Infrared studies with substituted orftho-bromophenols—part II

Spectrochimica

Acta, 1963, Vol. 19, pp. 463 to 474. Pergamon PressLtd. Prlntedin NorthernIreland

Infrared studies with substituted ortho-bromophenols-Part II. The effect of the solvent on the hydroxyl stretching frequency* IVOR BROWN,

G. EGLINTON

The University,

and M. MARTIN-SMITH?

Glasgow W.2,

(Receaved

26 July

Scotland

1962)

Abstract-Selected o&o-bromophenols have been studled to determme the effect of the solvents hexane, carbon tetrachlorlde, chloroform, carbon dlsulphide and acetonitrlle upon the frequencies, half band widths and apparent molecular extinction coefficients of the hydroxyl stretching absorptions. The “an&u-bonded”, OH . - * - Br, hydroxyl frequencies undergo smaller solvent shifts than those of the free hydroxyl groupings. The effect of temperature on the hydroxyl stretching absorptions of 2-bromophenol and 2,4-dibromo-6-t-butylphenol m tetrachloroethylene is reported and the hydroxyl stretching absorptlons of a number of substituted o&ho-bromophenols in ether-carbon tetrachlorlde mixtures have been measured. The phenolic hydroxyl of o&&o-bromophenols shows greater preference for mtramolecular (OH . * * . Br) bonding, rather than mtermolecular (OH . * * OEt,) hydrogen bonding. Comparisons are made with a number of 2-alkyl- and 2,6-dialkylphenols and the results discussed.

Part I of the present series [I], the results of studies on the positions and the intensities of t’he hydroxyl stretching absorptions of a number of substituted orthobromophenols in carbon tetrachloride solution were reported. Where the 6-position remained unsubstituted, the expected double hydroxyl absorption arising from the co-existence of the two conformations I and II (R=H, X=Br) [2] was observed, but where the 6-position carried a substituent-even a methyl group-only a single hydroxyl absorption peak, attributable to the intramolecularly hydrogen bonded conformation (I; R = alkyl, X = Br) was apparent. The complete absence of absorption attributable to conformation II in the 6-substituted compounds (R = alkyl, X = Br) is of interest in view of the fact that 2-t-butylphenol shows two peaks arising from conformations I and II (R = t-Bu, X = H) [3]. IN

q$

R&x

I

II

The present paper is concerned with the influence of solvent and competitive intermolecular hydrogen-bond formation upon the hydroxyl absorptions of these ortho-bromophenols. * Joint contribution from the Department of Chemistry and The Experimental Pharmacology Division, Institute of Physiology, the University, Glasgow, TV.2. t Present address: School of Pharmacy, The Royal College of Science and Technology, Glasgow, C. 1.

[I] I. BROWN, G. EGLINTON and M. MARTIN-SMITH, Spectrochinz. Acta 18, 1593 (1962). [2] L. PAULINC, J. Am. Chem. Sot. 58, 94 (1936). [3] N. A PUTTNA~X,J. Chew Sot. 5100 (1960). 463

I. BROWN, G. EGLINTON and M. MARTIN-SMITH

464

MEASUREMENTS AND RESULTS

Variution of solvent Our results for three representative ortho-bromophenols in hexane, carbon tetrachloride, chloroform, carbon disulphide and acetonitrile are summarized in Table 1. The frequencies (Y, cm-l), half band widths (AY(~,~)~,cm-l) and apparent molecular extinction coefficients (a,, l.moleelcm-l) were determined as previously described [l]. Also shown in Table 1 are the corresponding data for phenol and two pairs of 2-alkyl and 2,6-dialkyl phenols. These results are in good agreement with earlier studies of the effect of solvent upon the hydroxyl absorption of various simple phenols and hindered 2,6dialkylphenols [4]. Of the solvents selected acetonitrile is by far the most polar and the most basic, being able to associate with phenol molecules through intermolecular hydrogen bonding, and two distinct effects on passing from the other solvents to acetonitrile are immediately apparent. In the case of 2-t-butylphenol (compound 3), 2-bromophenol (compound 6) and 2-bromo-4-isopropylphenol (compound 7) the double hydroxyl absorption arising from the co-existence of the conformations corresponding to I and II which is present in all the other solvents, disappears in acetonitrile giving rise to a single broad intense band shifted to lower frequency. This is illustrated in Fig. 1 which shows the variation in the shape and intensity of the hydroxyl absorption with solvent in 2-bromophenol. On the other hand with 2,6-di-t-butylphenol (compound 5) and 2,4-dibromo-3-methyl-6-t-butylphenol (compound 8) two hydroxyl absorption bands can be detected in acetonitrile whilst there is but a single peak in the other solvents. This is illustrated in Fig. 2 which shows the variation in the shape and intensity of the hydroxyl absorption with solvent in 2,4-dibromo-3-methyl-6-t-butylphenol. These effects on passing to acetonitrile from the other solvents are due to intermolecular hydrogen bonding between the acetonitrile molecules and the molecules of the phenol. In the case of compounds 3, 6 and 7, in which there is relatively little steric hinderance to the approach of solvent molecules to the OH function, intermolecular hydrogen bonding occurs readily in acetonitrile. The shape and position of the broad absorption peak make it apparent that intermolecular bonding is occurring to the virtual exclusion of the intramolecular hydrogen bonding seen in the other solvents, although the existence of a small proportion of the intramolecularly bonded species can not be ruled out. With compounds 5 and 8 there is considerable steric hinderance to intermolecular hydrogen bonding, with the result that two species are present in acetonitrile. The low-frequency peak arises from the intermolecularly hydrogen bonded species whilst the high-frequency peak arises from unassociated phenol molecules. Compound 8 was also examined in 50 y0 acetonitrilecarbon tetrachloride solution in order to obtain sufficient resolution of the two bands for accurate measurement, and these are the values recorded in Table 1. It is readily apparent from Table 1 that the hydroxyl stretching frequencies generally decrease steadily in the order hexane > carbon tetrachloride > chloroform - carbon disulphide, whereas the apparent half band widths rise in a different order, namely hexane < carbon tetrachloride = carbon disulphide < chloroform. This [4] K.U.

INGOLD and D.R.TAYLoR,CU~J.

Chem.

39,220(1959).

M

8

2.6-diKe

.l

2,4-diBr-3-Me-6.t-Bu

2,0-di-t-Bu

None

~-MO

1

2

Phenol with suhstituents

3503

3629 3553

3621 3823

Y

10.5

13 12

13.5 16.5

~“wm

*

l

%

353

* 305

O-05 x Hexane, 0.5 mm

19

220

* 240

165 15

3618 5 3647

3500

205 *

%

17 17

A*WZIO

3612 3612

y

0005 Ed CCig, 5 mm

convenience.

30

40 22

3609 3641

3499

32 93

%~/,,a

165

IQ5

*

l

150

%

0 05 ix CHCI,, 0.5 mm

3599 3601

Y

235

3492

18 5

15

3609 3642

28 315

wl/z,o

0~0125 M CS,, 2 mm

3594 3506

P

stretchmg absorptions of substituted phenols in various solvents

* Not measured. t Asymmetrical band. : Measured by band reflexion. F These values involve a strong intermolecular association with tile solvent and are placed here for h = Hi&-frequency side I = Low-frequency side. The value for phenol m hexane is of a saturated solution. The values for compound 8 in acetonitrde are for a 0,025 M solutlou m 50Yb CHsCN-Ccl,. Values in parentheses arc approxmate.

____

IlO.

Compountl

Table 1. Hydroxyl

243

242

*

154 *

EC%

Awz,a

3496 6 (3442)

&:

165:

34cofc

l

rot

3624

l

3143Th$ 140 * *

y

~a

25

100

I

175 ’

0 05M rH,CY. t 0 5 mm

5

g

io E-

6 $I!

4

;;:

$ Fp

I. BROW, G. EGLINTONand M. M&TIN-SMITH

3600 Fig. 1. Hydroxyl absorption of 2-bromophenol (Table 1, compound 6) in nhexane, chloroform, carbon disulphide and acetonitrile solutions. 1, C&E,,; 2, CRC&; 3, CHeCN; (all 0.05 M in O-5mm cells); 4, CS,; (0.0125 M in 2 mm cell).

3500

3400

3300

Fig. 2. Hydroxyl absorption of 2,4-dibromo-3-methyl-B-t-butylphenol (Table 1, compound 8) in n-hexane, chloroform, carbon disulphide, carbon tetrachloride-8cetonitrile (1: 1) and acetonitrile solutions. 1, C,H,,; 2, CHCl,; 3, CH,CN; {all 0.05 M in 05 mm cells); 4, CS, (0.0125 M in 2 mm cell); 5, CH,CN-CC&( 1: 1) (0.025 M in 0.5 mm cell).

in the half band widths makes impossible any comparison of apparent extinction coefficients, but the changes in frequencies and half band widths with solvent are listed in Table 2 in which they have been so tabulated that, where different conformations exist, the bands are assigned to a given conformation. It is signi~oant that the ‘%&a-bonded” hydroxyl frequencies (compounds 6, 7 and 8) consistently undergo much smaller solvent shifts than those of the free hydroxyl groups. Presumably this is a reflexion of the existence of a lower degree of solvation in the “infra-bonded” nonformation I where the hydroxyl group is both more electrically neutral and more restricted conformationally and hence less solvent sensitive than is the hydroxyl group in the alternative tram conformation II. It is of interest that in 2-t-butylphenol (compound 3) the hydroxyl absorption arising from the conformation in which the hydroxyl group is directed towards the alkyl group is also less

variation

solvent sensitive than is the hydroxyl absorption ascribable to the conformation in which the hydroxyl group is directed away from the t-butyl group. It is only when

the alkyl group is t-butyl or buttressed iso-propyl in oaks-alkyl phenols that such dual hydroxyl absorption is encountered-lower alkyl groups give but a single peak. Our values for compound 3 are in good agreement with those previously reported for Z-t-butyl-4-methyIphenol[4]. The buttressed “~~~~u-bonded”hydroxyl group of compound 8 not surprisingly shows the smallest solvent shifts and in this connexion it is to be noted that oarbonyl frequencies in strongly chelated systems

Infra-red studies with substituted ortho-bromophenols-II

467

468

I. Bnowx,

G. E~LINTON and M. MARTIK-SMITH

where the basieity of the oxygen atom is reduced, also undergo negligible solvent shifts [S] as does the hydroxyl frequency in 2-nitrophenol [S]. The solvent shifts observed for the non “intro-bonded” conformations of 2-bromophenol (compound 6) and 2-bromo-4”isopropylphenol (compound 7) closely parallel those for phenol (compound l), but are significantly larger in magnitude, in agreement with the greater acidity of the halo compounds. Examination of 2-bromophenol in both carbon tetraohloride and acetonitrile singly, and in several different binary mixtures of these two solvents indicates that dielectric constant effects are of minor importance. Thus after plotting the shifts in frequenoy of both the “i@a-bonded” and “inter-bonded” bands relative to the carbon tetrachloride value against the calculated dielectric constant of the mixed solvent, there is found to be a change of about one wave number per four dielectric constant units for the “in&a” band and about one wave number per two units for the “inter” band. The extrapolated value for the “inter” band to acetonitrile in 100 % carbon ~tra~hloride at E = 2.24 was found to be 3386 cm-l. The width of the hydroxyl absorption band does not seem to be related to the position of @II). Thus it is apparent from Table 1 that in n-hexane, although the absorption of the “i&u-bonded” hydroxyl group of compound 8 occurs some 150 cm-r lower than the “free” hydroxyl of compound 5, the former band is slightly narrower than the latter. Since hydrogen-bonded hydroxyl bands are almost invariably quoted as being broader than the “free” with AY~~,~)~ proportional to Av our observation may be a significant pointer to the lack of freedom of the heavily buttressed hydroxyl group. This behaviour (also observed between compounds 1 and 8) continues in carbon tetrachloride and carbon disulphide, but in chloroform the hydroxyl absorption of compound 8 broadens like that of phenol without breaking the “&&a” bond, i.e. chloroform causes broadening without much shift in wavelength. This broadening effect is seen throughout the series studied except for compound 5 which has a heavily substituted symmetrical molecule. This may mean that the effect is due to an asymmetrical arrangement of polar ohlorofo~ molecules around the hydroxyl group despite the fact that the proton is still bound to the bromine atom. The formation of specific associations between chloroform and esters and ketones has recently been discussed in some detail by WH~TSELand KAC~RISE[ 71. It should be pointed out that ohloroform solutions were found to absorb atmospheric moisture with great rapidity giving rise to strong absorptions at 3608 cm-i = 20 cm-l) and 1603 cm-r (AyuIsjB= 16 cm-i) due to solvated water mole(A ~~~~~~~ cules, and although the utmost care was taken to work with anhydrous solutions, inaccuracies in the me~urements made in chloroform solutions are possible. Temperature studies

Two compounds, 2-bromophenol and 2,4-~bromo-6-t-butylphenol, were examined in tetrachlorethylene at various temperatures in order to discern any changes in the relative proportions of the “free” and “in&t-bonded” hydroxyl absorptions. The results are shown in Table 3. ~-[5] C.J. W. BROOKS,G.EGLINTON and J.F.RIoRxAN,J. Chem.Soc, 661 (1961). [S] L.J. BEIAAMY%nd H.E.H~LL~M, T~uns.FamdaySoc. 55,220 (1959). [?j K. B. WHETSELandR. E. ~~~~~~,~~ec~~oc~~~. Actu 18,315, 329, 341(1902).

Infrared Table 3. Temperature

studies with substituted

studies on 2-bromophenol and 2,4-dibromo-6-t-butylphenol tetrachlorethylene 2-Bromophenol

“Free”

Temp.

469

o&o-bromophenols-II

2,4-Dibromo-6-t-butylphenol “Free”

“In&a-bonded”

(“C)

r

Avu/z,n

E,

24” 40” 70” 90”

(3603) (3603) (3603)

* * *

6 1:

y 3526.5 3527 3529.5

m

A~/zr.

e,

18.5 19.5 19

171 161 157

-

“In&a-bonded” A ~u/z,o Y 3507.5 3508 3509 3511

19.5 19.5 19.5 19.5

&Cl 207 202 195 191

Figures m parentheses are approximate. * Not measured. The apparent extinction coefficients (e,) have been rounded to the nearest whole number.

With both compounds only very small effects, qualitatively similar to those found for the four ortho-halophenols by JONES and WATKINSON [S], were observed, but we consider that the relative changes in the proportion of the ‘%&a-bonded” and the “free” hydroxyl absorptions in 2-bromophenol are too small for accurate measurements of bond energies. There is a small but definite increase in the frequency of the single hydroxyl absorption arising from the intramolecularly hydrogen-bonded conformation I (R = t-Bu, X = Br) of 2,4-dibromo-6-t-butylphenol. There was no evidence of the appearance of absorption due to conformation II (R = t-Bu, X = Br) despite the slight fall in apparent molecular extinction coefficient of the bonded absorption with increase of temperature. COMPETITIVE INTERMOLECULAR HYDROGEN-BONDING STUDIES IN ETHERCARBON TETRACHLORIDE MIXTURES As an extension of the studies with acetonitrile solutions, certain compounds were examined in diethyl ether-carbon tetrachloride mixtures in order to ascertain the effect of competitive intermolecular hydrogen bonding on the position and intensity of the hydroxyl stretching absorptions. Similar studies of phenol-ether association have been previously reported for a number of simple 2-alkyl and 2,6dialkyl phenols [9] in which the factors limiting the extent of intermolecular hydrogen bonding were the bulk of the groups in the ortho positions of the phenol molecule and the size of the alkyl groups of the ether. In addition to selected o&o-bromophenols our studies were extended to include 2-fluorophenol and 2-fluoro4-nitrophenol in an endeavour to provide further verification that the single hydroxyl absorption peak (3592 cm-l in Ccl,) [l] of 2-fluorophenol arises from conformation I(R = H, X = F) (cf. Refs. 10-11). Sufficient ether was employed to ensure that some intermolecular phenol-ther hydrogen bonding was present and under these conditions three distinct hydroxyl absorption bands were possible-“free”, “intra” and “inter” bands. Table 4 gives the frequencies, half band widths and apparent molecular extinction coefficients for these three absorptions arising from species III, IV and V. Also shown is the [S] D. A. K. JONES and J. G. WATKINSON, Chem. & Ind. (London) 661 (1960). [9] L. J. BELLAMY, G. EGLINTON and J. F. MORMAN, J. Chem. Sot. 4762 (1961). [lo] A. W. BAKER, J. Am. Chem. Sot. 80, 3598 (1958). [ 11J A. W. BhKER and W. W KAEDING, J. Am. Chem. Sot. 81,5904 (1959).

470

I. BROWN, G. EGLINTON and M. MARTIN-SMITH

percentage of phenol non-bonded to ether, (rounded to the nearest 5 per cent), as calculated by dividing the molecular extinction coefhcient for the “free” or the intramolecularly hydrogen bonded absorption in pure carbon tetrachloride by the corresponding value for the ether-carbon tetrachloride mixture. The relative abundance of the various species within any one series of compounds may be interpreted in terms of the strength of the hydrogen bonds and steric effects (principally in relation to the entropy of the system). Also included in the Table are the values for phenol itself, 2-methylphenol, 2-t-butylphenol, 4-isopropylphenol and 4-t-butylphenol. The appearance of the absorption bands is illustrated in Figs. 3 and 4. All three absorptions are present only in ortho-bromophenols lacking a substituent in the 6-position. Ortho-bromophenols possessing a 6-substituent lack a “free” band st,o.

“intra” cm-l ca. 3520 (sharp)

‘free” ca. 3610 (sharp)

III

IV

“inter”

ca. 3330 (broad)

V

Fig. 3. Hydroxyl

absorptions of phenol, 2-bromophenol and 2-bromo-C-nitro-phenol (Table 4, Compounds 1, 6 and 18) in ether-carbon tetrachloride solutions. 1, phenol; 2,2-bromophenol; 3, S-bromo4nitrophenol. (All substances 0.003 M in 0.474 M ether-WI, in 5 mm cells). 3600

3500

3400

3300

3200

Fig. 4. Hydroxyl absorptions of 6-alkyl substituted 2-bromo-4-nitrophenols (Table 4, compounds 18-21) 1, ccl&in ether-carbon tetrachloride solutions. ether background (balanced). Substituent in 6position: 2, H; 3, methyl; 4, isopropyl; 5, t-butyl. (All substances 0.003 M m 2.3 M ether-ccl, in 5 mm cells).

0

3600

3500

3400

3300

3200

Fig. 5. Hydroxyl absorption of 4-t-butylphenol in 1, 4-t-butylether-carbon tetrachloride solution. phenol (Table 4, Compound 1O)in 0.47 M ether-ccl,; 2, as 1, with excess water added (saturated). (Both solutions 0.00125 M in 1 cm quartz cells).

I?4 R4

H iPr tBu H H

H iPr tBu Br Br Br Br H NO, NO, NO, NO, NO, NO, Br Br Br Br

R2

H H H Me tBu

Br Br Br Br Br Br Br F F Br Br Br Br Br Br Br Br Br

H H H H Me ePr tBu H H H H Me iPr tBu H Me iPr tBu

H H H H H

R6 3613 3615 3614 3613 3648 I3607 3608 (3603) 3609

Y

15

*

-

-

60 65 60 60 10 60 10 *

%

21 17 17 19 * 17 * *

“Free” (IV) A vu/a)0 V

22 22 26 25 20 20 19 21 30 24 22 22 24 24 28: 24 25 26

-

-

“In&a” (III) A VWWl

130 100 135 110 160 155 185 40 20 80 65 135 150 170 60 100 100 135

%

phenols in ether-carbon

3530 3535 3533 3529 3527 3523 3509 3592 3570 3514 3512 3509 3504 3489 35285 3527 3522 3509

stretching absorptions of substituted

(3164)§ (3210)s (3210) (3210) (3229) (3295) (3300) (3334)

3338 (3275) (3280) (3285) (3230) (3292) (3300) (3315) (3257) (3134) (3166)

3337 3349 3345 3340

V

(V)

150 br br br 280: br br br 2802 330: 270$ 275: br br br 2401 br br br

140:. 135; 140 170

A V(llalS

“Inter”

115 35 25 35 65 25 20 20 50 135 110 125 35 45 40 130 55 45 35

150 120 120 110

E,

tetrachloride solutions

80 65 90 65 85 80 85 20 10 35 25 60 65 75 35 55 55 60

30 55 50 35

%

non-bonded to ether

or “intra”.

(I

4

(a)--examined in 0.47 M ether-CC& solution. (b)--examined in 2.3 M ether-Ccl, solution. $ Measured by band reflexion. §Asymmetrical band. Values in parentheses. * Not measured. t 0.1 M phenol in 0.5 M ether-Ccl,. are approximate. br = broad. All measurements on ether-carbon tetrachloride solutions were run on the scale 4 mm = 10 cm- l. The recorded band positions cannot be as accurate as those examined in pure carbon tetrachloride since the broad OH * * . . 0 bands are dficult to place to &lo cm-‘. E, in CCI, X 100 The percentages have been rounded to the nearest five units. yo Unbound = E in ether_CC1 where E, is with respect to either “free”

1864 18(b) 19(b) 20(b) 21(b) 12(b) 13(b) 14(b) 15(b)

16(a) 17(a)

14(a) 15(a)

13(a)

W 7(a) 11(a) 12(a)

3tW

Wa)

l(a) w4 lob)

Compound

OH

Table 4. Hydroxyl

472

I. BROWN, G. EGLINTON and M. MARTIN-SMITH

and the alkyl phenols lack an “intra” band. The “free” and “h&a” bands are at virtually the same frequencies as in pure carbon tetrachloride solution [cf. Ref. 11. The intermolecular band appears in some cases to extend up to ca. 3600 cm-l. Solutions approximately 0.47 M (ea. 5% V/V) and 2.3 M (ca. 25% V/V) with respect to ether were employed, but the equilibrium did not seem unduly sensitive to the ether molarity as 0.67 M ether solutions gave results closely similar to those obtained in 0.47 M ether solutions. This is to be expected since in either solution the ether-phenol molecular ratio is in excess of 100 : 1. In the series 2-methylphenol, phenol, 2-bromophenol, 2,4dibromophenol and 2-bromo-4-nitrophenol (compounds 2,1,6,12 and 18 respectively) the increasing acidity of the phenolic hydroxyl group as the series is ascended is evident from the progressive shift to lower frequencies of the intermolecularly bonded band. Consideration of the apparent extinction coefficients and of the percentage of “free” or inter-molecularly bonded form present is, however, complicated by the presence of an intramolecular hydrogen bond in compounds 6,12 and 18, which bear orthobromine atoms. In these three compounds there is competition between the bromine atom and the ether molecules for the right to hydrogen bond to the proton but this competition is, of course, not present in compounds 1 and 2. A particularly good illustration of this effect is obtained by comparing compounds 2 and 6 since the methyl group (2.0 A) and the bromine atom (1.95 A) are of almost the same radius. Introduction of a 2-bromine atom into compound 1, i.e. to form compound 6, decreases drastically the amount of inter-bonding despite the already noted acidifying effect of the bromine atom, as the hydroxyl group shows greater preference for the formation of the intramolecular hydrogen bond. However, introduction of another bromine atom in the 4 position to form compound 12, while resulting in a further increase in acidity of the hydroxyl now increases the amount of inter-bonding over compound 6. Replacement of this 4-bromine atom by a nitro group so increases the acidity of the hydroxyl group that compound 18 attains the same degree of interbonding as compound 2 despite the competition afforded by the 2-bromine atom. A similar increase in acidity pertains between 4nitrophenol and phenol. Here measurements were made in ether-chloroform [12] and it was found that the more acidic nitro derivative gave a greater increase in intensity and a greater shift to lower frequency of the hydroxyl absorption due to the species associated with the ether. It would therefore seem that increased acidity of the hydroxyl group favours intermolecular hydrogen bonding rather than intramolecular hydrogen bonding. These effects are illustrated in Fig. 3. A related situation has been reported in the salicylaldehyde series where pyridine was used to induce competitive hydrogen bonding [13]. 4-Alkyl substitution which produces a slight decrease in acidity has very little effect on the association of either phenol or of 2-bromophenol with ether, but in the two principal series shown in Table 4 (compounds 12-l 5 and compounds 18-21) a S-alkyl group reduces considerably the extent of intermolecular association to ether, presumably by restricting access to the hydroxyl group and by virtue of the greater [12] G. EGLINTON. Unpublished (1959). [13] C. J. W. BROOKS and J. F. MORMAN, J. Chem. Sec. 3372 (1961).

Infra-red &udies with substituted

o&o-bromophenols-II

473

strength of the intramolecular hydrogen bond through steric compression of the OH-Br distance [l]. The effect of increasing bulk in the ortho substituent on ether bonding in these series is illustrated in Fig. 4. In the Paru-bromo series (compounds 12-15) there is a rise in frequency of the associated band which parallels the fall in intensity consequent upon the increasing bulk of the S-alkyl substiutent, the order being H < methyl < isopropyl < t-butyl. This is to be anticipated in view of the increased restriction offered to the approach of the ether molecule and is in marked contrast to the lowering in frequency of the intramolecular bond already discussed. The fall in intensity is perhaps the most sensitive measure of the 6-alkyl effect. In several of the experiments a rather broad symmetrical band of variable intensity was sometimes observed near 3500 cm-i (see also [9]) and its appearance is illustrated in Fig. 5. This band was found to be due to the presence of dissolved water in association with the ether molecules. (H-O-H . . . . . Et,0 or some similar complex). Water alone added to ether-carbon tetrachloride solution absorbs at = 100 cm-l), strong. 3614 em-r (A~~~,~~~ = 30 cm-l), weak, and 3494 cm-l (AY~~,~~~ It is difficult to exclude all traces of water during the preparation of solutions because both ether and some of the phenols are highly deliquescent and consequently the quantitative data recorded in Table 4 may not be entirely accurate.

2-Fluorophenol Unlike the other ortho-halophenols which show two hydroxyl stretohing peaks attributable to the cis and trons conformations I and II (R = H, X = Cl, Br or I) [lo,1 1,141, 2-fluorophenol shows but a single hydroxyl absorption band in carbon tetrachloride solution at 3592 cm-l [I]. This unusual value has led to queries as to whether it represents a bonded or a non-bonded hydroxyl absorption. BAKER and KAEDINQ [l I], on the basis of a series of elegant competitive intramole~ul~ hydrogen bonding studies in asymmetrical 2,6-dihalophenols, came to the conclusion that this single band in 2-fluorophenol was due to conformation I (R = H, X = F)-the intramolecularly hydrogen bonded conformation-since 2,4-dibromo-6-fluorophenol absorbs at 3574 and 3522 cm-l in carbon tetrachloride. The low-frequency band at 3522 cm-l being assignable to OH . . . Br bonding leaves the high-frequency absorption for the conformation in which the OH is directed towards the fluorine atom; its value of 3574 cm-l being very near the value of 3584 cm-l for 2-fluorophenol strongly indicates that the latter is a bonded peak. Similarly 2,6-~fluorophenol is known to absorb at 3586 cm-l in tetrachloroethylene 1151 and since measu~ments in the two solvents are comparable-for instance we find that 2,4-dibromo-6-t-butylphenol absorbs at 3507.5 cm-l in tetraohlorethylene and at 3509 cm-l in carbon tetraohloridethis observation provides further evidence for the “intro-bonded” conformation. Since the hydrogen bond in 2-fluorophenol is weaker than that in 2-bromophenol and yet there is evidence of the free conformation in the latter, it would be expected that there would be even more free conformation II present in the former. BAKER [lo] suggests that if it is indeed present its non-appearance is due to its occurrence [la]

M. M. DAVIES, Trans. B'aradag Sot. 36,333 (1940);

J. C&m. Phzp. 21, 1606 (1953); H. HOYER, Chew. [15] J. G. WATEINSO~;. Personal communication

(1961).

G. ROSSMY, W. L~TTKE and R. MECKE, Ber. 89, 146 (1956).

I. BROWN, G. EGLINTONand M. M~TIN~SMITH

474

on the high frequency side of the main band with a very small wave-number separation. Since we have shown in our studies that the frequency of the free band is more sensitive to change in solvent than the intra band, it was initially reasoned that examination in hexane might reveal the presence of a free band by means of an increased separation. Hexane, however, is known to favour the less polar conformation, in this case the intra bonded form. On examination of 2-fluorophenol in n-hexane, only one band at 3603 cm-l was observed with no evidence of a shoulder on the high-frequency side. We therefore turned out attention to a comparison of 2-fluorophenol (compound 16) and Z-~uoro-4-nitrophenol (compound 17) in ether-carbon tetrachloride solution since the aci~fying effect of the 4-nitro-coup would be expected to produce effects similar to those observed with the ort&-bromophenols. As is evident from Table 4 both the fluoro compounds display a much greater degree of inter-bonding than their bromo-counterparts, and the inter bands are at slightly lower frequencies. This latter fact might indicate that 2fluorophenol is more aoidio than 2-bromophenol but in view of the opinion [l I] that the reverse is probably true, it is presumably a reflexion of the difliculty in measuring the exact position of these broad bands. The increased amount of interbonding is probably largely due to the decreased strength ofthe OH.. . . F bond relative to the OH . . . . Br bond, the hydroxyl group thereby being more accessible to ether molecules. Our studies seem to show that 2-fluorophenol in ether-carbon tetrachloride solution behaves in a qualitatively similar fashion to 2-bromophenol, though the hydroxyl group in the former is more freely able to bond to ether molecules. This is unlikely to be a steric effect since it has been shown [9] that groups smaller in size that t-butyl have negligible effects in this oonexion. It may point to the decreased strength of the hydrogen bond (OH . . . . F) or to the possibility that the hydroxyl group is simply oriented towards the highIy polar fluorine atom without actually being bound to it. EXPERTMENTAL

All spectra were measured with a Unicam S.P.100 double-beam spectrophotometer equipped with an SP 130 sodium-chloride prism grating double monochromator operated under vacuum conditions by the general procedure used previously El]_ Carbon tetraohlorid~ (“AnalaR”), carbon disulphide (“AnalaR”) and n-hexane (spectroscopic grade) were used without further purification. Chloroform (“AnalaR”) was freed from ethanol by two successive passages through blue silica gel immediately before use. The M and B anhydrous diethyl ether (sodium-dried) in chloroform and carbon tetrachloride solutions were examined immediately after preparation. Acetonitrile was purified by successive prolonged treatments with potassium hydroxide, calcium chloride and phosphorus pentoxide, followed by ~stillation. Compounds 6-8 and 11-21 were obtained as indicated in Part I of this series [I]. Compounds 1,2,4,5,9 and 10 were purified from commercial samples. Compound 3 was prepared as previously described [ 161. thank Mrs F. LAWRIEfor most of the infra-red measurementsand one of us (I. B.) gratefully acknowledgesthe tenure of&fellowship awardedby Messrs. Glaxo Labor&tories Ltd.

Acknowledgements-We

[lS]

H. HART, J. Am. Chem.. See. 71, 1966 (1949).