A Concise Synthesis of Racemic Lorcaserin

Article abstract of DOI:10.1071/CH15667

We report herein the concise synthesis of racemic lorcaserin (±)-1, which is an anti-obesity drug. The synthetic route involved the key synthesis of an asymmetrical imide intermediate 11 and its efficient reduction. Imide 11 was synthesized directly by the reaction of nitrile 9 with acid 10. The reducing system of NaBH4, AlCl3, and trimethylsilyl chloride efficiently fulfilled the reduction of imide 11 to amine 8a, which could be converted to (±)-1 via Friedel-Crafts reaction as reported. This route afforded 69% two-step yield of 8a from 9 via 11, and the concise synthesis of (±)-1 was completed in three steps. This route offers an alternative pathway to the synthesis of (±)-1 and its analogues.

Full text of DOI:10.1071/CH15667

CSIRO PUBLISHING
Aust. J. Chem.
Full Paper
http://dx.doi.org/10.1071/CH15667
A Concise Synthesis of Racemic Lorcaserin
A
,
C
A
,
C
B
,
D
,
A D
Bin Xu,
Jincai Su,
Jing Wang,
and Guo-Chun Zhou
A
School of Pharmaceutical Sciences, Nanjing Tech University, Nanjing 211816,
Jiangsu, China.
B
College of Sciences, Nanjing Agricultural University, Nanjing 210095, Jiangsu, China.
C
These authors contributed equally to the work.
D
Corresponding authors. Email: wangjing@njau.edu.cn; gczhou@njtech.edu.cn
We report herein the concise synthesis of racemic lorcaserin (ꢀ)-1, which is an anti-obesity drug. The synthetic route
involved the key synthesis of an asymmetrical imide intermediate 11 and its efficient reduction. Imide 11 was synthesized
directly by the reaction of nitrile 9 with acid 10. The reducing system of NaBH , AlCl , and trimethylsilyl chloride
4
3
efficiently fulfilled the reduction of imide 11 to amine 8a, which could be converted to (ꢀ)-1 via Friedel–Crafts reaction as
reported. This route afforded 69 % two-step yield of 8a from 9 via 11, and the concise synthesis of (ꢀ)-1 was completed in
three steps. This route offers an alternative pathway to the synthesis of (ꢀ)-1 and its analogues.
Manuscript received: 21 October 2015.
Manuscript accepted: 14 December 2015.
Published online: 28 January 2016.
Introduction
Then, (ꢀ)-1 was obtained by Friedel–Crafts reaction of chloride
8a or bromide 8b under the catalysis of 1.5 equiv. of anhydrous
AlCl in 1,2-dichlorobenzene or 5.1 equiv. of anhydrous AlCl3
Obesity has become one of the most important public health
challenges in modern society. As a primary risk factor for type 2
diabetes, obesity is closely associated with the metabolic
syndrome and is linked to low-living quality and reduced life-
3
without solvent. Finally, resolution of (ꢀ)-1 by L-(þ)-tartaric
acid afforded lorcaserin (1). Apart from these processes, there
have been synthetic routes to prepare (ꢀ)-1 starting from
[
1]
span. Serotonin (5-HT) mediates or regulates a wide variety
of behaviours. Its diverse effects are related to the extensive
projections of serotonergic neurons of at least 14 different
receptor subtypes. Among all those subtypes, current research
studies suggest that the role of serotonin in appetite control is
[
7]
[8]
4-chlorophenylacetic acid and its esters.
Results and Discussion
[
As amine 2, alcohol 3 or bromide 4, 4-chlorophenylacetic
5]
[6]
[6]
[2]
largely related to the hypothalamic serotonin 5-HT2C receptor.
[7]
[8]
acid, and its esters used in the previous processes are not
primary raw materials and were derived from 4-chloro-
phenylacetonitrile (9), in our synthetic route (Scheme 2), 9 was
used as a starting material to synthesize key intermediate
asymmetrical imide 11, and imide 11 was efficiently reduced by
Lorcaserin (1), a highly selective 5-HT2C receptor agonist, has
been approved in 2012 by US FDA as the first approved anti-
obesity drug since orlistat was approved in 1999. Unlike the
earlier non-selective serotonin agonists, lorcaserin is a novel and
selective 5-HT2C full agonist with 15–100-fold higher affinity
over those of 5-HT2A and 5-HT2B receptors, which mediate the
NaBH -Lewis acid reducing system to afford amine 8a. After
4
[3]
that, amine 8a was transformed into the target compound (ꢀ)-1
pathogenesis of serotonergic valvulopathy.
In addition to the Heck coupling approach
Friedel–Crafts reaction
[
5,6]
[
4a-c]
by Friedel–Crafts reaction as reported.
Because this
and olefin
for the synthesis of racemic lorca-
serin ((ꢀ)-1), two major synthetic routes via Friedel–Crafts
[
4d]
synthetic route avoids unnecessary transformation and total
synthesis of (ꢀ)-1 is complete in three steps, it has potential
application in the industrial production of lorcaserin (1).
[
5,6]
reaction of (ꢀ)-1 were reported in previous literature
and
[
5]
outlined in Scheme 1. In one approach, amide compound 5
was obtained by the reaction of 2-chloropropionyl chloride with
amine 2, and then 5 was cyclized into intermediate 6 via Friedel–
Synthesis of Asymmetrical Imide 11 by the Reaction
of Nitrile 9 and Acid 10
Crafts reaction catalyzed by 3.0 equiv. of anhydrous AlCl , and
3
In our study streamline shown in Scheme 2, asymmetrical imide 11
is the key intermediate. Though cyclic imide is well known for
thereafter, reduction of amide 6 into (ꢀ)-1 was accomplished by
[9]
[10,11]
borane tetrahydrofuran (BH -THF). In the another synthetic
3
synthesis and application, few available methods
reported for the synthesis of acyclic imide, and only symmetrical
have been
[
5,6]
route,
(
bromination of alcohol 3 by phosphorous tribromide
PBr ) gave bromide 4. Alkyl amination of 1-amino-2-propanol
[10b,10d]
imide
or special asymmetrical imide with greatly different
have been prepared by the reaction of a nitrile
3
[10b]
acid–nitrile pairs
by bromide 4 was fulfilled in refluxing toluene to afford alcohol
. Chlorination of 7 by thionyl chloride (SOCl ) afforded
7
with an acid in the presence of Br o¨ nsted acid or/and Lewis acid.
However, these methods are not suitable for the synthesis of
conventional asymmetrical imide, for example, 11 in this work.
2
intermediate chloride 8a. Alternatively, bromination of alcohol
by thionyl bromide (SOBr ) provided intermediate bromide 8b.
7
2
Journal compilation Ó CSIRO 2016
www.publish.csiro.au/journals/ajc

B
B. Xu et al.
Cl
Cl
OH
OH
X
H
N
H
N
NH2
4
and
2
and
COCl
Cl
PBr3 X ꢁ OH,
Toluene
reflux, 87 %
Acetonitrile
O
X ꢁ NH2,
2
3
4
7
Cl
Cl
0
ꢀC to rt, 88 %
5
98 %
X ꢁ Br,
SOCl , 81 %
AlCl , neat
2
3
or SOBr , 63 %
150ꢀC, 54 %
2
∗
Y
Cl
1
O
H
N
AlCl , 1,2-dichlorobenzene,
3
BH -THF
rt, 70 %
Cl
NH
3
120–140
ꢀ
C, 86–89 %
HCl
NH
or AlCl , neat, 120–140
ꢀ
C
3
(
ꢂ)-
Resolution
Scheme 1. Previously reported synthetic routes of racemic lorcaserin ((ꢀ)-1) and Lorcaserin (1).
1
(1R)-1
Cl
90–91 %
8
8
a
, Y ꢁ Cl
, Y ꢁ Br
6
b
1
) NaBH , AlCl ,
4 3
Cl
TMSCl, THF, 70ꢀC
AlCl3
dichlorobenzene
O
TMSCl, TFA, Tf O
anhydrous ZnCl2
2
H
N
CN
2) HCl(g) in
ethyl acetate, 95 %
7
8%
50ꢀC, 72 %
ꢃ
HO
10
(ꢂ)-1
8a
Cl
O
O
or
Cl
Table 1
Cl
Table 2
9
11
AlCl , neat, 68 %
3
Scheme 2. Synthetic method of racemic lorcaserin (ꢀ)-1 via asymmetrical imide 11 employed in this work.
To synthesize asymmetrical imide 11 by the reaction of
nitrile 9 with acid 10, a series of reaction conditions were
explored (Table 1). In our initial trial reaction of nitrile 9 with
catalyzed by 1.0 equiv. of anhydrous ZnCl , 1.5 equiv. of 10 and
2
TMSCl, and 0.5 equiv. of TFA and Tf O with respect to 9
2
afforded 51 % of 11 (Table 1, entry 10). The reaction catalyzed
by 1.5 equiv. of 10 and TMSCl, 1.0 equiv. of anhydrous ZnCl2
and TFA, 0.5 equiv. of Tf O produced 48 % of 11 (Table 1, entry
2
-chloropropionic acid 10, sulfuric acid, hydrochloric acid, and
p-toluenesulfonic acid were used to activate the reactivity of 10,
however, only hydrolysis products of carboxylic acid and amide
from 9 were isolated (data not shown). Subsequent explorations
gave mild yields e.g. at 08C, the reaction of 1.5 equiv. of 10 with
2
11). The reaction performed with 1.5 equiv. of 10 and TMSCl,
and 1.0 equiv. of anhydrous ZnCl , TFA, and Tf O afforded
2
2
44 % of 11 (Table 1, entry 12). However, the simultaneous
increases of TFA and Tf O to 1.0 equiv. and anhydrous ZnCl to
1
fonic anhydride (Tf O) (Table 1, entry 1) or 1.0 equiv. of Tf O in
.0 equiv. of 9 catalyzed by 1.0 equiv. of trifluoromethanesul-
2
2
1.5 equiv. with 1.5 equiv. of 10 and TMSCl reduced the yield of
11 to 26 % (Table 1, entry 13). It was fortunate that the reaction
of 9 in 1.0-g scale and in 2.0-g scale with 1.5 equiv. of 10 in the
presence of 1.5 equiv. of TMSCl and TFA, and 1.0 equiv. of
anhydrous ZnCl and Tf O produced good yields (70 % and
2
2
[
10d]
the presence of 1.0 equiv. of trifluroacetic acid (TFA)
Table 1, entry 2) afforded 33 % or 48 % of the desired product
1. Using unchanged quantities of 10 and 9, the reaction
(
1
catalyzed by 1.0 equiv. of Tf O and 2.0 equiv. of TFA (Table 1,
2
2
2
entry 3) or 2.0 equiv. of Tf O and 1.0 equiv. of TFA (Table 1,
2
73 %, respectively) of the desired product 11 (Table 1, entries 14
and 15). Repeated experiments using the same ratio of 9 to 10,
TMSCl, anhydrous ZnCl , TFA, and Tf O produced the same
yields of 11 i.e. 70–75 %. The scale up reaction involving 10.0 g
of starting nitrile 9 under the same conditions as entries 14 and
15 (Table 1) achieved 72 % of 11 in 72 h (Table 1, entry 16),
suggesting that the reaction in larger scale needed longer
reaction times.
entry 4) provided 38 % or 29 % of 11. It is noted that the increase
of TFA (Table 1, entry 3) or Tf O to 2.0 equiv. produced more
by-products.
2
2
2
Then, anhydrous ZnCl was introduced into the reaction. It
2
was observed that the reaction catalyzed by 0.5 equiv. of anhy-
drous ZnCl and 1.0 equiv. of Tf O at 08C and 258C did not occur
2
2
(data not shown) and at 508C (Table 1, entry 5) afforded only
17 % of 11 with non-negligible by-products and a large amount
of unreacted 9. Reducing the amount of Tf O to 0.5 equiv. and
the addition of 1.5 equiv. of trimethylsilyl chloride (TMSCl) and
Based on entries 14 and 15 (Table 1), some variations were
investigated. Using 0.5 equiv. of Tf O (Table 1, entry 17), the
2
2
reaction afforded a reduced yield of 11 to 38 %. Meanwhile,
lowering the amount of anhydrous ZnCl to 0.5 equiv. afforded
0
anhydrous ZnCl produced 26 % of 11 and considerably lesser
.5 (Table 1, entry 6) or 1.0 (Table 1, entry 7) equiv. of
2
58 % of 11 (Table 1, entry 18). Indeed, reducing the amount of
TFA to 1.0 equiv. afforded a lower yield (44 %) of 11 (Table 1,
entry 12). These data suggested that the suitable combination of
TMSCl, anhydrous ZnCl , TFA, and Tf O was important for the
2
amounts of by-products. However, the reaction performed in the
absence of Tf O did not proceed (Table 1, entry 8), whereas the
2
reaction performed in the absence of anhydrous ZnCl depleted
2
2
2
9
Tf O accelerated and/or promoted the reaction and also gener-
and gave a mixture (Table 1, entry 9). It was suggested that
reaction, and the optimal ratios of 9 to 10, TMSCl, anhydrous
ZnCl , TFA, and Tf O were found to be 1.0–1.5, 1.5, 1.0, 1.5,
2
2
2
ated more by-product/s. However, TMSCl and anhydrous ZnCl2
reduced the activity of Tf O and instigated the production of
1.0, respectively, which is the key point for achieving high
reaction yields.
2
asymmetrical imide 11 or TMSCl and anhydrous ZnCl played a
2
It was found that the above solvent-free reaction became a
thick mass, thus making stirring difficult at 10–18 h after the
reaction started. Considering that efficient stirring could shorten
the reaction time and enhance the reaction yield, solvents DCM,
role in inhibiting the generation of by-product/s.
Thereafter, we focussed on the optimization of the ratio of 9
to 10, TMSCl, anhydrous ZnCl , TFA, and Tf O. The reaction
2
2

Synthesis of Lorcaserin
C
A,B
Table 1. Reaction conditions for the preparation of 11 from 9 and 10 (Scheme 2)
C
D,E
Yield [%]
Entry
T [8C]
Time [h]
TMSCl [equiv.]
ZnCl
2
[equiv.]
TFA [equiv.]
Tf
2
O [equiv.]
1
2
3
4
5
6
7
8
9
0
0
72
72
72
72
24
48
48
48
48
48
48
48
48
48
48
72
48
48
24
/
/
/
1.0
1.0
1.0
2.0
1.0
0.5
0.5
/
33
48
38
29
17
26
26
NR
M
/
/
1.0
2.0
1.0
/
0
/
/
0
/
/
50
50
50
50
50
50
50
50
50
50
50
50
50
50
50
/
0.5
0.5
1.0
1.0
/
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
/
/
/
/
0.5
0.5
0.5
1.0
1.0
1.0
1.0
1.0
0.5
1.0
1.0
1
1
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
8
9
1.0
1.0
1.0
1.5
1.0
1.0
1.0
1.0
0.5
1.0
0.5
1.0
1.0
1.0
1.5
1.5
1.5
1.5
1.5
1.5
51
48
44
26
70
73
72
38
58
29
A
Optimization reactions (1.0 equiv. of 9 and 1.5 equiv. of 10) were performed in 1.0-g scale of 9 except for reactions in entries 12, 15, and 17 that were
performed in 2.0-g scale of 9, and in entry 16 that were performed in 10.0-g scale of 9. The progress of the reaction was monitored by TLC.
B
‘
/’ means without the substance.
C
ZnCl
2
is anhydrous ZnCl
2
.
D
Isolated yield of 11.
E
NR means no reaction and M means a mixture.
toluene, and THF were successively selected as reaction sol-
vents. However, the reaction performed in DCM or in THF did
not proceed, and the reaction performed in toluene did not work
well and gave a mixture (data not shown). Therefore, the neat
reaction is the only choice in the current stage. Moreover, using
low temperatures (08C or 258C) was not good for the reaction,
whereas the reactions performed at 708C became complicated
In this work, TFA and Tf O could independently or together
2
catalyze this type reaction; however, they did not work very well
for asymmetrical imide 11. Even though the concerted or step-
by-step mechanism of this reaction and the roles of Tf O, TFA,
2
ZnCl , and TMSCl in this reaction are unclear, the suitable
2
combination of Tf O, TFA, ZnCl , and TMSCl with two starting
2
2
materials of nitrile 9 and acid 10 is the key for achieving higher
yields of asymmetrical imide 11. This observation suggests that
using the required optimal ratio is good for producing asymmet-
rical imide 11 in the competition between the formation of
symmetrical and asymmetrical imides.
(
1
data not shown) under the same reaction conditions as entries
4 and 15 (Table 1). These results indicated that elevated
temperatures facilitated the formation of undesired impurities.
Furthermore, it was observed that the reaction performed for
2
4
4 h (Table 1, entry 19), when compared with that performed for
8 h, dramatically reduced the yield of 11 to 29 % under the
Reduction of Asymmetrical Imide 11 to Amine 8a
same reaction conditions as entries 14 and 15 (Table 1). This
result demonstrated that maintaining the heating condition at
After completion of the synthesis of imide intermediate 11, the
reduction of 11 to 8a was the next key step. First, the reducing
system of NaBH in the presence of TMSCl was selected to
4
reduce imide 11 to amine 8a; however, no amine product 8a was
[
12]
5
08C was necessary for the completion of the reaction even
though the stirring was not efficient. On the other hand,
substitution of anhydrous ZnCl with anhydrous FeCl , anhy-
drous FeCl , or anhydrous AlCl led to no reaction by FeCl and
detected even though 10.0 equiv. of NaBH and 20.0 equiv. of
4
2
2
TMSCl were used (Table 2, entry 1). Then, we introduced
metallic Lewis acid (LA) to the reducing system and five dif-
ferent ratios of NaBH or KBH and anhydrous reagents of
3
3
2
anhydrous AlCl (data not shown) or the desired product 11 by
3
FeCl3 even though 9 was consumed completely (data not
shown). It was concluded that there is currently no sign of
improving the reaction by other Lewis acids rather than anhy-
drous ZnCl2.
4
13]
4
[
AlCl , ZnCl or MgCl
2
were explored (Table 2, entries 2–6).
However, no detectable product 8a was obtained up to 5.0 equiv.
3
2
of NaBH with 4.0 equiv. of anhydrous AlCl (Table 2, entry 7).
4
3
It is worth noting that purification of 11 was based on the
solubility difference, whereby the desired product 11 displays a
relatively better solubility in acetonitrile and DCM than the
by-products including symmetrical imide 2-(4-chlorophenyl)-
N-(2-(4-chlorophenyl)acetyl)acetamide (12) (Supplementary
Material). Generally, a large scale (e.g. starting from 10.0 g of 9)
of dry crude products, prepared according to entries 14 and 15
Subsequently, TMSCl was introduced into the reduction.
Addition of 0.5 or 1.5 equiv. of TMSCl to 6.0 equiv. of NaBH4
and 4.0 equiv. of anhydrous AlCl afforded 58 % or 57 % of
amine product 8a (Table 2, entries 8 and 9). Although the
increase of TMSCl from 0.5 to 1.5 equiv. did not affect the
yield of 8a, it was improved to 67 % when 4.0 equiv. of TMSCl
was used (Table 2, entry 10). In contrast, reducing the amount of
3
(Table 1), was extracted at 408C by acetonitrile or DCM, and
then crystallization of soluble materials from petroleum/ethyl
acetate provided pure asymmetrical imide 11 in 70–73 % yield.
anhydrous AlCl from 4.0 to 3.0 equiv. made the reduction much
3
less efficient (Table 2, entry 11) though 6.0 equiv. of NaBH and
4
4.0 equiv. of TMSCl were used.

D
B. Xu et al.
A,B
Table 2. Reaction conditions
for the reduction of imide 11 into
Conclusions
In summary, a concise synthetic route of racemic lorcaserin
amine 8a
(
ꢀ)-1 via an imide intermediate 11 has been developed by
E
Temperature [8C] Time [h] Yield [%]
F
Entry
Equivalents
C
starting from cheap and primary raw materials. In this approach,
asymmetrical essential skeleton imide 11 was obtained by the
relatively mild reaction of primary raw nitrile 9 with acid 10,
which could be a general process to prepare asymmetrical imide.
Subsequently, highly efficient reduction of imide 11 to amine 8a
was accomplished by NaBH /AlCl /TMSCl reducing system.
D
LA TMSCl
MBH
4
1
2
3
4
5
6
7
8
9
10
4
–
20
70
70
95
95
95
95
70
70
70
70
70
70
70
70
70
38
12
24
24
24
24
24
24
24
24
24
24
24
24
24
0
0
2
–
2.5
3
2.5
2
–
0
–
0
4
3
4
2
–
0
After the cyclization of 8a proceeded by AlCl , (ꢀ)-1 was
3
[
5,6]
6
3
–
0
afforded as reported.
two-step yield of 8a from 9 via 11, and the total synthesis of
Accordingly, the route afforded 69 %
5
4
–
,10
58
57
67
15
95
96
58
31
6
4
0.5
1.5
4
(
(
ꢀ)-1 from 9 via 11 by this route was fulfilled in three steps
Scheme 2). As starting from the reaction of primary raw nitrile 9
6
4
1
1
1
1
1
1
0
1
2
3
4
5
6
4
with acid 10 avoided unnecessary transformations and made the
synthesis of (ꢀ)-1 more concise and efficient, this route
6
3
4
6
4
5.5
5.5
5.5
5.5
(Scheme 2) should be attractive in cost and practicability and has
potential for large-scale industrial synthesis.
6
4
4
4
6
2
Experimental
A
All optimization reactions were performed with the same scale of starting
Commercially available materials and solvents were purchased
from Alfa Aesar, Energy Chemical, JiaXing Assent Chemical
Co. or Xilong Chemical Co. Anhydrous DCM was treated with
material 11 (0.5 g) except for reaction in entry 13 (10.0 g).
B
Reactions in entries 3–6 were performed in THF/toluene (1 : 1), whereas
the remaining reactions were performed in THF.
C
In entries 3, 4, and 6, MBH
refers to NaBH
2
In entries 3 and 4, MgCl was used as LA; in entries 5 and 6, ZnCl was used
4
refers to KBH
4
, whereas in the remaining
CaH and distilled, anhydrous THF was refluxed with Na and
2
1
13
entries, MBH
4
4
.
benzophenone and distilled under nitrogen. H and C NMR
spectra were recorded on a Bruker AV-400 spectrometer at 400
and 100 MHz, respectively, in [D6]DMSO and CD OD.
D
2
as LA; and in the remaining entries, AlCl
E
3
was used as LA.
3
Bath temperature.
Chemical shifts (d) are reported in parts per million (ppm)
relative to the solvent, and coupling constants (J) are expressed
in Hertz (Hz). High-resolution electrospray ionization mass
spectrometry (ESI MS; in the positive ion or negative ion
acquisition mode) analysis was performed on an Applied Bio-
systems Q-STAR Elite ESI-LC-MS/MS mass spectrometer.
HPLC analysis was performed on a Sunfire C18 column
(4.6 ꢁ 250 mm, 5 mm) eluted by 88 % methanol with 12 %
0.02 M solution (pH 8.0) of potassium hydrogen phosphate at a
F
Isolated yield of salt 8a.
Interestingly, the reduction yield of 8a was dramatically
increased to 95 % when 5.5 equiv. of TMSCl, 6.0 equiv. of
NaBH , and 4.0 equiv. of AlCl with respect to 11 were used
4
3
(
found that reducing the amount of NaBH from 6.0 to 4.0 equiv.
Table 2, entry 12). Compared with the above experiment, it was
4
ꢂ1
sharply decreased the yield of 8a to 58 % (Table 2, entry 14), and
using 2.0 rather than 4.0 equiv. of anhydrous AlCl produced
rate of 0.8 mL min and detected at 220 nm. Melting points
(uncorrected) were measured using an YRT-3 melting point
apparatus (Shanghai, China).
3
only 31 % of 8a (Table 2, entry 15). These experimental results
indicated that the ratio of 11, NaBH , anhydrous AlCl , and
4
3
TMSCl was crucial to the reduction and the optimal ratio was
: 6 : 4 : 5.5. Considering two carbonyl moieties of the imide 11,
the applied equiv. (relative to 11) of reductants were in reason-
able range. Under the optimal reaction conditions, the scale-up
reduction of 10.0 g of compound 11 generated 95.6 % of salt 8a
4-Chloro-N-(2-chloro-1-oxopropyl)-benzeneacetamide
(11) and 2-(4-Chlorophenyl)-N-(2-(4- chlorophenyl)acetyl)
acetamide (12) Obtained by the Reaction of Nitrile 9
and Acid 10
1
In a typical synthesis, 8.5 mL 2-chloropropionic acid 10
(98.7 mmol, 1.5 equiv.), 12.5 mL TMSCl (98.6 mmol,
(
Table 2, entry 13).
1
7
.5 equiv.), 9.0 g anhydrous ZnCl2 (66.0 mmol, 1.0 equiv.),
.5 mL TFA (100.0 mmol, 1.52 equiv.), 11.0 mL Tf O
RC ey ac cl i tzi oa nt ion of Amine 8a into (ꢀ)-1 by Friedel–Crafts
2
(66.0 mmol, 1.0 equiv.), and 10.0 g 4-chlorophenylacetonitrile 9
[
5,6]
According to the reported procedure,
hydrochloride salt of 8a proceeded to the intramolecular cycli-
the corresponding
(66.0 mmol) were added sequentially at 08C. The reaction was
stirred at 08C for 1 h and at room temperature (rt) for additional
1 h before heating to 508C (bath temperature). After reacting at
508C for 72 h, the reaction mixture was treated with 100 mL
deionized water and 100 mL petroleum ether and stirred at room
temperature for 3 h. The resulting precipitate was filtered and
equilibrated with ethyl acetate and sat. sol. NaHCO
organic phase was separated and washed with sat. sol. NaHCO
and dried over anhydrous Na SO . The reaction mixture was
zation by the Friedel–Crafts reaction using 1.5 equiv. of AlCl as
3
the catalyst at 1308C in 1,2-dichlorobenzene and finally, ,78 %
of (ꢀ)-1 was provided (Procedure A, see Experimental).
Alternatively, the intramolecular Friedel–Crafts reaction of 8a
[5]
was carried out in solvent-free conditions; however, instead of
, and the
3
[
using more than 5 equiv. of anhydrous AlCl as reported, the
5]
3
3
reaction was catalyzed by only 1.5 equiv. of anhydrous AlCl at
3
2
4
1
Experimental). Identical H NMR spectra and HPLC results
308C for 5 h to afford 68 % of (ꢀ)-1 (Procedure B, see
then filtered, evaporated, dried under vacuum, and the dried
solid was extracted with 100 mL acetonitrile at 408C, and the
undissolved solids of 12 and other by-product/s were filtered.
The filtrate was evaporated and the solid was re-crystallized
1
(
Supplementary Material) of (ꢀ)-1 were obtained to those
[
5,6]
reported.

Synthesis of Lorcaserin
E
from ethyl acetate and petroleum ether to afford desired product
1
DMSO, 400 MHz) 11.15 (1H, br s), 7.38 (2H, d, J 8.4), 7.27 (2H,
d, J 8.5), 4.85 (1H, q, J 6.7), 3.89 (2H, s), 1.55 (3H, d, J 6.7). dC
308C under vacuum to provide hydrochloride salt of (ꢀ)-1. dH
([D6]DMSO, 400 MHz) 9.69 (1H, br s), 9.28 (1H, br s), 7.28–7.18
(3H, m), 3.46 (1H, d, J 7.0), 3.30 (1H, dd, J 13.2, 7.5), 3.21 (2H, d,
J 13.4), 3.02 (1H dd, J 15.7, 7.0), 2.89 (2H, m), 1.34 (3H, d, J 7.3).
The H NMR spectra and HPLC results (Supplementary Material)
[5,6]
of (ꢀ)-1 were identical to those reported.
1 as a white powder (12.4 g, 72.2 %), mp 131ꢂ1338C. d ([D6]
H
1
(
CD OD, 101 MHz) 173.4, 171.3, 134.1, 132.3, 131.8 (2C),
3
þ
1
cacld for C H Cl NNaO 282.0065. Pure 12 as a white solid
29.6 (2C), 54.9, 43.8, 20.8. HRMS m/z 282.0059 [M þ Na] ;
1
1
11
2
2
was obtained by silica gel column chromatography eluted by
DCM, mp 208.9–211.98C. dH ([D6]DMSO, 400 MHz) 11.03
Acknowledgements
This work was supported in part by the Natural Science Foundation of China
(
1H, br s), 7.36 (4H, d, J 8.4), 7.26 (4H, d, J 8.5), 3.83 (4H, s).
þ
(
30973621) and Six Major Talents of Jiangsu Province of China (2014).
HRMS m/z 344.0216 [M þ Na] ; cacld for C H Cl NNaO
1
6
13
2
2
3
44.0221.
References
[
1] Prospective Studies Collaboration, Lancet 2009, 373, 1083. doi:10.1016/
S0140-6736(09)60318-4
2] J. C. Halford, E. J. Boyland, C. L. Lawton, J. E. Blundell, J. A. Harrold,
Drugs 2007, 67, 27. doi:10.2165/00003495-200767010-00004
Synthesis of Hydrochloride Salt of 2-(4-Chlorophenyl)-N-
ethyl-N-2-propylchloride (8a)
[
Typically, 8.7 g NaBH (230.7 mmol, 6.0 equiv.) was suspended
4
in 150 mL anhydrous THF under a nitrogen atmosphere and in
an ice bath, and 20.5 g AlCl (153.8 mmol, 4.0 equiv.) was added
to the reaction mixture in three batches. After the mixture was
stirred at room temperature for 1 h to yield a white sticky solid,
[3] W. J. Thomsen, A. J. Grottick, F. Menzaghi, H. Reyes-Saldana,
S. Espitia, D. Yuskin, K. Whelan, M. Martin, M. Morgan, W. Chen,
H. Al-Shamma, B. Smith, D. Chalmers, D. Behan, J. Pharmacol. Exp.
Ther. 2008, 325, 577. doi:10.1124/JPET.107.133348
3
[
4] (a) J. Smith, B. Smith, World Patent WO 2003/086306 A2 2003.
b) B. M. Smith, J. M. Smith, J. H. Tsai, J. A. Schultz, C. A. Gilson,
10.0 g compound 11 (38.4 mmol) in 100 mL THF and then
26.5 mL TMSCl (209.0 mmol, 5.5 equiv.) were added sequen-
(
S. A. Estrada, R. R. Chen, D. M. Park, E. B. Prieto, C. S. Gallardo,
D. Sengupta, W. J. Thomsen, H. R. Saldana, K. T. Whelan,
F. Menzaghi, R. R. Webb, N. R. A. Beeley, Bioorg. Med. Chem. Lett.
tially to the reaction mixture. After the addition, the reaction
mixture was heated to 708C (bath temperature) and stirred for 1
day, and the reaction was carefully quenched with water (50 mL)
in an ice bath after cooling to room temperature. The mixture pH
was adjusted to 14 by 30 % NaOH solution, and the reaction
mixture was stirred for ,30 min then extracted with DCM
2
005, 15, 1467. doi:10.1016/J.BMCL.2004.12.080
c) B. M. Smith, J. M. Smith, J. H. Tsai, J. A. Schultz, C. A. Gilson,
(
S. A. Estrada, R. R. Chen, D. M. Park, E. B. Prieto, C. S. Gallardo,
D. Sengupta, P. I. Dosa, J. A. Covel, A. Ren, R. R. Webb, N. R. A.
Beeley, M. Martin, M. Morgan, S. Espitia, H. R. Saldana, C. Bjenning,
K. T. Whelan, A. J. Grottick, F. Menzaghi, W. J. Thomsen, J. Med.
Chem. 2008, 51, 305. doi:10.1021/JM0709034
(
2 ꢁ 150 mL). The organic layer was washed with 150 mL sat.
sol. NaHCO and 100 mL sat. brine, dried over anhydrous
3
Na SO , and evaporated under reduced pressure to give a pale
2
4
(
d) Q. Zhu, J. Wang, X. Bian, L. Zhang, P. Wei, Y. Xu, Org. Process
yellow liquid. In an ice bath, the resulting liquid was dissolved in
00 mL ethyl acetate before hydrogen chloride gas was passed
Res. Dev. 2015, 19, 1263. doi:10.1021/ACS.OPRD.5B00144
5] (a) B. W. Burbaum, C. A. Gilson III, S. Aytes, S. A. Estrada,
D. Sengupta, B. Smith, M. Rey, U. Weigl, World Patent WO 2005/
019179 A2 2005.
1
[
through the solution and stirred for ,30 min to produce 9.9 g
hydrochloride salt of 8a as a white solid (96 % yield). d ([D6]
DMSO, 400 MHz) 9.07 (2H, br), 7.41 (2H, d, J 8.4), 7.30 (2H, d,
J 8.4), 4.56–4.46 (1H, m), 3.40 (1H, m), 3.29–3.22 (1H, m),
H
(
b) B. Smith, C. Gilson III, J. Schultz, J. Smith, World Patent WO 2005/
003096 A1 2005.
3
.21–3.13 (2H, m), 3.03–2.93 (2H, m), 1.53 (3H, d, J 6.6).
[6] U. Weigl, F. Porstmann, C. Straessler, L. Ulmer, U. Koetz, World
Patent WO 2007/120517 A2 2007.
[
7] T. Gharbaoui, S. K. Tandel, Y. Ma, M. Carlos, J. R. Fritch, World
Patent WO 2008/070111 A2 2008.
Synthesis of (ꢀ)-8-Chloro-1-methyl-2,3,4,5-tetrahydro-1H-
3
-benzazepine ((ꢀ)-1) from 8a by Intramolecular Friedel–
[
8] X. Wang, X. J. Z. Li, Y. G. Li, Y. Gao, X. H. Lv, X. L. Shen, Z. Wang,
X. L. Huang, China Patent CN 103333111 A 2013.
[
5,6]
Crafts Reaction
In Procedure A, the mixture of 3.0 g hydrochloride salt of 8a
11.2 mmol), 2.2 g anhydrous AlCl (16.8 mmol), and 20.0 mL
[9] M. K. Hargreaves, J. G. Pritchard, H. R. Dave, Chem. Rev. 1970, 70,
39. doi:10.1021/CR60266A001
10] (a) D. Davidson, H. Skovronek, J. Am. Chem. Soc. 1958, 80, 376.
4
(
3
[
1
,2-dichlorobenzene was stirred at 1308C for 16 h; or in Pro-
cedure B (neat), the mixture of 3.0 g hydrochloride salt of 8a
11.2 mmol) and 2.2 g anhydrous AlCl (16.8 mmol) was stirred
doi:10.1021/JA01535A032
(
b) W. S. Durrell, J. A. Young, R. D. Dresdner, J. Org. Chem. 1963, 28,
31. doi:10.1021/JO01038A060
c) R. A. Slavinskaya, T. N. Sumarokova, T. A. Kovaleva, M. K.
Nasirdinova, Zh. Org. Khim. 1982, 18, 2243.
d) V. A. Vinokurov, E. G. Gaevoi, L. A. Ryazanova, R. A. Karakhanov,
Zh. Org. Khim. 1986, 22, 1783.
(
3
8
at 1308C for 5 h. After cooling to room temperature, the reaction
was quenched by slow addition of 5.0 mL deionized water in an
ice bath and then 50 mL 2.0 N HCl and 30 mL (Procedure A) or
(
(
50 mL (Procedure B) ethyl acetate was added and stirred for
1 h. The aqueous layer was separated and the pH was adjusted to
14 by 30 % NaOH solution before extraction with 150 mL DCM.
[11] (a) H. Zohreh, S. Peyman, A. Z. Mohammad, Y. Maryam, Synlett
2007, 812.
(
b) M. Nasr-Esfahani, M. Montazerozohori, N. Filvan, J. Serb. Chem.
Soc. 2012, 77, 415. doi:10.2298/JSC110511168N
c) M. M. Khodaei, E. Nazari, Tetrahedron Lett. 2012, 53, 2881.
doi:10.1016/J.TETLET.2012.03.121
d) J. Shang, A. Pourvali, J. R. Cochrane, C. A. Hutton, Aust. J. Chem.
015, 68, 1854. doi:10.1071/CH15504
The organic layer was washed with 100 mL sat. sol. NaHCO3
and 100 mL sat. brine, dried over anhydrous Na SO , and
2
4
(
evaporated under reduced pressure to give a yellow liquid
78.2 % yield by Procedure A or 68.6 % yield by Procedure B).
(
(
Free base of (ꢀ)-1: d ([D6]DMSO, 400 MHz) 7.18–7.10 (3H,
H
2
m), 3.06–2.97 (1H, m), 2.89–2.68 (5H, m), 2.58 (1H, dd, J 13.1,
[
12] A. Giannis, K. Sandhoff, Angew. Chem., Int. Ed. Engl. 1989, 28, 218.
doi:10.1002/ANIE.198902181
[13] Y. Qiu, F. Zhang, C. Zhang, Tetrahedron Lett. 2007, 48, 7595. doi:10.
7
.2), 1.23 (3H, d, J 7.3). Free base of (ꢀ)-1 was dissolved in ethyl
acetate saturated with hydrogen chloride and stirred for 1 h at 08C,
and the precipitated solid was collected by filtration and dried at
1016/J.TETLET.2007.08.118