LIGHT ABSORPTION STUDIES

PART XIV. T H E ULTRAVIOLET ABSORPTION SPECTRA OF PHENOLS1

ABSTRACT
A spectral analysis of benzene derivatives is extended to s ~ ~ b s t i t ~phenols.
~ t e d The disc~lssion
emphasizes hydrogen bonding and steric effects which occur in phenols. Some of the phenolic
absorption bands are ascribed to locally escited states.
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

INTRODUCTION
T h e effect of the hydroxy group on the electronic spectra of benzene derivatives is of
interest for a number of reasons. First, phenols are known to give rise t o various forms
of hydrogen bonding, and hence the investigation of spectral effects which may be
ascribed to hydrogen bonding affords further information concerniilg the nature of the
hydrogen bond. Next, compounds which m a y be regarded as substituted phenols, for
example flavones, frequently occur a s natural products, and hence a n understanding of
pheilolic spectra assists in the identification and quantitative estimation of such naturally
occurring compounds. Lastly, previous papers in this series have attempted to analyze
the spectra of benzene derivatives in terms of mesomeric and steric interactions, and
i t is therefore of interest t o test these hypotheses on the observed spectra of phenols.
For personal use only.

T h e ultraviolet spectra of a number of phenols have previously been investigated by

other worlters (1, 2, 3, 4, 5 ) , but whenever convenient we have redetermined the spectra
in order to ensure near-identical conditions.

T H E SPECTRUM O F PHENOL
Phenols generally exhibit two absorption bands in the spectral region under investi-
gation, t h a t is, between 200 and 360 mp. T h e band a t longer wavelength is referred to
a s the C-band, and the band a t shorter wavelength a s the B-band. This assignment
corresponds to the assignment for anilines and is based on similar reasoning (cf. ref. 6).
T h e d a t a for phenol are listed in Table I.
T h e bands are somewhat altered with change of solvent, and i t has been shown by
Coggeshall and Lang (3) t h a t some of these spectral changes do not depend on stable
hydrogen-bonded complexes, although there is evidence t h a t some of the observed
spectral changes are caused b y some type of hydrogen bonding (5). Various forms of
intermolecular hydrogen bonding are certainly possible, involving both the hydrogen
atom of the hydroxy group and the oxygen atom. T h e wavelength displacement between
the spectra in the vapor phase a n d in cyclohexane solution must also result from some
solute-solvent interaction; but the latter interaction is unlikely t o be associated with
hydrogen bonding because similar displacements are observed for phenols and anilines
(cf. ref. 6). In more concentrated solutions of phenol in cyclohexane, yet other spectral
effects occur, presumably caused by intermolecular hydrogen bonding between phenol
~nolecules(see ref. 4).
T h e fine structure of the C-band for phenol or anisole in cyclohexane or similar inert
media is fairly characteristic of these chro~nophores,and can often be recognized in
substituted, including sterically hindered, phenols (see also later section). T h e fine
structure disappears if the phenolic species no longer contributes appreciably t o the
liVIanicscript receiued Nouenzber 10, 1956.
Contribution front the Memorial Uniuersity of Newfoundland, St. John's, Newfoundland.
Can. J. Chem. Vol. 37 (1959)
 DEARDEN A N D FORBES: LIGHT ABSORPTION STUDIES. XIV 1295

TABLE I
Absorption maxima of phenol in various solvents
(Values in italics represent inflections in this and subsequent tables)

B-Band C-Band
Solvent Xrnnx(m~) ernnr Arnsx(m~) h2.~ Source
Vapor phase 205 t
I ;:: t 'L
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

Cyclohexane 210 6000 1400 *

(B-band concn. 2200
= 2.3 XlO-" moles/].) 276 2100
Ether

Ethanol

Water 210 6000 1500 *

1250
0.1 N HCI
For personal use only.

N NaOH
Conc. H,SO( 9500 261
ca. 27'1
278-279
*This series of papers. tIntensities not determined.

observed spectrum; as for example in the C-band of m-nitro- or m-amino-phenol, where

the compound absorbs preferentially as nitrobenzene or aniline (see Table 11). T h e fine
structure of the C-band also disappears for intrainolecularly hydrogen-bonded molecules
such a s salicylaldehyde (see Table 11), b u t the fine structure is noticeable for a number
of orthosubstituted phenols even in ethanolic solution (see Table 111). The most probable
explanatioil seems to be that this fine structure is caused by vibrational sublevels and
that a hydrogen bond may obscure this fine structure. Hence, the fine structure is
frequently reduced if the phenol is associated with a solvent molecule by means of
an intermolecular hydrogel1 bond (see Table I ) or if the phenol molecule is intra-
molecularly hydrogen-bonded, but the fine structure does become evident if steric
interactions hinder the formation of a hydrogen bond (see sectioil on orthosubstituted
phenols).
In sodium hydroxide solution the phenolic spectrum is radically altered, and the
spectrum is assumed to correspond to the phenolate ion. For phenol in basic medium,
compared with cyclohexane solution, a wavelength displacement of 24 mp is observed
for the B-band and a displacement of 16 mp is observed for the C-band (see Table I). As
expected, similar wavelength displacements are not observed for the spectrum of anisole
(see ref. 7).

T H E SPECTRA O F PARASUBSTITUTED PHENOLS

T h e spectra of substituted phenols are listed in Table 11.
Table I1 lends support to the naming of the B-bands, that is, as bands corresponding
to the so-called B-bands in benzene, acetophenone, nitrobenzene, benzoic acid, and
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

TABLE I1
Absorption maxima of substituted phenols

Para isomer Meta isomer Ortho isomer

--

B-Band C-Band Phenolic B-band 2nd B-band C-Band Pllenolic B-band 2nd B-band C-Band
-- - --- -
For personal use only.

Substituent Solvent hrnnx(lnp) Emnx hrnnx(mp) Emnx hmnx(m~)Emns hmax(mp)Emax X~nnx(mp) Emax hmax(mp) Emax hm:,x(mp)~m:,x hmux(mp) Emnx

-OH Cyclohexane 223 4400*

/
(
ca. 286
289
292
ca. 299
2600* ca.217
2800
2900
2200
6800* -

-1 ca.268
273.5
279.5
1660*
2150
1800
214 6800 - ca. 270 2200

-0CHa Cyclohexane 224 7000 j ca. 286 2900

288 3150
291 3100
298 2350
-NHz Cycloherane* 235 234 6800 284 2300 235 7000 288
Water 229 229 G400 278 1900 229 6150 281
pH 3 or
0.1 A7 HCI 218.5 217 6300 ca. 918 269 1900 210 6500t
616q { ca. 275 1700
-CHs Cyclohexane 220 214 6100
{ ::: 1600
1850
213 7000

-C2Hn Cyclohexane 7250

I ca. 267
270
273
276
282
1400
1600
1800
2200
2000
ca. 265

{ :::
-CH(CHz)

-C(CHs) s
2 Cyclohexane

Cyclohexane
6850

6800
I
{
ca. 267
ca. 270
ca. 272
275
281
ca. 268
274
280
1560
1600
1700
2100
1900
1GOO
2100
2050
1 :::
ca. 264

{
ca. 265
268.5
275
 DEAR DEN A N D FORBES: LIGHT ABSORPTION STUDIES. XIV 1297

N W
O O r-
N N N
N LO-m
mr-r-
W N N
(om01 m a -
w.
(or-r- o r - m
W N N N N N
w
N
m
I
r-
r-
m
Nu2
a
m m
-
W
a
h h
0 0
m m

-u
-
- -
-
Z O O
w o o
0 0
0 0
X 88 8 88
2 2
4 4
2 F% 2 %2
Y m
ma, o h $ $E 2 N
I m
W
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

-N NN N
N w NwN N N N N h N

0 0 0 0 0 0 0 0
- *
- =
o w 0 w o w 0 0 8 8s 8 88
"3'2'2 Z E R E2 w
a 0e 4r- n 2 %E

-- -
w
W W N
(oar-
W N O
O r - h
WNrn N h 2 2% N ""
W N N N N N
(Or-h
N N N O
m 0
4 m m m k: E Z
d d
-+
For personal use only.
 1298 CANADIAN JOURNAL O F CHEMISTRY. VOL. 37. 1959

benzaldehyde, although the excited state of phenol may represent a locally excited state
(see below). This follows because the gradual displacement of the band for a series of
compounds such a s phenols indicates t h a t the main absorption band in phenol (and
iilcideiltally aniline) is of similar nature for the compounds listed in Table 11. Further,
as shown in Parts X I and XI1 (10, l l ) , this same band, for example in P-hydroxyaceto-
phenone, call likewise be related to the B-band in acetophenone. This confirms the
suggestioil t h a t all the above-mentioned absorption bands m a y be described a s B-bands.:"
T h e B-band in phenol is shown to be displaced t o loilger wavelength on parasubstitu-
tion. I i the parasubstitueilt is electron-withdrawi~lg,this displaceinent is greater, quali-
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

tatively related to the electron-withdrawii~g effect of the substituent, and the usual
high-intensity band corresponding t o a n extended coiljugated system is obtained. T h e
wavelength displacements with respect to pheilol in cyclohexaile are parallel t o the
displacements in the aniline series (6). T h e qz~a7ttitativerelation observed between spectra
of parasubstituted nitrobenzenes, acetophenones, and beilzaldehydes (10, 11) is, however,
absent for the displacemeilts observed between parasubstituted ailisoles and phenols,
since in the latter example the straight line plot of the wavelength displacements does
not pass through the origin (see followiilg paper).
As will also be shown in the followiilg paper and elsewhere, this difference can be
rationalized by assuming t h a t the B-band in phenol involves transitions which should
be associated t o a n appreciable extent with locally excited states of type I. In support
of this hypothesis the following m a y be cited: (i) the B-band of phenol a t 210 m p in
For personal use only.

cyclohexane is displaced b y about 7 mp from the corresponding band in benzene, a

wavelength displacement similar t o t h a t associated with the short-range interaction
of the hydroxy group (see sectioil on metasubstituted phenols); (ii) if phenol is
assumed to absorb a t longer wavelength, because hypothetical structures of type I1
contribute inore appreciably to the electronic excited state, the above-mentioned relation-
ship between wavelength displacemeilts in parasubstituted anisoles and phenols becomes
more nearly quantitative (see following paper); (iii) steric effects in the B-band of
phenol are different froin those which would be ailticipated for a band associated with
trailsitioils to excited states of type I1 (see section on orthosubstituted phenols); (iv) the
anomalous nature of the wavelength displaceilleilts for phenols has previously been
noted b y Doub and Vandellbelt (8) ; and (v) the bathochromic wavelength displacement
obtained for the spectra of inetasubstit~ited phenols compared with the spectrum of
phenol when the metasubstituent is electron-ivitl~dra\i~ingcan be explaiiled since a n
electron-withdrawii~gsubstituent has been assumed to facilitate resonance forms of type
I1 (see ref. G) and this therefore call accouilt for the ailoinalous batl~ochroi~zicwave-
length displacenleilt (see the cliscussion of the spectra of inetasubstituted phenols).
*Alter?tatioely, it has been proposed tlzal tlte B-band of c o ~ t p o ~ ~ lti dk es, for exa7~tple,nitroben,zene sl~oztldbe
related to the C-bond of bcttze?te, that i s to the baud occurri~tgi?t be?tze7te crt ca. 250 ?up. Wepster ( l d ) , ilt the
course of a disczrssiort 072 steric effects, szcggests that this latfcr argz~?itentis sz~pportedbeca~rsefor sterically ltirtdered
~titrobenze~ces the absorption band irt Izexane rentai?cs ?tear 250 nip, that i s , ?tear tlte C-bazd absorptiort of ben-
zene. Ozrr i?itrrpretation i s to regard t h i s band i?t sz~bstitz~ted~titrobeitse~tesa s a nitrobenze7tc B - b a ~ z d ,since the
latter band also occzirs ?tear 250 ntp. 5TJe bclieoc tlcat o z ~ rargn~rizentreceives szrpport front tlze absorptioi~~ i t a x i m a
of sterically hindered p - n i t r o a ~ t d i r ~ ei sl t ct/l.anol. Tltcse 717axi717a are all reported to occur nzar 380 nlp ( I d ) , that
i s , ?tear the i17axiv1al absorl>tion of p-nitron?tiline, a r ~ dtlte ? i ~ a x i m for a iltcsc exa?iaples d o ?tot tltereforc revert to
a atasinzn~ntnzar 250 nip.
 DEARDEN A X D FORBES: LIGHT ABSORPTION STUDIES. XIV 1299

I t may be noted that this hypothesis assumes that the absorption of parasubstituted
phenols such as p-cresol, hydroquinone, or p-bromophenol may still iinvolve locally
excited states of type I with both substituents exerting only a secondary effect. However,
in compouilds of this type, structures involving appreciable resonance interaction with
a t least one substituent seem more probable because the second substituent facilitates
additional electronic interaction. The relevant structures may be represented by resonance
forms of types I11 and IV, and coilsideratioil of structure IV also suggests that structure
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

V cannot be ruled out. The actual contributing structures will depend on the relative
over-all electronegativities of the two substituents in the ground and excited states. For
phenols in basic media, a different type of resonance interaction presumably occurs in
the excited state because of the replacemeilt of the phenol group by the phenolate
ion. Since this resonance interaction will be reflected in the maximal wavelengths i t
follows that the mesomeric interaction between the hydroxy group and the benzene
ring is less than the mesomeric interaction between the amino group and the benzene
ring. Extending the argument one step further, the maximal absorptions suggest an
order of mesomeric interaction PhO- > P h N H ? > PhOH > PhNH3+.
For personal use only.

The C-band shows the expected changes if the parasubstitueilt is electron-withdrawing,

namely, with increase of mesomeric interaction the C-band usually becomes progressively
less pronouilced (see Table I1 and Ref. 13). If the parasubstituent is electron-donating,
the fine structure of the C-band is frequently enhanced, the effect being most pronounced
for the spectra of p-methyl- and p-fluoro-phenol in cyclohexane. I t is worth noting t h a t
the fine structure of the C-band in these latter compounds is coilsiderably more pro-
nounced than for the corresponding meta isomers (see Table 11). This is surprising
since increased molecular symmetry is generally assumed t o simplify the absorption
spectra, and one would therefore have expected the fine structure of the C-band to be
less pronounced in the para isomer than in the meta isomer.

T H E SPECTRA OF METASUBSTITUTED PHENOLS

In metadisubstituted beilzeile derivatives direct resonance interaction is ruled out by
definition, and hence we would expect two B-bands correspondiilg to the two mono-
substituted benzene derivatives (cf. refs. 6, 10, 11). Both these bands may be modified
because of the other substituent, either by means of secondary short-range electroilic
interactions (cf. ref. 7), or possibly because of the buttressing effect (cf. refs. 14, 15). I t
is, of course, possible that one of the two B-bands may be obscured, either because of
overlap by the other B-band or because of A-band absorption (cf. ref. 13), and these
factors may also prevent a reliable band assignment for some examples. On the other
hand, characteristic properties like the fine structure of the phenolic C-band assist in
the band assignments.
These generalizations are illustrated for phenols in Table 11. If the non-phenolic sub-
stituent is electron-withdrawii~g,it is found t h a t the non-phenolic B-bands are normally
displaced to longer wavelength on account of the secondary interactioil of the hydroxy
group, and this displacement is of the order of 7 mp for a number of compouilds (cf.
ref. 7). This wavelength displacement of ca. 7 mp suggests that the hydroxy group tends
 1300 CANADIAN JOURNAL O F CHEMISTRY. VOL. 37, 1959

to facilitate the absorption of the non-phenolic chromophore. The hypothesis may be

illustrated by structures of type VI, by analogy with similar structures used in previous
arguments (cf. ref. 10). In basic media the secondary interaction is considerably increased,
because of the replacement of the hydroxy group by the 0--entity, and bathochromic
wavelength displacements of up to 25 mp (for m-nitrophenol compared with nitroben-
zene) are observed.
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

As has previously been noted by Lemon (2, cf. also ref. 7),* the effect of alkali on
the spectra of hydroxybenzaldehydes and hydroxyacetophenones is greatest in the para
isomers.
Our data also show that for the B-bands the displacement in the para isomers in
basic media is much more pronounced. For example, the phenolic B-band of m-nitro-
phenol on passing from neutral to alkaline solution shows a wavelength displacement
of 23.5 mp (see Table II), comparable with the displacement of 24 mp for phenol (see
For personal use only.

Table I). In the para isomer, presumably because appreciable primary resonance inter-
action occurs, the displacement of the B-band, caused by the addition of alkali, is ca.
90 mp (see Table 11). These observations may be of some practical importance. For
example, if the B-band of an acetyl-substituted phenol on changing the solvent from
neutral to basic medium does not show a bathochromic wavelength displacement greater
than ca. 25 mp, it may be assumed t h a t no uninhibited, direct mesomeric interaction
occurs between the hydroxy and acetyl groups.
The effect of the second substituent on the phenolic B-band is also illustrated for
some of the compounds listed in Table 11. Unfortunately, the relevant absorption band
frequently occurs a t too short a wavelength to be identified with certainty. However,
for a few compounds the band assignment presents little difficulty, as for example, in
m-nitrophenol where the phenolic absorption is appreciably displaced to longer wave-
length (see Table 11). This latter displacement has previously been ascribed to the
nitrosubstituent facilitating the phenolic absorption (10). This bathochromic displace-
ment for the phenolic B-band in compounds like m-nitrophenol may be related to the
changes in the integrated absorption intensities for the infrared hydroxyl absorption in
phenols, since the integrated absorptioil intensity for compounds like m-nitrophenol is
greater than for phenol (17, 18), and since increased integrated absorption intensities,
like bathochromic wavelength displacements in the B-band, may be related to increased
m e s ~ m e r i cinteractions.
I t should be noted t h a t when the metasubstituent is electron-withdrawing, the phenolic
B-band appears sometimes to be moved to longer wavelength (see Table 11), whereas the
anilinic B-band under these conditions is moved to shorter wavelength (see ref. 6). More-
*Lemon arrives at this concllision, although for the ortho and meta isomers he considers what we believe to be
the C-band, whereas for the para isomers he considers the B-band. Another example of what we believe i s a n
incorrect assignment i s provided by the reported data for the intense o-hydroxyacetopheno?le band near 315-330
m p , which has been described as a B-band (K-band, see ref. 16), whereas we wozrld regard it as a displaced
C-band.
 DEARDEN AND FORBES: LIGHT ABSORPTION STUDIES. XIV 1301

over, these wavelength displacements are parallel in the correspondillg infrared bands. For
example, in m-nitroaniline compared with aniline (19), the N H 2 stretching vibration is
moved to higher frequency, that is to shorter \vavelength, whereas the OH stretching
vibration in m-nitrophenol compared with phenol is moved to lower frequency. These
anomalous wavelength displacements for phenols can be explained by assuming that
the B-band of phenol occurs a t unusually short wavelength because of the contribution
of locally excited states of type I (see section on parasubstituted phenols). In this way,
the above-mentioned comparisons involving phenol are rendered invalid, in the sense
that metasubstituted phenols compared with hypothetical phenol molecules containing
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

more similar contributions of locally excited states than aniline may still afford the
anticipated hypsochromic wavelength displacement.
If the above assignments are accepted, examination of the relative intensities of the
two B-bands suggests that the phenolic B-band absorption usually occurs preferentially.
For example, in m-nitrophenol, compared with the two monosubstituted parent com-
pounds, the nitrobenzene B-band absorption decreases, whereas the phenolic absorption
increases in intensity. A similar relative intensity increase is observed for the aniline
B-band in m-nitroaniline ( G ) , although the relative increase in the intensity of the
phenolic B-band is less than for the anilinic B-band. These observations may be rational-
ized by a converse of the mechanism postulated to account for the increased extinction
coefficient in associated benzoic acid inolecules (see ref. 20). There it was shown that
an electron-withdrawing substituent, in the absence of direct resonance interaction,
For personal use only.

gives rise to an increased maximal extinction coefficient. Likewise, in metasubstituted

anilines or phenols the nitro group may be assumed to increase the probability of the
phenolic or anilinic transition, and conversely, the m-hydroxy and the m-amino sub-
stituents may be assumed to have the effect of relatively weakening the nitrobenzene
absorption, the effects being inore pronounced for the meta-amino substituent, which
donates electrons more effectively.
Some of the C-bands in metasubstituted phenols show the expected changes. For
example, if the metasubstituent is electron-withdrawing the C-bands are enhanced
compared with the C-bands in para isomers (cf. ref. 13).

T H E SPECTRA O F ORTHOSUBSTITUTED PHENOLS

The B-bands of the ortho isomers frequently resemble the B-bands of the meta isomers.
For example, for o-nitrophenol in 0.1 iV NaOH where intramolecular hydrogen bonding
is ruled out, the nitrobenzene band a t 282 mp, E = 4300, is similar t o the nitrobenzene
B-band in the meta isomer which occurs a t 291 mp, E = 4500 (9). The phenolic B-band
is considerably reduced in intensity and occurs in the ortho isomer as an inflection a t
250 mp, E = 5000, whereas in the meta isomer it occurs a t 251.5 mp, E = 11,000 (9, and
see Table 11). For o-nitrophenol in basic solution there is therefore no evidence for
appreciable resonance interaction. Similar spectral effects are observed in other examples
listed in Table 11, and the examples of 2,4- or 2,4,6-substituted nitrobenzenes when an
ortho-alkyl substituent does not contribute to the direct resonance interaction, have
previously been mentioned in this paper in the discussion of parasubstituted phenols.
Effects of this type may also be noted in the B-bands of a number of parasubstituted
phenols in neutral solution when, on introduction of two t-butyl groups in the ortho-
positions, no appreciable wavelength displacement is observed, but a decreased extinction
coefficient indicative of steric interactions is obtained (21).
 1302 CXSADI.AK J0URN.-IL O F CIIEIMISTRY. VOL. 37, 1959

This absence of resonance interaction may be ascribed to steric interactions, and

steric interactions are also indicated by a number of other physical properties, such
as the decreased iilterinolecular hydrogen bond formation ill sterically hindered phenols
(3, 15, 22, cf. also 23), and more generally by the failure of sterically hindered phenols to
exhibit ilormal phenolic properties (cf. for example, ref. 21). These observed steric
effects maj7 be associated, a t least partly, with the two unshared electron pairs on the
oxygen atom, since a steric effect caused entirely by the more distant hydrogen atom
of the hydrosy group xvould not account satisfactorily for the observed steric effects
(cf. ref. 24, p. 360). The B-band in orthosubstituted phenols does not normally undergo
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

a pronounced waveleilgth displacement even if the orthosubstituents are bulky as in

diortho-t-butylphenol (see Table 111). T h e absence of this spectral effect, characteristic
of steric inhibition of resonance in many other conjugated systems, has already received
an explanation (see section on parasubstituted phe~lols,which tentatively ascribes the

TABLE 111
Absorption~masirnaof orthosubstit~~ted
phenols in cyclohesane and ethanol

Cyclohexane Ethanol
Orthosubstitnents Xrnns(n1~) C ~ : L X Atr,.,x (111~) Enlnx

H,H (B-Band) 210 6000 218.5 6000

For personal use only.

1400
2200
2100
{ ca. 266
271
276

( ca. 264 1500 271

1 270
276
2100
2000
( co. 276

{
[
ca. 263
268.5
275
1400
2050
2100
( :::
CH3,CH3(B-Band) 211-212 7900 ca. 214 8600

CH(CH3)z,CH(CH3)2 (B-Band) 211 9200 cn. 211, 8000

CH (CH3)?,CH(CH3)?(C-Band) 1900 270.5 1850
1950 ( ca. 275 1700
C(CH313,C(CH3)3
(B-Bancl) 21 1 7500 ca. 212 7600

ca. 265 1550 274.5 2350

2350
2450
{ ca. 280 2000

212 8700 ca. 214 8450

H,Br (C-Band) 1750 275.5 2750
2700 ( CU. 281 2360
 DEARDEN A N D FORBES: LIGHT ABSORPTION STUDIES. XIV 1303

phe~lolicabsorption to transitions involving locally excited states). Steric interactions

are, however, indicated in the ultraviolet spectra by the absence of the usual wavelength
displacements between solutions in cyclohexane and ethanol. T h a t is, whereas between
solutions of phenol in cyclohexane and ethanol a wavelength displacement of 8 mp is
observed in the B-band (see Table I ) , no such displacement is observed for solutions
of strongly sterically hindered phenols (see Table 111 for the spectra of various ortho-
substituted phenols in cyclohexane and ethanol). Our explanation, follo\ving the suggestion
of Coggeshall and Lang (3), is to ascribe a reduced wavelength displacement to steric
interactions inhibiting solvent-solute interactions between phenol and solvent nlolecules.
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

Yet another spectral effect which may be associatecl with steric interactions is the
progressive increase in extinction coefficient of the C-band for the spectra of ortho-
substituted phenols in alltaline solution with increasecl size of the orthosubstituent (cf.
ref. 13). The relevant C-bancl data are listed in Table IV, and Table IV also shows t h a t
the B-baiids are not appreciably affected under these conditions.

TABLE IV
Absorption maxima of orthosubstitutecl phenols in allcaline solution
(iV aqueous NaOH)
For personal use only.

*Solutiol~contains ca. 1% ethanol.

Next, in neutral solution, there is evidence for intramolecular hyclrogen bonding, and
the spectral effects ascribed to hydrogen bonding are more pronounced for orthosubsti-
tuted phenols than for the corresponding anilines (cf. ref. 6). The evidence is obtained
by comparing the B-bands of a number of orthosubstituted phenols with the B-bands
of the lneta isomers, where the former B-bands show a bathochromic wavelength dis-
placement accompanied by an increased absorption intensity.
Since these spectral changes are much more pronounced in orthosubstituted phenols
than in the corresponding anisoles (see following paper) it is assumed that the spectral
changes must be caused, a t least partly, by intramolecular hydrogen bonding. Evidence
for an intramolecular hydrogen bond in, for example, o-nitrophenol is also provided
by the appreciably displaced hydroxyl band in the infrared region a t 3240 cm-I for
o-nitrophenol compared with the band a t 3590 cm-' for the meta and para isomers
(23, 26, cf. also refs. 27, 28, 29). Further evidence is provided by the C-H bending
frequency a t 748 cm-I for o-nitrophenol (30).

EXPERIMENTAL
The ultraviolet absorption spectra were determined by standard methods using a
Unicam SP5OO spectrophotometer. For each compound a t least two independent sets
of observations were made. Some of the spectra have previously been described in
this series.
 1304 CANADIAN JOURNAL O F CHEMISTRY. VOL. 37, 1959

The pheilols were mostly commercial materials, purified by distillation or recrystalliza-

tioil until their boiling points or melting points showed thein to be sufficiently pure. The
solvents used were spectroanalyzed cyclohexaile (Fisher) or commercially available
absolute ethanol. T h e spectra of the amiilopheilols and dihydroxybeilzenes were deter-
mined as quickly as possible to prevent oxidation. With this end in mind, the air in
the flaslts containing these solutioils was also replaced by nitrogen.

ACICNOWLEDGRIENTS
T h e authors are greatly indebted to Miss Floreilce Jackmail for very competent
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 14.139.187.67 on 02/13/19

techilical assistance in the deterinillation of some of the spectral data. They also wish
to thank the Fisheries Research Board of Canada for fillailcia1 support, since the major
part of this work was carried out under a contract between the Fisheries Research
Board of Canada and Memorial University of Newfouildland. T h e authors further
acknowledge the general assistailce of the National Research Council in support of
these studies.

REFERENCES
1. BURAWOY, A. and CHAMBERLAIN, J. T. J. Chem. Soc. 2310 (1952).
2. LEMON,H. W. J . Am. Chem. Soc. 69, 2998 (1947).
3. COGGESHALL, N. D. and LANG,E. M. J. Am. Chern. Soc. 70, 3283 (1948).
4. ICEUSSLER,V. VON. Z. Elektrochem. 58, 136 (1954).
5. NAGAKURA, S. and BABA,H. J. Am. Chem. Soc. 74, 5693 (1952).
For personal use only.

6. FORBES,W. F. and LECKIE,I. R. Can. J. Chem. 36, 1371 (1958).

7. FORBES,W. F. and RALPH,A. S. Can. J. Chem. 36, 869 (1958).
8. DOUB.L. and VANDENBELT. T. M. T. Am. Chem. Soc. 69. 2714 (1947).
9. DOUB; L. and VANDENBELT; j. M. j. Am. Chem. SOC. 71; 2414 j1949j.
10. FORBES,W. F. Can. J. Chem. 36, 1350 (1958).
11. DEARDEN, J . C. and FORBES,W. F. Can. J. Chem. 36, 1362 (1958).
12. WEPSTER,B. M. Progress in stereochemistrv. Vol. 2. Edited by, W. Klvne and P. B. D. de la
Mare. Butterworth Scientihc Publications, London. 1958. pp. -112 and -110.
13. FORBES,W. F., MUELLER, W. A., RALPH,A. S., and TEMPLETOX, J . F. Can. J. Chem. 35, 1049
(1957).
F O R B E S , ' ~F.. and MUELLER,W. A. J . Am. Chem. Soc. 79, 6495 (1957).
SEARS,W. C. and KITCHEN,L. J. J. Am. Chenl. Soc. 71, 4110 (1949).
KANEXIWA, N. J. Pharm. Soc. Japan, 76, 253 (1956).
17. BROWK,T. L. J. Phys. Chem. 61, 820 (1957).
18. STONE,P. J. and THOMPSON, H. W. Spectrochim. Xcta, 10, 17 (1957).
19. KRUEGER.P. T. and THOMPSON. H. W. Proc. Rov. Soc. A. 243. 143 (19571.
20. FORBES.W. F: and TEMPLETON. T. F. Can. 1. Cilem. 36. 180 (1968).\ -
21. COHEX,'L.A. J. org. Chen~.22: 1333 (1957f
22. BOWMAN, R. S., STEVENS,D. R., and BALDWIN, \Ir.E. J . An]. Chem. Soc. 79, 87 (1057).
23. CARLTON, J. K. and BRADBURY, W. C. J. Am. Chem. Soc. 78, 1069 (1956).
24. WEPSTER.B. R/I. Rec. trav. chim. 76. 357 (1957).
25. HOYER,H. Ber. 89, 146 (1956).
26. GINETTI.Y. Naturwiss. 41. 333 (1954).
27. FRANCK,' B., I-IORM~~NN, H.,'and SCIIEIBE, S. Ber. 90, 330 (1957).
28. FLEET, M. ST. C. Spectrochim. Acts, 10, 21 (1957).
29. FRAXCEL,R. J. J. Am. Chem. Soc. 74, 1265 (1952).
30. IC~oss,R. D., FASSEL,V. A,, and M.~RGOSHES, %I. J . Am. Chem. Soc. 78, 1332 (1956).