Microflow fluorinations of benzynes: Efficient synthesis of fluoroaromatic compounds

Article abstract of DOI:10.1248/cpb.c18-00578

Fluorinated aromatic compounds are found in a variety of biologically active compounds, including clinical drugs and agrochemicals. Therefore, the synthesis of aryl fluorides is particularly important in the medicinal and process chemistry fields. In this

Full text of DOI:10.1248/cpb.c18-00578

Vol. 66, No. 12
Chem. Pharm. Bull. 66, 1153–1164 (2018)
1153
Regular Article
Highlighted Paper selected by Editor-in-Chief
Microflow Fluorinations of Benzynes: Efficient Synthesis of
Fluoroaromatic Compounds
Takashi Ikawa,* Shigeaki Masuda, and Shuji Akai*
Graduate School of Pharmaceutical Sciences, Osaka University; 1–6 Yamadaoka, Suita, Osaka 565–0871, Japan.
Received July 30, 2018; accepted September 20, 2018
Fluorinated aromatic compounds are found in a variety of biologically active compounds, including
clinical drugs and agrochemicals. Therefore, the synthesis of aryl fluorides is particularly important in the
medicinal and process chemistry fields. In this paper, we report a method for the synthesis of aryl fluorides
by benzyne fluorination under microflow conditions using the efficient Comet X-01 micromixer. In compari-
son to our previously reported method under ordinary batch conditions, this approach facilitates a signifi-
cant reduction in reaction times, to ca. 10s, as well as increases in the yields of fluoroarenes (by up to 51%).
Key words benzyne; microflow reaction; aromatic fluorination; nonafluorobutanesulfonate; silyl group; tetra-
butylammonium fluoride
A variety of fluorinated compounds are employed in the (60–86%) of the aryl fluorides.
pharmaceutical and agrochemical industries.1–8) In particular,
In this paper, we report an extension of our previous ben-
aryl fluorides, such as enzalutamide and gefitinib, among oth- zyne batch-fluorination system,35) to microflow synthesis63)
ers, are the most widely used in clinical settings and have re- (Chart 1-B). The latter method not only achieved very short
ceived considerable attention from medicinal chemists.9) How- reaction times (ca. 10s) at room temperature, but also increas-
ever, the construction of C(sp2)–F bonds on arenes remains es in the yields of fluorinated benzenes 1 of up to 51%. In ad-
an ongoing challenge as most available methodologies employ dition, some fluorinated products 1 were exclusively obtained
harsh reaction conditions.10–15) Therefore, the incorporation of under microflow conditions.
fluorine in biologically active aromatic compounds has most
easily been achieved at the onset of the synthesis through the
use of fluorinated building blocks. Given these limitations, the
development of efficient methods for the fluorination of func-
tionalized aromatic compounds under mild conditions has re-
ceived increased attention over the latest decade.16–27) In fact,
our research group recently developed a method for the direct
preparation of fluorobenzenes 1 from 2-(trimethylsilyl)phenols
2
28); this method involves O-nonafluorobutanesufonylation fol-
lowed by the generation of the benzyne 4,29–34) and the imme-
diate nucleophilic addition of a fluoride ion at 4, to yield 1. In
this method, Bu4NF(t-BuOH)4 serves a dual role that generates
the benzyne and then fluorinates it35) (Chart 1-A). However,
further studies were required to improve this method as the
yields of aryl fluorides were relatively low, owing to the poor
nucleophilicity of the fluoride ion and the instability of 4.
Continuous flow synthesis has become more prevalent in the
field of organic synthesis, both in industry and academia.36–59)
Flow chemistry addresses many of the challenges faced in
standard synthetic procedures as it allows for efficient mixing,
easy scale-up, safe handling of hazardous chemicals, and rig-
orous temperature management. In the context of fluorination
chemistry, microflow systems enable the safe and efficient use
of poisonous fluorine gas for aromatic fluorination. Moreover,
flow systems prevent the formation of by-products through
the efficient release of the heat associated with exothermic
fluorination reactions.60,61) In the meantime, Buchwald and
colleagues reported the use of a CsF packed-bed microflow
Chart 1. (A) Our Previous Work: Benzyne Fluorination under Batch
Conditions; (B) This Work: Benzyne Fluorination under Microflow Con-
ditions
Results and Discussion
In the first instance, we investigated two types of mixers,
reactor for the Pd-catalyzed fluorination of aryl triflates.62) namely T-shaped (α06 or α04) and Comet X-01,64–66) for these
Although a relatively short residence time (20min) is required, fluorination reactions. The generation of 3,5-di-tert-butyl-
this method still requires a large excess of expensive CsF and benzyne (4a) from 2,4-di-tert-butyl-6-(trimethylsilyl)phenyl
temperatures as high as 120°C to obtain reasonable yields nonafluorobutanesulfonate (nonaflate) (3a) and the subse-
*To whom correspondence should be addressed. e-mail: ikawa@phs.osaka-u.ac.jp; akai@phs.osaka-u.ac.jp
© 2018 The Pharmaceutical Society of Japan

1154
Chem. Pharm. Bull.
Vol. 66, No. 12 (2018)
quent fluorination using 2.2 equiv of Bu4NF(t-BuOH)4 were the Comet X-01 mixer was used (we assumed that the con-
monitored (Table 1, Entries 1–12). The concentrations of 3a sumed 3a was completely transformed into 4a, except for that
in tetrahydrofuran (THF) and Bu4NF(t-BuOH)4 in THF were converted into 5a), which was significantly higher than that
set to 0.10 and 0.22M, respectively. The flow rates of 3a (V3a) in Entry 2. These results suggest that the generated benzyne
and Bu4NF(t-BuOH)4 (VF) were set to be the same (V3a=VF) in molecules were more efficiently trapped by the fluoride ion
most of all trial reactions. The residence time was determined using the Comet X-01 than the α06 mixer. Hence, the Comet
by the length of the outlet tube, which had an inner diameter X-01 mixer was chosen for further optimization studies. The
of 0.96mm. All reactions provided the desired, mono-fluori- use of a longer outlet tube (224cm) resulted in an improved
nated product, 1,3-di-tert-butyl-5-fluorobenzene (1a); however, yield of 1a (76%), while the recovered 3a decreased to only
the formation of the Thia-Fries rearrangement side-product 2% (Entry 8). These results suggest that both benzyne gen-
5a could not be avoided under most of all conditions. In all eration and fluorination proceeded in the Comet X-01 mixer
cases, the starting material 3a was recovered to a greater or as well as in the outlet tube. It is worth noting that the yield
lesser. Nevertheless, differences in the yields of the desired of 1a (76%) under flow conditions (Entry 8) was higher than
product 1a were noted. The yield of 1a was 44% at a flow rate that under the original batch conditions (55% yield of 1a over
(V3a=VF) of 2.0mL/min using the α06 T-shaped mixer at room a reaction time of 10s, see: Entry 13).35,67) This difference
temperature (Entry 1). Increases in the flow rate to 5.0mL/min is attributed mainly to the greater fluorination efficiency of
and 8.0mL/min resulted in increases in the yield of 1a to 54 4a under these optimized flow conditions (yield of 1a based
and 56%, respectively (Entries 2 and 3). In contrast, the use of on generated 4a: 97% (Entry 8) and 68% (Entry 13)). The
the α04 T-shaped mixer at a flow rate (V3a=VF) of 5.0mL/min Thia-Fries rearrangement was completely suppressed at 0°C,
resulted in a dramatic drop in the yield of 1a to 7% (Entry 4); however, the yield dropped to 57% (70% based on generated
therefore, the use of this mixer was not further pursued. The 4a) (Entry 9). A higher temperature, such as 60°C, resulted in
52% yield of 1a obtained using the Comet X-01 mixer (Entry a decrease in the yield of 1a to 68% (84% based on generated
7), at the same flow rate of 5.0mL/min, was comparable to 4a) (Entry 10).
the 54% yield obtained using α06 (Entry 2); however, the
Next, we examined a similar reaction with a 1:3 flow rate
yield of 5a and the recovery of 3a increased to 16 and 28%, (V3a=2.5mL/min, VF=7.5mL/min) of the solution of 3a (0.20M
respectively. Nevertheless, our attention was drawn to the high in THF) and that of Bu4NF(t-BuOH)4 (0.147 M in THF) with
yield (93%) of 1a (Entry 7), based on the generated 4a, when keeping the molar ratio of the combined solution and the total
Table 1. Optimizing the Reaction Conditions for Benzyne Fluorination under Flow Conditions
V (mL/min)
Yield (%)c)
Length
(cm)b)
Residence Time
(s)
Yield of 1a based
on generated 4ad)
Entry
Mixera)
V3a
VF
1a
5a
Recovered 3a
1
2
α06
90
90
2.0
5.0
2.0
5.0
9.8
3.9
2.4
3.9
10.4
5.2
1.0
9.7
9.7
9.7
1.0
9.7
44
54
7
8
4
6
49
63
64
8
α06
3
α06
90
8.0
8.0
56
10
8
2
4
α04
90
5.0
5.0
7
2
5
Comet X-01
Comet X-01
Comet X-01
Comet X-01
Comet X-01
Comet X-01
Comet X-01
Comet X-01
24
0.50
1.0
0.50
1.0
39
20
17
16
20
nd
18
15
16
11
24
28
2
57
92
93
97
70
84
91
95
6
24
54
7
24
5.0
5.0
52
76 (74)e)
8
224
224
224
24
5.0
5.0
9f )
10g)
11
12
5.0
5.0
57
19
1
5.0
5.0
68
2.5h)
2.5h)
7.5h)
7.5h)
71
7
224
78
2
13
Batchi)
—
—
—
10
55 (53)e)
19
ndj)
68
a) A solution of 3a in THF (0.10M, 5.0mL/min) and a separate solution of Bu4NF(t-BuOH)4 in THF (0.22M, 5.0mL/min) were mixed at room temperature using the
α06, α04, and Comet X-01 mixers. b) Length of the outlet tube with an inner diameter of 0.96mm. c) Determined by 1H-NMR analysis of the crude reaction mixture using
1,1,2,2-tetrachloroethane as the internal standard. d) In determining the yield of 1a based on generated 4a, the total amount of recovered 3a, and 3a that was converted into 5a,
was first subtracted, and the remaining amount of 3a was then used to calculate the yield of 1a. e) Isolated yield. f) The reaction was conducted at 0°C. g) The reaction was
conducted at 60°C. h) A solution of 3a in THF (0.20M, 2.5mL/min) and a separate solution of Bu4NF(t-BuOH)4 in THF (0.147M, 7.5mL/min) were mixed. i) 2.0mmol of 3a
and 4.4mmol of Bu4NF(t-BuOH)4 were reacted at 60°C for 10s in a flask. j) Not detected.

Vol. 66, No. 12 (2018)
Chem. Pharm. Bull.
1155
flow rate in the outlet tube (V3a+F=10mL/min) (Entries 11 and a mixture of meta-1f and ortho-1f in a 6.4:1 ratio under both
12). Importantly, the initial reaction rate became obviously batch and flow conditions (Entry 5). In this case, the regio-
higher (Entry 11) than that at the 1:1 flow rate (Entry 7), chemistry is believed to be mainly controlled by the steric
whereas, the yield of 1a (78, 95% based on generated 4a) for bulkiness of the C3-silyl group of 4f.68)
the residence time of 9.7s using a longer outlet tube (224cm)
Despite extensive examples in the literature involving the
(Entry 12) was eventually comparable to that of Entry 8. reactions of halobenzynes with a range of arynophiles,79–92)
Complete regioselectivity was observed during the formation the nucleophilic fluorination of halobenzynes 4h–4k, gener-
of 1a, which is attributable to the steric effect of the large tert- ated from 3h, 3i, 3j′, and 3k′, with Bu4NF(t-BuOH)4 under
butyl group at the C3 position in 4a.
normal batch conditions produced complex mixtures, and
Having optimized the flow conditions (Table 1, Entry 8), the yields of the expected fluorobenzenes 1 were very poor
the substrate scope of the reaction was examined. For this (Table 3). In stark contrast, the reaction of 5-chloro-3-(tert-
purpose, the benzyne precursors, 2-(trimethylsilyl)phenyl non- butyldimethylsilyl)benzyne (4h), generated from 3h, with
aflates 3b–3f and triflate 3g′ were used and their immediate Bu4NF(t-BuOH)4 under the optimized flow conditions pro-
fluorinations were examined (Table 2). Symmetrical benzynes duced 1h in 51% yield as a mixture of two regioisomers
4b, 4c, and 4e, generated from 3b, 3c, and 3e (Entries 1, 2 and (meta-1h/ortho-1h=1.2:1, Entry 1). Multi-substituted fluoro-
4), and asymmetrically substituted benzynes 4d, 4f, and 4g, benzenes 1i (meta-1i/ortho-1i=1.5:1, Entry 2) and 1j (single
from 3d, 3f, and 3g′ (Entries 3, 5 and 6), respectively, were product, Entry 3) were similarly prepared from 3i and 3j′ in
trapped by fluoride ions to produce the corresponding fluori- yields of 53 and 50%, respectively. Despite the success of
nated aromatic compounds 1b–g. It is noteworthy that all re- this methodology, the diiodobenzyne precursor 3k′ could not
actions under these flow conditions provided fluoroarenes 1 in be converted into the fluorodiiodobenzene 1k, rather a com-
higher yields than those under batch conditions (Table 2). Fur- plex mixture of products resulted (Entry 4). The completely
thermore, perfect regioselectivities were observed during reac- regioselective formation of meta-1j from 4j is attributed to
tions involving 3-(benzyloxy)benzyne 4d and 3-borylbenzyne synergism involving the C3-bromine atom that acts as an
4g, to afford meta-1d and ortho-1g, respectively, under both inductive electron-withdrawing substituent and is sterically
batch and flow conditions (Entries 3 and 6). In these cases, bulky; consequently it directs the nucleophile into the least-
the reactive sites of the benzynes were completely controlled hindered meta position (Entry 3). The preference for meta-1h
by the electron-withdrawing and electron-donating inductive over ortho-1h (1.2:1 ratio) was lower than that observed for 1f
effects of the C3-alkoxy and C3-boryl group, respectively.68–78) (meta-1f/ortho-1f=6.4:1, Table 2, Entry 5), due to differences
The fluorination of 3-(tert-butyldimethylsilyl)benzyne 4f gave in the electronic natures of their C5-substituents. Indeed, the
Table 2. Comparing the Substrate Scopes of Benzyne Fluorination under Flow and Batch Conditionsa

1156
Chem. Pharm. Bull.
Vol. 66, No. 12 (2018)
Table 3. Scope and Limitations of Halogenated-Benzyne Fluorinationsa
chlorine atom in a 5-chlorobenzyne is known to preferentially fluoride (NfF) (0.30M) in MeCN/THF (2:1) was mixed with a
drive nucleophilic addition to the C2 position93,94); hence the 0.020M solution of Bu4NPh3SiF2 (TBAT) in MeCN/THF (2:1)
additional chlorine substituent in 4h hampers the preferential at 60°C to generate 2-(trimethylsilyl)phenyl nonaflate 3a in
formation of meta-1h. On the other hand, the 2,5-dioxoranyl the first mixer and outlet tube (10m, residence time of 94s).
group at the C5 position of 4f has very little effect on the This reaction mixture was subsequently mixed with a solution
observed regioselectivity; therefore, the silyl group mainly of Bu4NF(t-BuOH)4 (0.22 M) in THF in the second mixer and
drives the addition of the fluoride ion to the C1 position due outlet tube (224cm, residence time of 9.7s) to form benzyne
to its steric bulkiness. Similar logic can be used to explain 4a, which was immediately trapped by fluoride to afford aryl
the formation of the mixture of regioisomers observed for 1i fluoride 1a in 72% isolated yield.
(meta-1i/ortho-1i=1.5:1, Entry 2).
When compared to batch conditions, significant improve-
2-(Trimethylsilyl)phenyl trimethylsilyl ether 6a, which is ments in the yields of aryl fluorides 1 were observed under
a new benzyne precursor that was recently reported by our microflow conditions (Tables 1–3). The use of microflow
group,95) was also successfully used in the benzyne fluorina- chemistry addressed various challenges associated with these
tion reaction under microflow conditions using three syringes reactions. The generation of benzynes from 3 using Bu4NF(t-
with two Comet X-01 micromixers, as shown in Chart 2. BuOH)4 is very fast, even at room temperature, and the re-
Hence, a solution of 6a (0.20M) and nonafluorobutanesulfonyl sulting benzynes are extremely reactive, which can lead to
Chart 2. Generation of Benzyne 4a from Silyl Ether 6a and Its Fluorination under Flow Conditions

Vol. 66, No. 12 (2018)
Chem. Pharm. Bull.
1157
decomposition and/or self-polymerization. Furthermore, the reactor. The yields of aryl fluorides obtained under flow condi-
fluoride ion is a poor nucleophile, and the fluorination step tions were generally higher than those obtained under batch
can be reversible24) (Chart 3). Microflow conditions facilitate conditions. To the best of our knowledge, this is the first re-
highly efficient mixing, rapid fluoride addition, and immediate port involving fluoride-ion-mediated generation of benzynes
protonation, to provide good yields. On the other hand, under under microflow conditions. The use of this microflow fluo-
batch conditions (Table 1, Entry 13), the reaction mixture is rination procedure en route to biologically active molecules,
not completely homogenous in the microscopic sense, which and in other benzyne reactions, is currently underway in our
is more obvious during and immediately following the addi- laboratory.
tion of Bu4NF(t-BuOH)4 (Fig. 1-A). When the concentration
of the fluoride ion in the microscopic vicinity of a generated
Experimental
benzyne is low, the reactive benzyne molecule may polymer-
General All reactions were carried out under an atmo-
ize or react with other arynophiles such as adventitious water, sphere of argon or nitrogen. A flask containing a stirrer bar
generated phenols, and solvents, resulting in poor yields of 1. and fitted with a three-way stopcock was used as the reac-
In comparison, a microflow system at a high flow rate affords tor. Anhydrous THF and MeCN were purchased from Kanto
greater reaction-mixture homogeneity due to highly efficient Chemical Co., Inc., Japan, and purified with a GlassContour™
mixing. Hence, benzynes react rapidly with fluoride ions fol- solvent purification system (Nikko Hansen & Co., Ltd., Japan)
lowed by immediate protonation by t-BuOH (or adventitious using two columns packed with activated molecular sieves.
water) to produce 1 in high yields (Fig. 1-B). The application Commercial 18-crown-6 was purified by recrystallization
of two different flow rates may realize more efficient mix- from MeCN. Bu4NF(t-BuOH)4,103) benzyne precursors 3a–3f,
ing and made the benzyne generation and fluorination faster 3g′ and 3i35) were prepared according to a literature proce-
(Table 1, Entry 11). This efficient mixing was particularly dure. All other reagents were purchased from Kanto Chemical
beneficial during the microflow reactions of halogenated ben- Co., Inc., FUJIFILM Wako Pure Chemical Industries, Ltd.,
zynes 4h–j (Table 3, Entries 1–3), resulting in the suppression Japan, Tokyo Chemical Industry Co., Ltd., Japan, Sigma-
of characteristic side-reactions of halobenzynes, such as the Aldrich Co., LLC., U.S.A., Nacalai Tesque, Inc., Japan or
halogen dance reaction,96) Thia-Fries rearrangement,93,97,98) and Kishida Chemical Co., Ltd., Japan, and were used without fur-
the further formation of other benzynes.99–102) The reaction of ther purification. Flash chromatography was performed using
3,5-iodobenzyne (4k) seems to be too difficult to be achieved 60N spherical neutral (40–50µm) silica gel purchased from
even under microflow conditions probably because of the high Kanto Chemical Co., Inc. All reactions were monitored by
leaving ability of the iodo group.
TLC on glass-backed 0.2-mm silica gel 60F254 plates (Merck),
In summary, we developed a method for the synthesis of and compounds were visualized under UV light (254nm).
aromatic fluorinated compounds that involves the generation The Comet X-01 micromixing device and dual syringe pumps
of a benzyne and its immediate fluorination in a microflow (Catamaran HII-10) were obtained from Techno Applica-
tions Co., Ltd., 34-16-204 Hon, Denenchofu, Oota, Tokyo,
145-0072, Japan (e-mail: yukio-matsubara@nifty.com). The
α06 (channel width 600µm) and α04 (Channel width 400µm)
micromixing devices were purchased from MiChS Co., Ltd.,
Japan.
Analytical Methods Melting points were recorded on a
Büchi M-565 or a Yanagimoto melting point apparatus and
are uncorrected. IR spectra were acquired on a Shimadzu
FTIR-8400S and Shimadzu IRAffinity-1S. 1H-NMR and
Chart 3. Steps in the Fluorination of a Benzyne
13C-NMR spectra were recorded on a Jeol JMN-ECA-500
Fig. 1. Illustrating the Microscopic Differences between the Batch Reaction (A) and the Microflow Reaction (B)

1158
Chem. Pharm. Bull.
Vol. 66, No. 12 (2018)
Chart 4. Flow Reactor for Tables 1–3
(1H: 500MHz, 13C: 125MHz, 19F: 470MHz), a Jeol JMN- with hexane and washed with water. The aqueous phase was
ECS-400 (1H: 400MHz, 13C: 100MHz, 19F: 376MHz), or extracted twice with hexane. The combined organic phase
a Jeol ECS-300 (1H: 300MHz, 13C: 75MHz, 19F: 282MHz) was washed with a saturated NaCl solution and dried over
spectrometer. Chemical shifts are reported in ppm and are ref- anhydrous Na2SO4. The mixture was filtered and the solvents
erenced against the chemical shifts of residual protonated sol- were removed under reduced pressure. The crude product was
vent (1H) or deuterated solvent (13C) (an internal standard was purified by silica-gel flash column chromatography (eluent:
not used for the 19F-NMR spectra as the JMN-ECA and ECS hexane, a mixture of hexane and EtOAc, or CH2Cl2) to afford
instruments do not require this). GC was performed using a the fluorinated product 1.
Shimadzu GC-2010 instrument. Mass spectra were acquired
1,3-Di-tert-butyl-5-fluorobenzene (1a)104) Following the
on a Jeol JMS-S3000 (matrix assisted laser desorption/ioniza- General Procedure A (Flow Conditions (Table 1, Entry
tion (MALDI)), a Jeol JMS-700 (FAB and electron ionization 8)), 2,4-di-tert-butyl-6-(trimethylsilyl)phenyl nonaflate (3a)35)
(EI)), or a JMS-T100TD (APCI) spectrometer. “Yield” refers (0.10 M in THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) were
to the isolated yield of a compound exhibiting at most only mixed under microflow conditions, and the mixture was
1
trace peaks in its H-NMR spectrum that are not attributable collected for 24s [total volume 4.0mL, 0.20mmol of 3a,
1
to its assigned structure. H-NMR spectra and melting points 0.44mmol of Bu4NF(t-BuOH)4]. 1,1,2,2-Tetrachloroethane
(where applicable) of all known compounds were recorded. (21 µL, 0.20mmol) was added to a crude mixture, and the
All new products were further characterized by high-resolu- yields of the products were determined based on 1H-NMR
tion (HR)-MS.
data as follows, 1a: 76%, 5a: 20% and 3a: 2%. The crude
General Procedure A for the Optimized Flow Synthe- mixture was purified by flash column chromatography on sil-
sis of Aryl Fluoride 1 from Nonaflate 3 (Tables 1–3 and ica gel (hexane/EtOAc=20:1) to provide the titled compound
1
Chart 4) A 0.10M THF solution of 2-(trimethylsilyl)phenyl 1a (31mg, 0.15mmol, 74%) as a colorless oil. The H-NMR
nonaflate 3 (1.0 equiv) and a 0.22M THF solution of Bu4NF(t- spectra of the obtained material was identical to that in our
BuOH)4 (2.2 equiv) were mixed and reacted at room tempera- previous report.35)
ture in a continuous flow reactor consisting of a dual syringe
Following the General Procedure B (Batch Conditions (Table
pump set to 5.0mL/min, a Comet X-01 mixing element, 1, Entry 13)), a mixture of 2,4-di-tert-butyl-6-(trimethylsilyl)-
and an outlet tube (inner diameter 0.96mm, outer diameter phenyl nonaflate (3a)35) (1.1g, 2.0mmol), Bu4NF(t-BuOH)4
1.56mm, length 224cm, and residence volume 1.62mL). The (1.3g, 4.4mmol) in THF (40mL, 0.050M) was stirred for 10s
effective residence time was 9.7s. The crude reaction mixture at 60°C. 1,1,2,2-Tetrachloroethane (0.21mL, 0.20mmol) was
was collected into a test tube containing ca. 1mL of a saturat- added to a crude product (1H-NMR yield, 1a: 55%, 5a: 19%
ed NH4Cl solution for 24s [total volume 4.0mL, 0.20mmol of and 3a: not detected). The crude mixture was purified by flash
3, 0.44mmol of Bu4NF(t-BuOH)4] from at least 10s after the column chromatography on silica gel (hexane) to provide the
commencement of solution flow. The mixture was extracted titled compound 1a (0.22g, 1.1mmol. 53%) as a colorless oil.
1
with hexane and washed with water. The aqueous phase was The H-NMR spectra of the obtained material was identical to
extracted twice with hexane. The combined organic phase that in our previous report.35)
was washed with a saturated NaCl solution and dried over
2,4-Di-tert-butyl-6-(nonafluorobutanesulfonyl)phenol
anhydrous Na2SO4. The mixture was filtered and the solvent (5a) (Table 1, Entry 13) 5a was obtained from above men-
was removed under reduced pressure. The crude product was tioned reaction mixture as a white solid (0.14g, 0.29mmol,
1
purified by silica-gel flash column chromatography (eluent: 15%). Rf: 0.6 (hexane/EtOAc=10:1). mp: 64–66°C. H-NMR
hexane, a mixture of hexane and EtOAc, or CH2Cl2) to afford (400MHz, CDCl3) δ: 1.31 (9H, s), 1.42 (9H, s), 7.55 (1H, d,
the fluorinated product 1.
J=2.5Hz), 7.77 (1H, d, J=2.5Hz), 8.98 (1H, brs). 13C-NMR
General Procedure B for the Batch Synthesis of Aryl (100MHz, CDCl3) δ: 29.2, 30.3, 34.6, 35.8, 104.2–122.8
Fluoride 1 from Nonaflate 3 (Tables 1–3) A flame-dried (4C, m), 113.0, 125.5, 134.9, 139.6, 143.2, 156.1. 19F-NMR
flask was charged with 2-(trialkylsilyl)phenyl nonaflate 3 (1.0 (376MHz, CDCl3) δ: −126.08–(−125.82) (2F, m), −121.12–
equiv) and a stirrer bar, capped with a rubber septum, and (−120.82) (2F, m), −112.82–(−112.68) (2F, m), −80.81–
evacuated and back-filled with nitrogen. Anhydrous THF (−80.65) (3F, m). IR (neat): 3381, 1471cm−1. HR-MS (APCI):
(0.050M) was added via a syringe, and the mixture was heated m/z Calcd for C18H22F9O3S [M]+: 488.1062. Found: 488.1066.
to 60°C. Bu4NF(t-BuOH)4 (2.2 equiv) was quickly added by
4-Fluoro-1,2-dimethoxybenzene (1b) (Table 2, Entry
opening the septum. After stirring at 60°C for the indicated 1)35) Following the General Procedure A (Flow Conditions),
amount of time, the reaction mixture was cooled to room 4,5-dimethoxy-2-(trimethylsilyl)phenyl nonaflate (3b)35) (0.10 M
temperature and then passed through a short pad of silica in THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) were mixed
gel using EtOAc as the eluent. The mixture was extracted under microflow conditions, and the mixture was collected

Vol. 66, No. 12 (2018)
Chem. Pharm. Bull.
1159
for 12s [total volume 2.0mL, 0.10mmol of 3b, 0.22mmol of General Procedure A (Flow Conditions), 2-(1H-naphtho[1,8-
Bu4NF(t-BuOH)4]. n-Decane (19µL, 0.10mmol) was added de][1,3,2]diazaborinin-2(3H)-yl)-6-(trimethylsilyl)phenyl triflate
to the reaction mixture and diluted with EtOAc (ca. 5mL). (3g′)35) (0.10 M in THF) and Bu4NF(t-BuOH)4 (0.22M in THF)
A part of the mixture was filtered through a silica gel pad were mixed under microflow conditions, and the mixture
and the filtrate was measured by GC (91% GC yield of 1b). was collected for 24s [total volume 4.0mL, 0.20mmol of 3g′,
The crude product was purified by flash column chromatog- 0.44mmol of Bu4NF(t-BuOH)4]. The crude product was puri-
raphy on silica gel (hexane/Et2O=1:1) to provide the titled fied by flash column chromatography on silica gel (hexane/
compound 1b (11mg, 70µmol, 71%) as a colorless oil. The EtOAc=5:1) to provide the titled compound ortho-1g (34mg,
1
1H-NMR spectra of the obtained material was identical to that 0.13mmol, 66%) as a colorless solid. The H-NMR spectra of
in our previous report.35)
the obtained material was identical to that in our previous re-
4-Fluoro-1,2-bis(benzyloxy)benzene (1c) (Table 2, Entry port.35)
2)35) Following the General Procedure A (Flow Conditions),
1-Fluoro-3-chloro-5-(tert-butyldimethylsilyl)benzene (meta-
4,5-bis(benzyloxy)-2-(trimethylsilyl)phenyl nonaflate (3c)35) 1h) and 1-Fluoro-4-chloro-2-(tert-butyldimethylsilyl)benzene
(0.10 M in THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) were (ortho-1h) (Table 3, Entry 1) Following the General Proce-
mixed under microflow conditions, and the mixture was col- dure A (Flow Conditions), 4-chloro-2-(tert-butyldimethylsilyl)-6-
lected for 24s [total volume 4.0mL, 0.2mmol of 3c, 0.44mmol (trimethylsilyl)phenyl nonaflate (3h) (0.10M in THF) and
of Bu4NF(t-BuOH)4]. The crude product was purified by flash Bu4NF(t-BuOH)4 (0.22M in THF) were mixed under microflow
column chromatography on silica gel (hexane/EtOAc=10:1) to conditions, and the mixture was collected for 24s [total vol-
provide the titled compound 1c (46mg, 0.15mmol, 74%) as a ume 4.0mL, 0.20mmol of 3h, 0.44mmol of Bu4NF(t-BuOH)4].
1
colorless solid. The H-NMR spectra of the obtained material The crude product (meta-1h/ortho-1h=1.2:1, determined by
1
was identical to that in our previous report.35)
300MHz H-NMR analysis) was purified by flash column chro-
1-Fluoro-3,5-bis(benzyloxy)benzene (meta-1d) (Table 2, matography on silica gel (hexane/EtOAc=5:1) to provide a
Entry 3)35,105) Following the General Procedure A (Flow 1.3:1 mixture of the titled compounds, meta-1h and ortho-1h
1
Conditions), 3,5-bis(benzyloxy)-2-(tert-butyldimethylsilyl)phenyl (24mg, 0.11mmol, 51%) as a colorless oil. H-NMR (300MHz,
nonaflate (3d)35) (0.10 M in THF) and Bu4NF(t-BuOH)4 (0.22M CDCl3) δ: 0.26 (6H×4/9, s), 0.30 (6H×5/9, s), 0.86 (9H×4/9, s),
in THF) were mixed under microflow conditions, and the mix- 0.88 (9H×5/9, s), 6.85–6.95 (1H×4/9, m), 7.03–7.08 (2H×5/9,
ture was collected for 24s [total volume 4.0mL, 0.20mmol of m), 7.20 (1H×5/9, dd, J=1.5, 1.5Hz), 7.25–7.30 (2H×4/9, m).
3d, 0.44mmol of Bu4NF(t-BuOH)4]. The crude product was 13C-NMR (100MHz, CDCl3) δ: −6.3, −5.4, 16.8, 17.2, 26.3,
purified by flash column chromatography on silica gel (hex- 26.5, 116.37 (d, J=29.0Hz), 116.43 (d, J=24.0Hz), 119.0 (d,
ane/EtOAc=10:1) to provide the titled compound meta-1d J=18.0Hz), 126.3 (d, J=33.5Hz), 129.0 (d, J=2.5Hz), 129.8 (d,
(45mg, 0.15mmol, 73%) as a colorless solid. Rf: 0.5 (hexane/ J=2.5Hz), 130.9 (d, J=8.5Hz), 134.5 (d, J=8.5Hz), 135.7 (d,
1
EtOAc=6:1). mp: 89–92°C. The H-NMR spectra of the ob- J=12.0Hz), 142.8 (d, J=9.5Hz), 162.2 (d, J=252.0Hz), 165.6 (d,
tained material was identical to that in our previous report.35)
J=241.0Hz). 19F-NMR (283MHz, CDCl3) δ: −112.0–(−111.8)
4-Fluoro-1,2-dimethylbenzene (1e) (Table 2, Entry (m), −100.5–(−100.3) (m). IR (neat): 1574, 1253, 1231cm−1.
4)35) Following the General Procedure A (Flow Conditions), HR-MS (EI) Calcd for C12H1835ClFSi [M]+: 244.0845. Found
4,5-dimethyl-2-(trimethylsilyl)phenyl nonaflate (3e) (0.10M in 244.0849, 244.0855 (Two regioisomers, ortho-1h and meta-
THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) were mixed under 1h were separated by gas chromatography and measured by
microflow conditions, and the mixture was collected for 12s HR-MS).
[total volume 2.0mL, 0.1mmol of 3e, 0.22mmol of Bu4NF(t-
Following the General Procedure B (Batch Condi-
BuOH)4]. n-Decane (19µL, 0.10mmol) was added to the reac- tions), a mixture of 4-chloro-2-(tert-butyldimethylsilyl)-6-
tion mixture and diluted with EtOAc (ca. 5mL). A part of the (trimethylsilyl)phenyl nonaflate (3h) (0.12g, 0.20mmol),
mixture was filtered through a silica gel pad and the filtrate Bu4NF(t-BuOH)4 (0.25g, 0.44mmol) in THF (4.0mL, 0.050M)
was measured by GC (57% GC yield of 1e). The retention was stirred for 1h at 60°C. However, the titled compounds,
time was identical with the commercial one.
meta-1h and ortho-1h could not be detected in a crude mix-
1
3-(1,3-Dioxolan-2-yl)-1-fluoro-5-(tert-butyldimethylsilyl)- ture by H-NMR spectroscopy.
benzene (meta-1f)35) and 4-(1,3-Dioxolan-2-yl)-1-fluoro-2-(tert-
1-Fluoro-3-iodo-5-(trimethylsilyl)benzene (meta-1i) and
butyldimethylsilyl)benzene (ortho-1f)35) (Table 2, Entry 5) 1-Fluoro-4-iodo-2-(trimethylsilyl)benzene (ortho-1i) (Table
Following the General Procedure A (Flow Conditions), 2-(tert- 3, Entry 2) Following the General Procedure A (Flow Con-
butyldimethylsilyl)-4-(1,3-dioxolan-2-yl)-6-(trimethylsilyl)- ditions), 4-iodo-2,6-bis(trimethylsilyl)phenyl nonaflate (3i)35)
phenyl nonaflate (3f)35) (0.10 M in THF) and Bu4NF(t-BuOH)4 (0.10 M in THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) were
(0.22M in THF) were mixed under microflow conditions, mixed under microflow conditions, and the mixture was
and the mixture was collected for 24s [total volume 4.0mL, collected for 24s [total volume 4.0mL, 0.20mmol of 3i,
0.20mmol of 3f, 0.44mmol of Bu4NF(t-BuOH)4]. The crude 0.44mmol of Bu4NF(t-BuOH)4]. The crude product (meta-
1
product (meta-1f/ortho-1f=6.4:1, determined by 300MHz 1i/ortho-1i=1.5:1, determined by 300MHz H-NMR analysis)
1H-NMR analysis) was purified by flash column chromatog- was purified by flash column chromatography on silica gel
raphy on silica gel (hexane) to provide a 6.4:1 mixture of the (hexane/EtOAc=5:1) to provide a 2:1 mixture of the titled
titled compounds, meta-1f and ortho-1f (33mg, 0.12mmol, compounds, meta-1i and ortho-1i (31mg, 0.11mmol, 53%) as
60%) as a green oil. The 1H-NMR spectra of the obtained mate- a brown oil. 1H-NMR (300MHz, CDCl3) δ: 0.25 (9H×2/3,
rial was identical to that in our previous report.35)
s), 0.29 (9H×1/3, m), 6.75 (1H×1/3, dd, J=9.0, 8.5Hz), 7.13
2-(2-Fluorophenyl)-2,3-dihydro-1H-naphtho[1,8-de] [ 1 ,3 ,2 ] - (1H×2/3, ddd, J=8.5, 2.5, 1.5Hz), 7.39 (1H×2/3, ddd, J=8.5,
diazaborinine (ortho-1g)35) (Table 2, Entry 6) Following the 2.5, 1.5Hz), 7.54–7.57 (1H×2/3, m), 7.58–7.64 (2H×1/3, m).

1160
Chem. Pharm. Bull.
Vol. 66, No. 12 (2018)
Chart 5. Flow Reactor for Chart 2
13C-NMR (100MHz, CDCl3) δ: −1.3, −1.1, 88.2 (d, J=3.0Hz), The effective residence time for step 1 was 94s, and for step
84.5 (d, J=6.5Hz), 117.5 (d, J=27.0Hz), 119.1 (d, J=18.0Hz), 2 was 9.7s. The crude reaction mixture was collected into
125.0 (d, J=23.0Hz), 130.2 (d, J=33.0Hz), 137.7 (d, J=3.5Hz), ca. 1mL of a saturated NH4Cl solution for 24s [total vol-
140.0 (d, J=8.0Hz), 143.7 (d, J=11.5Hz), 146.3 (d, J=3.5Hz), ume 4.0mL, 0.20mmol of 6a, 0.30mmol of NfF, 20µmol of
162.0 (d, J=255.0Hz), 167.2 (d, J=242.0Hz). IR (neat): 1277, Bu4NPh3SiF2, 0.44mmol of Bu4NF(t-BuOH)4] from at least
1258cm−1. HR-MS (FAB, NBA) Calcd for C9H12FSiI [M]+: 100s after the commencement of solution flow. The mixture
293.9731. Found 293.9730.
was extracted with hexane (three times), and the combined
Following the General Procedure B (Batch Conditions), a organic phase was dried over anhydrous Na2SO4. The organic
mixture of 4-iodo-2,6-bis(trimethylsilyl)phenyl nonaflate (3i)35) phase was filtered and concentrated under reduced pressure.
(0.26g, 0.40mmol), Bu4NF(t-BuOH)4 (0.49g, 0.88mmol) in The crude product was purified by flash column chromatogra-
THF (8.0mL, 0.050M) was stirred for 1h at 60°C. However, phy (hexane/EtOAc=20:1) to provide 1,3-di-tert-butyl-5-fluo-
only 5% of the titled compounds, meta-1i and ortho-1i was robenzene 1a104) as a colorless solid (30mg, 0.14mmol, 72%).
1
1
detected in a crude mixture by H-NMR spectroscopy using The H-NMR spectrum of the obtained material was identical
1,1,2,2-tetrachloroethane as an internal standard (meta-1i– to that in our previous report.35)
ortho-1i=1.5:1).
General Procedures C and D for the Synthesis of
1-Bromo-3-tert-butyl-5-fluorobenzene (meta-1j)106) (Table Benzyne Precursors 3h, 3j′ and 3k′
3, Entry 3) Following the General Procedure A (Flow Con-
ditions), 2-bromo-4-tert-butyl-6-(trimethylsilyl)phenyl triflate
General Procedure C
An oven dried flask was charged with 2-bromophenol (1.0
(3j′) (0.10M in THF) and Bu4NF(t-BuOH)4 (0.22 M in THF) equiv) and capped with an inlet adapter with a 3-way stopcock
were mixed under microflow conditions, and the mixture and then evacuated and back-filled with argon. Anhydrous
was collected for 48s [total volume 8.0mL, 0.40mmol of 3j′, THF (0.10–0.50M), Et3N (1.5 equiv) and Me3SiCl (1.5 equiv)
0.88mmol of Bu4NF(t-BuOH)4]. The crude product (meta- were added via syringes and the reaction mixture was stirred
1
1j/ortho-1j=>98:2, determined by 300MHz H-NMR analy- for a few hours at room temperature. The reaction mixture
sis) was purified by flash column chromatography on silica was concentrated under reduced pressure. Hexane was added
gel (hexane/CH2Cl2=100:1) to provide the titled compound to the residue and the mixture was filtered through a silica gel
meta-1j (46mg, 0.20mmol, 50%) as a brown oil. 1H-NMR pad and washed with hexane. The solution was evaporated to
(300MHz, CDCl3) δ: 1.30 (9H, s), 7.01 (1H, ddd, J=11.0, 2.5, give 2-bromophenyl trimethylsilyl ether. Without purification
1.5Hz), 7.05 (1H, ddd, J=8.0, 2.5, 1.5Hz), 7.28 (1H, dd, J=1.5, of the obtained material, anhydrous THF (0.10–0.33M) was
1.5 Hz).
added to the flask and the mixture was cooled to −78°C. n-
Following the General Procedure B (Batch Conditions), BuLi (1.6M hexane solution, 1.2 equiv) was added dropwise at
mixture of 2-bromo-4-tert-butyl-6-(trimethylsilyl)phenyl −78°C and the reaction was allowed to warm up to room tem-
a
triflate (3j′) (87mg, 0.20mmol), Bu4NF(t-BuOH)4 (0.25g, perature and stirred for several hours. To the reaction mixture
0.44mmol) in THF (4.0mL, 0.050M) was stirred for 1h at was added a saturated aqueous solution of NH4Cl for quench-
60°C. However, the titled compound meta-1j could not be de- ing. The mixture was extracted with EtOAc (this process was
1
tected in a crude mixture by H-NMR spectroscopy.
repeated three times) and combined organic phase was dried
Procedure for the Flow Synthesis of 1,3-Di-tert-butyl- over anhydrous Na2SO4. The organic phase was filtered and
5-fluorobenzene (1a)104) from Silyl Ether 6a (Charts 2 and concentrated under reduced pressure. The residue was purified
5) A MeCN/THF (2:1) solution of 2-(trimethylsilyl)-4,6- by flash column chromatography on silica gel (hexane/EtOAc)
di-tert-butylphenyl trimethylsilyl ether (6a) and NfF [6a: to provide 2-(trimethylsilyl)phenol 2.
0.20M (1.0 equiv), NfF: 0.30M (1.5 equiv), V=2.5mL/min]
and a 20mM solution of Bu4NPh3SiF2 (0.10 equiv) in THF
General Procedure D
An oven dried flask was charged with 2-(trialkylsilyl)phenol
(V=2.5mL/min) were mixed and reacted at 60°C in a continu- 2 (1.0 equiv), 18-crown-6 (1.0 equiv) and capped with rub-
ous flow reactor consisting of two dual syringe pumps, one set ber septum, and then evacuated and back-filled with argon.
to 2.5mL/min and the other to 5.0mL/min, two Comet X-01 Anhydrous THF (0.10M) and sodium hydride (NaH) (60% in
devices, an outlet tube with a 7.9mL residence volume (inner mineral oil, 1.5 equiv) was added into the flask, and the reac-
diameter 1.00mm, outer diameter 2.00mm, and length 10m), tion mixture was stirred for a few minutes. NfF (1.5 equiv)
and another outlet tube with a 1.6mL residence volume (inner was added via a syringe, and the resulting mixture was stirred
diameter 0.96mm, outer diameter 1.56mm, and length 2.24m). at 60°C. After the reaction completed, the mixture was cooled
To this stream was added a stream of Bu4NF(t-BuOH)4 (2.2 to 0°C. Water was added into the reaction mixture and the
equiv) in THF (0.22M, V=5.0mL/min) at room temperature. mixture was extracted with hexane (this process was repeated

Vol. 66, No. 12 (2018)
Chem. Pharm. Bull.
1161
three times). Organic phase was combined and it was dried to warm up to room temperature and stirred for 5h. To the
over anhydrous Na2SO4. The mixture was filtered and con- reaction mixture was added a saturated aqueous solution of
centrated under reduced pressure. The residue was purified by NH4Cl for quenching. The mixture was extracted with hexane
flash column chromatography on silica gel (hexane/EtOAc) to (this process was repeated three times) and combined organic
afford 2-(trialkylsilyl)phenyl nonaflate 3.
phase was dried over anhydrous Na2SO4. The organic phase
4,5-Dimethyl-2-(trimethylsilyl)phenyl Nonaflate (3e) was filtered and concentrated under reduced pressure. The
(Table 2, Entry 4) Following General Procedure C, a mix- residue was purified by flash column chromatography on silica
ture of 4,5-dimethyl-2-bromophenol35) (0.26g, 1.3mmol), Et3N gel (hexane) to provide 4-chloro-2-(tert-butyldimethylsilyl)-6-
(0.24mL, 1.7mmol), Me3SiCl (1.7mL, 20mmol) was stirred (trimethylsilyl)phenol (2h) (9.0g, 76%) as a colorless oil.
in anhydrous THF (4.3mL, 0.30M) for 1h at room tempera-
Following General Procedure D, a mixture of 4-chloro-2-
ture. To the obtained 2-bromophenyl trimethylsilyl ether were (tert-butyldimethylsilyl)-6-(trimethylsilyl)phenol (2h) (3.0g,
added THF (4.3mL, 0.30M) and n-BuLi (1.6M hexane solution, 9.7mmol), NaH (60% in mineral oil, 0.58g, 15mmol),
0.98mL, 1.6mmol), and stirred for 1h at room temperature. 18-crown-6 (2.5g, 9.7mmol) and NfF (2.5mL, 15mmol) was
The crude mixture was purified by flash column chromatogra- stirred in THF (50L, 0.20M) at reflux for 12h. The crude
phy on silica gel (hexane/EtOAc=10:1) to give 4,5-dimethyl- product was purified by column chromatography (hexane) to
2-bromophenyl trimethylsilyl ether (2e)35) as a colorless oil provide the titled compound 3h (4.9g, 85%) as a colorless
1
1
(0.23g, 90%). Rf: 0.5 (hexane). H-NMR (400MHz, CDCl3) δ: oil. Rf: 0.8 (hexane). H-NMR (500MHz, CDCl3) δ: 0.35 (9H,
0.27 (9H, s), 2.15 (6H, s), 6.65 (1H, s), 7.25 (1H, s). Following s), 0.39 (6H, s), 0.80 (9H, s), 7.47–7.49 (2H, m). 13C-NMR
General Procedure D, a mixture of 4,5-dimethyl-2-bromo- (125MHz, CDCl3) δ: −4.0, 0.5, 18.2, 26.9, 106.5–126.9 (4C,
phenyl trimethylsilyl ether (2e)35) (0.19g, 1.0mmol), NaH (60% m), 133.7, 135.3, 137.6, 137.9, 138.3, 152.7. 19F-NMR (470MHz,
in mineral oil, 0.12g, 3.0mmol), 18-crown-6 (0.79g, 3.0mmol) CDCl3) δ: −125.71–(−125.56) (m), −120.86–(−120.73) (m),
and NfF (0.52mL, 3.0mmol) was stirred in THF (5.0mL, −106.21–(−106.12) (m), −80.54–(−80.45) (m). IR (neat): 1403,
0.2M) at reflux for 5h. The crude reaction mixture was puri- 1570, cm−1. HR-MS (APCI): m/z Calcd for C19H2735ClF9O3SSi2
fied by flash column chromatography on silica gel (hexane) to [M+H]+: 597.0759. Found: 597.0781.
provide the titled compound 3e as a colorless oil (0.36g, 76%).
2-Bromo-4-tert-butyl-6-(trimethylsilyl)phenyl
Triflate
1
Rf: 0.8 (hexane). H-NMR (400MHz, CDCl3) δ: 0.26 (9H, s), (3j′) (Table 3, Entry 2) A round-bottom flask was charged
2.18 (3H, s), 2.19 (3H, s), 6.99 (1H, s), 7.15 (1H, s). 13C-NMR with 4-tert-butyl phenol (15g, 0.10mol) and stir bar. CH2Cl2
(100MHz, CDCl3) δ: 0.8, 19.1, 20.0, 106.5–119.4 (4C, m), was added and it was cooled to 0°C. Br2 (11mL, 0.22mol) was
120.6, 129.2, 136.1, 136.9, 140.4, 153.4. 19F-NMR (470MHz, added to the mixture over 10min. The mixture was wormed
CDCl3) δ: −125.82–(−125.64) (m), −120.79–(−120.64) (m), to room temperature and it was stirred for 4h. the mixture
−110.01–(−109.82) (m), −80.73–(−80.60) (m). IR (neat): 1422, was evaporated and the residue was purified by column chro-
1352 cm−1. HR-MS (APCI): m/z Calcd for C15H18F9O3SSi matography (hexane/AcOEt=10:1) to provide 2,6-dibromo-
1
[M+H]+: 477.0597. Found: 477.0626.
4-tert-butylphenol108) (29g, 92%) as a white solid. H-NMR
4-Chloro-2-(tert-butyldimethylsilyl)-6-(trimethylsilyl)- (400MHz, CDCl3) δ: 1.26 (9H, s), 5.71 (1H, brs), 7.42 (2H,
phenyl Nonaflate (3h) (Table 3, Entry 1) Following Gen- s). An oven dried flask was charged with 2,6-dibromo-
eral Procedure C, a mixture of 2,6-dibromo-4-chlorophenol107) 4-tert-butylphenol108) (5.0g, 16mmol) and capped with an
(14g, 50mmol) was dissolved in THF (0.20L, 0.25M). Me3SiCl inlet adapter with a 3-way stopcock and then evacuated and
(9.5mL, 75mmol) and Et3N (11mL, 75mmol) were added to back-filled with argon. Anhydrous THF (81mL, 0.20M), Et3N
the solution and stirred at room temperature for 1h. The reac- (3.4mL, 24mmol) and Me3SiCl (3.1mL, 24mmol) were added
tion mixture was concentrated under reduced pressure. Hexane via syringes and the reaction mixture was stirred for 1h at
was added to the residue and the mixture was filtered through room temperature. The reaction mixture was concentrated
a silica gel pad and washed with hexane. The solution was under reduced pressure. Hexane was added to the residue
evaporated to give 2,6-dibromo-4-chlorophenyl trimethylsilyl and the mixture was filtered through a silica gel pad and
ether (18g, 98%) as a colorless oil. Without purification of the washed with hexane. The solution was evaporated to give
obtained material, anhydrous THF (0.25L, 0.2M) was added 2,6-dibromo-4-tert-butylphenyl trimethylsilyl ether. Without
to the flask and the mixture was cooled to −78°C. n-BuLi purification of the obtained material, anhydrous Et2O (0.16L,
(2.5 M hexane solution, 24mL, 1.2 equiv) was added dropwise 0.10 M) was added to the flask and the mixture was cooled to
at −78°C and the reaction was allowed to warm up to room −78°C. n-BuLi (2.6M hexane solution, 6.8mL, 18mmol) was
temperature and stirred for 1h. Then the mixture was added added dropwise at −78°C and stirred for 1h. To the reaction
tert-butyldimethylsilyl chloride (9.0g, 60mmol) and stirred mixture was added Tf2O (4.0mL, 24mmol) and stirred for ad-
for 12h. Water (ca. 0.15L) was added to the mixture and it ditional 4h at −78°C. Saturated aqueous solution of NaHCO3
was extracted with hexane (three times). The organic phase was added to the reaction mixture for quenching and it was
was dried over anhydrous Na2SO4 and concentrated under warmed to room temperature. The mixture was extracted
reduced pressure. The crude product was purified by column with hexane (this process was repeated three times) and
chromatography (hexane) to provide 4-chloro-6-bromo-2- combined organic phase was dried over anhydrous MgSO4.
(trimethylsilyl)phenyl tert-butyldimethylsilyl ether (15g, 77%) The organic phase was filtered and concentrated under re-
as a colorless oil. 4-Chloro-6-bromo-2-(trimethylsilyl)phenyl duced pressure. The residue was purified by flash column
tert-butyldimethylsilyl ether (15g, 38mmol) was dissolved in chromatography on silica gel (hexane) to provide the titled
anhydrous THF (0.19L, 0.2M) and the mixture was cooled compound 3j′ (6.5g, 92%) as a brown solid. Rf: 0.8 (hexane).
to −78°C. n-BuLi (2.5M hexane solution, 18mL, 45mmol) mp: 39–41°C. 1H-NMR (400MHz, CDCl3) δ: 0.40 (9H, s),
was added dropwise at −78°C and the reaction was allowed 1.32 (9H, s), 7.48 (1H, d, J=2.5Hz), 7.64 (1H, d, J=2.5Hz).

1162
Chem. Pharm. Bull.
Vol. 66, No. 12 (2018)
13C-NMR (100MHz, CDCl3) δ: 0.1, 31.1, 34.8, 115.9, 118.6 tography (hexane) to provide the titled compound 6a as a
(1C, q, J=160Hz), 132.6, 133.0, 136.5, 146.5, 152.3. 19F-NMR white solid (16g, 99%). mp: 64–65°C. 1H-NMR (300MHz,
(470MHz, CDCl3) δ:−71.63 (s). IR (neat): 1398, 1572cm−1. CDCl3) δ: 0.30 (9H, s), 0.32 (9H, s), 1.28 (9H, s), 1.40 (9H,
HR-MS (APCI): m/z Calcd for C14H2179BrF3O3SSi [M+H]+: s), 7.24 (1H, d, J=2.5Hz), 7.37 (1H, d, J=2.5Hz). 13C-NMR
433.0111. Found: 433.0107.
(75MHz, CDCl3) δ: 1.3, 2.9, 31.3, 31.6, 34.2, 35.1, 126.5, 130.1,
2,4-Diiodo-6-(trimethylsilyl)phenyl Triflate (3k′) (Table 130.5, 138.6, 142.5, 156.5. IR (neat): 1413, 1254, 1223cm−1.
3, Entry 4) An oven dried flask was charged with HR-MS (MALDI) Calcd for C20H38OSi2 [M]+: 350.2456.
2,4,6-triiodophenol109) (5.0g, 11mmol) and capped with an Found 350.2440.
inlet adapter with a 3-way stopcock and then evacuated and
back-filled with argon. Anhydrous THF (53mL, 0.20M), Et3N
Acknowledgments This work was financially supported
(2.2mL, 16mmol) and Me3SiCl (1.0mL, 16mmol) were added by the JSPS KAKENHI (Grant numbers 16K08164, 15J06024,
via syringes and the reaction mixture was stirred for 1h at and 16H01151/18H04411 (Middle Molecular Strategy))
room temperature. The reaction mixture was concentrated and Japan Agency for Medical Research and Development
under reduced pressure. Hexane was added to the residue and (AMED) (17am0101085j0001), and the Research Foundation
the mixture was filtered through a silica gel pad and washed for Pharmaceutical Sciences. We would also like to thank Mr.
with hexane. The solution was evaporated to give 2,4,6-tri- Matsubara of Techno Applications for kindly providing us
iodophenyl trimethylsilyl ether. Without purification of the with the flow reactor parts.
obtained material, anhydrous Et2O (0.11L, 0.10M) was added
to the flask and the mixture was cooled to −78°C. n-BuLi
Conflict of Interest The authors declare no conflict of
(2.6 M hexane solution, 4.5mL, 12mmol) was added drop- interest.
wise at 0°C and stirred for 1h. To the reaction mixture was
added Tf2O (2.7mL, 16mmol) and stirred for additional 1.5h
Supplementary Materials The online version of this ar-
at room temperature. Saturated aqueous solution of NaHCO3 ticle contains supplementary materials.
was added to the reaction mixture for quenching and it was
warmed to room temperature. The mixture was extracted with
hexane (this process was repeated three times) and combined
organic phase was dried over anhydrous MgSO4. The organic
phase was filtered and concentrated under reduced pressure.
The residue was purified by flash column chromatography
on silica gel (hexane) to provide the titled compound 3k′
(5.5g, 94%) as a white solid. Rf: 0.8 (hexane). mp: 46–48°C.
1H-NMR (300MHz, CDCl3) δ: 0.37 (9H, s), 7.75 (1H, d,
J=2.0Hz), 8.21 (1H, d, J=2.5Hz). 13C-NMR (100MHz,
CDCl3) δ: 0.15, 91.6, 95.0, 118.5 (CF3, q, J=320.5Hz), 140.2,
145.5, 149.9, 151.2. 19F-NMR (470MHz, CDCl3) δ: −71.65 (s).
IR (neat): 1215, 1408cm−1. HR-MS (FAB, NBA): m/z Calcd for
C10H11F3I2O3SSiNa [M+Na]+: 572.8132. Found: 572.8131.
Preparation of 2-(Trimethylsilyl)-4,6-di-tert-butylphenyl
Trimethylsilyl Ether (6a) (Chart 2) An oven-dried
round bottom flask was charged with 2-bromo-4,6-di-tert-
butylphenol68) (13g, 46mmol) and a stirrer bar. The flask
was equipped with a three-way stopcock and evacuated and
back-filled with nitrogen gas (three times). THF (50mL),
Me3SiCl (8.8mL, 69mmol), and Et3N (9.6mL, 69mmol) were
sequentially added into the flask via a syringe. The mixture
was stirred at room temperature for 1h. The reaction mix-
ture was evaporated under reduced pressure, and the residue
was filtered through a silica gel pad. The silica gel pad was
washed with hexane and the combined solution was concen-
trated under reduced pressure to provide 2-bromo-4,6-di-
tert-butylphenyl trimethylsilyl ether, which was used in the
next reaction without further purification. 0.12L of anhydrous
THF was added to the flask containing 2-bromo-4,6-di-tert-
butylphenyl trimethylsilyl ether (0.40M) and the solution was
cooled to −78°C under nitrogen. n-BuLi (2.6M hexane solu-
tion, 21mL, 55mmol) was added dropwise into the mixture at
−78°C, after which it was stirred at room temperature for 2h.
Me3SiCl (8.8mL, 69mmol) was added to the reaction mixture
at the same temperature, with stirring at this temperature for
15min. The mixture was concentrated under reduced pressure.
The residue was purified by silica-gel flash column chroma-
References and Notes
1) “Bioorganic and Medicinal Chemistry of Fluorine,” ed. by Bégué
J.-P., Bonnet-Delpon D., John Wiley & Sons, Inc., Hoboken, NJ,
2008.
2) “Fluorine in Medicinal Chemistry and Chemical Biology,” ed. by
Ojima I., Wiley-Blackwell, Chichester, 2009.
3) Purser S., Moore P. R., Swallow S., Gouverneur V., Chem. Soc.
Rev., 37, 320–330 (2008).
4) Filler R., Saha R., Future Med. Chem., 1, 777–791 (2009).
5) O’Hagan D., J. Fluor. Chem., 131, 1071–1081 (2010).
6) Schäfer A., Wellner A., Strauss M., Schäfer A., Wolber G., Gust
R., J. Med. Chem., 55, 9607–9618 (2012).
7) Wang J., Sánchez-Roselló M., Aceña J. L., del Pozo C., Sorochin-
sky A. E., Fustero S., Soloshonok V. A., Liu H., Chem. Rev., 114,
2432–2506 (2014).
8) Fujiwara T., O’Hagan D., J. Fluor. Chem., 167, 16–29 (2014).
9) McGrath N. A., Brichacek M., Njardarson J. T. J., Chem. Educa-
tor, 87, 1348–1349 (2010).
10) Grushin V. V., Marshall W. J., Organometallics, 27, 4825–4828
(2008).
11) Grushin V. V., Acc. Chem. Res., 43, 160–171 (2010).
12) Furuya T., Klein J. E. M. N., Ritter T., Synthesis, 2010, 1804–1821
(2010).
13) Liang T., Neumann C. N., Ritter T., Angew. Chem. Int. Ed., 52,
8214–8264 (2013).
14) Campbell M. G., Ritter T., Chem. Rev., 115, 612–633 (2015).
15) Champagne P. A., Desroches J., Hamel J.-D., Vandamme M., Pa-
quin J.-F., Chem. Rev., 115, 9073–9174 (2015).
16) Watson D. A., Su M., Teverovskiy G., Zhang Y., García-Fortanet
J., Kinzel T., Buchwald S. L., Science, 325, 1661–1664 (2009).
17) Tang P., Wang W., Ritter T., J. Am. Chem. Soc., 133, 11482–11484
(2011).
18) Nemoto H., Nishiyama T., Akai S., Org. Lett., 13, 2714–2717
(2011).
19) Nemoto H., Takubo K., Shimizu K., Akai S., Synlett, 23, 1978–
1984 (2012).
20) Fier P. S., Hartwig J. F., Science, 342, 956–960 (2013).
21) Wang K.-P., Yun S. Y., Mamidipalli P., Lee D., Chem. Sci., 4,
3205–3211 (2013).
22) Yoshida H., Yoshida R., Takaki K., Angew. Chem. Int. Ed., 52,

Vol. 66, No. 12 (2018)
Chem. Pharm. Bull.
1163
59) Atobe M., Tateno H., Matsumura Y., Chem. Rev., 118, 4541–4572
(2018).
8629–8632 (2013).
23) Milner P. J., Kinzel T., Zhang Y., Buchwald S. L., J. Am. Chem.
Soc., 136, 15757–15766 (2014).
24) Zeng Y., Li G., Hu J., Angew. Chem. Int. Ed., 54, 10773–10777
(2015).
25) Zeng Y., Hu J., Synthesis, 48, 2137–2150 (2016).
26) Takubo K., Furutsu K., Ide T., Nemoto H., Ueda Y., Tsujikawa
K., Ikawa T., Yoshimitsu T., Akai S., Eur. J. Org. Chem., 2016,
1562–1576 (2016).
60) Sandford G., “12–Continuous Flow Selective Direct Fluorination
Using Fluorine Gas in Modern Synthesis Processes and Reactivity
of Fluorinated Compounds,” 1st Edition, Progress in Fluorine Sci-
ence, ed. by Groult H., Leroux F., Tressaud A, Elsevier, 2017, pp.
339–348.
61) Amii H., Nagaki A., Yoshida J., Beilstein J. Org. Chem., 9, 2793–
2802 (2013).
62) Noël T., Maimone T. J., Buchwald S. L., Angew. Chem. Int. Ed.,
50, 8900–8903 (2011).
27) Schimler S. D., Cismesia M. A., Hanley P. S., Froese R. D. J.,
Jansma M. J., Bland D. C., Sanford M. S., J. Am. Chem. Soc., 139,
1452–1455 (2017).
28) Ikawa T., Nishiyama T., Nosaki T., Takagi A., Akai S., Org. Lett.,
13, 1730–1733 (2011).
63) To the best of our knowledge, only one other paper reported ben-
zyne generation under flow conditions, in which the generated
benzyne was immediately used for the coupling with functional-
ized aryl lithiums. Nagaki A., Ichinari D., Yoshida J., J. Am.
Chem. Soc., 136, 12245–12248 (2014).
29) Wenk H. H., Winkler M., Sander W., Angew. Chem. Int. Ed., 42,
502–528 (2003).
64) While the channel junctions of α-04 (channel width 400µm) and
α-06 (channel width 600µm) are simple T-shape, the channel
shape of Comet-X01 is much more complicated to realize efficient
mixing.
30) Peña D., Pérez D., Guitián E., Angew. Chem. Int. Ed., 45, 3579–
3581 (2006).
31) Yoshida H., Ohshita J., Kunai A., Bull. Chem. Soc. Jpn., 83, 199–
219 (2010).
65) Mason B. P., Price K. E., Steinbacher J. L., Bogdan A. R.,
McQuade D. T., Chem. Rev., 107, 2300–2318 (2007).
66) Tanaka K., J. Synth. Org. Chem. Jpn., 68, 124–135 (2010).
67) The reverse slow addition of a solution of 3a (1.0mol/L) in THF
32) Kitamura K., Aust. J. Chem., 63, 987–1001 (2010).
33) Tadross P. M., Stoltz B. M., Chem. Rev., 112, 3550–3577 (2012).
34) García-López J.-A., Greaney M. F., Chem. Soc. Rev., 45, 6766–
6798 (2016).
to a solution of Bu
35) Ikawa T., Masuda S., Nishiyama T., Takagi A., Akai S., Aust. J.
Chem., 67, 475–480 (2014).
36) Yoshida J., Chem. Commun., 2005, 4509–4516 (2005).
37) Yoshida J., Nagaki A., Yamada T., Chem. Eur. J., 14, 7450–7459
(2008).
4NF(t-BuOH)4 in THF (2.2mol/L) for 1h gave a
similar result (56% yield of 1a and 16% of 5a in). The slow addi-
tion of a solution of Bu4NF(t-BuOH)4 in THF (2.2mol/L) to a solu-
tion of 3a in THF (1.0mol/L) for 1h gave lower yield of 1a (33%)
and similar yield of 5a (15%).
38) Nieuwland P. J., Koch K., van Harskamp N., Wehrens R., van Hest
J. C. M., Rutjes F. P. J. T., Chem. Asian J., 5, 799–805 (2010).
39) Yoshida J., J. Chem. Rec., 10, 332–341 (2010).
40) Yoshida J., Takahashi Y., Nagaki A., Chem. Commun., 49, 9896–
9904 (2013).
41) Books on flow microreactor synthesis: Ehrfeld W., Hessel V., Löwe
H., “Microreactors,” Wiley-VCH, Weinheim, 2000.
42) Hessel V., Hardt S., Löwe H., “Chemical Micro Process Engineer-
ing,” Wiely-VCH Verlag, Weinheim, 2004.
68) Akai S., Ikawa T., Takayanagi S.-i., Morikawa Y., Mohri S.,
Tsubakiyama M., Egi M., Wada Y., Kita Y., Angew. Chem. Int.
Ed., 47, 7673–7676 (2008).
69) Ikawa T., Takagi A., Kurita Y., Saito K., Azechi K., Egi M., Kaki-
guchi K., Kita Y., Akai S., Angew. Chem. Int. Ed., 49, 5563–5566
(2010).
70) Ikawa T., Tokiwa H., Akai S., J. Synth. Org. Chem. Jpn., 70,
1123–1133 (2012).
71) Takagi A., Ikawa T., Saito K., Masuda S., Ito T., Akai S., Org.
Biomol. Chem., 11, 8145–8150 (2013).
72) Ikawa T., Takagi A., Goto M., Aoyama Y., Ishikawa Y., Itoh Y.,
Fujii S., Tokiwa H., Akai S., J. Org. Chem., 78, 2965–2983 (2013).
73) Takagi A., Ikawa T., Kurita Y., Saito K., Azechi K., Egi M., Itoh
Y., Tokiwa H., Kita Y., Akai S., Tetrahedron, 69, 4338–4352
(2013).
43) Yoshida J., “Flash Chemistry. Fast Organic Synthesis in Microsys-
tems,” Wiley-Blackwell, 2008.
44) “Micro Process Engineering,” ed. by Hessel V., Renken A.,
Schouten J. C., Yoshida J., Wiley-VCH Verlag, Weinheim, 2009.
45) “Microreactors in Organic Chemistry and Catalysis,” 2nd ed., ed.
by Wirth T., Wiley-VCH Verlag, Weinheim, 2013.
46) Fukuyama T., Rahman M. T., Sato M., Ryu I., Synlett, 2008,
151–163 (2008).
74) Ikawa T., Urata H., Fukumoto Y., Sumii Y., Nishiyama T., Akai S.,
Chem. Eur. J., 20, 16228–16232 (2014).
47) Hartman R. L., Jensen K. F., Lab. Chip, 9, 2495–2507 (2009).
48) McMullen J. P., Jensen K. F., Annu. Rev. Anal. Chem., 3, 19–42
(2010).
49) Yoshida J., Kim H., Nagaki A., ChemSusChem, 4, 331–340 (2011).
50) Wiles C., Watts P., Green Chem., 14, 38–54 (2012).
51) Kirschning A., Kupracz L., Hartwig J., Chem. Lett., 41, 562–570
(2012).
52) McQuade D. T., Seeberger P. H., J. Org. Chem., 78, 6384–6389
(2013).
53) Elvira K. S., Solvas X. C., Wootton R. C. R., deMello A. J., Nat.
Chem., 5, 905–915 (2013).
54) Pastre J. C., Browne D. L., Ley S. V., Chem. Soc. Rev., 42, 8849–
8869 (2013).
55) Baxendale I. R., J. Chem. Technol. Biotechnol., 88, 519–552
(2013).
56) Yoshida J., Basics of Flow Microreactor Synthesis, Springer, 2015.
57) Movsisyan M., Delbeke E. I. P., Berton J. K. E. T., Battilocchio C.,
Ley S. V., Stevens C. V., Chem. Soc. Rev., 45, 4892–4928 (2016).
58) Plutschack M. B., Pieber B., Gilmore K., Seeberger P. H., Chem.
Rev., 117, 11796–11893 (2017).
75) Ikawa T., Kaneko H., Masuda S., Ishitsubo E., Tokiwa H., Akai S.,
Org. Biomol. Chem., 13, 520–526 (2015).
76) Ikawa T., Masuda S., Takagi A., Akai S., Chem. Sci., 7, 5206–5211
(2016).
77) Ikawa T., Sumii Y., Masuda S., Wang D., Emi Y., Takagi A., Akai
S., Synlett, 29, 530–536 (2018).
78) Kaneko H., Ikawa T., Yamamoto Y., Arulmozhiraja S., Tokiwa H.,
Akai S., Synlett, 29, 943–948 (2018).
79) Biehl E. R., Nieh E., Hsu K. C., J. Org. Chem., 34, 3595–3599
(1969).
80) Moreau-Hochu M. F., Caubere P., Tetrahedron, 33, 955–959
(1977).
81) Ghosh T., Hart H., J. Org. Chem., 53, 3555–3558 (1988).
82) Hart H., Ghosh T., Tetrahedron Lett., 29, 881–884 (1988).
83) Wickham P. P., Reuter K. H., Senanayake D., Guo H., Zalesky M.,
Scott W. J., Tetrahedron Lett., 34, 7521–7524 (1993).
84) Gokhale A., Schiess P., Helv. Chim. Acta, 81, 251–267 (1998).
85) Kirkham J. D., Delaney P. M., Ellames G. J., Row E. C., Harrity J.
P. A., Chem. Commun., 46, 5154–5156 (2010).
86) Bronner S. M., Goetz A. E., Garg N. K., J. Am. Chem. Soc., 133,

1164
Chem. Pharm. Bull.
mun., 49, 7602–7604 (2013).
Vol. 66, No. 12 (2018)
3832–3835 (2011).
87) Goetz A. E., Garg N. K., Nat. Chem., 5, 54–60 (2013).
88) Hendrick C. E., McDonald S. L., Wang Q., Org. Lett., 15, 3444–
3447 (2013).
89) Medina J. M., Mackey J. L., Garg N. K., Houk K. N., J. Am.
Chem. Soc., 136, 15798–15805 (2014).
99) Shi J., Qiu D., Wang J., Xu H., Li Y., J. Am. Chem. Soc., 137,
5670–5673 (2015).
100) Qiu D., He J., Yue X., Shi J., Li Y., Org. Lett., 18, 3130–3133
(2016).
101) Li L., Qiu D., Shi J., Li Y., Org. Lett., 18, 3726–3729 (2016).
102) Xu H., He J., Shi J., Tan L., Qiu D., Luo X., Li Y., J. Am. Chem.
Soc., 140, 3555–3559 (2018).
90) Yoshida S., Nagai A., Uchida K., Hosoya T., Chem. Lett., 46,
733–736 (2017).
91) Kou K. G. M., Pflueger J. J., Kiho T., Morrill L. C., Fisher E. L.,
Clagg K., Lebold T. P., Kisunzu J. K., Sarpong R., J. Am. Chem.
Soc., 140, 8105–8109 (2018).
92) Lv C., Wan C., Liu S., Lan Y., Li Y., Org. Lett., 20, 1919–1923
(2018).
93) Charmant J. P. H., Dyke A. M., Lloyd-Jones G. C., Chem. Com-
mun., 2003, 380–381 (2003).
94) Sun L., Nie J., Zheng Y., Ma J.-A., J. Fluor. Chem., 174, 88–94
(2015).
103) Kim D. W., Jeong H.-J., Lim S. T., Sohn M.-H., Angew. Chem. Int.
Ed., 47, 8404–8406 (2008).
104) Schlosser M., Guio L., Leroux F., J. Am. Chem. Soc., 123, 3822–
3823 (2001).
105) Stadlbauer S., Ohmori K., Hattori F., Suzuki K., Chem. Commun.,
48, 8425–8427 (2012).
106) Tredwell M., Preshlock S. M., Taylor N. J., Gruber S., Huiban M.,
Passchier J., Mercier J., Génicot C., Gouverneur V., Angew. Chem.
Int. Ed., 53, 7751–7755 (2014).
95) Ikawa T., Masuda S., Nakajima H., Akai S., J. Org. Chem., 82,
4242–4253 (2017).
107) Suresh P., Annalakshmi S., Pitchumani K., Tetrahedron, 63, 4959–
4967 (2007).
96) For a recent example of the halogen dance reaction, see: Hayashi
Y., Okano K., Mori A., Org. Lett., 20, 958–961 (2018).
97) Dyke A. M., Gill D. M., Harvey J. N., Hester A. J., Lloyd-Jones
G. C., Muñoz M. P., Shepperson I. R., Angew. Chem. Int. Ed., 47,
5067–5070 (2008).
108) Podgoršek A., Stavber S., Zupan M., Iskra J., Tetrahedron, 65,
4429–4439 (2009).
109) Gallo R. D. C., Ferreira I. M., Casagrande G. A., Pizzuti L.,
Oliveira-Silva D., Raminelli C., Tetrahedron Lett., 53, 5372–5375
(2012).
98) Hall C., Henderson J. L., Ernouf G., Greaney M. F., Chem. Com-