Open air palladium catalyzed cyanation-the use of PMHS to protect from oxygen

Article abstract of DOI:10.1016/j.tetlet.2007.02.018

PMHS (polymethylhydrosiloxane) used in catalytic amounts has a remarkable ability to prevent catalyst poisoning by oxygen contamination during palladium catalyzed cyanation reactions. The procedure is applicable to a wide range of substrates and is so effective that it allows the reactions to be run fully open to the atmosphere.

Full text of DOI:10.1016/j.tetlet.2007.02.018

Tetrahedron Letters 48 (2007) 2555–2557
Open air palladium catalyzed cyanation—the use of PMHS
to protect from oxygen
Michael T. Martin,* Bing Liu, Bobby E. Cooley, Jr. and John F. Eaddy
GlaxoSmithKline, Chemical Development—Synthetic Chemistry, Five Moore Drive, PO Box 13398,
Research Triangle Park, North Carolina 27709, USA
Received 4 January 2007; revised 5 February 2007; accepted 6 February 2007
Available online 9 February 2007
Abstract—PMHS (polymethylhydrosiloxane) used in catalytic amounts has a remarkable ability to prevent catalyst poisoning by
oxygen contamination during palladium catalyzed cyanation reactions. The procedure is applicable to a wide range of substrates
and is so effective that it allows the reactions to be run fully open to the atmosphere.
Ó 2007 Elsevier Ltd. All rights reserved.
O
OMe
Conversion of aryl halides to aryl cyanides is a common
and useful synthetic tool. Palladium catalyzed cyanation
reactions are perhaps the most direct methods to pro-
duce this transformation. However, the palladium cata-
lyzed cyanation of aryl halides is an often capricious
reaction and lacks robustness due to catalyst poisoning
resulting in stalled reactions.1 One common cause of cat-
alyst poisoning is contamination by oxygen which has
given rise to many procedures that call for stringent de-
gassing of solvents or even continuous sparging.2 We
have discovered that a simple and economical addi-
tive—polymethylhydrosiloxane (PMHS)—is capable of
imparting the reaction with a remarkable protective
effect from catalyst poisoning due to oxygen. Addition
of as little as 1 wt % PMHS to these reactions allows
them to progress even when fully exposed to the atmo-
sphere and should therefore make the reaction robust
while removing the need for a stringent degassing proto-
col when used in a process run under a normal inert
atmosphere.
O
OMe
NC
Zn(CN)2
Br
Pd2(dba)3 / DPPF
NMP
Cl
Cl
Cl
Cl
2
1
GW713646X
GW713643X
ination from low-level amounts of oxygen either from
the solvent or introduced during setup and monitoring.
One solution we investigated was to rigorously sparge
the reaction solvent with Argon before and during the
process. While this achieved a more reliable reaction,
it was cumbersome and expensive on pilot plant scale.
To this end, we sought to identify a chemical method
to minimize the deleterious effects of oxygen poisoning.
There are various reports of adding chemical or even
electrochemical reductants to palladium catalyzed cya-
nations in order to protect from catalyst oxidation,
but none seemed to offer the level of success we were
hoping to achieve.3 We felt the ultimate success would
be to achieve complete conversion while running a reac-
tion fully exposed to air. Upon screening a variety of
additives as potential reductants (Ph3P, Zinc, (EtO)3P,
FeCl2, CuI and Et3SiH), we found that indeed the addi-
tion of triethylsilane provided complete conversion
while the reaction was open to the atmosphere. PMHS
was found to be equally effective in this reaction and be-
came the focus of this work based on the fact that it was
inexpensive, non-toxic and safe to handle as compared
During the course of our development of a large-scale
process for the palladium catalyzed cyanation of 1, an
intermediate in the synthesis of a potential non-nucleo-
side reverse transcriptase inhibitor, we were plagued
by unexpected reaction failures. The reaction was unre-
liable and often required multiple charges of catalyst
and ligand to afford complete conversion. Many of these
stalled reactions could be directly attributed to contam-
*
Corresponding author. Tel.: +1 919 483 8229; fax: +1 919 315 8735;
e-mail: mtm39695@gsk.com
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.02.018

2556
M. T. Martin et al. / Tetrahedron Letters 48 (2007) 2555–2557
to triethylsilane.4 It was discovered that addition of only
1 wt % PMHS (0.06 equiv based on 60 g/mol hydride)
was sufficient to fully progress the reaction to comple-
tion even when run fully exposed to the air.
Table 1. Open air cyanation of aryl bromides in the presence of PMHS
Starting materiala
Time
(h)
Temperature
Isolated
yield
(°C)
1.5
1.5
3
80
80
61b
94
91
97
89
63
CF3
O
Br
This finding was confirmed on kilogram scale when the
reaction was conducted without the use of nitrogen or
solvent degassing. To further reduce the cost of this
chemical stage on scale, we investigated the use of less
expensive Pd(OAc)2 as an alternative to the Pd2(dba)3
catalyst. With only 0.9 mol % Pd(OAc)2 and
0.12 mol % DPPF, we found that the reaction reliably
proceeded to completion in the presence of catalytic
PMHS. By eliminating the need for subsequent charges
of catalyst and ligand and utilizing the less expensive
Pd(OAc)2 system, we were able to significantly lower
the cost of this reaction on pilot plant scale.
Br
O
80
Br
MeO
O
1
120
120
120
N
Br
H
3
With these results in mind, we set out to test this proto-
col on a variety of representative aryl bromides to deter-
mine the scope of this process. Table 1 lists the aryl
bromides that were subjected to cyanation reactions
using this protocol. In all cases the reactions were run
completely open to the atmosphere and without degas-
sing the solvent. The reactions generally went to comple-
tion in 1–3 h. Some of the less reactive substrates
(electron rich aryl halides for instance) required elevated
temperatures to achieve complete conversion. The
PMHS does not appear to accelerate the reaction or
increase the reactivity of the catalyst system as com-
pared to the non-PMHS reaction conditions. As with
most palladium catalyzed processes, the catalyst/ligand
combination has a finite lifetime under the reaction con-
ditions and indeed this cyanation reaction can be viewed
as a competition between catalyst lifetime and reaction
progression. Given that these reactions were run open
to the atmosphere, it is impressive to see that they were
able to fully progress before the catalyst was sufficiently
compromised to stall the reaction. Note that for these
MeO
Br
Br
3
MeO
O
Br
2
3
80
80
78
95
Br
N
a All reactions were carried out on 1-g scale in round bottom flasks
fully open to air. The reactions were followed by HPLC and products
characterized by 1H NMR and 13C NMR.
b Product is volatile.
0.3
Addition of catalyst and ligand
representative cases
a higher loading of PMHS
(10 wt %) was utilized in order to insure that both acti-
vated and deactivated substrates would progress to com-
pletion. A loading of 1 wt % is sufficient to allow open
air reactions to go to completion in many cases and
therefore should be completely effective against spurious
air contamination in a typical reaction run under an
inert atmosphere.
0.25
57% by HPLC
0.2
0.15
0.1
89% by HPLC
94% by HPLC
In order to more fully demonstrate the effectiveness of
PMHS in this process, we utilized ReactIR5 to monitor
the cyanation of bromobenzotrifluoride run under both
inert conditions and fully exposed to air. We chose
Pd2(dba)3 as the catalyst source for these comparisons
as it is already in the correct oxidation state to initiate
the catalytic cycle. Figure 1 shows that the reaction per-
formed under inert conditions was complete within 1 h
whereas, the reaction exposed to air stalled after only
30 min. When the same reaction was performed exposed
to air but in the presence of 10 wt % PMHS, the reaction
progressed at essentially the same rate as the inert atmo-
sphere reaction. Note that the absorption of the PMHS
at 1073 cmÀ1 overlaps that of the aryl C–Br bond stretch
and therefore the relative concentration curves for the
0
10
20
30
40
50
60
70
time (min)
Open to Air w/PMHS
open to air w/out PMHS
inerted w/out PMHS
Figure 1.
innerted reaction and PMHS treated reactions do not
overlay precisely.
The extent of reaction completion was verified by
HPLC. A possible explanation could involve a silyl
hydride reduction of the oxidized palladium back to
palladium(0). However, in reactions that were purpo-

M. T. Martin et al. / Tetrahedron Letters 48 (2007) 2555–2557
2557
sefully stalled by exposure to air prior to addition of the
PMHS, the addition of PMHS failed to restart the
catalytic cycle. In these stalled reactions the addition
of either fresh Pd2(dba)3 or DPPF alone also failed to
restart the reactions, suggesting that both the catalyst
and ligand were irreversibly inactivated. The addition
of both fresh catalyst and ligand did restart the reaction
as expected. These observations demonstrate that the
PMHS is capable of maintaining the palladium in its
correct oxidation state and thus protecting the catalyst
from oxidative inactivation yet it is not capable of actu-
ally reactivating catalyst that has been deactivated.
cant advancement in achieving a higher level of
robustness for these reactions. The use of PMHS in con-
junction with other palladium catalyzed reactions
known to be highly air sensitive is currently under inves-
tigation in our labs.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2007.02.018.
In a typical procedure, 3-bromoacetophenone (1.0 g,
5.3 mmol), zinc cyanide (0.32 g, 2.7 mmol), NMP
(5 ml) and water (0.5 ml) are stirred in a one neck round
bottom flask open to the atmosphere. The contents are
heated to 80 °C and treated with PMHS (0.10 g) fol-
lowed by a slurry of Pd(OAc)2 (11.3 mg, 0.05 mmol)
and DPPF (36.2 mg, 0.0065 mmol) in NMP (0.5 ml).
The contents are heated at 80 °C for aproximately 2 h
and monitored for completion by HPLC. The reaction
is cooled and the contents placed on a silica gel column
and eluted with 50% heptane/ethyl acetate. The purified
product is isolated as an off white solid, (596 mg, 78%)
by evaporation of volatiles. All products were fully char-
acterized by H NMR, 13C NMR and HPLC.
References and notes
1. Anderson, B.; Bell, E.; Ginah, F.; Harn, K.; Pagh, L.;
Wepsiec, J. J. Org. Chem. 1998, 63, 8224–8228.
2. (a) Chidambaram, R. Tetrahedron Lett. 2004, 45, 1441–
1444; (b) Maligres, P.; Waters, M.; Fleitz, F.; Askin, D.
Tetrahedron Lett. 1999, 40, 8193–8195.
3. (a) Davison, J. B.; Peerce-Landers, P. J.; Jasinski, R. J.
J. Electrochem. Soc. 1983, 130, 1862; (b) Ramnauth, J.;
Bhardwaj, N.; Renton, P.; Rakhit, S.; Maddaford, S.
Synlett 2003, 2237–2239.
4. Lawrence, N.; Drew, M.; Bushell, S. J. Chem. Soc., Perkin
Trans. 1 1999, 3381–3391.
5. In situ IR measurements were taken using a Mettler-Toledo
ReactIRTM 4000 with DiCompTM Probe. The consumption
This discovery of using PMHS as a protectant in palla-
dium catalyzed cyanation reactions represents a signifi-
of aryl bromide was profiled at 1073 cmÀ1
.