Structural diversity, spectral characterization and computational studies of Cu(I) complexes with pyridylamide ligands

Inorganica Chimica Acta 446 (2016) 150–160

Contents lists available at ScienceDirect

Inorganica Chimica Acta journal homepage: www.elsevier.com/locate/ica

Structural diversity, spectral characterization and computational studies of Cu(I) complexes with pyridylamide ligands Ethan P. McMoran a, Manuel N. Dominguez b, Ethan M. Erwin a, Douglas R. Powell c, Gerard T. Rowe b,⇑, Lei Yang a,⇑ a b c

Department of Chemistry, University of Central Arkansas, Conway, AR 72035, USA Department of Chemistry & Biochemistry, University of South Carolina-Aiken, Aiken, SC 29801, USA Department of Chemistry and Biochemistry, University of Oklahoma, Norman, OK 73019, USA

a r t i c l e

i n f o

Article history: Received 1 February 2016 Received in revised form 7 March 2016 Accepted 8 March 2016 Available online 14 March 2016 Keywords: Pyridylamide ligands Cu(I) complexes Crystal structures Cyclic voltammetry Computational studies

a b s t r a c t Cu(I) complexes supported by pyridylamide ligands N-2-acetamidopyridine (Haap), N-(3-pyridyl)nicotinamide (3-pna), N-(2-pyridyl)isonicotinamide (2-pina) and N-(4-pyridyl)picolinamide (4-ppa) were synthesized and characterized. The reaction of CuCl with the deprotonated form of Haap yielded the complex [Cu2(aap)2]n (1) with a one-dimensional polymeric structure. Complexes [Cu(3-pna)3(NCCH3)](HF)(SbF6) (2) and [Cu(4-ppa)(NCCH3)2]BF4 (5) displayed mononuclear structures with tetrahedral geometries on Cu (I) centers. Dinuclear complex [Cu2(2-pina)3(NCCH3)4](SbF6)2 (3) and tetranuclear cluster [Cu4(2pina)4(NCCH3)4](SbF6)4 (4) were afforded from the same reaction in CH3CN and DMF, respectively. Detailed analysis of X-ray crystal structures and Hirshfeld surface maps suggested that the weak noncovalent interactions and reaction solvents imposed a collaborative effect on the structural conformations of the Cu(I) complexes. The electrochemical properties of the complexes were investigated by cyclic voltammetry. Computational studies focusing on complex 1 revealed its unique electronic structure that corresponds to the interesting spectroscopic and electrochemical properties. Ó 2016 Elsevier B.V. All rights reserved.

1. Introduction Bioinspired synthetic complexes modeling the functions of metalloproteins active sites have attracted great interest due to their novel structural features and potential catalytic activities [1]. Previous research work has demonstrated that the rational selection and design of the ligand platform played a critical role of controlling the activities and structures of these model complexes [1d,1e,2]. So far, the major ligand design strategies mainly focused on the manipulation of denticity, steric effect, coordination donors and electron properties of the ligand backbone. Recently, the effect of the secondary sphere coordination environment on the function and physical properties of the metal complexes received growing attention due to the fact that the local environment of a protein pocket often has a profound impact on the function of the active site [1a]. Therefore, in an effort to advance the modeling chemistry of metalloproteins, functional groups that can introduce weak noncovalent interactions (e.g. hydrogen bonds, p


p interactions) were incorporated into ligand platforms to provide an auxiliary ⇑ Corresponding authors. Fax: +1 (501)4503623. E-mail addresses: [email protected] (G.T. Rowe), [email protected] (L. Yang). http://dx.doi.org/10.1016/j.ica.2016.03.011 0020-1693/Ó 2016 Elsevier B.V. All rights reserved.

environment to enhance the functions. For instance, a series of preorganized tripodal ligands with H-bond donors and acceptors have been reported to mimic the secondary coordination spheres of the catalytic sites of some iron and manganese enzymes [3]. The interesting chemistry from these studies has significantly contributed to the current understanding about the reactivity of model complexes and biological function of metalloproteins. As part of our ongoing work on synthesizing biologicallyrelated copper complexes with novel structures, spectroscopic properties and catalytic activities, pyridylamide ligands were employed as the supporting platforms to achieve the dual functions: (a) the pyridyl groups act as the main coordination sites to mimic the histidine donors in the active sites of copper containing proteins; (b) amide groups promote the formation of inter- or intramolecular weak interactions to simulate the outer-sphere coordination environment. Our previous research has demonstrated the versatile coordination behavior of the pyridylamide ligands with divalent transition metal ions [4]. X-ray crystallography and spectroscopic characterization showed that pyridyl nitrogens were excellent coordination donors, meanwhile the –NH–C (O)– amide groups produced diverse weak interactions in the local environment by using either H-bond donor (N–H group) or

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

acceptor (C@O group). Generally, the neutral –NH–C(O)– may use the C@O oxygen as the additional coordination donor, and the N–H often forms intra- or intermolecular hydrogen bond interactions. In the deprotonated form, the amidate –N–C(O)– has an anionic nitrogen with better r-donating abilities that can potentially increase the nuclearity of the metal complex and significantly change the architecture of the metal complexes [4b]. Therefore, by adjusting the functional groups of the ligands and reaction conditions, we were able to control the structures of the complexes and fine tune the geometry of the metal center(s) to obtain the desired secondary coordination sphere. This tunability makes pyridylamide ligands serve as the ideal systems for our work on examining the properties associated with the outersphere coordination environment in copper complexes. Herein we report the synthesis and structural characterization of new Cu(I) complexes supported by a group of pyridyl-secondary amide ligands (Scheme 1). Combined analysis based on X-ray crystal structures, Hirshfeld surface maps and theoretical studies demonstrated how weak coordination bonds, hydrogen bonds and p


p interactions impact the ligand binding modes, coordination environments and spectroscopic properties of the copper complexes. 2. Experimental 2.1. Physical measurements 1

H NMR spectra were recorded on Bruker AVANCE 300 MHz spectrometer at room temperature. Chemical shifts (d) were referenced to residual solvent signal. Elemental analyses were carried out by Atlantic Microlabs, Norcross, GA. The UV–Vis spectrum was recorded on a Cary 50 spectrometer. FT-IR spectra were collected on a Nicolet Magna 560 FT-IR spectrometer with an ATR attachment. Cyclic voltammograms were measured using EG&G Princeton Applied Research Scanning Potentiostat with a threeelectrode cell (Pt disk working electrode, Pt wire auxiliary electrode and aqueous Ag/AgCl glass reference electrode). All measurements were performed in either DMF or CH3CN degassed solution containing 1 mM analyte and 0.1 M tetrabutylammonium tetrafluoroborate at room temperature with the N2 protection. Recrystallized ferrocene was used as the internal standard. 2.2. Chemicals All reagents were obtained from commercial sources and used as received. Ligand N-2-acetamidopyridine (Haap), N-(4-pyridyl) picolinamide (4-ppa), N-(3-pyridyl)nicotinamide (3-pna) and N(2-pyridyl)isonicotinamide (2-pina) were synthesized according to literature [5]. 2.3. Synthesis of the metal complexes 2.3.1. [Cu2(aap)2]n (1) CuCl (0.036 g, 0.37 mmol) was added to a stirring mixture of Haap (0.050 g, 0.38 mmol) and NaH (0.011 g, 0.38 mmol) in 2 mL CH3CN. The yellow suspension was stirred for 1 h and 20 mL of

151

Et2O was added to the mixture. The solid was extracted into 2 mL DMF and the suspension was filtered. Vapor diffusion of Et2O into the yellow filtrate at room temperature led to the formation of yellow crystals suitable for X-ray crystallographic characterization (0.051 g, 68% yield). Anal. Calc. for C14H14N4O2Cu2: C, 42.31; H, 3.55; N, 14.10. Found: C, 42.78; H, 3.56; N, 14.24. FT-IR (cm1): 1604, 1557, 1472, 1430, 1359, 1323, 1288, 1264, 1168, 1124, 1053, 1021, 1003, 976, 872, 783, 745, 692, 619, 565, 530, 427. 1H NMR (DMSO-d6, 298 K): d = 8.38 (m, 2H), 8.03 (m, 2H), 7.60 (m, 2H), 6.86 (m, 2H), 2.22 (s, 6H) ppm. UV–Vis [DMF, kmax, nm (e, M1 cm1)]: 323 (4800). 2.3.2. [Cu(3-pna)3(NCCH3)](HF)(SbF6) (2) A solution of [Cu(NCCH3)4]SbF6 (0.039 g, 0.084 mmol) in 1 mL CH2Cl2 was added to a stirred solution of 3-pna (0.050 g, 0.25 mmol) in 1 mL CH2Cl2. The yellow suspension was stirred for 6 h and 15 mL of Et2O was added to the mixture. The resulting orange powder was collected and washed with Et2O (5 mL  3). The powder was extracted into 2 mL CH3CN and the turbid solution was filtered. Vapor diffusion of Et2O into the light yellow filtrate at room temperature led to the formation of yellow crystals suitable for X-ray crystallographic characterization (0.048 g, 60% yield). Anal. Calc. for C35H31N10O3F7SbCu: C, 43.88; H, 3.26; N, 14.62. Found: C, 43.62; H, 3.59; N, 14.71. FT-IR (cm1): 2366, 1678, 1628, 1598, 1584, 1544, 1484, 1468, 1437, 1367, 1350, 1310, 1240, 1177, 1157, 1050, 912, 888, 782, 727, 696, 635, 629, 552, 537, 523, 465, 433, 416. 1H NMR (DMSO-d6, 298 K): d = 11.05 (s, 3H), 8.41 (m, 4H), 8.22 (m, 5H), 7.84 (m, 5H), 7.65 (m, 5H), 7.22 (m, 5H), 2.04 (s, 3H) ppm. 2.3.3. [Cu2(2-pina)3(NCCH3)4](SbF6)2 (3) A solution of [Cu(NCCH3)4]SbF6 (0.079 g, 0.17 mmol) in 1 mL CH2Cl2 was added to a stirring solution of 2-pina (0.050 g, 0.25 mmol) in 1 mL CH2Cl2. The solvent was removed under reduced pressure after 1 h and the resulting orange solid was washed with Et2O (5 mL  3). The solid was extracted into 2 mL CH3CN and the turbid solution was filtered. Vapor diffusion of Et2O into the light yellow filtrate at room temperature led to the formation of yellow crystals suitable for X-ray crystallographic characterization (0.059 g, 52% yield). Anal. Calc. for C41H39N13O3F12Sb2Cu2: C, 36.20; H, 2.89; N, 13.38. Found: C, 36.44; H, 2.50; N, 13.26. FT-IR (cm1): 2357, 1686, 1600, 1576, 1523, 1496, 1464, 1433, 1417, 1369, 1305, 1262, 1240, 1155, 1124, 1097, 1062, 1003, 898, 856, 777, 757, 693, 657, 623, 545, 519, 453. 1H NMR (DMSO-d6, 298 K): d = 11.14 (s, 3H), 8.20 (m, 12H), 7.88 (m, 5H), 7.20 (m, 7H), 2.03 (s, 12H) ppm. 2.3.4. [Cu4(2-pina)4](SbF6)4 (4) A solution of [Cu(NCCH3)4]SbF6 (0.116 g, 0.25 mmol) in 1 mL DMF was added to a stirred solution of 2-pina (0.050 g, 0.25 mmol) in 1 mL DMF. The red orange solution was stirred for 4.0 h and 20 mL of Et2O was added to the mixture. The resulting orange solid was washed with Et2O (5 mL  3). Dissolving the product in DMF and vapor diffusion of Et2O into the clear red orange solution at room temperature led to the formation of red orange crystals suitable for X-ray crystallographic characterization (0.106 g, 72%

Scheme 1. Pyridylamide ligands Haap, 3-pna, 2-pina and 4-ppa applied in this work.

152

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

yield). Anal. Calc. for C44H36N12O4F12Sb4Cu4: C, 29.92; H, 2.05; N, 9.52. Found: C, 30.38; H, 1.95; N, 9.25. FT-IR (cm1): 1649, 1590, 1537, 1471, 1437, 1387, 1361, 1311, 1286, 1254, 1158, 1098, 1061, 858, 781, 694, 655, 628, 537, 451, 418. 1H NMR (DMSO-d6, 298 K): d = 11.15 (s, 4H), 8.19 (m, 10H), 8.01 (m, 10H), 7.92 (m, 12H) ppm. 2.3.5. [Cu(4-ppa)(NCCH3)2]BF4 (5) A solution of [Cu(NCCH3)4]BF4 (0.039 g, 0.13 mmol) in 1 mL CH3CN was added to a stirring solution of 4-ppa (0.050 g, 0.25 mmol) in 1 mL CH3CN. The yellow solution was stirred for 1.5 h and the 15 mL Et2O was added to the mixture. The resulting yellow powder was washed with Et2O (5 mL  3). Dissolving the product in CH3CN and vapor diffusion of Et2O into the clear yellow solution at room temperature led to the formation of yellow crystals suitable for X-ray crystallographic characterization (0.048 g, 59% yield). Anal. Calc. for C26H24N8O2F4BCu: C, 49.50; H, 3.83; N, 17.76. Found: C, 49.67; H, 3.48; N, 17.49. FT-IR (cm1): 2362, 1709, 1642, 1611, 1588, 1554, 1508, 1475, 1437, 1324, 1279, 1201, 1025, 997, 804, 757, 691, 620, 606, 584, 521, 474, 450, 430. 1 H NMR (DMSO-d6, 298 K): d = 11.07 (s, 2H), 8.48 (m, 5H), 7.90 (m, 6H), 7.27 (m, 5H), 2.04 (s, 6H) ppm. 2.4. X-ray crystallography X-ray crystallographic data were collected on crystals with dimensions of 0.40  0.14  0.14 mm for 1, 0.44  0.24  0.16 mm for 2, 0.75  0.06  0.02 mm for 3, 0.58  0.23  0.16 mm for 4, 0.30  0.180  0.17 mm for 5. Data were collected at 100 K using a diffractometer with a Bruker APEX CCD area detector [6] and graphite-monochromated Mo Ka radiation (k = 0.71073 Å). All five structures were solved by direct methods and refined by full-matrix least-square methods on F2 [7]. The crystal parameters, data collection and refinement are summarized in Table 1. The selected bond lengths and angles of complexes 1–5 are listed in Table 2. Non-hydrogen atoms were refined with anisotropic displacement parameters. The severely disordered solvent molecules in complexes 2 and 4 were modeled using the Squeeze program [8]. 2.5. Hirshfeld surface analysis In order to investigate the important intra- and intermolecular interactions within or among the coordination units, the Hirshfeld surface analysis was performed by using Crystal Explorer (version

3.0) software [9]. The calculation was based on the CIF files of the X-ray crystal structures. 2.6. Computational studies All DFT calculations of complex 1 were carried out using Orca 3.0.3 for Linux [10]. Molecular orbitals were visualized using IboView for Windows [11]. The solid-state structure of 1 was adapted into a monomeric form by replacing the intermolecular amide-carbonyl ligands of Cu1 with DMF amide-carbonyl groups. Geometry optimizations were carried out at the PBE/def2-TZVP level of theory [12,13] with Grimme’s dispersion correction (D3BJ) [14] and the conductor-like screening model (COSMO) [15] for DMF included (Table S1–S3). To account for BSSE all calculations employed the geometrical counterpoise correction (gCP) [16]. For each optimized structure, the assignment of an energetic minimum was confirmed through numerical frequency calculations (i. e., no imaginary frequencies). Unrestricted natural orbital analysis was employed for visualization of the unpaired spin orbital of the 1-electron oxidized form of 1. The electronic spectrum of 1 was predicted with TDA-TD-DFT calculations using the PBE-optimized geometry at the B3LYP/ def2-TZVP level of theory [17] with the N,N0 -dimethylformamide solvation environment modeled using the COSMO solvation model (Table S4) [15]. Transition energies were not corrected for nonequilibrium solvation effects. Visualization of selected electronic excitations was achieved through the use of natural transition orbitals [18]. Natural transition orbitals allow an excited state with significant contributions from several electronic transitions to be represented as a pair of orbitals that can be interpreted as the hole and particle associated with that excited state. 3. Results and discussion 3.1. Synthesis and general characterization Compound 1 was prepared from the reaction of CuCl with the ligand Haap deprotonated by NaH in CH3CN, followed by the recrystallization of the product in DMF. The reactions of Cu(I) salts with the neutral ligand Haap only provided uncharacterizable products. Complexes 2 and 5 with unexpected mononuclear structures were prepared from the reactions of neutral ligands with [Cu(NCCH3)4]X (X = SbF6 for complex 2 and BF4 for complex 5) in CH3CN. It seemed that reaction stoichiometry involving ligand and Cu(I) salt had no effect on the structures of the two complexes. Serious disproportionation happened during the reactions in DMF,

Table 1 Summary of crystal data and refinement parameters for complexes 1–5.

Formula Fw T (K) Space group a (Å) b (Å) c (Å) a (°) b (°) c (°) Z V (Å3) qcalc (g/cm3) l (mm1) R1 [I > 2r(I)] wR2 [I > 2r(I)] Goodness-of-fit (GOF) on F2

1

2

3

4

5

C14H14N4O2Cu2 397.34 100(2) C2/c 10.676 (4) 14.793(6) 9.959(4) 90 115.738(5) 90 4 1416.8(10) 1.863 3.013 0.0178 0.0516 1.005

C35H31N10O3F7SbCu 998.03 100(2) P21/n 17.0675(10) 8.6294(5) 27.5407(17) 90 106.231(2) 90 4 3894.6(4) 1.702 1.327 0.0617 0.1676 1.001

C41H39N13O3F12Sb2Cu2 1360.43 100(2)  P1

C59H71Cu4 F24N17O9Sb4 2359.48 100(2)  P1

7.7092(8) 18.1869(19) 19.716(2) 116.206(2) 96.719(2) 93.887(3) 2 2440.9(4) 1.851 2.054 0.0482 0.1068 1.006

11.0043(8) 20.5642(15) 20.6321(15) 119.718(2) 93.178(2) 90.258(3) 2 4045.9(5) 1.937 2.461 0.0623 0.1776 0.994

C28H27B CuF4N9O2 671.93 100(2) P21/n 7.4162(7) 19.6203(19) 20.753(2) 90 92.635(2) 90 4 3016.5(5) 1.480 0.792 0.0521 0.1507 0.999

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

and no characterizable products were isolated. Complexes 3 and 4 were synthesized through the similar reactions between neutral 2pina and [Cu(NCCH3)4]SbF6, but the application of different solvents afforded very different structures. All six complexes were air sensitive and turned to green or blue powder in a few days when exposed to air. 3.2. X-ray crystal structures and Hirshfeld surface analysis In our previous work, the ligand Haap showed a bidentate coordination mode, in which the pyridyl nitrogen and amide oxygen chelated to the same metal center to form a mononuclear complex, meanwhile the N–H group acted as a H-bond donor [4a]. In this work, the application of a strong base (NaH) deprotonated the N–H group so the nitrogen in the amidate –N–C(O)– group joined the coordination sphere of the Cu(I) center due to its strong r donating effect. The X-ray crystal structure of complex 1 presents a stair-like one-dimensional chain architecture. The left picture

Table 2 Selected bond lengths (Å) and angles (°) of complexes 1–5. Complex 1 Cu1B–N1B Cu1B0 –N2B Cu1B


O1A0 Cu1B


Cu1B0 N1B–Cu1B–N1B0 N1B–Cu1B–O1A0 N1B0 –Cu1B–O1A0 N2B–Cu1B0 –N2B0

1.9392(12) 1.8931(12) 2.448(2) 2.490 166.90(6) 91.70(6) 98.36(7) 179.33(6)

Cu1B–N1B0 Cu1B0 –N2B0 Cu1B


O1C

1.9392(12) 1.8931(12) 2.448(2)

O1A0 –Cu1B–O1C N1B–Cu1B–O1C N1B0 –Cu1B–O1C

79.82(7) 98.36(7) 91.70(4)

2.023(3) 2.024(3) 103.81(13) 104.98(14) 105.89(14)

Cu1–N3B Cu1–N1D N3A–Cu1–N3C N3B–Cu1–N3C N3C–Cu1–N1D

2.071(3) 2.033(3) 126.97(13) 106.79(13) 106.80(14)

2.39(6) 2.34

N1C–H


F1

2.10(5)

2.049(3) 2.029(4) 2.057(3) 1.977(3)

Cu1–N3C Cu1–N1F Cu2–N3B Cu2–N1E

2.049(3) 1.942(3) 2.057(4) 1.964(4)

116.48(13) 119.63(14) 106.69(13) 107.55(13) 109.94(14) 106.83(14)

N1A–Cu1–N1G N3C–Cu1–N1G N1G–Cu1–N1F N3A–Cu2–N1D N3B–Cu2–N1D N1D–Cu2–N1E

102.61(15) 96.94(14) 112.10(14) 105.43(14) 103.68(14) 122.50(14)

Complex 4 Cu1–N1A Cu1–O1A Cu2–N3A Cu3–N3B Cu3–O1C Cu4–N1D Cu1


Cu2 Cu3


Cu4 N1A–Cu1–N3D N3D–Cu1–O1A N3A–Cu2–O1B N3B–Cu3–N1C N3B–Cu3–O1C N3C–Cu4–O1D

1.901(5) 2.135(4) 1.897(5) 1.899(5) 2.171(4) 1.920(5) 8.917 8.898 165.3(2) 103.39(17) 103.87(17) 165.0(2) 104.96(18) 102.42(19)

Cu1–N3D Cu2–N1B Cu2–O1B Cu3–N1C Cu4–N3C Cu4–O1D Cu2


Cu3 Cu4


Cu1 N1A–Cu1–O1A N3A–Cu2–N1B N1B–Cu2–O1B N1C–Cu3–O1C N3C–Cu4–N1D N1D–Cu4–O1D

1.906(4) 1.913(5) 2.170(4) 1.926(5) 1.897(5) 2.158(4) 8.919 8.911 90.70(17) 167.0(2) 89.10(17) 89.96(18) 165.1(2) 90.84(18)

Complex 5 Cu1–N3A Cu1–N1C N3A–Cu1–N3B N3A–Cu1–N1D N3B–Cu1–N1D

2.057(2) 1.980(2) 109.72(8) 107.79(9) 106.23(9)

Cu1–N3B Cu1–N1D N3A–Cu1–N1C N3B–Cu1–N1C N1C–Cu1–N1D

2.069(2) 1.962(2) 102.94(8) 105.34(9) 124.35(9)

Complex 2 Cu1–N3A Cu1–N3C N3A–Cu1–N3B N3A–Cu1–N1D N3B–Cu1–N1D H-bonds F1–H


N1B C11–H


F1 Complex 3 Cu1–N1A Cu1–N1G Cu2–N3A Cu2–N1D Cu1


Cu2 N1A–Cu1–N3C N1A–Cu1–N1F N3C–Cu1–N1F N3A–Cu2–N3B N3A–Cu2–N1E N3B–Cu2–N1E

153

in Fig. 1 illustrates the chain structure formed by three dimeric units A, B and C that are connected with each other through long Cu–OC@O bonds (2.448 Å, dashed lines). These weak inter-unit bonds cause a severe distortion of the ligand–metal moiety characterized by a large N1B0 –Cu1B–Cu1B0 –N2B dihedral angle of 15.5°. The vertical distance between the two neighboring units is about 3.417 Å. The right picture of Fig. 1 shows the core structure of one repeating dimeric unit (unit B). In each dimeric unit, two deprotonated ligands sandwich two Cu(I) atoms through Npy and Namidate donors in a trans fashion, while the two C@O groups extend to the coordination sphere of two other dimeric units (units A and C) through weak Cu–OC@O bonds (2.448 Å). The pyridylamidate motif has a l2–j2(N,N) coordination mode, which is very rare in pyridylamide ligand systems. The Cu1B0 has a linear geometry with a N2B–Cu1B0 –N2B0 angle of 180° and a Cu–N distance of 1.8931 (12) Å. The Cu1B shows an interesting seesaw geometry when the Cu


Cu interaction is ignored. Two pyridyl nitrogens occupy the axial positions with a N1B–Cu1B–N1B0 bond angle of 166.90 (6)° and the copper center is pivoted by two OC@O donors from unit A (above) and unit C (below) with an O1A0 –Cu1B–O1C bond angle of 79.82(7)°. The weak Cu–OC@O bond (2.448 Å) is significantly longer than Cu–Npy bond (1.939 Å), possibly due to the structural constraint of the binuclear entity. The calculated four-coordinate structural index s4 of Cu1B is 0.67, and it confirms the seesaw geometry with a C2v symmetry [19]. The Cu


Cu distance (2.490 Å) is shorter than the sum of copper covalent radii (2.64 Å) [20], indicating the weak interaction between the copper centers. Complex 2 shows a mononuclear structure with a HF guest molecule embraced by three neutral 3-pna ligands (Fig. 2). The HF molecule may come from the decomposition of SbF 6 anions with trace amount of water [21]. The tetrahedral geometry of the Cu(I) center is fulfilled by three N3py donors and one CH3CN molecule with an average Cu–N distance of 2.038 Å. The structural index s4 of Cu1 is 0.90, which is very close to the standard tetrahedral geometry (s4 = 1.00) [19]. Despite the availability of multiple coordination donors, the ligand 3-pna displays a monodentate coordination mode, in which the N3py is the only binding site to the copper center. The most interesting feature of complex 2 is the encapsulated HF molecule stabilized in the center of the oval shaped cage with three prominent longitudinal ridges formed by three neutral ligands. The fluorine atom is directed toward the copper center with a Cu


F distance of 4.554 Å, and the hydrogen H1 is directed toward the open side of the cage. This particular orientation is apparently facilitated by the multiple intermolecular hydrogen bond interactions shown in Fig. 2 as the dashed lines. The N2C–H group and H11C from one ligand interact with F1 through H


F1–H1 hydrogen bonds of 2.105 and 2.345 Å, respectively. On the opposite side, the pyridyl nitrogen donors N1A and N1B from two different ligands form hydrogen bonds with the H1 atom (N1A


H1–F1 = 2.425 Å and N1B


H1–F1 = 2.386 Å). Although it looks like the four hydrogen bonds observed in the X-ray structure seemingly will generate a rotation torque to tilt the HF molecule, the guest molecule stays aligned with the direction of the ridges. The reason is the weak attractions from the other hydrogen bond donors and acceptors to the HF molecule, including H11A


F1 (2.735 Å), H11B


F1 (2.968 Å) and N1C


H1 (2.555 Å). The Hirshfeld surface mapped with dnorm for HF guest molecule confirms the presence of these weak interactions (Fig. 2 middle and right). The circled red spots 1, 2, 3 (Fig. 2 middle: front view) and 4 (Fig. 2 right: back view) correspond to the hydrogen bonds C–H11C


F1, N2C–H


F, N1A


H1–F1 and N1B


H1–F1 observed in the X-ray crystal structure (Fig. 2 left). The color of the circled areas 5, 6 and 7 is much lighter than the observed strong hydrogen bonds, but still visible to suggest the presence of the

154

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

Fig. 1. X-ray crystal structures of the one-dimensional chain (left) and a single unit (right) of complex 1. Thermal ellipsoids are shown at the 50% level. Hydrogen atoms are omitted for clarity. The dotted lines represent the weak interaction between amide oxygen and Cu(I) center.

Fig. 2. X-ray crystal structure of complex 2 (left) and Hirshfeld surface of HF molecule mapped with dnorm (middle: front view; right: back view). Thermal ellipsoids are shown at the 50% level. Most hydrogen atoms are omitted for clarity. The dotted lines represent the hydrogen bond interactions.

aforementioned weak interactions. It is noteworthy that, although the N2A–H and N2B–H groups are available H-bond donors, no hydrogen bond interactions were observed between either of these two groups and the HF molecule. Instead, both groups form intermolecular hydrogen bonds with amide C@O acceptors from the neighboring mononuclear units. This may be attributed to the small size of fluorine atom that only allows the limited number of hydrogen bonds with the surrounding donors. The three ligand backbones in each mononuclear unit maintain a nearly coplanar conformation, partially facilitated by the intramolecular hydrogen bond interactions between the C@O group and one of the C–H groups on 2-pyridyl ring (C@O


H = 2.22  2.36 Å, Fig. 2 left). The coplanar conformation leads to the formation of interesting p


p stacking interactions shown in Fig. 3 left. Two lower ligands from the unit A stack on top of two ligands from units B and C respectively though the closest C


C distances of 3.009 and 3.327 Å. In the Hirshfeld surface mapped with shape index for the unit A (Fig. 3 right), these contacts are represented by four circled areas characterized by paired red and blue triangles [22]. The synthetic procedures of 3 and 4 were similar, but the solvent effects from CH3CN and DMF led to the formation of two

different Cu(I) clusters with distinct structural features. In the Xray crystal structure of 3 (Fig. 4 left), a horseshoe-like dinuclear structure with three neutral 2-pina ligands was observed. A bridging ligand connects Cu1 and Cu2 through a l2-N2py,N4py coordination mode that results in a Cu


Cu distance of 8.622 Å. The other two ligands bind to the two copper centers through the N4-py donors only, leaving the rest of backbone free. For each copper, the coordination sphere is fulfilled by two CH3CN molecules and two pyridyl nitrogen donors with an average Cu–N distance of 2.016 Å. Although the CH3CN molecules in the starting material [Cu(NCCH3)4]SbF6 are usually labile ligands, only two CH3CN molecules are replaced by ligand 2-pina. This is probably due to the CH3CN solvent used in the synthesis (see Section 3.3). The structural index s4 values of Cu1 and Cu2 centers are 0.88 and 0.90, suggesting tetrahedral geometries of the metal sites [19]. Interesting intramolecular hydrogen bonds were observed in the crystal structure of 3. The N2A–H group from the bridging ligand interacts with the coordinated N1FCH3CN donor through a H


N distance of 2.515 Å, causing a shorter Cu1–N1F bond length (1.942(3) Å) than the other three Cu1–N bonds (2.042 Å). Meanwhile, the O1AC@O atom acts as the H-bond acceptor to interact with the C4A–H group through a H


O distance of 2.237 Å.

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

155

Fig. 3. p


p interactions among the three mononuclear units A, B and C of complex 2 (left) and Hirshfeld surface mapped with shape index for unit A (right). The four circled red and blue triangle pairs correspond to the p


p interactions generated by the two lower co-planar ligands in unit A. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4. X-ray crystal structures of 3 (left) and 4 (right). Thermal ellipsoids are shown at the 50% level. Most hydrogen atoms are omitted for clarity.

The third intramolecular hydrogen bond occurs between the O1BC@O atom and C4B–H group on the terminal 2-pina ligand of the Cu2 center (H


O distance = 2.225 Å). The presence of these interactions helps the two ligand backbones maintain a nearly coplanar conformation with a distortion angle of 17.1° between the two pyridyl rings. However, the terminal 2-pina ligand on the Cu1 is severely distorted along the amide linker, leading to the formation of a large torsion angle of 70.0° between the two pyridyl groups. The X-ray crystal structure of the tetracationic metallo-macrocycle of complex 4 is depicted by the right picture in Fig. 4. It shows a novel co-planar square structure with four ligands connecting with four Cu(I) centers alternately. The four neutral 2-pina ligands display the same l2-j2(N2-py,OC@O) coordination mode, in which the N2-py donor and amide C@O group chelate to one Cu(I) center, and the N4-py donor coordinates to another Cu(I) in a perpendicular fashion to form one of the four edges of the square. The average Cu


Cu distance is around 8.911 Å. The Cu(I) centers have the same N2O coordination environment with s3 values (three-coordinate structural index) in a range of 0.215–0.250 [23], suggesting distorted T-shape geometries of the Cu(I) sites. It is noteworthy to point out that the average Cu–O bond distance (2.158 Å) is significantly larger than the average Cu–N bond

(1.907 Å). This difference may be attributed to the structural restraint of the six-member chelating ring involving the Cu(I) center. The distance between the outer vertices is about 15.990 Å, while the cavity formed by the cluster has a size of 8.044  8.044 Å. As a result of the coplanar conformation, the tetranuclear metallo-macrocycle complex shows an interesting packing structure. Fig. 5 illustrates the packing architecture among three pairs (pairs 1, 2 and 3) of tetranuclear units. For each pair (e.g. pair 1), the two units overlap with each other in a nearly perfect face-to-face fashion with an intra-pair vertical distance of 3.247 Å. Meanwhile, both units in pair 1 partially overlap with the lower unit of pair 2 and the upper unit of pair 3 separately, through an inter-pair vertical distance of 2.947 Å. In order to obtain a better understanding of interactions among the units, Hirshfeld surface analysis was performed for one tetranuclear cluster. The Fig. 6 shows the front and back views of the Hirshfeld surface mapped with the shape index. The characteristic red and blue triangle pairs in the circled areas indicate the presence of multiple p


p interactions [21]. The full overlap of the intra-pair units leads to better contacts shown in 9 circles (Fig. 6 left), and the partial overlap of the inter-pair units affords three p


p interacting sites (Fig. 6 right).

156

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

Fig. 5. Packing structure of 4. Thermal ellipsoids are shown at the 50% level. Most hydrogen atoms are omitted for clarity.

Fig. 6. Hirshfeld surface of 4 mapped with shape index (left: front view; right: back view).

is slightly distorted with a torsion angle of 23.42° between the two pyridyl groups. The mononuclear unit also features some interesting intermolecular hydrogen bonds. The left picture in Fig. 8 shows the one-dimensional supramolecular chain architecture. The key factor for this conformation is the connecting function of the intermolecular hydrogen bonds involving C@O, N–H and C–H groups among the mononuclear units (black circle). The Hirshfeld surface of the mononuclear unit mapped with dnorm (Fig. 8 right) clearly shows the corresponding circled red dots at the positions of the two hydrogen bond donors. These effective interactions may block the coordination of the C@O group and might also contribute to the monodentate mode of ligand 4-ppa in complex 5. 3.3. Coordination modes of ligands

Inspired by the novel structures of complexes 3 and 4 based on ligand 2-pina, we prepared ligand 4-ppa, in which the positions of N–H and C@O groups on the amide linker were exchanged. We had hoped the ligand 4-ppa will instill new structural features in copper complexes. To our surprise, only a mononuclear complex was afforded from the reaction of 4-ppa with Cu(I) salt in CH3CN. Reactions with different ligand to metal ratios and in different solvents, such as DMF and acetone, failed to give any characterizable products. Fig. 7 shows the X-ray structure of this compound, in which the 4-pyridyl nitrogen donor is the only coordination site. The tetrahedral geometry of the Cu(I) center has a s4 value of 0.89 [19] and average Cu–N distance of 2.017 Å. One of the ligands has a coplanar conformation on the backbone, but the other one

Fig. 7. X-ray crystal structure of 5. Thermal ellipsoids are shown at the 50% level. Most hydrogen atoms are omitted for clarity.

Three coordination modes have been observed for ligand Haap in our work so far (Scheme 2). The monodentate mode jNpy was observed in a paddle wheel Cu(II) complex prepared from a reaction of Haap with Cu(OAc)2 in DMF [4b]. When the neutral Haap reacted with Zn(II) salt, a chelating mode (jNpy,OC@O) with pyridyl nitrogen and amide oxygen as the coordination donors was observed from the X-ray crystal structure recrystallized from MeOH [4a]. In both cases, the N–H group stays free to form hydrogen bonds with coordinated acetate or triflate anion. In this work, an interesting l3-(jNpy;jNamidate;jOC@O) mode was observed in the 1-D chain structure of complex 1. This is due to the coordination of amidate groups to the copper centers after the treatment of Haap with NaH. Although the same l3-(jNpy;jNamidate;jOC@O) mode of ligand Haap was proposed for Pd(II) complexes based on spectroscopic evidence [24], complex 1 remains as the first X-ray characterized example with such a coordination mode. Although the complexes 2–5 were prepared from the similar treatment of [Cu(NCCH3)4]X (X = BF4 or SbF6) with the neutral ligands, the diverse structural features observed from the X-ray crystallography characterization are particularly intriguing to us. The analysis based on the structures of these complexes revealed that the solvent and hydrogen bond interactions could be important factors to influence the coordination modes of ligands 3-pna, 2-pina and 4-ppa. Table 3 shows the coordination modes observed (blue) and expected but not observed (red) of the three ligands. For ligands 3-pna and 4-ppa, only the monodentate coordination mode (jNpy) was observed in this work, in which the 3-pyridyl (3-pna) and 4-pyridyl (4-ppa) nitrogens are the sole coordination donors in complexes 2 and 5. The attempts to isolate products with

157

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

Fig. 8. Intermolecular hydrogen bonds among the mononuclear units (left) of 5 and Hirshfeld surface mapped with dnorm (right).

Scheme 2. Coordination modes of the neutral ligand Haap (left and middle) and the anionic ligand app- (right).

l2-(jNpy;jNpy) and l2-(jNpy,OC@O;jNpy) modes were unsuccessful. For ligand 2-pina, all three modes were observed from the Xray structures of complexes 3 and 4, depending on the solvents employed. From the observed coordination modes of the three ligands, the C@O group remains free when CH3CN is used, and it acts as an efficient intramolecular hydrogen bond acceptor. This behavior may

be attributed to the preference of soft Cu(I) ions towards CH3CN solvent molecules with softer Lewis basicity over the harder C@O groups from the ligands. Therefore, the ligands exhibit a monodentate coordination mode in complexes 2, 3 and 5. Interestingly, when DMF was used, the chelating-bridging mode l2-(jNpy,OC@O; jNpy) was observed (complex 4), in which the 2-pyridyl nitrogen and C@O group bite to the Cu(I) and the 4-pyridyl nitrogen bridges to another Cu(I). This is possibly due to two reasons: (a) unlike CH3CN, the DMF molecule is a less competitive ligand for Cu(I) due to the hard Lewis basicity of oxygen, therefore the chelate effect from the ligand provides greater structural stability; (b) the hydrogen bond interactions imposed by the DMF molecules with the amide N–H group may reinforce the chelating mode by pulling the hydrogen away from the Cu(I) center. Indeed, the X-ray crystal structure of 4 shows the N–H groups of the tetranuclear cluster interacting with surrounding non-coordinated DMF molecules through strong intermolecular hydrogen bonds (N–H


O = 1.95  2.01 Å). It is noteworthy to mention that the same l2-(jNpy,OC@O;jNpy) has also been observed in a polymeric

Table 3 Coordination modes (blue: observed; red: unobserved) of ligands 3-pna, 2-pina and 4-ppa. Ligands

Coordination modes (blue: observed; red: expected but not observed)

jNpy

l2-(jNpy;jNpy)

l2-(jNpy,OC@O;jNpy)

3-pna 2 (CH3CN)

2-pina 3 (CH3CN)

3 (CH3CN) 4 (DMF)

4-ppa 5 (CH3CN)

158

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

Cu(II) structure supported by ligand 2-pina, which was prepared from the reaction in DMF [4c]. The conformation of ligand 4-ppa in complex 5 is quite unique among the three ligands with the monodentate coordination modes (jNpy). In addition to the intramolecular hydrogen bond between C@O and C–H groups, a second intramolecular hydrogen bond between 2-pyridyl nitrogen and N–H group was also observed in X-ray crystal structure (Fig. 8 left). Apparently this ‘‘locked” conformation makes the monodentate mode of 4-ppa even more favored than the desired l2-(jNpy;jNpy) bridging mode or l2-(jNpy,OC@O;jNpy) chelating-bridging mode in the reaction with Cu(I) in CH3CN. Although we cannot compare complex 5 with any Cu(I) products from the reactions in DMF due to the serious disproportionation problem, the structure of 5 still indicates the neutral 4-ppa ligand may not be a good candidate for the construction of polynuclear metal complexes. 3.4. Electrochemistry The electrochemical properties of the Cu(I) complexes were investigated by cyclic voltammetry (CV) in dry and degassed organic solvents with the protection of N2 gas. No redox waves were observed from ligands under the same experimental

conditions. Recrystallized ferrocene was used as an internal standard and tetrabutylammonium tetrafluoroborate was used as the supporting electrolyte. Complex 1 showed a reversible couple at E1/2 = 0.067 V (Epa = 0.142 V, Epc = 0.008 V, Fig. 9 left) with a peak current ratio ipa/ipc = 1.20, suggesting a chemically reversible redox process. A second redox couple, which is partially reversible, was observed at E1/2 = 0.502 V (Epa = 0.652 V, Epc = 0.352 V, Fig. 9 right) with an ipa/ipc value of 1.55. It has been established that a Cu(I) center with a linear geometry has a better resistance towards oxidation [25]. Therefore, the wave at E1/2 = 0.502 V is assigned as the Cu(I)/Cu (II) couple of the Cu1B0 center with a linear geometry (Fig. 1 right). For the Cu1B center, the weak coordination of two C@O groups may stabilize the Cu(II) center during the oxidation, causing a lower potential at E1/2 = 0.067 V. In addition, the better reversibility of Cu1B than Cu1B0 can also be explained by the geometry difference of the two metal centers. The see-saw geometry of the four-coordinated Cu1B requires less geometry adjustment to accommodate both +1 and +2 oxidation states, but the oxidation of the linear Cu(I) center to Cu(II) may lead to a significant change of the coordination environment, which is less favored for the reduction. The theoretical calculation supports our assignment and rationalization (see Section 3.5).

Fig. 9. Two redox couples at E1/2 = 0.067 V (left) and E1/2 = 0.502 V (right) in DMF. Scanning rate = 100 mV/s.

Fig. 10. Variation of peak current with the scan rate (left) and plot of ipc vs. the square root of scanning rate (right) of complex 1.

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

In order to obtain a better understanding of the reversible behavior at E1/2 = 0.067 V, multiple scans at different scanning rates were preformed and the ipc data were plotted against the square root of the scanning rate (Fig. 10). The DEp values slightly increased with the increased scanning rate, which has also been observed for the redox couple of internal standard ferrocene/ferrocenium (Fc/Fc+), suggesting the redox behavior of complex 1 is electrochemically reversible. The data fitting for the plot of ipc against the square root of the scanning rate shows a straight line with an R2 value of 0.9999, suggesting a diffusion-controlled pathway for the reduction/oxidation of the copper center. Complex 2 shows two related peaks in degassed CH3CN in the 2.0  +2.0 V window (Fig. S1). The oxidation peak at 0.22 V is assigned as the oxidation of the Cu(I) center to Cu(II). This process may lead to the coordination of CH3CN solvent molecules, leading to a geometry change to stabilize the metal center with +2 oxidation state. Therefore, the reduction peak at 0.79 V comes from the reduction of the solvent-derived species. For complex 3, a large anodic peak was observed around 0.2 V in CH3CN within the 1.0  0.2 V window (Fig. S2 left). Scanning the compound in the same conditions within a smaller window (0.7  +0.2 V) didn’t show such a peak. Therefore, the process might be attributed to the deposition of Cu(0) generated from the reduction of the Cu(I) species. In fact, a thin brown Cu(0) film was indeed observed on the surface of working electrode after the measurements. Complexes 4 and 5 showed similar reduction and deposition behavior in DMF and CH3CN respectively (Fig. S2 right and Fig. S3).

159

to achieve wavefunction convergence, the dicupric form of 1 required the addition of two DMF ligands to Cu2 (vide infra). Electrochemical studies of 1 showed that it could be oxidized in two steps, presumably both one-electron processes. Because the two Cu(I) centers in 1 have different coordination numbers and geometries, it is natural to wonder if their differences cause one of the metals to oxidize to the +2 state before the other. LFSE arguments (vide supra) point to the Cu1 with its higher coordination number and N2O2 coordination environment to be better able to stabilize the +2 oxidation state. Electronic structure calculations also support this assignment. UNO analysis of the one-electron oxidized form of 1 reveals that the unpaired electron resides in an orbital with a high degree of Cu1 antibonding dz2 character (Fig. 12). Mulliken analysis of the one-electron oxidized state

3.5. Computational studies Among the five Cu(I) complexes reported in this work, complex 1 exhibited interesting spectroscopic properties and electrochemical behaviors arising from its unique bimetallic structure. In an effort to gain further understanding of complex 1, we performed DFT calculations on the compound at the PBE/def2-TZVP level of theory. The solid-state structure of 1 includes weak coordination between Cu1 and the amide carbonyl oxygen atoms (Cu–O = 2.448 Å) from adjacent molecular units. To simplify the calculation and to better model the solution behavior, the amide carbonyl ligands have been replaced with two O-bound DMF molecules. The optimized geometry of this simplified structure can be seen in Fig. 11A, and it is a faithful representation of the solid-state structure. The calculated Cu1


Cu2 distance is contracted with respect to the crystal structure by only 0.016–2.474 Å, and the dihedral angle (N21–Cu1–Cu2–N22) between the copper atoms and their respective N-donor ligand atoms has increased from 15.54° to 28.68°. This increase in the twist angle is likely due to the lack of intermolecular interactions in our calculations. We also calculated optimized structures of the one- and two-electron oxidized forms of 1 that can be seen in Fig. 11B and C. In order

Fig. 11. DFT minimized geometries of 1 in (CuICuI)–(DMF)2 (A), (CuICuII)–(DMF)2 (B), and (CuIICuII)–(DMF)4 (C) forms. Cu


Cu distance is displayed on each structure. For clarity, hydrogen atoms are omitted in all structures, and DMF Nmethyl groups are not shown in C.

Fig. 12. Singly-occupied molecular orbital of the one-electron oxidized form of 1 from UNO analysis. Numbers in parenthesis indicate the Mulliken spin populations of each atom that has a contribution greater than 2%.

Fig. 13. TD-DFT predicted electronic spectrum of complex 1 with individual excitations (red lines) and by modeling each electronic transition as a Gaussian curve with a standard deviation of 0.4 eV (black line). Individual excitation contributions to the total absorbance indicated with vertical lines. Inset: the natural transition orbital hole-particle pair corresponding to the most intense electronic excitation at 312 nm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

160

E.P. McMoran et al. / Inorganica Chimica Acta 446 (2016) 150–160

shows that a plurality (43.4%) of the unpaired spin resides on Cu1, and inclusion of Cu1’s primary coordination sphere in the total increases this value to 72.6%. We encountered significant problems attempting to optimize the structure of dicupric 1 starting with the structure of one-electron oxidized 1. All attempts to obtain a wavefunction failed for this structure. Only by adding two DMF ligands to Cu2 were we able to achieve SCF convergence (Fig. 11C). This computational issue provides something of an explanation for the irreversibility of the second oxidation process in cyclic voltammetry experiments. The two-coordinate Cu2 atom is unable to support the divalent state without the stabilizing effect of additional DMF ligands. If the second oxidation of complex 1 follows a CE or EC mechanism in which the coordination number of Cu2 must increase through solvent ligand binding, this would lead to irreversible behavior in the CV. Complex 1 exhibits an intense absorbance in its UV–Vis spectrum at 323 nm with a molar absorptivity of 4800 M1 cm1 (Fig. S4). We sought to model this excitation with TD-DFT to determine its physical origin. The PBE functional employed for geometry optimizations did not perform well in this case, so we screened several functionals to find a suitable level of theory to predict the UV spectrum. The hybrid B3LYP functional performed the best in the TD-DFT calculations, and produced the predicted spectrum shown in Fig. 13. The predicted absorbance maximum (312 nm) matches the experimental value fairly well, and the predicted molar absorptivity (4322 M1 cm1) is also accurate within 10% of the actual value. The 6th excited state is the most intense, and was chosen for further analysis to determine the origin of the absorption. The natural transition orbital method [19] provides a way to simplify the interpretation of electronic excitations that involve linear combinations of several transitions between frontier orbitals. NTO analysis of the 6th excitation (Fig. 13, inset) reveals that it has a large degree of Cu-3d to pyridine p⁄ (MLCT) character, which is consistent with the high extinction coefficient. 4. Conclusions The pyridylamide ligands employed in this work exhibited diverse coordination modes towards Cu(I) centers. The anionic amidate nitrogen in the deprotonated Haap coordinates to the Cu (I) center as a strong r donor (complex 1), while the neutral N–H groups in complexes 2–5 act as H-bond donors to form intra- or intermolecular hydrogen bond interactions. Detailed analysis of complexes 1–5 demonstrated how weak non-covalent interactions and reaction conditions collaboratively impacted the coordination environments and structural properties of the complexes. One interesting observation was the supramolecular architecture stabilized by the p


p stacking interactions among the coplanar ligand backbones (complexes 2 and 4). These results enhanced our understanding on the correlation between the rational ligand design and the structural/functional properties of the metal complexes. Electrochemical studies showed the interesting reversible redox couples in complex 1, which could be attributed to the stepwise oxidation/reduction of the two Cu(I) centers with different coordination environments. Theoretical work suggested that these spectroscopic and electrochemical properties of complex 1 arise from its unique electronic structure featured by close Cu


Cu and strong Cu–ligand interactions. Acknowledgments This work was supported by the start-up fund provided by the College of Natural Sciences and Mathematics (CNSM) and

University Research Council Fund (URCF) from University of Central Arkansas, Arkansas Science & Technology Authority (ASTA) from the State of Arkansas. Appendix A. Supplementary material CCDC 1420197–1420201 contains the supplementary crystallographic data for complexes 1–5. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www. ccdc.cam.ac.uk/data_request/cif. Supplementary data associated with this article can be found, in the online version, at http:// dx.doi.org/10.1016/j.ica.2016.03.011. References [1] (a) S.A. Cook, E.A. Hill, A.S. Borovik, Biochemistry 54 (2015) 4167; (b) J. Serrano-Plana, I. Garica-Bosch, A. Company, M. Costas, Acc. Chem. Res. 48 (2015) 2397; (c) N. Gagnon, W.B. Tolman, Acc. Chem. Res. 48 (2015) 2126; (d) W.N. Oloo, L. Que Jr., Acc. Chem. Res. 48 (2015) 2612; (e) S. Hematian, I. Garcia-Bosch, K.D. Karlin, Acc. Chem. Res. 48 (2015) 2462. and references therein. [2] (a) E.A. Lewis, W.B. Tolman, Chem. Rev. 104 (2004) 1047; (b) A.L. Gavrilova Brice Bosnich, Chem. Rev. 104 (2004) 349. [3] (a) S.A. Cook, A.S. Borovik, Acc. Chem. Res. 48 (2015) 2407. and references therein; (b) E.M. Matson, J.A. Bertke, A.R. Fout, Inorg. Chem. 53 (2015) 4450; (c) H. Yamagishi, S. Nabeya, T. Ikariya, S. Kuwata, Inorg. Chem. 54 (2015) 11584; (d) J.M. Blacquiere, M.L. Pegis, S. Raugei, W. Kaminsky, A. Forget, S.A. Cook, T. Taguchi, J.M. Mayer, Inorg. Chem. 53 (2015) 9242; (e) Y.J. Park, E.M. Matson, M.J. Nilges, A.R. Fout, Chem. Commun. 51 (2015) 5310; (f) C.M. Moore, D.A. Quist, J.W. Kampf, N.K. Szymczak, Inorg. Chem. 53 (2014) 3278. [4] (a) E.P. McMoran, J.A. Goodner, D.R. Powell, L. Yang, Inorg. Chim. Acta 421 (2014) 465; (b) M.A. Pauly, E.M. Erwin, D.R. Powell, G.T. Rowe, L. Yang, Polyhedron 102 (2015) 722; (c) E.P. McMoran, C. Mugenzi, K. Fournier, M. Draganjac, D. Tony, K. Jeong, D.R. Powell, L. Yang, J. Coord. Chem. 69 (2016) 375. [5] (a) M. Kondo, A. Asami, H. Chang, S. Kitagawa, Cryst. Eng. 2 (1999) 115; (b) T.S. Gardner, E. Wenis, J. Lee, J. Org. Chem. 19 (1954) 753; (c) D.K. Kumar, D.A. Jose, P. Dastidar, A. Das, Langmuir 20 (2004) 10413; (d) C.B. Aakeröy, S.V. Panikkattu, B. DeGHaven, J. Desper, Cryst. Growth Des. 12 (2012) 2579; (e) M.F. El-Shazly, Inorg. Chim. Acta 40 (1980) 1. [6] (a) Data Collection: SMART Software Reference Manual, Bruker-AXS, 5465 E, Cheryl Parkway, Madison, WI 53711–5373, USA, 1998; (b) Data Reduction: SAINT Software Reference Manual, Bruker-AXS, 5465 E, Cheryl Parkway, Madison, WI 53711–5373, USA, 1998. [7] (a) G.M. Sheldrick, SHELXTL Version 6.10 Reference Manual, Bruker-AXS, 5465 E, Cheryl Parkway, Madison, WI 53711-5373 USA, 2000; (b) International Tables for Crystallography Volume C, Tables 6114, 4268 and 4242, Kluwer, Boston, 1995. [8] P. van der Sluis, A.L. Spek, Acta Cryst. A46 (1990) 194. [9] S.K. Wolff, D.J. Grimwood, J.J. McKinnon, M.J. Turner, D. Jayatilaka, M.A. Spackman. University of Western Australia, Available online at: www. hirshfeldsurface.net (accessed 09 July 2015). [10] F. Neese, Wiley Interdisciplinary Reviews: Computational Molecular Science 2 (2012) 73. [11] G. Knizia, J.E. Klein, Angew. Chem., Int. Ed. 54 (2015) 5518. [12] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297. [13] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868. [14] (a) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 132 (2010) 154104; (b) S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 32 (2011) 1456. [15] A. Klamt, G. Schüürmann, J. Chem. Soc., Perkin Trans. 2 (1993) 799. [16] H. Kruse, S. Grimme, J. Chem. Phys. 136 (2012) 154101. [17] A.D. Becke, J. Chem. Phys. 98 (1993) 5648. [18] R.L. Martin, J. Chem. Phys. 118 (2003) 4775-4774. [19] L. Yang, D.R. Powell, R.P. Houser, Dalton Trans. (2007) 955. [20] B. Cordero, V. Gómez, A.E. Platero-Prats, M. Revés, J. Echeverría, E. Cremades, F. Barragán, S. Alvarez, Dalton Trans. (2008) 2832. [21] M. Ponikvar, B. Zˇemva, J.F. Liebman, J. Fluorine Chem. 123 (2003) 217. [22] J.J. McKinnon, M.A. Spackman, A.S. Mitchell, Acta Cryst. B60 (2004) 627. [23] H.R. Khavasi, A.A. Tehrani, Inorg. Chem. 52 (2013) 2891. [24] S.A. Al-Jibori, A.I. Abdullah, T.A.K. Al-Allaf, Trans. Metal Chem. 32 (2007) 398. [25] (a) I. Sanyal, K.D. Karlin, R.W. Strange, N.J. Blackburn, J. Am. Chem. Soc. 115 (1993) 11259; (b) T.N. Sorrell, D.L. Jameson, J. Am. Chem. Soc. 105 (1983) 6013.