Conformations of allylic fluorides and stereoselectivities of their Diels-Alder cycloadditions

Article abstract of DOI:10.1021/jo0016024

The preparations of new allylic fluorides from the corresponding alcohols are reported. Conformational analysis is achieved by comparison of experimental NMR measurements with theoretical (B3LYP) calculations of relative energies of conformers and J_H,H and J_H,F coupling constants. The Diels-Alder reactions of allylic fluorides are investigated experimentally and theoretically. The stereoselectivities of the reactions were determined by NMR analysis and, in one case, by X-ray crystallography. Theoretical predictions of stereoselectivity based upon transition state modeling provided good agreement with experiment. Theoretical models for allylic fluorides and transition state conformations are reported.

Full text of DOI:10.1021/jo0016024

2374
J . Org. Chem. 2001, 66, 2374-2381
Con for m a tion s of Allylic F lu or id es a n d Ster eoselectivities of Th eir
Diels-Ald er Cycloa d d ition s
Danielle Gre´e, Laurent Vallerie, and Rene´ Gre´e*
Laboratoire de Syntheses et Activations de Biomolecules, associated with CNRS Ecole Nationale
Superieure de Chimie de Rennes, 35700 Rennes, France
Loic Toupet
Groupe Matiere Condense´e et Materiaux, associated with CNRS, Universite´ de Rennes,
Avenue du General Leclerc, 35042 Rennes, France
Ilyas Washington, J ean-Pierre Pelicier, Mercedes Villacampa, J ose´ Mar´ıa Pe´rez, and
K. N. Houk*
Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569
houk@chem.ucla.edu
Received November 9, 2000
The preparations of new allylic fluorides from the corresponding alcohols are reported. Conforma-
tional analysis is achieved by comparison of experimental NMR measurements with theoretical
(B3LYP) calculations of relative energies of conformers and J H,H and J H,F coupling constants. The
Diels-Alder reactions of allylic fluorides are investigated experimentally and theoretically. The
stereoselectivities of the reactions were determined by NMR analysis and, in one case, by X-ray
crystallography. Theoretical predictions of stereoselectivity based upon transition state modeling
provided good agreement with experiment. Theoretical models for allylic fluorides and transition
state conformations are reported.
In tr od u ction
preparation of allylic fluorides probably accounts for the
surprising lack of information about such systems.
We report here the preparation of new electrophilic
alkenes, 1 and 2, selected as prototypes of allylic fluo-
rides. By use of experiment and quantum mechanical
methods, we have explored the conformations of these
molecules and the stereoselectivities of their Diels-Alder
cycloaddition reactions. These results establish the rela-
tive roles of steric and electronic effects on the reaction
stereoselectivities of allylic fluorides and provide general
rules for the prediction.
The stereoselectivities of additions to chiral alkenes are
of wide interest, both from a mechanistic and synthetic
point of view.1 Introduction of fluorine in organic mol-
ecules induces significant modifications of their physical,
chemical, and biological properties due to the small size
and very high electronegativity of this atom.2 However,
very little is known about the influence of allylic fluorines
on addition stereoselectivities. For relatively simple
systems suitable for modeling studies, we have found
literature reports on the stereoselectivity of an osmyla-
tion of a compound containing allylic fluoride3 and an
interesting experimental and theoretical study of epoxi-
dation.4 General rules to predict the stereochemical
outcome of reactions of allylic fluorides would be useful.
The lack of regiocontrolled synthetic methods for the
Syn th esis
The new allylic fluorides, 1 and 2, were prepared by
direct fluorination of corresponding allylic alcohols, as
outlined in Schemes 1 and 2. Derivatives 4a , 4b,5 and
4c are easily obtained in two steps from protected
lactaldehyde 3,6 while 1d was prepared from diol 5.7
Fluorination of allylic alcohols, 4a -4c, by diethylamino
sulfur trifluoride (DAST)8 is completely regio- and stereo-
controlled giving only the E isomers 1, as established by
NMR. No allylic transposition was observed.9 Fluorina-
* To whom correspondence should be sent.
(1) See recent Chemical Reviews issue for diastereoselection: Chem.
Rev. 1999, 5.
(2) (a) Chemistry of Organic Fluorine Compounds II; Hudlicky, M.,
Pavlath, A. E., Eds.; ACS Monograph 187; American Chemical
Society: Washington, DC, 1995. (b) Experimental Methods in Organic
Fluorine Chemistry; Katazume, T., Yamazaki, T., Eds.; Kodansha:
Tokyo, 1998. (c) Biomedicinal Aspects of Fluorine Chemistry; Filler,
R., Kobayashi, Y., Eds.; Elsevier: Amsterdam, 1982. (d) Mann, J . J .
Chem. Soc. 1987, 16, 381. (e) Welch, J . T. Tetrahedron 1987, 43, 3123.
(f) Welch, J . T. Selective Fluorination in Organic and Bioorganic
Chemistry; ACS Symposium Series 456; American Chemical Society:
Washington, DC, 1991. (g) Ojima, I.; McCarthy, J . R.; Welch, J . T.
Biomedical Frontiers of Fluorine Chemistry; ACS Symposium Series
639; American Chemical Society: Washington, DC. 1996.
(3) Davis, F. A.; Qi, H.; Sundarababu, G. Tetrahedron Lett. 1996,
37, 4345.
(5) For another synthesis of 4b see: Busque, P.; De March, P.;
Figueredo, M.; Font, J .; Monsalvatde, M.; Virgili, A.; Alvarez-Larena,
A.; Piniella, J . F. J . Org. Chem. 1996, 61, 8578.
(6) Massad, S. K.; Hawkins, L. D.; Baker, D. C. J . Org. Chem. 1983,
48, 5180.
(7) Inokuchi, T.; Matsumoto, S.; Nishiyama, T.; Torii, S. J . Org.
Chem. 1990, 55, 462.
(8) Middleton, W. J . J . Org. Chem. 1975, 40, 575-578. Hudlicky,
M. Org. React. 1988, 35, 513.
(9) For results on phenyl substituted and polyunsaturated systems,
see: Boukerb, A.; Gree, D.; Laabassi, M.; Gree, R. J . Fluorine Chem.
1998, 88, 23.
(4) Fujita, M.; Ishizuka, H.; Ogura, K. Tetrahedron Lett. 1991, 32,
6355.
10.1021/jo0016024 CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/09/2001

Conformations of Allylic Fluorides
J . Org. Chem., Vol. 66, No. 7, 2001 2375
Sch em e 1
Sch em e 2
tion of 6 gave not only 1d (41% yield) but also 1d ′ (29%)
resulting from a methyl migration in a carbocation
intermediate. These derivatives were separated by chro-
matography. Methylenemalonate derivative 2 was pre-
pared in two steps from diethyl mesoxalate 7 (Scheme
2). It is worth noting that, in that case, fluorination of 8
occurs with a complete allylic transposition. In fluorina-
tion of allylic alcohols with DAST, competition between
the reaction with and without transposition (formal SN
and SN′, respectively) is frequently observed.8 The reac-
tions of 4, 6, and 7 gave only the most conjugated, and
probably the most stable, compounds 1 and 2.10
F igu r e 1. B3LYP/6-31G* optimized geometries of 3-fluoro-
1-butene. Energies are in kcal/mol; populations were calculated
using the Boltzmann distribution at 298 K; coupling constants
for H2-H3 were calculated using the Karplus relationship
(3J HH ) 7 - 1 cos R + 5 cos 2R).
These conformers were found to be higher in energy than
A by 0.5 and 1.4 kcal/mol, respectively. In conformation
C there is steric repulsion between the CH3 and the CH2
group. From the relative energies of the conformers, their
percentages at 25 °C were calculated using a Boltzmann
distribution to be 65.6%, 28.2%, and 6.2% for A, B, and
C respectively. Using the Karplus relationship (3J HH
)
3
7 - 1 cos R + 5 cos 2R)13 a J H coupling of 6.3 Hz was
2H3
predicted for compound 1a at room temperature. Experi-
mentally, a value of 6.4 Hz was measured. Calculated
relative energies for the three conformations of 4-fluoro-
trans-2-pentene, the compound with a trans-methyl
group added, were similar. In this case, a J H coupling
3H9
of 6.1 Hz was predicted, while a value of 6.7 Hz was
measured.
Gr ou n d Sta te Con for m a tion a l Stu d ies. Previous
studies11 dealing with the conformational properties of
allylic fluorides predicted a very low preference (0.5-0.7
kcal/mol) for eclipsing of the C-F linkage with the double
bond. Exceptions occur only for systems with an electron-
withdrawing group at the â position: for 1a , the calcu-
lated energy difference between the ground state (CF
eclipsed) and higher energy conformer (CH eclipsed) was
reported to be 1.8 kcal/mol.11
Computational studies were performed here to confirm
these results and to extend them to other type 1 E and
Z allylic alkenes. Hybrid density functional theory with
the B3LYP/6-31G* basis12 set was used for optimizations
of 3-fluoro-1-butene and its 1-substituted derivatives.
Optimization of 3-fluoro-1-butene gave three conform-
ers, A, B, and C, shown in Figure 1. Conformation A, in
which the fluorine is eclipsed with the double bond, is
lowest in energy. In conformations B and C the fluorine
is skew to the double bond, and either H or Me is eclipsed.
Electron withdrawing groups at the â position have a
dramatic effect on the relative energies of various con-
formers and, consequently, on the coupling constant.
Calculations for trans-1-cyano-3-fluoro-butene (1a ) gave
three conformations, D, E, and F , shown in Figure 2.
Calculations predict a large preference for D, which has
the fluorine eclipsed with the double bond. Conformers
E and F are now higher in energy by 1.8 and 2.7 kcal/
mol, respectively. This is consistent with the earlier
results from Kahn et al.11 In D, there is maximum
electron donation from the CH and CC bonding orbitals
into the electron-deficient cyanoalkene π* orbital. In the
other conformers, there is less opportunity for donation
(12) All calculations used Gaussian 94 (Revision A.1). Frisch, M.
J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; J ohnson, B. G.; Robb,
M. A.; Cheeseman, J . R.; Keith, T. A.; Peterson, G. A.; Montgomery, J .
A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J .
V.; Foresman, J . B.; Cioslowski, J .; Stefanov, B. B.; Nanayakkara, A.;
Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;
Andres, J . L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J .;
Binkley, J . S.; Defrees, D. J .; Baker, J .; Stewart, J . P.; Head-Gordon,
M.; Gonzalez, C.; Pople, J . A. Gaussian 94, revision A.1; Gaussian,
Inc.: Pittsburgh, PA, 1995.
(10) For another example, see: Li, Y.; Luthman, K.; Hacksell, U.
Tetrahedron Lett. 1992, 43, 4487.
(11) Kahn, S. D.; Pau, C. F.; Chamberlin, A. R.; Hehre, W. J . J . Am.
Chem. Soc. 1987, 109, 650 and references therein.
(13) Macomber, S. R. A Complete Introduction to NMR Spectroscopy;
Wiley: New York, 1998.

2376 J . Org. Chem., Vol. 66, No. 7, 2001
Gre´e et al.
Ta ble 1.
F igu r e 2. B3LYP/6-31G* optimized geometries for trans-1-
cyano-3-fluoro-1-butene.
F igu r e 3. B3LYP/6-31G* optimized geometries for the methyl
ester of 2.
into the π* orbital. Because the allylic CH is skew in the
highly populated D, the predicted coupling constant drops
to 4.2 Hz, compared to a measured value of 3.5 Hz. The
microwave spectrum of 1-cyano-3-fluoro-1-butene was
recently reported.14 The conformation with the allylic CF
eclipsed with the alkene was found to be the predominant
structure.
preference for G in which the fluorine is anti to the double
bond, over H, and in which it is syn. In G the electrone-
gative fluorine is far away from the oxygens of the ester.
The calculated J HH coupling constant for the methyl ester
is 9.3 Hz compared to the measured value of 6.6 Hz for
2.
Additional calculations are summarized in Table 1.
Calculations for a trans-â-formyl and the trans-carboxylic
acid also predict a large preference for the eclipsed
conformation. Here coupling constants are predicted to
be 4.3 and 3.3 Hz for the acid and aldehyde, respectively.
Experimentally, a value of 4.0 Hz is measured for both
4-fluoro-2-trans-pentenal and trans-4-fluoro-2-pentenoate.
With a donor substituent on the alkene, there is little
preference for the fluoride to be eclipsed or staggered with
the double bond. Calculations predict a 0.3 kcal/mol
preference for the CH eclipsed over the CF eclipsed, since
there is a favorable interaction between the electron-rich
The pattern of conformations of these allylic fluorides
is readily understandable, and the calculations reproduce
the experimental coupling constants, except in this last
case.
Diels Ald er Cycloa d d ition s of Allylic F lu or id es.
The reaction of 2,3-dimethyl-1,3-butadiene with 1c at 60
°C proceeded to give a 76:24 mixture of adducts 9c and
9c′ in 85% overall yield (Scheme 3). These derivatives
were separated by chromatography. Neither reversibility
nor epimerization was observed when each stereoisomer
was heated separately under the reaction conditions. The
stereochemistry of 9c could be confirmed unambiguously
in the following way: epoxidation of 9c gave a 2:1
mixture of 10c and 10c′, which was separated by chro-
matography. The epoxide 10c′ gave crystals suitable for
X-ray diffraction studies. The X-ray structure establishes
the stereochemical relationships shown in 10c′.
3
alkene π orbital and the out of plane σ* CF. A J HH
coupling constant of 9.2 Hz is predicted.
Figure 3 shows three calculated low energy conforma-
tions for the methyl ester of 2. There is a 1.2 kcal/mol
(14) Kassi, S.; Gre´e, D.; Gre´e, R.; Duflot, D.; Petitprez, D.; Wlodarc-
zak, G. J . Mol. Spectrosc. 2000, 202, 19.

Conformations of Allylic Fluorides
J . Org. Chem., Vol. 66, No. 7, 2001 2377
F igu r e 4. PM3 optimized geometries for 9c (J ) and 9c′ (K).
Sch em e 3
F igu r e 5. B3LYP/6-31G* TS for the reaction of butadiene
with 3-fluorobutene.
mixture). It is isolated in 49% yield. The stereochemistry
of 11 was deduced from the NMR data which showed a
3
3
low J H (3.1 Hz) and a high J H (23.4 Hz).
aHb
aF
Tr a n sition Sta tes a n d Ster eoselectivity Mod els.
The reaction of butadiene with 3-fluoro-1-butene was
used as a model reaction to explore these Diels-Alder
reactions theoretically. For this reaction there are twelve
possible transition structures; each of the three staggered
conformers of the allylic fluoride can add to butadiene
exo or endo and to the top or bottom of the diene. Figure
5 shows B3LYP/6-31G* exo transition structures. Struc-
tures on the left column give the products similar to 9a -
9d ; while those on the right column produce 9a ′-9d ′.
Configuration of the stereogenic center in the drawings
of 3-fluoro-1-butene in Figure 5 are the opposite of the
products in Figure 4. The reactants used experimentally
were racemic.
Transition structure conformation L, in which the
fluorine is inside and methyl is anti, has the lowest
energy. The second lowest in energy is M, by 0.6 kcal/
mol. In M the fluorine is outside and the methyl is anti.
Calculations predict a small preference for the formation
of the product from F-inside, and Me-anti, in accord with
the experiment. In conformations N and O the fluorine
is anti or inside with respect to the double bond, and the
methyl is in the more crowded outside position. Struc-
tures N and O are 2.5 and 1.3 kcal/mol, respectively,
higher in energy than L. In conformations P and Q the
methyl is inside, this places the fluorine either outside
or anti and perpendicular to the double bond. Structures
P and Q are 2.2 and 3.1 kcal/mol, respectively, higher in
energy than L.
The product stereochemistries were supported by NMR
data, which allowed identification of products from
related reactions. Isomer 9c′ has a very low 3J H
aHb
coupling constant (2.0 Hz) and a high 3J H (30.0 Hz),
aF
3
while isomer 9c has opposite trends: J H H ) 7.7 Hz and
a
b
3J H ) 8.0 Hz. Here Ha and Hb refer to the H geminal to
aF
F and the methine vicinal to F. The conformations that
are consistent with these assignments are shown in
Figure 4. The minor product 9c′ (K) has the CF bond anti
3
to the ring CH bond and consequently a large J HF ) 30
Hz. The CH bonds are gauche and have a small coupling.
This conformer is found to place the CH near the COPh
substituent and to place the Me in the least crowded
position. The major product 9c (J ), has the Me in the
least crowded position, while it places the CF bond near
3
the COPh substituent. Consequently the J HH is large, a
result of the antiperiplanar arrangement.
The cycloaddition of 1d gave similar results. An 81:19
ratio of adducts 9d and 9d ′ was obtained in 90% yield.
The NMR spectra (see experimental) are very similar to
9c and 9c′ and could be used for isomer identification.
Cycloaddition with 1a proved to be slow (only 40%
reaction after 10 days) leading to a 1:1 mixture of 9a and
9a ′. The stereochemistries of these adducts were estab-
lished using NMR, by analogy with 9c and 9c′. No Diels-
Alder reaction was observed with ester 1b; only starting
material was recovered in that case.
Figure 6 shows four exo B3LYP/6-31G* transition
states for the cycloaddition of 4-fluoropent-2-en-1-al, a
model system for reactions of 1c and 1d , with 2,3-
dimethyl-1,3-butadiene. Structures on the left column
give the diastereomeric products similar to 9a -9d while
those on the right column produce 9a ′-9d ′. Conformation
R, in which the fluorine is inside, has the lowest energy.
Under similar reaction conditions, the cycloaddition of
the doubly activated 2 is complete in 3 days, and a single
adduct, 11, is obtained (NMR control of the crude reaction

2378 J . Org. Chem., Vol. 66, No. 7, 2001
Gre´e et al.
result of steric crowding between the methyl substituent
and an ester substituent. Structure X shows anti attack
when the fluorine is perpendicular to the double bond.
Structure X is 4.8 kcal/mol higher in energy than W.
Con clu sion
We have presented experimental and theoretical re-
sults on the simple formation of chiral, allylic fluorides
from their corresponding alcohols and their stereoselec-
tive Diels-Alder reactions. We have shown that the
B3LYP/6-31G* level accurately reproduces geometries
and Diels-Alder transition structures for allylic fluorides
in accord with our experimental stereochemical data. In
chiral E allylic fluorides, with an electron-withdrawing
group at the â position, the fluorine has a large preference
to be eclipsed with the double bond.
In Diels-Alder transition states, there is a significant
preference for the alkyl group at the stereogenic center
to be anti with respect to the forming bond and the
fluorine to be inside. A cis substituent on the double bond
produces a reversal in selectivity, since the fluorine now
adopts the outside position. Earlier studies with allylic
ethers found anti alkyl and inside alkoxy preferred,15 and
the same preference is found here with the fluorine
adopting the role of the “inside-alkoxy” group. Indeed,
the overal stereochemical features are nearly the same
for the corresponding allylic ethers and fluorides.
F igu r e 6. B3LYP/6-31G* TS for the cycloaddition of butadi-
ene with 4-fluoropen-2-en-1-al.
Exp er im en ta l Section
Gen er a l P r oced u r es. All dry solvents were freshly dis-
tilled under nitrogen from the recommended drying agent.
Toluene was distilled from sodium, ether and THF were
distilled from sodium benzophenone ketyl, and dichloromethane
was distilled from CaH2. Pentane or low boiling (bp < 60 °C)
petroleum ether were used for chromatography. Other re-
agents and solvents were used as received from the commercial
suppliers. All reactions were performed with dry solvents and
under a dry nitrogen atmosphere unless specified. External
bath temperatures are reported. Melting points are uncor-
rected. FTIR was recorded on NaCl plates as thin film.
Elemental analyses were done at the Service de Microanalyse
(ICSN, Gif-sur-Yvette, France). MS was performed by the
Centre Re´gional de Mesures Physiques de l’Ouest (Rennes).
1H, 13C, and 19F NMR spectra were recorded in the specified
solvent.
Allylic F lu or id es 1a to 1c. Wittig Rea ction s of 3. A
solution of protected lactaldehyde6 (3) (2.4 g, 12 mmol) and
corresponding phosphorane (18 mmol, 1.5 equiv) in toluene
(40 mL) was heated under stirring at 70 °C for 3 h (derivatives
1a and 1b) or for 5 h (1c). After evaporation of toluene, ether
was added to the reaction mixture. After filtration of Ph3PO
and evaporation of the solvent the olefins were obtained, as a
1:1 mixture of the E and Z isomers for derivative a , as a 7:3
mixture of E and Z isomers for compound 1b and as a single
E isomer in the case of 1c. All these isomers were separated
by flash chromatography on silica gel using as eluent a 15:85
mixture of ether and petroleum ether.
F igu r e 7. B3LYP/6-31G*//PM3 TS for the reaction of 2 with
2,3-dimethyl-1,3-butadiene.
The TS is more asynchronous compared to 3-fluorobutene
with the longer forming bond to the carbon alpha to the
carbonyl. Transition state S is second lowest in energy
by 1.0 kcal/mol. In both cases, the methyl adopts the anti
position. In R the fluorine is inside, and in S the fluorine
is outside. There is now a larger preference for formation
of the F-inside, Me-anti transition state. In T and U the
methyl is in the more crowded outside position, and
fluorine is anti (T, 4.1 kcal/mol) or inside (U, 1.8 kcal/
mol).
Figure 7 shows four exo B3LYP/6-31G*//PM3 transi-
tion states for the highly stereoselective cycloaddition of
the diester, 2, with 2,3-dimethyl-1,3-butadiene. The
lowest energy transition structure, W, has the methyl
anti as usual, but now the fluorine is outside. This
transition structure leads to the product 11. Structure
V is 3.2 kcal/mol higher in energy and gives the unob-
served isomer. Calculations predict a large preference for
formation of isomer 11 in accord with the experiment.
The higher energy of V is due to repulsions between the
carbonyl oxygen with the inside fluorine. Placing the
methyl group on the inside gives Y, which is 6.9 kcal/
mol higher in energy than W. This energy difference is a
ter t-Bu tyld im eth ylsilyl Eth er s: Olefin a (E). 950 mg
1
(38%), Rf ) 0.5. H NMR (400 MHz, CDCl3) δ: 6.68 (dd, 1H, J
) 16.0, J ) 3.0); 5.56 (dd, 1H, J ) 16.0, J ) 2.2); 4.30 (bm,
1H); 1.19 (d, 3H, J ) 6.7); 0.84 (s, 9H); 0.02 (s, 3H); 0.01 (s,
3H). 13C NMR (100 MHz, CDCl3) δ: 158.1; 117.6; 97.7; 67.6;
25.7; 23.4; 18.1; -4.8; -5.0. IR (film): 2957; 2932; 2890; 2860;
2223 cm-1
.
(15) (a) Houk, K. N.; Moses, S. R.; Wu, Y,; Rondan, N. G.; J ager, V.;
Schohe, R. Fronczek, F. R. J . Am. Chem. Soc. 1984, 106, 3880. (b)
Haller, J .; Niwayama, S.; Duh, H.; Houk, K. N. J . Org. Chem. 1997,
62, 5728 and references therein.

Conformations of Allylic Fluorides
J . Org. Chem., Vol. 66, No. 7, 2001 2379
1
Olefin a (Z). 950 mg (38%), Rf ) 0.6. H NMR (400 MHz,
CDCl3) δ: 6.34 (dd, 1H, J ) 11.2, J ) 8.1); 5.17 (dd, 1H, J )
11.2, J ) 1.0); 4.62 (bm, 1H); 1.19 (d, 3H, J ) 6.4); 0.80 (s,
9H); 0.0 (s, 3H); -0.02 (s, 3H). 13C NMR (100 MHz, CDCl3) δ:
158.1; 115.3; 97.1; 67.8; 25.8; 23.7; 18.1; -4.8; -5.0. IR (film):
identical to (E) isomer.
2990; 2927; 2858; 2228; 1705; 1642 cm-1. HRMS(EI) for C5H6-
NF (M+) calcd 99.0484, found 99.0487.
1b. Bp: 70°/20 mmHg. 105 mg (80%), Rf ) 0.45 (ether/
petroleum ether 2:8). 1H NMR (400 MHz, CDCl3) δ: 6.91 (ddd,
1H, J HF ) 19.8, J ) 15.8, J ) 4.0); 6.04 (dt, 1H, J ) 15.8, J HH
) J HF ) 1.5); 5.29 (dqdd, 1H, J HF ) 47.9, J ) 6.6, J ) 4.0, J
) 1.5); 3.75 (s, 3H); 1.44 (dd, 3H, J HF ) 23.6, J ) 6.6). 13C
NMR (100 MHz, CDCl3) δ: 166.5; 146.3 (J CF ) 18.7); 120.2 (J CF
) 11.1); 87.8 (J CF ) 170.5); 51.8; 20.6 (J CF ) 22.9). 19F NMR
(376 MHz, CDCl3) δ: -177.6 (dqd J HF ) 48.0, J HF ) 23.6, J HF
) 20.0). Anal. Calcd for C6H9O2F: C, 54.54, H, 6.82. Found: C,
54.36; H, 6.71.
Olefin b (E). 1.74 g (60%), Rf ) 0.5. 1H NMR (400 MHz,
CDCl3) δ: 6.87 (dd, 1H, J ) 15.8, J ) 4.0); 5.94 (dd, 1H, J )
15.8, J ) 1.5); 4.39 (bm, 1H); 3.67 (s, 3H); 1.18 (d, 3H, J )
6.6); 0.84 (s, 9H); 0.07 (s, 3H); 0.06 (s, 3H). 13C NMR (100 MHz,
CDCl3) δ: 167.2; 152.2; 118.5; 67.6; 51.4; 25.7; 23.5; 18.2; -4.9.
IR (film): 2956; 2932; 2890; 2859; 1729; 1650 cm-1
.
1
1
Olefin b (Z). 750 mg (25%), Rf ) 0.7. H NMR (400 MHz,
CDCl3) δ: 6.19 (dd, 1H, J ) 11.7, J ) 7.6); 5.63 (dd, 1H, J )
11.7, J ) 1.6); 5.41 (dqd, 1H, J ) 7.6, J ) 6.1, J ) 1.6); 3.68
(s, 3H); 1.22 (d, 3H, J ) 6.1); 0.87 (s, 9H); 0.05 (s, 3H); 0.04 (s,
3H). 13C NMR (100 MHz, CDCl3) δ: 167.1; 155.1; 116.3; 65.5;
51.2; 25.8; 23.5; 18.1; -4.7; -4.8. IR (film): identical to (E)
isomer.
1c. 125 mg (70%), Rf ) 0.5 (ether/petroleum ether 3/7). H
NMR (400 MHz, CDCl3) δ: 8.00-7.95 (m, 2H); 7.60-7.44 (m,
3H); 7.19 (ddd, 1H, J ) 15.5, J ) 1.7, J HF ) 0.9); 7.02 (ddd,
1H, J HF ) 20.7, J ) 15.5, J ) 3.6); 5.39 (dqdd, 1H, J HF ) 48.0,
J ) 6.7, J ) 3.6, J ) 1.7); 1.55 (dd, 3H, J HF ) 23.6, J ) 6.7).
13C NMR (100 MHz, CDCl3) δ: 189.9; 145.8 (J CF ) 17.6); 137.4;
133.1; 128.7; 128.6; 123.5 (J CF ) 9.2); 88.4 (J CF ) 171.4); 20.8
(J CF ) 22.9). 19F NMR (376 MHz, CDCl3) δ: -177.3. IR (film):
3064; 2986; 2934; 1677; 1633; 1598; 1580; 1449 cm-1. HRMS-
(EI) for C11H11OF (M+) calcd 178.0794, found 178.0799.
Allylic F lu or id e 1d . Allylic Alcoh ol 6. 3,3-Dimethyl-1,2-
butanediol (5) (25 mmol) was oxidized following Torii’s pro-
cedure7, using Ca (OCl)2 as oxidant. The reaction yielded 3 g
of crude hydroxyaldehyde, directly used for the next step.
Starting from this crude aldehyde, the Wittig reaction was
done under previous conditions (toluene, 70 °C, 3h). After a
flash chromatography on silica gel using a 50:50 mixture of
ether and petroleum ether as eluent, allylic alcohol 6 was
isolated as a viscous oil which crystallized slowly in the
refrigerator.
Olefin c. 2.56 g (78%), Rf ) 0.3. 1H NMR (400 MHz, CDCl3)
δ: 8.0-7.90 (m, 2H); 7.60-7.52 (m, 1H); 7.50-7.42 (m, 2H);
7.14 (dd, 1H, J ) 15.3, J ) 1.5); 7.04 (dd, 1H, J ) 15.3, J )
3.0); 4.58 (qdd, 1H, J ) 6.6, J ) 3.0, J ) 1.5); 1.32 (d, 3H, J )
6.6); 0.96 (s, 9H); 0.11 (s, 6H). 13C NMR (100 MHz, CDCl3) δ:
190.8; 152.1; 137.9; 132.7; 128.5; 128.4; 122.7; 68.1; 25.8; 23.5;
18.2; -4.8; -4.9. IR (film): 2956; 2890; 2858; 1820; 1673; 1622
cm-1
.
Allylic Alcoh ols 4a a n d 4b (Desilyla tion ). To the preced-
ing olefin a or b (3 mmol of pure E isomer) in acetone (8 mL)
was added Amberlyst 15 (0.7 g) and H2O (0.2 g). The reaction
mixture was stirred for 3 h at 60 °C (for derivative a ) and at
40 °C (in the case of 1b), then it was filtered and solid Na2-
CO3 was added. After filtration and evaporation of solvents,
the allylic alcohols were purified by chromatography on silica
gel using a 50:50 mixture of ether and petroleum ether as
eluent.
6. F ) 74 °C (hexane). 2.8 g (51% overall yield, 2 steps), Rf
1
) 0.35. H NMR (400 MHz, C6D6) δ: 8.08-8.05 (m, 2H); 7.39
(dd, 1H, J ) 15.3, J ) 4.8); 7.30-7.15 (m, 4H); 3.9 (d, 1H, J )
4.8); 1.0 (s, 9H). 13C NMR (100 MHz, CDCl3) δ: 190.6; 148.1;
137.7; 132.9; 128.6; 125.5; 79.4; 35.8; 25.7. Anal. Calcd for
1
4a . 260 mg (90%), Rf ) 0.3. H NMR (400 MHz, CDCl3) δ:
6.77 (dd, 1H, J ) 16.1, J ) 4.0); 5.67 (dd, 1H, J ) 16.1, J )
2.0); 4.48 (m, 1H); 2.20 (d, 1H, J ) 4.5); 1.34 (d, 3H, J ) 6.6).
13C NMR (100 MHz, CDCl3) δ: 157.5; 117.3; 98.2; 67.0; 22.4.
C
14H18O2: C, 77.06; H, 8.25. Found: C, 76.86; H, 8.27.
F lu or in a tion of 6. To a solution of DAST (0.16 mL, 1.2
mmol) in a mixture of pentane (3 mL) and toluene (1 mL) was
added, at room temperature under nitrogen, allylic alcohol 6
(220 mg, 1 mmol) in small portions. After 10 min stirring, the
same Na2CO3 work up as in the preceding cases was applied.
The fluorides 1d and 1d ′ were separated by flash chromatog-
raphy on silica gel using a 10:90 mixture of ether and pentane
as eluent.
IR (film): 3420; 2979; 2228; 1635 cm-1
.
4b. 292 mg (75%). The spectroscopical data were in agree-
ment with literature.5
Allylic Alcoh ol 4c. To the preceding silyl ether (478 mg,
1.74 mmol) in THF (3 mL) were added acetic acid (9 mL) and
water (3 mL). The reaction mixture was stirred at room
temperature for 48 h. After addition of a saturated Na2CO3
solution and extraction with ether, the organic phase was dried
(MgSO4) and evaporated. The allylic alcohol 4c was purified
by flash chromatography on silica gel using a 50:50 mixture
of ether and petroleum ether as eluent.
1d . 92 mg (41%), Rf ) 0.64 (ether/petroleum ether: 20/80).
1H NMR (400 MHz, CDCl3) δ: 8.02-7.70 (m, 5H); 7.21 (dd,
1H, J ) 15.5, J ) 1.5); 7.10 (ddd, 1H, J HF ) 22.2, J ) 15.5, J
) 3.7); 4.87 (ddd, 1H, J HF ) 47.3, J ) 3.7, J ) 1.5); 1.04 (d,
9H, J HF ) 1.3). 13C NMR (100 MHz, CDCl3) δ: 189.6; 142.7
(J CF ) 18.3); 137.4; 133.0; 128.6; 125.5 (J CF ) 10.3); 98.5 (J CF
) 180.6); 35.6 (J CF ) 19.3); 25.2 (J CF ) 4.5). 19F NMR (376
MHz, CDCl3) δ: -189.8 (dd, J HF ) 47.3, J HF ) 20.5). HRMS-
(EI) for C14H17OF (M+) calcd 220.1263, found 220.1247.
1d ′. 64 mg (29%), Rf ) 0.53 (ether/petroleum ether: 20/80).
1H NMR (400 MHz, CDCl3) δ: 7.98-7.47 (m, 5H); 7.03 (dd,
1H, J ) 15.6, J ) 8.0); 6.94 (d, 1H, J ) 15.6); 2.70 (dqt, 1H,
J HF ) 13.6, J ) 6.9); 1.40 (d, 6H, J HF ) 21.7); 1.20 (d, 3H, J )
6.9). 13C NMR (100 MHz, CDCl3) δ: 190.7; 149.2 (J CF ) 6.2);
137.7; 132.7; 128.6; 126.6; 96.4 (J CF ) 171.1); 46.6 (J CF ) 23.0);
24.9 (J CF ) 24.5); 24.5 (J CF ) 24.6); 141 (J CF ) 5.5). 19F NMR
(376 MHz, CDCl3) δ: -140.8 (d sept., J HF ) 21.3, J HF ) 12.5).
HRMS(EI) for C14H17OF (M+) calcd 220.1263, found 220.1267.
Allylic F lu or id e 2. Allylic Alcoh ol 8. To a solution of ethyl
mesoxalate (7) (2 g, 11.5 mmol) in dry THF (10 mL) was added,
dropwise under nitrogen at -50 °C, propenylmagnesium
bromide (18 mL of a 1 M solution in THF, 1.5 equiv). The
reaction mixture was stirred for 30 min at -40 °C before
addition, at this temperature, of a saturated NH4Cl solution
(20 mL). After extraction with ether, the organic phase was
washed, dried (MgSO4), and evaporated. Alcohol 8 (80:20
mixture of E and Z isomers) was isolated by flash chromatog-
raphy on silica gel using a 50:50 mixture of ether and
petroleum ether as eluent.
1
4c. 230 mg (75%), Rf ) 0.2. H NMR (400 MHz, CDCl3) δ:
8.00-7.90 (m, 2H); 7.60-7.40 (m, 3H); 7.13 (dd, 1H, J ) 15.3,
J ) 1.0); 7.05 (dd, 1H, J ) 15.3, J ) 4.1); 4.60 (bm, 1H); 2.99
(bs, 1H); 1.38 (d, 3H, J ) 6.6). 13C NMR (100 MHz, CDCl3) δ:
191.0; 151.5; 137.4; 132.9; 128.53; 128.49; 123.0; 67.4; 22.7.
IR (film): 3422; 2975; 1667; 1623; 1598 cm-1. HRMS(EI) for
C
11H12O2 (M+) calcd 176.0837, found 176.0831.
F lu or in a tion of 4a -4c. To a solution of DAST (0.19 mL,
1.2 mmol) in CH2Cl2 (2 mL) was added, under nitrogen at room
temperature, a solution of allylic alcohol 4a -4c (1 mmol) in
CH2Cl2 (2 mL). After the mixture was stirred for 10 min, solid
Na2CO3 (250 mg) was added first. After filtration, the organic
phase was washed with a saturated Na2CO3 solution and dried
(MgSO4). The solvent was removed by distillation at atmo-
spheric pressure, and the allylic fluorides were purified by
flash chromatography on silica gel using as eluent a 10:90
mixture of ether and pentane.
1a . 50 mg (50%), Rf ) 0.4 (ether/petroleum ether 2:8). 1H
NMR (400 MHz, CDCl3) δ: 6.72 (ddd, 1H, J HF ) 19.8, J ) 16.3,
J ) 3.4); 5.67 (ddd, 1H, J ) 16.3, J ) 1.9, J HF ) 1.7); 5.23
(dqdd, 1H, J HF ) 47.4, J ) 6.7, J ) 3.5, J ) 2.0); 1.47 (dd, 3H,
J HF ) 23.7, J ) 6.7). 13C NMR (100 MHz, CDCl3) δ: 152.2 (J CF
) 17.9); 116.5; 99.6 (J CF ) 15.2); 87.4 (J CF ) 174.9); 20.3 (J CF
) 22.5). 19F NMR (376 MHz, CDCl3) δ: -180.2. IR (film):

2380 J . Org. Chem., Vol. 66, No. 7, 2001
Gre´e et al.
8. (4:1 E/Z mixture) 1.3 g (52%), Rf ) 0.43. 1H NMR (400
MHz, CDCl3) δ: 6.07 (dq, 1H E, J ) 15.3, J ) 6.6); 5.93 (d, 1H
Z +E, J (Z) ) 11.5); 5.84 (dq, 1H, Z, J ) 11.5, J ) 7.1); 4.28
(q, 4H, Z +E, J ) 7.1); 3.96 (bs, 1H, Z); 3.90 (bs, 1H, E); 1.80-
1.75 (m, 3H, Z +E); 1.30 (t, 3H, Z +E, J ) 7.1). 13C NMR (100
MHz, CDCl3) δ: 170.2; 132.4 (Z); 128.4 (E); 125.4 (E); 124.7
(Z); 62.8; 17.7; 14.3; 14.0. IR (film): 3482; 2984; 2920; 1739;
1658; 1632 cm-1. HRMS(EI) for C10H16O5 (M+) calcd 216.0997,
found 216.1007.
J ) 0.7); 3.82 (td, 1H, J ) 10.8, J ) 5.4); 2.52 (dtdd, 1H, J HF
) 30.2, J ) 11.0, J ) 5.8, J ) 0.7); 2.36 (m, 1H); 2.15 (m, 3H);
1.69 (s, 3H); 1.62 (s, 3H); 0.99 (s, 9H). 13C NMR (100 MHz,
CDCl3) δ: 205.3; 137.5; 133.2; 128.7; 128.3; 125.1; 123.3; 100.4
(J CF ) 178.8); 44.7 (J CF ) 1.4); 37.2 (J CF ) 19.9); 36.8; 35.5
(J CF ) 19.2); 30.8 (J CF) 7.4); 26.8 (J CF ) 4.8); 19.1; 18.6. 19F
NMR (376 MHz, CDCl3) δ: -204.0 (dd, J HF ) 47.4, J HF ) 30.2).
HRMS(EI) for C20H27OF (M+) calcd 302.2046, found 302.2051.
The same reaction applied to 2 (75 mg, 0.34 mmol) gave,
after 72 h, a single adduct 11 (NMR control 80% reaction with
20% remaining 2), purified by flash chromatography on silica
gel using a 20:80 mixture of ether and petroleum ether as
eluent.
F lu or in a tion of 8. Starting from 8 the same DAST
fluorination procedure as before was used. It gave allylic
fluoride 2 together with a small amount (around 15-20%) of
unidentified, nonfluorinated byproducts. Derivative 2 was
isolated by flash chromatography on silica gel using a 98:2
mixture of toluene and ether as eluent.
11. 51 mg (49%), Rf ) 0.7. 1H NMR (400 MHz, CDCl3) δ:
4.90 (dqd, 1H, J HF ) 48.3, J ) 6.6, J ) 3.0); 4.25-4.12 (m,
4H); 2.60 (d, 1H, J ) 16.8); 2.48 (dtd, 1H, J HF ) 23.4, J ) 7.1;
J ) 3.0); 2.39 (d, 1H, J ) 16.8); 2.27 (dd, 1H, J ) 18.3, J )
7.1); 2.11 (dd, 1H, J ) 18.3, J ) 7.1); 1.63 (s, 3H); 1.61 (s, 3H);
1.37 (dd, 3H, J HF ) 23.9, J ) 6.6); 1.24 (t, 6H, J ) 7.1). 13C
NMR (100 MHz, CDCl3) δ: 171.5; 170.3; 124.1 (J CF ) 0.8);
122.2; 90.6 (J CF ) 169.7) 61.3; 61.2; 56.7 (J CF ) 1.5); 42.8 (J CF
) 19.8); 36.7 (J CF ) 1.5); 29.0 (J CF ) 8.8); 20.3; (J CF ) 23.7);
18.7; 18;5; 13.90; 13.89. 19F NMR (376 MHz, CDCl3) δ: -181.3
(dqt, J HF ) 48.3, J HF ) 23.9). IR (film): 2984; 2920; 2861; 1731
cm-1. HRMS(EI) for C16H25O4F (M+) calcd 300.1737, found
300.1745.
1
2. 120 mg (55%), Rf ) 0.8 (ether/petroleum ether: 1/1). H
NMR (400 MHz, CDCl3) δ: 6.96 (dd, 1H, J HF ) 17.9, J ) 6.6);
5.50 (dqt, 1H, J HF ) 49.1, J ) 6.6); 4.35-4.20 (m, 4H); 1.52
(dd, 3H, J HF ) 23.7, J ) 6.6); 1.33 (t, 3H, J ) 7.1); 1.31 (t, 3H,
J ) 7.1). 13C NMR (100 MHz, CDCl3) δ: 164.4; 163.4; 146.3
(J CF ) 24.9); 128.1 (J CF ) 8.8); 86.9 (J CF ) 165.7); 61.8; 61.7;
20.5 (J CF ) 23.5); 14.0. 19F NMR (376 MHz, CDCl3) δ: -173.2
(dqd, J HF ) 47.5; J HF ) 23.7; J HF ) 17.8). HRMS(EI) for
C
10H15O4F (M+) calcd 218.0954, found 218.0958.
Diels-Ald er Cycloa d d ition s. The representative proce-
dure was as follows: a solution of 1c (0.150 g, 0.84 mmol) in
dimethylbutadiene (4 mL) was heated at 60 °C for 19 h. A
control using 19F NMR established that the reaction was
complete and gave a 76:24 ratio of 9c and 9c′. After evapora-
tion of excess diene, the two cycloadducts were separated by
flash chromatography on silica gel using 20:80 mixture of ether
and petroleum ether as eluent.
The same reaction conditions applied to 1a gave, after 10
days, only 40% cycloaddition with a 1:1 mixture of stereoiso-
mers 8a and 9a (NMR control on the crude reaction mixture).
These derivatives could not be completely separated by flash
chromatography on silica gel; only mixtures enriched in each
stereoisomer could be obtained.
1
1
8a + 9a . 31%. H NMR (400 MHz, CDCl3) δ: 5.03 (dqd, 1H
9c. 142 mg (65%), Rf ) 0.4. H NMR (400 MHz, CDCl3) δ:
syn, J HF ) 48.3, J ) 6.1, J ) 2.0); 4.60 (dqt, 1H anti, J HF
)
8.02-7.45 (m, 5H); 4.62(ddq, 1H, J HF ) 47.7, J ) 7.7, J ) 6.2);
3.64 (td, 1H, J ) 8.8, J ) 5.7); 2.49 (m, 1H, J HF ) 8.0, J ) 8.8;
J ) 7.7; J ) 5.8); 2.29 (dd, 1H, J ) 17.1, J ) 8.8); 2.18 (dd,
1H, J ) 17.1, J ) 5.4); 2.09 (dd, 1H, J ) 17.2, J ) 5.8); 1.85
47.8, J ) 6.6); 2.81 (td, 1H anti, J ) 6.6, J ) 6.1); 2.68 (ddd,
1H syn, J ) 11.2, J ) 9.7, J ) 7.6); 2.40-1.68 (m, 5H); 1.56
and 1.52 (bs, 6H); 1.31 (dd, 3H, J HF ) 23.9, J ) 6.6). 13C NMR
(100 MHz, CDCl3) δ: 125.1 and 124.2; 122.7 and 122.5; 122.0
and 121.4; 90;3 (anti, J CF ) 169.3); 89.6 (syn, J CF ) 170.3) 41.4
(d, syn, J CF ) 20.0); 40.9 (anti, J CF ) 19.8); 35.1 (syn); 32.8
(anti); 30.4 (syn, J CF ) 5.8); 28.7 (syn, J CF ) 3.6); 28.6 (anti,
J CF ) 3.4); 26.6 (anti, J CF ) 8.2); 19.1 (anti); 19.0 (syn); 18.7
(anti); 18.5 (syn); 18.1 (anti, J CF ) 23.2); 18.0 (syn, J CF ) 22.6).
19F NMR (376 MHz, CDCl3) anti δ, -178.3 (dqd, J HF ) 48.1,
(dd, 1H, J ) 17.1, J ) 8.2); 1.66 (bs, 6H); 1.28 (dd, 3H, J HF
)
24.9, J ) 6.2). 13C NMR (100 MHz, CDCl3) δ: 203.5; 136.9 (J CF
) 1.2); 132.8; 128.6; 128.2; 124.2; 123.2 (J CF ) 0.6); 92.5 (J CF
) 168.0); 43.0 (J CF ) 5.1); 41.6 (J CF ) 18.5); 34.5; 31.5 (J CF
)
6.9); 18.9; 18.6; 18.3 (J CF ) 23.3). 19F NMR (376 MHz, CDCl3)
δ: -171.4 (dqd, J ) 47.7; 24.9; 8.0). IR (film): 2983; 2915;
2859; 1680; 1596; 1580 cm-1. HRMS(EI) for C17H21OF (M+)
calcd 260.1576, found 260.1579.
J HF ) 24.1, J HF ) 7.3); syn δ, -191.3 (ddq, J HF ) 48.3, J HF
)
29.0, J HF ) 23.6). HRMS(EI) for C11H16NF (M+) (mixture of
8a + 9a ) calcd 181.1267, found 181.1268.
9c′. 44 mg (20%), Rf ) 0.5. 1H NMR (400 MHz, C6D6) δ:
8.06-7.10 (m, 5H); 4.85 (dqd, 1H, J HF ) 49.3, J ) 6.1, J )
2.0); 3.80 (td, 1H, J ) 11.2, J ) 5.1); 2.40-2.30 (m, 2H); 2.35-
2.20 (m, 1H, J HF ) 30.0, J ) 11.2, J ) 5.3, J ) 2.0); 2.04-1.88
(m, 2H); 1.60 (s, 3H); 1.50 (s, 3H); 1.05 (dd, 3H, J HF ) 23.9, J
) 6.1). 13C NMR (100 MHz, C6D6) δ: 203.8; 137.8; 133.0; 128.8;
128.7; 124.7; 124.1; 89.9 (J CF ) 168.3); 43.4 (J CF ) 3.1); 41.2
(J CF ) 19.2); 36.9; 29.3 (J CF ) 3.4); 19.0; 18.6; 18.1 (J CF ) 22.8).
Ep oxid es 10c a n d 10c′. To a solution of 9c (30 mg, 0.16
mmol) in CH2Cl2 (2 mL) was added a solution of metachlo-
roperbenzoic acid (60 mg, 0.48 mmol) in CH2Cl2 (2 mL) and
anhydrous Na2CO3 (10 mg). The reaction mixture was stirred
at room temperature for 1.5 h. After filtration, the organic
phase was washed with a saturated Na2CO3 solution and then
with water and dried (MgSO4). After evaporation of the
solvent, the diastereoisomeric epoxides (2:1 mixture, NMR
control) were separated by flash chromatography on silica gel
using a 40:60 mixture of ether and petroleum ether as eluent.
19F NMR (376 MHz, C6D6) δ: -190.9 (ddq, J HF ) 49.3, J HF
)
30.0, J HF ) 23.9). IR (film): 2984; 2912; 2860; 1678; 1596; 1580
cm-1 Anal. Calcd for C17H21OF: C, 78.39; H, 8.07. Found: C,
78.42; H, 8.13.
1
10c. 16 mg (50%), Rf ) 0.5. H NMR (400 MHz, CDCl3) δ:
The same reaction starting from 1d (0.250 g, 1.14 mmol)
gave, after 48 h, a 80:20 mixture (NMR control) of 9d and 9d ′;
they were separated by flash chromatography on silica gel
using a 10:90 mixture of ether and petroleum ether as eluent.
8.00-7.90 (m, 2H); 7.60-7.40 (m, 3H); 4.56 (dqt, 1H, J HF
)
47.1, J ) 6.3); 3.32 (td, 1H, J ) 10.3, J ) 7.0); 2.56 (qdd, 1H,
J HF ) J ) 11.1, J ) 6.4, J ) 4.7); 2.19 (dd, 1H, J ) 14.8, J )
4.7); 2.17 (dd, 1H, J ) 15.5, J ) 10.2); 2.06 (dd, 1H, J ) 15.5,
J ) 7.1); 1.64 (dd, 1H, J ) 14.8, J )11.2); 1.40 (s, 3H); 1.33 (s,
3H); 1.21 (dd, 3H, J HF ) 24.8, J ) 6.3). 13C NMR (100 MHz,
CDCl3) δ: 202.7; 137.0; 132.9; 128.7; 128.1; 92.1 (J CF ) 169.1);
61.9 (J CF ) 0.7); 60.9; 42.7 (J CF ) 5.3); 38.4 (J CF ) 19.0); 34.6;
32.4 (J CF ) 5.5); 20.7; 19.6; 18.0 (J CF ) 23.4). 19F NMR (376
1
9d . F ) 92°. 190 mg (55%), Rf ) 0.60. H NMR (400 MHz,
C6D6) δ: 8.13-8.09 (m, 2H); 7.27-7.21 (m, 3H); 3.94 (dd, 1H,
J HF ) 48.6, J ) 7.9); 3.72 (td, 1H, J ) 9.4, J ) 5.7); 2.87 (m,
1H, J HF ) 2.5); 2.33 (dd, 1H, J ) 16.5, J ) 2.2); 2.18 (dd, 1H,
J ) 16.5, J ) 5.1); 2.13 (dd, 1H, J ) 16.8, J ) 5.4); 1.83 (dd,
1H, J ) 16.8, J ) 8.7); 1.64 (s, 3H); 1.60 (s, 3H); 1.00 (d, 9H,
J HF ) 1.4). 13C NMR (100 MHz, C6D6) δ: 202.3 (J CF ) 2.9);
137.7 (J CF ) 2.5); 132.2; 128.6; 128.5; 124.6; 123.1 (J CF ) 1.8);
105.7 (J CF ) 181.2);43.0 (J CF ) 12.1); 38.0 (J CF ) 6.0); 37.2
(J CF ) 18.1); 35.4; 35.2 (J CF ) 20.5);25.9 (J CF ) 5.5); 18.9; 18.6.
19F NMR (376 MHz, C6D6) δ: -176.1 (d, J HF ) 48.5). Anal.
Calcd for C20H27OF: C, 79.47; H, 8.94. found: C, 79.39; H, 8.84.
MHz, CDCl3) δ: -172.9 (dqd, J HF ) 48.0, J HF ) 24.0; J HF
)
11.9). IR (film): 3440; 2982; 2956; 2929; 2854; 2380; 1677;
1596; 1580 cm-1. HRMS(EI) for C17H21O2F (M+) calcd 276.1525,
found 276.1525.
10c. F ) 65 °C (ether + pentane). 10 mg (31%), Rf ) 0.3.
1H NMR (400 MHz, CDCl3) δ: 8.00-7.90 (m, 2H); 7.60-7.45
(m, 3H); 4.68 (dqt, 1H, J HF ) 47.8, J ) 6.3); 3.67 (td, 1H, J )
9.1, J ) 4.7); 2.34 (b sext, 1H, J HF ) J ) 8.1); 2.16 (dd, 1H, J
) 15.0, J ) 4.7); 1.94 (dd, 1H, J ) 15.0, J ) 9.2); 1.89 (dd, 1H,
1
9d ′. 50 mg (15%), Rf ) 0.65. H NMR (400 MHz, CDCl3) δ:
8.05-8.01 (m, 2H); 7.60-7.47 (m, 3H); 4.08 (dd, 1H, J HF ) 47.4,

Conformations of Allylic Fluorides
J . Org. Chem., Vol. 66, No. 7, 2001 2381
J ) 15.8, J ) 7.3); 1.77 (dd, 1H, J ) 15.8, J )8.6); 1.39 (s,
3H); 1.33 (s, 3H); 1.22 (dd, 3H, J HF ) 24.8, J ) 6.3). 13C NMR
(100 MHz, CDCl3) δ: 203.0; 136.2; 133.2; 128.7; 128.3; 92.0 (J CF
Ack n ow led gm en t. We are grateful to the National
Institute of General Medical Sciences and National
Institutes of Health and for a joint grant from the
National Science Foundation (U.S.A.) and the Centre
National de la Recherche Scientifique (France) for
financial support of this research.
) 167.4); 62.1; 61.2 (J CF ) 0.6); 40.5 (J CF ) 19.2); 40.0 (J CF
5.1); 34.1; 30.1 (J CF ) 6.8); 20.7; 19.5; 17.7 (J CF ) 23.0). 19F
NMR (376 MHz, CDCl3) δ: -170.2 (dqd, J HF ) 47.8, J HF
)
)
24.8; J HF ) 8.0). IR (film): 3450; 3061; 2986; 2928; 2854; 1680;
1596; 1580 cm-1. Anal. Calcd for C17H21O2F: C, 73.98; H, 7.87;
found: C, 73.88; H, 7.66. HRMS(EI) for C17H22O2F (M + H+)
calcd 277.1604, found 277.1604.
J O0016024