Synthesis of propylene glycol monomethyl ether over Mg/Al hydrotalcite catalyst

Article abstract of DOI:10.1007/s10562-010-0335-y

Mg-Al hydrotalcites with different Mg/Al molar ratios were prepared and characterized by XRD, FT-IR, SEM and BET analyses. The calcined hydrotalcite with Mg/Al molar ratio of 4.0 (LDO Mg/Al 4.0) exhibited the highest catalytic activity in the synthesis of propylene glycol methyl ether (PM). The catalytic activity relating to the amount of the basic sites and crystallinity depended on the Mg/Al molar ratio. The optimal equilibrium of acid-base property and high crystallinity made the LDO Mg/Al 4.0 an excellent catalyst in the reaction. Etherification of propylene oxide (PO) with methanol over the LDO Mg/Al 4.0 was researched. The optimized reaction conditions were as follows: 140 °C, catalyst amount 0.9 wt%, methanol/PO molar ratio 4.0 and 6 h. The PO conversion and PM selectivity were 93.2 and 97.4%, respectively. Above all, almost all the PM was 1-methoxy-2-propanol, for no 2-methoxy-1-propanol was detected by GC analysis in the reaction products, and the catalyst could be reused for five times.

Full text of DOI:10.1007/s10562-010-0335-y

Catal Lett (2010) 137:94–103
DOI 10.1007/s10562-010-0335-y
Synthesis of Propylene Glycol Monomethyl Ether Over
Mg/Al Hydrotalcite Catalyst
•
Hong-yan Zeng Ya-ju Wang Zhen Feng
•
•
•
Kui-yi You Ce Zhao Jin-wei Sun Ping-le Liu
•
•
Received: 18 November 2009 / Accepted: 29 March 2010 / Published online: 10 April 2010
Ó Springer Science+Business Media, LLC 2010
Abstract Mg–Al hydrotalcites with different Mg/Al
molar ratios were prepared and characterized by XRD, FT-
IR, SEM and BET analyses. The calcined hydrotalcite with
Mg/Al molar ratio of 4.0 (LDO Mg/Al 4.0) exhibited the
highest catalytic activity in the synthesis of propylene
glycol methyl ether (PM). The catalytic activity relating to
the amount of the basic sites and crystallinity depended on
the Mg/Al molar ratio. The optimal equilibrium of acid–
base property and high crystallinity made the LDO Mg/Al
4.0 an excellent catalyst in the reaction. Etherification of
propylene oxide (PO) with methanol over the LDO Mg/Al
4.0 was researched. The optimized reaction conditions
were as follows: 140 °C, catalyst amount 0.9 wt%, meth-
anol/PO molar ratio 4.0 and 6 h. The PO conversion and
PM selectivity were 93.2 and 97.4%, respectively. Above
all, almost all the PM was 1-methoxy-2-propanol, for no
2-methoxy-1-propanol was detected by GC analysis in the
reaction products, and the catalyst could be reused for five
times.
glycol ether, it is expected as a safe substitute for toxic
ethylene glycol ether [2]. Propylene glycol monomethyl
ether (PM) is a member of the family of glycol ether with a
wide variety of consumer products and industrial applica-
tions. PM is used in the chemical, agricultural, automotive,
paint, lacquer, and varnish industry [3]. Its predominant use
is as a solvent in various manufacturing processes and as a
chemical building block for the manufacture of propylene
glycol methyl ether acetate. PM is produced commercially
by reacting propylene oxide (PO) with methanol in the
presence of catalyst. Due to the asymmetry of propylene
oxide, the epoxide ring of propylene oxide may open either
at the C–O bonds, so that the reaction yields a mixture of
two isomers with the a-isomer, 1-methoxy-2-propanol
(PPM) and the b-isomer, 2-methoxy-1-propanol. Among
the products, PPM has a lower toxicity than 2-methoxy-1-
propanol, so high selectivity is required for the reaction [2].
Acid catalysts, such as BF3, heteropolyacids and solid
acids, can be used for the reaction and appear to exhibit
good catalytic activity, but they inevitably produce unde-
sirable 2-methoxy-1-propanol and other byproducts. The
epoxy ring of PO may preferentially open at the least ste-
rically hindered position over the basic catalyst and lead to
most secondary alcohol PPM [2, 4–6]. Homogenous base
catalysts such as sodium methoxide and NaOH, show high
activity and selectivity, but they suffer from the drawbacks
of separation, liquid waste treatment and corrosion [4, 7].
The replacement of currently used homogeneous alkaline
catalysts by solid catalysts can result in catalyst reuse and
waste stream reduction and make the process of indus-
trial production environmentally benign and economical
because of their non-corrosiveness and easy separation
from products [8]. Zhang et al. [4] reported that MgO
which showed PO conversion 71.0% and PPM selectivity
92.0% was found to be more active and selective than CaO
Keywords Mg–Al hydrotalcite ꢀ Propylene oxide ꢀ
Propylene glycol monomethyl ether ꢀ Catalyst
1 Introduction
Glycol ether is a component of many commercial products,
such as air freshener, antifreeze, cosmetic product and
plastics [1]. Because of the low toxicity of propylene
H. Zeng (&) ꢀ Y. Wang ꢀ Z. Feng ꢀ K. You ꢀ C. Zhao ꢀ
J. Sun ꢀ P. Liu
College of Chemical Engineering, Xiangtan University,
Xiangtan 411105, Hunan, China
e-mail: hongyanzeng99@hotmail.com; hyzeng@xtu.edu.cn
123

Synthesis of Propylene Glycol Monomethyl Ether Over Mg/Al Hydrotalcite Catalyst
95
for the formation of 1-methoxy-2-propanol. Al2O3 showed
high PO conversion 98.2%, but low PPM selectivity 75.0%.
Acid-catalyzed reaction
500 °C for 7 h in a muffle furnace, which were designated
as LDOs.
2.2 Catalyst Characterization
HO δ+
Powder X-ray diffraction (XRD) patterns were obtained in
the 2h range 3–70° using a Rigaku D/MAX-3C instrument
with Cu Ka source (k = 0.1541 nm), and the X-ray pat-
terns were indexed according to the JCPDS database.
FT-IR spectrum was recorded on Perkin-Elmer Spectrum
One B instrument using KBr pellet technique. Scanning
electron micrograph (SEM) was obtained with a JEOL
JSM-6700F instrument. The specific surface areas were
calculated using the Brunauer–Emmett–Teller (BET)
method based on the N2 adsorption isotherms. To measure
the basic strengths of the Mg–Al hydrotalcites, the fol-
lowing Hammett indicators were used: bromothymol blue
(H- = 7.2), phenolphthalein (H- = 9.8), 2,4-dinitroani-
line (H- = 15.0) and nitroaniline (H- = 18.4). About
0.05 g of samples were shaken with a methanol solution of
Hammett indicators (5 mL) and left to equilibrate for 2 h
after which no further color changes were observed. The
basic strength is quoted as being stronger than the weakest
indicator which exhibits a color change, but weaker than
the strongest indicator that produces no change. To mea-
sure the basicity of the Mg–Al hydrotalcites, the method of
Hammett indicator-benzene carboxylic acid titration was
used [12].
OR
O
H+
CH3-CH -CH2
+OR
H+
CH3-CH-CH2 HOR
CH3-CH-CH2-OH
H
δ
Basic-catalyzed reaction
O δ
CH3-CH-CH2
OR
OH
O
CH3-CH-CH2 -OR
CH3-CH-CH2-OR
δ
In recent years, some patents and papers have shown
that synthetic anionic double hydroxide clays are efficient
and selective catalysts for the synthesis of glycol ether [2,
9, 10]. However, the hydrotalcites have not been studied
sufficiently so far. Only a few papers covering the influence
of acid-base properties of layered double oxides on the
synthesis of PM have been published, and in the literatures,
the PPM selectivity are low [2, 7]. Recently, it was found
that Mg–Al hydrotalcites showed the high activity and
selectivity to the desired product PPM in the synthesis of
PM. In this study, a series of Mg–Al hydrotalcites with
various Mg/Al molar ratios were studied for improving the
activity and selectivity of the Mg–Al hydrotalcite catalyst.
The characterization of the Mg–Al hydrotalcites was
studied with XRD, FT-IR, SEM, BET and Hammett titra-
tion method and the relationship between structure and
catalytic performance was discussed, which might give
fundamental insight into the catalytic mechanism of the
catalyst.
2.3 Catalyst Test
The reactions were carried out in a sealed stainless-steel
autoclave (500 mL), equipped with mechanical stirring and
sampling device. In the reactor, an amount of methanol,
propylene oxide and the catalyst were poured before
closing. After purging for 10 min with N2 flow, the mixture
was heated to the reaction temperature. When the reaction
was over, the mixture was cooled down to room tempera-
ture. The solid catalyst was filtered, washed with methanol,
then dried, and used for the recycling test. The reaction
products were analyzed on a gas chromatograph (Agilent
6980N with a flame ionization detector) equipped with a
30-m DB-1 capillary column. The carrier gas N2 (19.0 mL
min-1) and the oven temperature program consisted of:
start at 40 °C, ramp at 10 °C min-1 to 200 °C (16 min)
with injector and detector temperature 250 °C. The
n-hexane was used as the internal standard.
2 Experimental
2.1 Catalyst Preparation
All the Mg–Al hydrotalcites were prepared by urea method
-
(urea/NO3 molar ratio of 4.0) [11]. An aqueous solution
containing various Mg/Al molar ratios of Mg(NO3)2ꢀ6H2O
and Al(NO3)3ꢀ9H2O was placed into a three-neck flask.
The solution was maintained at 105 °C for 12 h under
stirring (300 rpm), and then aged statically at the same
temperature for another 18 h. The formed solids were
collected by filtration and washed with deionized water,
and subsequently dried at 100 °C for 18 h, which were
denoted as LDHs. Part of the LDHs were calcined at
2.4 Orthogonal Method
A 4-factor plus 4-level L16 (44) orthogonal experimental
design was applied to determine the optimal conditions
[13]. The orthogonal experimental design was an effective
123

96
H. Zeng et al.
samples in the composition range Mg2?/Al3? = 0.25 to
6.0 were similar, and a hydrotalcite-like layered double
hydroxide material (LDH; JCPDS file 22-700) was
observed in each case, giving a series of (00l) peaks
appearing as narrow symmetric lines at low angle, corre-
sponding to the basal spacing and high order reflections.
The purity of the phase, however, depended on the Mg/Al
molar ratio. X-ray diffraction pattern of Mg/Al molar ratios
of 0.25 and 0.5 showed the presence of AlOOH (boehmite)
as a phase impurity with LDH as the major phase. Mg/Al
molar ratios of 1.0 and 4.0 showed the X-ray pattern of
single phase LDH compound which matched totally with
the standard LDH pattern [15]. Generally, the sharpness
and intensity of XRD were considered to be proportional to
the crystallinity [16, 17]. Although the purely layered
structure with characteristic and symmetric reflections
could be obtained when Mg/Al molar ratio was equal to
1.0, the lower intensities and broader half-height widths at
low 2h angle were observed which indicated that the
crystallinity of LDH Mg/Al 1.0 was poorer than that of
the LDH Mg/Al 4.0. In the sample of LDH Mg/Al 6.0, the
formation of MgCO3 in addition to LDH was revealed.
The appearance of MgCO3 as an impurity was due to the
increase of magnesium content. A closer examination of
Fig. 1 revealed that the peak sharpness and intensity
increased as the Mg/Al molar ratio was added from 0.25
to 4.0, indicating an improved crystallinity, and then
decreased when the Mg/Al molar ratio was above 4.0. So,
the sample LDH Mg/Al 4.0 possessed the highest
crystallinity.
Table 1 Factors and levels of the L16(44) orthogonal experiments
Level Factor
Temperature Methanol/PO molar Catalyst
Time
(°C)
T
ratio
M
amount (wt%) (h)
A
t
1
2
3
4
80
100
120
140
4:1
3:1
2:1
1:1
0.6
0.9
1.2
1.5
2
4
6
8
method for investigating the effect of the factors on the PO
conversion and PM selectivity by reducing the experi-
mental numbers. Furthermore, the optimal reaction condi-
tions and the significance of every factor’s effect on the PO
conversion and PM selectivity could be obtained. Four
independent experimental variables, namely reaction tem-
perature (T), methanol/PO molar ratio (M), catalyst amount
(wt% of total reactants, A) and reaction time (t) were
selected as controlled factors. The data were analyzed by
analysis of variance (ANOVA). The factors and levels in
the orthogonal experiment were given in Table 1. The PO
conversion and PM selectivity were defined as the fol-
lowing equations, respectively.
mPOactual
PO conversion ¼
PM selectivity ¼
ꢁ 100%
mPOtheoretical
mPM
ꢁ 100%
m
where the mPOactual (g) and mPOtheoretical (g) were the mass
of PO reacted actually and theoretically, respectively. The
mPM and m were the mass of PM obtained actually and PM
produced by mPOactual theoretically, respectively.
2.5 Statistical Analysis
♣
♣
Hydrotalcite
♦
MgCO3
♥
AlOOH
All the experiments were carried out three times in order to
determine the variability of the results and to assess the
experimental errors. In this way, the arithmetical averages
and the standard deviations were calculated for all the
results.
♣
♣
♣
♦
♦
♣
♣
♣
♣
♣
♣
♣
e
♣
♣
♣
♣
d
c
♣
♣
♣
♣
♣
♣
3 Results and Discussion
3.1 Microstructural Studies
3.1.1 XRD Analysis
♣
♣
♥
♥
♣
♣
♣
♣
b
a
♥
♥
♥
♣
♣
♣
♣
0
10
20
30
2θ (°)
40
50
60
70
Figure 1 demonstrated the XRD patterns of the Mg/Al
hydrotalcites with different Mg/Al molar ratios. Indexing
of the diffraction peaks was done using standard JCPDS
file 22-700 [14]. Clearly, the structural properties of these
Fig. 1 X-ray diffraction patterns of the LDHs samples. a Mg/Al 0.25,
b Mg/Al 0.5, c Mg/Al 1.0, d Mg/Al 4.0, e Mg/Al 6.0
123

Synthesis of Propylene Glycol Monomethyl Ether Over Mg/Al Hydrotalcite Catalyst
97
band of the carbonate anions was observed at 1370 cm-1
(t3) for all the LDHs samples. For the Mg/Al molar ratios
of 0.25, 0.5 and 6.0, this peak split into a double band
indicating a lowering of carbonate anion symmetry and
the disordered nature of the interlayer, which caused the
removal of the degeneracy of the t3 and t4 modes [15]. The
appearance of some very faint absorption at 670–700 cm-1
for the Mg/Al molar ratios of 0.25 and 0.5 implying the
disordered nature, and the band at 687 cm-1 for Mg/Al
molar ratio of 4.0 was strongest in all other samples,
showing that the above interpretation was justified. The
lowering of the symmetry also caused the activation of the
t1 mode around 1050 cm-1, inactive when the carbonate
anion retained its full symmetry [15, 19, 20]. The t1
A
a
b
c
d
e
B
a
b
c
d
e
vibrational mode of carbonate ions appeared at 1060 cm-1
.
The above assumption was supported by the result that the
t1 mode was found for the Mg/Al molar ratios of 0.25, 0.5,
1.0 and 6.0, but not for the Mg/Al molar ratios of 4.0. The
results showed that the LDH Mg/Al 4.0 possessed high
symmetry and ordered nature of the interlayer anions.
The bands around 553 and 770 cm-1 could be assigned to
translation modes of the hydroxyl groups mainly influenced
by the trivalent aluminum but also influenced by probably
Mg2? in its coordination [19]. It could be seen that the
band at 770 cm-1 were clearly observed for all the Mg/Al
molar ratios, and the band at 553 cm-1 for the Mg/Al
molar ratios of 1.0, 4.0 and 6.0, but not for 0.25 and 0.5.
The intensity of the two bands was the highest for the Mg/
Al molar ratios of 4.0. The appearance of absorption band
at around 445 cm-1(d O–M–O) characteristic of lattice
vibrations of [Mg, Al] octahedral sheets demonstrated the
crystallization of the LDHs [21]. The intensity of the sharp
peak at 445 cm-1 increased with decrease in Al concen-
tration in the hydrotalcites with Mg/Al molar ratios of
0.25–4.0, and then decreased in the hydrotalcite with Mg/
Al molar ratios of 6.0. The observed discrepancies among
the LDHs with various Mg/Al molar ratios in FT-IR
spectra were consistent with those in XRD patterns
(Fig. 1). The samples with Mg/Al molar ratios of 1.0 and
4.0 showed pure the hydrotalcite pattern while the others
showed impurities of AlOOH (Mg/Al 0.25 and 0.5) and
MgCO3 (Mg/Al 6.0).
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber (cm-1)
Fig. 2 FT-IR spectra of the LDHs and LDOs samples. A LDHs
samples, B LDOs samples; a Mg/Al 0.25, b Mg/Al 0.5, c Mg/Al 1.0, d
Mg/Al 4.0, e Mg/Al 6.0
3.1.2 FT-IR Spectra
The FT-IR spectra of the Mg–Al hydrotalcites with dif-
ferent Mg/Al molar ratios in the region 400–4000 cm-1
were shown in Fig. 2. For all the hydrotalcites with dif-
ferent Mg/Al molar ratios, the broad band was found at
around 3450 cm-1 (m1-OH-1) and it was ascribable to the
brucite-like layers (OH- stretching vibration), caused by
the interlayer water molecules and hydroxyl groups in the
brucite-like layers [18]. In lower frequency region
(Fig. 2A), a peak at about 1650 cm-1 (d-HOH) in all the
LDHs samples could be attributed to the bending mode of
interlayer water [18, 19]. However, a shoulder around
1650 cm-1, or of a double band in the region 1640–
1660 cm-1, was observed in the cases of Mg/Al molar
ratios of 0.25, 0.5 and 1.0 (B1.0). This might indicate the
presence of a very small amount of aluminum not incor-
porated in the hydrotalcite structure [19]. The carbonate
anion in a symmetric environment was characterized by a
D3h planar symmetry, with three IR active absorption
bands, as well as in the case of the free carbonate anion. In
most hydrotalcites the three bands were observed at 1350–
1380 cm-1 (t3), 850–880 cm-1 (t2) and 670–690 cm-1
(t4) [15]. This was almost what was observed for every
hydrotalcite irrespective of the nature of the octahedral
sheets suggesting full symmetry for the interlayer anions
[20]. The three bands were observed only in the samples
with Mg/Al molar ratios of 1.0 and 4.0, which implied the
Mg/Al molar ratios of 1.0 and 4.0 had a rather symmetric
structure for the interlayer anions. The main absorption
The spectra of the calcined samples (LDOs) were dif-
ferent from those of the precursors (LDHs) (Fig. 2B). The
(m1-OH-1) mode appeared around 3450 cm-1, where
the intensity of the band was obviously lower than that of
the uncalcined samples (LDHs). The weak band at around
1640 cm-1 was observed, but the shoulder in the region
disappeared for the LDO Mg/Al 0.25, 0.5 and 1.0 (B1.0).
The band appeared at around 1390 cm-1 (t3) which was
about 20 cm-1 higher than that of the LDHs, and no
splitting was found for all the LDO samples. The presence
of a band of 678 cm-1 was only in the LDO Mg/Al 4.0 and
123

98
H. Zeng et al.
Fig. 3 SEM images of the LDHs and LDOs samples. A LDH Mg/Al 0.25, B LDH Mg/Al 4.0, C LDH Mg/Al 6.0, D LDO Mg/Al 0.25, E LDO
Mg/Al 4.0, F LDO Mg/Al 6.0
6.0 samples. The peak at 452 cm-1 stood for the lattice
vibrations of [Mg, Al] octahedral sheets, but also appeared
only in the LDO Mg/Al 4.0 and 6.0 samples. There was a
peak appearing at 864 cm-1 only in the LDO Mg/Al 4.0
sample, which implied the LDO Mg/Al 4.0 remained the
hydrotalcite layered structure. Calcination at 500 °C
induced dehydration, dehydroxylation and decarbonation
leading to the formation of mixed oxides of MgO and
Al2O3, and the FT-IR spectra of calcined samples reflected
these changes [22].
morphology and the size of the LDO Mg/Al 4.0 particles,
similar to the result reported by Lei et al. [23]. This was in
agreement with the assumption from IR that the LDO Mg/
Al 4.0 remained the hydrotalcite layered structure
(Fig. 2B). When the LDH sample with Mg/Al 0.25 was
prepared, strong textural change was observed. The sample
had irregular platelets forming aggregates only with little
hexagonal plate-shaped particles (Fig. 3A). The calcination
at 500 °C made the layered structure of the LDO Mg/Al
0.25 disappear (Fig. 3D) and the inorganic layers were no
more identified, which indicated calcination led to the
destruction of the layered structure. For the sample LDH
Mg/Al 6.0, individual platelets were still well observed
with irregular and imperfect platelet crystals besides the
perfect hexagonal platelets were observed (Fig. 3C). The
typical hexagonal platelets were still identified after the
calcination, and a lot of small particles and platelet
aggregates appeared in the Fig. 3F, in accordance with the
IR result for the LDO Mg/Al 6.0.
3.1.3 SEM Images
In order to investigate the morphology for the LDHs and
LDOs, the LDHs and LDOs with different Mg/Al molar
ratios 0.25, 4.0 and 6.0 were observed with SEM (Fig. 3).
Thin flat crystals indicating the layered structure were
observed for all the LDH samples. The plate-plate over-
lapping of crystallites gave rise to interfaces that could
accommodate extrinsic surface water, as well as other
adsorbates. All the particles of the LDH Mg/Al 4.0 showed
well-developed hexagonal plates with narrow size distri-
bution (2–4 lm) and the samples could be seen to be made
up of individual platelet particles in line with the typical
morphology for hydrotalcite-like materials (Fig. 3B). This
was in agreement with the XRD and FT-IR results men-
tioned above. The morphology and particle size of the
calcined material LDO with Mg/Al 4.0 (Fig. 3E) was
basically similar to those of the LDH precursor (Fig. 3B),
where the hexagonally lamellar morphology was integrally
preserved after calcination at 500 °C. It was interesting
to note that the calcination process did not alter the
3.1.4 BET Surface Area
Table 2 showed the results of the BET for the LDHs and
LDOs. The Mg/Al molar ratio in the samples had a strong
influence on the surface areas. With increase in the Mg/Al
molar ratio, the BET areas of the materials decreased from
172.92 to 53.14 m2 g-1 for the LDHs samples, and from
237.92 to 92.03 m2 g-1 for the LDOs samples. This
increase in surface area after calcination correlated with the
removal of water and CO2 from the interlayer space and an
increase in the pore volume. This behavior was similar to
other hydrotalcite-like compounds and their homologous
123

Synthesis of Propylene Glycol Monomethyl Ether Over Mg/Al Hydrotalcite Catalyst
99
Table 2 BET surface area and
Sample
LDHs
LDOs
base strength of Mg–Al
hydrotalcites with different Mg/
Al molar ratios
SBET (m2 g-1
)
Basic strength (H-)
SBET (m2 g-1
)
Basic strength (H-)
Mg/Al 0.25
Mg/Al 0.5
Mg/Al 1.0
Mg/Al 2.0
Mg/Al 3.0
Mg/Al 4.0
Mg/Al 5.0
Mg/Al 6.0
172.92
132.64
111.12
–
–
237.92
215.67
201.24
–
–
–
–
7.2 \ H- \ 8.2
8.2 \ H- \ 9.8
9.8 \ H- \ 12.2
9.8 \ H- \ 12.2
8.2 \ H- \ 9.8
7.2 \ H- \ 8.2
9.8 \ H- \ 12.2
12.2 \ H- \ 15.0
15.0 \ H- \ 17.2
15.0 \ H- \ 17.2
12.2 \ H- \ 15.0
9.8 \ H- \ 12.2
98.72
80.16
–
189.23
165.87
–
53.14
92.03
calcination products [16]. The calcination of hydrotalcite-
like samples at 500 °C resulted in mesoporous mixed
oxides with high surface areas. The LDO Mg/Al 0.25 had
the highest surface area (237.92 m2 g-1), while the LDO
Mg/Al 6.0 had the lowest surface areas. This difference
might be attributed to Al content in the Mg–Al hydrotalcite
samples.
4.0
A
H
H
7.2-9.8
9.8-15.0
3.0
2.0
1.0
Total basicity
4.0
B
3.1.5 Surface Basicity
3.0
2.0
1.0
Though both basic and acidic sites are present in LDHs
[23], it is reported that the synthesis of 1-methoxy-2-pro-
panol should be closely related to the basic sites with
moderate strength [4]. The basic strengths of various LDHs
and LDOs were determined by using Hammett indicator
and their basic strengths were in the ranges of 7.2–12.2 and
9.8–17.2, respectively (Table 2). The basicity of the
Mg–Al hydrotalcites with different Mg/Al molar ratios was
shown in Fig. 4. The main basic sites with H- in the range
of 7.2–9.8 (weak basic sites) and the other sites with H- in
the range of 9.8–15.0 (medium basic sites) were observed
(Fig. 4), where all the samples had very few number of
strong basic sites. The total basicity of the sample was
increased gradually with the Mg/Al molar ratio and came
up to the maximum value at the Mg/Al molar ratio of 3.0.
However, with further increase in the Mg/Al molar ratio it
was observed that the basicity was decreased. It was found
that the maximum (2.5 mmol g-1 for the LDH and
3.6 mmol g-1 for the LDO) of the basic sites was at the
Mg/Al molar ratio of 3.0. The similar results were reported
by Xie et al. [24]. The basic strength and basicity of the
LDHs samples were significantly lower than those of
the LDOs. The total basicity in the ranges of 7.2–9.8 from
the LDOs increased by 1.3–1.9 times compared with those
from the LDHs. It was particularly noteworthy that the
basicity in the ranges of 9.8–17.2 of the LDOs rose by
about 3.1–7.7 times comparing with that of the LDHs,
implying that the basicity in the ranges of 9.8–17.2 was
significantly raised after calcination at 500 °C. There was
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Mg/Al molar ratio
Fig. 4 Basicity of Mg/Al hydrotalcites with different Mg/Al molar
ratios. A LDHs, B LDOs
no marked difference in the basicity of the LDOs in the
ranges of 9.8–17.2 (medium basic sites) between the LDO
Mg/Al 3.0 and LDO Mg/Al 4.0 (Fig. 4B).
3.2 Catalytic Performance
The calcined hydrotalcites (LDOs) with different Mg/Al
molar ratios were chosen for the catalytic test. The catalytic
activity was tested in the synthesis of PM from methanol
and PO (Fig. 5). The experimental results indicated that the
catalytic activity (PO conversion and PM selectivity) was
improved with increase in Mg content when the Mg/Al
molar ratio was below 4.0, but when the Mg/Al molar ratio
was above 4.0, the catalytic activity dropped. Clearly, such
a tendency was basically correlated to that for the variation
of the LDOs basicity in the ranges of 9.8–17.2 with various
Mg/Al ratios (Fig. 4B). Another reason for the enhanced
activity in the LDO Mg/Al 4.0 might be that the precursor
LDH Mg/Al 4.0 showed better crystallinity in the XRD and
123

100
H. Zeng et al.
3.3 Orthogonal Experiment
100
90
80
70
60
50
40
100
90
80
70
60
50
40
Besides the pronounced effect of Mg/Al molar ratios on the
catalytic activity of the catalyst, a number of factors such
as temperature, reactant and catalyst concentrations, and
reaction time could also affect the synthesis of PM since it
was a liquid–solid two-phase reaction with a complicated
reaction mechanism. The effects of temperature, methanol/
PO molar ratio, catalyst amount and reaction time were
studied. In this work, an orthogonal experimental design
was conducted to investigate some pivotal reaction engi-
neering parameters, which exerted strong influence on the
PO conversion and PM selectivity.
PM selectivity
PO conversion
As listed in Table 1, an orthogonal table of L16(44) was
adopted to examine four variables as follows: temperature
(T), methanol/PO molar ratio (M), catalyst amount (wt% of
total reactants, A) and reaction time (t) at different levels.
The observed PO conversion and PM selectivity varied
remarkably with different parameters, suggesting that the
optimization of reaction conditions for improving PO
conversion and PM selectivity was a significant techno-
economical necessity. The relative importance of factors to
process indexes and the optimal set of factor levels were
determined by margin and variance analysis according to
the results shown in Table 3. In terms of the PO conversion
from Table 4, the range of the factors was T [ M [ t [ A,
namely temperature (RPO = 158.8) was the most important
factor. The second important variable was the methanol/PO
molar ratio (RPO = 108.2), the third was reaction time
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Mg/Al molar ratio
Fig. 5 Effect of Mg/Al molar ratio on the activity of the LDOs
catalyst. Methanol/PO molar ratio 4.0, 1.0 wt% catalyst amount at
140 °C for 6 h
SEM data and the loss of the lamellar structure had pre-
viously been linked to poor activity [25]. The high crys-
tallinity resulted in a regular arrangement of hydroxyl
groups on the surface and a high density of appropriately
acid–base balance [23]. The high catalytic activity and
selectivity of the LDOs for the synthesis of PM were
attributed to the presence of a high density of the medium
basic sites. Calcination at high temperature decomposes the
hydrotalcite into interactive, high surface area and well-
dispersed mixed Mg–Al oxides which presents basic sites
that are associated to structural hydroxyl groups as well as
strong Lewis basic sites associated to O2-Mn? acid–base
pairs [24]. The weak basic sites were associated with OH
groups and the medium basic sites were assigned on M–O
pairs [2, 4, 7]. The LDO Mg/Al 4.0 showed the highest PO
conversion (92.7%) and PM selectivity (97.4%). The LDO
Mg/Al 3.0 had the highest total basicity in all LDOs
samples as well as similar basic strength (Table 2) and
basicity in the ranges of 9.8–17.2 (medium basic sites)
(Fig. 4B). However, the performance of the LDO Mg/Al
3.0 was lower than that of the Mg–Al LDO 4.0. This dis-
crepancy might well be due to the reaction types and their
influence on the reactivity of the acid–base sites playing a
role in the mechanism. Under the conditions used in the
present work, part of the medium basic sites of the LDO
Mg/Al 3.0 likely did not take part in the reaction, but good
correspondence existed between the number of the acid–
base sites and the catalytic performance for the LDO Mg/
Al 4.0. More appropriately acid–base balance on the sur-
face of the Mg/Al 4.0 than Mg/Al 3.0 might result in a
higher catalytic activity. So, the LDO Mg/Al 4.0 served as
the catalyst for further researches on the optimization.
Table 3 Design and results of the orthogonal experiments
No.
T
M
A
t
PO
conversion (%)
PM
selectivity (%)
1
80
4:1
2:1
1:2
1:4
4:1
2:1
1:2
1:4
4:1
2:1
1:2
1:4
4:1
2:1
1:2
1:4
1.5
0.6
0.9
1.2
0.6
0.9
1.2
1.5
0.9
1.2
1.5
0.6
1.2
1.5
0.6
0.9
4
6
8
2
2
4
6
8
8
2
4
6
6
8
2
4
66.5
56.6
29.6
18.0
47.0
38.7
33.8
41.8
82.1
57.6
47.2
53.5
92.4
76.5
69.4
81.8
97.3
94.5
93.1
93.8
96.5
94.3
89.9
92.2
98.1
93.6
87.1
87.4
96.5
91.2
86.5
84.6
2
80
80
3
4
80
5
100
100
100
100
120
120
120
120
140
140
140
140
6
7
8
9
10
11
12
13
14
15
16
T temperature (°C), M methanol/PO molar ratio, A catalyst amount
(wt%), t time (h)
123

Synthesis of Propylene Glycol Monomethyl Ether Over Mg/Al Hydrotalcite Catalyst
101
Table 4 Margins of the conversion and PM selectivity, RPO and RPM
Factors
PO conversion (%)
Sum of each level
PM selectivity (%)
Sum of each level
RPO
RPM
1
2
3
4
1
2
3
4
T
M
A
t
170.7
288.1a
226.5
192.0
161.3
229.4
232.2a
234.1
240.4
179.9
201.8
236.3a
320.1a
195.0
231.9
230.0
158.8
108.2
30.4
378.7a
388.4a
364.9
370.4
372.9
373.6
370.1
363.3
366.2
356.6
373.8a
368.3
358.8
358
19.9
31.8
8.9
367.8
374.6a
44.3
11.3
T temperature (°C), M methanol/PO molar ratio, A catalyst amount (wt%), t time (h), RPO range of PO conversion, RPM range of PM selectivity
a
The optimum level
(RPO = 44.3). It was concluded that the optimal conditions
were 140 °C, methanol/PO molar ratio 4.0, and catalyst
amount 0.9 wt% for 6 h. As for the PM selectivity, the
range of the factors was M [ T [ t [ A, namely metha-
nol/PO molar ratio (RPM = 31.8) was the first effective
factor, followed by temperature (RPM = 19.9), and then
reaction time (RPM = 11.3), within the scope of selected
reaction conditions. The optimum conditions were 80 °C,
methanol/PO molar ratio 4.0, and catalyst amount 1.2 wt%
for 8 h (Table 4).
experiments. The optimal conditions were predicted
through the orthogonal experiment, but the optimal tem-
perature and methanol/PO molar ratio were on the edge of
the two levels in the orthogonal experiment. Hence, further
studies in the extended range were conducted as follows,
for the purpose of interpreting the reaction and improving
operation of the process.
3.4 Influence of Temperature
The significance of each factor was further checked
using analysis of variance (ANOVA) in Tables 5 and 6.
For the PO conversion, temperature was highly significant
(**), and methanol/PO molar ratio was significant (*),
which corresponded with the margin analysis (Table 4).
The variance analysis of the PM selectivity was different to
that of the PO conversion, where methanol/PO molar ratio
and temperature was highly significant (**). No obvious
effect of reaction time and catalyst amount on the PO
conversion and PM selectivity was observed (Table 4), as
indicated by the low F value (Tables 5, 6). The most likely
reason was that the reaction tended to equilibrium within
the designed level of time.
The effect of the reaction temperature (from 80 to 160 °C)
on the PO conversion and PM selectivity was studied
(Fig. 6). Clearly, temperature strongly affected the PO
conversion in the reaction. At moderate temperatures
between 80 and 100 °C, the PO conversion was relatively
low, which was possibly due to that PO might not be
activated at a low reaction temperature. The PO conversion
100
90
80
70
60
50
40
30
20
100
95
90
85
80
75
In order to test the speculation on the time, the reaction
time from 0.5 to 8 h was investigated in the conditions of
140 °C, catalyst amount 0.9 wt% and methanol/PO molar
ratio 4.0. It was found that the PO conversion (0.5 h:
29.5%, 1.0 h: 80.3%, 2.0 h: 89.1%) significantly increased
when reaction time was increased from 0.5 to 2 h, and then
weakly increased when the time was more than 2 h (4.0 h:
92.8%, 6.0 h: 93.1%, 8.0 h: 94.1%). At the same time, PM
selectivity (0.5 h: 97.5%, 1.0 h: 97.5%, 2.0 h: 97.4%) was
stable when reaction time was increased from 0.5 to 2 h,
and then slightly descended when the time was more than
2 h (4.0 h: 97.2%, 6.0 h: 97.0%, 8.0 h: 95.8%). The results
confirmed our supposition on the effect of reaction time.
In view of production cost, the catalyst amount 0.9 wt%
and reaction time 6 h were determined for further
PO conversion
PM selectivity
80
100
120
Temperature (°C)
140
160
Fig. 6 Effect of reaction temperature on PO conversion and PM
selectivity. Methanol/PO molar ratio 4.0, 0.9 wt% catalyst amount for
6 h
123

102
H. Zeng et al.
3.6 Reusability of the Catalyst
increased sharply with increase in temperature then
reached a plateau value, where a nearly maximum PO
conversion of 92.8% was obtained at 140 °C. Contrarily,
the PM selectivity was decreased slowly with increase in
temperature. When temperature was above 140 °C, the PM
selectivity gradually declined because side reactions
increased, implying that higher temperature were not
favorable to the reaction. It had an optimal value, under
which the effect of synergism was the best. As a result,
140 °C was selected as the feasible reaction temperature to
achieve high PO conversion and PM selectivity.
In order to evaluate the stability of the catalyst, reusability
of the catalyst was studied under the optimal conditions.
The catalyst was separated out by filtration under reduced
pressure in the end of the reaction and then reused for the
next run under the same conditions. The results as shown in
Table 7 indicated that the PO conversion and PM selec-
tivity was not affected even at the five runs with the reused
catalyst. This finding implied that the catalyst could be
recovered and recycled efficiently. Furthermore, the reus-
ability of the catalyst would be further improved when the
amount of the reactants and catalyst was increased, which
needed to be further investigated.
3.5 Influence of Methanol/PO Molar Ratio
Based on the results above (Table 4, 5, 6), the reaction time
(6 h) and catalyst amount (0.9 wt%) could be fixed, but the
methanol/PO molar ratio needed to be tested further. In
order to get the optimum factor, the six methanol/PO molar
ratios, namely 3.0, 4.0, 5.0, 6.0, 7.0 and 8.0, were chosen
and the experimental runs were conducted at 140 °C
(Fig. 7). The optimum molar ratio was found to be 4.0. As
represented in Fig. 7, by increasing methanol/PO molar
ratio from 3.0 to 4.0, PO conversion and PM selectivity
were smoothly increased. When the methanol/PO molar
ratio was above 4.0, the PO conversion and PM selectivity
had no significantly increase. Therefore, we could conclude
that to elevate the PO conversion and PM selectivity, an
excess methanol feed was effective to a certain extent. So,
the optimum molar ratio was 4.0.
4 Conclusions
The Mg/Al hydrotalcites were prepared by urea method.
The effects of the Mg/Al molar ratio on the catalytic per-
formance of the LDOs in the PM synthesis were studied.
The LDO Mg/Al 4.0 catalyst revealed high activity for the
synthesis of PM from PO and methanol. The catalyst
showed the highest PO conversion of 93.2% and PM
selectivity of 97.4%. An optimal balance of acid–base
properties and high crystallinity made the LDO Mg/Al 4.0
an excellent catalyst in the synthesis reaction. As shown
above, the activity might be further improved by optimiz-
ing reaction conditions. The reaction conditions for the
Table 5 Results of the
ANOVA analysis from PO
conversion
Source
Sum of squares
of variance
Degree
of freedom
Mean
square
F value
p
Significance
T
4068.33
1718.98
145.22
361.36
389.55
6683.44
3
3
3
3
1356.11
572.99
48.41
10.44
4.41
0.37
0.93
[0.01
[0.05
**
*
T temperature (°C), M
M
methanol/PO molar ratio, A
catalyst amount (wt%), t time
(h), F0.01(3,12) = 5.95 and
F0.05(3,12) = 3.49
A
t
120.45
Error
Total
* Significant, ** Highly
significant
Table 6 Results of the
ANOVA analysis from PM
selectivity
Source
Sum of squares
of variance
Degree
of freedom
Mean square
F value
p
Significance
T
55.3
3
3
3
3
18.43
56.025
3.542
5.59
7.16
21.76
1.38
[0.01
[0.01
**
**
M
168.075
10.625
16.575
7.725
T temperature (°C), M
A
methanol/PO molar ratio, A
catalyst amount (wt%), t time
(h), F0.01(3,12) = 5.95 and
F0.05(3,12) = 3.49
t
2.15
Error
Total
258.3
** Highly significant
123

Synthesis of Propylene Glycol Monomethyl Ether Over Mg/Al Hydrotalcite Catalyst
103
Acknowledgement This work was supported by Scientific Research
Fund of Hunan Provincial Education Department of China though
key-project (No. 08A080).
Table 7 The reusability of the LDO Mg/Al 4.0 catalyst
Used number
PO conversion (%)
PM selectivity (%)
1
2
3
4
5
93.3
93.2
92.5
92.7
91.9
97.4
97.8
97.4
97.2
97.0
References
1. Hootha MJ, Herberta RA, Hasemana JK, Orzecha DP, Johnsonb
JD, Buchera JR (2004) Toxicology 204:123
2. Cheng W, Wang W, Zhao Y, Liu L, Yang J, He M (2008) Appl
Clay Sci 42:111
3. Corley RA, Gies RA, Wu H, Weitz KK (2005) Toxicol Lett
156:193
4. Zhang W, Wang H, Li Q, Dong Q, Zhao N, Wei W, Sun Y (2005)
Appl Catal A 294:188
5. Chitwood HC, Freure BT (1946) J Am Chem Soc 68:688
100
95
100
95
¨
6. Martins L, Holderich W, Cardoso D (2008) J Catal 258:14
7. Zhang W, Wang H, Wei W, Sun Y (2005) J Mol Catal A 231:83
8. Fujita S-I, Bhanage BM, Aoki D, Ochiai Y, Iwasa N, Arai M
(2006) Appl Catal A 313:151
90
90
9. Atkins MP, Jones W, Chibwe M, US Patent 5,110,992 (1992)
10. Malherbe F, Besse J-P, Wadel SR, Smith WJ (2000) Catal Lett
67:197
85
85
11. Zeng HY, Deng X, Wang YJ, Liao KB (2009) AIChE J 55:1229
12. Yang Z, Xie W (2007) Fuel Processing Technol 88:631
13. Raj CBC, Quen HL (2005) Chem Eng Sci 60:5305
14. JCPDS X-ray powder diffraction file, no. 22-700 (1986)
15. Cavani F, Trifiro F, Vaccari A (1991) Catal Today 11:173
16. Kannan S, Narayanan A, Swamy CS (1996) J Mater Sci 31:2353
17. Costantino U, Marmottini F, Nocchetti M, Vivani R (1998) Eur J
Inorg Chem 10:1439
PO conversion
PM selectivity
80
80
75
8.0
75
3.0
4.0
5.0
6.0
7.0
Methanol/ PO molar ratio
18. Dos Reis MJ, Silverio F, Tronto J, Valim JB (2004) J Phys Chem
Solids 65:487
19. Kloprogge TJ, Frost RL (1999) J Solid State Chem 146:506
20. Hernandez-Moreno MJ, Ulibarri MA, Rendon JL, Serna CJ
(1985) Phys Chem Miner 12:34
Fig. 7 Effect of methanol/PO molar ratio on PO conversion and PM
selectivity
system were optimized to maximize the PO conversion
(93.2%) and PM selectivity (97.4%) and almost all the PM
in the reaction products was PPM. Optimum reaction
conditions were obtained with methanol/PO molar ratio of
4.0, 0.9% catalyst (wt% of total reactants) at 140 °C for
6 h. The catalyst brought advantages such as high catalytic
activity, easy separation by simple filtration, possible cat-
alyst recycling and use of non-toxic and inexpensive cat-
alyst. It is probable that the solid base catalyst becomes a
practical alternative to soluble bases.
21. Adachi-Pagano M, Forano C, Besse J-P (2003) J Mater Chem
13:1988
22. Rao MM, Reddy BR, Jayalakshmi M, Jaya VS, Sridhar B (2005)
Mater Res Bull 40:347
23. Lei XD, Zhang FZ, Yang L, Guo XX, Tian YY, Fu SS, Li F,
Evans DG, Duan X (2007) AIChE J 53:932
24. Xie WL, Peng H, Chen LG (2006) J Mol Catal A 246:24
25. Prinetto F, Tichit D, Teissier R, Coq B (2000) Catal Today
55:103
123