A bis-salicylaldoximato-copper(II) receptor for selective sulfate uptake

Abstract

The preferential uptake of sulfate over the industrially important anions, chloride and nitrate and structurally similar dihydrogen phosphate has been achieved in aqueous media with a dicopper salicylaldoxime complex.

---

Binding and extraction of oxoanions such as sulfate remains a challenging task yet is of upmost importance due to the biological, environmental and commercial relevance. Specific binding can be accomplished with highly pre-organised molecules such as indolocarbazole oligomers,¹ macrocyclic or macrobicyclic ligands.² However, complex formation is often slow, which has disadvantages in circulation industrial processes such as solvent extraction. Flexible, open chain ligands³ or metallocyclic supramolecules⁴ proffer a suitable alternative. The later have additional advantages; the host structure is formed in situ and benefits from additional improvements in strength and selectivity by direct metal–host binding.⁵ Ligands capable of existing in a zwitterionic form offer further opportunities,⁶ since controlled binding of both the cation and attendant anion can be triggered by a change in the pH of the solution.⁷

The ligand L, N,N′-dimethyl-N,N′-hexamethylenedi-(3-hydroxyiminomethyl-2-hydroxy-5-tert-butylbenzylamine), consisting of two 5-tert-butylsalicylaldoxime molecules, forms a helical twisted 2 ∶ 2 metallo-macrocycle upon Cu(II)-complexation, which can encapsulate anions (Fig. 1).⁸ In this work, we show that the extraction of the hydrophilic sulfate anion from aqueous solution into a water-immiscible organic phase is possible using this receptor; even in the presence of high concentrations of industrially relevant anions such as chloride and nitrate, which, based on their solvation energies should be preferentially extracted. In addition, we show unequivocally that anion shape is not a factor in this anti-Hofmeister behaviour as the structurally similar yet more hydrophobic dihydrogenphosphate anion is not captured in preference to sulfate within this system.

Fig. 1 Anion encapsulated Cu₂L₂ metallo-macrocycle.

In order to probe sulfate uptake over chloride, nitrate or phosphate, two sets of liquid–liquid extraction experiments were undertaken. The first was chosen to match the industrially relevant, Bulong circuit,⁹e.g. an aqueous solution containing sodium sulfate and sodium chloride was vigorously stirred for 24 h with an equal volume of chloroform containing the “copper only” complex [Cu₂(L-2H)₂].⁸ The intramolecular hydrogen bonding present around the copper centre within these pseudo-macrocyclic systems imparts improved stability to these complexes to the degree that ESMS analysis in this experiment gave a representative indication of the organic solution composition. Thus, anion uptake within the core could be conveniently analysed by ESMS with a strong isotopic peak pattern at 660 (m/z) indicative of the presence of the dicationic complex species [SO₄⊂Cu₂L₂]²⁺. In a second set of experiments a large 100-fold excess of chloride, nitrate or phosphate was used with parallel results, i.e. the presence of the dicationic complex species was again observed in the ESMS in all cases with no observation of the alternative anion within the core.

The similarity in the molecular mass and the isotopic distribution of sulfate and dihydrogen phosphate prevents their discrimination via low resolution ESI-MS. However, support for the preferred uptake of sulfate over dihydrogen phosphate is given by a carefully calibrated HR-MS study. The observed signal of the +2 fragment at 665.2882 m/z (lowest-mass) and the isotopic pattern is in agreement to the theoretical model of [SO₄⊂Cu₂L₂]²⁺ (C₆₄H₁₀₀N₈O₈Cu₂SO₄)²⁺m/z = 665.2886 (Fig. 2), with no indication of a peak associated with encapsulated dihydrogen phosphate species at ∼+0.01.

Fig. 2 Experimental high resolution ESI-MS isotope pattern for the dicationic fragment [SO₄Cu₂L₂]²⁺ (top) and the expected theoretical isotope pattern (bottom).

UV-vis titration experiments of the “copper only” complex [Cu₂(L–2H)₂]‡ in a 1 ∶ 1 mixture of isopropanol/1,2-dichloroethane were undertaken with nitrate and the halide anions to underline the observed stronger affinity towards sulfate. The calculated stability constants§ show the expected trend with smaller log K values for both nitrate and the halides as compared with the recently reported values for the tetrahedral anions sulfate and dihydrogen phosphate.⁸ It is worth noting that based on the complex stability data for sulfate and phosphate, a preference in binding these anions is not expected (Table 1).

Table 1 Calculated stability constants for anions binding with [Cu₂(L–2H)₂]

Anion	log K
a Ref. 8. b No anion encapsulation.
H2PO4^−	4.53 ± 0.03^a
SO4^2^−	4.43 ± 0.15^a
NO3^−	4.38 ± 0.06
I^−	4.01 ± 0.05
Br ^−	3.54 ± 0.08
Cl ^−	—^b

---

To emphasise the preference of the receptor to bind sulfate over either chloride, nitrate or phosphate, L in methanol was mixed in a 1 ∶ 1 ∶ 4 ratio of CuSO₄ and either NaCl, NaNO₃ in water or in a 1 ∶ 1 ∶ 2 ratio with CuSO₄ and K₂HPO₄ in water. Crystallisation of the resulting bulk materials resulted in the isolation of the mixed anion complexes [SO₄⊂Cu₂L₂]Cl₂, [SO₄⊂Cu₂L₂](NO₃)₂ and [SO₄⊂Cu₂L₂](H₂PO₄)₂.¶ An addition experiment utilising a mixture of three potential anions SO₄²⁻, NO₃⁻ and HPO₄²⁻ also resulted in the formation of [SO₄⊂Cu₂L₂](NO₃)₂ as evidenced by a space group check which matched exactly the data obtained for the above sulfate/nitrate complex. Each of the structures exhibit the 2 ∶ 2 metal to ligand helical assembly common to these lengthy bis-salicylaldoxime ligands (Fig. 3). In all cases the four tertiary amines on the complex are protonated as evidenced from identification of the accompanying anions. With the exception of the phosphate containing complex, X-ray structures of the complexes confirmed the encapsulation of SO₄²⁻ at the complex core in preference to the other anions (Fig. 3). There are subtle differences between the basic [SO₄⊂Cu₂L₂]²⁺ dicationic core based on the Cu–Cu distances and helical twist angles of the individual complexes (Table 2). Interestingly, direct reaction of L with Cu(NO₃)₂·3H₂O results in a NO₃⁻ encapsulated complex [NO₃⊂Cu₂L₂](NO₃)₃ (Fig. S5, ESI‡) with similar Cu–Cu distances (Table 2), indicating that the Cu₂L₂ core with its additional intra-hydrogen bonding has limited flexibility and probably plays a part in the overall stability of the bound sulfate complex. For the phosphate complex, both data quality and the similarity in size between the dihydrogen phosphate and sulfate anions preclude direct identification of sulfate at the core of this helicate, however, additional evidence for encapsulation of SO₄²⁻ over H₂PO₄⁻ is given by HR-MS and elemental analysis on the bulk crystalline sample as used in the X-ray analysis. A detailed description of the structures is given in the ESI.‡

Fig. 3 Comparison of the [SO₄⊂Cu₂L₂]²⁺ dication for (a) the chloride, (b) nitrate and (c) dihydrogen phosphate mixed anion complexes emphasising the strong similarities between each.

Table 2 Comparison of Cu–Cu bond length and helix twist angle between the various complexes

Complex	Cu–Cu/Å	O–Cu–Cu–O/°
a Ref. 8.
[SO4⊂Cu2L2]Cl2	6.601(2)	127.3(1)
[SO4⊂Cu2L2](NO3)2	6.537(2)	125.4(1)
[SO4⊂Cu2L2](H2PO4)2	6.560(1)	126.4(2)
[NO3⊂Cu2L2](NO3)3	6.607(2)	129.8(2)
[BF4⊂Cu2L2](BF4)3	6.938(2)^a	120.0(4)^a

---

The stability of the sulfate core is due to a number of factors including, copper sulfate coordination and an electropositive cavity of protonated ammonium groups. However, it is the enhanced stability of the basic helicate structure brought about by intramolecular hydrogen bonding around the copper centre, coupled with an appropriate span between the copper metal centres that allows this complex to bind sulfate strongly.

Investigations of any sulfate/phosphate discrimination point to a preferred uptake of sulfate over phosphate which is noteworthy given the minor difference between the association constants (Δlog = 0.05).

Work continues in the development of systems related to [Cu₂(L–2H)₂] which have both a reduced strap size and more rigidity. Such minor variations appear to drastically change the anion preference.

This work was supported by grants from Massey University. For the high resolution mass spectroscopy data we thank Mrs Pat Gread and Prof. Bill Henderson from the University of Waikato. The MacDiarmid Institute for Advanced Materials and Nanotechnology and the Tertiary Education Commission of New Zealand provided funding for the Rigaku Spider optics and detector.

Notes and references

1. J.-i. Kim, H. Juwarker, X. Liu, M. S. Lah and K.-S. Jeong, Chem. Commun., 2010, 46, 764–766.
2. B. A. Moyer, L. H. Delmau, C. J. Fowler, A. Ruas, D. A. Bostick, J. L. Sessler, E. Katayev, G. D. Pantos, J. M. Llinares, A. Hossain, S. O. Kang and K. Bowman-James, Adv. Inorg. Chem., 2007, 59, 175–204; L. R. Eller, M. Stepien, C. J. Fowler, J. T. Lee, J. L. Sessler and B. A. Moyer, J. Am. Chem. Soc., 2007, 129, 11020–11021; K. Bowman-James, Acc. Chem. Res., 2005, 38, 671–678; C. A. Ilioudis and J. W. Steed, J. Supramol. Chem., 2003, 1, 165–187; K. Choi and A. D. Hamilton, Coord. Chem. Rev., 2003, 240, 101–110; V. Amendola, L. Fabbrizzi, C. Mangano, P. Pallavicini, A. Poggi and A. Taglietti, Coord. Chem. Rev., 2001, 219–221, 821–837; P. A. Gale, Coord. Chem. Rev., 2003, 240, 191–221; P. A. Gale and R. Quesada, Coord. Chem. Rev., 2006, 250, 3219–3244; P. Mateus, R. Delgado, P. Brandao and V. Felix, J. Org. Chem., 2009, 74, 8638–8646.
3. K. Wichmann, B. Antonioli, T. Soehnel, M. Wenzel, K. Gloe, K. Gloe, J. R. Price, L. F. Lindoy, A. J. Blake and M. Schroeder, Coord. Chem. Rev., 2006, 250, 2987–3003; G. Hennrich and E. V. Anslyn, Chem.–Eur. J., 2002, 8, 2218–2224; M. H. Filby and J. W. Steed, Coord. Chem. Rev., 2006, 250, 3200–3218; B. Wu, J. Liang, J. Yang, C. Jia, X.-J. Yang, H. Zhang, N. Tang and C. Janiak, Chem. Commun., 2008, 1762–1764; I. Ravikumar, P. S. Lakshminarayanan, M. Arunachalam, E. Suresh and P. Ghosh, Dalton Trans., 2009, 4160–4168; F. Zhuge, B. Wu, J. Liang, J. Yang, Y. Liu, C. Jia, C. Janiak, N. Tang and X.-J. Yang, Inorg. Chem., 2009, 48, 10249–10256; R. Custelcean and P. Remy, Cryst. Growth Des., 2009, 9, 1985–1989; R. J. Warr, A. N. Westra, K. J. Bell, J. Chartres, R. Ellis, C. Tong, T. G. Simmance, A. Gadzhieva, A. J. Blake, P. A. Tasker and M. Schroder, Chem.–Eur. J., 2009, 15, 4836–4850.
4. A. Kumar, S.-S. Sun and A. J. Lees, Coord. Chem. Rev., 2008, 252, 922–939; R. S. Forgan, J. E. Davidson, F. P. A. Fabbiani, S. G. Galbraith, D. K. Henderson, S. A. Moggach, S. Parsons, P. A. Tasker and F. J. White, Dalton Trans., 2010, 39, 1763–1770; J. W. Steed, Chem. Soc. Rev., 2009, 38, 506–519.
5. M. Mahoney Joseph, U. Nawaratna Gayathri, M. Beatty Alicia, J. Duggan Peter and D. Smith Bradley, Inorg. Chem., 2004, 43, 5902–5907; M. Mahoney Joseph, M. Beatty Alicia and D. Smith Bradley, Inorg. Chem., 2004, 43, 7617–7621.
6. P. A. Tasker, C. C. Tong and A. N. Westra, Coord. Chem. Rev., 2007, 251, 1868–1877.
7. S. G. Galbraith, L. F. Lindoy, P. A. Tasker and P. G. Plieger, Dalton Trans., 2006, 1134–1136.
8. M. Wenzel, G. B. Jameson, L. A. Ferguson, Q. W. Knapp, R. S. Forgan, F. J. White, S. Parsons, P. A. Tasker and P. G. Plieger, Chem. Commun., 2009, 3606–3608.
9. K. Soldenhoff, N. Hayward and D. Wilkins, EPD Congress 1998, Proc. Sess. Symp., held at the TMS Annual Meeting, San Antonio, United States of America, 1998, pp. 153–165.
10. H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbuehler, Talanta, 1985, 32, 1133–1139; H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbuehler, Talanta, 1985, 32, 257–264; H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbuehler, Talanta, 1985, 32, 95–101; H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbuehler, Talanta, 1986, 33, 943–951.
11. P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A: Found. Crystallogr., 1990, A46, 194–201.

---

Footnotes

† This article is part of the ‘Emerging Investigators’ themed issue for ChemComm.

‡ Electronic supplementary information (ESI) available: Ligand synthesis, crystal packing, table of weak interactions of the crystal structures, crystal structure of [NO₃⊂Cu₂L₂](NO₃)₃·2H₂O, Job plots of the UV-Vis experiments and HR-MS spectra of [SO₄⊂Cu₂L₂](H₂PO₄)₂. CCDC 783210–783213. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c0cc02230f

§ Multi-varied analysis of the spectra (SPECFIT/32™),¹⁰ assuming a 1 ∶ 1 anion to [Cu₂L₂]⁴⁺ binding model as supported by Job plots (see ESI‡).

¶ Characterisation of complexes: [SO₄⊂Cu₂L₂]Cl₂·2.5H₂O: green solid, yield 41%, MS(ESI) m/z 666 [Cu₂L₂SO₄]²⁺; anal. calc. for C₆₄H₁₀₀Cu₂N₈O₁₂Cl₂S·2.5H₂O: C 53.06, H 7.31, N 7.74%; found: C 53.02, H 7.18, N 7.65%. Green plate crystals suitable for X-ray diffraction analysis were isolated by slow diffusion of diethyl ether into an acetonitrile/ethanol solution (1 ∶ 1) containing the complex over one week. Crystal data for C₆₆H₁₀₈Cl₂Cu₂N₈O₁₄S, M = 1467.64, monoclinic, a = 19.805(4) Å, b = 20.320(4) Å, c = 21.840(4) Å, β = 109.52(3)°, V = 8283(3) ų, T = 100 K, P21/c (no. 14), Z = 4, 72 361 reflections measured, 14 014 unique (R_int = 0.0821) of which 14 014 were used in the calculations, R₁ = 0.0650 (9908 with F > 2σ(F)). Disordered solvent regions in this complex were treated in the manner described by van der Sluis and Spek,¹¹ as 234 e⁻ per cell, approx. to 3.25(CH₃OH) (=72 e⁻) per formula unit. [SO₄⊂Cu₂L₂](NO₃)₂·H₂O·3EtOH·MeCN: green solid, yield 88%, MS(ESI) m/z 666 [Cu₂L₂SO₄]²⁺; anal. calc. for C₆₄H₁₀₀Cu₂N₁₀O₁₈S·H₂O·3EtOH·MeCN: C 52.29, H 7.50, N 9.32%, found: C 52.43, H 7.21, N 9.11%. Green block crystals suitable for X-ray diffraction analysis were grown by slow diffusion of diethyl ether into an acetonitrile/ethanol mixture (1 ∶ 1) over 3 days. Crystal data for C₇₂H₁₂₁Cu₂N₁₁O₂₂S, M = 1651.94, triclinic, a = 10.464(2) Å, b = 18.668(4) Å, c = 21.678(4) Å, α = 78.77(3)°, β = 85.67(3)°, γ = 85.96(3)°, V = 4135.0(14) ų, T = 123 K, P1̄ (no. 2), Z = 2, 71 567 reflections measured, 12 207 unique (R_int = 0.0921) of which 12 207 were used in the calculations, R₁ = 0.0667 (8871 with F > 2σ(F)). [SO₄⊂Cu₂L₂](H₂PO₄)₂: yield 55%, MS(ESI) m/z 666 [Cu₂L₂SO₄]²⁺, anal. calc. for C₆₄H₁₀₄Cu₂N₈O₂₀P₂S: C 50.35, H 6.87, N 7.34%, found: C 50.20, H 7.10, N 7.20%. Green chip crystals suitable for X-ray diffraction analysis were grown by slow diffusion of diethyl ether into an acetonitrile/methanol/ethanol mixture (1 ∶ 1 ∶ 1) over 2 days. Crystal data C₆₄H₁₀₀Cu₂N₈O₂₀P₂S, M = 1522.60, orthorhombic, a = 20.2187(4) Å, b = 21.9011(4) Å, c = 38.781(3) Å, V = 17172.6(13) ų, T = 123 K, Pbca (no. 61), Z = 8, 126 497 reflections measured, 13 160 unique (R_int = 0.1223) of which 13 160 were used in the calculations, R₁ = 0.0758 (7141 with F > 2σ(F)). Disordered solvent regions were treated in the manner described by van der Sluis and Spek,¹¹ as 590 e⁻ per cell, approximating to 1(CH₃CH₂OH), 2(CH₃OH) and 1(H₂O) (=72 e⁻) per formula unit. [NO₃⊂Cu₂L₂](NO₃)₃·2H₂O: yield 92%, MS(ESI) m/z 649 [Cu₂L₂NO₃–H]²⁺; anal. calc. for C₆₄H₁₀₀Cu₂N₁₂O₂₀·2H₂O: C 50.55, H 6.89, N 11.05%, found: C 50.59, H 6.79, N 10.89%. Green chip crystals suitable for X-ray diffraction analysis were grown by slow diffusion of diethyl ether into an acetone/ethanol mixture (1 ∶ 1) over 1 week. Crystal data for C₆₄H₁₀₀Cu₂N₁₂O₂₀, M = 1484.64, monoclinic, a = 11.847(2) Å, b = 19.317(4) Å, c = 37.831(8) Å, β = 93.95(3)°, V = 8637(3) ų, T = 123 K, P21/c (no. 14), Z = 4, 84 887 reflections measured, 13 237 unique (R_int = 0.1092) of which 13 237 were used in the calculations, R₁ = 0.0860 (9141 with F > 2σ(F)). Disordered solvent regions were treated in the manner described by van der Sluis and Spek,¹¹ as 250 e⁻ per cell, approximating to 2(CH₃CH₂OH) and 0.75(H₂O) (=61.5 e⁻) per formula unit.

---