Characterization and oxidative addition reactions for iridium cod complexes

Article abstract of DOI:10.1080/00958972.2016.1261402

Three different [Ir(LL′)(cod)] complexes (LL′?=?N-aryl-N-nitrosohydroxylaminato) (cupf), trifluoroacetylacetonato (tfaa), and (methyl 2-(methylamino)-1-cyclopentene-1-dithiocarboxylato-κN,κS) (macsm)) were synthesized, characterized, and their rates of oxidative addition with methyl iodide were determined. Formation of an isosbestic point during the oxidative addition of methyl iodide with the complexes containing tfaa and cupf as bidentate ligands indicated formation of only one product, while an increase in absorbance maximum observed for macsm confirms that the same reaction between the complex and methyl iodide occurs. Kinetic results for all complexes, except [Ir(tfaa)(cod)], showed simple second-order kinetics with a zero intercept (within experimental error). Rates of oxidative addition for bidentate ligands in acetonitrile showed an increase of an order of magnitude with a change in the type of bidentate ligands. Computational chemistry using density functional theory calculations showed that the oxidative addition reaction proceeds through a “linear” transition state with the methyl iodide unit tilted towards the LL′-bidentate ligand.

Full text of DOI:10.1080/00958972.2016.1261402

Journal of Coordination Chemistry
ISSN: 0095-8972 (Print) 1029-0389 (Online) Journal homepage: http://www.tandfonline.com/loi/gcoo20
Characterization and oxidative addition reactions
for iridium cod complexes
W. Purcell, J. Conradie, S. Kumar & J. A. Venter
To cite this article: W. Purcell, J. Conradie, S. Kumar & J. A. Venter (2017) Characterization and
oxidative addition reactions for iridium cod complexes, Journal of Coordination Chemistry,
70:1, 10-24, DOI: 10.1080/00958972.2016.1261402
To link to this article: http://dx.doi.org/10.1080/00958972.2016.1261402
Accepted author version posted online: 15
Nov 2016.
Published online: 29 Nov 2016.
Submit your article to this journal
Article views: 16
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gcoo20
Download by: [FU Berlin]
Date: 01 January 2017, At: 05:59

Journal of Coordination Chemistry, 2017
Vol. 70, no. 1, 10–24
http://dx.doi.org/10.1080/00958972.2016.1261402
Characterization and oxidative addition reactions for iridium
cod complexes
W. Purcell, J. Conradie, S. Kumar and J. A. Venter
department of Chemistry, university of the free state, Bloemfontein, south africa
ABSTRACT
ARTICLE HISTORY
received 26 July 2016
accepted 3 october 2016
Three different [Ir(LL′)(cod)] complexes (LL′
=
N-aryl-N-
nitrosohydroxylaminato) (cupf), trifluoroacetylacetonato (tfaa),
and (methyl 2-(methylamino)-1-cyclopentene-1-dithiocarboxylato-
κN,κS) (macsm)) were synthesized, characterized, and their rates of
oxidative addition with methyl iodide were determined. Formation
of an isosbestic point during the oxidative addition of methyl iodide
with the complexes containing tfaa and cupf as bidentate ligands
indicated formation of only one product, while an increase in
absorbance maximum observed for macsm confirms that the same
reaction between the complex and methyl iodide occurs. Kinetic
results for all complexes, except [Ir(tfaa)(cod)], showed simple second-
order kinetics with a zero intercept (within experimental error). Rates
of oxidative addition for bidentate ligands in acetonitrile showed
an increase of an order of magnitude with a change in the type of
bidentate ligands. Computational chemistry using density functional
theory calculations showed that the oxidative addition reaction
proceeds through a “linear” transition state with the methyl iodide
unit tilted towards the LL′-bidentate ligand.
KEYWORDS
iridium(i); bidentate ligands;
cyclooctadiene; oxidative
addition; methyl iodide
1. Introduction
The use of iridium complexes as catalysts in the synthesis of large-scale organic compounds
are relatively new compared to other metals. [IrI2(CO)2]− is used by British Petroleum in its
well-known Cativa process [1–4] for the production of acetic acid from methanol, while an
Ir-xyliphos complex is used by Novartis [5–7] to synthesize the herbicide (S)-metolachlor.
Another very important iridium catalyst is the Crabtree catalyst ([Ir(cod)(PCy3)(py)][PF6]) [8,
9], which showed a 100 times increase in hydrogenation activity compared to the Wilkinson
catalyst ([Rh(PPh3)3Cl]) and even allowed for hydrogenation of tri- and tetra-substituted
CONTACT W. Purcell
purcellw@ufs.ac.za
© 2016 informa uK limited, trading as taylor & francis Group

JOURNAL OF COORDINATION CHEMISTRY
11
alkenes. The complex also showed excellent stereoselectivity in reduction of exocyclic meth-
ylene groups compared to more traditional catalysts such as [Rh(ndb)(dppb)]BF4
(dppb = 1,4-bis(diphenylphosphino)butane), [Rh(PPh3)3Cl], and Pd/C [10].
The iridium(I) cyclooctadiene dimer [Ir(cod)Cl]2 is prepared with reaction between
Na2[IrCl6]·6H2O or H2[IrCl6]·6H2O and cycloocta-1,5-diene [11, 12]. This iridium organometallic
complex can undergo bridge-splitting and cod substitution reactions to synthesize a number
of new iridium organometallic complexes or the complex can also facilitate oxidative addition
reactions. [Ir(cod){(P(OPh)3}2]+ [13] is an example of a product formed after a bridge-splitting
reaction while isolation of [IrCl(CO)2(PPh3)] [12] is an example of a cod substitution reaction.
Examples of oxidative addition reactions include the direct reaction of HCl or HBr to produce
[Ir(cod)HCl2]2 [14] and [Ir(cod)HBr2]2 [15].
Oxidative addition is a key reaction in the cyclic catalytic conversion of methanol to acetic
acid in both the Monsanto and the Cativa processes [16]. This reaction is followed by methyl
migration, reductive elimination of the acetyl iodide, and the final step is its reaction with
water to produce acetic acid and the original metal complex is regenerated. The rate and
the mechanism of the oxidative addition step for both Rh and Ir square planar complexes
are influenced by electronic (type, ring size and atom combination of bidendate ligands),
steric (cone angle of phosphine ligands), and solvent effects. The spontaneous methyl migra-
tion step observed [17, 18] for most of the complexes complicate the kinetic study of these
reactions with the formation of different isomers in solution, the presence of equilibria and
solvent-assisted reaction pathways [19].
Kinetic study of the oxidative addition reactions of [Ir(L,L′)(cod)] complexes would simplify
the overall reaction with no CO insertion taking place, the formation of only one isomer and
simplified mechanisms with possible zero intercepts. A kinetic study of the oxidative addition
reaction between [Ir(sasac)(cod)] and CH3I in different solvents produces simplified kinetics [20].
In this study, three different iridium(I) cod complexes [Ir(LL′)(cod)] were synthesized, with
LL′ = N-aryl-N-nitrosohydroxoylaminato (cupf)), trifluoroacetylacetonato (tfaa), and (methyl
2-(methylamino)-1-cyclopentene-1-dithiocarboxylato-κN,κS) (macsm) (see scheme 1), to
determine the electronic influence of different donor ligands on the oxidative addition step
for these complexes. A computational chemistry study on the geometries and energies of
the reactants, transition state and possible products of the [Ir(LL′)(cod)] + CH3I reaction, shed
further light on the oxidative addition process.
2. Experimental
2.1. General considerations
Unless otherwise stated, all chemicals were reagent grade and used without purification
and all preparations were carried out in air. The bidendate ligands cupf
(N-aryl-N-nitrosohydroxylaminato) and tfaa (1,1,1-trifluoropentane-2,4-dione) were com-
mercially obtained and used without purification. IR spectra were recorded with a Hitachi
270–50 spectrophotometer while the NMR spectra (TMS as internal standard) were obtained
at 293 K on a Bruker 600 MHz spectrometer. The methyl iodide was stabilized by silver foil
to prevent decomposition and used in a well-ventilated fume cupboard. The CHN analyses
were performed on a LECOTruspec micro-analyzer, while the iridium analysis was performed
on a Shimadzu ICPS-7510 ICP-OES sequential plasma spectrometer.

12
W. PURCELL ET AL.
(a)
(b)
(c)
CF3
S
O
Ir
S
N
O
O
N
N
Ir
Ir
O
Ph
Scheme 1. different iridium(i) cod complexes synthesized in the study: (a) [ir(cupf)(cod)], (b) [ir(tfaa)
(cod)] and (c) [ir(macsm)(cod)].
The UV/Visible spectra and kinetic runs were performed on a Hitachi 150–20 UV/Visible
spectrophotometer in a thermostatically controlled holder (0.1 °C), which has a capacity of
6 cells. The solvents used for the kinetic runs were all purified and dried using the prescribed
methods [21]. All the complexes were tested for stability in the different solvents prior to
the kinetic runs. Typical experimental conditions were [M(cod)(LL′)] = 5.0 × 10−4 M, and [CH3I]
varied between 0.075 and 0.75 M, which ensured good pseudo-first-order kinetics. The
observed first-order rate constants (kobs) were calculated for at least two half-lives from the
above plots using the equation At = A∞ + (A0 − A∞)e(−kobs*t) with A0, At, and A∞ the absorbance
at time 0, t and infinity, respectively, using a nonlinear least squares program [22].
2.2. Synthesis
Slight variations of previously published synthetic routes [23] were introduced in the prepa-
ration of macsm.
2.2.1. (Methyl 2-(methylamino)-1-cyclopentene-1-dithiocarboxylato-κN,κS) (macsm)
[23]
Cyclopentanone (4.2 g, 0.05 mol) was mixed with an aqueous methylamine (31 g, 0.2 mol)
solution. The mixture was cooled to 5 °C and stirred for 120 min after which carbon disulfide
(4.6 g, 0.06 mol) was gradually introduced and stirred for another 120 min. The yellow pre-
cipitate was removed by filtration and washed with water. The dried product was added to
8-mL concentrated acetic acid and heated to 50 °C for a few minutes. The mixture was cooled
to room temperature and 20 mL of water was introduced. The yellow solid was then removed
by filtration, washed with water and dried, yield 25%. About 3.6 g of this product was dis-
solved in 6-mL 40% aqueous methyl amine solution, 30-mL methanol was added and
refluxed for 180 min. The volume of the solution was reduced under reduced pressure, the
residue added to 70 mL of water and the precipitate removed by precipitation. Hydrochloric
acid (10%) was added to the filtrate, which resulted in formation of a yellow solid product.
About 1 g of this product was dissolved in 5 mL of 6 M NaOH, an excess (≈1 mL) of methyl
iodide was slowly added with constant stirring for 2 h resulting in formation of a yellow solid
product, which was filtered and dried. Yield 58%, M.P 127–129 °C, Elemental analysis for
C8H13NS2 (Mr = 187.05): Found C, 51.34; H, 6.89; N, 7.32; S, 34.10%; Calcd C, 51.29; H, 6.99; N,
7.48; S, 34.23%; IR spectra: ν = 2971 (C–H, str), 1598 (C=N), 613 (C–S) cm−1, 1H NMR (600 MHz),
CDCl3, 20 °C: δ 1.89 (2H, m), 2.59 (3H, s), 2.69 (2H, m), 2.82 (2H, m), 3.07 (3H, s) ppm. 13C NMR
(600 MHz), CDCl3, 20 °C: δ 16.6, 20.6, 31.4, 33.0, 33.4, 118.3, 170.1, 197.3.

JOURNAL OF COORDINATION CHEMISTRY
13
2.2.2. (η4-1,5-Cyclooctadiene-N-aryl-N-nitrosohydroxylaminato)iridium(I), [Ir(cupf)
(cod)]
[IrCl(cod)]2 (0.5 g, 0.9 mmol) was dissolved in 20-mL dimethyl formamide (DMF). Cupf (0.45 g,
2.5 mmol) was dissolved in 6-mL DMF and slowly added while continuously stirring the
solution. The color changed from yellow to bright red. Water was added dropwise and a
yellow product precipitated from the solution, which was removed with centrifuging. The
product was then redissolved in minimum acetone and re-precipitated with the addition of
water and isolated with centrifuging. The yellow product was washed with water and dried
over P2O5 in a vacuum desiccator. Yield: 68%; IR spectra: ν = 3080 (Ar-C-H, str), 749 (=C–H
bending), 1455 (N–N=O) cm−1; MS; m/z 437.23 [M]+, 438.28 [M + H]+; Elemental analysis for
C14H17IrN2O2 (Mr = 437.52): Found C, 38.21; H, 3.89; N, 6.37%; Calcd C, 38.43; H, 3.92; N, 6.40%;
1H NMR (600 MHz), CDCl3, 20 °C: δ 1.74 (4H, m), 2.25 (4H, m), 4.24 (2H, t), 4.30 (2H, t), 7.47 (1H,
m), 7.49, (2H, m), 7.98 (2H, m), ppm. 13C NMR (600 MHz), CDCl3, 20 °C: δ 30.8, 31.0, 57.4, 58.4,
121.1, 129.4, 130.7, 138.2.
2.2.3. (η4–1,5-Cyclooctadiene)(iodo)(methyl)(N-aryl-N-nitrosohydroxylaminato)
iridium(III), [Ir(cupf)(cod)(CH3)I]
An excess of CH3I (10 mL) and [Ir(cupf)(cod)] (0.2 g, 0.47 mmol) was mixed in 2 mL acetone.
The solution was sealed for 60 min. The seal was removed and the solution was concentrated
in a vacuum desiccator. The concentrated solution was kept at −10 °C and an orange pre-
cipitate was isolated from the solution. Yield: 63%; IR spectra: ν = 3054 (Ar–C–H, str), 766
(=C–H bending), 1460 (N-N=O) cm−1. MS; m/z 579.50 [M]+, 580.01 [M + H]+; Elemental analysis
for C15H20IIrN2O2 (Mr = 579.45): Found C, 31.64; H, 3.46; N, 4.79%; Calcd C, 31.09; H, 3.48; N,
1
4.83%; H NMR (600 MHz), CDCl3, 20 °C; δ 1.39 (1H, m), 1.48 (1H, m), 1.70 (1H, m), 2.40
(1H, m), 2.47 (1H, m,), 2.49 (1H, m), 2.49 (1H, m), 2.58(1H, m), 3.01 (3H, s), 4.39 (1H, m), 4.61
(1H, m), 4.72 (1H, m), 5.40 (1H, m), 7.48 (1H, m), 7.52 (2H, m), 8.06 (2H, m). 13C NMR (600 MHz),
CDCl3, 20 °C; δ 8.6, 28.8, 29.0, 30.2, 33.7, 77.2, 79.0, 95.2, 97.1, 121.8, 129.4, 130.7, 139.1.
2.2.4. (η4–1,5-Cyclooctadiene-trifluoroacatylacetonato)iridium(I), [Ir(tfaa)(cod)]
The procedure for synthesis of [Ir(tfaa)(cod)] was identical to that of [Ir(cupf)(cod)]. Yield:
61%; IR spectra: ν = 2945 (C–H str), 774 (=C–H bending), 1302 (C–F), 1142 (C–O) cm−1; MS;
m/z 453.16 [M]+, 454.13 [M + H]+; Elemental analysis for C13H16F3IrO2 (Mr = 453.5): Found C,
1
34.01; H, 3.84%; Calcd C, 34.39; H, 3.56%; H NMR, (600 MHz), CDCl3, 20 °C: δ 1.60 (4H, m),
2.07 (3H, s), 2.19 (4H, m), 4.10 (2H, m), 4.14 (2H, m), 5.87 (1H, s) ppm. 13C NMR (600 MHz),
CDCl3, 20 °C: δ 29.03, 31.03, 31.10, 60.34, 61.13, 97.47, 118.1 167.54, 194.31.
2.2.5. (η4–1,5-Cyclooctadiene)(iodo)(methyl)(trifluoroacatylacetonato)iridium(III),
[Ir(tfaa)(cod)(CH3)I]
The procedure for the synthesis of [Ir(tfaa)(cod)(CH3)I] was identical to that of [Ir(cupf)(cod)
(CH3)I]. Yield: 60%; IR spectra: v = 3014 (C–H, str), 801(=C–H bending), 1296 (C–F), 1146 (C–O)
cm−1; MS; m/z 597.25 [M]+, 598.21 [M + H]+; Elemental analysis for C14H19F3IrIO2 (595.42):
Found C, 28.07; H, 3.37%; Calcd C, 28.24; H, 3.22%; 1H NMR (600 MHz), CDCl3, 20 °C: δ 1.92
(2H, m), δ 2.16 (3H, s), 2.17 (2H, m), 2.23 (3H, s), 2.63 (2H, m), 3.09 (2H, m), 4.29 (2H, m), 5.27
(2H, m), 5.97 (1H, s). 13C NMR (600 MHz), CDCl3, 20 °C: δ 17.2, 28.4, 28.5, 29.3, 33.1, 33.2, 99.3,
116.2, 167.1 194.9.

14
W. PURCELL ET AL.
2.2.6. (η4–1,5-Cyclooctadiene-methyl-2-(methylamino)-1-cyclopentene-1-
dithiocarboxylato)iridium(I), [Ir(macsm)(cod)]
The procedure for the synthesis of [Ir(macsm)(cod)] was identical to that of [Ir(cupf)(cod)].
Yield: 64%; IR spectra: ν = 2974 (C–H, str), v = 1602 (C=N), 792 (=C–H bending), 621 (C–S)
cm−1; MS; m/z 486.25 [M]+, 487.26 [M + H]+; Elemental analysis for Ir(macsm)(cod)], C16H24IrNS2
(486.73)): Found C, 39.20; H, 5.01; N,2.75; S, 12.99%; Calcd C, 39.40; H, 5.17; N, 2.87; S, 13.15%;
1H NMR (600 MHz), CDCl3, 20 °C; δ 1.83 (2H, m), 1.91 (2H, m), 1.97 (2H, m), 2.17 (2H, m), 2.38
(2H, m), 2.58 (3H, s), 2.76 (2H, m), 2.86 (2H, m), 3.49 (3H, s), 4.07 (2H, m), 4.14 (2H, m). 13C NMR
(600 MHz), CDCl3, 20 °C; δ 17.82, 21.05, 29.97, 32.08, 34.54, 38.67, 44.18, 59.17, 69.84, 125.74,
160.53, 172.88.
2.2.7. (η4–1,5-Cyclooctadiene-methyl-2-(methylamino)-1-cyclopentene-1-
dithiocarboxylato)(iodo)(methyl)iridium(III), [Ir(macsm)(cod)CH3I]
The procedure for the synthesis of [Ir(macsm)(cod)(CH3)I] was identical to that of [Ir(cupf)
(cod)(CH3)I]. Yield: 64%; IR spectra: v = 2927 (C–H, str), 1640 (C=N), 796 (=C–H bending), 636
(C–S) cm−1; MS; m/z 628.18 [M]+, 629.19 [M + H]+; Elemental analysis for C17H27IrINS2 (628.66):
Found C, 32.21; H, 4.40; N, 2.30; S, 10.02%; Calcd C, 32.43; H, 4.48; N, 2.22; S, 10.18%; 1H NMR
(600 MHz), CDCl3, 20 °C; δ 1.73 (1H, m), 1.83 (3H, s), 1.85 (1H, m), 1.97 (1H, m), 2.17 (1H, m),
2.23 (1H, m), 2.46 (3H, s), 2.50 (1H,m), 2.60 (1H, m), 2.63 (1H, m), 2.70 (1H, m), 2.75 (1H, m),
2.78 (1H, m), 2.83 (1H, m), 3.08 (1H, m), 3.23 (3H, s), 3.30 (1H, m), 4.31 (1H, m), 4.59 (1H, m),
4.98 (1H, m), 5.63 (1H, m). 13C NMR (600 MHz), CDCl3, 20 °C; δ 6.1, 17.0, 20.8, 27.0, 29.5, 31.6,
34.1, 35.3, 38.2, 45.2, 80.1, 80.8, 92.0, 94.1, 122.6, 157.1, 171.6.
2.3. Kinetics
The UV/visible spectrum of the oxidative addition reactions of [Ir(tfaa)(cod)] and [Ir(cupf)
(cod)] with methyl iodide in acetonitrile showed one reaction with formation of an isosbestic
point at 365 nm (figure 1(a)) and 375 nm, respectively (figure 1(b)), while the same reaction
between [Ir(macsm)(cod)] and methyl iodide (figure 1(c)) showed only an increase in absorp-
tion over the whole spectrum.
Results from the initial methyl iodide concentration variations in acetonitrile indicated
linear relationships for all three complexes with an intercept close to zero as illustrated by
Figure 1. spectralchangewiththeadditionofCh3i to(a) [ir(tfaa)(cod)], (b) [ir(cupf)(cod)] and(c) [ir(macsm)
(cod)] in acetonitrile at 298 K and 10 min intervals, [Ch3i] = .2 m.

JOURNAL OF COORDINATION CHEMISTRY
15
Figure 2. kobs vs. [Ch3i] in acetonitrile at different temperatures for the reaction [ir(macsm)(cod)] + Ch3i,
[ir(macsm)(cod)] = 5.0 × 10−4 mol l−1
.
H3C
k1
k-1
L'
Ir
L'
L
Ir
I
+ CH3I
L
Scheme 2. Proposed reactions between [ir(ll′)(cod)] and Ch3i (ll′ = cupf, tfaa and macsm).
Table 1. summary of the kinetic data and activation parameters for the oxidative addition reactions of
different [ir(ll)(cod] complexes and Ch3i in acetonitrile at 25° C.
k1 (103) (M−1s−1)
k−1(104) (s−1)
ΔH≠ (kJ mol−1)
ΔS≠ (J K−1 mol−1)
Complex
[ir(cupf)(cod)]
2.98(4)
1.1(5)
31(4)*
92(28)**
37(2)
−189(15)*
−3(3)**
−166(7)
−174(7)
[ir(tfaa)(cod)]
[ir(macsm)(cod)]
4.4(2)
2.84(7)
–
–
30(2)
* forward reaction.
** reverse reaction.
oxidative addition of [Ir(macsm)(cod)] and CH3I in acetone (figure 2). These results led to a
very simple reaction mechanism (scheme 2).
The rate law for this reversible reaction is given by equation 1:
R = k1[Ir(LL�)(cod)][CH3 I]−k−1[Ir(LL�)(cod)(CH3)I]
(1)
The observed rate constant after integration and using pseudo-first-order conditions is given
by equation 2:
[
]
kobs = k1 CH3I + k−1
(2)
A summary of the calculated rate constants as well as activation parameters is given in
table 1. These values indicate that the k-1 values for the different complexes, with the excep-
tion of [Ir(cupf)(cod)] in acetonitrile, are all zero within experimental error.
2.4. Computational study
Calculations were done using the density functional theory (DFT) as implemented in the
Amsterdam Density Functional 2013 program system [24]. The PW91 (Perdew-Wang 1991)

16
W. PURCELL ET AL.
[25] GGA (Generalized Gradient Approximation) functional, a Slater-type TZP (triple-ζ plus
polarization) basis set, a fine mesh for numerical integration, and full geometry optimizations
with tight convergence criteria were used. All structures were calculated as singlet states
(S = 0) and the geometry optimizations were all symmetry-unconstrained. Scalar relativistic
and solvent effects were taken into account using the ZORA [26] (Zero-Order Regular
Approximation) formalism and the COSMO (Conductor like Screening Model) model of sol-
vation [27], respectively. The type of cavity used for the solvent effects was Esurf [28] and
the solvent used was acetonitrile (ε0 = 37.5) to match the experimental conditions. All tran-
sition states and minima were confirmed with frequency analyses. Zero-point energy and
thermal corrections (vibrational, rotational and translational) were made to the electronic
energies in the calculation of the thermodynamic parameters. Enthalpies (H) and Gibbs free
energies (G) were calculated from:
U = Eel + EIE
(3)
H = U + RT(for gas phase); or H = U(in solution)
(4)
G = H−TS
(5)
where U = the total energy, Eel = total electronic energy, and EIE = internal energy (sum of
vibrational, rotational, translational energies including the zero-point energy correction),
R = ideal gas constant, T = temperature and S = entropy. The entropy (S) was calculated from
the temperature-dependent partition function in ADF at 298.15 K.
3. Results and discussion
3.1. Synthesis
All six iridium complexes were synthesized and characterized by 1H and 13C NMR, microa-
nalysis and infrared spectroscopy. The 1H NMR data are summarized in table 2. The eight cod
methylene protons for all the isolated complexes are observed between δ 1.39 and 3.09,
while the protons in two of the three isolated oxidative addition products are shifted down-
field. The four cod double bond protons in the starting compounds appear only as two peaks,
suggesting similar chemical environments on each side of the square planar compound,
but these peaks shift downfield and are split into four peaks for two of the three oxidative
addition products. This confirms the difference in the chemical environment above and
below the plane, as well as on either side of the oxidative addition product after coordination
of CH3 and I to the metal center. The absence of splitting in the tfaa complex suggests insig-
nificant changes in the chemical environment above and below the plane after oxidative
addition. The methyl groups in the oxidative addition products appear between δ 1.83 and
3.01 ppm confirming coordination of methyl iodide to the metal center in the different
isolated oxidative addition products.
The 13C NMR of [Ir(cupf)(cod)(CH3)I] indicated the presence of 13 carbons with the aliphatic
carbons assigned between 28.8 and 97.1 ppm and the aromatic carbons between 121.1 and
139.1 ppm. The APT experiment (ATP = Attached Proton test) for the same complex indicates
four carbons with an uneven number of protons (negative mode) and nine carbons with an

JOURNAL OF COORDINATION CHEMISTRY
17
Table 2. summary of the 1h spectra of all the isolated complexes.
1H spectra/δ ppm
Methylene
protons (cod)
Double bond
protons (cod)
Complex
Aromatic proton
Methyl group
Methine
[ir(cupf)(cod)]
[ir(cupf)(cod)(Ch3)i]
1.74–2.25
1.39–2.58
4.24, 4.30
4.39, 4.61, 4.72,
5.40
7.47 – 7.98
7.48 – 8.08
3.01
ir(tfaa)(cod)]
[ir(tfaa)(cod)(Ch3)i]
[ir(macsm)(cod)]
1.60–2.19
1.92–3.09
1.96–2.40
1.83–2.86*
1.97–3.08
4.10, 4.14
4.29, 5.27
4.07, 4.14
2.07
2.16, 2.23
2.58, 3.49
5.87
5.97
[ir(macsm)(cod)(Ch3)i]
4.31, 4.59, 4.98,
5.63
1.83, 2.46, 3.23
1.73–2.83*
*Cyclopentane ring.
Bold: methyl group after oxidative addition.
Figure 3. the solid state structure of (a) [ir(sacac)(cod)(Ch3)i] [29] and (b) the dft optimized geometry
[ir(cupf)(cod)(Ch3)i]. selected interatomic distances (Å) are shown to highlight the methyl group (noesy)
of the oxidative addition product and the proton(s) of the double bond of cod. non h and C atoms are
indicated with symbols and the hydrogens with a small black dot.
1
even number (positive mode) of protons. The H–1H NOESY experiment indicates strong
coupling between the methyl of the oxidative addition product at δ 3.01 and the proton of
the double bond of cod at δ 4.39, establishing that the methyl group is adjacent to these
two protons. The HMBC experiment also confirmed long-range coupling between the oxi-
dative addition methyl at δ 3.01 and the protons with the cyclooctadiene double bonds.
Close inspection of the crystal structure of [Ir(sacac)(cod)(CH3)I] [29] confirmed that the
orientation of one of the methyl protons is in the direction of the cod C–C bond with the
two H(methyl)–H(double bond) interactions almost equal to each other (3.006 and 3.013 Å, respec-
tively) endorsing that long-range coupling was observed (figure 3). This result is consistent
with trans addition of CH3I to [Ir(cupf)(cod)], similar to what was found for the oxidative
addition product of CH3I to [Ir(sacac)(cod)] [29].
Surprisingly the [Ir(tfaa)(cod)(CH3)I] 13C NMR spectrum shows the presence of only 10 of
the 14 carbons in the product, which corroborates the nine carbons (instead of 13) observed
for [Ir(tfaa)(cod)] as starting complex. This observation suggests, as was observed with the
1H NMR spectrum, that the chemical environment changes insignificantly before and after
oxidative addition, producing four equivalent cod carbons and therefore fewer carbons than
expected. The APT experiment indicates seven carbons with an even number of protons

18
W. PURCELL ET AL.
and three carbons with an odd number of protons, as expected for the methine and the two
methyl groups in the final product. The NOESY experiment indicates a diagonal interaction
between the new methyl group at δ 2.23 and two indistinguishable protons of the double
bond at δ 4.29. The HMBC experiment confirms long-range coupling/interaction between
the oxidative addition methyl group and a proton for one of the double bonds in
cyclooctadiene.
The 13C NMR spectrum of [Ir(macsm)(cod)(CH3)I] showed the presence of 17 carbons in
the oxidative product, while the APT experiments showed 10 carbons with odd number of
1
protons, and 7 carbons with an even number of protons (positive mode). H–1H NOESY
indicated diagonal interaction between the methyl group at δ2.46 of the oxidative addition
product and the proton of the double bonds at δ 4.98 and 5.63, which was confirmed with
HMBC. The NOESY coupling between the methyl group of the oxidative addition product
and the proton of the double bonds for [Ir(tfaa)(cod)CH3I] and [Ir(macsm)(cod)(CH3)I] indi-
cates that the CH3 groups are above the square plane of the Ir(I) complex.
The cod protons at 1.4–5.7 ppm compares favorably with those obtained for [Ir(triazalato)
(cod)] (triazolato = 3.5-bis(pyridine-2-yl)-1,2,4 triazolate (bpt-NH) and 4-amino-3,5-
bis(pyridine-2-yl)-1,2,4 triazolate (bpt)) [30], and [Ir(S,O)(cod)] (S,O = N-methyl-p-
methoxybenzathiohydroxaminato and 1-hydroxi-2-pyridinato ligands) [31], which were
observed between 1.8 and 3.8, and 1.7 and 4.4 ppm, respectively.
3.2. Kinetics
The initial UV/Vis scans after the addition of CH3I confirm that the three complexes react
with methyl iodide. The isolation of the products in acetone clearly indicates formation of
only one product, the Ir(III) alkyl product according to scheme 1. The kinetic results for all of
the oxidative addition reactions between CH3I and the different complexes in acetonitrile
showed simple second-order kinetics with zero intercepts for macsm and tfaa. The small
intercept observed for the reaction between CH3I and [Ir(cupf)(cod)] can either be a reversible
or a parallel reaction (table 1). Only two possible reactions can occur parallel to the main
oxidative addition, namely a solvent reaction with the square planar complex followed by
the reaction of the solvated complex with the CH3I to form the final oxidative addition com-
plex or oxidative addition via a cis mechanism which differs in reaction rates. A solvent
reaction pathway is highly unlikely since all the complexes were very stable in the different
solvents (paragraph 2.1). DFT calculations (paragraph 3.3) predict two different approaches
of CH3I to the metal complex (“linear” or “bent”) which both lead to trans addition while a
“front”addition leads to cis addition. Interestingly the DFT calculations predict cis products
for [Ir(cupf)(cod)] which have energy very similar to that of the trans product, but the acti-
vation energies of the“linear”(trans addition) and“front”(cis addition) transition states indi-
cate that the “linear” transition state is energetically favored above the “front” transition,
making cis addition highly unlikely. The intercept observed for the [Ir(cupf)(cod)] is therefore
regarded as a reversible reaction and not a parallel reaction. The rate of oxidative addition
for [Ir(macsm)(cod)] was the fastest (2.84(7) × 10−3 M s−1) of the three complexes and larger
by a factor 4 and 13 than [Ir(cupf)(cod)] and [Ir(tfaa)(cod)], respectively. The large negative
entropy of activation indicates associative activation during the transition state. The forward
reaction for [Ir(cupf)(cod)] is entropy controlled while the reverse reaction is enthalpy
controlled.

JOURNAL OF COORDINATION CHEMISTRY
19
Table 3. summary of the kinetic data for the oxidative addition reactions between [ir(ll)(cod)] and Ch3i
in different solvents and bidentate ligands with different ring sizes at 25° C.
k1 (×102) (M−1 s−1)
Solvent
Ref.
Complex
Donor atoms
Ring size
[ir(macsm)(cod)]
[ir(sacac)(cod)]
[ir(tfaa)(cod)]
[ir(cupf)(cod)]
[ir(anmetha)(cod)]
[ir(hpt)(cod)]
[ir(bpt-nh)(cod)]
[ir(anmetha)(cod)]
[ir(hpt)(cod)]
s,n
s,o
o,o
o,o
s,o
s,o
n,n
s,o
s,o
n,n
n,n
6
6
6
5
5
5
6
5
5
5
6
2.84(7)
0.52(3)
0.44(2)
0.298(4)
2.69(6)
2.2(2)
1.44(7)
0.943(10)
0.693(17)
0.35(1)
0.0919(4)
acetonitrile
acetonitrile
acetonitrile
acetonitrile
nitromethane
nitromethane
dichloromethane
acetone
this study
[20]
this study
this study
[31]
[31]
[30]
[31]
[31]
[30]
[30]
acetone
acetone
Benzene
[ir(bpt)(cod)]
[ir(bpt-nh)(cod)]
A comparison of the rate of oxidative addition for the different complexes (table 3) in
acetonitrile indicates the order N,S > S,O [20] > O,O for the bidentate donors, which corre-
sponds favorably with the oxidative addition order of N,S > N,O > S,O > O,O [32] observed
for a number of [Rh(LL′)(CO)(PPh3)] complexes containing different donor combinations
(LL′ = bidentate ligand of charge −1 and with donor L and L′). This increase in the oxidative
addition rate is attributed to an increase in the nucleophilicity of the metal complex as a
result of the better σ-bonding properties of nitrogen and sulfur compared to oxygen. A
comparison with [Ir(bpt-NH)(cod)] [30] (triazolato complexes) may be questionable since
the reactions were performed in different solvents than in the current study (dichlorometh-
ane and benzene); however, it is evident that the oxidative addition rate of this complex is
in the same order of magnitude (1.44(7) × 10−2 M−1 s−1 in dichloromethane) as those in the
current study.
DFT calculations for [M(triazolato)(cod)] [30] (M = Ir(I) and Rh(I)) predicted cis methyl
iodide addition compared to the trans addition to [Ir(sacac)(cod)] and [Ir(acac)(cod)], which
was also experimentally confirmed by the respective crystal structure determinations [29,
33]. A change in transition state and therefore a different final product is not surprising if
the substantial steric interaction between the cod protons and the methyl group is taken
into account, which is evident from the large deviation from linearity for the trans addition
products. The similarity between the tfaa and sacac/acac and to a certain extent macsm,
however, renders the above-mentioned argument questionable and more research is needed
to conclusively prove or disprove the argument.
3.3. Computational chemistry
No solid state X-ray data on the geometry of [Ir(cupf)(cod)], [Ir(tfaa)(cod)] and [Ir(macsm)
(cod)] are available [34], only powder diffraction data of [Ir(tfaa)(cod)] [35]. Thus, to get more
insight into the structure of the iridium(I) cod complexes of this study, DFT computational
chemistry was done to determine the optimized minimum energy geometries of [Ir(cupf)
(cod)], [Ir(tfaa)(cod)], and [Ir(macsm)(cod)], as well as the iridium(III) products of oxidative
addition of these complexes. Different views of the DFT optimized geometries of the [Ir(LL′)
(cod)] complexes are shown in figure 4. The pseudo aromatic ring formed by Ir and the LL′-
ligand is near planar in all three structures with a dihedral angle of −5.8° (SCCN in cupf),
−0.2° (OCCO in tfaa), and −1.3° (ONNO in macsm). The two C=C bonds of cod are nearly

20
W. PURCELL ET AL.
Figure 4. three different views of the PW91/tZP optimized geometries of (a) [ir(cupf)(cod)], (b) [ir(tfaa)
(cod)] and (c) [ir(macsm)(cod)]. selected bond lengths (Å) are indicated. the dotted gray lines in the top
structures highlight the staggered conformation of the two Ch2Ch2– moieties of the cod ligand. the dotted
gray lines in the middle structures highlight the nearly parallel conformation of the two C=C bonds of
the cod ligand. non h and C atoms are indicated with symbols and the hydrogens with a small black dot.

JOURNAL OF COORDINATION CHEMISTRY
21
CH3
I
CH3
Ir
I
CH3
L
Ccod
Ccod
L
CH3
Ccod
L
I
L
Ccod
CH3
L
Ccod
I
Ir
Ir
Ir
Ir
L'
L'
L'
Ccod
L'
L'
I
Ccod
Ccod
Ccod
Ccod
trans
cis 1
cis 2
cis 3
cis 4
Scheme 3. schematic illustrations of the trans and cis isomers of [ir(ll′)(Ch3)(i)(cod)].
Table 4. PW91/tZP dft calculated electronic (e), enthalpy (h), and Gibbs free (G) energies (kJ mol−1) of
the possible transition states (ts) and possible products of the [ir(ll′)(cod)] + Ch3i oxidative addition
reaction, relative to the energy of the reactants as 0.
[Ir(cupf)(cod)]
[Ir(tfaa)(cod)]
[Ir(macsm)(cod)]
∆H
∆G298 K
0.0
∆E
0.0
∆H
∆G298 K
∆E
0.0
∆H
∆G298 K
∆E
0.0
−7.3
ir(i) + Ch3i
ir(iii)-alkyl
cis 1
ir(iii)-alkyl
cis 2
ir(iii)-alkyl
cis 3
ir(iii)-alkyl
cis 4
ir(iii)-alkyl
trans
ts front
ts trans
0.0
−47
0.0
11.8
0.0
−42
0.0
5.9
0.0
−55.0
−17.7
−50.9
−21.0
−49.8
605.7
−54.8
−16.8
−56.3
−16.6
−60.8
142.3
2
52.1
−9
−43
−12
−39
570
40.5
14.9
−8
−46
−8
43.2
7.4
−2.0
6
56.5
48.4
−10.1
13.5
−1
23
38.7
43.8
0.5
65.0
12.9
−47
147
−45.0
195.3
−35
193
13.5
621.0
189.3
241.3
29.0 31 (31 exp) 82.5 (87 exp)
29.8 35 (37 exp) 77.8 (86 exp) 27.6 29 (30 exp) 80.0 (82 exp)
parallel when coordinated to Ir, while the two CH2CH2- moieties of cod adopt a staggered
conformation. The Ir-Ccod bonds are slightly longer (ca. 0.04 Å) in [Ir(macsm)(cod)] than in
[Ir(cupf)(cod)] and [Ir(tfaa)(cod)]. The Ir-O bonds in [Ir(cupf)(cod)] with the more strained
five-membered ring are slightly shorter (0.02–0.03 Å) than in [Ir(tfaa)(cod)] with a six-mem-
bered ring.
Oxidative addition of methyl iodide to [Ir(LL′)(cod)] leads to an iridium(III) alkyl complex
[Ir(LL′)(CH3)(I)(cod)]. The CH3 group and I in [Ir(LL′)(CH3)(I)(cod)] can either be trans (trans
product) or cis to each other (cis product). Four cis isomers are possible when the LL′-bidentate
ligand is unsymmetrical (scheme 3). The relative energies of the optimized geometries of
the five [Ir(LL′)(CH3)(I)(cod)] isomers for LL′ = cupf, tfaa and macsm are given in table 4.
The trans isomers have the lowest energies (most stable) for LL′ = tfaa and macsm, while for
LL′ = cupf the energy of the cis 1, cis 3 and trans isomers are very near to each other.
Both experimental and theoretical data strongly support the SN2 mechanism for the oxi-
dative addition of methyl iodide to square planar complexes [36]. During the SN2 attack, the
methyl iodide can either approach the metal center in a linear way with the iodine directed
away from the metal (leading to a “linear” transition state), or with C-I bond in a side-ways
fashion (leading to a“bent”transition state) [37]. Both the“linear”and“bent”transition states
lead to a five-coordinate methyl complex (of which the charge increases by 1) and a free
iodide ion (of charge −1). After coordination of the iodide, a trans oxidative addition complex
is formed [37]. A cis oxidative addition complex results from a “front” transition state (TS)
where the C–I bond of methyl iodide also approaches the metal center in a side-ways fashion,

22
W. PURCELL ET AL.
but then leads to a concerted three-center oxidative addition with simultaneous metal-I
and metal-C formation and I-C breaking [37].
The calculated activation energies of the “linear” and “front” transition states for the oxi-
dative addition reaction [Ir(LL′)(cod)] + CH3I for LL′ = cupf, tfaa and macsm is given in table
4. No “bent” transition state could be located. The “linear” transition state is energetically
favored, while the activation barrier of the“front”transition state is too high to occur. Trans
[Ir(LL′)(CH3)(I)(cod)] products are thus formed for all three complexes of this study, including
LL′ = cupf where the energy of the cis 1 and cis 3 isomers are slightly lower than the energy
of the trans isomer. This result is in agreement with computational chemistry and experi-
mental results [20] for the related oxidative addition reaction [Ir(LL′)(cod)] + CH3I for
LL′ = sacac (sacac = thioacetyl acetonato), which was found to proceed through a “linear”
transition state giving trans-[Ir(cod)(sacac)(CH3)(I)] in the solid state [29]. The NMR results of
the oxidative addition products, presented in a previous section, are also consistent with
trans oxidative addition products for the three complexes of this study.
The optimized trans-[Ir(LL′)(CH3)(I)(cod)] products have a distorted octahedral structure
with the I–Ir–CH3 angle (156–158°) deviating from linearity, possibly due to steric repulsion
between the CH3 or I and the olefinic hydrogens (figure 5). These results correlate very well
with the experimentally obtained I–Ir–CH3 bond angles of 156.06° and 156.58°, which were
reported for [Ir(cod)(sacac)(CH3)(I)] and [Ir(cod)(acac)(CH3)(I)] [33]. The optimized geometries
of the“linear”transition state of the [Ir(LL′)(cod)] + CH3I oxidative addition of each complex
are shown in figure 6. During the “linear” transition state of the oxidative addition process,
iridium approaches the methyl iodide trans to the leaving iodide with formation of the
Figure 5. PW91/tZP optimized geometries of the trans oxidative addition product of [ir(ll′)(cod)] + Ch3i
oxidative addition reaction for ll′ = (a) cupf, (b) tfaa, and (c) macsm. selected bond lengths (Å) and
angles (deg) are indicated. non h and C atoms are indicated with symbols and the hydrogens with a
small black dot.
Figure 6. PW91/tZP optimized geometries of the linear transition state of [ir(ll′)(cod)] + Ch3i oxidative
addition reaction for ll′ = (a) cupf, (b) tfaa, and (c) macsm. selected bond lengths (Å) are indicated. non
h and C atoms are indicated with symbols and the hydrogens with a small black dot.

JOURNAL OF COORDINATION CHEMISTRY
23
Ir-methyl bond and simultaneous dissociation of the Me-I bond. The methyl iodide is tilted
towards LL′-bidentate ligand, probably due to steric influence of the cod ligand. The com-
putational chemistry calculated activation free energy of the “linear” transition state is in
good agreement with experimental values (table 4). The DFT calculated enthalpy values in
table 4 show that the oxidative addition reaction is exothermic, the product Ir(III)-alkyl trans
is 35–47 kJ mol−1 more stable than the separate reactants, CH3I and the Ir(I) complex. DFT
calculated large negative activation entropy values (−172, −145 and -170 J K−1 mol−1 for
[Ir(cupf)(cod)], [Ir(tfaa)(cod)], and [Ir(macsm)(cod)], respectively) indicate an associative acti-
vation during the transition state.
4. Conclusion
We have synthesized three different [Ir(LL′)(cod)] complexes (LL′ = N-aryl-N-nitrosohyd-
roxylaminato) (cupf), trifluoroacetylacetonato (tfaa) and (methyl 2-(methylamino)-1-cyclo-
pentene-1-dithiocarboxylato-κN, κS) (macsm)). All the complexes were characterized by 1H
and 13C NMR spectroscopy, IR, elemental microanalysis, and mass spectrometry. Their rates
of oxidative addition with methyl iodide were determined. The computational study of [Ir(LL′)
(cod)] + CH3I oxidative addition reaction for LL′ = cupf, tfaa and macsm, in agreement with
1H NMR results, showed that the oxidative addition reaction proceeds through a “linear”
transition state with a calculated activation energy in agreement with experimental values.
The transition state corresponds to a SN2 nucleophilic attack by the metal compound on
methyl iodide. A concerted three center “front” transition state structure exists as well, but
is not energetically favored.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This work was supported by the Norwegian Supercomputing Program (NOTUR) through a grant of
computer time [grant number NN4654 K] (JC); the South African National Research Foundation; and
the Central Research Fund of the University of the Free State, Bloemfontein, South Africa.
References
[1] T.W. Dekleva, D. Foster. Adv. Catal., 34, 81 (1986).
[2] A. Haynes, P.M. Maitlis, G.E. Morris, G.J. Sunley, H. Adams, P.W. Badger, C.M. Bowers, D.B. Cook, P.I.P.
Elliott, T. Ghaffar, H. Green, T.R. Griffin, M. Payne, J.-M. Pearson, M.J. Taylor, P.W. Vickers, R.J. Watt.
J. Am. Chem. Soc., 126, 2847 (2004).
[3] P.M. Maitlis, A. Hayes, G.J. Sunley, M.J. Howard. J. Chem. Soc., Dalton Trans., 2187 (1996). doi:
10.1039/DT9960002187.
[4] P.D. Pavlechko, IHS Chemical Report, Review (2103). Available online at: www.ihs.com/pdf/
RW2014-01_toc_183772110917062932.pdf (accessed 16 March 2015).
[5] H-U. Blaser. Adv. Synth. Catal., 344, 17 (2002).
[6] H.-U. Blaser, B. Pugin, F. Spindler, M. Thommen. Acc. Chem. Res., 40, 1240 (2007).
[7] R. Dorta, D. Broggini, R. Stoop, H. Rüegger, F. Spindler, A. Togni. Chem. Eur. J., 10, 267 (2004).
[8] R.H. Crabtree. Acc. Chem. Res., 12, 331 (1979).

24
W. PURCELL ET AL.
[9] R.H. Crabtree, H. Felkin, T. Fillebeen-Khan, G.E. Morris. J. Organomet. Chem., 168, 183 (1979).
[10] J.M. Bueno, J.M. Coterón, J.L. Chiara, A. Fernández-Mayoralas, J.M. Fiandor, N. Valle. Tetrahedron
Lett., 41, 4379 (2000).
[11] S. Hietkamp, D.J. Stufkens, K. Vrieze. J. Organomet. Chem., 139, 189 (1977).
[12] G. Winkhuis, H. Singer. Z. Naturforsch., Teil B, 20, 602 (1965).
[13] M. Green, T.A. Kuc, S.H. Taylor. J. Chem. Soc., A, 2334 (1971).
[14] S. Hietkamp, D.J. Stufkens, K. Vrieze. J. Organomet. Chem., 122, 419 (1976).
[15] J.-Y. Chen, J. Halpern. J. Am. Chem. Soc., 93, 4939 (1971).
[16] D. Astruc. Organometallic Chemistry and Catalysis, p. 435, Springer Heidelberg Berlin, New York
(2000).
[17] J.G. Leipoldt, E.C. Steynberg, R. Van Eldik. Inorg. Chem., 26, 3068 (1987).
[18] J.A. Venter, J.G. Leipoldt, R. Van Eldik. Inorg. Chem., 30, 2207 (1991).
[19] (a) M.M. Conradie, J. Conradie. Inorg. Chim. Acta, 361, 208 (2008); (b) M.M. Conradie, J. Conradie.
Inorg. Chim. Acta, 361, 2285 (2008); (c) N.F. Stuurman, J. Conradie. J. Organomet. Chem., 694, 259
(2009); (d) M.M. Conradie, J. Conradie. Inorg. Chim. Acta, 362, 519 (2009).
[20] W. Purcell, J. Conradie, S. Kumar, J.A. Venter. J. Organomet. Chem., 801, 80 (2016).
[21] D.D. Perrin, W.L.F. Armarego. Purification of Laboratory Chemicals, 3rd Edn, Pergamon Press, Oxford
(1988).
[22] Scientist, Version 2, MicroMath Scientific Software, St. Louis, MO (2004).
[23] B. Bordas, P. Sohar, G. Matolcsy, P. Berencsi. J. Org. Chem., 37, 1727 (1972).
[24] ADF2013, SCM, Theoretical Chemistry, Vrije Universiteit, Amsterdam, The Netherlands, The ADF
program system uses methods described in: G. teVelde, F.M. Bickelhaupt, E.J. Baerends, C.F. Guerra,
S.J.A. van Gisbergen, J.G. Snijders, T. Ziegler. J. Comput. Chem., 22, 931 (2001).
[25] (a) J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C. Fiolhais. Phys. Rev.
B, 46, 6671 (1992); (b) Erratum: J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson,
D.J. Singh, C. Fiolhais. Phys. Rev. B, 48, 4978 (1993).
[26] (a) E. van Lenthe, A. Ehlers, E.J. Baerends. J. Chem. Phys., 110, 8943 (1999); (b) E. van Lenthe,
E.J. Baerends, J.G. Snijders. J. Chem. Phys., 99, 4597 (1993); (c) E. van Lenthe, E.J. Baerends, J.G.
Snijders. J. Chem. Phys., 101, 9783 (1994); (d) E. van Lenthe, J.G. Snijders, E.J. Baerends. J. Chem.
Phys., 105, 6505 (1996); (e) E. van Lenthe, R. van Leeuwen, E.J. Baerends, J.G. Snijders. Int. J.
Quantum Chem., 57, 281 (1996).
[27] (a) A. Klamt, G. Schüürmann. J. Chem. Soc., Perkin Trans. 2, 799, (1993); (b) A. Klamt. J. Phys. Chem.,
99, 2224 (1995); (c) A. Klamt, V.J. Jonas. J. Chem. Phys., 105, 9972 (1996).
[28] J.L. Pascual-ahuir, E. Silla, I. Tuñon. J. Comput. Chem., 15, 1127 (1994).
[29] Y.M. Terblans, S.S. Basson, W. Purcell, G.J. Lamprecht. Acta Crystallogr., Sect. C, 51, 1748 (1995).
[30] A. Muller, J. Conradie, W. Purcell, S.S. Basson, J.A. Venter. S. Afr. J. Chem., 63, 11 (2010).
[31] M. Theron, E. Grobbelaar, W. Purcell, S.S. Basson. Inorg. Chim. Acta, 358, 2457 (2005).
[32] (a) A. Roodt, G.J.J. Steyn. Recent Research Developments in Inorganic Chemistry, 2, 1 (2000); (b) J.G.
Leipoldt, S.S. Basson, L.J. Botha. Inorg. Chim. Acta, 168, 215 (1990).
[33] S.S. Basson, J.G. Leipoldt, W. Purcell, J.B. Schoeman. Acta Cryst. C, 45, 2000 (1989).
[34] Cambridge Structural Database (CSD), Version 5.36, May 2015 update.
[35] L. Davignon, A. Dereigne, R. Bonnaire, J.M. Manoli. J. Less Common Met., 25, 75 (1971).
[36] (a) T.R. Griffin, D.B. Cook, A. Haynes, J.M. Pearson, D. Monti, G.E. Morris. J. Am. Chem. Soc., 118, 3029
(1996); (b) M.M. Conradie, J. Conradie. Dalton Trans., 40, 8226 (2011).(c) J.J.C. Erasmus, J. Conradie.
Dalton Trans., 42, 8655 (2013); (d) M.M. Conradie, J. Conradie. J. Organomet. Chem., 695, 2126
(2010); (e) M.M. Conradie, J. Conradie. S. Afr. J. Chem., 61, 102 (2008). Available online at: http://
www.journals.co.za/sajchem/.
[37] M. Feliz, Z. Freixa, P.W.N.M. van Leeuwen, C. Bo. Organometallics, 24, 5718 (2005).