Science and Technology in Catalysis 2002 443

Copyright 9 2003 by Kodansha Ltd.

98
Improved Base Catalysts for industrial
Condensation Reactions

F King, G J Kelly, E H Stitt

Synetix, PO Box 1, Belasis Avenue, Billingham, Cleveland, TS23 1LB

1. INTRODUCTION
Industrially condensation reactions are of great importance in the production of
a number of key compounds. These include 2-ethyl hexanol, methyl isobutyl ketone
(MIBK) and Guerbet alcohols. Over 1.5 million tonnes of these chemicals are
produced worldwide every year using homogeneous bases such as NaOH and
Ca(OH)2 [ 1]. It has been estimated for these compounds [2] that 30% of the selling
price is product purification, recovery and waste treatment. For every 10 tonnes of
product formed the current homogeneous catalysts generate about 1 tonne of spent
catalyst. High capital costs are also associated with the handling of strong
homogeneous bases such as 30% caustic. The sustainability of the existing process
can be greatly improved by the use of a fixed-bed solid base catalyst especially if used
in conjunction with a fixed-bed hydrogenation catalyst.

1.1 Solid Base Catalysts

Solid base catalysts have a number of advantages over conventional
homogeneous (NaOH, KOH, Ca(OH)2) systems as follows:J3,4] High catalytic
activity and selectivity are often observed, solid base catalysts do not corrode reaction
vessels or reactors, repeated use is possible, easier separation from the products, less
waste than produced from stoichiometfic reagents, lower operation and maintenance
costs, no acid or base containing waste streams, lower energy synthesis, solid base
catalysts do not catalyse cracking reactions. Although solid base catalysts have a
much lower environmental impact than conventional liquid systems, this alone is
often not enough to persuade users to switch from liquid to solid systems. To switch
from a liquid based system to a solid system the user has to see the advantages of
using solid systems in terms of quality of product, selectivity of process and the
overall process economics and for new plants in reduced capital. Solid base catalysis
is an area of chemistry that offers an opportunity for exploitation if suitable catalysts
and processes can be identified and developed [5].

1.2 Reactions Networks

The product from the aldol condensation of n-butanal is usually an
ct,~-unsaturated aldehyde, 2-ethyl hexenal, from the dehydration of the aldol product.
Cu or Ni catalysts are the conventional catalysts for fully hydrogenating the
ct,15-unsaturated aldehyde through to the saturated alcohol [6]. In the presence of a
444 F. King et al.

base butanone is capable of losing a proton from the adjacent methylene group or the
terminal methyl group, Tautomerisation of the two possible carbanions formed from
the proton abstraction leads to two regioisomeric enolates. The enolates formed can
then attack a polarised carbonyl group to form a carbon-carbon bond to form two
aldol products. Starting from butanone the point of proton abstraction and the
direction of dehydration can lead to a wide range of possible products.

2. E X P E R I M E N T A L
2.1 Catalyst Preparation
A 4wt% Na/SiO2 catalyst was prepared by an impregnation method using nitrate
salts. Fuji Q10 silica spheres (2-3 mm) were used which have a surface area of 359 m 2
g-1 and a pore volume 1.01 cm 3 gl. A Pd doped version of this catalyst was also
prepared by co-impregnating the Fuji Q 10 silica with a mixture of Na and Pd nitrates
to produce a catalyst of the following formulation 0.1 wt% Pd/4wt% NaJSiO2. Both
catalysts were dried overnight at 100 ~ after impregnation and calcined at 450 ~ for
3 hours. A CuO/ZnO hydrogenation catalyst was prepared by co-precipitation and
contained 35 wt% CuO.

2.2 Catalyst Testing

Reaction testing was carried out in a stainless steel fixed bed microreactor
system. The catalysts were crushed to a particle size of 0.6 to l m m before testing. To
carry out the experiments the catalysts were brought up to reaction temperature in a
flow of carrier gas (H2 or N2 - 50 ml rain "t) before initiating the organic feed. Before
commencing testing, the CuO/ZnO hydrogenation catalyst, where used, was reduced
in a 10% H//N2 flow at 160 ~ over a period of 24 hrs. In the single bed experiments
with aldehyde a 3 ml bed of catalyst was used. In the single bed experiments with
ketone a 4 ml bed of catalyst was used. In the dual bed experiment with the ketone
two 2 ml beds of catalyst were used which were touching.

3. R E S U L T S AND DISCUSSION
3.1 Aldehyde reactions
The aldol condensation reaction of n-butanal was carried out over the 4wt%
Na/SiO2 and the 0. lwt% P d / 4 wt% Na/SiO2 catalysts. The system was rart at 7 barg
and a n-butanal liquid feed of 0.05 ml min t . The results are shown in Tables 1 and 2.

Time Temp H2 flow rate Conversion Selectivity Product Spectrum

(hrs) (~ (ml min"a) (%) (%) % 2-ethyl ....
% 2-ethyl % 2-ethyl
hexenal hexanal hexanoi
2 350 42 . . . . . . .
45
. . . . . .
72 99.0 0.5 0.5
4 350 42 58 86 98.4 1,3
. . .
0.3
.

6 350 42 50 89 99.5 0.2 0.3

22 350 42 ...... 38 .L 9~.L 99.2 0.6 0,2
30 350 21 40 91 i'6b ~-
,.
Off 0.0
46 350 21 41 92 99.5 0.5 0.0
53 400 " 21 43 I00 99.8. . . .
0.2 0.0
70 400 21 40 89 . . . .
99.6 0.2 0.2
72 450 21 47 79 99.7 0.3 0.0

Table I Reaction of n-butanal over Na/SiO2 catalyst

 445

Time Temp H2 flow rate Conversion Selectivity Product Spectrum

(hrs) (~ (ml min "l) (%) (%) % 2-ethyl % 2-ethyl % 2-ethyl
hexenal hexanal hexanol
4 350 50 u ,
82 81 . . . . . 30.2 69.2 0.6
6 350 50 . .
58
. . . .
92 13.6 86.0 0.4
8 350 50 . .
49 . .
94 7.0 92.5 0.5
24 350 50 42 100 4.5 94.9 0.6
31 350 25 39 96 ........
6.6 92.6 0.8
48 350 25 34 100 10.6 88.6 0.8
55 400 25 . . . . . .
40 . . . . . . .
86 12.2 87.0 0.8
72 400 25 32 92 16.2 83.1 0.7
77 450 25 38 75 25.4 73.9 0.7

Table 2 Reaction of n-butanal over PdfNa/SiO2 catalyst

Although the O. l wt% Pd containing catalyst initially has a higher conversion of

n-butanal it undergoes a more rapid deactivation than the Na/SiO2 catalyst. The other
major difference between the two catalysts is in the product spectrum achieved. The
Na/SiO2 catalyst produces almost exclusively (>99%) the ot,13-unsaturated aldehyde
despite carrying out the reaction in a hydrogen carrier stream. The 0. lwt% Pd
containing catalyst produces a high selectivity towards the saturated aldehyde product
2-ethyl hexanal (max 94.9%) with less than 2% of the butanal feed hydrogenated to
either n-butanol or 2-ethyl hexanol [6]. The reason for the more rapid deactivation of
the Pd/Na/SiO2 catalyst are not clear.

3.2 Ketone Reactions

The aldol condensation reaction ofbutanone was carried out over the 4wt%
Na/SiO2 catalyst (Table 3). The reaction was ran at 5 barg, a butanone liquid feed of
0.05 ml min "1 and a N2 carrier flow of 50 ml min "1. The major product was 5-methyl-
4-hepten-3-one, with both isomers being formed. This indicates an initial proton
abstraction from the terminal methyl group of MEK. Only very low levels of the
comparable 3,4-dimethyl-4-hexen-2-one product formed via the abstraction of the
internal proton were detected. Two major unidentified condensation products with
molecular weight of 126 were detected. The reaction ofbutanone over a dual bed
reactor stream with a 4wt% Na/SiO2 catalyst followed by a bed of CuO/ZnO
hydrogenation catalyst produces essentially two main components; 5-methylheptan-3-
one and 5-methylheptan-3-ol (Table 4). The former is the partially hydrogenated form
of the latter and therefore both compounds can be regarded as being derived from 5-
methyl-4-hepten-3-one. It can be also inferred from this result that the two
unidentified aldol products with molecular weight 126 produced over the Na/SiO2
catalysts are cis and trans-5-methyl-5-hepten-3-one because subsequent
hydrogenation of these compounds results in the formation of 5-methylheptan-3-ol.
The dual bed reactor system appears to be more active than the base catalysts alone
[7].
% SELECTIVITY
Component 325~ 350oC 3§162 400~
Conversion (%) 1.3 4.1 "'6.S "
t
.
7.8
3,4-dimethy.l-47hexen-2-one 1.0 1.3 1.5 1.5
5-methyleneheptan-3-one 2.1 3.1 3.5 3.5
Unknown 1(_!26) == 10.5 14.1 14.9 12.2
5-methyl-4-hepten-3-one(a) 15.5 19.2 19.9 19.3
446 F. King et al.

" Unknown 2(126) " 19.7 18.0

5-methy!~-hepten-3-one(b). 24.4 " 32.0 "31.0 " 33.0
3-methyl-3-hepte.n-2-one.. 4.3 . ..0.9 0.1 0.0
U.nknown 3(I 26) 8.9 0.6 . .
0.0 . .
0.0
Unknown 4(180) 0.9 2.8 6.5 10.9
L
Unknown 5(180) 7.5 ......2.3 0.9 0.5
Unknown 6(180).. 9.9 4.0 1.6 1.I

Table 3 Reaction ofbutanone over 4wt% Na/SiO2

% SELECTIVITY
Component . . . . . 300 C 325~ 350~ 375~
Conversion (%) '~' .L 4.5 ." 7.7 10.1 10.6
3,4-di.methylhexan-2-one 0.4 0.6 0.8 1.0 . . . .

5-methylheptan-3-one. 16.5 24.4 31.6 44.6

5-methylheptan-3-ol 80.9 73.4 65.1 54.3
.. Unknown1(12.6) . . . . . . 0.5 0.5 0.6 0.0
Unknown 2(126) 0.6 0.6 0.9 0.0
Unknown 3(126) 1.1 0.5 1.1 0.1

Table 4 Reaction ofbutanone over 4wt% Na/SiO2 followed by CuO/ZnO

catalyst

4. Conclusions
The viability has been demonstrated for solid fixed-bed basic materials to carry
out industrially important aldol condensations. This gives the opportunity to eliminate
the use of 30% caustic solution and dramatically reduce the amount of organically
contaminated waste caustic produced. The solid base catalysts can be run
continuously on-line for hundreds on hours at high conversions and selectivities. The
use of solid base catalysts also allows the easier integration of the condensation step
with any subsequent hydrogenation steps. The reaction of an aldehyde over a
Na/SiO2 solid base catalyst produces an c~,13unsaturated aldehyde intermediate. This
can be selectively hydrogenated to the saturated aldehyde by the addition of small
amounts of Pd to the solid base catalyst or fully hydrogenated to an alcohol over a
conventional CuO/ZnO hydrogenation catalyst. With the reaction ofbutanone there is
a high selectivity (max >96%) for reaction at the terminal position of the
unsymmetrical ketone. The combination of aldol condensation and hydrogenation
offers the possibility of the removal of a process step and consequent capital saving.

References
[ i ] K. Weisssermel and H.-J. Arpe, Industrial Organic Chemistry, Wiley, 1997
[2] J. J. Spivey and M. R. Gogate, Research Triangle Institute, US EPA Grant,
Pollution Prevention in Industrial Condensation Reactions, 1996
[3] K. Tanabe, Solid Acids and Bases, Kodansha, Tokyo and Academic Press,
New York London, 1970
[4] H. Hattori, Heterogeneous Catalysis and Fine Chemicals III, Elsevier, 1993
[5] G J Kelly, F King and M Kett, 'Waste Elimination in Condensation Reactions of
Industrial Importance', Green Chemistry, in press
[6] G. J. Kelly, Patent Application number WO 00/31011, June 2000
[7] G. J. Kelly, Patent Application number WO01/87812, Nov 2001