Synthesis and characterization of new paramagnetic tetraaryl derivatives of chromium and molybdenum

Article abstract of DOI:10.1016/j.jorganchem.2007.03.028

The low-temperature reaction of [CrCl₃(thf)₃] with LiC₆H₃Cl₂-2,6 yields the organochromium(III) compound [Li(thf)₄][Cr^III(C₆H₃Cl₂-2,6)₄] (1) in 48% yield. The homoleptic, anionic species [Cr^III(C₆H₃Cl₂-2,6)₄]^- is electrochemically related to the neutral one [Cr^IV(C₆H₃Cl₂-2,6)₄] (2) through a reversible one-electron exchange process (E_1/2 = 0.16 V, ΔE_p = 0.09 V, i_pa/i_pc = 1.18). Compound 2 was isolated in 74% yield by chemical oxidation of 1 with [N(C₆H₄Br-4)₃][SbCl₆]. Attempts to prepare the salt [NBu₄][Cr^III(C₆Cl₅)₄] (4) by direct arylation of [CrCl₃(thf)₃] with LiC₆Cl₅ in the presence of [NBu₄]Br gave the organochromium(II) salt [NBu₄]₂[Cr^II(C₆Cl₅)₄] (3) instead, as the result of a reduction process. The salt [NBu₄][Cr^III(C₆Cl₅)₄] (4) was cleanly prepared by comproportionation of 3 and [Cr^IV(C₆Cl₅)₄]. The reaction of [MoCl₄(dme)] with LiC₆Cl₅ in Et₂O solution proceeded with oxidation of the metal center to give the paramagnetic (S = 1/2), five-coordinate salt [Li(thf)₄][Mo^VO(C₆Cl₅)₄] (5). The crystal and molecular structures of 1 and 2 have been established by X-ray diffraction methods. The magnetic properties of 1 and 4 (S = 3/2) as well as those of 2 (S = 1) have been established by EPR spectroscopy as well as by ac and dc magnetization measurements.

Full text of DOI:10.1016/j.jorganchem.2007.03.028

Journal of Organometallic Chemistry 692 (2007) 3236–3247
www.elsevier.com/locate/jorganchem
Synthesis and characterization of new paramagnetic tetraaryl
derivatives of chromium and molybdenum
a
´
´
´
Pablo J. Alonso, Juan Fornies, M Angeles Garcıa-Monforte, Antonio Martın,
*
´
Babil Menjon , Conrado Rillo
Instituto de Ciencia de Materiales de Arago´n, Facultad de Ciencias, Universidad de Zaragoza–C.S.I.C., C/ Pedro Cerbuna 12, E-50009 Zaragoza, Spain
Received 24 December 2006; received in revised form 14 March 2007; accepted 14 March 2007
Available online 24 March 2007
Abstract
The low-temperature reaction of [CrCl3(thf)3] with LiC6H3Cl2-2,6 yields the organochromium(III) compound [Li(thf)4]-
[CrIII(C6H3Cl2-2,6)4] (1) in 48% yield. The homoleptic, anionic species [CrIII(C6H3Cl2-2,6)4]ꢀ is electrochemically related to the neutral
one [CrIV(C6H3Cl2-2,6)4] (2) through a reversible one-electron exchange process (E1/2 = 0.16 V, DEp = 0.09 V, ipa/ipc = 1.18). Compound
2 was isolated in 74% yield by chemical oxidation of 1 with [N(C6H4Br-4)3][SbCl6]. Attempts to prepare the salt [NBu4][CrIII(C6Cl5)4] (4)
by direct arylation of [CrCl3(thf)3] with LiC6Cl5 in the presence of [NBu4]Br gave the organochromium(II) salt [NBu4]2[CrII(C6Cl5)4] (3)
instead, as the result of a reduction process. The salt [NBu4][CrIII(C6Cl5)4] (4) was cleanly prepared by comproportionation of 3 and
[CrIV(C6Cl5)4]. The reaction of [MoCl4(dme)] with LiC6Cl5 in Et2O solution proceeded with oxidation of the metal center to give the
paramagnetic (S = 1/2), five-coordinate salt [Li(thf)4][MoVO(C6Cl5)4] (5). The crystal and molecular structures of 1 and 2 have been
established by X-ray diffraction methods. The magnetic properties of 1 and 4 (S = 3/2) as well as those of 2 (S = 1) have been established
by EPR spectroscopy as well as by ac and dc magnetization measurements.
ꢀ 2007 Elsevier B.V. All rights reserved.
Keywords: Organochromium; Organomolybdenum; Homoleptic compounds; EPR spectroscopy; Cyclic voltammetry; Electronic exchange
1. Introduction
this kind of compounds, EPR spectroscopy is an invaluable
tool as it is able to provide direct information on the elec-
tronic structure of the unpaired electrons of the d3 CrIII
metal ion. Moreover, given that the spin state is, as a rule,
S = 3/2, the relevant EPR parameters would be expected to
be highly sensitive to small changes in the coordination
sphere of the paramagnetic center as has been found in
classic coordination chemistry.
Homoleptic r-organochromium(III) derivatives bearing
no additional ancillary ligands can be considered still today
as rare chemical species [1]. To the best of our knowledge
just a few precedents have been isolated for the following
stoichiometries: [CrIIIR3] (R = CH(SiMe3)2 [2]), [CrIIIR4]ꢀ
(R = C6Cl5 [3], C6H2Me3-2,4,6 [4], C6H4OMe-2 [5]),
[CrIIIR5]2ꢀ (R = Ph [6,7], C6F5 [8]), and [CrIIIR6]3ꢀ
(R = Me [9], Ph [6,10], C6H4Ph–4 [5]). All of these com-
plexes have an open-shell electronic structure [11] and are
far from conforming to the 18-electron (or Effective
Atomic Number) rule. For the correct characterization of
Compound [Li(thf)4][CrIII(C6Cl5)4] was found to exhibit
interesting electrochemical behavior, with the following
three homoleptic species connected by consecutive one-elec-
tron exchange processes: [CrIV(C6Cl5)4]/[CrIII(C6Cl5)4]ꢀ/
[CrII(C6Cl5)4]2ꢀ [3a]. This rich behavior was possible
because of the versatility of the C6Cl5 group as a ligand,
which is able to act as a standard monodentate r-bonded
ligand and also as a poor chelating one by establishing a sec-
ondary bonding interaction with the metal center through
*
Corresponding author. Tel.: +34 976762293; fax: +34 976761187.
´
E-mail address: menjon@unizar.es (B. Menjon).
0022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.03.028

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3237
one of its ortho-Cl substituents. We sought to test whether
the related C6H3Cl2-2,6 group, that also bears ortho-Cl sub-
stituents but has a markedly different electron-withdrawing
character could also exhibit this ability to match the elec-
tronic and stereochemical requirements of chromium in dif-
ferent oxidation states. In this paper we describe the
synthesis of the new pseudo-octahedral anionic species
[CrIII(C6H3Cl2-2,6)4]ꢀ, and compare its electronic and elec-
trochemical properties with those of the known
[CrIII(C6Cl5)4]ꢀ anion. One-electron oxidation of the
[CrIII(C6H3Cl2-2,6)4]ꢀ anion gave the neutral species
[CrIV(C6H3Cl2-2,6)4], which has been structurally charac-
terized. Our attempts to prepare an analogous homoleptic
molybdenum derivative gave the five-coordinate oxo spe-
cies [MoVO(C6Cl5)4]ꢀ instead, as the result of an oxidation
process.
properties (see below). The IR spectrum shows well-
defined, sharp absorptions, including two strong ones at
887 and 1043 cmꢀ1, which can be assigned to the symmetric
and asymmetric vibrations of the C–O–C link within the
[Li(thf)4]+ cation [12]. In the absence of any detailed vibra-
tional study of the 2,6-dichlorophenyl group and deriva-
tives thereof, a reliable assignment of the remaining
absorptions in the IR spectrum of 1 cannot be made. The
MS (FABꢀ) of 1 shows the peaks corresponding to the
parent anion [Cr(C6H3Cl2-2,6)4]ꢀ and to fragments with
the formula [Cr(C6H3Cl2-2,6)4ꢀnCln]ꢀ (n = 1,2), deriving
from the stepwise replacement of chlorinated aryl groups
by Cl atoms. The isotopic distribution observed for all
those peaks is in good agreement with the suggested
formulation.
The redox behavior of 1 has been studied by electro-
chemical methods. The room temperature cyclic voltam-
mogram of 1 in CH2Cl2 solution measured between ꢀ1.6
2. Results and discussion
and +1.6 V (Fig. 1) shows an oxidation peak, (Ep)ox
0.20 V, which is recovered in the returning scan: (Ep)red
=
=
2.1. Synthesis and electrochemical behavior of the
homoleptic polychloroarylchromium compounds 1–4
0.11 V. This couple corresponds to an electrochemically
The low-temperature reaction (between ꢀ78 and 0 ꢁC)
of the halo-complex [CrCl3(thf)3] with LiC6H3Cl2-2,6 in
Et2O, followed by CH2Cl2 extraction and subsequent treat-
ment with thf and Et2O led to the formation of the organo-
chromium(III) compound [Li(thf)4][CrIII(C6H3Cl2-2,6)4]
(1), which was isolated as a violet solid in 48% yield
(Scheme 1). Compound 1 is highly sensitive to oxygen
and moisture and also has a rather limited thermal stabil-
ity, decomposing readily at room temperature both in solu-
tion and in the solid state. With regard to stability and
solubility properties, the behavior of 1 is similar to that
observed for the analogous pentachlorophenyl derivative
[Li(thf)4][CrIII(C6Cl5)4] [3]. Complex 1, however, is appre-
ciably less soluble in thf than the corresponding penta-
chlorophenyl homologous species.
0.06
CrIV/CrIII
0.04
0.02
I
0.00
-0.02
-0.04
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
The 2,6-dichlorophenyl-derivative 1 has been character-
ized by analytical and spectroscopic methods. Character-
ization of 1 also includes a detailed study of its magnetic
E [V]
Fig. 1. The CV of 1 in CH2Cl2 solution.
Cl
R
R
Cl
a + b + c
CrIII
Cl
Cl
[Li(thf)4]
[CrCl3(thf)3]
+ LiC6H3Cl2-2,6
1, violet, 48% yield
d
R
+ [N(C6H4Br-4)3][SbCl6]
R = C6H3Cl2-2,6
CrIV
R
R
R
2, brown, 74% yield
Scheme 1. Experimental procedure to obtain 1 and 2: (a) reaction in Et2O at ꢀ78 ꢁC; (b) extraction in CH2Cl2; (c) addition of thf; (d) in CH2Cl2 at 0 ꢁC.

3238
P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
reversible semisystem with E1/2 = 0.16 V, DEp = 0.09 V
and ipa/ipc = 1.18. By comparing this behavior with that
exhibited by the perchlorinated species [Li(thf)4][-
CrIII(C6Cl5)4] [3a], it is apparent that the potential required
to oxidize the anionic [CrIIIR4]ꢀ homoleptic species to the
corresponding [CrIVR4] neutral ones sharply decreases on
going from R = C6Cl5 (E1/2 = 0.76 V) to R = C6H3Cl2-
2,6 (E1/2 = 0.16 V) (cf. the vanadium system [13]). An elec-
tron exchange process associated with the [CrIIIR4]ꢀ/
[CrIIR4]2ꢀ couple—which was detectable for R = C6Cl5—
was not observed for R = C6H3Cl2-2,6 in the measured
range. Both observations are attributable to the signifi-
cantly smaller electron-withdrawing ability of the
C6H3Cl2-2,6 group in comparison with that associated with
the perchlorinated one.
(Scheme 2). The [NBu4]+ cation has in general a better
behavior in organometallic and coordination chemistry
than the [Li(thf)4]+ cation, the latter being hygroscopic,
prone to undergo dissociation of thf molecules and hence
showing a more pronounced electrophilic character. As
expected, the salt [NBu4][CrIII(C6Cl5)4] (4) shows an
enhanced stability in comparison with the [Li(thf)4]-
[CrIII(C6Cl5)4] salt. These two salts exhibit the same elec-
trochemical behavior. The EPR spectrum of 4 as well as
its magnetic properties will be discussed below.
2.2. Structural characterization of the homoleptic
polychloroarylchromium compounds 1 and 2
The crystal and molecular structures of 1 and 2 were
determined by X-ray diffraction methods. Crystal data
and other details of the structure analyses are presented
in Table 1. Crystals of 1 consist of separate [Li(thf)4]+ cat-
ions and [CrIII(C6H3Cl2-2,6)4]ꢀ anions with no sign of any
kind of covalent interaction between them. The [Li(thf)4]+
cation shows the standard tetrahedral (T–4) arrangement
of thf molecules around the Li+ ion as found in other cases
[14]. The structure of the [CrIII(C6H3Cl2-2,6)4]ꢀ anion is
depicted in Fig. 2 and a selection of interatomic distances
The neutral homoleptic species [CrIV(C6H3Cl2-2,6)4] (2)
was obtained by treatment of 1 with [N(C6H4Br-4)3][SbCl6]
as the result of a one-electron oxidation process (Scheme
1). Complex 2 was isolated as an analytically pure brown
solid in 78% yield. Compared with its perchlorinated
homologous species [CrIV(C6Cl5)4] [3a], complex 2 exhibits
a substantially higher thermal stability. This feature,
together with the appreciably higher solubility of 2 in the
most common organic solvents, enabled us to obtain single
crystals suitable for X-ray diffraction purposes and thus to
establish its molecular structure (see below).
Table 1
The electrochemical behavior of 2 in CH2Cl2 solution is
qualitatively similar to that discussed above for complex 1.
Hence, it can be concluded that 1 and 2 are actually the two
chemical species involved in that electrochemical process.
The salt [NBu4][CrIII(C6Cl5)4] could not be obtained by
direct reaction of [CrCl3(thf)3] with LiC6Cl5 in the presence
of [NBu4]Br. Under these conditions, arylation of the metal
center takes place together with a formal one-electron
reduction process (Scheme 2), eventually yielding [NBu4]2-
[CrII(C6Cl5)4] (3). The targeted organochromate(III) salt
[NBu4][CrIII(C6Cl5)4] (4) was eventually obtained in ca.
70% yield by a clean comproportionation reaction between
the homoleptic organochromium species in the adjacent
oxidation states, [NBu4]2[CrII(C6Cl5)4] (3) and [CrIV(C6Cl5)4]
Summary of crystallographic data and structure refinement for 1 and 2
1
2
Empirical formula
FW
C
931.29
40H44Cl8CrLiO4
C24H12Cl8Cr
635.94
T (K)
150(2)
150(2)
k (pm)
71.073
71.073
Crystal system
Space group
a (pm)
b (pm)
c (pm)
Orthorhombic
P212121
1472.12(17)
1478.06(11)
1955.51(18)
90
4.2550(7)
4
1.454
Monoclinic
P21/c
1053.6(4)
1093.9(3)
2079.5(5)
92.31(2)
2.3947(12)
4
b (ꢁ)
V (nm3)
Z
dc (g cmꢀ3
F(000)
)
1.764
1264
1916
Diffractometer
l (mmꢀ1
h Range (ꢁ)
Enraf Nonius CAD4
R
R
R
)
0.811
1.384
+ [NBu ]Br
4
CrII
[NBu ]
[CrCl (thf) ]
+
LiC Cl
6 5
2.08–24.99
0 6 h 6 17
0 6 k 6 17
ꢀ23 6 l 6 23
6407
2.10–24.98
0 6 h6 12
0 6 k6 12
ꢀ24 6 l 6 24
5967
4 2
3
3
R
a + b
Index range
3, salmon pink, 28.5% yield
R = C Cl
6
5
Number of reflections
collected
Number of unique
reflections
refinement method
Final R indices
[I > 2r(I)]a
c
IV
+ [Cr (C Cl ) ]
6
5 4
5964 (Rint = 0.0448)
4170 (Rint = 0.2009)
Cl
Cl
Cl
R
Full-matrix least-squares on F2
R
Cl
CrIII
Cl
R1 = 0.0543,
R1 = 0.0848,
wR2 = 0.1203
R1 = 0.2645,
wR2 = 0.1599
Cl
Cl
Cl
[NBu ]
4
wR2 = 0.1109
R1 = 0.1094,
wR2 = 0.1303
1.008
Cl
R indices (all data)
Cl
Goodness-of-fit on F2b
0.982
4, violet, 67% yield
P
P
P
P
a
R1 = (jFoj ꢀ jFcj)/ jFoj;
wR2=[ w(Fo2 ꢀ Fc2)2/ w(F2c)2]1/2
;
w = [r2(Fo2) + (g1P)2 + g2P]ꢀ1; P = (1/3) Æ [max{Fo2,0} + 2F2c].
Scheme 2. Experimental procedure to obtain 3 and 4: (a) reaction in Et2O
at ꢀ78 ꢁC; (b) extraction in CH2Cl2; (c) in CH2Cl2 at 0 ꢁC.
P
b
Goodness-of-fit = [ w(F2o ꢀ Fc2)2/(nobs ꢀ nparam)]1/2
.

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3239
Table 2
Selected bond distances and angles and their estimated standard devia-
tions for [Li(thf)4][Cr(C6H3Cl2-2,6)4] (1)
Bond distances (pm)
Cr–C(1)
Cr–C(7)
Cr–C(13)
Cr–C(19)
208.1(7)
215.0(7)
215.7(7)
209.8(7)
287.5(2)
299.2(3)
138.9(9)
138.2(10)
177.2(7)
174.7(7)
138.6(10)
138.9(10)
176.4(8)
176.1(8)
138.6(10)
139.0(10)
176.6(7)
177.4(8)
139.8(10)
140.4(10)
176.2(7)
175.5(7)
Crꢁ ꢁ ꢁCl(1)
Crꢁ ꢁ ꢁCl(3)
C(1)–C(2)
C(1)–C(6)
C(2)–Cl(1)
C(6)–Cl(2)
C(7)–C(8)
C(7)–C(12)
C(8)–Cl(3)
C(12)–Cl(4)
C(13)–C(14)
C(13)–C(18)
C(14)–Cl(5)
C(18)–Cl(6)
C(19)–C(20)
C(19)–C(24)
C(20)–Cl(7)
C(24)–Cl(8)
Fig. 2. Thermal ellipsoid diagram (50% probability) of the anion of 1.
and angles is given in Table 2. The CrIII center is sur-
rounded by four 2,6-dichlorophenyl groups with Cr–C dis-
tances comprised between 208.1(7) and 215.7(7) pm. The
presence of ortho-Cl substituents in the aryl group enables
it to act as a poor chelate ligand by establishing secondary
bonding interactions with the metal atom in two of the four
2,6-dichlorophenyl rings, thus rendering a six-coordinate
CrIII center. The two short Crꢁ ꢁ ꢁClortho contacts [Crꢁ ꢁ ꢁCl(1)
287.5(2) pm and Crꢁ ꢁ ꢁCl(3) 299.2(3) pm] make the corre-
sponding aryl rings swing sidewards as evidenced by the
significantly different Cr–Cipso–Cortho angles within each
ring: Cr–C(1)–C(2) 110.0(5)ꢁ vs. Cr–C(1)–C(6) 137.7(5)ꢁ
and Cr–C(7)–C(8) 111.0(5)ꢁ vs. Cr–C(7)–C(12) 136.7(7)ꢁ.
No elongation of the corresponding Cortho–Clortho bonds
is observed. The C(19)-to-C(24) ring can be considered as
a typically monodentate aryl ligand, since its two Cr–
Cipso–Cortho angles are equal within the experimental error
[Cr–C(19)–C(20) 125.4(5)ꢁ and Cr–C(19)–C(24) 123.3(5)ꢁ]
and the two ortho-Cl substituents are practically equidis-
tant from the metal center (345.0 and 351.1 pm). Although
the C(13)-to-C(18) ring is appreciably swung [Cr–C(13)–
C(14) 129.8(5)ꢁ vs. Cr–C(13)–C(18) 118.7(6)ꢁ], the corre-
sponding ortho-Cl substituents are too far from the metal
center (>329 pm) to be considered as indicative of any
bonding interaction.
The four r-bonded C atoms and the two weakly inter-
acting ortho-Cl atoms define a heavily distorted octahedron
(OC-6) around the CrIII center with a fairly large value of
continuous shape measure, S(OC-6) = 7.65 [15–18]. The
formation of two strained four-membered metallacycles
with small Cipso–Cr–Clortho angles [C(1)–Cr–Cl(1) 63.0(2)ꢁ
and C(7)–Cr–Cl(3) 60.6(2)ꢁ] is the main source for distor-
tion in the coordination polyhedron. Thus, the remaining
angles between cis donor atoms are between 77.2(2)ꢁ and
113.73(7)ꢁ, while the angles between trans standing atoms
are comprised between 143.3(2)ꢁ and 168.6(2)ꢁ. The overall
structure of the [CrIII(C6H3Cl2-2,6)4]ꢀ anion is quite simi-
lar to that observed for the perchlorophenyl homologous
species [CrIII(C6Cl5)4]ꢀ [3]. The metal environment in the
Bond angles (ꢁ)
C(1)–Cr–C(7)
C(1)–Cr–C(13)
C(1)–Cr–C(19)
C(1)–Cr–Cl(1)
108.0(3)
93.9(3)
104.3(3)
63.0(2)
168.6(2)
143.3(3)
95.2(3)
82.9(2)
60.6(2)
107.7(3)
81.5(2)
96.4(2)
C(1)–Cr–Cl(3)
C(7)–Cr–C(13)
C(7)–Cr–C(19)
C(7)–Cr–Cl(1)
C(7)–Cr–Cl(3)
C(13)–Cr–C(19)
C(13)–Cr–Cl(1)
C(13)–Cr–Cl(3)
C(19)–Cr–Cl(1)
C(19)–Cr–Cl(3)
Cl(1)–Cr–Cl(3)
Cr–C(1)–C(2)
165.4(2)
77.2(2)
113.73(7)
110.0(5)
137.7(5)
112.3(7)
115.1(6)
122.9(6)
111.0(5)
136.7(7)
112.0(7)
116.9(6)
119.9(7)
129.8(5)
118.7(6)
111.4(7)
119.0(5)
118.5(6)
125.4(5)
123.3(5)
111.3(6)
120.3(6)
120.3(6)
Cr–C(1)–C(6)
C(2)–C(1)–C(6)
C(1)–C(2)–Cl(1)
C(1)–C(6)–Cl(2)
Cr–C(7)–C(8)
Cr–C(7)–C(12)
C(8)–C(7)–C(12)
C(7)–C(8)–Cl(3)
C(7)–C(12)–Cl(4)
Cr–C(13)–C(14)
Cr–C(13)–C(18)
C(14)–C(13)–C(18)
C(13)–C(14)–Cl(5)
C(13)–C(18)–Cl(6)
Cr–C(19)–C(20)
Cr–C(19)–C(24)
C(20)–C(19)–C(24)
C(19)–C(20)–Cl(7)
C(19)–C(24)–Cl(8)
former is, however, slightly less distorted than in the latter,
for which an even larger value of continuous shape mea-
sure was obtained, S(OC-6) = 8.95 [15–17].

3240
P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
Table 3
Selected bond distances and angles and their estimated standard devia-
tions for [Cr(C6H3Cl2-2,6)4] (2)
Bond distances (pm)
Cr–C(1)
Cr–C(7)
Cr–C(13)
Cr–C(19)
C(1)–C(2)
C(1)–C(6)
C(2)–Cl(1)
C(6)–Cl(2)
C(7)–C(8)
C(7)–C(12)
C(8)–Cl(3)
C(12)–Cl(4)
C(13)–C(14)
C(13)–C(18)
C(14)–Cl(5)
C(18)–Cl(6)
C(19)–C(20)
C(19)–C(24)
C(20)–Cl(7)
C(24)–Cl(8)
204.1(11)
203.2(11)
203.6(11)
205.0(11)
140.6(15)
139.4(15)
174.7(11)
177.0(12)
141.9(14)
137.2(14)
173.8(12)
174.0(11)
138.2(14)
140.5(15)
177.0(12)
173.4(11)
138.4(14)
140.0(15)
177.2(11)
171.7(12)
Fig. 3. Thermal ellipsoid diagram (50% probability) of 2.
The structure of the neutral species [CrIV(C6H3Cl2-2,6)4]
(2) is depicted in Fig. 3, where it can be seen that the CrIV
center forms just four Cr–C r-bonds. The CrIV–C distances
are between 203.2(11) and 205.0(11) pm, well in the range
of the structurally characterized [CrIVR4] related species
(R = CH2CMe3 [19], CH2CMe2Ph [20], Cy [21] and
CPh@CMe2 [22]). These CrIV–C distances in 2 are shorter
than the corresponding CrIII–C distances in 1 as expected
for the different oxidation state of the metal center in each
case. The coordination polyhedron in 2 can be described as
a quite regular tetrahedron (T-4) with a very small value of
continuous shape measure, S(T-4) = 0.06 [15–17]. A closer
look at the structure, however, reveals a clear elongation of
the T-4, as two of the C–CrIV–C angles are smaller than
expected for the ideal geometry (109.5ꢁ), while the remain-
ing four are larger (Table 3). In order to explain this distor-
tion it is not necessary to invoke electronic effects
associated with the d2 electron configuration of the CrIV
center. Similar distortions have also been found in other
homoleptic, neutral [MIV(C6Cl5)4] species with closed-shell
(d0 or d10) electron configurations (M = Ti, Sn [23]) as well
as in related [MIII(C6Cl5)4]ꢀ anions (M = Ti [24], V [25])
with open-shell electron configurations (d1 for TiIII and
d2 for VIII). Such distortions have been attributed to steric
problems in arranging four bulky and highly anisotropic
aryl groups, R, around the metal center in a T-4 shape.
Bond angles (ꢁ)
C(1)–Cr–C(7)
112.3(5)
101.2(4)
115.5(4)
111.7(5)
102.4(5)
114.2(5)
114.1(8)
132.8(9)
112.9(10)
118.7(9)
119.2(9)
114.5(8)
130.6(9)
114.9(10)
118.1(9)
123.4(9)
116.7(9)
128.5(8)
114.6(11)
117.7(9)
122.9(8)
117.2(9)
127.7(8)
115.1(10)
118.0(9)
122.6(9)
C(1)–Cr–C(13)
C(1)–Cr–C(19)
C(7)–Cr–C(13)
C(7)–Cr–C(19)
C(13)–Cr–C(19)
Cr–C(1)–C(2)
Cr–C(1)–C(6)
C(2)–C(1)–C(6)
C(1)–C(2)–Cl(1)
C(1)–C(6)–Cl(2)
Cr–C(7)–C(8)
Cr–C(7)–C(12)
C(8)–C(7)–C(12)
C(7)–C(8)–Cl(3)
C(7)–C(12)–Cl(4)
Cr–C(13)–C(14)
Cr–C(13)–C(18)
C(14)–C(13)–C(18)
C(13)–C(14)–Cl(5)
C(13)–C(18)–Cl(6)
Cr–C(19)–C(20)
Cr–C(19)–C(24)
C(20)–C(19)–C(24)
C(19)–C(20)–Cl(7)
C(19)–C(24)–Cl(8)
0
0
~
2.3. EPR spectra of the homoleptic polychloroarylchromium
compounds
apparent spin S = 1/2 and an orthorhombic effective g
tensor as will be commented on briefly.
The fundamental term for the free CrIII ion (i.e., a d3 ion
under spherical symmetry) is the quartet 4F (S = 3/2).
When located in an octahedral environment, this term
The X-band EPR spectra of polycrystalline samples of 1
and 4 registered at liquid nitrogen temperature (77.3 K) are
shown in Fig. 4 together with that of the related species
[Li(thf)4][CrIII(C6Cl5)4] for comparison purposes. On the
right side of this figure, calculated spectra are shown con-
sidering a spin S0 = 1/2 and the principal g0i values (i = 1,
2, 3) given in Table 4. The observed features can be inter-
preted as due to an S = 3/2 paramagnetic species with
splits into two orbital triplets (4T2 and T1) plus an orbital
4
4
singlet, A2, which becomes the fundamental term under
this symmetry field. To describe the magnetic properties
of our CrIII systems through their corresponding spin
Hamiltonian, the low-field limit approximation has proven
to be particularly suitable [3]. In this case, the zero field

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3241
Fig. 4. Liquid nitrogen experimental (left) and simulated (right) X-band EPR spectra of powder samples of the following compounds:
[Li(thf)4][CrIII(C6Cl5)4] (a), [Li(thf)4][CrIII(C6H3Cl2-2,6)4] (b), and [NBu4][CrIII(C6Cl5)4] (c). Every spectrum was measured up to 1.5 T but no additional
signals were detected above the magnetic field indicated in the trace. The starred signal at g ꢂ 2.00 is assigned to some minoritary radical species.
ciated with the (S = 3/2)-system as well as of the ortho-
rhombicity parameter g. In the low-field approximation
and according to the selection rule jDmsj = 1, the spectrum
Table 4
Relevant electronic parameters derived from the analyses of the EPR
spectra of polycrystalline samples of the homoleptic polychloroaryl CrIII
(d3) derivatives registered at 77.3 K
should contain mainly transitions within the {W 1/2} dou-
Compound
g10
g20
g30
g
g
e
~
blet. Assuming that the principal axes of the g and D ten-
[Li(thf)4][CrIII(C6Cl5)4]a
4.377 3.456 1.943 1.968 0.079
sors coincide, the relationship between the apparent gi0
values within the {W 1/2} doublet and the true gi values,
are given by
[Li(thf)4][CrIII(C6H3Cl2-2,6)4] (1) 4.385 3.550 1.960 1.991 0.070
[NBu4][CrIII(C6Cl5)4] (4)
4.200 3.725 1.960 1.982 0.040
a
See Ref. [3]. The effective principal g-values obtained in frozen CH2Cl2
solutions are as follows: gx0 ¼ 4:02, gy0 ¼ 3:55 and g0z ¼ 1:946, which give a
value for the orthorhombicity parameter g = 0.08.
g0x ¼ gx 1 þ
!
1 ꢀ 3g
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
;
1 þ 3g2
!
1 þ 3g
g0y ¼ gy 1 þ
;
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
splitting (ZFS) term within the spin Hamiltonian,
1 þ 3g2
e
HZFS ¼ S ꢁ D ꢁ S, is the dominant one, while the Zeeman
!
~
term, HZe ¼ lBB ꢁ g ꢁ S, can be treated merely as a pertur-
2
g0z ¼ gz
ꢀ 1 :
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
bation thereof. In this kind of approximation, HZFS is
taken as a 0th-order term and can be written with respect
to its principal axes (X,Y,Z) as
1 þ 3g2
It is reasonable to assume that the introduction of geomet-
ric distortions in the octahedral environment of the CrIII
paramagnetic center will induce additional splitting in the
excited states only, as the fundamental one is an orbital sin-
glet. If the low symmetry effects and hence the anisotropy
ꢀ
ꢁ
1
HZFS ¼ S ꢁ D ꢁ S ¼ D SZ ꢀ SðS þ 1Þ þ gðS2X ꢀ SY2 Þ : ð1Þ
2
e
3
This means that, even in the absence of any applied mag-
~
netic field, the four spin states associated with the funda-
induced in the true g tensor are not very large, we can still
4
mental term A2 split into two Kramers doublets {W
}
make use of an isotropic g factor (i.e. gx = gy = gz = g).
Thus, a set of g0i values could in principle be derived from
any given couple of g and g values taking g01 ¼ gy0 , g20 ¼ gx0
and g03 ¼ g0z.
1/2
and {W 3/2}, separated by an energy difference
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
DE ¼ 2D 1 þ 3g2. In the low-field approximation, the
DE value is significantly larger than the energy resulting
from the electronic Zeeman interaction and hence, the
two Kramers doublets can be considered as completely
independent systems with regard to the EPR spectroscopy.
Following this reasoning, the estimated values of g and g
yielding the set of g0i values best fitting the experimental
spectra are given in Table 4. The values thus derived for
the isotropic g factor are close to that of the free electron
[ge = 2.0037(2)], as is generally observed in other CrIII spe-
cies. The value of the orthorhombicity parameter g for 1 is
As a result, each of0 these Kramers doublets can be assigned
0
~
an apparent spin S = 1/2 and is described by an effective g
~
tensor, which is itself a function of the true g tensor asso-

3242
P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
similar to that obtained for the homologous pentachlor-
ophenyl derivative [Li(thf)4][CrIII(C6Cl5)4]: 0.070 vs. 0.079
[3a]. This observation is in keeping with the similar geomet-
ric environment found for the CrIII center in the two
[Li(thf)4][CrIIIR4] salts (R = C6H3Cl2-2,6 and C6Cl5). A
smaller value of g was observed for the salt [NBu4][-
CrIII(C6Cl5)4] (4; Table 4). Pronounced effects in the EPR
parameters had already been reported to occur in systems
with S > 1/2 (as is the case of the CrIII ion) caused by small
changes in the surroundings of the paramagnetic center,
which include lattice effects derived from the nature of
the counterion [26]. Also the temperature is known to
induce modifications in the ZFS contribution through the
spin coupling to the lattice vibrations [27]. In fact, the
EPR spectra of 1 and 4 are observed to broaden on rising
the temperature—the two low-field features becoming
unresolved. Careful simulation of these spectra is still pos-
sible, allowing us to infer a decrease of g as the temperature
increases. However, the specific value of D cannot be deter-
mined from these EPR spectra.
Fig. 6. Dependence of the dc magnetization, M/NAlB, vs. the reduced
applied magnetic field (lBB/kBT) of 1 (solid symbols) and 2 (open
symbols) measured at 1.8 K (triangles) and 5 K (squares).
figures for the sake of clarity. The fact that the plots at dif-
ferent temperatures given in Fig. 6 do not fit a universal
curve can be interpreted as a further manifestation of the
ZFS effect already invoked to explain the EPR spectra of
the CrIII compounds. Accordingly, a simple Curie law can-
not be used to describe the thermal evolution of the in-
phase ac magnetic susceptibility of these compounds,
v(T), as already noted for the related compounds
[Li(thf)4][CrIII(C6Cl5)4] and [CrIV(C6Cl5)4] [3a]. As in those
cases, a polynomial expansion of v(T) in powers of Tꢀ1 has
been used for analyzing the data shown in Fig. 7:
The room temperature EPR spectrum of 2 is shown in
Fig. 5. No significant modifications are observed when
the temperature goes down to 10 K. The spectrum is simi-
lar to but still broader than that observed for the homolo-
gous species [CrIV(C6Cl5)4] [3a]. Since, in the case of non-
Kramers systems, this type of spectra is poorly informative,
the magnetic characterization of 2 will rely mainly on mag-
netism measurements (see below).
2.4. Magnetism measurements
ꢂ
ꢃ
C
T
k1 k2
T
vðTÞ ¼ v0 þ
1 þ
þ
þ ꢁ ꢁ ꢁ ;
ð2Þ
The spin states of the 2,6-dichlorophenyl compounds 1
and 2 have been established by isothermal dc magnetiza-
tion measurements, M(H), at low temperatures (Fig. 6).
The saturation tendency at 1.8 K indicate spin states
S = 3/2 and S = 1 for the CrIII (1; d3) and CrIV (2; d2) spe-
cies, respectively. The magnetic behavior of 4 is practically
superimposable to that of 1 and hence, it is not given in the
T2
where v0 accounts for the temperature independent contri-
bution to the magnetic susceptibility, vTIP—including the
diamagnetic contribution—, C is the Curie constant for a
simple paramagnet, and kj (j = 1,2,. . .) are constants con-
taining information on the ZFS contribution. Dashed lines
Fig. 7. AC magnetic susceptibility of 1 (filled circles) and 2 (empty circles)
as a function of the inverse of temperature. The dashed lines represent the
fit of Eq. (2) truncated in the Tꢀ3 term (see text).
Fig. 5. Room temperature X-band EPR spectra of powder samples of
[CrIV(C6Cl5)4] (a) and [CrIV(C6H3Cl2-2,6)4] (b).

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3243
in Fig. 7 correspond to a fit of the experimental data of
v(T) vs. Tꢀ1 (filled or empty circles) to this polynomial
expansion truncated in the Tꢀ3 term. The values thus ob-
tained for C and for the high-temperature magnetic mo-
In the case of a non-Kramers S = 1 entity (CrIV), and
e
~
assuming again coaxiality of the g and D tensors, it is also
possible to derive a suitable expression relating l(T) with
the ZFS parameters D and g. Since l(T) was found to be
practically insensitive to variations of g, it is convenient
to consider an axial case (g = 0). Under all these assump-
tions, and neglecting any anisotropic contribution to the
ment per molecule, l , are collected in Table 5 together
1
with the estimated g values.
Plots of l(T) vs. T for compounds 1 and 2 are given in
Fig. 8. In the absence of ZFS contribution, l should be
temperature independent. However, significant departures
from this ‘‘horizontal’’ behavior are observed for both
compounds at low temperatures. In the case of an S = 3/2
~
g tensor, the following expression is obtained [28]:
ꢀ
ꢁ
1=2
4dꢀ1ð1 ꢀ eꢀdÞ þ 2eꢀd
lðTÞ ¼ g
;
ð4Þ
2eꢀd þ 1
~
entity, and assuming that the principal axes of the g and
where d = D/kBT. A fit of this expression to the experimen-
tal data of 2 gave g = 1.912(2) and D = ꢀ139(2) GHz as
the values yielding the best agreement (solid line in the low-
er trace of Fig. 8). A similar analysis of the previously re-
ported [CrIV(C6Cl5)4] data [3a] gave the following
optimized values: g = 1.850(1) and D = ꢀ 46.7(8) GHz.
e
D tensors coincide, the following expression can easily be
derived:
ꢀ
ꢂ
ꢃ
ꢂ
ꢃ
ꢂ
ꢃꢁ
1=2
2kBT
DE
DE
2kBT
DE
lðTÞ ¼ lB C1 þ C2
tanh
þ C3 tanh
;
2kBT
ð3Þ
where DE = 2D(1 + 3g2)1/2 as above. The solid line in
Fig. 8 (upper trace) represents the best fit of this expression
to the 1 data, obtained for DE = 191(31) GHz. A similar
analysis of the data corresponding to 4 and [Li(thf)4]
[CrIII(C6Cl5)4] (not shown in the Figure), yielded the values
DE = 142(10) GHz and DE = 75(4) GHz, respectively. It is
interesting to note that in all these cases, the splitting be-
tween the {W 1/2} and {W 3/2} doublets is higher than
the microwave frequency, in agreement with our interpre-
tation of the EPR data (see above).
2.5. Synthesis and characterization of
[Li(thf)4][MoVO(C6Cl5)4] (5)
We have tried to prepare homoleptic haloaryl-deriva-
tives of molybdenum. However, all our attempts using dif-
ferent Mo starting products and arylating conditions have
failed so far: either unidentified reaction products were
obtained or unreacted starting materials were recovered
from the reaction medium. Only by reaction of
[MoCl4(dme)] with LiC6Cl5 in Et2O was a defined product
obtained, which was identified as [Li(thf)4][MoVO(C6Cl5)4]
(5). Compound 5 results from a combined process entailing
both arylation and oxidation of the metal center (Eq. (5)).
Table 5
Parameters derived from magnetization measurements
pffiffiffiffiffiffiffiffi
þthf
Compound
C
l
S
hg2i gEPR
½MoCl4ðdmeÞꢃ þ LiC6Cl5 ! ½LiðthfÞ4ꢃ½MoVOðC6Cl5Þ4ꢃ
1
[emu molꢀ1 K] [lB]
5; garnet; 40% yield
[Li(thf)4][CrIII(C6H3Cl2-
2,6)4] (1)
[CrIV(C6H3Cl2-2,6)4] (2)
[NBu4][CrIII(C6Cl5)4] (4)
1.85(1)
3.85
3/2 1.986 1.991
ð5Þ
0.972(4)
1.82(2)
2.89
3.91
1
1.972
3/2 1.970 1.982
–
Complex 5 was identified by analytical and spectroscopic
methods. The MS (FABꢀ) shows a peak corresponding
to the parent species [MoVO(C6Cl5)4]ꢀ (m/z 1102) together
with various anionic fragments derived thereof, all of them
showing the appropriate isotopic pattern for the suggested
stoichiometry. An absorption in the IR spectrum of 5
attributable to the m(Mo@O) mode could not be unambig-
uously assigned due to the rich vibrational pattern of both
the C6Cl5 groups and the thf molecule. Information about
the geometry of the [MoVO(C6Cl5)4]ꢀ anion was obtained
from the combination of X-ray diffraction data and EPR
spectroscopic results.
Poor-quality single crystals of [Li(C8H16O4)2][MoVO-
(C6Cl5)4] (50; C8H16O4 = 12-crown-4 ether) were obtained,
but the X-ray diffraction data collected were not of suffi-
cient quality to carry out a complete structural determina-
tion of the complex. Nevertheless, its general shape and the
connectivity could be established [29]. The geometry of the
[MoVO(C6Cl5)4]ꢀ anion, depicted in Fig. 9, can be
described as square-pyramidal (SPY-5) with the C6Cl5
groups hellicoidally arranged in the basal plane and the
Fig. 8. Effective magnetic moment per molecule in the low-field region
(linear region) as a function of temperature for 1 (solid symbols) and 2
(open symbols). Solid lines represent the evolution predicted according to
Eqs. (3) and (4) in each case (see text).

3244
P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
oxygen atom located in the apex of the pyramid. This coor-
dination environment is similar to that found for the
related [MoVOX4]ꢀ anions (X = Cl, Br, I) [30]. The pro-
tecting effect of the bulky C6Cl5 groups can reasonably
account for the stability of this base-free, five-coordinate
species, which shows no tendency to coordinate a further
been observed in highly ionic species such as the
[MoVOF4]ꢀ anion and other oxofluoromolybdenum deriv-
atives [31]. Our experimental results for complex 5, how-
ever, show the opposite trend, i.e., g^ < gi. This
‘‘inverted’’ situation had already been found to occur in
several oxomolybdenum derivatives including the closely
related [MoVOCl4]ꢀ anion [30] as well as the series of six-
coordinate anions [MoVOX5]2ꢀ (X = Cl, Br, I). The differ-
ent approaches taken to explain this ‘‘inverted’’ behavior
point to metal-ligand covalency as the main factor causing
it [32]. The C6Cl5 group is considered to have an electron-
withdrawing effect, that is weaker than that assigned to the
Cl atom [33]. Given this relationship, the EPR spectro-
scopic properties of 5 can also be rationalized in terms of
the Mo–C6Cl5 bond covalency.
ligand rendering
derivative.
a
more saturated, six-coordinate
The EPR spectrum of a polycrystalline sample of 5
(Fig. 10) shows no temperature dependence within the
120–300 K range. It can be assigned to a S = 1/2 entity
~
with an axial g tensor with the following principal values:
gi = 1.996(1) and g^ = 1.969(1). The calculated spectrum
using these values and a lorentzian lineshape with peak-
to-peak
separations
of
DHpp(i) = 1.7 mT
and
DHpp(^) = 0.7 mT is also shown in Fig. 10.
Based on a raw crystal-field model, the ground-state
orbital for a SPY-5 species such as the [MoVO(C6Cl5)4]ꢀ
anion (ideal C4v symmetry), should be expected to be the
dxy one with the following relative values for the corre-
3. Concluding remarks
We have demonstrated that the 2,6-dichlorophenyl
group is also able to establish secondary bonding interac-
tions with the metal center through one of its ortho-Cl sub-
stituents, thus acting as a poor chelating ligand, C6H3Cl2-
2,6-jC,jCl2. The presence of just two highly electronega-
tive Cl substituents in the phenyl ring makes it markedly
less electron-withdrawing in character than the analogous
perchlorinated group C6Cl5. This difference is evidenced
by the potential, E1/2, of the one-electron exchange process
operating in the [CrIVR4]/[CrIIIR4]ꢀ couple, which drops
from 0.76 V when R = C6Cl5 to 0.16 V when
R = C6H3Cl2-2,6. Given also the different solubility of
the metal complexes containing these two polychlorophe-
nyl groups under consideration, they can be used to induce
the desired global properties in the final compound.
~
sponding g tensor: gi < g^ < ge. This behavior has indeed
According to the similar geometries observed for the
[CrIIIR4]ꢀ anions (pseudo OC-6), the d3 Kramers salts
[Li(thf)4][CrIIIR4] have been found to behave as S = 3/2
species with similar magnetic properties—both macro-
scopic (bulk magnetization measurements) and micro-
scopic (EPR). The T-4 derivative [CrIV(C6H3Cl2-2,6)4] (2)
has a magnetism corresponding to an S = 1 species.
Fig. 9. Schematic drawing of the anion of 5.
Finally, a SPY-5 geometry has been established for the
d1 (S = 1/2) anion [MoVO(C6Cl5)4]ꢀ by the combined anal-
ysis of EPR spectroscopic and X-ray diffractrometric data.
4. Experimental
All reactions and manipulations were carried out under
purified argon using Schlenk techniques. Solvents were
dried by standard methods and distilled prior to use. The
metal complexes [CrCl3(thf)3] [34], [CrIV(C6Cl5)4] [3a],
and [MoCl4(dme)] [35] as well as the organolithium
reagents LiC6H3Cl2-2,6 [13b] and LiC6Cl5 [36] were pre-
pared as described elsewhere. The aminium salt
[N(C6H4Br-4)3][SbCl6] was purchased (Aldrich) and used
as received. Elemental analyses were carried out with a Per-
kin–Elmer 2400-Series II microanalyzer. IR spectra of KBr
discs were recorded on a Perkin–Elmer 883 spectropho-
Fig. 10. Experimental (solid line) and simulated (dashed line) X-band
EPR spectrum of a powder sample of 5 at 120 K.

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3245
tometer in the range 4000–200 cmꢀ1. Mass spectra were
recorded on a VG-Autospec spectrometer using the stan-
dard Cs-ion FAB (acceleration voltage: 35 kV).
15 cm3 caused the precipitation of a salmon pink solid (3)
which was filtered and vacuum dried. Slow diffusion of a
Et2O (40 cm3) layer onto the mother liquor at ꢀ30 ꢁC gave
a second fraction of 3, which was thus isolated in 28.5%
overall yield (1.00 g, 0.65 mmol). Anal. Found: C 43.0, H
4.2, N 1.7; C56H72Cl20CrN2 requires: C 43.8, H 4.7, N
1.8%. IR (KBr, cmꢀ1): ~mmax 1306 (s), 1276 (vs), 1227 (m),
1041 (m), 880 (w; [NBu4]+), 808 (s; C6Cl5: X-sensitive
vibration) [37], 739 (w; [NBu4]+), and 658 (s). MS (FABꢀ):
m/z 1040 [Cr(C6Cl5)4]ꢀ, 828 [Cr(C6Cl5)3Cl]ꢀ, 616
4.1. Synthesis of [Li(thf)4][CrIII(C6H3Cl2-2,6)4] (1)
[CrCl3(thf)3] (1.04 g, 2.77 mmol) suspended in Et2O
(20 cm3) at ꢀ78 ꢁC was stepwise added via cannula to a
suspension of LiC6H3Cl2-2,6 (ca. 16.6 mmol) in Et2O
(50 cm3) under stirring. The temperature of the reaction
mixture was allowed to rise to 0 ꢁC and then maintained
for 3 h in an ice bath. At this point a violet solid had
formed, which was filtered, washed with Et2O (3 · 5 cm3)
and extracted in CH2Cl2 (70 cm3) at 0 ꢁC. The extract
was concentrated to dryness and the resulting residue was
treated with thf (20 cm3). The violet solid formed (1) was
filtered, washed with Et2O (5 cm3) and vacuum dried. Slow
diffusion of a Et2O (40 cm3) layer onto the mother liquor at
ꢀ30 ꢁC gave a second fraction of 1, which was thus isolated
in 48% overall yield (1.25 g, 1.34 mmol). Anal. Found: C
50.2, H 4.75; C40H44Cl8CrLiO4 requires: C 51.6, H 4.8%.
[Cr(C6Cl5)2Cl2]ꢀ,
[Cr(C6Cl5)2]ꢀ.
581
[Cr(C6Cl5)2Cl]ꢀ,
and
546
4.4. Synthesis of [NBu4][CrIII(C6Cl5)4] (4)
To a solution of 3 (0.49 g, 0.32 mmol) in CH2Cl2
(40 cm3) at 0 ꢁC, was added solid [CrIV(C6Cl5)4] (0.34 g,
0.32 mmol). After 90 min stirring, the reaction mixture
was concentrated to ca. 10 cm3. Addition of n-hexane
caused the precipitation of a violet solid, which was filtered
and vacuum dried (4: 0.55 g, 0.43 mmol, 67.2% yield).
Anal. Found: C 36.2, H 2.7, N 1.0; C40H36Cl20CrN
requires: C 37.2, H 2.8, N 1.1%. IR (KBr, cmꢀ1): ~mmax
1320 (s), 1309 (s), 1281 (vs), 1219 (w), 1058 (m), 881 (w;
[NBu4]+), 822 (s; C6Cl5: X-sensitive vibration) [37], 739
(w; [NBu4]+), and 669 (s). MS (FABꢀ): m/z 1040
[Cr(C6Cl5)4]ꢀ, 828 [Cr(C6Cl5)3Cl]ꢀ, 793 [Cr(C6Cl5)3]ꢀ, 616
IR (KBr, cmꢀ1): mmax 1577 (w), 1554 (m), 1539 (s), 1460
~
(m), 1399 (vs), 1247 (m), 1232 (sh), 1164 (m), 1118 (s),
1043 (vs; C–O–Casym) [12], 918 (sh), 887 (s; C–O–Csym
)
[12], 768 (vs), 751 (vs), 692 (m) and 683 (m). MS (FABꢀ):
m/z 632 [Cr(C6H3Cl2-2,6)4]ꢀ, 522 [Cr(C6H3Cl2-2,6)3Cl]ꢀ,
487 [Cr(C6H3Cl2-2,6)3]ꢀ, 412 [Cr(C6H3Cl2-2,6)2Cl2]ꢀ, and
377 [Cr(C6H3Cl2-2,6)2Cl]ꢀ.
[Cr(C6Cl5)2Cl2]ꢀ,
[Cr(C6Cl5)2]ꢀ.
581
[Cr(C6Cl5)2Cl]ꢀ,
and
546
4.2. Synthesis of [CrIV(C6H3Cl2-2,6)4] (2)
4.5. Synthesis of [Li(thf)4][MoVO(C6Cl5)4] (5)
The addition of [N(C6H4Br-4)3][SbCl6] (0.47 g,
0.57 mmol) to a solution of 1 (0.53 g, 0.57 mmol) in
CH2Cl2 (15 cm3) at 0 ꢁC caused the immediate precipita-
tion of a brown solid. After 30 min of further stirring in
an ice bath, the solid was filtered, washed with CH2Cl2
(3 · 5 cm3), and vacuum dried (2: 0.27 g, 0.43 mmol, 74%
yield). Anal. Found: C 45.0, H 1.9; C24H12Cl8Cr requires:
C 45.3, H 1.9%. IR (KBr, cmꢀ1): ~mmax 1552 (w), 1542 (m),
1462 (w), 1404 (vs), 1245 (w), 1172 (m), 1137 (m), 1100
(w), 784 (m), 772 (vs), 765 (s), and 686 (w). MS (FABꢀ):
m/z 632 [Cr(C6H3Cl2-2,6)4]ꢀ, 522 [Cr(C6H3Cl2-2,6)3Cl]ꢀ,
487 [Cr(C6H3Cl2-2,6)3]ꢀ, 412 [Cr(C6H3Cl2-2,6)2Cl2]ꢀ, and
377 [Cr(C6H3Cl2-2,6)2Cl]ꢀ.
[MoCl4(dme)] (1.70 g, 4.45 mmol) suspended in Et2O
(10 cm3) at ꢀ78 ꢁC was stepwise added via cannula to a
solution of LiC6Cl5 (ca. 22.3 mmol) in Et2O (80 cm3) under
stirring. The suspension was allowed to warm up to 0 ꢁC
and after about 4 h of stirring, the by then red solid was fil-
tered, washed with Et2O (3 · 5 cm3), and extracted in
CH2Cl2 (50 cm3) still at 0 ꢁC. The extract was concentrated
to dryness and the resulting residue was redissolved in thf
(5 cm3). The slow diffusion at ꢀ30 ꢁC of a Et2O layer
(40 cm3) into the preceding solution yielded 5 as a garnet
solid (2.50 g, 1.78 mmol, 40% yield). Anal. Found: C
34.1, H 2.5; C40H32Cl20LiMoO5 requires: C 34.2, H 2.3%.
IR (KBr, cmꢀ1): mmax 2979 (m), 2883 (m), 1461 (w), 1314
~
4.3. Synthesis of [NBu4]2[CrII(C6Cl5)4] (3)
(s), 1285 (vs), 1273 (vs), 1196 (w), 1133 (w), 1121 (m),
1066 (m), 1044 (s; C–O–Casym) [12], 915 (sh), 889 (s; C–
O–Csym) [12], 829 (m; C6Cl5: X-sensitive vibration) [37],
and 674 (s). MS (FABꢀ): m/z 1102 [Mo(C6Cl5)4O]ꢀ, 890
[Mo(C6Cl5)3ClO]ꢀ, 608 [Mo(C6Cl5)2O]ꢀ, and 396
[Mo(C6Cl5)ClO]ꢀ.
[CrCl3(thf)3] (0.86 g, 2.29 mmol) suspended in Et2O
(15 cm3) at ꢀ78 ꢁC was stepwise added via cannula to a
solution of LiC6Cl5 (ca. 11.5 mmol) in Et2O (50 cm3) under
stirring. The temperature of the reaction mixture was
allowed to rise and by ca. ꢀ35 ꢁC, solid NBu4Br (0.74 g,
2.29 mmol) was added. The whole mixture was allowed
to reach 0 ꢁC and then it was stirred for further 3 h in an
ice bath. The salmon pink solid in suspension was filtered,
washed with Et2O (3 · 5 cm3) and extracted in CH2Cl2
(60 cm3) at 0 ꢁC. Concentration of the extract to ca.
4.6. Synthesis of [Li(C8H16O4)2][MoVO(C6Cl5)4] (50)
By using the procedure described for the synthesis of 5,
compound 50 was prepared starting from [MoCl4(dme)]
(1.00 g, 3.05 mmol) and LiC6Cl5 (ca. 15.25 mmol). The

3246
P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
CH2Cl2 extract was concentrated to ca. 10 cm3 and a few
drops of 12-crown-4 ether (C8H16O4) were added. The slow
diffusion of a Et2O layer (25 cm3) into that solution yielded
50 as a garnet solid (2.51 g, 1.71 mmol, 56% yield). Anal.
Found: C 32.5, H 2.45; C40H32Cl20LiMoO9 requires: C
4.9. Magnetic measurements
Magnetic measurements of polycrystalline samples were
carried out using a Quantum Design SQUID-Based Mag-
netometer MPMS-XL5. The magnetometer was calibrated
using standard palladium and dysprosium oxide reference
samples supplied by Quantum Design. The accuracy of
the measurements is better than 1%. Magnetic susceptibil-
ity measurements were performed from 1.8 to 265 K at
10 Hz and 4.1 Oe amplitude.
32.7, H 2.2%. IR (KBr, cmꢀ1): mmax 2915 (m), 2872 (m),
~
1447 (w), 1365 (w), 1315 (s), 1285 (vs), 1273 (vs), 1245
(w), 1196 (w), 1135 (vs), 1098 (vs), 1066 (m), 1025 (s),
926 (m), 856 (m), 827 (m; C6Cl5: X-sensitive vibration)
[37] and 674 (s). MS (FABꢀ): m/z 1102 [Mo(C6Cl5)4O]ꢀ,
1067 [Mo(C6Cl5)3(C6Cl4)O]ꢀ, 608 [Mo(C6Cl5)2O]ꢀ, and
396 [Mo(C6Cl5)ClO]ꢀ.
4.10. Electrochemistry
4.7. X-ray crystallography
Electrochemical studies were carried out using an
EG&G model 273 potentiostat in conjunction with a
three-electrode cell, in which the working electrode was a
platinum disc, the auxiliary electrode a platinum wire and
the reference was an aqueous saturated calomel electrode
(SCE) separated from the test solution by a fine-porosity
frit and an agar bridge saturated with KCl. Where possible,
solutions were 5 · 10ꢀ4 mol dmꢀ3 in the test compound
and 0.1 mol dmꢀ3 in [NBu4][PF6] as the supporting electro-
lyte. At the end of each voltammetric experiment, [Fe(g5-
C5H5)2] was added to the solution as an internal standard
for potential measurements. Under the conditions used,
the Eo value for the couple [Fe(g5-C5H5)2]+–[Fe(g5-
C5H5)2] was 0.47 V.
Crystal data and other details of the structure analyses
of compounds 1 and 2 are presented in Table 1. Suitable
crystals were mounted at the end of quartz fibers and held
in place with a fluorinated oil. All diffraction measurements
were made at 150 K on an Enraf-Nonius CAD-4 diffrac-
tometer, using graphite monochromated MoKa radiation.
Unit cell dimensions were determined from 24 centered
reflections in the range 20.9ꢁ < 2h < 28.2ꢁ for 1 and in the
range 17.2ꢁ < 2h < 31.5ꢁ for 2. For both structures, dif-
fracted intensities were measured in a quarter of reciprocal
space for 4.0ꢁ < 2h < 50.0ꢁ by x scans in the case of 1, and
by x/h scans for 2. Three check reflections remeasured
after every 300 ordinary reflections showed no decay of
the diffracted intensities over the period of data collection.
Absorption corrections were applied based on azimuthal
scan data (526 for 1 and 476 for 2). Lorentz and polariza-
tion corrections were applied.
The structures were solved by direct methods and
refined using the SHELXL-97 program [38]. All non-hydro-
gen atoms were assigned anisotropic displacement parame-
ters. The hydrogen atoms were constrained to idealized
geometries and assigned isotropic displacement parameters
equal to 1.2 times the Uiso values of their respective parent
atoms. In the case of 2, due to the low intensity of the col-
lected data, there were some problems with the anisotropic
thermal parameters of the C(6), C(14), and C(22) atoms
and thus, a common set of anisotropic thermal parameters
were used for each of these atoms and their adjacent car-
bon neighbors. Full-matrix least-squares refinement of
these models against F2 converged to final residual indices
given in Table 1.
Acknowledgments
This work was supported by the Spanish MEC (DGI)/
FEDER (Projects CTQ2005-08606-C02-01, BFU2005-
07422-C02-02, and MAT2005-1272) and the Gobierno de
´
´
´
Aragon (Grupo de Excelencia: Quımica Inorganica y de
los Compuestos Organometa´licos). We are indebted to Prof.
Dr. S. Alvarez (Universitat de Barcelona) for kindly pro-
viding values of continuous shape measure.
Appendix A. Supplementary material
CCDC 629497 and 629498 contain the supplementary
crystallographic data for this paper. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html, or from the Cambridge Crystallo-
graphic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223-336-033; or e-mail: depos-
it@ccdc.cam.ac.uk. Supplementary data associated with
this article can be found, in the online version, at
doi:10.1016/j.jorganchem.2007.03.028.
4.8. EPR measurements
EPR data were taken in a Bruker ESP380 spectrometer
working at X-band. Measurements at liquid nitrogen tem-
perature (77.3 K) were registered using a quartz immersion
Dewar. The magnetic field was measured with a Bruker
ER035M gaussmeter. A Hewlett–Packard HP5350B fre-
quency counter was used to determine the microwave fre-
quency. The samples were introduced in fused quartz
tubes.
References
[1] M.J. Carney, in: R.H. Crabtree, D.M.P. Mingos (Eds.), in: K.H.
Theopold (Ed.), Comprehensive Organometallic Chemistry III, vol. 5,
Elsevier Ltd., Oxford, UK, 2007, pp. 291–390 (Chapter 5.05);
M.J. Morris, in: E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.),
Comprehensive Organometallic Chemistry II, vol. 5, Elsevier Science,
Ltd, Oxford, UK, 1995, pp. 471–549 (Chapter 8);

P.J. Alonso et al. / Journal of Organometallic Chemistry 692 (2007) 3236–3247
3247
[6] R. Davis, L.A.P. Kane-Maguire, in: G. Wilkinson, F.G.A. Stone,
E.W. Abel (Eds.), Comprehensive Organometallic Chemistry, vol. 3,
Pergamon Press, Oxford, UK, 1982, pp. 953–1077, Section 26.2.
[2] G.K. Barker, M.F. Lappert, J.A.K. Howard, J. Chem. Soc., Dalton
Trans. (1978) 734;
[22] C.J. Cardin, D.J. Cardin, J.M. Kelly, R.J. Norton, A. Roy, B.J.
Hathaway, T.J. King, J. Chem. Soc., Dalton Trans. (1983) 671.
´
´
´
´
[23] I. Ara, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B. Menjon,
Chem. Eur. J. 10 (2004) 4186.
´
´
´
[24] P.J. Alonso, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B.
´
Menjon, Chem. Commun. (2002) 728, see also Ref. [23].
G.K. Barker, M.F. Lappert, J. Organomet. Chem. 76 (1974) C45.
´
´
´
´
´
´
[3] (a) P.J. Alonso, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B.
[25] P.J. Alonso, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B.
Menjo´n, Chem. Commun. (2001) 2138, see also Ref. [13].
[26] B.B. Garrett, M.T. Holbrook, J.A. Stanko, Inorg. Chem. 16 (1977)
1159;
Menjo´n, C. Rillo, Chem. Eur. J. 8 (2002) 4056;
(b) P.J. Alonso, L.R. Falvello, J. Fornies, M.A. Garcıa-Monforte,
A. Martın, B. Menjon, G. Rodrıguez, Chem. Commun. (1998) 1721.
[4] W. Seidel, G. Kreisel, Z. Anorg. Allg. Chem. 426 (1976) 150.
[5] F. Hein, D. Tille, Z. Anorg. Allg. Chem. 329 (1964) 72.
´
´
´
´
´
M.T. Holbrook, B.B. Garrett, Inorg. Chem. 15 (1976) 150.
[27] F.J. Owens, in: F.J. Owens, C.P. Poole, H.A. Farah (Eds.), Magnetic
Resonance of Phase Transitions, Academic Press, New York, 1979
(Chapter 6);
[6] F. Hein, K. Schmiedeknecht, J. Organomet. Chem. 5 (1966) 454.
[7] E. Muller, J. Krause, K. Schmiedeknecht, J. Organomet. Chem. 44
¨
´
(1972) 127.
R. Alcala, P.J. Alonso, R. Cases, J. Phys. C: Solid State Phys. 16
´
´
´
[8] P.J. Alonso, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B.
(1983) 4693.
´
Menjon, Organometallics 24 (2005) 1269.
[28] R.L. Carlin, Magnetochemistry, Springer-Verlag, Berlin, 1986, pp.
21–27, Section 2.3.
[29] Crystal data for 50: Monoclinic, space group P21/n, a = 1466.4(1) pm,
b = 3092.2(1) pm, c = 1468.2(1) pm, b = 90.27(1)ꢁ, V = 6.659(1) nm3;
h range for data collection: 2.07–24.97ꢁ; reflections collected: 12174;
[9] J. Krausse, G. Marx, J. Organomet. Chem. 65 (1974) 215;
E. Kurras, J. Otto, J. Organomet. Chem. 4 (1965) 114.
[10] M.M. Olmstead, P.P. Power, S.C. Shoner, Organometallics 7 (1988)
1380;
F. Hein, R. Weiss, Z. Anorg. Allg. Chem. 295 (1958) 145.
[11] R. Poli, J. Organomet. Chem. 689 (2004) 4291;
R. Poli, Chem. Rev. 96 (1996) 2135.
[12] G.M. Barrow, S. Searles, J. Am. Chem. Soc. 75 (1953) 1175;
L.J. Bellamy, third ed.The Infra-Red Spectra of Complex Molecules,
vol. 1, Chapman and Hall, Ltd, London, 1975, pp. 129–140 (Chapter 7).
[13] (a) A reduction of the E1/2 potential was also observed in the [VIVR4]/
[VIIIR4]ꢀ couple on decreasing the chlorination degree of the R
group.;
independent reflections: 11691 (Rint = 0.0460); final R indices
[I > 2r(I)]: R1 = 0.1026, wR2 = 0.2993; (all data) R1 = 0.1716,
wR2 = 0.3197. All the atoms were refined anisotropically, but a large
amount of unassigned electron density remained due to the poor
quality of the crystal.
[30] J. Beck, M. Koch, Z. Anorg. Allg. Chem. 632 (2006) 756;
J. Beck, M. Koch, Z. Anorg. Allg. Chem. 630 (2004) 394;
J.C. Fettinger, S.P. Mattamana, R. Poli, Acta Crystallogr., Sect. C 54
(1998) 184;
´
´
´
(b) P.J. Alonso, J. Fornies, M.A. Garcıa-Monforte, A. Martın, B.
J.C. Gordon, S.P. Mattamana, R. Poli, P.E. Fanwick, Polyhedron 14
(1995) 1339;
´
Menjon, Chem. Eur. J. 11 (2005) 4713.
´
[14] W.N. Setzer, P. von Rague Schleyer, Adv. Organomet. Chem. 24
A.J. Blake, S. Parsons, A.J. Downs, C. Limberg, Acta Crystallogr.,
Sect. C 51 (1995) 571;
(1985) 353.
[15] The continuous shape measure, S, gives a quantitative assessment on
the extent to which the coordination environment of a real metal
complex approaches a given ideal polyhedron. The smaller the value
of S, the better the agreement between the real molecule and the
suggested model, being S = 0 for perfect coincidence. For the
definition of this structural parameter and its use in four- or six-
coordinate complexes, see Ref. [16].
C.H. Schumacher, F. Weller, K. Dehnicke, Z. Anorg. Allg. Chem.
495 (1982) 135;
P. Klinzing, A. El-Kholi, U. Muller, K. Dehnicke, K. Findeisen, Z.
¨
Anorg. Allg. Chem. 569 (1989) 83;
C.D. Garner, L.H. Hill, F.E. Mabbs, D.L. McFadden, A.T. McPhail,
J. Chem. Soc., Dalton Trans. (1977) 853.
[31] K.K. Sunil, M.T. Rogers, Inorg. Chem. 20 (1981) 3283.
[32] V.I. Murav’ev, Russ. J. Coord. Chem. 31 (2005) 837;
V.I. Murav’ev, Russ. J. Coord. Chem. 31 (2005) 609;
S. Patchkovskii, T. Ziegler, J. Chem. Phys. 111 (1999) 5730;
J. Swann, T.D. Westmoreland, Inorg. Chem. 36 (1997) 5348;
D.M. Sabel, A.A. Gewirth, Inorg. Chem. 33 (1994) 148;
K.K. Sunil, J.F. Harrison, M.T. Rogers, J. Chem. Phys. 76 (1982)
3087.
[33] W.A. Sheppard, J. Am. Chem. Soc. 92 (1970) 5419.
[34] P. Boudjouk, J.-H. So, in: R.N. Grimes (Ed.), Inorganic Syntheses,
vol. 29, John Wiley & Sons, New York, 1992, pp. 108–111.
[35] C. Persson, C. Andersson, Inorg. Chim. Acta 203 (1993) 235.
[36] M.D. Rausch, F.E. Tibbetts, H.B. Gordon, J. Organomet. Chem. 5
(1966) 493.
[16] (a) S. Alvarez, Dalton Trans. (2005) 2209;
(b) J. Cirera, P. Alemany, S. Alvarez, Chem. Eur. J. 10 (2004) 190;
(c) S. Alvarez, D. Avnir, M. Llunell, M. Pinsky, New J. Chem. 26
(2002) 996;
(d) M. Pinsky, D. Avnir, Inorg. Chem. 37 (1998) 5575.
[17] M. Llunell, D. Casanova, J. Cirera, J.M. Bofill, P. Alemany, S.
Alvarez, M. Pinsky, D. Avnir, SHAPE (version 1.1), Universitat de
Barcelona, The Hebrew University of Jerusalem, Spain, Israel.
[18] If the Crꢁ ꢁ ꢁClortho interactions in complex 1 are disregarded, the
remaining CrC4 skeleton can alternatively be described as having a
sawhorse arrangement—related to the so-called W-TBPY-5 (see Ref.
[16b]).
[19] C. Schulzke, D. Enright, H. Sugiyama, G. LeBlanc, S. Gambarotta,
G.P.A. Yap, L.K. Thompson, D.R. Wilson, R. Duchateau, Organo-
metallics 21 (2002) 3810.
[20] V. Gramlich, K. Pfefferkorn, J. Organomet. Chem. 61 (1973) 247.
[21] P. Stavropoulos, P.D. Savage, R.P. Tooze, G. Wilkinson, B. Hussain,
M. Motevalli, M.B. Hursthouse, J. Chem. Soc., Dalton Trans. (1987)
557.
´
´
[37] R. Uson, J. Fornies, Adv. Organomet. Chem. 28 (1988) 219–297;
E. Maslowsky Jr., Vibrational Spectra of Organometallic Com-
pounds, John Wiley & Sons, New York, 1977, pp. 437–442.
[38] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal
Structures from Diffraction Data, University of Go¨ttingen, Go¨ttin-
gen, Germany, 1997.