Ind. Eng. Chem. Res.

2010, 49, 8125–8134 8125

Kinetics of Heterogeneous Catalytic Epoxidation of Propene with Hydrogen

Peroxide over Titanium Silicalite (TS-1)
Sang Baek Shin and David Chadwick*
Department of Chemical Engineering, Imperial College London, London SW7 2AZ, United Kingdom

The kinetics of the heterogeneous catalytic epoxidation of propene to propene oxide with hydrogen peroxide and
titanium silicalite (TS-1) as the catalyst was studied under mild conditions in an autoclave reactor using methanol/
water mixtures as the solvent. The effects of reactant concentration, solvent composition, catalyst loading, stirring
speed, and reaction temperature on the propene oxidation were investigated. The initial rates increased almost
linearly with catalyst loading. The apparent orders of reaction with respect to hydrogen peroxide and propene
were found to be 0.67 (0.17-0.70 wt %) and 0.63 (0.3-0.8 bar), respectively. The apparent activation energy was
found to be 25.8 kJ mol-1. Methanol content in the solvent was shown to have a dual influence on the reaction
rate, enhancing the solubility of propene and inhibiting the epoxidation reaction by competitive adsorption.

1. Introduction only a limited number of works have been devoted to detailed

Propene oxide is an important chemical intermediate in the
chemical industry that is mainly used for the production of
polyether polyols, propene glycols, and propene glycol ethers.1,2
The propene oxide industry has employed two different types of
commercial processes for several decades: the chlorohydrin process
and the hydroperoxidation process. However, both of these
processes have multiple reaction steps and several essential
disadvantages: The chlorohydrin process suffers from environ-
mental liabilities caused by a toxic oxidant and waste. On the other
hand, the economy of the hydroperoxidation process primarily
depends on the marketability of coproducts. Therefore, direct
epoxidation of propene with ex situ or in situ produced hydrogen
peroxide has recently attracted much attention as a more environ-
mentally benign and profitable chemical process.1-10 In this
approach, the reaction is carried out under mild conditions, and
theoretically, only water is generated as byproduct. Even though
the first commercial-scale propene oxide plant based on hydrogen
peroxide recently started its operation,11 a number of studies are
still being conducted to gain a better understanding of catalysis
and develop more efficient reactor concepts. It seems clear that
the next generation after the hydrogen peroxide-to-propene oxide
(HPPO) process would be the direct gas-phase epoxidation of
propene by oxygen or a mixture of oxygen and hydrogen, but
currently available catalysts have difficulty in attaining sufficient
conversion for commercial application.1,2,7,12-19
Titanium silicalite (TS-1) catalyst has been widely investigated
because it can promote the formation of propene oxide with a high
selectivity of greater than 95% and a high hydrogen peroxide
conversion.3,4,9,10,20-27 Clerici and co-workers first demonstrated
that TS-1 can be used efficiently as a catalyst for the epoxidation
of various olefins, including propene, using hydrogen peroxide in
methanol under mild conditions.20 Most TS-1 catalysts have been
prepared following the procedure described in the original patent,28
but a variety of attempts have also been made to find easier and
less expensive methods for preparing TS-1 catalyst.29-37 Several
catalysts involving Pd and Pt supported on TS-1 have also been
studied for the liquid-phase epoxidation of propene and olefins.4,38-47
Although the liquid-phase epoxidation of propene using hydrogen
peroxide and TS-1 catalyst has been studied for several decades,
* To whom correspondence should be addressed. Tel.: +44 (0)20
7594 5579. Fax: +44 (0)20 7594 5637. E-mail: d.chadwick@
imperial.ac.uk.
10.1021/ie100083u  2010 American Chemical Society
Published on Web 07/27/2010
kinetics studies.48,49
In this article, we report the kinetics of the epoxidation of
propene to propene oxide using hydrogen peroxide as the oxidant,
methanol/water as the solvent, and titanium silicalite as the catalyst
under operating conditions different from those used in previous
studies. In particular, the reaction conditions were extended to lower
concentrations of reactants and milder pressures. Previous studies
have examined methanol/water and isopropanol solvents,48,49 which
were found to increase the reaction rate as a result of the enhanced
solubility of propene. In the present study, unexpected results were
obtained concerning the effect of the methanol content where
methanol was found to inhibit the initial rates. It is shown that this
effect can be explained by a reaction model that includes a term
for methanol competitive adsorption and allows for the increased
solubility of propene as a function of methanol concentration.
Furthermore, the data are demonstrated to favor a Langmuir-
Hinshelwood-type formulation of the rate equation.
2. Experimental Section
2.1. Catalyst Preparation. TS-1 catalyst was prepared ac-
cording to the method described in the literature.20,28 A solution
of 1.5 g of tetraethyl orthotitanate (TEOT, g95.0%, Fluka) was
gradually added to 45 g of tetraethyl orthosilicate (TEOS, g99.0%,
Fluka) with magnetic stirring for 30 min. An aliquot of 100 g of
20 wt % tetrapropylammonium hydroxide solution (TPAOH,
Fluka) was then gradually added to the mixed solution under stirring
for an additional 30 min. After the mixture had been kept at 60 °C
for 3 h, 56 g of deionized water was added to the solution. The
composition of the final solution was the same as the molar reagent
ratios employed by Clerici et al.:20 SiO2/TiO2 ) 32.7, TPAOH/
SiO2 ) 0.46, and H2O/SiO2) 35. The resulting solution was
transferred to an autoclave in an oven and crystallized at 175 °C
for 48 h without stirring. After the solution had been cooled to
room temperature, the crystalline product was separated by
centrifugation. Washing with deionized water and separation were
repeated five times. The solid product was dried at 100 °C for 2 h
and then finally calcined at 550 °C for 5 h in an air-atmosphere
furnace. The resulting catalyst was ground into a powder with a
pestle and mortar.
2.2. Characterization. The crystallographic structure of TS-1
was identified on a PANalytical X’Pert PRO X-ray powder
diffractometer equipped with a nickel filter and Cu KR radiation
8126 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010

(λ ) 1.541 Å). Bragg angles between 5° and 50° were scanned

with a 0.0042° step size. Fourier transform infrared (FT-IR) spectra
were measured on a Perkin-Elmer Spectrum 1 spectrometer using
the KBr wafer technique. UV-visible measurements were per-
formed on a Perkin-Elmer Lambda 25 spectrometer. Scanning
electron microscopy (SEM) images were recorded on a Leo Gemini
1525 instrument to inspect the morphology of the catalyst. The
lattice structure of TS-1 was confirmed by transmission electron
microscopy (TEM) image taken on a JEOL JEM 2010 instrument.
Thermogravimetric analysis (TGA) was performed on a TA Q500
thermogravimetric analyzer by measuring weight changes of
catalyst under nitrogen gas. The temperature was raised from 30
to 1000 °C at a rate of 5 °C min-1. Brunauer-Emmett-Teller
(BET) analysis data was obtained by nitrogen adsorption and
desorption in a Micromeritics TriStar 3000 apparatus.
2.3. Epoxidation of Propene. A 200 mL thick-walled glass
Büchi autoclave system constructed by Strata Technology Ltd. was
used for batch experiments. The catalytic oxidations were carried Figure 1. X-ray diffraction pattern of TS-1 catalyst.
out in a magnetic-drive-stirred autoclave reactor that was immersed
in a water bath to control the reaction temperature. In a typical
run, a mixture consisting of 0.68 g of 35 wt % hydrogen peroxide,
35 g of methanol, and 34.32 g of deionized water was placed in
the autoclave reactor. A small sample was taken to determine the
initial concentration of hydrogen peroxide before the introduction
of the catalyst. Then, 0.28 g of TS-1 catalyst was added, and the
reactor was closed. The rotational speed of the gas entrainment
stirrer was set at 1000 rpm and maintained throughout experiment.
The reactor was heated to a predetermined temperature (40 °C).
When the set temperature was reached, the reactor was charged
with 10% propene in nitrogen at a predetermined total pressure (7
bar) to initiate the reaction. The reaction products were sampled
at regular time intervals through a dip tube. It is noted that a range
of low concentrations of hydrogen peroxide and propene was
employed in this work, whereas most previous studies conducted
in a batch reactor have used pure propene.9,10,20,23,26,49,50
The concentrations of hydrogen peroxide were determined by
iodometric titration. Liquid samples were analyzed on a gas
chromatograph (Shimadzu GC-2014) equipped with a flame Figure 2. FT-IR spectrum of TS-1 catalyst.
ionization detector and a capillary column (30 m × 0.32 mm) with 3. Results and Discussion
nitroterephthalic acid-modified polyethylene glycol (0.25 µm) as
the stationary phase. Propene oxide was the main product; propene 3.1. Catalyst Characterization. The analysis results showed
glycol and propene glycol monomethyl ether were the byproducts. good agreement with the data indicated in the original patent.28
Conversion of hydrogen peroxide, utilization of hydrogen peroxide, Figure 1 shows the XRD pattern of TS-1 catalyst in which single
selectivity to propene oxide, selectivity to byproduct, and yield of peaks at 24.4° and 29.3° demonstrate a conversion from a
propene oxide are defined by the following equations monoclinic symmetry (silicalite) to an orthorhombic symmetry
(titanium silicalite). The peak intensity magnitudes are known
nH0 2O2 - nH2O2 to be related to the crystallinity of TS-1, which strongly affects
XH2O2 ) (1) the catalyst activity, but it was not easy to predict the degree of
nH0 2O2 activity only from the peak intensity magnitudes.51 Figure 2
shows the FT-IR spectrum of TS-1 catalyst. There is an
nPO + nbyproducts absorption band at 971.4 cm-1 that is not present in the spectrum
UH2O2 ) (2) of either pure silicalites or titanium oxides. In general, the
nH0 2O2XH2O2 intensity of this band increases with the amount of titanium that
substitutes the silicon in the framework of silicalites.
nPO Figure 3 shows the UV-vis spectrum of TS-1. An absorption
SPO ) (3) band near 215 nm indicates four-coordinated tetrahedral Ti
nPO + nbyproducts
species in TS-1. No other peaks are observed, which demon-
nbyproducts strates that there is no octahedral Ti or anatase TiO2.9,52
Sbyproducts ) (4) In the SEM images shown in Figure 4, well-defined, uniform
nPO + nbyproducts cubic crystallites with rounding edges are observed to be
distributed homogeneously. It is known that TS-1 crystallizes
YPO ) XH2O2SPO (5) in the form of slightly rounded cubes when crystallization occurs
in a static autoclave and that this shape is more active for the
The initial rates of the chemical reaction were calculated from epoxidation of propene than other shapes such as spheres and
tangential slopes at t ) 0 using exponential curve fitting. twinned coffins.51,53 The size of the particles was in the range

Figure 3. UV-vis spectrum of TS-1 catalyst.
Figure 4. SEM micrographs of TS-1 catalyst.
of 200-300 nm. The activity of TS-1 is also strongly influenced
by the size of the catalyst particles: the smaller the particle size,
the more effective the catalyst.51,53 The TS-1 prepared in this
study therefore had a suitable shape and particle size to exhibit
sufficient activity for the epoxidation of propene largely free
of intraparticle mass-transfer resistance. The TEM micrograph
in Figure 5 exhibits well-ordered lattice fringes of the MFI
structure of TS-1, which is indicative of high crystallinity. This
also confirms that the TS-1 catalyst had uniformly sized
nanopores.
Figure 6 shows the TGA profile of the TS-1 catalyst. The
weight loss up to 100 °C and the total weight loss up to 1000
°C were 1.2% and 2.9%, respectively. The small weight loss
Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010 8127
Figure 5. TEM micrograph of TS-1 catalyst.
Figure 6. TGA profile of TS-1 catalyst.
up to 1000 °C indicates the good thermal stability of TS-1. The
weight loss up to 100 °C should be attributed to the loss of
physisorbed water. The small weight loss up to 100 °C suggests
that the prepared TS-1 catalyst had a relatively high hydrophobic
property, which is known to enhance the activity of TS-1.3,9,10
BET analysis gave a surface area of 421 m2 g-1 and a pore
volume of 0.255 cm3 g-1, which are comparable to the values
reported in the literature.3,7,9,10,54,55
3.2. TS-1 Performance at Normal Operating Conditions.
Prior to the propene epoxidation experiments in the Büchi
autoclave system, the effects of the reactor metal and catalyst
on the decomposition of hydrogen peroxide were examined. The
decomposition of hydrogen peroxide by TS-1 catalyst was
observed to be 3.3% after 5 h at normal operating conditions,
and the effect of the reactor metal was negligible (only 0.3%
after 5 h).
Figure 7 shows the performance of TS-1 at normal operating
conditions in the autoclave reactor. TS-1 catalyst prepared in
this study showed a high catalytic activity comparable to the
results reported in the previous literature.20,49 After 5 h, 92%
of the hydrogen peroxide had been consumed. The selectivity
to propene oxide gradually decreased with time to 93% after
5 h. TS-1 is a weak acid catalyst, which can lead to the solvoysis
of propene oxide. All of the observed byproducts were generated
8128 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010
Figure 7. Propene epoxidation in an autoclave reactor at normal operating
conditions: (a) concentrations of hydrogen peroxide, propene oxide, and
byproduct and (b) conversion of hydrogen peroxide, utilization of hydrogen
peroxide, selectivity to propene oxide, selectivity to byproduct, and yield
of propene oxide. Reaction conditions: 40 °C, 7 bar (10% C3H6/N2), 0.35
wt % H2O2, 50 wt % CH3OH, 0.28 g of TS-1, 1000 rpm stirring.
from propene oxide by the cleavage of the oxirane ring by
nucleophiles.21 In the hydrolysis of propene oxide, propene
glycol is formed from water, and propene glycol monomethyl
ether is formed from reaction with methanol. The utilization of
hydrogen peroxide was kept between 88% and 96% for 5 h.
The main byproducts were 1-methoxy-2-propanol and 2-meth-
oxy-1-propanol, and small amounts of 3-methoxy-1-propanol
and 1,2-propanediol were also detected.
3.3. Mass-Transfer Limitations. The epoxidation of propene
is a heterogeneous gas-liquid-solid-catalyzed reaction. To
investigate the true kinetics, mass-transfer resistances between
the phases must be removed. In this study, a self-aspirating
stirrer was used to intensify the contact between the phases by
recycling gas inside the autoclave. The mass-transfer resistances
were controlled by the level of turbulence induced by stirring.
A Rushton turbine type of impeller with six vertical blades was
employed to generate large vortices. The Reynolds number (Re)
was 33000 at an agitation speed of 1000 rpm. Therefore, the
system was operated in a completely turbulent flow regime (Re
> 1000056). The system was maintained in turbulent flow even
at 500 rpm (Re ) 16500) but fell into laminar flow below 300
rpm.
The effect of the agitation speed of the stirrer on the initial
rate of reaction was examined in the range of 100-1500 rpm.
Figure 8. Effect of agitation speed on the initial rate of propene epoxidation.
Reaction conditions: 40 °C, 7 bar (10% C3H6/N2), 0.35 wt % H2O2, 50 wt
% CH3OH, 0.28 g of TS-1, 100-1500 rpm stirring.
As can be seen in Figure 8, the rate of reaction was independent
of agitation speed when the agitation speed was greater than
500 rpm. This is clear evidence for the absence of gas-liquid
mass-transfer resistance at and above a stirring speed of 500
rpm. Thus, 1000 rpm was employed for the speed of agitation
in all further experiments.
The Mears criterion allows for the estimation of possible
mass-transfer limitations from the bulk fluid to the surface of
the catalyst.57 The Mears criterion calculated for the reactor
system was less than 0.15, which shows that external diffusion
resistance can be neglected.
To check the possibility of mass-transfer limitations, the
effectiveness factor was calculated from the Thiele modulus and
the effective diffusivity. The effectiveness factor was very close
to 1, which demonstrates that internal mass diffusion inside the
catalyst particle had negligible resistance.57 These findings
confirm that the chemical reaction on the surface of the catalyst
was rate limiting. The equations and calculated parameter values
used to check for mass-transfer limitations are shown in the
Appendix.
3.4. Catalyst Loading. The effects of catalyst loading on
the initial rate and selectivity of propene epoxidation were
studied in a range of catalyst concentrations (Figure 9). The
initial rate was found to be proportional to the catalyst loading,
and this linearity reconfirms the absence of mass-transfer
resistance. In Figure 9a, it seems that the data point at the largest
reciprocal of catalyst loading has a significant upward deviation
from a regression line. This is because, at a low level of catalyst
loading, the measured rate of reaction in a batch reactor can be
affected easily by several factors such as slight catalyst loss
from splashing.
The initial rate increased linearly with the amount of catalyst,
but the selectivity to propene oxide decreased with an increase
in the catalyst loading, particularly at higher catalyst loadings
(Figure 9b). The selectivity to propene oxide after 5 h was as
low as 86% when 0.5 g of TS-1 was used.
3.5. Initial Concentration of Hydrogen Peroxide. The
effect of the concentration of hydrogen peroxide on the initial
rate is shown in Figure 10. The results indicate that the order
of reaction is 0.67 with respect to hydrogen peroxide. This
finding is consistent with that of Clerici et al., who reported
that the reaction order with respect to hydrogen peroxide is
between zero and unity.20 Also, a similar fractional order can
be indirectly confirmed when a power-law model is applied to

Figure 9. Effect of catalyst loading on (a) initial rate of propene epoxidation
and (b) selectivity to propene oxide. Reaction conditions: 40 °C, 7 bar (10%
C3H6/N2), 0.35 wt % H2O2, 50 wt % CH3OH, 0.14-0.50 g of TS-1, 1000
rpm stirring.
Figure 10. Effect of initial hydrogen peroxide concentration on the initial
rate of propene epoxidation. Reaction conditions: 40 °C, 7 bar (10% C3H6/
N2), 0.17-0.70 wt % H2O2, 50 wt % CH3OH, 0.28 g of TS-1.
the results presented by Liang and colleagues.49 However, Thiele
et al. reported that hydrogen peroxide was consumed according
to pseudo-first-order kinetics.21
Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010 8129
Figure 11. Effect of initial propene partial pressure on the initial rate of
propene epoxidation. Reaction conditions: 40 °C, 3-8 bar (10% C3H6/N2),
0.35 wt % H2O2, 50 wt % CH3OH, 0.28 g of TS-1.
As the initial concentration of hydrogen peroxide increased,
the selectivity to propene oxide decreased. The selectivity to
propene oxide after 5 h was 84% when the initial concentration
of H2O2 was 0.2 mol kg-1, which is twice the hydrogen peroxide
concentration at normal operating conditions. These results
suggest that the initial concentration of hydrogen peroxide
should be adjusted according to the propene feed concentration
in order to minimize side reactions of propene oxide in
commercial applications.
3.6. Initial Propene Partial Pressure. The results obtained
at various initial propene partial pressures (0.3-0.8 bar) are
presented in Figure 11, which indicates that the order with
respect to propene was 0.63. This result differs from those
reported in some published studies. Almost first order can be
obtained with respect to propene when a power-law model is
applied to the results presented by Liang et al.,49 and Lee et
al.58 reported that propene follows pseudo-first-order kinetics
over methyltrioxorhenium catalyst. This discrepancy might be
due to different solvents and higher concentrations of reactants
employed in the previous studies. The effect of the solvent is
discussed in the next section.
The production rate of propene oxide increased with increas-
ing propene partial pressure, as expected, but the selectivity to
propene oxide was highest when the propene partial pressure
was 0.7 bar. A possible explanation for this is that a propene
partial pressure of 0.7 bar provides the stoichiometric amount
of propene for the initial concentration of hydrogen peroxide,
which is favorable for maintaining high selectivity.
3.7. Concentration of Solvent. Good catalytic performance
can be achieved in solvents that are able to dissolve both propene
and hydrogen peroxide. It is known that methanol is the best
among many kinds of protic or aprotic solvents.20,21,25,26,50
Therefore, methanol/water mixtures were used as the solvent
in this study.
The variations in the propene oxide concentration and
selectivity with varying weight percentage of methanol are
presented in Figure 12. As shown in Figure 12b, there was only
a slight difference in the selectivity to propene oxide when the
water content was below 50%, but the selectivity decreased
dramatically as the water content increased beyond this con-
centration. In particular, the formation of 1,2-propanediol was
greatly promoted by water. This result is consistent with those
of previous researchers.20,25,48
8130 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010
Figure 12. Effect of methanol concentration on (a) propene oxide production
and (b) selectivity to propene oxide. Reaction conditions: 40 °C, 7 bar (10%
C3H6/N2), 0.35 wt % H2O2, 0-70 wt % CH3OH, 0.28 g of TS-1.
It is known that the reaction rate of propene epoxidation
generally increases with increasing methanol content because
the solubility of propene in methanol is higher than that in
water.3,20,25,48 Contrary to expectations, in the present study,
the initial rate of propene oxide production was found to
decrease with increasing amount of methanol with a high degree
of linearity, as shown in Figure 12a. Interestingly, the initial
rate showed a maximum at low methanol concentration and was
fairly high even when there was no methanol (Figure 12a inset).
However, in water, the productivity to propene oxide for 5 h
was very low when compared with that in methanol/water
mixtures. The final concentration of propene oxide after 5 h
was highest at 50% methanol content because side reactions
were promoted at less than 50% methanol and the reaction rate
was low at greater than 50% methanol.
These results showing a decreasing rate with increasing
methanol content are quite different from previously reported
work on the influence of methanol and indicate that methanol
can act as an inhibitor through competitive adsorption. The
observation of a slight maximum at low methanol content also
points to the more conventional effect of enhancing the solubility
of propene. It seems likely that this inhibiting effect of methanol
was magnified by the mild conditions used in the present study
and masked in earlier works that used higher concentrations of
hydrogen peroxide and higher propene pressures.
Table 1. Henry’s Law Constants of Propene in Water and Methanol
at 40 °C59-61
solvent HE (atm) HE [mol/(L · atm)] HE,MeOH/HE,H2O
-3
H2O 6667 8.3 × 10 31.3
MeOH 90.5 2.6 × 10-1
Table 2. Additional Measurement of Initial Rates
H2O2 (wt %) PC3H6 (bar) CH3OH (wt %) r0 (mol L-1 gcat-1 min-1)
0.35 2.8 50 0.0165
1.05 0.7 50 0.0103
1.05 2.8 10 0.0167
Previous kinetic studies have sought to fit the rate of this
reaction to Eley-Rideal or Langmuir-Hinshelwood model
equations. For example, Liang et al.49 concluded that an
Eley-Rideal model with adsorption of hydrogen peroxide and
free propene was the best fit to their experimental data. The
proposed rate equation also allowed for competitive adsorption
by propene and the product propene oxide.49 To gain better
insight into the effects of the reactants, product, and methanol
concentration on the initial reaction rate, Eley-Rideal and
Langmuir-Hinshelwood model equations incorporating a term
for competitive adsorption of methanol and allowance for the
enhancement of propene solubility were used to fit to the present
experimental data. For example, a Langmuir-Hinshelwood
(L-H) model can be expressed by eq 6, where HE,H2O and
HE,MeOH are Henry’s law constants for water and methanol,
respectively. The denominator of eq 6 can be modified in the
case of a dual-site L-H model to give eq 7.
r0 )
kK1K2CH2O2(φH2OHE,H2O + φMeOHHE,MeOH)PC3H6
(6)
[1 + K1CH2O2 + K2(φH2OHE,H2O + φMeOHHE,MeOH)PC3H6 +
K3CMeOH]2
kK1K2CH2O2(φH2OHE,H2O + φMeOHHE,MeOH)PC3H6
r0 ) (7)
(1 + K1CH2O2 + K3CMeOH)[1 + K2(φH2OHE,H2O +
φMeOHHE,MeOH)PC3H6 + K4CMeOH]
An Eley-Rideal model with free propene can be represented
by eq 8, where allowance is also made for competitive
adsorption of propene following Liang et al.49 A similar equation
can be derived for the case where hydrogen peroxide is free.
r0 )
kK1CH2O2(φH2OHE,H2O + φMeOHHE,MeOH)PC3H6
1 + K1CH2O2 + K2(φH2OHE,H2O + φMeOHHE,MeOH)PC3H6 + K3CMeOH
(8)
Table 1 summarizes the Henry’s law constants (HE) of
propene with respect to water and methanol that were used to
consider the relative amounts of dissolved propene in the
different solvent mixtures.59-61
To augment the analysis, three additional measurements of
initial rates were made (Table 2). Eley-Rideal and L-H model
equations were evaluated, and statistical analysis was carried
out using all of the measured initial rates and a nonlinear least-
squares function of software R. Converged solutions were not
found for the Eley-Rideal models with competitive adsorption
of methanol as in eq 8. The nonlinear regression analysis showed
that the initial rate data could be fitted best by a form of L-H
model equation (Figure 13). There is good agreement between
 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010 8131

Figure 13. Experimental initial rates and fitting by the Langmuir-Hinshelwood models: (a) parity plot and (b) comparison with respect to methanol content.

Table 3. Kinetic Parameters and Statistical Results

t test (p value)
model kK1K2 K1 K2 K3 K4 F test (p value) RSS σ2
(I) L-H single-site 3.70 × 10-3 8.35 × 10-4 4.96 × 10-2 2.02 × 10-5 - 6.14 × 10-9 7.38 × 10-6 4.92 × 10-7
(II) L-H dual-site 1.53 × 10-5 2.83 × 10-3 1.90 × 10-3 1.11 × 10-2 1.18 × 10-5 3.11 × 10-10 2.21 × 10-6 1.47 × 10-7

Table 4. Parameters with 95% Confidence Interval for L-H Single- Table 5. Comparison of Conditions Used for Investigations into the
and Dual-Site Models at 40 °C Effect of Methanol Content
L-H single-site L-H dual-site ref(s) CH2O2 (mol/L) PC3H6 (atm) catalyst (g/L) temp (°C)
parameter units model model
this study 0.1 0.7 4 40
k mol L-1 gcat-1 min-1 1.66 ( 0.98 0.20 ( 0.056 Wang et al.,48 0.45 4 12 60
K1 L mol-1 9.74 ( 4.63 9.66 ( 5.31 Li et al.25
K2 L mol-1 4.15 ( 4.07 20.57 ( 10.65 Chen et al.3 1 1 9 25
K3 L mol-1 0.57 ( 0.17 0.12 ( 0.08 Clerici et al.20 0.5-1.75 4-8 4-8 20-60
K4 L mol-1 - 0.96 ( 0.26
propene and hydrogen peroxide concentrations spanning the
the experimental and predicted initial rates. The kinetic param- present and previously reported conditions, using eq 6 and the
eters and the results of t and F tests are given in Table 3. parameters of Table 4 (similar set of curves can be generated
According to the F test, the L-H dual-site model (model II) from eq 7). From this simulation, it can be seen that the
is statistically more significant than the single-site model (model dependence of the initial rates of propene epoxidation on the
I). However, model I also gives a small residual sum of squares methanol content can follow various trends, either increasing
(RSS) and variance (σ2), which leads to good fitting to the initial or decreasing, according to the levels of reactant concentrations.
rate data. Thus, it can be concluded that methanol competes This means that the increases in reaction rate with increasing
with propene and hydrogen peroxide for adsorption on single amount of methanol observed in the previous studies were a
sites or on dual sites of the catalyst surface. Based solely on consequence of the higher reactant concentrations used, which
the present results, it is not possible to differentiate unambigu- overcame the effect of competitive adsorption by methanol.
ously between the two L-H models. The confidence intervals
for model I and model II are presented in Table 4.
It is known that the high activity of TS-1 in propene
epoxidation is attributable in part to its hydrophobicity as it
preferentially adsorbs the less polar molecules propene and
H2O2. Methanol would be expected to be preferentially absorbed
on the active sites of TS-1 from water. The values of the
adsorption constants in Table 4 are in the order propene ≈ H2O2
> methanol, which is consistent with the polarities of these
species. Although the adsorption constant of methanol is
relatively small, the methanol concentration is large, giving rise
to a significant effect on the reaction rate. The TS-1 catalyst
prepared in this study contained only tetrahedral Ti and no
anatase TiO2 phase or extraframework Ti. On balance, the dual-
site L-H model (II) would seem to be more consistent with
the known structure of TS-1.
In Table 5, the operating conditions used in this work are
compared with those of previous studies that investigated the
effect of methanol content. Presented in Figure 14 are initial Figure 14. Initial rates of propene epoxidation at various reactant and methanol
rates with respect to methanol content calculated at various concentrations predicted by the Langmuir-Hinshelwood model (eq 6).
8132 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010
Figure 15. Effect of reaction temperature on (a) propene oxide production
and (b) selectivity to propene oxide. Reaction conditions: 21-50 °C, 7 bar
(10% C3H6/N2), 0.35 wt % H2O2, 50 wt % CH3OH, 0.28 g of TS-1.
3.8. Reaction Temperature. Figure 15 shows the effect of
the reaction temperature on propene epoxidation. The initial rate
increased with an increase in temperature, but the selectivity to
propene oxide decreased, which led to low production rate of
propene oxide as the reaction proceeded. It is noteworthy that
there was a significant difference in the selectivity between 40
and 50 °C. In particular, a large amount of 1,2-propanediol was
produced at 50 °C.
The Arrhenius plot shown in Figure 16 gives an activation
energy 25.8 kJ mol-1 for propene epoxidation over TS-1
catalyst. This apparent activation energy is close to that reported
over TS-1 in isopropanol by Liang et al.49
4. Conclusions
TS-1 catalyst prepared according to the original patent was
characterized by several analysis techniques and showed good
performance for the selective oxidation of propene to propene
oxide in an autoclave reactor.
The change in operating conditions had a profound effect on
the propene oxide concentration and selectivity. The selectivity
to propene oxide decreased with increasing catalyst loading,
reactant concentration, water content, and reaction temperature,
which, in turn, led to a decrease of the propene oxide
concentration. The apparent orders of reaction with respect to
hydrogen peroxide and propene were found to be 0.67 and 0.63,
respectively. The apparent activation energy was determined
Figure 16. Arrhenius plot of propene epoxidation. Reaction conditions:
21-50 °C, 7 bar (10% C3H6/N2), 0.35 wt % H2O2, 50 wt % CH3OH, 0.28 g
of TS-1.
to be 25.8 kJ mol-1. The concentration of solvent including
methanol and water had a tradeoff effect on the concentration
of propene oxide because the relative amounts of methanol and
water affected the reaction rate and selectivity to propene oxide
differently, that is, the reaction rate decreased but the selectivity
increased with increasing methanol content.
The initial rates of propene oxide production were observed
to decrease with increasing methanol content from a maximum
at low methanol concentrations. This effect was shown to be
due to the enhancement of propene solubility by methanol and
the opposing effect of competitive adsorption by methanol.
Langmuir-Hinshelwood single-site and dual-site rate equations
were found to give a good fit to the initial rate data.
Acknowledgment
Financial support by EPSRC (U.K.) is gratefully acknowledged.
The authors also thank Haseong Kim at Imperial College London
for the help with and discussion of the statistical analysis.
Nomenclature
Ci ) concentration of species i (mol kg-1 or mol L-1)
HE,i ) Henry’s law constant of propene for solvent i (mol L-1
atm-1)
k ) reaction rate constant (mol L-1 gcat-1 min-1)
Ki ) adsorption equilibrium constant of species i (L mol-1)
ni ) number of moles of species i (mol)
ni0 ) initial number of moles of species i (mol)
Pi ) pressure of species i
r0 ) initial reaction rate (mol kg-1 h-1 or mol L-1 gcat-1 min-1)
Si ) selectivity to product i
Ui ) utilization of species i
Xi ) conversion of species i
Yi ) yield of species i
Greek Letter
φi ) volume fraction of solvent i in a solvent mixture
Appendix
A1. Liquid-Solid Mass-Transfer Limitations
The Wilke-Chang equation62 was used to calculate the diffusion
coefficient of propene in a mixture of water and methanol
 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010 8133
-8
7.4 × 10 (φMB) T 1/2 (2) Cheng, W. G.; Wang, X. S.; Li, G.; Guo, X. W.; Zhang, S. J. Highly
DAB ) Efficient Epoxidation of Propylene to Propylene Oxide over TS-1 Using
ηBVA 0.6 Urea + Hydrogen Peroxide as Oxidizing Agent. J. Catal. 2008, 255 (2),
343–346.
The dimensionless numbers, such as the Schmidt number, (3) Chen, L. Y.; Chuah, G. K.; Jaenicke, S. Propylene Epoxidation with
Reynolds number, and Sherwood number, and the mass-transfer Hydrogen Peroxide Catalyzed by Molecular Sieves Containing Framework
Titanium. J. Mol. Catal. A: Chem. 1998, 132 (2-3), 281–292.
coefficient were calculated according to the following equations57
(4) Laufer, W.; Meiers, R.; Holderich, W. Propylene Epoxidation with
ν Hydrogen Peroxide over Palladium Containing Titanium Silicalite. J. Mol.
Sc ) Catal. A: Chem. 1999, 141 (1-3), 215–221.
DAB (5) Beckman, E. J. Production of H2O2 in CO2 and Its Use in the Direct
Synthesis of Propylene Oxide. Green Chem. 2003, 5 (3), 332–336.
Nd2
Re ) (6) Liu, X. W.; Wang, X. S.; Guo, X. W.; Li, G.; Yan, H. S. Regeneration
ν of Lamina TS-1 Catalyst in the Epoxidation of Propylene with Hydrogen
Peroxide. Catal. Lett. 2004, 97 (3-4), 223–229.
Sh ) 2 + 0.6Re1/2Sc1/3 (7) Wang, Q. F.; Wang, L.; Chen, J. X.; Wu, W.; Mi, Z. T. Deactivation
kcdp and Regeneration of Titanium Silicalite Catalyst for Epoxidation of
Sh ) Propylene. J. Mol. Catal. A: Chem. 2007, 273 (1-2), 73–80.
DAB (8) Zhao, J. L.; Zhou, J. C.; Su, J.; Guo, H. C.; Wang, X. S.; Gong,
W. M. Propene Epoxidation with In-site H2O2 Produced by H2/O2 Non-
A Mears criterion of less than 0.15 indicates that there are equilibrium Plasma. AIChE J. 2007, 53, 3204–3209.
no liquid-solid mass-transfer limitations57 (9) Park, S.; Cho, K. M.; Youn, M. H.; Seo, J. G.; Baeck, S. H.; Kim,
T. J.; Chung, Y. M.; Oh, S. H.; Song, I. K. Epoxidation of Propylene with
-rA′ FpRn Hydrogen Peroxide over TS-1 Catalyst Synthesized in the Presence of
Cm ) Polystyrene. Catal. Lett. 2008, 122 (3-4), 349–353.
kcCA (10) Park, S.; Cho, K. M.; Youn, M. H.; Seo, J. G.; Jung, J. C.; Baeck,
S. H.; Kim, T. J.; Chung, Y. M.; Oh, S. H.; Song, I. K. Direct Epoxidation
The values of these parameters are collected in Table A1. of Propylene with Hydrogen Peroxide over TS-1 Catalysts: Effect of
Table A1. Calculation Results for Liquid-Solid Mass-Transfer Hydrophobicity of the Catalysts. Catal. Commun. 2008, 9 (15), 2485–2488.
Limitations (11) Short, P. L. BASF, Dow Open Novel Propylene Oxide Plant. Chem.
Eng. News 2009, 87 (11), 21.
parameter description units value (12) Klemm, E.; Dietzsch, E.; Schwarz, T.; Kruppa, T.; de Oliveira,
-1 A. L.; Becker, F.; Markowz, G.; Schirrmeister, S.; Schutte, R.; Caspary,
DAB diffusion coefficient cm s 2
2.5 × 10-5 K. J.; Schuth, F.; Honicke, D. Direct Gas-phase Epoxidation of Propene
Sc Schmidt number - 248 with Hydrogen Peroxide on TS-1 Zeolite in a Microstructured Reactor. Ind.
Re Reynolds number - 33029 Eng. Chem. Res. 2008, 47 (6), 2086–2090.
Sh Sherwood number - 687 (13) Oyama, S. T.; Zhang, X. M.; Lu, J. Q.; Gu, Y. F.; Fujitani, T.
kc mass-transfer coefficient cm s-1 685.9 Epoxidation of Propylene with H2 and O2 in the Explosive Regime in a
-rA′ rate of reaction kmol kgcat-1 s-1 6.684 × 10-5 Packed-bed Catalytic Membrane Reactor. J. Catal. 2008, 257 (1), 1–4.
Cm Mears criterion - 5.71 × 10-12 (14) Nijhuis, T. A.; Visser, T.; Weckhuysen, B. M. Mechanistic Study
A2. Internal Mass-Transfer Limitations into the Direct Epoxidation of Propene over Gold/Titania Catalysts. J. Phys.
Chem. B 2005, 109 (41), 19309–19319.
To calculate the effectiveness factor, the effective diffusivity (15) Zwijnenburg, A.; Makkee, M.; Moulijn, J. A. Increasing the Low
Propene Epoxidation Product Yield of Gold/titania-based Catalysts. Appl.
and Thiele modulus were calculated using the following Catal. A: Gen. 2004, 270 (1-2), 49–56.
equations,57 with approximate values for the porosity, constric- (16) Sinha, A. K.; Seelan, S.; Tsubota, S.; Haruta, M. Catalysis by Gold
tion factor, and tortuosity Nanoparticles: Epoxidation of Propene. Top. Catal. 2004, 29 (3-4), 95–
102.
DABεpσ (17) Suo, Z. H.; Jin, M. S.; Lu, J. Q.; Wei, Z. B.; Li, C. Direct Gas-
De ) Phase Epoxidation of Propylene to Propylene Oxide Using Air as Oxidant
τ
on Supported Gold Catalyst. J. Nat. Gas Chem. 2008, 17 (2), 184–190.


k1FcSa (18) Bravo-Suarez, J. J.; Lu, J.; Dallos, C. G.; Fujitani, T.; Oyama, S. T.
φ1 ) R Kinetic Study of Propylene Epoxidation with H2 and O2 over a Gold/
De Mesoporous Titanosilicate Catalyst. J. Phys. Chem. C 2007, 111, 17427–
17436.
3
η) (φ1 coth φ1 - 1) (19) Yap, N.; Andres, R. P.; Delgass, W. N. Reactivity and Stability of
φ12 Au in and on TS-1 for Epoxidation of Propylene with H2 and O2. J. Catal.
2004, 226 (1), 156–170.
The values of these parameters are collected in Table A2. (20) Clerici, M. G.; Bellussi, G.; Romano, U. Synthesis of Propylene
Oxide from Propylene and Hydrogen Peroxide Catalyzed by Titanium
Table A2. Calculation Results for Internal Mass-Transfer Silicalite. J. Catal. 1991, 129 (1), 159–167.
Limitations (21) Thiele, G. F.; Roland, E. Propylene Epoxidation with Hydrogen
parameter description units value Peroxide and Titanium Silicalite Catalyst: Activity, Deactivation and
Regeneration of the Catalyst. J. Mol. Catal. A: Chem. 1997, 117 (1-3),
εp catalyst porosity - 0.4 351–356.
σ constriction factor - 0.8 (22) Mantegazza, M. A.; Petrini, G.; Spanò, G.; Bagatin, R.; Rivetti, F.
τ tortuosity - 3 Selective Oxidations with Hydrogen Peroxide and Titanium Silicalite
De effective diffusivity m2 s-1 2.66 × 10-10 Catalyst. J. Mol. Catal. A: Chem. 1999, 146, 223–228.
k1 rate constant m3 m-2 s-1 2.11 × 10-10 (23) Guo, X. W.; Wang, X. S.; Liu, M.; Li, G.; Chen, Y. Y.; Xiu, J. H.;
Fc density of catalyst kg m-3 2370 Zhuang, J. Q.; Zhang, W. P.; Bao, X. H. Epoxidation of Propylene with
Sa surface area of catalyst m2 kg-1 421000 Dilute H2O2 over Titanium Silicalite Containing Trace Aluminum. Catal.
φ1 Thiele modulus - 0.003488
Lett. 2002, 81 (1-2), 125–130.
η internal effectiveness factor - 0.9999
(24) Li, G.; Wang, X. S.; Yan, H. S.; Liu, Y. H.; Liu, X. W. Epoxidation
Literature Cited of Propylene Using Supported Titanium Silicalite Catalysts. Appl. Catal.
A: Gen. 2002, 236 (1-2), 1–7.
(1) Nijhuis, T. A.; Makkee, M.; Moulijn, J. A.; Weckhuysen, B. M. (25) Li, G.; Meng, J. W.; Wang, X. S.; Guo, X. W. Effect of Solvents
The Production of Propene Oxide: Catalytic Processes and Recent Develop- on Propene Epoxidation Catalyzed by Titanium Silicalite. React. Kinet.
ments. Ind. Eng. Chem. Res. 2006, 45 (10), 3447–3459. Catal. Lett. 2004, 82 (1), 73–80.
8134 Ind. Eng. Chem. Res., Vol. 49, No. 17, 2010
(26) Liu, X. W.; Wang, X. S.; Guo, X. W.; Li, G. Effect of Solvent on
the Propylene Epoxidation over TS-1 Catalyst. Catal. Today 2004, 93-95,
505–509.
(27) Clerici, M. G. TS-1 and Propylene Oxide, 20 Years Later. Oil Gas
Eur. Mag. 2006, 32 (2), 77–82.
(28) Taramasso, M.; Perego, G.; Notari, B. Preparation of porous
crystalline synthetic material comprised of silicon and titanium oxides. U.S.
Patent 4,410,501, 1983.
(29) Kraushaar, B.; van Hooff, J. H. C. A New Method for the
Preparation of Titanium-Silicalite (TS-1). Catal. Lett. 1988, 1 (4), 81–84.
(30) Gao, H. X.; Suo, J. S.; Li, S. B. An Easy Way to Prepare Titanium
Silicalite-1 (TS-1). J. Chem. Soc., Chem. Commun. 1995, (8), 835–835.
(31) Serrano, D. P.; Uguina, M. A.; Ovejero, G.; Vangrieken, R.;
Camacho, M. Synthesis of TS-1 by Wetness Impregnation of Amorphous
SiO2-TiO2 Solids Prepared by the Sol-Gel Method. Microporous Mater.
1995, 4 (4), 273–282.
(32) Uguina, M. A.; Serrano, D. P.; Ovejero, G.; Vangrieken, R.;
Camacho, M. Preparation of TS-1 by Wetness Impregnation of Amorphous
SiO2-TiO2 Solids: Influence of the Synthesis Variables. Appl. Catal. A:
Gen. 1995, 124 (2), 391–408.
(33) Gontier, S.; Tuel, A. Synthesis of Titanium Silicalite-1 Using
Amorphous SiO2 as Silicon Source. Zeolites 1996, 16 (2-3), 184–195.
(34) Li, G.; Guo, X. W.; Wang, X. S.; Zhao, Q.; Bao, X. H.; Han, X. W.;
Lin, L. W. Synthesis of Titanium Silicalites in Different Template Systems
and Their Catalytic Performance. Appl. Catal. A: Gen. 1999, 185 (1), 11–
18.
(35) Kim, K. Y.; Ahn, W. S.; Park, D. W.; Oh, J. H.; Lee, C. M.; Tai,
W. P. Microwave Synthesis of Titanium Silicalite-1 Using Solid Phase
Precursors. Bull. Korean Chem. Soc. 2004, 25 (5), 634–638.
(36) Ahn, W. S.; Lee, K. Y. Extensions in the Synthesis and Catalytic
Application of Titanium Silicalite-1. Catal. SurV. Asia 2005, 9 (1), 51–60.
(37) Chen, P.; Chen, X. S.; Tanaka, K.; Kita, H. A Novel and Less-
Expensive Preparation of Titanium Silicalite-1 Membrane. Chem. Lett. 2007,
36 (8), 1078–1079.
(38) Meiers, R.; Dingerdissen, U.; Holderich, W. F. Synthesis of
Propylene Oxide from Propylene, Oxygen, and Hydrogen Catalyzed by
Palladium-Platinum-Containing Titanium Silicalite. J. Catal. 1998, 176
(2), 376–386.
(39) Meiers, R.; Holderich, W. F. Epoxidation of Propylene and Direct
Synthesis of Hydrogen Peroxide by Hydrogen and Oxygen. Catal. Lett.
1999, 59 (2-4), 161–163.
(40) Hoelderich, W. F. ’One-Pot’ Reactions: A Contribution to Envi-
ronmental Protection. Appl. Catal. A: Gen. 2000, 194, 487–496.
(41) Hoelderich, W. F.; Kollmer, F. Oxidation Reactions in the Synthesis
of Fine and Intermediate Chemicals Using Environmentally Benign Oxidants
and the Right Reactor System. Pure Appl. Chem. 2000, 72 (7), 1273–1288.
(42) Grey, R. A.; Rangasamy, P. Aqueous epoxidation process using
modified titanium zeolite. U.S. Patent 6,194,591, 2001.
(43) Jenzer, G.; Mallat, T.; Maciejewski, M.; Eigenmann, F.; Baiker,
A. Continuous Epoxidation of Propylene with Oxygen and Hydrogen on a
Pd-Pt/TS-1 Catalyst. Appl. Catal. A: Gen. 2001, 208 (1-2), 125–133.
(44) Laufer, W.; Hoelderich, W. F. Direct Oxidation of Propylene and
Other Olefins on Precious Metal Containing Ti-catalysts. Appl. Catal. A:
Gen. 2001, 213 (2), 163–171.
(45) Danciu, T.; Beckman, E. J.; Hancu, D.; Cochran, R.; Grey, R.;
Hajnik, D.; Jewson, J. Direct Synthesis of Propylene Oxide with CO2 as
the Solvent. Angew. Chem., Int. Ed. 2003, 42, 1140–1142.
(46) Wang, Q. F.; Wang, L.; Mi, Z. T. Influence of Pt-Pd/TS-1 Catalyst
Preparation on Epoxidation of Olefins with Hydrogen Peroxide. Catal. Lett.
2005, 103 (1-2), 161–164.
(47) Chen, Q. L.; Beckman, E. J. One-pot Green Synthesis of Propylene
Oxide Using in Situ Generated Hydrogen Peroxide in Carbon Dioxide.
Green Chem. 2008, 10 (9), 934–938.
(48) Wang, X. S.; Guo, X. W.; Li, G. Synthesis of Titanium Silicalite
(TS-1) from the TPABr System and Its Catalytic Properties for Epoxidation
of Propylene. Catal. Today 2002, 74 (1-2), 65–75.
(49) Liang, X. H.; Mi, Z. T.; Wu, Y. L.; Wang, L.; Xing, E. H. Kinetics
of Epoxidation of Propylene over TS-1 in Isopropanol. React. Kinet. Catal.
Lett. 2003, 80 (2), 207–215.
(50) Zhang, Z. G.; Kang, J. N.; Wang, Y. Q. Effects of Organic Solvent
Addition on the Epoxidation of Propene Catalyzed by TS-1. React. Kinet.
Catal. Lett. 2007, 92, 49–54.
(51) van der Pol, A. J. H. P.; van Hooff, J. H. C. Parameters Affecting
the Synthesis of Titanium Silicalite-1. Appl. Catal. A: Gen. 1992, 92 (2),
93–111.
(52) Fan, W. B.; Duan, R. G.; Yokoi, T.; Wu, P.; Kubota, Y.; Tatsumi,
T. Synthesis, Crystallization Mechanism, and Catalytic Properties of
Titanium-Rich TS-1 Free of Extraframework Titanium Species. J. Am.
Chem. Soc. 2008, 130 (31), 10150–10164.
(53) van der Pol, A. J. H. P.; Verduyn, A. J.; van Hooff, J. H. C. Why
Are Some Titanium Silicalite-1 Samples Active and Others Not. Appl. Catal.
A: Gen. 1992, 92 (2), 113–130.
(54) Keshavaraja, A.; Ramaswamy, V.; Soni, H. S.; Ramaswamy, A. V.;
Ratnasamy, P. Synthesis, Characterization, and Catalytic Properties of
Micro-Mesoporous, Amorphous Titanosilicate Catalysts. J. Catal. 1995,
157 (2), 501–511.
(55) Li, Y. G.; Lee, Y. M.; Porter, J. F. The Synthesis and Characteriza-
tion of Titanium Silicalite-1. J. Mater. Sci. 2002, 37 (10), 1959–1965.
(56) Sinnott, R. K. Coulson & Richardson’s Chemical Enginering, Vol.
6: Chemical Engineering Design; Butterworth-Heinemann: Oxford, U.K.,
2005.
(57) Fogler, H. S. Elements of Chemical Reaction Engineering; Prentice-
Hall: Upper Saddle River, NJ, 1992.
(58) Lee, H.-J.; Shi, T.-P.; Busch, D. H.; Subramaniam, B. A Greener,
Pressure Intensified Propylene Epoxidation Process with Facile Product
Separation. Chem. Eng. Sci. 2007, 62 (24), 7282–7289.
(59) Yaws, C. L.; Yang, H.-C. Thermodynamic and Physical Property
Data; Gulf Publishing Company: Houston, TX, 1992.
(60) Miyano, Y.; Fukuchi, K. Henry’s Constants of Propane, Propene,
Trans-2-butene and 1,3-Butadiene in Methanol at 255-320 K. Fluid Phase
Equilib. 2004, 226, 183–187.
(61) Perry, R. H.; Green, D. W. Perry’s Chemical Engineers’ Handbook;
McGraw-Hill: New York, 2007.
(62) Reid, R. C.; Prausnitz, J. M.; Poling; B. E. The Properties of Gases
& Liquids; McGraw-Hill: New York, 1998.
ReceiVed for reView January 13, 2010
ReVised manuscript receiVed May 28, 2010
Accepted July 1, 2010
IE100083U