OPTICALACTIVITYOF SATURATED CYCLICCOMPOUXDS
Oct. 20, 1959
B
+k(A
B
- H)(R - H)
-k(B
- H)(R - H) b
a XXIV
Permolecular Asymmetry.-I t is to be recognized that additional kinds of screw patterns of polarizability will be possible. Optically active allenes, spiranes and biphenyls show such patterns when viewed along their axes of asymmetry. These screw patterns, in effect, run through the molecule and will be called permolecular patterns; such patterns may be important in rigid and cyclic com-
[CONTRIBUTION FROM THE
5483
pounds. It is found necessary to invoke this concept in the treatment of carbohydrates (part 11). Acknowledgment.-The author wishes to acknowledge the helpful advice and encouragement given him during the preparation and repeated revisions of this manuscript by Professors H. C. Brown, N. Kornblum, R. A. Benkeser and R. A. Sneen of the Department of Chemistry and by Professor G. N. Wollan of the Department of Mathematics of Purdue University. He is particularly grateful to the Purdue Research Foundation and to the Research Corporation for grants which aided greatly in the preparation of this manuscript; he wishes to express his appreciation to the Department of Chemistry of the University of Wisconsin for its hospitality during the summer of 1957. LAFAYETTE,
IND.
DEPARTMENT OF CHEMISTRY,
PURDUE UNIVERSITY]
The Optical Activity of Saturated Cyclic Compounds1 BY JAMES H. BREWSTER RECEIVED SEPTEMBER 2, 1958 It is shown that the principles of conformational asymmetry and the empirical constants used in estimating the sign and magnitude of rotation of acyclic compounds (part I ) can also be used in estimating the sign and magnitude of rotation of saturated cyclic compounds.
A flexible chain compound will generally have a relatively small rotation because its molecules can assume many conformations with different and opposed rotatory powers. It was shown in part I that the rotations of many open-chain compounds of known absolute configuration can be estimated by use of a simple system of conformational analysis and a series of empirical rotation constants for asymmetric conformations about individual bonds ; this treatment provides an explanation for the regularities observed in the rotations of acyclic compounds. The relatively large rotations of ring compounds can, then, be ascribed to their near rigidity which allows a display of the full rotatory powers of their asymmetric conformations.8 We may, accordingly, test the rotation constants obtained in part I by determining whether they can be used in estimating the sign and magnitude of rotation of cyclic compounds. It is shown in the present paper that these constants can, indeed, be used for this purpose, whence it follows that indirect support for the system of conformational analysis used in part I has been obtained. An important result of this test is the demonstration that, although the Marker Rules2 cannot be applied as such to cyclic compound^,^ the principles which lead to rules equivalent to those of Marker can be used successfully in the cyclic series so that the apparent lack of corre(1) A Useful Model of Optical Activity, Part 11; part I, preceding paper, p. 5475. (2) R . E. Marker, THISJOURNAL, I S , 976 (1936). (3) W. 3. Kauzmann, J. E. Walter and H. Byring, Chcm. Reor., 86, 338 (1940). (4) J. A. Mills, Ckemisfry & Industry, 218 (1953).
spondence between the cyclic and acyclic series4 disappears.6 Cyclohexane Derivatives.-Each of the ring bonds of the chair form of cyclohexane is asymmetric, alternating dextro (Ia) and levo (Ib); since these rotatory effects cancel completely, the optical activity of cyclohexane derivatives can be ascribed wholly to the asymmetric conformations made by the substituents with ring atoms or with one another.6 An isolated equatorial substituent, replacing H * in I a or Ib, has no effect on the conformational asymmetry of any of the bonds; an isolated axial substituent, replacing H in I a or Ib, causes equal and opposite changes in the asymmetry of the two nearby ring bonds but should have no effect on more distant bonds. Accordingly the epimers I I a and IIb should have essentially the same rotation and this rotation should be close t o ( 5 ) It should be noted that the present treatment of cyclic compounds was, in many important respects, anticipated by D H. Whiffen, i b i d . , 964 (19513, in a discussion of the rotatory properties of the hydroxytetrahydropyrans, the cyclitols and certain of the terpenes. It should also he noted, however, that the present treatment leads to several major simplifications and extensions of Whiffen’s work which would have been impossible were all of Whiffen’s conclusions accepted (see text, below). I t is, thus, one of the burdens of this paper to show that the principles presented here and in part I lead to results comparable to those obtained by Whiffen, the net result being to provide a broader justification of the rsscnfial points of his thesis. The author wishes to make it clear that no part of the present paper is to be regarded as an attack on Whiffen’s important pioneering contribution. (6) Asymmetrically distributed substituents may produce asymmetric distortions of the ring which could make the ring itself optically active. These effects will probably be small but might account for the small rotations observed in compounds otherwise expected to be optically inactive. These possible effects will, for the present, he ignored.
5484
JAMES
1\/IOLECULAR
H. B
TABLE I ROTATIONS O F 3-SUBSTITUTED CHOLESTAXES
Substituent
CI
[MID
B
a
..
93"
F
c1 Br NHz NMez OH OCHZ OBz COzH COnCHa CONHn CN CHaCOzH CHsCOiCHa CH?CHzOH CHzCH2CH3 CH(COzH)* CH( C02CHs): CHpC( CH8)nOH CHzC( CeHa)zOH CH=C( CH3)n (OCH3)z S*CzH,
906 110"
124" 129" 10IjC 91d 97" 73" [MIS:*
112c 96"J 90" 85" 99" 1218 128O 109' 1098 99' 86" 76"
llsa 1000 11V 91' 840 1218 109"
105"
prediction. The molecular rotations of a wide variety of 3-substituted cholestanes, as shown in Table I , average to +lOO", in good agreement with the value for cholestane itself (+93'). The average deviation (excluding the seemingly anomalous compounds in part b) from +93' is 160j9the largest effects, on the whole, being shown by axial ( a ) substituents of high polarizability.
H
H
+k(C
- H)'
-k(C
- H'i2
b
i
I
IO?
142h 1OO'G
102" 77h x9h
b NHAc 155' j2C.C NMe3'I 184d 0.kc 129' 56' 42 CHs CH=C( C6Hs)z 64h 22h Calculated from average values for specific rotation as cited by J. P. Mathieu and A. Petit, in "Tables de Constantes et DonnCes XumCriques. 6. Constantes Selection6es. Pouvoir Rotatoire Naturel. I. StCroides," Masson C. W. Shoppee and G. H. R . Sumet Cie., Paris, 1956. mers, J . Chem. Soc., 4813 (1957). C. W .Shoppee, D. E. Evans, H. C. Richards and G. H . R . Summers, ibid., 1649 R . D. Haworth, J . McKenna and R. G. Powell, (1956). ibid., 1110 (1953). 6 D. P. Dodg:on and R . DI Haworth, ibid.,67 (1952). J I,. LBbler, T'. Cernp and F. Sorm, Coll. Czech. Chem. Comm., 19, 1249 (1954). G. Roberts, C. W. Shoppee and R . J . Stephenson, J . Chem. SOC.,2705 (1954). C. W.Shoppee and R . J. Stephenson, ibid., 2230 (1954). R . H . Baker, L . S. Minckler and Q . R. Petersen, THISJOURNAL, 77, 3644 (1955). i R . M . Evans, et al., J . Chem. Soc., 1529 (1958). D. K. Fukushima, S. Lieberman C. \V. and B. Praetz, THISJOURNAL, 72, 5205 (1950). Shoppee, B. D. .igashe and G. H . R . Summers, J . Chevn. SOC.,3107 (1957).
b
t
I1
99' 137k
-,. I \ H
that of the compound in which X = H.7 The relatively small rotations of the 3-methyl cyclohexanols (IIIa and IIIb)8 are in accord with this (7) T h i s conclusion was reached earlier by IV. >I. Stokes and \V. Bergmann, J . Ovg. C h e m . , 17, 1194 (1952), and by W. Klyne and W. h l . Stokes, J . Chem. Soc., 1979 (1954), o n somewhat more intuitive grounds. I t would be naive t o expect t h a t both epimers would have exactly t h e same rotation as t h e parent compound since permolecular asymmetry and second-order atomic asymmetry effects (see part I ) could still be displayed. Indeed these minor effects might most profitably be studied in systems such as these, though such studies could be complicated by t h e possibility t h a t asymmetric distortions of rina geometry will also produce detectable rotatory effects 6 (8) T h e two (-)-alcohols can be oxidized t o the (+)-ketone [G. .I. Gough, H. Hunter and J. Kenyon, i b i d . 20.52 (1926); A . K. h,facbeth and J. A . Mills, ibid., 205 (194711; t h e absolute configuration of t h e ketone is established by its oxidation t o 8-methyladipic acid (see part 111, refs. 11 and 12). Relative configurations (cis and frans) are assigned in accord with t h e findings of H. L. Goering and C. Serres, THIS
b
a
1-
Adjacent substituents can produce a marked conformational asymmetry ; the size of this effect can be estimated by a consideration of the rotational changes which would accompany a stepwise building up of the structure in question from a parent cornpound which is optically inactive or has a known J O U R N A L , 74, 590s ( l L J i 2 l . and o f D S. N u v c e and D. I 3 Ilenney, ibzd., 74. 6912 (1952). (9) An error of 0.01Oin t h e observed rotation of a 1 % solution of a compound with a molecular weight of 400 would produce an error of this magnitude in t h e molecular rotation.
OPTICAL*ICTIVITY OF SATURATED CYCLIC COMPOCNDS
Oct. 20. 1959
rotation. Thus, IV is the optically inactive parent compound of the 1,2-disubstituted cyclohexanes having the group A equatorial. Replacement of the equatorial hydrogen at position 2 (H* in Va) by B produces the conformation Vb, which has a rotation + k ( A - H) (B - H) higher than Va. The conformation about the bond 2-3 is not altered (see above). Hence diequatorial VI, the product of this transformation, should have the indicated rotation, I V being inactive. Similarly, if the substituent enters axial, to produce VI1 (C axial), the new substituent produces equal and opposite conformational changes relative to the ring atoms but interaction of the substituents make the conformations about bonds 1-2 and 2-3 more levorotatory by a total of - k ( A - H) (C - H). The other rotational changes shown in VI-IX can be worked out in a parallel way; these expressions, used in conjunction with the data of Table I, part I, constitute predictions of the sign and magnitude of rotation of the simple compounds shown or of the rotational shifts which should accompany conversion of analogs of IV to the corresponding analogs of VI-IX. For the most part, these predictions parallel the empirical conclusion of Klyne and Stokes7 that X will be dextrorotatory when C* is substituted. Note, however, that in the present picture one form of VI should be inactive and that in some cases VII, VI11 and I X could be levorotatory. The present rules should be much more general than that of Klyne and stoke^,^ which was developed for thepa rticular case where the substituent A (in X) is a hydroxy group.
Q:
H
A
diequatorial : S k ( A - H)(B - H ) diaxial: 0
A equatorial: -kfA - H l ( C - H ) A axial : + k ( A - HliC - H)
A equatorial: + k ( A - H ) ( B - C) A axial: +k(A - H)(C - H )
A equatorial : + k ( A - H)(B - C) -k(D - H ) ( B - H ) A axial: S k ( A - H)(C - H ) +k(D - H)(B - C ) IX
D A
VIII
I
I
/
CH2 C*
y
g’ \A
For a test of these predictions we require rotation constants beyond those given in Table I, part I, as, for example, for the conformation X I . Such constants can be obtained by suitable combinations of those given in Table I, part I ; thus, from the values given for conformations XI1 and X I I I , we can estimate the value for X I to lie between 36’ and 48’. lo (10) T h e rotation constants for conformations X I 1 and X I I I are rounded t o the nearest five degrees.
54s5
k(o - H ) 2 = [ k ( O - H)(C - H)I2 - (50 f 2.5)‘ 60 f 2 5 k(C - H)’
Whiffens has shown that a value of +45’, which lies in the calculated range, can be used to calculate the sign and magnitude of rotation of eight polyhydroxy cyclohexanesll; this value, being suitably rounded, will be used here. It is important to note that estimates of this nature could not possibly be made by use of Whiffen’s treatment. In Table I1 are presented the rotatory shifts ( [ h l ] o ~ - ”to) be expected when a hydroxy group is placed on rings A, B or C of the steroid nucleus12 (ring D is treated separately below). This table can be used in predicting the rotatory effects of other substituents simply by replacing the numbers *50 by the value =kb(X - H)(C - H) as taken from Table I, part I. Additional values must, of course, be used when two adjacent substituents are considered. It is seen that for most cases the rotational difference between epimers is very closely predicted and that, in general, the directions of rotational shifts (relative to the unsubstituted compound) are correctly predicted, indicating the general correctness of this treatment. It will also be noted that sizeable discrepancies occur when predictions of individual rotation shifts are made; the fact that most of these discrepancies disappear when epimers are compared shows that they are produced by a persistent “anomaly.” A large mass of rotational data which have been accumulated suggests very strongly that this anomaly is produced by a flattening of the rings (to relieve conformational pressure exerted by the angular methyl groups on@-substituentsa t Cz, Cq, Cg, Cg and Cll and to relieve crowding of l@-, 7@- and 1la-substituents) ; these results will be presented in due time. I n the meantime, i t is of interest t o observe the extent to which correct predictions can be made by use of the presently accepted “full-chair” conformation model of the steroid nucleus. In Table I1 are presented rotations as predicted and as observed for monoamino-, hydroxy- and halo-cholestanes (A/B trans) (calculated from [MI f-93 for cholestane). Excluding the three italicized values, which are anomalous in magnitude but not in direction of shift, the average deviation from predicted values is 20’. I n Table I V are presented predicted and observed values for the hydroxycholanic acids. (and their esters) (A/B cis) ; this table includes a (11) See Whiffen’s paper6 for a tabulation of these results. The calculations for the cyclitols are in every respect identical t o those t h a t would be made under the present treatment. I n his further calculations, however, Whiffen uses an additional value of +32O for X I when it includes a glycosidic hydroxy group, and also a value of + 3 4 ” fot X I 1 (compare t h e present value of 50’). In these two regards, Whiffen’s treatment stands in contradiction t o t h a t presented here (see below). It is shown below the present values are adequate for calculating rotations of hydroxytetrahydropyrans and of terpene and steroid alcohols, and are, in most cases, superior t o those of Whiffen (12) The absolute configuration of the steroid nucleus has been shown t o be t h a t given below by t h e work of W. G. Dauben, D F. Dickel, 0. Jeger and V. Prelog, H e h . Chim. Acto, 36, 325 (1953), B. Riniker, D. Arigoni and 0. Jeger, abid., 37, 546 (1954); J. W. Cornforth, I. Youhotsky and G. Popjak, Nature, 173, 536 (1954); I(. Brenneisen, C. T a m m and T. Reichstein, Helv. Chim. Acto, SO, 1233 (1956). This relationship was predicted from rotational analogies with the terpenes by J. A. Mills, J . Chem. SOC.,4976 (1952); ref. 4.
5486
JAMES
H. BREWSTER
TABLE I1 ROTATORY EFFECTS OF HYDROXY GROUPS IN RINGS A , 13 A N D C OF THE STEROID NUCLEUS Position
A / R trans
la
[111,p
1MIOH-H
Calcd.
+50
Obsd."
$35 17 +37 49 $5 -2 -75 22 -25b
-
Calcd.
+
TABLE I11 ROTATIONS OF MONOSUBSTITUTED CHOLESMOLECULAR TANES
Obsd.
SubPosistituent tion
+
None ?iH*
0 50 52 0 0 0 - 12 B 3a 0 0 0 +7 B 40 -50 - 100 -97 50 B 0 5ff 6a +59 50 -50 -50 - 105 100 B 7a -50 -59 $110 - 100 - 169 50 B A/B cis lff 0 - 50 50 B 2a 0 0 0 B 0 3a $30 0 0 +1 B $29 4ff -50 - 100 50 B 50 58 - 100 6a -50 0 -50 - 107 +7 B -50 -79 7a - 174 - 100 $95 50 B Both series 50 8B -50 90 -50 -29 llff - 100 - 12.5 96 +50 B 12a 50 93 0 50 50 43 B Average values as cited by W. Klyne in E. A. Braude and F. C. Nachod, "Determination of Organic Structures by Physical Methods," Academic Press, Inc., New York, 1955, p. 111. b See Table V. N. Y.,
B
2a
+
+ +
+
-
+ + ++ +
+ + +
+
+
number of vicinal diols The persistent anomaly shown by 3a-hydroxy compounds in this series has been taken into account by use of the value [MI +l38" for 3a-hydroxycholanic acid (to be compared with the value f87' for cholanic acid). Excluding the two italicized values, which appear to be anomalous, the average deviation from the calculated values is 21'. A 50-hydroxy substituent also produces a small but persistent anomaly; as seen in Table V, rotations for a number of polyhydroxy cholestanes can be calculated (average deviation 15") when this effect has been taken into account. The data in Table VI provide a striking illustration of the "steroid anomaly" when individual rotations are calculated and an even more striking illustration of the disappearance of this anomaly when rotations of epimers are compared. It seems fair to conclude that the success with which differences in rotation of epimers can be calculated in this series provides strong support for the present treatment of optical activity. It must equally be concluded that the magnitude of individual discrepanciesl,indicates that some persistent factor is not being taken into account in these calculations; this factor could be: (a) an essential error in
frUflS)a"
[bf1
7-
a
Calcd.
Obsd.
Calcd.
Obsd.
93 93 148 38 148 93 93 93 143 43 143 93 93 - 77 263
110" 1126 233" 24c 209" 76e 13Zf 90 1209 31 202" goh 110 -183 313
93 2 3 4 6
-
i
OH
1 2 3 4 6 7
F
3
+
+
Vol. 81
93 I 1 105b 93 41" 38 14gc 148 - 77" 38 143 135d; 128" 93 105/; 10la 97 93 23O 43 143' 143 43 43 --(I
c1
3 124 93 6 208 263 7 - 77 -86 Br 3 93 129 7 -87 -90 LIa All rotations in chloroform. Unless otherwise noted, all rotations calculated from data cited by J. P. Mathieu and A. Petit, ref. a , Table I. C. W. Shoppee, R. J. W. Cremlyn, D. E. Evans and G. H . R. Summers, J . Chem. Soc., 4364 C. W. Shoppee, D. E. Evans, H. C. Richards and (1957). G. H. R. Summers, ibid., 1649 (1956). C.1%'. Shoppee, D. E. Evans and G. H. R . Summers, ibid., 97 (1957). H. B. Henbest and R. A. L. Wilson, ibid., 3289 (1956). e P. Striebel and C. Tamm, Helv. Chim. Acta, 37, 1094 (1954). f A. Fiirst and P. Plattner, ibid., 32, 275 (1949). 0 D. S . Jones, J. R . Lewis, C. W.Shoppee and G. H. R . Summers, J . Chem. Sac., 2876 (1955). C. W. Shoppee and G. H. K. Summers, ibid., 4813 (1957).
the present treatment, (b) the occurrence of permolecular asymmetry effects or (c) the use of an incorrect conformational model for the steroid nucleus. As pointed out above, work in progress points strongly to the last-named possibility. OH
OY
k(0
- H)' = +45" XI
(C
k ( 0 - H) - H ) = +50" XI1
:+s
k(C
- H)* = +60" XI11
These principles can also be applied to the alkyl cyclohexanols, but it must be remembered that some of these compounds can have two conformations of roughly the same energy. Thus, trans-2rnethylcyclohexanol (in configuration XIVa) l 3 has a molecular rotation of +44.2', in acceptable agreement with the value of +50° predicted for the diequatorial isomer. On the other hand, the related cis isomerla XIVb has a rotation of -15', (13) Relative configurations are assigned on t h e ground t h a t sodium reduction of t h e ketone will give chiefiy the diequatorial (trans) isomer. Both (+)-XIVaand (-)-XIVb can be oxidized to the (+)-ketone*. the toluenesulfonate of (+)-XIVagives, i n hot formamide, a levorotatory olefin: the optical activity of which is probably due t o the presence of (-)-3-methylcyclohexene. T h e (+)-form of the olefin has been obtained from (+)-3-methylcyclohexanone [M. Mousseron, R. Richaud and R. Granger, Bull. 3oc. chim. Frunce, [51 13, 222 (1946)l.t h e absolute configuration of which has been established (references 11-13 in part 111).
OPTICALACTIVITYOF SATURATED CYCLIC COMPOUNDS
Oct. 20, 1959
TABLE IV MOLECULAR ROTATIONS OF HYDROXYCHOLANIC ACIDS(A/B cis) [MID
Substituents
Calcd."
Acid
0bsd.b methyl ester
....
87 80 C 95 A 137 M 138 132 E 3a-OH 88 33 M 142 M 3a-OH, 6a-OH 138 145 M 3a-OH, 6p-OH 88 45 E 49 c 3a-OH, 7a-OH 188 224 E 3a-OH, 7p-OH 88 86 E 88 E 3a-OH, Ila-OH 188 2 1 6 M 199 A Ba-OH, llp-OH 188 208 E 227 A 3a-OH, 12a-OH 138 151 D 176 A 3a-OH, 12p-OH 3a-OH, 7a-OH, 12a-OH 138 1 4 7 E 106 E 3a-OH, lla-OH, lZa-OH 93 3a-OH, Ila-OH, 126-OH 133 188 E 3a-OH, 116-OH, 12a-OH 238 221 E 3a-OH, 116-OH, 126-OH 143 175 E 87 97E 36-OH 37 12 M ; 20 E 3P-OH, 6a-OH 87 58 M 3/3-OH, 6p-OH 7a-OH 37 26 C 87 106 E icu-OH, 12a-OH lla-OH 37 194 -4 42 13 D 50 M lla-OH, 12a-OH 92 227 M IlD-OH, 12P-OH 137 164 A 12a-OH 12p-OH 87 143 A a Rotations for cholanic acid derivatives containing a 3ahydroxy group are calculated from the value [MIDf 138 for the parent hydroxy acid; all other rotations are calculated from the value [MID +87 for cholanic acid. These values were chosen to give the best fits with the data cited above. I n these calculations it is assumed that the acids and methyl esters will have equal molecular rotations. Average values calculated from data taken from the tables of J. P. Mathieu and A. Petit, ref. a , Table I : A, acetone; C, chloroform; D, dioxane; E , ethanol; M, methanol.
5487
TABLE VI EPIMERIC VICINAL ~YU~S-DISUBSTITUTED CHOLESTANES Substituent
Conformation
Calcd.
[MID
Obsd.
IMI..-..
Calcd.
Obsd.
ee 48 73" 2a-OH, 3p-OH $45 +60 aa 93 133" 2p-OH, 3a-OH - 24 63b ee 20r-C1,3p-OH $103 93 166b +117 aa Zp-Cl, 3a-OH - 57 56b ee 2a-Br, 36-OH +150 4-149 aa 93 205b Pp-Br, 3a-OH -235 -2ib ee 2a-C1, 3p-Cl f258 93 261b f 3 2 8 aa 28-C1, 3a-Cl -327 -81b ee 2a-Br. 3p-Cl f420 $408 93 32ib aa 2p-Br, 3a-Cl -327 -76b ee 2a-C1, 3p-Br +3i9 93 303b $420 aa 26-C1, 3a-Br -447 -152b ee 2a-Br, 36-Br f540 +556 aa 93 4046 26-Br. 3a-Br 88 79d ee 36-OH, 4a-OH 4-55 -13 aa 143 66e 3a-OH, 46-OH 433 192b ee 36-Br, 4a-Br -160 -160 273 32b aa 3a-Br, 4p-Br ee +138 +125' 56-OH, 6a-OH -95 -137 f43 -12' aa 5a-OH, 6P-OH 613 261" ee 5p-Br, 6a-Br -87 -215' -700 -476 aa 5a-Br, 66-Br 5 C. W. Shoppee, D. N. Jones and G. H. R . Summers, J . Chem. Soc., 3100 (1957). * G. H . Alt and D. H . R. Barton, ibid., 4284 (1954). C. A. Grob and S . Winstein, Helv. Chim.Acta, 35, 782 (1952). L. F. Fieser and R . Stevenson, THISJOURNAL, 76, 1728 (1954). e P . A. Plattner, H. Heusser and A. B. Kulkarni, Helv. Chim. Acta, 31, 1822 (1948). f D. N . Jones, J . R. Lewis, C. 1%'. Shoppee and G. H . R . Summers, J . Chem. Sac., 2876 (1955).
conformation to one of the latter; conclusions of this nature are, necessarily, very approximate. I n the terpene series14 (Table VII), i t is to be expected that the trans-p-menthane derivatives will have their alkyl groups essentially completely in TABLE V the equatorial orientation. If so, the rotational MOLECULAR ROTATIONS O F CHOLESTAN-5a-OLS difference between epimeric alcohols, A(A - B ) , [MIAdditional substituents Calcd. Obsd. should be about 100'. The observed identical .... 68 39 C"'d values of +log0 in the menthol and carvomenthol 2a-OH 68 79 C" series indicate that the group R1 (see figures in 68 69 C" 3a-OH Table VII) makes no more contribution to rotatory G8 77 C" 36-OH power than do the substituents in 3-alkylcyclo4a-OII OR 57 cb hexanols (see 111). They indicate further that the 46-0H 118 109 Cb isopropyl group in the menthol series produces no 6a-011 68 01 Cb significant rotatory efect.14 In the other trans6p-OH 18 -12 Cb menthane derivatives, the rotation difference is 118 183 Pa 3a-OH, 4p-011 also close to 100' (observed +102', +105'); 68 88 Dn 3p-OH, 6a-OH these values are to be compared with the dif3p-OH, 66-OH 18 13 E" ference of +68' expected from Whiffen's value of Calculdted from data tabulated by J. P. Mathieu and -4. +34' for XII." Conformational mixtures are to 0
Petit, ref. a , Table I. D. N. Jones, J. R. Lewis, C. W. Shoppee and G. H. R . Summers, J. Chem. SOC., 2876 (1955). Calculated from the value + 6 8 O for cholestan-5a-01, C, chloroform; D, dioxane; chosen to give the best fit. E, ethanol; P , pyridine.
intermediate between the value of -50' expected for the conformation having an equatorial methyl group and the value of +50° expected for the conformation with an equatorial hydroxy group (see VIT). Indeed, the observed rotation indicates the presence of roughly two molecules of the former
(14) The absolute configurations of t h e rnenthones and carvornenthones are established by the work cited in references 11-13 of part 111. T h e relationships of t h e alcohols and amines t o the ketones and tbelr relative configurations are reviewed in J. L. Simonsen "The Terpenes," Vol. I, 1947, Cambridge University Press, Cambridge. The relative and absolute configurations of the dihydrocarveols and their hydrates have been established by the work of H. Schmidt, Chem. Ber., 83, 193 (1950); 88,463 (1955). I n menthol and isomenthol, all conformations of the isopropyl groups are "prohibited" by the rules of conformational analysis given in part I ; this group may either be free t o rotate (and so act as a cylindrically symmetrical substituent) or be frozen in ita symmetrical conformation. In t h e neo-sezies, the symmetrical conformation of this group is the only one "allowed."
5488
JAMES H.
hIOLECULAR
RorAlroxs
Vol. 81
BREWSTER
TABLE, VI1 TERPEYE hLCOHOLS .4KD
OF
AMINES
1
+61I +55 x
-OH 2"-
-OH --SH,
- C-CH, -C(CH8):
-OH -OH
-CH3 -CH(
I
OH a Data froni references cited
ill
+77 f69 +43 +20
-j i ! - 55
+lUC! +I10
-31
+lo8 t 92 108 62
-23 -65 - 42
+XI +55
(1 iJ
+
+
+
$2 +45 +28 +22
+50 +55
-50 - 55
+ 0.21 + 3.6 $55
+52 +36
footnote 14.
be expected in those members of the cis-menthane the iiisertion of single substituents oii the otherseries in which the two small substituents could be wise unsubstituted D ring of steroids is provided equatorial when the large one is axial. Thus, the by the data in Table IX; on the whole the agreerotation of neoiso-menthol (+0.21") suggests that ment is satisfactory. It is a simple matter t o the two forms are present in nearly equal amounts as extend this treatment to more highly substituted does the rotation of iso-carvomenthol ( + B o ) .compounds, such as estriols and androstan-3,16,17The rotations of iso-dihydrocarveol (+43') and triols (XXIV) where excellent agreement is obits hydrate (+41') are out of line, being consider- served. ably larger than expected. The present analysis Monocyclic cyclopentanes are, in principle, more indicates t h a t the rotations of the amines should complex to analyze since ten puckered conformabe parallel to those of the corresponding alcohols as, tions (of possibly unequal energy) should be conon the whole. they are. sidered. A major simplification is provided by the assumption t h a t the average conformation will be equivalent t o the fully planar conformation. To Q.OH c H., QOHCH, illustrate the utility of this approach, let US consider the nepetic acids,ls having the relative and ~
[MID +41."08 a
XI\'
[MID -15" b
.A stringent test of the utility of the present
conformational asymmetry effects a s a result of the sterically asymmetric environment produced by t h e angular methk-1 proup a t CIS. These effects will be considered in more detail in a m o r e complete review of rotatory effects in t h e steroid series (17) These calculations can only be made by assuming t h a t the rota tory power of the skew chain of four atoms, X-C-C-Y, increases with angular separation t o goo and then decreases t o zero a t 180", in t h e lrans confolmation, as, for example, in t h e simple expression
treatment is provided by its application to the prediction of the rotator>- effects of substituents in ring D of the steroids. The five-membered ring is forced into a puckered conformation by fusion to ring C; its most probable conformation is one [M] = -+ sin d . k ' X Y in which C13, C14 and Clti are coplanar, with C13 which will be used here. For correlation rvith values presented above and Cli below the plane.!5 This conforma- Table I , part I, this expression becomes tion is awkward to handle under the present treatment but we may approximate i t by noting that i t is intermediate between the two sinipler conforma- T h e values used for commonly ohserved angular separdtlons are then tions, XVa, in which Cla is the only nori-planar a [a11 atom. and XI,7b.in which C14is the non-planar atom. O o , 180' 0 Rotatory effects will then be the ai-erage of those predicted froni these two conformations. The bond conformations are as shown in XI'I-XXIII. I n Table VI11 are summarized the rotatory changes 2k 90 to be expected when a hydrogen atom a t C14)Clj, d3xy C16 or C1: is replaced by a simple ~ u b s t i t u e n t . ' ~ ~ ~ ' On thls basis, Corresponding staggered and eclipsed conformation5 ;1coniparison of predicted with observed values for should have t h e same rotatory power. ill
(1.5) See C. \V. Shoppee, K. H. Jenkins and G. H. R . I . Cheiii. Soc., 3048 (1958), for a review of this uoint
Summers,
(16) hlorr complex substituents, as f w n r i in the etio acids, t h e pregnan-2O-ones, the clir,lestanr- mid the c h ,lciliic .*ciils, clio\\ iidditioniil
(18) S. M. McElvain and E. J. Eisenbiaun, THIS JOURNAL, 1 7 , 1.599 (1955): E. J. Eisenbraun and S . M. McElvain, i b i d . , 7 7 , 3383 (1955): R. B. Rates, E . J. Eisenbraun and S. lf.McKlvnin, ibid., 80, 3413, 3120 (1!)3S)
OPTICALACTIVITYOF SATURATED CYCLIC COMPOUNDS
Oct. 20. 1959
5489
,CH,
Predicted: +k(C02H Predicted: + k ( C H)2; 1-135" H)(CO*H - H ) Observed: [MID 15O0l8 -k(C02H - H)2;-450 Observed: [MID -60"18 a b
+
CH3
CH3.
a::;:0. -C OzH "C02H
Predicted: +k(C - H,) (C02H - H ) ; +90 Observed: [MID +12Oo1* C XXP
Predicted: 0 d
given the relative configurations of the camphoric acids (XXVII),l 9 their absolute configurations can be assigned, this leading to an absolute configuration for (+)-camphor (XXVl) in agreement with the conclusioris of Fredga. 2o
CIs-Ci6 (XVb) XXT
wx
Predicted: +k(C H)2; $60'
HO --.
s H
Y
H
Calcd.
..--____(AT]D--
Obsd.
Calcd.
-- -Obsd.
0 -59 E"
226 E" a-OH -56 170 167 E" n-OH @-OH -34 192 ai-OH -24 &OH -6 E' 202 190 E d 8-OH 8-OH +39 4-40 E* 265 254 E" a L. Ruzicka, V . Prelog and P . Wieland, Helu. Chim. Acta, 2 8 , 1609 (1945); 28,250 (1945). * iY.S. Leeds, D. K. Fukushima and T. F. Gallagher, THISJOURNAL, 76, 2943 A. Butenandt and U. Westphal, 2 . physiol. (1954). Chem., 223, 147 (1934). A . Butenandt and J . S. L . Browne, ibid., 216,49 (1933). e M . N.Huffmanand H . H . Darby, THISJOURNAL, 66, 150 (1944). XXII.
ai-OH
absolute configurations shown in XXV. I t is seen, from these calculations that, given the relative configurations, the absolute configurations could be assigned simply from the signs of rotation. Alternatively, given the absolute orientation of the methyl group and the cis, trans relationships of the carboxy groups, the remaining stereochemical relationship (of the methyl group and the adjacent carboxy group) could, in each case, be decided from the sign and magnitude of rotation. Similarly,
\
CO,H
-
Observed: [MID +95.5' XXVIIa
-
Predicted: +k(C H)2 2k(C - H)(COlH - H ) ; - 120" Observed: [h'f]D -97.2" XXVIIb
Absolute configurations can now readily be as signed to a number of other dibasic acids (XXVIITThe configuration assigned to ( - ) =I). ? I -% trans-caronic acid (XXXI) is consistent with its formation from (+)-trans-chrysanthemic acid, 2 1 the configuration of which has been established2'; note particularly the important rotatory effect of the gem-dimethyl group. Barton and NickonZ6 have assigned the absolute configuration X X X I I to caryophyllene, basing their conclusions on the rotational properties of rearrangement products and the rules of Stokes and Bergmann and of Klyne and Stokes7 The same conclusion is reached somewhat more directly by noting that the (19) 0.Aschan, A n n . , 316, 209 (1901);R'. Hartmann, Ber., 21, 221 (1888). (20) A. Fredga and J. K. Miettinen, A d a Chrin S r n v d . , 1, 371 (1947). (21) H. Staudinger and L. Ruzicka, H e i v . C h i m . A&, 7 , 201 (1924). (22) L. Crombie and S. H. Harper, J . Chem. SOC.,470 (1954). (23) L. J. Goldsworthy a n d W. H. Perkin, i b i d . , 105, 2639 (1914). (24) L. J. Goldsworthy, i b i d . . 125, 2012 (1924). (25) E.Buchner and R . v. d . Heide, B e r . . 38, 3112 (1908). (26) D.H.K . Barton and A . Nickon, .I. C h r i r i . .SO( 1BI,J f l O i l )
5490
JAMES
H.BREWSTER
VOl. 81
TABLE VI11 CALCULATION O F ROTATORY SHIFTS FUR SUBSTITUTION I N Substituent
XVa
17P
0
- H)(C
n
1GP
4 3
+k(X l&Y
Average
- H)
+O.Sk(X - €I)(C - 11)
0 -k(X
--2k(X
11
0
k(X 17a
RI>G
(X- H) XVb
- H)(C
- 13)
- H)(C - H )
k k ( X - H)(C - 11)
4
- H)(C - H )
-R(X - Hj(C
- H)
- 1-1)
*!(X
- H)(C -
+k(X
- H)(C - H )
fl
- l . l k ( X - II)(C
H)
-0.3k(X
- II)(C - II)
-0 XR(X
- I1)(
-k(X - H ) ( C - 11) 15p
-k(S
- II)(C
- 1%)
-2k -(X
4 3
*(X
- H)(C - €1)
0
+k(X
- H)(C - H)
Fk(S - H ) ( C - H )
fl
15a
+k(Y
- H)(C
f2rZ(S- H ) ( C - 13)
+k(X
- €I)(C - H )
fl
4
c-11)
4
- II)
L k ( X - II)(C 14a
- H)(C - H)
0
0
0
0
- H)
+l.Bk(X - II)(C
-
ii:
+I.Sk(X - H ) ( C - H)
TABLE IX ROTATORY SHIFTSFOR SUBSTITUTION IS RINGU, OF STEROIDS
lenic acid (XXXIV)*'e is distinctly less dextrorotatory. It will be noted that the rotations predicted on the assumption that the ring is planar do not agree well with those observed, but that rotaA[Mlx- A Sub-a8tions predicted on the assumption that the rings are stituent Calcd. Obsd. Calcd. Obsd. puckered, as shown in XXXIII and XXXIV, agree 14-CHa $108 +73 & 17" quite 15-OH +so +73b -40 -11Ob AS a final example of the application of this 16-OH -15 -36c -15 -2OC treatment t o alicyclic compounds, consider the 17-OH -55 -50d +40 +20d two alcohols (+)-borneol and (-)-isoborneol 17-CHs - 66 +48 $20 +.z 10" formed by reduction of (+)-camphor (XXVI).'' a Calculated from data for 14 a-niethylcholestan-3~-ol, its acetate and ketone (D. H. R. Barton, D. A. J. Ives and The parent hydrocarbon is optically inactive, as is B. IC. Thomas, Chemistry 6 Industry, 1180 (1953)) and for the conformation about the bond between CI and 4,4,14a-trirnethylcholestanolderivatives (D. H. R. Barton, C z in each alcohol; the bond between CZ and cs J. E. Page and E. W.Warnhoff, J . Chem. SOL, 2715 (1954); is, however, asymnietric in opposite senses in the D. H . R. Barton and B. R. Thomas, ibid., 1842 (1953); W. Voser, M. Montavon, H. H. Gijnthard, 0. Jeger and L. two alcohols and is the sole conformational source Ruzicka, Helv. Chim. Acta, 33, 1893 (1950)). * Single of optical activity. On this basis, all derivati17cs compounds in methanol; reference compound in dioxane; of (+)-camphor having the relative corifiguratioll data of S. Bernstein, L. I. Feldman, W . S. Allen, R. H. XXXVa should have [M!D = k ( X - H) Blank and C. E. Linden, Chemistry 6 Industry, 111(1956). (C - Hj while those having configuration XXXVb Values cited by C. W.Shoppee, R. H. Jenkins and G. H. R. Summers, J . Chem. Soc., 3048 (1958). Average values as should have an equal but opposite rotation. T h e cited by W. KIyne, ref. Q, Table 11. e Average of three val- assignment of configuration XXXVa t o (+)ues: data of L. Ruzicka, P. Meister and V. Prelog, Helv. borneol is consistent with the conclusion reached Chim. Acta, 30, 867 (1947); S. A. Julia and H. Heusser, by Asahina, et al.ao on chemical grounds. Since ibid., 35,2080(1952).
+
dimethyl esters of the degradation products : trans-nor-caryophyllenic acid (XXXIIIa) ,27 transcaryophyllenic acid (XXXIIIb) and transhomocaryophyllenic acid (XXXIIIc) are all strongly dextrorotatory while cis-homocaryophyl-
*'
*'
(27) (a) W.C . Evans, G. R . Ramage and J. L. Simonsen, J. Chem. Soc., 1806 (1934); (b) G. R . Ramage and J. L. Simonsen, ibtd., 533 (1935); (c) T. L Dawson and G. R. Ramage, ibid., 3523 (1950); ( d ) H. N. Rydon, i b i d . , 593 (1936); 1340 (1937); (e) V. Jarolim, hl. Streibl, L. Dolejs and F. Sorm, Coil. Cscch. Chcm. Comm., 22, 1266
(1957).
(28) Electron diffraction studies indicate puckered structures f o r octafluorocyclobutane [H. P. Lemaire a n d R . L. Livingston, J . Chon. P h r s . , 18, 5F9 (1960)] and for cyclobutane [J. D . Dunitz and v. Schomaker, ibid., 20, 1703 (1954)l; X-ray studies indicate a similar structure for octachlorocyclobutane [T. B. Owen and J. L. Hoard, Acto C r y s f . ,4, 172 (1951)]. In each case dihedral angles of about 20' have been suggested. In t h e present system, one puckered form, in which a methyl group and t h e carbomethoxy group across t h e ring are crowded together, should be much less stable t h a a t h e other puckered fcjrrn, in which these two groups are quite far apart. T h e diagrams X X X I I I and X X X I V show t h e more stable puckered forms. (29) A. Ilaller, A n n . chim., [ V I ] 27, 392 (1892). (30) Y.Asahina, hl. Ishidate and T. Sano. Ber., 69, 313 (1936).
Oct. 20, 1959
OPTICAL ACTIVITY OF
r--fCozH Predicted: -k(COnH - H)2; Predicted: -k(CO,H - 135' H)a; -135" Observed: [MID -138OZ8 Observed: [MID -179°24 XXVIII XXIX
a
a
A02H
.,C02H
-
'C02H Predicted: k(C02H H)' - 2k(C - H ) (COrH - H ) ; -45' Observed: l. M. l ~-52.5OZ1 XXXI
5491
are assigned bornyl configurations (XXXVa). The (-)-halides formed by addition of H X to ( +)-camphenea7 (XXXVII) are assigned isobornyl configurations (XXXVb) in analogy with the formation of (-)-isobornyl acetate from (+)camphene.88 These conclusions are consistent with the present rule and with the views of Nevell, de Salas and Wilson39 on the stereochemistry of these additions.
CH3--- CH
'CO2H Predicted: -k(COyH H)'; -135' Observed: [MID -llOos
xxx
SATURATED CYCLIC COMPOUNDS
x--
@
H--
6 b
a
X -OH -SH -NH2 -C1 f -Br -I
+ 58 (toluene)20 + 36 (EtOAc)'*
[MID
+ 7 3 (ale)" 58 (ale)% + 67 ( a l ~ ) ~ ~ + 96%
-
33 (toluene)"
- 67 (alc)32
- 72 (ether)" - 13537
XXXV
XXXII B I
COiCH3 n
a b c
0 1
2
I
I
M1
[MID Predicted Planar Puckered +45' 1-138' 4-30' +go' $30' f90' XXXIII
Observed +11Q0(MeOH)2' +95"(MeOH)*7 +9g0ne H
CH,
I
I (1(.M.CO.M,
-Predicted [ M 1 ~ Planar Puckered -60" +30° XXXIV
Observed $35-78
(+)
($1
XXXVI
XXVII
Tetrahydropyran Derivatives.-The principles which permit the present major simplification and extension of Whiffen's treatment of conformational asymmetry5 appear to be contradicted a t three important points by Whiffen's work, These contradictions can be met only by showing that the present treatment allows a satisfactory calculation of the rotations of the pyranose sugars and their methyl glycosides by use of empirical rotation constants consistent (as some of Whiffen's are not) with the principles used here. It is shown below that these rotations can, indeed, be calculated using six such constants (rounded to the nearest Jive degrees) ; Whiffen used seven unrelated and unrounded constants in the same calculations. One of the present constants [k(C- H) (0- H) = +50°] corresponds in form t o Whiffen's constant H,5for which he gives a value of +34°42; as shown above and in part I, the present value 1s suitable for use with acyclic alcohols and is generally
the reduction of camphor with sodium gives predominantly borneol,31it is to be expected that the predominant products of reduction of the oxime32 (36) J. L. Kondakov and S. Saprikin, Bull. soc. chim. France, [iv] or thio d e r i ~ a t i v ewith ~ ~ active metals will be 37, 726 (1925). (37) Pariselle, C o m p f . rend., 180, 1832 (1925). bornyl derivatives; configurations assigned in this (38) J. Bertram and H. Walbaum. J. p r a k l . Chcm., liil 49, 1 (1894). way are in agreement with the present rule. Since (39) T. P. Nevell, E. de Salas and C. L. Wilson, J . Chcm. SoC.. (+)-bornyl acetate is formed by addition of acetic 1188 (1939). acid to (+)-pinene3* (XXXVI), the (+)-halides (40) R. H. Pickard and W. 0. Littlebury. ibid., 91, 1973 (1907). , ~ ~P. F. Frankland and F. Barrow, ibid., 9S, 2017 (1909). formed by addition of H X to ( + ) - ~ i n e n e ~ ~ (41) (31) E. Beckmann, J. prakt. Chem., [iil S I , 31 (1897). (32) M. 0. Forster, J . Chcm. Soc.. 73, 388 (1898); M. 0. Forster and J. Hart-Smith, i b i d . , 77, 1152 (1900). (33) H. Wuyts, Be?., 36, 863 (1903). (34) G. Bouchardat and J. Lafont, A n n . chim., [vi] 16, 236 (1889). (85) F. H. Thurber and R. C. Thielke, TEISJOURNAL., 63, 1030 (1931).
(42) Whiffen's value, together with the value for his constant F (+45O) [ k ( O H)% presents several contradictions with the present
-
work. If the numerical manipulations used here are allowed, these two values lead to: k(C H)S = +25'. which is badly out of line with the value of +BOe used in part I and suggests that carbon is less polarizable than oxygen. Acceptance of Whiffen's value, then, would require abandonment of the numerical manipulations or of the value
-
R(C - H)**
+60".
5492
JAMES
Vol. 81
H. BREWSTER
better than Whiffen's value when terpene and steroid alcohols are considered. The second of the present constants [K(O - H)? = + 4 5 O ] corresponds in form and magnitude to Whiffen's constant F and is, as Whiffen has ~ h o w n , ~suitable ~" for calculating the rotations of the cyclitols; as shown above, i t forms a consistent series with the values for k ( C - H)(O - H) and k(C - H)2usedinpartl. This constant will also be used to compute the rotation of asymmetric conformations about bond C, - Cz, where one of the hydroxy groups is glycosidic. Whiffen has used a constant G ( + 3 Z 0 ) for this purpose; acceptance of his value would require a repudiation of the central principle that the rotatory power of the asymmetric conformation XXXVIT1 is independent of the nature of I!.
*-
group is in a sterically asymmetric environment; the two "allowed" conformations of this group as found in D-hexoses are seen in XLII. The em-
4
+k(O -
H)2
r
=
- k ( C - H)(O - H )
$45"
+iV XLII
pirical value f25' is used here for this system; The positive sign Whiffen uses a value of +30'. of this value is consistent with the smaller size of the ether oxygen as compared to Cd and its substituents; the magnitude of this constant cannot be predicted. X final constant is required for calculations of the rotation of methyl glycosides. This constant reflects the fact that the xnethoxy group TABLE X
[MI = k ( X - HI(Z - H j XXXT'III
KWI'ATIOX C O K S T A N T S FOR C S E XVlTH C A R B O H Y D R A T E S
The tetrahydropyran ring, in contrast to cyclohexane, has an axis of polarizability difference (between 0 and (24). If the axial substituents at C1 and CSare different (as in XXXIX), they will form an axis of polarizability difference which will form a screw pattern of polarizability with the ring axis. This pattern, as seen in projection in XLI, has the character of a left-handed screw and should (part I) be dextrorotatory. Since the conformational asymmetry effect of the hydroxy group is --kH(O - H), weniay expect X X X I X to be strongly dextrorotatory ; the empirical value used here is + l O O o , to be compared with LVhiffen's An axial hydroxy group a t value J = +113'. C2 or C4as shown in XL produces the same permolecular pattern XLT, here opposed by the weak net levorotatory conformational effect: - k ( C - Oj (0 - H) = GO. -5'. The total rotatory effect of XIJ is estimated here to be + G O o ; Whiffen's constant I (+-13") fills the corresponding role in his calculation^.^^ =ln additional constant is required for the hexoses since the hydroxy methyl
+
Pl'otation Present
k(O - H)(O - H j
.
Whiffen6
---
Whiffens
+45" +32"
4\ G
A[ 3 1 1 - 7
1 (,
+loo"
XL
+
is in a sterically asymmetric eiiviroiinient. 111 the $-D-hexosides (XLIII), for example, the asynimetric conformation XLV should predominate, this conformation should be levorotatory but its inagnitude of rotation cannot be predicted. -1 value of - 1 0 3 O is used here (Vi-hiffen uses - 100°); a value of +loa" is, then, required for a-D-hexosides (XLJV).
+eoc
..
-
...
__
XLI, ( + J
143) Acceptance 01 Whiffeo's i ) i l e v p r e l d i o i i of t h i s ri,iistant HS being d u e wholly t o conformational asymmetry effects \vould be altogether incclnsivtent irith t h e preqent treatment.
i45"
+
Y ~
Present
k(C - H ) ( O - H ) H t34" +50" XXXIX" J +113" 100" XL" I +43" f6O" XLII 180" +2,7" XLV" - 100" -1 0 5 O a These values can be, roughly, calculated bj- letting the permolecular effect be f65" and by assigning the value: k( E - H)(C - H ) = -Zoo, where E represents an uriihared pair of electrons on oxygen. On this basis X X X I X heconies: 65 k ( C - E)(O - H ) - k ( O - H ) ( C - HI = +107; XL becomes: 65 - k ( C - O ) ( O - H ) = 60 (uiing t h e value k(O - H,," = 45) and XLV becornea: - k ( C E ) ( O - H ) = -92 . The empirical values give hetter fit5 and are used in Table XI.
>H
SXXIS,
-50'
1-1
S L Y , --kC(O - € = I) --1rXo
OPTICAL ACTIVITY
Oct. 20, 1959 TABLEXI MOLECULAR K O I‘A PIUNS
OF
A N D METHYL
HYDROXVTETRAHYDROPYRAXS GLYCOSIDES [&f]D O f
-[MID-Calcd.
Hydroxytetrahydropyran
Obsd.‘
6-n-Ribose a-D-XybSe @-D-Xylose a-D-.kabinose ,3-D-.%rabinose d-D-Lyxose l,j-Anh?.dro-D-arabitol L’-Deox?.-B-D-x? lose
- 45 - 35 +145 +141 - 45 -105 -116 -295 -286 -10b -109 - 150 - 132 - 45 - 34
a-D-Glucose p’-D-Glucose or-n-Mannose $-D-Mannose a-D-Galactose J-n-Galactose a-D-Gulose fi-D-Gulose d-D-XllOSe a-D-Talose d-D-Talose
+ 30 + 34
1,5-Anhydro-usorbitol galactitol mannitol talitol ’-Deoxya-D-glucose B-D-glucose a-D-galactose p-D-galactose
i 220
+202
+ 25
i 53 - 30 - 31 +270 +272 80 95 115 135 - 55 0 30 122 75 20 24
+
+
+ +
+
+
+
+
+
+ i5
4- 71)
4- 125
+12fi
- 75 - 25
- 19
-
methyl glycoside Calcd Obsd
3-250 -150 0 -400
+253 -108 - 28 -403
+325 +309 - 75 - 66 130 154 - 135 - 135 380 $375 - 25 0
+
+ +
- 160
- 16.2
83
+ 130 + 30 + 180
+ 80 + 6 i
[CONTRIBUTION FROM
+%3
