Selective hydrogenation of 3,3-dimethylbutanoyl chloride to 3,3-dimethylbutyraldehyde with silica supported Pd nanoparticle catalyst

Article abstract of DOI:10.1016/j.catcom.2011.01.025

A novel method for selective hydrogenation of 3,3-dimethylbutanoyl chloride (DMBC) to 3,3-dimethylbutyraldehyde (DMBA) with silica supported Pd nanoparticle catalyst (Pd/SiO₂) is developed. The catalysts were characterized by Fourier transform infrared spectroscopy, X-ray powder diffraction, N₂ physisorption and transmission electron microscopy. The performance of the Pd/SiO₂ catalyst was compared with Pd/C and Pd/BaSO₄ catalysts with or without being pretreated by quinoline-sulfur. The Pd/SiO₂ catalyst activated at 80 °C by bubbling hydrogen in cyclohexane for 1 h showed the highest yield of DMBA. For 3 wt.% Pd/SiO₂, the yield of DMBA reached 84.6%, which exhibited much higher value than Pd/C and Pd/BaSO₄ catalysts.

Full text of DOI:10.1016/j.catcom.2011.01.025

Catalysis Communications 12 (2011) 813–816
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Selective hydrogenation of 3,3-dimethylbutanoyl chloride to
3,3-dimethylbutyraldehyde with silica supported Pd nanoparticle catalyst
⁎
Sifang Li , Guoqin Chen, Lan Sun
Department of Chemical and Biochemical Engineering, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel method for selective hydrogenation of 3,3-dimethylbutanoyl chloride (DMBC) to 3,3-dimethylbutyr-
aldehyde (DMBA) with silica supported Pd nanoparticle catalyst (Pd/SiO2) is developed. The catalysts were
characterized by Fourier transform infrared spectroscopy, X-ray powder diffraction, N2 physisorption and
transmission electron microscopy. The performance of the Pd/SiO2 catalyst was compared with Pd/C and
Pd/BaSO4 catalysts with or without being pretreated by quinoline–sulfur. The Pd/SiO2 catalyst activated at
80 °C by bubbling hydrogen in cyclohexane for 1 h showed the highest yield of DMBA. For 3 wt.% Pd/SiO2, the
yield of DMBA reached 84.6%, which exhibited much higher value than Pd/C and Pd/BaSO4 catalysts.
© 2011 Published by Elsevier B.V.
Received 24 November 2010
Received in revised form 21 January 2011
Accepted 25 January 2011
Available online 3 February 2011
Keywords:
Hydrogenation
3,3-dimethylbutyraldehyde
Catalysis
Pd nanoparticle
1. Introduction
catalyst. Therefore, attempts to dispense with the use of catalyst
poisons acquire considerable significance. It is known to control the
3,3-Dimethylbutyraldehyde (DMBA) is an important intermediate
for the preparation of neotame which is known as an extremely
potent sweetener because its sweetening potency has been reported
to be 8000 times that of sucrose [1]. In addition, DMBA is also used to
produce pharmaceuticals such as pyrone derivatives [2]. DMBA can be
synthesized by the following methods: (a) oxidation of 1-chloro-3,
3-dimethylbutane with DMSO [3]; (b) hydrolysis of 1,1-dichloro-3,
3-dimethylbutane [4]; (c) oxidation of 3,3-dimethylbutanol with
2,2,6,6-tetramethyl-1-piperidinyloxy and a free radical [5]; (d) re-
duction of 3,3-dimethylbutyric acid with palladium acetate and tri-
p-tolyphosphine [6]. Each of these reported protocols suffered from
the use of expensive reagents or low product yields. Therefore, the
development of simple, convenient and economical approaches for
the synthesis of DMBA is still needed.
Aldehydes can be synthesized by Rosenmund reduction of the
corresponding acid chlorides in the presence of a supported Pd
catalyst. Pospisek and Blaha [7] reported preparation of DMBA by
hydrogenation of 3,3-dimethylbutanoyl chloride (DMBC) in the
presence of Pd/BaSO4 catalyst. It is known that a suitable catalyst
poison such as quinoline–sulfur is often used to prevent successive
hydrogenation of aldehyde to alcohol in Rosenmund reduction [8–15].
Catalyst poison deactivates the catalyst for undesired reaction but
causes much difficulty in purification of the aldehyde and reuse of the
structure [16], particle size [17] and pore size [18] of a metal oxide by
using organic compounds in sol–gel method. Ethylene glycol and
acetylacetone were used in the preparation of Pd/SiO2 catalyst [19].
Residual organics in the catalyst promoted the selectivity in the
hydrogenation of benzoyl chloride [20].
The aim of this research is to develop a method for synthesis of
DMBA by selective hydrogenation of DMBC in the presence of silica
supported Pd nanoparticle catalyst. Polyethylene glycol (PEG 200)
and tetrahydrofuran were used in the preparation of the catalyst via
sol–gel method. X-ray powder diffraction (XRD), transmission
electron microscope (TEM), N2 physisorption and Fourier transform
infrared spectroscopy (FT-IR) were used to characterize the catalysts.
The selectivity of the Pd/SiO2 was compared with Pd/C and Pd/BaSO4.
The effect of activation conditions on the catalytic activity was
investigated.
2. Experimental section
2.1. Materials
3 wt.% Pd/BaSO4 was prepared according to reference [21], 3 wt.%
Pd/C was procured from Baoji Rock Nonferrous Metal CO. Ltd., China.
Dinitrodiamminepalladium, 3,3-dimethylbutyric acid, thionyl chlo-
ride, benzene, tetraethylorthosilicate (TEOS), ethanol, PEG 200,
tetrahydrofuran and cyclohexane were of analytical grade. H2 and
N2 (both purities were more than 99.99%) were obtained from Linde
Gases CO. Ltd., China.
⁎
Corresponding author. Tel./fax: +86 592 2186195.
E-mail address: sfli@xmu.edu.cn (S. Li).
1566-7367/$ – see front matter © 2011 Published by Elsevier B.V.
doi:10.1016/j.catcom.2011.01.025

814
S. Li et al. / Catalysis Communications 12 (2011) 813–816
2.2. Synthesis of DMBC
Then the solution was added to 7.3 ml of TEOS dissolved in ethanol
(4.5 ml) and stirred for 1 h at 80 °C. A mixture of deionized water
(2 ml) and ethanol (2 ml) was added and the resulting viscous
solution was aged at 80 °C for 24 h to turn into a gel. The gel was dried
at 80 °C in vacuum for 8 h. Prior to hydrogenation reaction, some
catalysts were activated in an atmosphere of hydrogen at different
temperatures (80 °C, 160 °C and 200 °C) for 4 h, while some other
samples were activated in cyclohexane by bubbling hydrogen into the
solution at 80 °C for 1 h. The Pd/C and Pd/BaSO4 catalysts with or
without quinoline–sulfur were also used for comparison. The
inhibited catalyst was prepared by adding 20 μl of quinoline–sulfur
(containing quinoline 85.7 g/l and sulfur 14.3 g/l) for 1 g catalyst to
the reaction mixture.
In a 500 ml three-necked flask equipped with reflux condenser,
magnetic stirrer and thermometer, 100 ml of thionyl chloride was
added dropwise to 3,3-dimethylbutyric acid (117 g). The mixture was
refluxed at 80 °C in an oil bath for 3 h. Sulfur dioxide and hydrogen
chloride escaping from the condenser were absorbed in the alkali
solution. After the completion of the reaction, the residual thionyl
chloride wasremoved under vacuum. Then 20 ml of anhydrous benzene
was added, and the mixture of benzene and thionyl chloride was
removed under vacuum. This process was repeated 3 times. Finally,
DMBC was distilled under reduced pressure (20 kPa) at 79–81 °C.
2.3. Catalyst preparation
2.4. Catalyst characterization
3 wt.% Pd/SiO2 catalysts were prepared via sol–gel method.
Dinitrodiamminepalladium and TEOS were used as precursors for
palladium and silica. PEG 200 and tetrahydrofuran were employed as
the Pd-dispersant. The catalyst was prepared as follows: 0.13 g of
dinitrodiamminepalladium was added into a mixture of PEG 200
(7 ml) and tetrahydrofuran (1.5 ml) at 80 °C under vigorous stirring.
Fourier transform infrared spectroscopy (FT-IR) of the catalysts
was measured by KBr pellet method on a Nicolet Nexus FT-IR
spectrophotometer in the wave number range 4000–450 cm−1. X-ray
powder diffraction (XRD) of the samples was recorded on an X'Pert
PRO X-ray diffractometer with Cu Kα radiation (λ=0.15406 nm)
Fig. 1. TEM images of 3 wt.% Pd/SiO2 catalysts activated at different conditions: (a) activated at 80 °C in cyclohexane; (b) activated at 80 °C; (c) activated at 160 °C; (d) activated at
200 °C.

S. Li et al. / Catalysis Communications 12 (2011) 813–816
815
Table 1
with a scanning angle (2θ) range from 10° to 60°, voltage and current
Physical property of Pd/SiO2 catalysts.
of 40 kV and 30 mA. Surface areas calculated by the BET method were
determined from nitrogen adsorption–desorption isotherms at liquid
nitrogen temperature by using a Micromeritics TriStar 3000 instru-
ment. The morphologies of the catalysts were characterized by a
TECNAI F30 HRTEM transmission electron microscope (TEM).
Catalyst
Activation
temperature (°C)
BET surface
area (m2/g)
Pore volume
(cm3/g)
Average pore
diameter (nm)
Pd/SiO2
80a
80
160
200
80a
80a
264
251
296
317
930
127
0.43
0.57
0.63
0.64
0.62
0.46
27.96
26.22
25.19
25.07
22.04
29.13
Pd/C
Pd/BaSO4
2.5. Hydrogenation of DMBC
a
A three-necked flask (250 ml) equipped with reflux condenser,
mechanical stirrer and thermometer was used as the reactor. 100 g of
cyclohexane was charged into the reactor, and 2 g of the catalyst was
suspended in the cyclohexane via vigorous stirring under a nitrogen
atmosphere. The mixture was heated up to 80 °C by immersing the
reactor in a thermostatic oil bath. Then 20 g of DMBC was added and
hydrogen saturated with the solvent vapor in advance was bubbled
through the mixture at a steady flow rate at atmospheric pressure.
Hydrogen chloride from the reaction was trapped in a water sink. The
conversion of DMBC was measured by titrating the hydrogen chloride
absorbed in the water with NaOH solution. Samples of the reaction
mixture were withdrawn periodically for analysis. The yield of DMBA
was determined by mixing 0.5 g sample with saturated 2,4-dinitro-
phenylhydrazine solution to form a yellow precipitate of 2,4-
dinitrophenylhydrazone which was collected, dried and weighed
[12,22].
Catalyst was activated in cyclohexane.
particle size and distribution to Pd/SiO2, but Pd/BaSO4 has larger Pd
particle size and wider distribution than Pd/SiO2.
Table 1 shows the surface areas of 3 wt.% Pd/SiO2 activated at
80 °C, 160 °C and 200 °C, which were calculated by the BET method.
Pd/SiO2 activated at 80 °C in cyclohexane shows slightly larger surface
area than that activated at the same temperature without the solvent.
As the temperature is elevated, surface areas and pore volumes of the
catalysts increase, while the average pore diameters decrease. The
increase of the surface areas can be attributed to the loss of organics
on the surface of the catalyst.
Fig. 3 shows the XRD patterns of the support silica and Pd/SiO2
catalysts activated under different conditions. From XRD, peaks due to
metallic palladium (for example, Pd (111) and (200) around 2θ=40.6
and 47, respectively) become a little sharper with an elevation of the
activation temperature, indicating that the palladium particles grow.
The average size of palladium particles, which is calculated from the
full width at half maximum of the XRD peak at 2θ=40.6 using the
Scherer equation [23], increases from 4 nm to 9 nm with the elevation
of the activation temperature from 80 °C to 200 °C. These XRD
observations are roughly consistent with the TEM results.
FT-IR spectra of the Pd/SiO2 catalysts were recorded (Fig. 4). The FT-
IR spectra showed the typical vibration modes from silica, namely: υas
Si―O―Si at 1088 cm−1, υ Si―O (H) at 969 cm−1, υs Si―O―Si at
795 cm−1 and δ Si―O―Si at 463 cm−1. In addition, the corresponding
vibrations are due tothe presence of organic compounds in the catalysts.
For example, υas CH3 and υasCH2 at 2955 and 2886 cm−1 could be
identified, respectively. The presence of the organic residues in the
studied samples could be correlated to the organics used in the sol–gel
process. Even after seven runs, the catalyst still shows organic residues
from FT-IR curves in Fig. 4(d). Thus, it can be concluded that organic
residues could be stable embedded in the silica gel matrices.
3. Results and discussion
3.1. Catalyst characterization
Fig. 1 shows TEM images of 3 wt.% Pd/SiO2 activated at 80 °C in
cyclohexane as well as 80 °C, 160 °C and 200 °C without the solvent.
The TEM images reveal the dispersion of Pd nanoparticles on silica.
The Pd particle size and distribution were found to vary depending on
the activation conditions employed in the preparation process (Fig. 2).
Pd/SiO2 activated at 80 °C in cyclohexane shows the narrowest
particle distribution, while that activated at 200 °C exhibits the
widest particle distribution. As the Pd/SiO2 was activated at 80 °C in
cyclohexane, the particles were suspended in the liquid, which
avoided the aggregation of palladium due to good mass and heat
transfer. The average Pd particle size increases from 3.4 nm to 7.0 nm
with the elevation of the activation temperature from 80 °C to 200 °C.
This is probably because the organic residues in the surface of the
catalyst decrease during the heating and result in aggregation of
palladium particles. It is worth mentioning that Pd/C has similar Pd
40
80°C (cyclohexane)
e
d
80°C
160°C
30
200°C
c
20
b
10
0
a
10 15 20 25 30 35 40 45 50 55 60
2θ(°)
0
2
4
6
8
10
12
14
16
18
Particle size (nm)
Fig. 3. XRD patterns of 3 wt.% Pd/SiO2 catalysts treated under different conditions: (a)
support; (b) activated at 80 °C in cyclohexane; (c) activated at 80 °C; (d) activated at
160 °C; (e) activated at 200 °C.
Fig. 2. Particle size distribution of catalysts activated at different conditions.

816
S. Li et al. / Catalysis Communications 12 (2011) 813–816
Table 3
a
b
Effect of activation conditions on the hydrogenation of DMBC.
Activation temperature (°C)
Conversion
(%)
Selective
(%)
Yield
(%)
c
80a
80
160
200
93
92
94
95
91
87
80
75
84.6
80.0
75.2
71.3
d
DMBC 20 g; catalyst dosage, 10 wt.% based on DMBC; solvent, cyclohexane, 100 g;
temperature, 80 °C; hydrogen flow rate, 120 ml/min; stirring rate, 1000 rpm; reaction
time, 6 h.
a
Catalyst was activated in cyclohexane.
was measured to be 259 m2/g after seven cycles of reaction, which
showed no significant changes.
4000
3500
3000
2500
2000
1500
1000
500
Wavenumbers (cm-1)
Fig. 4. FT-IR spectra of 3 wt.% Pd/SiO2 catalysts: (a) activated at 80 °C in cyclohexane;
(b) activated at 160 °C; (c) activated at 200 °C; (d) after seven runs.
4. Conclusions
DMBA was synthesized with high yields via selective hydrogena-
tion of DMBC using Pd/SiO2 catalyst prepared by sol–gel method. TEM
and XRD indicate that Pd nanoparticles were highly dispersed on the
silica. Narrow particle size distribution was observed for Pd/SiO2
catalyst activated at 80 °C by bubbling hydrogen in cyclohexane. FT-IR
demonstrates the presence of the organic residues in the catalyst. The
yield of DMBA could achieve 84.6% by 3 wt.% Pd/SiO2 catalyst, which
showed much higher catalytic activity than Pd/C and Pd/BaSO4
catalysts.
3.2. Catalytic activity
In the hydrogenation over a supported Pd catalyst, mass transfer
resistance may be serious [24,25]. In order to overcome the mass
transfer resistance, the effects of stirring rate and hydrogen flow rate
were studied. Above the stirring rate of 1000 rpm and the hydrogen
flow rate of 120 ml/min, there were no significant changes in the
conversions of DMBC, which indicates the absence of mass transfer
resistance.
It is known that the nature of a support have a significant effect on
the catalytic performance [26,27]. In Rosenmund reduction, the most
common supports used are activated carbon and barium sulfate. The
performance of Pd/SiO2 catalyst in the hydrogenation of DMBC to
DMBA was compared with Pd/C and Pd/BaSO4 catalysts. Results are
summarized in Table 2. Pd/SiO2 catalyst exhibits much higher
selectivity and yield of DMBA than Pd/C and Pd/BaSO4 catalysts
with or without catalyst poison. The main by-product is 3,3-
dimethylbutanol, which is formed by the subsequent reduction of
DMBA. It is evident that a suitable catalyst poison such as quinoline–
sulfur will enhance the selectivity with a lower conversion.
Table 3 shows the results of hydrogenation of DMBC over 3 wt.%
Pd/SiO2 catalysts activated under different conditions. The catalyst
performance was remarkably affected by the activation conditions.
The catalyst activated in cyclohexane at 80 °C for 1 h showed the
highest catalyst performance. Increase of the activation temperature
of the catalyst was accompanied with the decrease of selectivity to
DMBA. The yield of DMBA dramatically decreased for the catalyst
activated at 200 °C, which was attributed to the growth and
aggregation of palladium particles as shown in Fig. 1.
Acknowledgement
The authors thank Xiamen Fine Chemical Technology Company for
the financial support of this study.
References
[1] C. Nofre, J.M. Tinti, Food Chem. 69 (2000) 245–257.
[2] J. Domagala, M. Pyrone, US 5,808,062 (1998).
[3] Z. Guo, R. Sawyer, I. Prakash, Synth. Commun. 31 (2001) 3395–3399.
[4] I. Prakash, H. Estates, Z. Guo, US 5,994,593 (1999).
[5] I. Prakash, WO 2,005,082,825 (2005).
[6] I. Prakash, L.A. Robert, K.S. Tanielyan, US 7,030,281 (2006).
[7] J. Pospisek, K. Blaha, Coll. Czech. Chem. Commun. 52 (1987) 514–521.
[8] K.W. Rosenmund, F. Zetzsche, Ber. Dtsch. Chem. Ges. 51 (1918) 585–594.
[9] E.B. Hershberg, J. Cason, Org. Synth. 21 (1941) 84–88.
[10] T. Ito, K. Watanabe, Bull. Chem. Soc. Jpn. 41 (1968) 419–423.
[11] A.W. Burgstahler, L.O. Weigel, C.G. Shaefer, Synthesis 11 (1976) 767–768.
[12] S. Affrossman, S.J. Thomson, J. Chem. Soc. 5 (1962) 2024–2029.
[13] J.A. Peters, H.V. Bekkum, Recl. Trav. Chim. Pays-Bas 100 (1981) 21–24.
[14] J.A. Peters, H.V. Bekkum, Recl. Trav. Chim. Pays-Bas 90 (1971) 1323–1325.
[15] A.I. Rachlin, H. Gurien, D.P. Wagner, Org. Synth. 51 (1971) 8–9.
[16] C.M. Liu, X.T. Zu, Q.M. Wei, L.M. Wang, J. Phys. D: Appl. Phys. 39 (2006)
2494–2497.
[17] M. Galceran, M.C. Pujol, M. Aguiló, F. Díaz, J. Sol–Gel Sci. Technol. 42 (2007) 79–88.
[18] R. Linacero, M.L. Rojas-Cervantes, J.D. López-González, J. Mater. Sci. 35 (2000)
3269–3278.
[19] S. Tanaka, F. Mizukami, S. Niwa, M. Toba, K. Maeda, H. Shimada, K. Kunimori, Appl.
Catal. A 229 (2002) 165–174.
Reuse of the catalyst is of great industrial relevance, therefore,
3 wt.% Pd/SiO2 activated at 80 °C in cyclohexane was used to test the
reusability of the catalyst. The Pd/SiO2 catalyst was subjected to seven
consecutive runs under the same conditions as Table 3. The
conversion was still above 90% and the surface area of the catalyst
[20] S. Tanaka, F. Mizukami, S. Niwa, M. Toba, G. Tasi, H. Shimada, K. Kunimori, Appl.
Catal. A 229 (2002) 175–180.
[21] A.I. Vogel, Vogel's Textbook of Practical Organic Chemistry, 5th ed.Longman,
London, 1989.
[22] X. Jia, X. Liu, J. Li, P. Zhao, Y. Zhang, Tetrahedron Lett. 48 (2007) 971–974.
[23] M. Bhagwat, P. Shah, V. Ramaswamy, Mater. Lett. 57 (2003) 1604–1611.
[24] V.G. Yadav, S.B. Chandalia, Org. Proc. Res. Dev. 1 (1997) 226–232.
[25] J.A. Bennett, R.P. Fishwick, R. Spence, J. Wood, J.M. Winterbottom, S.D. Jackson, E.
H. Stitt, Appl. Catal. A 364 (2009) 57–64.
[26] J. Panpranot, K. Pattamakomsan, J.G. Goodwin, P. Praserthdam, Catal. Commun. 5
(2004) 583–590.
[27] A. Quintanilla, J.J.W. Bakker, M.T. Kreutzer, J.A. Moulijn, F. Kapteijn, J. Catal. 257
(2008) 55–63.
Table 2
Hydrogenation of DMBC using various Pd catalysts.
Catalysts
Reaction time
(h)
Conversion
(%)
Selectivity
(%)
Yield
(%)
Pd/SiO2
Pd/C
6
6
6
10
10
93
89
86
81
80
91
75
74
88
84
84.6
66.8
63.6
71.3
67.2
Pd/BaSO4
Pd/Ca
a
Pd/BaSO4
DMBC 20 g; catalyst dosage, 10 wt.% based on DMBC; solvent, cyclohexane, 100 g;
temperature, 80 °C; hydrogen flow rate, 120 ml/min; stirring rate, 1000 rpm.
a
Catalysts with quinoline–sulfur.