian).
6. K.Z. Gumargalieva, G.E. Zaikov and Yu.V. Moiseev, Khim. Fizika, 14 (10), 29 (1995) (in
Russian).
7. H.C. Flemming, Polym. Degrad. Stability, 59 (1–3), 309 (1998).
8. M.S. Fel’dman, S.I. Kirsh and V.M. Pozhidaev, in: Biological Basis of Protection of Industrial
Materials from Biodamage, N. Novgorod (1991) (in Russian).
9. V.F. Smirnov and A.S. Semicheva, Conf. Biological Problems of Ecological Materials, Penza
(1995) (in Russian).
10. F.B. Oppermann, S. Pickartz and A. Steinbuchel, Polym. Degrad. Stability, 59 (1), 337 (1998).
11. A.M. Gallardo-Moreno, M.L. Gonzalez, J.M. Bruque and C. Pijrez-Giraldo, 1st Int. Meeting on
Appl. Phys. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 249,
No.1–3, 99 (2004).
12. Microbiological Adhesion and Aggregation, ed. by K.C. Marshall, Berlin etc.: Springer Verlag
(1984).
13. V.A. Fomin and V.V. Guzeev, Plast. Massy, 2, 42 (2001) (in Russian).
14. L. Richard, W. Bizzoco, R. Bass, Thuy T. Vuong, James B. Vahl, Corona L. Hoang and Melina
M. Diaz, J. Microbiol. Meth., 55 (3), 787 (2003).
15. Bing-Mu Hsu and Chihpin Huang, Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 201 (1), 201 (2002).
16. M. Mercier-Bonin, K. Ouazzani, P. Schmitz and S. Lorthois, J. of Colloid and Interface Sci.,
271 (2), 342 (2004).
17. J. Schauersberger, M. Amon, D. Aichinger and Apostoulos Georgopoulos, J. of Cataract and
Refractive Surgery, 29 (2), 361 (2003).
18. A.D. Zimon, Adhesion of Dust and Powders, Moscow: Khimiya (1976) (in Russian).
19. A.M. Raichur and S.P. Vijayalakshmi, Fuel, 82 (2), 225 (2003).
20. H. Onose, T. Miyazaki and S. Nomoto, J. Dent. Res., 59 (7), 1179 (1980).
21. A.S. Gordon, S.M. Gerchakov and L.R. Udey, Canad. J. Microbiol., 27 (7), 698 (1981).
22. P.J. Boyle, M. Walch and R. Mitchell, VIII Int. Congr. Microbiology: Program and Abstracts,
Boston: Int. Union of Microbiol. Soc., 75 (1982) (in Russian).
340 K.Z. Gumargalieva, I.G. Kalinina, S.A. Semenov and G.E. Zaikov
10 Controlled Release of
Aseptic Drug from
Poly(3-hydroxybutyrate)
Films: A Combination of
Diffusion and Zero-order
Kinetics
Introduction
Dramatically increasing costs of hydrocarbon raw materials stimulate the development of
new polymers, which do not depend on production of oil and gas. Fermentative biosynthesis
of poly(3-hydroxybutyrate) (PHB) and its homologues poly(3-hydroxyalkonoates) (PHAs)
is based on using renewable substrates. Hydrocarbon wastes of food and wine/juice indus-
tries (sugars, molasses, starch etc.) are the basic “structural material” for bacterial PHB (and
PHA). Utilization of hydrocarbons in biosynthesis of PHA is ecologically efficient.
In the recent decade, PHB and its copolymers have begun to be productively used in
medicine [e.g., 1–3]. For example, composites of PHB are characterized by a high biocom-
patibility to bone tissues that enables their use as bioresorbable osteo implants [4]. Modified
PHB works as a highly effective scaffold in tissue engineering and promotes proliferation,
adhesion and production of cells [5, 6]. PHB–PEG blends [7] have a good hemocompati-
bility. It is reported [8] that, in contact with blood, surfaces of PHB and PHBcoHV films
do not activate hemostatic changes on the cell level. Great progress for poly(4-hydroxybu-
tyrate) (P4HB) is observed in cardio implantation [9]. Artificial heart valves produced by
stereolithography [10] with P4HB and controlled by x-ray tomography have demonstrated
a relevant combination of mechanical properties and hemocompatibility [11]. Fabrication
of a stent based on PHB has been reported [12].
For this reason, PHB and its derivatives can be considered to be new promising med-
ical materials for tissue engineering [13], design of osteoprostheses with replacement of
biodegradable material by germinated bone tissues [14], and hemocompatible coatings for
cardiovascular surgery [15]. Within the framework of this paper, it should be emphasized
that there is a broader area of PHB applications: design of matrices, reservoirs, and micro/
nano-particles for controlled drug release [16–18]. In this case, information on the biocom-
patibility, rate of resorption and diffusion characteristics of polymer systems is required.
The aim of this paper is to design and study a therapeutic PHB system loaded with an aseptic
(furacillin) and intended for the release of the drug into biological aqueous media.
Recently [19] we have shown that water diffusion in PHB films 40-60 µm thick is com-
pleted in several tens of minutes, whereupon the films absorbed the limiting equilibrium
concentration of water (ca. 1 wt. %). Structural relaxation in PHB under humid conditions
is achieved after a longer period of time (nearly 1000 min). We have investigated the kinet-
ics of the release for several tens of days, therefore, to a first approximation, water transport
phenomena in PHB are not essential. However, the long-term kinetics of drug release from
PHB films has an intricate form and requires special analysis for both diffusion modeling
and drug delivery applications.
1 Experimental
We used PHB from Biomer (Lot F16). The initial PHB powder was solved in chloroform
under long-term boiling. The hot polymer solution was filtered, and the molecular weight
R.Y. Kosenko, Y.N. Pankova, A.L. Iordanskii, A.P. Bonartsev, and G.E. Zaikov 343
of PHB was determined by the viscosimetry technique in accordance with the procedure
described in [20]. The averaged value of MW is 183.5×103 g/mol. As an aseptic drug we
used furacillin (MW = 198 g/mol)
The basic characteristics of the polymers include density = 1.25 g/cm3, Tm = 178°C,
Tg = 9°C, degree of crystallinity = 70% (determined by WAXS data).
Films of PHB containing furacillin were prepared in a two-stage procedure. 1 g of
powder was mixed with 50 ml of dioxan and boiled in a retort with reverse glass refrigerator
as well. Then after cooling and removing dioxan by a vacuum pump, PHB and Fr were
solved in chloroform, which was followed by casting the film. The thickness of cast PHB
films varied from 120±10 µm to 180±15 µm, and the concentration of loaded Fr changed
in the sequence 0.5 > 1.0 > 1.5 > 1.75 > 2.0 > 3.0 > 5.0 wt. %. The drug release profiles of
PHB were registered in water and phosphate buffer (pH = 7.4) by the UV technique using
a Beckman DU65 UV spectrophotometer at 25°C.
0.45
0.40 5
0.35
0.30
0.25 3
Dt
0.20
0.15
0.10 1
0.05
0.00
0 5 10 15 20 25 30
time, days
Figure 1 Kinetic profiles of aseptic release. The figures show the initial concentrations of the drug
(wt. %); Dt is the optical density of the drug in a liquid medium at 373 nm.
344 Controlled Release of Aseptic Drug from Poly(3-hydroxybutyrate) Films
1.0 1
2
0.8
3
0.6
Gt/G00
0.4
0.2
0.0
0 5 10 15 20 25 30 35 40
(Time)1/2, (h)1/2
Figure 2 Impact of diffusion in the total drug release kinetics for PHB films (thickness, 180 µm).
Initial concentrations of the aseptic are (1) 1.75 wt. %, (2) 2.5 wt. %, (3) 3 wt. %; (5) 5 wt. %.
where D is the drug diffusion coefficient, cm2/sec; k is the kinetic constant of hydrolytic
destruction, sec –1; Ct is the concentration of the drug in the polymer, wt. %; x (cm) and t
(sec) are the coordinate and time of diffusion, respectively.
After subtracting the linear kinetic term (k · t) from the integral profile of release
Ct – kt ≡ Gt, (2)
The solution of Eq. 1a provided that Gt/Goo > 0.5 has the classical Fick’s form
where L is film thickness, cm; Gt is the concentration of the drug available for diffusion,
and other parameters were determined in Eq. 1.
Finding the logarithms of the left- and right-hand side parts of Eq. 3 and solving this
equation graphically in coordinates log (1 – Gt/Goo) – t:
we can calculate the diffusion coefficients of Fr in the PHB films. Numerical subtraction
of the linear contribution to the release kinetics from the total concentration of the drug
R.Y. Kosenko, Y.N. Pankova, A.L. Iordanskii, A.P. Bonartsev, and G.E. Zaikov 345
3.0
2.0
1.5
1.0
0.5
0.0
0 1 2 3 4 5
Total furacillin concentration, %
Figure 3 Relation between total and mobile concentrations of the drug in PHB films.
(shown in Fig. 1) was performed to separate the diffusion and kinetic components. After
computer calculation, the proper diffusion contribution is shown in Fig. 2. Here the results
are presented in classical diffusion coordinates: Gt/Goo – t1/2. We emphasize two features
of drug release profiles: 1) the linear character of the data in the above diffusion coordinates
holds within the limits 0 < Gt/Goo < 0.65 that attests a predominance of diffusion at the ini-
tial stage of release; 2) the slope of the lines in the same coordinates depends on the initial
concentration of the drug that demonstrates its concentration diffusivity dependence. Lim-
iting (equilibrium) values of the drug (Goo) dissolved in PHB films are needed to plot the
curves in Fig. 2. Additionally, these values show the portion of the drug taking part in mo-
lecular diffusion.
The isotherm of Fr sorption as the dependence of free diffusing concentration of the
drug on the total loaded drug concentration is shown in Fig. 3. From this figure it follows
that up to 3 wt. % both concentrations are related by a linear function. The deviation from
linearity is observed at the maximal concentration (5 wt. %) of the loaded Fr. At this point
the drug forms the proper phase in the polymer as yellow crystals, while the effect of phase
formation does not distort the manner of the kinetic curve (See Fig.2, the curve for 5 wt. %).
Summarizing the results presented in Figs. 1 and 2, we can estimate effective diffusion
coefficients for all initial concentrations of the drug. Figure 4 shows the concentration de-
pendence of diffusivities which has a maximum defined clearly in the drug concentration
area 1.0–1.5 wt. %. The rising branch of the curve D(C) results likely from disordering of
the PHB structure after drug loading. In contrast, the dropping branch is related to drug crys-
tal formation in the PHB matrix that occurs during the decrease of low-molecular-weight
component mobility. The formation of Fr crystals in PHB has been observed recently in our
work by Krivandin with WAXS technique [21].
Above we have mentioned that, simultaneously with the diffusion kinetics, the linear
kinetic process of Fr release is observed. In this case, the greater the initial concentration
of the loaded drug is, the higher the constant rate of drug release is. More informatively this
effect is shown in Fig. 5, where the exponential dependence of the degradation constant (k)
on the drug concentration is observed. Simultaneously with the measurement of the
346 Controlled Release of Aseptic Drug from Poly(3-hydroxybutyrate) Films
3.0
2.0
1.5
1.0
0.5
0.0
0 1 2 3 4 5
Total furacillin concentration, %
Figure 4 Dependence of release destruction constant (k) on loaded drug concentration for PHB
films.
1.6
Drug diffusion coefficient × 1010, cm2/s
1.2
0.8
0.4
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Aseptic concentration, wt. %
concentration of the drug desorbing PHB from films, we have determined the loss of weight
for polymer samples. The gravimetric measurements have shown that the polymer sample
loses its weight also in accordance with the linear law. Initial polymer without the drug has
the stable weight throughout the entire time of release. Preliminary experiments show that,
in contrast to enzymatic biodegradation of PHB going on the polymer surface [19], hydro-
lytic destruction involves all accessible volume of PHB that is supported by the increase in
brittleness and the decrease in the strength of PHB films.
R.Y. Kosenko, Y.N. Pankova, A.L. Iordanskii, A.P. Bonartsev, and G.E. Zaikov 347
3 Conclusion
We proposed a new polymer system for aseptic controlled release that includes films of
PHB and furacillin. The release goes on simultaneously in accordance with the kinetic
(polymer degradation) and diffusion mechanism. The rates of the kinetic mode for release
obey zero degree curves relative to time. The diffusion mode, which determines the profile
of release in the initial range of time (about the first week), was analyzed in greater detail.
The dependences both of diffusion coefficients and kinetic constant (k) of release on the
drug concentrations were demonstrated. These results are requisite for the further develop-
ment of drug release systems with multi-component action when several drugs simulta-
neously have a local action on biological tissues and cells.
4 Acknowledgement
The authors thank the Biomer Company (Krailing FRG) for the donation of PHB. Funding
for this research was provided partly by the Russian Foundation for Basic Research with
grants Nos 06-04-49339 and 08-03-00929-a.
References
1. R.W. Lenz and R.H. Marchessault, Biomacromolecules, 6 (1), 1–8 (2005).
2. D. Williams, Medical Device Technology, 16 (1), 9–10 (2005).
3. Y. Doi and C. Abe, Macromolecules, 23 (15), 3705–3707 (1990).
4. Y. Liu and M. Wang, Current Appl. Phys., 7 (5), 547–554 (2007).
5. R. Sodian, S.P. Hoerstrup, J.S. Sperling, D.P. Martin, S. Daebritz, J.E. Mayer Jr. and J.P. Vacanti,
ASAIO J., 46 (1), 107–110 (2000).
6. Biomaterials Science: An Introduction to Materials in Medicine, ed. by B.D. Ratner, A.S.
Hoffman, F.J. Schoen and J.E. Lemons, San Diego: Academic Press (1996).
7. G.X. Cheng, Z.J. Cai and L. Wang, J. Mater. Sci., 14, 1073–1078 (2003).
8. V.I. Sevastianov, N.V. Perova, E.I. Shishatskaya, G.S. Kalacheva and T.G. Volova, J. Biomater.
Sci. Polymer Ed., 14, 1029–042 (2003).
9. D.P. Martin and S.F. Williams, Biochem Eng. J., 16, 97–105 (2003).
10. R. Sodian, M. Loebe, A. Hein, D.P. Martin, S.P. Hoerstrup, E.V. Potapov et al., ASAIO J., 48,
12–16 (2002).
11. R. Sodian, S.P. Hoerstrup, J.S. Sperling et al., Ann. Thorac. Surg., 70, 140–144 (2000).
12. M. Unverdorben, A. Spielberger, A. Schywalsky et al., Cardiovasc. Intervent. Radiol., 25,
127–132 (2002).
13. G.T. Köse, S. Ber, F. Korkusuz et al., Biomaterials, 24, 4998–5007 (2003).
14. L.J. Chen and M. Wang, Biomaterials, 23, 2631–2639 (2002).
15. G.-Q. Chen and Q.Wu, Biomaterials, 26 (33), 6565–6578 (2005).
16. D.P. Martin, F. Skraly and S.F. Williams, US Patent 403242 (2003).
17. C.W. Pouton and S. Akhtar, Adv. Drug Deliver. Rev., 18, 133–162 (1996).
18. D. Sendil, I. Gürsel, D.L. Wise and V. Hasirci, J. Controlled Release, 59, 207–217 (1999).
19. A.L. Iordanskii and P.P. Kamaev, Polym. Science, 41B (1–2), 39–43 (1999).
20. R.H. Marchessault, K. Okamura and C.J. Su, Macromolecules, 3 (6), 735–740 (1970).
21. O.V. Shatalova, A.V. Krivandin, and A.L. Iordanskii, 6th Eur. Symposium on Polymer Blends.
Program and Abstracts. May 16–19, Max-Planck-Institut fur Polymerforschung (1999).
348 Controlled Release of Aseptic Drug from Poly(3-hydroxybutyrate) Films
Авторам:
11 Transport of Water
as a Structurally Sensitive
Process Characterizing the
Morphology of
Biodegradable Polymer
Systems
Introduction
Professor Nikolai M. Emanuel very attentively and, one could say, tremulously treated
young scientists. Direct contacts with Professor Emanuel were infrequent but became turn-
ing and, sometimes, key points of the scientific life of an inexperienced researcher. In our
case, his conceptions about diffusion processes playing not an accessory but a dominating
role in multi-phase biosystems served the basis of a scientific direction “Transport Phenom-
ena in Structurally Organized Polymer Systems”.
Why is it, in particular, diffusion? Why is it the diffusion namely of water? The human
habitat including its biological structures contains a considerable amount of water. Varia-
tion of its concentration and its functionalization determine all processes of vital functions
[1]. Concerning polymer systems, especially if they are used as biomedical systems (vas-
cular prostheses, matrixes for the controlled release of medicines, osteoprosthesis, new-
generation vascular stents, etc.) the necessity arises to obtain information about the state
water is in and the rate of its motion in intercrystal regions of polymer products [2].
From the fundamental point of view the diffusion of bipolar water molecules may be
considered as the motion of a specific probe extremely sensitive to polar functional groups
of macromolecules forming the polymer matrix. Moreover, water has a noticeable affinity
to various hydrophilic impurities occurring in the polymer at some concentration (remnants
of catalysts, plasticizers, monomers and other accompanying components of synthesized
materials or the residual content of albumens, lipids and slats in polymers of natural origin
[3]). Due to the small size of water molecules, such a probe overpasses nonpolar polymer
regions quickly enough. But its diffusion rate is significantly decreased in polar regions (for
example, near albumens’ and polyamides’ amine groups [4]) due to participation in the for-
mation and redistribution of hydrogen bonds in such hydrophilic polymers as cellulose,
PVA, etc. [5].
Investigations of water transport in polymer mixtures (where, as a consequence of par-
tial components’ compatibility, the influence of structural factors on the mechanical, barrier
and other operational properties is especially pronounced) take on a special significance. In
this work we try to show how the morphological characteristics of polymer mixtures influ-
ence the water diffusion parameters and how the latter correlate with the operational char-
acteristics of mixtures on the example of mixed compositions of:
1 Experimental
Materials. A superfine powder of poly-3-oxybutyrate (PHB) was kindly provided by
Biomer (Krailing, Germany). For preparation of mixed films from the solvent the powder
was preliminary dissolved in chloroform and the polymer solution was filtered through a
glass Schott 160 filter. The solubility of initial polymer powder in chloroform turned to be
Transport of Water as a Structurally Sensitive Process 351
close to 0.01 k/ml. The thickness of films was 100±5 µm. X-ray crystallinity of polymer
was 70% and according to the data of the DSC method it was 75%.
After filtration, the molecular mass of polymer films was determined by viscosimetry
using the following equation [6]:
[η ] = 7.7·10 –5 M ω
0.82
, (1)
where [η ] is the intrinsic viscosity of the polymer solution in chloroform. The viscosity-
average molecular mass of polymer, MW, turned to be equal to 150·103 g/mol. Granular
low density polyethylene (LDPE) (State Standard 583-020) was characterized by a molec-
ular mass of 250·103 and density 0.92 g/cm3. In this work, we also used polyvinyl alcohol
(PVA) of Russian brand 8/27 with the residual content of non-saponated acetate groups
8.2% and the concentration of sodium acetate 0.04 wt. %. The molecular mass of PVA is
equal to 64,000 g/mol and the melting temperature is 146°C.
Mixing in extruder. Mixture of LDPE with PHB: firstly, the preliminary mixing in a
ball mill was carried out. The ratios of the components (PHB/LDPE) were: 2/98, 4/96, 8/92,
16/84, and 32/68 wt. %. Blending was carried out in a single-auger extruder (ARP-20) with
the frequency of auger rotation 100 min –1 provided that the temperature of the melt was
not higher than 182°C. The plane-parallel slot was regulated in such a way so as to obtain
films 50–60 µm thick. A similar procedure was carried out when obtaining mixtures of
PHB/polyvinyl alcohol films (PVA, Russian brand 8/27) of structures 10/90, 20/80, 30/70,
and 50/50 wt. %. A similar procedure of extrusion was carried out for initial individual
polymers: PVA, LDPE and PHB.
Methods of analysis. The kinetic curves of water sorption / desorption and equilibri-
um sorption were measured by vacuum gravimetry using a quartz McBen’s microbalance
with the quartz spiral sensitivity of 0.67 mg/mm. The permeability of water vapors was reg-
istered in a traditional two-chamber cell described in [7]. The results of X-ray diffraction
at large angles were obtained with the help of a diffractometer with a linear-coordinate de-
tector constructed at the Joint Institute of Nuclear Research (Dubna, Russia). Since extru-
sion films are characterized by the orientation of both macromolecules and structural
elements in particular crystallites, such orientation was estimated by measuring their polar-
ization spectra. In investigations, the plane of the catalyst was oriented either along or trans-
versely to the direction of extrusion.
Spectral characteristics and round dichroism were measured on an IFS-48 Bricker IR
spectrometer.
2 Hydrophobization of poly-(3-hydroxybutyrate)
Differences in water diffusion behavior when comparing synthetic (gas and oil derivatives)
and natural (products of animals’ and plants’ vital functions) polymers are more clearly re-
vealed on the molecular and crystal level. For hydrophobic polymers (low density PE,
polypropylene, polysiloxane, synthetic rubbers, etc.), analysis of the chemical structure of
macromolecules shows the damage of the primary chemical structure and accumulation of
oxygen-containing functional groups in the polymer chain as a result of imperfect condi-
tions of polymer production. The appearance of groups “introduced” into the hydrophobic
matrix leads to an increase of equilibrium sorption of water and to a decrease of its effective
diffusion coefficient. The diffusion mobility is reduced as a result of water molecules’
352 A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and G.E. Zaikov
a
8
0
0 5 10 15 30 35
PHB, %
2.6
b
2.4
2.2
1/R (1228 cm_1)
2.0
1.8
1.6
1.4
0 5 10 15 20 25 30 35
PHB, %
Figure 1 Comparison of inverse permeability (a) and dichroism (b) values as a function of the struc-
ture of the polymer mixture for films based on LDPE–PHB.
interaction with the damaged polar groups of the polymer. For hydrophobic polymers of
natural origin, the structure of which is formed on the cell and enzyme levels, the damage
of the primary structure is less typical and may occur only at later technological stages, e.g.,
in processing or mixing in extruders.
Under normal conditions, the solubility of water in saturated hydrocarbons is limitedly
low (several mils of percent) [8]. In hydrocarbons C7 –C16 the mole fraction of water (Xw)
is written as the following equation:
where Vw is the water molar volume, ∆δ is the difference between the solubilities of water
and hydrocarbon, R and T are the gas constant and temperature, respectively. The authors
of [8] by extrapolation of eq. (1) showed that at 25°C the values of Xw did not exceed 10–6
mol/l (~2·10–5 wt. %), so this value might be considered as an ideal solubility of water in
the hydrocarbon matrix.
Transport of Water as a Structurally Sensitive Process 353
In both synthetic (polyolefines) and natural hydrocarbon polymers the limit of the sol-
ubility of water is significantly higher and is significantly changed depending on the poly-
mer production method. Systematic investigations of water diffusion in films of low and
high density PE showed that the emergence of damaged fragments on the polymer chain
influenced water molecules’ sorption and diffusion [9]. Immobilization of water molecules
on such polar groups statistically distributed in the hydrocarbon nonpolar matrix leads to a
noticeable decrease of the effective diffusion coefficient of water.
where Dwf is the diffusion coefficient of a single water molecule (free diffusion), Spf is the
concentration of introduced (damaged) groups, KH and KL are the Henry and Langmuir con-
stants of sorption, respectively.
With the aim of obtaining biodegradable polymer materials able to destruct in soil but
to keep their transport and mechanical properties during long exploitation period we inves-
tigated new polymer compositional mixtures of poly-3-oxybutyrate and LDPE [10]. For
such mixtures, water permeability depends on PHB/PE, the ratio of components’ concen-
trations, see Fig. 1a where the value of inverse permeability is presented, i.e. water diffusion
resistance as a function of PHB concentration. Introduction of carbonyl groups of POB into
the nonpolar matrix of polyethylene (LDPE) enables regulation of transport parameters.
The dependence “water resistance vs. POB concentration” described by the extreme and
non-additive function is in good agreement with the segment orientation acquired during
the extrusion process during the melting of PE/POB mixtures. The degree of orientation is
determined by the method of Fourier spectroscopy under construction of the dichroism
function [11] and is presented in Fig. 1b.
The reason for the extremums to appear on the “dichroism–mixture structure” curves
may be the covering of two processes oppositely influencing the orientation of macromol-
ecules in crystal regions (and consequently the orientation of crystals). On the one hand,
with the increase of the content of rigid molecules of PHB in the melt the tendency of or-
dering all molecules while passing through the slot of the extruder is increased. On the other
hand, the viscosity of the mixture in this case is increased and consequently the time nec-
essary for structural elements’ (including crystals’) orientation is also increased. At low
concentrations of PHB (>8 wt. %) the orientation process prevails. And at higher PHB con-
centrations the tendency of disordering of the structural organization begins to be exhibited
more clearly. The decrease of the degree of crystallinity of both components observed in-
dependently by the DSC method confirms this fact.
The results of X-ray analysis for large and small angles showed the presence of crystal
axial a-texture realized in investigated films for both PE and PHB [12]. Crystallites belong-
ing to both polymers are oriented preliminary in such a way that axes a of elementary crystal
cells of both mixture components coincide with the direction of extrusion, and axes c coin-
ciding with macromolecules’ orientation in crystallites’ elementary cells are oriented trans-
versely to this direction. Such a position of crystallites is characteristic of all compositions
studied. That is why a combination of earlier obtained X-ray conceptions about the crystal
structure of the POB/PE mixture and the results of polarization IR Fourier spectroscopy
suggest that the cause of inversion of the dichroism function is probably the difference in
the angles of transition moment for methyl groups belonging to PE (=0°C) and POB
(=90°C), respectively.
354 A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and G.E. Zaikov
3 Hydrophilization of poly-(3-hydroxybutyrate)
The results obtained by the DSC method for mixtures and initial samples of PHB and PVA
are presented in Table 1.
Table 1 Glass-transition temperature of mixture compositions with various contents of PHB.
As we mentioned above, all samples were obtained by the extrusion method as de-
scribed in the experimental part. Analysis of the thermograms showed that there was a sta-
ble low-temperature transition in the interval 24–54°C reflecting, most probably, the
transition from the glassy to rubberlike state at the glass-transition temperature of ~37°C.
We should note that for initial polymer components the corresponding glass-transition
temperatures lie in the region of 24.1°C and 53.9°C for PHB and PVA, respectively. The
existence in the mixture compositions of only one phase transition of the second order
should testify to the interaction between two polymers [13] forming the mixture and to their
possible partial compatibility in inter-crystallite regions of mixture films. A combination in
polymer crystallite regions (their part is large and is equal to more than 70%) most likely
does not occur since on DSC thermograms two independent peaks belonging correspond-
ingly to PHB and PVA are observed.
The occurrence of independent crystallite phases of each of mixture-forming polymers
was observed by the method of wide angle X-ray scattering (WAXS). In the interval of PHB
concentration from 0 up to 30% each of components forms its own, slightly oriented crystal
phase that is confirmed by X-ray diffractograms of the mixtures; they represent a super-
position of diffractograms of individual polymers: PHB (degree of crystallinity ~70%) and
PVA (degree of crystallinity ~30%).
At all polymer ratios forming the given mixtures, the parameters of the elementary
orthorhombic crystal cell of PHB remain constant and have the following values, respec-
tively: a = 0.576 nm, b = 1.32 nm, and c = 0.596 nm [14]. X-ray reflexes for PVA show
that crystalline regions of polyvinyl alcohol are in gamma form well described earlier in
domestic literature by L.L. Razumovskaya et al. [15]. In particular, the reflex S = 2.21
nm –1 belongs to this modification of PVA structure.
The form of diffractograms obtained at various angles of X-ray irradiation incidence
in relation to sample plane shows that for all samples containing PVA the axial cylindrical
texture is observed where the axis of texture coincides with direction of extrusion through
the slot. In samples containing 10 and 20% of PHB, the axial texture is observed also for
poly-3-hydroxybutyrate. However, in contrast to PVA where macromolecules are oriented
along the direction of extrusion, in crystal regions of poly-3-hydroxybutyrate the axis a of
elementary crystal cell coincides with the direction of extrusion and, consequently, biopoly-
mer macromolecules are preliminarily arranged transversely to the axis of extrusion [16].
Diffractograms of mixture compositions containing 30 and especially 50% of PHB
show the absence of texture of the latter and so, these sample are preliminarily isotropic.
The transition from textured to isotropic crystal structures observing in the region of 30%
of PHB is probably related to the phase inversion of samples occurring in the mentioned
region of mixture structure. A sharp change of structural and physical characteristics of
polymer mixtures occurs namely in the region of phase inversion, see, e.g., [17].
Transport of Water as a Structurally Sensitive Process 355
4
PHB(110)
PHB(020) 1
3 PVA(110)
2
I, rel. units
PHB(111)
2
0
1.0 1.5 2.0 2.5 3.0 3.5
S, 1/nm
6
PHB(020)
PHB(110)
4
I, rel. units
PVA(110)
2 3
2
2
1
1, 3
0
1.0 1.5 2.0 2.5 3.0 3.5
S, 1/nm
Figure 2 WAXS diffraction of extrusion films prepared by mixing PHB and PVA (upper, 80:20%
and lower, 70:30%). The vector of X-ray irradiation is directed along the direction of extrusion (1),
transversely (2) and at an angle of 20º (3) to the direction of extrusion. Results and Discussion were
prepared by O.V. Shatalova and A.V. Krivandin (IBCS RAS).
Taking into account the results presented above, we studied the physicomechanical
characteristics of polymer mixtures with various contents of PHB. It is important to note
that on the curves “rupture stress–structure of mixture” and “elongation at rupture–
structure of mixture” the sharp changes of values of mechanical parameters are observed,
and in the same concentration interval (~ 30% PHB), where the above mentioned transition
occurs. Besides, the minimum of the extreme dependence of elasticity modulus on PHB
concentration in the mixture is situated in the same concentration interval. A decrease of
elasticity modulus in the region of low concentrations of PHB is caused by the reduction of
segment mobility of interaction macromolecules of PHB and PVA. A further increase of
elasticity modulus values reflects the increase of structural units’ content responsible for
material strength and namely of crystallites belonging to poly-3-hydroxybutyrate. The to-
tality of the results of mechanical tests for mixture films presented in Figure 3 confirms the
physical concept of structural inversion, which concerns both crystal and inter-crystal (con-
ditionally amorphous) regions of mixture compositions.
Transformation of amorphous polymer regions as a result of phase inversion should
be reflected on the level of diffusion processes and especially under diffusion and
356 A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and G.E. Zaikov
100
250
Elongation at rupture, %
75 200
Rupture stress, MPa
150
50
100
50
25
0
0 20 40 60
PHB content, wt. %
Figure 3 Dependence of the tensile strain and rupture stress on the structure of a PHB–PVA poly-
mer mixture.
40
g/cm3 . molecular mass of mercury
35
30 T = 20°C
Water solubility,
25
20
15
10
5
0
0 20 40 60 80 100
PHB concentration, wt. %
200
of mercury 150
100
50
0 20 40 60 80 100
PHB concentration, wt. %
Figure 5 Extreme dependence of water permeability through films based on PHB–PVA mixtures.
of these OH groups of polymer. The contribution of the ester group of PHB is significantly
lower and in this particular case we can neglect it. Actually, the equilibrium sorption of wa-
ter in investigated mixtures similarly increases monotonously with the rise of PVA concen-
tration or with the decrease of the PHB content (Figure 4). Characteristic inflections and
extreme points on the water sorption–mixture structure dependence are not observed and
consequently the equilibrium sorption is sensitive mainly to the primary chemical structure
of polymer, to be more exact, to the chemical structure of the polymer mixture. Thus, a mo-
notonous increase of water equilibrium sorption with the change of the mixture structure is
determined by the substitution of ester functional groups belonging to PHB and possessing
a low affinity to water molecules by more hydrophilic groups (hydroxyl) belonging to PVA.
In contrast to equilibrium sorption data the values of normalized water flows (perme-
ability) have a characteristic degraded maximum in the concentration region corresponding
approximately to 40% of PHB as shown in Figure 5. Obviously, in this case as a result of
rebuilding the structure the maximum accumulation of structural defects occurs due to
which the increase of water flow is observed.
In the interval of concentrations preceding this point (10–30% PHB) there is no
additivity of the permeability parameter and even a small addition of polyester (10%)
significantly reduces the values of water permeability first of all at the expense of segment
interaction between PHB and PVA macromolecules forming the mixture. This conclusion
is also confirmed by the shift of mixture glass-transition temperature. We want to note once
again that such interactions are realized in intercrystalline or better to say amorphous re-
gions and do not affect the crystalline regions of polymer. The descending line of the water
permeability–mixture structure dependence is approximately described by a linear function
and consequently obeys a property–structure additive dependence. So, in this region of
structure each of components independently leaks water flow and, thus, segment interaction
revealed in the concentration region below the point of structural transition is practically
absent.
When one analyzes the dependence of water permeability on structure, one should
remember that from the physical point of view permeability represents the product of two
other physical parameters: sorption capacity and diffusion mobility. We considered water
358 A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and G.E. Zaikov
30 100
25
Diffusion coefficients 108, cm2/sec
=σ = Dw
15
50
10
5 25
0
0 10 20 30 40 50
PHB concentration, wt. %
Figure 6 Similarity of the behavior of the diffusion coefficients and rupture stress for PHB–PVA
mixture compositions.
sorption above, and as the parameter characterizing the mobility of water molecules in mix-
ture films we chose the coefficient of water diffusion. In Figure 6 we show these coeffi-
cients in the form of a dependence on mixture structure. An analogous dependence for
rupture stress is presented in the same figure for clarity. Both the diffusion and mechanical
characteristics of mixture compositions are characterized by clear inflection points situated
in the region of structural transition at 30% of PHB that coincides with the results of X-ray
analysis. As in the case of mechanical tests in the region of low content of PHB the diffusion
behavior of the system resembles the diffusion in pure PVA (the region 0–20%) whereas
at concentrations exceeding 30% of PHB the diffusion coefficients of mixture compositions
sharply exceed the diffusion coefficients of water in PHB (Dw = 1–10-8 cm2/sec [19]).
In accordance with the results of water permeability measurements we assume that
high values of the diffusion coefficients are caused by structure defects more clearly re-
vealed as a result and after phase transition. Earlier, by WAXS, EPR, Fourier IR spectros-
copy and vacuum gravimetry methods we showed [20, 21] that crystalline transition from
anisotropic into isotropic state of PHB led to the change of water diffusion mobility. In the
case of the change of mixture structure we also observe the transition from the textured state
into a more disordered one, i.e. isotropic. Besides, an increase of defects of mixture com-
positions at a relatively high content of PHB does not make it possible to exclude the pos-
sibility of partial water leakage through the defect zones in the matrix analogously to the
case observed for PHB–LDPE mixtures [22].
Intensive studying of particularities of water transfer in mixture compositions on the
basis of poly-3-hydroxybutyrate represents a necessary stage in the investigation of such
fundamental processes as physical ageing at high dampness, corrosion stability of polymer
membranes; these results should also promote the development of novel ecologically and
biologically compatible materials for medicine and agriculture.
A complex of methods combining structural (DSC, IR Fourier polarization spectros-
copy), mechanical and diffusion experiments was used in this work. By varying the nature
Transport of Water as a Structurally Sensitive Process 359
of the second mixing component (its hydrophilicity) we may control the morphology of the
mixture and, consequently, the rate of diffusion processes. For example, mixing of PHB
with a representative of the class of hydrophobic polymers (LDPE) does not give compat-
ible compositions. Matrices of PHB–LDPE mixtures are partially crystalline, oriented in
relation to extrusion direction. The direct correlation between the diffusion resistance of
film to water flow and the degree of orientation of macromolecules is shown. On the con-
trary, hydrophilization of PHB by mixing with PVA gives the possibility of preparing novel
mixture films partially compatible in intercrystalline regions. The region of structural tran-
sition (at 30% of PHB) where mechanical and transport characteristics are sharply changed
was found.
References
1. E.A. Vogler, Biological Properties of Water, in: Water in Biomaterials. Surface Science, Wiley:
Chichester–New York–Weinheim–Brisbane–Singapore–Toronto, Ch. 1, 3 (2001).
2. Water in Polymers, ACS Symposium Series 127, ed. by Rowland (1980).
3. J.H. Lee, T. Li and K. Park, Solvation Interactions for Protein Adsorption to Biomaterial
Surfaces, in: Water in Biomaterials. Surface Science, ed. by M. Morra, Wiley: Chichester–New
York–Weinheim–Brisbane–Singapore–Toronto, Ch. 5, 127 (2001).
4. G.E. Zaikov, A.L. Iordanskii and V.S. Markin, Diffusion of Electrolytes in Polymers, Ser. New
Concepts in Polymer Science, VSP Science Press: Utrecht–Tokyo (1988).
5. A.L. Iordanskii, P.P. Kamaev, A.A. Ol’khov, and A.M. Wasserman, Desalination, 126, 139,
(1999).
6. R.H. Marchessault, K. Okamura and C.J. Su, Macromolecules, 3 (6), 735 (1970).
7. J. Crank and G.S. Park, Diffusion in Polymers, Academic Press: London–New York (1968).
8. B. Schatzberg, J. Phys. Chem., 3 (67), 775 (1963).
9. D.W. McCall, D.C. Douglass and L. Blyler, Macromolecules, 17, 1644 (1984).
10. A.A. Ol’khov, A.L. Iordanskii, S.V. Vlasov et al., Russian Polymer News, 5 (1), 34 (2000).
11. R. Zbinden, Infrared Spectroscopy of High Polymers, Academic Press: New York (1966).
12. A.A. Ol’khov, S.V. Vlasov, L.S. Shibryaeva and A.L. Iordanskii, Polymer Sci. Ser. A., 42 (4),
676 (2000).
13. D.R. Paul and C.B. Bucknal, Polymer Blends, Wiley: New York, 1 (2000).
14. A.L. Iordanskii, A.A. Ol’khov, O.V. Shatalova, G.E. Zaikov, and U.J. Hanggi, Water Diffusion,
Crystalline Structure, and Mechanical Properties of Novel PHB–PVA Blends, ed by
A.L. Iordanskii, O.V. Startsev and G.E. Zaikov, Nova Science: New York, Ch. 9, 213 (2004).
15. L.L. Razumova, O.V. Shatalova, A.L. Iordanskii and G.E. Zaikov, Vysokomol. Soed., 38A, 1271
(1976) (in Russian).
16. A.A. Ol’khov, S.V. Vlasov, A.L. Iordanskii, G.E. Zaikov and V.M. Lobo, J. Appl. Polymer, 90,
1471 (2003).
17. J.P. Runt, Crystalline Polymer Blends, in: Polymer Blends, ed. by D.R. Paul and C.B. Bucknall,
Wiley: New York, 1, 167 (2000).
18. A.E. Chalykh, Diffusion in Polymers, Khimiya: Moscow (1984) (in Russian).
19. P.P. Kamaev, I.I. Aliev, A.L. Iordanskii and A.M. Wasserman, Polymer (UK), 42, 515 (2001).
20. P.P. Kamaev, A.L. Iordanskii, I.I. Aliev, A.M. Wasserman and U.J. Hänggi, Desalination, 126,
153 (1999).
21. A.L. Iordanskii and P.P. Kamaev, Polymer. Sci. Ser. B, 41 (1–2), 39 (1999).
22. A.L. Iordanskii, From Traditional to Novel Environmentally Friendly Polymers, in: Water
Transport in Synthetic Polymers, ed. by A.L. Iordanskii, O.V. Startsev, and G.E. Zaikov, Nova
Science: New York, Ch. 1, 1 (2004).
360 A.L. Iordanskii, Yu.N. Pankova, R.Yu. Kosenko, A.A. Ol’khov and G.E. Zaikov
статья одобрена автором
больше не смотреть
Всё сделано - только номера страниц
Keywords: electrospinning, fiber diameter, image analysis, direct tracking, distance trans-
form
Introduction
Conventional fiber spinning (like melt, dry and wet spinning) produce fibers with diameters
in the range of a micrometer. In recent years, electrospinning has gained much attention as
a useful method to prepare fibers in nanometer diameter range. These ultrafine fibers are
classified as nanofibers. The unique combination of high specific surface area, extremely
small pore size, flexibility and superior directional strength makes nanofibers a preferred
material form for many applications. Proposed uses of nanofibers include wound dressing,
drug delivery, tissue scaffolds, protective clothing, filtration, reinforcement and micro-
electronics.
362 M. Ziabari, V. Mottaghitalab and A.K. Haghi
α
d
O
In the electrospinning process, a polymer solution held by its surface tension at the end
of a capillary tube is subjected to an electric field. Charge is induced on the liquid surface
by an electric field. Mutual charge repulsion causes a force directly opposite to the surface
tension. As the intensity of the electric field is increased, the hemispherical surface of the
solution at the tip of the capillary tube elongates to form a conical shape known as the Tay-
lor cone. When the electric field reaches a critical value at which the repulsive electric force
overcomes the surface tension force, a charged jet of the solution is ejected from the tip of
the cone. Since this jet is charged, its trajectory can be controlled by an electric field. As
the jet travels in the air, the solvent evaporates, leaving behind a charged polymer fiber
which lays itself randomly on a collecting metal screen. Thus, continuous fibers are laid to
form a nonwoven fabric. Figure 1 illustrates the electrospinning setup [1–6].
The properties of electrospun nonwoven webs depend on the nature of the component
fiber as well as its structural characteristics such as fiber orientation [7–12], fiber diameter
[13], pore size [15], uniformity and other structural features [16]. Analyzing the electrospun
webs yield results and information, which will help researchers in improving the quality and
predicting the overall performance of electrospun webs. Some of the reasons for character-
ization may be process control, process development and product or quality control. Fiber
diameter is the most important structural characteristics in electrospun nonwoven webs. De-
pending on the process and material variables, the diameter of fibers produced by electro-
spinning varies. Almost all researches who have done characterization, have reported the
effects of processing variables on electrospun fiber diameter. There is no standard technique
to measure the fiber diameter and analyze its distribution. This explains the need to study
the fiber diameter of electrospun webs. Recently, image analysis has been used to identify
A Novel Technique for Measurement of Electrospun Nanofiber 363
fibers and measure structural characteristics in nonwovens. However, the accuracy and lim-
itations of these techniques have not been verified.
The objective of our research is to use image analysis for measuring electrospun fiber
diameter.
1 Methodology
It is assumed that the lines are infinitely long (continuous filament) so that, at least in
the image plane, all lines intersect the boundaries.
The aim is to obtain unbiased arrays spatially homogeneous for infinitely long lines.
Lately it was discovered by Pourdeyhimi et al. [7] that the best way to simulate nonwovens
of continuous fibers is through the second method. Under this scheme, a line with a speci-
fied thickness is defined by the perpendicular distance d from a fixed reference point O
located in the center of the image and the angular position of the perpendicular α. Distance
d is limited to the diagonal of the image [7, 17]. Figure 2 demonstrates this procedure.
Several variables are allowed to be controlled during the simulation:
1) Web density: can be controlled using the line density which is the number of lines to
be generated in the image.
2) Angular density: useful for generating fibrous structures with specific orientation
distribution. The orientation may be sampled from either a normal or a uniform
random distribution.
3) Distance from the reference point: varies between 0 and the diagonal of the image,
restricted by the boundary of the image and is sampled from a uniform random
distribution.
4) Line thickness (fiber diameter): is sampled from a normal distribution. The mean
diameter and its standard deviation are needed.
5) Image size: can also be chosen as required.
to understand the behavior of fiber diameter and fiber diameter distribution in the electro-
spun web as impacted by the processing parameters. The first step in determining fiber di-
ameter is to produce a high quality image of the web at a suitable magnification using
electron microscopy techniques, called micrograph. The methods for measuring electro-
spun fiber diameter are described in the following sections.
a) b)
c)
Figure 4 a) A simple simulated image, b) its skeleton overlaid on its distance transform, c) a histo-
gram of fiber diameter distribution obtained by distance transform.
distance transformed image by recording the intensities to compute the diameter at all
points along the skeleton. This method was proposed by Pourdeyhimi et al. [13]. Figure 4
shows a simple simulated image, its skeleton overlaid on its distance transform and the his-
togram of fiber diameter obtained by this method.
d
y
a) b)
Figure 5 a) Direct tracking, b) fiber diameter from the number of vertical and horizontal pixels.
a) b)
Figure 6 a) A simple simulated image which is labeled, b) histogram of fiber diameter distribution
obtained by direct tracking.
a) b)
c)
[18–21]. As it is shown in Fig. 7, global thresholding resulted in some broken fiber seg-
ments. This problem was solved using local thresholding.
2 Experimental
Electrospun nonwoven webs used as real webs in image analysis obtained from electrospin-
ning of PVA with an average molecular weight of 72,000 g/mol, purchased from Merck, at
different processing parameters. The micrographs of the webs were obtained using a Philips
XL-30 environmental scanning electron microscope under magnification of 10,000× after
being gold coated. Figure 8 shows micrographs of electrospun webs used as real webs.
R1 R2
R3 R4
R5
simulation. For the line thickness of 15, the standard deviation of the diameter obtained
from the direct tracking method is closer to that of the simulation. However, in this case the
distance transform measured the average diameter more accurately. For the simulated webs
with a line thickness of more than 15 in the first set, the direct tracking method resulted in
a better estimation of the mean and standard deviation of fiber diameter. This is due to the
fact that as the lines get thicker, there is a higher possibility of branching during the skele-
tonization (or thinning), and these branches remain even after pruning. Although these
branches are small, their orientation is typically normal to the fiber axis, thus causing a wid-
ening of the distribution.
Table 1 Structural characteristics of first set images.
C1 0–360 30 5
C2 0–360 30 10
C3 0–360 30 15
C4 0–360 30 20
C5 0–360 30 25
A Novel Technique for Measurement of Electrospun Nanofiber 369
C1 C2
C3 C4
C5
Mean Standard
V1 0–360 30 15 2
V2 0–360 30 15 4
V3 0–360 30 15 6
V4 0–360 30 15 8
V5 0–360 30 15 10
Furthermore, the distance transform method fails in measuring the diameter of fiber in
intersections. Intersections cause to overestimate fiber diameter. Since in the direct tracking
method an image is divided into parts where single fibers exist, the effect of intersections
which causes an inaccurate measurement of fiber diameter is eliminated. Therefore, there
will be a better estimate of fiber diameter.
370 M. Ziabari, V. Mottaghitalab and A.K. Haghi
V1 V2
V3 V4
V5
In the second set, irrespective of the line thickness in the simulation, for all simulated
webs direct tracking resulted in a better measurement of mean and standard deviation of
fiber diameter. For real webs, mean and standard deviation of fiber diameter for direct track-
ing were closer to those of the manual method, which concurs with the trends observed for
the simulated images.
C1 C2 C3 C4 C5
Simulation mean 5 10 15 20 25
standard 0 0 0 0 0
Distance transform mean 5.486 10.450 16.573 23.016 30.063
standard 1.089 2.300 5.137 6.913 10.205
Direct tracking mean 5.625 11.313 17.589 22.864 29.469
standard 1.113 2.370 4.492 5.655 7.241
A Novel Technique for Measurement of Electrospun Nanofiber 371
C1 C2
C3 C4
C5
V1 V2 V3 V4 V5
V1 V2
V3 V4
V3 V4
V5
R1 R2 R3 R4 R5
R1 R2
R3 R4
R5
4 Conclusion
Fiber diameter is the most important structural characteristics in electrospun nonwoven
webs. The typical way of measuring electrospun fiber diameter is through the manual meth-
od which is a tedious, time-consuming and an operator-based method and cannot be used
as an automated technique for quality control. We investigated the use of image analysis
for determining the fiber diameter and developed an automated method called direct track-
ing. Since this is a new technique, its accuracy needs to be evaluated using samples with
known characteristics. To that end, the µ-randomness procedure was used in order to sim-
ulate electrospun nonwoven webs. Based on this scheme, two sets of simulated images,
each containing 5 webs, were generated. The first set had random orientation with increas-
ing constant diameter. For the second set, the diameter values were sampled from normal
distributions with a mean of 15 and standard deviation ranging from 2 to 10 pixels. We
374 M. Ziabari, V. Mottaghitalab and A.K. Haghi
compared our method with the distance transform method. For all the simulated webs with
varying diameter and for those with constant diameter more than 15, the direct tracking
method resulted in the mean and standard deviation closer to the simulation. However, for
the simulated webs with smaller constant diameter, the distance transform measured the
mean and standard deviation of fiber diameter more accurately. The results suggest that the
direct tracking method is an accurate, direct measurement technique, because it extracts the
fiber diameter for the samples by tracking fixed segment of the fiber and eliminates the ef-
fect of intersections.
We have demonstrated the general applicability of the method using real webs. Five
real electrospun nonwoven webs obtained by electrospinning of PVA were used. Since the
methods needed binary images as input, the images first had to be segmented. A local
thresholding method together with Ostu’s method was employed in order to automatically
compute the appropriate threshold. The results obtained for real webs confirm the trends
suggested by simulated images. The results show that the use of image analysis in order to
determine the fiber diameter in electrospun nonwoven webs has been successful.
References
1. A. K. Haghi and M. Akbari, Phys. Stat. Solidi (a), 204, 1830–1834 (2007).
2. J. Doshi and D.H. Reneker, J. Electrostatics, 35, 151–160 (1995).
3. D.H. Reneker and I. Chun, Nanotechnology, 7, 216–223 (1996).
4. H. Fong and D.H. Reneker, Electrospinning and the Formation of Nanofibers, in: D.R. Salem,
Structure Formation in Polymeric Fibers, Hanser, Cincinnati, pp. 225–246 (2001).
5. Th. Subbiah, G.S. Bhat, R.W. Tock, S. Parameswaran and S.S. Ramkumar, J. Appl. Polymer Sci.,
96, 557–569 (2005).
6. W. Zhang, Z. Huang, E. Yan, Ch. Wang, Y. Xin, Q. Zhao and Y. Tong, Mater. Sci. Eng. A, 443,
292–295 (2007).
7. B. Pourdeyhimi, R. Ramanathan and R. Dent, Textile Res. J., 66, 713–722 (1996).
8. B. Pourdeyhimi, R. Ramanathan and R. Dent, Textile Res. J., 66, 747–753 (1996).
9. B. Pourdeyhimi, R. Dent and H. Davis, Textile Res. J., 67, 143–151 (1997).
10. B. Pourdeyhimi and R. Dent, Textile Res. J., 67, 181–190 (1997).
11. B. Pourdeyhimi, R. Dent, A. Jerbi, S. Tanaka and A. Deshpande, Textile Res. J., 69, 185–192
(1999).
12. B. Pourdeyhimi and H.S. Kim, Textile Res. J., 72, 803–809 (2002).
13. B. Pourdeyhimi and R. Dent, Textile Res. J., 69, 233–236 (1999).
14. B. Xu and Y.L. Ting, Textile Res. J., 65, 41–48 (1995).
15. A.H. Aydilek, S.H. Oguz and T.B. Edil, J. Comput. Civil Eng., 280–290 (2002).
16. R. Chhabra, Int. Nonwoven J., 43–50 (2003).
17. M.S. Abdel-Ghani and G.A. Davis, Chem. Eng. Sci., 40, 117–129 (1985).
18. R.C. Gonzalez and R.E. Woods, Digital Image Processing, 2nd edn, Prentice Hall (2001).
19. B. Jähne, Digital Image Processing, 5th revised and extended edn, Springer (2002).
20. M. Petrou and P. Bosdogianni, Image Processing: the Fundamentals, Wiley (1999).
21. J. Serra, Image Analysis and Mathematical Morphology, Academic Press, London (1982).
статья одобрена автором
больше не смотреть
Всё сделано - только номера страниц
Nanofibers produced by the electrospinning method are widely used for drug
delivery, as tissue scaffolding materials and filtration purposes where specific pore
characteristics are required. For continued growth in these areas, it is critical that
nanofibers be properly designed for these applications to prevent failure. Most of
the current methods only provide an indirect way of determining the pore
structure parameters and contain inherent disadvantages. In this study, we
developed a novel image analysis method for measuring the pore characteristics of
electrospun nanofiber webs. Five electrospun webs with different pore
characteristics were analyzed by this method. The method is direct and rapid and
presents valuable and comprehensive information regarding the pore structure
parameters of webs. Two sets of simulated images were generated to study the
effects of web density, fiber diameter and its variations on pore characteristics.
The results indicated that web density and fiber diameter significantly influence
the pore characteristics whereas the effect of fiber diameter variations was
insignificant.
Introduction
Fibers with a diameter of around 100 nm are generally classified as nanofibers. What makes
nanofibers of great interest is their extremely small size. Nanofibers compared to conven-
tional fibers, with higher surface area to volume ratios and smaller pore size, offer an op-
portunity for use in a wide variety of applications. To date, the most successful method of
376 M. Ziabari, V. Mottaghitalab and A.K. Haghi
1 Methodology
The porosity, εV, is defined as the percentage of the volume of the voids, Vv, to the total
volume (voids plus constituent material), Vt, and is given by
Vv
εV = ×100 .
Vt
Image Analysis of Pore Size Distribution in Electrospun Nanofiber Webs 377
Figure 2 Equivalent opening size, Oi, based on (a) equivalent area, (b) equivalent size.
Similarly, the Percent Open Area (POA), εA, that is defined as the percentage of the
open area, Ao, to the total area At, is given by
Ao
εA = × 100 .
At
Oi = (4 Ai / π )1/ 2 .
This diameter is greater than the side dimension of a square opening. A spherical par-
ticle with that diameter will never pass the opening (Fig. 2a) and may therefore not be con-
sidered as an equivalent dimension or equivalent diameter. This will only be possible if the
diameter corresponds to the side of the square area (Fig. 2b). However, not all openings are
squares, yet the equivalent square area of openings is used to determine their equivalent di-
mension because this simplified assumption results in one single opening size from the open
area. It is the diameter of a spherical particle that can pass the equivalent square opening,
hence the equivalent opening or pore size, Oi, results from
Oi = ( Ai )1/ 2 .
From the EOSs, Pore Size Distribution (PSD) and an equivalent diameter for which a
certain percentage of the opening have a smaller diameter (Ox, pore opening size that x per-
cent of pores are smaller than that size) may be measured.
The PSD curves can be used to determine the uniformity coefficient, Cu, of the inves-
tigated materials. The uniformity coefficient is a measure for the uniformity of the openings
and is given by
Cu = O60 O10 .
378 M. Ziabari, V. Mottaghitalab and A.K. Haghi
The ratio equals unity for uniform openings and increases with decreasing uniformity
of the openings [7].
Pore characteristic is one of the main tools for evaluating the performance of any non-
woven fabric and for electrospun webs as well. Understanding the link between processing
parameters and pore structure parameters will allow for better control over the properties
of electrospun fibers. Therefore there is a need for the design of nanofibers to meet specific
application needs. Various techniques may be used to evaluate pore characteristics of po-
rous materials including sieving techniques (dry, wet and hydrodynamic sieving), mercury
porosimetry and flow porosimetry (bubble point method) [8, 9]. As one goes about selecting
a suitable technique for characterization, the associated virtues and pitfalls of each tech-
nique should be examined. The most attractive option is a single technique which is non-
destructive, yet capable of providing a comprehensive set of data [10].
pore sizes and shapes deviate from the real ones. The method is suitable for measuring pore
sizes of woven geotextiles with fairly uniform pore sizes and shapes and is not appropriate
for electrospun nanofiber webs of different pore sizes.
In the later case, cross sectional image of nonwoven geotextile was used to calculate
the pore structure parameters. A slicing algorithm based on a series of morphological op-
erations for determining the mean fiber thickness and the optimal position of the uniform
slicing grid was developed. After recognition of the fibers and pores in the slice, the pore
opening size distribution of the cross sectional image may be determined. The method is
useful for measuring pore characteristics of relatively thick nonwovens and cannot be ap-
plied to electrospun nanofiber webs due to extremely small size.
Therefore, there is a need for developing an algorithm suitable for measuring the pore
structure parameters in electrospun webs. In response to this need, we have developed a new
image analysis based method and presented in the following.
In this method, a binary image of the web is used as an input. First of all, voids con-
nected to the image border are identified and cleared using morphological reconstruction
[13], where mask image is the input image and marker image is zero everywhere except
along the border. Total area which is the number of pixels in the image is measured. Then
the pores are labeled and each considered as an object. Here the number of pores may be
obtained. In the next step, the number of pixels of each object as the area of that object is
measured. Having the area of pores, the porosity and EOS regarding to each pore may be
calculated. The data in pixels may then be converted to nm. Finally PSD curve is plotted
and O50, O95 and Cu are determined.
a) b)
c)
using local thresholding. Note that, since the process is extremely sensitive to noise con-
tained in the image, before the segmentation, a procedure to clean the noise and enhance
the contrast of the image is necessary.
α
d
O
No.1 No.2
No.3 No.4
No.5
One of the most important features of simulation is that it allows several structural
characteristics to be taken into consideration with the simulation parameters. These param-
eters are: web density (controlled as line density), angular density (sampled from a normal
or random distribution), distance from the reference point (sampled from a random distri-
bution), line thickness (sampled from a normal distribution) and image size.
2 Experimental
Nanofiber webs were obtained from electrospinning of PVA with average molecular weight
of 72000 g/mol (Merck) at different processing parameters for attaining different pore
characteristics. Table 1 summarizes the electrospinning parameters used for preparing the
webs. The micrographs of the webs were obtained using a Philips XL-30 environmental
scanning electron microscope under magnification of 10000× after being gold coated. Fig-
ure 5 shows micrographs of the electrospun webs.
Table 1 Electrospinning parameters used for preparing nanofiber webs.
1 8 15 20 0.4
2 12 20 15 0.2
3 8 15 20 0.2
4 8 10 15 0.3
5 10 10 15 0.2
Image Analysis of Pore Size Distribution in Electrospun Nanofiber Webs 383
the porosity, O50 and O95 decrease. Cu also decreases with respect to porosity, that is to say
increasing the uniformity of the pores. The number of pores has an increasing trend with
decreasing porosity.
The image analysis method presents valuable and comprehensive information regard-
ing to pore structure parameters in nanofiber webs. This information may be exploited in
384 M. Ziabari, V. Mottaghitalab and A.K. Haghi
preparing the webs with needed pore characteristics to use in filtration, biomedical appli-
cations, nanoparticle deposition and other purposes. The advantages of the method are list-
ed below:
1) The method is capable of measuring pore structure parameters in any nanofiber webs
with any pore features and it is applicable even when other methods may not be
employed.
2) It is so fast. It takes less than a second for an image to be analyzed (with a 3 GHz
processor).
3) The method is direct and so simple. Pore characteristics are measured from the area
of the pores which is defined as the number of pixels of the pores.
4) There is no systematic error in measurement (such as assuming pores to be cylindrical
in mercury and flow porosimetry and the errors associated with the sieving methods
which were mentioned). Once the segmentation is successful, the pore sizes will be
measured accurately. The quality of images affects the segmentation procedure.
High-quality images reduce the possibility of poor segmentation and enhance the
accuracy of the results.
5) It gives a complete PSD curve.
6) There is no cost involved in the method and minimal technical equipments are needed
(SEM for obtaining the micrographs of the samples and a computer for analysis).
7) It has the capability of being used as an on-line quality control technique for large
scale production.
8) The results obtained by image analysis are reproducible.
9) It is not a destructive method. A very small amount of sample is required for
measurement.
1 0–360 20 5
2 0–360 30 5
3 0–360 40 5
4 0–360 20 10
5 0–360 30 10
6 0–360 40 10
7 0–360 20 20
8 0–360 30 20
9 0–360 40 20
Image Analysis of Pore Size Distribution in Electrospun Nanofiber Webs 385
Mean Standard
1 0–360 30 15 0
2 0–360 30 15 4
3 0–360 30 15 8
4 0–360 30 15 10
No.1 No.2
No.3 No.4
No.5 No.6
No.7 No.8
No.9
No.1 No.2
No.3 No.4
Pore structure parameters of simulated webs were measured using the image analysis
method. Table 5 summarizes the pore characteristics of the simulated images in the first set.
For webs with the same density, increasing the fiber diameter resulted in a decrease in O95,
number of pores and porosity. No particular trends were observed for O50 and Cu. Figures
9 and 10 show the PSD curves of the simulated images in the first set. As the web density
increases, the effects of fiber diameter are less pronounced since the PSD curves of the webs
become closer to each other.For webs with the same fiber diameter, the density increase re-
sulted in a decrease in O50, O95, Cu and porosity whereas the number of pores increased
with the density.
Table 5 Pore characteristics of the first set of simulated images.
a)
b)
c)
Figure 9 PSD curves of the first set of simulated images; effect of density, images with the diameter
of a) 5, b) 10, c) 20 pixels.
Table 6 summarizes the pore characteristics of the simulated images in the second set
set. No significant effects for variation of fiber diameter on pore characteristics were
observed. Suggesting that average fiber diameter is determining factor not variation of
388 M. Ziabari, V. Mottaghitalab and A.K. Haghi
a)
b)
c)
Figure 10 PSD curves of the first set of simulated images; effect of fiber diameter, images with the
density of a) 20, b) 30, c) 40 lines.
diameter. Figure 11 shows the PSD curves of the simulated images in the second set.
4 Conclusion
The evaluation of electrospun nanofiber pore structure parameters is necessary as it facili-
tates the improvement of the design process and its eventual applications. Various
Image Analysis of Pore Size Distribution in Electrospun Nanofiber Webs 389
Figure 11 PSD curves of the second set of simulated images, the effect of fiber diameter variation.
techniques have been developed to assess pore characteristics in porous materials. Howev-
er, most of these methods are indirect, have inherent problems and are not applicable for
measuring pore structure parameters of electrospun webs. In this investigation, we have
successfully developed an image analysis-based method as a response to this need. The
method is simple, comprehensive and fast and directly measures the pore structure param-
eters.
The effects of web density, fiber diameter and its variation on pore characteristics of
the webs were also explored using some simulated images. As the fiber diameter increased,
O95, the number of pores and porosity decreased. No particular trends were observed for
O50 and Cu. Increasing the density resulted in a decrease in O50, O95, Cu and porosity
whereas the number of pores increased with density. The effects of variation of the fiber
diameter on pore characteristics were insignificant.
References
1. A.K. Haghi and M. Akbari, Phys. Stat. Solidi (a), 204, 1830–1834 (2007).
2. D.H. Reneker and I. Chun, Nanotechnology, 7, 216–223 (1996).
3. D.R. Salem, Structure Formation in Polymeric Fibers, Hanser, Cincinnati, Chapter 6, H. Fong
and D.H. Reneker, Electrospinning and the Formation of Nanofibers (2001).
4. Th. Subbiah, G.S. Bhat, R.W. Tock, S. Parameswaran and S.S. Ramkumar, J. Appl. Polymer Sci.,
96, 557–569 (2005).
5. A. Frenot and I.S. Chronakis, Curr. Op. Coll. Interface Sci., 8, 64–75 (2003).
6. Ch.L. Casper, J.S. Stephens, N.G. Tassi, D.B. Chase and J.F. Rabolt, Macromolecules, 37,
573–578 (2004).
7. W. Dierickx, Geotext. Geomembr., 17 (4), 231–245 (1999).
8. S.K. Bhatia and J.L. Smith, Geosynthetics Int., 3 (2), 155–180 (1996).
9. S.K. Bhatia, J.L. Smith and B.R. Christopher, Geosynthetics Int., 3 (3), 301–328 (1996).
10. S.T. Ho and D.W. Hutmacher, Biomaterials, 27, 1362–1376 (2006).
11. A.H. Aydilek and T.B. Edil, Geotechn. Testing J., 27 (1), 1–12 (2004).
12. A.H. Aydilek, S.H. Oguz and T.B. Edil, J. Comput. Civil Eng., 280–290 (2002).
13. R.C. Gonzalez and R.E. Woods, Digital Image Processing, Prentice Hall, New Jersey, second
390 M. Ziabari, V. Mottaghitalab and A.K. Haghi
edn (2001).
14. B. Jähne, Digital Image Processing, Springer, England, 5th revised and extended edn (2002).
15. H.S. Kim and B. Pourdeyhimi, Int. Nonwoven J., 15–19 (Winter 2000).
16. M.S. Abdel-Ghani and G.A. Davis, Chem. Eng. Sci., 117 (1985).
17. B. Pourdeyhimi, R. Ramanathan and R. Dent, Textile Res. J., 66 (11), 713–722 (1996).
AUTHORS:
- Figure 13 still not mentioned in the text
- a list of keywords is missing
- two Figures 22 (the second Fig. 22 -- > Fig. 23?)
- check the items highlighted in red; approve the complete paper
14 Electrospun Biodegradable
and Biocompatible Natural
Nanofibers: A Detailed
Review
1 Introduction
An emerging technology of manufacturing of thin natural fibers is based on the principle
of the electrospinning process. In conventional fiber spinning, the mechanical force is ap-
plied to the end of a jet, whereas in the electrospinning process the electric body forces act
on an element of charged fluid. Electrospinning has emerged as a specialized processing
technique for the formation of submicron fibers (typically between 100 nm and 1 µm in di-
ameter), with high specific surface areas. Due to their high specific surface area, high po-
rosity and small pore size, the unique fibers have been suggested for a wide range of
applications.
Electrospinning of natural fibers offers unique capabilities for producing novel natural
nanofibers and fabrics with controllable pore structure. Current research effort has focused
on understanding the electrospinning of natural fibers in which the influence of different
governing parameters are discussed.
Electrospinning [1–10] is an economical and simple method used in the preparation of
polymer fibers. The fibers prepared by this method typically have diameters much smaller
than is possible to attain using standard mechanical fiber-spinning technologies [11].
Electrospinning of polymer solutions has gained much attention in the last few years as a
392 A.K. Haghi and R.K. Haghi
cheap and straightforward method to produce nanofibers [12 – 16]. Electrospinning differs
from the traditional wet/dry fiber spinning in a number of ways, of which the most striking
differences are the origin of the pulling force and the final fiber diameters. The mechanical
pulling forces in the traditional industrial fiber spinning processes lead to fibers in the mi-
crometer range and are contrasted in electrospinning by electrical pulling forces that enable
the production of nanofibers [14]. Depending on the solution properties, the throughput of
single-jet electrospinning systems ranges around 10 ml/min. This low fluid throughput may
limit the industrial use of electrospinning. A stable cone-jet mode followed by the onset of
the characteristic bending instability, which eventually leads to a great reduction in the jet
diameter, necessitate the low flow rate [6, 17]. When the diameters of polymer fiber mate-
rials are shrunk from micrometers (e.g. 10 – 100 µm) to submicrons or nanometers, there ap-
pear several amazing characteristics such as a very large surface area to volume ratio (this
ratio for a nanofiber can be as large as 103 times of that of a microfiber), flexibility in sur-
face functionalities, and superior mechanical performance (e.g. stiffness and tensile
strength) compared with any other known form of material. These outstanding properties
make the polymer nanofibers optimal candidates for many important applications [18].
These include filter media [19–21], composite materials [18, 22], biomedical applications
(tissue engineering scaffolds [23–27], bandages [28], drug release systems [29–30]), pro-
tective clothing for the military [31–34], optoelectronic devices and semiconductor mate-
rials [35–38], biosensors/chemosensors [39–42].
contraction of the surface charges to the counter electrode cause a force directly opposite
to the surface tension [46]. As the intensity of the electric field is increased, the hemispher-
ical surface of the fluid at the tip of the capillary tube elongates to form a conical shape
known as the Taylor cone [47]. Further increasing the electric field, a critical value is at-
tained with which the repulsive electrostatic force overcomes the surface tension and the
charged jet of the fluid is ejected from the tip of the Taylor cone [18, 48]. The jet exhibits
bending instabilities due to repulsive forces between the charges carried with the jet. The
jet extends through spiraling loops, as the loops increase in diameter the jet grows longer
and thinner until it solidifies or collects on the target [49].
2.1.1 Viscosity
The viscosity range of different polymer solutions which are spinnable is different. One of
the most significant parameters influencing the fiber diameter is the solution viscosity. A
higher viscosity results in a large fiber diameter. Figure 2 shows the representative images
of beads and beaded fibers for solutions with a range of viscosity. Beads and beaded fibers
are less likely to be formed for more viscous solutions. The diameter of beads becomes big-
ger and the average distance between beads on fibers longer as viscosity increases. Mean-
while, the shape of beads gradually changes from spherical to spindle-like [51].
20
Figure 3 Solution conductivity effects on the diameter of the electrospun P(LLA-CL) (70/30
wt. %) fibers [50].
solution concentration or the corresponding viscosity is too high due to the difficulty in liq-
uid jet formation [56–58].
Figure 4 shows the morphology of fibers obtained from solutions of 6, 8, 10 and 12
wt. % concentration at applied voltage 10 kV with a constant tip-to-collector distance of
10 cm. Spindle-like beads formed for fibers obtained from solution of 6 wt. %.
Electrospun Biodegradable and Biocompatible Natural Nanofibers 395
a b c d
Figure 4 The morphology of nanofibers at concentrations from 6% to 12% with a constant tip-to-
collector distance of 10 cm at applied voltage 10 kV. (The values below the images and the brackets
show the average fiber diameter (nm) and the standard deviation of fiber diameter. The scale bars for
fibers from solution from 6 to 10 wt.% and 12 wt.% are 2 mm and 5 mm, respectively.). above: Dis-
tribution of fiber diameters obtained at 10 kV with a tip-to-collector distance of 10 cm. The concen-
trations for (a), (b), (c) and (d) are 6, 8, 10 and 12 wt. %, respectively [55].
Molecular weight also has a significant effect on the rheological and electrical properties
such as viscosity, surface tension, conductivity and dielectric strength. It has been reported
that too low molecular weight polymers tend to form beads rather than fibers and high mo-
lecular weight polymers yield fibers with a larger average diameter [59].
Note that in Figure 5, as the molecular weight increases, several changes can be ob-
served in the electrospun structure of PVA. At a molecular weight of 9000–13,000 g/mol,
Figure 5 Photographs showing the typical structure in the electrospun polymer for various molec-
ular weights. The samples were obtained at the center of the deposition area. (a) 9000–10,000 g/mol;
(b) 13,000–23,000 g/mol; and (c) 31,000–50,000 g/mol (solution concentration: 25 wt. %) [58].
396 A.K. Haghi and R.K. Haghi
Figure 6 Solution viscosity and surface tension as a function of polymer solution concentration
[60].
the fibrous structure is not completely stabilized and a bead-on-string structure is obtained,
indicating the resistance of the jet to extensional flow.
Surface tension seems more likely to be a function of solvent compositions, but is negligibly
dependent on the solution concentration. Different solvents may contribute different sur-
face tensions. However, not necessarily a lower surface tension of a solvent will always be
more suitable for electrospinning. Generally, surface tension determines the upper and low-
er boundaries of electrospinning window if all other variables are held constant. Formation
of droplets, beads and fibers can be driven by the surface tension of the solution and lower
surface tension of the spinning solution helps electrospinning to occur at a lower electric
field [57].
The above variables are not independent of each other. Solution viscosity is a function of
both concentration and polymer molecular weight. These two parameters also affect the
number of entanglements [52].
Figure 6 shows the viscosity and surface tension of solution dissolved in formic acid;
they were 400–5550 cps and 49–68 mN/m, respectively. These values increased with in-
creasing of concentration. Specifically, the viscosity increased from 1400 to 5100 cps for
solution concentration from 25 to 30 wt. % [60].
Figure 7 Effect of NaCl amount in the PVA solution on fiber morphology (DH = 98%, voltage =
5 kV, tip–target distance = 10 cm; flow rate = 0.2 ml/h). NaCl amount based on H2O: (a) 0.05%, (b)
0.10%, (c) 0.15%, (d) 0.2%. Original magnification 10,000 [55].
expected to be significant. As the charges carried by the jet increased, higher elongation
forces were imposed to the jet under the electrical field, resulting in smaller beads and thin-
ner fiber diameters (Fig. 8).
A comparison of the diameter of the electrospun fibers with the electrical conductivity
of the solutions is shown in Fig. 9. There was a significant drop in the diameter of the elec-
trospun polymer fibers when the electrical conductivity of the solution increased. Beads
were also observed due to low conductivity of the solution, which results in insufficient
elongation of a jet by electrical force to produce uniform fiber.
Electrospun polymer nanofibers with the smallest fiber diameter were obtained with
the highest electrical conductivity. This suggests that the drop in the size of the fibers was
due to the increased electrical conductivity [50].
Figure 10 SEM images of PAA nanofibers with different concentrations of NaCl [61].
observed from all prepared PAA nanofibers. With increasing the concentration of NaCl, the
relative viscosity slightly decreased. Variations in viscosity may be caused by the chain
conformation change of PAA in solution upon adding NaCl. With increasing the ionic
strength of the solution, the chain conformation of PAA changes from the extended linear
conformation to the coil conformation because PAA is a polyelectrolyte. Although relative
viscosity decreases slightly with increasing the amount of NaCl [61].
The effect of NaCl addition on the morphology of electrospun PVA fibers was shown
in Fig. 11. Even a little sodium chloride added in the solution made its conductivity to in-
crease sharply. Net charge density carried by the jet in the electrospinning process can be
Electrospun Biodegradable and Biocompatible Natural Nanofibers 399
Figure 11 Effect of NaCl amount in the PVA solution on fiber morphology (DH = 98%, voltage =
5 kV, tip–target distance = 10 cm; flow rate = 0.2 ml/h). NaCl amount based on H2O: (a) 0.05%, (b)
0.10%, (c) 0.15%, (d) 0.2%. Original magnification 10,000 [56].
affected by the conductivity of the solution. PVA fiber diameters were gradually decreased
from 214±19 nm to 159±21 nm with increasing content of NaCl from 0.05% to 0.2%, for
the higher net charge density increased the electrical force exerted on the jet and led to de-
creased fiber diameter. When 0.2% NaCl was added in PVA solution, many small particles,
perhaps sodium chloride crystals, were observed on the SEM micrograph (Fig. 11d).
2.1.8 Solvent
The effects of solvents and their properties on electrospinnability of as-prepared polysty-
rene (PS) solutions and the morphological appearance of as-spun PS fibers were investigat-
ed qualitatively by means of a scanning electron microscope (SEM) in [62]. The eighteen
solvents used were benzene, t-butyl acetate, carbon tetrachloride, chlorobenzene, chloro-
form, cyclohexane, decahydronaphthalene (decalin), 1,2-dichloroethane, dimethylforma-
mide (DMF), 1,4-dioxane, ethylacetate, ethylbenzene, hexane, methylethylketone (MEK),
nitrobenzene, tetrahydrofuran (THF), 1,2,3,4-tetrahydronaphthalene (tetralin) and toluene.
DMF was the best solvent to dissolve PS pellets within six hours, while others were found
to dissolve PS pellets within one to three days, with the exception of hexane which was not
able to dissolve PS pellets after seven days. Only PS solutions in 1,2-dichloroethane, DMF,
ethylacetate, MEK, and THF could produce fibers with high enough productivity, while PS
solutions in benzene, cyclohexane, decalin, ethylbenzene, nitrobenzene, and tetralin were
not spinnable [62].
Figure 12 Effect of voltage on morphology and fiber diameter distribution from a 7.4% PVA/water
solution (DH = 98%, tip–target distance = 15 cm, flow rate = 0.2 ml/h). Voltage: (a) 5 kV, (b) 8 kV,
(c) 10 kV, (d) 13 kV. Original magnification 10,000 [56].
capillary tip to the grounded target when all other variables (conductivity, dielectric con-
stant and flow rate of solution to the capillary tip) are held constant [16].
With the increase of the electrical potential the resulting nanofibers became rougher.
It was already reported that a diameter of electrospun fibers was not significantly affected
by an applied voltage. This voltage effect was particularly diminished when the polymer
concentration was low. According to earlier works, higher voltage was reported to induce
not only a larger diameter but also a smaller diameter. Applied voltage may affect some fac-
tors such as the mass of polymer fed out from a tip of needle, elongation level of a jet by
an electrical force, morphology of a jet (a single jet or multiple jets), etc. A balance among
these factors may determine a final diameter of electrospun fibers. It is also noted that bead-
Electrospun Biodegradable and Biocompatible Natural Nanofibers 401
Figure 13 Main effect plots of applied voltage on average fiber diameter [55].
ed fibers have been found to be electrospun with too high a level of applied voltage. Al-
though voltage effects show different tendencies, the voltage did not show a significant role
in controlling the fiber morphology [59].
A series of experiments were carried out when the applied voltage was varied from 5
to 13 kV and the tip to target distance was held at 15 cm. The results are shown in Fig. 12.
There was a slightly increase in average fiber diameter with increasing applied electric field.
A considerable amount of thin fibers with diameters below 150 nm were found when the
applied voltage was above 10 kV. A narrow distribution of fiber diameters was observed at
a lower voltage of 5 kV, while a broad distribution in the fiber diameter was obtained at
higher applied voltages of 10 – 13 kV. Increasing the applied voltage, i.e., increasing the
electric field strength will increase the electrostatic repulsive force on the fluid jet which
favors the thinner fiber formation. On the other hand, the solution will be removed from the
capillary tip more quickly as jet is ejected from the Taylor cone. This results in the increase
402 A.K. Haghi and R.K. Haghi
of the fiber diameter. Corona discharge was observed at voltages above 13 kV, making elec-
trospinning impossible [56]. Also, an increase in voltage from 9.2 kV to 25 kV did not in-
duce a significant change in the mean diameter of the fibers [54]. The diameter of
electrospun PAN fibers did not change significantly over the range of applied voltage for
the various solution concentrations in the experimental region [56].
SEM micrographs of nanofibers electrospun at different electrospinning voltage from
a constant polymer concentration of 5 wt. % are shown in Fig. 14a. A fiber diameter tended
to decrease with increasing electrospinning voltage, although the influence was not as great
as that of polymer concentration. Figure 14b shows fiber diameter as a function of electro-
spinning voltage. The increase of the electrospinning voltage causes an increase of the elec-
trostatic stress on the jet, which may be analogous to an increase of the draw rate in
conventional fiber spinning [64].
Figure 15 Effect of flow rate of 7% PVA water solution on fiber morphology (DH = 98%, voltage
= 8 kV, tip–target distance = 15 cm). Flow rate: (a) 0.1 ml/h; (b) 0.2 ml/h; (c) 0.3 ml/h. Original mag-
nification 10,000 [56].
Electrospun Biodegradable and Biocompatible Natural Nanofibers 403
Figure 16 Volume feed rate effects on the diameter of the PLLA (Mw: 300 K) fibers electrospun
from solutions with different polymer concentration [50].
control the morphology of electrospun polymer fibers [9]. In the case of low-molecu-
lar-weight liquid, when a high electrical force is applied, formation of droplets can occur.
A theory proposed by Rayleigh explains this phenomenon. As evaporation of a droplet
takes places, the droplet decreases in size. Therefore, the charge density of its surface is in-
creased. This increase in charge density due to Coulomb repulsion overcomes the surface
tension of the droplet and causes the droplet to split into smaller droplets. However, in the
case of a polymer solution (high molecular weight liquid) the emerging jet does not break
up into droplets, but is stabilized and forms a string of beads connected by a fiber. As the
concentration is increased, a string of connected beads is seen, and with a further increase
there is reduced bead formation until only smooth fibers are formed. And sometimes spin-
dle-like beads can form due to the extension caused by the electrostatic stress. The changing
of fiber morphology can probably be attributed to a competition between surface tension
and viscosity. As concentration was increased, the viscosity of the polymer solution in-
creased. The surface tension attempted to reduce the surface area per unit mass, thereby
caused the formation of beads/spheres. Viscoelastic forces resisted the formation of beads
and allowed for the formation of smooth fibers. Therefore formation of beads at lower poly-
mer solution concentration (low viscosity) occurred where surface tension had a greater in-
fluence than the viscoelastic force. However, bead formation was reduced and finally
eliminated at higher polymer solution concentration, where viscoelastic forces had a greater
influence in comparison with surface tension. But when the concentration was too high,
high viscosity and rapid evaporation of solvent made the extension of jet more difficult,
thicker and nonuniform fibers were formed [55].
Figure 17 Effect of tip–target distance on fiber morphology from a 7.4% PVA / water solution (DH
= 98%, voltage = 5 kV, flow rate = 0.2 ml/h). Tip–target distance: (a) 8 cm, (b) 10 cm, (c) 12 cm, (d)
15 cm. Original magnification 10,000 [56].
Figure 18 Processing map obtained based on the systematic parameter study: (a) jet elongation / an
electrical force (affected by electrical conductivity of solvents, applied voltage), (b) mass of polymer
(affected by polymer concentration, applied voltage, volume feed rate) [50].
ning at 8 –15 cm of the tip–target distance. It was assumed that solution jets were elongated
and solidified quickly after they flowed out of the needle tip because of the high conduc-
tivity of fully hydrolyzed PVA used [56].
Based on the processing parameter studies, all the parameters’ effects on the morphol-
ogy of the electrospun nanofibers were summarized in a processing map (Fig. 18). A suit-
able level of processing parameters must be optimized to electrospin polymers into
Electrospun Biodegradable and Biocompatible Natural Nanofibers 405
nanofibers with a desired morphology and the parameters’ levels are dependent on the prop-
erties of polymers and solvents used in each electrospinning process. Understanding of how
each of processing parameter affects the morphology of the electrospun nanofibers is es-
sential. All the parameters have been divided into two groups; i.e. one with parameters
which affect the mass of polymer fed out from a tip of needle, and the other with parameters
which affect an electrical force during electrospinning. Polymer concentration, applied
voltage and volume feed rate were considered to affect the mass of polymer. Polymer con-
centration and feed rate directly reflect the mass of polymer. Increased polymer concentra-
tion and feed rate tend to bring more mass of polymer into the polymer jet. It is noteworthy
that the minimum polymer concentration to electrospin uniform fibers was determined by
the molecular weight of polymer. High molecular weight of polymer provides a sufficient
level of solution viscosity to produce a uniform jet during electrospinning even when poly-
mer concentration is relatively low. Applied voltage reflects the force to pull a solution out
from the needle hence higher applied voltage causes more solution coming out. On the other
hand, it was considered that solution electrical conductivity and applied voltage affect a
charge density thus an electrical force, which acts to elongate a jet during electrospinning.
Hence, higher solution electrical conductivity and applied voltage increase the jet elonga-
tion. Therefore, it is summarized that electrospun fibers with a smaller diameter can be pro-
duced with lower polymer concentration, feed rate and applied voltage when the effect of
the mass of polymer dominates to determine the final diameter of electrospun fibers, while
a smaller diameter of fibers can be electrospun with higher solution electrical conductivity
and applied voltage when the effect of the jet elongation is dominant. For both cases,
non-uniform / beaded fibers were found if the parameters were either too high or too low.
In fact, applied voltage affects both the polymer mass and jet elongation; however, the ef-
fect is not as dominant as the other parameters for controlling the morphology of electro-
spun fibers. It must be noted that polymer concentration, molecular weight and solution
electrical conductivity play a prime role in determining the morphology of electrospun fi-
bers. Polymer fibers with a smaller diameter can be electrospun with higher electrical con-
ductivity of solution and lower polymer concentration which can be further decreased by
higher molecular weight of polymer [50].
instability. Reneker et al. modeled the polymer jet by a linear Maxwell equation.
Like-charge repulsion generates a bending force that destabilizes the jet. Hohman et al. [67]
built an electrohydrodynamic instability theory, and predicted that under favorable condi-
tions, a nonaxisymmetric instability prevails over the familiar Rayleigh instability and a
varicose instability due to electric charges. In theoretical work to date, the rheology of the
polymer jet has been represented by a Newtonian viscosity [67, 68], a power-law viscosity
[66] and the linear Maxwell equation [9, 69, 71].
The jet is governed by four steady-state equations representing the conservation of
mass and electric charges, the linear momentum balance and Coulomb’s law for the field
E [9]. Mass conservation requires that
π R2υ = Q, (1)
where E is the z component of the electric field, K is the conductivity of the liquid, and I is
the constant total current in the jet. The momentum equation is formulated by considering
the forces on a short segment of the jet (Fig. 19):
d d γ
(π R 2 ρυ 2 ) = π R 2 ρ g + [π R (− P + τ zz )] + ⋅ 2π RR ′ + 2π R (tte − tne R ′) , (3)
dz dz R
where τzz is the axial viscous normal stress, p is the pressure, γ is the surface tension, and
t et and t en are the tangential and normal tractions on the surface of the jet due to electricity.
The prime indicates the derivative with respect to z, and R´ is the slope of the jet surface.
The ambient pressure has been set to zero. The electrostatic tractions are determined by the
surface charge density and the electric field:
ε σ 2 ε′−ε 2
tne = ( En2 − Et2 ) ≈ − E , (4)
2 2ε 2
tte = σ Et ≈ σ E , (5)
where ε and ε are the dielectric constants of the jet and the ambient air, respectively, En
Electrospun Biodegradable and Biocompatible Natural Nanofibers 407
and Et are the normal and tangential components of the electric field at the surface, and ||*||
indicates the jump of a quantity across the surface of the jet. We have used the jump con-
ditions for En and Et: ||ε En|| = ε E – εEn = σ, ||Et|| = E t – Et = 0, and assumed that εEn <<
(see Ganan–Calvo [13]) and Et ≈ E. The overbar indicates quantities in the surrounding air.
The pressure p(z) is determined by the radial momentum balance, and applying the normal
force balance at the jet surface leads to
γ
− p + τ rr = tne − , (6)
R
3 d γ R′ σσ ′ 2σ E
ρυυ ′ = ρ g + (η R 2υ ′) + + + (ε − ε ) EE ′ + . (7)
R 2 dz
R 2 ε ′ R
4 Natural fibers
The fibers in modern textile manufacture can be classified into two groups: natural and
man-made fibers. A vast classification of fibers is illustrated in Fig. 20. Natural fibers are
those provided by nature in a ready-made fibrous form. Natural fibers can be subdivided
into three main classes, according to the nature of their source:
• vegetable fibers
• animal fibers
• mineral fibers.
Vegetable fibers include cotton – the most important of all textile fibers – together with
flax, hemp, jute and other fibers, which are produced by plants. They are based on cellulose,
the material used by nature as a structural material in the plant world.
408 A.K. Haghi and R.K. Haghi
Animal fibers include wool and other hair-like fibers, and fibers such as silk, produced
as filaments by cocoon-spinning creatures. These animal fibers are based on proteins, the
complex substances from which much of the animal body is made [72].
5.1 Introduction
The silk fiber protein is synthesized by the silk gland cells and stored in the lumen of silk
gland. Subsequently, it is converted into silk fibers. Each strand of silk fiber is a double
structure with two fibroin filaments covered by sericin (Fig. 21).
In fibroin, alanine and glycine together account for 70% of the total composition,
whereas in sericin they make up about 15%. The chief component of sericin is another ami-
no acid, serine (30% of the total).
Silk is the only natural fibre which exists as a continuous filament. Each Bombyx mori
cocoon can yield up to 1600 meters of filament. These can be easily joined together using
the adhesive qualities of sericin to form a theoretically endless filament.
The silk fiber’s triangular cross-section gives it excellent light reflection capability.
The silk fiber is smooth, unlike those of wool, cotton and others. This is one of the rea-
sons why silk fabrics are so lustrous and soft.
Silk can absorb moisture up to 30% of its weight without creating a damp feeling.
When moisture is absorbed, it generates ‘wetting-heat’ which helps to explain why silk is
comfortable to wear next to the skin.
Silk has a tenacity of approximately 4.8 grams per denier, slightly less than that of ny-
lon.
Silk has poor resistance to ultraviolet light and for this reason is only recommended
for those curtains that are lined or not exposed to direct sunlight [73].
Figure 22 ATR-IR spectra of as-prepared and methanol-treated SF matrices: (a) as-spun SF-N; (b)
methanol-treated SF-N; (c) as-cast SF-F; (d) SF-M [75].
and each is interchangeable under certain conditions. Crystallization of SF involves the con-
formational transition that can be easily induced by simple physical (thermal), mechanical
or chemical treatments. The most common method to convert the distorted conformation
(random coil or silk I) of SF into the more stable β-sheet (silk II) conformation is to treat
the SF film with an organic solvent. It is well known that organic solvents, particularly
methanol, are highly effective in crystallizing SF from a distorted conformation to the
α-sheet.
Infrared spectroscopy (IR) has been often applied to study the molecular conformation
of silk fibroin fibers or films. The sensitive absorption bands on the IR spectrum are located
in the spectral regions of ~1625 cm–1 (amide I), ~1528 cm–1 (amide II), ~1230 cm–1 (amide
III) and ~700 cm–1 (amide V). To characterize the structure of SF matrices, the ATR-IR
spectra in the amide I and II regions are examined. The IR spectra of SF matrices are shown
in Fig. 22(a)–(c). The as-spun SF-N matrix was characterized by absorption bands at 1651
cm–1 (amide I) and 1528 cm–1 (amide II), attributed to the random coil conformation, as
shown in Fig. 22a. The methanol-treated SF-N matrix showed strong β-sheet absorptions
at 1622 and 1514 cm–1 within a methanol-treating time of 10 min, indicating that the ran-
dom coil conformation of the SF nanofibers rapidly converted into β-sheet structure
(Fig. 22(b)) [75].
On the contrary, both as-prepared SF-F and SF-M matrices showed strong β-sheet ab-
sorptions at 1622 and 1514 cm–1 without methanol treatment, as shown in Fig. 22(c) and
(d). The raw SF had highly ordered intermolecular hydrogen bonds (β-sheet), and the SF
film took a mainly β-sheet conformation, when cast from the formic acid solution.
Solid-state 13C NMR has been shown to be a more effective analytical tool for dem-
onstrating the formation of β-sheets in polypeptides and proteins, because the isotropic 13C
NMR chemical shifts of carbon atoms in proteins are sensitive to the β-sheet’s secondary
structure. It is well established that SF conformations are dependent upon the species of
silkworms and conditions of sample preparation. In particular, it has been reported that fi-
broin from Bombyx mori adopts two dimorphic structures, silk I and silk II. The silk II form
is identified by the 13C chemical shifts of glycine (Gly), serine (Ser), and alanine (Ala) that
are indicative of β-sheets, while the silk I form produces chemical shifts that are associated
with a loose helix or distorted β-turn. However, when compared with silk II, the less stable
silk I shows a relatively unresolved structure, and the conformation of the soluble form of
410 A.K. Haghi and R.K. Haghi
Figure 22 13C CP/MAS NMR spectra (a) and the expanded 13C NMR spectra of the methyl alanine
region (b) of SF matrices. The dotted arrow indicates the chemical shift of alanine in the β-sheet con-
formation [75].
of liquid fibroin (by losing water) inside the silk gland. One of them corresponds to the crys-
tal form of fibrous fibroin. The fibroin molecular chain which constitutes the spatial crystal
form is considered to be the intermediate form during the course of shifting the molecular
chain of liquid silk to the fibrous fibroin molecule chain, that is, β-chain. It is therefore
named the α-type of fibroin. The condition for the appearance of a type thereafter include
solidification when there is no stress from liquid silk and the temperature being less than
40°C. It becomes β-type when the temperature is above 50°C. Both α- and β-types of none
stretched are insoluble in water and the α-type is softer than the β-type [76].
• β-structure of silk
β-structure is the structure where the polypeptide chain is elongated. The structure can
be the type where all the molecular chains run in the same direction and form a parallel
pleated sheet or the type where the molecular chains run in the alternate direction and form
an antiparallel chain pleated sheet. In the case of a β-structure, there are three important fea-
tures, namely the period that is repeating period of the polypeptide chain, the spacing of the
molecular chain in the sheet and the distance between the sheets [76].
5.3.1 Degumming
The natural gum sericin is normally left on the silk during reeling, throwing and weaving.
It acts as a size, which protects the fiber from mechanical injury. The gum is removed from
finished yarns or fabrics, usually by boiling with soap and water [73]. The fibers were heat-
ed in an aqueous Na2CO3 (0.02 M) or 0.5% (w/w) NaHCO3 and rinsed with water to extract
sericin [75, 77].
Figure 23 The morphology of fibers at electric field of 3 kV/cm at concentrations from 5 to 19.5%
with a constant spinning distance of 7 cm. The figure also shows the average, standard deviation,
maximum and minimum values of the fiber diameter [77].
tion plate distance of 7 cm. At a 8% concentration, less than a 30 nm diameter fibers were
formed with beads (drops of polymer over the woven mesh); they were not uniform and
were branched off (Fig. 25). At a 10% concentration with 5 cm spinning distance and 2, 3
and 4 kV/cm electric fields, drops were formed instead of fibers. Continuous fibers were
Electrospun Biodegradable and Biocompatible Natural Nanofibers 413
Figure 25 The morphology of fibers at electric field of 4 kV/cm at concentrations from 5 to 19.5%
with a constant spinning distance of 7 cm [77].
obtained above 12% regardless of electric field and distance (Figs. 24 and 25). At 19.5%,
the average fiber diameter was much larger than that of fibers spun at lower concentrations.
In the short distance as well as low concentration (10%), the solution reaches the collection
plate before the solvent fully evaporates. This explains the formation of droplets and beads
at the low concentration and distance. Fewer beads were observed in electrospun fibers at
higher concentration. Increase in the regenerated silk concentration in the formic acid in-
creases the solution viscosity. At low concentrations beads are formed instead of fibers and
at high concentrations the formation of continuous fibers are prohibited because of the in-
ability to maintain the flow of the solution at the tip of the needle resulting in the formation
of larger fibers. Continuous nanofibers were obtained above 12% regardless of electric field
and distance and at higher concentration of 19.5% the average fiber diameter was larger
than at lower concentrations [77].
The SEM micrographs of nanofibers electrospun from SF solutions with different con-
centrations or viscosities ranged from 3% to 15% by weight are shown in Fig. 26. At a con-
centration below 9% by weight, beads or beaded fibers were generated by electrospinning.
Continuous nanofibers can be obtained at a concentration above 12% by weight, and this
414 A.K. Haghi and R.K. Haghi
Figure 26 SEM micrographs of electrospun SF nanofibers with concentration of (a) 3%, (b) 6%, (c)
9%, (d) 12%, or (e) 15% by weight [75].
Figure 27 The relationship between mean fiber diameter and electric field with concentration of
15% at spinning distances of 5, 7 and 10 cm (please refer to text for explanation) [77].
Electrospun Biodegradable and Biocompatible Natural Nanofibers 415
Figure 28 The relationship between the fiber diameter and the concentration at three electric fields
(2, 3, 4 kV/cm) [77].
Figure 28 shows that concentration apparently has a greater effect on the fiber diameter
than electric field. The multiple regression analysis was carried out to evaluate the contri-
bution of concentration and electric field on the fiber diameter [77].
Sukigara et al. [78] used RSM analysis (Response Surface Methodology) of the ex-
perimental results to develop a processing window which will produce nanoscale regener-
ated silk fibers by the electrospinning process. RSM is used in situations where several
variables influence a feature (called the response) of the system. The steps in the procedure
are described briefly as follows.
ζ i − [ζ Ai + ζ Bi ] / 2
xi = ,
[ζ Ai + ζ Bi ] / 2
where ζAi and ζBi refer to the high and low levels of the variables ζi; respectively.
3. Determination of the empirical model by multiple regression analysis to generate
theoretical responses (y): The second-order model is widely used in RSM. The general
equation for response η of the second-order model is given by:
k k k
η = βο + ∑ βi xi + ∑ βii xi2 + ∑ ∑ βij xi x j ,
i =1 i =1 i< j =2
416 A.K. Haghi and R.K. Haghi
Figure 29 RSM procedure to optimize the electrospinning condition of regenerated silk [78].
where k is the number of factors, xi are the coded variables and β are coefficients.
4. Calculation of the coefficients β to fit the experimental data as closely as possible.
The relationship between the response and the variables is visualized by a response
surface or contour plot to see the relative influence of the parameters, to find an optimum
parameter combination, and to predict experimental results for other parameters. For two
variables, when k = 2, the empirical model from the general equation (2) becomes
The RSM procedure to optimize the process parameters for the electrospinning silk is
shown in Fig. 29.
Sukigara et al. [78] designed a factorial experiment by using two factors (electric field
and concentration). For a quadratic model, experiments must be performed for at least three
levels of each factor. These levels are best chosen equally spaced. The two factors (silk con-
centration and electric field) and three levels resulted in nine possible combinations of fac-
tor settings. A schematic of the experimental design is shown in Fig. 30 (A) and (B).
5.5 Characterization
The conformational changes of the secondary structure of silk fibroin, which occur during
the electrospinning process, were analyzed by Raman spectroscopy. The spectra for pris-
tine, degummed and electrospun fibers are shown in Fig. 31 (A). Pristine and degummed
silk fibroins display characteristic conformational bands in the range 1650–1667 and
Electrospun Biodegradable and Biocompatible Natural Nanofibers 417
Figure 30 Experimental design (A) spinning distance 7 cm and (B) spinning distance 5 cm. The
values at the coordinate points show the mean fiber diameter (nm) of 100 measurements and coded
values are shown in brackets (electric field and concentration). NF: no fiber formation [78].
1241–1279 cm–1, which correspond to amide I and complex amide III, respectively. In this
study, the amide I (random coil) pristine band was observed at 1665 cm–1 and the amide III
(β-sheet) pristine band at 1231 cm–1. These well-defined bands were chosen because they
give a clear indication of changes in the secondary structures from random to β-sheet. The
degummed silk also shows absorption bands at these wavelengths. No significant spectral
changes were observed indicating that the fibroin conformation is unchanged during the de-
gumming process. The Raman spectra of the electrospun fiber are essentially the same as
that of the pristine and degummed fibers although minor bands and some differences in
peak intensities appear. This shows that the electrospinning process preserves the natural
conformation of fibroin. Figure 31 (B) shows that the amide I (1665 cm–1, random) to amide
III (1228 cm–1, β-sheet) ratio of the electrospun fiber is less than that of the pristine fiber.
This means that the electrospun fiber has a higher β-sheet content than the pristine fiber.
pristine
degummed
electrospun
Figure 31 (A) Raman spectroscopy of pristine, degummed and electrospun nonwoven silk mat. (B)
Secondary structural compositions of silk fibroin showing the fraction of amide I to amide III confor-
mations [79].
and cellulose xanthate (rayon) can be easily molded or drawn into fibers for textile appli-
cations, for designing composite materials (safety glass), as thermoplastics etc. [80].
Figure 32 SEM images of electrospun CTA fibers: (a) magnification ×1000, (b) magnification
×20,000, and (c) fractured cross-section of porous CTA fibers [81].
a porous CTA fiber. Isolated pores were also observed inside the fiber, which were larger
than those in the fiber surface. In liquid–liquid demixing of polymer solution, the resulting
polymer morphology is dependent on the phase separation mechanism. The structure
formed by the spinodal decomposition (SD) mechanism shows interconnected pores,
whereas the nucleation and growth (NG) mechanism mainly results in isolated spherical
pores [81].
Figure 33 shows SEM photographs of ultrafine CTA fibers electrospun from different
ratios of MC/ EtOH (90 / 10, 85 / 15, and 80 / 20 v/v). All ultrafine CTA fibers were elec-
trospun with the same electrospinning condition for a CTA solution in MC. It was found
that addition of EtOH changed not only the pore structure but also the fiber diameter. In the
case of MC/EtOH (90 / 10), the resulting CTA fibers had interconnected pores in the range
of 200–500 nm, indicating that the phase separation proceeded according to the SD mech-
anism. However, non-porous corrugated fibers were obtained from other mixed solvents
(Figs. 33(b) and (c)) [81].
Figure 33 SEM images of ultrafine CTA fibers electrospun using different volume ratios of
MC/EtOH: (a) 90 / 10, (b) 85 / 15, and (c) 80 / 20 [81].
7 Concluding remarks
Nanotechnology has become in recent years a topic of great interest to scientists and engi-
neers, and is now established as the prioritized research area in many countries. The reduc-
tion of size to the nanometer range brings an array of new possibilities in terms of material
properties, in particular with respect to achievable surface to volume ratios. Electrospinning
of natural fibers is a novel process for producing superfine fibers by forcing a solution
through a spinnerette with an electric field. A comprehensive review on this technique has
been made in this paper. Based on this review, many challenges exist in the electrospinning
process of natural fibers, and a number of fundamental questions remain open. The electro-
spinning technique provides an inexpensive and easy way to produce natural nanofibers on
420 A.K. Haghi and R.K. Haghi
low basis weight, small fiber diameter and pore size. It is hoped that this chapter will pave
the way toward a better understanding of the application of electrospinning of natural fibers.
Nevertheless, among the biodegradable and biocompatible polymers, SF was exten-
sively studied as one of the candidate materials for biomedical applications, because it has
several distinctive biological properties including good biocompatibility, biodegradability,
and minimal inflammatory reaction. One of the promising applications of SF in biomedical
engineering are scaffolding materials for tissue engineering. It was reported that SF matri-
ces could be useful for the culture of fibroblasts and osteoblasts as well as stem cells, and
could enhance the adhesion, growth, and differentiation of the cells in a manner similar to
that of collagen matrices. Nonwoven matrices of electrospun nanofibers could be prepared
from a regenerated SF solution, and the matrices were effective in cell attachment and
spreading of normal human keratinocytes and fibroblasts [82].
References
1. Y.Q. Wan, Q. Guo and N. Pan, Int. J. Nonlinear Sci. Num. Simul., 5, 5–8 (2004).
2. J.H. He, Y.Q. Wan and J.Y. Yu, Int. J. Nonlinear Sci. Num. Simul., 5 (3), 243–52 (2004).
3. J.H. He, Y.Q. Wan and J.Y. Yu, Int. J. Nonlinear Sci. Num. Simul., 5 (3), 253–261 (2004).
4. J.H. He and Y.Q. Wan, Polymer, 45 (19), 6731–6734 (2004).
5. X.-H. Qin, Y.Q. Wan and J.H. He, Polymer, 45 (18), 6409–6413 (2004).
6. S.A. Therona, E. Zussmana and A.L. Yarin, Polymer, 45, 2017–2030 (2004).
7. M.M. Demir, I. Yilgor, E. Yilgor and B. Erman, Polymer, 43, 3303–3309 (2002).
8. A.M. Ganan-Calvo, J Aerosol Sci., 30 (7), 863–872 (1999).
9. J.J. Feng, J. Non-Newtonian Fluid Mech., 116, 55–70 (2003).
10. Y.M. Shin, M.M. Hohman, M.P. Brenner and G.C. Rutledge, Polymer, 42, 9955–9967 (2001).
11. J. He, Y. Wan and J.-Y. Yu, Polymer, 46, 2799–2801 (2005).
12. P.K. Baumgarten, J. Colloid Interface Sci., 36, 71–79 (1971).
13. J. Doshi and D.H. Reneker, J. Electrostat., 35, 151–160 (1995).
14. D.H. Reneker, A.L. Yarin, H. Fong and S. Koombhongse, J. Appl. Phys., 87, 4531–4547 (2000).
15. M. Bognitzki, W. Czando, T. Frese, A. Schaper, M. Hellwig, M. Steinhart, A. Greiner and J.H.
Wendorff, Adv. Mater., 13, 70–72 (2001).
16. E. Zussman, A. Theron and A.L. Yarin, Appl. Phys. Lett., 82, 973–975 (2003).
17. S.A. Theron, A.L. Yarin, E. Zussman and E. Kroll, Polymer, 46, 2889–2899 (2005).
18. Z.-M. Huang, Y.-Z. Zhang, M. Kotak and S. Ramakrishna, Composites Sci. Technol., 63,
2223–2253 (2003).
19. D. Emig, A. Klimmek and E. Raabe, US Patent 6395046 (2002).
20. D. Groitzsch and E. Fahrbach, US Patent 4,618,524 (1986).
21. H.L. Schreuder-Gibson, P. Gibson, K. Senecal, M. Sennett, J. Walker, W. Yeomans et al., J. Adv.
Mater., 34 (3), 44–55 (2002).
22. M.M. Bergshoef and G.J. Vancso, Adv. Mater., 11, 1362–1365 (1999).
23. Z. Ma, M. Kotaki, R. Inai and S. Ramakrishna, Tissue Eng., 11, 101 (2005).
24. C.Y. Xu, R. Inai, M. Kotaki and S. Ramakrishna, Biomaterials, 25, 877 (2004).
25. H. Yoshimoto, Y.M. Shin, H. Terai and J.P. Vacanti, Biomaterials, 24, 2077 (2003).
26. M. Shin, O. Ishii, T. Sued and J.P. Vacanti, Biomaterials, 25, 3717 (2004).
27. Z. Ma, M. Kotaki, Th. Yong, Wei He and S. Ramakrishna, Biomaterials, 26, 2527–2536 (2005).
28. H.J. Jin, S. Fridrikh, G.C. Rutledge and D. Kaplan, Abstracts of Papers American Chemical
Society, 224 (1–2), 408 (2002).
29. Y.K. Luu, K. Kim, B.S. Hsiao, B. Chu and M. Hadjiargyrou, J. Control. Release, 89, 341 (2003).
30. G. Verreck, I. Chun, J. Rosenblatt, J. Peeters, A.V. Dijck, J. Mensch, M. Noppe and M.E.
Brewster, J. Contr. Release, 92, 349 (2003).
31. P.W. Gibson, H.L. Schreuder-Gibson and D. Riven, AIChE J., 45 (1), 190–195 (1999).
32. P.W. Gibson, H.L. Schreuder-Gibson and C. Pentheny, J. Coated Fabrics, 28, 63 (1998).
33. P.W. Gibson, H.L. Schreuder-Gibson and D. Rivin, Coll. Surf. A: Physicochem. Eng. Aspects,
187, 469–481 (2001).
Electrospun Biodegradable and Biocompatible Natural Nanofibers 421
34. H.L. Schreuder-Gibson, P.W. Gibson, K. Senecal, M. Sennett, J. Walker, W. Yeomans et al.,
J. Adv. Mater., 34 (3), 44–55 (2002).
35. I.D. Norris, M.M. Shaker, F.K. Ko and A.G. Macdiarmid, Synth. Metals, 114 (2), 109–114
(2000).
36. C.M. Waters, T.J. Noakes and C.I. Pavery, US Patent, 5088807 (1992).
37. K.J. Senecal, L. Samuelson, M. Sennett and G.H. Schreuder, US Patent Application Publication
0045547 (2001).
38. K.J. Senecal, D.P. Ziegler, J. He, R. Mosurkal, H. Schreuder-Gibson and L.A. Samuelson,
Photoelectric Response from Nanofibrous Membranes, Materials Research Society Symposium
Proceedings,708, pp. 285–289 (2002).
39. X. Wang, C. Drew, S.H. Lee, K.J. Senecal, J. Kumar and L.A. Samuelson, Nano Lett., 11,
1273–1275 (2002).
40. B. Ding, J.H. Kim, Y. Miyazaki and S.M. Shiratori, Sens. Actuators B: Chem., 101, 373–380
(2004).
41. P.I. Gouma, Rev. Adv. Mater. Sci., 5, 147–154 (2003).
42. K. Sawicka, P. Goum and S. Simon, Sensors and Actuators B, 108, 585–588 (2005).
43. J.M. Deitzel, J. Kleinmeyer, J.K. Hirvonen et al., Polymer, 42, 8163–8170 (2001).
44. H. Fong and D.H. Reneker, Electrospinning and Formation of Nanofibers, in: Salem, D.R. (ed.),
Structure Formation in Polymeric Fibers, Munich: Hanser, pp. 225–246 (2001).
45. K. Fujihara, M. Kotak and S. Ramakrishna, Biomaterials, 26, 4139–4147 (2005).
46. X. Fang and D.H. Reneker, J. Macromol. Sci. Phys. B, 36, 169–173 (1997).
47. G.I. Taylor, Proc. R. Soc. London, Ser. A, 313, 453–475 (1969).
48. El-R. Kenawy, G.L. Bowlin, K. Mansfield, J. Layman, D.G. Simpsonc, E.H. Sanders and G. E.
Wnek, J. Contr. Release, 81, 57–64 (2002).
49. S.F. Fennessey, J.R. Farris, Polymer, 45, 4217–4225 (2004).
50. S.-H. Tan, R. Inai, M. Kotaki and S. Ramakrishna, Polymer, 46, 6128–6134 (2005).
51. H. Fong, I. Chun and D.H. Reneker, Polymer, 40, 4585–4592 (1999).
52. S.L. Shenoy, W.D. Bates, H.L. Frisch and G.E. Wnek, Polymer, 46, 3372–3384 (2005).
53. H. Fong and D.H. Reneker, J. Polym. Sci. B: Polym Phys., 37 (24), 3488–3493 (1999).
54. S. Kidoaki, K. Kwon and T. Matsuda, Biomaterials, 26, 37–46 (2005).
55. S.Y. Gu, J. Ren and G.J. Vancso, Eur. Polymer J., 41 (11), 2559–2568 (2005).
56. C.H. Zhang, X. Yuan, L. Wu, Y. Han and Jing Sheng, Eur. Polymer J., 41, 423–432 (2005).
57. X. Geng, O.-H. Kwon and J. Jang, Biomaterials, 26, 5427–5432 (2005).
58. A. Koski, K. Yim and S. Shivkumar, Mater Lett., 58, 493–497 (2004).
59. J.M. Deitzel, J. Kleinmeyer, D. Harris and T.N. Beck, Polymer, 42, 261–272 (2001).
60. Y.J. Ryu, Hak Yong Kim, K.H. Lee, H.C. Park and D.R. Lee, Eur. Polymer J., 39, 1883–1889
(2003).
61. B. Kim, H. Park, S.-H. Lee and W.M. Sigmund, Mater. Lett., 59, 829–832 (2005).
62. T. Jarusuwannapoom, W. Hongrojjanawiwat, S. Jitjaicham, L. Wannatong, M. Nithitanakul, C.
Pattamaprom, P. Koombhongse, R. Rangkupan and P. Supaphol, Eur. Polymer J., 41, 409–421
(2005).
63. C.J. Buchko, L.C. Chen, Y. Shen and D.C. Martin, Polymer, 40, 7397–7407 (1999).
64. X.M. Moa, C.Y. Xub, M. Kotakib and S. Ramakrishna, Biomaterials, 25, 1883–1890 (2004).
65. A.F. Spivak and Y.A. Dzenis, Appl. Phys. Lett., 73, 3067 (1998).
66. D.H. Reneker, A.L. Yarin, H. Fong and S. Koombhongse, J. Appl. Phys., 87, 4531–4547 (2000).
67. M.M. Hohman, M. Shin, G. Rutledge and M.P. Brenner, Phys. Fluids, 13, 2201 (2001).
68. M.M. Hohman, M. Shin, G. Rutledge and M.P. Brenner, Phys. Fluids, 13, 2221 (2001).
69. D.H. Reneker, A.L. Yarin, H. Fong and S. Koombhongse, J. Appl. Phys., 87, 4531 (2000).
70. A.L. Yarin, S. Koombhongse and D.H. Reneker, J. Appl. Phys., 89, 3018 (2001).
71. Y.M. Shin, M.M. Hohman, M.P. Brenner and G.C. Rutledge, Appl. Phys. Lett., 78, 1149 (2001).
72. J.G. Cook, Handbook of Textile Fibers: I. Natural Fibers, third edn, Merrow Publisher (1964).
73. R.R. Franck, Silk, Mohair, Cashmere and Other Luxury Fibres, Woodhead Publishing (2001).
74. D. Kaplan, W.W. Adams, B. Farmer and C. Viney (editors), Silk Polymers: Materials Science
and Biotechnology, Washington DC: ACS Symposium Series, p. 544 (1994).
75. B.-M. Min, L. Jeong, Y.S. Nam, J.-M. Kim, J.Y. Kim and W.H. Park, Int. J. Biol. Macromol., 34,
281–288 (2004).
76. N. Hojo, Structure of Silk Yarn, Part A. Biological and Physical Aspects, Science Publisher,
422 A.K. Haghi and R.K. Haghi
USA (2000).
77. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus and F. Ko, Polymer, 44, 5721–5727 (2003).
78. S. Sukigara, M. Gandhi, J. Ayutsede, M. Micklus and F. Ko, Polymer, 45, 3701–3708 (2004).
79. J. Ayutsede, M. Gandhi, S. Sukigara, M. Micklus, H.-E. Chen and F. Ko, Polymer, 46,
1625–1634 (2005).
80. E. Entchevaa, H. Biena, L. Yina, C.-Y. Chunga, M. Farrella and Y. Kostovc, Biomaterials, 25,
5753–5762 (2004).
81. S.O. Han, W.K. Son, J.H. Youk, T.S. Lee and W.H. Park, Materials Lett., 59 (24–25),
2998–3001 (2005).
82. K E. Park, S.Y. Jung, S.J. Lee, B.-M. Min and W.H. Park, Int. J. Biol. Macromol., 38, 165–173
(2006).
Nomenclature
R = radius of the jet
R ' = slope of the jet surface
Q = constant volume flow rate
I = constant total current in the jet
σ = surface charge density
υ = velocity
E = z component of the electric field
K = conductivity of the liquid
γ = surface tension
Et = tangential components of the electric field at the surface
En = normal components of the electric field at the surface
g = gravity
τ zz = axial viscous normal stress
P = pressure
t ne = normal traction on the surface of the jet due to electricity
t te = tangential traction on the surface of the jet due to electricity.
ε = dielectric constants of the jet
ε– = dielectric constants of the ambient air
η = viscosity