[(Triethylenetetramine)copper(II)]-μ-cyanido-κ2 N:C-[bis(cyanido-κC)copper(I)]

The title compound, [Cu2(CN)3(C6H18N4)] or [Cu(trien)(CN)Cu(CN)2], where trien is triethylenetetramine, is a mixed-valence complex crystallizing as discrete molecules, with CuI and CuII ions linked by a bridging cyanide group. The CuII ion is in a square-pyramidal coordination environment, with the N atoms of the tetradentate trien ligand occupying the basal positions and Cu—N bond lengths in the range 2.028 (4)–2.047 (4) Å. An N-bonded cyanide group is in the apical position, with a slightly longer Cu—N bond length of 2.127 (4) Å. The CuI ion exhibits a trigonal–planar coordination geometry, bonded to the C atoms of the bridging cyanide group and two terminal cyanide groups with Cu—C bond lengths in the range 1.925 (4)–1.948 (5) Å. In the crystal, hydrogen bonding involving the tertiary N—H groups of the trien ligand and N atoms of symmetry-related terminal cyanide groups links molecules into a ribbon extending in the b-axis direction.


Comment
The structure determination of the title compound was undertaken as part of a series of synthetic and structural studies of mixed-valence copper cyanide complexes containing amine bases. The coordinated amines stabilize the divalent copper atoms against reduction by the cyanide groups. In this study, the synthesis involved the linear tetradentate base triethylenetetramine (trien), under conditions expected to produce a polymeric structure, as in Williams et al. (1972); Colacio et al. (2002); Kim et al. (2005). Instead the crystal structure is made up of discrete molecules with terminal cyanide groups that are not involved in covalent polymeric linkages, and is similar to structures reported by: Yuge et al. (1998);Pickardt et al. (1999); Pretsch et al. (2005).
The molecules contain a divalent and a monovalent copper atom bridged by a cyanide group. The divalent copper atom, Cu2, shows square-pyramidal coordination, with the four N atoms of the tetradentate ligand occupying the basal positions, and the N atom of the bridging cyanide group in the apical position. Cu2 lies 0.490 (2) Å out of the best plane through the four nitrogen atoms of the ligand, in the direction of the apical N atom. The three Cu-N-C-C-N-Cu chelate rings have torsional angles of 49.9 (5) °, 51.7 (6) ° and 53.1 (6) °. The monovalent copper atom, Cu1, shows trigonal planar coordination to the carbon atoms of the bridging and two terminal cyanide groups, with bond angles close to 120 ° and Cu1 almost coplanar with the three cyanide carbon atoms, lying 0.039 (3) Å above their plane.
The bridging C-N bond length of 1.131 (6) Å is not significantly different from the terminal C-N bond lengths. This C-N group is linearly bonded to the two copper atoms, with the angle Cu1-C-N = 177.9 (4)° and C-N-Cu2 = 174.0 (4)°. The linear geometry differs from that found in the one-dimensional polymer [Cu(dien)CN] + , (Corfield & Yang, 2012) where both copper atoms are divalent, and the C-N-Cu angle is non-linear at 146.5 (2)°. The Jahn-Teller lengthening of the axial Cu-N distance is greater in the polymer, with Cu-N = 2.340 (3) Å versus 2.127 (4) Å in the present structure.
Two prominent hydrogen bonds are seen linking N-H bonds from the trien ligand and nitrogen atoms of neighboring terminal cyanide groups. The N7-H7.. . N2(1 -x,1 -y,1 -z) hydrogen bond joins the two molecules in the unit cell into a centrosymmetric dimer, while the N10-H10···N3(1 -x,-y,1 -z) hydrogen bonds link these dimers into a ribbon extending along the direction of the b axis, as shown in figure 2. Hydrogen bonds between terminal cyanide groups and N -H bonds are also seen in the similar compound studied by Pretsch et al. (2005).
Intermolecular contacts appear normal, with the shortest H-H intermolecular distance found at 2.53 Å. There are short contacts, however, between the C atoms of the terminal cyanide groups and H atoms of neighboring molecules, with C3 -H13B(-x, -y, 1 -z) at 2.42 Å, and C2-H4B(1 + x, y, z) at 2.55 Å.

Experimental
The compound was prepared by addition of 20 ml of a solution containing 0.020 mol of CuSO 4 . 5H 2 O and 0.020 mol of trien to 20 ml of a solution containing 0.040 mol of NaCN. A yield of approximately 1.0 g of blue crystals in the form of supplementary materials sup-2 Acta Cryst. (2012). E68, m1532-m1533 diamond plates elongated along b was obtained. Total copper was analyzed iodometrically: found 35.7%; calculated 36.2%. Cyanide was analyzed by AgNO 3 titration of the HCN gas evolved on addition of 15M HNO 3 to a sample, as in Cooper & Plane (1966): found 20.3%; calculated 22.2%. This corresponds to a recovery rate of 91%, which was typical for our test analyses by this method. CN stretching frequencies were found at 2089 (s, doublet) and 2110 (m) cm -1 , with a Perkin-Elmer 710B spectrophotometer.

Refinement
All 18 hydrogen atoms of the diethylenetriamine ligand were found unambiguously from a difference Fourier map, and were initially refined freely. In the final refinements, hydrogen atoms were constrained to idealized positions by SHELXL (Sheldrick, 2008). No improvement was found by refining the NH atoms independently.
Refinements with anisotropic temperature factors for Cu, N and C atoms and constrained hydrogen atom parameters converged smoothly, but a difference Fourier synthesis at this stage showed a pattern of peaks and holes of 1.0-1.2 e/A 3 associated with the copper atoms. Further, examination of the observed and calculated structure factors showed a clear anisotropic variation in scale factor. This anisotropy was modeled by using the program XABS2, (Parkin et al., 1995) to modify the observed structure factors. The structure presented here is based upon subsequent refinements in SHELXL (Sheldrick, 2008) that lowered R 1 from 0.0544 to 0.0402 for all reflections and R 2 from 0.1552 to 0.1201. Changes in atomic parameters were small: the maximum shift was 1.6σ, with only 22 of the 162 atomic parameters changing by more than 1σ. The anisotropy in the scale factor and the noise in the final difference Fourier map were both lowered.
There is no evidence of disorder between the C and N atoms of the cyanide groups. Assignments of C and N atoms in the cyanide groups were verified after the last stage of the refinement, when the C and N atoms for each group were in turn inverted. In each case, R values increased, and the U values became unlikely.

[(Triethylenetetramine)copper(II)]-µ-cyanido-κ 2 N:C-[bis(cyanido-κC)copper(I)]
Crystal data where P = (F o 2 + 2F c 2 )/3 (Δ/σ) max < 0.001 Δρ max = 0.51 e Å −3 Δρ min = −0.56 e Å −3 Special details Experimental. crystal A: 2θ 0-35°; 32-50°; 47-79°. small crystal, showing visible decomposition after data collection, with 13% loss of intensities; no absorption correction. crystal B: 2θ 0-40°; 65-90°; 86-120°. larger crystal, mounted in capillary tube; no fall-off of intensities of standard reflections, but intensities fluctuated with an e.s.d. of 3%. Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.