Toward total spontaneous resolution of sec-Butylzinc complexes

Article abstract of DOI:10.1021/om9007828

Di-sec-butylzinc has been used to synthesize [Zn(s-Bu)₂L ₂] complexes, where L denotes amine ligands: e.g., pyridine (py), 2-picoline (2-pic), and 3-picoline (3-pic). The crystal structures of [Zn(sBu)₂(py)₂] (1), [Zn(s-Bu)₂(2-pic) ₂] (2), and [Zn(s-Bu)₂(3-pic)₂] (3) were determined. Reaction of di-sec-butylzinc with the bidentate amine ligands N, N, N', N' -tetramethylethylenediamine (TMEDA) and N, N, N', N' - tetraethylethylenediamine (TEEDA) afforded the chiral complexes [Zn(s-Bu) ₂(tmeda)] (4) and [Zn(s-Bu)₂(teeda)] (5); their crystal structures were determined. Since they coordinate two s-Bu groups each, complexes 1-5 may exist both as chiral and meso molecules in solution, but the crystalline complexes were found to be exclusively chiral. The enantiomerization rate of 1-5 in benzene was found to be slow on the NMR time scale. Chiral crystalline complexes containing the Zn(s-Bu)₂ moiety are of particular interest, since they display reactive chirogenic α-carbon atoms. The identification of a Zn(s-Bu)₂ complex such as 1-5, which crystallized as a conglomerate of chiral crystals, would afford a simple and cheap way to optical resolution of a chiral organometallic reagent, without the need to use enantiopure ligands. Using preferential crystallization, it should be possible to perform total spontaneous resolution, which may be regarded as a form of absolute asymmetric synthesis. However, all complexes isolated so far, including the dinuclear alkylzinc amide [Zn(s-Bu)(Me₂N(CH ₂)₂NH)]₂ (6), crystallize in centrosymmetric space groups. As a remedy, recrystallization of Zn(s-Bu)² complexes from different solvents was studied: in at least two cases, viz. [Zn(s-Bu) ₂(3-pic)₂] ? C₇H₈ (7) and [Zn(s-Bu)₂(3,5-lut)₂] ? C₇H₈ (8), it was found that recrystallization from toluene gave rise to clathrates. 2009 American Chemical Society.

Full text of DOI:10.1021/om9007828

Organometallics 2010, 29, 177–183 177
DOI: 10.1021/om9007828
Toward Total Spontaneous Resolution of sec-Butylzinc Complexes
€
Anders Lennartson,* Anna Hedstrom, and Mikael Hakansson*
Department of Chemistry, University of Gothenburg, SE-412 96 Gothenburg, Sweden
Received September 8, 2009
Di-sec-butylzinc has been used to synthesize [Zn(s-Bu)2L2] complexes, where L denotes amine
ligands: e.g., pyridine (py), 2-picoline (2-pic), and 3-picoline (3-pic). The crystal structures of [Zn(s-
Bu)2(py)2] (1), [Zn(s-Bu)2(2-pic)2] (2), and [Zn(s-Bu)2(3-pic)2] (3) were determined. Reaction of
di-sec-butylzinc with the bidentate amine ligands N,N,N0,N’-tetramethylethylenediamine (TMEDA)
and N,N,N0,N’-tetraethylethylenediamine (TEEDA) afforded the chiral complexes [Zn(s-Bu)2-
(tmeda)] (4) and [Zn(s-Bu)2(teeda)] (5); their crystal structures were determined. Since they
coordinate two s-Bu groups each, complexes 1-5 may exist both as chiral and meso molecules in
solution, but the crystalline complexes were found to be exclusively chiral. The enantiomerization
rate of 1-5 in benzene was found to be slow on the NMR time scale. Chiral crystalline complexes
containing the Zn(s-Bu)2 moiety are of particular interest, since they display reactive chirogenic
R-carbon atoms. The identification of a Zn(s-Bu)2 complex such as 1-5, which crystallized as a
conglomerate of chiral crystals, would afford a simple and cheap way to optical resolution of a chiral
organometallic reagent, without the need to use enantiopure ligands. Using preferential crystal-
lization, it should be possible to perform total spontaneous resolution, which may be regarded as a
form of absolute asymmetric synthesis. However, all complexes isolated so far, including the
dinuclear alkylzinc amide [Zn(s-Bu)(Me2N(CH2)2NH)]2 (6), crystallize in centrosymmetric space
groups. As a remedy, recrystallization of Zn(s-Bu)2 complexes from different solvents was studied: in
at least two cases, viz. [Zn(s-Bu)2(3-pic)2] C7H8 (7) and [Zn(s-Bu)2(3,5-lut)2] C7H8 (8), it was found
3
3
that recrystallization from toluene gave rise to clathrates.
Introduction
or crystallization-induced asymmetric transformation.4 We
have reported virtually complete total spontaneous resolu-
tion, for example, of a seven-coordinate samarium complex5
and more recently of five- and nine-coordinate complexes.6
Since the product is stereochemically labile, it will racemize
on redissolution, but an organometallic reagent resolved
through total spontaneous resolution may still be used in
the synthesis of stereochemically inert organic compounds,
as long as the reaction proceeds under controlled conditions:
e.g., in a solvent-free reaction. Several prochiral substrates
have been resolved by total spontaneous resolution and are
found to give rise to optically active organic products
through chemical reactions,7-11 but each system will usually
allow only one or a few similar products. Also, the products
are often highly specialized and less likely to find a practical
application. It would be more advantageous to subject a
general organometallic reagent to total spontaneous resolution.
We have reported on the use of organometallic reagents in
absolute asymmetric synthesis: i.e. the synthesis of optically
active compounds without the use of optically active starting
materials or catalysts.1,2 Our method is based on preparing
stereochemically labile organometallic reagents that crystal-
lize in a Sohncke space group3 and thus undergo sponta-
neous resolution on crystallization. Spontaneous resolution
means that the two enantiomers crystallize separately, and a
conglomerate of chiral crystals will be obtained. If the whole
batch of crystals originates from a single crystal nucleus and
fast enantiomerization in solution maintains the racemic
composition of the solute, a racemate may be converted
completely to a pure crystalline enantiomer in quantitative
yield. This may be referred to as total spontaneous resolution
*To whom correspondence should be addressed. E-mail: hson@
chem.gu.se (M.H.).
(1) (a) Feringa, B. L.; Van Delden, R. A. Angew. Chem., Int. Ed. 1999,
38, 3419. (b) Mislow, K. Collect. Czech. Chem. Commun. 2003, 68, 849.
(c) Soai, K.; Sato, I.; Shibata, T.; Komiya, S.; Hayashi, M.; Matsueda, Y.;
Imamura, H.; Hayase, T.; Morioka, H.; Tabira, H.; Yamamoto, J.; Kowata, Y.
Tetrahedron: Asymmetry 2003, 14, 185.
(2) (a) Vestergren, M.; Gustafsson, B.; Davidsson, O.; Hakansson,
M. Angew. Chem., Int. Ed. 2000, 39, 3435. (b) Vestergren, M.; Eriksson, J.;
Hakansson, M. Chem. Eur. J. 2003, 9, 4678. (c) Lennartson, A.; Olsson, S.;
Sundberg, J.; Hakansson, M. Angew. Chem., Int. Ed. 2009, 48, 3137.
(3) Flack, H. D. Helv. Chim. Acta 2003, 86, 905.
(5) Lennartson, A.; Vestergren, M.; Hakansson, M. Chem. Eur. J.
2005, 11, 1757.
(6) (a) Lennartson, A.; Hakansson, M. Angew. Chem., Int. Ed. 2009,
48, 5869. (b) Lennartson, A.; Hakansson, M. CrystEngComm 2009, 11,
1979.
(7) Penzien, K.; Schmidt, G. M. J. Angew. Chem. 1969, 81, 628.
(8) Green, B. S.; Heller, L. Science 1974, 185, 525.
(9) Gerdil, R.; Liu, H.; Bernardinelli, G. Helv. Chim. Acta 1999, 82,
418.
(10) Sakamoto, M.; Kobaru, S; Mino, T.; Fujita, T. Chem. Commun.
(4) Jacques, J.; Collet, A.; Wilen, S. H. In Enantiomers, Racememates
2004, 1002.
and Resolutions; Wiley: New York, 1984; p 369ff.
(11) Johansson, A.; Hakansson, M. Chem. Eur. J. 2005, 11, 5238.
r
2009 American Chemical Society
Published on Web 12/14/2009
pubs.acs.org/Organometallics

178 Organometallics, Vol. 29, No. 1, 2010
Lennartson et al.
Scheme 1
Scheme 2
Preferential crystallization is a useful technique on the
laboratory scale as well as for industrial applications on
the multiton scale.4
Results and Discussion
Unlike ZnMe2, ZnEt2, Zn(i-Pr)2, and Zn(n-Bu)2, Zn(s-Bu)2
is not commercially available. It can, however, be easily
prepared from inexpensive starting materials. The synthesis
of primary dialkylzinc compounds from alkyl iodides and
zinc was discovered by Frankland in 1848,14-17 and in 1929
Noller reported the synthesis of primary dialkylzinc com-
pounds in good yields with the use of different zinc-copper
couples and mixtures of alkyl iodides and bromides followed
by vacuum distillation.18 However, when secondary alkyl
groups (e.g., i-Pr and s-Bu) were used, only gaseous products
were obtained. In 1944, Soroos and Morgana modified
Noller’s procedure to allow synthesis of dialkylzinc com-
pounds from secondary alkyl halides in good yields.19 A 1:1
mixture of alkyl bromide and alkyl iodide was added slowly
under controlled-temperature conditions to an excess of
zinc-copper couple, and the product was distilled, keeping
the pressure and temperature as low as possible. The
zinc-copper couple used was turnings prepared from zinc
and a bronze brazing rod. Using Soroos and Morgana’s
procedure, but with an activated zinc-copper couple pre-
pared from zinc powder and copper sulfate,20 we were able to
obtain Zn(s-Bu)2 in 72% yield. Zn(s-Bu)2 is a colorless
pyrophoric liquid but may be stored under a nitrogen atmo-
sphere at -80 °C (where it forms a white solid) for at least 2
years. Thus, Zn(s-Bu)2 is a convenient starting material for
our purposes.
Complexes between Zn(s-Bu)2 and Pyridines. Addition of
pyridine (py), 2-picoline (2-pic), and 3-picoline (3-pic) to neat
Zn(s-Bu)2, followed by addition of hexane and cooling, gave
crystals of [Zn(s-Bu)2(py)2] (1), [Zn(s-Bu)2(2-pic)2] (2), and
[Zn(s-Bu)2(3-pic)2] (3). All three complexes are air-sensitive;
while crystals of 1 are colorless, those of 2 and 3 are yellow.
The crystal structures of 1-3 were determined by single-
crystal X-ray diffraction and demonstrate that the molecular
structures of 1-3 are similar; the coordination geometries
around the Zn atoms are distorted tetrahedral (Figure 1).
The N-Zn-N angles are significantly smaller than 109.5°:
the N-Zn-N angles in 1-3 are 87.31(6), 91.94(8), and
89.08(9)°, respectively (Tables 1 and 2). The C-Zn-C angles
We have reported on how such organometallic reagents may
be used in absolute asymmetric synthesis: octahedral
Grignard reagents displaying chirogenic magnesium atoms2a
gave chiral secondary alcohols on reaction with aldehydes,
and enantiomeric excesses (ee) up to 22% were obtained.2b
The low ee may be due to incomplete resolution of the
reagent or decomposition of the crystals during the course
of reaction, but an important factor is probably also low
enantioselectivity of the nucleophilic attack. Our solution to
this problem is to employ organometallic reagents displaying
stereochemically labile chirogenic R-carbon atoms. If such a
reagent could be resolved by total spontaneous resolution, a
reaction with an achiral substrate such as CO2 could trap the
configuration of the reactive chirogenic carbon center, with-
out the need to form a new chirogenic center (Scheme 1).
From a single reagent, a large variety of optically active
products could ideally be obtained, and such reactions could
be expected to proceed with a high degree of stereochemical
retention (or inversion). Indeed, we have found that the
reaction of such a reagent, di(1-η1-indenyl)di(3-picoline)-
zinc, with N-chlorosuccinimide (NCS) in the presence of
methanol and p-benzoquinone gave 1-chloroindene in high
yield and relatively high enantiomeric excess (89%) for
selected single crystals (Scheme 2).2c Bulk samples have given
up to 71% ee without optimization of the crystallization
conditions. During the course of that study,2c we investi-
gated several potentially useful systems, and we now present
the results concerning di-sec-butylzinc.
In comparison to other organometallic reagents bearing
chirogenic R-carbon atoms, such as organolithium reagents,
chiral secondary organozinc reagents will racemize more
slowly in solution.12 If the enantiomerization rate is too slow
at ambient temperature for total spontaneous resolution to
occur conveniently, the rate could be increased by crystal-
lizing at elevated temperatures or a catalyst could be added.
However, even if total spontaneous resolution would turn
out to be difficult, a conglomerate phase of a Zn(s-Bu)2
complex would still be attractive, since resolution through
preferential crystallization would be possible and convenient
access to enantiopure organozinc reagents should be useful.13
(14) Frankland, E. Philos. Trans. R. Soc. 1852, 142, 417.
(15) Frankland, E. Philos. Trans. R. Soc. 1855, 145, 259.
(16) Frankland, E. Experimental Researches in Pure, Applied and
Physical Chemistry; John Van Voorst: London, 1877.
(17) Seyferth, D. Organometallics 2001, 20, 2940.
(18) Noller, C. R. J. Am. Chem. Soc. 1929, 51, 594.
(19) Soroos, H.; Morgana, M. J. Am. Chem. Soc. 1944, 66, 893.
(20) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis; Wiley:
New York, 1967.
(12) (a) Guijarro, A.; Rieke, R. D. Angew. Chem., Int. Ed. 2000, 39,
1475. (b) Guijarro, A. In The Chemistry of Organozinc Compounds; Wiley:
Chichester, U.K., 2006; p 193ff.
(13) (a) Cote, A.; Charette, A. B. J. Am. Chem. Soc. 2008, 130, 2771.
(b) Lu, Z.; Chai, G.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 6045. (c) Hupe,
E.; Calaza, M.; Knochel, P. J. Organomet. Chem. 2003, 680, 136.

Article
Organometallics, Vol. 29, No. 1, 2010 179
Figure 1. Molecular structures of 1-3. Thermal ellipsoids enclose 50% probability.
Table 1. Selected Bond Distances in 1-5, 7, and 8
are larger than 109.5°; i.e., the Zn(s-Bu)2 moiety is distorted
toward the linear arrangement which is observed in Lewis-
base free dialkylzincs.21,22 The C-Zn-C angles in 1 and 3
are 134.68(8) and 133.78(12)°, respectively, while the
C-Zn-C angle in 2 is significantly smaller at 124.07(12)°.
It thus appears that steric repulsions induced by a substituent
at the 2-position of the pyridine ring forces the complex to
approach a more ideal tetrahedral coordination geometry in
2, in comparison with 1 and 3. The only aliphatic ZnR2
complex with pyridine-type ligands reported in the Cam-
bridge Structural Database (CSD)23 is [ZnMe2(dmap)2]
(dmap = 4-(dimethylamino)pyridine), which has C-Zn-C
angles of 137.2(1) and 137.7(1)° for the two independent
molecules in the asymmetric unit, the corresponding
N-Zn-N angles being 97.4(1) and 90.9(1)°, respectively.24
˚
˚
compd
Zn1-C2, Zn1-C6 (A)
Zn1-N1, Zn1-N2 (A)
1
2
3
4
5
7
8
2.0506(19), 2.041(2)
2.050(2)
2.2365(16), 2.1964(17)
2.2561(16)
2.040(3), 2.045(3)
2.0405(17)
2.046(2)
2.234(2), 2.210(3)
2.3221(13)
2.3830(16)
2.037(3), 2.035(3)
2.028(4), 2.032(4)
2.215(3), 2.216(3)
2.233(3), 2.217(3)
There seems to be no previous reports on crystal structures of
Zn(s-Bu)2 complexes, and the structural knowledge on dia-
lkylzinc complexes in the crystalline state is rather limited in
general. A search in the CSD revealed only approximately 10
structures of ZnMe2 complexes and about the same number
for ZnEt2 complexes. Only a few examples of crystal struc-
tures of higher dialkylzinc complexes have been reported. We
recently reported the first crystal structure of a Zn(i-Pr)2
complex, [Zn(i-Pr)2(tmeda)],25 and the crystal structure of
(21) Almenningen, A.; Helgaker, T. U.; Haaland, A.; Samdal, S. Acta
Chem. Scand. 1982, A36, 159.
(22) Lewinski, J.; Dranka, M.; Bury, W.; Sliwinski, W.; Justyniak, I.;
Lipkowski, J. J. Am. Chem. Soc. 2007, 129, 3096.
(23) Allen, F. H. Acta Crystallogr. 2002, B58, 380.
(24) Thomas, F.; Schulz, S.; Nieger, M. Angew. Chem., Int. Ed. 2005,
44, 5668.
(25) Lennartson, A.; Hedstroem, A.; Hakansson, M. Acta Crystallogr.
2007, E63, m123.

180 Organometallics, Vol. 29, No. 1, 2010
Lennartson et al.
Table 2. Selected Bond Angles in 1-5, 7, and 8
N-Zn-N (deg)
compd
C-Zn-C (deg)
C-Zn-N (deg)
1
2
3
4
5
7
8
134.68(8)
124.07(12)
133.78(12)
141.61(9)
141.26(11)
133.74(13)
134.17(17)
87.31(6)
91.94(8)
89.08(9)
79.42(7)
80.39(7)
87.86(10)
88.41(12)
105.46(8), 108.15(7), 107.00(7), 104.10(7)
110.27(7), 107.78(8)
109.12(11), 103.32(11), 103.56(10), 109.07(11)
106.29(6), 103.02(6)
103.54(7), 105.81(7)
103.10(12), 104.05(12), 109.15(11), 109.51(12)
104.37(15), 104.84(16), 107.62(15), 108.01(15),
tallization.5,11 Unfortunately, even though complexes 1-3 in
principle could form such intermolecular CH-π or π-π
interactions, no directed intermolecular interactions can be
found in any of the crystals.
Complexes between Zn(s-Bu)2 and Aliphatic Diamines. The
N-isopropyl-N,N0,N0-trimethylethylenediamine ligand forms a
crystalline complex with ZnEt2, which is chiral due to the
chirogenic N atoms.28 Other bidentate aliphatic amines, such
as TMEDA, should also be suitable ligands. Indeed, [Zn(s-
Bu)2(tmeda)] (4) is readily synthesized from Zn(s-Bu)2 and
TMEDA and forms colorless air-sensitive needles belonging
to space group C2/c (Figure 3, Tables 1 and 2). The coordi-
nation geometry around the central Zn atom in 4 is even
more distorted than in complexes 1-3, with a C-Zn-C
angle of 141.61(9)°. This may be due to steric influences,
since the repulsions between the s-Bu groups and the amine
appear to be smaller in 4 as compared to 1-3. The corre-
sponding C-Zn-C angles are 130.46(8)° in [Zn(i-Pr)2-
(tmeda)]25 and 135.8(3)° in [ZnMe2(tmeda)].26 Molecules of
4 are chiral, since the two s-Bu groups display the same
configuration and the conformational chirality of the pyr-
idine ligands in 1-3 is replaced by a conformational chirality
of the chelate ring. The S configuration at the s-Bu groups
corresponds to a λ conformation29 of the chelate ring system.
Further experiments using similar bidentate amines as li-
gands did not yield crystalline products, except in the case of
the TEEDA (N,N,N0,N0-tetraethylethylenediamine) ligand:
[Zn(s-Bu)2(teeda)] (5) could be isolated but was found to
crystallize in the centrosymmetric space group C2/c
(Figure 3, Tables 1 and 2). Moreover, the methyl groups
of the s-Bu ligands are disordered over two positions,
rendering the complex unsuited for total spontaneous reso-
lution.
Protic Ligands. The reaction of Zn(s-Bu)2 with N,N-
dimethylethylendiamine gave crystals of the dinuclear
complex [Zn(s-Bu)(Me2N(CH2)2NH)]2 (6) (see Scheme 3,
Figure 4, and Table 3). Only one of the two H atoms is
abstracted from the amine, and the ligand thus still retains
one H atom attached to N. Complex 6 is centrosymmetric,
and the two s-Bu groups are related by inversion symmetry
and the molecules are present in their achiral meso form.
Similar dinuclear alkylzinc amides have been reported: e.g.,
[MeZn(NMe(CH2)2NMe2)]2 and [EtZn(NMe(CH2)2NMe2)]2.30
Repeating the synthesis of 6 in the presence of pyridine did
not give rise to a mononuclear complex; instead, crystals of 6
were recovered.
Figure 2. Left-handed (a) and right-handed (b) conformations
of the pyridine ligands in 1-3.
Lewis-base free Zn(t-Bu)2 as well as the complexes formed
with 1,2-bis(4-pyridyl)ethane and 1,2-bis(4-pyridyl)ethene
were recently reported.22 The crystal structure of [Zn(neo-
pentyl)2(tmeda)] has also been reported.26
The R-carbon atoms in 1-3 are only slightly distorted
from tetrahedral coordination geometries, and the two s-Bu
groups in each complex display the same configuration. It
thus appears that Zn(s-Bu)2 has a tendency to prefer the
formation of chiral complexes with pyridine ligands, rather
than forming meso complexes; so far we have not character-
ized any meso-Zn(s-Bu)2 complex, although the presence of
diastereomers of Zn(s-Bu)2 has been detected in solution by
13C NMR spectroscopy.27 An additional sense of chirality is
generated by the orientation of the pyridine ligands, which
gives rise to a sense of helical chirality (Figure 2). A left-
handed conformation of the Zn(py)2-moiety in 1-3 corre-
sponds to an R configuration at the s-Bu groups. Although
the molecules of 1-3 are chiral, the crystals are not: the
complexes crystallize in centrosymmetric space groups. The
unit cells therefore contain both S,S and R,R molecules in
equal amounts. Crystals of 1 belong to space group P21/c
with one molecule in the asymmetric unit. Complex 2 crystal-
lizes in space group C2/c. The molecules in 2 are located on
2-fold axes, and the asymmetric unit will therefore consist of
half a molecule, making the two s-Bu groups crystallogra-
phically equivalent. Complex 3 crystallizes in P1. We have
previously noted that weak intermolecular interactions could
be instrumental in transferring stereochemical information,
which should increase the probability of conglomerate crys-
Enantiomerization in Solution. Inversion at secondary
alkyl carbons bonded to zinc has been reported to be slow
on the NMR time scale.12 Our NMR data are consistent with
(28) Johansson, A.; Wingstrand, E.; Hakansson, M. J. Organomet.
Chem. 2005, 690, 3846.
(29) Jensen, K. A. Inorg. Chem. 1970, 9, 1.
(30) Malik, M. A.; O’Brien, P.; Motevalli, M.; Jones, A. C. Inorg.
Chem. 1997, 36, 5076.
(26) O’Brien, P.; Hursthouse, M. B.; Motevalli, M.; Walsh, J. R.;
Jones, A. C. J. Organomet. Chem. 1993, 449, 1.
(27) Mynott, R.; Gabor, B.; Lehmkuhl, H.; Doering, I. Angew.
Chem., Int. Ed. Engl. 1985, 24, 335.

Article
Organometallics, Vol. 29, No. 1, 2010 181
Figure 3. Molecular structures of 4 and 5. The methyl groups of the s-Bu ligands in 5 are disordered over two sites (C4a and C4b).
Thermal ellipsoids enclose 50% probability.
Scheme 4
this: the two methylene hydrogen signals from the s-Bu
Figure 4. Molecular structure of [Zn(s-Bu)(Me2N(CH2)2NH)]2
groups in 1-6 show no dynamic features at ambient tem-
perature, indicating slow enantiomerization in benzene. We
could not find any meso complexes in the solid state, but it is
likely that they exist in solution.12,27 That we still obtain only
R,R and S,S complexes in high yield can be explained by
(6). Thermal ellipsoids enclose 50% probability.
Table 3. Selected Bond Distances and Bond Angles in 6a
˚
Bond Distances (A)
exchange of s-Bu groups from the meso form via a dimeric
complex,12 as indicated in Scheme 4.
C(2)-Zn(1)
N(1)-Zn(1)
2.036(5)
2.199(4)
N(2)-Zn(1)
Zn(1)-N(2)i
2.064(4)
2.052(4)
Such an exchange process could possibly also be catalyzed
on the surface of the growing crystals. In other words,
inversion at the R-carbon is not necessary for crystallization
of exclusively chiral molecules from a solution containing
meso complexes. However, for total spontaneous resolution
to be viable, racemization on the macroscopic time scale
must occur during crystallization; if elevated temperature is
not sufficient, a suitable catalyst can be utilized.
Bond Angles (deg)
C(2)-Zn(1)-N(2)i
C(2)-Zn(1)-N(2)
N(2)i -Zn(1)-N(2)
128.02(19)
123.1(2)
91.01(14)
C(2)-Zn(1)-N(1)
N(2)i -Zn(1)-N(1)
N(2)-Zn(1)-N(1)
118.45(18)
101.74(15)
83.59(14)
a Symmetry code: (i) -x þ 1, -y, -z þ 1.
Scheme 3
Formation of Clathrates with Toluene. Experiments were
conducted using other solvents such as THF and toluene in
order to either improve crystal quality or obtain phases with
cocrystallized solvent, which would give new chances for
spontaneous resolution to occur. When compound 3 was
recrystallized from toluene, crystals of a new phase, [Zn(s-
Bu)2(3-pic)2] C7H8 (7), were indeed obtained. As expected,
3
the crystal structure exhibits [Zn(s-Bu)2(3-pic)2] complexes

182 Organometallics, Vol. 29, No. 1, 2010
Lennartson et al.
(Prolabo), 2-picoline (Fluka), 3-picoline (Fluka), 3,5-lutidine
(E. Merck), N,N-dimethylaminoethylamine (Fluka), N,N,N0,
N0-tetramethylethylenediamine (Aldrich), N,N,N0,N0-tetraethy-
lethylenediamine (Aldrich), N,N,N0,N0-tetramethylpropylene-
diamine (Aldrich), 3-(N,N-dimethylamino)-propan-1-ol (Aldrich),
and N,N,N0-trimethylethylenediamine (Acros Organics) were
distilled from CaH2 prior to use. Complexes 1-6 are not suitable
for elemental analyses, since they are highly sensitive when
isolated from the mother liquor. NMR spectra were recorded
using a Varian Unity 400 MHz spectrometer.
Zinc-Copper Couple. Zinc-copper couple was prepared
using published procedures,20 with minor modifications. Zinc
powder (49.2 g, 0.75 mol) was stirred for 1 min in 3% hydro-
chloric acid (2 ꢀ 40 mL). The zinc powder was then washed with
water (5 ꢀ 40 mL), 2% aqueous copper sulfate solution (2 ꢀ 75 mL),
water (5 ꢀ 40 mL), absolute ethanol (4 ꢀ 30 mL), and diethyl
ether (5 ꢀ 30 mL). The zinc-copper couple was dried in vacuo
and stored under nitrogen. Yield: 44.1 g.
Zn(s-Bu)2. Di-sec-butylzinc was prepared according to pub-
lished procedures,19 with minor modifications. A 100 mL
Schlenk flask was equipped with a pressure-equalizing addition
funnel and a reflux condenser and charged with zinc-copper
couple (14 g, ca. 0.2 mol). A mixture of 2-iodobutane (6.2 mL,
0.054 mol) and 2-bromobutane (5.9 mL, 0.054 mol) was placed
in the addition funnel, and a small amount of halide mixture
(ca. 2 mL) was added to the zinc-copper couple at 50 °C with
stirring. The temperature was raised to 95 °C, and within 5 min
white fumes appeared. The hot oil bath was replaced by an oil
bath of ambient temperature, and the halide mixture was
added dropwise over 3 h. The reaction mixture was stirred
overnight at ambient temperature, which resulted in a clear,
almost colorless liquid and excess black zinc-copper couple.
The mixture was first evaporated to remove any butyl halide
and then carefully distilled at 10-3 mbar, condensing the
product at -196 °C. The temperature was maintained at
100 °C until distillation was almost complete and then raised
to 140 °C to recover the last trace of product. Yield: 6.9 g
(72%). 1H NMR (400 MHz, C6D6): δ 1.64 (m, 2H, CH2), 1.50
(m, 2H, CH2), 1.20 (d, 6H, CHCH3), 1.00 (t, 6H, CH2CH3),
0.72 (m, 2H, CH).
[Zn(s-Bu)2(py)2] (1). Pyridine (0.40 mL, 4.9 mmol) was added
to neat Zn(s-Bu)2 (0.40 mL, 2.4 mmol) at ambient temperature
and the mixture heated gently. The mixture was evaporated in
vacuo to remove any excess of Zn(s-Bu)2 or pyridine, diluted
with hexane (3.0 mL), and cooled to -80 °C. After a few hours,
colorless, irregular crystals had formed. Yield: 0.77 g (82%). 1H
NMR (400 MHz, C6D6): δ 8.31 (m, 4H, o-CH), 6.93 (m, 2H, p-
CH), 6.59 (m, 4H, m-CH), 1.98 (m, 2H, CH2), 1.89 (m, 2H,
CH2), 1.50 (d, 6H, CHCH3), 1.28 (m, 6H, CH2CH3), 1.00 (m,
2H, CH).
[Zn(s-Bu)2(2-pic)2] (2). The synthesis was analogous to that of
1, using 2-picoline (0.48 mL, 4.9 mmol) and neat Zn(s-Bu)2 (0.40mL,
2.4 mmol). Yellow crystals formed overnight at -80 °C. Yield: 0.77 g
(81%). 1H NMR (400 MHz, C6D6): δ 8.14 (m, 2H, o-CH), 6.49 (m,
2H, p-CH), 6.90 (m, 4H, m-CH), 2.27 (s, 6H, CCH3), 1.87 (m, 2H,
CH2CH),1.76(m,2H,CH2CH),1.36(m,6H,CHCH3), 1.24 (m, 6H,
CH2CH3), 0.89 (m, 2H, CH).
[Zn(s-Bu)2(3-pic)2] (3). The synthesis was analogous to that
of 1 and 2, using 3-picoline (0.60 mL, 6.1 mmol) and neat
Zn(s-Bu)2 (0.40 mL, 2.4 mmol). Yellow crystals formed over-
night at -80 °C. Yield: 0.49 g (55%). 1H NMR (400 MHz,
C6D6): δ 8.22 (s, 2H, o-CH), 8.18 (d, 2H, o-CH), 6.75 (m, 2H,
p-CH), 6.55 (dd, 2H, m-CH), 1.98 (m, 2H, CH2), 1.87 (m, 2H,
CH2), 1.69 (s, 6H, CCH3), 1.51 (d, 6H, CHCH3), 1.28 (dt, 6H,
CH2CH3), 1.02 (m, 2H, CH).
Figure 5. Disordered toluene molecules cocrystallize with
[Zn(s-Bu)2(3-pic)2] molecules in 7, forming a clathrate.
with disordered toluene molecules in the lattice (Figure 5).
Unfortunately, the crystals are not chiral. A second clathrate
was obtained when a solution of Zn(s-Bu)2 and 3,5-lutidine
(3,5-lut) in toluene was cooled to -80 °C: yellow monoclinic
crystals of [Zn(s-Bu)2(3,5-lut)2] C7H8 (8) were obtained.
3
Again, the crystal structure is centrosymmetric and the
lattice consists of [Zn(s-Bu)2(3,5-lut)2] and toluene mole-
cules. Clathrates 7 and 8 are composed of chiral molecules,
as in 1-5, and not of the corresponding meso forms. It is
clear that this willingness to form clathrates points out
a potential route to spontaneous resolution. Moreover,
cocrystallization of solvent molecules is interesting with
regard to formation of porous materials and in-crystal
selective reactions.
Conclusions
In search for new conglomerates to use in absolute asym-
metric synthesis, we have prepared complexes between di-
sec-butylzinc and amine ligands. In solution, meso com-
plexes probably coexist with chiral R,R and S,S complexes,
although enantiomerization is slow on the NMR time scale
at ambient temperature. Despite this, the crystal structures
of all Zn(s-Bu)2 complexes exhibited exclusively chiral mo-
lecules. However, no chiral crystals could be found. A
preliminary study of the influence of different solvents on
the crystallization of Zn(s-Bu)2 complexes revealed at least
two cases where toluene gave rise to clathrates; this indicates
a possible future route to chiral crystals.
Experimental Section
General Considerations. All experiments were performed
under a nitrogen atmosphere using standard Schlenk techniques.
All glass equipment was dried at 130 °C overnight. Toluene was
distilled from sodium/benzophenone shortly prior to use; hex-
ane was distilled from sodium/benzophenone/tetraglyme
[Zn(s-Bu)2(tmeda)] (4). The synthesis was analogous to that of
1-3, using N,N,N0,N0-tetramethylethylenediamine (0.39 mL,
2.6 mmol) and neat Zn(s-Bu)2 (0.44 mL, 2.6 mmol). Colorless
€
shortly prior to use. Zinc powder (Riedel-de Haen), copper(II)
1
sulfate pentahydrate (KEBO), 2-iodobutane (Aldrich), and
2-bromobutane (Fluka) were used as received. Pyridine
crystals formed overnight at -80 °C. Yield: 0.77 g (95%). H
NMR (400 MHz, C6D6): δ 2.00 (dq, 4H, CH2CH), 1.91 (s, 12H,

Article
Organometallics, Vol. 29, No. 1, 2010 183
Table 4. Crystal and Refinement Data for 1-8
param
1
2
3
4
5
6
7
8
M
formula
337.79
365.85
365.85
295.80
349.89
419.26
449.92
486.03
ZnC18H28N2 ZnC20H32N2 ZnC20H32N2 ZnC14H34N2 ZnC18H40N2 Zn2C16H40N4 ZnC27H32N2 ZnC29H44N2
space group
˚
˚
b/A
˚
c/A
P21/c
C2/c
P1
C2/c
C2/c
P21/n
P1
P21/c
a/A
9.3518(14)
13.822(2)
14.401(2
90
92.670(6)
90
1859.4(5)
4
1.207
15.794(4)
7.8917(13)
17.541(4)
90
115.96
90
1965.7(7)
4
1.236
10.216(2)
10.645(5)
11.735(5)
75.73(3)
65.33(2)
65.33(2)
1049.7(7)
2
16.865(5)
9.1648(16)
13.088(3
90
124.06(2)
90
1675.9(7)
4
1.172
18.546(5)
9.3694(19)
13.010(4)
90
114.528(7)
90
2056.7(9)
4
1.130
11.998(4)
7.218(3)
12.460(4)
90
99.112(11)
90
1065.4(6)
2
1.307
10.787(7)
11.093(8)
11.889(2)
71.019(12)
72.879(13)
85.313(16)
1285.5(13)
2
13.4621(18)
12.1885(15)
17.359(2)
90
96.490(5)
90
2830.0(6)
4
1.141
0.885
100(2)
R/deg
β/deg
γ/deg
3
˚
V/A
Z
Dc/g cm-3
μ(Mo KR)/mm-1
T/K
1.157
1.172
100(2)
1.162
0.969
100(2)
1.317
1.251
1.451
1.192
2.254
100(2)
0.4 ꢀ
100(2)
0.4 ꢀ
100(2)
0.4 ꢀ
100(2)
0.4 ꢀ
100(2)
0.2 ꢀ
cryst size/mm
0.3 ꢀ
0.4 ꢀ
0.3 ꢀ
0.2 ꢀ 0.1
2.04-25.50
11 833
0.4 ꢀ 0.1
3.47-25.50
6052
0.2 ꢀ 0.2
1.92-25.00
6359
0.2 ꢀ 0.2
4.45-24.99
4612
0.2 ꢀ 0.1
4.23-25.49
6419
0.1 ꢀ 0.1
3.57-24.99
5901
0.2 ꢀ 0.2
1.89-25.00
7844
0.3 ꢀ 0.1
3.55-25.00
17 077
θ range/deg
no. of rflns collected
no. of unique rflns (Rint
no. of params
)
3364 (0.0339) 1727 (0.0355) 3280 (0.0482) 1258 (0.0273) 1879 (0.0346) 1714 (0.0787) 4014 (0.564) 4552 (0.0308)
254
1.089
190
1.085
108
1.072
214
1.039
78
1.036
113
1.093
104
0.965
298
1.078
GOF on F2
final R1(F)a (I > 2σ(I))/ 0.0309/0.0778 0.0310/0.0792 0.0414/0.1074 0.0229/0.0589 0.0295/0.0711 0.0496/0.1204 0.0498/0.1308 0.0634/0.161
wR2(F2)b
R1a/wR2(F2)b (all data) 0.0349/0.0794 0.0321/0.0796 0.0477/0.1128 0.0236/0.0591 0.0334/0.0723 0.0599/0.1263 0.0512/0.1316 0.0681/0.165
P
P
P
P
a R1(F) = (||Fo| - |Fc||)/ |Fo|. b wR2(F2) = { [w(Fo2 - Fc2)2]/ [w(Fo2)2]}1/2
.
NCH3), 1.72 (s, 4H, NCH2), 1.60 (dd, 6H, CHCH3), 1.33
(dt, 6H, CH2CH3), 0.52 (m, 2H, CH).
analysis, in order to establish whether any of the complexes
displayed concomitant polymorphism.2c,31,32 Crystals of 1-8
were selected using a microscope at low temperature.33 Data
were recorded at 100 K using a Rigaku R-AXIS IIc area detector
[Zn(s-Bu)2(teeda)] (5). The synthesis was analogous to that of
1-4, using N,N,N0,N0-tetraethylethylenediamine (0.53 mL,
2,5 mmol) and neat Zn(s-Bu)2 (0.40 mL, 2.4 mmol). Colorless,
opaque crystals were formed after a few days at -80 °C. H
˚
with graphite-monochromated Mo KR radiation (λ = 0.710 73 A)
1
from a Rigaku RU-H3R rotating anode, operating at 50 kV and
90 mA. A total of 90 oscillation photos with a rotation angle of
2° was collected. Data were processed using the CrystalClear soft-
ware package,34 and an empirical absorption correction was applied
using the REQAB program. The structures were solved using the
program SIR-9235 and refined using full-matrix least-squares calcu-
lations on F2 with the SHELXL-97 program.36 All non-hydrogen
atoms were refined anisotropically, and hydrogen atoms were
included in calculated positions and refined using a riding model.
Structures were drawn using PLUTON37 and ORTEP338 for
Windows. SIR-92, SHELXL-97, and ORTEP3 were enclosed in
the WinGX software package.39 Crystal and refinement data are
given in Table 4.
NMR (400 MHz, C6D6): δ 2.46 (d, 4H, NCH2CH2), 2.46 (q, 8H,
NCH2CH3), 1.72 (m, 4H, CH2CH), 1.35 (d, 6H, CHCH3),
1.14 (t, 6H, CH2CH3), 0.94 (t, 12H, NCH2CH3), 0.68 (m, 2H,
CH).
[Zn(s-Bu)(Me2N(CH2)2NH)]2 (6). N,N-Dimethylethylendia-
mine (0.27 mL, 2.5 mmol) was added dropwise to Zn(s-Bu)2
(0.40 mL, 2.5 mmol) in hexane (1.0 mL). After a few minutes at
ambient temperature, the clear colorless solution was cooled to
-30 °C, which resulted in small white needles after a few hours.
Yield: 0.42 g (51%). 1H NMR (400 MHz, C6D6): δ 2.95 (m, 4H,
NHCH2), 2.19 (m, 4H, CH2N(CH3)2), 2.05 (m, 4H, CH2CH3),
1.89 (s, 12H, NCH3), 1.67 (dd, 6H, CHCH3), 1.31 (dt, 6H,
CH2CH3), 0.75 (m, 2H, CH).
[Zn(s-Bu)2(3-pic)2] C7H8 (7). Crystalline 3 was dissolved in
3
Acknowledgment. Financial support from the Swedish
Research Council (VR) is gratefully acknowledged.
toluene at ambient temperature followed by cooling to -80 °C.
Yellow crystals of 7 were deposited after a few hours in good yield.
[Zn(s-Bu)2(3,5-lut)2] C7H8 (8). 3,5-Lutidine (0.58 mL,
3
5.0 mmol) was added to neat Zn(s-Bu)2 (0.40 mL, 2.4 mmol)
at ambient temperature and heated gently. The mixture was
evaporated in vacuo, and toluene (0.7 mL) was added. Yellow
crystals of 8 formed in high yield overnight at -80 °C.
Supporting Information Available: CIF files giving crystal-
lographic data for 1-8 and figures giving 1H NMR spectra for
compounds 1-6. This material is available free of charge via the
Internet at http://pubs.acs.org.
X-ray Crystallography. All bulk samples were investigated by
selecting several crystals from each sample for X-ray diffraction
(34) CrystalClear 1.3, Rigaku Corp., Tokyo, Japan, 1998-2001.
(35) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Gualardi, A.
J. Appl. Crystallogr. 1993, 26, 343.
(36) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(37) Spek, A. L. Acta Crystallogr. 2009, D65, 148.
(38) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(39) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837.
(31) Bernstein, J.; Davey, R. J.; Henck, J.-O. Angew. Chem., Int. Ed.
1999, 38, 3441.
(32) Lennartson, A; Wiklund, T.; Hakansson, M. CrystEngComm
2007, 9, 856.
(33) Hakansson, M. Inorg. Synth. 1998, 32, 222.