Regio- and stereoselective ring opening of 2,3-diaryl oxiranes by LiBr/Amberlyst 15: A new stereocontrolled access to 1,2-diaryl-2-bromo alcohols

Article abstract of DOI:10.1021/jo048045w

(Chemical Equation Presented) Both symmetrical and non symmetrical trans-2,3-diaryloxiranes are regio- and stereoselectively opened by the LiBr/Amberlyst 15 system. In the case of symmetrical trans-stilbene oxide, the syn-versus anti-bromohydrins ratio ranged between 88/12 and 30/70, by varying the reaction temperature from 20 to -30°C. In the case of nonsymmetrical para -substituted trans-2,3-diaryloxiranes, the regioselectivity is determined by electronic effects. If one phenyl bears a strong electron-withdrawing group (as NO₂ or CF₃), the nucleophilic attack is totally on the β-carbon with respect to the substituted phenyl ring. With one phenyl bearing a strong electron-releasing group (OCH₃), the regioselectivity is reversed. Ab initio calculation at the DFT/B3LYP/6- 31G* level, run on protonated epoxide structures, supports the formation of a cationic acyclic intermediate. Application of the method on ortho-methoxy and ortho-nitro 2,3-diaryloxiranes afforded the syn-bromohydrins in excellent yield, via regio- and stereoselective opening at either α- or β-carbon, respectively.

Full text of DOI:10.1021/jo048045w

Regio- and Stereoselective Ring Opening of 2,3-Diaryl Oxiranes by
LiBr/Amberlyst 15: A New Stereocontrolled Access to
1,2-Diaryl-2-bromo Alcohols
Arlette Solladie´-Cavallo,† Paolo Lupattelli,*,‡ and Carlo Bonini‡
Laboratoire de Ste´re´ochimie Organome´tallique associe´ au CNRS, ECPM/Universite´ L. Pasteur,
25 rue Becquerel, 67087-Strasbourg, France, and Dipartimento di Chimica, Universita` degli studi della
Basilicata, via N. Sauro 85, 85100-Potenza, Italy
lupattelli@unibas.it
Received November 3, 2004
Both symmetrical and nonsymmetrical trans-2,3-diaryloxiranes are regio- and stereoselectively
opened by the LiBr/Amberlyst 15 system. In the case of symmetrical trans-stilbene oxide, the syn-
versus anti-bromohydrins ratio ranged between 88/12 and 30/70, by varying the reaction temper-
ature from 20 to -30 °C. In the case of nonsymmetrical para-substituted trans-2,3-diaryloxiranes,
the regioselectivity is determined by electronic effects. If one phenyl bears a strong electron-
withdrawing group (as NO2 or CF3), the nucleophilic attack is totally on the â-carbon with respect
to the substituted phenyl ring. With one phenyl bearing a strong electron-releasing group (OCH3),
the regioselectivity is reversed. Ab initio calculation at the DFT/B3LYP/6-31G* level, run on
protonated epoxide structures, supports the formation of a cationic acyclic intermediate. Application
of the method on ortho-methoxy and ortho-nitro 2,3-diaryloxiranes afforded the syn-bromohydrins
in excellent yield, via regio- and stereoselective opening at either R- or â-carbon, respectively.
Introduction
dihalides and 1,2-diols. In the past few years, new and
milder procedures have been proposed, which used silyl
halides in the presence of different promoters,2 elemental
halogen with various catalysts,3 borane halogenides,4 and
metal halides with Lewis acid systems.5 Among these
Vicinal halo alcohols have attracted the interest of
organic chemists both for their usefulness as versatile
building blocks, as well as key intermediates in the
synthesis of halogenated marine natural products. Al-
though a variety of new procedures have been reported,
most of them have some limitations, in terms of either
harsh reaction conditions or low regioselectivity in the
opening of unsymmetrical epoxides.1 The most common
method for the synthesis of 1,2-halohydrines from 1,2-
epoxides is their ring opening either with hydrogen
halides or with hydrohalogenic acid. These procedures
are generally associated with byproducts such as vic-
(2) (a) Xu, L.-W.; Li, L.; Xia, C.-G.; Zhao, P.-Q. Tetrahedron Lett.
2004, 45, 2435-2438. (b) Paek, S. H.; Shim, S. C.; Cho, C. S.; Kim,
T.-J. Synlett 2003, 849-851. (c) Nakajima, M.; Saito, M.; Uemura, M.;
Hashimoto, S. Tetrahedron Lett. 2002, 43, 8827-8829.
(3) (a) Soroka, M.; Goldeman, W.; Malysa, P.; Stochaj, M. Synthesis
2003, 2341-2344. (b) Shargi, H.; Eskandari, M. M. Tetrahedron 2003,
53, 8509-8514. (c) Shargi, H.; Paziraee, Z.; Niknam, K. Bull. Korean
Chem. Soc. 2002, 23, 1611-1615. (d) Diaz, D.; Martin, T.; Martin, V.
S. J. Org. Chem. 2001, 66, 7231-7233.
(4) (a) Guindon, Y.; Therien, M.; Girard, Y.; Yoakim, C. J. Org.
Chem. 1987, 52, 1680-1686. (b) Joshi, N. N.; Srebnik, M.; Brown, H.
C. J. Am. Chem. Soc. 1988, 110, 6246-6248. (c) Bell, T. W.; Ciaccio, J.
A.; Heller, E.; Talbot, A. Synlett 1991, 248-250.
† ECPM/Universite´ L. Pasteur.
‡ Universita` degli studi della Basilicata.
(1) (a) See: Halohydrines and derivatives. In Fiesers’ Reagents for
Organic Synthesis; Smith, J. G., Fieser, M., Eds.; New York, 1990. (b)
Bonini, C.; Righi, G. Synthesis 1994, 225. (c) Larock, R. C. Compre-
hensive Organic Transformations; VCH: New York, 1989; pp 508-
509.
(5) (a) Kotsuki, H.; Shimanouchi, T. Tetrahedron Lett. 1996, 37,
1845-1848. (b) Amantini, D.; Fringuelli, F.; Pizzo, F.; Vaccaio, L. J.
Org. Chem. 2001, 66, 4463-4467. (c) Sabitha, S.; Satheesh, R.;
Rajkumar, M.; Reddy, Ch. S.; Yadav, J. S. Tetrahedron Lett. 2001, 42,
3955-3958.
10.1021/jo048045w CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/01/2005
J. Org. Chem. 2005, 70, 1605-1611
1605

Solladie´-Cavallo et al.
TABLE 1. LiBr/Amb 15 Promoted Ring Opening of trans-Stilbene Oxide in Different Reaction Conditions
Amb/substrate
mg/mmol
conv
(%)a
yield
(%)b
entry
LiBr/substrate
T
time
syn/antia
1
2
3
4
5
6
7
2/1
2/1
4/1
4/1
4/1
4/1
4/1
440
220
440
220
220
440
0
rt
rt
rt
-15 °C
-25 °C
-30 °C
rt
1 h
6 h
50 min
6 h
18 h
2 h 30 min
4 h
100
81
100
80
70
100
95
88/12
83/17
75/25
47/53
50/50
30/70
80
95
80
70
95
no reaction
a Calculated by 1H NMR analysis of the crude mixture. b Isolated.
SCHEME 1
During our study on metal halides-promoted oxiranyl
ring opening of (E)-2,3-diaryloxiranes, we found interest-
ing results in terms of regio- and stereoselectivity, using
either MgBr2 or the NaBr/Amberlyst 15 system on either
(E)-3-phenyl-2-(2-pyridyl)oxirane or (E)-2-(2-fluorophen-
yl)-3-phenyloxirane.13
FIGURE 1. 2-Halo-1,2-diaryl alcohols as versatile intermedi-
ates.
The subsequent study on (E)-stilbene oxide using
different metal bromides with or without Amberlyst 15
has led to a new straightforward stereodivergent syn-
thesis to chiral syn- and anti-2-amino-1,2-diphenyletha-
nols.14
These preliminary results prompted us to study, more
deeply, both the influence of reaction conditions on the
stereoselectivity in the opening of the model (E)-stilbene
oxide and the regioselectivity in the opening of nonsym-
metrical substituted 2,3-diaryloxiranes.
methods, the use of elemental halogens suffers from low
conversions, prolonged reaction times, and poor regiose-
lectivity.6 Lithium halides have been shown to be versa-
tile reagents because of their availability at low cost.
They also facilitate ring opening under mild conditions.
Nevertheless, the use of strong Lewis acids often results
in low yield of halohydrins when applied to acid sensitive
substrates. Recently, the use of ionic liquids has broad-
ened the application of such procedures.7 Although there
are many methods, very few are suitable to regioselective
opening of nonsymmetrical 2,3-disubstituted oxiranes,
and, in such cases, regio- and stereoselective openings
are usually performed on substrates bearing chelating
groups.8
In this context, 2,3-diaryloxiranes are challenging
substrates in terms of regio- and stereoselectivity of the
ring opening, due to the small (if any) differences of
benzylic carbons reactivity. From a synthetic point of
view, they represent useful starting materials, because
the corresponding 2-halo-1,2-diaryl alcohols A are ver-
satile intermediates for the synthesis of functionalized
diaryl compounds, as, for instance, â-fluorobromides B,9
and â-amino alcohols C (Figure 1). These latter com-
pounds have recently received great attention, due to
their use, in optically active form, as chiral auxiliaries
in asymmetric synthesis,10 chiral stationary phases for
HPLC,11 and chiral ligands in asymmetric catalysis.12
Results and Discussion
Among the different metal halides, LiBr was the
reagent of choice for its high solubility in the reaction
medium (acetonitrile) and its high reactivity, even at low
temperatures. We considered also the optimum match
between thermodynamic stability and chemical reactivity
of the produced bromohydrins, if compared with the
corresponding chlorohydrins or iodohydrins.
At the beginning, we analyzed the influence of different
parameters (substrate/bromide ratio, substrate/Am-
berlyst 15 ratio, temperature, reagents concentration) on
the stereoselectivity in the opening of (E)-stilbene oxide
4 with the LiBr/Amberlyst 15 system (Scheme 1). The
most significant results are collected in Table 1.
As shown, different substrate/LiBr or substrate/Am-
berlyst ratios have usually little, if any, effect on the
overall reaction rate, conversion, and/or stereoselectivity
(compare entries 1-3).15 On the other hand, by changing
the reaction temperature, we noted a dramatic effect on
(6) (a) Konaklieva, M. I.; Dahl, M. L.; Turos, E. Tetrahedron Lett.
1992, 33, 7093-7096. (b) Sharghi, H.; Natimi, H. Bull. Chem. Soc. Jpn.
1999, 72, 1525-1531.
(7) Yadav, J. S.; Reddy, B. V.; Reddy, Ch. S.; Rajasekhar, K. Chem.
Lett. 2004, 33, 476-477.
(8) See: Righi, G.; Pescatore, G.; Bonadies, F.; Bonini, C. Tetrahe-
dron 2001, 57, 649-5656 and references therein.
(9) Hamman, S.; Benaissa, T.; Beguin, C. G. J. Chem. Res., Miniprint
1992, 360-373.
(10) Williams, R. M. In Advances in Asymmetric Synthesis; Hassner,
A., Ed.; JAI Press: 1995; Vol. 1, pp 45-94 and references therein.
(11) Yuki, Y.; Saigo, K.; Tachibana, K.; Hasegava, M. Chem. Lett.
1986, 1347-1350.
(12) (a) Hegedus, L. S. Acc. Chem. Res. 1995, 28, 299-305. (b)
Brown, B.; Hegedus, L. S. J. Org. Chem. 1998, 63, 8012-8018.
(13) Solladie´-Cavallo, A.; Lupattelli, P.; Marsol, C.; Isarno, T.;
Bonini, C.; Caruso, L.; Maiorella, A. Eur. J. Org. Chem. 2002, 1439-
1444.
(14) Lupattelli, P.; Bonini, C.; Caruso, L.; Gambacorta, A. J. Org.
Chem. 2003, 68, 3360-3363.
(15) Changes in reagent concentration had no apparent effect on
the reaction.
1606 J. Org. Chem., Vol. 70, No. 5, 2005

1,2-Diaryl-2-bromo Alcohols
SCHEME 2
The results of the opening reactions with the LiBr/
Amberlyst 15 system, at room temperature (Scheme 4),
are collected in Table 2.
The relative (anti vs syn) stereochemistry was deter-
mined by alkaline ring closure of the bromohydrins
mixture and determination of the trans or cis structure
of the epoxides obtained using 1H NMR. The regiochem-
istry was determined either after GC-MS analysis of the
crude mixtures by identification of characteristic frag-
ments or by NOESY experiments on the mixtures.
SCHEME 3
Characterization of Bromohydrins 6aI and 6aII.
The structure 6a was assigned to both the major II and
the minor I isomers by NOESY experiments, which
identified a correlation between the aryl protons at C2
and the proton at C5, assigned to CH(OH) (Figure 2).
Characterization of Bromohydrins 6bI and 6bII.
The structure 6b was assigned to both the major II and
the minor I isomers after GC-MS analysis of the crude
mixture, by identification of the characteristic fragments
(peaks m/z )175 and m/z ) 170 (79Br), 172 (81Br)) (Figure
3).
the stereoselectivity of the ring opening. At room tem-
perature, the corresponding bromohydrins were obtained
in high yield, with a syn/anti ratio ranging between 75/
25 and 88/12 (entries 1-3). By lowering the temperature,
the anti-bromohydrin increased, until becoming the
major reaction product below -25 °C (entries 4-6).16
A
The mixture of 6c, 7c, and 7e was characterized in a
similar manner.
similar stereoselective opening “with retention of config-
uration” at the oxiranyl carbon atom had been already
observed using different metal halides (MgBr2, NaBr,
KBr) and Amberlyst 15 at room temperature.14 These
more recent results could be explained assuming the
formation of an acyclic cationic-type intermediate, which
competes with a classical SN2 opening. Moreover, the syn-
bromohydrin resulted more favorably in these conditions.
By lowering the temperature, the formation of the acyclic
intermediate becomes slower, an SN2 opening with inver-
sion of configuration becomes competitive, and the syn/
anti ratio of the products decreases. The hypothesis of a
syn-bromohydrin favored by the formation of an acyclic
intermediate was supported by the reaction of (Z)-
stilbene epoxide 4, which, in the presence of the LiBr/
Amb 15 system at room temperature, also led quantita-
tively to syn-bromohydrin 2, via nucleophilic opening
with inversion of configuration (Scheme 2).
As it can be seen in Table 2, regioselectivity is affected
by substituents electronic properties. In the oxiranes with
one phenyl ring bearing a strong electron-withdrawing
group (NO2, CF3: entries 1 and 2), only the regioisomer
6 was detected in the reaction mixture, in which the
attack of the nucleophile is totally on the â-carbon with
respect to the substituted phenyl ring.
On the other hand, fluorine substitution (entry 3) gave
rise to both R and â openings with a low regioselectivity
toward regioisomer 7c (6/7 ) 40/60). In the presence of
OCH3 (electron-releasing group) on the phenyl ring, the
regioisomer 7d (R-opening) was the only one observed.
In this case, the reaction was performed at -30 °C,
obtaining bromohydrins 7d in quantitative yield. Due to
the lability of the products, they were characterized as
1
acetates, by H NMR and GC-MS analysis, after being
quenched in situ by addition of a 1/1 Ac2O/Pyr mixture.
In these reaction conditions, we obtained also 25% of 8
(Figure 4), as byproduct, which was the only product
when the reaction was performed at room temperature.
The existence of a possible thermodynamic equilibra-
tion between the two bromohydrins was excluded by the
complete unreactivity of the anti-bromohydrin in the
same reaction conditions (LiBr/substrate 4/1, Amb 15/
substrate 440 mg/mmol, rt 24 h).
In the case of trans-â-methyl styrene epoxide (entry
5), the reaction was completely regioselective toward 7,
via an attack of the bromide on the most electrophilic
benzylic carbon. All of these results match a general
model in which the regioselectivity of the bromide attack
is determined by the relative electrophilicity of the
oxiranyl carbons.
To theoretically support the observed high regioselec-
tivity, we considered the protonated epoxides 5a(H+) and
5d(H+) (Figure 5) as model structures for the reactive
intermediate.
With these results in hands, we then studied the
behavior of nonsymmetrical para-substituted 2,3-diaryl-
oxiranes, because it still lacks an efficient regioselective
opening method for such substrates. Thus, we prepared
different p-substituted (E)-2,3-diaryloxiranes 5 in good
to excellent yields, by the reaction of benzylidensulfur
ylide, generated from the corresponding sulfonium salts
under phase-transfer conditions, with the appropriate
aryl aldehyde (Scheme 3).17
SCHEME 4
J. Org. Chem, Vol. 70, No. 5, 2005 1607

Solladie´-Cavallo et al.
TABLE 2. LiBr/Amb 15 Promoted Ring Opening of Nonsymmetrical 2,3-Substituted Oxiranes, at Room Temperature
(Except Entry 4)
â-opening
6IIb
R-opening
7Ib
entrya
epoxide
R
Y
reaction time
6Ib
7IIb
1
2
3
4
5
5a
5b
5c
5d
5e
Ph
NO2
CF3
F
OCH3
H
1 h
31
38
18c
69
57
22c
Ph
45 min
5
Ph
40 min
2 h (T ) -30 °C)
1 h
38c
22c
Ph
95 (dr ) 3/2)d
CH3
19c
81c
a All reactions were performed at rt, with LiBr/substrate ) 4/1, Amb 15/substrate ) 440 mg/mmol. The conversions were all higher
than 99% and were determined by 1H NMR spectroscopy of the crude product. b Determined by 1H NMR spectroscopy of the crude product.
c Determined by GC/MS analysis of the crude together with GC/MS and 1H NMR analysis of the alkaline ring closure of the bromohydrins
mixture. d The bromohydrins were characterized as acetates, after quenching the reaction in situ by adding a 1:1 Ac2O/Pyr mixture at
low temperature (see Experimental Section).
FIGURE 2. Bromohydrins 6aI and 6aII.
FIGURE 3. Bromohydrins 6bI and 6bII and characteristic
fragments.
FIGURE 6. Minimized structures for 5a(H+)A and 5a(H+)-
B.
FIGURE 4. Structure of the byproduct 8.
tively, we calculated a difference in stability of 4.1 kcal/
mol in favor of structure 5a(H+)A (Figure 6). Such a value
would give rise to a complete regioselective ring opening,
which was indeed observed.
On the other hand, the same comparison between the
free energies of p-OCH3 cationic structures 5d(H+)A and
5d(H+)B provided a difference in stability of 7.6 kcal/
FIGURE 5. Model structures for the reactive intermediate.
We first noted that these structures invariably con-
verged to acyclic cationic structures, if submitted to
semiempirical calculations at the PM3 level. Thus, we
first carried out a conformational search on the two
couples of opened cationic structures by this level of
calculation, using a SPARTAN 02 package. Each con-
former found was then submitted to ab initio calculations
at the DFT/B3LYP/6-31G* level using a GAUSSIAN 03
package.18 Comparing the predicted free energies of the
two opened p-NO2 cationic structures 5a(H+)A and
5a(H+)B,19 which would give products 6 and 7, respec-
(18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma,
K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,
Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03; Gaussian, Inc.: Pittsburgh,
PA, 2003.
(16) At -30 °C with LiBr/substr. 4/1 and Amb 15/substr. 220 mg/
mmol, we had syn/anti ) 20/80; see ref 14.
(17) Akgu¨n, E.; Glinski, M. B.; Dhawan, K. L.; Durst, T. J. Org.
Chem. 1981, 46, 2730-2734.
(19) Each structure was confirmed as a real minimum by noting
the absence of imaginary values in the list of length frequencies.
1608 J. Org. Chem., Vol. 70, No. 5, 2005

1,2-Diaryl-2-bromo Alcohols
SCHEME 6
can be explained by a pure electronic effect. The ex-
tremely high stereoselectivity toward syn product, if
compared with that observed for para-substituted com-
pounds, indicates a likely stereoorienting effect of the
ortho substituent on the cationic intermediate, which
leads to the product with retention of configuration.
Conclusions
Symmetrical and nonsymmetrical trans-2,3-diarylox-
iranes represent useful substrates in the regio- and
stereoselective ring opening with the LiBr/Amberlyst 15
system. The high syn stereoselectivity observed in the
case of trans-stilbene oxide, together with the electronic
effect of substituents on the regioselectivity in the case
of nonsymmetrical substrates,20 allowed us to prepare,
in high yield, ortho-methoxyaryl and ortho-nitroaryl
bromohydrins, as key intermediates in the synthesis of
new polydentate ligands.
FIGURE 7. Minimized structures for 5d(H+)A and 5d(H+)-
B.
SCHEME 5
Experimental Section
(E)-Stilbene oxide 1, (Z)-stilbene oxide 4, and (E)-1-
phenylpropylene oxide 5e were prepared by oxidation of
commercial trans-stilbene, cis-stilbene, and trans-â-methyl
styrene with mCPBA, in CH2Cl2 at room temperature. The 1H
NMR spectra were consistent with those of commercially
available compounds.
(E)-2-(4-Nitrophenyl)-3-phenyloxirane 5a was obtained
in 80% yield after chromatographic purification (n-hexane/
Et2O ) 7/3). The 1H and 13C NMR spectra were consistent with
literature data.21
mol in favor of 5d(H+)B (Figure 7), which accounts for
the total opposite regioselectivity (only 7 was observed)
found in the experimental data.
Concerning the stereoselectivity of the ring opening,
there was a general trend in favoring anti-bromohydrins,
whatever the para-substituent on the phenyl ring. This
indicates a reliance between the stereoselectivity of ring
opening and electronic effects of substituents, even if it
is much less effective than in the regioselectivity.
Such interesting results, together with the previous
ones13 on substituent-induced regioselectivity, prompted
us to investigate ortho-substituted nonsymmetric 2,3-
diaryloxiranes, as trans 9 and 10, which were prepared,
in high yield, from benzyliden-dimethylsulfur ylide and
either o-anisaldheyde or 2-nitrobenzaldehyde, respec-
tively (Scheme 5).
Both epoxides were treated with the Li/Br Amb 15
system, in acetonitrile at room temperature (Scheme 6).
ortho-Methoxy epoxide 9 afforded the syn-bromohydrin
11 in excellent yield, via a regio- and stereoselective
opening at the R-carbon, with retention of configuration.
On the other hand, ortho-nitro epoxide 10 showed a
complete inversion of regioselectivity, leading to syn-
bromohydrin 13, almost quantitatively.
(Z)-2-(4-Nitrophenyl)-3-phenyloxirane was detected in the
crude of alkaline ring closure of a 31/69 6aI/6aII mixture. 1H
NMR (300 MHz, CDCl3), in mixture with 5a: overlapping
signals but δ ) 4.42 (A part of AB system, 3JAB ) 7.0 Hz, 1H),
3
4.49 (B part of AB system, JAB ) 7.0 Hz, 1H).
(E)-3-Phenyl-2-(4-trifluoromethylphenyl)oxirane 5b was
obtained in 80% yield after chromatographic purification (n-
hexane/CH2Cl2/benzene ) 8/1/1). 1H NMR (500 MHz, CDCl3):
δ ) 3.88 (d, 3J ) 1.4 Hz, 1H), 3.97 (d, 3J ) 1.4 Hz, 1H), 7.40-
3
3
7.43 (m, 5H), 7.50 (d, J ) 8.1 Hz, 2H), 7.67 (d, J ) 8.1 Hz,
2H). 13C NMR (125 MHz, CDCl3): δ ) 62.0, 63.1, 124.0 (q,
1JCF ) 273 Hz), 125.5, 125.8, 128.3, 128.6, 128.7, 130.4 (q, 2JCF
) 32 Hz), 136.5, 141.1. MS (m/z, %): 264 [M+] (100), 246 (46),
235 (97). Anal. Calcd for C15H11F3O: C, 68.18; H, 4.20.
Found: C, 68.3; H, 4.2.
10% of (Z) epoxide was also detected in the crude. 1H NMR
(300 MHz, CDCl3), in mixture with 5b: overlapping signals
but δ ) 4.64 (A part of AB system, 3JAB ) 7.0 Hz, 1H), 4.67 (B
(20) For a recent study on the electronic effect on the regioselectivity
in the opening of aryloxiranes with acetylides, see: Shindo, M.;
Sugioka, T.; Shishido, K. Tetrahedron Lett. 2004, 9265-9268.
(21) Imuta, M.; Ziffer, H. J. Org. Chem. 1979, 44, 2505-2509.
In these cases, the effects of the substituents on regio-
and stereoselectivity match. Again, the regioselectivity
J. Org. Chem, Vol. 70, No. 5, 2005 1609

Solladie´-Cavallo et al.
part of AB system, 3JAB ) 7.0 Hz, 1H). MS (m/z, %): 264 [M+]
(100), 246 (45), 235 (99).
6aI, 31%), 5.29 (d, 3J ) 6.4 Hz, 1H, CHOH, 6aII, 69%), 7.20-
3
7.40 (m, CH arom., 6aI+6aII), 7.47 (d, J ) 8.6 Hz, 2H, CH
arom., 6aII, 69%), 8.04 (d, 3J ) 8.6 Hz, 2H, 6aI, 31%), 8.15 (d,
3J ) 8.6 Hz, 2H, 6aII, 69%). 13C NMR (75 MHz, CDCl3): δ )
58.2 (CHBr, 6aI), 63.1 (CHBr, 6aII), 77.1 (CHOH, 6aII), 77.5
(CHOH, 6aI), 123.2, 126.5, 127.8, 128.0, 128.2, 128.4, 128.6,
128.7, 128.8, 129.1, 130.1, 136.6, 137.4, 145.9, 146.6.
(E)-2-(4-Fluorophenyl)-3-phenyl-oxirane 5c was ob-
tained in 67% yield after chromatographic purification (pe-
1
troleum ether/Et2O ) 9/1). H NMR (500 MHz, CDCl3): δ )
3
3.86 (brs, 1H), 3.89 (brs, 1H), 7.09 (d, J ) 10 Hz, 1H), 7.11
3
(d, J ) 10 Hz, 1H), 7.32-7.42 (m, 7H). 13C NMR (125 MHz,
2
CDCl3): δ ) 62.2, 62.7, 115.5 (d, JCF ) 22 Hz), 125.4, 127.1
syn-2-Bromo-2-phenyl-1-(4-trifluoromethylphenyl) eth-
anol 6bI was obtained in 35% yield after chromatographic
purification (n-hex/Et2O/benzene ) 8/2/1). 1H NMR (500 MHz,
CDCl3): δ ) 3.17 (brs, 1H), 5.10 (brs, 2H), 7.25-7.40 (m, 6H),
3
4
(d, JCF ) 14 Hz), 128.4, 128.5, 132.8 (d, JCF ) 3 Hz), 136.8,
162.7 (d, 1JCF ) 245 Hz). MS (m/z, %): 214 [M+] (46), 213 (40),
196 (24), 185 (100). Anal. Calcd for C14H11FO: C, 78.49; H,
5.18. Found: C, 49.2; H, 4.90.
3
7.46 (d, J ) 8.1 Hz, 2H), 7.58 (m, 1H). 13C NMR (125 MHz,
1
CDCl3): δ ) 63.7, 77.7, 123.9 (q, 1JCF ) 271 Hz), 125.1 (q, 3JCF
) 3.8 Hz), 126.7, 127.2, 127.5, 128.3, 128.6, 128.9, 130.2 (q,
2JCF ) 32 Hz), 137.6, 142.6. MS (m/z, %): 327 [M + 2]+ (2),
325 [M+] (2), 175 (100), 172 (26), 170 (26). Anal. Calcd for
C15H12BrF3O: C, 52.20; H, 3.50. Found: C, 53.3; H, 3.6.
29% of (Z) epoxide was also isolated. H NMR (500 MHz,
CDCl3): δ ) 4.35 (brs, 1H), 4.37 (brs, 1H), 6.88 (d, 3J ) 10
Hz, 1H), 6.90 (d, 3J ) 10 Hz, 1H), 7.14-7.23 (m, 7H). 13C NMR
2
(125 MHz, CDCl3): δ ) 59.1, 59.6, 114.8 (d, JCF ) 22 Hz),
126.8, 127.6, 127.8, 128.5 (d, 3JCF ) 14 Hz), 130.1, 134.1, 163.7
1
(d, JCF ) 245 Hz). MS (m/z,%): 214 [M+] (44), 213 (40), 196
anti-2-Bromo-2-phenyl-1-(4-trifluoromethylphenyl) eth-
anol 6bII was obtained in 55% yield after chromatographic
purification (n-hex/Et2O/benzene ) 8/2/1). 1H NMR (500 MHz,
CDCl3): δ ) 2.55 (brs, 1H), 5.07 (d, 3J ) 6.4 Hz), 5.25 (brd, 3J
(23), 185 (100).
(E)-2-(4-Methoxyphenyl)-3-phenyl-oxirane 5d was ob-
tained in 70% yield after recrystallization in petroleum ether.
The 1H NMR and MS spectra were consistent with literature
data.22
3
) 6.4 Hz), 7.25 (m, 2H), 7.33-7.44 (m, 5H), 7.58 (d, J ) 8.1
Hz, 2H). 13C NMR (125 MHz, CDCl3): δ ) 58.4, 77.5, 125.1,
126.1 (q, 1JCF ) 271 Hz), 127.5, 128.3, 128.4. MS (m/z, %): 327
[M + 2]+ (2), 325 [M+] (2), 175 (100), 172 (26), 170 (26). Anal.
Calcd for C15H12BrF3O: C, 52.20; H, 3.50. Found: C, 53.4; H,
3.7.
(Z)-1-Phenylpropylene oxide was detected in the crude of
alkaline ring closure of a 19/81 7eI/7eII mixture. 1H NMR (300
MHz, CDCl3), in mixture with 5e: overlapping signals but δ
3
3
3
) 1.10 (d, J ) 6.0 Hz), 3.37 (dd, J1 ) 6.0 Hz, J2 ) 4.0 Hz),
3
4.09 (d, J ) 4.0 Hz).
The syn- and anti-2-bromo-2-(4-methoxyphenyl)-1-
phenyl-ethanols 7dI and 7dII were obtained as the only
reaction products in the crude and were analyzed by 1H NMR.
1H NMR (500 MHz, CDCl3), 63/37 mixture of two diastereo-
isomers: δ ) 3.78 (s, 3H, OCH3, 63%), 3.82 (s, 3H, OCH3, 37%),
4.66 (A part of AB system, JAB ) 6.5 Hz, 1H, 63%), 4.69 (B
part of AB system, JAB ) 6.5 Hz, 1H, 63%), 4.78 (A part of AB
system, JAB ) 5.5 Hz, 1H, 37%), 4.82 (B part of AB system,
(E)-2-(2-Methoxyphenyl)-3-phenyloxirane 9 was ob-
tained in 65% yield after chromatographic purification (pe-
1
troleum ether/Et2O ) 9/1). H NMR (300 MHz, CDCl3): δ )
3
3
3.85 (s, 3H), 3.86 (d, J ) 1.8 Hz, 1H), 4.34 (d, J ) 1.8 Hz,
1H), 6.94 (d, 3J ) 8.8 Hz, 1H), 7.06 (dd, 3J1 ) 7.5 Hz, 3J2 ) 7.5
Hz, 1H), 7.33-7.45 (m, 7H). 13C NMR (75 MHz, CDCl3): δ )
55.2, 58.2, 62.2, 110.1, 120.6, 124.9, 125.6, 128.0, 128.3, 128.8,
137.4, 158.0. MS (m/z,%): 226 [M+] (20), 209 (22), 197 (41),
91 (100). Anal. Calcd for C15H14O2: C, 79.62; H, 6.24. Found:
C, 79.8; H, 6.3.
3
JAB ) 5.5 Hz, 1H, 37%), 6.77 (d, J ) 8.0 Hz, 2H, 63%), 6.86
(d, 3J ) 8.5 Hz, 2H, 37%), 7.05 (d, 3J ) 8.0 Hz, 2H, 63%), 7.13
3
(m, 1H), 7.19 (d, J ) 8.5 Hz, 2H, 37%), 7.24-7.34 (m, 4H).
1
34% of (Z) epoxide was also isolated. H NMR (300 MHz,
The syn- and anti-2-bromo-2-(4-methoxyphenyl)-1-
phenyl-ethanol acetates were obtained by quenching the
reaction mixture with 1 mL of a Ac2O/pyridine 1/1 mixture,
3
CDCl3): δ ) 3.72 (s, 3H), 4.42 (A part of AB system, JAB
)
3
4.4 Hz, 1H), 4.46 (B part of AB system, JAB ) 4.4 Hz, 1H),
6.70 (d, 3J ) 8.2 Hz, 1H), 6.84 (dd, 3J1 ) 7.5 Hz, 3J2 ) 7.5 Hz,
1
at -30 °C and usual workup. H NMR (300 MHz, CDCl3) in
3
3
1H), 7.12-7.21 (m, 6H), 7.28 (dd, J1 ) 7.5 Hz, J2 ) 1.6 Hz,
1H). 13C NMR (75 MHz, CDCl3): δ ) 55.1, 57.3, 59.2, 109.7,
119.7, 126.5, 127.1, 127.3, 128.0, 128.4, 134.9, 157.5. MS (m/
z, %): 226 [M+] (20), 209 (22), 197 (41), 91 (100). Anal. Calcd
for C15H14O2: C, 79.62; H, 6.24. Found: C, 79.7; H, 6.2.
(E)-2-(2-Nitrophenyl)-3-phenyloxirane 10 was obtained
in 90% yield after chromatographic purification (petroleum
ether/Et2O ) 4/1). 1H NMR (500 MHz, CDCl3): δ ) 3.83 (d, 3J
mixture (76/24) with ketone 8: δ ) 3.74 (s, 3H, OCH3, 48%),
3.80 (s, 3H, OCH3, 8, 24%), 3.82 (s, 3H, OCH3, 28%), 4.24 (s,
3
3
2H, 8, 24%), 5.18 (d, J ) 8.5 Hz, 1H, 28%), 5.19 (d, J ) 9.5
Hz, 1H, 48%), 6.23 (d, 3J ) 9.5 Hz, 1H, 48%), 6.29 (d, 3J ) 8.5
Hz, 1H, 28%), 6.70-7.60 (m), 8.03 (d, 3J ) 6.6 Hz, 2H, 8, 24%).
MS (for both the acetates) (m/z, %): 350 [M + 2]+ (1), 348 [M+]
(1), 290 (4), 288 (4), 201 (17), 199 (17), 137 (100).
MS (for 8) (m/z, %): 226 [M]+ (26), 121 (100), 105 (83).
The syn- and anti-3-bromo-3-phenyl-2-propanols 7eI
3
) 3.0 Hz, 1H), 4.54 (d, J ) 3.0 Hz, 1H), 7.40-7.45 (m, 5H),
7.54 (m, 1H), 7.75 (m, 2H), 8.20 (d, 3J ) 11 Hz, 1H). 13C NMR
(125 MHz, CDCl3): δ ) 59.9, 62.1, 124.7, 125.8, 126.0, 127.0,
128.5, 128.6, 128.6, 133.9, 134.3, 136.1. MS (m/z, %): 225 [M
- 16]+ (1), 135 (100). Anal. Calcd for C14H11NO3: C, 69.70; H,
4.70. Found: C, 69.5; H, 4.6.
1
and 7eII were obtained quantitatively after the workup. H
NMR (500 MHz, CDCl3), 81/19 mixture of 7eII/7eI: δ ) 1.13
3
3
(d, J ) 6.5 Hz, 3H, CH3, 7eI, 19%), 1.37 (d, J ) 5.5 Hz, 3H,
CH3, 7eII, 81%), 2.26 (brs, 1H, OH, 7eI and 7eII), 4.24 (m,
1H, 7eI and 7eII), 4.88 (d, 3J ) 7.0 Hz, 1H, 7eI, 19%), 4.90 (d,
3J ) 6.0 Hz, 1H, 7eII, 81%), 7.33-7.49 (m, 5H, 7eI and 7eII).
MS (for both 7eI and 7eII) (m/z, %): 216 [M + 2]+ (1), 214
[M+] (1), 172 (32), 170 (32), 91 (100).
(Z)-2-(2-Nitrophenyl)-3-phenyloxirane was detected in the
crude of alkaline ring closure of 13. 1H NMR (300 MHz,
3
3
CDCl3): δ ) 4.59 (d, J1 ) 4.5 Hz, 1H), 4.81 (d, J ) 4.5 Hz,
1H), 7.20 (m, 5H), 7.76 (m, 1H), 7.57 (d, 3J ) 10 Hz, 1H), 7.96
3
(d, J ) 10 Hz, 1H).
syn-2-Bromo-2-(2-methoxyphenyl)-1-phenyl-ethanol 11
and syn-2-Bromo-1-(2-methoxyphenyl)-2-phenyl ethanol
12 were obtained quantitatively after the workup. 1H NMR
(300 MHz, CDCl3), 72/28 mixture of 11/12: δ ) 2.83 (brs, 1H),
3.66 (s, 3H, OCH3, 11, 72%), 3.73 (s, 3H, OCH3, 12, 28%), 5.14
Spectra of syn- and anti-2-bromo-1,2-diphenylethanols 2 and
3 were compared with known data.14
The syn- and anti-2-Bromo-1-(4-nitrophenyl)-2-phen-
ylethanols 6aI and 6aII. Characterization by 1H NMR of the
1
3
3
mixture 6aI/6aII ) 31/69. H NMR (300 MHz, CDCl3): δ )
(d, J ) 8.0 Hz, 1H, 11, 72%), 5.28 (d, J ) 6.8 Hz, 1H, 12,
28%), 5.75 (d, 3J ) 8.0 Hz, 1H, 11, 72%), 5.76 (d, 3J ) 6.8 Hz,
1H, 12, 28%), 6.71 (d, 3J ) 8.4 Hz, 1H, 11, 72%), 6.83 (d, 3J )
8.4 Hz, 1H, 12, 28%), 6.92 (dd, 3J1 ) 7.6 Hz, 3J2 ) 7.5 Hz, 1H,
2.80 (brs, 1H, OH, 6II, 69%), 3.30 (brs, 1H, OH, 6aI, 31%),
3
3
5.06 (d, J ) 8.5 Hz, 1H, CHBr, 6aI, 31%), 5.07 (d, J ) 6.4
3
Hz, 1H, CHBr, 6aII, 69%), 5.15 (d, J ) 8.5 Hz, 1H, CHOH,
3
3
11, 72%), 7.00 (dd, J1 ) 7.6 Hz, J2 ) 7.5 Hz, 1H, 12, 28%),
7.17-7.22 (m, 6H), 7.53 (dd, 3J ) 7.5 Hz, 4J ) 1.3 Hz, 1H, 11,
(22) Majima, T.; Tojo, S.; Ispida, A.; Takamuko, S. J. Org. Chem.
1996, 61, 7793-7800.
3
72%), 7.60 (d, J ) 7.5 Hz, 1H, 12, 28%).
1610 J. Org. Chem., Vol. 70, No. 5, 2005

1,2-Diaryl-2-bromo Alcohols
The crude was acetylated by a 1/1 Ac2O/pyridine mixture
at room temperature. After usual workup, the reaction mixture
(d, 3J ) 4.8 Hz, 1H), 7.31 (m, 3H), 7.43 (m, 3H), 7.65 (m, 1H),
3
7.85 (d, J ) 11.4 Hz, 1H). 13C NMR (75 MHz, CDCl3): δ )
1
was analyzed via H NMR and GC-MS.
62.1, 72.6, 124.6, 128.2, 128.3, 128.7, 128.8, 128.9, 129.4, 129.6,
129.9, 133.1, 135.5, 138.2. Anal. Calcd for C14H12BrNO3: C,
52.20; H, 3.75. Found: C, 52.1; H, 3.6.
syn-2-Bromo-2-(2-methoxyphenyl)-1-phenyl-ethanol and
syn-2-Bromo-1-(2-methoxyphenyl)-2-phenyl-ethanol Ac-
etates. 1H NMR (300 MHz, CDCl3), 72/28 mixture of ac-
etates: δ ) 2.20 (s, 3H, CH3COO), 3.70 (s, 3H, OCH3, 72%),
3.72 (s, 3H, OCH3, 28%), 5.79 (d, 3J ) 9.0 Hz, 1H, CHBr, 72%),
Acknowledgment. We thank the University of
Basilicata for financial support, CNRS-France for a
3-month position to P.L., and Dr. E. Giorgio for helpful
discussions on theoretical calculations.
3
3
5.81 (d, J ) 9.0 Hz, 1H, CHBr, 28%), 6.33 (d, J ) 9.0 Hz,
1H, CHOAc, 72%), 6.35 (d, 3J ) 9.0 Hz, 1H, CHOAc, 28%),
3
6.68 (brd, J ) 8.3 Hz, 1H), 6.88 (m, 2H), 7.17-7.33 (m, 5H),
7.44 (m, 1H). 13C NMR (75 MHz, CDCl3): δ ) 52.2, 55.4, 55.6,
58.1, 77.3, 110.9, 120.7, 120.8, 126.8, 126.9, 127.0, 127.3, 127.4,
127.7, 128.1, 128.3, 128.4, 129.8, 130.0, 130.3, 139.2, 140.0,
155.9, 156.6. MS (for both the acetylated products) (m/z, %):
350 [M + 2]+ (1), 348 [M+] (1), 290 (7), 288 (7), 269 (73), 107
(100).
Supporting Information Available: Spectral data of
compounds 5b-d, 6a,b, 7d,e, 8, 9, 10, 11, 12, 13. 1H and 13
C
NMR spectra of key compounds. Energies and Cartesian
coordinates for structures 5a(H+)A, 5a(H+)B, 5d(H+)A, and
5d(H+)B. This material is available free of charge via the
Internet at http://pubs.acs.org.
syn-2-Bromo-1-(2-nitrophenyl)-2-phenyl ethanol 13 was
1
obtained in 95% yield after chromatographic purification. H
3
NMR (300 MHz, CDCl3): δ ) 5.41 (d, J ) 4.8 Hz, 1H), 5.70
JO048045W
J. Org. Chem, Vol. 70, No. 5, 2005 1611