Elucidation of the wide range of reaction pathways that accompany the electrochemical oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)X] (dpm = Ph2PCH2PPh2; X = Cl, Br)

Article abstract of DOI:10.1021/ic980796r

The electrochemical oxidation of cis,mer-[Mn(CO)₂(η¹-dpm)(η²-dpm)Br] (dpm = Ph₂PCH₂PPh₂), or (cis,mer)⁰, has been examined in dichloromethane (0.1 M Bu₄NPF₆) by voltammetric, bulk electrolytic, in situ and ex situ spectroelectrochemical and simulation techniques. On the voltammetric time scale at 20 °C, the neutral 18-electron cis,mer Mn(I) species is oxidized to the corresponding 17-electron cation which at slow scan rates isomerizes to the trans cation. Simulations are consistent with a rate constant of 3.1 ± 0.3 s^-1 for this isomerization process. Monitoring the reaction by in situ IR spectroscopy at low-temperature enables the identification of the v(CO) bands of all four species ((cis,mer)⁰; (cis,mer)⁺; (trans)⁰; (trans)⁺) in the resultant square reaction scheme that is operative under these thin layer electrolysis conditions. Additionally, 17-electron cis,fac-[Mn(CO)₂(η¹-dpm)(η ²-dpm)Br]⁺ and its 18-electron (cis,fac)⁰ counterpart, generated by a redox-induced catalytic isomerization reaction, are detected and characterized by IR spectroscopy (v(CO)). Room-temperature bulk oxidative electrolysis experiments reveal that the trans cation, generated in bulk solution from the (cis,mer)⁺ and (cis,fac)⁺ isomers, slowly ejects bromide with a rate constant of 1.6 × 10^-3 s^-1 to form trans-[Mn(CO)₂(η²-dpm)₂]⁺. The equivalent voltammetry in acetonitrile is complicated by an additional competing kinetic step which is attributed to reaction of this cation with the solvent. However, the major product formed upon oxidation at room temperature is still the trans cation. Less detailed studies on the oxidation of cis,mer-[Mn(CO)₂(η¹-dpm)(η²-dpm)Cl] only show significant differences under conditions of bulk electrolysis after trans-[Mn(CO)₂(η²-dpm)₂]²⁺ is formed via expulsion of Cl^-.

Full text of DOI:10.1021/ic980796r

Inorg. Chem. 1999, 38, 2005-2011
2005
Elucidation of the Wide Range of Reaction Pathways That Accompany the Electrochemical
Oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)X] (dpm ) Ph2PCH2PPh2; X ) Cl, Br)
John C. Eklund,† Alan M. Bond,*,† Ray Colton,† David G. Humphrey,† Peter J. Mahon,† and
Jacky N. Walter‡
Department of Chemistry, Monash University, Clayton, Victoria 3168, Australia, and Department of
Chemistry, La Trobe University, Bundoora, Victoria 3083, Australia
ReceiVed July 10, 1998
The electrochemical oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br] (dpm ) Ph2PCH2PPh2), or (cis,mer)0,
has been examined in dichloromethane (0.1 M Bu4NPF6) by voltammetric, bulk electrolytic, in situ and ex situ
spectroelectrochemical and simulation techniques. On the voltammetric time scale at 20 °C, the neutral 18-electron
cis,mer Mn(I) species is oxidized to the corresponding 17-electron cation which at slow scan rates isomerizes to
the trans cation. Simulations are consistent with a rate constant of 3.1 ( 0.3 s-1 for this isomerization process.
Monitoring the reaction by in situ IR spectroscopy at low-temperature enables the identification of the ν(CO)
bands of all four species ((cis,mer)0; (cis,mer)+; (trans)0; (trans)+) in the resultant square reaction scheme that is
operative under these thin layer electrolysis conditions. Additionally, 17-electron cis,fac-[Mn(CO)2(η1-dpm)(η2-
dpm)Br]+ and its 18-electron (cis,fac)0 counterpart, generated by a redox-induced catalytic isomerization reaction,
are detected and characterized by IR spectroscopy (ν(CO)). Room-temperature bulk oxidative electrolysis
experiments reveal that the trans cation, generated in bulk solution from the (cis,mer)+ and (cis,fac)+ isomers,
slowly ejects bromide with a rate constant of 1.6 × 10-3 s-1 to form trans-[Mn(CO)2(η2-dpm)2]+. The equivalent
voltammetry in acetonitrile is complicated by an additional competing kinetic step which is attributed to reaction
of this cation with the solvent. However, the major product formed upon oxidation at room temperature is still
the trans cation. Less detailed studies on the oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Cl] only show
significant differences under conditions of bulk electrolysis after trans-[Mn(CO)2(η2-dpm)2]2+ is formed via
expulsion of Cl-.
Introduction
(c) The cationic species may react to form an entirely new
class of product.9,11
Electrochemical oxidation of neutral 18-electron transition
metal organometallic complexes containing combinations of
carbon monoxide and phosphine ligands has received much
attention.1-10 Upon one-electron oxidation, a number of reaction
routes for the initially formed isostructural 17-electron cation
are possible on either the voltammetric (short) or bulk elec-
trolysis (long) time scales:
(a) The resulting cation is highly stable. In this case a
chemically reversible one-electron process is observed.6-9
(b) The cation may isomerize to give the well-known square
reaction scheme.3,5,9
In this present study we have examined in detail the
electrochemical oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-
dpm)Br] (Chart 1, 1A) in dichloromethane and acetonitrile by
voltammetric, bulk electrolytic and spectroelectrochemical
techniques. As noted above, in principle on oxidation of 1A,
the resulting isostructural cation (1A+) may isomerize, eject a
ligand and/or undergo internal nucleophilic attack by the pendant
phosphorus atom associated with the η1-dpm ligand or external
attack by the solvent. We show that all of these processes and
also cross redox reactions occur during the course of oxidation
of 1A so that the electrochemistry is unusually complex. To
obtain kinetic parameters and verify postulated mechanisms, the
commercial package, DigiSim,12 is used to simulate the cyclic
voltammetric response in both dichloromethane and acetonitrile.
Less detailed studies are reported on the cis,mer-[Mn(CO)2(η1-
dpm)(η2-dpm)Cl] (1B) system to examine the influence of the
halide ligand on the redox chemistry.
† Monash University.
‡ La Trobe University.
* To whom correspondence should be sent. Tel.: +61 3 9905 4507.
Fax: +61 3 9905 4597. E-mail: alan.bond@sci.monash.edu.au.
(1) Geiger, W. E. Acc. Chem. Res. 1995, 28, 351.
(2) Geiger, W. E. Prog. Inorg. Chem. 1985, 33, 275.
(3) Compton, R. G.; Barghout, R.; Eklund, J. C.; Fisher, A. C.; Bond, A.
M.; Colton, R. J. Phys. Chem. 1993, 97, 1661.
(4) Bond, A. M.; Colton, R.; McCormick, M. Inorg. Chem. 1977, 16,
155.
(5) Bond, A. M.; Grabaric, B. S.; Grabaric, Z. Inorg. Chem. 1978, 17,
1013.
(6) Bond, A. M.; Colton, R.; McGregor, K. Inorg. Chem. 1986, 25, 2378.
(7) Bagchi, R. N.; Bond, A. M.; Colton, R.; Creece, I.; McGregor, K.;
Whyte, T. Organometallics 1991, 10, 2611.
Experimental Details
Reagents, Compounds, and Solvents. Acetonitrile (Mallinckrodt,
Biolab Scientific Pty Ltd. (HPLC grade, 99.9%)) and dichloromethane
(Mallinckrodt. (HPLC grade, 99.9%)) were dried for at least 12 h over
molecular sieves. The electrolytes were tetrabutylammonium hexa-
fluorophosphate (Bu4NPF6)13 and Bu4NClO4 (South Western Analytical,
(8) Bond, A. M.; Colton, R.; Gable, R. W.; Mackay, M. F.; Walter, J. N.
Inorg. Chem. 1997, 36, 1181.
(9) Bond, A. M.; Colton, R. Coord. Chem. ReV. 1997, 166, 161.
(10) Wimmer, F. L.; Snow, M. R.; Bond, A. M. Inorg. Chem. 1974, 13,
1617.
(11) Bond, A. M.; Colton, R.; McGregor, K. Organometallics 1990, 9, 1227.
(12) Rudolph, M.; Reddy, D. P.; Feldberg, S. W. Anal. Chem. 1994, 66,
589.
10.1021/ic980796r CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/09/1999

2006 Inorganic Chemistry, Vol. 38, No. 9, 1999
Eklund et al.
Chart 1a
Figure 1. Cyclic voltammograms obtained at a scan rate, 200 mV s-1
for the oxidation of 1A (1.0 mM) in dichloromethane (0.1 M Bu4NPF6)
at a 1 mm diameter Pt macrodisk electrode; (a) 20 °C; (b) -40 °C.
Conventional bulk electrolysis experiments combined with coulom-
etry were undertaken with either a PAR model 273 potentiostat or a
BAS 100A electrochemical analyzer using a large platinum basket
working electrode, a platinum gauze auxiliary electrode separated from
the test solution by a salt bridge and the same reference electrodes
used in the voltammetric studies.
31P NMR spectra were recorded on a Bruker AM300 spectrometer
operating at 121.4 MHz, and chemical shifts were referenced against
an external aqueous solution of 85% H3PO4. Ex situ Infrared spectra
were recorded on Perkin-Elmer FT-IR 1720X, Perkin-Elmer 1430 IR,
or Bio-Rad FT-infrared spectrometers. In situ IR spectroelectrochemical
experiments were carried out using a modified IR reflection-absorption
spectroscopy (IRRAS) cell,17,18 mounted on a specular reflectance
accessory located in the sample compartment of a Bruker IFS 55 FTIR
spectrometer. The electrode arrangement consisted of a polished
platinum disk working electrode (r ) 2.5 mm), a platinum gauze
auxiliary electrode and a silver wire pseudo-reference electrode.
Electrolyses were carried out by stepping the potential of the working
electrode from a rest potential to a potential sufficient to cause
electrolysis, and single scan IR spectra (resolution 1.0 cm-1) were
collected as a function of time. The cell was thoroughly flushed with
nitrogen prior to addition of the sample and maintained under a nitrogen
atmosphere throughout the experiments.
a For compounds above, X ) Br (A), Cl (B).
electrometric grade). Compounds 1A and 1B were prepared by standard
literature procedures.14,15 The assignment of the cis,mer isomeric form
was made on the basis of IR and 31P NMR data.14,15 All solutions were
thoroughly purged of oxygen by outgassing with nitrogen that had been
presaturated with solvent. The absence of incident irradiation16 was
ensured by wrapping all solution vessels in aluminum foil. Experiments
were conducted at either 20 ((1) or -40 ((1)°C.
Instrumentation and Procedures. Cyclic voltammetric experiments
employed 1.0 or 3.0 mm diameter platinum or glassy-carbon macrodisk
or 10 or 25 µm diameter platinum or 10 µm diameter glassy carbon
microdisk working electrodes in conjunction with Ag/Ag+ (0.01 M
AgNO3, 0.1 M Bu4NClO4 in acetonitrile) or Ag/AgCl (saturated LiCl,
0.1 M Bu4NPF6 in dichloromethane) reference electrodes and a platinum
wire counter electrode. All potentials are quoted relative to the reversible
potential for the ferricenium/ferrocene couple (Fc+/0). Rotating disk
electrode experiments used a platinum macrodisk electrode (diameter,
3.0 mm) which was rotated by a variable speed rotator (Metrohm 628-
10).
X-band EPR measurements were made with a Varian E-12 spec-
trometer. Positive ion electrospray mass spectroscopic experiments were
conducted on a Micromass Platform II quadrupole spectrometer.
Portions of the approximately 1 mM solutions used in dichloromethane
experiments were diluted 1:10 with acetonitrile. The diluted solution
was injected directly into the spectrometer and the solution was
delivered to the vaporization nozzle of the electrospray ion source, via
an acetonitrile mobile phase, at a flow rate of 5 µL min-1. Voltages at
the first skimmer electrode were the minimum necessary to maintain a
stable spray (typically 10-25 V).
To simulate cyclic voltammetric responses DigiSim V2.112 was run
on a 120 MHz Pentium PC.
A Cypress Systems (model CYSY-1R) potentiostat in conjunction
with a 386 personal computer or an ADI Instruments Maclab4e/
potentiostat system controlled by a Macintosh Powerbook microcom-
puter were used for voltammetric measurements.
Results and Discussion
A. Voltammetric Studies on cis,mer-[Mn(CO)2(η1-dpm)-
(η2-dpm)Br] (Chart 1, Compound 1A) in Dichloromethane.
Figure 1 depicts a low scan rate, cyclic voltammogram (scan
(13) Fry, A. J. In Laboratory Techniques in Electroanalytical Chemistry,
2nd ed.; Kissinger, P. T., Heineman, W. R., Eds.; Marcel Dekker:
New York, 1996; Chapter 15.
(14) Carriedo, G. A.; Riera, V.; Santamaria, J. J. Organomet. Chem. 1982,
234, 175.
(15) Carr, S. W.; Shaw, B. L.; Thornton-Pett, M. J. Chem. Soc., Dalton
Trans. 1985, 2131.
(16) Eklund, J. C.; Bond, A. M.; Colton, R.; Humphrey, D. G.; Mahon, P.
J.; Morrison, R. J., work in progress.
(17) Best, S. P.; Clark, R. J. H.; Cooney, R. P.; McQueen, R. C. S. ReV.
Sci. Instrum. 1987, 58, 2071.
(18) Best, S. P.; Ciniawsky, C. A.; Humphrey, D. G. J. Chem. Soc., Dalton
Trans. 1996, 2945.

Reaction Pathways of Electrochemical Oxidation
Inorganic Chemistry, Vol. 38, No. 9, 1999 2007
rate, 200 mV s-1) obtained at 20 °C for oxidation of 1A in
dichloromethane (0.1 M Bu4NPF6). Over the potential range
-0.60 to 0.40 V (vs Fc+/0) and in the first cycle, an irreversible
response is observed (process 1) with an oxidative peak potential
(EoxP) of +0.20 ((0.02) V. In addition, a reductive feature
(process 2′) is evident at EredP, -0.44 ((0.02) V, with its
corresponding oxidation peak (process 2) being detected in
second and subsequent cycles at Eoxp, -0.36 ((0.02) V. As
the scan-rate increases, a reduction peak current associated with
process 1 (described as process 1′) is detected (Figure S1).
Concurrently, the current magnitude of process 2′ decreases
relative to that of process 1. The scan-rate dependence suggests
that the product of process 1 reacts to form a new species that
in turn undergoes a chemically reversible electron-transfer
process at less positive potentials (i.e. to generate processes 2
and 2′ on second and subsequent cycles). If the reaction
temperature is decreased to -40 °C, redox couple 1 becomes
chemically reversible (see Figure 1), while processes 2 and 2′
disappear, indicating that the rate of the reaction following the
initial charge-transfer process is slowed upon lowering the
temperature.
Process 1 was further studied using rotating-disk electrode
(RDE) which showed that the oxidation exhibits Levich
behavior19 over the rotation speed range of 500-3000 rpm to
give a diffusion coefficient of (7 ( 1) × 10-6 cm2 s-1. Analysis
of the limiting current-mass transport behavior suggests that
process 1 involves a one-electron charge-transfer process at all
rotation rates. Fast scan-rate voltammetry using a 25 µm
diameter platinum microdisk electrode enables the reversible
potential (E°F) to be determined for process 1 (assuming equal
diffusion coefficients for reactant and product). In the scan rate
range of 2000-5000 mV s-1 the measured half-wave potential
(E1/2 ≈ E°F) is independent of scan rate, as expected when the
homogeneous kinetics following the one-electron-transfer pro-
cess are out-run. Thus, an E°F value for the formal reversible
potential of +0.16 ((0.01) V is obtained at 20 °C. The same
reversible potential (E1/2 ≈ 0.16 V) was obtained under near
steady-state conditions at platinum or glassy carbon microdisk
electrodes. These microdisk data also imply that the homoge-
neous kinetics also may be out-run under these near steady-
state conditions, implying that the value of the first-order rate
Figure 2. In situ IR spectra obtained during electrolysis experiments
in an IRRAS cell commencing with 1A (1.0 mM) in dichloromethane
(0.1 M Bu4NPF6) at -40 °C. (a) Change in IR absorbance, relative to
that of the initial IR spectrum of 1A during oxidative electrolysis at
+0.25 V (the v and V arrows signify that the IR intensity initially
increases and then decreases). (b) Change in IR absorbance relative to
that of the IR spectrum formed in (a) of 2A+ during reductive
electrolysis at -0.55 V.
of fac-[Mn(CO)3(η2-dpm)Cl]3-5 where six-line spectra were
observed and attributed to the formation of low-spin Mn(II)
species, fac-[Mn(CO)3(η2-dpm)Cl]+ and mer-[Mn(CO)3(η2-
dpm)Cl]+.4
The oxidation of 1A at -40 °C was monitored by IR
spectroscopy using an IRRAS cell. Upon oxidation, the two
ν(CO) bands from 1A at 1936 and 1866 cm-1 decrease in
intensity while unresolved bands initially grow at approximately
2031 and 2022 cm-1 and at 1960 and 1950 cm-1 (Figure 2a).
These new ν(CO) bands are attributed to the generation of two
cationic cis isomeric forms of oxidized 1A since the difference
in frequency of about 80 cm-1 relative to 1A is consistent with
this interpretation.21-24 The two possible cis cationic isomers
are the cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ (1950 and 2022
cm-1) and cis,fac-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ (1960 and
2031 cm-1) species now referred to as compounds 1A+ and
2A+, respectively (Chart 1). After initially increasing with time
(seconds) the intensity of the overlapping ν(CO) bands from
compounds 1A+ and 2A+ eventually decreases. The intense
single ν(CO) band observed to grow at 1971 cm-1 is assigned
to the formation of the trans-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+
constant (k1, see below) must be of the order of 1 s-1 20
.
An additional irreversible oxidative process is observed at
Eox ) +1.20 ((0.10) V (scan rate 100 mV s-1). The process
P
of interest in this paper is the first one-electron step, so no further
detail on the suggested MnII/MnIII oxidation process is presented.
B. Monitoring of Bulk Electrolysis Experiments on cis,-
mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br] Using a Combination
of in Situ and ex Situ Voltammetric and Spectroscopic
Techniques. Bulk oxidative electrolyses experiments on solu-
tions of 1A in dichloromethane (0.1 M Bu4NPF6) were
conducted at 20 and -40 °C in both conventional and thin-
layer IRRAS cells. Unless otherwise stated, the potential of the
platinum working electrode was held at +0.25 V which is
sufficiently positive to achieve oxidation at process 1. At both
20 and -40 °C process 1 was determined by coulometry
(exhaustive electrolysis conditions at a platinum basket elec-
trode) to be a one-electron process. The product(s) formed was
examined by ex situ EPR measurements and found to contain
six lines of equal intensity indicative of Mn(II) (55Mn, I ) 5/2).
The EPR spectrum is very similar to that found after oxidation
(21) Shaw, M. J.; Geiger, W. E. Organometallics 1996, 15, 13.
(22) Connelly, N. J.; Freeman, M. J.; Orpen, A. G.; Sheehan, A. R.;
Sheridan, J. B.; Sweigart, D. A. J. Chem. Soc., Dalton Trans. 1985,
1019.
(23) Atwood, C. G.; Geiger, W. E.; Bitterwolf, T. E. J. Electroanal. Chem.
1995, 397, 279.
(24) Connelly, N. G.; Kitchen, M. D. J. Chem. Soc., Dalton Trans. 1977,
931.
(19) Southampton Electrochemistry Group. Instrumental Methods in Elec-
trochemistry; Ellis Horwood: London, 1990.
(20) Montenegro, M. I. Res. Chem. Kinet. 1994, 2, 299.

2008 Inorganic Chemistry, Vol. 38, No. 9, 1999
Eklund et al.
and is replaced by a single ν(CO) band at 1891 cm-1 which is
assigned to the formation of the neutral 18-electron 3A (trans)
species (Figure 2b).
The 31P NMR spectrum of the product formed by the bulk
oxidative-reductive electrolysis sequence obtained in a conven-
tional cell was recorded at -40 °C with Bu4NClO4 being used
instead of Bu4NPF6 as the supporting electrolyte. Four lines of
equal intensity were observed as expected for formation of 3A
(Table 1). The 31P NMR signal at -31.8 ppm (trans-[Mn(CO)2-
(η1-dpm)(η2-dpm)Br]) is consistent with retention of the pendant
phosphorus. In addition, steady-state microdisk voltammetry
after this electrolysis sequence produced only one wave with
an identical half-wave potential and limiting current (now for
process 2) to that observed prior to reduction. However,
importantly the position of zero current indicates that the species
in solution is now in the reduced form (Figure 3c). When this
reduced solution was warmed to 20 °C, steady-state voltam-
mograms indicate that concomitantly the limiting current for
process 2 decreases and that for process 1 grows until process
Figure 3. Microdisk near steady-state voltammograms obtained at -40
°C with a 25 µm diameter platinum electrode (scan rate 10 mV s-1
)
from an initially 1.0 mM solution of 1A in dichloromethane (0.1 M
Bu4NPF6) (a) before electrolysis, (b) after electrolysis at + 0.25 V,
and (c) after reduction of the solution formed in (b) at -0.55 V.
2 is not observed. In addition the ν(CO) band at 1891 cm-1
,
(compound 3A+, Chart 1). An additional minor product with
ν(CO) bands at 1878 and 1945 cm-1 detected as an intermediate
also disappeared upon complete electrolysis. Since 1A (cis,-
mer) has IR bands at 1866 and 1936 cm-1, the transient species
with a similar IR spectrum is attributed to the generation of a
small amount of cis,fac-[Mn(CO)2(η1-dpm)(η2-dpm)Br] (com-
pound 2A). If only partial rather than exhaustive electrolysis is
undertaken at -40 °C, then the IR spectrum shown in Figure
S2 is observed. This spectrum implies that catalytic isomeric
conversion of 1A (cis,mer) to 2A (cis,fac) occurs in good yield
in the presence of only trace amounts of the 1A+ (cis,mer)
cation.
assigned to 3A (trans), decreases and two new bands grow at
1936 and 1866 cm-1 as expected for formation of 1A (cis,-
mer), which must therefore be the thermodynamically favored
18-electron form of [Mn(CO)2(η1-dpm)(η2-dpm)Br].25
When bulk electrolysis experiments were conducted at 20
°C in the conventional cell the pink color associated with the
formation of 3A+ gradually disappeared when the oxidized
solution was left to stand (no applied potential) and the solution
became pale yellow. Cyclic voltammetric monitoring of this
experiment after completion of the electrolysis (Figure S3)
reveals that the major response detected is associated with redox
couple 2. However, a previously undetected reversible process
(couple 3) is also observed with a half-wave potential of 0.37
((0.02) V. Further monitoring by cyclic voltammetry shows
that couple 3 continues to grow until 30 min post-bulk
electrolysis, redox couple 2 is not detected while couple 3 is
extremely well defined. The 31P NMR spectrum from bulk
electrolysis in dichloromethane (0.1 M Bu4NClO4) obtained at
this time shows a single resonance having a chemical shift of
35.3 ppm. In situ IR spectroelectrochemical oxidation experi-
ments in the IRRAS cell at room temperature lead to a decrease
in the intensity of the 1A (cis,mer) ν(CO) bands at 1937 and
1866 cm-1 and concomitant growth of a single sharp band at
1971 cm-1 due to formation of 3A+ (trans), whereas the IR
spectrum of the solution oxidized in a conventional cell,
collected some 30 min after oxidation, exhibits a single carbonyl
stretch at 1916 cm-1. The IR and NMR data of this latter species
are similar to that reported14 for the trans-[Mn(CO)2(η2-dpm)2]+
cation or compound 4A+ (trans) (1916 cm-1 and 33 ppm,
respectively) which suggests that this compound is formed
slowly from 3A+. This assignment was supported by the
positive-ion electrospray mass spectrum obtained from the bulk
electrolyzed solution which produced a strong signal at 879 (m/
e), as expected for the 4A+ cation. The simulated and
experimental mass spectra for this formulation of the cation were
in excellent agreement.
Presumably the 2+/0 (cis,fac) redox couple has a similar
reversible potential to that for the 1+/0 (cis,mer) process so that
2A formed in a bulk electrolysis experiment is oxidized to 2A+
which like the cis,mer cation is also then isomerized to the trans
cationic isomer (see above), as shown by the detection of the
ν(CO) band at 1971 cm-1 (Figure S2). However, after only a
small amount of electrolysis, all species required for catalytic
isomeric conversion of 1A (cis,mer) to 2A (cis,fac) are present
in bulk solution. The reactions summarizing the catalytic
generation of 2A after formation of small amounts of 1A+ from
1A in bulk solution at early stages in the low-temperature
oxidation in the thin-layer IR spectroelectrochemical cell are:
1A+ (cis,mer) f 2A+ (cis,fac)
(1)
2A+ (cis,fac) + 1A (cis,mer) f
2A (cis,fac) + 1A+ (cis,mer) (2)
Under the conditions of electrolysis at -40 °C in a conven-
tional cell, only reductive process 2′ was observed in the cyclic
and steady-state (Figure 3) microdisk voltammetry after exhaus-
tive electrolysis, and there was no spectroscopic or voltammetric
evidence of liberated Br-, CO or dpm. From steady-state
microdisk electrode voltammetry, the magnitude of the reductive
current associated with process 2′ is the same as the equivalent
current associated with process 1 before bulk electrolysis (Figure
3c). Thus, all electrolysis data are consistent with the formation
of an essentially quantitative yield of the trans cation (3A+)
with 1A+ (cis,mer) and 2A+ (cis,fac) cations and neutral 2A
(cis,fac) species being generated at earlier stages of the
experiment.
On the basis of all the above evidence, it is therefore
postulated that redox couple 3 corresponds to the reaction
E
4+ (trans) a 42+ (trans) + e-
(3)
This process is analogous to that of the reversible [Mn(CO)2-
(η2-dpe)2]0/+ reaction which has a half-wave potential of +0.60
On rereduction in the IRRAS cell at - 0.55 V and -40 °C,
the IR signal at 1971 cm-1 for 3A+ (trans) decays with time
(25) Mingos, D. M. P. J. Organomet. Chem. 1979, 179, C29.

Reaction Pathways of Electrochemical Oxidation
Inorganic Chemistry, Vol. 38, No. 9, 1999 2009
Table 1. IR and 31P NMR Spectroscopic Data Obtained for the Species Associated with the Electrochemical Oxidation of
cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br] in Dichloromethane
ν(CO)/cm-1
δ (31P)/ppm
cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br]
cis,fac-[Mn(CO)2(η1-dpm)(η2-dpm)Br]
cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ a
cis,fac-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+ a
trans-[Mn(CO)2(η1-dpm)(η2-dpm)Br]
trans-[Mn(CO)2(η1-dpm)(η2-dpm)Br]+
trans-[Mn(CO)2(η2-dpm)2]+
1866
1936
1945
2022
2031
50.3
35.0
49.3
0.3
-28.1
-31.8
1878
1950
1960
1891
1971
1916
53.2
35.3
19.0
a Deduced by reference to ν(CO) data obtained for the isostructural neutral analogues.
((0.10) V (vs Fc+/0).10 The significantly more positive potential
for the dpe analogue (+0.60 vs +0.37 V) is in line with the
general observation that dpe species are oxidized at more
positive potentials than their dpm counterparts.6,26,27
Analysis of the rate of increase of ν(CO) at 1916 cm-1 (4A+)
and decrease of ν(CO) 1971 cm-1 (3A+) using the IRRAS cell
by a first-order rate law gave a rate constant of 1.6 ((0.2) ×
10-3 s-1 at 20 °C (k2, see below) for the conversion 3A+ to
4A+.
All of the evidence obtained from these bulk electrolysis
experiments imply that process 1, at short voltammetric time
domains, must correspond to the oxidation of 1A (cis,mer) to
the isostructural 17 electron cationic manganese(II) species 1A+
(cis,mer) and that process 1′ is the reverse reduction process,
so that couple 1 is the reaction:
dication is a strong oxidant and can oxidize bromide to bromine,
so that steps 4-6 may be followed by:
C
C
3A+ (trans) k2(slow)8 4A2+ (trans) + Br-
4A2+ (trans) + Br- k3(fast)8 4A+ (trans) + 1/2Br2
(7)
(8)
where reaction 7 is the rate-determining step and k2 has a value
of 1.6 ((0.2) × 10-3 s-1
.
Thus, at 20 °C the overall oxidation of cis,mer-[Mn(CO)2-
(η1-dpm)(η2-dpm)Br] on the time scale of bulk electrolysis is
represented by the reaction
cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br] f
trans-[Mn(CO)2(η2-dpm)2]+ + 1/2Br2 + e- (9)
E
1A (cis,mer) a 1A+ (cis,mer) + e-
(4)
However, 1A apparently is unstable on longer time scales so
that under slow scan rate cyclic voltammetric conditions, 1A+
(cis,mer) isomerizes to 3A+ (trans). This result is analogous to
oxidation of cis-[Cr(CO)2(η2-dpe)2]+ which yields the cis cation
at short times, but the trans-[Cr(CO)2(η2-dpe)2]+ isomer on long
time-scale experiments.28 Thus, if processes 2 and 2′ are
associated with the redox couple:
C. Simulation of the Room-Temperature Voltammetric
Oxidation of cis,mer-[Mn(CO)2(η1-dpm)(η2-dpm)Br]. The
cyclic voltammetric response obtained at 20 °C was simulated
according to the square reaction scheme described by eqs 4-6
in combination with the slow bromide expulsion step described
by eq 7 using the DigiSim voltammetric simulation package.12
The formal potentials for the redox processes 1/1′ (0.16 ( 0.02
V) and 2/2′ (-0.40 ( 0.03 V) were obtained as described above.
Double-layer capacitance values of 25 µF cm-2 and uncom-
pensated resistances of 2000 Ω were used in the simulations
which are typical of the solvent/electrolyte system of interest.29-31
The isomerization process was assumed to be completely
irreversible. Thus, solely for the purposes of the simulation, a
very high (1010), but chemically insignificant value of the
equilibrium constant was used for the C step. The kinetic
parameter (k2) used for the debromination of trans-[Mn(CO)2-
(η1-dpm)(η2-dpm)Br]+ was the value obtained from IR experi-
ments, although it eventuated that this rate constant is too slow
to be significant at the scan rates employed in experimental
studies. The diffusion coefficients of all the species in the
electrode reaction mechanism were set to 7 × 10-6 cm2 s-1
(the value determined for compound 1A using microdisk and
rotating disk voltammetry).
E
3A (trans) a 3A+ (trans) + e-
(5)
then it follows that the slow scan rate voltammetric oxidation
of 1A occurs in dichloromethane via an EC scheme:
E
C
1A (cis,mer) a 1A+ (cis,mer) + e-
(4)
(6)
k1
1A+ (cis,mer)
9
8 3A+ (trans)
This EC reaction scheme assumes that the catalytic process
which generates 2A (cis,fac) under bulk electrolysis conditions
is too slow to be significant on the voltammetric time scale.
Additionally, the reaction 1A+ (cis,mer) f 2A+ (cis,fac) may
occur more rapidly than reaction 6 and hence not be detected
under these conditions (not rate determining) or else it may not
be detectable because of lack of resolution of the [1A]+/0 and
[2A]+/0 processes.
As can be seen in Figure 4 excellent agreement is obtained
between experiment and theory at scan rates of 100 and 1000
mV s-1 for the proposed mechanism in dichloromethane when
the heterogeneous charge-transfer rate constants for electron-
transfer in steps 4 and 5 were both set at 0.06 ((0.03) s-1 and
when a k1 value of 3.1 ((0.3) s-1 is used. Similarly good fits
Since the 18-electron 4+ compound is the final stable product
of bulk electrolysis, it can be concluded that the 17-electron
3A+ slowly releases bromide to give 4A2+. However, this trans
(26) Szczepura, L. F.; Giambra, J.; See, R. F.; Lawson, H.; Janik, T. S.;
Jircitano, A. J.; Churchill, M. R.; Takeuchi, K. J. Inorg. Chim. Acta
1995, 239, 77.
(27) Bond, A. M.; Colton, R.; Humphrey, D. G.; Mahon, P. J.; Snook, G.
A.; Tedesco, V.; Walter, J. N. Organometallics 1998, 17, 2977.
(28) Bond, A. M.; Colton, R.; Daniels, F.; Fernando, D. R.; Marken, F.;
Nagaosa, Y.; Van Steveninck, R. F. M.; Walter, J. N. J. Am. Chem.
Soc. 1993, 115, 9556.
(29) Kadish, K. M.; Ding, J. Q.; Malinski, T. Anal. Chem. 1984, 56, 1741.
(30) Bond, A. M.; Oldham, K. B.; Snook, G. A. Monash University, 1998.
These resistance values were obtained via AC impedance measure-
ments on a cell of similar configuration to the one used in this study.
(31) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals
and Applications; John Wiley: New York, 1980.

2010 Inorganic Chemistry, Vol. 38, No. 9, 1999
Eklund et al.
Figure 4. Comparison of experimental and simulated (according to
the mechanism described by eqs 4-7 and the parameters given in the
text) voltammograms obtained at a 1 mm diameter platinum macrodisk
electrode for the oxidation of 0.92 mM 1A at 20 °C in dichloromethane
(0.1 M Bu4NPF6). (a) Scan rate, 100 mV s-1 (k1 ) 2.9 s-1), (b) scan
rate, 1000 mV s-1 (k1 ) 3.4 s-1).
Figure 5. Experimental and simulated (according to the mechanism
described by eqs 4-6 and 10 and the parameters given in the text)
voltammograms obtained at a 1 mm diameter platinum macrodisk
electrode using a scan rate of 100 mV s-1 for the oxidation of 1.04
mM 1A at 20 °C in acetonitrile (0.1 M Bu4NPF6). (a) k1 ) 0.5 s-1
;
single C step (eq 6) EC mechanism, (b) k1 ) 2 s-1 and k4 ) 4 s-1; two
were obtained for other scan rates in the range of 100-2000
mV s-1 and concentrations of 1A (0.2-1.0 mM).
parallel C steps (eqs 6 and 10) mechanism.
D. Voltammetry and Bulk Electrolysis in Acetonitrile.
Qualitatively, the voltammetry of 1A in acetonitrile is similar
to that in dichloromethane. Thus, (i) the reversible E1/2 value
for the [1A]+/0 couple is +0.26 ((0.02) V (vs Fc+/0) as
determined from fast scan rate cyclic voltammetric measure-
ments at a microdisk electrode, (ii) the E1/2 value for the [3A]+/0
couple is -0.26 ((0.02) V (slow scan rate voltammetry at a
macrodisk electrode), and (iii) the irreversible Mn(II) to Mn-
(III) oxidation process has a peak potential of +0.53 ((0.05)
dichloromethane, results from the presence of a homogeneous
process (additional to the isomerization step), C′ (eq 10), which
removes 1A+ from near to the electrode surface:
k4
C′
1A+ (cis,mer)
98 products
(10)
If this additional homogeneous chemical step is included into
the simulations with a first-order or pseudo-first-order rate
constant, k4, of 4 s-1, reasonable agreement is obtained between
the experimental and theoretical voltammograms (Figure 5b).
Due to the substantial nucleophilicity of acetonitrile,15 it is
suggested that the 1A+ (cis,mer) cation may react with this
solvent at a similar rate to that for the isomerization process in
an analogous manner to that of voltammetrically generated [Cr-
V at a scan rate of 100 mV s-1
.
However, it can be seen (Figure 5a) that poor agreement
between experiment and simulation is observed in acetonitrile
when the reaction scheme represented by eqs 4-6 is used
because the experimentally observed peak current values for
both processes 2′ and 1′ are both lower than the simulated
values. Furthermore, these peak currents cannot be simulated
because if the value of k1 (the only variable parameter in the
simulation) is increased, the peak currents for processes 1′ and
2′ will decrease and increase, respectively. In this solvent, the
parameters utilized in the dichloromethane voltammetric simula-
tions were used to calculate theoretical voltammograms, except
for the diffusion coefficient of 1.0 ((0.1) × 10-5 cm2 s-1
obtained from rotating-disk experiments, an uncompensated
resistance of 500 Ω, the reversible potentials given above and
the debromination step (eq 7) was neglected. It is suggested
that the failure to simulate the peak currents associated with
both processes 1′ and 2′, with the mechanism found in
6
+ 32
(CO) (η -arene)]
or [Cr(CO)4(η2-dpm)]+ 33 or chemically
3
generated [Cr(CO)4(η2-dpe)]+
17-electron cations which
34
undergo nucleophilic attack by CH3CN and elimination of CO
ligands.
In situ IR oxidative spectroelectrochemical experiments at
20 °C on 1A in acetonitrile (0.1 M Bu4NPF) at +0.35 V, reveal
that the ν(CO) bands at 1935 and 1867 cm-1 are replaced by a
(32) Compton, R. G.; Barghout, R.; Eklund, J. C.; Fisher, A. C.; Davies,
S. G.; Metzler, M. R. J. Chem. Soc., Perkin Trans. II 1993, 39.
(33) Bond, A. M.; Colton, R.; Kevekordes, J. E.; Panagiotidou, P. Inorg.
Chem. 1987, 26, 1430.
(34) Connor, J. A.; Riley, P. I.; Rix, C. J. J. Chem. Soc., Dalton Trans.
1977, 1317.

Reaction Pathways of Electrochemical Oxidation
Inorganic Chemistry, Vol. 38, No. 9, 1999 2011
strong single band at 1973 cm-1 and a weak band at 1874 cm-1
.
the Mn(II) unit is not a sufficiently strong oxidant to induce
elimination of Cl•, so that Cl- must be ejected, leaving the
central manganese atom in oxidation state II. In this case, the
reduction of trans-[Mn(CO)2(η2-dpm)2]2+ to the Mn(I) trans-
[Mn(CO)2(η2-dpm)2]+ complex must occur heterogeneously at
the electrode surface.
The band at 1973 cm-1 corresponds to the formation of 3A+ as
deduced by comparison with IR data obtained in dichlo-
romethane (1971 cm-1). It is postulated that the weak band at
1874 cm-1 results from a carbonyl ligand of 1A+ being replaced
by acetonitrile in a pseudo-first-order process. However, the
identity of this finally formed acetonitrile substituted species
or any intermediates has not been determined.
Conclusions
E. Electrochemical Oxidation of cis,mer-[Mn(CO)2(η1-
dpm)(η2-dpm)Cl] in Dichloromethane (Chart 1, Compound
1B). Comparative voltammetric and electrolysis experiments
also were conducted on 1B in dichloromethane at 20 °C. This
species exhibits a voltammetric response that is closely related
to the bromo-anologue. Thus, while the potentials of the
At 20 °C, slow scan rate cyclic voltammetry and bulk
electrolysis experiments suggest that 1A+ (cis,mer) formed upon
oxidation of 1A (cis,mer) rapidly isomerizes to 2A+ (trans),
the overall first-order rate constant for this isomerization reaction
being 3.1 ((0.3) s-1. On standing in solution after bulk
electrolysis experiments, 2A+ slowly reacts at 20 °C to give 4+
(trans) and bromine, the rate constant for the expulsion of
bromide being 1.6 ((0.2) × 10-3 s-1. Under conditions of bulk
electrolysis at -40 °C, IR evidence for the 17-electron 1A cation
is obtained. Additionally, at this temperature IR evidence for
catalytic generation of 2A (cis,fac) as well as its isostructural
cation, 2A+, have been obtained. This appears to be the first
time that evidence has been obtained for the cis,fac isomeric
forms in [Mn(CO)2(η1-dpm)(η2-dpm)Br] type systems. In the
thermodynamic sense, cis,mer is the favored isomeric form in
the 18-electron configuration, whereas the trans geometry is the
favored form in the oxidized 17-electron system. However,
interconversions between the two thermodynamically favored
states may occur via the cis,fac geometric configuration.
In acetonitrile, the isomerization processes are complicated
by what is believed to be a parallel reaction of the oxidized
cations with the solvent.
processes for 1B are slightly more positive (Eox ) +0.20 V
p
(process 1); Eox ) -0.34 V, Ered ) -0.42 V (processes 2
p
p
and 2′), scan-rate ) 200 mV s-1) than for 1A, bulk oxidative
electrolysis at 0.30 V also led to 4+ (trans) being detected as
the final stable product by voltammetric, IR, and electrospray
mass spectroscopic measurements. However, since released
chloride is much more difficult to oxidize than bromide,35 42+
does not convert chloride to chlorine. In this case, it was
therefore necessary to maintain the electrode potential at 0.30
V until the solution had changed color from pink to pale yellow
in order to obtain a quantitative yield of the 4+ cation. At the
potential of +0.30 V any 42+, formed by expulsion of Cl- from
the 3B+ (trans) produced via isomerization of 1B+ (cis,mer),
is immediately reduced at the electrode surface to 4+ to give
the overall reaction:
1B (cis,mer) f 4+ (trans) + Cl-
(11)
The oxidation of 1B in dichloromethane parallels that of the
bromine analogue on the voltammetric time scale. However,
the lower redox reactivity of released Cl- leads to differences
in the reaction mechanism on the bulk electrolysis time scale.
A possible alternative mechanistic description to explain the
long time-scale bulk oxidative electrolysis of 1A would be to
propose that 3A+ undergoes an oxidatively induced reductive
elimination reaction (akin to those observed for [Cp2MoCl2]36,37
and [Cp2TiCl2]38) in which a bromine atom rather than the
bromide anion is ejected and replaced by the pendant phosphine
arm of the η1-dpm ligand. However, for the chloride analogue,
Acknowledgment. We thank the Sir Robert Menzies Centre
for Australian Studies for awarding a Postdoctoral Fellowship
to J.C.E. and the A.R.C. for financial support of the project, as
well as an Endeavour Fellowship for D.G.H. We also thank
Adrian van den Bergen for synthesis of cis,mer-[Mn(CO)2(η1-
dpm)(η2-dpm)Br] and Dr. Georgii Lazarev for assistance with
the EPR measurements.
(35) Mussini, T.; Faita, G. Encycloepedia of Electrochemistry of the
Elements; Bard, A. J., Ed.; Marcel Dekker: New York, 1975; Vol. 1,
Chapters 1 and 2.
(36) Kotz, J. C.; Vining, W.; Coco, W.; Rosen, R.; Dias, A. R.; Garcia,
M. H. Organometallics 1983, 2, 68.
(37) Compton, R. G.; Booth, J.; Eklund, J. C. J. Chem. Soc., Dalton Trans.
1994, 1711.
Supporting Information Available: Figures S1-S3 are available
as Supporting Information. This material is available free of charge
via the Internet at http://pubs.acs.org.
(38) Anderson, J. E.; Sawtelle, S. M. Inorg. Chem. 1992, 31, 5345.
IC980796R