no water is added then catalyst performance drops off rapidly on
the second and third runs, however with 10 mol% water high
yields of > 95% are afforded over three consecutive cycles
(ESI†). At the relatively high catalyst loadings (10 mol%) used
in these preliminary experiments the counter ion effects
observed in the homogeneous system are not observed.
In conclusion, we have demonstrated that silver( )–phosphine
I
complexes partnered by carborane anions based on [CB11H12]
are effective and active catalysts in a hetero-Diels–Alder
transformation, and that water dramatically effects the observed
rate of reaction. These catalysts may also be supported on
commercially available resin to give active and recyclable
Lewis acid catalysts. Full details of the synthesis, solution and
solid state structures of 1 and 2 along with comparisons of the
catalytic performance when other common anions are partnered
Fig. 2 Relative rates of reaction CD2Cl2 solutions, 0.1 mol% 1, 2 and
(PPh3)Ag(OTf), 50 mol% added H2O. Also shown: 1 with no added
water.
with silver(
)–phosphines will be reported in due course.10
I
A. S. W. thanks the Royal Society for a University Research
Fellowship. The EPSRC and JERI are thanked for providing
funds for the purchase of a diffractometer. The referees are
thanked for useful comments.
shown here, and showed a very similar time dependent profile
to (PPh3)Ag(OTf).] It is clear that complex 1 gives the fastest
rate of catalysis, the reaction being complete in ~ 15 min,
yielding a turnover frequency (TOF) of ca. 4000 h21. Complex
2 is slower than 1, with complete reaction effected after 50 min,
while (PPh3)Ag(OTf) does not attain 100% conversion, even
after 80 min. Although these results reflect the previously
enunciated relative weakly coordinating properties of carborane
anions,2 especially the halogenated examples,4 it is pleasing to
see that these ideas can be applied to a synthetically useful
transformation, using well defined catalysts.
Importantly, and somewhat surprisingly, we have found that
water plays an important role in this reaction. When these NMR
experiments are performed in rigorously dry solvent (CD2Cl2,
vacuum distilled from CaH2), no catalysis occurs (Fig. 2).
Addition of a substoichiometric (1 ml, 50 mol%, unoptimised)
amount of H2O to the sample immediately initiates catalysis.
While water-accelerated catalysis is becoming appreciated
more widely,5 we currently can only speculate as to its rôle in
this system. These results, however, implicate a polarised,
silver-bound water molecule in the catalytic process, with the
resulting Lewis-assisted Brønsted-acid similar to that pre-
viously reported in lanthanide catalysed aromatic electrophilic
substitutions6 and the catalytic role of coordinated water in
certain zeolites.7 Consonant with this idea, when the reaction
was repeated in the presence of the hindered base 2,6-di-tert-
butyl-4-methylpyridine no product was formed, while a control
experiment taking I and II with just 50 mol% water also resulted
in no product formation. On the bench, it is probably
adventitious water that is made available to the reaction.
Catalyst 1 is also selective for imines over aldehydes in this
reaction. A competition experiment demonstrated preferential
activation of imine I in the presence of an equimolar amount of
benzaldehyde. Thus, after 15 min at room temperature the
product arising from reaction with the imine was isolated in
70% yield whilst the aldehyde adduct was produced in only 5%
yield.
Notes and references
‡ NMR data (CD2Cl2 solutions, 22 °C, 300 MHz). Complex 1: 1H{11B}, d
7.52–7.22 (15H, m, C6H5), 2.73 (1H, s br, CHcage), 2.43 (5H, BH). 11B{1H},
d 22.09 (1B), 26.59 (5B), 216.88 (5B). 31P{1H}, d 16.52 [1P, d d,
J(Ag109P) 766, J(Ag107P) 664 Hz]. IR/cm21 (KBr): 2608vs (BH), 2593s
(BH). Calc. for C19H21B11AgPBr6: C, 23.1; H, 2.12. Found: C, 22.6; H
2.17%. Complex 2: H{11B}, d 7.52–7.29 (15H, m, C6H5), 2.55 (1H, s br,
CHcage), 2.25 (1H, s br, BH), 1.85 (10H, 5 + 5 coincidence, BH): 11B{1H}.
d 210.27 (1B, s br), 211.18 (5 B), 212.02 (5 B). 31P{1H}, d 18.70 [1P, dd,
J(Ag109P) 795, J(Ag107P) 691 Hz]. IR/cm21 (KBr): 2565vs (BH), 2517s
(sh) (BH), 2372m (BH). Calc. for C19H27B11AgP: C, 44.5; H, 5.30. Found:
C, 44.3; H, 5.19%.
§ Crystallographic data: for 1: C19H21AgB11Br6P, M
=
986.57, l
=
=
0.71073 Å, monoclinic, space group P21/c, a
=
8.8950(10), b
24.4140(4), c = 14.4170(3) Å, b = 102.0720(10)°, U = 3061.60(9) Å3, Z
= 4, T = 150(2) K, Dc = 2.140 g cm23, m = 8.554 mm21, F(000) = 1848,
crystal: 0.10 3 0.10 3 0.05 mm, 7253 unique reflections (Rint = 0.0519),
R1 = 0.0358, wR2 = 0.0783 [I > 2s(I)]. For 2: C19H27AgB11P, M = 513.16,
¯
l = 0.71073 Å, triclinic, space group P1, a = 10.282(2), b = 10.988(2), c
=
11.308(2) Å, a
= 76.22(3), b = 75.25(3), g = 75.61(3)°, U =
1175.7(4) Å3, Z = 2, T = 150(2) K, Dc = 1.450 g cm23, m = 0.932 mm21
,
F(000) = 516, crystal 0.20 3 0.20 3 0.10 mm, 5374 unique reflections (Rint
= 0.0387), R1 = 0.0316, wR2 = 0.0757 [I > 2s(I)].
suppdata/cc/b1/b106719b/ for crystallographic data in CIF or other
electronic format.
¶ General procedure: ~ 3 mmol g21 polymer-bound PPh3 (Fluka, 200–300
mesh) with a slight excess of Ag[Y] in CH2Cl2 was used to generate the
polymer bound catalyst. After washing with CH2Cl2 and drying in vacuo,
catalytic runs were performed using ca. 10 mol% catalyst and 10 mol% H2O
in CH2Cl2 (5 ml) (1 h, room temp.). The support was recycled by cannula
filtration of the supernatent under argon and drying in vacuo.
1 Lewis Acids in Organic Synthesis, ed. H. Yamamoto, Wiley-VCH,
Weinheim, 2000; P. Buonora, J. C. Olsen and T. Oh, Tetrahedron, 2001,
57, 6099.
Polymer supported Lewis acids have been recently attracting
significant attention as they represent one method of generating
clean, efficient and re-usable catalysts.8 However, the in-
corporation of the Lewis acidic sites onto the polymer often
requires a multistep synthesis. Given that one of the most
popular supports is polymer bound triphenylphosphine,9 we
were interested to see if the high activity and efficiency shown
by complexes 1 and 2 could be transferred to this support.
2 C. A. Reed, Acc. Chem. Rev., 1998, 31, 133.
3 D. D. Ellis, A. Franken, P. A. Jellis, J. A. Kautz and F. G. A. Stone, J.
Chem. Soc., Dalton Trans., 2000, 2509; D. D. Ellis, J. C. Jeffery, P. A.
Jellis, J. A. Kautz and F. G. A. Stone, Inorg. Chem., 2001, 40, 2041.
4 C. A. Reed, K. C. Kim, R. D. Bolskar and L. J. Mueller, Science, 2000,
289, 101.
5 S. Ribe and P. Wipf, Chem. Commun., 2001, 299.
6 A. G. M. Barrett, D. C. Braddock, J. P. Henschke and E. R. Walker, J.
Chem. Soc., Perkin Trans. 1, 1999, 873.
7 H.-M. Kao, C. P. Grey, K. Pitchumani, P. H. Lakshminarasimhan and V.
Ramamurthy, J. Phys. Chem. A., 1998, 102, 5627.
8 S. V. Ley, A. R. Baxendale, R. N. Bream, P. S. Jackson, A. G. Leach,
D. A. Longbottom, M. Nesi, J. S. Scott, R. I. Storer and S. J. Taylor, J.
Chem. Soc., Perkin Trans. 1, 2000, 3815; S. Itsuno, in Lewis Acids in
Organic Synthesis, ed. H. Yamamoto, Wiley-VCH, Weinheim, 2000.
9 A. C. Comely, S. E. Gibson and N. J. Hales, Chem. Commun., 2000,
305.
Stirring a CH2Cl2 solution of Ag[Y][Y
= CB11H12 3,
CB11H6Br6 4, OTf 5] with commercially available resin
(Fluka) afforded a material that was an efficient catalyst in all
three cases for the hetero-Diels–Alder reaction under investiga-
tion.¶ Moreover, all the resins were shown to be re-usable over
at least three catalyst runs ( > 95% isolated yield, vide infra). In
concert with this, low leaching levels (0.3% Ag, by AAS) where
also determined, while the supernatant from freshly prepared
and filtered supported catalyst afforded only trace product
( < 5%) when used in the reaction. These supported catalysts
also show a significant dependence on the presence of water. If
10 C. Hague, N. J. Patmore, J. H. Cotgreave, M. F. Mahon, C. G. Frost and
A. S. Weller, manuscript in preparation.
Chem. Commun., 2001, 2286–2287
2287