Selective nitroaldol condensations over heterogeneous catalysts in the presence of supercritical carbon dioxide

Article abstract of DOI:10.1021/jo801650p

(Chemical Equation Presented) At 40-60°C, in the presence of heterogeneous catalysts based on Al₂O₃, supercritical carbon dioxide not only acts as a good solvent for the reaction of aromatic and aliphatic aldehydes with 1-nitroalkanes but, most importantly, it also allows the selectivity to be tuned between the competitive formation of β-nitroalcohols and nitroalkenes (from the Henry reaction and the nitroaldol condensation, respectively). In particular, when the pressure (and the density) of the supercritical phase is enhanced from 80 to 140 bar, the nitroalkene's selectivity increases, on average, from ～60 to >90%. Experiments show that, in the same pressure range, a steep increase of the concentration profiles of reactant aldehydes takes place. By contrast, under solvent-free conditions, the reaction usually proceeds with a higher conversion, but nitroalkanols are the major products.

Full text of DOI:10.1021/jo801650p

Selective Nitroaldol Condensations over Heterogeneous Catalysts in
the Presence of Supercritical Carbon Dioxide
Roberto Ballini,† Marco Noe`,‡ Alvise Perosa,‡ and Maurizio Selva*,‡
Dipartimento di Scienze Chimiche dell’UniVersita` di Camerino, Via Sant’Agostino 1-62032 Camerino,
Italy, and Dipartimento di Scienza Ambientali dell’UniVersita` Ca’ Foscari, Calle Larga S. Marta
2137-30123 Venezia, Italy
selVa@uniVe.it
ReceiVed July 25, 2008
At 40-60 °C, in the presence of heterogeneous catalysts based on Al2O3, supercritical carbon dioxide
not only acts as a good solvent for the reaction of aromatic and aliphatic aldehydes with 1-nitroalkanes
but, most importantly, it also allows the selectivity to be tuned between the competitive formation of
ꢀ-nitroalcohols and nitroalkenes (from the Henry reaction and the nitroaldol condensation, respectively).
In particular, when the pressure (and the density) of the supercritical phase is enhanced from 80 to 140
bar, the nitroalkene’s selectivity increases, on average, from ∼60 to >90%. Experiments show that, in
the same pressure range, a steep increase of the concentration profiles of reactant aldehydes takes place.
By contrast, under solvent-free conditions, the reaction usually proceeds with a higher conversion, but
nitroalkanols are the major products.
SCHEME 1
Introduction
The reaction of nitroalkanes with carbonyl derivatives is
amongthekeysynthetictoolsfortheconstructionofcarbon-carbon
bonds in organic chemistry.1 The reaction is typically carried
out under basic conditions, to allow the formation of ꢀ-nitroal-
kanols. However, when primary nitroalkanes (1) are involved,
two competitive processes may occur (Scheme 1): the formation
of(i)nitroalkanols(3,patha)throughtheso-callednitroaldol-Henry
reaction and of (ii) conjugated nitroalkenes (4, path b) via the
nitroaldol condensation.
Both products 3 and 4 are of great synthetic interest.
ꢀ-Nitroalcohols undergo a variety of chemical transformations
for the synthesis of ꢀ-aminoalkanols, R-hydroxyketones, R-ni-
troketones, and other synthetic intermediates,1,2 while nitroalk-
enes are good Michael acceptors and dienophiles in Diels-Alder
reactions,3 as well as versatile starting materials to prepare
different functionalities.1c,3,4 Compounds 4 are also recognized
for their biological activity.5
* Author to whom correspondence should be addressed. Phone: +39 041
236 8687. Fax: +39 041 236 8584.
In the most classical arrangement, the nitroaldol reactions
are performed under homogeneous conditions, in the presence
† Universita` di Camerino.
‡ Universita` Ca’ Foscari.
(1) (a) Rosini, G. In ComprehensiVe Organic Synthesis; Trost, B. M., Ed.;
Pergamon: Oxford, 1999; Vol. 1. (b) Luzzio, F. A. Tetrahedron 2001, 57, 915–
945. (c) Ono, N. In The Nitro Group in Organic Synthesis; Feuer, H., Ed.; Wiley-
VCH: New York, 2001.
(2) Ballini, R.; Bosica, G.; Fiorini, D.; Palmieri, A. Tetrahedron 2005, 61,
8971–8993.
(3) (a) Kabalka, G. W.; Guindi, L. H. M.; Varma, R. S. Tetrahedron 1990,
46, 7443–7457. (b) Barret, A. G. M. Chem. Soc. ReV. 1991, 20, 95–127. (c)
Barrett, A. G. M.; Graboski, G. G. Chem. ReV. 1986, 86, 751–762. (d) Perekalin,
V. V.; Lipina, E. S.; Berestovitskaya, V. M.; Efremov, D. A. Nitroalkenes; John
Wiley & Sons: Chichester, UK, 1994.
(4) Ballini, R.; Bosica, G. Synthesis 1994, 723–726.
8520 J. Org. Chem. 2008, 73, 8520–8528
10.1021/jo801650p CCC: $40.75 2008 American Chemical Society
Published on Web 10/08/2008

SelectiVe Nitroaldol Condensations
SCHEME 2
of a base catalyst in an organic solvent. Alternatively, hetero-
geneous catalysts have been introduced which allow for solvent-
free synthesis.6 Whatever the methodology, the selective
formation of ꢀ-nitroalkanols or of conjugate nitroalkenes very
often requires a careful choice of reaction parameters (temper-
ature, reaction time, basic strength of catalyst, etc.). Particularly,
the preparation of nitroalkenes, which usually needs stronger
conditions,1,4 implies multistep procedures.
As a part of our research program on eco-friendly carbon-
carbon bond forming methodologies,7 we decided to investigate
the activity and, most of all, the selectivity of nitroaldol
reactions/condensations in the presence of supercritical carbon
dioxide as the solvent. This is a largely unexplored area: to the
best of our knowledge, only one recent paper has been reported
on the reaction of p-cyanobenzaldehyde with 1-nitropropane
carried out at 40 °C, under different CO2 pressures, including
the subcritical state (Scheme 2).8
zeolites, supported acids, and metals immobilized on organic/
inorganic matrixes such as Deloxan, Amberlyst, alumina, and
silica. It should be noted here that basic Al2O3 as such or mixed
with KF is a convenient heterogeneous catalyst also for
nitroaldol reactions.16
These observations in conjunction with our interest for green
synthetic methods using scCO217 have inspired the present work.
We wish to report here that, in the presence of heterogeneous
catalysts, preferably based on or supported on basic Al2O3, the
reaction of aromatic and aliphatic aldehydes with both nitroet-
hane and 1-nitrohexane takes place efficiently in dense carbon
dioxide; even more intriguingly, the process is extremely
selective toward the formation of the corresponding nitroalkenes
(4; Scheme 1, path b). This study also provides good evidence
that the reaction selectivity is tuned by the CO2 pressure, and
that the overall reaction outcome is strictly related to the trend
of the solubility of aldehydes in the supercritical medium.
The use of supercritical CO2 (scCO2) was claimed to induce
a dramatic shift of the stereoselectivity of the Henry reaction,
with respect to the syn/anti ratio usually observed in more
conventional conditions (solventless, toluene, or MeCN as
solvents). All tests were referred to a homogeneous catalyst such
as Et3N.9 Supercritical CO2, however, due to the low viscosity
(η) and the high diffusivity (D) (0.01-0.03 mPa·s and ∼0.07
× 10-6 m2 s-1, respectively),10 is very efficient at penetrating
meso- and microporous structures which are typical of the
conventional supports used in the heterogeneous catalysis.
Mostly for this reason, the use of scCO2 as a solvent/carrier is
reported to improve the mass transfer (and the reaction rate/
selectivity) of a variety of different processes, including
alkylations,10,11 etherifications and esterifications,12 hydrogena-
tions and hydroformylations,13,14 and oxidations15 catalyzed by
Results
Aromatic Aldehydes. Initial experiments were carried out
using both benzaldehyde (2a) and 4-nitrobenzaldehyde (2b) as
model substrates. The relative amounts of reactants and of the
catalyst were chosen according to procedures already described
by us.2 A mixture of the aldehyde (2a: 152 µL, 1.5 mmol, or
2b: 0.23 g, 1.5 mmol), RCH2NO2 (1a: R ) CH3, 130 µL, 1.8
mmol, or 1b: R ) CH3(CH2)4, 260 µL, 1.8 mmol), and
commercial basic Al2O3 (0.3-0.45 g) was charged in a 30 mL
autoclave, which was then loaded with CO2 at 100 bar. The
reactor was electrically heated at the desired temperature (40
or 60 °C), and the mixture was kept under magnetic stirring.
After 18 h, the autoclave was cooled to rt and slowly purged (1
h). The mixture was then analyzed by 1H NMR, which showed
products from both the Henry reaction and the nitroaldol
condensation (Scheme 3).18
(5) (a) Brown, A. W. A.; Robinson, D. B. W.; Hurtig, H.; Wenner, B. J.
Can. J. Res. 1948, 26D, 177–187. (b) Bousquet, E. W.; Kirby, J. E.; Searle,
N. E. U.S. Patent 2,335,384, 1943, 38, 2834; Chem. Abstr. 1944, 38, 2834. (c)
Brian, P. W.; Grove, J. F.; McGowan, J. C. Nature 1946, 158, 876–877. (d)
McGowan, J. C.; Brian, P. W.; Hemming, H. G. Ann. Appl. Biol. 1948, 35,
25–36. (e) Bocobo, F. C.; Curtis, A. C.; Block, W. D.; Harrel, E. R.; Evans,
E. E.; Haines, R. F. Antibiot. Chemother. 1956, 6, 385–390. (f) Schales, O.;
Graefe, H. A. J. Am. Chem. Soc. 1952, 74, 4486–4490. (g) Dann, O.; Moller,
F. F. Chem. Ber. 1949, 82, 76–92. (h) Karker, R. J. U.S. Patent 2,889,246, 1959;
Chem. Abstr. 1959, 53, 17414i. (i) Zee-Cheng, K.; Cheng, C. J. Med. Chem.
1969, 12, 157–161.
The same reactions were also performed under solvent-free
conditions (solFC) in a conventional 5 mL glass reactor
thermostatted by an oil bath.
(14) (a) Tadd, A. R.; Marteel, A.; Manson, M. R.; Davies, J. A.; Abraham,
M. A. Ind. Eng. Chem. Res. 2002, 41, 4514–4522. (b) Marteel, A. E.; Tack,
T. T.; Bektesevic, S.; Davies, J. A.; Manson, M. R.; Abraham, M. A. EnViron.
Sci. Technol. 2003, 37, 5424–5431.
(6) Ballini, R.; Palmieri, A. Curr. Org. Chem. 2006, 10, 2145–2169.
(7) Project “Ecofriendly organic syntheses mediated by new catalytic systems”
funded by MUR (Italian Ministry of Unversity and Research); grant #
2006031724_002.
(8) (a) Parratt, A. J.; Adams, D. J.; Clifford, A. A.; Rayner, C. M. J. Chem.
Soc., Chem. Commun. 2004, 2720–2721. (b) Rayner, C. M. Org. Process Res.
DeV. 2007, 11, 121–132.
(9) Et3N is very soluble in scCO2: Jessop, P. G.; Hsiao, Y.; Ikariya, T.;
Noiory, R. J. Am. Chem. Soc. 1996, 118, 344–355.
(15) Fan, L.; Watanabe, T.; Fujimoto, K. Appl. Catal. 1997, 158, L41-L46.
(16) Ballini, R.; Bosica, G.; Forconi, P. Tetrahedron 1996, 52, 1677–1684.
(17) (a) DeSimone, J.; Selva, M.; Tundo, P. J. Org. Chem. 2001, 66, 4047–
4049. (b) Selva, M.; Tundo, P.; Perosa, A. Tetrahedron Lett. 2002, 43, 1217–
1219. (c) Loris, A.; Perosa, A.; Selva, M.; Tundo, P. J. Org. Chem. 2003, 68,
4046–4051. (d) Selva, M.; Tundo, P.; Perosa, A.; Dall’Acqua, F. J. Org. Chem.
2005, 70, 2771–2777. (e) Tundo, P.; Loris, A.; Selva, M. Green Chem. 2007, 9,
777–779. (f) Maggi, R.; Bertolotti, C.; Orlandini, E.; Oro, C.; Sartori, G.; Selva,
M. Tetrahedron Lett. 2007, 48, 2131–2134.
(10) (a) Gross, T.; Chen, L.; Lu¨demann, H.-D. In Supercritical Fluids as
SolVents and Reaction Media; Brunner, G., Ed.; Elsevier: Amsterdam, 2004;
Chapter 2.1. (b) Baiker, A. Chem. ReV. 1999, 99, 453–473.
(11) Hitzler, M. G.; Smail, F. R.; Ross, S. K.; Poliakoff, M. Chem. Commun.
1998, 359–360.
(12) (a) Gray, W. K.; Smail, F. R.; Hitzler, M. G.; Ross, S. K.; Poliakoff,
M. J. Am. Chem. Soc. 1999, 121, 10711–10718. (b) Vieville, C.; Mouloungui,
Z.; Gaset, A. Ind. Eng. Chem. Res. 1993, 32, 2065–2068.
(18) Although many methods for the preparation of nitroalkenes are known,
they generally produce (E)-nitroalkenes which are difficult to convert into the
corresponding Z-isomers. At most, mixtures of the E- and Z-products are obtained,
where (Z)-olefins are always minor products (see ref 3 and: (a) Knochel, P.;
Seebach, D. Synthesis 1982, 1017–1018. (b) Ono, N.; Kamimura, A.; Kawai,
T.; Kaji, A. J. Chem. Soc., Chem. Commun. 1987, 1550–1551. (c) Stanetty, P.;
Kremslehner, M. Tetrahedron Lett. 1998, 39, 811–812. (d) Fioravanti, S.;
Pellacani, L.; Tardella, P. A.; Vergari, M. C. Org. Lett. 2008, 10, 1449–1451.
In Scheme 3, (E)-nitroolefins are indicated (characterization data are reported
in Supporting Information).
(13) Hyde, J. R.; Licence, P.; Carter, D.; Poliakoff, M. Appl. Catal. A 2001,
222, 119–131.
J. Org. Chem. Vol. 73, No. 21, 2008 8521

Ballini et al.
SCHEME 3
Bearing this in mind, two sets of experiments were devised:
(i) in the first, the reactions of aldehydes 2a,b with nitroethane
catalyzed by basic Al2O3 were examined at 60 °C in scCO2 by
varying the pressure of the supercritical phase over the range
of 80-150 bar (dCO ) 0.19-0.56 g/mL); (ii) in the second,
2
operating at the same temperature (60 °C) and within the same
pressure interval, the concentration of both compounds 2a,b in
scCO2 was measured. In particular, reactions were performed
according to the procedure defined in Table 1. Concentration
measurements were carried out in a 90 mL stainless steel
autoclave equipped with a calibrated loop and needle valves
for sample withdrawals; the reactor was loaded with the
aldehyde (2a: 2.5 g; 2b: 1.0 g) and CO2 at 80 to 150 bar.23
The relatively low boiling point of nitroethane (114-115 °C)
did not allow reproducible measures of its concentration in the
supercritical medium. However, the visual inspection of mixtures
Results are reported in Table 1.
Three main aspects were recognized from the analysis of
Table 1: (i) At both 40 and 60 °C, conversions of both aldehydes
(2a,b) were usually lower in supercritical CO2 with respect to
the solvent-free conditions (compare entries 1-2, 3-4, 5-6,
7-8, and 9-10). This difference substantially disappeared for
reactions of 4-nitrobenzaldehyde at 60 °C that showed rather
similar conversions (76-77%) regardless of the medium and
of the nature of the nitroalkane (1a or 1b) used (entries 11-14).
On average, conversions were higher at 60 °C than at 40 °C.19
(ii) Compound 2b was more reactive than 2a. Under the same
conditions, the conversion of benzaldehyde was 1.5-3 times
lower than that of 4-nitrobenzaldehyde (compare entries 1-4
to 9-12, and 7-8 to 13-14). (iii) The reaction selectivity (S%)
toward nitroalkenes 4a,b′ was increased considerably when
experiments were run under CO2 pressure rather than in solvent-
free conditions. Remarkable cases were observed for both
aldehydes; for example, in the reaction of benzaldehyde, S%
for compounds 4a,a′ went from 49 to 79% and from 55 to 92%,
respectively (entries 3-4 and 7-8). Likewise, at comparable
high conversions of 4-nitrobenzaldehyde (76-77%), the reaction
selectivity (S% for compounds 4b,b′) increased from 39 to 86%
and from 36 to 62%, respectively (entries 11-12 and 13-14).
To continue the investigation, experimental conditions were
set according to these preliminary results. In particular, to reach
reasonably high conversions, a temperature of 60 °C20 and the
use of 1-nitroethane as the nucleophile were chosen. Other
possible effects of supercritical CO2 on the reaction selectivity
were then analyzed.
Effect of CO2 Pressure on Reactivity and Solubility of
Aromatic Aldehydes. Supercritical carbon dioxide is often
described as a tunable solvent;21 in fact, its solvating ability is
strictly related to its physicochemical properties (density and
viscosity), which can be widely adjusted through small variations
of pressure and temperature. For example, at 35 °C, the density
and viscosity of CO2 practically triple (d: from 0.28 to 0.81
g/mL, η: from 0.022 to 0.075 mPa ·s) when the pressure is
increased from 75 to 150 bar.22
24
of nitroethane/scCO2 revealed that homogeneous solutions
occurred at P g 90 bar.
Figures 1 and 2 report the result for benzaldedyde and
4-nitrobenzaldehyde, respectively. For a convenient comparison,
each figure displays on a single plot the trend of aldehyde
conversion and reaction selectivity (S% for nitroalkenes 4a,b′)
(black and red profiles) and of the molar concentration of
aldehydes 2a,b in scCO2 (blue profile) versus the pressure of
the supercritical phase. Data of reactions carried out under
solvent-free conditions (P
) 0) are also indicated.
CO2
Results of Figures 1 and 2 confirmed the data of Table 1.
For both reactions, the increase of the CO2 pressure had opposite
effects on the conversion of aldehydes and on the selectivity:
the first (conversion) always dropped, while the formation of
products 4a,b became more and more favored.
This was particularly manifest for benzaldehyde (Figure 1);
as the pressure was increased from 80 to 140 bar, the substrate
conversion decreased from 55 to 25% (black curve). At the same
time, the selectivity of the nitroaldol condensation (compound
4a) was dramatically improved from 50 to 92% (red curve).25
It should be noted that, in the same pressure range, the molar
concentration of benzaldehyde dissolved in scCO2 increased by
nearly 11-fold (blue curve), and at 140 bar, it approximated
the experimental limit value (0.26 M).23
In the case of 4-nitrobenzaldehyde (2b, Figure 2), the increase
of the CO2 pressure resulted in modest variations of both
conversion and selectivity, which were reduced and enhanced,
respectively, from 76 to 50%, and from 81 to 88% (black and
red curves).25 Under solvent-free conditions, however, the
selectivity for 4b was only 39% (see also entry 11, Table 1).
The molar concentration of aldehyde 2b in scCO2 was also
augmented by the increase of pressure (up to 3.3 × 10-2 M at
150 bar), though it remained well below the expected maximum
limit of 7.5 × 10-2 M.23
This trend of reactivity/selectivity was corroborated by
additional experiments in which 4-chlorobenzaldehyde (2c) was
(19) A number of organic reactions including condensations and dehydrative
condensation processes can be catalyzed/promoted by Al2O3, and very different
catalyst loadings are reported; see, for example: Kabalka, G. W.; Pagni, R. M.
Tetrahedron 1997, 53, 7999–8065. In this work, Al2O3 was used in a two-fold
weight excess with respect to aldehydes to allow sufficiently fast reactions.
However, in a single reaction of benzaldehyde and nitroethane carried out under
the conditions of entry 6 in Table 1, Al2O3 was reduced to 1/3 of its original
amount (weight ratio Al2O3:2a ) 0.67); the nitroaldol condensation proceeded
albeit with a lower conversion (28% after 18 h; S% ) 82). This indicates that
Al2O3 can be referred to as a catalyst.
(23) Concentration measurements in scCO2 were carried out through gravi-
metric analyses which required rather large amounts (1-2.5 g) of aldehydes
(details are in the Experimental Section). For this reason, the autoclave (90 mL)
used for these tests was bigger than that used for reactivity experiments. Under
such conditions, if both compounds 2a,b were completely dissolved in the
supercritical phase, then the maximum expected molar concentrations would
have been 0.26 and 0.075 mol/L for benzaldehyde and 4-nitrobenzaldehyde,
respectively.
(20) For the same reason, the reaction of 4-nitrobenzaldehyde with 1-nitro-
hexane was performed only at 60 °C (entries 13 and 14, Table 1).
(21) Chemical Synthesis Using Supercritical Fluids; Jessop, P. G., Leitner,
W., Eds.;Wiley-VCH: Weinheim, Germany,1999; Chapter 1.
(24) Both the top and the bottom of the autoclave were equipped with sapphire
windows, by which the content of the reactor could be examined at any given
pressure.
(25) Since no measures were made between 0 and 80 bar, dotted lines
indicated both conversion and selectivity trends.
(22) http://webbook.nist.gov.
8522 J. Org. Chem. Vol. 73, No. 21, 2008

SelectiVe Nitroaldol Condensations
a
TABLE 1. Reactions of Aldehydes 2a,b with Both 1-Nitroethane and 1-Nitrohexane in the Presence of Basic Al2O3 and scCO2
products (%)
b
entry
aldehyde
RCH2NO2
T (°C)
PCO (bar)c
conv (%)d
3e
4
sel (%)f
2
1
2
3
4
5
6
7
8
2a: X ) H
1a: R ) Me
40
40
60
60
40
40
60
60
40
40
60
60
60
60
0
100
0
100
0
100
0
100
0
100
0
100
0
38
16
55
42
34
5
42
25
82
47
76
77
76
76
20
5
28
9
17
2
19
3
72
17
45
11
49
29
18
11
27
33
17
3
23
23
10
30
30
66
27
47
48
69
49
79
50
60
55
92
14
64
39
86
36
62
2a: X ) H
1a: R ) Me
2a: X ) H
1a: R ) Me
1a: R ) Me
2a: X ) H
2a: X ) H
1b: R ) n-C5H11
1b: R ) n-C5H11
1b: R ) n-C5H11
1b: R ) n-C5H11
1a: R ) Me
2a: X ) H
2a: X ) H
2a: X ) H
9
2b: X ) NO2
2b: X ) NO2
2b: X ) NO2
2b: X ) NO2
2b: X ) NO2
2b: X ) NO2
10
11
12
13
14
1a: R ) Me
1a: R ) Me
1a: R ) Me
1b: R ) n-C5H11
1b: R ) n-C5H11
100
a All reactions were carried out for 18 h in the presence of basic Al2O3 as a catalyst (the weight ratio of Al2O3:aldehyde was 2). Each experiment was
repeated twice; the values for conversions, selectivities, and yields were the average of two runs that did not differ more than 5-7%. b The molar ratio
of 1a,b:2a,b was 1.2. c Odd entries (1, 3, 5, 7, 9, 11, and 13): reactions carried out under solvent-free conditions. d The conversion was referred to the
aldehyde; it was determined by 1H NMR. e Total amounts of syn/anti isomers (determined by 1H NMR). f Selectivity (%) toward the formation of
nitroalkenes 4a,b′. This was calculated by the following expression: sel (%) ) [4/(3 + 4)] × 100, where the amounts of compounds 3 (total of syn/anti
1
isomers) and 4 were estimated from H NMR spectra.
FIGURE 2. Left ordinate: the effect of CO2 pressure on the conversion
FIGURE 1. Left ordinate: the effect of CO2 pressure on the conversion
of 4-nitrobenzaldehyde and on the selectivity (S% for compound 4b)
of benzaldehyde and on the selectivity (S% for compound 4a) of the
of the reaction of 2b with nitroethane (black and red curves) at 60 °C.
reaction of 2a with nitroethane (black and red curves) at 60 °C. Right
ordinate: molar concentration of benzaldehyde in scCO2 at different
pressures and at 60 °C (blue curve).
Right ordinate: molar concentration of 4-nitrobenzaldehyde in scCO2
at different pressures and at 60 °C (blue curve).
dehyde (2d), 2-phenylpropionaldehyde (2e), 3-phenylbutyral-
set to react under the conditions of Figures 1 and 2 (2c: 0.21 g,
1.5 mmol; nitroethane: 130 µL, 1.8 mmol; basic Al2O3: 0.42 g;
T ) 60 °C, t ) 18 h; CO2: 80-140 bar). Results are reported
in Figure 3.
The behavior was rather similar to that of benzaldehyde; in
the range of 80-140 bar, the conversion of 2c dropped from
56 to 13%, while the selectivity for 4c [4-ClC6H4CHd
C(NO2)CH3: E-isomer18] increased from 65 to 91%.25 The
concentration of 4-chlorobenzaldehyde in the supercritical phase
was not measured; though, at 60 °C, mixtures of 2c/scCO2 gave
homogeneous solutions at P g 120 bar.24
dehyde (2f), n-decylaldehyde (2g), and n-heptylaldehyde (2h)
were investigated under the conditions of Table 1. Some
preliminary experiments showed that aliphatic aldehydes were
more reactive than aromatic ones. For this reason, a complete
set of reactivity data of compounds 2d-h was gathered only at
40 °C. A mixture of aldehyde (1.5 mmol), CH3CH2NO2 (1.8
mmol), and basic Al2O3 (the weight ratio of Al2O3:2 was 2)
was charged in a 30 mL autoclave, which was then loaded with
CO2 at 100 bar. After 18 h at 40 °C, the autoclave was cooled
to room temperature and slowly purged (1 h). The reaction
1
mixture was analyzed by H NMR.
In all cases, NMR analyses showed that the diastereomeric
excess (de, anti) in the formation of nitroaldol isomers of
compounds 2a-c was not substantially affected by the CO2
pressure; de varied from ∼30 to ∼40% on going from 80 to
140 bar. Under solvent-free conditions, it was only slightly
higher (45%).
The same reactions were also performed under solvent-free
conditions (solFC) in a conventional 5 mL glass reactor
thermostatted by an oil bath.
The Henry reaction and the nitroaldol condensation were the
sole observed processes for all substrates 2d-h (Scheme 4).
Nitroalkenes (4d-h) were obtained as E-isomers.18
Results are reported in Table 2. In the table, the selectivity
(%) was expressed toward the formation of nitroalkenes 4d-h.
Aliphatic Aldehydes. In the presence of basic Al2O3 as a
catalyst, the reactions of nitroethane with 3-phenylpropional-
J. Org. Chem. Vol. 73, No. 21, 2008 8523

Ballini et al.
TABLE 2. Reactions of Aldehydes 2d-h with 1-Nitroethane in the
a
Presence of Basic Al2O3 and scCO2
FIGURE 3. The effect of CO2 pressure on the conversion of 4-chlo-
robenzaldehyde (2c) and on the selectivity (S% for compound 4c) in
the reaction with nitroethane at 60 °C.
a All reactions were carried out for 18 h at 40 °C, in the presence of
basic Al2O3 as a catalyst (the weight ratio of Al2O3:aldehyde was 2).
Entries 1-6 and 8-10: experiments were repeated twice; the values for
conversions, nitroalkene amount, and selectivity were the average of two
runs that did not differ more than 5-7%. Entries 7-9: experiments
were run once. The molar ratio of nitroethane:2d-h was 1.2. b Odd
entries (1, 3, 5, 7, and 9): reactions carried out under solvent-free
conditions. d The conversion referred to the aldehyde; it was determined
by 1H NMR. e Amounts of nitroalkenes 4d-h (determined by 1H
NMR). f Selectivity (%) toward the formation of nitroalkenes 4d-h.
This was calculated by the following expression: sel (%) ) [4/(3 + 4)]
× 100, where quantities of compounds 3 (total of syn/anti nitroaldol
SCHEME 4. Aliphatic Aldehydes Used in the Reaction with
Nitroethane and Basic Al2O3
1
isomers) and 4 were estimated from H NMR spectra.
The presence of dense CO2 had the same twofold effect
observed for aromatic aldehydes. The conversion of aliphatic
aldehydes was generally lower (44-60%) in the supercritical
medium with respect to solvent-free conditions (85-93%). By
contrast, the selectivity for products 4d-h was much higher in
CO2 (84-97%) than in its absence (30-74%) (compare entries
2, 4, 6, 8, and 10 to 1, 3, 5, 7, and 9).
The effects of the CO2 pressure were further investigated by
studying the reactions of 3-phenylpropanal (2d) and n-hepty-
laldehyde (2g) with nitroethane at 40 °C in the range of 70-120
bar (dCO ) 0.19-0.72 g/mL). In the same pressure interval,
2
also the concentration of 2d (2.5 g) in scCO2 was measured
through the procedure described above for aromatic aldehydes.
Figures 4 and 5 report the result for compounds 2d and 2g,
respectively. The trend of aldehyde conversion and reaction
selectivity (S% for nitroalkenes 4a and 4g) (black and red
curves) and of the molar concentration of 2d in scCO2 (blue
profile) are shown against the pressure of the supercritical phase.
FIGURE 4. Left ordinate: the effect of CO2 pressure on the conversion
of 3-phenylpropionaldehyde (2d) and on the selectivity (S% for
compound 4d) in the reaction with nitroethane (black and red curves)
at 40 °C. Right ordinate: molar concentration of 2d in scCO2 at different
pressures and at 40 °C (blue curve).
Data of reactions carried out under solvent-free conditions (PCO
) 0) are also indicated.
for aromatic aldehydes (90-120 bar, dCO ) 0.23-0.44 g/mL
2
2
at 60 °C; Figures 1 and 3). In addition, the concentration profile
of 3-phenylpropionaldehyde in scCO2 (blue curve, 40 °C, Figure
4) appeared shifted by nearly 20 bar with respect to that obtained
for benzaldehyde (blue curve, 60 °C, Figure 1). At 120 bar, the
concentration limit value of 2d was reached (0.21 M).22
Different Catalysts. The reactions of nitroethane with
benzaldehyde (2a), 4-nitrobenzaldehyde (2b), and 3-phenyl-
propionaldehyde (2d) were explored also in the presence of three
different Al2O3-based catalysts such as KF supported on basic
The reaction conversion and the selectivity (S% for products
4d-g) were decreased and enhanced, respectively, by the
increase of the CO2 pressure, in a fashion that closely resembled
the behavior reported in Figures 1-3.24 However, Figures 4
and 5 displayed that sharp variations of both reaction parameters
(conversion and selectivity) occurred in the pressure range of
80-100 bar (dCO ) 0.28-0.62 g/mL at 40 °C), which was, on
2
average, 10-20 bar below the corresponding interval observed
8524 J. Org. Chem. Vol. 73, No. 21, 2008

SelectiVe Nitroaldol Condensations
Although far from being exhaustive, this analysis suggested
that basic catalysts offered the best compromise between
conversion and selectivity; especially for reactions performed
in the supercritical medium, basic Al2O3 appeared more suitable
for aromatic aldehydes, while KF/Al2O3 for 3-phenylpropanal.
Reaction conditions were not optimized; other possible effects
such as the amount of catalysts, the reaction time, the temper-
ature, etc. were not investigated. However, to further substantiate
the results, some of the products reported on Tables 1 and 2
and Figure 6 were isolated and their yields (Y, %) were
calculated. In particular, nitroalkenes 4a,b and 4d were purified
by FCC (eluant: petroleum ether/ethyl acetate in 4:1 v/v) after
the reaction of the corresponding aldehydes over basic Al2O3
and KF/Al2O3.27
In addition, two more experiments were carried out with both
2-phenylpropionaldehyde and 3-phenylbutyraldehyde (2e and
2f, respectively), which were set to react under the conditions
of Figure 6C (40 °C, KF/Al2O3 catalyst). Products 4e and 4f
were isolated accordingly.
FIGURE 5. The effect of CO2 pressure on the conversion of n-
heptylaldehyde (2g) and on the selectivity (S% for compound 4g) in
the reaction with nitroethane at 40 °C.
Yields (Y, %) of compounds 4a,b and 4d-f are reported in
Table 3.
Al2O3, neutral Al2O3, and acid Al2O3 (details on these catalytic
systems are reported in the Experimental Section). Reaction
conditions were those of Table 1 for compounds 2a,b and of
Table 2 for compound 2d (2: 1.5 mmol; nitroethane: 1.8 mmol;
catalyst: 2 in a 2 weight ratio; Table 1: 60 °C; Table 2: 40 °C).
All aldehydes were set to react for 18 h, under solvent-free
conditions and in scCO2 at 100 bar (30 mL autoclave).
Results are reported in Figure 6A-C for the case of 2a, 2b,
and 2c, respectively. For each catalyst, the reaction conversions
observed under solFC and in supercritical CO2 are shown by
red and green bars, respectively. The corresponding selectivities
toward nitroalkenes 4a, 4b, and 4d are indicated with gray and
blue columns. Figure 6A-C also displays data of reactions
catalyzed by basic alumina (Tables 1 and 2).
Within limits of the experimental errors, the isolated yields
of products 4 were in good agreement with values of conversions
and selectivities determined by the 1H NMR analyses of
reactions of Table 3.
Discussion
Reactivity of Aromatic and Aliphatic Aldehydes. In both
scCO2 and under solvent-free conditions, the results of reactions
catalyzed by basic Al2O3 (Tables 1 and 2 and Figures 1-5)
indicate that aliphatic aldehydes have similar reactivity (Tables
2 and 3) and are, on average, more reactive than aromatic ones.
Moreover, the conversion of 4-nitrobenzaldehyde is higher than
that of benzaldehyde. This trend is substantially confirmed also
by Table 3 and Figure 6, which compare the reactivity of
compounds 2a-g over different Al2O3-based catalysts, and by
the results reported by Akutu et al.28 for the nitroaldol reaction
carried out over a variety of solid base catalysts (MgO, CaO,
KF/Al2O3, etc.).
This behavior is possibly explained by factors affecting the
reactivity of the carbonyl carbon, such as (i) the higher steric
hindrance expected for aromatic aldehydes with respect to
aliphatic ones and (ii) the general increase of the -CO
electrophilic character induced by electron-withdrawing sub-
stituents (i.e., the NO2 group). To our knowledge, these aspects
have not been further detailed in the literature, for both the
Henry reaction and the nitroaldol condensation.
The following aspects emerged from this investigation:
(i) Conversion. Regardless of conditions, different catalysts
did not considerably affect the conversions of aromatic alde-
hydes, which ranged from 30 to 50% and from 60 to 75% for
benzaldehyde and 4-nitrobenzaldehyde, respectively (Figure
6A,B, red and green bars). An exception was the reaction of 2a
over acid alumina that proceeded to a very low extent in scCO2
(∼10%, Figure 6A).
By contrast, the conversion of 3-phenylpropanal (on average,
higher than for aromatic substrates) showed modest variations
in the absence of solvent (80-95%; Figure 6C, red bars), while
it significantly changed in the supercritical phase, from a
maximum of 87% on KF/Al2O3 to only 22% on acidic Al2O3
(green bars).
Effects of the CO2 Pressure on the Reaction Outcome. In
the supercritical phase, most of the CO2 properties are affected
by its pressure. In particular, the density, which at a reduced
temperature (Tr) of 1.1, increases by around 80% when the
reduced pressure (Pr) is augmented from 1 to 3.29 Accordingly,
a sharp enhancement of the solubility of organic compounds in
CO2 is usually observed in the proximity and above the critical
point. This behavior may offer an explanation for the trend of
(ii) Selectivity. The selectivity (S%) for nitroalkenes 4 was
always better in scCO2 than under solvent-free conditions
(compare blue vs gray bars). Nonetheless, especially for
aromatic substrates, it decreased rather constantly from basic
to neutral and acid catalysts; in this sequence, S% dropped from
80 to ∼20%, and from 86 to 54% for benzaldehyde and
4-nitrobenzaldehyde, respectively (Figure 6A,B, blue bars). In
the case of 3-phenylpropanal, in dense CO2, a sharp decrease
of the selectivity was observed only over acid alumina (27%,
Figure 6C).
(27) Products 4a,b and 4d were purified by FCC (details are in the
Experimental Section).
(28) (a) Akutu, A.; Kabashima, H.; Seki, T.; Hattori, H. Appl. Catal. A 2003,
247, 65–74. (b) Hattori, H. J. Jpn. Petr. Inst. 2004, 47, 67–81.
(29) Lucien, F. P.; Foster, N. R. In Chemical Synthesis Using Supercritical
Fluids; Jessop, P. G., Leitner, W., Eds.; Wiley-VCH: New York, 1999; Chapter
1.2, pp 39-53 Tr ) T/Tc; Tc ) critical temperature ) 31.1 °C; Pr ) P/Pc; Pc )
critical pressure ) 73.8 bar.
(26) Figure 6A,B: Experiments were repeated twice for the four sets of
catalysts. Figure 6C: Experiments were repeated twice except for acid Al2O3
that was run once. For duplicate reactions, the values for conversions and
selectivity were the average of two runs that did not differ more than 5-7%.
J. Org. Chem. Vol. 73, No. 21, 2008 8525

Ballini et al.
FIGURE 6. The reaction of nitroethane with (A) benzaldehyde, (B) 4-nitrobenzaldehyde, (C) 3-phenylpropionaldehyde in the presence of four
different catalysts.26
conversions shown by Figures 1-5. Although in this investiga-
tion, the equipment used for the reactivity tests and the
concentration measures is not strictly comparable (see Experi-
mental Section), it is incontrovertible that, when the operating
pressure is increased (from 70 to 140 bar), the steep rise
observed in the solubility profiles of aldehydes in scCO2 (Figures
1 and 4, blue curves) closely mirrors the drop of the corre-
sponding conversions (black curves). In other words, if the
concentration of reagents in the supercritical phase increases,30
then they are plausibly desorbed/removed from the catalytic
surface, so that the reaction rate is reduced. A similar negative
effect of the pressure was observed by others and by us in the
investigation of Diels-Alder cycloadditions and of nucleophilic
displacements, carried out in dense CO2 in the presence of silica-
supported catalysts.17c,31 It should be noted, however, that also
for the homogeneously catalyzed reaction of 4-cyanobenzalde-
hyde with 1-nitropropane (Scheme 2) a decrease of the reaction
rate was induced by high pressures (g120 bar).8 Hence, under
these conditions, the liquid-like density of CO2 does not
completely rule out a “conventional” dilution effect.
4, plateau of the blue profile) appears shifted to lower pressures
(∼110 bar) with respect to the corresponding point for benzal-
dehyde (2a: ∼140 bar, Figure 1). The trend cannot be ascribed
to the different polarity of the substrates,32 but it is rather due
to the tunable density of CO2. In fact, concentration measures
for compounds 2a and 2d were performed at 60 and 40 °C,
where the supercritical phase reaches comparable densities (and
solvating power as well) in the range of 0.56-0.67 g/mL at
pressures of 140 and 110 bar, respectively.
Another situation occurs for 4-nitrobenzaldehyde (2b, Fig-
ure 2), whose low solubility in scCO2 plausibly comes from
the combined effects of two properties: the higher polarity of
2b with respect to 2a and 2d and, most of all, the low vapor
pressure of 2b (mp 103-106 °C).25 This latter property is only
partially balanced by the increase of the density of CO2 up to
150 bar (dCO ) 0.6 g/mL, at 60 °C). In fact, concentration
2
measures reported for 2-nitrobenzaldehyde in supercritical CO2
claim that a complete miscibility of this substrate is reached
only at dCO of 0.9 g/mL (25 °C, 175 bar).33
2
The tunable features of CO2 are likely to account for the
The above reasoning also clarifies why the maximum
solubility of 3-phenylpropionaldehyde (2d) in scCO2 (Figure
high selectivity observed in the formation of nitroalkenes 4.
(32) The dielectric constants (ε) of benzaldehyde and of cinnamaldehyde
are 17.85 and 17.72, respectively ( CRC Handbook of Chemistry and Physics,
75th ed.; Lide, D. P., Ed.; CRC Press: Boca Raton, FL, 1994);ε is not available
for 3-phenylpropionaldehyde (2d); however, the structure similarity between 2d
(hydroxycinnamaldehyde) and cinnamaldehyde strongly suggests that the two
compounds should have also a comparable permittivity.
(30) Nitroethane is assumed to be completely dissolved at P g 90 bar (see
Results).
(31) (a) Weinstein, R. D.; Renslo, A. R.; Danheiser, R. L.; Tester, J. W. J.
Phys. Chem. B 1999, 103, 2878–2887.
8526 J. Org. Chem. Vol. 73, No. 21, 2008

SelectiVe Nitroaldol Condensations
TABLE 3. Isolated Yields of Nitroalkenes 4a,b and 4d-fa
CO2 is certainly chemisorbed over the basic sites of solid
Al2O3. Although this acid-base interaction is expected to
modify or even to poison, to some extent, the catalytic
surface,24 it may also alter the reaction pathways in favor of
nitroalkene products.
Activity of Different Catalysts. In supercritical CO2, basic
Al2O3 and KF/Al2O3 show better catalytic performances than
neutral and, particularly, acid aluminas (Figure 6 and Table 3).
This is ascribed to the different modes of reactant activation:
in the first step of the reaction, basic catalysts efficiently generate
stable nitronate carbanions from nitroalkanes,1 while neutral/
acid systems allow only a modest enhancement of the electro-
philic character of aldehydic carbonyls.
Conclusions
In the current literature, the large number of publications
involving both the Henry reaction and the strictly related
nitroaldol condensation proves how these transformations
continue to attract interest in several fields of organic
synthesis. Although these reactions are often studied sepa-
rately, a new frontier of this chemistry lies, in our view, in
the study of innovative and possibly more eco-compatible
conditions, able to modulate the selectivity between the two
processes. After the pioneering paper reporting on the Henry
reaction performed in scCO2,8 this work represents the first
extensive investigation in this area. In particular, the emerging
original aspect is that, in the presence of heterogeneous
catalysts, supercritical CO2 can not only act as a green solvent
for the reaction of both aromatic and aliphatic aldehydes with
1-nitroalkanes but also shift the selectivity (up to 97%)
toward the formation of nitroalkenes (from the nitroaldol
condensation). On the other hand, under solvent-free condi-
tions, nitroaldols (from the Henry reaction) are the major
products. This result is achieved by the combination of the
tunable properties of CO2 and the nature of the catalysts used.
Among the tested aluminas, basic systems perform better than
neutral and acid ones; however, only if the pressure (and
the density) of the supercritical phase is increased from 80
to 140 bar, then the nitroalkene’s selectivity increases, on
average, from ∼60 to more than 90%. Several effects, such
as the local density enhancement, the high diffusivity, and
the weak acidic character of the supercritical phase, are likely
to account for this behavior.
a All reactions were carried out for 18 h at 60 and 40 °C in scCO2
(100 bar). Each experiment was repeated twice; the values for
conversions, selectivity, and isolated yields were the average of two runs
that did not differ more than 5-7%. b The weight ratio of
catalyst:aldehyde was 2. The molar ratio of nitroethane:2 was 1.2. c The
conversion referred to the aldehyde, and it was determined by 1H NMR.
d Selectivity (%) toward the formation of nitroalkenes 4a,b and 4d-f
was calculated as reported in Table 2. e Isolated yields of compounds 4
were obtained after their purification by FCC.
This result is not necessarily associated with the trend of
the concentration of reactants in the supercritical phase. In
fact, the reactions of 4-nitrobenzaldehyde are selective even
though the substrate possesses a limited solubility in scCO2
over the whole range of pressures explored here (Table 1
and Figure 2). A plausible hypothesis for the selectivity
control should consider other peculiar aspects of CO2: (i)
Above the critical point (Pc ) 74 bar, Tc ) 31 °C), the so-
called clustering or local density enhancement gives rise to
special solvation effects not ordinarily found in conventional
liquid mixtures.10b Accordingly, a modest variation of
pressure alters both the solutes/solvent and the transition
states/solvent interactions that may affect the product dis-
tribution. In this respect, a different partitioning of nitroaldol/
nitroalkene products between scCO2 and the catalyst is
expected; the more polar Henry product (nitroaldol) is likely
to reside over the Al2O3 surface (rather than in the super-
critical solvent), where the dehydration takes place. (ii) The
high diffusivity (D) of scCO2 not only favors the mass
transfer of reagents over solid catalysts10 but it also helps
the desorption of products and co-products such as water that,
in the present case, is generated by the nitroaldol condensa-
tion (Scheme 1).34 The CO2-mediated removal of water from
the catalyst pores has been reported as a crucial factor for
addressing the selectivity of oxidation reactions carried out
in scCO2 over Pd supported on Al2O3.35 (iii) Thanks to its
weak acidity, carbon dioxide is an excellent probe to
characterize heterogeneous base catalysts.36 In particular,
under the conditions investigated here for nitroaldol reactions,
Experimental Section
SAFETY WARNING: Operators of high-pressure equipment
should take proper precautions to minimize the risks of personal
injury.37
Reactions Carried out in an Autoclave. In a typical experiment,
a stainless-steel 30 mL reactor equipped with sapphire windows
on both top and bottom apertures, a pressure gauge, two inlet and
purge valves, and a thermocouple was charged with nitroethane
(1a, 130 µL, 1.8 mmol), an aldehyde (2a-h, 1.5 mmol), and the
catalyst (basic, neutral, or acid Al2O3 as well as KF on Al2O3: the
weight ratio of catalyst:aldehyde was 2). The autoclave was then
pressurized with CO2 (SFC/SFE grade) at approximately 60 bar
by using an automatic syringe pump (ISCO model 260 D), and it
was electrically heated at the desired temperature (40 or 60 °C).
The reaction mixture was magnetically stirred. The final pressure
(80-140 bar; see Tables 1-3 and Figures 1-6) was reached by
(33) Dandge, D. K.; Heller, J. P.; Wilson, K. V. Ind. Eng. Chem. Prod. Res.
DeV. 1985, 24, 162–166.
(34) It should be noted that water has a limited, but not negligible, solubility
in scCO2: Toews, K. L.; Shroll, R. M.; Wai, C. M. Anal. Chem. 1995, 67, 4040–
4043. Plausibly, the maximum amount of water produced by the nitroaldol
condensation is completely dissolved in CO2.
(35) (a) Caravati, M.; Grunwaldt, J.-D.; Baiker, A. Phys. Chem. Chem. Phys.
2005, 7, 278–285. (b) Caravati, M.; Grunwaldt, J.-D.; Baiker, A. Appl. Catal. A
2006, 298, 50–56.
(37) Jessop, P. G.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc. 1996, 118,
344–355.
(36) Hattori, H. Chem. ReV. 1995, 95, 537–550.
J. Org. Chem. Vol. 73, No. 21, 2008 8527

Ballini et al.
slowly adding the remaining CO2 to the reactor. Then, the reaction
was allowed to proceed for 18 h. After cooling to rt, CO2 was slowly
vented by bubbling it into diethyl ether (5 mL) cooled at 0 °C. The
content of the cell was washed with additional ether (5 mL), and
the catalyst was filtered. The solvent of the combined ethereal
solutions was removed in vacuo, and finally, the residues were
analyzed by H NMR.
The same procedure was also used by replacing 1-nitroethane
with 1-nitrohexane (1b, 260 µL, 1.8 mmol; see Table 1).
(M+, 5%), 161 (M+ - NO2 - H, 19), 132 (20), 115 (M+ - 2NO2,
42), 115 (M+ - H - 2NO2, 100), 63 (22).
(E)-2-Nitro-5-phenyl-2-pentene, 4d:41 1H NMR (CDCl3) δ
7.33-7.14 (m, 7H), 2.82 (t, 2H, JH-H ) 7.5 Hz), 2.55 (q, 2H, J )
7.7 Hz), 2.05 (s, 3H); MS (EI), m/z (relative int) 191 (M+, <1%),
144 (M+ - NO2 - H, 10), 91 [M+ - CH2CHC(CH3)NO2, 100],
65 (10).
1
(E)-2-Nitro-4-phenyl-2-pentene, 4e:42 1H NMR (CDCl3) δ
7.36-7.22 (m, 6H), 3.71 (dq, 1H, J ) 6.9 Hz, J′ ) 10.3 Hz), 2.23
(d, 3H, JH-H ) 1.0 Hz), 1.47 (d, 3H, JH-H ) 6.9 Hz); MS (EI),
Reactions Carried out under Solvent-Free Conditions. The
reactions of 1-nitroethane with aldehydes 2a-h were also carried
out under solvent-free conditions, by adjusting a procedure previ-
ously reported by us:2 a 5 mL test tube was charged with
1-nitroethane (1a, 130 µL, 1.8 mmol), an aldehyde (2a-h, 1.5
mmol), and the catalyst (basic, neutral, or acid Al2O3 as well as
KF on Al2O3: the weight ratio of catalyst:aldehyde was 2). The
reaction mixture was then magnetically stirred and heated at 40 or
60 °C in an oil bath for 18 h. After cooling to rt, the catalyst was
filtered and washed with diethyl ether (5 mL). The solvent of the
clear solution was removed in vacuo, and finally, the residue was
m/z (relative int) 191 (M+, 5%), 145 (M+ - NO2, 12), 144 (M+
NO2 - H, 13), 130 (M+ - NO2 - Me, 26), 129 (58), 105 [M+
CHC(CH3)NO2, 100], 91 (25), 77 (53), 51 (37).
-
-
(E)-2-Nitro-5-phenyl-2-hexene, 4f: 1H NMR (CDCl3) δ
7.33-7.17 (m, 5H), 7.08 (t, 1H, J ) 7.95 Hz), 2.95 (m, 1H, JH-H
) 7.0 Hz), 2.50 (m, 2H), 2.04 (d, 3H, JH-H ) 0.9 Hz), 1.34 (d,
3H, JH-H ) 7.0 Hz); 13C{1H} NMR (CDCl3) δ 148.3, 145.1, 134.2,
128.6, 126.7, 126.6, 39.3, 36.8, 21.5, 12.4; MS (EI), m/z (relative
int) 205 (M+, <1%), 105 [M+ - CH2CHC(CH3)NO2, 100], 91 (5),
77 (56), 51 (8). Anal. Calcd for C12H15NO2: C, 70.22; H, 7.37; N,
6.82. Found: C, 70.26; H, 7.33; N, 6.85.
1
analyzed by H NMR.
Concentration of Benzaldehyde (2a), 4-Nitrobenzaldehyde
(2b), and 3-Phenylpropionaldehyde (2d) in scCO2: Figures 1,
2, and 4. In a typical experiment, the aldehyde (2a: 2.5 g, 23.6
mmol; 2b: 1.0 g, 6.6 mmol; 2d: 2.5 g, 18.7 mmol) was charged in
a 90 mL stainless-steel autoclave, which was heated at 40 °C (for
2d) or at 60 °C (for 2a,b) and brought to the desired pressure of
CO2 in the range of 80-150 bar (see also ref 21). The mixture
was kept under stirring at 750 rpm. After 30 min, the agitation
was stopped, and an aliquot of the mixture was withdrawn through
a stainless-steel calibrated loop of 1.12 mL fitted to the autoclave
head. The loop was vented into diethyl ether (5 mL) and washed
thoroughly with additional ether (5 mL). The solvent of the
combined ethereal solutions was removed in vacuo, and the amount
of the liquid or solid residue of 2 was determined gravimetrically.
The measure was repeated twice for each of the chosen pressures.
Characterization and Isolation of Products. The formation of
nitroaldols 3a-h and nitroalkenes 4a-h was confirmed by the 1H
NMR spectra of reaction mixtures. The spectroscopic data were in
good agreement with those reported in the literature.38
In the case of reactions of nitroethane with aldehydes 1a,b
(entries 4 and 12, Table 1), 1d (entry 2, Table 2), and 1e,f (Figure
6c), carried out in scCO2, the corresponding nitroalkenes 4a,b and
4d-f were also isolated. At the end of each experiment, the
resulting reaction mixtures were purified by FCC on silica gel using
an eluant solution of petroleum ether/ethyl acetate (4:1 v/v). Isolated
yields of compounds 4a, 4b, 4d, 4e, and 4f were 51 (0.12 g), 66
(0.21 g), 72 (0.21 g), 67 (0.19 g), and 63% (0.19 g), respectively.
The 1H NMR and GC/MS spectra of pure nitroalkenes are reported
below. A 13C{1H} NMR spectrum was recorded for the new
compound 4f.
(E)-2-Nitro-1-phenylpropene, 4a:39 1H NMR (CDCl3) δ 8.11
(1H, s), 7.47 (5H, m), 2.48 (3H, s); MS (EI), m/z (relative int) 163
(M+, 8%), 146 (9), 117 (M+ - NO2, 42), 116 (30), 115 (M+ - 2H
- NO2, 100), 105 (32), 91 (42), 77 (14), 65 (17), 51 (24).
(E)-2-Nitro-1-(4-nitro)phenylpropene, 4b:40 1H NMR (CDCl3)
δ 8.31 (d, 2H, J ) 8.5 Hz), 7.59 (d, 2H, J ) 8.5 Hz), 8.09 (br s,
1H), 2.46 (d, 3H, JH-H ) 1.0 Hz); MS (EI), m/z (relative int) 208
Acknowledgment. MIUR (Italian Ministry of University and
Research) is gratefully acknowledged for financial support
(Grant No. 2006031724_002).
1
Supporting Information Available: H NMR for 4a, 4b,
4d, 4e, and 4f in CDCl3 and 13C NMR spectra of the new
compound 4f. This material is available free of charge via the
Internet at http://pubs.acs.org.
(38) (a) Bulbule, V. J.; Deshpande, V. H.; Velu, S.; Sudalai, A.; Sivasankar,
S.; Sathe, V. T. Tetrahedron 1999, 55, 9325–9332. (b) Dumez, E.; Faure, R.;
Dulcere, J.-P. Eur. J. Org. Chem. 2001, 13, 2577–2588. (c) Ballini, R.; Bosica,
G.; Marcantoni, E.; Vita, P.; Bartoli, G. J. Org. Chem. 2000, 65, 5854–5857.
(d) Mourad, M. S.; Varma, R. S.; Kabalka, G. W. J. Org. Chem. 1985, 50, 133–
135. (e) Tokunaga, Y.; Ihara, M.; Fukumoto, K. J. Chem. Soc., Perkin Trans. 1
1997, 207–210. (f) Ono, N.; Maruyama, K. Bull. Chem. Soc. Jpn. 1988, 61,
4470–4472.
JO801650P
(40) (a) Ballini, R.; Bigi, F.; Gogni, E.; Maggi, R.; Sartori, G. J. Catal. 2000,
191, 348–343. (b) Caldarelli, M.; Habermann, J.; Ley, S. V. J. Chem. Soc., Perkin
Trans. 1 1999, 107–110.
(41) Ballini, R.; Castagnani, R.; Petrini, M. J. Org. Chem. 1992, 57, 2160–
2162.
(42) Chounan, Y.; Ono, Y.; Nishii, S.; Kitahara, H.; Ito, S.; Yamamoto, Y.
Tetrahedron 2000, 56, 2821–2831.
(39) McAnda, A. F.; Roberts, K. D.; Smallridge, A. J.; Ten, A.; Trewhella,
M. A. J. Chem. Soc., Perkin Trans. 1 1998, 501–504.
8528 J. Org. Chem. Vol. 73, No. 21, 2008