Structural diversity in imidazolidinone organocatalysts: A synchrotron and computational study

Article abstract of DOI:10.1107/S0108270107051396

(5)-1-(Methylaminocarbonyl)-3-phenylpropanaminium chloride (52·HCl), C₁₀H₁₅N₂O⁺·Cl^-, crystallizes in the orthorhombic space group P2₁2₁2 ₁ with a single formula unit per asymmetric unit. (5R/S)-5-Benzyl-2,2,3-trimethyl-4-oxoimidazolidin-1-ium chloride (R3 and S3), C₁₃H₁₉N₂O⁺·Cl^-, crystallize in the same space group as S2-HCl but contain three symmetry-independent formula units. (R/S)-5-Benzyl-2,2,3-trimethyl-4- oxoimidazolidin-1-ium chloride monohydrate (R4 and S4), C₁₃H ₁₉N₂O⁺-Cl^-·H₂O, crystallize in the space group P2₁ with a single formula unit per asymmetric unit. Calculations at the B3LYP/6-31G(d,p) and B3LYP/6-311G(d,p) levels of the conformational energies of the cation in R3, S3, R4 and S4 indicate that the ideal gas-phase global energy minimum conformation is not observed in the solid state. Rather, the effects of hydrogen-bonding and van der Waals interactions in the crystal structure cause the molecules to adopt higher-energy conformations, which correspond to local minima in the molecular potential energy surface.

Full text of DOI:10.1107/S0108270107051396

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
recrystallization. Both of these factors are expected to be
highly relevant to possible use of the materials in both
industrial and laboratory settings.
ISSN 0108-2701
The catalyst studied in the current work is a typical example
of the class (Ahrendt et al., 2000), is commercially available,
and has been described as `an inexpensive, bench-stable solid
that is readily handled by experimentalists or automated
systems' (Beeson & MacMillan, 2005). We therefore selected
this compound for a full structural characterization, which is of
interest both to con®rm the reported molecular structure of
this material, and with a view to rationalizing its solid-state
properties, including particle morphology and ¯ow, solubility,
stability, etc. Synchrotron radiation was employed in order to
gain the best quality structural information from these weakly
scattering materials and to ensure that our study was not
biased in favour of solid forms which readily form single
crystals. A simpli®ed reaction scheme is given below.
Structural diversity in imidazolidinone
organocatalysts: a synchrotron and
computational study
Jonathan C. Burley,a* Ryan Gilmour,b Timothy J. Priorc
and Graeme M. Dayd
aSchool of Pharmacy, University of Nottingham, University Park, Nottingham
NG7 2RD, England, bETH Zurich, Swiss Federal Institute of Technology,
È
Laboratorium fur Organische Chemie, ETH Honggerberg, Wolfgang-Pauli-Strasse 10,
È
È
CH-8093 Zurich, Switzerland, cDepartment of Chemistry, University of Hull,
È
Cottingham Road, Kingston-upon-Hull HU6 7RX, England, and dThe University
Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW, England
Correspondence e-mail: jonathan.burley@nottingham.ac.uk
Received 17 August 2007
Accepted 17 October 2007
Online 14 December 2007
(S)-1-(Methylaminocarbonyl)-3-phenylpropanaminium chloride
(S2ÁHCl), C10H15N2O+ÁCl , crystallizes in the orthorhombic
space group P212121 with a single formula unit per asymmetric
unit. (5R/S)-5-Benzyl-2,2,3-trimethyl-4-oxoimidazolidin-1-ium
chloride (R3 and S3), C13H19N2O+ÁCl , crystallize in the same
space group as S2ÁHCl but contain three symmetry-indepen-
dent formula units. (R/S)-5-Benzyl-2,2,3-trimethyl-4-oxoimi-
dazolidin-1-ium chloride monohydrate (R4 and S4),
C13H19N2O+ÁCl ÁH2O, crystallize in the space group P21 with
a single formula unit per asymmetric unit. Calculations at the
B3LYP/6±31G(d,p) and B3LYP/6±311G(d,p) levels of the
conformational energies of the cation in R3, S3, R4 and S4
indicate that the ideal gas-phase global energy minimum
conformation is not observed in the solid state. Rather, the
effects of hydrogen-bonding and van der Waals interactions in
the crystal structure cause the molecules to adopt higher-
energy conformations, which correspond to local minima in
the molecular potential energy surface.
Comment
A number of research groups have reported on the utility of
chiral imidazolidinone catalysts for a range of enantioselective
organic reactions (Ahrendt et al., 2000; Wilson et al., 2005;
Paras & MacMillan, 2001; Jen et al., 2000; Marigo et al., 2005;
Kunz & MacMillan, 2005; Beeson & MacMillan, 2005; Brochu
et al., 2004). Several reviews of the ®eld of organocatalysis
exist (AlmasËi et al., 2007; Bolm et al., 2005; Gaunt et al., 2007;
Tsogoeva, 2007; List, 2006). Enantiomeric excesses greater
than 90% are regularly achieved in relatively high yielding
reactions that are dif®cult to accomplish by other means.
However, there are to date very few structural studies of these
materials, and no reports of their stability to storage or
S2ÁHCl
[(S)-1-(methylaminocarbonyl)-3-phenylpropan-
aminium chloride] adopts the orthorhombic space group
P212121 with a single formula unit per asymmetric unit
(Fig. 1a). All chemically intuitive hydrogen-bonding oppor-
tunities are satis®ed in the solid state (Fig. 2a). Sheets of
hydrogen bonds are formed, which lie parallel to the ab plane.
The chloride ion is hydrogen bonded by three protonated
amine groups [N16Á Á ÁCl1
= 3.094 (2), 3.149 (2) and
Ê
3.166 (2) A], and the carbonyl group of the amide unit is
hydrogen bonded to the amide H atom on an adjacent mol-
Ê
ecule [N11Á Á ÁO10 = 3.018 (3) A]. The hydrophobic (phenyl)
o10 # 2008 International Union of Crystallography
DOI: 10.1107/S0108270107051396
Acta Cryst. (2008). C64, o10±o14

organic compounds
and hydrophilic (amide) fragments of the molecule form
layers, which are segregated along the c axis. Solid-state
packing forces appear to have a slight in¯uence on the mol-
ecular conformation; comparison of the observed molecular
conformation with one computed using a gas-phase molecular
mechanics model shows no unusual features, the two confor-
mations being broadly similar. The primary differences arise
because of a change in the C6ÐC7ÐC8ÐC9 torsion angle
from 98.15ꢀ in the gas phase (calculated as described in the
Experimental) to 54.65ꢀ in the solid-state, which is presumably
driven by a combination of crystal packing (optimal space
®lling) requirements and intermolecular hydrogen bonding of
the amide group.
R3 and S3 [(5R/S)-5-benzyl-2,2,3-trimethyl-4-oxoimidazo-
lidin-1-ium chloride] both crystallize in the orthorhombic
space group P212121 with three formula units per asymmetric
unit (S3 is shown in Fig. 1b). The three organic molecules
adopt different but related conformations, which differ
primarily as a result of changes in the orientation of the
imidazolidinone fragment with respect to the phenyl ring via
CÐC bond rotation. These differences in conformation do not
affect the hydrogen-bonding topology, with all three organic
cations forming two NÐHÁ Á ÁCl hydrogen bonds from the
protonated N atom to the free chloride ion (Fig. 2b). Neither
the carbonyl group nor the N atom of the amide group are
involved in classical hydrogen bonding for any of the three
distinct molecules, which is in contrast to the extensive
hydrogen bonding observed for the precursor S2ÁHCl. Two of
the three independent formula units are linked to one another
by hydrogen bonding through atoms Cl1 and Cl2 via NÐ
HÁ Á ÁCl interactions and thereby form a chain of hydrogen
bonds which runs parallel to the a axis. The third symmetry-
independent formula unit again forms NÐHÁ Á ÁCl hydrogen
bonds to create a chain that runs parallel to both the a axis and
the other hydrogen-bonding chain, but in this case the chain
comprises only a single symmetry-independent molecule and a
single symmetry-independent chloride ion (Cl1). The hydro-
philic sections of the structure (Cl ions and the polar parts of
the organic cation) are layered in the ac plane and are sepa-
rated by the hydrophobic phenyl rings. As expected, no
signi®cant differences exist between the crystal packing of the
R and S isomers.
The presence of three formula units with different mol-
ecular conformations in the asymmetric unit is curious, and
could result from either different minima on the molecular
energy surface or distortions away from the same minimum
caused by differences in packing forces experienced by the
three independent molecules in the unit cell. To distinguish
between these possibilities, we have performed calculations
designed to elucidate the gas-phase conformers and their
relative energies. R3 and S3 have two torsional degrees of
freedom, which we label ' (torsion angle C6ÐC7ÐC8ÐC9)
and ꢀ (torsion angle C5ÐC6ÐC7ÐC8). Initial low-level
calculations (using the molecular mechanics model Tripos 5.2)
indicate that three distinct conformational energy minima
exist in the gas phase (Fig. 3). The three minima were then re-
optimized at higher levels of theory for more reliable relative
energies, revealing large energy differences between the
conformations: the relative energies are 0, +10.67 [+11.29] and
1
+32.57 [+33.95] kJ mol
for conformations i, ii and iii,
respectively, at the B3LYP/6±31G(d,p) [B3LYP/6±311G(d,p)]
levels. The good quantitative agreement between the two
methods indicates that the effect of basis set superposition
error on the relative conformational energies is small, while
the stability of the most folded conformation i might even be
underestimated owing to the known poor treatment of
nonbonded interactions in density functional theory calcula-
tions (van Mourik et al., 2006). As typical energy differences
1
between polymorphs are of the order of 1±10 kJ mol
(Bernstein, 2002), we might only expect the lowest energy
conformation to be observed in the crystal structure.
However, it is clear from Fig. 3 that two molecules (A and C)
adopt conformations related to the intermediate-energy
calculated conformation ii, while molecule B is within the
Figure 1
The asymmetric units of (a) S2ÁHCl, (b) S3 and (c) S4. Displacement ellipsoids are shown at the 50% probability level.
C10H15N2O+ÁCl , C13H19N2O+ÁCl and C13H19N2O+ÁCl ÁH2O o11
ꢁ
Acta Cryst. (2008). C64, o10±o14
Burley et al.

organic compounds
potential energy well of the highest-energy calculated
conformation iii, although signi®cantly distorted away from
the minimum. It therefore appears that in order to optimize
the lattice energy, the three molecules avoid the lowest-energy
folded gas-phase conformation in favour of the two open
conformations. The conformational energy penalty of at least
18 kJ mol 1 (2 Â 11 kJ mol 1 for A and C, 33 kJ mol 1 for B)
is presumably compensated by the formation of hydrogen
bonds and van der Waals interactions in the crystal structure.
Both R4 and S4 [(R/S)-5-benzyl-2,2,3-trimethyl-4-oxoimi-
dazolidin-1-ium chloride monohydrate] crystallize in the
noncentrosymmetric monoclinic space group P21 with a single
formula unit per asymmetric unit (S4 is shown in Fig. 1c). The
conformation of the cation (ꢀ = 109.3ꢀ and ' = 173.4ꢀ)
corresponds most closely with minimum ii (Fig. 3), and it
therefore appears that, as with R3 and S3, intermolecular
factors including hydrogen-bonding and van der Waals inter-
actions outweigh intramolecular ones in determining the
conformation of the cation in the solid state. Overall, the
hydrogen bonding leads to the formation of tapes which run
parallel to the b axis (Fig. 2c). Again, the hydrophobic and
hydrophilic parts of the structure are segregated into chains
which follow the hydrogen-bonding topology. Again, no
unexpected differences exist between the crystal packing of
the R and S isomers.
In conclusion, computational and experimental studies
indicate that intermolecular rather than intramolecular forces
are primarily responsible for the molecular conformations
observed in the solid state for this class of compounds.
Experimental
(5S)-5-Benzyl-2,2,3-trimethylimidazolidin-4-one was synthesized
from commercially available (S)-phenylalanine methyl ester hydro-
chloride according to the literature method described by Ahrendt et
al. (2000). Treatment of (S)-phenylalanine methyl ester hydro-
chloride with ethanolic methylamine furnished the intermediate (S)-
phenylalanine N-methylamide hydrochloride, which was not isolated,
but heated to re¯ux in MeOH and acetone with catalytic
p-toluenesulfonic acid (p-TSA). The cyclized (S)-imidazolidinone
product was precipitated as the HCl salt and recrystallized from
2-propanol to yield a microcrystalline white needle-like material,
which was identical to that previously characterized. The R enan-
tiomer was prepared in an identical manner starting from commer-
cially available (R)-phenylalanine methylamide hydrochloride. This
material displayed identical 1H and 13C NMR spectra to the S
enantiomer, and had an equal but opposite optical rotation value, as
expected. A simpli®ed reaction scheme is shown in the Comment
section.
Figure 2
Hydrogen-bonding patterns observed in (a) S2ÁHCl (the hydrogen-
bonded sheet is viewed side-on for clarity), (b) S3 (only one of the three
chains is shown for clarity) and (c) S4. (Colour key for the electronic
version of the paper: carbon grey, hydrogen white, nitrogen blue, oxygen
red and chlorine green.)
Initial attempts to grow large crystals by slow evaporation (several
months, 277 K) from CHCl3 were successful in this aim, but crystal-
lographic characterization indicated that ring opening of the imida-
zolidinone had occurred to yield (R/S)-phenylalanine methylamide
hydrochloride. As the crystal structure of this material had not been
reported, a full data set was collected on the S isomer and is reported
here. Crystals of (5R/S)-5-benzyl-2,2,3-trimethylimidazolidin-4-one
hydrochloride were grown by rapid evaporation (ca 2 d, ambient
temperature) from CHCl3. Samples of (5R/S)-5-benzyl-2,2,3-trime-
thyl-4-oxoimidazolidin-1-ium chloride monohydrate were crystallized
from 2-propanol as a microcrystalline powder and correspond to the
`raw' reagent described in the synthesis; these were subjected to
standard benchtop storage conditions for several months after
crystallization.
Single crystals were selected, mounted at the end of two-stage glass
®bres and studied at 150 K on station 9.8 or station 16.2 SMX of the
UK Synchrotron Radiation Source, Daresbury. Routine data collec-
tion involved three series of ! scans. Data were corrected for beam
decay and absorption using a method based on equivalents. Further
details can be found in the relevant section of the CIF that accom-
panies this paper.
Computational studies were performed to investigate the confor-
mational energy landscape of the molecular cations. Both for
protonated (R/S)-phenylalanine methylamide and for the imidazoli-
dinium cation, initial calculations were performed using a simple
molecular mechanics model (Tripos 5.2), as implemented in Ghemical
Figure 3
Calculated (Tripos 5.2) conformational energy landscape for the (S)-5-
benzyl-2,2,3-trimethyl-4-oxoimidazolidin-1-ium cation, and energy-mini-
mized calculated conformations. Numbers refer to calculated conforma-
tions, letters to those observed experimentally. See Comment and Fig. 1
for further details.
È È
(Hassinen & Perakyla, 2001). For the imidazolidinium cation, for
o12 Burley et al. C10H15N2O+ÁCl , C13H19N2O+ÁCl and C13H19N2O+ÁCl ÁH2O
Acta Cryst. (2008). C64, o10±o14
ꢁ

organic compounds
which the potential energy landscape was relatively complex, the
three local minima located in the initial molecular mechanics search
were investigated at a higher level of theory. The three molecular
mechanics energy minima were energy minimized using both B3LYP/
6±31G(d,p) and B3LYP/6±311G(d,p) levels of theory, giving re®ned
gas-phase molecular structures and their relative energies. All density
functional theory calculations were performed using the program
CADPAC (Amos, 1995).
Data collection
Bruker D8 diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2003)
Tmin = 0.949, Tmax = 0.990
48955 measured re¯ections
13171 independent re¯ections
8896 re¯ections with I > 2ꢃ(I)
Rint = 0.126
Re®nement
R[F2 > 2ꢃ(F2)] = 0.071
wR(F2) = 0.171
S = 1.03
13171 re¯ections
488 parameters
6 restraints
H atoms treated by a mixture of
independent and constrained
re®nement
3
Ê
Compound S2ÁHCl
Crystal data
Áꢄmax = 0.46 e A
3
Ê
0.32 e A
Áꢄmin
=
Absolute structure: Flack (1983),
5804 Friedel pairs
Flack parameter: 0.09 (6)
C10H15N2O+ÁCl
Z = 4
Synchrotron radiation
Mr = 214.69
Orthorhombic, P212121
Ê
ꢁ = 0.69040 A
ꢂ = 0.32 mm
T = 150 (2) K
1
Ê
a = 4.9758 (7) A
Ê
b = 8.6213 (13) A
Compound R4
Ê
c = 25.521 (4) A
0.15 Â 0.07 Â 0.07 mm
3
Ê
V = 1094.8 (3) A
Crystal data
C13H19N2O+ÁCl ÁH2O
3
Ê
V = 724.2 (3) A
Data collection
Mr = 272.77
Monoclinic, P21
a = 9.6320 (19) A
Z = 2
Synchrotron radiation
Bruker D8 diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2003)
Tmin = 0.953, Tmax = 0.978
12758 measured re¯ections
3619 independent re¯ections
2971 re¯ections with I > 2ꢃ(I)
Rint = 0.091
Ê
Ê
ꢁ = 0.79770 A
ꢂ = 0.26 mm
T = 150 (2) K
1
Ê
b = 7.0446 (14) A
Ê
c = 11.093 (2) A
ꢅ = 105.82 (3)ꢀ
0.08 Â 0.03 Â 0.01 mm
Re®nement
R[F2 > 2ꢃ(F2)] = 0.058
wR(F2) = 0.157
S = 1.00
3619 re¯ections
141 parameters
4 restraints
H atoms treated by a mixture of
independent and constrained
re®nement
Data collection
Bruker D8 diffractometer
6282 measured re¯ections
3480 independent re¯ections
2485 re¯ections with I > 2ꢃ(I)
Rint = 0.049
Absorption correction: multi-scan
(SADABS; Sheldrick, 2003)
Tmin = 0.979, Tmax = 1.000
3
Ê
Áꢄmax = 0.58 e A
3
Ê
0.57 e A
Áꢄmin
=
Absolute structure: Flack (1983),
1447 Friedel pairs
Flack parameter: 0.01 (9)
(expected range = 0.978±0.999)
Re®nement
R[F2 > 2ꢃ(F2)] = 0.056
wR(F2) = 0.130
S = 0.96
3480 re¯ections
182 parameters
5 restraints
H atoms treated by a mixture of
independent and constrained
re®nement
Compound R3
Crystal data
C13H19N2O+ÁCl
3
Ê
Áꢄmax = 0.30 e A
Z = 12
Synchrotron radiation
3
Ê
0.39 e A
Áꢄmin
=
Mr = 254.75
Orthorhombic, P212121
Ê
Absolute structure: Flack (1983),
1560 Friedel pairs
Flack parameter: 0.15 (9)
ꢁ = 0.69040 A
ꢂ = 0.26 mm
T = 150 (2) K
1
Ê
a = 7.1167 (14) A
Ê
b = 19.237 (4) A
Ê
c = 30.370 (6) A
Ê
V = 4157.7 (14) A
0.1 Â 0.08 Â 0.04 mm
3
Compound S4
Data collection
Bruker D8 diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2003)
Tmin = 0.974, Tmax = 0.990
24370 measured re¯ections
9772 independent re¯ections
7354 re¯ections with I > 2ꢃ(I)
Rint = 0.071
Crystal data
C13H19N2O+ÁCl ÁH2O
3
Ê
V = 725.57 (18) A
Mr = 272.77
Monoclinic, P21
a = 9.6425 (14) A
Z = 2
Synchrotron radiation
Ê
Ê
ꢁ = 0.79770 A
ꢂ = 0.26 mm
T = 150 (2) K
Re®nement
1
Ê
b = 7.0517 (10) A
R[F2 > 2ꢃ(F2)] = 0.056
wR(F2) = 0.131
S = 0.99
9772 re¯ections
488 parameters
6 restraints
H atoms treated by a mixture of
independent and constrained
re®nement
Ê
c = 11.0895 (16) A
ꢅ = 105.796 (2)ꢀ
0.04 Â 0.04 Â 0.01 mm
3
Ê
Áꢄmax = 0.28 e A
Data collection
3
Ê
0.23 e A
Áꢄmin
=
Bruker D8 diffractometer
Absorption correction: multi-scan
(SADABS; Sheldrick, 2003)
Tmin = 0.987, Tmax = 0.990
6377 measured re¯ections
3330 independent re¯ections
2438 re¯ections with I > 2ꢃ(I)
Rint = 0.052
Absolute structure: Flack (1983),
4169 Friedel pairs
Flack parameter: 0.00 (6)
Compound S3
Re®nement
Crystal data
C13H19N2O+ÁCl
R[F2 > 2ꢃ(F2)] = 0.054
wR(F2) = 0.122
S = 0.98
3330 re¯ections
181 parameters
5 restraints
H atoms treated by a mixture of
independent and constrained
re®nement
Z = 12
Synchrotron radiation
Mr = 254.75
Orthorhombic, P212121
3
Ê
Ê
ꢁ = 0.69040 A
ꢂ = 0.26 mm
T = 150 (2) K
Áꢄmax = 0.27 e A
1
3
Ê
a = 7.123 (1) A
Ê
0.38 e A
Áꢄmin
=
Ê
b = 19.232 (3) A
Absolute structure: Flack (1983),
1395 Friedel pairs
Flack parameter: 0.01 (9)
Ê
c = 30.384 (4) A
0.2 Â 0.06 Â 0.04 mm
3
Ê
V = 4162.3 (10) A
C10H15N2O+ÁCl , C13H19N2O+ÁCl and C13H19N2O+ÁCl ÁH2O o13
ꢁ
Acta Cryst. (2008). C64, o10±o14
Burley et al.

organic compounds
Beeson, T. D. & MacMillan, D. W. C. (2005). J. Am. Chem. Soc. 127, 8826±
8828.
Bernstein, J. (2002). Polymorphism in Molecular Crystals. Oxford: Clarendon
Press.
Bolm, C., Rantanen, T., Schiffers, I. & Zani, L. (2005). Angew. Chem. Int. Ed.
44, 1758±1763.
Brochu, M. P., Brown, S. P. & MacMillan, D. W. C. (2004). J. Am. Chem. Soc.
126, 4108±4109.
For all compounds, data collection: APEX2 (Bruker, 2004); cell
re®nement: APEX2; data reduction: APEX2; program(s) used to
solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to
re®ne structure: SHELXL97 (Sheldrick, 1997); molecular graphics:
WinGX (Version 1.64; Farrugia, 1999); software used to prepare
material for publication: SHELXL97.
Bruker (2004). APEX2. Bruker AXS Inc., Madison, Wisconsin, USA.
Farrugia, L. J. (1999). J. Appl. Cryst. 32, 837±838.
Flack, H. D. (1983). Acta Cryst. A39, 876±881.
Gaunt, M. J., Johansson, C. C. C., McNally, A. & Vo, N. T. (2007). Drug Discov.
Today, 12, 8±27.
The authors thank CCLRC (now STFC) for access to
microcrystal diffraction facilities at Daresbury SRS.
È
È
Hassinen, T. & Perakyla, M. (2001). J. Comput. Chem. 22, 1229±1242.
Jen, W. S., Wiener, J. J. M. & MacMillan, D. W. C. (2000). J. Am. Chem. Soc.
122, 9874±9875.
Kunz, R. K. & MacMillan, D. W. C. (2005). J. Am. Chem. Soc. 127, 3240±3241.
List, B. (2006). Chem. Commun. pp. 819±824.
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: SF3055). Services for accessing these data are
described at the back of the journal.
Â
Marigo, M., Franzen, J., Poulsen, T. B., Zhuang, W. & Jùrgensen, K. A. (2005).
J. Am. Chem. Soc. 127, 6964±6965.
Mourik, T. van, Karamertzanis, P. G. & Price, S. L. (2006). J. Phys. Chem. A,
110, 8±12.
Paras, N. A. & MacMillan, D. W. C. (2001). J. Am. Chem. Soc. 123, 4370±4371.
Sheldrick, G. M. (1997). SHELXS97 and SHELXL97. University of
References
Ahrendt, K. A., Borths, C. J. & MacMillan, D. W. C. (2000). J. Am. Chem. Soc.
122, 4243±4244.
Â
AlmasËi, D., Alonso, D. & Najera, C. (2007). Tetrahedron Asymmetry, 18, 299±
365.
Amos, R. D. (1995). CADPAC. Version 6.0. University of Cambridge,
England. Contributions from I. L Alberts, J. S. Andrews, S. M. Colwell, N. C.
Handy, D. Jayatilaka, P. J. Knowles, R. Kobayashi, N. Koga, K. Laidig, P. E.
Maslen, C. W. Murray, J. E. Rice, J. Sanz, E. D. Simandiras, A. J. Stone &
M. D. Su.
È
Gottingen, Germany.
Sheldrick, G. M. (2003). SADABS. Version 2.10. University of Gottingen,
Germany.
È
Tsogoeva, S. B. (2007). Eur. J. Org. Chem. pp. 1701±1716.
Wilson, R. M., Jen, W. S. & MacMillan, D. W. C. (2005). J. Am. Chem. Soc. 127,
11616±11617.
o14 Burley et al. C10H15N2O+ÁCl , C13H19N2O+ÁCl and C13H19N2O+ÁCl ÁH2O
Acta Cryst. (2008). C64, o10±o14
ꢁ