Development of a scalable process for the crop protection agent isoclast

Article abstract of DOI:10.1021/acs.oprd.5b00007

A scalable process to the insecticide Isoclast manufactured by Dow AgroSciences LLC is described. The process involves the de novo construction of a fully elaborated pyridine sulfide using enamine-mediated cyclization followed by two efficient and inexpensive oxidations to introduce the sulfoximine.

Full text of DOI:10.1021/acs.oprd.5b00007

Article
pubs.acs.org/OPRD
Development of a Scalable Process for the Crop Protection Agent
Isoclast
Kim E. Arndt,‡ Douglas C. Bland,*,† Nicholas M. Irvine,‡ Stacey L. Powers,† Timothy P. Martin,‡
James Russell McConnell,† David E. Podhorez,† James M. Renga,‡ Ronald Ross,‡ Gary A. Roth,†
Brian D. Scherzer,† and Todd W. Toyzan†
†Dow Chemical Company, 1710 Building, Midland, Michigan 48674, United States
‡Process Chemistry, Dow AgroSciences LLC, 9330 Zionsville Road, Indianapolis, Indiana 46268, United States
ABSTRACT: A scalable process to the insecticide Isoclast manufactured by Dow AgroSciences LLC is described. The process
involves the de novo construction of a fully elaborated pyridine sulfide using enamine-mediated cyclization followed by two
efficient and inexpensive oxidations to introduce the sulfoximine.
ring construction to produce functionalized pyridine ring
systems exist in the literature.4 However, examples of 6-
(trifluoromethyl)pyridines prepared by de novo methods are
relatively rare, particularly those that can be readily function-
alized to produce useful intermediates for the preparation of
Isoclast.5 Several different approaches were explored, but the
most promising of those from a chemical complexity and
commercial feasibility standpoint was enamine-mediated
Michael additions.
INTRODUCTION
■
Isoclast Active (1, Scheme 1) is a fast acting sap-feeding
insecticide commercialized by Dow AgroSciences in 2013.1
Scheme 1. Discovery Chemistry Route to Isoclast Active (1)
This article will discuss our cyclization studies focused on
enamines derived from 3-(methylthio)butanal (6) to make
pyridine sulfide 10, followed by the development and scale-up
of conditions for the inexpensive introduction of the
sulfoximine to make Isoclast.
RESULTS AND DISCUSSION
Initially designed to control a broad spectrum of sap-feeding
insects, Isoclast has shown no signs of cross-resistance to
existing commercial products.2
■
A very attractive intermediate in the synthesis of Isoclast was
the trifluoromethylpyridine with a fully elaborated alkyl sulfide
side chain, 10 (Scheme 2). It was envisioned that this
compound could likely be obtained from commercially
available 3-(methylthio)butanal (6), a secondary amine (7), a
2-trifluoroacetyl vinyl ether (9), and an ammonia source.
The first synthesis of Isoclast was accomplished by a four-
step sequence starting from commercially available 5-
chloromethyl-2-(trifluoromethyl)pyridine (2, Scheme 1).1a
Chloride displacement with sodium methyl mercaptide gave
the sulfide (3) in good yield, followed by the conversion to the
sulfilimine (4) with cyanamide and iodobenzene diacetate. The
sulfilimine (4) was then oxidized to the corresponding
sulfoximine (5) with m-chloroperoxybenzoic acid (mCPBA)
in aqueous ethanol under basic conditions. Finally, low-
temperature methylation with iodomethane and potassium
hexamethyldisilazide gave Isoclast as a mixture of diaster-
eomers. The overall yield (four chemical steps) was a low 2.4%
after isolation by silica gel chromatography. Although this route
was optimized to provide gram quantities of Isoclast for initial
field trials, an aggressive commercialization plan included
rapidly escalating kilogram sample requirements.
Scheme 2. Enamine Michael/Cyclization Approach to
Pyridine Sulfide
Since the cost of 5-chloromethyl-2-(trifluoromethyl)pyridine
(2) was considered to be prohibitive for scale-up, an
investigation of alternative starting points for this material
was initiated using existing proprietary pyridine building
blocks.3 Concurrently, efforts also began surveying various de
novo pyridine synthesis routes. Numerous examples of de novo
Received: January 12, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Several cyclic and noncyclic enamines were compared in a
small/quick qualitative study in order to explore the effect of R1
and R2.
Scheme 3. Pyrrolidine Enamine/Michael addition/
Cyclization Approach to Pyridine Sulfide
The noncyclic secondary amines used in the study consisted
of dimethyl, diethyl, dipropyl, and di-isopropyl (Scheme 2). In
all cases, low yields of pyridine sulfide 10 were observed along
with significant quantities of 11. The cyclic secondary amines
included pyrrolidine, piperazine, piperidine, and morpholine. In
these cases, as suggested in the literature, pyrrolidine was
significantly superior.6 A third small qualitative study explored
R4 of enone 9 as a variable (Me, Et, i-Pr, n-Bu, i-Bu, t-Bu, and
cyclohexyl). For reasons not clear to us, ethyl and isopropyl
vinylogous esters were superior to the others. The use of
aldehyde 6, pyrrolidine, and 4-ethoxy-1,1,1-trifluorobut-3-en-2-
one (9, R4 = Et) initially afforded pyridine sulfide 10 in about
55% yield following silica gel chromatography. Additionally,
higher yields of pyridine sulfide 10 (increases of 5−10%) were
obtained when enamine 13 (derived from pyrrolidine) was
added to 4-ethoxy-1,1,1-trifluorobut-3-en-2-one (9, R4 = Et)
versus adding enone 9 to enamine 13 (Scheme 3).
Early sulfoximine analogue syntheses by the discovery
chemists introduced the sulfilimine functionality by first
reaction of a sulfide with sodium azide in H2SO4 to give the
corresponding unsubstituted sulfilimine, followed by alkylation
with cyanogen bromide. A later procedure was developed using
iodobenzene diacetate and cyanamide,7 which was used during
small sample preparations (Scheme 1, 3 to 4).
Perhaps the most troublesome aspect of the original
synthesis of Isoclast was the oxidation of the sulfilimine to
the corresponding sulfoximine. Although mCPBA and
potassium carbonate were used successfully to produce
numerous small-scale samples, we found chemical yields for
this reaction to be unpredictable upon scale-up. As with the
sulfilimine chemistry, we believed that cost, together with safety
concerns, made this approach unsuitable for large-scale sample
production. Dimethyldioxirane produced from oxone and
acetone under phase transfer conditions8 was used successfully
on small scale, but it was rejected for scale-up based on
economic and reactive chemistry concerns. Initial efforts were
focused on catalytic ruthenium-based sodium periodate
oxidations.9 This method proved to be more robust than the
mCPBA approach and was eventually scaled to multikilograms.
Initial Kilogram Scale-Up. As described above, the original
de novo synthesis of pyridine sulfide 10 utilized acetonitrile as
solvent for enamine formation, Michael addition to give diene
15, and subsequent ammonium acetate cyclization. The original
synthesis used about 2.5 equiv of pyrrolidine in acetonitrile.
Upon scale-up, toluene was the preferred solvent for enamine
formation because dehydration of this system was more
efficient and thus allowed for the use of about 1.1 equiv of
pyrrolidine (Scheme 3). The reaction was worked up by simple
filtration followed by rotary evaporative removal of the toluene.
The Michael addition and cyclization chemistry worked well in
acetonitrile and were not altered for this sample campaign. The
workup consisted of dilution of the reaction mixture with water
followed by extraction with hexane. Solvent removal followed
by vacuum distillation provided pure pyridine sulfide 10 in 65−
70% yield (based on aldehyde 6).
Several comments about the analytical procedures used to
monitor this chemistry are appropriate at this time. Enamine 13
was unstable toward GC analysis. Since it has a weak
chromaphore, HPLC with UV detection was difficult. The
most effective method for monitoring the conversion of
aldehyde 6 to enamine 13 was 1H NMR analysis. Since
analysis of enamine 13 was difficult via chromatographic
methods, again, the best way to monitor the reaction of 13 with
enone 14 was 1H NMR. Each of the vinyl protons of enone 14,
enamine 13, the two major isomers of intermediate 15, and
impurity 16 can be observed in the downfield region of the
NMR spectra. Since intermediate 15 has a strong chromaphore,
its disappearance can be monitored by normal-phase HPLC
(UV detection) during the course of the cyclization reaction.
Pyridine sulfide 10 was prepared in multiple runs using 1 L
reaction equipment (see Experimental Section). In addition, an
analytical standard of sulfide 10 was prepared for use in
quantitative analyses. Pyridine sulfide (10) process streams
were assayed via GC analysis using dibutyl phthalate as the
internal standard.
The iodobenzene diacetate procedure proved to be an
excellent way to form sulfilimine 17 (Scheme 4). The reactants
Scheme 4. Oxidation of Pyridine Sulfide 10 to Isoclast Active
(1)
were mixed together in acetonitrile at 10 °C, and the mixture
was allowed to warm to room temperature. Near 20 °C, a mild
exotherm occurred, which raised the temperature into the low
30s, and during this time, the mixture became a clear light
orange solution. LC analysis would indicate >98% area
conversion to the more polar sulfilimines 17.
Workup involved addition of water followed by three
extractions of the acetonitrile/water sulfilimine (17) solution
with hexanes to remove the majority of byproduct
iodobenzene. Solvent exchange into dichloromethane gave a
sulfilimine solution ready for the following oxidation. Addi-
tional LC analysis indicated two sulfilimine isomers were
present in a 30:70 area ratio. These isomers could be separated
and characterized (see Experimental Section). Although
providing a well-controlled and high-yielding synthesis of 17,
the use of iodobenzene diacetate presented the problems of
As had been observed previously, the major impurity
generated by this chemistry was enamine-one 16. This impurity
was generated during the enamine mediated Michael addition
to enone 14 and was easily removed by distillation of pyridine
sulfide 10.
B
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
using an expensive reagent as well as disposal or recycle of large
quantities of iodobenzene on a process scale.
of enone 14 with enamine 13 gave a much lower yield for the
cyclization step. It was assumed that some of the toluene must
be removed in order to azeotropically dry the toluene/13
solution and remove excess pyrrolidine prior to reaction with
enone 14.
Sulfilimine 17 was next oxidized to Isoclast using a catalytic
amount of RuCl3 with a full equivalent of NaIO4 in a CH2Cl2/
water two-phase system (Scheme 4). The sodium periodate was
mixed with water, followed by addition of ruthenium(III)
chloride. To this dark mixture was added the CH2Cl2/
sulfilimine (17) solution over about 2 h with rapid stirring.
The reaction was exothermic, with the temperature maintained
below 30 °C by control of the addition rate. At the onset of
addition, a vigorous off-gassing occurred, which subsided within
5 min. A black color formed on the exposed glass surface of the
reaction vessel and also up into the N2 bubbler. The cause of
this off-gassing and black color formation was not known;
however, it was possibly a result of in situ formation of volatile
ruthenium tetroxide. Over the course of the sulfilimine (17)
addition, the reaction mixture became black and very thick.
During some runs, the oxidation would appear to stall. In these
cases, the reaction could be coaxed to completion by the
addition of more RuCl3 and/or NaIO4.
Workup of this thick dark mixture needed to be performed
immediately upon completion of the oxidation. If the reaction
mixture was allowed to sit, then product 1 would start to
crystallize from the mixture, compounding problems with its
isolation. When the oxidation was complete, as determined by
LC analysis, more water was added to the mixture to affect
phase separation. This phase separation was hindered by a large
amount of black rag, and a flashlight had to be used for
interface determination. The rag/aqueous layer was re-extracted
with CH2Cl2. The organics were then combined, water was
added, and excess oxidant was destroyed by the addition of
aqueous sodium metabisulfite. The organics were concentrated,
and the gray paste-like solid recrystallized from hot 2-propanol
to give Isoclast as a gray solid, with yields generally in the upper
70% range across the two step oxidation procedure from
pyridine sulfide 10. In this manner, about 200 g of Isoclast
could be prepared. It should be noted that the grey color of the
product was caused by residual ruthenium. The ruthenium level
in the material was about 270 ppm, as determined via neutron
activation analysis. Although this oxidation of sulfilimine 17 to
the desired product, Isoclast, provided acceptable yield, the
processing issues encountered during workup, the iodine waste
generated, and the observations made during this oxidation
would make the use of RuCl3/NaIO4 a poor candidate for
further scale-up.
Second-Generation Process Development. Synthesis
of Pyridine Sulfide 10. During the course of our initial sample
campaign, it became evident that we could not meet the
required long-term cost of manufacturing utilizing the
chemistry described above. The original synthesis of pyridine
sulfide 10 worked well and required only minor modifications
in order to deliver a more streamlined process.
Our original preparation of enamine 13 used 1.1 equiv of
pyrrolidine, potassium carbonate, and toluene as the solvent.
Upon completion of the reaction, the potassium carbonate was
removed by filtration, and the toluene was removed by rotary
evaporation. For future sample campaigns, we demonstrated
that not all of the toluene needed to be removed prior to
forwarding enamine 13 to the Michael reaction with enone 14.
Concentration of the toluene/enamine mixture (ca. 35 °C/20
mmHg) to approximately 45−55 wt % 13 provided a solution
that could be used directly in the Michael reaction. One
experiment in which no toluene was removed prior to reaction
In the early process development stages, the Michael reaction
and subsequent cyclization were conducted in acetonitrile.
Subsequently, it was determined that toluene may be
substituted for acetonitrile with no deleterious effect on the
yield of 10. The use of toluene provided two major process
advantages. First, the toluene solution of enamine 13 may be
used directly in the following Michael reaction. Second, the
workup of the post cyclization reaction was simplified. In our
previous workup, the post cyclization acetonitrile solution of
pyridine sulfide 10 was diluted with water and extracted with
several portions of hexanes. The hexanes were then removed on
a rotary evaporator. The current workup consists of washing
the postcyclization toluene solution once with water followed
by removal of the toluene by rotary evaporation. The
concentrated crude sulfide is then purified by vacuum
distillation. The average yield from aldehyde 6 was 66%
following distillation with an average purity of 98 wt % (GC
internal standard method). The distillation bottoms were
composed of unidentified tars and impurity 16. Differential
scanning calorimetry (DSC) and accelerated rate calorimetry
(ARC) data on the distillation bottoms indicated a potential
exothermic decomposition starting as low as 170 °C. For this
reason, distillations were kept at an internal pot temperature of
less than 155 °C. The 66% yield over this three-step sequence
was noteworthy when one considers the number of bonds
being formed and broken while utilizing a minimal number of
physical operations.
Oxidation of Pyridine Sulfide 10 to Sulfilimine 17. With the
anticipation of further scale-up and eventual manufacture, an
alternative synthesis of sulfilimine 17 was desired. Two main
routes are commonly used to prepare sulfilimines,9 involving
either an oxidative imination of a sulfide or a nucleophilic
displacement of the corresponding sulfinimidoyl halide
(halosulfonium intermediate). There are few literature
examples7 for the formation of N-cyanosulfilimines from
sulfides. Discovery chemists introduced this functionality by
first reaction of a sulfide with sodium azide in H2SO4 to give
the unsubstituted sulfilimine, followed by alkylation with
cyanogen bromide. As described above, a later procedure was
developed using iodobenzene diacetate and cyanamide, which
was used during sample campaigns.
Both of these routes are examples of the oxidative imination
approach to sulfilimines, and they proceed through postulated
nitrene intermediates. The first route was unattractive by virtue
of potentially explosive sodium azide and the use of the highly
toxic alkylating agent cyanogen bromide. In the second route, a
postulated cyanonitrene intermediate was generated. Although
it is a well-behaved and high-yielding procedure, the high cost
and uncertain availability of iodobenzene diacetate suggested
the search for an alternative procedure. In addition, this
procedure presented the problematic disposal or recycle of a
large quantity of iodobenzene on a process scale. Swern
reported the generation and interception of this cyanonitrene
intermediate from N-sodio-N-chlorocyanamide to form N-
cyanosulfilimines.10 However, N-sodio-N-chlorocyanamide was
generated below −50 °C using t-butyl hypochlorite, and it did
not appear to be attractive from a manufacturing perspective.
Therefore, work was focused on the second approach to
C
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
sulfilimines, which involved nucleophilic displacement of a
generated sulfonium halide intermediate.
to the reaction mixture resulted in more sulfoxide (19)
formation.
A qualitative screening study was undertaken in the
laboratory that involved the generation of a sulfonium halide
intermediate 18 (Scheme 5) and its subsequent displacement
The promising results with calcium hypochlorite prompted a
reinvestigation using sodium hypochlorite, which would
eliminate insoluble calcium salt precipitation. Higher concen-
tration bleach solutions (10−12 wt %) are known to lose
strength over time, which might explain the problem using
commercial pool bleach. The reaction was subsequently
conducted using commercial bleach (6 wt %, 5.7% active via
titration). On a 1 mmol scale, cyanamide was dissolved in
acetonitrile and cooled to 0 °C, and the bleach was added. After
stirring cold for 15 min, sulfide 10 (dissolved in acetonitrile)
was added and allowed to warm to room temperature. LC
analysis indicated sulfide 10 was gone, with major sulfilimine 17
isomers present and only 6% (area LC) sulfoxide 19 isomers.
The mechanism and kinetics for the formation of sulfilimines
have been discussed7 and appear to be quite complicated. A
simplified mechanism is proposed in Scheme 5 and is based
upon analogy with the Mann−Pope reaction for the formation
of N-p-tosylsulfilimines. The chlorinating agent (bleach) can
react with the cyanamide to give chlorocyanamide or directly
with sulfide 10, leading to a chlorosulfonium intermediate 18.
Pyridine sulfide 10 can also react with the chlorocyanamide to
give the analogous chlorosulfonium intermediate 18. These
chlorosulfonium intermediates 18 are believed to react with
their counterion (or water), leading to sulfilimine 17 and
sulfoxide 19. The apparent competition (rates) of these two
pathways will determine the ratio of sulfilimine 17 to sulfoxide
19.
Further studies led to a change in the order of addition of the
reagents, which simplified the scale-up of this process. It was
found that sulfide 10 could be premixed with the cyanamide in
acetonitrile and cooled, followed by adding bleach, eliminating
one continuous slow addition. The reaction was quickly scaled
(0.5 mol), and the original workup procedure (extraction into
CH2Cl2) was modified to accommodate the new NaMnO4
oxidation (discussed in the following section). However,
sulfoxide (19) levels were found to have increased from 6%
(area LC) on the 1 mmol scale to 7−9% (area LC) on the
larger scales. The reasons for this increase were not known.
Two 0.1 mol reactions were conducted using 50% aqueous
cyanamide instead of the solid form. In addition to the lower
cost, use of this solution would make loading into larger vessels
more convenient and safer. However, in both cases, the level of
sulfoxide 19 increased to 12−13% (area LC). The initial
increase in water concentration appeared to have a detrimental
effect on this reaction.
Scheme 5. Possible Mann−Pope Type Pathway for
Sulfilimine Formation
with cyanamide or its sodium salt. The variables studied
included halogenating agent (N-chlorosuccinimide, trichloroi-
socyanuric acid, 1,3-dichloro-5,5-dimethylhydantoin, and N-
bromosuccinimide), nucleophile (cyanamide or sodium salt),
solvent, order of addition of reagents, and temperature. The
goal of this study was to rapidly identify a potential synthesis of
sulfilimine 17, which could later be developed into a procedure
suitable for scale-up. The reaction mixtures were analyzed by
HPLC (see Experimental Section for conditions), looking for
disappearance of sulfide 10 and formation of sulfilimine isomers
17. The major byproducts formed during these reactions were
the sulfoxide isomers 19, which were always present in >10%
area. In some cases, sulfoxides 19 proved to be the major
product. As a consequence, a successful procedure using these
halogenating agents was not realized. The formation of
cyanosulfilimines using 4 equiv of I2 with cyanamide and
sodium t-butoxide appeared in the literature11 after the
completion of this current study.
Earlier research indicated that sodium hypochlorite solution
(bleach) could be used with cyanamide in acetonitrile to give
sulfilimine 17. When this experiment was repeated using
commercial pool bleach obtained locally, significant amounts
(>10% area LC) of sulfoxides 19 were obtained, and large
amounts of unreacted sulfide 10 remained. Use of larger
quantities of bleach to drive the reaction to completion resulted
in the formation of ketone 20. Apparently under alkaline
conditions (Scheme 6), the sulfimimine undergoes isomer-
Oxidation of Sulfilimine 17 to Isoclast. Of particular
interest from a process scale-up perspective was the need to
replace ruthenium chloride/sodium periodate in the final
oxidation step. Although ruthenium chloride/sodium periodate
gave the desired final product in good yields, there have been
several caveats to this process. Removal of the entrained
ruthenium species in the final product requires treatment with
alumina followed by recrystallization from 2-propanol. Also,
noticeable off-gassing and deposition of a black residue into the
headspace of the reactor has been observed while running this
chemistry. In addition, difficulties encountered during workup
and phase separation of the reaction mixture dictated an
alternative synthetic procedure.
Scheme 6. Plausible Pathway to Ketone 20
ization and hydrolysis to the ketone (20). The use of solid
calcium hypochlorite appeared to give better conversion to
sulfilimine 17 with less formation of sulfoxides 19, but the
reactions could not be forced to completion. Over the course of
the reaction, calcium salts would gum out and eventually
prevent good mixing of the reaction mixture. Addition of water
There are some literature examples12 that describe the use of
potassium permanganate as an oxidant in the formation of
sulfoximines. However, none of these references contain
D
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
examples of substrates with pyridine ring functionality or a
nitrile group on the sulfilimine nitrogen. With respect to
Isoclast, some preliminary work demonstrated that catalytic
reaction of sulfilimine 17 with 4 equiv of peracetic acid and 6
mol % of manganese dioxide (MnO2) generated Isoclast in
about 56% yield of crude isolated product. While these
preliminary experiments were encouraging, MnO2 is known
to be a catalyst for peroxide decomposition. It was noticed
during this experiment that the reaction temperature increased
quickly, and a water bath was needed to moderate the internal
temperature. During this reaction, a purple color was observed
(possibly indicating the presence of a Mn7+ species). This
prompted efforts to study Mn7+ (e.g., some permanganate salt)
for the final oxidation step.
Either sodium or potassium permanganate should feasibly
work equally in this oxidation step; however, the cheaper
sodium salt was chosen based on the cost of future scale-up.
Commercial 40 wt % aqueous sodium permanganate was
utilized for these studies. The studies began with looking at
two-phase liquid−liquid oxidations. A previously prepared
solution (using the iodobenzene diacetate route) of sulfilimine
17 in methylene chloride (∼11−15 wt %) was utilized for these
initial process studies. This sulfilimine solution was cooled in an
ice−water bath, and then the aqueous permanganate solution
(1 mol equiv) was slowly added via an addition funnel. It was
determined that running the reaction with about 0.95 mol equiv
of permanganate relative to sulfilimine 17 gave clean
conversion, acceptable yields, and a manageable workup
scheme. The reaction was very exothermic, so the permanga-
nate addition rate was adjusted to maintain the internal reaction
temperature at the desired setting (typically, 10−20 °C). The
reaction mixture was analyzed by reverse-phase HPLC, and the
reaction was subsequently quenched with sodium bisulfite to
remove excess oxidant. The organic layer was separated and
concentrated on a rotovap to obtain crude Isoclast as a white
solid.
One concern with this oxidation strategy was the manganese
byproducts formed during the redox chemistry. As the
permanganate reagent was slowly added, the Mn7+ species
presumably became reduced to MnO2. Over time, as the
permanganate was slowly added, the reaction mixture became
more viscous from manganese byproduct formation. This
brown precipitate exhibited fine particles that were difficult to
filter quickly as well as remove entrained sulfoximine product.
In addition, the MnO2 precipitate exhibited a willing propensity
to deposit a thick rind layer onto the walls of the glass reactor.
From a scale-up perspective, formation of such a thick rind
layer could possibly insulate the reaction mixture from the
appropriate heat transfer needed by the cooling jacket in larger
reaction vessels. It was found that by reversing the addition
order of reagents and running the reaction more dilute these
process issues were eliminated.
Using this approach, the reaction vessel was charged with 40
wt % sodium permanganate followed by additional water and
methylene chloride (see Experimental Section for details), and
then the sulfilimine/methylene chloride solution was slowly
added via addition funnel. The addition rate was adjusted so
that the internal reaction temperature could be maintained
between 10 and 20 °C. The solution viscosity appeared lower
with this reverse addition approach. When the bisulfite quench
was added to the reaction vessel, there was no apparent rind
layer after 30 min of additional stirring. The reaction was then
filtered, and the product was isolated as described in the
Experimental Section.
Several impurities have been observed in the starting
sulfilimine solution and the crude isolated sulfoximine product.
Figure 1 summarizes the main impurities than can be easily
Figure 1. Potential relevant impurities associated with the Isoclast
process.
detected by HPLC. Sulfoxide 19 was a byproduct formed
during the preparation of sulfilimine 17. Sulfone 21 was the
direct oxidation product of sulfoxide 19 during the
permanganate chemistry. Controlling the formation of sulfone
21 could be achieved through suppressing the amount of
sulfoxide 19 formation during the sulfilimine preparation.
Sulfilimine 22 was likely formed when sulfilimine 17 was
exposed to more acidic reaction conditions. Formation of this
impurity could likely be suppressed by controlling the pH of
the reaction.
Methyl ketone 20 was an impurity that was ubiquitous to
both oxidation steps. Methyl ketone (20) formation can be
suppressed by controlling the pH during permanganate
oxidation of sulfilimine 17. The biggest breakthrough in
understanding pH dependence came from the following
observations. Observation 1: When pure/crystalline sulfilimine
isomers are dissolved into methylene chloride and then treated
with 40 wt % sodium permanganate (usual reaction
concentration), there was considerably more generation of
methyl ketone 20 than in the standard process. Observation 2:
Most of the sulfilimine/methylene chloride solutions utilized
for permanganate oxidations were prepared from the
iodobenzene diacetate chemistry. Toward the end of the
permanganate studies, an alternative oxidation approach to
preparing the sulfilimine using commercial bleach was under
development (see previous discussion). When these sulfilimine
solutions (prepared from bleach) where subject to the
permanganate oxidation conditions, a higher percent area by
HPLC analysis of methyl ketone 20 was observed as compared
to that from the standard process.
During the sodium permanganate oxidations, sodium
hydroxide was generated, leading to a highly alkaline mixture.
Under these conditions, methyl ketone 20 formation was
observed. In the iodobenzene diacetate route to sulfilimine 17,
2 equiv of acetic acid are generated from this oxidation step and
carried through to the permanganate oxidation, in effect
neutralizing the sodium hydroxide that was formed. If
sulfilimine/methylene chloride solutions prepared from the
bleach route were subsequently spiked with acetic acid, then the
permanganate oxidation gave clean conversion to the desired
sulfoximines with very little methyl ketone present. Therefore,
further oxidations required the addition of acetic acid to the
sulfilimine solutions prior to slow addition to the aqueous
permanganate.
Oxidations to Prepare Isoclast. As previously mentioned,
Isoclast was produced utilizing the iodobenzenediacetate
oxidation of pyridine sulfide 10 to sulfilimine 17 followed by
sodium permanganate oxidation in CH2Cl2 to the desired
product. During the early stages of process development, the
E
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
bleach/cyanamide procedure was investigated, and it was
decided to scale this newly developed chemistry (see
Experimental Section). This new procedure also simplified
workup of the sulfilimine reaction by eliminating extraction into
methylene chloride, thus eliminating its use. The acetonitrile
sulfilimine solution was used directly in the permanganate
oxidation. The main impurity found in this material was sulfone
21. During our studies using this newly developed procedure, it
was decided to wash a batch of the final filtered Isoclast with
25% aq 2-propanol vs straight H2O to improve the overall
purity of the product. In this case, a 60% yield of material was
obtained with an improved purity of 96.2 wt %.
to room temperature and stirred for about 5 h. A small sample
was removed and analyzed via H NMR, which indicated that
1
the reaction was complete. The mixture was filtered, and the
solid was washed with toluene. The filtrate was concentrated on
the rotary evaporator (20 mmHg/35 °C), leaving 13 as a pale
yellow liquid (677 g). The material was used in the subsequent
step. 1H NMR (300 MHz, CDCl3/major isomer >95%) δ 1.33
(d, J = 6.6 Hz, 3H), 1.81 (m, 4H), 2.00 (s, 3H), 2.99 (m, 4H),
3.22 (m, 1H), 3.98 (dd, J = 13.5, 9.0 Hz, 1H), 6.26 (d, J = 13.5
Hz, 1H).
5-[1-(Methylthio)ethyl]-2-(trifluoromethyl)pyridine
(10). A 1 L three-necked, round-bottomed flask was fitted with
a condenser, a mechanical stirrer, and an addition funnel. The
vent from the condenser was connected to a 2 L trap that could
be purged with N2. The vent from the trap went to a 1 L
jacketed (cold water) flask with a subsurface dip tube. The 1 L
flask contained 10% bleach. The vent from the 1 L flask was
SUMMARY
■
An efficient synthesis of Isoclast has been discovered,
developed, and implemented. The synthesis included a de
novo pyridine ring construction to provide an advanced
intermediate. Two efficient and inexpensive oxidations of this
intermediate provided a concise, low-cost synthesis of Isoclast
Active15 1.
1
connected to a recirculating scrubber column packed with /4
in. ceramic saddles. Bleach was pumped through this scrubber
column. The column was vented in a hood. The system was
purged with N2 via the 2 L purge trap. To the reaction vessel
were added CH3CN (694 g) and ETFBO (9, 737 g, 4.39 mol).
The solution was cooled to below 15 °C, and then the enamine
(8, 677 g, ca. 3.95 mol) was added dropwise, keeping the
temperature at or below 15 °C. The dark reaction mixture was
warmed to room temperature and stirred overnight. A small
EXPERIMENTAL SECTION
■
The 3-(methylthio)butanal was purchased from R.C. Treatt
(>97% purity). 14 was purchased from Solvay (>98% purity).
Sodium permanganate was purchased from the Carus
Corporation. Iodobenzene diacetate was purchased from
Deepwater Chemical. All solvents were HPLC grade and
purchased from Fisher Scientific. All other raw materials were
ACS grade purchased from Aldrich Chemical.
1
sample was removed and analyzed by H NMR. The NMR
spectra indicated that the reaction was complete. To the vessel
was added ammonium acetate (461 g, 5.98 mol), and the dark
slurry was heated at 80 °C and held there for 30 min. At that
time, HPLC analysis indicated that the reaction was complete
(less than 1% 15 remained). The dark mixture was cooled to
room temperature and sparged with N2 for 1.5 h. Acetonitrile
(128 g), water (485 g), and hexane (502 g) were added. After
stirring for 30 min, the phases were separated, and the aqueous
phase was extracted again with hexane (365 g). The combined
organic phases were concentrated on the rotary evaporator (20
mmHg, 35 °C), leaving a very dark liquid. The material was
distilled through a 12 cm Vigreux column, and product 10 was
collected at about 90 °C overhead temperature at 2 mmHg as a
yellow liquid (593 g, >98 wt % via internal standard GC assay).
1H NMR (400 MHz, CDCl3) δ 1.63 (d, J = 7.2 Hz, 3H), 1.95
(s, 3H), 3.94 (q, J = 7.1 Hz, 1H), 7.68 (d, J = 8.0 Hz, 1H), 7.90
(dd, J = 8.0, 2.0 Hz, 1H), 8.66 (d, J = 2.0 Hz, 1H). GC-MS m/z
(% relative intensity, ion) 221 (30%, M+), 174 (100%), 154
(85%).
N-Cyano-S-[1-(6-trifluoromethyl-3-pyridinyl)ethyl]-S-
methylsulfilimine (17).13 Procedure 1. In a 3 L four-necked,
round-bottomed flask, a mixture of 221 g (1.0 mol) of 5-[1-
(methylthio)ethyl]-2-(trifluoromethyl)pyridine (10) and 42 g
(1.0 mol) of cyanamide in 1200 mL of acetonitrile was cooled
below 10 °C with an ice bath. To this solution was added 322 g
(1.0 mol) of iodobenzene diacetate all at once. The yellow
slurry was allowed to stir below 10 °C for 10 min, and then the
ice bath was removed. The reaction mixture slowly warmed to
room temperature over 1.5 h and then slowly exothermed from
22 to 30 °C over the next 0.5 h. The reaction was now a clear
orange solution. The reaction mixture was allowed to return to
room temperature, when LC analysis indicated that there was
<1% (area) starting sulfide. The reaction solution was
transferred to a 5 L round-bottomed flask with a mechanical
stirrer and bottom drain, and 800 mL of water was added.
Excess oxidant was destroyed by adding ∼20 mL of an aqueous
The pyridine sulfide (10) could be assayed by GC analysis
using dipropyl phthalate as the internal standard: Agilent HP-1
(30 m × 0.32 mm, 0.25 μm film), temp 1 = 50 °C, time 1 = 2
min, rate = 30 °C/min, temp 2 = 300 °C, 24 psi He pressure.
The disappearance of diene 15 could be monitored using
normal-phase HPLC: Zorbax RX-SIL (4.6 × 250 mm, 5 μm),
60% hexane/40% t-butylmethyl ether, 1 mL/min, 225 nm
detection. GC/MS analyses were conducted using an Agilent
5975 Inert mass selective detector at 70 eV ionization potential.
The GC column and temperature program are as described
above.
Isoclast Active (1) and sulfilimine 17 could be assayed by
HPLC using o-toluic acid as the internal standard: Agilent
Zorbax RX-C8 (250 mm × 4.6 mm, 5 μm), 25% solvent A =
90:10:0.1 CH3CN/CH3OH/formic acid, 75% solvent B =
90:10:0.1 water/CH3OH/formic acid, 1 mL/min, 32 °C, 230
or 260 nm.
NMR spectra were recorded on a Bruker Avance 400 MHz
1
(400.13 MHz for H; 100.62 for 13C) or a Bruker 300 MHz
(300.13 MHz for 1H; 75.5 MHz for 13C) instrument. Chemical
shifts are reported in parts per million (ppm) (δ) relative to
tetramethylsilane. Multiplicities are reported as follows: s
(singlet), d (doublet), t (triplet), and m (multiplet). Coupling
constants (J) are reported in Hz.
3-(Methylthio)-1-(pyrrolidin-1-yl)-1-butene (13). A 5 L
three-necked, flask round-bottomed flask with thermowell was
fitted with a N2 inlet/addition funnel, a mechanical stirrer, and
a stopper. After the vessel was flushed with N2, it was loaded
with granular potassium carbonate (278 g, 2.01 mol), toluene
(2091 g), and pyrrolidine (317 g, 4.45 mol). The slurry was
cooled to about 10 °C, and the aldehyde 6 (411 g, 3.99 mol)
was added dropwise, keeping the temperature at or below 15
°C. After the addition was complete, the mixture was warmed
F
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
solution of sodium metabisulfite, testing the mixture with starch
iodide paper. To the mixture was added 800 mL of hexanes,
and the mixture was stirred for 5 min and separated. The
bottom aqueous layer was returned to the flask, 400 mL of
water was added followed by 400 mL of hexanes, and the
mixture was stirred for 5 min and separated. The aqueous was
again returned to the round-bottomed flask and extracted a
third time with 400 mL of hexanes. The hexane extracts
(containing the bulk of iodobenzene) were discarded. The
aqueous layer was concentrated in vacuo (to remove
acetonitrile) until a cloudy two-phase mixture was obtained.
This mixture was extracted two times (700 mL, 300 mL) with
dichloromethane, and the organics were combined and dried
overnight over MgSO4. After filtration, the dichloromethane
solution (1560 g) was used directly in the following oxidation.
LC analysis indicated a 92 area % (28:64 ratio) of two
sulfilimines isomers.
Purification of the Two Sulfilimine Isomers. Isomer A.
Sulfilimine solution from a typical reaction (40 mL, ∼16 wt %
theory) was concentrated in vacuo and exposed to high vacuum
to give a thick orange/amber oil. This oil was dissolved in 10
mL of EtOAc, and 10 mL of hexanes was added. To the cloudy
mixture was added 1 mL of EtOAc to give back a clear solution.
The flask was scratched with a glass rod to induce
crystallization. The mixture was cooled in a refrigerator for 1
h, filtered, and exposed to high-vacuum drying to give 1.2 g of a
white powder. mp 115−117 °C. >99% (area) LC of the first
eluting isomer. 1H NMR (400 MHz, DMSO-d6) δ 8.79 (d, J =
2 Hz, 1H), 8.15 (dd, J = 8, 2 Hz, 1H), 8.02 (d, J = 8 Hz, 1H),
4.73 (q, J = 7 Hz, 1H), 2.66 (s, 3H), 1.77 (d, J = 7 Hz, 3H). 13C
NMR (101 MHz, DMSO-d6) δ 150.89, 146.61 (q, J = 34.3 Hz),
139.16, 133.33, 122.89, 120.80, 120.56 (m), 120.16, 56.35,
30.49, 14.93.
The organics were combined, and this acetonitrile/sulfilimine
solution was used directly in the following oxidation. LC
analysis indicated a 95 area % (40:54 ratio) of two sulfilimines
isomers.
N-Cyano-S-[1-(6-trifluoromethyl-3-pyridinyl)ethyl]-S-
methylsulfoximine (1).14 Procedure 1. To 1500 mL of water
was added 214 g (1.0 mol) of sodium periodate, and the
mixture was stirred for 15 min. Initially, the solids dissolved to
give a cloudy mixture, but they later reappeared toward the end
of 15 min (hydrate?). To this white mixture was added 930 mg
(4.5 mmol) of ruthenium(III) chloride hydrate, and the dark
brown mixture was stirred for 5 min. To this rapidly stirred
mixture was added dropwise a solution of ∼1.0 mol sulfilimine
in 1000 mL of dichloromethane over 1.75 h. Note that during
the first several minutes of sulfilimine solution addition a vigorous
off-gassing was seen, as evidenced by rapid bubbling through the N2
bubbler. Also at this time, a black residue would form on the
exposed surfaces of the reaction flask along with black color
formation in the N2 bubbler. The cause of this off-gassing and black
color formation was not known. The temperature during this
addition rose from 16 to 29 °C. During some of these
oxidations, a thick flocculent solid would form; in others, the
reaction remained fairly thin with a minimal amount of solids
present. The reaction mixture was monitored by LC analysis,
and in some of the oxidations, additional small portions of
sodium periodate would need to be added to complete the
sulfilimine to sulfoximine oxidation. The reason why some of
the oxidations stalled before completion was not known. Upon
completion, the reaction mixture was transferred to a 5 L
round-bottomed flask with mechanical stirrer and bottom drain,
and 1.5 L of water was added. After stirring for 10 min, the
mixture was allowed to settle, and the layers were separated
(dark mixture with solids present; a flashlight was used to aid in
determining the interface). The aqueous mixture was re-
extracted with 200 mL of dichloromethane, and the aqueous
mixture was discarded. The organics were combined and
returned to the workup vessel, and 500 mL of water was added.
Excess oxidant was destroyed by adding ∼40 mL of an aqueous
solution of sodium metabisulfite, testing the mixture with starch
iodide paper. The layers were separated, and the organics were
dried over MgSO4, filtered, and concentrated in vacuo to give a
paste-like gray solid. This solid was heated and dissolved in 500
mL of hot 2-propanol, and the flask was placed in a refrigerator
overnight. The slurry was filtered and rinsed with 2-propanol to
give 260 g of a gray solid. The solid was air-dried in a hood
overnight and further dried in a vacuum oven at 40 °C to give
217 g (78% yield for two steps) of a gray solid; 30:70 (area)
ratio of isomers by LC analysis.
Isomer B. The filtrate from above was concentrated in vacuo
to give a thick amber oil [LC analysis indicated an 82 area %
15:67 ratio) of two sulfilimines isomers]. This oil was purified
by flash chromatography on silica, eluting with 5% EtOH in
CHCl3. Some minor colored material was discarded first. The
major sulfilimine isomer (second eluting isomer by LC) was
collected next, concentrated in vacuo, and exposed to high-
vacuum drying to give 3.2 g of thick amber oil. This oil was
slurried and scratched with 20 mL of Et2O, cooled in a
refrigerator, filtered, and exposed to high-vacuum drying to give
2.48 g of a white powder. mp 78−80 °C. >99% (area) LC of
1
the second eluting isomer. H NMR (400 MHz, DMSO-d6) δ
8.89 (d, J = 2 Hz, 1H), 8.22 (dd, J = 8, 2.0 Hz, 1H), 8.01 (d, J =
8 Hz, 1H), 4.78 (q, J = 7 Hz, 1H), 2.74 (s, 3H), 1.75 (d, J = 7
Hz, 3H). 13C NMR (101 MHz, DMSO-d6) δ 150.50, 146.61 (q,
J = 34.3 Hz), 138.58, 134.99, 122.83, 120.89 (m), 120.61,
120.10, 56.84, 29.82, 12.64.
Purification for Analytical Sample. The gray color
(ruthenium) could be removed from the product by treatment
with neutral alumina. Contact time should be kept to a
minimum to prevent isomerization of the sulfoximine isomers.
For example, 385 g of the gray sulfoximine [98 area % (31:67
ratio) of isomers by LC] was heated and dissolved in 2.5 L of
EtOAc. To this dark solution was added 190 g of neutral
alumina (Aldrich, Brockmann I, neutral activated), and the
mixture was stirred for 10 min. The mixture was filtered warm
through Celite, concentrated in vacuo, and air-dried in a hood
to give 378 g of a light tan solid [99 area % (32:67 ratio) of
isomers by LC]. A total of 1128 g of sulfoximine treated in the
above manner was heated and dissolved in 2.5 L of i-PrOH.
The solution was allowed to slowly cool to 30 °C with stirring.
The mixture was filtered at 30 °C and rinsed with 2-propanol to
Procedure 2. In a 2 L four-necked, round-bottomed flask a
solution of 110.6 g (0.475 mol, 95% assay) of 5-[1-
(methylthio)ethyl]-2-(trifluoromethyl)pyridine and 25.2 g
(0.6 mol) of cyanamide in 600 mL of acetonitrile was cooled
to −5 °C with a salt/ice bath. To this solution was added 750 g
(0.575 mol) of 5.7 wt % aqueous commercial bleach dropwise
over 45 min, with the temperature kept below 0 °C. The yellow
reaction mixture was allowed to stir at −1 °C for 30 min. To
the light orange mixture was added 9.5 g (0.05 mol) of sodium
metabisulfite in 25 mL of water to destroy any remaining
oxidant, and the two-phase mixture was allowed to settle. An
insoluble material was present in the aqueous phase. This
aqueous phase was re-extracted with 2 × 50 mL of acetonitrile.
G
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
give 1397 g of an off-white solid. This material was dried in a
hood overnight and then exposed to vacuum oven drying at 40
°C to give 1052 g of an off-white powder. mp 116−128 °C.
within 5 min. The mixture was stirred cold for 1 h and filtered
to give 147.6 g of a white solid. The product was air-dried in a
hood for 2 days to give 116.5 g of product and was further dried
in a vacuum oven at 35 °C to give 116.5 g (88% yield) of a
white powder. LC analysis indicated a 95 area % (43:52 ratio)
of two isomers, with the major impurity being the sulfone
(3.5% area).
1
34:66 area ratio of isomers by LC. H NMR (300 MHz,
DMSO-d6) δ 8.58 (s, 1H), 8.26 (m, 1H), 8.03 (d, J = 8 Hz,
1H), 5.37 (m, 1H), 3.49, 3.47 (2s, 3H), 1.88 (m, 3H).
Procedure 2. In a 5 L four-necked, round-bottomed flask, a
mixture of 400 mL of dichloromethane, 400 mL of water, and
320 mL (1.25 mol) of a 40% aq solution of NaMnO4 (Aldrich)
was cooled to 13 °C with an ice bath. To this rapidly stirred
mixture was added dropwise a solution of (∼1.0 mol)
sulfilimine in 1000 mL of dichloromethane (∼1560 g) over
1.75 h. During this time, the ice bath was lowered or raised to
maintain a reaction temperature of 13−20 °C. A gritty film
appeared to form on portions of the round-bottomed flask. A
10 min postreact LC sample indicated no remaining sulfilimine
starting material. After stirring for 30 min at 15 °C, a solution of
570 g (3.0 mol, 3 equiv) of sodium metabisulfite in 900 mL of
water was added with rapid stirring over 1.5 h: Note that this is
very exothermic, as the temperature rose from 15 to 28 °C
rapidly at first. The dark reaction mixture color gradually
lightened, and a brownish flocculent mixture was obtained by
the conclusion of bisulfite addition. No film or rind remained
on the flask sides at this point. The mixture was stirred at room
temperature (23 °C) for 30 min and was then filtered through
Whatman coarse wet-strengthened filter paper. The brown
paste-like solid was patted down and rinsed with two wet cake
volumes of dichloromethane (wet cake wt 500 g). The clear
two-phase mixture was transferred to a 4 L separatory funnel,
and the bottom organics were collected. The aqueous layer was
re-extracted with 30 mL of dichloromethane, and the organics
were combined with the first cut (∼1800 mL volume). The
solution was concentrated in vacuo to give 275 g of a white
solid. This solid was air-dried overnight in a hood to give 260 g
and finally in a vacuum oven at 40 °C to give 259 g (93% yield)
of a white solid. LC analysis indicated a 98 area % (30:68 ratio)
of two isomers.
Procedure 3. In a 2 L four-necked, round-bottomed flask, a
mixture of 100 mL of acetonitrile, 200 mL of water, and 160 g
(0.45 mol) of a 40% aq solution of NaMnO4 (Aldrich) was
cooled to 15 °C with an ice bath. To this rapidly stirred mixture
was added dropwise a solution of (∼0.475 mol) sulfilimine in
∼700 mL of acetonitrile over 50 min. During this time, the ice
bath was lowered or raised to maintain a reaction temperature
near 19 °C. The reaction was allowed to postreact for 45 min.
The dark mixture was cooled to 12 °C, and a solution of 171 g
(0.9 mol) of sodium metabisulfite in 300 mL of water was
added with rapid stirring over 15 min: Note that this very
exothermic, as the reaction temperature rose from 12 to 28 °C
rapidly at first. The dark reaction mixture color gradually
lightened, and an off-white flocculent mixture was obtained by
the conclusion of bisulfite addition. A small dark rind remained
on the flask sides at this point, but it dissipated within 10 min.
The mixture was stirred at room temperature (23 °C) for 30
min and was then filtered through Whatman filter paper. The
off-white solid was rinsed with 50 mL of acetonitrile. The clear
two-phase mixture was transferred to a 2 L separatory funnel,
and the bottom aqueous layer was discarded. The upper yellow
organic layer was concentrated in vacuo to ∼50 wt % product.
At this point, the product was a two-phase mixture, and a small
amount of crystalline material could be seen on the sides of the
flask. This mixture was poured onto 300 mL of rapidly stirred
water in an ice bath, and the white insoluble phase solidified
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: dcbland@dow.com. Tel: (989) 638-6946.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors would like to thank analytical sciences group
within Dow for developing process analytical methods.
DEDICATION
■
This article is dedicated in memory of Kim Arndt, whose
passion and energy for process chemistry development have
inspired many of his colleagues.
REFERENCES
■
(1) (a) Loso, M. R.; Nugent, B. M.; Huang, J. X.; Rogers, R. B., Zhu,
Y.; Renga, J. M.; Hegde, V. B.; Demark, J. J. Insecticidal N-substituted
(6-haloalkylpyridin-3-yl)alkyl sulfoximines. US Patent 7,687,634 B2,
2010. (b) Loso, M. R.; Watson, G. B.; Babcock, J.; Kramer, V. J.; Zhu,
Y.; Nugent, B. M.; Thomas, J. D. Modern Crop Protection Compounds,
2nd ed.; John Wiley & Sons, Inc.: New York, 2012; Vol. 3, p 1226.
(2) Zhu, Y.; Loso, M. R.; Watson, G. B.; Sparks, T. C.; Rogers, R. B.;
Huang, J. X.; Gerwick, B. C.; Babcock, J. M.; Kelley, D.; Hegde, V. B.;
Nugent, B. M.; Renga, J. M.; Denholm, I.; Gorman, K.; DeBoer, G. J.;
Hasler, J.; Meade, T.; Thomas, J. D. J. Agric. Food Chem. 2011, 59,
2950.
(3) Ross, R.; Renga, J. M.; Bland, D. C.; Roth, G.; Fung, A. P.; Davis,
C. S. Methods for preparing 3-substituted-6-trifluoromethylpyridines from
6-trichloromethyl halogenated pyridines. US Patent 8,785,650 B2, 2014.
(4) For a review, see Henry, G. D. Tetrahedron 2004, 60, 6043.
(5) (a) Cooke, J. W. B.; Coleman, M. J.; Caine, D. M.; Jenkins, K. P.
Tetrahedron Lett. 1998, 39, 7965. (b) Desrosiers, J. N.; Kelly, C. B.;
Fandrick, D. R.; Nummy, L.; Campbell, S. J.; Wei, X.; Sarvestani, M.;
Lee, H.; Sienkiewicz, A.; Sanyal, S.; Zeng, X.; Grinberg, N.; Ma, S.;
Song, J. J.; Senanayake, C. H. Org. Lett. 2014, 16, 1724.
(6) Cook, G. A. Enamines: Synthesis, Structure and Reactions; Marcel
Dekker, Inc.: New York, 1988; p 7.
(7) Kemp, J. E. G.; Ellis, D.; Closier, M. D. Tetrahedron Lett. 1979,
20, 3781.
(8) Gaggero, N.; D’Accolti, L.; Colonna, S.; Curci, R. Tetrahedron
Lett. 1997, 38, 5559.
(9) (a) Gilchrist, T. L.; Moody, C. J. Chem. Rev. 1977, 77, 409.
(b) Oae, S.; Furukawa, N. Sulfilimines and Related Derivatives;
American Chemical Society: Washington, DC, 1983. (c) Reggelin,
M.; Zur, C. Synthesis 2000, 1, 1.
(10) Hutchins, M. G. K.; Swern, D. J. Org. Chem. 1982, 47, 4847.
(11) Mancheno, O. G.; Bistri, O.; Bolm, C. Org. Lett. 2007, 9, 3809.
(12) (a) Kresze, G.; Wustrow, B. Chem. Ber. 1962, 95, 2652.
(b) Claus, P. K.; Rieder, W.; Hofbauer, P.; Vilsmaier, E. Tetrahedron
1975, 31, 505. (c) Claus, P. K.; Hofbauer, P.; Rieder, W. Tetrahedron
Lett. 1974, 15, 3319. (d) Johnson, C. R.; Rigau, J. J. J. Org. Chem. 1968,
33, 4340. (e) Shimizu, H.; Sugimoto, A.; Kataoka, T.; Hori, M.
Heteroat. Chem. 1995, 6, 167.
(13) Podhorez, D. E.; Ross, R., Jr.; McConnell, J. R. Process for the
preparation of certain substituted sulfilimines. US Patent 7,868,027 B2,
2011.
H
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(14) (a) Arndt, K. E.; Bland, D. C.; Podhorez, D. E.; McConnell, J. R.
Process for the oxidation of certain substituted sulfilimines to insecticidal
sulfoximines. US Patent 7,511,149 B2, 2009. (b) Bland, D. C.; Adaway,
T. J.; Podhorez, D. E. Process for the Preparation of Certain Substituted
Sulfilimines. US Patent 8,193,222 B1, 2012.
(15) Isoclast is a trademark of The Dow Chemical Company (DOW)
or an affiliated company of Dow.
I
DOI: 10.1021/acs.oprd.5b00007
Org. Process Res. Dev. XXXX, XXX, XXX−XXX