t-Bu-Amphos-RhCl3·3H2O: A highly recyclable catalyst system for the cross-coupling of aldehydes and aryl- and alkenylboronic acids in aqueous solvents

Article abstract of DOI:10.1039/b509406b

The combination of t-Bu-Amphos and RhCl₃·3H₂O gave the first highly recyclable catalyst for the coupling of aryl- and vinylboronic acids with aldehydes in aqueous solvents. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b509406b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
t-Bu-Amphos–RhCl3?3H2O: a highly recyclable catalyst system for the
cross-coupling of aldehydes and aryl- and alkenylboronic acids in
aqueous solvents{
Rongcai Huang and Kevin H. Shaughnessy*
Received (in Cambridge, UK) 5th July 2005, Accepted 18th July 2005
First published as an Advance Article on the web 3rd August 2005
DOI: 10.1039/b509406b
sources in combination with hydrophilic phosphines, such as
TPPTS and t-Bu-Amphos, have been shown to promote the
addition of arylboronic acids to activated alkenes in aqueous
solvents.13–17 To the best of our knowledge, there are no reports of
the use of hydrophilic catalysts for the addition of organoboron
reagents to aldehydes in aqueous solvents.
The combination of t-Bu-Amphos and RhCl3?3H2O gave the
first highly recyclable catalyst for the coupling of aryl- and
vinylboronic acids with aldehydes in aqueous solvents.
Addition of carbon nucleophiles to carbonyl compounds, such as
in Grignard–Barbier reactions, is a classic method for carbon–
carbon bond formation. Rhodium-catalyzed addition of air- and
moisture-stable organoboron or organotin reagents offers a more
functional group tolerant method than the traditional Grignard–
Barbier reagents.1 In addition, by use of chiral ligands, modest
enantioselectivities have been achieved with this methodology.2–4
Oi et al.5 and Miyaura et al.2 initially reported the rhodium-
catalyzed addition of arylstannanes and arylboronic acids to
aldehydes, respectively. Ueda and Miyaura6 subsequently showed
that sterically demanding alkylphosphines, such as t-Bu3P gave
more active catalysts than the chelating arylphosphines used in his
original report. Strongly s-donating N-heterocyclic carbene
ligands have also been shown to be highly effective in these
coupling reactions.7,8
The coupling of benzaldehyde and phenylboronic acid in 1 : 1
water–acetonitrile at 80 uC was chosen as the initial model reaction
on which to optimize the catalytic system (eqn (1), Table 1). Initial
scouting reactions indicated that sodium hydroxide was required
for the reaction to proceed and that 80 uC was the optimal
temperature for acceptable reaction rates. Using these conditions
as a starting point, the effects of changing the rhodium source,
ligand, and solvent system were explored. Catalysts derived from
sulfonated arylphosphines (TPPTS and TXPTS) were inactive
under these conditions (Table 1, entries 2 and 3). Use of 1 : 1
toluene–water gave the same yield (84%) as 1 : 1 water–acetonitrile,
while use of water as the only solvent gave slightly higher yields
(90%). RhCl3?3H2O, [RhCl(cod)]2, and Rh(acac)(coe)2
(coe 5 cyclooctene) all gave nearly identical yields when used
as the rhodium source in coupling reactions carried out in water
(88–90%, Table 1, entries 5–7). Commercially available and air-
stable RhCl3?3H2O was chosen as the most convenient catalyst
precursor for further study.
We have recently reported the use of sterically demanding,
water-soluble alkylphosphines, such as t-Bu-Amphos, to give
active and recyclable catalysts for aqueous-phase palladium-
catalyzed cross-coupling reactions.9,10 While aqueous–organic
solvent systems are commonly used in the Rh-catalyzed arylation
of aldehydes, the use of hydrophobic ligands precludes retention of
the catalyst in the aqueous-phase. By using a hydrophilic ligand,
we envisioned that the rhodium catalyst would be retained in the
aqueous-phase of an aqueous–organic biphase, allowing the
catalyst to be easily recovered and potentially recycled. Li
et al.11,12 have reported ligand-free rhodium-catalyzed additions
of organoboron and tin reagents to aldehydes in water, although
the recyclability of this catalyst was not described. Rhodium
ð1Þ
Using the t-Bu-Amphos–RhCl3?3H2O catalyst system, phenyla-
tions of a variety of aldehydes were carried out in 1 : 1 water–
acetonitrile at 80 uC (Table 2). Benzhydrol was obtained from
benzaldehyde in 82% isolated yield under these conditions, while
Table 1 Catalyst optimization studiesa
Entry Rh salt
Ligand
Solvent
Yield (%)b
1
2
3
4
5
6
7
a
RhCl3?3H2O
t-Bu-Amphos CH3CN–H2Oc 85
RhCl3?3H2O
RhCl3?3H2O
RhCl3?3H2O
RhCl3?3H2O
[RhCl(cod)]2
TPPTS
TXPTS
CH3CN–H2Oc
CH3CN–H2Oc
0
0
t-Bu-Amphos Toluene–H2Od 84
Department of Chemistry and the Center for Green Manufacturing, The
University of Alabama, Box 870336, Tuscaloosa, AL 35487-0336, USA.
E-mail: kshaughn@bama.ua.edu; Fax: +1 205 348 9104;
Tel: +1 205 348 4435
{ Electronic supplementary information (ESI) available: Experimental
procedures, product characterization data, and 1H and 13C NMR spectra
of new compounds. See http://dx.doi.org/10.1039/b509406b
t-Bu-Amphos H2O
t-Bu-Amphos H2O
90
88
90
Rh(acac)(coe)2 t-Bu-Amphos H2O
Benzaldehyde (1 equiv.), phenylboronic acid (1.5 equiv.), NaOH
(2 equiv.), Rh salt (0.01 equiv.), ligand (0.01 equiv.), 1.5 mL solvent,
b
c
d
80 uC, 1 h. GC yield. 1 : 1 CH3CN–H2O. 1 : 1 toluene–H2O.
4484 | Chem. Commun., 2005, 4484–4486
This journal is ß The Royal Society of Chemistry 2005

View Article Online
moderate yields (23–55%) of coupling products (Table 3,
entries 3–5). Longer reaction times gave only marginally higher
yields in these reactions. An optimized system with an electron
deficient benzaldehyde derivative and an electron rich arylboronic
acid gave a 90% yield of the benzhydrol product. 2-Tolylboronic
acid gave similar yields to sterically unhindered arylboronic acids
(Table 3, entry 6). E-1-Octenylboronic acid also gave good yields
in couplings with benzaldehyde derivatives to give E-1-aryl-2-
alken-1-ols (Table 3, entries 7–9). An attempt to alkylate
benzaldehyde with 1-butylboronic acid was unsuccessful, however.
Although some limitations were encountered in exploring the
substrate scope, the complementary requirements for the aldehyde
and boronic acid allow most desired products to be obtained in
good yields. For example, the use of 4-cyanophenylboronic acid
and benzaldehyde gave a 23% yield of 4-cyanobenzhydrol, while
the same compound could be made in 86% yield by placing the
electron withdrawing group on the aldehyde electrophile.
Similarly, 1-aryl-1-alkanols could not be prepared from benzalde-
hyde and an alkylboronic acid, but could be prepared in moderate
yield by using an arylboronic acid and an aliphatic aldehyde.
One important advantage to using a hydrophilic catalyst in an
aqueous-biphasic solvent mixture is the potential to easily recover
and recycle the aqueous catalyst containing phase. The recycl-
ability of the t-Bu-Amphos–Rh catalyst system was explored in the
model coupling of benzaldehyde and phenylboronic acid. The
initial reaction was run using 2 mol% RhCl3?3H2O–t-Bu-Amphos
in water at 80 uC for 1 h. Upon completion, the reaction mixture
was extracted with deoxygenated ethyl acetate. Analysis of the
organic phase by GC indicated that benzydrol was produced in
79% yield (Table 4). The aqueous phase containing the catalyst
was transferred to a new vial under nitrogen containing fresh
substrates and base. The vial was then heated at 80 uC for 1 h and
the product recovered as before. This process was repeated for a
total of 9 reaction cycles. As the recycling study progressed, salts
began to precipitate from the aqueous phase. After completion of
the sixth cycle, the aqueous phase was allowed to stand overnight.
The aqueous supernatant was then removed from the precipitated
salts and used for cycles 7–9.
Table 2 Phenylboronic acid addition to aldehydes
Entry
1
Aldehyde
PhCHO
Product
Yield (%)a
82
78b
2
3
4
p-CN-C6H4CHO
p-Br-C6H4CHO
p-F-C6H4CHO
86
77
78
5
6
p-MeO-C6H4CHO
o-Me-C6H4CHO
59
78
7
8
C8H17CHO
50
52
c-C6H11CHO
9
t-BuCHO
0
Table 3 Coupling of aldehydes and boronic acids
a
1 mol% RhCl3?3H2O–t-Bu-Amphos, 1 : 1 H2O–CH3CN, 80 uC, 2 h.
Average isolated yield of two runs. Reaction run in water.
b
the same reaction in water gave a 78% yield (Table 2, entry 1).
Benzaldehyde derivatives with electron withdrawing groups gave
good yields of benzhydrol derivatives (Table 2, entries 2–4). In
contrast, incorporation of an electron releasing group gave lower
yields (Table 2, entry 5), even with long reaction times.
Incorporation of an ortho-methyl substituent on the aldehyde
had no effect on the reaction yield. Heptaldehyde and cyclohex-
anecarbaldehyde gave modest yields of coupled product, while
pivaldehyde gave no conversion (Table 2, entries 7–9). The lower
electrophilicity of the aliphatic aldehydes likely accounted for the
lower yields obtained with these substrates.
Entry
RCHO
R9B(OH)2
Yield (%)a
1
2
3
4
5
6
7
8
9
10
a
PhCHO
p-MeO-C6H4B(OH)2
p-MeO-C6H4B(OH)2
p-F-C6H4B(OH)2
p-NC-C6H4B(OH)2
p-HO2C-C6H4B(OH)2
o-Me-C6H4B(OH)2
(E-1-octenyl)B(OH)2
(E-1-octenyl)B(OH)2
(E-1-octenyl)B(OH)2
n-C4H9B(OH)2
79
90
55
23
28
86
84
71
73
0
p-CN-C6H5CHO
PhCHO
PhCHO
PhCHO
PhCHO
PhCHO
p-CN-C6H5CHO
o-Me-C6H5CHO
PhCHO
A contrasting electronic effect was seen when substituted
arylboronic acids were used (Table 3). Electron rich 4-methoxy-
phenylboronic acid gave good yields of coupled product (Table 3,
entry 1), while electron deficient arylboronic acids gave low to
1 mol% RhCl3?3H2O–t-Bu-Amphos, 1 : 1 H2O–CH3CN, 80 uC, 2 h.
Average isolated yield of two runs.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 4484–4486 | 4485

View Article Online
Table 4 Recycling of the t-Bu-Amphos–RhCl3 catalyst systema
unsaturated LRh(OH) complex. The t-Bu-Amphos–RhCl3 pre-
catalyst could give this type of active species upon reduction of
Rh(III) to Rh(I) with phenylboronic acid. We are currently
investigating the nature of the active species in this system.
In conclusion we have shown that the t-Bu-Amphos–
RhCl3?3H2O catalyst system provides an effective catalyst for the
addition of aryl- and vinylboronic acids to both aryl and aliphatic
aldehydes in aqueous solvents. Significantly the catalyst could be
used for 9 reaction cycles with little change in the yield of
benzhydrol.
Cycle
Yield (%)b
Cycle
Yield (%)b
1
2
3
4
5
a
79
83
90
85
90
6
7
8
9
90
86
74
76
Coupling of benzaldehyde (0.3 mmol) and phenylboronic acid
(0.33 mmol), 2 mol% t-Bu-Amphos–RhCl3, water (1.5 mL), 80 uC,
1 h. Cycles 2–9 used the aqueous catalyst solution from the previous
b
cycle. GC yield of benzhydrol obtained by ethyl acetate extraction
of the reaction mixture.
We thank the National Science Foundation (CHE 0124255) for
financial support of this work.
The catalyst proved to be highly stable in the recycling study,
during which the benzhydrol yield ranged from 74 to 90% with an
average yield of 84%. The yield in the initial run was slightly lower
than we typically see in single runs, possibly due to incomplete
extraction of the benzhydrol product. After the first cycle, the
yields varied from run to run, but remained between 83 and 90%.
Notably, the filtration after the sixth reaction cycle did not appear
to have a significant effect on catalyst activity. Only in the 8th and
9th cycles do the yields drop slightly. Even with this drop in yield,
the catalyst appears to retain a significant level of activity. To the
best of our knowledge, this work represents the first report of a
recyclable catalyst for the addition of organometallic reagents to
aldehydes. Nine cycles with a single catalyst charge is the highest
number of cycles that we are aware of in a Rh-catalyzed coupling
reaction.16,17
Notes and references
1 K. Fagnou and M. Lautens, Chem. Rev., 2003, 103, 169–196.
2 M. Sakai, M. Ueda and N. Miyaura, Angew. Chem., Int. Ed., 1998, 37,
3279–3281.
3 C. Moreau, C. Hague, A. S. Weller and C. G. Frost, Tetrahedron Lett.,
2001, 42, 6957–6960.
4 T. Focken, J. Rudolph and C. Bolm, Synthesis, 2005, 429–436.
5 S. Oi, M. Moro and Y. Inoue, Chem. Commun., 1997, 1621–1622.
6 M. Ueda and N. Miyaura, J. Org. Chem., 2000, 65, 4450–4452.
7 A. Fu¨rstner and H. Krause, Adv. Synth. Catal., 2001, 343, 343–350.
¨
8 I. Ozdemir, S. Demir and B. C¸ etinkaya, J. Mol. Catal. A: Chem., 2004,
215, 45–48.
9 R. B. DeVasher, L. R. Moore and K. H. Shaughnessy, J. Org. Chem.,
2004, 69, 7919–7927.
10 R. B. DeVasher, J. M. Spruell, D. A. Dixon, G. A. Broker, S. T. Griffin,
R. D. Rogers and K. H. Shaughnessy, Organometallics, 2005, 24,
962–971.
11 T. Huang, Y. Meng, S. Venkatraman, D. Wang and C.-J. Li, J. Am.
Chem. Soc., 2001, 123, 7451–7452.
12 C.-J. Li and Y. Meng, J. Am. Chem. Soc., 2000, 122, 9538–9539.
13 M. Lautens, A. Roy, K. Fukuoka, K. Fagnou and B. Martin-Matute,
J. Am. Chem. Soc., 2001, 123, 5358–5359.
14 M. Lautens and J. Mancuso, Org. Lett., 2002, 4, 2105–2108.
15 M. Lautens and M. Yashida, J. Org. Chem., 2003, 68, 762–769.
16 R. Amengual, V. Michelet and J.-P. Geneˆt, Tetrahedron Lett., 2002, 43,
5905–5908.
17 E. Genin, V. Michelet and J.-P. Geneˆt, J. Organomet. Chem., 2004, 689,
3820–3830.
The fact that the catalyst could be used for nine reaction cycles
at 80 uC shows that the active species is robust. Furthermore, the
active catalyst species must be soluble in water, since decantation
from precipitated salts after the 6th cycle had little effect on the
yield of the subsequent cycle. Therefore the involvement of a
heterogeneous rhodium catalyst, either as the active species or a
reservoir for soluble Rh nanoparticles, can likely be excluded in
this system. Ueda and Miyaura6 have proposed that the active
species in his t-Bu3P–Rh(I) catalyst system is a coordinatively
4486 | Chem. Commun., 2005, 4484–4486
This journal is ß The Royal Society of Chemistry 2005