Oxidation of fluoroalkyl alcohols using sodium hypochlorite pentahydrate [1]

Article abstract of DOI:10.1016/j.jfluchem.2020.109719

Fluoroalkyl alcohols are effectivity oxidized to the corresponding fluoroalkyl carbonyl compounds by reaction with sodium hypochlorite pentahydrate in acetonitrile in the presence of acid and nitroxyl radical catalysts. Although the reaction proceeded slower under a nitroxyl radical catalyst- free condition, the desired carbonyl compounds were obtained in high yields. For the reaction with fluoroalkyl allylic alcohols, the corresponding α,β-epoxyketone hydrates were obtained in high yields.

Full text of DOI:10.1016/j.jfluchem.2020.109719

Journal of Fluorine Chemistry 243 (2021) 109719
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Oxidation of fluoroalkyl alcohols using sodium hypochlorite
pentahydrate [1]
,
Masayuki Kiriharaa *, Katsuya Suzukia, Kana Nakakuraa, Katsuya Saitoa, Riho Nakamuraa,
Kazuki Tujimotoa, Yugo Sakamotoa, You Kikkawaa, Hideo Shimazub, Yoshikazu Kimurac
a Department of Materials and Life Science, Shizuoka Institute of Science and Technology, 2200-2 Toyosawa, Fukuroi, Shizuoka 437-8555, Japan
b R&D Department of Chemicals, Nippon Light Metal Company, Ltd., Kambara, Shimizu-ku, Shizuoka 421-3203, Japan
c Research and Development Department, Iharanikkei Chemical Industry Co., Ltd., Kambara, Shimizu-ku, Shizuoka 421-3203, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Fluoroalkyl alcohols are effectivity oxidized to the corresponding fluoroalkyl carbonyl compounds by reaction
with sodium hypochlorite pentahydrate in acetonitrile in the presence of acid and nitroxyl radical catalysts.
Although the reaction proceeded slower under a nitroxyl radical catalyst- free condition, the desired carbonyl
compounds were obtained in high yields. For the reaction with fluoroalkyl allylic alcohols, the corresponding
Oxidation
Fluoroalkyl alcohol
Sodium hypochlorite pentahydrate
TEMPO
α
,β-epoxyketone hydrates were obtained in high yields.
Fluoroalkyl allylic alcohol
Fluoroalkyl α,β-epoxyketone hydrates
1. Introduction
tetramethylpiperidin-1-oxyl) (Scheme 2) [5c,f]. This reaction efficiently
provides the corresponding carbonyl compounds in high yields, and the
experimental procedure is simple and easy. Moreover, this method is
environmentally benign because the postoxidation waste is harmless
“table salt” (NaCl) [5,6].
Fluoroalkyl carbonyl compounds, represented by the trifluoromethyl
ketones, are very important and attractive building blocks for the syn-
theses of fluorine-containing organic molecules in several fields of sci-
ence & industry (medicinal chemistry, pharmaceutical science, material
science, agricultural chemistry, etc.) [2]. Several methods of generating
fluoroalkyl carbonyl compounds have been developed. Among them, the
oxidation of fluoroalkyl alcohols is one of the most useful methods [3].
The secondary fluoroalkyl alcohols are easily prepared by the
Ruppert-Prakash reaction {the reaction of aldehydes with trimethylsi-
lylated fluoroalkanes [(trifluoromethyl)trimethylsilane (TMSCF3),
(difluoromethyl)trimethylsilane (TMSCHF2), etc.]} [4], however, the
fluoroalkyl alcohols are generally hard to be oxidized due to the strong
electron withdrawing fluoroalkyl groups. Therefore, very strong oxi-
dants, such as the Dess-Martin reagent [3a,b] or chromium(VI) [3c], are
required to obtain the fluoroalkyl ketones from the secondary fluo-
roalkyl alcohols (Scheme 1). Although these oxidants afford the desired
products, the Dess-Martin reagent is expensive and explosive, and
chromium(VI) is highly toxic. To overcome these drawbacks of the
conventional methods of oxidation of fluoroalkyl alcohols, several new
reactions have recently been developed [3d-i].
During the course of our study of the oxidation by NaOCl⋅5H2O, we
found that secondary alcohols bearing fluoroalkyl groups were effi-
ciently oxidized to the corresponding ketones by NaOCl⋅5H2O catalyzed
by TEMPO [1]. Recently, Leadbeatear and Eddy et al. reported that 2,2,
6,6-tetramethylpiperidine-4-acetamido-hydroxyammonium
tetra-
fluoroborate (4-AcO-TEMPOH⋅BF4) is a good catalyst for the oxidation
of alcohols by NaOCl⋅5H2O [6d]. Although secondary fluo-
roalkylalcohols were also effectively oxidized to the corresponding ke-
tones by their method [6d], 4-AcO-TEMPOH⋅BF4 is not commercially
available and has to be prepared from the relatively expensive 4-AcO--
TEMPO. We further examined the oxidation of fluoroalkylalcohols
with NaOCl⋅5H2O, and found that commercially available and inex-
pensive TEMPO is a good catalyst for oxidation of several fluo-
roalkylalcohols including some primary alcohols. For the secondary
fluoroalkylalcohols, NaOCl⋅5H2O provided the desired ketones under a
catalyst-free condition. We also found that fluoroalkyl allylic alcohols
afforded the corresponding
α,β-epoxyketone hydrates. This article will
We have developed the oxidation of alcohols with sodium hypo-
chlorite pentahydrate (NaOCl⋅5H2O) catalyzed by TEMPO (2,2,6,6-
describe the full details of our results (Scheme 3).
* Corresponding author.
E-mail address: kirihara.masayuki@sist.ac.jp (M. Kirihara).
https://doi.org/10.1016/j.jfluchem.2020.109719
Received 10 October 2020; Received in revised form 24 December 2020; Accepted 25 December 2020
Available online 6 January 2021
0022-1139/© 2021 Elsevier B.V. All rights reserved.

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
2. Results and discussion
2.1. Examinations of the reaction conditions
The reaction conditions were evaluated using 1,1,1-trifluoro-2-(4-
methoxyphenyl)ethanol (1a) as a substrate (Table 1). The reaction of
1a with NaOCl⋅5H2O produced the desired ketone (2a) in 85 % yield
under the standard conditions for the oxidation of non-fluorinated al-
cohols (1 mol% TEMPO, 2.4 eq NaOCl⋅5H2O, in dichloromethane) (run
1). The investigation of the solvent effect suggested that several solvents
can be used for this oxidation except for acetone (runs 2–8). Acetone
rapidly reacted with NaOCl⋅5H2O, and all of 1a remained unreacted
(run 6). The best solvent for the oxidation of 1a is acetonitrile (run 8).
The reaction of 1a at room temperature provided some by-products
resulting in a decrease in yield (run 9). The use of 1.2 eq. NaOCl⋅5H2O
afforded 2a in low yield (run 10).
Scheme 1. Oxidation of fluoroalkyl alcohols.
For the reaction with 0.1 mol% TEMPO, 2a was obtained in a good
yield (84 %), though the reaction proceeded slowly (run 11).
Next, the types of nitroxyl radical catalysts were evaluated in
acetonitrile (Table 2). All the catalysts were effective for the oxidation of
1a with NaOCl⋅5H2O (runs 1–4). Interestingly, although the reaction
rate was slower, 2a was obtained in a high yield from the reaction with
NaOCl⋅5H2O under a nitroxyl radical-free condition (run 5) [7,8].
For the TEMPO-catalyzed oxidation of ordinary alcohols, [4e, f]
NaOCl⋅5H2O was a far more effective oxidant than the conventional
aqueous 13 % NaOCl. A similar result is anticipated for the oxidation of
fluoroalkyl secondary alcohols. Thus, the reaction of 1a with the con-
ventional aqueous 13 % NaOCl proceeded very slowly and the desired
2a was obtained in a poor yield (Table 3 run 2). In our previous studies
of the oxidation using NaOCl as an oxidant, we admitted that the
oxidizing ability of NaOCl largely depends on the pH; the main differ-
ence between NaOCl⋅5H2O and the conventional aqueous 13 % NaOCl is
their pH (NaOCl⋅5H2O: 10~11, conventional aqueous 13 % NaOCl:
>13). Actually, aqueous NaOCl (pH 10.7) prepared from the conven-
tional aqueous 13 % NaOCl and hydrochloric acid was effective for the
oxidation of 1a (run 5) and exhibited a reactivity similar to the aqueous
NaOCl (pH 10.7) prepared from NaOCl⋅5H2O and water (run 3). On the
other hand, aqueous NaOCl (pH 13) prepared from NaOCl⋅5H2O, water
and sodium hydroxide reacted with 1a very slowly (run 5) as in the case
of the conventional aqueous 13 % NaOCl (run 2).
Scheme 2. Oxidation of alcohols with NaOCl⋅5H2O catalyzed by TEMPO. .
Scheme 3. Oxidation of fluoroalkylalcohols using NaOCl⋅5H2O.
Table 1
Examination of the reaction conditions.
Run
Solvent
TEMPO
NaOCl・5H2O
Time
Yield (%)a
1
CH2Cl2
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
1 mol%
0.1 mol%
2.4 eq
2.4 eq
2.4 eq
2.4 eq
2.4 eq
2.4 eq
2.4 eq
2.4 eq
2.4 eq
1.2 eq
2.4 eq
1 h 15 min
30 min
30 min
1 h 15 min
2 h
85
2
DCEb
78
3
Toluene
BTFc
80
4
80
5
Hexane
Acetone
EtOAc
CH3CN
CH3CN
CH3CN
CH3CN
80
6
2 h
0
7
1 h 45 min
15 min
15 min
1 h
87
8
99 (81)d
9
85e
79
10
11
1 h 15 min
84
a
Yields based on 19F NMR spectra of the crude reaction mixture using 4-chlorobenzotrifluoride as the internal standard.
b
c
DCE: 1,2-Dichloroethane.
BTF: Benzotrifluoride [(trifluoromethyl)benzene].
d
e
The number in parentheses is referred to as the isolated yield.
Reaction at room temperature.
2

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 2
Examination of the N-oxyl radical.
Run
N-oxyl radical
Time
Yield (%)a
1
2
3
4
5
TEMPO
15 min
30 min
15 min
2 h
99 (81)b
4-Methoxy-TEMPO
AZADO
81
86
keto-ABNO
–
85
5 h
89 (73)b
a
b
Yields based on 19F NMR spectra of the crude reaction mixture using 4-chlorobenzotrifluoride as the internal standard.
The numbers in parentheses refer to the isolated yield.
Table 3
Comparison of NaOCl ⋅ 5H2O vs. conventional 13 % aq. NaOCl with varying pH.
Run
NaOCl
pH
Additive
Time
Yield(%)a
1
2
3
4
5
NaOCl・5H2O
–
–
20 min
24 h
99 (81b)
13 % NaOCl(aq) [conventional aq. solution]
13 % NaOCl(aq) [prepared from NaOCl⋅5H2O]
13 % NaOCl(aq) [prepared from NaOCl⋅5H2O]
13 % NaOCl(aq) [conventional aq. solution]
13
–
18b
10.7
13
–
45 min
24 h
quant
37
NaOH
HCl
10.7
45 min
83b
a
b
Yields based on 19F NMR ratio.
Isolated yield.
KHSO4 might be used as an acid to neutralize NaOCl into HOCl,
substituent (1) were treated with NaOCl⋅5H2O (2.4 eq.) and potassium
hydrogen sulfate (5 mol%) in acetonitrile in the presence of the TEMPO
catalyst at 0 ◦C (Table 5) [9]. In all cases, the fluorinated alcohols 1 were
rapidly oxidized to produce the desired 2 or the corresponding hydrate
(gem-diol) 2′ in the presence of TEMPO in high yields. The fluorinated
alcohols having electron deficient aromatic rings (1c, 1f) provided the
hydrate (2′c, 2′f) as the products (entries 3, 6).
which is a stronger oxidant. The treatment of 1a without KHSO4 or with
other acids (CF3CO2H, CH3CO2H) gave 2a in slightly lower yields
(Table 4).
2.2. Reaction of fluoroalkyl secondary alcohols bearing an aromatic
substituent catalyzed by TEMPO
Several fluoroalkyl secondary alcohols bearing an aromatic
Table 4
Effects of the acid catalyst.
Run
Acid catalyst
Time
Yield(%)a
1
2
3
4
KHSO4
none
20 min
30 min
15 min
20 min
99
90
80
83
CF3COOH
AcOH
a
Yields based on 19F NMR spectra of the crude reaction mixture using 4-chlorobenzotrifluoride as the internal standard.
3

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
2.5. Reaction of fluoroalkyl primary alcohols, trifluoroactaldehyde
ethylhemiacetal, and hexafluoroisopropanol in acetonitrile-d3
Table 5
Oxidation of fluoroalkyl secondary alcohols bearing an aromatic substituent.
The oxidations of several other types of fluorinated alcohols were
next attempted. The reaction was performed in acetonitrile-d3 (CD3CN)
and analyzed by NMR (Table 8). The reaction of primary alcohols (1o,
1p) provided the corresponding carboxylic acids in good yields. The
ethyl hemiacetal of trifluoroacetaldehyde was oxidized to ethyl tri-
fluoroacetate (2q) in 77 %. In the case of hexafluoroisopropanol (1 r),
the desired hydrate of hexafluoroacetone (2r’) was obtained in 79 %
yield. These results suggested that NaOCl⋅5H2O is a very good oxidant
for the synthesis of perfluoroketones, perfluorocarboxylic acids, and
esters of the perfluorocarboxylic acids. Unfortunately, these alcohols
(1o-r) were completely inert to the reaction with NaOCl⋅5H2O under the
TEMPO-free conditions.
2.6. Reaction of fluoroalkyl alcohols bearing an allylic substituent:
Synthesis of 1,1,1-trifluoromethyl α,β-epoxyketone hydrates
As shown in the previous chapter (2–4), the 1,1,1-trifluoromethyl
,β-epoxyketone hydrate (3′n) was efficiently obtained from the reac-
α
tion of the fluorinated allylic alcohol (1n) with NaOCl⋅5H2O catalyzed
by TEMPO (Table 6 entry 5) [10]. Since 1,1,1-trifluoromethyl
α
,β-epoxyketone hydrates (3′) seem to be attractive synthetic building
blocks, the generality of this reaction was then evaluated.
Several 1,1,1-trifluoromethyl allylic alcohols (1n-x) were treated
with NaOCl⋅5H2O (2.4 eq) in the presence of the TEMPO-catalyst
(Table 9). The allylic alcohols having an electron-deficient aromatic
ring (1s-u) efficiently produced the desired α,β-epoxyketone hydrates
(3′s-u) in high yields (entries 2–4). In all cases, the corresponding
enones or enone hydrates were not obtained at all. Unfortunately, an
allylic alcohol having an electron-rich aromatic ring (1v) provided a
complex mixture (entry 5) affected by the electrophilic chlorination of
the aromatic ring.
aIsolated yield.
2.7. Consideration of the reaction mechanism of the reaction of
b19F NMR yield using 4-chlorobenzotrifluoride as the internal standard.
fluoroalkyl alcohols with NaOCl⋅5H2O
2.3. Reaction of fluoroalkyl secondary alcohols bearing an aliphatic
substituent
The reaction mechanism of this reaction is almost the same as that of
the oxidation of ordinary alcohols with NaOCl⋅5H2O catalyzed by
TEMPO and KHSO4 (Scheme 4). First, hypochloric acid (HOCl) is pro-
duced from the reaction of NaOCl⋅5H2O and KHSO4. HOCl oxidizes
TEMPO to produce the active species (A) and HCl. The produced HCl
then reacts with NaOCl to form HOCl and NaCl. Since the oxidation of
alcohols with NaOCl⋅5H2O/TEMPO/KHSO4 occurs under acidic to
neutral conditions, the reaction proceeds via the transition state B [11].
The lone pair of the nitrogen atom attacks the hydrogen atom of the
hydroxy group in the alcohol, therefore, more acidic fluorinated alco-
hols are good substrates for this reaction.
The reaction of fluoroalkyl alcohols bearing an aliphatic substituent
with NaOCl⋅5H2O was then evaluated (Table 6). Unfortunately, fluo-
roalkyl alcohols having an alkyl group (1 j, k) were almost inactive
under the standard reaction conditions (entries 1 and 2). An alcohol
bearing a cyclopropyl moiety (1 l) was efficiently oxidized to afford the
desired ketone (2 l) without decomposition of the cyclopropane-ring
(entry 3). A trifluoromethyl propargyl alcohol (1 m) also rapidly reac-
ted with NaOCl⋅5H2O to provide the corresponding ketone (2 m) in 42 %
yield, but together with unidentified byproducts (entry 4). In the case of
a trifluoromethyl allylic alcohol (1n), the epoxyketone hydrate (3n) was
obtained in 82 %, and formation of the expected enone (2n) or enone-
hydrate (2′n) was not observed (entry 5). The reaction mechanism
will be shown later (Scheme 6).
2.8. Consideration of the reaction mechanism of the fluoroalkyl allylic
alcohol (1n)
The mechanism of the formation of α,β-epoxyketone hydrate (3′n)
from 1n was investigated. The epoxide (3′n) might be produced from
the corresponding enone (2n) which was provided from the oxidation of
1n. We prepared 2n according to the literature [3j], and examined the
reaction of 2n with NaOCl⋅5H2O both in the presence and absence of
TEMPO. The corresponding 3′n was obtained in both cases (Scheme 5)
[12].
2.4. Reaction of fluoroalkyl secondary alcohols without TEMPO
As shown by run 5 in Table 2, 1a was oxidized to the ketone (2a) by
the reaction with NaOCl⋅5H2O without the TEMPO-catalyst [8].
Therefore, the oxidations of more secondary fluorinated alcohols (1)
were examined with NaOCl⋅5H2O and 5 mol% of KHSO4 in CH3CN in the
absence of TEMPO (Table 7). Although the reaction rate was slower, the
same products (2 or 3) were obtained in moderate to high yields.
The plausible mechanism is shown in Scheme 6. The fluorinated
allylic alcohol (1n) is first oxidized to the corresponding enone (2n).
Further oxidation of 2n might produce the epoxyketone, which is iso-
lated as the hydrated form of 3n (Scheme 6). Although the reactive
species is HOCl prepared from NaOCl⋅5H2O with acid, the lone pair of
the nitrogen atom of active formed TEMPO (Scheme 4, A) acts as base to
4

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 6
Oxidation of fluoroalkyl secondary alcohols bearing an aliphatic substituent with NaOCl⋅5H2O catalyzed by TEMPO.
aIsolated yield.
b19F NMR yield using 4-chlorobenzotrifluoride as the internal standard.
react with HOCl providing ClOꢀ . Therefore, the reaction in the presence
of TEMPO catalyst provided the better result in Scheme 4.
JEOL (JNM-EX400) spectrometer as solutions in CDCl3 using TMS, CFCl3
or the residual solvent peak as the internal standard, and the coupling
constants (J) are given in hertz (Hz). The IR spectra were recorded using
a Jasco IR-8300 FT-IR spectrophotometer. The mass spectra were
recorded by a Shimadzu GCMS-QP1100EX spectrometer (EI) or JEOL
JMS-T100LC spectrometer (ESI).
Interestingly, the reaction of a non-fluorinated analog (4) with
NaOCl⋅5H2O under the same reaction conditions for a longer reaction
time (4 h 32 min) only provided the simple enone (5) in a lower yield
with recovery of the starting material (4) (Table 10). The electron
withdrawing trifluoromethyl moiety is essential in order to form an
α
,β-epoxyketone.
4.1. Standard experimental procedure for the oxidation of fluorine-
containing alcohols using NaOCl⋅5H2O catalyzed by TEMPO
3. Conclusion
Potassium hydrogen sulfate (6.7 mg, 0.05 mmol) and TEMPO (1.7
mg, 0.01 mmol) were added to a stirred solution of a fluoroalkyl alcohol
(1) (1.0 mmol) in acetonitrile (5 mL) at room temperature, and the
resulting mixture was cooled to 0 ºC. Sodium hypochlorite pentahydrate
(395 mg, 2.4 mmol) was then added to the reaction mixture, stirred
under the same conditions, and monitored by TLC. After the starting
material (1) had been consumed, water (5 mL) was added, and the
resulting mixture was extracted with ethyl acetate (20 mL×3). The
combined organic extract was washed with brine (50 mL), dried over
sodium sulfate, and evaporated. The 19F-NMR yields were calculated by
using 4-chlorobenzotrifluoride as the internal standard. The crude
product was purified by silica-gel column chromatography using hex-
ane/ethyl acetate as the eluent to provide the fluoroalkyl ketone (2).
Several fluoroalkyl alcohols (1) are effectively oxidized by the re-
action with NaOCl⋅5H2O catalyzed by TEMPO to afford the corre-
sponding fluoroalkyl ketones and fluoroalkyl carboxylic acid in high
yields, except for fluoroalkyl alcohols bearing simple alkyl groups. In the
case of the secondary fluoroalkyl alcohols, the desired fluoroalkyl ke-
tones are also obtained in good yields. The α,β-epoxyketones were effi-
ciently obtained by the reaction of fluoroalkyl allylic alcohols with
NaOCl⋅5H2O catalyzed by TEMPO.
4. Experimental section
General Information: All reagents were purchased from Nacalai
Tesque, Wako Pure Chemicals Industries, Kanto Kagaku, Kishida Re-
agents Chemical Co., Tokyo Chemical Industry or Aldrich, and used
without any further purification. Most of the starting materials (1a~i,
1j~n, 1s~x) were prepared from the reaction of (trimethylsilyl)tri-
fluoromethane [or (trimethylsilyl)difluoromethane] with the corre-
sponding aldehydes according to the literature [13]. For the starting
materials in Table 8 (1o~r), commercially available reagents were used.
Melting points were measured by a Yanaco micro melting point appa-
ratus (MP-J3) and are uncorrected. The NMR spectra were recorded by a
4.2. Standard experimental procedure for the oxidation of fluorine-
containing alcohols using NaOCl⋅5H2O without the TEMPO catalyst
Potassium hydrogen sulfate (13.7 mg, 0.10 mmol) was added to a
stirred solution of a fluoroalkyl alcohol (1) (2.0 mmol) in acetonitrile
(5 mL) at room temperature, and the resulting mixture was cooled to
0 ºC. Sodium hypochlorite pentahydrate (791 mg, 4.8 mmol) was added
to the reaction mixture, stirred under the same conditions, and
5

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 7
Oxidation of fluoroalkyl secondary alcohols with NaOCl⋅5H2O without TEMPO-catalyst.
aIsolated yield.
monitored by TLC. After the starting material (1) had been consumed,
water (5 mL) was added, and the resulting mixture was extracted with
ethyl acetate (20 mL×3). The combined organic extract was washed
with brine (50 mL), dried over sodium sulfate, and evaporated. The
residue was purified by silica-gel column chromatography using hex-
ane/ethyl acetate as the eluent to produce the fluoroalkyl ketone (2).
4.5. 2,2,2-Trifluoro-1-(4-nitrophenyl)ethane-1,1-diol (2′c) [16]
Pale yellow crystals.
mp: 80–82 ºC (lit. 79–80 ºC [16])
1H NMR (CD3OD) δ: 7.90 (2H, dd, J = 22.0, 8.8 Hz), 8.23 (2H, m).
13C NMR(CD3OD) δ: 95.84 (q, J = 31.7 Hz), 122.20 (q, J = 287.6 Hz),
122.72, 129.02, 141.17, 148.27.
19F NMR(CD3OD) δ: -82.98 (s).
4.3. 2,2,2-Trifluoro-1-(4-methoxyphenyl)ethan-1-one (2a) [14]
4.6. 2,2,2-Trifluoro-1-(3-methoxyphenyl)ethan-1-one (2d)[14]
Pale yellow oil.
1H NMR (CDCl3) δ: 3.89 (3H, s), 7.00 (2H, apparent doublet, J = 8.8
Hz), 8.06 (2H, apparent doublet, J = 8.8 Hz).
13C NMR (CDCl3) δ: 55.6, 114.4, 115.4 (q, J = 291.3 Hz), 122.7,
132.7, 165.4, 178.7 (q, J = 34.4 Hz).
Pale yellow oil.
1H NMR (CDCl3) δ: 3.88 (3H, s), 7.26 (1H, m), 7.46 (1H, t, J = 7.8
Hz), 7.57 (1H, s), 7.67 (1H, d, J = 7.8 Hz).
13C NMR (CDCl3) δ: 55.52, 113.97, 116.64 (q, J = 291.4 Hz), 122.26,
122.72, 130.10, 131.07, 159.98, 180.54 (q, J = 34.5 Hz).
19F NMR (CDCl3) δ: -71.73 (s).
19F NMR (CDCl3) δ: -71.48 (s).
4.4. 2,2,2-Trifluoro-1-phenylethan-1-one (2b) [15]
4.7. 2,2,2-Trifluoro-1-(2-methoxyphenyl)ethan-1-one (2e) [14]
Pale yellow oil.
1H NMR (CDCl3) δ: 7.55 (2H, apparent triplet, J = 7.2 Hz), 7.72 (1H,
apparent triplet, J = 8.0 Hz), 8.09 (2H, apparent doublet, J = 8.8 Hz).
13C NMR (CDCl3) δ: 116.67 (q, J = 291.7 Hz), 129.06, 129.90,
130.06, 135.49, 181.05 (t, J = 35.2 Hz).
Pale yellow oil.
1H NMR (CDCl3) δ: 3.91 (3H, s), 7.04 (2H, m), 7.59 (2H, m), 7.68
(1H, d, J = 8.0 Hz).
13C NMR (CDCl3) δ: 56.24, 112.07, 116.23 (q, J = 291.3 Hz), 120.66,
121.67, 131.30, 135.29, 159.82, 183.12 (q, J = 36.2 Hz).
19F NMR (CDCl3) δ: -71.23 (s).
6

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 8
Oxidations of several other types of fluorinated alcohols.
a19F-NMR yield using 4-chlorobenzotrifluoride as the internal standard.
bContaining the corresponding sodium salt.
19F NMR (CDCl3) δ: -74.97 (s).
4.11. 2,2-Difluoro-1-(4-methoxyphenyl)ethan-1-one (2i) [19]
4.8. 2,2,2-Trifluoro-1-(pyridine-2-yl)ethane-1,1-diol (2′f) [17]
Pale yellow oil.
1H NMR (CDCl3) δ: 3.88 (3H, s), 6.24 (1H, t, J = 53.6 Hz), 6.98 (2H,
d, J = 8.7 Hz), 8.05 (2H, d, J = 8.7 Hz).
Pale yellow crystals
13C NMR (CDCl3) δ: 55.3, 111.0 (t, J = 253.0 Hz), 114.1, 124.3,
131.8, 164.8, 185.7 (t, J = 24.8 Hz).
mp: 63–65 ºC (lit. 64–66 ºC [17])
1H NMR (CDCl3) δ: 7.45 (1H, m), 6.xx (2H, s), 7.75 (1H, dd, J = 7.7,
0.78 Hz), 7.88 (1H, td, J = 7.7, 1.2 Hz), 8.6 (1H, m).
13C NMR (CDCl3) δ: 95.14 (q, J = 32.6 Hz), 123.96 (q, J = 286.6 Hz),
125.08, 128.72, 137.99, 147.48, 152.17.
19F NMR (CDCl3) δ: -121.98 (d, J = 57.6 Hz).
4.12. 1,1,1-Trifluoro-4-phenylbutan-2-one (2j) [20]
19F NMR (CDCl3) δ: -83.18 (s).
It could not be isolated due to small amount.
The 19F-NMR peak of this crude product was identical to that in the
literature. [20]
4.9. 2,2,2-Trifluoro-1-[2-(benzyloxy)oxy)phenyl]ethan-1-one (2g)
19F NMR (CDCl3) δ: -79.767 (s).
Pale yellow oil.
1H NMR (CDCl3) δ: 5.10 (2H, s), 6.96–7.02 (2H, m), 7.28–7.50 (6H,
m), 7.47 (1H, t, J = 7.2 Hz), 7.65 (1H, d, J = 8.0 Hz).
13C NMR (CDCl3) δ: 70.91, 113.51, 116.45 (q, J = 291.4 Hz), 121.03,
122.07, 127.45, 128.33, 128.77, 129.11, 131.49, 135.92, 135.95,
183.29 (q, J = 36.4 Hz).
4.13. 2,2,2-Trifluoro-1-(2-phenylcyclopropyl)ethan-1-one (2l) [3h]
Pale yellow oil.
1H NMR (CDCl3) δ: 1.72–1.75 (1H, m), 2.74–2.81 (1H, m), 2.49–2.53
(1H, m), 2.74–2.81 (1H, m), 7.13–7.15 (2H, m), 7.25–7.34 (3H, m).
13C NMR (CDCl3) δ: 20.75, 27.23, 32.72, 114.33 (q, J = 290.4 Hz),
126.44, 127.40, 128.73, 138.16, 189.45 (q, J = 36.4 Hz).
19F NMR (CDCl3) δ: -79.19 (s).
19F NMR (CDCl3) δ: -74.03 (s).
IR (neat) cmꢀ 1: 3070, 3038, 2933, 2879, 1712, 1597, 1490, 1453,
1276, 1175, 1011, 940, 752, 700, 667, 526.
MS (EI) (m/z): 280 (M+).
HRMS (ESI) calcd for C15H11F3O2Na [(M+Na) +], 303.0603, found :
303.0603.
4.14. 1,1,1-Trifluoro-4-phenylbut-3-yn-2-one (2m) [21]
The 1H-, 13C- and 19F-NMR peaks of this crude product were identical
to those in the literature [21]. The product decomposed during silica-gel
column chromatography.
4.10. 1-(9-Anthracenyl-)2,2,2-trifluoroethan-1-one (2h) [18]
Yellow crystals.
1H NMR (CDCl3) δ: 7.42ꢀ 1.66 (5H, m).
mp: 80–82 ºC (lit. 81–84 ºC [18])
13C NMR (CDCl3) δ: 83.36, 100.52, 113.41 (q, J = 288.4 Hz), 118.06,
128.92, 132.51, 133.93, 166.92 (q, J = 42.1 Hz).
19F NMR (CDCl3) δ: -78.29 (s).
1H NMR (CDCl3) δ: 7.45–7.58 (4H, m), 7.74 (2H, d, J = 8.4 Hz), 8.06
(2H, d, J = 8.4 Hz), 8.59 (1H, s).
13C NMR (CDCl3) δ: 116.00 (q, J = 293.4 Hz), 123.85, 125.86,
127.31, 127.85, 128.72, 128.94, 130.68, 130.92, 191.10 (q, J = 38.3
Hz).
19F NMR (CDCl3) δ: -76.49 (s).
7

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 9
Synthesis of 1,1,1-trifluoro-α,β-epoxyketone hydrates.
aIsolated yield.
b19F-NMR yield using 4-chlorobenzotrifluoride as the internal standard.
Scheme 4. Plausible reaction mechanism.
Scheme 5. Reaction of 2n with NaOCl.
Scheme 6. Reaction mechanism of epoxidation of 2n with NaOCl.
8

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
Table 10
Comparison of the fluorinated allylic alcohol (1n) vs the non-fluorinated allylic alcohol (4).
Entry
R
Time
Staring Material 1n or 4 (%)a
Enone 2n or 5 (%)a
α
,β-epoxyketone 3′n or 6 (%)a
1
2
CF3
27 min
0
0
89
0
CH3
4 h 32 min
83
17
a
Isolated yield.
4.15. 2,2,2-Trifluoro-1-[(2RS, 3SR)-3-phenyloxiranyl]ethenone-1,1-diol
4.21. Standard experimental procedure for the synthesis of 1,1,1-
trifluoromethylꢀ ,βꢀ epoxyketone hydrates
(3n) [22]
α
Colorless crystals, mp: 139–140 ºC (lit.130–140 ºC [22])
1H NMR (CDCl3) δ: 3.48 (1H, d, J = 1.4 Hz), 4.03 (1H, d, J = 1.4 Hz),
4.20–4.26 (2H, br), 7.29–7.41 (5H, m)
The 1,1,1-trifluoromethylꢀ α,βꢀ epoxyketone hydrates were synthe-
sized according to the same procedure for the oxidation of fluorine-
containing alcohols using NaOCl⋅5H2O catalyzed by TEMPO.
13C NMR (CDCl3) δ: 55.40, 60.43, 92.25 (q, J = 32.6 Hz), 121.76 (q, J
= 287.5 Hz), 125.74, 128.69, 128.95, 134.36.
4.22. 2,2,2-Trifluoro-1-[(2RS, 3SR)-3-(4-nitrophenyl)oxiranyl]ethane-
19F NMR (CDCl3) δ: -84.83 (s).
1,1-diol (3′s)
Pale yellow crystals. mp: 79-82 ℃
4.16. Standard experimental procedure for the oxidation of fluorine-
1H NMR (Acetone-d6) δ: 3.03 (1H, s), 3.46 (1H, d, J = 1.8 Hz), 4.27
(1H, d, J = 1.8 Hz), 6.40 (1H, brd, J = 6.4 Hz), 7.65 (2H, d, J = 9.0 Hz),
8.26 (2H, d, J = 9.0 Hz).
containing alcohols using NaOCl⋅5H2O catalyzed by TEMPO in CD3CN
Potassium hydrogen sulfate (6.7 mg, 0.05 mmol), TEMPO (1.7 mg,
13C NMR (Acetone-d6) δ: 53.79, 61.81, 91.73 (q, J = 31.6 Hz),
124.06 (q, J = 287.5 Hz), 124.36, 127.81, 144.53, 148.86.
19F NMR(Acetone-d6) δ: -83.65 (3F, s).
0.01 mmol), and 4-chlorobenzotrifluoride (134.5 μL, 1.00 mmol, as the
internal standard) were added to a stirred solution of a fluoroalkyl
alcohol (1) (1.00 mmol) in CD3CN (5 mL) at room temperature, and the
resulting mixture was cooled to 0 ºC. Sodium hypochlorite pentahydrate
(395 mg, 2.4 mmol) was then added to the reaction mixture and stirred
under the same conditions. Part of the reaction solution (0.4 mL) was
collected, the 19F NMR was measured, and the yield of 2 was calculated.
IR (neat) cmꢀ 1: 3445, 3084, 2990, 1605, 1525, 1346, 1165, 1116,
1036, 908, 849, 721, 514, 453
MS (EI) (m/z): 261 [(M--H2O)+]
HRMS (ESI) calcd for C10H6F3NO4Na [(M+Na--H2O) +], 284.0141,
found: 284.0138.
4.17. Trifluoroacetic acid (2o) [23]
4.23. 1-[(2RS, 3SR)-3-(4-Fluorophenyl)oxiranyl]-2,2,2-trifluoroethane-
1,1-diol (3′t) [22]
The 19F-NMR peaks of this sample were identical to those of com-
mercial reagents.
Colorless crystals. mp: 76–78 ℃ (lit. 79–81 ℃ [22]).
1H NMR (CDCl3) δ: 3.43 (1H, d, J = 1.60 Hz), 3.71 (1H, br), 3.91 (1H,
br), 3.99 (1H, d, J = 1.60 Hz), 7.05-7.12 (2H, m), 7.25-7.33(2H, m).
13C NMR (CDCl3,) δ: 55.03, 59.73, 90.83 (q, J = 33.1 Hz), 115.83 (d,
J = 21.1 Hz), 122.29 (q, J = 285.6 Hz), 127.65 (d, J = 8.5 Hz), 129.98
(d, J = 2.8 Hz), 163.20 (d, J = 248.2 Hz).
19F NMR (CD3CN) δ: -75.3~ -75.4 (br).
4.18. 2,2-3,3,4,4,4-Heptafluorobutylic acid (2p) [23]
The 19F-NMR peaks of this sample were identical to those of com-
mercial reagents.
19F NMR (CDCl3) δ: -84.81 (3F, s), -112.75-112.68 (1F, m).
19F NMR (CD3CN) δ: -62.22 (3F, t, J = 8.7 Hz).ꢀ 117.44 (2F, q, J =
8.7 Hz), -126.82 (2F, brs).
4.24. 1-[(2RS, 3SR)-3–(4-Bromophenyl)oxiranyl]-2,2,2-
trifluoroethane-1,1-diol (3′u) [22]
4.19. Ethyl trifluoroacetate (2q) [23]
Colorless crystals. mp: 76–78 ℃ (lit. 77–78 ℃ [22])
1H NMR (CDCl3) δ: 3.40 (1H, d, J = 1.8 Hz), 3.96 (1H, d, J= 1.8 Hz),
7.1 8 (2H, d, J = 8.4 Hz). 7.52 (2H, d, J = 8.4Hz).
13C NMR (CDCl3) δ: 54.95. 59.78, 90.80 (d, J = 32.9 Hz), 122.28 (q, J
= 286.6 Hz), 123.11, 127.43, 131.93, 133.39.
1
The H-, 13C- and 19F-NMR peaks of this sample were identical to
those of commercial reagent.
1H NMR (CD3CN) δ: 1.32 (3H, t, J = 8.0 Hz), 4.39 (2H, q, J = 7.2 Hz).
13C NMR (CD3CN) δ: 13.91, 65.58, 115.28 (q, J = 284.7 Hz), 157.92
(q, J = 41.2 Hz).
19F NMR(CDCl3) δ: -84.83 (s)
19F NMR (CD3CN) δ: -75.2 (s).
4.25. 1-[(2RS, 3SR)-3-Butyloxiranyl]-2,2,2-trifluoroethane-1,1-diol
(3′x)
4.20. 1,1,1,3,3,3-Hexafluoro-2,2-propanediol (2′r) [24]
Pale yellow oil.
The 19F-NMR peak of this sample was similar to that (measured in
CDCl3) reported in the literature. [24]
1H NMR (CDCl3) δ: 0.92 (3H, t, J = 5.2 Hz), 1.36–1.74 (6H, m), 3.05
(1H, td, J = 5.7, 2.2 Hz), 3.14 (1H, d, J = 2.2 Hz).
13C NMR (CDCl3) δ: 13.81, 22.26, 27.55, 30.50, 56.02, 56.61, 90.86
19F NMR (CD3CN) δ: -82.4 (s)
9

M. Kirihara et al.
Journal of Fluorine Chemistry 243 (2021) 109719
(d) T. Okada, H. Matsumuro, S. Kitagawa, T. Iwai, K. Yamazaki, Y. Kinoshita,
Y. Kimura, M. Kirihara, Synlett 26 (2015) 2547;
(q, J = 32.6 Hz), 122.42 (q, J = 286.2 Hz).
19F NMR (CDCl3) δ: -84.86 (s).
(e) T. Okada, H. Matsumuro, T. Iwai, S. Kitagawa, K. Yamazaki, T. Akiyama,
T. Asawa, Y. Sugiyama, Y. Kimura, M. Kirihara, Chem. Lett. 44 (2015) 185;
(f) T. Okada, T. Asawa, Y. Sugiyama, M. Kirihara, T. Iwai, Y. Kimura, Synlett
(2014) 596;
IR (neat) cmꢀ 1: 3365, 2348, 1450, 1259, 1187, 1110, 1066, 984,
913, 850, 722, 590.
HRMS (ESI) calcd for C8H11F3O2 [(M-H2O) +], 197.1767, found:
197.1761.
(g) S. Kitagawa, H. Mori, T. Odagiri, K. Suzuki, Y. Kikkawa, R. Osugi, S. Takizawa,
Y. Kimura, M. Kirihara, SynOpen 3 (2019) 21;
(h) M. Kirihara, R. Osugi, K. Saito, K. Adachi, K. Yamazaki, R. Matsushima,
Y. Kimura, J. Org. Chem. 84 (2019) 8330;
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgments
(i) N. Hakuto, K. Saito, M. Kirihara, Y. Kotsuchibashi, Polymer Chem. 11 (2020)
2469.
[6] (a) Organic syntheses using NaOCl⋅5H2O were reported by other groups:
M. Uyanik, N. Sasakura, M. Kuwahata, Y. Ejima, K. Ishihara Chem. Lett. 44 (2015)
381;
(b) T. Hirashita, Y. Sugihara, S. Ishikawa, Y. Naito, Y. Matsukawa, S. Araki, Synlett
29 (2018) 2404;
(c) A. Watanabe, K. Miyamoto, T. Okada, T. Asawa, M. Uchiyama, J. Org. Chem.
83 (2018) 14262;
This study was supported by the Tokai Foundation for Technology.
We grateful to Nippon Light Metal Co., Ltd., for supplying NaOCl⋅5H2O.
We also gratefully thank the Tosoh Finechem Corporation for providing
(trimethylsilyl)trifluoromethane and (trimethylsilyl)difluoromethane.
(d) S.A. Miller, K.A. Bisset, N.E. Leadbeatera, N.A. Eddya, Eur. J. Org. Chem
(2019) 1413;
(e) A. Porcheddu, F. Delogu, L.D. Luca, C. Fattuoni, E. Colacino, Beilstein J. Org.
Chem. 15 (2019) 1786.
[7] R.V. Stevens, K.T. Chapman, C.A. Stubbs, W.W. Tam, K.F. Albizati, Tetrahedron
Lett. 23 (1982) 4647. Conventional aqueous NaOCl has been known as a selective
oxidant for none-fluorinated secondary alcohols under catalyst-free conditions.
Primary alcohols are almost inactive under the same conditions;.
[8] Recently, Hirashita et al. reported that secondary alcohols react with NaOCl⋅5H2O
in acetonitrile under catalyst-free conditions [6b].
References
[1] (a) Part of this study has been reported at the Fluorine Conference of Japan:
K. Suzuki, H. Shimazu, K. Nakakura, K. Saito, Y. Kikkawa, Y. Kimura, M. Kirihara,
in: The 40th Fluorine Conference of Japan, November 14–15, in Tottori; P-33,
2017;
[9] In the cases of 1a, b, f, similar results were reported by Leadbeatera and Eddya
[6d].
(b) M. Kirihara, K. Suzuki, H. Shimazu, K. Nakakura, K. Saito, Y. Kikkawa,
Y. Kimura, in: The 41th Fluorine Conference of Japan, October 26, in Hirosaki; O-
26, 2018.
[10] Although the α,β-epoxyketone hydrate 3’n was produced from the reaction of 1n
with NaOCl⋅5H2O in the absence of the TEMPO-catalyst, the reaction proceeded
slower and the yield of 3’n was lower (Table 7, entry 6). The TEMPO-free condition
[2] (a) W. Wu, Z. Weng, Synthesis 50 (2018) 1958;
(b) C.B. Kelly, M.A. Mercadantea, N.E. Leadbeater, Chem. Commun. 49 (2013)
11133.
is not attractive as a synthetic method of α,β-epoxyketones (3).
[11] (a) J.M. Bobbitt, Tcimail, Tokyo Chemical Industry, Ltd. No. 146, 2010, p. 2.
April;
[3] (a) R.J. Linderman, D.M. Graves, Tetrahedron Lett. 28 (1987) 4259;
(b) R.J. Linderman, D.M. Graves, J. Org. Chem. 54 (1989) 661;
(c) S.V. Ley, A. Madin, Comp. Org. Synth. 7 (1991) 256;
(b) W.F. Bailey, J.M. Bobbitt, K.B. Wiberg, J. Org. Chem. 72 (2007) 4504.
[12] 19F NMR yield using 4-chlorobenzotrifluoride as the internal standard. Several
chlorinated by-products were produced in both cases.
´
´
(d) V. Kesavan, D. Bonnet-Delpon, J. Begue, A. Srikanth, S. Chandrasekaran,
[13] (a) R. Krishnamurti, D.R. Bellew, G.K.S. Prakash, J. Org. Chem. 56 (1991) 984;
(b) Y. Zhao, W. Huang, J. Zheng, J. Hu, Org. Lett. 13 (2011) 5342.
[14] C.B. Kelly, M.A. Mercadante, T.A. Hamlin, M.H. Fletcher, N.E. Leadbeater, J. Org.
Chem. 77 (2012) 8131.
Tetrahedron Lett. 41 (2000) 3327;
(e) Z. Mei, T. Omote, M. Mansour, H. Kawafuchi, Y. Takaguchi, A. Jutand,
S. Tsuboi, T. Inokuchi, Tetrahedron 64 (2008) 10761;
(f) C.B. Kelly, M.A. Mercadante, T.A. Hamlin, M.H. Fletcher, N.E. Leadbeater, J.
Org. Chem. 77 (2012) 8131;
[15] G.K.S. Prakash, J. Hu, M.M. Alauddin, P.S. Conti, G.A. Olah, J. Fluorine. Chem. 121
(2003) 239.
(g) Y. Tanaka, T. Ishihara, T. Konno, J. Fluorine Chem. 137 (2012) 99;
(h) Y. Kadoh, M. Tashiro, K. Oisaki, M. Kanai, Adv. Synth. Catal. 357 (2015) 2193;
(i) V.A. Pistritto, J.M. Paolillo, K.A. Bisset, N.E. Leadbeater, Org. Biomol. Chem.
16 (2018) 4715;
[16] M.M. Kremlev, A.I. Mushta, W. Tyrra, D. Naumann, H.T.M. Fischer, Y.
L. Yagupolskii, J. Fluorine Chem. 128 (2007) 138.
[17] R.L. Salvador, M. Saucier, Tetrahedron 27 (1971) 1221.
[18] W.H. Pirkle, D.L. Sikkenga, M. Pavlin, J. Org. Chem. 42 (1977) 384.
[19] I. Pravsta, M. Zupana, S. Stavber, Synthesis 18 (2005) 3140.
[20] D. Yang, Y. Zhou, N. Xue, J. Qu, J. Org. Chem. 78 (2013) 4171.
[21] P.V. Ramachandran, B. Gong, A.V. Teodorovic’, H.C. Brown, Tetrahedron
Asymmetry 5 (1994) 1061.
(j) H. Cheng, Y. Pei, F. Leng, J. Li, A. Liang, D. Zou, Y. Wua, Y. Wub, Tetrahedron
Lett. 54 (2013) 4483.
[4] S. Krishnamoorthy, G.K.S. Prakash, Synthesis 49 (2017) 33946.
[5] (a) Organic syntheses using NaOCl⋅5H2O were reported by us: M. Kirihara,
T. Okada, T. Asawa, Y. Sugiyama, Y. Kimura J. Synth. Org. Chem. Jpn. 78 (2020)
11;
[22] C. Zheng, Y. Li, Y. Yang, H. Wang, H. Cui, J. Zhang, G. Zhaoa, Adv. Synth. Catal.
351 (2009) 1685.
(b) M. Kirihara, T. Okada, Y. Sugiyama, M. Akiyoshi, T. Matsunaga, Y. Kimura,
Org. Process Res. Dev. 21 (2017) 1925;
[23] This compound is commercially available.
[24] T. Hayasaka, Y. Katsuhara, T. Kume, T. Yamazaki, Tetrahedron 67 (2011) 2215.
(c) T. Okada, T. Asawa, Y. Sugiyama, T. Iwai, M. Kirihara, Y. Kimura, Tetrahedron
72 (2016) 2818;
10