Electrochemistry of group VI metal carbonyl compounds containing 1,1′-bis(diphenylphosphino)ferrocene

Article abstract of DOI:10.1021/om049735t

The oxidative electrochemistry of a variety of homo- and bimetallic group VI metal carbonyl compounds containing 1,1′-bis(diphenylphosphino) ferrocene (dppf) with the general formulas M(CO)₅(dppf) (1a-c), M(CO)₄(dppf) (2a-c), and (CO)₅M(dppf)M'(CO)₅ (3a-f) (M, M' = Cr, Mo, W) was investigated. The effect of the group VI metal and the coordination mode of the dppf ligand on the potential at which the oxidation of the group VI metals and the ferrocene backbone of dppf occurred was examined.

Full text of DOI:10.1021/om049735t

Organometallics 2004, 23, 4655-4660
4655
Electr och em istr y of Gr ou p VI Meta l Ca r bon yl
Com p ou n d s Con ta in in g
1,1′-Bis(d ip h en ylp h osp h in o)fer r ocen e
Amanda C. Ohs,†,‡ Arnold L. Rheingold,§ Michael J . Shaw,| and Chip Nataro*,†
Department of Chemistry, Lafayette College, Easton, Pennsylvania 18042, Department of
Chemistry and Biochemistry, University of California, La J olla, California 92093, and
Department of Chemistry, Southern Illinois University, Edwardsville, Illinois 62026
Received April 12, 2004
The oxidative electrochemistry of a variety of homo- and bimetallic group VI metal carbonyl
compounds containing 1,1′-bis(diphenylphosphino)ferrocene (dppf) with the general formulas
M(CO)5(dppf) (1a -c), M(CO)4(dppf) (2a -c), and (CO)5M(dppf)M′(CO)5 (3a -f) (M, M′ ) Cr,
Mo, W) was investigated. The effect of the group VI metal and the coordination mode of the
dppf ligand on the potential at which the oxidation of the group VI metals and the ferrocene
backbone of dppf occurred was examined.
In tr od u ction
A variety of reactions have been catalyzed using
transition-metal compounds that contain the ligand 1,1′-
bis(diphenylphosphino)ferrocene (dppf).1 A number of
reasons for the enhanced catalytic activity of these
compounds have been presented: in particular, the bite
angle of dppf and the redox-active ferrocene backbone.1
The oxidative electrochemistry of dppf has been studied,
and unlike its close relative ferrocene, the product of
dppf oxidation undergoes a chemical reaction.2 However,
the dppf-based oxidation often becomes reversible when
F igu r e 1. Monodentate, bidentate, and bridging coordina-
dppf is coordinated to a metal center, as seen in the
oxidative electrochemistry of [MCl2(dppf)] (M ) Pd, Pt).3
Although electrochemical studies have been performed
on compounds containing a dppf ligand, none have in-
vestigated the effect of varying dppf coordination modes
on the electrochemistry of the ferrocene backbone.
The most commonly found modes of dppf coordination
are monodentate, bidentate, and bridging.1,4 A variety
of homo- and bimetallic group VI metal carbonyl com-
pounds with varying dppf coordination modes have been
prepared (Figure 1).5 It is surprising that the electro-
chemistry of the entire series has not been explored,
since this series presents dppf bound to a series of metal
centers in a variety of coordination modes. In fact, only
one of these compounds, Mo(CO)4(dppf), has been in-
vestigated electrochemically; the oxidation shows a
tion modes of dppf.
chemically irreversible wave at 0.42 V versus Fc0/+ in
THF.6 The effect of group VI metals on the electrochem-
istry of coordinated dppf has been investigated in a
series of compounds, [MCln{(µ-dppf)M′(CO)5}n] (M ) Au,
M′ ) Cr, Mo, W, n ) 1; M ) Pd, Pt; M′ ) Cr, Mo, W, n
) 2).7 Data from this series have shown that there does
not appear to be a significant difference in the potential
at which the ferrocene backbone undergoes oxidation
as M′ varies. However, in both series, the dppf ligand
bridges two different metal centers; thus, the effect of
varying coordination modes on the oxidative electro-
chemistry could not be examined.
The purpose of this work is to investigate how the
three known coordination modes of dppf to group VI
metal carbonyls and the variation of the group VI metal
present influence the electrochemistry of the compound.
The oxidative electrochemistry of 12 different dppf-
containing compounds was examined, and the effect of
the dppf coordination mode (monodentate, bidentate, or
bridging) on the ferrocene-based oxidation of a dppf
ligand was determined. In addition, the relationship
between the coordination mode of the dppf ligand and
the potential at which the oxidation of the group VI
* To whom correspondence should be addressed. E-mail: nataroc@
lafayette.edu.
† Lafayette College.
‡ Current address: Department of Chemistry and Biochemistry,
University of California, La J olla, CA 92093.
§ University of California.
| Southern Illinois University.
(1) (a) Gan, K. S.; Hor, T. S. A. In Ferrocenes; Togni, A., Hayashi,
T., Eds.; VCH: New York, 1995. (b) Colacot, T. J . Platinum Met. Rev.
2001, 45, 22.
(2) Nataro, C.; Campbell, A. N.; Ferguson, M. A.; Incarvito, C. D.;
Rheingold, A. L. J . Organomet. Chem. 2003, 673, 47.
(3) Corain, B.; Longato, B.; Favero, G.; Ajo`, D.; Pilloni, G.; Russo,
U.; Kreissl, F. R. Inorg. Chim. Acta 1989, 157, 259.
(4) Bandoli, G.; Dolmella, A. Coord. Chem. Rev. 2002, 209, 161.
(5) Hor, T. S. A.; Phang, L.-T. J . Organomet. Chem. 1989, 373, 319.
(6) DuBois, D. L.; Eigenbrot, C. W., J r.; Miedaner, A.; Smart, J . C.;
Haltiwanger, R. C. Organometallics 1986, 5, 1405.
(7) Phang, L.-T.; Au-Yeung, S. C. F.; Hor, T. S. A.; Khoo, S. B.;
Zhou, Z.-Y.; Mak, T. C. W. J . Chem. Soc., Dalton Trans. 1993, 165.
10.1021/om049735t CCC: $27.50 © 2004 American Chemical Society
Publication on Web 08/27/2004

4656 Organometallics, Vol. 23, No. 20, 2004
Ohs et al.
Ta ble 1. Cr ysta l Da ta a n d Str u ctu r e Refin em en t
metal occurs was explored. During the course of this
work, alternative syntheses for compounds 2a , 3d -f,
and the related compound Cr(CO)4(dppm) were devel-
oped. In addition, compounds 2a and 3a were charac-
terized by spectroelectrochemistry and the structure of
compound 2a was determined by X-ray crystallography.
Deta ils for Cr (CO)4(F e(C5H4P P h 2)2)‚CH2Cl2
formula
formula wt
cryst syst
space group
a, Å
C39H30Cl2CrFeO4P2
803.32
triclinic
P1h
10.1832(14)
b, Å
c, Å
10.9023(17)
17.987(3)
Exp er im en ta l Section
R, deg
82.471(3)
76.908(3)
84.290(3)
1923.2(5)
1.387
2
820
Gen er a l P r oced u r es. Preparative reactions were carried
out under an atmosphere of argon using Schlenk techniques.
Solvents were purified under nitrogen using standard meth-
ods. Hexanes and methylene chloride (CH2Cl2) were refluxed
over CaH2 and then distilled. HPLC grade CH2Cl2, from
Aldrich, used in electrochemical measurements was refluxed
over CaH2 and distilled under an atmosphere of argon.
Tetrahydrofuran (THF) was distilled from potassium ben-
zophenone ketyl. The metal hexacarbonyls, dppf, decameth-
ylferrocene (Fc*), and acetylferrocene were purchased from
Strem. Cr(CO)6 was freshly sublimed prior to use. Tetrabu-
tylammonium hexafluorophosphate (NBu4PF6), trimethyl-
amine N-oxide (TMNO), and silica gel (200-400 mesh, 60 Å)
were purchased from Aldrich. Tris(4-bromophenyl)aminium
hexafluorophosphate, [N(C6H4Br)3][PF6], was prepared accord-
ing to the literature procedure.8
â, deg
γ, deg
V, Å3
D
Z
calcd, g cm-3
F(000)
cryst size, mm
T, K
wavelength, Å
abs coeff, mm-1
θ range for data collecn, deg
index ranges
0.40 × 0.30 × 0.10
223(2)
0.710 73
0.920
2.11-23.26
-11 e h e 5, -12 e k e 12,
-19 e l e 9
7401
5245 (R(int) ) 0.0274)
none
0.9137 and 0.7099
full-matrix least squares on F2
5245/0/415
no. of rflns collected
no. of indep rflns
abs cor
max and min transmissn
refinement method
no. of data/restraints/params
goodness of fit on F2
final R indices (I > 2σ(I))
R indices (all data)
P r ep a r a tion of Com p ou n d s. Syn th esis of 1a -c, 2a -c,
a n d 3a -c. These compounds were prepared according to
literature methods with minor modifications in the purification
process.5 Instead of using thin-layer chromatography, purifica-
tion was accomplished with column chromatography, on silica
gel, using mixtures of hexanes and CH2Cl2.
1.042
R1 ) 0.0351, wR2 ) 0.0794
R1 ) 0.0440, wR2 ) 0.0808
largest diff peak and hole, e Å-3 0.446 and -0.414
Alter n a te Syn th esis of 2a . Compound 2a was prepared
in 65% yield by the photolysis of a mixture of Cr(CO)6 and
dppf in THF.
dissolved in a minimum amount of CH2Cl2. The solution
immediately turned purple and was stirred for 20 min to
ensure that the reaction was complete. IR (CH2Cl2): νCO (cm-1
)
Syn th esis of 3d ,f. These compounds were prepared using
a slight modification of the literature synthesis.9 In separate
flasks, equimolar quantities of the appropriate metal hexa-
carbonyl were reacted with TMNO in a 3:8 mixture of CH2Cl2
and THF. These solutions were then added to an equimolar
amount of dppf in a 3:8 mixture of CH2Cl2 and THF. Following
removal of the solvent, the compounds were purified by column
chromatography.
Syn th esis of 3e. A solution of W(CO)6 in THF was irradi-
ated in a photolysis vessel with a 5.5 W mercury lamp. An
equimolar amount of Cr(CO)6 was similarly reacted in a second
photolysis vessel. The solutions were then added to an equimo-
lar amount of dppf in THF. The product was purified by
column chromatography on silica gel.
Syn th esis of Cr (CO)4d p p m . Cr(CO)6 (0.035 g, 0.16 mmol)
and dppm (0.027 g, 0.071 mmol) were placed in a photochemi-
cal reaction vessel equipped with a stir bar. THF (10.0 mL)
was added, and the solution was cooled by flowing cold water
through a jacketed quartz immersion well. Argon was bubbled
through the solution, and the reaction mixture was irradiated
for 45 min using a 5.5 W mercury lamp. The solution was
transferred to a Schlenk flask and stirred for 2 h. The volume
was then reduced in vacuo, and the product was purified by
column chromatography on silica gel. The product was eluted
using a mixture of 1:10 CH2Cl2 and hexanes. This yielded 7.2
mg (26% yield) of the desired product, as confirmed by
comparing the IR spectrum to the literature data.10
2071 (m) and 1966 (m).
Cr ysta llogr a p h y. Crystals of 2a were obtained by dissolv-
ing the solid in a minimum amount of CH2Cl2, layering the
solution with hexanes, and then cooling the mixture in a
freezer. Crystals were found to belong to the triclinic crystal
system, and centrosymmetry was initially assumed and later
demonstrated to be correct. Crystallographic data are pre-
sented in Table 1. The asymmetric unit consists of one
Cr(CO)4dppf molecule and a molecule of CH2Cl2 (electron count
41, expected 42). The latter was found to be highly disordered
and treated as a diffuse electron density using the routine
SQUEEZE (A. Spek, Platon Library). All non-hydrogen atoms
were refined anisotropically, and hydrogen atoms were treated
as idealized contributions. An absorption correction using
SADABS was applied to the data. All computations used
software provided by the Bruker Corp. (Madison, WI).
Electr och em ica l Mea su r em en ts. Electrochemical studies
were performed in 10.0 mL of CH2Cl2 at ambient temperature
using a Princeton Applied Research Model 263-A potentiostat.
A three-electrode configuration was used with a 1.5 mm glassy-
carbon-disk working electrode, a Pt-wire counter electrode, and
a nonaqueous Ag|AgCl reference electrode separated from the
solution by a fine glass frit. Prior to use, the working electrode
1
was polished first with a 1 µm diamond paste followed by a /4
µm diamond paste; after each polishing, the electrode was
rinsed with acetone. The experiments were carried out under
a moisture-free argon atmosphere, and the argon was bubbled
through a presaturation tube containing CH2Cl2. Solutions of
the analyte were 1.0 mM, and NBu4PF6 (0.1 M) was used as
the supporting electrolyte. Scans were made at 50 mV/s and
from 100 to 1000 mV/s at 100 mV/s increments. Normal pulse
voltammetry experiments were conducted at a scan rate of 5
mV/s.
Ch em ica l Oxid a tion of 2a . Acetylferrocenium hexafluo-
rophosphate was prepared by reacting equimolar amounts
(0.013 mmol) of acetylferrocene and [N(C6H4Br)3][PF6] in a
minimal amount of CH2Cl2. This solution was then added
dropwise to a solution of 2a (9.0 mg, 0.013 mmol) already
(8) Eberson, L.; Larsson, B. Acta Chem. Scand., Ser. B 1987, 41,
367. Eberson, L.; Larsson, B. Acta Chem. Scand., Ser B 1986, 40, 210.
(9) Hor, T. S. A.; Phang, L.-T. Polyhedron 1990, 9, 2305.
(10) Colquhoun, I. J .; Grim, S. O.; McFarlane, W.; Mitchell, J . D.;
Smith, P. H. Inorg. Chem. 1981, 20(8), 2516.
Sp ectr oelectr och em istr y. Fiber-optic spectroelectrochem-
ical experiments were performed using a Bruker Tensor 27
FTIR modified with a Remspec fiber-optic infrared system
equipped with a low-temperature probe and liquid-nitrogen-

Metal Carbonyl Compounds Containing Ferrocene
Organometallics, Vol. 23, No. 20, 2004 4657
cooled MCT detector. Current measurements were made, and
potentials were controlled using a three-electrode system
attached to an EG&G PAR 263 computer-controlled poten-
tiostat.
In a typical experiment, the cell was set up as described in
the literature with the fiber-optic waveguide bringing the IR
beam to and from the electrode surface.11 Supporting electro-
lyte (NBu4PF6, 0.40 g) and dry CH2Cl2 (10.0 mL) were
introduced, and the solution was degassed with a slow stream
of dry, deoxygenated nitrogen for 5 min. Baselines were
recorded for both the electrochemical (cyclic voltammetry, 200
mV/s) and the IR parts of the experiment. Enough sample was
introduced to the cell to yield a 1.0 mM solution of analyte.
Cyclic voltammograms were recorded at 200 mV/s to determine
the appropriate potentials for spectroelectrochemical analysis.
To obtain the difference spectra, a background IR spectrum
was recorded for 1 min; the potential of the electrode was set
to convert the analyte into its electrode product, and after 20
s of electrolysis, a 1 min IR scan was recorded with the
potential still being applied.
F igu r e 2. ORTEP diagram of 2a with ellipsoids drawn
at the 30% probability level. Select bond distances (Å) and
angle (deg): Cr-P(1) ) 2.4393(9), Cr-P(2) ) 2.3984(9),
Cr-C (av) ) 1.864, C-O (av) ) 2.306; Cr-C-O (av) )
174.0.
Resu lts a n d Discu ssion
Syn th esis. Compound 2a was prepared photochemi-
cally from Cr(CO)6 and dppf in THF at 0 °C. The syn-
theses of compounds 3d ,f were carried out by generating
equimolar amounts of the requisite M(CO)5(thf) com-
plexes and adding them to dppf. The literature method
for preparing 3d requires the preparation of 1b, which
is then reacted with Cr(CO)5(thf).9 Similarly, 3f is
prepared from 1c and Mo(CO)5(thf).9 While the reported
yields appear to be significantly greater (33% for 3d and
26% for 3f)9 than those reported in this work, this is
somewhat misleading. The yields of 1b,c are 3% and
27%, respectively.5 On the basis of the initial amount
of dppf used, the overall yield of 3d from the literature
preparation is 0.9%, while that of 3f is 7%. In addition,
the literature syntheses require two purification steps,
whereas our syntheses of these compounds give similar
yields but require only one column.
Ta ble 2. P a r a m eter s of d p p f Coor d in a tion in
M(CO)4d p p f (M ) Cr , Mo, W)
Cr
Mo
W
ref
this work
96.67(3)
179.6
64.02
36
12
13
P-M-P (deg)
95.28(2)
179.03
67.32
41.9
95.24(5)
178.54d
66.94d
40.7d
a
XA-M′-XB (deg)
P-Fe-P (deg)
τb (deg)
av P-M (Å)
P-P (Å)
2.4189
3.614
0.042
2.560
3.784
0.022
2.5480
3.764d
0.049d
av δPc (Å)
a
b
centroid-Fe-centroid. The torsion angle CA-XA-XB-CB,
where C is the carbon bonded to the P and X is the centroid.
c Deviation of the P atom from the Cp plane, a positive value
d
meaning the P is closer to Fe. Determined from the published
structural data using ORTEP-3 for Windows.23
A new photochemical synthesis of 3e was developed,
giving the product in 24% yield. This yield exceeds the
literature method, which required isolation of 1a fol-
lowed by addition of W(CO)5(thf).9 Our photochemical
synthesis is significantly less time-consuming, requiring
2 h for the photolysis of the hexacarbonyls, as opposed
to the 3.5 h required in the TMNO synthesis. Other
advantages of the photochemical synthesis include the
use of milligram quantities of reactants instead of grams
and the need for only one column separation, rather
than two.
Str u ctu r e. During the preparation of 2a , crystals
suitable for X-ray analysis were obtained. The struc-
ture of 2a was determined, completing the series of
M(CO)4(dppf) (M ) Cr, Mo, W) structures. The
ORTEP diagram shows that the chromium atom adopts
containing compounds.1a,4 The ideal τ angle for the
synclinal staggered conformation is 36°,4 which is the
angle found for 2a . This ideal angle has never been seen
in a pseudo-octahedral compound and has only been
reported in two other structures, both of which are
pseudo square planar.14 It is interesting that 2a exhibits
an ideal τ angle and yet exhibits the greatest distortion
of the group VI metal center from the ideal octahedral
geometry.
Electr och em istr y. Cyclic voltammetry was used to
determine the potential at which oxidation occurs for
the complexes. The decamethylferrocene (Fc*)/deca-
methylferrocenium (Fc*+) couple was used as an inter-
nal standard in all measurements, because the ferrocene
(Fc) couple overlaps the range of potentials for the
compounds studied. All potentials were referenced to
Fc/Fc+ by subtracting 0.55 V from the potentials refer-
a
distorted-octahedral geometry (Figure 2). The
P(1)-Cr-P(2) angle is 96.67(3)°, which is significantly
larger than the analogous angle in the molybdenum
(95.28(2)°)12 and tungsten (95.24(5)°)13 derivatives. In
addition, a variety of bonding parameters typically re-
ported for dppf-containing compounds are presented for
the entire series of M(CO)4dppf compounds in Table 2.
All of the dppf ligands for this series of compounds
are synclinal staggered, which is defined as having a τ
angle between 25 and 45°.1a,4 This arrangement is the
most common configuration for η2 mononuclear dppf-
(11) Shaw, M. J .; Henson, R. L.; Houk, S. E.; Westhoff, J . W.; J ones,
M. W.; Richter-Addo, G. B. J . Electroanal. Chem. 2002, 534, 47.
(12) Butler, I. R.; Cullen, W. R.; Kim, T. J .; Rettig, S. J .; Trotter, J .
Organometallics 1985, 4, 972.
(13) Song, L. C.; Liu, J . T.; Hu, Q. M.; Wang, G. F.; Zanello, P.;
Fontani, M. Organometallics 2000, 25, 5342.
(14) (a) Longato, B.; Pilloni, G.; Valle, G.; Corain, B. Inorg. Chem.
1988, 27, 956. (b) Vasapollo, G.; Toniolo, L.; Cavinato, G.; Bigoli, F.;
Lanfranchi, M.; Pellinghelli, M. A. J . Organomet. Chem. 1994, 481,
173.

4658 Organometallics, Vol. 23, No. 20, 2004
Ohs et al.
Ta ble 3. P oten tia ls (V) a t 100 m V/s for th e
Oxid a tion of Com p ou n d s 1a -c, 2a -c, a n d 3a -f vs
F c0/+ a
compd
E°′ (Fe)
ir/if
Ep
ir/if
Ep
ir/if
1a
1b
1c
2a
2b
2c
3a
3b
3c
3d
3e
3f
0.33
0.32
0.34
0.35
0.33
0.37
0.50
0.51
0.52
0.50
0.51
0.52
0.86
0.75
0.81
0.99
0.99
0.93
0.99
0.68
0.48
0.74
0.65
0.68
0.90
1.02
1.00
0.21
0.96
1.02
0.92
1.05
1.09
0.98
1.03
1.05
1.29
1.35
1.38
a
All data were obtained from 1.0 mM solutions of the compound
in CH2Cl2 with 0.1 M NBu4PF6 as the supporting electrolyte.
F igu r e 3. IR spectra after electrochemical oxidation of 2a .
The spectra were obtained using NBu4PF6 (0.1 M) as the
supporting electrolyte in 10.0 mL of CH2Cl2. Cyclic vol-
tammograms were recorded at 200 mV/s with a 1.0 mM
solution of the analyte.
enced to Fc*/Fc*+.15 The potentials at which oxidation
occurs were measured for each compound and are listed
in Table 3.
There were two waves in the oxidative electrochem-
istry of the η1 compounds (1). The wave at the less
positive potential was assigned as a dppf-based oxi-
dation, while the second wave was assigned as oxida-
tion of the group VI metal. The potential at which
the oxidation of the coordinated dppf in 1 occurs is
shifted by approximately +0.1 V as compared to free
dppf. This shift is significantly less than the shift seen
in [MCl2(dppf)] (M ) Pd, Pt), which is approximately
0.4 V.3 For all three compounds, the second wave was
irreversible at all scan rates employed. The first wave
was reversible if the potential was switched prior to
reaching the second wave but was not completely
reversible when the second wave was scanned.
sten (2c) analogues. There are two reversible waves in
the oxidation of 2a , and they have similar peak currents.
From normal pulse voltammetry data the slope of the
plot of E vs log((Id - I)/I) is expected to be 59 mV/n,
and for the first wave the slope was determined to be
63 mV, indicating a one-electron process.19 The first
wave occurs at a potential considerably lower than that
observed for other dppf-containing compounds in this
study; however, the second oxidation is within the range
of the potentials observed for the dppf-based oxidations
of the other compounds. Therefore, it was essential to
assign the oxidation waves in 2a .
To assign the waves, spectroelectrochemical measure-
ments were performed on 2a (Figure 3). When the
potential is set more positive than the first oxidation,
peaks at 2063 and 1961 cm-1 appear, while peaks at
2005 and 1882 cm-1 diminish. This shift of approxi-
mately 65 cm-1 is characteristic of oxidation of a metal
with carbonyl ligands.20 When the potential is set more
positive than the second oxidation, no significant change
is seen in the IR spectrum. Further support for the first
oxidation being chromium-based comes from the chemi-
cal oxidation of 2a with acetylferrocenium hexafluoro-
phosphate, which was chosen since it undergoes oxida-
tion at a potential (0.27 V) between the first and second
waves of 2a .21 After the chemical oxidation of 2a , new
νCO bands were observed in the IR spectrum at 2063
and 1966 cm-1, similar to the observations made during
the spectroelectrochemical experiment.
The electrochemistry of an analogous compound,
Cr(CO)4(dppm), was studied as additional support for
the first oxidation being chromium-based. Cr(CO)4-
(dppm) exhibits one chromium-based oxidation wave at
0.22 V. In addition, the shifts in the IR spectrum of
Cr(CO)4(dppm)+ compared to Cr(CO)4(dppm) should
be similar to those of 2a + versus 2a if the first oxida-
tion of 2a is occurring at the chromium center. The IR
The oxidations of the group VI metals of 1 occur at
approximately 1 V and have peak currents similar to
those of the dppf-based waves. The potentials are less
positive than those of the corresponding metal hexa-
carbonyls but significantly more positive than for other
M(CO)5PR3 compounds. In CH3CN, the oxidation of
Cr(CO)5PMe3 occurs at 0.62 V in CH3CN, while that of
Cr(CO)6 is 1.1 V.16 The chromium-based oxidation of 1a
is anticipated to be less positive than that of Cr(CO)6,
since an electron-withdrawing CO ligand is replaced by
an electron-donating phosphine ligand. However, the
potential was shown to be significantly more positive
than for other Cr(CO)5PR3 compounds, which can be
rationalized two ways. First, dppf is not as donating as
other phosphines.17 The second possibility is that the
dppf-based oxidation occurs prior to the group VI metal
oxidation and the dppf+ ligand is anticipated to be an
even weaker donor than dppf. A similar trend was noted
in the oxidative electrochemistry of M(CO)5(FcPh2P) (M
) Cr, Mo, W) where the Fc group of the phosphine is
oxidized prior to the group VI metal.18
In the series of η2 compounds, the oxidative electro-
chemistry of the chromium compound (2a ) is quite
different from that of the molybdenum (2b) and tung-
(15) Camire, N.; Mueller-Westerhoff, U. T.; Geiger, W. E. J . Orga-
nomet. Chem. 2001, 823, 637.
(16) Bond, A. M.; Carr, S. W.; Colton, R.; Kelly, D. P. Inorg. Chem.
1983, 22, 989.
(19) For reversible reactions at slow scan rates, this approach is
valid. Bard, A. J .; Faulkner, L. R. In Electrochemical Methods:
Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001; p
282.
(17) O’Connor, A. R.; Nataro, C. Organometallics 2004, 23, 615.
(18) Kotz, J . C.; Nivert, C. L.; Lieber, J . M. J . Organomet. Chem.
1975, 91, 87.
(20) Camire, N.; Nafaddy, A.; Geiger, W. E. J . Am. Chem. Soc. 2002,
124, 7260.
(21) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 7.

Metal Carbonyl Compounds Containing Ferrocene
Organometallics, Vol. 23, No. 20, 2004 4659
F igu r e 5. Plot of E°′ values of the ferrocene-based oxida-
tion (R2 ) 0.971) for compounds 3a -f versus the average
position of the metals in group VI.
F igu r e 4. Cyclic voltammogram of 3a measured at 100
mV/s in a 1.0 mM solution of the analyte in CH2Cl2 with
NBu4PF6 (0.1 M) as the supporting electrolyte.
and no significant shift in the νCO bands was observed.
However, when the potential was set more positive than
the second oxidation, a shift of about 71 cm-1 was
observed in the νCO bands. This magnitude of shift is
characteristic of an oxidation of a metal with carbonyl
ligands.20 These results, along with the amplitude of the
waves, are indicative of the first wave being dppf-based
and the second being chromium-based.
spectrum of [Cr(CO)4(dppm)][BF4] has been reported,
and the peaks shift by approximately 70 cm-1 22
similar
,
to spectroelectrochemical and chemical oxidation of 2a .
On the basis of these studies it can be concluded that
the first wave in the oxidation of 2a is chromium-based
and the second is dppf-based.
Since the potential for the first wave in 3b-f is
similar to that of 3a , it is likely that the first wave in
3b-f is due to oxidation of the dppf ligand and the
remaining wave(s) are due to oxidation of the group VI
metal. The dppf oxidation of 3b-f is reversible if the
waves at higher potential are not scanned. The group
VI metal oxidations of 3a -c occur at approximately 1
V, which is similar to the potential at which the group
VI metals for 1a -c undergo oxidation. The similar
potentials of 1a and 3a , as well as the simultaneous
oxidation of both chromium centers, imply that the
group VI metal centers in 3a are electrochemically
isolated. However, this apparent electrochemical isola-
tion is not observed in the asymmetric systems, 3d -f.
There appears to be a positive shift in the potential
at which the group VI metal centers are oxidized in
comparing 3d -f to the symmetric compounds 3a -c.
This could be due to the formation of a dication prior to
the final oxidation of 3d -f, whereas in 3a -c, two
simultaneous one-electron oxidations of the monocation
occur.
In this series of compounds, there are two separate
points to consider. The first is the effect of changing the
group VI metal on the potential at which the ferrocene
backbone is oxidized. For a given coordination mode, the
group VI metal seems to have a minimal impact on the
ferrocene-based oxidations. A similar observation was
made in the oxidative electrochemistry of a series of
tungsten and molybdenum compounds with monoden-
tate ferrocenylphosphines (FcPh2P, Fc2PhP, and Fc3P).18
The second consideration is the effect of the coordination
mode on the potential at which oxidation occurs. In
comparison to 1a and 3a , 2a has the fewest carbonyl
ligands and the most phosphorus donors coordinated to
the group VI metal center. Therefore, it is not surprising
that the oxidation of 2a occurs at the least positive
potential for the chromium center in the series. A
The oxidative electrochemistry of 2b has previously
been examined and was reported to be irreversible.6 In
this study the oxidation of 2b was found to be reversible.
The differences in the reversibility are most likely due
to the solvent used in each analysis. The earlier study
used THF, which is a much more coordinating solvent
than CH2Cl2. Mo(CO)6 has a shown tendency to form a
seven-coordinate intermediate; therefore, it is likely that
the irreversibility reported for the oxidation of 2b in
THF is due to the formation of a highly reactive seven-
coordinate intermediate.18 Similar to this study, the
previous study of the oxidation of 2b found only one
oxidation wave.6 Since the potentials in 2b,c are similar
to that of the dppf-based oxidation in 2a , it is likely that
these are dppf-based oxidations. It is unclear why there
are no waves observed for the group VI metal.
Finally, the oxidative electrochemistry of the bridging
species (3) was investigated. The cyclic voltammogram
of compound 3a (Figure 4) exhibits unusual behavior
in this series of compounds, since it displays two
reversible oxidations. The second wave in the cyclic
voltammogram of 3a has a peak current approxi-
mately twice the magnitude of the first wave. Using the
plots of E vs log((Id - I)/I) from normal pulse voltam-
metry, the first wave had a slope of 60 mV (a one-
electron process) and the second wave had a slope of
120 mV (a two-electron process). On the basis of the
oxidative electrochemistry of dppf and the metal hexa-
carbonyl complexes, there are three redox-active metal
centers in 3a . To assign the two waves to the appropri-
ate metal center, 3a was studied using spectroelectro-
chemistry.
A solution of 3a was placed in the electrochemical cell
and the potential set more positive than the first
oxidation wave. The IR spectrum was then obtained,
(22) Ashford, P. K.; Baker, P. K.; Connely, N. G.; Kelly, R. L.;
Woodley, V. A. J . Chem. Soc., Dalton Trans. 1982. 477.
(23) Farrugia, L. J .; J . Appl. Crystallogr. 1997, 30, 565.

4660 Organometallics, Vol. 23, No. 20, 2004
Ohs et al.
similar effect can be seen when comparing Mo(CO)5-
(FcPh2P) and Mo(CO)4(FcPh2P)2, in which the addition
of another ferrocenylphosphine group causes the mo-
lybdenum oxidation to be shifted approximately 0.4 V
less positive.18 The dppf-based oxidations for 3a -f are
significantly different than those observed for 1a -c,
with a shift to more positive potentials of more than 0.1
V. This is due to the presence of two withdrawing metal
centers bound to the dppf ligand.
is bridging two metal centers and least positive when
bound in an η1 coordination mode.
Con clu sion
The electrochemistry of a series of group VI metal
carbonyl compounds containing dppf in a variety of
coordination modes has been investigated. Compounds
1a -c, 2a , and 3a -c exhibit two oxidation waves, while
compounds 2b and 2c display one wave and 3d -f show
three waves. Spectroelectrochemistry and chemical
oxidation methods were employed to assign the waves
to either the ferrocene backbone or the group VI metal.
With the exception of 2a the wave(s) at more positive
potential(s) is assigned to the group VI metal(s).
Trends between electrochemistry, dppf coordination
mode, and the group VI metal center were explored. For
compounds 1a -c and 2a -c, the dppf-based oxidations
occur at approximately the same potentials for the
chromium and molybdenum compounds and slightly
more positive for the tungsten compounds. A similar
trend is noted in the potentials for the dppf-based
oxidation of [(OC)5M(µ-dppf)]2M′Cl2 (M) Cr, Mo, W; M′
) Pd, Pt).7 The potentials for dppf-based oxidation of
3a -c become more positive on going down group VI.
This opposes the trend seen in (OC)5M(µ-dppf)AuCl
(M) Cr, Mo, W),7 but it is unclear why this difference
occurs. The dppf-based oxidations of 3d -f occur at
potentials similar to those for the symmetric compounds
3a -c. A plot of the potentials for the dppf-based
oxidations vs the average position of the group VI metal-
(s) shows that the potential of the asymmetric dimers
is approximately the average of those of the parent
symmetric dimers (Figure 5). For all of the compounds
the dppf-based oxidation is most positive when the dppf
Ack n ow led gm en t. A.C.O. and C.N. wish to thank
the donors of the Petroleum Research Fund, adminis-
tered by the American Chemical Society, for partial
funding, the Academic Research Committee at Lafayette
College for partial funding, and the Kreske Foundation
for the purchase of the J oel Eclipse 400 NMR. M.J .S.
wishes to thank the NSF for support (Grant No.
0213297).
Su p p or tin g In for m a tion Ava ila ble: Full tables of crystal
data, atomic coordinates, thermal parameters, and bond
lengths and angles for 2a as a CIF file and text giving details
of the syntheses of 2a and 3d -f. This material is available
free of charge via the Internet at http://pubs.acs.org.
OM049735T