Reactivity of Hg(II) ortho-cyano-aminothiophenolate coordination polymers [Hg{SC6H3XN(C≡N)}]n with cyclic aromatic amines, Ph3P = E (E = O, S) and [HgCl2(κ2-phen)]

Abstract We report reactions of Hg(II) ortho-cyano-aminothiophenolate coordination polymers, [Hg{SC6H3XN(C≡N)}]n, with donor ligands. Addition of cyclic aromatic amines, such as pyridine (L), leads to breakdown of the polymeric structure to give mononuclear four-coordinate complexes, [Hg{κ2 -SC6H3XN(C≡N)}L2], while 2,2’-bipy and 1,10-phen (L2) similarly afford mononuclear chelate complexes [Hg{κ2 -SC6H3XN(C≡N)}(κ2-L2)]. In contrast, Ph3P = E (E = O, S) form 1:1 complexes which, on the basis of related phosphine adducts, are formulated as binuclear [Hg{µ-SC6H3XN(C≡N)}(κ1-E = PPh3)]2. Reaction of [Hg{SC6H4N(C≡N)}]n with [HgCl2(κ2 -phen)] affords an adduct tentatively formulated as [Hg{κ2-SC6H4N(C≡N)}(µ-Cl)2Hg(κ2-phen)] in which the two chlorides bridge the Hg.Hg vector. Graphical Abstract


Introduction
Extended inorganic systems [1], as exemplified by 3 D metal-organic frameworks (MOFs) [2,3], have become an area of intense research over the past 25þ years having a wide range of potential applications [4]. The simplest such systems are 1 D coordination polymers which are known for a wide range of metals [5] but are especially prevalent for Hg(II) [6]. We recently reported the synthesis of a small series of Hg(II) coordination polymers, [HgfSC 6 H 3 XN(CN)g] n (X ¼ H, Me, Cl, NO 2 , Br) (1a-e), such chemistry is not always so simple. Thus with HgCl 2 , addition of pyridine leads to formation of a coordination polymer in which chlorides bridge six-coordinate Hg(II) centers [11]. Addition of two equivalents of pyridine to 1 b-c in CHCl 3 resulted in formation of [HgfSC 6 H 3 XN(CN)g(py) 2 ] (2 b-c) in good yields. For 2 b, integration of the methyl resonance against signals in the aromatic region showed a ratio of pyridine to ocap of 2:1, and this was supported by the elemental analysis. While this does not unequivocally rule out the possibility of dimeric or polymeric structures, we strongly favor formation of mononuclear species (Scheme 2). With pyrazole (pz), related complexes [HgfSC 6 H 3 XN(CN)g(pz) 2 ] (3 b-c) were similarly formed. We next turned our attention to reactions with 2-aminopyridine (2-ampy) and 2-aminopyrimidine (2ampym), both of which have the potential to bind through either the ring nitrogen or the amine group. Complexes 4 b-c and 5 b-c resulted, for which NMR data clearly showed a 2:1 ratio of ocap to added ligand and also coordination of the pyridyl exclusively, with the NH 2 moiety remaining unbound, as shown by the presence of a broad signal at ca. d 6 ppm in the 1 H NMR spectra integrating to four protons. Similar reactions were also carried out with 1a but the poor solubility of the products meant that good quality characterizing data could not be obtained.
2.18 and 2.27, respectively, and confirm the ocap and diamine ratio of 1:1. Unfortunately we have been unable to confirm the precise structure using X-ray diffraction despite repeated attempts to grow suitable crystals. We also prepared the chloro-and nitro-derivatives 6c-d and 7c-d, respectively, in the hopes of growing suitable crystals, but again we were unsuccessful.

Reactions with Ph
We next turned our attention to coordination of other potential two-electron donor ligands in the form of Ph 3 P ¼ O and Ph 3 P ¼ S, which we hoped would allow us to probe the nature of hard and soft preference for oxygen and sulfur, respectively. Sun and co-workers [17] reported the synthesis and crystal structure of mononuclear [HgBr 2 (j 1 -O ¼ PPh 3 ) 2 ] for which well-defined m(P ¼ O) vibrations are observed in the IR spectrum at 1187, 1164 and 1151 cm À1 , while the corresponding chloride has also been briefly reported [18].
Addition of ca. 2.5 equivalents of Ph 3 P ¼ O or Ph 3 P ¼ S to 1 led to formation of new products 8a-b and 9 b,d characterized analytically as adducts with a ratio of Hg to phosphine-chalcogenide of 1:1. For all, a singlet is observed in the 31 Pf 1 Hg NMR spectrum, for example at 25.5 ppm for 8a, while for the same complex IR bands at 1166(s) and 1118(s) cm À1 are in accord with the expected positions for m(P ¼ O) vibrations. The 1 H NMR spectra of both 8 b and 9 b show a single methyl environment at d 2.28 and integration suggests that there is an ocap to added ligand ratio of 1:1 despite the excess of Ph 3 P ¼ E used. Once again, our inability to grow suitable single crystals means that the precise molecular structures remain unknown. We favor that shown in Scheme 4 on the basis of the 1:1 ratio of ligands and this structural type is adopted by PPh 3 derivatives [7,8]. Irrespective of the actual molecular structures is the observation that both of these ligands readily bind to Hg(II) but fail to break the mercury-thiolate bridges, similar to the chemistry observed for phosphines.

Reaction of 1a with HgCl 2 (j 2 -phen)
From the preceding sections it is clear that the polymeric nature of 1 can be disrupted upon addition of two-electron donor ligands. Preliminary studies suggest that binuclear complexes are also available following a similar strategy whereby the donor ligand(s) are coordinated to a second metal center. To test this we reacted 1a with [Hg(j 2 -phen)Cl 2 ] in EtOH. Both reagents were only sparingly soluble, but a slow deepening of the color suggested that they were slowly reacting. After 2 h a bright yellow Scheme 3. Reactions of 1 with 2,2'-bipyridine (bipy) and 1,10-phenanthroline (phen).
precipitate formed which was collected and dried to afford what we suggest is the unsymmetrical bimetallic [Hgfj 2 -SC 6 H 4 N(CN)g(m-Cl) 2 Hg(j 2 -phen)] (10) (Scheme 5). The 1 H NMR spectrum clearly shows formation of a new product containing equal amounts of phen and ocap ligands, while elemental analysis is consistent with the stoichiometry shown. We have been unable to grow crystals to confirm this formulation but are further developing this concept with the aim of preparing a range of heterobimetallic ocap complexes.

Summary and conclusion
In this contribution we have prepared and characterized a range of new Hg(II) orthocyano-aminothiophenolate (ocap) complexes by reaction of the coordination polymers [HgfSC 6 H 3 XN(CN)g n ] with N, O, and S-donor ligands. With both monodentate and bidentate N-donor ligands, mononuclear complexes result in which the ocap ligand binds in a simple chelate fashion through nitrogen and sulfur and the pendent cyanide ligand remains uncoordinated. This contrasts with our earlier investigation of these polymers with mono and bidentate phosphines in which the majority of products were either binuclear ocap-bridged complexes or coordination polymers in which a j 1 ,j 1 -diphosphine linked Hg(ocap) centers [7,8]. The predominance of mononuclear species models the chemistry we have recently found with Pd(II)-ocap complexes. In contrast, both Ph 3 P ¼ E (O, S) react in a 1:1 stoichiometry to afford species which we propose on the basis of the analogous PPh 3 chemistry [7,8] to be ocap-bridged dimers. Somewhat unexpectedly, addition of [HgCl 2 (j 2 -phen)] to [HgfSC 6 H 4 N(CN)g n ] also results in breakdown of the polymeric network to afford a bs(chloride)-bridged dimer formulated as [Hgfj 2 -SC 6 H 4 N(CN)g(m-Cl) 2 Hg(j 2 -phen)]. In the analogous phosphine chemistry [7,8] we were able to show that, while several species may be accessible in solution, interconverting via a number of (unresolved) fluxional processes (as evidence by broad NMR spectra), in the solid state a range of different ocap coordination modes were accessible. In marked contrast, all complexes reported here show sharp and well-resolved NMR spectra, in accord with static solution structures (on the NMR timeframe) but frustratingly, we have been unable to unequivocally establish coordination modes due to our inability to grow crystals of suitable quality. Attempts to do this are ongoing as we further develop the coordination chemistry of this novel redox-active ligand.

General procedures
1 H and 31 Pf 1 Hg NMR spectra were recorded on a Varian Unity spectrometer. IR spectra were recorded on either a Shimadzu FT-IR 8400 spectrophotometer from 400-4000 cm À1 as KBr discs and from 200-600 cm À1 as CsI discs or on Bruker Tensor 28 spectrometer with a platinum ATR unit in Martin-Luther Universit€ at Halle-Wittenberg, Germany. Elemental analyses were carried out at Al Al-Bayt University, Jorden using a Euro vector EURO EA300 elemental analyzer. Melting points were measured on a Gallenkamp melting point apparatus and are uncorrected. All commercially pure chemicals were used as received. Complexes 1a-d were prepared as previously outlined [8].

Synthesis of [HgfSC 6 H 3 XN(CN)gL 2 ] (2-5)
To a suspension of 1 b (0.200 g, 0.554 mmol) in CHCl 3 (10 ml), pyridine (py) (0.08 g, 1.108 mmol) was added, and the mixture was refluxed for 3 h. The yellow solution was set aside to slowly evaporate at room temperature. A yellow solid formed which was collected, washed with Et 2 O and dried under vacuum. Recrystallization from DMSO gave a yellow powder.

Preparation of [HgfSC 6 H 3 XN(CN)g(j 2 -L)] (6-7)
A solution of 2,2 0 -bipyridine (bipy) (0.073 g, 0.467 mmol) in EtOH (10 ml) was added to a suspension of 1a (0.162 g, 0.467 mmol) in ethanol (10 ml). The mixture was stirred for 3 h. The yellow precipitate formed was filtered off, washed with EtOH and dried under vacuum to give a yellow powder. A solution of 2,2 0 -bipyridine (bipy) (0.088 g, 0.554 mmol) in CHCl 3 (10 ml) was added to a suspension of 1 b (0.200 g, 0.554 mmol) in CHCl 3 (10 ml). The mixture was refluxed for 3 h. The yellow solution was slow evaporated at room temperature giving a yellow precipitate which was filtered, washed with diethyl ether and dried under vacuum. The precipitate was recrystallized from DMF to give a pale-yellow powder. Complexes 6c and 6e were prepared and isolated using a similar method.
6b: Yellow, yield: 0.242 g (68% A solution of 1,10-phenanthroline (phen) (0.050 g, 0.277 mmol) in CHCl 3 (10 ml) was added to a suspension of 1 b (0.100 g, 0.277 mmol) in CHCl 3 (10 ml). The mixture was refluxed for 3 h. The yellow solution was slow evaporated at room temperature giving a yellow precipitate which was filtered, washed with Et 2 O and dried under vacuum. The yellow precipitate was recrystallized from DMF to give a pale-yellow powder.