Structure-induced catalysis enhancement of Cu-amino catalysts for rapidly selective oxidation of sulfides in the presence of H₂O₂

Abstract

With benzyl phenyl sulfide as a model substrate and H₂O₂ as an oxidant, high selectivity toward the desired sulfoxide is obtained when using two Cu^II-amino acid complexes as catalysts, while strong stereospecific obstruction in a 1D helical array of complex 2 enables speedy sulfoxidation with excellent conversion.

---

Selective oxidation of sulfides to sulfoxides is a very important reaction that deserves much attention because sulfoxides are important chiral auxiliaries in asymmetric synthesis and valuable pharmaceuticals.^1,2 Avoiding over oxidation of the sulfoxides to their sulfones is also a significant issue for the selective oxidation of sulfides.³ The use of H₂O₂ as a selected oxidant offers many advantages including the fact that it is a cheap and readily available reagent, with water as the only by product.^4,5 The oxidation of prochiral sulfides to sulfoxides has been catalyzed by a range of iron complexes,⁶ titanium complexes,⁷ and vanadyl complexes,^8,9 as well as copper complexes of Schiff base ligands.¹⁰ Most of the metal-based catalysts are discrete complexes with a combination of Schiff base ligands and metal salts.^6,11–15 Enantioselective sulfoxidation has been studied using catalysts of chiral metal–organic frameworks (CMOFs).^16,17

A porous homochiral MOF, namely [Zn₂(bdc)(L-lac)(DMF)]·DMF (L-lac = L-lactic acid, and H₂bdc = 1,4-benzenedicarboxylic acid), showed remarkable catalytic activity in the oxidation of thioethers to sulfoxides using urea hydroperoxide (UHP) or H₂O₂ for the substrates with small substituents. Even higher conversion (up to 100%) and chemoselectivity was observed with 90% H₂O₂ as the oxidant in a mixed solvent system of CH₂Cl₂/CH₃CN. Unfortunately, no or little asymmetric induction was found in the catalytic sulfoxidation reactions due to the fact that the chiral ligand (L-lac) was not effective for chiral induction in the sulfoxidation promoted by the metal ions.^16,18 Meanwhile, coordination polymers (CPs), especially 1D CPs, have seldom been involved in catalysis.^19,20 Very recently, the selective oxidation of methylphenylsulfide with chiral 1D coordination polymers of salen manganese metalloligands and lanthanide knots, with excellent conversion (100%) and good chemoselectivity (88%), was observed using iodosobenzene as the oxidant,²⁰ which is expensive and leads to complicated by-products and related environmental problems.

On the other hand, natural amino acids and their derivatives are promising candidates for the construction of numerous interesting coordination architectures^21,22 and application in heterogeneous catalysis and proton conductivity.^23,24 At this stage, Rosseinsky and co-workers reported an aspartic acid based chiral MOF with a 1,2-bis(4-pyridyl)ethylene (bpe) co-ligand showing a 3D pillared layer structure. The protonated framework by HCl endorses the Brönsted acid catalytic activity for the methanolysis of cis-2,3-epoxybutane, confirming the heterogeneous nature of the catalytic system.²⁵ A serine-based chiral MOF with a 2D grid network was designed and synthesized by Wu and co-workers toward the catalytic performance of 1,2-addition of α,β-unsaturated ketones and the Biginelli reaction.²⁴ In the pioneering work to control the orientation of coordination helices, N-(2-pyridylmethyl)-aspartic acids (2L-pasp and 2D-pasp) with differing α-carbon chiralities have been utilized to assemble with Cu^II salts into infinite single helices within the coordination architecture.²⁶ However, examples of 1D coordination polymers based on such pyridine–amino acid functional reduced Schiff bases have not been documented to date in the aspect of heterogeneous catalysis.

Taking all of these observations into account, we decided to reinvestigate the coordination of helical chains based on the complexation of amino acid ligands with copper(II) ions. A suitable modification to the promising salen and salan ligands in combination with an appropriate copper(II) source may provide the desired results. In this work, two pyridine-amino reduced Schiff base copper complexes, namely {[Cu(2L-pasp)(H₂O)]·3.5H₂O}_n [1, 2L-H₂pasp = N-(2-pyridylmethyl)-L-aspartic acid] and {[Cu(3L-vgly)(OAc)(H₂O)]·2H₂O}_n [2, 3L-vgly = N-(3-pyridylmethyl)-L-valine acid]‡ were synthesized and utilized for the selective oxidation of sulfides under green conditions using aqueous hydrogen peroxide as the oxidant and alcohol as the solvent.

The chiral pyridylmethyl–amino acid derivatives were prepared by the Schiff base condensation of pyridinecarboxaldehyde with the enantiopure amino acid, followed by reduction with NaBH₄, as described elsewhere.²⁷ The complexation of the reduced Schiff base ligands with copper salts (1 : 1 molar ratio) in dioxane/water afforded two blue block shaped crystals of 1 and 2. The products are stable in air and insoluble in water and common organic solvents, and were formulated on the basis of X-ray single-crystal diffraction and elemental analysis (see ESI for details†).

Compound 1 crystallizes in the orthorhombic space group P2₁2₁2 and exhibits a 1D chiral coordination chain structure. The basic coordination unit is [Cu(2L-pasp)(H₂O)] with one copper ion, one 2L-pasp ligand, and one coordinated water molecule, as well as four water guests (Fig. 1a). The Cu^II center is coordinated by three O atoms, two from carboxylate groups and one from lattice water, and two N atoms, one from an amino group and one from a pyridyl group, with a square-pyramidal geometry (τ = 0.16).²⁸ Each 2L-pasp chelates one copper atom through the pyridyl, amino and carboxylate groups and further bridges a neighboring Cu^II ion through another carboxylate to form a 2₁ helical chain with a pitch of 11.249(1) Å along the crystallographic a-axis (Fig. 1b). The coordination assembly is the same as previously reported for the Cu 2L-pasp complex,²⁶ but with a reduced number of lattice water molecules and reduced crystal symmetry. Robust hydrogen-bonding interactions were found to interlink adjacent helical chains into a 3D supramolecular network (Fig. S1†).

Fig. 1 Views of 1 showing the coordination environment of Cu^II with labeling of the asymmetric atoms (a) and the 1D helix (b).

Complex 2 crystallizes in the monoclinic P2₁ space group and has a similar 1D chiral coordination structure to that of 1, but more twisted. The asymmetric unit contains one copper ion, one 3L-vgly ligand, one acetate anion and one coordinated water molecule as well as two water guests (Fig. 2a). Each Cu^II cation coordinates to two carboxyl groups (one from 3L-vgly, the other from acetate), one amine group, one pyridyl group and one aqua ligand, adopting a nearly ideal square-pyramidal geometry (τ = 0.09). Each 3L-vgly ligand bridges two neighboring copper atoms via the pyridyl group and the carboxylate group to extend into 1D polymeric chains running along the crystallographic b-axis (Fig. 2b). In comparison with complex 1, there exist large isopropyl substituents on the opposite sides of the square-pyramids, and the helical pitch [16.263(4) Å] is longer than that of 1 because of the 3-positional pyridyl coordination in 2. Furthermore, hydrogen-bonding between the aqua ligand and carboxylate group of 3L-vgly (O3–H3A⋯O2) joins neighboring helical chains into a 2D network, which is extended to a 3D supramolecular framework by O3–H3B⋯O5 interactions with lattice water included (Fig. 2c).

Fig. 2 (a) The coordination environment of Cu^II with labeling of the asymmetric atoms. (b) View of 2 showing the 1D helix. (c) Hydrogen-bonding network with the inclusion of water guests.

Both complexes are air stable under ambient conditions and thermogravimetric experiments were implemented to investigate their thermal stabilities. The thermogravimetric (TG) curves of 1 and 2 (Fig. S2†) reveal that the first weight loss of 23.1% and 13.2% (calculated: 22.1% for 1 and 14.1% for 2) from room temperature to ca. 135 and 140 °C, respectively, can be ascribed to the stepwise release of guest and coordinated water molecules. There is a short plateau (135–190 °C) in the TG curve of 1, indicating the thermal stability of the activated polymeric coordination molecules. In the case of complex 2, a similar but shorter plateau (140–160 °C) to that in 1 also shows that the remaining active structure is intact with an open Cu^II center. After that, the sharp weight losses of 1 and 2 indicate the decomposition of the remaining structures with comparable trends.

The selective oxidation of sulfides to sulfoxides is one of the most fundamental organic transformations due to the fact that sulfoxides are important intermediates for various pharmaceutically active compounds.³ Two copper complex catalysts were utilized to react with related sulfides with 30% H₂O₂ as the oxidant to test their catalytic ability. Initially, benzyl phenyl sulfide was chosen as a model substrate to optimize the reaction conditions. The effects of different reduced Schiff base catalysts, reaction time and solvent media were investigated in an attempt to optimize the sulfoxidation conditions. The catalytic performances of copper complex catalysts are listed in Table 1. To select the proper solvent, oxidation with 30% hydrogen peroxide as the oxidant was studied in a variety of organic solvents such as methanol, ethanol, acetonitrile, toluene and dichloromethane. Complex 1 shows modest catalysis performance either in CH₃CN or C₂H₅OH (Table 1, entries 1 and 2), and longer reaction time leads to higher conversion of sulfide (Table 1, entry 6). Complex 2 exhibits excellent catalytic performance in a short time using CH₃OH, C₂H₅OH or CH₃CN as the solvent (Table 1, entries 3–5). As an environmental friendly solvent, water was also taken into account (Table 1, entries 7 and 12). The result indicated that pure water is not an effective solvent for the catalysis, nevertheless the mixed H₂O/C₂H₅OH solvents result in good conversion and selectivity. Interestingly, pure non-polar solvents such as toluene and dichloromethane caused rather low conversion (Table 1, entries 10 and 11). Meanwhile, incorporation of ethanol in the non-polar solvents increases the conversion of benzyl phenyl sulfide in the fast catalysis procedure (Table 1, entries 13–15).

Table 1 Selective oxidation of benzyl phenyl sulfide to the corresponding sulfoxide with different Cu-based catalysts and various solvent systems^a

Entry	Solvent	Catalyst	Time (h)	Conversion^b (%)	Selectivity^b (%)	TON^c
a Reaction conditions: sulfide (1.0 mmol), catalyst (5 mol%), solvent (10 mL), 30% H2O2 (3.0 mmol), 30 °C.b Conversion (%) and selectivity (%) were determined by HPLC.c TON, turn over number.
1	CH3CN	1	0.5	25	>99	5
2	C2H5OH	1	0.5	24	>99	5
3	CH3CN	2	0.5	98	>99	20
4	CH3OH	2	0.5	98	>99	20
5	C2H5OH	2	0.5	99	>99	20
6	CH3CN	1	8	86	>99	17
7	H2O	2	6	13	98	3
8	THF	2	6	18	>99	4
9	EtOAc	2	6	4	97	1
10	CH2Cl2	2	6	5	>99	1
11	Toluene	2	6	1	>99	0
12	H2O/C2H5OH (1 : 1)	2	1	99	98	20
13	CCl4/C2H5OH (1 : 1)	2	1	55	99	11
14	THF/C2H5OH (1 : 1)	2	1	53	99	11
15	CH2Cl2/C2H5OH (1 : 1)	2	1	87	99	17
16	C2H5OH	Blank	24	8	>99	2
17	C2H5OH	Cu(NO3)2·2.5H2O	24	71	>99	14
18	C2H5OH	Cu(OAc)2·H2O	24	20	>99	4
19	C2H5OH	3L-vgly + Cu(OAc)2·H2O	24	35	>99	7

---

Consequently, the optimum conditions were established as catalyst 2 (5 mol%), 30% H₂O₂ (3.0 mmol) and ethanol (10 mL) at 30 °C by varying the reaction factors (see ESI, Table S2–5†), if enantioselectivity was not taken into account. With the proper reaction conditions established, the substrate scope was extended to a series of structurally diverse sulfides such as methylphenyl sulfide, 4-nitrophenylmethyl sulfide, 4-methoxyphenylmethyl sulfide, dibutyl sulfide, diphenyl sulfide, and the sulfide intermediate of omeprazole (Table 2). The S-enantiomer of the omeprazole esomeprazole is a proton-pump inhibitor for the treatment of gastroesophageal reflux disease, and it was successfully synthesized using chemical or biological oxidations.³ Noticeably, it turns out that either small alkyl or bulky diphenyl groups lead to modest conversion for sulfides (Table 2, entries 1–5), and electron rich thioether substrates result in lower selectivity. Meanwhile, semirigid groups with flexible methylene provide decent conversion and selectivity (entries 6 and 7 in Table 2), which is favorable for the selective synthesis of omeprazole. The better catalytic properties of 2 than those of 1 may be ascribed to the inherent nature of complex 2, and its suitable supramolecular window (Fig. 2c) available for shape selection.

Table 2 Substrate scope of the sulfoxidation^a

Substrate	Product	Conversion^b (%)	Selectivity^b (%)	TON^c
a Reaction conditions: sulfide (1.0 mmol), catalyst 2 (5 mol%), ethanol (10 mL), 30% H2O2 (3 equiv.), 30 °C, 0.5 h.b Conversion (%) and selectivity (%) were determined by HPLC.c TON, turn over number.
		47	91	9
		46	92	9
		65	86	13
		78	80	16
		50	>99	10
		>99	>99	20
		88	99	18

---

The stability and recyclability of catalyst 2 were confirmed by XRD and SEM studies (Fig. 3 and 4). It is notable that 2 presented modest stability in acetonitrile (better than ethanol), being reused for 4 cycles without significant loss in its activity. A simple filtration testified that the reaction is mainly a heterogeneous catalysis with a slight homogeneous component (Fig. 5). Despite the good conversion and chemoselectivity, very low asymmetric induction was observed in the catalytic sulfoxidation (around 20%, Table S2–S5†). This phenomenon may be attributed to the inadequate steric effect of the chiral centers. In comparison, recently reported manganese and lanthanide 1D chiral coordination polymers of a salen ligand²⁰ lead to better conversion but lower chemoselectivity for the sulfoxidation of methylphenyl sulfide and 4-nitrophenylmethyl sulfide. Although the lanthanide complex contains bulk substituent groups, the enantioselectivity is relatively low for the sulfoxidation (up to 25%).

Fig. 3 XRD patterns of complex 2 before and after catalysis.

Fig. 4 SEM images of complex 2 before and after catalysis.

Fig. 5 Filtration experiment for complex 2. Conversions and selectivities are given as a function of time.

For the catalysts in this work, the Cu^II centers adopt a similar square-pyramidal coordination geometry and dissimilar environments. However, each of them has a terminal aqua ligand, which was assumed to activate equivalent H₂O₂ oxidant and avoid the over oxidation of sulfide, leading to excellent chemoselectivity for all reactions. Nevertheless, complex 2 has remarkable catalytic activity with size and shape selectivity, and high conversion in the fast sulfoxidation of thioethers bearing rigid groups with flexible methylene.

In summary, two copper coordination polymers based on pyridylmethyl amino acids have been synthesized and structurally characterized. They exhibit dissimilar coordination environments of the Cu^II centers but similar 1D helical coordination assemblies. Their catalytic activity in sulfoxidation was evaluated. The reaction parameters, such as the reaction time and solvent, were varied to improve the conversion and selectivity for benzyl phenyl sulfide. This method represents the first example of reduced Schiff amino complexes catalyzing the sulfoxidation under environmentally friendly and mild conditions (with H₂O₂ as the oxidant). Both of the copper complexes exhibit catalytic activity with high conversion and chemoselectivity. Complex 2 promoted rapid catalysis and was favorable to the chemoselective synthesis of omeprazole. The investigation of the enantioselective synthesis of prochiral sulfoxides is underway to screen coordination polymers based on pyridylmethyl amino acids in oxidation processes with environmentally friendly oxidants.

Acknowledgements

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (21201026 and 21302014), the Qing Lan Project and a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), the Jiangsu Key Laboratory of Advanced Catalytic Materials and Technology (BM2012110).

Notes and references

1. M. C. Carreno, G. Hernandez-Torres, M. Ribagorda and A. Urbano, Chem. Commun., 2009, 6129–6144.
2. G. E. O’Mahony, A. Ford and A. R. Maguire, J. Sulfur Chem., 2013, 34, 301–341.
3. E. Wojaczynska and J. Wojaczynski, Chem. Rev., 2010, 110, 4303–4356.
4. H. Srour, P. Le Maux, S. Chevance and G. Simonneaux, Coord. Chem. Rev., 2013, 257, 3030–3050.
5. G. La Sorella, L. Sperni, G. Strukul and A. Scarso, Adv. Synth. Catal., 2016 DOI:10.1002/adsc.201600430 , ahead of print.
6. H. Egami and T. Katsuki, J. Am. Chem. Soc., 2007, 129, 8940–8941.
7. L. Postigo, M. Ventura, T. Cuenca, G. Jimenez and B. Royo, Catal. Sci. Technol., 2015, 5, 320–324.
8. C. Drago, L. Caggiano and R. F. W. Jackson, Angew. Chem., Int. Ed., 2005, 44, 7221–7223.
9. A. Lazar, P. Sharma and A. P. Singh, Microporous Mesoporous Mater., 2013, 170, 331–339.
10. G. E. O’Mahony, A. Ford and A. R. Maguire, J. Org. Chem., 2012, 77, 3288–3296.
11. P. K. Bera, N. Gupta, S. H. R. Abdi, N. H. Khan, R. I. Kureshy and H. C. Bajaj, RSC Adv., 2015, 5, 47732–47739.
12. Z.-Z. Li, S.-Y. Yao, J.-J. Wu and B.-H. Ye, Chem. Commun., 2014, 50, 5644–5647.
13. W. Dai, G. Li, L. Wang, B. Chen, S. Shang, Y. Lv and S. Gao, RSC Adv., 2014, 4, 46545–46554.
14. P. K. Bera, P. Kumari, S. H. R. Abdi, N.-U. H. Khan, R. I. Kureshy, P. S. Subramanian and H. C. Bajaj, RSC Adv., 2014, 4, 61550–61556.
15. P. Kelly, S. E. Lawrence and A. R. Maguire, Synlett, 2007, 1501–1506.
16. D. N. Dybtsev, A. L. Nuzhdin, H. Chun, K. P. Bryliakov, E. P. Talsi, V. P. Fedin and K. Kim, Angew. Chem., Int. Ed., 2006, 45, 916–920.
17. W. Xuan, C. Ye, M. Zhang, Z. Chen and Y. Cui, Chem. Sci., 2013, 4, 3154–3159.
18. A. L. Nuzhdin, D. N. Dybtsev, V. P. Fedin and G. A. Bukhtiyarova, Dalton Trans., 2009, 10481–10485.
19. W. L. Leong and J. J. Vittal, Chem. Rev., 2011, 111, 688–764.
20. M. Yadav, A. Bhunia, S. K. Jana and P. W. Roesky, Inorg. Chem., 2016, 55, 2701–2708.
21. X. L. Yang and C. D. Wu, CrystEngComm, 2014, 16, 4907–4918.
22. E. V. Anokhina, Y. B. Go, Y. Lee, T. Vogt and A. J. Jacobson, J. Am. Chem. Soc., 2006, 128, 9957–9962.
23. S. C. Sahoo, T. Kundu and R. Banerjee, J. Am. Chem. Soc., 2011, 133, 17950–17958.
24. M. Wang, M.-H. Xie, C.-D. Wu and Y.-G. Wang, Chem. Commun., 2009, 2396–2398.
25. M. J. Ingleson, J. P. Barrio, J. Bacsa, C. Dickinson, H. Park and M. J. Rosseinsky, Chem. Commun., 2008, 1287–1289.
26. S.-P. Wu and C.-H. Lee, CrystEngComm, 2009, 11, 219–222.
27. Q.-Q. Zhang, Z.-H. Zhang, B.-H. Qu, Q. Chen and M.-Y. He, Inorg. Chim. Acta, 2014, 418, 59–65.
28. G. A. Senchyk, A. B. Lysenko, H. Krautscheid, E. B. Rusanov, A. N. Chernega, K. W. Kraemer, S.-X. Liu, S. Decurtins and K. V. Domasevitch, Inorg. Chem., 2013, 52, 863–872.

---

Footnotes

† Electronic supplementary information (ESI) available: Crystallographic data, TG curves, extra experimental procedures, and extra figures and tables including NMR data of sulfoxides and catalytic oxidations. CCDC 1491826 and 1491827. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra22393a

‡ Crystal data for 1: C₂₀H₃₈Cu₂N₄O₁₇, M_r = 733.62, orthorhombic, space group P2₁2₁2 (no. 18), a = 11.2493(11), b = 16.5313(17), c = 8.4611(8) Å, V = 1573.5(3) ų, Z = 2, D_c = 1.548 Mg m⁻³, Flack parameter, 0.029(5); μ = 1.429 mm⁻¹, R(int) = 0.0207; goodness-of-fit on F² = 1.072; R₁ = 0.0219, wR₂ = 0.0645. Crystal data for 2: C₁₃H₂₄CuN₂O₇, M_r = 383.88, monoclinic, space group P2₁ (no. 4), a = 7.2507(16), b = 16.263(4), c = 7.3756(16) Å, V = 868.0(3) ų, Z = 2, D_c = 1.469 Mg m⁻³, Flack parameter, 0.038(8); μ = 1.293 mm⁻¹, R(int) = 0.0173; goodness-of-fit on F² = 0.844; R₁ = 0.0210, wR₂ = 0.0616.

---