Syntheses and electrochemistry of (p-XC6H4O) 6W (1-X, X = H, CH3, OCH3, Cl, Br, OH, OCH 2Ph) and (p-XC6H4O)5W(OC 6H4OH) (X = H, CH3, OCH3, Cl, Br): An approach to electrocatalytic CH bond activation

Article abstract of DOI:10.1016/j.poly.2004.08.006

Hexaphenoxides (p-XC₆H₄O)₆W (1-X, X = H, CH₃, OCH₃, OCH₂Ph, Cl, Br) serve as starting materials for (p-XC₆H₄O)₅W(OC₆H ₄OH) (6-X, X = OH, OMe, Me, H, Br, Cl), whose electrochemical oxidations were studied in reference to CH bond activation. Alcoholysis of W(OMe)₆ afforded (p-PhCH₂OC₆H ₄O)₆W (1-OCH₂Ph), which could be hydrogenated (10% Pd/C, 1 atm H₂) to prepare (p-HOC₆H ₄O)₆W (1-OH). Related alcoholyses of WCl₆ with HOC₆H₄-p-X provided the hexaphenoxides (p-XC ₆H₄O)₆W (1-X, X = H, CH₃, OCH ₃, Cl, Br) through minor modifications of literature procedures. Acid catalyzed treatment of 1-X with p-HOC₆H₄OCH₂Ph provided a mixture of substitution products (p-XC₆H ₄O)_6-xW(OC₆H₄OCH₂Ph) _x (x = 1, 5-X) that could be hydrogenated (10% Pd/C, 1 atm H ₂) to a mixture of hydroxylated products. Chromatography permitted isolation of (p-XC₆H₄O)₅W(OC₆H ₄OH) (6-X, X = H, 19%; Me, 29%; OCH₃, 19%; Cl, 12%; Br, 11%) in modest yields. Hexaphenoxides 1-X and 6-X manifested two electrochemical reduction waves whose positions were a function of para-substituent. When oxidized, 6-X and 1-OH were proposed to behave as W(V)quinone mimics, albeit at potentials capable of oxidizing hydrocarbons as shown via a thermochemical cycle. If the proposed transients (p-XC₆H₄O) ₅W(OC₆H₄O) (7-X) were generated, degradation was apparently competitive with CH bond activation. The structure of oligomeric {K[(p-ClC₆H₄O)₆W]}_∞ (8-Cl) is addressed, and comments on the nature of radical CH bond activation in this and related systems are presented.

Full text of DOI:10.1016/j.poly.2004.08.006

Polyhedron 23 (2004) 2841–2856
www.elsevier.com/locate/poly
Syntheses and electrochemistry of (p-XC6H4O)6W
(1-X, X = H, CH3, OCH3, Cl, Br, OH, OCH2Ph) and
(p-XC6H4O)5W(OC6H4OH) (X = H, CH3, OCH3, Cl, Br):
an approach to electrocatalytic CH bond activation
Orson L. Sydora, Jonas I. Goldsmith, Thomas P. Vaid, Abigail E. Miller,
*
Peter T. Wolczanski , Hector D. Abrun˜a
Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell University Ithaca, NY 14853, USA
Received 21 May 2004; accepted 25 August 2004
Available online 25 September 2004
Abstract
Alcoholysis of W(OMe)6 afforded (p-PhCH2OC6H4O)6W (1-OCH2Ph), which could be hydrogenated (10% Pd/C, 1 atm H2) to
prepare (p-HOC6H4O)6W (1-OH). Related alcoholyses of WCl6 with HOC6H4-p-X provided the hexaphenoxides (p-XC6H4O)6W
(1-X, X = H, CH3, OCH3, Cl, Br) through minor modifications of literature procedures. Acid catalyzed treatment of 1-X with p-
HOC6H4OCH2Ph provided a mixture of substitution products (p-XC6H4O)6ꢀxW(OC6H4OCH2Ph)x (x = 1, 5-X) that could be
hydrogenated (10% Pd/C, 1 atm H2) to a mixture of hydroxylated products. Chromatography permitted isolation of (p-XC6H4O)5-
W(OC6H4OH) (6-X, X = H, 19%; Me, 29%; OCH3, 19%; Cl, 12%; Br, 11%) in modest yields. Hexaphenoxides 1-X and 6-X man-
ifested two electrochemical reduction waves whose positions were a function of para-substituent. When oxidized, 6-X and 1-OH
were proposed to behave as W(V)quinone mimics, albeit at potentials capable of oxidizing hydrocarbons as shown via a thermo-
chemical cycle. If the proposed transients (p-XC6H4O)5W(OC6H4O) (7-X) were generated, degradation was apparently competitive
with CH bond activation. The structure of oligomeric {K[(p-ClC6H4O)6W]} (8-Cl) is addressed, and comments on the nature of
1
radical CH bond activation in this and related systems are presented.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
only autoxidation processes mediated by cobalt and
manganese have been shown to be commercially viable,
and these are rather specialized applications (e.g.,
The direct conversion of alkanes to functionalized
products is a tremendously important, yet very difficult
chemical transformation facing the organotransition
metal community. The stoichiometric activation of car-
bon–hydrogen bonds has now been achieved in numer-
ous systems [1,2], but the capability to catalyze this
reaction concomitant with oxidation of the substrate
has rarely been successfully realized [3–14]. To date,
t
cC6H12 ! HO2C(CH2)4CO2H, tBuH ! BuOOH, p-
MeC6H4Me ! p-HO2CC6H4CO2H, etc.) that have a
minimum of selectivity issues [15,16].
Organometallic approaches to C–H bond activation
have generated at least one major conceptual advance.
In stark contrast to known free radical chemistry,
organometallic systems tend to oxidatively or otherwise
add stronger CH bonds selectively over weaker. While
some have argued for steric interpretations of these
events, it seems clear that this CH bond activation selec-
tivity stems from the greater D(M–R) derived from
*
Corresponding author. Tel.: +1 607 255 7220; fax: +1 607 255
4137.
E-mail address: ptw2@cornell.edu (P.T. Wolczanski).
0277-5387/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2004.08.006

2842
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
activation of the stronger bond. For example, in Eqs. (1)
and (2), if DG0rxn ꢁ DHr0xn and the bonds in LnMH (Eq.
(1)) or LnMXH (Eq. (2)) are relatively unaffected by
R1 versus R2, then DHr0xn ¼ fDðM ꢀ R2Þ ꢀ DðMR1Þg
þfDðR1HÞ ꢀ DðR2HÞg. As long as the differences in
D(M–Rn) are greater than the differences in D(H–Rn)
(i.e., DD(M–Rn)/DD(H–Rn) > 1), selectivities opposite
to that of free radical activations prevail: sp-CH > Ar-
H P sp2-CH, including cPrH > MeH > 1ꢁsp3-CH > 2ꢁ
sp3-CH > 3ꢁsp3-CH, with benzylic and allylic CH bonds
often modest exceptions [17–19]. It seems clear that
when Green [20] discovered the proclivity of transiently
generated tungstenocene to activate arene CH bonds
(Eq. (3)), it was this selectivity that was first being wit-
nessed. In most systems involving direct oxidative addi-
tion of carbon–hydrogen bonds, this selectivity is
modest because binding of RnH precedes C–H bond
cleavage [17,18]. In cases where the actual CH activation
is rate-determining [20], there is considerable optimism
that problems such as methane to methanol and acetic
acid [6,7], and terminal versus internal sp3-CH activa-
tion and functionalization of acyclic alkanes will ulti-
mately be solved [13,14].
The thermodynamic feasibility of CH bond activa-
tion can be predicted with a thermochemical cycle de-
vised by Parker and Tilset that use physically
measurable parameters ði:e:; E0ox; pKa; CÞ to determine
the homolytic bond dissociation energies (BDEs) of
X–H species [28]. Fig. 1 illustrates such a cycle in which
CH bond activation can be done electrocatalytically [4]
provided the reactive intermediate, LnMX, is generated
at a potential positive enough to abstract a hydrogen
atom. The sequence gives the BDE (sol) of the X–H
bond, which constitutes a reasonable thermodynamic
limit of the CH bond that can be broken through H
atom abstraction by LnMX. Mayer [29] has used similar
cycles in rationalizing oxidations of permanganate [30]
and MMO mimics [31]. Herein is explored the genera-
tion of a WVI(semiquinone)/WV(quinone) as a function-
ality that should be capable of the desired abstraction.
Unfortunately, this specific system failed to generate
products of CH activation, but a closer look at its fea-
tures yields some surprises regarding CH bond activa-
tions by LnMX.
2. Results
LnMðHÞR2 þ R1H ꢀ LnMðHÞR1 þ R2H
LnMðXHÞR2 þ R1H ꢀ LnMðXHÞR1 þ R2H
ð1Þ
ð2Þ
2.1. Syntheses
2.1.1. (p-PhCH2OC6H4O)6W (1-OCH2Ph) and (p-HO-
C6H4O)6W (1-OH)
hm
PhH
Cp2WH2 !½Cp2Wꢂ ! Cp2WðHÞPh
ð3Þ
In an earlier project designed with the synthesis of
inorganic dendrimers, the robust character of the hexa-
aryloxidetungsten coordination environment was
exploited as shown in Scheme 1. Treatment of
W(OMe)6, prepared from W(OMe)4Cl2, MeOH and
NEt3 [32], with p-HOC6H4OCH2Ph in toluene for 24 h
Concerted types of C–H bond activations have been
separated into various classes such as r-bond metathe-
sis, oxidative addition, 1,2-RH-addition and electrophi-
lic activation [1]. While each represents a particular class
of reactions, it is clear that electrophilic attack at the C–
H bond, whether in a binding or cleavage event, is a
common thread among these paths. Free radical and
metalaradical (e.g., LnM + RH ! LnMH + Rꢁ, LnMX
+ RH ! LnMXH + Rꢁ, etc.) mechanisms for alkane
activation may also yield some control, provided selec-
tivity is built into the metal framework as in nature, or
the precise D(RnH) limits of activation can be controlled
and utilized.
Electrocatalytic Metallaradical
+
-
L MX
n
RH
H
+ e
23.06 E˚
H(sol)
Eox
Many bioinorganic systems, most notably methane
monoxygenase [21], cytochrome P-450 [22], and related
oxidases [23–25], have been examined in the context of
CH bond activation. Some exciting recent results have
been obtained by Que, who has employed the tetram-
ethylcyclam ligand in stabilizing an Fe(IV) oxo complex
[26], and has expanded his ligand arsenal to pentaden-
tate ligands whose ligated FeIV@O groups are capable
of hydroxylating cyclohexane and substrates containing
weaker CH bonds [27]. Note that the metal functionality
(MX) that reacts with RH need not be an oxo or multi-
ple-bonded entity, but the transfer of a hydrogen atom
to MX must satisfy certain thermodynamic require-
ments for it to occur with reasonable facility.
.
L MX-H
n
R
+
L MXH
n
-
L MX-H
+
e
23.06 E˚
n
ox(MXH)
+
+
2.303RT pK
L MXH
n
L MX
n
+
H (sol)
a
.
+
-23.06 E˚
red
H (sol)
H (sol)
.
BDFE (sol)
L MX-H
n
L MX
n
+
H (sol)
with appropriate assumptions and entropic corrections:
BDE (sol) = 23.06 E˚ + 1.36 pK + C
ox(MXH)
a
(for CH CN, C ~ 57(2) kcal/mol)
3
@ +1.4 V, if pK ~ 0, BDE (sol) ~ 89(2) kcal/mol
a
@ +2.1 V, if pK ~ 0, BDE (sol) ~ 105(2) kcal/mol
a
Fig. 1. Electrocatalytic metallaradical.

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2843
at 213 ꢁC in a bomb reactor yielded the polybenzyl
ether, (p-PhCH2OC6H4O)6W (1-OCH2Ph, 91%). Depro-
tection via hydrogenation (H2 (1 atm), Pd/C) in ethanol/
toluene ultimately yielded dark red (p-HOC6H4O)6-
WÆ2THFÆC6H6 ((1-OH)Æ2THFÆC6H6) upon crystalli-
zation from THF/benzene. A structural investigation
of (1-OH)Æ2THFÆC6H6 revealed a dense 3-D hydro-
gen-bonding network linking phenolic and THF sites.
The addition of Lewis bases such as 4,40-bipyridine
and 1,2-di-4-pyridylethane to 1-OH led to the hexabasic
mononuclear complex (4,40-bipyÆHOC6H4O)6W (2), the
doubly interpenetrating 3-D network [{W(OC6H4-
OH)6}{4,40-(NC5H4)2(CH2CH2)}2]1 (3), and the triply
interpenetrating 3-D network [{W(OC6H4OH)6}{4,40-
est variants of Funk and Baumann [35], and Mortimer
and Strong [36]. Treatment of tungsten hexachloride
with ꢁ7 equiv. of p-XC6H4OH in refluxing chloroben-
zene for 2 d formed (p-XC6H4O)6W (1-X, X = H,
82%; CH3, 60%; OCH3, 40%; Cl, 87%; Br, 68%) in var-
ying yields.
X
X
O
135˚C
WCl
+
6
O
2d, -6 HCl
X
O
W
O
O
X
ClC H
~7
X
6
5
O
OH
1-X
X
X
(NC5H4)2(CH2CH2)}3 ÆTHF] (4ÆTHF), according to
1
X-ray crystallographic studies [33].
ð4Þ
Since an entire [(p-XC6H4O)5W] unit was consid-
ered as a tunable group for the ensuing electrochemi-
cal study, a protocol for the replacement of one
phenoxide from 1-X was needed. As Scheme 2 reveals,
(p-XC6H4O)6W reacted with HOC6H4OCH2Ph in
refluxing chlorobenzene with catalytic tosylic acid to
While the network solids were intriguing, the stabili-
zation of pendant hydroxy functionalities on a high va-
lent early transition metal center was a striking
observation. Earlier studies of group 4 and 5 derivatives
suggested that self-condensation was prevalent, yet no
oligomerization chemistry was evident for (p-HO-
C6H4O)6W (1-OH). Now that the hydroxy functional
group could be engineered into the W(VI) center, could
reactivity other than hydrogen bond formation be
exploited?
yield
a
mixture of substituted products, (p-
XC6H4O)6ꢀxW(OC6H4O-CH2Ph)x (x = 1, 5-X), as
determined by TLC. The mixture of substituted prod-
ucts was hydrogenated (10% Pd/C, 1 atm H2) in an
EtOH/toluene mixture to a mixture of hydroxylated
products, (p-XC6H4O)6ꢀxW(OC6H4OH)x (x = 1, 6-X),
also determined by TLC. The mono-hydroxylated spe-
cies 6-X were separated from the mixture by column
chromatography and recrystallized from ethanol
(X = H, 19%; Me, 29%; OCH3, 19%; Cl, 12%; Br,
11%).
2.1.2. (p-XC6H4O)6W (1-X), (p-XC6H4O)5W(OC6H4-
OCH2Ph) (5-X) and (p-XC6H4O)5W(OC6H4OH) (6-
X, X = H, CH3, OCH3, Cl, Br)
para-Substituted tungsten hexaphenoxide complexes
were synthesized via modification of methods developed
by Beshouri and Rothwell [34], which in turn were mod-
W(OMe)
6
O
O
213ÞC
+
O
O
toluene
-6 MeOH
6
HO
O
O
O
W
O
O
O
O
O
H (1 atm), Pd/C
2
O
1-OCH Ph
EtOH/toluene
24 h, 23ÞC
2
HO
OH
N
N
.
W{(OC H OH) (4,4'-bipy)}
6
2
1-OH
6
4
O
W
O
O
O
OH
O
HO
3
4
O
[{W(OC H OH) }{(4,4'-NC H ) (C H )} ]
6 4 6 5 4 2 2 4 2
+
N
N
HO
[{W(OC H OH) }{(4,4'-NC H ) (C H )} ]
6
4
6
5
4 2
2 4 3
OH
Scheme 1.

2844
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
X
X
X
X
1-X
TosH
O
W
O
O
O
O
135˚C
X
O
O
X
X
O
O
W
O
O
ClC H
6
5
O
O
-HOC H X
6
4
X
X
5-X
X
X
+
1.25 HO
O
H (1 atm), Pd/C
2
+ other substitution products
EtOH/toluene
6 h, 23ÞC
X
X
X
X
O
O
O
O
column
X
O
W
O
OH
X
O
W
O
O
OH
chromatography
O
O
O
X
X
X
X
X = OMe, 19%
= Me, 29%
= H, 19%
+ other substitution products
6-X
= Cl, 12%
= Br, 11%
Scheme 2.
2.2. Electrochemistry
2.2.1. Quinone
X
X
6-X
O
The 2-electron oxidation of hydroquinone in aceto-
nitrile is irreversible and produces quinone with the re-
lease of two protons as shown in Eq. (5) The first step
of this oxidation is a one electron oxidation of hydroqui-
none (H2Q) to the radical cation (H2Q+, Eq. (6)) fol-
lowed by rapid proton loss to form the resonance
stabilized semiquinodal radical (HQ, Eq. (7)) [37,38].
If the [(p-XC6H4O)5W]+ unit of (p-XC6H4O)5W(O-
C6H4OH) (6-X) is considered as a proton ‘‘replace-
ment’’, than an analogous one electron oxidation
pathway to a W(VI) semiquinone/W(V) quinone species
can be envisaged (Eq. (8)).
O
X
OH
O
W
O
O
O
X
X
X
X
O
O
E˚
-e
ox
X
O
W
O
O
CH CN
3
-
+
-
HO
OH
O
O + 2 H + 2 e
O
+
-H
E˚ = + 1.4 V
ox
(vs. NHE)
O
7-X
X
ð5Þ
X
ð8Þ
Application of the cycle shown in Fig. 1 to hydro-
quinone (HQ = LnMXH) indicates that HQ
CH CN
3
+
-
HO
OH
HO
OH + e
+
H Q
2
H Q
2
ðE0ox ꢁ þ1:4 V versus NHEÞ could activate a CH bond
of <89 (2) kcal/mol. However, if the E0ox potential of a
hydroquinone like system could be raised to about
+2.1 V (versus NHE), then activation of a 105(2)
kcal/mol CH bond is possible. Would the para-substit-
uents on 6-X yield such an increase in oxidation
potential?
ð6Þ
CH CN
+
OH
3
+
HO
HO
HQ
O
+ H
+
H Q
2
ð7Þ

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2845
[40]. Lau and coworkers have found that in both water
and acetonitrile oxidation of p-XC6H4OH (X = MeO,
tBu, Me, Cl, CN, H) by trans-[RuVI(L)(O)2]2+
(L = 1,12-dimethyl-3,4:9,10-dibenzo-1,12-diaza-5,8-dioxa-
cyclopentadecane) proceeds via
a hydrogen atom
abstraction mechanism. The unstable phenoxy radical
was further oxidized to p-benzoquinone and 4,40-biphe-
noquinone via rapid competing steps [41]. Electrochem-
ical studies on the mechanism of phenol oxidation
suggest that the first step is an irreversible one electron
oxidation of phenol followed by proton loss forming
the resonance stabilized phenoxy radical shown in Eq.
(12) [42]
Fig. 2. Cyclic voltammogram of hydroquinone in acetonitrile: 0.5 mM
QH2, 0.1 M TBAH (tetra-n-butylammonium hexafluorophosphate),
Ag wire reference, 100 mV/s.
Fig. 2 shows the oxidation of hydroquinone at plati-
num electrodes in acetonitrile. The peak potential of the
irreversible oxidation wave (wave I, +1.40 V versus
NHE) and irreversible reduction wave (wave II,
+0.55 V versus NHE) were similar to the values reported
by Chambers and Eggins [37] (wave I, +1.36 V; wave II,
+0.37 V versus NHE, 2 V/s), who found that the peak
potential and width for wave II was highly dependent
on scan rate and pH. Further investigation into the
behavior of wave I led to a proposed mechanism where-
by reversible one electron oxidation of hydroquinone
(QH2) forms the radical cation QH2+ (Eq. (6)) followed
by rapid, irreversible deprotonation to the protonated
semiquinone QH (Eq. (7)). A hemiketal intermediate
(QH)2 then forms via HQ dimerization (Eq. (9)), fol-
lowed by disproportionation to yield Q and H2Q (Eq.
(10))
-
+
-e ,-H
OH
O
CH CN
3
O
O
ð12Þ
2.2.3. (p-XC6H4O)6W (1-X, X = H, CH3, OCH3, Cl, Br,
OH), (p-XC6H4O)5W(OC6H4OH), and p-XC6H4OH
The impetus for attempting this project stemmed
from three observations. First, cyclic voltammetry on
(p-HOC6H4O)6W (1-OH) revealed multiple irreversible
oxidation waves at ꢁ1.29 V (versus NHE) and 1.63 V
(Table 1). According to Fig. 1, if these events were indic-
ative of a process such as that in Eq. (8), which is ex-
pected to be irreversible in analogy to H2Q/Q
electrochemistry, then C–H bonds of considerable
strength could conceivably be attacked. In order to
simplify the system, and to modulate the oxidation
potentials of interest through para-substitution, (p-
XC6H4O)5W(OC6H4OH) (6-X, X = H, CH3, OCH3,
Cl, Br) were synthesized. Second, a 0.0–2.3 V (versus
Ag wire) cyclic voltammogram of (p-BrC6H4O)5W(O-
C6H4OH) (6-Br), which was chosen because it possesses
a second oxidation wave at ꢁ+1.95 V (versus NHE),
showed a pronounced increase in current when excess
Ph3CH (3–10 equiv.; BDE ꢁ 81 kcal/mol) [43] was
added to the solution. In concert with this experiment,
the addition of dihydroanthracene (BDE ꢁ 79 kcal/
mol) [31] to a solution of (p-ClC6H4O)5W(OC6H4OH)
(6-Cl) caused a similar increase in current at ꢁ+1.9 V
(versus NHE) during another 0.0–2.3 V scan.
OH
CH CN
3
2 HO
O
O
O
(QH)
2
HQ
OH
ð9Þ
OH
HO
OH
H
Q
CH CN
3
3
+
O
O
(QH)
2
OH
O
O
Q
ð10Þ
Wave II is the irreversible two electron reduction of
the Q/H+ mixture back to hydroquinone (Eq. (11)) [9]
eꢀ
Hþ
Q þ Hþ þ eꢀ ! HQ ! QHꢀ ! QH2
ð11Þ
2.2.2. Phenols
The oxidation of phenols can lead to a variety of
products depending on the oxidant, catalyst, or substitu-
tion pattern of the phenol [39]. Treatment of 2,6-xylenol
with a copper pre-catalyst in the presence of dioxygen
yielded either diphenoquinone or polyphenylene ether
depending on the pyridine to cuprous chloride ratio
In view of the possibility of electrocatalytic C–H
bond activation, some other cases seemed puzzling. Un-
der similar conditions, no evidence of toluene
(BDE ꢁ 89.8(6) kcal/mol)
[43]
or acetonitrile

2846
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
Table 1
E0red and E0ox values (V vs. NHE, CH3CN)a for p-XC6H4OH, (p-XC6H4O)6W (1-X) and (p-XC6H4O)5W(OC6H4OH) (6-X); X = H, Me, OMe, Cl, Br
Compound vs. NHE (V)
Ered(1)
Ered(2)
Eox(1)
Eox(2)b
(p-BrC6H4O)6W (1-Br)
(p-BrC6H4O)5W(OC6H4OH) (6-Br)
p-BrC6H4OH
ꢀ0.17
ꢀ0.16
ꢀ1.61
ꢀ1.22
ꢀ1.19
1.88c
1.32b
2.03
2.2
1.95
(p-ClC6H4O)6W (1-Cl)
(p-ClC6H4O)5W(OC6H4OH) (6-Cl)
p-ClC6H4OH
ꢀ0.17
ꢀ0.28
ꢀ1.25
ꢀ1.28
ꢀ1.35
1.84
1.41
1.95
2.2
1.91
(C6H5O)6W (1-H)
(C6H5O)5W(OC6H4OH) (6-H)
C6H5OH
ꢀ0.4
ꢀ1.6
1.83
1.45
2.24
d
d
ꢀ0.26
ꢀ1.89
ꢀ0.59
ꢀ0.54
ꢀ1.85
e
f
(p-MeC6H4O)6W (1-Me)
(p-MeC6H4O)5W(OC6H4OH) (6-Me)
p-MeC6H4OH
ꢀ1.70
ꢀ1.62
1.47
1.26
1.73
(p-MeOC6H4O)6W (1-OMe)
(p-MeOC6H4O)5W(OC6H4OH) (6-OMe)
p-MeOC6H4OH
ꢀ0.54
ꢀ0.56
ꢀ1.57
ꢀ1.5
1.33
1.17
1.52
1.71g
1.72g
ꢀ1.63
(p-HOC6H4O)6W (1-OH)
p-HOC6H4OH
ꢀ0.58
ꢀ1.75
1.29c
1.40
1.63
0.55
a
From cyclic volatammetry referenced to Ag wire with 0.1 M added tetra-n-butylammonium hexafluorophosphate (TBAH); the tabulated
values are vs. NHE because the thermodynamic quantities in Fig. 1 use NHE as a reference. Typical samples were ꢁ0.5 mM, and all scan rates were
100 mV/s.
b
Appear as shoulders atop a generally featureless, increasing current.
Multiple waves.
c
d
Scan was only from ꢀ1.6 to +1.8 V.
e
Possible shoulder at +1.9 V.
Scan to 2.0 V, current still increasing.
Other minor waves visible.
f
g
(BDE ꢁ 94.8(21) kcal/mol) [43] activation by any 6-X
species was noted, despite oxidation potentials that sug-
gested such events were feasible according to the calcu-
lations of Fig. 1. It was also plausible that toluene might
be activated by hydride transfer, but while a cycle simi-
lar to that in Fig. 1 can be constructed [44], not enough
thermodynamic parameters are known for RH ! R+
+ Hꢀ to determine its feasibility. Regardless, cyclic vol-
tammograms of 6-X in the presence of toluene were un-
changed. Finally, Fig. 4 illustrates another CV of (p-
BrC6H4O)5W(OC6H4OH) (6-Br) taken from ꢀ2.2 to
+2.3 V; when cycled in the presence of Ph3CH, this vol-
tammogram was unchanged. Whatever was generated in
Fig. 3 does not persist when the scan is taken to ꢀ2.2
V for the longer period associated with it (Fig. 4).
Table 1 lists the electrochemical reduction and
oxidation waves attributed to (p-XC6H4O)6W (1-X,
X = H, CH3, OCH3, Cl, Br, OH), (p-XC6H4O)5W(O-
C6H4OH) (6-X), and p-XC6H4OH. In general, the
phenols exhibit an irreversible oxidation wave (two
in some cases) that increases in potential according
to substituent from p-donors to r-donors to r-with-
drawing groups: p-X = OH (easiest) < OMe < Me
< Cl < Br < H. The oxidation wave appears as a fea-
tureless increase in current that is initiated at high pot-
entials. The differences based on substituent are
substantial, span 0.84 V, and the parent phenol is oxi-
dized at an anomalously high potential of +2.24 V.
Presumably Cl < Br because the former has some p-in-
fluence that contradicts its r-effect. Ignoring hydroqui-
none and its special characteristics (vide supra), the
phenols also exhibit a quasi-reversible reduction wave.
Here the inductive contributions appear to play a
greater role than any p-effects because the ease of
reduction follows p-X = Cl (easiest) > OMe > Br
> Me > H, and the range of potentials is 0.64 V.
p-Type electrons on substituents are energetically clo-
ser to arene p-bonding orbitals than arene p*-orbitals.
As a consequence, p-effects play a greater role facilitat-
ing oxidation of the phenols. The reduction trend sim-
Fig. 3. Cyclic voltammogram of (p-BrC6H4O)5W(OC6H4OH) (6-Br)
in CH3CN showing the increase in current in successive scans after
addition of Ph3CH (ꢁ5 mM): ꢁ0.5 mM 6-Br, 0.1 M TBAH (nBu4PF6),
Ag wire reference, 100 mV/s.

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2847
Fig. 4. Cyclic voltammogram of (p-BrC6H4O)5W(OC6H4OH) (6-Br) in acetonitrile from ꢀ2.2 to +2.3 V: ꢁ0.5 mM 6-Br, 0.1 M TBAH (tetra-n-
butylammonium hexafluorophosphate), Ag wire reference, 100 mV/s.
ply reflects inductive effects, which will tend to lower
both the donor p-orbitals and the receptor p*-orbitals
of the phenols as the substituents become more elec-
tron-withdrawing.
oxidation, there is a large increase in current, often with
no readily discernable peak. The second oxidation
(Eox(2)), if it is clear, appears as a wave atop this in-
crease. In the cyclic voltammogram of 1-Br shown in
Fig. 5, both oxidation waves are atop the current in-
crease. The Eox(2) values, when clearly observed, reflect
the previous trend. The similarities in Eox(2) and the oxi-
dation waves of the parent phenols strongly suggest that
the current increase observed is due to degradation of
1-X and the release of phenol via phenoxy radical
As previously observed for related compounds by
Beshouri and Rothwell [34], (p-XC6H4O)6W (1-X,
X = H, CH3, OCH3, Cl, Br, OH) exhibit two reversible
reduction waves corresponding to the formation of ani-
ons [1-X]ꢀ and dianions [1-X]2ꢀ. In contrast to the
phenols, these reductions reflect a substantial influence
by p-components, which render the reductions more dif-
ficult. The trend of ease of the first reduction for 1-X is
p-X = Br ꢁ Cl (easiest) > H > OMe > OH > Me, and
the range of potentials is 0.42 V. Inductively, the oxy-
genated substituents should be on par with Cl, but their
p-donor capability renders them more difficult to re-
duce. The trend of the second reduction has roughly
the same range (0.53 V) and the trend is also similar:
p-X = Br (easiest) > Cl > OMe > H > Me > OH. p-Do-
nors increase the p-donating ability of the phenoxides,
which results in an increase in energy of the t2g receptor
orbitals on the pseudo-octahedral complexes. This coun-
teracts the tendency of r-withdrawing substituents to
lower the t2g set. The reductions can certainly be ration-
alized as metal in character, and quite different from the
phenols.
The oxidations of (p-XC6H4O)6W (1-X, X = H, CH3,
OCH3, Cl, Br, OH) are complicated. A relatively clean
irreversible wave (Eox(1)) occurs for p-X = OH <
OMe < Me < H ꢁ Cl < Br, which is basically the same
trend – corresponding to the same effects – seen for
the aforementioned phenol oxidations. The values are
slightly less positive by 0.1–0.4 V, consistent with re-
moval of an electron from ligand-based orbitals that
are more ‘‘phenolate’’ in character. During and after this
X
X
X
X
O
O
-e
O
O
O
O
X
X
X
W
O
X
O
O
W
O
O
O
X
X
X
X
X
X
O
MeCN
O
N
CMe
O
X
W
O
O
O
X
X
X
1-X, X = H, CH3, OCH3, Cl, Br, OH
ð13Þ
The reductions of (p-XC6H4O)5W(OC6H4OH) (6-X,
X = H, CH3, OCH3, Cl, Br) are essentially the same as

2848
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
Fig. 5. Cyclic voltammogram of (p-BrC6H4O)6W (1-Br) in acetonitrile from ꢀ2.2 to +2.3 V: ꢁ0.5 mM 1-Br, 0.1 M TBAH (tetra-n-butylammonium
hexafluorophosphate), Ag wire reference, 100 mV/s.
those of 1-X within 0.1 V. The oxidations appear to re-
veal the problem in implementing electrocatalytic C–H
bond activation. Each complex has an irreversible oxi-
dation around +1.2 to 1.5 V that appears essentially at
the onset of a large current increase. This is likely to
be the oxidation of the p-HOC6H4O-ligand, since it is
very close to the oxidations of 1-OH and that of hydro-
quinone. The second oxidation, when observed atop the
current increase, is near that of Eox(2) of 1-X and the
phenols. It is plausible that initial oxidation occurs at
the p-HOC6H4O-ligand, perhaps via a process like that
revealed in Eq. (8). If the W(V) quinone product (or
W(VI) semiquinone) degrades before H atom abstrac-
tion can occur to regenerate 6-X, degradation would
lead to phenol loss. While the initial oxidation wave
can be attributed to the HOC6H4O-ligand, a wave that
would correspond to the reduction of free quinone is
not evident, thus it is likely that protons generated upon
oxidation are playing a role in the degradation.
W(V) quinone/W(VI) semiquinone complex, (p-
XC6H4O)5W(OC6H4O) (7-X, X = H, CH3, OCH3, Cl,
Br, OH) was generated yet unstable under the electro-
chemical conditions. Early work suggested that the in
situ preparation of 7-X could be undertaken by utilizing
a W(V) precursor. Funk et al. indicated that the reduc-
tion of (p-XC6H4O)6W (1-X, X = H, CH3, Cl) by Raney
nickel in the presence of potassium metal produced the
five coordinate tungsten(V) aryloxides, ‘‘(p-XC6-
H4O)5W’’ [45]. Despite product characterization that
included molecular weight determination, EA, EPR
and polarography, a five coordinate complex of this type
would seem to be sterically capable of dimerizing to
form [(p-XC6H4O)4W]2(l-OC6H4X)2. Later work by
Wilkinson and coworkers [46] showed that reduction
of W(OPh)6 by metallic lithium, sodium, or potassium
in THF produced the octahedrally coordinated hexa-
phenoxotungstate(V) ion [W(OPh)6]ꢀ as [Li(thf)2]
[W(OPh)6], Na[W(OPh)6]ÆNaOPh, and K[W(OPh)6],
respectively. The monomers [Et4N][W(OPh)6], formed
by treating Na[W(OPh)6]ÆNaOPh with excess [Et4N]Br,
and [Li(thf)2][W(OPh)6] were structurally characterized.
EPR spectra were collected in frozen tetrahydrofuran at
100K for all of the [W(OPh)6]ꢀ compounds as well as
‘‘(PhO)5W’’, and the values appeared similar. Wilkinson
and coworkers [46] proposed that ‘‘(PhO)5W’’ in THF
was most likely the axially distorted WV monomer
W(OPh)5(thf), although no further characterization
was reported.
Detection of phenol atop the general increase in cur-
rent, which probably reflects the production of a num-
ber of oxidizable species, is a likely signature of
catastrophic degradation.
2.3. Attempted generation of (XC6H4O)5W(OC6H4O)
(7-X)
Although electrocatalytic C–H bond activation via
Fig. 1 was not realized, it was still possible that the

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2849
Table 2
Crystallographic data for triclinic {K[(p-ClC6H4O)6W]} (8-Cl)
Following the Funk protocol [45], reduction of (p-
ClC6H4O)6W (1-Cl) with Raney Ni (Et2O:EtOH;
40:1) generated a red solution that turned yellow-or-
ange upon addition of K0. The addition of quinone
caused an immediate color change to purple-red,
1
Formula
Formula weight
C36H24O6Cl6KW
494.10
ꢀ
Space group
˚
P1
k (A)
0.71073
7.7888 (4)
10.3052 (6)
11.8888 (7)
100.7200 (10)
105.4990 (10)
96.7620 (10)
889.26 (9)
173(2)
1
and a black solid precipitated. H NMR spectroscopy
˚
a (A)
˚
revealed only the formation of 1-Cl. Since the solvent
system appeared incompatible, the procedure was re-
done in Et2O and THF-d8 with similar results. As
Eq. (14) shows, in Et2O the reduction procedure
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
3
˚
yields {K[(p-XC6H4O)6W]} (8-X, X = H, 44%; Cl
1
61%) which can be isolated as orange crystals in
61% yield
V (A )
T (K)
Z
1
1.845
qcalc (g/cm3)
l (mmꢀ1
)
3.862
0.0218
ðp-xC6H4OÞ W ½1-XðX ¼ H; ClÞꢂ
6
R[I > 2r(I)]a
RaneyNi;Kꢃ
w2[I > 2r(I)]b
Goodness-of-fitc
Largest differential peak (e A
0.0558
1.028
!
fK½ðp-ClC6H4OÞ Wꢂg1½8-XðX ¼ H; ClÞꢂ
6
Et2O=EtOHðꢁ30:1Þ
ꢀ3
˚
)
1.740
ꢀ1.188
ð14Þ
ꢀ3
˚
Largest differential hole (e A
)
P
P
a
b
c
R1 = [|Fo|ꢀ|Fc|]/ |Fo|.
Attempts to remove one of the phenoxide ligands from
{K[(p-ClC6H4O)6W]} (8-Cl) by electrophilic attack
P
P
wR2 = [ w(|Fo| ꢀ |Fc|)2/ xFo
2 1/2
] .
P
1
with [Me3O][BF4] or Me3SiCl only resulted in oxidation
to (p-ClC6H4O)6W (1-Cl). The phenoxide ligand resisted
GOF = [ w(|Fo| ꢀ |Fc|)2/(n ꢀ p)]1/2, n = number of independent
reflections (5665 of 12570, Rint = 0.0237), p = number of parameters
(277).
1
substitution by PhCN, but H NMR spectroscopy indi-
cated that benzonitrile breaks up the infinite chain into a
monomeric species, [(p-ClC6H4O)6W][K(NCPh)4] (9-Cl,
Eq. (15))
3. Discussion
fK½ðp-ClC6H4OÞ Wꢂg1 ð8-ClÞ
6
3.1. Radical CH bond activation
benzene-d6
!
½ðp-ClC6H4OÞ Wꢂ½KðNCPhÞ ꢂ ð9-ClÞ
ð15Þ
6
4
PhCN
3.1.1. Electrocatalyst lifetime
The premise of this project, electrocatalytic CH bond
activation, was not realized. In an electrocatalytic proc-
ess [4], either reduction or oxidation (this case) serves to
regenerate a complex that undergoes a redox process,
hence the electrode takes the place of a chemical redox
partner that would be consumed (e.g., dioxygen) in the
reaction. Rapid degradation of the electrochemically
generated W(V) quinone or alternately described
W(VI) semiquinone radical intermediate prior to H
atom abstraction was considered as one possible cause
of failure. In Fig. 3, the short potential range (0.0–
2.3V) allowed degradation products (probably phenols)
to be regenerated with a H atom source present, whereas
the longer potential range in Fig. 4 (ꢀ2.2 to 2.3 V) and
longer scan time yielded no behavior that could be mis-
construed as ‘‘electrocatalytic’’.
2.4. X-ray crystal structure of {K[(p-ClC6H4O)6W]}
(8-Cl)
1
Crystallographic data for {K[(p-ClC6H4O)6W]} (8-
1
Cl) are presented in Table 2 and pertinent interatomic
distances and angles are listed in Table 3 As Fig. 6.
reveals, 8-Cl forms
a
zig-zag chain of [(p-
ClC6H4O)6W]ꢀ anions and potassium cations. The
potassium cation is coordinated by a distorted square
plane of oxygen atoms (\O–K–Ocis = 55.8(1)ꢁ,
124.2(1)ꢁ; \O–K–Otrans = 180ꢁ) in addition to axial
coordination by two asymmetrically bound aryl rings
of two different [(p-ClC6H4O)6W]ꢀ units (Fig. 7). A
small distortion from octahedral coordination is ob-
served in the WO6 core (\O1–W1–O2 = 87.8(1)ꢁ;
\O1–W1–O3 = 91.4(1)ꢁ;
\O2–W1–O3 = 92.4(1)ꢁ)
In retrospect, the lifetime of the intermediate that
must do the H atom abstraction must be substantial,
and this factor was underappreciated in this study.
For example, according to an Evans–Polanyi plot
[48], the activation of dihydroanthracene by transient
(p-XC6H4O)5W(O-C6H4O) (7-X, X = H, CH3, OCH3,
Cl, Br, OH) that was formed from E0ox ꢁ þ1:4 V
(Fig. 8). The average W–O bond distance (1.955(1)
˚
A) is longer than the average distance of normal tung-
˚
sten(VI) aryloxide bonds (1.889(14) A) [33,47] reflect-
ing partial population of the p* orbitals of d1
pseudo-octahedral W(V). A shorter terminal W–O
˚
bond distance (1.946(1) A) is found in comparison to
˚
the bridged W–O distances (d(W1–O1) = 1.960(2) A;
˚
d(W1–O2) = 1.959(1) A).
(Fig. 1) would occur at roughly 10 Mꢀ1 ꢀ1, hence
s
the lifetime of 7-X would need to be around a tenth

2850
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
Table 3
˚
Selected interatomic distances (A) and angles (ꢁ) pertaining to {K[(p-ClC6H4O)6W]} (8-Cl)
1
Interatomic distances
W–O1
W–O2
W–O3
(K–O)ave
1.874 (3)
1.926 (3)
1.913 (2)
2.904 (2)
K1–C13
K1–C15
K1–C17
3.139 (2)
3.538 (2)
3.516 (2)
K1–C14
K1–C16
K1–C18
3.270 (2)
3.652 (2)
3.257 (2)
Interatomic angles
(O–W1–O)trans
O2–W1–O3
180
92.4 (1)
132.3 (1)
O1–W1–O2
W1–O1–C1
O1–K–O2
87.8 (1)
135.0 (1)
55.8 (1)
O1–W1–O3
W1–O2–C7
O1–K–O20
91.4 (1)
139.1 (1)
124.2 (1)
W1–O3–C13
Fig. 6. Molecular view of the extended structure of {K[(p-ClC6H4O)6W]} (8-Cl). W, purple; K, blue; O, red; Cl, green; C, black.
1
Fig. 7. Coordination environment of K cation in {K[(p-ClC6H4O)6W]} (8-Cl).
1
of a second; at 2.1 V, activation would occur at ꢁ106
Mꢀ1 sꢀ1 and its lifetime would need to be in the
microsecond range. All oxidation waves found for (p-
XC6H4O)5W(OC6H4OH) (6-X, X = H, CH3, OCH3,
Cl, Br; X = OH, 1-OH) on the 100 mV/s timescale
were irreversible, and no reproducible reduction waves
were ever seen in this region, even at higher scan
speeds (100 V/s). Given the failure to achieve activa-
tion, no further elucidation of the electrochemistry or
product analyses were attempted.
3.1.2. Solvent stability and ionization energy
An attempt at CH bond activation using an electro-
catalytic cycle requires the use of a solvent that will sup-
port an electrolyte and not suffer activation. One might
question the utility of such a process given the rather
low BDEs of polar solvents such as acetonitrile
(BDE ꢁ 94.8(21) kcal/mol, ꢁ19 M) [43]. From the cal-
culations in Fig. 1 acetonitrile should have been an
excellent substrate for the putative quinone based chem-
istry projected for 6-X.

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2851
-2
Ph CH
3
PhCH CH
2
3
Ph-i-Pr
-4
-6
PhCH
3
Me HCCHMe
2
2
-8
c
HexH
-10
-12
-14
CH Cl
2
2
CH CN
3
Fig. 8. Molecular view of the [(p-ClC6H4O)6W]ꢀ ion in {K[(p-
ClC6H4O)6W]} (8-Cl).
8
11
12
9
10
13
1
IE (eV)
Fig. 9. Correlation of Que et al. hydrocarbon activation rates (k2 in
Mꢀ1 ꢀ1) by [(Bn-tpen)FeO]2+ vs. IE (eV), assuming its degradation
s
Que has examined the remarkable oxidation of hydro-
carbons by non-heme Fe(IV) oxo compounds in acetonit-
rile, thereby providing evidence to the contrary [27]. Que
et al. generate [{(pyCH2)2NCH2CH2N-(CH2py)2}FeO]2+
([(Bn-tpen)FeO]2+) in acetonitrile and measure second-
order rate constants for the activation of several hydro-
carbons, including cyclohexane, apparently without
competition from solvent activation. They correlate the
logk(act) versus the BDE(substrate) in the usual fashion,
but had acetonitrile been included as a ‘‘substrate’’, it
would stand out as a dramatically outlying point. If one
estimates the reaction of [(Bn-tpen)FeO]2+ with acetonit-
rile as the compoundꢀs decomposition rate, one can
rationalize the inexplicable stability of acetonitrile to H
atom abstraction.
rate is that of CH3CN activation. An estimate of methylene chloride
activation based on its IE is given as an open circle.
+
-
L MX + HR
n
L MX + HR
n
(a)
(b)
Fig. 9 plots the lnk2 (k2 in Mꢀ1 sꢀ1) versus IE (eV) for
the substrates used and acetonitrile, assuming its activa-
tion rate is that of the Fe(IV) oxoꢀs degradation in the
absence of added substrate [27]. A clean correlation is
observed, with R2 = 0.955; CH2Cl2, which is not used
in Queꢀs study, but has seen use in radical-based sys-
tems, is given a predicted activation lnk2 based on its
IE. This plot is a reasonable alternative to the more con-
ventional logk2 versus BDE employed by Que, and has
the obvious advantage of being able to include acetonit-
rile, yet what is its interpretation?
In Fig. 10, two reaction paths for H atom abstraction
are given: one implicates a synchronous transfer of pro-
ton and electron, and one reveals a transition state (e.g.,
the half-way point) that consists of greater electron
transfer character than proton transfer. In such an in-
stance, electron withdrawing groups on the substrate,
such as the CN on acetonitrile or the chlorides on
CH2Cl2 can attenuate the susceptibility to H atom
abstraction, despite BDEs that would lead one to pre-
dict ready activation. Note that a non-synchronous
-
+
L MX + HR
L MXH + R
n
n
Fig. 10. Reaction coordinate for H atom transfer to LnMX compar-
ing: (a) direct path where the proton and electron travel simultaneously
and (b) a path that possesses more electron transfer character at the
transition state.
transfer could lower the frequencies of transition state
vibrations that contribute heavily to the kinetic isotope
effect (KIE) for H atom transfer (i.e., frequencies of a
transition state with RHꢁ+ character should be lowered
considerably from that of Rꢁ). It has been shown that
such low frequency vibrations are a significant contribu-
tor to some anomalously high KIEs [49].
3.2. Phenoxide complexes
The stability of tungsten para-hydroxy-phenoxide
complexes to degradation via oligomerization and

2852
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
condensation permitted the synthesis of (p-XC6H4O)5-
W(O-C6H4OH) (6-X, X = H, CH3, OCH3, Cl, Br;
X = OH, 1-OH) for this study. Their stability to exces-
sive proton-induced ligand exchange, and chromatog-
raphy conditions verifies their robust character. The
complexes and their hexaphenoxide precursors exhibit
clean, reversible reductions to anionic and dianionic
species, but oxidations were irreversible and induced
degradation. It was disappointing that variation of
the (p-XC6H4O)5W fragment failed to substantially
change the oxidation potential of the –OC6H4OH
group, which was the first oxidation wave in each
6-X, despite the wide variation in the oxidation poten-
tials of the parent phenols. It was hoped that the
variation in phenol oxidation potential would be trans-
mitted to some extent to the W center and amplified
given the five p-XC6H4O ligands, but this was not
observed.
If one views the WOC6H4OH unit as a ‘‘phenoxide
anion’’, precedent suggests that the premise of this study
was thermodynamically doomed. Cheng et al. calculated
several p-hydroquinone anion bond strengths to be
ꢁ66–74 kcal/mol, which are values incommensurate
with the proposed CH bond activation. This work is cer-
tainly consistent with the relatively steady, irreversible
E0oxð1Þ values for 6-X, which varied by only 0.28 V.
Clearly, while the concept of electrocatalytic CH
bond activation according to Fig. 1. remains appar-
ent, a system possessing higher oxidation potentials
and greater stability of the oxidation product is
needed for fruition. The more common cycle for
Fig. 1 involves the pKa of LnMXH and the E0ox of
LnMXꢀ, but in this study the compounds were cho-
sen as hydroquinone mimics, hence the pKa of
LnMXH+ should be <0, i.e., dissociation should occur
spontaneously upon oxidation. Since the 1.36 pKa fac-
tor is rather modest, this was not considered to be a
serious detriment to the proposed CH bond activa-
tion, but it certainly could be a source of overestima-
tion in terms of limiting the strength of the bonds
potentially activated if the pKa is much less than zero.
For another view of this point, consider the more
conventional way of estimating the OH bond strength
of (p-XC6H4O)5W(O-C6H4OH) (6-X). First, the pKa
of 6-X would be determined (e.g., the pKa of 6-X
would be >0), then the oxidation potential of
(p-XC6H4O)5W(OC6H4Oꢀ) would be measured. It
is likely that the oxidation potential for this anionic
species would be far less positive than for 6-X, there-
by compensating for the pKa increase. Unfortunately,
attempts to generate a stable anion of 6-X via depro-
tonation failed, and mild bases afforded H-bonded
coordination networks [33]. It is likely that (p-
XC6H4O)5W(O-C6H4Oꢀ), which could also be consid-
ered as [(p-XC6H4O)5WIV(OC6H4O)]ꢀ, is an unstable
species.
4. Experimental
4.1. General considerations
All manipulations were performed using either glove-
box or high vacuum line techniques. Hydrocarbon sol-
vents containing 1–2 mL of added tetraglyme, and
ethereal solvents were distilled under nitrogen from pur-
ple sodium benzophenone ketyl and vacuum transferred
from same prior to use. C6D6 and acetonitrile were dried
˚
over activated 4 A molecular sieves, vacuum transferred
and stored under N2. All other chemicals were pur-
chased from commercial sources and used as received.
All glassware was oven dried, and NMR tubes for sealed
tube experiments were additionally flame-dried under
dynamic vacuum. All phenols were purchased from Ald-
rich Chem. Co. and recrystallized prior to use.
NMR spectra were obtained using INOVA-400 and
Unity-500 spectrometers, and chemical shifts are re-
ported relative to C6D6 (1H, 7.15; 13C{1H}, 128.39) or
acetone-d6 (1H, 2.05; 13C{1H}, 29.92). Infrared spectra
were recorded on a Nicolet Impact 410 spectrophotom-
eter interfaced to a Gateway PC. Elemental analyses
were performed by Oneida Research Services, Whites-
boro, NY, or Robertson Microlit Laboratories, Madison,
NJ. Magnetic moments were determined in C6D6 or
thf-d8 at room temperature using Evans method with
an applied diamagnetic correction.
4.2. Procedures
4.2.1. (p-PhCH2OC6H4O)6W (1-OCH2Ph)
W(OCH3)6 (1.150 g, 3.11 mmol) and 4-benzyloxyphe-
nol (5.00 g, 25.0 mmol) were combined in a large glass
bomb with 40 mL toluene. The bomb was lowered into
a salt bath until the solvent level was covered and the
temperature settled at 173ꢁC. After 24 h, the reaction
was judged incomplete by a TLC using toluene as the
eluting solvent. W(OC6H4OCH2C6H5)6 has an Rf of
0.70 while the Rf of W(OC6H4OCH2C6H5)5(OCH3) is
0.65 and other less substituted products have lower
Rfs. The temperature was gradually increased, until after
24 h at 213 ꢁC the reaction was complete except for a
trace of a lower Rf spot. Toluene was added to the reac-
tion mixture and it was washed with 3 · 350 mL 5%
NaOH (aq). The deep red solution was dried over
Na2SO4 and the solvent stripped on a rotary evaporator.
The resulting dark red, very viscous oil was placed under
1
high vacuum for several hours. H NMR showed clean
1-OCH2Ph and toluene. The yield of 4.41 g translates to
3.89 g 1-OCH2Ph when the presence of toluene was ac-
counted for, giving an actual yield of 91%.
4.2.2. (p-HOC6H4O)6W (1-OH)
1-OCH2Ph(3.85g, 2.79mmol)wasdissolvedin125mL
toluene/125 mL ethanol along with 500 mg 10% PdC. The

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2853
mixture was stirred under 1 atm H2 and checked periodi-
cally by TLC using 100% ethyl acetate as the eluting sol-
vent. Incompletely deprotected products have higher Rfs
(0.85 or greater) than the final product (Rf = 0.80). After
4 h, the reaction was mostly complete, but required 24 h
until only a trace spot at Rf = 0.85 remained. The reaction
mixture was filtered through celite and the celite washed
with ethanol. Volatileswereremovedon arotary evapora-
tor and the product transferred to a 100 mL flask with
THF. The THF was stripped and dry THF was added
andremoved.DryTHFwasaddedandthenstrippedtwice
more to remove traces of water. The product was crystal-
lized from 50 mL benzene/3–4 mL THF and dried under
vacuum for 12 h (1.778 g, 60% based on (p-HOC6H4O)6-
WÆ2THFÆC6H6) (1-OHÆ2THFÆC6H6). 1H NMR in ace-
tone-d6 indicated pure 1-OH with 2.0 equivalents of
THF and 0.9 equivalents of benzene. Elemental analysis
indicated some loss of solvent before combustion: Anal.
Calc. for W(OC6H4OH)6 Æ1.5 THF: C, 53.29; H, 4.47.
Found: C, 53.11; H, 4.55%.
(e) (C6H5O)6W (1-H). TLC (25% ethyl acetate:hex-
1
ane): Rf = 0.60. H NMR (acetone-d6): d 6.90 (t, J = 8
Hz, CHp), 6.96 (d, J = 8 Hz, CH), 7.27 (t, J = 8 Hz,
CHm). 13C{1H} NMR (acetone-d6): d 121.16 (Co),
124.59 (Cp), 129.81 (Cp), 129.88 (Cm), 162.69 (Cipso).
4.2.4. General (p-XC6H4O)5W(OC6H4OH) (6-X, X =
OMe, Me, H, Cl, Br)
A 250 mL round bottom flask was charged with 1-X
(5.4 mmol), 4-(benzyloxy)phenol (10 mmol), p-toluene-
sulfonic acid monohydrate (1.1 mmol), and chloroben-
zene (100 mL). The solution was refluxed under an
argon atmosphere for 1–3 h and the reaction was moni-
tored by TLC (silica plate) for optimization of the mono-
substituted product (p-XC6H4O)5W(OC6H4OCH2Ph)
(5-X). Toluene (100 mL) was added to reaction solution
and the mixture was washed three times with both 5%
NaOH(aq) and de-ionized water. It was dried over
MgSO4, and the solvent was removed under reduced
pressure. The resulting red oil was dissolved in a mixture
of toluene (50 mL) and absolute ethanol (50 mL) fol-
lowed by an argon purge. 10% Palladium on carbon
(0.125 g) was added and the heterogeneous solution
was stirred under hydrogen (1 atm) for 5 h. Analysis
by TLC of the hydrogenated solution versus the starting
solution indicated complete hydrogenation. The catalyst
was removed by filtration and the volatiles were removed
on a rotary evaporator. Purification was accomplished
by column chromatography on silica using an appropri-
ate mixture of ethyl acetate and hexane as the mobile
phase. Further purification by recrystallization from
methanol was performed when 0possible.
4.2.3. General (p-XC6H4O)6W (1-X, X = H, CH3,
OCH3, Cl, Br)
To a 250 mL flask attached to a reflux condenser was
added WCl6 (5 g, 12.6 mmol). A sidearm flask contain-
ing p-XC6H4OH (81.7 mmol) was connected to the flask,
which was attached to an argon line equipped with a
bubbler. 100 mL of chlorobenzene was added and the
phenol was gradually tapped into the flask and HCl
(g) was evolved. The solution was refluxed for 12–24 h
followed by an argon purge to remove residual HCl
(g). The solution was washed twice with 10% NaOH
(aq) and deionized water, then dried over MgSO4. The
solvent was removed under reduced pressure, and the
product crystallized from petroleum ether (X = Me, Cl,
Br) or ethanol (X = OMe, H).
(a) (p-MeOC6H4O)5W (O-C6H04OH) (6-OMe).
TLC(50% ethyl acetate:hexane): Rf = 0.50. 1H NMR
(acetone-d6): d 3.73 (s, 15H, Me), 6.71 (d, J = 9 Hz,
2H, H0 or H0 ), 6.80 (m, 12H, Hm and C0Ho0 or C0H0m),
o
m
(a) (p-MeOC6H4O)6W (1-CH3). TLC (50% ethyl
acetate:hexane): Rf = 0.59. 1H NMR (acetone-d6): d
3.72 (s, Me), 6.78 (d, J = 9 Hz, CH), 6.85 (d, J = 9 Hz,
CH). 13C{1H} NMR (acetone-d6): d 55.85 (Me),
114.67 (Cm), 121.91 (Co), 156.71 (Cp), 156.93 (Cipso).
(b) (p-MeC6H4O)6W (1-CH3). TLC (25% ethyl ace-
tate:hexane): Rf = 0.66. 1H NMR (acetone-d6): d 2.33 (s,
Me), 6.83 (d, J = 9 Hz, CH), 7.05 (d, J = 9 Hz, CH).
13C{1H} NMR (acetone-d6): d 20.69 (Me), 120.96 (Co),
130.17 (Cm), 133.77 (Cp), 160.78 (Cipso).
6.86 (m, 10H, CHo), 8.20 (s, 1H, OH). 13C{1H} NMR
(acetone-d6): 55.83 (Me), 114.63 (Cm), 115.98 (C0m),
121.90 (Co), 122.02 (C0o), 154.43 ðCp0 Þ, 156.25 ðCi0psoÞ,
156.60 (Cp), 156.98 (Cipso). Anal. Calc for
C41H40O12W: C, 54.2; H, 4.5. Found: C, 54.2; H, 4.4%.
ðbÞ ðp-MeC6H4OÞ W ðOC06H04OHÞ ð6-MeÞ. TLC (25%
5
ethyl
acetate:hexane):
Rf(5-Me) = 0.44;
Rf(6-
1
Me) = 0.18. H NMR (acetone-d6): d2.32( s, 15H, Me),
6.71 (d, J = 9 Hz, 2H, C0Ho0 or C0H0m), 6.82 (m, 12H,
Ho and H0o or Hm0 ) 7.05 (d, J = 9Hz, 10H, Hm), 8.31 (s,
1H, OH). 13C{1H} NMR (acetone-d6): 20.86 (Me),
115.93 ðC0 Þ, 120.92 (Co), 122.20 ðC0 Þ, 130.16 (Cm),
(c) (p-ClC6H4O)6W (1-Cl). TLC (25% ethyl ace-
1
tate:hexane): Rf = 0.73. H NMR (acetone-d6): d 7.01
m
o
(d, J = 9 Hz, CH), 7.32 (d, J = 9 Hz, CH). 13C{1H}
NMR (acetone-d6): d 122.63 (Co), 129.81 (Cp), 130.04
(Cm), 160.79 (Cipso).
133.57 (Cp), 154.81 ðC0 Þ, 156.13 ðC0 Þ, 160.85 (Cipso).
p
ipso
Anal. Calc for C41H40O7W: C, 59.4; H, 4.9. Found: C,
59.5; H, 4.8%.
(d) (p-BrC6H4O)6W (1-Br). TLC (15% ethyl ace-
1
(c) (p-BrC6H4O)5W (OC06H40 OH)(6-Br). TLC (3%
ethyl acetate:hexane): Rf(5-Br) = 0.75; Rf(6-Br) = 0.44.
1H NMR (acetone-d6): d 6.71 (m, 2H, C0H0 or C0H0 ),
tate:hexane): Rf = 0.68. H NMR (acetone-d6): d 6.96
(d, J = 9 Hz, CH), 7.47 (d, J = 9 Hz, CH). 13C{1H}
NMR (acetone-d6): d 123.07 (Co), 133.07 (Cm), 161.24
(Cipso), Cp obscured.
o
m
6.94 (m, 12H, Ho and C0Ho0 or C0Hm0 ) 7.46 (m, 10H,
Hm), 8.62 (s, 1H, OH).

2854
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
(d) (p-ClC6H4O)5W (OC06H40 OH) (6-Cl). TLC (3%
4.2.8. Electrochemistry: general procedure
ethyl acetate:hexane): Rf(5-Cl) = 0.70; Rf(6-Cl) = 0.41.
4-Chlorophenol could not be separated from 6-Cl by
column chromatography or crystallization and impeded
definitive spectroscopic 0characterization.
The working electrode was a 1mm diameter platinum
disk electrode sealed in glass. In cases where fast sweep
rates were employed, a 75 lm platinum disk electrode,
sealed in glass, was used. All measurements were carried
out inside a glove-box (Vac. Atmospheres) at 25 ꢁC.
Acetonitrile (Burdick & Jackson; distilled in glass) and
(e) (C6H5O)5W (OC6H04OH) (6-H). TLC (25%
ethyl acetate:heptane): Rf(5-H) = 0.48; Rf(6-H) = 0.25.
1H NMR (acetone-d6): d 6.72 (d, J = 8 Hz, 2H,
C0H0o or C0H0m), 6.93 (m, 17H, CHo, CHm and
C0H0o or C0H0m), 7.26 (m, 10H, d), 8.32 (s, 1H, OH).
˚
dried over 4 A molecular sieves was used as the solvent.
Tetra-n-butyl ammonium hexafluorophosphate (TBAG;
G.F. Smith) used as supporting electrolyte was recrystal-
lized 3· from absolute ethanol and dried under vacuum
for 96 h. A Princeton Applied Research Model 173
Potentiostat with a model 175 Programmer were used.
Data at sweep rates at or below 500 mV/s were recorded
on a Soltec X–Y recorder. At faster sweep rates, data
were recorded on a Tektronix digital storage oscillo-
scope. A background scan of solvent and supporting
electrolyte out to the potential limits of the experiment,
was always varied out. A sweep rate dependence of the
peak current was carried out for all the materials in-
volved and in all cases the current was proportional to
the square root of the sweep rate (over a typical range
of 50–1000 mV/s) as anticipated for a diffusion control-
led process. A silver wire was employed in order to en-
sure rigorous exclusion of water, especially since
experiments were done inside a dry box. Reproducibility
of the peak potentials – even those that are irreversible –
was roughly 10–15 mV.
13C{1H} NMR (acetone-d6): 115.99 ðC0 Þ, 121.09 (Co),
m
122.30 ðC0 Þ, 124.31 (Cp), 129.83 (Cm), 154.31 ðC0 Þ,
p
155.94 ðC0iposoÞ, 162.81 (Cipso).
4.2.5. {K[(p- ClC6H4O)6W]} (8-Cl)
1
A 25 mL round bottom flask was charged with 1-
Cl (0.300 g, 0.316 mmol), Raney Nickel (0.075 g, 1.28
mmol), diethyl ether (15 mL), and ethanol (0.40 mL).
Potassium metal (0.02 g, 0.5 mmol) was added and
the purple-red solution was stirred for 2 h. The
resulting heterogeneous orange solution was filtered,
and the filtrate layered with pentane yielding the
product as orange crystals (0.191 g, 61%). Anal. Calc.
for C36H24O6Cl6KW: C, 43.8; H, 2.5; Cl, 21.5.
Found: C, 44.0; H, 2.5; Cl, 20.6%.
4.2.6. {K[(C6H5O)6W]} (8-H)
1
A 25 mL round bottom flask was charged with 1-H
(0.200 g, 0.269 mmol), Raney Nickel (0.050 g, 0.852
mmol), diethyl ether (10 mL), and ethanol (0.30
mL). Potassium metal (0.02 g, 0.5 mmol) was added
and the purple-red solution was stirred for 2 h result-
ing in a yellow precipitate and light yellow solution.
The solid was collected by filtration and separated
from the remaining Raney Nickel by dissolution in
minimal THF followed by filtration and crystallization
4.2.9. X-ray crystal structure determination: {K[(p-
ClC6H4O)6W]} (8-Cl)
1
X-ray diffraction quality single crystals were ob-
tained by layering a saturated diethyl ether solution
of 8-Cl with pentane. One orange crystal
(0.3 · 0.15 · 0.1 mm) was selected, coated in polyiso-
butylene, and placed under a 173 K nitrogen stream
on the goniometer head of a Siemens SMART CCD
Area Detector system equipped with a fine-focus
1
by slow evaporation (0.093 g, 44%). H NMR (thf-d8):
d 4.96 (m1/2 ꢁ 20 Hz, 6H, Hp), 6.63 (m1/2 ꢁ 40 Hz, 12H,
CH), 9.28 (m1/2 ꢁ 20 Hz, 12H, CH). The 1H NMR
spectrum is similar to that obtained by Wilkinson
and coworkers [45] (acetone-d6) for an independent
synthesis of KW(OPh)6. leff(293K) = 1.5 lB (Evansꢀ
method in thf-d8).
˚
molybdenum X-ray tube (k = 0.71073 A). Preliminary
diffraction data revealed a triclinic crystal system. A
hemisphere routine was used for data collection. The
data was subsequently processed with the Bruker
SAINT program and the space group was determined
ꢀ
to be P1. The data was corrected for absorption with
4.2.7. Preparation of [(p-ClC6H4O)6W][K(PhCN)4]
(9-Cl)
SADABS, solved by direct methods, completed by sub-
sequent difference Fourier syntheses and refined by
full-matrix least-squares procedures (SHELXTL). All
non-hydrogen atoms were refined with anisotropic
displacement parameters and hydrogen atoms were in-
A 10 mm O.D. glass tube was charged with 8-Cl
(0.081 g, mmol) and benzonitrile (4 mL). The solution
was degassed, sealed, and heated at 100 ꢁC for 5 d with
no color change. The product was isolated as an orange
solid (0.072 g, 63%) after solvent removal followed by
trituration with pentane. 1H NMR (thf-d8): d 7.12
(m1/2 ꢁ 160 Hz, 12H, CH), 7.51 (t, J = 8 Hz, 8H, PhCN
Hp or Hm), 7.63 (t, J = 8 Hz, 4H, PhCN Hp or Hm), 7.69
(d, J = 8 Hz, 8H, PhCN Ho), 9.18 (m1/2 ꢁ 10 Hz, 12H,
CH).
1
cluded at calculated positions.
1
Cambridge Crystallographic Data Centre deposition number:
CCDC 247677.

O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
2855
[20] M.L.H. Green, P.J. Knowles, J. Chem. Soc., Chem. Commun.
(1970) 1677.
Acknowledgements
[21] E.A. Ambundo, R.A. Friesner, S.J. Lippard, J. Am. Chem. Soc.
124 (2002) 8770–8771, and references therein.
The National Science Foundation and the Depart-
ment of Chemistry and Chemical Biology at Cornell
University are gratefully acknowledged for support.
T.P.V. thanks the NSF for an Inorganic Materials
Traineeship. We thank a referee for pointing out the
work of Cheng, Handoo, Xue and Parker [50].
[22] M. Newcomb, R. Shen, Y. Lu, M.J. Coon, P.F. Hollenberg,
D.A. Kopp, S.J. Lippard, J. Am. Chem. Soc. 124 (2002) 6879–
6886, and references therein.
[23] (a) P. Chaudhuri, M. Hess, U. Flo¨rke, K. Wieghardt, Angew.
Chem. Int. Ed. 37 (1998) 2217–2220;
(b) P. Chaudhuri, M. Hess, J. Muller, K. Hildenbrand, E. Bill, T.
Weyhermuller, K. Wieghardt, J. Am. Chem. Soc. 121 (1999)
¨
9599–9610.
[24] (a) R.C. Pratt, T.D.P. Stack, J. Am. Chem. Soc. 125 (2003) 8716–
8717;
References
(b) Y.D. Wang, J.L. DuBois, B. Hedman, K.O. Hodgson, T.D.P.
Stack, Science 279 (1998) 537–540.
[1] J.A. Labinger, J.E. Bercaw, Nature 417 (2002) 506–514.
[2] S. Stahl, J.A. Labinger, J.E. Bercaw, Angew. Chem. Int. Ed. Engl.
37 (1998) 2180–2192.
[25] C.R. Goldsmith, R.T. Jonas, T.D.P. Stack, J. Am. Chem. Soc.
124 (2002) 83–96.
[26] (a) J.-U. Rohde, J.-H. In, M.H. Lim, W.W. Brennessel, M.R.
[3] (a) A.E. Shilov, G.B. Shulꢀpin, Activation and Catalytic Reactions
of Saturated Hydrocarbons in the Presence of Metal Complexes,
Kluwer Academic Press, Dordrecht, Netherlands, 2000;
(b) A.E. Shilov, A.A. Shteinmann, Coord. Chem. Rev. 24 (1977)
97–143.
Bukowski, A. Stubna, E. Munck, W. Nam, L. Que Jr., Science
¨
299 (2003) 1037–1039;
(b) M.H. Lim, J.-U. Rohde, A. Stubna, M.R. Bukowski, M.
Costas, R.Y.N. Ho, E. Munck, W. Nam, L. Que Jr., Proc. Natl.
¨
Acad. Sci. 100 (2003) 3665–3670.
[27] J. Kaizer, E.J. Klinker, N.Y. Oh, J.-U. Rohde, W.J. Song, A.
[4] M.S. Freund, J.A. Labinger, N.S. Lewis, J.E. Bercaw, J. Mol.
Catal. 87 (1994) L11–L16.
Stubna, J. Kim, E. Munck, W. Nam, L. Que Jr., J. Am. Chem.
¨
Soc. 126 (2004) 472–473.
[5] M. Lin, C. Shen, E.A. Garcia-Zayas, A. Sen, J. Am. Chem. Soc.
123 (2001) 1000–1001.
[28] (a) D.D.M. Wayner, V.D. Parker, Acc. Chem. Res. 26 (1993)
287–294;
[6] R.A. Periana, O. Mironov, D. Taube, G. Bhalla, C.J. Jones,
Science 301 (2003) 814–818.
(b) M. Tilset, V.D. Parker, J. Am. Chem. Soc. 111 (1989) 6711–6717;
M. Tilset, V.D. Parker, J. Am. Chem. Soc. 112 (1990) 2843;
(c) V.D. Parker, K.L. Handoo, F. Roness, M. Tilset, J. Am. Chem.
Soc. 113 (1991) 7493–7498.
[7] R.A. Periana, D.J. Taube, S. Gamble, H. Taube, T. Satoh, H.
Fujii, Science 280 (1998) 560–564.
[8] R.F. Jordan, D.F. Taylor, N.C. Baenziger, Organometallics 9
(1990) 1546–1557.
[29] J.M. Mayer, Acc. Chem. Res. 31 (1998) 441–450.
[30] K.A. Gardner, J.M. Mayer, Science 269 (1995) 1849–1851.
[31] (a) K. Wang, J.M. Mayer, J. Am. Chem. Soc. 119 (1997) 1470–1471;
(b) J.R. Bryant, J.M. Mayer, J. Am. Chem. Soc. 125 (2003) 10351–
10361, and references therein.
[9] C.G. Jia, D.G. Piao, J.Z. Oyamada, W.J. Lu, T. Kitamura, Y.
Fujiwara, Science 287 (2000) 1992–1995.
[10] T. Matsumoto, R.A. Periana, D.J. Taube, H. Yoshida, J. Mol.
Catal. A: Chem. 180 (2002) 1–18.
[11] (a) K.M. Waltz, J.F. Hartwig, J. Am. Chem. Soc. 122 (2000)
11358–11369;
[32] L.B. Handy, K.G. Sharp, F.E. Brinckman, Inorg. Chem. 11
(1972) 523–531.
(b) H.Y. Chen, S. Schlecht, T.C. Temple, J.F. Hartwig, Science
287 (2000) 1995–1997;
[33] T.P. Vaid, O.L. Sydora, R.E. Douthwaite, P.T. Wolczanski, E.B.
Lobkovsky, J. Chem. Soc. Chem. Commun. (2001) 1300–1301.
[34] S.M. Beshouri, I.P. Rothwell, Inorg. Chem. 25 (1986) 1962–1964.
[35] H. von Funk, W. Baumann, Z. anorg. Chem. 231 (1937) 264–268.
[36] P.I. Mortimer, M.I. Strong, Aust. J. Chem. 18 (1965) 1579–1587.
[37] B.R. Eggins, J.Q. Chambers, J. Electrochem. Soc.117(1970) 186–192.
[38] M.W. Lehmann, D.H. Evans, J. Electroanal. Chem. 500 (2001)
12–20, and references therein.
(c) C.E. Webster, Y.B. Fan, M.B. Hall, D. Kunz, J.F. Hartwig, J.
Am. Chem. Soc. 125 (2003) 858–859.
[12] (a) R.E. Maleczka, F. Shi, D. Holmes, M.R. Smith, J. Am.
Chem. Soc. 125 (2003) 7792–7793;
(b) J.Y. Cho, M.K. Tse, D. Holmes, R.E. Maleczka, M.R. Smith,
Science 295 (2002) 305–308.
[13] C.M. Jensen, Chem. Commun. (1999) 2443–2449.
[14] (a) K.B. Renkema, Y.V. Kissin, A.S. Goldman, J. Am. Chem.
Soc. 125 (2003) 7770–7771;
[39] L. Papouchado, R.W. Sandford, G. Petrie, R.N. Adams, J.
Electroanal. Chem. 65 (1975) 275–284.
[40] C. Iwakura, M. Tsunaga, H. Tamura, Electrochem. Acta 17
(1972) 1391–1400.
(b) K. Krogh-Jespersen, M. Czerw, N. Summa, K.B. Renkema,
P.D. Achord, A.S. Goldman, J. Am. Chem. Soc. 124 (2002)
11404–11416.
[41] D.T.Y. Yiu, M.F.W. Lee, W.W.Y. Lam, T.-C. Lau, Inorg. Chem.
42 (2003) 1225–1232.
[15] G.W. Parshall, S.D. Ittel, Homogeneous Catalysis: The Applica-
tions and Chemistry of Catalysis by Soluble Transition Metal
Complexes, 2nd ed., Wiley, New York, 1992.
[16] R.A. Sheldon, J.K. Kochi, Metal-catalyzed Oxidations of Organic
Compounds, Academic Press, New York, 1981.
[17] W.D. Jones, E.T. Hessell, J. Am. Chem. Soc. 115 (1993) 554–562.
[18] (a) P.O. Stoutland, R.G. Bergman, S.P. Nolan, C.D. Hoff,
Polyhedron 7 (1988) 1429–1440;
[42] J.A. Richards, D.H. Evans, J. Electroanal. Chem. 81(1977)171–187.
[43] J. Berkowitz, G.B. Ellison, D. Gutman, J. Phys. Chem. 98 (1994)
2744–2765, and references therein.
[44] D.E. Berning, B.C. Noll, D.L. DuBois, J. Am. Chem. Soc. 121
(1999) 11432–11447.
[45] H. Funk, H. Matschiner, H. Naumann, Z. Anorg. Allg. Chem.
340 (1965) 75–81.
[46] J.I. Davies, J.F. Gibson, A.C. Skapski, G. Wilkinson, W.-K.
Wong, Polyhedron 1 (1982) 641.
(b) R.A. Periana, R.G. Bergman, J. Am. Chem. Soc. 108 (1986)
7332–7346.
[47] (a) M.L. Listemann, J.C. Dewan, R.R. Schrock, J. Am. Chem.
Soc. 107 (1985) 7207–7208;
[19] (a) J.L. Bennett, P.T. Wolczanski, J. Am. Chem. Soc. 119 (1997)
10696–10719;
(b) M.A. Lockwood, P.E. Fanwick, O. Eisenstein, I.P. Rothwell,
J. Am. Chem. Soc. 118 (1996) 2762–2763.
(b) J.L. Bennett, T.P. Vaid, P.T. Wolczanski, Inorg. Chim. Acta
270 (1998) 414–423.

2856
O.L. Sydora et al. / Polyhedron 23 (2004) 2841–2856
[48] (a) J.K. Kochi (Ed.), Free Radicals, Wiley, New York, 1973;
(b) J.H. Knox, in: Z.B. Alfassi (Ed.), Chemical Kinetics
of Small Organic Radicals, CRC Press, Boca Raton, FL,
1988.
[49] L.M. Slaughter, P.T. Wolczanski, T.R. Klinckman, T.R. Cundari,
J. Am. Chem. Soc. 122 (2000) 7953–7975.
[50] J.-P. Cheng, K.L. Handoo, J. Xue, V.D. Parker, J. Org. Chem. 58
(1993) 5050–5054.