Studies on polyoxo and polyperoxo-metalates: Part 7. Lanthano- and thoriopolyoxotungstates as catalytic oxidants with H2O2 and the X-ray crystal structure of Na8[ThW10O36]·28H2O

Article abstract of DOI:10.1016/S0022-328X(00)00308-9

The effectiveness of salts of [Ln^IIIW₁₀O₃₆]^9- (Ln = Y, La, Ce, Pr, Sm, Eu, Gd, Dy, Er, Lu) and [M^IVW₁₀O₃₆]^8- (M = Ce, Th) as catalysts with H₂O₂

Full text of DOI:10.1016/S0022-328X(00)00308-9

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 607 (2000) 146–155
ꢀ
Studies on polyoxo and polyperoxo-metalates
Part 7. Lanthano- and thoriopolyoxotungstates as catalytic
oxidants with H O and the X-ray crystal structure of
2
2
Na [ThW O ]·28H O
8
10 36
2
William P. Griffith *, Neil Morley-Smith, Helena I.S. Nogueira, Abdel G.F. Shoair,
Maria Suriaatmaja, Andrew J.P. White, David J. Williams
Department of Chemistry (Inorganic Chemistry), Imperial College of Science, Technology and Medicine, London SW 7 2AY, UK
Received 13 March 2000; received in revised form 5 May 2000; accepted 15 May 2000
Dedicated to Professor Martin Bennett, FRS, a great chemist and a good friend since our joint undergraduate and postgraduate days at
Imperial College.
Abstract
III
9−
IV
8−
The effectiveness of salts of [Ln W O ]
(Ln=Y, La, Ce, Pr, Sm, Eu, Gd, Dy, Er, Lu) and [M W O ]
(M=Ce, Th)
10
36
10 36
as catalysts with H O for alcohol oxidations and alkene epoxidations has been studied. It appears that catalysis arises from the
2
2
8
−
polyperoxotungstates formed from H O . The X-ray crystal structure of the title complex shows that in the [ThW O ]
anion
2
2
10 36
the thorium has square antiprismatic geometry in which eight oxygen atoms from two W O moieties form vertex-sharing bonds;
5
18
31
1
Raman data suggest that the structure of the anion is retained in aqueous solution. New P[ H]-NMR data for
III
11−
IV
10−
[
Ln {PW O } ]
(Ln=Y, La, Ce, Pr, Nd, Sm, Eu, Tb, Ho, Er, Yb, Lu) and [Ce {PW O } ]
in the solid state, in water
1
1
39
2
11 39 2
and in H O solution are presented; these species have also been used for oxidation catalysis. © 2000 Elsevier Science S.A. All
2
2
rights reserved.
Keywords: Oxidation catalysis; X-ray crystal structure; Lanthanides; Thorium; Polyoxotungstates
+
1. Introduction
R=cetylypyridinium, (Me(CH ) NC H ) ) catalyse
2 15 5 5
oxidations of benzyl alcohol to benzaldehyde and 2-oc-
tanol to octanone [4–7]. Heavier rather than lighter
lanthanides were found to be slightly more effective, the
reverse being the case for epoxidation of cyclooctene;
the cerium complex is an anomalously poor catalyst [8].
Attempts were made to correlate such effects with the f
orbital occupancy of the lanthanides [5,9].
There is currently much interest in the use of polyox-
ometalates as useful catalysts for specific organic oxida-
tions [2]; this paper concerns the use of lanthanopoly-
oxotungstates and a thoriopolyoxotungstate as cata-
lysts for alcohol and alkene oxidations with H O as
2
2
co-oxidant.
It has been reported that sodium salts of
In preliminary work we showed that oxidation catal-
ysis with sodium and mixed sodium–tetramethylammo-
9
−
IV
8−
[
LnW O ]
(Ln=Nd, Sm) and [Ce W O ]
will
1
0
36
10 36
9
−
IV
8−
catalyse, with H O , the oxidation of cyclohexanol to
nium salts of [LnW10
O
36
]
and [Ce W10
refluxing BuOH–H O , is equally well effected by
2 2
O
36
]
in
2
2
t
cyclohexanone [3], and that R H [LnW O ] salts
7
2
10 36
(
Ln=Y, La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Ho, Er, Yb;
tungstate in the absence of lanthanide [10]. Here we
re-examine claims [2–9] made for the efficacy of
III
9−
IV
8−
[
Ln W O ]
and [Ce W O ]
as catalysts with
1
0
36
10 36
ꢀ
H O for the oxidation of alcohols and alkenes using
their sodium, cetylpyridinium and n-hexylammonium
2
2
For Part 6, see Ref. [1].
Corresponding author. fax: +44-20-7594-5804.
*
E-mail address: wgriffith@ic.ac.uk (W.P. Griffith).
salts. We have prepared and similarly used salts of
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00308-9

W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
147
IV
8−
[
Th W O ]
and report for the first time the X-ray
2. Results and discussion
1
0
36
crystal structure of its sodium salt and its Raman
spectra. New spectroscopic data for lanthanophospho-
III
9−
2.1. Sodium salts of [Ln W O ]
[M W O ]
10 36
and of
10
36
III
11−
IV
8−
polyoxotungstates [Ln {PW O } ]
, earlier studied
(M=Ce, Th)
11
39
2
as oxidation catalysts with H O2 [11], are also
2
presented.
These are made by reaction of Na [WO ] and the
2
4
lanthanide nitrate at pH 7.0–7.5 [12]; the first to be
crystallographically characterised was Na H -
Ce W O ]·30H O [13]. The stoicheiometries of
6
2
IV
[
10
36
2
Na H [SmW O ]·28H O [14] and Na H[GdW O ]·
6
3
10 36
2
8
10 36
3
0H O [15] have been established by single-crystal X-
2
ray studies. Thus it is clear that the sodium and water
contents can be variable; analyses of our products are
most consistent with Na [W O ]·nH O, with n varying
9
10 36
2
between 20–38.
The less effective catalytic properties of salts of
IV
8−
[
Ce W O ]
noted by Kera et al. [8] and by us for
1
0
36
cetylpyridinium salts (see below) could arise from
tighter bonding between the tetravalent cerium centre
and the eight donor oxygen atoms of the two W O
5
18
moieties to which it is coordinated. To explore this
possibility we made another complex with a tetravalent
centre, the thoriopolyoxotungstate N [ThW O ]·
8
10 36
2
8H O.
2
2
.2. Na [ThW O ]·28H O; formulation, X-ray crystal
8
10 36
2
structure and Raman spectra
Reaction of thorium nitrate with hot aqueous sodium
tungstate at pH 7.3, a method similar to those used for
III
9−
IV
8−
making [Ln W O ]
and [Ce W O ] , was re-
10
36
10 36
Fig. 1. Ball-and-stick representation of the anionic [ThW10O36]8−
thoriopolyoxotungstate unit.
ported in 1971 to give Na [ThW O ]·25H O [16], sub-
4
8
28
2
sequently reformulated as Na [W O ]·27H O [17].
6
10 35
2
Although it has been suggested [18] that a better formu-
lation would be N [ThW O ]·30H O, no X-ray data
have been reported. We have repeated the literature
preparations [16,17] and obtained crystals which, after
X-ray analysis, proved the formulation to be
Table 1
Selected bond lengths (A
8
10 36
2
a
,
) for 1
ThꢀO(20)
ThꢀO(40)
W(1)ꢀO
W(1)ꢀO(12)
W(1)ꢀO(14)
W(2)ꢀO
W(2)ꢀO(12)
W(2)ꢀO(23)
W(3)ꢀO
W(3)ꢀO(13)
W(3)ꢀO(30)
W(4)ꢀO
W(4)ꢀO(14)
W(4)ꢀO(40)
W(5)ꢀO
W(5)ꢀO(15)
W(5)ꢀO(45)
2.444(12)
ThꢀO(30)
ThꢀO(50)
2.455(14)
2.447(13)
1.714(12)
1.920(13)
1.91(2)
1.717(12)
1.789(12)
1.954(14)
1.722(14)
1.942(13)
1.944(14)
1.703(14)
1.961(14)
1.937(14)
1.742(14)
1.953(13)
1.792(14)
2.454(14)
2.340(13)
1.92(2)
Na [ThW O ]·28H O.
W(1)ꢀO(1)
W(1)ꢀO(13)
W(1)ꢀO(15)
W(2)ꢀO(2)
W(2)ꢀO(20)
W(2)ꢀO(25)
W(3)ꢀO(3)
W(3)ꢀO(23)
W(3)ꢀO(34)
W(4)ꢀO(4)
W(4)ꢀO(34)
W(4)ꢀO(45)
W(5)ꢀO(5)
W(5)ꢀO(25)
W(5)ꢀO(50)
8
10 36
2
This X-ray analysis shows the thorium complex to be
IV
1.922(13)
2.326(14)
2.013(14)
1.934(14)
2.319(13)
1.988(13)
1.785(14)
2.354(14)
1.99(2)
isomorphous with the Na H [Ce W O ]·30H O [13]
6
2
10 36
2
IV
and Na [U W O ]·30H O [19] analogues, the only
8
10 36
2
disparities being in the assessment of the numbers of
sodium ions and water molecules present. The thorium
structure matches most closely that of the uranium
complex, having ten tungsten atoms and eight distinct
sodium cations per thorium centre; only 28 water
molecules, however, were identified. The structure (Fig.
1.792(14)
2.312(13)
1.99(2)
1
) has crystallographic C2 symmetry about an axis
passing through the thorium atom and normal to the
W(1)ꢀThꢀW(1%) vector. The geometry at thorium is
almost perfect square antiprismatic, the degree of stag-
ger between the upper and lower faces being less than
1.929(13)
a
The oxo ligand oxygen atoms carry the same numbers as their
parent tungsten atoms; the bridging oxygen atoms carry the numbers
of the two tungsten atoms being bridged, the lower numbered tung-
sten always coming first, e.g. the oxygen bridging W(1) and W(5) is
O(15) and that bridging W(2) and W(3) is O(23), etc.
1
° from ideal.
The ThꢀO distances (Table 1) are typical and lie in
,
the range 2.444(12)–2.455(14) A, cf. between 2.29 and

148
W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
contacts to the ‘core’ oxygen atom) and the ThO8
square antiprism are all vertex linked. The sodium
cations are all octahedrally coordinated (though signifi-
cantly distorted) and edge-sharing. Only two of these
NaO octahedra contact the polyoxo anion, in each
6
case sharing an oxo ligand apex [O(3) and O(4), bonded
to W(3) and W(4), respectively], approach to the other
oxo ligands being by water molecules.
The Raman spectrum of solid Na [ThW O ]·28H O
8
10 36
2
is shown in Fig. 3(a) and that of its saturated aqueous
solution in Fig. 3(b). Although there is substantial
solid-state splitting in the w(WꢁO) region near 1000
−
1
cm
the general similarity of the two profiles suggests
retention of the structure of the anion in aqueous
solution.
2
.3. Oxidations with H O catalysed by
2 2
III
9−
IV
8−
[Ln W O ]
and [M W O ]
10 36
(M=Ce, Th)
10
36
2.3.1. With sodium salts
Our earlier work on catalysis of oxidations by
n−
sodium salts of [LnW O ]
was incomplete since a
10
36
variety of sodium and mixed sodium-tetra-alkylammo-
nium salts were used [10]. In Table 2 we report oxida-
tions of benzyl alcohol to benzaldehyde and cyclo-
hexanol to cyclohexanone, catalysed by sodium salts of
III
9−
IV
8−
[
Ln W O ]
and [M W O ]
(M=Ce, Th) in
1
0
36
10 36
CHCl –H O in the absence of a phase-transfer
3
2
2
reagent. Oxidations of 2-octanol by H O in the pres-
2
2
ence of these species gave only small yields of 2-oc-
tanone (B5%) as did oxidations of cyclo-octene to
cyclo-octene oxide (B10%). These species do catalyse
oxidations with H O2 but not very efficiently, pre-
2
sumably due to the monophasic nature of the solvent in
Fig. 2. Polyhedral representation of the octa-anionic [ThW10O36]8−
which they are soluble (aqueous H O ). Oxidations of
2 2
benzyl alcohol give only benzaldehyde; no attack on the
benzene ring is observed, although others have reported
that in some cases of oxidation of this alcohol, ring
attack may be observed [21].
moiety showing the vertex sharing of the square pyramidal WO and
square antiprismatic ThO8 units.
5
IV
2
.32 A in Na [U W O ]·30H O [19]. The WꢀO dis-
,
8
10 36
2
III
9−
IV
8−
tances fall, as expected, into four distinct groups with
The [Ln W O ]
and [M W O ]
salts are
10
36
10 36
those to the terminal oxo ligands in the range
made from Na [WO ] and the lanthanide or thorium
2
4
1
.703(14)–1.742(14) A
tungsten atoms 1.91(2)–2.013(14) A
tungsten and thorium 1.785(14)–1.792(14) A
those to the ‘core’ oxygen atom are between 2.312(13)
and 2.354(14) A. It is interesting to note that the ‘core’
oxygen atom (O) lies only 0.04 A out of the plane of its
,
, those bridging between two
, those between
whilst
nitrate at pH 7.3; use of a Na [WO ] solution at pH 7.3
2 4
,
in the absence of lanthanide or thorium gives the
‘blank’ species referred to in Table 2, and clearly this is
,
III
9−
as effective
a
catalyst as [Ln W O ]
or
10
36
17
IV
8−
183
,
[M W O ] . From published
W and O-NMR
10 36
,
data it is known that in aqueous tungstate at pH
7.0–7.5 the predominant species are the interchanging
equatorially bonded tungsten atoms W(2), W(3), W(4)
and W(5) in the direction of W(1); the non-bonded
6
−
10−
paratungstates [W O ]
and [H W O ]
[22,23].
7
24
2
12 42
OꢀTh distance is 3.07 A. The geometries at each tung-
,
The Raman spectra of such paratungstates in solution
2
−
sten centre exhibit the normal distortion with the metal
atom being displaced in the direction of its oxo ligand.
We have observed this effect in other oxotungstates and
polyperoxotungstates [20].
(made by adjusting aqueous [WO4] to pH 7) are very
different from that of aqueous tungstate, but we find
that paratungstate solutions in excess H O give Raman
2
2
spectra which am very similar in profile to that [24] of
2
−
In the context of the ‘polyhedral’ structure (Fig. 2),
[W O (O ) (H O) ] . The latter species is known to
2 3 2 4 2 2
the five WO square pyramids (discounting the long
catalyse the oxidation in H O of alcohols [24–26] and
2 2
5

W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
149
2
−
of alkenes [25]. Further, we show in Fig. 3(c) the Raman
spectrum of a solution of Na [ThW O ]·28H O in
[W O (O ) (H O) ]
these systems.
is the effective catalyst for all
2
3
2 4
2
2
8
10 36
2
excess H O ; this is essentially identical in profile with
2
2
−
1
that of paratungstate in H O (the band at 880 cm
is
2.3.2. With quaternary phase-transfer amine cations
Kera et al. have made number of salts
R H [LnW O ] and R [Ce W O ] (R=cetylypyri-
2
2
due to free H O ), while this in turn is very similar to that
a
2
2
9
−
IV
of [SmW O ]
in H O [10]. We conclude that the
1
0
36
2 2
7
2
10 36
8
10 36
2
−
+
spectrum in Fig. 3(c) arises from [W O (O ) (H O) ]
.
dinium, (Me(CH ) NC H ) ) [4–8], finding their ele-
2
3
2 4
2
2
2 15 5 5
III
Of the strongest Raman bands in Fig. 3(c), that at 968
mental analyses to be R H [Ln W O ] (Ln=La, Ce,
7 2 10 36
−
1
cm arises from w(WꢁO) and those at 858, 622 and 566
Pr, Nd, Sm, Eu, Gd, Dy, Ho, Er, Yb) [6,8] and the
cerium(IV) complex as R [CeW O ] [5]. They have
−
1
s
cm
are likely to be due to w(OꢀO), w (W(O )) and
2
8
10 36
as
w (W(O )), respectively of the coordinated peroxo lig-
used them as catalysts with CHCl –aqueous H O for
2
3 2 2
18
ands, using our previous assignments for normal and
substituted [W O (O ) (H O) ]
O
oxidations of alcohols and alkenes [4–8]. We have
2
−
III
9−
[24].
made such salts of [Ln W O ]
(Ln=La, Ce, Pr,
2
3
2 4
2
2
10 36
IV
8−
From these Raman data and the fact that
Sm, Eu, Gd, Dy, Ho, Er, Lu) and [M W O ]
10
36
9
−
paratungstate in H O is as effective as [LnW O ]
(M=Ce, Th). Our analyses agree in general with those
of Kera et al. although the samarium and holmium
complexes analyse as R H[LnW O ], and the cerium
2
8−
2
10 36
IV
and [M W O ]
(M=Ce, Th) as oxidation cata-
lysts (as is also the case for their salts with quaternary
ammonium cations, see Section 2.3.2) it is clear that
1
0
36
8
10 36
and thorium complexes as R [MW O ].
8
10 36
Fig. 3. (a) Raman spectrum of solid Na [ThW10O36]·28H O, (b) Raman spectrum of a saturated aqueous solution of Na [ThW10O36]·28H O, (c)
8
2
8
2
3
3
Raman spectrum of a solution of an aqueous solution of Na [ThW10O36]·28H O (0.2 g in 5 cm ) and 0.2 cm aqueous 30% H O . All spectra run
8
2
2
2
with a frequency-doubled Nd-YAG laser at 15 mW power.

150
W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
Data on these salts as catalysts for oxidations of
equally good results, once more suggesting that lan-
thanides are not necessary for these oxidations. We
were unable to isolate pure samples of the n-hexylam-
monium salts although the lanthanum salt gave analy-
benzyl alcohol, cyclohexanol, 2-octanol and cyclo-
octene are given in Table 2. At first we used the
literature conditions (a 6 h treatment with 1:1 CHCl3–
H O ) [6] but found that oxidations for 3 h gave very
n
2
2
ses close to the formula (( C H ) N) H [LaW O ].
6
13 4
5
4
10 36
2
−
similar yields of products. In general the trends in our
yields of oxidation products are similar to those re-
ported [4–8] but, unlike the previous work, we report
catalytic turnover numbers, calculated using the molec-
ular weights of the lanthanopolyoxotungstates. The
previous workers did not carry out ‘blank’ runs in
which a polyoxotungstate was used in place of a lan-
Again, [W O (O ) (H O) ]
is likely to be the effec-
2
3
2 4
2
2
tive catalyst in all these reactions rather than
III
9−
IV
8−
[
Ln W O ]
or [M W O ]
10 36
.
1
0
36
III
11−
2
.4. Salts of [Ln {PW O } ]
11 39 2
Studies on the oxidation of cyclohexanol to cyclohex-
thanopolyoxotungstate, apart from showing in one in-
III
11−
anone in the presence of [Ln {PW O } ]
(Ln=
11
39 2
n
stance that ( Bu N) [W O ] was less effective than
4
4
10 32
La, Pr, Nd, Sm, Gd) with H O in various solvents
2
2
R H [LnW O ] [8]. We carried out ‘blank’ oxidations
7
2
10 36
have recently been reported and it was suggested that
these are effective catalysts [27]. We have, however,
in the absence of lanthanide by adding cetylpyridinium
2
−
chloride to [WO4]
at pH 7.3, conditions analogous to
31
shown previously from P-NMR and Raman data,
those used for preparation of R H [LnW O ], after
7
2
10 36
that catalysis of the oxidation of primary alcohols to
aldehydes, secondary alcohols to ketones and alkenes
which H O was added. The product (likely to be a
2
2
mixture of cetylpyridinium salts of the paratungstates
III
11−
to epoxides by [Ln {PW O } ]
(Ln=Y, La, Ce,
1
1
39 2
6
−
10−
[
W O ]
and [H W O ]
), gives yields and
turnovers comparable — and in some cases superior
to those found for R H [LnW O ] and
R [Ce W O ]. This suggests, as foreshadowed in our
IV
10−
7
24
2
12 42
Pr, Sm, Tb, Yb) and [Ce {PW O } ]
in CHCl3–
1
1
39
2
n
H O in the presence of (( C H ) N)Cl is effected by
2
2
6
13 4
—
3−
7
2
10 36
[
(
(PO ){WO(O ) } ]
H O)}]
2
,
[(PO ){WO(O ) } {WO(O ) -
4 2 2 2
4
2 2
4
2 2
2−
IV
8
10 36
3−
and [(PO (OH)){WO(O ) } ] . These latter
3 2 2 2
earlier work [10] and with the oxidations above for
sodium salts that the lanthanide has no catalytic influ-
ence — in fact in some cases it may be inhibitory.
Table 2 indicates that cerium and heavier lanthanide
complexes are generally somewhat less effective as cata-
three species are formed by degradation of
III
11−
IV
10−
[
Ln {PW O } ]
and [Ce {PW O } ]
by
1
1
39
2
11 39 2
H O2 rather than by the lanthanophosphopolyoxo-
2
tungstates themselves [11].
In Table 3 we report for the first time P[ H] MAS
3
1
1
III
9−
lysts than other [Ln W O ]
species for oxidation
1
0
36
III
NMR data for solid K [Ln {PW O ) ]·nH O salts
11
11 39 2
2
of benzyl alcohol, cyclohexanol and 2-octanol. On the
grounds of its known redox potential [12],
(
Ln=Y, La, Ce, Pr, Nd, Sm, Eu, Tb, Ho, Er, Yb, Lu),
IV
31
1
for K [Ce {PW O } ]·nH O and P[ H] spectra for
their solutions in aqueous H
III
9−
8−
10
11 39
2
2
[
[
Ce W O ]
is probably oxidised by H O2 to
1
0
36
2
31
1
IV
2 2
O . Our P[ H] data for
Ce W O ] . The heavier lanthanides have smaller
1
0
36
their aqueous solutions agree well with literature values
sizes due to the lanthanide contraction, while the radius
of cerium(IV) is 0.15 A smaller than that of cerium(III).
31
[
28,29]. The P resonances are shifted little from solid
,
to solution, suggesting that the structures of the anions
of the solids are retained by their solutes; likewise, the
Differences in catalytic effectiveness may arise from
stronger LnꢀO bonding via a greater partial ionic char-
acter between the smaller central atom to the two
W O moieties, thus rendering degradation by H O to
III
Raman spectrum of K [Sm {PW O } ] is very simi-
11
11 39 2
lar to that of its aqueous solution [11], also suggesting
minimal change of structure in aqueous solution. Only
5
18
2
2
polyperoxotungstates more difficult; such an effect is
likely to be even more pronounced for cerium(IV) since
a greater central atom charge is also involved. The
31
one P peak is observed for the aqueous solutions,
suggesting equivalence of the positions of the two phos-
phorus atoms, with chemical shifts varying from +69
(Er) to −201 ppm (Tb). For some solid samples two
IV
8−
complex [Th W O ] is comparable in catalytic effi-
1
0
36
III
9−
ciency with [Ln W O ]
so this latter factor is
clearly less important, perhaps because thorium(IV) is
slightly larger (ca. 0.05 A) than cerium(IV).
Oxidations of benzyl alcohol, cyclohexanol, 2-oc-
1
0
36
31
peaks were observed in the MAS P-NMR, probably
due to slight asymmetry in the structure with respect to
,
the two {PW O } moieties. Although we and others
11 39
III
9−
have been unable to obtain single crystals of any of
tanol and cyclo-octene with [Ln W O ]
and
10
36
IV
8−
[
M W O ]
(M=Ce, Th) in the presence of the
these salts suitable for X-ray analysis, in the related
1
0
36
n
IV
phase transfer catalyst (( C H ) N)Cl in CHCl –
complex Cs12[U {GeW11
O
39
}
2
]·13H O
2
the crystal
6
13 4
3
aqueous H O are listed in Table 2. Yields and
structure [30] shows that the {GeW11
O39} units form a
2
2
turnovers are considerably superior to those observed
for cetylypyridinium salts, but again ‘blank’ oxidations
distorted square antiprismatic arrangement of their
donor oxygen atoms around the central uranium atom
such that there are slight inequivalences of the two Ge
atoms.
2
−
carried out using [WO4]
initially at pH 7.3 (before
addition of H O ) in the absence of lanthanide give
2
2

W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
151
31
1
In Table 3 the P[ H]-NMR spectra of a number of
III
11−
other [Ln {PW O } ]
(Ln=Y, La, Ce, Tb, Lu)
11
39 2
IV
10−
and [Ce {PW O } ]
complexes in H O are given.
11
39
2
2
2
On measuring the NMR spectra as soon as possible
after addition of H O , in each case a single peak is
2
2
seen significantly shifted from that of the parent lan-
thanophosphopolyoxotungstate. These peaks may arise
4
−
from mononuclear complexes, [Ln{PW O }]
and
11
39
IV
3−
[
Ce {PW O }] , formed as intermediates during
11
39
degradation of their binuclear parents by H O : such
2
2
31
P shifts are also observed for a 1:1 mixture of the
lanthanide nitrate and [PW O ] . Although we have
7
−
11
39
been unable to isolate any solids, formation of related
mononuclear species has been proposed by Peacock
and Weakley [12]; Maksimov et al have isolated a
complex formulated as H [PW CeO ] [31], though this
5
11
40
was not characterised.
After 24 h these peaks disappear and are replaced by
others, listed as A, B, C and D in Table 3. Peak A
3
−
probably arises from (PO4) . Peaks B, C and D are
most clearly seen for the samarium complex, and their
1
83
W satellites allowed their identification as being due
to the phosphopolyperoxotungstates [(PO ){WO-
4
3−
3
−
(
O ) } ] , [(PO ){WO(O ) } {WO(O ) (H O)}]
and
2
2
4
4
2 2
−
2
2 2
2
2
[(PO (OH)){WO(O ) } ] , respectively [11]. Similar
3
2 2 2
assignments for B, C and D have been made for
7
−
3−
[
PW O ]
[32] and for [PW O ]
[33] in H O .
1
1
39
12 40
1−
2 2
1
Raman spectra of [Sm{PW O } ]
also showed that
11
39 2
this anion is degraded in aqueous H O2 to give a
2
spectrum very similar to that of the above phospho-
polyperoxotungstates [11].
III
11−
It is noteworthy that, for [Ce {PW O } ]
and
11
39 2
IV
10−
31
[
Ce {PW O } ]
the P resonances change much
more slowly than do those of [Ln {PW O } ]
11 39 2
III
11−
11
39 2
after addition of H O , consistent with our observation
2
2
that the cerium species are less effective oxidation cata-
III
9−
lysts [11] than the other [Ln W O ]
complexes. As
10
36
9−
III
IV
8−
is the case for the [Ce W O ] –[Ce W O ]
1
0
36
10 36
couple, redox potential data suggest that H O oxidises
2
2
10−
III
11−
IV
[
Ce {PW O } ]
to [Ce {PW O } ]
[12].
1
1
39
2
11 39 2
These species catalyse oxidations of alcohols and alke-
nes with H O much more slowly than do other
2
2
III
11−
[
[
Ln {PW O } ]
species, as seen above for
1
1
39
9−
2
III
IV
8−
Ce W O ]
and [Ce W O ]
compared with
complexes. Presumably, again, the
breakup by H O of these binuelear species is retarded
10
36
10 36
9
−
other [LnW O ]
1
0
36
2
2
by the smaller, more highly charged cerium(IV) centre,
31
and this is reflected in the P-NMR data.
3
. Conclusions
The sodium, cetylpyridinium and n-hexylammonium
III
9−
IV
8−
salts [Ln W O ] and [M W O ] (M=Ce, Th)
10
36
10 36
have been evaluated as catalysts for the oxidation of

152
W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
Table 3
3
1P-NMR data for K11[LnIII{PW11O39} ]·nH O salts
2
2
[Ln{PW11O39}2]11− [Ln{PW11O39}]4−
[Ln{PW11O39}2]11− in H O
2 2
a
Ln:
K11[Ln{PW11O39}2]
(solid)
(aqueous solution)
(aqueous solution)
A e
B f
C g
D h
E i
Y
La
−13.3
−12.2
−11.4
−18.2
−12.8
−13.4
−23.3
−15.3
−0.3
−14.7
−11.7
−0.2
0.16
−0.65s b
−0.41s
0.82s
−1.46s
−1.16s
2.00sh
2.20sh
−2.84s
−2.43s
−11.8, −12.1
−16.1, −20.8
−13.3, −13.4
−14.4, −17.4
III
c
Ce
Ce
Pr
Nd
Sm
Eu
Tb
Ho
Er
1.36
1.34
IV
−13.5
−19.5
−25.9
−15.7
4.6
−237.6
−96.9
87.1
0.85s
−14.9
−0.09 −0.63s
−1.14
−2.62s
−174.9
−201.6
−96.8
69.0
−16.56b d −40.28b
Yb
Lu
[
29.9
35.7
−12.72
30.0
−12.65
−3.18b
−0.49s
−0.32s
−6.06b
−1.21s
− 13.94b
−2.39s
0.03
0.24
PW11O39]7 –H O2
−
2
a
b
c
d
e
f
In excess of 30% H O (H O –W 12:1).
2
183
2
2
2
s=band with two
W satellites.
sh=band with two shoulders from 183W satellites.
183
b=broad band, probably due to
W satellites.
3−
A: (PO4)
.
B: [(PO (OH)){WO(O ) }]2−.
3
2 2
g
h
i
C: [(PO ){WO(O ) } {WO(O ) (H O)}]3−.
4 2 2 2 2 2 2
3−
D: [(PO ){WO(O ) } ]
.
4
2 2 4
E: Other species.
2
31
alcohols and the epoxidation of cyclo-octene in
CHCl –H O . It appears that the lanthanide or tho-
MHz) as H O solution. Solid-state MAS P-NMR
2
spectra were recorded at 121.4 MHz (7.05 T) on a
Bruker MSL300 spectrometer using a standard Bruker
magic angle sample spinning (MAS) probe with double-
bearing rotation mechanism. The samples were studied
as polycrystalline powders in zirconia rotors (4 mm
external diameter) and MAS frequencies at 8–13 kHz
with stability better than 910 Hz) were used.
transverse magnetisation was prepared using a single
pulse excitation and H high-power decoupling was
applied during acquisition. Spectra were recorded at
3
2
2
rium hetero-atom plays no effective role in such cataly-
sis (and indeed in some cases has an inhibitory effect);
2
−
degradation to [W O (O ) (H O) ]
occurs and it is
2
3
2 4
2
2
this species that is almost certainly responsible for all
the oxidations.
3
1
3
1
P[H]-NMR spectra have been measured of solid
(
P
III
IV
K [Ln {PW O } ]·nH O salts, K [Ce {PW O } ]·
1
1
11 39
2
2
10
11 39
2
nH O and of their solutions in water and aqueous
1
2
H O . For the latter, the data suggest that they are
2
2
degraded to phosphopolyperoxotungstates which,
rather than the lanthanophosphopolyoxotungstates, are
responsible for the oxidation catalysis.
31
ambient probe temperature. All P chemical shifts are
given relative to aqueous 85% H PO solution.
Raman spectra were measured on a Perkin–Elmer
760X Fourier Transform Raman instrument with 1064
3
4
1
nm. Nd–YAG excitation with a power of 2 W, and on
a Dilors LabRam Infinity instrument with a 532 nm.
frequency-doubled Nd–YAG excitation at 15 mW. The
GC data were obtained on a Perkin–Elmer Autosystem
instrument using a Perkin–Elmer stainless-steel column
packed with 5% Carbowax 20M on Chromosorb WHO
AW (DCMS treated). Microanalyses were obtained
from North London University (C, H, N) or from
ICPAES measurements on an ARL instrument (Na,
Ln, W) by Mr. B. Coles of the Geology Department,
Imperial College.
4
. Experimental
4
.1. General procedures
Lanthanide salts, sodium tungstate, benzyl alcohol,
-octanol, cyclohexanone and cyclo-octene were pur-
2
chased from Aldrich and hydrogen peroxide (AnalaR
0%) from BDH; all were used without further
purification.
The solution P[ H] data were measured on a JEOL
3
3
1
1
31
1
ESX 270 spectrometer ( P, 109.25 MHz, H, 270.05

W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
III 9−
153
4
.2. Preparation of sodium salts of [Ln W O ]
yellow solution colourless. The mixture was then cooled
to 5°C for a day, after which white crystals of the
product were collected.
10
36
IV
9−
and [M W O ]
(M=Ce, Th)
1
0
36
2
2
−
For the lanthanide complexes an adaptation of the
method of Peacock and Weakley [12] was used. To 5.0
Anal. Found: H, 1.0; K, 11.0; O , 18.6. Calc.: H,
1.2; K, 11.3; O2 , 18.4%.
2
−
3
g (15 mmol) of Na [WO ]·2H O in 5 cm of water,
2
4
2
adjusted to pH 7.0 with glacial acetic acid, was added
the lanthanide nitrate (1.5 mmol) in 2 cm of water and
4.4. Preparation of cetylpyridinum salts of
3
III
9−
IV
9−
[Le W O ]
and [M W O ]
10 36
(M=Ce, Th)
10
36
the solution heated to 90°C with stirring until dissolu-
tion was complete (ca. 15 min). The solution was
filtered hot and left to cool; crystals of the product were
filtered off and dried.
3
Cetylpyridinium chloride (1.0 g, 2.8 mmol) in 10 cm
of water was added to a solution of 1.0 g (0.32 mmol)
of Na [LnW O ]·nH O in 5 cm of water. The precip-
3
9
10 36
2
The ‘blank’ material (i.e. a mixture of sodium
itate was filtered, washed with water and dried. The
solution was left to cool and the product filtered off
and dried. The ‘blank’ material (i.e. a mixture of the
6
−
10−
paratungstates ([W O ]
and [H W O ]
[22,23])
7
24
2
12 42
was made by following the above procedure but omit-
ting addition of the lanthanide salt.
6
−
cetylpyridinium salts of the paratungstates [W O ]
7
24
10−
The thorium salt was made by an adaptation of the
literature method [16,17] by using exactly the procedure
as above, thorium. nitrate Th(NO ) ·6H O replacing
and [H W O ]
[22,23]) was made by following the
2
12 42
above procedure but omitting addition of the lan-
thanide salt.
3
4
2
the lanthanide nitrate. The product was recrystallised
from water.
4.4.1. Analytical data
In the analyses below R=cetylpyridinium,
+
2 15 5 5 21 38
+
4.2.1. Analytical data
(Me(CH ) (NC H ) , C H N .
Na [YW O ]·20H O. Yield 10%. (Anal. Found: Na,
R H [YW O ]. Yield 63%. Anal. Found: C, 38.3;
7 21 10 31
9
10 36
2
7.1; Y, 2.5. Calc.: Na, 6.7; Y, 2.9%).
H, 6.0; N, 2.1. Calc.: C, 38.1; H, 5.9; N, 2.1%.
Na [LaW O ]·28H O. Yield 51%. (Anal. Found: La
R H [LaW O ]. Yield 62%. Anal. Found: C, 39.6;
9
10 36
2
7
2
10 36
4
5
.3; Na, 6.3; W, 56.9. Calc.: La, 4.3; Na, 7.1; W,
6.3%).
H, 6.4; N, 1.6. Calc.: C, 37.7; H, 5.8; N, 2.0%.
R H [CeW O ]. Yield 60%. Anal. Found: C, 35.0;
7
2
10 36
Na [CeW O ]·38H O. Yield 26%. (Anal. Found:
Ce, 3.8; Na, 5.8; W, 51.1. Calc.: Ce, 4.7; Na, 6.0; W,
H, 6.0; N, 2.0. Calc.: C, 37.6; H, 5.8; N, 2.1%.
9
10 36
2
IV
R [Ce W O ]. Yield 62%. Anal. Found: C, 40.7;
8
10 36
53.3%).
H, 6.2; N, 2.2. Calc.: C, 40.4; H, 6.1; N, 2.2%.
Na [PrW O ]·28H O. Yield 55%. (Anal. Found:
R H[SmW O ]. Yield 63%. Anal. Found: C, 39.3;
9
10 36
2
8
10 16
Na, 7.8; Pr, 3.4; W, 55.9. Calc.: Na, 6.2; Pr, 4.3; W,
H, 6.0; N, 2.0. Calc.: C, 40.3; H, 6.1; N, 2.2%.
56.3%).
R H [EuW O ]. Yield 62%. Anal. Found: C, 37.8;
7
2
10 36
Na [SmW O ]·30H O. Yield 25%. (Anal. Found:
H, 5.6; N, 1.8. Calc.: C, 37.6; H, 5.8; N, 2.1%.
9
10 36
2
Na, 7.0; Sm, 4.8; W, 54.0. Calc.: Na, 6.3; Sm, 4.5; W,
R H [GdW O ]. Yield 60%. Anal. Found: C, 37.8;
7
2
10 36
55.5%).
H, 5.7; N, 2.0. Calc.: C, 37.3; H, 5.8; N, 2.1%.
Na [EuW O ]·20H O. Yield 36%. (Anal. Found:
R H [DyW O ]. Yield 63%. Anal. Found: C, 37.3;
9
10 36
2
7
2
10 36
Eu, 4.4; Na, 6.5; W, 51.4. Calc. Na, 6.3; Sm, 4.65; W,
H, 6.0; N, 2.2. Calc.: C, 37.3; H, 3.8; N, 2.1%.
56.3%).
R H[HoW O ]. Yield 62%. Anal. Found: C, 39.7;
8
10 36
Na [GdW O ]·30H O. Yield 20%. (Anal. Found:
H, 6.6; N, 2.1. Calc.: C, 40.2; H, 6.1; N, 2.2.
9
10
3
2
Gd, 4.5; Na, 6.7; W, 50.5. Calc.: Gd, 4.7; Na, 6.2; W,
R H [ErW O ]. Yield 64%. Anal. Found: C, 37.6;
7
2
10 36
55.4%).
H, 5.7; N, 2.0. Calc.: C, 37.4; H, 5.8; N, 2.1%.
Na [HoW O ]·30H O. Yield 20%. (Anal. Found:
R H [LuW O ]. Yield 62%. Anal. Found: C, 37.6;
9
10 36
2
7
2
10 36
Ho, 4.2; Na, 7.4; W, 48.4. Calc.: Ho, 4.9; Na, 6.2; W,
H, 5.8; N, 2.1. Calc.: C, 37.4; H, 5.7; N, 2.1%.
55.3%).
R [ThW O ]. Yield 62%. Anal. Found: C, 39.4; H,
8
10 36
6
.3; N, 1.7. Calc.: C, 39.3; H, 6.1; N, 2.2%.
4
.3. Preparation of K [W O (O ) (H O) ]·2H O
2
2
3
2 4
2
2
2
4
.5. Preparation of potassium salts of
III
11−
IV
10−
The method is adapted from that of Stomberg [34].
[Le {PW O } ]
and [Ce {PW O } ]
11
39
2
11 39 2
To potassium hydroxide (1.0 g, 17 mmol) in the mini-
mum amount of water was added WO ·H O (2 g. 7
The method of Haraguchi et al. [35] was used. Hy-
drated H [PW O ]·nH O (4.3 g, 1.5 mmol) was dis-
3
2
3
12 40
2
3
mmol) and the mixture stirred until all of the solid had
dissolved. The solution was filtered and hydrogen per-
solved in hot water (5 cm ) and a solution (0.75 mmol
in 2 cm of water) of the lanthanide salt was added
3
3
oxide (7 cm of a 30% aqueous solution) was added
(LaCl ·7H O,
Ce(NO ) ·6H O,
(NH ) [Ce(NO ) ],
3
2
3 3
2
4 2 3 6
dropwise and sufficient dilute HCl added to render the
Ln(NO ) ·6H O (Ln=Y, Pr, Sm, Tb, Yb). A solution
3 3 2

154
W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
3
of potassium acetate (5.0 g, 50 mmol) in water (5 cm )
was added, and the pH adjusted to 7 by addition of
acetic acid dropwise with vigorous stirring. After filtra-
tion the solution was left in the refrigerator and crystals
of the complexes recovered.
strate (benzyl alcohol, 2-octanol, cyclohexanone or cy-
n
clo-octene; 5 mmol) and (( C H ) N)Cl (0.045 g, 0.12
6
13 4
mmol) and the mixture heated to 60°C. A similar
procedure was used for the ‘blank’ but no lanthanide
3
nitrate was used. Then 30% aqueous H O (20 cm ) was
2
2
added slowly over a period of 10 min. Refluxing was
continued, with stirring, for 3 h. The mixture was then
cooled and the organic layer separated. The aqueous
4
.5.1. Analytical data
K [Y{PW O } ]·16H O. Yield 84%. Anal. Found:
1
1
11 39
2
2
3
layer was washed with chloroform (3×25 cm ) and the
K, 7.7; P, 0.9; Y, 1.4. Calc.: K, 7.0; P, 1.0; Y, 1.4%.
K [La{PW O } ]·35H O. Yield 88%. Anal. Found:
K, 6.3; La, 1.8; P, 0.9. Calc.: K, 6.6; La, 2.1; P, 1.0%.
combined chloroform solutions dried over MgSO and
4
1
1
11 39
2
2
filtered. The filtrate was then analysed by GC.
III
K [Ce {PW O } ]·20H O. Yield 78%. Anal.
1
1
11 39
2
2
Found.: Ce, 1.9; K, 7.0; P, 0.8. Calc.: Ce, 1.9; K, 6.8; P,
.0%.
K [Ce {PW O } ]·20H O. Yield 51%. Anal.
5
. Crystallography
1
IV
5.1. Crystal data for Na
[ThW10O36]·28H O
2
1
0
11 39
2
2
8
Found: Ce, 2.1; K, 6.4; P, 0.8. Calc.: Ce, 2.2; K, 6.2; P,
.0%.
K [Pr{PW O } ]·20H O. Yield 69%. Anal. Found:
1
M=3334.9, monoclinic, space group C2/c (no. 15),
a=18.144(3), b=18.638(3), c=18.443(2) i=
A,
,
1
1
11 39
2
2
3
−3
K, 6.9; P, 0.9; Pr, 1.8; W, 64.2. Calc.: K, 6.8; P, 1.0; Pr,
.2; W, 64.4%.K [Sm{PW O } ]·20H O. Yield 69%.
95.79(1)°, V=6205(1) A
,
, Z=4, Dcalc=3.570 g cm
v(Mo–K )=21.02 cm , F(000)=5944, T=203 K;
,
−
1
2
a
1
1
11 39
2
2
Anal. Found: K, 6.8; Sm, 2.3; P, 0.9; W, 64.1. Calc.: K,
clear blocks, 0.40×0.23×0.23 mm, Siemens P4/PC
diffractometer, ꢀ-scans, 5473 independent reflections.
The structure was solved by direct methods and the
6
1
2
6
.8; Sm, 2.4; P, 1.0; W, 64.3%.K [Tb{PW O } ]·
11 11 39 2
2H O. Yield 87%. Anal. Found: K, 6.9; P, 0.9; Tb,
2
.3; W, 66.4. Calc.: K, 7.0; P, 1.0; Tb, 2.6; W,
5.7%.K [Yb{PW O } ]·17H O. Yield 89%. Anal.
non-hydrogen atoms were refined anisotropically using
2
full-matrix least-squares based on F to give R
=0.056,
1
1
11 39
2
2
1
Found: K, 6.8; P, 0.9; W, 64.6; Yb, 2.3. Calc.: K, 6.9;
P, 1.0; W, 64.6%; Yb, 2.8%.
wR
[ꢀF
2
=0. 136 for 4349 independent observed reflections
ꢀ\4|(ꢀF ꢀ), 2q550°] and 386 parameters [36].
o
o
III
9 10 36 2
4
.6. Oxidations with Na [Ln W O ]·nH O and
6. Supplementary material
N [ThW O ]·28H O
8
10 36
2
Full crystallographic details have been deposited with
the Cambridge Crystallographic Data Centre, CCDC
no. 141115. Copies of this information can be obtained
free of charge from The Director, CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK (fax: +44-1223-
336033; e-mail: deposit@ccdc.cam.ac.uk or www: http:/
3
In a 100 cm flask was placed the catalyst (0.08
3
mmol), CHCl (20 cm ) the substrate (benzyl alcohol,
-octanol, cyclohexanone or cyclo-octene; 5 mmol) and
the mixture was heated to 60°C. Then 30% aqueous
H O (20 cm ) was added slowly over a period of 10
3
2
3
2
2
min. Refluxing was continued, with stirring, for 3 h.
The mixture was then cooled and the organic layer
separated. The aqueous layer was washed with chloro-
/
www.ccdc.cam.ac.uk).
3
Acknowledgements
form (3×25 cm ) and the combined chloroform solu-
tions dried over MgSO and filtered. The filtrate was
then analysed by GC.
4
We thank the Calouste Guibenkian Foundation
H.I.S.N.) and the Egyptian Ministry of Education
A.G.F.S.) for grants for postgraduate work. We thank
the University of London Intercollegiate Research Ser-
vice for support of the Raman instruments at Imperial
College and the solid state NMR instrument at Univer-
sity College London, and Dr Abil Aliev, University
College London, for help with the MAS NMR studies.
(
(
III
7 2 10 36
4
.7. Oxidations with R H [Ln W O ] and
IV
R [M W O ] (R=cetylpyridinium, M=Ce, Th)
8
10 36
The procedure was similar to that above, the cety-
lypyridinium salt (0.25 g, 0.6 mmol) replacing
III
Na [Ln W O ]·nH O.
9
11 36
2
III
9−
4
.8. Oxidations with [Ln W O ]
and
10
36
References
IV
8−
n
[M W O ]
(M=Ce, Th) with (( C H ) N)Cl
6 13 4
1
0
36
[
1] Part 6: N.M. Gresley, W.P. Griffith, A.J.P. White, D.J.
Williams, J. Chem. Soc. Dalton Trans. (1997) 89.
3
III
9−
In a 100 cm flask was placed [Ln W O 1
or
1
0
36
IV
8−
3
[
M W O ]
(0.08 mmol), CHCl (20 cm ), the sub-
[2] I.V. Kozhevnikov, Chem. Rev. 98 (1998) 171.
1
0
36
3

W.P. Griffith et al. / Journal of Organometallic Chemistry 607 (2000) 146–155
155
[
[
[
[
[
[
[
3] R. Shiozaki, H. Goto, Y. Kera, Bull. Chem. Soc. Jpn. 66 (1993)
790.
4] R. Shiozaki, A. Inagaki, A. Nishino, H. Kominami, Y. Kera,
Kidorui 26 (1995) 274.
[20] W.P. Griffith, B.C. Parkin, A.J.P. White, D.J. Williams, J.
Chem. Soc. Dalton Trans. (1995) 3131.
[21] K.F. Wedermeyer, Houben-Weyl: Methoden der Organischen
Chemie, vol. 6/1c, Thieme, Stuttgart, 1976, p. 30.
2
5] R. Shiozaki, H. Kominami, H. Kera, Synth. Comm. 26 (1996)
[22] J.J. Hastings, O.W. Howarth, J. Chem. Soc. Dalton Trans.
1
663.
(
1992) 209.
6] R. Shiozaki, A. Inagaki, A. Ozaki, H. Kominami, S. Yamaguchi,
J. Ichihara, Y. Kera, J. Alloy. Compd. 261 (1997) 132.
7] Y. Kera, Y. Mochizuki, S. Yamaguchi, J. Ichikara, H. Komi-
nami, Kidorui 32 (1998) 308.
8] R. Shiozaki, A. Inagaki, H. Kominami, S. Yamaguchi, J. Ichi-
tara, Y. Kera, J. Mol. Catal. 124 (1997) 29.
[
[
23] R.I. Maksimovskaya, K.G. Burtseva, Polyhedron 4 (1985) 1559.
24] N.J. Campbell, A.C. Dengel, C.J. Edwards, W.P. Griffith, J.
Chem. Soc. Dalton Trans. (1989) 1203.
[
[
[
[
[
[
[
[
[
25] A.J. Bailey, W.P. Griffith, B.C. Parkin, J. Chem. Soc. Dalton
Trans. (1995) 1833; 1203.
26] O. Bortolini, V. Conte, F. di Furia, G. Modena, J. Org. Chem.
9] R. Shiozaki, A. Wagaki, A. Nishio, E. Nishio, M. Maekawa, H.
Kominami, Y. Kera, J. Alloys Compounds 24 (1996) 193.
5
1 (1986) 2661.
27] V. Indira, S. Gopinathan, C. Gopinathan, Indian J. Chem. 37A
1998) 816.
28] M.A. Fedotov, B.Z. Pertsikov, D.K. Danovich, Polyhedron 9
1990) 1249.
[
[
[
[
10] W.P. Griffith, R.G.H. Moreea, H.I.S. Nogueira, Polyhedron 15
1996) 3493.
11] N.M. Gresley, W.P. Griffith, A.C. Laemmel, H.I.S. Nogueira,
B.C. Parkin, J. Mol. Catal. 117 (1997) 185.
(
(
(
29] G.M. Maksimov, G.N. Kustova, K.I. Matveev, T.P. Lazarenko,
Sov. J. Coord. Chem. 16 (1990) 117.
30] C.M. Tourn e´ , G.F. Tourn e´ , M.C. Brianso, Acta Crystallogr.
Sect. B 36 (1980) 2012.
31] G.M. Maksimov, R.I. Maksimovskaya, I.V. Kozhevnikov, Russ.
J. Inorg. Chem. 37 (1992) 1180.
32] D.C. Duncan, R.C. Chambers, E. Hecht, C.L. Hill, J. Am.
Chem. Soc. 117 (1995) 681.
12] R.D. Peacock, T.J.R. Weakley, J. Chem. Soc. Sect. A (1971)
1
836.
13] J. Iball, J.N. Low, T.J.R. Weakley, J. Chem. Soc. Dalton Trans.
1974) 2021.
14] T. Ozeki, T. Yamase, Acta Crystallogr. Sect. C 50 (1994) 327.
15] T. Yamase, T. Ozeki, M. Tosaka, Acta Crystallogr. Sect. C 50
1994) 1849.
16] G.H. Marcu, I. Todorut, A. Botar, Rev. Roum. Chim. 16 (1971)
335.
17] A.V. Botar, T.J.R. Weakley, Rev. Roum. Chim. 18 (1973) 1155.
18] L.P. Kazanskii, A.M. Golubev, I.I. Baburinaq, E.A.
Torchenkova, V.I. Spitsyn, Bull. Acad. Sci. USSR Div. Chem.
Sci. (1978) 1956.
(
[
[
(
[
33] L. Salles, C. Aubry, B. Thouvenot, F. Robert, C. Dor e´ mieux-
Morin, G. Chottard, H. Ledon, Y. Jeannin, J.M. Br e´ geault,
Inorg. Chem. 33 (1994) 871.
1
[
[
[34] R. Stomberg, Acta Chem. Scand. 22 (1968) 1076.
[35] N. Haraguchi, Y. Okaue, T. Isobe, Y. Matsuda, Inorg. Chem. 33
(1994) 10125.
[36] SHELXTL PC version 5.03, Siemens Analytical X-Ray Instru-
ments, Inc., Madison, WI, 1994.
[
19] A.M. Golubev, L.A. Muradyan, L.P. Kazanskii, E.A.
Torchenkova, V.I. Simonov, V.I. Spitsyn, Sov. J. Coord. Chem.
3
(1977) 715.