Synthesis and electrochemistry of platinum complexes of hydroquinon-2-ylmethyl- and p-benzoquinon-2-ylmethyldiphenylphosphine

Article abstract of DOI:10.1039/a900610i

Platinum(II) complexes of new (hydroquinon-2-ylmethyl)diphenylphosphine (PPh₂thqH₂) and diphenyl(quinon-2-ylmethyl)phosphine (PPh₂tq) ligands have been studied. Reaction of (2,5-dimethoxybenzyl)diphenylphosphine (PPh₂dmb) with [PtCl₂(PhCN)₂] afforded [PtCl₂(PPh₂dmb)₂] 1a. Metathesis reactions gave the bromo (1b) and iodo (1c) congeners. Deprotection of the hydroquinone groups in 1a using boron tribromide followed by treatment with base produced the O,P-chelated hydroquinonate phosphine complex, cis-[Pt(O,P-PPh₂thqH)₂] 2. Reaction of 2 with hydrobromic acid afforded cis-[PtBr₂(PPh₂thqH)₂] 3 which can be oxidised to give the quinone phosphine complex cis-[PtBr₂(PPh₂tq)₂] 4. Unlike previously reported quinone phosphine complexes, 4 is robust and stable to hydrolysis; its reduction with excess of zinc and dilute hydrobromic acid produced cis-[Pt(O,P-PPh₂thqH)₂(ZnBr₂)] 5. Crystal structure analyses of 2·2dmf, 4·0.5dcm and 5·2dmf were performed. The electrochemistries of 1b, 3, and 4 have been characterised by cyclic voltammetry and controlled potential electrolyses. Cyclic voltammograms of 4 in the presence of dilute hydrobromic acid exhibit a four-electron cathodic process, attributed to reduction to 3; those of 3 show an anodic process attributed to oxidation to 4. The electrochemistry of 4 under aprotic conditions is extraordinary. Although there are two well separated, pendant quinone substituents only a single one-electron reduction process is observed. The reduction affords a radical species (6^?-) which has been characterised by cyclic voltammetry, and by EPR and UV/Vis/NIR spectroscopy. It is argued from the available data that 6^?- is a novel platinum(IV) complex with bound hydroquinonate and semiquinonate (sq) groups, namely [PtBr₂(O,P-PPh₂thq)(O,P-PPh₂tsq ^?)]^-, and a possible mechanism for its remarkable formation is discussed.

Full text of DOI:10.1039/a900610i

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and electrochemistry of platinum complexes of
hydroquinon-2-ylmethyl- and p-benzoquinon-2-ylmethyl-
diphenylphosphine
Seri Bima Sembiring,a Stephen B. Colbran*b and Donald C. Craigb
a Department of Chemistry, Faculty of Mathematics and Natural Science,
University of North Sumatra, Medan, Indonesia
b School of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia.
E-mail: s.colbran@unsw.edu.au
Received 22nd January 1999, Accepted 22nd March 1999
Platinum() complexes of new (hydroquinon-2-ylmethyl)diphenylphosphine (PPh2thqH2) and diphenyl(quinon-2-
ylmethyl)phosphine (PPh2tq) ligands have been studied. Reaction of (2,5-dimethoxybenzyl)diphenylphosphine
(PPh2dmb) with [PtCl2(PhCN)2] afforded [PtCl2(PPh2dmb)2] 1a. Metathesis reactions gave the bromo (1b) and iodo
(1c) congeners. Deprotection of the hydroquinone groups in 1a using boron tribromide followed by treatment with
base produced the O,P-chelated hydroquinonate phosphine complex, cis-[Pt(O,P-PPh2thqH)2] 2. Reaction of 2 with
hydrobromic acid afforded cis-[PtBr2(PPh2thqH)2] 3 which can be oxidised to give the quinone phosphine complex
cis-[PtBr2(PPh2tq)2] 4. Unlike previously reported quinone phosphine complexes, 4 is robust and stable to hydrolysis;
its reduction with excess of zinc and dilute hydrobromic acid produced cis-[Pt(O,P-PPh2thqH)2(ZnBr2)] 5. Crystal
structure analyses of 2ؒ2dmf, 4ؒ0.5dcm and 5ؒ2dmf were performed. The electrochemistries of 1b, 3, and 4 have been
characterised by cyclic voltammetry and controlled potential electrolyses. Cyclic voltammograms of 4 in the presence
of dilute hydrobromic acid exhibit a four-electron cathodic process, attributed to reduction to 3; those of 3 show an
anodic process attributed to oxidation to 4. The electrochemistry of 4 under aprotic conditions is extraordinary.
Although there are two well separated, pendant quinone substituents only a single one-electron reduction process is
observed. The reduction affords a radical species (6 Ϫ) which has been characterised by cyclic voltammetry, and by
᝽
EPR and UV/Vis/NIR spectroscopy. It is argued from the available data that 6 Ϫ is a novel platinum() complex with
᝽
Ϫ
ؒ
bound hydroquinonate and semiquinonate (sq) groups, namely [PtBr2(O,P-PPh2thq)(O,P-PPh2tsq )] , and a possible
mechanism for its remarkable formation is discussed.
This paper describes the preparation and electrochemistry of
some transition metal (specifically platinum) complexes of
novel phosphine ligands with p-hydroquinone or p-quinone
substituents. p-Quinones along with their redox products,
p-semiquinones and p-hydroquinones, comprise perhaps the
quintessential organic electron and hydrogen transfer systems.1
For example, electron transfer reactions between transition
metal centres and p-quinone cofactors are vital for all life,
occurring in key biological processes as diverse as the oxid-
ative maintenance of biological amine levels,2,3 tissue (col-
lagen and elastin) formation,3,4 photosynthesis5,6 and aerobic
(mitochondrial) respiration.6,7 Nevertheless, in comparison to
the extensive co-ordination chemistry for chelate-stabilised,
o-(hydro/semi)quinone σ-donor ligands,8,9 there are relatively
few transition metal σ-complexes of p-(hydro/semi)quinone
ligands,10 thus giving some incentive for this study. However, the
main impetus was provided by the observations that small
amounts of a p-quinone promoter can prolong the efficacy and
lifetime of certain homogeneous transition metal catalysts.
Examples include Shell’s olefin-oligomerisation process11 and
the palladium-catalysed alternating copolymerisation of alkenes
and carbon monoxide to afford polyketones with unusual,
useful properties of potential commercial value.12 In both
processes, phosphine ligands are used to stabilise the catalytic-
ally active transition metal centres. The exact role(s) of the
p-quinone promoter has not been established, but possibilities
for its action include as a hydride acceptor, as an oxidant of
“dead-end” lower oxidation-state species to regenerate active
catalyst, and as oxidant promoting more of the catalytically
active species. Recent results revealing that p-quinones can be
used as hydrogen acceptors in cycles for transition metal-
catalysed oxidations of organic substrates provide further
incentive for this work.13 Moreover, p-quinones are commonly
used in organic synthesis as organic dehydrogenation/oxidation
reagents.1,14
Taken together the above results suggest potential uses for
transition metal complexes of quinone phosphine ligands,
including, for example, in homogeneous oxidations of organic
substrates. In response, we instigated studies of the palladium
and platinum co-ordination chemistry of the first phosphine
ligands with p-hydroquinone or p-quinone substituents. Some
early results have been communicated.15–19 Under rigorously
aprotic conditions, oxidation of the hydroquinone phosphine
complexes [MX2(PPh2hqH2)2] (M = Pd or Pt; PPh2hqH2 =
2-diphenylphosphino-1,4-hydroquinone; X = Cl, Br or I)
produced the quinone phosphine complexes [MX2(PPh2q)2]
(PPh2q = 2-diphenylphosphino-1,4-benzoquinone). The com-
plexes are electrochemically active; for example, four-electron,
two-proton reduction of the quinone phosphine complexes
[MX2(PPh2q)2] afforded the chelated hydroquinonate phos-
phine complexes cis-[M(O,P-PPh2hqH)2].15,20 However, our
previously reported quinone phosphine complexes are suscep-
tible to hydrolytic loss of the quinone group, probably because
of the direct attachment of the quinone groups to the phos-
phorus atoms.16,19,20 This is a major limitation to the further
development and use of these systems. For example, chemically
reversible interconversion between the hydroquinone/quinone
redox states is expected in buffered acidic aqueous solutions
but can not be observed because the quinone phosphine lig-
ands decompose.20 A more (hydrolytically) stable quinone
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1543

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for cis-[Pt(O,P-PPh2-
thqH)2]ؒ2dmf (2ؒ2dmf) with estimated standard deviations (e.s.d.s.) in
parentheses
phosphine ligand system was therefore sought. We thought that
an intervening methylene group would isolate the phosphorus
and redox-active hydroquinone/quinone centres leading to
increased stability for the quinone phosphine complexes. To
this end, we now report a study of the platinum co-ordin-
ation chemistry of new p-hydroquinone-/p-quinone phosphine
ligands.
Pt–PA
Pt–O1A
Pt–PB
Pt–O1B
PA–C7A
PA–C8A
PA–C14A
2.229(1)
2.044(3)
2.232(1)
2.030(3)
1.830(4)
1.815(3)
1.817(3)
O1A–C1A
C1B–C2B
PB–C7B
1.340(5)
1.399(6)
1.830(4)
1.815(3)
1.818(3)
1.334(5)
1.379(5)
PB–C8B
PB–C14B
O1B–C1B
O2B–C4B
OMe
OH
O
Ph2P
Ph2P
Ph2P
3
4
3
4
3
4
PA–Pt–O1A
PA–Pt–PB
PA–Pt–O1B
89.5(1)
101.0(1)
168.7(1)
O1A–Pt–PB
O1A–Pt–O1B
PB–Pt–O1B
168.6(1)
79.7(1)
90.0(1)
6
6
6
OMe
OH
O
PPh2dmb
PPh2thqH2
PPh2tq
ments of the other peaks were made on the basis of their
relative intensities compared to those of the methylene reson-
ances. The relative amounts of cis and trans isomers deduced
from 31P-{1H} and 1H NMR spectra were consistent in all cases.
Treatment of complex 1a with BBr3 followed by methanolic
sodium carbonate gave a pale yellow powder which analyses as
cis-[Pt(O,P-PPh2thqH)2]ؒ1.5H2O (2ؒ1.5H2O), Scheme 1. The
complex is poorly soluble and dimethylformamide (dmf) was
the only solvent found in which it readily dissolved. Although it
initially dissolved in dimethyl sulfoxide, within a few seconds a
yellow fibrous powder began to precipitate. Elemental analysis
of this is in accord with the formulation 2ؒ2H2Oؒ2(CH3)2SO
suggesting that 2 forms a dimethyl sulfoxide solvate hydrate
that, unusually, is insoluble in dimethyl sulfoxide. Formation of
the hydrate provides a convenient method for purification of 2.
The electrospray mass spectrum of 2 shows a prominent peak at
m/z 810 for the [M ϩ H]ϩ ion. A broad hydroxy band at
3283 cmϪ1 is seen in the IR spectrum. The NMR spectra are
unexceptional and confirm that 2 has cis stereochemistry: for
example, the 1H NMR spectrum reveals a “filled in” cis methyl-
ene doublet at δ 4.11 and the 31P-{1H} NMR spectrum exhibits
a singlet at δ 28.08 flanked by a satellite doublet with a cis
31P–195Pt coupling constant of 3822 Hz. Although not tested, it
is likely that the initial product of the reaction of 1a with BBr3
followed by methanol quenching of the borate intermediates
is trans-[PtBr2(PPh2thqH2)2], and that reaction of this with
sodium carbonate leads to closure of the O,P-chelate rings to
afford 2ؒnH2O.
A single crystal analysis of complex 2ؒ2dmf was undertaken
on pale yellow crystals grown from diethyl ether saturated–dmf
solution. The molecular structure and labelling scheme is
shown in Fig. 1. The platinum() ion is planar and cis-co-
ordinated by two O,P-chelate hydroquinonate phosphine lig-
ands. In the crystal structure, a molecule of dmf hydrogen
bonds to the terminal hydroquinonate OH group of each ligand
[O (dmf) ؒ ؒ ؒ O (hydroquinonate) 2.62, 2.70 Å]. The significant
distortion from normal square planar geometry around the
platinum ion [P–Pt–P 101.0(1) and O–Pt–O = 79.7(1)Њ] arises
because of the constrained ligand “bite” [(P–Pt–O)average
89.8(1)Њ]. Other key bond lengths and angles (Table 1) are
unexceptional and similar to those found in other platinum()
complexes with a cis-O2P2 donor set.24
Chart 1
Results and discussion
Synthetic studies
Ligands. Complexes of the ligands (hydroquinone-2-yl-
methyl)diphenylphosphine (PPh2thqH2) and (p-benzoquinone-
2-ylmethyl)diphenylphosphine (PPh2tq, Chart 1) were targeted
for this study. The entry point to these was provided by the
protected precursor (2,5-dimethoxybenzyl)diphenylphosphine
(PPh2dmb, Chart 1) which was prepared as follows. 2,5-Dimeth-
oxybenzylbromide, obtained by methylation of 2,5-dihydroxy-
toluene with iodomethane–potassium carbonate followed by
bromination of the intermediary 2,5-dimethoxytoluene with N-
bromosuccinamide, was treated with magnesium filings that had
been precrushed dry under nitrogen [precrushing is essential to
avoid the Wurtz coupled product21] to give the 2,5-dimethoxy-
benzyl Grignard reagent which was treated with chlorodiphe-
nylphosphine at Ϫ80 ЊC to afford PPh2dmb in excellent yield
(88%). No attempt was made to obtain the free hydroquinone
and quinone phosphines PPh2thqH2 and PPh2tq respectively,
because we anticipated that complexes of hydroquinone/
quinone phosphines would be best prepared by first forming
complexes of phosphines with a protected hydroquinone/
quinone substituent (PPh2dmb in the present case) and only
then deprotecting the hydroquinone/quinone substituents(s).
Platinum(II) complexes. Platinum() complexes of PPh2dmb
were easily prepared. Reactions of two equivalents of PPh2dmb
with [PtCl2(PhCN)2] in dichloromethane (dcm) gave good
yields of the phosphine complex [PtCl2(PPh2dmb)2] 1a an off-
white solid, 95%, eqn. (1). Metathesis reactions of 1a in acetone
[PtCl2(PhCN)2] ϩ 2 PPh2dmb →
[PtCl2(PPh2dmb)2] ϩ 2 PhCN (1)
with a large excess of potassium bromide or potassium iodide
afforded the bromo (1b) and iodo (1c) analogues in near quanti-
tative yields, eqn. (2).
[PtCl2(PPh2dmb)2] ϩ 2 KX →
[PtX2(PPh2dmb)2] ϩ 2 KCl (2)
Reaction of complex 2 with hydrobromic acid gave cis-
[PtBr2(PPh2thqH2)2] 3 in good isolated yield (>80%), Scheme 1.
Analytical and spectroscopic data for 3 accord with its form-
ulation: for example, IR spectra show a broad hydroxy band at
3272 cmϪ1, the 31P{-1H} NMR spectrum exhibits a singlet at
δ 12.18 flanked by a satellite doublet with a cis 31P–195Pt coup-
ling constant of 3720 Hz, and the 1H NMR spectrum reveals a
cis methylene doublet at δ 4.30 with flanking satellites due to
195Pt coupling. Treating a suspension of 3 in methanol with
excess of sodium carbonate gave back 2 in near quantitative
yield, showing that 2 and 3 can be cleanly interconverted,
Scheme 1. No evidence was found for the trans analogues of 2
and 3 in these reactions. The results reveal that binding of the
The elemental analyses of complexes 1a–1c were all consist-
1
ent with their respective formulations. The 31P-{1H} and H
NMR spectra reveal that they were isolated as mixtures of trans
and cis stereoisomers. The peaks for the cis and trans isomers
in the 31P-{1H} NMR spectra were directly assigned from the
31P–195Pt coupling constants {for [PtX2(PR3)2] complexes,
1J(PPt) typically is less than 2800 Hz for trans isomers and
greater than 3000 Hz for cis isomers}.22–24 Notably, in 1H NMR
spectra, the α-methylene resonances of the trans isomers [where
J(PP) is large (>500 Hz)] appear as “virtual” 1:2:1 triplets
whereas “filled in” 1:1 doublets were observed for the cis
isomers [where J(PP) is typically small (<80 Hz)].25 Assign-
1544
J. Chem. Soc., Dalton Trans., 1999, 1543–1554

View Article Online
hydroquinone groups and O,P-chelate formation can be pH-
controlled, and that the cis stereochemistry of the platinum ion
is unchanged during these reversible transformations. This
behaviour is much the same as found for transition metal com-
plexes of other potentially O,P-chelate, alcohol- or phenol-
substituted phosphine ligands.23,24
strong peak at m/z 887 for the [M Ϫ Br]ϩ ion in the electrospray
mass spectrum, a strong quinone C᎐O band at 1657 cmϪ1 in the
᎐
IR spectrum, a singlet at δ 10.23 flanked by a satellite doublet
with a cis 31P–195Pt coupling constant of 3685 Hz in the 31P-
{1H} NMR spectrum and a methylene doublet at δ 4.07
1
(flanked by satellite peaks due to 195Pt–H coupling) in the H
Phosphines undergo Michael addition reactions with quin-
ones, making these groups incompatible in an uncomplexed,
“free” ligand.27 Complex formation prevents these reactions by
“tying up” the phosphorus lone pair, and there should be no
impediment to preparation of a complex with quinone phos-
phine ligands. In accord with these expectations, the p-quinone
phosphine complex [PtBr2(PPh2tq)2] 4 was obtained by oxid-
ation of 3 with two equivalents of strongly oxidising 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in 97% yield,
Scheme 1. Partial elemental analytical data for 4 are consistent
with its formulation. Notable spectroscopic data include the
NMR spectrum, again indicative of cis stereochemistry. In
order to test the stability of 4 to hydrolysis, a d6-acetone solu-
tion of the complex was treated with a few drops of 0.1 M
1
hydrobromic acid and the fate of 4 monitored by H and 31P-
{1H} NMR spectroscopy. The solution was stable. No decom-
position or other reaction of 4 was observed.
An X-ray analysis of light brown crystals of complex
4ؒ0.5CH2Cl2 obtained from dcm solution saturated with vapour
from a 1:1 mixture of diethyl ether–pentane confirms that the
platinum() ion is co-ordinated by two cisoid toluquinone
phosphine and two bromo ligands, Fig. 2. The distorted square
planar geometry about the platinum ion [Br–Pt–Br 86.6(1);
P–Pt–P 98.8(1); Br–Pt–P 85.9, 89.0(1)Њ] is unremarkable for cis-
bis(phosphine)platinum() complexes, as are the metal–ligand
bond lengths (see Table 2).23,24 The quinone substituents lie on
opposite sides of the platinum co-ordination plane and display
the expected quinonoid C–C, C᎐C and C᎐O distances1 (Table
᎐
᎐
2); they are unequivocally quinones.
Quinones can be reduced by zinc under mildly acidic condi-
tions to the corresponding hydroquinones.1 The oxidation of
complex 3 (by DDQ) to 4 has already been described. In order
to establish the chemical reversibility of the 3/4 couple, we
treated a solution of 4 in acetone with excess of zinc powder
and 0.1 M hydrobromic acid. After 2 h the excess of zinc was
removed by filtration and the product precipitated with water.
Recrystallisation from dmf–diethyl ether afforded bimetallic
cis-[Pt(O,P-PPh2thqH)2(ZnBr2)]ؒ2dmf (5ؒ2dmf); the dimer
formally is the adduct of the Lewis acid ZnBr2 and 2 (Scheme
1). Only a small sample of 5 was prepared and a partial elem-
ental analysis for C, H was not obtained. However, ICP analysis
reveals a 2:1:1 ratio for P:Pt:Zn, and the electrospray mass
Fig. 1 An ORTEP26 plot of cis-[Pt(O,P-PPh2thqH)2 2. The thermal
ellipsoids are drawn at the 10% probability level.
OMe
Cl
Ph2P
Pt
OMe
P
Cl
OMe
Ph2
(i) BBr3, CH2Cl2
(ii) Na2CO3, MeOH
1a
OMe
Br
Br
Ph2
P
Ph2
P
OH
OH
OH
Pt
Pt
P
P
HBr(aq), acetone
Na2CO3, MeOH
Ph2
Ph2
HO
O
O
OH
OH
2
3
HBr(aq), acetone
DDQ, acetone
Br
Br
Ph2
P
Ph2
P
O
O
O
O
Pt
Pt
Zn
5
Zn, HBr(aq), acetone
P
P
HO
O
O
OH
Ph2
Ph2
Br
Br
4
Scheme 1
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1545

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for cis-[PtBr2-
(PPh2tq)2]ؒ0.5dcm (4ؒ0.5dcm) with e.s.d.s. in parentheses
Table 3 Selected bond lengths (Å) and angles (Њ) for cis-[Pt(O,P-
PPH2thqH)2(ZnBr2)]ؒ2dmf (5ؒ2dmf) with e.s.d.s. in parentheses
Pt–Br1
Pt–Br2
PA–C1A
PA–C8A
PA–C14A
O1A–C3A
O2A–C6A
C1A–C2A
C2A–C3A
C2A–C7A
C3A–C4A
C4A–C5A
C5A–C6A
C6A–C7A
2.477(1)
2.469(1)
1.855(7)
1.833(5)
1.835(4)
1.204(8)
1.211(8)
1.497(9)
1.497(10)
1.338(8)
1.477(10)
1.335(10)
1.448(11)
1.468(10)
Pt–PA
Pt–PB
2.255(2)
2.254(2)
1.840(7)
1.824(5)
1.815(4)
1.219(8)
1.213(8)
1.504(9)
1.490(9)
1.317(9)
1.451(11)
1.335(10)
1.475(10)
1.481(10)
Pt–PA
Pt–O1A
Pt–PB
Pt–O1B
Zn–Br1
Zn–Br2
Zn–O1A
Zn–O1B
2.217(3)
2.080(5)
2.219(2)
2.060(6)
2.349(2)
2.341(2)
2.014(6)
2.011(6)
PA–C7A
PA–C8A
PA–C14A
O1A–C1A
PB–C7B
PB–C8B
PB–C14B
O1B–C1B
1.829(9)
1.807(6)
1.808(6)
1.362(10)
1.800(9)
1.803(6)
1.819(6)
1.382(10)
PB–C1B
PB–C8B
PB–C14B
O1B–C3B
O2B–C6B
C1B–C2B
C2B–C3B
C2B–C7B
C3B–C4B
C4B–C5B
C5B–C6B
C6B–C7B
PA–Pt–O1A
PA–Pt–PB
PA–Pt–O1B
O1A–Pt–PB
O1A–Pt–O1B
PB–Pt–O1B
91.7(2)
100.0(1)
167.5(2)
168.1(2)
76.2(2)
Br1–Zn–Br2
Br1–Zn–O1A
Br1–Zn–O1B
Br2–Zn–O1A
Br2–Zn–O1B
O1A–Zn–O1B
116.0(1)
105.7(2)
120.5(2)
123.2(2)
108.2(2)
78.8(2)
92.2(2)
Br1–Pt–Br2
Br1–Pt–PA
Br1–Pt–PB
86.6(1)
85.9(1)
174.3(1)
Br2–Pt–PA
Br2–Pt–PB
PA–Pt–PB
169.7(1)
89.0(1)
98.9(1)
Fig. 3 An ORTEP plot of cis-[Pt(O,P-PPh2thqH)2(ZnBr2)] 5. Details
as in Fig. 1.
and O-Zn-O 78.8(2)Њ]. The bond lengths (Table 3) are not
remarkable,24 and are similar to the corresponding distances
for 2. In the crystal structure each of the terminal hydro-
quinonate OH groups is hydrogen bonded to a molecule of dmf
[there are two dmf per molecule of 5; O (dmf) ؒ ؒ ؒ O (hydro-
quinonate) 2.56, 2.66 Å]. The cobalt() iodide adduct of
Fig. 2 An ORTEP plot of cis-[PtBr2(PPh2tq)2] 4. Details as in Fig. 1.
a bis(phosphinoenolate)nickel() complex, cis-[Ni{O,P-[Ph2-
28
ؒ ؒ ؒ
ؒ ؒ ؒ
PCH C( O)(4-MeC H )]} (CoI )], exhibits square planar
᎐᎐ ᎐᎐
6
4
2
2
spectrum shows prominent peaks at m/z 1031, 516, 952 and
809 corresponding to the [M Ϫ H]ϩ, [M]2ϩ, [M–Br]ϩ and [M–
ZnBr2]ϩ ions. The 31P-{1H} NMR spectrum exhibits a singlet at
δ 29.66 flanked by a cis satellite doublet [1J (31P–195Pt) 3851 Hz]
nickel and pseudotetrahedral cobalt centres giving overall a
structure comparable to that of 2.
The preparation of complex 5 from 4 confirms that the latter
complex undergoes four-electron reduction to produce a
hydroquinone phosphine complex. With the benefit of hind-
sight, it is not surprising that 5 was isolated rather than 3 which
was targeted. A large excess of zinc powder was used, which
would consume the hydrobromic acid and form zinc bromide.
Under the final conditions pertaining in the reaction, no acid
and an excess of zinc and zinc bromide, loss of the bromo
ligands from 3 is expected and 5 is a logical product. The reverse
reaction, 5 to 3, goes cleanly. Addition of a drop of 0.1 M
hydrobromic acid to 5 in d6-acetone afforded 3 in quantitative
yield (as judged by NMR spectroscopy).
1
and the H NMR spectrum shows the requisite peaks for two
equivalent cisoid hydroquinonate phosphine ligands and for the
dmf solvate molecules.
The structure of the metal dimer 5 was established by X-ray
crystallography, Fig. 3, and reveals that the ortho-oxygen atoms
of the hydroquinonate groups bridge the platinum and zinc
centres. Analogously to the structure of 2 (see above), the
platinum() ion in 5 is O,P-co-ordinated by two cisoid hydro-
quinonate phosphine (PPh2thqH) ligands [the sum of the bond
angles about the platinum ion (Table 3) is 695Њ compared to
720Њ for a perfect square planar centre; the bond angles P–Pt–P
100.0Њ and O-Pt-O 76.2Њ are little changed from those in 2]. The
zinc ion is co-ordinated by two bromo ligands and by the two
bridging ortho-oxygen atoms of the hydroquinonate phos-
phine ligands in distorted-tetrahedral fashion [the sum of the
bond angles about the zinc ion of 652.3Њ compares with 657Њ for
a perfect tetrahedron, but the individual bond angles (Table 3)
are considerably different from 109.5Њ, e.g. Br–Zn–Br 116.0(1)Њ
Electrochemical studies
The electrochemistries of complexes 1b, 3 and 4 have been
characterised by cyclic voltammetric and bulk controlled-
potential electrolysis experiments. Cyclic voltammograms were
recorded in anhydrous dcm or dmf with 0.1 M tetrabutyl-
ammonium hexafluorophosphate at a platinum disc working
1546
J. Chem. Soc., Dalton Trans., 1999, 1543–1554

View Article Online
electrode. All potentials are quoted relative to the ferrocenium–
ferrocene couple. Except where otherwise stated, all description
and potentials refer to measurements at 295 K run at a scan rate
(ν) of 100 mV sϪ1. The following information, from voltam-
mograms recorded under these conditions, is given for com-
parison with the electrochemistry of the platinum complexes.
p-Hydroquinone29 (H2hq) in dmf exhibits an irreversible
anodic peak at ϩ0.46 V for oxidation to p-benzoquinone
(q)-Hϩ and a daughter cathodic peak at Ϫ0.76 V for reduction
of q-Hϩ to give back q. p-Benzoquinone29 features a reversible
Ϫ
one-electron quinone–semiquinone radical anion (q–sq
)
᝽
couple at Ϫ0.91 V in dcm and at Ϫ0.87 V in dmf followed at
more negative potential by a quasireversible one-electron semi-
quinone anion–hydroquinone dianion (sq Ϫ–hq2Ϫ) couple at ca.
᝽
Ϫ1.45 V [∆Ep = 165 mV cf. ∆Ep(Fcϩ–Fc) = 86 mV] in dcm and
at Ϫ1.69 V [∆Ep = 220 mV cf. ∆Ep(Fcϩ–Fc) = 60 mV] in dmf.
Bromide ion30 (as 2.0 × 10Ϫ3 M Et4NBr) exhibits an anodic
peak at ϩ0.40 V in dcm and at ϩ0.32 V in dmf for the irrevers-
Ϫ
ible oxidation of BrϪ to Br3 followed by the quasireversible
Br3Ϫ–Br2 couple at ϩ0.62 V [∆Ep = 210 mV] in dcm and at
ϩ0.61 V [∆Ep = 190 mV] in dmf.
Cyclic voltammograms of complex 1b in dcm reveal an
irreversible reduction at Ϫ2.1 V and two broad irreversible
oxidation processes at ϩ0.82 and ϩ1.05 V followed by a
quasireversible couple at ca. ϩ1.3 V. By comparison with the
electrochemistry of [Pt(PR3)2X2] complexes where PR3 is an
electrochemically inactive phosphine and X is a halide ligand,
the irreversible reduction for 1b is attributed to the PtII–Pt0
couple, and the two irreversible anodic peaks to metal-centred
oxidations of the cis and trans isomers, respectively.31 The
quasireversible couple at more positive potential is assigned to
oxidation of the dimethoxyphenyl substituents.32
Fig. 4 Cyclic voltammogram (four cycles) of complex 4 in dcm–0.1 M
[NBun ][PF6] at a freshly polished platinum-disc working electrode;
4
scan rate = 100 mV sϪ1 and temperature = 295 K.
A cyclic voltammogram of complex 3 in dmf solution
revealed a broad anodic peak at ϩ0.65 V attributed to the
four-electron oxidation of the hydroquinone substituents29 to
afford 4-Hϩ, eqns. (3) and (4), and a daughter cathodic peak at
[PtBr2(PPh2thqH2)2] →
[PtBr2(PPh2tq)2] ϩ 4 eϪ ϩ 4 Hϩ (3)
[PtBr2(PPh2tq)2] ϩ 2 Hϩ → [PtBr2(PPh2tq-H)2]2ϩ (4)
Ϫ0.74 V for reduction of this product. The synthesis of 4 by
oxidation of 2 with 2 equivalents of DDQ provides support
for this assignment.
Figs. 4 and 5 show cyclic voltammograms of complex 4
recorded in dcm and dmf, respectively, at a freshly polished
platinum working electrode. The initial scans are clearly similar
in both solvents. However, subsequent scans in dcm are compli-
cated by stripping and adsorption behaviour resulting from
deposition of reduction product(s) onto the electrode,
behaviour not observed in dmf. In both solvents, scans to
negative potentials display an irreversible reduction process, PI
(at Ϫ0.86 V in dcm and at Ϫ0.76 V in dmf), which upon scan
inversion gives rise to two irreversible oxidation processes, PIII
and PIV (at ϩ0.05 and ϩ0.23 V, respectively, in dcm and at
Ϫ0.26 and ϩ0.02 V, respectively, in dmf). In dcm, PIV has a
cathodic counterpart, PV [∆Ep = 230 mV; i(PV)/i(PIV) ≈ 0.4 with
ν = 100 mV sϪ1], e.g. Fig. 4. In dmf, PV is barely discernible, e.g.
see Fig. 5. Peaks PII–PV are not present unless PI is traversed
first. In some cases in dmf, shoulders for the peaks of a weak
quasireversible couple can be discerned in the tail of PI. Cyclic
voltammograms were also recorded at scan rates between 50
mV sϪ1 and 10 V sϪ1 in both solvents (e.g. Fig. 6) and in dcm at
Ϫ78 ЊC. As the scan rate increases PI moves to more negative
potential, and plots of peak current versus ν1/2 are linear for PI–
PIV. At ambient temperature the following general changes are
noted as the scan rate (ν) is increased. (1) The reduction process
shows some chemical reversibility with PII, the anodic counter-
part of PI, growing in relative magnitude as the scan rate is
Fig. 5 Cyclic voltammogram (four cycles) of complex 4 in dmf–0.1 M
[NBun ][PF6]. Other details as in Fig. 4.
4
increased. (2) The relative magnitudes of PIII and PIV change
with PIII growing at the expense of PIV. At Ϫ78 ЊC the voltam-
mograms display PI–PIII only, with the PI–PII couple becoming
more chemically reversible and PIII relatively smaller at higher
scan rates.
Accurate measurements in both dcm or dmf reveal that the
first oxidation exhibited by BrϪ ion occurs to positive potential
of PIV, and cyclic voltammograms of equimolar complex 4 and
BrϪ ion in dcm exhibit PI–PV well clear of the peaks for the BrϪ–
Ϫ
Br3 oxidation and the Br3Ϫ–Br2 couple. The observations are
noteworthy because they indicate that neither PIII nor PIV ori-
ginates from oxidation of BrϪ ion. Moreover in dcm solutions
peaks for oxidation of BrϪ ion were never observed (e.g. see Fig.
6); BrϪ ion is not a product of PI–PIV. However, in dmf solution
a broad anodic peak was observed at ≈ϩ0.35 V in the tail of the
anodic solvent discharge (marked by the asterisk in Fig. 7),
which is close to the potential measured for the BrϪ–Br3Ϫ oxid-
ation. We ascribe the broad ϩ0.35 V peak to oxidation of a
small amount of bromide ion produced by preceding solvation
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1547

View Article Online
Fig. 6 Cyclic voltammograms of complex 4 in dmf–0.1 M [NBun ]-
Fig. 8 The Vis/NIR spectra of complex 4 in dmf–0.4 M [NBun ][PF6]
at 295 K prior to (–––) and after (——) controlled potential elect4rolysis
at Ϫ1.2 V.
4
[PF6] at 295 K and different scan rates. The platinum-disc working
electrode was polished between the recording of each voltammogram.
The arrows mark the potentials of the BrϪ → Br3Ϫ and Br3Ϫ → Br2
oxidations under identical conditions.
Controlled-potential coulometry reveals that PI is an one-
electron process. Two bulk electrolyses of complex 4 (1.0 mM)
in dmf at a Pt-gauze electrode held at Ϫ1.2 V were performed
and caused the solution to change from pale yellow to dark red-
grey and consumed 0.98 and 1.03 Faraday molϪ1. A third bulk
electrolysis at Ϫ1.0 V and half the concentration of 4 (0.5 mM)
consumed 0.89 Faraday molϪ1. In cyclic voltammograms
recorded after exhaustive electrolysis initial cathodic scans
reveal no peak out to Ϫ2.3 V, i.e. PI disappears, but PIII and PIV
appear in the reverse anodic scans, Fig. 7. Bulk electrolysis of
fully reduced 4 in dmf at ϩ0.27 V, i.e. positive of PIV, consumed
1.12 Faraday molϪ1, i.e. PIII and PIV together consume one elec-
tron per mol. The voltammogram after this reoxidation showed
peaks PI, PIII and PIV, albeit with PI broadened compared to
that in cyclic voltammograms of 4 at a freshly polished platinum
working electrode.
The reduction product of complex 4 was further character-
ised by UV/Visible/NIR, EPR and NMR spectroscopic studies.
Fig. 8 displays Visible/NIR spectra of 4 prior to and after
exhaustive reduction at PI. Both 4 and its reduction product
display a strong UV band [at 308 (ε ≈ 2790) and at 315 nm
(ε ≈ 6070 MϪ1 cmϪ1), respectively]. Whereas 4 only exhibits a
weak band in the visible region at 785 nm (ε ≈ 15 MϪ1 cmϪ1), the
reduction product shows a broad band at 850 nm (ε ≈ 300 MϪ1
cmϪ1) with shoulders at ca. 790 and 960 nm and a peak at 490
nm (ε ≈ 395 MϪ1 cmϪ1) on the tail of the intense UV band.
The radical nature of the reduction product of complex 4
was confirmed by EPR spectroscopy. First a solution of 4 in
d2-dcm was treated with excess of cobaltocene [4 equivalents;
E1/2 (CoCp2–CoCp2ϩ) = Ϫ1.35 V in dmf]. A brown solid immedi-
ately precipitated, a result consistent with the deposition of the
reduction product observed in cyclic voltammograms of 4 in
dcm. The suspension of the reduction product in d2-dcm exhib-
ited a strong isotropic EPR signal at g = 2.002 which slowly
decayed over several days whilst an anisotropic signal at g = 5.2
slowly appeared. Next 4 in d4–dmf was titrated with cobalto-
cene. Addition of 1 equivalent gave a black solution with a
Fig. 7 Cyclic voltammograms of complex 4 in dmf–0.4 M [NBun ][PF6]
at 295 K prior to (a) and after (b) controlled potential electrol4ysis at
Ϫ1.2 V.
of 4 by dmf, the chemical step–electrochemical step (CE)
mechanism33 described by eqns. (5) and (6). Consistent with
C: [PtBr2(PPh2tq)2] ϩ dmf
[PtBr2-n(dmf)n(PPh2tq)2]nϩ (n = 1 or 2) ϩ nBrϪ (5)
E: 3 BrϪ → Br3Ϫ ϩ 2 eϪ
(6)
this interpretation, the peak is present in initial scans to positive
potentials, prior to scanning through PI–PV, and the current
from the oxidation remains about the same whether PI–PV are
traversed or not. Conductivity measurements indicate that 4
is a non-electrolyte in dmf solution and NMR spectra of 4 in
d7-dmf show peaks for a single species only. Taken together
these results imply that 4 is unsolvated [i.e. the equilibrium (5)
lies firmly to the left] with solvation only occurring, within the
CV timescale, when an electrode potential sufficient to oxidise
BrϪ ion is applied, eqn. (6).
1
reddish hue. The H NMR spectrum of the solution exhibited
only peaks for cobaltocenium ion and the protio impurities in
the solvent and the EPR spectrum of the frozen solution at 77
K revealed a broad isotropic peak at g = 2.003, Fig. 9. Hyper-
fine coupling was not resolved. The intensity of the EPR signal
did not change upon addition of further cobaltocene (up to 5
equivalents). Solutions produced by exhaustive reductive bulk
electrolysis of 4 exhibited the same EPR signal. The insolubility
of the radical product in dcm and other solvents with a suitable
1548
J. Chem. Soc., Dalton Trans., 1999, 1543–1554

View Article Online
reported for a bis(quinone) and arises because the unpaired
Ϫ
electron in the monoanion I is fully delocalised over both
᝽
quinone centres. Inspection of crude molecular models of 4,35
starting with the parameters from its crystal structure and
leaving the cis-PtBr2P2 core in fixed position, indicates that
rotations about the P–Pt bonds and the bonds to the benzylic
carbon atoms lead to minimum attainable distances between
the quinone rings of 5.3 Å with the rings parallel and 4.2 Å with
the rings perpendicular. The quinone groups remain too far
apart for significant π interaction although a small splitting of
Ϫ
the two q–sq couples is expected due to electrostatic effects.
᝽
Ϫ
The absence of any reduction processes for 6 (to over 1.5 V
᝽
more negative potential than the q–sq Ϫ couple for 4) therefore
᝽
points to both quinone groups in 4 being involved in the single
chemically irreversible, one-electron reduction process. How
can the addition of a single electron affect the two widely
separated quinone groups?
Fig. 9 The EPR spectrum of the product (6 Ϫ) from reduction of
complex 4 with cobaltocene (1 equivalent). Conditions: dmf glass,
T = 77 K, ν = 9.497 GHz.
O
O
O
O
᝽
IR window prevented measurement of its IR spectrum over the
I
region for quinone C᎐O bands. Attempts to isolate solid prod-
᎐
uct for further analysis were thwarted by decomposition, con-
sistent with the radical product being extremely oxygen and/or
moisture sensitive.
The above data reveal the following. Peak PI can confidently
be assigned to a quinone-centred reduction of complex 4 by
Scheme 2 presents a possible mechanism for reduction of
complex 4, a solution to this enigma. The first step is one-
electron reduction of a quinone substituent, i.e. the expected
Ϫ
q–sq couple.† The nucleophilicity of semiquinones has been
᝽
comparison with the cyclic voltammograms of 1b (devoid of
measured to increase by up to six orders of magnitude com-
pared to the parent quinones.36 The second step is addition of
the nucleophilic semiquinone substituent to the electrophilic
Ϫ
electrochemistry in this region) and p-benzoquinone (q–sq
᝽
couple at about the same potential). The dependence of PII on
scan rate indicates that a fast chemical step depletes the semi-
platinum() centre8 to afford {4*} Ϫ. This is the anticipated
᝽
quinone product of PI (4 Ϫ) and generates a product radical
᝽
first step in a ligand exchange reaction at a square planar
species (6 Ϫ; we assign this as an anion because bromide ion is
᝽
platinum() centre, and such a reaction would normally pro-
Ϫ
ceed by loss of a leaving ligand from {4*} (e.g. BrϪ ion).
not produced).
᝽
Both peaks PIII and PIV remain in cyclic voltammograms of
However, the platinum centre is now electron rich (bound by the
semiquinonate anion) and we posit that the conveniently
placed, second quinone substituent oxidatively adds to the
Ϫ
6
(whether produced by exhaustive reductive electrolysis or
᝽
by cobaltocene reduction of 4). Clearly these anodic peaks are
due to 6 Ϫ. Cooling a sample slows down chemical steps and
᝽
platinum centre to afford 6 Ϫ, which is suggested to be
᝽
recording a cyclic voltammogram at low temperature can be
thought of as an alternative to scanning at very fast scan rates.
Only PI–PIII are observed in voltammograms of complex 4 at
low temperature and PIII is therefore assigned to irreversible
an octahedral platinum() (hydroquinonate)(semiquinonate)
Ϫ
ؒ
species, namely [PtBr2(O,P-PPh2thq)(O,P-PPh2tsq )] . The
Ϫ
ؒ
platinum() tautomer, [PtBr2(O,P-PPh2tsq )2] , is discounted
Ϫ
for 6 on the basis of the isotropic EPR signal without 195Pt
᝽
oxidation of 6 Ϫ. Peak PIV can not arise from further oxidation
᝽
hyperfine coupling [we also note that genuine platinum()
complexes are rare37].
The proposed structure for 6 neatly accounts for the EPR
of the product(s) produced at PIII as this would consume more
than one electron per mol. The results are consistent, however,
Ϫ
᝽
with pre-equilibration between 6 Ϫ, which is oxidised at PIII,
᝽
spectrum, a ligand-centred radical, and the intense, broad band
in the Vis/NIR spectrum (which is atypical for a simple plat-
inum complex37 or for a simple semiquinone38); it could arise
from either an intervalence charge transfer transition (νIT in
Scheme 2) or a ligand-to-metal charge transfer (hq2Ϫ→PtIV)
transition.9 Moreover, the proposed structure is entirely analo-
gous to those of recently described [CoIII(N–N)(catechol-
ate)(o-semiquinonate)] (N–N = dinitrogen chelate ligand)
complexes which display both intervalance charge transfer and
ligand-to-metal charge transfer {catecholate→CoIII} transi-
tions as well as characteristic temperature-dependent, spin
transitions to [CoII(N–N)(o-semiquinonate)2] species.9 The
major difference between the structures of these cobalt species
and a second species, which is oxidised at PIV, with the pre-
equilibrium favouring 6 and being rapid at ambient tem-
perature and slow compared to the CV timescale at low
temperature (Ϫ78 ЊC). Possibilities for the chemical step include
solvation or isomerisation reactions.
Ϫ
᝽
That the reduction of complex 4 is firmly an one-electron
process and produces a radical product (6 Ϫ) that can not be
᝽
further reduced (no cathodic processes before the tail of the
cathodic discharge at Ϫ2.3 V and no further reduction with
excess of cobaltocene, see above) is (at first) puzzling! There are
two quinone groups in 4 and a q–sq Ϫ couple is anticipated for
᝽
each one.29,34 If the two quinone groups were non-interacting,
the q–sq Ϫ couples would be coincident and appear as a single
᝽
two-electron process. For closer quinone groups electrostatic
† If the quinone substituents in complex 4 are sufficiently isolated for
simultaneous reduction of both to give 42Ϫ, a diradical with two semi-
quinone substituents, then addition of the semiquinone substituents
to the metal centre and the cross-reduction of parent 4 (within the
effects and electronic delocalisation might lead to consecutive,
individual q–sq Ϫ couples, but even in the limit of full electronic
᝽
delocalisation between the redox centres the couples should be
separated by ӷ500 mV.34 For example, the helical bis(quinone) I
CV timescale) would lead to 6 Ϫ and the observed overall one-electron
᝽
displays consecutive reversible q–sq Ϫ couples separated by 470
᝽
reduction stoichiometry. Scheme 2 shows only one of several isomers
Ϫ
mV.34a,b This is the largest separation of the q–sq couples
Ϫ
possible for [PtBr2(O,P-PPh2thq)(O,P-PPh2tsq )]Ϫ 6
.
ؒ
᝽
᝽
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1549

View Article Online
O
O
O
O
O
O
Br
Br
± e–
E1/2 (q/sq• –
PtII
PPh2
Br
PtII
O
PtII
O
PPh2
PPh2
Br
Br
Br
)
Ph2P
Ph2P
Ph2P
O
O
O
O
4• –
{4*}• –
4
O
O
O
O
Br
Br
PtIV
O
PtIV
O
PPh2
PPh2
Br
Ph2P
Br
Ph2P
e–
hνIT
O
O
6• –
Scheme 2
Ϫ
and that proposed for 6 is that the 5 oxygen atoms (see
᝽
Ϫ
Chart 1) in 6 are not bound to the metal centre and remain
᝽
susceptible to protonation and subsequent reaction(s), perhaps
accounting for the observed instability of this reduction
product.
Fig. 10 shows the cyclic voltammogram of complex 4 in dmf
in the presence of dilute hydrobromic acid. On adding the acid
a new reduction process (PI*) with a peak current ≈7.5 times
larger than that for PI (i.e. prior to addition of acid) appears at
Ϫ1.05 V. Also, PII–PV disappear and a new anodic peak (PII*) is
found at Ϫ0.62 V (compare Figs. 5 and 10). The Randles–
Sevcˇik eqn. (7)33 describes the peak current in a cyclic voltam-
∞
ip = Ϫ(2.69 × 105)n3/2Co D1/2ν1/2
(7)
∞
mogram, where Co is the bulk concentration of the species
undergoing oxidation/reduction, D its diffusion coefficient and
n the number of electrons transferred. In the present experi-
ment only the latter parameter changes on adding the acid. The
increase in the peak current indicates, therefore, a transition
from an one- to a four-electron reduction process (43/2 = 8).
Accordingly, PI* is attributed to reduction of 4 to afford 3, i.e.
the reverse of eqn. (3), and PII* to the reverse process. This is
consistent with the voltammetry displayed by benzoquinones
under acidic conditions and with the preparation of 5 from 4
(see above). Semiquinones disproportionate under protic condi-
tions, often at near to diffusion controlled rates,39 explaining the
switch from one- to four-electron reduction behaviour when
acid is added to a solution of 4.29
Fig. 10 Cyclic voltammogram of complex 4 and ferrocene in dmf–0.1
M [NBun ][PF6] at a freshly polished platinum-disc working electrode
4
after addition of 5% v/v 0.1 M hydrobromic acid; scan rate = 100 mV
s
Ϫ1 and temperature = 295 K.
The co-ordination of the hydroquinone groups to the platinum
centre in these complexes can be reversibly controlled by pH
adjustment. There are few surprises here; other studies of tran-
sition metal complexes with potentially O,P-chelate alcohol or
phenol-substituted phosphine ligands have demonstrated simi-
lar pH control of oxygen co-ordination to the metal centre.23,24
The quinone phosphine complex 4 is easily prepared by oxid-
ation of the hydroquinone phosphine precursor 3 and is not
susceptible to hydrolytic loss of the two quinone groups, unlike
Conclusions
Synthetic strategies to the hydroquinone diphenylphosphine
(PPh2thqH2) complexes of platinum() have been developed.
1550
J. Chem. Soc., Dalton Trans., 1999, 1543–1554

View Article Online
related complexes where the quinone substituents are directly
bonded to the phosphorus atoms.16,19,20 The stability of 4 sug-
gests other transition metal complexes of quinone phosphines
should be readily available and we anticipate rich chemistry for
these.
electrode comprised of a commercial Ag–AgCl mini-reference
electrode (Cyprus Systems, Inc. EE008) but filled with the
electrolyte solution to be used in the experiment [rather than
AgCl saturated 3 M KCl(aq) solution], a freshly polished
platinum disc (1 mm diameter) working electrode and
a platinum wire as the auxiliary electrode. Freshly polished
platinum working electrodes were prepared from commercial
mini-electrodes (Cyprus Systems, Inc. EE041) by grinding with
SiC emery paper (600 mesh), then successively polishing with 6
and 1 µm diamond slurries, and finally with 0.2 µm alumina
slurry. Between each grinding and polishing step, and after final
polishing, the electrode was sonicated in doubly distilled water
for 5 min. The electrodes were then rinsed with the solvent to be
used and thoroughly dried. The solvents used for electro-
chemical measurements, dcm and dmf, were highest quality
anhydrous grade sealed under argon (Aldrich) and were used as
Most remarkable is the electrochemistry of complex 4.
Under protic conditions, chemical and electrochemical reduc-
tion of 4 is a four-electron process (two electrons per quinone
group) and produces hydroquinone phosphine complexes (e.g.
2 and 5). This behaviour is as expected. In stark contrast, 4 is
cleanly reduced under aprotic conditions by one electron to a
radical product (6 Ϫ) that can not be further reduced. The
᝽
dichotomy between one-electron reduction and the absence of
quinone-centred electrochemistry in the product leads us to
argue that reduction of 4 is accompanied by co-ordination of
both quinone substituents to the platinum ion. It is proposed
that a concomitant, extraordinary redistribution of electrons
obtained. The support electrolyte was 0.1 M [NBun ][PF6].
4
between the ligand and metal redox centres leads to 6 Ϫ with a
Solutions were deoxygenated by flushing with high purity
nitrogen (presaturated with solvent) and then blanketed with
a cover of nitrogen for the duration of the experiment. An
electrochemical scan of the solvent electrolyte system was
always recorded before the addition of the compound to ensure
that there were no spurious signals. All potentials are quoted
relative to the ferrocenium–ferrocene (Fcϩ–Fc) couple which
was measured in situ as an internal reference.
᝽
platinum() ion bound by one hydroquinonate and one semi-
quinonate ligand (i.e., the addition of one electron to 4
causes two-electron oxidation of platinum and net three-electron
reduction of the quinone groups!). The mechanism, Scheme 2,
follows logically from the increased nucleophilicity of the semi-
quinone substituent produced by quinone-centred reduction of
Ϫ
4, and the suggested structure of 6 neatly accounts for its
᝽
reactivity, and electrochemical and spectroscopic properties. We
Controlled potential coulometry was carried out in a con-
ventional three-compartment “H”-cell adapted so that it could
be loaded and sealed under an inert atmosphere (high purity
dinitrogen). An Ag–AgCl quasi-reference electrode (the same
as used in CV experiments) was placed in the working com-
partment along with the platinum gauze (5 × 2 cm2) working
electrode and a platinum disc mini-electrode for running CV
experiments. The counter electrode was a platinum gauze (4 × 2
cm2) separated from the working compartment by two fine-
porosity glass frits. During the electrolyses the solutions in the
working compartment were stirred magnetically with a Teflon-
coated stirring bar. The concentration of the substrate was
believe that the formation of 6 Ϫ demonstrates the proclivity of
᝽
quinone pendants, upon reduction to their semiquinone or
hydroquinone anion counterparts, to bind and transfer elec-
trons with a transition metal centre.8–10 In closing, we re-
emphasise that such processes are of pivotal biological
importance,2–7 although the present work in no way attempts to
model biological systems.
Experimental
Reactions were routinely carried out under an atmosphere of
dry dinitrogen using standard Schlenk and cannula techniques.
Solvents were distilled from the appropriate drying agent under
dinitrogen immediately prior to use: dcm and acetonitrile from
P2O5 and then from CaH2; acetone from KMnO4 and then from
anhydrous B2O3; hexanes from sodium wire; diethyl ether and
tetrahydrofuran from sodium benzophenone ketyl; dmf was
dried over calcium hydride and then twice distilled under
reduced pressure; methanol and ethanol were distilled from
magnesium turnings. Chemicals were obtained from com-
mercial sources (usually Aldrich) and used as obtained.
Microanalyses for C, H and N were performed by the
University of New South Wales microanalytical service. Induct-
ively coupled plasma (ICP) analyses for other elements (P, Pt
and Zn) employed a GBC Integra ICP-AES multi-channel
instrument. Prior to analysis, samples were dried at 35 ЊC for 48
h under vacuum (0.2 mmHg) over phosphorus pentaoxide.
Quoted melting points are uncorrected. The EI and ES mass
spectra were recorded using a VG Quattro mass spectrometer;
the carrier stream for ES was 1% acetic acid in 1:1 acetonitrile–
water. The 1H and 13C NMR spectra were obtained in the des-
ignated solvents on a Bruker AC300F (300 MHz) instrument,
31P NMR spectra on a Bruker ACP300 spectrometer operating
at 121.46 MHz and were referenced relative to external 85%
phosphoric acid. The EPR spectra were recorded on a Bruker
EMX 10 spectrometer, IR spectra as paraffin mulls on a
Perkin-Elmer 580B spectrometer and electronic spectra using
a CARY 5 spectrophotometer in the dual beam mode.
1.0 mM and the support electrolyte was 0.4 M [NBun ][PF6].
4
Two test electrolyses were performed using these conditions:
oxidation of ferrocene to ferrocenium ion and reduction of
cobaltocenium hexafluorophosphate to cobaltocene consumed
1.04 and 0.99 Faraday molϪ1 respectively.
Preparations
(2,5-Dimethoxbenzyl)diphenylphosphine (PPh2dmb). Step 1:
2,5-dimethoxytoluene. Iodomethane (33 mL, 0.53 mol) was
added by syringe to a mechanically stirred mixture of 2,5-
dihydroxytoluene (30 g, 0.24 mol) and potassium carbonate
(73.5 g, 0.24 mol) in degassed dmf under a dinitrogen atmos-
phere. The solution mixture was stirred for 16 h. Cooled water
(0 ЊC) was added to the resulting pink solution until all the
potassium carbonate had dissolved. The red solution was
extracted with diethyl ether several times and the combined
extracts washed successively with 2.5 M sodium hydroxide and
water. The diethyl ether phase was dried over anhydrous mag-
nesium sulfate and the solvent removed giving a red liquid.
Distillation under 20 mmHg pressure at 120 ЊC gave a colour-
less liquid, 2,5-dimethoxytoluene (17.50 g, 50%). The com-
1
pound was identified from its NMR spectrum:41 δH (CDCl3)
6.83–6.73 (3 H, m, C6H3), 3.84 (3 H, s, OCH3), 3.81 (3 H, s,
OCH3) and 2.31 (3 H, s, CH3).
Step 2: 2,5-dimethoxybenzyl bromide. CAUTION: This prod-
uct irritates the skin and eyes and must be handled with due
care. A solution of 2,5-dimethoxytoluene (17.50 g, 0.12 mol) in
carbon tetrachloride (25 mL) was added to a suspension of
N-bromosuccinamide (22.5 g, 0.12 mmol) and benzoyl peroxide
(10 mg) in carbon tetrachloride (200 mL). The solution mixture
was mechanically stirred and irradiated with a 200 watt lamp
for 5 h. The hot solution was filtered and the volume of the
Electrochemical measurements were recorded using
a
Pine Instrument Co. AFCBP1 Bipotentiostat interfaced to and
controlled by a Pentium computer. Data were transferred
to a Power Macintosh computer for processing using the
IGORPRO 2.0TM software.40 For CV measurements, a standard
three electrode configuration was used with a quasi-reference
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1551

View Article Online
solution reduced to ca. 100 mL. The resulting solution was
cooled (ice–methanol bath) and gave the product as an off
white solid (14 g, 50% ), mp 75 ЊC (lit.42: 75–76 ЊC) (Found: C,
46.90; H, 4.75. Calc. for C9H11BrO2: C, 46.75; H, 4.76%).
δη (CDCl3) 6.91 (2 H, m, C6H3), 6.82 (1 H, s, C6H3), 4.54 (2 H, s,
CH2), 3.85 (3 H, s, OCH3) and 3.77 (3 H, s, OCH3).
3.39. C42H42I2O4P2Pt requires C, 44.96; H, 3.75%); NMR
spectra show two isomers. δH (CDCl3), cis-[PtI2(PPh2dmb)2]
(65%) 7.75 (2 H, m, C6H3), 7.15 (8 H, m, Ph), 7.09 (8 H, m, Ph),
6.92 (4 H, m, Ph), 6.82 (2 H, m, C6H3), 6.46 [2 H, d, J(HH) 9,
C6H3], 4.52 [4 H, d, J(PH) 12 Hz, CH2], 3.90 (6 H, s, OCH3) and
2.98 (6 H, s, OCH3); and trans-[PtI2(PPh2dmb)2] (35%) 7.56
(8 H, m, Ph), 7.40 (8 H m, Ph), 7.29 (6 H, m, Ph and C6H3), 6.68
(2 H, m, C6H3), 6.52 [2 H, d, J(HH) 9, C6H3], 4.28 [4 H, t, J(PH)
9 Hz, CH2], 3.52 (6 H, s, OCH3) and 3.10 (6 H, s, OCH3).
δP (CDCl3) 6.66 [s, J(31P–195Pt) 3545, cis isomer] and 1.55 [s,
J(31P–195Pt) 2395 Hz, trans isomer].
Step 3: PPh2dmb. Magnesium turnings (8 g, 0.33 mol) were
stirred dry in a 250 mL Schlenk flask under a dinitrogen atmos-
phere for 48 h. Grey crushed magnesium powder was pro-
duced.21 The flask was then connected to a pressure equalising
dropping funnel. Freshly distilled diethyl ether (50 mL) was
added to the magnesium powder, the mixture was cooled to
0 ЊC and a solution of 2,5-dimethoxybenzyl bromide (3 g, 13
mmol) in diethyl ether (100 mL) added dropwise to the
centre of the vortex created by rapid stirring. Addition of the
2,5-dimethoxybenzyl bromide solution took 2 h. The mixture
was stirred for 2 h and then the solution was filtered via a
cannula into a solution of chlorodiphenylphosphine (2.33 mL,
13 mmol) in diethyl ether at 0 ЊC. A white precipitate formed
immediately. After stirring for 16 h the reaction mixture was
quenched with aqueous ammonium chloride (3 g in 50 mL of
water). The diethyl ether phase was collected and dried with
magnesium sulfate. Removing the solvent gave an oily residue.
Recrystallisation from methanol yielded the product as a white
solid (3.85 g, 88%), mp 50 ЊC (Found: C, 75.20; H, 6.50.
C21H21O2P requires C, 75.00; H, 6.25%). δH (CDCl3) 7.45–7.31
(10 H, m, Ph), 6.74 [1 H, d, J(HH) 9, C6H3], 6.68 [1 H, dd,
J(HH) 9 and 2 Hz, C6H3], 6.39 (1 H, m, C6H3), 3.69 (3 H, s,
OCH3), 3.57 (3 H, s, OCH3) and 3.44 (2 H, s, CH2). δP (CDCl3)
Ϫ11.65 (s). m/z (EI-MS) 336 (Mϩ).
[Pt(O,P-PPh2thqH)2] 2. Boron tribromide (1.5 mL, 15 mmol)
was added to a stirred solution of [PtCl2(PPh2dmb)2] (0.50 g,
0.11 mmol), in dcm (25 mL) at 0 ЊC under a dinitrogen atmos-
phere. The solution changed from light yellow to red brown on
addition of the boron tribromide. After 16 h the solvent was
removed. The resulting foamy yellow solid was treated with
distilled methanol (15 mL) and then sodium carbonate (1.5 g)
was added. After 3 h much light yellow precipitate had formed.
This was collected by filtration, washed with water to remove
the excess of sodium carbonate, rinsed with diethyl ether and
air dried to yield the product, as a light yellow solid (420 mg,
96%), mp 180 ЊC (decomp.) (Found: C, 54.43; H, 4.06.
C38H32O4P2Ptؒ1.5H2O requires C, 54.55; H, 4.19%). This com-
pound dissolved in (CH3)2SO to produce a clear light yellow
solution, but almost immediately a yellow solvate hydrate pre-
cipitated. [Found: C, 50.74; H, 4.69. C38H32O4P2Ptؒ2(CH3)2SOؒ
2H2O requires C, 50.35; H, 4.79%]. The solvate hydrate and the
original product both dissolved in dmf. m/z (ES-MS, dissolved
in dmf) 810 (M ϩ H). δH [(CD3)2NCDO] 8.07 (1 H, s, OH), 8.01
(1 H, s, OH), 7.35–7.19 (20 H, m, Ph), 6.62 [2 H, d, J(HH) 6,
C6H3], 6.43 [2 H, d, J(HH) 6 Hz, C6H3], 6.16 (2 H, s, C6H3) and
4.11 [4 H, d, J(PH) 13.5 Hz CH2]. δP [(CD3)2NCDO] 28.08
[s, J(31P–195Pt) 3822 Hz].
[PtX2(PPh2dmb)2] complexes. X = Cl (1a). A solution of
[PtCl2(PhCN)2] (0.17 g, 0.36 mmol) in dcm (3 mL) was added to
a solution of PPh2dmb (0.25 g, 0.74 mmol) in dcm (10 mL) and
the resulting clear solution stirred for ca. 15 min. The solvent
was then reduced to ca. 3 mL. Dropwise addition of diethyl
ether (40 mL) precipitated an off white solid (0.32 g, 95%), mp
248 ЊC (Found: C, 53.77; H, 4.62. C42H42Cl2O4P2Pt requires C,
53.73; H, 4.48%); NMR spectra show two isomers. δH (CDCl3),
cis-[PtCl2(PPh2dmb)2] (95%) 7.81 (2 H, m, C6H3), 7.18 (8 H, m,
Ph), 7.09 (8 H, m, Ph), 6.94 (4 H, m, Ph), 6.81 [2 H, d, J(HH) 9,
C6H3], 6.41 [2 H, d, J(HH) 9, C6H3], 4.16 [4 H, d, J(PH) 12 Hz,
CH2], 3.90 (6 H, s, OCH3) and 2.98 (6 H, s, OCH3); and trans-
[PtCl2(PPh2dmb)2] (5%) 7.60 (8 H, m, Ph), 7.38 (8 H, m, Ph),
7.30 (6 H, m, Ph and C6H3), 6.65 (2 H, m, C6H3), 6.52 [2 H, d,
J(HH) 9, C6H3], 4.23 [4 H, t, J(PH) 8 Hz, CH2], 3.51 (6 H, s,
OCH3) and 3.03 (6 H, s, OCH3). δP (CDCl3) 11.53 [s, J(31P-195Pt)
3780, cis isomer] and 15.99 [s, J(31P–195Pt) 2565 Hz, trans
isomer].
[PtBr2(PPh2thqH2)2] 3. Concentrated hydrobromic acid (0.5
mL) was added to a suspension of [Pt(O,P-PPh2thqH)2] (0.21 g,
0.26 mmol) in acetone (20 mL). The solution was stirred for
15 min, then the volume of the solvent was reduced to 5 mL
and water added to give the product, [PtBr2(PPh2thqH2)2], as an
off white solid (0.21 g, 81%), mp 264 ЊC (Found: C, 47.90; H,
3.83. C38H34Br2O4P2PtؒC3H6O requires C, 47.81; H, 3.89%).
δH [(CD3)2CO] 8.30 (2 H, s, OH), 7.75 (2 H, s, C6H3), 7.44 (2 H,
s, OH), 7.21 (12 H, m, Ph), 6.93 (8 H, m, Ph), 6.60 [2 H, d,
J(HH) 9, C6H3], 6.47 [2 H, d, J(HH) 9, C6H3] and 4.30 [4 H, d,
J(PH) 12 Hz, CH2]. δP [(CD3)2CO)] 12.18 [s, J(31P–195Pt) 3720
Hz]. νmax/cmϪ1 (OH) 3272s and 3261s (paraffin mull).
X = Br (1b). A mixture of [PtCl2(PPh2dmb)2] (0.10 g, 0.1
mmol) and sodium bromide (0.10 g, 1.00 mmol) was stirred in
dcm (25 mL) for 2 h. The undissolved solid was removed by
filtration and the solvent removed from the filtrate to produce
a pale yellow solid (0.09 g, 80%), mp 246 ЊC (Found: C,
49.14; H, 4.48. C42H42Br2O4P2Pt requires C, 49.07; H, 4.09%);
NMR spectra show two isomers. δH (CDCl3), cis-[PtBr2-
(PPh2dmb)2] (95%) 7.80 (2 H, m, C6H3), 7.18 (8 H, m, Ph), 7.09
(8 H, m, Ph), 6.94 (4 H, m, Ph), 6.83 [2 H, dd, J(PH) 12 and 3,
C6H3], 6.46 [2 H, d, J(HH) 9, C6H3, 2H], 4.32 [4 H, d, J(PH)
12 Hz, CH2], 3.91 (6 H, s, OCH3) and 2.98 (6 H, s, OCH3); and
trans-[PtBr2(PPh2dmb)2] (5%) 7.56 (8 H, m, Ph), 7.37 (8 H, m,
Ph), 7.29 (6 H, m, Ph and C6H3), 6.68 (2 H, m, C6H3), 6.53 [2 H,
d, J(HH) 9, C6H3], 4.24 [4 H, t, J(PH) 8 Hz, CH2], 3.53 (6 H, s,
OCH3) and 3.08 (6 H, s, OCH3). δP (CDCl3) 10.97 [s, J(31P–
195Pt) 3725, cis isomer] and 15.30 [s, J(31P–195Pt) 2485 Hz, trans
isomer].
[PtBr2(PPh2tq)2] 4. A solution of DDQ (45 mg, 0.20 mmol)
in acetone (2 mL) was added to a solution of [PtBr2(P-
Ph2thqH2)2] (0.10 g, 0.10 mmol) in acetone (15 mL), and the
resulting solution stirred for 15 min. The solvent was then
removed and the residue extracted with dcm (20 mL). The sol-
vent was removed from the extract giving the product as a red-
brown solid (0.10 g, 97%), mp 140 ЊC (Found: C, 47.60; H, 3.21.
C38H30Br2O4P2Pt requires C, 47.16; 3.10%). m/z (ES-MS) 887
[M Ϫ Br]ϩ. δH (CDCl3) 7.45 (8 H, m, Ph), 7.32 (4 H, m, Ph),
7.14 (8 H, m, Ph), 7.09 (2 H, m, C6H3), 6.65 [2 H, dd, J(HH)
10 and 2.5, C6H3], 6.53 [d, J(HH) 10, C6H3] and 4.07 [d, J(PH)
13 Hz, CH2]. δP (CDCl3) 10.23 [s, J(31P–195Pt) 3685 Hz]. Con-
ductivity: ΛM (1.0 mM in dmf) = 3.7 S cm2 molϪ1. νmax/cmϪ1
(CO) 1657s (paraffin mull).
Reduction of [PtBr2(PPh2tq)2] 4 with cobaltocene. Several
experiments were performed, all in sealable EPR/NMR tubes
which could be fitted to a vacuum manifold. In a typical
experiment, a sealable NMR tube was charged with [PtBr2-
(PPh2tq)2] (20 mg, 0.02 mmol) and cobaltocene (3.9 mg, 0.02
mmol) and attached to the vacuum manifold. Freeze-thaw
X = I (1c). Reaction of [PtCl2(PPh2dmb)2] (0.10 g, 0.10
mmol) with potassium iodide (0.20 g, 1.2 mmol) using the
method outlined for [PtBr2(PPh2dmb)2] gave the product as a
red-brown solid (0.09 g, 80%) mp 230 ЊC (Found: C, 44.65; H,
1552
J. Chem. Soc., Dalton Trans., 1999, 1543–1554

View Article Online
degassed CD2Cl2 transferred to the tube by cooling it with
liquid nitrogen whilst under vacuum. The NMR tube was
sealed and the CD2Cl2 thawed. A brown precipitate in a clear
CCDC reference number 186/1399.
See http://www.rsc.org/suppdata/dt/1999/1543/ for crystallo-
graphic files in .cif format.
1
yellow solution formed. The H NMR spectrum showed only
peaks for protio-solvent impurities (δ 5.32) and cobaltocenium
ion (δ 5.88). The EPR spectrum showed a strong isotropic
signal at g = 2.002. In experiments using dmf as the solvent,
deoxygenated dmf was introduced to the evacuated EPR tube
via cannula.
Acknowledgements
S. B. S. thanks the University of North Sumatra for a period of
leave.
References
[Pt(O,P-PPh2thqH)2(ZnBr2)] 5. Hydrobromic acid (3 mL, 0.1
M) was added dropwise to a stirred suspension of zinc powder
(200 mg) in a solution of [PtBr2(PPh2tq)2] (30 mg, 0.03 mmol)
in acetone (25 mL). After 1 h the mixture was filtered to remove
the excess of zinc and the solvent concentrated to ca. 10 mL.
Water was added to precipitate a faun powder which was col-
lected by filtration, washed with water and diethyl ether, and
dried in vacuo (30 mg, 90%). ICP analysis: P:Pt:Zn ratio
≈ 2:1:1. m/z (ES-MS) 516 (M2ϩ), 809 (M Ϫ ZnBr2]ϩ), 952
([M Ϫ Br]ϩ) and 1031 ([M Ϫ H]ϩ). δH [(CD3)2CO] 7.55–7.27
(20 H, m, Ph), 6.50 (2 H, m, C6H3), 6.18 (2 H, br, C6H3), 5.61
[2 H, d, J(PH) 10, C6H3] and 3.86 [4 H, d, J(PH) 13 Hz, CH2].
δP [(CD3)2CO] 29.66 [s, J(31P–195Pt) 3851 Hz]. Crystals of this
sample were obtained on recrystallisation from dmf under an
atmosphere of diethyl ether, and the formulation of 5 rests on
the above data, on a crystal structure analysis of 5ؒ2dmf and on
the reaction of 5 with hydrobromic acid (see next).
1 Z. Rappoport and S. Patai (Editors), The Chemistry of the
Quinonoid Compounds, Wiley, New York, 1988, vol. II, Parts 1 and
2; S. Patai (Editor), The Chemistry of the Quinoid Compounds,
Wiley, New York, 1974, vol. I, Parts 1 and 2.
2 D. M. Dooley, M. A. McGuirl, D. E. Brown, P. N. Turoski,
W. S. McIntire and P. F. Knowles, Nature (London), 1991, 349,
262; D. M. Dooley, R. A. Scott, P. F. Knowles, C. M. Colangelo,
M. A. McGuirl and D. E. Brown, J. Am. Chem. Soc., 1998, 120,
2599.
3 J. P. Kliman, Chem. Rev., 1996, 96, 2541; W. S. McIntire, Annu. Rev.
Nutrition, 1998, 18, 145.
4 S. X. Wang, M. Mure, K. F. Medzihradsky, A. L. Burlingame,
D. E. Brown, D. M. Dooley, A. J. Smith, H. M. Kagan and J. P.
Klinman, Science, 1996, 273, 1076; L. I. Smithmungo and H. M.
Kagan, Matrix Biology, 1998, 16, 387.
5 J. Deisenhofer, O. Epp, K. Micki, R. Huber and H. Michel, Nature
(London), 1985, 318, 618; H. Kurreck and M. Huber, Angew. Chem.,
Int. Ed. Engl., 1995, 34, 849.
6 D. G. Nichols and S. J. Ferguson, Bioenergetics 2, Academic Press,
New York, 1992.
7 S. Iwata, J. W. Lee, K. Okada, J. K. Lee, M. Iwata, B. Rasmussen,
T. A. Link, S. Ramaswamy and B. K. Jap, Science, 1998, 281, 64;
U. Brandt, Biochim. Biophys. Acta., Bioenerg., 1998, 1364, 85.
8 C. G. Pierpont and R. M. Buchanan, Coord. Chem. Rev., 1981, 38,
45; C. G. Pierpont and C. Lange, Prog. Inorg. Chem., 1994, 41, 331.
9 D. M. Adams, A. Dei, A. L. Rheingold and D. N. Hendrickson,
J. Am. Chem. Soc., 1993, 115, 8221; O. S. Jung, D. H. Jo, Y. A. Lee,
Y. S. Sohn and C. G. Pierpont, Angew. Chem., Int. Ed. Engl., 1996,
35, 1694; D. M. Adams and D. N. Hendrickson, J. Am. Chem. Soc.,
1996, 118, 11515; O.-S. Jung, D. H. Jo, Y.-A. Lee, B. J. Conklin and
C. G. Pierpont, Inorg. Chem., 1997, 36, 19.
Reaction of [Pt(O,P-PPh2thqH)2(ZnBr2)] 5 with hydrobromic
acid. A sample of [Pt(O,P-PPh2thqH)(ZnBr2] (25 mg) was dis-
1
solved in d6-acetone (0.5 mL) in an NMR tube. After H and
31P-{1H} NMR spectra had been acquired, 0.1 M hydrobromic
acid (ca. 0.05 mL) was added, the solution shaken and the
NMR spectra re-run. The 1H and 31P-{1H} NMR spectra
revealed that 5 cleanly reacted to give 3.
Crystallography
10 R. A. Klein, P. Witte, R. Vanbelzen, J. Fraanje, K. Goubitz,
M. Numan, H. Schenk, J. M. Ernsting and C. I. Elsevier, Eur. J.
Inorg. Chem., 1998, 319; T. E. Keyes, R. J. Forster, P. M. Jayaweera,
C. G. Coates, J. J. Mcgarvey and J. G. Vos, Inorg. Chem., 1998, 37,
5922; M. Handa, T. Nakao, M. Mikuriya, T. Kotera, R. Nukada
and K. Kasuga, Inorg. Chem., 1998, 37, 149; T. P. Vaid, E. B.
Lobkovsky and P. T. Wolczanski, J. Am. Chem. Soc., 1997, 119,
8742; R. A. Klein, C. J. Elsevier and F. Hartl, Organometallics, 1997,
16, 1284; A. Kunzel, M. Sokolow, F.-Q. Liu, H. W. Roesky,
M. Noltemeyer, H.-G. Schmidt and I. Usón, J. Chem. Soc., Dalton
Trans., 1996, 913; S. L. Scott, A. Bakac and J. H. Espenson, J. Am.
Chem. Soc., 1992, 114, 4605; T. van der Graaf, D. J. Stufkens,
J. Vichová and A. Vlcˇek, Jr., 1991, 401, 305; S. Ernst, P. Hael,
J. Jordanov, W. Kaim, V. Kasack and E. Roth, J. Am. Chem. Soc.,
1989, 111, 1733; M. Hanaya and M. Iwaizumi, Organometallics,
1989, 8, 672; A. Vlcˇek, Jr., J. Klíma and A. A. Vlcˇek, Inorg. Chim.
Acta, 1983, 69, 191; J.-T. M. Tuchagues and D. N. Hendrickson,
Inorg. Chem., 1983, 22, 2545.
11 W. Keim, Chem.-Ing.-Tech., 1984, 56, 850; personal communication.
12 E. Drent and H. M. Budzelaar, Chem. Rev., 1996, 96, 663.
13 K. Itami, A. Palmgren, A. Thorarensen and J. E. Backvall, J. Org.
Chem., 1998, 63, 6466; H. Grennberg and J. E. Backvall, Chem. Eur.
J., 1998, 4, 1083; A. Palmgren, A. Thorarensen and J. E. Backvall,
J. Org. Chem. 1998, 63, 3764; C. Jonasson and J. E. Backvall,
Tetrahedron Lett., 1998, 39, 3601; K. Bergstad, H. Grennberg and
J. E. Backvall, Organometallics, 1998, 17, 45; J. Vicente, A. Arcas,
D. Bauutista and G. B. Shul’pin, J. Chem. Soc., Dalton Trans., 1997,
1505; R. Z. Chen and J. J. Pinatello, Environ. Sci. Technol., 1997, 31,
2399; T. Yokota, S. Fujibayashi, Y. Nishiyama, S. Sakaguchi and
Y. Ishii, J. Mol. Catal. A., 1996, 114, 113; C. Jia, P. Muller and
M. Mimoun, J. Mol. Catal. A., 1995, 101, 127; G. B. Shul’pin,
M. M. Bochkova and G. V. Nizova, J. Chem. Soc., Perkin Trans. 2,
1995, 1465.
Crystal data for [Pt(O,P-PPh2thqH)2]ؒ2dmf (2ؒ2dmf). Pt(C19-
H16O2P)2ؒ(2C3H7NO), M 955.9, triclinic, space group P1, a
¯
9.958(8), b 14.336(14), c 15.438(9) Å, α 101.40(4), β 93.85(4), γ
108.79(3)Њ, V 2025(3) Å3, Dc 1.57 g cmϪ3, Z 2, µ(Mo-Kα) 36.25
cmϪ1, T = 294 K. Crystal size 0.10 × 0.12 × 0.19 mm, 2θmax 48Њ,
minimum and maximum transmission factors 0.50 and 0.70.
The number of reflections was 5424 considered observed
[I > 3σ(I)] out of 6350 unique data, with Rmerge 0.013 for equiv-
alent reflections. Final residuals R, RЈ were 0.023, 0.032 for the
observed data.
Crystal data for [PtBr2(PPh2tq)2]ؒ0.5CH2Cl2 (4ؒ0.5CH2Cl2).
Pt(C19H15O2P)2Br2ؒ(0.5CH2Cl2), M 1010.0, monoclinic, space
group P21/c, a 10.908(5), b 17.329(5), c 20.523(10) Å, β
105.06(2)Њ, V 3746(3) Å3, Dc 1.79 g cmϪ3, Z 4, µ(Mo-Kα) 60.90
cmϪ1, T = 294 K. Crystal size 0.09 × 0.13 × 0.19 mm, 2θmax 44Њ,
minimum and maximum transmission factors 0.53 and 0.76.
The number of reflections was 3303 considered observed
[I > 3σ(I)] out of 4878 unique data, with Rmerge 0.018 for equiv-
alent reflections. Final residuals R, RЈ were 0.024, 0.031 for the
observed data.
Crystal data for [Pt(O,P-PPh2thqH)2(ZnBr2)]ؒ2dmf (5ؒ2dmf).
Pt(C19H16O2P)2ZnBr2ؒ(2C3H7NO),
M
1181.1, monoclinic,
space group P21/c, a 13.030(9), b 30.701(13), c 12.460(9) Å, β
113.81(3)Њ, V 4560(5) Å3, Dc 1.72 g cmϪ3, Z 4, µ(Mo-Kα) 54.85
cmϪ1, T = 294 K. Crystal size 0.09 × 0.19 × 0.20 mm, 2θmax 44Њ,
minimum and maximum transmission factors 0.38 and 0.51.
The number of reflections was 3569 considered observed
[I > 3σ(I)] out of 5572 unique data, with Rmerge 0.017 for
equivalent reflections. Final residuals R, RЈ were 0.033, 0.042
for the observed data.
14 M. Hudlicky, ACS Monogr., 1990, 186.
15 S. B. Sembiring, S. B. Colbran and D. C. Craig, Inorg. Chem., 1995,
34, 761.
16 S. B. Sembiring, S. B. Colbran, D. C. Craig and M. L. Scudder,
J. Chem. Soc., Dalton Trans., 1995, 3731.
J. Chem. Soc., Dalton Trans., 1999, 1543–1554
1553

View Article Online
17 S. B. Sembiring, S. B. Colbran, R. Bishop and A. D. Rae, Inorg.
Chim. Acta, 1995, 228, 109.
18 S. B. Sembiring, S. B. Colbran and L. R. Hanton, Inorg. Chim. Acta,
1992, 202, 67.
19 S. B. Sembiring, S. B. Colbran and D. C. Craig, Inorg. Chim. Acta,
1990, 176, 225.
29 B. R. Eggins and J. Q. Chambers, J. Electrochem. Soc., 1970, 117,
186; N. Gupta and H. Linshitz, J. Am. Chem. Soc., 1997, 119, 6384;
J. Q. Chambers in ref. 1, vol. I, ch. 14 and vol. II, ch. 12.
30 P. Castellonese and G. Launay, Bull. Soc. Chim. Fr., 1978, 7/8, I317;
B. E. Conway, Y. Phillips and S. Y. Qian, J. Chem. Soc., Faraday
Trans., 1995, 283.
20 S. B. Sembiring, Zainal, S. B. Colbran and D. C. Craig, manuscript
in preparation.
21 K. V. Baker, J. M. Brown, N. Hughes, A. J. Skarnulis and A. Sexton,
J. Org. Chem., 1991, 56, 698.
22 J. A. Rahn, D. O. O’Donnell, A. R. Palmer and J. H. Nelson, Inorg.
Chem., 1989, 28, 2631.
31 G. Mazzocchin, G. Bontempelli, M. Nicolini and B. Crociani,
Inorg. Chim. Acta, 1976, 18, 159; J. A. Davies and V. Uma,
Inorg. Chim. Acta, 1983, 76, L305; J. Electroanal. Chem., 1983, 158,
13; G.-A. Mazzocchin and G. Bontempelli, J. Electroanal. Chem.,
1984, 179, 269; J. A. Davies and V. Uma, J. Electroanal. Chem., 1984,
179, 273.
23 H. D. Empsall, B. L. Shaw and B. L. Turtle, J. Chem. Soc., Dalton
Trans., 1976, 1500; T. B. Rauchfuss, Inorg. Chem., 1977, 16, 2967;
H. D. Empsall, P. N. Heys and B. L. Shaw, J. Chem. Soc., Dalton
Trans., 1978, 257; P. Braunstein, D. Matt, D. Nobel, F. Balegroune,
S.-E. Bouaoud, D. Grandjean and J. Fischer, J. Chem. Soc., Dalton
Trans., 1988, 353; A. W. G. Platt and P. Pringle, J. Chem. Soc.,
Dalton Trans., 1989, 1193; P. Bergamini, S. Sostero, O. Traveso,
T. J. Kemp and P. Pringle, J. Chem. Soc., Dalton Trans., 1989, 2017;
L. Kollar and G. S. Zalontai, J. Organomet. Chem., 1991, 421, 341;
E. M. Georgiev, H. tom Dieck, G. Hahn, G. Petrov and M. Kirilov,
J. Chem. Soc., Dalton Trans., 1992, 1311.
24 C. D. Montgomery, N. C. Payne and C. J. Willis, Inorg. Chem., 1987,
26, 519; N. W. Alcock, A. W. G. Platt and P. Pringle, J. Chem. Soc.,
Dalton Trans., 1987, 2273; Inorg. Chim. Acta, 1987, 128, 215;
J. Chem. Soc., Dalton Trans., 1989, 139; A. Varshney, M. L. Webster
and G. M. Gray, Inorg. Chem., 1992, 31, 2580.
32 A. Zweig, W. G. Hodgson and W. H. Jura, J. Am. Chem. Soc.,
1964, 86, 4124; V. Le Berre, L. Angely, J. Simonet, G. Mousset and
M. Bellec, J. Electroanal. Chem., 1987, 218, 173.
33 A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fund-
amentals and Applications, Wiley, New York, 1980.
34 (a) B. Yang, L. Liu, T. J. Katz, C. A. Liberko and L. R. Miller, J. Am.
Chem. Soc., 1991, 113, 8993; (b) A. L. Sargent, J. Almlof and
C. A. Liberko, J. Phys. Chem., 1994, 98, 6114; (c) S. Hunig,
K. Sinzger, M. Kemmer, U. Langohr, H. Rieder, S. Soderholm,
J. U. Vonschutz and H. C. Wolf, Eur. J. Org. Chem., 1998, 9, 1977.
35 CS Chem3DProTM V3.2, Cambridgesoft Corp., Cambridge, MA,
1996.
36 P. S. Rao and E. Hayon, J. Phys. Chem., 1973, 77, 2274.
37 D. M. Roundhill, Comprehensive Coordination Chemistry, Pergamon,
Oxford, 1987, vol. 5, ch. 52, pp. 351–532.
38 T. Itoh, Chem. Rev., 1995, 95, 2351.
25 F. H. Allen and S. N. Sze, J. Chem. Soc. A, 1971, 2054.
26 C. K. Johnson, ORTEP II, report ORNL-5138, Oak Ridge National
Laboratory, Oak Ridge, TN, 1976.
39 E. T. Denisov and I. V. Khudyakov, Chem. Rev., 1987, 87, 1313.
40 IGORPRO 2.0TM, Wavemetrics Inc., Lake Oswego, OR, 1995.
41 K. Yoshida, M. Shigi and T Fueno, J. Org. Chem., 1975, 40, 63.
42 A. T. Shulgin and E. M. Gal, J. Chem. Soc., 1953, 1316.
27 F. Rameriz, Pure Appl. Chem., 1964, 9, 337.
28 J. Andrieu, P. Braunstein, M. Drillon, Y. Dusausoy, F. Ingold,
P. Rabu, A. Tiripicchio and F. Ugozzoli, Inorg. Chem., 1996, 35,
5986.
Paper 9/00610I
1554
J. Chem. Soc., Dalton Trans., 1999, 1543–1554