Aromaticity of Polycyclic Conjugated Hydrocarbons

Milan Randic*
National Institute of Chemistry, Ljubljana, Slovenia
Received December 13, 2001
Contents
I. Prologue 3451
II. Introduction 3455
III. Dilemmas 3456
A. Qualitative versus Quantitative Approaches 3456
B. Observables versus Non-observables 3458
C. Structural Criteria versus Properties as Criteria 3459
D. Valence Bond Theory versus Molecular Orbital
Theory
3460
E. On Interlocking of the MO and the VB
Methods
3461
F. Chemical Graph Theory versus Quantum
Chemistry
3462
G. Clar 6n Rule versus Hu ckel 4n + 2 Rule 3464
H. Hydrocarbons versus Heteroatomic Systems 3465
IV. Hidden Treasures of Kekule Valence Structures 3466
A. Conjugated Circuits 3467
B. Innate Degree of Freedom 3470
C. Clar Structures 3472
V. Graph Theoretical Approach to Chemical Structure 3473
A. Metric 3473
B. Chemical Graphs 3473
C. Isospectral Graphs 3473
D. Embedded Graphs 3475
E. Partial Ordering 3476
VI. On Enumeration of Benzenoid Hydrocarbons 3477
VII. Kekule Valence Structures Count 3479
A. Non-branched Cata-condensed Benzenoids 3481
B. Branched Cata-condensed Benzenoids 3482
C. Benzenoid Lattices 3482
D. Peri-condensed Benzenoids 3483
E. Miscellaneous Benzenoids 3484
F. The Approach of Platt 3485
G. Computer Programs for Calculating K 3485
H. Transfer-Matrix Method 3486
I. Use of Recursion Relations 3486
J. Use of Signed Matrices 3487
VIII. Enumeration of Conjugated Circuits 3488
IX. Approximate Approaches versus Ambitious
Computations
3490
A. Semi-empirical Valence Bond Approaches for
Benzenoid Hydrocarbons
3490
1. PaulingWheland Valence Bond Approach 3490
2. PaulingWheland Resonance Theory 3491
3. HerndonSimpson Model 3491
4. Hierarchical VB Schemes 3491
X. Conjugated Circuits Model 3492
A. Monocyclic Conjugated Systems 3493
B. Polycyclic Conjugated Systems 3493
XI. Expression for Molecular Resonance Energy 3494
XII. Benzenoids: Systems with Only 4n + 2
Conjugated Circuits
3494
XIII. Non-benzenoid Systems 3504
A. Biphenylene and Related Non-benzenoid
Systems
3506
B. Non-benzenoid Systems with Odd Rings
Having Only 4n + 2 Conjugated Circuits
3507
C. Non-benzenoid Systems with Odd Rings
Having 4n Conjugated Circuits
3508
D. Summary on Applications of the Conjugated
Circuits Method to Hydrocarbon and Carbon
Chemistry
3508
* Professor Emeri tus, Department of Mathemati cs and Computer
Sci ence, Drake Uni versi ty, Des Moi nes, I A 50311. Permanent
address: 3225 Ki ngman Rd., Ames, I A 50311. Fax: (515) 292-
8629.
Milan Randic is a native of Croatia, born in 1930, and is a citizen of the
United States and Croatia. He studied theoretical physics at the University
of Zagreb (1954) under Professor Ivan Supek (a student of Werner
Heisenberg). Supek introduced him to the book by Linus Pauling, The
Nature of the Chemical Bond, with a request to get involved in quantum
chemistry. He got his Ph.D. degree in Cambridge in 1958, studying high-
resolution infrared molecular spectra with Dr. Norman Sheppard. At the
Rudjer Boskovic Institute in Zagreb, Croatia, in 1960, he founded the
theoretical chemistry group. Since 1971, he has visited several U.S.
universities. From 1980 to 2000, he was in the Department of Mathematics
and Computer Science at Drake University, Des Moines, Iowa. Among
the awards he has received are the City of Zagreb Science Award, the
Boris Kidric Science Award (of Republic of Slovenia), The Governors
Science Award (Iowa), and the Skolnik Award (American Chemical
Society). He is a member of the Croatian Academy of Sciences and Arts.
His continuing interest is to advance a better understanding of chemical
graph theory. Since January 1, 2000, he is Professor Emeritus at Drake
University, while continuing a three-months-a-year visitatation at the
National Institute of Chemistry in Ljubljana, Slovenia. His other interests
include development of a universal language, NOBEL.
PHOTOGRAPH CAPTION: The author is in full agreement with Plutarch
(46119 A.D.), who characterized wine as, Of drinks s the most useful;
Among medication s the most tasteful; Among food s the most
enjoyable.
3449 Chem. Rev. 2003, 103, 34493605
10.1021/cr9903656 CCC: $44.00 2003 American Chemical Society
Published on Web 07/29/2003
XIV. Quantum Chemical Justification of the Conjugated
Circuits Model
3509
XV. On a Diagnostic Use of Conjugated Circuits 3512
XVI. Resonance Graphs 3512
XVII. Aromaticity versus Anti-aromaticity 3516
XVIII. On Classification of Polycyclic Conjugated
Hydrocarbons
3517
XIX. Fully Aromatic Hydrocarbons 3520
XX. Less than Fully Aromatic Hydrocarbons 3522
A. Degree of Aromaticity 3523
B. On the Boundary of Aromaticity 3524
C. Degree of Anti-aromaticity 3525
D. Clar Valence Structures and Aromaticity 3526
XXI. Biphenylenes 3527
XXII. Fully Anti-aromatic Hydrocarbons 3528
XXIII. Aromaticity/Anti-aromaticity 3529
XXIV. ABC of Aromaticity 3530
A. A of Aromaticity 3530
B. B of Aromaticity 3531
C. C of Aromaticity 3532
XXV. Local Aromaticity 3532
A. The Approach of Polansky and Derflinger 3532
B. Approaches Based on Bond Orders 3533
C. Graph Theoretical Approach to Benzene
Character of Fused Benzene Rings
3535
D. On Quantitative Interpretation of Clar Valence
Structures
3535
E. On Neglect of Local Aromaticity 3536
F. Alternative Ring Indices 3537
XXVI. Ring Currents 3539
A. Ring Currents as Ring Descriptors 3539
B. Nuclear-Independent Chemical Shifts (NICS) 3540
C. A Graph Theoretical View on Ring Currents 3541
D. Conjugated Circuits Currents 3542
XXVII. Graph Theoretical Ring Resonance Energies 3544
XXVIII. The Most Aromatic Aromatic Compounds 3544
A. Fully Benzenoid Hydrocarbons 3544
B. Giant Benzenoids 3545
XXIX. Clars Aromatic Sextets 3547
A. Geometrical Definition of Clars Valence
Formulas
3547
B. Mathematical Definition of Clars Valence
Formulas
3549
XXX. Quantum Chemical Justification for Clars
Structures
3549
XXXI. On Interpretation of Molecular Resonance Energy 3552
XXXII. Other Points 3552
A. Benzenoid Hydrocarbons Revisited 3553
B. Biphenylenes Revisited 3556
1. Parity of Kekule Valence Structures 3556
2. Degrees of Freedom of Kekule Valence
Structures of Non-benzenoids
3557
C. Dilemma: Kekule Structures or Clar
Structures?
3558
D. On Kekulene and Superaromaticity 3560
XXXIII. Clar Structures Revisited 3561
A. Exceptions to the Rule 3561
B. Generalized Clar Structures 3562
C. Sextet Polynomial 3563
D.
Canonical Clar Structures 3563
E. Resonant Clar Structures 3564
F. Graphical Construction of Clar Structures 3565
XXXIV. Fullerenes 3566
XXXV. Challenges and Unsolved Problems 3570
A. Challenges 3570
B. Unsolved Problems 3571
XXXVI. Concluding Remarks 3572
XXXVII. Epilogue 3573
A. Quantum Chemical and Graph Theoretical
Analysis of [n]Phenylenes
3574
B. Clar Structures for Non-benzenoid Hydrocarbons 3576
XXXVIII.Chemical Graph TheorysProspects and Retrospect 3577
XXXIX. Tribute to Experimental Chemistry 3579
XL. Apologies 3581
XLI. Acknowledgments 3584
XLII. Appendix 1. Biographical Notes 3584
Arthur Cayley (18211895) 3584
James Joseph Sylvester (18141897) 3585
August Kekule (18291896) 3585
Charles A. Coulson (19101974) 3586
Erich Clar (19021987) 3586
Oskar E. Polansky (19191989) 3587
Michael J. S. Dewar (19181997) 3587
XLIII. Appendix 2. Short Historical Comments on Graph
Theory in Mathematics and Chemistry
3588
A. Graph Theory in Mathematics 3588
B. Graph Theory in Chemistry 3589
C. Further Readings 3589
D. Personal Note 3589
XLIV. Appendix 3. ErrorssOmissions 3590
XLV. Appendix 4. ErrorssPrevention 3591
XLVI. Note Added in Proof 3594
A. Clar Structures of C
60
3594
B. Benzenoid [10]Cyclophenacene Belt 3594
C. Historical Remark on Clar -Sextets 3595
D. More Support for Clars -Aromatic Sextets 3595
XLVII. References 3595
PRACTICAL POINTS: Mol ecul ar structures are
numbered accordi ng to the fi gure i n whi ch they are
shown by a fracti on x/X, where x i s the number of
the mol ecul e and X i s the number of the fi gure
showi ng i ts skel eton. I n the l i st of references, i n the
case of revi ew arti cl es we have gi ven the ti tl e of each
arti cl e as thi s offers some i nformati on on the par-
ti cul ar content of the revi ew. I n addi ti on, we may
have i ndi cated the l ength of the revi ew by l i sti ng the
ful l page range of such arti cl es.
I f a man can preach a better sermon, write a
better book, or make a better mouse-trap than
his neighbor, though hebuild his housein the
woods, theworld will makea beaten path tohis
door.
R. W. Emerson (1803-1882),
Ameri can essayi st and poet,
a l eadi ng fi gure i n
Ameri can l i terary hi story.
3450 Chemical Reviews, 2003, Vol. 103, No. 9 Randic
a mol ecul e. That the presence of 4n conjugated
ci rcui ts i s detri mental to mol ecul ar stabi l i ty has been
wi tnessed al ready for aromati c compounds i n whi ch
4n + 2 conjugated ci rcui ts domi nate and enforce the
pl anar geometry for the systems i n whi ch 4n conju-
gated ci rcui ts are al sopresent. Hafner and Schnei der
553
consi dered al kyl deri vati ves of aceheptyl ene and
observed that these compounds depart somewhat i n
thei r properti es from the standard non-benzenoi d
conjugated systems. Cl earl y thi s departure i s due
to the presence of anti -aromati c 4n contri buti ng
conjugated ci rcui ts.
For a compound to qual i fy as anti -aromati c, how-
ever, we must have a domi nant anti -aromati c con-
tri buti on. Hafner and co-workers
655
desi gned a cl ever
way to arri ve at anti -aromati c compounds by forc-
i ng a mol ecul e that has undergone the Jahn-Tel l er
di storti on back i nto a more symmetri cal geometri c
form. Thi s was accompl i shed by i ntroduci ng bul ky
tert-butyl substi tuents on the s-i ndacene skel etal
frame. Cal cul ati ons done by Gel l i ni et al .
656
show that
CC bonds i n s-i ndacene al ternate i n l ength, assumi ng
typi cal CC doubl e (around 1.32-1.33 ) and CC
si ngl e (1.47-1.49 ) bond l engths, respecti vel y, and
resul ts i n a structure havi ng C
2h
symmetry. Thi s
structure represents a more stabl e system than the
hypotheti cal structure wi thout bond al ternati on (hav-
i ng symmetry D
2h
). I n contrast, cal cul ati ons on
1,3,5,7-tetra-tert-butyl -s-i ndacene (TTBI ) show that
the more stabl e system i s the one wi th more equal -
i zed peri pheral CC bond l engths (around 1.40-1.41
), i n agreement wi th the avai l abl e X-ray data.
Hafner and co-workers, after careful anal ysi s of
i nfrared and Raman spectra of 1,3,5,7-tetra-tert-
butyl -s-i ndacene, concl uded that Thetert-butyl effect
on theTTBI structureresults in a larger delocaliza-
tion of the-electron densityover theindaceneplane,
giving rise to a molecular structure close to D
2h
symmetry.
656
Hence, because tert-butyl groups do not consti tute
conjugated fragments and are i gnored i n the conju-
gated ci rcui ts model , we may concl ude that modi -
fi ed s-i ndacene, that i s, 1,3,5,7-tetra-tert-butyl -s-
i ndacene, represents an anti -aromati c speci es. More
recentl y, Bal dri dge and Si egel
657
have theori zed that
cycl ooctatetraene (COT) can si mi l arl y be forced to
adopt a pl anar rather than a tub conformati on. A fl at
form for COT woul d make contri buti ons from 4n
conjugated ci rcui t be fel t and thus woul d approach
an anti -aromati c structure. Accordi ng to cal cul ati ons,
the two anneal ed bi cycl o[2.2.1]hexene fragments
woul d resul t i n a pl anar structure that woul d have
l ocal i zed -el ectrons. Matsuura and Komatsu
658,659
synthesi zed the compound, and i ndeed the X-ray
confi rmed the cal cul ati ons i ndi cati ng the presence of
CC doubl e bonds i n the endo conformati on wi th
respect to anneal ed bi cycl o[2.2.1]hexeno fragments.
Addi ng four anneal ed fragments resul ted i n a pl anar
structure wi th exocycl i c CC doubl e bonds wi th re-
spect to the fragments. Fi nal l y, by pl aci ng anneal ed
bi cycl o[2.2.1]hexeno fragments two bonds apart, one
prevents CC doubl e bond fi xati on, and as a resul t
one obtai ns a ful l y anti -aromati c modi fi ed cycl ooc-
tatetraene. These recent achi evements show that
anti -aromati ci ty that was for the most part fi cti -
ti ous i s becomi ng factual .
An i ndi rect proof that anti -aromati c compounds
are el usi ve comes from data on i nter-stel l ar com-
pounds.
660-662
I n vi ew of the l ow densi ty of matter
and extremel y l ow temperatures i n outer space,
structures that woul d be di ffi cul t to observe i n the
l aboratory may have l ong enough l i fe i n the i nter-
stel l ar space to be detected. Thus, for i nstance, among
others, the smal l est aromati c compound, cycl i c C
3
H
3
+
,
has been i denti fi ed i n the i nter-stel l ar space. The
search for anti -aromati c compounds i n the i nter-
stel l ar space thus appears to be an i nteresti ng
project. However, as of today, no anti -aromati c
compounds have been detected i n outer space,
663
al though, as i s wel l known, the i nter-stel l ar space i s
ri ch i n hydrocarbons.
I n discussing aromaticity/anti-aromaticity, the prob-
l em that needs better understandi ng i s why some
aromati c compounds (havi ng 4n conjugated ci rcui ts)
have modi fi ed properti es, as was the case wi th
Hafners hydrocarbons, and i n other compounds, l i ke
bi phenyl ene, 4n conjugated ci rcui ts apparentl y do not
show a vi si bl e mani festati on of the presence of 4n
conjugated ci rcui ts. We wi l l address thi s probl em i n
secti on XXXI I (Bi phenyl enes Revi si ted).
XXIV. ABC of Aromaticity
We can summari ze the graph theoreti cal approach
to aromati ci ty by focusi ng on the three i mportant
aspects pertai ni ng to aromati ci ty: (A) cl assi fi cati on
of compounds as ful l y aromati c or l ess aromati c; (B)
characteri zati on of the degree of aromati ci ty of ful l y
aromati c compounds; and (C) di scri mi nati on of the
l ocal aromati c features of l arger compounds. We wi l l
bri efl y outl i ne each of these three i mportant steps
for cl ari fi cati on of the concept of aromati ci ty as i t
appl i es to pol ycycl i c conjugated hydrocarbons. From
Tabl e 39, whi ch summari zes the ABC of aromati ci ty,
one can i mmedi atel y see that al l the three i ndi ces
are rel ated. I n parti cul ar, i t fol l ows that for aromati c
compounds A > 0, for azul ene and so-cal l ed azul e-
noi d compounds, bui l t from fi ve- and seven-member
ri ngs, B ) 0, whi l e for ri ngs not parti ci pati ng i n
conjugated ci rcui ts and thus not contri buti ng to
mol ecul ar RE, l i ke the central ri ngs of peryl ene and
bi santhene, C ) 0.
A. A of Aromaticity
As we have seen, aromati ci ty has been defi ned i n
terms of the presence and the absence of 4n + 2 and
4n conjugated ci rcui ts. The approach has l ed to a
numeri cal i ndex that esti mates the degree of aroma-
ti ci ty for compounds havi ng both 4n + 2 and the 4n
conjugated ci rcui ts. I t appears qui te natural to expect
Table 39. ABCs of Aromaticity
A )
RE(4n + 2) - RE(4n)
RE(4n + 2) + RE(4n)
aromati ci ty
B )
RE(R
1
) - RE(R
n*1
)
RE(R
1
) + RE(R
n*1
)
benzene character
C )
RRE(R
1
) - RRE(R
n*1
)
RRE(R
1
) + RRE(R
n*1
)
Cl ar i ndex
3530 Chemical Reviews, 2003, Vol. 103, No. 9 Randic
that an i ndex of aromati ci ty i s sensi ti ve to the
rel ati ve rol es of 4n + 2 and 4n conjugated ci rcui ts.
Constructi on of such an i ndex i s accompl i shed by
parti ti oni ng RE i nto contri buti ons ari si ng from 4n + 2
conjugated ci rcui ts, RE(4n + 2), and destabi l i zi ng
contri buti ons ari si ng from 4n conjugated ci rcui ts,
RE(4n). By fol l owi ng thi s reasoni ng, we arri ve at the
i ndex A gi ven i n Tabl e 39.
The i ndex A assi gns the val ue A ) 1 to al l
benzenoi d hydrocarbons, whi ch are thus al l charac-
teri zed as ful l y aromati c. However, as we know,
there are vari ati ons of the degree of aromati ci ty even
among benzenoi d hydrocarbons, though they may be
rel ati vel y smal l . We have al ready seen that ful l y
benzenoi d hydrocarbons, l i ke tri phenyl ene, di ben-
zopyrene, hexabenzocoronene, and other 6n -el ec-
tron systems, show unusual stabi l i ty. Hence, we need
a measure of aromati ci ty that can di fferenti ate
aromati c character among ful l y aromati c com-
pounds.
B. B of Aromaticity
I n order to characteri ze these smal l vari ati ons of
aromati c features of benzenoi d hydrocarbons, we
consi der another i ndex that wi l l measure the degree
of si mi l ari ty of a gi ven benzenoi d hydrocarbon to
benzene. Such an i ndex, referred to as the i ndex of
benzene character B of benzenoi d hydrocarbons, was
proposed i n 1987 as fol l ows:
664
Here, RE(R
1
) i s the part of the computed resonance
energy that comes from the presence of the conju-
gated ci rcui ts R
1
, whi l e RE(R
n*1
) i s the part of the
computed resonance energy that comes from the
presence al l other conjugated ci rcui ts. The denomi na-
tor, [RE(R
1
) +RE(R
n*1
)], i s i n fact the total mol ecul ar
RE, whi ch thus makes B a di mensi onl ess number,
al ways l ess than one, except for benzene, for whi ch
by defi ni ti on B ) 1.
I n Tabl e 40 we have col l ected B i ndi ces for a
number of smal l er benzenoi d hydrocarbons. For
compari son, we al so show the correspondi ng aroma-
ti ci ty i ndi ces based on bond l engths. Accordi ng to
Jul g,
665-667
one can deri ve an i ndex of aromati c
character from the di fferences between the actual CC
bond l engths i n a benzenoi d hydrocarbon and the
average CC bond di stance d usi ng the expressi ons
The numeri cal val ue of i ndex constructed i n thi s way
wi l l depend on the bond l engths used, whi ch coul d
be experi mental bond l engths or cal cul ated ones. One
of the two Jul gs i ndi ces shown i n Tabl e 40 i s based
on experi mental CC bond l engths, and the other i s
based on quantum chemi cal cal cul ati ons usi ng the
AM1 model (whi ch i s attri buted to Dewar). As we can
see from Tabl e 40, the magni tudes of the deri ved
i ndi ces based on bond l engths vary somewhat i n a
rel ati vel y smal l i nterval , whi l e graph theoreti cal l y
deri ved B i ndi ces vary i n the i nterval from about 1/2
to 1, the l argest possi bl e val ue assumed to bel ong to
benzene. A compari son between the B i ndex of
aromati ci ty and i ndi ces based on the approach of Jul g
shows some di fferences for i ndi vi dual benzenoi ds. For
exampl e, the B i ndi ces for naphthal ene, peryl ene, and
quaterryl ene, whi ch are al l formal l y bui l t by con-
necti ng naphthal ene uni ts by essenti al l y si ngl e CC
bonds, are the same. The Jul g i ndi ces based on AM1
cal cul ati ons for the same compounds show a decreas-
i ng trend, the l argest val ue bel ongi ng to naphthal ene.
Cl earl y, the decrease of Jul gs i ndex can be attri buted
to the i ncrease i n the number of essenti al l y si ngl e
CC bonds i n peryl ene and quaterryl ene, whi ch do not
parti ci pate i n conjugati on but neverthel ess i nfl uence
the average bond di stance d. For more detai l s on
Jul gs bond l ength i ndi ces, the reader shoul d consul t
a recent paper by Ki ral j and Ferrei ra
668
on predi cti ng
CC bond l engths i n pl anar benzenoi d pol ycycl i c
hydrocarbons, where one can al so fi nd a l i st of over
300 experi mental and cal cul ated bond l engths for
some 27 smal l er benzenoi d hydrocarbons.
One can construct other i ndi ces for benzene char-
acter of benzenoi d hydrocarbons, whi ch we wi l l
desi gnate as B i n order to di fferenti ate them from
the i ndex B. For exampl e, one can consi der the
Kekul e i ndex,
595
whi ch assi gns to i ndi vi dual Kekul e
val ence structures an i ndex deri ved from l ocal prop-
erti es of mol ecul ar orbi tal s, and take the average over
al l Kekul estructures. Even though for many Kekul e
structures thi s i ndex i s bi gger than the Kekul ei ndex
of benzene, the average Kekul ei ndex appears smal l er
for pol ycycl i c benzenoi d hydrocarbons than i t i s for
benzene. However, for an i ndex to refl ect benzene
character, one expects certai n trends among struc-
tural l y rel ated benzenoi ds to be sati sfi ed, such as
B ) [RE(R
1
) - RE(R
n*1
)]/[RE(R
1
) + RE(R
n*1
)]
J ) 1 - 255[/d]
2
)
i
[d
i
- d]
2
/N
Table 40. B Indices for a Number of Smaller
Benzenoid Hydrocarbons
Jul gs val ues
benzenoi d B i ndex exptl cal cd (AM1)
benzene 1.000 1.000 1.000
naphthal ene 0.751 0.932 0.928
anthracene 0.630 0.889 0.878
phenanthrene 0.778 0.878 0.928
tetracene 0.561 0.870 0.849
tri phenyl ene 0.801 0.906 0.946
chrysene 0.734 0.848 0.922
pyrene 0.630 0.916 0.899
peryl ene 0.751 0.877 0.890
di benzo[a,h]anthracene 0.742 0.972 0.906
pi cene 0.670 0.900 0.921
benzo[e]pyrene 0.712 0.877 0.890
pentacene 0.522 0.880 0.826
di benzo[a,c]anthracene 0.796 0.891 0.915
di benzo[fg,op]tetracene 0.800 0.881 0.944
benzo[ghi]peryl ene 0.649 0.875 0.921
coronene 0.585 0.955 0.933
quaterryl ene 0.751 0.889 0.877
hexabenzocoronene 0.741 0.910 0.848
kekul ene 0.690 0.877 0.881
B(naphthal ene) > B(anthracene) >
B(tetracene) > ...
B(ful l y benzenoi d hydrocarbons) >
B(benzenoi d hydrocarbons)
Aromaticity of Polycyclic Conjugated Hydrocarbons Chemical Reviews, 2003, Vol. 103, No. 9 3531
Whi l e the average Kekul e i ndex sati sfi es the fi rst
condi ti on, i t apparentl y fai l s to sati sfy the second,
because B(naphthal ene) > B(tri phenyl ene), naph-
thal ene havi ng mi grati ng -sextets and tri phenyl ene
bei ng a ful l y benzenoi d hydrocarbon.
Neverthel ess, the Kekul e i ndex, whi ch assi gns to
i ndi vi dual Kekul e val ence structures a numeri cal
val ue, i s of some i nterest, as i t refl ects the rel ati ve
i mportance of i ndi vi dual Kekul e val ence structures,
a topi c whi ch has recei ved some but apparentl y not
suffi ci ent attenti on i n the l i terature.
C. C of Aromaticity
Some cauti on i s requi red when consi deri ng bi gger
and bi gger mol ecul es, i ncl udi ng benzenoi d hydrocar-
bons, i n whi ch there coul d be consi derabl e vari ati ons
i n l ocal properti es. Al ready i n smal l er benzenoi ds,
such as peryl ene and bi santhene, i n whi ch the central
CC bonds are essenti al l y si ngl e (that i s, they are
si ngl e CC bonds i n al l Kekul e val ence structures),
we have porti ons of a mol ecul e that are not contri b-
uti ng to RE, and thence to aromati ci ty. By averagi ng
CC bond l engths, the contri buti ons from vari ous
ri ngs, or the contri buti ons from di fferent Kekul e
val ence structures, we may be di l uti ng the aromati c
characteri sti cs of a mol ecul e wi th spuri ous contri bu-
ti ons. I t does appear that useful characteri zati on of
benzenoi d hydrocarbons may fol l ow i f we consi der
i ndi vi dual benzene ri ngs, rather than taki ng the
average of such contri buti ons from al l ri ngs. Thi s
l eads us to a ri ng i ndex whi ch we refer to as C of
aromati ci ty, where C stands for Cl ar, whi ch i s defi ned
as
539
RRE(R
1
) i s the part of the computed ri ng resonance
energy of i ndi vi dual benzene ri ngs that comes from
the presence of the conjugated ci rcui ts R
1
, whi l e
RRE(R
n*1
) i s the part of the computed contri buti on
to the resonance energy for the consi dered ri ng that
comes from the presence of al l other conjugated
ci rcui ts contri buti ng to RE to the parti cul ar benzene
ri ng. The denomi nator [RRE(R
1
) + RRE(R
n*1
)] i s i n
fact the total ri ng resonance energy (RRE), whi ch
thus makes C a di mensi onl ess number, al ways l ess
than one, except for benzene, for whi ch by defi ni ti on
C ) 1. I n the next secti on we gi ve numeri cal
exampl es of the RRE.
XXV. Local Aromaticity
I t i s not onl y among benzenoi ds that di fferent
mol ecul es show di fferent degrees of aromati c char-
acter, but wi thi n si ngl e pol ycycl i c conjugated hydro-
carbons, di fferent ri ngs show di fferent l ocal aroma-
ti ci ti es. Dewar expl i ci tl y menti oned extendi ng the
noti on of aromati ci ty cri teri a to i ndi vi dual ri ngs i n
pol ycycl i c systems.
669
The pi oneeri ng work on bri dg-
i ng the gap between cal cul ated MO resul ts and
characteri zati on of l ocal aromati ci ty of benzene ri ngs
was attri buted to Pol ansky and Derfl i nger,
582
whose
work deserves more attenti on. They found some
justi fi cati on for Cl ars model of l ocal i zed benzenoi d
regi ons i n pol ycycl i c conjugated benzenoi ds, and they
deri ved a ri ng i ndex characteri zi ng i ndi vi dual ben-
zene ri ngs of a mol ecul e. Thi s ri ng i ndex i s deter-
mi ned from computed MO coeffi ci ents when the MOs
of the system are expanded i n sets of MOs of each
ri ng. The benzene character for benzene ri ngs of
numerous smal l er benzenoi d hydrocarbons reported
by Pol ansky and Derfl i nger was based on usi ng the
HMO approach, but conceptual l y the approach i s
qui te general and can be extended to more sophi sti -
cated MO cal cul ati ons.
A. The Approach of Polansky and Derflinger
I n the standard MO approach to conjugated hy-
drocarbons (e.g., i n the HMO cal cul ati ons), the mo-
l ecul ar orbi tal s are expressed as a l i near combi nati on
of atomi c orbi tal s. However, i t i s possi bl e, as Pol an-
sky and Derfl i nger outl i ned,
582
to express the same
mol ecul ar orbi tal s i n terms of mol ecul ar orbi tal s of
the benzene ri ng. As a resul t, i nstead of obtai ni ng
i nformati on on the contri buti ons of mol ecul ar orbi tal s
to the bond orders, one obtai ns the contri buti ons of
a set of benzene orbi tal s to i ndi vi dual benzene ri ngs
of pol ycycl i c benzenoi d hydrocarbons. I n other words,
the benzene character descri bed by Pol ansky and
Derfl i nger, is theprojection of occupied -MOs in a
given hexagon L of a polycyclic benzenoid hydrocar-
bon onto the three occupied MOs of a benzene
moleculelocated on that position.
523
I n Fi gure 92 we i l l ustrate for a col l ecti on of smal l er
benzenoi d hydrocarbons the benzene ri ng i ndi ces as
reported by Pol ansky and Derfl i nger.
582
The resul ts
are qui te i nteresti ng, i f not astoundi ng. As one can
see, there are consi derabl e vari ati ons among i ndi -
vi dual ri ngs wi thi n a mol ecul e and between ri ngs i n
di ffer ent mol ecul es. I f we i gnor e di phenyl and
hexaphenyl , i n whi ch benzene ri ngs are bri dged by
a si ngl e CC bond rather than fused, and i n whi ch
hi gher val ues for the ri ng i ndi ces were found than
i n benzene, the ri ng i ndi ces of Pol ansky and Der-
fl i nger are smal l er than the val ue for benzene,
varyi ng between about 0.680 to 0.950. Observe al so
that the vari ati ons are more pronounced i n some
mol ecul es than others. A more careful exami nati on
of the resul ts of Pol ansky and Derfl i nger shows that
the aromati c character of i ndi vi dual ri ngs i s far from
uni form. Moreover, the vari ati ons show some anti ci -
pated and some unexpected regul ari ti es. As a rul e,
termi nal ri ngs i n benzenoi d hydrocarbons show
greater si mi l ari ty to benzene, suggesti ng that the
l ocal aromati ci ty may be i nfl uenced to a consi derabl e
degree by speci fi cs of mol ecul ar peri phery patterns.
Hi ghl y si gni fi cant are the observed l arge di fferences
i n benzene character among many adjacent ri ngs.
Thi s i s parti cul arl y vi si bl e i n the cases of tri -
phenyl ene (6/92), tetrabenzanthracene (15/92), di ben-
zopyrene (13/92), and hexabenzocoronene (20/92).
Observe that al l the menti oned benzenoi ds are those
that Cl ar cl assi fi ed as ful l y benzenoi d. I n compari -
son wi th ri ngs i n other benzenoi ds shown i n Fi gure
92, they al l show a l arge benzene character for
benzene ri ngs whi ch are the si tes of aromati c -sex-
tets. I t i s si gni fi cant vi ndi cati on for Cl ars model of
C ) [RRE(R
1
) - RRE(R
n*1
)]/
[RRE(R
1
) + RRE(R
n*1
)]
3532 Chemical Reviews, 2003, Vol. 103, No. 9 Randic
(643) Fol wer, P. W.; Stei ner, E.; Cadi ol i , B.; Zanasi , R. J . Phys. Chem.
1998, 102, 7297.
(644) Gomes, J. A. N. F.; Mal l i on, R. B. Aromati ci ty and Ri ng Currents.
Chem. Rev. 2001, 101, 1349-1383.
(645) Gossauer, A.; Rexhausen, H. J . Chem. Soc., Chem. Commun.
1983, 275.
(646) Plavsic, D.; Nikolic, S.; Trinajstic, N. J . Mol. Struct. (THEOCHEM)
1992, 277, 213.
(647) Chung, A. L.; Dewar, M. J. S. J . Chem. Phys. 1965, 42, 756.
(648) Bresl ow, R. Angew. Chem., I nt. Ed. Engl. 1968, 7, 565.
(649) Bresl ow, R. Acc. Chem. Res. 1973, 6, 393.
(650) Bal l y, T.; Masamune, S. Tetrahedron 1980, 36, 343.
(651) Mai er, G. Angew. Chem., I nt. Ed. Engl. 1986, 27, 309.
(652) Katz, T. J.; Strauss, H. L. J . Chem. Phys. 1960, 32, 1873.
(653) I rngarti nger, H.; Ni xdorf, M. Angew. Chem., I nt. Ed. Engl. 1983,
22, 403.
(654) Jahn, H. A.; Tel l er, E. Proc. R. Soc. (London) 1937, A161, 220.
(655) Hafner, K.; Stowasser, B.; Kri mmer, H.-P.; Fi scher, S.; Bohm,
M. C.; Li nder, H. J. Angew. Chem., I nt. Ed. Engl. 1986, 25, 630.
(656) Gel l i ni , C.; Cardi ni , G.; Sal vi , P. R.; Marconi , G.; Hafner, K. J .
Phys. Chem. 1993, 97, 1286.
(657) Bal dri dge, K. K.; Si egel , S. S., ci ted i n: Rouhi , A. M. Chem. Eng.
News 2001, March 12, 55.
(658) Matsuura, A.; Komatsu, K. J . Am. Chem. Soc. 2001, 123, 1775.
(659) Matsuura, A.; Komatsu, K. J . Am. Chem. Soc. 2001, 123, 1768.
(660) Dul ey, W. W.; Wi l l i ams, D. A. I nterstellar Chemistry; Academi c
Press: New York, 1984.
(661) Leger, A.; dHendecourt, L. I n PolycyclicAromaticHydrocarbons
and Astrophysics; Leger, A., et al ., Eds.; NATO ASI Seri es C,
191; Rei del Publ .: Norwel l , MA, 1987; pp 223-254.
(662) Hudgi ns, D. M. I nterstel l ar Pol ycycl i c Aromati c Compounds and
Astrophysi cs. Polycyclic Arom. Compd. 2002, 22, 469-488.
(663) Ai hara, J.-I .; Hi rama, M. I nternet Electron. J . Mol. Des. 2002,
1, 52.
(664) Randi c, M.; Ni kol i c, S.; Tri najsti c, N. Gazz. Chim. I tal. 1987,
117, 69.
(665) Jul g, K. I n Aromaticity, Pseudoaromaticity, Anti-Aromaticity;
Bergmann, E. D., Pul l man, B., Eds.; I srael Academy of Sci ences
and Humani ti es: Jerusal em, 1971; p 383.
(666) Schl eyer, P. v. R.; Freeman, P. K.; Ji ao, H.; Gol dfuss, B. Angew.
Chem., I nt. Ed. Engl. 1995, 34, 337.
(667) Lewi s, D.; Peters, D. Facts and Theories of Aromaticity; Mac-
mi l l an Press: London, 1975; pp 10-14.
(668) Ki ral j, R.; Ferrei ra, M. M. C. J . Chem. I nf. Comput. Sci. 2002,
42, 508.
(669) Dewar, M. J. S. Angew. Chem., I nt. Ed. Engl. 1971, 10, 761.
(670) Hosoya, H., recent pri vate communi cati on.
(671) Randi c, M. Acta Chim. Slov., to be submi tted.
(672) Randi c, M. Croat. Chem. Acta, to be submi tted.
(673) Dewar, M. J. S.; Tri najsti c, N. Collect. Czech. Chem. Commun.
1970, 35, 3136.
(674) Pl anck, M. Survey of Physical Theory; Dover Publ .: New York,
1993 (transl ated by R. Jones and D. H. Wi l l i ams from Phys-
i kal i sche Rundbl i cke, ori gi nal l y publ i shed by Methuen: London,
1925).
(675) Si gmund Freud (1856-1939) was an Austri an physi ci an who
founded psychoanal ysi s and revol uti oni zed modern psychi atry.
He was honored i n 1930 wi th the Goethe Pri ze for Li terature.
He moved to Engl and to avoi d the dangers of Nazi occupati on
of Austri a and di ed i n London.
(676) Buhl , M.; Hi rsch, A. Spheri cal Aromati ci ty of Ful l erenes. Chem.
Rev. 2001, 101, 1153-1183.
(677) Mi tchel l , R. H. Measuri ng Aroamti ci ty by NMR. Chem. Rev.
2001, 101, 1301-1315.
(678) Krygowski , T. M.; Cyranski , M. K. Structural Aspects of Aro-
mati ci ty. Chem. Rev. 2001, 101, 1385-1419.
(679) Katri tzky, A. R.; Jug, K.; Oni ci u, D. C. Quanti tati ve Measures
of Aromati ci ty for Mono-, Bi -, and Tri cycl i c Penta- and Hexa-
atomi c Heteroaromati c Ri ng Systems and Thei r I nterrel ati on-
shi ps. Chem. Rev. 2001, 101, 1421-1449.
(680) De Proft, F.; Geerl i ngs, P. Conceptual and Computati onal DFT
i n the Study of Aromati ci ty. Chem. Rev. 2001, 101, 1451-1464.
(681) Sl ayden, S. W.; Li ebman, J. F. The Energeti cs of Aromati c
Hydrocarbons: An Experi mental Thermochemi cal Perspecti ve.
Chem. Rev. 2001, 101, 1541-1566.
(682) Kruszewski , J.; Krygowski , T. M. Tetrahedron Lett. 1972, 3839.
(683) Krygowski , T. M. J . Chem. I nf. Comput. Sci. 1993, 33, 70.
(684) Krygowski , T. M.; Cyranski , M. Tetrahedron 1972, 52, 1713.
(685) Krygowski , T. M.; Cyranski , M.; Ci esi el ski , A.; Swi rska, B.;
Leszczynski , P. J . Chem. I nf. Comput. Sci. 1996, 36, 1135.
(686) Li , S.; Ji ang, Y. J . Am. Chem. Soc. 1995, 117, 8401.
(687) Maynau, D.; Sai d, M.; Mal ri eu, J. P. J . Am. Chem. Soc. 1983,
105, 5244.
(688) Hai gh, C. W.; Mal l i on, R. B. Croat. Chem. Acta 1989, 62, 1.
(689) Anusooya, Y.; Chakrabarti , A.; Pati , S. K.; Ramasesha, S. I nt.
J . QuantumChem. 1987, 70, 503.
(690) Gomes, J. A. N. F.; Mal l i on, R. B. J . Org. Chem. 1981, 46, 719.
(691) Subramani an, G.; Schl eyer, P. v. R.; Ji ao, H. Organometallics
1997, 16, 2362.
(692) Paul i ng, L. J . Chem. Phys. 1936, 4, 673.
(693) Jonathan, N.; Gordon, S.; Dai l ey, B. P. J . Chem. Phys. 1962,
36, 2443.
(694) Memory, J. D. Biochim. Biophys. Acta 1963, 66, 168.
(695) Jung, D. E. Tetrahedron 1969, 25, 129.
(696) Mal l i on, R. B. J . Mol. Spectrosc. 1970, 35, 491.
(697) Haddon, R. C. Tetrahedron 1972, 28, 3613.
(698) Fi geys, H. P.; Defay, N.; Marti n, R. H.; McOmi e, J. F. M.; Ayers,
B. E.; Chadwi ck, J. B. Tetrahedron 1976, 32, 2571.
(699) Ai hara, J. PureAppl. Chem. 1982, 54, 1115.
(700) Randi c, M. Acta Chim. Slov. 1997, 44, 361.
(701) Randi c, M.; Guo, X. I nt. J . QuantumChem. 1999, 74, 251.
(702) Randi c, M.; Guo, X., to be publ i shed.
(703) The word regressi on came from earl y stati sti cal observati ons
that the hei ght of chi l dren of tal l parents i s hi gh, but not as
hi gh as that of thei r parents and si mi l arl y that the hei ght of
short chi l dren i s short, but not as short as that of thei r parents.
Thus, hei ght as a property transmi tted to the next generati ons
does not progress but rather regresses.
(704) Hoffmann, R. Physica 1991, A 177, 22.
(705) Doeri ng, W. E.; Roth, W. R. Tetrahedron 1963, 19, 715.
(706) Doeri ng, W. E.; Roth, W. R. Angew. Chem. 1963, 75, 27.
(707) Schroder, G. Chem. Ber. 1964, 97, 3140.
(708) Merenyi , R.; Oth, J. F. M.; Schroder, G. Chem. Ber. 1964, 97,
3150.
(709) Randi c, M. J . Mol. Struct. (THEOCHEM) 1991, 229, 139.
(710) Randi c, M.; Pl avsi c, D.; Tri najsti c, N. Struct. Chem. 1991, 2, 543.
(711) Hei senberg, W. Theory, Cri ti ci sm and Phi l osophy. I n From a
Life of Physics; Li fshi tz, E. M., Ed.; Worl d Sci enti fi c Co.:
Si ngapore, 1989; p 55.
(712) Zander, M. Polycyclische Aromatien; Teubner Studi enbucher,
B.G. Teubner: Stuttgart, 1995.
(713) Brauchl e, Ch.; Voi tl ander, J. Tetrahedron 1982, 38, 279.
(714) Trenti ni , F. V.; Voi tl ander, J.; Brauchl e, Ch. Mol. Phys. 1991,
74, 553.
(715) Brauchl e, Ch.; Voi tl ander,; Vogl er, H. Z. Naturforsch. 1981, 36a,
651.
(716) Bi ermann, D.; Schmi dt, W. J . Am. Chem. Soc. 1980, 102, 3163.
(717) Bi ermann, D.; Schmi dt, W. J . Am. Chem. Soc. 1980, 102, 3173.
(718) Vol l hardt, K. P. C. Angew. Chem. 1984, 96, 525.
(719) Kohn, M. C. Bull. Math. Biol. 1986, 48, 417.
(720) Schmal z, T. G. A Val ence Bond Vi ew of Ful l erenes. I n Valence
Bond Theory; Cooper, D. L., Ed.; El sevi er: Amsterdam, 2002;
Chapter 17, pp 535-564.
(721) Kl ei n, D. J. Resonati ng Val ence-Bond Theori es for Carbon
-Networks and Cl assi cl e/Quantum Connecti ons. I n Valence
Bond Theory; Cooper, D. L., Ed.; Theoreti cal and Computati onal
Chemi stry 10; El sevi er: Amsterdam, 2002; pp 447-502.
(722) Kl ei n, D. J.; Li u, I nt. J . QuantumChem. 1994, S28, 50.
(723) Si r Arthur Conan Doyl e (1859-1930) was Scotti sh physi ci an and
wri ter. The quotati on i s a statement made by the character
Sherl ock Hol mes i n The Si gn of Four, Chapter 6, 1890.
(724) Anderson, P. G. I n Fibonacci Numbers and Their Applications;
Phi l i ppou, A. N., Bergum, P. G., Horadam, A. F., Eds.; Rei del
Publ .: Dordrecht, 1986; p 2.
(725) Shavi tt, I . I n TheUnitary group for theevaluation of electronic
energy matrix elements; Hi nze, J., Ed.; Lecture Notes i n Chem-
i stry 22; Spri nger-Verl ag: New York, 1981
(726) Robb, M. A.; Ni azi , U. Comput. Phys. Rep. 1984, 1, 127.
(727) Wi l cox, C. F., Jr. Tetrahedron Lett. 1968, 795.
(728) Di as, J. R. J . Math. Chem. 1992, 9, 253.
(729) Di as, J. R. J . Mol. Struct. (THEOCHEM) 1990, 206, 1.
(730) Gutman, I .; Tri najsti c, N. Croat. Chem. Acta 1975, 47, 35.
(731) Wi l cox, C. F., Jr. J . Am. Chem. Soc. 1969, 91, 1969.
(732) Herndon, W. C. J . Am. Chem. Soc. 1976, 98, 887.
(733) Gutman, I .; Tri najsti c, N.; Wi l cox, C. F., Jr. Tetrahedron 1975,
31, 143.
(734) Wi l cox, C. F., Jr.; Gutman, I .; Tri najsti c, N. Tetrahedron 1975,
31, 147.
(735) Cvetkovi c, D.; Gutman, I . Tri najsti c, N. J . Mol. Struct. 1975,
28, 443.
(736) DAmato, S. S.; Gi marc, B.; Tri najsti c, N. Croat. Chem. Acta
1981, 54, 1.
(737) Randi c, M. Mol. Phys. 1977, 34, 849.
(738) Kri vka, P.; Tri najsti c, N. Collect. Czech. Chem. Commun. 1985,
50, 291.
(739) Kl ei n, D. J.; Schmal z, T. G.; El -Basi l , S.; Randi c, M.; Tri najsti c,
N. J . Mol. Struct. 1988, 179, 99.
(740) Gutman, I .; Randi c, M.; Tri najsti c, N. Rev. Roum. Chim. 1978,
23, 383.
(741) Wi l cox, C. F., Jr. J . Org. Chem. 1989, 54, 5035.
(742) Babi c, D.; Tri najsti c, N. J . Mol. Struct. (THEOCHEM) 1994, 314,
321.
(743) Kovacek, D.; Margeti c, D.; Maksi c, Z. B. J . Mol. Struct.
(THEOCHEM) 1993, 285, 195.
(744) Berri s, B. C.; Hovakeemi am, G. H.; Lai , Y.-H.; Mestdagh, H.;
Vol l hardt, K. P. C. J . Am. Chem. Soc. 1985, 107, 5670.
(745) Di ercks, R.; H.; Vol l hardt, K. P. C. Angew. Chem., I nt. Ed. Engl.
1986, 25, 266.
3602 Chemical Reviews, 2003, Vol. 103, No. 9 Randic