Syntheses, Structures, and Reactivity of Polynuclear Pyrazolato Copper(I) Complexes, Including an ab-Initio XRPD Study of [Cu(dmnpz)]3 (Hdmnpz = 3,5-Dimethyl-4-nitropyrazole)

Article abstract of DOI:10.1021/ic970875n

The reaction of [Cu(CH₃CN)₄](BF₄) with 3,5-dimethyl-4-nitropyrazole (Hdmnpz) in the presence of triethylamine yields the new copper(I) complexes [Cu(dmnpz)]₃ (1) and (Et₃NH)₂[Cu₄

Full text of DOI:10.1021/ic970875n

4284
Inorg. Chem. 1998, 37, 4284-4292
Syntheses, Structures, and Reactivity of Polynuclear Pyrazolato Copper(I) Complexes,
Including an ab-Initio XRPD Study of [Cu(dmnpz)]3 (Hdmnpz )
3,5-Dimethyl-4-nitropyrazole)
G. Attilio Ardizzoia,*,1a Sergio Cenini,1a Girolamo La Monica,1a Norberto Masciocchi,1b
Angelo Maspero,1a and Massimo Moret*,1b
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, and Dipartimento di Chimica
Strutturale e Stereochimica Inorganica, Centro CNR, Universita` di Milano, Via Venezian 21,
I-20133 Milano, Italy
ReceiVed July 11, 1997
The reaction of [Cu(CH3CN)4](BF4) with 3,5-dimethyl-4-nitropyrazole (Hdmnpz) in the presence of triethylamine
yields the new copper(I) complexes [Cu(dmnpz)]3 (1) and (Et3NH)2[Cu4(dmnpz)6] (2), depending on the
experimental conditions. The reactivity of 1 and 2 toward neutral ligands such as triphenylphosphine, cyclohexyl
isocyanide (RNC), and carbon monoxide has been investigated. In particular, both complexes readily react with
RNC, giving the dinuclear complexes [Cu(dmnpz)(RNC)]2 (4) and [Cu(dmnpz)(RNC)2]2 (5), depending on the
copper/RNC ratio, and with PPh3, affording the dimeric derivative [Cu(dmnpz)(PPh3)]2 (6). The crystal and
molecular structure of 1 has been determined ab initio using X-ray powder diffraction data from conventional
laboratory equipment. Crystals of 1 are monoclinic, C2/c, a ) 20.057(2) Å, b ) 13.816(2) Å, c ) 7.883(1) Å,
â ) 95.912(4)°; RF and Rwp 0.067 and 0.039, respectively, for Rietveld refinement on 3900 data points collected
in the range 17 < 2θ < 95° (Cu KR radiation). Crystals of 1 contain planar trimers, with the copper atoms
bridged by exo-bidentate ligands and short intermolecular Cu‚‚‚Cu contacts (3.329(7) Å). For complexes 2 and
4-6, single-crystal X-ray diffraction studies have been performed. Crystals of 2 are monoclinic, P21/c, a )
11.026(1) Å, b ) 13.456(2) Å, c ) 18.668(5) Å, â ) 92.91(2)°, Z ) 2. The ionic packing of 2 contains tetranuclear
complexes, with the copper atoms connected by six dmnpz ligands. Crystals of 4 are triclinic, P1h, a ) 7.418(1)
Å, b ) 9.780(1) Å, c ) 11.177(3) Å, R ) 109.61(2); â ) 101.34(3)°, γ ) 105.82(1)°, Z ) 2. Crystals of 5 are
triclinic, P1h, a ) 9.506(4) Å, b ) 9.957(2) Å, c ) 11.658(4) Å, R ) 86.73(2)°, â ) 79.54(2)°, γ ) 82.47(4)°,
Z ) 1. Crystals of 6 are triclinic, P1h, a ) 11.379(2) Å, b ) 13.592(2) Å, c ) 14.409(3) Å, R ) 81.22(1)°, â )
85.21(1)°, γ ) 87.55(1)°, Z ) 2. 4, 5, and 6 are binuclear complexes bearing one RNC, two RNC, and one
triphenylphosphine ligands on each copper atom, respectively. The steric requirement of the dmnpz ligands, in
the presence of RNC or PPh3, forces the inner [Cu2(dmnpz)2] core of the three complexes into markedly different
conformations.
Introduction
copper(I) complexes containing exo-bidentate pyrazolate ligands,
focusing our interest on the ability of these derivatives, of
general formula [Cu(pz*)]n, to act as selective catalysts in the
oxidation of organic substrates such as aliphatic and aromatic
amines.5
The synthesis of di- and polynuclear complexes having
ligands which maintain the metal centers in close proximity is
an important objective in coordination metal chemistry. Interest
in such systems arise mainly because of their potential role in
multi-metal-centered catalysis in both biological and industrial
reactions.2 The major effort in this field is to find convenient
binucleating ligands able to attain this goal. Deprotonated
pyrazoles (pyrazolate anions) represents a significant class of
binucleating ligands because of their capability of coordinating
in an exo-bidentate fashion.3,4
With the aim of obtaining a deeper insight into the under-
standing of the steric and electronic factors that govern the
nuclearity and the chemical behavior of this class of copper(I)
derivatives, we began a systematic study using different
pyrazoles whose donor capabilities and steric accessibility were
modified by introducing substituents on the heterocyclic ring,
demonstrating the structural versatility caused by the nature of
the pyrazolate ligands.6
In the recent past, we have been particularly interested in
the synthesis, structural characterization and reactivity of
* Corresponding author. Tel. 39 2 266 80 675. Fax: +39 2 236 27 48.
E-mail: attilio@csmtbo.mi.cnr.it.
(4) A limited number of endo-bidentate pyrzolate complexes has been
structurally characterized: Eigenbrot, C. W., Jr.; Raymond, K. N.
Inorg. Chem. 1981, 20, 1553; 1982, 21, 2653. Ro¨ttger, D.; Erker, G.;
Grehl, M., Fro¨lich, R. Organometallics 1994, 13, 3897. Guzei, I. A.;
Baboul, G. A.; Yap, G. P. A.; Rheingold, A. L.; Schegel, H. B.; Winter,
C. H. J. Am. Chem. Soc. 1997, 119, 3387.
(5) (a) Ardizzoia, G. A.; Angaroni, M.; La Monica, G.; Cariati, F.; Cenini,
S.; Moret, M.; Masciocchi, N. Inorg. Chem. 1991, 30, 4347. (b)
Ardizzoia, G. A.; Cenini, S.; La Monica, G.; Masciocchi, N.; Moret,
M. Inorg. Chem. 1994, 33, 1458.
(1) (a) Dipartimento di Chimica Inorganica, Metallorganica e Analitica.
(b) Dipartimento di Chimica Strutturale e Stereochimica Inorganica.
(2) (a) Poilblanc, R. Inorg. Chim. Acta 1982, 62, 75. (b) Tejel, C.; Villoro,
J. M.; Ciriano, M. A.; Lo´pez, J. A.; Eguiza´bal, E.; Lahoz, F. J.;
Bakhmutov, V. I.; Oro, L. A. Organometallics 1996, 15, 2967.
(3) (a) Trofimenko, S. Chem. ReV. 1972, 72, 497. (b) Trofimenko, S. Prog.
Inorg. Chem. 1986, 34, 115. (c) La Monica, G.; Ardizzoia, G. A. Prog.
Inorg. Chem. 1997, 46, 151.
S0020-1669(97)00875-6 CCC: $15.00 © 1998 American Chemical Society
Published on Web 08/01/1998

Synthesis, Structure and Reactivity of Complexes
Inorganic Chemistry, Vol. 37, No. 17, 1998 4285
In this paper we continue with our study, reporting the
synthesis of two new homoleptic copper(I) pyrazolate com-
plexes, the trinuclear [Cu(dmnpz)]3 and the tetranuclear anionic
[Cu4(dmnpz)6]2- species, derived from reaction of [Cu(CH3-
CN)4](BF4) with 3,5-dimethyl-4-nitropyrazole (Hdmnpz), their
full structural characterization, and their reactivity with neutral
ligands. The full structural characterization of the dimeric
species obtained by reaction of these derivatives with cyclohexyl
isocyanide and triphenylphosphine is also reported.
erations, the formation of the neutral trinuclear species [Cu-
(dmnpz)]3 (1) is not unexpected.
The formation of the ionic species 2 is more surprising. The
synthetic method described above (eq 1) has been verified to
work well for a large series of pyrazoles.8 In all cases, the
formation of homoleptic, neutral [Cu(pz*)]n derivatives (Hpz*
) pyrazole, n ) ∞;9 3,5-dimethylpyrazole, n ) 3;7a 3,5-
diphenylpyrazole, n ) 47b) has been observed, independent of
the Cu/Hpz*/Et3N ratios employed. The formation of (Et3NH)2-
[Cu4(dmnpz)6] seems to be driven by the peculiar electronic
properties of the 3,5-dimethyl-4-nitropyrazole rather than by
steric effects. Notably, 3,5-dimethylpyrazole, possessing a steric
hindrance comparable to that of Hdmnpz but deeply differing
in the electronic properties,10 gives rise to only the neutral
trinuclear [Cu(dmpz)]3 derivative.7a,8
Behavior of 2 in Solution. Complex 2 is quite stable when
dissolved in deoxygenated acetone in the presence of free
triethylamine. In the absence of Et3N, fast dissociation of the
dmnpz anion takes place and quantitative formation of the white
trimeric [Cu(dmnpz)]3 derivative is observed. The reaction is
reversible, since by adding Et3N to the reaction mixture,
complex 1 easily dissolves restoring 2. The overall process can
be regarded as an equilibrium reaction between 1 and 2 (eq 3).
Results and Discussion
Synthesis of the Homoleptic Pyrazolato Complexes 1 and
2. The reaction of [Cu(CH3CN)4](BF4) and 3,5-dimethyl-4-
nitropyrazole in acetone in the presence of triethylamine leads
to the formation of two distinct products, depending on the
experimental conditions.
The highly insoluble, white trinuclear derivative [Cu(dmnpz)]3
(1) was obtained employing a 1:1 copper/dmnpz molar ratio
(eq 1).
3[Cu(CH3CN)4](BF4) + 3Hdmnpz + 3Et3N -CH CN8
3
+ 3(Et3NH)BF4 (1)
[Cu(dmnpz)]3
(1)
3(Et NH) [Cu (dmnpz) ] 4[Cu(dmnpz)]
3
h
+
3
2
4
6
It is useful to note that the copper/dmnpz ratio required in
eq 1 can be attained by carrying out the reaction either with a
stoichiometric amount of Hdmnpz (and excess of Et3N) or,
alternatively, with excess Hdmnpz and stoichiometric triethyl-
amine. In the presence of both Hdmnpz and Et3N in excess,
the orange ionic (Et3NH)2[Cu4(dmnpz)6] (2) species was isolated
instead. Indeed, the formation of 2 can be achieved by means
of a true titration, adding Et3N dropwise into an acetone solution
of [Cu(CH3CN)4](BF4) and Hdmnpz; the formation of complex
1, as a white suspension, was initially observed and is followed
(in the presence of excess Et3N) by its gradual dissolution to
give a clear, orange solution of 2. This observation, together
with the fact that reaction of 1 with excess Hdmnpz in the
presence of Et3N quantitatively affords complex 2 (eq 2),
strongly indicates that 1 is the intermediate species in the
formation of 2.
(2)
(1)
6(Et3NH)(dmnpz) (3)
A further acid-base equilibrium involving proton transfer
in (Et3NH)(dmnpz) accounts for the observation that free Et3N
stabilizes solutions of 2 (eq 4).
(Et3NH)(dmnpz) h Hdmnpz + Et3N
(4)
To evaluate the role of the cation in the stabilization of
complex 2, we have prepared a series of derivatives of this
species with cations not possessing acidic protons, (cat)2-
[Cu4(dmnpz)6], (cat ) K, 3a; nBu4N, 3b; [N(PPh3)2], 3c). These
derivatives can be directly obtained employing the appropriate
n
base (alcoholic KOH or Bu4NOH) in the synthetic procedure
described for 2 or by metathetic exchange from 2 with the
appropriate base or salt (KCl, KOH, nBu4NOH, nBu4NCl,
[N(PPh3)2]Cl) (see Experimental Section). This latter observa-
Et3N
4[Cu(dmnpz)]
3(Et NH) [Cu (dmnpz) ]
3 + 6Hdmnpz
8
3
2
4
6
1
tion, together with spectroscopic evidence (IR and H NMR)
(1)
(2)
and reactivity analogies, strongly suggests that, in complexes
3a-c, the [Cu4(dmnpz)6]2- anion retains its identity, irrespective
of the countercation. As expected, the new derivatives 3a-c
are quite stable in solution (in absence of oxygen), and
conductivity measurements revealed the predictable behavior
as 1:2 electrolytes.11
(2)
It is well-known that group 11 metals in the +1 oxidation
state form neutral binary compounds of general formula
[Cu(pz*)]n with the deprotonated form of several pyrazoles
(Hpz*). The few structurally characterized systems have been
found to possess a variety of molecular arrangements and
oligomeric or polymeric structures, with a definite tendency of
forming trimeric species.7 Thus, on the basis of these consid-
(8) La Monica, G.; Ardizzoia, G. A. Inorg. Synth. 1997, 31, 299.
(9) Masciocchi, N.; Moret, M.; Cairati, P.; Sironi, A.; Ardizzoia, G. A.;
La Monica, G. J. Am. Chem. Soc. 1994, 116, 7668.
(10) Differences between Hdmpz and Hdmnpz are outlined by the
corresponding pK values, representative of the acidity and basicity of
diazoles:
(6) (a) Ardizzoia, G. A.; Beccalli, E. M.; La Monica, G.; Masciocchi, N.;
Moret, M. Inorg. Chem. 1992, 31, 2706. (b) Angaroni, M.; Ardizzoia,
G. A.; La Monica, G.; Beccalli, E. M.; Masciocchi, N.; Moret, M. J.
Chem. Soc., Dalton Trans. 1992, 2715.
Hpz* + H2O h (pz*)- + H3O+
(7) (a) Ehlert, M. K.; Rettig, S. J.; Storr, A.; Thompson, R. C.; Trotter, J.
Can. J. Chem. 1990, 68, 1444. (b) Raptis, R. G.; Murray, H. H.;
Fackler, J. P. Jr. Inorg. Chem. 1988, 27, 26. (c) Ehlert, M. K.; Storr,
A.; Thompson, R. C. Can. J. Chem. 1992, 70, 1121. (d) Ehlert, M.
K.; Rettig, S. J.; Storr, A.; Thompson, R. C.; Trotter, J. Can. J. Chem.
1989, 67, 1970. (e) Ehlert, M. K.; Rettig, S. J.; Storr, A.; Thompson,
R. C.; Trotter, J. Can. J. Chem. 1991, 69, 432. (f) Ehlert, M. K.; Rettig,
S. J.; Storr, A.; Thompson, R. C.; Trotter, J. Can. J. Chem. 1992, 70,
2161. (g) Bovio, B.; Bonati, F.; Banditelli, G. Inorg. Chim. Acta 1984,
87, 25.
(pK1): 15.00 (Hdmpz), 10.65 (Hdmnpz)
(H2pz*)+ + H2O h Hpz* + H3O+
(pK2): 4.06 (Hdmpz), - 0.45 (Hdmnpz)
See: Catalan, J.; Abboud, J. L. M.; Elguero, J. In AdVances in
Heterocyclic Chemistry; Katritzky, A. R., Ed.; Academic Press: New
York, 1987; Vol. 41, p 187.
(11) Geary, W. J. Coord. Chem. ReV. 1971, 7, 81.

4286 Inorganic Chemistry, Vol. 37, No. 17, 1998
Ardizzoia et al.
Figure 2. ORTEP plot of the [Cu4(dmnpz)6]2- anion, with partial
labeling scheme. Thermal ellipsoids are drawn at the 30% probability
level. Hydrogen atoms are omitted for clarity.
the metal atoms. The coordination at the metal centers is
completed by intertriangle contacts involving the copper atoms
of a symmetry-related (-1) molecule (Cu‚‚‚Cu, 3.329(7) Å),
giving rise to a stepped-ribbon secondary structure of triangular
complexes; a similar arrangement has also been found for all
other [M(pz*)]3 (pz* ) a generic pyrazolate) complexes
previously cited but [Cu(dmpz)]3 and [Cu(tmpz)]3, for which
these aurophilic interactions only produce dimeric supra-
molecular moieties. Among silver pyrazolates, it has recently
been shown that also the [Ag(pz)]39 and [Ag(dmpz)]312a trimers
show a secondary structure which extends throughout the crystal.
Interestingly, while most of the group 11 binary pyrazolates
show nuclearities and packing features similar to their purely
organic counterparts (obtained by formal substitution of the
N-M-N fragments by N-H‚‚‚N linkages),9 this is not true
for 1, since Hdmnpz is known to form, in the solid state, helical
(hydrogen bonded) polymers.14
The molecular structure of anion 2 is shown in Figure 2
together with a labeling scheme. Selected bond parameters are
reported in Table 1. The tetranuclear anion sits about a
crystallographic inversion center, thus having a strictly planar
metal core imposed by symmetry. The four copper atoms are
arranged at the corners of a parallelogram with side lengths of
3.324(1) (Cu(1)‚‚‚Cu(2)) and 3.019(1) (Cu(1)‚‚‚Cu(2′)) Å
(primed atoms were generated with the (-x, -y, 1 - z)
symmetry operation), well beyond the range normally accepted
for a Cu-Cu covalent bond. The internal angles of the
parallelogram are significantly different from 90°, Cu(2)-
Cu(1)-Cu(2′) and Cu(1)-Cu(2)-Cu(1′) being 101.92(2) and
78.08(2)°, respectively. The cationic compound [Cu4(pydz)6]2+ 15
is strictly akin to anion 2, containing a tetranuclear copper
framework bridged by six pyridazine ligands. However, this
complex exhibits a more regular metal arrangement with Cu‚
‚‚Cu interactions of 3.264(1) and 3.096(1) Å and Cu-Cu-Cu
angles equal to 90° within 1°. Each copper center in 2 is
trigonally coordinated by nitrogen atoms belonging to the 3,5-
dimethyl-4-nitropyrazolate ligands, bridging all edges of the
parallelogram. Similarly to Munakata’s complex,15 the long
Cu-Cu edges are doubly bridged by two dmnpz ligands placed
on opposite sides with respect to the Cu4 plane, in a nearly
orthogonal orientation (with dihedral angles between the dmnpz
rings (N(1)-N(2)) or (N(7)-N(8)) least-squares planes and Cu4
of 84.65(9) and 86.97(8)°, respectively). On the contrary, the
Figure 1. ORTEP plot of the [Cu(dmnpz)]3 (1) molecule, with partial
labeling scheme. Relevant bond distances (Å) and angles (deg) are
Cu1-N1, 1.87(4); Cu2-N2, 1.89(5); Cu2-N4, 1.83(4); Cu1‚‚‚Cu2,
3.185(11); Cu2‚‚‚Cu2′, 3.255(13); N1-Cu1-N1′, 175(4); N2-Cu2-
N4, 177(2); Intermolecular contact, Cu1‚‚‚Cu2′′, 3.329(7) (primed and
1
double-primed atoms generated by the -x, y, /2 - z and -x, -y, -z
symmetry operations, respectively).
Complexes 1 and 2 have been fully characterized by means
of X-ray diffraction. Due the marked insolubility of 1 in organic
solvents, preventing the growth of crystals suitable for a single-
crystal X-ray diffraction study, its structure has been solved ab
initio by means of X-ray powder diffraction analysis (XRPD).12
Complex 2, to the best of our knowledge, is the first structurally
characterized homoleptic anionic pyrazolate.13
Crystal and Molecular Structures of 1 and 2. The crystal
structure of 1 contains neutral trimeric copper(I) molecules with
idealized 6hm2 symmetry. The molecular structure of 1 is
depicted in Figure 1 together with a partial labeling scheme
and selected bonding parameters. A detailed analysis of its
geometry cannot be done, since it is known that XRPD studies
of complex molecules need the introduction of numerous
geometrical restraints and therefore cannot be compared to a
single-crystal X-ray analysis; however, important chemical
features such as stoichiometry, molecular shape, crystal packing,
and metal-metal interatomic distances are reasonably well
determined. Complex 1, analogously to many other homoleptic
pyrazolates of group 11 metals,7,9 exhibits a triangular arrange-
ment of metal atoms with all edges bridged by exo-bidentate
dmnpz ligands with Cu(1)-Cu(2) (3.185(11) Å) and Cu(2)-
Cu(2′) (3.255(13) Å) (since the molecule lies around a 2-fold
axis, primed atoms were obtained by the symmetry operation
1
-x, y, /2 - z). The copper atoms are linearly coordinated by
two nitrogen atoms of dmnpz ligands, which lie in the plane of
(12) (a) Masciocchi, N.; Cairati, P.; Sironi, A. Powder Diffr. 1998, 13, 35
and references therein. (b) Masciocchi, N.; Sironi, A. J. Chem. Soc.,
Dalton Trans. 1997, 4643.
(13) The synthesis and characterization of a series of anionic complexes
of 3,5-pyrazoledicarboxylic acid (H3dcbpz) as a dinucleating trianionic
ligand, i.e. [M2(cod)2(dcbpz)]-, [M2(CO)4(dcbpz)]- (M ) Rh, Ir), and
[M2(dcbpz)]2- (M ) Pt(II), Pd(II), Ni(II), Cu(II)) have been reported.
However, X-ray structure determinations evidenced that the trianionic
dcbpz ligand coordinates to the two metal centers via the pyrazole N
atom and the carboxylate O atoms. See: (a) Bayon, J. C.; Net, G.;
Esteban, P.; Rasmussen, P. G.; Bergstrom, D. F. Inorg. Chem. 1991,
30, 4771. (b) Bayon, J. C.; Esteban, P.; Net, G.; Rasmussen, P. G.;
Baker, K. N.; Hahn, C. W.; Gumz, M. M. Inorg. Chem. 1991, 30,
2572.
(14) Foces-Foces, C.; Cano, F. H.; Elguero, J. Gazz. Chim. It. 1993, 123,
477.
(15) Maekawa, M.; Munakata, M.; Kuroda-Sowa, T.; Nozaka, Y. J. Chem.
Soc., Dalton Trans. 1994, 603.

Synthesis, Structure and Reactivity of Complexes
Inorganic Chemistry, Vol. 37, No. 17, 1998 4287
Table 1. Selected Bond Distances (Å) and Angles (deg) for
Compounds 2, 4, 5, and 6
found in polymeric Hdmnpz for which a value of 1.34(1) Å
has been found. This phenomenon, as already pointed out by
Ehlert et al.,7a could be associated with a Cu(dπ)-dmnpz(π*)
interaction between metal and the nitropyrazole orbitals of acidic
character. None of the dmnpz rings deviate significantly from
planarity, all having the nitro group almost coplanar with the
ring.
Reactivity of 1 with Cyclohexyl Isocyanide (RNC) and
Triphenylphosphine. The reaction of complex 1 with cyclo-
hexyl isocyanide (RNC) in diethyl ether leads to the isolation
of two different products depending on the copper/RNC molar
ratio. When the reaction is carried out with a Cu/RNC ratio of
1:1 the dimeric species [Cu(dmnpz)(RNC)]2 (4) was isolated
(eq 5).
compound 2
compound 6
Cu(1)‚‚‚Cu(2)
3.3238(8) Cu(1)‚‚‚Cu(2)
3.1813(6)
2.1877(7)
1.986(2)
1.969(2)
2.1863(7)
1.957(2)
2.020(2)
1.390(3)
1.394(3)
Cu(1)‚‚‚Cu(2′)
Cu(1)-N(1)
Cu(1)-N(4)
Cu(1)-N(7)
Cu(2)-N(2)
Cu(2)-N(5′)
Cu(2)-N(8)
N(1)-N(2)
N(4)-N(5)
N(7)-N(8)
3.019(1) Cu(1)-P(1))
1.981(3) Cu(1)-N(1)
1.977(3) Cu(1)-N(4)
1.975(3) Cu(2)-P(2)
1.980(3) Cu(2)-N(2)
1.955(3) Cu(2)-N(5)
1.951(3) N(1)-N(2)
1.400(4) N(4)-N(5)
1.391(4) N(1)-Cu(1)-P(1) 119.42(6)
1.402(4) N(1)-Cu(1)-N(4) 109.94(8)
Cu(1)‚‚‚Cu(2)‚‚‚Cu(1′) 78.08(2)
Cu(2)‚‚‚Cu(1)‚‚‚Cu(2′) 101.92(2)
N(4)-Cu(1)-P(1) 128.25(6)
N(2)-Cu(2)-P(2) 134.62(6)
N(2)-Cu(2)-N(5) 109.80(8)
N(5)-Cu(2)-P(2) 113.48(6)
2[Cu(dmnpz)]
3[Cu(dmnpz)(RNC)]
3 + 6RNC f
(5)
2
(1)
(4)
compound 4
compound 5
Cu‚‚‚Cu′
3.540(1)
1.974(3)
1.972(3)
1.869(6)
3.728(2)
2.037(2)
2.034(2)
1.930(3)
1.937(4)
1.389(3)
1.141(4)
1.459(4)
1.150(4)
1.520(4)
Analytical and spectroscopic data are in agreement with the
assigned formulation. In particular, the IR spectrum of 4 shows
only one absorption due to ν(NC) stretching at 2166 cm-1, a
value close to those found in a series of derivatives of general
formula [Cu(pz*)(RNC)]2 synthesized by us in the past years
(Hpz* ) pyrazole, Hpz16; 3,5-dimethylpyrazole, Hdmpz16;
3,5-diphenylpyrazole, Hdppz5a; 3,5-dicarbomethoxypyrazole,
Hdcmpz6a). Complex 4 was fully characterized by means of a
single-crystal X-ray diffraction study (see later).
When complex 1 was reacted with excess RNC, a different
derivative, analyzing as Cu(dmnpz)(RNC)2 (5), was obtained.
The IR spectra of 5 registered either in the solid state or in
solution shows two absorptions attributable to ν(NC) (2165 and
2143 cm-1, Nujol; 2169 and 2155 cm-1, CH2Cl2 solution)
indicating the presence of two RNC ligands coordinated to a
copper center. Two possible structures can account for the
assigned formulation: a mononuclear species having a dmnpz
group coordinated in a rare mono-dentate fashion (A) or a
dinuclear species having a central core, Cu-[N-N-]2-Cu, as
in derivative 4, but possessing two tetracoordinated copper(I)
centers (B).
Cu-N(1)
Cu-N(2′)
Cu-C(6)
Cu-C(13)
N(1)-N(2)
N(4)-C(6)
N(4)-C(7)
N(5)-C(13)
N(5)-C(14)
1.388(4)
1.142(7)
1.463(5)
N(1)-Cu-N(2′)
C(6)-Cu-N(1)
C(6)-Cu-N(2′)
C(13)-Cu-N(1)
C(13)-Cu-N(2′)
C(6)-Cu-C(13)
N(4)-C(6)-Cu
C(6)-N(4)-C(7)
N(5)-C(13)-Cu
C(13)-N(5)-C(14)
113.78(13)
121.5(2)
124.6(2)
109.87(9)
108.62(12)
108.84(12)
108.60(12)
107.65(13)
113.24(14)
176.3(3)
173.4(3)
176.3(3)
176.5(3)
178.3(7)
176.9(7)
short edges of the metal parallelogram are bridged by only one
dmnpz ligand nearly coplanar with the Cu4 core (dihedral angle
between the dmnpz (N(4)-N(5)) least-squares plane and Cu4
) 8.5(2)°). Given the role of the bridging exo-bidentate dmnpz
ligands and the actual values of the Cu-Cu distances, any
interaction between the copper atoms is weak at best. The two
crystallographically independent copper atoms, while being both
trigonally coordinated, exhibit different degrees of pyramidal-
ization. Indeed, Cu(1) is placed within experimental error in
the same plane of its coordinating atoms N(1), N(4), and N(7)
(deviation 0.003(2) Å), while Cu(2) is well off the plane defined
by N(2), N(5′), and N(8) (deviation 0.795(3) Å), with the apex
of the pyramid pointing away from the center of the molecule.
The distortion of the coordination environment is also reflected
by the deviation of the copper atoms from the dmnpz ligands
planes (Cu(1), 0.095(6) and 0.174(6) Å out of the N(1)-N(2)
and N(7)-N(8) dmnpz planes, respectively; Cu(2), 0.559(6) and
0.609(6) Å out of the N(1)-N(2) and N(7)-N(8) dmnpz planes,
respectively). [For Cu(2) there is also an in plane distortion
with two Cu-N interactions significantly shorter than the third
one (Cu(2)-N(5′) 1.955(3), Cu(2)-N(8) 1.951(3), Cu(2)-N(2)
1.980(3) Å).] The Cu(1)(N-N)2Cu(2) six-membered metallo-
cycle displays a sofa conformation with Cu(2) 0.586(3) Å out
of the Cu(1)/N(1)/N(2)/N(7)/N(8) average plane (maximum
deviation 0.038(2) Å). The dmnpz ligands bridging the long
side of the tetracopper parallelogram face each other at a
distance of about 3.4 Å, allowing some π-π stabilizing
intramolecular interaction. The N-N bond distances of the
three ligands are markedly longer (average 1.398 Å) than those
A
B
The 1H NMR of 5, registered in CD2Cl2 at room temperature
shows, beside the signals attributable to the cyclohexyl rings,
only one signal at 2.62 ppm derived from the methyl groups of
the pyrazolate ligand. The spectrum does not change signifi-
cantly on lowering the temperature, as instead expected for a
structure of type A, due the well-known scrambling of the
nitrogen atoms that equalizes the 3 and 5 sites on the pyrazolate
(16) Ardizzoia, G. A.; Angaroni, M.; La Monica, G.; Masciocchi, N.; Moret,
M. J. Chem. Soc., Dalton Trans. 1990, 2277.

4288 Inorganic Chemistry, Vol. 37, No. 17, 1998
Ardizzoia et al.
Figure 3. ORTEP plot of the [Cu(dmnpz)(RNC)]2 (4) molecule, with
partial labeling scheme. Thermal ellipsoids are drawn at the 30%
probability level. Hydrogen atoms are omitted for clarity.
ligands coordinated in a monodentate mode.17 Hence, a
structure type B for complex 5 in solution seems the more
plausible. The same structure was found in the solid state by
means of a single-crystal X-ray diffraction study (see later).
Complex 5 can also be obtained by reacting 4 with RNC (eq
6), and despite the highly steric crowding at the copper(I) centers
it is very stable in solution, not showing any dissociation of
the cyclohexyl isocyanide ligands.
Figure 4. ORTEP plot of the [Cu(dmnpz)(RNC)2]2 (5) molecule, with
partial labeling scheme. Thermal ellipsoids are drawn at the 30%
probability level. Hydrogen atoms are omitted for clarity.
RNC
RNC8
2[Cu(dmnpz)]3
3[Cu(dmnpz)(RNC)]
8
2
(1)
(4)
3[Cu(dmnpz)(RNC)2]2
(6)
(5)
Also in this case the peculiar electronic properties of the
dmnpz ligand play a crucial role in determining the nature of
the final product. Indeed, the related [Cu(dmpz)]3 derivative,
possessing a molecular arrangement and steric hindrance
comparable to that of complex 1, reacts with RNC giving only
the dinuclear, tricoordinated, copper(I) derivative [Cu(dmpz)-
(RNC)]2 even in the presence of excess cyclohexyl isocyanide.16
The reaction of complex 1 with PPh3 exclusively affords the
dinuclear species [Cu(dmnpz)(PPh3)]2 (6), also in the presence
of an excess of ligand (eq 7).
Figure 5. ORTEP plot of the [Cu(dmnpz)(PPh3)]2 (6) molecule, with
partial labeling scheme. Thermal ellipsoids are drawn at the 30%
probability level. Hydrogen atoms are omitted for clarity.
The C-Cu-N angles average to 123.0°, the N(1)-Cu-N(2′)
angle being 113.8(1)°, similarly to that observed for complex
[Cu(3,5-dmpz)(RNC)]216 (123.46(6) and 113.0(1)°, respectively).
As in the latter complex, also in the present case the six-
membered Cu-[N-N]2-Cu metallocycle is nearly planar with
maximum deviation from the least-squares plane of 0.024(2) Å
for atoms N(1) and N(2). Also, the two independent Cu-N
bond distances (average 1.973 Å) are approximately equal to
2[Cu(dmnpz)]
3[Cu(dmnpz)(PPh )]
3 + 6PPh3 f
(7)
3
2
(1)
(6)
Complex 6 is quite stable in solution (in absence of O2), and
was characterized by a single-crystal X-ray diffraction analysis.
The same derivatives obtained according to reactions 5-7
can be isolated employing the tetranuclear anionic complexes
(cat)2[Cu4(dmnpz)6] (2, 3a-c), instead of [Cu(dmnpz)]3 (1). In
these cases, the formation of (cat)(dmnpz) as byproduct of the
reactions was evidenced.
Crystal and Molecular Structures of 4, 5 and 6. Crystals
of 4, 5, and 6 contain discrete dimeric complexes packed with
normal van der Waals interactions. Their molecular structure
are reported in Figures 3, 4, and 5, respectively, with a partial
labeling scheme. Selected bond parameters are reported in
Table 1.
Complex 4 lies on an inversion center and contains two
copper(I) atoms linked by two exo-bidentate dmnpz ligands;
also connected to each metal atom there is one cyclohexyl
isocyanide ligand. Therefore the copper atoms attain a planar
trigonal coordination with a small deviation from the plane of
atoms C(6), N(1) and N(2′) (0.032(5) Å) (primed atoms are
related by the symmetry operator 1 - x, -y, -z) and a non
bonding Cu‚‚‚Cu distance of 3.540(1) Å. The dimeric complex
possesses an overall idealized 2/m symmetry with the 2-fold
axis bisecting the nitro groups and the mirror plane passing
through the copper atoms and NC groups.
16
those evidenced within compound [Cu(3,5-dmpz)(RNC)]2
(1.960(2) Å).
Complex 5 lies around an inversion center but displays a
much higher idealized symmetry (mmm) with three orthogonal
mirror planes. The copper atoms are connected to two nitrogen
atoms of bridging 3,5-dimethyl-4-nitropyrazolates and two
cyclohexyl isocyanide ligands. The latter are on opposite sides
of the central Cu2N4 ring, thus generating a tetrahedral
coordination around the metal atoms. The increased steric
crowding around the metal centers with respect to complex 4
is evidenced by the Cu-N and Cu-C bond distances which
have an average value of 2.035 and 1.933 Å, respectively,
significantly longer than those observed in compound [Cu(3,5-
dmpz)(RNC)]216 and in 4. As a side effect, the intradimer Cu‚
‚‚Cu distance increases from 3.540(1) in 4 to 3.728(2) Å in 5
(which, to our knowledge, is the longest ever observed in µ2-
pyrazolate bridged dimers) and the N(1)-Cu-N(2′) angle drops
to 109.87(9)° on changing the coordination from planar trigonal
to tetrahedral. The six-membered cycle defined by the two
copper and the four nitrogen atoms also in this case is planar
(maximum deviation from the least-squares plane fit of 0.006(1)
Å).
(17) (a) Bandini, A. L.; Banditelli, G.; Bonati, F.; Minghetti, G.; Demartin,
F.; Manassero, M. J. Organomet. Chem. 1984, 269, 91. (b) Ro¨ttger,
D.; Erker, G.; Grehl, M.; Fro¨hlich, R. Organometallics 1994, 13, 3897.
Compound 6 is a dimeric molecule with two copper(I) ions
trigonally coordinated by two bridging dmnpz ligands and one

Synthesis, Structure and Reactivity of Complexes
Inorganic Chemistry, Vol. 37, No. 17, 1998 4289
Table 2. ν(CO) Absorptions (cm-1)of CO Adducts (7a-d) of
(cat)2[Cu4(dmnpz)6], 2 and 3a-c
at -20 °C, but rapidly lose CO if heated under vacuum (60 °C,
10-2 Torr), restoring the starting complexes (eq 8). This last
ν(CO)
4CO d
(cat) [Cu (dmnpz) ]
+
2
4
6
complex
cat
solution
Nujol
N2 or vacuum
(2, 3a-c)
7a
7b
7c
7d
Et3NH
K
Bu4N
2050
2050
2050
2050
2045
2052
2052
2044
(cat)2[Cu4(dmnpz)6(CO)4]
(8)
(7a-d)
[N(PPh3)2]
observation prompts us to suggest that the uptake of CO does
not cause a significant rearrangement of the molecular structure
of the starting complexes and that the carbonyl derivatives
maintain a tetranuclear arrangement. The thermal behavior
could not be easily explained if major rearrangements leading
to the tetrameric units were required.
In addition, a thermogravimetric study of the potassium
derivative, 7b, decarbonylation, revealed the presence of one
molecule of CO per copper center. On these basis, the
formulation of derivatives 7a-d as tetranuclear species contain-
ing tetracoordinated copper(I) centers seems plausible.
triphenylphosphine each. The presence of the methyl substit-
uents in positions 3 and 5 on the pyrazolates and of the bulky
triphenylphosphine makes the steric requirements demanding
with respect to the final conformation of the whole dimeric
molecule. Indeed, differently from compound 4, in which the
metal atoms, the pyrazolates and the NC groups lie in the same
plane, complex 6 shows a highly puckered Cu-[N-N]2-Cu
six-membered ring. Adopting an almost planar disposition of
the metallocycle, as observed in compound [Ag(pz)2(PPh3)2],18
would cause strong steric repulsion between the phenyl and
methyl hydrogen atoms. The conformation presently observed
guarantees the minimization of these unfavorable repulsions and
at the same time allows for an efficient intramolecular packing
of the six phenyl rings, with two of them (C(13x) and C(22x))
nearly parallel (dihedral angle 17.0(1)°). The coordination
around the copper atoms significantly deviates from planarity
with Cu(1) and Cu(2), 0.182(1) and 0.169(1) Å, respectively,
out of the pertinent coordination plane. A strong distortion of
the pyrazolate exo-bidentate bonding mode is also observed, in
which the copper atoms lie out of the pyrazolate average planes
with deviations ranging from a minimum of 0.163(4) Å for
Cu(2) with respect to dmnpz (N(1)-N(2)) up to 0.755(4) Å for
Cu(2) with respect to dmnpz (N(4)-N(5)). The resulting
conformation of the six membered Cu-[N-N]2-Cu ring is
twist-boat; it represents an additional example of the confor-
mational variety in this class of complexes, dictated by steric
and electronic requirements, which allowed the observation of
planar, boat, chair, and sofa conformations.
-
-
A similar behavior was also found for the previously reported
tetranuclear [Cu(OtBu)(CO)]4 copper(I) carbonyl complex,
which was shown to retain, after carbonylation, the nuclearity
of the parent [Cu(O-tBu)]4 complex, although with Cu4O4 cores
possessing different geometries (cubane vs planar).20
A further substantiation of this hypothesis is found in the
1
close similarity between the H and 13C NMR spectra of the
Reaction of (cat)2[Cu4(dmnpz)6] with Carbon Monoxide.
Complexes 2 (cat ) Et3NH) and 3 (cat ) K, 3a; Bu4N, 3b;
[N(PPh3)2], 3c) react with CO in acetone or CH2Cl2 affording
the carbonyl derivatives 7a-d as indicated by infrared spec-
troscopy (Table 2). The uptake of CO is very fast, as revealed
by the fading of the color of the solutions (from orange-yellow
to colorless) and it is easily reversed by bubbling N2 into the
solution. The carbonyl derivatives can be isolated in the solid
state by evaporating the solution in a flow of carbon monoxide.
Attempts to grow single crystals of the carbonyl derivatives
suitable for X-ray analysis were unsuccessful, resulting in
recrystallization of the starting complexes. The formulation of
these derivatives is hence inferred on the basis of spectroscopic
evidences. As reported in Table 2 the ν(CO) values are typical
for a terminal CO group bound to a copper(I) center in a
tetrahedral environment.19 In solution, all four derivatives show
the ν(CO) at 2050 cm-1 suggesting an identical formulation
for all species (i.e., independently on the cation). Moreover,
only a slight difference exists between the ν(CO) registered in
solution and in Nujol mulls, suggesting that the solution structure
is maintained also in the solid state.
starting complexes and those of the carbonylated species. The
1H NMR spectrum of complex 3a shows only one signal
attributable to the methyl groups of the dmnpz ligands (2.31
ppm); bubbling CO causes only a slight downfield shift of the
signal (2.56 ppm). This behavior is substantially unchanged
on lowering the temperature. The 13C NMR spectrum shows
similar results, as long as the signals attributable to the dmnpz
ligand are concerned. Unfortunately, the signals due to the
coordinated CO molecules went undetected (probably due to
the lability of the CO ligand), hiding helpful information.
The possible formation of a dinuclear species of formula
[Cu(dmnpz)(CO)]2, reminiscent of the structure of the RNC or
PPh3 derivatives (4 and 6), is, in our opinion, to be ruled out
on the basis of two observations: (i) the frequency of the ν(CO)
does not match those found in tricoordinated copper(I) centers,19
and (ii) the uptake of CO is not depleted by the presence in the
reaction medium of the (cat)(dmnpz) salts, as verified in the
case of (Et3NH)2[Cu4(dmnpz)6] (2). In fact, in this case, the
presence in solution of Et3N and Hdmnpz was found to be
necessary in order to prevent the precipitation of [Cu(dmnpz)]3.
A dinuclear pyrazolate complex, having only one CO group
coordinated in a terminal mode to a copper(I) center, [Cu2-
(dcmpz)2(py)2(CO)], was indeed recently reported by us and
crystallographically characterized.6a However, in that case, the
Complexes 7a-d are stable enough in the solid state at room
temperature if maintained under a CO atmosphere or under N2
(18) Ardizzoia, G. A.; La Monica, G.; Maspero, A.; Moret, M.; Masciocchi,
N. Inorg. Chem. 1997, 36, 2321.
(19) (a) Sorrell, T. N.; Jameson, D. L. J. Am. Chem. Soc. 1983, 105, 6013.
(b) Pasquali, M.; Floriani, C. In Copper Coordination Chemistry:
Biochemical and Inorganic PerspectiVes; Karlin, K. D., Zubieta, J.,
Eds.; Adenine Press: Duilderland, NY, 1983; p 311.
(20) (a) Tsuda, T.; Habu, H.; Horiguchi, S.; Saegusa, T. J. Am. Chem. Soc.
1974, 96, 5930. (b) Geerts, R. L.; Huffman, J. C.; Folting, K.; Lemmen,
T. H.; Caulton, K. G. J. Am. Chem. Soc. 1983, 105, 3503.

4290 Inorganic Chemistry, Vol. 37, No. 17, 1998
Ardizzoia et al.
C, 30.72; H, 3.07; N, 21.50. Found: C, 30.15; H, 2.94; N, 20.79.
Calcd for C62H108Cu4N20O12 (3b) (76% yield): C, 47.15; H, 6.84; N,
17.74. Found: C, 47.35; H, 6.70; N, 17.24.
presence of pyridine was necessary to “depolymerize” the
starting homoleptic (presumably trimeric) [Cu(dcmpz)]x com-
plex, affording the dinuclear [Cu(dcmpz)(py)]2 intermediate as
the actual reactive species.
Synthesis of {[N(PPh3)2]}2[Cu4(dmnpz)6], 3c. To a CH2Cl2 (5 mL)
suspension of complex 3a (150 mg, 0.128 mmol) was added [N(PPh3)2]-
Cl (147 mg, 0.256 mmol) under stirring. A white precipitate of KCl
rapidly formed. The precipitate was filtered off under nitrogen, and
the clear solution was evaporated to dryness, giving a yellow residue
that was washed with methanol and dried under vacuum (80% yield).
Anal. Calcd for C102H96Cu4N20O12P2 (3c): C, 58.06; H, 4.55; N, 23.28.
Found: C, 58.16; H, 4.38; N, 23.01.
Synthesis of [Cu(dmnpz)(RNC)]2, 4. To a suspension of complex
1 (100 mg, 0.164 mmol) in diethyl ether (10 mL) was added cyclohexyl
isocyanide (RNC) (61.0 µL, 0.46 mmol, Cu/RNC molar ratio 1/1) under
stirring. The white suspension was stirred for 2 h at room temperature.
Complex 4 was then collected by filtration, washed with diethyl ether,
and dried under vacuum (95% yield). Complex 4 can also be obtained
from the ionic complexes 2 and 3a-d; in this case, the presence of
free neutral pyrazole in the mother liquors was confirmed. Crystals
suitable for X-ray structure analysis were obtained by slow diffusion
of n-hexane into dichloromethane solution of 4. Anal. Calcd for
C24H34Cu2N8O4: C, 46.08; H, 5.44; N, 17.92. Found: C, 45.83; H,
5.47; N, 17.44.
Synthesis of [Cu(dmnpz)(RNC)2]2, 5. Complex 5 was obtained
as described for 4 but using excess cyclohexyl isocyanide (Cu/RNC
molar ratio 1/4) (92% yield). Crystals suitable for X-ray structure
analysis were obtained by slow diffusion of n-heptane into 1,2-
dichloroethane solution of 5. Anal. Calcd for C38H56Cu2N10O4: C,
54.09; H, 6.64; N, 16.61. Found: C, 53.80; H, 6.64; N, 16.20.
Synthesis of [Cu(dmnpz)(PPh3)]2, 6. Complex 1 (150 mg, 0.246
mmol) was suspended in acetone (10 mL) under nitrogen atmosphere,
and triphenylphosphine (579 mg, 2.21 mmol) was added under stirring.
The clear solution formed was then evaporated to dryness, yielding a
yellow solid. It was washed with ethanol and diethyl ether and dried
under vacuum (85% yield). Complex 6 can be alternatively obtained
from the ionic derivatives 2 and 3a-c. Crystals suitable for X-ray
structure analysis were obtained by slow diffusion of diethyl ether into
an acetone solution of 6. Anal. Calcd for C46H42Cu2N6O4P2: C, 59.29;
H, 4.51; N, 9.02. Found: C, 59.54; H, 4.53; N, 9.19.
Synthesis of Carbonyl Adducts. (Et3NH)2[Cu4(dmnpz)6(CO)4],
7a. Complex 2 (200 mg, 0.154 mmol) was dissolved under nitrogen
in the minimum volume of acetone (ca. 5 mL) in the presence of 0.5
mL of triethylamine. Carbon monoxide was then bubbled through the
solution. The orange solution rapidly turned colorless, and IR
monitoring showed the presence of a strong absorption at 2050 cm-1
attributable to the carbonyl species. A pale-yellow solid was obtained
removing the solvent in a flow of CO.
(cat)2[Cu4(dmnpz)6(CO)4] (cat ) K, 7b; Bu4N, 7c; [N(PPh3)2],
7d). These derivatives were obtained as described for 7a from acetone
solutions of complexes 3a-c (the presence of Et3N was in this case
unnecessary). The thermal lability of derivatives 7a-d precluded
elemental analysis based on combustion methods. The formulation of
these derivatives is given on the basis of spectroscopic and thermo-
gravimetric data (see text).
Thermogravimetry. Several TGA measurements carried out on
derivative 7b showed an average loss of weight of about 8% (required
for CO/Cu ) 1:8.7%). In a preparative experiment, 275.2 mg of 7b
was weighed in a glass bulb and then heated at 60 °C under vacuum
(10-2 Torr) for 4 h, obtaining 254.1 mg of 3a (∆m ) - 7.7%, CO/Cu
) 0.87).
Crystallography: XRPD Analysis. X-ray powder diffraction data
were obtained with Cu KR radiation (λ ) 1.5418 Å) on a Rigaku D-III
MAX horizontal-scan powder diffractometer equipped with parallel
(Soller) slits, a graphite monochromator in the diffracted beam a Na(Tl)I
scintillation counter, and a pulse height amplifier discrimination. The
generator was operated at 40 kV and 40 mA. Slits used: DS (1.0°);
AS (1.0°); RS (0.05°). The white powder (complex 1) was gently
ground in an agate mortar to ensure homogeneity in particle size and
then cautiously deposited on a quartz monocrystal (zero background
holder) with the aid of a binder (5% collodion in amyl acetate). Data
were collected at room temperature in the 5-95° range, in the θ-2θ
Conclusions
The synthesis, the structural properties and the reactivity
toward neutral ligands such as cyclohexyl isocyanide and
triphenylphosphine of the new homoleptic [Cu(dmnpz)]3 and
(cat)2[Cu4(dmnpz)6] copper(I) pyrazolato complexes have been
described. A comparison between [Cu(dmnpz)]3 and the
structural analogue [Cu(dmpz)]3 has shown that the electronic
properties of the dinucleating pyrazolate anions play a major
role compared with the steric ones as long as the reactivity of
the species is concerned; however, subtle steric features of the
ligand seems to drive the stereochemistry of the reaction
products (i.e., in particular the conformation of the Cu-[N-
N]2-Cu rings). Studies are in progress in order to verify these
hypotheses by employing suitable substituted pyrazoles.
In addition, the structural characterization of 1, which only
appears as fine microcrystalline powders, has shown that, if the
complexity of the structural problem is not too large, ab initio
XRPD methods from conVentional laboratory equipment may
afford valuable, otherwise inaccessible, structural information,
such as metal atom coordination, nuclearity of the species, and
its relevant crystal packing features.12b
Experimental Section
Solvents were dried and purified by standard methods. 3,5-
Dimethylpyrazole, cyclohexyl isocyanide, and triethylamine were used
as supplied (Aldrich Chemical Co.), and triphenylphosphine (Merck)
was recrystallized from methanol before use. [Cu(CH3CN)4](BF4)21
and 3,5-dimethyl-4-nitropyrazole22 were prepared according to literature
procedures. Infrared spectra were taken on a Bio-Rad FTIR 7
instrument, thermogravimetric analyses were obtained under nitrogen
1
atmosphere employing a Perkin-Elmer TGA 7 HIGH IE, and H and
13C NMR spectra were acquired on a Bruker AC-200 FT spectrometer.
Elemental analyses were performed at the Microanalytical Laboratory
of this University (C, H, N). All reactions were carried out under a
dry nitrogen atmosphere using standard Schlenk techniques.
Synthesis of [Cu(dmnpz)]3, 1. To an acetone solution of 3,5-
dimethyl-4-nitropyrazole (400 mg, 2.84 mmol) was added [Cu(CH3-
CN)4](BF4) (850 mg, 2.70 mmol) under stirring. The colorless solution
was stirred for 5 min, and triethylamine (1 mL) was added dropwise.
A white precipitate suddenly formed. After 15 min stirring, complex
1 was collected by filtration, washed with acetone, and dried under
vacuum (95% yield). Anal. Calcd for C15H18Cu3N9O6: C, 29.48; H,
2.95; N, 20.64. Found: C, 29.31; H, 2.70; N, 20.25.
Synthesis of (Et3NH)2[Cu4(dmnpz)6], 2. Triethylamine (2 mL) was
added dropwise to a solution of 3,5-dimethyl-4-nitropyrazole (1.00 g,
7.09 mmol) and [Cu(CH3CN)4](BF4) (700 mg, 2.22 mmol) in acetone
(10 mL). A white precipitate (complex 1) initially formed, and
subsequently the color of the suspension turned to orange. The
suspension was stirred for an additional half hour, and then the orange
solid was filtered off, washed with a small amount of acetone (5 mL),
and dried under vacuum (81% yield). Crystals suitable for X-ray
analysis were obtained by slow evaporation of an acetone solution of
2. Anal. Calcd for C42H68Cu4N20O12: C, 38.83; H, 5.24; N, 21.57.
Found: C, 38.59; H, 5.17; N, 21.51.
Synthesis of K2[Cu4(dmnpz)6], 3a, and (Bu4N)2[Cu4(dmnpz)6], 3b.
These compounds were obtained as described for 2 but using a methanol
solution of KOH or (Bu4N)OH, respectively, instead of triethylamine
or by treating a methanol suspension of 2 with alcoholic KOH or
(Bu4N)OH. Anal. Calcd for C30H36Cu4K2N18O12 (3a) (90% yield):
(21) Kubas, G. J. Inorg. Synth. 1979, 19, 90.
(22) Elguero, J.; Jacquier, R.; Tien Duc, H. C. N. Bull. Soc. Chim. France
1966, 3727.

Synthesis, Structure and Reactivity of Complexes
Inorganic Chemistry, Vol. 37, No. 17, 1998 4291
Figure 6. Final Rietveld refinement plot for [Cu(dmnpz)]3 (1), with peak markers and difference plot at the bottom.
Table 3. Summary of Crystal Data for 1 and X-ray Powder
mode, step scan with ∆2θ ) 0.02° and fixed time 15 s. The lowest
2θ 13 peak positions (located by standard peak-search methods) were
fed to the trial and error indexing routines of TREOR.23 The program
found a reasonable agreement within the monoclinic system, with the
following parameters and figures of merit: a ) 20.07 Å; b ) 13.81
Å; c ) 7.89 Å; â ) 95.9°; M(13) ) 45; F(13) ) 72 (0.004; 33). No
rational transformations of this unit cell into higher symmetries could
be found. Systematic absences indicated C2/c as the probable space
group, which was subsequently confirmed by successful refinement.
ALLHKL24 was used to extract intensities in the 5-65° (2θ) range.
SIRPOW25 provided the initial atomic coordinates of two independent
copper atoms, one of which on the 2-fold axis. Remaining atoms were
located by difference Fourier techniques and geometrical model
building. The final refinements were performed using GSAS,26 which
allows bond parameters (Cu-N, N-N, C-N, N-O, and C-C) to be
restrained to known values; the relative contribution of the soft restraints
(listed in the Supporting Information) to the total ø2 was about 11%.
An unconstrained refinement was also performed but resulted, after
convergence, in a significant spread of chemically equivalent distances
and was therefore rejected. However, its associated esd’s are probably
more reasonable; therefore, all values reported in the text and in the
tables have been computed on the basis of these esd’s, obtained after
removal of the constraints. Low angle data (2θ < 17°, six reflections
only), being those most affected by instrumental aberrations (such as
for fixed-slit Bragg-Brentano geometry, loss of primary beam intensity,
and axial divergence effects), were discarded. Two isotropic thermal
parameters were refined, one for the metal atoms (0.058(2) Å2) and
one for the lighter ones (0.106(4) Å2). The contribution of the hydrogen
atoms to the scattering factors was neglected. Atomic scattering factors
were taken from the internal library of GSAS.26 The background was
refined using a six-term cosine Fourier series; a pseudo-Voigt function
was found to best fit the peak shapes. The angular dependence of the
peak-widths was modeled using the Caglioti et al.27 dependence, with
Diffraction Structural Analysis Parameters
empirical formula
fw
crystal system
space group
a, Å
b, Å
c, Å
C15H18Cu3N9O6
610.95
monoclinic
C2/c
20.051(2)
13.814(2)
7.8829(8)
95.905(5)
2171.9(6)
4
1224
1.868
293(2)
RigakuD-III/max
0.02
â, deg
V, Å3
Z
F(000)
D(calc), g cm-3
temperature, K
diffractometer
step, deg
time per step, s
2θ range, deg
no. of data points collected
no. of independent reflections
no. of data/restraints/parameters
15
17.0-95.0
3900
1047
3900/45/62
0.067
a
RF
b
Rp, Rwp
0.028, 0.041
a RF ) ∑||Fo| - |Fc||/∑|Fo|, where the sum runs over all space-
group permitted reflections. b Rp ) ∑|yi,o - yi,c|/∑yi,o; Rwp ) ∑wi(yi,o
- yi,c)2/∑wiyi,o2, where the sum runs over all measured data points. yi,o
and yi,c are the observed and calculated intensities at each ith point,
respectively, and wi is a statistical weight, taken as 1/yi,o.
crystals of 2, 4, 5, and 6 were mounted in air on a glass fiber tip and
then onto a goniometer head. Table 4 contains details of the crystal
data, data collection, and refinement procedures. Single-crystal X-ray
diffraction data were collected at room temperature (i) for 5 on a
Siemens SMART CCD three-circle area detector and (ii) for 2, 4, and
6 on an Enraf-Nonius CAD4 diffractometer using graphite-monochro-
matized Mo Ka radiation (λ ) 0.710 73 Å). Unit cell parameters and
an orientation matrix were obtained from least-squares refinement on
44 reflections measured in three different sets of 15 frames each in the
range 2 < θ < 23° for 5, while for the other compounds the setting
angles of 25 randomly distributed intense reflections with 10 < θ <
14° were processed by least-squares fitting.
For 5 the intensity data were collected using the ω scan technique
within the limits 4 < 2θ < 55°. Approximately a full sphere of data
was collected with frame width set to 0.3°, the sample-detector distance
fixed at 5.5 cm, and a beam exposure of 20 s per frame. The first 100
frames were recollected at the end of data collection to monitor crystal
decay, which was not observed. The collected frames were then
the V and Z (tan θ dependent) parameters set to zero. Final Rp, Rwp
,
and RF values were 0.028, 0.041, and 0.067, respectively. Figure 6
shows the final Rietveld refinement plot. Table 3 contains a summary
of crystal data and refinement parameters.
Crystallography: Single-Crystal Analysis of 2, 4, 5 and 6. (a)
Collection and Reduction of Single-Crystal X-ray Data. Suitable
(23) Werner, P. E.; Eriksson, L.; Westdahl, M. J. Appl. Crystallogr. 1985,
18, 367.
(24) Pawley, G. S. J. Appl. Crystallogr. 1981, 14, 357.
(25) Cascarano, G.; Favia, L.; Giacovazzo, C. J. Appl. Crystallogr. 1992,
25, 310.
(26) Larson, A. C.; Von Dreele, R. B. Lansce, Ms-H805; Los Alamos
National Laboratory: Los Alamos, New Mexico, 1990.
(27) Caglioti, G.; Paoletti, A.; Ricci, F. P. Nucl. Instrum. 1958, 3, 223.

4292 Inorganic Chemistry, Vol. 37, No. 17, 1998
Ardizzoia et al.
Table 4. Summary of X-ray Single-Crystal Data and Structure Refinement Parameters
2
4
5
6
empirical formula
fw
space group
a, Å
b, Å
c, Å
C42H36Cu4N20O12
1267.06
P21/c
11.026(1)
13.456(2)
18.668(5)
C24H34Cu2N8O4
625.68
P1h
C38H56Cu2N10O4
844.00
P1h
C46H42Cu2N6O4P2
931.88
P1h
7.418(1)
9.780(1)
11.177(3)
109.61(2)
101.34(3)
105.82(1)
696.8(2)
1
1.57
1.491
293(2)
0.0477, 0.1295
9.506(4)
9.957(2)
11.658(4)
86.73(2)
79.54(2)
82.47(4)
1075.1(6)
1
1.04
1.304
293(2)
0.0330, 0.0905
11.379(2)
13.592(2)
14.409(3)
81.22(1)
85.21(1)
87.55(1)
2193.7(7)
2
1.09
1.411
293(2)
0.0284, 0.0719
R, deg
â, deg
92.91(2)
γ, deg
V, Å3
2766.1(9)
2
1.59
1.521
293(2)
0.0412, 0.1184
Z
µ, mm-1
F(calc), g cm-3
temperature, K
R1, wR2a
a R1 ) ∑||Fo| - |Fc||/∑|Fo|; wR2 ) [∑w(Fo - Fc2)2/∑wFo ]1/2
.
2
4
processed by software SAINT for integration; an absorption correction
was applied together with decay/merging (SADABS) on the 12 334
with exception of the cation carbon atoms in 2 (which belonged to
two different disordered models having 70 and 30% occupancy factors,
respectively) and the minor component (14%) of the disordered
cyclohexyl isocyanide ligand in 4. To stabilize the refinements, the
disordered fragments required contraints on the C-C bond distances
and angles. Hydrogen atoms were included in calculated positions and
refined riding on their parent carbon atoms with isotropic displacement
parameters which were 1.5 (methyls, with two equally occupied
disordered models) or 1.2 (phenyls or cyclohexyl isocyanides) times
that of the pertinent carbon atom. No hydrogen atoms were added to
the disordered triethylammonium cation in 2 and the disordered
cyclohexyl isocyanide in 4. The final conventional agreement indexes
are listed in Table 4, together with other experimental details.
2
2
(Rσ ) ∑[σ(Fo )]/∑Fo ) ) 0.0353) collected reflections, 4809 of which
2
are unique, with Rint ) ∑|Fo2 - Fmean |/∑Fo2 ) 0.0198. For the crystals
examined using the CAD4 diffractometer the data collections were
performed by the ω scan method with variable scan speed and variable
scan range. The crystal stability under diffraction was checked by
monitoring three standard reflections every 180 min. In all cases, crystal
decay was not observed. An empirical absorption correction was
applied using ψ scans of three suitable reflections having ø values close
to 90°.28 The measured intensities were corrected for Lorentz,
2
polarization, background, and decay effects and reduced to Fo .
(b) Solution and Structure Refinement. The structures were solved
by direct methods (SIR9229) and difference Fourier methods and were
2
subsequently refined by full-matrix least-squares against Fo using
Acknowledgment. This work was supported by the Minis-
tero dell’Universita` e della Ricerca Scientifica e Tecnologica
(MURST) and by the Italian Consiglio Nazionale delle Ricerche
(CNR). The technical support of Mr. G. Mezza is also
acknowledged.
2
2
reflections with Fo g 3σ(Fo ) and the program SHELXL9330 on a
Silicon Graphics Indigo computer. Scattering factors for neutral atoms
and anomalous dispersion corrections were taken from ref 31. All
nonhydrogen atoms were given anisotropic displacement parameters,
(28) North, A. C. T.; Phillips, D. C.; Mathews, F. S. Acta Crystallogr.
1968, A24, 351.
(29) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435.
(30) Sheldrick, G. M. SHELXL-93: Program for structure refinement;
University of Go¨ttingen: Germany, 1994.
Supporting Information Available: Fractional atomic coordinates,
bond distances and angles, anisotropic displacement parameters, list
of geometrical restraints, and a full table of crystal data and structural
refinement parameters are supplied (29 pages). Ordering information
is given on any current masthead page.
(31) International Tables for Crystallography; Kluwer: Dordrecht, 1992;
Vol. C, Tables 4.2.6.8 and 6.1.1.4.
IC970875N