Mononuclear and binuclear copper(I) complexes ligated by bis(3,5-diisopropyl-1-pyrazolyl)methane: Insight into the fundamental coordination chemistry of three-coordinate copper(I) complexes with a neutral coligand

Inorg. Chem. 2007, 46, 10607−10623

Mononuclear and Binuclear Copper(I) Complexes Ligated by

Bis(3,5-diisopropyl-1-pyrazolyl)methane: Insight into the Fundamental

Coordination Chemistry of Three-Coordinate Copper(I) Complexes with

a Neutral Coligand^†

Kiyoshi Fujisawa,*^,‡Yuki Noguchi,^‡Yoshitaro Miyashita,^‡Ken-ichi Okamoto,^‡and Nicolai Lehnert*^,§

Department of Chemistry, Graduate School of Pure and Applied Sciences, UniVersity of Tsukuba,

Tsukuba 305-8571, Japan, Department of Chemistry, UniVersity of Michigan, 930 N. UniVersity,

Ann Arbor, Michigan 48109

Received June 10, 2007

By using the neutral bidentate nitrogen-containing ligand, bis(3,5-diisopropyl-1-pyrazolyl)methane (L1′′), the

copper(I) complexes [Cu(L1′′)₂](CuCl₂) (1CuCl₂), [Cu(L1′′)₂](ClO₄) (1ClO₄), [Cu(L1′′)]₂(ClO₄)₂(2ClO₄), [Cu(L1′′)]₂-

(BF₄)₂(2BF₄), [Cu(L1′′)(NCMe)](PF₆) (3PF₆), [Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄), [Cu(L1′′)(PPh₃)](PF₆) (4PF₆), [

{

Cu-

(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄), and the copper(II) complexes [{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂] (6), and [Cu(L1′′)Cl₂]

(7) were systematically synthesized and fully characterized by X-ray crystallography and by IR and ¹H NMR

spectroscopy. In the case of copper(II), ESR spectroscopy was also applied. In comparison with the related neutral

tridentate ligand L1′, bis-chelated copper(I) complexes and binuclear linear-coordinated copper(I) complexes are

easy to obtain with L1′′, like 1CuCl₂, 1ClO₄, 2ClO₄, and 2BF₄. Importantly, stronger and bulkier ligands such as

acetonitrile (3PF₆) and especially triphenylphosphine (4ClO₄and 4PF₆) generate three-coordinate structures with

a trigonal-planar geometry. Surprisingly, for the smaller ligand carbon monoxide, a mononuclear three-coordinate

structure is very unstable, leading to the formation of a binuclear complex (5ClO₄) with one bridging perchlorate

anion, such that the copper(I) centers are four-coordinate. The same tendency is observed for the copper(II) bis-

(

µ−hydroxo) compounds 6, which is additionally bridged by two perchlorate anions. Both copper(II) complexes 6

and 7 were obtained by molecular O₂oxidation of the corresponding copper(I) complexes. A comparison of the

new copper(I) triphenylphosphine complexes 4ClO₄and 4PF₆with corresponding species obtained with the related

tridentate ligands L1′ and L1 (8ClO₄and 9, respectively) reveals surprisingly small differences in their spectroscopic

properties. Density functional theory (DFT) calculations are used to shed light on the differences in bonding in

these compounds and the spectral assignments. Finally, the reactivity of the different bis(pyrazolyl)methane complexes

obtained here toward PPh₃, CO, and O₂is discussed.

Introduction

rings, using straightforward synthetic procedures.^1,2There-

fore, the transition-metal complexes with these ligands

have raised interest among many researchers, and they

continue to be the subject of many studies, especially in

organometallic chemistry.³In particular, the coordination

chemistry of copper(I) complexes with low coordination

numbers is of increasing interest, for example two-coordinate

structures.⁴

Since Trofimenko’s first report in 1970, poly(pyrazolyl)-

alkanes, especially bis(pyrazolyl)methane, have been popular

polydentate nitrogen donor ligands.¹Their coordination

behavior can easily be adjusted by changing the electronic

and steric characteristics of the substituents on the pyrazolyl

^†Dedicated to Professor O. Yamauchi on the occasion of his 70th

birthday.

* To whom correspondence should be addressed. E-mail: kiyoshif@

chem.tsukuba.ac.jp (K.F.), lehnertn@umich.edu (N.L.).

^‡University of Tsukuba.

(2) A recent book and reviews: (a) Trofimenko, S. Chem. ReV. 1993, 93,

943-980. (b) Trofimenko, S. Scorpionates: The Coordination

Chemistry of Polypyrazolylborate Ligands, Imperial College Press:

London, U.K., 1999. (c) Parkin, G. Chem. Commun. 2000, 1971-

1985.

^§University of Michigan.

(1) Trofimenko, S. J. Am. Chem. Soc. 1970, 92, 5118-5126.

10.1021/ic7011342 CCC: $37.00

Published on Web 11/14/2007

Inorganic Chemistry, Vol. 46, No. 25, 2007 10607

Fujisawa et al.

Recently, a number of investigations of copper complexes

with neutral bidentate nitrogen-containing ligands using

pyrazolyl donors have been carried out. Chou and co-workers

structurally characterized the 2:1 co-crystal of the mixed

copper(I)/copper(II) complexes [Cu^I(H₂Cpz₂)(NCMe)₂](ClO₄)

and [Cu^II(H₂Cpz₂)₂(OClO₃)₂], the perchlorate bridged bis-

(µ-hydroxo) dicopper(II) complex [{Cu^II(H₂Cpz₂)}₂(µ-OH)₂-

(µ-ClO₄)](ClO₄), and the bis-chelated copper(I) complex

[Cu^I(H₂Cpz₂)₂](ClO₄), using bis(pyrazolyl)methane (H₂Cpz₂)

as a ligand.⁵They also determined the crystal structures of

the related mononuclear copper(I) complexes, [Cu^I{H₂C-

(3,5-Me₂pz)₂}(NCMe)](ClO₄), [Cu^I{H₂C(3,5-Me₂pz)₂}(CO)-

(OClO₃)], and [Cu^I{H₂C(3,5-Me₂pz)₂}(PPh₃)](ClO₄) with the

methyl-substituted ligand bis(3,5-dimethyl-1-pyrazolyl)-

methane (H₂C(3,5-Me₂pz)₂) and investigated the reactivity

of these compounds.⁶A few other substituted N2 ligands

with pyrazolyl rings have also been applied, such as

diphenylbis(pyrazolyl)methane (Ph₂Cpz₂), 1,3-bis(5-meth-

ylpyrazolyl)propane (bmpp), and 1,3-bis(3,5-dimethylpyra-

zolyl)propane (bdpp), and the corresponding copper(II)

complexes [Cu^II(Ph₂Cpz₂)(NO₃)₂(H₂O)], [Cu^II(Ph₂Cpz₂)Cl₂],

[Cu^II(bmpp)(NO₃)₂], and [Cu^II(bdpp)Cl₂] were structurally

characterized.^7,8

Figure 1. Structures of the ligands discussed in this work: bis(3,5-

diisopropyl-1-pyrazolyl)methane (L1′′), hydrotris(3,5-diisopropyl-1-pyra-

zolyl)borate anion (L1^-), and tris(3,5-diisopropyl-1-pyrazolyl)methane (L1′).

In our previous studies, we have focused on tridentate

nitrogen-containing ligands such as the neutral ligand tris-

(3,5-diisopropyl-1-pyrazolyl)methane (L1′) and the corre-

sponding anionic ligand hydrotris(3,5-diisopropyl-1-pyrazolyl)-

borate anion (L1^-, Figure 1). Four-coordinate copper(I)

complexes were prepared with these ligands and character-

ized.⁹These results provide insight into second coordination

sphere effects caused by the different ligands employed; that

is, the influence of the total charge, the steric hindrance, and

electronic effects of the ligands on the structures, properties,

and the reactivity of the complexes. These findings prompted

us to attempt the comparison of the structures, spectroscopic

properties, and reactivities of these copper(I) complexes with

corresponding compounds containing the bidentate ligand

bis(3,5-diisopropyl-1-pyrazolyl)methane H₂C(3,5-iPr₂pz)₂(L1′′,

Figure 1). It can be expected that the change in coordination

number of the supporting coligand from tridentate to biden-

tate will have a great influence on the properties and

reactivities. In this work, the syntheses of the copper(I)

complexes [Cu(L1′′)₂](CuCl₂) (1CuCl₂), [Cu(L1′′)₂](ClO₄)

(1ClO₄), [Cu(L1′′)]₂(ClO₄)₂(2ClO₄), [Cu(L1′′)]₂(BF₄)₂(2BF₄),

[Cu(L1′′)(NCMe)](PF₆) (3PF₆), [Cu(L1′′)(PPh₃)](ClO₄)

(4ClO₄), [Cu(L1′′)(PPh₃)](PF₆) (4PF₆), and [{Cu(L1′′)-

(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄) with H₂C(3,5-iPr₂pz)₂as a

bidentate nitrogen-containing ligand and the reactivity of the

obtained copper(I) complexes toward PPh₃, CO, and O₂are

investigated. A special focus is put on the preparation of

low-coordinate copper(I) complexes, that is, three-coordinate

compounds with acetonitrile, phosphine, and CO as the third

ligands. For this purpose, we have investigated the factors

that favor or disfavor the formation of three-coordinate

copper(I) and how the metal avoids this unfavorable coor-

dination number with a neutral bidentate N2 coligand. The

copper(II) complexes [{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂] (6) and

[Cu(L1′′)Cl₂] (7)¹⁰have been obtained by the reaction of

copper(I) precursors with molecular O₂. All of the complexes

were characterized by IR, far-IR, and NMR/ESR spectros-

copy, and by X-ray crystallography. The properties of the

copper(I) triphenylphosphine complexes 4ClO₄and 4PF₆are

compared to those of [Cu(L1′)(PPh₃)](ClO₄) (8ClO₄) and

[Cu(L1)(PPh₃)] (9),^11bwhich contain the tridentate ligands

L1′ and L1^-(vide supra). Density functional theory (DFT)

calculations were performed to shed light on the spectro-

scopic properties and assignments of these phosphine

(3) (a) Shiu, K.-B.; Yeh, L.-Y.; Pemg, S.-M.; Cheng, M.-C. J. Organomet.

Chem. 1993, 460, 203-211. (b) Tang, L.; Wang, Z.; Xu, Y.; Wang,

J.; Wang, H.; Yao, X. Polyhedron 1998, 17, 3765-3769. (c) Sa´nchez,

G.; Serrano, J. L.; Pe´rez, J.; Ram´ırez de Arellano, M. C.; Lo´pez, G.;

Molins, E. Inorg. Chim. Acta 1999, 295, 136-145. (d) Tsuji, S.;

Swenson, D. C.; Jordan, R. F. Organometallics 1999, 18, 4758-4784.

(e) Arroyo, N.; Go´mez-de la Tore, F.; Jalo´n, F. A.; Manzano, B. R.;

Moreno-Lara, B.; Rodr´ıguez, A. M. J. Organomet. Chem. 2000, 603,

174-184. (f) Tang, L.-F.; Zhao, S.-B.; Jia, W.-L.; Yang, Z.; Song,

D.-T.; Wang, J.-T. Organometallics 2003, 22, 3290-3298.

(4) (a) Sanyal, I.; Karlin, K. D.; Strange, R. W.; Blackburn, N. J. J. Am.

Chem. Soc. 1993, 115, 11259-11270. (b) Voo, J. K.; Lam, K. C.;

Rheingold, A. L.; Riordan, C. G. Dalton Trans. 2001, 1803-1805.

(c) Himes, R. A.; Park, G. Y.; Barry, A. N.; Blackburn, N. J.; Karlin,

K. D. J. Am. Chem. Soc. 2007, 129, 5352-5353.

(5) Chou, C.-C.; Su, C.-C.; Tsai, H. -L.; Lii, K. -H. Inorg. Chem. 2005,

44, 628-632.

(10) Lehnert, N.; Cornelissen, U.; Neese, F.; Ono, T.; Noguchi, Y.;

Okamoto, K.; Fujisawa, K. Inorg. Chem. 2007, 46, 3916-3933.

(11) (a) Kitajima, N.; Fujisawa, K.; Moro-oka, Y.; Toriumi, K. J. Am. Chem.

Soc. 1989, 111, 8975-8976. (b) Kitajima, N.; Fujisawa, K.; Fujimoto,

C.; Moro-oka, Y.; Hashimoto, S.; Kitagawa, T.; Toriumi, K.; Tatsumi,

K.; Nakamura, A. J. Am. Chem. Soc. 1992, 114, 1277-1291. (c)

Baldwin, M. J.; Root, D. E.; Pate, J. E.; Fujisawa, K.; Kitajima, N.;

Solomon, E. I. J. Am. Chem. Soc. 1992, 114, 10421-10431.

(6) Chou, C.; Su, C. -C.; Yeh, A. Inorg. Chem. 2005, 44, 6122-6128.

(7) Shaw, J. L.; Cardon, T. B.; Lorigan, G. A.; Ziegler, C. J. Eur. J. Inorg.

Chem. 2004, 1073-1080.

(8) Schuitema, A. M.; Engelen, M.; Koval, I. A.; Gorter, S.; Driessen,

W. L.; Reedijk, J. Inorg. Chim. Acta 2001, 324, 57-64.

(9) Fujisawa, K.; Ono, T.; Ishikawa, Y.; Amir, N.; Miyashita, Y.;

Okamoto, K.; Lehnert, N. Inorg. Chem. 2006, 45, 1698-1713.

10608 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

steric interactions between the bulky ligands L1′′, L1′, L1^-, and

the large ligand PPh₃are taken into account. To handle molecules

of this size (∼100 atoms), application of the LanL2DZ basis set is

necessary for the geometry optimizations and frequency calcula-

tions. The calculated structures show good overall agreement with

experiment. The obtained vibrational frequencies exhibit no imagi-

nary modes. The LanL2DZ basis set applies Dunning/Huzinaga full

double-ú (D95)¹⁵basis functions on first-row and Los Alamos

effective core potentials plus DZ functions on all of the other

atoms.¹⁶In addition, the structures of the species [Cu(L1′′)]⁺, [Cu-

(L1′)]⁺, and [Cu(L1)] as well as free PPh₃were also fully optimized

using B3LYP/LanL2DZ for the calculation of complex formation

energies. For this purpose, single point calculations using B3LYP/

TZVP on the LanL2DZ structures were also performed. The TZVP

basis set¹⁷has been applied as implemented in G03.

Preparation of Complexes. [Cu{H₂C(3,5-iPr₂pz)₂}₂](CuCl₂),

[Cu(L1′′)₂](CuCl₂) (1CuCl₂). Acetone (60 cm³) was added to a

mixture of L1′′ (588 mg, 1.86 mmol)¹⁰and CuCl (202 mg, 2.04

mmol) in a glovebox. After stirring for 24 h, the solution was

filtered, and the filtrate was evaporated under vacuum. A yellowish-

white powder was obtained. Recrystallization from acetonitrile at

-30 °C afforded colorless crystals. Yield: 65% (501 mg, 0.603

mmol). Anal. Calcd for C₃₈H₆₄N₈Cl₂Cu₂: C, 54.92; H, 7.76; N,

13.48. Found: C, 54.83; H, 7.65; N, 13.45. IR (cm^-1): 3199w,

2965vs, 1545s, 1469s, 1383s, 1271vs, 1183m, 1068m, 1058m, 807s.

Far-IR (cm^-1): 583m, 536s, 516w, 470m, 408vs, 399vs, 367w,

323w, 302w, 253w, 226w, 179w, 170w. ¹H NMR (CD₂Cl₂): 1.04

(d, J_HH) 6.2 Hz, 24H, CH(CH₃)₂), 1.23 (d, J_HH) 6.8 Hz, 24H,

CH(CH₃)₂), 2.74 (br, 4H, CH(CH₃)₂), 3.02 (sept, J_HH) 6.8 Hz,

4H, CH(CH₃)₂), 6.03 (s, 4H, 4-H(pz)), 6.05 (s, 4H, H₂C). ¹³C NMR

(CD₂Cl₂): 22.79 (CH(CH₃)₂), 23.23 (CH(CH₃)₂), 26.28 (CH(CH₃)₂),

28.32 (CH(CH₃)₂), 57.00 (H₂C), 101.7 (pz-4C), 153.3 (pz-3C),

165.3 (pz-5C).

complexes. The systematic comparison of copper(I) com-

plexes with the related coligands L1′′, L1′, and L1^-that have

identical isopropyl-substituted pyrazolyl rings leads to de-

tailed insight into the structural favors and reactivities of

these compounds, as a function of the applied coligand.

Experimental Section

Materials. Preparation and handling of all of the complexes were

performed under an argon atmosphere using standard Schlenk tube

techniques or in a VAC inert atmosphere glovebox containing argon

gas. Dichloromethane and acetonitrile were distilled from phos-

phorus pentoxide and calcium hydride prior to use, respectively.

Diethyl ether and heptane were carefully purified by refluxing/

distilling under an argon atmosphere over sodium benzophenone

ketyl.¹²Chloroform and octane were spectroscopic grade and were

used after bubbling with argon gas. Anhydrous solvents, [Cu-

(MeCN)₄](PF₆), AgClO₄, and AgPF₆were purchased from Aldrich

Chemical Co., Inc. and stored in a glovebox. CuCl was purified

according to a published method.^{11 13}C-enriched carbon monoxide

and NMR solvents were obtained from Cambridge Isotope Labo-

ratories, Inc. An autoclave (TVS-1, 300 cm³) was used for ligand

syntheses. Silica gel was obtained from Merck KGaA. Other

reagents were purchased from Wako Pure Chemical Ind. Ltd. and

used without further purification. The copper(I) perchlorate complex

[Cu{HC(3,5-iPr₂pz)₃}(OClO₃)] ([Cu(L1′)(OClO₃)])⁹and the copper-

(I) acetonitrile complex [Cu{HB(3,5-iPr₂pz)₃}(NCMe)] ([Cu(L1)-

(NCMe)])⁹were prepared by published methods. Although the

synthetic procedure for [Cu{HB(3,5-iPr₂pz)₃}(PPh₃)] ([Cu(L1)-

(PPh₃)]) (9)^11bhas already been reported, this complex was prepared

here by a modified method. The synthesis of H₂C(3,5-iPr₂pz)₂(L1′′)

has been reported before by us¹⁰and Tang.^3fCaution! Perchlorate

salts are potentially explosiVe and should be handled with care.

Instrumentation. IR and far-IR spectra were recorded on KBr

pellets in the 4000-400 cm^-1region and on CsI pellets in the 650-

150 cm^-1region, respectively, using a JASCO FT/IR-550 spec-

trophotometer. Abbreviations used in the description of vibrational

data are as follows: vs, very strong; s, strong; m, medium; w, weak.

¹H NMR (600 MHz), ¹³C NMR (150 MHz), and ³¹P NMR (242

MHz) spectra were obtained on a Bruker AVANCE-600 NMR

spectrometer at room temperature (296 K) unless stated otherwise.

¹H and ¹³C chemical shifts were reported as δ values downfield

from the internal standard tetramethylsilane. ³¹P chemical shifts were

referenced relative to an external standard of PPh₃. ESR spectra

were recorded on a Bruker EMXT ESR spectrometer as frozen

solutions at ∼ -135 °C in quartz tubes (diameter 5 mm) using the

liquid-nitrogen temperature controller BVT 3000. UV-vis absorp-

tion spectra at room temperature in the 240-1100 nm range were

recorded on a JASCO V-570 spectrophotometer. The elemental

analyses (C, H, and N) were performed by the Department of

Chemistry at the University of Tsukuba.

[Cu{H₂C(3,5-iPr₂pz)₂}₂](ClO₄), [Cu(L1′′)₂](ClO₄) (1ClO₄).

[Cu(L1′′)]₂(ClO₄)₂(2ClO₄) (133 mg, 0.139 mmol) was dissolved

in acetonitrile (15 cm³) in a glovebox. After stirring for 24 h, the

solvent was evaporated under vacuum. Recrystallization from

chloroform/heptane (1:2) at -50 °C yielded colorless crystals.

Yield: 45% (50.2 mg, 0.0631 mmol). Anal. Calcd for C₃₈H₆₄N₈-

ClCuO₄‚0.1CHCl₃: C, 56.64; H, 8.00; N, 13.87. Found: C, 56.37;

H, 7.86; N, 13.77. IR (cm^-1): 2965vs, 2930s, 2869m, 1546s,

1461s, 1382m, 1274s, 1144s, 1113vs, 1090vs, 796m, 625m. Far-

IR (cm^-1): 623vs, 584m, 536w, 472m, 406m, 325w, 248w, 226w,

(14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,

M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,

K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,

V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G.

A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;

Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,

H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;

Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.

E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J.

W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;

Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;

Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari,

K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;

Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;

Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;

Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;

Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.

Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT, 2004.

(15) Modern Theoretical Chemistry; Dunning, T. H., Jr.; Hay, P. J.;

Schaefer, H. F., III., Eds.; Plenum Press: New York, U.S.A., 1976.

(16) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283. (b)

Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284-298. (c) Hay,

P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.

Calculations. Spin-restricted density functional theory calcula-

tions using Becke’s three-parameter hybrid functional with the

correlation functional of Lee, Yang, and Parr (B3LYP¹³) were

performed using the program package Gaussian 03.¹⁴The structures

of the three complexes, [Cu(L1′′)(PPh₃)]⁺, [Cu(L1′)(PPh₃)]⁺, and

[Cu(L1)(PPh₃)], were fully optimized without any simplifications

using B3LYP/LanL2DZ. This is necessary to ensure that all of the

(12) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory

Chemicals, 4th ed.; Butterworth-Heinemann: Oxford, U. K., 1997.

(13) (a) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100. (b) Becke, A.

D. J. Chem. Phys. 1993, 98, 1372-1377. (c) Becke, A. D. J. Chem.

Phys. 1993, 98, 5648-5652.

(17) Scha¨fer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571-

2577.

Inorganic Chemistry, Vol. 46, No. 25, 2007 10609

Fujisawa et al.

1

169w, 154w. H NMR (CDCl₃): 0.967 (d, J_HH) 6.9 Hz, 24H,

mixture of [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) (134 mg, 0.140 mmol) and

PPh₃(77 mg, 0.293 mmol) in a glovebox. After stirring for 2 h,

the solution was filtered and the filtrate was evaporated under

vacuum, yielding a white powder. Recrystallization from chloroform/

diethyl ether (1:2) at -50 °C gave colorless crystals. Yield: 56%

(121 mg, 0.163 mmol). Method b: 4ClO₄could also be obtained

by the reaction between [{Cu(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄)

(103 mg, 0.101 mmol) and PPh₃(55.6 mg, 0.212 mmol) in

dichloromethane (10 cm³). Recrystallization from dichloromethane/

octane (1:2) at -50 °C gave a white powder. Yield: 39% (58.9

mg, 0.0794 mmol). Anal. Calcd for C₃₇H₄₇N₄ClCuO₄P: C, 59.91;

H, 6.39; N, 7.55. Found: C, 59.49; H, 6.34; N, 7.43. IR (cm^-1):

3126w, 3055w, 2970vs, 2933m, 2872m, 1547s, 1479s, 1435s,

1272s, 1094vs, 747s, 696s, 622s, 528s, 507m. Far-IR (cm^-1):

623vs, 583w, 529vs, 506vs, 499s, 444m, 425m, 404w, 376w, 256w,

CH(CH₃)₂), 1.32 (d, J_HH) 6.8 Hz, 24H, CH(CH₃)₂), 2.29 (br, 4H,

CH(CH₃)₂), 3.24 (sept, J_HH) 6.8 Hz, 4H, CH(CH₃)₂), 5.94 (s, 4H,

4-H(pz)), 6.25 (s, 4H, H₂C). ¹³C NMR (CDCl₃): 22.38 (CH(CH₃)₂),

23.16 (CH(CH₃)₂), 25.58 (CH(CH₃)₂), 27.63 (CH(CH₃)₂), 57.00

(H₂C), 99.28 (pz-4C), 152.1 (pz-3C), 160.6 (pz-5C).

[Cu{H₂C(3,5-iPr₂pz)₂}]₂(ClO₄)₂, [Cu(L1′′)]₂(ClO₄)₂(2ClO₄).

A solution of AgClO₄(110 mg, 0.532 mmol) in tetrahydrofuran

(10 cm³) was added to a solution of [Cu(L1′′)₂](CuCl₂) (1CuCl₂)

(221 mg, 0.266 mmol) in dichloromethane (15 cm³) in a glovebox.

After stirring for 1.5 h, the solution was filtered, and the filtrate

was concentrated under reduced pressure. Recrystallization from

dichloromethane/tetrahydrofuran/heptane (1:2:3) at -50 °C gave

colorless crystals. Yield: 98% (250 mg, 0.261 mmol). Anal. Calcd

for C₃₈H₆₄N₈Cl₂Cu₂O₈‚0.25CH₂Cl₂: C, 46.87; H, 6.63; N, 11.43.

Found: C, 46.68; H, 6.76; N, 11.59. IR (cm^-1): 3129w, 2968vs,

2933s, 2872s, 1546s, 1481m, 1461m, 1290m, 1267m, 1091vs, 622s.

Far-IR (cm^-1): 623vs, 552w, 536w, 464w, 418w, 336w, 298w,

211w, 197w, 186w, 168w, 160w. ¹H NMR (CD₂Cl₂): 1.15 (d, J_HH

) 6.9 Hz, 24H, CH(CH₃)₂), 1.18 (d, J_HH) 6.9 Hz, 24H, CH(CH₃)₂),

2.82 (sept, J_HH) 6.9 Hz, 4H, CH(CH₃)₂), 2.91 (sept, J_HH) 6.9

1

188m, 167w, 158m. H NMR (CDCl₃): 0.856 (d, J_HH) 6.9 Hz,

12H, CH(CH₃)₂), 1.33 (d, J_HH) 6.8 Hz, 12H, CH(CH₃)₂), 2.56

(sept, J_HH) 6.9 Hz, 2H, CH(CH₃)₂), 3.29 (sept, J_HH) 6.9 Hz,

2H, CH(CH₃)₂), 5.96 (s, 2H, 4-H(pz)), 6.51 (s, 2H, H₂C), 7.46-

7.53 (m, 15H, PPh₃). ¹³C NMR (CDCl₃): 23.02 (CH(CH₃)₂), 23.11

(CH(CH₃)₂), 25.49 (CH(CH₃)₂), 28.59 (CH(CH₃)₂), 57.44 (H₂C),

99.14 (pz-4C), 129.2 (PPh₃), 129.3 (PPh₃), 130.9 (PPh₃), 131.0

(PPh₃), 131.2 (PPh₃), 133.6 (PPh₃), 133.7 (PPh₃), 153.9 (pz-3C),

161.9 (pz-5C). ³¹P NMR (CDCl₃): 12.11 (s, PPh₃).

Hz, 4H, CH(CH₃)₂), 6.21 (s, 4H, 4-H(pz)), 6.70 (br, 4H, H₂C). ¹³

C

NMR (CD₂Cl₂): 22.52 (CH(CH₃)₂), 23.14 (CH(CH₃)₂), 26.83 (CH-

(CH₃)₂), 30.01 (CH(CH₃)₂), 61.08 (H₂C), 102.6 (pz-4C), 156.5 (pz-

3C), 164.0 (pz-5C).

[Cu{H₂C(3,5-iPr₂pz)₂}(PPh₃)](PF₆), [Cu(L1′′)(PPh₃)](PF₆)

(4PF₆). The preparation was carried out by the same method as

applied for 4ClO₄using [Cu(L1′′)(NCMe)](PF₆) (3PF₆) (120 mg,

0.212 mmol) and PPh₃(58 mg, 0.222 mmol) in dichloromethane

(15 cm³). Recrystallization from dichloromethane/diethyl ether (1:

2) at -50 °C gave colorless crystals. Yield: 38% (63.5 mg, 0.0807

mmol). Anal. Calcd for C₃₇H₄₇N₄CuF₆P₂: C, 56.45; H, 6.02; N,

7.12. Found: C, 56.23; H, 5.99; N, 7.00. IR (cm^-1): 3135w, 3054w,

2969s, 2934m, 2874m, 1548m, 1480s, 1436s, 1273s, 1097m,

1058m, 842vs, 746s, 695s, 558s, 531s. Far-IR (cm^-1): 643s, 620w,

558vs, 530s, 506s, 494s, 445m, 427w, 401w, 303w, 258w, 247w,

[Cu{H₂C(3,5-iPr₂pz)₂}]₂(BF₄)₂, [Cu(L1′′)]₂(BF₄)₂(2BF₄). The

preparation was carried out by the same method as applied for

2ClO₄using [Cu(L1′′)₂](CuCl₂) (1CuCl₂) (205 mg, 0.247 mmol)

in dichloromethane (15 cm³) and AgBF₄(96 mg, 0.493 mmol) in

tetrahydrofuran (10 cm³) in a glovebox. Recrystallization from

dichloromethane/tetrahydrofuran/heptane (1:2:3) at -50 °C gave

colorless crystals. Yield: 67% (154 mg, 0.165 mmol). Anal. Calcd

for C₃₈H₆₄N₈B₂Cu₂F₈‚0.25CH₂Cl₂: C, 48.11; H, 6.81; N, 11.73.

Found: C, 47.99; H, 6.84; N, 11.86. IR (cm^-1): 3136w, 2967vs,

2934s, 2872s, 1546s, 1478m, 1463m, 1290m, 1275m, 1057vs,

806m. Far-IR (cm^-1): 641s, 628w, 583w, 550w, 520vs, 471w,

1

176w, 161m. H NMR (CDCl₃): 0.873 (d, J_HH) 6.9 Hz, 12H,

1

413w, 253w, 171w, 156s. H NMR (CDCl₃): 1.22 (br, 24H, CH-

CH(CH₃)₂), 1.33 (d, J_HH) 6.8 Hz, 12H, CH(CH₃)₂), 2.56 (sept,

J_HH) 6.8 Hz, 2H, CH(CH₃)₂), 3.22 (sept, J_HH) 6.7 Hz, 2H,

CH(CH₃)₂), 5.98 (s, 2H, 4-H(pz)), 6.35 (s, 2H, H₂C), 7.47-7.52

(m, 15H, PPh₃). ¹³C NMR (CDCl₃): 22.99 (CH(CH₃)₂), 25.49 (CH-

(CH₃)₂), 28.61 (CH(CH₃)₂), 57.10 (H₂C), 99.28 (pz-4C), 129.3

(PPh₃), 129.4 (PPh₃), 130.8 (PPh₃), 131.0 (PPh₃), 133.5 (PPh₃),

133.6 (PPh₃), 153.9 (pz-3C), 162.0 (pz-5C). ³¹P NMR (CDCl₃):

12.20 (s, PPh₃), -138.9 (sept, PF₆).

(CH₃)₂), 1.26 (br, 24H, CH(CH₃)₂), 2.93 (br, 4H, CH(CH₃)₂), 3.06

(br, 4H, CH(CH₃)₂), 6.22 (s, 4H, 4-H(pz)), 6.79 (s, 4H, H₂C). ¹³C

NMR (CDCl₃): 22.37 (CH(CH₃)₂), 22.91 (CH(CH₃)₂), 26.25

(CH(CH₃)₂), 29.30 (CH(CH₃)₂), 61.37 (H₂C), 102.0 (pz-4C), 156.3

(pz-3C), 163.5 (pz-5C).

[Cu{H₂C(3,5-iPr₂pz)₂}(NCMe)](PF₆), [Cu(L1′′)(NCMe)](PF₆)

(3PF₆). A solution of L1′′ (212 mg, 0.669 mmol) in dichlo-

romethane (10 cm³) was added to a solution of [Cu(MeCN)₄](PF₆)

(274 mg, 0.736 mmol) in acetonitrile (20 cm³). After stirring for 1

h, the solvent was evaporated under vacuum. The resulting solid

was extracted with dichloromethane (25 cm³). The obtained solution

was evaporated under vacuum, and a yellowish white powder was

obtained. Recrystallization from dichloromethane/diethyl ether (1:

2) at -30 °C afforded colorless crystals. Yield: 72% (272 mg,

0.481 mmol). Anal. Calcd for C₂₁H₃₅N₅CuF₆P: C, 44.56; H, 6.23;

N, 12.37. Found: C, 44.53; H, 6.26; N, 12.45. IR (cm^-1): 2966s,

2933m, 2872m, 1545m, 1465m, 1384m, 1274s, 1183w, 1108w,

843vs, 558s. Far-IR (cm^-1): 645s, 559vs, 536w, 503w, 477m,

411w, 377w, 281w, 214w, 205w, 161m. ¹H NMR (CD₂Cl₂): 1.20

(d, J_HH) 7.0 Hz, 12H, CH(CH₃)₂), 1.22 (d, J_HH) 6.8 Hz, 12H,

CH(CH₃)₂), 2.31 (s, 3H, CH₃CN), 2.96 (sept, J_HH) 6.8 Hz, 4H,

CH(CH₃)₂), 5.94 (s, 2H, 4-H(pz)), 5.99 (s, 2H, H₂C). ¹³C NMR

(CD₂Cl₂): 23.14 (CH(CH₃)₂), 26.27 (CH(CH₃)₂), 28.83 (CH(CH₃)₂),

57.08 (H₂C), 100.4 (pz-4C), 153.2 (pz-3C), 162.7 (pz-5C).

[Cu{H₂C(3,5-iPr₂pz)₂}(PPh₃)](ClO₄), [Cu(L1′′)(PPh₃)](ClO₄)

(4ClO₄). Method a: Dichloromethane (15 cm³) was added to a

[{Cu{H₂C(3,5-iPr₂pz)₂}(CO)}₂(µ-ClO₄)](ClO₄), [{Cu(L1′′)-

(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄). [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) (150

mg, 0.156 mmol) was dissolved in dichloromethane (15 cm³) in a

glovebox. The solution was cooled to -78 °C in an argon

atmosphere, and the argon gas was then replaced by a CO

atmosphere. The solution was allowed to warm up to room

temperature. After stirring for 3 days, the solution was filtered and

the filtrate was concentrated under reduced pressure. Recrystalli-

zation from dichloromethane/heptane (1:2) at -50 °C under a CO

atmosphere gave colorless crystals. Yield: 41% (64.8 mg, 0.0638

mmol). Anal. Calcd for C₄₀H₆₄N₈Cl₂Cu₂O₁₀‚2CH₂Cl₂: C, 42.58;

H, 5.78; N, 9.46. Found: C, 42.61; H, 5.97; N, 9.47. IR (cm^-1):

3135w, 3071w, 2970vs, 2934s, 2872m, 2104vs, 1545s, 1462s,

1397s, 1385s, 1278vs, 1095vs, 1029s, 808m, 728s, 623s. Far-IR

(cm^-1): 624vs, 587w, 542m, 516m, 473w, 415m, 376w, 324w,

1

301w, 201m, 188m. H NMR (CD₂Cl₂): 1.26 (d, J_HH) 6.8 Hz,

24H, CH(CH₃)₂), 1.27 (d, J_HH) 6.8 Hz, 24H, CH(CH₃)₂), 2.97

(sept, J_HH) 6.8 Hz, 4H, CH(CH₃)₂), 3.05 (sept, J_HH) 6.9 Hz,

4H, CH(CH₃)₂), 6.00 (s, 4H, 4-H(pz)), 6.06 (s, 4H, H₂C). ¹³C NMR

10610 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

(CD₂Cl₂): 22.90 (CH(CH₃)₂), 22.93 (CH(CH₃)₂), 26.03 (CH(CH₃)₂),

28.96 (CH(CH₃)₂), 55.98 (H₂C), 99.96 (pz-4C), 152.8 (pz-3C),

162.3 (pz-5C), 172.9 (CO).

133.8 (PPh₃), 153.1 (pz-3C), 161.5 (pz-5C). ³¹P NMR (CDCl₃):

13.79 (br, s, PPh₃).

[Cu{HB(3,5-iPr₂pz)₃}(PPh₃)], [Cu(L1)(PPh₃)] (9). The prepa-

ration was carried out by the same method¹⁰as applied for 4ClO₄

using [Cu(L1)(NCMe)]⁹(123 mg, 0.215 mmol) and PPh₃(59.2 mg,

0.226 mmol) in dichloromethane (15 cm³) in a glovebox. Recrys-

tallization from dichloromethane/acetonitrile (1:2) at -30 °C gave

colorless crystals. Yield: 36% (61.0 mg, 0.0771 mmol). Anal. Calcd

for C₄₅H₆₁N₆BCuP: C, 68.30; H, 7.77; N, 10.62. Found: C, 67.97;

H, 7.78; N, 10.37. IR (cm^-1): 3071w, 3055w, 2962vs, 2927s,

2865s, 2531m, 1536s, 1469s, 1458s, 1434s, 1392s, 1379s, 1295s,

1172vs, 1041s, 783s, 698s, 527s. Far-IR (cm^-1): 528vs, 512vs,

489s, 437m, 413m, 403m, 395m, 305w, 261w. ¹H NMR (CDCl₃):

0.72 (d, J_HH) 6.8 Hz, 18H, CH(CH₃)), 1.24 (d, J_HH) 6.9 Hz,

18H, CH(CH₃)₂), 2.69 (sept, J_HH) 6.8 Hz, 3H, CH(CH₃)₂), 3.53

(sept, J_HH) 6.7 Hz, 3H, CH(CH₃)₂), 5.67 (s, 3H, 4-H(pz)), 7.25-

7.36, 7.61-7.64 (m, 15H, PPh₃). ¹³C NMR (CDCl₃): 23.77 (CH-

(CH₃)₂), 23.88 (CH(CH₃)₂), 26.17 (CH(CH₃)₂), 28.42 (CH(CH₃)₂),

95.09 (pz-4C), 128.39 (PPh₃), 128.45 (PPh₃), 129.38 (PPh₃), 134.23

(PPh₃), 134.33 (PPh₃), 135.34 (PPh₃), 135.55 (PPh₃), 154.48 (pz-

3C), 157.99 (pz-5C). ³¹P NMR (CDCl₃): 10.73 (br, s, PPh₃).

[{Cu{H₂C(3,5-iPr₂pz)₂}}₂(µ-OH)₂(µ-ClO₄)₂], [{Cu(L1′′)}₂(µ-

OH)₂(µ-ClO₄)₂] (6). [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) (95.5 mg, 0.0996

mmol) was dissolved in dichloromethane in a Schlenk tube and

cooled to -78 °C in an argon atmosphere. The argon gas was then

replaced by an O₂atmosphere, and the solution was allowed to

warm up to room temperature. After stirring for 24 h, the solvent

was concentrated under reduced pressure. Recrystallization from

dichloromethane/heptane (1:2) at -30 °C gave two kinds of crystals.

The main product was 6 obtained as blue crystals, which were

manually separated. A small amount of green crystals remained,

the nature of which is not known. Yield: 24% (24.0 mg, 0.0242

mmol). Anal. Calcd for C₃₈H₆₆N₈Cl₂Cu₂O₁₀: C, 45.96; H, 6.70;

N, 11.28. Found: C, 45.82; H, 6.54; N, 11.12. IR (cm^-1): 3536s,

2969vs, 2932s, 2871m, 1549s, 1484s, 1467s, 1404m, 1386m, 1275s,

1183m, 1096vs, 812m, 623m. Far-IR (cm^-1): 624vs, 548w, 519m,

460w, 422w, 306w, 213w. UV-vis absorption (CH₂Cl₂, λ_max, nm)

(ꢀ/M^-1cm^-1) 256 (8950), 355 (1360), 563 (100). ESR (CH₂Cl₂)

silent.

[Cu{H₂C(3,5-iPr₂pz)₂}Cl₂], [Cu(L1′′)Cl₂] (7).¹⁰Method a: [Cu-

(L1′′)₂](CuCl₂) (1CuCl₂) (110 mg, 0.132 mmol) was dissolved in

dichloromethane in a Schlenk tube and stirred in air. The clear

solution of 1CuCl₂turned dark yellow over 24 h. The solution

was filtered, and the filtrate was concentrated under reduced

pressure. Recrystallization from dichloromethane/heptane (1:2) at

-30 °C gave orange crystals. Yield: 51% (30.5 mg, 0.0676 mmol).

Method b: 7 could also be obtained as follows: L1′′ (202 mg,

0.639 mmol) dissolved in dichloromethane (15 cm³) was added to

a solution of CuCl₂‚2H₂O (122 mg, 0.716 mmol) in acetone (20

cm³). The color of the solution gradually turned to dark yellow.

After stirring for 1 h, the solvent was removed under vacuum. The

residue was extracted with dichloromethane, and the obtained

organic solution was filtered. The organic solvent was then

concentrated under reduced pressure. Recrystallization from dichlo-

romethane/heptane (1:2) at -30 °C gave orange crystals. Yield:

72% (208 mg, 0.461 mmol). Anal. Calcd for C₁₉H₃₂N₄Cl₂Cu: C,

50.61; H, 7.15; N, 12.42. Found: C, 50.36; H, 7.17; N, 12.24. IR

(cm^-1): 3123w, 2967vs, 2931s, 2870m, 1547s, 1462s, 1441s, 1398s,

1384s, 1279s, 1184m, 1062m, 810s, 657w. Far-IR (cm^-1): 645m,

629m, 590w, 545s, 470w, 315vs, 181m, 171m. UV-vis absorption

(CH₂Cl₂, λ_max, nm) (ꢀ/M^-1cm^-1) 289 (3720), 375 (1330), 911 (140).

ESR (CH₂Cl₂:CH₂ClCH₂Cl ) 1:1) g_av) 2.14.

[Cu{HC(3,5-iPr₂pz)₃}(PPh₃)](ClO₄), [Cu(L1′)(PPh₃)](ClO₄)

(8ClO₄). The preparation was carried out by the same method as

applied for 4ClO₄using [Cu(L1′)(OClO₃)]⁹(173 mg, 0.275 mmol)

and PPh₃(75.6 mg, 0.288 mmol) in dichloromethane (15 cm³) in

a glovebox. Recrystallization from dichloromethane/diethyl ether

(1:2) at -50 °C gave colorless crystals. Yield: 47% (116 mg, 0.130

mmol). Anal. Calcd for C₄₆H₆₁N₆ClCuO₄P: C, 61.94; H, 6.89; N,

9.42. Found: C, 61.75; H, 6.93; N, 9.30. IR (cm^-1): 3130w, 3055w,

2965vs, 2931s, 2869m, 1556s, 1469s, 1435s, 1397s, 1385s, 1291s,

1233s, 1094vs, 825s, 697s, 622m, 530m. Far-IR (cm^-1): 623vs,

531vs, 505vs, 497s, 444w, 425w, 395w, 366w, 305w, 273w, 258w.

¹H NMR (CDCl₃): 0.83 (d, J_HH) 6.8 Hz, 18H, CH(CH₃)), 1.38

(d, J_HH) 6.9 Hz, 18H, CH(CH₃)₂), 2.62 (sept, J_HH) 6.7 Hz, 3H,

CH(CH₃)₂), 3.49 (sept, J_HH) 6.6 Hz, 3H, CH(CH₃)₂), 5.98 (s, 3H,

4-H(pz)), 7.44-7.54 (m, 15H, PPh₃), 8.27 (s, 1H, HC). ¹³C NMR

(CDCl₃): 22.78 (CH(CH₃)₂), 23.34 (CH(CH₃)₂), 25.88 (CH(CH₃)₂),

28.51 (CH(CH₃)₂), 66.93 (HC), 99.25 (pz-4C), 129.0 (PPh₃), 129.1

(PPh₃), 130.6 (PPh₃), 132.8 (PPh₃), 133.1 (PPh₃), 133.7 (PPh₃),

Reaction of [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) with ¹³CO. The ¹³CO-

labeled complex was prepared by the same method as the

corresponding unlabeled complex 5ClO₄using 2ClO₄and ¹³CO.

Reaction of [Cu(L1′′)(NCMe)](PF₆) (3PF₆) with O₂. In a

Schlenk tube, 3PF₆was dissolved in dichloromethane and cooled

to -78 °C under an argon atmosphere. The argon was then replaced

by dioxygen, and the solution was allowed to warm up to room

temperature. After stirring for 24 h, the solvent was removed under

vacuum.

X-ray Data Collection and Structural Determination. Crystal

data and refinement parameters for the investigated ligand (L1′′)

and the copper complexes are given in Tables 1 and 2. The

diffraction data for all of the complexes were measured on a Rigaku/

MSC Mercury CCD system with graphite monochromated Mo KR

(λ ) 0.71069 Å) radiation at low temperatures (-69 to -90 °C).

Each crystal was mounted on the tip of a glass fiber using a heavy-

weight oil. The unit cell parameters of each crystal from six image

frames were retrieved using the Rigaku Daemon software and

refined with CrystalClear on all of the observed reflections.¹⁸Data

using 0.5° intervals in φ and ω for 20-45 s/frame were collected

with a maximum resolution of 0.77 Å (55°, 744 oscillation images).

The data were corrected for Lorentz and polarization effects. An

empirical absorption correction was applied for each complex.^18-20

Structures were solved by direct methods (SIR 92²¹or SIR 97²²).

The positions of the metal atoms and their first coordination sphere

were located from the E-map; other non-hydrogen atoms were found

in alternating difference Fourier syntheses.²³Least-squares refine-

ment was carried out anisotropically during the final cycles

(18) Pflugrath, J. W. CrystalClear, version 1.3; Acta Crystallogr. Sect. D

1999, 55, 1718-1725.

(19) CrystalStructure 3.70: Crystal Structure Analysis Package; Rigaku

and Rigaku/MSC: Tokyo, 2005.

(20) Watkin, D. J.; Prout, C. K.; Carruthers, J. R.; Betteridge, P. W. Crystal

Issue 10; Chemical Crystallography Laboratory: Oxford, U.K.,

1996.

(21) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A. SIR 92;

J. Appl. Crystallogr. 1993, 26, 343-350.

(22) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.; Giaco-

vazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna,

R. SIR 97; J. Appl. Crystallogr. 1999, 32, 115-119.

(23) Beurskens, P. T.; Admiraal, G.; Beurskens, G.; Bosman, W. P.; de

Gelder, R.; Israel, R.; Smits, J. M. M. DIRDIF-99, Technical Report

of the Crystallography Laboratory; University of Nijmegen: Nijmegen,

The Netherlands, 1999.

Inorganic Chemistry, Vol. 46, No. 25, 2007 10611

Fujisawa et al.

Table 1. Crystallographic Data of Ligand L1′′ and of the Complexes 1CuCl₂, 1ClO₄, 2ClO₄, 2BF₄, and 3PF₆

1ClO₄

2ClO₄

2BF₄

complex

formula

L1′′

1CuCl₂

·5CHCl₃

·2CH₂Cl₂·C₄H₈O

·2CH₂Cl₂·C₄H₈O·2H₂O

3PF₆

C₁₉H₃₂N₄

316.49

C₃₈H₆₄Cl₂Cu₂N₈C₄₃H₆₉Cl₁₆CuN₈O₄C₄₄H₇₆C_l6Cu₂N₈O₉C₄₄H₈₀B₂Cl₄Cu₂F₈N₈O₃C₂₁H₃₂CuF₆N₅P

830.98

fw

1392.86

P2₁/n (no. 14)

14.1946(4)

23.0068(5

20.3309(5)

90

97.0640(4)

90

6589.1(3)

4

1200.94

1211.68

563.03

space group

a (Å)

b (Å)

P2₁/c (no. 14)

8.8156(10)

19.959(2)

10.9411(15)

90

96.308(7)

90

1913.4(4)

4

P1h (no. 2)

8.9380(2)

9.2290(2)

26.3640(9)

80.693(5)

81.471(5)

84.762(5)

2117.32(10)

2

C2/c (no. 15)

20.5140(17)

13.6093(9)

21.0771(18)

90

100.4450(14)

90

5786.8(8)

4

C2/c (no. 15)

20.4617(14)

13.6674(7)

20.8773(10)

90

99.7430(10)

90

5754.3(6)

4

Pna2₁(no. 33)

20.370(10)

13.493(7)

9.657(5)

90

c (Å)

R (deg)

â (deg)

γ (deg)

V (Å³)

Z

90

2654(2)

4

reflns collected/unique 15 365/4362

16 297/9179

51 519/15 039

19 083/6468

4263 (I > 3σ(I))

359

22 359/6564

4618 (I > 3σ(I))

359

20 336/3156

2442 (I > 3σ(I))

339

obsd reflns

3240 (I > 3σ(I)) 4920 (I > 3σ(I)) 4533 (I > 5σ(I))

240

reflns params

518

0.069, 0.082

744

0.058, 0.068

final R, Rw^a

0.054, 0.051

0.095, 0.107

0.076, 0.094

0.055, 0.054

^aR ) ∑||F_o| - |F_c||/∑|F_o|; Rw ) [(∑w(|F_o| - |F_c|)²/∑wF_o)]^1/2, w ) 1/σ²(|F_o|).

2

Table 2. Crystallographic Data of the Complexes 4ClO₄, 4PF₆, 5ClO₄, 6, 7, and 8ClO₄

5ClO₄

6

7

8ClO₄

complex

formula

4ClO₄

4PF₆

·2CH₂Cl₂

·6CH₂Cl₂

·2CH₂Cl₂

·1.5CH₂Cl₂

C₃₇H₄₇ClCuN₄O₄P C₃₇H₄₇CuF₆N₄P₂C₄₂H₆₈Cl₆Cu₂N₈O₁₀C₄₄H₇₈Cl₁₄Cu₂N₈O₁₀C₂₁H₃₆C_l6CuN₄

741.78

C₄₇.₅H₆₄Cl₄CuN₆O₄P

1019.40

fw

787.29

1184.86

1502.58

620.81

space group

a (Å)

b (Å)

P1h (no. 2)

10.3789(4)

18.1638(7)

20.1690(7)

98.9760(17)

99.6730(17)

90.6630(15)

3699.5(2)

4

P1h (no. 2)

10.49370(10)

13.2790(2)

14.9003(2)

82.208(5)

76.009(5)

71.340(4)

1904.96(4)

2

P2₁/c (no. 14)

12.6250(2)

19.2085(4)

24.0421(5)

90

105.0230(2)

90

5631.11(19)

4

P1h (no. 2)

9.76550(10)

12.83230(10)

15.41210(10)

68.413(8)

78.870(9)

71.475(8)

1696.47(2)

1

P1h (no. 2)

12.931(5)

13.192(5)

20.405(9)

78.57(2)

77.268(2)

64.711(16)

3048.4(20)

4

Pbca (no. 61)

19.3853(4)

20.7432(4)

25.8555(5)

90

c (Å)

R (deg)

â (deg)

γ (deg)

V (Å³)

Z

90

10 396.8(4)

8

reflns collected/unique 29 531/16 422

15 440/8426

5389 (I > 3σ(I)) 4139 (I > 5σ(I))

498

43 053/12 713

13 166/7509

4355 (I > 5σ(I))

391

24 687/13 686

15 216 (I > 5σ(I)) 5186 (I > 5σ(I))

649

78 630/11 886

obsd reflns

8300 (I > 3σ(I))

959

reflns params

681

0.042, 0.053

660

final R, Rw^a

0.068, 0.075

0.057, 0.068

0.068, 0.072

0.085, 0.105

0.1656, 0.1995

2

^aR ) ∑||F_o| - |F_c||/∑|F_o|; Rw ) [(∑w(|F_o| - |F_c|)²/∑wF_o)]^1/2, w ) 1/σ²(|F_o|).

(CrystalStructure).^19,20Hydrogen atoms were placed in calculated

positions but not refined anisotropically. Strong remaining peaks

in 1ClO₄, 2ClO₄, 2BF₄, 6, 7, and 8ClO₄are due to disordered

solvent molecules. Hydrogen atoms of solvent water molecules in

2BF₄were not placed. Chlorine atoms of one of the solvent

molecules CHCl₃in 1ClO₄and CH₂Cl₂in 2ClO₄were highly

disordered; therefore, sets of them have equal occupancy. The

structure of 8ClO₄could not be determined at a very good quality

because of a high degree of disorder even at low temperature

(-90 °C). The absolute configuration of 3PF₆was confirmed from

the values of the Flack parameters.²⁴

zolyl)methane have been prepared,²⁶which further underlines

the fundamental importance of the coordination chemistry

of these ligands. The synthesis of H₂C(3,5-iPr₂pz)₂(L1′′) was

reported previously in ref 3f. We have prepared this ligand

by an alternative method using an autoclave, and, this way

we were able to obtain crystals for X-ray analysis. This

method can lead to higher yields and substantially less

darkening of the reaction mixture, compared to the procedure

reported before.^3f

To avoid the reaction of the copper(I) complexes with

dioxygen, all of the compounds were prepared in a glovebox.

Schematic drawings of all of the transformations performed

in this work are presented in Scheme 1. Details are available

from the Experimental Section and Section C. Complexes

[Cu(L1′′)₂](CuCl₂) (1CuCl₂) and [Cu(L1′′)(NCMe)](PF₆)

(3PF₆) were obtained from L1′′ and the corresponding

copper(I) salts. [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) and [Cu(L1′′)]₂-

(BF₄)₂(2BF₄) were synthesized by the reaction of 1CuCl₂

with AgClO₄and AgBF₄, respectively. [Cu(L1′′)(PPh₃)]-

(ClO₄) (4ClO₄) and [Cu(L1′′)(PPh₃)](PF₆) (4PF₆) were

Results and Discussion

A. Syntheses and Structures. General Synthetic Pro-

cedures. The coordination chemistry of bis(pyrazolyl)-

methanes toward transition-metal ions has received increasing

attention, and many transition-metal complexes containing

bis(pyrazolyl)methanes have been synthesized and character-

ized in recent years.^3-7,10,25,26These ligands are especially

valuable for the preparation of two- and three-coordinate

transition-metal complexes, which are active in catalysis due

to their unsaturated coordination environment.⁴In addition,

a number of heteroscorpionate ligands based on bis(pyra-

(25) (a) Michaud, A.; Fontaine, F.-G.; Zargarian, D. Inorg. Chem. Acta

2006, 359, 2592-2598. (b) Boho, N.; Zargarian, D. Inorg. Chem. 2007,

46, 299-308.

(26) Otero, A.; Ferna´ndes-Baeza, J.; Antinolo, A.; Tejeda, J. Lara-Sa´nchez,

A. Dalton Trans. 2004, 1499-1510.

(24) Flack, H. D. Acta. Crystallogr., Sect. A 1983, 39, 876-878.

10612 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

Scheme 1. Schematic Drawings of All of the Reactions Performed in This Work

obtained from the reaction of 2ClO₄and 3PF₆with PPh₃,

respectively. [{Cu(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄), [Cu-

(L1′′)₂](ClO₄) (1ClO₄), and [{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂]

(6) were synthesized by the reaction of 2ClO₄with CO,

MeCN, and O₂, respectively. The copper(II) chloro complex

[Cu(L1′′)Cl₂] (7) was synthesized by stirring 1CuCl₂in

dichloromethane in air. [Cu(L1′)(PPh₃)](ClO₄) (8ClO₄) and

[Cu(L1)(PPh₃)] (9) were generated by the reaction of

PPh₃with [Cu(L1′)(OClO₃)]⁹and [Cu(L1)(NCMe)],⁹respec-

tively.

Structures and Insight from Complex Syntheses. An

ORTEP drawing of the crystal structure of ligand L1′′ is

shown in Figure S1 in the Supporting Information (structural

parameters: see Table S1 in the Supporting Information). The

averaged C(methylene)-N distance and N-C(methyl-

ene)-N angle are 1.450 Å and 113.7°, respectively. These

values are almost identical to those of the corresponding

neutral tridentate ligand L1′ (1.447 Å and 112.7°).

The copper complexes 1CuCl₂, 1ClO₄, 2ClO₄, 2BF₄,

3PF₆, 4ClO₄, 4PF₆, 5ClO₄, 6, and 7 with L1′′ were also

obtained as crystals suitable for X-ray crystallography. The

ORTEP drawings of these complexes are shown in Figures

2 and 3, with exception of 1ClO₄, 2BF₄, and 4PF₆, which

are presented in Figure S2 in the Supporting Information.

Selected bond distances and angles are listed in Table 3. The

preliminary structure of 8ClO₄is shown in Figure S3 in the

Supporting Information (structural parameters: see Table S2

in the Supporting Information).

In complex [Cu(L1′′)₂](CuCl₂) (1CuCl₂) containing the

unusual counterion (Cu^ICl₂)^-, the copper(I) ion is in a

distorted tetrahedral environment with two bidentate

ligands L1′′ bound, showing Cu-N distances ranging from

1.993(5) to 2.120(6) Å and N-Cu-N angles ranging from

95.6(2) to 134.6(2)° (Figure 2, top left). The overall structure

is similar to that of the previously reported complex [Cu-

{H₂C(pz)₂}₂](ClO₄).⁵The dihedral angle between the two

chelating planes (N11-Cu1-N21 and N31-Cu1-N41) is

94.1°, which is somewhat larger than the 90.0° expected for

a tetrahedral geometry. Importantly, a three-coordinate

complex of type [Cu(L1′′)(Cl)] is therefore avoided by

copper(I) in this case by undergoing a ligand disproportion-

ation reaction:

2 [Cu(L1′′)(Cl)] f [Cu(L1′′)₂]⁺+ (CuCl₂) ^-

Correspondingly, treatment of the tridentate ligand L1′

with Cu^ICl results in the clean formation of [Cu(L1′)(Cl)].⁹

A similar ligand disproportionation process is observed

during the formation of [Cu(L1′′)₂](ClO₄) (1ClO₄), is shown

in Figure S2, top in the Supporting Information. In this case,

the formation of [Cu(L1′′)(NCMe)]⁺is avoided (cf. Scheme

1) by the generation of [Cu(L1′′)₂]⁺and [Cu(NCMe)₄]⁺(not

-

isolated). 1ClO₄with ClO₄as a counterion has a very

similar structure to that of 1CuCl₂(cf. Table 3). The dihedral

angle of the two chelating planes (N11-Cu1-N21 and

N31-Cu1-N41) is only 91.3° in this case. The difference

Inorganic Chemistry, Vol. 46, No. 25, 2007 10613

Fujisawa et al.

Figure 2. Crystal structures of the cationic parts of [Cu(L1′′)₂](CuCl₂) (1CuCl₂) (top left), [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) (top right), [Cu(L1′′)(NCMe)](PF₆)

(3PF₆) (bottom left), and [Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄) (bottom right). The hydrogen atoms are omitted for clarity. All of the ellipsoids are drawn at the

50% probability level. In the case of 4ClO₄, two crystallographically independent molecules are present, whose structural features are essentially identical.

Molecule 1 is presented here.

between 1CuCl₂and 1ClO₄relates to somewhat stronger

interactions between the copper(I) center and the counterion

in 1ClO₄.

Because the N-Cu-N angles of [Cu(HBpz₃)]₂and [Cu{HB-

(3,5-Me₂pz)₃}]₂are not linear (N-Cu-N ) ∼140°), the

Cu‚‚‚Cu distances are distinctively shorter in these cases

compared to 2ClO₄and 2BF₄(Cu‚‚‚Cu distance: 2.66 Å in

[Cu(HBpz₃)]₂, 2.51 Å in [Cu{HB(3,5-Me₂pz)₃}]₂, 2.97 Å in

2ClO₄, and 2.91 Å in 2BF₄). These differences again relate

to the different coordination number (bidentate vs tridentate)

and charge (neutral vs anionic) of L1′′ and these hydrotris-

(pyrazolyl)borate ligands. Correspondingly, using the neutral

tridentate ligand L1′, treatment of [Cu(L1′)(Cl)] with AgClO₄

results in the straightforward formation of the tetrahedral

complex [Cu(L1′)(OClO₃)].⁹Therefore, to obtain the desired

three-coordinate complexes of type [Cu(L1′′)(X)]Z with a

neutral bidentate ligand, stronger ligands such as acetonitrile

or triphenylphosphine are needed to prevent this dimer

formation. Note that related complexes with less-hindered

bis(pyrazolyl)methanes (H₂Cpz₂or H₂C(3,5-Me₂pz)₂) have

Interestingly, the molecular structures of [Cu(L1′′)]₂(ClO₄)₂

(2ClO₄) and [Cu(L1′′)]₂(BF₄)₂(2BF₄) correspond to binuclear

linear-coordinated copper(I) complexes with two bridging

-

L1′′ ligands and two ClO₄or two BF₄counterions,

respectively (Figure 2, top right; and Figure S2, center in

the Supporting Information). Importantly, the N-Cu-N

angles are close to 180° in 2ClO₄and 2BF₄. In this case,

the formation of the three-coordinate complex [Cu(L1′′)-

(ClO₄)] is avoided by dimerization (ligand disproportionation

does not happen here because [Cu(ClO₄)₂]^-with the weak

ligand perchlorate is unstable). The same situation leading

to the formation of [Cu(L1′′)]₂²⁺is encountered with the non-

-

coordinating counterion BF₄. Iber and co-workers previ-

ously reported the structures of binuclear linear-coordinated

copper(I) complexes such as [Cu(HBpz₃)]₂and [Cu{HB(3,5-

Me₂pz)₃}]₂.²⁷In these cases, monoanionic tridentate ligands

(27) (a) Arcus, C. S.; Wilkinson, J. L.; Mealli, C.; Marks, T. J.; Iber, J. A.

J. Am. Chem. Soc. 1974, 96, 7564-7565. (b) Mealli, C.; Arcus, C.

S.; Wilkinson, J. L.; Marks, T. J.; Iber, J. A. J. Am. Chem. Soc. 1976,

98, 711-718.

-

such as HBpz₃and HB(3,5-Me₂pz)₃have been applied.

10614 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

Figure 3. Crystal structures of the cationic parts of [{Cu(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄) (top), [{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂] (6) (bottom left), and

[Cu(L1′′)Cl₂] (7) (bottom right). The hydrogen atoms are omitted for clarity. All of the ellipsoids are drawn at the 50% probability level. In the case of 7,

two crystallographically independent molecules are present, whose structural features are essentially identical. Molecule 1 is presented here.

not been obtained, indicating that the alkyl substitution of

L1′′ is also critical for the nature of the complexes formed.^5,6

much steric repulsion, which leads to similar Cu^I-N(MeCN)

distances in all of these complexes.

The copper(I) acetonitrile complex 3PF₆finally has the

desired trigonal-planar structure (total angle around the

copper(I) ion: 360°) as shown in Figure 2, bottom left. As

mentioned above, the formation of this complex is enabled

by the strong ligand acetonitrile combined with the strictly

The molecular structures of [Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄)

and [Cu(L1′′)(PPh₃)](PF₆) (4PF₆) with the strong ligand

triphenylphosphine are both trigonal planar (total angles

around the copper(I) ion: 357.2° in 4ClO₄and 358.2° in

4PF₆) as shown in Figure 2, bottom right and in Figure S2,

bottom in the Supporting Information. Note that, in the case

of 4ClO₄, two crystallographically independent complex

molecules exist. The copper(I) ion is slightly deviated from

the N-N-P(PPh₃) plane by 0.170 and 0.208 Å in 4ClO₄,

and by 0.155 Å in 4PF₆toward the counterion (Cu‚‚‚O-

(ClO₄): 4.83 and 5.25 Å in 4ClO₄, Cu‚‚‚F(PF₆): 4.45 Å

in 4PF₆). The steric hindrance between the iPr groups of

L1′′ and the bulky PPh₃ligand leads to a slight lengthening

of the average Cu-N distance (2.008 Å in 4ClO₄and 2.006

Å in 4PF₆) and asymmetric bidentate coordination of L1′′

(∆ ) 0.042 (Cu1) and 0.028 Å (Cu2) in 4ClO₄and 0.035 Å

in 4PF₆) as compared to the three-coordinate acetonitrile

complex 3PF₆(Cu-N ) 1.997 Å and ∆ ) 0.004 Å).

Interestingly, the bulky iPr groups of L1′′ seem to keep the

counteranion away from copper, and, this way, play an

important rolein achieving three-coordinate copper com-

-

non-coordinating counterion PF₆. The geometry of this

complex is very similar to [Cu{H₂C(3,5-Me₂pz)₂}(NCMe)]-

(ClO₄) published before with a less-hindered bidentate

ligand.⁶The deviation of the copper(I) ion from the N11-

N21-N31 plane in 3PF₆is only 0.008 Å, which is smaller

than the value of 0.078 Å observed for [Cu{H₂C(3,5-Me₂-

pz)₂}(NCMe)](ClO₄). This might relate to the stronger

-

interaction of copper(I) with perchlorate (Cu‚‚‚O(ClO₄) )

3.181 Å) in the latter case compared to the Cu(I)‚‚‚F

-

interaction in 3PF₆(Cu1‚‚‚F45(PF₆) ) 4.939(4) Å). Fur-

thermore, the difference in alkyl substituents of these ligands

seems to contribute to this difference. The Cu^I-N(MeCN)

distance of 1.870(5) Å in 3PF₆is similar to the reported

values for [Cu(L1′)(NCMe)](PF₆)⁹(1.887(5) Å) and [Cu-

{H₂C(3,5-Me₂pz)₂}(NCMe)](ClO₄)⁶(1.879(5) Å). Because

of the small size of MeCN, this ligand does not experience

Inorganic Chemistry, Vol. 46, No. 25, 2007 10615

Fujisawa et al.

Table 3. Selected Bond Distances (Å) and Angles (Degrees) for Complexes Characterized by X-ray Crystallography^a

[Cu(L1′′)₂](CuCl₂) (1CuCl₂)

Cu1-N11

1.999(5)

2.120(6)

107.6(2)

95.6(2)

Cu1-N21

N11-Cu1-N21

N21-Cu1-N31

2.209(5)

98.6(2)

112.3(2)

Cu1-N31

N11-Cu1-N31

N21-Cu1-N41

1.993(5)

134.6(2)

105.4(2)

Cu1-N41

N11-Cu1-N41

N31-Cu1-N41

[Cu(L1′′)₂](ClO₄) (1ClO₄)^b

Cu1-N11

2.186(7)

2.001(6)

109.9(2)

97.2(2)

Cu1-N21

N11-Cu1-N21

N21-Cu1-N31

1.972(5)

96.1(2)

111.5(2)

Cu1-N31

N11-Cu1-N31

N21-Cu1-N41

2.112(6)

101.7(2)

136.4(2)

Cu1-N41

N11-Cu1-N41

N31-Cu1-N41

[Cu(L1′′)]₂(ClO4)₂(2ClO₄)^c

Cu1-N11

1.868(5)

176.0(2)

Cu1-N21′

1.874(5)

Cu1‚‚‚Cu1′

2.9714(12)

2.9061(9)

N11-Cu1-N21′

[Cu(L1′′)]₂(BF₄)₂(2BF₄)^b,c

Cu1-N21′

Cu1-N11

N11-Cu1-N21

1.878(3)

174.58(18)

1.874(3)

[Cu(L1′′)(NCMe)](PF₆) (3PF₆)

Cu1-N11

N11-Cu1-N21

Cu1-N31-C31

1.999(5)

96.1(2)

175.3(6)

Cu1-N21

N11-Cu1-N31

1.995(5)

138.0(2)

Cu1-N31

N21-Cu1-N31

1.870(5)

125.8(2)

[Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄)

Cu1-N11

2.024(5)

1.998(6)

93.3(2)

93.4(2)

Cu1-N21

1.982(6)

2.026(5)

122.87(18)

139.60(16)

Cu1-P31

2.171(2)

2.177(2)

141.59(16)

123.68(19)

Cu2-N51

Cu2-N61

Cu2-P61

N11-Cu1-N21

N51-Cu2-N61

N11-Cu1-P31

N51-Cu2-P61

N21-Cu1-P31

N61-Cu2-P61

[Cu(L1′′)(PPh₃)](PF₆) (4PF₆)^b

Cu1-N11

N11-Cu1-N21

1.988(3)

93.60(15)

Cu1-N21

N11-Cu1-P31

2.023(4)

140.67(13)

Cu1-P31

N21-Cu1-P31

2.1726(12)

123.95(10)

[{Cu(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄)

Cu1-N11

2.017(6)

2.252(5)

1.802(10)

1.115(13)

126.5(3)

105.6(3)

128.6(3)

98.7(2)

Cu1-N21

2.030(7)

1.989(6)

2.311(5)

97.3(3)

Cu1-C31

1.816(10)

2.002(6)

1.104(12)

125.1(3)

99.9(2)

96.2(2)

97.6(2)

178.6(10)

Cu1-O71

Cu2-N41

Cu2-N51

Cu2-C61

C61-O62

N21-Cu1-C31

O71-Cu1-C31

N41-Cu2-C61

N51-Cu2-O72

Cu2-O72

C31-O31

N11-Cu1-N21

N11-Cu1-O71

Cu1-C31-O31

N51-Cu2-C61

O72-Cu2-C61

N11-Cu1-C31

N21-Cu1-O71

N41-Cu2-N51

N41-Cu2-O72

Cu2-C61-O62

95.8(2)

178.3(9)

125.8(3)

102.9(3)

[{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂] (6)^d

Cu1-N11

2.018(6)

1.921(3)

Cu1-N21

1.995(4)

2.428(4)

90.3(2)

Cu1-O1

1.922(4)

3.137(5)

Cu1-O1′

Cu1-O31

Cu1-O32′

Cu1···Cu1′

2.9466(10)

95.80(19)

79.89(18)

99.84(19)

94.54(19)

81.19(15)

N11-Cu1-N21

N21-Cu1-O1

Cu1-O1-Cu1′

N21-Cu1-O31

O1-Cu1-O32′

N11-Cu1-O1

N21-Cu1-O1′

N11-Cu1-O31

N21-Cu1-O32′

O31-Cu1-O1′

172.03(19)

170.83(18)

92.41(19)

91.01(16)

93.66(16)

N11-Cu1-O1′

O1-Cu1-O1′

N11-Cu1-O32′

O1-Cu1-O31

O1′-Cu1-O32′

93.3(2)

100.1(2)

92.96(18)

72.97(19)

[Cu(L1′′)Cl₂] (7)

Cu1-N11

1.971(4)

2.234(2)

Cu1-N21

1.999(5)

1.967(4)

2.228(2)

101.35(16)

103.67(6)

97.82(16)

102.77(6)

Cu1-Cl11

2.2214(14)

2.023(5)

90.2(2)

110.45(15)

90.3(2)

Cu1-Cl12

Cu2-N31

Cu2-N41

Cu2-Cl22

2.2157(14)

127.81(13)

125.30(14)

131.38(15)

128.16(14)

Cu2-Cl21

N11-Cu1-N21

N21-Cu1-Cl11

N31-Cu2-N41

N41-Cu2-Cl21

N11-Cu1-Cl11

N21-Cu1-Cl12

N31-Cu2-Cl21

N41-Cu2-Cl22

N11-Cu1-Cl12

Cl11-Cu1-Cl12

N31-Cu2-Cl22

Cl21-Cu2-Cl22

109.41(13)

^aEstimated standard deviations indicated in parentheses. ^bStructures are presented in the Supporting Information. ^cSymmetry code: -x + 1, y, -z + ¹/₂

+ 1. ^dSymmetry code: -x + 1, -y, -z + 1.

plexes. This can be seen by comparison with [Cu{H₂C(3,5-

Me₂pz)₂}(PPh₃)](ClO₄), where the Cu‚‚‚O(ClO₄) distance

The molecular structure of 5ClO₄consists of a binuclear

[{Cu(L1′′)(CO)}₂(µ-ClO₄)]⁺unit, where two {Cu(L1′′)-

(CO)}⁺subunits are bridged by ClO₄^-(Figure 3, top). This

way, the formation of the three-coordinate complex [Cu-

(L1′′)(CO)]⁺is avoided. This is possible as a result of the

small size of CO, which allows a direct dimerization of the

subunits, and the presence of a coordinating counterion,

which functions as a bridge. Nevertheless, it is surprising to

find a dimeric species in this case, because coordination of

-

is only 2.574 Å. The Cu-P(PPh₃) distances of 2.171(2) and

2.177(2) Å in 4ClO₄, 2.174(2) Å in 4PF₆, and 2.173(1) Å

in [Cu{H₂C(3,5-Me₂pz)₂}(PPh₃)](ClO₄)⁶are extremely simi-

lar. Selected bond distances and angles of these compounds

are summarized in Table 4, which also provides a comparison

with other PPh₃complexes with related nitrogen-containing

coligands.²⁸

10616 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

Table 4. Selected Structural Parameters for Copper(I) Triphenylphosphine Complexes with Bidentate or Tridentate Coligands

Cu-N (av)

N-Cu-N

(av) (deg)

Cu-P

(Å)

δ (³¹P)

(ppm)

(Å)

ref

neutral Nn (n ) 2 or 3) coligands

[Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄)

[Cu(L1′′)(PPh₃)](PF₆) (4PF₆)

[Cu{H₂C(3,5-Me₂pz)₂}(PPh₃)](ClO₄)

[Cu(L1′)(PPh₃)](ClO₄) (8ClO₄)

2.008(6)

2.006(4)

2.019(6)

2.085(10)

93.4(2)

93.6(1)

95.0(2)

89.0(4)

2.174(2)^a

2.1726(12)

2.171(2)

2.159(4)

12.11

this work

6

12.20

c

13.79^b

this work

anionic N3 coligand

[Cu{HB(3,5-Me2pz)₃}(PPh₃)]

2.098(14)

90.6(5)

2.166(6)

11.87^b

28

^aAverage distance. ^bBroad. ^cNot reported.

Table 5. Selected Structural Parameters and Properties of Copper(I)-Carbonyl Complexes with Bidentate and Tridentate Coligands

Cu-N

Cu-C

(Å)

C-O

(Å)

Cu-C-O

(deg)

ν(C-O)

δ (¹³C)

(ppm)

(av) (Å)^a

(cm^-1

)

ref

neutral Nn (n ) 2 or 3) coligands

[{Cu(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄)

[Cu{H₂C(3,5-Me₂pz)₂}(CO)(OClO₃)]

[Cu(NH(py)₂)(CO)(OClO₃)]

[Cu(en)(CO)(PhBPh₃)]

2.010(7)

2.004(3)

1.991(1)

2.041(4)

2.080(6)

1.809(10)^a

1.806(5)

1.808(2)

1.806(6)

1.777(5)

1.778(10)

1.110(13)^a

178.5(10)^a

2104

2108

2110

2117

2107

2100

172.9

this work

1.124(6)

176.8(4)

173.9

6

29

32

9

b

1.110(7)

1.127(6)

1.133(9)

178.3(5)

176.7(5)

176.8(9)

[Cu(L1′)(CO)](ClO₄)

[Cu{HC(3-tBupz)₃}(CO)](PF₆)

173.7

b

34

anionic Nn (n ) 2 or 3) coligands

[Cu{N[(C₃F₇)C(Dipp)N]₂}(CO)]

[Cu{N[(C₃F₇)C(Ftfmp)N]₂}(CO)]

[Cu{N[(C₃F₇)C(C₆F₅)N]₂}(CO)(NCMe)]

[{Cu(CO)}₂(TC-5,5)]

1.9401(15)

1.957(3)

2.0366(12)

1.934(6)

1.782(2)

1.138(2)

1.115(5)^a

1.124(2)

1.124(10)^a

178.3(2)

178.4(4)

176.68(12)

177.1(8)^a

2109

2128

2108

2071

172.9

30a

30b

31

b

1.816(5)^a

1.8333(17)

1.760(9)^a

b

169.9, 174.1

[Cu(L1)(CO)]

2.018(6)

2.046(4)

1.769(8)

1.118(10)

1.120(13)

178.6(9)

178.3(5)

2056

2083

9,11,35

36

178.7, 183.0

b

[Cu{HB(pz)₃}(CO)]

1.765(11)

^aAveraged value. ^bNot reported.

perchlorate and CO at the same time indicates that the simple

four-coordinate monomer [Cu(L1′′)(CO)(OClO₃)] should also

be able to form. In fact, corresponding four-coordinate

complexes have been isolated for other ligands (see below).

Somehow, the bridged dimer is energetically favored under

the synthetic conditions of 5ClO₄, but the exact reason for

(en)(CO)(PhBPh₃)]³²(the structure of this latter complex

indicates a weak interaction between copper(I) and the

counterion, where the metal is displaced by 0.26 Å from the

N₂C plane). In contrast, the Cu-C(CO) distances of the

three-coordinate complexes with anionic coligands tend to

be shorter: 1.782(2) Å in [Cu{N[(C₃F₇)C(Dipp)N]₂}(CO)],^30a

1.813(5) and 1.818(5) Å in [Cu{N[(C₃F₇)C(Ftfmp)N]₂}-

(CO)],^30band 1.771(7) and 1.749(9) Å in [{Cu(CO)}₂(TC-

5,5′)].³¹Thus, real three-coordinate carbonyl complexes are

very rare.³³Selected bond distances and angles of copper-

(I)-carbonyl complexes are summarized in Table 5, includ-

ing species with neutral and anionic tridentate nitrogen-

containing ligands.^9,11b,34-36Interestingly, the Cu^I-C(CO)

distances of the complexes with tridentate ligands are slightly

shorter, indicating a somewhat stronger π-backbond in these

cases.

-

this dimerization is not known. Cu(I)-O(ClO₄) distances

are 2.252 (Cu1) and 2.311 Å (Cu2). By using other bidentate

nitrogen-containing ligands, the copper(I) carbonyl com-

plexes [Cu{H₂C(3,5-Me₂pz)₂}(CO)(OClO₃)],⁶[Cu(NH(py)₂)-

(CO)(OClO₃)] (py ) 2-pyridyl),²⁹[Cu{N[(C₃F₇)C(Dipp)N]₂}-

(CO)] (Dipp ) 2,6-diisopropylphenyl),^30a[Cu{N[(C₃F₇)C-

(Ftfmp)N]₂}(CO)] (Ftfmp ) 2-fluoro-6-trifluoromethyl-

phenyl),^30b[Cu{N[(C₃F₇)C(C₆F₅)N]₂}(CO)(NCMe)],^30b[{Cu-

(CO)}₂(TC-5,5′)] (TC ) tropocoronand),³¹and [Cu(en)-

(CO)(PhBPh₃)]³²were structurally characterized. Importantly,

three-coordinate complexes are only obtained with anionic

coligands. All of these complexes show linear Cu-C-O

units (cf. Table 5), which is the normal case. The Cu-C(CO)

distances in the compounds with neutral coligands are

similar: 1.816(10) Å (Cu1) and 1.802(10) Å (Cu2) in 5ClO₄,

1.806(5) Å in [Cu{H₂C(3,5-Me₂pz)₂}(CO)(OClO₃)],⁶1.808(2)

Å in [Cu(NH(py)₂)(CO)(OClO₃)],²⁹and 1.806(6) Å in [Cu-

In the copper(II) complex [{Cu(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂]

(6), two [Cu(L1′′)]²⁺units are bridged by two OH^-and two

ClO₄^-anions to form a binuclear structure (Figure 3, bottom

left). The copper(II) ions are symmetry-equivalent and exhibit

a square pyramidal coordination (τ ) 0.02),³⁷with two

(33) To our knowledge, no three-coordinate Cu(I)-carbonyl complex with

a neutral bidentate coligand has been characterized by X-ray crystal-

lography to this date.

(28) Kitajima, N.; Koda, T.; Hashimoto, S.; Kitagawa, T.; Moro-oka, Y.

J. Am. Chem. Soc. 1991, 113, 5664-5671.

(29) Thompson, J. S.; Whitney, J. F. Inorg. Chem. 1984, 23, 2813-2819.

(30) (a) Dias, H. V. R.; Singh, S. Inorg. Chem. 2004, 43, 5786-5788. (b)

Dias, H. V. R.; Singh, S.; Flores, J. A. Inorg. Chem. 2006, 45, 8859-

8861.

(31) Villacorta, G. M.; Lippard, S. J. Inorg. Chem. 1987, 26, 3672-3676.

(32) Pasquali, M.; Floriani, C.; Gaetani-Manfredotti, A. Inorg. Chem. 1980,

19, 1191-1197.

(34) Reger, D. L.; Collins, J. E.; Rheingold, A. L.; Liable-Sands, L. M.

Organometallics 1996, 15, 2029-2032.

(35) Imai, S.; Fujisawa, K.; Kobayashi, T.; Shirasawa, N.; Fujii, H.;

Yoshimura, T.; Kitajima, N.; Moro-oka, Y. Inorg. Chem. 1998, 37,

3066-3070.

(36) (a) Bruce, M. I.; Ostazewski, A. P. P. J. Chem. Soc., Dalton Trans.

1973, 2433-2435. (b) Churchill, M. R.; DeBoer, B. G.; Rotella, F.

J.; Salah, O. M. A.; Bruce, M. I. Inorg. Chem. 1975, 14, 2051-2056.

Inorganic Chemistry, Vol. 46, No. 25, 2007 10617

Fujisawa et al.

nitrogen atoms of L1′′, two oxygen atoms of OH^-, and one

-

oxygen atom of ClO₄bound. The interaction between

copper and the second bridging perchlorate is very weak

-

(Cu-O(ClO₄) distance: 3.137(5) Å), leading to the square

pyramidal geometry. The Cu1-O1-Cu1′ angle is 100.1-

(2)° and the Cu1‚‚‚Cu1′ distance is 2.9466(10) Å. The

molecular structure of 6 is different from [{Cu(H₂Cpz₂)}₂-

(µ-OH)₂(µ-ClO₄)](ClO₄) reported before.⁵Whereas this

complex has a butterfly {Cu₂(OH)₂}²⁺core with a dihedral

angle of 153.1°, 6 possesses a planar {Cu₂(OH)₂}²⁺structure

(dihedral angle: 180°). These differences relate to the steric

hindrance of the iPr substituents of the pyrazolyl rings: if

the {Cu₂(OH)₂}²⁺core in 6 would have a butterfly structure,

the iPr groups would get too close to each other. Using a

neutral tridentate ligand like in [{Cu(L1′)}₂(µ-OH)₂](CF₃-

SO₃)₂,³⁸no coordination of the counterion is observed due

to the additional nitrogen donor present. Here, a planar {Cu₂-

(OH)₂}²⁺core is again observed.

In [Cu(L1′′)Cl₂] (7), the copper(II) ion is four-coordinate

with L1′′ and two chloride ions bound (Figure 3, bottom

right). The average Cu-Cl bond distance of 2.225(2) Å in

7 is slightly shorter than that of 2.258 Å in the five-coordinate

complex [Cu(L1′)Cl₂] with a neutral tridentate ligand.³⁹

Figure 4. ³¹P NMR spectra of copper(I) triphenylphosphine complexes

in the PPh₃resonance region: [Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄), [Cu(L1′′)-

(PPh₃)](PF₆) (4PF₆), [Cu(L1′)(PPh₃)](ClO₄) (8ClO₄), and [Cu(L1)(PPh₃)]

(9). ³¹P NMR shifts are referenced relative to an external standard of PPh₃

(0 ppm, top).

In the copper(I) triphenylphosphine complex [Cu(L1′)-

(PPh₃)](ClO₄) (8ClO₄), the copper(I) ion is distorted tetra-

hedrally coordinated by three nitrogen atoms of L1′ and one

phosphorus atom of PPh₃(Figure S3 and Table S2 in the

Supporting Information). Interestingly, the Cu-P distance

of 2.159(4) Å in 8ClO₄is shorter than that of 4ClO₄

(2.174(2) Å) and 4PF₆(2.1726(12) Å), and also the value

of 2.166(6) Å reported for [Cu{HB(3,5-Me₂pz)₃}(PPh₃)],²⁸

which contains an anionic hydrotris(pyrazolyl)borate ligand

(Table 4). This is further discussed in the next section. In

the case of [Cu(L1)(PPh₃)] (9), crystals suitable for high-

quality X-ray analysis could not be obtained.

the neutral tridentate ligand L1′ appears at 13.79 ppm. These

differences in the ³¹P NMR properties of 4ClO₄(or 4PF₆)

and 8ClO₄are related to the difference in coordination

number of these complexes. The somewhat larger chemical

shift for 8ClO₄indicates that PPh₃is a stronger donor to

the metal in this complex, which is surprising on the basis

of the larger coordination number of copper in this case. This

finding is in agreement with the DFT calculations, which

show an increase in the atomic charge on phosphorus in [Cu-

(L1′)(PPh₃)]⁺compared to [Cu(L1′′)(PPh₃)]⁺(cf. Table 6).

³¹P NMR spectra of 9 with the anionic tridentate ligand L1^-

show a broad resonance at 10.73 ppm. The broad nature of

this signal is likely due to an instability of this complex in

solution. In contrast, the relatively sharp signals of 4ClO₄

and 4PF₆indicate more stable Cu-P bonds with ligand L1′′.

Correspondingly, 9 has the smallest calculated formation

energy of all of the PPh₃complexes investigated here (cf.

Table 6). A similar behavior is also observed for [Cu{HB-

(3,5-Me₂pz)₃}(PPh₃)].²⁸

In general, ¹³C chemical shifts for terminal CO ligands in

classical metal carbonyls appear in the range of 184-223

ppm.^9,40In comparison, terminal CO ligands in nonclassical

metal carbonyls show smaller chemical shifts in the range

of 140-189 ppm.^9,40In 5ClO₄, the signal of the carbonyl

carbon is observed at 172.9 ppm. ¹³C chemical shifts of other

copper(I) carbonyl complexes ligated by either bidentate or

tridentate nitrogen-containing ligands are listed in Table 5,

which also lists the C-O stretching frequencies of these

complexes. The ¹³C chemical shift obtained for 5ClO₄is

B. Properties. NMR Spectroscopy. H and ¹³C NMR

1

spectra were measured for the ligand L1′′ and for all the

obtained copper(I) complexes. In all cases, the appropriate

signals of the protons and carbon atoms of the bidentate or

tridentate nitrogen-containing ligands are observed (see

Experimental Section).

The ³¹P NMR spectra of the copper(I) triphenylphosphine

complexes 4ClO₄, 4PF₆, 8ClO₄, and 9 show a single

resonance for PPh₃(cf. Figure 4). ³¹P chemical shifts of PPh₃

in these complexes appear downfield compared to free

triphenylphosphine, indicating a decrease in shielding due

to the presence of the metal, which is the expected case for

a σ-donor ligand like phosphine. As mentioned before, the

complexes 4ClO₄and 4PF₆are trigonal planar with similar

structures, which is consistent with their almost identical ³¹

P

chemical shifts of 12.11 ppm (4ClO₄) and 12.20 ppm (4PF₆).

In comparison, the ³¹P chemical shift of PPh₃in 8ClO₄with

(37) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor, G.

C. J. Chem. Soc., Dalton Trans. 1984, 1349-1356.

(38) Fujisawa, K. manuscript in preparation.

(39) Fujisawa, K.; Ono, T.; Aoki, H.; Ishikawa, Y.; Miyashita, Y.; Okamoto,

K.; Nakazawa, H.; Higashimura, H. Inorg. Chem. Commun. 2004, 7,

330-332.

(40) (a) Weber, L. Angew. Chem., Int. Ed. Engl. 1994, 33, 1077-1078.

(b) Abuke, F.; Wang, C. Coord. Chem. ReV. 1994, 137, 483-524. (c)

Willner, H.; Abuke, F. Angew. Chem., Int. Ed. Engl. 1997, 36, 2402-

2425. (d) Strauss, S. H. J. Chem. Soc., Dalton Trans. 2000, 1-6. (e)

Lupinetti, A. J.; Strauss, S. H.; Frenking, G. Prog. Inorg. Chem. 2001,

49, 1-112.

10618 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

similar to that of the mononuclear copper(I) carbonyl

complex [Cu{H₂C(3,5-Me₂pz)₂}(CO)(OClO₃)] (173.9 ppm).⁶

The reason for this similarity is that, in these two complexes,

copper(I) is coordinated by two nitrogen donors, one carbon

H₂C(3,5-Me₂pz)₂: Me) and the different types of nitrogen-

donor groups in these ligands only play a minor role for the

properties of the copper(I) centers. Furthermore, the value

of ν(C-O) in 5ClO₄is unexpectedly close to those of the

related copper(I) carbonyl complexes [Cu(L1′)(CO)](ClO₄)⁹

at 2107 cm^-1and [Cu{HC(3-tBupz)₃}(CO)](PF₆)³⁴at 2100

cm^-1. This means that the electron density at the copper(I)

center in 5ClO₄, with two nitrogen and one oxygen donors

(from perchlorate) coordinated, is comparable to that of [Cu-

(L1′)(CO)](ClO₄) with three nitrogen donors instead. These

results are consistent with the obtained ¹³C chemical shifts

of these compounds described above, which are also similar.

The Cu-CO stretching frequency in 5ClO₄is observed at

415 cm^-1, as identified by ¹³CO labeling, in the case of which

this band shifts to 399 cm^-1.

-

atom of CO, and one oxygen atom of ClO₄, in a distorted

tetrahedral geometry. However, the ¹³C chemical shifts of

these compounds are also unexpectedly close to that of the

related complex [Cu(L1′)(CO)](ClO₄) (173.7 ppm), which

contains a tridentate ligand. Therefore, ¹³C chemical shifts

simply do not seem to be very sensitive to the structures of

the complexes and the nature of the coligands applied (see

also ref 9).

Vibrational Spectroscopy. IR and far-IR spectra of all

of the complexes prepared here were measured. In the case

of the acetonitrile complex 3PF₆, the C-N stretching mode

of the coordinated acetonitrile ligand could not be observed.

This seems surprising, but a similar observation was made

in the case of [Cu(L1′)(NCMe)](PF₆). In comparison,

ν(C-N) is observed at 2255 cm^-1for [Cu(L1)(NCMe)] with

the anionic N3 ligand L1^-.⁹The reason for this difference

is not clear.

The far-IR spectra of the triphenylphosphine complexes

4ClO₄and 4PF₆show three strong bands near 500 cm^-1

(Figure S4 in the Supporting Information), which are

assigned to Whiffen’s y vibrations (out-of-plane bending of

the phenyl rings).⁴¹A second group of three medium-intense

bands appears near 450-400 cm^-1, which most likely

correspond to Whiffen’s t vibrations. The so-called u and x

vibrations of PPh₃appear as weak bands around 250-270

cm^-1. Unfortunately, the ν(Cu-P) vibration could not be

identified from the spectra. The reason for this finding is

discussed in the DFT section below. The far-IR spectra of

8ClO₄and 9 show spectral features similar to PPh₃, as

discussed above (cf. Figure S5 in the Supporting Informa-

tion).

Copper(I) carbonyl complexes are stabilized by back-

donation of copper(I) d electrons into the antibonding π*

orbitals of CO. Correspondingly, the C-O stretching vibra-

tion ν(C-O) is a sensitive indicator of the extent of back-

donation, or vice versa, the electron density available at the

metal center.^9,42In general, copper(I) carbonyl complexes

show intense IR bands around 2100 cm^-1assignable to

ν(C-O). The carbonyl complex 5ClO₄exhibits the C-O

stretch at 2104 cm^-1. Upon ¹³C isotope labeling, this band

shifts to 2050 cm^-1(Figure S6 in the Supporting Informa-

tion), which confirms this assignment. A comparison of C-O

stretching frequencies in 5ClO₄and other carbonyl com-

plexes with neutral bidentate nitrogen-containing ligands,

such as [Cu{H₂C(3,5-Me₂pz)₂}(CO)(OClO₃)],⁶[Cu{NH-

(py)₂}(CO)(OClO₃)],²⁹and [Cu(en)(CO)(PhBPh₃)],³²is given

in Table 5. Surprisingly, in these three cases, ν(C-O) is also

observed at 2108-2117 cm^-1. This result indicates that the

nature of the pyrazolyl alkyl substituents (L1′′: iPr and

The far-IR and UV-vis absorption spectra of the

copper(II) chloro complex 7 are shown in Figures S7 and

S8 in the Supporting Information, respectively. The chloride-

to-copper(II) CT transition of 7 is assigned to the absorption

band at 375 nm, whereas the feature at 911 nm corresponds

to a d-d transition. The UV-vis spectrum of [{Cu(L1′′)}₂-

(µ-OH)₂(µ-ClO₄)₂] (6) shows two bands at 355 and 563 nm

(Figure S9 in the Supporting Information), which are

assigned to a hydroxide-to-copper(II) CT transition and a

d-d transition, respectively, on the basis of a comparison

with [{Cu(L1)}₂(µ-OH)₂].⁴³Interestingly, whereas 6 and

[{Cu(L1)}₂(µ-OH)₂]¹⁰have planar {Cu₂(OH)₂}²⁺cores and

diamagnetic ground states (S ) 0, from ESR), [{Cu(H₂-

Cpz₂)}₂(µ-OH)₂(µ-ClO₄)](ClO₄) with a butterfly {Cu₂-

(OH)₂}²⁺core has a ferromagnetic ground state (S ) 1, from

ESR).⁵

C. Reactivity. In this work, the reactivity of 2ClO₄and

3PF₆toward PPh₃, CO, and O₂is the main focus of the

investigation, because these complexes show unusual geo-

metric structures, binuclear linear and trigonal planar,

respectively. The reactivity of each complex is described in

Scheme 1.

First, the reactivity of 2ClO₄and 3PF₆with PPh₃is

discussed. The reaction between 2ClO₄and 2 equiv of PPh₃

results in the rapid formation of the triphenylphosphine

complex 4ClO₄. Similarly, 4PF₆is generated by the reaction

of 3PF₆with 1 equiv of PPh₃. Because phosphines are strong

ligands to copper(I) and PPh₃is also quite bulky, three-

coordinate copper(I) complexes are easily obtained with this

ligand. Interestingly, as shown in Figure S10 in the Sup-

porting Information, the additive IR spectrum of the starting

complex 2ClO₄and PPh₃is similar to the IR data of the

generated triphenylphosphine complex 4ClO₄. This behavior

is also observed for 4PF₆(Figure S11 in the Supporting

Information). Importantly, this means that from the IR

spectra, it is not clear whether PPh₃coordinates or not. This

problem is solved by ³¹P NMR spectroscopy (cf. Figure 3),

which shows distinct ³¹P chemical shifts for 4ClO₄and 4PF₆

(41) (a) Shobatake, K.; Postmus, C.; Ferraro, J. R.; Nakamoto, K. Appl.

Spectrosc. 1969, 23, 12-16. (b) Pettinari, C.; Marchetti, F.; Polimante,

R.; Cingolani, A.; Portalone, G.; Colapietro, M. Inorg. Chim. Acta

1996, 249, 215-229.

(42) Schneider, J. L.; Carrier, S. M.; Ruggiero, C. E.; Young, V. G., Jr.;

Tolman, W. B. J. Am. Chem. Soc. 1998, 120, 11408-11418.

(43) Fujisawa, K.; Kobayashi, T.; Fujita, K.; Kitajima, N.; Moro-oka, Y.;

Miyashita, Y.; Yamada, Y.; Okamoto, K. Bull. Chem. Soc. Jpn. 2000,

73, 1797-1804.

Inorganic Chemistry, Vol. 46, No. 25, 2007 10619

Fujisawa et al.

Table 6. Calculated Properties and Formation Energies of Cu(I)-Phosphine Complexes

bond distances (Å)

charge^b

P(PPh₃)

complex formation energy^c

complex

Cu-N (av.)

Cu-P

ν(Cu-P)/ PPh₃(Inv.)^a

LanL2DZ

TZVP

[Cu(L1′′)(PPh₃)]⁺

[Cu(L1′)(PPh₃)]⁺

[Cu(L1)(PPh₃)]

2.052

2.148

2.122

2.323

2.329

2.343

537

534

531

+0.56

+0.57

+0.60

-35.3

-29.8

-19.2

-38.3

-32.8

-21.5

^aIn cm^-1

.

^bCalculated Mulliken charge on phosphorus. ^cCalculated using the equation: [Cu(L)]^0/++ PPh₃f [Cu(L)(PPh₃)]^0/+. In kcal/mol.

-

compared to free PPh₃. From NMR, PPh₃certainly coordi-

nates to copper(I), consistent with the X-ray structural

analyses.

with PF₆. Additionally, acetonitrile is a strong ligand for

copper(I) and, therefore, competes with CO and is harder to

replace. To synthesize a mononuclear copper(I) carbonyl

complex, [Cu(L1′′)(CO)]X, 2BF₄was then utilized. After

reaction of this compound with CO, the solvent was removed

and the IR spectrum was measured showing ν(C-O) at 2104

cm^-1(Figure S16 in the Supporting Information). Recrys-

tallization was carried out under a CO atmosphere, but no

crystals exhibiting ν(C-O) in the IR spectrum could be

obtained. This again shows that the coordinated perchlorate

anion in 5ClO₄plays an important role in stabilizing the

CO complex. As mentioned above, the ¹³C chemical shift

of CO and ν(C-O) in 5ClO₄are both similar to those of

[Cu(L1′)(CO)](ClO₄) (cf. Table 5), but the latter complex

with the tridentate ligand is clearly more stable in solution.⁹

The reason for this difference is that the structural integrity

of 5ClO₄depends on the labile perchlorate bridge, which is

most likely easily lost in solution, leading to an equilibrium

with different mononuclear species. Our main hypothesis is

that the three-coordinate complex [Cu(L1′′)(CO)]⁺is formed

in this process, which is very labile toward the loss of CO.

This way, CO leaks out of the system, and only CO-free

compounds are finally obtained. In comparison, four-

coordinate [Cu(L1′)(CO)]⁺constitutes a stable unit in

solution.

Next, the reactivity of 2ClO₄and 3PF₆toward O₂is

discussed. 2ClO₄was dissolved in dichloromethane and

cooled to -78 °C under an argon atmosphere. The argon

gas was then replaced by O₂, and the solution was allowed

to warm up to room temperature. During this time, the color

of the solution turned gradually from colorless to deep green.

From the reaction mixture, 6 was obtained and characterized

by X-ray crystallography as well as by ESR and UV-vis

absorption spectroscopy. The reaction of 3PF₆with O₂has

also been performed, but surprisingly, only very small

amounts of oxidative products were obtained. This means

that three-coordinate 3PF₆is basically stable toward O₂.

In comparison, the cationic complex [Cu(L1′)(NCMe)]PF₆

with the neutral tridentate ligand L1′ reacts slowly with O₂

to give the µ-η²:η²peroxo complex [{Cu(L1′)}₂(µ-η²:η²-O₂)]-

(PF₆)₂(ca. ∼10% conversion),⁹whereas the neutral complex

[Cu(L1)(NCMe)] with the anionic tridentate ligand L1 reacts

readily with O₂to give the µ-η²:η²peroxo complex [{Cu-

(L1)}₂(µ-η²:η²-O₂)].^9,11On the basis of these findings, part

of the lack of reactivity of 3PF₆relates to the charge of the

complex, which shifts the redox potential. Two-coordinate

2ClO₄still reacts much slower with O₂than [Cu(L1′)-

(NCMe)]X, which is in agreement with the literature.^4a,4c

Unfortunately, no oxygenated intermediate could be obtained

in the reaction of the Cu(I)-L1′′ complexes with O₂. The

Second, the reaction of 2ClO₄with CO is evaluated. After

the reaction of 2ClO₄with CO, the solvent was removed

under vacuum, and an IR spectrum of the crude product was

measured. The IR data show ν(C-O) as a broad band at

2104 cm^-1(Figure S12, middle in the Supporting Informa-

tion). Recrystallization of this material led to colorless

crystals, which, however, did not exhibit ν(C-O) in the IR

spectrum anymore (Figure S12, bottom in the Supporting

Information). X-ray crystallography then showed that the

final product corresponds to 1ClO₄, where two bidentate

ligands L1′′ are bound to copper, but CO is not present. This

result indicates a surprising instability of the Cu(I)-CO bond

in these systems, whereas this bond is usually quite strong.

Correspondingly, if the solvent was not removed after the

reaction of 2ClO₄with CO, and the recrystallization was

performed under a CO atmosphere, colorless crystals were

obtained that show the intense ν(C-O) band at 2104 cm^-1.

X-ray crystallography then confirmed that the obtained

material was the binuclear copper(I) carbonyl complex

5ClO₄. From these results, 5ClO₄seems to be very unstable

toward the loss of CO in solution. In the same way, if 5ClO₄

was dissolved in acetonitrile and stirred under an argon

atmosphere, ν(C-O) disappeared in the IR spectrum (Figure

S13, bottom in the Supporting Information), which means

that CO was removed. This behavior is also observed for

[Cu(L1′)(CO)]⁺with a neutral tridentate ligand but not for

[Cu(L1)(CO)].⁹CO in 5ClO₄was also readily displaced by

PPh₃(Figure S14 in the Supporting Information). To further

investigate this, the reaction of 3PF₆with CO was performed.

After the reaction, the solvent was removed under vacuum,

and the IR spectrum of the resulting crude product was

measured, showing a very small band for ν(C-O) around

2090 cm^-1(Figure S15, middle in the Supporting Informa-

tion). Importantly, this frequency is somewhat different from

5ClO₄. Because PF₆^-is a non-coordinating, this band could

belong to the elusive, three-coordinate species [Cu(L1′)-

(CO)]⁺or the corresponding four-coordinate acetonitrile

complex. From the literature, the latter possibility is much

more likely (see ref 29). If recrystallization of this material

was performed under argon, CO was removed as a ligand,

as indicated in the IR spectrum in Figure S15, bottom in the

Supporting Information. From these results, it becomes clear

that the reactivity of 3PF₆toward CO is even lower than

that of 2ClO₄. The reason for this is again the apparent

instability of [Cu(L1′′)(CO)]⁺: whereas perchlorate in 2ClO₄

serves as an additional ligand to copper, preventing the

formation of a three-coordinate complex, this is not possible

10620 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

reason for this is not clear, especially because corresponding

side-on peroxo complexes could be obtained easily with L1′

as mentioned above, but also with simple bidentate amine

ligands.^44,45One possibility is that the dioxygen complex with

L1′′ is too unstable, such that the decomposition of this

species yielding simple copper(II) complexes is always faster

than its generation. In this case, the intermediate would never

appear at a significant concentration. We also tried to

generate an oxygen intermediate by the reaction of [{Cu-

(L1′′)}₂(µ-OH)₂(µ-ClO₄)₂] (6) with hydrogen peroxide at low

temperature. However, no intermediates were observed.

Recently, exactly this observation has been reported for [Cu₂-

(H3m)(CO)₂]²⁺(H3m ) dinucleating hexaaza macrocyclic

Figure 5. Overview of the properties of copper(I) complexes with the

ligands L1′′, L1′, and L1^-.

2⁺

derivates)⁴⁶and [Cu(baib)]₂(baib ) Schiff base deriva-

for PPh₃. The latter result is in agreement with the instability

of 9 in solution as observed by ³¹P NMR. This trend is

strikingly confirmed by the properties of the corresponding

carbonyl complexes, where (based on the observed C-O

stretching frequencies), the electron-rich complex with L1^-

shows a stronger Cu(I)-CO bond than the compound with

L1′.⁹This inverse trend in bond stabilities compared to PPh₃

is due to the fact that phosphines are strong donors, whereas

CO is an acceptor that prefers electron-rich metals. On the

basis of the results obtained here, L1′′ leads to the most

unstable CO complex, again in agreement with this trend.

Figure 5 summarizes these findings.

tives)⁴⁷upon reaction with O₂. Finally, we explored the

reaction of [Cu(L1′′)₂](CuCl₂) (1CuCl₂) with O₂. Recrys-

tallization gave orange crystals of [Cu(L1′′)Cl₂] (7).¹⁰

A

similar behavior was reported before for [Cu{Ph₂C(pz)₂}₂]-

(CuCl₂).⁷

The reaction of 2ClO₄with acetonitrile surprisingly did

not yield the three-coordinate complex [Cu(L1′′)(NCMe)]-

(ClO₄), but 1ClO₄. In contrast, this reaction proceeds readily

with the analogous complex [Cu(L1′)(OClO₃)]. As described

above, this is due to a ligand disproportionation reaction in

the case of L1′′ to avoid the formation of a three-coordinate

complex (cf. Section A).

Another interesting problem is the lack of a distinct Cu-P

stretching mode in the vibrational spectra of the PPh₃

complexes, which would be expected to occur as a new,

additional band compared to the additive spectra of the

precursor [Cu(L)]^0/+and PPh₃. However, no such band is

observed experimentally. Surprisingly, this finding is actually

confirmed by the DFT calculations on the three PPh₃

complexes investigated here. Triphenylphosphine shows three

bands in the 500 cm^-1region, which belong to C-P-C

bending vibrations. In ideal C₃symmetry, these would

actually lead to an A₁and an E component. The A₁vibration

is special, because it corresponds to the PR₃inversion mode,

where the phosphorus atom moves perpendicular to the plane

formed by the three attached carbons. Hence, this mode

actually corresponds to a Cu-P stretch in the resulting

Cu(I)-PPh₃complex, where the phosphorus atom then

moves along the Cu-P vector. Importantly, this A₁vibration

occurs at higher energy than the E components (in the

calculations), which indicates that the bands at 529/530 cm^-1

in 4ClO₄/4PF₆(Calcd: 537 cm^-1), at 531 cm^-1in 8ClO₄

(Calcd: 534 cm^-1), and at 528 cm^-1in 9 (Calcd: 531 cm^-1)

have to be assigned to this mode. Other PPh₃vibrations

(especially the A₁-symmetric P-C stretch) also contain a

Cu-P stretching component. Importantly, this also implies

that the mode around 530 cm^-1is not very diagnostic for

the strength of the Cu-P bond.

D. DFT Calculations on Cu(I)-PPh₃Complexes. To

gain further insight into the properties of the PPh₃complexes

and the vibrational assignments, DFT calculations on the

three complexes [Cu(L1′′)(PPh₃)]⁺, [Cu(L1′)(PPh₃)]⁺, and

[Cu(L1)(PPh₃)] were applied without any structural simpli-

fications. The obtained optimized structures show very good

overall agreement with experiment. Table 6 shows that the

average Cu-N distances are also reproduced reasonably well.

In contrast, the Cu-P distances are obtained too long

(consistently about 0.15 Å longer compared to experiment),

which is due to the lack of polarization functions in the

LanL2DZ basis set, as observed before.⁴⁸As shown in Table

6, the calculated formation energies for these three PPh₃

complexes show an interesting trend where,

[Cu(L1′′)(PPh₃)]⁺(∆E ) -38 kcal/mol) >

[Cu(L1′)(PPh₃)]⁺(-33) . [Cu(L1)(PPh₃)] (-22)

indicating that the complex with bidentate L1′′ has the most

electron-poor copper(I), which therefore binds to the phos-

phine donor the strongest, whereas anionic L1^-leads to the

most electron-rich copper(I), which has the lowest affinity

(44) (a) Stack, T. D. P. Dalton Trans. 2003, 1881-1889. (b) Mirica, L.

M. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004, 104, 1013-

1045.

(45) Cole, A. P.; Mahadevan, V.; Mirica, L. M. M.; Ottenwaelder, X.; Stack,

T. D. P. Inorg. Chem. 2005, 44, 7345-7364.

Conclusions

(46) Coast, M.; Ribas, X.; Poater, A.; Valbuena, J. M. L.; Xifra, R.;

Company, A.; Duran, M.; Sola`, M.; Llobet, A.; Corbella, M.; Uso´n,

M. A.; Mah´ıa, J.; Solans, X.; Shan, X.; Benet-Buchholz, J. Inorg.

Chem. 2006, 45, 3569-3581.

(47) Mandal, S.; Mukherjee, R. Inorg. Chim. Acta 2006, 359, 4019-4026.

(48) (a) Lehnert, N.; Tuczek, F. Inorg. Chem. 1999, 38, 1671-1682. (b)

Tuczek, F. AdV. Inorg. Chem. 2004, 56, 27-53.

In this work, eight copper(I) complexes and two

copper(II) complexes with the neutral bidentate nitrogen-

containing coligand L1′′ were systematically synthesized and

characterized, and their reactivity was evaluated in compari-

son to analogous complexes with the related neutral tridentate

Inorganic Chemistry, Vol. 46, No. 25, 2007 10621

Fujisawa et al.

ligand L1′ published before.⁹The second focus of this article

is the synthesis of mononuclear three-coordinate copper(I)

complexes of the type [Cu(L1′′)(X)](Z), where X ) MeCN,

CO, and PPh₃; Z ) counterion.

Importantly, our results demonstrate how the structures

and reactivities of copper(I) complexes are affected by

changing the coordination number of the coligand from

neutral tridentate (L1′) to neutral bidentate (L1′′). Whereas

the treatment of L1′ with CuCl results in the formation of

[Cu(L1′)(Cl)],⁹the same reaction with L1′′ yields the

complex [Cu(L1′′)₂](CuCl₂) (1CuCl₂), which corresponds to

a disproportionation of type,

teristically different from those obtained for [Cu(L1′)(PPh₃)]-

(ClO₄) (8ClO₄) and [Cu(L1)(PPh₃)] (9), which both contain

N3 tridentate ligands. Other three-coordinate structures such

as [Cu(PPh₃)₃]^{+ 49}and [Cu(SPh)₃]^{2- 50}are observed in the

case of phosphine or hindered anionic ligands. In addition,

three-coordinate structures were recently reported using

hindered tert-butyl isocyanide as a ligand⁵¹and hindered

52

anionic â-diketoiminate {Ph[CCH₂(Dipp)N]₂}^-and flu-

orinated bis(pyrazolyl)borate {H₂B(3,5-(CF₃)₂pz)}^{- 53}ligands.

These examples are also consistent with our finding that

three-coordinate copper(I) complexes are easily obtained with

bidentate anionic ligands or phosphines, whereas neutral

bidentate nitrogen-donor ligands require steric hindrance, and

the obtained complexes are still unstable.

2 CuCl + 2 L1′′ f 2 [Cu(L1′′)(Cl)]^#f

-

[Cu(L1′′)₂]⁺+ CuCl₂

As indicated above, the reaction of 2ClO₄with CO leads

to the generation of the binuclear carbonyl complex [{Cu-

(L1′′)(CO)}₂(µ-ClO₄)](ClO₄) (5ClO₄), and no three-coordi-

nate carbonyl complex of type [Cu(L1′′)(CO)]⁺could have

been obtained.³³On the other hand, [Cu(L1′)(CO)](ClO₄)⁹

is readily generated by using the tridentate ligand L1′. These

experimental results show that the stability of 5ClO₄is much

lower than that of [Cu(L1′)(CO)](ClO₄), although the

electronic structure of the Cu(I)-CO bond is very similar

in these compounds, as indicated by the C-O stretching

frequencies and the ¹³C chemical shifts. The reason for the

low stability of 5ClO₄in solution is that the integrity of this

complex depends on the presence of the weak perchlorate

bridge. Loss of this bridge, which should be easily possible

in solution, leads to the generation of free [Cu(L1′′)(CO)]⁺,

which we have designated as ligand disproportionation

reaction. This is our first example where a coordination

number of 3 is avoided by copper(I). Obviously, if a

reasonably strong ligating counterion Z is present, the

formation of a four-coordinate complex and a two-coordinate

-

anion of type CuZ₂is in fact energetically faVored oVer

the three-coordinate species. A similar process is observed

for the reaction of [Cu(L1′′)]₂(ClO₄)₂(2ClO₄) with aceto-

nitrile. Here, the formation of the three-coordinate species

[Cu(L1′′)(NCMe)]⁺is avoided:

[Cu(L1′′)]₂²⁺+ MeCN (excess) f 2[Cu(L1′′)(NCMe)]^{+ #}

f [Cu(L1′′)₂]⁺+ [Cu(NCMe)_n(ClO₄)] (not isolated)

This complex can only be obtained with the non-

2+

which readily loses CO to form two-coordinate [Cu(L1′′)]₂

or other four-coordinate complexes.

-

coordinating counterion PF₆(3PF₆; vide infra). The syn-

The trends observed in the stabilities and spectroscopic

properties of the copper(I) complexes with L1′′, L1′, and

L1^-, and the donor and acceptor ligands PPh₃and CO are

nicely reproduced by the DFT calculations. Taken together,

these results show that the electron density at the copper

center increases in the order: L1′′ < L1′ < L1^-(cf. Figure

5). Hence, L1′′ should give rise to the strongest Cu(I)-

phosphine interaction, whereas L1^-leads to the strongest

Cu(I)-CO bond. The DFT calculations also explain the

absence of a discrete Cu-P stretching band in the IR spectra.

In fact, several internal PPh₃vibrations of idealized A₁

symmetry (especially the C-P-C bend, but also the P-C

stretch) obtain Cu-P stretching contributions in the corre-

sponding complexes, which means that the Cu-P stretching

coordinate is extremely delocalized over a number of PPh₃

modes.

thesis of the starting complex of this reaction, [Cu(L1′′)]₂-

(ClO₄)₂, is yet another example for this general reaction

scheme. The three-coordinate complex [Cu(L1′′)(OClO₃)]

presumably formed as an intermediate in the synthetic

procedure is unstable, and instead, the two-coordinate

complex [Cu(L1′′)]₂(ClO₄)₂is isolated from the reaction

mixtures. With CO as a ligand, a perchlorate-bridged dimer

is formed, where each copper(I) is four-coordinate. Cor-

respondingly, when a non-coordinating counterion is used,

no stable CO complex can be isolated. These results present

strong evidence that three-coordinate copper(I) with L1′′ as

a neutral coligand is energetically unstable, and hence, does

not willingly form in contrast to complexes that contain

monoanionic bidentate nitrogen-donor ligands such as

{N[(C₃F₇)C(Dipp)N]₂}^{- 30a}and {N[(C₃F₇)C(Ftfmp)N]₂}^-.^30b

Even these complexes easily form four-coordinate species

with coordinated acetonitrile.

(49) For example, Healy, P. C.; Hanna, J. V. Acta Crystallogr. 2003, E59,

m384-m386.

(50) Fujisawa, K.; Imai, S.; Suzuki, S.; Moro-oka, Y.; Miyashita, Y.;

Yamada, Y.; Okamoto, K. J. Inorg. Biochem. 2000, 82, 229-238.

(51) Zhou, L.; Powell, D.; Nicholas, K. M. Inorg. Chem. 2007, 46, 2316-

2321.

(52) (a) Spencer, D. J. E.; Aboelella, N. W.; Reynolds, A. M.; Holland, P.

L.; Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 2108-2109. (b)

Spencer, D. J. E.; Reynolds, A. M.; Holland, P. L.; Jazdzewski, B.

A.; Dubac-Toia, C.; Pape, L. L.; Yokota, S.; Tachi, Y.; Itoh, S.;

Tolman, W. B. Inorg. Chem. 2002, 41, 6307-6321.

(53) Dias, H. V. R.; Richey, S. A.; Diyabalange, H. V. K.; Thankamani, J.

J. Organomet. Chem. 2005, 690, 1913-1922.

As mentioned above, a three-coordinate copper(I) complex

could only be obtained with acetonitrile in the presence of

-

the non-coordinating counterion PF₆. However, this picture

completely changes when phosphines are used: the strong

and bulky ligand PPh₃forms three-coordinate copper(I)

complexes very easily. The resulting compounds can be

characterized best by ³¹P NMR spectroscopy. Complexes

[Cu(L1′′)(PPh₃)](ClO₄) (4ClO₄) and [Cu(L1′′)(PPh₃)](PF₆)

(4PF₆) with L1′′ show ³¹P chemical shifts that are charac-

10622 Inorganic Chemistry, Vol. 46, No. 25, 2007

Copper Complexes with a Neutral N2-Type Coligand

Finally, whereas [Cu(L1′)(NCMe)](ClO₄) and [Cu(L1′)-

(OClO₃)] with the tridentate nitrogen-containing ligand L1′

react readily with O₂to give oxygen intermediates such as

the µ-η²:η²peroxo complex,⁹no corresponding oxygen

intermediate could be obtained in the presence of the

bidentate ligand L1′′. Presumably, the stability of these

intermediates is too low with L1′′, which facilitates fast

decomposition even at low temperature. In addition, three-

coordinate 3PF₆is quite stable toward O₂at room temper-

ature, which is completely unexpected.

All of these surprising differences originate from the

change in the coordination number of the supporting ligand

from three to two. On the basis of all of these results

combined with the previous findings, it is now possible

to efficiently control the structures and reactivities of

copper(I) complexes by changing the coordination number,

the charge, and the steric demand of the supporting coligand.

We are currently pursuing studies to investigate the effect

of different pyrazolyl side chains in bis(pyrazolyl)methane

type ligands on the structures and reactivities of the resulting

copper(I) and copper(II) complexes.

Acknowledgment. This research was supported in part

by JSPS (Scientific Research (B), Grants 14350471 and

17350043) and by MEXT (Scientific Research on Priority

Areas, Grant 19020011) to K.F.

Supporting Information Available: Crystal structures of L1′′,

1ClO₄, 2BF₄, 4PF₆, and 9 (Figures S1-S3), selected structural

parameters (Tables S1-S2), IR, far-IR, and UV-vis spectra

(Figures S4-S16), and the Cartesian coordinates of fully optimized

[Cu(L1′′)(PPh₃)]⁺, [Cu(L1′)(PPh₃)]⁺, and [Cu(L1)(PPh₃)] (Tables

S3-S5). Crystallographic data is available in CIF format. This

material is available free of charge via the Internet at http://

pubs.acs.org. Crystallographic data have been deposited at the

CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K., and copies

can be obtained on request, free of charge, by quoting the

publication citation and the deposition numbers (620276-620287).

IC7011342

Inorganic Chemistry, Vol. 46, No. 25, 2007 10623

Article Doi

Mononuclear and binuclear copper(I) complexes ligated by bis(3,5-diisopropyl-1-pyrazolyl)methane: Insight into the fundamental coordination chemistry of three-coordinate copper(I) complexes with a neutral coligand

DOI: 10.1021/ic7011342

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ic7011342

Full text of DOI:10.1021/ic7011342

Products guided by the article

R&D Labs maybe for 1001191-33-8

Relevant to this article

Hot Product