758
S. M. Laaman et al.
SHORT PAPER
der N2 at ambient temperature for 30 min. To the yellow suspension
was added the furan (1 mmol) and the mixture heated at reflux until
the starting material was consumed (TLC) (see Table). The solvent
was evaporated under reduced pressure and the resulting residue
dissolved in CH2Cl2 (15 mL), washed with aq HCl (2M, 2 x 10 mL),
H2O (2 x 10 mL) and dried (MgSO4). The organic solvent was
evaporated under reduced pressure and the crude product purified
by flash chromatography to give the products recorded in the Table
and whose properties are summarised below:
Reactions of 2,5-Diphenylfuran (2a)
then these transformations would become particularly at-
tractive. We have now treated the above furans with the
Katz reagent and find that it works exactly like the trimer
1; the isothiazoles are produced very cleanly and conve-
niently in even higher yields (Table). Thus following the
earlier work,5 a solution of ethyl carbamate (4.3 mmol),
thionyl chloride (4.3 mmol) and pyridine (2.0 mL) in ben-
zene (20 mL) was stirred at room temperature for 30 min;
the furan (1 mmol) was added and the mixture heated at
reflux until the furan was consumed (TLC; mostly within
Flash chromatographic purification (5% EtOAc/light petroleum)
gave 5-benzoyl-3-phenylisothiazole3b (3a) as a clear oil (279 mg,
98%).
1H NMR: d = 7.40–7.59 (5H, m, ArH), 7.64–7.70 (1H, m, ArH),
7.95–8.01 (5H, m, ArH).
MS (CI): m/z = 266 (M+ + 1, 100%), 265 (M+, 58), 188 (M+ – Ph,
65), 105 (PhCO+, 42).
IR (neat): n = 3062 (Ar-H), 2923, 2854, 1646 (C=O), 1577 (C=C),
1496, 1450, 1388, 1268, 1114, 1025, 844, 767, 713, 690, 590 cm–1.
6–24 h). There was no reaction in the absence of pyridine The standard procedure was used with the omission of pyridine and
no reaction was observed.
nor, surprisingly, if the pyridine was replaced by Hünig’s
base or 2,6-di-tert-butyl-4-methylpyridine, and use of less
thionyl chloride lowered the yield.
The standard procedure was modified by replacing pyridine with
Hünig’s base or 2,6-di-tert-butyl-4-methylpyridine8 and no reaction
was observed.
The similarity of the results with both sets of reagents sug-
gests the same reaction pathway through a common inter-
mediate, most reasonably NSCl. To gain more insight into
the mechanism we compared the reaction of a highly po-
larised unsymmetrical furan, 2-(4-methoxyphenyl)-5-(4-
nitrophenyl)furan (2d), with the trimer and with the stan-
dard carbamate–thionyl chloride mixture. Both gave the
same single product, 5-(4-methoxybenzoyl)-3-(4-nitro-
phenyl)isothiazole (3d), regiospecifically and in high
yield, with none of the 5-nitrobenzoyl isomer. This obser-
vation, in the trimer reaction,3c was considered to favour a
mechanism involving electrophilic substitution of the
The standard procedure modified by using 1 mol equiv of SOCl2
gave 5-benzoyl-3-phenylisothiazole3b (3a) as a clear oil (23 mg,
8%) identical to that described above, and recovered 2,5-diphenyl-
furan (194 mg, 78%).
Reaction of 2,5-Bis(4-methylphenyl)furan (2b)9
Flash chromatographic purification (20% EtOAc/light petroleum)
gave 5-(4-methylbenzoyl)-3-(4-methylphenyl)isothiazole3b (3b) as
a tan solid (278 mg, 94%), mp 101–102 °C (lit.,3b mp 100–101.5
°C).
1H NMR: d = 2.41 (3H, s, Me), 2.47 (3H, s, Me), 7.29–7.37 (4H, m,
ArH), 7.85–7.91 (4H, m, ArH), 7.98 (1H, s, 4-H)
furan ring by NSCl to give a b-thiazyl derivative which re- MS (CI): m/z = 293 (M+, 85%), 202 (12), 119 (55).
arranges to the isothiazole; it provides further support for
a common reaction pathway.
IR: n = 1650, 1605, 1500, 1418, 1405, 1311, 1285, 1211, 1203,
1179, 1111, 1069, 1042, 1016, 968, 953, 934 cm–1.
Whatever the precise mechanism, the mixture described
by Katz and co-workers5 is a most convenient reagent for
this one-pot synthesis of isothiazoles, which is a useful ad-
dition to the limited number of good routes to this mono-
cyclic ring system.7 It remains to be seen whether the
same reagents will duplicate a wider range of trimer reac-
tions.3,6
Reaction of 2,5-Di-tert-butylfuran (2c)10
Flash chromatographic purification (10% EtOAc/light petroleum)
gave 3-tert-butyl-5-pivaloylisothiazole (3c) as colourless prisms
(132 mg, 66%), mp 29–31 °C (lit.,3b mp 29–31 °C).
1H NMR: d = 1.35 and 1.37 (2 x 9H, s, 2 x t-Bu), 7.62 (1H, s, ArH).
MS (EI): m/z = 225 (M+, 21%), 210 (16), 168 [(M-But)+, 62], 141
[(M-t-BuCN)+, 34], 57 (t-Bu, 28).
IR: n = 3000, 2967, 2929, 2904, 2868, 1672, 1506, 1470, 1396,
1360, 1265, 1236, 1212, 1150, 1115, 974, 900, 831, 799 cm–1.
Melting points were taken on a Kofler hot-stage apparatus, infrared
spectra recorded on a Unicam Research Series 1 FTIR instrument as
liquid films (for oils) or KBr discs, NMR spectra in CDCl3 solution
with TMS as internal standard on a JEOL 270 spectrometer and
mass spectra on a Kratos MS8ORF. TLC was performed using
Merck silica 60F254 plates and flash chromatography was conducted
using Fluka silica gel 60. Light petroleum had bp 60–80 °C.
Reaction of 2-(4-Methoxyphenyl)-5-(4-nitrophenyl)furan (2d) 3c
Flash chromatographic purification (20% EtOAc/light petroleum)
gave 5-(4-methoxybenzoyl)-3-(4-nitrophenyl)isothiazole3c (3d) as
light yellow needles (219 mg, 76%), mp 159–160°C (lit.,3c mp 158–
160 °C).
1H NMR: d = 3.93 (3H, s, MeO), 7.05 (2H, d, J = 8.9, ArH), 8.02
(2H, d, J = 8.9, ArH), 8.08 (1H, s, 4-H), 8.16 (2H, d, J = 8.9, ArH),
8.34 (2H, d, J = 8.9, ArH).
Reaction of Ethyl Carbamate, Thionyl Chloride and Pyridine
with Various Furans; General Procedure
A solution of ethyl carbamate (383 mg, 4.3 mmol), SOCl2 (512 mg,
4.3 mmol) and pyridine (2.0 mL) in benzene (20 mL) was stirred un-
MS (EI): m/z = 340 (M+, 79%), 135 (100).
Synthesis 1999, No. 5, 757– 759 ISSN 0039-7881 © Thieme Stuttgart · New York