Enantioselective separation on naturally chiral metal surfaces: D,L-aspartic acid on Cu(3,1,17)(R&S) surfaces.

Homochirality of amino acids, sugars, proteins, and DNA is one of the biochemical hallmarks of life on Earth. Its origins have been debated for decades. Given that minerals such as quartz were probably the first enantiomerically pure materials on Earth, it has been suggested that such materials served as chiral substrates for enantiospecific surface chemistry. Enantiospecific adsorption on the naturally chiral surfaces of minerals has been demonstrated, however, the enantiomeric excesses have been only 1–2 % with a couple of examples suggesting enantiomeric excesses of ca. 10%. 2] Higher enantioselectivities have been observed for adsorption on chiral surfaces of organic crystals. Highly enantioselective adsorption could be of enormous importance, if observed on catalytically active materials such as metals. Herein, we demonstrate that naturally chiral metal surfaces can yield much higher enantioselectivities than minerals and we provide the first definitive proof of enantioselective separation by a naturally chiral metal surface. Chiral surfaces can be created by adsorption of chiral molecules on achiral substrates or by cleavage of crystals with bulk chiral structures. 2, 4] Somewhat counter-intuitively, chiral surfaces can also be generated from achiral materials such as metals. As an example, the two non-superimposable mirror images of the Cu(3,1,17) surface are illustrated in Figure 1. The kinked step structures of these surfaces lack symmetry and, therefore, they are chiral. The first experimental observation of enantioselectivity on naturally chiral surfaces used cyclic voltammetry measurements to show that l-glucose oxidizes more rapidly than d-glucose on the Pt(643) surface, and vice versa on Pt(643). c] Gellman et al. showed that the enantiospecific desorption kinetics of R-3-methylcyclohexanone from chiral Cu(643) surfaces can result in a kinetic separation of 3-methylcyclohexanone. A fundamental challenge in enantioselective surface chemistry and catalysis is to understand the molecular origins of enantioselectivity. The enantiospecific interactions between chiral molecules and chiral surfaces manifest themselves in enantiospecific differences in adsorption energetics and reaction kinetics. Thus, detection of enantiospecificity on chiral surfaces is the first step toward understanding the origins of enantioselectivity. However, such work has been hampered by the difficulty in differentiating two adsorbed enantiomers. Early work made use of different radioisotopes in the two enantiomers. Here, we demonstrate the use of C labeling for enantiodiscrimination of chiral adsorbed species. Enantioselective separation of gas phase d,l-aspartic acid by equilibrium adsorption on naturally chiral Cu(3,1,17) surfaces was demonstrated and quantified by measuring the ratios of the enantiospecific adsorption equilibrium constants, K S K S 1⁄4 K R K R . Measurements of adsorption isotherms (adsorbate coverage as a function of pressure at constant temperature, q(P) j T) can be used to determine adsorption equilibrium constants and adsorption free energies. As illustrated in Figure 2, exposure of a racemic mixture of dand l-enantiomers to the Ror S-enantiomers of a chiral surface results in the adsorption of non-racemic mixtures onto the surfaces. The relative coverages of the enantiomers, qS qS 1⁄4 qR qR , then provide quantitative measurements of the enantiospecific difference in the adsorption free energy: DDG 1⁄4 DGS DGS 1⁄4 DGR DGR . Figure 1. Ball model depiction of the Cu(3,1,17) chiral single-crystal surfaces. The Cu(3,1,17) surfaces have kinked steps formed by (110) and (111) microfacets, separated by (100) terraces. The enantiomorphous surfaces are nonsuperimposable because the three microfacets forming each kink have opposite rotational orientation. The kinked structures with a clockwise and counterclockwise rotation, (111)!(100)!(110), are designated R and S, respectively.