Novel ring-cleaving reaction of 4-nitro-1,1,2,2,9,9,10,10-octafluoro[2.2]paracyclophane with nucleophiles

Article abstract of DOI:10.1016/j.jfluchem.2008.08.003

When 4-nitro-AF4 is treated with nucleophiles such as alkoxides and cyanide, a novel ring opening, cyclophane destroying reaction is observed whereby, via an S_NAr mechanism, the nucleophile attacks the bridgehead aryl carbon vicinal to the nitro group with subsequent aryl-CF₂ bond cleavage.

Full text of DOI:10.1016/j.jfluchem.2008.08.003

Journal of Fluorine Chemistry 129 (2008) 1133–1138
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Novel ring-cleaving reaction of 4-nitro-1,1,2,2,9,9,10,10-
octafluoro[2.2]paracyclophane with nucleophiles
*
William R. Dolbier Jr. , Yian Zhai, Wei Xu, Will Wheelus, Florian Dulong, Efram Goldberg,
Ion Ghiviriga, Merle A. Battiste
Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, United States
A R T I C L E I N F O
A B S T R A C T
Article history:
When 4-nitro-AF4 is treated with nucleophiles such as alkoxides and cyanide, a novel ring opening,
cyclophane destroying reaction is observed whereby, via an SNAr mechanism, the nucleophile attacks the
bridgehead aryl carbon vicinal to the nitro group with subsequent aryl-CF2 bond cleavage.
ß 2008 Elsevier B.V. All rights reserved.
Received 9 July 2008
Received in revised form 12 August 2008
Accepted 13 August 2008
Available online 23 August 2008
Keywords:
[2.2]Paracyclophane
AF4
Nucleophilic substitution
Fluorine reactivity effects
1. Introduction
product isolated from this reaction (Fig. 2) that the [2.2]para-
cyclophane system was no longer present.
Over the last decade, our broad studies of octafluoro[2.2]par-
acyclophane (AF4) and its derivatives have demonstrated that the
highly electron-deficient aromatic rings of AF4 exhibit a rich
diversity of chemical behavior, including clean, high yield electro-
philic substitution and disubstitution [1,2] unique aryne reactivity
[3] and excellent SRN1-reactivity with soft nucleophiles such as
phenyl thiolate and malonate anion (Scheme 1) [4]. In the present
communication, we wish to report the unexpected discovery of an
unusual ring-opening reaction of nitro-AF4 when it is treated with
hard nucleophiles, such as alkoxides and cyanide ion.
When the reaction with t-BuOÀ was worked up and isolated by
column chromatography, phenol 1a was isolated in 83% yield
(Scheme 2). The product was characterized by 1H, 13C, and 19F NMR,
and by elemental analysis (Fig. 3). Examination of the crude product
mixture indicated that more than one compound bearing a CF2H
group was present, but once the mixture was subjected to isolation
and column chromatography, only the phenol remained. When
other alkoxides, such as methoxide or trifluoroethoxide were used
instead of t-butoxide, in each case only the phenol product could be
isolated. Table 1 provides the results that were obtained from the
reactions of 4-nitro-AF4 with various nucleophiles.
Reactions of [2.2]paracyclophanes that lead to ring-opening
destruction of the paracyclophane system are not common. Early
in their studies of the parent hydrocarbon [2.2]paracyclophane
(PCP), Cram and coworkers reported that, due to the system’s
inherent strain of about 31–33 kcal/mol, the system undergoes
reversible thermal homolytic cleavage of one of the benzylic C1–C2
bonds to form bis-benzylic diradical 2, which could be intercepted
in a number of ways, including by hydrogen atom transfer, as
shown in Scheme 3 [5]. All other examples of ring-opening of
[2.2]paracyclophanes have involved SET oxidation of PCP, pro-
cesses that proceed via radical cation intermediates. Such radical
cations have been generated either electrochemically [6], under
Birch reduction conditions [7], or via treatment of PCP with
cerium(IV) ammonium nitrate (CAN) [8] or the strong one-electron
2. Results and discussion
Thus, upon treatment of 4-nitro-1,1,2,2,9,9,10,10-octafluor-
o[2.2]paracyclophane with KO-t-Bu in t-BuOD, an experiment
designed to induce deuterium exchange at the 5-position, it was
found that instead of exchange the reaction resulted in total
destruction of the [2.2]paracyclophane structure. As is the case for
4-nitro-AF4 (Fig. 1), all monosubstituted derivatives of AF4 exhibit
four characteristic AB systems in their 19F NMR spectra. It was
apparent upon examination of the 19F NMR spectrum of the
* Corresponding author.
E-mail address: wrd@chem.ufl.edu (W.R. Dolbier Jr.).
0022-1139/$ – see front matter ß 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2008.08.003

1134
W.R. Dolbier Jr. et al. / Journal of Fluorine Chemistry 129 (2008) 1133–1138
could possibly derive from a reductive SET process, via a radical
anion intermediate, we believe that the overall results, as
discussed below, are more consistent with the SNAr mechanism
depicted in Scheme 5.
The high reactivity of the substrate, nitro-AF4, towards
nucleophiles can be attributed to the great electron deficiency
of its nitro-bearing benzene ring. This electron deficiency is
reflected by its reduction potential. The respective half-wave
potentials for AF4, iodo-AF4 and nitro-AF4, as indicated by cyclic
voltammetry experiments, are À2.02, À1.69 and À0.86 (V versus
saturated calomel reference electrode, SCE). The former two
reductions were irreversible, whereas the relative stability of the
radical anion of nitro-AF4 allows its reduction to be a reversible
process. The addition of the nucleophile at the bridgehead position
would be activated by the nitro group, by the two electron deficient
bridge fluorinated groups, and by relief of ring strain resulting from
this attack. Additional relief of ring strain would also facilitate the
second step of the mechanism, the novel opening of the
paracyclophane ring system, as would formation of ÀCF2CF2Ar,
which should not be a bad leaving group and which would be
quickly protonated in the presence of cosolvent alcohol.
Scheme 1. SRN1 reactivity.
oxidant, nitrosonium tetrafluoroborate (NOBF4) [9]. As exemplified
by the reaction of PCP with CAN in Scheme 4, all of these oxidation
reactions involve cleavage of the C1–C2 benzylic bond of radical
cation 3 to form the bis-benzylic distonic radical cation 4, which
can be trapped either before of after a second oxidative step to form
products such as 5.
Until the current work, there have been no reports of reactions
that led to cleavage of an aryl-bridge bond, nor have there, to our
knowledge, been any examples of either nucleophile-induced or
reductive ring-cleaving reactions of [2.2]paracyclophanes.
Although it is possible that the process that we have observed
Of course, neither the expected t-butyl nor the methyl ether
product is actually isolated from the reaction. Instead, the phenol
1a is isolated in both cases. We rationalize this fact on the basis
Fig. 1. 19F NMR spectrum of 4-nitro-AF4.
Fig. 2. 19F NMR spectrum of product 1b.

W.R. Dolbier Jr. et al. / Journal of Fluorine Chemistry 129 (2008) 1133–1138
1135
Scheme 2. Ring opening reaction of 4-nitro-AF4.
Fig. 3. Assignment of peaks in 1H, 13C, and 19F NMR spectra.
Scheme 3.
that the presence of the nitro group and two perfluoroalkyl
(bridge) groups on the ether product benzene ring should make the
phenolate anion produced by either an SN2 or an E2 reaction quite a
good leaving group. Therefore, under the highly basic, highly
nucleophilic conditions of the reaction, conversion of the methyl or
t-butyl ethers to the observed phenol product should be quite
facile (Scheme 6).
Indirect evidence for this hypothesis was derived from the
results that were obtained when either cyanide or trifluoroeth-
Table 1
Results for reaction of nucleophiles with nitro-AF4
Nucleophile
(equiv.)
Solvent
Temperature
Time
(h)
Product
(%)
(8C)
t-BuOÀ (6.0)
CH3OÀ (6.5)
CF3CH2OÀ (18)
CNÀ (2.9)
t-BuOH/n-Bu2O (1:1)
CH3OH/n-Bu2O 1:1
THF/CH3CH2OH 1:1
CH3CN
110
110
80
18
19
8
1a (78%)
1a (73%)
1a (67%)
1b (52%)
80
0.5
Scheme 4.

1136
W.R. Dolbier Jr. et al. / Journal of Fluorine Chemistry 129 (2008) 1133–1138
Scheme 5. Suggested mechanism.
Scheme 6. Formation of phenol product 1a.
oxide were used as nucleophile in the reaction. As Scheme 1
indicates, when cyanide was used as nucleophile, the product that
was obtained (in 52% yield after chromatography) retained the
cyano group. Cyanide is not considered to be a good SET
nucleophile, which speaks in favor of the SNAr mechanism, and
the reaction is carried out in the absence of any other reactive base
or nucleophile, so it remains intact.
product, in contrast to the t-butyl ether cannot undergo an E2
elimination to the phenol, and it will be much less reactive than the
methyl ether with respect to SN2 chemistry.
3. Conclusions
In conclusion, nitro-AF4 has been found to be highly receptive
to attack by nucleophiles, such reactions resulting in an
unprecedented overall ring-cleaving process that destroys the
[2.2]paracyclophane entity. These reactions appear to proceed via
an addition–elimination process (SNAr) with initial attack at the
aromatic bridgehead carbon vicinal to the nitro substituent, with
the intermediate delocalized carbanion undergoing rate-deter-
mining fragmentation of the aryl bridge bond with resultant
extrusion of the tetrafluoroethyl anion, which is then protonated.
Although results are presented for use of only a limited number of
nucleophiles, it seems likely that the process will have some
generality.
When trifluoroethoxide was used as the nucleophile in reaction
with nitro AF4, although the phenol 1a was still the only product
observed (67%) after chromatographic purification, an NMR
analysis (19F) of the crude reaction product indicated that the
major observable product after 24 h at 80 8C was not the phenol,
but what appeared to be the trifluorethyl ether, as evidenced by a
significant signal for its CF3 group at À74.8 ppm, a signal clearly
distinct from the analogous signal for the remaining trifluor-
oethanol, which was at À79.9 ppm. A chemical shift of À75 ppm is
consistent with expectations for a trifluoroethyl aryl ether [10]. It
was somewhat surprising to us that the relatively non-basic
nucleophile, CF3CH2OÀ, would be as effective as it was in the SNAr
process. However, trifluoroethoxide has previously been shown to
be an effective nucleophile in SNAr processes [11], and it was used
at a considerably larger concentration than were the other
alkoxides. Also, in all likelihood the rate determining step in this
SNAr process is NOT the nucleophilic addition step, but the second,
ring-cleavage step. Thus the relative nucleophilicity of the
attacking nucleophile should not be very relevant with regard to
the relative efficacy of the overall reaction. The trifluoroethyl ether
4. Experimental
4.1. General
H1–C13 correlation spectra were recorded on a Varian Inova
spectrometer equipped with a 5 mm indirect detection probe,
operating at 500 MHz for 1H and at 125 MHz for 13C. Chemical
shifts are reported in ppm relative to TMS.

W.R. Dolbier Jr. et al. / Journal of Fluorine Chemistry 129 (2008) 1133–1138
1137
19F and H1-F19 HOESY spectra were recorded on a Varian
Mercury spectrometer operating at 300 MHz for 1H and
282 MHz for 19F. 19F chemical shifts are reported in ppm
relative to CFCl3.
HCl. Multiple extractions were performed using diethyl ether/
water and washing with brine. The organic layers were
combined and were dried over anhydrous MgSO4, then filtered.
The solvent was evaporated under vacuum and the residue
purified by silica gel column chromatography eluting initially
with neat hexanes. The product was then eluted incrementally
with increasingly polar mixtures of hexanes/dichloromethane
ending in a ratio of 5:1, respectively. The product was eluted
as a yellow band. Upon evaporation the crystals appeared
as bright yellow rosettes. The like fractions were combined
and the phenol product was obtained with a yield of 0.35 g
(67%).
4.2. 4-Nitro-AF4 with t-butoxide: 2-nitro-4-(1,1,2,2-tetrafluoro-2-(4-
(1,1,2,2-tetrafluoroethyl)phenyl)ethyl)phenol (1a)
A 50 mL three-necked round bottom flask was charged with 4-
nitro-AF4 (1.20 g, 3.02 mmol), potassium t-butoxide (2.01 g,
17.91 mmol, 5.93 equiv.), 10 mL anhydrous t-butanol, and 10 mL
n-butyl ether. The mixture was heated to 110 8C and was
maintained at this temperature overnight. Upon cooling, the
mixture was acidified to a pH of 5 by dropwise addition of
concentrated HCl. Multiple extractions were performed using
diethyl ether/water and washing with brine. The organic layers
were combined, dried over anhydrous MgSO4 and then filtered.
The solvent was evaporated under vacuum and the residue
purified by silica gel column chromatography eluting initially with
neat hexanes. The product was then eluted incrementally with
increasingly polar mixtures of hexanes/dichloromethane ending in
a ratio of 5:1, respectively. The desired product eluted as a yellow
band. Upon evaporation, the crystals appeared as bright yellow
rosettes. Like fractions were combined, and the phenol product
4.5. 4-Nitro-AF4 with cyanide ion: 2-nitro-4-(1,1,2,2-tetrafluoro-2-
(4-(1,1,2,2-tetrafluoroethyl)phenyl)ethyl)benzonitrile (1b)
In a three necked round bottom flask, bearing a condenser,
sodium cyanide (0.37 g, 7.55 mmol) was stirred in acetonitrile
(10 mL) for 30 min. Then 4-nitro-AF4 (1.01 g, 2.54 mmol),
dissolved in acetonitrile (5 mL), was added slowly to the mixture
drop wise over a period of about 10 min. The reaction medium
turned instantly light yellow at addition of the first drop, later
inducing a strong orange color. The mixture was then heated to
reflux (ꢀ80 8C) for 30 min, checking the efficiency of the reaction
by TLC samples, until the starting material had totally
disappeared (Rf = 0.81 in hexane 2:1 ethyl acetate, compared
to Rf = 0.89 for 4-nitro AF4). At this time, the reaction medium,
which had turned to dark orange, was permitted to cool to room
temperature, and it was then poured into ethyl acetate (30 mL).
The solids in the flask, sodium cyanide in excess, were dissolved
in water (30 mL), which had been poured into the ethyl acetate
mixture. The aqueous layer, brown, was then washed with more
ethyl acetate, leading to an emulsion that needed nearly ten
minutes to be separated. Similar additional washings continued
until no product was seen by TLC. The organic layers were
combined, dried by magnesium sulfate, and filtered, with the
magnesium sulfate salts being washed by ethyl acetate. After
rotary evaporation, a dark brown solid was obtained, and it was
chromatographed (silica gel) using hexane 2:1 ethyl acetate to
give 0.56 g (52%) of light-brown product 1b: mp 139 8C; 1H NMR
(1a) was obtained in a yield of 0.98 g (78%): 1H NMR,
J = 54 and 2.1 Hz, 1H), 7.25 (d, J = 9 Hz, 1H), 7.71 (d, J = 9 Hz, 1H),
7.67 (s, 4H), 8.31 (s, 1H), 10.76 (s, OH); 13C NMR,
110.3 (tt, J = 251
d 5.95 (tt,
d
and 44 Hz), 115.3 (tt, J = 250 and 45 Hz), 115.6 (tt, J = 250 and
45 Hz), 116.1 (tt, J = 250 and 45 Hz), 120.8 (s), 123.0 (t, J = 25 Hz),
124.8 (t, J = 6.6 Hz), 127.0 (t, J = 6 Hz), 127.5 (t, J = 6 Hz), 133.1 (t,
J = 26 Hz), 133.5 (s), 133.5 (t, J = 26 Hz), 135.6 (T, J = 6 Hz), 157.0
(s); 19F NMR,
d
À111.0 (t, J = 9 Hz), À111.6 (t, J = 9 Hz), À114.2 (s),
À134.4 (d, J = 54 Hz); Anal. Calcd. for C16H9F8NO3: C, 46.3; H, 2.2;
N, 3.4. Found: C, 46.5; H, 2.3; N, 3.2.
4.3. 4-Nitro-AF4 with methoxide
A 50 mL three-necked round bottom flask was charged with 4-
nitro-AF4 (1.15 g, 2.90 mmol), sodium methoxide (1.02 g,
18.88 mmol, 6.51 equiv.), 10 mL anhydrous methanol, and 10 mL
n-butyl ether. The mixture was heated to 110 8C and was
maintained at this temperature overnight. Upon cooling, the
mixture was acidified to a pH of 5 by dropwise addition of
concentrated HCl. Multiple extractions were performed using
diethyl ether/water and washing with brine. The organic layers
were combined and were dried over anhydrous MgSO4, then
filtered. The solvent was evaporated under vacuum and the residue
purified by silica gel column chromatography eluting initially with
neat hexanes. The product was then eluted incrementally with
increasingly polar mixtures of hexanes/dichloromethane ending in
a ratio of 5:1, respectively. The product was eluted as a yellow
band. Upon evaporation, the crystals appeared as bright yellow
rosettes. The like fractions were combined and the phenol product
was obtained with a yield of 0.85 g (71%).
(499 MHz, CDCl3),
d
8.56 (s, 1H), 8.10 (d, 3J = 8.0 Hz, 1H), 8.05 (d,
2
3
3J = 8.2 Hz, 1H); 7.74 (s, 4H), 5.99 (tt, JHF = 54 Hz, JHF = 1.8 Hz);
19F NMR (470 MHz, CDCl3),
d
À110.75 (t, JFF = 9.7 Hz), À111.71
3
3
2
(t, JFF = 9.7 Hz), À114.06 (s), À134.24 (d, JHF = 53.9 Hz); 13C
NMR (126 MHz, CDCl3),
d
148.8, 136.8 (t, 2J = 42 Hz), 136.1, 133.5
(t, 2J = 42 Hz), 132.9, 132.5 (t, 2J = 45 Hz), 127.7, 127.2, 124.7,
116.0 (tt, 1J = 253, 2J = 51 Hz), 115.3 (tt, 1J = 250, 2J = 40 Hz), 115.0
(tt, 1J = 255, 2J = 51 Hz) 114.2, 111.1, 110.2 (tt, 1J = 250,
2J = 54 Hz); Anal. Calcd. for C17H8F8N2O2: C, 48.1; H, 1.9; N,
6.6. Found: C, 48.6; H, 1.9; N, 6.2; HRMS m/z calcd. for
C
17H8F8N2O2Na (M+Na): 447.0350, found 447.0364.
4.6. Electrochemistry
The cyclic voltammetry (CV) experiments were performed on
an EG&G Potentiostat/Galvanostat (Model 273A). All experiments
were carried out with 5 mM solution in CH3CN containing 0.1 M
tetra-n-butylammonium perchlorate (TBAP) as supporting elec-
trolyte, and were performed with a 0.02 cm2 Pt button working
electrode, a Pt wire counter electrode and a silver wire pseudo
reference electrode (calibrated with ferrocene and expressed as V
versus SCE) in an argon filled dry box. The half wave potential E1/2
refers the saturated calomel Ag wire as reference electrode, the
system was calibrated with ferrocene/ferrocenium (Fc/Fc+) redox
system (E1/2(FC) = 0.3927 V versus SCE.
4.4. 4-Nitro-AF4 with 2,2,2-trifluoroethoxide
A 50 mL three-necked round bottom flask was charged with
sodium hydride (0.43 g, 17.92 mmol), 10 mL anhydrous THF, and
then 2,2,2-trifluoroethanol (1.30 mL, 1.81 g, 18.09 mmol). The
mixture was heated to 80 8C and was maintained at this
temperature for 2 h. To this was added 4-nitro-AF4 (0.50 g,
1.26 mmol) and 10.00 mL anhydrous ethanol and the reaction
run an additional 6 h at 80 8C. Upon cooling, the mixture
was acidified to a pH of 5 by dropwise addition of concentrated

1138
W.R. Dolbier Jr. et al. / Journal of Fluorine Chemistry 129 (2008) 1133–1138
[4] K. Wu, W.R. Dolbier Jr., M.A. Battiste, Y.-A. Zhai, Mendeleev Commun. (2006) 146–
Acknowledgment
147.
[5] H.J. Reich, D.J. Cram, J. Am. Chem. Soc. 91 (1969) 3517–3526.
[6] T. Sato, K. Torizuka, M. Shimizu, Y. Kurihara, N. Yoda, Bull. Chem. Soc. Jpn. 52
(1979) 2420–2423.
[7] J.L. Marshall, T.K. Folson, Tetrahedron Lett. 12 (1971) 757–760.
[8] W. Adam, M.A. Miranda, F. Mojarrad, H. Sheikh, Chem. Ber. 127 (1994) 875–
879.
The authors are grateful to Specialty Coating Systems, Inc. for a
generous donation of AF4.
References
[9] S. Sankararaman, H. Hopf, I. Dix, P.G. Jones, Eur. J. Org. Chem. (2000) 2711–
2716.
[10] T.D. Quach, R.A. Batey, Org. Lett. 5 (2003) 1381–1384.
[11] X.Y. Xu, X.H. Qian, Z. Li, G.H. Song, J. Fluorine Chem. 88 (1998) 9–13.
[1] A.J. Roche, W.R. Dolbier Jr., J. Org. Chem. 64 (1999) 9137–9143.
[2] A.J. Roche, W.R. Dolbier Jr., J. Org. Chem. 65 (2000) 5282–5290.
[3] W.R. Dolbier Jr., Y.-A. Zhai, W. Wheelus, M.A. Battiste, I. Ghiviriga, M.D. Bartberger,
J. Org. Chem. 72 (2007) 550–558.