Chiral self assembled monolayers as resolving auxiliaries in the crystallization of valine

Authors

  • Aniruddh Singh,

    1. Department of Chemical and Biological Engineering, Illinois Institute of Technology, 10 W. 33rd St. Chicago, Illinois 60616
    Current affiliation:
    1. Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts, 02139.
    Search for more papers by this author
  • Allan S. Myerson

    Corresponding author
    1. Department of Chemical and Biological Engineering, Illinois Institute of Technology, 10 W. 33rd St. Chicago, Illinois 60616
    • Department of Chemical and Biological Engineering, Illinois Institute of Technology, 10 W. 33rd St. Chicago, Illinois 60616, Telephone: + 1 312 5673101; Fax: +1 312 5678857.
    Search for more papers by this author

Abstract

Chiral drugs are a subgroup of drug substances that contain one or more chiral centers. For reasons of safety and efficacy, the pure enantiomer is usually preferred over the racemate in many marketed dosage forms. Thus, resolution of racemic mixtures is an active area of research. In this work, chiral self assembled monolayers (SAMs) on gold were employed as resolving auxiliaries in the crystallization of the amino acid valine. Results showed the ability to obtain one enantiomer in excess on the crystals grown on the chiral SAMs when starting with racemic solutions. The enantiomer obtained in excess was the one having opposite chirality to the monolayer being used. In addition, it was possible to obtain crystals of the pure enantiomer when starting with a solution having an enantiomeric excess value of 50%. Control experiments carried out without chiral SAMs showed that at equilibrium, mixtures of the pure enantiomer and racemic compound were obtained under these conditions. The enantiomer obtained on the chiral SAMs was the one that was initially present in excess regardless of the chirality of the monolayer being used. © 2010 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 99:3931–3940, 2010

INTRODUCTION

Chirality is defined as the geometric property responsible for nonidentity of an object with its mirror image. A chiral object may exist in two enantiomorphic forms which are mirror images of each other. Objects that are super imposable on their mirror images are achiral. At the molecular level, there are achiral as well as chiral molecules. Chiral molecules exist in two enantiomeric forms; the term enantiomorphic is generally applied to macroscopic objects. The oldest known manifestation of molecular chirality is the optical activity, the property, that is, exhibited by the rotation of the plane of polarization of light. The two enantiomers of a given compound have opposite optical activity. One is positive or dextrorotatory, while the other is negative or levorotatory.1

An equimolar mixture of two enantiomers whose physical state is unspecified or unknown is called a racemate. The separation of the two enantiomers that constitute a racemate is called a resolution or an optical resolution. Crystalline racemates may belong to one of three different classes, namely, conglomerate, racemic compound, or pseudoracemate. A conglomerate is a mechanical mixture of the crystals of the two pure enantiomers. The most common type of crystalline racemate is that in which the two enantiomers are present in equal quantities in a well-defined arrangement within the crystal lattice. The resultant homogeneous solid phase is called a racemic compound. The third possibility corresponds to the formation of a solid solution between the two enantiomers coexisting in an unordered manner in the crystal. This solid solution is known as a pseudoracemate.

Chiral drugs are a subgroup of drug substances that contain one or more chiral centers. More than one-half of marketed drugs are chiral.2 In some cases, chiral drugs are marketed as racemates as it is frequently costly to resolve a racemic mixture into pure enantiomers. Opposite enantiomers often differ significantly in their pharmacological,3 toxicological,4 pharmacodynamic, and pharmacokinetic5, 6 properties. From the points of view of safety and efficacy, the pure enantiomer is preferred over the racemate in many marketed dosage forms.7 Thus, resolution of enantiomer mixtures is an active area of research.

Racemic compound systems account for more than 90% of all racemic mixtures8 and cannot be resolved by direct crystallization. A common route to resolution is to react the racemic mixture with an optically pure resolving agent, such as an acid or a base, to give a mixture of diastereomers which are nonenantiomeric and hence have different physical properties. The phase diagram of such a mixture is generally asymmetric and crystallization will yield crystals enriched in one of the isomers.

Chiral surfaces and interfaces have received considerable interest in recent years because of their importance in separation of enantiomers.9 Cleavage of quartz or calcite, materials with chiral bulk structures, leads to surfaces that are naturally chiral. Chiral surfaces can also be created by anchoring or adsorption of a chiral molecule on a nonchiral surface.10–12 The ability to differentiate between enantiomers of a chiral molecule is one of the most interesting properties of a chiral surface. The use of both naturally chiral surfaces and chiral self-assembled monolayers as resolving auxiliaries in the crystallization of racemates has been previously reported. Naturally chiral bulk crystalline surfaces have been shown to display chiral selectivity when immersed in a racemic solution.13 Self-assembled monolayers of (+)-L and (−)-D cysteine on gold substrates were used as resolving auxiliaries in the crystallization of rac-glutamic acid (a conglomerate) and rac-histidine (a racemic compound).14, 15 Similarly, chiral discrimination between (R)- and (S)-thalidomide enantiomers was achieved using SAMs of (R)- and (S)-1,1′-binaphthalene-2, 2′-dithiol (BNSH).16 Chiral discrimination between (D)- and (L)-phenylalanine enantiomers was achieved using SAMs of (BNSH).17

In this work, we used self assembled monolayers (SAMs) of L-cysteine, D-cysteine, N-acetyl-L-cysteine, L-glutathione, and D-penicillamine on gold as resolving auxiliaries in the crystallization of the amino acid valine which crystallizes as a racemic compound.

MATERIALS AND METHODS

Materials

L-Cysteine (LC) was obtained from Acros Organics, Morris Plains, NJ. D-Cysteine (DC), L-glutathione (LG), D-penicillamine (DP), N-acetyl-L-cysteine (NALC), DL-valine, d-valine and L-valine were purchased from Sigma–Aldrich Corp., St Louis, MO. Deionized water was obtained from a Barnstead Nanopure Infinity water purification system. Titanium (99.995%) and gold pellets (99.999%) were purchased from Kurt J. Lesker Company, Clairton, PA. Table 1 shows the chemical structures of the chiral SAMs used in this work.

Table 1. Chemical Structures of the Chiral Self-Assembled MonolayersThumbnail image of

Gold Surfaces and SAMs Preparation

Microscope glass slides were immersed in “piranha solution” (3:1 concentrated H2SO4/30% H2O2) for 30 min. Caution: Piranha solution reacts violently with organic materials and should be handled with extreme care. The glass slides were then thoroughly washed with water and rinsed with copious amounts of ethanol and blown dry with nitrogen. Gold surfaces were prepared by evaporation of titanium onto glass slides, followed by evaporation of gold. The deposition of metals on the glass slides was carried out using an electron beam evaporator (Thermionics Vacuum Products, Clawiter Hayward, CA, base vacuum of 10−7 torr). The slides were first coated with a thin layer of titanium (∼100 Å) to promote adhesion followed by a layer of gold (∼500 to 1000 Å). The SAMs were formed on the gold surfaces by immersing the substrates overnight in 10 mM solutions of L-cysteine, D-cysteine, N-acetyl-L-cysteine, L-glutathione, and D-Penicillamine in deionized water. After removal from solution, the substrates were rinsed with copious amounts of deionized water and blow dried with nitrogen. Self assembled monolayers of cysteine on gold have been previously studied using X-ray photoelectron spectroscopy (XPS) and Scanning Tunneling Microscopy (STM).14, 18 It has been reported that cysteine molecules are absorbed onto gold through the sulfur atom, while leaving the other two groups (COOH and NH2) free. STM images showed that LC chemisorption induces the lifting of the equation image reconstruction of the Au(111) surface and that the LC adlayer adopts a hexagonal equation image structure.18 The self-assembly of NALC on gold has been studied by in situ attenuated total reflection infrared (ATR-IR) spectroscopy, polarization modulation infrared reflection absorption spectroscopy (PM-IRRAS), and a quartz crystal microbalance (QCM)19 and it has been reported that the resulting structure of the adsorbate layer is quite complex. Based on the initial ATR-IR spectrum, the orientation of the molecule within the adsorbate layer was determined. The orientation analysis showed that the amide group is tilted with respect to the surface. In such an orientation, assuming that the sulfur binds to the gold, the carboxylic acid group is in proximity to the surface. Adsorption of LG on gold surfaces has been investigated by PM-IRRAS and ATR-IR.20 PM-IRRAS was used to study ex situ the adsorbate layer prepared from aqueous solutions at different pH, whereas ATR-IR was applied to study in situ adsorption from ethanol in the presence and absence of acid and base. On the basis of such investigations it was proposed that in the absence of ions in water the molecule forms a T shape with the two arms (Glu and Gly parts) of glutathione far apart from each other. The molecule is firmly anchored on the gold surface via the thiol group of the cys part. The self-assembly of DP on gold has been investigated by ATR-IR spectroscopy 21. The infrared spectra gave evidence for a three-point interaction between DP and the gold surface involving the thiol, amine, and carboxylate functional groups and density functional theory calculations supported this finding. In this adsorption mode, the thiol is bridged between two gold atoms and both the amine and carboxylate are involved in surface interaction. It has also been suggested that the three-point surface interaction makes DP an attractive candidate for chiral surface modification.21

Crystallization Experiments from Supersaturated Racemic Solutions

To conduct crystallization experiments supersaturated solutions of DL-valine in deionized water were prepared. In each experiment DL-valine in excess was slurried in deionized water in 100 mL glass jars or in 20 mL scintillation vials. The slurries were heated to dissolve and then cooled to room temperature to prepare supersaturated solutions. Gold surfaces with self-assembled chiral monolayers were then immersed in these solutions. Once crystals were observed on the chiral surfaces, they were carefully removed for characterization. Control experiments were also carried out for each experimental condition. In these experiments valine was crystallized from bulk solution without any chiral surfaces being immersed, to confirm that the results being obtained were due to the effect of the SAMs. Microscope images of the crystals on the chiral SAMs were acquired with a polarized light microscope (Nikon Eclipse ME600). The crystals were then removed from the surfaces and ground into a fine powder for characterization with X-ray Powder Diffraction (XRPD). Since preferred orientation significantly affects the relative peak intensities observed in XRPD,22 the samples were ground to less than 10 µm size particles. XRPD patterns were obtained with a Rigaku Miniflex diffractometer using monochromatic Cu Kα radiation with a nickel filter (λ = 1.54 Å) generated at 30 kV and 15 mA. The data were collected from 2 to 40° with a step size of 0.1° at a scan rate of 1.0°/min. Aluminum sample holders with a zero background silicon plate were used to carry out measurements.

The XRD patterns of the samples were compared with reference patterns of DL-valine and L-valine shown in Figure 1. Since the XRPD patterns of pure enantiomers are identical, the reference pattern of D-valine is the same as L-valine.

Figure 1.

XRPD patterns of DL-valine and L-valine.

A comparison of the reference patterns of DL-valine and L-valine shows that it is possible to differentiate between the pure enantiomer and the racemate using XRPD as the racemate is a racemic compound with a XRPD pattern that is different from the pure enantiomer. The presence of a crystalline diffraction line unique to the pure enantiomer (at ∼29.5° 2θ) was also noted and this high intensity peak was defined as the characteristic peak of the pure enantiomer. The XRPD pattern of the samples crystallized on the chiral surfaces may match either DL-valine, indicating that no chiral resolution has taken place; or the pure enantiomer, indicating that even though we started with a supersaturated solution of DL-valine, only one of the pure enantiomers has crystallized on the chiral surfaces. A third option is that an enantiomeric excess (EE) of one of the enantiomers is present in the crystals. In this case the XRPD pattern of the crystals would show characteristic peaks corresponding to both the racemate and the pure enantiomers.

To identify which enantiomer was present in excess or was obtained in the pure form we used a Rudolph Research Analytical Autopol IV Automatic Polarimeter (Model APIV/6W w/Temptrol) with an accuracy of ±0.001° using a sodium lamp operating at 589 nm. The solids obtained were dissolved in deionized water and the optical rotation of these solutions was measured in the polarimeter at 20°C. To measure the EE value of the solids obtained on the chiral SAMs, a calibration curve (optical rotation vs. EE) was first prepared for EE values ranging from 10% to 50% EE and the values of optical rotation were noted for each of these solutions at a particular concentration. Next a solution of the unknown solids (at the same concentration) was prepared in deionized water and its optical rotation was measured and the EE value was obtained from the calibration curve.

Obtaining a Ternary Solubility Diagram for the Valine–Water System

Ternary solubility diagrams of chiral compounds in a solvent give valuable information about the solubility of pure enantiomers, racemic mixtures and mixtures of the two enantiomers of the compound in varying compositions (different enantiomeric excess values) in the solvent. These diagrams are useful for optimizing crystallization conditions. However we were unable to find a ternary solubility diagram for the valine–water system at 25°C in the literature, although one at 60°C has been reported.23 To obtain a solubility diagram we prepared undersaturated solutions of varying concentrations of DL-valine in 10 mL water at 25°C. These solutions were prepared in 20 mL scintillation vials that were kept at constant temperature using a water bath. The water bath used was a NESLAB RTE Refrigerated Bath/Circulator capable of being operated at a controlled temperature from −25 to +150 ± 0.1°C. Pure L-valine was added to these undersaturated solutions in small steps of 10 mg each and after each addition the solution was allowed to equilibrate for atleast 24 h. When the solids added did not dissolve, the solubility was noted as the concentration of the solution before the final addition. The data obtained was thus within 1 mg/mL of the actual solubility point. Also, solubility of the pure enantiomer and the racemic compound was measured. For these measurements, excess solid was slurried in water for 24 h with constant stirring. A sample of the slurry was withdrawn and filtered with a 0.22 um syringe filter (Millipore Millex Sterile Syringe Filters) to obtain a saturated solution. A sample of known weight of this saturated solution was allowed to evaporate under ambient conditions and the weight of the solid present in the solution was thus obtained to measure the solubility.

Crystallization Experiments from Supersaturated Solutions with Enantiomeric Excess

To conduct crystallization experiments supersaturated solutions of valine in deionized water with an enantiomeric excess were prepared. In each experiment excess amounts of a mixture of DL-valine and L or D-valine was slurried in deionized water in 20 mL scintillation vials. As an example, to prepare a solution having a 50% enantiomeric excess of the L-enantiomer, at a concentration of 85 mg/mL, 425 mg of DL-valine and 425 mg L-valine were added to 10 mL water. The slurries were heated to dissolve and then cooled to room temperature to prepare supersaturated solutions. Gold surfaces with self-assembled chiral monolayers were then immersed in these solutions. Once crystals were observed on the chiral surfaces, they were carefully removed for characterization. Control experiments were also carried out for each experimental condition. Microscope images of the crystals on the chiral SAMs were acquired and the crystals were then removed from the surfaces and ground into a fine powder for characterization with X-ray Powder Diffraction (XRPD). The XRPD patterns of the samples were compared with reference patterns of DL-valine and L or D-valine. To identify which enantiomer was present in excess or was obtained in the pure form optical rotation measurements were carried out using a polarimeter.

RESULTS

Crystallization Experiments from Supersaturated Racemic Solutions

In experiment 1, crystals were observed on the chiral surfaces in all cases. The crystals were characterized with XRPD and it was found that although the pattern was very close to the reference pattern of DL-valine, the characteristic peak of the pure enantiomer (∼29.5° 2θ) was also observed. Based on the results obtained in experiment 1, two more experiments were carried out with lower initial supersaturation. An increase in the relative intensity of the characteristic peak of the pure enantiomer would indicate that decreasing the initial supersaturation promotes chiral resolution. On the other hand a decrease in the relative intensity would indicate that the lower supersaturation inhibits chiral resolution. Table 2 summarizes the relative intensity of the characteristic peak of the pure enantiomer for all experiments. Figure 2 shows the XRPD patterns of the crystals obtained on NALC SAMs in experiments 1, 2, and 3.

Table 2. Summary of Relative Intensity of 29.5° 2θ Peak
 SupersaturationLCNALCLGDPNo SAM
Experiment 10.31.57%1.31%2.23%1.05%0%
Experiment 20.183.16%6.79%2.07%9.54%0%
Experiment 30.046.99%9.77%N/AN/AN/A
Figure 2.

XRPD patterns of the crystals obtained on NALC SAMs in experiments 1, 2, and 3 (from top to bottom).

Ternary Solubility Diagram for the Valine–Water System

Table 3 shows the solubility data obtained for solutions having different enantiomeric excess values using the methods described previously. Figure 3 shows the ternary solubility diagram (enlarged top portion) of D-valine, L-valine, and water at 25°C.

Table 3. Solubility Data for Valine in Water at 25°C
Enantiomeric Excess%L/%DSolubility (mg/mL)
100% (pure enantiomer)100/059
84%92/863
70%85/1567.5
58%79/2172
48%74/2677.5
35%67.5/32.576.5
17%58.5/41.572
0% (racemic)50/5069
Figure 3.

Ternary solubility diagram of D-valine, L-valine, and water at 25°C.

Based on this solubility diagram we carried out experiments at compositions other than 50:50 D/L to see if it is possible to crystallize either of the pure enantiomers on the chiral monolayers.

Crystallization Experiments from Supersaturated Solutions with Enantiomeric Excess

Initially, we carried out a series of experiments starting with supersaturated solutions having varying enantiomeric excess (L-enantiomer) values. The results of these experiments are summarized in Table 4. The solution concentration in each experiment was 85 mg/mL.

Table 4. Results of Crystallization Experiments from Solutions with Initial EE
Initial L-EELCNALCLGDPBulk Solution (Control Expt.)
Initial solution EE (L-enantiomer)
 20%DL + L-solidDL + L-solidDL + L-solidDL + L-solidDL-solid
 40%DL + L-solidDL + L-solidDL + L-solidDL + L-solidDL-solid
 50%DL + L-solidDL + L-solidL-solidDL + L-solidDL + L-solid
 50% (repeat)DL + L-solidL-solidDL + L-solidDL + L-solidDL + L-solid
 60%L-solidL-solidL-solidL-solidL-solid
Initial D-EEDCDCDCBulk Solution (Control Expt.)
Initial solution EE (D-enantiomer)
 50%D-solidD-solidD-solidDL + D-solid
Additional Experiments with Initial D-EEDCBulk Solution (Control Expt.)Initial Solution EE (L-Enantiomer)DCBulk Solution (Control Expt.)
Initial solution EE (D-enantiomer)
 100%D-solidD-solid100%L-solidL-solid
 80%D-solidD-solid80%L-solidL-solid
 60%D-solidD-solid60%L-solidL-solid
 50%D-solidDL + D-solid50%L-solidDL + L-solid
 50% (repeat)DL + D-solidDL + D-solid50% (repeat)DL + L-solidDL + L-solid
 40%DL + D-solidDL-solid40%DL + L-solidDL-solid
 20%DL + D-solidDL-solid20%DL + L-solidDL-solid

Figure 4 shows an image of the pure L-enantiomer crystals obtained on the LG chiral monolayer and on the NALC chiral monolayer in the above experiments. Also shown are the XRPD patterns of the L-enantiomer crystals obtained in these experiments. Next, we carried out an experiment starting with 50% EE solutions, however the D-enantiomer of valine was used to create the EE. Also D-cysteine was used as the chiral SAM and the experiment was repeated thrice. In all three cases, we obtained pure D-enantiomer crystals on the chiral SAM. These results are summarized in Table 4. The solution concentration in each experiment was 85 mg/mL.

Figure 4.

XRPD patterns of L-enantiomer crystals grown on L-G and NALC SAMs and D-enantiomer crystals grown on DC SAM (from top to bottom) and the corresponding crystal images (from left to right).

Figure 4 shows an image of the pure D-enantiomer crystals obtained on the D-cysteine chiral monolayers in the above experiments. Also shown is the XRPD pattern of the D-enantiomer crystals obtained in these experiments.

Finally, a series of experiments were carried out using D-cysteine chiral SAMs and starting with solutions having an enantiomeric excess value ranging from 100% D excess to 100% L excess. Experiments with 50% D and L excess were both repeated twice. The results of these experiments are summarized in Table 4. The solution concentration in each experiment was 85 mg/mL.

DISCUSSION

Crystallization Experiments from Supersaturated Racemic Solutions

From the relative intensity of the characteristic peak of the pure enantiomer for all experiments shown in Table 2, it can be seen that on decreasing the initial supersaturation the relative intensity of the peak increased for crystals grown on LC, NALC, and DP monolayers. However, for crystals grown on LG monolayers the relative intensity decreased slightly. No crystals were obtained on LG and DP monolayers and in the vial without a chiral SAM (control experiment) in experiment 3. Control vials in experiments 1 and 2 yielded racemic crystals. It is important to note that these results are preliminary and further experiments at various conditions will need to be conducted to obtain a statistically significant amount of samples. Also, in our experiments with valine, when starting with racemic solutions, we observed that the D-enantiomer crystallizes in excess on the L-monolayers and vice versa. This observation matches results reported in the literature for the compound histidine.15 The approximate EE value, however, was only 7% in our experiments with valine compared to the ∼27–31% values reported for histidine.15

The results obtained in these experiments indicated that chiral monolayers may be used to promote the crystallization of a pure enantiomer in excess (enantiomeric excess) when crystallizing from a racemic solution. For the three supersaturation levels that were used, the lower values seem to promote chiral resolution more than the higher ones. Further experiments need to be carried out to determine if, for the system being studied, there are any trends in chiral resolution based on the monolayer being used.

Ternary Solubility Diagram for the Valine–Water System

From the data shown in Table 3, as expected for a racemic compound forming system, starting with pure enantiomer (100% EE) the solubility increased as the enantiomeric excess was decreased, reaching a maximum value close to the eutectic composition. After crossing the eutectic composition, the solubility decreased as we reached 0% enantiomeric excess (racemic solution). The eutectic point for the valine–water system has been reported at about 46% EE in the literature.23 In our measurements we obtained a maximum solubility value of 77.5 mg/mL at 48% EE. To confirm whether this value is close to the eutectic value another approach was used. Excess DL-valine and L-valine was slurried in deionized water for ∼48 h with constant agitation. At the end of this period, a sample of the slurry was withdrawn and filtered with a 0.22 um syringe filter (Millipore Millex Sterile Syringe Filters) to obtain a saturated solution. This solution should have an enantiomeric excess corresponding to the eutectic point (∼46% EE lit.). A sample of known weight of this saturated solution was allowed to evaporate under ambient conditions and the weight of the solid present in the solution was thus obtained to measure the solubility. The value obtained was 78 mg/mL which is very close to the maximum solubility (77.5 mg/mL at 48% EE) obtained in our measurements. The 78 mg/mL solubility data point is also included in the solubility diagram (Fig. 3) with the assumption that the EE value at this solubility is 46%.

Crystallization Experiments from Supersaturated Solutions with Enantiomeric Excess

We carried out a series of experiments starting with supersaturated solutions having varying enantiomeric excess values. We wanted to test the hypothesis that by entering the region of the phase diagram where saturated solution exists in equilibrium with a mixture of racemic compound and pure enantiomer, we may be able to crystallize only the pure enantiomer on the chiral SAMs.

As shown in Table 4, when the initial EE of the solution was 20% and 40% excess of the L-enantiomer, a mixture of DL-solid and L-solid was obtained on the SAMs while only DL-solid was obtained in the control experiment in the bulk solution when no SAM was used (indicating that we are in the region of the phase diagram where DL-solids exist in equilibrium with saturated solution). Significantly, when we increased the solution EE to 50%, we were able to obtain pure L-enantiomer on the L-glutathione SAM. In the control experiment a mixture of racemic compound and pure L-enantiomer was obtained (indicating that we are in the region of the phase diagram where DL + L solids exist in equilibrium with saturated solution). When we repeated the experiment, pure L-enantiomer was obtained on the N-acetyl-L-cysteine monolayer while mixtures of racemic compound and pure enantiomer were obtained in all other cases. Finally, on increasing the solution EE to 60% we obtained pure L-enantiomer on all SAMs and also in the control experiment (indicating that we are now in the region of the phase diagram where L-solids exist in equilibrium with saturated solution). Another interesting result was that when we started with solutions having an initial EE, the enantiomer obtained on the chiral SAM in excess or in pure form is the one that was present in excess in the initial solution and not the one having opposite chirality to the monolayer as was the case when starting with racemic solutions, for example, when starting with 50% EE solutions, having the D-enantiomer of valine as the one in excess, we obtained pure D-enantiomer crystals on chiral D-cysteine SAMs.

While carrying out a series of experiments using D-cysteine chiral SAMs and starting with solutions having an enantiomeric excess value ranging from 100% D excess to 100% L excess, we were especially interested in knowing whether we could obtain both pure D and pure L-valine crystals on the D-cysteine monolayers. As shown in Table 4, we were able to crystallize both pure D-enantiomer and pure L-enantiomer crystals on the D-cysteine chiral SAM. The enantiomer that crystallized on the SAM was the one that was initially present in excess. It has been previously reported15 that in the case of crystallization of histidine on chiral SAMs D and L-cysteine, the D-enantiomer crystallizes in excess on L-cysteine monolayers and vice versa when starting with racemic supersaturated solutions. As stated previously, in our experiments with valine, when starting with racemic solutions, similar behavior was observed with the D-enantiomer crystallizing on the L-monolayers in excess and vice versa. When the initial solution was enriched in a particular enantiomer, that enantiomer was obtained in excess (for 20% and 40% initial EE experiments) or in pure form (for 50%, 60%, 80%, and 100% initial EE experiments). Interestingly out of the two experiments that were carried out using 50% EE of each enantiomer in the initial solution, pure enantiomer was only obtained once each. Similarly, in the previous set of experiments (Table 4), pure L-enantiomer was obtained on the L-glutathione monolayer the first time the experiment was carried out and on the N-acetyl-L-cysteine monolayer when the experiments were repeated. Further experiments and studies are needed to determine the reasons behind these behaviors.

CONCLUSIONS

Chiral self-assembled monolayers were used as resolving auxiliaries in the crystallization of valine. We were able to obtain one enantiomer in excess on the crystals obtained on the chiral SAMs when starting with racemic solutions. The enantiomer obtained in excess had the opposite chirality of the monolayer used. On decreasing the supersaturation the relative intensity of the characteristic XRD peak of the pure enantiomer increased, indicating that chiral resolution was enhanced. However, further experiments need to be carried out to obtain a statistically significant number of samples to confirm these trends. In addition crystals of the pure enantiomer were obtained when starting with a solution having an EE value of 50%. Control experiments carried out without chiral SAMs resulted in mixtures of the pure enantiomer and racemic compound. The enantiomer obtained on chiral SAMs at conditions of large enantiomeric excess was the one initially present in excess regardless of the chirality of the monolayer being used. Further experiments need to be carried out to understand the mechanism behind these observations. The results of this work however, indicate the chiral SAM's should be further explored as a potential method for enantiomer resolution.

Ancillary