Dicopper Complexes with a Dissymmetric Dicompartmental Schiff Base-Oxime Ligand: Synthesis, Structure, and Magnetic Interactions

Article abstract of DOI:10.1021/ic9601145

A dicopper(II) complex with an unsymmetrical dicompartmental ligand containing bridging phenolic oxygen atoms was synthesized by stepwise condensation, in the presence of copper(II) ions, of 2,6-diformyl-4-methylphenol and diaminopropane, followed by hydroxylamine. The complex was isolated and characterized as Cl^-, [Cu₂Cl₄]^2-, and ClO₄^- derivatives. A byproduct of the reaction, a copper(II) complex with a substituted salycilaldoxime, was also structurally characterized. The structure of the binuclear cation of major interest is nearly planar, with the copper(II) ions slightly displaced to opposite sides of the plane of the donor atoms (<0.20 ?). [Cu₂Cl₄]^2- is an unusual anion, having been characterized previously in only five other compounds. It is flat, with both copper-(I) ions in trigonal planar environments and two chlorides bridging the two copper(I) centers. X-ray structure determinations of the [Cu₂Cl₄]^2- and ClO₄^- derivatives of the dicopper complex reveal a strong tendency for the formation of coordination dimers and tetramers. These aggregates arise because an oxime oxygen from one binuclear cationic complex coordinates axially to the copper atom of another such complex. In the case of the [Cu₂Cl₄]^2- salt, the anion forms additional bridges between binuclear complexes, resulting in polymeric chains. In the ClO₄^- salt, however, aggregation is limited to the tetramer because of terminal axially coordinated methanol molecules. Temperature-dependent magnetic susceptibility measurements show strong antiferromagnetic interaction (ca. -700 cm^-1) between the two copper(II) ions within one binuclear cationic complex. No magnetic interactions have been observed between binuclear units. Cyclic voltammograms were measured in dimethylformamide (DMF) solution and show two separate reduction waves at -1.47 and -1.04 V, as well as an oxidation wave at +0.04 V (vs Fc⁺/Fc). The mixed-valent Cu^II/Cu^I complex is thermodynamically stable toward disproportionation.

Full text of DOI:10.1021/ic9601145

510
Inorg. Chem. 1997, 36, 510-520
Dicopper Complexes with a Dissymmetric Dicompartmental Schiff Base-Oxime Ligand:
Synthesis, Structure, and Magnetic Interactions
Elena V. Rybak-Akimova,† Daryle H. Busch,*,† Pawan K. Kahol,‡ Nicholas Pinto,‡
Nathaniel W. Alcock,§ and Howard J. Clase§
Department of Chemistry, University of Kansas, Lawrence, Kansas 66045, Department of Chemistry,
Wichita State University, Wichita, Kansas 67208, and Department of Chemistry, University of Warwick,
Coventry CV4 7AL, United Kingdom
ReceiVed January 31, 1996X
A dicopper(II) complex with an unsymmetrical dicompartmental ligand containing bridging phenolic oxygen
atoms was synthesized by stepwise condensation, in the presence of copper(II) ions, of 2,6-diformyl-4-methylphenol
and diaminopropane, followed by hydroxylamine. The complex was isolated and characterized as Cl-, [Cu2Cl4]2-
,
-
and ClO4 derivatives. A byproduct of the reaction, a copper(II) complex with a substituted salycilaldoxime,
was also structurally characterized. The structure of the binuclear cation of major interest is nearly planar, with
the copper(II) ions slightly displaced to opposite sides of the plane of the donor atoms (<0.20 Å). [Cu2Cl4]2- is
an unusual anion, having been characterized previously in only five other compounds. It is flat, with both copper-
(I) ions in trigonal planar environments and two chlorides bridging the two copper(I) centers. X-ray structure
determinations of the [Cu2Cl4]2- and ClO4- derivatives of the dicopper complex reveal a strong tendency for the
formation of coordination dimers and tetramers. These aggregates arise because an oxime oxygen from one
binuclear cationic complex coordinates axially to the copper atom of another such complex. In the case of the
[Cu2Cl4]2- salt, the anion forms additional bridges between binuclear complexes, resulting in polymeric chains.
In the ClO4- salt, however, aggregation is limited to the tetramer because of terminal axially coordinated methanol
molecules. Temperature-dependent magnetic susceptibility measurements show strong antiferromagnetic interaction
(ca. -700 cm-1) between the two copper(II) ions within one binuclear cationic complex. No magnetic interactions
have been observed between binuclear units. Cyclic voltammograms were measured in dimethylformamide (DMF)
solution and show two separate reduction waves at -1.47 and -1.04 V, as well as an oxidation wave at +0.04
V (Vs Fc+/Fc). The mixed-valent CuII/CuI complex is thermodynamically stable toward disproportionation.
Introduction
The rich chemistry of complexes of dicompartmental ligands
binding identical metals (and to a lesser extent different metals)
has been applied to the investigation of magnetic exchange
interactions between the two metal ions, the stabilization of
mixed-valent species, and the activation of small molecules.1-6
Particularly significant examples of this type are planar ligands
with imine or amine donors and bridging phenolic oxygens,
which are usually referred to as Robson-type ligands (I in Figure
1; Figure 2a).5,6 In these ligands, substituents X, Y, and Z can
easily be varied, giving both symmetric and dissymmetric
ligands. Dicopper complexes have been especially significant,
because of their interesting properties and the relative simplicity
of their synthesis.5,6 In most cases the two copper centers are
strongly antiferromagnetically coupled.3,4,5b
Figure 1. (I) Robson-type binuclear complex. (II) Dissymmetric
complex of ligand (L) containing Schiff base and oxime compartments.
† University of Kansas.
‡ Wichita State University.
§ University of Warwick.
X Abstract published in AdVance ACS Abstracts, January 15, 1997.
(1) Guerriero, P.; Tamburini, S.; Vigato, P. A. Coord Chem. ReV. 1995,
139, 17.
(2) Fraser, C.; Ostrander, R.; Rheingold, A. L.; White, C.; Bosnich, B.
Inorg. Chem. 1994, 33, 324.
(3) Cairns, C. J.; Busch, D. H. Coord. Chem. ReV. 1986, 69, 1. Groh, S.
E. Israel J. Chem. 1976-1977, 15, 277. Zanello, P.; Tamburini, S.;
Vigato, P. A.; Mazzocchin, G. A. Coord. Chem. ReV. 1987, 77, 165.
(4) Kahn, O. Struct. Bonding (Berlin) 1987, 68, 89.
(5) (a) Vigato, P. A; Tamburini, S.; Fenton, D. E. Coord. Chem. ReV.
1990, 106, 25. (b) Casselato, U.; Vigato, P. A; Fenton, D. E.; Vidali,
M. Coord. Chem. ReV. 1979, 8, 199.
Figure 2. Different types of metal atom alignments in polynuclear
complexes.
Recently, the chemistry of Robson-type complexes has been
expanded by increasing the number of donor atoms in the
ligands and, consequently, the number of metal ions coordinated
by what are now multicompartmental ligands. For example,
(6) Atkins, A. J.; Black, D.; Blake, A. J.; Marin-Becerra, A.; Parsons, S.;
Ruiz-Ramirez, L.; Schro¨der, M. Chem. Commun. 1996, 457.
S0020-1669(96)00114-0 CCC: $14.00 © 1997 American Chemical Society

Cu2 Complexes with a Schiff Base-Oxime Ligand
Inorganic Chemistry, Vol. 36, No. 4, 1997 511
Scheme 1
complexes with four,7-11 six,12,13 eight,7a-c and twelve12 metals
have been described, and some of their magnetic interactions
have been characterized. In most of these compounds, however,
magnetic interaction is limited to the pairs of metal ions. Thus,
the cores of four-metal clusters (type b, Figure 2) are best
described as “dimers of dimers”, containing pairs of magneti-
cally coupled ions with no significant magnetic interaction
between the pairs,9,11 despite the apparent symmetry of the
simple representation of their structure. Similarly, dimers of
Figure 2a type dinuclear complexes do not show significant
magnetic interaction between dinuclear units.14,15 To the extent
of our knowledge, the only example of extended magnetic
interaction throughout such polynuclear complexes has been
observed for a “benzene-like” cyclic array of six metal ions.12
A more favorable situation can be expected if metal chains rather
than metal cores are present in the structure. Attempts have
recently been made to connect two dicompartmental Robson-
type ligands by means of an organic group,15,16 although the
coordination chemistry of such ligands has not been extensively
studied, and tetranuclear complexes of type c (Figure 2) were
not obtained.
We suggest a different approach to the construction of metal
chains based on Robson-type ligands, using two bimetallic units
connected to each other by a fifth metal ion (type d, Figure 2).
This requires dicompartmental ligands with additional donor
atoms capable of coordination with the “extra” metal ion.
Oxime groups are appropriate candidates, providing “external”
donors in the form of their oxygen atoms, while the oxime
nitrogen atoms are bound to the metal ions within a dicom-
partmental ligand. Some examples of polynuclear complexes
with bridging oxime ligands have been reported.17,18 Here we
report the synthesis and characterization of a binuclear building
block (II, Figure 1) for the type d (Figure 2) complexes. Despite
the asymmetric nature of the ligand, the dicopper complexes II
(Figure 1), which are characterized in this paper, are shown to
retain the advantages of symmetrical Robson-type complexes;
e.g., strong antiferromagnetic coupling within a binuclear cation
and stabilization of mixed-valent CuIICuI species toward dis-
proportionation. The additional donors present in the oxime
groups play the predicted role of bridging groups between
binuclear units, giving rise to coordination tetramers or poly-
mers.
(7) (a) McKee, V.; Tandon, S. S. J. Chem. Soc., Chem. Commun. 1988,
385. (b) McKee, V.; Tandon, S. S. Inorg. Chem. 1989, 28, 2901.
(c) McKee, V.; Tandon, S. S. J. Chem. Soc., Dalton Trans. 1991,
221. (d) McKee, V.; Tandon, S. S. J. Chem. Soc., Chem. Commun.
1988, 1334.
(8) Nanda, K. K.; Venkatsubramanian, K.; Majumdar, D.; Nag, K. Inorg.
Chem. 1994, 33, 1581.
(9) Sakijama, H.; Tokuyama, K.; Matsumura, Y.; Ohkawa, H. J. Chem.
Soc., Dalton Trans. 1993, 2329.
(10) Motoda, K.-i.; Sakiyama, H.; Matsumoto, N.; Ohkawa, H.; Fenton, D.
E. J. Chem. Soc., Dalton Trans. 1995, 3419. Sakiyama, H.; Motoda,
K.; Ohkawa, H.; Kida, S. Chem. Lett. 1991, 1133.
Results and Discussion
(11) Edwards, A. J.; Hoskins, B. F.; Robson, R.; Wilson, J. C.; Moubaraki,
B.; Murray, K. S. J. Chem. Soc., Dalton Trans. 1994, 1837. Grannas,
M. J.; Hoskins, B. F.; Robson, R. Inorg. Chem. 1994, 33, 1071.
Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Markiewicz, A.;
Murray, K. S.; Robson, R. Inorg. Chem. 1992, 31, 3585. Bell, M.;
Edwards, A. J.; Hoskins, B. F.; Kachab, E. H.; Robson, R. J. Am.
Chem. Soc. 1989, 111, 3603. Bell, M.; Edwards, A. J.; Hoskins, B.
F.; Kachab, E. H.; Robson, R. J. Chem. Soc., Chem. Commun. 1987,
1852.
(12) Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Benelli, C. Inorg.
Chem. 1995, 34, 5507. Tandon, S. S.; Thompson, L. K.; Bridson, J.
N. J. Chem. Soc., Commun. 1992, 911.
(13) Hoskins, B. F.; Robson, R.; Smith, P. J. Chem. Soc., Chem. Commun.
1990, 488.
Synthesis. We have synthesized and characterized three
dicopper complexes of an asymmetric ligand containing a Schiff
(16) Tandon, S. S.; Thompson, L. K.; Bridson, J. N. Inorg. Chem. 1993,
32, 32. Perez, M. A.; Bermejo, J. M. An. Quim. 1993, 89, 105. Perez,
M. A.; Bermejo, J. M. J. Org. Chem. 1993, 58, 2628.
(17) Krebs, C.; Winter, M.; Weyhermu¨ller, T.; Bill, E.; Wieghardt, K.;
Chaudhuri, P. J. Chem. Soc., Chem. Commun. 1995, 1913. Colacio,
E.; Dominguez-Vera, J. M.; Escuer, A.; Kivekas, R.; Romerosa, A.
Inorg. Chem. 1994, 33, 3914. Birkelbach, F.; Winter, M.; Florke,
U.; Haupt, H.-J.; Butzlaff, C.; Lengen, M.; Bill, E.; Trautwein, A. X.;
Wieghardt, K.; Chaudhuri, P. Inorg. Chem. 1994, 33, 3990. Ruiz,
R.; Lloret, F.; Julve, M.; Munoz, M. C.; Bois, C. Inorg. Chim. Acta
1994, 219, 179. Shiozuka, M.; Matsumoto, N.; Ohkawa, H. Bull. Chem.
Soc. Jpn. 1992, 67, 1988.
(14) Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; McKee, V.; Downard,
A. J. Inorg. Chem. 1992, 31, 4635.
(15) Tandon, S. S.; Thompson, L. K.; Bridson, J. N.; Bubenik, M. Inorg.
Chem. 1993, 32, 4621.
(18) Lineau, D.; Oshio, H.; Ohkawa, H.; Kida, S. J. Chem. Soc., Dalton
Trans. 1990, 2283.

512 Inorganic Chemistry, Vol. 36, No. 4, 1997
Rybak-Akimova et al.
Figure 3. Crystal structure of [Cu2L]2[Cu2Cl4]: front view of one dicopper cation, showing atomic numbering.
base fragment on one side and two oxime groups on the other
(II, Figure 1). The complexes contain the same cation [Cu2L]+
and different anions, chloride perchlorate, and [Cu2Cl4]2-
.
Perchlorate anion is not coordinated to the binuclear cation, the
Cu2Cl42- is bound to the Cu(II) in the binuclear cation through
chloride bridges in the solid complex, and the coordination mode
of Cl- in the chloride derivative has not been studied (see the
discussion of the structures below). The ligand L has two
different compartments, so a two-step synthesis was used for
the dicopper complex (Scheme 1). A similar approach has
previously been used to make other dicopper complexes with
unsymmetrical ligands.2,19-22
The product of the first step, CuL1, is insoluble in most
common organic solvents, and because this interferes with the
subsequent reaction, the dicopper complex was chosen to build
the oxime coordination compartment of the complex. In this
process the next step involves addition of a second copper atom
to give a soluble intermediate. After incorporation of the second
copper(II) ion into an empty compartment, the coordinated
aldehydes were transformed into oxime groups by condensation
with hydroxylamine. The reactions of metal complexes have
been widely used to make macrocyclic complexes, but they are
not generally employed in the synthesis of complexes with
oxime ligands, where the ligand is usually prepared prior to
complex formation. Our example demonstrates that the metal
ion promoted synthesis of oxime complexes can compete with
traditional methods.
In the case of polyfunctional compounds, the main reaction,
condensation of hydroxylamine with the free aldehyde groups
present in [Cu2L1]2+, is accompanied by a parallel reaction in
which azomethine bonds of the preformed Schiff base are
cleaved, apparently with the replacement of diaminopropane by
hydroxylamine. As a result, an unexpected byproduct, [Cu-
(L2)2] (Scheme 1), a copper(II) complex with two substituted
salicylaldoxime fragments, was isolated in a crystalline form.
The high reactivity of aldehyde and/or azomethine groups
toward substitution reactions is confirmed by the easy formation
of acetals; the dimethylacetal groups in Cu(L2)2 were formed
under mild reaction conditions, by reaction with the solvent
methanol.
Figure 4. Crystal structure of [Cu2L]2[Cu2Cl4]: side view of the
complex, showing the linkage into chains through oxime and anion
bridges.
the complex with a different anion ([Cu2Cl4]2-). This copper-
(I) anion is probably formed by the reduction of copper(II) by
hydroxylamine, and the yield of the tetrachlorodicuprate(I) salt
can be increased if one extra equivalent each of copper(II)
chloride and of hydroxylamine hydrochloride is used in the
second reaction step. The perchlorate salt was also obtained,
by the addition of NaClO4 to a solution of the chloride
derivative. Caution: perchlorate salts are often explosiVe.
Characterization. The synthesized binuclear complexes
were characterized by elemental analysis (including Cu analysis),
mass spectrometry (positive FAB), IR and electronic spectros-
copy and, for three compounds, by crystal structure determi-
nation. The main peak in the mass spectrum of each complex
corresponds to the molecular ion [Cu2L]+ with the isotopic
pattern characteristic of a dicopper compound. The IR spectra
support the reactions shown in Scheme 1; the spectra of all three
salts look similar, and the frequencies for [Cu2L]2[Cu2Cl4] are
cited here. The carbonyl stretch present at 1670 cm-1 in the
IR of the precursor [CuL1]21,23 disappears in the spectrum of
[Cu2L]+. The CdN stretch of the new azomethine bond in the
The main product of the reaction (Scheme 1) precipitates in
the form of [Cu2L]Cl. It is accompanied by a small amount of
(19) Fraser, C.; Bosnich, B. Inorg. Chem. 1994, 33, 338.
(20) Karunakaran, S.; Kandaswamy, M. J. Chem. Soc., Dalton Trans. 1994,
1595.
(21) Ohkawa, H.; Kida, S. Bull. Chem. Soc. Jpn. 1972, 45, 1759.
(22) Casellato, U.; Fregona, D.; Sitran, S.; Tamburini, S.; Vigato, P. A.
Inorg. Chim. Acta 1985, 110, 181.
(23) Gagne´, R. R.; Spiro, C. L.; Smith, T. J.; Hamann, C. A.; Thies, W.
R.; Shiemke, A. K. J. Am. Chem. Soc. 1981, 103, 4073.

Cu2 Complexes with a Schiff Base-Oxime Ligand
Inorganic Chemistry, Vol. 36, No. 4, 1997 513
Table 1. Crytallographic Data
[Cu2L]2+[Cu2Cl4]2- (A)
[Cu2L]4+[ClO4]4-‚4MeOH (B)
[Cu(L2)2] (C)
-
formula
Mr
syst
space group
a/Å
b/Å
[C21H21Cu2N4O4]2+[Cu2Cl4]2-
653.93
triclinic
P1h
8.0452(11)
8.7160(10)
16.913(2)
96.471(9)
93.088(8)
105.273(10)
1132.4(2)
1
[C88H100Cu8N16O20]4+[ClO4]4
2608.04
triclinic
P1h
8.637(7)
15.875(11)
19.44(2)
106.51(6)
97.50(7)
96.77(6)
2499(4)
1
C22H28CuN2O8
512.0
monoclinic
P21/c
12.273(5)
11.790(7)
8.048(5)
90
97.41(4)
90
c/Å
R/deg
â/deg
γ/deg
U/Å3
1154.8(11)
2
1.472
Z
D(calc/(g cm-3
T/K
)
1.918
295(2)
1.733
293(2)
293(2)
cryst. dimens/mm
wavelength, Å
µ(Mo KR)/mm-1
R1(obs F)a
0.84 × 0.21 × 0.092
0.709 26
3.066
0.037
0.100
0.06 × 0.10 × 0.18
0.709 26
1.869
0.0541
0.141
0.40 × 0.22 × 0.11
0.709 26
0.996
0.052
0.141
R2w (all F2)a
2
2
a R indices based on F2 are denoted R2 and those based on F are denoted R1, e.g., R2w ) [∑[w(Fo - Fc2)2]/∑[w(Fo )2]]1/2 and R1 ) ∑||Fo| -
|Fc||/∑|Fo|.
oxime fragment probably overlaps with that of the Schiff base
fragment, since only one strong band is observed at 1628 cm-1
in the spectrum of [Cu2L]+. Two NsO bands (characteristic
for oxime complexes and usually observed at 1060-1250
cm-1 24) appear in the spectrum of [Cu2L]+ after the completion
of the second step in Scheme 1: a new band appears at 1192
cm-1, and three absoptions are observed in the spectrum of
[Cu2L]+ at 1068, 1097, and 1115 cm-1, while only two bands
are present in the same region in the spectrum of [CuL1] (at
1082 and 1107 cm-1). A weak feature sometimes observed
for oxime complexes in the region between 2300 and 2700
cm-1, and attributed to the hydrogen-bonded fragment OsHsO,24
is not present in the spectra of our complexes. Strong bands at
1090 and 625 cm-1 in the spectrum of the perchlorate salt are
due to that anion.
The electronic spectra of [Cu2L]Cl and [Cu2L]2[Cu2Cl4],
measured in methanol solutions in the range 330-1200 nm,
are very similar. The chloride derivative gave the following
λmax (ꢀ) values (extinction coefficients per one complex
cation): 385 (3300), 600 (broad, 120), 700 sh (ca. 80), 945
(25), 1150 (20) nm. For the tetrachlorodicuprate(I) complex,
very similar spectra were recorded: 395 (2700), 610 (broad,
110), 700 sh (ca. 80), 945 (53), 1150 (73) nm. For dicopper
complexes with Robson-type ligands, an intense band in the
near-UV region is usually assigned to an intraligand transition
of the CdN fragment overlapping with a copper-phenolic
oxygen charge transfer transition, while a moderate band at
about 600 nm is assigned to d-d transitions.14,19-21,25-32 The
latter band is sometimes less broad and more intense than that
observed for our asymmetric complex. It might be suggested
that, in our complex, the copper ions in the oxime and Schiff
base compartments are not equivalent and absorb at slightly
different energies. However, almost the same ligand-field
spectrum was observed for a symmetrical analog of [Cu2L]+
(Figure 1, X ) Me, Y ) H, Z ) (CH2)3): 600 nm (90), 700(sh)
nm (60).25,27 An explanation for this spectrum (as well as an
alternative explanation for the spectrum of [Cu2L]+) requires
the Cu(II) ions to be five-coordinated, since two bands are
usually observed at similar wavelengths for five-coordinated
copper complexes, with the high-energy band more intense than
the low-energy one.33 A remarkable feature of the spectrum
of [Cu2L]2[Cu2Cl4] is its significant absorbance in the near-IR
region, which can be tentatively assigned to near-IR d-d bands
in tetragonal-pyramidal Cu(II) complexes.33
Crystal Structures. (A) [Cu2L][Cu2Cl4] The dicopper
complex cation with the asymmetric Schiff base-oxime ligand
is planar, with the copper(II) ions slightly displaced to opposite
sides of the plane of the donor atoms (0.19 Å for the Schiff
base compartment and 0.16 Å for the oxime compartment)
(Figures 3 and 4). The CusN and CusO distances are
comparable with those in analogous complexes,1,2,7a-c,14,28,34-43
with slightly shorter bond lengths in the oxime compartment
than on the Schiff base side (Table 2). The copper-copper
separation is 3.07 Å. The bridging angles CusO(Ph)sCu
within the binuclear cation are equal to 103.17(8)° (Cu(1)sO-
(6)sCu(2)) and 102.29(8)° (Cu(1)sO(17)sCu(2)). The ge-
(33) Lever, A. B. P. Inorganic Electronic Spectroscopy; Elsevier: Am-
sterdam, 1984.
(34) Lacroix, P.; Kahn, O.; Gleizes, A.; Valade, L.; Cassoux, P. NouV. J.
Chem. 1984, 8, 643.
(35) Lacroix, P.; Kahn, O.; Theobald, F.; Leroy, J.; Wakselman, C. Inorg.
Chim. Acta 1988, 142, 129.
(36) Gagne´, R. R.; Henling, L. M.; Kistenmacher, T. J. Inorg. Chem. 1980,
19, 1226.
(37) Hoskins, B. F.; McLeod, N. J.; Schaap, H. A. Aust. J. Chem. 1976,
29, 515.
(24) Egneus, B. Talanta 1972, 19, 1387. Burger, K.; Ruff, I.; Ruff, F. J.
Inorg. Nucl. Chem. 1965, 27, 179.
(25) Pilkington, N. H.; Robson, R. Aust. J. Chem. 1970, 23, 2225.
(26) Gagne´, R. R.; Koval, C. A.; Smith, T. J. J. Am. Chem. Soc. 1977, 99,
8367.
(27) Mandal, S. K.; Nag, K. J. Chem. Soc., Dalton Trans. 1983, 2429.
(28) Mandal, S. K.; Thompson, L. K.; Nag, K.; Charland, J.-P.; Gabe, E.
J. Inorg. Chem. 1987, 26, 1391.
(29) Gagne´, R. R.; Koval, C. A.; Smith, T. J.; Cimolino, M. C. J. Am.
Chem. Soc. 1979, 101, 4571.
(30) (a) Ohkawa, H.; Nishio, J.; Ohba, M.; Tadokoro, M.; Matsumoto, N.;
Koikawa, M.; Kida, S.; Fenton, D. E. Inorg. Chem. 1993, 32, 2949.
(b) Ohkawa, H.; Tadokoro, M.; Aratake, Y.; Ohba, M.; Shindo, K.;
Mitsumi, M.; Koikawa, M.; Tomono, M.; Fenton, D. E. J. Chem. Soc.,
Dalton Trans. 1993, 253.
(38) Mandal, S. K.; Thompson, L. K.; Nag, K.; Charland, J.-P.; Gabe, E.
J. Can. J. Chem. 1987, 65, 2815.
(39) Mandal, S. K.; Thompson, L. K.; Newlands, M. J.; Gabe, E. J.; Nag,
K. Inorg. Chem. 1990, 29, 1324.
(40) Mandal, S. K.; Thompson, L. K.; Newlands, M. J. Gabe, E. J. Inorg.
Chem. 1989, 28, 3707.
(41) Carlisle, W. D.; Fenton, D. E.; Roberts, P. B.; Casellato, U.; Vigato,
P. A.; Graziani, R. Transition Met. Chem. 1986, 11, 292.
(42) Brychcy, K.; Drager, K.; Jens, K.-J.; Tilset, M.; Behrens, U. Chem.
Ber. 1994, 127, 465.
(31) Long, R. C.; Hendrickson, D. N. J. Am. Chem. Soc. 1983, 105, 1513.
(32) Mandal, S. K.; Nag, K. J. Chem. Soc., Dalton Trans. 1984, 2141.
(43) Brychcy, K.; Drager, K.; Jens, K.-J.; Tilset, M.; Behrens, U. Chem.
Ber. 1994, 127, 1817.

514 Inorganic Chemistry, Vol. 36, No. 4, 1997
Rybak-Akimova et al.
Table 2. Selected Bond Lengths (Å) and Angles (deg)
(i) For [Cu2L]2+[Cu2Cl4]2- (A)a
Cu(1)-N(20)
Cu(1)-N(2)
Cu(1)-O(17)
Cu(1)-O(6)
Cu(1)-O(21)1
Cu(2)-N(13)
Cu(2)-O(6)
Cu(2)-N(9)
1.947(2)
1.955(2)
1.958(2)
1.960(2)
2.388(2)
1.965(2)
1.971(2)
1.972(2)
Cu(2)-O(17)
Cu(2)-Cl(2)
Cu(3)-Cl(1)
Cu(3)-Cl(2)
Cu(3)-Cl(1)2
Cu(3)-Cu(3)2
Cl(1)-Cu(3)2
1.996(2)
2.6462(9)
2.2115(11)
2.1516(9)
2.4664(11)
2.9937(10)
2.4664(11)
N(20)-Cu(1)-N(2)
N(20)-Cu(1)-O(17)
N(2)-Cu(1)-O(17)
N(20)-Cu(1)-O(6)
N(2)-Cu(1)-O(6)
O(17)-Cu(1)-O(6)
N(20)-Cu(1)-O(21)1
N(2)-Cu(1)-O(21)1
96.43(9)
O(6)-Cu(2)-O(17)
N(9)-Cu(2)-O(17)
Cu(1)-O(6)-Cu(2)
76.65(7)
161.92(9)
103.17(8)
92.57(8)
162.42(9)
169.94(8)
92.16(8)
77.79(7)
88.43(8)
91.66(8)
Cu(1)-O(17)-Cu(2) 102.29(8)
N(13)-Cu(2)-Cl(2)
O(6)-Cu(2)-Cl(2)
N(9)-Cu(2)-Cl(2)
O(17)-Cu(2)-Cl(2)
Cl(2)-Cu(3)-Cl(1)
97.20(8)
90.38(6)
99.02(7)
95.03(6)
142.75(5)
O(17)-Cu(1)-O(21)1 103.69(7)
O(6)-Cu(1)-O(21)1
N(13)-Cu(2)-O(6)
N(13)-Cu(2)-N(9)
O(6)-Cu(2)-N(9)
N(13)-Cu(2)-O(17)
96.57(7)
166.85(9)
Cl(2)-Cu(3)-Cl(1)2 116.48(4)
Cl(1)-Cu(3)-Cl(1)2 100.63(4)
Figure 5. Crystal structure of [Cu2L][ClO4]‚MeOH dimer, showing
atomic numbering:
97.50(10) Cu(3)-Cl(1)-Cu(3)2
79.37(4)
108.49(4)
91.87(8)
91.94(8)
Cu(3)-Cl(2)-Cu(2)
ligand (2.388(2) Å) or a chloride (2.6462(9) Å). The structure
is generally similar to those of analogous complexes with
Robson-type dicompartmental ligands, but the present complex
is the first with unsymmetrical ligands (with the exception of
the Bosnich ligand which, despite containing compartments
designed for six- and four-coordination, gives rise to two five-
coordinate Cu ions in its dicopper complex2).
The coordinated chloride is part of the complex [Cu2Cl4]2-
anion. This is an unusual anion, only previously characterized
in five other compounds,14,44 and, significantly, two of these
contain complex cations resembling that in the present example.
The [Cu2Cl4]2- anion is flat, with both copper(I) ions in trigonal
planar environments and two chlorides bridging the two Cu(I)
centers. The anion has a center of inversion, though the CuCl3
units are not symmetrical, their CusCl distances differing by
0.5 Å.
The [Cu2Cl4]2- anion bridges two [Cu2L]+ cations, and the
oxime-copper link to the other copper produces a chain
structure (Figure 4), with the CusO(oxime)sN(oxime) angle
equal to 108.5(2)°. Somewhat similar interactions have been
observed for other dioxime complexes, e.g., that of copper(II)
with dimethylglyoxime.45 A few Robson-type complexes are
also known in which one ligand appears to be axially coordi-
nated to the copper ion in another ligand. For example, a
benzene ring is situated 2.55 and 3.04 Å above the copper(I) in
one example,36 while in others the oxo group from a µ-oxotet-
rameric cation is axially coordinated to the copper ion in the
next ligand7a-c or the phenolic oxygen is axially coordinated
to Cu(II).30a These two trends (the dimerization of copper(II)
oximes and stack formation by means of axial coordination for
Robson-type complexes) clearly favor the formation of dimers
for oxime-containing, Robson-type complexes.
(ii) For [Cu2L]+[ClO4]-‚MeOH (B)b
Cu(11)-N(102)
Cu(11)-O(117)
Cu(11)-O(106)
Cu(11)-N(120)
Cu(11)-O(101)3
Cu(12)-N(113)
Cu(12)-O(106)
Cu(12)-O(117)
Cu(12)-N(109)
Cu(12)-O(221)
1.928(7)
1.955(5)
1.961(6)
1.977(7)
2.319(6)
1.962(8)
1.971(6)
1.975(6)
1.980(7)
2.269(6)
Cu(21)-N(220)
Cu(21)-N(202)
Cu(21)-O(217)
Cu(21)-O(206)
Cu(21)-O(2S)
Cu(22)-N(213)
Cu(22)-N(209)
Cu(22)-O(217)
Cu(22)-O(206)
Cu(22)-O(1S)
1.950(7)
1.954(7)
1.963(5)
1.967(5)
2.344(7)
1.961(7)
1.964(7)
1.976(5)
1.9962(6)
2.406(7)
N(102)-Cu(11)-O(117) 170.0(3) N(220)-Cu(21)-N(202) 97.2(3)
N(102)-Cu(11)-O(106)
O(117)-Cu(11)-O(106)
N(102)-Cu(11)-N(120)
O(117)-Cu(11)-N(120)
93.9(3) N(220)-Cu(21)-O(217) 92.5(3)
77.2(2) N(202)-Cu(21)-O(217) 170.3(3)
96.2(3) N(220)-Cu(21)-O(206) 166.8(3)
91.0(3) N(202)-Cu(21)-O(206) 92.4(3)
O(106)-Cu(11)-N(120) 159.7(2) O(217)-Cu(21)-O(206) 77.9(2)
N(102)-Cu(11)-O(101)3 91.7(2) N(220)-Cu(21)-O(2S)
O(117)-Cu(11)-O(101)3 95.0(2) N(202)-Cu(21)-O(2S)
O(106)-Cu(11)-O(101)3 105.6(2) O(217)-Cu(21)-O(2S)
N(120)-Cu(11)-O(101)3 91.7(3) O(206)-Cu(21)-O(2S)
94.1(3)
94.4(3)
85.8(3)
94.1(3)
N(113)-Cu(12)-O(106) 163.0(3) N(213)-Cu(22)-N(209) 97.5(3)
N(113)-Cu(12)-O(117)
O(106)-Cu(12)-O(117)
N(113)-Cu(12)-N(109)
O(106)-Cu(12)-N(109)
91.9(3) N(213)-Cu(22)-O(217) 93.0(3)
76.5(2) N(209)-Cu(22)-O(217) 169.5(3)
98.3(3) N(213)-Cu(22)-O(206) 168.0(3)
91.8(3) N(209)-Cu(22)-O(206) 92.6(3)
O(117)-Cu(12)-N(109) 167.3(3) O(217)-Cu(22)-O(206) 77.0(2)
N(113)-Cu(12)-O(221)
O(106)-Cu(12)-O(221)
98.1(3) N(213)-Cu(22)-O(1S)
96.4(2) N(209)-Cu(22)-O(1S)
96.3(3)
89.6(2)
89.6(2)
90.3(2)
O(117)-Cu(12)-O(221) 101.8(2) O(217)-Cu(22)-O(1S)
N(109)-Cu(12)-O(221)
84.3(3) O(206)-Cu(22)-O(1S)
Cu(11)-O(106)-Cu(12) 103.1(3) Cu(21)-O(206)-Cu(22) 102.2(3)
Cu(11)-O(117)-Cu(12) 103.2(3) Cu(21)-O(217)-Cu(22) 102.9(3)
Cu(11)-O(117)-Cu(12) 103.2(3)
(iii) For [Cu(L2)2] (C)c
Cu(1)-O(6)
Cu(1)-N(2)
O(1)-N(2)
N(2)-C(3)
1.865(3)
1.946(4)
1.401(4)
1.289(5)
C(5)-O(6)
C(8)-O(9)
C(8)-O(10)
1.323(5)
1.399(5)
1.417(5)
(B) [Cu2L]ClO4‚MeOH. Four dinuclear cations, similar to
that just described, are linked in a centrosymmetric tetramer
by axial coordination of oxime oxygens to copper(II) ions
(Figures 5 and 6). The geometry of all copper(II) centers is
square-pyramidal, with the elongated fifth bonds to axially
coordinated methanol molecules (in “terminal” cations) or oxime
oxygens (in “central” cations). The terminal axially-coordinated
methanol molecules prevent the formation of a polymeric chain.
O(6)4-Cu(1)-N(2)
O(6)-Cu(1)-N(2)
C(3)-N(2)-O(1)
a Symmetry transformations used to generate equivalent atoms
(indicated by superscript 1 or 2): (1) -x, -y + 1, -z + 1; (2) -x,
-y, -z. b Symmetry transformations used to generate equivalent atoms
(indicated by superscript 3): -x, -y + 2, -z. c Symmetry transforma-
tions used to generate equivalent atoms (indicated by superscript 4):
-x + 1, -y + 1, -z.
88.14(14)
91.86(14)
113.0(3)
C(3)-N(2)-Cu(1)
O(1)-N(2)-Cu(1)
C(5)-O(6)-Cu(1)
127.9(3)
119.1(3)
129.1(3)
(44) Banci, L.; Bencini, A.; Dei, A.; Gatteschi, D. Inorg. Chim. Acta 1984,
84, L11. Haasnoot, J. G.; Favre, T. L. F.; Hinrich, W.; Reedijk, J.
Angew. Chem., Int. Ed. Engl. 1988, 27, 856. Andersson, S.;
Hakansson, M.; Jagner, S. Inorg. Chim. Acta 1993, 209, 195. Escriche,
L.; Lucena, N.; Casabo, J.; Teixidor, F.; Kivekas, R.; Sillanpaa, R.
Polyhedron 1995, 14, 649.
ometry of both copper ions is square-pyramidal, each forming
an elongated fifth bond, either to an oxime oxygen of a different
(45) Vaciago, A.; Zambonelli, L. J. Chem. Soc. A 1970, 218.

Cu2 Complexes with a Schiff Base-Oxime Ligand
Inorganic Chemistry, Vol. 36, No. 4, 1997 515
Figure 6. Crystal structure of [Cu2L][ClO4]‚MeOH: side view of the tetrameric structure.
observed.20,30b,43,48 To produce extensively coupled polynuclear
magnetic materials on the basis of Robson-type ligands, it is
important to modify the ligands in a way which will provide
additional opportunities for building extended networks of
magnetically coupled centers. At the same time, such modifica-
tions should not destroy the strong coupling within binuclear
building blocks, which could accompany the introduction of
asymmetry into the ligand. For this reason, it is important to
find out whether or not the asymmetric Shiff base-oxime ligand
environment supports strong anitferromagnetic interaction be-
tween two copper(II) centers. The compond [Cu2L]2[Cu2Cl]4
has a crystal structure which appears to be favorable for
antiferromagnetic exchange, because it is very similar to the
structure of symmetrical Robson complexes where strong
antiferromagnetic interaction does occur.37,39,40 The structure
of the cation in the chloride derivative of the dicopper complex
is presumably similar to that of the binuclear cation in the
tetrachlorodicuprate(I) complex. Electronic non-equivalency of
two copper centers might, however, decrease the exchange
coupling between them.3,4 The experimentally determined room
temperature magnetic moments for both the chloride and
tetrachlorodicuprite derivatives of the new asymmetric Schiff
base-oxime dicopper complex are practically the same (0.65
B.M. per copper ion) and indicate antiferromagnetic coupling.
A cryomagnetic study was undertaken to determine the values
of coupling constants, with the results shown in Figure 8. The
measurements covered the temperature range from 80 to 450
K for [Cu2L]Cl (Figure 8a) and from 5 to 450 K for [Cu2L]2[Cu2-
Cl4] (Figure 8b). The behavior of the two complexes below
room temperature is very similar. The data for [Cu2L]Cl,
measured over the temperature range 80-300 K, were fitted to
the Bleaney-Bowers expression49
Figure 7. Crystal structure of a byproduct [Cu(L2)2].
The existence of this tetrameric structure in the absence of
bridging anions demonstrates that coordination of oxime oxygen
to copper ions alone is enough for cation aggregation. The
coordination modes of the copper(II) ions are similar to those
in Figure 3, but the relative nonplanarity (as compared to
compound A) of the individual units is notable (Figure 6).
However, the deviations of the Cu(II) ions from N2O2 plane
are reasonably small: 0.195 Å (Cu(11)) and 0.165 Å (Cu(12))
for central compartments (values very close to those for the
polymeric tetrachlorodicuprite salt); 0.078 Å (Cu(21) and 0.055
Å (Cu(22)). The CusCu separation in both terminal and central
cations is 3.08 Å. The CusO(Ph)sCu angles are as follows:
103.2(3)° (Cu(11)sO(117)sCu(12)) and 103.1(3)° (Cu(11)sO-
(106)sCu(12)) for the central cation; 102.2(3)° (Cu(21)sO-
(206)sCu(22)) and 102.9(3)° (Cu(21)sO(217)sCu(22)) for the
terminal cation. The bond angle CusO(oxime)sN(oxime) is
130.3(5)° for central-terminal cations and 108.4(4)° for central-
central cations.
(C) [Cu(L2)2]. This mononuclear compound is a copper(II)
complex with two substituted salicylaldoxime-like anionic
ligands (Figure 7). The Cu(II) ion is located in a square-planar
environment, with two phenolate oxygens and two oxime
nitrogens coordinated in a trans-geometry. The two functional
groups of the ligand that are not bound to the metal ion have
been converted into dimethylacetal groups, and there is no oxime
oxygen bridging. Selected bond lengths and angles are listed
in Table 2.
Ng2â2
3kBT
øm ) (1 - xp)øBB
+
S(S + 1)xp + NR
(
)
Ng2â2
2J
-1
øBB
)
3 + exp -
(1)
(
)
[
]
kBT
kB(T - θ)
using the isotropic Heisenberg exchange Hamiltonian H )
-2JS1S2 for two interacting spins equal to -1/2. øm in the above
equation is expressed as emu/mole Cu, NR is the temperature
independent paramagnetism, xp denotes the contribution due to
monomeric impurity spins, and θ is a Weiss-like correction term
to account for possible magnetic interactions between spin pairs
or dimers.
Magnetic Measurements. For planar Robson-type dicopper
complexes, strong antiferromagnetic interactions have usually
been reported.14,25,28,30a,34,35,38-43,46,47 However, in several cases,
especially for complexes with asymmetric ligands of distorted
geometry, only weak antiferromagnetic interactions were
(46) Mandal, S. K.; Thompson, L. K.; Nag, K. Inorg. Chim. Acta 1988,
149, 247.
(47) Lambert, S. L.; Hendrickson, D. N. Inorg. Chem. 1979, 18, 2683.
(48) Timken, M. D.; Marritt, W. A.; Hendrickson, D. N.; Gagne, R. A.;
Sinn, E. Inorg. Chem. 1985, 24, 4202.
(49) Bleaney, B.; Bowers, K. D. Proc. R. Soc. London A 1952, 214, 451.

516 Inorganic Chemistry, Vol. 36, No. 4, 1997
Rybak-Akimova et al.
ligand unit is stable in the chloride derivative, the initial stage
of the decomposition almost certainly involves the copper(I) in
the anion. Electron transfer between anionic copper(I) and
copper(II) atoms in the cation could account for these observa-
tions. Formally, the spin state would be unchanged by this
transition, but instead of two antiferromagnetically coupled
copper(II) ions in the dicompartmental ligand, two more or less
independent copper(II) ions would exist, one in the cation and
one in the anion. When cooled to room temperature, samples
that have been heated above 400 K show two weak and broad
signals, with g ) 2.34 and g ) 2.15, in their EPR spectra, a
behavior supporting this hypothesis. Further decomposition
probably takes place, because the magnetization curve at high
temperature is not completely reversible.
A comparison of the coupling constants (2J) for our com-
plexes with analogous values for other Robson-type dicopper
complexes show that they fall in the usual range of 2J values
between -600 and -900 cm-1 14,25,28,30a,34,35,38,43,46-47
(The
.
same exchange parameter is generally referred to as 2J, but
sometimes as J; all the numbers in this paper correspond to 2J
in eq 1 and in the Hamiltonian H ) -2JS1S2). For example,
the energy separation between singlet and triplet states in the
chloride derivative of the symmetrical Schiff base dicopper
complex (I in Figure 1, X ) Me, Y ) H, Z ) (CH2)3) has
been reported to be 588,47 722,39 or about 900 cm-1 30b
(estimation from the data in ref 25). Despite the different
estimates, it should be noted that our values for the new
unsymmetrical complex are practically the same as the numbers
for its symmetrical analog. This similarity can be rationalized
on the basis of magnetostructural correlations.
Figure 8. Temperature dependence of magnetic susceptibilities of (a)
[Cu2L]Cl‚2H2O and (b) [Cu2L]2[Cu2Cl4].
A standard least-squares fitting of the data for [Cu2L]Cl
yielded the following values for the five parameters: g ) 2.088
( 0.005, 2J ) -698 ( 10 cm-1, NR ) (59 ( 4) × 10-6 emu/
mol Cu, θ ) 0, xp ) 0. All of the data, up to the highest
measured temperature, were fitted to the above parameters.
However, for the compound [Cu2L]2[Cu2Cl4], only the data up
to 296 K were included for fitting. A fitting analysis gave the
following values for the five parameters: g ) 2.099 ( 0.005,
2J ) -690 ( 10 cm-1, NR ) (60 ( 4) × 10-6 emu/mol Cu,
θ ) (1.2 ( 0.5) × 10-3 K, xp ) 0.0101 ( 0.0005. The curve
drawn to 500 K denotes the øT behavior corresponding to the
parameter values obtained above.
Even though it is evident from the crystal structure of
[Cu2L]2[Cu2Cl4] that two dicopper(II) cations are bound to each
other, forming a dimer of dimers, no significant magnetic
interaction between the dimers can be seen from the magnetic
measurements (the θ-correction included in the Bleaney-
Bowers equation to account for possible intradimer interactions
was found to be equal to 0 for one compound and negligibly
small for another one). It is possible, however, that the
antiferromagnetic interaction within the dimer is too strong for
any further weak interactions to be identified.
The two compounds show different behaviors at high
temperatures. The experimental data for [Cu2L]Cl show a good
fit to the curve calculated from the low-temperature data, with
reversible susceptibility changes and no signs of decomposition.
This demonstrates that the dicopper-organic ligand framework
is thermally stable at least to 450 K. The behavior of the
tetrachlorodicuprate(I) derivative at high temperatures is dif-
ferent. The experimental magnetic susceptibility above 320 K
is significantly higher than that calculated from the low-
temperature parameters (Figure 8b), indicating the appearance
of extra spin components, and the changes at high temperature
are not reversible. The measurements were conducted in Vacuo
(under 0.1 Torr of He gas), so aerial oxidation of copper(I) in
the [Cu2Cl4]2- anion can be excluded. Since the dicopper-
Magnetostructural correlations have been very successful in
the description of the Cu-Cu coupling in binuclear complexes
with flexible noncyclic bridging ligands.3,50 It has been shown
that the degree and even sign of this coupling is primarily
determined by the Cu-O-Cu angle; if this exceeds 97.5°, an
antiferromagnetic interaction is observed; if less, a ferromagnetic
interaction.50 The other significant factors are the degree of
pyramidal distortion at the bridging oxygen atom (the more
planar the oxygen environment, the higher the antiferromagnetic
coupling), the CusCu separation (not very important when a
bridging ligand is involved in antiferromagnetic coupling), and
the degree of coplanarity of the planes formed by each copper
ion, and the two bridging oxygens (the antiferromagnetic
interaction is stronger when the “halves” of the binuclear
complex are coplanar).14,20,30,34,35,39,40,42,43,46 Since the dx -y
2
2
orbitals of copper(II) ions are involved in the magnetic exchange
pathway, it has been suggested that displacement of Cu out of
the plane of the ligand will decrease the coupling con-
stant.34,38,39,46
Although the application of quantitative magnetostructural
correlations to dicopper complexes with rigid macrocyclic
Robson-type ligands is limited, because different structural
features are not independent from each other, the same structural
factors are expected to play a role in magnetic interactions. Their
relative importance, however, might differ from that for acyclic
complexes. For example, a decrease of the CusOsCu bridging
angle from 104.5° to 96.3° leads to only a modest decrease of
the coupling constant (from about -800 to -689 cm-1 40,41
sin
)
contrast with the behavior of nonrigid complexes.50 Moreover,
an attempt to correlate the J values with CusO(Ph)sCu angles
yielded a graph with significantly scattered data points (R2 )
0.52) and with the slope of its least-squares straight line much
smaller than the corresponding values for hydroxide- and
(50) Crawford, V. H.; Richardson, H. W.; Wasson, J. R.; Hodgson, D. J.;
Hatfield, W. E. Inorg. Chem. 1976, 15, 2107.

Cu2 Complexes with a Schiff Base-Oxime Ligand
Inorganic Chemistry, Vol. 36, No. 4, 1997 517
alkoxide-bridged complexes.51 Significant decreases of the
values of exchange constants were observed in cases of a large
(0.58 Å) displacement of copper(II) out of the ligand plane (to
-41 cm-1),43 pronounced ligand twist in a clathrochelate ligand
(-66 cm-1),48 and distortion of the Cu2O2 fragment from
planarity due to displacement of metal ions from the ligand plane
to the same side of the ligand (-124 cm-1).30a
The various structural factors summarized above provide a
foundation for considering the structure of the new dicopper
Schiff base-oxime complex from the point of view of magnetic
exchange coupling between the two copper ions. The dicopper
complex studied here shows neither steric hindrance (which
might have led to a significant twist of the ligand) nor out-of-
range values for its CusOsCu angles (102.3° and 103.2° for
our complex, and 102.3°, 103.6°, or 104.5° for different salts
of its symmetrical analog37,39,40). CusCu separation (3.07 Å
for our complex and from 3.091 to 3.313 for different salts of
a symmetrical Robson complex37,39,40) and the displacement of
the copper ions from their coordination planes (0.16 and 0.19
Å for our complex and from 0.019 to 0.21 Å for different salts
of a Robson complex37,39,40) is also similar for our Schiff base-
oxime complex and its symmetrical analog (X ) CH2, Y ) H,
Z ) Z′ ) (CH2)3, (Figure 1). Possible unfavorable factors
include different displacements of the two kinds of Cu(II) ions
from the planes of their donor atoms in the different compart-
ments of the ligand (here 0.19 Å in the Schiff base compartment
and 0.16 Å in the oxime compartment). This leads to a slight
distortion of the Cu2O2 fragment from planarity. From the data
available it is uncertain if this different displacement of Cu for
different compartments is a feature of the asymmetric ligand
or arises because the two copper ions have different axial ligands
(chloride and oxime oxygen, respectively). The extra negative
charge of the oxime ligand in comparison with the usual
Robson-type ligands should also be considered. Electronic
effects in symmetrical ligands either have little influence on
the magnetic exchange between two copper(II) ions34,35,42,43 or
lead to decreases in the coupling constants in the case of
electron-withdrawing substitutents.51,52 Consequently, the electron-
donating properties of the oxime group can be favorable for
antiferromagnetic exchange. However, the value of the anti-
ferromagnetic interaction constant is dependent on the energy
difference between the magnetic orbitals of the two copper
centers; the largest overlap of magnetic orbitals will occur for
two centers of the same energy,3 with the antiferromagnetic
interaction decreased for two nonidentical sites. From the results
of magnetic measurements it is obvious that these factors are
not significant enough to decrease the magnetic coupling
between two copper(II) ions. It can be concluded that the
asymmetrical Schiff base-oxime ligand (II, Figure 1) provides
an efficient pathway for magnetic interaction between two metal
ions. The antiferromagnetic coupling within this binuclear
cation is practically the same as that in analogous symmetrical
Schiff base ligands.
Figure 9. Cyclic voltammogram of [Cu2L]Cl‚2H2O (10-3 M solution
in DMF, 0.1 M N(Et)4PF6, scan rate 50 mV/s).
∆E ) 60 mV) and the more negative wave quasi-reversible
(E1/2 ) 1.47 V, ∆E ) 100 mV); one irreversible oxidation wave
is also seen (which can be quasi-reversible under certain
conditions) (E1/2 ) +0.04, ∆E ) 90 mV). Two reversible or
quasi-reversible reduction waves are typical of dicopper com-
plexes with Robson-type ligands14,19,20,23,27-32,35,38,40,42,43,46,53-55
and are assigned to the stepwise reduction of the copper(II)
ions: the first wave corresponds to the CuIICuII/CuIICuI couple,
and the second wave, to the CuIICuI/CuICuI couple. The
difference between the half-wave potentials of these two
1
processes (∆E ) E1/2 - E1/22) is a measure of the relative
stability of mixed-valent CuIICuI species:
CuIICuII + CuICuI ) 2CuIICuI, Kcon
ln Kcon ) (nF/RT)∆E
For dicopper complexes with symmetrical ligands, the value of
the conproportionation constant has been suggested as a measure
of the electron delocalization in a mixed-valence complex.23,53
This approach obviously cannot be applied to asymmetrical
complexes with two copper ions in different coordination
environments. However, Kcon still characterizes the stability of
the mixed-valent complex. Despite the expectation that spin
delocalization is less efficient in asymmetric complexes, the
differences in half-wave potentials for the two reductions (and
consequently the Kcon values) are sometimes very high for highly
distorted complexes with ligands possessing significantly dif-
ferent coordination sites (for a ∆E of ∼0.7 V, Kcon ∼ 1012-
1013).19,20,30a
The electrochemical reduction data for the asymmetrical
Schiff base-oxime complex [Cu2L]Cl can be compared with
the data for an analogous symmetrical Schiff base complex
(Figure 1, I, X ) Me, Y ) H, Z ) (CH2)3). For the latter
compound, the half-wave potentials measured in the same
solvent (DMF) were reported to be -1.31 and -0.92 V; ∆E )
0.39 V.23 Both potentials are more negative for the oxime-
containing ligand, which can be attributed to the stabilization
of the higher oxidation state (and destabilization of the lower
oxidation state) by the additional negative charge on the ligand.
This behavior, while consistent with the electrochemistry of
copper oxime complexes,56,57 differs from that reported by
Addison54 who found that potentials are practically the same
Electrochemistry. Cyclic voltammograms were measured
in dimethylformamide (DMF) solution for the complex [Cu2L]-
Cl, as well as for the precursor mononuclear complex [CuL1]
(see Scheme 1). Only one reversible reduction wave with E1/2
) -1.34 V, ∆E ) 80 mV was observed for the mononuclear
compound. For the binuclear Schiff base-oxime complex
[Cu2L]Cl, two separate reduction waves were observed (Figure
9), with the more positive wave reversible (E1/2 ) -1.04 V,
(51) Thompson, L. K.; Mandal, S. K.; Tandon, S. S.; Bridson, J. N.; Park,
M. K. Inorg. Chem. 1996, 35, 3117.
(52) MacLachlan, M. J.; Park, M. K.; Thompson, L. K. Inorg. Chem. 1996,
35, 5492.
(53) Gagne´, R. R.; Spiro, C. L. J. Am. Chem. Soc. 1980, 102, 1443.
(54) Addison, A. W. Inorg. Nucl. Chem. Lett. 1976, 12, 899.
(55) Mandal, S. K.; Adhikary, B.; Nag, K. J. Chem. Soc., Dalton Trans.
1986, 1175.

518 Inorganic Chemistry, Vol. 36, No. 4, 1997
Rybak-Akimova et al.
Table 3. Atomic Coordinates (×104) and Equivalent Isotropic
direction by introducing negatively charged oxime groups into
the ligand, without loss of stability of the mixed-valent CuIICuI
complex.
Displacement Parameters (Å2 × 103) for [Cu2L]2+[Cu2Cl4]2- (A)
x
y
z
U(eq)a
An interesting feature in the electrochemical behavior of
[Cu2L]Cl is the presence of an oxidation wave +0.04 V. This
wave is irreversible when CVA is registered in the range +0.6
to -2.1 V. However, it becomes quasi-reversible for scans from
+0.6 to -0.6 V and then back (scan rate 50, 100, or 200 mV/
s). This one-electron oxidation wave can be tentatively assigned
to CuIICuII-CuIIICuII oxidation, with oxidation of the Cu(II) in
the oxime compartment. The potential of this process is
significantly lower than the oxidation potential of dicopper
complexes with amine dicompartmental ligands,28 reflecting the
ability of oxime ligands to stabilize high oxidation states. Thus,
for Cu(dmgH)2 (dmgH ) dimethylglyoximato) in CH2Cl2 the
following potentials have been reported: +0.38 V (oxidation),
-1.26 V (reduction)57 (potentials recalculated Vs Fc+/Fc). For
these two processes ∆E ) 1.64 V, compared to 1.51 V for the
oxime compartment of our complex, supporting the suggested
assignment of this oxidation wave.
Cu(1)
Cu(2)
Cu(3)
Cl(1)
Cl(2)
O(1)
N(2)
C(3)
C(4)
C(5)
O(6)
C(7)
C(8)
N(9)
C(10)
C(11)
C(12)
N(13)
C(14)
C(15)
C(16)
O(17)
C(18)
C(19)
N(20)
O(21)
C(22)
C(23)
C(24)
C(25)
C(26)
C(27)
C(28)
C(29)
130.4(4)
-1263.5(4)
205.2(8)
-2160.7(12)
1296.2(10)
1441(3)
4246.6(3)
4426.0(3)
1424.7(6)
192.9(12)
3461.4(10)
2503(2)
2357(3)
902(3)
3840.9(2)
2131.7(2)
617.3(3)
-173.7(6)
1509.1(5)
4986.3(12)
4326.9(12)
4084(2)
3403.3(15)
2840.5(14)
2931.4(10)
2179(2)
1551(2)
1454.5(13)
735(2)
330(2)
766(2)
24(1)
26(1)
64(1)
57(1)
43(1)
41(1)
28(1)
32(1)
27(1)
24(1)
26(1)
28(1)
32(1)
30(1)
43(1)
46(1)
59(1)
35(1)
37(1)
29(1)
25(1)
28(1)
28(1)
30(1)
27(1)
33(1)
34(1)
34(1)
47(1)
35(1)
35(1)
34(1)
45(1)
34(1)
302(3)
-575(4)
-1849(3)
-2178(3)
-1390(2)
-3377(3)
-3839(4)
-3226(3)
-3996(4)
-2673(5)
-2077(6)
-1079(3)
-97(4)
1005(4)
1121(3)
198(2)
2233(3)
2492(3)
1784(3)
2112(3)
378(3)
1359(3)
2929(2)
620(3)
1462(3)
2931(3)
3401(4)
4512(4)
6166(4)
6285(3)
7691(3)
8258(3)
7322(3)
5788(2)
8072(3)
7288(3)
5817(3)
5394(2)
-1255(3)
-1979(3)
-3727(3)
-1023(3)
9889(3)
10636(3)
12397(3)
9701(3)
1551.0(14)
1796(2)
2533(2)
3148.1(14)
3077.5(10)
3840(2)
Conclusions
A new strategy has been suggested for building chains of
magnetically coupled metal ions in which binuclear complexes
are designed so that they can be bound to each other by a fifth
metal ion. As building blocks for such a strategy, dicopper(II)
complexes have been synthesized and characterized with an
asymmetric Robson-type ligand, consisting of Schiff base and
oxime compartments. X-ray structure determination shows that
the asymmetry of the dicompartmental ligand does not introduce
significant distortion of the planar structure of binuclear cations,
with Cu-donor atom bond lengths and angles close to those for
analogous symmetrical Robson complexes. As a result, a strong
antiferromagnetic interaction (ca. -700 cm-1) between two
copper(II) centers exists within the binuclear cations. Cyclic
voltammograms of the newly synthesized complexes show one
oxidation and two reduction waves, with relative stabilization
of higher oxidation states of copper. The two consecutive
reduction waves are significantly separated from each other (by
430 mV), making mixed-valent CuIICuI species thermodynami-
cally stable toward disproportionation. As anticipated for this
design, the oxime oxygens show significant tendencies to be
coordinated to other metal ions. In the absence of a fifth metal,
they coordinate axially to the copper ions from other binuclear
cations. This results in the formation of the coordination
tetramers (for the perchlorate salt) or coordination polymers (for
the [Cu2Cl4]2- salt, where the unusual anions provide additional
bridging possibilities). Consequently, the dicopper complexes
with the new unsymmetrical dicompartmental ligand retain the
main advantages of symmetric Robson complexes (strong
magnetic coupling and stabilization of mixed-valent species)
and add the potential to extend the number of metal ions
coordinated with it and coupled to each other. The syntheses
of heterodinuclear complexes with this dicompartmental ligand
and pentanuclear complexes resulted from their interaction with
the fifth metal ion, as well as their cryomagnetic studies, are
currently in progress in our laboratories.
4530(2)
4611.1(12)
5347.8(11)
3310(2)
2678(2)
2625(2)
2116(2)
2622(2)
3298(2)
3386(2)
-2754(4)
-3969(4)
-4928(5)
-4247(4)
1946(4)
3012(4)
3962(4)
3135(4)
3895(2)
a U(eq) is defined as one-third of the trace of the orthogonalized Uij
tensor.
for complexes of ligands with X ) Me, Y ) Me, Z ) (CH2)3
and X ) Me, Y ) Me, Z ) OBF2O-; he concluded that the
negative charge of the ligand does not influence the reduction
potential of the copper complex. However, this result can also
be explained by the shift to more positive redox potentials for
copper complexes with BF2-bridged oxime ligands compared
to simple copper oximes.56
Structural variation in dicopper complexes with Robson-type
ligands causes a variety of effects in their electrochemical
behavior. For example, changes in the Y substituents (I, Figure
1) lead to significant changes in the second (more negative)
redox potential, but not the first potential.27 On the other hand,
reduction of azomethine bonds and variations in the lengths of
the polymethylene chains Z in the reduced ligands lead to
variations in the first, but not the second, potential.30b In our
complex, introduction of negatively charged Z (OsHsO-
instead of (CH2)3) in one compartment of the ligand leads to
negative shifts in the redox potential of the copper in both
compartments. The stability of the mixed-valence species is
quite high (∆E ) 0.43 V, Kcon ) 1.9 × 107), slightly higher
than the stability of the CuIICuI complex in the symmetrical
Schiff base analog (∆E ) 0.39 V, Kcon ) 5.6 × 105).23
Previously reported alterations in Robson-type dicopper
complexes extended the range of redox potentials in the positive
direction,35,42-43 particularly targeting materials with spin
exchange along infinite cation-anion chains.35 Our results show
that the range of redox potentials can be extended in the negative
Experimental Section
Electronic absorption spectra were recorded on a Varian Model 2300
recording spectrophotometer. IR spectra were obtained from KBr disks
using a Perkin-Elmer Model 1600 Fourier transform spectrometer. ESR
spectra were measured using a Bruker ESP 300 E ESR spectrometer,
(56) Gagne´, R. R.; Allison, J. L.; Gall, R. S.; Koval, C. A. J. Am. Chem.
Soc. 1977, 99, 7170.
(57) Chaudhuri, P.; Winter, M.; Della Vedova, B. P. C., et al. Inorg. Chem.
1991, 30, 2148.

Cu2 Complexes with a Schiff Base-Oxime Ligand
Inorganic Chemistry, Vol. 36, No. 4, 1997 519
Table 4. Atomic Coordinates (×104) and Equivalent Isotropic Displacement Parameters (Å2 × 103) for [Cu2L]+[ClO4]-‚MeOH (B)
x
y
z
U(eq)a
x
y
z
U(eq)a
Cu(11)
Cu(12)
O(101)
N(102)
C(103)
C(104)
C(105)
O(106)
C(107)
C(108)
N(109)
C(110)
C(111)
C(112)
N(113)
C(114)
C(115)
C(116)
O(117)
C(118)
C(119)
N(120)
O(121)
C(122)
C(123)
C(124)
C(125)
C(126)
C(127)
C(128)
C(129)
Cu(21)
Cu(22)
O(201)
N(202)
C(203)
C(204)
C(205)
O(206)
C(207)
157.2(13)
1235.8(13)
-1821(7)
-1676(8)
-2733(10)
-2811(11)
-1643(10)
-363(7)
-1893(11)
-814(12)
454(10)
1218(14)
2910(14)
3607(12)
3248(9)
4321(12)
4312(11)
3041(11)
1793(6)
3122(10)
1849(11)
609(9)
9372.7(7)
9568.2(7)
9508(4)
9665(4)
10074(5)
10349(6)
10271(5)
9898(4)
10589(5)
10561(6)
10240(5)
10271(9)
10317(8)
9571(7)
9385(5)
9052(6)
8722(5)
8715(5)
9078(3)
8271(5)
8109(6)
8469(4)
8201(4)
10739(6)
11039(6)
11427(7)
10962(6)
8356(6)
7949(6)
7541(7)
7919(6)
6812.0(7)
4957.1(7)
8497(4)
7731(4)
7677(6)
6943(6)
6166(6)
6032(3)
5528(6)
728.4(5)
2348.0(5)
-513(3)
221(4)
38(1)
39(1)
44(2)
38(2)
39(2)
43(2)
34(2)
39(2)
41(2)
49(3)
47(2)
93(4)
86(4)
67(3)
45(2)
47(2)
41(2)
39(2)
34(1)
36(2)
46(2)
39(2)
55(2)
44(2)
49(3)
66(3)
47(2)
45(2)
47(2)
69(3)
49(2)
38(1)
42(1)
60(2)
40(2)
43(2)
37(2)
42(2)
37(1)
44(2)
C(208)
N(209)
C(210)
C(211)
C(212)
N(213)
C(214)
C(215)
C(216)
O(217)
C(218)
C(219)
N(220)
O(221)
C(222)
C(223)
C(224)
C(225)
C(226)
C(227)
C(228)
C(229)
O(1S)
-1703(11)
-1067(9)
-819(16)
218(19)
188(15)
207(9)
751(11)
979(10)
586(10)
-58(6)
887(10)
4690(7)
4331(5)
3413(7)
3091(8)
3091(6)
3966(4)
4033(6)
4780(5)
5602(6)
5771(3)
6300(6)
7199(6)
7503(4)
8380(4)
7056(6)
6436(7)
6591(7)
5674(6)
4661(6)
5297(6)
5156(7)
6113(6)
5561(4)
5594(8)
6134(5)
6325(8)
6302(2)
7072(5)
6182(17)
6322(19)
5573(9)
5776(23)
6594(26)
5812(27)
7962(2)
7665(6)
7212(8)
8503(7)
8366(10)
4244(5)
3702(4)
3618(7)
3177(7)
2453(5)
2329(4)
1764(5)
1489(5)
1813(5)
2431(3)
1501(5)
1785(5)
2318(4)
2455(3)
4943(5)
5310(5)
5963(5)
5053(5)
854(5)
54(3)
53(2)
93(4)
111(5)
73(3)
45(2)
46(2)
37(2)
37(2)
38(1)
37(2)
43(2)
40(2)
52(2)
48(3)
48(3)
67(3)
48(2)
44(2)
43(2)
60(3)
46(2)
65(2)
83(4)
81(2)
107(5)
64(1)
112(3)
133(14)
215(24)
146(16)
168(22)
182(23)
129(15)
72(1)
137(4)
204(6)
196(6)
251(9)
530(5)
1292(5)
1848(4)
1695(3)
2583(5)
3194(5)
3222(4)
3951(5)
4038(6)
3627(5)
2836(4)
2500(5)
1718(5)
1179(5)
1375(3)
454(5)
-157(5)
-125(3)
-768(3)
1490(5)
2192(6)
2353(5)
2727(5)
1514(5)
792(6)
553(10)
-128(8)
-340(7)
-3501(11)
-3675(11)
-4526(12)
-3036(11)
1625(10)
1921(10)
2680(12)
1539(10)
2051(8)
3377(13)
-3270(8)
-4712(14)
4432(4)
3745(10)
5274(31)
5364(46)
3194(21)
4220(58)
6015(21)
3763(32)
2181(4)
524(5)
-137(5)
862(5)
-437(8)
3773(3)
3426(6)
2045(4)
1986(8)
8181(2)
8373(5)
8800(10)
7678(17)
7910(19)
8628(17)
8210(26)
7466(11)
4807(2)
4059(5)
5011(6)
5081(7)
5006(6)
-4109(10)
-4394(12)
-5861(12)
-3237(12)
5658(11)
5755(11)
7197(11)
4477(11)
-979.0(13)
-493.3(13)
-2175(9)
-2001(8)
-2672(10)
2739(10)
-2061(10)
-1354(6)
-2227(11)
C(1S)
O(2S)
C(2S)
Cl(1)
O(11)
O(12)b
O(13)b
O(14)b
O(12A)c
O(13A)c
O(14A)c
Cl(2)
587(6)
284(5)
2907.7(6)
3010.6(6)
3308(4)
3490(4)
4030(5)
4346(4)
4086(5)
3516(3)
4456(5)
O(21)
O(22)
O(23)
O(24)
1938(13)
2231(16)
3539(12)
977(13)
a U(eq) is defined as one-third of the trace of the orthogonalized Uij tensor. b Occupancy 0.58(3). c Occupancy 0.42(3).
Table 5. Atomic Coordinates (×104) and Equivalent Isotropic
fitted with a Model 179 digital coulometer and a Princeton Applied
Research Model 175 Universal programmer. The scan rate range was
50-200 mV/s, with all measurements performed in 2 × 10-3 M DMF
solutions containing 0.1 M tetrabutylammonium hexafluorophosphate
as a supporting electrolyte. Ferrocene was used as an internal standard,
and all potentials are quoted versus the Fc+/Fc couple.
Room temperature magnetic moments were measured using a
Johnson Matthey magnetic susceptibility balance calibrated with
mercury tetrathiocyanatocobaltate(II). Cryomagnetic studies used a
“force magnetometer” in an external magnetic field of 5 kG, which
has previously been described.58
Displacement Parameters (Å2 × 103) for [Cu(L2)2] (C)
x
y
z
U(eq)a
Cu(1)
O(1)
N(2)
C(3)
C(4)
C(5)
O(6)
C(7)
C(8)
5000
5000
0
42(1)
69(1)
48(1)
48(1)
43(1)
41(1)
50(1)
44(1)
44(1)
57(1)
48(1)
69(2)
50(1)
52(1)
47(1)
69(2)
70(2)
3279(3)
4305(3)
4666(3)
5698(3)
6366(3)
6112(2)
7355(3)
8048(3)
8914(2)
8439(2)
9056(4)
7646(4)
6998(4)
6035(3)
7353(4)
9513(4)
3533(3)
4010(3)
3723(4)
4103(3)
4920(4)
5390(3)
5253(3)
6123(4)
6433(3)
5766(3)
4740(5)
4784(4)
3968(4)
3657(4)
3477(5)
7388(5)
917(4)
1487(5)
3004(6)
3864(5)
3198(5)
1706(4)
4178(5)
3449(5)
4668(4)
1951(4)
2125(7)
5739(6)
6412(6)
5467(5)
8132(6)
4193(6)
All chemicals were of reagent grade. 2,6-Diformyl-4-methylphenol
was synthesized as described in ref 59, and the monocopper complex
with ligand L1 (Scheme 1), according to ref 23.
O(9)
O(10)
C(11)
C(12)
C(13)
C(14)
C(15)
C(16)
Synthesis of [Cu2L]Cl‚2H2O. To 0.125 g of [CuL1] suspended in
10 mL of methanol was added with stirring 0.050 g (1 equiv) of CuCl2‚
2H2O; a clear brown-green solution was formed in 20 min. To this
solution 0.040 g (2 equiv) of NH2OH‚HCl was added, followed by
0.12 mL (3 equiv) of triethylamine. The mixture was stirred at ambient
temperature for 30 min, when a green precipitate appeared. After
stirring overnight, the green precipitate was removed by filtration or
centrifugation. The precipitate was washed with small portions of
methanol, followed by ether, and dried under vacuum; yield, 0.083 g
(48%; 67% for a preparation at 10-fold scale). The compound can be
recrystallized from methanol. MS (positive FAB, methanol-NBA):
519, 521 ([Cu2L]+. IR (medium and strong bands only): 3417 (broad),
a U(eq) is defined as one-third of the trace of the orthogonalized Uij
tensor.
with diphenylpicrylhydrazyl (DPPH) as a standard. Fast atom bom-
bardment (FAB) mass spectra were obtained in matrices of magic bullet
and nitrobenzyl alcohol, with argon as a fast atom beam. Cyclic
voltammetry (CVA) was performed using a three-electrode system
consisting of a glassy carbon working electrode, platinum auxiliary
electrode, and silver reference electrode. Voltammograms were
recorded with a Princeton Applied Research Model 173 potentiostat,
(58) Kahol, P. K.; McCormick, B. J. J. Phys.: Condens. Matter 1991, 3,
7963.
(59) Ullmann, F.; Brittner, K. Berichte 1909, 42 (B.II), 2539.

520 Inorganic Chemistry, Vol. 36, No. 4, 1997
Rybak-Akimova et al.
2923, 1629, 1564, 1479, 1455, 1421, 1386, 1310, 1284, 1235, 1192,
1113, 1096, 1071, 813, 760, 645, 568, 516 cm-1. Anal. Calcd for
C21H25N4O6Cu2Cl: C, 42.61; H, 4.26; N, 9.46; Cu, 21.47. Found: C,
43.17; H, 4.20; N, 9.20; Cu, 21.63.
2922, 2861, 1634, 1595, 1568, 1456, 1419, 1387, 1309, 1272, 1240,
1196, 1090 (very strong), 994, 940, 876, 813, 758, 625, 572, 518 cm-1
.
Crystal Structure Analysis. Crystal data are given in Table 1 for
compounds A-C. Crystals of A were well-formed dark brown plates,
green in transmitted light; for B, small brown blocks; for C, dark brown
plates.
Isolation of [Cu(L2)2]. On allowing the filtrate from the synthesis
of [Cu2L]Cl to stand overnight in an open flask, a brown-green solid
crystallized. Filtration and careful washing with methanol gave [Cu-
(L2)2] as a brown crystalline solid; yield, about 5%. MS (positive FAB,
CHCl3-NBA): 510, 512 (molecular ion); 480; 448. IR (medium and
strong bands only): 3430 (broad), 3198, 2987, 2934, 2825, 1682 (w),
1627, 1558, 1456, 1444, 1370, 1290, 1253 (w), 1228, 1194, 1118, 1047,
(A) Data Collection and Processing. Siemens P3R3 (P4 for A)
four-circle diffractometer, λ ) 0.710 73 Å, ω-2θ mode, with scan
range of (1.0 (2θ) around the KR1 - KR2 angles, scan speed 3-29 deg
min-1, depending upon the intensity of the 2 s prescan; backgrounds
measured at each end of the scan for 0.25 of the scan time. Cell
constants obtained by least-squares refinement on diffractometer angles
for 15 (30 for A) automatically centered reflections. Three standard
reflections were monitored every 200 reflections and showed no
decrease during data collection. Reflections were corrected for
absorption effects by the Gaussian method for A and by the analytical
method using ABSPSI for C;60 no correction for B.
1025, 999, 887, 863, 804, 700, 668, 619, 570, 531 cm-1
.
Synthesis of [Cu2L]2[Cu2Cl4]. A small amount of this compound
crystallizes out of the filtrate and the methanol washings in the synthesis
of the chloride salt. A better yield is achieved by using an excess of
copper chloride and hydroxylamine in the chloride synthesis. After
mixing 1 equiv of [CuL1], 1 equiv of CuCl2‚2H2O, 2 equiv of NH2-
OH‚HCl, and 3 equiv of NEt3 in methanol, the mixture was stirred for
several minutes until the precipitate started to form. At this stage, 1
equiv (0.050 g) of CuCl2‚2H2O was added, followed by 0.5 equiv of
NH2OH‚HCl. After the mixture was stirred for 30 min, a brown-green
precipitate separated from the solution. This was redissolved in 20
mL of methanol, undissolved brown precipitate centrifuged off, and
the solvent evaporated slowly in air. The diamond-shaped crystals of
the tetrachlorodicuprate(I) salt were filtered, washed with a small
amount of ethanol and with ether, and dried under vacuum; yield, 0.042
g (22%). MS (positive FAB, methanol-NBA): 519, 521 ([Cu2L]+).
IR (medium and strong bands only): 3393 (broad), 2923, 1628, 1594,
1567, 1455, 1419, 1304, 1275, 1237, 1192, 1115, 1097, 1068, 1002,
812, 758, 645 cm-1. Anal. Calcd for C42H42N8O8Cu6Cl4: C, 38.51;
H, 3.23; N, 8.55; Cu, 29.11. Found: C, 38.93; H, 3.45; N, 8.71; Cu,
28.44.
(B) Structure Analysis and Refinement. For A, heavy atoms were
located by the Patterson interpretation section of SHELXTL PLUS61
and the light atoms then found by E-map expansion and successive
Fourier syntheses; B and C were solved by direct methods using
SHELXTL PLUS61 (TREF). Hydrogen atoms were added at calculated
positions and refined using a riding model. For A, the oxime-hydrogen
was not located. Anisotropic displacement parameters were used for
all non-H atoms; H atoms were given isotropic displacement parameters
2
U ) 0.08 Å2. The weighting scheme was wcalc ) 1/[σ2(Fo ) + (aP)2
+ bP], where P ) (Fo2 + 2Fc2)/3. Refinement used SHELXTL-93.62
For A, the large difference peak and hole, +2.041 and -1.337 e
Å-3, lie close to Cu(3) and probably represent an unresolved disorder
of the bridging group. For B, three oxygen atoms of one perchlorate
group were disordered (occupancies 0.58(3) and 0.42(3); ClsO and
OsO distances were restrained to approximate equality). For C, the
structure solution revealed a centrosymmetric structure with Cu on an
inversion center. Selected bond lengths and angles in Table 2, and
atomic coordinates are given in Tables 3-5.
Synthesis of [Cu2L](ClO4)‚MeOH. Caution: Perchlorates are
potentially explosiVe. A 0.1 g amount of [Cu2L]Cl‚2H2O was dissolved
in a minimum amount of methanol (about 30 mL). A solution of 0.20
g of NaClO4‚H2O in 5 mL of methanol was added. A dark green
precipitate started to form immediately. The mixture was kept for 1
h; then the microcrystalline precipitate was filtered, washed with small
amounts of methanol and ether, and dried in air; yield, 0.080 g (70%).
X-ray quality crystals were obtained by a similar procedure, but an
equivalent amount of sodium perchlorate was added to the dilute
methanol solution of [Cu2L]Cl‚2H2O. Slow evaporation of the solvent
resulted in crystal formation. MS (positive FAB, methanol-NBA):
519, 521. IR (medium and strong bands only): 3446 (broad), 3013,
Acknowledgment. This material is based upon work sup-
ported by the NSF under EPSCoR Grant OSR-9255223. The
government has certain rights in this material. This work also
received matching support from the state of Kansas. We would
like to thank the reviewers for careful reviewing of our
manuscript.
Supporting Information Available: Tables listing full bond lengths
and angles, thermal parameters, and H-atom coordinates, and complete
information on crystal data, collection, and refinement parameters (27
pages). Ordering information is given on any current masthead page.
(60) Alcock, N. W.; Marks, P. J. J. Appl. Crystallogr. 1993, 27, 200.
(61) Sheldrick, G. M. SHELXTL PLUS user’s manual; Nicolet Instrument
Co.: Madison, WI, 1986.
(62) Sheldrick, G. M. J. Appl. Crystallogr., in press.
IC9601145