Synthesis, separation and structure determination of atropisomeric ketimines: precursors of new potential chiral ligands and phase-transfer catalysts

Article abstract of DOI:10.1016/j.tetasy.2008.06.024

Sterically crowded chiral biaryl ketimines have been prepared via nucleophilic addition onto a substituted benzonitrile or 8-cyanoquinoline and subsequent treatment of the resulting methylenimines either with methyl iodide, or with a chiral primary amine. Atropisomerism due to the hindered rotation of the aryl groups about the C-C bond adjacent to C{double bond, long}N has been evidenced in all cases. Physical separation and full characterisation (NMR and X-ray analysis) of atropisomeric imines have been accomplished.

Full text of DOI:10.1016/j.tetasy.2008.06.024

Tetrahedron: Asymmetry 19 (2008) 1523–1531
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Synthesis, separation and structure determination of atropisomeric ketimines:
precursors of new potential chiral ligands and phase-transfer catalysts
b
Abdelaziz Retmane a, Said Gmouh a, Marc Runghen b, Jean-Yves Valnot b, , Jacques Maddaluno ,
*
Loïc Toupet c, Hassan Oulyadi b, Jamal Jamal Eddine a,
*
a Département de Chimie Université Hassan II Ain-Chock Faculté des Sciences, B.P.5366 Maârif Casablanca, Morocco
b UMR CNRS 6014 and FR CNRS 3038, IRCOF, Université and INSA de Rouen, 76821 Mont Saint Aignan Cedex, France
c UMR CNRS 6626 and FR CNRS 2108, Université de Rennes 1, campus de Beaulieu, 35042 Rennes Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Sterically crowded chiral biaryl ketimines have been prepared via nucleophilic addition onto a substi-
tuted benzonitrile or 8-cyanoquinoline and subsequent treatment of the resulting methylenimines either
with methyl iodide, or with a chiral primary amine. Atropisomerism due to the hindered rotation of the
aryl groups about the C–C bond adjacent to C@N has been evidenced in all cases. Physical separation and
full characterisation (NMR and X-ray analysis) of atropisomeric imines have been accomplished.
Ó 2008 Published by Elsevier Ltd.
Received 27 March 2008
Revised 11 June 2008
Accepted 20 June 2008
Available online 12 July 2008
1. Introduction
tions. The alkylation of this substrate was chosen as a model, and it
has become a reference, which gives an indication about the per-
Asymmetric phase-transfer catalysis (APTC) has become an
attractive concept as it allows us to perform organic reactions un-
der mild and safe conditions using inexpensive reagents. A large
number of APTC reactions employing chiral catalysts are described
in the literature1 prominently featuring a structurally rigidified
cinchona alkaloid system.2 Such a performance is somewhat
intriguing and has triggered the interest of several research groups
either to understand the chiral induction mechanism or to design
and synthesise sterically and electronically tunable catalysts.
Among the non-alkaloid chiral catalysts reported, chiral C2-sym-
metric binaphthyl ammonium and C2-symmetric guanidinium
salts have been found to be of great value.3 In general, enantio-
selective catalysis requires catalysts possessing multipoint interac-
tion with the substrate in the transition state for good chiral
induction. A survey of the literature reveals that this evidence
applies to APTC and that, in most cases, the reaction temperature
remains a controlling factor for stereoselectivity. Over the course
of these interesting contributions, we became interested in a
rational design of new chiral phase-transfer agents and ligands.
We have reported earlier the sterically hindered and remark-
ably stable ketiminium salt 1 (Fig. 1) as being a versatile phase-
transfer catalyst.4 The catalytic properties of 1 were investigated
in the C-alkylation of a glycine-derived imine in non-chiral condi-
formance of newly designed chiral phase-transfer catalysts. Imini-
um salt 1 displayed comparable catalytic properties with respect to
previously described phase-transfer catalysts such as onium salts,5
crown ethers6 or metal complexes.7 The asymmetric version of the
catalytic reaction required the design and synthesis of the required
hindered chiral iminium salts.8 In this context, some possibilities
for the introduction of chirality have been envisaged depending
on (1) the position of the chiral vector on the imine framework
(2 or 3 and 4, Fig. 1) and (2) the potential chirality due to restricted
rotation.9
2. Results and discussion
The nucleophilic addition of 1-naphthyllithium onto (R)-2-(sec-
butoxy)-benzonitrile 5 and subsequent quenching of the lithio-
imine intermediate with methyl iodide (Scheme 1) gave the chiral
N-methyl imine 6, a precursor of iminium salt 2. The quenching
conditions have been found to determine the number of observed
stereoisomers of imine 6.
Thus, when N-methylation was conducted at low temperature
(ꢀ50 °C kinetic conditions), 13C NMR of a deuteriochloroform solu-
tion of the crude product showed two sets of imino 13C@N signals
at d (168.6, 169.7) and (176.9, 177.1). The chemical shift difference
of 10 ppm indicates the presence of a pair of both syn and anti geo-
metrical configurations.10 The N-methyl group displayed, four dif-
ferent signals in the 1H NMR 300 MHz spectra at d 3.38, 3.33, 3.07
and 2.79 ppm with a slight doubling of the two upfield signals (7%,
64%, 13% and 16%, respectively) and thus confirmed the presence of
* Corresponding authors. Tel.: +33 235522439; fax: +33 235522971 (J.Y.V.); tel.:
+212 65101701; fax: +212 22230674 (J.J.E.).
E-mail addresses: jean-yves.valnot@univ-rouen.fr (J.-Y. Valnot), e.j.eddine@
fsac.ac.ma (J. J. Eddine).
0957-4166/$ - see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2008.06.024

1524
A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
N
N+
I-
OH
I-
I-
N+
N+
N
O
OH
O
4
1
2
3
Figure 1.
Li
CN
N
N
O
CH3I
CH3
Li
O
O
5
6
Scheme 1.
more than two diastereoisomers. Interestingly, when the deuterio-
chloroform solution was heated to 60 °C for 15 min and allowed to
cool to room temperature, the recorded 1H NMR spectrum showed
the complete disappearance of the N–CH3 signal at 3.38 ppm in
favour of the most upfield shifted one. On the other hand, the pro-
portions of the other individual signals remain unchanged. This
observation indicates, that the steric environment imposed by
the aromatic groups in 6 seems to reach the level of the required
hindrance that allows the ambient temperature detection of atrop-
isomeric structures for both syn and anti configurations. However,
it is not large enough to inhibit interconversion, at least for one
isomer.
The atropisomerism about the C@ N bond has previously been
reported by Boyd and Jennings in the kinetic and mechanism stud-
ies of the isomerisation of chiral and achiral N-alkyl imines derived
from aryl alkyl ketones.11 Further conformational investigation by
dynamic NMR and molecular mechanic calculations, described by
Lunazzi et al.,12 provided interesting and reliable information with
regard to the stereomutation pathway of such enantiomers.
Fortunately, methylation of the lithioimine intermediate at
20 °C provided in excellent yield the crude imine 6 whose 13C
NMR and 1H NMR spectra showed only the first pair of 13C@N sig-
nals (168.5, 169.7) and the two N–CH3 signals at 3.33 and 3.07 ppm
(ca. 8:2). It is worth emphasising that TLC (2:8 EtOAc/heptane) of
the crude mixture showed imine 6 as two separate spots with a
Furthermore, heating a deuteriochloroform solution of pure 6a,
which is stable as a solid, resulted in the progressive appearance
of isomer 6b until equilibrium was reached (6a:6b = 6:4) after
48 h (Fig. 2d). Isomerisation of the same solution at room tempera-
ture did not exceed 10% after one week. On the other hand, a syn-
structure could be found from the X-ray analysis of a crystal of pure
6a grown in diluted heptane/ether (3:1) (Fig. 3).13 Also noticeable
from this structure is the out of plane position of both aryl moieties,
(C10, C11, C13, C18 = ꢀ61.5) and (C1, C10, C11, C13 = ꢀ48.5), sug-
gesting that their rotation may imply atropisomerism.
Complementary structural information on the stereoisomeric
nature of components 6a and 6b in solution came from the high-
field NMR investigation. Therefore, classical high-field 500 and
600 MHz homo- and hetero- nuclear experiments were under-
taken in order to assign all the protons and sp2 carbons of the indi-
vidual isomers. Of particular importance was the downfield shift
(8.52 ppm) noticed for H8 (numbering according to the corre-
sponding carbon atom of the X-ray spectrum) of the naphthyl
group of isomer 6a. The same proton is shifted 0.7 ppm upfield
in isomer 6b (Fig. 2a and b). In the same way, the ortho-protons
of the phenoxy group H14 display signals centred at 7.15 ppm
for isomer 6a and 7.86 ppm for isomer 6b. This cross shift can be
related to the orientation of the two rings relative to each other.
The absence of a NOESY correlation between the aromatic protons
in both cases, is in line with the X-ray results and in accordance
with conformations in which the naphthyl and the phenyl groups
are twisted out of the imino plane. More importantly, the intense
NOESY correlation observed for both isomers 6a and 6b between
N–CH3 and the ortho-protons of the phenoxy group H14 clearly
indicates their spatial proximity and the orientation of the
N-methyl group cis to the chiral aryl moiety, suggesting stereo-
isomers with a syn-configuration for both 6a and 6b (Fig. 4).
Conversely, only a faint correlation between N–CH3 and the
naphthyl group of 6a was observed, suggesting that, in chloroform,
the alkyl chain is not affected by the naphthyl group proximity.
small
DRf. Careful chromatographic separation helped us to isolate
the most eluted isomer 6a which crystallised upon solvent re-
moval. On the other hand, isomer 6b could be recovered as a col-
ourless oil, which does not crystallise upon standing. 1H and 13C
NMR unambiguously revealed that both components have the
imine 6 skeletal and that isomer 6a showed N–CH3 and 13C@N at
d 3.33 and 168.5 ppm, respectively. The pure fractions of 6b exhib-
ited 13C@N and N–CH3signals at 169.7 and 3.07 ppm, respectively,
but the contribution from isomer 6a reached ca. 10% after 24 h of
standing in solution at ambient temperature (Fig. 2).

A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
1525
Figure 3. X-ray crystallographic structure of pure 6a.
Figure 2. (a) 1H NMR of pure 6a; (b) 1H NMR of 6b with progressive contribution of
6a; (c) 13C NMR of pure 6a; (d) 13C NMR of pure 6a after 48 h of heating.
Another point of concern is the strong correlation between H17
and H19 in both 6a and 6b. For isomer 6b, the NOESY experiment
allowed us to gain insight into the space orientation of the chiral
moiety since two clear sets of correlations are observed with the
aromatic protons (Fig. 5). We suggest that the downfield shift ob-
served for H8, as well as the upfield shift of the chiral auxiliary pro-
tons, are the results of the rotation of the naphthyl group (variation
of the dihedral angle C10, C11, C13, C18), which brings the alkyl
chain in a closer proximity to the naphthyl ring current. On the
other hand, the slight doubling observed for the N–CH3 group
can be attributed to the existence of two stable rotamers, on the
NMR timescale, which are different only as regards the orientation
of the alkyl side chain referring to the naphthyl group because of a
switching orientation around H19–H17.
The X-ray and NMR studies prompt us to conclude that compo-
nents 6a and 6b are conformers of the same geometric stereoiso-
mer syn. Therefore, interconversion noticed in ageing or heating
solutions can be attributed to conformational equilibrium related
to slow motion of the naphthyl about the chiral axis C10-C11,
and not to syn ? anti isomerisation (Fig. 6).
Figure 4. NOESY chart of a mixture of isomers 6a and 6b showing correlation
between NCH3–H(14) and H(17)–H(19).
Subtraction of 6a and 6b signals from the 1H and 13C NMR
spectra of the kinetically formed mixture allowed us to assign
the non-separated anti stereoisomers. It has to be mentioned
that the upfield N-methyl signal shows a slight doubling similar
to the one observed for the same group of isomer 6b. This pheno-
menon can also be attributed to the existence of stable isoenerge-
tique conformations in solution. Therefore, the fast isomerisation
upon heating of the kinetically formed crude mixture, together

1526
A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
helped to notice that the chiral auxiliary orientation, referring to
the naphthalene moiety, controls the pseudocycle formation
through favourable interaction with oxygen pro R doublet in both
cases (Fig. 7). Under kinetic conditions, the formation of anti iso-
mer could be the result of methylation of open species.
This first result confirms that isolation of stable rotamers of
ketimines is feasible when a sufficient hindrance is provided. In
an attempt to increase the stability, we came to the more hindered
ketimine 8, arguing that the quinoline nitrogen ring could contrib-
ute to favourable electronic interactions. The precursor of iminium
salt 3 was prepared in the same way by the addition of the chiral 1-
naphthyllithium derivative onto 8-cyanoquinoline and subsequent
quenching with methyl iodide (Scheme 2).
Interestingly, in this case, the quenching conditions have not
been found to control the nature and number of observed stereo-
isomers. After the reaction was complete, TLC (1:9, EtOAc/Hep-
tane) of the crude mixture showed two well-separated spots. 1H
NMR of each isolated isomer showed the N–CH3 signals at
d = 3.99 ppm for the most eluted 8a and d 3.86 ppm for the less
eluted isomer 8b. Alternatively, proton H8 of the naphthyl group
of 8a is shifted 1 ppm upfield compared to the same proton of
8b (7.09 vs 8.12 ppm). Again, careful NMR investigation revealed
that the methyl group faces the naphthyl group in both cases,
and by analogy with the preceding results, we postulate the pres-
ence of two rotamers of one single geometric isomer with a syn
configuration. Additionally, no trace of isomerisation between 8a
and 8b conformers was observed. However, while 8b is stable upon
heating, the 1H NMR of an ageing solution of 8a showed the com-
plete transformation of the N–CH3 signal, which split into three
sets of multiplets between 3.06 and 3.97 ppm.
Chiral ketimine 4, featuring the N-linked chiral moiety, was
prepared following an adapted transimination protocol15 between
2-(S)-aminobutanol and 1-naphthyl-1-(2-hydroxyphenyl)-methyl-
idenamine 9 (Scheme 3).
Imine 4 has been isolated as a crystallised mixture of two insep-
arable isomers (7:3). While stable in the crystallised form, isomeri-
sation in CDCl3 to a 1/1 mixture has t1/2 = 1 h at 27 °C. Highfield
NMR spectroscopy (1H, 13C, HMQC, COSY and NOESY) helped to dif-
ferentiate two sets of naphthyl protons and outlined an intense
NOESY effect between the naphthyl group and the hydroxyalkyl
chain protons (Fig. 8), which clearly evidenced the geometric
anti-configuration. The correlations observed for both isomers
Figure 5. NOESY chart expansion for pure 6b underlying correlations between the
alkyl chain and the naphthyl group protons.
with the unchanged individual proportions of 6a and 6b, evidenced
both the absence of syn ? anti isomeriztion and the presence of
anti atropisomers. These observations are in accordance with the
earlier reported conformational studies on hindered imines, which
concluded that barrier to interconversion of atropisomers is signifi-
cantly lower than the barrier to syn ? anti isomerisation.12
The observed stereoisomer distribution, depending on methyl
iodide-quenching conditions, is obviously related to the structure
of the intermediate lithioimine species in solution. Thus, under
thermodynamic conditions, exclusive formation of the syn
stereoisomers 6a and 6b implies alkylation cis to the phenoxy
group. This requires the nitrile face differentiation during the
nucleophilic addition conducting to the less-hindered lithioketi-
mines. A possible explanation of the resulting stereoselectivity is
the rigidification under thermodynamic conditions of the interme-
diate lithioimine by further intramolecular electronic interaction
between oxygen and lithium atoms. The study of molecular models
C9
C9
C1
9
1
C1
Projection along
C11-C10
10
O
C13
NCH3
11
13
18
N
C13
NCH3
14
Syn, aS , R
NCH3 > C13 > C9 > C1
6a
C1
C1
C9
9
1
C13
NCH3
10
O
11
C9
Syn, aR, R
13
C13
NCH3
N
6b
Figure 6. Interconversion 6a–6b by rotation of the naphthyl group and configuration assignment.14

A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
1527
H
H
Me
Me
Me
Me
O
O
I
I
Li
Li
N
N
N
OR*
6b and/ or 6a
6a
Figure 7. Possible lithioimine intermediate species in THF solution under thermodynamic conditions.
Br
N
O
H8
1- n-BuLi, THF, -78˚C
N
2- Cyanoquinoline
CH3
3- CH3I, -50˚C to RT
O
7
8
Scheme 2.
21
4
1
20
OH
9
11
H2N
OH
18
NH
N
19
13
12
14
OH
17
9
4
Scheme 3.
between H16 of the phenol moiety and the hydroxymethyl, and
the minor one between H13 and H18, are probably due to the spa-
tial proximity of the two hydroxy groups.
13 kcal/mol. Moreover, rotation of the phenol group for the same
geometric isomer showed a less stable conformer (ꢀ14.52 kcal/
mol) for which the hydroxy group is located close to the nitrogen.
Again, these studies allowed us to conclude that atropisomerism is
due to the naphthyl-group slow motion.
Finally, independent methylation of the separated isolated-con-
formers (6a, 6b) and (8a, 8b) in acetonitrile led to iminium salts 2
and 3, respectively. It has to be noticed that each pair led to two
iminium salts with a different specific rotation. On the other hand,
1H NMR spectra were identical except for the methyl groups which
displayed two pairs of slightly doubled singlets attributable to a
partial conformational interconversion during the quaternisation
process. The difference of specific rotations for the salts of each
pair indicates that this interconversion is slow. Atropisomeric sta-
bility therefore requires design of more sterically hindered chiral
iminium salts.
These NMR observations are consistent with the results of semi-
empirical AM1 calculation.16 First, the energy barrier computed for
the syn/anti interconversion via trigonal N-inversion has been
determined to be 20.7 kcal/mol. The sp2 N-inversion is the unique
pathway considered since tautomerisation and acid-catalysed
mechanisms do not apply for imine 4. Secondly, the energetic pro-
file associated with the naphthyl and phenol rotation about the
respective adjacent C–C bonds (variation of the dihedral angles
a
= [C1, C10, C11, C12] and a0 = [C13, C12, C11, C10], respectively)
for both syn- and anti-stereoisomers, showed the lowest minima
for the anti-imine owing to naphthyl rotation (ꢀ12.86 and
ꢀ14.79 kcal/mol) (Fig. 9). The interconversion process between
the two stable conformers requires a free energy of activation of

1528
A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
3. Conclusion
While only a 5% ee was observed with iminium salts, the in situ
generated chiral metal complex derivative of 4 revealed enantio-
selection in the -allylation with ees reaching 17%. Optimisation
a
of enantiomeric excess necessitated the incorporation of crowded
biaryls for hindered rotation and chiral auxiliaries, which differen-
tiate the iminium salts faces for better ion-pairing. This will cer-
tainly be beneficial for both atropisomeric homogeneity and the
steric environment around the metal coordination sphere or the
charged nitrogen atom.
4. Experimental
4.1. General
Uncorrected melting points were determined on a Büchi 510
instrument. Optical rotations were determined at 25 °C on Per-
kin–Elmer 341 polarimeter using the sodium D line and concentra-
tions are given in g/100 ml. IR spectra were recorded on Perkin–
Elmer 16 PC FT instrument (in cmꢀ1). 1H and 13C NMR spectra were
recorded on Bruker AC 300 spectrometer at 300 and 75 MHz,
respectively.1D and 2D NMR experiments were performed at 500
or 600 MHz on Bruker Avance DMX 500 or Bruker Avance DMX
600. Chemical shifts are in ppm referenced to TMS in CDCl3. Mass
spectra were determined on ATI Unicom automass spectrometer
and ThermoFinnigan SSQ 7000 GC/MS. Elemental analyses were
performed on Thermo Finnigan Elementary Analyser Flash EA
1112. Analytical thin layer chromatography (TLC) was performed
on Merck Kieselgel 60 F254 glass plates (0.25 mm) and compounds
were visualised by UV light (254 nm) or phosphomolybdic acid–
ethanol (22.1 g–180 ml) spray. Column chromatography was car-
ried out on Merck Kieselgel. Chiral HPLC was performed on Kon-
tron 420 and JASCO 875 equipped with Whelk-01 column,
eluent: hexane/iPrOH = 90:10, 1 ml/min, retention times: 11 (S):
5.37 min; 11Ò: 7.46 min.5b Chemicals were purchased from either
Sigma–Aldrich or Fluka and used as received. Solvents were dis-
tilled according to known procedures.
Figure 8. NOESY chart of isomers of imine
configuration and the presence of two rotamers.
4 evidencing the anti-geometric
4.2. (R)-2-sec-Butoxybenzonitrile 5
To a solution of 2-cyanophenol (1.54 g, 13 mmol), triphenyl-
phosphine (3.7 g, 14.14 mmol) and 2-(S)-sec-butanol (1 ml,
14.1 mmol) in anhydrous ether (60 ml) was slowly added with stir-
ring a solution of diisopropylazodicarboxylate (DIAD) (2.84 g,
14.1 mmol) in anhydrous ether (15 ml). The resulting suspension
was stirred at room temperature for 3 h, after which the white pre-
cipitate was filtered off and the solvent was removed in vacuo. The
residue was distilled under low pressure to give compound 5 as
Figure 9. Computer generated structure of the stable rotamers of anti imine 4.
The catalytic properties of the four isolated chiral iminium salts
and ligand 4 were thereafter investigated in the asymmetric allyla-
tion of benzophenone imine 10 under L/L and S/L phase-transfer
conditions, respectively (Scheme 4).
colourless oil (2.35 g, 80%) Eb. 160–163 °C/10 mmHg; [
a
] = ꢀ54 (c
1.2, CHCl3); 1H NMR (300 MHz, CDCl3) d (ppm): 0.87 (t, J = 6.5 Hz,
2 or 3, (10 mol %), PhMe
*
N
CO2tBu
N
CO2tBu
50% NaOH, Allyl bromide, RT
Yield 76%, ee 5%
11
10
Cu(OAc)2, 4 (10 mol %), PhMe, NaOHs (1.5 eq)
Allyl bromide, RT, Yield 75%, ee 17%
Scheme 4. Allylation of Schiff base 10 under phase-transfer conditions.

A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
1529
3H), 1.23 (d, J = 6.0 Hz, 3H), 1.55 (m, 2H), 4.3 (m, 1H), 6.69–7.45 (m,
4H); 13C NMR (75 MHz, CDCl3) d (ppm): 9.1, 18.6, 28.6, 76.2, 102.3,
113.3, 116.3, 120.0, 133.3, 133.9, 159.7; IR (neat): 1263, 1462,
1270, 1590.3, 2226.9, 2935.9, 2974.4; EIMS 175(M+), 146, 120,
91, 57.
13C NMR (75 MHz, CDCl3) d (ppm): 11.0 (C22), 18.5 (C20), 28.0
(C21), 41.9 (C12), 74.0 and 73.9 (C19), 112.5 and 113.0 (C17),
120.2 (C15), 124.5 (C1), 125.5 (C2), 126.6 (C6), 127.0 (C7), 127.8
(C8), 128.6 (C3), 128.7 (C5), 131.1 (C14), 132.0 (C13), 133.5 and
133.3 (C16), 137.7 and 137.9 (C10), 139.0 (C9), 156.5 and 156.5
(C18), 169.94 and 169.98 (C11); IR (neat): 1628; EIMS 317 (M+),
246, 217, 154, 127.
4.3. 1-Bromo-2-((R)-sec-butoxy)naphthalene 7
Following the above procedure using 1-bromo-2-naphthol in-
stead of 2-cyanophenol, compound 7 was isolated as a colourless
4.4.2. N-Methylation at ꢀ50 °C
The lithioimine was cooled to ꢀ50 °C and methyl iodide
(1.2 equiv) was added at once. After 2 h of stirring between
ꢀ50 °C and ꢀ30 °C TLC (2:8, EtOAc/heptane) showed two heavy
and less resolved spots both having roughly equal intensities. Fol-
lowing the above work-up, crude mixture of diastereoisomers of
imine 6 was isolated as brownish oil (95%). Data for the non-sepa-
rated anti stereoisomers: 1H NMR (300 MHz, CDCl3) d (ppm): 1.1
(br s, 3H), 1.3 (br s, 3H), 1.72 (m, 2H), 2.79 and 3.38 (2s, 3H), 3.9
(m, 1H), 6.59–7.95 (m, 9H), 8.01–8.28 (m, 2H); 13C NMR
(75 MHz, CDCl3) d (ppm): 12.6, 22.0, 30.3, 43.0, 71.5, 113.9,
115.7, 122.0, 125.2, 126.2, 126.4, 126.5, 127.1, 128.6, 129.4,
131.7, 132.5, 133.9, 134.1, 135.6, 155.1, 176.9, 177.1.
oil (94%); Eb. 140 °C/0.8 mmHg; [
a
] = ꢀ2.1 (c 0.16, CHCl3); 1H
NMR (300 MHz, CDCl3) d (ppm): 0.95 (t, J = 8.7 Hz, 3H), 1.26 (d,
J = 9.6 Hz, 3H), 1.60 (m, 2H) 4.32 (m, 1H), 7.11–8.39 (m, 6H); 13C
NMR (75 MHz, CDCl3) d (ppm): 10.2, 20.0, 29.6, 78.6, 117.7,
124.8, 126.8, 127.0, 128.3, 129.0, 134.0, 134.3, 156.5; IR (neat):
945, 1242, 1266, 1462, 1593, 1624, 2932, 2971, 3065; EIMS
279(M+), 255, 199, 172, 157, 145, 115, 57.
4.4. N-Methyl-1-(2-(R)-sec-butoxyphenyl)-1-(1-naphthyl)
methylenimine 6
To a solution of 1-bromonaphthalene (935 mg, 4.54 mmol) in
anhydrous ether (5 ml) was added n-BuLi (1.2 equiv) at ꢀ78 °C
under an N2 atmosphere. The temperature was allowed to rise to
ꢀ50 °C for a complete halogen–metal exchange before a solution
of 5 (1.02 g, 5.4 mmol) in anhydrous ether (1 ml) was added. The
reaction mixture was then left to warm up to ambient temperature
under stirring.
4.5. N-Methyl-1-(8-quinolinyl)-1-(2-(R)-sec-butoxynaphtyl)-
methylenimime 8
Following the above halogen–metal exchange procedure and N-
methylation at ꢀ50 °C, compounds 7 (1.13 g, 4.05 mmol) and 8-
cyanoquinoline (620 mg, 4.04 mmol) gave imines 8a (535 mg,
36%) and 8b (370 mg, 25%).
4.4.1. N-Methylation at 20 °C
Data for 8a: yellow powder; mp 120–122 °C; Rf = 0.18 (1:9
Methyl iodide (920 mg, 1.2 equiv) was added to the lithioimine
and the resulting mixture was left to react for 15 min, after which
TLC (2:8 EtOAc/iPrOH) showed imine 6 as two spots (Rf = 0.55 and
0.6), the most eluted atropisomer 6a being the major component.
The solvent was removed under reduced pressure, after which
ether (5 ml) was added, the suspension was filtered off and the fil-
trate was concentrated to dryness. The pure-collected column
chromatography (2:8 EtOAc/Heptane) fractions afforded a first
crop of 6a as a white powder (930 mg). A second crop of isomer
6a (78 mg) crystallised out upon cooling overnight. Heptane wash-
ing and subsequent drying provided pure 6a (1.05 g, 73%); mp
EtOAc/cyclohexane); [
a]
D = ꢀ18.6 (c 1, CHCl3); 1H NMR (300 MHz,
CDCl3) d (ppm):1.03–01.25 (t, J = 7.9, 3H), 1.26–1.33 (d, J = 8.1,
3H), 1.55–1.72 (m, 1H), 1.74–1.86 (m, 1H), 3.99 (s, 3H), 4.40–4.49
(m, 1H), 7.14–7.17 (m, 2H), 7.22–7.29 (m, 1H), 7.3–7.35 (m, 1H),
7.37–7.55 (m, 2H), 7.67–7.85 (m, 4H), 8.12 (m, 1H), 10.5 (m, 1H);
13C NMR (75 MHz, CDCl3) d (ppm): 9.1, 19.3, 29.2, 31.9, 61.0,
117.5, 117.7, 119.2, 121.3, 123.3, 123.9, 124.1, 125.6, 125.7, 125.8,
125.9, 126.0, 127.5, 127.6, 127.6, 128.6, 128.7, 134.3, 153.7,
167.5; IR (KBr) 1211, 1424, 1558, 3682; EIMS 368 (M+), 354, 298,
268, 211, 169,155, 83, 58. Anal. Calcd for C25H24N2O: C, 81.49; H,
6.57; N, 7.60. Found: C, 81.52; H, 6.55; N, 7.63.
135–137 °C; [
a
]
D = ꢀ59 (c 1, CHCl3); 1H NMR (500 MHz, CDCl3) d
Data for 8b: a colourless oil Rf = 0.14; [
a
]
D = ꢀ10.5 (c 1, CHCl3);
(ppm): 0.74 (m, 3H, H22), 1.13 (m, 3H, H20), 1.52 (m, 1H, H21),
1.54 (m, 1H, H21), 3.41 (s, 3H, N–CH3), 4.38 (m, 1H, H19), 6.96
(m, 2H, H15,17), 7.15 (d, J = 6.5 Hz, 1H, H14), 7.34 (d, J = 6.4 Hz,
1H, H16), 7.38 (m, 2H, H1,2), 7.5 (m, 2H, H7,6), 7.81 (d, J = 6.5 Hz,
1H, H3,), 7.84 (d, J = 6.5 Hz, 1H, H5), 8.61 (d, J = 6.5 Hz, 1H, H8);
13C NMR (125 MHz, CDCl3) d (ppm): 10.1(C22), 19.3 (C20), 29.4
(C21), 42.2 (C12), 74.7 (C19), 112.8 (C17), 120.2 (C15), 125.0
(C1), 126.0 (C6), 126,6 (C7), 126.8 (C8), 127.4 (C2), 128.2 (C4),
128.4 (C5), 129.4 (C3), 129.6 (C14), 130.2 (C16), 131.7 (C9), 134.3
(C10), 139.6 (C13), 154.8 (C18), 168.6 (C11); IR (KBr): 1263,
1462. Anal. Calcd for C22H23NO: C, 83.24; H, 7.30; N, 4.41. Found:
C, 83.48; H, 7.32; N, 4.38.
1H NMR (300 MHz, CDCl3) d (ppm): 0.39–0.5 (m, 3H), 0.80–1.02 (m,
5H), 3.86 (s, 3H), 4.0–4.20 (m, 1H), 7.05–7.15 (m, 3H), 7.25–7.35
(m, 3H), 7.4–7.55 (m, 2H), 7.75–7.9 (m, 4H); 13C NMR (75 MHz,
CDCl3) d (ppm): 9.2, 19.4, 29.3, 31.9, 61.1, 117.7, 119.3, 123.4,
124.0, 124.1, 125.0, 125.6, 125.7, 125.8, 125.9, 126.0, 127.0,
127.6, 128.6, 128.7, 129.1, 134.4, 153.6, 167.5; IR (neat): 1211,
1424, 1514, 3682; EIMS 368 (M+), 354, 298, 268, 211, 169,155,
83, 58.
4.6. 1-(1-Naphthyl)-1-(2-(1-hydroxyphenyl))methylenimine 9
After a complete halogen–metal exchange using 1-bromonaph-
thalene (1.1 g, 5.378 mmol), 2-cyanophenol (230 mg, 0.5 equiv)
was added at ꢀ50 °C and the reaction mixture was allowed to warm
up to room temperature (2 h) before adding methanol (1 ml). Stir-
ring was continued for 30 min, water (1 ml) was added and the
resulting solution was concentrated to dryness under vacuo. Col-
umn chromatography (3:7 EtOAc/heptane) purification afforded
The washing filtrate and the impure fractions were combined,
after which the solvent was removed under vacuo, and a second
column chromatography provided pure fractions of isomer 6b as
a colourless oil (330 mg, 23%); [
a
]
D = ꢀ32 (c 1, CHCl3); 1H NMR
(600 MHz, CDCl3) d (ppm): 0.1 (t, J = 7.7 Hz, 1.5H, H22), 0.21 (d,
J = 5.8 Hz, 1.5H, H200), 0.5 (m, 0.5H, H210 + 0.5H, H21), 0.55 (t,
J = 7.7 Hz, 1.5H, H220), 0.59 (m, 0.5H, H21), 0.66 (d, J = 5.8 Hz,
1.5H, H20), 0.74 (m, 0.5H, H210), 3.14 and 3.15 (2s, 3H, N–CH3),
3.9–4.05 (m, 1H, H19), 6.72 and 6.73 (2d, J = 8.4 Hz, 1H, H17),
6.9–7.02 (m, 1H, H15), 7.14–7.2 (2d, J = 7.0 Hz, 1H, H1), 7.3–7.35
(dd, J = 7 and 8.4 Hz, 1H, H16), 7.37–7.45 (m, 1H, H2), 7.5–7.55
(m, 2H, H6,7), 7.8–7.84 (m, 3H, H3,8,14), 7.85–7.87 (m, 1H, H5);
methylenimine
9 as a
yellow oil (964 mg, 73%). 1H NMR
(300 MHz, CDCl3) d (ppm): 6.53 (m, 1H), 6.8 (m, 1H), 6.83 (m, 1H)
6,99–7.84 (m, 8H), 9.44 (s, 1H), 14.63 (s, 1H); 13C NMR (75 MHz,
CDCl3) d (ppm): 118.3, 118.6, 119.8, 125.2, 125.5, 125.7, 126.9,
127.4, 128.8, 130.0, 130.4, 132.6, 133.8, 134.00, 137.0, 163.6,
181.7; IR (neat): 1624, 3548; EIMS 246 (M+), 230, 217, 128, 91, 69.

1530
A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
4.7. N-(2(S)-1-Hydroxybut-2-yl)-1-naphtyl-1-(20-hydroxy-
phenyl)methylenimine 4
383, 368, 298, 129. Anal. Calcd for C26H27IN2O: C, 61.18; H, 5.33;
N, 5.49. Found: C, 61.27; H, 5.01; N, 5.45.
A solution of 9 (500 mg, 1.56 mmol) and 2-(S)-aminobutanol
(139 mg, 1.56 mmol) in 1,2-dichloroethane (5 ml) was stirred at
reflux overnight. Column chromatography (3:7 EtOAc/heptane)
purification of the residual oil obtained after solvent removal affor-
ded imine 4 (mixture of two stereoisomers, 7:3) as yellow powder
4.10. Allylation of imine 10
4.10.1. Liquid/liquid system
To a solution of 10 (100 mg, 0.34 mmol), a solution of 2 or 3
(10 mol %) and allyl bromide (105 mg, 0.87 mmol) in toluene
(3 ml) was added at room temperature, and under stirring, 50%
aqueous KOH (1.2 ml) was added. The mixture was then vigorously
stirred at room temperature until the reaction was complete (TLC,
10:90 EtOAc/hexane). Ether (3 ml) was added and the organic layer
was washed with water and dried over MgSO4. Evaporation of sol-
vents and purification of the residual oil by column chromatogra-
(400 mg, 80%); mp 138–140 °C; [a]
D = +76.3 (c 0.535, CHCl3); 1H
NMR (500 MHz, CDCl3) d (ppm): 0.63–0.68 (major) and 0.77–0.82
(minor) (2t, 5.9 Hz, 3H, H21), 1.43–1.57 (m, 3H, H20 and OH),
3.0–3.25 (m, 1H, H18,), 3.52–3.61 (m, 2H, H19), 6.57 (m, 1H,
H14), 6.63 (d, J = 7.5 Hz, 1H, H13), 7.05 (d, J = 8.7 Hz, 1H, H16),
7.28 (m, 1H, H15), 7.37 (d, J = 5.8 Hz, 0.4H, H1), 7.45 (dd, J = 3.1
and 6.2 Hz, 1H, H7), 7.48 (d, J = 6.0 Hz, 0.6H, H1), 7.57 (dd, J = 6.9
and 7.8 Hz, 1H, H6), 7.61–7.64 (m, 1H, H2,8), 7.69 (d, J = 8.7 Hz,
1H, H8), 7.96 (d, J = 8.7 Hz, 1H, H5), 8.04 (d, J = 8.2 Hz, 1H, H3),
15.73 (minor) and 15.75 (major) (2s, 1H, Ph–OH); 13C NMR
(125 MHz, CDCl3) d (ppm): both stereoisomers: 10.92 and 10.99
(C21), 25.8 and 26.2 (C20), 64.3 and 64.5 (C18), 66.0 and 66.4
(C19), 117.9 (C14), 118.6 (C16), 120.3 (C12), 125.4 and 125.5
(C2), 126.0 and 126.06 (C5), 126.07 and 126.91 (C1), 126.98 and
127.1 (C7), 127.2 and 127.5 (C6), 128.9 (C8), 129.7 and 129.8
(C3), 130.8 and 130.9 (C9), 132.02 and 132.06 (C13), 131.86 and
132.24 (C2), 133.09 (C9), 133.1 (C15), 133.71 and 133.74 (C4),
163.9 (C17), 174.9 and 175.1 (C11); EIMS 320 (M+1), 288, 260,
246, 231, 169, 141, 127, 101, 77. Anal. Calcd for C21H21NO2: C,
78.99; H, 6.58; N, 4.38; Found: C, 79.05; H, 6.42; N, 4.35.
phy (EtOAc/hexane = 1:9) gave
colourless oil (80 mg, 70%).
a-allylglycine schiff base 11 as a
4.10.2. Solid/liquid system
A solution of imine 4 (10 mg, 0.034 mmol) and Cu(OAc)2
(8.16 mg, .068 mmol) in methanol (2 ml) was refluxed for 2 h.
The metal complex which crystallised out was collected by filtra-
tion, washed with the minimum of water and methanol, dried
and added to a suspension of imine 10 (100 mg, 0.34 mmol), allyl
bromide (105 mg, 0.87 mmol) and solid KOH (57 mg, 1.02 mmol)
in toluene (3 ml). After 6 h of stirring at room temperature, ether
(5 ml) was added, the solid was removed by filtration and the fil-
trate was washed with water and dried over MgSO4. Evaporation
of solvents and purification of the residual oil by column chroma-
tography (EtOAc/hexane = 1:9) gave product 11 as a colourless oil
(86 mg, 76%). Chiral HPLC showed 2-(S)-allylglycine iminoester as
the major enantiomer with 17% ee.
4.8. N,N-Dimethyl-1-naphthyl-1-(2-(R)-sec-butoxyphenyl)
methyleniminium iodide 2
Acknowledgements
A mixture of imine 6a or 6b (50 mg, 0.13 mol) and methyl io-
dide (55 mg, 3 equiv) in acetonitrile (2 ml) was stirred for 1 h at
room temperature. Acetonitrile was evaporated under reduced
pressure, ether (3 ml) was then added and the solid was collected
by filtration and washed with ether to yield iminium salt 2 as a
white powder in both cases (69 mg, 73%).
We are grateful to the Comité Mixte Inter Universitaire Franco-
Marocain (A.I. 185/SM/99) and the Université de Haute Normandie
(Grant 395072G to A.R) for the financial support.
References
Quaternisation of 6a: mp 195–197 °C; [
CH2Cl2).
Quaternisation of 6b: mp 195–197 °C; [
a
]
D = ꢀ19 (c 0.31,
1. For recent reviews, see (a) Maruoka, K.; Ooi, T. Chem. Rev. 2003, 103, 3013–
3028; (b) O’Donnell, M. J. Acc. Chem. Res. 2004, 37, 506–517; (c) Albanese, D.
Mini-Rev. Org. Chem. 2006, 3, 195–217; (d) Ooi, T.; Maruoka, K. Angew. Chem.,
Int. Ed. 2007, 46, 4222–4266; (e) Hashimoto, T.; Maruoka, K. Chem. Rev. 2007,
107, 5656–5682.
a
]
D = ꢀ12 (c 0.31,
CH2Cl2); 1H NMR (300 MHz, CDCl3) d (ppm): 1.73 (m, 3H), 1.98
(m, 3H), 2.25 (m, 2H), 3.28 and 3.30 (2s, 3H), 3.52 and 3.58 (2s,
3H), 3.9 (m, 1H), 6.59–7.95 (m, 9H), 8.01–8.28 (m, 2H); 13C NMR
(75 MHz, CDCl3) d (ppm): 12.6, 13.4, 21.1, 22.2, 30.2, 30.5, 71.5,
72.1, 117.7, 117.9, 119.0, 128.6, 129.4, 131.7, 132.5, 133.8, 134.4,
135.6, 138.9, 155.1, 156.3, 196.8, 198.3; IR (KBr): 1628; IMS 332
(M+-I), 318, 302, 246, 217, 154, 127. Anal. Calcd for C23H26NOI: C,
60.13; H, 5.66; N, 3.05. Found: C, 60.10; H, 5.42; N, 3.08.
2. Kacprzak, K.; Garonski, J. Synthesis 2001, 961–998.
3. (a) Kita, T.; Georgieva, A.; Hashimoto, Y.; Nataka, T.; Nagasawa, K. Angew.
Chem., Int. Ed. 2002, 41, 2832–2834; (b) Maruoka, K.; Wang, Y.-G. Org. Process
Res. Dev. 2007, 11, 628–632 and earlier references.
4. Gmouh, S.; Jamal-Eddine, J.; Valnot, J. Y. Tetrahedron 2000, 56, 8361–8366.
5. (a) O’Donnell, M. J.; Bennett, W. D.; Wu, S. J. Am. Chem. Soc. 1989, 111, 2353–
2355; (b) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997, 119, 12414–
12415; (c) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1997, 38, 8595–8598.
6. Cram, D. J.; Sogah, G. D. Y. J. Chem. Soc., Chem. Commun. 1981, 625–628.
7. (a) Belokon, Y. N.; Kochetkov, K. A.; Churkina, T. D.; Ikonnikov, S.; Vyskocil, S.;
Kagan, H. B. Tetrahedron: Asymmetry 1999, 10, 1723–1728; (b) Belokon, Y. N.;
North, M.; Kublitski, V. S.; Ikonnikov, N. S.; Krasik, P. E.; Maleev, V. I.
Tetrahedron Lett. 1999, 40, 6105–6108; (c) Belokon, Y. N.; Davies, R. G.;
North, M. Tetrahedron Lett. 2000, 41, 7245–7248.
4.9. N,N-Dimethyl-1-(8-quinolinyl)-1-(2-(R)-sec-butoxy-
naphtyl)-methyleniminium iodide 3
8. Hindered cyclic chiral iminium salts have been reported to catalyse the alkene
epoxidation: Page, P. C. B.; Buckley, B. R.; Barros, D.; Blacker, A. J.; Marples, B.
A.; Elsegood, M. R. J. Tetrahedron 2007, 63, 5386–5393. and references therein.
9. Gmouh, S.; Jamal Eddine, J.; Maddaluno, J. F.; Oulyadi, H.; Valnot, J.-Y. Abstracts
of Papers, 225th ACS National Meeting, New Orleans, LA, United States, March
23–27, 2003.
Quaternisation of 8a or 8b (50 mg, 0.135 mmol) following the
above procedure gave iminium salt 3 as a white solid in both cases
(58 mg, 85%).
Quaternisation of 8a: mp 186–187 °C; [
Quaternisation of 8b: mp 186–187 °C; [
a
]
]
D = ꢀ17 (c 1, CHCl3).
a
D = ꢀ11.5 (c 1, CHCl3).
10. Casarini, D.; Lunazzi, L.; Macciantelli, D. J. Chem. Soc., Perkin Trans. 2 1992,
1363–1370.
1H NMR (300 MHz, CDCl3) d (ppm): 1.13 (t, J = 7.3 Hz, 3H), 1.21
(m, 3H), 1.28 (m, 2H), 3.01 and 3.32 (2s, 3H), 4.02 (m, 1H), 7.27–
7.89 (m, 9H), 8.01–8.28 (m, 3H); 13C NMR (75 MHz, CDCl3) d
(ppm): 12.2, 14.2, 22.1, 26.2, 33.5, 70.3, 73.1, 117.1, 118.0, 119.7,
127.6, 127.8, 128.6, 128.7, 129.4, 132.0, 132.6, 134.0, 134.3,
136.4, 138.9, 153.7, 155.1, 157.3, 179.3, 195.8, 199.8; IR (KBr):
1190, 1260, 1270, 1300, 1520, 1620, 3245; EIMS 383,21 (M+-I),
11. (a) Boyd, D. R.; Al-Showiman, S.; Jennings, W. B. J. Org. Chem. 1978, 43, 3335–
3339; (b) Hamor, T. A.; Jennings, W. B.; Proctor, L. D.; Tolley, M. S.; Boyd, D. R.;
Mullan, T. J. Chem. Soc., Perkin Trans. 2 1990, 25–30; (c) Le Gac, S.; Monnier-
Benoit, N.; Metoul, L. D.; Petit, S.; Jabin, I. Tetrahedron: Asymmetry 2004, 15,
139–145.
12. Guerra, A.; Lunazzi, L. J. Org. Chem. 1995, 60, 7959–7965.
13. X-ray data for isomer 6a:
C22H23NO, Mr = 317.41, monoclinic, P21, a =
7.5510(2), b = 15.5062(4), c = 8.2576(2) Å, b = 108.036(1)°, V = 919.35(4) Åꢀ3
,

A. Retmane et al. / Tetrahedron: Asymmetry 19 (2008) 1523–1531
1531
ıtZ = 2, Dx = 1.297 Mg mꢀ3, k(MoK
T = 293 K. The sample (0.38 ꢁ 0.25 ꢁ 0.18 mm) was studied on a NONIUS Kappa
CCD with graphite monochromatised MoK radiation. The cell parameters
were obtained with Denzo and Scalepack (Otwinowski & Minor, 1997) with 10
frames (psi rotation: 1° per frame). The data collection (Nonius, 1999)
(2hmax = 60°, 183 frames via 2.0° omega rotation and 14 s per frame, range
HKL: H 0, 9 K 0, 20 L ꢀ10, 10) gives 11,180 reflections. The data reduction with
Denzo and Scalepack (Otwinowski and Minor, 1997) led to 2197 independent
a
) = 0.71073 Å,
l
= 0.69 cmꢀ1, F(000) = 340,
the full-matrix least-square techniques (use of F square magnitude; x, y, z, bij
for O, C and N atoms, x, y, z in riding mode for H atoms; 218 variables and 1842
2
a
observations with I > 2.0
r
(I); calcd w ¼ 1=½r2ðFo2Þ þ ð0:096PÞ þ 0:028Pꢂ, where
P ¼ ðF2o þ 2Fc2Þ=3 with the resulting R = 0.046, Rw = 0.130 and Sw = 1.055
(residual around solvent molecules)
Dq
< 0.20 e Åꢀ3).Atomic scattering
factors from International Tables for X-ray crystallography (1992). Ortep
views realised with PLATON98 (Spek, 1998). All the calculations were performed
on a Pentium NT Server computer. CCDC 676002.
reflections from which 1842 with I > 2.0
r
(I). The structure was solved with SIR
-
14. Prelog, V.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1982, 21, 567–583.
15. O’Donnell, M. J.; Polt, R. L. J. Org. Chem. 1982, 47, 2663–2666.
16. MOPAC 2007, James J. P. Stewart, Stewart computational Chemistry, Colorado
Spring, CO, USA. http://OpenMOPAC.net.
97 (Altomare et al., 1998) which reveals the non-hydrogen atoms. After
anisotropic refinement, many hydrogen atoms may be found with a Fourier
Difference. The whole structure was refined with SHELXL97 (Sheldrick, 1997) by