# Fileset

[2024_ApplClaySci_Packing material for HPLC（MDR）.pdf](https://mdr.nims.go.jp/filesets/6ceb64b3-2806-4c2e-b865-be25c83802bf/download)

## Creator

Akihiko Yamagishi, [Kenji Tamura](https://orcid.org/0000-0001-6578-0923), Masumi Kamon, Jun Yoshida, Hisako Sato

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Use of an Ion-exchange Adduct of Synthetic Hectorite and Chiral Copper(II) Complex as a Packing Material for Chromatographic Resolution](https://mdr.nims.go.jp/datasets/41d8bf31-25f7-42b5-8af9-62cd2f78c955)

## Fulltext

1  Use of an Ion-exchange Adduct of Synthetic 1 Hectorite and Chiral Copper(II) Complex as a 2 Packing Material for Chromatographic  3 Resolution 4   5 Akihiko Yamagishi,a Kenji Tamura,*b Masumi Kamon,b Jun Yoshida c and Hisako Sato*d  6  7 a Toho University, School of Medicine, Ohta-ku, Tokyo 143-8540, Japan  8 b National Institute for Materials Science, Environmental Circulation Composite 9 Materials Group, Tsukuba, 305-0044, Japan 10 c Nihon University, Department of Chemistry, College of Humanities & Sciences, 11 Setagaya-ku, Tokyo 156-8550, Japan 12 d Ehime University, Department of Chemistry, College of Science, Matsuyama, Ehime 13 790-8577, Japan,  14  15  16  17 *Corresponding author: Kenji Tamura (TAMURA.Kenji@nims.go.jp) 18 Tel: +81-29-860-4370, Fax: +81-29-860-4667 19  20  21 2  Abstract 22      A spherically shaped particle of synthetic hectorite (denoted as Na-HEC) was ion-23 exchanged with a divalent Cu(II) complex, [Cu(SS-oxa)]2+ (SS-oxa = SS-2,2′-24 isopropylidene-bis(4-phenyl-2-oxazoline)). The material is denoted as [Cu(SS-25 oxa)]2+/HEC. A column for high performance liquid chromatography (HPLC) was 26 prepared by packing 4.0 g of [Cu(SS-oxa)]2+/HEC into a stainless tube (25 cm x 0.4 cm 27 (i.d.)). When tris(acetylacetonato)cobalt(III) (denoted as [Co(acac)3]) was eluted by 28 methanol at the flow rate of 0.2 mLmin-1 and 4 oC, the compound was separated to ∆- and 29 Λ-enantiomers nearly to the baseline separation. Organic molecules with two hydroxyl 30 groups such as 1,1’-binaphthyl-2,2’-diol were also resolved partially. The results 31 promised the practical utility of the clay column chromatography for obtaining 32 enantiomeric compounds. 33  34  35  36  37  38 Keywords: Ion-exchange adducts; synthetic hectorite; chiral copper(II) complexes; high 39 performance liquid chromatograph; optical resolution 40  41   42 3  1. Introduction 43 A smectite clay mineral exhibits unique characters when it adsorbs cationic species 44 (Ogawa and Kuroda, 1995; Shichi and Takagi, 2000; Schoonhedyt, 2014). As a two-45 dimensional adsorbent, the clay mineral binds a cationic species either on the external 46 surfaces or in the interlayer. A layer is charged negatively due to the isomorphous 47 substitution of Si(IV) or Al(III) by lower valent metal ions. A negatively charged sites 48 exists at high surface density (c. a. one negative charge (e-) per 100 nm2). Moreover, the 49 external surface of a layer is characterized by the two-dimensional periodicity of 50 phyllosilicate networks. Owing to these characters, an adsorbed molecule exhibits the 51 following unique features: (i) it takes uniform orientation with respect to a layer surface 52 and (ii) it is arranged under two-dimensional regularity.  53      From the viewpoint of molecular recognition, the above features ((i) and (ii)) lead 54 to the hypothesis that a pre-adsorbed molecule interacts with an approaching molecule, 55 recognizing its shape in cooperative manners. In case that a pre-adsorbed molecule is 56 chiral, it may interact in stereoselective manners (Yoshida et al., 2020a, 2020b; Sato et al. 57 2018a, 2018b; Yamagishi and Sato, 2012; Kotkar and Ghosh, 1987). Prompted by this 58 view, we developed clay column chromatography for optical resolution (Yamagishi et al., 59 2022, 2021, 2012, 1996, 1992; Yamagishi, 1986,1985,1981; Yamagishi and Ohnishi, 60 1982; Kakegawa and Yamagishi, 2006; Okada et al., 2018). In the attempts, an ion-61 exchange adduct of a smectite clay mineral and a chiral cationic molecule was prepared. 62 The substance was used as a packing material of a column in liquid chromatography. 63 When a chiral molecule is eluted through the column, it may interact with the pre-64 adsorbed chiral metal complex stereoselectively. The interactions affect the rate of elution, 65 4  depending on the chirality relation between pre-adsorbed and eluted molecules. 66 Accordingly the opposite enantiomers are eluted at different elution volumes to achieve 67 optical resolution. As a successful example, the column packed with an ion-exchange 68 adduct of ∆- or Λ-[Ru(phen)3]2+ (phen = 1, 10-pehnanthroline) and a synthetic hectorite 69 was proved to resolve a wide scope of inorganic and organic chiral molecules (Yamagishi 70 and Sato, 2012). Since a smectite clay mineral itself is achiral, a pre-adsorbed chiral 71 molecule (e. g. ∆- or Λ-[Ru(phen)3]2+) is denoted as a “chiral director”. The column is 72 stable at high temperature (> 30 oC) for a long time and exhibits high separation capacity 73 in comparison to the known commercially available chiral columns. The mechanism of 74 chiral recognition was revealed by vibrational circular dichroism spectroscopy (Sato et 75 al., 2022, 2018a, 2018b). 76      In the present work, a column was developed using a chiral copper(II) as a chiral 77 director. The used complex was [Cu(SS-oxa)]2+ (SS-oxa = 2,2′-isopropylidene-bis(4-78 phenyl-2-oxazoline)). The complex is known to act as a homogeneous asymmetric 79 catalyst for various organic reactions (Johnson et al., 2000). Moreover, there is a report 80 that the complex is immobilized in a smectite clay mineral and employed as a 81 heterogeneous asymmetric catalyst (Fraile et al., 2009; Sato, et al., 2021). An adduct was 82 prepared by ion-exchanging synthetic hectorite with [Cu(SS-oxa)]2+. Using the column 83 packed with the ion-exchange adduct, optical resolution was attempted by eluting a 84 racemic mixture of tris(acetylacetonato)cobalt(III) (denoted as [Co(acac)3]) by methanol. 85 As a result, a racemic mixture of [Co(acac)3] was separated to ∆- and Λ-enantiomers 86 nearly completely. The mechanism of resolution was investigated by the theoretical 87 simulations. The same column resolved partially an organic molecule with two hydroxyl 88 5  groups such as 1,1’-binaphthyl-2,2’-diol (denoted as BINAL). The results promised the 89 practical utility of the present column for obtaining enantiomeric inorganic and organic 90 compounds chromatographically. 91  92 2. Experimental 93 2.1. Materials: S,S-2,2'-(dimethylmethylene)bis(4-phenyl-2-oxazoline) (denoted as S,S-94 oxa), copper(II)trifluoromethanesulfonate (denoted as [Cu(TFMS)2]), dibenzoylmethane 95 (denoted as dbmH), racemic tris(acetyacetonato)metal(III) (denoted as [M(acac)3] : M = 96 Co, Cr, Ru, Ir and Rh), 1,1’-binaphthyl-2, 2’-diol (denoted as BINAL), R- and S-97 phenylethane-1, 2-diol, racemic 1-phenylethanol, racemic 1-phenylethyamine and SS- 98 and RR-1, 2-diphenylethane-1, 2-diols were purchased from Tokyo Kasei Co. Ltd. 99 (Japan) and used as received. 1,1’-binaphthyl-2, 2’-dibromomethoxy (denoted as BINAL-100 Br) was synthesized in our laboratory. As shown in Fig. S1, the purity of the compound 101 was confirmed by being eluted on a chiral column (RU-1 Ceramosphere, Shiseido Co. 102 Ltd. Japan). Used methanol was of reagent (Tokyo Kasei Co. Ltd. (Japan)). 103 Synthetic hectorite (Laponite RD; denoted as Na-HEC) was purchased from BYK 104 Additives & Instruments (U. K.). Its elemental composition and cation exchange capacity 105 (CEC) are stated to be (Na0.37Ca0.01)[(Mg2.80Li0.19)(Si3.96O10)(OH)2] and 0.65 milli-106 equivalents per gram, respectively. The spherical particle of Na-HEC (c.a. 5 µm in 107 averaged diameter) was prepared by the spray-drying method as described previously 108 (Sakuma et al., 2018). The specific surface area and pore volume for the spherical particle 109 were estimated by nitrogen adsorption−desorption isotherms (Fig. S2, Table. S1). 110 2.2 Instrument: The prepared clay particles were characterized using XRD (Rigaku 111 6  Ultima IV) with Cu Kα radiation (λ = 0.15418 nm) under conditions of 40 kV and 30 mA, 112 and at scanning rate of 2 °/min. Surfaces of the clay particles were observed by scanning 113 electron microscopy (Quanta 600 SEM, FEI). The surface areas of the samples were also 114 estimated from nitrogen adsorption measurements (BELSORP-max, MicrotracBEL 115 Corp.). UV-visible electronic spectra were recorded using a UV–vis spectrophotometer 116 (U-2810, Hitachi Ltd., Japan). Electronic circular dichroism (ECD) spectra were 117 measured using a polarimeter (J-720, JASCO Corporation, Japan). High performance 118 liquid chromatography (HPLC) was performed using a Gulliver HPLC system (JASCO, 119 Japan). For checking the purity of a used compound, a commercial column 120 (Ceramosphere RU-1 (SHISEIDO, Japan)) was used.  121 2.3 Preparation of a chiral clay column: [Cu(SS-oxa)](TFMS)2 was prepared by mixing 122 equimolar amounts of copper(II) trifluoromethanesulfonate (or  [Cu(TFMS)2]) and SS-123 2,2’-isopropylidene-bis(4-phenyl-2-oxazoline) (or SS-oxa) in methanol. [Cu(SS-124 oxa)(dbm)](TFMS) was prepared by mixing equimolar amounts of [Cu(SS-oxa)](TFMS) 125 and dbmH in methanol. The preparation and identification of [Cu(SS-oxa)(dbm)](TFMS) 126 was described elsewhere (Sato et al., 2020). The progresses of these reactions were 127 monitored by the change of UV-vis spectra. 128 For preparing a column, 3.3 g of the spherical particles of Na-HEC was dispersed in 129 50 mL of methanol. A methanol solution containing 1.9 × 10-3 mole of [Cu(SS-130 oxa)](TFMS)2 was added slowly under stirring. The clay particle was coloured blue, 131 indicating that [Cu(SS-oxa)]2+ was ion-exchanged with Na+ ion in Na-HEC spontaneously. 132 The resultant material was denoted as [Cu(SS-oxa)]2+/HEC. After filtering the suspension, 133 the solid precipitate or [Cu(SS-oxa)]2+/HEC  (4.0 g) was dried under air. 4.0g of [Cu(SS-134 7  oxa)]2+/HEC was packed into a stainless tube (25 cm × 4 mm (i.d.)). XRD patterns and 135 SEM images were measured before and after the ion-exchanging of Na-HEC particles 136 with [Cu(SS-oxa)]2+ (Figs. S3 and S4). The dead volume (Vd) of the column was 137 determined to be 1.67 mL by eluting 10 µL of chloroform. In the similar way, a column 138 of [Cu(SS-oxa)(dbm)]+/HEC (5 cm × 0.4 cm (i.d.)) was prepared. Its Vd was determined 139 to be 0.52 mL. 140  141  142 Chart 1. Molecular structures of (a) [Cu(SS-oxa)]2+, (b) [Cu(SS-oxa)(dbm)]+ , (c) Λ-143 [Co(acac)3] and (d) ∆-[Co(acac)3].  144  145 8  2.4 Computational methods: The association structure between [Cu(SS-oxa)]2+ and 146 [Ru(acac)3] was investigated by the theoretical simulation. The method was based on the 147 dynamic DFT calculation. The same approach has been taken previously in several 148 examples to study the details of intermolecular interaction of a resolved molecule and a 149 chiral director fixed on a solid surface (Sato et al., 2020; Sato et al., 2021; Yamagishi et 150 al., 2021; Yamagishi et al., 2022). The calculation was made for the association model 151 between [Cu(SS-oxa)]2+ and ∆- or Λ-[Co(acac)3]. Gaussian 16 program (C.02) was used 152 (Frisch et al., 2019). Geometry optimization was performed at the DFT (Density 153 Functional Theory) at the level of 6-31G (d,p.) for all atoms. For an intercalation model, 154 the upper and lower clay layers were replaced with a SiO42- anion. At the initial 155 optimization, the anion was fixed, and the other was positionally optimized. After 156 convergence, the positions of all the molecules were freely moved for optimization. 157 Finally geometry optimization was performed at the DFT-d3 (Density Functional Theory 158 and empirical dispersion) at the level of 6-31G (d,p.) for all atoms. 159  160 Results and Discussion 161 3.1. Resolution of racemic tris(acetylacetonato)cobalt(III): For examining the 162 resolution capability of the prepared [Cu(SS-oxa)]2+/HEC column, racemic 163 tris(acetylacetonato)cobalt(III) (denoted as [Co(acac)3]) was eluted by methanol. 164 [Co(acac)3] is neutral and possesses a nearly spherical shape. Chirality arises from the 165 helical twisting of three acetyacetonato (denoted by acac-) ligands around a central Co(III) 166 ion (Chart 1). Its helicity is denoted as ∆∆∆ or (simply as ∆) or ΛΛΛ (simply as Λ), 167 9  depending on the twisting direction. Since the molecule possesses no functional group on 168 the outer surface, it is difficult to be optically resolved by use of resolving reagents such 169 as camphor or tartaric acid. The enantiomer is stable against racemization at room 170 temperature. In case that the molecule is resolved on a chiral column, it indicates that 171 there exists a microscopic cavity in a column material to recognize the asymmetric 172 structure of an eluted molecule. 173 Curve (A) in Figure 1 shows the chromatogram when [Co(acac)3] (1 × 10-8 mole) 174 was eluted at 4℃ and the flow rate of 0.2 mLmin-1. Two peaks were obtained nearly on 175 the baseline separation. Collecting an eluted solvent at the first and second peaks, their 176 electronic circular dichroism (ECD) spectra were measured. It was concluded that the 177 first and second bands corresponded to the ∆- and Λ-enantiomers of [Co(acac)3], 178 respectively. The elution order was also confirmed by flowing ∆- and Λ-enantiomers 179 separately. The separation factor (α) was given by 180 α = (V2-Vd)/(V1-Vd)     (1) 181 in which V1, V2 and Vd were the elution volumes of the less and more retained peaks and 182 the dead volume (1.67 mL), respectively. The value of α was calculated to be 1.11. The 183 value was comparable to those of commercially available columns (e.g. IC Chiralpak (25 184 cm × 4.6 mm (i.d.)) (Daicel Chemical Ind, Ltd. Japan) (Shen and Okamoto, 2016).  185 10   186 Fig. 1. Chromatogram when [Co(acac)3] (1 × 10-8 mole)) was eluted at (A) 4, (B) 15 and 187 (C) 30 ℃. The samples were injected as a volume of 50 µL. Flow rate was 0.2 188 mLmin-1. The elution was monitored by electronic absorbance at 300 nm. 189  190      Separation behavior was examined by changing elution conditions such as 191 temperature, flow rates and injected amounts. Curves (A(, (B) and (C) in Figure 1 192 compare the chromatograms when the temperature changed from 4, 15 and 30 ℃, 193 respectively. On raising temperature, elution volumes decreased with the increase of 194 11  overlapping of two peaks. A corrected elution volume, (Vi - Vd) (i = 1 and 2), is related 195 to the thermodynamic parameters of the adsorption equilibrium as below: 196 In (Vi -Vd) = ∆H/RT - ∆S/R, （2）         197 in which R, ∆H and ∆S are gas constant, the enthalpy and entropy changes of adsorption, 198 respectively. By plotting In (Vi -Vd) against 1/T (Fig. S5), ∆H and ∆S are calculated as 199 shown in Table 1. According to the results, Λ-[Co(acac)3] is adsorbed by [Cu(SS-200 oxa)]2+/HEC stronger than ∆-[Co(acac)3] with the larger enthalpy and smaller  entropy 201 changes of adsorption.  202  203 Table 1. Thermodynamic parameters of adsorption of ∆- or Λ-[Co(acac)3] on a column 204 of [Cu(SS-oxa)]2+/HEC. 205  ∆−[Co(acac)3] Λ-[Co(acac)3] difference ∆H kJmol-1 17.27 17.40 0.13 ∆S JK-1mol-1 - 44.02 - 43.78 - 0.24  206 Figures 2 (A), (B) and (C) show the chromatograms when [Co(acac)3] (1 8 × 10-8 207 mole) was eluted at the flow rates of 0.1, 0.2 and 0.3 mLmin-1, respectively. The samples 208 were injected as a volume of 50 µL. On raising a flow rate, two peaks were more 209 overlapped until they were fused nearly to one broad band. According to the results in 210 Figure 2 (C), the time required for attaining adsorption equilibration (denoted as tp) was 211 larger than the time for an injected molecule to pass through the column at the flow rate 212 of 0.3 mLmin-1. From the dead volume of the column (1.65 mL) and the flow rate value, 213 tp was estimated to be larger than 5 minutes.  214 12   215  216 Fig. 2. Chromatogram when [Co(acac)3] was eluted at 30 ℃ at the flow rate of 217 (A) 0.1, (B) 0.2 and (C) 0.3 mLmin-1. Elution was monitored by the 218 electronic absorbance at 300 nm. 219  220 Figures 3 (A), (B) and (C) show the chromatograms when [Co(acac)3] was eluted 221 at the injected amounts of 1.0 × 10-7, 1.0 × 10-6 and 1.6 × 10-6 mole, respectively. The 222 flow rate was 0.2 mLmin-1 and temperature 30 oC. An injected volume was 50 µL. On 223 13  raising the injected amount, two peaks were overlapped until they were fused to one broad 224 band. Thus the maximum amount of mounted [Co(acac)3] (denoted by Mmax) was 1.0 × 225 10-7 mole for two peaks to appear separately in the chromatogram. The amount of pre-226 adsorbed [Cu(SS-axo)]2+ complexes was 1.5 × 10-3 mole. This was much larger than Mmax. 227 Thus only a small part of pre-adsorbed [Cu(SS-axo)]2+ complexes were effective for 228 resolution. Probably most of [Cu(SS-axo)]2+ complexes were present within the inner 229 parts of HEC particles so that they were unable to be contact with eluted [Co(acac)3]. It 230 will be important to identify the location of adsorbed [Cu(SS-oxa)]]2+ within a clay 231 material (HEC). It is proposed to apply the solid state NMR to discriminate between the 232 Cu(II) complexes on an external surface and in the interlayer space. The chemical shift 233 of peaks assigned to C13 atoms in [Cu(SS-oxa)]2+ is expected to give the information on 234 its location. 235 14   236 Fig. 3. Chromatogram when [Co(acac)3] was eluted when the injected amount was 237 changed from (A) 1.0 × 10-7, (B) 1.0 × 10-6 and (C) 1.6 × 10-6 mole.  The flow rate 238 was 0.2 mL min-1 and the temperature was 30 °C. Elution was monitored by 239 electronic absorbance at 300 nm. 240  241      Chromatographic behavior was compared for a series of tris(acetylacetonato) 242 metal(III) complexes (or [M(acac)3]: M = Cr, Ir, Ru and Rh). The complexes were eluted 243 at 4 ℃ and a flow rate of 0.2 mLmin-1. Figures 4 (A), (B), (C) and (D) show the 244 15  chromatograms of [Cr(acac)3], [Ir(acac)3], [Ru(acac)3] and [Rh(acac)3], respectively. The 245 separation factor (α) is given in Table 2. From the table, α depended crucially on the kind 246 of a central metal ion. [Co(acac)3] (Figure 1 (A)) was resolved most efficiently. [Ir(acac)3], 247 [Ru(acac)3] and [Rh(acac)3] were partially resolved. [Cr(acac)3] was not resolved to the 248 detectable level by means of ECD measurements. From the ECD spectra of the fractions 249 before and after the peak positions, elution order was from ∆ to Λ for [M(acac)3] (M = 250 Co, Ir, Ru and Rh). Table 1 includes the ionic radius of a three-valent central metal ion 251 (or M(III)). Based on these, [Co(acac)3] is the smallest in size. This might be the reason 252 that [Co(acac)3] was resolved at highest separation factor. Alternatively it is suggested 253 that there exists a cavity around [Cu(SS-axo)], whose size matches with [Co(acac)3]. 254 Other complexes were difficult to be resolved because they were too bulky to enter the 255 cavity completely. [Cr(acac)3] was not resolved in spite of its size smaller than that of 256 [M(acac)3] (M = Ir, Ru and Rh). One reason might lie in the high spin multiplicity of 257 Cr(III) (or S = 3/2). Since Cu(II) has a spin of 1/2, it is suspected that the magnetic 258 repulsion between Cr(III) and Cu(II) inhibited the occurrence of effective chiral 259 discrimination. 260  261 16   262 Fig. 4. Chromatogram when [M(acac)3] was at the flow rate of 0.2 mLmin-1 and 263 temperature 4 oC. The injected amount was 1.0 × 10-7 for all runs. Elution was 264 17  monitored by electronic absorbance at 330 nm, (A) [Ir(acac)3]), 320 nm; (B) 265 [Ru(acac)3], 270 nm; Rh(acac)3], 320 nm; [Cr(acac)3], 330 nm, respectively.  266  267 Table 2. Chromatographic results for a series of [M(acac)3]. The elution conditions are 268 stated in the caption of Figure 4.** 269 Complex V1 -Vd (mL) V2 -Vd (mL) Separation factor Ionic radius (M3+) (pm) * [Ir(acac)3] 5.03 5.38 1.07 82 [Ru(acac)3] 6.19 6.6 1.07 82 [Rh(acac)3] 6.33 6.58 1.04 81 [Cr(acac)3] 4.73 4.73 1 76 *Cited from Shannon R. D. (1976) Act. Crystallogr. A32, 751. 270 ** An error in estimating the elution volume was 5%. 271  272      In order to obtain a support for the existence of a microscopic cavity in [Cu(SS-273 axo)]+2/HEC, the chromatographic behavior of [Co(acac)3] was compared using a column 274 packed with [Cu(SS-axo)(dbm)]+/HEC (5 cm x 0.4 cm (i.d.)). In this column, a vacant 275 space around Cu(II) ion was already occupied by dbm-. On this column, [Co(acac)3] was 276 eluted nearly at the volume close to Vd (dead volume: 0.35 mL) (not shown). The results 277 supported that an empty space around a Cu(II) ion in [Cu(SS-axo)]2+/HEC acted as a 278 microscopic cavity for recognizing the chirality of an eluted molecule.  279 Resolution of chiral organic molecules: In order to extend the scope of targets, a number 280 of chiral organic molecules were eluted on the present column. There have been several 281 18  columns developed for resolving organic molecules chromatographically (Wang et al., 282 2023; Hoyas et al., 2021; Wahab et al., 2017; Shen and Okamoto, 2016). For the purpose, S-, 283 R- and racemic 1,1’-bibaphthyl-2,2’-diols (denoted as R-, S- and rac-BINAL, 284 respectively) were eluted. BINAL possesses two helically twisted phenyl groups (Chart 285 2). The molecule was anticipated to interact with the two phenyl groups of [Cu(SS-oxa)]2+ 286 through π−π interactions. Figures 5 (A), (B) and (C) show the chromatograms when 287 racemic, R- and S-BINAL (1 x 10-8 mol) were eluted at 30 oC and a flow rate of 0.2 288 mLmin-1, respectively. In case of (A), two peaks appeared. They were partially 289 overlapped. The ECD spectra of the solutions collected at the first and second peaks 290 showed that they contained R-BINAL and S-BINAL as an excess enantiomer, 291 respectively. The results were consistent when R- and S-enantiomers were eluted 292 separately ((B) and (C) in Figure 5, respectively).  293 19   294 Chart 2. Molecular structures of (A) BINAL, (B) BINAL-Br, (C) 1-phenylethane-1, 2-295 diol, (D) 1-phenylethanol, (E) 1-phenylethyamine and (F) 1, 2-diphenylethane-296 1, 2-diol. 297  298 20   299 Fig. 5. Chromatograms when (A) racemic BINAL, (B) R-BINAL and (C) S-BINAL were 300 eluted at 30 oC and a flow rate of 0.2 mLmin-1. The injected amount was 1×10-8 301 mole. The elution was monitored by electronic absorbance at 270 nm.  302  303      For comparison, 1,1’-bibaphthyl-2,2’-dibromomethoxy (BINAL-Br) (Chart 2) was 304 eluted under the same conditions. The chromatogram showed one single band nearly at 305 21  the dead volume (or at 1.80 mL) (not shown). The ECD spectra of the solutions collected 306 at the former and latter parts of the band showed no ECD activity. Thus BINAL-Br 307 showed no affinity towards the column. It was suspected that the presence of OH groups 308 were necessary for a molecule to be resolved by the column. One possibility was that the 309 two OH groups in BINAL coordinated weakly a Cu(II) ion in [Cu(SS-oxa)]2+, while no 310 such interaction was possible in case of BINAL-Br.  311      Prompted by the above hypotheses, other compounds with one or two OH groups 312 were eluted. The results are summarized in Table 3. As a result, 1-phenylethane-1,2-diol 313 was resolved partially, validating the hypothesis. When 1-phenylethylamine was mounted, 314 it was not eluted at all. It indicated that the amino group in the molecule coordinated 315 strongly with a Cu(II) ion in the column. 316  317 Table 3. Chromatographic results for organic compounds. The elution conditions were 318 the same as in Figure 5. 319 Compound V1 -Vd (mL) V2 -Vd (mL) separation factor 1,1’-bibaphthyl-2,2’-diol 0.53 0.93 1.76 1,1’-bibaphthyl-2,2’-dibromomethoxy 0.02 0.02 1 1-phenylethyl-1,2-diol 1.40 (S) 1.73 (R) 1.24 1-phenylethanol 0.5 0.5 1 1-phenylethylamine Not eluted Not eluted   1, 2-diphenylethane-1, 2-diol 1.17 (SS) 1.25 (RR) 1.07 (*) An error in estimating the elution volume was 5%. 320 22   321 Theoretical simulation for the binding of [Co(acac)3] on the column: the binding of ∆- 322 or Λ-[Co(acac)3] on the column packed with [Cu(SS-oxa)]+/HEC was studied 323 theoretically.  The simulated system contained ∆- or Λ-[Co(acac)3], [Cu(SS-oxa)]+ and 324 one SiO42- anion. Here, SiO42- was included in the place of a clay layer to maintain the 325 condition of charge neutrality.  326      Figures 7 (A) and (B) show the optimum structures when the ∆- and Λ-enantiomers 327 of [Co(acac)3] are associated with [Cu(SS-oxa)]2+ in the presence of SO42-. Both of the 328 enantiomers are located close to a central Cu(II) ion. The Co(III) – Cu(II) distance is 0.32 329 and 0.31 nm, respectively. Two phenyl groups in [Cu(SS-oxa)]2+ orient so as to surround 330 the enantiomers. One remarkable difference is that the angle between the phenyl groups 331 for a pair of ∆-[Co(acac)3]/[Cu(SS-oxa)]2+is large (< 150 degree) for the ∆-enantiomer, 332 while it is small (> 110 degree) for the Λ-enantiomer. This implies that the steric 333 hindrance is larger for a pair of ∆-[Co(acac)3]/[Cu(SS-oxa)]2+ than for a pair of Λ-334 [Co(acac)3]/[Cu(SS-oxa)]2+. This leads to the lower stabilization energy for the former 335 pair than for the latter pair. In fact, the stabilization energy in Λ-[Co(acac)3]/[Cu(SS-336 oxa)]2+ is calculated to be lower by 15.7 kJmol-1 than that in ∆-[Co(acac)3]/[Cu(SS-337 oxa)]2+.Thus the calculation predicts that the Λ-enantiomer binds with [Cu(SS-oxa)]2+ 338 stronger than the ∆-enantiomer. This is in accord with the experimental results. The 339 calculation implied that the stabilization was achieved by the “induced-fit” of the two 340 phenyl moieties in [Cu(SS-oxa)]2+ when Λ-[Co(acac)3] entered the cavity around a Cu(II) 341 ion. No such “induced-fit” took place in case of ∆-[Co(acac)3]. 342 23  (A) 343  344 (B) 345  346 Fig. 6. The optimum structures of enantiomeric [Co(acac)3] associated with [ Cu(SS-347 oxa)]2+ in the presence of SO42-; (A) ∆-[Co(acac)3]/[Cu(SS-oxa)]2+ and (B) Λ-348 [Co(acac)3]/[Cu(SS-oxa)]2+. 349  350 4. Conclusions  351 Chromatographic resolution was performed by use of a column packed with an ion-352 exchange adduct of synthetic hectorite and a chiral Cu(II) complex. The used chiral 353 24  complex was [Cu(SS-oxa)]2+ (SS-oxa=SS-2,2′-isopropylidene-bis(4-phenyl-2-oxazoline)). 354 The column resolved a racemic mixture of [Co(acac)3] (acacH = acetylacetone) to ∆- and 355 Λ-enantiomers nearly to the baseline separation. With the help of theoretical simulation, 356 it was concluded that the resolution was realized by the occupation of the enantiomers in 357 a cavity around a Cu(II) ion. As for organic compounds, the column exhibited the 358 resolution ability toward an organic molecule with two hydroxyl groups. It was suggested 359 that the molecule binds with a Cu(II) ion in a stereoselective way through coordinating 360 interactions.  361  362 Author Contributions 363 All authors contributed equally.  364  365 Conflicts of interest 366 There are no conflicts to declare.  367  368 Acknowledgements 369 This work was financially supported by the Japan Society for the Promotion of 370 Science (JSPS) KAKENHI Grant (JP22H02033, JP22K0526 and JP20K21090). The 371 computations were performed in collaboration with the Research Centre for 372 Computational Science, Okazaki, Japan (Project:23-IMS-C160). 373  374 25  References 375 Fraile, J. M., García, J. I., Mayoral, J. A., 2009. Noncovalent immobilization 376 of enantioselective catalyst. Chem. Rev. 109, 360-417.  377 https://doi.org/10.1021/cr800363y 378 Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, 379 J.R., Scalmani, G., Barone, V., Petersson, G.A., Nakatsuji, H., Li, X., Caricato, M., 380 Marenich, A.V., Bloino, J., Janesko, B.G., Gomperts, R., Mennucci, B., Hratchian, 381 H. P., Ortiz, J.V., Izmaylov, A.F., Sonnenberg, J.L., Williams-Young, D., Ding, F., 382 Lipparini, F., Egidi, F., Goings, J., Peng, B., Petrone, A., Henderson, T., 383 Ranasinghe, D., Zakrzewski, V.G., Gao, J., Rega, N., Zheng, G., Liang, W., Hada, 384 M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., 385 Honda, Y., Kitao, O., Nakai, H., Vreven, T., Throssell, K., Montgomery, J.A.J., 386 Peralta, J.E., Ogliaro, F., Bearpark, M.J., Heyd, J.J., Brothers, E.N., Kudin, K.N., 387 Staroverov, V.N., Keith, T.A., Kobayashi, R., Normand, J., Raghavachari, K., 388 Rendell, A.P., Burant, J. C., Iyengar, S.S., Tomasi, J., Cossi, M., Millam, J.M., 389 Klene, M., Adamo, C., Cammi, R., Ochterski, J.W., Martin, R.L., Morokuma, K., 390 Farkas, O., Foresman, J.B., Fox, D.J., 2019. Gaussian 16, Revision C.02. 391 Hoyas, S., Roscioni, O. M., Tonneaux, C., Gerbaux, P., Cornil, J., Muccioli, L., 2021. 392 Peptoids as a chiral stationary phase for liquid chromatography: Insights from 393 molecular dynamics simulations. Biomacromolecules, 22, 2573−2581. 394 https://doi.org/10.1021/acs.biomac.1c00302 395 Johnson. J. S., Evans. D. A., 2000. Chiral bis(oxazoline) copper(II) complexes: 396 Versatile catalysts for enantioselective cycloaddition, aldol, michael, and 397 https://doi.org/10.1021/acs.biomac.1c0030226  carbonylene reactions. Acc. Chem. Res., 2000, 33, 325–335. 398 https://doi.org/10.1021/ar960062n 399 Kakegawa, N., Yamagishi, A., 2005. Coadsorption studies of tris(1,10-phenanthroline) 400 ruthenium(II) and N-methylated alkaloid cation by laponite with an application 401 for a chiral column packing material. Chem. Mater., 17, 2997-3003.  402 https://doi.org/10.1021/cm048285s 403 Kotkar, D. Ghosh, D. P., 1987. Effect of pH on the chromatographic resolution of 404 carboxy derivatives of tris(bipyridyl)ruthenium(II) on a chirally modified 405 montmorillonite column. Inorg. Chem. 26, 208-209. 406 https://doi.org/10.1021/ic00248a041 407 Ogawa, M., Kuroda, K., 1995. Photofunctions of intercalation compounds. Chem. Rev. 408 95, 399-438. https://doi.org/10.1021/cr00034a005 409 Okada, T., Mutsuki, O., Tajima, K., Yamakami, T., Hisako Sato, H., 2018. Variation in 410 thickness of a layered silicate on spherical silica particles affected HPLC chiral 411 chromatographic resolution. Appl. Clay Sci., 163, 72-80.  412 https://doi.org/10.1016/j.clay.2018.07.017 413 Sakuma, K., Tamura, K., Minagawa, K., 2018. “Doughnut”-like clay microparticles 414 fabricated using a hybrid method of spray drying and centrifugal disc atomization. 415 Chem. Lett., 47, 68-70. https://doi.org/10.1246/cl.170891 416 Sato, H., Tamura, K., Takimoto, K., Yamagishi, A., 2018a. Solid state vibrational 417 circular dichroism towards molecular recognition: Chiral metal complexes 418 intercalated in a clay mineral. Phys. Chem. Chem. Phys. 20, 3141-3147. 419 https://doi.org/10.1039/c7cp05114j 420 https://pubs.acs.org/doi/10.1021/cr00034a00527  Sato, H. Takimoto, K., Mori, H., Yamagishi, A., 2018b. Stereoselective interactions as 421 manifested by vibrational circular dichroism spectra: Interplay between chiral 422 metal complexes Co-adsorbed in a montmorillonite clay. Phys. Chem. Chem. 423 Phys. 20, 25421-25427. https://doi.org/10.1039/c8cp04753g 424 Sato, H. Takimoto, K., Yoshida, J., Yamagishi, A., 2020. Vibrational circular dichroism 425 towards asymmetric catalysis: Chiral induction in substrates coordinated with 426 copper(II) Ions. Phys. Chem. Chem. Phys. 22, 24393-24398. 427 https://doi.org/10.1039/d0cp04827e 428 Sato, H., Takimoto, K., Yoshida, J., Watanabe, Y., Yamagishi, A., 2021. Solid-state 429 vibrational circular dichroism as applied for heterogenous asymmetric catalysis: 430 Copper (II) complexes immobilized in montmorillonite. Chem. Lett. 50, 896-898. 431 https://doi.org/10.1246/cl.200937 432 Sato, H., Takimoto, K., Yamagishi, A., Yoshida, J., Hara, M., 2022. Vibrational circular 433 dichroism spectroscopy toward intercalation compounds of sodium 434 montmorillonite: evidences for molecular packing of enantiopure monovalent 435 Ir(III) complexes within interlayer spaces. Appl. Clay Sci., 228, 106621. 436 https://doi.org/10.1016/j.clay.2022.106621 437 Schoonheydt, R. A., 2014. Functional hybrid clay mineral films. Appl. Clay Sci. 96, 9-438 21. https://doi.org/10.1016/j.clay.2014.04.010  439 Shannon, R. D., 1976. Revised effective ionic radii and systematic studies of 440 interatomic distances in halides and chalogenides. Acta Cryst. A32, 751-767.  441 https://doi.org/10.1107/S0567739476001551 442 Shen, H., Okamoto, Y., 2016. Efficient separation of enantiomers using stereoregular 443 https://www.sciencedirect.com/science/article/pii/S0169131714001252#!https://doi.org/10.1107/S056773947600155128  chiral polymers. Chem. Rev., 116, 1094−1138. 444 https://doi.org/10.1021/acs.chemrev.5b00317  445 Shichi, T., Takagi, K., 2000. Clay minerals as photochemical reaction fields. J. 446 Photochem. Photobiol. C: Photochem. Rev. 1, 113-130. 447 https://doi.org/10.1016/S1389-5567(00)00008-3 448 Yamagishi, A. Soma, M.,1981. Optical resolution of metal chelates by use of adsorption 449 on a colloidal clay. J. Am. Chem. Soc., 103, 4640-4642. 450 https://doi.org/10.1021/ja00405a086 451 Yamagishi, A. Ohnishi, R. 1982. Clay column chromatography for optical resolution: 452 Initial resolutions of bis(acetylacetonato)(glycinato)cobalt(III) and 453 (acetylacetonato)bis(glycinato)cobalt(III) on a Δ-tris(1,10-phenanthroline) 454 nickel(II)-montmorillonite column. Inorg. Chem. 21, 4233-4236. 455 https://doi.org/10.1021/ic00142a024 456 Yamagishi, A. 1985. Chromatographic resolution of enantiomers having 457 aromatic groups by an optically active clay-chelate adduct. J. Am. Chem. 458 Soc. 107, 732-734. https://doi.org/10.1021/ja00289a051 459 Yamagishi, A.,1986. Optical resolution and racemization reaction of 460 tris(bathophenanthrolinedisulfonato)iron(II): Absence of an intramolecular 461 racemization path in aqueous solution. Inorg. Chem. 25, 55-57. 462 https://doi.org/10.1021/ic00221a015 463 Yamagishi, A., Makino, H., Nakamura, Y., Sato, H., 1992. Separation of isomers of 464 cobalt(III) complexes by liquid chromatography on a column packed with a clay, 465 https://doi.org/https://doi.org/10.1021/ja00405a08629  ruthenium(II) complex adduct.Clay. Clay Miner. 40, 359-361. 466 https://doi.org/10.1346/CCMN.1992.0400318 467 Yamagishi, A., Taniguchi, M., Imamura, Y., Sato, H., 1996. Clay column 468 chromatography for optical resolution: Selectivities of L-[Ru(phen)3]2+ and L- 469 [Ru(bpy)3]2+ Laponite columns toward 1,1'-binaphthol. Appl. Clay Sci. 11, 1-10. 470 https://doi.org/10.1016/0169-1317(96)00010-5 471 Yamagishi, A., Sato, H., 2012. Stereochemistry and molecular recognition on the 472 surface of a smectite clay mineral. Clay. Clay Miner. 60, 411-419. 473 https://doi.org/10.1346/CCMN.2012.0600407 474 Yamagishi, A. Takimoto, K., Tamura, K., Sato, F., Sato, H., 2021. Chiral discrimination 475 of dansylated alanine methyl ester on a modified clay surface: vibrational circular 476 dichroism approach. Bull. Chem. Soc. Jpn., 94, 2711-2717. 477 https://doi.org/10.1246/bcsj.20210291 478 Yamagishi, A., Yamamoto, S., Takimoto, K., Tamura, K., Kamon, M., Sato, F., Sato, 479 H., 2022. Clay column chromatography for optical resolution: a series of 480 derivatized amino acids. Bull. Chem. Soc. Jpn., 95, 961–967. 481 https://doi.org/10.1246/bcsj.20220077 482 Yoshida, J., Tateyama, K., Kasahara, Y., Yuge, H., 2020a. Stabilization of oxidized 483 ruthenium complexes by adsorption on clay minerals. App. Clay Sci. 199, 1, 484 105869. https://doi.org/10.1016/j.clay.2020.105869 485 Yoshida, J., Tateyama, K., Yuge, H., Hara, M., 2020b. Hexagonal ordering of 486 racemic Ni(II) complexes in the interlayer space of a clay mineral. Chem. 487 Commun. 56, 10670-10673. https://doi.org/10.1039/D0CC04977H.  488 https://doi.org/10.1016/0169-1317(96)00010-530  Wahab, M. F. Patel, D. C., Wimalasinghe, R. M., Armstrong, D. W., 2017. 489 Fundamental and practical insights on the packing of modern high efficiency 490 analytical and capillary column. Anal. Chem. 89, 8177−8191. 491 https://doi.org/10.1021/acs.analchem.7b00931 492 Wang, F., Li, X., Jiang, S., Han, J., Wu, J., Yan, M., Yao, Z., 2023. Enantioselective 493 behaviors of chiral pesticides and enantiomeric signatures in foods and the 494 environment. J. Agric. Food Chem. 71, 12372−12389. 495 https://doi.org/10.1021/acs.jafc.3c02564 496  497  498  499  500  501  502  503  504  505  506  507  508  509  510  511 https://doi.org/10.1021/acs.jafc.3c0256431  Supporting Information 512 Use of an Ion-exchange Adduct of Synthetic Hectorite and Chiral 513 Copper(II) Complex as a Packing Material for Chromatographic  514 Resolution 515   516 Akihiko Yamagishi,a Kenji Tamura,*b Masumi Kamon,b Jun Yoshida c and Hisako Sato* 517 d  518  519 a Toho University, School of Medicine, Ohta-ku, Tokyo 143-8540, Japan  520 b National Institute for Materials Science, Environmental Circulation Composite 521 Materials Group, Tsukuba, 305-0044, Japan 522 c Nihon University, Department of Chemistry, College of Humanities & Sciences, 523 Setagaya-ku, Tokyo 156-8550, Japan 524 d Ehime University, Department of Chemistry, College of Science, Matsuyama, Ehime 525 790-8577, Japan,  526  527  528 *Corresponding author: Kenji Tamura (TAMURA.Kenji@nims.go.jp) 529 Tel: +81-29-860-4370, Fax: +81-29-860-4667 530  531 32  Contents 532  533 Figure S1. The electronic circular dichroism spectra (ECD) of the eluted fractions of 1,1’-534 binaphthyl-2, 2’-dibromomethoxy (denoted as BINAL-Br).  535  536 Figure S2. Curves of nitrogen adsorption−desorption isotherms for Laponite particles.  537  538 Table. S1 BET surface area (As) and the total pore volume (Vp) of Laponite particles. 539  540 Figure S3. XRD patterns of (a) Laponite particles (or Na-HEC) and (b) [Cu(SS-541 oxa)]2+/Laponite particles (or [Cu(SS-oxa)]2+/HEC).  542  543 Figure S4. SEM images of particles for (a)/(b) Na-HEC and (c)/(d) [Cu(SS-oxa)]2+/HEC. 544  545 Figure S5. The Arrhenius plots of the natural logarithm of elution volume versus the 546 reciprocal of temperature. 547  548   549 33   550 Fig. S1.  The electronic circular dichroism spectra (ECD) of 1,1’-binaphthyl-2, 2’-551 dibromomethoxy (denoted as BINAL-Br). BINAL-Br was synthesized in our laboratory. 552 The purity of the compound was checked by eluting on a chiral column (RU-1 553 Ceramosphere (Shiseido Ind. Co.)). Two separate peaks were obtained. The ECD spectra 554 of the first and second peaks are shown below. The mirror image relation was obtained 555 to confirm their purity. 556  557  558 Figure S1. The ECD spectra of the first (black) and second fractions (red) of eluted 559 BINAL-Br. The first and second fractions were corresponded to R- and S-560 enantiomers, respectively.  561  562  563 -150-100-50050100150350 400 450 500 550 600 650 700θ / mdegWavelength / nm34   564 Fig S2. Curves of nitrogen adsorption−desorption isotherms for Laponite particles: (open 565 circles) adsorption profile and (solid circles) desorption profile. Based on the results, 566 Table. S1 lists the BET surface area (As,BET) and pore volume (Vp). 567  568  569 Figure S2. Curves of nitrogen adsorption−desorption isotherms for Laponite particles: 570 (open circles) adsorption profile and (solid circles) desorption profile.  571  572 Table S1. BET surface area (As) and the total pore 573 volume (Vp) of Laponite particles. 574  575 A s，BET 320 (m2 g-1)Vp 0.14 (cm3 g-1)35   576 Fig S3. XRD patterns of (a) Laponite particles (or Na-HEC) and (b) [Cu(SS-577 oxa)]2+/Laponite particles (or [Cu(SS-oxa)]2+/HEC), respectively. The weak peak 578 observed around 2θ~60.8° corresponds to the (060) reflection with a d-spacing of 0.15 579 nm, indicating trioctahedral smectite. The basal spacing, d(001), was determined to be (a) 580 1.3 nm and (b) 1.43 nm, respectively. From the results, the basal spacing expanded by c.a. 581 0.13 nm. Moreover the sharpening of the peak (001) implied that the improvement of 582 ordering of layer stacking on the adsorption of [Cu(SS-oxa)]2+ ions.  583  584  585 Figure S3. XRD patterns of (a) Laponite particles (or Na-HEC) and (b) [Cu(SS-586 oxa)]2+/Laponite particles (or [Cu(SS-oxa)]2+/HEC), respectively.  587  588 36   589 Fig S4. SEM images of particles for (a)/(b) Na-HEC and (c)/(d) [Cu(SS-oxa)]2+/HEC, 590 respectively. The images indicated that the particle maintained a spherical shape after 591 adsorbing [Cu(SS-oxa)]2+.  592  593  594  595 Figure S4. SEM images of particles for (a)/(b) Na-HEC and (c)/(d) [Cu(SS-596 oxa)]2+/HEC. 597  598  599 37  Fig S5. The Arrhenius plots of the natural logarithm of the elution volume versus the 600 reciprocal of temperature. The plot gives the thermodynamic parameters of adsorption 601 equilibrium as below: 602 In (Vi -Vd) = ΔH/RT - ΔS/R        603 in which R, ΔH and ΔS are gas constant, the enthalpy and entropy changes of adsorption, 604 respectively. 605  606 1.51.61.71.81.922.12.22.30.0033 0.00335 0.0034 0.00345 0.0035 0.00355 0.0036 0.00365LnP1/T / K1.51.61.71.81.922.12.22.30.0033 0.00335 0.0034 0.00345 0.0035 0.00355 0.0036 0.00365LnP1/T / K  607 Figure S5. The Arrhenius plots of the natural logarithm of elution volume versus the 608 reciprocal of temperature: filled black circles (peak 1) and red triangles (peak 2). The data 609 are taken from the temperature dependence of the elution chromatogram of [Co(acac)3] 610 (Figure 1 in the text). 611  Kakegawa, N., Yamagishi, A., 2005. Coadsorption studies of tris(1,10-phenanthroline)　ruthenium(II) and N-methylated alkaloid cation by laponite with an application for a chiral column packing material. Chem. Mater., 17, 2997-3003.　 https://doi.org/10.... Kotkar, D. Ghosh, D. P., 1987. Effect of pH on the chromatographic resolution of carboxy derivatives of tris(bipyridyl)ruthenium(II) on a chirally modified montmorillonite column. Inorg. Chem. 26, 208-209. https://doi.org/10.1021/ic00248a041 Ogawa, M., Kuroda, K., 1995. Photofunctions of intercalation compounds. Chem. Rev. 95, 399-438. https://doi.org/10.1021/cr00034a005 Okada, T., Mutsuki, O., Tajima, K., Yamakami, T., Hisako Sato, H., 2018. Variation in thickness of a layered silicate on spherical silica particles affected HPLC chiral chromatographic resolution. Appl. Clay Sci., 163, 72-80.　 https://doi.org/10.1016/... Sakuma, K., Tamura, K., Minagawa, K., 2018. “Doughnut”-like clay microparticles fabricated using a hybrid method of spray drying and centrifugal disc atomization. Chem. Lett., 47, 68-70. https://doi.org/10.1246/cl.170891 Shichi, T., Takagi, K., 2000. Clay minerals as photochemical reaction fields. J. Photochem. Photobiol. C: Photochem. Rev. 1, 113-130. https://doi.org/10.1016/S1389-5567(00)00008-3 Yamagishi, A. Soma, M.,1981. Optical resolution of metal chelates by use of adsorption on a colloidal clay. J. Am. Chem. Soc., 103, 4640-4642. https://doi.org/10.1021/ja00405a086 Yamagishi, A. Ohnishi, R. 1982. Clay column chromatography for optical resolution: Initial resolutions of bis(acetylacetonato)(glycinato)cobalt(III) and (acetylacetonato)bis(glycinato)cobalt(III) on a Δ-tris(1,10-phenanthroline) nickel(II)-montmorillo... Yamagishi, A.,1986. Optical resolution and racemization reaction of tris(bathophenanthrolinedisulfonato)iron(II): Absence of an intramolecular racemization path in aqueous solution. Inorg. Chem. 25, 55-57. https://doi.org/10.1021/ic00221a015 Yamagishi, A., Makino, H., Nakamura, Y., Sato, H., 1992. Separation of isomers of cobalt(III) complexes by liquid chromatography on a column packed with a clay, ruthenium(II) complex adduct.Clay. Clay Miner. 40, 359-361. https://doi.org/10.1346/CCMN.1...