Electrochemistry of ferrocenylphosphines FcCH2PR2 (Fc=(η5-C5H5)Fe(η5- C5H4); R=Ph, CH2OH and CH2 CH2CN), and some phosphine oxide, phosphine sulfide, phosphonium and metal complex derivatives

Article abstract of DOI:10.1016/S0022-328X(03)00309-7

Electrochemical studies of the free ferrocenylphosphine ligands FcCH₂PR₂ (Fc=(η⁵-C₅ H₅)Fe(η⁵-C₅H₄); R=Ph, CH₂OH and CH₂CH₂CN) and some phosphine oxide, phosphine sulfide, phosphonium and metal derivatives are described. The free ligands exhibit complex voltammetric responses due to participation of the phosphorus lone pair in the redox reactions. Uncomplicated ferrocene-based redox chemistry is observed for P^V derivatives and when the ligands are coordinated in complexes cis-PtCl₂[FcCH₂P(CH₂OH) ₂], PdCl₂[FcCH₂P(CH₂OH) ₂], [Au{FcCH₂P(CH₂OH)₂} ₂]Cl, RuCl₂(η⁶-C₁₀ H₁₄)[FcCH₂P(CH₂OH)₂] and RuCl₂(η⁶-C₁₀H₁₄) (FcCH₂PPh₂). The reaction pathways of the free ligands after one-electron oxidation have been examined in detail using voltammetry, NMR spectroscopy and electrospray mass spectrometry. Direct evidence for formation of a P-P bonded product is presented.

Full text of DOI:10.1016/S0022-328X(03)00309-7

Journal of Organometallic Chemistry 676 (2003) 62Á72
www.elsevier.com/locate/jorganchem
Electrochemistry of ferrocenylphosphines FcCH2PR2 (Fcꢀ
/
(h5-
Ph, CH2OH and CH2CH2CN), and some
C5H5)Fe(h5-C5H4); Rꢀ
phosphine oxide, phosphine sulfide, phosphonium and metal complex
derivatives
/
Alison J. Downard a,*, Nicholas J. Goodwin b, William Henderson b
a Department of Chemistry, University of Canterbury, Private Bag 4800, Christchurch, New Zealand
b Department of Chemistry, University of Waikato, Private Bag 3105, Hamilton, New Zealand
Received 6 March 2003; received in revised form 6 March 2003; accepted 17 April 2003
Abstract
Electrochemical studies of the free ferrocenylphosphine ligands FcCH2PR2 (Fcꢀ
/
(h5-C5H5)Fe(h5-C5H4); Rꢀ
/Ph, CH2OH and
CH2CH2CN) and some phosphine oxide, phosphine sulfide, phosphonium and metal derivatives are described. The free ligands
exhibit complex voltammetric responses due to participation of the phosphorus lone pair in the redox reactions. Uncomplicated
ferrocene-based redox chemistry is observed for PV derivatives and when the ligands are coordinated in complexes cis-
PtCl2[FcCH2P(CH2OH)2], PdCl2[FcCH2P(CH2OH)2], [Au{FcCH2P(CH2OH)2}2]Cl, RuCl2(h6-C10H14)[FcCH2P(CH2OH)2] and
RuCl2(h6-C10H14)(FcCH2PPh2). The reaction pathways of the free ligands after one-electron oxidation have been examined in
detail using voltammetry, NMR spectroscopy and electrospray mass spectrometry. Direct evidence for formation of a Pꢀ
/P bonded
product is presented.
# 2003 Elsevier Science B.V. All rights reserved.
Keywords: Ferrocenyl group; Phosphine; Electrochemistry; NMR spectroscopy; Ferrocenium reaction
1. Introduction
play quite complex redox chemistry which is attributed
to involvement of the PIII lone pair electrons.
One of the best known properties of ferrocenes is their
ability to undergo reversible one-electron oxidation.
Adding a phosphine sustituent to the ferrocene allows
incorporation of the redox-active ferrocenyl moiety into
transition metal complexes, often leading to an increase
in the reactivity of the complex. Coordinated ferroce-
nylphosphine ligands usually exhibit simple, reversible
electrochemistry arising from one-electron oxidation of
the ferrocene center [1]. This well-behaved redox chem-
istry accounts, in part, for their popularity as ligands. In
contrast, free ferrocenylphosphine ligands typically dis-
1,1’-bis(Diphenylphosphino)ferrocene (dppf) is the
most widely used and studied ferrocenylphosphine
ligand. The electrochemistry of the free ligand [2Á
and of a large number of metal complexes of dppf have
been reported [1], however, only Pilloni et al. [2] have
examined the electrochemistry of dppf in detail. Com-
ments have also been made concerning the electrochem-
istry of other ferrocenylphosphine ligands in which the
phosphine moiety is directly linked to a cyclopentadie-
nyl ring; their redox chemistry appears to be
qualitatively similar to that of dppf [8Á . Recently
new synthetic routes to ferrocenylmethylphosphine
ligands [Fe(h5-C5H5){h5-C5H4CH2PR2}] (FcCH2PR2)
[18Á
and FcCH(CH3)PR2 [23] have been developed
* Corresponding author. Tel.: ꢁ
2110.
E-mail
Downard).
/
64-3-364-2501; fax: ꢁ
/
64-3-364-
leading to the synthesis of series of compounds with a
carbon spacer between the cyclopentadienyl ring and the
P atom. The reactivities and complexation chemistry of
address:
a.downard@chem.canterbury.ac.nz (A.J.
0022-328X/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00309-7

A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
63
FcCH2PR2 (Rꢀ
been investigated [18,24,25] and some chemistry of
FcCH(CH3)PR2 (RꢀCH2OH, CH2CH2CN and CH2-
/H, CH2OH, CH2CH2CN and Ph) have
/
(NC4H8O)) has been established [26,27]. However, there
are few reports concerning the redox properties of these
compounds. The oxidation potential of FcCH2P(O)(O-
CH3)OH has been reported [28], and of FcCH2P-
(CH2OH)2, FcCH2P(O)(CH2OH)2 and FcCH2P(S)-
(CH2OH)2 attached to silicon surfaces [29]. In the latter
study each compound exhibited a chemically reversible
oxidation process which was attributed to the ferrocenyl
moiety, however, the ferrocenyl layers were not stable to
repeat redox cycling. Most recently, gold and silver
complexes of FcCH2PPh2 were found to exhibit rever-
sible one-electron electrochemistry based on the ferro-
cenyl moiety, and the redox chemistry of the free ligand
was reported to resemble that of dppf [25].
In this work, we expand the study of the electro-
chemistry of ferrocenylmethylphosphine ligands to in-
clude FcCH2P(CH2OH)2, FcCH2P(CH2CH2CN)2 and
some phosphine oxide, phosphine sulfide and phospho-
nium derivatives. Several palladium, platinum and
ruthenium complexes of the ligands are also examined.
A detailed investigation of the redox chemistry of the
free ligands using bulk electrolysis, 31P NMR and
electrospray mass spectrometry (ESMS) reveals some
similarities and some important differences to the redox
chemistry of dppf.
Fig. 1. Cyclic voltammograms obtained using a Pt disk electrode, scan
100 mV sꢂ1 of (a) 7.5 mM FcCH2PPh2, (b)
1 mM
FcCH2P(CH2OH)2 and (c) 1 mM FcCH2P(CH2CH2CN)2.
rateꢀ
/
negative direction. Cyclic voltammograms of the com-
pounds recorded in dichloromethane solution show the
same features except that the second oxidation peak for
FcCH2P(CH2OH)2 is small and sharp and there is an
associated large and sharp cathodic peak indicative of
adsorbed species. Considering the major oxidation peak,
there is a linear relationship between the anodic peak
current, ipa1, and (scan rate)1/2 over the range 20Á500
mV sꢂ1 indicating a diffusion controlled process.
Comparison of the responses with that of equimolar
FcCH2P(O)(CH2OH)2 under the same conditions (see
below) enables the process to be assigned to the transfer
of one-electron. Based on the observed E1/2 values and
the redox chemistry of other ferrocenylphosphines, the
first oxidation peak is assumed to correspond to
oxidation of the ferrocenyl moiety. E1/2 values are
negative of the FcHꢁ/FcH couple, consistent with the
electron donating effect of the methylene substituent.
For all compounds the chemical reversibility of the
major oxidation process depends on the concentration
and potential scan rate. As shown in Fig. 2 for
FcCH2P(CH2OH)2, the cathodic to anodic peak current
ratio for the first oxidation process decreases with
decrease in scan rate and with increase in concentration
from 0.5Á5 mM. The current for the second oxidation
peak relative to the first increases with decreasing scan
rate and increasing concentration, reaching a maximum
of 0.5 at slow scan rate and high concentration (Fig. 2d).
2. Results
2.1. Cyclic voltammetry of ferrocenylphosphines
FcCH2PPh2, FcCH2P(CH2OH)2 and
FcCH2P(CH2CH2CN)2
Fig. 1 shows cyclic voltammograms of FcCH2PPh2,
FcCH2P(CH2OH)2 and FcCH2P(CH2CH2CN)2 re-
corded at scan rateꢀ
100 mV sꢂ1. Electrochemical
/
data are given in Table 1; unless noted, identical
electrochemical behaviour was observed using Pt and
glassy carbon (GC) working electrodes. 31P NMR data
are also included in Table 1 for convenience.
All ferrocenylphosphines show a major oxidation
process near 0 V vs FcHꢁ/FcH followed by a second
process with lower current at ca. 200 mV higher
potential. Under the experimental conditions of Fig. 1,
the second process is very minor for FcCH2P-
(CH2CH2CN)2 but it is more significant at lower scan
rate or higher concentration. When the potential is
scanned in the negative direction after first traversing
the oxidation steps, one (FcCH2PPh2 and FcCH2P-
(CH2CH2CN)2) or two (FcCH2P(CH2OH)2) broad,
irreversible reduction peaks of relatively low current
are observed at potentials between ꢂ
/
1.3 and ꢂ1.7 V.
/
These peaks are absent in scans commencing in the

64
A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
Table 1
Cyclic voltammetric and 31P NMR data for ferrocenylphosphines, phosphine oxides, sulfides and metal complexes
Chemical shift (d) b
a
Compound
Epa1
Epc1
E1/2(1)
Epa2 Epc2 E1/2(2) Epc
FcCH2PPh2
0
0.03
ꢂ
/
0.07
0.06
0.11
0.03
ꢂ
/
0.04 0.20 0.11 0.16
0.02 0.22 0.14 0.18
0.07 0.16 0.10 0.13
ꢂ
/
1.2 c
1.3 c
ꢂ
/
11.8
22.1
19.3
FcCH2P(CH2CH2CN)2
FcCH2P(CH2OH)2
FcCH2P(O)(CH2OH)2
FcCH2P(S)(CH2OH)2
FcCH2P(O)Ph2
ꢂ
/
ꢂ
/
ꢂ
ꢂ/
/
ꢂ
/
ꢂ
/
0.03
ꢂ
/
ꢂ
/
1.4 c, ꢂ
/
1.7 c
ꢂ
/
0.04
0.05
0.06
0.17
0.15
0.17
ꢂ
/
0
45.8
44.0
ꢂ
/
0.03 0.01
ꢂ
/
0.01 0.02
29.0
[(FcCH2)2P(CH2OH)2]Cl
[FcCH2P(Bz)(CH2OH)2]Br
[FcCH2P(H)Ph2]BF4
0.08
0.08
0.11
0.13
0.12
0.14
ꢂ
/
2.7 d, ꢂ
/
1.5 e
1.6 e
22.30 f
26.0 f
6.31
ꢂ
/
2.7 d, ꢂ
/
ꢂ
/
1.3 c,d, ꢂ
/
1.9 d
ꢂ
/
0.8 e, ꢂ
/
1.3 c,e
cis-PtCl2[FcCH2P(CH2OH)2]
PdCl2[FcCH2P(CH2OH)2]
0.07
0.08
0.09
0.04
ꢂ
0
/
0.02 0.03
0.04
7.8 f
32.4 f
36.2 f
25.9
[Au{FcCH2P(CH2OH)2}2]Cl
RuCl2(h6-C10H14)[FcCH2P(CH2OH)2]
RuCl2(h6-C10H14)(FcCH2PPh2)
0.01
0.03 0.01
0.15 0.12 0.71 0.58 0.64
0.05
ꢂ
/
0.76 0.66 0.71
ꢂ
/
0.09
ꢂ/
ꢂ/
28.8
CH3CNÁ0.1 M [Bu4N]BF4 or [Bu4N]PF6; scan rateꢀ
100 mV sꢂ1; potentials are in V vs FcHꢁ/FcH; except where noted, potentials are
independent of electrode material.
/
a
Broad irreversible peak.
Unless noted otherwise, solvent is CDCl3.
b
c
Only present after scanning through oxidation processes.
GC electrode.
d
e
Pt electrode.
d6-DMSO.
f
FcCH2PPh2:
/
FcCH2P(CH2CH2CN)2B
/
FcCH2P(CH2OH)2.
In order to gain more insight into the reaction
pathway, attempts were made to simulate the cyclic
voltammetry of FcCH2P(CH2OH)2 using the DIGISIM
software package (Bioanalytical Systems) [30]. A variety
of mechanisms were considered and calculated voltam-
mograms were compared with those obtained at 22 8C
using 0.5Á5 mM FcCH2P(CH2OH)2 and scan rates
ranging from 20Á500 mV sꢂ1. Despite extensive efforts,
a good match between experimental and calculated
voltammograms could not be obtained, suggesting a
more complex system than is apparent from casual
inspection of the voltammograms.
2.2. Cyclic voltammetry of ferrocenylphosphine oxides
FcCH2P(O)Ph2 and FcCH2P(O)(CH2OH)2 and sulfide
FcCH2P(S)(CH2OH)2
Fig. 2. Cyclic voltammograms obtained using a Pt disk electrode of
(a), (c) 0.5 mM FcCH2P(CH2OH)2 and (b), (d)
FcCH2P(CH2OH)2. (a), (b) Scan rateꢀ
100 mV sꢂ1; (c), (d) scan
rateꢀ
20 mV sꢂ1
5
mM
/
/
.
The phosphine oxides and sulfide exhibit a simple,
chemically reversible one-electron oxidation process
with E1/2 values slightly positive that of FcHꢁ/FcH
(Table 1). There are no other redox processes within the
solvent limit. The one-electron nature of the redox
change is assigned by comparison of the current with
that for oxidation of equimolar ferrocene. The cyclic
voltammetric DEp values are in the range 65Á74 mV as
found for oxidation of ferrocene under the same
These results are consistent with a mechanism based
on a bimolecular reaction involving the initially formed
ferrocenium to give products which can be reversibly
oxidized at a more positive potential and irreversibly
reduced at quite negative potentials (an ECE mechan-
ism). The relative reaction rates of the oxidized ferroce-
nylphosphines can be deduced from the currents for the
second oxidation peak (at constant concentration and
scan rate) and appear to increase in the order:
conditions (scan rateꢀ
/
100 mV sꢂ1; concentration Â
1
/
mM). Thus, it is assumed that the electron transfer is

A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
65
reversible and that there is a contribution to DEp from
uncompensated solution resistance.
For each of these compounds the electron transfer
step is located on the ferrocenyl moiety. The potentials
of the redox processes are typical for oxidation of
ferrocene derivatives [1,28] and it is well known that
phosphine oxides and sulfides cannot undergo phos-
phine-based oxidations at low potentials [31]. As found
for other ferrocenylphosphine compounds, formation of
a PV center appears to prevent follow-up chemical
reactions of the ferrocenium [8,9,13,14]. The more
positive E1/2 values compared with the parent ferroce-
nylphosphines are attributed to the relative electron
withdrawing properties of PIII and PV.
2.3. Cyclic voltammetry of the ferrocenylphosphonium
salts [FcCH2P(Bz)(CH2OH)2]Br,
[FcCH2P(H)Ph2]BF4 and [(FcCH2)2P(CH2OH)2]Cl
Fig. 3. Cyclic voltammograms obtained using a Pt disk electrode of 1
Cyclic voltammograms of [FcCH2P(Bz)(CH2OH)2]Br
and [FcCH2P(H)Ph2]BF4 show a reversible one-electron
oxidation couple. For [(FcCH2)2P(CH2OH)2]Cl the
peak currents are approximately doubled and there is
a slight broadening of the peaks arising from two very
closely-spaced couples. E1/2 values for the phosphonium
oxidations are between 0.12 and 0.14 V vs FcHꢁ/FcH
(Table 1). For all compounds, the oxidation is assigned
to the ferrocene/ferrocenium couple. As expected, there
are no chemical complications due to involvement of the
PV center but the positive charge of the phosphonium
shifts the E1/2 values are significantly positive compared
with those for the phosphine oxides and sulfide.
Phosphoniums [FcCH2P(Bz)(CH2OH)2]Br and [(Fc-
CH2)2P(CH2OH)2]Cl also exhibit an irreversible reduc-
mM [(FcCH2)2P(CH2OH)2]Cl. (a), (b) GC electrode, scan rateꢀ
/
200
mV sꢂ1 and (c) Pt disk electrode, scan rateꢀ200 mV sꢂ1
/
.
gous to those described for [FcCH2P(Bz)(CH2OH)2]Br
and [(FcCH2)2P(CH2OH)2]Cl. In addition, after scan-
ning through the oxidation peak of [FcCH2P(H)Ph2]BF4
a small irreversible reduction peak with low current
appears at Epcꢀ
/
ꢂ1.3 V. This most likely has the same
/
origin as the irreversible reduction process observed for
FcCH2PPh2 suggesting that oxidation of [FcCH2P-
(H)Ph2]BF4 and FcCH2PPh2 leads to the same pro-
duct(s).
2.4. Cyclic voltammetry of ferrocenylphosphine
complexes cis-PtCl2[FcCH2P(CH2OH)2],
PdCl2[FcCH2P(CH2OH)2],
[Au{FcCH2P(CH2OH)2}2]Cl, RuCl2(h6-
C10H14)[FcCH2P(CH2OH)2] and RuCl2(h6-
C10H14)(FcCH2PPh2)
tion peak at Epcꢀ
V (Pt electrode), scan rateꢀ
/
ꢂ
/
2.7 V (GC electrode) and Â
/
ꢂ1.5
/
/
100 mV sꢂ1. By compar-
ison with the oxidation peak currents, these reductions
appear to involve the transfer of one-electron although
definitive assignment is not possible due to the broad-
ness of the peaks. These processes are assigned to
phosphonium reduction. As shown in Fig. 3, scanning
through the reduction process gives rise to a small peak
The Pt(II), Pd(II) and Au(I) complexes of
FcCH2P(CH2OH)2 have simple, clean electrochemistry
in acetonitrile solution. All complexes exhibit a single
reversible oxidation process with E1/2 slightly positive of
FcHꢁ/FcH (Table 1). As observed for other ferrocenyl-
phosphine metal complexes, involvement of the phos-
phorus lone pair in coordination to the metal shuts
down the chemical complications seen for the free
ligands giving ferrocenyl-based redox processes
[1,10,15,25,34]. The observation of a single redox
process for [Au{FcCH2P(CH2OH)2}2]Cl which has
two ferrocenylphosphine ligands indicates that interac-
tion between the ferrocene groups is very small. The
Ru(II) complexes of FcCH2P(CH2OH)2 and FcCH2-
PPh2 exhibit a reversible one-electron oxidation process
at low potentials followed by a second oxidation process
Â200 mV less positive than the major oxidation
/
process. The new peak is at a similar potential to the
parent ferrocenylphosphine consistent with the expected
reductive cleavage of the phosphoniums to yield phos-
phines [32]. The reason for the strong dependence of the
reduction peak potential on electrode material is not
understood and was not investigated further. However,
it is interesting to note that very similar behaviour was
observed for the reduction of FcCH2NMe2Hꢁ at Au, Pt
and GC electrodes [33]. The reductive electrochemical
response is more complex for [FcCH2P(H)Ph2]BF4.
Broad irreversible reduction peaks are seen at Epcꢀ
1.9 V (GC electrode) and ꢂ0.8 V (Pt electrode),
scan rateꢀ
100 mV sꢂ1. These processes appear analo-
/
ꢂ
/
/
/

66
A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
controlled current method was usually preferred. Most
studies were based on FcCH2PPh2 which showed clean
formation of reaction products compared with more
complex behaviour for FcCH2P(CH2OH)2. Electrolyses
were performed using 1 and 5 mM solutions of the
ferrocenylphosphine; monitoring of the electrolyses by
cyclic voltammetry and steady state voltammetry
showed that the results did not depend on concentration
and hence 5 mM ferrocenylphosphine was used to
facilitate rapid 31P NMR analysis of the electrolysis
solutions. Rapid analysis was important because irre-
versible changes in solution composition were seen over
the course of 1 h. For comparison, bulk electrolyses
were also performed in dichloromethane and DMF
yielding very similar voltammetric and 31P NMR data
to those obtained in acetonitrile solutions.
Controlled current oxidation of FcCH2PPh2 in ace-
tonitrile solution by a charge corresponding to 1 F
molꢂ1 results in a golden solution with a green tint.
Voltammetric changes during the electrolysis are shown
in Fig. 5AaÁd. There is a progressive decrease in the
current due to the main oxidation peak and after
oxidation by one-electron, a broad peak, (labelled (i)
Fig. 4. Cyclic voltammogram obtained using a Pt disk electrode, scan
rateꢀ
200 mV sꢂ1 of 0.9 mM RuCl2(h6-C10H14)(FcCH2PPh2).
/
at higher potentials which is not fully chemically
reversible at scan rates 5
500 mV sꢂ1 (Fig. 4). The
first redox process is assumed to be ferrocene-based, and
/
the second due to oxidation of the Ru(II) center.
2.5. Bulk electrolysis with 31P NMR analysis of the redox
products of FcCH2PPh2 and FcCH2P(CH2OH)2
in Fig. 5Ad) with low current is observed at Â
Concomitantly, there is an increase in current due to a
reversible oxidation process with E1/2 0.15 and an
irreversible reduction with Epcꢀ 1.25 V. These are
/
0 V.
The cyclic voltammetry described above strongly
suggests that the complexity seen in the electrochemistry
of the free ferrocenylphosphines is due to the presence of
the PIII center with its lone pair of electrons. In order to
further probe the redox chemistry, detailed investiga-
tions using bulk electrolysis with voltammetric and 31P
NMR analysis of reaction products were undertaken.
Results for these experiments are given in Table 2.
Controlled potential and controlled current bulk
electrolyses of FcCH2PPh2 and FcCH2P(CH2OH)2
ꢀ
/
/
ꢂ
/
labelled (ii) and (iii), respectively in Fig. 5Ad. Compar-
ison of Fig. 5Ac and d shows that when the potential
scan is initiated in the negative direction first and
traverses peak (iii), the current at (i) is larger than
when the scan is initiated in the positive potential
direction. Evidently a species oxidized at (i) is generated
by reduction process (iii). The potentials of the rever-
sible couples (i) and (ii) suggest ferrocene-based electron
transfers; steady-state voltammetry at a microelectrode
electrode shows that in the oxidized solution, the
ferrocene centers are at least 90% in the neutral (non-
oxidized) form. When the oxidized bulk electrolysis
were performed in acetonitrile ꢂ0.1 M [Bu4N]BF4
/
solutions. When investigating the first oxidation step,
controlled potential electrolyses were undertaken at an
applied potential between the first and second oxidation
process. After passing the same amount of charge,
results were identical for controlled potential and
controlled current electrolyses and hence the more rapid
solution is briefly reduced at ꢂ
/
0.4 V (charge passed is
less than 0.1 F molꢂ1) prior to obtaining the steady
Table 2
Cyclic voltammetric and 31P NMR data for products formed by oxidative bulk electrolysis of FcCH2PPh2 and FcCH2P(CH2OH)2
a
Starting compound
FcCH2PPh2
Epa1
Epc1
E1/2(1)
Epa2
Epc2
E1/2(2)
Epc
Chemical shift (d)
0.03
ꢂ0.03
/
0
0.18
0.11
0.15
ꢂ/
1.25
22.7, 23.6 (Jꢀ
/
118 Hz)
108.2, 109.1 (Jꢀ
/
118 Hz)
FcCH2P(CH2OH)2
0.04
ꢂ
/
0.03
0.01
0.13
0.07
0.10
ꢂ
ꢂ/
/
1.6 b, ꢂ
/
1.8 b
21.9, 22.7 (Jꢀ103 Hz)
24.1
114.8, 115.6 (Jꢀ
100 mV sꢂ1; potentials are in V vs FcHꢁ/FcH; except where noted, potentials are
/
1.4 c, ꢂ1.8 c
/
/
102 Hz)
CH3CNÁ0.1 M [Bu4N]BF4; voltammetry: scan rateꢀ
/
independent of electrode material.
a
Irreversible peak.
GC electrode.
Pt electrode.
b
c

A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
67
Fig. 5. (A) Cyclic voltammograms obtained using a Pt disk electrode, scan rateꢀ
(a) before electrolysis, (b) after oxidative electrolysis with chargeꢀ
0.5 F molꢂ1 and (c), (d) after oxidative electrolysis with chargeꢀ
(B) 31P NMR spectra of 5 mM FcCH2PPh2 in CH3CNÁ0.1 M [Bu4N]BF4 (a) after oxidative electrolysis with chargeꢀ0.5 F molꢂ1 and (b) after
oxidative electrolysis with chargeꢀ 0.4 V prior to obtaining spectra.
1.0 F molꢂ1. The solutions were briefly reduced at ꢂ
/
100 mV sꢂ1 in CH3CNÁ0.1 M [Bu4N]BF4 of 5 mM FcCH2PPh2
1.0 F molꢂ1
/
/
.
/
/
/
state voltammograms, the ferrocene centers are con-
verted fully to the neutral form and the solution loses its
green tint.
contains only the two doublets (Fig. 5Bb). Note that two
doublets of equal intensity are always observed after
bulk electrolysis, regardless of the relative sizes of
processes (i), (ii) and (iii) in steady-state and cyclic
voltammograms.
Despite attempts to exactly reproduce experimental
conditions, the relative peak currents for processes (i)Á
(iii) varied significantly between experiments. The cur-
rents for processes (ii) and (iii) were similar but the
current for couple (i) varied between ca. 10 and 80% that
of (ii) (or (iii)). Regardless of these variations, the total
current due to oxidation of ferrocene centers in the
oxidized solution was approximately half that in the
initial solution, as measured by steady-state voltamme-
try.
Oxidation of the starting compound by 2 F molꢂ1
yields a green solution with the same voltammetric
features described above. Consistent with the solution
colour, steady-state voltammetry confirms that the
ferrocene centers are now in their oxidized form.
Fig. 5B shows the changes in the 31P NMR spectra
which accompany the one-electron oxidative bulk elec-
trolysis. There is a progressive loss of the signal due to
The voltammograms of Fig. 5Ac and d suggest that
the reduction step at ꢂ1.25 V may regenerate starting
/
phosphine. Hence solutions which had been oxidized by
a charge corresponding to 1 F molꢂ1 were reduced
(using controlled current reduction) by a charge corre-
sponding to 1 F molꢂ1. Voltammetry of the final
solutions confirmed formation of a ferrocenylphosphine
or ferrocenylphosphine oxide, albeit at lower concentra-
tion than the initial ferrocenylphosphine. The steady-
state current due to oxidation of ferrocene centers
increased compared to the oxidized solution but it was
significantly less than that in the starting solution. 31P
NMR analysis of the final solutions showed starting
ferrocenylphosphine and
a
variable amount of
FcCH2P(O)Ph2 (dꢀ29.0 ppm). There were no other
/
signals. Based on the 31P NMR signals, the amount of
the phosphine oxide formed during reduction of the
bulk solution varied between experiments from none to
ca. 30% that of regenerated starting compound. These
results did not appear to depend on the relative currents
of redox processes (i), (ii) and (iii) in the oxidized
starting compound (dꢀ
growth of two equal-sized doublets at dꢀ
108.2, 109.1 ppm (for each doublet, Jꢀ
/
ꢂ
/
11.8 ppm vs H3PO4) and the
22.7, 23.6 and
118 Hz). No
/
/
other signals are seen in the spectrum. After oxidation
by a charge corresponding to 1 F molꢂ1, the spectrum

68
A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
solution. Even when no FcCH2P(O)Ph2 was detected in
the final solution, cyclic voltammetry showed that
recovery of the starting material was close to only 50%.
Voltammetric and 31P NMR analysis of the bulk
electrolysis products of FcCH2P(CH2OH)2 reveal more
complex behaviour for this phosphine after one-electron
oxidation (Fig. 6). In addition to oxidation processes at
2.6. Chemical oxidation of FcCH2PPh2 and ESMS and
1H NMR spectroscopy of oxidation products
Electrochemical monitoring of the reaction of
[FcH]BF4 with FcCH2PPh2 indicated that FcHꢁ oxi-
dizes the ferrocenylphosphine leading to the same
products as oxidative bulk electrolysis. However, all
attempts to isolate reaction products after chemical and
electrochemical oxidation were unsuccessful due to the
gradual formation of intractable brown decomposition
E1/2
sible reduction processes are seen with Epcꢀ
1.7 V. 31P NMR analysis of the oxidized solution (Fig.
6c) reveals two doublets ((dꢀ21.9, 22.7 and 114.8, 115.6
ppm, each with Jꢀ102 Hz) and an additional singlet at
dꢀ24.1 ppm. Reduction of the oxidized solution led to
Â
/
ꢂ
/
0.05 and 0.095 V, two broad merged irrever-
/
ꢂ
/
1.5 and
ꢂ
/
1
materials. ESMS and H NMR spectroscopy, coupled
/
with chemical oxidation of FcCH2PPh2, were used to
further characterize the oxidation products. ESMS
spectra (positive ion mode) were obtained of four
samples: FcH, [FcH]BF4, FcCH2PPh2, and FcCH2PPh2
mixed with [FcH]BF4. After accounting for signals
directly arising from [FcH]BF4 and FcH, the spectra
of the latter two samples were very similar. The principal
peaks seen in the spectra of samples containing
/
/
precipitation. After collecting the precipitate and dissol-
ving in DMF, no signals were observed in the 31P NMR
spectrum.
FcCH2PPh2 were: m/z 199.1 [FcCH2]ꢁ, 384.0 [2M]2ꢁ
400.1 [Mꢁ Hꢁ
O]ꢁ, 569.0 [MꢁPPh2]ꢁ, 783.0 [2Mꢂ
O]ꢁ and 855.1 [2MꢁBF4]ꢁ. The peaks at m/z 384.0
,
/
/
/
/
/
and 400.1 dominated the spectra.
1H NMR spectra of acetonitrile solutions of
[FcH]BF4 and FcCH2PPh2 were more complex and
less well-resolved than that of the starting ferrocenyl-
phosphine. The phenyl and ferrocenyl signals were
shifted downfield, as found for phosphine oxide and
phosphonium derivatives. The methylene resonance at
3.2 ppm in FcCH2PPh2 was absent in spectra of the
oxidation products and new singlets appeared at 1.33
and 4.75 ppm. The latter signal could correspond to the
methylene resonance shifted downfield by a neighbour-
ing phosphonium center. 31P NMR analysis of the same
sample showed the expected doublets and several
additional signals which could not be assigned. Hence
further analysis of the 1H NMR spectra was not
attempted.
3. Discussion
Voltammetric and 31P NMR analyses of the solutions
of FcCH2PPh2 after oxidation by bulk electrolysis (Fig.
5) suggest formation of product(s) containing at least
two types of ferrocene centers (giving redox couples (i)
and (ii)) and two types of P centers (giving two doublets
in the 31P NMR). The identical NMR Pꢀ
/
P coupling
constants and the simultaneous formation of both
NMR signals (and their simultaneous loss when the
oxidized solution is reduced) indicate that the two P
environments are in a single molecule. On the other
hand there is no evidence for formation of a product
containing two ferrocene centers. Based on the voltam-
metry of the oxidized solutions (Fig. 5A), and the cyclic
voltammetry of FcCH2P(CH2OH)2 which shows that
Fig. 6. Cyclic voltammograms (a, b) obtained using a Pt disk
electrode, scan rateꢀ
7.5 mM FcCH2P(CH2OH)2 (a) before electrolysis, (b) after oxidative
electrolysis with chargeꢀ
1.0 F molꢂ1. (c) 31P NMR spectrum of the
solution in (b).
/
100 mV sꢂ1 in CH3CNÁ0.1 M [Bu4N]BF4 of
/

A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
69
the peak current for the oxidation of the product reaches
a maximum of half that of the parent compound (Fig.
2d), the product appears to have only one electroactive
ferrocene center. After reduction of the bulk oxidized
solution, voltammetry is consistent with the regenera-
tion of approximately half the initial amount of starting
ferrocenylphosphine while starting material is the only
31P NMR-active compound. Considering these observa-
tions we tentatively propose that one ferrocene center is
cleaved during formation of the product. The cleaved
ferrocenyl moiety may polymerize or undergo other
follow-up reactions giving the broad, low current
process near 0 V (couple (i), Fig. 5Ad). A polymeriza-
tion product would account for the broadness of the
peaks and the lower total current after reaction. ESMS
spectra are consistent with these suggestions showing
products arising from a dimerization reaction. However,
the very different conditions of the conventional bulk
electrolysis and electrospray techniques do not allow
firm parallels to be drawn. The alternative possibility
methylene cation FcCH2ꢁ is a reasonable proposition
based on the known of stability of a-ferrocenyl carboca-
tions [38].
2[FcCH2PPh2]ꢁ0 FcCH2PꢁPh2PPh2 ꢁFcCHꢁ2
FcCHꢁ2 0 follow-up reactions
(1)
(2)
However, apart from observation of FcCH2ꢁ in ESMS
spectra, there is no direct evidence for the involvement
of this species. Likely follow-up reaction products such
as bis(ferrocenyl)ethane and [(FcCH2)2PPh2]ꢁ are not
detected in NMR or ESMS spectra and hence formation
of FcCH2ꢁ cannot be confirmed.
For FcCH2P(CH2OH)2 there is an additional oxida-
tion product to those seen for FcCH2PPh2 and we
propose a ferrocenylphosphonium with an oxidation
potential that closely coincides with that of couple (ii),
an irreversible reduction process close to that of the Pꢀ
/
P
product and an NMR signal at d 24.1 ppm (singlet).
This is presumably a species containing either a single P,
or two identical P centers.
that the Pꢀ/P containing product contains one electro-
active ferrocene center and one which is electroinactive
cannot be discounted. However, we have no suggestions
as to the structure or mechanism of formation of such
an electroinactive ferrocene moiety.
As the only other ferrocenylphosphine to be subjected
to detailed electrochemical investigations, dppf forms an
interesting comparison with the compounds described in
this work. Since publication of Pilloni et al.’s study of
dppf [2], similar electrochemistry has been reported for
various ferrocenylphosphines with P directly bonded to
the cyclopentadienyl ring and similar reaction pathways
have been assumed [8Á . Thus, dppf is an important
benchmark compound. Pilloni et al. proposed that the
redox chemistry of dppf involves an initial one-electron
oxidation of the ferrocenyl moiety followed by an
intramolecular charge transfer from the PIII center. A
bimolecular reaction between two dppfꢁ species leads
to the eventual formation of the phosphonium dppfHꢁ
and phosphine oxide dppfO. These species were identi-
fied by 31P NMR analysis of the oxidative bulk
On the basis of their potentials and behaviour on
reduction, redox couple (ii) and peak (iii) (Fig. 5Ad),
and the corresponding processes in voltammograms of
the starting ferrocenylphosphine (Fig. 1a), can reason-
ably be assigned to a ferrocenylphosphonium product.
Initial formation of (R3P-PR3)ꢁ or (R3P-PR3)2ꢁ after
electrochemical oxidation of PR3 is a well-established
reaction pathway [35,36] and hence we propose a Pꢀ
/P
bonded phosphonium product which incorporates only
one ferrocenyl moiety. An ion with m/z 569.0 in the
ESMS spectra is consistent with formation of
FcCH2PꢁR2PR2. However, consideration of the 31P
NMR spectra of the reaction products suggests that a
simple phosphoniumÁphosphine species is an unlikely
candidate for the final oxidation product. The chemical
shifts are not consistent with a phosphine center and the
coupling constant is within the range expected for PVÁ
PV rather than PVÁPIII coupling [37]. A more reasonable
product on the basis of the coupling constants is
FcCH2PꢁR2P(O)R2 which could form via reaction
with adventitious water. However, while both a phos-
phonium and phosphine oxide center would be expected
to have a chemical shift at the region of the upfield
signal of the product, neither can account for the
downfield signal. Hence we are unable to confidently
assign a structure for the phosphorus center giving rise
to the downfield 31P NMR signals.
electrolysis product solution. A dimer, with two Pꢀ
/
P
bonds, was tentatively proposed as a short-lived inter-
mediate.
For the ferrocenylmethylphosphines of the present
study, initial oxidation of the ferrocene center is also
proposed and subsequent reactions must involve intra-
molecular charge transfer from PIII giving a reduced
(neutral) ferrocene. For these compounds direct evi-
dence for formation of a dimeric species is provided by
31P NMR analysis and is supported by ESMS data.
However, reaction of the dimer to give FcCH2Pꢁ(H)R2
and/or FcCH2P(O)R2 as primary oxidation products is
ruled out by electrochemical and 31P NMR data. The
different fate of the initially formed dimeric species for
the ferrocenylmethylphosphines is tentatively suggested
to be due to the accessibility of FcCH2ꢁ as a relatively
stable cleavage product. Thus, this study lends support
to the interpretation of the electrochemistry of dppf and
Considering the suggestion that the reaction pathway
involves formation of FcCH2PꢁR2PR2, the simplest
reaction that can account for such a species is shown in
Eqs. (1) and (2). Fragmentation to give the ferrocenyl-

70
A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
its analogs and at the same time it has uncovered a new
reaction pathway for ferrocenylphosphine ligands.
4.2. Materials
Solvents used for electrochemistry were analytical
reagent grade and were dried either by passing through
a column of activated neutral alumina or by twice
distilling from CaH2. [Bu4N]PF6 (Aldrich) was recrys-
tallized from ethanol/water and dried under vacuum at
80 8C. [Bu4N]BF4 was prepared as described previously
[39]. FcCH2P(CH2OH)2, FcCH2P(O)(CH2OH)2, Fc-
CH2P(S)(CH2OH)2, FcCH2P(CH2CH2CN)2 and Fc-
CH2PPh2, Pt, Pd, Au and Ru complexes of
FcCH2P(CH2OH)2 and Ru complexes of FcCH2PPh2
were synthesised as described previously [18,20,24].
[FcH]BF4 was prepared by oxidation of ferrocene with
benzoquinone [40].
4. Experimental
4.1. Physical measurements
ESMS was performed in the positive ion mode using a
Micromass LCT mass spectrometer with the probe
maintained at 3200 V and 150 8C. The source was
operated at 80 8C with cone voltageꢀ20 V. Acetonitrile
/
was used to dissolve samples and as the mobile phase.
31P NMR spectra were acquired on a Varian XL-300
spectrometer operating at 121.45 MHz. Phosphorus
chemical shifts are reported relative to external 85%
H3PO4 at d 0.0 ppm. For analysis of non-deuterated
electrolysis solutions, an external lock signal was
provided through use of a D2O insert. Immediately
after acquisition of a sample spectrum, a few transients
were obtained from a sample of D2O containing an 85%
4.3. Preparation of FcCH2P(O)Ph2
To a solution of 100 mg (0.26 mmol) of FcCH2PPh2
in water (5 ml), methanol (35 ml) and dichloromethane
(10 ml) was added 190 mg of 6% H2O2 (0.335 mmol).
After stirring for 1 h solvent was removed under reduced
pressure without heating until product precipitated.
Diethyl ether was added and the organic layer was
1
H3PO4 insert, enabling referencing of the sample. H
and 13C spectra were recorded on a Bruker AC300
1
spectrometer operating at 300.13 Hz for H and 75.47
washed with water (3ꢃ10 ml). During washing, suffi-
/
Hz for 13C.
All electrochemical experiments were performed in a
cient dichloromethane was added to prevent precipita-
tion of the product from the organic layer. Solvent was
removed giving the crude product in quantitative yield
as a yellow powder. Recrystallization from dichloro-
methane/petroleum spirits at 4 8C gave the product as a
brown powder (86 mg, 0.215 mmol, 83%). Elemental
Anal. Found: C, 68.4; H, 5.1%. Calc. for C23H21FeOP:
C, 69.0; H, 5.3%. 31P{1H} NMR (CDCl3): d 29.0 (s). 1H
nitrogen atmosphere at 2292 8C. Unless stated other-
/
wise, measurements were made in acetonitrile solutions
with 0.1 M supporting electrolyte. Identical electroche-
mical and NMR results were obtained using [Bu4N]PF6
and [Bu4N]BF4 electrolytes and these were used inter-
changeably for voltammetric experiments. Cyclic vol-
tammograms were obtained with
a
PAR 173
NMR (CDCl3): d 3.42 (d, FcCH2P, Jꢀ13 Hz, 2H), 4.01
/
potentiostat with Model 273 interface, a Par Model
175 universal programmer and a Graphtec WX1200
recorder. Working electrodes were GC or Pt disks
(s, 4H, C5H4), 4.08 (s, 5H, C5H5), 7.39Á7.70 (m, 10H,
C6H5). 13C{1H} NMR (CDCl3): d 33.38 (d, FcCH2P,
Jꢀ
C5H4-2), 77.77 (d, C5H4-1, Jꢀ
Jꢀ11 Hz), 131.27 (d, C6H5-m, Jꢀ
C6H5-p), 133.06 (s, C6H5-i).
/
67 Hz), 68.02 (s, C5H4-3), 68.90 (s, C5H5), 69.85 (s,
3 Hz), 128.43 (d, C6H5-o,
9 Hz), 131.76 (s,
(areaꢀ
/
7.0 and 0.8 mm2, respectively) and Pt disk
/
microelectrodes (10 mm diameter). Working electrodes
were polished with 1 mm diamond paste prior to
obtaining each voltammogram. The auxiliary electrode
was a Pt wire and a Ag/Agꢁ (0.01 M in CH3CNÁ0.1 M
[Bu4N]PF6) reference electrode was used. Potentials are
reported vs ferrocenium/ferrocene (FcHꢁ/FcH) after
referencing to ferrocene in a separate solution.
Controlled-potential and controlled-current electro-
lyses were performed using a PAR Model 273A
potentiostat/galvanostat and a three compartment cell.
The Pt mesh working and auxiliary electrodes were
separated by a porosity 5 glass frit and the working and
reference electrodes were separated by a Vycor frit. A
nitrogen atmosphere was maintained in the cell and the
solution was stirred throughout the electrolysis. Condi-
tions for controlled current electrolyses were: 1 mM
/
/
4.4. Preparation of [(FcCH2)2P(CH2OH)2]Cl
All solution manipulations were carried out in the
absence of air in an N2 atmosphere. FcCH2OH (0.503 g,
2.33 mmol) was dissolved in 20 ml dry dichloromethane,
oxalyl chloride (0.4 ml, 5 mmol) was added and the
reaction stirred for 90 min. Solvent was removed under
vacuum and the solid redissolved in 20 ml dry dichlor-
omethane. FcCH2P(CH2OH)2 (0.681 g, 2.33 mmol) was
added and the reaction stirred overnight. The product
thus formed was dried under vacuum and methanol (7
ml) added. The slurry was heated to 50 8C, supernatant
removed and diethyl ether (40 ml) was used to pre-
ferrocenylphosphine: Iapp
sphine: Iapp 5 mA.
ꢀ
/
1 mA; 5 mM ferrocenylpho-
cipitate the product. After cooling to ꢂ
product precipitated as a yellow microcrystalline solid
/
30 8C the
ꢀ
/

A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
71
(0.354 g, 0.672 mmol, 28.9%). Elemental Anal. Found:
C, 54.4; H, 5.4%. Calc. for C24H28ClFe2O2P: C, 54.7; H,
References
1
5.3%. 31P{1H} NMR (d6-DMSO): d 22.3 (s). H NMR
[1] A. Togni, T. Hayashi (Eds.), Ferrocenes: Homogeneous Catalysis,
Organic Synthesis, Materials Science, VCH, Weinheim, 1995.
[2] G. Pilloni, B. Longato, B. Corain, J. Organomet. Chem. 420
(1991) 57.
(d6-DMSO): d 3.40 (s, 2H, OH), 3.52 (d, FcCH2P, Jꢀ
/
13 Hz, 4H), 4.20 (d, PCH2O, Jꢀ6 Hz, 4H), 4.26 (s,
/
10H, C5H5), 4.27 (unres. t, 4H, C5H4), 4.39 (unres. t,
4H, C5H4). 13C{1H} NMR (d6-DMSO): d 21.37 (d,
[3] J.C. Kotz, C.L. Nivert, J. Organomet. Chem. 52 (1973)
387.
[4] T.M. Miller, K.J. Ahmed, M.S. Wrighton, Inorg. Chem. 28 (1989)
2347.
FcCH2P, Jꢀ
72.41 (s, C5H4-3), 72.95 (s, C5H5), 73.46 (s, C5H4-2),
78.68 (s, C5H4-1).
/
37 Hz), 54.49 (d, PCH2O, Jꢀ53 Hz),
/
[5] C.E. Housecroft, S.M. Owen, P.R. Raithby, B.A.M. Shaykh,
Organometallics 9 (1990) 1617.
[6] B. Corain, B. Longato, G. Favero, D. Ajo, G. Pilloni, U. Russo,
F.R. Kreissl, Inorg. Chim. Acta 157 (1989) 259.
[7] D.L. DuBois, C.W. Eigenbrot, Jr., A. Miedaner, J.C. Smart, R.C.
Haltiwanger, Organometallics 5 (1986) 1405.
4.5. Preparation of [FcCH2P(H)Ph2]BF4
All solution manipulations were carried out in the
absence of air in an N2 atmosphere. To a solution of 100
mg FcCH2PPh2 (0.26 mmol) dissolved in 5 ml diethyl
ether, a solution of HBF4 in acetonitrile (5.6 ml, 0.27
mmol) was added dropwise with stirring. Petroleum
spirits (20 ml) and a small amount of diethyl ether were
added, causing the product to separate as an oil. The
supernatant was removed and the oil washed with
[8] J. Podlaha, P. Stepnicka, J. Ludvik, I. Cisarova, Organometallics
15 (1996) 543.
[9] P. Stepnicka, I. Cisarova, J. Podlaha, J. Ludvik, M. Nejezchleba,
J. Organomet. Chem. 582 (1999) 319.
[10] I.R. Butler, M. Kalaji, L. Nehrlich, M. Hursthouse, A.I.
Karaulov, K.M.A. Malik, J. Chem. Soc. Chem. Commun.
(1995) 459.
[11] P. Zanello, A. Cinquantini, M. Fontani, M. Giardiello, G. Giorgi,
C.R. Landis, B.F.M. Kimmich, J. Organomet. Chem. 637Á639
(2001) 800.
petroleum spirits (3ꢃ
for 2 days, giving the product as a yellow oil (0.113 g,
0.239 mmol, 92%). 31P NMR (CDCl3): d 6.31 (d, 1J(Pꢀ
H)ꢀ
511 Hz). 1H NMR (CDCl3): d 4.57Á4.70 (m, 11 H,
FcCH2P, C5H5 and C5H4), 7.99Á8.27 (m, 10H, C6H5).
13C{1H} NMR (CDCl3): d 23.82 (d, FcCH2P, Jꢀ
43
Hz), 69.55 s, C5H4-3), 69.89 (s, C5H5), 70.03 (s, C5H4-2),
78.15 (d, C5H4-1, Jꢀ611 Hz), 115.43 (d C6H5-i, Jꢀ82
Hz), 130.20 (d, C6H5-o, Jꢀ13 Hz), 133.67 (d, C6H5-m,
Jꢀ10 Hz), 135.14 (s, C6H5-p).
/5 ml), then dried under vacuum
[12] P. Zanello, G. Opromolla, G. Giorgi, G. Sasso, A. Togni, J.
Organomet. Chem. 506 (1996) 61.
/
[13] A. Gref, P. Diter, D. Guillaneux, H.B. Kagan, New J. Chem. 21
(1997) 1353.
/
[14] D.A. Durfey, R.U. Kirss, M.R. Churchill, K.M. Keil, W.
Feighery, Synth. React. Inorg. Met. Org. Chem. 32 (2002) 97.
[15] A. Masson-Szymczak, O. Riant, A. Gref, H.B. Kagan, J.
Organomet. Chem. 511 (1996) 193.
/
/
/
[16] R. Horikoshi, T. Mochida, R. Torigoe, Y. Yamamoto, Eur. J.
Inorg. Chem. (2002) 3197.
/
/
[17] P. Stepnicka, R. Gyepes, O. Lavastre, P.H. Dixneuf, Organome-
tallics 16 (1997) 5089.
[18] N.J. Goodwin, W. Henderson, B.K. Nicholson, Inorg. Chim.
Acta 295 (1999) 18.
4.6. Preparation of [FcCH2P(Bz)(CH2OH)2]Br
[19] N.J. Goodwin, W. Henderson, B.K. Nicholson, J. Fawcett, D.R.
Russell, J. Chem. Soc. Dalton Trans. (1999) 1785.
[20] N.J. Goodwin, W. Henderson, B.K. Nicholson, J.K. Sarfo, J.
Fawcett, D.R. Russell, J. Chem. Soc. Dalton Trans. (1997) 4377.
[21] N.J. Goodwin, W. Henderson, B.K. Nicholson, J. Chem. Soc.
Chem. Commun. (1997) 31.
A solution of 100 mg of FcCH2P(CH2OH)2 dissolved
in methanol (10 ml) was heated to 45 8C and benzyl
bromide (0.20 ml, 1.7 mmol) was added dropwise with
stirring. Solvent was removed under reduced pressure
leaving a yellow oil which gave the crude product as a
yellow powder after treatment with methanol/diethyl
[22] N.J. Goodwin, W. Henderson, J.K. Sarfo, J. Chem. Soc. Chem.
Commun. (1996) 1551.
ether at ꢂ
NMR (d6-DMSO): d 26.0 (s). 13C{1H} NMR (d6-
DMSO): d 19.41 (d, PCH2Bz, Jꢀ38 Hz), 23.93 (d,
FcCH2P, Jꢀ38 Hz), 52.37 (d, PCH2O, Jꢀ55 Hz),
/
30 8C (0.075 g, 0.165 mmol, 49%). 31P{1H}
[23] T.V.V. Ramakrishna, A.J. Elias, A. Vij, J. Organomet. Chem. 602
(2000) 125.
[24] N.J. Goodwin, W. Henderson, Polyhedron 17 (1998) 4071.
[25] E.M. Barranco, O. Crespo, M.C. Gimeno, A. Laguna, P.G.
Jones, B. Ahrens, Inorg. Chem. 39 (2000) 680.
[26] T.V.V. Ramakrishna, A.J. Elias, J. Organomet. Chem. 637Á639
(2001) 382.
/
/
/
70.32 (s, C5H4-3), 70.81 (s, C5H5), 71.23 (s, C5H4-2),
78.31 (s, C5H4-1), 129.75 (s, C6H5-p), 130.07 (d, C6H5-i,
[27] K. Muralidharan, N.D. Reddy, A.J. Elias, Inorg. Chem. 39 (2000)
3988.
Jꢀ
Hz).
/
8 Hz), 130.87 (s, C6H5-m), 139.91 (d, C6H5-o, Jꢀ5
/
[28] M.E.N.P.R.A. Silva, A.J.L. Pombeiro, J.J.R. Frausto da Silva, R.
Hermann, N. Deus, T.J. Castilho, M.F.C.G. Silva, J. Organomet.
Chem. 421 (1991) 75.
[29] R.D. Eagling, J.E. Bateman, N.J. Goodwin, W. Henderson, B.R.
Horrocks, J. Chem. Soc. Dalton Trans. (1998) 1273.
[30] M. Rudolph, D.P. Reddy, S.W. Feldberg, Anal. Chem. 66 (1994)
589A.
Acknowledgements
We thank Rewi Thompson and Bruce Clark for
NMR and ESMS measurements, respectively.
[31] H. Maeda, H. Ohmori, Acc. Chem. Res. 32 (1999) 72.

72
A.J. Downard et al. / Journal of Organometallic Chemistry 676 (2003) 62Á72
[32] K.S.V. Santhanam, Chem. Organophosphorus Compd. 1 (1990)
103.
[36] E.V. Nikitin, A.S. Romakhin, V.A. Zagumennov, Y.A. Babkin,
Electrochim. Acta 42 (1997) 2217.
[33] R.F. Winter, G. Wolmershauser, J. Organomet. Chem. 570 (1998)
201.
[37] K.R. Dixon, in: J. Mason (Ed.), Multinuclear NMR, Plenum
Press, New York, 1987, p. 369.
[34] J.E. Kingston, L. Ashford, P.D. Beer, M.G.B. Drew, J. Chem.
Soc. Dalton Trans. (1999) 251.
[38] J.J. Dannenberg, M.K. Levenberg, J.H. Richards, Tetrahedron 29
(1973) 1575.
[35] W.B. Gara, B.P. Roberts, J. Chem. Soc. Perkin Trans. 2 (1978)
150.
[39] A.J. Downard, Langmuir 16 (2000) 9680.
[40] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.