Trimerization of aldehydes with one α-hydrogen catalyzed by sodium hydroxide

Article abstract of DOI:10.2478/s11696-013-0450-5

Trimerization of 2-methylpropanal (isobutyraldehyde) is a simple and effective method to synthesize 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate and 2,2,4-trimethyl-1,3-pentanediol-3-monoisobutyrate which are often used as film forming auxiliaries in paints. The use of solid sodium hydroxide as a catalyst provides an excellent yield of above 85 % after the optimization of the reaction time and the catalyst dosage. Furthermore, trimerization of four other aldehydes with one α-hydrogen catalyzed by solid sodium hydroxide can also take place and the yield of 1,3-diol monoesters reaches 50-70 %. Trimerization of aldehydes with one α-hydrogen can be explained by a three-step reaction mechanism: (i) aldol condensation of aldehyde; (ii) crossed Cannizzaro reaction; and (iii) esterification of carboxylic acid and alcohol.

Full text of DOI:10.2478/s11696-013-0450-5

Chemical Papers
DOI: 10.2478/s11696-013-0450-5
SHORT COMMUNICATION
Trimerization of aldehydes with one α-hydrogen catalyzed
by sodium hydroxide
Yu-Gang Li, Chen-Xi Luo, Chao Qian, Xin-Zhi Chen*
Key Laboratory of Biomass Chemical Engineering of Ministry of Education, Zhejiang University, 310027 Hangzhou, China
Received 28 March 2013; Revised 23 May 2013; Accepted 5 June 2013
Trimerization of 2-methylpropanal (isobutyraldehyde) is a simple and effective method to
synthesize 2,2,4-trimethyl-1,3-pentanediol monoisobutyrate and 2,2,4-trimethyl-1,3-pentanediol-3-
monoisobutyrate which are often used as film forming auxiliaries in paints. The use of solid sodium
hydroxide as a catalyst provides an excellent yield of above 85 % after the optimization of the
reaction time and the catalyst dosage. Furthermore, trimerization of four other aldehydes with one
α-hydrogen catalyzed by solid sodium hydroxide can also take place and the yield of 1,3-diol mo-
noesters reaches 50–70 %. Trimerization of aldehydes with one α-hydrogen can be explained by a
three-step reaction mechanism: (i) aldol condensation of aldehyde; (ii) crossed Cannizzaro reaction;
and (iii) esterification of carboxylic acid and alcohol.
c
ꢀ 2013 Institute of Chemistry, Slovak Academy of Sciences
Keywords: trimerization, isobutyraldehyde, aldehydes, sodium hydroxide
Trimerization of aliphatic aldehydes with one α-
hydrogen is of fundamental importance in synthetic
organic chemistry granting access to 1,3-diol mono-
esters widely used as coalescing agents in the paint
industry (T¨ormäkangas & Koskinen, 2001; Gallagher
et al., 2008). Our work was focused on 2,2,4-trimethyl-
1,3-pentanediol monoisobutyrate (Va) and its isomers
(2,2,4-trimethyl-1,3-pentanediol-3-monoisobutyrate;
VIa) synthesized by the trimerization of 2-methylpro-
panal (isobutyraldehyde). Since Va is a glycol ester
with a high boiling point and it is hardly soluble in
water, it has drawn attention as a solvent and plasti-
cizer having remarkable properties such as very good
miscibility with resins, low solidifying point, excellent
viscosity stability, and low soiling property in latex
emulsions and paints (Takasu et al., 1972; Duccini et
al., 2007). In recent years, it has been used as a film
forming auxiliary in paints (Swan, 2005). VIa has the
same industrial applications as Va and it is not neces-
sary to separate them in industrial scale.
condensation of isobutyraldehyde catalyzed by vari-
ous catalysts is a convenient method and the reac-
tion with high atom economy is friendly to the envi-
ronment. However, the catalysts used to promote the
self-condensation of isobutyraldehyde are usually ex-
pensive and not easily accessible, [Al(OC2H5)4]2Mg,
[Al(OC3H7)4]2Mg, [Al(OC4H9)4]2Mg (Kulpinski &
Nord, 1943; Villani & Nord, 1947), LiWO2 (Vil-
lacorta & Filippo, 1983), 1,3-diol monoalcoholates
(T¨ormäkangas & Koskinen, 2001), Fe3(CO)12 (Ito et
al., 1983), Cp*2Sm(thf)2 (Miyano et al., 1998). Prepa-
ration of the above-mentioned catalysts is usually
complicated and costly which limits their industrial
application. Jian et al. (2003a) disclosed a method to
obtain Va. The reaction was carried out in two steps
in the presence of a 45 mass % sodium hydroxide so-
lution. The condensation reaction was conducted at
temperatures in the range of 20–30◦C for 3 h, and at
60◦C for 2 h in the second step. After the removal of
the aqueous phase by delamination, the organic phase
was purified by fractional distillation under reduced
pressure. Tic et al. (2012) introduced a method of con-
tinuous production of isobutyraldehyde. The reaction
With the field of applications widening, the de-
velopment of general and efficient methods for the
synthesis of Va is an active area of research. Self-
*Corresponding author, e-mail: chemtec@163.com

ii
Y. G. Li et al./Chemical Papers
Table 1. Effect of the catalyst amount and the reaction time on the yield of the isobutyraldehydea trimerization at 60◦C
n(NaOH)
Time
h
x(Ia)
x(IIa)
x(IIIa)
x(IVa)
x(Va), x(VIa)
x(VIIa)
Yieldb
%
Entry
mol
mole %
1
2
3
4
5
6
7
8
9
0.025
0.038
0.050
0.063
0.073
0.063c
0.063
0.063
0.063
4
4
4
4
4
4
2
3
5
84.4
55.9
40.4
17.0
16.1
41.1
61.4
41.0
16.8
1.9
2.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
2.1
3.4
4.8
5.9
9.1
8.8
2.9
5.0
6.3
2.0
3.3
4.1
5.7
8.8
7.7
2.6
4.1
5.7
9.5
34.6
50.0
71.1
65.7
41.2
32.9
49.0
70.5
0.0
0.1
0.7
0.3
0.3
1.1
0.2
0.9
0.6
23.2
59.3
72.8
86.2
82.1
63.6
58.8
71.9
85.4
a) Amount of isobutyraldehyde was 120 mL (1.3 mol); b) yield of Va and VIa determined by GC; c) 6.25 g of 40 mass % sodium
hydroxide solution was used.
was carried out in the presence of a mixture of sodium
hydroxide and sodium isobutyrate in the mass ratio of
3 : 1 as the catalyst, first at temperatures in the range
of 58–62◦C for 8–12 min and then at temperatures in
the range of 88–92◦C for 3–7 min. After the removal
of the catalyst, the organic phase was purified by frac-
tional distillation under reduced pressure. However, in
either case, the yield obtained was unsatisfactory.
As it is well known, the sodium hydroxide solu-
tion destroys esters via hydrolysis, thus, water is sup-
posed to have a negative effect on the reaction yield.
To achieve higher yields, the use of water should be
avoided and solid sodium hydroxide was used as the
catalyst instead of the sodium hydroxide solution. Al-
though the sodium hydroxide solution is broadly used
as a catalyst for the self-condensation of Ia, this is the
first time sodium hydroxide has been used directly as
the catalyst.
This paper provides a detailed study of the trimer-
ization of isobutyraldehyde, Va, and VIa catalyzed by
sodium hydroxide with excellent yield and proposes
a three-step reaction mechanism for the first time.
Trimerization of other aldehydes with one α-hydrogen
was also studied.
Starting materials and reagents were of reagent
grade (> 98 %) and used as received. GC (Ag-
ilent 1705A; Agilent, USA) and GC-MS (TRACE
G2000/TRACE MS, ThermoQuest, USA) were used
to analyze the reaction mixtures. 1H NMR (400 MHz)
and 13C NMR (100 MHz) spectra were recorded on
a Bruker AC-P400 spectrometer (USA). The spectra
were measured in CDCl3, relative to TMS as an in-
ternal standard. All reactions were accomplished in
pre-dried glassware (150◦C, 4 h).
stopped after 4 h by removing the heat source and
the reaction mixture was cooled to room temperature.
Gaseous HCl (0.074 mol) was introduced into the re-
action vessel at the rate of 60 mL min−1 for 30 min to
transform sodium hydroxide and sodium isobutyrate
into sodium chloride. After sodium chloride was re-
moved by filtration, the reaction mixtures were ana-
lyzed using gas chromatopraphy and GC-MS. Relative
contents of the components were determined by the
area normalization method. The yield of Va and VIa
was determined by an external standard method as
described below. First, standard samples of nine dif-
ferent concentrations around the probable concentra-
tion of the analyte were analyzed by GC at the same
injection volume to obtain a scatter diagram with a
regression line, with the abscissa representing the con-
centration of the standard sample and its ordinate the
peak area. Then, the reaction mixture was analyzed
by GC at the same injection volume as the standard
samples and the concentration was calculated consid-
ering the peak area.
The filtrate was purified using a fractional col-
umn of 10 cm with corrugated wire gauze packing in
four steps. Unreacted Ia was collected in the range
of 63–65◦C at atmospheric pressure. Isobutyric acid
(IVa) was collected in the range of 146–149◦C at
atmospheric pressure. 2,2,4-Trimethyl-1,3-pentanediol
(IIIa) was collected in the range of 120–122◦C at
1.33 kPa (absolute pressure). Va and VIa were col-
lected in the range of 127–129◦C at 1.33 kPa (abso-
lute pressure). The remainder was 2,2,4-trimethyl-1,3-
pentanediol diisobutyrate (VIIa). Va and VIIa were
separated by column chromatography (EA/Hex; ϕr =
1 : 4).
To a 250 mL three-necked flask with a reflux
condenser, thermometer, and a mechanic stirrer Ia
(120 mL, 1.3 mol) was added. Then, flake sodium hy-
droxide (2.5 g, 0.063 mol) was added in batches over
2 h because of the exotermicity of the reaction. Dry
and pure sodium hydroxide was protected from air in
a desiccator before being added to the reaction. The
reaction was carried out at 60◦C. The reaction was
In the experiments, the amount of sodium hydrox-
ide was varied from 1.0 g (0.025 mol) to 3.0 g (0.073
mol) to study its effect on the reaction (Table 1).
Formation of 3-hydroxy-2,2,4-trimethylpentanal (IIa)
and IIIa was observed and the conversion of Ia was
poor when the amount of sodium hydroxide was below
1.5 g (0.038 mol). When the amount of sodium hydrox-
ide was increased, IIa disappeared and the amount of

Y. G. Li et al./Chemical Papers
iii
Fig. 1. Trimerization of Ia and aldehydes with α-hydrogen catalyzed by sodium hydroxide.
IIIa increased considerably (Table 1). In addition, for-
mation of IVa was observed after a sufficient amount
of gaseous HCl, originally intended to be used to
achieve complete transformation of sodium hydroxide
into sodium chloride, was added. Furthermore, when
the reaction mixtures were analyzed with GC at 15
min intervals during the reaction, it was observed that
the amount of IIIa showed an increasing and then a
decreasing trend, indicating that the formation of the
desired product, Va and VIa, can be explained by a
three-step reaction mechanism (Fig. 1): (i) aldol con-
densation (Han et al., 2000; Mahrwald et al., 1998) of
Ia providing IIa; (ii) IIa not containing α-hydrogen re-
action with the original aldehyde (Ia) to give IIIa and
IVa via a crossed Cannizzaro reaction (Swain et al.,
1979; Bikbaeva et al., 1975). It is necessary to state
that in some cases, the Cannizzaro and Tishchenko
reactions (Fouquet et al., 1979) are competitive with
each other. The balance between them can be con-
trolled by the proper choice of catalyst, reaction tem-
perature and solvents. Experimental results showed
that IIa reacted with the original aldehyde to give
IIIa and IVa via a Cannizzaro reaction when catalyzed
by flake sodium hydroxide at 60◦C. The crossed Can-
nizzaro reaction did not proceed to completion unless
enough sodium hydroxide was added into the reaction,
thus, IIa could not be transformed into IIIa completely
and the conversion of Ia was poor when the amount
of sodium hydroxide was below 1.5 g (0.038 mol); (iii)
esterification of IIIa and IVa producing Va or VIa. At
the same time, VIIa was prepared as a side product
by esterification at the 1 : 2 mole ratio of IIIa to IVa.
Because of the reversibility of the esterification reac-
tion, the conversion of IIIa and IVa was incomplete.
Unreacted IVa was turned into sodium isobutyrate by
sodium hydroxide, so it could not be observed unless a
sufficient amount of HCl was added into the reaction.
More IVa was turned into sodium isobutyrate when
the amount of sodium hydroxide increased, which led
to the shortage of IVa when the esterification of IIIa
and IVa took place. The amount of IIIa which was
not transformed to Va or VIa increased, as shown
above. Spectral data of the corresponding compounds
are summarized in Table 2.
Experimental data are shown in Table 1. The yield
of Va and VIa increased with the increasing amount
of sodium hydroxide (up to 0.063 mol). When the
amount of sodium hydroxide was above 0.063 mol, the
yield did not increase obviously. Therefore, the value
of 0.063 mol was chosen for further experiments. In
case of a very short reaction time, the yield was not
high, the period of 4 h was found to be optimal to
achieve a good yield. As predicted, water had a nega-
tive effect on the yield. When sodium hydroxide solu-
tion was used as the catalyst instead of solid sodium
hydroxide, the yield was lower and the content of IIIa
was high.
The effect of IVa present in the starting materials
on the reaction was also studied. IVa consumes sodium
hydroxide and the reaction needs enough sodium hy-
droxide, therefore, more sodium hydroxide had to be
added when the amount of IVa in Ia increased.
On basis of the above results, many aldehydes
with one α-hydrogen such as 2-methyl-butyraldehyde
(Jian et al., 2003b) (Ib), 2-methyl-pentanal (Ic), D,L-
2-phenylpropionaldehyde (Id), 2-ethyl-butyraldehyde
(Ie) can to react in the presence of sodium hydroxide
at 60◦C (Table 3).

iv
Y. G. Li et al./Chemical Papers
Table 2. Spectral data of the corresponding compounds
Compound
Spectral data
IIIa
1H NMR (CDCl3), δ: 3.47 (m, 2H, J = 10.6 Hz), 3.39 (d, 1H), 2.81 (s, 2H), 1.94 (m, 1H), 1.04 (d, 3H), 0.98 (d,
3H), 0.93 (s, 6H)
13C NMR (CDCl3), δ: 82.7, 73.2, 39.2, 29.1, 23.5, 20.0, 17.0
IVa
Va
1H NMR(CDCl3), δ: 2.59 (m, 1H), 1.17 (d, 6H)
13C NMR (CDCl3), δ: 183.2, 33.6, 18.4
1H NMR (CDCl3), δ: 3.94 (m, 2H, J = 10.9 Hz), 3.22 (d, 1H), 2.54 (m, 1H), 1.85 (m, 1H), 1.15 (d, 6H), 0.97 (d,
3H), 0.93 (s, 3H), 0.91 (s, 3H), 0.87 (d, 3H)
13C NMR (CDCl3), δ: 177.4, 79.3, 71.5, 39.3, 34.0, 28.9, 23.4, 22.0, 20.5, 19.2, 19.2, 16.6
VIa
1H NMR (CDCl3), δ: 4.79 (d, 1H), 3.11 (m, 2H, J = 11.6 Hz), 2.64 (m, 1H), 2.09 (m, 1H), 1.21 (d, 6H), 1.01 (s,
3H), 0.97 (d, 3H), 0.94 (d, 3H), 0.88 (s, 3H)
13C NMR (CDCl3), δ: 178.2, 79.4, 69.8, 40.0, 34.8, 28.2, 22.7, 22.3, 19.6, 19.5, 17.6
VIIa
1H NMR (CDCl3), δ: 4.74 (d, 1H), 3.86 (m, 2H, J = 11.0 Hz), 2.61 (m, 1H), 2.57 (m, 1H), 2.01 (m, 1H), 1.21 (d,
3H), 1.19 (d, 3H), 1.16 (d, 6H), 0.98 (s, 3H), 0.96 (d, 3H), 0.93 (s, 3H), 0.91 (d, 3H)
13C NMR (CDCl3), δ: 176.8, 176.7, 79.3, 69.8, 38.7, 34.6, 34.0, 28.5, 23.3, 21.9, 21.4, 19.3, 19.3, 19.2, 18.8, 17.9
Table 3. Trimerization of aldehydes with one α-hydrogen cat-
sory and analytical evaluations of paints with and without
texanol. Environment Science & Technology, 42, 243–248.
DOI: 10.1021/es071555y.
Han, Z. F., Yorimitsu, H., Shinokubo, H., & Oshima, K. (2000).
A highly effective aldol reaction mediated by Ti(O-n-Bu)4/t-
BuOK combined reagent. Tetrahedron Letters, 41, 4415–
4418. DOI: 10.1016/s0040-4039(00)00642-0.
Ito, K., Kamiyama, N., Nakanishi, S., & Otsuji, Y. (1983). Selec-
tive trimerization of aliphatic aldehydes catalyzed by polynu-
clear carbonylferrates. Chemistry Letters, 12, 657–660. DOI:
10.1246/cl.1983.657.
Jian, X. J., Zhang, J., Zhen, L. L., Liu, F. S., & Liu, Y. (2003a).
Chinese Patent No. 1429809A. Beijing, China: State Intellec-
tual Property Office of China.
alyzed by NaOHa
Aldehyde
Time/h
Products
Yieldb/%
Ib
Ic
Id
Ie
7
7
7
8
Vb, VIb
Vc, VIc
Vd, VId
Ve, VIe
63.3
59.7
58.2
51.8
a) Mole ratio of aldehyde/NaOH was 20 : 1 and reaction tem-
perature was 60◦C; b) yield determined by GC.
Jian, X. J., Zhang, J., Zhen, L. L., Liu, F. S., & Liu, Y. (2003b).
Chinese Patent No. 1429659A. Beijing, China: State Intellec-
tual Property Office of China.
Kulpinski, M. S., & Nord, F. F. (1943). Essential steps in the
catalytic condensation of aldehydes; New synthesis of gly-
col esters. Journal of Organic Chemistry, 8, 256–270. DOI:
10.1021/jo01191a007.
Mahrwald, R., Costisella, B., & Gu¨ndogan, B. (1998). Stereo-
control in aldol addition—synthesis of syn and anti 3-hydroxy
aldehydes. Synthesis, 1998, 262–264. DOI: 10.1055/s-1998-
2031.
The reaction is similar to the trimerization of Ia
(Fig. 1). It can also be explained by a three-step re-
action mechanism: (i) aldol condensation of aldehyde;
(ii) crossed Cannizzaro reaction; and (iii) esterification
of carboxylic acid and alcohol. In general, the longer
is the carbon chain of an aldehyde, the more difficult
is the reaction and the longer reaction time is needed
(Table 3).
Acknowledgements. The authors gratefully acknowledge the
funding support by a grant from the Research Fund for the Doc-
toral Program of Higher Education of China (20120101110062)
and The Low Carbon Fatty Amine Engineering Research Cen-
ter of the Zhejiang Province (2012E10033).
Miyano, A., Tashiro, D., Kawasaki, Y., Sakaguchi, S., & Ishii,
Y. (1998). Trimerization of aliphatic aldehydes to 1,3-diol
monoesters catalyzed by Cp*2Sm(thf)2. Tetrahedron Letters,
39, 6901–6902. DOI: 10.1016/s0040-4039(98)01447-6.
Swain, C. G., Powell, A. L., Sheppard, W. A., & Morgan, C.
R. (1979). Mechanism of the Cannizzaro reaction. Journal
of the American Chemical Society, 101, 3576–3583. DOI:
10.1021/ja00507a023.
References
Swan, P. (2005). An introduction to coalescing aids and East-
man film forming technologies. In Proceedings of the 36th In-
ternational Conference on Coatings Technology, May 23–25,
2005 (pp. 265–278). Pardubice: Czech Republic. Univerzita
Pardubice.
Takasu, T., Sasaki, T., & Yamamoto, A. (1972). U.S. Patent
No. 3,703,541. Washington, DC, USA: United States Patent
and Trademark Office.
Tic, W. J., Guzialowska-Tic, J., Sutor, E., & Bulanda, W.
(2012). European Patent No. EP 2436667. Munich, Germany:
European Patent Office.
T¨ormäkangas, O. P., & Koskinen, A. M. P. (2001). Fast aldol-
Tishchenko reaction utilizing 1,3-diol monoalcoholates as the
Bikbaeva, G. G., Kislina, I. S., & Vinnik, M. I. (1975). Mech-
anism of cross Cannizzaro reaction of 2,4-dimethoxybenz-
aldehyde with formaldehyde in aqueous KOH solutions. Bul-
letin of the Academy of Sciences of the USSR Division of
Chemical Science, 24, 1861–1865. DOI: 10.1007/bf00930152.
Duccini, Y., Sanfilippo, A., & Ugazio, S. P. J. (2007). European
Patent No. EP 1547573. Munich, Germany: European Patent
Office.
Fouquet, G., Merger, F.,
¨
Platz, R. (1979). Uber die
&
Tischtschenko-Reantion von Aldolen. Liebigs Annalen der
Chemie, 10, 1591–1601. DOI: 10.1002/jlac.197919791019. (in
German)
Gallagher, M., Dalton, P., Sitvarin, L., & Preti, G. (2008). Sen-

Y. G. Li et al./Chemical Papers
v
catalysts. Organic Process Research & Development, 5, 421–
425. DOI: 10.1021/op0001335.
Villacorta, G. M., & Filippo, J. S., Jr. (1983). Lithium tung-
sten dioxide promoted Claisen–Tishchenko condensation of
aromatic and aliphatic-aldehydes. Journal of Organic Chem-
istry, 48, 1151–1154. DOI: 10.1021/jo00156a001.
Villani, F. J., & Nord, F. F. (1947). Essential steps in the cat-
alytic condensation of carbonyl compounds. III. The action
of metallic alkoxides. Journal of the American Chemical So-
ciety, 69, 2605–2607. DOI: 10.1021/ja01203a011.