650
L. KUHN,P. SCHLEYER, W. BAITINGER, JR.,
Reaction of Cumylpotassium with t-Nitrocumene.-Cumylpotassium was prepared by the method of Morton22 from 7.8 g. (0.20 g.-atom) of potassium sand, 24.0 g. (0.20 mole) of cumene, and 10.6 g. (0.10 mole) of 1-chloropentane in 75 ml. of heptane. t-riitrocumene (16.8 g., 0.10 mole) was added with stirring t o t h e deep red cumylpotassium solution while maintaining t h e temperature at -10 to -5" during its addition. Stirring at this temperature was continued for 1 hr. after the addition was completed. t-Butyl alcohol, 20 ml., was then added t o destroy a n y excess potassium. T h e solution was filtered t o give 16.0 g. of colorless solid which assayed for a mixture of 7.25 g. of potassium chloride and 9.20 g . of potassium nitrite, in yields of 97 and 107y0, respectively. T h e filtrate, 79.0 g., contained 2.6 X 10-4 mole of di-t-cumylnitroxide having a nitrogen h.c.c. of 14.9 gauss. Analysis of the filtrate by v.p.c. showed t h e presence of bicumyl (7.6 g., 427, yield based on reacted t-nitrocumene), a-methylstyrene (0.9 g . ) , cumene (13.3 g.), t-nitrocumene (4.0 g.), and four unidentified components (combined weight, 1.9 9 . ) which were not further investigated. Reaction of Cumylpotassium with t-Nitrobutane.21a-t-Nitrobutane (10.2 g., 0.10 mole) was added t o 0.10 mole of cumylpotassium in 75 ml. of heptane a t -15" over a period of 1 hr. After stirring for 0.5 hr., the reaction mixture was allowed t o warm t o room temperature and t-butyl alcohol (20 ml.) was added t o destroy excess potassium. The mixture was hydrolyzed with 100 ml. of water, extracted with ether, and the extract dried over magnesium sulfate. Evaporation of the solvent in vucuo resulted in the loss of t-nitrosobutane and left a red oily residue which was analyzed b y v.p.c. Among the 22 components detected, heptane (5.08 g . ) , t-nitrobutane (0.091 g.), cumene (13.8 g . ) , di-t-butylnitroxide (0.585 g.), tri-t-butylhydroxylamine (1.99 g . ) , and bicumyl (0.461 9 . ) were identified by their retention times relative t o o-dichlorobenene. The remaining components, ranging in amount from 0.030 to 0.390 g., were not further investigated. (22) A. A. Morton and E . J. Lanpher, J . Org. Chem., 23, 1636 (1958)
AND
L. EBERSON
Vol. 86
Reaction of t-Butylmagnesium Chloride with &Nitrobutane.t-Sitrobutane (50.0 g., 0.485 mole), dissolved in a n equal volume of ether, was added t o the Grignard reagent prepared from 46.5 g. (0.5 mole) of t-butyl chloride and 12.2 g. (0.5 mole) of magnesium. T h e Grignard solution was cooled t o -78' prior to t h e addition of the nitro compound. During its addition, over the course of 1 hr., the reaction mixture turned green and a solid precipitated. The reaction mixture was stirred at 25" for 0.5 hr. after which it was hydrolyzed b y t h e addition of water a t 0". Hydrolysis was accompanied by the evolution of isobutylene. The organic layer was separated and dried over anhydrous magnesium sulfate. Analysis b y e.s.r. showed it t o contain about l o F 2 mole of di-t-butylnitroxide. Other products, identified b y v.P.c., were t-nitrosobutane, t-butyl alcohol, t-butyl nitrate, and tri-t-butylhydroxylamine. p,p'-Dianisy1nitroxide.-The Grignard reagent was prepared from p-bromoanisole (61 g., 0.326 mole) and 8.7 g. (0.358 mole) of magnesium in 1 1. of 1: 1 ether-benzene by the procedure of Keid.23 It was cooled in ice and p-nitroanisole (42.4 g., 0.326 mole) was added over a I-hr. period. .kt the end of this time 200 ml. of water was added slowly until the magnesiurn salts had coagulated. The organic layer was decanted from the magnesium salts and evaporatively distilled a t 30". The residue showed a typical nitroxide e.s.r. signal. All attempts a t purification, however, failed. Small amounts of di-p-anisyl and di-panisylamine were shown to be present in this residue b y comparison of its infrared spectrum with those of authentic samples.
Acknowledgment.-The authors are grateful to Dr. W. G. Hodgson for his measurement of the e.s.r. spectra and magnetic susceptibilities, Dr. J. H. Deonarine and his group for microanalyses, Dr. J . Lancaster for n.m.r. spectra, and Mr. E. S. Everett for the V.P.C. analyses. (23) J. C. Reid and H. B. Jones, J . Biol. C h e m . , 1 7 4 , 427 (1948)
[CONTRIBUTION FROM THE U. S. ARMYBALLISTICS RESEARCH LABORATORIES, ABERDEESPROVING GROUNDS, M D .; FRICK CHEMICAL LABORATORY, P R I X C E T O X UNIVERSITY, PRINCETOS, N . J . ; AND THE UXIVERSITY O F LUND,LUND,S W E D E N ]
Conformational Effects and Hydrogen Bonding in 1,4-Diols1 B Y LESTERP. K U H N P, ~A U L
VON
R.s C H L E Y E R , 3 WILLIAM F.BAITINGER, J R . , 4 AND LENNART EBERSON' RECEIVEDACGUST5, 1963
Study of the influence of conformation and configuration on intramolecular hydrogen bonding in diols has been extended t o 1,4-dihydroxybutane derivatives The 3 p infrared spectra of 57 butane-1,4-diols substituted only upon the 2- and 3-positions -acyclic, unsaturated, monocyclic, and bicyclic examples-were examined in detail. In a formal sense these 1,4-diols resemble the 1,Z-diols studied earlier b y similar techniques, t h e -CHzOH groups of 1,4-diols replacing the -OH groups of 1,2-diols; however, t h e two series of compounds behave quite differently. In l,2-diols the spectral shifts ( A v ) due t o intramolecular hydrogen bonding decrease with increasing azimuthal angle between OH groups, but t h e reverse is true for l,.l-diols. A maximum AV appears to be reached when the azimuthal angle between adjacent C H 2 0 H groups in 1,4-diols is about 90'. A conformational analysis of 1,4-diols with different azimuthal angles provides an explanation for this behavior; optimum hydrogen bonding interactions can be achieved only in certain geometrical arrangements Evidence is presented t h a t more than one major conformation permitting hydrogen bonding is present in certain of the diols. The influence of other structural variations on spectral details is discussed.
Intramolecular hydrogen bonding is particularly sensitive to changes in molecular geometry. For 1,2diols, the most extensively studied class of compounds, the magnitude of the infrared spectral shifts ( A v ) of the fundamental OH stretching vibrational bands due to intramolecular association have been correlated with configurational and conformational In 1,2(1) Presented a t t h e Fourth Delaware Valley Regional Meeting, American Chemical Society, Philadelphia, P a , J a n , , 1962, Abstracts, p 77, and a t the Ninth Conference on Reaction hlechanisms, Brookhaven, pi. Y ,Sept., 1962. (2) Ballistics Research Laboratories, paper V of a series on hydrogen bonding, paper I V : L. P . Kuhn and R . E Bowman, Spectrochim. A d a . 1 1 , 650 (1961). Also see L. P K u h n and G . G . Kleinspehn, J . O r g . C h e m . , 2 8 , 721 (1963) (3) Princeton University. Alfred P. Sloan Research Fellow. Paper XI of a series o n hydrogen bonding; paper X : A , Allerhand and P . v a n R . Schleyer. J . A m Chem. S o c . , 83 1715 (1963). (4) Princeton University. American Cyanamid Junior Research Fellow, 1960-1962; P\-ational Institutes of Health Fellow, 1962-1963. ( 5 ) University of Lund. (6) (a) L. P . K u h n , J A?n C h e m . Soc., 1 4 , 2492 (1952); (b) ;bid., 1 6 , 4393 (1954); (c) i b i d . , 80,5930 (19.58). ( 7 ) P . von R . Schleyer, ibid.,83,1368 (1961), and references cited therein. ( 8 ) For a review, see h f . Tich9, C h e m . List?, 6 4 , 506 (1960).
diols, the azimuthal angle, 4, between two C-0 bonds on adjacent carbon atoms is of prime importance in determining A V " ~ ;values ranging from Av = 0 cm.-' for compounds where the hydroxyl groups are too far apart to permit intramolecular interaction (e.g., transcyclopentane-1,2-diol, 90" < < 120°)6,9to AV = 103 cm.-' (cis-exo-norbornane-2,3-diolj + = 0") have been reported (Table 11). Most 1,2-diols have 4 azi-
+
(9) F . T'. Brutcher, J r . , and W. Bauer, Jr., J . A m Chem. Soc , 8 4 , 2236 (1962), propose a modified equation relating A v and O H . . . O distances for 1,2-diolst o replace the original Kuhnea equation now known to he in error. This modified equation is not successful for higher diols, however, and a more general relationship is under investigation ( b y L. P . K 1. (10) H.Kwart and W. G . Vosburgh, ibid., 1 6 , ,5400 (1954); H . Kwart and G . C Gatos, ibid , 8 0 , 881 (1958). H Krieger, A n n . A c a d . Sci F e n n i c a e , A I I , No. 109 (1961). C j also S J Angyal and R . J Young, J . A m . Chem. Soc., 81, 5467 (1959); G.Jacob and G. Ourisson, Bull so(, c h i n . F r a n c e , 734 (19.58); T . Takeshita and M Kitajima, Bull. Chem. Soc. J a p a n . 32, 985 (1959); and E. L. Eliel and C Pillar, J . A m . Chem. S o c , 77, 3600 (1955), 8 4 , 4999 (1962). 65 cm.-' (11) Even though 4 E 0' in cis-cyclobutane-1,2-diols,A v because of t h e greater O H , , 0 distance caused by distortion of the C-C-C ring angles t o 9 0 0 R . Criegee and K . Noll, A n n , 6 2 1 , 1 (1959); E . J. hloriconi, W . F O'Connor, I,. P. K u h n , E A . Keneally, and F.T. Wallen berger, J . A m Chem. Soc., 81,6472 (1959). See Table I V .
Feb. 20, 1964
DIO DIOLS
HYDROGEN BONDING IN
65 1
muthal angles near 60” and Au values between 30 and 50 cm. - I ; hence, conformational differences can be detected in nonrigid systems. In acyclic 1,2-disubstituted ethylene glycols, nonbonded repulsions result in larger Au’s for dl- or tkreo- (I) than for meso- or erytkro- (11) isomers (Table III).688s12 The reverse is true for cyclohexane-1,2-diols where, probably due to a flattening of the ring from the regular geometry usually a ~ s u m e d , ’cis-ea-diols ~ give a larger Au than trans-ee isomers.6,8 The same is found, of course, for rings smaller than six membered ; for larger rings, Aulransfirst becomes larger than Aucis in the cyclodecane-1,2-diols (Table IV). H R R
R$c
R$c
H
R
H
R
dl or threo Ib, X Ia, X = OH IIIa, X = CHsOH IIIb, X
H
= =
OH CHzOH
meso or erythro 11, X = OH I\’, X = CHzOH
A second structural effect on Au recognized in 1,2diols is the alteration of geometry due to substituents, particularly bulky ones (e.g., lj1,2,2-tetra-t-butylethylene glycol, Au = 170, the largest observed value for 1,2-diols; Table V).6 C-C-C angle distortion (“Thorpe-Ingold effect”), suggested as a possible factor, has been assessed by similar techniques in a study of 1,3-di0ls.~ Very few spectral measurements on 1,4-diols have been reported in the literature; less than a dozen alian phatic examples are k n o ~ n . ~ ~ ~We ~ ’here ~ - ’report ~ investigation on 2- and 3-substituted butane-1,4-diols. In a formal sense these resemble 1,2-diols with the OH groups of the latter replaced by CHzOH groups. The spectral behavior of analogously constituted 1,2- and 1 , P diols is completely diferent, however; possible reasons for this discrepancy are considered in this paper. Experimental Procedures and Results Preparation of Diols.-The diols listed in Table I were prepared using accepted methods outlined in detail in the literature.’* I n general the following procedure was employed: a n excess of lithium aluminum hydride was slurried in anhydrous ether, and the carboxylic acid, anhydride, or ester precursor of the desired diol in ether solution was added at such a rate as t o maintain gentle reflux. After the addition was complete the reaction mixture was refluxed for 1 hr. and then decomposed either with water or with saturated sodium sulfate solution. T h e ether was removed and the products were distilled if liquids or recrystallized if solids. Diacids occasionally were converted to their methyl esters with diazomethane before reduction. I n the case of some of the diols (listed by a n asterisk in Table I ) the reductions were run on such a small scale, owing to the availability of only small quantities of the precursors, t h a t it was not feasible t o purify the diols obtained. For these examples the infrared spectra were run on the crude material, it having been demonstrated in many other (12) G . Chiurdoglu, R . D e Groote, W . Masschelein, and M . H . Van Rissegheim, Bull. soc chim. Belges, TO, 342 (1961). (13) T h e C-C-C angles in cyclohexane are about 111.5’ and not 109.5’; V. A. Atkinson and 0. Hassel, Acla Chem. Scand., 13, 1737 (19.59); V . A. Atkinson, ibid , 15, 599 (1961). (14) M. Mousseron, M . hlousseron, and M. Granier, Bull. soc. chim. France, 1418 (1960). ( 1 5 ) (a) I. Sicher, F. SipoS, and J. JonaS, Coll Czech. Chem. C o m m . , 16, 262 (1961); (b) A. B. Foster, A H Haines, and M. Stacey, T ~ l r a h e d r o n16, . 177 (1961). (16) 1).W. Davidson, Can. J . Chem., 39, 2139 (1961); J Rigaudy and P. Courtot, Telrahedron l e l f e r s , No. 3, 9.5 (1961); R . D . Stolow. J . A m . Chem. Soc ,63,2592 (19611, P. Courtot, A n n chim., [a], 197 (1963); R . D. Stolow and M. M. Bonaventura, J A m . Chem. Soc., 35, 3630 (1963); J. Tadanier, J . Org. C h e m . . 38, 17.14 (1963); J. FajkoS, J . Joska. J . Piiha, and F . Sorm, Coll. Czech. Chem. Commun., 28, 2337 (1Y63); J . Pifha, J. Joska, and J , FajkoS, ibid., 28, 2011 (1903). (17) H . Christol, hl. Levy, and Y. Pietrasanto, Bull. chim. soc. France, 1132 (1963). (18) S G. Gaylord, “Reduction With Complex Metal Hydrides,” Interscience Publishers, Inc., New York, N.Y., 1956, Chapters 8 and 9,
3476
3636
I
3636
3700
3600
3dOO
3400
3700
3600
3500
3400
Fig. 1.-Representative hydroxyl region infrared spectra of butane-1,4-diol derivatives; concentrations all 0.005 M in CCl, solution, 1-cm. cells: A, butane-1,4-diol (1); B, 2-norbornenecis-endo-5,6-dimethanol (47); C, cyclooctane-cis-1,2-dimethanol (42); D , cycloheptane-cis-1,2-dimethanol(40). instances t h a t small amounts of impurities had little or no effect on the position of the peaks. Where possible, the physical constants of the diols were determined and are reported in Table I ; agreement with the literature was good for known compounds. Sources of Compounds.-Compounds 1, 21a, 24, and 47 were commercially available and were redistilled or recrystallized before use. Compounds 4 through 20 were prepared by reduction of the appropriate carboxylic acids.19 cis-Cyclopentane-1,2-dicarboxylic acid, furnished by Dr. N . L. Allinger,zo~ was used for the preparation of compounds 31 and 32. Compounds 2 7 , 29 and 30 were supplied by Dr. A. T . Blomquist and Dr. A . C . Cook. Compounds 41 and 43 were donated by Dr. J. Sicher, as were the carboxylic acid precursors for the ris examples, 40 and 42.16 Compound 55 was furnished by Dr. J. Meinwald; hydrogenation gave compound 54. Compounds 2, 33, 34, 37, 44, 45, 46, 50, and 51 were prepared by the catalytic hydrogenation of compounds22, 35, 36, 38, 47, 48, 49, 52, and 53, respectively. T h e remaining compounds were prepared by reduction of appropriate precursors or by following literature methods. Infrared Spectral Procedures.-The infrared ciirves were obtained in CClr solutions by the same procedures outlined previo~sly.~ T h~e~ l,4-diols showed a strong tendency t o associate intermolecularly and it was desirable to use very dilute solutions, 0.002 M or less, t o avoid interference from the dimer band which comes a t about the same position as the intramolecularly bonded peak. For this purpose 2-cm. matched silica cells were employed in several instances in place of the 1-cm. cells ordinarily used. The recent availability to us of a Perkin-Elmer 421 grating spectrometer equipped with scale expansion has permitted the examination of very dilute solutions in I-cm. cells. Agreement with the earlier results has been excellent; the sharp, free peaks were reproduced with a n accuracy of f l cm.-’ and the broader, bonded peaks t o f 2 cm.-l. Spectral shifts, A v , can be reproduced within 2 cm.-l and intensiy ratios within 10%. Data are summarized in Table I. Spectral Features and Complications.-Ideally, the high resolution 3 p spectra of 1,4-diols substituted only on the 2- and 3positions should consist of two peaks. The higher frequency “free” peak would be sharp and a broad lower frequency band would be expected due t o intramolecular OH . . . 0 hydrogen bonding. Many of the compounds examined here, such as butane-1,4-diol itself (Fig. l A ) , exhibited this uncomplicated behavior. Besides the band positions, the separation between them (Av,in cm.-’) is of interest since it is usually considered to be related to the enthalpy of the hydrogen bond and to the OH . . . 0 d i ~ t a n c e provided ,~ that the compounds are of the same general type.*Ob These data are included in Table I . In a 1,4-diol a certain fraction of molecules will be intramolecularly hydrogen bonded. b u t only one of the two OH groups will be acting a s a proton donor; the other will be “free.” The rest of the molecules will not be internally associated; both OH groups will be “free.” The free peak intensity comprises both types of systems and will be relatively insetisitive to the degree of intramolecular hydrogen bonding taking place. The bonded peak will be due only t o internal association and should serve as a
~-
(19) L . Eberson, Acfa Chem Scand., 1 3 , 40 (1959). (20) (a) N. 1, Allinger and V. B. Zalkow, J . A m . Chem. S o c , 83, 1144 (1961); (b) G. C Pimentel and A. L RIcClellan, “The Hydrogen Bond,” W. H . Freeman and Co , San Francisco, Calif., 1960.
L. KUHN,P. SCHLEYER, W. BAITINGER, JR.,
652
AND
L. EBERSON
Vol. 86
TABLE I PHYSICAL AND SPECTROSCOPIC PROPERTIES OF DIO DIOLS
Compound
NO.
ConBguration
M.p. or b.p. (mm.), 'C.
Lit. ref.
n*QD
Carbon, % Hydrogen, % Calcd. Found Calcd. Found
viree
vbonded
A"
363Sb (3634)s (3640)18 3636'
3477 (3478)' (3484)'6 3477 3546sh 3477 3534sh 3482c 3466 3473 3466 3476d 3463 3515 3476 3476 3463 3472 3460 3458 3436 3509 3476
159 (156)' (156) La 159 YO 159 102 153 167 159 166 150( 160)" 169 114(121)6
Free/ bonded intensity ratio"
Acyclic 1,4-diols 1
Butane-
2
2-Methylbutane-
3
2.2-Dimethylbutane-
4 6 6
2,3-Dimethylbutane-
133.5 (16)
94-95 (0.5)
dl
2,3-Diethylbutane-
dl
meso
21
1.4488 22 23
* meso
7
1.4460
124-126 125-127 130-132 130-182 89-9 1 42-43 175-178 109-110 96-98 96-98 84-86 125-126 217-219 92-93
(8) (8)
(8) (8)
24 1 4521 24 1.4548 1.4537
3636'
1 4528
60.98 65.71 65.71 68.91 68.91 71 29 71.2Y 72.98 72 98 75.53 75.53
6 1 22 11.94 66 01 1 2 . 4 1 65 2 12 41 69 04 1 2 . 7 3 68 8 12 73 7 1 . 2 8 12 87 71 21 12 87 72.66 13.13 74.20 13.13 75.3 11 89 74.4 11 89
12 16
12 58 12 6 12 09 12 6 13 00 13 10 12 84 13 06 11 9 11 8
71.29
71.40
12.87
13 10
9.93
9 97
8 9 10 I1 12 13 14 15 16 17
2,3-Jliisopropylbutane-
dl meso
2,3-Di i-butylbutane-
di meso
2,3-L)ineopentylbutane-
dl meso dl meso
le
2,2,3,3-Bisdimethylenebutane-
78-79
65.57
67.4
19
3,2,3.3-Bistrimethylene-
70.71
10.66
2,2,3.3-Bistctramethylcne-
60-61, then solidifies remelts a t 75-76 92-93
70.54
20
72 68
72.6
11.18 11 2
2,3-Dicyclohexylbutane2,2 3 3-Tetramethylbutane2,2,3,3-Tetraethylbutane-
butane-
m.m.p. 65-90
25
10 5
3635 3633 3632" 3632 3626 3632 3629 3631 3627 3627 3623 3627 3630 3632 3620 (3610)j 3633"
15,5
151(160Ie 164( 173Ie 151(164)' 167 ( 176)e 172 196 123 144(160)'
2.6
( 2 6)" 1 5 5.0 1.8 3.0 2.5 1.1 1.0 0.8 0.7 1.3 1.7 1.1 1.9 0.Y 0.7 1.3
0.9 0.9 1 .o
3481 3555w 3468
152 78 165
0.9
3498
130
5.5
0
...
3480
148
2.1
3540
91
4.1
360d
0
...
..
3610g
(17)j 0
...
3619j 3625 3634 3635 3628
3512 None 3507 3498 3509
107(124)"
1.1
127 137 1 19( 127)'
1.2 1.2
3498
125(138)'
...
3501 3475 3533sh 3502 3534sh 3491 3551 3472 3534 3517 3496
131 159
1.2 1 .o 2.5 2.2 3.3 2.1 3.8 1.4 2.1 2.1 3.6
3633
butane-
Unsaturated acyclic diols 21
2-Butene-
21a
cis
98-100 ( 0 7)
lrans
97 ( 0 . 2 )
22
2-Methyl-2 butene-
cis
100-101 (0.5)
23
2,3-Dimethyl,2-butene-
cis
90-91 (0.1)
24
2 3-Dibromo-2-butene-
lrans
24a
1.4779
3628shb+' 3622 3635sh' 3626 3628shb,L' 3623 3631shb*g 3621 3623b,g 3615
27 1.4686
28 29
114.8-116.0
2 Butyne-
26
30
43-45
31
..
Monocyclic diols 25 26 27
3,3-Dimethylcyclopropane1,2-dimethanol Cyclobutane- 1,2-dimethanol
28 29
30
3,4-DiphenylcyclobutaneCIS.1 2-dimethanolh 3 Cyclobutene-1,Zdimethanol Cyclopentane- 1,Z-dimethanol
cis
trans cish trans
123-124 (8) 123-124 (8) 103-104 (0.5)
1.4700 1.4701
64.62 64.62
1.4728
62.04
63.53 63.60 61.79
10.77 10.75 1 0 . 7 7 10.80 10.41
10.38
32
Cyclohexane-1 ,Z-dimethanol
CIA
42-43
6, 33
3623 3630sh 3632 3634 3634
34
Cyclohexane.l,2-dimethanol
irans
51-52
6,33
3635
35
4-Cyclobexene-1,2dimethanol
cis
31 32 33
36
cish
cis !ram
!ran9
37
4-Methylcyclohexane-cis-1,2dimethanol
38 38
4-?rlethyl-4-cyclohexene-1,2-
dimethanol
*
18
*
18
141-113 (1.7)
1.5080
113-115 (0.25)
1 , 5 0 6 3 35
3838"
34
31336~
* cis trans
40
Cycloheptanc 12-dimethanol
41 42
transi Cyclooctane-l,2-dimethanol cis
43
tiand
cis
142-144 ( 0 , 7 ) 134-136 (0.73
3636d 1.4967 36 1.5021
69.19
69 26
10 32
10.45
3640
3477 3536 351 1 3522 3498 3484 3476 3540sh 3480
3623' 3623' 363Jd 3625' 3623' 3636d 3623 363Bd
3498 3487 3496 3494 3487 3503 3490 3477
3636d 363tid
*
3636
*
3642 3636
0
101
139 101 144' 87 166'" 102 119m 140' 159'" 100
125" 114 138 158 160 96 160
...
..
1.6 2.2 2.2 1.9 1.9 2.8 1.8 2.5
Bicyclic diols 44
45 46 47 48 49 50 51
Norbornane-2,3-dimethanol
2-Norbornene-5,6dimethanol Bicyclo[2.2.2]octane-2,3dimethanol
endo-cis 61.6-62.6 c x o cis trans 119-120( 0 . 3 ) endo-cis 84 4-85 6 exo-cis 126-128 (0.3) trans 149-150 (1) cis 89 .4-90. 2 trans 76.3-77.0
*
37,38 37 ,5065 37 37,38 ,5203 37 ,5123 37 38
69.13
69.02
10.33
10.42
70.10 70.10 70 55 70.55
69.69 69 55 70.61 70.71
9.15 9.33 9.15 9.36 10.65 10.60 10.65 10.94
125( 138)'
136(1 49)e 139 131(142)e 136(149)e 133 133(146)' 159
1.0 1 .o 1 4 1.1 1 0 1.7 0.9 1.6
HYDROGEN BONDING IN DIOLS
Feb. 20. 1964
653
TABLE I (Continued)
K O
Compound
52 63 54 55
2-Bicyclo[2 2.2]octene-5,6dimethanol Compound V Compound V I
Configuration
M . p . or h . p . ( m m . ) , OC.
ando-cis 103.0-103.8 !raws 72 5-73.0 CIS
Lit. ref. 38
*
Carbon, Calcd Found 71.39 71.39
Hydrogen, 3 '5 Calcd. Found
vree
pbonded
9.70 9.70
3623' 3638d
3485 3488 3,505 3480
71.76 71.80
9.59 9.59
3639 3635
CiSO
3602w
i\U
Free/ bonded intensity ratio"
138(l5ljC
1 0
1.50
1 7 1 6 1 1
134 155 33p
*
This compound prepared in insufficient quantity for characterization; see text. a Extinction coefficient (peak height) free peak/ Peak unsymmetrical on the high frequency side. Peak unsymmetrical bonded peak. * 2-cm. cells used for these determinations. Band probably due t o on the low frequency side. e Av calculated from an assumed free peak position of 3636 cm.-l (see text). intramolecular hydrogen bonding t o t h e cyclopropane ring.7,33 0 Allylic or propargylic alcohol hydrogen b 0 n d i n g ~ ~ 3AV ~ ~computed ~~; Pure samples supplied by Dr. A . T.Blomquist and Dr. A. G. Cook, Cornell University. 1 Pure from high frequency shoulder. samples supplied by Dr. J , Sicher, Czechoslovak Academy of Science, Prague. The infrared spectra of compounds 40-43 were identical with similar curves by Dr. Sicher. 3 Hydrogen bonding t o bromine.8 Previously reported6 t o have Av = 137 cm.-l; also A v = 140 c n - 1 has been reported for another compound of this type." Previously reported6 t o have Av = 152 cm.-1. tis-4-t-Butylcyclohexane-trans-l,2-dimethanol ( t h e all-equatorial isomer) has been reported15 t o have A v = 152 cm.-'. A U = 171-176 cm.-l A v = 127 cm.-' has been reported for two compounds of this type.14 Pure has been reported for three compounds of this type.14 sample supplied by Dr. J . Meinwald, Cornell University. P Hydrogen bonding t o the double bond.8,3g,40
'
CHzOH
CHzOH (54)
v
much more sensitive indicator of the extent to which this is occurring. The inate intensity of a n OH . . . 0 peak is greater than t h a t of the free peak,*Ob so often both are of comparable height (Fig. 1 ) . Ideally, areas under each of these bands should be compared, b u t i t is easier t o measure their relative extinction coefficients. Such free/bonded intensity ratios (Table I ) should be related to the proportions of molecules engaged in intramolecular hydrogen bonding, for a given type of system. This assumption, a convenient one, is based on the supposition t h a t the extinction coefficients of both free and bonded peaks remain more or less constant from one 1,4-diol to another. This assumption does not appear unreasonable t o us. Primary alcohols are known t o have similar free peak extinction coefficient^.^^ Since the Av's of the 1,4-diols examined are roughly comparable, it is possible t h a t their extinction coefficients are also similar, but this is less amenable t o direct test. Obviously, if a diol had no bonded peak, it is not intramolecularly hydrogen bonded. We feel t h a t the distinct difference in free/bonded intensity ratios displayed by examples A and B in Fig. 1indicate a significant difference in the extents of intramolecular association between the two compounds. S o t all t h e 1,4-diols examined displayed an ideal two-peak (21) "Beilstein," 3rd Ed , Yo1 I , p. 2172; H Adkins and H. R Billica, J . A m . Chem. S o c , 70, 3121 (1948) (22) A. F . Shepard and J . I< Johnson, ibid., 64, 4385 (1932); B Wojcik and H Adkins. ibid., 66, 4939 (1933) (23) I
