Conformational studies by dynamic NMR, 77: Stereomutation of the enantiomers of hindered O-substituted oximes

Article abstract of DOI:10.1002/1099-0690(200010)2000:20<3439::AID-EJOC3439>3.0.CO;2-J

As anticipated by Molecular Mechanics calculations, the (E) and (Z) isomers of diaryl ketone oximes containing a bulky substituent (PhS or Ph₂N) in the ortho position of the phenyl ring, display different conformational preferences. Whereas the (E) isomers exhibit a plane of symmetry at any accessible temperature, the (Z) isomers exist as a pair of stereolabile enantiomers that were detected by low-temperature NMR spectroscopy in a chiral environment. In a number of O-substituted oximes, the enantiomerisation barriers of the (Z) isomers were determined by monitoring the line shape of the NMR signals of diastereotopic methylene hydrogen atoms as a function of temperature. The four stereoisomers, generated by the combination of the conformational axial chirality with the configurationally stable chirality of a stereogenic centre, have been all detected in a specifically substituted oxime and monitored in an appropriate chiral environment. The NMR results in solution were confirmed by X-ray diffraction measurements in the solid state.

Full text of DOI:10.1002/1099-0690(200010)2000:20<3439::AID-EJOC3439>3.0.CO;2-J

FULL PAPER
Conformational Studies by Dynamic NMR, 77[‡]
Stereomutation of the Enantiomers of Hindered O-Substituted Oximes
Rino Leardini,[a] Lodovico Lunazzi,[a] Andrea Mazzanti,*[a] Hamish McNab,[b] and
Daniele Nanni*[a]
Keywords: Conformational analysis / Dynamic NMR / MM calculations / X-ray diffraction
As anticipated by Molecular Mechanics calculations, the (E)
and (Z) isomers of diaryl ketone oximes containing a bulky
substituent (PhS or Ph2N) in the ortho position of the phenyl
ring, display different conformational preferences. Whereas
the (E) isomers exhibit a plane of symmetry at any accessible
temperature, the (Z) isomers exist as a pair of stereolabile
enantiomers that were detected by low-temperature NMR
spectroscopy in a chiral environment. In a number of O-sub-
stituted oximes, the enantiomerisation barriers of the (Z) iso-
mers were determined by monitoring the line shape of the
NMR signals of diastereotopic methylene hydrogen atoms as
a function of temperature. The four stereoisomers, generated
by the combination of the conformational axial chirality with
the configurationally stable chirality of a stereogenic centre,
have been all detected in a specifically substituted oxime and
monitored in an appropriate chiral environment. The NMR
results in solution were confirmed by X-ray diffraction meas-
urements in the solid state.
Introduction
the CϭNO plane. In the (E) isomer, on the contrary, the
substituted ring might still be accommodated in an arrange-
ment nearly coplanar to the CϭNO moiety. Such a qualit-
ative prediction has been substantiated by calculations[2]
carried out on diaryl ketone oximes bearing quite a bulky
substituent in the ortho position of one of the two phenyl
rings, as in the case of 1, 2.
The prototype of diaryl ketone oximes, namely benzo-
phenone oxime (Ph2CϭNOH), is expected to display two
different values for the dihedral angles formed by the plane
of the CϭNO moiety with that of each phenyl ring, owing
to the different steric hindrance experienced by the two aro-
matic groups. Although an X-ray structure analysis of
benzophenone oxime itself is not available, molecular mech-
anics calculations[2] suggest that in the isolated molecule,
the ring in the (E) configuration should be nearly coplanar
and the ring in the (Z) configuration significantly twisted
with respect to the CϭNO moiety plane, the calculated di-
hedral angles being 20° and 54°, respectively. The crystal
structures obtained for related molecules such as O-[1-
(benzotriazol-1-yl)-2-methylpropyl]benzophenone oxime,
[3a]
benzophenone O-(2,3,5,6-tetrafluoro-4-pyridyl)oxime
and the two isomeric forms of para-bromobenzo-
[3b]
phenone O-picryloxime,[3c] confirm that the (Z) rings are
Whereas in the calculated ground state (GS) conforma-
tion of the (E) isomers (1b, 2b) the plane of the ortho-substi-
tuted phenyl group turns out to be coplanar with that of
the CϭNOH moiety, in the GS conformation of the (Z)
isomers (1a, 2a) the ortho-substituted phenyl group is calcu-
lated to lie in a plane orthogonal to that of the CϭNOH
moiety. Contrary to the (E) isomers, the (Z) isomers are
therefore chiral molecules and should exhibit a pair of (M)/
(P) conformational enantiomers if the corresponding in-
terconversion barriers are sufficiently large to be amenable
to an experimental verification. A similar situation has been
reported to occur in the (Z) isomers of some ortho-substi-
tuted N-alkyl-C-arylimines.[4]
twisted much more than the (E) rings. As a further example,
the two isomeric para-chlorobenzaldehyde oximes
[3d]
have
the (E) ring coplanar with the plane of the CϭNO moiety
whereas the (Z) ring is twisted out of that plane.
Benzophenone oximes bearing an ortho substituent in
one of the two phenyl groups can also exist as (E) or (Z)
isomers. In the latter case the substituted ring should be
twisted to the point of becoming essentially orthogonal to
[‡]
Part 76: Ref.[1]
[a]
Department of Organic Chemistry ‘‘A.Mangini’’, University of
Bologna,
Viale Risorgimento 4, 40136 Bologna, Italy
Fax: (internat.) ϩ 39-051/209-3654
E-mail: mazzand@ms.fci.unibo.it
The calculated[2] energy profile for the rotation about the
ArϪCN bond (Ar being the ortho-substituted phenyl
group) of 1 and 2 indicates that there are two different
[b]
Department of Chemistry, University of Edinburgh,
Edinburgh, EH9 3JJ, UK
Eur. J. Org. Chem. 2000, 3439Ϫ3446
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/1020Ϫ3439 $ 17.50ϩ.50/0
3439

R. Leardini, L. Lunazzi, A. Mazzanti, H. McNab, D. Nanni
FULL PAPER
transition states (TS) for such an enantiomerisation process
in the (Z) isomers. One (i.e. TS1) corresponds to a planar
situation where the X substituent is syn ( ϭ 0°). The other
transition state (i.e. TS2) corresponds to a planar situation
where the X substituent is anti ( ϭ 180 °) with respect to
the OH group (Table 1). In order to detect these stereolabile
enantiomers, the rotation rate corresponding to the lower
one of the two barriers has to be made sufficiently slow
with respect to the time scale of the technique employed for
their detection. Depending on the nature of the substituent
X, either TS1 or TS2 may correspond to the energy level
yielding the rate-determining step. In the case of 1a (X ϭ
SPh), for instance, the calculated enantiomerisation barrier
is 9.4 kcal molϪ1 (Table 1): Accordingly, at temperatures
close to Ϫ90 °C the corresponding life-time should be long
enough to allow detection by NMR spectroscopy. Clearly,
only in an appropriate chiral environment would separate
NMR signals become observable for these enantiomers.
Table 1. Relative energies (E, in kcal molϪ1) and dihedral angles
(θ) between the ortho-substituted ring and the CϭNO plane in the
ground (GS) and transitions states (TS) of the (Z) isomers 1a and
2a as obtained by MM calculations, see ref.[2]
Scheme 1. Top: X-ray diffraction structure of 1a; bottom: MMX-
calculated structures for the two most stable conformational min-
ima of 1a; the relative energies E are in kcal molϪ1
Compound
GS
E (θ)
TS1
E (θ)
TS2
E (θ)
90° torsion angle between the CϭNOH moiety and the sub-
stituted phenyl in the (Z) isomer is likely to relieve the strain
which occurs in the (E) isomer 1b, where the substituted
phenyl remains coplanar, while the less encumbering unsub-
stituted phenyl ring becomes orthogonal. Under thermo-
dynamic equilibrium conditions, such a large energy differ-
ence would lead to a negligible amount of the (E) isomer
1a
2a
0 (97 °)
13.5 (0 °)
10.2 (0 °)
9.4 (180 °)
11.7 (180 °)
0 (Ϫ102 °)
Results and Discussion
Both the (E) and (Z) isomers of oxime 1 were obtained so that the relatively large yield of the minor isomer (about
approximately in 1:5 proportion, by treating hy- 20%) seems to indicate that kinetic control is somehow op-
a
droxylamine with the corresponding ketone (see Experi- erating in the reaction process. This is supported by the ob-
mental Section). An X-ray diffraction determination did servation that once the isolated isomer 1b is dissolved in a
show that the more abundant isomer (m.p. 139 °C) corre- chlorinated solvent (chloroform or dichloromethane), a
sponds to the (Z) isomer 1a. In Scheme 1 (top) it is shown rapid isomerisation does take place and only the more
how the plane of the ortho-substituted phenyl group is in- stable form 1a can be observed (both isomers apparently
deed orthogonal to that of the CϭNOH moiety, as indic- display, in these solvents, identical 1H and 13C NMR
ated by calculations. The calculated structure corresponding spectra). The synϪanti interconversion of oximes is known
[5]
to the absolute energy minimum is essentially equal to that to be an acid-catalysed process
so that traces of HCl,
experimentally determined, the only difference concerning likely to be present in chlorinated solvents, greatly acceler-
the position of the phenyl ring of the SPh substituent ate such interconversions. On the other hand, when hydro-
(Scheme 1, bottom left). Actually, a theoretical structure carbons are used as solvents, this process is much slower
identical to the experimental one was also found to corre- and in [D8]toluene for instance, the isomer 1b exists suffi-
spond to a conformational minimum (Scheme 1, bottom ciently long to yield NMR spectra different from those of
right), but its energy value appeared to be slightly higher 1a.
(by 0.9 kcal molϪ1). The barrier to rotation about the PhϪS
In CD2Cl2, the 13C NMR spectrum of the (Z) isomer 1a
bonds is quite low, so that the crystal packing apparently displays a single line for the quaternary CϭN carbon atom
favours a conformation which is expected to be moderately at any temperature down to Ϫ85° C. However, in the chiral
less stable in the isolated molecule. As can be seen in environment generated by the addition of an appropriate
Scheme 1, the steric effects of the more crowded conformer amount of an enantiomerically pure Pirkle’s alcohol, such
[6]
adopted in the crystal are undoubtedly larger than those of as (R)-l-ArϪCH(OH)CF3 (Ar ϭ 9-anthryl),
this line
the most stable theoretical conformer, but its more compact splits into a 1:1 doublet (line separation 0.1 ppm) at about
shape probably fits better into the crystal cell. Ϫ85 °C. On warming to 20 °C, a single line is observed
The energy value calculated for the (E) isomer 1b is much again: this proves that the (Z) isomer 1a is comprised of
higher (by 3.7 kcal molϪ1) than for the (Z) isomer 1a. The two stereolabile enantiomers that can be detected only at
3440
Eur. J. Org. Chem. 2000, 3439Ϫ3446

Conformational Studies by Dynamic NMR, 77
FULL PAPER
low temperatures because they interconvert too rapidly at
ambient temperature. This type of experiment, however,
could not be carried out in the case of the (E) isomer 1b
because the Pirkle’s alcohol turned out to be sufficiently
acid to catalyse the rapid interconversion of 1b into 1a even
in [D8]toluene.
Contrary to the case of oxime 1, the (E) isomer 2 does
not interconvert into the (Z) isomer (see Experimental Sec-
tion) in chlorinated solvents. It might be argued that the
catalytic process is hampered by the presence of the amino
group which neutralises possible traces of acid present in
these solutions. Due to this occurrence, we could check that
in a CD2Cl2 solution of the (E) isomer 2b, the signal of the
CϭN carbon atom does not split at Ϫ85 °C in the presence
of excess enantiomerically pure Pirkle’s alcohol, whereas
under the same conditions, the corresponding signal of the
(Z) isomer 2a does split (separation of 0.6 ppm), much in
the same way as reported for the (Z) isomer 1a. It is thus
conclusively ascertained that the (Z) isomers adopt a chiral
conformation whilst the (E) isomers do not.
Figure 1. Temperature dependence of the 400-MHz 1H NMR spec-
trum of 3a in a CD2Cl2 solution containing a 300:1 molar excess
of an enantiomerically pure Pirkle’s alcohol (see text); on the right,
the computer simulation is shown
Although we have proved the existence of two stereolabile
enantiomers for the (Z) isomers 1a and 2a, we were not able
to determine the corresponding enantiomerisation barriers,
owing to the very low signal to noise ratio of the CϭN
quaternary carbon lines. To overcome this difficulty, we pre-
pared the O-methyl derivatives of both the isomers of 1 (i.e.
3a and 3b), which are quite stable and do not interconvert
so easily into each other. Incidentally this fact seems to sup-
port the hypothesis that the interconversion of 1b into 1a is
facilitated by the intermediacy of a C-nitroso tautomeric
form,[7] that obviously cannot occur in O-alkyl derivatives.
In the presence of the mentioned Pirkle’s alcohol,[6] one
1
of the two O-methyl isomers (3a) displays a methyl H sig-
nal which broadens on lowering the temperature and even-
tually splits into a pair of lines below Ϫ15 °C, whereas the
corresponding signal of the other isomer (3b) does not split
even at Ϫ100 °C. Thus the chiral structure (Z) and the achi-
ral structure (E) could be confidently assigned to 3a and to
3b. The spectral simulation (Figure 1) yields the rate con-
stants, hence the free energy of activation, for the enantiom-
erisation process (14.9 Ϯ 0.3 kcal molϪ1, as in Table 2).
An alternative method to measure the enantiomerisation
barriers requires the introduction of a probe sensitive to the
chirality, for instance the methylene moiety, which displays
diastereotopic hydrogen atoms in asymmetric molecules.
For this reason we synthesised a number of (Z) isomers
(4a؊9a) containing a CH2 group and prepared, for com-
parison, the corresponding (E) isomers 6bϪ9b.
Figure 2. Experimental (left) and computer-simulated (right) 1H
methylene signal of 8a as a function of temperature in CDCl3 at
300 MHz
It is worth mentioning that contrary to the case of oximes
1 and 2, the reaction leading to oxime 8 yields an amount
of the (E) isomer larger than that of the (Z) isomer (see
Experimental Section): a further indication that the reac-
tion is under kinetic control.
1
The H NMR spectra of the (E) isomers 6b؊9b always
display a single line for the CH2 hydrogen atoms at any
Eur. J. Org. Chem. 2000, 3439Ϫ3446
3441

R. Leardini, L. Lunazzi, A. Mazzanti, H. McNab, D. Nanni
FULL PAPER
Table 2. Experimental free energies of activation (∆G϶ in kcal molϪ1) for the enantiomerisation of the (Z) isomers 3a؊10a; the shift
separations (∆ν) are in Hz at the appropriate temperature (°C)
Compound
∆G϶
Monitored signal
∆ν (T)
J [Hz]
Solvent
Frequency [MHz]
[a]
3a
4a
5a
6a
7a
8a
9a
14.9
14.4
14.8
14.8
14.9
14.2
14.5
14.5
14.7[b]
CH3
4.8 (Ϫ30°)
8.5 (Ϫ30°)
92 (Ϫ20°)
70 (Ϫ10°)
26 (Ϫ20°)
18 (Ϫ50°)
53 (Ϫ20°)
22 (Ϫ20°)
13.1 (Ϫ20°)
Ϫ
CD2Cl2
CD2Cl2
CDCl3
CDCl3
CDCl3
CD2Cl2
CDCl3
400
400
300
400
400
300
300
CH2
Ϫ10.2
Ϫ17.3
Ϫ16.2
Ϫ16.6
Ϫ14.8
Ϫ16.4
Ϫ14.2
ϩ6.3
CH2
CH2
CH2
CH2
CH2ϪCOO
CH2ϪPh
CH3ϪCH
10a
CD2Cl2
400
[a]
[b]
In the presence of a 300:1 molar excess of an enantiopure Pirkle’s alcohol (see text). Ϫ In this case the barrier actually refers to an
epimerisation process about the chiral axis.
attainable temperature: For instance even at Ϫ135 °C in difference (0.1 kcal molϪ1) experimentally observed for the
dimethyl ether, the 400-MHz NMR spectrum of 8b showed analogous pair 6a and 7a.
that these hydrogen atoms remain enantiotopic, as expected
Having demonstrated that the (Z) isomers are chiral and
for a nonchiral molecule. On the contrary, in all the (Z) have enantiomerisation barriers high enough to be experi-
isomers 4aϪ9a anisochronous methylene signals were ob- mentally measurable, it can consequently be anticipated
served on cooling, since the molecular asymmetry renders that the introduction of a configurationally stable chiral
these hydrogen atoms diastereotopic. In a typical example, centre would generate a pair of diastereoisomers in unequal
reported in Figure 2, the single CH2 line of 8a splits into proportions. For instance, if the OH hydrogen atom of the
an AB-type spectrum below Ϫ20 °C, and the accompanying oxime 1a is replaced by a CHMeEt group, yielding the cor-
computer simulation allowed us to determine the appropri-
ate rate constants for the enantiomerisation process. X-ray
diffraction confirmed that 8a corresponds to a (Z) structure
where the ortho-substituted phenyl group has its plane or-
thogonal to that of the CϭNO moiety.
The free energies of activation (∆G϶) for these com-
pounds are given in Table 2. Within the experimental errors
they are temperature-independent, suggesting that the ac-
tivation entropy is negligible, as is usually observed in con-
formational processes. Also, the barriers measured by mon-
itoring the CH2 signals cover a very restricted range of
values (from 14.2 to 14.8 kcal molϪ1) for all the examined
derivatives containing the PhS substituent (i.e. 4aϪ6a, 8a,
9a), regardless of the group bonded to the oxygen atom of
the oxime. This is a clear indication that the same motion,
i.e. rotation of the ortho-substituted phenyl ring, is respons-
ible for the observed dynamic processes.
When the diphenylamino derivative 7a is investigated, its
enantiomerisation barrier appears to be essentially equal to
that of the corresponding derivative 6a (Table 2), despite
the apparent larger dimension of the Ph2N with respect to
the PhS moiety. The theoretical study of the rotation path-
way carried out on the simpler corresponding oxime 2a
(Table 1), had shown that the transitions state TS1 was the
one related to the lowest energy level, rather than TS2 as in
the case of the PhS-substituted oxime 1a. The inter-system
crossing of the transition states thus makes the correspond-
ing barrier lower than that expected if the same TS2 trans-
ition state had been adopted by both 1a and 2a. The calcu-
lated difference between the two barriers (Table 1) in fact
becomes only 0.8 (i.e. 10.2Ϫ9.4), rather than 2.3 (i.e.
11.7Ϫ9.4) kcal molϪ1. Within the approximations of this
Scheme 2. Theoretical structures of the enantiomers (PS), (MR) of
diastereoisomer 10aЈ (top) and of the enantiomers (PR), (MS) of
type of calculations a theoretical difference as small as 0.8
the diastereoisomer 10a" (bottom); the calculated relative energies
kcal molϪ1 compares reasonably well with the negligible E are in kcal molϪ1
3442
Eur. J. Org. Chem. 2000, 3439Ϫ3446

Conformational Studies by Dynamic NMR, 77
FULL PAPER
responding derivative 10a, calculations predict that both 10a", which comprises the pair (PR)/(MS). Indeed most of
1
such diastereoisomers (10aЈ and 10a") correspond to a min- the H NMR aliphatic signals of 10a in CD2Cl2 broaden
imum energy. These diastereoisomers result from the com- on cooling, eventually yielding two almost equally intense
bination of the stereolabile axial chirality (M) with the con- spectra below 0 °C. Computer line-shape simulation indic-
figurationally stable chirality (R) or (S) of the stereogenic ated that the barrier for the interconversion of the more
centre at the asymmetric carbon atom:[8] They are enanti- (51%) into the less (49%) stable[9] diastereoisomer (14.7 kcal
omerically related, to the pair derived from the combination molϪ1 as in Table 2) is essentially the same as those meas-
of the (P) with the (S) or (R) chirality, as shown in ured for the other derivatives. In Figure 3 it is shown how
Scheme 2.
the doublet (JMe,CH ϭ 6.3 Hz) observed at ambient temper-
According to the theory, the diastereoisomer 10aЈ, com- ature for the methyl group bonded to the CHEt moiety of
prising the enantiomers (PS) and (MR) (Scheme 2, top), 10a splits at Ϫ30 °C into two doublets separated by
should be slightly more stable (by only 0.4 kcal molϪ1) than 0.035 ppm. When at the same temperature the environment
is rendered chiral by addition of an appropriate amount of
the mentioned Pirkle’s alcohol, [6] the two pairs split further,
yielding four pairs of lines since all four stereoisomers (PS),
(MR), (PR), (MS), as in Scheme 2, now display distinguish-
able NMR spectra.
Experimental Section
Materials: The oximes 1, 2, and 8 were prepared by treacting the
appropriate ketone with hydroxylamine hydrochloride in EtOH and
pyridine.[10] After heating at reflux for a few hours, the volume was
reduced and the residue poured into iced water, extracted with di-
ethyl ether and dried (sodium sulfate). After removal of the solvent,
the residue was crystallised from light petroleum ether/benzene.
Typical yields were in the range 70Ϫ80%.
Phenyl[(2-phenylsulfanyl)phenyl]methanone Oxime (1): Two isomers
(E), (Z), in a 1:5 proportion, were obtained (total yield 85%) from
[11]
phenyl[(2-phenylsulfanyl)phenyl]methanone
(26.1 g, 90 mmol)
and hydroxylamine hydrochloride (23.3 g, 76.4 mmol). They were
separated by chromatography; eluent: light petroleum ether (40Ϫ70
°C)/diethyl ether. Ϫ (Z) Isomer (1a): M.p. 138Ϫ140° C. Ϫ MS; m/z
(%): 305 (23) [Mϩ], 288 (100), 212 (22), 77 (25). Ϫ 1H NMR
(CDCl3, 200 MHz): δH
ϭ 7.16Ϫ7.40 (m, 12 H, aromatic),
7.42Ϫ7.50 (m, 2 H, aromatic), 9.12 (br. s, 1 H, OH). Ϫ 13C NMR
(CD2Cl2, 100.6 MHz): δC ϭ 126.8, 127.2, 127.81, 128.77, 128.9,
129.4, 129.9, 130.1, 131.5, 132.3, 133.7, 134.2, 134.5, 135.3, 156.5.
Ϫ 1H NMR ([D8]toluene, 75.5 MHz, at Ϫ30 °C): δC ϭ 157.30 (Cϭ
N). Ϫ (E) Isomer (1b): M.p. 111Ϫ113 °C. Ϫ MS; m/z (%): 305
(23) [Mϩ], 288 (100), 212 (22), 77 (25). Ϫ 13C NMR ([D8]toluene,
75.5 MHz, at Ϫ30 °C): δC ϭ 156.7 (CϭN). The difference of
0.6 ppm between the CϭN signals of 1a and 1b was also checked
by recording the spectrum of a [D8]toluene solution containing a
mixture of both isomers. Ϫ C19H15NOS (305.4): calcd. C 74.73, H
4.95, N 4.59, S 10.50; found C 74.65, H 4.92, N 4.65, S 10.46.
[(2-Diphenylamino)phenyl]phenylmethanone Oxime (2): Two isomers
(E), (Z), in a 1:3 proportion, were obtained (total yield 82%) from
[12]
[(2-diphenylamino)phenyl]phenylmethanone
(10.5 g, 30 mmol)
and hydroxylamine hydrochloride (8.9 g, 24.6 mmol). They were
separated by chromatography; eluent: light petroleum ether (40Ϫ70
°C)/diethyl ether. Ϫ (Z) Isomer (2a): M.p. 150Ϫ152 °C. Ϫ MS; m/z
(%): 364 (92) [Mϩ], 347 (36), 332 (100), 272 (20), 167 (12), 77 (16).
Figure 3. At ambient temperature, the 1H NMR spectrum (in
CD2Cl2 at 300 MHz) of the (Z) isomer 10a displays a doublet (J ϭ
6.3 Hz) for the methyl group bonded to the CHEt moiety (top); at
Ϫ30 °C, two doublets (separated by 0.035 ppm) due to the diaster-
eoisomers 10aЈ and 10a" are observed (middle trace); underneath
is reported the spectrum (also at Ϫ30 °C) in a chiral environment
(15:1 molar excess of Pirkle’s alcohol[6] showing the four doublets
expected for the four stereoisomers of Scheme 2; the two doublets
(downfield) due to the pair of enantiomers of the minor (49%) dias-
tereoisomer are separated by 0.009 ppm, those for the major (51%)
diastereoisomer (upfield) are separated by 0.007 ppm
Ϫ
1H NMR (CDCl3, 300 MHz): δH ϭ 6.86Ϫ6.78 (m, 5 H, aro-
matic), 7.00Ϫ7.06 (m, 4 H, aromatic), 7.12Ϫ7.26 (m, 6 H, aro-
matic), 7.32Ϫ7.40 (m, 4 H, aromatic), 8.30 (br. s, 1 H, OH). Ϫ 13C
NMR (CDCl3, 75.5 MHz): δC ϭ 122.5, 123.69, 123.74, 127.1,
127.7, 127.8, 128.6, 129.1, 129.5, 130.1, 131.6, 135.5, 146.5, 147.5,
156.5. Ϫ IR: ν˜max ϭ 3560 cmϪ1. Ϫ C25H20N2O (364.4): calcd. C
82.39, H 5.53, N 7.69; found C 82.33, H 5.50, N 7.73. Ϫ (E) Isomer
(2b): M.p. 161Ϫ163 °C. Ϫ MS; m/z (%): 364 (98) [Mϩ], 347 (39),
Eur. J. Org. Chem. 2000, 3439Ϫ3446
3443

R. Leardini, L. Lunazzi, A. Mazzanti, H. McNab, D. Nanni
129.2, 129.3, 129.9, 130.9, 131.7, 132.2, 133.1, 135.6, 137.1, 137.7,
FULL PAPER
332 (100), 272 (22), 167 (11), 77 (13). Ϫ 1H NMR (CDCl3,
300 MHz): δH ϭ 6.64Ϫ6.70 (m, 3 H, aromatic), 6.80Ϫ7.36 (m, 15 155.5. Ϫ All the other O-substituted oximes were prepared by treat-
H, aromatic), 7.42Ϫ7.48 (dd, 1 H, aromatic), 8.74 (br. s, 1 H, OH).
ing the (E) or the (Z) isomer of the appropriate oxime with a se-
13C NMR (CDCl3, 75.5 MHz): δC ϭ 122.2, 123.2, 124.7, 127.4, lected halide in DMSO with K2CO3. The temperature varied de-
128.68, 128.72, 129.3, 129.5, 130.4, 132.2, 132.6, 134.4, 146.4, pending on the reactants: typical yields were in the range 70Ϫ80%.
Ϫ
147.6, 156.2. Ϫ IR: ν˜max ϭ 3560 cmϪ1
.
(Z)-Phenyl[(2-phenylsulfanyl)phenyl]methanone O-Neopentyloxime
(4a): The reaction (yield 80%) was carried out at 80 °C using 1-
bromo-2,2-dimethylpropane (1.7 g, 4.5 mmol) and oxime 1a (1.5 g,
5 mmol). Ϫ MS; m/z (%): 375 (14) [Mϩ], 288 (100), 212 (12), 184
2-Phenyl-1-[2-(phenylsulfanyl)phenyl]-1-ethanone Oxime (8): Two
isomers (E), (Z), in a 3:1 proportion (total yield 75%) were ob-
tained from 2-phenyl-1-[2-(phenylsulfanyl)phenyl]-1-ethanone
(6.1 g, 20 mmol) and hydroxylamine hydrochloride (4.8 g,
15.0 mmol). They were separated by chromatography; eluent: light
petroleum ether (40Ϫ70 °C)/diethyl ether. Ϫ (Z) Isomer (8a): M.p.
100Ϫ101° C. Ϫ MS; m/z (%): 319 (25) [Mϩ], 302 (100), 224 (35),
212 (9), 211(15), 210 (13), 92 (8), 91 (85), 77 (17). Ϫ 1H NMR
(CDCl3, 200 MHz): δH ϭ 3.80 (2 H, br. s, CH2), 6.81Ϫ6.88 (m, 1
H aromatic), 7Ϫ10Ϫ7.30 (m, 13 H, aromatic), 8.28 (br. s, 1 H,
OH). Ϫ 13C NMR (CDCl3, 50.3 MHz): δC ϭ 40.6, 125.5, 125.70,
125.76, 127.19, 127.23, 127.9, 128.0, 128.4, 129.8, 131.5, 132.4,
134.6, 134.9, 135.1, 156.5. Ϫ (E) Isomer (8b): M.p. 90Ϫ91 °C. Ϫ
MS; m/z (%): 319 (23) [Mϩ], 302 (100), 224 (19), 212 (19), 211(28),
210 (25), 92 (53), 91 (89), 77 (36). Ϫ 1H NMR (CDCl3, 200 MHz):
δH ϭ 4.18 (s, 2 H, CH2), 7.08Ϫ7.30 (m, 14 H, aromatic), 9.18 (br.
s, 1 H, OH). Ϫ 13C NMR (CDCl3, 50.3 MHz): δC ϭ 33.9, 125.1,
125.5, 126.1, 127.1, 128.04, 128.11, 128.2, 128.8, 130.3, 130.9,
134.3, 134.6, 134.8, 136.1, 157.1. Ϫ C20H17NOS (319.4): calcd. C
75.20, H 5.36, N 4.38, S 10.04; found C 75.26, H 5.39, N 4.34,
S 10.07.
1
(13), 77 (41). Ϫ H NMR (CDCl3, 200 MHz): δH ϭ 0.89 (s, 9 H,
tBu), 3.90 (s, 2 H, CH2), 7.13Ϫ7.38 (m, 12 H, aromatic), 7.47Ϫ7.55
(m, 2 H, aromatic). Ϫ 13C NMR (CDCl3, 50.3 MHz): δC ϭ 27.52,
32.86, 85.34, 127.44, 127.59, 127.66, 128.96, 129.63, 129.74, 129.89,
132.23, 132.89, 135.51, 136.23, 136.50, 136.99, 155.44.
Ϫ
C24H25NOS (375.5): calcd. C 76.76, H 6.71, N 3.73, S 8.54; found
C 76.81, H 6.74, N 3.68, S 8.51.
(Z)-Phenyl[(2-phenylsulfanyl)phenyl]methanone O-(3,3-Dimethyl-2-
oxobutyl)oxime (5a): The reaction (yield 84%) was carried out at
20 °C using 1-bromo-3,3-dimethylbutan-2-one (1.75 g, 4.2 mmol)
and oxime 1a (1.5 g, 5 mmol). Ϫ MS; m/z (%): 403 (15) [Mϩ], 290
1
(21), 289 (29), 288 (100), 212 (34), 184 (25), 77 (39), 57 (85). Ϫ H
NMR (CDCl3, 300 MHz): δH ϭ 1.13 (s, 9 H, tBu), 4.86 (br. m, 2
H, CH2), 7.12Ϫ7.40 (m, 11 H, aromatic), 7.41Ϫ7.50 (m, 3 H, aro-
matic). Ϫ 13C NMR (CDCl3, 75.5 MHz): δC ϭ 26.9, 41.8, 75.2,
127.5, 127.7, 127.94, 128.86, 129.5, 130.00, 130.06, 130.2, 132.1,
133.2, 135.1, 135.7, 136.5, 136.6, 157.4, 211.1. Ϫ IR: ν˜max ϭ 1720
(CϭO) cmϪ1. Ϫ C25H25NO2S (403.5): calcd. C 74.41, H 6.24, N
3.47, S 7.95; found C 74.36, H 6.21, N 3.50, S 7.99.
2-Phenyl-1-[2-(phenylsulfanyl)phenyl]-1-ethanone:
Thiophenol
(13.4 g, 122 mmol) was added dropwise to an EtOH solution of
sodium ethoxide (6.8 g, 100 mmol), and the mixture was warmed
for 20 min at 50 °C. The solvent was removed and the residue
dissolved in dimethylformamide. Subsequently 1-(2-chlorophenyl)-
2-phenyl-1-ethanone[13] (23.2 g, 100 mmol) was added dropwise
and the solution was heated at reflux for 3 h. The volume of the
solution was reduced and, after addition of water, the organic res-
idue was extracted with diethyl ether, dried, treated with carbon
and filtered. The compound (16.7 g, yield 55%) was crystallised
from light petroleum ether/benzene. Ϫ M.p. 72Ϫ73 °C. Ϫ MS; m/z
(%): 304 (14) [Mϩ], 213 (100), 184 (28). Ϫ 1H NMR (CDCl3,
200 MHz): δH ϭ 4.28 (s, 2 H, CH2), 6.93 (dd, 1 H, J1 ϭ 7.9 Hz,
J2 ϭ 1.4 Hz, aromatic), 7.08Ϫ7.49 (m, 12 H, aromatic), 7.80 (dd,
1 H, J1 ϭ 7.4 Hz, J2 ϭ 1.9 Hz, aromatic). Ϫ 13C NMR (CDCl3,
50.3 MHz): δC ϭ 47.8, 125.3, 127.5, 129.2, 129.3, 129.5, 130.1,
130.2, 130.6, 132.5, 134.1, 135.0, 135.1, 135.6, 142.1, 199.6. Ϫ IR:
ν˜max ϭ 3000, 1660 cmϪ1. Ϫ C20H16OS (304.4): calcd. C 78.91, H
5.30, S 10.53; found C 78.86, H 5.33, S 10.49.
(Z)-tert-Butyl
2-[({Phenyl[2-(phenylsulfanyl)phenyl]methylidene}-
amino)oxy]acetate (6a): The reaction (yield 91%) was carried out at
20 °C using tert-butyl chloroacetate (1.9 g, 4.53 mmol) and oxime
1a (1.5 g, 5 mmol). Ϫ MS; m/z (%):419 (16) [Mϩ], 288 (100), 212
(13), 184 (14), 77 (15), 57 (33). Ϫ 1H NMR (CDCl3, 300 MHz):
δH ϭ 1.48 (s, 9 H, tBu), 4.50 (br. m, 2 H, CH2), 7.12Ϫ7.39 (m, 12
H, aromatic), 7.43Ϫ7.50 (m, 2 H, aromatic. Ϫ 13C NMR (CDCl3,
50.3 MHz): δC ϭ 28.2, 71.7, 81.5, 127.04, 127.11, 127.2, 128.3,
128.9, 129.41, 129.45, 129.55, 131.5, 132.6, 134.7, 135.1, 135.7,
135.8, 156.7, 169.0.
C25H25NO3S (419.5): calcd. C 71.57, H 6.01, N 3.34, S 7.64; found
C 71.62, H 6.05, N 3.37, S 7.58.
Ϫ IR: ν˜max ϭ . Ϫ
1750 (CϭO) cmϪ1
tert-Butyl
(E)-2-[({Phenyl[2-(phenylsulfanyl)phenyl]methylidene}-
amino)oxy]acetate (6b): The reaction was carried out at 50 °C using
tert-butyl chloroacetate and oxime 1b in the same proportion as in
the previous case. Ϫ MS; m/z (%): 419 (15) [Mϩ], 288 (100), 212
(10), 184 (24), 77 (12), 57 (33). Ϫ 1H NMR (CDCl3, 200 MHz):
δH ϭ 1.47 (s, 9 H, tBu), 4.62 (s, 2 H, CH2), 7.10Ϫ7.40 (m, 12 H,
m, aromatic), 7.58Ϫ7.65 (m, 2 H, aromatic). Ϫ 13C NMR (CDCl3,
50.3 MHz): δC ϭ 28.1, 71.8, 81.6, 126.3, 127.4, 127.8, 129.1, 129.4,
129.5, 130.1, 130.9, 131.5, 132.4, 133.0, 135.4, 137.2, 137.4, 147.0,
Phenyl[2-(phenylsulfanyl)phenyl]methanone O-Methyloxime (3):
Two isomers (E), (Z) in a 1:5 proportion (yield 85%) were obtained
[11]
by treating phenyl[(2-phenylsulfanyl)phenyl]methanone
(2.9 g,
10 mmmol) with O-methylhydroxylamine hydrochloride[10] (2.7 g,
8.5 mmol) in EtOH and pyridine. They were separated by chroma-
tography; eluent: light petroleum ether(40Ϫ70 °C)/diethyl ether. Ϫ
169.0. Ϫ IR: ν˜max ϭ 1745 (CϭO) cmϪ1
.
tert-Butyl (Z)-2-[({[2-(Diphenylamino)phenyl]phenylmethylidene}-
(Z) Isomer (3a): MS; m/z (%): 319 (16) [Mϩ], 288 (100), 184 (15), amino)oxy]acetate (7a): The reaction (yield 80%) was carried out at
77(13). Ϫ 1H NMR (CDCl3, 300 MHz): δH ϭ 3.92 (s, 3 H, Me), 40 °C using tert-butyl chloroacetate (3.8 g, 8.0 mmol) and oxime
7Ϫ18Ϫ7.60 (m, 14 H, aromatic). Ϫ 13C NMR (CDCl3, 75.5 MHz): 2a. Ϫ MS; m/z (%): 478 (85) [Mϩ], 422 (35), 347 (100), 243 (28),
1
δC ϭ 62.6, 126.4, 127.3, 127.8, 129.0, 129.2, 129.3, 129.9, 130.8, 167 (16), 57 (49). Ϫ H NMR (CDCl3, 300 MHz): δH ϭ1.42 (s, 9
131.7, 132.2, 134.64, 135.3, 135.9, 136.5, 155.6. Ϫ C20H17NOS H, tBu), 4.23 (br. s, 2 H, CH2), 6.74Ϫ6.90 (m, 6 H, aromatic),
(319.4): calcd. C 75.20, H 5.36, N 4.38, S 10.04; found C 75.22, H
7.05Ϫ7.36 (m, 12 H, aromatic), 7.49Ϫ7.54 (dd, 1 H, aromatic). Ϫ
5.32, N 4.43, S 10.01. Ϫ (E) Isomer (3b): MS; m/z (%): 319 (9)
13C NMR (CDCl3, 75.5 MHz): δC ϭ 28.1, 71.1, 81.3, 122.2, 123.6,
1
[Mϩ], 288 (100), 184 (16), 77(11). Ϫ H NMR (CDCl3, 300 MHz): 123.7, 127.30, 127.35, 127.6, 128.4, 129.0, 129.8, 130.2, 131.9,
δH ϭ 3.99 (s, 3 H, Me), 7.15Ϫ7.58 (m, 14 H, aromatic). Ϫ 13C 135.5, 146.0, 147.5, 156.3, 169.0. Ϫ C31H30N2O3 (478.6): calcd. C
NMR (CDCl3, 75.5 MHz): δC ϭ 62.4, 126.4, 127.80, 127.83, 129.0,
77.80, H 6.32, N 5.85; found C 77.76, H 6.29, N 5.90.
Eur. J. Org. Chem. 2000, 3439Ϫ3446
3444

Conformational Studies by Dynamic NMR, 77
FULL PAPER
tert-Butyl
(E)-2-[({[2-Diphenylamino)phenyl]phenylmethylidene}-
X-ray Diffraction
amino)oxy]acetate (7b): The reaction was carried out as above at
40 °C using tert-butyl chloroacetate and oxime 2b. Ϫ MS; m/z (%):
478 (80) [Mϩ], 422 (35), 347 (100), 243 (25), 167 (11), 57 (48). Ϫ
1H NMR (CDCl3, 300 MHz): δH ϭ 1.42 (s, 9 H, tBu), 4.38 (s, 2 H,
CH2), 6.68Ϫ6.75 (m, 4 H aromatic), 6.84Ϫ6.92 (m, 2 H, aromatic),
7.02Ϫ7.36 (m, 12 H, aromatic), 7.44Ϫ7.49 (dd, 1 H, aromatic). Ϫ
13C NMR (CDCl3, 75.5 MHz): δC ϭ 28.1, 71.3, 81.2, 122.0, 123.0,
124.7, 127.2, 128.5, 128.9, 129.1, 129.5, 130.3, 132.3, 132.8, 134.3,
146.3, 147.5, 156.4, 168.5.
¯
Crystal Data of 1a: C19H15NOS (305.38), triclinic, P1, Z ϭ 2, a ϭ
˚
7.6341(5), b ϭ 8.9532(6), c ϭ 12.3720(8) A, α ϭ 71.584(2), β ϭ
3
˚
87.455(2), γ ϭ 87.344(2)°, V ϭ 801.06(9) A , Dcalcd. ϭ 1.266 g
cmϪ3, F(000) ϭ 320, µMoϭ 0.203 cmϪ1, T ϭ 293 K. Data were
collected using a graphite-monochromated Mo-Kα radiation (λ ϭ
˚
0.71073 A) range 1.74° Ͻ θ Ͻ 25.00°. Of 6741 reflections measured,
2825 were found to be independent (Rint ϭ 0.0303), 2227 of which
were considered as observed [I Ͼ 2σ(I)], and were used in the re-
finement of 160 parameters leading to a final R1 of 0.0472 and a
final Rall of 0.0593. The structure was solved by direct methods
and refined by full-matrix least squares on F2, using SHELXTL 97
program packages. In refinements weights according to the scheme
w ϭ [σ2(F2o) ϩ (0.0948P)2 ϩ 0.0360P]Ϫ1 were used where P ϭ (F2o
ϩ 2F2c)/3. The hydrogen atoms, including the OH hydrogen atom,
were located by geometric calculations and refined using a ‘‘riding’’
method; wRobs and wR2 were equal to 0.1411 and 0.1472, respect-
ively. The goodness of fit parameter S was 1.087. Largest difference
tert-Butyl
(Z)-2-[({Phenyl[2-(phenylsulfanyl)phenyl]ethylidene}-
amino)oxy]acetate (9a): The reaction (yield 80%) was carried out at
80 °C using tert-butyl chloroacetate (1.7 g, 3.98 mmol) and oxime
1a (1.5 g, 5 mmol). Ϫ MS; m/z (%): 433 (5) [Mϩ], 302 (100), 224
(10), 211 (10), 91 (67), 57 (26). Ϫ 1H NMR (CDCl3, 200 MHz):
δH ϭ 1.50 (s, 9 H, tBu), 3.84 (br. s, 2 H, CH2), 4.38 (br. s, 2 H,
CH2), 6.92Ϫ6.98 (m, 1 H, aromatic), 7.10Ϫ7.35 (m, 13 H, aro-
matic). Ϫ 13C NMR (CDCl3, 50.3 MHz): δC ϭ 27.9, 41.3, 71.0,
81.2, 126.4, 126.9, 128.1, 128.6, 128.7, 128.8, 129.3, 130.3, 132.4,
132.9, 135.7, 136.34 136.6, 157.4, 168.9. Ϫ IR: ν˜max ϭ 1750 (Cϭ
O) cmϪ1. Ϫ C26H27NO3S (433.6): calcd. C 72.03, H 6.28, N 3.23,
S 7.39; found C 72.00, H 6.25, N 3.20, S 7.44.
peak and hole was 0.256 and Ϫ0.256 eAϪ3. Crystallographic data
˚
(excluding structure factors) for the structure(s) reported in this
paper have been deposited with the Cambridge Crystallographic
Data Center as supplementary publication no. CCDC-141678.
Copies of the data can be obtained free of charge on application
to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK [Fax: (in-
ternat.) ϩ 44-1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
tert-Butyl
(E)-2-[({Phenyl[2-(phenylsulfanyl)phenyl]ethylidene}-
amino)oxy]acetate (9b): The reaction was carried out as above at
60 °C by using tert-butyl chloroacetate and oxime 1b. Ϫ M.p.
68Ϫ69 °C. Ϫ MS; m/z (%): 433 (6) [Mϩ], 302 (100), 224 (7), 211
(9), 91 (55), 57 (18). Ϫ 1H NMR (CDCl3, 200 MHz): δH ϭ 1.52 (s,
9 H, tBu), 4.28 (s, 2 H, CH2), 4.68 (s, 2 H, CH2), 6.90Ϫ7.38 (m,
14 H, aromatic). Ϫ 13C NMR (CDCl3, 50.3 MHz): δC ϭ 28.2, 35.8,
71.5, 81.6, 126.4, 126.5, 127.3, 128.4, 129.20, 129.28, 129.5, 129.9,
131.8, 131.9, 135.86, 135.91, 135.99, 136.7, 159.0, 169.1. Ϫ IR:
Crystal Data of 8a: C20H17NOS (319.41), monoclinic, P21/c, Z ϭ
˚
4, a ϭ 9.0807(5), b ϭ 13.4441(8), c ϭ 13.7268(8) A, β ϭ 95.878(2)°,
V ϭ 1680.75(12) A , Dcalcd. ϭ 1.262 g cmϪ3, F(000) ϭ 672, µMoϭ
3
˚
0.196 cmϪ1, T ϭ 293 K. Data were collected using a graphite-mon-
˚
ochromated Mo-Kα radiation (λ ϭ 0.71073 A) range 2.12° Ͻ θ
ν˜max ϭ 1745 (CϭO) cmϪ1
.
Ͻ 25.00°. Of 13827 reflections measured, 2963 were found to be
independent (Rint ϭ 0.0462), 2241 of which were considered as ob-
served [I Ͼ 2σ(I)], and were used in the refinement of 172 para-
meters leading to a final R1 of 0.0494 and a final Rall of 0.0628.
The structure was solved by direct methods and refined by full-
matrix least squares on F2, using SHELXTL 97 program packages.
In refinements weights according to the scheme w ϭ [σ2(F2o) ϩ
(0.0827P)2]Ϫ1 were used, where P ϭ (F2o ϩ 2F2c)/3. The hydrogen
atoms were located by geometric calculations and refined using a
‘‘riding’’ method; wRobs and wR2 were equal to 0.1572 and 0.1636,
respectively. The goodness of fit parameter S was 1.250. Largest
(Z)-Phenyl[2-(phenylsulfanyl)phenyl]methanone
O-(1-Methylpro-
pyl)oxime (10a): The reaction was carried out at 40 °C using 2-
bromobutane (2.65 g, 7.3 mmol) and oxime 1a (3.0 g, 10 mmol). Ϫ
MS; m/z (%): 361 (12) [Mϩ], 288 (100), 212 (7), 197 (15), 184 (20),
1
77 (23). Ϫ H NMR (CDCl3, 300 MHz): δH ϭ 0.88 (t, 3 H, Me),
1.20, (d, 3 H, Me), 1Ϫ40Ϫ1.78 (m, 2 H, CH2), 4.18Ϫ4.25 (m, 1 H,
CH), 7.10Ϫ7.38 (m, 12 H, aromatic), 7.45Ϫ7.55 (m, 2 H, aromatic).
Ϫ
13C NMR (CDCl3, 75.5 MHz): δC ϭ 9.8, 19.3, 28.4, 81.2, 126.71,
126.76, 127.0, 128.2, 128.8 (probably comprising two carbon
atoms), 128.9, 129.3, 131.3, 132.5, 134.8, 136.03, 136.2, 136.8,
154.3. Ϫ C23H23NOS (361.5): calcd. C 76.42, H 6.41, N 3.87, S
8.87; found C 76.35, H 6.38, N 3.84, S 8.92.
Ϫ3
˚
difference peak and hole was 0.279 and Ϫ0.263
e
A
.
Crystallographic data (excluding structure factors) for the struc-
ture(s) reported in this paper have been deposited with the Cam-
bridge Crystallographic Data Center as supplementary publication
no. CCDC-141678 (1a) and CCDC-141679 (8a). Copies of the data
can be obtained free of charge on application to CCDC, 12 Union
Road, Cambridge CB2 1EZ, UK [Fax: (internat.) ϩ 44-1223/336-
033; E-mail: deposit@ccdc.cam.ac.uk].
NMR Measurements: Spectra were recorded at 200, 300 (Gemini,
Varian) or 400 MHz (Mercury, Varian) for the H frequency. The
1
temperatures were calibrated by means of an Ni/Cu thermocouple
inserted in the probe before the measurements. Line-shape simula-
tions were performed by means of a PC version of the DNMR 6
program.[14] In the spectra obtained in the presence of a chiral solv-
ating agent (CSA), the shift separation (∆ν) is temperature-depend-
ent. By means of an empirical equation this separation was related
to the temperature in the range where the rotation rate was un-
doubtedly negligible. The ∆ν values were subsequently extrapolated
into the temperature range where the exchange rate occurs, al-
lowing one to obtain reliable line-shape simulations. In order to
Acknowledgments
We are grateful to I.Co.C.E.A., CNR, Bologna for access to the
observe sufficiently separated 1H NMR signals for the enantiomers, 400-MHz NMR spectrometer and to Dr. E. Dimartino, University
a very large molar excess (up to 300:1) of the Pirkle’s alcohol[6]
used as CSA was needed. Thus in these experiments, the concentra-
tion of the substrate had to be quite low (typically 10Ϫ3 ) in order
to have a concentration of CSA lower than the saturation point.
of Bologna, for helping with the X-ray diffraction determinations.
Financial support was obtained from MURST (National Project
‘‘Stereoselection in Organic Synthesis’’) and from the University of
Bologna (Progetto triennale di Ateneo 2000Ϫ2002).
Eur. J. Org. Chem. 2000, 3439Ϫ3446
3445

R. Leardini, L. Lunazzi, A. Mazzanti, H. McNab, D. Nanni
FULL PAPER
[6]
[7]
[8]
(R)-l-1-(9-Anthryl)-2,2,2-trifluoroethanol as in: W. H. Pirkle, J.
Am. Chem. Soc. 1966, 88, 1837.
[1]
S. Grilli, L. Lunazzi, A. Mazzanti, J.Org.Chem. 2000, 65, 3563.
[2]
The MMX force-field, as implemented in the Program PC
V. Enchev, G. Ivanova, A. Ugrinov, G. D. Neykov, J. Molec.
Struct. 1999, 608, 149 and references quoted therein.
[8a] D. Casarini, E. Foresti, F. Gasparrini, L. Lunazzi, D. Mac-
Model, Serena Software, Bloomington IN, was used.
[3]
[3a] A. R. Katritzky, J. Jiang, J. V. Greenhil, P. J. Steel, J. Chem.
Soc., Perkin Trans 1 1992, 3055. Ϫ [3b] R. E. Banks, W. J. Jondi,
ciantelli, D. Misiti, C. Villani, J. Org. Chem. 1993, 58, 5674.
[8b]
R. G. Pritchard, A. E. Tipping, Acta Crystallogr., Sect. C:
Ϫ
F. Gasparrini, L. Lunazzi, A. Mazzanti, M. Pierini, M.
[3c]
Cryst. Struct. Commun. 1995, 51, 291. Ϫ
J. D. McCullogh
Pietrusiewicz, C. Villani, J. Am. Chem. Soc. 2000, 121, 4773.
Slightly different proportions (10aЈ ϭ 54% and 10a" ϭ 46%)
were observed in CDCl3.
Jr., I. C. Paul, D. Y. Curtin, J. Am. Chem. Soc 1972, 94, 883.[3d]
K. Folting, W. N. Lipscomb, B. Jerslev, Acta Crystallogr. 1964,
17, 1263 and references quoted therein.
[9]
[10]
[11]
A. I. Vogel, Practical Organic Chemistry, 4th ed., Longmans,
London, 1969, p. 345.
G. Capozzi, G. Melloni, G. Modena, J. Chem. Soc., Perkin
Trans. 1 1973, 2250.
[4]
D. R. Boyd, S. Al-Showiman, W. B. Jennings, J. Org. Chem.
1978, 43, 3335. D. R. Boyd, W. B. Jennings, L. C. Waring, J.
Org. Chem. 1986, 51, 992.
[12]
[13]
[14]
[5] [5a]
B. Staskun, J. Org. Chem. 1968, 33, 3031.
S. S. Jenkins, J. Am. Chem. Soc. 1933, 55, 703.
QCPE program no. 633, Indiana University, Bloomington, IN.
Received March 21, 2000
C. G. McCarty, in The chemistry of the carbon-nitrogen
double bond (Ed.: S. Patai), Interscience Publishers, London,
[5b]
New York, 1970 ch. 9, p. 389. Ϫ
W. B. Jennings, V. E. Wil-
son, in Acyclic Organonitrogen Stereodynamics (Eds.: J. B. Lam-
bert, Y. Takeuchi), WCH, New York, 1992, ch 6, p. 193.
[O00146]
3446
Eur. J. Org. Chem. 2000, 3439Ϫ3446