205

THEMOLECULARADsoRPTIoNAREASoFHYDROCARBON
GASES ON CHARCOAL1
Bv N{. A. Nev2 eNo J. L. MonnrsoN

Abstract
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

Introduction
For personal use only.

of he th-vdrocarbons on the various adsorbents'

Experimental
Malerials
Nitrogen was obtained from tank nitrogen, by passing the tank
nitrogen
driers,
successively through chromous chloride solution (to remove oxygen),
and a liquid air traP.
r Manuscriqt receiaeil Noaember 9, 1948'
Contri.bul,ionfromtheDeparlmen|,ofChemislry,[Jniaersityo!A!!er!a'E-ilmonlon,Al,berto.
2 Grod.u.ote stu.ilent. Holdir at the time oJ a Buisory uniler
the National' Research Council'
of Canaila.
 206 :ANADTAN JIrRNAL oF RESEARCH. vol. 27, sEC. B.

Pure methane rvas obtained by the low temperature fractionation (liquid

air)
of Viking natural gas (9I/6 methane, 3.30/6 other hydrocarbons, S.I/s
nitrogen) using a column described by Freeth and Verschoyle (12).
Pure ethane was prepared by the hyclrorysis of ethyrmagnesium bromide.
The gas was passed through scrubbers ancl fractionated inlne same
way as
the methane.
Helium (98.2%) from the ohio Chemical company was used. prior to
storage in a bulb, the helium was purified by being pissed through
an activated
charcoal trap cooled in liquid air.
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

Ethylene (99.57d, propane (gg.9Td, normal butane (gg.gTd, and iso_

butane (997d, obtained from the ohio chemical company, were purified
further by fractionating each once at 40o to 50. c. beiow-their respective
boiling points.
commercial acetylene was purified in the same manner as the methane and
ethane. Prior to fractionation, the gas was passed through scrubbers to
remove soluble impurities (10, p. 162).
The charcoals were samples of a series of coconut charcoals of different
degrees of activation produced by the Standard chemical company
of
Montreal. The series is the same as that used by Lemieux and Morriso'
in whose paper some data on their properties are given.
it+;,
Method,
The adsorption isotherms of the gases were determined by means of an
For personal use only.

apparatus similar to the one described by Emmett (6, p. 3). The principal
volumes, other than the adsorption bulb, were water-jacketed to control
the
gas temperature. These temperatures were measured to 0.5o accuracy.
Pressures were measured to the nearest 0.05 cm. of mercury.
Samples of 0.20 to 0.25 gm. charcoal were placed in an adsorption bulb
of
about 2.5 ml. volume. The dead space was determined at 0" c. with helium.
when the adsorptions of the various gases were measured, temperatures
several degrees above the boiling points of the respective gases had to be
used. Although the theory of the B.E.T. equatitn suggests the boiling
points of the adsorbing gases as the adsorption temperaiurcs, the higher
temperatures were necessary because it was found that charcoal exhibited
the phenomenon of capillarv condensation at the lower temperatures. The
boiling points of the gases and the adsorption temperatures are given in
Table I.
The nitrogen bath contained liquid air. Acetylene, ethylene, and ethane
temperatures were obtained by immersing a liquid air finger in an acetone
bath; propane and isobutane temperatures were obtained tv arv ice acetone
baths, and normal butane by an ice-water bath.
-
A special constant temperature bath had to be devised for methane. A
pyrex test tube wound with 5 ft. of 20 gauge nichrome resistance wire was
fitted into a larger test tube. The assembry was immersed in liquid air,
 NAYANDMoRR.TSoN:ADS)RPTI}NoFHYDR}CARjBoNG.4sEsoNCHARC}AL20T

with the inner test tube as the cooling chamber for the
charcoal bulb' A
.opp".-.on.tantanthermocouplewasfixedtotheadsorptionbulbandaregu-
the temperature
latecl electric current through the nichrome I'v-ire controlled
to within + 1" C'
TABLE I
Borr,rNc PoTNTS -{ND ADSoRPTToN TDMPERATURDS oF GASEous ADSoRBATES

IemP. oI
adsorption bath,
oK,
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

83
Nitrogen
Methane ttL7 r28
t91 .4 198
Acetylene 183
Ethylene 169.3
184.9 195
Ethane 234
Propane 23r.0
272.8 273
n-Butane 265
Isobutane 263.0

Some attempt was made to avoid an arbitrary choice of the time of equilib-
rium. This arbitrariness arises from the porous nature of coconut step
charcoal'
and about
At first, an arbitrary time of 10 min. for the initial adsorption
five minutes for each subsequent adsorption step was used' However'
muchmoreconsistentdatawereobtai.'edwhentheestimatedequilibrium
made in 'some preliminary
times were based on experiinental measurements
For personal use only.

time studies of the adsorption process'

1 and 6 were
The rates of adsorption of several of the gases on Charcoals
measured, by following the pressure drop with
time after admission of the gas

E
l

E
E
b c

TIME

Frc. 1. Ty\ical pressure-tine curtte for t'he adsorpt'ion Proeess'

gave
to the adsorption bulb. A plot of the pfessure against time always
curves typically represented by Fig' 1'
pure
It is considered that the first part ab of the curve mainly"represents
diffusion into
adsorption and the second part bc represents adsorption after
 208 :ANADTAN J\rRNAL oF RESEARCE. voL. 27, sEC. B.

TABLE II
TruB elr,orroo (urlrurns) FoR ADsoRprroN EeurLrBRruM To BE ATTATNED

Steps in Charcoal No. 1

Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

adsorption
process CrH.* n-CrHrc isa-CrHro* CrHo CtHr* Nr*

I 30 20 15 10 l0 25 30
a 25
10 10 10 15 15 20
J 5 4
A 4 10 15
5 5 4 4 5 10
5 5 A
4 q
.) 5
6 5 ^ 4 J 5

Charcoal No. 6

1 25 30 30 20
2 15 20 20 10
.) 10
A
10 10 5
5 10 10 5
(
For personal use only.

5
6 5 q
5

The times allotted to the other gases and charcoals were estimated
by
interpolation.
Results
All the gases used in these experiments gave practically straight lines
when
the data u'ere plotted according to the linear form of Langmuir;s
equation;
.1
P/a" : p/a^
**
in which z" is the volume of gas (reduced to S.T.p.) adsorbed
at the pressure,
p' In Figs. 2 and 3, the plots of the equation are shown for nitrogen and
normal butane. These and similar prots were used to calculate
represents the volume adsorbed in a single molecular layer,
z-, which
and, in the case
of charcoal, the maxim.um adsorptio n 3). The experimental
e, ;. 292; results
are given in this form in Table III.
The volumes in Table III were not corrected for deviations from
ideality.
calculations on normal butane, which exhibits the largest deviations,
did not
give a measurable difference from the ideal case, probabry
because the adsorp-
tions are relatively large and the dead space is relatively
small (abou t 2.4 ml.
at S.T.P.) (7).
 NAY AND MORRISON: ADSORPTION OF HYDROCARBON G45ES ON CHARCOAL

j
(,
r
c
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

t
L

_a
a\

Frc. 2. Langmuir plols for adsorption of nitrogen on charcoal' series'

For personal use only.

E
U
0-
J
t
L
j

Frc. 3. Langmuir plots for adsorplion of normalbutane on charcoal series'

 210 :ANADTAN JqvENAL oF r':ESEARCH. voL. 22. sEC. B.

TABLE III
Maxruuu ADSoRprroN oF GASES rN MILLTMoLEs pER cRAM oF crrARcoAL

Charcoal number
Adsorbate

N: 5-08 6.72 7 .08 9.30 9 .98 11 .83 12.35 13.77

CzH, 5.50 6.01 6. 10 7 .72 8 .39 9.57 10. 09 11.04
CrHo 4. 03 5.06 5.18 6.70 7 .02 8.20 8. 63 9.46
CHr 4.28 5.82 6.12 7.75 8.38 9.76
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

10.18 10.96
CrHo 3.28 4.25 4.44 5.87 6.18 /.JJ 7 .79 8.37
CrHo r.77 2.73 2.88 3.96 4.39 5.47 5.59 6.13
a-CnHro 1 .36 2.20 2.39 J .IJ s.70 4.55 4.77 5.49
isa-CrHro 1. 18 1.59 r.78 2.70 3.46 4.02 4.24 4.66

Discussion
The nitrogen ?a values are different from the values given in the paper of
Lemieux and Morrison (14) for the sanre charcoal series. The differences
arise from sampling. Larger samples cannot be used in this type of work,
so that the differences are unavoidable. However, they are not significant
to the present paper.
Ad,sor pl,ion and.
Difusion
The preliminary time studies indicate that two processes are occurring,
For personal use only.

which for convenience will be called primary and secondary adsorptions.

Although not completely separable, these processes are apparently of suffi-
ciently different rates to be partially separated. The primary adsorption
is by far the more rapid process. The experiments were carried out at such
a pace as to make the secondary adsorption effect negligible, an objective
believed to have been achieved in the case of the more active Charcoals 5 to g.
For these charcoals, the points of the Langmuir plots were in each case on
one straight line. on the other hand, for the less active charcoals, and parti-
cularly for the larger adsorbate molecules (cf. z-butane in Fig. 3) there is a
tendency for the points to lie on a curve concave to the pressure axis. The
curvature may be explained by the secondary adsorption process which
would tend to be cumulative-increasingvo at a more rapid rate than would
have been given by the primary adsorption process alone.
The less active charcoals are also characterized by having a larger proportion
of smaller pores (11, 14). Thus the secondary adsorption process probably
involves diffusion into very small pores.
Orientation anil Area of Adsorbed, Molecul,es
In Fig. 4, the maximurn adsorptions of the straight chain hydrocarbon gases
are compared in the same way as Lemieux and Morrison (14) had compared
the maximum adsorptions of the aliphatic acids. The a^ for each gas is
plotted against the u- of normal butane, for each charcoal.
 NAYANDMORRISON:ADSORPTIONOFEYDROCARISONGASESONCHARCOAL2II

In the case of the adsorption of the aliphatic acids from solution on the same
charcoal,
charcoal series, similar plots gave parallel lines at low activation of the
gases
and converging lines at high activation. In contrast, the hydrocarbon
give diverging lines with increasing charcoal activity'
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16
For personal use only.

246
VM, MILLIMOLES PER GM.

onil Prooane
FIc. 4. Comparison of a^ , the marimum ailsorptions, oJ methane' etkane'
with thot of normal bul'ane.

charcoal surface.
 212 .ANADIAN J\zRNAL oF REIEARCH. voL. 22, sEC. B.

rption are les

ntation. ies
p.1246), of
equal to
z- (N")
::
?-(9as)
X 16.2,

. The average molecular adsorption areas are based

The exclusion of the other values is suggested by
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

TABLE IV
Mor,ocr,'r,en ADSoRprroN AREAS oF GAS MoLEcuLES rN seuARE ANGSTR.MS

Charcoal number
Adsorbate Average

CzHz 1( n 18.8 19.5 19. 3 20.0 19.8 10R+ 0.3

20.2
CzHa 20.4 22.r 22.5 23.0 23.4 23.2 .23.6 23.1 + 0.3
CHr 19.2 18.7 18.7 19.1 19.3 19.6 19 .7 20.4 + 0.4
C:Ho 25.1 25.6 25.8 2s.7 26.2 26.1 25.7 26.7 25.9 + o.4
CrHe 46.5 39.9 39.8 38. 36.8 35. 3s .8 36.+
1 1 36.0 + 0.6
For personal use only.

n-Cftln 60. 5 49.5 48.0 46.6 43.7 42 1 42.0 +0.7 42 .1 + 0.8

lso-C+Hro 69.7 06. 5 64.4 55.8 46.8 t1 1 Ai /
47 .9 + 0.4
 NAvANDMORRISON:ADS)RPTI}NoFEYDR}CAR]goNGAsEsoNCHARC1AL2|3
that
increase with charcoal activation (acetylene and ethylene)' (2) those
(propane
remain the same (methane and ethane) and (3) those that
decrease
values are reached at higher
and the two butanes). In every case, steady
probably the explanation for the development of the
charcoal activations.
steadyvaluesisthatallthegasesincludingnitrogenreachaboutthesame
internal surface for each of Charcoals 5 I o 8'
TheresultsinTablelVsuggestthatforthehydrocarbonmoleculescon-
(1) acetylene and ethylene'
sidered here there are three grorrp. of thicknesses,
(2) methane and ethane and (3) propane and the two butanes'
It would
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

appearthat,comparedwiththenitrogenmolecule'thefirstgroupisthinner'
the second about the same, and the third thicker. If
the average molecular
adsorptionareasinTablelVareincorrect,itmaybearguedthatthecorrect
molecular adsorption areas would be less for molecules
thicker than nitrogen
and greater for molecules thinner than nitrogen'
ThemolecularadsorptionareaswillprobablyberelatedtovanderWaals the
for some of
areas. Pauling (15, p. 189) gives the van der Waals radii
gases discussed here. The van der Waals
and not to the thickness of molecules in
layer). Nevcrthelcss' some indica of
van der Waals diameters. Thus, n gen
and methane and ethane about 4 A ;k'
the same van der \Vaals radius as the
For personal use only.

end contact of the molecules' Howe

with its nonrotation about the double
for ethane and a thickness similar t
supPort this suggcstion.
Further,ifpropaneu''eretolieasflataspossibleonthecharcoalsurface'
ethane' Experi-
it would have the same molecular thickness as methane and two butanes.
as the
mentally, however, propane has about the same thickness
forces holding propane to the charcoal surface are
Apparently, the adsorpiion
not sufficient to restrict free rotation aborrt it" single
carbon-carbon bonds.

Comparisons may fg
tion areas (all referred
lene (8), normal butane
For normal butane ad
borators (1) found values of 39' 7, 4l'4' 43'7 and
45'1 A'' On the same
from 37'8 to.40'6^Ar'
adsorbents they found areas for l-butene ranging
butane and 22'2 A2
Emmett and Cines (S) oUtuitt"a areas of 49 it" for normal of 52 and
found (9) areas
for acetylene, both on porous glass' Further' they A2 on a
and an area of 42
41 A2 for normal butane on two samples of zincdust,
Harkins and Jura
carbon black. From their work with several absorbents, for
A'fot normal butane' Similarly' from results
(13) assign
".rnru"^of-i+ and Emmett (5) assign a
various metal foils and powders, Davis, DeWitt'
;i;;i;;. s AJ ro. .,o.-.1 butane and' 42' 7 A' for l-butene'
 214 CANADIAN JOURNAL OF RESEARCH. VOL,27, SEC. A.

References
1. BrrnB, \.^4,, Pgllp", M. H., Srurru, W. R., and \,Vnronr.r.,
69:2294. 1947.
C. B. J. Am. Chem. Soc.
Can. J. Res. Downloaded from www.nrcresearchpress.com by MICHIGAN STATE UNIV on 10/16/16

2' BnuN-c.u'n, S. The adsorption of gases and vapours, Vor. 1. princeton

Princeton, N.J. 194J. univ. press,
3' BnuNAusn, s. and Euunrr, p. H. J. Am. chem. soc. 59 :26g2.
rg37.
4. BnuNeuBn, S., Euuorr, p. H., and Tnr,r,nn, E.
J. Am. Chem..Soc. 60 :309. 193g.
5' D'n'vrs, R' T., Duwrrr, T' w., and Euunrr, p. H. phys. & colloid.
1947. J. chem. 51 :1232.
For personal use only.