Microwave accelerated Suzuki coupling of chloro-aryl phosphine-oxides: A method for introducing diversity into phosphine ligands

Article abstract of DOI:10.1016/j.molcata.2008.01.001

There is a demand for methodology that can rapidly produce families of catalysts for catalyst-structure-performance studies. The results reported here show microwave accelerated Suzuki coupling of a range of boronic acids with chloro-aryl phosphine-oxides can give high yields of arylated phosphine oxides. Chloro-aryl phosphine-oxide substrates have been found to be unreactive in previous studies, but in these examples, the combination of a highly active (but short lived) catalyst and instantaneous microwave heating at 140 °C allows the reactions to proceed. These cross-coupling reactions are being used as the basis for a rapid synthesis of phosphine ligands (readily available by microwave-assisted reduction of the phosphine oxides).

Full text of DOI:10.1016/j.molcata.2008.01.001

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
Microwave accelerated Suzuki coupling of chloro-aryl phosphine-oxides:
A method for introducing diversity into phosphine ligands
Karen Damiana, Matthew L. Clarkea,∗, Christopher J. Cobleyb
a School of Chemistry, University of St Andrews, EaStCHEM, St Andrews, Fife KY16 9ST, UK
b Dowpharma, Chirotech Technology Limited, A subsidiary of The Dow Chemical Company, 162 Cambridge Science Park,
Milton Road, Cambridge CB4 0GH, UK
Received 29 October 2007; received in revised form 20 December 2007; accepted 4 January 2008
Available online 11 January 2008
Abstract
There is a demand for methodology that can rapidly produce families of catalysts for catalyst–structure–performance studies. The results
reported here show microwave accelerated Suzuki coupling of a range of boronic acids with chloro-aryl phosphine-oxides can give high yields of
arylated phosphine oxides. Chloro-aryl phosphine-oxide substrates have been found to be unreactive in previous studies, but in these examples,
the combination of a highly active (but short lived) catalyst and instantaneous microwave heating at 140 ◦C allows the reactions to proceed. These
cross-coupling reactions are being used as the basis for a rapid synthesis of phosphine ligands (readily available by microwave-assisted reduction
of the phosphine oxides).
© 2008 Elsevier B.V. All rights reserved.
Keywords: Palladium; Cross-coupling; Microwave; Phosphine ligands; Diversity orientated synthesis
1. Introduction
will undergo palladium catalysed Suzuki, Heck, carbonylation
and amination reactions to deliver new phosphine oxides (and
subsequently phosphines) [3]. Unfortunately, we could not use
the Xiao protocols directly as a general method for the rapid
generation of phosphine libraries, since the use of a bromo-aryl
functionality severely limits the types of phosphine structure
that can be readily accessed. Suzuki coupling of chloro-aryl
phosphine-oxides, on the other hand, could form the basis of a
rapid method to make families of phosphines since they are sta-
ble to the manipulations needed to prepare a much larger range
of mono- and di-phosphines (Grignard chemistry, Pd-catalysed
P–C bond forming reactions, lithiation etc.). In this paper, we
report the successful use of microwave-assisted Suzuki cross-
coupling on chloro-aryl phosphine-oxides substrates, and an
example demonstrating microwave-assisted P O bond reduc-
tion to provide the corresponding phosphines.
Microwave heating has made a significant impact in
medicinal chemistry laboratories, since it allows the rapid
synthesis of structurally related families of compounds for
drug structure-activity studies [1]. Palladium catalysed cross-
coupling reactions between aryl bromides and nucleophiles
using microwave heating have been heavily utilised. Reaction
times are cut down from ∼1 day to 30 min or much less. Due to
the subtlety and unpredictability of catalyst–structure–activity
relationships in homogeneous catalysis, there is a similar
demand for rapid syntheses of structurally related libraries of
catalysts. There has been considerable worldwide research effort
in this area [2], since rapid identification of effective catalysts is
essential for their use in the pharmaceutical industry.
We were not aware of any reports in which microwave accel-
erated cross-coupling had been used as the key step to prepare
families of phosphine ligands. We were intrigued by the very
interesting reports of Xiao and co-workers who demonstrated
that, under conventional heating, bromo-aryl phosphine-oxides
2. Experimental
2.1. General information
∗
All chemicals and solvents are standard laboratory grade,
obtained from commercial sources and were used as received.
Corresponding author. Tel.: +44 1334 463850; fax: +44 1334 463808.
E-mail address: mc28@st-andrews.ac.uk (M.L. Clarke).
1381-1169/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2008.01.001

K. Damian et al. / Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
47
Table 1
Dry, degassed solvents were used for reactions unless otherwise
Screening of catalysts for cross-coupling reaction shown in Scheme 1a
indicated. Normal grade solvents were used for chromatography.
Solvents were removed by rotary evaporation on a Heidolph
labrota 4000. Flash column chromatography (eluents given in
brackets) were performed using Davisil silica gel Fluorochem
60 A, particle size 35–70 ␮m. All NMR spectra were recorded
on Bruker Avance 300 instruments.
Entry
Precatalyst
1 (%)
2 (%)
3 (%)
4 (%)
1
2
3
4
5
Pd(OAc)2
100
15
0
0
0
0
49
0
>95
>95b
0
15
92
<5
0
0
21
8
0
0
7
6
5
8
˚
Microwave reactions were carried out in a Biotage® Initiator
using 5 ml heavy-walled reactor vials equipped with an air tight
seal. The temperature is measured by an infrared temperature
probe that measures the temperature on the surface of the vial.
The pressure is measured by direct reading of the deflection of
the septa on the vial using a load cell behind the inner part of
the cavity lid.
a
Reactions were conducted in MeCN with 10 equivalents of PhB(OH)2 and
CsF as base; 5 mol% catalyst at 140 ◦C in a Biotage Discover microwave oven
for 30 min. Values refer to yield determined by 31P NMR spectroscopy directly
from reaction mixture. The relative NMR responses (1:1:1:1), and the relative
positions of the peaks were assigned by spiking experiments.
b
Product isolated in 67% yield after column chromatography.
Infrared spectra were recorded on a PerkinElmer Paragon
1000 Spectrum GX FT-IR system. Compounds were analysed
as KBr disks. Mass spectra were recorded on a Waters Micro-
mass GCT (Time of flight) fitted with lockspray for accurate
mass (ESI) or GCT (CI) instruments. Only major peaks are
reported and intensities are quoted as percentages of the base
peaks. The ligands/precatalysts 5–7 are commercially available.
The ligand and precatalyst 8 were prepared using a method
previously developed by us [6]. Brief details are reproduced
here for convenience. N-Methylpiperazine (0.48 g; 4.8 mmol)
and triethylamine (0.48 g; 6.6 mmol) were placed in a schlenk
tube equipped with a rubber septum under a dry nitrogen atmo-
sphere. Dry diethyl ether (20 ml) was then added. The solution
was cooled down to 0 ◦C and dicyclohexylchlorophosphine (1 g;
4.3 mmol) was added dropwise. The solution was then stirred
at room temperature overnight, after which time phosphorus
NMR demonstrated a quantitative conversion to the desired lig-
to 90/10); (220 mg; 0.43 mmol; 67%). 1H NMR (300 MHz,
CDCl3) δ 7.29–7.42 (m, 9H, ArH), 7.53–7.56 (m, 6H, ArH),
7.62–7.56 (m, 12H, ArH); 13C NMR (75.5 MHz, CDCl3) δ
127.6 (ArC), 127.7 (ArC), 127.8 (ArC), 128.6 (ArC), 129.4
(ArC), 130.9 (ArC), 132.3 (ArC), 133.1 (d, 2JC–P 10.2 Hz, ArC),
140.3 (ArC), 145.2 (d, JC–P 2.6 Hz, ArC); 31P{ H} NMR
(121.5 MHz) δ 29.7; IR (KBr) νmax: 1596 (m), 1544 (w), 1480
(m), 1443 (w), 1390 (m), 1264 (w), 1183 (s), 1117 (s), 1007
(m); LRMS (ES+) m/z 507.2 ((M+H)+, 100%); HMRS (ES+):
m/z calcd for C36H28OP: 507.1878, found: 507.1889.
4
1
2.3. Example procedure for Suzuki coupling on substrates
9a and 9b
To a microwave vial was added 9a (0.1 g; 0.29 mmol),
PhB(OH)2 (0.2 g; 1.7 mmol), CsF (0.5 g; 3.4 mmol) and catalyst
(8) (7 mg; 0.0014 mmol). The air was displaced with nitro-
gen and dry MeCN added (3 ml). The reaction was heated
in a microwave for 30 min at 140 ◦C. This was repeated in
further 5 vials, in order to obtain larger samples for characteri-
sation. The reaction mixtures were then combined, water was
added and the organic layer extracted with dichloromethane
(3 ml × 50 ml), dried (Na2SO4) and concentrated. The prod-
uct (10a) was isolated by column chromatography (elution
mixture: ethyl acetate/hexane, from 30/70 to 90/10) (454 mg;
1.05 mmol; 62%) 1H NMR (300 MHz, CDCl3) δ 7.22–7.49 (m,
1
and (31P{ H} NMR (121.5 MHz, C6D6) δ 76.12). If required,
the ligand can be isolated by Schenk filtration and removal of
volatiles and stored for long periods of time (many months)
under a nitrogen atmosphere. To prepare air stable precatalyst,
8, 10 ml of the ligand solution were transferred to a Schlenk flask
containing allylpalladium chloride dimer (0.2 g; 0.0014 mmol)
and the resulting solution stirred for 3 h. The yellow precipitate
that formed was filtered and washed with ether. (31P{ H} NMR
1
(121.5 MHz; CDCl3) δ 96.87).
1
16H, ArH), 7.63–7.69 (m, 4H, ArH), 0.79 (d, JH–H 12.1 Hz,
2.2. Example procedure for Suzuki coupling reactions
2H, ArH), 7.87 (s, 1H, ArH); 13C NMR (75.5 MHz, CDCl3) δ
127.7 (ArC), 128.4 (ArC), 129.1 (d, 2JC–P 12.0 Hz, ArC), 129.4
(ArC), 129.9 (ArC), 130.0 (ArC), 132.2 (ArC), 132.5 (ArC),
132.6 (ArC), 133.6 (ArC), 140.4 (ArC), 142.5 (d, 1JC–P 12.4 Hz,
The triarylphosphine oxide (1) (0.15 g; 0.4 mmol), together
with the precatalyst (2.5 mol%), ArB(OH)2 (0.38 g; 3.1 mmol)
and base (10 eq.) indicated in Table 1 were dissolved in the
solvent (5 ml), placed in a microwave vial, and sealed under
nitrogen before heating in the microwave. The conversion val-
ArC); 31P{ H} NMR (121.5 MHz) δ 30.57; IR (KBr) νmax: 1591
1
(w), 1572 (w), 1498 (w), 1452 (w), 1412 (m), 1438 (m), 1190 (s),
1115 (s), 1135 (s), 1023 (w); LRMS (ES+) m/z 453.2 ((M+Na)+,
100%); HMRS (ES+): m/z calcd for C30H23ONaP: 453.1384,
found: 453.1384.
1
ues shown in Table 1 were calculated using 31P{ H} NMR
peak identities confirmed by spiking experiments, and rela-
tive responses (1:1:1:1) confirmed by comparing accurately
measured authentic samples of starting material and products.
(121.4 MHz; C6D6): (2) δP 29.7; (3) δP 29.8; (4) δP 29.9; starting
material (1) δP 29.2.
All the other products were prepared following the proce-
dure described above, using the quantities of CsF and ArB(OH)2
indicated in Scheme 2.
(10b) Reaction run on 200 mg of starting material (4 × 50),
(234 mg; 0.48 mmol; 83%) 1H NMR (300 MHz, CDCl3)
δ 3.64 (s, 6H, 2×OCH3), 6.84–6.94 (m, 4H), 7.16–7.28
(m, 10H, ArH), 7.35–7.79 (m, 7H, ArH); 13C NMR
The reaction was also run on 250 mg of starting material
(5 mg × 50 mg) and the product (2) was isolated by column chro-
matography (elution mixture: ethyl acetate/hexane, from 30/70

48
K. Damian et al. / Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
133.0 (ArC), 133.4 (d, 3JC–P 2.6 Hz, ArC), 133.6 (ArC), 133.8
(75.5 MHz, CDCl3) δ 55.5 (OCH3), 111.2 (ArC), 120.9
(ArC), 128.4 (d, JC–P 12.0 Hz, ArC), 129.1 (ArC), 129.5
2
1
(ArC), 139.0 (d, JC–P 12.7 Hz, ArC), 156.7 (ArC); 31P NMR
1
{ H}(121.5 MHz) δ 30.54; IR (CDCl3) νmax (cm−1): 1898 (w),
(ArC), 130.7 (ArC), 130.9 (ArC), 131.7 (ArC), 131.8 (ArC),
2
132.3 (d, JC–P 9.8 Hz, ArC), 133.6 (ArC), 134.2 (ArC),
1820 (w), 1601 (w), 1497 (m), 1465 (m), 1437 (s), 1398 (m),
1261 (m), 1239 (m), 1182 (s), 1119 (s), 1058 (w), 1027 (m);
LRMS (ES+) m/z 407.1 ((M+Na)+, 100%); HMRS (ES+): m/z
calcd for C25H21O2NaP: 407.1177, found: 407.1170.
1
1
138.4 (d, JC–P 13.1 Hz, ArC), 156.4 (ArC); 31P{ H} NMR
(121.5 MHz) δ 30.56; IR (KBr) νmax: 1598 (w), 1493 (m),
1410 (m), 1278 (w), 1247 (w), 1247 (m), 1192 (s), 1117
(s), 1022 (m); LRMS (ES+) m/z 513.2 ((M+Na)+, 100%);
HMRS (ES+): m/z calcd for C32H27O3NaP: 513.1596, found:
513.1394.
(10g) Reaction run on 300 mg (3 × 100 mg), (214 mg;
1
0.56 mmol; 58%). H NMR (300 MHz, CDCl3) δ 3.70 (s, 3H,
1
OCH3), 6.83 (d, JH–H 8.7 Hz, 2H, ArH), 7.33–7.45 (m, 10H,
1
(10c) Reaction run on 400 mg of starting material (4 × 100),
ArH), 7.58–7.64 (m, 5H, ArH), 7.84 (d, JH–H 12.5 Hz, 1H,
(444 mg; 0.90 mmol; 79%) 1H NMR (300 MHz, CDCl3) δ 3.73
ArH); 13C NMR (75.5 MHz, CDCl3) δ 55.7 (OCH3), 114.7
1
2
(s, 6H, 2xOCH3), 6.87 (d, JH–H 8.6 Hz, 4H, ArH), 7.36–7.48
(ArC), 128.7 (ArC), 129.0 (d, JC–P 12.1 Hz, ArC), 129.3 (d,
2JC–P 12.8 Hz, ArC), 130.5 (ArC), 130.6 (ArC), 130.7 (ArC),
132.2 (ArC), 132.4 (ArC), 132.6 (ArC), 132.8 (ArC), 133.9 (d,
(m, 10H, ArH), 7.63–7.72 (m, 6H, ArH), 7.79 (s, 1H, ArH);
13C NMR (75.5 MHz, CDCl3) δ 55.4 (OCH3), 114.4 (ArC),
128.4 (ArC), 128.5 (ArC), 128.7 (ArC), 132.1 (ArC), 132.2
1
1JC–P 37.9 Hz, ArC), 141.5 (d, JC–P 11.98 Hz, ArC), 160.0
1
(ArC); 31P{ H} NMR (121.5 MHz) δ 30.53; IR (KBr) νmax
:
1
(ArC), 132.6 (ArC), 133.1 (d, JC–P 26.6 Hz, ArC), 134.3
1
1
(ArC), 141.6 (d, JC–P 12.6 Hz, ArC), 159.6 (ArC); 31P{ H}
NMR (121.5 MHz) δ 30.85; IR (KBr) νmax: 1608 (m), 1513
(s), 1437 (m), 1288 (w), 1254 (s), 1181 (s), 1117 (m), 1118
(m), 1029 (w); LRMS (ES+) m/z 513.2 ((M+Na)+, 100%);
HMRS (ES+): m/z calcd for C32H27O3NaP: 513.1596, found:
513.1600.
1602 (s), 1570 (w), 1513 (w), 1412 (m), 1339 (s), 1310 (s),
1245 (s), 1171 (s), 1108 (w), 1025 (m); LRMS (ES+) m/z 407.1
((M+Na)+, 100%); HMRS (ES+): m/z calcd for C25H21O2NaP:
407.1177, found: 407.1172.
(10h) Reaction run on 100 mg of starting material
1
(2 × 50 mg), (139 mg; 0.31 mmol; 98%). H NMR (300 MHz,
(10d) Reaction run on 300 mg of starting material
CDCl3) δ 6.95–7.07 (m, 5H, ArH), 7.24–7.30 (m, 2H, ArH),
7.36–7.50 (10H, ArH), 7.59–7.66 (m, 5H, ArH), 7.86 (d, 1JH–H
12.6 Hz, 1H, ArH);13CNMR(75.5 MHz, CDCl3)δ117.9(ArC),
118.1 (ArC), 122.5 (ArC), 127.5 (d, 3JC–P 2.3 Hz, ArC), 127.6
(ArC), 127.9 (d, 2JC–P 12.7 Hz, ArC), 128.8 (ArC), 129.2 (ArC),
129.3 (ArC), 129.5 (ArC), 129.6 (ArC), 131.0 (ArC), 131.1
(ArC), 130.7 (ArC), 131.4 (ArC), 132.1 (ArC), 132.8 (ArC),
1
(6 × 50 mg), (400 mg; 0.65 mmol; 75%) H NMR (300 MHz,
CDCl3) δ 6.95–7.07 (m, 10H, ArH), 7.25–7.30 (m, 4H, ArH),
7.38–7.51 (m, 10H, ArH), 6.64–7.83 (m, 7H, ArH); 13C NMR
2
(75.5 MHz, CDCl3) δ 118.0 (d, JC–P 13.5 Hz, ArC), 122.6
(ArC), 127.5 (ArC), 127.6 (ArC), 127.7 (ArC), 127.8 (ArC),
127.9 (ArC), 128.8 (ArC), 130.5 (ArC), 131.0 (ArC), 131.1
(ArC), 132.0 (d, 2JC–P 8.7 Hz, ArC), 133.4 (ArC), 133.8 (ArC),
133.8 (ArC), 139.8 (d, 1JC–P 11.9 Hz, ArC), 155.7 (ArC), 156.4
1
(ArC); 31P{ H} NMR (121.5 MHz) δ 30.50; IR (KBr) νmax
:
1
140.4 (d, JC–P 12.6 Hz, ArC), 155.7 (ArC), 156.5 (ArC);
1
31P{ H} NMR (121.5 MHz) δ 30.74; IR (KBr) νmax (cm−1):
1589 (w), 1509 (w), 1489 (m), 1436 (w), 1239 (m), 1193 (w),
1118 (m); LRMS (ES+) m/z 447.2 ((M+H)+, 100%); HMRS
(ES+): m/z calcd for C30H23O2NaP: 469.1333, found: 469.1320.
1588 (m), 1508 (m), 1488 (s), 1437 (w), 1238 (s), 1195 (w),
1119 (w); LRMS (ES+) m/z 637.3 ((M+Na)+, 100%); HMRS
(ES+): m/z calcd for C42H32O3P: 615.2089, found: 615.2092.
(10e) Reaction run on 200 mg of starting material
2.4. Procedure of microwave-assisted reduction of (10c)
1
(4 × 50 mg), (179 mg; 0.51 mmol; 79%). H NMR (300 MHz,
CDCl3) δ 6.96–7.91 (m, 19H, ArH); 13C NMR (75.5 MHz,
CDCl3) δ 127.23 (ArC), 127.9 (ArC), 128.6 (d, 2JC–P 12.1 Hz,
ArC), 128.8 (ArC), 128.9 (ArC), 130.6 (ArC), 130.7 (d,
Phosphine oxide (10c) (65 mg; 0.13 mmol) was placed in
a microwave vial and sealed under an inert atmosphere and
dry MeCN added, followed by dry Et3N (131.5 mg; 1.3 mmol)
and HSiCl3 (88 mg; 0.65 mmol). The reaction was heated in a
microwave at 145 ◦C for 10 min. MeCN was removed under
reduced pressure and the crude residue was dissolved in dry
DCM and passed through a short pad of silica under nitrogen.
The solvent was removed under reduced pressure and the NMR
sample was prepared using dried and degassed CDCl3 (51 mg;
0.11 mmol; 83%).
2
3JC–P 2.9 Hz, ArC), 130.8 (d, JC–P 10.13 Hz, ArC), 131.8
(ArC), 132.1 (ArC), 132.2 (ArC), 133.2 (ArC), 133.9 (ArC),
1
1
140.0 (ArC), 141.6 (d, JC–P 11.9 Hz, ArC); 31P{ H} NMR
(121.5 MHz) δ 30.46; IR (KBr) νmax: 1471 (w), 1438 (m), 1398
(w), 1185 (s), 1120 (s); LRMS (ES+) m/z 377.1 ((M+Na)+,
100%); HMRS (ES+): m/z calcd for C24H19ONaP: 377.1071,
found: 377.1067.
(10f) Reaction run on 100 mg of starting material
(11) 1H NMR (300 MHz, CDCl3) δ 3.75 (s, 6H, 2×OCH3),
6.85–6.88 (m, 4H, ArH), 7.25–7.41 (m, 15H, ArH), 7.61 (t, 1H,
ArH); 13C NMR (75.5 MHz, CDCl3) δ 55.4 (OCH3), 114.25
(ArC), 125.7 (ArC), 128.6 (d, 2JC–P 6.9 Hz, ArC), 130.3 (ArC),
130.5 (ArC), 133.3 (ArC), 133.7 (ArC), 133.9 (ArC), 137.0 (d,
1
(2 × 50 mg), (104 mg; 0.27 mmol; 85%). H NMR (300 MHz,
CDCl3) δ 3.63 (s, 3H, OCH3), 6.84–6.94 (m, 2H, ArH),
7.18–7.26 (m, 2H, ArH), 7.34–7.48 (m, 7H, ArH), 7.56–7.67
(m, 6H, ArH), 7.73–7.77 (m, 1H, ArH); 13C NMR (75.5 MHz,
CDCl3) δ 55.8 (OCH3), 111.6 (ArC), 121.4 (ArC), 128.6 (ArC),
1
1JC–P 10.1 Hz, ArC), 138.2 (d, JC–P 10.7 Hz, ArC), 141.3 (d,
128.9 (d, JC–P 12.2 Hz, ArC), 129.6 (ArC), 130.9 (d, JC–P
1JC–P 7.6 Hz; ArC), 159.3 (ArC); 31P{ H} NMR (121.5 MHz) δ
2
2
1
9.6 Hz, ArC), 131.2 (ArC), 131.6 (d, JC–P 13.5 Hz, ArC),
3.23; LRMS (CI+) m/z 475.17 (M+H)+; HMRS (CI+): m/z calcd
2
3
2
132.3 (d, JC–P 2.5 Hz, ArC), 132.6 (d, JC–P 9.9 Hz, ArC),
for C32H28O2P: 475.1815, found: 475.1827.

K. Damian et al. / Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
49
3. Results and discussion
It is well known that aryl chlorides are significantly more
reluctant cross-coupling partners and it is only recently that a
fewrapidmicrowave-assistedarylchloridecross-couplingshave
been developed [4]. Xiao has previously noted that chloro-aryl
phosphine-oxides do not participate in Suzuki cross-coupling
under conventional conditions [3b]. In our hands, the best
resultforcross-couplingoftris(4-chlorophenyl)phosphineoxide
under conventional conditions made use of [Pd(OAc)2]/5 and
gave only a 43% isolated yield of the mono-arylated product,
4 after 48 h in refluxing toluene. No diarylated or triarylated
products were detected.
Fig. 1. Catalysts used in cross-coupling reactions.
Using tris(4-chlorophenyl)phosphine oxide as a model sub-
strate, a catalyst screen was carried out under microwave heating
1
conditions with product conversion analysed by 31P{ H} NMR
The conditions used here are clearly a highly suitable proce-
dure for cross-coupling some of the most sluggish aryl chloride
substrates, as judged by comparing the results shown in Table 1
and Scheme 2 with the failure of Buchwald’s remarkably active
(and other) catalysts under conventional heating. There has been
some debate as to the origin of the microwave effect on Suzuki
coupling reactions. The reactions reported here do not seem
possible with conventional heating. In our view, the microwave
effect is connected to the reaction vessel reaching an unusually
hightemperaturein<5 s, allowingthereactiontoproceedrapidly
before the relatively short lived Pd-mono-phosphine catalyst
decomposes.
Due to the subtlety of ligand effects in asymmetric catal-
ysis, it would be desirable to rapidly prepare chiral ligands
with novel P-aryl substituents from a readily available precursor
and this was the initial ambition in this project. However, the
synthesis of bis(3,5-dichlorophenyl)chloro-phosphine always
resulted in some difficulty to remove impurities that made
conversion into the desired chiral di(3,5-dichlorophenyl) sub-
stituted phosphines problematic. However, there may be other
spectroscopy [5]. Precatalysts 5–8 were selected since they are
readily available and have all shown activity in cross-coupling
of deactivated aryl chlorides [4i,6,7,8].
Selected reactions are shown in Table 1 (Fig. 1). The main
findings are that the use of the [Pd(dcymepip)(allyl)Cl] (8) and
Buchwald catalyst system 5 enabled the first effective cross-
coupling of chloro-aryl phosphine-oxides, producing triarylated
product 2 in good yield. Both these catalysts are significantly
moreactive thanthe palladacycle-based catalyst 7 or [Pd(OAc)2]
alone, which was ineffective. This is a relatively rare example of
a reaction that will only work if microwave heating is employed,
at least using the catalysts screened to date.
Catalyst 8 was subsequently selected for further syn-
thetic studies. After optimisation, it was pleasing to find
that this protocol could be used to couple a range of
boronic acids with 3-chlorophenyl(diphenyl)phosphine oxide
and the potentially more challenging double coupling of
3,5-dichlorophenyl(diphenyl)phosphine oxide in high yield
(Scheme 2).
Scheme 2. Microwave accelerated cross-coupling of chloro-aryl phosphine-
oxides: variation of aryl chloride and boronic acid. For coupling of 9a, 6
equivalents of Ar(B(OH)2) and 12 equivalents of CsF were used. For 9b, 3
equivalents of ArB(OH)2/6 equivalents CsF were applied. Yields are for isolated
purecompoundsafterchromatography:conversionsareconsiderablyhigher, and
often approach 100% (phosphine oxides do not elute well on silica).
Scheme 1. Reaction used for optimisation of the procedure.

50
K. Damian et al. / Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
References
[1] (a) M. Larhed, C. Moberg, A. Hallberg, Acc. Chem. Res. 35 (2002) 717;
(b) C.O. Kappe, Angew. Chem. Int. Ed. 43 (2004) 6250;
(c) P. Lidstro¨m, J.P. Tierney, Microwave-Assisted Organic Synthesis, Black-
well, Oxford, UK, 2005;
(d) P. Lidstrom, J.P. Tierney, B. Wathey, J. Westman, Tetrahedron 57 (2001)
9225.
[2] A variety of approaches towards high-throughput catalysis synthesis are
being explored;
(a) C. Gennari, U. Piarulli, Chem. Rev. 103 (2003) 3071;
(b) M.T. Reetz, Angew. Chem. Int. Ed. 40 (2001) 284;
M.T. Reetz, Angew. Chem. 113 (2001) 292;
(c) M.T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew. Chem. Int. Ed.
42 (2003) 790;
M.T. Reetz, T. Sell, A. Meiswinkel, G. Mehler, Angew. Chem. 115 (2003)
814;
Scheme 3. Microwave accelerated reduction of one of the phosphine oxides.
(d) C. Bolm, C. Tanyeli, A. Grenz, C.L. Dinter, Adv. Synth. Catal. 344 (2002)
649;
(e) S.R. Gilbertson, S.E. Collibee, A. Agarkov, J. Am. Chem. Soc. 122(2000)
6522;
(f) J.R. Porter, W.G. Wirschun, K.W. Kuntz, M.L. Snapper, A.H. Hoveyda,
J. Am. Chem. Soc. 122 (2000) 2657;
(g) X.-B. Jiang, L. Lefort, P. Elspeth-Goudriaan, A.H.M. DeVries, P.W.N.M.
Van Leeuwen, J.G. De Vries, J.N.H. Reek, Angew. Chem. Int. Ed. 45 (2006)
1223;
X.-B. Jiang, L. Lefort, P. Elspeth-Goudriaan, A.H.M. DeVries, P.W.N.M.
Van Leeuwen, J.G. De Vries, J.N.H. Reek, Angew. Chem. 118 (2006)
1629;
(h) M. Weis, C. Waloch, W. Seiche, B. Breit, J. Am. Chem. Soc. 128 (2006)
4188;
diphosphine systems that lend themselves to using this proto-
col.
In order to demonstrate that the procedure can produce
phosphines, the reduction of one of the oxides was inves-
tigated. Using the simple mono-phosphine oxide 10c, we
investigated the reduction from the crude reaction mixture
after the Suzuki coupling. However, this one-pot reaction did
not proceed, presumably due to some incompatibility of the
SiCl3H reducing agent with the boron waste in the Suzuki
reaction. However, the reduction does take place on purified
phosphine oxides, and moreover, P O reduction at 145 ◦C in
MeCN using microwave heating allows this reduction to take
place in much shorter times than are typically employed using
conventional heating [9]. Phosphine oxide 10c was reduced
to phosphine 11 in 83% yield using a microwave acceler-
ated P O reduction. (Scheme 3). The faster reaction time is
merely due to the ability to heat MeCN beyond its boiling
point.
(i) J.N.H. Reek, M. Roder, P.E. Goudriaan, P.C.J. Kamer, P.W.N.M. Van
Leeuwen, V.F. Slagt, J. Organomet. Chem. 690 (2005) 4505;
(j) J.G. deVries, L. Lefort, Chem. Eur. J. 12 (2006) 4722;
(k) M.L. Clarke, J.A. Fuentes, Angew. Chem. Int. Ed. 46 (2007) 930.
[3] (a) C. Baillie, W. Chen, J. Xiao, Tetrahedron Lett. 42 (2001) 9085;
(b) L. Xu, J. Mo, C. Baillie, J. Xiao, J. Organomet. Chem. 687 (2003)
301;
(c) C. Baillie, J. Xiao, Tetrahedron 60 (2004) 4159.
[4] For microwave accelerated cross-couplings of deactivated substrates, see;
(a) R.B. Bedford, C.P. Buts, T.E. Hurst, P. Lidstro¨m, Adv. Synth. Catal. 346
(2004) 1627;
4. Conclusions
(b) N.E. Leadbeater, M. Marco, Org. Lett. 4 (2002) 2973;
(c) Y. Wang, D.R. Sauer, Org. Lett. 6 (2004) 2793;
(d) M.L. Clarke, Adv. Synth. Catal. 347 (2005) 303;
(e) E. Ferrer Flegeau, M.E. Popkin, M.F. Greaney, Org. Lett. 8 (2006) 2495;
(f) E. Alacid, C. Na´jera, Adv. Synth. Catal. 348 (2006) 945;
(g) E. Alacid, C. Na´jera, Adv. Synth. Catal. 348 (2006) 2085;
(h) H. Gold, M. Lahred, P. Nilsson, Synlett (2005) 1596;
(i) M.L. Clarke, J.A. Fuentes, M.B. France, E.J. Milton, G.J. Roff, Bielstein
J. Org. Chem. 3 (2007), art. 18.
We have shown that a series of substrates that were thought
to be unreactive in Suzuki cross-coupling can indeed be coupled
in high yield using microwave heating and an appropriate cata-
lyst. In addition, we also report here an example of a microwave
accelerated P O reduction to complete the process. The over-
all process uses two microwave accelerated reactions that each
take 30 min or less to potentially produce a range of ligands
from readily available chloro-aryl phosphine oxide precursors.
The methodology should be quite advantageous in the syn-
thesis of a collection of simple 3,5-diarylphenyl substituted
phosphines. More structurally complex multidentate phosphine
libraries will require a convenient synthesis of the chloro-aryl
precursors in order to offer significant advantages over standard
methods.
[5] Phosphine oxides 1–4 each have a different 31P NMR chemical shift, as
revealed by spiking experiments. The arylated products and starting materi-
als also have similar response factors by NMR as expected.
[6] (a) M.L. Clarke, D.J. Cole-Hamilton, J.D. Woollins, J. Chem. Soc. Dalton
Trans. (2001) 2721;
(b) M.L. Clarke, in: S.M. Roberts, J. Xiao, T. Pickett, S. Miller (Eds.), Cat-
alysts for the Fine Chemicals Industry, vol. 3, Wiley, 2004;
(c) M.L. Clarke, A.M.Z. Slawin, J.D. Woollins, Polyhedron 22 (2003)
19.
[7] T.E. Barder, S.D. Walker, J.R. Martinelli, S.L. Buchwald, J. Am. Chem. Soc.
127 (2005) 4685, and ref’s therein.
[8] W.A. Herrman, C. Brossmer, C.P. Reisinger, T.H. Riermeier, K. Ofele, M.
Beller, Chem. Euro. J. 3 (1997) 1357.
Acknowledgements
[9] P O bond reductions typically take place in refluxing xylene or toluene over
extended periods of time. Other experiments carried out in these laboratories
show that these reactions can be quite fast for phosphines that are resistant to
air oxidation (and presumably have a low reduction potential for the oxide).
The authors would like to thank Dowpharma for funding and
the EPSRC for an equipment grant (HPLC) and the use of the
EPSRC National Mass Spectrometry Service.

K. Damian et al. / Journal of Molecular Catalysis A: Chemical 284 (2008) 46–51
51
Phosphine oxides with a higher reduction potential require long reaction
times and forcing conditions. The microwave method is likely to be more
useful for these latter phosphine oxides. For examples of P=O reduction
using conventional heating, see;
(b) T. Hayashi, S. Hirate, K. Kitayama, H. Tsuji, A. Torii, Y. Uozumi, J. Org.
Chem. 66 (2001) 1441;
(c) Y. Uozumi, A. Tanahashi, S. Lee, T. Hayashi, J. Org. Chem. 58 (1993)
(1945);
(a) E. Gorobets, B.M.M. Wheatley, J.M. Hopkins, R. McDonald, B.A. Keay,
Tetrahedron Lett. 46 (2005) 3843;
(d) Y. Uozumi, N. Suzuki, A. Ogiwara, T. Hayashi, Tetrahedron 50 (1994)
4293.