Cu(II) homogeneous and heterogeneous catalysts for the asymmetric Henry reaction

Article abstract of DOI:10.1016/j.molcata.2010.03.013

New homogeneous Cu(II) complexes have been prepared, fully characterised and tested for the asymmetric Henry reaction. Heterogeneous Cu(II) catalysts have been synthesised for comparison purposes.

Full text of DOI:10.1016/j.molcata.2010.03.013

Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Cu(II) homogeneous and heterogeneous catalysts for the asymmetric
Henry reaction
Matthew D. Jonesa,∗, Christine J. Coopera, Mary F. Mahonb,∗∗, Paul. R. Raithbya,
David Apperleyc, Joanna Wolowskad, David Collisond
a Department of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK
b Bath Chemical Crystallography Unit, Department of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK
c Durham University, Department of Chemistry, Solid State NMR Service, Durham DH1 3LE, UK
d University of Manchester, School of Chemistry, Manchester M13 9PL, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
New Cu(II) complexes have been prepared and characterised via single crystal X-ray diffraction studies,
EPR spectroscopy, elemental analysis and high resolution mass spectrometry. In all cases 1:2 (copper
to ligand) stoichiometric complexes were isolated. The homogeneous Cu(II) complexes were tested for
the asymmetric Henry reaction. Conversions in excess of 70% were obtained with enantioselectivities
in the range 0–78%. Heterogeneous Cu(II) catalysts have been prepared. In these cases high conversions
were obtained. However, after prolonged reaction time the main product observed was 1,3-dinitro-2-
phenyl propane. The formation of this product can be curtailed by both decreasing the catalyst loading
and employing shorter reaction times.
Received 8 January 2010
Received in revised form 1 March 2010
Accepted 7 March 2010
Available online 12 March 2010
Keywords:
Homogeneous and heterogeneous catalysts
Copper
Supported catalysts
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
catalysts for this process remains limited. However, there have
been successes with catalysts supported on PEG polymers, Wang
The Henry (nitroaldol) reaction is a fundamental carbon–carbon
bond forming process in organic chemistry and is a key step in many
syntheses [1,2]. In recent years the asymmetric variant of the Henry
reaction has become a powerful method in the enantioselective
preparation of carbon–carbon bonds. This transformation can be
catalysed by Cu(II) complexes [3–14]. For example, Bandini et al.
have investigated a series of C2-symmetric oligothiophene Cu(II)
systems [15]. Cu(II) bis-oxazolines have also shown promise in this
area [10,16]. Blay et al. have successfully demonstrated the use of
C1-symmetric camphor derived ligands for the nitroaldol reaction
of nitromethane and bromonitromethanes [17–20]. Cu(II) reagents
based on (−)-sparteine are also active catalysts [21]. Recently,
Constable demonstrated that Cu(II)-salen systems are effective
with yields and enantioselectivities significantly enhanced with
the addition of a further equivalent of Cu(OAc)2 [22]. However,
the Henry reaction is by no means limited to Cu(II), for example
Zn(II) [23–25], Cr(III) [26,27], La(III) [28] and Co(II) [29] have all
been shown to catalyse the reaction. The use of heterogeneous
type resins and dendrimers [30–32]. In this paper we have pre-
pared five new Cu(II) complexes, all of which were characterised
by single crystal X-ray diffraction. These complexes were tested
for the asymmetric addition of nitromethane to benzaldehyde.
Three classes of supported heterogeneous catalysts have been pre-
pared; their characterisation (via solid-state NMR spectroscopy and
EPR spectroscopy) is presented together with initial results for the
asymmetric Henry reaction.
2. Experimental
2.1. General procedures
1
1H and 13C{ H} NMR spectra were recorded on a Bruker 300
or 250 MHz spectrometer, and referenced to residual solvent peaks
(CDCl3). Coupling constants are given in Hertz. Elemental analysis
was performed by Mr. A.K. Carver at the Department of Chemistry,
University of Bath. (1R,2R)-1,2-Diaminocyclohexane was resolved
from the commercially available trans-1,2-diaminocyclohexane by
the method of Jacobsen and co-workers [33]. ICP analysis was per-
formed by Medac Ltd. Nanoporous carbon was purchased from
Aldrich (<50 nm particle size, surface area > 100 m2 g−1) and used
as received.
∗
Corresponding author. Tel.: +44 01225 384908; fax: +44 01225 386231.
Corresponding author.
E-mail addresses: mj205@bath.ac.uk (M.D. Jones),
∗∗
m.f.mahon@bath.ac.uk (M.F. Mahon).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.03.013

M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
9
2.2. X-ray crystallography
(Rint = 0.0313). Final GooF = 1.052, R1 = 0.0263, wR2 = 0.0614,
R
indices based on 9626 reflections with I > 2ꢂ (refinement on F2),
595 parameters, 3 restraints. ꢃ = 1.201 mm−1. Absolute structure
parameter = −0.006(5).
Data were collected on a Nonius Kappa CCD diffractometer
using Mo-K␣ radiation (ꢀ = 0.71073 Å) at a temperature of 150(2) K,
except for Cu(4)2·(OTf)2 which was collected on a Xcalibur, Atlas
diffractometer using CuK␣ radiation (ꢀ = 1.54184 Å) at 100(2) K. All
structures were solved by direct methods and refined on all F2
data using the SHELXL-97 suite of programs. All hydrogen atoms
were included in idealised positions and refined using the rid-
ing model. Refinements were generally straightforward with the
following exceptions and points of note. For Cu(2)2·(OTf)2 excel-
lent convergence was obtained once pseudo-merohedral twinning
(47%) about the −1 0 1 direct lattice direction had been addressed.
The unit cell metrics in concert with the extent and nature
of the twinning suggested higher symmetry (orthorhombic ‘C’,
or monoclinic ‘C’) initially, neither of which produced a credi-
ble model. For Cu(2)2·(OTf)2 the ADPs for C(25) are slightly less
isotropic than desirable, but efforts to model disorder in this
region of the electron density map afforded no improvement in
convergence. Despite copious recrystallisation efforts only small
crystals (0.05 mm × 0.05 mm × 0.01 mm) for Cu(4)2·(OTf)2 were
obtained, due to this weak diffraction the data was truncated to
ꢁ = 62.26 (CuK␣). Nonetheless, the structure is unambiguous. For
Cu(2)2·(OTf)2 HR-ESI Calc. for [M2+] 325.6739 found 325.6736.
Calc. for C42H52CuF6N4O6S2 C, 53.07; H, 5.51; N, 5.89. Found C,
52.3; H, 5.44; N, 5.74. FT-IR (solid cm−1) 3224w, 3156w, 2948w,
1497w, 1455m, 1287m, 1241s, 1147s, 1026s, 987m, 761m, 634
s. C42H52CuF6N4O6S2, M = 950.54, 0.35 mm × 0.20 mm × 0.10 mm,
monoclinic, P21, a = 11.7300(1) Å, b = 35.5410(4) Å, c = 11.7290(1) Å,
ˇ = 114.6240(1)◦,
V = 4445.10(7) Å3,
Z = 4,
Dc = 1.420 g/cm3,
F0 0 0 = 1980, 2ꢁmax = 54.9◦, 35572 reflections collected, 35,590
unique. Final GooF = 1.186, R1 = 0.0467, wR2 = 0.1288, R indices
based on 33,459 reflections with I > 2ꢂ (refinement on F2), 1100
parameters,
1
restraint. ꢃ = 0.661 mm−1
.
Absolute structure
parameter = −0.009(9).
Cu(3)2·(OTf)2 HR-ESI Calc. for [M2+
] 385.6950, found
385.6950. Calc. for C46H60CuF6N4O10S2 C, 51.60; H, 5.65; N,
5.23. Found C, 51.5; H, 5.63; N, 5.21. FT-IR (solid cm−1) 3327w,
3225w, 2944w, 1595m 1454m, 1290s, 1241s, 1221s, 1158s,
1025s, 758m, 706m, 631s.
C46H60CuF6N4O10S2, M = 1070.64,
0.10 mm × 0.10 mm × 0.10 mm, monoclinic, P21, a = 11.5320(5) Å,
b = 18.6300(1) Å, c = 11.8630(6) Å, ˇ = 102.643(3)◦, V = 2486.9(2) Å3,
Z = 2, Dc = 1.430 g/cm3, F0 0 0 = 1118, 2ꢁmax = 50.0◦, 28,133 reflec-
tions collected, 8616 unique (Rint = 0.1396). Final GooF = 1.029,
2+
the Cu(5)2 cation the copper centre was modelled over two sites
in a 80:20 ratio. Multi-scan absorption corrections were applied to
the data on merit.
R1 = 0.0637, wR2 = 0.1220,
R indices based on 5199 reflec-
tions with I > 2ꢂ (refinement on F2), 681 parameters,
1
2.3. Solid-state NMR
restraint. ꢃ = 0.605 mm−1
−0.008(18).
.
Absolute structure parameter =
Solid-state NMR spectra were recorded at the EPSRC national
solid-state NMR service centre (Durham University) on a Varian
VNMRS 400 MHz spectrometer (100.562 MHz for 13C), using the
cross-polarization pulse sequence (contact time 3.0 ms and recycle
delay 1.0 s), with TPPM decoupling. A spinning rate of 10.0 kHz was
employed.
Cu(4)2·(OTf)2 Calc. for C46H60CuF6N4O6S2 C, 54.88; H, 6.01; N,
5.57. Found C, 53.5; H, 5.88; N, 5.49. FT-IR (solid cm−1) 3256w,
2945w, 1494w, 1459w, 1284s, 1247s, 1231s, 1163m, 1024s,
921m, 751m, 742m, 637s. C93H124Cu2F12N8O13S4, M = 2045.32,
0.05 mm × 0.05 mm × 0.01 mm, monoclinic, P21, a = 11.4238(7) Å,
b = 32.3577(8) Å, c = 13.0790(4) Å, ˇ = 96.545(4)◦, V = 4803.1(3) Å3,
Z = 2, Dc = 1.414 g/cm3, F0 0 0 = 2144, CuK␣ radiation, ꢀ = 1.54184 Å,
2ꢁmax = 124.5◦, 22,418 reflections collected, 10,864 unique
2.4. EPR
(Rint = 0.0735). Final GooF = 0.878, R1 = 0.0466, wR2 = 0.0789,
R
All the measurements were performed using a Bruker EMX spec-
trometer at X-band (∼9.4 GHz) and K-band (∼24.0 GHz) at room
temperature and at 120 K. The samples were measured as powders
at 290 and 120 K, fluid solutions at 290 K and frozen solution at
120 K. The simulations of the spectra were performed using Bruker
XSophe computer simulation software (version 1.1.4).
indices based on 7323 reflections with I > 2ꢂ (refinement on F2),
1194 parameters, 1 restraint. ꢃ = 2.106 mm−1. Absolute structure
parameter = 0.027(19).
Cu(5)2·(OTf)2·H2O HR-ESI Calc. for [M2+
] 159.5956; found
159.5954. Calc. for C16H34CuF6N4O7S2 C, 30.21; H, 5.39; N,
8.81. Found C, 30.2; H, 5.38; N, 8.75. FT-IR (solid cm−1
3416w, 3245w, 2963w, 1665w, 1604w, 1461w, 1250s,
1226m, 1027s, 943m, 760m, 632s. 16H34CuF6N4O7S2,
)
2.5. Ligand preparation and characterisation
C
M = 636.13, 0.40 mm × 0.15 mm × 0.10 mm, tetragonal, P41212,
a = b = 12.6100(1) Å, c = 16.2480(2) Å, V = 2583.63(4) Å3, Z = 4,
Dc = 1.635 g/cm3, F0 0 0 = 1316, 2ꢁmax = 54.9◦, 51,652 reflections col-
lected, 2957 unique (Rint = 0.0434). Final GooF = 1.075, R1 = 0.0226,
wR2 = 0.0573, R indices based on 2885 reflections with I > 2ꢂ
The ligands were prepared via standard procedures [34,35], a
typical procedure for 3 and 4 are given in supporting information.
2.6. Complex preparation and characterisation
(refinement on F2), 169 parameters, 3 restraints. ꢃ = 1.094 mm−1
Absolute structure parameter = −0.013(10).
.
A typical procedure for Cu(1)2·(OTf)2. 1 (0.82 g, 2.8 mmol) was
dissolved in MeOH (10 mL) to which Cu(OTf)2 (0.5 g, 1.4 mmol)
was added. This was stirred for 1 h and the solution removed in
vacuo. The resulting blue powder was recrystallised from MeOH
and Et2O at −20 ◦C and a crop of deep blue crystals was obtained
2.7. Synthesis of heterogeneous catalysts
To prepare the amine grafted material, silica (60 Å Davisil grade)
was initially reacted with (MeO)3Si(CH2)3NH2 [36,37].
after
2
days. Cu(1)2·(OTf)2. HR-ESI Calc. for [M2+
] 145.5800
found 145.5807. Calc. for C14H28CuF6N4O6S2 C, 28.47; H, 4.75; N,
9.49. Found C, 28.2; H, 4.97; N, 9.12. FT-IR (solid cm−1) 3267w,
2935w, 1603m, 1589m 1494m 1468m, 1289s, 1239s, 1154m
1118m, 919m, 757s, 635s. C28H56Cu2F12N8O12S4, M = 1180.13,
0.40 mm × 0.30 mm × 0.25 mm, triclinic, P1, a = 8.6140(1) Å,
b = 11.6740(2) Å, c = 11.7340(2) Å, ˛ = 81.757(1)◦, ˇ = 87.113(1)◦,
␥ = 89.344(1)◦, V = 1166.29(3) Å3, Z = 1, Dc = 1.680 g/cm3, F0 0 0 = 606,
2ꢁmax = 55.0◦, 22,337 reflections collected, 10,000 unique
A Amine functionalised silica (10 g, 10 mmol loading of NH2)
was suspended in CH2Cl2 (100 mL), and terephthalaldehyde (1.34 g,
10 mmol) added. The mixture was stirred at room temperature for
2 h before the silica material was collected and washed with CH2Cl2
(3× 50 mL) and dried. C, 8.60; H, 1.39; N, 1.27.
B The previously synthesised silica material (A) (3.9 g) was sus-
pended in CH2Cl2 (50 mL) and (R,R)-1,2-diaminocyclohexane (0.4 g,
3.9 mmol) added. The mixture was stirred at room temperature for

10
M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
4 h before the silica material was collected and washed with CH2Cl2
(3× 50 mL) and dried. C, 8.78; H, 1.90; N, 2.33.
3. Results and discussion
C The previously synthesised silica material (B) (1.5 g) was sus-
pended in CH2Cl2 (50 mL) and benzaldehyde (0.15 mL, 1.5 mmol)
added. The mixture was stirred at room temperature for 4 h
before the silica material was collected and washed with CH2Cl2
(3× 50 mL) and dried, C, 8.74; H, 1.73; N, 1.78. Copper cata-
lysts: the previously synthesised silica material (C) (1.0 g) and
Cu(OTf)2 (0.5 g, 1.5 mmol) were placed in a flask under argon,
and methanol (40 mL) added. The mixture was stirred at room
temperature for 1 h before the silica material was collected.
The product was washed with methanol (3× 50 mL) until the
washings were clear and dried. C, 6.10; H, 1.36; N, 1.23, Cu
0.28%.
3.1. Preparation and characterisation of Cu(II) complexes
Ligands 1–4 were prepared via standard literature procedures.
Ligands of this type have shown promise for the asymmetric Henry
reaction [22]. Ligand 5 was purchased from Aldrich. All complexes
prepared are shown in Scheme 1.
All complexes were characterised by single crystal X-ray diffrac-
tion, see Fig. 1 for the molecular structure of the Cu(2)2 cation.
2+
This paper reports the crystallographic characterisation of all the
Cu(II) complexes. Cu(1)2·(OTf)2 crystallises in the triclinic space
group P1 and contains two copper centres in the asymmetric
unit. The copper centre is coordinated by two diamine ligands
and one weakly coordinated OTf− counter-ion giving the cop-
per centre a coordination number of five. The Cu–N distances of
approximately 2.0 Å are analogous to literature precedent [39–42].
Cu(2)2·(OTf)2 crystallises in the monoclinic space group P21, with
two Cu(II) centres in the asymmetric unit. Two diamine moieties
coordinated to each Cu(II) centre and no Cu–OTf interactions are
observed. The copper centres are in a highly distorted square pla-
nar geometry, as exemplified by the range of N–Cu(1)–N angles
present in the structure {cis interactions 85.25(16)–102.64(18)◦
and for trans 149.78(17)–152.11(16)◦}. Cu(3)2·(OTf)2 crystallises in
the monoclinic space group P21. In this case the Cu(II) centre is best
described has having a slightly distorted square planar geometry.
Cu(4)2·(OTf)2 crystallises in the monoclinic space group P21 with
two Cu(II) centres in the asymmetric unit. The Cu(II) centre again is
pseudo square planar, see supporting information for selected bond
distances and angles. For both Cu(3)2·(OTf)2 and Cu(4)2·(OTf)2 two
diamine ligands are coordinated to the Cu(II) centre together with
a weakly coordinating OTf− counter-ion.
There are significant differences in the coordination geometries
between Cu(2)2·(OTf)2 and either Cu(3)2·(OTf)2 or Cu(4)2·(OTf)2.
These are manifested by analysis of the Cu–NH–CH2–CAr torsion
angles. For Cu(2)2·(OTf)2 these are in the range 47.6–65.7◦, how-
ever for Cu(3)2·(OTf)2 the analogous angles are 85.9–169.3◦ and
for Cu(4)2·(OTf)2 59.7–177.3◦. There is also a significant difference
between the angle of the planes formed from N(1)–Cu(1)–N(2)
and N(3)–Cu(1)–N(4) which are close to parallel (i.e. 180◦) for
Cu(3)2·(OTf)2 and Cu(4)2·(OTf)2. However, for Cu(2)2·(OTf)2 the
analogous angle is 139◦. Cu(5)2·(OTf)2·H2O crystallised in the
tetragonal space group P41212, Fig. 1. The copper centre is best
D
Analogous to C except o-OMe-benzaldehyde (0.18 mL,
1.5 mmol) was added. C, 8.63; H, 1.55; N, 1.67. Copper cata-
lyst: analogous method as detailed above. C, 6.43; H, 1.30; N,
1.24.
E Analogous to C except o-tolualdehyde (0.17 mL, 1.5 mmol) was
added. C, 8.86; H, 1.66; N, 1.79. Copper catalyst: Analogous method
as detailed above C, 6.11; H, 1.33; N, 1.18.
2.8. Preparation of Cu-exchanged zeolite Y
Zeolite HY (10 g) was stirred with a solution of Cu(OAc)2
(1.57 g, 7.86 mmol) in distilled water (30 mL) for 24 h at room
temperature. The zeolite material was collected by vacuum fil-
tration, dried under vacuum at 100 ◦C and calcined at 550 ◦C,
pXRD of the calcined material was analogous to that of pure zeo-
lite HY. Cu-exchanged zeolite Y (0.36 g, 0.090 mmol Cu, based
on ICP results of 1.6 wt% Cu in the sample) was placed under
argon, and to this a solution of chiral ligand (0.075 mmol) in
CH2Cl2 (10 mL) was added. After stirring at room temperature
for 3 h, benzaldehyde (0.1 mL, 1.0 mmol), nitromethane (0.55 mL,
10 mmol) and triethylamine (35 ␮l, 0.25 mmol) were added, and
the mixture stirred at room temperature for a pre-determined
time.
2.9. Preparation of Cu on carbon catalysts [38]
Nanoporous carbon (500 mg) was suspended in methanol
(10 mL) to which the copper complex (2.35 × 10−5 mol) was added.
This was stirred overnight, filtered, washed with methanol (2×
20 mL) and dried in vacuo.
2.10. Typical catalytic procedure
Under argon EtOH (10 mL) was added to a Schlenk flask, to
which the Cu(II) homogeneous catalyst was added (0.05 mmol)
or the desired amount of heterogeneous catalyst and the solution
stirred. Benzaldehyde (0.1 mL, 1.0 mmol), nitromethane (0.55 mL,
10 mmol) and NEt3 (35 ␮L, 0.25 mmol) were added and the solu-
tion stirred for the appropriate amount of time. After the desired
time the reaction was filtered through a plug of silica and the sol-
vents were removed in vacuo. 1H NMR spectroscopy was used to
determine the conversion by analysis of the 1H integral of PhCHO
at 9.94 ppm to the 1H integral of PhCH(OH)CH2NO2 at 5.45 ppm.
For the heterogeneous catalysts a 1H quintet {Ph(CH)(CH2NO2)2}
at 4.25 ppm and a 1H doublet at 7.90 ppm for {PhCH = CHNO2}
were accounted for in the selectivity measurements. The enan-
tiomeric excess was determined by HPLC using an Agilent Compact
1120 LC with UV detection (254 nm). A flow rate of IPA:hexane
(1:9) at 1 mL/min was used with a Lux Cellolose-1 column,
the retention times were 16 and 19 min for the two enan-
tiomers.
Scheme 1. Complexes prepared in this study.

M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
11
Table 1
Catalytic results for the homogeneous catalysts.
b
Catalysta
Time/h
Equiv. NEt3
Conv.c
eed
0
60 (S)
43 (S)
68 (S)
72 (S)
78 (S)
36 (R)
Cu(1)2·(OTf)2
Cu(2)2·(OTf)2
Cu(3)2·(OTf)2
Cu(3)2·(OTf)2
Cu(3)2·(OTf)2
Cu(4)2·(OTf)2
Cu(5)2·(OTf)2
6
6
4
6
6
6
4
0.25
0.25
0.5
0.25
0.13
0.25
0.5
68
45
75
59
20
26
76
a
The catalyst:benzaldehyde:nitromethane molar ratio was 0.05:1:10.
Equivalent (with respect to benzaldehyde).
b
c
Conversion as determined by 1H NMR spectroscopic analysis.
d
Determined by chiral HPLC, the absolute configuration was determined by opti-
cal rotation data.
per catalysts. The procedure was also attempted with free ligand
4 (without base) as there are examples of organocatalysed Henry
reactions [47]. In this case after 6 h a 40% conversion was obtained
but it failed to induce any enantioselectivity in the product. All the
Cu(II) catalysed transformations were performed using 1 mmol of
benzaldehyde, 10 mmol of nitromethane, 0.05 mmol of catalyst and
0.5, 0.25, 0.13 mmol of the promoter triethyl amine, Table 1. The
conversions were determined via 1H NMR spectroscopy and the
enantiomeric excess from HPLC. Cu(1)2·(OTf)2 was relatively active
for the Henry reaction but failed to induce any selectivity in the
product [15]. However, Cu(2–5)2·(OTf)2 were successful in confer-
ring enantioselectivity into the process, with the highest ee being
78% for Cu(4)2·(OTf)2. It is observed that increasing the steric bulk of
the substituent on the aromatic ring (H to Me to OMe) has the effect
of increasing the enantioselectivity in the product. Decreasing the
amount of the NEt3 promoter increased the selectivity, but at the
expense of conversion. In this case it was seen that without NEt3 no
activity was observed. During the catalysis itself the counter-ion is
presumably no longer coordinated to the copper centre to facilitate
the coordination of benzaldehyde and nitromethane to allow the
catalysis to proceed.
2+
2+
Fig. 1. Molecular structures of Cu(2)2 (top) and Cu(5)2 ·H2O cation (bottom),
the triflate counter-ions and all hydrogen atoms except those bound to nitrogen
atoms and water have been removed for clarity. Atoms labelled with a suffix A are
generated by the −x + 1,−y + 1,−z + ½ symmetry operation.
3.3. Preparation, characterisation and catalysis of the
heterogeneous systems
described as square based pyramidal. In this case a molecule
of water is coordinated to the metal centre with a Cu(1)–O(1)
distance of 2.3076(15) Å. The complexes were also analysed
via mass spectrometry and IR spectroscopy. Mass spectrome-
try afforded the parent M2+ ion, noteworthy for Cu(1)2·(OTf)2
a peak at 440.1128 (Calc. 440.1130) was also observed indicat-
ing that one OTf counter-ion maybe coordinated in solution. To
further characterise these materials EPR spectroscopy was per-
formed on Cu(2)2·(OTf)2 and Cu(5)2·(OTf)2·H2O (see supporting
information for further details and for a table of g and A val-
ues). For Cu(5)2·(OTf)2·H2O the simulation of the frozen solution
spectrum at X-band at 130 K gave an axial set of g values
Various supported systems have been utilised and are described
herein for the asymmetric Henry reaction. Specifically the reaction
we are studying is the addition of benzaldehyde to nitromethane
[30–32]. Important to this study is that silica and zeolites materials
have not been used extensively as supports for this reaction, there-
fore we attempted to heterogenize the ligands described herein to
these supports [48,49]. The heterogeneous catalysts were prepared
1
as shown in Fig. 2. 13C{ H} CP/MAS solid-state NMR spectroscopic
analyses are in agreement with the desired structures, Fig. 3 [36].
For example A has resonances at 9, 23, 58 and 63 ppm for the
CH2’s of the propyl tether and unreacted OMe [36]; the aromatic
peaks are centred at 127 and 138 ppm, the imine carbon is seen at
161 ppm and the carbonyl at 194 ppm. After reaction with (1R,2R)-
1,2-diaminocyclohexane new resonances at 25, 33 and 43 ppm are
observed for the cyclohexane ring, and as expected there is a dis-
appearance of the CHO peak observed in A. For C, D and E there is
an increase in the intensity of the aromatic resonances compared
to B which would be expected after the reaction with the benzalde-
hyde derivative. For D there is a new peak at 54 ppm for the OMe
group and additional intensity for the peak at 160 ppm arising from
the aromatic carbon attached to the methoxy group. For E there is
a new resonance at 17 ppm consistent with the presence of the
methyl substituent and an increase in intensity of the resonance at
138 ppm from the aromatic carbon bound to the methyl group. To
further characterise these materials the solid Cu2+ catalysts, where
R = OMe or Me, were analysed via EPR spectroscopy. The simulation
and A values with g = 2.05, g = 2.20 and A = 52.6 × 10−4 cm−1
,
A = 187.6 × 10−4 cm−⊥1, which are in agreement with literature
||
⊥
||
precedent for complexes with analogous structural motifs as those
described herein [43–46]. For Cu(2)2·(OTf)2 simulation of the frozen
solution spectrum at X-band at 130 K gave an axial set of g val-
ues and A values with g = 2.05, g = 2.20 and A = 23.9 × 10−4 cm−1
,
⊥
||
⊥
A = 180.8 × 10−4 cm−1. For both complexes the analysis was rel-
||
atively complex, with evidence of multiple Cu(II) sites present,
which could arise from coordination of the triflate anion or sol-
vent molecules. For Cu(2)2·(OTf)2 there was also evidence for dimer
formation in solution.
3.2. Homogeneous catalysis
The catalytic activity for the asymmetric nitroaldol coupling of
nitromethane and benzaldehyde was investigated for the five cop-

12
M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
Fig. 2. Preparation of the heterogeneous Cu(II) catalysts. (i) 1,4-Benzenedicarboxaldehyde/MeOH, (ii) (1R,2R)-diaminocyclohexane, (iii) benzaldehyde, 2-
methylbenzaldehyde or 2-methoxybenzaldehyde/MeOH, (iv) Cu(OTf)2/MeOH.
of the powder spectrum of samples at X-band, at 290 and 130 K, are
consistent with the presence of Cu(II) in the material. Both hetero-
diates [54]. These alkanes have been shown to be produced from
heterogeneous primary amine/tertiary amine catalysts [50]. Sig-
nificantly, acid surfaces have been shown to enhance the activity
via a cooperative mechanism involving free amines on the surface
and potentially by activating the intermediate ␤-nitrostyrene to
attack by nitromethane [52]. This is presumably a possible expla-
nation for the observation of this by-product in this case, as the
support is acidic and in these systems there are potentially free
amine (–NH2) sites on the surface remaining from unreacted moi-
eties. Although not the main thrust of this work it should be noted
that the formation of 1,3-dinitro alkanes is typically performed at
elevated temperatures [51]. In an attempt to reduce the undesired
side-reaction 100 and 200 mg of catalyst were used for 24 h, Table 2.
Noteworthy, aswitch in enantioselectivity ongoing fromthehomo-
geneous to the heterogeneous systems was observed. Bandini et al.
observed (with the same absolute stereochemistry in the backbone)
a change in selectivity on going from an amine to the analogous
imine with his oligothiophene Cu(II) systems [15]. We suggest that
this is also the case here as the silica supported heterogeneous cat-
geneous catalysts have the same parameters g = 2.05, g = 2.25 and
⊥
||
A
⊥
= 23.0 × 10−4 cm−1, A = 191.2 × 10−4 cm−1. These are analogous
||
to Cu(II)-imine systems in the literature supporting the structures
proposed in Fig. 2 [46]. These values are also analogous to the homo-
geneous catalysts prepared in this study.
The solid-supported systems were tested for the Henry
reaction (1:10:0.25 molar ratio of benzaldehyde:nitromethane:
triethylamine) using 200 mg of the supported system and for a
period of 72 h.
Good conversions ca. 100% were observed, as there was no
evidence of benzaldehyde in the 1H NMR spectrum. However,
the major product formed in the reaction was 1,3-dinitro-2-
phenyl propane as opposed to the desired nitroaldol species,
␤-nitrostyrene was also observed [50–53]. 1,3-Dinitro alkanes
have been shown to be key building blocks in the synthesis of
HIV-protease inhibitors and other biologically important interme-
Table 2
Catalytic results for the heterogeneous systems.
Catalyst
Time/h
Loading/mga
Conv.b
Selectivityb
eec
Cu–H
Cu–H
Cu–OMe
Cu–OMe
Cu–Me
Cu–Me
24
24
24
24
24
24
200
100
200
100
200
100
99
95
86
80
99
76
64
78
70
82
72
73
5 (R)
10 (R)
31 (R)
22 (R)
34 (R)
25 (R)
a
The benzaldehyde:nitromethane:NEt3 molar ratio was 1:10:0.25.
Conversion/selectivity as determined by 1H NMR analysis.
Determined by chiral HPLC, the absolute stereochemistry was determined by
b
c
Fig. 3. 13C{ H} CP/MAS solid-state NMR spectroscopic analysis for the heteroge-
1
optical rotation data.
neous supported systems. See Fig. 2 for the structures of A, B, C, D, E.

M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
13
Table 3
of nitromethane and benzaldehyde. Heterogeneous copper con-
taining catalysts were also prepared which also showed good
conversions but modest selectivities. Future work is aimed at
improving the selectivity of the heterogeneous systems and at
developing base free catalysed processes.
Catalytic results for the zeolite heterogeneous catalysts.
Catalysta
Time/h
Conv.b
Selectivityb
eec
Cu–H
Cu–OMe
Cu–Me
24
24
24
71
47
49
91
81
100
44 (S)
22 (S)
42 (S)
a
The catalyst:benzaldehyde:nitromethane molar ratio was 0.05:1:10, 35 ␮L of
Supporting information
NEt3 used loading of Cu-exchanged zeolite HY 0.36 g.
b
Conversion/selectivity as determined by 1H NMR spectroscopic analysis.
Determined by chiral HPLC, the absolute stereochemistry was determined by
Full experimental details (EPR, synthesis procedures and repre-
sentative NMRs) and crystal data in the .cif format are available.
CCDC numbers 751,962–751,966 contain the supplementary crys-
tallographic data for this paper and can be obtained free of
charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data request/cif.
c
optical rotation data.
Table 4
Catalytic results for the carbon heterogeneous catalysts.
b
Catalysta
Equiv. NEt3
Time/h
Conv.b
Selectivityb
eec
2+
Cu(2)2 on C
0.25
0.25
0.25
0.13
0.13
48
48
48
48
48
96
95
80
91
73
94
94
94
98
100
48 (S)
47 (S)
33 (S)
61 (S)
32 (S)
Acknowledgments
2+
Cu(3)2 on C
2nd use
2+
MDJ knowledges the EPSRC (EP/F037864) for funding his first
grant and for a PhD studentship to CJC. The EPSRC is also thanked
for funding for the solid-state NMR and EPR spectroscopic services,
respectively.
Cu(3)2 on C
2nd use
a
The benzaldehyde:nitromethane:NEt3 molar ratio was 1:10:0.25, 100 mg of het-
erogeneous catalyst used.
b
Conversion/selectivity as determined by 1H NMR spectroscopic analysis.
c
Determined by chiral HPLC, the absolute stereochemistry was determined by
Appendix A. Supplementary data
optical rotation data.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2010.03.013.
alysts are imines whereas the homogeneous are amines [15]. There
is also an increase in selectivity on reducing the reaction time and
catalyst loading and modest enantioselectivites were observed. The
residue from the Cu–Me catalyst with 200 mg for 24 h contained
14 ppm of Cu (from ICP measurements) in the product, indicating
that minimal leaching had occurred. Attempts to prepare and iso-
late the imine variants of the complexes proved unsuccessful. In fact
when the imine version of 2 was reacted with Cu(OTf)2 crystals of
Cu(1)2·(OTf)2 were isolated, implying that the imine functionality is
not stable under these conditions in the homogeneous phase. This
coupled with the low ee’s and selectivity of these heterogenised
catalysts led us to investigate the possibility of using other supports.
It has been demonstrated that Cu-exchanged zeolite HY with
bis(oxazoline) ligands are effective for the carbonyl-ene reaction
[55]. In an attempt to improve the selectivity in our Henry system
towards the nitroaldol product (as there would be no free amine
on the surface to aid the formation of the 1,3-dinitro alkanes), the
diamine ligands were heterogenised on copper exchanged zeolite
HY, Table 3 [55]. Compared to the silica supported system there
was a slight increase in both selectivity (in this case the major by-
product was ␤-nitrostyrene) and enantioselectivity. The residue
from the Cu–Me catalyst contained 12 ppm of Cu in the product,
implying the catalysis is heterogeneous. When the silica and zeo-
lite systems were reacted for 72 h, near quantitative conversions
were observed, although the selectivities decreased significantly.
It has been shown that simple catalysts can be anchored to car-
bon supports via ionic interactions [38]. Carbon was chosen as this
support has no Br␾nsted acid sites to facilitate the production of
the side products. The catalysts were prepared by stirring a slurry of
carbon with the copper complexes in methanol. High conversions
and selectivities were obtained and good enantioselectivities, even
with low amounts of NEt3, Table 4. On reuse of the catalyst, unfor-
tunately there was a significant reduction in both activity and ee,
implying that theses species may not be truly heterogeneous.
References
[1] F.A. Luzzio, Tetrahedron 57 (2001) 915–945.
[2] D. Seebach, A.K. Beck, T. Mukhopadhyay, E. Thomas, Helv. Chim. Acta 65 (1982)
1101–1133.
[3] S. Jammi, P. Saha, S. Sanyashi, S. Sakthivel, T. Punniyamurthy, Tetrahedron 64
(2008) 11724–11731.
[4] R. Kowalczyk, L. Sidorowicz, J. Skarzewski, Tetrahedron-Asymmetry 19 (2008)
2310–2315.
[5] M. Colak, T. Aral, H. Hosgoren, N. Demirel, Tetrahedron-Asymmetry 18 (2007)
1129–1133.
[6] S.K. Ginotra, V.K. Singh, Org. Biomol. Chem. 5 (2007) 3932–3937.
[7] M. Sedlak, P. Drabina, R. Keder, J. Hanusek, I. Cisarova, A. Ruzicka, J. Organomet.
Chem. 691 (2006) 2623–2630.
[8] Y. Xiong, F. Wang, X. Huang, Y.H. Wen, X.M. Feng, Chem. Eur. J. 13 (2007)
829–833.
[9] C.S. Gan, G.Y. Lai, Z.H. Zhang, Z.Y. Wang, M.M. Zhou, Tetrahedron-Asymmetry
17 (2006) 725–728.
[10] D.A. Evans, D. Seidel, M. Rueping, H.W. Lam, J.T. Shaw, C.W. Downey, J. Am.
Chem. Soc. 125 (2003) 12692–12693.
[11] Y.D. Zhang, L. Xiang, Q.W. Wang, X.F. Duan, G.F. Zi, Inorg. Chim. Acta 361 (2008)
1246–1254.
[12] Q.T. Nguyen, J.H. Jeong, Polyhedron 25 (2006) 1787–1790.
[13] Q.T. Nguyen, J.H. Jeong, Polyhedron 27 (2008) 3227–3230.
[14] G.Q. Zhang, E. Yashima, W.D. Woggon, Adv. Synth. Catal. 351 (2009) 1255–1262.
[15] M. Bandini, F. Piccinelli, S. Tommasi, A. Umani-Ronchi, C. Ventrici, Chem. Com-
mun. (2007) 616–618.
[16] C. Christensen, K. Juhl, K.A. Jorgensen, Chem. Commun. (2001) 2222–2223.
[17] G. Blay, E. Climent, I. Fernandez, V. Hernandez-Olmos, J.R. Pedro, Tetrahedron-
Asymmetry 17 (2006) 2046–2049.
[18] G. Blay, E. Climent, I. Fernandez, V. Hernandez-Olmos, J.R. Pedro, Tetrahedron-
Asymmetry 18 (2007) 1603–1612.
[19] G. Blay, L.R. Domingo, V. Hernandez-Olmos, J.R. Pedro, Chem. Eur. J. 14 (2008)
4725–4730.
[20] G. Blay, V. Hernandez-Olmos, J.R. Pedro, Chem. Commun. (2008) 4840–4842.
[21] H. Maheswaran, K.L. Prasanth, G.G. Krishna, K. Ravikumar, B. Sridhar, M.L. Kan-
tam, Chem. Commun. (2006) 4066–4068.
[22] E.C. Constable, G.Q. Zhang, C.E. Housecroft, M. Neuburger, S. Schaffner, W.D.
Woggon, N. J. Chem. 33 (2009) 1064–1069.
[23] A. Bulut, A. Aslan, O. Dogan, J. Org. Chem. 73 (2008) 7373–7375.
[24] S.F. Lu, D.M. Du, J.X. Xu, S.W. Zhang, J. Am. Chem. Soc. 128 (2006) 7418–7419.
[25] B.M. Trost, V.S.C. Yeh, Angew. Chem. Int. Ed. Engl. 41 (2002) 861.
[26] R. Kowalczyk, L. Sidorowicz, J. Skarzewski, Tetrahedron-Asymmetry 18 (2007)
2581–2586.
4. Conclusions
[27] A. Zulauf, M. Mellah, E. Schulz, J. Org. Chem. 74 (2009) 2242–2245.
[28] H. Sasai, T. Suzuki, S. Arai, T. Arai, M. Shibasaki, J. Am. Chem. Soc. 114 (1992)
4418–4420.
[29] Y. Kogami, T. Nakajima, T. Ikeno, T. Yamada, Synthesis – Stuttgart (2004)
1947–1950.
Five Cu(II) homogeneous complexes have been prepared and
characterised via single crystal X-ray diffraction studies. The com-
plexes were shown to be active for the asymmetric coupling

14
M.D. Jones et al. / Journal of Molecular Catalysis A: Chemical 325 (2010) 8–14
[30] M. Bandini, M. Benaglia, R. Sinisi, S. Tommasi, A. Umani-Ronchi, Org. Lett. 9
(2007) 2151–2153.
[44] P. Comba, T.W. Hambley, M.A. Hitchman, H. Stratemeier, Inorg. Chem. 34 (1995)
3903–3911.
[31] M. Gaab, S. Bellemin-Laponnaz, L.H. Gade, Chem. Eur. J. 15 (2009) 5450–5462.
[32] T. Kehat, M. Portnoy, Chem. Commun. (2007) 2823–2825.
[33] J.F. Larrow, E.N. Jacobsen, Y. Gao, Y.P. Hong, X.Y. Nie, C.M. Zepp, J. Org. Chem.
59 (1994) 1939–1942.
[45] M. Becker, F.W. Heinemann, F. Knoch, W. Donaubauer, G. Liehr, S. Schindler, G.
Golub, H. Cohen, D. Meyerstein, Eur. J. Inorg. Chem. (2000) 719–726.
[46] S. Abry, A. Thibon, B. Albela, P. Delichere, F. Banse, L. Bonneviot, N. J. Chem. 33
(2009) 484–496.
[34] T. Kylmala, N. Kuuloja, Y.J. Xu, K. Rissanen, R. Franzen, Eur. J. Org. Chem. (2008)
4019–4024.
[47] T. Marcelli, R.N.S. van der Haas, J.H. van Maarseveen, H. Hiemstra, Angew.
Chem., Int. Ed. Engl. 45 (2006) 929–931.
[35] M.D. Jones, M.F. Mahon, J. Organomet. Chem. 693 (2008) 2377–2382.
[36] M.D. Jones, M.G. Davidson, C.G. Keir, L.M. Hughes, M.F. Mahon, D.C. Apperley,
Eur. J. Inorg. Chem. (2009) 635–642.
[37] K.A. Utting, D.J. Macquarrie, N. J. Chem. 24 (2000) 591–595.
[38] C.F.J. Barnard, J. Rouzaud, S.H. Stevenson, Org. Process Res. Dev. 9 (2005)
164–167.
[48] A.P. Bhatt, K. Pathak, R.V. Jasra, R.I. Kureshy, N.U.H. Khan, S.H.R. Abdi, J. Mol.
Catal. A: Chem. 244 (2006) 110–117.
[49] B.M. Choudary, M.L. Kantam, B. Kavita, J. Mol. Catal. A: Chem. 169 (2001)
193–197.
[50] T. Hara, S. Kanai, K. Mori, T. Mizugaki, K. Ebitani, K. Jitsukawa, K. Kaneda, J. Org.
Chem. 71 (2006) 7455–7462.
[39] A.P. Cole, V. Mahadevan, L.M. Mirica, X. Ottenwaelder, T.D.P. Stack, Inorg. Chem.
44 (2005) 7345–7364.
[40] E. Coronado, C. Gimenez-Saiz, A. Nuez, V. Sanchez, F.M. Romero, Eur. J. Inorg.
Chem. (2003) 4289–4293.
[41] C. Pariya, F.L. Liao, S.L. Wang, C.S. Chung, Polyhedron 17 (1998) 547–554.
[42] C. Pariya, K. Panneerselvan, C.S. Chung, T.H. Lu, Polyhedron 17 (1998)
2555–2561.
[43] S. Sen, M.K. Saha, S. Mitra, A.J. Edwards, W. Clegg, Polyhedron 19 (2000)
1881–1885.
[51] K. Komura, T. Kawamura, Y. Sugi, Catal. Commun. 8 (2007) 644–648.
[52] K. Motokura, M. Tada, Y. Iwasawa, Angew. Chem., Int. Ed. Engl. 47 (2008)
9230–9235.
[53] S.L. Poe, M. Kobaslija, D.T. McQuade, J. Am. Chem. Soc. 128 (2006) 15586–15587.
[54] R. Ballini, L. Barboni, D. Fiorini, G. Giarlo, A. Palmieri, Chem. Commun. (2005)
2633–2634.
[55] N.A. Caplan, F.E. Hancock, P.C.B. Page, G.J. Hutchings, Angew. Chem., Int. Ed.
Engl. 43 (2004) 1685–1688.