Reactions of nucleophiles with perfluoro[2.2]paracyclophane

Article abstract of DOI:10.1021/jo9014535

(Chemical Equation Presented) The aromatic rings of perfluoro[2.2] paracyclophane are extremely reactive with respect to nucleophilic substitution reactions. This paper emphasizes products of monosubstitution by hydroxide, alkoxide, thiolate, enolate, and

Full text of DOI:10.1021/jo9014535

pubs.acs.org/joc
Reactions of Nucleophiles with Perfluoro[2.2]paracyclophane
Lianhao Zhang, Katsu Ogawa, Ion Ghiviriga, and William R. Dolbier Jr.*
Department of Chemistry, P.O. Box 117200, University of Florida, Gainesville, Florida 32611-7200
wrd@chem.ufl.edu
Received July 7, 2009
The aromatic rings of perfluoro[2.2]paracyclophane are extremely reactive with respect to nucleo-
philic substitution reactions. This paper emphasizes products of monosubstitution by hydroxide,
alkoxide, thiolate, enolate, and amine nucleophiles. All reactions appear to proceed via SNAr
mechanisms; reactivity issues are discussed including the effect of substituents on reactivity and
regiochemistry of substitution.
SCHEME 1. Reactivity of Hexafluorobenzene and Pentafluoro-
pyridine with Nucleophiles
Introduction
Fluorine substituents on alkenes or aromatic rings are
known to significantly enhance the electron deficiency of
these unsaturated systems and as a result increase their
reactivity toward nucleophiles.1-3 Trifluoromethyl substi-
tuents, although not as effective as a nitrile group, are even more
effective activating groups for such nucleophilic attack.3-5
Such activation derives, of course, from the ability of these
groups to stabilize the carbanion “Meisenheimer” intermedi-
ates that would be formed, for example, during a nucleo-
philic aromatic substitution reaction proceeding via an SNAr
mechanism. Such a process, proceeding by a carbanion
intermediate, is the most common by which nucleophilic
aromatic substitution reactions occur, such mechanism
being facilitated by substituents that will increase the elec-
tron deficiency of the system.
trifluoromethyl-substituted analogues, such as perfluoro-
toluene.3,4,8 In a kinetic study of the reactivity of perfluoro-
polymethylbenzenes toward nucleophiles,4 it was observed
that, in its reaction with -OCH3/HOCH3 at 25 °C, perfluoro-
toluene is 7000 times as reactive as hexafluorobenzene.
Adding a second (para) trifluoromethyl group (as in per-
fluoro-p-xylene) leads to a somewhat lower reactivity, but it
is still 2900 times more reactive than hexafluorobenzene.
Perfluoro[2.2]paracyclophane (F8) has recently been
synthesized9 and because its aromatic rings resemble those
of perfluoro-p-xylene, it should be expected to exhibit high
reactivity toward nucleophiles, although the reactivity may
be somewhat diminished because of the nonplanarity of F8’s
benzene rings.
Because of the presence of multiple fluorine substituents,
hexafluorobenzene and pentafluoropyridine exhibit high
reactivity toward nucleophiles (Scheme 1),2,3,6,7 as do
*To whom correspondence should be addressed.
(1) Chambers, R. D.; Vaughan, J. F. S. In Organofluorine Chemistry.
Fluorinated Alkenes and Reactive Intermediates; Chambers, R. D., Ed.;
Springer: Berlin, Germany, 1997; Vol. 192, pp 1-88.
(2) Brooke, G. M. J. Fluorine Chem. 1997, 86, 1–76.
(3) Chambers, R. D.; Martin, P. A.; Sandford, G.; William, D. L. H.
J. Fluorine Chem. 2008, 129, 998–1002.
(4) Rodionov, P. P.; Osina, O. I.; Platonov, V. E.; Yakobson, G. G. Bull.
Soc. Chim. Fr. 1986, 986–992.
(5) Shein, S. M.; Evstifeev, A. V. Zh. Obshch. Khim. 1968, 38, 487–492.
(6) Bellas, M.; Price, D.; Suschitzky, H. J. Chem. Soc. C 1967, 1249–1254.
(7) Chambers, R. D.; Hutchinson, J.; Musgrave, W. K. R. J. Chem. Soc.
1964, 3736–3739.
(8) Aroskar, E. V.; Chaudhry, M. T.; Stephens, R.; Tatlow, J. C. J. Chem.
Soc. 1964, 2975–2981.
(9) Dolbier, W. R., Jr.; Xie, P.; Zhang, L.; Xu, W.; Chang, Y.; Abboud,
K. A. J. Org. Chem. 2008, 73, 2469–2472.
DOI: 10.1021/jo9014535
r
Published on Web 08/10/2009
J. Org. Chem. 2009, 74, 6831–6836 6831
2009 American Chemical Society

Zhang et al.
JOCArticle
SCHEME 2. Reactions of F8 with Nucleophiles
Results and Discussion
system that will strongly inhibit further reaction of a nucleo-
phile with the ring bearing the O-. Not only that, but the
impact of the O- must also be significantly transmitted to the
other benzene ring of the paracyclophane, since a second
strong hydroxide nucleophile is also not observed to add to
that ring either. As a weaker donor, the methoxy substituent
of 1b appears to inhibit its reaction with a second equivalent
of methoxide, but its influence is not sufficiently strong to
prevent substitution of the other, unsubstituted benzene ring
of 1b. The observed result was consistent with results ob-
tained by Tatlow and co-workers in their study of the
reaction of perfluoro-p-xylene with methoxide ion.8 In con-
trast, the results obtained from the reaction of F8 with
thiophenolate anion clearly indicate that the SPh substituent
of the putative monoadduct must activate that ring toward
addition of a second nucleophile. Such results are consistent
with the previously observed formation of only p-bis-
(phenylthio)-2,3,5,6-tetrafluorobenzene from the reaction
of either 1 or 2 equiv of phenylthiolate anion with hexafluoro-
benzene.10
Indeed, F8 proved to be very reactive with a large variety of
nucleophiles. In this paper, reactions that led mainly to
monosubstitution will be emphasized, with discussion cen-
tered on the definition of factors that favor monosubstitution.
Subsequent papers will deal with multisubstitution reactions
of F8, the regiochemistry of multisubstitution, and character-
ization of the multisubstituted products, including detailed
multidimensional NMR analysis of these products.
When F8 was allowed to react at room temperature with
up to 8 equiv of NaOH in aqueous THF a single, mono-
hydroxy product 1a was formed in 99% yield (Scheme 2). A
similar reaction of 2.2 equiv of NaOMe in THF yielded the
monosubstituted product 1b in 49% yield along with 14% of
a dimethoxy product, the pseudo-para isomer 2. Providing
further contrast, a reaction of F8 with 1 equiv of sodium
thiophenolate yielded no monosubstituted product at all.
Instead, the p-bis(phenylthio) product 3 was obtained in
30% yield along with small amounts of tetrakis- and hexakis-
(phenylthio) products.
All three of the above nucleophiles were highly reactive in
their respective substitution reactions with F8, with the
reactions being complete after 2 days at room temperature
for hydroxide and methoxide. The reaction with phenyl
thiolate required only 1 day. The differences exhibited by
these nucleophiles with respect to multiple substitution can
be explained based on the variable effects of the different
substituents (O-, OMe, and SPh) of the monoadducts on
their reactions with a second equivalent of nucleophile.
Substitution of fluorine by hydroxide to form phenol deri-
vative 1a will, of course, under the reaction conditions
actually form the deprotonated phenolate anion, and the
TABLE 1. Reaction of Nucleophiles with Perfluoro[2.2]paracyclophane
in THF at Room Temperature
reaction
time, h
product no. and
yield (%)
nucleophile
equiv
color
HO-
8
2
44
48
1a (99)
1b (49)
2 (14)
yellow
white
white
yellow
white
white
white
MeO-
C6H5S-
1
1
4
1.1
24
18
48
18
3 (30)
4-F-C6H4-O-
-CH(CO2Me)2
tert-
1c (77)
1d (73)
1e (54)
butyllithium
Et2NH
(CH2)4NH
PhCH2NHCH3
(CH3)2NH (aq)
4.4
2.2
2.2
2.2
20
24
24
1
1f (91)
1g (84)
1h (68)
1i (70)
yellow
yellow
yellow
yellow
O
- substituent will act as a powerful donor to the aromatic
(10) Langille, K. R.; Peach, M. E. J. Fluorine Chem. 1971/72, 1, 407–414.
6832 J. Org. Chem. Vol. 74, No. 17, 2009

Zhang et al.
JOCArticle
TABLE 2. Reaction of Bidentate Nucleophiles with Perfluoro[2.2]paracyclophane in THF at Room Temperature
bidentate nucleophile
equiv
reaction time, h
product no. and yield (%)
color
HOCH2CH2OH
1.1 (excess NaH)
2.2
2 (excess NaH)
1.1 (excess NaH)
1.1 (excess NaH)
2.2
48
24
48
48
18
18
18
1j (50)
1k (58)
4a (78)
4b (56)
4c (75)
5a (70)
5b (62)
white
NH2CH2CH2NHCH2Ph
1,2-dihydroxybenzene
1,2-dihydroxy-4-nitrobenzene
1,2-benzene-dithiol
NH2CH2CH2NH2
EtNHCH2CH2NHEt
yellow
light yellow
light yellow
brown
red
2.2
red
SCHEME 3. Reaction of F8 with Bidentate Nucleophiles
Dimethyl malonate anion behaves much like hydroxide in its
reaction with F8, forming only a monoadduct even when the
nucleophile is present in great excess. The reason for this is
much the same, since the monoadduct would become im-
mediately deprotonated to form the highly unreactive car-
banion. Other nucleophiles, generally oxygen and nitrogen
nucleophiles that serve to deactivate the ring to which they
become attached, have also been utilized in reaction with F8,
with the results from all of these reactions being summarized
in Table 1.
All of the above reactions were carried out preparatively
for times of between 18 and 48 h, but it was later determined
that the reactions with most of the nucleophiles were com-
plete in less than an hour. Indeed, while conducting relative
reactivity experiments it was determined that the reactions of
methoxide with F8, hexafluorobenzene, and pentafluoropyr-
idine in THF were all complete within 15 min. Under
conditions of direct competition, in the presence of equal
amounts of F8 and pentafluoropyridine, methoxide reacted
exclusively with pentafluoropyridine. Likewise, F8 reacted
exclusively in competition with hexafluorobenzene. Thus in
reactions with nucleophiles, pentafluoropyridine is much
more reactive than F8, which is itself much more reactive
than hexafluorobenzene.
All of the reaction mixtures were observed to develop a
yellow color during reaction. Among the monosubstituted
products, the ether, malonate, and tert-butyl products were
colorless, whereas the hydroxy, sulfide, and amine products
were various shades of yellow. The UV-vis absorption
spectra of these products show a progression toward longer
wavelength absorption as the substituent becomes increas-
ingly electron donating (see Figures SI-1a and 1b and SI-2 in
the Supporting Information).
addition-elimination mechanism involving formation of a
Meisenheimer (carbanion) intermediate, because of the ex-
treme electron deficiency of the F8 substrate and the obvious
electron-transfer ability of many of the nucleophiles in Table 1,
it was considered prudent to also consider the possibility
that the reactions might proceed via an electron-transfer,
free radical chain SRN1 mechanism. In pursuit of evidence
regarding this issue, the reduction potential of F8 was
determined via differential pulse voltammetry, and EPR
studies were carried out to determine whether the F8 radical
anion might be detected, either electrochemically or during
the course of any of the reactions of F8 with the various
nucleophiles.
Electrochemical characterizations of F8 and nitro-AF4
were performed by cyclic voltammetry (CV) and differential
pulse voltammetry (DPV) in acetonitrile with 0.1 M
tetrabutylammonium hexafluorophosphate (Bu4NPF6) as
a supporting electrolyte. The results are given in Table 3.
The cyclic voltammogram of F8 shows an irreversible
reduction wave with a peak at -1.24 V (vs. SCE)
(Figure 1a). The voltammogram of F8 shows anodic current
in the reverse scan, peaked at -0.80 V (vs. SCE), correspond-
ing to the oxidation of reduced species. In addition, no
change in the voltammogram was observed when scanning
Bidentate nucleophiles were also examined to determine
whether the intramolecular mode of reaction for the second
nucleophile might provide sufficient kinetic advantage to
observe cyclic products (Scheme 3 and Table 2). Reaction of
ethylene glycol in the presence of excess NaH led simply to
formation of the monoadduct 1j. Attempts to stimulate
cyclization by raising the temperature of the reaction led
only to decomposition. In contrast, catechol (o-dihydroxy-
benzene) gave the cyclic diether, 4a, resulting from consecu-
tive nucleophilic substitution of F8 by the o-phenolate
anions. Likewise, reactions of the primary and secondary
bis-amines, 1,2-diaminoethane and 1,2-di(ethylamino)ethane,
also resulted in formation of the respective cyclic disubsti-
tuted products 5a and 5b. These two compounds were both
red in color, with their UV bands extending past 500 nm
(see Figures SI-3 and SI-4 in the Supporting Information).
Although all of the reactions of F8 with nucleophiles can
be understood within the context of the conventional SNAr
J. Org. Chem. Vol. 74, No. 17, 2009 6833

Zhang et al.
JOCArticle
TABLE 3. Summary of Electrochemical Characterization of F8 and
Nitro-AF4a
compd
F8
nitro-AF4
Epc,b V
Epa,b V
Ered,c V
-1.24
-1.16
-0.80
-0.65
-1.12
-0.94
aPotentials are adjusted to vs. SCE. bCV. cDPV
cycles were repeated 10 times, which suggests the chemically
reversible nature of the redox process. The results suggest
probable delocalization of the charge through the stacked π-
system of F8 stabilizing the intermediate radical anion
species.
Because of the electrochemically irreversible nature of the
redox process, the DPV technique was utilized to determine
the reduction potential of F8. The DPV voltammogram
shows a reduction peak at -1.12 V (vs. SCE) (Figure 1b).
Although the presence of a possible second reduction wave
was observed, the results were inconclusive due to over-
lapping solvent reduction waves. The reduction potential
of F8 in acetonitrile (-1.12 V vs. SCE) is virtually the same as
those of nitrobenzene (-1.14 V) and p-fluoronitrobenzene
(-1.13 V),11 more negative than that of nitro-AF4 (the nitro
derivative of the bridge-fluorinated [2.2]paracyclophane,
AF4) (-0.94 V), but more positive than that of hexafluoro-
benzene (-2.2 V) or that reported for pentafluoropyridine
(-2.3 V).12 The reduction potential of perfluoro-p-xylene
has not yet been reported.
FIGURE 1. (a) Cyclic voltammogram (CV) of F8. (b) Differential
pulse voltammogram (DPV) of F8.
Halonitroaromatics generally do not undergo substitution
by the free radical chain SRN1 mechanism, mainly because
the intermediate nitro-stabilized aromatic radical anions
appear to be too stable to allow dissociation of halide to
form the propagating aryl radical in a kinetically competitive
manner.13,14 Moreover, fluoride has proved to be by far the
worst halide leaving group for an SRN1 reaction.15,16 Thus, it
seems unlikely that F8, which has a similarly stabilized
radical anion and only fluoride leaving groups, would parti-
cipate in a productive SRN1 reaction.
An attempt to directly observe the F8 radical anion by
EPR under conditions of electrochemical generation failed,
although the F8 was destroyed by the potential; nor could
this radical anion be detected in situ, that is during the
reaction of F8 with PhS-Naþ in THF. Electron transfer
obviously was occurring during the electrochemical experi-
ment; thus the lack of an observable EPR spectrum indicates
that the intermediate radical anion (and any other radical
species that are formed subsequently) must have been de-
stroyed too rapidly to be observed. All that one can conclude
by the lack of an EPR signal during the chemical reactions is
that if the reaction proceeds via an SET process, any radical
anion/radical intermediates must be too short-lived to be
observed in the experiment.
with fluorine having been replaced on the benzene rings. The
only products thus far identified derived from destruction of
the paracyclophane structure, which means that when F8’s
radical anion is formed, it prefers reactions other than loss of
aromatic fluoride. Again, this result appears to preclude
involvement of an SRN1 mechanism in the reaction of F8
with nucleophiles.
The intervention of any free radical chain mechanism
was additionally tested by an experiment in which the
reaction of nucleophile pyrrolidine with F8 was carried out
in the presence of 1 equiv of free radical trap, TEMPO. The
reaction was not inhibited and proceeded in a normal
manner. This result again speaks clearly against involvement
of a free radical chain process.
Lastly, the involvement of a free radical chain process in
the reaction of F8 with methoxide can be specifically ruled
out based on earlier work by Bunnett17 and Saveant,18 which
indicated that methoxide’s preferred reaction with aryl radi-
•
cals is hydrogen atom abstraction to form the CH2O-
radical anion. No such reductive reaction was observed in
any of our reactions.
All of these results lead one to conclude that the reactions
of F8 with nucleophiles cannot proceed via the free radical
chain SRN1 mechanism, and that the most probable mechan-
ism for these reactions is the SNAr mechanism proceeding via
its usual delocalized (Meisenheimer) carbanion intermedi-
ate. One cannot completely rule out electron transfer or at
least formation of a charge transfer complex as the initial
step of the nucleophilic substitution mechanism, since non-
radical chain SET processes, where an intermediate charge
The fate of the F8 radical anion was probed by examining
the reaction mixture after electrochemical reduction of F8 in
acetonitrile. No paracyclophane products could be observed
(11) Shalev, H.; Evans, D. H. J. Am. Chem. Soc. 1989, 111, 2667–2674.
(12) Beletskaya, I. P.; Artamkina, G. A.; Mil’chenko, A. Y.; Sazonov,
P. K.; Shtern, M. M. J. Phys. Org. Chem. 1996, 9, 319–328.
(13) Marquet, J.; Jiang, Z.; Gallardo, I.; Batile, A.; Cayon, E. Tetrahedron
Lett. 1993, 34, 2801–2804.
(14) Rossi, R. A. J. Chem. Educ. 1982, 59, 310–312.
(15) Bunnett, J. F. Acc. Chem. Res. 1978, 11, 413–420.
(16) Bunnett, J. F.; Creary, X. J. Org. Chem. 1974, 39, 3611–3612.
(17) Bunnett, J. F. Acc. Chem. Res. 1992, 25, 2–9.
(18) Amatore, C.; Badoz-:ambling, J.; Bonnel-Huyghes, C.; Pinson, J.;
Saveant, J.-M.; Thiebault, A. J. Am. Chem. Soc. 1979, 104, 1979–1986.
6834 J. Org. Chem. Vol. 74, No. 17, 2009

Zhang et al.
JOCArticle
transfer complex of radical anion and radical collapse within
the solvent cage to form the Meisenheimer complex, have
been proposed previously.19
-132.3 (dd, J1=56 Hz, J2=10 Hz, 1F), -133.2 (dt, J1=64 Hz,
J2=17 Hz, 1F), -134.6 (m, 1F), -135.3 (m, 1F). Anal. Calcd for
C22H4F16O: C 44.92, H 0.69. Found: C 45.24, H 0.72.
4-(Bis(carbomethoxyl)methyl)perfluoro[2.2]paracyclophane (1d).
To a solution of dimethyl malonate (161.7 mg, 1.2 mmol) in
anhydrous THF (10 mL) was added 60% sodium hydride
(48 mg, 1.2 mmol) and the mixture was stirred at rt for 10 min.
Then F8 (148.8 mg, 0.3 mmol) was added and the reaction
mixture was stirred at rt for 2 days, after which it was concen-
trated to dryness. The residue was purified by column chromato-
graphy (chloroform) to obtain 1d (80 mg, 73%) as a white
solid: mp 148-149 °C; 1H NMR δ 5.14 (s, 1H), 3.88 (s, 3H), 3.78
(s, 3H); 19F NMR δ -98.7 (ddd, J1=251 Hz, J2=25 Hz, J3=
12 Hz, 1F), -99.3 (ddd, J1=251 Hz, J2=21 Hz, J3=8 Hz, 1F),
-100.0 (ddd, J1=249 Hz, J2=25 Hz, J3=8 Hz, 1F), -100.1
(ddd, J1=249 Hz, J2=33 Hz, J3=8 Hz, 1F), -104.6 (dt, J1=251
Hz, J2=17 Hz, 1F), -105.4 (dddd, J1=251 Hz, J2=69 Hz, J3=
35 Hz, J4=8 Hz, 1F), -105.7 (ddt, J1=257 Hz, J2=69 Hz, J3=
12 Hz, 1F), -106.6 (ddt, J1=253 Hz, J2=75 Hz, J3=12 Hz),
-109.5 (dd, J1=73 Hz, J2=10 Hz, 1F), -122.1 (m, 1F), -129.0
(m, 2F), -131.8 (m, 1F), -134.1 (m, 1F), -134.5 (m, 1F). Anal.
Calcd for C21H7F15O4: C 41.47, H 1.16. Found: C 41.76, H 1.06.
Catechol Adduct of F8 (4a). To a solution of catechol (110 mg,
1 mmol) in anhydrous THF (10 mL) was added 60% sodium
hydride (88 mg, 2.2 mmol). The resulting reaction mixture was
stirred at room temperature for 10 min, after which F8 (248 mg,
0.5 mmol) was added. The mixture was stirred at rt overnight,
and then it was concentrated to dryness. The residue was
purified by column chromatography (hexanes) to obtain 4a
Conclusion
The aromatic rings of perfluoro[2.2]paracyclophane are
exceptionally receptive to nucleophilic substitution, and all
of the observations related to F8’s reactivity and regiochem-
istry of reactions with the various nucleophiles that have
been presented and discussed in this paper can be readily
rationalized within the framework of the SNAr mechanism.
Experimental Section
Perfluoro[2.2]paracyclophan-4-ol (1a). To a solution of so-
dium hydroxide (128 mg, 3.2 mmol) in water (0.5 mL) was added
tetrahydrofuran (THF) (8 mL) and perfluoro[2.2]paracyclo-
phane (F8) (198.4 mg, 0.4 mmol). The reaction mixture was
homogeneous and it was stirred at room temperature (rt) for
44 h and then concentrated to dryness. The residue was purified
by column chromatography (ethyl acetate) to obtain 1a (196 mg,
99.2%) as a yellow solid: mp 192-193 °C; 1H NMR (acetone-d6)
δ 3.75 (br s, 1H); 19F NMR (acetone-d6) δ -98.15 (d, J=255.0
Hz, 1F), -99.36 (ddm, J1=255.2 Hz, J2=25.1 Hz, 1F), -100.79
(dd, J1=247.0 Hz, J2=27.1 Hz, 1F), -101.27 (ddd, J1=251.0
Hz, J2 =26.8 Hz, J3 =10.4 Hz, 1F), -102.24 (J1 =249.0 Hz,
J2=62.3 Hz, 1F), -105.20 (dddd, J1=244.8 Hz, J2=72.7 Hz,
J3=16.6 Hz, J4=10.4 Hz, 1F), -106.54 (dddd, J1=247.0 Hz,
J2=68.5 Hz, J3=16.6 Hz, J4=10.2 Hz, 1F), -106.82 (dddd, J1=
251.0 Hz, J2=68.5 Hz, J3=18.6 Hz, J4=6.5 Hz, 1F), -131.96
(m, 1F), -134.80 (d, J=22.1 Hz, 1F), -136.44 (m, 1F), -137.98
(d, J=8.7 Hz, 1F), -138.82 (d, J=72.7 Hz, 1F), -144.13 (d, J=
78.9 Hz, 1F), -162.58 (br s, 1F); HRMS (CI) calcd for C16H1-
F15O 493.9788, found 493.9774.
4,7-Bis(phenylthio)perfluoro[2.2]paracyclophane (3). Follow-
ing the procedure used for 1a, a mixture of sodium benzene-
thiolate (29.4 mg, 0.2 mmol) and F8 (99.2 mg, 0.2 mmol) in
anhydrous THF (4 mL) was stirred at rt for 48 h. After column
chromatography (hexanes) 40 mg of product 3 was obtained
(30% based on F8, 60% based on thiolate) as a yellow solid: mp
122-124 °C; 1H NMR, δ 7.75 (m, 10H); 19F NMR δ -100.2 (dd,
J1=245 Hz, J2=12 Hz, 2F), -100.7 (dd, J1=251 Hz, J2=43 Hz,
2F), -101.0 (d, J=63 Hz, 2F), -102.2 (ddd, J1=245 Hz, J2=
66 Hz, J3=6.4 Hz, 2F), -103.3 (dddd, J1=252 Hz, J2=55 Hz,
J3=15 Hz, J4=6 Hz, 2F), -128.5 (dd, J1=43 Hz, J2=11 Hz,
2F), -134.3 (dddd, J1=54 Hz, J2=20 Hz, J3=6 Hz, J4=4 Hz,
2F). Anal. Calcd for C28H10F14S2: C 49.71, H 1.49. Found: C
49.84, H 1.64.
4-(4-Fluorophenoxy)perfluoro[2.2]paracyclophane (1c). To a
mixture of 4-fluorophenol (28 mg, 0.25 mmol) in anhydrous
THF (10 mL) was added 60% sodium hydride (11 mg, 0.275
mmol). The resulting reaction mixture was stirred for 30 min,
after which F8 (124 mg, 0.25 mmol) was added. The mixture was
stirred at rt overnight, and then it was concentrated to dryness.
The residue was purified by column chromatography (hexanes)
to obtain 1c (110 mg, 76.9%) as a white solid: mp 98-99 °C; 1H
NMR δ 7.01 (m, 2H), 6.81 (m, 2H); 19F NMR δ -99.46 (ddd,
J1=249.0 Hz, J2=29.1 Hz, J3=10.4 Hz, 1F), -100.4 (ddd, J1=
253 Hz, J2=31 Hz, J3=10 Hz, 1F), -100.6 (ddd, J1=251 Hz,
J2=29 Hz, J3=10 Hz, 1F), -101.0 (d, J=251 Hz, 1F), -104.5
(dd, J1 =245 Hz, J2 =62 Hz, 1F), -104.9 (ddt, J1 =249 Hz,
J2=73 Hz, J3=15 Hz, 1F), -105.4 (ddt, J1=251 Hz, J2=62 Hz,
J3=14 Hz, 2F), -119.0 (m, 1F), -122.0 (m, 1F), -131.7 (m, 2F),
1
(220 mg, 77.5%) as a light yellow solid: mp 162-163 °C; H
NMR δ 7.06 (dd, J1=6.6 Hz, J2=3.6 Hz, 2H), 6.93 (dd, J1=
6.6 Hz, J2=3.6 Hz, 2H); 19F NMR δ -99.6 (d, J=249 Hz, 2F),
-100.2 (ddd, J1=253 Hz, J2=25 Hz, J3=6 Hz, 2F), -103.8
(ddq, J1=249 Hz, J2=74. Hz, J3=10 Hz, 2F), -104.6 (dd, J1=
253 Hz, J2=62 Hz, 2F), -131.3 (m, 2F), -136.6 (d, J=62 Hz,
2F), -139.2 (d, J=71 Hz, 2F). Anal. Calcd for C22H4F14O2: C
46.66, H 0.71. Found: C 46.35, H 0.51.
Electrochemistry. The cyclic voltammetry (CV) experiments
were performed on a Bioanalytical Systems CW50 electroche-
mical analyzer at a sweep rate of 100 mV/s, using a platinum disk
working electrode, a platinum wire auxiliary electrode, and a
silver wire pseudoreference electrode. At the end of each scan,
ferrocene was added as internal standard and potentials are
referenced to the potential of ferrocene/ferrocenium redox
couple. The differential pulse voltammetry experiments were
performed with the same setup at a scan rate of 20 mV/s, a pulse
amplitude of 50 mV, and a pulse period of 200 ms. Sample and
pulse width were 17 and 50 ms, respectively. Solutions of
samples were prepared in acetonitrile. The supporting electro-
lyte was 0.10 M tetrabutylammonium hexafluorophosphate
(TBAPF6). The experimental potentials obtained vs. ferro-
cene/ferrocenium redox couple were corrected to the SCE
standard (correction factor of þ0.328 C)
Bulk Electrolysis of F8. Bulk electrolysis of F8 was performed
on a Bioanalytical Systems CV-27 cyclic voltammograph, using
a platinum gauze working electrode, a coiled platinum wire
auxiliary electrode, and a silver wire pseudoreference electrode.
Ferrocene was used as an internal standard for the reference
electrode potential. F8 (200 mg, 0.4 mmol) was dissolved in
20 mL of acetonitrile containing 0.10 M tetrabutylammonium
hexafluorophosphate (TBAPF6) as supporting electrolyte. The
solution was purged with nitrogen gas for 15 min. The potential
of the working electrode was kept at -1.1 V (vs. SCE) for 4 h
with continuous stirring. The solution turned darker brown
as the electrolysis proceeded. The current was ca. 10 mA for
the duration of the experiment. The resulting solution was
concentrated to dryness, then purified by column chromato-
graphy (silica gel, hexanes) to obtain two fractions. The first
(19) Bacaloglu, R.; Blasko, A.; Bunton, C.; Dorwin, E.; Ortega, F.;
Zucco, C. J. Am. Chem. Soc. 1991, 113, 238-246 and references cited therein.
J. Org. Chem. Vol. 74, No. 17, 2009 6835

Zhang et al.
JOCArticle
fraction was recovered F8 (120 mg), whereas the second fraction
was a mixture of reduced products. Both proton and fluorine
NMR spectra indicated an absence of aromatic C-H bonds and
the probable presence of a CH2 group resulting from reduction
of two geminal fluorines on one of the bridges of F8 to form a
CH2-CF2 bridge.
indicated that the methoxide anion had only reacted with the
pentafluoropyridine. The F8 was untouched.
A similar experiment designed to compare the reactivities of
F8 and hexafluorobenzene resulted in reaction of methoxide
only with the F8.
Reaction in the Presence of TEMPO. To a solution of
pyrrolidine (20.5 mg, 0.29 mmol) and 2,2,6,6-tetramethylpiper-
idine 1-oxyl (TEMPO, 45 mg, 0.29 mmol) in anhydrous THF
was added F8 (65.4 mg, 0.13 mmol). The resulting reaction
mixture was stirred for 1 h and then concentrated to dryness
and purified by column chromatography (silica gel, hexanes)
to provide 4-(pyrrolidin-1-yl)perfluoro[2.2]paracyclophane (1g)
(100% conversion, 84% isolated yield) as a yellow solid.
Competition Experiments. To a mixture of pentafluoropyr-
idine (50 mg, 0.295 mmol) and F8 (124 mg, 0.25 mmol) in
anhydrous THF (6 mL) was added NaOMe (8 mg, 0.148 mmol).
After stirring for 10 min at rt, a fluorine NMR of the mixture
Acknowledgment. The assistance of Prof. Alex Angerhofer
in attempting to observe the F8 radical anion in EPR
experiments is gratefully acknowledged.
Supporting Information Available: General experimental
information, UV-vis spectra of 1a-i,k, and the sodium salt of
1a, 4a-c, and 5a,b, proton and fluorine NMR spectra of all new
compounds, and details of experimental and characterization
data for compounds 1b, 2, 1e, 1f, 1g, 1h, 1i, 1j, 1k, 4b, 4c, 5a, and
5b. This material is available free of charge via the Internet at
http://pubs.acs.org.
6836 J. Org. Chem. Vol. 74, No. 17, 2009