(2S,3S)-2-(N,N-Dibenzylamino)-butane-1,3-diol refined using a multipolar atom model

Article abstract of DOI:10.1107/S0108270107043296

The crystal structure of the title compound, C₁₈H ₂₃NO₂, was determined using the experimental library multipolar atom model. The refinement showed a significant improvement of crystallographic statistical indices when com

Full text of DOI:10.1107/S0108270107043296

organic compounds
Acta Crystallographica Section C
Crystal Structure
Communications
tions with toluene as the sole by-product (Yoshida et al., 1988).
We report here the X-ray structure of (II), which was prepared
in good yield by reduction of the intermediate ester, (I).
ISSN 0108-2701
(2S,3S)-2-(N,N-Dibenzylamino)-
butane-1,3-diol refined using a
multipolar atom model
Krzysztof Ejsmont,a Jean-Pierre Joly,b Emmanuel Wengerc
and Christian Jelschc*
The molecular structure of (II) is depicted in Fig. 1. An
intramolecular O1ÐHO1Á Á ÁN1 hydrogen bond (Table 2)
aInstitute of Chemistry, University of Opole, Oleska 48, 45-052 Opole, Poland,
Ê
forms a ®ve-membered ring. Atom N1 lies about 0.407 A out
b
Â
Â
Á
Â
UMR 7565 (SRSMC) ± Structure et Reactivite des Systemes Moleculaires Complexes,
of the plane de®ned by the three neighbouring C atoms (C1,
C5 and C12), towards atom HO1 (Fig. 1). The sum of the
valence angles around the N atom is 338.9ꢀ, indicating sp3-
hybridization (328ꢀ for sp3 and 360ꢀ for sp2). The crystal
structure of (II) is presented in Fig. 2. Chains along the b axis
are formed by O2ÐHO2Á Á ÁO1i hydrogen bonds [symmetry
code: (i) x, y + 1, z; Table 2].
Â
Â
Â
Universite Henri Poincare Nancy 1, Faculte des Sciences, BP 239, 54506
Vandoeuvre les Nancy Cedex, France, and cLaboratoire de Cristallographie et
Á
Â
Â
Â
Modelisation des Materiaux Mineraux et Biologiques (LCM3B), CNRS, UMR 7036,
Â
Â
Â
Universite Henri Poincare Nancy 1, Faculte des Sciences, BP 239, 54506
Á
Vandoeuvre les Nancy Cedex, France
Correspondence e-mail: christian.jelsch@lcm3b.uhp-nancy.fr
Received 27 June 2007
Initially, in the independent atom model (IAM) re®nement,
a conventional spherical neutral atom model was applied.
Scale factors, atomic positions and displacement parameters
for all atoms were re®ned using the MoPro program (Guillot
et al., 2001; Jelsch et al., 2005) until convergence. In the
experimental library multipolar atom model (ELMAM)
re®nement, the same parameters were varied but a multipolar
charged atom model was applied. The electron-density para-
meters were transferred from the ELMAM library (Pichon-
Pesme et al., 2004; Zarychta et al., 2007) and subsequently kept
®xed. Riding restraints on H-atom B factors were applied
similarly in both re®nements, which were carried out with the
same intensity data and cut-off criterion [I/ꢁ(I) ꢁ 0]. The HÐ
X distances were restricted to standard values in X-ray and
Accepted 4 September 2007
Online 14 December 2007
The crystal structure of the title compound, C18H23NO2, was
determined using the experimental library multipolar atom
model. The re®nement showed a signi®cant improvement of
crystallographic statistical indices when compared with a
conventional spherical neutral atom re®nement.
Comment
Enantiomerically pure amino acids and their derivatives play
an important role as indispensable intermediates in the
pharmaceutical industry and agrochemistry (Breuer et al.,
2004, and references therein). Among these optically active
amino-alcohols, l- and/or d-threoninol have recently received
much attention as chiral parts of promising drugs against
chemokine-mediated diseases (Brough & McInally, 2004), for
the synthesis of acridine±DNA conjugates for site-selective
RNA scissors (Shi et al., 2005), as linkers for Methyl Red
incorporation into DNA (Kashida, Tanaka et al., 2006), as
asymmetric cis-platinum(II) complexes for cancer therapy
(van Rijt et al., 2006), as intercalators between pyrene moities
and an oligodeoxyribonucleotide for the detection of deletion
polymorphisms (Kashida, Asanuma et al., 2006) or between
10±23 DNAzyme and a binding arm for RNA cleavage activity
enhancement (Asanuma et al., 2006), and for the synthesis of
inhibitors of protein kinase C for the treatment of in¯amma-
tion (Peng et al. 2006) or of a small combinatorial library of
dihydroceramide analogs (Villorbina et al., 2007).
As part of a programme for the evaluation of new asym-
metric catalysts from the chiral pool (Waykole et al., 2007) (i.e.
amino acids, sugars, etc.), we have synthesized the new opti-
cally pure ꢀ-amino alcohol (II) in two steps from readily
available d-threonine. Diol (II) is a convenient source of
d-threoninol by catalytic hydrogenolysis under mild condi-
Figure 1
The molecular structure of (II), showing the intramolecular hydrogen
bond (dashed line). Displacement ellipsoids are drawn at the 50%
probability level and H atoms are shown as small spheres of arbitrary
radii.
o18 # 2008 International Union of Crystallography
DOI: 10.1107/S0108270107043296
Acta Cryst. (2008). C64, o18±o20

organic compounds
cally stirred dispersion of 98% d-threonine (2.02 g, 16.6 mmol) and
Na2CO3 (3.875 g, 19.5 mmol, ꢂ2.2 equivalents) in 75% aqueous
EtOH (50 ml) at room temperature. The resulting mixture was
re¯uxed for 2.5 h [the formation of (I) being monitored by SiO2 thin-
layer chromatography (TLC); RF = 0.43; AcOEt±cyclohexane 1:4],
cooled to room temperature, concentrated under reduced pressure,
and partitioned between dichloromethane and water (2 Â 50 ml). The
aqueous phase was extracted with dichloromethane (3 Â 50 ml). The
combined organic phases were washed with saturated aqueous
NaHCO3 (50 ml) and water (3 Â 50 ml), dried over MgSO4, and
®nally concentrated in vacuo to afford the crude ester (I) as a pale-
yellow syrup, which was used for the next step without further
puri®cation (yield 5.8 g, 14.9 mmol, ꢂ90%). An analytical sample was
isolated by TLC (SiO2, AcOEt±hexane 1:9) as a colourless gum.
[ꢀ]D = +148.9ꢀ (c = 2.6, CHCl3). MS (70 eV, EI+) calculated for
C25H27NO3 = 389; found m/z (%) = 390 (23) [M + H]+, 344 (32)
1
[M Ð C2H5O]+. IR (®lm): 1730 cm 1. H NMR (250 MHz, CDCl3,
298 K): ꢂ 7.15±7.55 (15H, m, aromatic), 5.35 (1H, d, J = 12.4 Hz,
CHHPh), 5.22 (1H, d, CHHPh), 4.07 (1H, m, H-ꢃ), 4.07 (2H, d, J =
13.2 Hz, NCHHPh), 3.5 (1H, s, OH), 3.39 (2H, d, NCHHPh), 3.11
(1H, d, J = 9.5 Hz, H-ꢀ), 1.1 (3H, d, J = 5.8 Hz, CH3). 13C NMR
(100 MHz, CDCl3): ꢂ 139.45 (C aromatic), 135.7 (C aromatic), 129.1,
128.7, 128.6, 128.5, 127.4 (CH aromatic), 67.2 (C-ꢃ), 66.3 (PhCH2O),
63.2 (C-ꢀ), 54.8 (PhCH2N), 19.2 (CH3). Elemental analysis calculated
for C25H27NO3: C 77.09, H 6.99, N 3.60%; found: C 76.91, H 6.81, N
3.67%.
Figure 2
The packing of (II), showing the arrangement of the chains.
neutron diffraction studies (Allen, 1986) in the IAM and
Ê
ELMAM re®nements, respectively (0.002 A distance ꢁ). The
ELMAM re®nement shows a good improvement in statistical
indexes when compared with the IAM re®nement; the R(F)
factor is reduced from 4.22 to 3.02% and wR(R) from 5.12 to
3.37%. The minimum and maximum peaks in the residual
electron density are 0.036 and 0.056 e A 3, respectively,
Ê
3
after the IAM re®nement and 0.028 and 0.039 e A after
For the preparation of (II), a solution of crude ester (I) (5.453 g,
14.0 mmol) in absolute ether (50 ml) was dropped over a period of
30 mm on to a magnetically stirred suspension of 95% LiAlH4 (0.78 g,
ꢂ19.5 mmol, 1.15 equivalents) in absolute tetrahydrofuran (THF,
50 ml) under argon cooled below 277 K. The mixture was stirred for
25 h at room temperature and re¯uxed for an additional hour. After
completion of the reaction (checked by SiO2 TLC; ethyl acetate±n-
hexane 1:1; RF ' 0.29), the mixture was cooled to room temperature,
quenched by slow addition of ethyl acetate (5 ml) and saturated
aqueous Na2SO4 (5 ml), and stirred overnight in air. The suspension
was ®ltered through sintered glass and the remaining salts thoroughly
washed with a dichloromethane±ethanol mixture (1:1, 100 ml). The
®ltrates were concentrated under reduced pressure and the residue
partitioned in dichloromethane±water (2 Â 50 ml), washed with
water (2Â 20 ml), dried over MgSO4, concentrated under reduced
pressure, and ®nally stored at 277 K. The resulting solids were
isolated by ®ltration and crystallized twice from ethyl acetate±
hexanes (1:19) to yield pure alcohol (II) (3.2 g, ꢂ73%) as white
crystals suitable for X-ray diffraction (m.p. 363±364 K). [ꢀ]D = +54.0ꢀ
(c = 1, CHCl3). MS (70 eV, EI+): m/z 286.1 [M + H]+. 1H NMR
(250 MHz, CDCl3/" D2O): ꢂ 7.2±7.42 (m, 10H, aromatic), 3.99 (2H, d,
J = 13.2 Hz, NCHHPh), 3.85 (1H, m, H-ꢃ), 3.8 (2H, d, J = 13.2 Hz, J =
5.8 Hz, CH2OD), 3.72 (2H, d, NCHHPh), 2.61 (1H, d, J = 8.8 Hz, H-
ꢀ), 1.15 (3H, d, J = 5.8 Hz, CH3).13C NMR (100 MHz, CDCl3): ꢂ 139.5
(C aromatic), 129.4, 128.7, 127.5 (CH aromatic), 65.5 (C-ꢃ), 64.8 (C-
ꢀ), 59.1 (PhCH2), 54.7 (CH2ÐOH), 20.4 (CH3). Elemental analysis
calculated for C18H23NO2: C 75.76, H 8.12, N 4.91%; found: C 75.79,
H 8.09, N 4.90%.
Ê
the ELMAN re®nement. The most signi®cant differences in
the geometry of compound (II) are, as expected, the bond
distances involving H atoms, which had different targets in the
ELMAN and IAM re®nements. When H atoms are not
considered, the r.m.s. deviation between the two structures is
Ê
0.0067 A, while the r.m.s. discrepancy for the bond lengths is
Ê
Ê
0.0055 A. The differences do not exceed 0.015 A for bond
distances and 0.5ꢀ for bond angles when H atoms are excluded.
The HÐCÐH angles show the strongest (2.9ꢀ) r.m.s. discre-
pancy, the XÐYÐH angles differ by 1.8ꢀ, while the XÐYÐZ
angles are less affected, with a low (0.2ꢀ) r.m.s. difference. The
s.u. values of the bond distances and angles are smaller with
the ELMAM re®nement owing to better R factor values
(Table 1).
The largest effect of the multipoles transfer on the crys-
tallographic structure is observed in the atomic thermal
motion. The r.m.s. difference over all the Uij parameters
reaches 16% between the two re®nement models. The Uij
values derived from the IAM re®nement have an r.m.s. value
10% larger on average than those derived from the ELMAM
re®nement. With the independent spherical atom model, the
displacement parameters are incorrect as they incorporate
some signi®cant deformation electron density due to improper
deconvolution between these two features (Jelsch et al., 1998).
Crystal data
3
Ê
C18H23NO2
V = 1670.4 (4) A
Z = 4
Mr = 285.37
Monoclinic, C2
a = 18.6701 (4) A
b = 6.652 (1) A
c = 16.060 (2) A
ꢃ = 123.126 (8)ꢀ
Experimental
Mo Kꢀ radiation
1
Ê
ꢄ = 0.07 mm
T = 100 (2) K
For the preparation of (2S,3S)-benzyl 2-(N,N-dibenzylamino)-3-
hydroxybutanoate, (I), benzyl bromide (6.6 ml, 55 mmol, 3.3
equivalents) was dropped over a period of 30 min on to a magneti-
Ê
Ê
0.35 Â 0.25 Â 0.20 mm
ꢃ
Acta Cryst. (2008). C64, o18±o20
Ejsmont et al. C18H23NO2 o19

organic compounds
Data collection
tion: CrysAlis RED; program(s) used to solve structure: SIR92
(Altomare et al., 1993); program(s) used to re®ne structure: MoPro
(Guillot et al., 2001; Jelsch et al., 2005); molecular graphics:
SHELXTL (Sheldrick, 1990); software used to prepare material for
publication: publCIF (Westrip, 2007).
Oxford Diffraction Xcalibur2
diffractometer
8273 measured re¯ections
2016 independent re¯ections
1966 re¯ections with I > 2ꢁ/(I)
Rint = 0.022
Re®nement
R[F2 > 2ꢁ(F2)] = 0.030
wR(F2) = 0.034
S = 1.04
2016 re¯ections
199 parameters
46 restraints
H-atom parameters constrained
This work was partially supported by the Agence Nationale
de la Recherche, programme Blanc.
3
Ê
Áꢅmax = 0.05 e A
3
Ê
0.03 e A
Áꢅmin
=
Supplementary data for this paper are available from the IUCr electronic
archives (Reference: DN3056). Services for accessing these data are
described at the back of the journal.
Table 1
Hydrogen-bond geometry (A, ).
ꢀ
Ê
DÐHÁ Á ÁA
DÐH
HÁ Á ÁA
DÁ Á ÁA
DÐHÁ Á ÁA
References
O1ÐHO1Á Á ÁN1
0.97
0.97
2.06
1.87
2.729 (3)
2.789 (2)
125
158
Allen, F. H. (1986). Acta Cryst. B42, 515±522.
Altomare, A., Cascarano, G., Giacovazzo, C. & Guagliardi, A. (1993). J. Appl.
Cryst. 26, 343±350.
Asanuma, H., Hayashi, H., Zhao, J., Liang, X., Yamazawa, A., Kuramochi, T.,
Matsunaga, D., Aiba, Y., Kashida, H. & Komiyama, M. (2006). Chem.
Commun. 48, 5062±5064.
O2ÐHO2Á Á ÁO1i
Symmetry code: (i) x; y ‡ 1; z.
Table 2
Selected geometric data for (II) (A, ).
ꢀ
Ê
È
Breuer, M., Ditrich, K., Habicher, T., Hauer, B., Kesseler, M., Sturmer, R. &
Zelinski, T. (2004). Angew. Chem. Int. Ed. 43, 788±824.
Brough, S. J. & McInally, T. (2004). Patent No. WO2003-GB400020030916,
1-24.
ELMAM is the experimental library multipolar atom model and IAM is the
independent atom model.
Flack, H. D. (1983). Acta Cryst. A39, 876±881.
Guillot, B., Viry, L., Guillot, R., Lecomte, C. & Jelsch, C. (2001). J. Appl. Cryst.
34, 214±223.
Jelsch, C., Guillot, B., Lagoutte, A. & Lecomte, C. (2005). J. Appl. Cryst. 38,
38±54.
Jelsch, C., Pichon-Pesme, V., Lecomte, C. & Aubry, A. (1998). Acta Cryst. D54,
1306±1318.
Kashida, H., Asanuma, H. & Komiyama, M. (2006). Chem. Commun. 26,
2768±2770.
Kashida, H., Tanaka, M., Baba, S., Sakamoto, T., Kawai, G., Asanuma, H. &
Komiyama, M. (2006). Chem. Eur. J. 12, 777±784.
Oxford Diffraction (2006). CrysAlis CCD and CrysAlis RED. Versions
1.171.31.7. Oxford Diffraction Ltd, Abingdon, Oxfordshire, England.
Peng, S., Zhao, M., Wang, C. & Gu, K. (2006). Chinese Patent CN1785995 A
20060614, 1-38.
Distance/angle
ELMAM
IAM
O1ÐC2
O2ÐC4
N1ÐC1
N1ÐC5
N1ÐC12
O1ÐC2ÐC1
O1ÐC2ÐC3
O2ÐC4ÐC1
C1ÐN1ÐC5
C1ÐN1ÐC12
C5ÐN1ÐC12
1.465 (2)
1.436 (2)
1.502 (2)
1.506 (1)
1.496 (2)
107.3 (2)
108.2 (2)
112.5 (3)
112.6 (3)
115.8 (3)
110.5 (3)
1.461 (3)
1.435 (3)
1.502 (3)
1.504 (1)
1.500 (4)
107.3 (4)
108.4 (4)
112.5 (5)
112.8 (5)
115.7 (4)
110.4 (5)
Pichon-Pesme, V., Jelsch, C., Guillot, B. & Lecomte, C. (2004). Acta Cryst. A60,
204±208.
Rijt, S. van, van Zutphen, S., den Dulk, H., Brouwer, J. & Reedijk, J. (2006).
Inorg. Chim. Acta, 359, 4125±4129.
Sheldrick, G. M. (1990). SHELXTL. Siemens Analytical X-ray Instruments
Inc., Madison, Wisconsin, USA.
At ®rst, a least-squares re®nement, based on |F2 |, was carried out
using the program SHELXL97 (Sheldrick, 1997), the H atoms being
constrained according to standard crystallography stereochemistry. A
least-squares re®nement, based on |F|, was then carried out with the
program MoPro (Guillot et al., 2001; Jelsch et al., 2005) using the
ELMAM approach (Zarychta et al., 2007). The re¯ection weights
were set equal to 1/ꢁ2(Fo). The Uiso(H) values were restrained to be
È
Sheldrick, G. M. (1997). SHELXL97. University of Gottingen, Germany.
Shi, Y., Machida, K., Kuzuya, A. & Komiyama, M. (2005). Bioconjug. Chem.
16, 306±311.
Villorbina, G., Canals, D., Carde, L., Grijalvo, S., Pascual, R., Rabal, O.,
Teixido, J., Fabrias, G., Llebaria, A., Casas, J. & Delgado, A. (2007). Bioorg.
Med. Chem. 15, 50±62.
Waykole, L. M., McKenna, J. J., Bach, A., Prashad, M., Repic, O. & Blacklock,
T. J. (2007). Synth. Commun. 37, 1445±1454.
Westrip, S. P. (2007). publCIF. In preparation.
Yoshida, K., Nakajima, S., Wakamatsu, T., Ban, Y. & Shibasaki, M. (1988).
Heterocycles, 27, 1167±1168.
Ê
1.2Ueq of the attached atom, with a standard deviation of 0.01 A. In
the absence of suitable anomalous scattering, the re®nement of the
Flack (1983) parameter led to inconclusive values; therefore Friedel
equivalent re¯ections were merged prior to the ®nal re®nements. The
absolute structure was set by reference to the known chirality of the
enantiomerically pure d-threonine used in the chemical synthesis.
Data collection: CrysAlis CCD (Oxford Diffraction, 2006); cell
re®nement: CrysAlis RED (Oxford Diffraction, 2006); data reduc-
Zarychta, B., Pichon-Pesme, V., Guillot, B., Lecomte, C. & Jelsch, C. (2007).
Acta Cryst. A63, 108±125.
ꢃ
o20 Ejsmont et al.
C18H23NO2
Acta Cryst. (2008). C64, o18±o20