Hydrogen and Halogen Bond Mediated Coordination Polymers of Chloro-Substituted Pyrazin-2-Amine Copper(I) Bromide Complexes

Abstract

A new class of six mono- (1; 3-Cl-, 2; 5-Cl-, 3; 6-Cl-) and di-(4; 3,6-Cl, 5; 5,6-Cl-, 6; 3,5-Cl-) chloro-substituted pyrazin-2-amine ligands (1–6) form complexes with copper (I) bromide, to give 1D and 2D coordination polymers through a combination of halogen and hydrogen bonding that were characterized by X-ray diffraction analysis. These Cu(I) complexes were prepared indirectly from the ligands and CuBr₂ via an in situ redox process in moderate to high yields. Four of the pyrazine ligands, 1, 4–6 were found to favor a monodentate mode of coordination to one Cu^I ion. The absence of a C6-chloro substituent in ligands 1, 2 and 6 supported N1–Cu coordination over the alternative N4–Cu coordination mode evidenced for ligands 4 and 5. These monodentate systems afforded predominantly hydrogen bond (HB) networks containing a catenated (μ₃-bromo)-Cu^I ‘staircase’ motif, with a network of ‘cooperative’ halogen bonds (XB), leading to infinite polymeric structures. Alternatively, ligands 2 and 3 preferred a μ₂-N,N’ bridging mode leading to three different polymeric structures. These adopt the (μ₃-bromo)-Cu^I ‘staircase’ motif observed in the monodentate ligands, a unique single (μ₂-bromo)-Cu^I chain, or a discrete Cu₂Br₂ rhomboid (μ₂-bromo)-Cu^I dimer. Two main HB patterns afforded by self-complimentary dimerization of the amino pyrazines described by the graph set notation R₂²(8) and non-cyclic intermolecular N–H∙∙∙N’ or N–H∙∙∙Br–Cu leading to infinite polymeric structures are discussed. The cooperative halogen bonding between C–Cl∙∙∙Cl–C and the C–Cl∙∙∙Br–Cu XB contacts are less than the sum of the van der Waals radii of participating atoms, with the latter ranging from 3.4178(14) to 3.582(15) Å. In all cases, the mode of coordination and pyrazine ring substituents affect the pattern of HBs and XBs in these supramolecular structures.

1. Introduction

Construction of supramolecular structures from small molecules that self-assemble using hydrogen bonds (HBs) and other non-covalent interactions is the ultimate goal of the crystal engineering discipline [1]. Hydrogen bonding, due to the smaller size and easy polarizable nature of H-atom, has become a reliable tool to fabricate highly symmetric and exotic solid-state networks with tunable properties for applications in biology [2], and materials sciences [3]. The H-bonding knowledge gleaned from organic co-crystals have been applied to self-assemble metal-organic/coordination networks [4,5,6,7]. In fact, under the “crystal engineering umbrella”, the design and synthesis of coordination compounds have received wide attention, due to their intriguing structural topologies [8,9,10]. Unlike organic co-crystals, engineering inorganic compounds is dependent on two principals; primary coordination sphere (metal-ligand interactions) and secondary coordination sphere (non-covalent interactions) [11]. Despite several factors (e.g., pH, temperature) [12,13] could influence these two “parameters” for structurally diverse outcomes, modulation of networks based on the metal-ions geometry, organic ligands and their functional groups stand out in coordination chemistry research [14,15,16,17,18,19,20,21]. This fact is due to reproducible outcomes, that reflect the strong metal-ligand coordination bonds, and can be achieved with the judicious choice of organic ligands and metal-ions [14,15,16,17,18,19,20,21]. In this context, for example, O- and N-atoms are typical donors for the coordination bond formation; the former are derived from functional groups such as –COOH, –SO₃H and phosphonates [22,23,24], and the latter primarily from N-heterocycles [25]. The combined use of these two groups render ligands a strong coordinating ability, and are well-known for the preparation of homometallic and heterometallic coordination compounds [25].

In this regard, aminopyrazine carboxylic acids (L), of the forms L, and L⁺, for the construction of hydrogen-bonded organic co-crystals [26,27,28,29,30,31], and L⁻ functioning as a polydentate ligand in the preparation of coordination compounds [32,33,34,35,36,37,38,39] are reported in the literature. The presence of aromatic N-atoms and carboxylic acid/carboxylate groups within the same ligand may enhance the N–M bond strengths, often assisted by the polydentate bonding nature of O-atoms [26]. However, if the –COOH group is replaced by an aprotic donor substituent such as chlorine, and in combination with copper halides, what will happen to N-atom coordination nature? Will the chlorine substituents and metal-bound halides establish halogen bonds (XBs) [40,41,42,43,44], and halogen∙∙∙halogen interactions [45,46,47,48]? This knowledge of XBs in metal complexes is derived from our previous experiences in halopyridine-Cu(I)/Cu(II) compounds [49,50,51,52]. When the bulky chlorines are installed close to an N-atom, can this affect the N–M coordination? How do the substituents mediate a hybrid topology containing –HN–H∙∙∙N_pz (pz = pyrazine) hydrogen bonds and C–Cl∙∙∙Br halogen bonds? To test our hypothesis, we synthesized six chloro-substituted pyrazin-2-amine ligands (1–6) using procedures reported in the literature [53,54,55,56,57], and each ligand was combined with CuBr₂ in a 1:2 metal:ligand ratio. The CuBr complexes of these ligands were obtained by exploiting the known redox activity of Cu(II) halides in the presence of organic carbonyl compounds [31]. This work represents the first systematic study of the metal-ligand, HB, and XB interactions of chloro-substituted pyrazin-2-amines with copper halides and recent results will be discussed.

2. Materials and Methods

All reagents and solvents, including 2, were obtained from commercial suppliers and were of at least reagent grade, and used as received, unless otherwise stated. The ligands 1 and 3–6 were prepared by modified literature procedures [53,54,55,56,57,58]. Full experimental, FT-IR spectroscopic details and the single-crystal [59,60,61,62,63,64,65,66] and powder X-ray experimental [67,68,69], and structure refinements for 1∙CuBr–6∙CuBr (CCDC numbers: 2001484–2001491) are given in the Supporting Information.

3. Results and Discussion

This new class of chloro-substituted pyrazin-2-amine ligands given in Chart 1 were made to react with cupric bromide (CuBr₂), either by refluxing in a 1:1:1 (v/v/v) mixture of acetone (Ace), ethanol (EtOH) and acetonitrile (MeCN) or 1:5 (v/v) mixture of Ace:EtOH to afford the desired Cu(I)Br complexes, via an in situ redox process, according to Scheme 1. An excess of pyrazine ligand was used to act as ligand and auxiliary base for the HBr liberated during the reduction of CuBr₂ to CuBr by acetone [70]. The reduction was accompanied by a color change from a deep blue-green solution characteristic of Cu(II) ions, to a clear, yellow solution upon refluxing for several minutes. In total, eight different Cu(I)Br coordination complexes obtained from the combination of these six ligands will be discussed.

X-ray quality crystals of the complexes were obtained by slowly concentrating the reaction mixtures by controlled evaporation. Interestingly, 2c∙CuBr was only obtained from a 1:5 (v/v) Ace and EtOH reaction mixture, together with 2b∙CuBr. The concomitant polymorphism did not allow the isolation of 2c∙CuBr as a phase pure material; however, 2b∙CuBr could be prepared independently by alternative methodologies in moderate yield. Moreover, attempts to recrystallize the mixture of 2c∙CuBr and 2b∙CuBr from MeCN afforded only complexes 2a∙CuBr and 2b∙CuBr. This suggests the 2c∙CuBr is only somewhat stable when prepared from a 1:5 (v/v) Ace:EtOH mixture and readily dissociates in the polar aprotic solvent MeCN. Optical microscopy could be routinely used to distinguish the different color and crystal habits of the three different polymorphs of (2a–c) CuBr (See Figure S1). Powder X-ray diffraction methods were used to demonstrate the solid-state structures and estimate the phase purity of the bulk material by using Pawley full pattern fittings. The FT-IR spectra of the isolated complexes were compared to the respective ligands, and display two specific regions from 3500–2800 cm⁻¹ and 1200–400 cm⁻¹ for the hydrogen-bonded N–H stretching [71] and fingerprint regions, respectively, which are distinctive for each ligand and complex (see Figures S10–S16).

In general, reactions between donor ligands (L) and CuX (X = Cl, Br, I) yield complexes with formula Cu_nX_nL_m, that display diverse structural topologies, containing rhomboid dimer, zigzag polymer, staircase polymer, closed cubane, and hexagon clusters, have been reported in the literature [72]. The structural diversification stems from the very nature of the ligand coordination modes and the tendency of copper halides to form clusters via µ₂- and µ₃-halide bridges [73]. In our eight complexes, we obtained three topologies exclusively, namely staircase polymer (1∙CuBr, 2a∙CuBr, 2c∙CuBr, 4∙CuBr, 5∙CuBr, and 6∙CuBr), rhomboid dimer (2b∙CuBr), and zigzag polymer (3∙CuBr), as depicted in Figure 1. The staircase and rhomboid structures feature well-known Cu∙∙∙Cu distances (Cuprophilic interactions) [74], ranging from 2.7547(13) to 3.086(4) Å, but will not be discussed further in the text.

The complexes 1∙CuBr and 2a∙CuBr both crystallize as clear, pale-yellow or colorless needles in an orthorhombic space group, Pna2₁ and P2₁2₁2₁, respectively. Both contain discrete 1D polymeric chains of the catenated (μ₃-bromo)-Cu^I ‘staircase’ that run parallel to the c-axis and a-axis in 1∙CuBr and 2a∙CuBr, respectively. In these chains, the Cu(I) ions have tetrahedral geometry, coordinated by three bromines and the pyrazine N1-atom. Since the pyrazine N4-atom is not involved in the N–Cu bond formation, it plays a central role in the HB formation involving a C2-amino group of an adjacent chain. The orthogonal arrangement of the discrete 1-D polymeric chains leads to a herringbone packing structure in both 1∙CuBr and 2a∙CuBr, as illustrated in Figure 2. The N4∙∙∙H–N HBs in 1∙CuBr of 2.245(11) Å, [∠N4∙∙∙H–N = 155.6(9)°] are shorter than in 2a∙CuBr [2.349(7) Å, ∠N4∙∙∙H–N = 133.4(5)°]. The XBs in 1∙CuBr and 2a∙CuBr complexes are, however, more comparable at 3.465(3) Å [∠Br∙∙∙Cl–C = 165.4(3)°] and 3.509(3) Å [∠Br∙∙∙Cl–C = 159.5(3)°], respectively. This network of HBs and XBs does lead to a complex molecular packing that extends in three dimensions.

The complex 6∙CuBr crystallizes in a monoclinic P2₁/c space group, and is composed of 1D polymeric chains of the catenated (μ₃-bromo)-Cu^I ‘staircase’ that run along the shortest unit cell a-axis. The asymmetric unit contains one bromide anion and a one Cu(I) cation coordinated by the sterically less hindered N1-atom of 6. Not surprisingly, the C3 and C5 chloro-substituents vicinal to the N4-atom render this ring nitrogen Cu-coordination passive. The orthogonal arrangement of the discrete 1D polymeric chains leads to a herringbone packing structure that is similar to 1∙CuBr and 2a∙CuBr (for 6∙CuBr, see Figure S2). The 1D chains are aligned by a more extensive network of XBs, afforded by the C3- and C5-Cl substituents (C3–Cl∙∙∙Br–Cu [3.4794(14) Å, ∠Br∙∙∙Cl–C = 162.44(19)°], C5–Cl∙∙∙Br–Cu [3.4178(14) Å, ∠Br∙∙∙Cl–C = 175.34(19)°]) and longer N4∙∙∙H–N of 2.511(4) Å, [∠N4∙∙∙H–N = 135.8(3)°] HBs, compared to 1∙CuBr and 2a∙CuBr.

The N1–Cu mode of coordination realized in 1∙CuBr, 2a∙CuBr, and 6∙CuBr allows the vicinal C2-amino group hydrogen and a Cu(I) bound bromide to form cyclic N–H∙∙∙Br–Cu HBs, with distances varying from ca. 2.609(7)–2.774(2) Å [∠Br∙∙∙H–N = 146(1)°–168.9(4)°], within the discrete polymeric (μ₃-bromo)-Cu^I ‘staircase’ chain (See Figure S3). These cyclic intra-chain N–H∙∙∙Br HBs afford a pseudo-helical type arrangement of the pyrazine ligands along the polymeric (μ₃-bromo)-Cu^I ‘staircase’ backbone. The HBs, in combination with the C–Cl∙∙∙Br–Cu XBs from neighboring chains, aids the stabilization of the catenated (μ₃-bromo)-Cu^I ‘staircase’ polymeric chains in these complexes.

The complexes 4∙CuBr and 5∙CuBr crystallize in a triclinic P-1 and monoclinic P2₁/n space groups, respectively. In both complexes, the catenated (μ₃-bromo)-Cu^I ‘staircase’ motif is preserved, however, the tetrahedral Cu(I) centers are coordinated by three bromides and a pyrazine N4-atom. In the case of 4∙CuBr, it crystallizes with an additional pyrazine molecule in the asymmetric unit that is not involved in coordination with the Cu(I) centers. The N4–Cu coordination mode found in both complexes allows the C2-amino group to participate in hydrogen-bonding with a neighboring pyrazine in a self-complimentary N–H∙∙∙N_pz pyrazine dimer. The resultant hydrogen-bonded dimers in 4∙CuBr and 5∙CuBr are described by the graph set notation R₂²(8), with HB parameters of ca. 2.308(12) Å, [166.4(7)°] and 2.138(5) Å [160.1(3)°], respectively (Figure 3). These Watson–Crick-like base pairing structures are characteristic of aminopyrazine derivatives [75,76]. The R₂²(8) hydrogen bonding in 4∙CuBr and 5∙CuBr leads to essentially linear chains that pack into laminar 2D polymeric sheets held together by the polymeric (μ₃-bromo)-Cu^I ‘staircase’ motif. The non-coordinating pyrazine in 4∙CuBr forms an additional set of R₂²(8) hydrogen-bonded dimers that cross-link the 2D sheets via non-cyclic N–H∙∙∙N-ring hydrogen bonds [2.219(11) Å ∠N–H∙∙∙N4, 177.7(8)°], into a 3D structure shown in Figure 3a. Although 4∙CuBr and 5∙CuBr lack the intra-chain cyclic N–H∙∙∙Br–Cu HBs (see Figure S4) found in 1∙CuBr, 2a∙CuBr, and 6∙CuBr, a network of non-cyclic inter-chain N–H∙∙∙Br–Cu and C–H∙∙∙Br–Cu HBs are established between neighboring chains in 5∙CuBr or the non-complexed pyrazine in 4∙CuBr. Overall, the N–H∙∙∙Br–Cu and C–H∙∙∙Br–Cu HBs together with the C–Cl∙∙∙Br–Cu, and C–Cl∙∙∙Cl–C XBs stabilize the polymeric μ₃-bromo ‘staircase’, and further cross-link the laminar 2D polymeric sheets into complex 3D structures. Notwithstanding the structural similarities between 4∙CuBr and 5∙CuBr, as determined by single-crystal X-ray diffraction, the solid-state structure of 4∙CuBr could not be demonstrated in the bulk material, as determined by powder X-ray diffraction. Full pattern Pawley analysis of 4∙CuBr suggests that the bulk material is effectively isomorphous with 6∙CuBr (See Table S6 and Figure S8). This suggests that additional polymorphs may exist where the Cu(I) center is coordinated by the sterically more congested N1 in 4. This possibility has also been observed in 3∙CuBr, and polymorphism has been demonstrated in 2(a–c)∙CuBr, as described below. Despite repeated attempts, under different conditions, alternative single-crystal structures of Cu(I) complexes of 4 have still not been realized.

The three polymeric structures 2b, 2c∙and 3∙CuBr all contain bridging µ₂-N,N’ pyrazine ligands, and display the three (CuBr)_n structural motifs outlined in Figure 1. Complex 2b∙CuBr crystallizes in the orthorhombic space group Pbcn as amber, prismatic blocks (See Figure S1a). In this complex, the pyrazine ligands are µ₂-N,N’ bridging between Cu₂Br₂ rhomboid dimer units to afford a distorted tetrahedral coordination geometry at the Cu(I) ion formed by two symmetry-equivalent bromines, and the N1- and N4-atoms of two different pyrazine ligands. This leads to a 2D dimensional honey-comb sheet that propagates in the ab plane, as illustrated in Figure 4a. The honey comb structure is similar to the [CuCl(μ-2,5-dimethylpyrazine-N,N′)]_n structure reported in the literature [77]. Adjacent layers are held together by C5–Cl∙∙∙Br–Cu [3.5105(7) Å, ∠Br∙∙∙Cl–C = 161.83(8)°] XBs and N–H∙∙∙Br–Cu [2.6402(2) Å, ∠Br∙∙∙H–N = 159.28(14)°] HBs, as given in Figure 4b. Moreover, these HBs and XBs, together with steric effects, prevent the formation of a catenated μ-bromo-Cu^I motif, as in the staircase polymers previously mentioned.

The bright yellow, needles of 2c∙CuBr crystallize in the monoclinic space group P2₁/n. The asymmetric unit consists of a Cu(I) ion; one bromide and one half of a pyrazine ligand. The C5-chlorine and C2-amine substituents exhibit a positional disorder with 50:50 occupancies. The distorted tetrahedral coordination environment of Cu(I) consists of one pyrazine nitrogen and three symmetry equivalent bromide anions. Ligand 2 is µ₂-N,N’ bridging to yield a 2D polymeric sheet containing the catenated (μ₃-bromo)-Cu^I ‘staircase’ chain structure, as shown in Figure 5. The polymeric structure of 2c∙CuBr is similar to that found in the 2D coordination polymer of 2-aminopyrazine with CuI (i.e., [Cu₂I₂(2-aminopyrazine)]_n) [78] and [2(μ-2,5-dimethylpyrazine-N,N′)Cu₂]_n where X = Br, I [77]. The adjacent sheets are linked by N–H∙∙∙Br–Cu [2.6054(17) Å, ∠Br∙∙∙H–N = 175(4)°] interactions between the C2-amino group and a Cu(I)-bound bromide. As a result of positional disorder, the C–Cl bond length associated with the 50% occupancy is close to the standard C–Cl bond distance, and results in a weak C5–Cl∙∙∙Br–Cu [3.582(15) Å, ∠Cl∙∙∙Br–Cu = 103.9(9)°] ’contact’. These weak interactions are likely triggered by the N–H∙∙∙Br–Cu interactions between the neighboring molecules in this laminar structure. Due to the disorder, there are possible HB and XB interactions that could alternate along a plane that passes through the amino- and chloro-substituents of the pyrazines of adjacent sheets, thus forming moderately short Cl∙∙∙Cl halogen contacts with R_XB = 0.88. These HBs and XB interactions further supports the catenated (μ₃-bromo)-Cu^I ‘staircase’ motif as a common structural feature in these complexes.

The complex 3∙CuBr, crystallized in the monoclinic space group P2₁/n. The distorted tetrahedral coordination environment of the Cu(I) is composed of two symmetry-related bromides and two crystallographically different N-atoms from two distinct pyrazine ligands. The Cu(I) ions are bridged by µ₂-N,N’ pyrazine ligands, forming 1D linear chains that run along the c-axis as illustrated in Figure 6a. These chains are further linked into a 2D sheet structure by polymeric (μ₂-bromo)-Cu^I-Br chains along the a-axis. The 2D layers are not flat, but instead adopt a slightly corrugated pattern because of the ∠N–Cu–N = 136.6(4)° at the distorted tetrahedron at the Cu(I) ion, as shown in Figure 6b. Two types of interactions assemble the layers into a 3D structure. First is the hydrogen bonding between a C2-amino group and Cu(I)-bound bromide from an adjacent layer [2.3615(10) Å, ∠Br∙∙∙H–N = 164.2 (7)°; 2.4846(12) Å, ∠Br∙∙∙H–N = 158.9(6)°], to afford a R₂²(8) ring motif from a set of two N–H∙∙∙Br HBs. The second interaction is C6–Cl∙∙∙Br–Cu halogen bonds [3.527(3) Å, ∠Br∙∙∙Cl–C = 158.8(4)°] between a chlorine on the pyrazine ring and bromide coordinated to Cu(I) on an adjacent layers, as shown in Figure 6b. Similar to the (μ₃-bromo)-Cu^I ‘staircase’ motif, the μ₂-bromo polymeric chains found in 3∙CuBr is fully stabilized by the network of HB and XBs, and is likely adopted because of the steric demands of the N1 coordination, compared to the more accessible N4-coordination mode in ligand 3.

4. Conclusions

In summary, eight Cu(I)Br coordination polymers have been synthesized, and their single-crystal structures characterized by using X-ray diffraction analysis. Three different types of (CuBr)_n arrangements with n ≥ 2 were realized. Six of the eight complexes form characteristic catenated μ₂-bromo staircase polymers that are reminiscent of those reported in related copper(I) halide complexes. Our findings indicate that a combination of cyclic and non-cyclic hydrogen bonds (HBs), between the C2-amino group and Cu(I) bound bromides (N–H∙∙∙Br–Cu) and halogen bonding (C–Cl∙∙∙Br–Cu) interactions, underpins the formation of catenated μ₂- and μ₃-bromo (CuBr)_n polymeric structures.

It was found that the mode of coordination by the dichloro-substituted aminopyrazines 4–6 to the Cu(I) could be readily predicted based on simple electronic and steric arguments. In the mono chloro-substituted pyrazines 1–3, these effects were less obvious, as evidenced by the realization of both monodentate (1∙CuBr and 2a∙CuBr), and polymeric structures (2b∙CuBr, 2c∙CuBr and 3∙CuBr). The polymeric 3∙CuBr suggests that there is little steric hindrance afforded by the combination of a C2-amino and C6-chloro substituents vicinal to the N1–Cu coordination site, and may indicate that this is not a major factor in determining N1-Cu coordination passiveness in 4∙CuBr and 5∙CuBr. This could also account for the unique polymeric (μ₂-bromo)-Cu^I chain in 3∙CuBr, which is able to accommodate the more sterically demanding N1–Cu coordination, and allows stabilizing cyclic N–H∙∙∙Br HBs.

Three polymorphs were isolated and structurally characterized based on the simple 2-amino-5-chloropyrazine ligand, 2. We demonstrate that concomitant polymorphism leads to both 1D and 2D polymeric structures based on the μ₃-bromo staircase polymer motif in 2a∙CuBr and 2c∙CuBr, respectively. The third polymorph, 2b∙CuBr, displays a unique honeycomb network composed of discrete Cu₂Br₂ rhomboid (μ₂-bromo)-Cu^I dimers and µ₂-N,N’ bridging ligands. These three polymorphs demonstrate the role of the C2-amino group in forming three types of hydrogen bonding patterns recognized in all eight of these complexes. The self-complementary, symmetry-related N–H∙∙∙N_pz pyrazine dimers described by the graph set notation R₂²(8) with distances varying from ca. 2.138(5)–2.308(12) Å [∠Br∙∙∙H–N = 160.1(3)°–177.7(8)°] were found in N4–Cu coordination structures 4∙CuBr and 5∙CuBr, and resulted in the formation of laminar 2D sheet structures. Based on the competitive binding sites in the simple mono-substituted aminopyrazine ligands, and the role of the amino-group to influence structural features through hydrogen bonding, we are currently investigating the formation of dihalopyrazine-copper complexes.

The Cu(I) halide complexes have attracted considerable attention because of their physical properties and functional applications in future technologies, such as solar energy conversion, light emitting devices, and possible sensing applications. As a result, there has been considerable focus on exploring and developing structural relationships that will provide insights for future materials design and synthesis. The interesting 2D-honeycomb network found in 2b∙CuBr is a particularly appealing architecture that could have applications in gas storage or separation, if the porosity can be tuned through ligand design. While systematic structural studies on Cu(I) halide and pyrazine derivatives have not received much attention to date, this work demonstrates a level of structural predictability not typically observed in Cu(I) halide complexes. We believe future investigations of halo-substituted aminopyrazines and related pyrazine derivatives will provide the basis for an empirical structural database that will aid future material designs.