Solid state structural and solution studies on the formation of a flexible cavity for anions by copper(I) and 1,2-bis(diphenylphosphino)ethane

Article abstract of DOI:10.1016/S0277-5387(02)01189-0

Copper(I) complexes of 1,2-bis(diphenylphosphino)ethane (dppe) with a stoichiometry Cu₂(dppe)₃(X)₂ [X ^- = CN^- (1), SCN^- (2), NO₃^- (3)] are obtained from direct reactions of CuX and dppe. The complexes are structurally and spectroscopically (NMR and IR) characterized. The structure of the [Cu₂(dppe)₃]²⁺ dication is similar to the structural motif observed in many other complexes with a chelating dppe and a bridging dppe connecting two copper centers. In complexes 1-3, the anions are confined to the cavity formed by the phosphines which force a monodentate coordination mode despite the predominant bidentate/bridging character of the anions. The coordination angles rather than the thermochemical radii dictate the steric requirement of anions. While the solution behavior of 3, with nitrate, is similar to complexes studied earlier, complexes with pseudohalides exhibit new solution behavior.

Full text of DOI:10.1016/S0277-5387(02)01189-0

Polyhedron 21 (2002) 2433Á2443
/
www.elsevier.com/locate/poly
Solid state structural and solution studies on the formation
of a flexible cavity for anions by copper(I) and
1,2-bis(diphenylphosphino)ethane
D. Saravanabharathi a, Monika b, Paloth Venugopalan b, Ashoka G. Samuelson a,*
a Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560 012, India
b Department of Chemistry, Panjab University, Chandigarh 160 014, India
Received 24 May 2002; accepted 13 August 2002
Abstract
Copper(I) complexes of 1,2-bis(diphenylphosphino)ethane (dppe) with a stoichiometry Cu2(dppe)3(X)2 [Xꢀ ꢁ
(2), NO3 (3)] are obtained from direct reactions of CuX and dppe. The complexes are structurally and spectroscopically (NMR
/
CNꢀ (1), SCNꢀ
ꢀ
and IR) characterized. The structure of the [Cu2(dppe)3]2ꢂ dication is similar to the structural motif observed in many other
complexes with a chelating dppe and a bridging dppe connecting two copper centers. In complexes 1Á3, the anions are confined to
/
the cavity formed by the phosphines which force a monodentate coordination mode despite the predominant bidentate/bridging
character of the anions. The coordination angles rather than the thermochemical radii dictate the steric requirement of anions.
While the solution behavior of 3, with nitrate, is similar to complexes studied earlier, complexes with pseudohalides exhibit new
solution behavior.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Copper(I); Anions; Dppe; Coordination mode; Solid state structures
1. Introduction
anions [5]. Since transition metal ions provide the
requisite electrostatic, non-covalent and covalent inter-
actions, they have been obvious choices for hosting
anions. In most cases, a covalent bond is formed
with the metal and is clearly the driving force making
anion capture an energetically favorable process [6].
Successful designs of anion specific receptors have based
their design on metal complexes with ligands capable of
hydrogen bonding to augment their interactions with
the anion [5b]. Rare examples have come to light, where
The design and development of molecules to function
as hosts for small molecules or ions is an important area
of research [1]. In particular, anion receptors and
sensors have gained attention due to their relevance in
environmental pollution [2] and in biological systems [3].
Designing hosts for binding anions in solution has
proved to be an arduous task due to the large solvation
energy to be overcome. Due to the variety in shapes and
sizes of anions [4], it is difficult to find a universal host
that would be compatible with different anions. Design-
ing a shape and size selective host for capturing a
specific anion is considerably more involved. In con-
junction with suitable ligands, metal complexes have
been found capable of anchoring or sensing a variety of
the anionÁhost interaction is based purely on non-
/
covalent interactions, with only electrostatic interactions
anchoring the anion to the metal complex [7]. Appro-
priate choice of ligands and metal ions can thus lead to
esoteric hosts for anions. To function as a sensor, the
molecule should possess an additional ability to signal
the binding event. A measurable property of the
receptor, such as its redox potential, or spectroscopic
behavior (absorption and or emission) needs to undergo
* Corresponding author. Tel.: ꢂ
0683
E-mail address: ashoka@ipc.iisc.ernet.in (A.G. Samuelson).
/
91-80-309-2663; fax: ꢂ91-80-334-
/
significant change to signal the presence of anions [1,5Á
/
7].
0277-5387/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 2 7 7 - 5 3 8 7 ( 0 2 ) 0 1 1 8 9 - 0

2434
D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á2443
/
We have been exploring the possibility of using
Encapsulation of these anions in a hydrophobic sheath
prevents bidentate interactions and introduces new
solution behavior.
copper(I) complexes of tertiary aryl phosphines to
design systems suitable for sensing anions [8]. While
most transition metal ions form rather rigid complexes
due to crystal field stabilization effects, closed shell
systems like Cu(I) have a plastic coordination sphere [9].
This allows for considerable deviations from the ideal
geometry required for stability in partly filled d orbitals.
The Cu(I) complex of diphenylphosphinohexane (dpph)
reported by Kitagawa et al. [10] hosts oxoanions of
different shapes and sizes between two copper centers
bridged by the ligand. Our recent studies have demon-
strated the efficiency of non-covalent interactions with
anions in modulating the conformation of phenyl rings
on the ligand such as bis diphenylphosphinomethane
(dppm) [8b]. The nuclearity of the Cu(I) dppm com-
plexes is controlled by the nature of anions. Oxoanions
form dimers having short contacts or weak interactions
between the anion and aromatic hydrogen atoms.
However, halides, hydroxide and pseudohalides lead to
very strong interactions of the anion with copper and
result in the formation of trinuclear copper clusters
bridged on the edges by dppm [8a]. Strong interactions
are a limitation when sensor applications are envisaged
due to possible deactivation on binding. Hence, alter-
native aryl phoshpines are being explored in our
laboratory. Here we report our studies on 1,2-bis(di-
phenylphosphino)ethane (dppe).
2. Results
2.1. Molecular composition and characterization
The composition of Cu(I) phosphine complexes is not
always determined by the stoichiometry of the reactants.
In several cases, it has been shown that halides and
phosphines form complexes with very different formulae
from those dictated by the reaction stoichiometry
[13g,13h]. In the current study, we found that starting
with ratios of dppe/Cu equal to 1.5 always gave copper
complexes with the same stoichiometry. Single crystal
X-ray crystallographic analysis confirmed the structures
proposed on the basis of spectroscopic data. The
complexes chosen for crystallographic examination
have anions that are potentially bidentate and are
predominantly found in bridging environments [15].
These compounds join the group of rare examples of
Cu(I) complexes where cyanide and thiocyanate anions
are found in a monodentate binding mode [16].
2.2. General description of the structures
Dppe is one of the most versatile phosphine ligands,
X-ray diffraction studies of single crystals of the
compounds made in this study confirm the formation
of molecular structures expected in this series [13,14].
Selected bond angles and lengths are given in Table 1. A
representative structure showing the molecular structure
of 2 is given in Fig. 1. There are two types of dppe
capable of stabilizing copper in its ꢂ1 state [11]. Unlike
/
dppm, this ligand can chelate as well as bridge copper
centers. We have earlier demonstrated the possibility of
obtaining both coordination modes upon altering the
ratio of the phosphine with respect to the amount of
Cu(I) [12]. A survey of Cu(I) dppe complexes indicates
repeated occurrence of a family of complexes with the
Table 1
Selected bond lengths and bond angles for 1, 2 and 3.
general stoichiometry Cu2(dppe)3X2 (Xꢁvarious mono
/
anions/guest) [13]. In all these structurally characterized
complexes, two Cu centers are bridged by a dppe ligand
and each metal center possesses one chelating dppe, with
the fourth coordination site available for the anion/guest
(X). It is found that [Cu2(dppe)3]2ꢂ forms a cavity
capable of accommodating a wide variety of anions of
different shapes and sizes. X ranges from chloride to
bulky thiolates. In one system, a neutral ligand aceto-
nitrile, is present in the place of X [14]. This prompted us
to prepare and study three new complexes with poten-
tially bidentate/bridging anions [15], CNꢀ, SCNꢀ and
NO3ꢀ. The complexes have been structurally character-
ized and their solution behavior examined. A critical
examination of the cavity to accommodate guests with
diverse steric demands has been made. Surprisingly,
copper complexes of phosphines with cyanide or thio-
cyanate in the monodentate mode are very few and the
complexes we have synthesized represent some of the
first examples with the anions in this environment.
˚
Bond lengths (A)
Bond angles (8)
1
2
3
Cu(1)ꢀC(3)
Cu(1)ꢀP(1)
Cu(1)ꢀP(2)
Cu(1)ꢀP(3)
1.964(4)
C(3)ꢀCu(1)ꢀP(1)
P(1)ꢀCu(1)ꢀP(2)
P(1)ꢀCu(1)ꢀP(3)
C(3)ꢀCu(1)ꢀP(2)
C(3)ꢀCu(1)-P(3)
P(2)ꢀCu(1)-P(3)
107.05(9)
112.76(3)
115.16(3)
120.73(9)
111.43(8)
89.22(3)
2.2808(8)
2.3040(9)
2.3276(8)
Cu(1)ꢀN(1)
Cu(1)ꢀP(1)
Cu(1)ꢀP(3)
Cu(1)ꢀP(2)
1.985(3)
N(1)ꢀCu(1)ꢀP(1)
P(1)ꢀCu(1)ꢀP(3)
P(1)ꢀCu(1)ꢀP(2)
N(1)ꢀCu(1)ꢀP(3)
N(1)ꢀCu(1)-P(2)
P(3)ꢀCu(1)-P(2)
108.79(9)
115.24(3)
117.01(3)
113.75(9)
111.44(9)
89.72(4)
2.2707(9)
2.2998(9)
2.3058(10)
Cu(1)ꢀO(3)
Cu(1)ꢀP(2)
Cu(1)ꢀP(1)
Cu(1)ꢀP(3)
2.070(3)
O(3)ꢀCu(1)ꢀP(1)
P(1)ꢀCu(1)ꢀP(2)
P(1)ꢀCu(1)ꢀP(3)
O(3)ꢀCu(1)ꢀP(2)
O(3)ꢀCu(1)-P(3)
P(2)ꢀCu(1)-P(3)
109.87(8)
121.04(4)
112.79(4)
116.25(8)
104.01(9)
89.84(4)
2.2697(10)
2.2684(10)
2.3285(10)

D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á
/2443
2435
inhibited as they are encapsulated within the [Cu2(dp-
pe)3]2ꢂcore.
It is interesting to note that there is a significant
relationship between the bending of the anion at the
coordinating atom and the dihedral angle of the
bridging dppe. The Cuꢀ
the angle of 174.38, where as the angle of Cuꢀ
the case of thiocyanate, and CuꢀOꢀN bond of the
nitrate are 1678 and 121.78, respectively. Similarly, the
torsion angles about CꢀC in chelating dppe is almost
similar in these three cases (CNꢀ 558, SCNꢀ 568, NO3
C (bridging
/CN bond is almost linear with
/
NꢀC in
/
/
/
/
ꢀ
578), but the torsion angles of Cuꢀ
/
Pꢀ
/
Cꢀ
/
dppe) vary significantly for the anions CNꢀ (56.38),
SCNꢀ (54.98), and NO3ꢀ (64.88). The relevance of these
differences to the steric requirements of the anion will be
evaluated later after discussing the solution state beha-
vior.
Fig. 1. The molecular structure of 2 at the 30% probability level.
(Hydrogen atoms and labels for phenyl rings are omitted for clarity.
Structure for 1 and 3 are given in supporting information.)
2.3. Spectroscopic characterization
ligands in each of the three molecules. A bridging dppe
holds two metal centers together and both metal centers
posses a second chelating dppe. All are centrosymmetric
molecules and the center of symmetry is located between
the methylene carbons of the bridging dppe. Thus both
metal centers are crystallographically equivalent. In all
these complexes, the copper atom adopts a distorted
tetrahedral geometry due to the constraint imposed by a
chelating dppe. The bite angles of the chelating dppe are
comparable to what has been observed in other struc-
tures [13,14]. They are remarkably constant (89.28,
89.78, 89.88 for anions CNꢀ, SCNꢀ, NO3ꢀ, respec-
2.3.1. IR spectroscopy
In the solid state, the IR spectra of all these complexes
show characteristic bands for the host framework and
the guest anions. The preferences of the host for
different anions are reflected in the lability of the
complexes in solution. A comparison of the solid state
spectra with the solution IR spectra offers useful
information in this regard. The stretching frequency
n(CꢁN) of the cyanide ion in complex 1 occurs at 2100
cmꢀ1 in the solid state and at 2105 cmꢀ1 in chloroform
solution. Compared to the value of the ionic or
uncomplexed cyanide [20] (which occurs at 2076
cmꢀ1), the complexed form shows an increase of 30
cmꢀ1. Similarly in complex 2, the n(CꢁN) mode of the
thiocyanate group is very intense and offers information
about the nature of anion binding. The thiocyanate
tively) and hence an invariant feature of the complex.
˚
P distances range from 2.27 to 2.33 A. The
The Cuꢀ
/
fourth coordination site is occupied by the anion. In the
case of complex 1 the C of CNꢀ coordinated to the
copper makes an angle of 107.058 with respect to the P1
atom of the bridging dppe and 120.718, 111.448,
respectively to the P2 and P3 atoms of the chelating
dppe. Similarly, the N1 atom of the thiocyanate complex
2 and the O3 atom in the nitrate complex 3 coordinate
group can bind to the metal through sulfur (Mꢀ
or nitrogen (MꢀNCS), or bridge through N and S (Mꢀ
NCSꢀM). The mode of coordination can be distin-
/
SCN),
/
/
/
guished from the vibrational frequencies of the SCN
group. Although the exact frequency is dependent on
the metal center, its oxidation state, and the ancillary
ligands [20], the S-bound or bridging thiocyanate groups
exhibit n(CꢁN) at higher frequencies than the N-bound
thiocyanate [21]. In the structurally characterized com-
plex 2, the SCN group binds through nitrogen. As
expected, in the solid state the value of n(CꢁN) stretch in
the present complex is observed at 2070 cmꢀ1. This is
lower in energy in comparison with other Cu(I) phos-
phine complexes possessing a bridging SCN [17] group.
This value is also higher than the stretching frequency of
the uncomplexed SCNꢀ ion (which occurs at 2053
cmꢀ1). In chloroform solution, the same complex
exhibits a band at 2080 cmꢀ1 and free thiocyanate
group is not detected (Fig. 2). Significant differences
between 1, 2 and 3 (vide infra) prompted us to examine
to the copper in a near tetrahedral environment. In the
˚
C distance (1.96 A)
case of the cyanide complex, the Cuꢀ
/
is comparable to the value found in a similar complex
˚
[16] (1.95 A, (PPh2N(H)PPh2)Cu(PPh3)CN). The Cuꢀ
/
N
(1.98 A) bond distance observed in the thiocyanate
˚
structure is in good agreement with the typical length of
a Cuꢀ
the Cuꢀ
than other copper complexes where nitrate is coordi-
/N bond [17]. In the structure containing nitrate,
˚
/
O bond distance is 2.07 A and slightly shorter
˚
nated to Cu (2.174(2) A for [Cu(PPh3)3NO3] [18] and
˚
2.206 A for [Cu(PPh2Me)3NO3] [19]. Thus no unusual
features are observed in these systems other than the
slightly shorter bond length in the case of the nitrate ion.
The predominantly bridging character of these anions is

2436
D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á2443
/
Fig. 2. The IR of complex 2 showing the n(CꢁN) vibrations. The broken
trace (in the bottom spectrum) indicates the effect of added dppe
ligand. Other IR spectra are given in supporting information.
Fig. 3. 31P NMR of complexes: (1) 1; (2) 2; and (3) 3 in CDCl3 at 293
K. Other NMR spectra are given in Section 6.
the analogous complex with azide Cu2(dppe)3(N3)2 (4).
The IR spectrum of complex 4 indicates a characteristic
band at 2033 cmꢀ1 in the solid state indicative of the
azide ion. In chloroform solution the band shifts to 2041
cmꢀ1. As in the case of the thiocyanate complex, no
band corresponding to the presence of free azide anion
could be seen.
The behavior of complex 3 is distinctly different from
that of complexes, 1, 2, and 4. In the solid state, the
nitrate group in complex 3 shows a band at 1287 cmꢀ1
close to the value reported for a monodentate nitrate ion
[20,21]. Other bands corresponding to this ion could not
be observed as they are obscured by bands of the
phosphine ligand in the same region. In chloroform
solution, however, the nitrate ion remains uncoordi-
nated. In solution, a stretching frequency of 1378 cmꢀ1
is observed for this complex very close to that of free
nitrate ions. This dissociation in solution is quite
2.4. 31P NMR spectroscopy
The compounds prepared in this study differ only in
their counterions and so one would not have expected
the 31P NMR of these complexes to be very different.
The solution 31P NMR of complex 3 (with NO3ꢀ), in
CDCl3 at room temperature consists of two sets of
peaks; a broad resonance at 5.0 ppm and a relatively
sharp peak at ꢀ
peak becomes narrower and the peak at ꢀ
/
9.6 ppm (Fig. 3(3)). At 223 K the broad
9.6 ppm splits
/
further. The observation of a temperature dependent
broad peak is consistent with the presence of a bis
chelated intermediate ([Cu(dppe)2]ꢂ) which has been
previously observed in the family of Cu2(dppe)3X2
complexes, as well as in some other Cu(I) dppe
complexes [13a,13f,14].
However, the room temperature 31P NMR of com-
surprising since the nitrateꢀ
/
Cu bond should be quite
pound 1 in CDCl3 consists of two broad signals at ꢀ
/
1.5
strong if the short CuꢀO bond, shorter than other
/
and ꢀ3.5 ppm (Fig. 3(1)). Surprisingly, there is no peak
/
comparable complexes, is any indication of bond
strength.
between 0 and 10 ppm, indicating the absence of the bis
chelated form. On lowering the temperature to 223 K,
The ability of the host [Cu2(dppe)3]2ꢂ to retain the
anion in its cavity is reflected by its reaction with dppe.
This was probed using IR spectroscopy. Excess dppe
(2.2 equiv.) was added to a chloroform solution of the
respective complex and the solution IR spectrum was
the intensity of the peak at ꢀ
expense of the other peak at ꢀ
excess dppe (2.5 equiv.) at room temperature results in a
new set of peaks with a major peak at ꢀ3.4 ppm and a
smaller peak at ꢀ0.5 ppm. Interestingly, the intensity
changes are reversed when the sample is cooled to 223 K
and only the peak at ꢀ0.5 ppm is observed. The free
/1.5 ppm increases at the
/3.5 ppm. Addition of
/
/
recorded. In the case of complex 1 the Cꢁ
/N stretch
remains unaffected and no band corresponding to ionic
CNꢀ ion was observed. In contrast, solutions of 2 and 4
indicate observable dissociation to give the free anion
/
dppe signal at 223 K undergoes an upfield shift similar
to what was observed for dpph by Kitagawa et al. [10].
The room temperature 31P NMR in CDCl3 of
thiocyanate (Fig. 3(2)), and azide compounds show
two broad signals merged together in each case, at
when excess dppe is added. The Cꢁ
/
N stretch of free
thiocyanate (Fig. 2), and NꢁN stretch of the free azide
/
ions are observed in addition to those of the complexed
species. Since the nitrate is uncomplexed in solution in
the absence of added dppe, one expects no change on
addition of dppe. Indeed this is exactly what is observed.
ꢀ
/
9.0 and ꢀ
/
6.5 ppm for the thiocyanate complex and
ꢀ
/
9.0, ꢀ5.7 ppm for the azide complex. Lowering the
/

D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á
/2443
2437
temperature to 223 K does not produce any new peaks
in both cases and only increases the separation between
the peaks present. The addition of excess dppe (2.5
equiv.) changes the solution equilibrium as indicated by
new broad signals, which are barely distinguishable
from the base line at room temperature. While at
of Sadler and coworkers [13f] has shown that a bis
chelated form [Cu(dppe)2]ꢂ is the major species with a
31P resonance at 10.0 ppm when Clꢀ is the counter
anion. In the presence of excess dppe the intensity of this
1
peak increased. The temperature dependent H NMR
studies were also consistent with the observation of
dissociative equilibria with the 2.42 ppm peak assignable
to [Cu(dppe)2]ꢂ, present over the entire range of
temperature and more predominant at lower tempera-
tures. A similar temperature dependent behavior has
been observed in two other complexes. In solutions of
[Cu2(dppe)3(CH3CN)2]2ꢂ, Vijayashree et al. [14] have
noted a similar equilibrium. More recently, Comba et al.
[13a] have studied the iodo complex Cu2(dppe)3I2 in
solution and reached the same conclusion. The observed
low temperature 31P NMR signals of the reported
complexes and those of the present study are collected
in Table 2.
room temperature these new signals appear at ꢀ
/6
ppm for thiocyanate and at ꢀ2 ppm for the azide
/
complex, lowering the temperature to 223 K shifts these
broad signals downfield with well-defined peak posi-
tions. At the low temperature limit, with excess dppe,
the SCN complex shows a clear peak at 6.6 ppm and the
azide complex at 7.5 ppm indicative of the formation of
the bis chelated complex in rapid equilibrium with the
non-chelated form.
2.5. 1H NMR spectroscopy
1
All the compounds prepared in this study indicate
1
temperature dependent chemical shifts in the H NMR
Based on the 31P, H NMR and IR spectra, the three
complexes 1, 2 and 4 appear to be unique in not forming
the bis chelated complex of Cu(I) at any temperature.
Further, complex 1 is different from 2 and 4. Even the
deliberate addition of excess dppe aimed to produce
[Cu(dppe)2]CN species (in analogy with the chloride
complex), was not successful. These results suggest that
the binding of the cyanide group to the metal is quite
strong and the anion is retained in the cavity even in the
presence of excess dppe. The formation of a species with
an ionic cyanide such as [Cu(dppe)2]CN is not favored.
The solution behavior can thus be described by a
fluxional processes involving breaking and closing of
the chelate ring and retention of the bridging coordina-
tion in the 31P NMR time scale. Chelation is normally a
stable coordination mode for five member rings, but the
angle subtended at the metal center (908), is much
smaller than what is required for an ideal tetrahedron.
This would cause bond weakening in this family of
complexes. Since there are two chelate rings, solution
equilibrium will naturally involve a number of species.
As a result of rapid equilibrium, only two sets
of phosphorous resonances are observed. A signal at
spectra. The room temperature NMR spectrum of
complex 1 indicates two broad resonances at 2.3 and
2.0 ppm for methylene protons. A broad multiplet is
observed for aromatic protons in the range 6.6Á7.8
/
ppm. Upon lowering the temperature, the methylene
signals start splitting, and at 223 K two clear signals are
obtained centered at 2.0 and 1.8 ppm. Likewise, the
aromatic protons also exhibit temperature dependence,
and at 223 K they show a complex multiplet in the range
6.4Á8.0 ppm.
/
Complexes 2 and 4 show nearly identical temperature
dependence and differ slightly from complex 1. At room
temperature, the methylene protons show two broad
peaks, 2.2 and 2.1 ppm for the thiocyanate complex and
for the azide complex they are 2.3 and 2.1 ppm. Gradual
lowering of the temperature splits the methylene reso-
nances into three clear signals at the low temperature
limit (2.2, 2.0, and 1.8 ppm for thiocyanate complex and
for azide complex they are 2.2, 2.1, and 1.9 ppm).
Similarly, the aromatic signals observed in the range
7.3Á
/
6.8 for both complexes 2 and 4 undergo further
6.6 ppm. Although
splitting at 223 K in the range 7.7Á
/
ꢀ3.5 ppm could be assigned to the P atoms of bridging
/
complex 3 shows changes with temperature, the changes
are distinctly different. Both methylene (2.4, 2.1 ppm)
dppe. The other peak at ꢀ1.5 ppm could be assigned to
/
the P atoms present in the partially broken chelating
ligand that undergoes exchange with the coordinated
arm. The broad nature of this signal is probably due to
many magnetically inequivalent P atoms in equilibrium.
On decreasing the temperature, a slowing down of the
exchange process leads to reduced line broadening and
increased intensity of the signals. The tenacious nature
of CNꢀ is revealed by the absence of [Cu(dppe)2]ꢂ even
at the low temperature limit.
and aromatic protons (7.3Á7.1 ppm) do not show fine
/
structure but do show slight changes in their chemical
shift on lowering the temperature.
3. Discussion
3.1. Solution behavior
Likewise, the 31P NMR of complexes 2 and 4 do not
indicate the presence of the [Cu(dppe)2]ꢂcation in
significant concentrations. The IR spectra are also
very clear that the free anions are not involved in
The solution behavior of complexes 1, 2, and 4 are
significantly different from 3 and those studied earlier
with counterions, Clꢀ, ClO4ꢀ, or Iꢀ. The elegant work

2438
D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á2443
/
Table 2
Summary of low temperature 31P NMR data for copper dppe complexes
Complex
T (K) [Cu(dppe)2]2ꢂ form (ppm)
Various forms of monodentate dppe (ppm)
References
Cu(dppe)2BF4 in CD2Cl2
Cu(dppe)2I in CD2Cl2
Cu(dppe)(CH3)2C2B4H5 in CD2Cl2
Cu2(dppe)3Cl2 in CD2Cl2
Cu2(dppe)3(CH3CN)2(ClO4)2 in acetone-d6
1 in CDCl3
198
ꢂ7.5
ꢂ7.5
ꢂ8.5
ꢂ10.5
ꢂ7.7
absent
absent
absent
ꢂ5.0
[13a]
[13a]
ꢀ9.5
173
188
233
223
223
223
223
ꢀ3.4
ꢀ4.3
ꢀ3.0
ꢀ3.5
ꢀ9.2
ꢀ9.1
ꢀ8.7
[26]
[13f]
ꢂ3.2
ꢀ1.5
ꢀ1.5
[14]
this work
this work
this work
this work
2 in CDCl3
4 in CDCl3
3 in CDCl3
ꢀ6.5
ꢀ5.7
ꢀ8.6
ꢀ9.0
ꢀ9.0, ꢀ9.7
significant concentrations as in the case of complex 1.
But in the presence of added dppe there are differences
in the IR spectrum and the 31P NMR pattern. This
suggests a different kind of solution processes from what
is observed for 1. The solution 31P NMR indicates a very
broad singlet in both cases (a major difference from the
cyano complex), at room temperature. These anions are
long enough for bridging the metal centers. This might
result in a breaking of the bridging/chelated dppe. Rapid
equilibration through the intermediacy of anion bridged
structures in solution could result in a single broad peak.
In other words, the dissociation in the case of the
thiocyanate and azide complexes could be promoted by
the extra bridging ability of the anion. First the anion
attacks the metal center resulting in a partially bonded
chelated/bridging dppe unit. These processes lead to the
formation of a weakly bound intermediate (S bound
SCN complex for example), which dissociates to give
[Cu(dppe)2]ꢂ. As a result, in the presence of excess
dppe, the equilibrium is shifted towards the bis chelated
complex. In this series, the nitrate complex 3 alone,
behaves similar to the chloride complex. The anion is
not held firmly in the cavity or coordinated to the
copper in solution. However, in the solid state, there is
no difference between the three complexes. Given the
choice between ambidentate anions and dppe, the latter
is preferred for bridging the metal centers.
to the cyanide. The predominance of the bis chelated
form at low temperatures is similar to what is observed
in the chloride complex and is a reflection of the
thermodynamic stability of this form. The solution
equilibrium unique to 2 and 4 is shown in Scheme 1.
On the basis of these results it is clear that the solid
state structure is not retained in solution in these
complexes except in the case of the cyanide. The cyanide
remains coordinated to the copper while most anions are
labile. The p accepting and soft nature of the cyanide
favors an inert complex. However, if these were the only
factors, one is hard put to explain the structure of the
nitrate and thiocyanate complexes. Binding with the
nitrate anion appears to be quite strong judging from
the distance between Cu and O, a hard donor. The
ambidentate SCN anion prefers to have nitrogen
coordination rather than sulfur, despite Cu(I) being
soft and thiophillic [22]. However, all these results are
consisent with the tendency of Cu(I) to balance the
number of soft donors and hard donors in its coordina-
tion sphere [23]. This tendency, along with favorable
short contacts (vide infra) with anions might be
responsible for the coordination of hard donors like N
and O to Cu in the solid state. The Cuꢀ/ONO2 bond in
complex 3 is expected to be strong (on the basis of solid
state results), it succumbs to competition from the soft
centers in solution and is not coordinated to Cu(I) in
ꢂ
1
These results including the H NMR spectra are best
their presence leading to the formation of Cu(dppe)2 .
understood on the basis of varying degree of lability
exhibited by these anions. They point out that the
cyanide complex 1 represents one extreme where the
anion remains completely coordinated to the copper.
Even in the presence of excess dppe, there is no hint of
formation of the bis chelated complex. The nitrate
complex 3 is the other extreme, typical of most
complexes in this series. Complexes 2 and 4 are
intermediate cases and represent unique examples where
additional anion bridged structures exist. The formation
of [Cu(dppe)2]ꢂ species only in the presence of added
What makes the [Cu2(dppe)3]2ꢂ core capable of
accepting large anions? The steric demands of aryl
phosphines and their relationship to the stability of
their metal complexes are quite well known. Usually the
cone angles of phosphines are the best criteria to assess
the steric demands of phosphines. The cone angles of
PPh2Me, and PPh3 are 1368, and 1458, respectively.
Since the dppe ligand is structurally very similar to
PPh2Me (differs only in the methyl group) one would
have expected a similar cone angle at the P center.
However, covalent tying of methylene units results in
significant reduction of cone angle and the cone angle of
dppe has been estimated to be 1258 [11]. This significant
ꢀ
dppe indicates that the SCNꢀ or N3 is less labile than
the complexes studied earlier but more labile compared

D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á
/2443
2439
Scheme 1. Equilibria unique to complexes 2 and 4.
reduction of steric demand around phosphorus atoms
upon chelation to the metal offers sufficient space for a
fourth coordination site involving rather bulky ligands.
reported structures. It is clear that the torsion angles
about the CꢀC bond (t1) (see Fig. 4) of the chelating
dppe are not very different for different anions. This is
in accord with the theoretical studies on five member
chelates. It was shown that the energy of the chelate
/
3.2. Analysis of the solid state structures
critically depends on the torsion angles about the Cꢀ
/C
bond rather than any other conformational change
within the chelate ring [25]. Accordingly, the torsion
angles t1 are found to be invariant in all the structures of
the Cu2(dppe)3 series.
The conformation associated with the bridging dppe
varies significantly. The torsion angles associated with
the bridging dppe (involving small anions) are listed in
Table 3. Gaughan et al. have found the torsion angle (t2)
We now discuss why this framework is flexible
making it an universal host for anions. Using the
programs PREQUEST and VISTA available in CSD [24]
we have analyzed the bond distances, angles and torsion
angles associated with this framework in the present and
of the bridging dppe (Cuꢀ
for describing the overall conformation of the molecule
C bond) is
/
Pꢀ
/
Cꢀ
/
C) as a key parameter
[13i]. An increase in t2 (rotation about the Pꢀ
/
expected to minimize intramolecular bumping. In the
case of small anions, an increase is observed in the
torsion angles of the bridging dppe (Cuꢀ
/
Pꢀ
/
Cꢀ/C) on
increasing the size requirements of the anion. It is
interesting to note that the changes in this torsion angle
correlates with the coordination angle a1 rather than the
size indicated by the thermochemical radii. If one
considers the solid angle swept away by the anion rather
than the volume occupied by the anion, the coordina-
tion angle at the metal is a critical factor. The solid angle
Fig. 4. The flexible torsion and bond angles of the host frame work.

2440
D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á2443
/
Table 3
Steric and conformational parameters in [Cu2(dppe)3X2] complexes
Anions X
Thermochemical radius (pm)
Torsion angle (8) t2 (CuꢀPꢀCꢀCbr
)
a1 (8)
a3 (8)
a2 (8)
ꢀ
NO3
Clꢀ
165
167
177
181
199
206
64.8
51.3
56.3
61.9
54.9
61.2
121.7
109.86
102.23
107.05
99.87
116.25
115.65
111.43
117.48
111.45
113.9
CNꢀ
174.3
134.4
161.7
ꢀ
N3
SCNꢀ
Iꢀ
108.78
109.80
spanned by a bent anion is more. Hence the steric
requirement of the anions are dictated by the coordina-
tion angles rather than the thermochemical radii. In
addition to this, the angles a2 and a3, which involve the
anion and the host frame work, show an inverse
correlation.
The steric demands of bulkier anions are more
difficult to interpret since a major portion of the anion
is outside the cavity formed by dppe ligands. Unlike
small anions, factors other than the coordination angles
influence the bridging dppe angles. This is understand-
able as there are aromatic rings in these anions which
are capable of interacting with the aromatic rings of
dppe. The deviation of the torsion angles in the case of
bulkier guests were attributed by Gaughan et al. to
several non-bonded interactions [13i].
The variation in the size of the cavity produced by the
three PPh2 units coordinated to the Cuꢂ to accommo-
date anions is a result of several factors. The flexibility
associated with the rotation of phenyl rings, the torsion
angles associated with the bridging dppe and the
possibility of different coordination angles around
Cu(I) lead to a fairly flexible host for anions. The
bond distances around copper and the metrical para-
meters associated with the chelating dppe remain con-
stant. The IR, and NMR spectra of these compounds in
solution suggest a complex solution behavior very much
dependant on the anion. A new process involving a
bridging anion rather than one bridged through dppe is
suggested to explain the behavior of azide and thiocya-
nate complexes. A bis chelated dppe complex is ob-
served only in the presence of added dppe in these
systems. However, the cyanide ion is distinctly different
and is the only one, which is kinetically inert. The
[Cu2(dppe)3]2ꢂ core is indeed unique in accommodating
a variety of anions of different shapes and sizes making
it an ideal universal host for anions.
Further modulation of cavity size can be achieved by
changing the spatial arrangement of the six phenyl rings
surrounding each metal center. Four hydrogens on these
phenyl rings roughly demarcate the cavity size. Some of
the aromatic hydrogen atoms are found to interact with
the anion through CꢀH p interactions. A list of short
/
contacts with explanatory figures is given in the
supporting material. These short contacts were also
noticed by Albano et al. in the case of Cu2(dppe)3Cl2
[13h]. It is the N and/or O atom of the anions that
exhibit weak interactions with the aromatic hydrogen
atoms. Thus the short contacts observed in the case of
the thiocyanate are between the N of SCNꢀ and the
aromatic hydrogen atoms. The bent coordination geo-
metry of the thiocyanate localizes the charge on the
more electronegative nitrogen end of the ligand resulting
in a lone pair being available for short contacts. It also
causes a deviation from the correlation between the
coordination angle with the torsion angle t2 in the solid
state.
5. Experimental
5.1. General remarks
Dichloromethane, petroleum ether (b.p. 608Á808),
/
and MeCN were purified and dried by conventional
methods. They were distilled under N2 and deoxyge-
nated before use. All manipulations were carried out
under an atmosphere of purified N2 Standard Schlenk
techniques were used. CuCN was obtained locally and
CuSCN, was freshly prepared before use [27]. Cupric
nitrate was obtained from Ranbaxy, India. The complex
Cu2(dppe)3(N3)2 (4) was synthesized using the reported
1
4. Conclusion
procedure [13j]. H NMR spectra were recorded on a
Bruker ACF 200 MHz or on a Bruker AMX 400 MHz
spectrometer with TMS as the internal reference.
31P{1H} NMR spectra were recorded on a Bruker
AMX 400 MHz spectrometer operating at 162 MHz.
IR spectra were recorded either in solid state as KBr
pellets or as CHCl3 solution on a Bruker spectrometer.
In the present study, direct reactions between dppe
and Cu(I) salts with different bidentate anions have been
explored. The resulting complexes have very similar
solid state structures and conform to the structural
motif that has been characterized for this stoichiometry.

D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á
/2443
2441
Table 4
Summary of crystallographic information for compounds 1Á
/
3
Crystal data
1
2
3
Empirical formula
Formula weight
Crystal system
Space group
C
80H72Cu2N2P6×2H2O
C80H72Cu2N2P6S2
1438.42
C78H72Cu2N2O6P6
1446.28
1410.32
monoclinic
C2/c
monoclinic
P21/n
monoclinic
P21/n
˚
a (A)
23.290(2)
17.662(1)
17.597(2)
92.02(1)
7234.0(11)
4
12.817(1)
19.404(2)
15.508(3)
103.85(1)
3744.7(9)
2
12.806(2)
17.377(3)
16.910(2)
105.66(1)
3623.3(9)
2
˚
b (A)
˚
c (A)
b (8)
3
˚
V (A )
Z
rcalc (Mg mꢀ3
)
1.295
0.768
1.276
0.795
1.326
0.773
Absorption coefficient (mmꢀ1
F(000)
Crystal size (mm)
)
2936
0.31ꢃ0.24ꢃ0.17
1492
0.27ꢃ0.25ꢃ0.24
1500
0.29ꢃ0.27ꢃ0.21
u Range (8)
1.83Á/23.50
1.86Á/24.01
1.78Á
/
24.00
Reflections collected
Independent reflections
Data/restraints/parameters
Final R indices [I ꢀ2s(I)]
Final R indices (all data)
Goodness-of-fit on F2
5492
5345 [Rintꢁ0.0260]
6128
5844 [Rintꢁ0.0155]
5952
5673 [Rintꢁ0.0174]
5673/0/424
R1 aꢁ0.0451, wR2 bꢁ0.0997
R1 aꢁ0.0743, wR2 bꢁ0.1091
1.022
5345/0/415
5844/0/415
R1 aꢁ0.0367, wR2 bꢁ0.0848
R1 aꢁ0.0536, wR2 bꢁ0.0931
1.032
R1 aꢁ0.0408, wR2 bꢁ0.0953
R1 aꢁ0.0585, wR2 bꢁ0.1042
1.042
ꢀ3
˚
Residual r (e A
)
0.343 and ꢀ0.216
0.280 and ꢀ0.237
0.285 and ꢀ0.349
a
R1ꢁ(a½½F0½ꢀ½Fc½½)/(a½F0½).
b
wR2ꢁ[a w(½F0½2ꢀ½Fc½2)2]/a½w½F0½2)2]1/2
.
5.2. Syntheses of complexes
white residue was washed with ether, and dried in
vacuum to get 3 in quantitative yield. Anal. Calc. for
[Cu2(C26H24P2)3(NO3)2]×
/
0.5CH2Cl2: C, 63.32; H, 4.90;
5.2.1. Cu2(dppe)3(CN)2 (1)
N, 1.88. Found: C, 63.70; H, 4.95; N, 1.95%. Suitable
crystals for X-ray diffraction studies were grown from
CH2Cl2 on layering with petroleum ether.
To a slurry of CuCN (0.02 g, 0.2 mmol) in CH2Cl2 (10
ml), 1.5 equiv. of dppe (0.133 g) was added and stirred
for 1 h. After complete dissolution of CuCN, the solvent
was removed under vacuum. The residue was then
washed with Et2O and dried in vacuum to give complex
5.3. X-ray crystallographic study
1
in
quantitative
yield.
CH2Cl2: C, 66.66; H, 5.07; N,
Anal.
Calc.
for
[Cu2(C26H24P2)3(CN)2]×
/
Crystals were mounted along the largest dimension
and used for data collection. The intensity data were
1.92. Found: C, 66.91; H, 5.19; N, 2.11%. Suitable
crystals for X-ray diffraction were grown from CH2Cl2
on layering with petroleum ether.
collected on a Siemens P4 single crystal diffractometer
˚
equipped with Mo source (lꢁ0.71073 A) and highly
/
oriented graphite monochromator. The lattice para-
meters and standard deviations were obtained by least-
squares fit to 40 reflections. The data were collected by
5.2.2. Cu2(dppe)3(SCN)2 (2)
A similar procedure was adopted with 0.03 g of
CuSCN and 0.15 g (1.5 equiv.) of dppe and the yields
2u Áu scan mode with a variable scan speed ranging
/
from 2.08 to a maximum of 60.08 minꢀ1. Three
reflections were used to monitor the stability and
orientation of the crystal and were re-measured after
every 97 reflections. Their intensities did not change
significantly during the data collection. The data were
corrected for LP factors and an empirical absorption
correction based on c scan method was also applied.
The structure was solved by direct methods using
SHELX-97 package [28] and also refined using the same.
All the non-hydrogen atoms were refined anisotropi-
cally. The hydrogen atoms were included in the ideal
positions with fixed isotropic U values and were riding
with their respective non-hydrogen atoms. The hydro-
were
quantitative.
Anal.
0.5CH2Cl2: C, 65.29; H 4.93;
Calc.
for
[Cu2(C26H24P2)3(SCN)2]×
/
N, 1.89. Found: C, 64.81; H, 4.95; N, 1.91%. Suitable
crystals for X-ray diffraction were grown from CH2Cl2
on layering with petroleum ether.
5.2.3. Cu2(dppe)3(NO3)2 (3)
To a solution of Cu(NO3)2 (0.03 g) in distilled MeCN
(15 ml), few strips of Cu foil were added and stirred.
After complete reduction of Cu(NO3)2 (as indicated by
the disappearance of blue color), 0.15 g of dppe was
added to the same solution and stirred for 20 min. After
filtration, the solvent was removed under vacuum. The

2442
D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á2443
/
Commun. (1998) 2109;
gen atoms of the water molecule were located from the
difference Fourier and included as such in the refine-
ment and were riding with water oxygen. A summary of
the details of structure determination is given in Table 4.
(c) P.L. Jones, K.J. Byrom, J.C. Jeffery, J.A. McCleverty, M.D.
Ward, Chem. Commun. (1997) 1361;
(d) K.T. Holman, M.M. Halihan, J.W. Steed, S.S. Juisson, J.L.
Atwood, J. Am. Chem. Soc. 117 (1995) 7848.
[8] (a) J.K. Bera, M. Nethaji, A.G. Samuelson, Inorg. Chem. 38
(1999) 218;
(b) J.K. Bera, M. Nethaji, A.G. Samuelson, Inorg. Chem. 38
(1999) 1725.
6. Supplementary material
[9] B.J. Hathaway, in: G. Wilkinson, R.D. Gillard, J.A. McCleverty
(Eds.), Comprehensive Coordination Chemistry, vol. 5, Pergamon
Crystallographic data for the crystal structures have
been deposited with the Cambridge Crystallographic
Press, Oxford, 1987, pp. 534Á593.
/
Data Centre, CCDC NOS. 183692Á183694 for com-
/
[10] S. Kitagawa, M. Kondo, S. Kawata, S. Wada, M. Maekawa, M.
Munakata, Inorg. Chem. 34 (1995) 1445.
[11] C.A. McAuliffe, in: G. Wilkinson, R.D. Gillard, J.A. McCleverty
(Eds.), Comprehensive Coordination Chemistry, vol. 2, Pergamon
pounds 1, 2 and 3, respectively. Copies of this informa-
tion may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK
Press, Oxford, 1987, pp. 984Á1066.
/
(fax: (internet) ꢂ44-1223-336033; e-mail: deposit@
/
[12] N. Vijayashree, Ph.D. Thesis, Indian Institute of Science, 1993.
[13] (a) P. Comba, C. Katsichtis, B. Nuber, H. Pritzkow, Eur. J. Inorg.
Chem. (1999) 777;
ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk/conts/
retrieving.html).
(b) M. Biagini-Cingi, Anna-Maria Manotti-Lanfredi, F. Ugoz-
zoli, A. Camus, N. Marsich, Inorg. Chim. Acta 262 (1997) 69;
(c) J. Romero, J.A. Garcia-Vazquez, A. Sousa, Y.D. Chang, J.
Zubieta, Inorg. Chim. Acta 245 (1996) 119;
Acknowledgements
(d) E.W. Ainscough, E.N. Baker, M.L. Brader, A.M. Brodie, S.l.
Ingham, J.M. Waters, J.V. Hanna, P.C. Healy, J. Chem. Soc.,
Dalton Trans. (1991) 1243;
We wish to thank CSIR, New Delhi for financial
assistance and the Chairman, Bioinformatics Center,
and Sophisticated Instrumentation Facility, Indian In-
stitute of Science, Bangalore, India for the data from
CSD and the NMR spectrometer, respectively.
(e) E.W. Ainscough, E.N. Baker, A.G. Bingham, A.M. Brodie,
C.A. Smith, J. Chem. Soc., Dalton Trans. (1989) 2167;
(f) S.J. Berners-Price, R.K. Johnson, C.K. Mirabelli, L.F.
Faucette, F.L. McCabe, P.J. Sadler, Inorg. Chem. 26 (1987) 3383;
(g) P. Fiaschi, C. Floriani, M. Pasquali, A. Chiesi-Villa, C.
Guastini, Inorg. Chem. 25 (1986) 462;
References
(h) V.G. Albano, P.L. Bellon, G. Ciani, J. Chem. Soc., Dalton
Trans. (1972) 1938;
[1] (a) P.D. Beer, P.A Gale, Angew. Chem., Int. Ed. Engl. 40 (2001)
486;
(i) A.P. Gaughan, K.S. Bowman, Z. Dori, Inorg. Chem. 11 (1972)
601;
(b) P.A. Gale, Coord. Chem. Rev. 199 (2000) 181.
[2] (a) B. Moss, Chem. Ind. (1996) 407;
(j) A.P. Gaughan, R.F. Ziolo, Z. Dori, Inorg. Chem. 12 (1971)
2776.
(b) C. Glidewell, Chem. Br. 26 (1990) 137.
[3] (a) P.J. Chakrabarti, J. Mol. Biol. 234 (1993) 463;
(b) H. Luecke, F.A. Quiocho, Nature 347 (1990) 402;
(c) J.J. He, F.A. Quiocho, Science 251 (1991) 1479.
[14] N. Vijayashree, A.G. Samuelson, M. Nethaji, Curr. Sci. 65 (1993)
57.
[15] (a) Y. Zhao, M. Hong, W. Su, R. Cao, Z. Zhou, A.S.C. Chan, J.
Chem. Soc., Dalton Trans. (2000) 1685;
[4] S.R. Gadre, C. Kolmel, I.H. Shrivastava, Inorg. Chem. 31 (1992)
¨
(b) D.J. Chesnut, J. Zubieta, Chem. Commun. (1998)
1707;
2279.
[5] (a) J.A. Atwood, K.T. Holman, J.W. Steed, Chem. Commun.
(1996) 1401;
(c) D.J. Chesnut, A. Kusnetzow, J. Zubieta, J. Chem. Soc.,
Dalton Trans. (1998) 4801;
(b) P.D. Beer, Acc. Chem. Res. 31 (1998) 71.
[6] (a) G.S. Papaefstathiou, S.P. Perlepes, A. Escuer, R. Vicente, M.
Font-Bardia, X. Solans, Angew. Chem., Int. Ed. Engl. 40 (2001)
884;
(d) M. Kabesova´, R. Boca, M. Meln´ık, D. Valigura, M. Dunaj-
Jurco, Coord. Chem. Rev. 140 (1995) 115.
[16] J. Ellermann, F.A. Knoch, K.J. Meier, M. Moll, J. Organomet.
Chem. 428 (1992) C44.
(b) V. Amendola, E. Bastianello, L. Fabbrizzi, C. Mangano, P.
Pallavicini, A. Perotti, A.M. Lanfredi, F. Ugozzoli, Angew.
Chem., Int. Ed. Engl. 39 (2000) 2917;
[17] (a) A.P. Gaughan, R.F. Ziolo, Z. Dori, Inorg. Chem. Acta 4
(1970) 640;
(b) N.K. Homsy, M. Noltemeyer, H.W. Roesky, H.-G. Schmidt,
G.M. Sheldrick, Inorg. Chem. Acta 90 (1984) L59.
[18] J.C. Dyason, L.M. Engelhardt, P.C. Healy, H.L. Klich, A.H.
White, Aust. J. Chem. 39 (1986) 2003.
(c) R.-D. Schnebeck, E. Freisinger, B. Lippert, Angew. Chem.,
Int. Ed. Engl. 38 (1999) 168;
(d) A. Knoepfler-Muhlecker, B. Scheffter, H. Kopacka, K. Wurst,
P. Peringer, J. Chem. Soc., Dalton Trans. (1999) 2525;
(e) S. Warzeska, R. Kramer, Chem. Commun. (1996) 499;
[19] M. Mathew, G.J. Palenik, A.J. Carty, Can. J. Chem. 49 (1971)
4119.
¨
(f) B. Hammerle, E.P. Muller, D.L. Wilkinson, G. Muller, P.
¨
¨
[20] K. Nakamato, Infrared Spectroscopy of Inorganic and Coordina-
tion Compounds, 2nd ed., Wiley, New York, 1978, p. 178.
[21] A.J. Carty, A. Efraty, Inorg. Chem. 8 (1969) 543.
[22] (a) R.G. Pearson, Chemical Hardness: Applications from Mole-
cules to Solids, Wiley-VCH, Weinheim, 1997, p. 3;
(b) R.G. Pearson, J. Chem. Edu. 45 (1968) 581.
Peringer, J. Chem. Soc., Chem. Commun. (1989) 1527;
(g) L. Fabbrizzi, P. Pallvicini, L. Parodi, A. Taglietti, Inorg.
Chim. Acta 238 (1995) 5.
[7] (a) T. Okubo, S. Kitagawa, M. Kondo, H. Matsuzaka, T. Ishii,
Angew. Chem., Int. Ed. Engl. 38 (1990) 931;
(b) K.T. Holman, G.W. Orr, J.W. Steed, J.L. Atwood, Chem.

D. Saravanabharathi et al. / Polyhedron 21 (2002) 2433Á
/2443
2443
[23] G. Pilloni, B. Corain, M. Degano, B. Longato, G. Zanotti, J.
Chem. Soc., Dalton Trans. (1993) 1777.
[27] G.H. Jeffery, J. Bassett, J. Mendham, R.C. Denney, Vogel’s
Textbook of Quantitative Chemical Analysis, 5th ed., Longman
UK, London, 1989, p. 495.
[24] Cambridge Structural Database, Version 5.21 Release, April 2001.
[25] J.R. Gollogly, C.J. Hawkins, Inorg. Chem. 8 (1969) 1168.
[26] L. Barton, P.K. Rush, Inorg. Chem. 24 (1985) 3413.
[28] G.M. Sheldrick, SHELX-97, Fortran Program for Crystal Structure
Refinement, 1997.