Journal of Structural Chemistry. Vol. 50, No. 1, pp. 52-59, 2009 Original Russian Text Copyright © 2009 by N. I. Giricheva, G. V. Girichev, N. P. Kuzmina, Yu. S. Medvedeva, and A. Yu. Rogachev
STRUCTURE OF THE Cu(salen) MOLECULE CuO2N2C16H14 ACCORDING TO GAS-PHASE ELECTRON DIFFRACTION DATA AND QUANTUM CHEMICAL CALCULATIONS N. I. Giricheva,1 G. V. Girichev,2 N. P. Kuzmina,3 Yu. S. Medvedeva,1 and A. Yu. Rogachev3
UDC 539.27
In the framework of synchronous gas-phase electron diffraction and mass spectrometry experiment, the saturated vapor of N,Nc-ethylenebis(salicylaldiminate) copper(II) CuO2N2C16H14 is studied at a temperature T 574(5) K. It is found that evaporation is congruent and the saturated vapor consists of monomeric molecules. Electron diffraction data are proved to correspond to the geometric model for the CuO2N2C16H14 molecule of ɋ2 symmetry with an almost planar structure of the CuN2O2 coordination fragment and internuclear distances rh1(Cu–O) = 1.917(13) Å and rh1(Cu–N) = 1.931(15) Å. The structural parameters obtained are compared to those quantum chemically calculated and molecular parameters in crystals. Keywords: structure, Schiff bases, CuO2N2C16H14, Cu(salen), electron diffraction, quantum chemical calculations.
The present work continues studies on the gas phase of 3d-metal complexes with Schiff bases. In the works [1, 2] electron diffraction was used to find the geometric configuration of free Ni(salen), Ni(acacen), and Cu(acacen) molecules, and the structure of the ɆN2Ɉ2 coordination fragment in three complexes was shown to be close to planar irrespective the number of 3d-electrons in central ions Ni2+(3d 8) and Cu2+(3d 9). This work presents the results of studying the saturated vapor of N,Nc-ethylenebis(salicylaldiminate) copper(II) by the synchronous gas-phase electron diffraction and mass-spectrometric experiment. The vapor composition and the geometric structure of the Cu(salen) molecule are analyzed. The experimental structural parameters of the molecule in the gas phase are compared to the parameters obtained in quantum chemical calculations and the molecular parameters in the crystal.
EXPERIMENTAL The Cu(II) complex was synthesized by the interaction of stoichiometric amounts of copper acetate and the Schiff base in the aqueous alcoholic medium. The precipitate of the Cu(salen) complex (dark green) was filtered on a Buchner funnel, washed with some amount of EtOH, and dried in air. The complex obtained was recrystallized from ethanol. Yield a80%. Cu(salen). IR spectrum: 3020 cm–1, 2954 cm–1, 2926 cm–1 (C–H), 1626 cm–1 (C–N + C–O), 1530 cm–1, 1468 cm–1, 1450 cm–1 (C–C). Electron diffraction experiment was performed on an EMR-100/APDM-1 complex apparatus [3, 4]. The Cu(salen) sample was evaporated at T 574(5) K from a molybdenum effusion cell with a nozzle diameter of 0.6 mm and a length of
1
Ivanovo State University. 2Ivanovo State University of Chemistry and Technology;
[email protected]. M. V. Lomonosov Moscow State University. Translated from Zhurnal Strukturnoi Khimii, Vol. 50, No. 1, pp. 58-65, January-February, 2009. Original Article Submitted February 11, 2008; revised June 16, 2008. 3
52
0022-4766/09/5001-0052 © 2009 Springer Science+Business Media, Inc.
TABLE 1. Mass Spectra of Cu(salen) and Cu(acacen) Vapors at the Ionizing Electron Energy of 50 V Measured Simultaneously with Electron Diffraction Patterns
m/e
Cu(salen), Ɍ 574(5) K Ion
Cu(acacen), Ɍ 425(5) K [2] Ion
Irel, %
m/e
Irel, %
63
Cu+
12
11
63 76
Cu+ C6H 4
55 19
89
C7H 5
12
103
NC7H 5
12
97
105
OC7H 5
16
105
ONC5H 7 CuOC2H+
119
ONC7H 5
14
133
ONC8H 7+ (L1/+ 2 )
100
111
ONC6H 9+ (L1/+ 2 )
24
145
CuN2C4H 6 ONC9H 9 CuO2C6H 2 ON2C10H10 CuONC8H 7+ (CuL1/+ 2 )
12
144
CuONC4H13
11
174
CuONC6H 9+ (CuL1/+ 2 )
100
285
+ CuO2N2C12H 18 (CuL+)
52
147 170 179 197 209 222 225 266 298 330
CuON2C8H 6 CuN2C10H10 CuO2N2C8H 6 O2N2C16H14 CuN2C16H14 + CuO2N2C16H 14 (CuL+)
6
8 13 12 44 6 4 3 5 2 40
1.6 mm and with the ratio of the evaporation area to the nozzle orifice area larger than 500. The cell temperature was measured by a WR-5/20 tungsten-rhenium thermocouple. Accelerating voltage was 80 kV. The vacuum in the diffraction chamber was held at 1.810–6 Torr during the electron diffraction measurements. Electron diffraction measurements were accompanied by recording mass spectra. In the mass spectrum of Cu(salen) vapors, we have not found ions with a weight larger than that of the molecular [CuO2N2C16H14]+ ion. All the other ions are the result of electron-impact dissociative ionization of monomeric molecules. This allows us to be sure that dimeric molecules and volatile impurities are absent in the vapor above Cu(salen) under conditions of the electron diffraction experiment. Table 1 presents the mass spectrum of the Cu(salen) vapor measured simultaneously with the electron diffraction patterns that indicates the monomeric composition of the vapor under study and the absence of volatile impurities. Electron diffraction patterns were measured at two nozzle-to-plate distances: L1 = 598 mm and L2 = 338 mm. Six electron diffraction patterns of the compound under study and two electron diffraction patterns of the ZnO crystal standard were measured at each distance. The electron diffraction pattern of ZnO was recorded twice (before and after electron diffraction measurements of the compound under study) in order to determine and control wavelength stability. Electron diffraction patterns of the gas were scanned on an automatic microphotometer [5] with a step of 0.1 mm along the plate diagonal. We scanned the area of 10u65 mm2; the number of equidistant scanning lines was 33. Total scattering intensities I(s) were obtained in ranges: s = 1.3-16.3 Å–1 and 3.6-28.4 Å–1. The experimental function sM(s) was calculated by the formula sM(s) = ((I(s) – G(s))/G(s)s, where G(s) is the smooth background line. 53
STRUCTURAL ANALYSIS In the structural analysis process, the following assumptions were made: — molecules of only one type CuO2N2C16H14 are present in the vapor; — the geometric configuration of the molecule belongs to the point group ɋ2; — all atoms of the aromatic fragment, including atoms of substituents, lie in one plane. The molecule model (see Fig. 1) was described by 36 independent parameters: internuclear distances r(C12–C13), r(C7–C8), r(C8–C14), r(C14–C15), r(C15–C16), r(C16–C17), r(C6–C17), r(C6–C8), r(C12–N2), r(N2–C7), r(O4–C6), r(Cu–N2), r(C12–H32), r(C14–H24), bond angles C13C12N2, C12N2C7, N2C7C8, C7C8C14, C8C14C15, C14C15C16, C15C16C17, H32C12C13, H32C12N2, H33C12N2, H33C12C13, H22C7N2, H24C14C8, H25C15C14, H26C16C15, H27C17C16, O4C6C8, and torsion angles W(N2C12C13N3), W(C7N2C12C13), W(C8C7N2C12), W(C14C8C7N2), W(H22C7N2C12). It seems impossible to determine simultaneously and independently 36 structural parameters from electron diffraction data due to their correlation. In order to diminish this correlation the following restrictions were introduced: — in the least-square analysis (LSA), 5 types of internuclear distances were varied independently: C–C, N–C, Cu– N, C–H, and O–C, the difference obtained in quantum chemical calculations being maintained between inequivalent distances of one type; — 5 bond angles were varied: C13C12N2, C12N2C7, N2C7C8, C7C8C14, and O4C6C8, the quantum chemical difference between ɋɋɋ angles in the benzene fragment being maintained and the values of ɇɋɋ and ɇɋN angles being fixed at quantum chemical parameters; — the values of 4 torsion angles were varied: W(N2C12C13N3), W(C7N2C12C13), W(C8C7N2C12), and W(C14C8C7N2). The value of the W(H22C7N2C12) angle was taken equal to the quantum chemical value and was not refined. Due to the restrictions imposed, the number of variable parameters was reduced to 14. These parameters along with 14 groups of amplitudes (Table 2) were independently refined in the LSA. The starting values for internuclear distances, bond and torsion angles were taken from the results for the Cu(salen) molecule UB3LYP/6-31G* calculated (see below). The starting values for vibrational amplitudes as well as vibrational corrections for internuclear distances 'r = rh1 – ra were calculated using the SHRINK program [6] with the force field and the molecule geometry obtained by the same quantum chemical method. Note that in the LS procedure, the corrections 'r were not varied. Phases and amplitudes of atomic scattering were taken from the work [7]. Fig. 2 depicts the molecular component of the experimental scattering intensity function sM(s) and the radial distribution function f (r)exp along with the theoretical functions corresponding to optimized values of structural parameters of the molecule. Table 2 presents the results of the LSA for the functions sM(s).
Fig. 1. Geometric model of the Cu(salen) molecule with atomic numbering. 54
TABLE 2. Values of ra-Parameters, Experimental and Calculated Vibrational Amplitudes with their Division into Groups, and Corrections for Internuclear Distances 'r (in Å)
Termɚ
ra
lexp
lcalcb
'rɫ = rh1 – ra
Group No.d
Termɚ
ra
lexp
lcalcb
'rɫ = rh1 – ra
Group No.d
C14–H24 C6–O4 N2–C7 N2–C12 C12–C13 Cu–O4 Cu–N2 C17–O4 C7–C12 N3–C12 N2–C8 C8–O4 N2–N3 O4–O5 C8–C16 Cu–C12 Cu–C6 Cu–C7 Cu–C8 N2–C14 C14–O4 N2–O5 C12–O4
1.125(2)e 1.321(4) 1.308 1.468(3) 1.544(1) 1.921 1.929(7) 2.334 2.318 2.392 2.386 2.474 2.561 2.730 2.807 2.846 2.858 2.916 3.282 3.637 3.723 3.825 4.156
0.077(5)f 0.051(3) 0.049(3) 0.060(3) 0.063(3) 0.104(3) 0.105(3) 0.054(3) 0.060(3) 0.063(3) 0.049(3) 0.054(3) 0.071(3) 0.168(4) 0.091(4) 0.106(4) 0.099(4) 0.101(4) 0.077(7) 0.081(9) 0.077(9) 0.118(9) 0.133(7)
0.075 0.043 0.041 0.052 0.055 0.069 0.070 0.064 0.069 0.072 0.059 0.063 0.080 0.147 0.070 0.085 0.078 0.080 0.089 0.072 0.067 0.109 0.106
0.0013 –0.0000 0.0007 0.0007 –0.0008 –0.0001 –0.0010 –0.0014 0.0053 0.0027 0.0052 0.0085 –0.0015 –0.0428 0.0039 0.0110 0.0151 0.0137 0.0165 0.0085 0.0152 0.0217 0.0352
1 2 2 2 2 3 3 4 4 4 4 4 4 5 5 5 5 5 6 7 7 7 8
Cu–C17 N2–C17 Cu–C14 C17–O5 N2–C16 Cu–C16 Cu–C15 N2–C21 N2–C18 C14–O5 C15–O5 C7–C21 N2–C20 C12–C19 N2–C19 C8–C18 C9–C15 C14–C20 C14–C19 C15–C20 C15–C19 C15–H29 H25–H29
4.144 4.306 4.695 4.913 5.062 5.292 5.538 6.000 6.075 6.216 6.808 7.004 7.066 7.149 7.116 7.911 8.045 9.731 10.126 10.284 10.827 11.916 13.002
0.117(7) 0.113(7) 0.136(6) 0.221(6) 0.124(6) 0.104(11) 0.106(11) 0.160(18) 0.157(18) 0.190(18) 0.231(22) 0.201(22) 0.188(22) 0.226(22) 0.187(22) 0.158(30) 0.192(30) 0.215(82) 0.186(82) 0.243(82) 0.208(82) 0.231(82) 0.252(82)
0.091 0.086 0.094 0.179 0.082 0.091 0.093 0.127 0.125 0.158 0.173 0.143 0.130 0.168 0.129 0.126 0.160 0.178 0.149 0.206 0.171 0.194 0.215
0.0261 0.0238 0.0279 0.0070 0.0270 0.0370 0.0387 0.0503 0.0435 0.0506 0.0538 0.0815 0.0623 0.0547 0.0600 0.1014 0.1015 0.1482 0.1567 0.1648 0.1792 0.2142 0.2520
8 8 9 9 9 10 10 11 11 11 12 12 12 12 12 13 13 14 14 14 14 14 14
a
Parameters ra and lexp are given for the most important terms of the molecule that belong to different structural peaks of the function f(r) (Fig. 3). b Amplitudes and corrections 'r are calculated by the force field obtained from UB3LYP/6-31G*calculations. c Corrections are calculated with regard to a nonlinear relation between natural and Cartesian vibrational coordinates within the approach [6] and were not varied in the LSA. d Numbers of groups of terms that belong to different peaks of the function f (r); augmentation of amplitudes for the terms in one group was taken equal in the LSA. e Values of VLSA are given for independent internuclear distances in parentheses. f Values of 3VLSA are given for vibrational amplitudes in parentheses.
QUANTUM CHEMICAL CALCULATIONS Quantum chemical calculations of the structure and the force field of the Cu(salen) molecule were performed to evaluate the starting geometric parameters and also to obtain the consistent estimation of the generalized vibrational amplitudes at the first stage of the the LSA of the function sM(s). Quantum chemical calculations were carried out by density functional theory (DFT) using the GAUSSIAN-03 program [8] with the B3LYP hybrid exchange-correlation potential [9]. Two variants of calculations were performed using the full electron 6-31G* basis set and also Stevens–Bash–Krauss pseudo-potentials (CEP-31G basis set) [10] on heavy atoms.
55
Fig. 2. Experimental (dots) and theoretical (solid line) functions of molecular intensity scattering sM(s) (a) and radial distribution f (r) (b). TABLE 3. Geometric Parameters of the Cu(salen) Molecule Obtained by Gas Electron and X-Ray Diffraction and Quantum Chemical Calculations Cu(salen) Parameter (r in Å, angles in GED, rh UB3LYP/ UB3LYP 1 deg) Rf = 3.2% CEP-31G 6-31G* r(C12–C13) p1a r(C7–C8) (p1) b r(C12–N2) p2 r(N2–C7) (p2) r(C–Hav) p3 r(C6–O4) p4 r(Cu–N2) p5 r(Cu–O4) r(N…N) r(O…O) r(N…O)
1.543(3)c 1.431(3) 1.469(8) 1.308(8) 1.131(5) 1.321(10) 1.927(17) 1.921(15) 2.559(17) 2.687(47) 2.815(18) e C13C12N2 p6 105.3(15) C12N2C7 p7 113.5(13)
1.554 1.450 1.490 1.324 1.098 1.336 1.972 1.920 2.655 2.759 2.809 107.3 120.7
1.538 1.427 1.461 1.300 1.092 1.297 1.941 1.891 2.616 2.765 2.786 107.2 121.3
XRD [16]
Parameter (r in Å, angles in deg)
1.503 1.432 1.483 1.283 — 1.325 1.941 1.906 2.615 2.701 2.780 108.0 121.0
N2C7C8 p8 C7C8C14 p9 O4C6C8 p10 NCuN OCuO NCuO M(N2C12C13N3) p11 M(C7N2C12C13) p12 M(C8C7N2C12) p13 M(C14C8C7N2) p14 M(O4N2O5N3) ¦cycle ¦(N)
d
Cu(salen) GED, rh1 UB3LYP/ UB3LYP XRDd Rf = 3.2% CEP-31G 6-31G* [16] 121.5(13) 117.2(5) 127.4(7) 83.2(7) 88.8(17) 94.1(7) –44.9(21)f –163.8(22) –172.4(40) –170.5(39) 176.0(97) 521.0(25) 355.9(24)
125.5 121.9 123.9 84.6 91.9 92.4 –43.0 –146.4 –177.8 –179.3 163.0 523.9 360.0
125.3 118.0 124.4 84.8 94.0 93.3 –43.8 –140.7 177.2 178.0 145.6 521.1 355.9
124.3 123.7 123.9 84.7 90.2 92.5 –41.32 –147.8 –177.5 –175.7 176.0 524.5 359.9
a
pi is independently refined parameter. (pi) was refined in the group with the parameter pi with a maintained quantum chemical difference between them. ɫ 2 2 2 1/ 2 Total error in the values of internuclear distances calculated by the formula V (Vscale (2.5V LS ) ) , where Vscale = 0.002r. d Mean values of parameters in crystal. e Error in values of bond angles equal to 2.5VLSA. f Error in values of torsion angles equal to VLSA. Correlation coefficients larger than 0.7: p4/p2 = –0.77, p8/p11 = –0.74, p11/p6 = 0.95, p12/p7 = 0.80, p13/p8 = 0.79, l(Cu…C)/p6 = 0.80. b
Full geometry optimization of the molecule corresponding to the energy minimum on the potential energy surface was carried out to a gradient value of 10–11 au with regard to symmetry restrictions. The molecule was assumed to have the ɋ2 symmetry axis. The correspondence between the optimized geometry and the energy minimum was tested in calculating vibrational frequencies. Multiplicity of the molecule was taken as two.
56
The force field of the Cu(salen) molecule required to evaluate vibrational amplitudes lcalc and corrections 'r = rh1 – ra for perpendicular vibrations given in Table 2 was calculated at the UB3LYP/6-31G* level of theory. Table 3 presents the results for the geometric structure of the Cu(salen) complex calculated using UB3LYP/6-31G* and UB3LYP/ɋȿɊ-31G method/basis set.
RESULTS AND DISCUSSION Vapor composition and mass-spectrum. Table 1 presents the mass spectrum of Cu(salen) along with the mass spectrum of the Cu(ɚɫɚɫen) complex studied previously [2] that has a similar CuN2O2 coordination fragment and the ethylene bridge. According to mass spectra, the saturated vapors of these compounds are monomeric. The mass spectra of both complexes exhibit 4 intense peaks each that are assigned to the molecular CuL+ ion, the CuL1/2 ion consisting of the metal atom and half of the ligand, the metal-free L1/2 ion corresponding to half of the ligand, and the atomic ion of copper Cu+.
Thus, the schemes of electron-impact dissociative ionization of both Cu(salen) and Cu(acacen) complexes prove to be similar. Features of the geometric structure of the free molecule and the molecule in the crystal. Unlike Ni(salen), Ni(acacen), and Cu(acacen), for the Cu(salen) complex X-ray diffraction data are absent for single crystals containing only Cu(salen) complexes. However, several works on the crystals whose composition includes Cu(salen) complexes are available from the Cambridge Crystallographic Database [11]. Table 4 lists the geometric parameters of the coordination fragment and the ethylene bridge in the free molecule (obtained in this work) and in ɋu(salen) molecules in different crystals. Depending on the crystal packing, the structure of the Cu(salen) complex changes considerably. So, for example, in the crystal whose unit cell includes the Cu2O7N4C32H34Gd3+ ion, three ClO14 ions, 1/2 of the C2H5NO2 molecule, and two CuO2N2C16H14 molecules, the structures of the latter are different (Table 4, “1st and 2nd molecules”) in both bond lengths and torsion angles that define the structure of the ethylene bridge and the coordination fragment [12]. In another crystal [13], the unit cell contains, apart from the Cu(salen) molecule, the 4-(NO2)C6H4(OH) molecule, the para-nitrophenol molecule being oriented by the NO2 group to the ethylene bridge of the Cu(salen) complex. Hence, the structure of the ethylene bridge distorts, and the angle M(N2C12C13N3) becomes 30q.
TABLE 4. Geometric Parameters of the Coordination Fragment and the Ethylene Bridge in a Free Molecule and ɋu(salen) Molecules in Different Crystals Parameter (r in Å, angles in deg)
GED rh1
r(ɋ12–ɋ13) M(N2C12C13N3) r(Cu–N2) r(Cu–O4) r(N…N) r(O…O) r(N…Ɉ) NCuN OCuO M(O4N2O5N3)
1.543(3) –44.9(21) 1.928(17) 1.921(15) 2.559(17) 2.687(47) 2.815(18) 83.1(7) 88.7(17) 176.0(97)
XRD [12] 1st molecule
2nd molecule
1.487 –39.7 1.947 1.953 2.645 2.705 2.886 85.5 90.6 160.9
1.420 –30.9 1.952 1.915 2.677 2.698 2.783 86.6 89.6 172.1
XRD [13]
XRD [14]
XRD[15]
1.485 –30.17 1.915 1.895 2.531 2.628 2.805 82.7 87.8 173.39
1.543 –39.53 1.931 1.890 2.605 2.679 2.797 84.8 90.3 152.89
1.503 –41.32 1.941 1.906 2.615 2.701 2.780 84.7 90.2 176.02
57
In the crystal [14] containing Cu(salen) and CS(NH2)2 molecules in the unit cell, the latter molecule is oriented by hydrogen atoms of amino groups to oxygen atoms of the Cu(salen) molecule. This results in the formation of the intermolecular hydrogen bond with a distance r(O…H) a 2 Å. Therefore, the CuN2O2 coordination fragment noticeably distorts. The torsion angle M(O4N2O5N3) is 153q, unlike the value of 176q for this angle in the free molecule (Table 4). Less dense crystal packing is observed for the crystal [15] whose unit cell contains Cu(salen) and CHCl3 molecules, the shortest distance between atoms of the neighboring molecules exceeding 3 Å. The structure of the Cu(salen) molecule in a similar crystal can be expected to be close to the structure of the free molecule. It is seen from Table 4 that the spread in values of internuclear Cu–N and Cu–O distances in different crystals reaches 0.06 Å. Thus, values of r(Cu–N) and r(Cu–O) in the free molecule are within the spread of distances in crystals. The same can be said about distances between unbonded Ɉ…Ɉ, N…N, N…O atoms. In many crystals, the ɋ–ɋ distance in the ethylene bridge is smaller than r(C–C) in the free molecule. A more noticeable distinction is observed in the values of torsion angles M(O4N2O5N3) and M(N2ɋ12ɋ13N3) defining the planarity of the coordination fragment and torsion of the ethylene bridge. In crystals [12] and [14], Cu(salen) molecules have a substantially non-planar coordination fragment, whereas in crystals [13] and [15] and one of the molecules of the crystal [12] this fragment is almost planar as in the free molecule. It should be noted that fragments with high structural nonrigidity (the coordination fragment and the ethylene bridge) are prone to major changes in the complex geometry caused by molecular packing effects in the crystal. In Table 3, parameters of the free molecule found by electron diffraction and also by quantum chemical calculations (UB3LYP/CEP-31G and UB3LYP/6-31G*) are compared to parameters of the Cu(salen) complex in the crystal with lower density packing [15]. As seen from Table 3, all parameters of the CuO2N2 coordination fragment in the crystal [15] coincide with similar parameters of the free molecule (within error limits of the electron diffraction experiment). Quantum chemical calculations at the UB3LYP/CEP-31G level result in the values of distances overestimated by 0.01-0.04 Å in both the ligand and the coordination fragment. Moreover, calculations at this level of theory predict high nonplanarity of the coordination fragment, while according to electron diffraction data, it has an almost planar structure. In the analysis of calculation results for the Cu(acacen) molecule with 6-31G* and CEP-31G basis sets, it is seen that the first basis set better reproduces the values of internuclear distances; however, the non-planarity of the coordination fragment calculated with this basis set is obviously overestimated in comparison with both the free molecule and the molecule in the crystal. Despite some differences between the results of experiments (GED and XRD) and calculations (UB3LYP), several features can be noted in the structure of the complex. Thus, the values of internuclear ɋ12–ɋ13 and ɋ12–N2 distances correspond to the representation of single ɋ–ɋ and ɋ–N bonds. The five-membered CuN2ɋ2 cycle has the form of a twistconformation with the largest torsion angle around the ɋ–ɋ bond (M(NCCN) = 44.9(21)q) and the smallest angle (M(NCuNC) = 14.1(7)q) around the Cu–N bond. The sum of internal angles in the cyclic fragment (Table 3, ¦cycle) is 521.0(25)q, which is noticeably smaller than the value of 540q, characteristic of a planar pentagon. At the same time, the sum of bond angles at the nitrogen atom is 355.9(24)q, which corresponds to almost planar coordination of N2–C12, N2–C7, and N2–Cu bonds (Table 3, ¦(N)). The N2–C12 and N2–C7 bonds have different lengths, the value of r(N2–C7) being characteristic of the double N = C bond and the value of r(N2–C12) of the single one. As noted above, two experimental methods (gas-phase electron and X-ray diffraction) yield an almost planar structure of the CuN2O2 coordination fragment (Table 3, M(O4N2O5N3) is close to 180q). The coordination fragment in the Ni(salen) complex has a similar structure [1], i.e., despite the different number of 3d-electrons in central Ni2+(3d 8) and Cu2+(3d 9) ions, both complexes tend to maintain the planar arrangement of M–O and M–N coordination bonds. The work was supported by RFBR (grant No. 07-03-00656ɚ).
REFERENCES 1. G. V. Girichev, N. I. Giricheva, N. P. Kuzmina, et al., J. Struct. Chem., 46, No. 5, 813-823 (2005). 2. G. V. Girichev, N. I. Giricheva, N. P. Kuzmina, et al., ibid., 49, No. 5, 871-882 (2008). 58
3. 4. 5. 6.
7. 8. 9. 10. 11.
12. 13. 14. 15.
G. V. Girichev, A. N. Utkin, and Yu. F. Revichev, Prib. Tekhn. Éksp., No. 2, 187-190 (1984). G. V. Girichev, S. A. Shlykov, and Yu. F. Revichev, ibid., No. 4, 167-169 (1986). E. G. Girichev, A. V. Zakharov, G. V. Girichev, and M. I. Bazanov, Izv. Vuz. Tekstil. Prom., No. 2, 142-146 (2000). V. A. Sipachev, J. Mol. Struct., 567/568, 67 (2001); V. A. Sipachev, ibid., 121, 143 (1985); V. A. Sipachev, in: Advances in Molecular Structure Research, I. Hargittai and M. Hargittai (eds.), JAI Press, New York, No. 5, 263 (1999). A. W. Ross, M. Fink, and R. L. Hildebrand, International Tables of Crystallography, C., Kluwer Acad. Publ., Dodrecht (1992), p. 245. M. J. Frisch, G. W. Trucks, H. B. Schlegel, et al., Gaussian-03, Revision B.03, Gaussian, Inc., Pittsburgh PA (2003). D. Becke, J. Chem. Phys., 98, 5648 (1993). W. Stevens, H. Basch, and J. Kraus, ibid., 81, 6026 (1984); W. J. Stevens, M. Krauss, H. Basch, and P. G. Jasien, Can. J. Chem., 70, 612 (1992); T. R. Cundari and W. J. Stevens, J. Chem. Phys., 98, 5555 (1993). F. H. Allen, Acta Crystallogr., B58, 380-388 (2002); J. van de Streek, ibid., B62, 567-579 (2006); A. G. Orpen, ibid., B58, 398-406 (2002); F. H. Allen and W. D. S. Motherwell, ibid., 407-422; R. Taylor, ibid., D58, 879-888; I. J. Bruno, J. C. Cole, P. R. Edgington, et al., ibid., B58, 389-397. A. Bencini, C. Benelli, A. Caneschi, et al., J. Am. Chem. Soc., 107, 8128 (1985). E. N. Baker, D. Hall, and T. N. Waters, J. Chem. Soc. A, 400 (1970). M. B. Ferrari, G. G. Fava, and C. Pelizzi, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 32, 901 (1976). E. N. Baker, D. Hall, and T. N. Waters, J. Chem. Soc. A, 406 (1970).
59
