Multiple pentafluorophenylation of 2,2,3,3,5,6,6-heptafluoro-3,6-dihydro- 2H-1,4-oxazine with an organosilicon reagent: NMR and DFT structural analysis of oligo(perfluoroaryl) compounds

Article abstract of DOI:10.1002/hlca.200690239

The reagent Me₃Si(C₆F₅) was used for the preparation of a series of perfluorinated, pentafluorophenyl-substituted 3,6-dihydro-2H-1,4-oxazines (2-8), which, otherwise, would be very difficult to synthesize. Multiple pentafl

Full text of DOI:10.1002/hlca.200690239

Helvetica Chimica Acta – Vol. 89 (2006)
2671
Multiple Pentafluorophenylation of 2,2,3,3,5,6,6-Heptafluoro-3,6-dihydro-2H-
1,4-oxazine with an Organosilicon Reagent: NMR and DFT Structural
Analysis of Oligo(perfluoroaryl) Compounds
by Masakazu Nishida*a), Haruhiko Fukayaa), Yoshio Hayakawaa), Taizo Onoa), Kotaro Fujiib), and
Hidehiro Uekusab)
a) National Institute of Advanced Industrial Science and Technology (AIST), 2266-98, Shimoshidami,
Moriyama-ku, Nagoya, 463-8560 Japan
(phone: +81-52-736-7329; fax: +81-52-736-7304; e-mail: m-nishida@aist.go.jp)
b) Department of Chemistry and Material Science, Tokyo Institute of Technology, Ookayama 2, Meguro-
ku, Tokyo, 152-8551 Japan
The reagent Me3Si(C6F5) was used for the preparation of a series of perfluorinated, pentafluoro-
G
phenyl-substituted 3,6-dihydro-2H-1,4-oxazines (2–8), which, otherwise, would be very difficult to syn-
thesize. Multiple pentafluorophenylation occurred not only on the heterocyclic ring of the starting com-
pound 1 (Scheme), but also in para position of the introduced C6F5 substituent(s) leading to compounds
with one to three nonafluorobiphenyl (C12F9) substituents. While the tris(pentafluorophenyl)-substituted
compound 3 could be isolated as the sole product by stoichiometric control of the reagent, the higher-sub-
stituted compounds 5–8 could only be obtained as mixtures. The structures of the oligo(perfluoroaryl)
compounds were confirmed by 19F- and 13C-NMR, MS, and/or X-ray crystallography. DFT simulations
of the 19F- and 13C-NMR chemical shifts were performed at the B3LYP-GIAO/6-31++G(d,p) level for
geometries optimized by the B3LYP/6-31G(d) level, a technique that proved to be very useful to accom-
plish full NMR assignment of these complex products.
Introduction. – Perfluorinated compounds are being used as a wide variety of engi-
neering polymers and surfactants on an industrial scale because of their unique thermal
and chemical properties. Especially, perfluoroaryl compounds show several unique
properties such as high electron affinity or a rather sterically demanding and p-elec-
tron-accepting nature compared with the corresponding non-fluorinated aryl systems.
Several reports showed that strong Lewis acidity of a perfluoroaryl ring is efficient
for catalytic reactions: perfluoroarylboranes as co-catalyst for the olefin polymeriza-
tion with group-4 metallocene alkyls [1], bifunctional perfluoroarylboranes for the acti-
vation of basic substrates, and the selective binding of anions [2]. For the purpose of
application of asymmetric catalysis, chiral bidentate (perfluoroaryl)phosphane ligands
were synthesized [3]. Polymers containing perfluorophenylene moieties at the main
chain are expected to be used in high-performance thermoplastics [4–6] and optical-
waveguide devices [7][8]. Recently, tetrakis(pentafluorophenyl)porphyrin was
reported to be applicable in various fields: imine aziridination [9], hydrogen peroxide
induced oxidation [10], phenylethyne-linked porphyrin dyads [11], and as matrices for
matrix-assisted laser-desorption-ionization (MALDI) mass spectrometry [12].
Fluorinated organosilicon compounds have been used as transfer reagents of fluo-
rinated substituents for preparing synthetic intermediates in the areas of agrochemicals,
ꢀ 2006 Verlag Helvetica Chimica Acta AG, Zürich

2672
Helvetica Chimica Acta – Vol. 89 (2006)
pharmaceuticals, etc. [13]. Trimethyl(trifluoromethyl)silane, Me3SiCF3, is the most
effective transfer reagent of a CF3 moiety for various organic molecules, especially
aldehydes and ketones [14]. A recent report has described the trifluoromethylation
of less-reactive carboxylates, thiocyanates, and selenocyanates [15]. An aromatic ana-
logue, Me3
[16], and several reports were published on the pentafluorophenylation of organic mol-
ecules by Me3Si(C6F5), e.g., perfluoro-olefins [17], perfluoro-aza-alkenes [18][19], or
A
G
N)3
G
AHCTREUNG
perfluorinated carbonyl compounds [20].
In a preliminary note [21], we described the reactions of Me3Si(C6F5) with highly
A
electronegative and bulky substrates, focusing on a different reactivity between per-
fluorinated 3,4-dihydro-2H-pyrrole and perfluorinated 3,6-dihydro-2H-1,4-oxazine
(1; see the Scheme below). This former study indicated that in the reaction of
Me3Si(C6F5) the pyrrole predominantly provides dimerized products, while the reaction
G
of the oxazine 1 affords C6F5-substituted derivatives without dimerization [21]. In the
substitution reaction of perfluorinated aliphatic imines, such as perfluoro-aza-alkenes,
oligomerization competes and a complex mixture is obtained. Differing from such per-
fluorinated imine systems, the suppression of the competing dimerization or oligome-
rization reaction found in the six-membered compound 1 prompted us to investigate in
detail multiple pentafluorophenylations with Me3Si(C6F5), which led to a series of inter-
G
esting perfluorinated products (Scheme) characterized spectroscopically and, in part,
also by density-functional-theory (DFT) calculations.
Results and Discussion. – 1. Synthesis. Our approach towards the synthesis of new
oligo- and poly(perfluoroaryl) materials was based on the reactions of compounds 1–3
with Me
ACHTREUNG
catalyst, is sensitive to moisture, all reactions with Me3
AHCTREUNG
anhydrous conditions in an inert atmosphere. Although the reactions proceeded well in
MeCN, tetraglyme, and other aprotic polar solvents, MeCN was the solvent of choice
since it allowed convenient product separation. The reactions were typically started
at 08 for 1 h, and then run at ambient temperature for 20 h. The results are summarized
in Table 1.
When equimolar amounts of Me3
mono-substituted compound 2 was obtained as the main product, accompanied by
small amounts of the bis-adduct 4 and the tris-adduct 3. When an excess of Me3Si(C6F5)
A
AHCTREUNG
was used, compound 3 was obtained in moderate-to-good yield, while 4 was still iso-
lated in low yield (Entries 2 and 3). To attempt selective preparations of compounds
3 and 4, reaction of the isolated mono-adduct 2 with Me3Si(C6F5) was next examined.
G
When used in equimolar amounts (Entry 4), compound 4 was obtained in somewhat
higher yield (16%; 28% based on consumed 2). However, adduct 3 was present in
the mixture in almost the same proportion (24%; 42% based on consumed 2). Fortu-
nately, the tris-adduct 3 was obtained as the sole product in 78% yield when using an
excess of the reagent (Entry 5).
The pentafluorophenylation by Me3Si(C6F5) is considered to proceed via an addi-
G
tion–elimination mechanism (AdN-E): attack of (C6F5)À at the C=N bond and succes-
sive elimination of FÀ. This mechanism works only when the C-atoms are directly
bonded to the N-atoms so that the nucleophiles are only introduced into the neighbor-

Helvetica Chimica Acta – Vol. 89 (2006)
2673
Scheme
i) Me3
N
G

2674
Helvetica Chimica Acta – Vol. 89 (2006)
Table 1. Reactions of 1–3 with Me3Si(C6F5). Conditions: at 08 for 1 h, then at r.t. for 20 h, unless noted
otherwise.
Entry
Compound
([mmol])
Me3
[mmol]
Si(C6F5)
KF
[mmol]
Productsa)
1
2
3
4
5
6
7b)
1 (3.3)
1 (1.1)
1 (3.2)
2 (0.44)
2 (0.41)
3 (0.30)
3 (0.30)
3.3
2.3
9.7
0.44
0.82
0.30
0.97
0.75
0.21
0.64
0.085
0.085
0.11
0.11
2 (41), 3 (7.0), 4 (6.4)
2 (29), 3 (52), 4 (3.3)
2 (7.5), 3 (64)
2 (43), 3 (24), 4 (16)
2 (4.0), 3 (78)
3 (72), 5 (4.5), 6 (3.3)
3 (5.6), 5 (8.3), 6 (3.5), 7a (7.8), 7b (3.0), 8 (0.6)c)
a) Yields are given in % (in parentheses) and were determined by 19F-NMR integration. b) At 508 for 48h.
c) Together with a series of dimerization products (see Fig. 2 and text).
ing C-atoms which is in contrast with the AdN-E reaction of perfluoro-olefin C=C sys-
tems, where multiple substitutions by nucleophiles around the perfluorocarbon skele-
tons are observed due to the migration of the C=C bonds, which leads to complex prod-
uct mixtures [22][23].
Because two electron-withdrawing C6F5 groups of the bis-adduct 4 enhance the
reactivity of the N=CÀF unit, this compound easily decomposes upon column chroma-
tography on silica gel (SiO2) or by Kugelrohr distillation, which explains the low yields
of 4 (Table 1). However, the tris-adduct 3 was found to be less reactive compared to 1, 2,
and 4, providing no tetrasubstituted anion (Scheme, Route A). This result is in strong
contrast with the Ruppert–Prakash reagent, Me3Si(CF3), which was previously reported
G
to give perfluorinated 2,2,6,6-tetramethylpiperidine in a similar fluorinated 1-azacyclo-
hexene system [24]. One more contrasting difference is expected for the reagent
Me3Si(C6F5) because of the possibility to further react at the para position of the intro-
G
duced pentafluorophenyl aromatics (Routes B and C). Thus, the reactivity of isolated 3
was further examined.
When compound 3 was reacted with 1 equiv. of Me3Si(C6F5), small amounts of the
G
expected perfluorinated biphenyl compounds 5 and 6 were obtained (Table 1, Entry 6).
When the reaction was conducted under more-severe conditions (508, 48h) with ca. 3
equiv. of the reagent, we were even able to isolate, apart from 5 and 6, the perfluori-
nated bis-biphenyl and tris-biphenyl analogues 7a,b and 8, respectively (Table 1,
Entry 7).
From a mechanistic point of view, it is noteworthy to point out two interesting fea-
tures of the nucleophilic attack of a (C6F5)À anion on the already introduced C6F5
ring(s): 1) the nucleophilic attack hardly occurred in the reactions with 2 and 4, but
in the reactions with the more-bulky compounds having at least three C6F5 groups,
i.e., compounds 3 and 5–7. This means that the N=CÀF unit is more reactive than
the para position of the C6F5 group. 2) The nucleophilic attack occurred only in para
position of the C6F5 ring directly bonded to the heterocycles, but not on the C6F5
ring of the perfluorinated biphenyl group. Hence, para substitution stopped at the
biphenyl stage, and any further extension of phenylene rings of the type C6F5(C6F4)n,
with n>1, did not occur. It is very surprising that such a subtle structural change on

Helvetica Chimica Acta – Vol. 89 (2006)
2675
the C6F5-ring environment leads to such a profound change in the reactivity of the para
position of the C6F5 ring in the AdN-E mechanism. Recently, the formation of 1-H-per-
fluoro-4,4’-polyphenyls (1-H(C6F4)nF) was reported by nucleophilic attack at the para
position of a C6F5 ring in the reaction of Me3
N
N
details were reported regarding compound characterization.
2. Compound Characterization. Compounds 2–8 were identified by 19F- and 13C-
NMR, DI-MS, and MALDI-TOF-MS. Adjacent F-atom peaks were confirmed by
the 19F{19F} homo-decoupling method. The correlations between 19F and 13C peaks
were determined by either 19F-detected 19F,13C-HMQC spectroscopy or partially
13C{19F} hetero-decoupling techniques with selective 19F decoupling. As an example,
the 19F,13C-HMQC spectrum of 3 is displayed in Fig. 1.
The material balance was very low in the reaction leading to the higher-substitution
products (Table 1, Entry 7) due to a considerable amount of residual materials formed.
Therefore, the residue was further analyzed by 19F-NMR and MALDI-TOF-MS. No
3,3,5,5-tetrasubstituted pentafluorophenyl derivative 9 was detected, though. However,
the MALDI-TOF experiments indicated the formation of some intriguing dimeric
products, consistent with a series of signals at m/z 422+148n (n=6–11), as shown in
Fig. 2. Since the separation of these dimers was very difficult, the underlying mecha-
nism is not clear yet. This dimerization reaction of the six-membered ring, however,
is a competitive reaction for the multiple pentafluorophenylation, and it could be
Fig. 1. 19F-Detected 19F,13C-HMQC Spectrum of Compound 3. The F2 and F1 axes refer to dF and dC,
resp.

2676
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 2. MALDI-TOF Mass spectrum of inseparable dimeric side products
one of the reasons why no perfluorinated terphenyl derivatives were obtained under
the more-severe reaction conditions.
The molecular structure of the tris-adduct 3 was unequivocally determined by X-ray
crystallography (Fig. 3). Selected bond lengths, bond angles, and dihedral angles are
listed in Table 2. For crystal data and structure-refinement details, see Table 4 in the
Exper. Part. In the asymmetric unit of the crystal, two molecules are independent. How-
ever, they can be fitted almost completely (RMS fit of 0.222 Š) by noncrystallographic
rotation–inversion of 166.388. Regarding the two geminal C6F5 substituents at C(12)1),
the distances C(1)ÀC(12) (1.533(2) Š) and C(12)ÀC(22) (1.543(2) Š), each between
an sp2 and an sp3 atom, are prolonged to an equivalent length of the C(sp3)ÀC(sp3)
bond C(12)ÀC(17) (1.542(2) Š), comparable to the molecular structure of
(C6F5)CH2OC(CF3)(C6F5)2 (CÀC 1.538(8) Š) [20]. Bond-angle distortions were also
observed around the hetero-atoms: C(15)ÀO(16)ÀC(17) 114.79(12)8, N(13)ÀC(14)À
C(15) 125.35(13)8, C(12)ÀN(13)ÀC(14) 118.60(12)8.
From a supramolecular point of view, compound 3 has a very interesting cap-like
structure. The selection of the C6F5 group for a supramolecular-recognition site is
based on the recent surge of reports on p–p stacking [26–28]. Further, we also expect
interesting CÀF···HÀC interactions when using molecular systems based on several
C6F5 groups in the vicinity of a nitrogen center, rather than in a scattered manner on
the periphery of a carbon skeleton.
1
)
Arbitrary atom numbering according to Fig. 3.

Helvetica Chimica Acta – Vol. 89 (2006)
2677
Fig. 3. X-Ray crystal structure of 3. Thermal ellipsoids are shown at the 50%-probability level.
Table 2. Selected Bond Lengths (Š), Bond Angles (8), and Torsion Angles (8) in the X-Ray Crystal
Structure of 31)
O(16)ÀC(15)
N(13)ÀC(14)
C(14)ÀC(15)
C(17)ÀC(12)
C(1)ÀC(12)
1.3753(18)
1.2616(19)
1.520(2)
1.542(2)
1.533(2)
O(16)ÀC(17)
N(13)ÀC(12)
C(14)ÀC(33)
C(12)ÀC(22)
1.3762(19)
1.4867(18)
1.496(2)
1.543(2)
C(15)ÀO(16)ÀC(17)
N(13)ÀC(14)ÀC(15)
N(13)ÀC(12)ÀC(22)
O(16)ÀC(15)ÀC(14)
N(13)ÀC(12)ÀC(17)
C(17)ÀC(12)ÀC(1)
114.79(12)
125.35(13)
102.68(11)
114.78(12)
109.19(12)
106.34(12)
C(14)ÀN(13)ÀC(12)
N(13)ÀC(14)ÀC(33)
N(13)ÀC(12)ÀC(1)
O(16)ÀC(17)ÀC(12)
C(17)ÀC(12)ÀC(22)
C(22)ÀC(12)ÀC(1)
118.60(12)
119.67(13)
109.93(12)
113.60(12)
117.01(12)
111.59(12)
N(13)ÀC(14)ÀC(15)ÀO(16)
N(13)ÀC(14)ÀC(33)ÀC(34)
N(13)ÀC(12)ÀC(1)ÀC(6)
À0.4(2)
N(13)ÀC(14)ÀC(33)ÀC(38)
À88.51(19)
92.80(18)
158.18(14)
N(13)ÀC(12)ÀC(22)ÀC(23)
62.02(16)
The end-cap structure of 3 most-likely also prevented further nucleophilic attack on
the C=N bond, which rationalizes why no tetra-adduct 9 was formed. This is also the
reason why additional pentafluorophenylation takes place at the para position of exist-
ing C6F5 substituent(s), which, in turn, gave rise to even deeper cap structures for 5–8.

2678
Helvetica Chimica Acta – Vol. 89 (2006)
3. DFT Calculations. Although NMR analysis is the most powerful tool for the iden-
tification of perfluorinated compounds, an assignment of each peak was difficult with
our compounds, not only in 19F-NMR, but also in 13C-NMR, especially in the case of
adducts with several C6F5 rings. We reported earlier that DFT calculations of various
fluoro compounds, when performed at the B3LYP level using the gauge-independent
atomic orbital (GIAO) level with the 6-31++G(d,q) basis set, is quite useful for the
estimation of chemical shifts [29] and also for conformational analysis of highly steri-
cally hindered polyfluoro-olefins [22]. With these powerful theoretical arsenals in
our hands, we, thus, confirmed the structures of compounds 2–8.
Optimized geometries of 2–4 were obtained by starting from modified structures
based on the X-ray crystal structure of 3 by removing or adding C6F5 rings at the
B3LYP/6-31G(d) level. On the basis of the geometries so determined, 19F- and 13C-
NMR shieldings were calculated at the B3LYP-GIAO/6-31++G(d,p) level (Table 3).
Note that, in the following, the atomic numbers are defined according to the IUPAC
priority rules. For compounds 2–4, the calculated (calc.) 19F-NMR shieldings showed
a very good relationship with experimental (exper.) values, r2 being 0.998( 2), 0.997
(3), and 0.994 (4) (Fig. 4). The differences between the experimentally determined
and the calculated values were larger for the C4F4NO ring (Dd(F) À7.58to À11.43)
than for the C6F5 rings (Dd(F) À2.95 to À5.98), except for the 3-F-atom (Dd(F)
À1.36) of the C4F4NO rings (FÀC=N)2).
The calculated and experimental 13C-NMR shifts of 2–4 also showed a very good
correlation, with r2 values of 0.991 (2), 0.995 (3), and 0.995 (4) (data not shown graphi-
cally). In contrast with the above 19F-NMR data, the Dd(C) values were smaller for the
C4F4NO ring (Dd(C) À2.99 to À4.44) than for the C6F5 rings (Dd(C) À7.14 to À10.15)
and for the C=N group of the C4F4NO ring (Dd(C) À9.70 to À11.37). In cases of close-
lying peaks, especially in the 13C-NMR experiments, reversals of signal orders between
calculated and experimental values were observed. Accordingly, for the compounds
having close-lying peaks, NMR prediction was not perfect, and, thus, the data had to
be carefully examined. In general, the calculated values were positively shifted for
19F-NMR shifts, while they were negatively shifted for 13C-NMR shifts due to the p-
electron effect, as observed before for highly branched fluoro-olefins [22].
The more-complicated structures 5–8 were also successfully corroborated by
means of 19F- and 13C-NMR analyses of experimental and DFT-calculated shielding
data. The geometries of these compounds were determined at the B3LYP/6-31G(d)
level, and the 19F- and 13C-NMR shifts were calculated from these geometries at the
B3LYP-GIAO/6-31++G(d,p) level. The experimentally determined 19F- and 13C-
NMR chemical shifts of 5–8 are given in the Exper. Part3). It should be useful to
pointed out some trends: 1) the absorption signals by o-F-atoms (2,6-F) of the 3-C6F5
moiety, i.e., the C6F5 group at the 3-position of the 1,4-oxazine ring2) appeared at ca.
d(F) À140, while those of the 5-C6F5 group appeared at ca. d(F) À135. 2) The p-F-
atom (4-F) of the 3-C6F5 group appeared at ca. d(F) À148, while that of the 5-C6F5
group resonated at ca. d(F) À149 ppm. 3) the m-F-atoms (3,5-F) of the 3- and 5-C6F5
2
)
)
Non-systematic C-atom numbering as indicated in formula 3 (Scheme).
Supplementary data regarding the calculated NMR chemical shifts of 5–8 can be obtained from the
corresponding author (M. N.).
3

Helvetica Chimica Acta – Vol. 89 (2006)
2679
Table 3. Experimentally Observed vs. Calculated 19F- and 13C-NMR Chemical Shifts (in ppm) of 2–4. For
details, see text and Exper. Part.
Group/Atom2)
2
3
4
obs.
calc.
obs.
calc.
obs.
calc.
C4
FnNOa)
A
2-F
3-F
5-F
6-F
3-C
À72.31
À83.12
À71.78
À82.07
À75.80
À84.43
À62.81
À78.45
À64.20
À88.94
À99.76
À89.68
À107.34
À99.11
À76.32
À87.75
A
2,6-F
3,5-F
4-F
À138.65
À158.68
À145.94
À142.11
À164.53
À151.85
À140.01
À159.42
À147.92
À142.69
À165.17
À153.26
5,5-(C6F5)2
2,6-F
3,5-F
4-F
À135.55
À159.49
À149.27
À139.01
À165.45
À154.81
À135.83
À159.46
À149.26
À139.18
À165.39
À154.59
C
4FnNOa)
G
2-C
3-C
5-C
6-C
3-C
1-C
2,6-C
3,5-C
4-C
112.20
158.23
107.27
113.57
108.57
148.33
102.85
109.45
113.25
149.78
66.36
109.54
140.08
63.38
109.72
149.33
66.34
106.63
137.96
63.31
119.03
115.02
119.03
115.02
6F5
105.16
144.72
137.96
144.04
97.14
134.59
128.20
134.32
106.20
145.03
137.86
143.52
97.53
135.16
127.94
134.04
5,5-(C6F5)2
1-C
2,6-C
3,5-C
4-C
110.18102.72
144.93
138.30
110.17
103.03
135.41
128.80
144.96
138.33
134.81
128.74
1422.38132.8 142.68132.78
a) For 2, 3, and 4, n=6, 4, and 5, resp. (see Scheme).
groups appeared at ca. d(F) À159. 4) The signals of the C12F9 group resonated at ca.
d(F) À135 (2,6-F); À137 (2’,6’-F); À160 (3’,5’-F), and at À149 (4’-F), while the 3,5-F
signals appeared at ca. d(F) À139 and À135 in the 3- and 5-C12F9 moieties, respectively.
The calculated 19F-NMR shifts nicely correlated with the experimental values, the cor-
relation coefficient being very high (0.995<r2 <0.998). When looking at the overall
correlation for the optimized geometries of 2–8, an overall correlation coefficient r2
of 0.995 was obtained (Fig. 4). Thus, the combined use of DFT-based and experimental
19F-NMR experiments is very reliable in assigning pentafluorophenyl (C6F5) and non-
afluorobiphenyl (C12F9) derivatives.
The following trends were observed in the experimental 13C-NMR spectra of 5–8:
the signals for 3,5-C, 2,6-C, 4-C, 2’,6’-C, 3’,5’-C, and 4’-C appeared at d(C) 137–145 and
those for 4-C, 1-C, and 1’-C appeared at d(C) 101–117. The corresponding calculated
values, derived at the B3LYP-GIAO/6-31++G(d,p) level, also showed good correla-

2680
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 4. Correlation between the experimentally determined and calculated 19F-NMR shieldings of 2–8
relative to CFCl3 as internal reference. A similar correlation was obtained for the corresponding 13C-
NMR data (not shown).
tions (0.990<r2 <0.995). Thereby, the Dd values showed some scatterings with regard
to the atoms 1-C and 1’-C connecting two C6F5 rings. Nevertheless, the overall correla-
tion coefficient for the 13C-NMR data of 2–8 was still high (r2 =0.992).
Conclusions. – The pentafluorophenylation of the perfluorinated 1,4-oxazine 1 with
Me
adduct 3, almost selectively obtained with an excess of Me3
ing cap structure, as confirmed by X-ray crystal-structure analysis. Although the forma-
tion of the tetrasubstituted compound 9 is conceivable by reacting 3 with Me3Si(C6F5),
ACHTREUNG
AHCTREUNG
AHCTREUNG
further pentafluorophenylation occurred only at the para position of the already intro-
duced C6F5 rings due to the bulky cap structure of 3, preventing nucleophilic attack at
the C=N bond of the 1,4-oxazine ring. Accordingly, five nonafluorobiphenyl-substi-
tuted compounds, 5–8, were obtained by multiple substitutions of the p-F-atoms of
the C6F5 groups.
The structures of 2–8 were determined by 19F- and 13C-NMR, MS, and/or X-ray
crystallography. Their assignments were confirmed by comparison of DFT-calculated
(B3LYP-GIAO/6-31++G(d,p)) vs. experimental 19F- and 13C-NMR chemical shifts.
Since the fundamentals for the structural analysis of such perfluoroaryl systems was estab-
lished in this paper, a further elongation of perfluoroaryl system will be targeted next.
We would like to thank Dr. Yutaka Tai (National Institute of Advanced Industrial Science & Tech-
nology) for providing access to the TOF-MS facility.

Helvetica Chimica Acta – Vol. 89 (2006)
2681
Experimental Part
General. The starting material 1 (=2,2,3,3,5,6,6-heptafluoro-3,6-dihydro-2H-1,4-oxazine) was pre-
pared by pyrolysis of potassium perfluoromorpholinoacetate [30] and purified by repeated trap-to-trap
distillation (final purity: 95%). Trimethyl(pentafluorophenyl)silane, Me
N
N
tion of bromopentafluorobenzene with chlorotrimethylsilane and phosphorus tris(diethylamide) [19]
using tetraglyme as a solvent (instead of benzonitrile). Spray-dried KF was purchased from Wako Pure
Chemical Industrial, Inc. All solvents were dried over 4-Š molecular sieves degassed by freeze-thaw
cycles. All reactions were carried out under Ar atmosphere and under anh. conditions. Gases and volatile
liquids were handled in a conventional Pyrex-glass vacuum system equipped with a Heise–Bourdon tube
gauge and a Televac thermocouple gauge. Gas chromatographic (GC) analyses were performed on a Shi-
madzuGC-17A instrument with a NEUTA BOND-1 column (60 m”0.25 mm; 1.5 mm; GL Science).
NMR Spectra were recorded on a Varian Unity-Inova-300 apparatus in CDCl3 in a 5-mm NMR tube
at 75.42 (13C) and 282.24 MHz (19F); d values rel. to CFCl3 as internal reference. The 13C-NMR spectra
were measured using WURST modulation for complete fluorine decoupling, and using continuous wave
for selective fluorine decoupling, resp. 19F-detected 19F,13C-HMQC spectra were recorded at 282.24 MHz
with full 13C decoupling using WALTZ-16 modulation and a relaxation delay of 2.4 s. GC/MS data were
obtained with a ShimadzuQP-5000 quadrupole mass spectrometer by electron-impact (EI) ionization at
70 eV on the GC column described above. Direct-inlet mass spectrometric (DI-MS) data were obtained
on a ShimadzuQP-1100EX quadrupole mass spectrometer in EI mode at 70 eV. MALDI-TOF-MS Data
were obtained on a Bruker Daltonics AutoFLEX time-of-flight (TOF) mass spectrometer operated
under matrix-assisted laser-desorption-ionization (MALDI) conditions (matrix: ꢁtrans-3-indole-acrylic
acidꢂ).
Reaction of 1 with an Equimolar Amount of Me3Si(C6F5). Spray-dried KF (39 mg, 0.42 mmol) in a
100-ml reaction vessel was dried at 80–908 under vacuum, and then anh. MeCN (5 ml) was added
using the vacuum line. The reaction vessel was cooled to À788, and Me
A
was added under Ar atmosphere. The mixture was then cooled to À1968 and treated with 1 (0.439 g,
2.10 mmol) using the vacuum-transfer method. The mixture was then stirred at 08 for 1 h, and then at
r.t. for 20 h. The volatile products were removed in vacuum at 08. Product 2 was obtained by trap-to-
trap distillation (508/1 mmHg) in 36% isolated yield. The distilling residue was extracted with CHCl3
(5 ml), and evaporation of the extract gave a mixture of 2 (41%), 3 (7%), and 4 (6.4%), as determined
by 19F-NMR. The sample for the identification of 2 was obtained by Kugelrohr distillation.
Data of 2,2,3,3,6,6-Hexafluoro-5-(pentafluorophenyl)-3,6-dihydro-2H-1,4-oxazine (2). 19F-NMR
(CDCl3)2): À72.31 (tm, J=12.7, 2-F of C4F6NO); À89.68 (tt, J=5.9, 5.6, 6-F of C4F6NO); À99.76 (br.
t, J=5.6, 5-F of C4F6NO); À138.65 (m, J=14.4, 12.7, 4.0, 2,6-F of C6F5); À145.94 (tt, J=21.5, 4.0, 4-F
of C6F5); À158.68 (m, J=21.5, 14.4, 3,5-F of C6F5). 13C-NMR (CDCl3)2): 105.16 (1-C of 3-C6F5); 107.27
(5-C of C4F6NO); 112.20 (2-C of C4F6NO); 113.57 (6-C of C4F6NO); 137.96 (3,5-C of 3-C6F5); 144.04
(4-C of 3-C6F5); 144.72 (2,6-C of 3-C6F5); 158.23 (3-C of C4F6NO). GC/EI-MS: 359 (9, M+), 193 (42),
124 (30), 100 (100), 93 (6), 69 (16).
Data of 3. See below.
Data of 2,2,5,6,6-Pentafluoro-3,3-bis(pentafluorophenyl)-3,6-dihydro-2H-1,4-oxazine (4). This com-
pound could not be isolated in pure form, neither by column chromatography (CC) on silica gel nor
by Kugelrohr distillation, because it easily decomposed during isolation. 19F-NMR (CDCl3)2): À62.81
(t, J=23.4, 3-F of C4F5NO); À75.80 (br. s, 2-F of C4F5NO); À78.45 (br. s, 6-F of C4F5NO); À135.83
(m, J=15.5, 2,6-F of C6F5); À149.26 (tt, J=21.6, 5.6, 4-F of C6F5); À159.46 (m, J=21.6, 15.6, 3,5-F of
C6F5). 13C-NMR (CDCl3)2): 66.34 (5-C of C4F5NO); 109.72 (2-C of C4F5NO); 110.17 (1-C of 5,5-
(C6F5)2); 119.03 (6-C of C4F5NO); 138.33 (3,5-C of 5,5-(C6F5)2); 142.68(4-C of 5,5-(C6F5)2); 144.96 (2,
6-C of 5,5-(C6F5)2); 149.33 (3-C of C4F5NO). GC/EI-MS: 507 (41, M+), 441 (24, [MÀCOF2]+), 422
(100, [MÀCOF3]+), 377 (17), 372 (39), 358(11), 346 (41), 327 (77), 296 (31), 274 (38), 258(12), 248
(27), 229 (11), 227 (11), 224 (28), 217 (77), 198 (42), 179 (42), 148 (42), 117 (26), 93 (25), 81 (15), 69 (37).
Reaction of 2 with 2 Equiv. of Me3Si(C6F5). Spray-dried KF (8mg, 0.085 mmol) in a 100-ml reaction
vessel was dried at 80–908 under vacuum, and then treated with anh. MeCN (1 ml) using the vacuum line.
The vessel was cooled at À788, and 2 (0.169 g, 0.41 mmol) was added under Ar atmosphere. The mixture

2682
Helvetica Chimica Acta – Vol. 89 (2006)
Si(C6H5) (0.198g, 0.82 mmol) via vacuum transfer. The
was then cooled to À1968, and treated with Me3
AHCTREUNG
mixture was stirred at 08 for 1 h, and then at r.t. for 20 h. The volatile products were removed in vacuum
from the reaction mixture at 508, and the distilling residue was extracted with CHCl3 (5 ml). Evaporation
of the extract gave a light-yellow solid, a mixture of 2 (4%) and 3 (78%), which was purified by CC (SiO2;
hexane) to afford pure 3 in 49% isolated yield.
Data of 2,2,6,6-Tetrafluoro-3,3,5-tris(pentafluorophenyl)-3,6-dihydro-2H-1,4-oxazine (3). 19F-NMR
(CDCl3)2): À71.78(br. s, 2-F of C4F4NO); À76.32 (br. s, 6-F of C4F4NO); À135.55 (br. s, 2,6-F of 5,5-
(C6F5)2); À140.01 (m, 2,6-F of 3-C6F5); À147.92 (tt, J=21.5, 4.0, 4-F of 3-C6F5); À149.27 (tt, J=21.5,
4.8, 4-F of 5,5-(C6F5)2); À159.42 (dm, J=21.5, 3,5-F of 3-C6F5); À159.49 (dm, J=21.5, 3,5-F of 5,5-
(C6F5)2). 13C-NMR (CDCl3)2): 66.36 (5-C of C4F4NO); 106.20 (1-C of 3-C6F5); 110.18(1-C of 5,5-
(C6F5)2); 113.25 (2-C of C6F4NO); 119.03 (6-C of C6F4NO); 137.86 (3,5-C of 3-C6F5); 138.30 (3,5-C of
5,5-(C6F5)2); 142.38(4-C of 5,5-(C6F5)2); 143.52 (4-C of 3-C6F5); 144.93 (2,6-C of 5,5-(C6F5)2); 149.78
(3-C of C6F4NO); 145.03 (2,6-C of 3-C6F5). DI-MS (EI): 655 (11, M+), 589 (19, [MÀCOF2]+), 396
(54), 346 (60), 327 (100), 296 (31), 229 (26), 217 (25), 193 (10), 179 (40), 148(13), 117 (10), 93 (13).
Reaction of 3 with an Equimolar Amount of Me3Si(C6F5)3. Spray-dried KF (10 mg, 0.11 mmol) in a
100-ml reaction vessel was dried at 80–908 under vacuum, and then treated with anh. MeCN (1 ml) using
the vacuum line. The reaction vessel was cooled at À788, and 3 (0.20 g, 0.31 mmol) was added under Ar
atmosphere. The mixture was cooled to À1968, and Me
A
uum transfer. The mixture was stirred at 08 for 1 h, and then at r.t. for 20 h. The volatile products were
removed in vacuum at 508, and the residue was extracted with CHCl3 (5 ml), which, after evaporation,
gave a deep-orange viscous liquid containing 3 (72%), 5 (4.5%), and 6 (3.3%), as determined by 19F-
NMR. Anal. samples for the identification of 5 and 6 were obtained by CC (SiO2; hexane/CH2Cl2 99 :1).
Data of 2,2,6,6-Tetrafluoro-5-(2,2’,3,3’,4’,5,5’,6,6’-nonafluorobiphenyl-4-yl)-3,3-bis(pentafluoro-
phenyl)-3,6-dihydro-2H-1,4-oxazine (5). 19F-NMR (CDCl3)2): À71.44 (br. s, 2-F of C4F4NO); À76.26
(br. s, 6-F of C4F4NO); À135.48(br. s, 2,6-F of C6F5); À135.74 (m, 2,6-F of C12F9); À137.25 (m, 2’,6’-F
of C6F5); À139.37 (m, 3,5-F of C12F9); À149.10 (tm, J=21.7, 4’-F of C12F9); À149.12 (tm, J=23.7, 4-F
of C6F5); À159.36 (m, 3,5-F of C6F5); À160.21 (m, 3’,5’-F of C12F9). 13C-NMR (CDCl3)2): 66.31 (5-C of
C4F4NO); 101.44 (1’-C of 3-C12F9); 110.00; (4-C of 3-C12F9) 110.10 (1-C of 3-C12F9); 110.14 (1-C of 5,5-
(C6F5)2); 113.27 (2-C of C4F4NO); 119.05 (6-C of C4F4NO); 138.03 (3’,5’-C of 3-C12F9); 138.28 (3,5-C of
5,5-(C6F5)2); 142.37 (4-C of 5,5-(C6F5)2); 142.93 (4’-C of 3-C12F9); 144.22 (2’,6’-C of 3-C12F9); 144.46 (3,
5-C of 3-C12F9); 144.48(2,6-C of 3-C 12F9); 145.03 (2,6-C of 5,5-(C6F5)2); 149.98(3-C of C F4NO). DI-
4
MS (EI): 803 (11, M+), 737 (17, [MÀCOF2]+), 544 (20), 475 (13), 396 (12), 377 (13), 365 (12), 346
(44), 341 (14), 327 (100), 296 (27), 217 (18), 179 (15).
Data of 2,2,6,6-Tetrafluoro-3-(2,2’,3,3’,4’,5,5’,6,6’-nonafluorobiphenyl-4-yl)-3,5-bis(pentafluoro-
phenyl)-3,6-dihydro-2H-1,4-oxazine (6). 19F-NMR (CDCl3)2): À71.58(br. s, 2-F of C4F4NO); À76.23
(br. s, 6-F of C4F4NO); À135.19 (br. s, 2,6-F of 5-C6F5); À135.19 (br. s, 2,6-F of C12F9); À135.90 (br. s,
3,5-F of C12F9); À137.16 (m, 2’,6’-F of C12F9); À139.97 (m, 2,6-F of 3-C6F5); À147.84 (t, J=21.5, 4-F of
3-C6F5); À149.12 (tm, J=21.5, 4-F of 5-C6F5); À149.23 (t, J=19.5, 4’-F of C12F9); À159.27 (m, 3,5-F of
5-C6F5); À159.34 (m, 3,5-F of 3-C6F5); À160.34 (m, 3’,5’-F of C12F9). 13C-NMR (CDCl3)2): 66.62 (5-C
of C4F4NO); 101.40 (1’-C of C12F9); 106.17 (1-C of 5-C6F5); 108.88 (1-C of 3-C6F5); 109.99 (1-C of
C12F9); 113.26 (2-C of C4F4NO); 116.48(4-C of C 12F9); 119.01 (6-C of C4F4NO); 137.85 (3,5-C of 3-
C6F5); 138.01 (3,5-C of 5-C6F5); 138.30 (3’,5’-C of C12F9); 142.47 (4-C of 5-C6F5); 142.93 (4-C of C12F9);
143.51 (4-C of 3-C6F5); 144.46 (2’,6’-C of C12F9); 144.83 (3,5-C of C12F9); 144.91 (2,6-C of 5-C6F5, 2,6-C
of C12F9); 145.00 (2,6-C of 3-C6F5); 149.97 (3-C of C4F4NO). DI-MS (EI): 803 (21, M+), 737 (44,
[MÀCOF2]+), 544 (31), 494 (57), 475 (100), 444 (27), 437 (15), 406 (18), 396 (19), 377 (13), 375 (10),
365 (14), 346 (18), 327 (47), 296 (15), 247 (23), 237 (43), 229 (43), 222 (22), 217 (33), 203 (12), 193
(17), 179 (47).
Reaction of 3 with 3 Equiv. of Me3Si(C6F5) under More-Harsh Conditions. Spray-dried KF (10 mg,
0.11 mmol) was placed in a 100-ml reaction vessel and dried at 80–908 under vacuum. Then, anh.
MeCN (1 ml) was added using the vacuum line, and the mixture was cooled at À788. Compound 3
(0.20 g, 0.31 mmol) was added under Ar atmosphere. The mixture was cooled at À1968, and treated
with Me3Si(C6F5) (0.23 g, 0.97 mmol) by vacuum transfer. The mixture was stirred at 08 for 1 h, and
N
then heated at 508 for 48h. The volatile products were removed in vacuo at 508, and the residue was

Helvetica Chimica Acta – Vol. 89 (2006)
2683
extracted with CHCl3 (5 ml). Evaporation of the extract gave a deep-brown viscous liquid (0.22 g), which
was subjected to CC (SiO2; 1. hexane/CH2Cl2 99 :1, 2. hexane/AcOEt 95 :5). The less-polar fractions
eluted with hexane/CH2Cl2 consisted of several products of similar Rf values, which were not further sep-
arated. By 19F-NMR analysis, in combination with DFT calculations, compounds 3 (5.6%), 5 (8.3%), 6
(3.5%), 7a (7.8%), 7b (3.0%), and 8 (0.6%)4) could be identified (see text). The more-polar compounds
trapped on top of the silica-gel column, eluted with hexane/AcOEt, formed a complex mixture of prod-
ucts (30 mg), as concluded by 19F-NMR. According to MALDI-TOF-MS experiments, it was obvious that
a homologous series of dimers had been formed (see text and Fig. 2).
Data of 2,2,6,6-Tetrafluoro-3,5-bis(2,2’,3,3’,4’,5,5’,6,6’-nonafluorobiphenyl-4-yl)-3-(pentafluoro-
phenyl)-3,6-dihydro-2H-1,4-oxazine (7a). 19F-NMR (CDCl3)2): À71.58(br. s, 2-F of C4F4NO); À76.20
(br. s, 6-F of C4F4NO); À135.05 (br. s, 2,6-F of C6F5); À135.22 (br. s, 3,5-F of 5-C12F9); À135.73 (m, 2,
6-F of 3-C12F9); À135.77 (m, 2,6-F of 5-C12F9); À137.11 (m, 2’,6’-F of 5-C12F9); À137.26 (m, 2’,6’-F of
3-C12F9); À139.32 (m, 3,5-F of 3-C12F9); À149.01 (tm, J=21.5, 4-F of C6F5); À149.14 (tm, J=19.8, 4’-F
of 5-C12F9); À149.21 (tm, J=21.5, 4’-F of 3-C12F9); À159.15 (m, 3,5-F of C6F5); À160.23 (m, 3’,5’-F of
5-C12F9); À160.31 (m, 3’,5’-F of 3-C12F9). 13C-NMR (CDCl3)2): 66.64 (5-C of C4F4NO); 101.40 (1’-C of
5-C12F9); 101.46 (1’-C of 3-C12F9); 108.93 (1-C of 5-C6F5); 109.96 (1-C of 3-C12F9); 110.04 (1-C of 5-
C12F9); 112.66 (4-C of 3-C12F9); 113.30 (2-C of C4F4NO); 116.50 (4-C of 5-C12F9); 119.03 (6-C of
C4F4NO); 138.02 (3’,5’-C of 3-C12F9); 138.04 (3’,5’-C of 5-C12F9); 138.33 (3,5-C of 5-C6F5); 142.46 (4-C
of 5-C6F5); 142.91 (4’-C of 5-C12F9); 142.95 (4’-C of 3-C12F9); 144.24 (2,6-C of 5-C6F5); 144.46 (3,5-C of
5-C12F9); 144.50 (2,6-C and 2’,6’-C of 3-C12F9; 2’,6’-C of 5-C12F9); 144.56 (2,6-C of 5-C12F9); 145.02 (3,5-
C of 3-C12F9); 150.24 (3-C of C4F4NO). DI-MS (EI): 951 (13, M+), 885 (24, [MÀCOF2]+), 692 (12),
544 (25), 494 (36), 475 (100), 444 (25), 437 (13), 406 (12), 377 (17), 365 (26), 346 (11), 341 (29), 327
(59), 296 (14), 272 (11), 237 (13), 217 (14), 179 (14).
Data of 2,2,6,6-Tetrafluoro-3,3-bis(2,2’,3,3’,4’,5,5’,6,6’-nonafluorobiphenyl-4-yl)-5-(pentafluoro-
phenyl)-3,6-dihydro-2H-1,4-oxazine (7b). 19F-NMR (CDCl3)2): À71.72 (br. s, 2-F of C4F4NO); À75.78
(br. s, 6-F of C4F4NO); À134.84 (br. s, 3,5-F of 5,5-(C12F9)2); À135.72 (br. s, 2,6-F of 5,5-(C12F9)2);
À137.14 (br. s, 2’,6’-F of 5,5-(C12F9)2); À139.99 (br. s, 2,6-F of C6F5); À147.79 (tm, J=21.5, 4-F of
C6F5); À149.19 (td, J=21.7, 3.95, 4’-F of 5,5-(C12F9)2); À159.31 (m, 3,5-F of C6F5); À160.30 (m, 3’,5’-F
of 5,5-(C12F9)2). 13C-NMR (CDCl3)2): 66.60 (5-C of C4F4NO); 101.39 (1’-C of 5,5-(C12F9)2); 109.93 (1-C
of 3-C6F5); 110.02 (1-C of 5,5-(C12F9)2); 113.26 (2-C of C4F4NO); 116.65 (4-C of 5,5-(C12F9)2); 119.55
(6-C of C4F4NO); 137.83 (3,5-C of 3-C6F5); 138.00 (3’,5’-C of 5,5-(C12F9)2); 142.89 (4’-C of 5,5-
(C12F9)2); 143.50 (4-C of 3-C6F5); 144.45 (3,5-C and 2’,6’-C of 5,5-(C12F9)2); 144.52 (2,6-C of 5,5-
(C12F9)2); 144.89 (2,6-C of 3-C6F5); 149.78(3-C of C 4F4NO). DI-MS (EI): 951 (8.9, M+), 885 (21,
[MÀCOF2]+), 692 (34), 672 (13), 622 (100), 592 (28), 554 (19), 544 (38), 525 (27), 517 (22), 516 (23),
513 (16), 494 (42), 489 (13), 485 (15), 475 (99), 444 (24), 437 (19), 420 (15), 406 (13), 377 (35), 365
(51), 358 (13), 346 (31), 341 (34), 327 (89), 321 (26), 312 (66), 311 (42), 296 (48), 277 (21), 272 (15),
261 (13), 237 (13), 229 (24), 217 (30), 179 (23).
2,2,6,6-Tetrafluoro-3,3,5-tris(2,2’,3,3’,4’,5,5’,6,6’-nonafluorobiphenyl-4-yl)-3,6-dihydro-2H-1,4-oxa-
zine (8). 19F-NMR (CDCl3)2): À71.46 (br. s, 2-F of C4F4NO); À75.71 (br. s, 6-F of C4F4NO); À134.84 (br.
s, 3,5-F of 5,5-(C12F9)2); À135.10 (m, 2,6-F of 3-C12F9); À135.68( m, 2,6-F of 5,5-(C12F9)2); À137.08( m, 2’,
6’-F of 5,5-(C12F9)2); À137.28( m, 2’,6’-F of 3-C12F9); À139.32 (br. s, 3,5-F of 3-C12F9); À149.13 (tm,
J=21.5, 4’-F of 3-C12F9); À149.17 (tm, J=21.5, 4’-F of 5,5-(C12F9)2); À160.29 (m, 3’,5’-F of 3-C12F9);
À160.31 (m, 3’,5’-F of 5,5-(C12F9)2). 13C-NMR (CDCl3)2): 66.91 (5-C of C4F4NO); 101.39 (1’-C of 3-
C12F9); 101.75 (1’-C of 5,5-(C12F9)2); 109.01 (1-C of 5,5-(C12F9)2); 110.06 (1-C of 3-C12F9); 113.29 (2-C
of C4F4NO); 115.08(4-C of 3-C 12F9); 116.24 (6-C of C4F4NO); 116.25 (4-C of 5,5-(C12F9)2); 138.02 (3’,
5’-C of 3-C12F9 and 3’,5’-C of 5,5-(C12F9)2); 142.85 (4’-C of 3-C12F9); 142.90 (4’-C of 5,5-(C12F9)2);
144.16 (3,5-C of 3-C12F9); 144.37 (3,5-C of 5,5-(C12F9)2); 144.49 (2,6-C of 5,5-(C12F9)2 and 2’,6’-C of 5,5-
(C12F9)2); 144.53 (2,6-C and 2’,6’-C of 3-C12F9); 150.44 (3-C of C4F4NO). MALDI-TOF-MS: 1099 (100,
M+), 1080 (65, [MÀF]+).
Yields determined by 19F-NMR integration.
4
)

2684
Helvetica Chimica Acta – Vol. 89 (2006)
Table 4. Crystal Data and Details of Structure Refinement of 3
Empirical formula
Formula weight
C22
655.23
N
G
Temperature
173(2) K
Wavelength
0.71075 Š
Crystal system, space group
Unit-cell dimensions
Monoclinic, P 21/c
a=12.1820(11) Š
b=13.0800(12) Š
c=27.184(3) Š
b=91.415(2)8
3
Volume
4330.2(7) Š
Z
8
Density (calc.)
Absorption coefficient
F(000)
2.010 Mg/m3
0.237 mmÀ1
2544
Data, restraints, parameters
Goodness-of-fit on F2
Final R for 8144 refl. (I>2s(I))
R Indices (all data)
Largest diff. peak and hole
9944, 0, 775
1.040
R1=0.0342, wR2=0.0840
R1=0.0444, wR2=0.0921
0.288, À0.247 e Š
À3
Single Crystal X-Ray Analysis of 3 (Table 4)5). Suitable crystals of 3 were obtained by recrystalliza-
tion from an aerobic hexane soln. at 58 of an anal. sample obtained by CC (SiO2; hexane). A single crystal
of dimension 0.30”0.24”0.16 mm was mounted on a glass capillary and used for diffraction-data collec-
tion on a Bruker SMART CCD system at a temp. of 173(2) K. A total of 57,200 reflections were measured
(1.50<q<27.528), with 9,944 reflections being unique (Rint =0.0349). Absorption correction by the
multi-scan method (SADABS) was applied to the data set, and the maximum and minimum transmission
factors were 0.963 and 0.861, resp. The structure was solved by direct methods (SHELXS-97), and the
non-H-atoms were refined anisotropically by full-matrix least-squares on F2 (SHELXL-97). The final
R1value was 0.0342 for 8,144 reflections with I>2s(I).
Computational Methods. Density-functional-theory (DFT) calculations were performed with the
Gaussian98program package [31]. All geometries were optimized at the B3LYP hybrid functional
[32][33] with the 6-31G(d) basis set. Isotropic NMR-shielding tensors were calculated at the B3LYP
level using the gauge-independent atomic orbital (GIAO) method [34–36] with the 6-31++G(d, p)
basis set. The chemical shifts d were calculated from the shielding (s) as d=sref Às, where sref is the
19F-NMR shielding of CFCl3 (sref =179.1618ppm). The calculated 13C-NMR shifts were derived as
above, but rel. to Me
ACHTREUNG
[1] E. Y.-X. Chen, T. J. Marks, Chem. Rev. 2000, 100, 1391.
[2] W. E. Piers, G. J. Irvine, V. C. Williams, Eur. J. Inorg. Chem. 2000, 2131.
[3] B. Hoge, C. Thçsen, T. Herrmann, P. Panne, I. Pantenburg, J. Fluorine Chem. 2004, 125, 831.
[4] A. A. Goodvin, F. W. Mercer, M. T. McKenzie, Macromolecules 1997, 30, 2767.
[5] K. Miyatake, K. Oyaizu, E. Tsuchida, A. S. Hay, Macromolecules 2001, 34, 2065.
[6] Z. Lu, L. Cheng, J. Li, K. Zhang, S. Yi, J. Qin, J. Polymer Sci., A 2004, 42, 925.
5
)
The crystallographic data of 3 have been deposited with the Cambridge Crystallographic Data Centre
as supplementary publication number CCDC-612410. Copies of the data can be obtained at http://
www.ccdc.cam.ac.uk/data_request/cif.

Helvetica Chimica Acta – Vol. 89 (2006)
2685
[7] H.-J. Lee, E.-M. Lee, M.-H. Lee, M.-C. Oh, J.-H. Ahn, S. G. Han, H. G. Kim, J. Polymer Sci., A 1998,
36, 2881.
[8] J.-P. Kim, W.-Y. Lee, J.-W. Kang, S.-K. Kwon, J.-J. Kim, J.-S. Lee, Macromolecules 2001, 34, 7817.
[9] Y. Li, P. W. H. Chan, N.-Y. Zhu, C.-M. Che, H.-L. Kwong, Organometallics 2004, 23, 54.
[10] E. Baciocchi, M. F. Gerini, A. Lapi, J. Org. Chem. 2004, 69, 3586.
[11] K. Tomizaki, A. B. Lysenko, M. Taniguchi, J. S. Lindsey, Tetrahedron 2004, 60, 2011.
[12] R. Arakawa, Y. Shimomae, H. Morikawa, K. Ohara, S. Okuno, J. Mass Spectrom. 2004, 39, 961.
[13] ꢁSynthetic Fluorine Chemistryꢂ, Eds. G. A. Olah, R. D. Chambers, G. K. S. Prakash, John Wiley &
Sons, New York, 1992, Chapt. 10.
[14] G. K. S. Prakash, A. K. Yudin, Chem. Rev. 1997, 97, 757.
[15] R. P. Singh, J. M. Shreeve, Tetrahedron 2000, 56, 7613.
[16] N. R. Patel, R. L. Kirchmeier, Inorg. Chem. 1992, 31, 2537.
[17] V. V. Bardin, V. A. Petrov, I. V. Stennikova, G. G. Furin, Zh. Org. Khim. 1989, 25, 52.
[18] V. V. Bardin, V. A. Petrov, V. K. Grinevskaya, E. I. Mysov, G. G. Furin, L. S. German, Zh. Org.
Khim. 1990, 25, 923.
[19] N. R. Patel, R. L. Kirchmeier, J. M. Shreeve, Inorg. Chem. 1993, 32, 4802.
[20] M. Nishida, A. Vij, R. L. Kirchmeier, J. M. Shreeve, Inorg. Chem. 1993, 34, 6085.
[21] M. Nishida, T. Ono, T. Abe, Nippon KagakuKaishi 2000, 817.
[22] M. Nishida, H. Fukaya, T. Ono, Eur. J. Org. Chem. 2003, 3648.
[23] M. Nishida, T. Ono, J. Fluorine Chem. 2003, 120, 93.
[24] R. E. Banks, M. K. Besheesh, N. J. Lawrence, D. J. Tovell, J. Fluorine Chem. 1999, 97, 79.
[25] W. Tyrra, D. Naumann, J. Fluorine Chem. 2005, 126, 1303.
[26] G. W. Coates, A. R. Dunn, L. M. Henling, J. W. Ziller, E. B. Lobkovsky, R. H. Grubbs, J. Am. Chem.
Soc. 1998, 120, 3641.
ˇ
[27] M. Gdaniec, W. Jankowski, M. J. Milewska, T. Połonski, Angew. Chem., Int. Ed. 2003, 42, 3903.
[28] T. Korenaga, Y. Kawauchi, T. Kosaki, T. Ema, T. Sakai, Bull. Chem. Soc. Jpn. 2005, 78, 2175.
[29] H. Fukaya, T. Ono, J. Comput. Chem. 2004, 25, 51.
[30] M. Nishida, H. Fukaya, T. Abe, J. Fluorine Chem. 1996, 76, 3.
[31] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakr-
zewski, J. A. Montgomery Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Dan-
iels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C.
Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D.
K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W.
Gill, B. Johnson, W.Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A.
Pople, ꢁGaussian 98, Revision A.7ꢂ, Gaussian, Inc., Pittsburgh PA, 1998.
[32] A. D. Becke, J Chem. Phys. 1993, 98, 5648.
[33] C. Lee, W.Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[34] R. Ditchfield, Mol. Phys. 1974, 27, 789.
[35] K. Wolinski, A. J. Sadlej, Mol. Phys. 1980, 41, 1419.
[36] K. Wolinski, J. F. Hilton, P. Pulay, J. Am. Chem. Soc. 1990, 112, 8251.
Received April 21, 2006