Asymmetric styrene dimerisation using mixed palladium-indium catalysts

Article abstract of DOI:10.1016/j.tet.2005.06.083

Catalysts formed in situ from mixtures of palladium acetate, indium(III) triflate and a chiral non-chelating bis(phosphite) ligand give good to excellent conversions and reasonable enantioselectivity in the asymmetric dimerisation of styrenes.

Full text of DOI:10.1016/j.tet.2005.06.083

Tetrahedron 61 (2005) 9799–9807
Asymmetric styrene dimerisation using mixed
palladium–indium catalysts
a
b
b,c
Robin B. Bedford,a, Michael Betham, Michael E. Blake, Andres Garces, Sarah L. Millara
*
´
´
and Sanjiv Prasharc
aSchool of Chemistry, University of Bristol, Cantock’s Close, Bristol BS8 1TS, UK
bDepartment of Chemistry, University of Exeter, Exeter EX4 4QD, UK
´ ´ ´
Departamento de Tecnologıa Quımica y Ambiental, E.S.C.E.T., Universidad Rey Juan Carlos, Mostoles (Madrid) 28933, Spain
c
Received 12 May 2005; revised 10 June 2005; accepted 24 June 2005
Available online 18 July 2005
Abstract—Catalysts formed in situ from mixtures of palladium acetate, indium(III) triflate and a chiral non-chelating bis(phosphite) ligand
give good to excellent conversions and reasonable enantioselectivity in the asymmetric dimerisation of styrenes.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
hemi-labile functions.2e One such ligand—ligand 3—
proves to be particularly effective in the asymmetric
hydrovinylation of challenging substrates such as
4-isobutylstyrene.2b,c
The asymmetric hetero-dimerisation of a-olefins with
ethene (Eq. 1)—the asymmetric hydrovinylation
reaction—is a well studied and topical area of research.1,2
It provides access to chiral building-blocks from simple,
cheap precursors with 100% atom economy; that is to say
that all the atoms of all the starting material are incorporated
into the product.
Chiral palladium catalysts can also be exploited, however,
their use is sometimes hampered by the deleterious
isomerisation of the desired chiral products to non-chiral
internal alkenes.2a,7
(1)
Typically, the most active and selective catalysts for this
class of reaction are nickel based with appropriate chiral
ligands.1 Ligands 1–3 are amongst the best ligands yet
reported, giving high activities and good to excellent
enantioselectivities. The original benchmark was set by
Wilke and co-workers who introduced the bis(azaphos-
pholene) ligand 1, which shows excellent activity and
stereoselectivity.3 Not only does this ligand perform well in
classical media, but Leitner and co-workers demonstrated
that catalysts containing 1 can be used in super-critical
CO2.4 Leitner’s group has also shown that phosphoramidite
ligands such as ligand 2, introduced by Ferringa and
co-workers,5 are remarkably active and selective.6 Rajan-
Babu and co-workers have extensively investigated the use
of a range of chiral monodentate phosphines with secondary
In contrast with hydrovinylation, the related asymmetric
homo-dimerisation of alkenes has been somewhat under-
investigated. One example of this class of reaction, which
has been explored to a limited extent is the intramolecular
cycloisomerisation of 1,6-dienes.8 Recently, Leitner and
co-workers demonstrated that nickel catalysts with appro-
priate chiral phosphoramidite or azaphospholene ligands
could give the exo-methylene carbocycle 4 in high
regioselectivity and good enantioselectivity from diethyl
diallylmalonate (Eq. 2).9
Keywords: Palladium; Indium; Catalysis; Styrene; Dimerisation;
Hydrovinylation.
*
Corresponding author. Fax: C44 117 925 1295;
e-mail: r.bedford@bristol.ac.uk
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.06.083

9800
R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
bis-phenols 11a and b, respectively, in the presence of
triethylamine (Scheme 1). Ligand 9 could also be prepared
by the reaction of the bis(chlorophosphite) 12a with (S)-
BINOL in the presence of triethylamine. The analogous
reaction with the bulkier analogue 12b proved unsuccessful.
The compounds 12 were synthesised from the appropriate
diols 11b and c and PCl3 in the presence of triethylamine.
(2)
Intermolecular asymmetric alkene dimerisation reactions
remain rare. One example of a non-selective reaction that is
potentially amenable to the development of an enantiose-
lective version is the dimerisation of styrenes (Eq. 3).
RajanBabu and co-workers found that when a typical nickel
catalyst system is used—[{NiBr(h3-allyl)}2]/PPh3/
AgOTf—poor conversion to the dimer is obtained, along
with substantial formation of polystyrene.2e In contrast,
palladium catalysts appear far more attractive; Shirakawa
and co-workers demonstrated that a catalyst formed in situ
from palladium acetate, triphenylphosphine and indium
triflate is particularly effective.10 We were interested to see
whether an asymmetric version of this reaction could be
realised, and we now report that mixed palladium–indium
systems containing non-chelating chiral bis(phosphites)
show considerable promise.
The 31P NMR spectrum for the ligand 8 shows a singlet at
d 145.6 ppm, very close to that reported for ligand 5
(d 144.8 ppm, C6D6).11 The 1H NMR spectrum shows
distinctive signals for the cyclohexyl ring in addition to
resonances for the aromatic protons. The 31P NMR
spectrum of ligand 9 shows a singlet at d 149.3 ppm, a
little down-field of those for 8 and 5, presumably due to the
presence of the ortho-methyl groups. As well as aromatic
signals, the 1H NMR spectrum of ligand 8 shows two methyl
signals at 1.44 and 2.28 corresponding to the xylyl and
bridgehead methyl groups, respectively.
2.2. Catalysis
(3)
The reaction chosen for the optimisation studies was the
dimerisation of unsubstituted styrene (Eq. 4).
2. Results and discussion
2.1. Synthesis of ligands
(4)
The chiral non-chelating bis(phosphite) ligand 5 was shown
by Faraone and co-workers to furnish the dimetalla-
macrocycles 6 and 7.11 We reasoned that any bimetallic
asymmetric activation of styrene may be best served by such
structural types and therefore, synthesized the analogous
ligands 8 and 9.
We initially performed a brief solvent optimisation study
using ligand 9, the results of which are summarised in
Table 1. The use of 1,4-dioxane gives excellent yield and
selectivity for the desired product dimer E-13a as well as
moderate enantioselectivity. Dichloromethane, DME and
THF give lower conversions and poorer selectivity for the
desired product versus isomerisation and/or oligomerisation
products. No reaction was observed in the absence of
solvent. Having established that the best results are obtained
in the dimerisation with dioxane as solvent, this was then
used for the rest of the studies.
The effect of varying the ratios of palladium, indium and
ligand 9 was examined next and the results are summarised
in Table 2. At 1 mol% loading of palladium and ligand 9, in
the absence of indium triflate, no activity is observed
(Table 2, entry 1), implying either that the active catalyst is
a heterobimetallic species or that the indium triflate is
Table 1. Solvent optimisation
Entry
Solvent
Conversion,
a
%
Yield of
13a, %b
ee %c
1
2
3
4
5
1,4-Dioxane
Dichloromethaned
DME
THF
No solvente
O99
90
82
70
0
99
68
57
45.5
0
30
23
22
18
_
Conditions: Pd(OAc)2 (1.0 mol%), In(OTf)3 (5 mol%), ligand
(1.0 mol%), styrene (4.36 mmol), solvent (5 ml), rt, 18 h.
9
a Conversion of styrene determined by GC (hexadecane internal standard).
b Determined by GC (hexadecane internal standard).
c Determined by HPLC.
d Ligand 9 (0.5 mol%), 2.5 mol% In(OTf)3.
e Ligand 9 (0.5 mol%).
These ligands were prepared by the reaction of the
phosphine chloride 10, derived from (S)-BINOL, with the

R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
9801
Scheme 1. Conditions: (i) Et3N, toluene, K40–90 8C, 18 h; (ii) PCl3, Et3N, toluene, K40 8C–rt, 18 h; (iii) (S)-BINOL, Et3N, toluene, K40 8C–rt, 18 h.
required as a catalyst activator. By 0.5 mol% indium
loading, good activity is seen (entry 3). Holding the
loadings of palladium and ligand 9 at 1 mol% and increasing
the amount of indium triflate leads to an increase in
stereoselectivity to a maximum of 37% at 2.5 mol% loading
(entry 5). Further, increasing the amount of indium leads to
an increase in activity (entries 6–8) but this is sometimes
offset by concomitant erosion in enantioselectivity. It may
be imagined that this is due to a competing dimerisation
mediated by the indium triflate. Indeed, when the reaction is
repeated in the absence of palladium acetate (entry 9), small
amounts of dimer 13a are produced as a racemic mixture.
By contrast, in the absence of the ligand 9, indium triflate
does not give any dimer 13a (entry 10). Some conversion to
the dimer 13a is seen using a palladium–indium mixture in
the absence of the phosphite ligand (entry 11).
very broad peak centred around d 2 ppm demonstrating that
the phosphite donor is perfectly capable of coordinating to
the indium centre. The broadening of the peak is due to the
quadrupolar nature of indium (96% IZ9/2). This is very
similar to the 31P spectroscopic data reported for
[In(OTf)3{P(OiPr}3], which shows a broad peak at
6.6 ppm.12 In addition a sharp peak is seen at d K1.5 ppm
(as well as minor peaks at 21.88, 41.66 and 56.30 ppm)
possibly corresponding to a hydrolysed species.
The spectrum of a 1:1 mixture of palladium acetate and
ligand 9 in dioxane/CDCl3 (3:1) shows two major peaks at d
98.71 and 99.05 as well as smaller peaks at 164.43, 102.03
and K2.23 ppm. The peaks between 98 and 103 ppm are
consistent with the formation of simple phosphite adducts of
Pd(II). The smaller, downfield shift we very tentatively
assign as a palladacycle resulting from C–H activation of
one of the ortho-methyl groups of the ligand by the
palladium acetate,13 whilst the high field resonance is
presumably due to hydrolysis of the ligand. Mixing
palladium acetate, indium triflate and ligand 9 in a 1:2.3:1
The data obtained are consistent with a heterobimetallic
active catalyst, but do such species form in the presence of
ligand 9? A 31P NMR spectrum of a 2.3:1 mixture of ligand
9 and indium triflate in 1,4-dioxane/CDCl3 (3:1) shows a
Table 2. Effect of varying relative ratios of Pd/In/ligand
Entry
Pd mol%
In(OTf)3 mol%
Ligand 9 mol%
Conversion, %a
Yield of 13a, %b
ee %c
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
0
0
1
0
1
1
1
1
1
1
1
1
1
0
0
!1
9
!1
9
nd
0.1
0.5
1
2.5
3
14 (R)
28 (R)
32 (R)
37 (R)
31 (R)
30 (R)
35 (R)
0
87
80
75
87
O99
75
24
55
55
81
80
75
87
O99
43
8
5
10
2.5
5
9
10
11
0
15
—
—
2.5
Conditions: styrene (4.36 mmol), dioxane (5 ml), rt, 18 h.
a Conversion of styrene determined by GC (hexadecane internal standard).
b Determined by GC (hexadecane internal standard).
c Determined by HPLC.

9802
R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
Table 3. Effect of varying Lewis acid co-catalyst
ratio in dioxane/CDCl3 (3:1) led to immediate and
substantial precipitation. A 31P NMR spectrum of this
mixture at a reaction time of 1.5 h shows a broad peak
centred around d 164 ppm, a smaller broad peak at 126 ppm
and a third at around K3 ppm (major species). The high-
field shift is again probably due to hydrolysis of the ligand
but we are not able to completely rule out the possibility that
it may be an indium–phosphite species, due to the broad
nature of the peak. When the reaction is left longer (hours)
the precipitate dissolves and the 31P NMR shows only one
sharp peak at K2.0 ppm. When the reaction is performed in
CDCl3, the precipitate is not observed and the 31P NMR
spectrum recorded within 45 min shows only the broad peak
at 165 ppm and a sharp peak at K1.99 ppm (major species).
Again, over the course of several hours, the 31P spectrum
reveals complete hydrolysis.
Entry
Lewis acid
Conversion,
a
Yield of
13a, %b
ee %c
%
1
2
In(OTf)3
In(OTf)3C2.7
(S)-BINOL
HOTf
Tf2O
Yb(OTf)3
AgOTf
Sc(OTf)3
Sn(OTf)2
La(OTf)3
Zn(OTf)2
InCl3
94
96
94
96
31 (R)
33 (R)
3
4
5
6
7
8
9
10
11
12
77
47
10
17
100
8
17
!1
4
45
43
2
13
67
8
!1
!1
!1
!1
23 (R)
22 (S)
27.5 (R)
21.5 (R)
21 (R)
20 (R)
nd
nd
nd
nd
SmCl3
!1
Conditions: Pd(OAc)2 (1.0 mol%), additive (2.5 mol%), ligand 9
(1.0 mol%), styrene (4.36 mmol), 1,4-dioxane (5 ml), rt, 18 h.
a Conversion of styrene determined by GC (hexadecane internal standard).
b Determined by GC (hexadecane internal standard).
c Determined by HPLC.
Taken together, the spectroscopic data show that palladium
competes effectively with indium triflate for the ligand 9
with little or no evidence for the formation of indium–
phosphite complexes in the presence of palladium. It is also
apparent that the hydrolysis of the ligand 9 is relatively
facile in the presence of either indium triflate or palladium
acetate. The reaction of the model ligand 14 with palladium
acetate (1:1, toluene, rt) again leads to the formation of a
small amount of a compound tentatively assigned as a
palladacycle formed by C–H activation of an ortho-methyl
group as well as substantial amounts (major compound) of
hydrolysis product.
of triflic acid (entry 4) then the opposite enantiomer of 13a
predominates.
A range of alternative Lewis acid co-catalysts was examined
and in all cases, where discernable, the same enantiomer (R)
of product dimer 13a predominates (Table 3, entries 5–12).
Of the other triflate salts tested, only scandium triflate shows
good conversion (entry 7), however, the selectivity to the
desired product is substantially lower than with indium
triflate. Both indium chloride and samarium chloride prove
ineffective.
The effect on activity of changing the chiral ligand was
examined next and the results are presented in Table 4.
Small changes in the non-coordinating bis(phosphite)
ligand’s structure appear to have a substantial impact on
both activity and enantioselectivity (entry 1). One signifi-
cant structural difference between the ligands 9 and 8 is the
absence in the latter of ortho-methyl functions on the
bridging bis-phenol group, indicating that steric bulk in
these positions is a prerequisite for decent performance.
However, as outlined above, attempts to produce analogues
of these ligands with ortho-tert-butyl groups have so far
proved unsuccessful.
If the indium triflate, either free or complexed with ligand 9,
plays a role in enantiodiscrimination then it may be
imagined that the use of a chiral indium analogue would
have an influence on the enantioselectivity of the reaction.
This could be either positive or deleterious, depending on
diastereomeric matching or mis-matching. BINOL-modi-
fied indium species, formed in situ from (S)-BINOL and
InCl3 give good to excellent enantioselectivity in allylation
of aldehydes with allyl tin reagents.14 Therefore, In(OTf)3
was pre-treated with (S)-BINOL in 1,4-dioxane and the
resultant mixture was used in the dimerisation of styrene
with palladium acetate and ligand 9. Surprisingly the results
from this experiment (Table 3, entry 2) do not show
particularly significant deviation from those obtained in a
control experiment performed in the absence of the BINOL
(entry 1). It is possible that the triflic acid produced in the
reaction of BINOL with indium triflate provides an
alternative activation pathway that coincidently gives
similar performance/activity to the non-BINOL modified
catalyst system. In order to test this, anhydrous triflic acid
was used in place of indium triflate in the standard reaction
(entry 3). Conversion remains high but selectivity for the
desired dimer is substantially reduced, as is enantioselec-
tivity. Interestingly, when triflic anhydride is used in place
The ligand 14 can be envisaged as half of the non-chelating
bis(phosphite), 9. As can be seen whilst activity remains
high using 14, selectivity for the dimeric product 13a is
vastly reduced (entry 2). Interestingly, the enantioselectivity
obtained, while reduced, is substantially higher than that
obtained with ligand 8, again supporting the idea that the
ortho-methyl groups are an important structural motif.
Replacing one of the two ortho-methyl groups with a proton
and the other with a tert-butyl leads to an erosion in both
activity and enatioselectivity (ligand 15; entry 3). The
incorporation of a 2-phenyl group leads to increased
selectivity for 13a but at the expense of enantioselectivity
(ligands 16 and 17; entries 4 and 5). Whilst two phenyl
groups in the 2- and 6-positions give substantially improved
selectivity for the dimer 13a compared with methyl groups
(compare entries 6 and 2) this is again is at the detriment of
enantioselectivity. It is interesting to note that, in this case,

R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
9803
Table 4. Effect of varying chiral ligands
Entry
Ligand (or preformed palladium complex)
Ratio Pd:In:ligand Conversion, Yield of
a
ee %c
(mol%)
%
13a, %b
1
1:2.5:1
52
33
5.5 (R)
2
1:2.5:2
93
11
29 (R)
14
3
1:2.5:2
60
10
12 (R)
15
4
1:5:2
75
64
1.5 (S)
16
5
1:5:2
84
63
1.5 (R)
17
18
2
6
1:5:2
O99
69
9 (S)
7
8
9
1:2.5:2
1:2.5:2
1:2.5:2
!1
!1
62
!1
!1
9
nd
nd
19
7 (S)

9804
R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
Table 4 (continued)
Entry
Ligand (or preformed palladium complex)
Ratio Pd:In:ligand Conversion, Yield of
a
ee %c
(mol%)
%
13a, %b
10
11
1:2.5:2
!1
!1
20.5 (R)
(R)
1:2.5:1
10
4
2.5 (S)
(R)
12
13
1:2.5:1
1:2.5:1
27
5
!1
n.d
(S,S)
3
3.5 (R)
(R)
14
1:2.5:1
RZEt!1
0
nd
(all-R)
15
16
RZiPr!1
0
2
nd
9.5 (S)
1:2.5:2
4
(all R)
(R,R)
17
18
1:2.5:1
1:2.5:1
51
89
51
49
12.5 (S)
0.5 (S)
(S,S)-20
19
1:2.5:1
97
73
0.5 (S)
21

R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
9805
Table 4 (continued)
Entry
Ligand (or preformed palladium complex)
Ratio Pd:In:ligand Conversion, Yield of
a
ee %c
—
(mol%)
%
13a, %b
20
Pd:InZ1:2.5
62
0
22
21
Pd:InZ1:2.5
70
2
nd
23
Conditions: Pd(OAc)2 (1.0 mol%), additive (2.5 mol%), ligand 9 (1.0 mol%), styrene (4.36 mmol), 1,4-dioxane (5 ml), rt, 18 h.
a Conversion of styrene determined by GC (hexadecane internal standard).
b Determined by GC (hexadecane internal standard).
c Determined by HPLC.
the opposite enantiomer of 13a is obtained compared with
using ligands 8, 9,14 and 15, despite the fact that they are all
derived from (S)-BINOL. This may be due to the fact that
the 2,6-diphenyl groups of ligand 18 show a pronounced
tendency to undergo p-interactions with a metal centre.15
This may well lead to the adoption of an alternative catalytic
manifold.
substrates. Both 4-fluorostyrene and 2-methylstyrene are
converted to the desired dimers in reasonable yields at rt
(Scheme 2). The enantioselectivity observed in the
dimerisation of 4-fluorostyrene is somewhat lower than
that with styrene and in contrast with the unsubstituted
substrate, the product 13b is formed exclusively as the
Z-isomer. The dimerisation of 2-methylstyrene yields the
E-isomer in an encouraging 56% ee.
Interestingly, in contrast with the monodentate ligands
14–18, the diastereomeric phosphoramidite ligands 2 and 19
show essentially no activity (entries 7 and 8). This is despite
the fact that these ligands are amongst the most effective
tested in asymmetric hydrovinylation reactions with nickel
catalysts.9,6 The other monodentate ligands tested also
proved to be poor (entries 9 and 10) as did the chelating
bisphosphines BPPFA, chiraphos, BINAP, Et-Duphos,
iPr-Duphos and Et-BPE (entries 11–16, respectively). This
is perhaps not surprising since RajanBabu and co-workers
found that the chelating bisphosphines BINAP and Me-
Duphos show no activity in asymmetric hydrovinylation
reactions.2e By contrast (K)-diop (entry 17) shows
moderate conversion with high regioselectivity and some
enantioselectivity. This may be due to increased size and
flexibility of the chelate ring. Indeed the more flexible
bidentate ligands 2016 and 21 (entries 18 and 19) both show
good activity and reasonable selectivity but give the desired
product as essentially racemic mixtures. Both of the
preformed ‘PCP’-pincer complexes 22 and 23 do not yield
the desired dimerisation product (entries 20 and 21).
Scheme 2. a Isolated yield, ee determined by HPLC. b Spectroscopic yield
determined by H NMR, isolated yieldZ20%, ee determined by HPLC.
1
The electron rich substrate 4-methoxystyrene does not give
any of the desired dimerisation product under the standard
conditions, but rather yields polymeric material. It appears
that the polymerisation is catalysed by an indium species;
control reactions with indium triflate with or without ligand
9 also give essentially quantitative conversion to polymeric
materials, whereas a reaction containing palladium acetate
and ligand 9 in the absence of indium leads only to
recovered starting material.
Having established that the initial catalyst system tested—
that formed in situ from palladium acetate, indium triflate
and ligand 9—shows the best performance and enantio-
selectivity. We briefly screened this against selected
3. Conclusions
In summary, we have realised the asymmetric dimerisation
of styrenes using mixed palladium–indium catalysts

9806
R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
supported by chiral non-chelating bis(phosphite) ligands.
While the enantioselectivities obtained with the current
catalyst systems are modest, the performances observed are
considerably better than those using some of the best ligand
systems reported for the related asymmetric hydrovinylation
reaction. In addition, the more successful catalyst systems
identified are amenable to simple modification, which holds
promise for future optimisation of both activity and
selectivity. These studies are ongoing in our group and the
results will be reported at a later date.
130.97, 131.69 (ArH), 132.15, 133.11, 133.36, 133.38,
147.30, 147.54, 147.65, 148.40 (ArC); 31P NMR (121 MHz,
CDCl3) d 149.33.
Method B. A mixture of the appropriate bisphenol 11a
(4.0 g, 16.1 mmol) and freshly distilled PCl3 (10 ml,
115 mmol) in toluene (80 ml) was cooled to K40 8C and
treated drop-wise with Et3N (8 ml, 58 mmol) in toluene
(20 ml). The reaction was allowed to warm to rt overnight.
The mixture was filtered through a pad of Celite, which was
then washed with toluene (2!20 ml) and the solvent
removed from the combined solutions in vacuo to give the
intermediate 12a as a white solid. This was held under
vacuum until all of the PCl3 had been removed as
determined by 31P NMR. Yield 6.67 g; 86%. 31P NMR
(CDCl3): d 200.8 ppm. This was used in the subsequent step
without further purification.
4. Experimental
4.1. General
All reactions and manipulations of air-sensitive materials
were performed under nitrogen, either in a glove-box or
using standard Schlenk techniques. Solvents were distilled
from appropriate drying reagents prior to use. Ligands
14–18 were prepared by modification of a procedure
reported for a closely related phosphite.17 Ligand 21 and
complexes 22 and 23 were prepared according to literature
methods.18,19 All other materials, except where noted, were
obtained commercially and used as received. GC analysis
was performed on a Varian 3800 GC fitted with a 25 m CP
Sil 5CB column and data were recorded on a Star
workstation. Enantioselectivity was determined by HPLC
on a Varian Prostar 210 fitted with a Chiralcel OD or OB
column.
A mixture of Et3N (8.0 ml, 57.1 mmol) and 12a (3.29 g,
6.76 mmol) in toluene (60 ml) was cooled to K40 8C and
treated drop-wise with a solution of (S)-binaphthol (3.90 g,
13.62 mmol) in toluene (20 ml). The reaction was allowed
to warm to rt overnight. The mixture was filtered through a
pad of Celite, which was then washed with toluene (2!
20 ml), the solvent removed from the combined solutions in
vacuo and the resultant solid was recrystallised (dichloro-
methane/hexane) to give ligand 9 as a white solid (5.12 g,
83%). Spectroscopic data as above.
4.2. Typical method for catalysis
Ligand 9 (0.041 g, 0.044 mmol) was added to a solution of
palladium acetate in 1,4-dioxane (0.022 M, 2.00 ml), the
solution was diluted with dioxane (2 ml) and stirred at rt for
5 min. The appropriate stryrene (4.36 mmol) was added
followed by indium triflate (0.061 g, 0.1085 mmol) in one
portion. More dioxane (1 ml) was used to wash down any
indium triflate stuck to the side of the reaction flask and the
reaction mixture was then stirred at rt for 18 h. The reaction
was quenched with water (25 ml), the resultant mixture
extracted with dichloromethane (3!25 ml), the organic
phase dried (MgSO4) and the solvent removed under
reduced pressure. The crude product was dissolved in
toluene (10 ml) and hexadecane (internal standard, 0.068 M
in toluene, 5.00 ml) was added. The conversion and yield
were determined by GC analysis.
4.1.1. Synthesis of ligand 8. A mixture of bisphenol, 11a,
(1.53 g, 5.70 mmol) and the chlorophosphite 10 (2.00 g,
5.70 mmol) in toluene (60 ml) was cooled to K40 8C and
then treated drop-wise with a solution of Et3N (3.0 ml,
21.4 mmol) in toluene (20 ml). The reaction was stirred and
allowed to warm to rt overnight. The mixture was filtered
through a pad of Celite, which was then washed with
toluene (2!20 ml) and the solvent removed from the
combined solutions in vacuo and the resultant solid was
recrystallised from dichloromethane/hexane to give com-
pound 8 as a white solid. Yield 3.48 g, 70%, C58H42O6P2
1
requires C, 77.7; H, 4.7%; Found: C, 74.60; H, 5.47. H
NMR (300 MHz, CDCl3): d 1.36–1.51 (8H, br s, CyH),
2.07–2.19 (2H, br s, CyH), 6.97–7.23 (16H, m, ArH), 7.25–
7.51 (8H, m, ArH), 7.75–7.96 (8H, m, ArH); 13C NMR
(75 MHz, CDCl3): d 20.43, 25.25, 36.24 (CH2), 44.05 (C-
(CH2)2), 114.00, 118.78, 118.88, 121.80 (ArH), 123.37
(ArC), 123.97, 124.19, 125.34, 125.68, 125.99, 126.52,
126.84, 126.89, 128.06, 128.83 (ArH), 130.19, 130.60,
131.49, 131.76, 143.45, 146.47, 146.52, 148.21, 148.32
(ArC); 31P NMR (121 MHz, CDCl3): d 145.63.
4.2.1. Compound 13a. Isolated by column chromatography
(silica, chloroform) as a yellow oil, 0.431 g, 48%. 1H NMR
(300 MHz, CDCl3) d 1.30 (3H, d, 3JHHZ7 Hz, CH3), 3.42–
3
3
3.50 (1H, dq, JHHZ5, 7 Hz, R3CH), 6.24 (1H, d, JHH
Z
5 Hz, ]CH), 6.25 (1H, s, ]CH), 6.98–7.20 (10H, m, ArH).
The ee was determined by HPLC, Chiralcel OD, heptane/
isopropanol 99.8:0.2, flow rate 0.9 ml/min, l 215 cmK1, Rf
S-isomerZ11.24 min, R-isomerZ11.87 min, absolute con-
figuration determined by order of Rf in comparison with
literature data.20
4.1.2. Synthesis of ligand 9. Method A. This was prepared
by an analogous method to that outlined above. Yield 87%.
C59H46O6P2 requires C, 77.6; H, 5.1%; Found: C, 75.5; H,
5.4. 1H NMR (300 MHz, CDCl3): d 1.44 (6H, s, CH3), 2.28
(12H, s, CH3), 6.83 (4H, s, ArH), 7.14–7.24 (4H, m, ArH),
1
7.28–7.40 (8H, m, ArH), 7.41 (4H, d, JHHZ8 Hz, ArH),
4.2.2. Compound 13b. Isolated by column chromatography
(silica, hexane then 5% dichloromethane in hexane), as a
yellow oil, 0.041 g, 77%. HRMS m/zZ244.1074; C16H14F2
requires 244.1064. 1H NMR (300 MHz, CDCl3) d 1.35 (3H,
d, 3JHHZ7 Hz, CH3), 3.52 (1H, m, R3CH), 6.10–6.20 (1H,
7.81–7.94 (8H, m, ArH); 13C NMR (75 MHz, CDCl3): d
18.81, 31.58 (CH3), 42.44 (C(CH3)2), 122.31, 122.49 (ArH),
123.40 (ArC), 125.70, 125.85, 126.72, 126.89, 127.52,
127.59, 127.84, 128.78, 128.87 (ArH), 130.32 (ArC),

R. B. Bedford et al. / Tetrahedron 61 (2005) 9799–9807
9807
3
3
5. Feringa, B. L. Acc. Chem. Res. 2000, 33, 346.
`
6. Francio, G.; Faraone, F.; Leitner, W. J. Am. Chem. Soc. 2002,
dd, JHHZ7, 16 Hz, ]CH), 6.26 (1H, d, JHH
Z
3
16 Hz,]CH), 6.85–6.89 (2H, d, JHHZ9 Hz, ArH), 6.91–
6.94 (2H, d, 3JHHZ9 Hz, ArH), 7.09–7.15 (2H, q, 3JHFZ5,
9 Hz, ArH). The ee was determined by HPLC, Chiralcel
OD, heptane/isopropanol 99.8:0.2, flow rate 0.9 ml/min, l
215 cmK1, Rf major isomerZ9.39 min, minor isomerZ
10.26 min, absolute configuration not determined.
124, 736.
7. (a) Albert, J.; Bosque, R.; Cadena, J. M.; Delgado, S.; Granell,
J.; Muller, G.; Ordinas, J. I.; Bardia, M. F.; Solans, X. Chem.
Eur.J. 2002, 8, 2279. (b) Englert, U.; Haerter, R.; Vasen, D.;
Salzer, A.; Eggeling, E. B.; Vogt, D. Organometallics 1999,
18, 4390. (c) Albert, J.; Cadena, M.; Granell, J.; Muller, G.;
Ordinas, J. I.; Panyella, D.; Puerta, C.; Sanudo, C.; Valerga, P.
4.2.3. Compound 13c. Isolated by column chromatography
(silica, chloroform), as a yellow oil, 0.01 g, 19%. HRMS m/
zZ236.1558; C18H20 requires 236.1565. 1H NMR
(300 MHz, CDCl3) d 1.36–1.39 (3H, d, JHHZ7 Hz, CH3),
2.23 (3H, s, ArCH3), 2.33 (3H, s, ArCH3), 3.75–3.84 (1H,
dq, JHHZ7, 7 Hz, R3CH), 6.10–6.17 (1H, dd, JHHZ7,
16 Hz, ]CH), 6.46–6.52 (1H, d, JHHZ16 Hz, ]CH),
¨
Organometallics 1999, 18, 3511. (d) Bayersdorfer, R.; Ganter,
B.; Englert, U.; Keim, W.; Vogt, D. J. Organomet. Chem.
1998, 552, 187.
3
´
8. (a) Bogdanovic, B. Adv. Organomet. Chem. 1979, 17, 104.
3
3
(b) Heumann, A.; Moukhliss, M. Synlett 1998, 1211. (c)
Heumann, A.; Moukhliss, M. Synlett 1999, 268.
3
3
7.03–7.11 (6H, m, ArH), 7.14–7.20 (1H, d, JHHZ7 Hz,
`
9. Boing, C.; Francio, G.; Leitner, W. Chem. Commun. 2005,
¨
ArH). The ee was determined by HPLC, Chiralcel OB,
heptane/isopropanol 98.0:2.0, flow rate 0.5 ml/min, l
254 cmK1, Rf major isomerZ11.24 min, minor isomerZ
14.36 min, absolute configuration not determined.
1456.
10. Tsuchimoto, T.; Kamiyama, S.; Negoro, R.; Shirakawa, E.;
Kawakami, Y. Chem. Commun. 2003, 852.
11. Arena, C. G.; Drommi, D.; Faraone, F.; Graiff, C.; Tiripicchio,
A. Eur. J. Inorg. Chem. 2001, 247.
12. MacDonald, C. L. B.; Corrente, A. M.; Andrews, C. G.;
Taylor, A.; Ellis, B. D. Chem. Commun. 2004, 250.
13. This assignment is, at the moment, based solely on the low-
field value of the chemical shift compared with simple Pd(II)
adducts, consistent with literature data for related five-
membered palladacycles formed on orthometallation of
triarylphosphites. See, for example: Bedford, R. B.; Hazel-
wood, S. L.; Limmert, M. E.; Albisson, D. A.; Draper, S. M.;
Scully, P. N.; Coles, S. J.; Hursthouse, M. B. Chem. Eur. J.
2003, 3126.
Acknowledgements
We thank the Universidad Rey Juan Carlos for the provision
of a postdoctoral grant (A. G.) and funding through Project
GVC-2004-12 (S. P., A. G.), the EPSRC (Advanced
Research Fellowship for RBB, PDRAs for MB and MEB,
DTA for SLM) and Johnson Matthey for the loan of
palladium acetate.
14. Teo, Y.-C.; Tan, K.-T.; Loh, T.-P. Chem. Commun. 2005,
1318.
References and notes
15. Bedford, R. B.; Betham, M.; Blake, M.E. Manuscript in
preparation.
1. For a recent review see: RajanBabu, T. V. Chem. Rev. 2003,
103, 2845.
16. Ligand 21 was prepared by the reaction of Ph2PCl with
(2S,20S)-1,10-(ethane-1,2-diyl)bis(pyrrolidine-2,1-diyl)dimethanol
in toluene with triethylamine as base. The chiral diol was
preparedaccordingtoaliteraturemethod:Xu, Q.;Wang,H.;Pan,
X.; Chan, A. S. C.; Yang, T.-K. Tetrahedron Lett. 2001, 6171.
17. Baker, M. J.; Pringle, G. J. Chem. Soc., Chem. Commun. 1991,
1292.
2. Recent references: (a) Shi, W.-J.; Xie, J.-H.; Zhou, Q.-L.
Tetrahedron: Asymmetry 2005, 16, 705. (b) Zhang, A.;
RajanBabu, T. V. Org. Lett. 2004, 6, 3159. (c) Zhang, A.;
RajanBabu, T. V. Org. Lett. 2004, 6, 1515. (d) Kumareswaran,
R.; Nandi, M.; RajanBabu, T. V. Org. Lett. 2003, 5, 4345.
(e) RajanBabu, T. V.; Nomura, N.; Jin, J.; Nandi, M.; Park, H.;
Sun, X. J. Org. Chem. 2003, 68, 8431.
18. Sablon, R.; Newton, C.; Dierkes, P.; Osborn, J. A. Tetrahedron
Lett. 1996, 37, 4933.
3. Wilke, G. Angew. Chem., Int. Ed. 1988, 27, 185.
4. (a) Wegener, A.; Leitner, W. Chem. Commun. 1999, 1583.
19. Bedford, R. B.; Betham, M; Blake, M. E.; Coles, S. J.;
Hursthouse, M. B.; Pilarski, L. T.; Pringle, P. G.; Wingad, R.
Manuscript in preparation.
`
¨
(b) Bosmann, A.; Francio, G.; Janssen, E.; Solinas, M.;
Leitner, W.; Wasserscheid, P. Angew. Chem., Int. Ed. 2001,
40, 2697.
20. Schwink, L.; Knochel, P. Chem. Eur. J. 1998, 4, 950.