Phosphine-functionalised polymer resins as Pd scavengers

Article abstract of DOI:10.1016/j.tetlet.2005.08.005

Three phosphine-functionalised polymer resins were used as scavengers of palladium catalysts from Buchwald-Hartwig aryl amination reactions. The purity of the products was assessed, and residual palladium analysed by ICP-AES. Scavenging efficiencies of up to >98.5% were demonstrated.

Full text of DOI:10.1016/j.tetlet.2005.08.005

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6911–6913
Phosphine-functionalised polymer resins as Pd scavengers
a
b,
*
´
Meritxell Guino and King Kuok (Mimi) Hii
aDepartment of Chemistry, KingÕs College London, Strand, London WC2R 2LS, UK
bDepartment of Chemistry, Imperial College London, South Kensington, London SW7 2AZ, UK
Received 13 June 2005; revised 25 July 2005; accepted 1 August 2005
Abstract—Three phosphine-functionalised polymer resins were used as scavengers of palladium catalysts from Buchwald–Hartwig
aryl amination reactions. The purity of the products was assessed, and residual palladium analysed by ICP-AES. Scavenging effi-
ciencies of up to >98.5% were demonstrated.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Palladium catalysis is a powerful tool in synthetic chem-
istry for C–C and C–X bond formations.1 In recent
years, significant advances have been made in the dis-
covery and design of highly active catalysts, which fall
into either one of the two categories: those that are ther-
mally stable and deliver extremely high turnovers, typi-
cally at temperatures of >100 ꢁC;2 or those that are
catalytically active under ambient temperatures.3 Natu-
rally, the latter systems are more desirable due to their
greater energy efficiency and better accommodation of
thermally sensitive moieties. Nonetheless, fairly high
catalytic loadings (typically 1–3 mol %) are usually re-
quired to provide acceptable rates.
H2-X
N
x = 0, 1
x = 1, 2
x = 2, 3
N
H
NH2
x
SH
OH
SH
N
N
S
OH
S
N
SH
4
5
Figure 1. Commercially available Pd scavenger resins.
The coordination of the homogeneous catalyst onto the
solid support is envisaged to be dependent on the nature
of the palladium species and the arresting moiety. Here-
in, we report the use of three commercially available
phosphine-functionalised polymer supports (PS-TPP,
PS-PPh2 and PS-PCy2, Fig. 2) as scavengers of Pd cata-
lysts from reaction mixtures.
As palladium catalysis gains greater popularity in fine
chemicals and pharmaceutical processes, the removal
and recovery of the precious metal from the products
have become important issues,4 due to the significant
cost of palladium, as well as the decreasing regulatory
limit of heavy metal residues in pharmacologically active
compounds.
PS
PS
PPh2
PS-TPP
Several palladium catalyst scavengers have been mar-
keted in recent years, including several polymer resins
(Fig. 1). These contain chelating polyamines 1–3,5 mac-
roporous resins functionalised with 2,4,6-trimercaptotri-
azine (MP-TMT) 4,6 or the macroporous resin beads
carrying a dithiothreitol functionality, 57 (Fig. 1). These
resins remove palladium efficiently from organic and
aqueous solutions.
PS-PPh2
PPh2
PS
P
PS-PCy2
Keywords: Palladium catalysis; Scavenger; Polymer resin; Phosphine.
*
Corresponding author. Tel.: +44 207594 1142; fax: +44 207594
5804; e-mail: mimi.hii@imperial.ac.uk
Figure 2. Phosphine-functionalised polymer resins.
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.08.005

6912
M. Guino´, Hii, K. K. / Tetrahedron Letters 46 (2005) 6911–6913
Ph
R
smoothly, with aniline or N-methylaniline, to furnish
N
H
Ph
+
R
the secondary and tertiary arylamine products 6a–e.
At the end of the reaction (5 h), the polymer resin was
added to the reaction mixture, and the resultant suspen-
sion was swirled gently overnight in an orbital shaker
(14 h). The polymer beads were recovered by filtration,
and the filtrate was subjected to a simple aqueous wash
to remove inorganic salts. After removal of solvents, the
composition of the arylamine was examined without
further purification.9
Br
Y
N
Y
Y = H, R = H, 6a
Pd2(dba)3 (0.5 mol%)
P(t-Bu)3 (0.8 mol%)
t-BuONa, toluene
RT, 5 h
Y = OMe, R = Me, 6b
Y = OMe, R = H, 6c
Y = CN, R = Me, 6d
Y = CN, R = Ph, 6e
Scheme 1. Buchwald–Hartwig aryl amination reaction.
In the present work, these scavengers were used to re-
cover residual catalyst from products of the aryl amina-
tion reaction (Scheme 1). As far as we are aware, the
attempted recovery of the catalysts from this system
has never been reported before using the other scaveng-
ers. This is thought to be particularly challenging, as the
system contains amine substrate and product that may
compete for binding to the metal centre. Hence, the
three polymer resins were chosen for the difference in
their Lewis basicity.
1H NMR spectroscopy was employed to confirm the
formation of the arylamine product and the presence
of any starting materials. Most of the time, the presence
of any palladium residues is clearly detectable as it gives
rise to coloured NMR samples, ranging from green,
orange to dark red-brown. 31P–{1H} NMR spectro-
scopy can be used to detect the presence of any phos-
phorus species in the solutions. Wherever applicable,
the purity of the isolated diarylamine products was
assessed by determining their melting points. Finally,
ICP-AES analyses were performed to determine the
amount of Pd residue and the efficiency of the recovery
process (Table 1).
Tri-tert-butylphosphine ligand was chosen for the
homogeneous reaction. Sterically bulky phosphine li-
gands are able to stabilise highly reactive coordinatively
unsaturated Pd species.3 Thus, we envisage that this will
also make it more susceptible to capture by the function-
alised supports.
As expected, high levels of Pd residue were present in the
products when the reactions were performed without the
scavenging procedure (entries 1 and 10).
The catalytic reactions were performed under ambient
conditions according to the established procedures.8
Using 1 mol % of Pd(0) with 0.8 mol % of P(t-Bu)3,
reactions of three aryl halides (4-bromobenzene, 4-
bromoanisole and 4-bromobenzonitrile) proceeded
Different equivalents of the polymer resins were em-
ployed as scavengers, to establish the optimal amount
required for the efficient removal of the palladium
Table 1. Removal efficiency of Pd from aryl amination reactions (Scheme 1)a
Entry
Y, R
Product
Scavengerb
Yield/%c
Mp/ꢁCd
dP/ppme
Residual Pdf
0.63%
1
2
3
4
5
6
7
8
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
H, H
6a
6a
6a
6a
6a
6a
6a
6a
6a
6b
6b
6b
6b
6b
6b
6b
6b
6b
6c
6d
6e
—
85
60
40
75
85
80—
80
8050n.d.
85
90—
100
100
80
75
50—
85
90—
70—
82
—
—
—
+63 s, +66 s
+66 s, +35 s
+66 s, +35 s
n.d.
PS-TPP (10)
PS-TPP (20)
PS-PPh2 (10)
PS-PPh2 (7.5)
PS-PPh2 (5)
PS-PCy2 (10)
PS-PCy2 (7.5–)51
PS-PCy2 (5)
—
PS-TPP (10)
PS-TPP (20)
PS-PPh2 (10)
PS-PPh2 (7.5)
PS-PPh2 (5)
PS-PCy2 (10)
PS-PCy2 (7.5)
PS-PCy2 (5)
PS-PPh2 (10)
PS-PPh2 (10)
PS-PPh2 (10)
50–51
50–51
+66 s, +63 s
50–51
<0.01%
n.d.
n.d.
9
48–49
+63 s, +66 s
—
—
—
—
+66 s, +63 s
—
n.d.
+66 s, +63 s
103–104
104–106
—
+66 s, +63 s
1.44%
+66 s, +63 s
+66 s, +63 s
n.d.
10OMe, Me
11
12
13
14
15
16
17
OMe, Me
OMe, Me
OMe, Me
OMe, Me
OMe, Me
OMe, Me
OMe, Me
OMe, Me
OMe, H
0.027%g
n.d.
n.d.
18
19
n.d.
n.d.
n.d.
0.05%
0.18%
0.10%
20CN, Me
21
82
84
CN, Ph
a The reactions were conducted with aryl halide (1 equiv), aromatic amine (1 equiv), Pd2(dba)3 (0.5 mol %), ligand (0.8 mol %) and NaO(t-Bu)
(1.5 equiv) in toluene, room temperature, 5 h.
b Value in parenthesis denotes equivalents used.
c Isolated yields, verified by 1H NMR.
d Mp (6a): 51–53 ꢁC, 6b and 6e are viscous oils at room temperature.
e 31P–{1H} NMR resonance(s), n.d. = not detected.
f Determined by ICP-AES analysis.
g 267 ppm in 50mg.

M. Guino´, Hii, K. K. / Tetrahedron Letters 46 (2005) 6911–6913
6913
catalyst. Triphenylphosphine-functionalised resin (PS-TPP)
was found to be an ineffective scavenger, even when it
was employed in large quantities (entries 2, 3, 11 and
12). The NMR samples were invariably coloured, and
the corresponding 31P NMR spectra indicated the pres-
ence of a large amount of phosphorus impurities. Fur-
thermore, the deployment of a large excess (20
equivalents per Pd) of the polymer also led to a decrease
in the recovered yield (entry 3), which was attributed to
the product being trapped within the polymer resins.
alytically active species in the homo- and heterogeneous
reactions.
In summary, we have demonstrated that certain phos-
phine-functionalised polymer supports, particularly the
inexpensive PS-PPh2, can be used to capture palla-
dium-catalysts effectively. The scavenging efficiency is
comparable to other commercially available Pd scaveng-
ers in less-challenging processes. We are currently inves-
tigating the use of these polymer resins as heterogeneous
supports in palladium catalysis, and the results will be
reported in due course.
In comparison, 7.5–10equiv of PS-PPh 2 (entries 4–6 and
13–15) and PS-PCy2 (entries 7–9 and 16–18) were found
to remove the catalysts effectively. Products were recov-
ered in high yields (between 80% and 90%) and good
product purity was obtained, as indicated by the agree-
ment between the mp of the product 6a with reported
values. Less than 0.01% of Pd was found in the product
6a (entry 4) and 0.027% in 6b (entry 13), corresponding
to scavenging efficiencies of >98.5% and 98%, respec-
tively (calculated by comparison to unscavenged reac-
tions—entries 1 and 10). These values were found to
be comparable to those reported for the ethylene
diamine-derived polymer scavengers 1–3 in the
Suzuki–Miyaura cross-coupling5a and that of a poly-
mer-supported chelate phosphine for the removal of
GrubbsÕ catalyst.10
Acknowledgements
The authors thank KingÕs College London and Imperial
College London for support and Johnson Matthey plc
for the provision of palladium salts.
References and notes
1. Handbook of Organopalladium Chemistry for Organic
Synthesis; Negishi, E., de Meijere, A., Eds.; Wiley-Inter-
science: New York, 2002.
2. Farina, V. Adv. Synth. Catal. 2004, 346, 1553–1582.
3. Christmann, U.; Vilar, R. Angew. Chem., Int. Ed. 2005, 44,
366–374.
4. Garrett, C. E.; Prasad, K. Adv. Synth. Catal. 2004, 346,
889–900.
5. (a) Urawa, Y.; Miyazawa, M.; Ozeki, N.; Ogura, K. Org.
Process Res. Dev. 2003, 7, 191–195; (b) Sigma–Aldrich
Chemfiles, Vol. 4, No. 1: Polymer-Supported Scavengers
and Reagents for Solution Phase Synthesis.
Attempts to reduce the amount of the scavengers to
5 equiv were clearly insufficient, as indicated by the pres-
ence of phosphorus residues in the recovered product
(entries 6, 9, 15 and 18).
However, when 4-bromobenzonitrile was employed as
the substrate, %Pd residue was particularly high (entries
20and 21), which is attributed to the complexation of
the nitrile moiety to Pd. Thus, the efficiency of Pd scav-
enging is dependent on the functional groups present on
the substrate.
6. (a) Ishihara, K.; Nakayama, M.; Kurihara, H.; Itoh, A.;
Haraguchi, H. Chem. Lett. 2000, 1218–1219; (b) Argonaut
technical note no. 515, 2003, MP-TMT.
7. Winder, R. Chem. Ind. 2004, 7.
8. Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy,
K. H.; Alcazar-Roman, L. M. J. Org. Chem. 1999, 64,
5575–5580.
PS-PPh2 and the Wang-aldehyde (PS-CHO) resins may
be used in tandem for the synthesis of secondary amines,
to remove Pd and excess amine, respectively (Scheme 2).
Thus, the reaction between bromobenzene and aniline
furnished diphenylamine in 75% yield, with high prod-
uct purity (NMR, GC–MS and mp). The procedure is
practically simple, and we envisage that it is amenable
to automated procedures for the construction of
libraries of arylamines.11
9. Typical reaction procedure: Aryl bromide (1.0mmol),
arylamine (1.0mmol), Pd 2(dba)3 (5.75 mg, 0.005 mmol),
P(tBu)3 (10wt % in hexane, 16 lL, 0.008 mmol) and
sodium tert-butoxide (150mg, 1.5 mmol) were weighed
into a round-bottomed flask or in a RadleyÕs carousel
reaction tube, under an atmosphere of N2. A stirrer bar
was added followed by anhydrous toluene (2 mL). The
resultant purple mixture was stirred at rt and monitored
by TLC. After complete consumption of starting materials
(typically about 5 h), polymer-bound phosphine resin was
added to the brown suspension, and the resultant mixture
was swirled in an orbital shaker at rt for 14 h (overnight).
The resin was recovered by filtration, and the filtrate was
diluted with diethyl ether and washed with H2O
(3 · 5 mL). After evaporation of the solvents, the product
was subjected to analyses (NMR, melting point and ICP-
AES) without further purification.
None of the recovered polymer beads (PS-PPh2 or PS-
PCy2) were found to be catalytically active at room tem-
perature, or at 80 ꢁC.12 As an excess of the polymer resin
was used in the scavenging process, this may be due to
the low Pd loading and the presence of a large effective
concentration of the phosphine moiety present in these
resin beads, which may prevent the formation of the cat-
10. Westhus, M.; Gonthier, E.; Brohm, D.; Breinbauer, R.
Tetrahedron Lett. 2004, 45, 3141–3142.
11. The recyclability of the Wang-aldehyde resin had been
1. Pd cat.
2. PS-PPh2
PS-CHO
PhBr
+
+PhNH2
Ph2NH
Ph2NH
´
demonstrated by one of us in an earlier paper: Guino, M.;
75% yield
´
Brule, E.; de Miguel, Y. R. J. Comb. Chem. 2003, 5, 161–
m.p. 50-51 oC
PhNH2
(1.3 equiv)
165.
12. No leaching of the Pd species into the homogeneous
solution was observed, even at high temperature.
Scheme 2. Successive scavenging.