The trifluoromethoxy group as a fluorine twin in the Diels-Alder reactions of halogenated quinones

Article abstract of DOI:10.1016/j.tetlet.2008.05.104

We describe here a study devoted to the comparison of the relative influence of chlorine, fluorine, and trifluoromethoxy substituents on the regiochemical outcome of the Diels-Alder reaction. For this purpose, we examined the behavior of mixed 'halogenated' quinones bearing these groups in their cycloadditions with simple dienes. Contrary to the expectation based on its known electronic properties, the trifluoromethoxy group behaves very much more like a fluorine than a chlorine atom in such reactions. On the basis on an endo transition state demonstrated here for these additions, we tentatively suggest that non-bonded interactions are the main factor controlling the regiochemistry.

Full text of DOI:10.1016/j.tetlet.2008.05.104

Tetrahedron Letters 49 (2008) 4575–4578
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The trifluoromethoxy group as a fluorine twin in the Diels–Alder
reactions of halogenated quinones
*
Emmanuel Magnier, Patrick Diter, Jean-Claude Blazejewski
Institut Lavoisier de Versailles, UMR CNRS 8180, Université de Versailles St Quentin en Yvelines, 45 avenue des Etats-Unis, 78035 Versailles Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
We describe here a study devoted to the comparison of the relative influence of chlorine, fluorine, and
trifluoromethoxy substituents on the regiochemical outcome of the Diels–Alder reaction. For this
purpose, we examined the behavior of mixed ‘halogenated’ quinones bearing these groups in their cyc-
loadditions with simple dienes. Contrary to the expectation based on its known electronic properties, the
trifluoromethoxy group behaves very much more like a fluorine than a chlorine atom in such reactions.
On the basis on an endo transition state demonstrated here for these additions, we tentatively suggest
that non-bonded interactions are the main factor controlling the regiochemistry.
Received 16 April 2008
Revised 20 May 2008
Accepted 21 May 2008
Available online 27 May 2008
Keywords:
Fluorine
Trifluoromethoxy
Trifluoromethyl ethers
Diels–Alder
Ó 2008 Elsevier Ltd. All rights reserved.
The trifluoromethoxy group enjoys a great interest among fluo-
rinated group in various applications thanks to its peculiar proper-
ties.1 However, besides the field of aromatic compounds,2 its
chemistry is poorly known mainly because non-aromatic trifluoro-
methyl ethers are far less common than aromatic ones.3,4 It has
been demonstrated that, at least in the aromatic series, its elec-
tronic properties lie between that of a chlorine and a fluorine
atom.2,5 Although some trifluoromethoxy substituted alkenes are
known,4,6 to our knowledge, there is no report on their reactivity
as dienophilic components in Diels–Alder reactions. Here, we pres-
ent our results concerning the regio- and stereochemical outcome
of the Diels–Alder addition of some halogenated quinones with
simple symmetrical dienes, allowing a comparison between ’halo-
gen’ substituents (chloro, fluoro, and trifluoromethoxy).
In order to overcome the leveling effect resulting from the
strong tendency of fluorobenzoquinone to give mainly addition
products toward its hydrogenated side under kinetic conditions,7,8
we selected to study the behavior of mixed ‘halogenated’ quinones
(2-chloro-6-fluoro and 2-chloro-6-trifluoromethoxy benzoquinon-
es) toughed to enable a finer analysis of the relative effects of these
halogen atoms.
O
O
OH
OH
SO2Cl2
Cl
X
X
Cl
NaClO2
H2SO4
X
i
Bu2NH
toluene
RT
1
3
2
a X = F
b X = OCF3
Scheme 1. Synthesis of quinones.
of 0 °C. Quinones 3a and 3b were obtained, respectively, in 31%
and 53% yield.12 Their Diels–Alder reactions with dimethylbutadi-
ene 4 or cyclopentadiene 7 occurred smoothly in dichloromethane
at room temperature over 24 h.13 After removal of the solvent, the
adducts were obtained in essentially quantitative crude yields as a
mixture of regioisomers 5 and 6 (or 8 and 9, respectively)14 (Table
1). These adducts proved unstable under the light as well as under
slightly acidic conditions, but can be kept safely at low tempera-
ture in the dark for several weeks. Attempted chromatographic
separations resulted in considerable decomposition and loss of
material.15
Structure determination relied primarily on NMR spectroscopy.
The distinction between the regioisomers derived from fluoroqui-
none 3a was straightforward: signals for the vinylic fluorine atoms
in 5a and 8a appeared at À109.5 and À107.0 ppm in the 19F NMR
spectra (compare with À110.2 ppm for quinone 3a), whereas those
for tertiary fluorine atoms in 6a and 9a were found at À153.2 and
À141.7 ppm. Moreover, the vinylic hydrogen atoms (H3 in 5a and
H7 in 8a) showed a coupling constant in the range 11–12 Hz with
Quinones 3 were readily obtained by selective ortho chlorina-
tion of phenols 19,10 followed by direct oxidation of chlorophenols
2 with sodium chlorite in an acidic medium using our earlier de-
scribed method for trifluoromethylphenols (Scheme 1).11
Oxidation of 2-chloro-6-fluorophenol 2a proved to be highly
exothermic and was consequently performed at À10 °C instead
* Corresponding author. Tel.: +33 1 3925 4465; fax: +33 1 3925 4452.
E-mail address: jcblaz@chimie.uvsq.fr (J.-C. Blazejewski).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.05.104

4576
E. Magnier et al. / Tetrahedron Letters 49 (2008) 4575–4578
Table 1
Regioselectivity of the cycloadditions
O
O
X
Cl
H
X
Cl
3, CH2Cl2
H8a (8b)
8b + 9b
+
3
RT, 24h
H
4
H4 (8b)
5
O
O
O
6
a X = F
b X = OCF3
H1 (8b)
O
H1 (9b)
H4 (9b)
3.10
Cl
X
H
H8a (9b)
4
3, CH2Cl2
X
Cl
4a
8a
+
7
RT, 24h
1
H
7
O
O
8
9
3.30
3.20
3.00
2.90
Entry
X (quinone 3)
Diene
Adducts (ratio)
Figure 2. Partial (tertiary protons region) 1H NMR spectrum (C6D6, 300 MHz) of
adducts mixture 8b and 9b.
1
2
3
4
F (3a)
OCF3 (3b)
F (3a)
4
4
7
7
5a:6a (95:5)
5b:6b (91:9)
8a:9a (78:22)
8b:9b (80:20)
OCF3 (3b)
exo transition state. 1H NMR spectra run in CDCl3 were useless for
this purpose, very important signals from H1, H4, and H8a being
highly crowded in a very narrow range (3.5–3.6 ppm). Spectra
run in C6D6 were more informative, signals from H8a could be
unambiguously located in all adducts (8a, d = 3.06; 9a, d = 2.74;
8b, d = 3.06; 9b, d = 3.14 ppm)18 and appeared as clean doublets
with a coupling constant in the range 3.9–4.4 Hz with H1 (see
Figs. 1 and 2).19 Such a coupling constant is highly diagnostic of
an exo position for H8a.20 Consequently all cycloadditions with
cyclopentadiene reported here occurs via the usual endo transition
state.
Considering the very mild conditions used, it can be safely
assumed that no retro Diels–Alder reactions occurred during these
cycloadditions. The ratio of regioisomers shown in Table 1 con-
cerns thus primary adducts and not an equilibrated mixtures of
regioisomers. On this ground, the regioselectivity observed in these
normal electron demand Diels–Alder reactions seems somewhat
amazing. Based on electronic arguments alone (inductive and res-
onance effects of the substituents),2 we should have observed an
intermediate behavior of the trifluoromethoxy group lying some-
where between that of a chlorine or a fluorine atom but not so
close to the latter.21
If an endo transition state could be assumed, as deduced above
for cyclopentadiene, during reactions of both dienes studied here,
it may be speculated that a chlorine atom may experience more se-
vere interactions with the diene substituents in such transition
state than the smaller fluorine atom. This may explain the results
observed with chlorofluorobenzoquinone 3a. But could this expla-
nation also apply for the presumably bulkier trifluoromethoxy
group? Obviously the Van der Waals volume of OCF3 could be con-
sidered far bigger than that of a fluorine atom. However, the con-
formational degree of freedom exhibited by the ether linkage,
may largely compensate this bulkiness handicap in some selected
situations.22 Moreover, contrary to the methoxy group, the OCF3
substituent is known to adopt a perpendicular position out of the
plane of an aromatic ring.1,6,23 One can tentatively assume that this
situation also hold for the case of trifluoromethoxybenzoquinones.
On this basis, it can be presumed that steric interactions with a tri-
fluoromethylether are mainly restricted to the oxygen atom.24
Knowing the nearly isosteric relationship between fluorine and
oxygen,25 we propose that, under such conditions, the trifluoro-
methoxy group and the fluorine atom behave similarly, their
comparable steric requirements overwhelming the relatively small
disparity of their electronic properties.
the vinylic fluorine atom, while these nuclei appeared as broad sing-
lets or faint doublets in adducts 6a and 9a. As expected, the fluorine
chemical shift difference between the adducts bearing a trifluoro-
methoxy group was less clear cut, but remained sufficiently large
to discriminate a vinylic OCF3 (d À58.6 to À58.8 ppm) from a tertiary
OCF3 group (d À51.6 to À52.2 ppm). Again these assignments were
corroborated by the coupling pattern of the vinylic atoms appearing
as quadruplets (JFH ca. 2 Hz) in 5b and 8b, and as singlets in 6b and
9b. Based on these assignments, and the relative integration of the
vinylic protons in the 1H NMR spectra, the regiochemical outcome
of the Diels–Alder reactions studied here is presented in Table 1.
It is clearly apparent from the data gathered in Table 1 that the
trifluoromethoxy group, in these cycloadditions, behaves very
much like a fluorine than a chlorine atom. Adducts 5 and 8 bearing
the chlorine atom in a tertiary position are predominantly ob-
tained. The selectivity observed is somewhat lower with the more
reactive cyclopentadiene and noticeably reversed between fluorine
and trifluoromethoxy substituents.
In the case of cyclopentadiene adducts, the problem of the endo
versus exo addition of quinones had to be resolved. Usually Diels–
Alder reactions of common quinones occur exclusively via an endo
transition state.16 It has been reported, however, that some fluori-
nated dienophiles8,17 gave sometimes the products arising from an
8a + 9a
H8 (8a)
H4 (8a)
H1 (8a + 9a)
H4 (9a)
H8 (9a)
3.30
3.20
3.10
3.00
2.90
2.80
2.70
Figure 1. Partial (tertiary protons region) 1H NMR spectrum (C6D6, 300 MHz) of
adducts mixture 8a and 9a.
We tried to rationalize these unexpected results by FMO calcu-
lations. In fact, such kind of computations have been performed

E. Magnier et al. / Tetrahedron Letters 49 (2008) 4575–4578
4577
and pooled with the hexane extracts (3 Â 100 mL) of the resulting aqueous
phase. The organic phase was dried (MgSO4), filtered, and allowed to crystallize
at À30 °C. The quinone 3a or 3b was obtained in three crops by filtration
followed each time by slight concentration of the hexane phase and recooling.
2-Chloro-6-fluoro-1,4-benzoquinone (3a): Yield 31%; mp 84–85 °C. 19F NMR
(CDCl3, 188 MHz) d: À110.2 (d, J = 9 Hz); 1H NMR (CDCl3, 300 MHz) d: 7.01 (d,
1H, J = 2.4 Hz), 6.50 (dd, 1H, J = 8.8, 2.4 Hz); 13C NMR (CDCl3, 75 MHz) d: 183.8
(d, J = 14 Hz), 172.4 (d, J = 25 Hz,), 159.6 (d, J = 292 Hz), 142.1 (d, J = 7 Hz), 133.8
(s), 115.5 (d, J = 10 Hz). HRMS: calculated for C6H235ClFO 159.9722; found:
159.97143 d À4.7 ppm.
with some success on related fluorinated systems (but not includ-
ing the trifluoromethoxy group).26 Our own FMO calculations at
the B3LYP/6-31G* level gave qualitative results in accordance with
the experimental results for quinone 3a but failed with the triflu-
oromethoxy substituted quinone 3b. Indeed, very recently, Lemal
caution for the risk associated with such ‘prediction about reaction
pathways’ for Diels–Alder reactions of o-Fluoranil.27
Thus, to date, explanations developed above may be considered
highly speculative and will constitute a challenge for more sophis-
ticated calculations.
Nevertheless, we have shown in this work, that in the Diels–Al-
der reaction of benzoquinones with simple dienes, the trifluoro-
methoxy group behaves like a fluorine twin. We think that these
results may prove highly useful for those who plan to introduce
an OCF3 group in a molecule by the means of a Diels–Alder
reaction.
2-Chloro-6-trifluoromethoxy-1,4-benzoquinone (3b): Yield 53%; mp 88–89 °C.
19F NMR (CDCl3, 188 MHz) d: À58.9 (d, J = 2 Hz); 1H NMR (CDCl3, 300 MHz) d:
7.05 (d, 1H, J = 2.3 Hz), 6.56 (quint, 1H, J = 2.3 Hz); 13C NMR (CDCl3, 75 MHz) d:
183.2, 172.0, 148.8 (q, J = 1 Hz), 142.7, 133.6, 119.9 (q, J = 265 Hz), 118.4 (q,
J = 2 Hz). HRMS: calculated for C7H235ClF3O3 225.9639; found 225.96349 d
À1.8 ppm.
13. No efforts were made to check the optimal reaction time.
14. General procedure for Diels–Alder reactions. A solution of the quinone 3a or 3b
(2.2 mmol) and a slight excess of the appropriate diene 4 or 7 (2.6 mmol) in
dichloromethane (25 mL), shielded from light with an aluminum foil, was
allowed to stir for 24 h under an argon atmosphere. After removal of the
solvent, a mixture of the adducts was obtained either as a pale yellow powder
(8a and 9a) or as an oil (mixture of 8b and 9b) in essentially quantitative yield.
8a-Chloro-2-fluoro-6,7-dimethyl-4a,5,8,8a-tetrahydro-[1,4]naphthoquinone (5a)
(major isomer): 19F NMR (CDCl3, 188 MHz) d: À109.5 (ddd, J = 10.9, 2.9,
1.6 Hz). 1H NMR (CDCl3, 300 MHz) d: 6.34 (ddd, 1H, J = 10.9, 1.3, 0.8 Hz), 3.42
(dddd, 1H, J = 10.9, 9.5, 7.3, 1.4 Hz), 3.02 and 2.48 (AB system, 2H, J = 16.9 Hz),
2.2–2.4 (m, 2H), 1.64 (m, 3H), 1.57 (m, 3H). 13C NMR (CDCl3, 75 MHz) d: 195.1
(d, J = 13 Hz), 183.6 (d, J = 22 Hz), 160.3 (d, J = 296 Hz), 122.7, 123.3, 116.6 (d,
J = 8 Hz), 67.4 (d, J = 5 Hz), 55.7, 38.8 (d, J = 2 Hz), 33.3, 18.5, 17.9. HRMS:
calculated for C12H1235ClFO2 242.0504; found 242.04996 d À2.0 ppm.
2-Chloro-8a-fluoro-6,7-dimethyl-4a,5,8,8a-tetrahydro-[1,4]naphthoquinone (6a)
(minor isomer): 19F NMR (CDCl3, 188 MHz) d: À153.2 (ddd, J = 29.5, 19, 9 Hz).
1H NMR (CDCl3, 300 MHz) d: 7.05 (s, 1H); other signals are obscured by signals
from the major isomer. 13C NMR (CDCl3, 75 MHz) d: 192.1 (d, J = 10 Hz), 187.3
(d, J = 21 Hz), 145.8 (d, J = 3 Hz), 137.6, 123.7 (d, J = 1 Hz), 120.0 (d, J = 2 Hz),
94.3 (d, J = 191 Hz), 51.6 (d, J = 21 Hz), 36.0 (d, J = 24 Hz), 27.7 (d, J = 2 Hz), 18.4.
8a-Chloro-2-trifluoromethoxy-6,7-dimethyl-4a,5,8,8a-tetrahydro-[1,4]naphtho-
quinone (5b) (major isomer): 19F NMR (CDCl3, 188 MHz) d: À58.6 (d, J = 2.3 Hz).
1H NMR (CDCl3, 300 MHz) d: 6.39 (m, 1H); 3.46 (ddd, 1H, J = 10.0, 7.1, 1.3 Hz);
3.05 and 2.51 (AB system, 2H, J = 17.2 Hz), 2.2–2.5 (m, 2H); 1.67 (br s, 3 H);
1.60 (br s, 3H). 13C NMR (CDCl3, 75 MHz) d:194.8, 183.2, 149.7 (q, J = 1 Hz),
123.7, 122.5, 120.1 (q, J = 264 Hz), 119.7 (q, J = 1 Hz), 67.5, 55.5, 39.1, 33.7, 18.6,
Acknowledgements
We gratefully thank Nathalie Toulemonde for preliminary
experiments, Violaine Chapuis and Sandrine Blondel for technical
assistance as well as Sami Lakhdar for preliminary computations.
References and notes
1. For
a recent review including preparation and properties of trifluoro-
methylethers, see: Leroux, F.; Jeschke, P.; Schlosser, M. Chem. Rev. 2005, 105,
827–856 and references cited therein.
2. Olah, G. A.; Yamato, T.; Hashimoto, T.; Shih, J. G.; Trivedi, N.; Singh, B. P.; Piteau,
M.; Olah, J. A. J. Am. Chem. Soc. 1987, 109, 3708–3713.
3. For some aspects of the chemical reactivity of alkyl trifluoromethylethers, see:
(a) Aldrich, P. E.; Sheppard, W. A. J. Org. Chem. 1964, 29, 11–15; (b) Barton, D. H.
R.; Hesse, R. H.; Jackman, G. P.; Pechet, M. M. J. Chem. Soc., Perkin Trans. 1 1977,
2604–2608 and references cited therein; (c) Yagupol’skii, L. M.; Alekseenko, A.
N.; Il’chenko, A. Ya. Ukr. Khim. Zh. 1978, 44, 52–55; (d) Kamil, W. A.; Haspel-
Hentrich, F.; Shreeve, J. M. Inorg. Chem. 1986, 25, 376–380; (e) Blazejewski, J.-
C.; Anselmi, E.; Wakselman, C. J. Org. Chem. 2001, 66, 1061–1063; (f) Agouridas,
V.; La¨ıos, I.; Cleeren, A.; Kizilian, E.; Magnier, E.; Blazejewski, J.-C.; Leclercq, G.
Bioorg. Med. Chem. 2006, 14, 7531–7538.
4. (a) Barton, D. H. R.; Hesse, R. H.; Jackman, G. P.; Ogunkoya, L.; Pechet, M. M. J.
Chem. Soc., Perkin Trans. 1 1974, 739–742; (b) Trofimenko, S.; Johnson, R. W.;
Doty, J. K. J. Org. Chem. 1978, 43, 43–47; (c) Krespan, C. G.; Smart, B. E. J. Org.
Chem 1986, 51, 320–326; (d) Blazejewski, J.-C.; Anselmi, E.; Wernicke, A.;
Wakselman, C. J. Fluorine Chem. 2002, 117, 161–166; (e) Lebedev, N. V.;
Berenblit, V. V.; Starobin, Yu K.; Gubanov, V. A. Russ. J. Appl. Chem. 2005, 78,
1640–1645.
5. Cacace, F.; Crestoni, M. E.; Di Marzio, A.; Fornarini, S. J. Phys. Chem. 1991, 95,
8731–8737.
6. Brey, W. S. J. Fluorine Chem. 2005, 126, 389–399.
7. Ansell, M. F.; Nash, B. W.; Wilson, D. A. J. Chem. Soc. 1963, 3012–3028.
8. Essers, M.; Haufe, G. J. Chem. Soc., Perkin Trans. 1 2002, 2719–2728.
9. Gnaim, J. M.; Sheldon, R. A. Tetrahedron Lett. 1995, 36, 3893–3896.
10. 2-Chloro-6-fluorophenol 2a is commercially available.
18.0. HRMS: calculated for
C d
13H1235ClF3O3 308.0422; found 308.04207
À0.3 ppm.
2-Chloro-8a-trifluoromethoxy-6,7-dimethyl-4a,5,8,8a-tetrahydro-[1,4] naphthoq-
uinone (6b) (minor isomer): 19F NMR (CDCl3, 188 MHz) d: À52.2 (d, J = 1.3 Hz).
1H NMR (CDCl3, 300 MHz) d: 6.94 (q, 1H), 3.36 (br t, 1H, J = 7.7 Hz), 2.86 (d, 1H,
J = 17.3 Hz), 1.62 (br s, 3H) other signals are obscured by signals from the major
isomer. 13C NMR (CDCl3, 75 MHz) d: 193.2, 186.0, 145.2, 136.0, 121.7, 122.7,
84.9, 52.5, 35.6 (q, J = 1 Hz), 31.1, 18.7, 18.2.
4a-Chloro-6-fluoro-1,4,4a,8a-tetrahydro-1,4-methano-naphthalene-5, 8-dione (8a)
(major isomer): 19F NMR (CDCl3, 188 MHz) d: À107.0 (d, J = 11.1 Hz). 1H NMR
(CDCl3, 300 MHz) d: 6.43 (d, 1H, J = 11.1 Hz), 6.19 (dd, 1H, J = 5.7, 2.5 Hz), 6.04
(dd, 1H, J = 5.7, 3.1 Hz), 3.60–3.50 (m, 3H), 2.08 (br d, 1H, J = 9.5 Hz), 1.85 (dt,
1H, J = 9.5, 1.6 Hz). 1H NMR (C6D6, 300 MHz) d: 5.68 (d, 1H, J = 11.4 Hz), 5.57
(dd, 1H, J = 5.7, 2.8 Hz), 5.44 (dd, 1H, J = 5.6, 3.1 Hz), 3.25 (m, 1H), 3.06 (d, 1H,
J = 3.9 Hz), 2.98 (m, 1H); 1.52 (br d, 1H, J = 9.5 Hz), 1.18 (dt, 1H, J = 9.5, 1.6 Hz).
13C NMR (CDCl3, 75 MHz) d: 194.5 (d, J = 14 Hz), 185.0 (d, J = 21 Hz), 162.9 (d,
J = 296 Hz), 138.7, 134.6, 122.1 (d, J = 10 Hz), 68.4 (d, J = 6 Hz), 62.8, 55.1 (d,
J = 2 Hz), 47.8, 47.1. HRMS: calculated for
226.01808 d À4.7 ppm.
C
11H835ClFO2 226.0191; found
Procedure for 2-chloro-6-trifluoromethoxyphenol (2b) Sulfuryl chloride (9 mL,
6-Chloro-4a-fluoro-1,4,4a,8a-tetrahydro-1,4-methano-naphthalene-5, 8-dione (9a)
(minor isomer): 19F NMR (CDCl3, 188 MHz) d: À141.7 (dd quint, J = 28.2, 6.6,
2 Hz). 1H NMR (CDCl3, 300 MHz) d: 6.95 (s, 1H), 6.27 (dt, 1H,J = 5.6, 2.3 Hz),
5.26 (dt, 1H, J = 5.6, 2.2 Hz), 3.42 (m, 1H), 3.26 (dd, 1H, J = 28.2, 4.0 Hz); 2.00
(dq, 1H, J = 9.4, 1.4 Hz), 1.82 (dq, 1H, J = 9.4, 1.5 Hz). 1H NMR (C6D6, 300 MHz)
d: 6.28 (s, 1H), 5.62 (dt, 1H, J = 5.8 Hz, J = 2.3 Hz), 5.96 (dt, 1H, J = 5.7 Hz,
J = 2.9 Hz), 3.12 (m, 1H); c.a. 2.96 (m, 1H), 2.74 (dd, 1H, J = 28.5 Hz, J = 4.0 Hz),
c.a. 1.49 (m, 1H); c.a. 1.18 (m, 1H). 13C NMR (CDCl3, 75 MHz) d: 193.2 (d,
J = 5 Hz), 185.8 (d, J = 21 Hz), 148.8 (d, J = 4 Hz), 140.8 (d, J = 3 Hz), 140.6, 132.1
(d, J = 8 Hz), 97.3 (d, J = 201 Hz), 58.1 (d, J = 21 Hz), 53.0 (d, J = 24 Hz), 47.3 (d,
J = 2 Hz), 47.0. HRMS: calculated for C11H835ClFO2 226.0191; found 226.01826
d À3.9 ppm.
112 mmol) was added dropwise to
a hot (70 °C) stirred solution of 2-
trifluoromethoxyphenol 1b (20 g, 112 mmol) in toluene (250 mL) containing
diisobutylamine (0.156 mL, 0.9 mmol). After 1 h stirring at 70 °C, and cooling,
most of the solvent was removed by rotatory evaporation. The residue was
diluted with ether (200 mL) and washed twice with 0.3 M sulfuric acid. After
drying (MgSO4) and removal of solvent, 21 g (99 mmol, 88% yield) of
chlorophenol 2b was obtained as an oil for suitable further oxidation. 19F
NMR (CDCl3, 188 MHz) d: À58.8 (d, J = 1.4 Hz); 1H NMR (CDCl3, 300 MHz) d:
7.30 (dd, 1H, J = 8.2, 1.5 Hz), 7.18 (d quint, J = 8.3, 1.5 Hz), 6.89 (t, 1H,
J = 8.2 Hz), 5.73 (s, 1H). 13C NMR (CDCl3, 75 MHz) d: 144.9, 137.1, 127.9, 121.7,
121.1, 120.6 (q, J = 259 Hz), 120.4. HRMS: calculated for C7H335ClFNa2O2
256.9573; found: 256.9596 d À1.4 ppm.
4a-Chloro-6-trifluoromethoxy-1,4,4a,8a-tetrahydro-1,4-methano-naphthalene-5,8-
dione (8b) (major isomer): 19F NMR (CDCl3, 188 MHz) d: À58.6 (d, J = 2.1 Hz).
1H NMR (CDCl3, 300 MHz) d: 6.51 (q, 1H, J = 2.1 Hz), 6.23 (dd, 1H, J = 5.6,
2.1 Hz), 6.07 (dd, 1H, J = 5.6, 3.0 Hz), 3.68—3.43 (m, 3H); 2.11 (br dq, 1H, J = 9.6,
ca 1Hz), 1.87 (br dt, 1H, J = 9.6, 1.7 Hz). 1H NMR (C6D6, 300 MHz) d: 6.11 (q, 1H,
J = 1.8 Hz), 5.61 (dd, 1H, J = 5.7, 2.8 Hz), 5.50 (dd, 1H, J = 5.7, 3.1 Hz), 3.26 (m,
1H); 3.06 (d, 1H, J = 3.9 Hz), 2.99 (m, 1H), 1.52 (br d, 1H, J = 9.5 Hz), 1.19 (dt, 1H,
J = 9.5, 1.7 Hz). 13C NMR (CDCl3, 75 MHz) d: 193.9, 184.5, 152.5 (q, J = 1Hz),
138.8, 135.0, 124.8 (q, J = 2 Hz), 119.9 (q, J = 265 Hz), 68.4, 62.4, 55.0, 47.6, 47.0.
HRMS: calculated for C12H835ClF3O3 292.0109; found 292.00995 d À3.1 ppm.
6-Chloro-4a-trifluoromethoxy-1,4,4a,8a-tetrahydro-1,4-methano-naphthalene-5,8-
dione (9b) (minor isomer): 19F NMR (CDCl3, 188 MHz) d: À51.6 (s).1H NMR
11. (a) Blazejewski, J.-C.; Dorme, R.; Wakselman, C. Synthesis 1985, 1120–1121; (b)
Blazejewski, J.-C.; Dorme, R.; Wakselman, C. J. Chem. Soc., Perkin Trans. 1 1987,
1861–1864.
12. General procedure for the preparation of quinones 3: A solution of sodium
chlorite hydrate (320 mmol) in water (85 mL) was added at once to
a
(magnetically) well-stirred suspension of the chlorophenol 2a or 2b
(80 mmol) in sulfuric acid (0.3 M, 270 mL) in an open flask cooled in an ice
bath (an ice salt bath was used for phenol 2a). The temperature quickly rose to
30–35 °C with strong evolution of chlorine dioxide vapors. The mixture was
stirred in the cold for an additional hour after the exotherm had ceased, then
degassed with continuous stirring (ice bath still present) under the vacuum of
a water aspirator. The bright yellow crystals formed were collected by filtration

4578
E. Magnier et al. / Tetrahedron Letters 49 (2008) 4575–4578
(CDCl3, 300 MHz) d: 6.95 (s, 1H); 6.30 (dd, 1H, J = 5.8, 2.7 Hz), 5.99 (dd, 1H,
19. In the case of 9a the signal for H8a was further splitted (J = 28.2 Hz) by
coupling with the vicinal fluorine atom, pointing thus to a cis relationship
between these nuclei (see Ref. 8).
20. Laszlo, P.; von Ragué Schleyer, P. J. Am. Chem. Soc. 1964, 86, 1171–1179.
21. Rozeboom, M. D.; Tegmo-Larsson, I.-M.; Houk, K. N. J. Org. Chem. 1981, 46,
2338–2345.
J = 5.8, 3.2 Hz), 3.68—3.43 (m, 3H,), 2.01 (br dt, 1H, J = 9.6, 1.5 Hz), 1.80 (br dt,
1H, J = 9.6, 1.6 Hz). 1H NMR (C6D6, 300 MHz) d: 6.36 (s, 1H); 5.55 (dd, 1H,
J = 5.7, 2.8 Hz), 5.28 (dd, 1H, J = 5.7, 3.2 Hz), 3.14 (d, 1H, J = 4.4 Hz), 3.11 (m,
1H), 2.91 (m, 1H); 1.35 (br d, 1H, J = 9.5 Hz), 1.06 (br dt, 1H, J = 9.6, 1.6 Hz). 13
C
NMR (CDCl3, 75 MHz) d: 193.0, 185.8, 148.9, 140.7, 139.6, 132.3, 121.2 (q,
J = 260 Hz), 86.6, 56.8 (q, J = 1 Hz), 55.1, 48.9, 47.6. HRMS: calculated for
22. Dimitrov, A.; Groß, U.; Rüdiger, St.; Storek, W.; Burdon, J. J. Fluorine Chem. 1996,
78, 1–5.
C
12H835ClF3O3 292.0109; found 292.01142 d 1.9 ppm.
15. The main products isolated resulted from elimination of the bridgehead
‘halogen’ substituent, with eventual further aromatization in the case of
dimethylbutadiene adducts.
23. (a) Leibold, C.; Reinemann, S.; Minkwitz, R.; Resnik, P. R.; Oberhammer, H. J.
Org. Chem. 1997, 62, 6160–6163; (b) Radice, S.; Tortelli, V.; Causà, M.;
Castiglioni, C.; Zerbi, G. J. Fluorine Chem. 1999, 95, 105–116; (c) Böhm, H.-J.;
Banner, D.; Bendels, S.; Kansy, M.; Kuhn, B.; Müller, K.; Obst-Sander, U.; Stahl,
M. ChemBioChem 2004, 5, 637–643.
24. Carcenac, Y.; Tordeux, M.; Wakselman, C.; Diter, P. New J. Chem. 2006, 30, 447–
457.
25. O’Hagan, D.; Rzepa, H. S. J. Chem. Soc., Chem. Commun. 1997, 645–652.
26. (a) Essers, M.; Mück-Lichtenfeld, C.; Haufe, G. J. Org. Chem. 2002, 67, 4715–
4721; (b) Hajduch, J.; Paleta, O.; Kv´ıca˘la, J.; Haufe, G. Eur. J. Org. Chem. 2007,
5101–5111.
´
16. Garcıa, J. I.; Mayoral, J. A.; Salvatella, L. Acc. Chem. Res. 2000, 33, 658–664.
17. (a) Wilson, R. M. J. Org. Chem. 1983, 48, 707–711. This Letter is in favor of an
endo transition state for the reaction of tetrafluoro-p-benzoquinone with
cyclopentadiene; (b) Crowley, P. J.; Percy, J. M.; Stansfield, K. Tetrahedron Lett.
1996, 45, 8237–8240; (c) Essers, M.; Birgit Wibbeling, B.; Haufe, G. Tetrahedron
Lett. 2001, 42, 5429–5433.
18. These values are concentration dependent. Consequently, concentrations of the
samples were adjusted to give the maximum dispersion of the signals of
interest in the 1H NMR spectra.
27. Lemal, D. M.; Ramanathan, S.; Shellito, J. J. Org. Chem. 2008, 73, 3392–3396.