Spin-Based Chiral Separations and the Importance of Molecule–Solvent Interactions

This work uses magneto-electrochemical quartz crystal microbalance methods to study the enantiospecific adsorption of chiral molecules onto a ferromagnetic substrate. The effects of solution conditions, pH, and solvent isotope composition indicate that the kinetics of the enantiomeric adsorption depend strongly on the charge state and geometry of the adsorbate, whereas no thermodynamic contributions to enantiospecificity are found. Density functional theory calculations reveal that an interplay between the adsorbate and solvent molecules is important for defining the observed enantiospecific preference with an applied magnetic field; however, it remains unclear if intermolecular vibrational couplings contribute to the phenomenon.

..8 Figure S12. Possible adsorption configurations and the dipole moments (green arrow) for N-acetyl-Lcysteine methyl ester in H 2 O on Au based on DFT calculations. The red, gray, brown and white spheres indicate the oxygen atoms, nitrogen atoms, carbon atoms, and hydrogen atoms. Note that this is a side view (in direction of a) of Figure 6 in the main manuscript.
..8 Figure S13. Possible adsorption configurations and the dipole moments (green arrow) for N-acetyl-Lcysteine methyl ester in H 2 O on Au based on DFT calculatiosn. The red, gray, brown and white spheres indicate the oxygen atoms, nitrogen atoms, carbon atoms, and hydrogen atoms. Note that this is a top view of Figure 6 in the main manuscript.  Figure S18. Optimized geometries of deprotonated cysteine (-1 H + ) with 4 explicit water molecules. A, B, C, and D have an -NH 3 + group while E, F, G, and H have an -NH 2 group. The most stable structure among all the sampled geometries is A. In terms of structures that have the -NH 2 group, the most stable structure is F. Note that the proton is delocalized in the water ring in structure F. .

Adsorption Isotherm
Cyclic voltammetry and QCM measurements were used to build the data for the adsorption isotherms. The data were collected by scanning from -0.4 V to -1.1 V versus saturated Ag|AgCl at a scan rate = 25 mV/s. Unless specified all measurements were made in a pH 8 phosphate buffer solution. The frequency change from the 30 th cycle, corresponding to the oxidative adsorption of cysteine on a gold substrate, was converted to a change in mass ( ) according to the Sauerbrey equation by a factor of -1.4 ∆ ng / Hz. The large number of cycles was needed to allow the instrument to reach equilibrium and give consistent results. Next, the mass change ( ) or concentration/ was plotted versus concentration and ∆ ∆ fit to a Langmuir -Freundlich isotherm model.

Linearization derivation of isotherm
Assuming that the adsorption is a reversible chemical reaction, the general form of adsorption isotherm can be expressed in a linear form: Thus, the maximum adsorption amount and the adsorption equilibrium constant can be calculated by the slope and intercept value of the linear plot. From the , the Gibbs free energy of the adsorption can be calculated by: where is the concentration of the solvent, which is usually close to the concentration of pure solvent for low concentration solutions.

Kinetic model derivation
Assuming a simple Langmuir model, the desorption rate can be written as where is the concentration of adsorbed cysteine on the surface (filled surface sites or coverage) and is the desorption rate constant. Thus, the rate constant extracted from the exponential fits for the desorption data correspond to . For the case of adsorption, we can write the rate as where is the concentration of cysteine in solution, is the adsorption rate constant, and is the ′ effective adsorption rate constant.
To quantify the rates for the desorption and adsorption processes, the time responses of the QCM frequency were fit to an exponential growth equation for desorption and an exponential decay = • / 1 + 0 equation for adsorption, where A, , and τ 1 and τ 2 were adjusted for a best fit to the = • -/ 2 + 0 0 data. Then the desorption rate constant was calculated as , and the effective adsorption rate

Underpotential deposition (UPD) of lead
Underpotential deposition (UPD) of lead on our substrates were performed using a 5 mM Pb(ClO 4 ) 2 , 10 mM of HClO 4 and 0.1 M of KClO 4 solution. Cyclic voltammograms (left panel) from 0.24 V to -0.45 V versus saturated Ag|AgCl in tandem with the QCM show that the lead adsorbs and gives rise to a cathodic peak at -0.31 V and an anodic peak at around -0.2 V. These redox potentials match well with the reported redox peaks for lead UPD on a Au(111) surface. 1 While the redox peaks using our electrode are broader, associated with other crystalline facets, the most intense, and hence largest contribution, arises from Au (111).  A B Figure S13. Possible adsorption configurations and the dipole moments (green arrow) for N-acetyl-L-cysteine methyl ester in H 2 O on Au based on DFT calculatiosn. The red, gray, brown and white spheres indicate the oxygen atoms, nitrogen atoms, carbon atoms, and hydrogen atoms. Note that this is a top view of Figure 6 in the main manuscript.

The choice of cysteine orientation
We explored a large number of possible geometries for cysteine adsorbed on the Au surface. We began by investigating which functional groups of cysteine are most important in adsorption. We explored geometries with (A) -COO -, (B) -S -, (C) -SH and -COO -, and (D) -SH, -COOH, and -NH2 interacting with the Au surfaces. Besides the default simulation settings mentioned in the manuscript, a polarizable continuum model (PCM) with the dielectric constant of 80 is also applied by the Environ addon of Quantum Espresso. The geometries before and after the optimizations are shown in Figure S14 and the energies for the optimized geometries are listed below the corresponding geometries.

The choice of number of explicit water molecules
To investigate whether the local solvation structure is important or not, explicit water molecules have been added to the simulation cell in the presence of PCM. The structures of adding 3,4, and 5 water molecules before and after the optimization are shown in Figure S15. The adsorption energy of added water molecule is calculated as the reaction energy of the following reaction: where structure has one extra explicit water molecule than . The adsorption energy of adding the 3 rd , S S 4 th , and 5 th water molecule is plotted in Figure S16 (the data point for the 3 rd water molecule is in fact the averaged adsorption energy of all the first three water molecules). Note that adding the 4 th water molecule causes a significant energy drop compared to adding 3 and the 5 th water molecule. Also, from Figure S15 B and C, we find that a local water-ring structure forms when the 4 th water molecule is added, and the structure is effectively unchanged when the 5 th water molecule is added. Consequently, we choose to add 4 explicit water molecules. S10 Figure S15. Optimized geometries of cysteine with explicit water molecules. A, B, and C: cysteine with 3, 4, and 5 explicit water molecules, respectively. Note that with 4 explicit water molecules, a local ring-shape structure forms. This structure effectively unchanges when the 5 th water molecule is added. Figure S16. Adsorption energy of adding another explicit water molecule. The energy of the 3 rd water molecule represents the average adsorption energy of the first three water molecules.

Geometries and energies of sampled structures
The sampled geometries of cysteine with 4 explicit water molecules after the optimization are shown in Figure S17, and the relative energy of each optimized geometry is listed below the corresponding structure. Figure S17. Several optimized geometries of cysteine with 4 explicit water molecules. Structure D is the most stable structure.
The optimized geometries and relative energies of the sampled deprotonated cysteine (-1 H + ) with 4 water molecules are shown in Figure S18. Among all sampled structures in Figure S18, structure A is the most stable one which has the -NH 3 + group. For structures with the -NH 2 group instead, structure F is the most stable one. Figure S18. Optimized geometries of deprotonated cysteine (-1 H + ) with 4 explicit water molecules. A, B, C, and D have an -NH 3 + group while E, F, G, and H have an -NH 2 group. The most stable structure among all the sampled geometries is A. In terms of structures that have the -NH 2 group, the most stable structure is F. Note that the proton is delocalized in the water ring in structure F.