Silver phosphanes partnered with carborane monoanions: Synthesis, structures and use as highly active Lewis acid catalysts in a hetero-Diels-Alder reaction

Article abstract of DOI:10.1002/1521-3765(20020503)8:9<2088::AID-CHEM2088>3.0.CO;2-L

Four Lewis acidic silver phosphane complexes partnered with [1-closo-CB₁₁H₁₂]^- and [1-closo- CB₁₁H₆Br₆]^- have been synthesised and studied by solution NMR and solid-state X-ray d

Full text of DOI:10.1002/1521-3765(20020503)8:9<2088::AID-CHEM2088>3.0.CO;2-L

FULL PAPER
Silver Phosphanes Partnered with Carborane Monoanions:
Synthesis, Structures and Use as Highly Active Lewis Acid Catalysts
in a Hetero-Diels Alder Reaction
Nathan J. Patmore, Catherine Hague, Jamie H. Cotgreave, Mary F. Mahon,
Christopher G. Frost,* and Andrew S. Weller*[a]
Abstract: Four Lewis acidic silver
are observed that are suggested to be
monomeric [Ag(PPh3)2][CB11H12] and
dimeric [Ag(PPh3)2(CB11H12)]2. With
the more weakly coordinating anion
[CB11H6Br6]À and one phosphane, com-
plex 3 [Ag(PPh3)(CB11H6Br6)] is iso-
required (0.1 mol%) coupled with turn-
over frequencies of 4000hÀ1 (quantita-
tive conversion to product after 15 min-
utes using 3 at room temperature).
Moreover, the reaction does not occur
in rigorously dry solvent as addition of a
substoichiometric amount of water
(50 mol%) is necessary for turnover of
the catalyst. It is suggested that a Lewis
assisted Br˘nsted acid is formed be-
tween the water and the silver. The
effect of changing the counterion to
[BF4]À, [OTf]À and [ClO4]À has also
been studied. Significant decreases in
reaction rate and final product yield are
observed on changing the anion from
[CB11H6Br6]À, thus demonstrating the
utility of weakly coordinating carborane
anions in organic synthesis.
phosphane complexes partnered with
[1-closo-CB11H12]À
and
[1-closo-
CB11H6Br6]À have been synthesised and
studied by solution NMR and solid-state
X-ray diffraction techniques. In the
complex [Ag(PPh3)(CB11H12)] (1), the
silver is coordinated with the carborane
lated.
Complex 4,
[Ag(PPh3)2-
(CB11H6Br6)], has been characterised
spectroscopically. All of the complexes
have been assessed as Lewis acids in the
hetero-Diels Alder reaction of N-ben-
zylideneaniline with Danishefsky×s di-
ene. Exceptionally low catalyst loadings
for this Lewis acid catalysed reaction are
À À
by two stronger 3c 2e B H Ag bonds,
À À
one weaker B H Ag interaction and a
very weak Ag ¥¥¥ Carene contact in the
solid state. In solution, the carborane
remains closely connected with the
‡
{Ag(PPh3)} fragment, as evidenced
by 11B chemical shifts. Complex 2
[Ag(PPh3)2(CB11H12)]2 adopts a dimeric
motif in the solid state, each carborane
bridging two Ag centres. In solution at
low temperature, two distinct complexes
Keywords: carboranes
compounds cycloaddition
homogeneous catalysis ¥ silver
¥
cluster
¥
¥
reports of anions such as A or related compounds utilised in
despite their potential advantages
over perfluorinated borates.[11, 12]
Introduction
10]
catalytic processes,[7
Monoanionic carborane anions based around [closo-1-
CB11H12]À (A, see Scheme 1) are among the most inert and
least coordinating anions currently known.[1, 2] The high
chemical stability and low nucleophilicity of these ions means
that they have been used to stabilise exotic cationic species,
which are not isolatable with other, more coordinating,
The coordination chemistry of cationic silver(i) phosphane
complexes (combined with a range of anions) has been
studied in some detail,[13 16] to the point that useful structural
predications may be made on the basis of spectroscopic
data.[17] Silver phosphanes exo-coordinated to polyhedral
anions.[3 Such anions also have the potential to act as
boranes are rarer,[18] but recent work, particularly from Stone
6]
21]
partners with cationic Lewis acidic transition metal complexes
that take part in various catalytic processes. Somewhat
surprisingly, given the almost ubiquitous use of [B(C6F5)4]À
and derivatives in such applications, there are only a few
et al.[19
has established this area. Related complexes in
which exo-coordinated silver is bound to ligands other than
phosphane,[22 25] or arene solvent molecules[26 29] or where the
silver is partnered with halogeno-substituted carbo-
ranes[27, 30 33] are also known.
One of us has a current interest in the synthesis, structures
and reactivity of Lewis acidic transition metal complexes
partnered with anion A and derivatives.[24, 25] Concurrently,
there is also interest in Lewis acid catalysis using silver(i) salts
as effective promoters of a wide-range of organic trans-
formations including allylation, aldol, ene and glycosylation
[a] Dr. C. G. Frost, Dr. A. S. Weller, N. J. Patmore, C. Hague
J. H. Cotgreave, Dr. M. F. Mahon
Department of Chemistry, University of Bath
Bath, BA27AY (UK)
Fax : (‡44)1225-826231
E-mail: c.g.frost@bath.ac.uk, a.s.weller@bath.ac.uk
2088
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2088 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9

2088 2098
We report here the synthesis and solution and solid-state
structural investigations for complexes of the formula
[Ag(PPh3)n(CB11H6Y6)] (n ˆ 1, 2; Y ˆ H, Br) and their use
as active catalysts in the hetero-Diels Alder reactions of N-
benzylideneaniline with Danishefsky×s diene. Aspects of this
work have been communicated previously.[42]
Ag
Ag
C
C
Br
Br
Br
Br
Br
Br
B
A
Results and Discussion
C
Synthesis and structures: Silver(i) salts of monoanionic
carborane anions are readily prepared and are air stable and
useful synthons for subsequent reactions. For example,
they can be used in silver salt metathesis reactions to
introduce a carborane anion into a metal×s coordination
sphere[23, 25, 29, 43, 44] or to generate synthetically useful salts.[28]
In addition they often have significant solubility in aromatic
solvents compared with other cations, facilitating more
complete characterisation of the cation/anion pair. In princi-
ple, they also provide a useful starting point for the
investigation of silver(i) complexes of the general formula
[Ag(L)n(CB11H6Y6)] (L ˆ two electron donor; Yˆ H, halo-
gen) as simple addition of the required ligand (L) results in
complex formation.
C
PPh3
PPh3
Ph3P
Ph3P
Ag
Ag
Ag
PPh3
1
C
= BH
= BBr
Br
2
C
C
Addition of slightly less than one equivalent (to avoid
formation of 2, vide infra) of PPh3 to a suspension of
Ag[CB11H12] in CH2Cl2 results in the formation of
[Ag(PPh3)(CB11H12)], 1, as an analytically pure solid in good
isolated yield after recrystallisation (Tables 1 and 2). The
solid-state structure of 1 is presented in Figure 1. The
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Ag
Br
Br
Ag
Ph3P
PPh3
PPh3
‡
{AgPPh3} fragment interacts with the carborane anion
4
3
À
À
through the three {BH} units that form the B(12) B(7) B(8)
polyhedral face, a coordination motif seen previously in
[(Cp*)ZrMe2-h3-(CB11H12)].[10] The silver phosphane frag-
ment does not sit over the centre of this face but is more
Scheme 1. Structures A and B and 1 4.
reactions.[34] Of particular note is the use of silver(i) BIN AP
([1,1'-binaphthalene]-2,2'-diylbis(diphenylphosphane)) com-
plexes in asymmetric aldol reactions,[35
localised towards B(8) and the corresponding Ag B bond
À
À
À
lengths reflect this: Ag(1) B(8) 2.504(1) ä, Ag(1) B(12)
38]
À
Mukaiyama aldol
2.569(3) ä, Ag(1) B(7) 2.619(2) ä. These bond lengths are
broadly similar to those previously reported for the related
reactions,[39] asymmetric allylations[40] and hetero-Diels
Alder reactions.[41] It is well established that these reactions
are accelerated by other Lewis acids (for example: Ti, B, Al,
Sn complexes). However, many of these established catalysts
are sensitive to air, water and product inhibition and
consequently are used at a low substrate/catalyst ratio. The
use of silver(i) phosphane complexes can provide a practical
solution, with many precatalysts being stable in air and
retaining activity in the presence of reaction product. Never-
theless, the best examples from the literature routinely
employ high catalytic loadings (5 10 mol%) to achieve
competitive rates and product yields. We are thus interested
in developing silver(i) Lewis acids complexed with carborane
anions in anticipation that the weakly coordinating nature of
the anions will reveal enhanced activity for these systems. The
hetero-Diels Alder reaction is one of the most useful
methods for the synthesis of bioactive heterocycles. Hence
we chose to test our new complexes in this challenging
reaction with the intention of identifying structure activity
relationships when the coordination sphere of the silver
(number of phosphanes, anion) is systematically changed.
complex
[2,2,2-(CO)3-2-PPh3-7,12-{Ag(PPh3)}-closo-2,1-
MoCB10H9][20, 21] (2.552(4) ä and 2.589(4) ä), in which the
Ag centre is bound dihapto by the cage and is the only other
closo-polyhedral borane structure reported with an exo-
{AgPR3} fragment, although there are others reported with
ligands other than phosphane,[22] with Ag metal bonds[21] and
À
with endo coordination of a silver fragment.[18] The location of
À
the silver phosphane fragment in relation to the B(8) B(12)
bond is such that the silver atom adopts a planar coordination
motif with respect to P(1), B(8) and B(12) with the sum of
relevant angles around Ag(1) 359.78. P(1) essentially lies on
the same plane formed by Ag(1), B(8) and B(12) (max
deviation from plane 0.083 ä for P(1)), resulting in an
apparent vacant coordination site transoid to H(7). Inspection
of the extended lattice packing diagram shows that there are
À
no Ag ¥¥¥ H B intermolecular interactions directed towards
this vacant site.[18, 20] However, there is a long intermolecular
interaction (3.348 ä) between Ag(1) and C(35)' of a phenyl
ring in a symmetry related molecule (Figure 2), and the arene
carbon approaches the silver from the reverse of the trigonal
Chem. Eur. J. 2002, 8, No. 9
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2089 $ 20.00+.50/0
2089

A. S. Weller, C. G. Frost et al.
FULL PAPER
Table 1. Crystal data and structure refinement for 1, 2 and 3.
1
2
3
empirical formula
Mr
C19H27AgB11P
513.16
C74H84Ag2B22P4
1550.85
C19H21AgB11Br6P
986.57
T [K]
l [ä]
150(2)
0.71073
170(2)
0.71069
150(2)
0.71073
crystal system
space group
a [ä]
b [ä]
c [ä]
a [8]
b [8]
g [8]
V [ä3]
Z
triclinic
P1
10.282(2)
10.988(2)
11.308(2)
76.22(3)
75.25(3)
75.61(3)
1175.7(4)
2
1.450
0.932
triclinic
P1
monoclinic
P21/c
8.89500(10)
24.4140(4)
14.4170(3)
90
102.0720(10)
90
3061.60(9)
4
≈
≈
11.6963(2)
13.1668(2)
14.1393(2)
96.8860(6)
106.8980(7)
110.5370(8)
1890.30(5)
1
1calcd [mgmÀ3
]
1.362
0.646
792
2.140
8.554
1848
m [mmÀ1
F(000)
]
516
crystal size [mm]
q range [8]
reflections collected
0.20 Â 0.20 Â 0.10
2.96 to 27.50
20702
0.25 Â 0.20 Â 0.17
3.38 to 27.50
22570
0.10 Â 0.10 Â 0.05
3.71 to 27.88
23164
independent reflections
reflections observed (>2s)
max and min transmission
data/restraints/parameters
goodness of fit on F 2
final R indices [I > 2s(I)]
R indices (all data)
5374 [R(int) ˆ 0.0387]
8605 [R(int) ˆ 0.0250]
7253 [R(int) ˆ 0.0519]
4538
7964
5880
0.9126 and 0.8355
5374/0/302
1.018
R1 ˆ 0.0316 wR2 ˆ 0.0757
R1 ˆ 0.0417 wR2 ˆ 0.0813
0.544 and À0.923
1.027, 0.981
8605/2/478
1.050
R1 ˆ 0.0289 wR2 ˆ 0.0738
R1 ˆ 0.0323 wR2 ˆ 0.0763
1.727 and À0.584
0.66 and 0.47
7253/0/349
1.041
R1 ˆ 0.0358 wR2 ˆ 0.0783
R1 ˆ 0.0521 wR2 ˆ 0.0855
1.081 and À1.181
largest diff. peak and hole [eäÀ3
]
Table 2. Selected bond lengths [ä] and angles [8] for the new
compounds 1 3.
Compound 1
À
À
À
Ag(1) P(1)
2.3625(7)
2.619(3)
2.504(3)
2.569(3)
2.36(2)
P(1) Ag(1) B(7)
142.11(6)
158.23(6)
160.32(6)
41.26(8)
40.37(8)
40.98(8)
À
À
À
Ag(1) B(7)
P(1) Ag(1) B(8)
À
À
À
Ag(1) B(8)
P(1) Ag(1) B(12)
À
À
À
Ag(1) B(12)
B(12) Ag(1) B(8)
À
À
À
Ag(1) H(7)
B(12) Ag(1) B(7)
À
À
À
Ag(1) H(8)
2.14(2)
2.26(2)
B(7) Ag(1) B(8)
À
Ag(1) H(12)
Compound 2
À
À
Ag(1) P(1)
2.4698(3)
2.4741(3)
3.494(2)
2.892(2)
Ag(1) H(12)'
2.17(2)
2.51(2)
130.90(1)
72.4(6)
À
À
Ag(1) P(2)
Ag(1) H(7)
À
À
À
Ag(1) B(7)
P(1) Ag(1) P(2)
À
À
À
Ag(1) B(12)'
B(7) Ag(1) B(12)'
Compound 3
À
À
À
Ag(1) P(1)
2.4032(10)
2.8710(5)
2.6963(5)
3.0953(5)
3.4901(5)
P(1) Ag(1) Br(7)
117.44(3)
155.55(3)
93.31(3)
157.15(1)
87.471(14)
À
À
À
Ag(1) Br(7)
P(1) Ag(1) Br(8)
À
À
À
Ag(1) Br(8)
P(1) Ag(1) Br(12)
À
À
À
Figure 1. Molecular structure of complex 1, showing the atom-numbering
scheme. Thermal ellipsoids are drawn at the 50% probability level.
Ag(1) Br(12)
P(1) Ag(1) Br(11)'
À
À
À
Ag(1) Br(11)'
Br(7) Ag(1) Br(8)
counterion or two-electron ligand (L)) molecules[14, 16] sug-
gesting a relatively strong AgP bond.
plane formed by B(8), B(12), Ag(1) and P(1). Although this
À
distance is very long compared with other reported silver
1
In solution, the H{11B} NMR spectrum of 1 shows a 1:1
arene bond lengths found in the solid-state structure of Ag
carboranes (ca. 2.5 2.6 ä), we suggest that it may be
significant, given the observed orientation of the silver
phosphane fragment with respect to the cage. Indeed, gas-
phase DFT calculations on 1 show that the silver adopts the
expected tetrahedral geometry when there are no intermo-
ratio of phosphane to carborane anion. Resonances due to
{BH} groups are observed at d ˆ 2.25 and d ˆ 1.85 in the ratio
1:10 (the latter is a 5‡5 coincidence), with no high-field
À
resonances potentially indicative of M À H B interactions
observed.[48] That only two {BH} resonances are observed
suggests that the {AgPPh3} fragment is fluxional over the
surface of the cage and affords time-averaged C5V symmetry
in solution. This facile process is evidenced by no significant
change observed in the spectrum when recorded at À908C.
lecular interactions,[45] while comparable weak silver arene
contacts have been previously identified.[46, 47] The
À
À
Ag(1) P(1) distance, at 2.3625(7) ä, is at the shorter end of
the range reported for [Ag(PPh3)L] or [Ag(PPh3)]X (X ˆ
2090
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2090 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9

Silver Phosphanes
2088 2098
other systems.[25, 44] Specifically, in the 11B{1H} NMR spectrum,
resonances are observed at d ˆ À10.3, d ˆ À11.2 and d ˆ
1
À12.0 in the ratio 1:5:5. This dichotomy between H and 11B
chemical shifts may be indicative of the cage-silver bonding
[49]
À
predominantly originating from Ag B interactions, with a
reduced contribution (or perhaps a more electrostatic inter-
action) from the cage hydrogens. It is noteworthy that AgH
interactions can be observed by 1H NMR spectroscopy in
certain cases and show both significant chemical shifts and
AgH coupling constants,[50] neither of which are observed for
complex 1, or previously in related Ag cage compounds.[19
21]
Given that only the antipodal [B(12)] and lower pentagonal
À
belt [B(7) B(11)] boron atoms appear significantly perturbed
in the 11B NMR spectrum on coordination of the {AgPPh3}
fragment, we suggest a mechanism to explain the observed
fluxionality that incorporates the metal phosphane precessing
À
around the five triangular faces formed between B(7) B(11)
and B(12). This mechanism is similar to that previously
‡[44]
observed for the movement of {Rh(cod)}
{Pt(tBu2P(CH2)3PtBu2)}
or
‡[51]
over the surface of [closo-
CB11H12]À. In the 31P{1H} NMR spectrum of 1, a peak centred
at d ˆ 18.70 is observed as a pair of concentric doublets, due to
109AgP and 107AgP coupling, having an average value for
J(AgP)average of 743 Hz. This large value is consistent with the
strong AgP bond seen in the solid state and is also indicative
of a low-coordinate silver phosphane species;[13] this suggests
the observed solution and solid-state structures of 1 are
similar.
Figure 2. Dimeric unit formed in the extended lattice by Ag ¥¥¥ Carene
À
contacts (Ag C(35)' 3.348 ä). Atom labelling and other details as
in Figure 1.
Related metallaboranes with exo-coordinated {AgPPh3} frag-
ments are also highly fluxional in solution.[20, 21] We were
initially surprised to observe that the 11B{1H} NMR spectrum
shows that both the antipodal [B(12)] and lower pentagonal
Addition of two equivalents of PPh3 to Ag[CB11H12]
gives
complex 2,
having
the
empirical
formula
[Ag(PPh3)2(CB11H12)], which has been characterised by
solution NMR spectroscopy and in the solid state by X-ray
diffraction. The extra PPh3 ligand has the effect of disturbing
the silver-cage bonding from 1, so that a dimeric, centro-
symmteric, [Ag(PPh3)2(CB11H12)]2 unit is now observed (Fig-
ure 4), in which each carborane unit bridges two silver centres.
Each cage interacts with the two silver centres through one
À
belt [B(7) (11)] borons undergo a significant upfield shift on
complexation with the {AgPPh3} fragment when compared
with Ag[CB11H12] (Figure 3), while there is no significant shift
in the {BH} resonances observed in the 1H{11B} NMR spectrum.
This was unexpected as coordination of an exo-metal frag-
ment results in significant upfield shifts of the exo-metal
coordinated {BH} units in both the 1H and 11B NMR spectra in
À
À
shorter (Ag(1) H(12)' 2.17(2) ä, Ag(1) B(12)' 2.892(2) ä)
À
and one longer (Ag(1) H(7) 2.51(2) ä,
À
À
Ag(1) B(7) 3.494(2) ä) Ag ¥¥¥ H B in-
teraction. This bonding mode to the two
silver centres by the antipodal [B(12)]
and a single lower pentagonal belt
[B(7)] vertex is similar to that observed
in the extended solid-state structure of
[Mo(Cp)(CO)3I ¥ Ag(CB11H12)]2.[24] Di-
meric and tetrameric silver carborane
complexes with cages other than
[CB11H12]À have also been previously
reported.[18, 20 22] As expected on moving
to a higher coordination number in
À
going from 1 to 2, the Ag P bond
lengths are relatively longer in the latter
À
complex (Ag(1) P(1) 2.4698(3) ä,
À
À
Ag(1) P(2) 2.4741(3) ä). The P(1)
À
Ag(1) P(2) angle of 130.90(1)8 is sim-
ilar to that observed in the three-
Figure 3. 11B NMR chemical shift ™stick∫ diagram comparing complexes 1 and 2 (solutions in CD2Cl2)
with Ag[CB11H12] (solution in [D6]acetone).
coordinate
complex
[Ag(PPh3)2-
(NCMe)][BF4] (129.37(2)8).[14]
Chem. Eur. J. 2002, 8, No. 9
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2091 $ 20.00+.50/0
2091

A. S. Weller, C. G. Frost et al.
FULL PAPER
which the six lower hydrogen atoms in the cage are replaced
with bromines, and it is considered to be significantly more
weakly coordinating than [CB11H12]À. While there are now a
significant number of solid-state structural investigations on
the silver salts of perhalogenated anions,[27, 30 33] there are few
studies in which both the solution and solid-state structures of
their transition metal complexes have been investigated.
Addition of one equivalent of PPh3 to a suspension of
Ag[CB11H6Br6] in CH2Cl2 affords the new complex
[Ag(PPh3)(CB11H6Br6)] (3). The solid-state structure of 3 is
presented in Figure 5. On first inspection, the silver(i) centre is
coordinated with the carborane anion through two shorter
À
(2.6963(5) ä and 2.8710(5) ä) Ag Br interactions and one
longer one (3.0953(5) ä), all of which fall comfortably within
the sum of the van der Waals radii for silver and bromine.
These bond lengths are of similar values to those previously
reported for silver complexes of [CB11H6Br6]À[31] and dibro-
moalkane complexes of silver.[52] The silver phosphane bond
length (2.4032(10) ä) is slightly longer than that found in 1.
The phosphane phenyl groups adopt a relatively uncommon
Figure 4. Molecular structure of complex 2, showing the atom-numbering
scheme. Phenyl hydrogen atoms are omitted for clarity. Thermal ellipsoids
are drawn at the 50% probability level.
À
C2v arrangement, a result of one phenyl group (C(2) C(7))
In solution at room temperature in 2, the carborane does
not interact significantly with the metal centre. This is
evidenced in the 11B NMR spectrum, which does not show
any appreciable perturbation from that found for
Ag[CB11H12], with three resonances observed at d ˆ À5.1,
À10.6 and À12.6 in the ratio 1:5:5 (Figure 3). These data
twisting to minimise interactions with Br(12). Given that the
silver is formally 4-connected in the asymmetric unit, the fact
that it appears to adopt a pseudotrigonal planar, as opposed to
a tetrahedral coordination motif (sum of relevant angles
around Ag(1) is 360.58) was on first inspection puzzling to us.
However, similar to 1, the packing diagram (Figure 6) reveals
a long interaction (3.4901(5) ä) between Ag(1) and Br(11)'
on a symmetry related cage in the lattice; this bromine
atom approaches Ag(1) approximately trans to Br(12)
À
reflect the significantly weaker Ag ¥¥¥ H B interactions appar-
ent in 2, relative to 1, in the solid state. In the room-
temperature 31P{1H} NMR spectrum of 2, a single broad
resonance centred at d ˆ 14.8 is observed, suggesting that the
phosphane ligands are undergoing an exchange process.
Progressive cooling affords two broad singlets at d ˆ 15.4
and d ˆ 13.2 at 08C, which gradually sharpen into two sets of
doublets at À608C, of approximately equal proportions,
centred at d ˆ 13.7 (J(AgP) ˆ 333 Hz) and d ˆ 13.6. The latter
resonance is further observed as a pair of concentric doublets,
À
À
(Br(12) Ag(1) Br(11)' 157.15(1)8). Although this distance is
long, the coordination motif around the silver suggests that it
is significant in the solid state. This interaction completes the
coordination sphere at the silver centre. It also results in
phenyl groups on symmetry related phosphanes having a close
À
approach in the lattice (Figure 6), and the Ag(1) P(1)
showing coupling to both 109Ag and 107Ag (J(AgP)average
ˆ
vector being canted from lying equidistant between Br(7)
519 Hz). These changes in the NMR spectrum on cooling
are consonant with the presence of two different complexes in
solution at low temperature, which interconvert at ambient
temperature. The species that reveals the larger AgP coupling
constant has a value very similar to that reported for two-
coordinate [Ag(PPh3)2][BF4] (J(AgP) ˆ 530 Hz),[14] while the
alternative species has a coupling constant intermediate
between the three-coordinate [Ag(PPh3)2Br] (J(AgP) ˆ
385 Hz)[15] and four-coordinate [Ag(PPh3)4][PF6] (J(107AgP)
ˆ 224 Hz)[13] silver phosphane complexes. On the basis of
these observations, we tentatively suggest that, at low temper-
ature an approximate 1:1 mixture consisting of the two-
coordinate, ion-pair separated, species [Ag(PPh3)2][CB11H12]
and the dimeric [Ag(PPh3)2(CB11H12)]2 is formed.
No evidence for the disproportionation of 2 to form
[Ag(PPh3)][CB11H12] and [Ag(PPh3)3][CB11H12] was seen in
the low-temperature 31P{1H} NMR spectrum.[13] The low-
temperature 11B{1H} NMR spectrum exhibits two broad peaks
at d ˆ À5.8 (1B) and d ˆ À11.6(10B), essentially unshifted
from room-temperature values.
The coordination chemistry of the anion [closo-CB11H12]À
can be usefully compared with that of [closo-CB11H6Br6]À, in
Figure 5. Molecular structure of complex 3, showing the atom-numbering
scheme. Thermal ellipsoids are drawn at the 50% probability level.
2092
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2092 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9

Silver Phosphanes
2088 2098
Figure 6. Extended solid-state structure apparent in complex 3: a) showing weak axial Ag ¥¥¥ Br interactions; b) space-filling diagram showing orientation of
phenyl groups in the solid state.
À
À
À
À
and Br(8) (P(1) Ag(1) Br(8) 155.55(3)8, P(1) Ag(1) Br(7)
117.44(3)8).
NMR spectroscopy and microanalysis. In solution at room
temperature, the 31P{1H} NMR spectrum of 4 has a broad
singlet at d ˆ 12.9, showing that the phosphane ligands are
undergoing exchange under ambient conditions. The corre-
sponding 1H{11B} NMR spectrum shows a 2:1 ratio of
phosphane to cage, with no signals observed at high field.
The low-temperature (À808C) 31P{1H} NMR spectrum shows
that one species is present at this temperature, due to two
concentric doublets centred around d ˆ 8.1. The magnitude of
the AgP coupling constant (J(AgP)average ˆ 240 Hz) is very
small and of the same order of that associated with
In solution, C5v symmetry is observed for the cage anion in
the 11B NMR spectrum, which shows that the {AgPPh3}
‡
fragment does not retain a rigid coordination motif with
respect to the cage. The 31P{1H} NMR spectrum of 3 is similar
to that of 1 with a resonance centred at d ˆ 16.52 is observed
as a concentric pair of doublets, J(AgP)average ˆ 715 Hz,
suggesting that similar structures are adopted in solution for
both complexes. This value is slightly smaller than that
À
observed for 1, in line with the slightly longer Ag P bond
‡
observed in the solid state for 3 compared with complex 1.[17]
Addition of two equivalents of PPh3 to Ag[CB11H6Br6]
results in the isolation of a white solid having the empirical
formula [Ag(PPh3)2(CB11H6Br6)] (4). Despite repeated at-
tempts, crystals suitable for X-ray diffraction could not be
obtained, so characterisation of complex 4 was limited to
[Ag(PPh3)4] .[13] To account for this, a low-temperature
limiting structure that has a sp3 hybridised silver(i) centre is
suggested, in which the hexabromo carborane anion is
coordinated to the silver by two of the bromines on the cage.
That there is a more intimate contact between the halogen-
ated cage and silver in 4 compared with the silver-cage contact
Chem. Eur. J. 2002, 8, No. 9
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2093 $ 20.00+.50/0
2093

A. S. Weller, C. G. Frost et al.
FULL PAPER
in 2 is further reflected in the
relative catalytic activities of
these two compounds, which
will be presented later. While
we have taken measures to use
pure samples (recrystallised
material used exclusively for
NMR and catalysis experi-
OMe
+
Ph
Ph
+
O
O
Ag complex (1 mol%)
CH2Cl2
N
N
+
H
Ph
O
Ph
O
Ph
H
Ph
TMSO
<5%
70%
II
IV
I
III
(1.02 equiv)
(1 equiv)
(1 equiv)
Scheme 3. Activation of imines in the presence of aldehydes by silver complex 3.
ments), we cannot completely rule out the presence of small
amounts of free PPh3 in 2 and 4, which would lead to rapid
phosphane exchange at room temperature.[13]
Kobayashi et al.,[53] the silver catalyst is thus an A-2 Lewis
acid (active, imine selective).
Prompted by these results we decided to follow the
1
reactions by H NMR spectroscopy, using catalyst loadings
of 0.1 mol%, in the anticipation of uncovering structure
activity relationships involving counterion and/or phosphane
ligands. This initially resulted in the unexpected observation
that the reaction between diene II and imine I did not proceed
(no reaction after ca. two hours at room temperature) when in
the NMR tube, using rigorously dried solvent (CD2Cl2
vacuum distilled from CaH2). However, addition of a
substoichiometric amount of water (ca. 50 mol%, 1 mL,
unoptimised) to these solutions resulted in rapid initiation
of the reaction. This is an example of a water-accelerated
Lewis acid catalysed reaction; such reactions, in which water
is an important partner, are being increasingly appreciated.[54]
Pertinent to this observation, modest yield enhancements on
addition of stoichiometric amounts of water to a hetero-
Diels Alder reaction catalysed by lanthanide trifluorometh-
anesulfonylamides[55] have been reported previously. Current-
ly, we are unsure about the exact role of the water in the
reaction. It is likely, however, that a polarised, silver-bound,
water molecule acts as a Lewis assisted Br˘nsted acid, similar
to that in lanthanide salt catalysed aromatic electrophilic
substitutions.[56] It is also similar to the role metal-coordinated
water is suggested to play in certain catalytic processes.[57] In
support of this, preliminary DFT calculations on the model
Catalysis: We have investigated the reaction between
N-benzylideneaniline (I) and Danishefsky×s diene (II)
(Scheme 2) with the silver complexes 1 to 4. Importantly, in
the absence of any catalyst there was only a trace (less than
5%) of product after 24 hours at room temperature. Initial
experiments using freshly distilled reagent grade solvents
revealed that the addition of 1 mol% of silver complex
promoted the efficient formation of the product (III) in less
than one hour at room temperature (Table 3). On the bench
under these conditions, complexes 1 to 4 were significantly
more active than [Ag(PPh3)(OTf)] and [Ag(PPh3)(BF4)]
whilst [Ag(PPh3)(ClO4)] was of comparable activity. Remark-
ably, the catalyst loading could be lowered to 0.1 mol% of
complex 3 and the reaction was still complete in less than
thirty minutes to afford an isolated yield of 99% product
(>2000 turnovershourÀ1).
Table 3. Yields of compound III. Catalyst (1 mol%); imine I (1.1 mmol);
Danishefsky×s diene (II) (1.65 mmol). Yields reported are after reaction for
60 minutes at room temperature and workup (see Experimental Section for
full details).
Complex
Isolated yield of III [%]
1
2
3
4
5
6
7
[Ag(PPh3)(BF4)]
[Ag(PPh3)(OTf)]
[Ag(PPh3)(ClO4)]
[Ag(PPh3)(CB11H12)] (1)
[Ag(PPh3)2(CB11H12)] (2)
[Ag(PPh3)(CB11H6Br6)] (3)
[Ag(PPh3)2(CB11H6Br6)] (4)
35
70
90
98
99
99
85
‡
ˆ
system [Ag(Me3P)(OH2)(MeNH CHMe)] indicate that a
coordinated water molecule plays an important function in
the silver promoted [2‡4] addi-
PMe3
H
H
tion of imine to diene. A putative
intermediate in this process is
shown in Scheme 4, in which the
water molecule forms a hydrogen
bond with the nitrogen atom.[45]
This proposed intermediate ac-
counts for both the observed
O
Ag
Me
N
Me
A competition experiment illustrated that silver complex 3
Scheme 4. Proposed inter-
mediate structure.
could preferentially activate imines in the presence of
aldehydes (Scheme 3). When Danishefsky×s diene (II) was
added to a 1:1 mixture of benzaldehyde (IV) and imine (I) in
the presence of 1 mol% of 3, the product arising from the
reaction, after 15 minutes at room temperature, was isolated
in 70% yield whilst the aldehyde adduct was produced in only
5% yield. According to the classification system suggested by
dependence on trace amounts of water in this reaction and
the strong counterion effect observed (vide infra), with both
the Lewis assisted Br˘nsted acidity (OH ¥¥¥ Nhydrogen bond)
and the need for a vacant site (alkene coordination) playing
important roles. Experimentally, what is clear is that trace
OMe
OMe
Ph
Ph
Ph
Ph
Ph
1 - 4 (0.1 to 1 mol%)
N
N
N
+
H2O (ca. 50 mol%)
CH2Cl2
H
Ph
O
TMSO
TMSO
NOT ISOLATED
II
I
V
III
Scheme 2. Reaction between N-benzylideneaniline (I) and Danishefsky×s diene (II) with the silver complexes 1 4.
2094
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 0947-6539/02/0809-2094 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9

Silver Phosphanes
2088 2098
amounts of water are needed for this reaction to proceed,
which is probably made available to the reaction, when at the
bench, by adventitious water.
both reduce the Lewis acidity of the metal centre while also
adding more steric bulk and blocking a potential vacant site. It
is not unexpected then, that complexes 2 and 4 are signifi-
cantly poorer catalysts than any of the monophosphane
complexes. Interestingly the relative reaction rates for the
bisphosphane complexes 2 and 4 are reversed from those seen
for the monophosphane/carborane complexes. Complex 2,
which incorporates the [CB11H12] anion, catalyses the reaction
significantly faster than 4. Although this is perhaps counter-
intuitive given the relative coordinating abilities of these two
anions, when the solution structures at À608C are compared
(vide supra), complex 4 shows a considerably smaller value
for J(AgP), suggesting a silver centre that has a higher formal
coordination number in 4 than in 2. This implies that the
[CB11H6Br6]À anion interacts significantly with the metal
centre at this temperature in 4. Although we do not have the
solid-state structure of 4 to hand,[59] this correlation between
the magnitude of J(AgP) at low temperature and the relative
rates of 2 and 4 is compelling. This difference in rate is also
reflected in the relative product yields on the bench after one
hour (entries 5 and 7 in Table 3).
The catalyst [Ag(PPh3)(BF4)] ceases to function after only a
few minutes of activity, giving only modest yields of cyclo-
addition product in both the NMR and the bench-top
experiments (Figure 7 and Table 3, entry 1). We suggest that
under the conditions used for catalysis (0.1 mol% catalyst,
50 mol% water), rapid (<2 minutes) silver(i) mediated hy-
drolysis of the [BF4]À anion occurs to afford an oxyborate
anion that binds strongly with the metal centre, shutting down
the catalytic cycle dramatically. In support of this, hydrolysis
of [BF4]À to afford coordinated oxyborates has been reported
previously.[60, 61]
That a silver-bound water molecule is strongly implicated as
a catalytic proton source in these reactions is demonstrated by
the following control experiments. Addition of just water to a
mixture of I and II affords no product, while under the
standard conditions used for catalysis addition of the hindered
base 2,6-di-tert-butyl-4-methylpyridine suppressed the reac-
tion completely.[56] Moreover, water need not be the unique
proton source. The reaction is easily repeated replacing an
alcohol (methanol, 20 mol%) for water, resulting in essen-
tially identical product yields and reaction times.
The results of these NMR experiments, in which water is
added, are shown in Figure 7 for the new complexes 1 through
4 and for comparison the complexes [Ag(PPh3)(BF4)],
[Ag(PPh3)(ClO4)][58] and [Ag(PPh3)(OTf)].[16] All of these
experiments were carried out at 0.1 mol% catalyst loading
with 50 mol% added water in solutions in CD2Cl2, with the
diene in 1.5 molar excess. The final products in the NMR tube
are a consistent mixture of intermediate V and final product
III (ca. 90:10 ratio), which on workup gives only the final
product. The reactions were monitored by selected peaks due
to intermediate V and the final product (see Experimental
Section). No hydrolysis of the imine to benzaldehyde was
observed under the conditions used, while the consumption of
imine I in each run followed the same time-dependent profile
(although inversed) as the increase in V and III.
It is clear from Figure 7 that complex 3 gives the fastest rate
of catalysis, complete conversion being apparent after 15 min-
utes. This equates to a turnover frequency (TOF) of
4000hourÀ1, which is excellent for a Lewis acid catalysed
reaction of this type. Complex 1, although a slower catalyst
than 3, also yields complete conversion of the imine after
ꢀ40 minutes. Not unsurprisingly the monophosphane com-
plexes, [Ag(PPh3)(ClO4)] and [Ag(PPh3)(OTf)], which con-
tain more nucleophilic counterions, are less active than 1 and
3. The addition of an extra PPh3 ligand would be expected to
Conclusion
The preparation of the four silver(i) phosphane complexes
partnered with the carborane anions [closo-CB11H12] and
Figure 7. Chart showing relative rates of reaction between I and II by using 0.1 mol% of catalysts 1 4 and [Ag(PPh3)(Y)] [Yˆ OTf, BF4, ClO4] and
50 mol% added H2O at room temperature in CD2Cl2. See text for other details.
Chem. Eur. J. 2002, 8, No. 9
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2095 $ 20.00+.50/0
2095

A. S. Weller, C. G. Frost et al.
FULL PAPER
À12.0 (5B); 11B d ˆ À10.3 (d, 5‡1 coincidence, J(HB) ˆ 118 Hz), À12.0 (d,
J(HB) ˆ 110 Hz, 5B); 31P{1H} (228C): d ˆ 18.70 (dd, J(Ag109P) ˆ 795,
J(Ag107P) ˆ 691 Hz, 1P); IR (KBr): nÄ ˆ 2565 (vs, BH), 2517 (shs, BH),
2372 cmÀ1 (m, BH); elemental analysis calcd (%) for C19H27B11AgP: C 44.5,
H 5.30; found: C 44.3, H 5.19.
[closo-CB11H6Br6] has allowed a study to be made of the
relationship between solid and solution structures and per-
formance as partners to cationic Lewis acids of these anions in
an organic transformation. Not unsurprisingly, given the now
well-documented[4, 5] inertness coupled with the low nucleo-
philicity of the hexahalageno anions and the precedent for
pronounced counterion effects in Lewis acid catalysed cyclo-
additions,[62 64] the monophosphane complex 3 is a significant-
ly better catalyst than both 1 and all of the other systems
examined. In addition, compound 3 (and presumably the
others, although not tested) is selective for imines and can be
used at significantly lower catalyst loadings than previously
reported for Lewis acid catalysed hetero-Diels Alder reac-
tions.
To us, the surprising result is that the relative ordering in
rate of catalysis between [CB11H12] and [CB11H6Br6] is
reversed when two phosphanes are coordinated to the silver
centre. Although this was unexpected, it is pleasing that a
gross correlation exists between observable NMR properties
in solution (magnitude of J(AgP)) and the catalytic activity.
Thus, complexes 1 and 3, which have large AgP coupling
constants, are better catalysts than 2 and 4, with the latter
complex exhibiting the poorest performance and the lowest
coupling constant by far (albeit measured at low temper-
ature). Overall this suggests that when considering catalyst
performance in these, and related systems, the influence of
counterion and ligands are not independent and need to be
considered together. It is however encouraging that useful
catalyst performance indicators may be gleaned from simple
NMR experiments, which could help in the identification of
promising catalytic systems. To this end, we are currently
investigating the applicability of complexes such as 3 in other
Lewis acid catalysed reactions and will report on these in due
course.
[Ag(PPh3)2(CB11H12)] (2): Ag[CB11H12] (0.075 g, 0.299 mmol) and PPh3
(0.158 g, 0.602 mmol) were stirred together in CH2Cl2 (15 mL) in the dark
for 1 hour. This solution was filtered and hexane (20 mL) added to the
filtrate to induce crystallisation. The colourless product was isolated by
decanting off the solvent and drying in vacuo (0.204 g, 88% yield). Crystals
suitable for an X-ray diffraction study were grown by redissolving a portion
of the solid product in a minimum of CH2Cl2, layering with hexanes and
then placing in a freezer overnight at À308C to yield white crystals.
1H{11B} (228C): d ˆ 7.45 7.18 (m, 30H; C6H5), 2.21 (brs, 1H; CHcage), 1.85
(brs, 1H; BHa), 1.58 (brs, 10H; 5‡5 coincidence, BH); 11B{1H} (CD2Cl2,
228C): d ˆ À5.1 (brs, 1B), À10.6 (brs, 5B), À12.6 (brs, 5B); 11B (228C):
d ˆ À5.1 (d, J(HB) ˆ 128 Hz, 1B), À10.6 (d, J (HB) ˆ 138 Hz, 5B), À12.6
(d, J(HB) ˆ 158 Hz, 5B); 31P{1H} (CD2Cl2, 228C): d ˆ 14.8 (s, br); 31P{1H}
(À608C): d ˆ 13.7 (brd, J(AgP) ˆ 333 Hz), 13.6 (dd, J(Ag109P) ˆ 554,
J(Ag107P) ˆ 483 Hz); IR (KBr): 2544 (vs, BH), 2442 (m, BH), 2396 cmÀ1
(m, BH); elemental analysis calcd (%) for C37H42B11AgP2: C 57.3, H 5.46;
found: C 57.2, H 5.53.
[Ag(PPh3)(CB11H6Br6)] (3): PPh3 (0.062 g, 0.236 mmol) was dissolved in
CH2Cl2 (10 mL) and added dropwise to a Schlenk flask charged with
Ag[CB11H6Br6] (0.200 g, 0.276 mmol) in the dark and with stirring. The
resulting solution was stirred overnight and filtered. The supernatent
solvent was removed in vacuo to leave a clear oil. Crystals suitable for an
X-ray diffraction study were grown by redissolving the product up in
minimum CH2Cl2, layering with hexane, then placing the sample in a
freezer overnight at À308C (0.194 g, 83% yield).
1H{11B}: d ˆ 7.52 7.22 (m, 15H; C6H5), 2.73 (brs, 1H; CHcage), 2.43 (5H;
BH); 11B: d ˆ À2.1 (brs, 1B, B), À6.6 (brs, 5B, BBr), À16.9 (d, J(HB) ˆ
157 Hz, 5B, BH); 31P{1H}: d ˆ 16.52 (d of d, J(Ag109P) ˆ 766, J(Ag107P) ˆ
664 Hz, 1P); IR (KBr): 2608 (vs, BH), 2593 cmÀ1 (s, BH); elemental
analysis calcd (%) for C19H21B11AgPBr6: C 23.1, H 2.12; found: C 22.6, H
2.17.
[Ag(PPh3)2(CB11H6Br6)] (4): The compounds Ag[CB11H6Br6] (0.250 g,
0.345 mmol) and PPh3 (0.181 g, 0.690 mmol) were stirred together in
CH2Cl2 (20 mL) in the dark for 1 hour. The solution formed was filtered
and solvent removed in vacuo to leave a white solid. This solid was
redissolved in minimum CH2Cl2, layered with hexane and placed in a
freezer overnight at À308C to afford a colourless microcrystalline powder
(0.246 g, 57%).
1H{11B}: d ˆ 7.45 6.87 (m, 30H; C6H5), 2.45 (brs, 1H; CHcage), 2.21 (brs,
5H); 11B: d ˆ 1.3 (brs, 1B, B), À6.8 (s, 5B), À17.2 (d, J(HB) ˆ 166 Hz, 5B);
31P{1H} (228C): d ˆ 8.1 (brs); 31P{1H} (À808C): d ˆ 8.06 (dd, J(109AgP) ˆ
256, 107J(AgP) ˆ 223 Hz); IR (KBr): 2956 (vs, BH), 2587 cmÀ1 (s, BH);
elemental analysis calcd (%) for C37H36B11AgP2Br6: C 35.7, H 2.89; found:
C 35.5, H 3.01.
Experimental Section
General: All manipulations were carried out under an argon atmosphere
by using standard Schlenk line or dry box techniques. CH2Cl2 was distilled
from CaH2 and hexane was distilled from sodium. Solution NMR spectra
were measured on Varian400 MHz and Bruker Advance300 MHz FT-
NMR spectrometers in solutions in CD2Cl2. Residual protio solvent was
used as a reference (CD2Cl2, d ˆ 5.33) in 1H NMR, BF3 ¥ OEt2 (external) in
11B NMR and 85% H3PO4 (external) for 31P NMR spectra. Coupling
constants are given in Hz. Elemental analyses were performed in-house at
the Department of Chemistry. The complexes Ag[CB11H12][26] and
Ag[CB11H6Br6][13] were prepared by the published literature routes or
variations thereof. All other chemicals were used as purchased from
Aldrich.
General experimental procedure for catalytic studies: The imine I (0.200 g,
1.1 mmol) in dichloromethane (1.5 mL) at room temperature was added to
a stirring solution of freshly recrystallised catalyst (1 mol%) in dichloro-
methane (3.5 mL). The reaction mixture was allowed to stir for 5 minutes
then Danishefsky×s diene (II) (320 mL, 1.65 mmol) was added dropwise.
After 60 minutes, the reaction was quenched with aqueous sodium
hydrogen carbonate and extracted with ethyl acetate. The crude product
was purified by chromatography on silica gel (hexane/ethyl acetate 4:1) to
afford the product III.
Preparations
[Ag(PPh3)(CB11H12)] (1): PPh3 (0.163 g, 0.650 mmol) was dissolved in
CH2Cl2 (10 mL) and added dropwise to a Schlenk flask charged with
Ag[CB11H12] (0.155 g, 0.591 mmol), in the dark and with stirring. This
solution was stirred overnight and then cannula filtered. The solvent was
removed in vacuo to leave a pale yellow solid. Colourless crystals suitable
for an X-ray diffraction study were grown by redissolving the product up in
minimum CH2Cl2, layering with hexane, then placing the sample in a
freezer overnight at À308C (0.281 g, 92% yield).
1H NMR: d ˆ 7.67 (dd, J(HH) ˆ 1, 7 Hz, 1H), 7.35 6.99 (m, 10H), 5.27 (dd,
J(HH) ˆ 1, 7 Hz, 2H), 3.29 (dd, J(HH) ˆ 7, 16 Hz, 1H), 2.77 (ddd, J(HH) ˆ
1, 3, 16 Hz, 1H).
General procedure for NMR tube reactions: Solutions of catalyst were
typically prepared by dissolving the compound (1 mg) in CD2Cl2 (1 mL)
with the use of an ultrasound bath to ensure complete catalyst dissolution,
although, by eye, the solids seemed to have dissolved completely. The
relevant quantity of catalyst solution to give a 0.1 mol% catalyst concen-
tration (i.e. 0.00011 mmol of catalyst) was taken from this standard solution
and placed in an NMR tube previously charged with N-benzylideneaniline
(20 mg, 0.11 mmol). Danishefsky×s diene (32 mL, 0.17 mmol) and water
(1 mL, 0.056 mmol) were added to the NMR tube, which was then shaken
1H{11B} (228C): d ˆ 7.52 7.29 (m, 15H; C6H5), 2.55 (brs, 1H; CHcage), 2.25
(brs, 1H; BH), 1.85 (10H; 5‡5 coincidence, BH); selected 1H{11B}
(À908C): d ˆ 2.59 (brs, 1H; CHcage), 2.34 (brs, 1H; BH), 1.94 (5H; BH),
1.76 (5H; BH); 11B{1H} (228C): d ˆ À10.3 (shbr, 1B), À11.2 (br, 5B),
2096
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2096 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9

Silver Phosphanes
2088 2098
vigorously before being placed in the NMR spectrometer and measure-
[20] D. D. Ellis, A. Franken, P. A. Jellis, J. A. Kautz, F. G. A. Stone, Dalton
Trans. 2000, 2509.
[21] D. D. Ellis, J. C. Jeffery, P. A. Jelliss, J. A. Kautz, F. G. A. Stone, Inorg.
Chem. 2001, 40, 2041.
[22] Y.-W. Park, J. Kim, Y. Do, Inorg. Chem. 1994, 33, 1.
[23] D. J. Liston, C. A. Reed, C. W. Eigenbrot, W. R. Scheidt, Inorg. Chem.
1987, 26, 2739.
[24] N. J. Patmore, J. W. Steed, A. S. Weller, Chem. Commun. 2000, 1055.
[25] N. J. Patmore, M. F. Mahon, J. W. Steed, A. S. Weller, Dalton Trans.
2001, 277.
[26] K. Shelly, D. C. Finster, Y. J. Lee, W. R. Scheidt, C. A. Reed, J. Am.
Chem. Soc. 1985, 107, 5955.
[27] S. V. Ivanov, A. J. Lupinetti, S. M. Miller, O. P. Anderson, K. A.
Solnstev, S. H. Strauss, Inorg. Chem. 1995, 34, 6419.
[28] Z. W. Xie, T. Jelinek, R. Bau, C. A. Reed, J. Am. Chem. Soc. 1994, 116,
1907.
[29] S. V. Ivanov, J. J. Rockwell, S. M. Miller, O. P. Anderson, K. A.
Solntsev, S. H. Strauss, Inorg. Chem. 1996, 35, 7882.
[30] C.-W. Tsang, Q. Yang, E. T.-P. Sze, T. C. W. Mak, D. T. W. Chan, Z.
Xie, Inorg. Chem. 2000, 39, 5851.
[31] Z. Xie, B.-M. Wu, T. C. W. Mak, J. Manning, C. A. Reed, J. Chem. Soc.
Dalton Trans. 1997, 1213.
ments were taken at timed intervals. The disappearance of the peak at d ˆ
ˆ
8.51 due to PhN CHPh was monitored, along with the growth of the peaks
centred at d ˆ 6.64 and d ˆ 6.54 [3H total, intermediate V] and d ˆ 5.27 [1H
total, final product III]. A plot for each catalyst, of time verses con-
sumption of I and time verses total concentrations of V and III showed
essentially the same profile. Repeat runs for all the catalysts tested showed
the same time-dependent profiles.
1
Intermediate V: H NMR: d ˆ 7.56 (d, J(HH) ˆ 13 Hz, 1H), 7.42 7.40 (m,
2H), 7.28 7.23 (m, 1H), 7.10 7.01 (m, 3H), 6.64 (m, 1H), 6.54 (m, 2H),
5.58 (d, J(HH) ˆ 12 Hz, 1H), 4.86 (q, J(HH) ˆ 6 Hz, 1H), 4.71 (d, J(HH) ˆ
6 Hz, 1H), 3.68 (s, 3H), 2.94 (dd, J(HH) ˆ 6 Hz, 22H), 0.144 (s, 9H).
X-ray crystallography: The crystal structure data for compounds 1 3 were
collected on a Nonius KappaCCD. Structure solution followed by full-
matrix least-squares refinement was performed by using the SHELX suite
of programs throughout.[65] Hydrogens were included at calculated
positions throughout, with the exception of H7 and H12 in compound 2.
These latter hydrogens were readily located in the penultimate difference
Fourier map and refined at a fixed distance 1.12 ä from parent atoms B(7)
and B(12), respectively. Plots were produced by using ORTEX.[66]
CCDC-171721 (1), 171769 (2) and 168828 (3) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cam-
bridge Crystallographic Data Centre, 12 Union Road, Cambridge
CB21EZ, UK; fax: (‡44)1223-336-033; or e-mail: deposit@ccdc.cam.ac.
uk).
[32] Z. Xie, R. Bau, C. A. Reed, Angew. Chem. 1994, 106, 2566; Angew.
Chem. Int. Ed. Engl. 1994, 33, 2433.
[33] T. Jelinek, P. Baldwin, W. R. Scheidt, C. A. Reed, Inorg. Chem. 1993,
32, 1982.
[34] A. Yanagisawa in Lewis Acids in Organic Synthesis (Ed.: H.
Yamamoto), Wiley-VCH, Weinheim, 2000.
[35] T. Hayashi, M. Sawamura, Y. Ito, Tetrahedron 1999, 48, 1992.
[36] A. Togni, S. D. Pastor, J. Org. Chem. 1990, 55, 1649.
[37] A. Yanagisawa, H. Nakashima, A. Ishiba, H. Yamamoto, J. Am.
Chem. Soc. 1996, 118, 4723.
Acknowledgements
[38] A. Yanagisawa, Y. Nakatuska, K. Asakawa, H. Kageyama, H.
Yamamoto, Synlett 2001, 69.
[39] M. Ohkouchi, D. Masui, M. Yamaguchi, T. Yamagishi, J. Mol. Catal. A
2001, 170, 1.
[40] A. Yanagisawa, Y. Matsumoto, H. Nakashima, K. Asakawa, H.
Yamamoto, J. Am. Chem. Soc. 1997, 119, 9319.
[41] S. Yoo, S. Saaby, R. G. Hazell, K. A. J˘rgensen, Chem. Eur. J. 2000, 6,
2435.
A.S.W. thanks the Royal Society for a University Research Fellowship and
equipment grant. The University of Bath is also thanked for financial
support (studentships to N.J.P. and C.H.). C.G.F. thanks Astra-Zeneca for
generous support from their strategic research fund. The use of the
Cambridge Structural Database at Daresbury service is acknowledged. The
EPSRC/JERI are acknowledged for funding for the diffractometer.
[42] C. Hague, N. J. Patmore, C. G. Frost, M. F. Mahon, A. S. Weller,
Chem. Commun. 2001, 2286.
[1] C. A. Reed, Acc. Chem. Res. 1998, 31, 133.
[43] D. J. Liston, Y. J. Lee, W. R. Scheidt, C. A. Reed, J. Am. Chem. Soc.
1989, 111, 6643.
[44] A. S. Weller, M. F. Mahon, J. W. Steed, J. Organomet. Chem. 2000,
614 615, 113.
[45] G. D. Ruggiero, A. S. Weller, I. H. Williams, unpublished results.
[46] H.-P. Wu, C. Janiak, G. Rheinwald, H. Lang, J. Chem. Soc. Dalton
Trans. 1999, 183.
[2] S. H. Strauss, Chem. Rev. 1993, 93, 927.
[3] I. Zharov, B. T. King, Z. Havlast, A. Pardi, J. Michl, J. Am. Chem. Soc.
2000, 122, 10253.
[4] C. A. Reed, N. L. P. Fackler, K. C. Kim, D. Stasko, D. R. Evans,
P. D. W. Boyd, C. E. F. Rickard, J. Am. Chem. Soc. 1999, 121, 6314.
[5] C. A. Reed, K. C. Kim, R. D. Bolskar, L. J. Mueller, Science 2000, 289,
101.
[47] C. Kleina, E. Graf, M. W. Hosseini, A. D. Cian, J. Fischer, Chem.
Commun. 2000, 239.
[6] A. J. Lupinetti, M. D. Havighurst, S. M. Miller, O. P. Anderson, S. H.
Strauss, J. Am. Chem. Soc. 1999, 121, 11920.
[48] S. A. Brew, F. G. A. Stone, Adv. Organomet. Chem. 1993, 35, 135.
[49] A. S. Weller, T. P. Fehlner, Organometallics 1999, 18, 447.
[50] H. Brunner, D. Mijolovic, B. Wrackmeyer, B. Nuber, J. Organomet.
Chem. 1999, 579, 298.
[7] H. W. Turner, European Patent Application 277003, 1988; [Chem.
Abstr. 1988, 110, 58291].
[8] G. G. Hlatky, R. R. Eckman, H. W. Turner, Organometallics 1992, 11,
1413.
[51] G. S. Mhinzi, S. A. Litster, A. D. Redhouse, J. L. Spencer, J. Chem.
Soc. Dalton Trans. 1991, 2769.
[9] G. G. Hlatky, H. W. Turner, R. R. Eckman, J. Am. Chem. Soc. 1989,
111, 2728.
[52] D. M. V. Seggen, P. K. Hurlburt, O. P. Anderson, S. H. Strauss, Inorg.
Chem. 1995, 34, 3453.
[10] D. J. Crowther, S. L. Borkowsky, D. Swenson, T. Y. Meyer, R. F.
Jordan, Organometallics 1993, 12, 2897.
[53] H. Kobayashi, T. Busujima, S. Nagayama, Chem. Eur. J. 2000, 6, 3491.
[54] S. Ribe, P. Wipf, Chem. Commun. 2001, 299.
[55] K. Mikami, O. Kotera, Y. Motoyama, H. Sakaguchi, Synlett 1995,
975.
[56] A. G. M. Barrett, D. C. Braddock, J. P. Henschke, E. R. Walker,
Perkin Trans. 1 1999, 873.
[11] W. V. Konze, B. L. Scott, G. J. Kubas, Chem. Commun. 1999, 1807.
[12] J. Powell, A. Lough, T. Saeed, J. Chem. Soc. Dalton Trans. 1997, 4137.
[13] E. L. Muetterties, C. W. Alegranti, J. Am. Chem. Soc. 1972, 94, 6386.
[14] R. E. Bachmann, D. F. Andretta, Inorg. Chem. 1998, 37, 5657.
[15] G. A. Bowmaker, Effendy J. V. Hanna, P. C. Healy, B. W. Skelton,
A. H. White, J. Chem. Soc. Dalton Trans. 1993, 1387.
[16] M. BardajÌ, O. Crespo, A. Laguna, A. K. Fischer, Inorg. Chim. Acta.
2000, 304, 7.
[57] N.-M. Kao, C. P. Grey, K. Pitchumai, P. H. Lakshminarasimhan, V.
Ramamurthy, J. Phys. Chem. 1998, 102, 5627.
[58] T. G. M. H. Dikkhoff, R. G. Goel, E. L. Muetterties, C. W. Alegranti,
Inorg. Chim. Acta 1980, 44, L72.
[17] G. A. Bowmaker, J. V. Hanna, C. E. F. Rickard, A. S. Lipton, Dalton
Trans. 2001, 20.
[59] Note added in proof: After the manuscript was submitted the solid-
[18] H. M. Colquohoun, T. J. Greenhough, M. G. H. Wallbridge, J. Chem.
Soc. Chem. Commun. 1980, 192.
[19] D. D. Ellis, P. A. Jellis, F. G. A. Stone, Organometallics 1999, 18, 4982.
state structure of compound 4 was determined and it shows
a
‡
monomeric {Ag(PPh3)2} fragment bound to one [CB11H6Br6] cage
Chem. Eur. J. 2002, 8, No. 9
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2097 $ 20.00+.50/0
2097

A. S. Weller, C. G. Frost et al.
FULL PAPER
by two lower pentagonal belt bromine atoms, consistent with the
structure proposed from the low-temperature 31P{1H} NMR spectrum.
[60] C. J. d. Reijer, M. Wˆrle, P. S. Pregosin, Organometallics 2000, 19, 309.
[61] G. Brauers, F. J. Feher, M. Green, J. K. Hogg, A. G. Orpen, J. Chem.
Soc. Dalton Trans. 1996, 3387.
[64] D. A. Evans, J. A. Murry, P. v. Matt, R. D. Norcross, S. J. Miller,
Angew. Chem. 1995, 107, 864; Angew. Chem. Int. Ed. Engl. 1995, 34,
798.
[65] G. M. Sheldrick, SHELX-97. Program for Refinement of Crystal
Structures, University of Gˆttingen, Gˆttingen (Germany), 1997.
[66] P. McArdle, J. Appl. Crystallogr. 1995, 28, 65.
[62] A. Lightfoot, O. Schinder, A. Pfaltz, Angew. Chem. 1998, 110, 3047;
Angew. Chem. Int. Ed. 1998, 37, 2897.
[63] E. P. K¸ndig, C. M. Saudan, G. Bernardinelli, Angew. Chem. 1999, 111,
1298; Angew. Chem. Int. Ed. 1999, 38, 1220.
Received: October 15, 2001 [F3613]
2098
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0809-2098 $ 20.00+.50/0
Chem. Eur. J. 2002, 8, N o. 9