Organic Process Research & Development 2008, 12, 116–119
Sulfur Contamination Due to Quenching of Halogenation Reactions with Sodium
Thiosulfate: Resolution of Process Problems via Improved Quench Protocols
Yanqiao Xiang, Pierre-Yves Caron, Brett M. Lillie, and Rajappa Vaidyanathan*
DeVelopment Science and Technology, Pfizer Inc., Eastern Point Road, Groton, Connecticut 06340, U.S.A.
Abstract:
During efforts to develop the manufacturing process for 1,
it was found that the Migita coupling reaction of 2 with 3 was
capricious. While some batches of 6-iodoindazole (2) purchased
from external vendors underwent the desired coupling using 1
mol % catalyst loading, other batches required significantly
higher catalyst loadings (up to 4 mol %). Surprisingly, the
HPLC potencies and purity profiles of the good and offending
batches of 2 were comparable, and there was no discernible
difference between the batches by routine analytical methods
(1H NMR, GC, HPLC, ROI, and heavy metals assay). However,
it was found that the presence of an event at ca. 115 °C in the
differential scanning calorimetry (DSC) thermogram of 2
correlated with poor performance in the coupling reaction. All
of the offending batches of 2 that required a higher catalyst
loading exhibited a DSC event at 115 °C, whereas the good
batches did not.
Assuming that 2 is synthesized via a Sandmeyer reaction
(Scheme 2),3,4 it would be safe to surmise that a reducing agent
was used to quench the iodination reaction. The most common
quenching agents used for such reactions are sulfur-based
reductants, usually sodium thiosulfate or sodium bisulfite. On
the basis of known melting points, it was hypothesized that
the event at 115 °C could be due to the presence of either iodine
(mp ) 113 °C) or elemental sulfur (mp ) 112–120 °C
depending on the allotropic form). Small quantities of iodine
and elemental sulfur (ca. 0.2 wt %) were separately spiked into
the Migita coupling reaction with a good batch of 2 in order to
determine if one of these compounds led to stalled or slow
reactions. It was found that the reactions with iodine proceeded
to completion with the usual 1 mol % charge of catalyst,
whereas the reactions spiked with sulfur required the addition
of excess catalyst. An HPLC method was quickly developed
for the detection and quantitation of elemental sulfur in 2 (vide
infra). It was found that all of the offending lots contained >0.2
wt % sulfur, while the good lots contained <0.01 wt % sulfur.
Based on further experimentation, appropriate specifications for
sulfur levels in 2 were set in order to ensure robust performance
in the Migita coupling step.
Many metal-mediated cross-couplings involve the use of organic
halides, which are usually accessed by halogenation reactions.
Cross-couplings are sensitive to the presence of impurities in the
halides. This paper describes the origin of one such problematic
impurity (sulfur) during the synthesis of organic halides and
proposes alternatives to minimize or eliminate its formation.
Introduction
Metal-mediated cross-couplings have gained widespread use
in organic synthesis, particularly in the pharmaceutical industry.1
These reactions typically involve coupling of an organic halide
with the appropriate coupling partner in the presence of a
catalytic amount of metal and ligand. Since very small quantities
of catalysts are used in these types of reactions, the presence
of trace amounts of catalyst poisons in the reaction mixture can
have a deleterious effect on the reaction rate and outcome. It is
therefore important to ensure that the coupling partners in these
cross-coupling reactions are of acceptable purity.
Most of the organic halides used in cross-coupling reactions
are synthesized via a halogenation reaction. Generally, the
workup of halogenation reactions involves the use of a reducing
agent to quench the excess halogen source. In addition to
choosing an effective quench, it is important to pay due attention
to the byproduct of the quench, since small amounts of residues
could have an impact on subsequent transformations of the
halogenated compounds. This paper describes the identification
of a catalyst poison arising from a reductive quench and
proposes alternative approaches to minimize or eliminate its
formation.
Background
Axitinib (1), a potent inhibitor of the vascular endothelial
growth factor (VEGF), is currently under investigation for the
treatment of a variety of tumors. The synthetic route to this
active pharmaceutical ingredient (API) is depicted in Scheme
1. The first step in the synthesis is a Migita coupling reaction
between commercially available 6-iodoindazole (2) and 2-mer-
capto-N-methylbenzamide (3) in the presence of Pd2(dba)3 and
xantphos to afford 4. Iodination of 4 followed by a Heck
reaction with 2-vinylpyridine furnishes the desired API 1.2
Results and Discussion
It is well-known that elemental sulfur can poison transition
metal catalysts. It is well-documented but seldom recognized
that sodium thiosulfate decomposes to elemental sulfur under
* To
whom
correspondence
should
be
addressed.
Email:
rajappa.vaidyanathan@pfizer.com.
(1) Dugger, R.; Ragan, J.; Ripin, D. Org. Process Res. DeV. 2005, 9, 253–
258.
(3) Edwards, L.; Isaac, M.; Maddaford, S.; Slassi, A.; Xin, T. WO
2001005758, 2001.
(4) Foloppe, N.; Fisher, L. M.; Francis, G.; Howes, R.; Kierstan, P.; Potter,
A. Bioorg. Med. Chem. 2006, 14, 1792–1804.
(2) Flahive, E.; Ewaniki, B.; Yu, S.; Higginson, P. D.; Sach, N. W.; Inaki,
M. QSAR Comb. Sci., 2007, 26, 679–685.
116
•
Vol. 12, No. 1, 2008 / Organic Process Research & Development
10.1021/op700227p CCC: $40.75
2008 American Chemical Society
Published on Web 01/01/2008