2-(p-Tolylsulfinyl)benzyl halides as efficient precursors of optically pure trans-2,3-Disubstituted aziridines

Article abstract of DOI:10.1002/chem.201000217

Aziridination of (R)-N-sulfinylaldimines (aryl, heteroaryl and alkenyl derivatives) with 2-(p-tolyl sulfinyl)benzyl iodide in the presence of sodium hexamethyl disilazide takes place with almost complete control of the stereoselectivity (facial and antil

Full text of DOI:10.1002/chem.201000217

DOI: 10.1002/chem.201000217
2-(p-Tolylsulfinyl)benzyl Halides as Efficient Precursors of Optically Pure
trans-2,3-Disubstituted Aziridines
Yolanda Arroyo,*[a] ꢀngela Meana,[a] M. Ascensiꢁn Sanz-Tejedor,*[a] Inꢂs Alonso,[b] and
Josꢂ Luis Garcꢃa Ruano*[b]
Dedicated to Professor Josꢀ Barluenga on the occasion of his 70th birthday
Abstract: Aziridination of (R)-N-sulf-
inylaldimines (aryl, heteroaryl and al-
kenyl derivatives) with 2-(p-tolyl sulf-
inyl)benzyl iodide in the presence of
sodium hexamethyl disilazide takes
place with almost complete control of
the stereoselectivity (facial and anti/
syn) and with very high yields affording
optically pure N-sulfinyl trans-2,3-di-
removal of both C- and N-p-tolylsulfin-
yl groups with tBuLi provides the cor-
responding trans-NH aziridines 8 with-
out affecting their optical purity. Some
experimental results suggest these pro-
cesses evolve through benzyl halocar-
benoids as intermediates, whereas the-
oretical calculations support the forma-
tion of benzyl carbanions. These results
have served for revising the mechanis-
tic aspects of the reactions of substitut-
ed 2-p-tolylsulfinyl benzylcarbanions
with electrophiles.
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
Keywords: asymmetric synthesis ·
aziridination · chiral aziridines
diastereoselectivity · sulfinylimines
·
ACHTUNGTRENNUNGsubstituted aziridines 7. Simultaneous
Introduction
Haloalkyl-,[2] a-halocyclopropyl,[3] a-haloalkylidene[4] and a-
haloallyllithium[5] carbenoids have been used as versatile re-
agents in organic synthesis and extensively analyzed by
spectroscopic methods and theoretical calculations.[1b,6] In
contrast, (a-halobenzyl)lithium carbenoids have received
much less attention due to their extremely strong tendency
to give eliminative dimerization.[1] The electrophilic charac-
ter of a-bromobenzylcarbenoids was first detected by Closs
and Moss[7] in the preparation of arylcyclopropanes from
ArCHBr2, alkyllithiums and alkenes. Later on, the reaction
of (benzylchloromethyl)lithium with alkenes was reported.[8]
With respect to the reactions with electrophiles other than
benzyl halides the only example in the literature is the tri-
methylsilylation of (benzylchloromethyl)lithium and (ben-
zylbromomethyl)lithium.[9] More recently, Florio et al. re-
ported that (3-pyridylchloromethyl)lithium behaved like a
benzylic a-chloroorganolithium and reacts with carbonyl
compounds and Schiffꢀs bases.[10]
a-Haloorganolithium (or organosodium) compounds, also
called halocarbenoids, belong to the synthetically most
useful reactive intermediates in organic synthesis because
they have both nucleophilic and electrophilic reactivities.[1]
Additionally, the tetravalent nature of carbenoids may
confer a high level of stereoselectivity to their reactions,
which is difficult to attain by using divalent free carbenes. a-
[a] Prof. Dr. Y. Arroyo, Dr. ꢁ. Meana, Prof. Dr. M. A. Sanz-Tejedor
Organic Chemistry Department
Escuela de Ingenierꢂas Industriales
Universidad de Valladolid
Paseo del Cauce, 59, 47011 Valladolid (Spain)
Fax : (+34)983423310
E-mail: atejedor@eis.uva.es
We have recently reported the almost completely stereo-
selective transfer of chiral benzyl groups to different electro-
philes by starting from 2-(p-tolylsulfinyl)benzyl carbanions
bearing alkyl,[11] trimethylsilyloxy,[12] and methylthio[13] sub-
stituents at the benzylic carbon atom. In all these reactions,
the stereochemical results were explained by assuming the
formation of chelated species, with the lithium intramolecu-
larly stabilized by the sulfinyl oxygen atom. According to
yarroyo@eis.uva.es
[b] Prof. Dr. I. Alonso, Prof. Dr. J. L. Garcꢂa Ruano
Organic Chemistry Department (C-I)
Universidad Autꢃnoma de Madrid
Cantoblanco, 28049 Madrid (Spain)
Fax : (+34)914973966
E-mail: joseluis.garcia.ruano@uam.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000217.
9874
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9874 – 9883

FULL PAPER
theoretical calculations, these carbanion–Li+ complexes
adopt a quasi-boat conformation with the metal restricting
the access of the electrophile to one of the faces of the nu-
cleophile,[13c,14] which justifies the complete stereoselectivity
observed in the transfer of the prochiral benzylcarbanions,
mainly controlled by a remote sulfinyl group. At this point,
we reasoned that if the ortho sulfinyl group was able to sta-
bilize the (a-halobenzyl)lithium carbenoids it would reduce
their tendency to the eliminative dimerization and highly
stereoselective reactions with electrophiles could be carried
out. As potential precursors of the stabilized carbenoids we
chose the 2-p-tolylsulfinylbenzyl halides (Scheme 1), and N-
synthesis of the haloderivatives (S)-2[17] and (S)-3[18] by reac-
tion of bromine or iodine with the tin derivative (S)-1,
which in its turn was obtained from enantiomerically en-
riched (S)-2-(p-tolylsulfinyl)toluene according to the proce-
dure previously reported by us[19] (Scheme 2).
Scheme 2. Synthesis of a-halo-2-(p-tolylsulfinyl)toluenes (S)-2 and (S)-3.
Optically pure N-p-tolyl and N-tert-butylsulfinylimines
(4a–m) were prepared from the corresponding aldehydes
according to the procedures reported by Davis[20] and
Ellman, respectively.[21] To find the optimal conditions for
the aziridination procedure, the influence of different bases
and halobenzyl derivatives as well as the addition sequence
were investigated (Scheme 3 and Table 1). The experiences
carried out by adding bromobenzyl derivative (S)-2
(1 equiv) to strong bases, such as lithium diisopropylamide
(LDA) or lithium hexamethyl disilazide (LHMDS;
1.5 equiv), and soon after the imine (R)-4a, were found to
Scheme 1. Possible intermediates for the reaction of 2-p-tolylsulfinylben-
zyl halides and N-sulfinylACTHNUTRGNEUGNimines to give aziridines.
sulfinylimines were selected as electrophiles for obtaining
aziridines, prevalent structures in biologically active mole-
cules[15] and versatile building blocks for selective organic
transformations by ring-opening processes.[16] Here we de-
scribe the results obtained in this study which have provided
one of the most efficient methods for obtaining trans-aziri-
dines in a highly stereoselective manner (Scheme 1). The
stereochemical results obtained in these reactions are differ-
ent to those observed from other ortho-sulfinylbenzyl carb-
ACHTUNGTRENNUNGanions previously studied, which suggests a different struc-
ture for the involved intermediates (maybe carbenoids).
However, theoretical calculations concerning the structure
of the possible intermediates
Scheme 3. Homocoupling reaction from (S)-2.
support the carbanionic struc-
Table 1. Influence of the base, the halogen and the substituent on the nitrogen atom on the aziridination pro-
cess.
tures as the most stable ones,
putting in question the exis-
tence of carbenoid intermedi-
ates. Finally, a good explanation
of the experimental results is
provided by the theoretical
study of the transition states
corresponding to the different
approaches of the reagents.
Entry
X
Base
Imine
Yield [%][a]
d.r.[b]
de [%][b] trans-7a/cis-6a
[c]
1
2
3
4
5
6
7
8
Br
Br
Br
Br
Br
Br
I
phosphacene
LDA
(R)-4a
(R)-4a
(R)-4a
(R)-4a
(S)-4a
4’a
–
–
–
40[d]
90
43:57
75:25
80:20
50:50
–
>98:>98
>98:>98
>98:>98
5:5
Results and Discussion
LHMDS
NaHMDS
LHMDS
LHMDS
LHMDS
NaHMDS
87
28[e]
To apply the new strategy, pre-
sumably involving carbenoid
structures stabilized by the sul-
finyl oxygen atom, the use of
the 2-p-tolylsulfinyl benzyl bro-
mide or iodide as precursors
[f]
–
–
(R)-4a
(R)-4a
90
89:11
96:4
>98:>98
>98:>98
I
89[g]
[a] Combined isolated yield for both cis-6a and trans-7a isomers. [b] Determined by 1H NMR spectroscopy
from the crude reaction. [c] Starting materials were recovered unaltered. [d] 50% of (S)-2 was recovered unal-
tered. [e] 65% of (S)-2 was recovered unaltered. [f] Complex mixture. [g] Isolated yield for the trans-7a
are required. We performed the isomer.
Chem. Eur. J. 2010, 16, 9874 – 9883
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9875

Y. Arroyo, M. A. Sanz-Tejedor, J. L. Garcꢂa Ruano et al.
be inefficient for preparing aziridines because, under these
conditions, alkene 5 is almost quantitatively formed. This
behaviour is compatible with the conversion of (S)-2 into
the carbenoid 2a, which subsequently would suffer a quick
homocoupling reaction followed by 1,2-elimination of
LiBr[22] to afford 5 (Scheme 3). Nevertheless, the formation
of 5 from a carbene intermediate cannot be excluded.
When 2a was generated from (S)-2 in the presence of the
imine (R)-4a (Barbier conditions),[23] the homocoupling re-
action is prevented. As summarized in Table 1, under these
conditions, a mixture of two aziridines, cis-6a and trans-7a,
were observed in all cases except when phosphacene was
used as a base. (Table 1, entry 1). LDA leads to an almost
equimolecular mixture of two diastereisomers, cis-6a and
trans-7a, in rather modest yield but with complete facial se-
lectivity for both diastereoisomers (entry 2). A dramatic im-
provement of the yield was observed when using LHMDS
as a base (90%, entry 3), which
tions carried out by starting from ioderivative (S)-3 and (R)-
4a produced a higher trans selectivity than those with (S)-2
(Table 1, cf. entries 7 and 8 with 3 and 4, respectively). Once
again, the best base was NaHMDS, which yielded a 96:4
mixture of trans-7a and cis-6a, with the isolation of optically
pure trans-7a in 89% yield achieved after chromatography
(entry 8). The trans/cis ratio dropped to 89:11 when we used
LHMDS as a base (entry 7). These results proved that the
diastereoselectivity is clearly dependent on the metal (Li or
Na) and the halogen atom (Br or I), thus allowing the possi-
bility of excluding the carbenes as intermediates in these re-
actions.
Once the optimal reaction conditions for the aziridination
process had been established, we studied the behaviour of
different (R)-N-sulfinylimines (4a–m) with (S)-3 to deter-
mine the scope of the reaction (Table 2). In all the experien-
ces, aziridination reactions were completed within one
also enhanced the trans/cis rela-
Table 2. Stereoselective aziridination of N-sulfinylimines (R)-4a–m with a-iodo-2-(p-tolylsulfinyl)toluene (S)-
3.
tionship to 75:25 maintaining
the excellent facial selectivity
(diastereomeric excess (de)
>98%). We achieved a further
increase in trans selectivity by
employing lithium hexamethyl
disilazide (NaHMDS), which
yielded an 80:20 mixture of
trans-7a and cis-6a without
modifying the yield and the
facial selectivity (entry 4).
At this point, we investigated
the influence of a change of the
configuration of the imine. The
reaction of (S)-2 with (S)-4a in
the presence of LHMDS yield-
ed an almost equimolecular
mixture of the four possible
aziridines in low yield (Table 1,
entry 5), which contrasts with
the good results obtained from
(R)-4a (entry 3). It proves that
the matched pair for this
double asymmetric induction
process is formed by the re-
agents with a different configu-
ration at the sulfur atom (see
later). Otherwise, we have also
checked the reaction of (S)-2
with the sulfonamide 4’a, which
afforded a complex mixture of
compounds the complete analy-
sis of which was not possible in
our hands (entry 6).
Entry
1
(R)-Imine/R1
4a/p-Tol
R2
Yield [%][a] trans-7
trans-7/cis-6[b]
de [%][b] trans-7/cis-6
89
96:4
>98
>98
33
–
–
2
3
4
5
4b/p-Tol
4c/p-Tol
4d/p-Tol
4e/p-Tol
88
61[c]
83
>98:2
65:35
87:13
97:3
>98
>98
–
>98
>98
94
6
7
4 f/p-Tol
4g/p-Tol
88
90
96:4
>98
>98
–
–
>98:2
8
9
4h/p-Tol
4i/p-Tol
90
82
>98:2
>98
>98
–
–
97:3
10
11
12
13
4j/p-Tol
4k/p-Tol
4l/tBu
59
67
72
79
72:28
80:20
85:15
90:10
>98
>98
>98
>98
42
10
>98
>98
4m/tBu
Finally, the influence of the
halogen atom at the benzylic
position on the course of the re-
1
[a] Determined by H NMR spectroscopy from the crude reaction. [b] Yield of isolated product. [c] Calculated
from the ratios (2S,3S)-7/trans-(2R,3R)-7 or cis-(2R,3R)-6/cis-(2R,3S)-6, measured by 1H NMR spectroscopy.
action was also studied. Reac- The second isomer is not detected in the crude reactions for those cases for which the de is higher than 98%.
9876
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9874 – 9883

Optically Pure trans-2,3-Disubstituted Aziridines
FULL PAPER
minute, at À788C, and aziridines trans-7 were obtained as
major or exclusive compounds in high yields and with total
facial selectivity over a range of substrates. The only excep-
tion was the reaction starting from the imine 4c, containing
a strong electron-withdrawing group, which was produced
with a low diastereomeric trans/cis ratio and with 33% de
for the trans isomer (Table 2, entry 3). The reactions with
arylACHTUNGTRENNUNGimines 4a, 4b, and 4e (containing neutral, electron-do-
nating or bulky groups) and heteroarylimine 4 f evolved
with an almost complete trans selectivity (trans/cis ratio
ranges from 96:4 to >98:2) in excellent isolated yields (88–
94%) (entries 1, 2, 5 and 6). For arylimine 4d, bearing the
bulkier 2-bromophenyl group, the trans/cis ratio slightly de-
creased to 87:13, although a high isolated yield in the opti-
cally pure trans-7d could be achieved (entry 4). The aziridi-
nation procedure was also found to be successful for alkeny-
laldimines 4g–i, furnishing the corresponding trans-aziri-
dines 7g–i as the exclusive (trans/cis >98:2; entries 7 and 8)
or major (trans/cis 97:3; entry 9) diastereoisomers. The reac-
tions with N-p-toluensulfinyl alkylimines 4j and 4k evolved
in moderate trans selectivity and with low de for the cis iso-
mers (entries 10 and 11). Fortunately, the use of the N-tert-
Scheme 4. C- and N-desulfinylation of aziridines trans-7.
column (98.5% enantiomeric excess (ee)), which indicates
that reaction conditions do not produce significant epimeri-
sation at the starting material.
To get a deeper understanding of the stereochemical re-
sults described herein it is necessary to compare them with
those observed in previously studied reactions starting from
a-methylsulfenyl- and a-dimethyl sulfonium benzyl carban-
ions, Li-(S)-9 and Li-(S)-10, respectively (Scheme 5). In the
first case, the matched pair is formed by the reagents with
the same configuration at sulfur, (S)-9 and (S)-4, and the ob-
tained b-sulfanylamines[13a] were subsequently transformed
into (2R,3R)-aziridines,[27] by following the synthetic se-
quence showed in Scheme 5, route a. These results could be
explained by assuming the reaction proceeded through the
chelated species A, with the lithium intramolecularly stabi-
lized by the sulfinyl oxygen atom, and with the sulfinyl
oxygen atom at the sulfinylimine adopting its most stable
(s)-cis conformation around the C=N bond (both hypothesis
were supported by theoretical calculations).[13c] Accordingly,
TS-1 must be the presumably most stable transition state,
because it would result from the transoid approach of both
reagents in which electronic repulsion between N and S
atoms is minimized, to their respectively less-hindered faces.
The second strategy previously used for synthesizing aziri-
dines involves the reaction of the ylide derived from (S)-10
with (R)-4[28] (Scheme 5, route b). In this case, the matched
pair is formed by reagents of the opposite configuration at
their sulfinyl centres and (2S,3S)-aziridines are obtained as
the major products. The stereochemical model so far pro-
posed to explain these results is based on the assumption
that the ylide reacts through the planar free carbanion B,
stabilized by the electron-withdrawing [SMe2]+ group and
displaying the sulfinyl oxygen atom far from the carbanionic
carbon atom. According to theoretical calculations,[28] TS-2
is now the most stable transition state because it would
result from the cisoid approach of both reagents (strongly
favoured by electrostatic attractions between N and S+) to
their respectively less-hindered faces.
butylsulfACHTUNGTRENNUNG
inylimines[24] 4l and 4m improved the trans selectiv-
ity and a complete facial selectivity was achieved for both
isomers (entries 12 and 13). These results allow us to con-
clude that reactions of (S)-3 with N-sulfinylaldimines pro-
vide one of the best so far reported methods to obtain opti-
cally pure trans-2,3-disubstituted aziridines.
The cis or trans stereochemistry assigned to aziridines 6
and 7, respectively, was deduced from the vicinal coupling
3
constants of the protons joined to C-2 and C-3. J2,3 values
for the cis isomers ranged between 6.8 and 7.8 Hz (dihedral
angle ꢀ08), whereas those for the trans isomers are much
lower (3.6–4.4 Hz, dihedral angle ꢀ1208). The absolute con-
figurations for compounds cis-6a and trans-7a were un-
equivocally established by X-ray diffraction studies. They al-
lowed us to assign the S configuration to both ring carbon
atoms at trans-7a and 2R,3S for cis-6a[25] (see the scheme in
Table 2). The similar behaviour observed in all the aziridina-
tion reactions summarized in Table 2, all of them evolving
with complete facial selectivity and excellent trans/cis diaste-
reoselectivity, suggests that the absolute configuration for
trans-aziridines 4b–m is also 2S,3S.
Deprotection of the obtained disulfinylated compounds is
required to provide NH-aziridines, which are valuable inter-
mediates to a range of synthetically useful compounds.[26] At
this point, it is important to note that there is not any one-
step protocol able to synthesize NH-aziridines and most of
the methods for the asymmetric synthesis of aziridines pro-
vide N-protected aziridines for which the substituents at the
nitrogen atom are usually difficult to remove. In our case, si-
multaneous C- and N-desulfinylation of the disulfinylated
aziridines 7 can be performed by reaction with 2.2 equiva-
lents of tBuLi in good yields. We have checked the efficien-
cy of this procedure in the case of compounds 7a, 7b, 7e
and 7h (Scheme 4). The optical purity of the NH-aziridine
8h was determined by HPLC with a Chiralcel OD-H
The results obtained in this paper (Scheme 5, route c)
starting from iododerivative (S)-3 are very similar to those
observed from sulfonium ylides, because the matched pair is
formed by reagents with different configurations at their
sulfur atoms and reactions afford almost exclusively (2S,3S)-
7 aziridines. However, it is not easy to assume a similar
mechanism for both transformations because the halogen
would not be able to stabilize a free carbanion similar to
Chem. Eur. J. 2010, 16, 9874 – 9883
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9877

Y. Arroyo, M. A. Sanz-Tejedor, J. L. Garcꢂa Ruano et al.
Scheme 5. Conversion of Li-(S)-9, Li-(S)-10, (S)-3 into the corresponding (2R,3R) or (2S,3S)-aziridines via TSs TS-1, TS-2 and TS-3.
that of B (due to electronic repulsion of the lone electron
pairs at the halogen and carbanionic carbon atoms) nor the
cisoid approach of the reagents yielding a transition state
similar to TS-2 (presumably unstabilized by the dipolar re-
pulsion of the C-I and C=N bonds). Thus, the formation of
intermediates different to both A and B species, maybe car-
benoids, emerged as one possible explanation of the ob-
served stereochemical results.
To understand the nature of the intermediates involved in
these reactions, the possible structures of the species result-
ing in the reaction of (S)-3 with LHMDS were studied theo-
retically at the DFT (B3LYP)[29] level by using the Gaussi-
an 03 program.[30,31] The most stable structures found for
model carbanion–Li+ complexes and free carbanions are
shown in Scheme 6. Dimethyl ether and dimethylamine
were used as simplified models for solvent and base, respec-
tively, and have been included as ligands for the lithium
atom. The tolyl group has also been simplified as a phenyl
one. Two stable structures have been found for carbanion–
Li+ complexes. Chelated species I, with the sulfinyl oxygen
and carbanionic carbon atoms coordinated to the lithium
atom, is more stable than II, for which the sulfinyl oxygen
atom has been shifted by a molecule of ether (Scheme 6).[32]
Structure I is identical to chelated species A depicted in
Scheme 5, and considered as the most reactive intermediate
in the case of benzylcarbanions joined to SMe, OMe or
alkyl groups. The lower stability of II could be a conse-
quence of the higher association ability of the sulfinyl
oxygen atom to the metal. Considering the question about
the “carbenoid character” of II, there is only a slight elonga-
Scheme 6. Molecular structures, representative distances of the hydrogen
bonds [ꢅ], and energies [kcalmolÀ1] of possible carbanionic species. The
first value indicates the relative energy, with ZPE correction included, of
À
I, II, IV (+LiACHTUNGTERN(NNUG NHMe2)CAHTUNGTNERNG(U OMe2)3 OMe2) and V (+LiACHTNUGETRNGN(UN NHMe2)ACHTUNGTRENNUNG(OMe2)2I)
with respect to III, which is considered as the most stable. Free-energy
correction is indicated in brackets, first in vacuo and second in THF
(mimicked by IEFPCM).
seems to be preferred probably due to the high delocaliza-
tion of the charge into the benzyl system. The search of
other structures exhibiting a clear carbenoid character has
been unsuccessful.
Another two structures have been considered in which
the carbanionic carbon atom is not directly joined to the
metal. The first one, III, is a free carbanion stabilized by a
hydrogen bond with the dimethylamine ligand. This type of
complex would probably be the first formed after the depro-
tonation step and, therefore, it could be considered as a pre-
cursor of the species I, in which the carbanion acts as a
tion[1b] of the C I bond (2.24 ꢅ) relative to the nonlithiated
À
species (2.22 ꢅ), which suggests that the carbanion character
9878
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9874 – 9883

Optically Pure trans-2,3-Disubstituted Aziridines
FULL PAPER
ligand of the metal, shifting one
molecule of Me2O. The compar-
ison of the calculated energy
values for I and III indicates
that the former is slightly more
stable in a vacuum, but the
latter one is the most stable (by
3 kcalmolÀ1) when considering
the effect of the solvent. We
have also studied the free carb-
anion structure IV, for which
the Li+ is not stabilizing the
carbanion, because
a similar
structure was invoked to satis-
factorily explain the selectivity
observed in the reaction of sul-
finylimines (R)-4 with the dime-
thylsulfonium
analogue.[28]
However in the present case,
free carbanion IV is considera-
bly less stable than the most
favoured structures (I or III),
even when solvent effects are
considered
(11.7 kcalmolÀ1),
thus contrasting with the small
difference existing (2.8 kcal
molÀ1) for similar species in the
case of the dimethylsulfonium
analogue.[28] Finally, we have
also studied the energy corre-
sponding to a carbene structure
V, but it is also much less stable
than all the Li+-stabilized com-
plexes previously considered.
Looking at the two most
stable structures, I and III, the
metal with its ligands are ori-
ented toward the si face of the
benzylic carbanion, thus impos-
Scheme 7. Molecular structures, representative distances [ꢅ] and energies [kcalmolÀ1] of possible transition
states between model anion III and model (E)-N-phenylsulfinylimine derived from acetaldehyde that would
yield trans-7 (TSa) and cis-6 (TSb). The first value indicates the relative energy, with ZPE correction included.
Free-energy correction is indicated in brackets
ing serious steric interactions in the approach of the electro-
philes to that face. However, the experimental results indi-
cate that the major trans-aziridines 7 result in the approach
of the electrophile to that face. This fact suggests a complex
situation also involving another type of interaction that is
different from the steric ones, which would be only under-
standable by studying the relative stability of the transition
states leading to the four possible diastereoisomers, that is,
re–re, re–si, si–re and si–si approaches of the carbanion and
imine reagents. We have been unsuccessful in the search of
the transitions states starting from I, but some of them
could be localized by starting from III. Scheme 7 shows the
TSs found in the reaction of the complex III with the (E)-N-
phenylsulfinylimine derived from acetaldehyde[33] (as the
simplest imine).[34]
with sulfinyl imine coordinated to the metal were not found.
The three TSs found (see Scheme 7) seem to derive from
the initial formation of a hydrogen bond between the sulfi-
nyl oxygen atom at sulfinamide and the aminic hydrogen
atom that was bonded to the carbanionic centre in III. After
À
this interchange of acceptor in the hydrogen bond of the N
H (not unexpected on the basis of the high ability of the sul-
finyl oxygen atom to form this type of bond) the approach
of the imine to the upper face of the carbanion (si–re ap-
proach, Scheme 7) produces TSa that will evolve into trans-
(2S,3S)-7. In this transition state, there are no significant
steric restrictions, the sulfinyl oxygen atom in the imine
adopts its most stable (s)-cis conformation and the groups
À
around the incipient C C bond show a slightly distorted
gauche conformation (N-C-C-I=À68.38). When the attack
of the imine bonded to the aminic hydrogen atom takes
place to the bottom face of the carbanion, two transition
states, TSb and TSc (resulting from the re–re and re–si ap-
In all these TSs the presence of stabilizing interactions
(hydrogen bonds) between the model imine and the ligands
of the metal is essential. TSs without these interactions or
Chem. Eur. J. 2010, 16, 9874 – 9883
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9879

Y. Arroyo, M. A. Sanz-Tejedor, J. L. Garcꢂa Ruano et al.
proaches, respectively, Scheme 7), were found. In TSb,
which evolves into the cis-(2R,3S)-6, the imine also adopts
the (s)-cis conformation and the groups around the newly
trans selectivity and good isolated yields, make this reaction
potentially useful in organic synthesis. Theoretical calcula-
tions support the carbanionic structure of the intermediates
despite some experimental results that suggested that carbe-
noid species could be involved. It has been possible to find
the transition states providing aziridines trans-(2S,3S)-7, ob-
tained as the major or exclusive isomers in these reactions,
À
formed C C bond show an almost antiperiplanar arrange-
ment (N-C-C-I=+161.38), which will probably be favoured
with respect to the gauche one by the low dipolar repulsion
À
À
of the C N and C I bonds. Additionally, it is stabilized by a
hydrogen bond between the iminic nitrogen atom and one
of the protons of the aryl group (d=2.53 ꢅ). These two fac-
tors would be able to explain the small difference in energy
between TSa and TSb calculated for N-phenyl methyliden-
and aziridines cis-ACTHNUTRGNEUNG(2R,3S)-6, obtained as the minor ones in
some reactions. The structural differences between these
transition states allow us to rationalize their energetic differ-
ences, which in their turn justify the observed stereoselectiv-
ity in these reactions.
ACHTUNGTRENNUNGimine, despite the twisted arrangement adopted by the re-
agents to reach TSb. The low trans/cis selectivity experimen-
tally observed for alkyl imines (72:28 ratio; Table 2,
entry 10) agrees with the similar energy calculated for TSa
and TSb. When the PhSO group at the imines is substituted
by the tBuSO one, it is expected that the stability of TSa
would be scarcely affected, whereas that of TSb would de-
crease on steric grounds (tBu/Ph interaction would affect
Experimental Section
General methods: Dry solvents and liquid reagents were distilled under
argon just prior to use. THF was distilled from Na/benzophenone ketyl,
CH2Cl2 and diisopropylamine were dried over P2O5 and KOH, respec-
tively. NaHMDS (1.0m solution in THF), nBuLi and tBuLi (2.5 and 1.5m
solution in hexanes, respectively) were purchased from commercial sup-
pliers and used as received. All reactions were carried out in flame- or
oven-dried glassware under an inert Ar atmosphere. Reactions were
monitored by TLC on commercially available precoated plates (Silica
Gel 60 F254). Flash column chromatography was performed by using by
Silica Gel 60 (230–400 mesh) and a SCX column. NMR spectra were ob-
tained in CDCl3 solutions at 300 and 75 MHz for 1H and 13C NMR spec-
troscopy, respectively, (chemical shifts (d) are given in ppm), and all cou-
pling constants, J, are given in Hz. Diastereoisomeric ratios were deter-
mined from the integration of well-separated signals in the 1H NMR
spectra of the crude reaction mixtures (H-C2 and H-C3 for cis/trans com-
pounds). Mass spectra were measured by electron impact (EI, 70 eV) or
FAB. Melting points were measured in open capillary tubes and they are
uncorrected. Characterization data of compounds 6a, 6c–f, 6i–m, 7b-f
and 7j and NMR spectra of compounds 6a, 6c–f, 6i–m, 7g–i, 7k–m, 8b,
8e and 8h as well X-ray structure and refinement details of compounds
6a and 7a can be found in the Supporting Information.
À
the hydrogen bond N···H Ar, see Scheme 7), thus explain-
ing the higher selectivity observed for tert-butylsulfinyl-
ACHTUNGTRENNUNG
À
À
À
À
(N Ciminic Me) and the carbanion (CBn Cipso Cortho) are dif-
ferent for TSa (17.78) and TSb (44.88). These differences
could explain the complete facial selectivity observed in re-
actions of the arylimines (Table 2), by assuming that TSa’
(Scheme 7) is stabilized by p-stacking interactions (these in-
teractions also explained the results found in reactions with
N-aryl derivatives of arylimines[4]), whereas it would not
possible for TSb because the relative arrangement of both
aryl groups is not appropriated. Finally, the re–si approach
of the imine to the bottom face of the carbanion generates
TSc, precursor of the trans-(2R,3R)-aziridines. This lack of
the stabilizing hydrogen bond of the iminic nitrogen with
the aromatic hydrogen atom in the transition state deter-
mines that the sulfinyl oxygen atom at the imine does not
adopt the most stable (S)-cis arrangement and the groups
General procedure for the reactions summarized in Table 2: NaHMDS
(1.2 equiv) at À788C was added dropwise to a stirred solution of (S)-a-
iodo-2-(p-tolylsulfinyl)toluene (0.15 mmol, 1.0 equiv) and N-sulfinylimine
((R)-4a–m, 0.15 mmol, 1.0 equiv) in THF (3 mL). When the reaction was
completed (one minute), the mixture was hydrolyzed at that temperature
with saturated aqueous NH4Cl solution (1 mL) and extracted with
CH2Cl2 (3ꢆ5 mL,). The combined organic layers were dried over MgSO4,
and the solvent was removed under reduced pressure. The residue was
purified by flash column chromatography.
À
around the incipient C C bond show a slightly distorted
gauche conformation (N-C-C-I=+68.08). All these facts
could be responsible for the lower stability of TSc with re-
spect to TSa or TSb (Scheme 7) determining the absence of
the (2R,3R)-aziridines in the reaction mixtures. Finally, we
were unable to find the transition state corresponding to the
si–si approach of the imine to the upper face of the carban-
ion, affording the cis-(2S,3R)-aziridines. These compounds
were only experimentally observed in the case of alkyl
imines (Table 2, entries 10 and 11).
trans-(2S,3 S)-2-Phenyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(p-tolu-
ene sulfinyl)]aziridine (7a): Eluent for chromatography: hexane/Et2O
1:4; yield: 89%; white solid; m.p. 118–1198C (hexane/Et2O); [a]D20
=
À145.4 (c=0.9 in CHCl3) (lit.[28] =À145.5 (c=0.6 in CHCl3)); spectro-
scopic data of compound 7a are coincident with those previously report-
ed: 1H NMR (300 MHz, CDCl3): d=8.04 (dd, 1H, J=7.8, 1.1 Hz), 7.56,
7.41, 7.24, 7.10 (two AA’BB’ systems, 8H), 7.51 (dd, 1H, J=7.8, 1.1 Hz),
7.38–7.28 (m, 7H), 4.35, 3.75 (2d, 2H, J=4.3 Hz), 2.37, 2.29 ppm (2s,
6H).
trans-(2S,3S)-2-Propenyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(p-
toluene sulfinyl)]aziridine (7g): Eluent for chromatography: hexane/Et2O
Conclusion
1
1:3; yield: 90%; yellow syrup; [a]2D0 =À236.0 (c=0.3 in CHCl3); H NMR
(300 MHz, CDCl3): d=8.00 (dd, 1H, J=7.9, 1.2 Hz), 7.59, 7.42, 7.23, 7.17
(two AA’BB’ systems, 8H), 7.44–7.42 (m, 1H), 7.27–7.20 (m, 1H), 6.90
(d, 1H, J=7.7 Hz), 4.67–4.63 (m, 2H), 4.12 (d, 2H, J=3.8 Hz), 2.66 (dd,
1H, J=3.8, 1.3 Hz), 2.36, 2.34 (2s, 6H), 1.77 ppm (s, 3H); 13C NMR
(75 MHz, CDCl3): d=143.5, 142.4, 141.8, 141.7, 141.5, 134.4, 134.2, 130.9,
129.9, 129.5, 128.4, 126.3, 125.6, 124.6, 124.5, 123.8, 52.8, 33.4, 21.4,
2-(p-Tolylsulfinyl)benzyl iodide (S)-3 is an efficient precur-
sor of enantiomerically pure trans-2-substituted-3-phenyl
aziridines, with aromatic, heteroaromatic, a,b-unsaturated
and alkyl N-sulfinylimines acting as eletrophiles at these
aziridination processes. The excellent facial selectivity, high
9880
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9874 – 9883

Optically Pure trans-2,3-Disubstituted Aziridines
FULL PAPER
1
18.1 ppm; HRMS: m/z: calcd for C18H18NOS: 296.1109 [M+ÀSOTol];
808C; [a]2D0 =+251.2 (c=0.2 in CHCl3); H NMR (300 MHz, CDCl3): d=
found: 296.1112.
7.38–7.27 (m, 5H), 7.21, 6.89 (AA’BB’ system, 4H), 3.82 (s, 3H), 3.07
(brs, 2H), 1.50 ppm (brs, 1H); 13C NMR (75 MHz, CDCl3): d=158.9,
139.7, 131.7, 128.6, 127.2, 126.6, 125.4, 114.0, 55.3, 43.3 ppm; HRMS: m/z:
calcd for C15H15NO: 225.1154; found: 225.1145.
trans-(2S,3S)-2-Cinnamyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(p-
toluene sulfinyl)]aziridine (7h): Eluent for chromatography: hexane/Et2O
1:3; yield: 90%; yellow solid; m.p. 68–698C (hexane/Et2O); [a]D20
=
1
(À)-(2S,3S)-2-(2-Naphthyl)-3-phenylaziridine (8e): Eluent for chroma-
À570.0 (c=0.1 in CHCl3); H NMR (300 MHz, CDCl3): d=8.06 (dd, 1H,
J=7.0, 0.8 Hz), 7.63, 7.38, 7.25, 7.01 (two AA’BB’ systems, 8H), 7.48–
7.24 (m, 7H), 6.32 (dd, 1H, J=15.7, 8.8 Hz), 6.31 (d, 1H, J=8.8 Hz),
4.18 (d, 2H, J=3.7 Hz), 2.75 (dd, 1H, J=8.8, 3.7, 1.3 Hz), 2.37, 2.07 ppm
(2s, 6H); 13C NMR (75 MHz, CDCl3): d=143.0, 142.3, 142.2, 141.8,
141.1, 135.7, 133.9, 136.8, 130.9, 130.1, 129.6, 128.6, 128.4, 126.7, 126.5,
125.4, 124.6, 123.8, 122.3, 53.4, 34.0, 21.4, 21.1 ppm; HRMS: m/z: calcd
for C23H20NOS: 358.1266 [M+ÀSOTol]; found: 358.1259.
tography hexane/Et2O 1:1; yield: 75%; white solid; m.p. 74–758C; [a]D20
=
1
À160.3 (c=0.3 in CHCl3); H NMR (300 MHz, CDCl3): d=7.87–7.78 (m,
4H), 7.52, 7.27 (m, 8H), 3.30, 3.20 (2s, 2H), 1.61 ppm (brs, 1H);
13C NMR (75 MHz, CDCl3): d=139.6, 137.0, 133.3, 132.8, 128.7, 128.4,
127.7, 127.6, 127.4, 126.3, 125.7 125.5, 124.4, 123.4 ppm (C-2 and C-3 are
missing); HRMS: m/z: calcd for C18H15N: 245.1205; found: 245.1193;
HPLC (98.5 ee, Daicel Chiralcel OD-H, hexane/IPA 90:10, 0.7 mLminÀ1
254 nm, 258C): tR =39 min (2S,3S).
,
trans-(2S,3S)-2-(1-Methylcinnamyl)-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-
[(R)-(p-toluenesulfinyl)]aziridine (7i): Eluent for chromatography:
hexane/Et2O 1:3; yield: 82%; yellow syrup (compound 7i is particularly
unstable in solution); 1H NMR (300 MHz, CDCl3): d=8.05 (dd, 1H, J=
7.9, 1.2 Hz), 7.61, 7.53, 7.30, 7.25 (two AA’BB’ systems), 7.52–7.40 (m,
3H), 7.35–6.90 (m, 5H), 6.53 (s, 1H), 4.04, 3.67 (2d, 2H, J=4.5 Hz),
2.39, 2.34 (2s, 6H), 1.83 ppm (s, 3H).
(À)-(2S,3S)-2-Cinnamyl-3-phenylaziridine (8h): Eluent for chromatogra-
phy hexane/Et2O 1:2.5; yield: 70%; white syrup; [a]2D0 =À104.7 (c=0.6 in
1
CHCl3); H NMR (300 MHz, CDCl3): d=7.40–7.20 (m, 10H, 6.70 (d, 1H,
J=15.8 Hz), 5.92 (dd, 1H, J=15.8, 7.7 Hz), 3.08 (s, 1H), 2.71 (d, 1H, J=
7.7 Hz), 1.60 ppm (brs, 1H); 13C NMR (75 MHz, CDCl3): d=139.3,
136.5, 131.8, 129.5, 128.6, 128.5, 127.6, 127.2, 126.1, 125.6 ppm (C-2 and
C-3 are missing); HRMS: m/z: calcd for C16H15N: 221.1205; found:
221.1199.
trans-(2S,3S)-2-Isopropyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(p-
toluene sulfinyl)]aziridine (7k): Eluent for chromatography: hexane/Et2O
1:2; yield: 67%; white syrup; [a]2D0 =À200.2 (c=1.0 in CHCl3); 1H NMR
(300 MHz, CDCl3): d=7.89 (dd, 1H, J=7.7, 0.8 Hz), 7.54, 7.47, 7.24, 7.19
(two AA’BB’ systems, 8H), 7.42, 7.32 (2dt, 2H, J=7.6, 1.0 Hz), 7.06 (d,
1H, J=7.5 Hz), 3.82 (d, 1H, J=3.8 Hz), 2.36, 2.34 (2s, 6H), 2.61 (dd,
1H, J=8.8, 3.8 Hz), 2.17–1.90 (m, 1H), 1.08, 0.74 ppm (2d, 6H, J=
6.6 Hz); 13C NMR (75 MHz, CDCl3): d=144.0, 142.6, 141.7, 141.6, 141.4,
134.9, 131.1, 130.0, 129.5, 129.4, 128.7, 126.1, 125.1, 124.7, 49.4, 30.3, 28.3,
21.3, 21.1, 20.4 ppm; HRMS: m/z: calcd for C18H20NO2S2: 346.0935 [M+
ÀSOTol]; found: 346.0942.
trans-(2S,3S)-2-Butyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(tert-
butane sulfinyl)]aziridine (7l): Eluent for chromatography: hexane/Et2O
1:2.5; yield: 72%; colourless oil; [a]2D0 =À182.5 (c=0.2 in CHCl3);
1H NMR (300 MHz, CDCl3): d=8.11 (dd, 1H, J=7.7, 1.5 Hz), 7.55–7.42
(m, 2H), 7.49, 7.23 (AA’BB’ system, 4H), 7.15 (d, 1H, J=7.5 Hz), 3.66
(d, 1H, J=3.9 Hz), 2.35 (s, 3H), 2.21–2.19 (m, 1H), 1.74–0.87 (3 m, 6H),
1.13 (s, 9H), 0.81 ppm (t, 3H, J=7.0 Hz); 13C NMR (75 MHz, CDCl3):
d=143.3, 142.1, 142.0, 135.7, 131.1, 130.1, 128.4, 126.6, 125.4, 124.1, 57.3,
49.6, 32.1, 29.8, 27.5, 22.4, 22.6, 21.4, 13.8 ppm; HRMS: m/z: calcd for
C19H22NO2S2: 360.1092 [M+ÀtBu]; found: 360.1093.
Acknowledgements
We thank the DGYCT (CTQ-2009-12168-BQU), Comunidad Autꢃnoma
de Madrid (S2009/PPQ-1634) and JCYL (VA085A08) for financial sup-
port and Dr. J.M. ꢁlvarez for X-ray elucidation. A generous allocation
of computer time at the Centro de Computaciꢃn Cientꢂfica de la UAM is
also acknowledged.
[1] For reviews, see: a) M. Braun, Angew. Chem. 1998, 110, 444–465;
Angew. Chem. Int. Ed. 1998, 37, 430–451; b) G. Boche, C. W. Loh-
renz, Chem. Rev. 2001, 101, 697–756; c) A. Basu, S. Thayumanavan,
Angew. Chem. 2002, 114, 740–763; Angew. Chem. Int. Ed. 2002, 41,
716–738; d) M. Braun in The Chemistry of Organolithium Com-
pounds (Eds.: Z. Rappoport, I. Marek), Wiley, New York, 2004,
Chapter 13, pp. 829–898.
[2] For leading and recent references, see: a) H. C. Stiasny, R. W. Hoff-
mann, Chem. Eur. J. 1995, 1, 619- 624; b) R. W. Hoffmann, H.-C.
Stiasny, J. Krꢇger, Tetrahedron 1996, 52, 7421–7434; c) M. Shimizu,
T. Hata, T. Hiyama, Heterocycles 2000, 52, 707–717; d) M. Shimizu,
T. Fujimoto, X. Y. Liu, H. Minezaki, T. Hata, T. Hiyama, Tetrahe-
dron 2003, 59, 9811–9823; e) P. R. Blakemore, S. P. Marsden, H. D.
Vater, Org. Lett. 2006, 8, 773–776; f) B. A. Pearlman, S. R. Putt,
J. A. Fleming, J. Org. Chem. 2006, 71, 5646–5677.
[3] For a review, see: A. De Meijere, M. Vonseebach, S. Zollner, S. I.
Kozhushkov, V. N. Belov, R. Boese, T. Haumann, J. Benetbuchholz,
D. S. Yufit, J. A. K. Howard, Chem. Eur. J. 2001, 7, 4021–4034, and
references therein.
[4] For recent references, see: a) J. Kvꢂcala, A. Pelter, Collect. Czech.
Chem. Commun. 2001, 66, 1508–1520; b) H. Yanagisawa, K. Miura,
M. Kitamura, K. Narasaka, K. Ando, Helv. Chim. Acta 2002, 85,
3130–3135; c) H. Yanagisawa, K. Miura, M. Kitamura, K. Narasaka,
K. Ando, Bull. Soc. Chim. 2003, 76, 2009–2026; d) Y. H. Li, L. Lu,
X. M. Zhao, Org. Lett. 2004, 6, 4467–4470; e) M. Braun, A. Hoh-
mann, J. Rahematpura, C. Buhne, S. Grimme, Chem. Eur. J. 2004,
10, 4584–4593.
[5] For recent references, see: a) G. J. Gordon, T. Luker, M. W. Tuckett,
R. J. Whitby, Tetrahedron 2000, 56, 2113–2129; b) S. Dixon, S. M.
Fillery, A. Kasatkin, D. T. E. Norton, R. J. Whitby, Tetrahedron
2004, 60, 1401–1416.
trans-(2S,3S)-2-Isopropyl-3-[(S)-2-(p-toluenesulfinyl)phenyl]-1-[(R)-(tert-
butane sulfinyl)]aziridine (7m): Eluent for chromatography: hexane/Et2O
1:2.5; yield: 79%; white syrup; [a]2D0 =À198.8 (c=1.0 in CHCl3);
1H NMR (300 MHz, CDCl3): d=8.05 (d, 1H, J=7.3 Hz), 7.54–7.44 (m,
2H), 7.49, 7.23 (AA’BB’ system, 4H), 7.12 (d, 1H, J=7.0 Hz), 3.74 (d,
1H, J=3.6 Hz), 2.34 (s, 3H), 2.17 (dd, 1H, J=9.7, 3.7 Hz), 1.90–1.83 (m,
1H), 1.14, 0.46 (2d, 6H, J=6.5 Hz), 1.10 ppm (s, 9H); 13C NMR
(75 MHz, CDCl3): d=143.4, 142.0, 141.8, 136.3, 131.3, 130.2, 128.5, 126.6,
125.7, 124.8, 57.5, 56.0, 31.4, 27.7, 21.3, 21.1, 20.6 ppm; HRMS: m/z: calcd
for C18H20NO2S2: 346.0935 [M+ÀtBu]; found: 346.0942.
Representative procedure for total desulfinylation of N-sulfinyl aziri-
dines: tBuLi (0.18 mL, 0.27 mmol, 1.5m in hexane, 2.2 equiv) was added
to a stirred solution of 7a, 7b, 7e and 7h (0.12 mmol) in THF (2 mL).
When the reaction was completed (15 min), the mixture was hydrolyzed
with saturated aqueous NH4Cl (1 mL) and extracted with CH2Cl2 (3ꢆ
3 mL). The combined organic layers were dried over MgSO4 and the sol-
vent was removed under reduced pressure. The residue was purified by
flash column chromatography.
(À)-(2S,3S)-2,3-Diphenylaziridine (8a): Eluent for SCX column chroma-
tography NH3/methanol (7m); yield: 75%; colourless syrup; [a]D20
=
À320.0 (c=0.5 in CHCl3) (lit.[35] =À328.8 (c=1.25 in CHCl3)); spectro-
scopic data of compound 8a are coincident with those previously report-
ed: 1H NMR (300 MHz, CDCl3): d=7.40–7.29 (m, 10H), 3.12 (s, 2H),
1.70 ppm (brs, 1H).
[6] For recent references, see: a) H. Hermann, J. C. W. Lohrenz, A.
Kuhn, G. Boche, Tetrahedron 2000, 56, 4109–4115; b) J. Kvicala, J.
Stambasky, S. Bohm, O. Paleta, J. Fluorine Chem. 2002, 113, 147–
(+)-(2S,3S)-2-(p-Methoxyphenyl)-3-phenylaziridine (8b): Eluent for
chromatography hexane/Et2O 1:3; yield: 67%; white solid; m.p. 79–
Chem. Eur. J. 2010, 16, 9874 – 9883
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9881

Y. Arroyo, M. A. Sanz-Tejedor, J. L. Garcꢂa Ruano et al.
154; c) L. M. Pratt, B. Ramachandran, J. D. Xidos, C. J. Cramer,
D. G. Truhlar, J. Org. Chem. 2002, 67, 7607–7612; d) L. M. Pratt, N.
Van Nguyen, L. T. Le, J. Org. Chem. 2005, 70, 2294–2298; e) L. M.
Pratt, L. T. Le, T. N. Truong, J. Org. Chem. 2005, 70, 8298–8302;
f) F. M. Bickelhaupt, H. L. Hermann, G. Boche, Angew. Chem. 2006,
118, 838–841; Angew. Chem. Int. Ed. 2006, 45, 823–826; g) K.
Ando, J. Org. Chem. 2006, 71, 1837–1850.
Luisi, M. T. Rochetti, J. Org. Chem. 2002, 67, 759–763; see also ref-
erence [1d].
[23] M. Julia, J. N. Verpaux, T. Zahneisen, Synlett 1990, 769–770.
[24] It is noteworthy that reactions of N-tert-butylsulfinylimines derived
from aromatic aldehydes with (S)-3 afforded complex mixtures of
compounds the complete analysis of which was not possible in our
hands.
[7] G. L. Closs, R. A. Moss, J. Am. Chem. Soc. 1964, 86, 4042–4053.
[8] R. A. Olofson, C. M. Dougherty, J. Am. Chem. Soc. 1973, 95, 581–
582.
[9] H. Andringa, Y. A. Heus-Kloos, L. Brandsma, J. Organomet. Chem.
1987, 336, C41-C43.
[10] a) S. Florio, L. Troisi, J. Org. Chem. 1996, 61, 4148–4150; b) V. Cap-
riati, S. Florio, R. Luisi, Synlett 2005, 1359–1369; c) V. Capriati, S.
Florio, F. M. Perna, Chem. Eur. J. 2009, 15, 7958–7979.
[25] CCDC-719589 contains the supplementary crystallographic data for
this compound. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. See also the Supporting Information.
[26] Recent reviews: a) Z. Zhijun, I. D. G. Watson, A. K. Yudin, J. Am.
Chem. Soc. 2005, 127, 17516–17529; b) I. D. G. Watson, A. K.
Yudin, Acc. Chem. Res. 2006, 39, 194–206. Others references, see:
c) L. Gentilucci, Y. Grijzen, L. Thijs, B. Zwanenburg, Tetrahedron
Lett. 1995, 36, 4665–4668; d) F. A. Davis, Y. Zhang, A. Rao, Z.
Zhang, Tetrahedron 2001, 57, 6345–6352; e) F. A. Davis, J. Deng, Y.
Zhang, R. C. Haltiwanger, Tetrahedron 2002, 58, 7135–7143; f) F. A.
Davis, Y. Wu, H. Yan, W. McCoull, K. R. Prasad, J. Org. Chem.
2003, 68, 2410–2419.
[27] Y. Arroyo, A. Meana, J. F. Rodrꢂguez, M. Santos, M. A. Sanz-Teje-
dor, J. L. Garcꢂa Ruano, Tetrahedron 2006, 62, 8525–8532.
[28] Y. Arroyo, A. Meana, J. F. Rodrꢂguez, M. A. Sanz-Tejedor, I.
Alonso, J. L. Garcꢂa Ruano, J. Org. Chem. 2009, 74, 4217–4224.
[29] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789;
b) A. D. Becke, J. Chem. Phys. 1993, 98, 1372–1377.
[30] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[11] a) J. L. Garcꢂa Ruano, M. C. CarreÇo, M. A. Toledo, J. M. Aguirre,
M. T. Aranda, J. Fischer, Angew. Chem. 2000, 112, 2848–2849;
Angew. Chem. Int. Ed. 2000, 39, 2736–2737; b) J. L. Garcꢂa Ruano,
J. Alemꢈn, J. F. Soriano, Org. Lett. 2003, 5, 677–680; c) J. L. Gar-
cꢂa Ruano, M. T. Aranda, J. M. Aguirre, Tetrahedron 2004, 60, 5383–
5392; d) J. L. Garcꢂa Ruano, J. Alemꢈn, J. A. Parra, J. Am. Chem.
Soc. 2005, 127, 13048–13054; e) J. L. Garcꢂa Ruano, J. Aleman, B.
Cid, Synthesis 2006, 687–686; f) J. L. Garcia Ruano, V. Marcos, J.
Alemꢈn, Angew. Chem. 2008, 120, 6942–6945; Angew. Chem. Int.
Ed. 2008, 47, 6836–6839; g) J. L. Garcꢂa Ruano, J. Alemꢈn, S.
Catalꢈn, V. Marcos, S. Monteagudo, A. Parra, C. Pozo, S. Fustero,
Angew. Chem. 2008, 120, 8059–8062; Angew. Chem. Int. Ed. 2008,
47, 7941–7944; h) J. L. Garcꢂa Ruano, A. Parra, V. Marcos, C. del
Pozo, S. Catalꢈn, S. Monteagudo, S. Fustero, J. Am. Chem. Soc.
2009, 131, 9432–9441; i) J. L. Garcꢂa Ruano, V. Marcos, J. Alemꢈn,
Synthesis 2009, 3339–3349.
[12] J. L. Garcꢂa Ruano, J. Alemꢈn, Org. Lett. 2003, 5, 4513–4516.
[13] a) Y. Arroyo, A. Meana, J. F. Rodrꢂguez, M. Santos, M. A. Sanz-Te-
jedor, J. L. Garcꢂa Ruano, J. Org. Chem. 2005, 70, 3914–3920; b) Y.
Arroyo, A. Meana, J. F. Rodrꢂguez, M. Santos, M. A. Sanz-Tejedor,
J. L. Garcꢂa Ruano, J. Org. Chem. 2007, 72, 1035–1038; c) Y.
Arroyo, A. Meana, J. F. Rodrꢂguez, M. A. Sanz-Tejedor, I. Alonso,
J. L. Garcꢂa Ruano, J. Org. Chem. 2009, 74, 764–772.
[14] J. L. Garcꢂa Ruano J. Alemꢈn, I. Alonso, A. Parra, V. Marcos, J.
Aguirre, Chem. Eur. J. 2007, 13, 6179–6195.
[15] a) S. R. Rajski, R. M. Williams, Chem. Rev. 1998, 98, 2723–2796;
b) T. J. Hodgkinson, T. J. M. Shipman, Tetrahedron 2001, 57, 4467–
4488; c) S. Fꢇrmeier, J. O. Metzger, Eur. J. Org. Chem. 2003, 649–
659.
[16] For reviews on this subject, see: a) D. Tanner, Angew. Chem. 1994,
106, 625–646; Angew. Chem. Int. Ed. Engl. 1994, 33, 599–619;
b) H. M. I. Osborn, J. Sweeney, Tetrahedron: Asymmetry 1997, 8,
1693–1715; c) A-H. Li, L.-X. Dai, V. K. Aggarwal, Chem. Rev. 1997,
97, 2341–2372; d) T. Ibuka, Chem. Soc. Rev. 1998, 27, 145–154;
e) W. McCoull, F. A. Davis, Synthesis 2000, 1347–1365; f) J. B. Swee-
ney, Chem. Soc. Rev. 2002, 31, 247–258; g) P. Mꢇller, C. Fruit,
Chem. Rev. 2003, 103, 2905–2919; h) X. E. Hu, Tetrahedron 2004,
60, 2701–2743; i) X. L. Hou, J. Wu, R. H. Fan, C. H. Ding, Z. B.
Luo, X. L. Dai, Synlett 2006, 181–193.
[17] J. L. Garcꢂa Ruano, J. Alemꢈn, A. Parra, A. Alcudia, C. Maya, J.
Org. Chem. 2009, 74, 2145–2152.
[18] Y. Arroyo, A. Meana, M. A. Sanz-Tejedor, J. L. Garcia Ruano, Org.
Lett. 2008, 10, 2151–2154.
[19] J. L. Garcꢂa Ruano, J. Alemꢈn, A. Padwa, Org. Lett. 2004, 6, 1757–
1760.
[20] F. K. Davis, R. E. Reddy, J. M. Szewczyk, G. V. Reddy, P. S. Portono-
vo, H. Zhang, D. Fanelli, R. T. Reddy, P. Zhou, P. J. Carrol, J. Org.
Chem. 1997, 62, 2555–2563.
[31] Geometries have been fully optimized by using the standard 6-
31G(d) basis set for all the atoms except I: a) R. Ditchfield, W. J.
Hehre, J. A. Pople, J. Chem. Phys. 1971, 54, 724–728; b) M. M.
Francl, W. J. Petro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J.
DeFrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654–3665; the
LANL2DZ basis set supplemented with an ad function of exponent
0.289 was used for iodine; c) P. J. Hay, W. R. Wadt, J. Chem. Phys.
1985, 82, 299–310; d) A. Hçllwarth, M. Bçhme, S. Dapprich, A. W.
Ehlers, A. Gobbi, V. Jonas, K. F. Kçhler, R. Stegmann, A. Veld-
kamp, G. Frenking, Chem. Phys. Lett. 1993, 208, 237–240; frequen-
cies and zero-point energy (ZPE) were also computed at the same
level of theory. Final energies have been obtained by using the more
extended 6-311+G** basis set for all atoms except iodine: e) A. D.
McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639–5648; f) R.
Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980,
72, 650–654; For I, a basis set of similar quality (SDB-aug-cc-
pVTZ) was used: g) J. M. L. Martin, A. Sundermann, J. Chem. Phys.
2001, 114, 3408–3420; relative free energies (in kcalmolÀ1) were
evaluated at the B3LYP/6-311+G** level with ZPE and entropy
corrections evaluated at 298 K by using the frequencies previously
calculated at the B3LYP/6-31G(d) level. Because of the importance
of solvent effects, especially when charged species are involved,
[21] J. A. Ellman, D. A. Cogan, T. P. Tang, G. Liu, T. D. Owens, J. Org.
Chem. 1999, 64, 1278–1284.
[22] The eliminative dimerization of lithium carbenoids to give alkenes
has been extensively studied. In these processes, the 1,2-elimination
of LiBr is very fast, see: a) B. J. Wakefield in Organolithium Meth-
ods, Academic Press, London, 1988; b) V. Capriati, S. Florio, R.
model structures I–V have also been optimized in
a dielectric
medium mimicking THF, by using the IEF-PCM model: h) E.
Cancꢉs, B. Mennucci, J. J. Tomasi, J. Chem. Phys. 1997, 107, 3032–
9882
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9874 – 9883

Optically Pure trans-2,3-Disubstituted Aziridines
FULL PAPER
3041; i) M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys.
Lett. 1998, 286, 253–260; in the case of model transition states, this
optimization including solvent effects was not possible.
[34] Due to the complexity expected for the TSs formed from complex
III, a preliminary study by starting from different relative orienta-
tions of model carbanion IV (Scheme 6) and the (E)-N-phenylsulfi-
nylimine derived from acetaldehyde was carried out (see the Sup-
porting Information for the corresponding structures of transition
states TS’a–d). The difference in energy between TS’a and TS’b
(2.0 kcalmolÀ1) predicts a trans/cis ratio of 97:3, much higher than
the 72:28 ratio experimentally observed with alkyl imines (Table 2,
entry 10).
[32] Complexes of type I and II, varying the N-Li-C-I dihedral angle,
and type III, varying the C-S-O-Li dihedral angle, turned out to be
less stable (see the Supporting Information for complexes Ib, IIb
and IIIb).
[33] This imine was used as a model in our previous study of the reaction
of dimethylsulfonium derivatives (see reference [28]). The rotamer
with the sulfinyl oxygen atom in a (s)-cis arrangement with respect
to the C=N bond was the most stable. However, solvent effects and
stabilizing interactions within the transition states can dramatically
[35] B. B. Lohray, J. R. Ahujas, J. Chem. Soc. Chem. Commun. 1991, 95–
97.
Received: January 26, 2010
À
modify the conformation around the N S bond.
Published online: April 12, 2010
Chem. Eur. J. 2010, 16, 9874 – 9883
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9883