J. Mol. Evol. 10, 161--166 (1977)

Journal of Molecular Evolution © by Springer-Verlag 1977

Stereoselective Crystallization Induced by Traces of Dissolved Optically Active Impurities K. L. Kovfics Institute of Biophysics, Biological Research Center, Hungarian Academy of Sciences, H-6701 Szeged, Hungary P.O. Box 521 Summary. Isothermal crystallization of D,L-sodium-ammonium tartrate with traces of different impurities admixed shows that the added chiral contaminations produce a preferential crystallization of the tartrate isomer of same handedness. The critical lowest concentration of effective seeding agents is 0 . 1 - 0 . 5 %. 1% optically active excess material induces 1 . 0 - 3 . 6 % optical purity in the deposited crystals. An analysis of the relevant data reported so far gives similar figures in different crystallization systems. The relation of the results to the suggested lattice energy difference between enantiomers is discussed. Key words: Induced resolution - Chiral influence - Sodium-ammonium tartrate - Origin of optical activity - Stereoselective crystallization.

Several different possible roles of crystals in chemical evolution have been recognized. Crystal structure provides a repetitive, organized matrix for accumulation of materials in a stable and pure form and it can serve as catalytic surface for prebiological reactions as well (Bernal, 1967). From the point of view of the origin of life's fascinating preference of one optical isomer .to its antipode the existence of separable mixture of asymmetric crystals seems to be essential (Northrop, 1957; Harrison, 1974). Numerous compounds are known to form racemic conglomerates i.e., from the saturated solution of the racemate a mixture containing identical amount of individual crystals of each antipode deposits. Once the ratio of L/D crystal seeds is shifted from 1 by any impact, the difference will be amplified as crystallization proceeds. Yamagata (1966) postulated a slight energy content difference between enantiomers. In the first a t t e m p t to substantiate the existence of a lattice energy difference Thiemann and Wagener (1970) precipitated D,L-sodium-ammonium tartrate. Optical activity of the same sense was found in every one of 10 trials. Other reports include polymerization (Thiemann and Darge, 1974) as well as crystallization of asparagine (Thiemann, 1974b). In the latter experiment the difference was established starting from both racemic and nonracemic conditions.

162

K.L. Kovacs

A considerable amount of work has been devoted to the study of resolutions by inserting crystals or solutions of optically active compounds into saturated solutions of racemates (Secor, 1963; Bonner, 1972). The method has been widely employed also in manufacturing optically active chemicals. F r o m the evolutionary point o f view such studies may represent an experimental model for amplification mechanisms o f any initial asymmetry via autocatalysis during chemical evolution. There are few investigations designed to obtain the quantitative relationship between the amount of added seeding material and its stereoselective effect. Thus Wald's (1957) statement about the surface of the Earth being contaminated with one enantiomer due to the presence of life is commonly accepted and cited as a consideration why truly unbiased stereochemical reactions can be performed only with great difficulties today. Studies, however, on what is the critical lowest concentration of optically active "impurities" which still cause stereoselectivity are missing from the literature. The purpose of the present investigation was to find out what amount of asymmetric impurity is able to make its influence felt in autocatalytic processes as well as to check the suggested crystal lattice energy difference between enantionmers.

Experimental Enantiomers of sodium-ammonium tartrate (Fluka), mandeleic acid (Norse), phenylalanine (Merck) and leucine (Reanal) were used without further purification. D,L-tartaric acid (Reanal) was recrystallized several times from water until it was optically inactive (checked by the UV region of its circular dichroism spectrum on a JASCO 40c spectropolarimeter). Pure tartaric acid was then converted to its sodiumammonium salt and recrystallized again. 60 g of D,L-sodium-ammonium tartrate was dissolved in 100 ml freshly distilled water giving a final volume of 1 2 5 - 1 3 0 ml. Crystallizations were conducted in small glass test-tubes. Each sample contained 2 ml of the racemic tartrate solution into which always the same volume of dissolved "impurity" was pipetted. Controls received distilled water. The tubes were randomly positioned into a holder that fits into a desiccator. The desiccator, filled with P205, was closed at atmospheric pressure and kept at 18+2 ° C. After four weeks the crystalline phase was separated carefully from the supernatant and dried over P205. Special care was taken to avoid external contamination throughout the above operations. The probes were weighed on a Sartorius balance. Optical activity of the crystals was determined after dissolution in water on a JASCO 40c spectropolarimeter at a fixed wavelength (215nm) at the CD maximum of the tartrate in N 2 atmosphere.

Results and Discussion The effect of traces of four different enantiomer pairs has been investigated. Figures 1 and 2 summarize the results. Each point on the curves represents the average of 10 individual crystallizations. The standard deviation is indicated at the point belonging to controls (i.e., no contamination added deliberately).

Impurities in Stereoselective Crystallization

16 3 Fig, 1. Induction of optical activity in the crystallization of D, L-sodiumammonium tartrate by added substances: ( • ) L(+)-sodiumammonium tartrate ( - D - ) L-leucine, (--x-) D(--)-sodium-ammonium tartrate, (--t--) D-leucine

2OO

C E g •o I00

"o

T o --L ~ - A .................

~

i

J

!

IJ

Control

-I()

-8

-6 -t, -2 log excess motcrial cone. [ M / I ]

Fig, 2. Induction of optical activity in the crystallization of D, L-sodiumammonium tartrate by added substances: ( - A - ) (+)-mandeleic acid; ( - o - ) L-phenylalanine, (--m-) ( - ) mandeleic acid, ( - 6 - ) D-phenylalanine

200

100

{o

5

-~00

-200

c~,o,

-6

-6

-~

-,.

-~

log CXC¢SS n'~terioI co~. [M/I}

164

K.L. Kov~cs

It is apparent that the added substances induce b o t h weight increase and optical activity depending on their concentration. R o u g h l y ten times more material is needed to cause an increase in accumulation of the tartrate than the limit where optical activity appears. Within the studied c o n c e n t r a t i o n range leucine did n o t increase the mass of the crystalline phase, although the stereoselective influence o f its optical isomers clearly observable. Nucleation of the tartrate isomer having the same handedness as the impurity is preferred. This is in line with earlier observations (Kipping and Pope, 1898; Ostromisslenskii, 1908; McKenzie, 1915; Secor, 1963). The results seem to indicate that 0 . 1 - 0 . 5 % chiral a d m i x t u r e c o n c e n t r a t i o n produces significant (at 2 % confidence level) optical activity in the deposited crystals of D , L - s o d i u m - a m m o n i u m tartrate, and the effect is fairly i n d e p e n d e n t o f the nature of the added material. The analysis of the few relevant reported data discloses that the above critical lowest c o n c e n t r a t i o n of effective impurities applies to other systems as well. Thus in T h i e m a n n ' s (1974) e x p e r i m e n t 0 . 1 % c o n c e n t r a t i o n of extra enantiomers of asparagine caused appearance o f optical activity in precipitated D,L-asparagine. A detailed study has been carried o u t on copper c o m p l e x e s of aspartic and glutamic acids (Harada, 1970 and references therein). In order to prepare up to 100 % optically pure crystals a considerably high c o n c e n t r a t i o n o f chiral seeding agent was used. AssuTable 1. Lowest critical effective concentrations of impurities in the isothermal crystallization of D, L-sodium-ammonium tartrate at 2 % significance level Weight Material

Effectivea Molar ratio X10 2

Average weight b g

Optical

Activity

Effectivea Molar ratio X10 2

Average (0)bls Optical deg cm 2 decimole-1 purity c (%)

D-(-)-sodiumammonium tartrate

1, 3

0.62

0.13

--135.6

0.9

L(+)-so diumammonium tartrate

1.3

0.56

0.13

+131.2

0.9

(-)-mandeleic acid

0.4

0.73

0.04

-249.2

1.6

(+)-mandeleic acid

0.4

0.77

0.04

+243.5

1.6

D-phenylalanine

1.3

0.45

0.40

-125.4

0.8

L-phenylalanine

1.3

0.46

0.13

+319,8

2.1

D-leucine

0. 57

1.3

-- 66.0

0.4

L-leucine

0.42

1.3

+ 74. 2

0. 5

a where significant (at 2% confidence level) deviation from controls is observed; moles of added substance/moles of tartrate b at 1.3x10 "2 molar ratio 2 c at 1% impurity concentration:(0)215 of pure D-sodium-ammonium tartrate is --6850 deg cm decimole - 1

Impurities in Stereoselective Crystallization

165

ming linear relationship between the concentration of added stereoisomers and the yield of resolution one can calculate that 1 % optically active amino acid in the system will result an optical purity of 1.1 ~2.0 % depending on the particular amino acid (for the calculation the data were taken from Harada, 1972a; Harada, 1972b). The same number in our work varies from 0 . 4 - 2 . 1 % (see Table 1); an excellent agreement in spite of the different experimental conditions. Pin cock (1974) induced resolution of 1, l ' - b i n a p h t h y l by adding chiral compounds to the melts at 1 3 0 - 1 4 5 ° C. 18 chemicals were checked but no clear picture was obtained. Certain impurities did not promote the stereoselective crystallization at all, in other cases both enantiomers o f the compound gave predominance of the same 1, l'binaphthyl isomer. Only administering of 5 % mandeleic acid provided absolute proof o f control of nucleation. From these data an optical purity o f 3 . 0 - 6 . 2 % can be estimated at 1% impurity concentration. This figure lies not too far from our results especially if the entirely different circumstances are also taken into account. The average o f the 40 control samples gives a small positive rotation (+4.4 deg cm 2 decimole -1 , standard deviation +10.6, i.e.; 0.06 % optical purity) which seemingly supports the existence of an intrinsic asymmetric property of enantiomers (Thiemann and Wagner, 1970). This is, however, misleading as in other experiments similarly minute but negative rotations were obtained (Kov~cs and Garay, 1975; Kov~cs, 1977). Therefore the nonzero rotation of controls is felt to be due to mere statistical fluctuation. An interesting phenomenon (Thiemann and Wagener, 1970) regarding the resistance of saturated tartrate solutions against infection has been checked and corroborated. Saturated sodium-ammonium tartrate was left repeatedly in an Erlenmayer flask covered only with aluminum foil on the laboratory bench. The solution displayed no sign of being infected (turbidity, etc.) and remained optically inactive during the entire 4 weeks of the experiment. We may conclude that either the rigorous precautions mentioned as a must in most of the related discussions (Wald, 1957; Thiemann, 1974a) do not affect the outcome of the experiment so markedly; or else saturated solutions posses special bactericid and/or fungicid properties. An early work done on tartrates (Kipping and Pope, 1898) points out that in completely open containers more D (natural isomer) crystals are formed than L (unnatural isomer) ones. The "spontaneous resolution" diminished when proper protection against dust particles from the atmosphere was executed. The symmetric pattern of the curves in Figure i shows that the magnitude of the effect of both antipodes is the same. Phenylalanine represents an exception, but the used D-phenylalanine sample turned out to be impure. Similarly to the crystallizations under racemic starting conditions at this point our results are again in contradiction with those of Thiemann (1974b). In the crystallization of D, L-asparagine he found a difference not only in the sign of induced optical activity but also in its magnitude upon adding extra L- or D-asparagine. The deviation has been explained as strong evidence illustrating an energy difference between left and right handed crystal lattices. However, the above observation with impure phenylalanine enantiomers may offer a simple reason for the asymmetric behavior of asparagine.

166

K.L. Kovacs

It is concluded that the asymmetric effect due to mixing of weak interactions with the electromagnetic ones is not experimentally testable at the present level of accuracy of measurements.

Acknowledgement. The technical assistance of Ms. M. Toth and Zso Weberis gratefuUyacknow-

ledged.

References

Bernal, J.D. (1967). The origin of life. London: Weidenfeld and Nicolson Bonner, W.A. (1972). Origins of molecular chirality. In: Exobiology, C. Ponnamperuma, ed., pp. 120--234. Amsterdam-London: North-Holland Harada, K. (1970). Naturwissenschaften 57, 114 Harada, K. (1972a). Bull Chem. Soc. Japan. 45, 2859 Harada, K. (1972b) Chem. Letters. 1057 Harrison, L.G. (1974). J. Mol. Evol. 4, 99 Kipping, F.S., Pope, W.J. (1898). J. Chem. Soc. 73, 606 Kov~cs, K.L., Garay, A.S. (1975). Nature 254, 538 Kov~cs, K.L. (1977). Proc. 2nd ISSOL Mecting, Kyoto, Japan McKenzie, A. (1915). J. Chem. Soc. 107, 440 Northrop, J.H. (1957). Proc. Nat. Acad. Sci. U.S. 43, 304 Ostromisslenskii, I. (1908). Ber. 41, 3035 Pincock, R.E., Branshaw, R.P., Perkins, R.R. (1974). J. Mol. Evol. 4, 67 Secor, R.M. (1963). Chem. Rev. 63, 297 Thiemann, W. (1974a). Naturwissenschaften 61,476 Theimann, W. (1974b). J. Mol. Evol. 4, 85 Thiemann, W., Darge, W. (1974). Origins of Life 5, 263 Thiemann, W., Wagener, K. (1970). Angew. Chem. 82, 776 Wald, G. (1957). Ann. N.Y. Acad. Sci. 69, 369 Yamagata, Y. (1966). J. Theoret. Biol. 11, 495 Received February 11, 1977

Stereoselective crystallization induced by traces of dissolved optically active impurities.

J. Mol. Evol. 10, 161--166 (1977) Journal of Molecular Evolution © by Springer-Verlag 1977 Stereoselective Crystallization Induced by Traces of Diss...
268KB Sizes 0 Downloads 0 Views