Scalable synthesis of a prostaglandin EP4 receptor antagonist

Article abstract of DOI:10.1021/jo1004197

The evolution of scalable, economically viable synthetic approaches to the potent and selective prostaglandin EP4 antagonist 1 is presented. The chromatography-free synthesis of multikilogram quantities of 1 using a seven-step sequence (six in the longest

Full text of DOI:10.1021/jo1004197

pubs.acs.org/joc
Scalable Synthesis of a Prostaglandin EP4 Receptor Antagonist
Danny Gauvreau,* Sarah J. Dolman, Greg Hughes, Paul D. O’Shea, and Ian W. Davies
Department of Process Research, Merck Frosst Centre for Therapeutic Research, 16711 Trans Canada Hwy,
ꢀ
Kirkland, Quebec, Canada H9H 3L1
danny_gauvreau@merck.com
Received March 8, 2010
The evolution of scalable, economically viable synthetic approaches to the potent and selective pro-
staglandin EP4 antagonist 1 is presented. The chromatography-free synthesis of multikilogram
quantities of 1 using a seven-step sequence (six in the longest linear sequence) is described. This
approach has been further modified in an effort to identify a long-term manufacturing route. Our
final synthesis involves no step requiring cryogenic (<-25 °C) conditions; comprises a total of four
steps, only three of which are in the longest linear synthesis; and features the use of two consecutive
iron-catalyzed Friedel-Crafts substitutions.
Introduction
initiative to develop a practical and economically viable
long-term manufacturing route to 1 are also disclosed.
The approach used to prepare 1 in support of medicinal
chemistry studies is outlined in Scheme 1.2 Treatment of 2,5-
dimethylthiophene 2 with NBS afforded dibromothiophene3.
Care had to be taken to protect this reaction from light to
avoid bromination of the methyl substituents. Furthermore,
compound 3 was found to be unstable and has a limited shelf
life. Selective monometal-halogen exchange at -78 °C, follo-
wed by addition of 4-trifluoromethylbenzaldehyde, afforded
benzylic alcohol 4. Treatment of this material with a mixture
of trifluoroacetic acid and triethylsilane afforded the reduced
product 5. A second metal-halogen exchange was conduc-
ted in a 2:1 mixture of diethyl ether and THF at -78 °C,
followed by the addition of carbon dioxide to afford carboxylic
acid 6. To prepare the amine fragment 8, 1,4-dicyanobenzene
(DCB) was subjected to titanium-mediated Kulinkovich cyclo-
propanation conditions developed by Szymoniak and Bertus
to afford a cyclopropylamino-cyanobenzene.3 Hydrolysis and
esterification of the remaining nitrile afforded 8 in 10-15%
As part of an ongoing program aimed at developing EP4
inhibitors for the treatment of disorders associated with
arachidonic acid metabolism,1,2 a number of potent and
selective inhibitors have been identified, among which the
substituted thiophene derivative 1 was found to be particu-
larly promising in preclinical studies.2 As such, a scalable
approach to this compound was required, and the develop-
ment of a chromatography-free synthesis of its diethyl
ammonium salt (1 DEA) on multikilogram scale is described
3
herein. Further improvements made as the result of an
(1) For references on the use of EP4 antagonists for treating inflamma-
tion, see: (a) McCoy, J. M.; Wicks, J. R.; Audoly, L. P. J. Clin. Invest. 2002,
110, 651. For atherosclerosis, see: (b) Cipollone, F.; Fazia, M. L.; Iezzi, A.;
Cuccurullo, C.; De Cesare, D.; Ucchino, S.; Spigonardo, F.; Marchetti, A.;
Buttitta, F.; Paloscia, L.; Mascellanti, M.; Cuccurullo, F.; Mazzetti, A.
Arterioscler., Throm., Vasc. Biol. 2005, 25, 1925. For cancer, see: (c) Yang,
L.; Huang, Y.; Porta, R.; Yanagisawa, K.; Gonzalez, A.; Segi, E.; Johnson,
D. H.; Narumiya, S.; Carbone, D. P. Cancer Res. 2006, 66, 9665. (d) Chell,
S. D.; Witherden, I. R.; Dobson, R. R.; Moorghen, M.; Herman, A. A.;
Qualthrough, D.; Williams, A. C.; Paraskeva, C. Cancer Res. 2006, 66, 3106.
(e) Ma, X.; Kundu, N.; Rifat, S.; Walser, T.; Fulton, A. M. Cancer Res. 2006,
66, 2923.
(3) Szymoniak, J.; Bertus, P. Chem. Commun. 2001, 1792–1793. For the
Kulinkovich cyclopropanation, see: Kulinkovich, O. G.; Sviridov, S. V.;
Vasilevski, D. A.; Pritytskaya, T. S. Zh. Org. Khim. 1989, 25, 2244–2245. For
a recent review, see: Kulinkovich, O. G.; de Meijere, A. Chem. Rev. 2000, 100,
2789–2834.
(2) Blouin, M.; Han, Y.; Burch, J.; Farand, J.; Mellon, C.; Gaudreault,
ꢀ
M.; Wrona, M.; Levesque, J.-F.; Denis, D.; Mathieu, M.-C.; Stocco, R.;
Vigneault, E.; Therien, A.; Clark, P.; Rowland, S.; Xu, D.; O’Neill, G.;
Ducharme, Y.; Friesen, R. J. Med. Chem. 2010, 53 (5), 2227–2238.
4078 J. Org. Chem. 2010, 75, 4078–4085
Published on Web 05/14/2010
DOI: 10.1021/jo1004197
r
2010 American Chemical Society

Gauvreau et al.
JOCArticle
SCHEME 1. Medicinal Chemistry Approach to 1
yield from DCB. The synthesis of 1 was completed by a
HATU-mediated coupling of 6 and 8, followed by hydrolysis
of the ester.
Results and Discussion
Development efforts were initiated by screening a variety
of conditions for effecting the Friedel-Crafts acylation of 2
with an acid chloride derived from 11 (Scheme 2).5 Mild
Lewis acids (ZnCl2, MgCl2, Zn(OTf)2) did not reproducibly
afford complete conversion, even after 18 h at 90 °C. In the
presence of stronger Lewis acids (AlCl3, TiCl4), acylation of
dimethylthiophene proceeded at 0 °C in common solvents
such as dichloromethane or 1,2-dichloroethane. However,
low yields and intractable byproducts were obtained when
the reactions were performed in these solvents, presumably
resulting from participation of these haloalkyl solvents in
Friedel-Crafts alkylations. Switching to chlorobenzene as
a solvent resolved this problem. As such, benzoic acid 11
was treated with oxalyl chloride in chlorobenzene, and then
dimethylthiophene and TiCl4 were added directly to the
reaction mixture, affording 92% yield of ketone 12 in a
one-pot procedure.
To develop a scalable approach to 1, a number of key issues
needed to be addressed: difficulties associated with the bromi-
nation of 2 and storage of 3; the number of cryogenic metal-
halogen exchange/electrophilic quench steps, and the use of
large amounts of a highly flammable solvent (Et2O) in the
formation of 6; poor yields in converting DCB to 8; and the cost
associated with the use of HATU as a coupling reagent. An
alternative retrosynthetic analysis was envisioned that would at
least partially address these concerns (Figure 1). As with the
medicinal chemistry approach to 1, the synthesis would be
completed by coupling 6 and 8, although more cost-effective
alternatives to HATU were desired. Installation of the 4-triflu-
oromethyl benzyl substituent by a Friedel-Crafts acylation/
reduction sequence would significantly improve efficiency and
obviate the preparation and storage of dibromothiophene 3.
Also, literature reports suggested that titanium-mediated cyclo-
propanations could be chemoselective for nitriles in the presence
of ester functional groups.4 This would allow 8 to be prepared
directly from the commercially available nitrile 10 in a single step.
SCHEME 2. Acylation of Dimethylthiophene
Conveniently, 12 could be reduced to 9 by treatment with
NaBH4 followed by Et3SiH and TFA at 90 °C (Scheme 3).
Unfortunately, efforts to introduce a carboxylic acid moiety
with oxalyl chloride or phosgene in the presence of a variety
of Lewis acids (AlCl3, TiCl4, ZnCl2, ZnI2) were unsuccessful.
(4) Szymoniak, J.; Bertus, P. Synlett 2003, 2, 265–267.
(5) Olah, G. A. Friedel-Crafts and Related Reactions; Interscience:
New York, 1965; Vol. III, part 2, pp 1518 -1593.
FIGURE 1. First generation retrosynthetic approach to 1.
J. Org. Chem. Vol. 75, No. 12, 2010 4079

Gauvreau et al.
JOCArticle
In light of this, a more sequential approach to the installation
of this functionality was evaluated. Ketone 12 was unreactive
toward direct bromination by NBS or elemental bromine.
However, in the presence of as little as 1 mol % ZnCl2, clean
conversion to 13 occurred upon treatment with bromine
within 30 min at 0 °C.6 Importantly, bromination of the
benzylic positions was not observed, even in the presence of
excess bromine. This transformation could be carried out on
unpurified ketone 12 to afford 13 in 94% yield. Exhaustive
ketone reduction could be achieved in a single operation by
treating 13 with triethylsilane and TiCl4 in 1,2-dichloroethane.
This reaction was found to be particularly sensitive to residual
chlorobenzene.7 However, if care was taken to remove the last
traces of Ph-Cl from crude 13 by stripping from toluene prior
to solvent switching into 1,2-dichloroethane, yields of >80%
could be consistently achieved.
In the medicinal chemistry approach to 1, the use of a mixed
diethyl ether and tetrahydrofuran solvent system was found
to be critical for a successful metal-halogen exchange of 5.
The resulting lithium anion was found to be unstable in THF
alone unless the reaction was performed at temperature
below -80 °C, which is difficult to achieve on multikilogram
scale. In other acyclic ethereal solvents such as tert-butyl-
methylether (TBME), very little metal-halogen exchange was
seen even at higher temperatures. The addition of 1.1 equiv
of TMEDA allowed for smooth metal-halogen exchange at
-65 °C. The anion formed under these conditions was found
to be stable up to -55 °C, although significant amounts of
the quenched anion product 9 were observed upon warming
above -55 °C. Alternative metal-halogen exchange strate-
gies (Grignard, Bu3MgLi) were also found to be ineffective.8
Quenching the anion by bubbling CO2 (gas) into the reaction
mixture at a rate such that the internal temperature did not
exceed -55 °C allowed carboxylic acid 6 to be formed in
79% yield.
decided to investigate the preparation of amine 8 from com-
mercially available methyl 4-cyanobenzoate 10. Using the
modified conditions reported by de Meijere (MeTi(OiPr)3,
Et2Zn, LiI or LiOiPr) failed to afford amine 8.9 Treating
nitrile 10 with Ti(OiPr)4, EtMgBr, and BF3 OEt2 at low
3
temperature (-78 to -40 °C) also failed to afford any desired
cyclopropylamine. However, at -25 °C, 94% conversion of
the starting nitrile after 1 h was observed, with a 43% assayed
yield of the desired cyclopropylamine (Scheme 4). Modifica-
tion of solvent, ratio of reagents, order of addition, and
replacement of BF3 OEt2 by lithium salts all failed to improve
3
upon the modest yield. However, this procedure did provide a
high degree of chemoselectivity with no cyclopropyl alcohol
being observed. To purify 8, the material was extracted into 3 N
HCl to remove nonbasic impurities. The HCl salt of 8 could
then be extracted into 2-methyltetrahydrofuran to separate the
desired material from more polar byproducts. A final purity
upgrade was achieved by preparation of the methanesulfonic
acid salt of 8, followed by salt break, to recover 92% of amine 8
in greater than 95% purity as judged by HPLC analysis.
SCHEME 4. Preparation of Cyclopropylamine 8 from Methyl
4-Cyanobenzoate
The synthesis of 1 was completed by a one-pot amidation-
hydrolysis procedure (Scheme 5). Thiophene acid 6 was con-
verted in situ to the desired acid chloride with oxalyl chloride
and catalytic DMF. The crude acid chloride was coupled
with 8 to afford an amide, which was directly saponified to
acid 1. The optimal solvent mixture for the hydrolysis was a
3:1 mixture of THF/MeOH, which allowed for complete
hydrolysis of the methyl ester within 2 h at 50 °C.10 Finally,
compound 1 was purified by crystallization of its diethyl-
amine salt from THF/TBME to afford the final compound
(>98% pure by HPLC analysis) in 81% yield from thiophene
SCHEME 3. Preparation of Acid 6 from Thiophene Ketone 12
(6) (a) Olah, G. A.; Kuhn, S. J.; Flood, S. H.; Hardie, B. A. J. Am. Chem.
Soc. 1964, 86 (6), 1039–1044. (b) Olah, G. A.; Kuhn, S. J.; Flood, S. H.;
Hardie, B. A. J. Am. Chem. Soc. 1964, 86 (6), 1044–1046. (c) Olah, G. A.;
Kuhn, S. J.; Hardie, B. A. J. Am. Chem. Soc. 1964, 86 (6), 1055–1060.
(d) Gnanapragasam, N. S.; Joseph, N. Curr. Sci. 1971, 40 (4), 83.
(e) Gnanapragasam, N. S.; Ramanujam, R.; Srinivasan, S. P. Curr. Sci.
1976, 45 (24), 862–863. (f) Clark, J. H.; Ross, J. C.; Macquarrie, D. J.;
Barlow, S. J.; Bastock, T. W. Chem. Commun. 1997, 1203–1204. (g) Ross,
J. C.; Clark, J. H.; Macquarrie, D. J.; Barlow, S. J.; Bastock, T. W. Org. Proc.
Res. Dev. 1998, 2 (4), 245–249. (h) Bastock, T. W.; Trenbirth, B.; Clark, J. H.;
Ross, J. Eur. Pat. Appl. EP 866046-A1, 1998.
(7) For reasons that are still unclear, with as little as 5 wt % chlorobenzene
in the solvent mixture, the yields of the reaction dropped to <40%, via
decomposition, presumably polymerization, as observed by HPLC.
(8) (a)Grignard, V. Bull. Soc. Chim. 1910, 7, 453. (b) Knochel, P.; Dohle, W.;
Gommermann, N.; Kneisel, F. F.; Kopp, F.; Korn, T.; Sapountzis, I.; Vu, V. A.
Angew. Chem., Int. Ed. 2003, 42, 4302–4320. (c) Abarbri, M.; Dehmel, F.;
Knochel, P. Tetrahedron Lett. 1999, 40, 7449–7453. (d) Kitagawa, K.; Inoue, A.;
Shinokubo, H.; Oshima, K. Angew. Chem., Int. Ed. 2000, 39 (14), 2481–2483.
(e)Dolman, S. J.;Gosselin, F.;O’Shea, P. D.;Davies, I. W.Tetrahedron2006, 62
(21), 5092–5098.
With the thiophene acid 6 in hand, an improved approach
to preparing the cyclopropylamine fragment 8 was thought.
Efforts to improve the synthesis of 8 from DCB failed to
afford more then a 15% yield of the required amine. As chemo-
selective titanium-mediated nitrile-cyclopropanations had
previously been demonstrated in the presence of an ester,4 we
(9) De Meijere, A.; Wiedemann, S.; Frank, D.; Winsel, H. Org. Lett. 2003,
5, 753–755.
(10) A 5:1 ratio of THF/MeOH required 7 h to achieve complete conversion.
4080 J. Org. Chem. Vol. 75, No. 12, 2010

Gauvreau et al.
JOCArticle
acid 6. Overall, this first large-scale synthesis of the diethyl-
amine salt of EP4 antagonist 1 was performed in six steps and
21% overall yield from (4-trifluoromethyl)benzoic acid 11,
producing 3.7 kg of the 1 DEA salt.
3
original conditions (H2[PtCl6], dioxane, 120 °C), no aryla-
tion product was observed for either 4-(trifluoromethyl)-
benzyl alcohol or its related acetate (Table 1, entries 1, 2).11a
When IrCl3 was used as catalyst,11a 81% conversion to desired
compound 9 was observed when trifluoromethylbenzylic
alcohol was used, but no product was obtained with the acetate
analogue (entries 3, 4). Conditions developed by Rueping
(Bi(OTf)3, DCE) afforded promising results with both the
benzylic alcohol and acetate (entries 5, 6; 74% and 36%
conversion, respectively).11b However, because of the high
cost of both IrCl3 and Bi(OTf)3 and the low yields observed,
we investigated less expensive catalysts such as FeCl3 and
ZnCl2.11c While no product was observed with 40 mol %
ZnCl2, we were pleased to observe modest reactivity with
20 mol % FeCl3 with both the benzylic alcohol and acetate
(entries 7, 8; 67% and 33% conversion, respectively). By
increasing the catalyst loadings from 20 to 45 mol %, com-
plete conversion was obtained within 16 h at 75 °C (entry 9).
After screening for additives, three acids (H3PO4, DL-tartaric
acid, and MsOH) were identified as viable co-catalysts,
whereas bases (Na2CO3, Et3N) were found to completely
suppress reactivity. Addition of 30 mol % MsOH allowed for
94% conversion with just 20 mol % FeCl3 (entry 10). The
impact of MsOH was even more pronounced for ZnCl2
as 61% conversion to 9 was observed in the presence of the
acid, while no desired product was observed in its absence
(entries 11, 12). Optimal conditions for this transformation
were identified as 2 equiv of 2,5-dimethylthiophene, with
40 mol % FeCl3 and 40 mol % MsOH in DCE at 55 °C. After
16 h, complete conversion of the benzylic alcohol was
observed with 70% assayed yield of 9. During the prepara-
tion of this manuscript, reports regarding the benzylation of
arenes using FeCl3 on benzyl ethers were published.12 These
conditions have not been tested.
SCHEME 5. One-Pot Amidation/Hydrolysis Sequence for the
Synthesis of 1
While the chemistry described above was acceptable for
the support of preclinical activities, there were opportunities
for improvement with regard to the development of a long-
term manufacturing route to 1. For instance, it might be possi-
ble to form the amide bond through a palladium-catalyzed
3-component coupling between thiophene bromide 5, cyclo-
propylamine 8, and carbon monoxide. Furthermore, the
substitution of the 2,5-dimethylthiophene by direct alkyla-
tion with an appropriately activated 4-trifluoromethylbenzyl
synthon would remove the need for the exhaustive ketone
reduction (Figure 2).
Thiophene 9 was readily brominated under the same
conditions as used in the first synthesis (vide supra) to afford
bromothiophene 5 in 93% yield (Scheme 6).13 The three-
component coupling was then briefly investigated. Using
conditions developed for the carbonylation of haloindoles
(PdCl2(PhCN)2, dppf, CO 300 psi, Et3N, 140 °C),14 52%
conversion to the amide 15 was observed in an unoptimized
reaction. The methyl ester was hydrolyzed as in the previous
SCHEME 6. Summary of Second Process Research Route to 1
FIGURE 2. Second retrosynthesis: three-component coupling
approach to 1.
Recent work by Beller and Rueping has shown that transi-
tion metals catalyze the alkylation of arenes by benzylic alco-
hols and acetates in good yields with low catalyst loading.11
However, all reported examples employed electron-rich
or -neutral alkylating agents coupled with a large excess
(3 to >100 equiv) of arene. To the best of our knowledge, the
alkylation of arenes by electron-deficient benzylic alcohols
or acetates, such as those bearing a trifluoromethyl substi-
tuent, has not been demonstrated. Furthermore, a significant
reduction in the excess of arene required for complete con-
version would be necessary in order for this transformation
to be economically viable on large scale. Using Beller’s
(11) (a)Beller, M.;Mertins, K.;Iovel,I.;Kischel, J.;Zapf, A. Angew.Chem.,
Int. Ed. 2005, 44, 238–242. (b) Rueping, M.; Nachtsheim, B. J.; Ieawsuwan, W.
Adv. Synth. Catal. 2006, 348, 1033–1037. (c) Beller, M.; Iovel, I.; Mertins, K.;
Kischel, J.; Zapf, A. Angew. Chem., Int. Ed. 2005, 44, 3913–3917.
J. Org. Chem. Vol. 75, No. 12, 2010 4081

Gauvreau et al.
JOCArticle
TABLE 1. Transition-Metal-Catalyzed Benzylation of 2,5-Dimethylthiophenea
entry
catalyst
additive
none
none
none
none
none
none
none
none
14a/b
temp (°C)
conversionb (%)
assay yield (%)
1
2
3
4
5
6
7
8
H2[PtCl6] (20%)c
H2[PtCl6] (20%)c
IrCl3 (20%)
a
b
a
b
a
b
a
b
a
a
a
a
75
75
75
75
75
75
75
75
75
50
70
70
0
0
81
0
74
36
67
33
99
94
0
0
0
36
0
40
22
nd
nd
62
67
0
IrCl3 (20%)
Bi(OTf)3 (20%)
Bi(OTf)3 (20%)
FeCl3 (20%)
FeCl3 (20%)
9
FeCl3 (45%)d
FeCl3 (20%)d
ZnCl2 (40%)d
ZnCl2 (40%)d
none
10
11
12
MsOH (30%)e
None
MsOH (20%)
61
nd
aAll reactions were carried out in DCE for a period of 16 h using 3 equiv of 2,5-dimethylthiophene. bConversions were measured by HPLC at 215 nm.
cReaction run in 1,4-dioxane. dReaction performed with only 2 equiv of 2,5-dimethylthiophene. eNo reaction occurred in the absence of the FeCl3 catalyst.
synthesis to afford 1 in 88% yield. By utilizing the three-
component coupling, in addition to direct alkylation of
2,5-dimethylthiophene, we were able to reduce the number
of steps in the longest linear sequence from six to four steps in
addition to avoiding cryogenic conditions.
While this approach was more efficient, the use of palladium
in the penultimate step could represent a problem, as its removal
is often tedious. With a straightforward preparation of the
thiophene fragment 9, a direct Friedel-Crafts amidation be-
tween this arene and isocyanate 16 would represent an expedient
approach to 1 (Figure 3). Amine 8 was readily and cleanly
converted to 16 using phosgene and Et3N. The isocyanate was
used without further purification in the Friedel-Crafts amida-
tion, as it was found to decompose rapidly on silica gel.
byproducts, 1.1 equiv of FeCl3 afforded 67% yield within
30 min at 50 °C. It should be noted that the product was
unstable to reaction conditions for prolonged periods; aging the
reaction mixture for 16 h caused a 17% decrease in yield (50%
assayed yield). It was therefore necessary to quench the reaction
as soon as sufficient conversion was achieved. Conveniently,
this amidation can be performed with crude thiophene 9 and
crude isocyanate 16 to afford comparable yields.15 The synthesis
was completed by LiOH-mediated saponification.
This final route to EP4 antagonist 1 affords 41% overall
yield, in just three linear steps, from commercially available
4-(trifluoromethyl)benzyl alcohol 14a (Scheme 7). Impor-
tantly, this approach relies only on nontoxic and inexpensive
transition metals, and none of the transformations require
reaction temperatures below -25 °C.
Conclusion
In summary, three alternatives to the original synthesis of
potent and selective EP4 antagonist 1 have been developed.
The first approach, which featured a stepwise preparation of
carboxylic acid 6, allowed for the preparation of multikilo-
gram quantities of 1. The second more efficient synthesis
involved an iron-catalyzed benzylation of 2,5-dimethylthio-
phene followed by a palladium-catalyzed three-component
coupling, which reduced the longest linear sequence to four
steps. Finally, an iron-catalyzed Friedel-Crafts amidation
of thiophene fragment 9 with isocyanate 16 allowed for the
preparation of the EP4 antagonist 1 with only three steps in
the longest linear sequence. This final approach involves
only nontoxic transition metals and no cryogenic steps.
FIGURE 3. Third retrosynthesis: Friedel-Crafts amidation
approach to 1.
The investigation of a variety of Lewis acids for the Friedel-
Crafts amidation of 9 with isocyanate 16 revealed that tita-
nium and boron reagents such as Ti(OiPr)4, TiCl4, and
BF3 Et2O afforded <30% conversion. Better results were
3
achieved with aluminum and iron chloride salts (AlCl3,
FeCl3), which catalyzed the transformation to 93-99%
conversion. While 2 equiv of AlCl3 was required to afford
only 34% yield of amide 15 contaminated by intractable
Experimental Section
(2,5-Dimethyl-3-thienyl)[4-(trifluoromethyl)phenyl]methanone
(12). A visually clean, 100-L 5-neck round-bottom flask connected
(12) Wang, B.-Q.; Xiang, S.-K.; Sun, Z.-P.; Guan, B.-T.; Hu, P.; Zhao,
K.-Q.; Shi, Z.-J. Tetrahedron Lett. 2008, 49, 4310–4312.
(13) Conditions used to brominate 9.
(14) Beller, M.; Kumar, K.; Zapf, A.; Michalik, D.; Tillack, A.; Heinrich,
T.; Bottcher, H.; Arlt, M. Org. Lett. 2004, 6, 7–10.
(15) Performing the Friedel-Crafts alkylation of 2,5-dimethylthiophene
with isocyanate 16 followed by arylation with benzylic alcohol 14a to afford
15 failed as a result of low conversion in the second reaction.
4082 J. Org. Chem. Vol. 75, No. 12, 2010

Gauvreau et al.
JOCArticle
SCHEME 7. Final Route to EP4 Antagonist 1
to a scrubber filled with 5 N NaOH (20 L) was charged with
chlorobenzene (45 L), benzoic acid 11 (5.96 kg, 31.4 mol), and
DMF (10 mL). Oxalyl chloride (2.87 L, 32.9 mol) was added
over 45 min. The mixture was heated with a steam bath until the
internal temperature reached 50 °C. The reaction maintained an
internal temperature of 45-50 °C. After 1 h, the cloudy reaction
mixture was assayed by HPLC of an aliquot, which indicated
96% conversion of acid 11 to the corresponding acid chloride.
After the internal temperature had dropped to 22 °C, dimethylthio-
phene (3.25 L, 28.5 mol) was added to the reactor at once, followed
by titanium(IV) chloride (3.44 L, 31.4 mol) over 1 h via the
addition funnel. The crude reaction mixture was transferred into
a visually clean 160-L extractor charged with 1 N HCl (60 L). After
vigorous stirring for 5 min, the phases were allowed to separate.
The organic layer (bottom) was removed, and the aqueous layer
was back-extracted with heptane (40 L). The organic phases were
combined and washed with half-brine (20 L), and then the solvent
volume was reduced in vacuo to afford a thin brown oil. HPLC
analysis determined the material to be 52.77 wt % ketone 12, or
6.49 min. HRMS calcd for C14H10BrF3OS: 361.9661, obsd
[MþH] 362.9667.
3-Bromo-2,5-dimethyl-4-[4-(trifluoromethyl)benzyl]thiophene (5).
A visually clean, 5-L 3-neck round-bottom flask was charged
with bromoketone 13 (0.43 kg, 86 wt %, 1.20 mol), triethylsilane
(0.48 L, 2.98 mol), and dichloroethane (2.2 L) and then cooled
until the internal temperature reached -8 °C. Titanium(IV)
chloride (0.13 L, 1.20 mol) was added over 1 h, keeping the
internal temperature below 14 °C. The crude reaction mixture
was transferred into a visually clean 5-L extractor charged with
1 N HCl (2 L). After vigorous stirring for 5 min, the phases were
allowed to separate. The organic (bottom) layer was removed,
and the aqueous layer was back-extracted with heptane (1 L).
The organic phases were combined and washed with half-brine
(1 L). The solvent was removed under vacuum to afford 800 g
thick brown oil. HPLC analysis determined the material to be
48 wt % bromoalkane 5, or 0.384 kg, a 88% assay yield. 1H NMR
(400 MHz, dmso-d6): δ 7.69-7.59 (m, 1 H), 7.31 (d, J=8.02 Hz,
1 H), 3.98 (s, 1 H), 2.32 (d, J=20.73 Hz, 3 H). 13CNMR(100MHz,
acetone-d6): δ 145.1, 134.5, 133.1, 131.1, 128.6 (q, J=31.9 Hz),
126.0 (q, J=3.7 Hz), 125.9, 124.4 (q, J=269.4 Hz), 112.2, 34.1,
14.9, 13.9. 19F NMR (375 MHz, acetone-d6): δ -62.9. IR (NaCl
plates): 3065.3 (w), 2921.6 (m), 2857.5 (w), 1617.2 (s), 1417.4
(m), 1326.5 (s). HPLC ret time: 7.41 min. HRMS calcd for
C14H12BrF3S: 347.9795, obsd [M þ H] 348.9821.
1
8.24 kg, a 92.4% assay yield. H NMR (400 MHz, acetone-d6):
δ 7.95-7.83 (m, 4 H), 6.83 (s, 1 H), 2.56 (s, 3 H), 2.45-2.32 (m,
3 H). 13C NMR (100 MHz, acetone-d6): δ 190.5, 147.7, 143.7,
136.4, 136.1, 130.3, 128.3, 126.2, 15.1, 14.6. 19F NMR (375 MHz,
acetone-d6): δ -62.9. IR (NaCl plate): 3061.1 (m), 2923.5 (m),
2862.1 (w), 1651.8 (s), 1556.5 (m), 1480.3 (s), 1327.5 (s). HPLC ret
time: 5.98 min. HRMS: calcd for C14H11F3OS 284.0483, obsd
[M þ H] 285.0556.
2,5-Dimethyl-4-[4-(trifluoromethyl)benzyl]thiophene-3-carboxylic
Acid (6). A visually clean, 50-L 5-neck round-bottom flask was
charged with bromoalkane 5 (4.00 kg, 37.6 wt %, 4.31 mol),
tetramethylethylenediamine (711 mL, 4.74 mol), and MTBE
(20 L) and then cooled until the internal temperature reached
-65 °C. n-BuLi (2.24 L, 2.5M, 5.60 mol) was added over 1 h,
keeping the internal temperature below -55 °C. The mixture
was stirred for 0.5 h, and then CO2 gas (∼ 300 g) was bubbled
into the reaction mixture over 1.5 h, again keeping the internal
temperature below -55 °C. After 1.5 h, HPLC analysis indi-
cated ∼85% CO2 incorporation (vs reduction). HCl (1 N, 13 L)
was charged directly to the reactor. The biphasic solution was
transferred into a visually clean 100-L extractor, and the phases
were allowed to separate. The aqueous layer was removed and
the organic layer collected. The aqueous layer was back-extracted
with MTBE (6 L). The organic phases were combined and
treated with 0.5 N KOH (13.0 L), and the aqueous layer collec-
ted. The organic phase was then back-extracted with 0.5 N
KOH (6.5 L). The combined aqueous layers were returned to
the extractor, which was also charged with MTBE (23 L). The
biphasic solution was acidified by addition of 6 N HCl (1.25 L)
until pH ∼1. The layers were allowed to separate, and the
organic layer was collected and then washed with half-brine
(13 L). The crude organic material was concentrated in vacuo,
flushing with heptane (10 L) to afford a yellow solid (∼4.5 kg).
(4-Bromo-2,5-dimethyl-3-thienyl)[4-(trifluoromethyl)phenyl]-
methanone (13). A visually clean, 100-L 5-neck round-bottom
flask connected to a scrubber filled with 5 N NaOH (20 L) was
charged with ketone 12 (13.27 kg, 52.7 wt %, 24.7 mol),
chlorobenzene (33 L), and zinc chloride (33.6 g, 0.25 mol) and
then cooled until the internal temperature reached 16 °C.
Bromine (3.94 kg, 24.7 mol) was added over 1 h, and the mixture
was aged for 15 min. The crude reaction mixture was transferred
into a visually clean 160-L extractor charged with 1 N HCl (45 L).
After vigorous stirring for 5 min, the phases were allowed to
separate. The organic (bottom) layer was removed, and the
aqueous layer back-extracted with heptane (25 L). The organic
phases were combined and washed with half-brine (20 L). The
solvent was then evaporated under reduced pressure to afford a
thin brown oil. HPLC analysis determined the material to be
80.0 wt % bromoketone 13, or 8.35 kg, a 93.6% assay yield.
1H NMR (400 MHz, acetone-d6): δ 7.98 (d, J = 8.13 Hz, 2 H),
7.90 (d, J=8.27 Hz, 2 H), 2.38 (s, 3 H), 2.35 (s, 3 H). 13C NMR
(100 MHz, acetone-d6): δ 191.9, 141.3, 140.4, 137.2, 134.5 (q,
J=36 Hz) 133.4, 130.9, 128.7 (q, J=269.4 Hz), 126.6 (q, J=4.7
Hz), 107.3, 14.4, 14.3. 19F NMR (375 MHz, acetone-d6): δ
-63.1. IR (NaCl plate): 3073.3 (w), 2922.6 (m), 2857.5 (w),
1670.3 (s), 1540.0 (s), 1484.6 (m), 1321.1 (s). HPLC ret time:
J. Org. Chem. Vol. 75, No. 12, 2010 4083

Gauvreau et al.
JOCArticle
The crude solid was charged to a visually clean, 50-L round-
bottom flask, suspended with heptane (23 L), and then cooled
via an ice/acetone bath until the internal temperature reached
2 °C. The slurry was stirred for 6 h and then filtered over a glass
frit, washing with cold heptane (1.25 L). The filter cake was
dried via house vacuum under nitrogen overnight and then
oven-dried at 50 °C for 24 h. A total of 1.22 kg of dry yellow
solid was collected. HPLC analysis indicated the material
and excess oxalyl chloride were removed under reduced pres-
sure. The residue was treated with THF (27 L, 10 mL/g) and
Hunig’s base (2.24 L, 1.50 equiv) and then cooled to 3 °C.
Cyclopropylamine 9 (1.88 kg, 1.15 equiv) was added to the
solution as a THF solution (5 L, 2 mL/g) over a period of 30 min.
The mixture was aged 30 min (99.8% conversion). To the
solution were added MeOH (4 mL/g, 10.7 L) and 4 N LiOH
(7.47 L, 3.5 equiv). The mixture was heated to 55 °C for 1.5 h and
then recooled to 22 °C. The reaction was quenched by the
addition of 2 N HCl (19 L, 7 mL/g). Organic solvents were
removed under reduced pressure, and the residue dissolved in
Me-THF (54 L, 20 mL/g). The biphasic mixture was transferred
to a 120-L extractor, and the layers were separated. The aqueous
layer was back-extracted using Me-THF (13 L, 5 mL/g). The
combined organic layers were washed with water (13 L, 5 mL/g);
the assay yield of the acid 1 showed to be 88.0% (3.56 kg). 1H
NMR (400 MHz, DMSO-d6): δ 12.75 (s, 1 H); 8.82 (s, 1 H); 7.71
(d, J=8.20 Hz, 2 H); 7.57 (d, J=8.00 Hz, 2 H); 7.24 (d, J=8.00
Hz, 2 H); 7.06 (d, J=8.20 Hz, 2 H); 4.00 (s, 2 H); 2.39 (s, 3 H);
2.27 (s, 3 H); 1.23-1.18 (m, 2 H); 1.08-1.04 (m, 2 H). 13C NMR
(126 MHz, DMSO-d6): δ 167.1 165.8 148.6 145.2 135.4 134.6
133.7 131.4 129.0 128.8 128.0 126.6 (q, J=31.5 Hz) 125.0 (q, J=
1
to be 87 wt % acid 6, 79% assay yield. H NMR (400 MHz,
acetone-d6): δ 7.55 (d, J=8.02 Hz, 1 H), 7.30 (d, J=7.97 Hz,
1 H), 4.29 (s, 1 H), 2.71-2.49 (m, 2 H). 13C NMR (100 MHz,
acetone-d6): δ 165.3, 146.7, 145.6, 136.7, 136.6, 132.4, 129.4,
125.7, 125.6, 33.3, 16.0, 12.8. 19F NMR (375 MHz, acetone-
d6): δ -62.2. IR (KBr pellet): 3023.5 (w), 2925.9 (m), 1674.0 (s),
1554.6 (w), 1480.4 (m), 1328.5 (s). HPLC ret time: 5.04 min.
HRMS calcd for C15H13F3O2S: 314.0588, obsd [M þ H]
315.0658.
Methyl 4-(1-Aminocyclopropyl)benzoate (8). A visually clean
100-L 5-neck round-bottom flask was charged with methyl 4-cyano-
benzoate 10 (2.60 kg, 1.00 equiv) and toluene (40 L, 15 mL/g).
The mixture was cooled to -25 °C, and then Ti(OiPr)4 (4.73 L,
1.00 equiv) was added to the solution over 5 min. Ethylmagne-
sium bromide (10.5 L, 2.0 equiv) was added over a period of
2 h. The mixture was aged at -20 °C for 30 min, and then
borontrifluoride diethyl ether (4.09 L, 2 equiv) was added over
40 min. The mixture was aged at -20 °C for 30 min, after which
HPLC showed 93% conversion. The reaction was quenched by
addition of 3 N HCl (40 L). The biphasic mixture was trans-
ferred to a 100-L extractor, and the layers were allowed to
separate. The aqueous layer was washed with toluene (13 L,
5 mL/g) and then aqueous extracted with 2-methyltetrahydro-
furan (Me-THF) (2 ꢀ 26 L) and 2 ꢀ 5 mL/g (2 ꢀ 13 L). The
organic extracts were combined and washed with 3 N NaOH
(26 L, 10 mL/g) and then brine (13 L, 5 mL/g). The assay yield of
the cyclopropylamine 8 was determined by HPLC and shown to
be 43.2% (1.334 kg). 1H NMR (400 MHz, CDCl3): δ 7.93 (d, J=
8.28 Hz, 2 H); 7.33-7.20 (d, J=8.28 Hz, 2 H); 3.86 (s, 3 H); 2.26
(s, 2 H); 1.17-1.10 (m, 2 H); 1.04-0.99 (m, 2 H). 13C NMR (101
MHz, CDCl3): δ 166.9 152.2 129.6 127.6 124.7 51.9 36.4 19.2.
IR: 3392.9 (w) 3325.0 (w) 3013.8 (w) 2955.2 (w) 1710 (s) 1609.6
(s) 1281.5 (s) 1107.8 (s). HRMS calcd for C11H13NO2: 191.09463,
obsd [M þ H] 192.10257. Melting point: 49.6 -50.4 °C.
3.7 Hz) 124.6 124.4 (q, J=272.0 Hz) 34.3 31.5 18.8 13.9 12.5. 19
F
NMR (377 MHz, DMSO-d6): δ -66.9. IR 3291.9 (bw) 2924.6
(w) 1696.3 (m) 1630.2 (m) 1325.8 (s) 1116.7 (m). HRMS calcd for
C25H22NO3SF3: 473.12725, obsd [M þ H] 474.13494. Melting
point: 254.1 -256.0 °C.
4-{1-[({2,5-Dimethyl-4-[4-(trifluoromethyl)benzyl]-3-thienyl}-
carbonyl)amino]cyclopropyl}benzoic Acid DEA (1 DEA). A vis-
ually clean 100-L 5-neck round-bottom flask was charged with
3
3
acid 1 (3.54 kg, 7.48 mol) and THF (21 L, 6 mL/g). Et2NH (1.18 L,
1.52 equiv) was added to the suspension. Acid 1 DEA salt seeds
3
(30.0 g) were added, and the salt crystallized out. MTBE (25 L)
was added over 2 h. The suspension was aged 13 h at room
temperature. The mixture was cooled to 3 °C, and MTBE (27 L,
8 mL/g) was added over 2 h. The suspension was aged for 1.5 h
and then filtered. The cake was rinsed with 1 ꢀ 7 L MTBE/THF
(2:1) and 2 ꢀ 7 L MTBE. The cake was dried on the frit for 62 h
under nitrogen and then dried in a vacuum oven at 60 °C for
20 h. The yield of 1 DEA was 3.76 kg (92%) as a beige solid.
3
1
The purity of the material by HPLC was 97.8 A%. H NMR
(500 MHz, DMSO-d6): δ 8.80 (s, 1 H); 7.68 (d, J=8.16 Hz, 2 H);
7.56 (d, J=8.00 Hz, 2 H); 7.24 (d, J=8.00 Hz, 2 H); 7.02 (d, J=
8.16 Hz, 2 H); 3.98 (s, 2 H); 3.33 (s, 2 H); 2.74 (q, J = 7.18 Hz,
4 H); 2.38 (s, 3 H); 2.26 (s, 3 H); 1.18-1.13 (m, 2 H); 1.09 (t, J=
7.18 Hz, 6 H); 1.03-0.96 (m, 2 H). 13C NMR (126 MHz,
DMSO-d6): δ 169.3 165.7 145.2 145.1 135.8 134.9 134.3 133.8
131.2 128.8 128.7 126.6 (q, J = 31.5 Hz) 125.0 (q, J = 3.3 Hz)
124.4 (q, J=271.6 Hz) 123.9 41.2 34.2 31.6 18.2 13.8 12.6 11.7.
19F NMR (377 MHz, C6D6): δ -66.9. IR: 3219.4 (bw) 2985.6
(w) 2922.9 (s) 1627.0 (m) 1522.4 (m) 1377.4 (s) 1323.9 (s) 1127.9
(m) 1066.4 (m). HRMS calcd for C25H22NO3SF3: 473.12725,
obsd [M þ H - Et2NH] 474.13427. Melting point: 254.1 -256.0 °C
(loss of Et2NH at 170 °C).
2,5-Dimethyl-3-[4-(trifluoromethyl)benzyl]thiophene (9). A vis-
ually clean 25-mL round-bottom flask was charged with 4-tri-
fluoromethylbenzyl alcohol 14a (257 mg, 1.46 mmol) and DCE
(1.2 mL). To the solution were added 2,5-dimethylthiophene 2
(0.333 mL, 2.92 mmol), MsOH (38 μL, 0.584 mmol), and FeCl3
(95 mg, 0.584 mmol). The mixture was put under nitrogen
atmosphere and heated to 55 °C. The mixture was aged at this
temperature for 16 h. The mixture was poured onto a saturated
solution of NH4Cl (20 mL) and diluted with MTBE (20 mL). The
aqueous layer was separated and then back-extracted with
MTBE (10 mL). Combined organic layers were washed with
brine (10 mL), dried with MgSO4, filtered, and concentrated
under reduced pressure. Assay yield of 9: 278 mg (1.028 mmol),
70%. 1H NMR (400 MHz, acetone-d6): δ 7.61 (d, J = 8.00 Hz,
Methyl 4-(1-Aminocyclopropyl)benzoate MsOH Salt (8 MsOH).
3
3
A visually clean 100-L 5-neck round-bottom flask was charged
with the cyclopropylamine 8 (2.63 kg, 1.00 equiv) and THF
(32 L, 12 mL/g). To the solution was added the MsOH (1.00 L,
1.12 equiv) as a THF (4.0 L, 1.5 mL/g) solution over a period of
2 h. The suspension was stirred at room temperature for 15 h.
The suspension was filtered, rinsed twice with cold THF (2 ꢀ
8 L, 2 ꢀ 3 mL/g), and then dried on the frit for 3 h and then in a
vacuum oven first at 50 °C for a period of 60 h. The yield of
material obtained was 3.93 kg, which was 94.4 wt % (yield =
92.9%). 1H NMR (400 MHz, DMSO-d6): δ 8.78 (s, 3 H); 7.96 (d,
J=8.24 Hz, 2 H); 7.50 (d, J=8.24 Hz, 2 H); 3.84 (s, 3 H); 2.35 (s,
3 H); 1.46-1.39 (m, 2 H); 1.33-1.22 (m, 2 H). 13C NMR (101
MHz, DMSO-d6): δ 165.8 143.3 129.3 128.8 126.4 52.3 39.7 35.6
13.9. IR: 3017.4 (s) 2959.7 (s) 2721.7 (m) 1710.0 (s) 1612.8 (s)
1546.4 (s) 1435.5 (s) 1290.6 (s). HRMS calcd for C11H13NO2:
191.09463, obsd [M þ H] 192.10424. Melting point: 229.2 -
230.7 °C.
4-{1-[({2,5-Dimethyl-4-[4-(trifluoromethyl)benzyl]-3-thienyl}-
carbonyl)amino]cyclopropyl}benzoic Acid (1). A visually clean
100-L 5-neck round-bottom flask equipped with a NaOH scrub-
ber was charged with thiophene acid 6 (2.95 kg at 91 wt % =
2.68 kg, 1.00 equiv), THF (16 L, 6 mL/g), and DMF (6.64 mL,
1 mol %). Oxalyl chloride (897 mL, 1.20 equiv) was added to
the solution over a period of 30 min at room temperature. The
mixture was aged at room temperature for 2 h, and then solvent
4084 J. Org. Chem. Vol. 75, No. 12, 2010

Gauvreau et al.
JOCArticle
2 H); 7.40 (d, J=8.00 Hz, 2 H); 6.45 (s, 1 H); 3.91 (s, 2 H); 2.32 (s,
6 H). 13C NMR (101 MHz, acetone-d6): δ 146.8 136.1 135.9 132.1
129.9 128.5 (q, J=31.8 Hz) 128.2 126.1 (q, J=3.8 Hz) 125.5 (q,
J=271.1 Hz) 34.4 15.0 12.9. 19F NMR (377 MHz, acetone-d6): δ
-66.9. IR: 2922.0 (w) 2861.9 (w) 1617.3 (m) 1324.0 (s) 1161.9 (m)
1123.3 (m) 1065.4 (m). HRMS calcd for C14H13SF3: 270.06901,
obsd [M þ H] 271.07628.
mixture was put under nitrogen atmosphere and then warmed to
50 °C for 15 min. The reaction mixture was then poured onto a
saturated solution of NH4Cl (20 mL) and diluted with Me-THF
(30 mL). The layers were separated, and the aqueous layer was
back-extracted with Me-THF (20 mL). Combined organic layers
were washed with brine (20 mL), dried with MgSO4, filtered,
and concentrated. Assay yield of 15: 90 mg (0.185 mmol), 67%.
By three-component-coupling amidation: In a hydrogenation bomb,
the thiophene bromide 5 (75 mg, 0.215 mmol) was dissolved in
anisole (3 mL). The cyclopropylamine 8 (62 mg, 0.322 mmol) was
added to the solution, followed by the addition of dppf (54 mg,
0.097 mmol), bis(benzonitrile)palladium(II) chloride (12 mg,
32 mmol), and Et3N (45 μL, 0.322 mmol). The mixture was put
under a carbon monoxide atmosphere (300 psi) and heated to
140 °C. The mixture was kept at this temperature for 16 h and then
checked by HPLC;. 52% conversion to the thiophene-amide 15 was
observed. 1H NMR (500 MHz, DMSO-d6): δ 8.83 (s, 1 H); 7.71 (d,
J=8.20 Hz, 2 H); 7.57 (d, J=7.96 Hz, 2 H); 7.23 (d, J=7.96 Hz,
2 H); 7.07 (d, J=8.20 Hz, 2 H); 4.01 (s, 2 H); 3.81 (s, 3 H); 2.39 (s,
Methyl 4-(1-Isocyanatocyclopropyl)benzoate (16). A visually
clean 25-mL round-bottom flask was charged with phosgene
(0.501 mL, 20 wt % toluene solution, 0.947 mmol) and DCM
(5 mL) and cooled to 0 °C. To the solution was added the
cyclopropylamine 8 (170 mg, 0.888 mmol) and Et3N (272 μL,
1.953 mmol) as a DCM (2.5 mL) solution over a period of 15 min.
The mixture was warmed to 20 °C and aged 30 min. The reaction
was quenched by the addition of 1 N HCl and diluted with DCM
(20 mL). The layers were separated, and the organic layers was
washed with brine (10 mL), dried with MgSO4, filtered, and
concentrated under reduced pressure. The residue was used
directly in the amidation. The material can be purified by flash
3 H); 2.27 (s, 3 H); 1.25-1.21 (m, 2 H); 1.11-1.04 (m, 2 H). 13
C
1
chromatography 80:20 hexanes/EtOAc. H NMR (400 MHz,
NMR (101 MHz, DMSO-d6): δ 166.5 166.3 149.6 145.7 135.8 135.2
134.2 131.9 129.3 129.2 127.3 127.1 (q, J= 31.7 Hz) 125.5 (q, J=
3.7 Hz) 125.2 124.9 (q, J=271.9 Hz) 52.4 34.8 32.0 19.4 14.4 13.0.
19F NMR (377 MHz, acetone-d6): δ -66.9. IR 3276.8 (w) 2923.0
(w) 1718.0 (s) 1638.2 (s) 1329.4 (m) 1118.7 (m) 1068.1 (w). HRMS
calcd for C26H24NO3SF3: 487.14290, obsd [M þ H] 488.15191.
Melting point: 220.9 -221.7 °C.
CDCl3): δ 7.97 (d, J = 8.17 Hz, 2 H); 7.27 (d, J=8.37 Hz, 2 H);
3.88 (s, 3 H); 1.42-1.40 (m, 2 H); 1.29-1.27 (m, 2 H). 13C NMR
(101 MHz, CDCl3): δ 166.5 146.2 129.8 128.7 124.3 121.6 52.0
38.4 19.5. IR: 2959.7 (w) 2285.1 (s) 1721.6 (s) 1609.6 (m) 1432.1 (m)
1281.5 (s) 1111.7 (s). HRMS calcd for C12H11NO3: 218.08117,
obsd [M þ H] 218.08102. Melting point =42.6 -43.2 °C.
Methyl 4-{1-[({2,5-Dimethyl-4-[4-(trifluoromethyl)benzyl-
3-thienyl}carbonyl)amino]cyclopropyl} Benzoate (15). By Friedel-
Craft amidation: A visually clean 10-mL round-bottom flask was
charged with thiophene-alkane 9 (82 mg, 0.304 mmol) and DCE
(1.5 mL). To the solution were added the cyclopropyl-isocyanate 16
(60 mg, 0.276 mmol) and the FeCl3 (49 mg, 0.304 mmol). The
1
Supporting Information Available: Copies of H and 13C
NMR spectra for compounds 1, 1 DEA, 5, 6, 8, 8 MsOH, 9, 12,
3
3
13, 15, and 16. This material is available free of charge via the
Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 12, 2010 4085