Oxidatively induced one-electron reductive elimination of pentafluorophenyl from [Mo(C6F5)(CO)5]-

Article abstract of DOI:10.1016/j.poly.2004.02.022

The anionic molybdenum(0) complex [Mo(C₆F₅)(CO) ₅]^- displays a one-electron oxidation by cyclic voltammetry that is only partially chemically reversible. The partial chemical reversibility is caused by a slow chemical reaction following oxidation. This reaction has been shown to be cleavage of the Mo-C₆F₅ bond in a one-electron reductive elimination process. Chemical oxidation by ferrocenium in the presence of nucleophiles yields known Mo(CO)₅L (L=Py, PPh₃) complexes, while oxidation with tropylium traps the C₆F₅ radical as C₇H₇C ₆F₅. NMR and mass spectrometry were used to identify the oxidation products. Simulated cyclic voltammograms have been used to estimate the rate constant for cleavage of the Mo-C₆F₅ bond in the presence of pyridine as approximately 700 M^-1s^-1.

Full text of DOI:10.1016/j.poly.2004.02.022

Polyhedron 23 (2004) 1371–1374
www.elsevier.com/locate/poly
Oxidatively induced one-electron reductive elimination
of pentafluorophenyl from [Mo(C6F5)(CO)5]ꢀ
*
Jungsook Kim, Stephen L. Gipson
Department of Chemistry, Baylor University, Waco, TX 76798, USA
Received 11 December 2003; accepted 15 February 2004
Available online 12 April 2004
Abstract
The anionic molybdenum(0) complex [Mo(C6F5)(CO)5]ꢀ displays a one-electron oxidation by cyclic voltammetry that is only
partially chemically reversible. The partial chemical reversibility is caused by a slow chemical reaction following oxidation. This
reaction has been shown to be cleavage of the Mo–C6F5 bond in a one-electron reductive elimination process. Chemical oxidation
by ferrocenium in the presence of nucleophiles yields known Mo(CO)5L (L ¼ Py, PPh3) complexes, while oxidation with tropylium
traps the C6F5 radical as C7H7C6F5. NMR and mass spectrometry were used to identify the oxidation products. Simulated cyclic
voltammograms have been used to estimate the rate constant for cleavage of the Mo–C6F5 bond in the presence of pyridine as
approximately 700 Mꢀ1 sꢀ1
.
Ó 2004 Elsevier Ltd. All rights reserved.
Keywords: Molybdenum complexes; Carbonyl complexes; Pentafluorophenyl; Oxidation; Bond cleavage; Reductive elimination
1. Introduction
ductive elimination via cleavage of the Mo–C6F5 bond
to yield the 16-electron Mo(CO)5 fragment and a pen-
tafluorophenyl radical.
For some years now we have been studying the
redox-induced chemistry of pentafluorophenylmolyb-
denum carbonyl complexes. We have shown that the
cycloheptatrienyl derivative Mo(C6F5)(CO)2(g-C7H7)
undergoes an inefficient oxidatively induced CO in-
sertion [1], while upon reduction it dimerizes by cou-
pling of the cycloheptatrienyl rings to give the ditropyl
complex [{Mo(C6F5)(CO)2}2(g6:g06-C14H14)]2ꢀ [2]. Dis-
placement of the ditropyl ligand from the dimeric
dianion by P(OMe)3 yields cis-mer-[Mo(C6F5)(CO)2-
{P(OMe)3}3]ꢀ [3], which undergoes other substitution
reactions and an oxidatively induced isomerization [4].
In an effort to find a more convenient entry into related
Mo(0) chemistry, we developed a synthesis of the
[Mo(C6F5)(CO)5]ꢀ anion [5]. We have now concluded
an investigation of the oxidative chemistry of this
complex, in which it was demonstrated that oxidation is
followed by relatively slow one-electron, homolytic re-
While the possibility of one-electron reductive elimi-
nation from transition metal organometallic com-
pounds has long been discussed, the literature contains
relatively few reported examples. The chemical oxida-
tion of R4Pb (R ¼ Me, Et), Me2Pt(PPh3)2 and
2ꢀ
Et2Pt(PMe2Ph)2 by IrCl6 is believed to be followed by
metal–carbon bond cleavage to yield alkyl radicals
2ꢀ
which subsequently react with additional IrCl6
to
yield RCl as the organic product of one-electron re-
ductive elimination [6,7]. Similarly, chemical or elec-
trochemical oxidation of tetramethylaurate [8] and
dimethylcobalt macrocyclic chelates [9] produces methyl
radicals. One-electron oxidation of several group 4
dicyclopentadienyl complexes Cp2MR2 (M ¼ Ti, Zr;
R ¼ Me, Ph, Bz) produces organic radicals which sub-
sequently dimerize [10–12]. Chemical or electrochemical
oxidation of CpꢁIr(DMSO)Me2 in DMSO as solvent
yields CH4 through Ir–CH3 bond cleavage followed by
hydrogen abstraction from adventitious water, while
oxidation of the rhodium analog yields instead the two-
* Corresponding author. Tel.: +254-710-3311; fax: +254-710-2403.
E-mail address: stephen_gipson@baylor.edu (S.L. Gipson).
0277-5387/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2004.02.022

1372
J. Kim, S.L. Gipson / Polyhedron 23 (2004) 1371–1374
electron reductive elimination product ethane [13].
Oxidation of CpꢁRuL2R (L ¼ PMe3, PMe2Ph, R ¼ Me;
L ¼ PMe3, R ¼ CH2CMe3) and CpRu(PPh3)2Me yield
RH via Ru–C bond cleavage followed by H-atom ab-
straction from one of the other ligands [14]. In a similar
study of the oxidations of CpRu(CO)(PR3)Me (R ¼ Cy,
Ph), it was concluded that coordination of solvent to the
17eꢀ cation precedes Ru–Me bond cleavage [15]. The
only reported example of one-electron reductive elimi-
nation from group 6 compounds is that of Cp2MX2
(M ¼ Mo, W; X ¼ halide, thiolate) [16]. We report here
the first oxidatively induced metal–carbon bond cleav-
age (one-electron reductive elimination) in a molybde-
num complex, [Mo(C6F5)(CO)5]ꢀ, which also has the
distinction of not having any hydrogen-containing
ligands.
3. Results and discussion
3.1. Cyclic voltammetry of PPN[Mo(C6F5)(CO)5]
The cyclic voltammetry of PPN[Mo(C6F5)(CO)5] in
CH2Cl2 displays a partially chemically reversible one-
electron oxidation at +0.66 V vs. Cpꢁ2Feþ/Cp2ꢁFe. The
ratio of the cathodic to the anodic peak currents for this
couple ranges from near zero at scan rates below 20 mV/
s to near unity at scan rates above 5 V/s. This change in
chemical reversibility suggests that the presumed prod-
uct of the oxidation, the neutral 17-electron complex
Mo(C6F5)(CO)5, undergoes a relatively slow chemical
reaction. In THF and MeCN solvents the chemical re-
versibility is even lower, with the peak current ratio for a
given scan rate decreasing in the order CH2Cl2 >
THF > MeCN. Given that this is the same order as the
coordinating ability of the solvents, these results suggest
that the decomposition of the 17-electron species in-
volves nucleophilic attack by solvent.
2. Experimental
The participation of nucleophiles in the chemical re-
action following oxidation of [Mo(C6F5)(CO)5]ꢀ is
further supported by the observation that the addition
of ligands such as pyridine and triphenylphosphine to
CH2Cl2 solutions of PPN[Mo(C6F5)(CO)5] causes a
decrease in chemical reversibility of the oxidation. At a
scan rate of 0.2 V/s, the peak current ratio for a 1.0 mM
solution of PPN[Mo(C6F5)(CO)5] in CH2Cl2 decreases
from 0.72 in the absence of pyridine to essentially zero
with 5.0 mM added pyridine. All of these observations
are consistent with an EC mechanism for the oxidation
of [Mo(C6F5)(CO)5]ꢀ.
2.1. Reagents
All reactions were carried out under argon using
Schlenk techniques or in a nitrogen-atmosphere glove-
box. All solvents were distilled from appropriate drying
agents and stored under nitrogen. The compounds
PPN[Mo(C6F5)(CO)5] [5] (PPN ¼ bis(triphenylphos-
phoranylidene)ammonium) and Cp2FeBF4 [17] were
prepared by published methods. All other reagents were
obtained commercially and were used as received.
2.2. Instrumentation
3.2. Chemical oxidation of PPN[Mo(C6F5)(CO)5] by
ferrocenium and tropylium
IR spectra were obtained using a Mattson Instru-
ments Genesis II FTIR and a cell with CaF2 windows
1
separated by a 0.1 mm spacer. H, 31P, and 19F NMR
In order to identify the products of the reaction of
Mo(C6F5)(CO)5 with nucleophiles, PPN[Mo(C6F5)-
(CO)5] was treated with the mild oxidants Cp2FeBF4
and C7H7BF4. The oxidations with ferrocenium (1
equiv.) were conducted in the presence of pyridine or
triphenylphosphine (2 equiv.) in CDCl3 as solvent.
NMR spectroscopy of the resulting solutions identified
spectra were obtained on a Bruker DPX 300 MHz
NMR spectrometer. NMR chemical shifts were refer-
1
enced to residual protons in the solvent (for H), 85%
H3PO4 external standard (for 31P), or CFCl3 external
standard (for 19F). Mass spectra were obtained on a VG/
Fisons Prospec 3000 double focusing spectrometer using
EI ionization. Peaks in the mass spectrum were refer-
enced to perfluorokerosene.
1
the products as Mo(CO)5L (by H and 31P, L ¼ Py or
PPh3), C6F5H (by 1H and 19F), and Cp2Fe (by 1H).
Thus, the 16-electron Mo(CO)5 fragment produced by
the Mo–C6F5 bond cleavage was trapped as a stable 18-
electron complex by the added ligand, while the C6F5
radical most likely abstracted a hydrogen atom from
adventitious water to yield the pentafluorobenzene.
The reaction of PPN[Mo(C6F5)(CO)5] with C7H7BF4
could not be performed in the presence of nucleophiles
since the tropylium reacted more rapidly with the nu-
cleophiles than with the Mo complex. Nevertheless, a
slow reaction between PPN[Mo(C6F5)(CO)5] and
C7H7BF4 did occur in the absence of added nucleophile
Cyclic voltammetry was performed using a Bioana-
lytical Systems BAS 100B/W electrochemical analyzer.
Solutions contained approximately 0.1 M [Bu4N]PF6
supporting electrolyte. Electrodes consisted of a 0.5 mm
Pt disk working electrode, Pt wire auxiliary electrode,
and Ag/AgCl reference electrode. Potentials were refer-
enced to the formal potential of the decamethylferr-
ocenium–decamethylferrocene couple (Cpꢁ2Feþ/Cp2ꢁFe),
which we measure as approximately )0.1 V vs. Ag/
AgCl. Simulations of the cyclic voltammetry were run
using Bioanalytical Systems DigiSim 3.0 software.

J. Kim, S.L. Gipson / Polyhedron 23 (2004) 1371–1374
1373
in CDCl3 and the NMR spectra of the product solution
were consistent with C7H7C6F5 being the pentafluor-
ophenyl-containing product. Isolation of this species by
preparative TLC followed by mass spectral analysis
gave a parent ion at m/z 258.11 (C7H7C6F5) and frag-
ments at 167.03 (C6F5) and 91.08 (C7H7). Thus the
oxidation of [Mo(C6F5)(CO)5]ꢀ by C7H7BF4 is believed
to involve the following sequence of reactions:
½MoðC6F5ÞðCOÞ ꢂꢀ þ C7H7þ ! MoðC6F5ÞðCOÞ þ C7H7Å
5
5
MoðC6F5ÞðCOÞ þ S ! MoðCOÞ S þ C6F5Å
5
5
C7HÅ7 þ C6FÅ5 ! C7H7C6F5
where S in the above reactions could be solvent, the BF4ꢀ
anion, or adventitious water.
3.3. Simulation of cyclic voltammetry
Fig. 1. Experimental (solid lines) and simulated (circles) cyclic vol-
tammograms of 1.0 mM PPN[Mo(C6F5)(CO)5] + pyridine in CH2Cl2/
0.1 M [Bu4N]PF6 at a 0.5 mm diameter Pt disk electrode at 0.2 V/s.
Concentrations of pyridine labeled in mM. Optimized rate constant for
In order to understand more about the mechanism of
the apparent one-electron reductive elimination of C6F5
from Mo(C6F5)(CO)5, we have simulated the cyclic
voltammetry of PPN[Mo(C6F5)(CO)5] in CH2Cl2 in the
presence of pyridine using the reactions in Scheme 1.
At reasonably fast scan rates in the absence of pyri-
dine, the electron transfer reaction appears chemically
and electrochemically reversible and can be simulated
using a slightly slow electron transfer with a rate con-
stant of 0.06 cm/s and a transfer coefficient of 0.5. Fig. 1
shows the effect of pyridine concentration on the
apparent chemical reversibility of experimental and
simulated cyclic voltammograms of 1.0 mM PPN[Mo-
(C6F5)(CO)5] at a scan rate of 0.2 V/s. In the absence of
pyridine, simulations gave an optimum fit using a first-
order rate constant for reaction (1) of 0.22 sꢀ1. This
reaction almost certainly involves some nucleophile, but
since this species can be neither conclusively identified
nor quantitated, the reaction was modeled as a first-
order process.
k1 labeled in sꢀ1 and for k2 in Mꢀ1 sꢀ1
.
Fig. 1, optimization of k2 gave values ranging from 482
to 823 Mꢀ1 sꢀ1. Generally excellent agreement between
experimental and simulated results was thus achieved
using the parameters shown in Table 1 and an average
value for k2 of approximately 700 Mꢀ1 sꢀ1. This corre-
sponds to a relatively slow reaction, much slower than
the elimination of methyl radical from dimethylco-
balt(IV) macrocycles [9], though significantly faster than
the reaction of CpRu(CO)(PR3)CH3 with CH3CN [15].
The mechanism of the oxidatively induced one-elec-
tron reductive elimination is proposed to involve nu-
cleophilic attack (association) of the incoming ligand at
the Mo, followed by extrusion of the C6F5 radical. Es-
sentially identical results could be obtained by modeling
the reaction with pyridine as either a rate-limiting pre-
equilibrium between Mo(C6F5)(CO)5 and pyridine
followed rapid extrusion of C6F5 or a rapid pre-equi-
librium followed by a rate-determining homolysis of the
Mo–C6F5 bond. Therefore, the combination of nucleo-
With all other parameters held constant at values
given in Table 1, simulations were then run to obtain the
second-order rate constant, k2, for the reaction of pyri-
dine with Mo(C6F5)(CO)5, reaction (2). Optimum fits
were obtained with k2 values ranging from 646 to 1041
M
ꢀ1 sꢀ1. Fig. 2 illustrates the effect of scan rate on the
Table 1
Parameters used in simulation of cyclic voltammetry of
PPN[Mo(C6F5)(CO)5] in CH2Cl2 in the presence of pyridine
apparent chemical reversibility of experimental and
simulated cyclic voltammograms of 1.0 mM
PPN[Mo(C6F5)(CO)5] in the presence of 5.0 mM pyri-
dine. Using the same values for other parameters as in
a
kE
0.5
0.06 cm sꢀ1
Electrode area
Double layer
capacitance
0.0023 cm2
0.0013 lF
k1
0.22 sꢀ1
Uncompensated
resistance
2000 X
Keq1
Keq2
2.2 ꢃ 106
2.7 ꢃ 108
a, transfer coefficient; kE, standard heterogeneous rate constant for
electron transfer; k1, first-order rate constant for reaction in absence of
nucleophile; Keq, equilibrium constants for reactions (1) and (2) (see
Scheme 1).
Scheme 1.

1374
J. Kim, S.L. Gipson / Polyhedron 23 (2004) 1371–1374
systems [15], further reaction of this species is believed to
involve prior association of the incoming ligand. The
resulting 19-electron species then undergoes cleavage of
the metal–carbon bond, effecting a one-electron reduc-
tive elimination reaction. The products of the reaction
have all been identified by a combination of IR, NMR,
and mass spectrometries.
Acknowledgements
We thank the Robert A. Welch Foundation (Grant
No. AA-1083) and Baylor University, in part, for sup-
port of this research.
References
[1] S.L. Gipson, K. Kneten, Inorg. Chim. Acta 157 (1989) 143.
[2] M. Su, S.L. Gipson, D.F. Mullica, E.L. Sappenfield, D.H.
Leschnitzer, J. Chem. Soc., Dalton Trans. (1993) 3323.
[3] S.L. Gipson, D.F. Mullica, E.L. Sappenfield, M.L. Hander, D.H.
Leschnitzer, J. Chem. Soc., Dalton Trans. (1992) 521.
[4] D.F. Mullica, S.L. Gipson, M. Su, E.L. Sappenfield, Polyhedron
14 (1995) 2215.
Fig. 2. Experimental (solid lines) and simulated (circles) cyclic vol-
tammograms of 1.0 mM PPN[Mo(C6F5)(CO)5] + 5.0 mM pyridine in
CH2Cl2/0.1 M [Bu4N]PF6 at a 0.5 mm diameter Pt disk electrode. Scan
rates labeled in V/s. Optimized rate constants labeled in Mꢀ1 sꢀ1
.
[5] W.J. Doucette, J. Kim, J.A. Kautz, S.L. Gipson, Inorg. Chim.
Acta 304 (2000) 237.
philic attack and bond homolysis were modeled as the
single second-order reaction shown in Scheme 1. Con-
sistent with qualitative observations of the cyclic vol-
tammetry, the reductive elimination in the presence of
added nucleophile is significantly more rapid than in
pure solvent. This observation lends support to the
proposed associative nature of the overall reaction.
[6] H.C. Gardner, J.K. Kochi, J. Am. Chem. Soc. 97 (1975) 1855.
[7] J.Y. Chen, J.K. Kochi, J. Am. Chem. Soc. 99 (1977) 1450.
[8] D. Zhu, S.V. Lindeman, J.K. Kochi, Organometallics 18 (1999)
2241.
[9] W.H. Tamblyn, R.J. Klingler, W.S. Hwang, J.K. Kochi, J. Am.
Chem. Soc. 103 (1981) 3161.
[10] S.L. Borkowsky, N.C. Baenziger, R.F. Jordan, Organometallics
12 (1993) 486.
[11] M.J. Burk, W. Tumas, M.D. Ward, D.R. Wheeler, J. Am. Chem.
Soc. 112 (1990) 6133.
[12] R.F. Jordan, R.E. LaPointe, C.S. Bajgur, S.F. Echols, R. Willett,
J. Am. Chem. Soc. 109 (1987) 4111.
4. Conclusions
[13] E. Fooladi, T. Graham, M.L. Turner, B. Dalhus, P.M. Maitlis, M.
Tilset, J. Chem. Soc., Dalton Trans. (2002) 975.
[14] P. Diversi, M. Fontani, M. Fuligni, F. Laschi, S. Matteoni, C.
Pinzino, P. Zanello, J. Organomet. Chem. 626 (2001) 145.
[15] T. Aase, M. Tilset, V.D. Parker, J. Am. Chem. Soc. 112 (1990)
4974.
On the basis of experimentally observed cyclic vol-
tammetry, simulations of this cyclic voltammetry, and
products observed from chemical oxidation, a mecha-
nism for the oxidatively induced one-electron reductive
elimination of C6F5 from PPN[Mo(C6F5)(CO)5] has
been proposed. An initial one-electron oxidation pro-
duces the 17-electron species Mo(C6F5)(CO)5. In
agreement with previously published results for other
[16] J.C. Kotz, W. Vining, W. Coco, R. Rosen, Organometallics 2
(1983) 68.
[17] D.N. Hendrickson, Y.S. Sohn, H.B. Gray, Inorg. Chem. 10 (1971)
1559.