Large-scale preparation of lodobenzene dichloride and efficient monochlorination of 4-aminoacetophenone

Article abstract of DOI:10.1021/op980024e

Large-scale monochlorination of 4-aminoacetophenone using iodobenzene dichloride is described. Special emphasis was given to the characterization of the iodobenzene dichloride and to the development of a practical procedure for handling this agent from the viewpoint of hazards. This process was successfully scaled up in a pilot plant.

Full text of DOI:10.1021/op980024e

Organic Process Research & Development 1998, 2, 270−273
Large-Scale Preparation of Iodobenzene Dichloride and Efficient
Monochlorination of 4-Aminoacetophenone
Atsuhiko Zanka,* Hiroki Takeuchi, and Ariyoshi Kubota
Technological DeVelopment Laboratories, Fujisawa Pharmaceutical Co. Ltd., 2-1-6 Kashima,
Yodogawa-ku, Osaka 532, Japan
Chart 1. Monochlorination of 4-aminoacetophenone
Abstract:
Large-scale monochlorination of 4-aminoacetophenone using
iodobenzene dichloride is described. Special emphasis was
given to the characterization of the iodobenzene dichloride and
to the development of a practical procedure for handling this
agent from the viewpoint of hazards. This process was
successfully scaled up in a pilot plant.
except for hydrogen chloride gas. We first tried to directly
react 4-aminoacetophenone with chlorine gas, but this only
gave mixtures of several byproducts, and no desired monochlo-
rinated product could be obtained. In further investigations,
it was found that treating chlorine gas with iodobenzene to
give iodobenzene dichloride first, followed by reaction with
4-aminoacetophenone was highly effective for selective
aromatic monochlorination. Whilst chlorination of anilines
by iodobenzene dichloride was reported by Murakami5 and
this agent also has proven to be applicable for several
valuable reactions in pharmaceutical chemistry,6 it has failed
to achieve wide use in organic synthesis mainly because of
lack of stability. To our knowledge, no application of this
agent on a large scale has been described in the literature.
Thus, the initial goal of process development involved
evaluation of the possibility of safely applying this agent to
a large scale by using exploratory chemistry prior to running
reactions on the pilot plant scale.
Introduction
The introduction of chlorine into an aromatic ring is a
commonly used reaction in organic synthesis and medicinal
chemistry. Many methods have been reported for carrying
out chlorination in the chemical industry.1 However, there
are few methods amenable to the large-scale production of
pharmaceutical intermediates in pilot plants, which usually
do not have any special equipment.
Recently, we were interested in preparation of 4-amino-
3-chloroacetophenone (2, Chart 1), which is a common
intermediate to many of the COX-II (cyclooxygenase-2)
selective inhibitors prepared in Fujisawa.2 Whilst several
methods are reported for chlorination of reactive aniline,3
there are few reports presenting practical aromatic chlorina-
tion for deactivated anilines that are amenable to a large-
scale synthesis in a pharmaceutical pilot plant.4 Amongst
several chlorinating reagents, N-chlorosuccinimide (NCS) is
especially attractive for a large-scale preparation, since this
agent is inexpensive, stable to storage for a long time, and
commercially available in bulk quantities. Thus, we first
selected and applied NCS to preparation of 2 by modifying
the methods reported by Niokson.4 Whilst effective for a
small-scale synthesis, several complications were identified
in the direct scale-up of this method. The reactions involved
several byproducts that were difficult to remove and were
identified as predominately side chain chlorinated products,
along with traces of nuclear and side chain polychlorinated
derivatives, and consequently resulted in low yield (∼40%).
To develop more efficient and cost-effective chlorination
methods, our efforts turned to a search for alternative
chlorinating reagents in greater detail. Amongst numerous
chlorinating agents, we directed our attention to chlorine gas
in that this agent is inexpensive, readily available, and
involves no material which is lost as waste to be treated,
Results and Discussion
Evaluation and Characterization of Iodobenzene Dichlo-
ride. Our efforts involved initially complete evaluation and
characterization of this agent, followed by designing a
chemical process after evaluating several aspects: studies
by differential thermal analysis (DTA), differential scanning
calorimetry (DSC), accelerating rate calorimetry (ARC), and
impact sensitivity test (IST). DTA was used to indicate the
decomposition profile under heated conditions. As shown
in Figure 1, decomposition of this agent started from about
100 °C, but the rate of heat flow was not considered serious.
The potential thermokinetic description could be estimated
by measuring heat liberated using DSC. The exotherm
monitored was due to the decomposition of agent and was
-8.26 kcal/mol (Figure 2). IST indicated that this reagent
may decompose under conditions of severe impact. The
detailed chemical hazard involved in running the reaction
was also evaluated using ARC. As shown in Figure 3, this
(1) McBee, E. T.; Hass, H. B. Ind. Eng. Chem. 1941, 33, 137.
(2) Tsuji, K.; Nakamura, K.; Konishi, N.; Okumura, H.; Matsuo, M. Chem.
Pharm. Bull. 1992, 40, 2399.
(3) Neale, R. S.; Schepers, R. G.; Walsh, M. R. J. Org. Chem. 1964, 29, 3390.
(4) Niokson, T. E.; Roche-Dolson, C. A. Synthesis 1985, 669.
(5) Murakami, M.; Inukai, M.; Koda, A.; Nakano, K. Chem. Pharm. Bull. 1971,
19, 1696.
(6) Barton, D. H. R.; Miller, E. J. Am. Chem. Soc. 1950, 72, 370.
270
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development
S1083-6160(98)00024-3 CCC: $15.00 © 1998 American Chemical Society and Royal Society of Chemistry
Published on Web 06/16/1998

that it was possible to deal with this reagent on a pilot scale
according to further investigations: (1) decomposition energy
was small and partial decomposition did not lead to total
reagent decomposition; (2) the impact unstability problem
was overcome by using a Bu¨chner funnel in the filtration
procedure; (3) this agent cleanly precipitated from methylene
chloride and could be air-dried in a light-excluded environ-
ment at room temperature; (4) chlorination reaction pro-
ceeded smoothly under cooled conditions whereby this agent
could be handled without major concerns for thermal
decomposition.
Practical Preparation of Iodobenzene Dichloride. Al-
though described methods in the literature are practical and
effective,8 there still existed serious shortcomings regarding
a scale-up procedure. For example, in pilot-plant work,
safety concerns should be addressed regarding treatment with
chlorine gas from the view of personal hazard. Therefore,
the process development campaign first began with defining
a practical procedure for reaction conditions for chlorine gas
with iodobenzene. In our earliest studies, chlorine gas was
directly bubbled into a mixture of iodobenzene and meth-
ylene chloride in order to prevent excess gas from leaking
from the vessel. However, the tube at the outlet was soon
clogged with products. Another potential scale-up problem
was the method for determining reaction completion. Excess
use might cause leakage of chlorine gas into a pilot plant;
however, not enough chlorine gas sharply dropped the yield.
Whilst the absorbed gas amount, roughly estimated by
measuring the weight of the steel bomb, was effective as an
indicator, a more exact end point of the reaction was required.
HPLC analysis of the remaining amount of iodobenzene was
most definite, but this required opening of the reactor in a
chlorine gas atmosphere. However, further investigations
indicated that the reaction speed was high enough not to
require directly bubbling chlorine gas into the reaction
mixture; hence, a closed system was also considered to be
applicable. In effect, an internal pressure increase in the
reactor was used to reveal the reaction end point. Since pilot
plants usually do not include reactors suitable for working
under pressure conditions, we designed the practical ap-
paratus depicted in Figure 4 and determined the end point
in the following three ways: (1) the change of the ethylene
glycol surface level (indicating internal small pressure
change), (2) the weight of the steel bomb, and (3) HPLC
analysis. After completion was determined, the excess
chlorine gas was substituted by nitrogen gas, and the
precipitate was filtered using a Bu¨chner funnel. The cake
was washed with fresh methylene chloride to remove
contained chlorine gas and mother liquor, followed by
ventilation with fresh air in a dark space. The scale of this
reaction was designed after considering three viewpoints:
absorption speed of chlorine gas, heat balance between
reaction exotherm, and cooling ability of the reaction vessel
and productivity. Actually, this procedure was successfully
conducted three times on a 20 kg scale.
Figure 1. DTA result.
Figure 2. DSC result.
Figure 3. ARC result.
agent was unaffected until the temperature reached 50 °C,
and the development of decomposition was slow above this
temperature. The pressure data also showed the same profile
(Figure 3). Other concerns were that this agent is rather
unstable to light7 and that complete removal of solvents from
the product by evaporation might decrease the quality.8
Despite these drawbacks, we finally reached the conclusion
(7) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; Wiley: New York,
1969; Vol. 1, p 505.
(8) Lucas, H. J.; Kennedy, E. R. Organic Syntheses; Wiley: New York, 1955;
Collect. Vol. III, p 482.
Practical Application of Iodobenzene Dichloride to
Chlorination of 4-Aminoacetophenone. On lab scale,
4-aminoacetophenone uneventfully reacted with iodobenzene
Vol. 2, No. 4, 1998 / Organic Process Research & Development
•
271

Figure 5. Reaction of 4-aminoacetophenone with iodobenzene
dichloride under adiabatic conditions.
byproducts, which caused purification problems downstream.
Removal of these impurities by recrystallization was not
effective enough and resulted in necessity for chromatog-
raphy purification. Therefore, we turned our effort to other,
more practical purification methods in more detail and found
that the hydrogen chloride salt of 2 selectively precipitated
from organic solvents. In early studies, the filtration was
very slow and practically impossible to apply on large scale;
however, further investigation revealed that the salt produced
from dioxane improved the filtration speed. Applying these
findings, this process was successfully scaled up to pilot
plant.
Figure 4. Construction of reactor.
dichloride in the presence of pyridine, involving a small
exotherm to afford 2 with small amounts of byproducts. This
procedure was developed into a scaled up process. Our first
concern for safety in applying this agent in a pilot plant
involved addressing the reaction exotherm. On a lab scale
(<10 g of 4-aminoacetophenone), the heat of reaction under
ice-cooled conditions almost completely dissipated to the
surroundings, so that no typical exotherm was seen. How-
ever, as is well-known,9 the heat losses from a full-scale
reactor are much less compared with those from laboratory
glassware. Especially, in the case of a rapid reaction using
rather unstable agents, underestimation of the heat evolved
can cause a reaction to run out of control. Therefore,
accurate information about heat evolved was essential prior
to running the pilot plant, and this should be estimated under
adiabatic conditions. In a run carried out in an adiabatic
reactor starting with 11.5 g of 4-aminoacetophenone, the
temperature exothermed from ∼3 to ∼53 °C over only 3
min (Figure 5). This result indicated that the reaction might
run away if the all starting materials were put into the
reaction vessel at the same time, but also that reaction heat
control could be achieved by addition speed. Our strategy
for an operational procedure involved the addition of small
amounts of iodobenzene dichloride (∼1 kg) under cooled
conditions (0 °C) and following the exthotherm of the
reaction, followed by cooling to 0 °C again. This procedure
was totally repeated 24 times with about a 10 min interval
between additions. After completion of addition, the pre-
cipitated pyridine hydrogen chloride was filtered and washed
with fresh THF. Next, we endeavored to remove unwanted
byproducts. The filtrate contained small amounts of several
Conclusions
In this paper, we have described a practical and scaled
up monochlorination of 4-aminoacetophenone using iodo-
benzene dichloride. This selective chlorinating agent was
characterized from the viewpoint of safety issues. The results
from DTA and DSC indicated the decomposition profile of
this agent and called for further investigations. The defining
safety data came from ARC studies and the fact that it was
possible to drive the chlorination reaction under cooled
conditions. These detailed safety evaluation approaches to
hazards are useful when evaluating an unstable reagent for
a process. The final process was successfully scaled up to
afford 24.8 kg of 2 in 87% yield of 94% purity.
Experimental Section
General Procedures. 4-Aminoacetophenone of pure
grade was commercially available from EMS-DOTTIKON
AG. Melting points were measured on a Thomas-Hoover
apparatus and are uncorrected. IR spectra were recorded on
a HORIBA FT-210 spectrometer. NMR spectra were
measured on a Bruker AC200P (1H, 200 MHz). Chemical
shifts are given in parts per million, and tetramethylsilane
was used as the internal standard. Mass spectra were
measured on a Hitachi Model M-80 mass spectrometer using
EI for ionization. Elemental analyses were carried out on a
Perkin-Elmer 2400 CHN elemental analyzer. HPLC analyses
were performed using a YMC GEL ODS 120 Å S-7 column
and an acetonitrile/water phase. The water component was
adjusted with KH2PO4 and Na2HPO4 to pH ) 7.8. Purity
of each obtained product was determined by comparison with
(9) Etchells, J. C. Org. Process Res. DeV. 1997, 1, 435.
272
•
Vol. 2, No. 4, 1998 / Organic Process Research & Development

purified authentic samples using quantitative HPLC. Re-
agents and solvents were used as obtained from commercial
suppliers without further purification. The DTA experiment
was carried out using a Seiko Model SSC/5200H. The DSC
experiment was performed on a Rigaku Model DSC-10A.
The ARC experiment was conducted on a CSI 851-001 using
a bomb with a 1/4 in. neck. The start temperature of the
run was 27 °C, with heat step of 3 °C, a search time of 10
min, and a slope sensitivity of 0.02 °C/min. The φ-factor
was 3.1.
Practical Preparation of Iodobenzene Dichloride. Chlo-
rine gas (9.0 kg, 127 mol) was introduced to iodobenzene
(20.0 kg, 98 mol) in methylene chloride (30 L) with stirring
at -3 to +4 °C as shown in Figure 4. When the reaction
was judged to be complete by HPLC, the excess chlorine
gas was exchanged with nitrogen gas and the precipitate was
filtered off, washed with fresh methylene chloride (6 L), and
dried at room temperature to afford iodobenzene dichloride
(25.3 kg, 94% yield) as a yellowish solid; mp 113-117 °C
(dec.) (lit.7 115-120 °C (dec.)).
was totally consumed. After the addition was complete, the
resulting mixture was further stirred at the same temperature
for 1 h. The precipitate was filtered off, and the cake was
washed with fresh THF (19.5 L). The filtrate was washed
with brine (160 L), sodium bisulfite (15 kg, 144 mol) in water
(130 L), brine (130 L), and then concentrated under reduced
pressure to ∼58.5 L. After the addition of dioxane (117 L),
hydrogen chloride in dioxane (4 mol/L, 78 L) was added
dropwise to this solution at 15-20 °C and stirring was
continued at 0 °C over 2 h. The precipitate was filtered off
and washed with dioxane (19.5 L). Drying under reduced
pressure afforded 4-amino-3-chloroacetophenone (2) as its
hydrogen chloride salt (24.8 kg, 87% yield) of 94% purity
as a yellowish solid: mp 165-166 °C (dec.); 1H NMR (200
MHz, DMSO-d6) 2.43 (s, 3H), 6.87 (d, 1H, J ) 4.2 Hz),
7.66 (dd, 1H, J ) 8.5, 2.0 Hz), 7.79 (d, 1H, J ) 2.0 Hz),
8.69 (br s, 3H); IR (KBr) 3064, 2296, 2008, 1692, 1601,
1574, 1555, 1521 cm-1; MS (EI) m/z 170, 136, 111, 75. Anal.
Calcd for C8H9N2OCl2: C, 46.63; H, 4.40; N, 6.80. Found:
C, 46.67; H, 4.38; N, 6.63.
Large-Scale Preparation of 4-Amino-3-chloroacetophe-
none Hydrogen Chloride Salt. Iodobenzene dichloride (1.5
kg, 5.46 mol) was added to a solution of 4-aminoacetophe-
none (19.5 kg, 144 mol) in a mixture of THF (195 L) and
pyridine (11.4 kg, 144 mol) at 0 °C and the internal
temperature smoothly reached about 3 °C. After recooling
to under 0 °C, further iodobenzene dichloride (1.5 kg, 5.46
mol) was added. Addition of iodobenzene dichloride (about
1.5 kg) was totally repeated 24 times at ∼10 min intervals.
In this way, 36.8 kg (144 mol) of iodobenzene dichloride
Acknowledgment
We especially thank Dr. David Barrett, Medicinal Chem-
istry Research Laboratories, Fujisawa Pharmaceutical Co.
Ltd., for his interest and ongoing advice in this work.
Received for review March 12, 1998.
OP980024E
Vol. 2, No. 4, 1998 / Organic Process Research & Development
•
273