Mechanistic investigation of the oxidative addition reactions between different iridium (I) cyclooctadiene ([Ir(LL′)(cod)]) complexes and iodomethane

Article abstract of DOI:10.1016/j.ica.2004.06.070

The oxidative addition reactions between two different [Ir(cod)(LL′)] complexes (LL′ = hpt and AnMetha) and iodomethane was kinetically investigated. The rate of oxidative addition was determined as 2.2(2) × 10^-2 and 2.69(6) × 10^-2 M^-1 s ^-1 for [Ir(cod)(hpt)] and [Ir(cod)(AnMetha)] in nitromethane respectively. The large negative entropy of activation for the above-mentioned reactions in different solvents clearly point to an associative mechanism. An intrinsic volume of activation of -30.5(3) and -28(3) cm³ mol ^-1 was determined for [Ir(cod)(hpt)] and [Ir(cod)(AnMetha)], respectively. A linear transition state with large charge separation and central ion contraction due to oxidation, contributes to the negative volume of activation.

Full text of DOI:10.1016/j.ica.2004.06.070

Inorganica Chimica Acta 358 (2005) 2457–2463
www.elsevier.com/locate/ica
Mechanistic investigation of the oxidative addition reactions
between different iridium (I) cyclooctadiene
([Ir(LL0)(cod)]) complexes and iodomethane
*
Melanie Theron, Ebeth Grobbelaar, Walter Purcell , Stephen S. Basson
*
Department of Chemistry, University of the Free State, Bloemfontein 9300, South Africa
Received 10 June 2004; accepted 28 June 2004
Available online 12 March 2005
Abstract
The oxidative addition reactions between two different [Ir(cod)(LL0)] complexes (LL0 = hpt and AnMetha) and iodomethane was
kinetically investigated. The rate of oxidative addition was determined as 2.2(2) · 10ꢀ2 and 2.69(6) · 10ꢀ2 Mꢀ1 sꢀ1 for [Ir(cod)(hpt)]
and [Ir(cod)(AnMetha)] in nitromethane respectively. The large negative entropy of activation for the above-mentioned reactions in
different solvents clearly point to an associative mechanism. An intrinsic volume of activation of ꢀ30.5(3) and ꢀ28(3) cm3 molꢀ1
was determined for [Ir(cod)(hpt)] and [Ir(cod)(AnMetha)], respectively. A linear transition state with large charge separation and
central ion contraction due to oxidation, contributes to the negative volume of activation.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Oxidative addition; Iridium; Cyclooctadiene; High pressure kinetics; Volume of activation
1. Introduction
Detailed kinetic studies using a number of addend
substrates like Hg(CN)2[2] and CH3I [3] all have large
negative entropies of activation indicative of the asso-
ciative nature of these oxidative additions. A dispute
which is often raised, especially for those reactions
involving haloalkanes is whether these reactions take
place via a free radical pathway or involve transition
states of a concerted three-centered (I) or two-step
SN2 nature (II).
Oxidative addition reactions of alkyl halides to
square planar Ir(I) and Rh(I) complexes are of general
importance in terms of the formation of metal alkyl
complexes and their role/application in catalysed indus-
trial and other processes. The use of [Ir(CO)Cl(PPh3)2]
(Vaska) in the hydrogenation of alkenes as well as the
polymerisation of styrene and the use of [Ir(cod)(o-
phen)Cl] in the conversion of methanol to acetic acid
are two illustrative examples of the application of irid-
ium (I) complexes in catalytic processes [1]. More than
often oxidative addition forms part of these catalytic cy-
cles and related mechanistic aspects are thus of prime
concern.
≠
δ+
CH
3
δ+
δ-
≠
I
Ir
CH
Ir
3
δ-
I
*
Corresponding authors. Fax: +27 51 430 7805.
E-mail addresses: purcellw.sci@mail.uovs.ac.za (W. Purcell), bas-
sonss.sci@mail.uovs.ac.za (S.S. Basson).
(I)
(II)
0020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.06.070

2458
M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
¼
solvational component, DV solv, due to the creation of
ionic charges during end-on addition of O2 to the
more reactive and coordinatively saturated five-coordi-
nate complex. Similarly, the oxidative addition rate of
CH3I onto [Ir(acac)- (cod)(Br)]ꢀ [3] is faster compared
to its four-coordinate precursor, [Ir(acac)(cod)] in line
with an increased Lewis basicity for the five-coordi-
nate complex [9].
The latter should produce an ionic intermediate
(III) which is difficult to detect and is usually con-
verted to the coordinatively saturated cis- or trans-oxi-
dative addition product in a fast consecutive step.
However, in the reaction of [Pt(Me)2(2,20-Bipy)] [4]
with CH3I
+
-
Ir
CH
We have in recent years investigated the oxidative
addition reactions of [Ir(b-diketone)(cod)] complexes
(b-diketone = 2,4-pentanedionato (acac) or 1,1,1-triflu-
oro-2,4-pentanedionato ligands (tfaa)) in different sol-
vents, as well as the bromide catalysis of the
oxidative addition reaction of CH3I to [Ir(acac)(cod)]
[3]. However, no detail solvent and pressure study
was performed on any of these Ir(I) complexes. It
was decided to synthesise new Ir(I)-cod complexes, con-
taining different bidentate ligands, in order to deter-
mine the effect of the latter variation on the rate of
oxidative addition reactions. In addition, we decided
to perform a pressure and solvent study on these com-
plexes to probe the transition state since the absence of
any carbonyl ligand in these complexes eliminate the
possibility of acyl formation and simplify the resulting
kinetics.
I
3
(III)
in acetonitrile it was possible, using a low temperature
1H NMR technique, to prove the existence of
[Pt(Me)3(2,20-Bipy)(NCCH3)]+Iꢀ at ꢀ40 ꢁC. It is thus
evident that a polar as well as a good nucleophilic sol-
vent may stabilise intermediates of this sort.
The effect of different solvents and elevated pressure
was consequently also studied in order to obtain more
information on the nature of the transition state.
Results from the pressure-dependence kinetic study
of the reaction between [Rh(Sacac)(CO)(PPh3)] [5]
(Sacac = thioacetylacetonato ligand) and CH3I for
example indicated an increase in reaction rate at high-
¼
er pressure and solvent polarity [6], but both DS and
¼
DV varied slightly for the range of solvent polarities
investigated. On the basis of the solvent independence
of DV (ꢀ13.2(5)–18.3(9) cm3 molꢀ1) it was concluded
¼
that this reaction proceeds via a concerted three-
centred transition state (I) since a contradictory and
significant solvent dependency would be expected for
an ion-paired intermediate via a SN2 process. The con-
clusion regarding a three-centred transition state for
the latter reaction is however not a clear cut case since
recent work indicated a more complicated rate con-
stant, which following the oxidative addition also in-
clude a carbonyl insertion step. The reaction between
[Rh(cupf)(CO)(PPh3)] [5] (cupf = cupferrate) and
CH3I indicated no correlation between the observed
rate and the polarity (contrary to the above), but a
2. Experimental
2.1. General considerations
Unless otherwise stated, all the chemicals used were
of reagent grade, and all the preparations were carried
out in air. The sodium salt of 1-hydroxy-2-pyridinethi-
one (hpt) was obtained from Merck Chemicals. The
NMR spectra were obtained at 293 K on a Bruker
300-MHz spectrometer.
¼
large increase in V from ꢀ17.3(8) to ꢀ25.4(7) cm3
2.2. Syntheses
molꢀ1 for non-polar to polar solvents, respectively,
were observed. The authors postulated that these val-
ues are more in agreement with the formation of an
ion-pair (II) intermediate which will be favoured by
polar solvents. The oxidative addition of hydrogen
and iodomethane onto the Vaska complex [7] resulted
2.2.1. N-methyl-4-methoxybenzothiohydroxamic acid
N-methyl-4-methoxybenzothiohydroxamic acid (An-
Metha) ligand was prepared using the procedures previ-
ously described [10,11]. 1H NMR results are reported in
Table 1.
¼
in the intrinsic volumes of activation, DV intr, of ꢀ18
and ꢀ17 cm3 molꢀ1, respectively. These were consid-
ered to be consistent with the simultaneous formation
of two Ir–H bonds for H2 but only one bond for
CH3I corresponding to transition state (II). Volumes
2.2.2. g4-Cycloocta-1, 5-diene(2-pyridinethiolato-N-
oxide-jO,jS) iridium (I), ([Ir(hpt)(cod)]) and g4-
Cycloocta-1,5-diene(N-methyl-4-methoxybenzo-
thiohydroxamato-jO,jS)-iridium (I),
([Ir(AnMetha)(cod)]) were prepared as follows:
About 0.45 mmol [Ir(cod)Cl]2 in 15 cm3 DMF and
0.9 mmol LL0 in 3 cm3 DMF was mixed and stirred
for 2 min. The slow addition of water precipitated a yel-
low product. The centrifuged yellow precipitate was
¼
of activation, DV , of ꢀ31 and ꢀ44 cm3 molꢀ1 for
[Ir(cod) (phen)]+ and [Ir(cod)(phen)(I)] [8] were respec-
tively found for the oxidative addition reactions with
dioxygen in methanol. The larger negative value is
ascribed to an additional contribution from the

M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
2459
Table 1
1H NMR data for different [Ir(LL0)(cod)] complexes in CDCl3
Complex
AnMetha
Cod (ppm)
LL0 (ppm)
CH3 (ppm)
d 6.88, 7.37 (CH), d 3.57 (CH3), d 3.82 (OCH3),
d 10.88 (OH)
[Ir(hpt)(cod)]
d 1.70, 1.82, 2.20, 2.24 (CH2)
d 6.85, 7.25, 7.88, 8.34 (CH), d 4.02, 4.28 (CH)
6.92, 7.27 (CH), d 3.82(OCH3), d 3.71 (CH3)
d 6.85, 7.24, 7.88, 8.33 (CH)
[Ir(AnMetha)(cod)]
[Ir(hpt)(cod)(CH3)I]a
[Ir(AnMetha)(cod)(CH3)I]a
a
d 1.8, 2.25 (CH2), d 3.96, 4.30 (CH)
d 1.72, 1.83, 2.21, 2.24 (CH2), d 4.04, 4.30 (CH)
d 1.81, 2.27 (CH2), d 3.98, 4.31 (CH)
d 2.17
d 2.20
d 6.93, 7.29 (CH), d 3.84(OCH3), d 3.72 (CH3)
Results obtained by subtracting the starting material spectra.
redissolved in warm acetone and precipitated by the
slow addition of water. The bright yellow product was
dried over P2O5 in a vacuum desiccator for ca. 15 h.
Yield: 233 mg, 63% ([Ir(hpt)(cod)]) and 312 mg, 70%
([Ir(AnMetha)(cod)]). H NMR results are reported in
Table 1.
followed at lower pressure (between 1 and 110 MPa
for all the other solvents) due to the crystallisation of
benzene at pressures exceeding 90 MPa. Because all
the plots of kobsd vs. [CH3I] gave non-zero intercepts,
a complete concentration variation was performed at
each pressure in order to obtain the correct volume of
activation for the forward and reverse rate. All reactions
were performed under pseudo-first-order conditions
with CH3I at least in tenfold excess. Typical experimen-
tal conditions were [Ir] = 1.0 · 10ꢀ3 M and [CH3I]
varied between 1.0 · 10ꢀ2 and 0.5 M. The observed
first-order rate constant were calculated from plots of
1
Both complexes were further characterised by means
of X-ray structure determinations [12,13].
2.2.3. (g4-Cycloocta-1, 5-diene)iodomethyl(2-
pyridinethiolato-N-oxide-jO,jS) iridium (III),
([Ir(hpt)(cod)(CH3)I]) and (g4-Cycloocta-1,5-
diene)iodomethyl(N-methyl-4-methoxybenzo-
ln(At · A ) versus time, where At and A are the absor-
1
1
bencies at time t and infinity, respectively. These plots
were linear for at least 2–3 half-lives of the reaction. A
non-linear least-squares program was used for the
calculations.
thiohydroxamato-jO,jS) iridium (III),
([Ir(AnMetha)(cod)(CH3)I]) were prepared as follows:
About 0.2 mmol of the iridium complex and 6 mmol
of CH3I (30· excess) was dissolved in 10 cm3 acetone.
The solution was heated after which it was left at room
temperature to precipitate. 1H NMR results are re-
ported in Table 1.
2.4. Reaction volumes and qp-values
Isolation of the product after the oxidative addition
reaction yielded a mixture of the starting material as
well as the product, confirming the existence of the equi-
librium that exists between reactant and product and
which is clearly illustrated by the different kinetic runs.
The mixture of reactant and product existed even in
the presence of a large excess of CH3I. The NMR results
for the product were obtained by subtracting or elimi-
nating the reactant spectra.
A locally manufactured Carlsberg dilatometer [14]
was used to determine the reaction volume for the reac-
tions between [Ir(LL0)(cod)] and CH3I. The dilatometer
which had a capillary with diameter = 0.5 mm was cali-
brated with double distilled water. The molar volume of
water was calculated using q-values calculated at differ-
ent temperatures (21–30 ꢁC) [15].
The change in molar volume (DVm) between two tem-
peratures was calculated by difference.
The constants for this dilatometer were calculated
from the straight line obtained from a plot of DVm vs.
change in capillary height using a non-linear least
squares program. Ethyleneglycol was used as solvent
in the temperature bath that enabled better temperature
control ( 0.01 ꢁC).
The qp-values (25 ꢁC) used in this study were ob-
tained from a linear regression calculation using three
values at different temperatures that were obtained from
[16]. The calculated qp-values (10ꢀ6 barꢀ1) at 25 ꢁC are
as follows: nitromethane, 1.698; acetone, 5.647; 1,2-
dichloroethane, 11.093; chloroform, 14.983; benzene,
16.263. The Kirkwood model [17], represented in Eq.
2.3. Kinetics
All the solvents used during the kinetic study were
purified and dried prior to its use. All the kinetic mea-
surements at atmospheric pressure were performed on
a GBC (model 916) double-beam UV/VIS spectropho-
tometer equipped with a thermostat cell holder (0.01
ꢁC). The high-pressure kinetics was performed on a
GBC (model 916) double-beam UV/VIS spectropho-
tometer equipped with a thermostat high-pressure cell.
The oxidative addition reactions were followed at ca.
400 nm for acetone, 1,2-dichloroethane, chloroform
and benzene and at 427 nm for nitromethane. The reac-
tion between [Ir(LL0)(cod)] and CH3I in benzene were
(2), was used to determine the solvent influence on
¼
DV solv
:

2460
M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
¼
¼
¼
ð1Þ
ð2Þ
DV ¼ DV intr þ DV solv
¼
¼ DV intr ꢀ ðN0Du2=r2Þqp:
3. Results and discussion
The oxidative addition reactions of the two different
[Ir(cod)(LL0)] complexes can be presented as follows:
k1
½IrðcodÞðLL0Þꢁ þ CH3I ꢀ ½IrðcodÞðLL0ÞꢁðCH3ÞIꢁ
ð3Þ
k
ꢀ1
Under the selected conditions these reactions gave the
expected second-order kinetics and the observed rate
constant, kobsd, depended linear on the iodomethane
concentration, see Fig. 1, i.e.,
kobsd ¼ k1½CH3Iꢁ þ k
:
ð4Þ
ꢀ1
Solvents with a large variety of dielectric constants (ꢀ)
were used in this study in order to try and find some
correlation between the volume of activation and the
solvent properties. The values of k as a function of tem-
perature, solvent and pressure are reported in Table 2.
The results in Table 2 clearly show that the rate of
oxidative addition depends to a certain degree on the
solvent in which it was performed. An increase of
Fig. 1. kobsd versus [CH3I] for the oxidative addition reaction of CH3I
with [Ir(hpt)(cod)] in acetone, k = 401 nm.
approximately a factor 9 was observed for the oxidative
addition reaction between [Ir(hpt)(cod)] and iodome-
thane as the solvents were changed from non-polar
Table 2
Summary of the kinetic and activation parameters for the oxidative addition reactions between the [Ir(LL0)(cod)] complexes and CH3I at 25 ꢁC with
LL = hpt and AnMetha
Solvent
Nitromethane
38.3
Acetone
20.7
1,2-Dichloroethane
Chloroform
4.9
Benzene
2.3
ꢀ
10.4
qp · 106 (25 ꢁC)
1.70
5.65
11.09
14.98
16.26
[Ir(hpt)(cod)]
k1 · 102 (Mꢀ1 sꢀ1 a
)
2.2(2)
0.7(1)
32(7)
49(5)
0.7(2)
0.85(5)
0.5(1)
16(4)
0.66(3)
1.42(3)
4.7(3)
53(1)
0.189(6)
0.60(2)
3.2(1)
k
ꢀ1 · 103 (sꢀ1 a
)
0.47(2)
14.7(7)b
53(4)
K (=k1/kꢀ1) (Mꢀ1
)
¼
DH1 ðkJ molꢀ1
Þ
46(4)
57(6)
¼
DS1 ðJ Kꢀ1 molꢀ1
Þ
ꢀ111(16)
5(39)
ꢀ109(5)
ꢀ131(13)
61(18)
ꢀ110(4)
ꢀ14.3(3)
ꢀ20(2)
21(1)
–
ꢀ107(19)
¼
DSꢀ1 ðJ Kꢀ1 molꢀ1
Þ
ꢀ50(7)
–
DV (k1) (cm3 molꢀ1
)
ꢀ30(3)
ꢀ79(20)
ꢀ74(7)
ꢀ30.5(3)
ꢀ26(3)
ꢀ22(2)
ꢀ18(1)
¼
DV (kꢀ1) (cm3 molꢀ1
)
–
–
–
–
–
–
–
–
–
¼
ꢀ1
3
ꢀ
DV ðcm mol
Þ
¼
DV intr ðcm3 molꢀ1
Þ
–
[Ir(AnMetha)(cod)]
k1 · 102 (Mꢀ1 sꢀ1 a
)
2.7(6)
0.41(19)
65(30)
59(4)
–
0.94(1)
0.97(4)
9.7(4)c
40(2)
1.10(1)
1.40(4)
7.9(2)
0.530(5)
2.24(2)
2.37(3)
59(4)
0.553(1)
0.63(4)
8.8(6)
k
ꢀ1 · 103 (sꢀ1 a
)
K (=k1/kꢀ1) (Mꢀ1
)
¼
DH1 ðkJ molꢀ1
Þ
57(0.5)
86(7)
59(1)
¼
DHꢀ1 ðkJ molꢀ1
Þ
64(4)
89(4)
78(2.5)
ꢀ96(4)
ꢀ44(8)
¼
DS1 ðJ Kꢀ1 molꢀ1
Þ
ꢀ76(13)
–
ꢀ152(8)
ꢀ89(12)
ꢀ92(2)
ꢀ10(23)
ꢀ91(13)
0.82(3)
¼
DSꢀ1 ðJ Kꢀ1 molꢀ1
Þ
DV (k1) (cm3ꢀm1 olꢀ1
)
ꢀ28(3)
–
–
–
–
–
–
–
–
¼
3
ꢀ
DV ðcm mol
Þ
ꢀ68(6)
a
Subscript 1 refer to forward reaction (k1) and ꢀ1 to reverse reaction (kꢀ1).
b
c
K = 9(1) Mꢀ1 spectrophotometric.
K = 23(3) Mꢀ1 spectrophotometric.

M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
2461
(benzene) to polar solvents (nitromethane) (see Fig. 2).
An increase of only a factor 4 were observed for the cor-
responding reaction between [Ir(AnMetha)(cod)] and
iodomethane. These results suggests that charge separa-
tion in the transition state is better assisted by the polar
solvents, i.e., linear transition state. This increase in the
forward rate constant, k1, is relative small compared to
the increase which was observed for the same reaction
between [Rh(Sacac)(CO)(PPh3)] and iodomethane [6].
A 20-fold increase was observed for this reaction when
it was performed in more polar solvents. The summary
of kinetic results for different Ir(I) complexes listed in
Table 3 show some important tendencies. The rate con-
stants cited are the ones that were done in the most polar
solvents. The following tendency was observed for the
different bidentate ligands [9]:
The large negative entropy of activation obtained for
the reactions of both Ir(I) complexes in all the different
solvents is indicative of an associative mechanism.
Large negative volumes of activation were obtained
for these reactions in the different solvents that are also
indicative that these reactions occur via an associative
mechanism. A steady increase in the volume of activa-
tion from non-polar to polar solvents was observed.
The plot of volume of activation vs. the calculated qp
(effect of pressure on the dielectric constant) values is
presented in Fig. 3 which clearly indicates a linear rela-
tionship between the volume of activation and the sol-
vent polarity. The straight line according to Eq. (2)
has an intercept which represents the intrinsic volume
of activation for the reaction. The results in Fig. 3
clearly indicates that the same transition state exists
within all the solvents used in this study and that the
reaction between [Ir(hpt)(cod)] and CH3I has an intrin-
macsm > AnMetha > hpt > Sacac > tfaa > cupf
sic volume of activation of ꢀ30.5(3) cm3 molꢀ1
.
This tendency is attributed to the electron donor abil-
ity of the ‘‘bite atoms’’ of the bidentate ligands bonded
to the metal ion. These ‘‘bite atoms’’ increase the elec-
tron density on the metal ion, i.e., increase in the Lewis
basicity, and thereby increase the rate of oxidative addi-
tion for these complexes.
Predictions about the transition state of these reac-
tions can be made by using the intrinsic volume of acti-
vation value of ꢀ30.5 cm3 molꢀ1for [Ir(hpt)(cod)] and
Fig. 3. The volume of activation versus the calculated qp for the
oxidative addition reaction of CH3I with [Ir(hpt)(cod)] at 25 ꢁC in the
different solvents.
Fig. 2. kobsd versus [CH3I] for the oxidative addition reaction of CH3I
with [Ir(hpt)(cod)] in acetone, benzene and nitromethane.
Table 3
Summary of the second-order rate constants for the oxidative addition reactions of different [Ir(LL0)(cod)] complexes with CH3I
Solvent
Donor atom
Ring size
(103) k1 (Mꢀ1 sꢀ1
)
Solvent
Ref.
[Ir(macsm)(cod)]
[Ir(AnMetha)(cod)]
[Ir(hpt)(cod)]
S–N
S–O
S–O
S–O
O–O
O–O
6
5
5
6
6
6
28.4(7)
26.9(6)
22(2)
acetonitrile
nitromethane
nitromethane
acetonitrile
acetonitrile
acetonitrile
[9]
this study
this study
[Ir(Sacac)(cod)]
[Ir(tfaa)(cod)]
[Ir(cupf)(cod)]
5.2(3)
4.4(2)
2.17(7)
[9]
[9]
[9]

2462
M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
ꢀ28(2) cm3 molꢀ1 for [Ir(AnMetha)(cod)]. The results in
Table 2 show the solvation component of the volume of
activation for the different solvents. For the reaction in
benzene, a positive value of +12.5 cm3 molꢀ1 is needed
to give an overall volume of activation of ꢀ18
cm3 molꢀ1. This solvation effect is probably due to the
decrease in non-polar solvent interaction with a highly
polar trigonal bipyramidal transition state with its large
charge separation during the formation of the transition
state, i.e., linear transition state.
(cod)(CH3)I] [20] clearly indicated the trans orientation
of the methyl and iodide in the structures. The I–Ir–
CH3 angle of 156ꢁ in both structures was within experi-
mental error the same and deviated significantly from
the expected 180ꢁ. This large deviation is attributed to
repulsion between the two trans ligands and the four
protons bonded to the four ethylene carbons of the
cod ligand. This repulsion between the cod and the
incoming CH3I as well as the relative inflexible geometry
of the four coordinated complex ([Ir(hpt)(cod)]), make a
three centre transition state between CH3I and Ir in the
transition state highly unlikely. The more favourable
transition state is therefor the linear SN2 transition state.
The SN2 intrinsic mechanism is also more favoured for
oxidative addition reactions as indicated by ab initio cal-
culations performed by Griffin et al. [21] for the reac-
tions between [M(CO)2I2]ꢀ (M = Rh, Ir) and CH3I.
The relative large positive value for the volume of acti-
ꢀ
The reaction volume ðDV Þ for the reaction between
[Ir(hpt)(cod)] and iodomethane was determined as
ꢀ74(7) cm3 molꢀ1. Due to the large experimental error
on the volume of activation for the reverse reaction
¼
(DV = ꢀ79(20) cm3 molꢀ1) no useful deduction could
be made from these values. Using the reaction volume
as well as the volume of activation and Eq. (5) the acti-
vation volume for the reverse reaction could be calcu-
lated [18]:
vation (+44 cm3 molꢀ1) of the reverse reaction (k
)
ꢀ1
are explained in terms of the breaking of two bonds
(Ir–I and Ir–CH3) as well as the volume increase due
to the reduction of the Ir(III) to Ir(I) for the reverse
reaction (reductive elimination) which takes place via a
dissociative mechanism.
¼
¼
DV ¼ DV forward ꢀ DV reverse
;
ð5Þ
ꢀ1
DV reverse ¼ þ44 cm3 mol
:
¼
The corresponding value for [Ir(AnMetha)(cod)] is
calculated as +40 cm3 molꢀ1. The volume profile for
the reaction between [Ir(hpt)(cod)] and CH3I is given
in Fig. 4, indicating the slightly asymmetric nature of
the intrinsic and solvational reorganisation in the transi-
tion state.
A possible explanation for the negative intrinsic vol-
ume of activation for both complexes can be attributed
to both electronic and bonding factors present in the
transition state. The combined effect of the nuclear con-
traction due to the oxidation of Ir(I) to Ir(III) during the
reaction as well as the new bond formation between the
iridium centre and the methyl group of CH3I adequately
explains the magnitude of the negative intrinsic volume
of activation for the forward reaction. The crystal struc-
tures of both [Ir(acac)(cod)(CH3)I] [19] and [Ir(Sacac)
A linear relationship between the volume of activa-
tion and qp values was also observed by Steiger and
Kelm for the oxidative addition reaction between [Ir(-
CO)(Cl)(PPh3)2] and CH3I [7]. A much smaller intrinsic
¼
volume of activation, DV = ꢀ17 cm3 molꢀ1, was ob-
tained for this reaction. It is interesting to note that
the contribution of the solvolysis component to the vol-
ume of activation was also the largest for the non-polar
solvents. In this case, contrary to our study, they ob-
served an increase in negative activation volume for
the non-polar solvents which they attributed to the for-
mation of a linear transition state. A volume of activa-
tion of ꢀ44 cm3 molꢀ1 was obtained for the cis
addition of O2 to the [Ir(phen)(cod)]+ [8] and it is well
known that this reaction takes place via a concerted
three-centre mechanism.
Acknowledgements
The authors thank the Research Fund of this Univer-
sity for financial support.
References
[1] R.S. Dickson, Homogeneous Catalysis with Compounds of
Rhodium and Iridium, D. Reidel Publishing Company, Dordr-
echt, 1985.
[2] G.J.J. Steyn, S.S. Basson, G.J. Leipoldt, G.J. van Zyl, J.
Organomet. Chem. 418 (1991) 113.
[3] S.S. Basson, J.G. Leipoldt, W. Purcell, J.B. Schoeman, Inorg.
Chim. Acta 173 (1990) 155.
Fig. 4. Volume profile for the reaction of [Ir(hpt)(cod)] with CH3I.

M. Theron et al. / Inorganica Chimica Acta 358 (2005) 2457–2463
2463
[4] C. Ducker-Benfer, R. van Eldik, A.J. Canty, Organometallics 13
¨
(1994) 2412.
[14] R. van Eldik, Inorganic High Pressure Chemistry; Kinetics and
Mechanism, Elsevier Science Publishers B.V., Amsterdam,
1986.
[5] J.A. Venter, J.G. Leipoldt, R. van Eldik, Inorg. Chem. 30 (1991)
2207.
[15] R.C. Weast (Ed.), CRC Handbook of Chemistry and Physics,
60th ed., CRC Press, Boca Raton, FL, 1980.
[16] N. Isaacs, Liquid Phase High Pressure Chemistry, Wiley, Chich-
ester, 1981.
[6] J.G. Leipoldt, S.S. Basson, L.J. Botha, Inorg. Chim. Acta 168
(1990) 215.
[7] H. Steiger, H. Kelm, J. Phys. Chem. 77 (1973) 290.
[8] D.J.A. de Waal, T.I.A. Gerber, W.J. Louw, R. van Eldik, Inorg.
Chem. 21 (1982) 2002.
[17] G.J. Kirkwood, J. Chem. Phys. 2 (1934) 351.
[18] R. Mohr, L.A. Mietta, Y. Ducommun, R. van Eldik, Inorg.
Chem. 24 (1985) 757.
[9] M. Theron, M.Sc. Thesis, Univ. of the Free State, Bloemfontein,
1993.
[19] S.S. Basson, J.G. Leipoldt, W. Purcell, J.B. Schoeman, Acta
Cryst. C45 (1989) 2000.
[10] K.A. Jensen, O. Buchardt, C. Chrisopheson, Acta Chem. Scand.
21 (1967) 1936.
[20] Y.M. Terblans, S.S. Basson, W. Purcell, G.J. Lamprecht, Acta
Cryst. C51 (1995) 1748.
[11] K.A. Jensen, C. Pedeson, Acta Chem. Scand. 15 (1961) 1087.
[12] M. Theron, W. Purcell, S.S. Basson, Acta Cryst. C52 (1996) 336.
[13] M. Theron, W. Purcell, S.S. Basson, Acta Cryst. C51 (1995) 1750.
[21] T.R. Griffin, D.B. Cook, A. Haynes, J.M. Pearson, D. Monti,
G.E. Morris, J. Am. Chem. Soc. 118 (1996) 3029.