Highly selective epoxidation of styrene using a transition metal-aluminium(III) complex containing the [MeAl(2-py)3]- anion (2-py = 2-pyridyl)

Article abstract of DOI:10.1039/b413488e

The reactions of [MeAl(2-py)₃Li-thf] (1) with FeCl₂ or Cp₂Mn in toluene-thf give simple access to the Group 13-transition metal heterometallic complexes [{MeAl(2-py)₃}₂M] [M = Fe (2), Mn (3)]; complex

Full text of DOI:10.1039/b413488e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Highly selective epoxidation of styrene using a transition metal–
aluminium(III) complex containing the [MeAl(2-py) ] anion (2-py 5
2
3
2
-pyridyl){
a
Carmen Soria Alvarez, Felipe Garc ´ı a, Simon M. Humphrey, Alexander D. Hopkins,*
a
a
a
a
b
a
a
Richard A. Kowenicki, Mary McPartlin, Richard A. Layfield, Robert Raja,* Michael C. Rogers,
a
a
Anthony D. Woods and Dominic S. Wright*
a
Received (in Cambridge, UK) 2nd September 2004, Accepted 7th October 2004
First published as an Advance Article on the web 25th November 2004
DOI: 10.1039/b413488e
2
The reactions of [MeAl(2-py) Li?thf] (1) with FeCl or Cp Mn
using the [MeAl(2-py)
]
3
anion, a ligand which combines the
3
2
2
in toluene–thf give simple access to the Group 13–transition
metal heterometallic complexes [{MeAl(2-py) M] [M 5 Fe
2), Mn (3)]; complex 2 has been shown to be a highly selective
styrene epoxidation catalyst in air.
features of negative charge found in tris-pyrazolylborate ligands
with the different donor characteristics of pyridyl counterparts.
The potential of these heterometallic species as homogeneous
catalysts is illustrated by the selective epoxidation of styrene by 2 in
air.
3 2
}
(
In the past two decades there has been considerable interest in tris-
1
pyridyl ligands of the general type shown in Fig. 1. Interest in
We recently reported the first potential source of a Group 13
2 4
tris-pyridyl ligand [MeAl(2-py) ] . Preliminary studies revealed,
3
2
these and in related tris-pyrazolylborates and -methanes has
however, that as a consequence of the polarity of the C–Al bond
one facet of the behaviour of this anion is its ability to function as
2
focused on their broad applications in coordination, organome-
tallic and bioinorganic chemistry as well as in homogeneous
catalysis. One potential difference between the tris-pyridyl and tris-
pyrazolyl ligands is the better s-donor and p-acceptor properties
a thermally-stable source of [2-py] , reaction of the [MeAl(2-
2
py)3] ion with CuCl giving [{Cu(2-py)}3]‘. Surprisingly, the
4
reactions of [{MeAl(2-py) Li?thf] (1) with FeCl or Cp Mn (1 : 1
3
2
2
1
of pyridine, which should lead to differences in coordination
or 1 : 2 equivalents) follow a completely different course, the
products being the heterometallic compounds [{MeAl(2-py) } Fe]
properties. Modification of the bridgehead groups/atoms (Y)
provides a further potential means by which metal coordination
properties of tris-pyridyl ligands can be tailored, both geome-
trically and electronically. To date, the vast majority of studies of
ligands of this type have involved those containing non-metallic
3
2
(2) and [{MeAl(2-py) } Mn] (3) in which transfer of the [MeAl(2-
2
3
2
py)3] anion intact to the Fe(II) and Mn(II) centres has occurred
(Scheme 1) (ESI{,§). The different outcome of the reaction
involving Cu(I) presumably results from the preference for a linear
1
bridgeheads, commonly Y 5 CX (X 5 H, OH, OR, NH ), N, P
2
ligand coordination geometry in [{Cu(2-py)} ] . The H NMR
3
‘
1
and PLO. Our interest in this area has focused on the
incorporation of p-block metals as the bridgehead. Such species
spectrum of 2 indicates the presence of at least one paramagnetic
species in solution (200–300 K). Magnetic measurements of solid 2
3
5
normally have larger bite angles than their non-metallic counter-
parts and can have variable oxidation states and potential
electrochemical activity at the bridgehead. The coordination of
such ligands to metal centres also provides a simple approach to
heterometallic compounds. We report here the successful applica-
tion of this approach in the synthesis of the Al(III)–transition metal
complexes [{MeAl(2-py) } Fe] (2) and [{MeAl(2-py) } Mn] (3)
confirm that the complex is paramagnetic at room temperature.
Only charge-transfer transitions can be observed in the UV/visible
spectrum of 2 in thf. Two of these absorptions [27 322 (e 5 ca.
2
1
3
8 000 mol dm cm ) and 23 175 (e 5 ca. 17 100 mol dm
21
21
3
21
cm )] have similar energies and extinction coefficients to those
2+
found previously in the solution spectra of [{Y(2-py) } Fe]
3
2
6a
3
2
3 2
(Y 5 CH, PLO). However, unlike the latter an extremely intense,
low-energy transition [19 627 cm (e 5 ca. 26 000 mol dm
21
21
3
21
cm )] is also found for 2, which is responsible for the deep-red
7
colour of the complex. Cyclic voltammetry of 2 shows the
Fig. 1 Structural arrangement found in tris-pyridyl ligands (e.g., Y 5 HC,
P, N).
{
Electronic supplementary information (ESI) available: §synthetic details
and characterisation data for complexes 2 and 3; "details of the catalytic
reaction and kinetic plot for the epoxidation of styrene at 80 uC. See http://
www.rsc.org/suppdata/cc/b4/b413488e/
*adh1002@cam.ac.uk (Alexander D. Hopkins)
rr243@cam.ac.uk (Robert Raja)
dsw1000@cam.ac.uk (Dominic S. Wright)
Scheme 1
1
98 | Chem. Commun., 2005, 198–200
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Scheme 2
6,9
˚
of the Fe–N bonds in 2 [2.054(3) A] compared to the much
˚
longer Mn–N bond distances in 3 [2.291(3)–2.305(3) A]. The bond
lengths in both complexes are, however, consistent with the
presence of high-spin Fe(II) and Mn(II) ions.
¯
Fig. 2 (a) Structure of the Fe(II) complex 2, of crystallographic 3
˚
symmetry. Selected bond lengths (A) and angles (u): Fe(1)–N(1) 2.054(3),
Al(1)–C(1) 1.969(3), Al(1)–C(11) 1.969(3), C(11)–N(1) 1.369(4),
Although a large number of catalytic systems based on a range
of transition metals have been reported for the oxidation of
alkenes, these have normally required expensive oxidants in the
…
Al(1) Fe(1) 3.334(7), intra-ligand
…
N
N
mean 3.01, C(11)–Al(1)–
C(11B) 101.8(1), C(1)–Al(1)–C(11) 116.3(1), C(11)–N(1)–Fe(1) 123.3(2),
N(1)–Fe(1)–N(1A) 180.0, N(1)–Fe(1)–N(1B) 94.2(1), N(1)–Fe(1)–N(1C)
1
0
reactions (such as H O and TBHP). These systems are
2
2
commonly not environmentally friendly, are not selective and
85.8(1). Symmetry: A 2 2 x, 2y, 2 2 z; B 1 2 y, 21 + (x 2 y), z; C 1 + y,
11
require high reactor pressures. Preliminary studies of the use of 2
as a homogeneous catalyst for the industrially important
epoxidation of styrene (Scheme 2) were undertaken at various
temperatures (Table 1) (ESI{,"). Most significant is the selective
formation of styrene oxide in this reaction under extremely mild
conditions using only air as the oxidant. For example, although
the conversion of styrene into styrene oxide is only 45.7% at 65 uC
in 1 h, the product is generated almost exclusively under these
conditions and little 1,2-diol is formed. In addition, no polymers
and no benzaldehyde are generated at this temperature. To the
best of our knowledge, 2 is one of the most selective epoxidation
catalysts for styrene yet observed, in the absence of an added
1
2 (x 2 y), 2 2 z. (b) Structure of the Mn(II) complex 3: Mn(1)–N(1)
2.305(3), Mn(1)–N(2) 2.300(3), Mn(1)–N(3) 2.296(3), Mn(1)–N(4)
2
.295(4), Mn(1)–N(5) 2.300(3), Mn(1)–N(6) 2.291(3), Al(1,2)–C(1,2) mean
…
.983(5), Al(1,2)–C(11–61) 2.010(4)–2.014(3), Mn(1) Al(1,2) mean 3.493,
1
…
intra-ligand N N mean 3.28, Ca–Al–Ca 103.5(2)–104.7(2), N–Mn(1)–N
8.2(2)–179.6(3). H atoms and the molecule of thf in the lattice of 3 have
8
been omitted for clarity.
presence of a quasi-reversible Fe(III)/Fe(II) couple of E1/2 5 20.08 V
together with an irreversible Fe(II)/Fe(I) couple at –1.13 V (relative
6b
to Ag/Ag ).
+
The low-temperature X-ray crystallographic studies of 2 (Fig. 2a)
and 3?thf (Fig. 2b){ show that both complexes have similar
structural arrangements in which octahedral Fe(II) and Mn(II) ions
12,13
oxidant.
For example, the most selective catalyst of this type
previously reported for the epoxidation of styrene gave a 45%
2
are coordinated by two [MeAl(2-py)
3
]
anions. Bis-coordinated
conversion to products with a selectivity of 65% styrene oxide (i.e.,
1
2
transition metal complexes of this type are common for tris-pyridyl
ligands containing non-metallic (C, N and P) bridgeheads, and the
well below the selectivity of 2 in this reaction). At this stage, the
mechanism of the reaction is not known. However, previous
studies suggest that it could involve the initial formation of a
2
+
2+
Fe(II) cations [{N(2-py)
3
}
2
Fe] , [{HOC(2-py)
3
}
2
Fe]
and
6a,8
2+
6a,9a
[{OLP(2-py)
3
}
2
Fe]
have been characterised previously.
binuclear m -O bridged Fe(III) complex as the active species.
2
2
In summary, we have shown that the [MeAl(2-py)3] anion can
However, 2 and 3 are the first examples of such bis-coordinated
complexes containing metallic bridgeheads and are the first
examples of neutral bis-coordinated complexes containing any
type of tris-pyridyl ligand. The neutrality of 2 and 3 makes them
be readily transferred on to transition metal centres, the
compounds 2 and 3 being the first bis-coordinated complexes
involving a metallic scorpionate ligand of this type. The observed
selectivity of the epoxidation reaction of styrene is particularly
worthy of note and provides future promise of the potentially
broader applications of these readily accessible bimetallic main
group–transition metal complexes in catalysis.
9
comparable with classical tris-pyrazolylborate complexes, such as
6a
the Fe(II) complex [{HB(pz) } Fe]. Although the Ca–Al–Ca
3
2
2
angles within the [MeAl(2-py)3] anions of 3 [103.5(2)–104.7(2)u]
are similar to those found in the Li precursor 1 [103.8(1)–
4
1
+
04.4(1)u], there is a noticeable compression of the bridgehead
We gratefully acknowledge the EPSRC (C. S. A., S. M. H.,
F. G., M. McP., M. C. R., D. S. W.), Churchill and Fitzwilliam
angle in 2 [101.8(1)u]. This difference is explained by the shortness
Table 1 Reaction conditions, conversions and product distributions for the reactions of styrene with dry air in the presence of 2. All reactions
were undertaken at a pressure of 45 bar under an inert atmosphere
Product selectivity
Experiment No.
Temp./uC
65 uC
Time/h
Conv. (%)
Epoxide
Diols
Polymers
Benzaldehyde
1
2
3
4
1
4
1
4
1
4
0.5
4
45.7
84.3
46.3
85.0
65.7
88.5
72.0
99.6
96.5
86.5
97.1
85.5
89.5
63.1
39.0
—
3.2
13.5
2.8
14.5
5.1
15.6
14.5
32.3
—
—
—
—
5.3
21.2
18.8
27.5
—
—
—
—
—
—
27.6
40.3
65 uC
80 uC
100 uC
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 198–200 | 199

View Article Online
2
D. L. Reger, Comments Inorg. Chem., 1999, 21, 1; S. Trofimenko,
Chem. Rev., 1993, 93, 943; S. Trofimenko, Scorpionates: The
Coordination Chemistry of Polypyrazolylborate Ligands, Imperial
College Press, London, 1999.
Colleges, Cambridge (Fellowship for A. D. H.), B.P. (R.R.), Clare
College, Cambridge (Denman Baynes Fellowship, R. A. L.) St.
Catharine’s College, Cambridge (Fellowship for A. D. W.), the
States of Guernsey and the Domestic and Millennium Fund (R. A.
K.), and the Cambridge European Trust (F. G.) for financial
support. We also thank Dr. J. Davies for collecting X-ray data on
complexes 2 and 3.
3 M. A. Beswick, M. K. Davies, P. R. Raithby, A. Steiner and
D. S. Wright, Organometallics, 1997, 16, 1109; M. A. Beswick,
C. J. Belle, M. K. Davies, M. A. Halcrow, P. R. Raithby, A. Steiner
and D. S. Wright, Chem. Commun., 1996, 2619; D. Morales, J. P e´ rez,
L. Riera, V. Riera and D. Miguel, Organometallics, 2001, 20, 4517.
4
F. Garc ´ı a, A. D. Hopkins, R. A. Kowenicki, M. McPartlin,
M. C. Rogers and D. S. Wright, Organometallics, 2004, 23, 3884.
a
Carmen Soria Alvarez, Felipe Garc ´ı a, Simon M. Humphrey,
a
a
a
a
Alexander D. Hopkins,* Richard A. Kowenicki, Mary McPartlin,
b
5 Details of the variable-temperature magnetic measurements of 2 and 3
will be included in a future full paper.
6 (a) P. A. Anderson, T. Astley, M. A. Hitchman, F. R. Keene,
B. Moubaraki, K. S. Murray, B. W. Skelton, E. R. T. Tiekink,
H. Toffund and A. H. White, J. Chem. Soc., Dalton Trans., 2000, 3505
a
Richard A. Layfield, Robert Raja,* Michael C. Rogers,
a
a
a
Anthony D. Woods and Dominic S. Wright*
a
a
University of Cambridge, Department of Chemistry, Lensfield Road,
Cambridge, UK CB2 1EW. E-mail: adh1002@cam.ac.uk;
rr243@cam.ac.uk; dsw1000@cam.ac.uk; Fax: +44 1223 336362;
Tel: +44 1223 763122
2+
(Fe–N range 1.947–1.981 A˚ ); (b) For [{HC(2-py) } Fe] and related
3
2
cations reversible processes occur with E1/2 for Fe(III)/Fe(II) ca. + 0.70 V.
7 The characteristic magenta colour of low-spin Fe(II) tris-pyrazolylbo-
b
School of Chemistry, University of North London, Holloway Road,
1
1
London, UK N7 8DB
rates has been ascribed previously to a A A T d–d transition in
1
g
1g
which intensity borrowing occurs with a nearby charge-transfer
absorption: J. P. Jesson, S. Trofimenko and D. R. Eaton, J. Am.
Notes and references
21
Chem. Soc., 1967, 89, 3158 (ca. 19 000 cm , e 5 57–90 mol dm
21
3
2
1
cm ).
8 E. S. Kucharski, W. R. McWinnie and A. H. White, Aust. J. Chem.,
1976, 31, 53.
{
group R3, Z = 3, a = b = 10.2291(5), c = 33.6643(18) A, V = 3050.5(3) A,
Crystal data for 2: C32
¯
H
30Al
2 6
FeN , M= 608.43, rhombohedral, space
˚
˚
2
1
m(Mo–Ka) = 0.438 mm , T = 180(2) K. Data were collected on a Nonius
KappaCCD diffractometer. Of a total of 5072 reflections collected, 670
were independent (Rint = 0.072). The structure was solved by direct
2
9
3
For Fe(II) complexes containing [RB(pz9) ] ligands: (a) J. D. Oliver,
D. P. Mullica, B. B. Hitchman and W. O. Milligan, Inorg. Chem., 1980,
9, 165; (b) D. M. Eichhorn and W. H. Armstrong, Inorg. Chem., 1990,
9, 3607; (c) S. Calogero, G. G. Lobbia, P. Cecchi, G. Valle and
2
1
2
methods and refined by full-matrix least squares on F (G. M. Sheldrick,
SHELX-97, G o¨ ttingen, 1997). Application of the SQUEEZE programme
in PLATON (version 1.07: A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7)
removed unassigned residual electron density due to disordered solvent.
Final R1 = 0.042 [I > 2s(I)] and wR2 = 0.113 (all data). CCDC 249604.
J. Friedl, Polyhedron, 1994, 13, 67; (d) Y. Sohrin, H. Kokusen and
M. Matsui, Inorg. Chem., 1995, 34, 3928; (e) A. L. Rheingold, G. P. Yap,
L. M. L. Liable-Sands, I. A. Guzei and S. Trofimenko, Inorg. Chem.,
1997, 36, 6261; (f) C. Janiak, S. Temizdemir, S. Dechert, W. Deck,
F. Girsaedies, J. Heinze, M. J. Kolm, T. G. Scharmann and
O. M. Zipffel, Eur. J. Inorg. Chem., 2000, 1229; (g) P. Gecchi,
M. Berrettoni, M. Giorgetti, G. G. Lobbia, B. Calogero and L. Slievano,
Inorg. Chim. Acta, 2001, 318, 67; (h) T. Kitano, Y. Sohrin, Y. Hata,
H. Wada, T. Hori and K. Ueda, Bull. Chem. Soc. Jpn., 2003, 76, 1365.
2 6
Crystal data for 3?thf; C36H38Al MnN O, M = 679.62, monoclinic, space
˚
group Cc, Z = 4, a = 9.2658(19), b = 24.167(5), c = 15.877(3) A, V =
˚
553.1(12) A, m(Mo–Ka) = 0.458 mm , T = 180(2) K. Data were collected
21
3
on a Nonius KappaCCD diffractometer. Of a total of 9511 reflections
collected, 4469 were independent (Rint = 0.067). The structure was solved
by direct methods and refined by full-matrix least squares on F (G. M.
Sheldrick, SHELX-97, G o¨ ttingen, 1997). Four residual maxima in the final
2
10 For recent examples, see: (a) M. C. White, A. G. Doyle and
E. N. Jacobsen, J. Am. Chem. Soc., 2001, 123, 7194; (b) L. Espinal,
S. L. Suib and J. F. Rusling, J. Am. Chem. Soc., 2004, 126, 7676 and
references therein.
˚
23
difference Fourier (ca. 2–3 eA ) could not be explained. Repeated data
collection failed to resolve this problem that was attributed to co-
crystallisation of small amounts of an unidentified material. The most
satisfactory refinement was obtained by arbitrarily assigning metal atoms
of 10% occupancy to the four sites of residual electron density. Final R1 =
11 J. R. Monnier, Appl. Catal. A, 2001, 221, 73.
12 Q. Tang, Y. Wang, J. Liang, P. Wang, Q. Zhang and H. Wan, Chem.
2+
0.069 [I > 2s(I)] and wR2 = 0.179 (all data). CCDC 249605. See http://
Commun., 2004, 440 (heterogeneous catalyst involving a Co -exchanged
faujasite zeolite).
www.rsc.org/suppdata/cc/b4/b413488e/ for crystallographic data in .cif or
other electronic format.
1
3 X. Meng, K. Lin, X. Yang, Z. Sun, D. Jiang and F.-S. Xiao, J. Catal.,
003, 218, 460 (heterogeneous catalyst involving Cu (OH)PO and
Cu O(PO ; most selective 14.9% conversion with 67.8% epoxide and
33.2% benzaldehyde).
2
2
4
1
L. F. Szezepura, L. M. Witham and K. J. Takeuchi, Coord. Chem. Rev.,
998, 174, 5.
4
4 2
)
1
2
00 | Chem. Commun., 2005, 198–200
This journal is ß The Royal Society of Chemistry 2005