CHEMISTRY & BIODIVERSITY ± Vol.

2 (2005) 1195

−Green× Synthesis of Nitro Alcohols Catalyzed by Solid-Supported

Tributylammonium Chloride in Aqueous Medium
by Zongting Wang* a ), Huiqing Xue b ), Shuchao Wang a ), and Cunguang Yuan a )
a
) School of Chemistry & Chemical Engineering, Chinese University of Petroleum, Dongying, 257062,
P. R. China (phone: ‡ 86-5 46-8 39 58 01; e-mail: [email protected])
b
) Laboratory of Chemistry, Shanxi College of Traditional Chinese Medicine, Taiyuan, 030024, P. R. China

A series of 1-aryl-2-nitroalkan-1-ols (5a ± m; Table 2) was prepared in yields of 64 ± 94% by Henry reaction
between aromatic aldehydes and nitroalkanes in the presence of a dual catalytic system of KOH and
polystyrene-supported tributylammonium chloride (PsTBAC; 2). The reactions were performed either in H2O
containing 10% of THF or, in the case of liquid reagents (aldehydes and nitroalkanes), in neat H2O at room
temperature. The immobilized phase-transfer catalyst, designed to be used for large-scale industrial processes,
could be easily separated from the products, and effectively re-used several times, basically without loss in
activity.

Introduction. ± The nitroaldol or Henry reaction, a classical method for the

construction of C C bonds, is a versatile method to prepare nitro alcohols [1] [2]. The
nitroaldol reaction can be catalyzed by alumina [3a], KF on alumina [3b], Amberlyst
A-21 [4a], Mg-Al hydrotalcite [4b], KG-60-NEt2 [4c], organic bases [5a,b], inorganic
bases, quaternary ammonium salts [5c,d], and other reagents. The reaction has been
performed both in protic and aprotic solvents, in aqueous media [6a], or without any
solvent (neat) [6b] [7]. However, there are also several drawbacks. The classical
nitroaldol reaction, when performed in organic solvents in the presence of base, has a
low selectivity toward different carbonyl compounds [8]. Also, basic reagents often
lead to side reactions, e.g., other aldol condensations or Cannizzaro reactions. And,
finally, volatile and toxic organic solvents have a negative impact on the environment.
In the context of rapidly developing Green Chemistry [9], one of the most-
important technical problems of industrial, homogenous phase-transfer catalysts
(PTCs) is the need to separate them from the reaction mixtures and to purify them for
re-use, or to dispose them properly. This problem not only increases cost, it may also
affect the purity of the products, and complicates scale up. Hence, the development of
less-expensive, more-efficient, recyclable PTCs is highly desirable. Immobilization of
the catalyst on a polymeric matrix, as in the case of the polystyrene (Ps)-supported
triethyl- or tributylammonium chlorides 1 (PsTEAC) and 2 (PsTBAC), respectively,
can provide a simple solution to these problems [10] [11].
In recent years, H2O has been recognized as an attractive medium for many organic
reactions [6a]. Aqueous media, relative to organic ones, are much less expensive, less
dangerous, and more environmentally friendly, and readily allow the control of pH and
the use of micro-aggregates (surfactants). Moreover, contrary to belief, the low

¹ 2005 Verlag Helvetica Chimica Acta AG, Z¸rich

1196 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)

solubility [12] of most organic reagents in H2O is not an obstacle to their reactivity,
which often is even higher in the absence of organic co-solvents.
Herein, we report an environmentally benign process for the simple, polymer-
supported phase-transfer catalysis of nitro alcohols in aqueous medium.

Results and Discussion. ± In a first experiment, different catalysts were compared in

the nitroaldol reaction of benzaldehyde (3a) with nitroalkanes 4 in H2O (Table 1). To a
solution of the aldehyde, 1.1 equiv. of the nitroalkane, one drop of 40% (w/w) KOH
soln., and the catalyst (1 mol-equiv. of Et3N, or 0.1 mol-equiv. of 1 or 2) were added in
H2O (5 ml) containing a small amount of THF (0.5 ml). The reactions were complete
within a few hours at room temperature. The polymer-supported catalysts, once de-
emulsified with NaCl, were filtered off, and the residue was washed with AcOEt. Then,
the filtrate was extracted with AcOEt (3  15 ml), the organic phase was dried
(Na2SO4 ) and concentrated, and the crude nitro alcohols 5, if necessary, were purified
by flash chromatography (SiO2 ; AcOEt/petroleum ether 1 : 10).

Table 1. Model Nitroaldol Reaction with Different Catalysts. Conditions: H2O/THF 100 : 10, r.t. The reactions
work similarly well in neat H2O for liquid reagents.

Entry Catalyst a ) Base Time [h] Product Isolated yield [%]

1 Et3N ± 6 5a 81
2 ± KOH 18 56
3 PsTEAC KOH 8 83
4 PsTBAC KOH 8 94
5 Et3N ± 6 5b 85
6 ± KOH 17 67
7 PsTEAC KOH 8 86
8 PsTBAC KOH 8 89
a
) PsTEAC (1) and PsTBAC (2), 0.1 mol-equiv.; Et3N, 1.0 mol-equiv.

We found that THF as a co-solvent is not really necessary for the reactions to
proceed. The results hardly changed in neat H2O, when the reactants were liquids at
ambient temperature. When we used an excess of nitroalkane (1.5 equiv.), the reaction
 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1197

was faster, but the yield only slightly increased. Since benzaldehyde is prone to air
oxidation, traces of benzoic acid should be neutralized in situ with base by adjusting a
pH of 8 ± 9, before the Henry reaction is started. Also, the reaction time should be kept
minimal to avoid side reactions.
As shown in Table 1, Et3N (Entries 1 and 5) is an efficient catalyst for nitroaldol
reactions. However, stoichiometric amounts and recycling problems limit its use in
industrial processes. When KOH, the base in the classical Henry reaction, was solely
used (Entry 2), the yield decreased due to side reactions. When PsTEAC or PsTBAC
(0.1 equiv.) were used in combination with a small amount of KOH, good results were
obtained. Thereby, PsTBAC (2) gave slightly higher yields of 5 compared to PsTEAC
(1).
Encouraged by these results, other aldehydes were examined as substrates under
similar conditions of the dual catalytic system PsTBAC/KOH. Thereby, 10% of THF
(0.5 ml) was added as a co-solvent, since the solid aromatic aldehydes had to be brought
into the aqueous solution. The results of these experiments are collected in Table 2.

Table 2. Synthesis of Nitro Alcohols 5 by Henry Reaction with PsTBAC (2). Condition, PsTBAC (2; 0.1 mol-
equiv.), KOH, H2O/THF 100 : 10, r.t.

Entry R1 R2 Time [h] Product Isolated yield [%] syn/anti a )

1 H H 8 5a 94 ±
2 4-Br H 7 5c 74 ±
3 4-Cl H 7 5d 79 ±
4 2-Cl H 6 5e 91 ±
5 4-NO2 H 7 5f 88 ±
6 2-MeO H 8 5g 78 ±
7 4-Me H 9 5h 64 ±
8 H Me 8 5b 86 1.96 : 1
9 4-Cl Me 7 5i 83 1.65 : 1
10 2-Cl Me 7 5j 91 1.04 : 1
11 4-NO2 Me 7 5k 90 1.28 : 1
12 2-MeO Me 7 5l 81 1.08 : 1
13 4-Me Me 9 5m 71 1.89 : 1
a
) Determined by 1H-NMR analysis.

All aromatic aldehydes tested reacted efficiently with nitromethane (4a) or

nitroethane (4b) in the presence of PsTBAC and KOH in H2O. Electron-withdrawing
atoms/groups, such as Cl (Table 2, Entries 3, 4, 9, and 10) and NO2 (Entries 5 and 11)
generally favored the reaction. Thereby, a Cl-atoms in ortho-position (Entries 4 and 10)
gave rise to higher yields than one in para-position (Entries 3 and 9). Electron-donating
atoms/groups such as Me (Entries 7 and 13) and Br (Entry 2) were less favorable in
1198 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005)

terms of yield. We also observed some stereochemical selectivities (Entries 8 ± 13),

which were similar to those obtained in related experiments reported previously [5a].
At the end of the model reaction between benzaldehyde and nitromethane, the
PsTBAC catalyst was filtered off (B¸chner funnel), washed with CH2Cl2 , dried under
vacuum, and re-used several times. We found that the catalyst produced similarly good
results in at least four further cycles (total: five runs), the product yields almost
remaining constant: 94, 90, 93, 92, and 88% yield, respectively.
In summary, we have shown that polystyrene-supported tributylammonium
chloride (PsTBAC; 2) efficiently catalyzes the nitroaldol reaction in aqueous medium,
and can be used for at least five such repetitive transformations. Thus, PsTBAC can be
considered a powerful −green× agent suited for the large-scale preparation of simple
nitro alcohols. Work is now in progress to design new inorganic- and organic-carrier-
supported catalysts, such as dendrimers, which could further improve the stereo-
chemical outcome of this reaction.

Experimental Part
General. All commercially available solvents and regents were used without further purification. Thin-layer
chromatography (TLC) was performed on silica-gel F 254 plates; visualization under UV light. Flash-column
chromatography (FC) was performed on silica gel (100 ± 200 mesh). 1H- and 13C-NMR Spectra were recorded at
400 and 100 MHz, resp., and were referenced to internal solvent signals; d in ppm, J in Hz. All compounds were
fully characterized, including GC/MS (Agilent 6890N, 5793N, HP-5).
Syntheses of the Catalysts PsTEAC (1) and PsTBAC (2). To a flask, Chloromethyl resin (5.0 g, 200 ±
400 mesh, 3.5 mmol Cl/g) was added in DMF (15 ml), and the resin was stirred for 2 h. Then, Et3N or Bu3N
(20 ml) was added, and the mixture was heated at 808 for 24 h, filtered, and dried in vacuum. The residue was
marinated in anh. EtOH, and stirred for 12 h at 808. Then, the solid-supported catalyst was dried in vacuum. The
concentration of onium cations, which is the same as that of the Cl anions, was determined indirectly by
titration of Cl with AgNO3 soln.
Typical Procedure for Nitroaldol Reactions Catalyzed by PsTBAC. To a soln. of, e.g., benzaldehyde (3a;
0.351 g, 3.31 mmol) in H2O (5 ml) and THF (0.5 ml) 1), nitroethane (4b; 0.272 g, 3.63 mmol), one drop of aq.
40% KOH soln., and PsTBAC (2; 0.20 g, ca. 10 mol-% of onium ions) were added. The mixture was stirred at r.t.
for 8 h. Then, excess H2O was added, the suspension was filtered (B¸chner funnel), and washed with AcOEt.
The aq. layer was extracted with AcOEt (3  15 ml), the org. phase was dried (Na2SO4 ) and concentrated, and
the crude product was purified by FC (SiO2 ; AcOEt/petroleum ether 1 : 10) to afford 89% of 2-nitro-1-
phenylpropan-1-ol (5b).
Data of 2-Nitro-1-phenylethanol (5a). 1H-NMR (CDCl3 ): 3.30 ± 3.70 (1 H); 4.36 ± 4.51 (2 H); 5.31 (1 H),
7.22 ± 7.80 (5 H). 13C-NMR (CDCl3 ): 70.74 (C(1)); 80.87 (C(2)); 125.46 ± 138.09 (Ph).
Data of 2-Nitro-1-phenylpropan-1-ol (5b). 1H-NMR (CDCl3 ; anti-isomer): 1.45 (3 H); 2.90 ± 3.30 (1 H);
4.72 (1 H); 5.32 (1 H); 7.31 ± 7.85 (5 H). 1H-NMR (CDCl3 ; syn-isomer): 1.26 (3 H); 2.90 ± 3.30 (1 H); 4.74 (1 H);
4.97 (1 H); 7.31 ± 7.85 (5 H).
Data of 1-(4-Bromophenyl)-2-nitroethan-1-ol (5c). 1H-NMR (CDCl3 ): 3.60 ± 4.00 (1 H); 4.47 ± 4.59 (2 H);
5.41 (1 H); 7.25 ± 7.72 (4 H).
Data of 1-(4-Chlorophenyl)-2-nitroethan-1-ol (5d). 1H-NMR (CDCl3 ): 3.30 ± 3.70 (1 H); 4.42 ± 4.55 (2 H);
5.38 (1 H); 7.27 ± 7.76 (4 H).
Data of 1-(2-Chlorophenyl)-2-nitroethan-1-ol (5e). 1H-NMR (CDCl3 ): 3.40 ± 3.80 (1 H); 4.40 (1 H); 4.60
(1 H); 5.78 (1 H); 7.22 ± 7.33 (3 H); 7.41 ± 7.61 (1 H).
Data of 2-Nitro-1-(4-nitrophenyl)ethan-1-ol (5f). 1H-NMR (CDCl3 ): 3.80 ± 4.20 (1 H); 4.64 (2 H); 5.64
(1 H); 7.62 ± 8.37 (4 H).

1) THF is required only in the case of solid aromatic aldehydes.

 CHEMISTRY & BIODIVERSITY ± Vol. 2 (2005) 1199

Data of 1-(2-Methoxyphenyl)-2-nitroethan-1-ol (5g). 1H-NMR (CDCl3 ): 3.30 ± 3.70 (1 H); 4.59 (2 H); 5.60
(1 H); 6.87 ± 7.79 (4 H).
Data of 1-(4-Methylphenyl)-2-nitroethan-1-ol (5h). 1H-NMR (CDCl3 ): 2.37 (3 H); 3.60 ± 4.00 (1 H); 4.42 ±
4.59 (2 H); 5.38 (1 H); 7.15 ± 7.74 (4 H).
Data of 1-(4-Chlorophenyl)-2-nitropropan-1-ol (5i). 1H-NMR (CDCl3 ; anti-isomer): 1.46 (3 H); 3.30 ± 3.70
(1 H); 4.63 (1 H); 5.35 (J ˆ 4.0, 1 H); 7.28 ± 7.81 (4 H). 1H-NMR (CDCl3 ; syn-isomer): 1.28 (3 H); 3.30 ± 3.70
(1 H); 4.75 (1 H); 5.00 (J ˆ 8.0, 1 H); 7.28 ± 7.81 (4 H).
Data of 1-(2-Chlorophenyl)-2-nitropropan-1-ol (5j). 1H-NMR (CDCl3 ; anti-isomer): 1.41 (3 H); 3.40 ± 3.80
(1 H); 4.08 (1 H); 5.84 (J ˆ 2.8, 1 H); 7.27 ± 7.64 (4 H). 1H-NMR (CDCl3 ; syn-isomer): 1.41 (3 H); 3.40 ± 3.80
(1 H); 4.87 (1 H); 5.60 (J ˆ 4.8, 1 H), 7.27 ± 7.64 (4 H).
Data of 2-Nitro-1-(4-nitrophenyl)propan-1-ol (5k). 1H-NMR (CDCl3 ; anti-isomer): 1.48 (3 H); 3.80 ± 4.20
(1 H); 4.10 (1 H); 5.59 (J ˆ 3.6, 1 H), 7.62 ± 8.40 (4 H). 1H-NMR (CDCl3 ; syn-isomer): 1.37 (3 H); 3.80 ± 4.20
(1 H); 4.80 (1 H); 5.22 (J ˆ 4.4, 1 H), 7.62 ± 8.40 (4 H).
Data of 1-(2-Methoxyphenyl)-2-nitropropan-1-ol (5l). 1H-NMR (CDCl3 ; anti-isomer): 1.42 (3 H); 3.30 ±
3.70 (1 H); 3.85 (3 H); 4.88 (1 H); 5.55 (J ˆ 3.6, 1 H); 6.86 ± 7.80 (4 H). 1H-NMR (CDCl3 ; syn-isomer): 1.30
(3 H); 3.30 ± 3.70 (1 H); 3.85 (3 H); 4.98 (1 H); 5.16 (J ˆ 8.8, 1 H); 6.86 ± 7.80 (4 H).
Data of 1-(4-Methylphenyl)-2-nitropropan-1-ol (5m). 1H-NMR (CDCl3 ; anti-isomer): 1.45 (3 H); 2.32
(3 H); 4.64 (1 H); 5.26 (J ˆ 4.4, 1 H); 7.13 ± 7.73 (4 H). 1H-NMR (CDCl3 ; syn-isomer): 1.24 (3 H); 2.32 (3 H);
4.73 (1 H); 4.93 (J ˆ 9.2, 1 H); 7.13 ± 7.73 (4 H).

This study was financially supported in part by the Natural Science Foundation of China (No. 29933050 and
20373082).

REFERENCES
[1] L. C. Henry, Acad. Sci. Ser. 1895, 120, 1265; L. Henry, Bull. Soc. Chim. Fr. 1895, 13, 99.
[2] G. Rosini, in −Comprehensive Organic Synthesis×, Ed. B. M. Trost, Pergamon Press, New York, 1991,
p. 321 ± 340.
[3] a) G. Rosini, R. Ballini, P. Sorrenti, Synthesis 1983, 1014; b) J. M. Melot, F. Texier-Boullet, A. Foucraud,
Tetrahedron Lett. 1986, 7, 493.
[4] a) R. Ballini, M. Patrini, G. Rosini, Synthesis 1987, 711; b) V. J. Bulbule, V. H. Deshpande, S. Velu, V. T.
Sathe, Tetrahedron 1999, 55, 9325; c) R. Ballini, G. Bosica, D. Livi, A. Palmieri, R. Maggi, G. Sartori,
Tetrahedron Lett. 2003, 44, 2271.
[5] a) I. Marao, F. P. Cossio, Tetrahedron Lett. 1997, 38, 6461; b) D. Simoni, R. Rondanin, M. Morini, R.
Baruchello, F. P. Invidiata, Tetrahedron Lett. 2000, 41, 1607; c) R. Ballini, G. Bosica, J. Org. Chem. 1997, 62,
425; d) V. J. Bulbule, G. K. Jnaneshwara, R. R. Deshmukh, H. B. Borate, V. H. Deshpande, Synth.
Commun. 2001, 31, 3623.
[6] a) R. Ballini, G. Sosica, J. Org. Chem. 1997, 62, 418; b) A. Bhattacharya, V. C. Purohit, Org. Process Res.
Dev. 2003, 7, 254.
[7] Environmentally Friendly Solvent-Free Processes: −Preparation of Nitro Alcohols. A Class of Valuable
Drug Intermediates by Henry Reaction×, American Chemical Society, 57th South-west Regional Meeting,
October 2001, pp. 17 ± 20.
[8] B. M. Trost, Science 1985, 227, 908.
[9] T. Cablewski, F. Faux, C. R. Straus, J. Org. Chem. 1994, 59, 3408; K. D. Raner, C. R. Strauss, R. W. Trainor,
J. S. Thorn, J. Org. Chem. 1995, 60, 2456.
[10] J. S. Fruchtel, G. Jung, Angew. Chem., Int. Ed. 1996, 35, 17; L. A. Thomson, J. A. Ellmann, Chem. Rev. 1996,
96, 555; P. H. H. Hermkens, H. C. J. Ottenhejm, D. Rees, Tetrahedron 1996, 52, 4527; R. Annunziata, M.
Benaglia, M. Cinquini, F. Cozzi, F. Tocco, Org. Lett. 2000, 2, 1737.
[11] M. Benaglia, A. Puglisi, F. Cozzi, Chem. Rev. 2003, 103, 3401; B. Thierry, J. C. Plaquevent, D. Cahard,
Tetrahedron: Asymmetry 2001, 12, 983; Q. L. Zeng, W. Weng, Y. Z. Jiang, Chemistry 2001, 69, 547; R.
Chinchilla, P. Mazon, C. Najera, Tetrahedron: Asymmetry 2000, 11, 3277.
[12] F. Fringuelli, G. Pani, O. Piermatti, F. Pizzo, Tetrahedron 1994, 50, 11 499.

Received April 15, 2005