A kinetic investigation of the oxidative addition reactions of the dimeric Bu4N[Ir2(μ-Dcbp)(cod)2] complex with iodomethane

Article abstract of DOI:10.1016/j.ica.2006.03.041

The kinetic results of the oxidative addition of iodomethane to Bu₄N[Ir₂(μ-Dcbp)(cod)₂] (Dcbp = 3,5-dicarboxylatepyrazolate anion) show that oxidative addition can occur via a direct equilibrium pathway (K₁ = 88(22) acetone, 51(3) 1,2-dichloroethane, 55(4) dichloromethane, 52(12) acetonitrile and 43(5) M^-1 chloroform) or a solvent-assisted pathway (k₂, k₃). Oxidative addition occurs mainly along the direct pathway, which is a factor 10-40 faster than the solvent-assisted pathway. The observed solvent effect cannot be attributed to the donosity or polarity of the solvents. The fairly negative ΔS^≠ value (-110(7) J K^-1 mol^-1) and the positive ΔH^≠ value (+47(2) kJ mol^-1) for the oxidative addition step are indicative of an associative process.

Full text of DOI:10.1016/j.ica.2006.03.041

Inorganica Chimica Acta 359 (2006) 3800–3806
www.elsevier.com/locate/ica
A kinetic investigation of the oxidative addition reactions of the
dimeric Bu4N[Ir2(l-Dcbp)(cod)2] complex with iodomethane
*
*
Ebeth Grobbelaar, Walter Purcell , Stephen S. Basson
Department of Chemistry, University of the Free State, Bloemfontein 9300, South Africa
Received 3 June 2005; accepted 31 March 2006
Available online 13 May 2006
Abstract
The kinetic results of the oxidative addition of iodomethane to Bu4N[Ir2(l-Dcbp)(cod)2] (Dcbp = 3,5-dicarboxylatepyrazolate anion)
show that oxidative addition can occur via a direct equilibrium pathway (K1 = 88(22) acetone, 51(3) 1,2-dichloroethane, 55(4) dichloro-
methane, 52(12) acetonitrile and 43(5) Mꢀ1 chloroform) or a solvent-assisted pathway (k2, k3). Oxidative addition occurs mainly along
the direct pathway, which is a factor 10–40 faster than the solvent-assisted pathway. The observed solvent effect cannot be attributed to
¼
¼
the donosity or polarity of the solvents. The fairly negative DS value (ꢀ110(7) J Kꢀ1 molꢀ1) and the positive DH value
(+47(2) kJ molꢀ1) for the oxidative addition step are indicative of an associative process.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Oxidative addition; Iridium; 3,5-Dicarboxylatepyrazolate; Iodomethane; Kinetics
1. Introduction
type of deactivation is the result of electron donation from
the metal ion in the lower oxidation state to the metal ion in
the higher oxidation state. The result of this deactivation is
that while one metal ion is activated for oxidative addi-
tion the other is activated for reductive elimination, a fea-
ture which can be very useful in the design of catalysts [13].
Despite this very attractive feature of the oxidative addi-
tion reactions of dimeric pyrazolyl-bridged iridium(I)
dimers little is known about the kinetics of these complexes.
A kinetic investigation of the oxidative addition reactions of
CH3I to Bu4N[Ir2(l-Dcbp)(cod)2] was thus undertaken to
gain knowledge about the mechanism and factors which
influence the oxidative addition reactions of dimeric com-
plexes. These results were also compared to the oxidative
addition reactions of the corresponding monomeric irid-
ium(I) b-diketone complexes done in our laboratory [14].
The possible applications of dimeric complexes such as
pyrazolyl-bridged iridium(I) dimers in catalysis have
increased tremendously during recent years [1]. In certain
cases such as in the hydrogenation of cyclohexene the cat-
alytic activity of the dimeric complexes are higher than that
of the corresponding monomeric complexes [2–5]. The
higher catalytic activity of the dimeric complexes is attrib-
uted to the electronic communication between the metal
ions through bridging ligands such as pyrazoles [6].
The presence of two metal ions which can undergo oxi-
dative addition give rise to different oxidative addition pat-
terns such as reversible and irreversible oxidative addition
and addition on only one or both metal ions [8]. Co-opera-
tive effects between the metal ions could also lead to metal–
metal bond formation and/or the migration of the added
fragments between the two metal ions [9–12]. In some cases
oxidative addition on one metal ion leads to the deactiva-
tion of the second metal ion for oxidative addition. This
2. Experimental
2.1. General considerations
*
All the preparations were performed under an atmo-
sphere of dry nitrogen and unless otherwise stated all the
Corresponding authors. Fax: +27 51 4307805 (W. Purcell).
E-mail address: purcellw.sci@mail.uovs.ac.za (W. Purcell).
0020-1693/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2006.03.041

E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
3801
chemicals were reagent grade and used without further
purification. IR spectra were recorded with a Hitachi
270-50 spectrophotometer while the NMR spectra were
obtained at 293 K on a Bruker 300 MHz spectrometer.
was redissolved in 3 cm3 chloroform, filtered and the sol-
vent removed with nitrogen gas. The product was dried
1
over P2O5 in a vacuum desicator for 14 h. Yield: 50%. H
NMR (CDCl3), d 6.90 ppm H(s), d 4.35 ppm CH(m)
(cod), d 2.61(m), 1.30–1.80 ppm CH2(m) (cod), 1.22 ppm
(d), d 1.38–1.48 ppm (m), d 3.28 CH2(m) Bu4N, d
1.1 ppm CH3(t) Bu4N, d 2.17 ppm CH3.
2.2. Syntheses
2.2.1. Preparation of bis-[l-chloro-g4-cycloocta-1,
5-dieneiridium(I)]
2.3. Kinetics
[Ir(l-Cl)(g4-cod)]2 was prepared by a modification of
the method published by Bezman et al. [15]. Hydrated iri-
dic acid, H2[IrCl6] Æ xH2O, 5 g (Next Chimica) and 3 g of
hydroquinone (Merck) were heated under reflux for 1 h
with 180 cm3 of an ethanol/water (2:1) mixture. The etha-
nol/water mixture was deoxygenated for half an hour prior
to usage. The reaction mixture was refluxed for a further 4
h on addition of 5 cm3 cis,cis-1,5-cyclooctadiene. Approxi-
mately 100 cm3 of solvent was distilled from the reaction
vessel and 10 cm3 of water was added, causing the crystal-
lisation of an orange solid. After cooling the reaction mix-
ture to room temperature, the solid was filtered and washed
thoroughly with cold methanol. The product was dried
over P2O5 in a vacuum desiccator for ca. 15 h. Yield:
72%, 1H NMR (CDCl3), d 4.18 ppm CH(m), d 2.20,
1.60 ppm CH2(m).
The solvents used during the kinetic study were purified
and dried prior to use by standard procedures [17]. The
kinetic measurements were performed at atmospheric pres-
sure on a GBC (Model 916) and a Varian (Cary 50) UV/
Vis spectrophotometer equipped with a thermostatted 6
cell changer (0.1 ꢁC). The oxidative addition reactions with
CH3I were followed at 425.0 nm with the exception of ace-
tonitrile and dichloromethane, which were followed at
420.0 nm. Typical experimental conditions were [Ir2-com-
plex] = 0.5–5.5 · 10ꢀ4 M and [CH3I] varied between
1.0 · 10ꢀ2 and 0.2 M thus ensuring good pseudo-first-order
plots of ln(At ꢀ A ) versus time where At and A are the
1
1
absorbencies at time t and infinity, respectively. The
observed pseudo-first-order rate constants where calculated
from the above plots by using a non-linear least-square
program. The spectrophotometric determination of the
equilibrium constant were performed similarly to previous
work [14] at k = 425 nm in acetone. The metal complex
solutions were freshly prepared before each experiment to
minimise the extent of the solvolysis reaction.
2.2.2. Preparation of tetrabutylammonium l-3,
5-dicarboxylatopyrazolatobis[g4-cycloocta-1,
5-dieneiridate(I)], Bu4N[Ir2(l-Dcbp)(cod)2]
´
The preparation of Bu4N[Ir2(l-Dcbp)(cod)2] by Bayon
et al. [16] was modified as follows: 0.2 mmol pyrazole-
3,5-dicarboxylic acid (Aldrich) was dissolved in 6 cm3
0.1 M Bu4NOH (Merck) and stirred in a nitrogen atmo-
sphere for ca. 1 min to form the pyrazole-3,5-dicarboxyla-
tepyrazolate anion. [Ir(l-Cl)(g4-cod)]2 (0.2 mmol) was
dissolved in 5 cm3 of acetonitrile and added dropwise
(slowly) to the pyrazole-3,5-dicarboxylatepyrazolate solu-
tion. The solution was stirred for a further 10 min in a
nitrogen atmosphere and the bright yellow solid was fil-
tered off. The product was washed with hexane and dried
over P2O5 in a vacuum desiccator for ca. 15 h. Yield:
3. Results and discussion
3.1. Synthesis
The slight variation in the synthesis of the [Ir(l-Cl)-
(g4-cod)]2 complex, namely the initial reduction of H2[IrCl6] Æ
xH2O with hydroquinone, followed by the addition of cod
resulted in the formation of a pure orange crystalline prod-
uct whereas the method of Bezman and Baird always yielded
a mixture of orange and brown crystals. The slight variation
in the synthesis of the Bu4N[Ir2(l-Dcbp)(cod)2] complex
also gave better yields. The neutralisation of the pyrazole-
3,5-dicarboxylic acid with Bu4NOH had the dual benefit
of neutralising the acid and acting as cation for the final
product. The m(OCO) of 1642 cmꢀ1 compares favourably with
60%. mOCO1642 cmꢀ1 1H NMR (CDCl3), d 6.85 ppm
,
H(s), d 4.10 ppm CH(m) (cod), d 2.17(m), 1.35–1.70 ppm
CH2(m) (cod), 1.23 ppm (d), d 1.35–1.70 ppm (m), d 3.20
CH2(m) Bu4N, d 1.00 ppm CH3(t) Bu4N.
´
2.2.3. Preparation of tetrabutylammonium
iodo-g4-cycloocta-1,5-dieneiridate(II)-l-3,
that obtained by Bayon et al. but is a little higher than the
value obtained for the corresponding rhodium(I) complex
(m(OCO) = 1631 cmꢀ1) indicating higher electron density in
the Rh–O bond compared to that in the Ir–O bond.
A small chemical shift for all the protons in the oxidative
addition product indicates a change in the chemical environ-
ment within the product compared to that of the starting
material. The methyl oxidative protons at d 2.17 ppm corre-
sponds well with the values 2.17 and 2.20 for the mono
nuclear oxidative addition products of [Ir(LL)(cod)]
complexes LL = 2-pyridinethiolato-N-oxide (hpt) and
5-dicarboxylatopyrazolatomethyl-g4-cycloocta-1,
5-dieneiridate(II), Bu4N[(cod)Ir(I)-l-Dcbp-Ir(cod)(CH3)]
Bu4N[Ir2(l-Dcbp)(cod)2(CH3)I] was prepared as fol-
lows: 0.02 mmol Bu4N[Ir2(l-Dcbp)(cod)2] was dissolved
in 5 cm3 chloroform (Merck) and a 40-fold excess CH3I
was added to this solution. The solution was heated to
35ꢁC and left for 2 h to allow the oxidative addition to
go to completion. The solution was filtered and the solvent
removed with nitrogen gas. The yellow/orange oily product

3802
E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
N-methyl-4-methoxybenzothiohydroxamate (AnMetha),
respectively [18].
Dcbp)(cod)2]-complex in solutions of acetonitrile, acetone,
1,2-dichloroethane, dichloromethane and chloroform.
The following rate law for the proposed mechanism in
Scheme 1 was deduced with the application of the steady
state approximation to the solvolysis intermediate, [Ir2(l-
Dcbp)(cod)2S]ꢀ [19]:
3.2. Kinetics
The proposed reaction mechanism for the oxidative
addition of CH3I to Bu4N[Ir2(l-Dcbp)(cod)2] (Scheme 1)
involves two different pathways namely the k1/kꢀ1 pathway
(direct route) and the k2/k3 or solvent-assisted pathway.
ꢀ
ꢁ
k2½Irꢁ þ k ½Pꢁ
ꢀ3
R ¼ k1½Irꢁ½CH3Iꢁ ꢀ k ½Pꢁ þ k2½ Irꢁ ꢀ k
ꢀ1
ꢀ2
k
ꢀ2 þ k3½CH3Iꢁ
ð1Þ
The k1 and k reaction steps can occur on their own or
ꢀ1
together with the solvent-assisted pathway.
where [Ir] = (Bu4N)[Ir2(l-Dcbp)(cod)2] and [P] = (Bu4N)-
[Ir2(l-Dcbp)(cod)2(CH3)(I)].
Although only one relative slow reaction was detected
for the oxidative addition of CH3I to Bu4N[Ir2(l-
Dcbp)(cod)2], the possibility of a fast reaction exists because
in the case of dimeric complexes both metal ions can
undergo oxidative addition. The presence of a fast reaction
was investigated with a stopped-flow method, but the
results were inconclusive. We therefore decided upon a
two-center addition of the iodomethane to give an Ir(II)–
Ir(II) complex with the possibility of a metal–metal bond
to fulfil 18 electron stability. Similar asymmetric alkyl
halide addition have been observed for [(PPh3)(CO)-
Ir(Me)-(l-pz)2-Ir(I)(PPh3)(CO)] [6] and [(g4-cod)Ir(Me)-
(l-pz)2-Ir(I)(g4-cod)] [7].
The isosbestic point, which was detected during the oxi-
dative addition reaction, implies that only one product is
formed during the reaction. The results for the oxidative
addition reaction between Bu4N[Ir2(l-Dcbp)(cod)2] and
CH3I is given in Table 2. The single reaction includes the
possibility of oxidative addition on only one metal ion or
simultaneous addition on both metal ions. We were unable
to characterise the reaction product by means of X-ray
crystallography so that the exact nature of oxidative addi-
tion (on both or on only one metal ion) is still unknown. A
solvent-assisted pathway was included in the proposed
mechanism because of the instability of the Bu4N[Ir2(l-
Integrating Eq. (1) under the experimental conditions
chosen, the observed pseudo-first-order rate constant for
Eq. (1) is given by the following equation:
ꢀ
ꢁ
k2 þ k
ꢀ3
kobs ¼ k1½CH3Iꢁ þ kꢀ1 þ k2 ꢀ k
ð2Þ
ꢀ2
k
ꢀ2 þ k3½CH3Iꢁ
The kꢀ3 pathway can be considered negligible small on the
grounds that oxidative addition via the solvent-assisted
pathway goes to completion so that Eq. (2) simplifies to
Eq. (3).
k2k3½CH3Iꢁ
kobs ¼ k1½CH3Iꢁ þ kꢀ1 þ
ð3Þ
k
ꢀ2 þ k3½CH3Iꢁ
Eq. (3) consists of the two different pathways namely the k1
and k pathway and the k2/k3 pathway. The direct path-
ꢀ1
way (k1/kꢀ1) can occur on its own or in conjunction with
the solvent-assisted pathway (k2/k3).
When the contribution of the solvent-assisted pathway
to the total rate is negligible, it means that oxidative addi-
tion mainly occurs along the direct pathway. If oxidative
addition occurs via the k1 and kꢀ1 pathways in the presence
of an excess CH3I ([CH3I] ꢂ [Complex]) the rate constant
for the direct pathway will be as follows:
kobs ¼ k1½CH3Iꢁ þ k
ð4Þ
ꢀ1
-
-
K1, k1
+ CH3I
+ CH3I
Ir
Ir
I
Ir
Ir
k-1
K2 + S
CH3
k3
k-3
k-2
k2
-
S
Ir
Ir
= μ-Dcbp3-
= cod
S = solvent
-
3
O
O
O
O
C
C
3
-
=
N
N
Dcbp
Scheme 1.

E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
3803
A graph of kobs versus [CH3I] will be linear with a slope
equal to k1 and an intercept equal to k
.
ꢀ1
The results of our study however have shown that the
solvent-assisted pathway contributes to the rate of oxida-
tive addition. An example of the solvolysis reaction
between Bu4N[Ir2(l-Dcbp)(cod)2] and the different solvents
used in this study is shown in Fig. 1 and the kobs values for
these reactions are listed in Table 1. A comparison between
the kobs value (1.3(3) · 10ꢀ3 sꢀ1 at [CH3I] = 0) for the oxi-
dative addition of CH3I to Bu4N[Ir2(l-Dcbp)(cod)2] in ace-
tone and the kobs value (4.56(5) · 10ꢀ4 sꢀ1) for the reaction
between acetone and Bu4N[Ir2(l-Dcbp)(cod)2] show that
the solvolysis reaction do not influence the rate of oxidative
addition unduly. The solvolysis reactions between
Bu4N[Ir2(l-Dcbp)(cod)2] and 1,2-dichloroethane, chloro-
form and dichloromethane furthermore show isosbestic
points (Table 1) which is indicative of the formation of
only one product. This implies that kꢀ2 ꢃ k2 so that
ksolv @ k2. The solvolysis complex can further react with
CH3I to form the same product which is obtained through
oxidative addition by the k1/kꢀ1 pathway. A further impor-
tant observation is the fact that the oxidative addition of
CH3I to Bu4N[Ir2(l-Dcbp)(cod)2] shows an isosbestic point
at 358.0 nm (Fig. 2) while the solvolysis reaction in Fig. 1
would probably show an isosbestic point at ꢄ340.0 nm.
The contribution of the solvent-assisted pathway to the
rate of oxidative addition was postulated as follows:
(a)
(b)
1.000
0.500
A
0.000
350.0
400.0
450.0
Wavelength (nm)
Fig. 2. (a) (Bu4N)[Ir2(l-Dcbp)(cod)2] without added CH3I in acetone;
(b) spectral change for the oxidative addition of CH3I to (Bu4N)[Ir2(l-
Dcbp)(cod)2] with 1 min intervals at 25 ꢁC.
(a) If k3[CH3I] ꢂ k in Eq. (2), then Eq. (3) changes to
ꢀ2
kobs ¼ k1½CH3Iꢁ þ kꢀ1 þ k2
ð5Þ
The intercept of Eq. (5) now contains an additional
contribution, namely k2. Eq. (5) thus implies that as
soon as the solvolysis intermediate is formed, it reacts
quite rapidly with CH3I to form the expected oxida-
tive addition product.
(b) If k3[CH3I] ꢃ k in Eq. (3), then Eq. (3) changes to
ꢀ2
kobs ¼ ðk1 þ k3K2Þ½CH3Iꢁ þ k
ð6Þ
1.000
ꢀ1
The slope of Eq. (6) now contains contributions from the
direct as well as the solvent-assisted pathway so that a graph
of kobs versus [CH3I] will have a slope of k1 + k3K2 and an
intercept equal to kꢀ1. Eq. (6) implies therefore that the
reaction between the solvolysis intermediate and CH3I is
quite slow.
0.500
A
Plots of the pseudo-first-order rate constants, kobs
,
according to Eqs. (5) and (6) for the oxidative addition
reactions against [CH3I] in the different solvents gave linear
relationships with non-zero intercepts (Fig. 3) implying
reversible oxidative addition.
0.000
400.0
500.0
Eqs. (5) and (6) contain either an additional contribu-
tion towards the slope or the intercept. Both Eqs. (5) and
(6) predict the same experimental results (a linear graph
with a non-zero intercepts). From this it is clear that the
mechanism, whether direct or a combination of direct
and solvent-assisted, could not be determined from the
kinetic results alone. Two further indicators, namely the
difference between the oxidative addition and solvolysis
rates and the isosbestic point which was observed for the
oxidative addition reaction, were used to propose the
mechanism in Scheme 1.
Wavelength (nm)
Fig. 1. Spectral changes of the Bu4N[Ir2(l-Dcbp)(cod)2] complex
(4.9 · 10ꢀ4 M) in acetone (10 min intervals, 25.0 ꢁC).
Table 1
The solvolysis rate constants for Bu4N[Ir2(l-Dcbp)(cod)2] in different
solvents at 25.0 ꢁC
Solvent
(103) kobs (sꢀ1
)
t1/2 (s)
Isosbestic point (nm)
1,2-Dichloroethane
Chloroform
Dichloromethane
Acetonitrile
0.87(3)
0.84(1)
0.64(3)
0.51(2)
0.456(5)
797
825
1083
1359
1520
357.0
359.0
352.0
Another possibility for the direct interaction between
the dinuclear iridium complex and methyl iodide is the nor-
mal SN2 mechanism. The reaction between the metal
Acetone

3804
E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
This is not the case with the latter being constantly larger
chloroform
32
28
24
20
16
12
8
which may be interpreted that the intercepts are more com-
plex in accordance with Eq. (5). However, the opposing
solvent polarity effects (Table 2) may also be indicative of
the existence of a neutral Ir(I)–Ir(III) intermediate species
(vide supra) in the SN2 process since the increase of solvent
polarities lead to an unexpected 4-fold decrease in oxida-
tive addition rates. If the Ir(I)–Ir(III) complex is main-
tained at steady state concentrations (as indicated by Eq.
(8)) then the linear kinetic behaviour in Fig. 3 could be rec-
onciled with the above-mentioned species’ existence.
The isosbestic point which was observed for the oxida-
tive addition of CH3I to (Bu4N)[Ir2(l-Dcbp)(cod)2] in ace-
tone under pseudo-first-order conditions indicates the
presence of only one reaction. If there is a solvolysis reac-
tion present the contribution of this second slow reaction
to the spectral changes observed in Fig. 2 is negligible small
so that the spectral change is only that of oxidative addi-
tion. This explains the clearly defined isosbestic point in
Fig. 2. Some uncertainty exists however because the isos-
bestic point for the solvolysis reaction in chloroform
(356.0 nm) and that for the oxidative addition reaction
(359.0 nm) in the same solvent are very close to one
another. The isosbestic point for the solvolysis reaction in
acetone when extrapolated occurs at a measurably lower
wavelength (ꢄ340.0 nm). This reinforces the assumption
that the spectral changes in Fig. 2 are only due to the oxi-
dative addition reaction.
dichloromethane
acetone
acetonitrile
4
0
0.00
0.05
0.10
0.15
0.20
0.25
[CH3I] (M)
Fig. 3. kobs vs. [CH3I] for the oxidative addition reaction of CH3I with
(Bu4N)[Ir2(l-Dcbp)(cod)2] in different solvents (25.0 ꢁC).
complex and the addendum molecule with the subsequent
product formation is given in the following equation:
½Ir2ðl-DcbpÞðcodÞ ꢁꢀ þ CH3I
2
k1
¢ ½ðcodÞIrIIIMeðl-DcbpÞIrIðcodÞꢁ;
k
ꢀ1
k4
Iꢀ !½ðcodÞðIÞIrIII Meðl-DcbpÞIrIðcodÞꢁꢀ
The formation of the intermediate, in close association
with iodide, is most probably much faster compared to
the second irreversible step (k1[CH3I] + kꢀ1) ꢂ k4 [20]
which results in the following rate law:
Rate ¼ k4K1½Ir2ðl-DcbpÞðcodÞꢀꢁ½CH3Iꢁ
The observed rate constant for the direct interaction be-
tween the metal complex and the addendum reduces to
Eq. (9) under pseudo-first-order conditions
ð7Þ
If Eq. (5) instead of Eq. (6) should hold, then the real
value of k can be determined by kꢀ1 = intercept ꢀ k2.
ꢀ1
For example a graph of kobs versus [CH3I] in acetone has
a slope of 7.0(1) · 10ꢀ2 Mꢀ1 sꢀ1 and an intercept equal to
1.3(2) · 10ꢀ3 sꢀ1 (Table 2). A value of 8(2) · 10ꢀ4 sꢀ1 is
ð8Þ
2
obtained for k when k2 is substituted with the solvolysis
ꢀ1
rate for acetone (4.56(5) · 10ꢀ4 sꢀ1). According to this K1
which is equal to k has a value of 88(22) Mꢀ1 which is sim-
k1
ilar to the spectroꢀp1hotometrically determined value for K1
of 11(2) Mꢀ1 if the error on both values is taken into
account.
kobs ¼ k4K1½CH3Iꢁ
ð9Þ
which results in a straight line for a plot of kobs versus
[CH3I] (see Fig. 3). The intercept which is obtained for
all the different solvents are still attributed to a solvent
pathway and which may include the k2 term (see Eq. (5)).
If this is true, the solvolysis rate constants and intercepts
(Table 2) should be comparable within experimental error.
To strengthen this argument further the following exper-
iment was performed: the complex, (Bu4N)[Ir2(l-Dcbp)-
(cod)2] (5.0 · 10ꢀ4 M) was dissolved in acetone and left
for an hour so that the solvated intermediate could form
(t1/2 = 25 min). This solution was then reacted further with
Table 2
Solvent effects for the oxidative addition of CH3I to Bu4N[Ir2(l-Dcbp)(cod)2] at 25.0 ꢁC
(102) k1 (Mꢀ1 sꢀ1
)
(103) Intercepta (sꢀ1
)
(103) k2 (sꢀ1
)
(103) k
(sꢀ1
)
Kc (Mꢀ1
)
b
a
Solvent
e
ꢀ1
Chloroform
4.9
8.9
10.4
20.7
14.6(4)
8.3(1)
7.7(1)
7.0(1)
4.2(4)
2.1(1)
2.4(1)
1.3(2)
0.84(1)
0.64(3)
0.87(3)
0.456(5)
3.4(4)
1.5(1)
1.5(1)
0.8(2)
43(5)
55(4)
51(3)
88(22)
11(2)d
52(12)
Dichloromethane
1,2-Dichloroethane
Acetone
Acetonitrile
38
3.58(7)
1.20(9)
0.51(2)
0.69(9)
a
k
ꢀ1 + k2.
Solvolysis rate constant.
K = k1/k
b
c
.
ꢀ1
d
Spectrophotometrically.

E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
3805
CH3I (0.015 M) and followed at 425.0 nm. A pseudo-first-
order rate constant of 1.35 · 10ꢀ3 sꢀ1 (=k3[CH3I]) was
obtained. This rate is a factor 3 faster than the rate
obtained for k2 (4.56(5) · 10ꢀ4 sꢀ1) implying that the
pathway. This non-linear orientation was due to steric
crowding of the cod ligand‘s methylene protons above
and below the plane formed by the iridium and the ancil-
lary bidentate ligand atoms. We envisaged the same phe-
nomenon in the present complex in that the rate of initial
nucleophilic attack of the metal upon the methyl carbon
in CH3I is slowed down due to the above-mentioned steric
crowding. This may also be the reason why the solvolysis
reaction becomes more pronounced relative to the oxida-
tive addition rate.
assumption of k3[CH3I] ꢂ k made in the deduction of
ꢀ2
Eq. (4) was justified. The fact that k3[CH3I] P k2 further
strengthens the argument for the existence of the k3
pathway.
This solvent-assisted oxidative addition reaction is fur-
thermore a factor 10 slower than the comparable rate for
the direct pathway under the same conditions. Similar
results were also obtained when this experiment was
repeated with chloroform as solvent. In this case oxidative
addition via the solvent-assisted pathway was a factor 40
slower than that for the direct pathway. Oxidative addition
therefore occurs mainly via the direct pathway especially
since the solvolysis rate constants for acetone and chloro-
form (k2 = 4.56(5) · 10ꢀ4 and 8.4(1) · 10ꢀ4 sꢀ1, respec-
tively) are relatively slow.
¼
¼
The activation parameters, DH and DS , for the oxida-
tive addition of CH3I to (Bu4N)[Ir2(l-Dcbp)(cod)2] (k1
pathway) are 47(2) kJ molꢀ1 and ꢀ110(7) JKꢀ1 molꢀ1
,
respectively. The activation parameters for the reductive
elimination reaction (k pathway) were not determined
ꢀ1
because the intercept values in Table 2 are a combination
of the k and the k2/k3 pathways. The fairly negative
ꢀ1
entropy of activation together with the positive enthalpy
of activation is indicative of an associative process which
is more enthalpy than entropy driven for the k1 pathway.
The enthalpy of activation for the k1 pathway is similar
to the values which was reported for the corresponding
[Ir2(l-pz)2(cod)2] complex [23]. There are however a big dif-
ference in the entropy of activation being zero within
experimental error in the case of the [Ir2(l-pz)2(cod)2] com-
plex (15(25) J Kꢀ1 molꢀ1). Both the entropy and the
enthalpy values of activation for the (Bu4N)[Ir2(l-
Dcbp)(cod)2] complex is also in agreement with the general
activation parameters which was reported for the corre-
sponding monomeric iridium(I) complexes [26].
The intercept, which was obtained from the plots of the
pseudo-first-order rate constants, kobs, versus [CH3I] in
the different solvents (Fig. 3), is therefore a combination of
the solvolysis reaction (k2 pathway) and reductive elimina-
tion (k pathway). The results in Table 2 of the solvent
ꢀ1
effects show that the rate of oxidative addition depends
to a slight degree on the characteristics of the different sol-
vents. The most important being a decrease in rate con-
stants with a simultaneous increase in dielectric constant.
The observed solvent effect could however not be attrib-
uted to the donosity of the solvents.
This is in contrast to the expected solvent effect for oxi-
dative addition reactions. Normally oxidative additions
are characterised by a 20–60-fold increase in the oxidative
addition rate when the solvents are varied [21]. An example
of such an increase in the oxidative addition rate is the 20-
fold increase in the rate constants for the oxidative addition
of CH3I to [Rh(acac)(CO)(PPh3)] in different solvents [22].
A comparison of the solvent effect for the k1 pathway in this
case, with that observed for the oxidative addition of CH3I
to the corresponding [Ir2(l-pz)2(cod)2] complex shows that
the solvent effect in our case is remarkably smaller than the
reported effect [23]. The increase in the oxidative addition
rate in this study is 4-fold while in the case of [Ir2(l-pz)2-
(cod)2] there is a 16-fold increase in the rate constants.
Our results also compare with an increase of approximately
a factor 9 which was observed for the oxidative addition
reaction between the monomeric [Ir(hpt)(cod)] complex
and iodomethane and an increase of a factor 4 for the cor-
responding reaction between [Ir(AnMetha)(cod)] and
iodomethane [18].
Acknowledgements
The authors thank the South African Foundation for
Research Development as well as the Research Fund of
this University for financial support.
References
[1] A.P. Sadimenko, S.S. Basson, Coord. Chem. Rev. 147 (1996) 247.
´
[2] M.P. Garcıa, M.A. Esteruelas, M. Martin, L.A. Oro, J. Organomet.
Chem. 467 (1994) 151.
[3] M. A Esteruelas, M.P. Garcıa, A.M. Lopez, L.A. Oro, Organomet-
allics 10 (1991) 127.
[4] M.P. Garcia, A.M. Lopez, M.A. Esteruelas, F.J. Lahoz, L.A. Oro, J.
Chem. Soc., Chem. Commun. (1988) 793.
[5] M.A. Esteruelas, M. P Garcia, A.M. Lopez, L.A. Oro, Organomet-
´
´
´
´
allics 11 (1992) 702.
[6] K.A. Beveridge, G.W. Bushnell, K.R. Dixon, D.T. Eadie, S.R.
Stobart, J. Am. Chem. Soc. 104 (1982) 920.
[7] A.W. Coleman, D.T. Eadie, S.R. Stobart, J. Am. Chem. Soc. 104
(1982) 922.
The solvent effect for the oxidative addition reactions
observed here is also strongly reminiscent to that of
[Ir(hpt)(cod)] and [Ir(AnMetha)(cod)] [18] where negative
intrinsic volumes of activation, supported by a significant
reduced angle (156ꢁ) for the trans-orientated H3C–Ir(III)-
I moiety in the oxidative addition products [24,25], pointed
[8] J.L. Atwood, K.A. Beveridge, G.W. Bushnell, K.R. Dixon, D.T.
Eadie, S.R. Stobart, M.J. Zaworotko, Inorg. Chem. 23 (1984)
4050.
[9] K.A. Beveridge, G.W. Bushnell, S.R. Stobart, Organometallics 2
(1983) 1447.
[10] D.O.K. Fjeldsted, S.R. Stobart, M.J. Zaworotko, J. Am. Chem. Soc.
107 (1985) 825.
[11] J.A. Bailey, S.L. Grundy, S.R. Stobart, Organometallics 9 (1990) 536.
¼
to a linear SN2 transition state of [Irꢅ ꢅ ꢅCH3–I] for the k1

3806
E. Grobbelaar et al. / Inorganica Chimica Acta 359 (2006) 3800–3806
[12] G.W. Bushnell, D.O.K. Fjeldsted, S.R. Stobart, J. Wang, Organo-
metallics 15 (1996) 3785.
[13] T.G. Schenk, C.R.C. Milne, J.F. Sawyer, B. Bosnich, Inorg. Chem.
24 (1985) 2338.
[14] S.S. Basson, J.G. Leipoldt, W. Purcell, J.B. Schoeman, Inorg. Chim.
Acta 173 (1990) 155.
[15] S.A. Bezman, P.H. Bird, A.R. Fraser, J.A. Osborn, Inorg. Chem. 19
(1980) 3755.
[19] A. Wojcicki, Adv. Organomet. Chem. 11 (1973) 87.
[20] J.H. Espenson, Chemical Kinetics and Reaction Mechanisms, 2nd
ed., McGraw-Hill, New York, 1995, pp. 86–90.
[21] S.S. Basson, J.G. Leipoldt, J.T. Nel, Inorg. Chim. Acta 167 (1984)
84.
[22] J.G. Leipoldt, S.S. Basson, L.J. Botha, Inorg. Chim. Acta 215 (1990)
168.
[23] R.D. Brost, S.R. Stobart, J. Chem. Soc., Chem. Commun. (1989) 498.
[24] S.S. Basson, J.G. Leipoldt, W. Purcell, J.B. Schoeman, Acta
Crystallogr., Sect. C 45 (1989) 2000.
´
[16] J.C. Bayon, G. Net, P. Esteban, P.G. Rasmussen, D.F. Bergstorm,
Inorg. Chem. 30 (1991) 4771.
[17] D.D. Perrin, W.L.F. Armarego, Purification of Laboratory Chemi-
cals, 3rd ed., Pergamon Press, Oxford, 1988.
[25] Y.M. Terblans, S.S. Basson, W. Purcell, G.J. Lamprecht, Acta
Crystallogr., Sect. C 51 (1995) 1748.
[18] M. Theron, E. Grobbelaar, W. Purcell, S.S. Basson, Inorg. Chim.
Acta 358 (2005) 2457.
[26] K.G. van Aswegen, J.G. Leipoldt, I.M. Potgieter, G.J. Lamprecht,
A. Roodt, G.J. van Zyl, Transit. Met. Chem. 16 (1991) 369.