3534
CARL
[CONTRIBUTION
Terpenoids.
XXIII.
F R O M THE
DJERASSI.4ND R.HODGES DEPARTMENT OF
Vol. 78 UNIVERSITY]
CHEMISTRY O F W A Y N E
Interconversion of Thurberogenin and Betulinic Acid’ BY CARLDJERASSI AND R. HODGES RECEIVED FEBRUARY 13, 1956
Reduction of the 30-nor-20-ketone V of thurberogenin ( I ) with calcium in liquid ammonia leads in poor yield t o the corresponding derivative ( V I ) of betulinic acid. Coupled with earlier evidence, this interconversion establishes the structure and stereochemistry of the cactus triterpenes thurberogenin ( I ) and stellatogenin (11). Some miscellaneous reactions involving the side chain and ring E of thurberogenin are also recorded.
Thurberogenin (I) and especially stellatogenin (11) are widely distributed among various cactus species and they represent triterpenes of unusual interest. Together with dumortierigenin6 they are the only naturally occurring triterpenoid lactones,6 and furthermore, thurberogenin and stellatogenin belong to the rare class of lupeol triterpenes of which only three members (lupeol, betulin and betulinic acid) were known7 prior to our investigations of cactus triterpenes. The structure assignments of these two substances, which had not been converted to a triterpene of known structure, rested on circumstantial evidence of the following type : The presence in thurberogenin (I) of the 3phydroxy-4,4-dimethyl moiety in ring A, of an isopropenyl side chain and of a five-membered lactone ring involving a tertiary hydroxyl group was establisheds rigorously. Since stellatogenin (11) can be dehydrated? to thurberogenin (I) without rearrangement and possesses an additional tertiary hydroxyl group, the relationship between the two triterpenes implicit in structures I and I1 is unequivocal. The co-occ~rrence~ of stellatogenin with oleanolic acid and especially with betulinic acid (111) and oxyallobetulin (IV)g suggests on biogenetic grounds that the carbonyl group of the lactone ring originates a t C-17. Assuming a normal triterpene skeleton, thurberogenin and stellatogenin must then be represented by structures I and I1 or the corresponding isomers with the lactone ring terminating a t C-13. The reasons for
our preference for I (and 11) are given in detail in our earlier paper8 and are based to a large extent upon the base-catalyzed E-homo rearrangement of the 30-nor-20-ketone V derived from thurberogenin (I), a reaction which would not have occurred if the lactone terminated a t C-13. ~
H l\l
0 1 ‘
1\10
/
AI/iV
AA/ / 1 I i‘c=o
pc=o
rrw H
A A’/3./
HO/
&v
1
R + . p
I1
1
~ OH
Iva, R =(
.H oxc
I
1
c , R = O
\/ 20; 19 7
.__
At/3../
1
,:HNd
H O i \ 4A/ :HV ’
I11
3.
J.
CHI
CH3
I c=o .~
I c=o . --
\
AA/
H~
‘0
/
AI/iV
1
\c=o
/+A/:\’ A C O / P
v
~‘COIH
i\/h /:v H HO/((&”
I
I
H
‘0
AI/W / 113H l\c=o
,A
=/ . .H
b,R
i s H 1
(1) Paper X X I I , C. Djerassi, A . Bowers, S . Burstein, H . Estrada, J , Grossman, J . Herran, A . J . Lemin. 4.Manjarrez and S . C Pakrashi, THISJ O U R N A L , 76, i312 (1956). (2) This investigation was supported by a research grant (No. G-3863) f r o m t h e Division of Research Grants of t h e National Institutes of Health, U. S. Public Health Service. (3) C. Djerassi, I,. E. Geller and 4.J. Lemin, THISJ O U R N A L , 7 6 , 2254 (1953). (4) C . Djerassi, L H. 1,iu. E F a r k a s , A . E. Lippman, A . J. Lemin, L. E. Geller, R. N. McDonald and B. J. Taylor, i b i d . , 77, 1200 (1955). ( 5 ) C. Djerassi, E. Farkas, A . J. Lemin, J . C. Collins and F . Walls. ibid., 78, 2969 (1964). (6) T h e crude glycosides of Lemaireoceucus dumovlieri (reference 5 ) and L . steiialus (reference 4) show a n infrared band (Xujol mull) a t 5.65 p attributable t o a five-membered lactone. Coupled with t h e evidence presented earlier (footnote 7 in reference 8) this would indicate t h a t t h e lactone ring was present in t h e plant material rather t h a n produced during t h e acid hydrolysis of t h e glycoside. (7) For re\.ieu-s see 0. Jeger in L . Zechmeister’s “Progress in t h e Chemistry of Organic Natural Products,” X’ol. V I I , Springer, Vienna, p 1, and D. H . R . Barton in E . H . Rotid’s “Chemistry of Carbon Compounds,” X’ol. I I B , Elsevier Press, Houston, Texas, 1953, p. 726. ( 8 ) C . Dierassi. E. Farkas, L . H. Liu and G. H. Thomas, THIS J O U R N A L , 77, ,5330 (1055). (0) T h e isolation o f oxyallobetulin is described in t h e Experimental purtion I t i i ni,t certxin whether t h i i substance ic present in t h e c u c t u ~or fi,rmed a i ;in ;irtif:rct frfim bctulinic ncid (luring t h e acid hydroly+* < > f1 he slycc,+i- a t the stellatogenin acetate, thurberogenin acetate and nor-ketone stages) in 125 cc. of dry toluene was added to 2 g. of calcium metal dissolved in 500 cc. of liquid ammonia (distilled from sodium). The mixture was stirred for 5 h r . a t Dry Ice temperature while adding 0.5-g. portions of calcium every 30 minutes. At the end of the reaction, 25 g. of ammonium (17) K. S. Leeds, D. G . Puknshima a n d T . F. Gallagher, THIS 7 6 , 2 2 6 5 (1954) (18) J. Schmidt-ThomC, .47igew. C h e m , 67, 71.5 (195.5); B e y . , 88, 3895 (1956). (19) I t should be remembered (reference 8) t h a t thurberogenin itself is not affected by alkali except for opening of t h e lactone ring and i h a t acidification results in immediate relactonization. 120) While t h e course of this reaction, especially with t h e unsaturated aldehyde, is obscure, i t nevertheless must be associated in some manner with t h e 19-hydroxyl substituent. As demonstrated in t h e Experimental portion of this paper, t h e unsaturated aldehyde derived from betulin diacetate is not affected by base except for loss of t h e acetoxyl groups. (21) Melting points were determined on t h e Gofler block. Cnless noted otherwise, rotations were measured in chloroform solution. All spectroscopic measurements were carried out by blrs. Dolores Phillips. T h e microanalyses were performed b y Spang Alicroanalytical T,nl,oratory, Plymouth, hlichignn. -JOURXAL,
Vol. '75
chloride was added, the ammonia was allowed to evaporate, the residue was diluted with water containing 20 cc. of acetic acid and extracted thoroughly with chloroform. The chloroform was extracted with cold 5y0 aqueous potassium hydroxide (in which lactones of the thurberogenin type are insoluble), the extracts were acidified with acetic acid and the product again removed with chloroform. This procedure was repeated in order to ensure that only acidic material was removed and yielded 96 mg. of crude acid VIa. Brief methylation with ethereal diazomethane followed by acetylation with acetic anhydride-pyridine ant1 Chromatography on 20 g. of alumina (deactivated with 3:; (if 10% acetic acid) yielded from the benzene eluates aftcr crystallization from methanol-chloroform 21 mg. Tf methyl 3acetoxy-3O-nor-2O-ketobetulinate (VIb), m.p. 21 5-218", [ a ] ~ -22' (c 0.24), rotatory dispersion12 (Fig. 1) in dioxane: [alioo -17", [ a l a a s -23', [a1276 -623", "nlax." [ a ] a o i . i +324". Identity with an authentic specimen (see belom j was confirmed by mixture melting point determination and identity of the infrared spectra. Anal. Calcd. for Cr4Hj005:C, 74.67; H, 9.79. Found: C, 71.82; H, 10.07. The original chloroform solution (after alkaline extraction) was evaporated to dryness, and the residue was refluxed for 3 hr. with 2.5yGmethanolic potassium hydroxide in order to convert any unreacted lactone V to the acidic E-homo rearrangement product.* The mixture was poured into water and extracted with Chloroform tu furnish 805 mg. of neutral material. Acetylation with acetic anhydride followed bl- chromatography over deactivated alumina and elution with benzene-ether ( 4 : 1) yielded a crystalline residue which was recrl-stallized several times frum methanol-chloroform; 1-ield, 105 mg., rn.p. 280-283' (with sublimation), [ a ] +26.6". ~ This product is apparently the triacetate VIIb of the tetrol VIIa and is probably epimeric a t C-20 with the corresponding derivative (n1.p. 220resulting from the lithium aluniinum hydride reduction of V. -4naI. Calcd. for CJ,Hi6O7:C, 71.39; H , 9.59. Found: C, 70.95; H. 9.34. .i 45 tng. sample of the triacetate VIIh was refluxed for 5 hr. with 5% methanolic potassium hydroxide and the resulting triol IYIa was purified by recrystallization from dilute methanol and by sublimation a t 230' and 0.01 m m . , m.p. 275-280' (with sublimation), [ a ]+15" ~ (methanol). Anal. Calcd. for CsuHao04: C, 75.28; H, 10.89. Found: C, 75.23; H, 10.95. Conversion of Betulin to Methyl 3-Acetoxy-30-nor-20ketobetulinate (VIb).-The required betulinic acid (111) methyl ester acetate was prepared b y a slight modification of the literature directionsz2 from the readily availahlc betulin which was obtained from birch hark. A mixture of 2 g. of betulin diacetate and 0.2 g. of potassium hydroxide in 1 : 1 methanol-dioxane (400 cc.) was kept for 24 hr. at room temperature. The crude product T V ~ S chxomatographed on deactivated a l k i n a and after removing 105 mg. of unreacted betulin diacetate with benzenehexane (4:1),the desired betulin 3-monoacetate (1.6 g . , m.p. 260-263') was eluted with benzene-ether (1:1). h small amount of betulin (210 mg.) could be recovered from the column by washing with ether. The betulin 3-monoacetate (1.5 g.) was oxidized with 230 mg. of chromium trioxide in 25 cc. of glacial acetic acid for 10 minutes at 25' and 15 minutes on the steam-bath. The crude acid was methylated with diazomethane anti purified by chromatography leading to 400 mg. of methyl &acetoxybetulitiate, 1n.p. 197-201". The above acetate methyl ester (370 mg.) was left stancling with 185 mg. of osmium tetroxide in 50 cc. of dioxane for 5 days a t 0' and the osmate ester was cleaved by the sodium sulfite technique. The crude product (305 mg.) was m i dized (24 hr., room temperature) without purification with periodic acid in methanol and the crude'solid was chromatr)graphed. Elution with benzene and recrystallization from methanol-chloroform furnished the desired nor-ketone T'Ib , 2 3 m.p. 214-216', [a10 -18' (c 1.0); rotatory dispersionz4 ( 2 2 ) I*. Ruzicka, A . €1. Lainberton and 1.:. \\-. Christie, N e i u . Chrirl. A c i n . 21, 1706 (1938). ( 2 3 ) T h e literaturc constants (reference 11) for thi, s~1l)stallcc prepared by another route are: m.p. 235', [alu - 16'. (24) N'c a r c grateful t o hfrs. R. Riniker for these measureinenti.
INTERCONVERSION OF THURBEROGENIN AND BETULINIC ACID
July 20, 1 9 X
3337
T h e bisnor acid X I a (45 mg.) was recovered unchanged (Fig. 1) in dioxane: [cy1700 -26", [01]jgg -26O, [ L Y ] Z K ~ -654", " after standing at room temperature for 12 hr. with 50 mg. max." [ 0 1 ] 3 0 7 . j 336". A n a l . Calcd. for ClzHsoO;: C, 74.67; H , 9.79. Found: of chromium trioxide16 in 5 cc. of glacial acetic acid. No neutral material was formed when 40 mg. of the acid X I a C, 74.63; H , 9.87. was heated at 50-60" for 1.5 hr. with 60 mg. of lead tetraIsolation of Oxyallobetulin (IV) .-The thurberogenin for the present work was prepared' by dehydration of crude acetate16,28in 10 cc. of acetic acid. Miscellaneous Experiments. ( a ) Reaction of 30-Aldestellatogenin 3-monoacetate, which in turn had been obthe unsaturated tained from a new lot of Lemazreocereus stellatus. I n one hydo Derivatives with Base.-When aldehyde (diacetoxylupenal), obtained by selenium dioxide instance, the resulting thurberogenin acetate was found to be contaminated by another substance which could not be oxidationzg of betulin diacetate, is refluxed for 2 hr. with 29, potassium carbonate solution in 1: 1 dioxane-water, there is separated by chromatography b u t which on repeated (10-15) recrystallization was obtained in a pure state. I t s obtained the free dihydroxy aldehydez9 accompanied by some 3-monoacetate, m.p. 240-245', 2.90, 5.80 infrared carbonyl bands at 5.65 and 5.78 p indicated the and 5.90 p. When the reaction time is extended beyond 4 presence of lactone and acetoxyl groups, and the substance was identified as oxyallobetulin acetate (IVb) (m.p. ca. hours, only the free diol is isolated. 360", [ a ] +55') ~ by direct comparison with an authentic Anal. Calcd. for C32H5001: C, 77.06; H , 10.11. Found: specimen2j and by conversion to oxyallobetulone (IVc) C, 77.57; H, 10.22. (m.p. 330-333') which is the most suitable derivative.26 Under comparable conditions (or refluxing with 27, Oxyallobetulin acetate (IVb) can be removed very readily potassium hydroxide for 8-12 h r . ) , the unsaturated aldeby chromatography prior to dehydration since stellatogenin hydes derived from thurberogenin yields over 90% of acidic acetate is considerably more polar. materialLgbut no crystals were obtained, even after chro19-Ketotrisnorbetulin Diacetate (VIII).27--A solution of matography of the methylated and acetylated product. 900 mg. of 3~,28-diacetoxy-19a-methoxycarbonyltrisnorlu- The same applied to the saturated aldehyde X I V and to thc panel3 in 200 cc. of benzene was added to a boiling solution unsaturated aldehyde X. of phenylmagnesium bromide prepared from 2 g. of magne( b ) Reaction of 30-Nor-20-ketothurberogenin Acetate sium, 6 cc. of bromobenzene and 20 cc. of ether. T h e ether (V) with Perbenzoic Acid.--A solution of 220 mg. of the was removed and the benzene solution mas refluxed for 6 hr. norketone V and 20 mg. of p-toluenesulfonic acid mono.kfter decomposition with ammonium chloride, the crude hydrate in 15 cc. of ethyl acetate was kept at 0" for 4 days product was passed through a column of alumina. T h e with 1.1 equivalents of perbenzoic acid in 1 cc. of chlorosolid (440 mg., m.p. 195199') apparently represented a form. Over 95% of unreacted starting material was remixture of the diphenylcarbinol and the diphenylethylene covered. T h e same results were observed in chloroform since it possessed A%": 250 mp, log e 3.97. solution in the absence of p-toluenesulfonic acid. A 195-mg. sample was dehydrated completely by refluxing ( c ) Reactions of 30-Nor-20-ketothurberogenin Acetate for 3 hr. with 25 cc. of acetic anhydride and 4 cc. of pyridine, (V) Oxime.-The Beckmann rearrangement was carried out and the product was purified by chromatography (m.p. with phosphorus oxychloride in pyridine solution,18 b u t 247-251°, A"$: 248 mp, log e 4.3), b u t no analytical sample less than 5Vo of crude, acidic material was produced. of the diphenylethylene was prepared. T h e total material dcetylation of the 3-acetoxy oxime8 with acetic anwas ozonized in chloroform solution a t 25' and the ozonide hydride-ppridine followed by recrystallization from methdecomposed in acetic acid solution with zinc dust. Chro- anol-chloroform led t o plates of the 3-acetoxy oxime matography on alumina removed benzophenone in the benacetate, m.p. 245-249' dec., 5.80 and 6.05 p. zene-hexane fractions while elution with ether-benzene Anal. Calcd. for Ca3Hlg?;O6: C, 71.32; H, 8.89. Found: furnished the desired trisnorketone VIII, which was reC, 71.20; H, 9.12. crystallized from methanol-chloroform; m.p. 246-250", I n a n attempt to prepare a 19-hydroxy-20-amine which [a]D+37.5", X ~ ~5.79 ~ ' and a 8.01 p . might be useful for certain ring E enlargement reactions, Anal. Calcdyfor C31HasOj: C, 74.36; H , 9.66. Found: 95 mg. of the oxime8 was refluxed for 2 hr. with an excess of C, 74.35; H , 9.65. lithium aluminum hydride in ether. Acetylation, chro33,Z8-Diacetoxy- 193-hydroxy- 19a-me thoxycarbonyltris- matography and recrystallization from methanol-chloroform norlupane (XIb).-&I stream of ozonized oxygen was passed produced 30 mg. of the oxime acetate VIId, m.p. 222-232' through a solution of 450 mg. of the triol-diacetoxy-aldehyde dec., 5.80 and 6.07 p . XS in 50 cc. of acetic acid for 40 minutes at 20". Water .4nal. Calcd. for CljHj5XOi: C , 69.83; H, 9.21; N, (50 cc.) was added, the solution was heated on the steam2.33. Found: C, 69.87; H , 9 . 1 1 ; S , 2 . 3 9 . bath for 5 minutes and the crystalline material (330 mg., m.p. 199-207') was collected. This appeared to be the ( d ) Attempted Dehydration of Dihydrothurberogenin intermediate ozonide and a n analytical sample mas obtained Triol Diacetate (VIIe).-The resistance of the tertiary by dissolving in ether, washing with dilute potassium bihydroxyl group in thurberogenin triol diacctate has alreadycarbonate and recrystallizing rapidly from dilute acetic been noteds and several experiments were now carried out in acid; colorless crystals (bright yellow when hot), m.p. 208order t o determine whether dehydration would be more 212', [ a ] +~l o , no high selective absorption in the ultrafacile in the dihydro analog lr1Ie. Xn uncrystallizable gum was obtained when the diacetate was refluxed for '7 violet, 2.90 and broad band a t 5.78-5.81 p . Anal. Calcd. for C3&109: C, 67.52; H, 8.67. Found: hr. with boron trifluoride etherate in benzene while the following experimental conditions resulted in 80-90$, C , 68.01; H , 8.86. recovery of starting material; refluxing for 2 hr. in benzene The mother liquors of the ozonide were evaporated t o solution with thionq-l chloride, refluxing for 30 minutes with dryness and crystallized from dilute acetic acid to furnish phosphorous oxychloride in pyridine, refluxing for 1 hour 80 mg. of the desired bisnor acid X I a , m.p. 251-260". with acetic anhydride and sodium acetate or shaking with an The same acid was produced in nearly quantitative yield excess of phosphorus pentachloride in petroleum ether for 30 when the crystalline ozonide was warmed for 5 minutes with minutes. a 5(% solution of 30y0 hydrogen peroxide in acetic acid. ( e ) Reactions of the E-Homo Rearrangement Product of For characterization purposes, a specimen of the acid was V.--Xny correlation of thurberogenin with a member of the methylated with diazomethane, passed through an alumina taraxastane group would require elimination of the 1901column and recrystallized from methanol-chloroform. The methyl ester X I b exhibited m.p. 171.5-173.5', [ a ] ~hydroxy group of the E-homo rearrangement product (as the 3-acetoxy methyl ester).s The hydroxyl group was not +19Y , -_ . Anal. Calcd. for C33Hj20;: C, 70.68; H, 9.35. Found: removed by zinc dust in refluxing acetic acid and since keto1 acetates react more readily under those c0nditions,3~a n C , 70.55; H, 9.25. attempt was make t o acetylate the 19a-hydroxy group under
+
(25) Prepared f r o m betulin via allobetulin formate according to H. Schulze and K. Pieroh, Ber., 55, 2332 (1922). ( 2 6 ) G. S.D a v y , T. G. Halsall, E. R. H. Jones and G. D. Meakins, J. Chrnz. S o c . , 2702 (1951). (27) This experiment was carried o u t by Dr. L H. Liu and hIr. R . S McDonald using the procedure employed earlier (reference 14) in t h e lupeol series.
(28) Cf.E . Rohrmann, R . G. Jones and H. A. Shonle, THISJOURNAL, 6 6 , 1856 (1944). (29) I,. Ruzicka, 31. Brenner and E. Rey. Heiv. Chiriz. A c t a , 2 6 , 161 (1942). (30) Cf.R . S. Rosenfeld and T. I?. Gallagher, THISJOURNAI., 77, 4307 (1955), and references cited.
3538
COMMUNICATIONS TO THE EDITOR
acid-catalyzed conditions which have proved effective with 17~z-hydroxy-20-ketosteroids,31but only unreacted starting material was recovered. (31) Cf.Huang-Minlon, E. Wilson, PI'. L. Wendler and M. Tishler, THIS J O U R N A L , 74, 5394 (1952); R. B. Turner, ibid., 76, 3489 (1953).
Vol. 7 8
Acknowledgment.--We should like to acknowledge the benefit of a stimulating discussion wit. Prof. Gilbert Stork of Columbia University. DETROIT,MICHIGAN
COMMUNICATIONS T O T H E EDITOR T H E RESOLUTION O F 0-ETHYL ETHYLPHOSPHONOTHIOLIC ACID
Sir:
presence of a reactive group directly attached to phosphorus in a resolved compound of this type provides one with a convenient tool applicable t o a study of the reactions and stereochemistry of the asymmetric phosphorus atom. Detailed reports on the resolution, reactions and stereochemistry of this and similar compounds will be published a t a later date.
We wish to record the successful resolution of 0-ethyl ethylphosphonothiolic acid (C?Hs(C*HsO)P(O)SH).l The compound was resolved by fractional recrystallization of its quinine salt (I) from acetone-ether. The more insoluble diastereoisoRESEARCH DIVISION meric salt (Ia) crystallized as a monohydrate : CHEMICAL DIRECTORATE OF RESEARCH HERBERT S.AARON prisms, m.p. 151-153O (with loss of its water of CHEMICAL WARFARE LABORATORIES JACOB I. MILLER hydration), [ a I g 6 D -96.6 + 0.S' (sobs -1.990 + ARMYCHEMICAL CENTER,MD. 0.015O, acetone, 2-dcm., c = 1.130), equiv. wt., RECEIVED M A Y21, 1956 492 (calcd. 497 for the monohydrate). IVhen vacuum dried over phosphorus pentoxide for three ELECTROPHORETIC DEMONSTRATION OF T H E hours a t looo,I a gave rise to anhydrous product. ISOMERIZATION O F BOVINE PLASMA ALBUMIN m.p. 15S-1GOo, [ c u ] ~ ~-9'7.6 D =k 0.6' (aohs -1.0'70 AT LOW pH 0.007', acetone, 1-dcm., c = 1.096), equiv. wt., Sir: 469 (calcd. 479). Recently much interest has been exhibited in a The acid was separated from I a as its sodium pronounced conformational change which takes salt in an essentially aqueous solution by treating place in bovine plasma albumin a t pH values acid I a in methanol with an equivalent amount of to the isoelectric point. I t was first suggested by aqueous base. T h a t the phosphorus atom main- Tanford' that expansion of the protein molecule tains its tetrahedral configuration in the anion is results upon titration with acid. Gutfreund and demonstrated by the optical activity of the product Sturtevant' demonstrated a slow thermal effect recovered from the sodium salt. The acid was re- upon adding acid to this protein. Yang and Foscovered by the addition of an equivalent amount of ter3 demonstrated a parallel and reversible endilute hydrochloric acid to the sodium salt solution hancement of the optical rotation and intrinsic and extraction of the product from the resulting viscosity acid to p H 4, and suggested that there solution with ether. The acid was characterized as exists an all-or-none equilibrium between two its dicyclohexylamine salt: m.p. 139- 1C10.5~ forms of the protein molecule. Tanford4 has re[a]?'D -7.11 zk 0.23' ( a n b e -0.153 =I= 0.003', cently shown evidence for an intermediate which methanol, 1-dcm., c = 2.150), found: C, 37.38; he terms the "expandable" form. H , 10.00 (calcd. for C16Hd403NP: C, 5'7.2s; H , \Ve have recently been successful in attaining 10.22). excellent resolution of two boundaries in the elecAfter the removal of a mixed middle crop of I, trophoretic patterns of this protein over the FH the more soluble diastereoisomeric salt (Ib) crys- range 4.6 to 3.5. Heterogeneity of plasma altallized as soft anhydrous needles: m.p. 1GG- 16S0, bumins in this pH range has been reported prc[CU]?~D -S1.'7 + 0.6' ( m o b s -1.613 f 0.012, ace- ~ i o u s l y . ~ -However, ~ we can now demonstrate tone, 2-dcm., c = 0.9868);equiv. wt., 475 (calcd. that this heterogeneity is due in the main to a fiH 479). The dicyclohexylamine salt of this enantio- dependent transition of the normal form of the promorph of the acid gave m.p. 158-160°, ( c ~ ) ' ~ D tein into a faster migrating form. presumably of +6.85 0.23' (sobs 0.221 + 0.008", methanol, higher positive charge. Results summarized in 1-dcm., c = 3.230), found: C, 57.30, H 10.02, mixed (1) C. Tanfurd, Proc. I o z , a A c a d . S c i . , 69, 200 (1932). melting point with the enantiomorphic dicyclo(2) H . Gutfreund and J . Sturtevant, THIS J O U R K A I . , 76, 5k17 hexylamine salt, above. 163- 165'. Racemic 0- (1933). ( 3 ) J . T. Y a n g and J . Ii. Foster, i b i d . , 76, I388 (19.51); 77, 2371, ethyl ethylphosphonothiolic acid forms a dicyclo3805 ( 1 9 ~ 5 ) . hexylamine salt, m.p. 166- 16s'. (1) C. Tanfurd, J . Buzzeil, D. Rands a n d S . Snanwii, i b i i i , 77, 1,421 This communication represents the first reported (195.5). resolution of a phosphorus acid, the optical ac( 5 ) J . Luetscher, ibid., 61, 2888 (1939). hriii , (A) D. Sharp, G. Cooper, J. Erickson and H . h'eurath, J . Bioi ti\ity of which is due solely to the presence of an 144, 139 (19L?). I n this paper i t was also shown that hetemjieneity asymmetric phosphorus atom. AIoreover, the c1is:rpl)ears b e l o w pH 3 . 5 , in uccord with oiir (Iwn r e i u l t . .
*
+
( 1 ) T h e preparation o f alkylph~,sphonath,~,lic a c ~ d . \\,I1 be d e w r l b p d in a forthcoming paper h > I* XT' Hufim,inn a n d r i ! iT
