1
410
O. Ilovich and J. Deutsch
Vol. 42
tors that contributed to low yield was the occurrence of a
polymerization reaction with formation of a black tar. In
order to minimize polymerization, the addition of 2-furan-
carboxaldehyde (1) was divided in two steps. First, the
reaction uses milder conditions when compared with the
condensation of ethylene acetal of 4-bromo-2-furancar-
boxaldehyde and the 17-ketosteroids in the presence of n-
butyllithium. Such mild conditions are of great interest in
condensation of 17-ketosteroids that contain a 14β-
hydroxy group that has been shown to be important in the
structure-activity relationship of cardioactive steroids
towards Na/K ATPase.
2
-furancarboxaldehyde (1) was added to the ice cold and
mechanically stirred aluminum chloride in a small volume
of methylene chloride till a homogeneous mixture of the
complex of the aluminum chloride and the 2-furancarbox-
aldehyde (1) was formed. The rest of the 2-furancarbox-
aldehyde (1) was added in additional 40 minutes and the
mixture was left to cool down in an ice bath. Further, the
bromine was added during the next 2 hours. Any attempts
to speed up the bromine addition ended in the production
of larger amount of black tar. Attempts to purify the prod-
uct by distillation at low pressure also produced low yields
due to polymerization. In order to avoid polymerization
during the purification process, the 4,5-dibromo-2-furan-
carboxaldehyde (2) was first protected as ethylene acetal
and the distillation process was replaced by flash chro-
matography on silica gel.
EXPERIMENTAL
2
-Furancarboxaldehyde, bromine, n-butyllithium, aluminum
chloride, 4,4,5,5-tetramethyl-1,3,2-dioxaborolane, benzene, ethyl-
ene glycol, PdCl (pph ) and silica gel (Merck, 230-400 mesh,
2
3 2
60Å) were purchased from Aldrich Chemical, (Milwaukee, IL).
All the solvents were purchased from Frutarom Israel. The 2-furan-
carboxaldehyde was distilled prior to use (95-96 °C at 10 mm Hg).
The mass (electron impact ionization) spectra were measured with
a TRACE GC/MS (Finnigan, ThermoQuest, San Jose, CA) operat-
ing at 70eV and with the ion source heated at 200 °C. The nmr
measurements were performed with a Varian T 300 spectrometer in
deuteriochloroform.
The debromination process with n-butyllithium was
carried out at a temperature of –45 to –55 °C because the
reaction at lower temperatures (-65 to –75 °C), caused the
formation of the unwanted 5-bromo-2-furancarboxalde-
hyde isomer (6). In other words, at a low temperature (-
4
,5-Dibromo-2-furancarboxaldehyde (2).
To mechanically stirred aluminum chloride (90 g, 680 mmols)
and cooled on ice, a solution of 2-furancarboxaldehyde (1) (13.5
g, 141 mmols) in methylene chloride (16 mL) was added slowly
over a period of 20 minutes. An additional (16.5 g, 172 mmols)
of 2-furancarboxaldehyde was added to the mixture over a
period of 40 minutes. Bromine (120 g, 750 mmols) was added
dropwise for 2 hours and the mixture was left stirring overnight
at room temperature. The reaction mixture was cooled on an ice
bath and ice was added slowly over a period of 1 hour. After the
exothermic reaction stopped, brine (500 mL) and 30% diethyl
ether in hexane (800 mL) was added with continuous stirring
until all the solid material was dissolved. The organic phase was
separated and the solvent removed at low pressure. Co-evapora-
tion with dry benzene (50 mL) yielded 60 g of crude 4,5-
dibromo-2-furancarboxaldehyde (yield: 75%). The product was
further reacted with ethylene glycol without purification since
attempts to distill the 4,5-dibromo-2-furancarboxaldehyde
7
0 °C) the elimination of the 4-bromo atom is as fast as
the elimination of the 5-bromo atom producing an equal
mixture of the two isomers. In contrast, at higher temper-
atures (-45 to –55 °C) only the 5-bromo elimination takes
place to give the desired 4-bromo-2-furancarboxaldehyde
isomer (4) exclusively (Table 1). When we consider a
plausible mechanism, we must take into account the fact
that this is not a simple case of a kinetic and thermody-
namic controlled reaction. The reaction is not reversible
and once a certain product mixture is formed, it will not
change due to time/temperature changes. The bromine in
the C5 position is the more hindered among of the two,
therefore the lithium-halogen exchange on the C5 posi-
tion requires a higher temperature in order to take place.
At lower temperatures the bromine atom in the C4 posi-
tion is equally reactive, creating up to an equal mixture of
the two products.
1
yielded in a black polymer; H nmr: δ = 9.48 (s, 1H, CHO), 7.22
13
ppm (s, 1H, C3-H); C nmr: δ = 105.16 (C3), 131.67 (C2),
123.86 (C4), 153.82 (C1), 176.21 ppm (C5); ms: m/z (%): 252
+
.
(50 M ), 254 (100), 256 (50).
Table 1
Temperature
-bromo-2-furancarboxaldehyde
-70 to -78 °C
50%
-60 to -70 °C
65%
-50 to -60 °C
78%
-45 to -55 °C
100%
4
The Suzuki boronate derivative was synthesized by a
variation of the reported procedure [13]. The reagents
were heated for 6 hours at 80 °C due to the low reactivity
of the hetero aromatic bromides in formation of the sp2 -
2-(4,5-Dibromo-2-furanyl)-[1,3]-dioxolane (3).
Crude 4,5-dibromo-2-furancarboxaldehyde (2) (60 g, 0.23
mols) was dissolved in benzene (300 mL), mixed with ethylene
glycol (70 g, 1.12 mols), concentrated sulfuric acid (2 mL) and
boiled overnight in a Dean Stark apparatus (10 mL of water was
released). Half of the solvent was removed at reduced pressure
and replaced with 20% diethyl ether in hexane (200 mL). The
3
sp bond. Reflux of the toluene was found to decrease the
reaction yield - again due to polymerization. The Suzuki
Ilovich, Ohad
