Studies on the Schardinger Dextrins. X. The Interaction of

X. The Interaction of Cyclohexaamylose, Iodine and Iodide. Part I. Spectrophotometric Studies1. B Y JOHN A. THO MA^ AND DEXTER FRENCH3. RECEIVED...
0 downloads 0 Views 578KB Size
6142

JOHN

-4. THOMA AND DEXTERF R E N C r r

JTOl. SO

the crystallizations. Whereas during earlier crystallizaIsomer C: A solution with c 1.0125 showed a n [CY]% tions the product melted a t 140-142', the purest isomer C +46.4' 15 minutes after dissolution of the compound. The (fine needles, different in appearance from those of isomer rotation values observed 4 hours thereafter and after a total A) from the final crvstallization melted sharDlv at 135-136". period of ca. 4.25 days were [ c Y ] ~ ~$54.3 D and +59.8", reAnal. Calcd. for C~tHieOa(isomer C): C,h?.83; H, 5.80. spectively. An infrared spectrum of the product (from the Found: C, 47.84; H, 5.81. solution kept 4.25 days), crystallized from ether-peiitane, T h e mother liquors from the original crystallization gave, appeared t o be identical with t h a t from isomer B. on concentration in vacuo, a second crop of crystals (0.58 g., Quantitative Comparison of the Infrared Absorption in 111.p. 112-115'). Repeated recrystallization of these from the Carbonyl Region by Isomers A, B, C and Tetra-O-acetyl/3-D-xylopyranose.-Freshly dried chloroform was used for ether-pentane ( 2 : l ) gave a minute amount of needles of isomer A (infrared spectral identity established). making the solutions. An NaC1 cell of thickness 1 = 0.0502 Optical Rotation Behavior of Isomers A, B and C in cm. mas used. The spectrophotometer slit width wils coiiSolutions.-A.C.S. reagent grade chloroform or pyridine stant a t 46 p during the measurenients. The results of tlic was used. Rotations were made in a half-decimeter po- measurements a t the different compnrnble coiice~itraticins All solutions were left standing at room are siim~nedt i p i n Table 11. larimeter tube. temperature for the periods stated. TADLE II Chloroform Solutions.-Isomer A: For a solution with c Apparent 1.015, a n initial value of [ a I z 5$108.4" ~ had not changed integrated in 46.5 hours. A solution with c 1.06 gave, after standing absorption I k L intensity ca. 14 days, a final [ a ] 2 +115.4'. 5~ I n each case, the prodx cl n ( x 10-41, uct, after expelling the CHCla in vacuo and crystallizing the unleX, Concn., c, ( = loglo AY.?)~ mole-.' residue from ether-pentane, gave a n infrared spectrum Compound cm. - I mole/liter ( T O / T ) ~cm. ~ ~- 1 ) 1. cm. - 2 identical with t h a t of unchanged isomer A . 12.54 0.710 4 6 . 0 0.018170 Isomer B: -4 solution with c 1.05 showed a n initial [ c Y ] ~ ~ D Isomer A 1745 ,009085 40.0 11.30 .3i0 $46.1" and after 28.5 hours a final [ C Y ] ~ ~+30.5'. D The . IS8 4 0 . 0 11.04 ,004543 infrared spectrum of the product, crystallized from ether-pentane, showed significant differences from t h a t of isomer ,018840 12.27 ,790 4 2 . 0 B, but did not reveal conclusive evidence of t h e appearance 11.58 Isomer B 1747 ,009421 ,423 3 7 . 0 of absorption bands characteristic of isomer C . A solution 10.57 ,004711 ,210 3 4 . 0 D after C Q . 14 days with c 1.00 showed a final [ c Y ] ~ ~$38.6' and a n infrared spectrum of the product was identical with 12.19 .018190 ,795 40.0 t h a t from the product of a similar mutarotation study of 11.30 ,009095 Isomer C 1745 ,421 35.0 isomer C. 9.60 ,202 3 1 . 0 ,004548 Isomer C: A solution with c 1.00 showed a n initial [ c Y ] ~ ~ D -20.0" which changed in ca. 22 hours to a value of [ c Y ~ Z ~ D 900 46 5 18 50 015720 +6.6". An infrared spectrum of the product, crystallized 456 40 0 16 18 Tetraacetyl- 1752 007860 from ether-pentane, showed enhancement of intensity of the x) lose 003930 220 36 0 11 03 peaks at 760 and 3328 cm.-' characteristic of isomer B. Ll solution with c 1.005 showed a final [ c Y ] ~ $32.8' ~D after Acknowledgments.-The author wishes to thank ca. 14 days and a n infrared spectrum of the product was Dr. R. N . Jones for advice received in making the identical with t h a t from the product of a similar mutarotation study of isomer B. quantitative infrared studies and for making avail5 : 1 Pyridine-Water Solutions.-Isomer A: X solution able information from the manuscript of referencc with c 1.04 showed a n initial [ a ] z 5 ~ $115.4' and after ca. 10 while still in press. Deep gratitude is expressed 3 days a n [ a ] * 5 ~ +55.8', which did not change on further standing. An infrared spectrum of the product, crystal- to Dr. R. W. Watson for the keen interest shown in lized from ether-pentane, was identical with t h a t of isomer this work and to Dr. C. T. Bishop for useful sugB. gestions and criticism received. The technical asIsomer B: A solution with c 1.0125 showed a n initial sistance of Mr. F. Rollin in making the infrared [cY]'.~D +50.4" and after ca. 4.25 days a n [ a I z 64-51.4'. ~ spectrograms is gratefully acknowledged. An infrared spectrum of the product, crystallized from etherOTTAWA 2, CASADA pentane, showed unchanged isomer B.

[COSTRIBUTION FROM

THE

DEPARTMEXT OF CHEMISTRY, IOWAS n m COLLEGE ]

Studies on the Schardinger Dextrins. X. The Interaction of Cyclohexaamylose, Iodine and Iodide. Part I. Spectrophotometric Studies1 B Y JOHN

A. THO MA^

AND

DEXTERF R E N C H 3

RECEIVED J U N E 23, 1958 The cyclohexaamylose, 12, I - system was studied a s a model for the starch, Iz, I- system. Using the method of continuous variation it was found t h a t u-dextrin in aqueous Ipforms a n aIzcomplex in the absence of I - and a n aIa-complex in the presence of I-. The method of continuous variation has been extended to some special ternary systems.

The unique molecular structure of a (cyclic moleIntroduction The Schardinger dextrins, and particularly cyclo- cule consisting of six 0-1 + 4-linked D-glucopyrahexaamylose (referred to hereafter as a ) , have been nose units) makes i t possible to form inclusion comof very considerable interest because of their abil- pound in solution as well as in the solid state. Thus ity to form highly colored, crystalline iodine com- systems containing a and complexing agents are plexes resembling the starch-iodine ~ o m p l e x . ~particularly favorable for study as models of the far more complicated starch systems. (1) T h i s work was supported in part b y grants from Union Carbide a n d Carbon Corporation and General Foods Corporation. Although the understanding of the starch-iodine (2) Union Carbide and Carbon Fellow 1957-1958. reaction has advanced substantially in recent years, (3) To whom inquiries concerning this article should be made. several aspects remain controversial. Some con(-1) D. French, Adg. Carbohydrate C h e m . , l a , 189 (195.7).

Nov. 20, 1958

CYCLOHEXAAMYLOSE WITH IODINE AND IODIDE

temporary workers5-’ have reported that when Iis added to starch-iodine solutions, i t plays an integral part in the reaction but i t remains uncertain whether or not starch can form an iodide-free iodine complex.8 It was thought that this question could be resolved readily by examining the absorption spectra of starch and dextrin solutions in aqueous IZ in the presence and absence of I- because complexes of 1 2 and those involving both 1 2 and I- should show characteristic spectra. To repress any Iwhich might arise from the hydrolysis of Iz,either HI03 or strong acids can be added. Equations (1) and (2) represent the reactions involved 12 103-

+ H20 = H f + I - + H I 0 + 51- + 6 H + 3Hz0 + 312

6143

mine c/b experimentally, the absorbancy (or some other property) of the ternary complex is measured in a set of solutions containing an amount x of B and an amount Co - x of C. Each solution also contains a fixed amount of A in large excess over that required for complex formation. When the absorbancy due to the ternary complex is plotted against x, the value of c/b is given by Xmax/(CO Xmax). After the ternary system has reached equilibrium the equations may be written A = A0 - complex = A O (5 ) B=r-P-q c = co - R” - cq K’ = A B / p K ” = AaBbCc/q

( 1)

(2)

Rundle and Frenchg have demonstrated the formation of a starch-iodine complex by absorption of I p vapors by helical amylose in the solid state. Since no iodide was added, i t was inferred that I- was not necessary for the reaction. However, even with “dry” amylose, i t now appears likely that Iwas produced either by the reduction of I p by impurities or by solvolysis of IZ by occluded HzO or alcohol, or the OH groups of the carbohydrate. Our preliminary investigations of the absorption spectra of a-12-1- solutions and the formation of a12 and a K I crystals4 suggested that the dextrin, IZ and I- reactions involve three binary systems (012, CUIand 13-l’) and one ternary system (aIa-). The extent of dissociation of the a-12-1- complex appeared to be very small, indicating that the determination of the stoichiometry of the complex would be amenable to the method of continuous variation.

(6)

(7) (8)

(9)

where A , B, C, p and p represent, respectively, the equilibrium concentrations of A, B, C, binary complex and ternary complex, and K’ and K ”are the dissociation constants for the binary and ternary complexes. Substituting 5 , 6, 7 and 8 into 9, differentiating with respect to x, and setting dg/dx = 0, the condition for a maximum, one obtains after simplification c/b = X m a x / ( Co - X,,,). Similarly, when A and C are continuously varied in the presence of a large excess of B, the combining ratio of A and C in the complex may be evaluated. However, when A and B are continuously varied in a large excess of C their apparent ratio will vary between a / b and 1 depending upon the magnitudes of the dissociation constants of the binary and ternary complex and the amount of C in the system. Only if a/b = 1 or the amount of binary complex formed is insignificant may the ratios of any two Theory Applications of the method of continuous vari- reactants in the ternary complex be found by conation and the implications and validity of the as- tinuously varying them in the presence of a large sumptions made in the theory of this method have excess of the third component. Similarly, if K” >> been critically reviewed by Wo1dbye.l’ Vosburghl2 K‘ and the ternary reaction is the preponderating and Katzin13 have extended the method to binary equilibrium, the combining ratio of any two comsystems in which successive dissociation constants ponents in the complex may be found by continfor the complex differ by several orders of magni- uously varying these components in the presence of tude. I n this paper, we extend the method of an added constant amount of the third (not necescontinuous variation to some special ternary systems sarily a large excess). Even though equilibrium constants may be unwhich may occur when a or starch and IZ and Ireact. The most general case which is treated in- favorable, i t is sometimes possible to choose convolves the competition of one binary system with a ditions under which the method of continuous variation is applicable to ternary systems. The ternary system and is represented a-12-1- system will serve as an example, and may A + B = A B (3) be represented by the series of equations U A 4- bB

+ CC

rl,BbC,

(4)

Iz f 1-

where a / b and c/b are, respectively, the ratios of A:B and C:B in the ternary complex. To deter( 5 ) G. A. Gilbert, J. V. R. Marriot, Trans. Favaday SOL.,44, 84 (1948). (6) R. J. Higginbotham, J . Textile Inst.. 40, t783 (1949). (7) D.L.Mould, Biochem. J.,58, 593 (1954). (8) E. 0.Forster, Ph.D. Thesis, Columbia University, 1951, reported that a and amylose could form 1s complexes in the absence of I-. These tests were conducted at $H 6 in 0.2 M KIOa, conditions which do not effectively prevent the formation of I- b y the hydrolysis of In. The appearance of Forster’s absorption spectra indicates that his complexes contained I (9) R. E. Rundle and D. French, THISJOURNAL, 66, 1696 (1943). (10) G.Jones and B. B. Kaplan, i b i d . , 60, 1845 (1928). (11) F. Woldhye. Acta Chem. Scand., 9, 299 (1955). (12) W. C. Vosburgh and G. R . Cooper, THISJOURNAL, 63, 437 (1941). (13) L. I Katzin and E . J . Gebert, ibid , 78, 5456 (1950).

-.

I3-i KI = [ I B ] [ I - ] / [ I J - ]

(10)

+ = K I [cy][Iz]./lcyI21 (11) 0112 + I - = 0 1 1 ~Kz ; = [cyIz][I-]/[cyIa-] (12) + I - 0 1 1 ~K; O= [cy][I-]/[011-1 (13) CUI-+ I? Ka = [cyI-][Iz]/[aI3-] (14) + Iz + I- = aIa-; Ks = [ C Y ] [ I ~ ] [ I - I / [ O(15) ~I~-I 12

0112;

(Y

=

cyIO-;

01

let concentration cy added = [cyt] and concentration 1 2 added = [12t] and concentration I- added = [C, - IS$]

then a t equilibrium if conditions are chosen such that [a]remains essentially constant as [I4 and [I-] are varied (a condition which can be achieved by adding a large excess of a or by choosing conditions such that the degree of formation of the complex is small) then a t equilibrium these equa-

6144

JOHN

A. THOMA AND DEXTER FRENCII

tioiis may be written where brackets represent concentrations [I21 = [ I ~ t l- [Is-] - [a121 - [NII-I

(10)

and

[[-I

= [C”

-

- 113-1

-

[CYr-l

-

Ia13-I

(17)

and [a1 = [ a t 1 (18) Now substituting 16, 17 and 19 into 11 and 13 wc obtain

[CY111 IJCl

+ ["til = [NIPl(kl) =

latl([ I s ] - [ I -1 -

and IaI-IlK

+

= iatl i

ICYtll

[NI-I(ki)

ic, - I,,]

-1,

(19)

[CYIs-II

(20)

jnI

=

- [I,

I -

but from 10 and 13 113-1 = K6[aI3-l’Kl[LY1 = ko[a13-I

(21)

Then substituting 21 into 19 and 20 we obtain [NTZI = ([ay,l/K,)I [ I % ] - Im13-1(1 K , ) ) (22) and [aI-l ( [ a t ] / & ){ [co - Izt1 - [a13-1(1 4- ha) 1 (23) Now substituting 16, 17 and 18 into 13 and substituting 21, 22 and 23 into the resulting equation we obtain after simplification and combining constants

+

(Kb)

Ik’[Co [.I?-] = { k ’ [ I ~ t ]- k”[~yIj-]] Iat] - .k”[0113-]] (24)

then differentiating with respect to I2t and setting d[aIa-]/dlzt = 0, the condition for a maximum, obtain after simplification: I 2 t max = G/2. 1

x. - p I - - t -

1701.

so

The a-12-l- system has been shown to be in rapid reversible equilibrium by potentiometric methods; these results will be reported in another paper. Experimental Materials.-The preparation of pure has been described elsewhere.4 All other reagents were of the purest grade commercially available and used without further purification. The absence of the blue starch-iodine complex when a saturated solution of Ipin 0.2 M HIO, was added to ail equal volume of 0.5% soluble starch ind~catedt h a t the I concentration was insignificant.14 The purity of stock a was determined hy measuring its optical rotation . 4 Apparatus.-For the spectral studies, a Beckman model DU spectrophotometer was used with 1.OO-cm. silica cells. Optical rotation was determined in a Rudolph precision polarimeter. Procedure.--A nearly saturated solution of stock IZ was made by shaking or stirring 1 2 crystals in water for two days at room temperature. T h e concentration of this solution was determined immediately before use by measuring its optical density’s when diluted 1 t o 5 a t 460 mp. Standard solutions of 1 2 mere then made b y appropriate dilution of the stock solution. Standard CY and KI solutions were made b y appropriate dilution of exactly weighed amounts of these reagents. During the course of experiments, it was frequently necessary to prepare standard IP solutions because exposure to air resulted in volatilization. All optical density measurements were made at room temperature in capped silica cells. The procedure employed for continuous variation studies has been described e l s e ~ l i e r e . ’ ~ ~ ~ ~

Results and Discussions Binary Systems.-Curves B and C in Fig. 1 depict, respectively, the absorption spectra of IZ in 0.08 M HI03 in the presence and absence of excess CY (8s. distilled water blank). The shift of the 1 2 maximum upoii addition of CY suggests the formation of an a-IZ complex. When the method of continuous variatiotl was applied tofthe CY-IZ system and the data plotted in Fig. 2, i t appears that I2 and a react in a 1: 1 ratio. In this study,

8 4

%

I

I

I_

300

400 500 Wave length, nip. Fig. 1.-Ultraviolet and visible absorption of a-12 and IZ systems in the presence of IIIO? and the CY-1. system in the ,21 12, 4.8 X Jf C Y ; absence of HIOs: A, 2 6 X 13, 2 G x 10-4 31 L, 4 s x 10-3 ,id 01, 0 08 N H I O ~ ; c, 2 G X lo-’ If IZ in 0.08 J f HIOd. Distilled water bras used for the blank.

By similar considerations, it can be shown that if Iz and CY are continuously varied in the presence of a constant amount of I- the method of continuous variation would show a maximum where the ratio of 1 2 to a equals one.

0.0

0.5

+

1.0

Ratiu 1 2 / 1 2 01. Fig. 2.-Continuous variation plot for the 01-12 system in 0.2 M HI03: A, observed optical density at 420 nip as 1 2 was varied from 0 to 4.57 X lov4 31; D, five times difference between the optical density of A and same solutioii minus a ; 0.2 M , HIOI served as a blank. (14) Under the conditions which Forster used (see ref. S), a deep blue complex was formed which was only slowly destroyed upon strong acidification. (1.5) A . D. Awlrey OIKI I< I, 2 5 X 10-5 11 I-, 1 2 X AI CY. Distilled water was used for a blank. 250

the spectrum of the a-12-1- solution resulted from free 1 3 - was dismissed because a t this dilution, without added a , the calculated14 optical density a t 285 mp would be only 0.006, compared to the observed optical density of 0.410. Urhen It and a were continuously varied in the presence of I- and the optical densities plotted against the ratio Iz/(Iz+a) (Pig. 4)) i t was apparent that I1 and a react in the ternary complex in a 1:1 ratio. Similarly when the optical densities of solutions continuously varied in I 2 and I- in constant a were plotted against the ratio 1-/12 I- (Fig. 5 ) , i t was apparent that 1 2 and I- reacted in a 1:1 ratio in the complex. These data supported the inference arrived a t from the shape of the absorption spectrum that the complex had the formula aIz-. Perchloric acid ( 0 3 M) was

+

0.5

0.0 Ratio I-/I2

1.0

+ I-.

Fig. 5.-Continuous variation plot of the a-12-1- system Ai' CY and 0 8 M HClOa: A, observed optical in 1.3 X density as I - was varied from 0 to 1.0 X lo-* 111a t 288 mp; E, 350 mp; C , 440 mp. 0.8 M HC104 served as a blank.

added to all solutions in this group to suppress the hydrolysis of IZaccording to equation 1. 111 When (Y was varied between lom3and and Iz and I- were varied over as large a range as was suitable for spectrophotometric measurement, all the maxima (other than that of I?)in the region of 350 to 650 mp could be attributed to I d - , a b or aI2. Because it has been shown that a! can form an IZ complex in the absence of I- and because a serves as a simple model for the amylose helix, i t will be of interest to examine the spectra of amylose-I2 solutions in the absence of I- to see if amylose and I2complex. From the results of this study, i t may be inferred that the absorption spectra of starch-iodine coni-

plexes will vary greatly depending on the ratio of iodine to iodide in the complex; moreover, it may be possible to determiiie this ratio quitc unam-

-__-

Vol. so

COMMUNICATIONS To THE EDITOR

6146

biguously by the method of continuous variations, if a system can be found in which equilihriuin exists. \XES, I O W Z

--____

___

-~

-

~

-

COMMUNICATIONS T O 1 H E EDITOR THE STRUCTURE OF DIOSCORINE

Sir : Dioscorine, C13H1902N, a water-soluble alkaloid, has been isolated from Dioscorea hirsuta Blume1s2 and Dioscorea hispida Dennst.".' The early structural investigations2 were continued by Pinder3-9 who has proposed three structures, including I, for dioscorine. No facts have been presented from which the structure of the nitrogenous nucleus and the location of the lactone ring present in the alkaloid could be deduced conclusively. In this communication we wish to present findings which establish structure I for dioscorine. Reduction of the alkaloid with lithium aluminum hydride gave a diol (11),1n.p. 121' (picrate, m.p. 164-165', lit.* 159-160'), which by ozonization was converted to glycolaldehyde and a hydroxyketone (111), m.p. 2 5 O , infrared peaks a t 3540 cm.-' (OH); 1705 cm:-l (C=O); 1405 cm.-' (COCH2) and 1360 cm.-l (COCH3) (picrate, m.p. 120-122'). On exposure to 0.1 N sodium hydroxide, 111 underwent retroaldol cleavage to acetone and a ketone (IV), infrared peak a t 1723 em.-' ( c s 2 ) , 1730 em.-' (Cc14); [o(Iz5D + l i o (c 1.43 in H2O) (di-p-tOlUOyl-D-tartrate, m.p. 149-151 ') , Treatment of IV with ethanedithiol and then desulfurization produced tropane (V), picrate, imp. 250-28s' (dec.) ; chloroplatinate, m.p. 2OS-21O0, identical with authentic samples.lo The ketone IV is therefore tropan-2-one or tropan-6-one. We initially preferred" the 6-substituted structure, mainly because of the position of the carbonyl band in the infrared of IV. Reduction with sodium borohydride or with hydrogen over Adanis catalyst yielded an alcohol (VI), m p . TOo (picrate, m.p. 262-263'; di-p-tOlUOyl-D-tartrate, m.p. 1G5-1GG0; di-p-toluoyl-L-tartrate, 1n.p. 1 6 4 7 0.5') which had an infrared spectrum entirely different from that of il mixture of Ga- and 6p-hydroxytropane, m.p. 65-81' (cf. ref. 9) prepared by desulfurization of GP-hydroxytropan-3-one thioketal. The three derivatives of the alcohol from dioscorine (I), however, were identical with those of 2a-hydroxy(1) H. W.Schutte, Chcm. Zeiitr., 68, 11, 130 (1897). (2) M . K. Gorter, Rec. ~ Y U P chim., . 30, 1 6 1 (1911). ( 3 ) A. R. Pinder, Nature, 168, 1090 (1931). (1) A . R. Pinder, J . Chem. Soc.. 2236 (1952i. (5) A. R. Finder, i b i d . , 1825 (1953). (6) A. R. Pinder. %bid., 1577 (1956). (7) A . K. Pinder, Chemistiy und I n i l t r s l v y , 1240 (1957). (8) A. R. Pinder, Telvnhedvon, 1, 301 (1957). ('2) J. B. Jones and A. R. Finder, Chemistry and I n d t ~ s t ~ y1000 , (1958). (10) K , Hess, B e y . , 51, 1001 (101s); R. Willstatter and 1'. Iglaiier,

ibid., 33, 1170 (1900). (11) G. Diichi, D. E . Ayer and I?. M. White, XVIth I n t w n . Cvngr. of Pure and Applied Chemistry, Paris, July, 1957.

tropane (VI). I 2 Similarly the di-p-toluoyl-D-tartrates of IV from cocaine and dioscorine were identical. The nuclear magnetic resonance spectrum of dioscorine (in CHC1,) has no peaks above 1200 C.P.S. l 3 and unactivated methyls are therefore absent. The configuration a t CZ then was determined : The saturated diol prepared by catalytic reduction of I1 had a pK, of 8.81 (Cellosolve) like 2a-hydroxytropane (VI), pK, 5.42, while @-hydroxytropane (1711) had pK, 9.62. The observed differences in basicity are similar to those of quinine and epiquinineL4and demonstrate that the tertiary hydroxyl in I1 has the n-configuration which is i n .\

1

[I

Ill T

\'Ill

agreement with the previously reported infrared evidence.* 2P-Hydroxytropane (VII), m.p. IOo, Y L ~ O D 1.4562 (picrate, n1.p. 263-265') was prepared by lithium aluminum hydride reduction of 2P,RP-epoxytropane (IX) (picrate, 1n.p. 248-254') which in turn was available by oxidation of tropidine (YIIP with trifluoroperacetic acid in il~etonitri1e.l~The conversion of both cocaine'? and dioscorine to the same tropan-2-one (XI-) demonstrates identical absolute configuration16 in the two alkaloids. We are indebted to Eli Lilly and Co. for financial support and to Prof. E. Schlittler for plant material. I)EPARTMENT OF CHEXISTRY 1) E A \ E R

--

TECHIYOLOGS c BXITI P. REYNOLDS-WARNIIOFF DIVAIN M LVHITE RECEIVED OCTOBER 6, 1958

?VIASSACHUSETTS 1NSTITUTE O F C A M B R I D G E , ?if~4SSACIICSETTS

(12) 31. R. Bell and d. Archer, accompanying paiier. TVe w i s l i 10 thank n r . S. Archer tor a sumple of this substance. (18) Relatixe to an arbitrarily assigned peak of 1000 c i).s for t h e aromatic proton of toluene. (14) V. Prelog and 0. IIYtiiger, Hf'la. C h i m . A r l t z , 33, '7021 [1931). (15) Method of G. Fodor, J T o t h , I. Koczor, P. Doho and I. Vincie, f h P m i s t y y and I n d u s i v y , 764 (1950). (le) E. Hardegger a n d H. Ott, Hela. Chim. Acta, 38, 312 (195;).