Electron transfer and chemical reactions associated with the oxidation of an extensive series of mononuclear complexes [M(CO)2(k 1-P-P)(k2-P-P)X] and binuclear complexes [{M(CO) 2(k2-P-P)X}2</sub

Article abstract of DOI:10.1021/om034388t

The electrochemical oxidation of extensive series of mononuclear and binuclear complexes was discussed. It was found that the synthesis of manganese and rhenium dicarbonyl halide with variety of diphosphene ligands was carried out. The manganese complexes

Full text of DOI:10.1021/om034388t

3164
Organometallics 2004, 23, 3164-3176
Electr on Tr a n sfer a n d Ch em ica l Rea ction s Associa ted
w ith th e Oxid a tion of a n Exten sive Ser ies of
Mon on u clea r Com p lexes [M(CO)2(K1-P -P )(K2-P -P )X] a n d
Bin u clea r Com p lexes [{M(CO)2(K2-P -P )X}2(µ-P -P )]
(M ) Mn , Re; P -P ) Dip h osp h in e or Rela ted Liga n d ;
X ) Cl, Br )
Alan M. Bond,*,1 Ray Colton,1 Adrian van den Bergen,1 and J acky N. Walter2
School of Chemistry, P.O. Box 23, Monash University, Victoria 3800, Australia, and
Department of Chemistry, La Trobe University, Bundoora, Victoria 3083, Australia
Received December 18, 2003
The electrochemical oxidation of an extensive series of cis,mer-M(CO)2(κ1-dpm)(κ2-P-P)X
(M ) Mn, Re; dpm ) Ph2PCH2PPh2; P-P ) dpm, Ph2PCH2CH2PPh2 (dpe), o-(Ph2P)2C6H4-
(dpbz); X ) Cl, Br) complexes has been investigated on both the voltammetric and bulk
electrolysis time scales. At short time domains or low temperatures, the manganese
complexes undergo a reversible one-electron oxidation to cis,mer-[Mn(CO)2(κ1-dpm)(κ2-P-
P)X]+. These compounds isomerize under slow scan rate voltammetric conditions at room
temperature to give trans-[Mn(CO)2(κ1-dpm)(κ2-P-P)X]+, which on even longer bulk elec-
trolysis time scales slowly lose X- to form a reactive trans-[Mn(CO)2(κ2-dpm)(κ2-P-P)]2+
intermediate. In turn, this complex is reduced either at the electrode surface (X ) Cl) or in
a homogeneous chemical reaction (X ) Br) to form trans-[Mn(CO)2(κ2-dpm)(κ2-P-P)]+, which
is the final product observed under conditions of bulk electrolysis. In contrast, the first
oxidation step for the corresponding rhenium complexes involves oxidation of the pendant
phosphorus atom and reaction with traces of water to give cis,mer-Re(CO)2(κ1-dpmO)(κ2-
dpm)X and then cis-[Re(CO)2(κ2-dpmO)2]+, with the rhenium center subsequently being
oxidized at more positive potentials. Methylation of the pendant phosphorus of cis,mer-Re-
(CO)2(κ1-dpm)(κ2-dpm)X gives cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)X]+. The ligand-based
oxidation pathway is blocked by the methylation, so that only the usual metal-centered Re-
(I)/Re(II)-based oxidation processes are observed in the voltammetry of these compounds.
The compounds cis,mer-Re(CO)2(κ1-ape)(κ2-ape)X and cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)X]+
(ape ) Ph2AsCH2CH2PPh2) exhibit electrochemistry similar to that of the dpm analogues.
The major products of the interaction of dpe with the metal pentacarbonyl halides (2:1) are
the new binuclear species {M(CO)2(κ2-dpe)X}2(µ-dpe), whose structures have been determined
by IR and 31P NMR spectroscopy and electrospray mass spectrometry (ESMS). The
stereochemistry of the manganese complexes is cis,fac, and upon voltammetric oxidation,
first one and then the second manganese atom are oxidized. Bulk oxidative electrolysis at
low temperature leads to the generation of cis,fac-[{Mn(CO)2(κ2-dpe)X}2(µ-dpe)]+ and cis,-
fac-[{Mn(CO)2(κ2-dpe)X}2(µ-dpe)]2+. The stereochemistry of the oxidized metal centers
isomerizes to trans+ at room temperature, but the binuclear structure is retained. Bulk
reductive electrolysis at low temperature after bulk oxidative electrolysis at room temperature
leads to the characterization of trans-{Mn(CO)2(κ2-dpe)X}2(µ-dpe). The binuclear rhenium
complexes cis,mer-{Re(CO)2(κ2-P-P)X}2(µ-P-P) (P-P ) dpe, Ph2P(CH2)3PPh2 (dpp)) were
isolated, and they each undergo two successive one-electron oxidations to generate the mono-
and dications without any isomerization on the voltammetric time scale. ESMS provides
independent proof of the binuclear nature of the oxidized species.
In tr od u ction
where subsequently the hapticity symbol, κ, is often
omitted for simplicity for these and other compounds
in which all the diphosphine ligands are bidentate. In
contrast, reactions with 2 mol equiv of P-P give a range
of dicarbonyl products which appear to be dependent
The interactions of the group 7 metal carbonyl halides
M(CO)5X (M ) Mn, Re; X ) Cl, Br, I) with 1 mol equiv
of a variety of diphosphines (P-P) have been investigated
extensively and give exclusively fac-M(CO)3(κ2-P-P)X,3-6
* To whom correspondence should be addressed. Fax: (61)(3) 9905
4597. E-mail: alan.bond@sci.monash.edu.au.
(1) Monash University.
(4) Reimann, R. H.; Singleton, E. J . Organomet. Chem. 1972, 38,
113.
(5) Edwards, D. A.; Marshalsea, J . J . Organomet. Chem. 1975, 96,
C50.
(2) La Trobe University.
(3) Osborne, A. G.; Stiddard, M. H. B. J . Chem. Soc. 1962, 4715.
(6) Colton, R.; McCormick, M. J . Inorg. Chem. 1977, 16, 155.
10.1021/om034388t CCC: $27.50 © 2004 American Chemical Society
Publication on Web 05/14/2004

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3165
IR and 31P NMR spectra agree with literature values,10,11 but
the yields were found to be significantly higher (90-95%). The
trans-[Mn(CO)2(κ2-dpe)2]X and trans-Mn(CO)(κ2-dpe)2X (X )
Cl, Br) complexes were synthesized by refluxing a mixture of
0.5-0.7 g of Mn(CO)5X and dpe in a 1:2 molar ratio for 5-7 h
in toluene. When the mixture was cooled to room temperature,
trans-[Mo(CO)2(κ2-dpe)2]X precipitated. Cooling to 0 °C and
standing overnight resulted in crystallization of trans-Mn(CO)-
(κ2-dpe)2X. Spectroscopic data for these complexes agree with
literature values. The mixed-diphosphine complexes Mn(CO)2-
(κ1-dpm)(κ2-P-P)Br (P-P ) dpe, dpbz; dpbz ) o-(Ph2P)2C6H4)
were prepared in a way similar to that for cis,mer-Mn(CO)2-
(κ1-dpm)(κ2-dpm)X by the interaction of 0.7 g of fac-Mn(CO)3-
(P-P)Br with a 1:1 molar mixture of dpm, except that a longer
reflux time of 22 h was employed in refluxing heptane. Yields
obtained were in the range of 90 ( 5%. Analogous reactions
of 0.5-0.7 g of Mn(CO)5X and dpe (1:2 molar ratio) yielded a
mixture of compounds with a reflux time of 13 h for the chloro
compounds and 22 h for the bromo compounds. After removal
of the solvent, recrystallization from dichloromethane/hexane
gave as the major product (85-90% yield) a solid which was
characterized (next section) as orange needles of the binuclear
complex cis,fac-{Mn(CO)2(dpe)X}2(µ-dpe). Evaporation of the
filtrate and several recrystallizations from dichloromethane
gave cis-Mn(CO)2(κ1-dpe)(κ2-dpe)X as a minor product (e5%
yield).
(iii) P r ep a r a tion of Rh en iu m Com p lexes. The com-
plexes cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)X (structure type I)
were prepared by the literature method described for the chloro
compound.10 Thus, 0.5 g of Re(CO)5XC and dpm (1:2 molar
ratio) were refluxed in methylene for 6 h. However, the 31P
NMR spectra showed the presence of weak resonances in
addition to those expected for cis,mer-Re(CO)2(κ1-dpm)(κ2-
dpm)X. A precipitate was collected by gravity filtration,
washed with hexane, and air-dried. Passing a dichloromethane
solution of the mixture of compounds through a silica column,
followed by addition of hexane to the eluted solution, gave an
85 ( 5% yield of the pure complex cis,mer-Re(CO)2(κ1-dpm)-
(κ2-dpm)X, obtained as a white solid. The impurity mentioned
above was isolated by further elution of the silica column with
acetone and subsequent removal of the solvent under vacuum
(e5% yield). It was characterized as the known11 cis,mer-Re-
(CO)2(κ1-dpmO)(κ2-dpm)X (structure II), previously generated
upon the identity of the ligand. Thus, Mn(CO)5X is
reported to react with dpm (dpm ) Ph2PCH2PPh2) to
give cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)X,7,8 while with
dpe (dpe ) Ph2PCH2CH2PPh2) the compounds trans-
[Mn(CO)2(κ2-dpe)2]X9 have been prepared. Analogous
products have been reported for rhenium.10-12
In this paper, we have synthesized manganese and
rhenium dicarbonyl halides with a wide variety of
diphosphine ligands. The dicarbonyl complexes prepared
include further examples of the aforementioned mono-
meric types, together with new binuclear species. Ex-
tensive oxidative electrochemical studies on both vol-
tammetric and bulk electrolysis time scales are described
for all the species. The electrochemical studies reveal a
wide range of reaction pathways for both the monomeric
and dimeric dicarbonyl compounds, with the details
depending on the metal (Mn or Re), the halide (Cl or
Br), and the coordination mode of the diphosphine
ligand. Compound characterization is achieved by a
combination of IR and 31P NMR spectroscopy, voltam-
metry, and electrospray mass spectrometry (ESMS).
Exp er im en ta l Section
(i) Ma ter ia ls. All solvents used in synthetic procedures
were distilled and dried analytical reagent grade, and all pre-
parations utilized 100 mL volumes of refluxing solvent under
a nitrogen atmosphere. Manganese and rhenium carbonyls
and phosphine ligands (Strem) were used as supplied. The
pentacarbonyl halides were prepared by interaction of the
carbonyls with halogen.13 Fac-M(CO)3(κ2-P-P)X (M ) Mn, R;
P-P ) diphosphine or related ligand; X ) Cl, Br) were prepared
by reaction of equimolar quantities of M(CO)5X (1 g) and ligand
in refluxing chloroform for 2 h (manganese complexes) or
toluene for 3 h (rhenium complexes). The solvent was removed
under vacuum and the solid recrystallized from dichlo-
romethane/hexane.
(ii) P r ep a r a tion of Ma n ga n ese Com p lexes. The known7,8
complexes cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)X (X ) Cl, Br)
(structure I) were prepared by reacting 0.5-0.7 g of the tri-
carbonyl species fac-Mn(CO)3(dpm)X with a 1:1 molar ratio of
dpm in refluxing heptane for approximately 16 h. After the
by bubbling dry air through a solution of cis,mer-Re(CO)2(κ1-
dpm)(κ2-dpm)X. The known10 complexes cis,mer-[Re(CO)2(κ1-
dpmMe)(κ2-dpm)X]I (structure III) were prepared in 85 ( 5%
mixture was cooled, the liquid was decanted and the precipi-
tate from this reaction was dissolved in dichloromethane and
passed through a column of silica in order to remove any
charged species present. Hexane was added to the eluted solu-
tion, which was cooled to 0 °C and left standing overnight.
Orange crystals of cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)X precipi-
tated from the solution and were collected by gravity filtration.
(7) Carr, S. W.; Shaw, B. L.; Thornton-Pett, M. J . Chem. Soc., Dalton
Trans. 1985, 2131.
(8) Carriedo, G. A.; Riera, V.; Santamaria, J . J . Organomet. Chem.
1982, 234, 175.
(9) Osborne, A. G.; Stiddard, M. H. B. J . Chem. Soc. 1965, 700.
(10) Carr, S. W.; Shaw, B. L.; Thornton-Pett, M. J . Chem. Soc.,
Dalton Trans. 1987, 1763.
(11) Carriedo, G. A.; Rodriguez, M. L.; Garcia-Granda, S.; Aguirre,
A. Inorg. Chim. Acta, 1990, 178, 101.
(12) Freni, M.; Valenti, V.; Guisto, D. J . Inorg. Nucl. Chem. 1965,
27, 2635.
(13) Colton, R.; McCormick, M. J . Aust. J . Chem. 1976, 29, 1657.
yield by the addition of (1.3:1 molar ratio) excess MeI to a
dichloromethane solution of 0.5-0.7 g of cis,mer-Re(CO)2(κ1-
dpm)(κ2-dpm)X. The solution was allowed to stand overnight
and the solvent then removed under vacuum. To investigate
the electrochemical properties of the methylated cationic

3166 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
species, it was necessary to replace iodide by PF6-, since iodide
is electrochemically active in the potential range of interest.
This was undertaken by pretreatment of an anion exchange
column with KPF6 and passing an acetone solution of the
methylated complex through the column. Reaction of 0.5-0.7
g of Re(CO)5X and dpe (1:2 molar ratio) in refluxing mesitylene
for 5 h yielded 80 ( 5% of the binuclear complexes cis,mer-
{Re(CO)2(dpe)X}2(µ-dpe), whose characterization is given in
the Results and Discussion. Upon cooling of the solution, a
white solid precipitated which was then collected by filtration
and washed with hexane. Reaction between 0.5 g of Re(CO)5X
and dpe (1:2 molar ratio) under the milder conditions of
refluxing xylene (4 h) yielded 90 ( 5% cis,mer-Re(CO)2(κ1-dpe)-
(κ2-dpe)X after the precipitate formed on cooling was dissolved
in dichloromethane and the solution chromatographed using
a silica column and dichloromethane as the eluent. cis,mer-
Re(CO)2(κ1-ape)(κ2-ape)X (ape ) Ph2AsCH2CH2PPh2) was pre-
pared similarly. Additon of a 1:3:1 molar excess of MeI to a
dichloromethane solution yielded cis,mer-[Re(CO)2(κ1-apeMe)-
(κ2-ape)X]I after the solution was allowed to stand overnight
and the solvent removed under vacuum. Despite extensive
efforts, no crystals suitable for X-ray structural analysis could
be obtained for any of the new compounds.
(iv) Electr och em ica l Meth od s. Dichloromethane (HPLC
grade) was pretreated with alumina to remove any acid
impurities. Dichloromethane (electrolyte) solutions gave sat-
isfactory blank voltammograms. Conventional voltammetric
measurements typically were obtained with 1.0 mM solutions
of compound in dichloromethane (0.1 M Bu4NPF6), using a
Cypress Systems (Lawrence, KS) Model CYSY-1 computer-
controlled electrochemical system or a BAS 100A electrochemi-
cal analyzer (Bioanalytical Systems, West Lafayette, IN). The
working electrode was a glassy-carbon disk (0.5 mm radius),
the auxiliary electrode was a platinum wire, and the reference
electrode was Ag/AgCl (saturated LiCl in dichloromethane (0.1
M Bu4NPF6)) separated from the test solution by a salt bridge.
The reversible voltammetry of an approximately 1.0 mM
ferrocene (Fc) solution in the same solvent was used as a
reference redox couple, and all potentials are quoted relative
to Fc+/Fc. Near steady-state voltammograms were recorded
using a 12.5 µm radius platinum-microdisk electrode. Solutions
were purged with solvent-saturated nitrogen before voltam-
metric measurements and then maintained under an atmo-
sphere of nitrogen during the measurements.
Bulk electrolysis experiments were undertaken in dichlo-
romethane with 0.4 M Bu4NPF6 or 0.4 M Bu4NclO4 as the
electrolyte with a BAS 100 A 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 electrode as used in the
voltammetric studies. Coulometric analysis of the bulk elec-
trolysis data was used to either determine the purity of the
compound or the n value of the electrode process if the purity
had been established by independent methods.
(v) Sp ectr oscop ic Meth od s. 1H and proton-decoupled 31P
and 13C NMR spectra were recorded on a Bruker AM 300
spectrometer (31P at 121.496 MHz in dichloromethane and 13C
at 75.469 MHz in CDCl3 solution). The high-frequency-positive
convention is used for chemical shifts with external 85% H3-
PO4 (31P) and internal TMS (1H and 13C) references. Infrared
spectra were recorded on Perkin-Elmer FT-IR 1720X and
Perkin-Elmer 1430 IR spectrometers.
delivered the solution to the vaporization nozzle of the elec-
trospray source at a flow rate of 5 µL min-1. Nitrogen was used
as the drying gas and for nebulization with flow rates of
approximately 3 L min-1 and 100 mL min-1, respectively. The
voltage on the first skimmer (B1) was usually 40 V, but higher
voltages were used to induce collisionally activated fragmenta-
tions as described in the text. Peaks are identified by the most
abundant mass in the isotopic mass distribution.
(vii) An a lysis of Solid s. Elemental analysis and proton
NMR spectra of the compounds, with
a few exceptions,
demonstrated that solvent obtained from the refluxing solvent
and/or solvent used for recrystallization or even MeI (when
relevant) was present. Attempts to fully remove solvent by
heating altered the solvent component but did not completely
remove solvent, according to elemental analysis, and/or caused
decomposition, except in the cases of Mn(CO)2(κ1-dpm)(κ2-dpe)-
Br, Mn(CO)2(κ1-dpe)(κ2-dpe)Br, and Re(CO)2(κ1-dpe)(κ2-dpe)Br.
In studies on the related Re(CO)(dpe)2Br compound, where an
X-ray structure was obtained, molecular modeling of the
crystal structure14 revealed the presence of hydrophobic chan-
nels that are believed to enable ready uptake of organic
solvents and formation of material with widely variable solvent
composition. A similar situation is believed to prevail with
many of the compounds prepared in this study. Elemental
microanalyses for compounds prepared and recrystallized from
different solvents were performed by Chemsearch, University
of Otago, Otago, New Zealand, and provided the following
representative data where 0, 0.5, 0.75, 1.0, and 1.5 mol of
dichloromethane are present from recrystallization from this
solvent. Anal. Found (calcd) for [Mn(CO)2(dpe)2]Br‚0.5CH2Cl2:
C, 63.7 (63.5); H, 4.9 (4.8). Found (calcd) for Mn(CO)2(κ1-dpm)-
(κ2-dpb)Br‚0.75CH2Cl2: C, 64.8 (64.5); H, 3.7 (3.5). Found
(calcd) for Mn(CO)2(κ1-dpm)(κ2-dpe)Br: C, 65.0 (65.4); H, 5.0
(4.8). Found (calcd) for Mn(CO)2(κ1-dpe)(κ2-dpe)Br: C, 62.5
(62.5); H, 4.6 (4.7). Found (calcd) for Re(CO)2(κ1-dpe)(κ2-dpe)-
Br: C, 54.1 (54.2); H, 3.9 (4.0). Found (calcd) for Re(CO)2(κ1-
dpe)(κ2-dpe)Br‚1.5CH2Cl2: C, 53.0 (53.5); H, 4.1 (4.4). Found
(calcd) for Re(CO)2(κ1-dpm)(κ2-dpm)Br‚CH2Cl2: C, 54.4 (54.2);
H, 4.2 (3.9). Found (calcd) for [Re(CO)2(κ1-dpmMe)(κ2-dpm)-
Br]I‚0.5CH2Cl2: C, 50.4 (50.7); H, 3.7 (3.8).
In view of the difficulty of obtaining solvent-free samples
for microanalysis or crystals suitable for X-ray structural
characterization, mass spectral data for new compounds
having the expected m/z values have been obtained (Tables 1
and 2). Importantly, in all cases, the agreement between
experimental and theoretical isotropic mass patterns was
excellent. 13C and 1H NMR data obtained for deuterated
acetonitrile solutions revealed the presence of dichloromethane,
chloroform, MeI, or refluxing solvent as appropriate in most
of the samples. However, 31P NMR data for dichloromethane
solutions of the compounds prepared by chemical means or
by bulk electrolysis (see later; Figures S1-S7, Supporting
Information) showed no evidence of phosphine ligand or any
other diamagnetic phosphine-containing impurities. Data
obtained by NMR and IR (Tables 3 and 4) are fully consistent
with the structures reported (see Results and Discussion).
Voltammetric analysis was used to confirm the presence or
absence (<0.1%) of free halide or phosphine. In the case of the
binuclear complexes, where variable solvent composition of
between 0.15 and 0.65 mol of solvent/mol of complex was
obtained on the basis of microanalysis data, quantitative
coulometric analysis confirmed that the purity used for
electrochemical and spectroscopic examination was 97 ( 1.5%,
assuming that the oxidation process employed for this purpose
was a 1.0 or 2.0 electron oxidation process (see below). Further
details of the characterization of all compounds are contained
in section A of the Results and Discussion.
(vi) Electr osp r a y Ma ss Sp ectr om etr y, Positive ion elec-
trospray mass spectra were obtained with a VG Bio-Q triple
quadrupole mass spectrometer using a water/methanol/acetic
acid (50:50:1) mobile phase. Solutions of the compounds (2.0
mM in dichloromethane) were mixed, if necessary, with
oxidant or sodium acetate, as described in the text. The mixed
solution then was diluted 1:10 with methanol and immediately
injected directly into the spectrometer via a Rheodyne injector
fitted with a 10 µL loop. A Pheonix 20 micro LC syringe pump
(14) Snook, G. A.; Bond, A. M.; Fletcher, S. J . Electroanal. Chem.
2003, 554-555, 157.

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3167
Ta ble 1. Electr osp r a y Ma ss Sp ectr om etr ic Da ta for Ma n ga n ese Com p lexes
compd
additive
NaOAc
ion
m/z
fac-Mn(CO)3(dpe)Br
[Na + Mn(CO)3(dpe)Br]+
641
537
983
895
879
961
983
895
879
879
975
1023
957
941
907
1488
1576
788
[Mn(CO)3(dpe)]+
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Br
NaOAc
[Na + Mn(CO)2(dpm)2Br]+
[Mn(CO)2(dpmO)(dpm)]+
[Mn(CO)2(dpm)2]+
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Br
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Cl
[H + Mn(CO)2(dpm)2Br]+
[H + Mn(CO)2(dpm)2Cl]+
[Mn(CO)2(dpmO)(dpm)]+
[Mn(CO)2(dpm)2]+
a
trans-[Mn(CO)2(dpm)2]PF6
[Mn(CO)2(dpm)2]+
cis-Mn(CO)2(κ1-dpm)(κ2-dpe)Br
cis-Mn(CO)2(κ1-dpm)(κ2-dpbz)Br
[H + Mn(CO)2(dpm)(dpe)Br]+
[H + Mn(CO)2(dpm)(dpbz)Br]+
[Mn(CO)2(dpmO)(dpbz)]+
[Mn(CO)2(dpm)(dpbz)]+
[Mn(CO)2(dpe)2]+
trans-[Mn(CO)2(dpe)2]Br
cis,fac-{Mn(CO)2(dpe)Cl}2(µ-dpe)
cis,fac-{Mn(CO)2(dpe)Br}2(µ-dpe)
NOBF4
NOBF4
[Mn2(CO)2(dpe)3Cl2]+
[Mn2(CO)2(dpe)3Br2]+
[Mn2(CO)2(dpe)3Br2]2+
a
From exhaustive electrolysis of cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)X.
Ta ble 2. Electr osp r a y Ma ss Sp ectr om etr ic Da ta for Rh en iu m Com p lexes
compd
additive
NaOAc
ion
m/z
c,m-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Cl]PF6
[Re(CO)2(dpmMe)(dpm)Cl]+
[dpmMe]+
1061
399
199
1113
1105
1043
1177
457
[Ph2PdCH2]+
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br
c,m-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]PF6
cis-[Re(CO)2(dpmO)2]ClO4
[Na + Re(CO)2(dpm)2Br]+
[Re(CO)2(dpmMe)(dpm)Br]+
[Re(CO)2(dpmO)2]+
c,m-[Re(CO)2(κ1-apeMe)(κ2-ape)Cl]PF6
[Re(CO)2(apeMe)(ape)Cl]+
[apeMe]+
[Ph2PdCH2]+
199
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Br
[H + Re(CO)2(apeO)(ape)Br]+
[H + Re(CO)2(ape)2Br]+
[Re(CO)2(apeMe)(ape)Br]+
[Na + Re(CO)2(dpm)(dpe)Br]+
[Na + {Re2(CO)4(dpe)3Cl2}]+
[Re2(CO)3(NO)(dpe)3Cl2]+
[Re2(CO)4(dpe)3Cl2]+
1223
1207
1221
1127
1773
1752
1750
1863
1840
1838
1882
1801
c,m-[Re(CO)2(κ1-apeMe)(κ2-ape)Br]PF6
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpe)Br
cis,mer-{Re(CO)2(dpe)Cl}2(µ-dpe)
cis,mer-{Re(CO)2(dpe)Cl}2(µ-dpe)
NaOAc
NaOAc
NOBF4
cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe)
cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe)
NaOAc
NOBF4
[Na + {Re2(CO)4(dpe)3Br2}]+
[Re2(CO)3(NO)(dpe)3Br2]+
[Re2(CO)4(dpe)3Br2]+
[Re2(CO)4(dpp)3Br2]+
[Re2(CO)4(dpp)3Br]+
cis,mer-{Re(CO)2(dpp)Br}2(µ-dpp)
NOBF4
Ta ble 3. IR (Ca r bon yl Region ) a n d 31P NMR Da ta in Dich lor om eth a n e for Ma n ga n ese Com p lexes Obta in ed
fr om Syn th esized Com p ou n d s or fr om Bu lk Electr olysis Exp er im en ts
compd
ν(CO) (cm-1
)
δ(31P) (ppm)
temp (NMR) (°C)
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Cl
trans-Mn(CO)2(κ1-dpm)(κ2-dpm)Cl
trans-[Mn(CO)2(κ1-dpm)(κ2-dpm)Cl]+
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Br
trans-Mn(CO)2(κ1-dpm)(κ2-dpm)Br
trans-[Mn(CO)2(κ1-dpm)(κ2-dpm)Br]+
trans-[Mn(CO)2(dpm)2]+
1940, 1867
1892
1976
1937, 1867
1891
1972
1916
1939, 1867
1889
1971
1940, 1867
1893
52.9, 36.6, 1.7, -28.6
56.2, 51.5, 19.7, -31.6
22
-40
50.3, 35.0, 0.3, -28.1
53.2, 49.3, 19.0, -31.8
22
-40
35.3
22
22
-40
cis-Mn(CO)2(κ1-dpm)(κ2-dpe)Br
trans-Mn(CO)2(κ1-dpm)(κ2-dpe)Br
trans-[Mn(CO)2(κ1-dpm)(κ2-dpe)Br]+
cis-Mn(CO)2(κ1-dpm)(κ2-dpbz)Br
trans-Mn(CO)2(κ1-dpm)(κ2-dpbz)Br
trans-[Mn(CO)2(κ1-dpm)(κ2-dpbz)Br]+
cis-Mn(CO)2(κ1-dpe)(κ2-dpe)Cl
cis-Mn(CO)2(κ1-dpe)(κ2-dpe)Br
trans-[Mn(CO)2(dpe)2]Cl
81.4, 53.0, 47.5, -28.2
98.9, 75.1, 54.4, -29.9
85.8, 61.1, 47.2, -28.3
102.8, 86.5, 53.2, -30.0
22
-40
1972
1937, 1863
1937, 1866
1897
81.7, 54.2, 52.8, -14.2
22
22
22
79.6, 52.9, 51.2, -13.8
71.6
trans-[Mn(CO)2(dpe)2]Br
trans-Mn(CO)(dpe)2Cl
1897
1826
71.9
73.2
22
22
trans-Mn(CO)2(dpe)2Br
1824
1937, 1863
1889
1969, 1937, 1863
1969
1937, 1865
1897
77.6
80.8 (1), 52.9 (2)
100.5, 76.7, 59.2
22
22
-60
cis,fac-{Mn(CO)2(dpe)Cl}2(µ-dpe)
trans-{Mn(CO)2(dpe)Cl}2(µ-dpe)
c,f(t)-[{Mn(CO)2(dpe)Cl}2(µ-dpe)]+
trans-[{Mn(CO)2(dpe)Cl}2(µ-dpe)]2+
cis,fac-{Mn(CO)2(dpe)Br}2(µ-dpe)
trans-{Mn(CO)2(dpe)Br}2(µ-dpe)
c,f(t)-[{Mn(CO)2(dpe)Br}2(µ-dpe)]+
trans-[{Mn(CO)2(dpe)Br}2(µ-dpe)]2+
79.1 (1), 50.9 (2)
99.0, 75.2, 57.8
22
-60
1965, 1937, 1865
1965
Resu lts a n d Discu ssion
generated by bulk electrolysis. Figures S1-S4 (Sup-
porting Information) contain 31P NMR spectra of Mn
complexes. The new mixed-diphosphine complexes Mn-
(CO)2(κ1-dpm)(κ2-P-P)Br (P-P ) dpe, dpbz; dpbz )
o-(Ph2P)2C6H4) exhibit two carbonyl IR stretches, indi-
A. Ch a r a cter iza tion of Com p ou n d s. (i) Ma n ga -
n ese Com p lexes. 31P NMR and IR spectroscopic data
for all manganese complexes isolated are summarized
in Table 3, as are data (where available) for compounds

3168 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
Ta ble 4. IR Da ta (Ca r bon yl Region ) a n d 31P NMR Da ta (22 °C) in Dich lor om eth a n e for Rh en iu m
Com p lexes Obta in ed fr om Syn th esized Com p ou n d s or fr om Bu lk Electr olysis Exp er im en ts
compd
ν(CO) (cm-1
)
δ(31P) (ppm)a
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Cl
cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Cl
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Cl]PF6
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br
cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Br
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]PF6
cis-[Re(CO)2(dpmO)2]+
cis,mer-{Re(CO)2(dpe)Cl}2(µ-dpe)
cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe)
cis,mer-{Re(CO)2(dpp)Br}2(µ-dpp)
cis,mer-Re(CO)2(κ1-dpe)(κ2-dpe)Cl
cis,mer-Re(CO)2(κ1-dpe)(κ2-dpe)Br
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Cl
cis,mer-Re(CO)2(κ1-apeO)(κ2-ape)Cl
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Cl]PF6
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Br
cis,mer-Re(CO)2(κ1-apeO)(κ2-ape)Br
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Br]PF6
1938, 1858
1942, 1864
1942, 1869
1941, 1862
1943, 1867
1943, 1870
1931, 1845
1943, 1857
1943, 1860
1951, 1857
1942, 1856
1943, 1862
1936, 1853
1938, 1856
1940, 1860
1941, 1859
1942, 1861
1942, 1866
9.5*, -21.1*, -27.9, -44.6
23.4, 6.8*, -21.8*,-43.0
18.7, 10.5*, -21.7*, -48.2
5.3*, -24.2*, -28.1, -47.7
23.9, 2.9*, -24.2*, -45.6
18.0, 5.1*, -25.6, -50.1
66.5, 31.5
34.4*, 18.5, 6.5*
31.4*, 15.3, 2.3*
-2.3*, -11.7*, -22.0
36.9*, 21.3, 7.4*, -12.3
33.3*, 17.6, 2.9*, -12.2
38.4*, 8.4*
37.4*, 10.5*
38.2*, 11.2*
35.8*, 4.0*
34.7*, 5.7*
35.1*, 6.1*
a
The asterisks indicate that these pairs of phosphorus atoms in each compound are strongly coupled (J ) 170-185 Hz), indicating a
mutually trans configuration.
cating a cis carbonyl geometry. The 31P NMR spectra
for all of these types of compounds show four resonances
of equal integrated intensities (Figure S1), and in each
case a signal is observed close to the resonance position
for free dpm, thus indicating that dpm is the κ1 ligand.
However, since coupling details could not be observed,
a distinction cannot be made between cis,fac- and cis,-
mer-Mn(CO)2(κ1-dpm)(κ2-P-P)Br, and therefore the com-
pounds are referred to as merely cis in Table 3. The
trans isomeric forms produced by bulk oxidative and
reductive electrolysis sequences (see later) of the cis
isomer exhibit (Table 3) a single carbonyl stretch and
four 31P NMR resonances of equal intensity (Figure S2),
as expected for the trans configuration.
(κ1-dpe)(κ2-dpe)Br. Again, fine structure could not be
discerned in the NMR spectrum of Mn(CO)2(κ1-dpe)(κ2-
dpe); therefore, these compounds also are referred to as
merely cis-Mn(CO)2(κ1-dpe)(κ2-dpe)X in Table 3. Details
of the interpretation of spectra of compounds detected
during the course of bulk electrolysis experiments (Table
3, Figure S4) are contained elsewhere in the text. The
absence of 31P coupling details in a number of cases
could arise from line broadening associated with the
presence of paramagnetic impurities. However, efforts
to remove paramagnetic impurities by ion chromatog-
raphy and bulk reductive electrolysis did not lead to
changed spectra, and coupling also could not be detected
at low temperatures.
The IR spectrum of the major product from the inter-
action of Mn(CO)5Br and dpe (1:2) shows two absorp-
tions of equal intensities, indicating a cis-dicarbonyl
configuration. The 31P NMR spectrum shows two strong
resonances at δ 79.1 and 50.9 ppm of integrated intensi-
ties 1:2 (Figure S3). As both of these signals are at much
higher frequency than the resonance of free dpe, the
spectrum clearly indicates that all of the phosphorus
atoms are coordinated to the metal. However, the 1:2
intensity ratio is not consistent with any monomeric
formulation and leads to the conclusion that the com-
pound is a dpe-bridged binuclear species. ESMS (Table
1) measurements, described in detail later, confirm the
binuclear nature of this compound with the empirical
formula Mn2(CO)4(dpe)3Br2. The IR and NMR data fix
the geometry as cis,fac, as shown in structure IV.
(ii) Rh en iu m Com p lexes. Specroscopic data for the
synthesized (31P NMR and IR) and electrochemically
generated rhenium complexes (IR) are summarized in
Table 4. 31P spectra of rhenium complexes are contained
in Figures S5-S7 (Supporting Information). The 31P
NMR spectra of the complexes cis,mer-Re(CO)2(κ1-dpm)-
(κ2-dpm)X consist of four resonances, two of which are
coupled strongly, which indicates nonequivalent trans
phosphorus atoms on rhenium, in agreement with the
previously published data.10 The 13C NMR spectrum of
a deuterated chloroform solution of cis,mer-Re(CO)2(κ1-
dpm)(κ2-dpm)Br shows two resonances in the carbonyl
region, one of which appears as a doublet, which
therefore is assigned to the carbonyl group trans to
phosphorus. The IR spectrum of the compound shows
the expected two carbonyl stretches, and all data are
consistent only with the previously proposed structure.10
The IR spectrum of the impurity mentioned in the
Experimental Section shows two stretches consistent
with a cis-dicarbonyl geometry, and its 31P NMR (Figure
S5) spectrum shows four resonances of equal intensities.
These are very similar to those previously reported11
for cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Br. Similar obser-
vations were made for the cis,mer-Re(CO)2(κ1-dpm)(κ2-
dpm)Cl system. The complexes cis,mer-[Re(CO)2(κ1-
dpmMe)(κ2-dpm)X]I were prepared by the addition of
MeI to cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)X, which me-
thylates the pendant phosphorus atom to give the
phosphonium salt. To investigate the electrochemical
properties of the methylated cationic species, it was
The minor product from the interaction of Mn(CO)5Br
and dpe (1:2) also shows two carbonyl stretches, again
indicating a cis-dicarbonyl arrangement. The 31P NMR
spectrum consists of four signals of equal integrated
intensities, one of which is close to the resonance of free
dpe, which is consistent with the formulation Mn(CO)2-
-
necessary to replace iodide by PF6 by ion exchange,

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3169
since iodide is electrochemically active in the potential
range of interest. The IR spectrum of cis,mer-[Re(CO)2-
(κ1-dpmMe)(κ2-dpm)Br]PF6 shows two absorption bands
very similar to those of cis,mer-Re(CO)2(κ1-dpm)(κ2-
dpm)Br. The 31P NMR spectrum of the cation is also
very similar to that of cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)-
Br, except that the signal due to the pendant phospho-
rus is no longer present and is replaced by a signal in
the region associated with phosphonium salts. Thus, the
spectroscopic data indicate that the pendant phosphorus
is methylated, and the remainder of the molecule is
unchanged.
The IR spectrum of the white solid obtained from the
reaction of Re(CO)5Cl and dpe (1:2) in refluxing mesi-
tylene showed two bands in the carbonyl region consis-
tent with a cis-dicarbonyl geometry. The 31P NMR
spectrum (Figure S6) consists of three signals of equal
intensities, all at considerably higher frequency than
the resonance of free dpe, indicating that all of the
phosphorus atoms are coordinated to the metal. Two of
the signals are doublets (J P-P ) 180 Hz), which is
indicative of trans nonequivalent phosphorus atoms on
rhenium. This spectroscopic evidence can be rationalized
in terms of a dpe-bridged binuclear complex, but with
cis,mer geometry rather than the cis,fac stereochemistry
assigned to the corresponding manganese complex. To
confirm the stoichiometry, Re(CO)5X and 1.5 mol of dpe
were reacted under the same conditions and the product
shows the same 31P NMR spectrum and no resonance
for fac-Re(CO)3(dpe)X (the known first product of this
reaction), indicating that enough dpe is present to react
with all the tricarbonyl complex. ESMS data (Table 2),
described in more detail later, confirm the empirical
formula Re2(CO)4(dpe)3Cl2. Similar observations were
made for the corresponding bromo complex. The only
structure consistent with all the spectroscopic and mass
spectrometric data is cis,mer-{Re(CO)2(dpe)X}2(µ-dpe)
(structure V), and this is also fully consistent with
geometry with the κ1-ape ligand being phosphorus-
bonded. Addition of MeI to this compound gives a
derivative for which the IR and 31P NMR spectra are
almost unaltered and are consistent with the formula-
tion cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)X]I. Confirma-
tion of stoichiometry is provided by ESMS. The inter-
pretation of IR and 31P NMR spectra (Table 4, Figure
S7) of Re compounds generated during the course of
bulk electrolysis experiments are provided elsewhere in
the text.
B. Electr och em ica l Stu d ies on cis,m er -Mn (CO)2-
(K1-d p m )(K2-P -P )X (X ) Cl, Br ) Com p lexes. (i) Cy-
clic Volta m m etr y. The oxidation of cis,mer-Mn(CO)2-
(κ1-dpm)(κ2-dpm)Xhas been reported in detailelsewhere.15
In view of the similarity of all of the cis,mer-Mn(CO)2-
(κ1-dpm)(κ2-P-P)X complexes, only a brief summary of
the voltammetry of some of these dicarbonyls needs to
be given. Cyclic voltammetric data obtained at a scan
rate of 200 mV s-1 in dichloromethane (0.1 M Bu4NPF6)
at specified temperatures are summarized in Table 5.
On the initial cycle of the potential, an oxidative
response (process 1) is observed at the potentials
indicated in Table 5. An associated reduction response
(process 1′) also is seen for most complexes (Table 5).
However, the smaller value of the reduction current
indicates that process 1 is not reversible in the chemical
sense at a scan rate of 200 mV s-1. A second reduction
peak (process 2′) also is observed on the first cycle of
the potential, while second and subsequent cycles reveal
a corresponding oxidation response (process 2). For
redox couple 2, the reduction and oxidation peak height
are equal, as expected for a chemically reversible
process. Scanning to more positive potentials reveals
additional oxidation responses, which are irreversible
under all conditions studied and will not be discussed
further.
The limiting oxidation current for process 1 at a
platinum-microdisk electrode (radius 12.5 µm) is similar
to that for an equimolar solution of fac-Mn(CO)3(κ2-dpe)-
Br, which is known to undergo a one-electron oxida-
tion.16 The charge-transfer process associated with redox
couple 1 is therefore assigned to the reaction
cis,mer-Mn(CO)2(κ1-dpm)(κ2-P-P)X h
cis,mer-[Mn(CO)2(κ1-dpm)(κ2-P-P)X]+ + e- (1)
in which the Mn(I) compound is oxidized to the isos-
tructural Mn(II) compound. However, cis,mer-[Mn(CO)2-
(κ1-dpm)(κ2-P-P)X]+ rapidly converts to the species
reduced in process 2′, which on the basis of data
contained in ref 15 is the electronically17 and sterically
preferred trans-[Mn(CO)2(κ1-dpm)(κ2-P-P)X]+. Redox
couple 2 may therefore be assumed to be due to the
reaction
electrochemical studies described later. A similar com-
plex was synthesized with dpp (dpp ) Ph2P(CH2)3PPh2).
Re(CO)2(κ1-dpe)(κ2-dpe)X is synthesized by reacting
Re(CO)5X in a 1:2 molar ratio with dpe under milder
conditions in refluxing xylene. The 31P NMR spectrum
consists of four resonances of equal intensities, three of
them at positions almost identical with those of cis,-
mer-{Re(CO)2(dpe)X}2(µ-dpe); the fourth is close to the
position of the resonance of free dpe, indicating the
compounds to be cis,mer-Re(CO)2(κ1-dpe)(κ2-dpe)X. The
compounds Re(CO)2(κ1-ape)(κ2-ape)X (ape ) Ph2AsCH2-
CH2PPh2) were prepared by reacting Re(CO)5X with 2
mol of ape in refluxing mesitylene. IR spectroscopy
indicates a cis-dicarbonyl arrangement, and the 31P
NMR spectrum consists of two doublets with a large
coupling constant, suggesting nonequivalent trans phos-
phorus atoms. The data are consistent with a cis,mer
trans-Mn(CO)2(κ1-dpm)(κ2-dpm)Br h
trans-[Mn(CO)2(κ1-dpm)(κ2-dpm)Br]+ + e- (2)
Overall, therefore, the voltammetric behavior is sum-
(15) Eklund, J . C.; Bond, A. M.; Colton, R.; Humphrey, D. G.; Mahon,
P. J .; Walter, J . N. Inorg. Chem. 1999, 38, 2005.
(16) Bond, A. M.; Colton, R.; McDonald, M. E. Inorg. Chem. 1978,
17, 2842.
(17) Mingos, D. M. P. Organomet. Chem. 1979, 179, C29.

3170 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
Ta ble 5. Volta m m etr ic Da ta (Sca n Ra te 200 m V s-1) a t a Gla ssy-Ca r bon -Disk Electr od e (Ra d iu s 0.5 m m ) for
Ma n ga n ese Com p lexes in Dich lor om eth a n e (0.1 M Bu 4NP F 6)
ox
red
compd
process
Ep (V)
Ep (V)
E1/2 (V)
∆Ep (mV)
temp (°C)
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Cl
1
1
2
3
0.196
0.204
-0.344
0.414
0.106
0.104
-0.422
0.320
0.151
0.154
-0.383
0.367
90
100
78
94
90
20
-30
20
20
20
(actually [Mn(CO)2(dpm)2]2+/+
cis,mer-Mn(CO)2(κ1-dpm)(κ2-dpm)Br
)
1
0.204
0.114
0.159
1
2
3
0.214
-0.333
0.428
0.112
-0.413
0.304
0.163
-0.373
0.366
102
80
124
-30
20
(actually [Mn(CO)2(dpm)2]2+/+
)
20
cis-Mn(CO)2(κ1-dpm)(κ2-dpe)Br
cis-Mn(CO)2(κ1-dpm)(κ2-dpbz)Br
cis,fac-{Mn(CO)2(dpe)Cl}2(µ-dpe)
1
1
2
1
1
2
(1)
(2)
1
2
3
4
(1)
(2)
1
2
3
0.208
0.228
-0.285
0.210
0.232
-0.233
0.182
0.310
0.235
0.365
-0.321
-0.200
0.191
0.332
0.247
20
-40
20
20
-40
20
20
20
-50
-50
20
20
20
20
0.120
-0.361
0.174
-0.323
108
76
0.118
-0.309
0.175
-0.271
114
76
0.103
0.221
-0.417
-0.290
0.169
0.293
-0.369
-0.245
132
144
96
90
cis,fac-{Mn(CO)2(dpe)Br}2(µ-dpe)
0.113
0.250
-0.368
-0.262
0.180
0.320
-0.323
-0.219
134
140
90
-50
-50
20
0.390
-0.278
-0.176
4
86
20
marized by a limited version of a square scheme:18
(ii) Bu lk Electr olysis. Bulk electrolyses of dichlo-
romethane (0.1 M Bu4NPF6) solutions of cis-Mn(CO)2-
(κ1-dpm)(κ2-P-P)X were performed initially at -40 °C to
prevent chelation of the dpm in the 17e cation.15 In each
case, the amount of charge passed corresponds to a one-
electron oxidation process when the potential is held
ox
close to Ep of process 1. Steady-state voltammograms
obtained at a Pt-microdisk electrode during the course
of these bulk oxidative experiments show that redox
couple 1 is absent after electrolysis, while reversible
redox couple 2 is observed as a reduction process. The
IR spectra of these solutions (Table 3) each display a
single strong carbonyl stretch corresponding to forma-
tion of trans-[Mn(CO)2(κ1-dpm)(κ2-P-P)X]+. Bulk reduc-
F igu r e 1. Oxidative cyclic voltammograms at a glassy-
carbon electrode (radius 0.5 mm) of 1.0 mM solutions of
(a) cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br and (b) cis,mer-Re-
(CO)2(κ1-dpmMe)(κ2-dpm)Br in dichloromethane (0.1 M Bu4-
NPF6) at 22 °C using a scan rate of 200 mV s-1
.
tion (-40 °C) of each solution at a potential slightly
red
reaction scheme given in eq 3 is verified by bulk
electrolysis and spectroscopic characterization of prod-
ucts at low temperatures.
more negative than Ep
for process 2 generated the
oxidative component of process 2 (shown by steady-state
voltammograms) via a one-electron process. The IR
spectra of these solutions, acquired quickly at room
temperature (Table 3), show a single carbonyl stretch
corresponding to trans-Mn(CO)2(κ1-dpm)(κ2-P-P)X. The
31P NMR spectrum (Figure S2) of each reduced solution
containing 0.4 M Bu4NCO4 as the electrolyte displays
four signals of equal intensities (Table 3), which is
further evidence that the reductive component of process
2 generates trans-Mn(CO)2(κ1-dpm)(κ2-P-P)X. At room
temperature, chelation of the pendant phosphorus in
trans-[Mn(CO)2(κ1-dpm)(κ2-P-P)X]+ occurs to give the
trans-[Mn(CO)2(κ2-dpm)(κ2-P-P)]+ complex.16 Thus, the
C. Electr och em ica l Stu d ies on cis,m er -Re(CO)2-
(K1-d p m )(K2-d p m )X a n d cis,m er -Re(CO)2(K1-d p m O)-
(K2-d p m )X. (i) Cyclic Volta m m etr y. A cyclic voltam-
mogram of a 1.0 mM solution of cis,mer-Re(CO)2(κ1-
dpm)(κ2-dpm)Br in dichloromethane (0.1 M Bu4NPF6)
at 22 °C shows an initial oxidation response occurring
at 0.472 V (process 1), which is unresolved from a second
response, process 2 (Figure 1a). At more positive po-
tentials a third response (process 3) is observed, and
the reductive direction reverse scan reveals some reduc-
tion current (process 3′). When the scan direction is
switched after either process 1 or 2, no reductionre-
sponse is observed. Since the oxidation of the rhenium
center occurs at a considerably more positive potential
(18) (a) Bond, A. M.; Colton, R. Coord. Chem. Rev. 1997, 166, 161.
(b) Bond, A. M.; Colton, R.; J ackowski, J . J . Inorg. Chem. 1975, 14,
274.

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3171
than the corresponding process for manganese, it was
considered possible that the first irreversible oxidation
response (process 1) might be associated with oxidation
of the pendant phosphorus atom to an oxide. This was
confirmed by examining the cyclic voltammetry of an
authentic sample of cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)-
Br, which gives processes 2 and 3 but not process 1.
Process 1 is therefore assigned to the overall reaction
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br +
H2O f cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Br +
2H+ + 2e- (4)
assuming that the reaction involves the initial formation
of a phosphonium cation radical and subsequent reac-
tion with traces of water in the system. This reaction
in eq 4 could not be prevented even when using
dichloromethane dried for extended periods of time over
molecular sieves or alumina. Process 2 may then be
assigned to the chemically irreversible oxidation of cis,-
mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Br. Similar observations
were made with cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Cl and
cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Cl. Further evidence
for the assignment of processes 1 and 2 and a discussion
of process 3 will be given below, after consideration of
data obtained by bulk oxidative electrolysis.
(ii) Bu lk Electr olysis. Due to the similar potentials
of processes 1 and 2, which prevented selective oxida-
tion, a solution of cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br in
dichloromethane (0.4 M Bu4NClO4) was partially oxi-
dized at a potential corresponding to the foot of process
1, and the solution was monitored by 31P NMR spec-
troscopy. The spectra acquired at various stages of the
electrolysis show the growth of signals due to cis,mer-
Re(CO)2(κ1-dpmO)(κ2-dpm)Br (Table 2). When the po-
tential is made more positive, these signals in the NMR
spectrum broaden and numerous other resonances are
observed. New voltammetric responses are observed
temporarily during this stage of the electrolysis, but
process 3, which is reversible, is the only permanent
response, as long as the applied electrolysis potential
is kept less positive than that of process 3. The 31P NMR
spectrum of the fully electrolyzed solution (Figure S7),
which shows only redox couple 3, has two resonances
of equal intensities. The IR spectrum shows two carbo-
nyl bands consistent with a cis-dicarbonyl geometry.
Exhaustive bulk electrolysis of a fresh solution of cis,-
mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br shows that the overall
reaction to produce redox couple 3 is an overall appar-
ently six-electron process.
F igu r e 2. Proposed mechanism for the formation of cis-
[Re(CO)2(dpmO)2]+.
in the exhaustively electrolyzed solutions, and redox
couple 3 is assigned to the reaction
[Re(CO)2(dpmO)2]+ h [Re(CO)2(dpmO)2]2+ + e- (5)
The mechanism proposed for the oxidation of cis,mer-
Re(CO)2(κ1-dpm)(κ2-dpm)Br is shown in Figure 2. Two
electrons are required to oxidize each of the two P(III)
atoms to P(V), one to oxidize Re(I) to Re(II) and one to
oxidize bromide to bromine. The final product of this
scheme is [Re(CO)2(dpmO)2]2+, but at the potential
applied to the electrode (less positive than process 3) it
will be reduced to [Re(CO)2(dpmO)2]+. The charge
passed in the reverse direction is not subtracted by the
instrument used; thus, an apparently six-electron reac-
tion is indicated. However, the overall reaction is
actually a five-electron oxidation process:
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br + 2H2O f
cis-[Re(CO)2(dpmO)2]+ + 1/2Br2 + 4H+ + 5e- (6)
In the case of bulk oxidative electrolysis of cis,mer-
Re(CO)2(κ1-dpm)(κ2-dpm)Cl, the final IR and 31P NMR
spectra are very similar to those derived from cis,mer-
Re(CO)2(κ1-dpm)(κ2-dpm)Br. However, the exhaustive
electrolysis experiment shows that five rather than six
electrons are apparently involved in the complete oxida-
tion in this case. Bulk electrolysis of cis,mer-Re(CO)2-
(κ1-dpmO)(κ2-dpm)Br also gave the same final product
associated with process 3 via an overall four-electron
process, while cis,mer-Re(CO)2(κ1-dpmO)(κ2-dpm)Cl is
exhaustively oxidized to also give the same product in
an overall three-electron process.
In the case of the chloro compound, the overall
reaction is very similar but the applied potential is not
sufficiently positive to oxidize the displaced chloride ion
to chlorine, thus leading to
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Cl + 2H2O f
cis-[Re(CO)2(dpmO)2]+ + Cl- + 4H+ + 4e- (7)
Another minor side reaction must occur in this case, as
the coulometric count approaches five rather than four
electrons. Obviously, as observed, the electron count is
expected to be less when the electrolysis commences
ESMS experiments and spectroscopic data (described
later) show that the ion [Re(CO)2(dpmO)2]+ is present

3172 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
with the partially oxidized species cis,mer-Re(CO)2(κ1-
dpmO)(κ2-dpm)X, rather than cis,mer-Re(CO)2(κ1-dpm)-
(κ2-dpm)X.
The two structures VIa and VIb for cis-[Re(CO)2-
(dpmO)2]+ are both consistent with the 31P NMR and
IR spectra, but 13C NMR spectroscopy should distin-
guish between them, since a carbonyl trans to oxygen
should give a singlet resonance, while one trans to
phosphorus should be a doublet. The 13C NMR spectrum
perature leads to enhanced reversibility in the first
process, a decrease in the current for process 2 relative
to that for process 1, until this more positive process is
barely detected at -70 °C. The charge-transfer step
associated with process 1 is assigned to the couple
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Cl h
cis,mer-[Re(CO)2(κ1-ape)(κ2-ape)Cl]+ + e- (9)
The temperature dependence of the voltammetric data
suggest that process 2 is associated with oxidation of a
species formed from a chemical step following the
charge-transfer process, the rate of which is slowed upon
cooling the solution. Similar observations were made for
the bromo complex, and data are summarized in Table
6.
(ii) Bu lk Electr olysis. Exhaustive bulk electrolysis
of a dichloromethane solution (0.4 M Bu4NClO4) of cis,-
mer-Re(CO)2(κ1-ape)(κ2-ape)Br was carried out at a
ox
potential slightly less than Ep of process 1. A cyclic
voltammogram of the resulting solution shows process
1 to be absent with process 2 present. However, a
steady-state voltammogram indicates that process 2 is
present as a reduction rather than an oxidation reaction.
Both the IR and 31P NMR spectra of this solution are
similar to those of the starting material (Table 4),
suggesting that the overall reaction may only involve
change at the pendant arsenic atom. ESMS (see later)
gives a clear indication that the cation cis,mer-[Re(CO)2-
(κ1-apeO)(κ2-ape)Br]+ is present in the solution; there-
fore, process 2 is assigned to the redox couple
(carbonyl region) of cis-[Re(CO)2(dpmO)2]+ is a singlet;
therefore, structure VIa is indicated.
D. Volta m m etr ic Stu d ies on cis,m er -[Re(CO)2(K1-
d p m Me)(K2-d p m )X]P F 6. If the mechanism for oxida-
tion of cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)X is correct,
then methylation of the pendant phosphorus in cis,mer-
Re(CO)2(κ1-dpm)(κ2-dpm)X to give cis,mer-[Re(CO)2(κ1-
dpmMe)(κ2-dpm)Br]PF6 should remove the phosphorus
oxidation reaction in the electrochemical studies. An
oxidative cyclic voltammogram of a 1.0 mM solution of
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]PF6 in dichlo-
romethane (0.1 M Bu4NPF6) shows an oxidative re-
sponse at 0.836 V (process 1), and on the reverse scan,
the corresponding reduction response (process 1′) is
observed (Figure 1b). Analysis of the steady-state vol-
tammogram at a platinum microelectrode indicates that
this is a one-electron process and is assigned to the
reaction
cis,mer-[Re(CO)2(κ1-apeO)(κ2-ape)Br]+ h
cis,mer-[Re(CO)2(κ1-apeO)(κ2-ape)Br]2+ + e- (10)
The product is in the reduced form because the applied
potential is less positive than the reversible potential
for process 2. The exact mechanism for the formation
of the κ1-apeO complex is not known, but it clearly
involves some reaction of the initially produced cis,mer-
[Re(CO)2(κ1-ape)(κ2-ape)Br]+, presumably with traces of
water in the system, to give an equation analogous to
that given for the oxidation of the cis,mer-[Re(CO)2(κ1-
dpm)(κ2-dpm)Br] reaction (see eq 4).
F . Electr och em ica l Stu d ies of cis,m er -[Re(CO)2-
(K1-a p eMe)(K2-a p e)X]+. An oxidative cyclic voltammo-
gram of a 1.0 mM solution of cis,mer-[Re(CO)2(κ1-
apeMe)(κ2-ape)Br]+ in dichloromethane (0.1 M Bu4PF6)
shows the presence of only a single reversible process
1. Analysis of a steady-state voltammogram indicates
that this is a reversible one-electron process, which is
therefore assigned to the redox couple
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]+ h
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]2+ + e- (8)
where Re(I) is oxidized to Re(II). The chloro complex
exhibits similar voltammetry, although it is interesting
that there is no evidence for isomerization to the trans
isomer on the voltammetric time scale.
E. Electr och em ica l Stu d ies on cis,m er -[Re(CO)2-
(K1-a p e)(K2-a p e)X. (i) Cyclic Volta m m etr y. An oxida-
tive cyclic voltammogram of a 1.0 mM solution of
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Cl in dichloromethane
(0.1 M Bu4NPF6) at 20 °C and with a scan rate of 200
mV s-1 shows a partially reversible process 1 at 0.578
V and a second smaller reversible process 2 at 0.726 V.
On second and successive scans, process 1 diminishes
while process 2 becomes dominant. Lowering the tem-
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Br]+ h
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Br]2+ + e- (11)
Since the pendant arsenic atom cannot be oxidized in
this compound, the absence of a second process is
consistent with the identification of process 2 made
above.
G. Electr och em ica l Stu d ies on cis,fa c-{Mn (CO)2-
(d p e)X}2(µ-d p e) (X ) Cl, Br ). (i) Cyclic Volta m m e-
tr y. The cyclic voltammetry of the binuclear species is

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3173
Ta ble 6. Volta m m etr ic Da ta (20 °C, Sca n Ra te 200 m V s-1) a t a Gla ssy-Ca r bon -Disk Electr od e (Ra d iu s 0.5
m m ) for Rh en iu m Com p lexes in Dich lor om eth a n e (0.1 M Bu 4NP F 6)
ox
red
initial compd
process
Ep (V)
Ep (V)
E1/2 (mV)
∆Ep
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Cl
1
2
3
1
1
2
3
1
1
2
3
1
2
3
1
2
3
1
2
1
1
2
1
0.525a
0.525a
0.875
0.837
0.472
0.549
0.898
0.836
0.492
0.634
1.028
0.496
0.640
1.020
0.500
0.608
0.956
0.578
0.726
0.724
0.608
0.740
0.720
0.783
0.749
0.829
0.793
92
88
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Cl]PF6
cis,mer-Re(CO)2(κ1-dpm)(κ2-dpm)Br
0.808
0.746
0.406
0.540
0.853
0.791
0.449
0.587
90
90
86
94
cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)Br]PF6
cis,mer-{Re(CO)2(dpe)Cl}2(µ-dpe)
cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe)
cis,mer-{Re(CO)2(dpp)Br}2(µ-dpp)
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Cl
0.408
0.548
0.452
0.594
88
92
0.410
0.512
0.455
0.560
90
96
0.488
0.638
0.630
0.514
0.648
0.630
0.533
0.682
0.677
0.561
0.694
0.675
90
88
94
94
92
90
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Cl]PF6
cis,mer-Re(CO)2(κ1-ape)(κ2-ape)Br
cis,mer-[Re(CO)2(κ1-apeMe)(κ2-ape)Br]PF6
a
The potentials of these processes overlap.
extremely complicated. There are always both oxidation
and reduction components, and the nature of the vol-
tammetry is critically dependent upon temperature.
Since detailed assignments cannot be made until the
spectroscopic characteristics of the system have been
determined after bulk electrolysis experiments (see
later), these initial voltammetric section responses will
be indicated as, for example, process (1) (indicating
complexity within process 1) and detailed mechanistic
discussion will be given later. Voltammetric data are
summarized in Table 5.
Oxidative cyclic voltammograms at a glassy-carbon-
disk electrode (radius 0.5 mm) for a 1.0 mM solution of
cis,fac-{Mn(CO)2(dpe)Br}2(µ-dpe) in dichloromethane
(0.1 M Bu4NPF6) are shown in Figure 3. The first scan
of the voltammogram at 20 °C (Figure 3a) displays
oxidation responses at 0.191 V (process (1)) and 0.332
V (process (2)), which are irreversible when the scan is
reversed after process (2). However, the reduction waves
are observed at more negative potentials (processes (3′)
and (4′)) and a second scan reveals the oxidative
components of the reversible redox couples (3) and (4).
Figure 3b shows an oxidative cyclic voltammogram in
which the potential is reversed immediately after
process (1). The reverse scan shows only the reduction
process (3′), and the related oxidation response, process
(3), is seen on the second scan. A steady-state voltam-
mogram at a platinum-microdisk electrode (radius 12.5
µm) at 20 °C for the two irreversible oxidation processes
(1) and (2) shows that their limiting currents are similar
to each other and also similar to the current observed
for the first one-electron oxidation of mer-{Cr(CO)3-
(dpe)}2(µ-dpe).19 Thus, it is concluded that processes (1)
and (2) each involve one electron on the voltammetric
time scale.
F igu r e 3. Oxidative cyclic voltammograms at a glassy-
carbon electrode (radius 0.5 mm) of a 1.0 mM solution of
cis,fac-{Mn(CO)2(dpe)Br}2(µ-dpe) in dichloromethane (0.1
M Bu4NPF6; scan rate 200 mV s-1): (a) at 20 °C; (b) at 20
°C, reversing scan after process 1; (c) at -50 °C.
(4′) are no longer observed. This indicates that the
species giving rise to redox couples (3) and (4) result
from chemical steps after oxidation processes (1) and
(2). The evidence presented earlier shows that redox
couple (3) is associated with a product formed from
chemically irreversible process (1) and redox couple (4)
is associated with chemically irreversible process (2).
In view of the common experience of isomerization
following oxidation,18 it is suggested that isomerization
of one metal center occurs after process (1) to give a
mixed-geometry binuclear complex (cis,fac0/ trans+) and,
following process (2), both metal centers become trans
({trans+}2), and it is these species which give rise to
A cyclic voltammogram of this solution at -50 °C is
shown in Figure 3c. Processes (1) and (2) are now
chemically reversible, and reduction processes (3′) and
(19) Bond, A. M.; Colton, R.; Cooper, J . B.; McGregor, K.; Walter, J .
N.; Way, D. M. Organometallics 1995, 14, 49.

3174 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
shows a single carbonyl stretch at 1965 cm-1, indicating
that trans carbonyl groups now are present on both
metal centers, and the compound is identified as trans-
[{Mn(CO)2(dpe)Br}2(µ-dpe)]2+
.
Exhaustive bulk electrolysis at 20 °C of a fresh
dichloromethane solution of cis,fac-{Mn(CO)2(dpe)Br}2-
ox
(µ-dpe), at a potential slightly more positive than Ep
for process (2), indicates that two electrons are passed
per molecule of compound, and the same final IR
spectrum is observed; therefore, the same result is
obtained whether the electrolysis is performed in either
a one- or two-stage manner.
When a bulk reduction of the fully oxidized solution
is carried out at 20 °C at a potential slightly more
negative than Epred for process (3), the amount of charge
passed indicates that this is a two-electron process. A
steady-state voltammogram shows that processes (3)
and (4) are absent after the reduction, but processes (1)
and (2) are observed. The current associated with
processes (1) and (2) is equal to that for processes (3)
and (4) before the reduction, and the voltammogram
shows the reductive components of processes (1) and (2)
are present in the solution. This result shows that, on
the longer time scale of coulometry, the trans geometry
on the reduced manganese atoms reverts back to the
original cis,mer geometry.
F igu r e 4. Steady-state voltammograms at a platinum
microelectrode (radius 12.5 µm) of a 1.0 mM solution of cis,-
fac-{Mn(CO)2(dpe)Br}2(µ-dpe) in dichloromethane (0.1 M
Bu4NPF6) at different stages during bulk electrolysis at 20
°C: (a) before electrolysis; (b) after exhaustive oxidative
ox
electrolysis at a potential slightly less positive than Ep
for process 1; (c) after exhaustive oxidative electrolysis at
ox
a potential slightly more positive than Ep for process 2.
This isomerization is prevented if the bulk reduction
experiment is carried out at -40 °C (after the two-
electron oxidation at 20 °C), since under these conditions
the steady-state voltammogram shows the reductive
components of processes (3) and (4). The IR spectrum
of this solution, acquired quickly at room temperature,
shows a single band at 1897 cm-1, which is close to the
position of the carbonyl stretch for trans-Mn(CO)2(κ1-
dpm)(κ2-dpm)Br (Table 3). The 31P NMR spectrum
acquired at -60 °C (Table 3, Figure S4), obtained under
the same conditions but with 0.4 M Bu4NClO4 as the
electrolyte, displays three signals of equal intensities,
all of which occur at higher frequency than the reso-
nance position of free dpe, indicating that all the
phosphorus atoms are coordinated to the metal centers.
Thus, the spectroscopic data confirm the presence of
trans-{Mn(CO)2(dpe)Br}2(µ-dpe) in solution after the
oxidation and reduction cycle at -40 °C. Similar results
are obtained for the corresponding chloro compound.
(iii) Deta iled Discu ssion of th e Electr od e P r o-
cesses Detected u n d er Differ en t Con d ition s. The
exact nature of the voltammetric responses observed
depends on the conditions, especially the temperature,
which determines the rate of isomerization of cis,fac+
to trans+. Consequently, the voltammetry will be dis-
cussed at different temperatures using a shorthand
notation which indicates the geometry and charge of the
Mn(CO)2(dpe)Br moieties at each end of the dpe bridg-
ing ligand.
redox couples (3) and (4). Support for the validity of
these assignments will be given in the next section.
(ii) Bu lk Electr olysis. Exhaustive bulk electrolysis
at 20 °C of a solution of cis,fac-{Mn(CO)2(dpe)Br}2(µ-
dpe) in dichloromethane (0.1 M Bu4NPF6) was under-
taken at a platinum-gauze working electrode at a
potential slightly less positive than Epox for process (1).
The amount of charge passed after this exhaustive
electrolysis corresponds to a one-electron-oxidation pro-
cess. A steady-state voltammogram of the solution
before electrolysis is shown in Figure 4a, and a volta-
mmogram after electrolysis is shown in Figure 4b.
Process (1) is absent after a one-electron electrolysis,
and the sign of the current indicates that it has been
replaced by the reductive component of process (2).
Process (3) remains and still involves the same number
of electrons in the charge-transfer process as did process
(1) before electrolysis. The IR spectrum of the oxidized
solution shows three carbonyl stretches. Two at 1937
and 1865 cm-1 are identical with those of the starting
material and are assigned to the carbonyl groups
coordinated to the unoxidized manganese atom in the
binuclear complex. The third stretch at 1965 cm-1 occurs
at a position very close to that observed for trans-[Mn-
(CO)2(κ1-dpm)(κ2-dpm)Br]2+ and is assigned to trans
carbonyl groups on the oxidized manganese atom. This
stable product of the one-electron oxidation of cis,fac-
{Mn(CO)2(dpe)Br}2(µ-dpe) is identified as cis,fac-Mn-
(CO)2(dpe)Br(µ-dpe)-trans-[Mn(CO)2(dpe)Br]+, simpli-
fied to cis,fac(trans)-[{Mn(CO)2(dpe)Br}2(µ-dpe)]+.
Further bulk electrolysis of this solution, at a poten-
tial slightly more positive than Epox for process (2), gives
a second one-electron oxidation on the basis of the
charge passed. Cyclic voltammograms show that process
(2) has been replaced by process (4). A steady-state
voltammogram of the oxidized solution (Figure 4c)
reveals that the oxidative components of processes (3)
and (4) are present in the solution. The IR spectrum
At -50 °C, the voltammetric oxidation of (cis,fac0)2
(Figure 3c) exhibits two chemically reversible couples:
These are the true couples 1 and 2. However, at 20 °C,
the isomerization of cis,fac+ to trans+ is fast on the
voltammetric time scale (Figure 3a); thus, the observed
(cis,fac0)2 h (cis,fac0)(cis,fac+) + e- (couple 1) (12)
(cis,fac0)(cis,fac+) h (cis,fac+)2 + e- (couple 2) (13)

[M(CO)2(κ1-P-P)(κ2-P-P)X] and [{M(CO)2(κ2-P-P)X}2(µ-P-P)]
Organometallics, Vol. 23, No. 13, 2004 3175
responses (1) and (2) are actually due to
place upon oxidation, which led to the conclusion that
significant charge delocalization does not appear to be
evident in the mixed valent (cis,fac0)(trans+) complexes.
The potential changes introduced by chemical irrevers-
ibility (processes (1) and (2) at 20 °C) and variation of
the geometry of the moiety at the other end of the bridge
(couple 3 and process (3)) appear to be relatively small.
H. Electr och em ica l Stu d ies of cis,m er -{Re(CO)2-
(d p e)X}2(µ-d p e). An oxidative cyclic voltammogram of
a 1.0 mM solution of cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe)
in dichloromethane (0.1 M Bu4NPF6) shows two revers-
ible couples 1 and 2 for all cycles of the potential at a
scan rate of 200 mV s-1, and there is no indication on
this time scale of any isomerization to the trans
geometry. A steady-state voltammogram indicates that
each process involves one electron, and these waves are
assigned to the consecutive oxidations of the two rhe-
nium atoms from Re(I) to Re(II)
(cis,fac0)2 h (cis,fac0)(cis,fac+) + e- fast8
(cis,fac0)(trans+) (process (1)) (14)
(cis,fac0)(trans+) h (cis,fac+)(trans+) + e- fast8
(trans+)2 (process (2)) (15)
Thus, on the reverse scan, the reduction processes in
Figure 3a are
(trans+)2 + e- h (trans0)(trans+) (couple 4) (16)
(trans0)(trans+) + e- h (trans0)2 (couple 3) (17)
and these are the true redox couples 3 and 4, because
isomerization of trans bulk to the starting material cis,-
fac0 is slow on the voltammetric time scale. Alterna-
tively, these couples 3 and 4 also may be observed after
exhaustive oxidative electrolysis of (cis,fac0)2 at a po-
tential corresponding to process (2), which also gener-
ates (trans+)2.
The processes shown in Figure 3b are subtly different.
When only the initial one electron oxidation is carried
out at room temperature, the product is (cis,fac0)(trans+)
and its reduction to (cis,fac0)(trans0) appears as process
(3′). Since isomerization of trans0 to cis,fac0 is slow on
the voltammetic time scale, the oxidation response (3)
is due to the reoxidation of (cis,fac0)(trans0) to (cis,fac0)-
(trans+) which occurs at a potential close to that of
couple 3 (eq 17). Couple 4 is the simplest response of
all and is given only by the process shown in eq 16.
The fact that the reversible potentials for the four
couples (eqs 12, 13, 15, and 16) and other electron-
transfer reactions involving (cis,fac0)(trans+) differ im-
plies that a large number of thermodynamically favored
second-order cross-redox reactions can occur, e.g.
cis,mer-{Re(CO)2(dpe)Br}2(µ-dpe) h
cis,mer-[{Re(CO)2(dpe)Br}2(µ-dpe)]+ + e- (18)
cis,mer-[{Re(CO)2(dpe)Br}2(µ-dpe)]+ h
cis,mer-[{Re(CO)2(dpe)Br}2(µ-dpe)]2+ + e- (19)
At more positive potentials, an irreversible oxidation
response (process 3) is observed, which probably in-
volves further oxidation of the metal centers. Similar
responses are observed for cis,mer-{Re(CO)2(dpp)Br}2-
(µ-dpp), and data for all compounds of this type are
summarized in Table 6.
I. Electr osp r a y Ma ss Sp ectr om etr y. (i) Ma n ga -
n ese Com p lexes. ESMS provides an independent
probe to verify the stoichiometry, but not the isomeric
form, of species in solution and is used in this paper to
provide additional verification of the identity of the
compounds generated electrochemically. For ionic com-
pounds, the cation may be observed directly without
problems, but also there are now a number of recognized
techniques to convert unobservable neutral species into
predictable and closely related ionic derivatives which
may be observed by ESMS.20 For example, neutral
species such as metal Schiff base complexes, ML2, may
be protonated by the mobile phase in the spectrometer
(frequently H2O/MeOH/HOAc 50:50:1) and the intact
ions [HML2]+ are readily observed.21 Alternatively,
metal ion addition (metal adduction) may sometimes be
successful; for example, in the presence of sodium ions
in the solution the neutral anticancer agent cis-PtCl2-
(NH3)2 gives the ion [Na + PtCl2(NH3)2]+.22 Such
methods also are frequently applicable with respect to
carbonyl complexes.
(trans+)2 + (cis,fac0)2 f (trans+,trans0) + (cis,fac+)
These reaction pathways can give rise to a range of
homogeneous reactions that are may be coupled to the
electron-transfer processes and hence modify (catalyze)
isomerization reactions. These second-order processes
have not been included in the above discussion of the
voltammetry, bulk electrolysis, or isomerization pro-
cesses.
The potentials of processes (1)-(4) are determined
predominantly by the ligand geometry of the metal atom
being oxidized/reduced (processes (1) and (2) for cis,fac
and processes (3) and 4 for trans) and the charges
involved in the process (+/0 and 2+/+). The potential
difference between cis,fac+/0 and trans+/0 (couples 1 and
3) is exactly the same (538 mV) as between cis,fac2+/+
and trans2+/+ (couples 2 and 4), and is similar to values
observed in other systems.18 The difference in potential
for successive oxidations for each geometry (couples 1
and 2 for cis,fac and couples 3 and 4 for trans) are also
equal at 124 mV, and again this value is typical for
successive oxidations in binuclear species of this kind.19
This relatively small difference in potential between
isostructural redox couples implies that the two redox
centers are not strongly interacting. This is consistent
with the earlier discussion of the IR changes that takes
All ESMS data for manganese complexes are sum-
marized in Table 1. Addition of sodium ions to a solution
of fac-Mn(CO)3(dpe)Br enables observation of the ion
[Na + Mn(CO)3(dpe)Br]+ (m/z 641). Another peak at m/z
537 is assigned to the ion [Mn(CO)3(dpe)]+ and is formed
by elimination of NaBr. Increasing the B1 voltage
(20) Colton, R.; D’Agostino, A.; Traeger, J . C. Mass Spectrom. Rev.
1995, 14, 79.
(21) Cardwell, T. J .; Colton, R.; Mitchell, S.; Traeger, J . C. Inorg.
Chim. Acta 1994, 216, 75.
(22) Poon, G. K.; Mistry, P.; Lewis, S. Biol. Mass Spectrom. 1991,
20, 687.

3176 Organometallics, Vol. 23, No. 13, 2004
Bond et al.
causes this peak to increase in relative intensity,
thereby demonstrating that it is formed from the intact
ion by collisional activation in the gas phase. Similarly,
addition of sodium ions to cis,mer-Mn(CO)2(κ1-dpm)(κ2-
dpm)Br gives the sodium adduct of the intact molecule
[Na + Mn(CO)2(dpm)2Br]+ (m/z 983), together with a
number of other ions, including [Mn(CO)2(dpm)2]+ (loss
of NaBr) and [Mn(CO)2(dpmO)(dpm)]+ (m/z 895). Oxida-
tion of phosphine ligands, especially those with pendant
phosphorus atoms, is a common observation in ESMS.24
A number of the Mn(CO)2(κ1-dpm)(κ2-P-P)X compounds
were examined by ESMS without any additives in the
solution, and in all cases the ions [H + Mn(CO)2(dpm)-
(P-P)X]+ are observed (Table 1). These are probably
simple protonated species, with the proton being derived
from the mobile phase. It is important to realize that
observation of such protonated species does not imply
that a considerable fraction of the solute is protonated;
indeed all the indications are that these manganese
carbonyl derivatives are not very basic. However, the
combination of the extreme sensitivity of mass spec-
trometry and the inability of the technique to detect the
neutral unprotonated molecules allow the observation
of minute proportions of protonated species. These in
turn allow confirmation of the stoichiometry of the
compound in solution.
dpe clearly shows a preference to coordinate both its
phosphorus atoms, either as in the µ-dpe dimers or,
under more forcing conditions, the κ2 complexes cis-[Re-
(CO)2(dpe)2]+ and trans-Re(CO)(dpe)2X.
The chemistry of the 17e M(II) species formed by
electrochemical oxidation is even more varied and shows
considerable differences between the chemistries of the
two metals. It is well established18 that the oxidation
potentials of complexes of the second-/third-row metals
are more positive than those of the corresponding first-
row metal derivatives. This has interesting conse-
quences in the case of the M(CO)2(κ1-dpm)(κ2-dpm)X
complexes, as the oxidation potential for the pendant
phosphine lies between the oxidation potentials for the
two metals. The manganese compounds are well-
behaved in the sense that the metal-based oxidation is
accompanied by the usual isomerization to a trans-
carbonyl arrangement, but there is the added feature
of slow displacement of the halo ligand to generate [Mn-
(CO)2(dpm)2]2+. The isomerization characteristics of cis,-
mer-[Re(CO)2(κ1-dpm)(κ2-dpm)X]+ are not known be-
cause of the rapid formation of cis,mer-[Re(CO)2(κ1-
dpmO)(κ2-dpm)X]+, but it is interesting that both this
cation and cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)X]+ do
not isomerize on the voltammetric time scale, and
analogous reactivities are observed for the ape com-
plexes.
The oxidant NOBF4 was added to the solutions of the
binuclear molecules cis,fac-{Mn(CO)2(dpe)X}2(µ-dpe) to
generate the cationic forms which then are observed,
thus confirming unequivocally the binuclear nature of
the compounds.
The binuclear manganese complexes cis,fac-{Mn-
(CO)2(dpe)X}2(µ-dpe) are arguably the most interesting
compounds discussed in this paper. The stability of the
singly and doubly oxidized binuclear complexes is
remarkable when compared with those derived from the
isoelectronic chromium complex mer,mer-{Cr(CO)3-
(dpe)}2(µ-dpe). In that system, the monocation is stable
on the synthetic time scale only at -78 °C, while the
dication is stable only on the voltammetric time scale
at that temperature.19 The oxidized manganese di-
nuclear compounds are not only stable at room temper-
ature, but each half of the molecule is well behaved
(almost independent of the other) in its isomerization
properties. In contrast, while the rhenium complexes
have notable stability for the oxidized species, they do
not seem to rapidly isomerize upon oxidation.
(ii) Rh en iu m Com p lexes. All ESMS data for rhen-
ium complexes are summarized in Table 2. Ionic species
such as cis,mer-[Re(CO)2(κ1-dpmMe)(κ2-dpm)X]PF6 and
cis-[Re(CO)2(dpmO)2]ClO4 all give a strong peak for
their intact cations. Neutral monomeric species such as
cis,mer-[Re(CO)2(κ1-dpm)(κ2-dpm)Br are observed as
either their sodium ion adducts or as protonated species.
ESMS provides unequivocal proof for the binuclear
nature of the species, formulated as cis,mer-{Re(CO)2-
(dpe)X}2(µ-dpe), since they were observed as both their
intact sodium adducts and as the intact corresponding
singly charged cation after addition of the one-electron
chemical oxidant NOBF4.
Gen er a l Discu ssion a n d Con clu sion s
Ack n ow led gm en t. We gratefully acknowledge the
Australian Research Council for financial support.
J .N.W. thanks the Australian Commonwealth Govern-
ment for a Post Graduate Research Award.
The chemistry of the diphosphine derivatives of the
group 7 carbonyl halides is very rich in both the 18e
and 17e oxidized states. In the 18e M(I) complexes, the
main variations detected in the chemistry are a function
of the ligands; dpm easily forms κ1/κ2 derivatives, while
Su p p or tin g In for m a tion Ava ila ble: Figures S1-S7,
containing NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
(23) Colton, R.; Dakternieks, D. Inorg. Chim. Acta 1993, 208, 173.
(24) Colton, R.; Harvey, J .; Traeger, J . C. Org. Mass Spectrom. 1992,
27, 1033.
OM034388T