Synthesis of highly branched perfluoroolefins that are super-congested via multi-substitution of trifluoromethyl groups: Trifluoromethylation of hexafluoropropene trimers with Ruppert-Prakash reagent

Article abstract of DOI:10.1016/j.jfluchem.2016.12.001

The reaction of hexafluoropropene trimers with Ruppert-Prakash (CF₃SiMe₃) reagent gave highly congested perfluoroolefins such as the mono-trifluoromethylation products F-2,4-dimethyl-3-isopropyl-2-pentene (P1) and E- and Z-forms of F-4,4-dimethyl-3-isopropyl-2-pentene (2E and 2Z), and the bis-trifluoromethylation product F-2,4,4-trimethyl-3-isopropyl-2-pentene (3). The E-form of 2 was also comprised of two rotamers. Various aprotic polar solvents were surveyed for this reaction, and it was found that the aprotic solvent DMI has a unique solvent effect to selectively give the mono-trifluoromethylated perfluoroolefin P1, a precursor for the persistent perfluoroalkyl radical F-3-isopropyl-2,4-dimethyl-3-pentyl, in very high yield.

Full text of DOI:10.1016/j.jfluchem.2016.12.001

Accepted Manuscript
Title: Synthesis of highly branched perfluoroolefins that are
super-congested via multi-substitution of trifluoromethyl
groups: Trifluoromethylation of hexafluoropropene trimers
with Ruppert-Prakash reagent
Author: Taizo Ono
PII:
S0022-1139(16)30492-4
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.jfluchem.2016.12.001
FLUOR 8909
To appear in:
FLUOR
Received date:
Revised date:
Accepted date:
27-5-2016
30-11-2016
2-12-2016
Please cite this article as: Taizo Ono, Synthesis of highly branched perfluoroolefins
that are super-congested via multi-substitution of trifluoromethyl groups:
Trifluoromethylation of hexafluoropropene trimers with Ruppert-Prakash reagent,
Journal of Fluorine Chemistry http://dx.doi.org/10.1016/j.jfluchem.2016.12.001
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Synthesis of highly branched perfluoroolefins that are
super-congested via multi-substitution of trifluoromethyl
groups: Trifluoromethylation of hexafluoropropene trimers
with Ruppert-Prakash reagent
Taizo Ono*
National Institute of Advanced Industrial Science and Technology (AIST) 2266-98, Anagahora,
Moriyama, Nagoya, Aichi, 463-8560, Japan
Corresponding author’s E-mail: t.ono@aist.go.jp;
Phone (81)-(0)52-736-7491, Fax (81)-(0)52-736-7224
Keywords: persistent perfluoroalkyl radical, hexafluoropropene trimers, rotamer, Ruppert-Prakash
reagent, trifluoromethylation, solvent effect, 1,3-dimethylimidazolidinone

2
Graphical abstract

Synthesis of highly-branched perfluoroolefines super-congested with multi-substitutions of
trifluoromethyl grouops: Trifluoromethylation of hexafluoropropene trimers with Ruppert-Prakash
reagent for the synthesis of a precursor of persistent perfluoroalkyl radical, F-3-isopropyl-2,4-dimethyl-
3-pentyl
Taizo Ono*
National Institute of Advanced Industrial Science and Technology (AIST), 2266-98, Anagahora, Moriyama,
Nagoya, Aichi, 463-8560, Japan
CF3SiMe3
CF3
CF3
"cat. F-"
RT
F-
F-
F-
F-
+
+
Aprotic polar
solvent
F3C
F3C
T-2, or T-3
3
P1
2Z, 2E
Mono-CF3 substitution
Bis-CF3 substitution
Solvent effects on the reaction of hexafluoropropene trimers (T-2 or T-3) with Ruppert-Prakash reagent
(CF3SiMe3) was investigated and found mono-trifluoromethylated product P1 was selectively produced
in 1,3-dimethylimidazolidinone in high yield. F- denotes all unmarked bond to fluorine.

Highlights
 Ｒeaction of hexafluoropropene trimers with Ruppert-Prakash reagent was
investigated.
 Mono- and bis-trifluoromethlation products were obtained.
 Solvent effects on the reaction was surveyed for various aprotic solvents.
 Prominent solvent effect was found for 1,3-dimethylimidazolidinone.
 High-yield selective synthesis of mono-trifluoromethylated F-olefin, a precursor
for the persistent perfluoroalkyl radical, F-2,4-dimethyl-3-isopropyl-3-pentyl was
realized.

3
Abstract
The reaction of hexafluoropropene trimers with Ruppert-Prakash (CF3SiMe3) reagent gave highly
congested
perfluoroolefins
such
as
the
and
mono-trifluoromethylation
products
of
F-2,4-dimethyl-3-isopropyl-2-pentene
(P1)
E- and Z-forms
F-4,4-dimethyl-3-isopropyl-2-pentene (2E and 2Z), and the bis-trifluoromethylation product
F-2,4,4-trimethyl-3-isopropyl-2-pentene (3). The E-form of 2 was also comprised of two rotamers.
Various aprotic polar solvents were surveyed for this reaction, and it was found that the aprotic
solvent DMI has a unique solvent effect to selectively give the mono-trifluoromethylated
perfluoroolefin
P1,
a
precursor
for
the
persistent
perfluoroalkyl
radical
F-3-isopropyl-2,4-dimethyl-3-pentyl, in very high yield.

4
Introduction
Persistent perfluoroalkyl radicals, F-3-isopropyl-2-methyl-3-pentyl (PFR-1) and
F-3-isopropyl-2,4-dimethyl-3-pentyl (PFR-2), which
Rf
are stable enough for distillation and GC analysis, were
F-
.
discovered by serendipity in the course of research on
artificial blood substitutes (Fig. 1) [1]. PFR-1 can be
PFR-1: Rf = F
PFR-2: Rf = CF3
easily prepared from a mixture of hexafluoropropene
(HFP)
trimers
comprised
(T-2)
of Fig. 1. Structures of persistent perfluoroalkyl
radicals, PFR-1 (F-3-ethyl-2,4-dimethyl-3-
pentyl) and PFR-2 (F-2,4-dimethyl-3-isopropyl-
F-3-isopropyl-4-methyl-2-pentene
and
3-pentyl): F- refers to all unmarked bonds being
to fluorine atoms.
R
F-3-isopropyl-2-methyl-2-pen
R-
F2
tene
(T-3)
by
direct
.
fluorination using F2 gas
(high yield) or even by
electrochemical fluorination
(low yield) [2]. Simple,
high-yield reactions using the
industrially available mixture
PFR-2
P1: R = CF3
P2: R = CH3
T-2
Scheme 1. Potential preparative methods to PFR-2 by the direct fluorination
of precursor olefins, F-2,4-dimethyl-3-isopropyl-2-pentene (P1) and 2-
methyl-F-4-methyl-3-isopropyl-2-pentene (P2): all unmarked bonds are to
fluorine atoms.
of HFP trimers can be easily scaled up, and thus PFR-1 has been repetitively prepared
on a kilogram scale in our hand with no trouble. Therefore, its industrial applications
have been pursued in various fields such as surface modification of polymers and
diamond-like coatings (DLCs) on various materials [3-4], ESR imaging and standard
[5-7], trifluoromethylation reagent [8], and radical initiator for the synthesis of
fluoropolymers [9-14]. Its unique physicochemical properties to form CT-complexes
and its photochemistry were recently reported [15-16]. In contrast to PFR-1, however,
PFR-2 was only detected via ESR measurements on a sample obtained from the thermal
reaction of PFR-1 and T-2. But PFR-2 was not obtained in large quantities, and thus its
practical synthesis is necessary for further characterization. With this background in
mind, we decided to find synthetic routes to precursor olefins such as F-2,4-dimethyl-3-
isopropyl-2-pentene (P1) and 2-methyl-F-4-methyl-3-isopropyl-2-pentene (P2) for the
preparation of PFR-2 (Scheme 1). If these precursors are available, their direct
fluorination could provide highly probable synthetic routes to PFR-2. However, our
prior work on the preparation of P2 showed that the reaction of methyl lithium or
methyl Grignard reagents with T-2 gave a miserable mixture with more than 10 various
methylated by-products [17]. Therefore, we next targeted the synthesis of P1. We
selected the Ruppert-Prakash reagent (R-P reagent) for this purpose. The

5
Ruppert-Prakash reagent (R-F reagent) is quite a useful trifluoromethylating agent, and
thus it has been used in various nucleophilic reactions. A recent new method for
preparing the R-P reagent based on the industrial waste trifluoromethane, CF3H,
empowers the reagent as a stronger industrial arsenal for wider applications [22]. The
intermediate trifluoromethyl anion has just recently been proved to be long-lived in
solution and was even characterized in the
C8C17SiMe3
CF3CF=CF2
solid state [23-25]. It is now highly
speculated as the true species for the
nucleophilic trifluoromethylation reaction
with the R-F reagent. Various reactions with
the R-F reagent have been developed due to
industrial needs for trifluoromethyl group(s)
CF3CF=CFC8F17
58 % yield
CF2
CF2
C8C17SiMe3
F2C
F
CF2
F2C
F
CF2
C8F17
F
90 % yield
Scheme 2. Prior work on the reaction of a higher homolog of
Ruppert-Prakash reagent (C8C17SiMe3) with perfluoroolefines.
in pharmaceutical, agrochemical, and materials science, and these developments have
been summarized in excellent reviews [26-29]. Generally, C=X (X = O, N) represents
the functional groups manipulated by this reagent. The most prominent and important
reaction is the addition of the
CH3
CH3
MeLi
or
MeMgX
CH3
trifluoromethyl anion to carbonyl
groups. In contrast, examples with
C=C double bonds have not been
studied much. Recently, C-H bond
activation with metal catalysis has
been developed to give a chance of
using CF3SiMe3 to introduce a
trifluoromethyl group to a C=C
double bond; however, examples of
Many multi-
substituted
by-products
(a)
+
+
CH3
T-2
T-2
PhO
OPh
OPh
OPh
NaOPh
DMF
PhOH
(b)
(c)
Et3N
DMF
NaOPh
DMF
OPh
T-3
Scheme 3. Prior art of reactions of T-2 with (a) C-nucleophiles (organometallic reagents
such as MeLi and MeMgX) and with (b) O-nucleophiles (PhOH/Et3N/DMF, then
NaOPh/DMF), and (c) of T-3 with NaOPh/DMF (all unmarked bonds are to fluorine
atoms).
C=C-F bond activation for reaction with the R-P reagent are still very scanty [30]. This
type of reaction has only been reported for the reaction of some perfluoroolefins with
the higher homologue of the R-P reagent, namely C8F17SiMe3, or with analogous
congeners such as CF2=CFSiMe3 and C6F5SiMe3 (Scheme 2) [31-34]. Thus, the reaction
of perfluoroolefins with CF3SiMe3 still has room to be studied in connection with our
concerns. HFP trimers are very susceptible toward heteroatom nucleophiles such as

6
metal alkoxides, thiolates, and amines, usually giving a mixture of multi-substitution
products [18], but in the particular case with phenolate anion, its nucleophilic attack can
be controlled by the conditions used. While the nucleophilic substitution stops at the
first stage with the reaction of PhOH and Et3N in DMF, leading to one simple
substitution product in high yield [19-21], with NaOPh in DMF, further nucleophilic
attack occurred to give the di-substituted product in good yield (Scheme 3).
A comparative study on the reactions of T-2 and T-3 with the R-P reagent will be reported in the
present work (Scheme 4).
Results and discussions
Since the bulkiness of the trifluoromethyl group is believed to be close to that of a tert-butyl group
[35], the nucleophilic approach of a trifluoromethyl anion would seemingly have a high steric barrier
with respect to the highly substituted HFP trimers T-2 and T-3.
We examined first the reaction of T-2 with the R-P reagent using DMF as solvent, which is generally
used for the RP reaction, and the results are summarized in Table 1. As a preliminary test, the
reaction was conducted in the absence of added fluoride anion. No reaction occurred at all, and the
starting T-2 was recovered (Run 1). Next the same reaction was conducted with the addition of a
catalytic amount of potassium bifluoride. The biphasic reaction comprising the upper layer (DMF +
CF3SiMe3) and lower layer (perfluoroolefin) proceeded at room temperature over 1 hour to give the
desired perfluoroolefin P1 in reasonable yield of 62.7% (Run 2). Therefore, the presence of fluoride
anion is essential for this reaction. The structure of P1 was confirmed to have two interlocking
19
iso-C3F7 groups by F-NMR spectroscopy (Fig. 2). The coupling constants are consistent with the
structure depicted in Fig. 2. In addition to mono- trifluoromethylation, further consecutive
trifluoromethylations
occurred
to
give
bis-trifluoromethylated
F-2,4,4-trimethyl-3-isopropyl-2-pentene (3) in
13.3% yield. With these bulky trifluoromethyl
groups surrounding the C=C bond, the formation of
3 having seven (7) trifluoromethyl groups jammed
in such close proximity of the C=C bond
CF3:
CF3:
-59.35
septet
quartet
-56.07
doublet
quartet
12.0 Hz
15.5 Hz
58.6 Hz
F3C
CF3
CF3
CF3
2CF3:
CF3
F
F: -157.5
CF3SiMe3
-70.11
qaurtet
doublet
F
+
broad
?
CF3
CF3
12.7 Hz
is unexpected. However, its formation is
?
multiplet
T-2
F: -153.6
septet
doublet
confirmed without doubt from its
19F-NMR spectrum (Fig. 3). It should
also be mentioned that one additional
2CF3:
-70.56
broad
CF3SiMe3
36.4 Hz
+
?
doublet
?
CF3
P1
T-3
Fig. 2 19F-NMR interpretation for the precursor perfluoroolefin
of PFR-2, F-2,4-dimethyl-3-isopropyl-2-pentene, P1.
product
F-(Z)-4,4,-dimethy-3-
Scheme 4. Present work on the reactions of T-2 and T-3 with Ruppert-Prakash
reagent. Solvent effects on this reaction were also surveyed (all unmarked bonds
are to fluorine atoms).
isopropyl-2-pentene (2Z) was observed

7
albeit in rather low yield (2.6%). The structure of 2Z is also confirmed by its 19F-NMR spectrum
(Fig. 3). The assignment of compound 2 to the Z-isomer is based on the key observations of both the
CF3 and F groups bonded the same terminal carbon atom of the C=C double bond, of which the
couplings are a dectet for the CF3 group (-66.29 ppm, 11.7 Hz) due to the three (3) neighboring CF3
groups of the tert-(CF3)3C moiety, and a septet for the F atom (-67.03 ppm, 39.7 Hz) due to the two
(2) neighboring CF3 groups
F: -67.03
septet
CF3:
"F-"
CF3:
"F-"
F: -66.29
decaplet
doublet
-55.09
39.7 Hz
-56.83
F-
F-
-
of iso-C3F7 group. The
observations of a dectet for
the F atom on the iso-C3F7
group due to coupling with
F-
doublet
quartet
T-2
4.2 Hz
septet
decaplet
F3C
CF3
12.9 Hz
T-3
F3C
F
15.5 Hz
CF3
15.5 Hz
F3C
11.7 Hz
C
"
Fc3at. F-"
"cat. F-"
2CF3:
CF3SiMe3
-70.40
CF
D3MF
DMF
CF3
2CF3:
-72.38
F3C
CF3SiMe3
RT
F
RT
F
doublet
minor
or
CF3
broad singlet
CF3
CF3
CF3
CF3
CF3
the
tert-(CF3)3C
moiety
F: -166.3
decaplet
doublet
major
F: -146.99
F-
no
F-
F3C
3CF :
3F-
-60.05
doublet
quartet
3CF3:
-57.98
broad singlet
decaplet
F3C
(-166.3 ppm, 26.7 Hz), and a
doublet for the two (2) CF3
groups of the iso-C3F7 group
due to coupling with the
previously mentioned F atom
31.7 Hz
2
26.7 Hz
3
P1
Z- or E-form
3
2Z
Scheme 5. The formation of the super-congested perfluoroolefins, P1and 2 having six (6)
trifluoromethyl substituents around a C=C bond and 3 with seven (7) trifluoromethyl
Fig. 3. 19F-NMR interpretation for the Z-form of F-4,4-dimethyl-3-isopropyl-2-pentene,
substituents. The main path for the formation of 3 is through 2. F- denotes all unmarked
2Z and F-2,4,4-trimethyl-3-isopropyl-2-pentene, 3.
bonds are to fluorine atoms.
on the terminal carbon atom of the olefin (-72.38 ppm, 39.7 Hz), suggest hindered rotation of the
iso-C3F7 group. The by-product 2Z undoubtedly comes from T-3, as the starting T-2 contains
approximately 10% of T-3 as an impurity. Since 2.6% of T-3 was consumed in the formation of 2Z,
7.4% should remain in the reaction mixture. However, more than 10% of T-3 remained in the
reaction mixture, actually ca. 14%. This inconsistency is explained by the existence of an
equilibrium between T-2 and T-3 under the reaction conditions (vide infra). The prolongation of the
reaction time to up to 12 hours does not change the yields of the products (Run 3). When the R-P
reagent was used in a 2:1 molar excess, the starting T-2 and T-3 were completely consumed, giving
P1 and 3 in 68.9 and 31.1% yields, respectively (Run 4). The formation of 3 is possible through both
P1 and 2Z in principle; however, we speculate its formation mainly through 2Z (vide infra). If this
speculation is correct, then 31.1% yield of 3 is contradictory with the T-3 content (10% as impurity
in the starting T-2). It is known that T-2 and T-3 are interconvertible under the presence of a fluorine
anion catalyst [36-37]. Such an interconversion was also confirmed in the case of our reaction when
run under the identical conditions but without the addition of the R-P reagent, and thus, the starting
mixture (T-2: 90%, T-3: 10%) became the mixture (T-2: 71.8%, T-3: 28.2%) enriched with T-3 after
1 hour of stirring at room temperature (Run 5). Further stirring shifted the composition up to almost
the thermodynamic equilibrium mixture (T-2: 33.7%, T-3: 66.3%). This fluoride anion catalyzed
shift of the composition through the intermediate F-di-isopropylethylmethanide anion explains the
production of 3 being more than the starting 10% (Scheme 5). All reactions were conducted by using
1 mL of DMF throughout with the exception of Run 6. When the volume of the solvent is increased

8
to 5 mL (Run 6), the yield of P1 increased from 62.7 to 69.3, and in response, the production of 3 is
somewhat suppressed to 6.5%. When the quantity of R-P reagent used is increased while keeping
the volume of the solvent unchanged (1 mL), the properties of the reaction medium approach closer
to the properties of the R-P reagent. Thus, when the scale of the reaction of Run 4 was doubled as in
Run 7 with the volume of the solvent remaining unchanged at 1 mL, the reaction media might be
quite different from DMF itself. Thus, under reaction conditions where DMF was almost like an
additive rather than a solvent, the reaction still proceeded well in a bilayer system to afford P1 and 3
in 65.5% and 34.5% yields, respectively, almost the same results as Run 4 (Run 7). This is true for
further increases in the quantities of the substrate and R-P reagent as in Run 8 (2 mmole of T-2 and
8 mmole of CF3SiMe3), resulting in some yield improvement of P1 up to 74.0% with less production
of 3 (Run 8). Potassium bifluoride was used as the fluoride source due to easy operation even under
moist atmosphere. When potassium fluoride was used instead as the fluoride source, the reaction
becomes slow in comparison with the potassium bifluoride case (Run 9). Interestingly, only P1 was
produced in 39.3% yield under the reaction conditions similar to Run 2 (Run 9). When the amount of
KF is increased to 0.3 mmole, under the same reaction conditions as Run 2, the yield was increased
to 77.7% with less production of 3 (Run 10). With an increase in the amount of R-P reagent, the
production of P1 increased, and the production of 3 decreased to less than 10%, suggesting
circumstantially that the formation of 3 is not through P1 but through 2. Otherwise such a high yield
of P1 with less 3 is not attainable.
We next studied the reaction of T-3 with the R-P reagent, and the results are summarized in Table
2. The reaction was conducted at first with 1.1 equivalent of CF3SiMe3 against T-3 (Run 1). The
mono-trifluoromethylated product 2Z was obtained in 23.3% yield, and the bis- trifluoromethylated
product 3 was also obtained in a comparative yield 21.1% (based on capillary gas chromatography).
Interestingly, in addition to 2Z one more unknown peak appeared very close to 2Z in the gas
chromatogram. The unknown peak was isolated by preparative scale GC using a glass column
described in the experimental section. Careful investigation of this sample by 19F-NMR spectroscopy
revealed that the peak was comprised two components in a molar ratio of 2 to 1. Both components
have the same structural units in common with 2Z, one CF3, one F, one iso-CF(CF3)2, and one
C(CF3)3 bonded to C=C bond, and thus, should be the geometrical isomer of 2Z, namely the E-form.
Then, why are there two components with the same structural units? All the chemical shifts are quite
similar each other, while the coupling patterns and constants are quite different. All of the features of
19F-NMR spectrum can be reasonably
F: -65.57
decaplet
doublet
F: -67.97
decaplet
doublet
6.8 Hz
F
CF3:
-64.32
doublet
CF3:
-68.47
septet
53.8 Hz
explained by the existence of
rotational isomers with hindered
rotation around the CF(CF3)2 moiety
10.4 Hz
15.5 Hz
F
CF3
CF3
29.4 Hz
30.8 Hz
F3C
F: -142.6
quartet
doublet
CF3
CF3
F
CF3
F3C
F3C
2CF3:
-71.34
quartet
CF3
F
CF3
CF3
CF3
2CF3:
-68.49
decaplet
F: -158.2
decaplet
doublet
3CF3:
-60.18
doublet
doublet
3CF3:
-62.15
doublet
septet
30.3 Hz
8.6 Hz
2E-R1
2E-R2
Fig. 4. 19F-NMR interpretation for two rotamers, 2E-R1 and 2E-R2 of the E-form of F-4,4-
dimethyl-3-isopropyl-2-pentene, 2.

9
bonded to C=C bond (Fig. 4). The major rotational isomer is the form of 2E with the iso-C3F7 group
rotation fixed in keeping the C-F bond toward the (CF3)3C group (2E-R1). The other rotational
isomer is the form of 2E with the iso-C3F7 group rotation fixed in keeping C-F bond toward the
neighboring CF3 group (2E-R2). We have previously observed the same kind of rotational isomers in
similar structures in the reaction of HFP trimers with methyl lithium or Grignard reagents [17] and
also with C6F5SiMe3 [34]. A prior investigation by 19F-NMR spectroscopy has even shown that the
starting material T-3 is a mixture of two rotamers [38]. A similar rotational lock has also been
previously seen in the olefinic system with four iso-propyl groups [39].
The sample of T-3 used for this reaction was very pure as high as 99.7%, and thus, the formation of
P1, which comes from T-2, should be less than 0.3%; however, it is actually obtained in 1.1% yield.
The interconversion of T-3 to T-2 was examined under the same reaction conditions without
CF3SiMe3, and it was confirmed that about 12% of T-3 was converted to T-2 (Run 4). This fluoride
anion catalyzed interconversion was consistent with the result of Run 1 (the total amount of P1 and
T-2 is ca. 12%). Next, we conducted the same reaction using 2.0 equivalent CF3SiMe3 against T-3
and found that the yield of the bis-trifluoromethylated product 3 increased to 59.7%, almost triple of
Run 1 with the yields of the other products such as P1, 2Z and 2E unchanged (1.0, 21.7, and 14.0%,
respectively) (Run 2). Almost all the starting T-3 was consumed (2.5% remained). A further increase
of the amount of CF3SiMe3 did not have much influence on the yield of 3 (53.0% vs. 59.7%), but it
did have an effect on the yields of 2E and 2Z (Run 3). Interestingly, the total yield of the
mono-substituted products (2Z + 2E) remained mostly unchanged (35.7% for Run 2 and 41.7% for
Run 3), but the yield of 2E decreased to 7.2% from 14.0% and the yield of 2Z increased to 34.5%
from 21.7%. Further successive additions of CF3SiMe3 and catalyst to the reaction mixture as well as
two more hours of reaction time lead to an increase yield of 3 to 65.7% from 53.0%, but the reaction
is not yet complete as some T-3 remains. It is possible to conclude that the reactivity of the
trifluoromethyl anion is roughly the same as that of the phenolate anion. Thus, as in the case of T-2
and PhOH/Et3N/DMF, and now in the case of T-3 and NaOPh/DMF, the same pattern of products is
observed.
Finally, the same reaction of T-2 with the R-P reagent was conducted in various solvents to see if a
solvent effect existed on the selectivity and product distribution. The results are summarized in Table
3. For sake of comparison, the results from a representative DMF reaction are included in the table
again (Run 1). A very small change in the structure of the solvent from dimethylformamide (DMF)
to dimethylacetamide (DMA) does not affect the reaction, and almost the same product distribution
was attained under the same reaction conditions (Run 2). The same result was also obtained in the
case of the N-methylpyrrolidone (NMP) (Run 3). On the other hand, when the reaction was
conducted in CH3CN, the reaction was rather slow when compared to those run in DMF, DMA, and
NMP, and thus, P1 was obtained only in 38.5% even after the reaction time was prolonged to 14

10
hours (Run 4). Interestingly, the starting material T-2 (only 0.9% recovered) was recovered primarily
in a form of T-3 (48.6%), suggesting that the reactivity of T-3 is low in CH3CN. Dimethoxyethane
(DME) and diethylene glycol dimethyl ether (diglyme) did not work for this reaction at all, and the
interconversion between T-2 and T-3 also did not occur. Therefore, the starting composition
remained unchanged (Runs 5 and 6). A higher congener of diglyme, triethylene glycol dimethyl ether
(triglyme) gave interesting results with rather selective formation of P1 (51.2%) with no less than
1.1% yield of 3 (Run 7). Interestingly, the percentages of T-2 and T-3 recovered are 40.6% and 7.1%,
respectively, which means that T-2 becomes P1 and T-3 becomes 3, independently. When a two
molar excess of CF3SiMe3 was used, the same independent reactions of T-2 and T-3 follow, leading
to 80.1% of P1 and 6.9% of 3 with recoveries of 9.4% of T-2 and 3.6% of T-3, respectively (Run 8).
More interestingly, in the case of 1,3-dimethylimidazolidinone (DMI), very selective formation of
P1 results along with a rather good yield of P1 of 67.2% (Run 9). Actually, DMI suppresses the
formation of 2 and 3 from T-3; however, it does not prevent the interconversion of T-2 to T-3, thus
leading to an increase in the content of T-3 to 21.6%. With this result in mind, the use of CF3SiMe3
was increased to 4 equivalence, and P1 was selectively formed again in a yield of 88.4% (Run 10).
Interestingly, T-2 and T-3 were recovered in 7.9% and 3.7%, respectively, which may be a kinetic
composition. Thus, the reaction conditions used in Run 10 are quite idealistic for the synthesis of P1,
a precursor for the persistent perfluoroalkyl PFR-2.
Not much difference in physico-chemical parameters exists between DMI (b.p. 226 ºC, m.p. 8.2 ºC,
dipole moment: 4.05~4.09, dielectric constant at 25 ºC: 37.6) and DMF (b.p. 153 ºC, m.p. -61 ºC,
dipole moment: 3.82, dielectric constant at 20 ºC: 36.7), so it is rather surprising to see such a
solvent effect, and no explanation can be offered at this time. Further study of this solvent effect is
necessary and is now underway.
Conclusions
It is concluded that hexafluoropropene trimers, both F-4-methyl-3-isopropyl-2-pentene (T-2) and
F-3-ethyl-2,4-dimethyl-2-pentene (T-3), are trifluoromethylated with Ruppert-Prakash reagent to
give highly congested perfluoroolefins such as the mono-trifluoromethylation products
F-2,4-dimethyl-3-isopropyl-2-pentene (P1) and F-4,4-dimethyl-3-isopropyl-2-pentene (2) and also
the bis-trifluoromethylation product F-2,4,4-trimethyl-3-isopropyl-2-pentene (3). A solvent effect on
this reaction was surveyed, and it was found that trifluoromethylation can be controlled by careful
selection of the solvent. Aprotic polar solvents such as DMF, DMA, NMP, CH3CN, triglyme, and
DMI are all usable for this trifluoromethylation, whereas DME and diglyme are not. DMA and NMP
gave almost the same results as DMF, but CH3CN retards the reaction rate and thus is not practical.
As solvents, triglyme and DMI are rather specific for the formation of P1. Interestingly enough for
our purpose, DMI has such a prominent nature to control this trifluoromethylation to give the

11
mono-trifluoromethylated product P1, which is the desired precursor perfluoroolefin for the
synthesis of PFR-2, selectively in high yield, by suppressing the consecutive trifluoromethylation to
the bis-trifluoromethylated product 3. All other similar, polar aprotic solvents that were tried did not
stop the trifluoromethylation reaction at the first stage and thus, gave mono- and
bis-trifluoromethylation products as a mixture. At present, no clear explanation is available for the
unique solvent effect of DMI, which is under further investigation.
Experimental
Hexafluoropropene trimers are prepared by the oligomerization of hexafluoropropene according to
the method reported by Haszeldine et al. [40]. Each isomer T-2 and T-3 was isolated using a
spinning-band distillation column. The trimer T-2 obtained is shown to be 90% pure, as analyzed by
capillary GC, with 10% T-3 as an impurity. The trimer T-3 is 99.7% pure with 0.3% T-2 as an
impurity. All the solvents were dried over molecular sieve (MS 4A), sodium, or CaH2 and distilled
before use. 19F-NMR spectra were measured with a Varian Inova 300 operated at 282.24 Hz using
CDCl3 as solvent, and the chemical shifts are reported on the scale, with negative shifts to low
frequency of the internal reference of CFCl3. A capillary column (NB-1 1.5 , 0.25 mm x 60 m)
was used for the analysis, and a glass column (3 mm x 3m) packed with 13.2% Fomblin® on
Chromosorb W (AW-DMCS, 60-80 mesh) was used for the analysis and also for isolating the
products in pure form.
Trifluoromethylation of T-2 (90% purity, 10% T-3) with Ruppert-Prakash reagent in DMF
T-2 (1 mmole, 450 mg), trifluoromethyl trimethylsilane (1.1 mmole, 156 mg), potassium bifluoride
(0.3 mmole, 23.4 mg), and DMF (1ml) were combined and vigorously stirred at room temperature
for 1 h. The lower layer was separated by pipetting and analyzed by capillary gas chromatography
(F-2,4-dimethyl-3-isopropyl-2-pentene (P1): 62.7%, F-4,4-dimethyl-3-isopropyl-2-pentene, (2Z):
2.6%, F-2,4,4-trimethyl-3-isopropyl-2-pentene (3): 13.3%, T-2; 7.1%, T-3: 14.2%). Each product
was isolated by using a glass column and subsequently identified.
19F-NMR (CDCl3) of (P1): -56.07 (3F, doublet quartet, J = 58.6, 12.0 Hz), -59.35 (3F, septet
quartet, J = 15.5, 12.0 Hz), -70.11 (6F, multiplet), -70.56 (6F, broad doublet, J = 36.4), -153.6 (1F,
septet doublet, J = 35.7, 12.7 Hz), -157.5 (1F, quartet doublet, J = 58.6, 12.0 Hz).
MS (m/z, %) of (P1) : 481 (M-F, 1.3), 393 (C9F16, 1.9), 343 (C8F13, 2.2), 293 (C7F11, 2.5), 243 (C6F9,
1.7), 181 (C4F7, 1.2), 155 (C5F5, 1.7), 124 (C4F4, 1.0), 119 (C2F5, 1.3), 100 (C2F4, 1.3), 93 (C3F3, 13),
69 (CF3, 100).
19F-NMR ( CDCl3) of (2Z): -60.05 (9F, doublet quartet, J = 26.7, 13.2 Hz), -66.29 (3F, dectet
doublet, J = 11.7, 4.3 Hz), -67.03 (1F, septet doublet quartet, 39.7, 15.5, 4.2 Hz), -72.38 (6F, doublet,
J = 39.5 Hz), -166.3 (1F, dectet doublet, J = 26.7, 15.5 Hz).

12
MS (m/z, %) of (2Z) : 393 (C9F15, 3.6), 343 (C8F13, 2.1), 293 (C7F11, 1.7), 255 (C5F3(CF3)2, 0.7), 243
(C6F9, 1.4), 205 (C5F4CF3, 1.2), 200 (C4F8, 0.9), 181 (C4F7, 1.0), 155 (C5F5, 1.0), 150 (C3F6, 0.6),
143 (C4F5, 1.3), 131 (C3F5, 1.1), 124 (C4F4, 1.2), 119 (C2F5, 2.3), 117 (C5F3, 0.7), 100 (C2F4, 1.1), 93
(C3F3, 1.8), 69 (CF3, 100), 50 (CF2, 0.8).
19F-NMR (CDCl3) of (3): -55.09 (3F, septet, J = 12.9), -56.83 (3F, dectet, J = 15.5 Hz), -57.98 (9F,
broad singlet), -70.40 (6F, broad singlet), -146.99 (1F, dectet, J = 31.7).
MS (m/z, %) of (3) : 443 (C10F17, 1.1), 393 (C9F15, 0.9), 355 (C5F2(CF3)2(C2F5), 0.7), 293 (C7F11,
1.0), 243 (C6F9, 0.6), 205 (C5F4CF3, 1.7), 200 (C4F8, 0.7), 181 (C4F7, 1.8), 155 (C5F5, 1.0), 131 (C3F5,
0.9), 124 (C4F4, 0.7), 117 (C5F3, 0.7), 100 (C2F4, 0.7), 93 (C3F3, 1.1), 69 (CF3, 100), 50 (CF2, 0.7).
Trifluoromethylation of T-3 (99.7% purity, 0.3% T-3) with Ruppert-Prakash reagent in DMF
T-3 (1 mmole, 450 mg), trifluoromethyl trimethylsilane (1.1 mmole, 156 mg), potassium bifluoride
(0.3 mmole, 23.4 mg), and DMF (1ml) were combined and vigorously stirred at room temperature
for 1 h. The lower layer was separated by pipetting and analyzed by capillary gas chromatography
(F-2,4-dimethyl-3-isopropyl-2-pentene (P1): 1.1%, F-(Z)-4,4-dimethyl-3-isopropyl-2-pentene (2Z):
23.3%,
F-(E)-4,4-dimethyl-3-isopropyl-2-pentene
(2E):
17.3%,
F-2,4,4-trimethyl-3-
isopropyl-2-pentene (3): 21.1%, T-2; 10.8%, T-3: 26.4%). 2E was isolated by preparative-scale GC.
19F-NMR analysis of the isolated 2E showed that the sample was comprised of two (2) rotational
isomers, 2E-R1 and 2E-R2 (see the detail in the text) with a peak area ratio of 2 to 1.
19F-NMR (CDCl3) of (2E-R1): -158.2 (1F, dectet doublet, J = 30.3, 6.8 Hz), -71.34 (2CF3, quartet,
J = 15.5 Hz), -68.47 (CF3, septet, J = 15.5 Hz), -67.97 (1F, dectet doublet, J = 30.8, 6.8 Hz), -60.18
(3CF3, doublet doublet, J = 30.8, 30.3 Hz).
19F-NMR (CDCl3) of (2E-R2): -142.6 (1F, quartet doublet, J = 53.8, 10.4 Hz), -68.49 (2CF3, dectet,
J = 8.6 Hz), -65.57 (1F, dectet doublet, J = 29.4 Hz, 10.4 Hz), -64.32 (CF3, doublet, J = 53.8 Hz),
-62.15 (3CF3, doublet septet, J = 29.4, 8.6 Hz).
Acknowledgements
This study was carried out based on the financial support from both AIST and Daikin Industries Co.
Ltd. Hexafluoropropene was kindly supplied by Daikin Industries Co. Ltd.

13
References
1
2
3
K. V. Scherer, Jr., T. Ono, K. Yamanouchi, R. Fernandez, P. Henderson, H. Goldwhite, J. Am.
Chem. Soc., 107 (1985) 718-719.
T. Ono, H. Fukaya, M. Nishida, N. Terasawa, T. Abe, J. Chem. Soc., Chem. Commun. (1996)
1579-1580.
T. Ono, E. Hayashi, M. Nishida, M. Ikeyama, S. Miyagawa, K. Omori, T. Sato, (2009)
JP4387765.
4
5
T. Ono, E. Hayashi, M. Nishida, (2009) JP4352125.
T. Ono, Y. Hayakawa, H. Yokoyama, Proceedings of International Symposium on EcoTopia
Science 2007, ISETS07 (2007) 85-86.
6
7
H. Yokoyama, T. Ono, Appl. Magn. Reson., 33-3 (2008) 197-205.
M. Okazaki, T. Ono, K. Komaguchi, N. Ohta, H. Fukaya, K. Toriyama, Appl. Magn. Reson., 36
(2009) 89-95.
8
9
A. Sato, J. Han, T. Ono, A. Wzorek, J. L. Acena, V. A. Soloshonok, Chem. Commun., 51 (2015)
5967-5970.
Y. R. Patil, T. Ono, B. Ameduri, ACS MacroLetters, 1 (2012) 315-320.
10 F. Boschet, T. Ono, B. Ameduri, Macromol. Rapid Commun., 33 (2012) 302-308.
11 Y. Patil, A. Alaaeddine, T. Ono, B. Ameduri, Macromolecules, 46 (2013) 3092-3106.
12 M. N. Wadekar, Y. R. Patil, B. Ameduri, Macromolecules, 47 (2014) 13-25.
13 G. Puts, G. Lopez, T. Ono, P. Crouse, B. Ameduri, RSC Adv., 5 (2015) 41544-41554.
14 C. M. Friesen, B. Ameduri, Macromolecules, 48 (2015) 7060-7070.
15 T. Ono, K. Ohta, J. Fluorine Chem., 167 (2014) 198-202.
16 S. R. Allayanov, D. A. Gordon, P. B. Henderson, R. E. Fernandez, V. E. Jackson, D. A. Dixon, J.
Fluorine Chem., 180 (2015) 240-247.
17 M. Nishida, H. Fukaya, T. Ono, Eur. J. Org. Chem. (2003) 3648-3658.
18 N. Ishikawa, J. Maruta, J. Synth. Org. Chem., 39 (1981) 51-62.
19 Y. Hayakawa, T. Ono, Eur. Polymer J. (2006) 1357-1361.
20 T. Ono, Y. Hayakawa, “Fluorine-Containing Synthons”, ACS Symposium Series 911, V. A.
Soloshonok, Ed.; American Chemical Society; Washington, DC (2005) Chapter 29, 498-513.
21 T. Ono, Y. Hayakawa, N. Yasuda, H. Uekusa, Y. Ohashi, “Current Fluoroorganic Chemistry.
New Synthetic Directions, Technologies, Materials and Biological Applications”, ACS
Symposium Series 949, V. A. Soloshonok, K. Mikami, T. Yamazaki, J. T. Welch, J. F. Honek,
Eds.; American Chemical Society; Washington, DC (2007) Chapter 10, 170-189.
22 G. K. S. Prakash, P. V. Jog, P. T. B. Batamack, G. A. Olah, Science, 338 (2012) 1324-1327.
23 G. K. S. Prakash, F. Wang, Z. Zhang, R. Haiges, M. Rahm, K. O. Christe, T. Mathew, G. A. Olah,
Angew. Chem. Int. Ed., 53 (2014) 11575-11578.

14
24 N. Santschi, R. Gilmour, Angew. Chem. Int. Ed., 53 (2014) 11414-11415.
25 A. Lishchynskyi, F. M. Miloserdov, E. Martin, J. Benet-Buchholz, E. C. Escudero-Adan, A. I.
Konovalov, V. V. Grushin, Angew. Chem. Int. Ed., 54 (2015) 15289-15293.
26 G. K. S. Prakash, A. K. Yudin, Chem. Rev. 97 (1997) 757-766.
27 G. K. S. Prakash, M. Mandal, J. Fluorine Chem., 112 (2001) 123-131.
28 J. Gawronski, N. Wascinska, J. Gajewy, Chem. Rev., 108 (2008) 5227-5252.
29 X. Liu, C. Xu, M. Wang, Q. Liu, Chem. Rev., 115 (2015) 683-730.
30 L.-L. Chu, F.-L. Qing, Acc. Chem. Res., 47 (2014) 1513-1522.
31 On C8F17SiMe3: W. B. Farnham, In Synthetic Fluorine Chemistry; G. A. Olah, R. D. Chambers,
G. K. S. Prakash, Eds.; Wiley; New York (1992) Chapter 11.
32 On CF2=CFSiMe3: V. A. Petrov, N. V. Shevchenko, V. K. Grinevskaya, Z. N. Parnes, V. V.
Bardin, G. G. Furin, L. S. German, M. E. Vol’pin, Izv. Nauk. SSSR, Ser. Khim, (1990) 2826.
33 On C6F5SiMe3: V. A. Petrov, V. V. Bardin, V. K. Grinevskaya, E. I. Mysov, G. G. Furin, L. S.
German, M. E. Vol’pin, Izv. Nauk. SSSR, Ser. Khim, (1990) 923.
34 M. Nishida, H. Fukaya, Y. Hayakawa, T. Ono, K. Fujii, N. Yasuda, H. Uekusa, Asian J. Org.
Chem., 5 (2016) 927-937.
35 M. Schlosser, D. Michel, Tetrahedron, 52 (1996) 99-108.
36 N. Ishikawa, A. Sekiya, Nippon Kagaku Kaishi, (1972) 2214-2216.
37 W. Brunskill, W. T. Flowers, R. Gregory, R. N. Haszeldine, Chem. Commun. (1970)
1444-1446.
38 F. Cavagna, C. Schumann, J. Magn. Reson., 22 (1976) 333-344.
39 O. Ermer, Angew. Chem. 95 (1983) 1010-1011.
40 W. Dmowsky, W. T. Flowers, R. N. Haszeldine, J. Fluorine Chem., 131 (1977) 1314–1321.

15
Fig.
1.
Structures
of
persistent
perfluoroalkyl
radicals,
PFR-1
(F-3-ethyl-2,4-dimethyl-3-pentyl) and PFR-2 (F-2,4-dimethyl-3-isopropyl-3-pentyl): F-
refers to all unmarked bonds being to fluorine atoms.
Fig. 2. 19F-NMR interpretation for the precursor perfluoroolefin of PFR-2,
F-2,4-dimethyl-3-isopropyl-2-pentene, P1.
Fig. 3. 19F-NMR interpretation for the Z-form of F-4,4-dimethyl-3-isopropyl-2-pentene,
2Z and F-2,4,4-trimethyl-3-isopropyl-2-pentene, 3.
19
Fig. 4. F-NMR interpretation for two rotamers, 2E-R1 and 2E-R2 of the E-form of
F-4,4-dimethyl-3-isopropyl-2-pentene, 2.
Scheme 1. Potential preparative methods to PFR-2 by the direct fluorination of
precursor
olefins,
F-2,4-dimethyl-3-isopropyl-2-pentene
(P1)
and
2-methyl-F-4-methyl-3-isopropyl-2-pentene (P2): all unmarked bonds are to fluorine
atoms.
Scheme 2. Prior work on the reaction of a higher homolog of Ruppert-Prakash reagent
(C8C17SiMe3) with perfluoroolefins.
Scheme 3. Prior art of reactions of T-2 with (a) C-nucleophiles (organometallic reagents
such as MeLi and MeMgX) and with (b) O-nucleophiles (PhOH/Et3N/DMF, then
NaOPh/DMF), and (c) of T-3 with NaOPh/DMF (all unmarked bonds are to fluorine
atoms).
Scheme 4. Present work on the reactions of T-2 and T-3 with Ruppert-Prakash reagent.
Solvent effects on this reaction were also surveyed (all unmarked bonds are to fluorine
atoms).
Scheme 5. The formation of the super-congested perfluoroolefins, P1and 2 having six
(6) trifluoromethyl substituents around a C=C bond and 3 with seven (7) trifluoromethyl
substituents. The main path for the formation of 3 is through 2. F- denotes all unmarked
bonds are to fluorine atoms.

16
Table 1. Trifluoromethylation of T-2 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures.
The reactions were all conducted by using 1 mL of DMF as solvent except Run 6. The
biphasic reaction system comprising an upper phase (CF3SiMe3 + DMF) and a lower
phase (HFP trimers) was vigorously stirred at room temperature for 1 hr. 1) T-2 and T-3
have very close boiling points, so it is very difficult to separate them and thus, the T-2
used for these experiments has T-3 as an impurity. The purity of T-2 is about 90%. 2)
The yields are not isolated yields, but rather the calculated yields based on peak area
ratios from gas chromatography. 3) The reaction was conducted by using 5 mL of DMF,
a volume that was five times larger than that of the standard reaction condition. 4) KF
was used instead of KHF2.
Table 2. Trifluoromethylation of T-3 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures.
All reactions were conducted by using 1 mL of DMF as solvent. The biphasic reaction
system comprising an upper phase (CF3SiMe3 + DMF) and a lower phase (T-3) was
vigorously stirred at room temperature for 1 hr. 1) T-3 used was 99.7% purity (0.3% is
T-2). 2) The yields are not isolated yields, but rather the calculated yields based on peak
area ratios from gas chromatography.
Table 3. Trifluoromethylation of T-2 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures in various solvents.
The standard reaction conditions: The said amount of reagents described in the table are
introduced into 1 mL of solvent and stirred at room temperature for 1 hr. 1) T-2 and T-3
have very close boiling points, so it is very difficult to separate them and thus, the T-2
used for these experiments has T-3 as an impurity. The purity of T-2 is about 90%. 2)
DMA: Dimethylacetamide, NMP: N-methylpyrrolidone, DME: 1,2-Dimethoxyethane,
Diglyme: Diethylene glycol dimethyl ether, Triglyme: Triethylene glycol dimethyl ether,
DMI: 1,3-Dimethylimidazolidinon. 3) The yields are not isolated yields, but rather the
calculated yields based on peak area ratios from gas chromatography. 4) In the case of
CH3CN, the reaction time was extended to 14 h due to the sluggish reaction.

17
Tables
DMF
RT
CF3
CF3
F-
F-
F-
F-
+
+
"cat. F-"
CF3SiMe3
F3C
F3C
3
P1
T-2
2Z
Table 1. Trifluoromethylation of T-2 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures.
cat.F-
Products yields 2)
Recovery
CF3SiMe Tim
T-2 1)
KHF2
(%)
Run
e
3
(mmole)
(mmole P1
)
2Z
3
T-2
T-3
(mmole)
(h)
1.0
1.0
1.0
1.0
2.0
1.1
1.1
1.1
2.0
-
12
1
-
-
-
-
90.6
7.1
6.4
-
9.4
14.2
16.7
-
1
2
3
4
5
0.3
0.3
0.1
2
62.7
60.8
68.9
-
2.6
13.3
13.5
31.1
-
12
1
2.6
-
-
-
-
-
-
-
-
1
71.8
33.7
8.1
-
28.2
66.3
16.1
-
13
1
-
-
6 3)
7
1.0
2.0
2.0
1.0
1.0
1.1
4.0
8.0
1.1
1.1
0.1
0.3
0.1
0.1
0.3
69.3
65.5
74.0
39.3
77.7
6.5
34.5
25.2
-
1
8
1
0.4
51.6
5.0
0.4
9.1
8.0
9 4)
10 4)
1
1
9.3
The reactions were all conducted by using 1 mL of DMF as solvent except Run 6. The
biphasic reaction system comprising an upper phase (CF3SiMe3 + DMF) and a lower phase
(HFP trimers) was vigorously stirred at room temperature for 1 hr. 1) T-2 and T-3 have very
close boiling points, so it is very difficult to separate them and thus, the T-2 used for these
experiments has T-3 as an impurity. The purity of T-2 is about 90%. 2) The yields are not
isolated yields, but rather the calculated yields based on peak area ratios from gas
chromatography. 3) The reaction was conducted by using 5 mL of DMF, a volume that was
five times larger than that of the standard reaction condition. 4) KF was used instead of
KHF2.

18
DMF
RT
CF3
CF3
F-
F-
F-
F-
+
+
"cat. F-"
CF3SiMe3
F3C
F3C
T-3
P1
3
2Z, 2E
Table 2. Trifluoromethylation of T-3 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures.
cat.F-
Products yields 2)
Recovery 2)
T-3 1)
(mmole) (mmole)
CF3SiMe3 Time
KHF2
(%)
Run
(h)
(mmole) P1
2Z
2E
3
T-2
T-3
1.0
1.0
1.0
1.1
2.0
4.0
+1
-
1
1
1
2
1
0.3
0.15
0.1
1.1 23.3 17.3 21.1 10.8 26.4
1
2
3
1.0 21.7 14.0 59.7 1.1
2.5
5.3
2.0
1.0 34.5 7.2 53.0
0.4 26.8 5.1 65.7
-
-
0.1
4
1.0
0.3
-
-
-
-
12.0 88.0
All reactions were conducted by using 1 mL of DMF as solvent. The biphasic reaction
system comprising an upper phase (CF3SiMe3 + DMF) and a lower phase (T-3) was
vigorously stirred at room temperature for 1 hr. 1) T-3 used was 99.7 % purity (0.3 % is T-2).
2) The yields are not isolated yields, but rather the calculated yields based on peak area
ratios from gas chromatography.

19
DMF
RT
CF3
CF3
F-
F-
F-
F-
+
+
"cat. F-"
CF3SiMe3
F3C
F3C
3
P1
T-2
2Z
Table 3. Trifluoromethylation of T-2 with Ruppert-Prakash reagent to produce highly
congested perfluoroolefins with super-congested structures in various solvents.
cat.F-
KHF2
(mmole)
0.3
Products yields 3)
Recovery
T-2 1)
CF3SiMe3 Solvent 2)
(%)
Run
(mmole) (mmole)
P1
2Z
3
T-2 T-3
1.0
1.0
1.0
1.0
1.1
1.1
1.1
2.0
DMF
DMA
NMP
62.7
61.2
61.8
38.5
2.7 13.3 7.1 14.2
0.6 13.7 8.0 16.5
1
2
3
4
0.3
0.3
1.6
1.0
11.0 8.0 17.6
11.0 0.9 48.6
CH3CN
0.3
4)
5
6
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.0
2.0
1.1
2.0
1.1
3.0
4.0
DME
Diglyme
Triglyme
Triglyme
DMI
0.1
0.1
0.1
0.1
0.1
0.1
0.1
-
-
-
-
-
-
-
-
-
-
90.6 9.4
90.6 9.4
-
7
51.2
80.1
67.2
86.2
88.4
1.1 40.6 7.1
6.9 9.4 3.6
8
9
-
-
-
11.1 21.6
9.9 3.9
7.9 3.7
10
11
DMI
DMI
The standard reaction conditions: The said amount of reagents described in the table are
introduced into 1 mL of solvent and stirred at room temperature for 1 hr. 1) T-2 and T-3 have
very close boiling points, so it is very difficult to separate them and thus, the T-2 used for
these experiments has T-3 as an impurity. The purity of T-2 is about 90%. 2) DMA:
Dimethylacetoamide, NMP: N-methylpyrrolidone, DME: 1,2-Dimethoxyethane, Diglyme:
Diethylene glycol dimethyl ether, Triglyme: Triehylene glycol dimethyl ether, DMI:
1,3-Dimethylimidazolidinon. 3) The yields are not isolated yields, but rather the calculated
yields based on peak area ratios from gas chromatography. 4) In the case of CH3CN, the
reaction time was extended to 14 h due to the sluggish reaction.