p-chlorobenzaldehyde with acetaldehyde. Indeed, compound
12 was formed more dominantly if the reaction 8 f 2 was
run in the presence of 0.1 equiv of acetaldehyde. Also, when
the reaction was run in 2-propanol instead of ethanol, no
traces of 12 were formed, supporting the likelihood that
ethanol was the source for acetaldehyde. We could exclude
oxygen as the oxidizing agent for ethanol. Irrespective of
whether the reaction was run under inert conditions or in
the presence of oxygen, compound 12 was formed in similar
amounts.17 Thus, acetaldehyde seemed to be formed in small
amounts from ethanol via hydride abstraction by a hydride
acceptor, most likely p-chlorobenzaldehyde.
Only recently have we encountered a similar hydride
transfer while investigating the condensation reaction of the
sodium lithium salt of tert-butyl acetoacetate with E-[3-(4-
fluorophenyl)-1-(1-methylethyl)-1H-indol-2-yl]-2-propenal in
the fluvastatin project.18 The hydride transfer of electron-
rich species to nonenolizable aldehydes has also been
reported by other authors.19
methane (307 mL) was added. After 3-10 h, the deep-red
reaction mixture was transferred to 15% aqueous hydrochlo-
ric acid (500 mL). After phase separation, the organic phase
was washed twice with water. Dichloromethane was distilled
off, while ethanol was added at the same rate. The crystal-
lized product was isolated by filtration to give 86 g (94%)
of pure 5.
5.3. 2-Dibutylamino-1-(2,7-dichloro-9H-fluoren-4-yl)-
ethanol (8). To a slurry of 5 (80 g, 0.257 mol) in ethanol
(240 mL) was added sodium borohydride (3.15 g, 0.083 mol)
portionwise at -5 to 5 °C within 1 h. As the viscosity
increased, good mixing was necessary to reach complete
conversion. After agitating for 1 h, dibutylamine (99.6 g,
0.771 mol) was added, and the mixture was heated up to
distill off the ethanol. The mixture was heated to 140 °C for
3 h, cooled down to below 90 °C, and quenched with 10%
aqueous sodium hydroxide (100 mL). After phase separation,
the organic phase was washed with brine (2 × 100 mL).
The organic layer was concentrated under reduced pressure
at 60-90 °C. After complete removal of dibutylamine,
ethanol (170 mL) was added, keeping the temperature above
60 °C. The solution was seeded at 50 °C and allowed to
crystallize at 0 °C to yield 84.5 g (81%) of pure 8.
5.4. 2-Dibutylamino-1-{2,7-dichloro-9-[1-(4-chlorophen-
yl)meth-(Z)-ylidene]-9H-fluoren-4-yl}ethanol (2). New Pro-
cess Conditions. A stirred mixture of 8 (80 g, 0.197 mol),
p-chlorobenzaldehyde (29 g, 0.207 mol), sodium hydroxide
(15.8 g, 0.394 mol), and ethanol (1160 mL) was heated up
within 1 h from 23 to 70 °C and immediately afterwards
cooled within 1 h from 70 to 38 °C. The mixture was agitated
at 38 °C for at least 4 h to complete crystallization. After
filtration, washing, and drying, 91.6 g (88%) of crude 2 was
isolated. When the reaction was run in the presence of
oxygen, compound 10 was obtained as the main product and
was isolated in crude form by silica gel chromatography.
1H NMR (400 MHz, chloroform-d) δ 0.98 (t, J ) 7.23 Hz,
6 H), 1.30-1.57 (m, 8 H), 2.43-2.58 (m, 3 H), 2.64-2.73
(m, 2 H), 2.82 (dd, J ) 12.95, 3.60 Hz, 1 H), 5.16 (dd, J )
10.23, 3.54 Hz, 1 H), 7.45 (d, J ) 1.39 Hz, 2 H), 7.55 (d, J
) 2.08 Hz, 1 H), 7.65 (t, J ) 1.26 Hz, 1 H), 7.80 (dd, J )
2.08, 0.57 Hz, 1 H); MS (ES+) m/z 420 (MH+).
5.5. New Process for the Crystallization of 2. A mixture
of crude 2 (15 g, 0.284 mol) and heptane fraction (75 mL)
was heated to reflux (94 °C). A solution was formed which
was clear filtered above 90 °C. The solution was cooled to
70 °C, seeded with pure 2, cooled within 4 h to 3 °C, and
stirred at that temperature for at least 1 h. The suspension
was filtered, washed, and dried to yield 13.9 g (93%) of pure
2.
5.6. Preparation of Compound 12. A stirred mixture of
8 (8.2 g, 20 mmol), p-chlorocinnamaldehyde 11 (3.0 g, 21.1
mmol), sodium hydroxide (0.69 g, 17.3 mmol), and ethanol
(159 mL) was heated up within 1 h from 23 to 38 °C. The
mixture was agitated at 38 °C for 18 h. The formed crystals
were isolated by filtration and recrystallized from ethyl
acetate and hexane. 1H NMR (500 MHz, DMSO-d6) δ 0.79
The formation of by-product 12 cannot be completely
prevented. However, by observing the optimal temperature
profile during the isomerization-crystallization in the
Knoevenagel condensation, it can be maintained at a level
low enough that its complete removal is guaranteed during
the recrystallization of 2.
4. Summary
The manufacturing process for lumefantrine 2 was
reworked. For the conversion of 5 to compound 8 a one-pot
process was developed, eliminating isolation of the undesir-
able epoxide 7. A significant increase in throughput was also
achieved by applying new reaction conditions for the
Knoevenagel condensation of 8 to 2 and for the recrystal-
lization of 2. The elimination of the somewhat increased level
of by-product 12 could be brought under control.
5. Experimental Section
5.1. 2,7-Dichloro-9H-fluorene (4). Chlorine (92 g, 1.30
mol) was passed into the suspension of fluorene (103 g, 0.618
mol) in acetic acid (661 g) at 35-45 °C. While fluorene
was dissolving, compound 4 started to crystallize from the
mixture. After the addition of chlorine was completed
(approximately 15 h), the mixture was heated to 90 °C to
liberate formed hydrochloric acid and to dissolve the product.
The mixture was cooled to room temperature, and the
precipitated product was isolated by filtration and drying to
yield 20 g (48%) of pure 4.
5.2. 2-Chloro-1-(2,7-dichloro-9H-fluoren-4-yl)ethanone
(5). A mixture of chloroacetyl chloride (41 g, 0.363 mol),
AlCl3 (59 g, 0.444 mol), and dichloromethane (118 mL) was
cooled to -5 °C. A solution of 4 (69 g) dissolved in dichloro-
(17) This is in sharp contrast to the formation of fluorenone 10. Compound 10
can be made to the dominant product if during the conversion of 8 f 2 air
is bubbled through the reaction mixture.
(18) Fuenfschilling, P. C.; Hoehn P.; Mutz J.-P. Organic Process Res. DeV. 2007,
11, 13.
(19) For lithium enolate of acetaldehyde as the reducing agent for nonenolizable
aldehydes, see: Di Nunno, L.; Scilimati, A. Tetrahedron 1988, 44, 3639.
344
•
Vol. 11, No. 3, 2007 / Organic Process Research & Development