Y. Ashikari, et al.
CatalysisTodayxxx(xxxx)xxx–xxx
was collected for 20 min. To the solution, water was added, and the
mixture was extracted by dichloromethane. The combined organic
phase was concentrated under reduced pressure, and the obtained solid
was washed by methanol to afford tert-butyl 6-(3-(1-adamantyl)-4-an-
isyl)-2-naphthoate (7) in 91 % yield (1.70 g, 3.62 mmol) as a white
solid. 1H NMR (400 MHz, CDCl3) δ 1.66 (s, 9 H), 1.80 (s, 6 H), 2.10 (s,
3 H), 2.18 (s, 6 H), 3.90 (s, 3 H), 6.99 (d, J =8.4 Hz, 1 H), 7.54 (dd,
J = 2.4, 8.4 Hz, 1 H), 7.60 (d, J =2.4 Hz, 1 H), 7.78 (dd, J = 2.0,
8.4 Hz, 1 H), 7.88 (d, J =8.8 Hz, 1 H), 7.97 (d, J =8.4 Hz, 1 H), 8.00
(br, 1 H), 8.03 (dd, J = 1.6, 8.4 Hz, 1 H), 8.54 (br, 1 H), 13C NMR
(100 MHz, CDCl3) δ 28.3, 29.1, 37.1, 37.2, 40.6, 55.1, 81.1, 112.0,
124.7, 125.6, 125.7, 125.9, 126.3, 127.9, 128.8, 129.6, 130.4, 131.2,
132.6, 135.7, 138.9, 141.0, 158.8, 166.0, HRMS (ESI) calculated for
Table 1
Flow Suzuki–Miyaura coupling reaction of phenylboronic acid (1) with 4-io-
dobenzonitrile (2) using HBPSG-Pd column reactor.
Entry
solvent
X eq.
1.1
base
Yield (%)
1
2
3
4
5
6
7
8
9
THF
Et3N
TBAF
Et3N
TBAF
Et3N
TBAF
KOH
KOH
KOH
1
1
1
7
12
44
45
61
83
Toluene
1.1
10 % MeOH/THF
1.1
C
32H36O3Na [M + Na+]: 491.2557, found 491.2541.
To 200 mL round-bottomed flask were added
3
3
a
7 (1.68 g,
20 % MeOH/THF
3.59 mmol, the whole amount of the product of the former synthesis),
trifluoroacetic acid (38 mL), and dichloromethane (115 mL). The mix-
ture was stirred at room temperature for 22 h then added to distilled
water (450 mL). After extraction (CH2Cl2, 50 mL), the combined or-
ganic phase was concentrated under reduced pressure to afford 6-(3-(1-
adamantyl)-4-anisyl)-2-naphthoic acid (adapalene) in 100 % yield
(1.48 g, 3.59 mmol) as a pale yellow solid. 1H NMR (400 MHz, DMSO-
d6) δ 1.76 (s, 6 H) 2.07 (s, 3 H), 2.14 (s, 6 H), 3.87 (s, 3 H), 7.12 (d, J
=8.8 Hz, 1 H), 7.58 (d, J =2.4 Hz, 1 H), 7.66 (dd, J = 2.4, 8.4 Hz, 1 H),
7.89 (dd, J = 2.0, 8.4 Hz, 1 H), 7.98 (dd, J = 1.6, 8.4 Hz, 1 H), 8.08 (d,
J =8.8 Hz, 1 H), 8.16 (d, J =8.8 Hz, 1 H), 8.22 (s, 1 H), 8.58 (s, 1 H),
13.05, br, 1 H). The spectrum was in good agreement with a previous
was accomplished at 60 °C, which indicated the novel column reactor
filled with HBPSG-Pd has higher reactivity than the previous one.
Since the optimized condition (Table 1 entry 9) was in hand, we
demonstrated continuous production using this column reactor. The
solution containing 1, 2 (3 eq), and KOH in 20 % methanol/THF was
passed through 10 cm of HBPSG-Pd column reactor dipped in 60 °C
water bath with a 1 mL/min flow rate. The 100 mL solution was con-
tinuously introduced and an aliquot was collected every 10 min. The
results are shown in Fig. 1. During 80 min the reaction yield remained
around 80 %, strongly suggesting high productivity and high reusability
of this column reactor. It should be mentioned that after a long-time
flow synthesis, all palladium might be finally removed from the support
[6b] although the reaction efficiency was not diminished after 80 min
operation in our reactor. Therefore, the preferred long operation for
larger scale experiments would be considered as a combination with an
appropriate catalyst recycling technology in near future.
3. Results and discussion
3.1. Flow Suzuki–Miyaura coupling reaction of phenylboronic acid with 4-
iodobenzonitrile using HBPSG-Pd column reactor
First we tried flow Suzuki–Miyaura cross coupling reaction of
commercially available reagents using HBPSG-Pd column. The combi-
nation of phenylboronic acid (1) with 4-iodobenzonitrile (2) was se-
lected as reagents to compare catalyst activity to our previous report
[8b]. Since the support was silica gel, we avoided highly polar solvent
such as water. Thus, we chose to use THF, toluene, and co-solvent of
methanol with THF as reaction solvent. In those solvent conditions,
inorganic base, which is typically used for conventional Suzuki–-
Miyaura coupling reaction, was not applicable because they are not
soluble to such less polar solvents. Thus, we used organic base such as
triethylamine (Et3N) and tri-n-butylammonium fluoride (TBAF) instead
of the inorganic base. The investigation was carried out using 10 cm
column reactor in a 60 °C water bath. The solution of 2, 1 (X equiv to 2),
and base (2 equiv to 1) in the respective solvent was introduced into the
column reactor by syringe pumps at the flow rate of 1 mL/min. The
Entries 1–5 of Table 1 indicate the combination of less polar sol-
vents with organic base was not applicable for flow Suzuki–Miyaura
cross coupling reaction. Only entry 6 (10 % MeOH/THF with TBAF)
yielded an amount of product 3 but the yield rapidly decreased during
continuous collection of the reacted solution (62 % to 44 %), suggesting
this condition may seriously damage the column reactor. Although
those reactions resulted in dismal yields, the trial of TBAF suggested
adequate reaction conditions would achieve a high yield of 3, so we
continued to optimize reaction conditions. We found that potassium
hydroxide, an inorganic base, was soluble to 10 % MeOH/THF solvent,
and using this condition, 45 % of product was obtained (entry 7). The
yield was increased when the amount of 1 was increased (entry 8).
Finally, 83 % yield was achieved with 3 equivalent of phenylboronic
acid in 20 % methanol/THF solution (entry 9). While our previous
column reactor required temperatures above 80 °C for over 80 % yield
of this particular transformation [8b], in the present study 83 % yield
3.2. Flow Suzuki–Miyaura coupling reaction of phenylborate with 4-
iodobenzonitrile using HBPSG-Pd column reactor
We next investigated the integration of lithiation, borylation, and
coupling reaction to enhance the process. We previously reported
Suzuki–Miyaura cross coupling reaction using lithium arylborate com-
plex, which is synthesized from halogen–lithium exchange reaction of
aryl halide with butyllithiums followed by reaction with boronic esters.
Because the products are lithium arylborate complexes, Suzuki reaction
can proceed without a base [13]. We envisaged that if our new column
reactor was applicable for the coupling reaction of such aryl borates, a
high productive integrated coupling process would be achieved.
Lithium triisopropoxy phenylborate (4) was synthesized from the
reaction of phenyl lithium, which was generated by halogen–lithium
exchange of bromobenzene with n-butyllithium, with triisopropyl bo-
rate, shown in Scheme 1. To the solution, 4-iodobenzonitrile (2) and
methanol were added to make the concentration of 2 to 0.033 M. The
resultant solution was introduced into the HBPSG-Pd column reactor
(10 cm) by syringe pump at the respective flow rates. The results are
The initial condition (flow rate: 0.5 mL/min, reaction temperature:
60 °C, entry 1) gave an excellent result where 2 was perfectly converted
to the desired product 3. Encouraged by this result, we increased the
flow rate to 1.0 mL/min in order to increase the volume per unit time.
However, disappointingly, the product yield was dramatically de-
creased, leading to decreased productivity (entry 2). Then we raised the
reaction temperature (entries 3 and 4). When applying higher tem-
perature, product yields were increased, and finally 99 % yield was
achieved at 97 °C (entry 4), doubling productivity. As expected, this
temperature produced higher flow rate with high productivity and in-
creased flow rates (entries 5–9). As anticipated, productivity increased
3