2-Dihydromethylpiperazinediium-MII (MII = Cu II, FeII, CoII, ZnII) double sulfates and their catalytic activity in diastereoselective nitroaldol (Henry) reaction

Article abstract of DOI:10.1039/c2dt31300f

A series of double dihydromethylpiperazinediium metallic sulfates 1-7 [H₂mpz]M^II(SO₄)₂·nH ₂O (mpz = 2-methylpiperazine, C₅H₁₂N ₂) are prepared by slow evaporation using a racemic (R/S)-mpz (for 1, 2) or enantiopure R-mpz (for 3), S-mpz (for 4-7) and sulfates of Cu ^II (for 5), Fe^II (for 1, 4), Co^II (for 7) and Zn^II (for 2, 3, 6), and characterized by infrared spectroscopy, elemental analysis and single crystal X-ray diffraction. The [M ^II(H₂O)₆]²⁺, [(R/S)-H ₂mpz]²⁺, [(R)-H₂mpz]²⁺ or [(S)-H₂mpz]²⁺ cations and 2SO₄^2- anions are linked together via two types of hydrogen bonds, Ow-Hw...O and N-H...O, leading to supramolecular arrangements. The use of the racemic 2-mpz provides alternatives in crystallization: a centrosymmetric structure where the enantiomers are related by an appropriate crystallographic symmetry operation or one where the enantiomers occupy the same crystallographic position, generating disorder. Compounds 1-7 act as diastereoselective catalysts for the nitroaldol (Henry) reaction. The diastereoselectivity can be regulated from exclusive threo to exclusive erythro isomer preparation with typical yields of 50-99%, depending on the catalyst and the substrate used.

Full text of DOI:10.1039/c2dt31300f

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
2-Dihydromethylpiperazinediium-MII (MII = CuII, FeII,
CoII, ZnII) double sulfates and their catalytic activity in
diastereoselective nitroaldol (Henry) reaction
Cite this: Dalton Trans., 2013, 42, 399
Houcine Naïli,*a Fadhel Hajlaoui,a Tahar Mhiri,a Tatiana C. O. Mac Leod,b
Maximilian N. Kopylovich,b Kamran T. Mahmudovb and Armando J. L. Pombeiro*b
A series of double dihydromethylpiperazinediium metallic sulfates 1–7 [H2mpz]MII(SO4)2·nH2O (mpz =
2-methylpiperazine, C5H12N2) are prepared by slow evaporation using a racemic (R/S)-mpz (for 1, 2) or
enantiopure R-mpz (for 3), S-mpz (for 4–7) and sulfates of CuII (for 5), FeII (for 1, 4), CoII (for 7) and ZnII
(for 2, 3, 6), and characterized by infrared spectroscopy, elemental analysis and single crystal X-ray diffrac-
tion. The [MII(H2O)6]2+, [(R/S)-H2mpz]2+, [(R)-H2mpz]2+ or [(S)-H2mpz]2+ cations and 2SO4 anions are
2−
linked together via two types of hydrogen bonds, Ow–Hw⋯O and N–H⋯O, leading to supramolecular
arrangements. The use of the racemic 2-mpz provides alternatives in crystallization: a centrosymmetric
structure where the enantiomers are related by an appropriate crystallographic symmetry operation or
one where the enantiomers occupy the same crystallographic position, generating disorder. Compounds
1–7 act as diastereoselective catalysts for the nitroaldol (Henry) reaction. The diastereoselectivity can be
regulated from exclusive threo to exclusive erythro isomer preparation with typical yields of 50–99%,
depending on the catalyst and the substrate used.
Received 11th April 2012,
Accepted 17th September 2012
DOI: 10.1039/c2dt31300f
www.rsc.org/dalton
cations and a sole anion, e.g. sulfate, is usually out of atten-
tion. Hence, it would be attractive to prepare and structurally
Introduction
The chemistry of organically templated metal sulfates is a characterize such mixed salts and demonstrate their stereo
rapidly expanding research field due to their structural diversi- (diastereo) activity in a useful reaction.
ties and potential applications as functional materials.1–3 For
The Henry or nitroaldol reaction (combination of nitroalk-
instance, the double sulfates of divalent metal and chiral ane and aldehyde, a very important synthetic tool for the crea-
organic amines exhibit interesting ferroelectric or non-linear tion of a carbon–carbon bond and up to two stereogenic
optical properties such as frequency doubling or second har- centers) can be used as a model reaction to check the catalytic
monic generation.4,5 The use of optically pure organic com- activity.11–14 Usually this reaction is carried out in the presence
pounds as catalysts in stereoselective organic transformations of strong bases, leading to dehydration with concomitant for-
is another currently growing area of research.6–10 Recently a lot mation of a nitroolefin. The interest in asymmetric versions of
of publications appeared where enantiopure organics are com- this reaction started growing after the work15,16 on the use of
bined with metal ions in complex catalysts combining stereo- chiral bimetallic lithium–lanthanum catalysts, whereby the
specificity of the former with the high activity of the latter. In control of stereochemistry of two newly generated carbon
particular, several chiral metal–organic frameworks (MOFs) centres has become possible. In the last two decades many
were found to be highly regio- and stereo-selective for a other asymmetric catalysts have been developed and nowadays
number of useful chemical transformations.7 Usually the pre- aldehydes or α-keto esters can be converted to the corres-
paration of such MOFs constitutes a rather complicated task, ponding nitroalkanol with good enantio- and diastereo-
while the application of the easy to prepare MII-complexes, selectivity.17–19 However, to our knowledge, salts containing
composed of organic (e.g. aminium) and inorganic (metal ion) jointly a metal and an asymmetric organic cation (e.g.
2-H2mpz2+) have not yet been applied as catalysts for this
reaction.
aLaboratoire de l’Etat Solide, Département de Chimie, Université de Sfax, BP 1171,
Thus, the main aims of the current work are as follows:
3000 Sfax, Tunisie. E-mail: houcine_naili@yahoo.com
(i) to synthesize and structurally characterize the salts or the
MII-complexes (MII = Fe, Co, Cu and Zn) containing either
racemic or chiral 2-H2mpz and different doubly charged metal
bCentro de Química Estrutural, Complexo I, Instituto Superior Técnico, Technical
University of Lisbon, Av. Rovisco Pais, 1049-001 Lisbon, Portugal.
E-mail: pombeiro@ist.utl.pt
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 399–406 | 399

View Article Online
Paper
Dalton Transactions
cations; (ii) to study the catalytic activity and diastereoselectiv- in the stability of the crystal structures by linking the organic
ity of the synthesized salts in the nitroaldol combination of and inorganic cations via Ow–H⋯O and N–H⋯O hydrogen
nitroethane with various aldehydes.
bonds. Within the intermolecular bonds, the N⋯O distances
vary between 2.767(3) and 2.866(3) Å in 1 and between 2.664(2)
and 2.910(3) Å in 2. The Ow⋯O distances range from 2.694(3)
to 2.810(2) Å and between 2.560(4) and 2.841(5) Å in 1 and 2,
respectively.
Results and discussion
Synthesis and characterization of the salts
Description of X-ray crystal structures of compounds 3–7
Compounds [(R/S)-C5H14N2][Fe(H2O)6](SO4)2 (1), [(R/S)-C5H14N2]-
[Zn(H2O)6](SO4)2 (2), [(R)-C5H14N2][Zn(H2O)6](SO4)2 (3), [(S)-
C5H14N2][Fe(H2O)6](SO4)2 (4), [(S)-C5H14N2][Cu(H2O)6](SO4)2
(5), [(S)-C5H14N2][Zn(H2O)6](SO4)2 (6), [(S)-C5H14N2][Co(H2O)6]-
(SO4)2 (7) (see Fig. 1 and Table 1) were obtained by slow
evaporation of water solutions containing metal(II) sulfates,
sulfuric acid and either racemic or chiral cyclic diamine
2-methylpiperazine.20–23 All the structures of 1–7 are related:
the asymmetric units in each compound consist of organic
cations, SO4 tetrahedra and one MII cation octahedrally coordi-
nated to six oxygen atoms, all being held together through
extensive hydrogen-bonding networks. However, the sym-
metries of the synthesized salts differ greatly. Compounds 1–2,
which were synthesized using racemic 2-mpz, crystallize in the
centrosymmetric space groups P21/n and P21/c. In contrast,
salts 3–7, prepared from enantiopure 2-mpz, crystallize in the
noncentrosymmetric polar space group P21.
Both structures and compositions are similar in salts 3–7
(Fig. 1). Each compound 3, 4, 6 and 7 contains a single unique
divalent metal (MII = FeII, CoII or ZnII), a chiral organic H2mpz
cation and two sulfate tetrahedra, all of which lie in general
positions, while compound 5 consists of isolated tetraaquabis
(sulfato-O) copper anions [Cu(SO4)2(H2O)4]2−, organic cations
[(S)-H2mpz]2+ and two free water molecules. The crystal struc-
tures belong to the polar, noncentrosymmetric space group
P21, crystal class 2 (C2) for which the only symmetry elements
are a series of 21 screw axes. The Flack parameters refined to
−0.028(9), 0.020(9), 0.001(2), −0.015(11) and −0.013(13) for 3,
4, 5, 6 and 7, respectively.
The possibility of pseudosymmetry with the inorganic com-
ponents of 3–7 was investigated. The organic cations were
removed from the crystallographic models, and PLATON30 was
used to probe for missing symmetry. The ADDSYM command
suggests the possibility of a missing additional symmetry
leading to the P21/c space group, with inversion symmetry
Description of X-ray crystal structures of compounds 1 and 2
1 and 2 (Fig. 1) are built from [MII(H2O)6]2+, (SO4)2− and dis- directly between the metal centers.
ordered [H2mpz]2+ ions linked together by an extensive three-
The position of the methyl groups on the 2-methylpiperazi-
dimensional H-bonding network similar to the related salts of nediium cation rings violates the pseudosymmetry found by
cobalt, nickel and manganese.21 The metal cations are located PLATON, does not recognize the contribution of a single CH3
in special positions on crystallographic inversion centers and group over the rest of the structures. In addition, the systema-
each is coordinated by six oxygen atoms from water molecules tic absences are not correct for P21/c. Specifically, the expected
of which three are crystallographically independent. The l = 2n + 1 absences in h0l reflections are not present. These
2-dihydromethylpiperazinediium cations are located about results confirm the occurrence of a noncentrosymmetric space
crystallographic inversion centers, with all atoms located in group and agree with the chemical pathway in which enantio-
general positions. The Fe(H2O)6 and Zn(H2O)6 octahedra are meric molecules have been used in compounds 3–7.
slightly distorted. Bond angles O–Fe–O fall in the range
The molecular arrangements are described by an alterna-
[86.59(7)–93.41(7)°], while the O–Zn–O angles range from tion of cationic and anionic layers. The three-dimensional
86.15(6)° to 93.85(3)°. The intermolecular MII⋯MII distance in packing of 3–7 shows similar polyhedra with slight variation of
1 is considerably longer than that found in 2 (7.083(2) Å and them and the organic cations. The resulting H-bonding net-
6.599(2) Å, respectively), possibly attributed to the radii of works can be alternatively described by three-dimensional
metallic cations and the nature of disorder in the organic supramolecular frameworks belonging to three different struc-
groups of 1 by comparison with 2. Moreover, two disorder ture types, forming channels in which the R or S-2-H2mpz
mechanisms are observed in the [H2mpz]2+ cations with two cations play a templating role. However, despite the simi-
orientations of both [(R)-H2mpz]2+ and [(S)-H2mpz]2+ enantio- larities in compositions in compounds 3–7, distinct differ-
mers. In compound 1 the organic cations are partly disordered ences in three-dimensional packing are observed between the
(disorder in the carbon of methyl groups with the occupancy ZnII (3 and 6), FeII (4), CuII (5) and CoII (7).20,22,23
factor equal to 0.5), while in 2 the C and N atoms are distribu-
Compounds 3, 4, 6 and 7 are all constructed from octahed-
ted between two positions related by the symmetry centre with rally coordinated transition metal cations coordinated by six
a refined site occupancy factor equal to 0.5. The disordered water molecules [M(H2O)6]2+, isolated sulfate anions (SO4)2−
,
[H2mpz]2+ cations reside between the inorganic layers, balan- and either [(R)-H2mpz]2+ or [(S)-H2mpz]2+ cations. The octa-
cing charge and participating in the extensive hydrogen- hedra environments of MII (MII = FeII, CoII and ZnII) are not
bonding network.
regular, as seen in other isostructural metal sulfates.2–4 O–MII–
The sulfate anions are stacked such that they form anionic O bond angles range by the transition metal with averages of
layers parallel to the cationic ones. They play an important role 87.05(8)°, 85.88(5)°, 86.87(1)° and 88.84(8)° observed for 3, 4, 6
400 | Dalton Trans., 2013, 42, 399–406
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 1 Ellipsoid plots with an atom labeling scheme for 1–7 (ellipsoids are drawn at the 50% probability level).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 399–406 | 401

View Article Online
Paper
Dalton Transactions
Table 1 List of compounds studied in this work
Space CCDC
Code group number References
Compound
[(R/S)-C5H14N2][Fe(H2O)6](SO4)2
[(R/S)-C5H14N2][Zn(H2O)6](SO4)2
[(R)-C5H14N2][Zn(H2O)6](SO4)2
[(S)-C5H14N2][Fe(H2O)6](SO4)2
[(S)-C5H14N2][Cu(H2O)6](SO4)2
[(S)-C5H14N2][Zn(H2O)6](SO4)2
[(S)-C5H14N2][Co(H2O)6](SO4)2
1
2
3
4
5
6
7
P21/c 828001 21
P21/n 859890 22
P21
P21
P21
P21
P21
859888 22
823975 20
724064 23
859889 22
823974 20
Scheme 1 Formation of nitroaldols by the Henry reaction.
product formation was observed in the absence of the catalysts
as well as in the presence of simple metal sulfates (Table 2,
entries 14–19), while the application of 1–7 allows us to obtain
mixtures of β-nitroalkanol diastereoisomers (threo and erythro
forms) in high yields up to 98% and high enantiomeric
excesses (up to 85% ee). 1 appears to be more active than the
other tested compounds (Table 2, entry 1), but it is not very
selective, while 3–7 demonstrate excellent diastereoselectivity
towards the erythro or the threo isomer (Table 2, entries 3–7).
The possible reasons for the different diastereoselectivity
being influenced by (i) enantiomeric structures of the com-
plexes R- (for 3) and S- (for 4–7); (ii) hydrogen bonding inter-
actions between the protonated imino groups in the mpz
(which can act as hydrogen bond donors) and the developing
O in the transition state (which can accept a hydrogen bond).
Such interactions could, if well organized, favor the formation
of a given product via transition state stabilization. Next we
have chosen 1 (due to its high activity), 3 (due to the erythro
selectivity) and 5 (due to the threo selectivity) for further study.
As a preliminary solvent screening demonstrates, methanol
is a better solvent than benzaldehyde or nitroethane, presum-
ably due to a tighter ion pairing in the course of association
with H-shift.14 We have also examined the effect of other sol-
vents (entries 8–13, Table 2). In the case of 1 the solvent
polarity influences the selectivity, while for 3 and 5 no effect
on the selectivity is observed. Generally, the yield depends on
solvent polarity and low yields are obtained in less polar sol-
vents, possibly due to low solubility of the catalysts.14 Control
reactions were carried out in the absence of catalysts 1–7 but
in the presence of ZnCl2, Zn(SO4)2, Co(SO4)2, Fe(SO4)2 or
Cu(SO4)2 (entries 14–19, Table 2). There is no reaction between
benzaldehyde and nitroethane in the absence of the metal
complex, even in the presence of metal salts (entries 14–19,
Table 2). Next the increase of the catalyst amount enhances
the product yield from 96 to 98% (1), from 5 to 11% (3) and
from 17 to 51% (5) for the respective amounts of 1 mol% and
3 mol% of these catalysts (entries 20–28, Table 2). The reaction
was time-dependent. Decreasing the reaction time led to sig-
nificant decrease in the conversion (entries 29–37, Table 2).
There is not an effect to selectivity of the reaction on reducing
the time. On the other hand, increasing the temperature in the
20–45 °C range not significantly improves the yield of
β-nitroalkanol for 24 h reaction time (entries 38–43), as well as
the selectivity of the reaction.
and 7, respectively. The [MII(H2O)6]2+ cations are separated
from one another through the shorter distances, 5.359(2),
7.035(3), 6.582(3) and 7.091(3) Å. The organic cations [(R)-
H2mpz]2+ or [(S)-H2mpz]2+ donate hydrogen bonds to neigh-
boring (SO4)2− tetrahedra. The N–H⋯O distances range
between 2.673(3) and 2.913(3) Å, while the Ow–H⋯O range
between 2.664(3) and 2.983(4) Å.
Compounds 3 and 6 are produced from enantiomeric com-
ponents and interestingly crystallize with different chiral unit
cells. Indeed, in the experiment, the concentration of sulfuric
acid is crucial in the synthesis of complexes 3 and 6. Surpris-
ingly, the high acidity (pH 1–1.5) supports the formation of
compound 3, while low acidity increases the formation of 6.
The crystal structure of [(S)-H2mpz][Cu(H2O)4](SO4)2·2H2O
(5) consists of isolated [Cu(SO4)2(H2O)4]2− anions, [(S)-H2mpz]2+
cations and free water molecules linked together by two types
of hydrogen bonds, N–H⋯O and Ow–H⋯O. The CuII atoms
are coordinated by two oxygen atoms of sulfate groups and
four corners of water molecules forming distorted octahedra
geometry. The CuII ions in this complex display so-called
[4 + 2] type of coordination that has often been observed,
which is consistent with a Jahn–Teller distortion.31 The O–
CuII–O angles range from 88.75(1)° to 91.24(2)°. The inter-
metallic centres are isolated from one another with a shortest
distance Cu⋯Cu = 7.558(2) Å. This MII⋯MII distance in 5 is
the longest one found in this work. Cooperative hydrogen
bond networks are present. Indeed, the intermolecular dis-
tances N⋯O vary from 2.742(6) to 2.855(4) Å and the Ow⋯O
range between 2.643(4) to 2.886(5) Å.
In the complexes 1–7, the S–O distances range from 1.462(2)
to 1.485(2) Å with an average of 1.473(2) Å. Slight differences
in the S–O bond lengths together with the slight deformation
of the anions reflect how the ionic radii of the MII cations
affect the supramolecular structures. The chemical formula of
these materials resembles that of the Tutton’s salts.2,3 Indeed,
although replacing monovalent metals by divalent organic
cations, there are strong similarities between the two families,
with difference resulting from the size, shape and charge of
the amino group involved in the present structures.
Catalytic activity of 1–7 in the Henry reaction
The reactions of a variety of para- (NO2 or OCH3) or ortho-
(CH3) substituted aromatic and aliphatic aldehydes with
nitroethane (Table 3; 1 and 5 were used as the catalyst
To optimize the reaction (Scheme 1) conditions, we examined
the association of benzaldehyde and nitroethane, varying the
catalyst amount, time, temperature and solvent (Table 2). No
402 | Dalton Trans., 2013, 42, 399–406
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 2 Catalytic activity of 1–7 in Henry reactiona
Entry
Catalyst
Time (h)
Amount of catalysts (mol%)
Temp (°C)
Solvent
Yieldb (%)
Selectivityc threo/erythro
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
1
3
5
1
3
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
24
5
10
15
5
10
15
5
10
15
24
24
24
24
24
24
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
—
1.0
1.0
1.0
1.0
1.0
1.0
3.0
4.0
1.0
3.0
4.0
1.0
3.0
4.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
35
45
35
45
35
45
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
THF
97.4
46.7
6.54
46.2
49.9
40.8
37.0
59.0
5.25
23.3
32.1
5.74
37.4
—
73 : 27
81 : 19
0 : 100
100 : 0
100 : 0
100 : 0
100 : 0
48 : 52
0 : 100
100 : 0
65 : 35
0 : 100
100 : 0
—
9
THF
THF
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
MeCN
MeCN
MeCN
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
5
Blank
ZnCl2
Zn(SO4)2
Co(SO4)2
Fe(SO4)2
Cu(SO4)2
—
—
—
—
—
—
—
—
—
—
1
1
1
3
3
3
5
5
5
1
1
1
3
3
3
5
5
5
1
1
3
3
5
5
95.8
98.2
98.2
4.61
11.3
11.5
17.4
51.2
51.6
42.1
69.2
87.3
3.5
72 : 28
73 : 27
72 : 28
0 : 100
0 : 100
0 : 100
100 : 0
100 : 0
100 : 0
72 : 28
73 : 27
73 : 27
0 : 100
0 : 100
0 : 100
100 : 0
100 : 0
100 : 0
73 : 27
72 : 28
0 : 100
0 : 100
100 : 0
100 : 0
6.7
9.3
22.2
35.4
44.7
99.1
99.3
13.6
14.0
55.7
57.3
a Reaction conditions: 1.0–4.0 mol% (0.1–0.4 μmol) of a catalyst precursor (typically 2 mol%), solvent (MeOH, THF, MeCN) (2 mL), nitroethane
(4 mmol) and benzaldehyde (1 mmol). b Determined by 1H NMR analysis (see Experimental). c Calculated by 1H NMR.
precursor) were also studied and shown to provide the respec- sluggish reactions were observed in the reaction of aldehyde
tive β-nitroalkanols with yields ranging from 35 to 99% and with the p-methoxy substituent (entry 5).
with selectivities up to 100% for the threo isomer. The nature
of substrates greatly influences the yields and selectivities.
Hence, the electron-withdrawing group (NO2) leads to lower
yields compared with those for the electron-donating OCH3
Conclusions
moiety (entries 3 and 4 vs. 5 and 6, Table 3). In the case of
2-methylbenzaldehyde, low selectivity and yield are observed,
probably due to steric hindrance (entries 7 and 8, Table 3). For
aliphatic aldehydes, a good yield is observed for acetaldehyde
the shortest one (entries 9–12).
Moreover, p-substituents on the aromatic ring of aldehyde
enhanced enantioselectivity, exhibiting nearly exclusive for-
mation of anti-diastereomers with 39–54% ee (entries 1, 3 and
5). The reaction of aldehyde with the electron-withdrawing
–NO2 group proceeded smoothly (entry 7), whereas the
The directed synthesis of noncentrosymmetric double dihydro-
methylpiperazinediium metallic sulfates was achieved through
the incorporation of the enantiomerically pure chiral amine
2-methylpiperazine. Despite the presence of crystallographic
disorder when using a racemic source of the amine, the synth-
eses containing either S or R-2-mpz result in five noncentro-
symmetric compounds. The thus synthesized salts [H2mpz]-
MII(SO4)2·nH2O can be used as effective (isolated yields up to
99%) and selective catalysts of the nitroaldol addition between
aromatic aldehydes and nitroethane in a methanol medium,
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 399–406 | 403

View Article Online
Paper
Dalton Transactions
Table 3 Henry reaction of various aldehydes and nitroethane with catalysts 1 and 5a
Entry
Substrate
Catalyst
Yieldb (%)
Threo/erythro ratioc
ee(anti)d (%)
1
2
1
5
97.4
49.9
73 : 27
100 : 0
47
82
3
4
1
5
83.2
35.4
65 : 35
100 : 0
39
77
5
6
1
5
98.5
53.8
78 : 22
100 : 0
54
85
7
8
1
5
79.7
27.3
75 : 25
100 : 0
52
81
9
CH3CHO
1
5
1
5
99.1
58.4
98.7
52.1
76 : 24
100 : 0
72 : 28
100 : 0
52
78
43
76
10
11
12
CH3CH2CHO
a Reaction conditions: 2.0 mol% of catalysts 1 (derived from racemic R/S-mpz) and 5 (derived from chiral S-mpz), methanol (2 mL), nitroethane
(4 mmol) and benzaldehyde (1 mmol), reaction time: 24 h. b Determined by 1H NMR analysis (see Experimental). c Calculated by 1H NMR.
d Determined by chiral HPLC analysis.
in contrast with some previously reported results in other solution. The compounds were observed in the positive mode
systems.11–19,24,25 The isolated yields of the final products (capillary voltage = 80–105 V).
depend on structural limitations and solvent polarity. Catalyst
3 is selective towards the erythro isomer, whereas 5 to the threo
one. Thus, comparison of the activities of 5 with those of other
catalysts11–19,24,25 indicates that the former has a promising
potential towards the Henry diastereoselective transformation,
being, to our knowledge, among the most selective (100%)
towards the threo isomer so far reported for the nitroaldol reac-
tion under mild conditions.
Synthesis
The compounds were synthesized by slow evaporation from
aqueous solutions.
[(R/S)-H2mpz][Fe(H2O)6](SO4)2 (1) or [(S)-H2mpz][Fe(H2O)6]-
(SO4)2 (4): 0.2779 g (1.00 × 10−3 mol) FeSO4·7H2O with
0.1000 g (1.00 × 10−3 mol) 2-mpz, 0.2128 g (2.17 × 10−3 mol)
H2SO4, and 4.9500 g (2.75 × 10−1 mol) of deionized water were
mixed under stirring for 15 min and left for slow evaporation.
Dark green prismatic crystals started to form after a few days.
Anal. calcd for C5H26FeN2O14S2 (M = 458.24): C, 13.11; H, 5.72;
N, 6.11. Found: C, 13.21; H, 5.61; N, 5.89. MS (ESI): m/z: 459.3
Experimental section
Materials
FeSO4·7H2O (99%, Merck), ZnSO4·7H2O (99%, Merck), CoSO4· [M + H]+. IR data (cm−1): N–H 1463, 1634; C–H 3014; S–O
7H2O (99%, Merck), CuSO4·5H2O (99%, Merck), 2-methylpiper- 1094, 1195; O–H 3356.
azine (95%, Aldrich), (R)-(−)-2-methylpiperazine (99%,
[(R/S)-H2mpz][Zn(H2O)6](SO4)2 (2), [(R)-H2mpz][Zn(H2O)6]-
Aldrich), (S)-(+)-2-methylpiperazine (99%, Aldrich) and H2SO4 (SO4)2 (3) or [(S)-C5H14N2][Zn(H2O)6](SO4)2 (6) were synthesized
(96%, Carlo Erba) and deionized water were used in the synth- similarly by joint evaporation of 0.2875 g (1.00 × 10−3 mol) of
eses. Infrared spectra (4000–400 cm−1) were recorded on a ZnSO4·7H2O, 0.1000 g (1.00 × 10−3 mol) of 2-mpz, 0.1962 g
1
BIO-RAD FTS 3000MX instrument in KBr pellets. The H and (2.00 × 10−3 mol) of H2SO4, and 4.9860 g (2.77 × 10−1 mol) of
13C NMR spectra were recorded at ambient temperature on a deionized water giving white prismatic crystals. Anal. calcd for
Bruker Avance II + 300 (UltraShield™ Magnet) spectrometer C5H26ZnN2O14S2 (M = 467.77): C, 12.84; H, 5.60; N, 5.99.
operating at 300.130 and 75.468 MHz for proton and carbon- Found: C, 12.74; H, 5.51; N, 5.49. MS (ESI): m/z: 468.3
13, respectively. The chemical shifts are reported in ppm using [M + H]+. IR data (cm−1): N–H 1451, 1634; C–H 3012; S–O
tetramethylsilane as the internal reference. Carbon, hydrogen, 1098, 1110; O–H 3344.
and nitrogen elemental analyses were carried out by the Micro-
[(S)-H2mpz][Cu(H2O)4](SO4)2·2H2O (5) was synthesized simi-
analytical Service of the Technical University of Lisbon. Elec- larly from 0.2496 g (1.00 × 10−3 mol) of CuSO4·5H2O, 0.3000 g
trospray mass spectra (ESI-MS) were run with an ion-trap (3.00 × 10−3 mol) of S-2-mpz, 0.2265 g (2.31 × 10−3 mol) of
instrument (Varian 500-MS LC Ion Trap Mass Spectrometer) H2SO4, and 4.9140 g (2.73 × 10−1 mol) of water giving blue
equipped with an electrospray ion source. For electrospray ion- prismatic crystals. Anal. calcd for C5H26CuN2O14S2 (M =
ization, the drying gas and flow rate were optimized according 465.94): C, 12.89; H, 5.62; N, 6.01. Found: C, 12.76; H, 5.41; N,
to the particular sample with 35 p.s.i. nebulizer pressure. Scan- 5.89. MS (ESI): m/z: 466.8 [M + H]+. IR data (cm−1): N–H 1457,
ning was performed from m/z 100 to 1200 in a methanol 1511; C–H 3017, S–O 1097, 1221; O–H 3421.
404 | Dalton Trans., 2013, 42, 399–406
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
[(S)-C5H14N2][Co(H2O)6](SO4)2 (7) was synthesized from
5: M = 465.94, monoclinic, space group P21, a = 7.5583(5) Å,
0.2811 g (1.00 × 10−3 mol) of CoSO4·7H2O, 0.3000 g (3.00 × b = 10.1721(6) Å, c = 10.7974(7) Å, β = 94.425(4)°, V = 827.67(9)
10−3 mol) of S-2-mpz, 0.2168 g (2.21 × 10−3 mol) of H2SO4, and Å3, Z = 2, Dcalc = 1.870 g cm−3, F000 = 486, μ = 1.646 mm−1
,
5.0040 g (2.78 × 10−1 mol) of water yielding red prismatic crys- 24 878 reflections collected, 6454 unique (Rint = 0.0463), R1 =
tals. Anal. calcd for C5H26CoN2O14S2 (M = 461.33): C, 13.02; 0.0452, wR2 = 0.1033 (I > 2σ).
H, 5.68; N, 6.07. Found: C, 12.93; H, 5.53; N, 5.94. MS (ESI):
6: M = 467.77, monoclinic, space group P21, a = 6.5819(2) Å,
m/z: 462.2 [M + H]+. IR data (cm−1): N–H 1507, 1496; C–H b = 11.0014(2) Å, c = 12.5229(3) Å, β = 101.489(1)°, V = 888.62(4)
3020; S–O 1091, 1186; O–H 3432.
Å3, Z = 2, Dcalc = 1.748 g cm−3, F000 = 488, μ = 1.686 mm−1
,
11 777 reflections collected, 3808 unique (Rint = 0.0358), R1 =
0.0384, wR2 = 0.1003 (I > 2σ).
X-ray structure determinations
7: M = 461.33, monoclinic, space group P21, a = 10.8707(5)
The X-ray quality single crystals of complexes 1–7 were Å, b = 7.8521(3) Å, c = 11.7389(5) Å, β = 116.3950(1)°, V =
immersed in cryo-oil, mounted on a Nylon loop and measured 897.55(7) Å3, Z = 2, Dcalc = 1.725 g cm−3, F000 = 514, μ =
at a temperature of 293 K. Intensity data were collected using a 1.272 mm−1, 10 290 reflections collected, 4319 unique (Rint
Bruker AXS-KAPPA APEX II diffractometer with graphite mono- 0.0309), R1 = 0.0379, wR2 = 0.0982 (I > 2σ).
chromated Mo-Kα (λ 0.71073) radiation. Data were collected
=
Catalytic activity studies
using omega scans of 0.5° per frame and full sphere of data
was obtained. Cell parameters were retrieved using Bruker
SMART software and refined using Bruker SAINT26a on all the
observed reflections. Analytical absorption corrections were
performed by modelling the crystal faces.26b
Structures were solved by direct methods by using the
SHELXS–97 package27 and refined with SHELXL–97.28 Calcu-
lations were performed using the WinGX System–Version
1.80.03.29 The aqua H atoms were located in a difference map
and refined with O–H distance restraints of 0.96(1) Å and H–H
restraints of 1.50(1) Å so that the H–O–H angle fitted to the
ideal value of a tetrahedral angle. H atoms bonded to C and
N atoms were positioned geometrically and allowed to ride on
their parent atoms, with C–H and N–H bonds fixed at 0.97 and
0.90 Å, respectively.
A typical reaction was carried out under air as follows: to
1.0–4.0 mol% (0.1–0.4 μmol) of a catalyst precursor (typically
2 mol%) contained in the reaction flask were added methanol
(2 mL), nitroethane (4 mmol) and aldehyde (1 mmol), in that
order. The reaction mixture was stirred for the required time at
room temperature and air atmospheric pressure. After evapo-
ration of the solvent, the residue was dissolved in DMSO-d6
and analyzed by 1H NMR. The yield of the β-nitroalkanol
product (relatively to the aldehyde) was established typically by
taking into consideration the relative amounts of these com-
1
pounds, as given by H NMR and previously reported.24,25 The
adequacy of this procedure was verified by repeating a number
of 1H NMR analyses in the presence of 1,2-dimethoxyethane as
an internal standard, added to the DMSO-d6 solution, which
gave yields similar to those obtained by the above method.
Moreover, the internal standard method also confirmed that
no side products were formed. The ratio between the threo and
Least squares refinements with anisotropic thermal motion
parameters for all the non-hydrogen atoms and isotropic for the
remaining atoms were employed. CCDC 828001 (1), 859890 (2),
859888 (3), 823975 (4), 724064 (5), 859889 (6), 823974 (7)
contain the supplementary crystallographic data for this paper.
1: M = 458.24, monoclinic, space group P21/c, a = 10.9273(2)
Å, b = 7.8620(10) Å, c = 11.7845(3) Å, β = 116.733(10)°, V =
904.20(3) Å3, Z = 2, Dcalc = 1.683 g cm−3, F000 = 480, μ =
1
erythro isomers was also determined by H NMR spectroscopy.
In the 1H NMR spectra, the values of vicinal coupling con-
stants (for the β-nitroalkanol products) between the α-N–C–H
and the α-O–C–H protons identify the isomers, being J = 7–9 or
3.2–4 Hz for the threo or erythro isomers, respectively.24,25 The
enantiomeric excess was determined by HPLC using a Shi-
madzu high performance liquid chromatograph with UV
detection (210 nm). In the HPLC (OD-H, 4 : 96 isopropanol–
hexane at 1 mL min−1) four peaks at 35.1, 51.5 for the anti
isomer and 43.9 and 57.1 for the syn isomer were observed.
1.133 mm−1, 13 422 reflections collected, 2290 unique (Rint
0.0431), R1 = 0.0540, wR2 = 0.1156 (I > 2σ).
=
2: M = 467.77, monoclinic, space group P21/n, a = 5.5988(1)
Å, b = 10.9613(2) Å, c = 12.5479(2) Å, β = 101.3850(1)°, V =
889.75(3) Å3, Z = 2, Dcalc = 1.746 g cm−3, F000 = 488, μ =
1.684 mm−1, 30 251 reflections collected, 3918 unique (Rint
0.0446), R1 = 0.0502, wR2 = 0.1288 (I > 2σ).
=
3: M = 467.77, monoclinic, space group P21, a = 10.8665(2)
Å, b = 7.8600(1) Å, c = 11.7029(2) Å, β = 116.2830°, V = 896.22(3)
Acknowledgements
Å3, Z = 2, Dcalc = 1.733 g cm−3, F000 = 488, μ = 1.672 mm−1
,
Grateful thanks are expressed to Dr T. Roisnel (Centre de
10 529 reflections collected, 3957 unique (Rint = 0.0482), R1 = Diffractométrie X, Université de Rennes I) for the assistance in
0.0302, wR2 = 0.0784 (I > 2σ). single-crystal X-ray diffraction data collection. This work has
4: M = 458.24, monoclinic, space group P21, a = 10.9444(4) been partially supported by the Foundation for Science and
Å, b = 7.8602(3) Å, c = 11.8020(4) Å, β = 116.170(2)°, V = 911.19(6) Technology (FCT) (projects PTDC/QUI-QUI/102150/2008 and
Å3, Z = 2, Dcalc = 1.670 g cm−3, F000 = 480, μ = 1.124 mm−1
,
PEst-OE/QUI/UI0100/2011) and by Tuniso-Portuguese Scientific
20 839 reflections collected, 8255 unique (Rint = 0.0345), R1 = and Technical Cooperation 2009 Program. The authors
0.0471, wR2 = 0.0880 (I > 2σ).
acknowledge the Portuguese NMR Network for providing
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 399–406 | 405

View Article Online
Paper
Dalton Transactions
access to the NMR facility, and IST Node of the Portuguese 14 M. N. Kopylovich, T. C. O. Mac Leod, K. T. Mahmudov,
Network of mass-spectrometry (Dr Conceição Oliveira) for the
ESI-MS measurements.
M. F. C. Guedes da Silva and A. J. L. Pombeiro, Dalton
Trans., 2011, 40, 5352–5361.
15 H. Sasai, T. Suzuki, S. Arai and M. Shibasaki, J. Am. Chem.
Soc., 1992, 114, 4418–4420.
16 M. Shibasaki and N. Yoshikawa, Chem. Rev., 2002, 102,
2187–2209.
References
17 C. Palomo, M. Oiarbide and A. Laso, Eur. J. Org. Chem.,
2007, 2561–2574.
1 M. Dan, J. N. Behera and C. N. R. Rao, J. Mater. Chem.,
2004, 14, 1257–1265.
18 J. Boruwa, N. Gogoi, P. P. Saikia and N. C. Barua, Tetra-
hedron: Asymmetry, 2006, 24, 3315–3326.
19 C. Palomo, M. Oiarbide and A. Mielgo, Angew. Chem., Int.
Ed., 2004, 43, 5442–5444.
20 F. Hajlaoui, H. Naïli, S. Yahyaoui, Alexander J. Norquist,
T. Mhiri and T. Bataille, J. Organomet. Chem., 2012, 700,
110–116.
21 F. Hajlaoui, H. Naïli, S. Yahyaoui, M. M. Turnbull,
T. Mhiri and T. Bataille, Dalton Trans., 2011, 40, 11613–
11620.
22 H. Naïli, J. Coord. Chem., 2012, 65, 1178–1188.
23 F. Hajlaoui, S. Yahyaoui, H. Naïli, T. Mhiri and T. Bataille,
Inorg. Chim. Acta, 2010, 363, 691–695.
24 A. Cwik, A. Fuchs, Z. Hella and J. Clacens, Tetrahedron,
2005, 61, 4015–4021.
25 V. J. Bulbule, V. H. Deshpande, S. Velu, A. Sudalai,
S. Sivasankar and V. T. Sathe, Tetrahedron, 1999, 55, 9325–
9332.
2 S. Yahyaoui, W. Rekik, H. Naïli, T. Mhiri and T. Bataille,
J. Solid State Chem., 2007, 180, 3560–3570.
3 C. N. Morimoto and E. C. Lingafelter, Acta Crystallogr., Sect.
B: Struct. Crystallogr. Cryst. Chem., 1970, 26, 335–341.
4 L. F. Kirpichnikova, L. A. Shuvalov, N. R. Ivanov, B. N. Prasolov
and E. F. Andreyev, Ferroelectrics, 1989, 96, 313–317.
5 E. A. Muller, R. J. Cannon, A. N. Sarjeant, K. M. Ok,
P. S. Halasyamani and A. J. Norquist, Cryst. Growth Des.,
2005, 5, 1913–1917.
6 A. Berkessel, H. Gröger and D. MacMillan, Asymmetric
Organocatalysis – From Biomimetic Concepts to Applications
in Asymmetric Synthesis, Wiley-VCH, 2005, p. 454.
7 (a) H. Pellissier, Recent Developments in Asymmetric Organo-
catalysis, Springer, 2010, p. 241; (b) M. K. Bhunia, S. K. Das,
M. M. Seikh, K. V. Domasevitch and A. Bhaumik, Poly-
hedron, 2011, 30, 2218–2226.
8 F. Song, C. Wang and W. Lin, Chem. Commun., 2011, 47,
8256–8258.
26 (a) Bruker, APEX2 & SAINT, Bruker, AXS Inc., Madison,
Wisconsin, USA, 2004; (b) J. De Meulenaer and H. Tompa,
Acta Crystallogr., 1965, 19, 1014–1018.
27 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 1990, 46, 467–473.
9 D. Dang, P. Wu, C. He, Z. Xie and C. Duan, J. Am. Chem.
Soc., 2010, 132, 14321–14323.
10 G. Nickerl, A. Henschel, R. Grünker, K. Gedrich and
S. Kaskel, Chem. Ing. Tech., 2011, 83, 90–103.
11 G. Rosini, in Comprehensive Organic Synthesis, ed.
B. M. Trost, I. Flemig, C. H. Heatcock, Pergamon,
New York, 1991, vol. 2, pp. 321–340.
28 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystal-
logr., 2008, 64, 112–122.
29 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
30 A. L. J. Spek, Appl. Crystallogr., 2003, 36, 7–13.
31 V. M. Petrusevski, Bull. Chem. Technol., 1990, 9, 13–19.
12 F. A. Luzzio, Tetrahedron, 2001, 57, 915–945.
13 P. Hammar, T. Marcelli, H. Hiemstra and F. Himo, Adv.
Synth. Catal., 2007, 349, 2537–2548.
406 | Dalton Trans., 2013, 42, 399–406
This journal is © The Royal Society of Chemistry 2013