New conditions for the synthesis of thiophenes via the Knoevenagel/Gewald reaction sequence. Application to the synthesis of a multitargeted kinase inhibitor

Article abstract of DOI:10.1016/j.tet.2006.07.008

Novel conditions have been developed for the preparation of substituted 2-aminothiophenes employing the Knoevenagel condensation followed by the Gewald reaction. The benefits of these conditions are their mildness, and the ease of product isolation. Thus,

Full text of DOI:10.1016/j.tet.2006.07.008

Tetrahedron 62 (2006) 11311–11319
New conditions for the synthesis of thiophenes via the
Knoevenagel/Gewald reaction sequence. Application to the
synthesis of a multitargeted kinase inhibitor
*
David M. Barnes, Anthony R. Haight, Thomas Hameury, Maureen A. McLaughlin,
Jianzhang Mei, Jason S. Tedrowy and Joan Dalla Riva Toma
GPRD Process Research and Development, Abbott Laboratories, 1401 Sheridan Road, North Chicago, IL 60064-6285, United States
Received 1 March 2006; revised 30 June 2006; accepted 3 July 2006
Available online 28 July 2006
This paper is dedicated to Professor David MacMillan on the occasion of his receipt of the Tetrahedron Young Investigator Award
Abstract—Novel conditions have been developed for the preparation of substituted 2-aminothiophenes employing the Knoevenagel con-
densation followed by the Gewald reaction. The benefits of these conditions are their mildness, and the ease of product isolation. Thus,
the Knoevenagel condensation is run in the presence of hexamethyldisilazane and acetic acid, which combine to perform the roles of
desiccant, and catalyst. The Gewald reaction is performed with inorganic base in THF/water, which suppresses byproduct formation. This
process has been employed in the total synthesis of a multitargeted kinase inhibitor.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Thienopyrimidine 1 is a multitargeted receptor tyrosine
kinase inhibitor, which demonstrates potent antitumor acti-
vity;3 retrosynthetically, its heterocyclic core arises via con-
densation of 2-aminothiophene 4a with formamide to form
the pyrimidine ring (Scheme 1). Knoevenagel condensation
of 40-nitroacetophenone (2a) with malononitrile, followed
by Gewald reaction with sulfur forms this aminothiophene
in a concise fashion. Initial attempts to accomplish the syn-
thesis of 2-aminothiophene 4a employing standard literature
conditions1 provided poor yields for both the Knoevenagel
and Gewald reactions and resulted in complicated mixtures
that were difficult to purify. In this paper, we describe new
conditions for both the Knoevenagel and Gewald reactions
and their application in the synthesis of 1. In addition, we de-
scribe the extension of these conditions to other substrates.
2-Aminothiophenes have demonstrated a broad spectrum of
uses, including pharmaceuticals, dyes, and agrochemical ap-
plications.1 Conceptually, the simplest and most convergent
preparation of this class of compounds is the condensation of
ketones with an activated nitrile and elemental sulfur, which
was first described in 1960s by Gewald and co-workers.2 Al-
though a one-pot procedure is well-established, the two-step
procedure in which an a,b-unsaturated nitrile is first pre-
pared by Knoevenagel condensation of a ketone or aldehyde
with an activated acetonitrile, followed by base-promoted
reaction with sulfur, has generally been found to result in
higher yields (Eq. 1).1
NH2
X
CN
X
O
S
(1)
S
2. Results and discussion
X
CN
R2
R2
base
R2
R
R
3a-i
R
4a-i
2.1. Knoevenagel condensation
2a-i
see Table 3 for R, R2
Literature preparation4 of 3a and similar ylidene com-
pounds involves mixing of ketone, malononitrile, and a
catalyst (e.g., ammonium acetate) in benzene or toluene
(Eq. 2). The mixture is then warmed to reflux and water is
removed via azeotropic distillation employing a Dean–Stark
apparatus.
* Corresponding author. E-mail: david.barnes@abbott.com
Present address: Chemical Process Research & Development, Amgen
Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, United States.
y
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.07.008

11312
D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
HN
OCN
CH3
NO2
7
CH3
HN
NO2
NO2
O
NH2
N
NH2
NC
N
Me
N
O
S
S
S
N
H2N
1
5
4a
2a
Scheme 1. Retrosynthesis of kinase inhibitor 1.
(Eq. 3).7 Thus simple premixing of HMDS (1.2 equiv relative
to ketone) with acetic acid8 prior to addition of 40-nitroaceto-
phenone (2a) and malononitrile (2.0 equiv), followed by
warming to 65 ꢀC affords 97% conversion in 5 h withw1%
dimer 8. Following aqueous workup, crystallization from tol-
uene/heptane afforded 90% yield of ylidene crystals, which
were stable at ambient conditions for at least three months.
NH2
CN
CN
NC
NC
CN
NC
CN
O
NH4OAc
Me (2)
Me
Me
O2N
PhH or toluene
O2N
O2N
NO2
2a
3a
8
NC
CN
Me
HN(TMS)2
HOAc
toluene
O
Ylidene 3a is thermally unstable, hydrolytically sensitive,
and decomposes in the presence of weak bases (i.e., sodium
bicarbonate). This instability is most likely due to the combi-
nation of an unhindered enolizable methyl group and an elec-
tron withdrawing nitro-substituted aromatic ring. Analogous
compounds derived from acetophenone or 40-nitropropio-
phenone are much less prone to degradation. At the elevated
temperatures required for azeotropic distillation, dimeriza-
tion to form adduct 8 was significant; this dimerization re-
sulted in highly variable yields of ylidene 3a ranging from
10–70% (Eq. 2). Using lower boiling solvents such as THF
or EtOAc resulted in even more ylidene dimerization. In
addition, ylidene 3a isolated by silica gel chromatography
from these standard Knoevenagel reaction conditions was
an unstable oil, which dimerized to 8 upon standing.
(3)
+ NC
CN
Me
65 °C
O2N
O2N
90%
2a
3a
2.2. Gewald reaction
Classical conditions for the Gewald thiophene synthesis in-
volve the reaction of ylidene and sulfur with an organic base
(i.e., triethylamine, morpholine) in ethanol or dimethyl
formamide (DMF). In the case of ylidene 3a these conditions
afforded mostly dimer 8 with thiophene 4a as a minor com-
ponent of the reaction mixture. A brief solvent screen (tolu-
ene, ethanol, tetrahydrofuran, DMF) for the reaction (using
morpholine as base) identified tetrahydrofuran (THF) as
showing the highest reactivity and selectivity toward product
formation although the reaction still contained a significant
amount of dimer 8 (30–40%).
Lowering the reaction temperature tow60 ꢀC in toluene did
show promise; however, the reaction stalled (w50% con-
version); longer reaction times resulted only in significant
dimerization. Isolated ylidene 3a hydrolyzes to starting
ketone 2a under mild acidic conditions (LiH2PO4, THF/
water) indicating that the condensation reaction is revers-
ible, and that the removal of the water is necessary to drive
the reaction to completion. Addition of inorganic desiccants
A base screen in THF at ambient temperature next showed
that inorganic bases are preferred to organic (Table 1). Aque-
ous sodium carbonate showed the highest reactivity and
selectivity; in the absence of water, sodium carbonate pro-
vided almost no reaction. Careful monitoring of the sodium
carbonate reaction showed that the reaction was mildly exo-
thermic and complete upon mixing of the base. Continued
exploration of inorganic bases highlighted aqueous NaHCO3
as a superior base for the preparation of thiophene 4a. In the
optimized procedure, 1 equiv of NaHCO3 in water is added
to a stirring mixture of sulfur and ylidene in THF; the reac-
tion is complete at the end of the addition. The reverse addi-
tion protocol did not offer any benefit to product purity and
gave non-reproducible results.
ꢀ
˚
(MgSO4, Na2SO4, 4 A molecular sieves) at 60 C failed to
improve the reaction. On the other hand, an organic desic-
cant, namely trimethylsilyl acetate (TMSOAc,5 1.5 equiv
relative to ketone) afforded full conversion with only 5%
dimerization in 6 h at 60 ꢀC (toluene, NH4OAc).
Aside from increasing theyield of the reaction, the significant
decrease in ylidene dimerization under these conditions was
surprising. Postulating that buffering of the reaction by acetic
acid was responsible, the Knoevenagel reaction was attemp-
ted using acetic acid as solvent.6 Without added TMSOAc
the reaction stalled at around 80% conversion after 6 h at
65 ꢀC, but the level of ylidene dimer 8 remained low
(w2%). Addition of TMSOAc (Eq. 3) afforded 100% conver-
sion in 6 h withw1.5% dimer 8 at 65 ꢀC. Hexamethyldisila-
zane (HMDS) could also be used to generate in situ not only
the TMSOAc, but also the NH4OAc promoter for the reaction
The effect of temperature on the impurity profile of the
reaction was also investigated. Four different reaction tem-
peratures were investigated under otherwise identical condi-
tions (Table 2). Cooling the reaction to 5 ꢀC slowed the rate of
product formation significantly while comparatively increas-
ing the rate of ylidene dimerization. Increasing the reaction

D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
11313
Table 1. Gewald reaction base screen (Eq. 4)a
stoichiometric NaHCO3 (0.2 equiv) is employed the reaction
goes to completion although with a slightly poorer purity
profile. HPLC analysis of the reactions with stoichiometric
versus catalytic NaHCO3, showed that while dimer 8 is
generated in both cases, prolonged stirring at 35 ꢀC in the
presence of 1.0 equiv of NaHCO3 significantly degrades it
to a secondary byproduct, which is easier to remove during
the crystallization of desired product 4a.
NH
2CN
CN
NH2
S
NC
CN
Me
NC
S8 (1.3 equiv)
base
THF, rt
NC
(4)
+
Me
O2N
O2N
O2N
3a
4a
8
NO2
Base
HPLC
HPLC
HPLC
PA%b
8
PA%b 3a
PA%b 4a
Using the optimized reaction conditions from the experi-
ments described above, the thermochemical profile of the re-
action was monitored with a MultiMaxÔ system (Fig. 1).10
The initial endotherm arises from simple mixing of the
sodium bicarbonate solution with THF. HPLC monitoring
of the reaction mixture during the endotherm period showed
that little reaction occurred (<10% PA 4a). Most of the
reaction occurs during the exotherm, with the majority of
the reaction (>85% PA 4a) complete once Tmax is reached.
Morpholinec
Pyridinec
i-Pr2NHc
Et3Nc
5.0
71.8
ND
0.1
1
88.5
0.7
ND
0.6
0.5
0.5
39.5
0.5
64.4
59.1
35.1
3.2
86.3
91.2
69.5
80.7
66.2
35.1
1.7
14.7
17.9
6.1
2.1
4.1
1.6
2.6
KOt-Buc
Na2CO3
c
Na2CO3 (aq)d
NaHCO3 (aq)d
K2CO3 (aq)d
K3PO4 (aq)d
NaOH (aq)d
2.1
2.5
An interesting trend was observed in the temperature profile
when the addition time was varied between 15 and 90 min.
With the slower addition rates of 60 and 90 min, the initial
endotherm (due to mixing) was followed by a return of inter-
nal temperature to the jacket temperature of 35 ꢀC for a brief
period of time, followed by the exotherm. As noted above,
the reaction is complete prior to complete addition of the
base solution. Samples of reaction mixtures pulled during
the plateau period between endo- and exotherm events
showed little reaction occurring. From these reactions we
determined the amount of bicarbonate solution that had
been added at the onset of the exotherm event to be 0.173
and 0.178 equiv of NaHCO3, respectively. At this time, the
underlying mechanism of this effect is not readily apparent,
although a base-mediated reorganization of S8 to a more re-
active sulfur species may be implicated.11
a
Base was added in one portion to sulfur and 3a suspended in THF (10 mL/
g 3a). Reaction samples were analyzed by HPLC.
Peak area percent of each component. ND¼not detected.
Monitored after 120 min.
b
c
d
Monitored after 10 min.
temperature, while affording no appreciable increase in reac-
tion rate, appeared to decrease the relative rate of dimer for-
mation. At 60 ꢀC there was no detectable formation of dimer
8, however, the reaction generated new unidentified impuri-
ties. Therefore, the optimal reaction temperature was deter-
mined to be 35–40 ꢀC. Following the reaction, the product
was crystallized from a THF/EtOH/water mixture to provide
aminothiophene 4a in 80–85% isolated yield (Eq. 6).
NH2
S8 (1.2 atom equiv.)
NaHCO3 (1.0 equiv.)
NC
CN
Me
NC
S
(6)
2.3. Other substrates
THF, H2O
O2N
O2N
This two-step procedure for the synthesis of 2-aminothio-
phenes was applied to other substrates.12 As can be seen in
Table 3, thiophenes with both electron rich (entries 2 and
7) and electron poor (entries 3–6) phenyl rings at the 4-posi-
tion are prepared successfully. In addition, a 4-heterocycle-
80 - 85% yield
3a
4a
Based upon the mechanism proposed by Gewald and others,
the reaction should be catalytic in base.1,9 Indeed, if sub-
Table 2. Temperature screen for the Gewald reaction (Eq. 5)a
NH
2 CN
NH2
NC
CN
Me
NC
S8 (1.2 equiv)
NaHCO3
NC
CN
S
(5)
+
Me
THF/H2O
O2N
O2N
O2N
3a
4a
8
NO2
Temperature,b ꢀC
Assay yield%c 4a
HPLC PA%d
8
5
55
94
94
92
16
1.4
0.4
0
22
39
60
a
Ylidene 3a and sulfur (1.2 atom equiv) were suspended in THF at the specified temperature. NaHCO3 solution was added over 30 s.
Internal temperature at start of NaHCO3 addition.
Reaction yield determined by HPLC assay in relative to a standard.
Peak area percent of each component.
b
c
d

11314
D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
41.0
40.0
39.0
38.0
37.0
36.0
35.0
34.0
33.0
15 min addn
30 min addn
1 hour addn
1.5 hour addn
12.00
22.00
32.00
42.00
Time (minutes)
Figure 1. Temperature profiles of Gewald reaction at different NaHCO3 addition rates.
in 72% yield over two steps from 40-nitroacetophenone.
This set the stage for a highly convergent synthesis in which
the isocyanate was added in the final step.
Table 3. Application of the Knoevenagel and Gewald conditions to other
substrates (Eq.7)
NH2
S
HN(TMS)2
HOAc
toluene
65 °C
NC
CN
NC
R2
NC
R2
CN
S8, NaHCO3
THF, water
(7)
O
R
Condensation with formamide was accomplished at elevated
temperatures to effect conversion to nitropyrimidine 5 in
88% yield.14 In early experiments the sulfur-linked dimer
9 was observed in varying amounts. A small amount of tri-
phenylphosphine (10 mol %) added to the reaction mixture
was sufficient to scavenge residual sulfur from the Gewald
reaction, providing an effective control of this impurity.
R
R
4a-i
R2
2a-i
3a-i
Entry
Ketone
R2
R
Ylidene,
yield%
Thiophene,
yield%
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2e
2f
2g
2h
2i
4-NO2Ph
Ph
3-NO2-Ph
4-MeSO2Ph
4-BrPh
2,6-F2Ph
4-MeOPh
2-Thiophene
Ph
H
H
H
H
H
H
H
H
Ph
3a, 90
3b, 82
3c, 83 (88)
3d, 76
3e, 86
3f, 64 (73)
3g, 82
3h, 56
4a, 80
4b, 40 (96)
4c, 81
4d, 58 (96)
4e, 73 (96)
4f, 58 (98)
4g, 80
NO2
NHEt
NHMe
N
NH2
N
N
S
N
NH2
N
NH2
N
4h, 46 (100)
4i, 78 (99)
S
NH2
3i, 80
S
N
N
Numbers in parentheses represent assay yields prior to isolation.13
S
S
NO2
9
10
11
substituted derivative (entry 8) as well as a 5-substituted
thiophene (entry 9) is accessible using this methodology.
Previous reports have detailed the use offormamidine acetate
as an effective reagent for the conversion of amino nitriles
to pyrimidines.15 While the use of this reagent (4 equiv)
in formamide did permit the reaction to be run at lower
temperatures (w125 ꢀC), we observed evolution of CO and
NH3 during the course of the reaction. In the absence of
2.4. Synthesis of thienopyrimidine 1
The two-step synthesis of 2-aminothiophenes was em-
ployed in the context of a synthesis of thienopyrimidine 1
(Scheme 2). As discussed above, thiophene 4a was prepared
NH2
NC
NC
CN
Me
O
Ph3P (10 mol%)
Formamide
S
NC
CN
S8, NaHCO3
THF, H2O, 35 °C
80%
Me
140 - 150 °C
HMDS, AcOH, 55 °C
O2N
O2N
O2N
90%
88%
2a
3a
4a
NO2
NH2
H
N
H
CH3
H3C
N C O
N
NH2
NH2
O
7
Raney Ni, H2
N
MeOH/THF
76%
N
NH2
N
EtOH/THF, 0 °C
S
S
N
N
94%
N
5
6
S
1
Scheme 2. Synthesis of thienopyrimidine 1.

D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
11315
Table 4. Solvent effects in isocyanate acylation (Eq. 10)a
formamidine acetate, the thermal decomposition of formam-
ide itself was not significant at <160 ꢀC.
NH2
H
N
H
CH3
Hydrogenation of the nitro group to the corresponding ani-
line 6 was accomplished over Raney nickel in THF/MeOH,
in 76% yield. The choice of methanol as the alcohol co-sol-
vent was based on the formation of N-ethyl derivative 10
(1–2%) when EtOH was employed, presumably via an oxida-
tion to acetaldehyde/reductive alkylation mechanism.16
When methanol was employed, less than 0.5% of N-alkylated
product 11 was observed.
H3C
N
C
O
N
NH2
N
O
7
N
(10)
NH2
N
solvent mixture (30 mL / g 6)
0 °C
S
N
6
1
S
Entry Solvent mixture
Isocyanate 7, 6, HPLC 1, HPLC 12, HPLC
equiv
PA%b
PA%b
PA%b
The final step required acylation of aniline 6 with 3-tolyliso-
cyanate 7. There are two reactive amine sites, although the
aniline (N1) proved more reactive than the aminopyrimidine
(N2), overacylation to form bis-acyl 12 (Eq. 8) was observed.
Under standard conditions (3:1 CH2Cl2/THF, 1.0 equiv iso-
cyanate, 0 ꢀC to rt),3 from which the product precipitated
early in the reaction, 3% of this impurity was formed at
99% conversion (HPLC peak area% levels at 265 nm). We
expected that thienopyrimidine 1 reacted with residual iso-
cyanate 7 at the end of the reaction and reasoned that an
alternate, sacrificial nucleophile might be added to the mix-
ture, which would not interfere with acylation of the aniline
nitrogen, but would compete with overacylation of 1 (Eq. 8)
to furnish an easily removable side-product (Eq. 9).
1
2
3
4
5
6
THF/EtOH (2:1)
THF/EtOH (2:1)
THF/MeOH (2:1) 1.25
THF/MeOH (2:1) 1.5
CH2Cl2/THF (3:1)c 0.5
1.0
1.25
2.2
0.2
0.4
0.1
19.4
13.9
97.1
98.7
98.8
98.4
78.4
85.6
0.2
0.4
0.2
0.4
1.1
0.08
THF/EtOH (2:1)
0.625
a
Isocyanate 7 was added dropwise to a mixture of 6 in the solvent mixture
(30 mL/g 6) over w5 min, maintaining the temperature at <2 ꢀC. Reac-
tions were assayed after 2 h.
Reactions were analyzed by HPLC at 265 nm.
Reaction was run at 0 ꢀC for 2 h, then warmed to rt for 2 h.
b
c
a sacrificial nucleophile of intermediate reactivity between
that of the two amines of aniline 6, so that excess isocya-
nate 7 is destroyed before it overacylates the product (even
N1
NH2
HN
HN
NCO
CH3
xs
HN
CH3
O
N2
CH3
HN
NH2
N
O
NCO
O
N
+
NH2
N
N
solvent
NuH
CH3
N
H
NH
N
X
S
N
CH3
(8)
(9)
S
S
6
7
1
12
NCO
H
N
xs
Nu
CH3
O
+ NuH
CH3
Alcohols react with isocyanates at a slower rate and possibly
via a slightly different mechanism than do amines.17 Due to
the inherent low nucleophilicity of the aminopyrimidine, an
alcohol might serve as an additive with the desired effect.
When ethanol was added to a reaction in THF (2:1 THF/
EtOH, 1.0 equiv isocyanate 7, 0 ꢀC), we observed good con-
version to 1 (2.2% remaining 6), with <0.2% overacylated
impurity 12 (Table 4, entry 1). Both MeOH and EtOH in
THF required extra isocyanate 7 to attain high levels of con-
version (<0.4% remaining 6), but a smaller excess was re-
quired with EtOH (1.25 equiv vs 1.5 equiv in MeOH for
>99.8% conversion). Stirring overnight at ambient tempera-
ture did not significantly affect impurity levels. The product
was crystallized from EtOH in 94% isolated yield.
with 1.25 equiv of isocyanate 7, only 0.4% of 12 is
formed). However, the lower levels of bis-acyl 12 even
at incomplete conversion (entries 5 and 6) indicate that
the alcohol is also affecting the relative reactivities of
the two nitrogen atoms, perhaps by selective hydrogen
bonding.18
In conclusion, we have described new reaction conditions
for the Knoevenagel condensation and Gewald thiophene
synthesis by which 40-nitroacetophenone (2a) is converted
to 2-aminothiophene 4a in 72% yield via crystalline inter-
mediates. This methodology has been extended to the use
of other alkyl aryl and alkyl heterocyclic ketones, where it
has also proven effective. Finally, we have applied this in
the context of a synthesis of multitargeted kinase inhibitor
1, which was prepared in 45% overall yield through five
steps.
Our results indicate that the alcohol co-solvent is playing
two roles in the acylation reaction. First, it functions as

11316
D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
3. Experimental
3.1.5. 2-[1-(4-Bromophenyl)ethylidene]-malononitrile
(3e). Reaction run using 40-bromoacetophenone (10.00 g,
50.3 mmol). Crystallization with heptane (88 mL) provided
10.55 g (86.0%) of yellow crystalline 3e. Mp¼93.5–
3.1. Knoevenagel condensation
1
Hexamethyldisilazane (1.2 equiv) was added to acetic acid
(0.67 mL/mmol ketone 2) at a rate to maintain the internal
temperature at or below 74 ꢀC. (Caution—this addition is
very exothermic!) The HMDS/acetic acid mixture is added
to a solution of ketone 2 and malononitrile (2 equiv) in acetic
acid (0.33 mL/mmol). After reaction completion, the mix-
ture is cooled to ambient temperature with chilled toluene
(1.67 mL/mmol, at 0 ꢀC) and diluted with water (1.33 mL/
mmol). The aqueous layer was separated and extracted
with toluene (0.67 mL/mmol). The combined organic ex-
tracts were washed four times with water and dried over
MgSO4. The products 3 were isolated by crystallization or
chromatography.
94.5 ꢀC; H NMR (400 MHz, CDCl3) d 2.44 (s, 3H), 7.23
(ddd, J¼8.89, 2.47, 2.23 Hz, 2H), 7.46 (ddd, J¼8.89, 2.47,
2.23 Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3) d 24.4,
85.1, 112.2, 112.3, 126.8, 128.5, 132.1, 134.2, 173.3 ppm;
MS (APCI+) m/z 247.0; Anal. Calcd for C11H7BrN2: C,
53.47; H, 2.86; N, 11.34. Found: C, 53.26; H, 2.87; N, 11.19.
3.1.6. 2-[1-(2,6-Difluorophenyl)ethylidene]-malononitrile
(3f). Reaction run using of 20,60-difluoroacetophenone
(10.21 g, 65.4 mmol). Crystallization with heptane (100 mL)
provided 8.48 g (63.5%) of yellow crystalline 3f. Mp¼
80–81 ꢀC; 1H NMR (400 MHz, CDCl3) d 2.78 (s, 3H), 7.22
(dd, J¼8.51, 7.96 Hz, 2H), 7.65 (tt, J¼8.51, 6.38 Hz, 1H)
ppm; 13C NMR (100 MHz, CDCl3) d 24.2, 91.4, 110.9,
111.0, 112.1, 112.1, 112.1, 112.3, 112.3, 112.3, 113.3,
113.5, 113.7, 132.8, 132.9, 133.0, 156.7, 156.8, 159.2,
159.3, 165.3 ppm; MS (APCI+) m/z 203.0 (MꢁH); Anal.
Calcd for C11H6F2N2: C, 64.71; H, 2.96; N, 13.72. Found:
C, 64.47; H, 2.83; N, 13.78.
3.1.1. 2-[1-(4-Nitrophenyl)ethylidene]-malononitrile
(3a). Reaction run using 40-nitroacetophenone (10.00 g,
60.6 mmol). Crystallization with heptane (93 mL) provided
11.49 g (89.0%) of yellow crystalline 3a. Mp¼87–88 ꢀC; 1H
NMR (400 MHz, CDCl3) d 2.68 (s, 3H), 7.69 (ddd, J¼9.09,
2.40, 2.23 Hz, 2H), 8.36 (dt, J¼9.16, 2.28 Hz, 2H) ppm; 13C
NMR (100 MHz, CDCl3) d 24.7, 87.6, 111.5, 111.5, 124.1,
128.1, 141.2, 149.1, 172.3 ppm; MS (APCI+) m/z 212.0
(MꢁH); Anal. Calcd for C11H7N3O2: C, 61.97; H, 3.31; N,
19.71. Found: C, 61.69; H, 3.30; N, 19.74.
3.1.7. 2-[1-(4-Methoxyphenyl)ethylidene]-malononitrile
(3g). Reaction run using 40-methoxyacetophenone (2.89 g,
19.3 mmol). Crystallization with heptane (30 mL) provided
3.14 g (82.3%) of yellow crystalline 3g. Mp¼79.5–
1
80.5 ꢀC; H NMR (400 MHz, CDCl3) d 2.89 (s, 3H), 4.15
3.1.2. 2-[1-(Phenyl)ethylidene]-malononitrile (3b). Reac-
tion run using acetophenone (9.7 mL, 83 mmol). Crystalliza-
tion with heptane provided 11.48 g (82.3%) of white
crystalline 3b. Mp¼94.5–95.5 ꢀC; 1H NMR (400 MHz,
CDCl3) d 2.43 (s, 3 H), 7.31 (m, 5 H) ppm; 13C NMR
(100 MHz, CDCl3) d 24.5, 84.6, 112.5, 112.5, 127.0, 128.8,
131.9, 135.5, 174.9 ppm; MS (APCI+) m/z 167.1 (MꢁH);
Anal. Calcd for C11H8N2: C, 78.55; H, 4.79; N, 16.66. Found:
C, 78.24; H, 4.47; N, 16.81.
(s, 3H), 7.26 (ddd, J¼9.43, 3.16, 2.64 Hz, 2H), 7.88 (ddd, J¼
9.50, 3.09, 2.64 Hz, 2H) ppm; 13C NMR (100 MHz, CDCl3)
d 24.1, 81.9, 113.1, 113.4, 114.2, 127.6, 129.5, 162.6,
173.4 ppm; MS (APCI+) m/z 197.1 (MꢁH); Anal. Calcd
for C12H10N2O: C, 72.71; H, 5.08; N, 14.13. Found: C,
72.38; H, 4.78; N, 14.07.
3.1.8. 2-[1-(2-Thiophene-yl)ethylidene]-malononitrile
(3h). Reaction run using 2-acetylthiophene (8.5 mL,
78.9 mmol). At the end of the reaction, ethanol (60 mL)
was added and the mixture was filtered. After drying the
solids were stirred in ethanol (45 mL) and filtered. Drying
in vacuo provided 7.67 g (56%) of crystalline 3h. Mp¼87–
3.1.3. 2-[1-(3-Nitrophenyl)ethylidene]-malononitrile
(3c). Reaction run using 30-nitroacetophenone (10.00 g,
60.6 mmol). Crystallization with heptane (93 mL) provided
10.76 g (83.4%) of yellow crystalline 3c. Mp¼99–100 ꢀC;
1H NMR (400 MHz, CDCl3) d 2.71 (s, 3H), 7.74 (t,
J¼8.03 Hz, 1H), 7.89 (ddd, J¼7.75, 1.78, 1.03 Hz, 1H),
8.36 (t, J¼1.85 Hz, 1H), 8.40 (ddd, J¼8.23, 2.13, 1.03 Hz,
1H) ppm; 13C NMR (100 MHz, CDCl3) d 24.6, 87.4, 111.5,
111.6, 122.1, 126.1, 130.3, 132.6, 136.9, 148.0, 171.9 ppm;
MS (APCI+) m/z 212.1 (MꢁH); Anal. Calcd for
C11H7N3O2: C, 61.97; H, 3.31; N, 19.71. Found: C, 61.04;
H, 3.10; N, 19.37.
1
88 ꢀC; H NMR (400 MHz, CDCl3) d 2.70 (s, 3H), 7.24
(dd, J¼4.87, 4.19 Hz, 1H), 7.77 (dd, J¼4.94, 0.82 Hz, 1H),
8.04 (dd, J¼4.05, 0.75 Hz, 1H) ppm; 13C NMR (100 MHz,
CDCl3) d 23.9, 78.7, 113.3, 113.8, 128.8, 133.2, 134.1,
137.9, 162.1 ppm; MS (APCI+) m/z 173.0 (MꢁH); Anal.
Calcd for C9H6N2S: C, 62.05; H, 3.47; N, 16.08. Found: C,
60.66; H, 2.99; N, 15.84.
3.1.9. 2-[1-(Phenyl)benzylidene]-malononitrile (3i). Re-
action run using deoxybenzoin (10.01 g, 51.0 mmol).
Crystallization with heptane (100 mL) provided 9.98 g
(80.0%) of white crystalline 3i. Mp¼78.7–79.7 ꢀC; 1H NMR
(400 MHz, CDCl3) d 4.31 (s, 2H), 7.10 (dd, J¼7.20,
2.26 Hz, 2H), 7.28 (m, 3H), 7.49 (m, 5H) ppm; 13C NMR
(100 MHz, CDCl3) d 43.4, 85.4, 112.4, 112.6, 127.5, 128.7,
131.6, 133.9, 134.4, 176.9 ppm; Anal. Calcd for C17H12N2:
C,83.58;H, 4.95;N,11.47.Found:C,83.27;H,4.78;N,11.33.
3.1.4. 2-[1-(4-Methylsulfonylphenyl)ethylidene]-malono-
nitrile (3d). Reaction run using 40-methylsulfonylacetophe-
none (10.00 g, 50.5 mmol). The product crystallized during
the reaction. After drying, the solids were stirred in ethanol
(50 mL) and filtered. Drying in vacuo provided 9.47 g
1
(76.2%) of crystalline 3d. Mp¼138.5–139.5 ꢀC; H NMR
(400 MHz, CDCl3) d 2.67 (s, 3H), 3.11 (s, 3H), 7.69 (d,
J¼8.37 Hz, 2H), 8.08 (d, J¼8.37 Hz, 2H) ppm; 13C NMR
(100 MHz, CDCl3) d 24.7, 44.5, 87.3, 111.6, 111.7, 128.0,
140.6, 143.1, 172.8 ppm; MS (APCI+) m/z 245.1 (MꢁH);
Anal. Calcd for C12H10N2O2S: C, 58.52; H, 4.09; N,
11.37. Found: C, 58.41; H, 3.78; N, 11.45.
3.2. Gewald reaction
Ylidene 3 and elemental sulfur (1.2 atom equiv) are sus-
pended in tetrahydrofuran (10 mL/g 3) and warmed to an

D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
11317
internal temperature of 35 ꢀC. A solution of sodium
bicarbonate (1.0 equiv in 5 mL of water/g 3) is added over
w1 h. The mixture is stirred at 35 ꢀC for approximately
30 min before the solution is transferred to a separatory fun-
nel and washed with 12.5% aqueous NaCl and 25% aqueous
NaCl. The products 4 were isolated by crystallization.
3.2.5. 2-Amino-4-(2,6-difluorophenyl)-3-thiophenecar-
bonitrile (4f). Reaction run using 2-[1-(2,6-difluorophe-
nyl)ethylidene]-malononitrile (3f, 2.0 g, 9.8 mmol). HPLC
assay following the reaction indicated a 98% yield. Crystal-
lization from 32 mL of 2:1:1 water/EtOH/THF, followed by
crystallization from 20 mL toluene yielded 1.35 g (58%)
of the desired product. Mp¼143.6–145.7 ꢀC; 1H NMR
(400 MHz, DMSO-d6) d 7.48 (tt, J¼8.4, 6.6 Hz, 1H), 7.27
(s, 2H), 7.19 (t, J¼8.0 Hz, 2H), 6.56 (s, 1H) ppm; 13C NMR
(100 MHz, DMSO-d6) d 164.3, 160.1 (1/2 of doublet), 157.7
(1/2 of doublet), 130.2 (t), 124.5, 115.1, 111.6, 111.3, 109.7,
84.5 ppm; MS (ESI+) m/z 237 (M+H); HRMS (ESI FTMS)
Calcd for C11H7F2N2S (M+H): 237.0293. Found: 237.0289.
3.2.1. 2-Amino-4-phenyl-3-thiophenecarbonitrile (4b).
Reaction run using 2-[1-(phenyl)ethylidene]-malononitrile
(3b, 2 g, 11.9 mmol). HPLC assay following the reaction
indicated a 96% yield. Crystallization from 15 mL of
toluene yielded 0.93 g (40% yield) of the desired product
as a white solid. Mp¼ 102.3–105.1 ꢀC (lit.19 100–102 ꢀC);
1H NMR (400 MHz, DMSO-d6) d 7.48–7.56 (m, 2H),
7.38–7.44 (m, 2H), 7.30–7.36 (m, 1H), 7.23 (s, 2H), 6.51
(s, 1H) ppm; 13C NMR (100 MHz, DMSO-d6) d 165.7,
137.9, 133.9, 128.1, 127.3, 126.4, 116.2, 104.7, 82.9 ppm.
MS (ESI+) m/z 201 (M+H); Anal. Calcd for C11H8N2S:
C, 65.97; H, 4.03; N, 13.99. Found: C, 65.96; H, 3.84; N,
13.95.
3.2.6. 2-Amino-4-(4-methoxyphenyl)-3-thiophenecarbo-
nitrile (4g). Reaction run using 2-[1-(4-methoxyphenyl)-
ethylidene]-malononitrile (3g, 1.98 g, 8.03 mmol). Crystalli-
zation from 28 mL of 2:1:1 water/EtOH/THF, followed
by trituration in 15 mL toluene yielded 1.39 g (80%) of the
desired product. Mp¼164–166.5 ꢀC (dec); 1H NMR
(400 MHz, DMSO-d6) d 7.45 (d, J¼8.9 Hz, 2H), 7.18 (s,
2H), 6.97 (d, J¼8.9 Hz, 2H), 6.39 (s, 1H), 3.77 (s, 2H) ppm;
13C NMR (100 MHz, DMSO-d6) d 165.6, 158.3, 137.6,
127.6, 126.5, 116.3, 113.6, 103.3, 83.1, 55.1 ppm; MS
(APCI+) m/z 231 (M+H); Anal. Calcd for C12H10N2OS: C,
62.59; H, 4.38; N, 12.16. Found: C, 62.40; H, 4.29; N, 12.05.
3.2.2. 2-Amino-4-(3-nitrophenyl)-3-thiophenecarbo-
nitrile (4c). Reaction run using 2-[1-(3-nitrophenyl)ethyl-
idene]-malononitrile
Crystallization from 25 mL of toluene yielded 2.34 g
(3c,
2.51 g,
11.8 mmol).
1
(81%) of the desired product. Mp¼194–197 ꢀC; H NMR
(400 MHz, DMSO-d6) d 8.34 (t, J¼2.1 Hz, 1H), 8.19 (ddd,
J¼8.2, 2.3, 1.0 Hz, 1H), 7.99 (ddd, J¼7.7, 1.8, 1.0 Hz, 1H),
7.72 (t, J¼8.0 Hz, 1H), 7.38 (s, 2H), 6.80 (s, 1H) ppm; 13C
NMR (100 MHz, DMSO-d6) d 166.1, 147.4, 135.20,
135.17, 132.7, 129.8, 122.0, 120.7, 115.9, 106.9, 82.2 ppm;
MS (ESI+) m/z 246 (M+H), 263 (M+NH4), 268 (M+Na);
Anal. Calcd for C11H7N3O2S: C, 53.87; H, 2.88; N, 17.13.
Found: C, 53.85; H, 2.64; N, 17.03.
3.2.7. 2-Amino-4-(2-thienyl)-3-thiophenecarbonitrile
(4h). Reaction run using 2-[1-(2-thienyl)ethylidene]-malo-
nonitrile (3h, 2.00 g, 11.5 mmol). HPLC assay following
the reaction indicated a 100% yield. Crystallization from
23 mL toluene yielded 1.09 g (46%) of the desired product.
Mp¼110–112 ꢀC (dec); 1H NMR (400 MHz, DMSO-d6)
d 7.49 (dd, J¼5.1, 1.2 Hz, 1H), 7.34 (dd, J¼3.6, 1.2 Hz,
1H), 7.30 (s, 2H), 7.09 (dd, J¼5.1, 3.6 Hz, 1H), 6.54 (s, 1H)
ppm; 13C NMR (100 MHz, DMSO-d6) d 165.7, 135.8, 130.5,
127.3, 124.9, 123.7, 116.0, 104.1, 82.2 ppm; MS (ESI+) m/z
207 (M+H); HRMS (ESI FTMS) Calcd for C9H7N2S2:
207.0045. Found: 207.0039; Anal. Calcd for C9H6N2S2: C,
52.40; H, 2.93; N, 13.58. Found: C, 51.45; H, 2.64; N, 13.20.
3.2.3. 2-Amino-4-(4-methanesulfonylphenyl)-3-thio-
phenecarbonitrile (4d). Reaction run using 2-[1-(4-meth-
anesulfonylphenyl)ethylidene]-malononitrile (3d, 1.4 g,
5.7 mmol). HPLC assay following the reaction indicated
a 96% yield. Crystallization from 22 mL of 2:1:1 water/
EtOH/THF, followed by crystallization from 10 mL toluene
yielded 0.91 g (58%) of the desired product. Mp¼195–
197 ꢀC; 1H NMR (400 MHz, DMSO-d6) d 7.96 (d, J¼
8.5 Hz, 2H), 7.78 (d, J¼8.5 Hz, 2H), 7.36 (s, 2H), 6.73 (s,
1H), 3.24 (s, 3H) ppm; 13C NMR (100 MHz, DMSO-d6)
d 166.1, 139.2, 138.6, 136.1, 127.2, 126.9, 115.9, 107.2,
82.3, 43.4 ppm; MS (ESI+) m/z 279 (M+H), 296
(M+NH4), 301 (M+Na); HRMS (ESI FTMS) Calcd for
C12H11N2O2S2 (M+H): 279.0256. Found: 279.0256.
3.2.8. 2-Amino-4,5-diphenyl-3-thiophenecarbonitrile
(4i). Reaction run using 2-[1-(phenyl)benzylidene]-malono-
nitrile (3i, 1.20 g, 4.9 mmol). HPLC assay following the re-
action indicated a 99% yield. Crystallization from 62 mL
of 4:1:1 water/EtOH/THF, followed by trituration with
15 mL toluene yielded 1.06 g (78%) of the desired product.
1
Mp¼203–205 ꢀC; H NMR (400 MHz, DMSO-d6) d 7.40
(s, 2H), 7.29–7.38 (m, 2H), 7.10–7.26 (m, 6H), 7.02 (ddd,
J¼6.4, 1.7, 1.5 Hz, 2H) ppm; 13C NMR (100 MHz,
DMSO-d6) d 163.3, 134.1, 133.7, 132.4, 128.8, 128.1,
127.6, 127.4, 126.4, 118.8, 115.7, 86.6 ppm; MS (ESI+) m/
z 277 (M+H); Anal. Calcd for C17H12N2S: C, 73.88; H,
4.38; N, 10.14. Found: C, 74.23; H, 4.25; N, 10.19.
3.2.4. 2-Amino-4-(4-bromophenyl)-3-thiophenecarbo-
nitrile (4e). Reaction run using 2-[1-(4-bromophenyl)ethyl-
idene]-malononitrile (3e, 1.98 g, 8.03 mmol). HPLC assay
following the reaction indicated a 96% yield. Crystallization
from 17 mL of 2:1:1 water/EtOH/THF, followed by crystal-
lization from 17 mL toluene yielded 1.67 g (73%) of the
desired product. Mp¼190–192 ꢀC; 1H NMR (400 MHz,
DMSO-d6) d 7.61 (d, J¼8.5 Hz, 2H), 7.47 (d, J¼8.7 Hz,
2H), 7.28 (s, 2H), 6.58 (s, 1H) ppm; 13C NMR (100 MHz,
DMSO-d6) d 165.9, 136.5, 133.1, 131.1, 128.4, 120.6,
116.0, 105.3, 82.5 ppm; MS (APCI+) m/z 279, 281
(M+H); Anal. Calcd for C11H7BrN2S: C, 47.33; H, 2.53;
N, 10.04. Found: C, 47.02; H, 2.25; N, 9.78.
3.3. Scale up of multitargeted kinase inhibitor 1
3.3.1. 2-[1-(4-Nitrophenyl)ethylidene]-malononitrile
(3a). Caution: malononitrile is highly toxic and readily
absorbed through the skin!
A round bottomed flask was charged with acetic acid (6 L),
malononitrile [2.6 kg], and 40-nitroacetophenone (3.0 kg).

11318
D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
To a separate flask was charged acetic acid (12 L) and
1,1,1,3,3,3-hexamethyldisilazane (HMDS) (3.55 kg) was
added at a rate to maintain the internal temperature at or be-
low 45 ꢀC. Caution: the addition of HMDS to acetic acid is
very exothermic!
146.0, 140.0, 135.5, 127.3, 123.5, 115.8, 108.0, 82.0 ppm;
MS (APCIꢁ) m/z 244 (MꢁH); HRMS (ESI FTMS) Calcd
for C11H8N3O2S (M+H): 246.0332. Found: 246.0327.
3.3.3. 4-Amino-5-(4-nitrophenyl)-thieno[2,3-d]pyrimi-
dine (5). To a round bottomed flask were charged thiophene
4a (2850 g, 11.62 mol), triphenylphosphine (243.8 g,
0.9296 mol), and formamide (48.6 kg). The reaction mixture
was warmed to an internal temperature of 145–150 ꢀC.
NOTE: This reaction has been shown to generate carbon
monoxide and ammonia due to decomposition of formamide
above 150 ꢀC. After 4 h 20 min the heating mantle was
turned off and the reaction mixture allowed to cool to
room temperature and stirred overnight. To the orange sus-
pension was added 42.6 L of water and the reaction mixture
stirred at room temperature for 1 h. The suspension was fil-
tered and washed with water (3ꢃ14.3 L). The filter cake was
then washed with acetone in portions until the filtrate was
pale yellow (5ꢃ5.8 L). Thienopyrimidine 5 was dried in va-
cuo at 60–70 ꢀC to yield 2.792 kg (88.2%) of an orange
The contents of the first flask were transferred to the acetic
acid/HMDS mixture and rinsed with acetic acid (300 mL).
The contents were warmed to 55 ꢀCꢂ10 ꢀC for 9 h 35 min,
then the heat source was turned off. The reaction was rapidly
cooled by charging pre-chilled toluene (30 L, chilled to 0 ꢀC)
to the flask. The reaction solution was then extracted with
water (23.9 L). The aqueous layer was back-extracted with
toluene (12 L). The combined organic layers were then
repeatedly washed with water (4ꢃ24 L) to remove acetic
acid and malononitrile. After drying over magnesium sulfate,
the toluene solution was concentrated in vacuo to approxi-
mately four volumes (mL/g) and the mixture was seeded
and the reaction was stirred for 15 min. Heptane (29.25 L)
was added over 4.5 h and the slurry was stirred overnight.
The slurry was then filtered and was washed with 6 L of
heptane and the product was dried in a vacuum oven at
room temperature for 8 h, to yield 3.486 kg (90.0% yield)
of the desired product.
1
solid. Mp>225 ꢀC; H NMR (400 MHz, DMSO-d6) d 8.35
(s, 1H), 8.32 (d, J¼8.9 Hz, 2H), 7.71 (d, J¼8.9 Hz, 2H),
7.67 (s, 1H) ppm; 13C NMR (100 MHz, DMSO-d6)
d 167.0, 157.6, 153.3, 146.5, 141.4, 132.6, 129.5, 123.4,
122.4, 111.8 ppm; MS (ESI+) m/z 273 (M+H), 227
(M+HꢁNO2); Anal. Calcd for C12H8N4O2S: C, 52.93; H,
2.96; N, 20.58. Found: C, 52.66; H, 2.57; N, 20.52.
3.3.2. 2-Amino-4-(4-nitrophenyl)-3-thiophenecarbo-
nitrile (4a). Ylidene 3a (3300 g), sulfur (599.0 g), and
tetrahydrofuran (29.4 kg) were charged to a flask. The
reaction contents were warmed to an internal temperature
of approximately 35 ꢀC. The reaction mixture was charged
with 7.0 kg of 7.3% aqueous NaHCO3 over a period of ap-
proximately 50 min maintaining the reaction temperature
below 65 ꢀC. An additional 10.8 kg of the sodium bicarbon-
ate solution was added to the reaction mixture at a rate,
which maintained the reaction temperature above 30 ꢀC.
The reaction mixture was maintained at 35 ꢀC for 30 min
post addition time. The heat source was turned off and
the reaction mixture was cooled to approximately 30 ꢀC.
NaCl solution (14.3 kg, 12.5%) was added in one portion
to the reaction mixture, which was stirred for 10 min and
then the layers were separated and the organics were
washed with an additional 15.3 kg of the 12.5% NaCl solu-
tion. The layers were separated and the organics were
washed with 25% NaCl solution (13.0 kg). The organics
were concentrated in vacuo to a volume of approximately
four volumes relative to theoretical yield of the product.
At this point the mixture is warmed to an internal temper-
ature of 60 ꢀC and ethyl alcohol (14.5 L) added at such
a rate to maintain internal temperature of 60 ꢀC. After seed-
ing, 23.2 L of water was added over approximately 5 h
while maintaining 60 ꢀC, then the slurry was cooled over-
night. The slurry was filtered and the solids were dried in
vacuo at 60 ꢀC.
3.3.4. 4-Amino-5-(4-aminophenyl)-thieno[2,3-d]pyrimi-
dine (6). A slurry of nitrothienopyrimidine 5 (2.8 kg) in
32 kg of tetrahydrofuran and 29 kg of methanol was added
to a mixture of Raney nickel (2.7 kg) and water (0.7 kg).
The slurry was rinsed with 32 kg tetrahydrofuran then hydro-
genated at 40 psi. After the reaction is complete, the reaction
mixture is filtered the filter cake is rinsed with 13.5 kg of tet-
rahydrofuran. After distilling to about 12 L, 15 kg of EtOH is
added and reconcentrated twice. Water (36 kg) is added over
1 h. The product was isolated by filtration, washed with
20.6 kg of water and dried in a vacuum oven at 55 ꢀC to yield
1.894 kg (76%) of the desired aminothienopyrimidine 6.
1
Mp¼185–188 ꢀC; H NMR (400 MHz, DMSO-d6) d 8.28
(s, 1H), 7.25 (s, 1H), 7.09 (d, J¼8.4 Hz, 2H), 6.66 (d, J¼
8.5 Hz, 2H), 5.37 (s, 2H) ppm; 13C NMR (100 MHz,
DMSO-d6) d 166.1, 157.8, 153.0, 148.4, 135.1, 129.1,
121.9, 118.5, 113.4, 113.0 ppm; MS (ESI+) m/z 243
(M+H); Anal. Calcd for C12H10N4S: C, 59.48; H, 4.16; N,
23.12. Found: C, 59.60; H, 3.97; N, 22.96.
3.3.5. N-[4-(4-Aminothieno[2,3-d]pyrimidin-5-yl)-
phenyl]-N0-(3-methylphenyl)urea (1). To a solution of an-
iline 6 (45.7 g, 189 mmol) in 450 mL of ethyl alcohol and
900 mL of THF at 0 ꢀC was added 3-methylphenylisocya-
nate (30 mL, 240 mmol) at such a rate to maintain the tem-
perature ꢄ2 ꢀC. The addition required 6 min. Following the
addition, the reaction mixture was stirred in an ice/water
bath for 2 h, then warmed to ambient for 2 h. The reaction
mixture was concentrated, then washed twice with 200 mL
EtOH to remove THF. The product was filtered and washed
with 300 mL EtOH, then dried at 60 ꢀC to yield 66.6 g (94%
yield) of the product as a white solid. Mp¼227–228 ꢀC. 1H
NMR (400 MHz, DMSO-d6) d 2.27 (s, 3H) 6.78 (d, J¼
7.27 Hz, 1H) 7.15 (t, J¼7.75 Hz, 1H) 7.24 (d, J¼8.65 Hz,
1H) 7.30 (s, 1H) 7.37 (d, J¼8.51 Hz, 2H) 7.40 (s, 1H) 7.59
In order to remove sulfur, the solids product were removed
from the vacuum oven and then charged into a flask. Toluene
(28.6 kg) was charged and the slurry was stirred for approx-
imately 1 h and then filtered. The filter cake was rinsed twice
with toluene (5.7 kg) and then dried invacuo at 60 ꢀC to yield
3.04 kg (80.0% isolated yield) of the desired product 4a.
1
Mp¼202–204 ꢀC; H NMR (400 MHz, DMSO-d6) d 8.27
(d, J¼9.1 Hz, 2H), 7.80 (d, J¼8.9 Hz), 7.40 (s, 2H), 6.82
(s, 1H) ppm; 13C NMR (100 MHz, DMSO-d6) d 166.2,

D. M. Barnes et al. / Tetrahedron 62 (2006) 11311–11319
11319
8. The addition of HMDS to AcOH is extremely exothermic
and should be performed carefully. The heat of reaction is
ꢁ535 J/g of HMDS. The calculated adiabatic temperature
rise is 75 ꢀC.
(d, J¼8.65 Hz, 2H) 8.32 (s, 1H) 8.64 (s, 1H) 8.84 (s, 1H)
ppm; 13C NMR (100 MHz, DMSO-d6) d 166.3, 157.7,
153.1, 151.8, 139.4, 138.9, 137.4, 134.2, 128.9, 128.3,
128.1, 122.2, 119.7, 118.3, 117.8, 115.0, 112.7, 21.3 ppm;
Anal. Calcd for C20H12N5OS: C, 63.98; H, 4.56; N, 18.65.
Found: C, 64.24; H, 4.35; N, 18.42.
9. Gewald, K.; Hofmann, I. J. Prakt. Chem. 1969, 311, 402–407;
Gewald, K.; Kleinert, M.; Thiele, B.; Hentschel, M. J. Prakt.
Chem. 1972, 314, 303–314; Gewald, K.; Gruner, M.; Hain,
U.; Sueptitz, G. Monatsh. Chem. 1988, 119, 985–992.
10. Each of the MultiMaxÔ experiments was set up as follows:
individual reactors were charged with 4.0 g of ylidene 3a,
696 mg of S8, and 40 mL of THF. A solution of 1.52 g
(1.0 equiv) of NaHCO3 was dissolved in 20 mL of H2O and
added over the designated time at the designated temperature
and stirring rate.
Acknowledgements
Drs. Steve Wittenberger and Steve King are gratefully
acknowledged for helpful discussions.
References and notes
11. Bhatnagar, S. K. Rubber News 1969, 8, 25–31.
12. The optimized conditions developed for 40-nitroacetophenone
(2a) were employed in these reactions, and were not further
optimized for the other substrates. It is possible that these con-
ditions do not represent the ideal process for other substrates,
in particular for those with a different electronic nature (for
example the 4-methoxy derivatives 2g and 3g).
13. The products were isolated by crystallization. As the crystalli-
zations were not optimized, the sometimes-large difference in
assay yield and isolated yield can be attributed to high loss to
the liquors.
1. Sabnis, R. W.; Rangnekar, D. W.; Sonawane, N. D.
J. Heterocycl. Chem. 1999, 36, 333–345.
2. Gewald, K.; Schinke, E.; Boettcher, H. Chem. Ber. 1966, 99,
94–100.
3. Dai, Y.; Guo, Y.; Frey, R. R.; Ji, Z.; Curtin, M. L.; Ahmed,
A. A.; Albert, D. H.; Arnold, L.; Arries, S. S.; Barlozzari, T.;
Bauch, J. L.; Bouska, J. J.; Bousquet, P. F.; Cunha, G. A.;
Glaser, K. B.; Guo, J.; Li, J.; Marcotte, P. A.; Marsh, K. C.;
Moskey, M. D.; Pease, L. J.; Stewart, K. D.; Stoll, V. S.;
Tapang, P.; Wishart, N.; Davidsen, S. K.; Michaelides, M. R.
J. Med. Chem. 2005, 48, 6066–6083.
14. Kandeel, Z. E.-S. Heteroat. Chem. 1996, 7, 29–33.
15. Freeman, F.; Kim, D. S. H. L. J. Org. Chem. 1991, 56, 4645–
4648 and references therein.
4. Milart, P.; Wilamowski, J.; Sepiol, J. J. Tetrahedron 1998, 54,
15643–15656 and references cited therein.
5. Kozuka, S.; Kitamura, T.; Kobayashi, N.; Ogino, K. Bull.
Chem. Soc. Jpn. 1979, 52, 1950–1952; Bassindale, A. R.;
Stout, T. Tetrahedron Lett. 1985, 26, 3403.
16. For the N-alkylation of anilines by alcohols over Raney nickel,
see: Rice, R. G.; Kohn, E. J. J. Am. Chem. Soc. 1955, 77, 4052–
4054.
6. HOAc has been used as an additive in reactions with azeotropic
removal of water. Buchstaller, H.-P.; Siebert, C. D.; Lyssy,
R. H.; Frank, I.; Duran, A.; Gottschlich, R.; Noe, C. R.
Monatsh. Chem. 2001, 123, 279–293.
7. Mason, P. S.; Smith, E. D. J. Gas Chromatogr. 1966, 4, 398–
400.
17. Schwetlick, K.; Noack, R.; Stebner, F. J. Chem. Soc., Perkin
Trans. 2 1994, 599–608.
18. A third mechanism, in which the alcohol selectively deacylates
N2 of bis-acyl 12, is ruled out because overnight stirring at
ambient temperature did not lower this impurity.
19. Gewald, K. Chem. Ber. 1965, 3571–3577.