Ammonia promoted barium sulfate catalyst for dehydration of lactic acid to acrylic acid

Article abstract of DOI:10.1039/c7ra10736f

The dehydration of lactic acid (LA) to acrylic acid over ammonia promoted barium sulfate was studied under various conditions. Interplanar spacing (d) calculated from the enlarged (121) diffraction peak of XRD patterns with the Bragg equation is influenced by preparation conditions, which determines the acid-base properties of the prepared barium sulfate. The present work focused on the preparation conditions such as alkaline agents, pH values and calcination temperatures, which affected the d value. It was found that aqueous ammonia was used as an alkaline agent at pH = 5 to synthesize barium sulfate with an appropriate d value, which displayed an excellent catalytic performance for LA dehydration to acrylic acid. In the presence of the prepared barium sulfate with an appropriate d value, the dehydration reaction of lactic acid proceeded efficiently, with 100% lactic acid conversion and ～82% acrylic acid selectivity. The unprecedented catalytic performance is due to a balance between acidic sites and basic sites existing on the surface of the prepared catalyst. The catalyst is very stable for at least 24 h. The deactivation catalyst can be easily regenerated as it is calcined at 500 °C for 10 h under the air atmosphere.

Full text of DOI:10.1039/c7ra10736f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Ammonia promoted barium sulfate catalyst for
dehydration of lactic acid to acrylic acid†
Cite this: RSC Adv., 2017, 7, 54696
Xinli Li, Zhi Chen, Ping Cao, Wenjie Pu, Weixin Zou, Congming Tang *a
a
a
a
b
c
c
and Lin Dong
The dehydration of lactic acid (LA) to acrylic acid over ammonia promoted barium sulfate was studied under
various conditions. Interplanar spacing (d) calculated from the enlarged (121) diffraction peak of XRD
patterns with the Bragg equation is influenced by preparation conditions, which determines the acid–
base properties of the prepared barium sulfate. The present work focused on the preparation conditions
such as alkaline agents, pH values and calcination temperatures, which affected the d value. It was found
that aqueous ammonia was used as an alkaline agent at pH ¼ 5 to synthesize barium sulfate with an
appropriate d value, which displayed an excellent catalytic performance for LA dehydration to acrylic
acid. In the presence of the prepared barium sulfate with an appropriate d value, the dehydration
reaction of lactic acid proceeded efficiently, with 100% lactic acid conversion and ꢀ82% acrylic acid
selectivity. The unprecedented catalytic performance is due to a balance between acidic sites and basic
sites existing on the surface of the prepared catalyst. The catalyst is very stable for at least 24 h. The
Received 28th September 2017
Accepted 24th November 2017
DOI: 10.1039/c7ra10736f
ꢁ
rsc.li/rsc-advances
deactivation catalyst can be easily regenerated as it is calcined at 500 C for 10 h under the air atmosphere.
Research on AA obtained from LA via dehydration has
made much progress in recent years. Hydroxyapatites offered
Introduction
Lactic acid (LA) derived from biomass attracts the interest of an excellent activity for the dehydration of LA to AA, and were
3
5–37
academia and industrial areas, since it is accessible, low cost intensively investigated by several groups.
These studies
focused on regulating the acid–base properties of catalysts by
can controlling the preparation conditions of catalysts, and
1
–5
6
7,8
9
and rich. For example, starch, cellulose, hemicellulose,
10
11
12
13
7,14–16
lignin, wheat straw, glucose, xylose and glycerol
be efficiently converted to LA via microbiology fermentation and disclosing the relation between acid–base properties and
catalytic conversion. LA can be used as a platform molecule to activity. Modied zeolites with alkali metals were also inves-
produce a variety of chemicals such as acrylic acid (AA), 2,3- tigated, and viewed as a class of catalytic materials with a high
1
8,22,38–42
pentanedione, acetaldehyde, propionic acid, pyruvic acid and efficiency.
Besides, oxysalt materials of alkali earth
4
3
44
poly-lactic acid under the given conditions including appro- metals and rare earth metal such as BaSO
4
,
Ba
(ref. 46) were
2,3- also used to catalyze the dehydration of LA, and offered an
2 2 7
P O ,
17–20
priate catalysts and reaction conditions.
can be obtained through the dehydration reaction of LA,
pentanedione obtained through the intermolecular condensa- excellent performance. Very recently, the work has investi-
2 2 7 3 4 4
For example, AA Sr P O modied with H PO (ref. 45) and LaPO
21,22
4
7
23–26
tion reaction of two molecules of LA,
through a decarbonylation/decarboxylation reaction of LA,
and propionic acid obtained through the deoxygenation reac- BaSO
acetaldehyde obtained gated the formation of BaSO
4
crystal defects by means of
ultrasound treatment and measured the acidity density of
crystals with different defects, which related to catalytic
27–30
4
31,32
tion of LA.
Among these chemicals, AA has the biggest activity for the dehydration of LA to AA. However, the research
6
market demand (around 5.1 ꢂ 10 t per year), and is widely used on constructing microstructure (especially in crystal defect) of
33,34
to produce acrylates, dispersants, polymers, and so on.
catalyst by synthesis conditions and the relation between
microstructure of catalyst and activity of catalytic dehydration
of LA, is very rare.
Accordingly, we tried to construct the crystal defect of BaSO4
by controlling aqueous solution acid–base property and calci-
nation temperature in the present work, and measured the
a
School of Chemistry and Chemical Engineering, Chongqing University of Technology,
Chongqing 40054, PR China. E-mail: tcmtang2001@163.com
b
Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, China
3 2
acidity density and basicity density by NH -TPD and CO -TPD,
Jiangsu Key Laboratory of Vehicle Emissions Control, Center of Modern Analysis, respectively. Based on these characterizations, we further dis-
West Normal University, Nanchong, Sichuan, 637002, PR China
c
Nanjing University, Nanjing 210093, PR China
closed the relationship between acid–base property of catalyst
induced by crystal defects and activity.
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c7ra10736f
1
54696 | RSC Adv., 2017, 7, 54696–54705
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
preparation conditions had an impact on ne structure of
Results and discussion
Characterization
BaSO . For example, as for alkaline reagents, the peak (121) of
4
the sample treated with aqueous ammonia moved toward le in
XRD, FT-IR and EDS. Fig. 1 showed the XRD patterns of the biggest distance, while it remained at the standard position
ꢁ
samples prepared with different methods including alkaline (2q ¼ 28.746 ) as the sample was treated with ethylenediamine
reagents, pH values and calcination temperatures. According to (Fig. 2A). As for pH values, it was found that all samples dis-
47
recent report, the preparation methods involved in solvents played an right shi for (121) peak except for sample treated
and ultrasonic treatments in synthesis of BaSO crystals can with pH ¼ 5. For calcination temperatures, the peak (121)
4
efficiently produce crystal defects, generating some novel moved toward right gradually with an increase of calcination
47
properties for improving catalytic performances. Based on the temperatures. According to previous work, the le or right
enlightenment of this work, it is expected that varying prepa- shis of the diffraction peak was attributed to the defect-
ration parameters such as alkaline reagents, pH values and induced lattice expansion or contraction, respectively. The
calcination temperatures can achieve a controllable synthesis of interplanar spacing was calculated from Bragg equation, and
BaSO
patterns of the prepared samples, we can clearly observe the point defects, cations escape from their normal sites in ionic
4
crystals with some defects. According to normal XRD the results were listed in Tables 1–3. According to the theory of
48,49
characteristic diffraction peaks, which matched well with the crystal, become interstitial atoms and produce vacancies.
standard BaSO sample (JCPDS 24-1035). In other words, BaSO
Vacancies and interstitial atoms always form in pairs, and
4
4
samples can be easily synthesized at different preparation provide sites with unbalanced electrons. The escaped cations
conditions. However, the enlarged (121) diffraction peak for are electropositive and act as Lewis acid, and the electronegative
these samples differed each other (see Fig. 2), suggesting that vacancies can donate electrons and act as basic sites. As for
Fig. 1 XRD patterns of samples (A: treated with different alkaline reagents at pH ¼ 5; B: treated with aqueous ammonia at different pH values; and
C: treated with aqueous ammonia and pH ¼ 5 at different calcination temperatures).
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 54696–54705 | 54697

View Article Online
RSC Advances
Paper
Fig. 2 The enlarged (121) diffraction peak of samples (A: treated with different alkaline reagents at pH ¼ 5; B: treated with aqueous ammonia at
different pH values; and C: treated with aqueous ammonia and pH ¼ 5 at different calcination temperatures).
dehydration reaction of LA to AA, it has been conrmed that addition, EDS measurement was also carried out, and showed
acid–base properties of catalysts play a crucial role during the in Fig. S1 and Table S1.† Ba, S and O elements were clearly
17,36,37,50
catalytic process.
observed in all samples, while other elements such as N were
Fig. 3 showed the FT-IR spectra of the prepared samples with not observed, indicating a complete removal of impurity from
different methods including alkaline reagents, pH values and samples.
calcination temperatures. The characteristic absorption bands
Textural and acid–base properties of samples. Textural and
ꢃ1
ꢃ1
ꢃ1
occurred at 1077 cm , 1116 cm , 1208 cm , which are acid–base properties of samples were measured and the results
ꢃ1
ascribed to S]O stretching vibration; occurred at 610 cm and were shown in Tables 1–3 and Fig. S2–S4.† From the results
ꢃ1
6
46 cm , which are ascribed to O]S–O bending mode. In shown in Table 1, the specic surface area of samples treated
Table 1 Textural and acid–base properties of barium sulfate treated with different alkaline reagents at pH ¼ 5
Surface acidity
Surface basicity
2
ꢃ1
3
ꢃ1
a
b
ꢃ1
ꢃ1
Alkaline reagent
S
BET (m g
)
Vol (cm g
)
Pore size (nm)
d (nm)
(mmol g
)
(mmol g
)
ꢃ1
ꢃ1
ꢃ1
ꢃ1
—
9.8
8.2
7.3
7.4
1.1 ꢂ 10
1.1 ꢂ 10
1.4 ꢂ 10
1.2 ꢂ 10
3.0
3.4
3.4
0.3104
0.3107
0.3103
0.3105
18.9
15.1
23.3
15.2
32.4
51.2
94.4
90.1
Aqueous ammonia
Ethylenediamine
n-Butylamine
30.4
a
b
Calculated from desorption branch data on the Barrett–Joyner–Halenda (BJH) model. Interplanar spacing, determined by enlarged (121)
diffraction peak of XRD patterns with Bragg equation.
54698 | RSC Adv., 2017, 7, 54696–54705
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
Table 2 Textural and acid–base properties of barium sulfate treated with aqueous ammonia at different pH values
Surface acidity
Surface basicity
2
ꢃ1
3
ꢃ1
a
b
ꢃ1
ꢃ1
pH value
S
BET (m g
)
Vol (cm g
)
Pore size (nm)
d (nm)
(mmol g
)
(mmol g
)
ꢃ1
ꢃ1
ꢃ1
ꢃ1
ꢃ1
—
3
5
7
9
9.8
9.5
8.2
9.0
8.9
1.1 ꢂ 10
1.0 ꢂ 10
1.1 ꢂ 10
1.1 ꢂ 10
1.4 ꢂ 10
3.0
4.3
3.4
3.4
7.4
0.3104
0.3094
0.3107
0.3092
0.3098
18.9
18.8
15.1
9.2
32.4
56.3
51.2
96.3
57.3
13.0
a
b
Calculated from desorption branch data on the Barrett–Joyner–Halenda (BJH) model. Interplanar spacing, determined by enlarged (121)
diffraction peak of XRD patterns with Bragg equation.
with alkaline reagents during the aging process reduced in decreased with an increase of calcination temperature. Interest-
comparison with the sample treated without alkaline reagent. ingly, the acidity density of samples also decreased with an
Organic bases such as ethylenediamine and n-butylamine have increase of calcination temperature while their corresponding
ꢃ1
ꢃ1
a greater effect than inorganic base (aqueous ammonia) for basicity density varied between 28.3 mmol g and 65.4 mmol g
.
reducing the specic surface area of sample. Interplanar
spacing determined by enlarged (121) diffraction peak of XRD
patterns with Bragg equation varied from 0.3103 to 0.3107 nm,
Activity
indicating that alkaline reagents can efficiently control the
formation of crystal defect. Among these alkaline reagents,
Preparation conditions of catalyst
Alkaline reagents vs. catalytic performance. Effect of alkaline
aqueous ammonia displayed the highest defect of crystal (cor- reagents on catalytic performance of the dehydration reaction
responding d ¼ 0.3107 nm). Compared with the sample treated of LA was investigated, and the results were shown in Table 4.
without alkaline reagent, all samples treated with alkaline Initially, we measured the activity of the catalyst treated without
reagents decreased in acidity except for ethylenediamine, while alkaline reagents, and the performance including LA conver-
they all increased in basicity.
sion ¼ 90.9% and AA selectivity ¼ 59.7% was acceptable. It was
In the selected alkaline reagent of aqueous ammonia, effect expected that further improvement of catalytic performance was
of pH values on textural and acid–base properties of samples achieved by controlling the acid–base environment in aging
4
was shown in Table 2. Initially, the specic surface area of process of BaSO crystal. It was clearly seen that all the used
samples decreased with an increase of pH values, while it alkaline reagents displayed a positive effect for improving the
increased slightly with further increase of pH values. The lowest catalytic performance except for ethylenediamine. Especially,
2
ꢃ1
specic surface area of sample was achieved in 8.2 m g as the the aqueous ammonia offered the most excellent catalytic
pH value was adjusted to 5. It was noted that interplanar performance (LA conversion ¼ 100% and AA selectivity ¼
spacing determined by enlarged (121) diffraction peak of XRD 81.2%). The area-specic catalytic rates were also measured,
patterns with Bragg equation increased compared to standard and shown in Table 4. LA consumption rate increased with
value as the pH value was adjusted to 5, and others decreased. using alkaline reagents, and AA formation rate occurred in
The acidity density of samples decreased with an increase of pH increase or decrease, indicating that parallel side reactions
values except for pH ¼ 9, and their corresponding basicity existed. According to further observation, the selectivities to
ꢃ1
ꢃ1
density varied between 32.4 mmol g and 96.3 mmol g
.
side reactions such as formation of propionic acid and acetic
Under the xed pH ¼ 5 with aqueous ammonia, effect of acid became higher, resulting in decrease of AA selectivity.
calcination temperature on textural and acid–base properties of
samples was depicted in Table 3. The specic surface area, pore CO
Combined with characterizations by XRD, NH
-TPD, catalytic performance was found to relate to acid–
crystals induced by microstructures.
diffraction peak of XRD patterns with Bragg equation of samples According to the measurements of interplanar spacing and
acid–base properties, defects of BaSO crystals were produced
3
-TPD and
2
volume and interplanar spacing determined by enlarged (121) base properties of BaSO
4
4
Table 3 Textural and acid–base properties of barium sulfate treated with aqueous ammonia and pH ¼ 5 at different calcination temperatures
Surface acidity
Surface basicity
ꢁ
2
ꢃ1
3
ꢃ1
a
b
ꢃ1
ꢃ1
Calc. temp. ( C)
S
BET (m g
)
Vol (cm g
)
Pore size (nm)
d (nm)
(mmol g
)
(mmol g
)
ꢃ1
ꢃ1
ꢃ2
ꢃ2
—
8.2
8.6
4.1
3.7
1.1 ꢂ 10
1.4 ꢂ 10
9.5 ꢂ 10
2.6 ꢂ 10
3.4
8.3
6.9
3.4
0.3107
0.3102
0.3093
0.3096
15.1
14.0
8.4
51.2
65.4
28.3
56.1
5
6
7
00
00
00
3.7
a
b
Calculated from desorption branch data on the Barrett–Joyner–Halenda (BJH) model. Interplanar spacing, determined by enlarged (121)
diffraction peak of XRD patterns with Bragg equation.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 54696–54705 | 54699

View Article Online
RSC Advances
Paper
Fig. 3 FT-IR spectra of samples (A: treated with different alkaline reagents at pH ¼ 5; B: treated with aqueous ammonia at different pH values;
and C: treated with aqueous ammonia and pH ¼ 5 at different calcination temperatures).
by deviation of interplanar spacing to normal value, and caused in which crystal formed and grew, and we observed the forma-
different acid–base properties. Pioneering work has demon- tion of crystal defects. Due to the crystal defects, the BaSO
strated that acid–base properties of catalysts are crucial to crystals treated with different alkaline reagents offered different
4
17,22,36,37
catalytic dehydration of LA to AA.
Recently, the work has acid–base properties. From the results shown in Table 4, cata-
disclosed that both ultrasound treatment and aging time can lytic performance decreased with an increase of acidity density;
efficiently control the formation of crystal defects, which result and it was also affected by basicity, and medium basicity density
47
in different acidity. In the present work, alkaline reagents were favored the catalytic performance for dehydration of LA to AA.
used to control the acid–base environment of aqueous solution For example, ethylenediamine treated sample displayed the
a
Table 4 Effect of alkaline reagents on catalytic performance of the dehydration reaction of LA
Area-specic catalytic rate
b
ꢃ1
ꢃ2
Sel. [%]
(mmol h
m )
LA conv.
[%]
Carbon balance
[%]
Alkaline reagents
AA
AD
PA
AC
PD
LA consumption
AA formation
—
90.9
100
89.1
99.5
59.7
81.2
48.6
69.1
16.7
12.1
21.1
20.9
17.1
4.5
23.1
7.1
1.8
1.3
4.8
1.6
1.4
0.3
1.0
0.6
93.6
99.6
89.3
99.7
601
759
678
709
359
616
330
490
Aqueous ammonia
Ethylenediamine
n-Butylamine
a
ꢃ1
ꢃ1
Catalyst, 0.38 mL, 0.48–0.59 g, uncalcination, particle size: 20–40 meshes, carrier gas N
2
: 1.2 mL min , feed ow rate: 1.4 mL h , LA feedstock:
0 wt% in water, reaction temperature, 400 C, TOS: 2–4 h. LA: lactic acid, AA: acrylic acid, AD: acetaldehyde, PA: propionic acid, AC: acetic acid,
PD: 2,3-pentanedione.
ꢁ
b
2
54700 | RSC Adv., 2017, 7, 54696–54705
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
a
Table 5 Effect of pH values modulated with aqueous ammonia on catalytic performance of the dehydration reaction of LA
Area-specic catalytic rate
b
ꢃ1
ꢃ2
Sel. [%]
(mmol h
m )
LA conv.
[%]
pH
AA
AD
PA
AC
PD
Carbon balance
LA consumption
AA formation
—
3
5
7
9
90.9
82.5
100
92.5
97.5
59.7
60.4
81.2
67.8
69.7
16.7
22.1
12.1
22.0
17.5
17.1
13.1
4.5
7.5
9.4
1.8
2.5
1.3
1.5
1.6
1.4
0.5
0.3
0.9
1.2
93.6
82.7
99.6
84.9
100
601
491
759
615
668
359
297
616
417
466
a
ꢃ1
ꢃ1
Catalyst, 0.38 mL, 0.48–0.55 g, uncalcination, particle size: 20–40 meshes, carrier gas N
2
: 1.2 mL min , feed ow rate: 1.4 mL h , LA feedstock:
ꢁ
b
2
0 wt% in water, reaction temperature, 400 C, TOS: 2–4 h. LA: lactic acid, AA: acrylic acid, AD: acetaldehyde, PA: propionic acid, AC: acetic acid,
PD: 2,3-pentanedione.
poorest activity (LA conversion ¼ 89.1% and AA selectivity ¼ catalyst surface. Compared with two catalysts treated with pH ¼
ꢃ1
4
8.6%) due to the highest acidity density (23.3 mmol g ). Under 3 and 9, respectively, they have similar basic site densities (56.3
ꢃ1
ꢃ1
the identical acidity conditions, aqueous ammonia treated mmol g and 57.3 mmol g ), and different acidic site densities
ꢃ1
ꢃ1
ꢃ1
sample has lower basicity density (51.2 mmol g ) than n- (18.8 mmol g and 13.0 mmol g ). Whereas, the latter had
ꢃ1
butylamine (90.1 mmol g ), while the former offered higher better catalytic performance (LA conversion ¼ 97.5% and AA
selectivity of AA (81.2%) than the latter (69.1%) although LA selectivity ¼ 69.7%) than the former (LA conversion ¼ 82.5%
conversion was close to each other.
and AA selectivity ¼ 60.4%), indicating that appropriately
pH values vs. catalytic performance. In the screening of alka- lowering acidic site density favored the catalytic dehydration
line reagents, aqueous ammonia was the most excellent alka- reaction of LA to AA.
line reagent for improving the catalytic performance. The pH
Calcination temperature vs. catalytic performance. Table 6
values were adjusted using aqueous ammonia to observe effect showed effect of calcination temperature on the performance of
of pH values on catalytic performance, and the results were catalysts. LA conversion decreased from 100% to 50.1% as the
shown in Table 5. From the results shown in Table 5, the calcination temperature increased from uncalcination to
ꢁ
ꢁ
selectivity to AA on the catalysts treated with aqueous ammonia 600 C. With further increase of calcination temperature to
in the range of pH ¼ 3–5 has been improved in comparison with 700 C, LA conversion began to increase to 75.9%. Besides, AA
the catalyst treated without any alkaline reagent. Conspicu- selectivity also reduced with an increase of calcination
ously, the selectivity to AA achieved as high as 81.2% when pH temperature. Although area-specic catalytic rate (LA
ꢃ1
ꢃ2
value was adjusted to 5 using aqueous ammonia. It was noted consumption: 1227 mmol h
m
and AA formation: 724 mmol
ꢃ1
ꢃ2
ꢁ
that the characteristic diffraction peak of crystal face (121)
h
m
) on the catalyst calcined at 700 C was the highest
moved toward le direction from the standard position (2q ¼ among all the catalysts calcined at different temperatures (ca.
ꢁ
ꢁ
2
8.746 ) with aqueous ammonia adjusting to pH ¼ 5, while ꢀ700 C), LA conversion as well as AA selectivity is relatively
2
ꢃ1
others with pH ¼ 3, 7 and 9 moved toward right direction. Lyu lower due to its low specic surface area (only 3.7 m g far
2
ꢃ1
and Wang also observed the characteristic diffraction peak lower than others (ꢀ8 m g )). In order to better understand
crystal face (121) moved toward right direction as the BaSO the effect of calcination temperature on the catalytic perfor-
4
47
crystals were treated with ultrasound means. Whether the mance, we can draw support from the XRD characterization
characteristic diffraction peak of crystal face (121) moved shown in Fig. 2C and the acid–base site density shown in Table
toward le or right direction, these BaSO crystals produced 3. With an increase of calcination temperature, characteristic
4
defects, leading to different acid–base properties located on the diffraction peak of crystal face (121) also gradually moved
a
Table 6 Effect of calcination temperature on catalytic performance of the dehydration reaction of LA
Area-specic catalytic rate
b
ꢃ1
ꢃ2
Sel. [%]
(mmol h
m )
Calc. temp.
LA conv.
[%]
ꢁ
[
C]
AA
AD
PA
4.5
6.8
20.7
10.1
AC
PD
Carbon balance
LA consumption
AA formation
—
100
100
50.1
75.9
81.2
70.8
56.3
59.0
12.1
20.1
17.8
26.1
1.3
1.5
4.0
1.8
0.3
0.50
0.9
2.2
99.6
99.9
85.1
82.3
759
804
667
1227
616
569
375
724
5
6
7
00
00
00
a
ꢃ1
ꢃ1
Catalyst, 0.38 mL, 0.45–0.57 g, particle size: 20–40 meshes, carrier gas N
2
: 1.2 mL min , feed ow rate: 1.4 mL h , LA feedstock: 20 wt% in water,
ꢁ
b
reaction temperature, 400 C, TOS: 2–4 h. LA: lactic acid, AA: acrylic acid, AD: acetaldehyde, PA: propionic acid, AC: acetic acid, PD: 2,3-
pentanedione.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 54696–54705 | 54701

View Article Online
RSC Advances
Paper
a
Table 7 Effect of reaction temperature on performance of the dehydration reaction of LA
Area-specic catalytic rate
b
ꢃ1
ꢃ2
Sel. [%]
(mmol h
m )
Reaction temp.
LA conv.
[%]
ꢁ
[
C]
AA
AD
9.2
10.9
12.1
14.2
19.7
PA
4.5
4.8
4.5
AC
PD
Carbon balance
LA consumption
AA formation
3
3
4
4
4
50
70
00
20
40
69.7
98.5
100
100
100
82.1
81.7
81.2
68.8
53.9
0.9
1.1
1.3
3.1
4.1
0.2
0.2
0.3
1.1
1.7
95.5
99.8
99.6
96.3
91.5
499
705
759
774
759
410
576
616
533
409
10.1
13.2
a
ꢃ1
ꢃ1
Catalyst, 0.38 mL, 0.49–0.53 g, uncalcination, particle size: 20–40 meshes, carrier gas N
2
: 1.2 mL min , feed ow rate: 1.4 mL h , LA feedstock:
b
2
0 wt% in water, TOS: 2–4 h. LA: lactic acid, AA: acrylic acid, AD: acetaldehyde, PA: propionic acid, AC: acetic acid, PD: 2,3-pentanedione.
toward right direction, suggesting that high temperature catalytic rate also increased with an increase of reaction
favored the formation of crystal defects. As a result, different temperature. For this reason, LA conversion increased with an
acid–base properties of catalysts were achieved by calcination.
increase of reaction temperature. However, as the reaction
ꢁ
Furthermore, we correlated the catalytic performance of temperature was above 400 C, LA conversion remained
catalysts with acid–base properties of catalysts. Firstly, we constant (100%). As for AA selectivity, it gradually decreased
compared the catalyst without calcination with the catalyst with an increase of reaction temperature while byproduct
ꢁ
calcined at 700 C, and found that they had similar basic site selectivity increased, indicating that the side reactions were
ꢃ1
ꢃ1
density (51.2 mmol g and 56.1 mmol g ), and different acidic accelerated at higher reaction temperature. For example, pro-
ꢃ1
ꢃ1
site density (15.1 mmol g vs. 3.7 mmol g ). The catalytic pionic acid selectivity and acetic acid selectivity achieved 13.2%
ꢁ
performance of the former (LA conversion ¼ 100% and AA and 4.1%, respectively, far higher than those obtained at 350 C
selectivity ¼ 82.1%) was obviously more excellent than the latter (4.5% and 0.9%).
(
LA conversion ¼ 75.9% and AA selectivity ¼ 59.0%), suggesting
Catalyst colour became darker with an increase of reaction
ꢁ
that too low acidic site density was unfavourable for the dehy- temperature, and displayed a black colour especially at 440 C
dration of LA to AA. The poorest performance (LA conversion ¼ (shown in Fig. S5†), and these used catalysts were calcined
ꢁ
5
0.1% and AA selectivity ¼ 56.3%) was observed over the cata- under the air atmosphere at 500 C for 10 h to obtain light
ꢁ
lyst calcined at 600 C, indicating that too low basic site density colour again, suggesting that the substance formed on the
had also a disadvantage for the dehydration of LA to AA.
Reaction conditions
catalyst surface may be carbon or coke, which covered the active
sites, leading to decrease of AA selectivity.
Reaction temperature vs. reaction performance. According to
Weight hourly space velocity (WHSV) vs. reaction performance.
the discussion on preparation conditions of catalysts in the Weight hourly space velocity (WHSV) is generally used to eval-
52,53
preceding sections, we obtained the optimal preparation uate the performance of heterogeneous catalyst.
Table 8
conditions for BaSO
4
with crystal defects (alkaline reagent: showed the inuence of LA WHSV on reaction performance.
ꢁ
aqueous ammonia, pH ¼ 5.). Next, effect of reaction conditions The reaction was conducted at 400 C with LA ow rate changed
ꢃ1
ꢃ1
on the dehydration of LA would be investigated. As an impor- from 1.0 to 8.4 mL h (corresponding WHSV ¼ 2.0–16.8 h ).
tant factor determining the reaction rate and reaction selec- For LA conversion, it almost remained at 100% when LA WHSV
51
ꢃ1
ꢃ1
tivity, reaction temperature was rstly discussed. From effect was lower than 2.8 h . As LA WHSV increased from 2.8 h to
of reaction temperature on dehydration of LA shown in Table 7, 16.8 h , LA conversion gradually decreased from 100% to
ꢃ1
LA conversion increased with an increase of reaction tempera- 70.5%. AA selectivity, unlike LA conversion, increased fast from
ꢁ
ꢁ
ꢃ1
ture from 350 C to 400 C. For a given catalyst, area-specic 71.3% to 81.2% with an increase of LA WHSV from 2.0 h to 2.8
a
Table 8 Effect of weight hourly space velocity (WHSV) on performance of the dehydration reaction of LA
Area-specic catalytic rate
b
ꢃ1
ꢃ2
Sel. [%]
(mmol h
m )
LA conv.
[%]
ꢃ1
WHSV (h
)
AA
AD
PA
AC
PD
Carbon balance
LA consumption
AA formation
2
2
5
8
1
.0
.8
.6
.4
100
100
92.2
80.5
70.5
71.3
81.2
78.2
77.8
77.7
15.1
12.1
11.7
12.5
12.6
8.6
4.5
6.1
7.1
7.2
1.9
1.3
1.4
1.4
1.6
0.5
0.3
0.4
0.4
0.3
98.5
99.6
90.2
97.1
97.3
542
759
1399
1833
3210
386
616
1094
1426
2494
6.8
a
ꢃ1
2
Catalyst, 0.38 mL, 0.5 g, uncalcination, particle size: 20–40 meshes, carrier gas N : 1.2 mL min , LA feedstock: 20 wt% in water, reaction
ꢁ
b
temperature, 400 C, TOS: 2–4 h. LA: lactic acid, AA: acrylic acid, AD: acetaldehyde, PA: propionic acid, AC: acetic acid, PD: 2,3-pentanedione.
54702 | RSC Adv., 2017, 7, 54696–54705
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
Fig. 4, LA conversion slightly decreased within initial 24 h on
stream while aer experienced 24 h on stream, LA conversion
drastically reduced to near 85%. Similarly, AA selectivity was
inuenced by the time on stream. For example, AA selectivity also
slightly decreased from about 81.2% to 76.5% within the initial
reaction time (1–24 h). Catalytic performance reduced with time
on stream, indicating a deactivation of catalyst. However, the
deactivated catalyst can be easily regenerated aer a simple
ꢁ
calcination at 500 C for 10 h under the air atmosphere because
the resultants on the catalyst surface causing in deactivation are
carbon or coke (also shown in Fig. 4, cycle 2). In addition, XRD,
FT-IR and TG characterizations of the spent catalyst and the fresh
catalyst was performed, and the results were shown in Fig. 5, S6
and S7.† It was clearly seen that the structure of BaSO crystal
4
remained well during the catalytic reaction.
Fig. 4 The stability of catalyst with time on stream. Catalyst, 0.38 mL,
Experimental
Materials
0
2
.5 g, uncalcination, particle size: 20–40 meshes, carrier gas N : 1.2
ꢃ1
ꢃ1
mL min , feed flow rate: 1.4 mL h , LA feedstock: 20 wt% in water,
reaction temperature, 400 C. The used catalyst in cycle 1 was
ꢁ
ꢁ
calcined at 500 C for 10 h to remove the carbon deposits from its Lactic acid (85–90 wt%), was purchased from Chengdu Kelong
surface, and then was used again in cycle 2.
Chemical Reagent Co. and was used for the dehydration reaction
of lactic acid to synthesize AA without further purication. Sul-
phuric acid (H
ammonia, n-butylamine, and ethylenediamine were purchased
2 4 2
SO ), barium hydroxide (Ba(OH) ), aqueous
ꢃ1
h
. It is known that LA contact time on the surface of BaSO
4
28
catalyst shortens with the enhancement of LA WHSV. The from Energy Chemical Company. Acrylic acid, propionic acid,
appropriately shortened contact time favored the selective acetic acid, acetaldehyde, 2,3-pentanedione and n-butanol,
formation of AA, indicating that dehydration reaction of LA is together with hydroquinone were obtained from Sigma-Aldrich.
relatively faster than other side reactions. Area-specic catalytic Acrylic acid, propionic acid, acetic acid, 2,3-pentanedione and
rate quickly increased with an increase of LA solution WHSV. acetaldehyde were used for gas chromatograph reference mate-
ꢃ1
ꢃ2
For example, LA consumption rate was 542 mmol h
m
at LA rials, and n-butanol was adopted as internal standard material.
ꢃ1
ꢃ1
solution WHSV ¼ 2.0 h while it attained to 3210 mmol h
Hydroquinone (0.3 wt%) was used as a polymerization inhibitor.
ꢃ2
ꢃ1
m
at LA solution WHSV ¼ 16.8 h . For that reason, 70.5% of
ꢃ1
LA conversion was obtained at high WHSV of 16.8 h
.
Preparation of catalysts
2
BaSO catalyst (blank): rstly, 25.20 g of Ba(OH) was fully dis-
Stability of catalyst. Fig. 4 showed the stability of the catalyst
with time on stream. The experiment was performed at 400 C
with LA concentration of 20 wt% and feed ow rate of 1.4 mL
. It is known that long-term stability is a very important
characteristic for a heterogeneous catalyst. From the cycle 1 of
ꢁ
4
solved in 150 mL distilled water. Under stirring at room
ꢃ1
temperature, diluted H
2
SO
4
solution with 0.2 mol L
was
ꢃ1
h
dropwisely added to the obtained Ba(OH)
2
aqueous solution to
54
form a white precipitate of barium sulfate at pH ¼ 2, and aged
for 15 h. Subsequently, the white precipitate was rinsed at least
three times to remove dissociative H SO using distilled water
2
4
ꢁ
and dried at 120 C in the air circulating oven for 3 h.
BaSO catalyst treated with alkaline agents: the obtained
BaSO precipitates with above preparation method were aged
for 15 h at pH ¼ 5 using different alkaline agents. Specially,
using aqueous ammonia to adjust pH values, BaSO precipi-
4
4
4
tates were aged at pH ¼ 3, 7, and 9 for 15 h, respectively.
Catalyst characterization
Powder X-ray diffraction measurement was conducted on
a Dmax/Ultima IV diffractometer operated at 40 kV and 20 mA
with Cu-Ka radiation. The FTIR spectra of the catalysts were
ꢃ1
recorded in the range of 500–4000 cm on a Nicolet 6700 spec-
trometer. The EDS measurements of the catalysts were performed
using a scanning electron microscope (SEM, JSM-6510). The
specic surface areas of catalysts were measured through
Fig. 5 XRD patterns of fresh BaSO
4
catalyst and spent BaSO
4
catalysts. nitrogen adsorption at 77 K using Autosorb IQ instrument. Prior
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 54696–54705 | 54703

View Article Online
RSC Advances
Paper
ꢁ
to adsorption, the samples were treated at 250 C under vacuum
Conclusions
for 6 h and the specic surface area was calculated according to
the Brunauer–Emmett–Teller (BET) method. Surface acid and BaSO crystals with different defects were prepared by control-
4
base properties of the samples were estimated by NH -TPD and ling the preparation conditions such as alkaline reagents, pH
3
CO
2
-TPD, respectively, on a Quantachrome Instrument. The values and calcination temperature. It was found that these
ꢃ1
sample (ca. 50–60 mg) was purged with dry Ar (50 mL min
purity >99.999 vol%) at 400 C for 1.0 h, followed by reducing the Among the tested alkaline reagents, aqueous ammonia dis-
,
4
BaSO crystals with different defects offered different activities.
ꢁ
furnace temperature to room temperature, and switching to played an evidently promotional effect for dehydration of LA to
a ow of 8 vol% NH
3
/Ar or 10 vol% CO
2
/Ar for 1 h to execute NH
3
AA since the BaSO crystal had an appropriate acid–base prop-
4
or CO adsorption. Then, NH or CO adsorbed on the sample is erty caused by crystal defect. Besides, pH values and calcination
2
3
2
ꢁ
ꢁ
ꢃ1
desorbed in the range of 80–800 C at a rate of 10 C min
.
temperature can also control the formation of crystal defect,
which determined the acid–base property of catalyst. Relatively
less acidity density and enhanced basicity density favored the
dehydration of LA to AA. Under the optimal reaction conditions,
Catalyst evaluation
The vapor-phase dehydration reaction of lactic acid to acrylic
acid over the catalysts was carried out at 350–440 C under
atmospheric pressure in a vertical xed-bed tubular quartz
reactor (40 cm ꢂ 4 mm (i.d.)), mounted in a cylindrical elec-
trical furnace. The catalyst (0.50–0.60 g, 20–40 meshes) was
placed in the middle of the reactor and quartz wool was placed
ꢁ
the stability of BaSO
pH ¼ 5 was evaluated. Encouragingly, LA conversion reduced in
5% (from 100% to 85%) within 28 h on stream.
4
catalyst treated with aqueous ammonia at
1
Conflicts of interest
in both ends. Before catalyst evaluation, the catalyst was pre-
ꢁ
treated at the required reaction temperature (ca. 400 C) for There are no conicts to declare.
ꢃ1
1.0 h under high purity N (0.1 MPa, 1.2 mL min ). The feed-
2
stock (20 wt% solution of lactic acid) was then pumped into the
Acknowledgements
preheating zone (lactic acid aqueous solution ow rate, 1.4 mL
ꢃ1
h
) and driven through the catalyst bed by nitrogen. The liquid
This work was supported by Sichuan Province Science and
Technology Department application foundation project
products were condensed using ice-water bath and analyzed off-
line using a SP-6890 gas chromatograph with a FFAP capillary
column connected to a FID. Quantitative analysis of the prod-
ucts was carried out by the internal standard method using n-
butanol as the internal standard material. GC-MS analyses of
(2017JY0188), and the opening foundation of Jiangsu Key
Laboratory of Vehicle Emissions Control (OVEC 032).
the samples were performed using Agilent 5973N Mass Selective Notes and references
Detector attachment. The conversion of lactic acid and the
selectivity toward acrylic acid or other by-products are calcu-
lated as follows with eqn (1) and (2), respectively.
1
2
3
4
5
P. Maki-Arvela, I. L. Simakova, T. Salmi and D. Y. Murzin,
Chem. Rev., 2014, 114, 1909–1971.
J. C. Serrano-Ruiz, R. Luque and A. Sepulveda-Escribano,
Chem. Soc. Rev., 2011, 40, 5266–5281.
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–
2502.
W. P. Deng, Y. Wang and N. Yan, Current Opinion in Green
and Sustainable Chemistry, 2016, 2, 54–58.
S. Siankevich, S. Mozzettini, F. Bobbink, S. Ding, Z. Fei,
N. Yan and P. J. Dyson, ChemPlusChem, DOI: 10.1002/
cplu.201700344.
G. Juodeikiene, D. Vidmantiene, L. Basinskiene,
D. Cernauskas, E. Bartkiene and D. Cizeikiene, Catal.
Today, 2015, 239, 11–16.
n
0
ꢃ n
1
Conversion=% ¼
ꢂ 100;
(1)
n
0
n
p
Selectivity=% ¼ n
ꢂ 100
(2)
0
ꢃ n
1
where n is the molar quantity of lactic acid fed into reactor, n
0
1
p
is the molar quantity of lactic acid in the effluent, and n is the
molar quantity of lactic acid converted to acrylic acid or other
byproducts such as acetaldehyde, propionic acid, acetic acid,
6
2,3-pentanedione.
Area-specic catalytic rate is dened as previous references
22,28
reported,
and is determined with eqn (3) and (4).
7
8
P.
Wattanapaphawong,
P.
Reubroycharoen
and
LA consumption rate ¼
A. Yamaguchi, RSC Adv., 2017, 7, 18561–18568.
ꢃ1
Y. L. Wang, W. P. Deng, B. J. Wang, Q. H. Zhang, X. Y. Wan,
Z. C. Tang, Y. Wang, C. Zhu, Z. X. Cao, G. C. Wang and
H. L. Wan, Nat. Commun., 2013, 4, 2141.
amount of LA consumed per hour in the reactor ðmmol h Þ
2
surface area of catalyst in the reactor ðm Þ
(
3)
9
C. Gao, C. Q. Ma and P. Xu, Biotechnol. Adv., 2011, 29, 930–
9
39.
AA formation rate ¼
1
1
0 I. Delidovich, P. J. C. Hausoul, L. Deng, R. Pfutzenreuter,
M. Rose and R. Palkovits, Chem. Rev., 2016, 116, 1540–1599.
1 Y. M. Zhang, X. R. Chen, J. Q. Luo, B. K. Qi and Y. H. Wan,
Bioresour. Technol., 2014, 158, 396–399.
ꢃ1
amount of AA formed per hour in the reactor ðmmol h Þ
(4)
2
surface area of catalyst in the reactor ðm Þ
54704 | RSC Adv., 2017, 7, 54696–54705
This journal is © The Royal Society of Chemistry 2017

View Article Online
Paper
RSC Advances
12 X. C. Wang, Y. L. Song, C. P. Huang, F. B. Liang and 34 C. M. Tang, Y. Zeng, X. G. Yang, Y. C. Lei and G. Y. Wang, J.
B. H. Chen, Green Chem., 2014, 16, 4234–4240. Mol. Catal. A: Chem., 2009, 314, 15–20.
13 L. D. Ye, X. D. Zhou, M. S. Bin Hudari, Z. Li and J. C. Wu, 35 B. Yan, L. Z. Tao, Y. Liang and B. Q. Xu, ACS Catal., 2014, 4,
Bioresour. Technol., 2013, 132, 38–44. 1931–1943.
14 C. A. Ramirez-Lopez, J. R. Ochoa-Gomez, M. Fernandez- 36 Y. Matsuura, A. Onda and K. Yanagisawa, Catal. Commun.,
Santos, O. Gomez-Jimenez-Aberasturi, A. Aonso-Vicario and 2014, 48, 5–10.
J. Torrecilla-Soria, Ind. Eng. Chem. Res., 2010, 49, 6270–6278. 37 V. C. Ghantani, S. T. Lomate, M. K. Dongare and
1
1
1
1
1
2
2
2
2
2
2
2
2
5 H. X. Yin, H. B. Yin, A. L. Wang, L. Q. Shen, Y. Liu and
Y. Zheng, J. Nanosci. Nanotechnol., 2017, 17, 1255–1266.
6 G. Y. Yang, Y. H. Ke, H. F. Ren, C. L. Liu, R. Z. Yang and
W. S. Dong, Chem. Eng. J., 2016, 283, 759–767.
7 B. Yan, L. Z. Tao, A. Mahmood, Y. Liang and B. Q. Xu, ACS
Catal., 2017, 7, 538–550.
8 X. H. Zhang, L. Lin, T. Zhang, H. O. Liu and X. F. Zhang,
Chem. Eng. J., 2016, 284, 934–941.
9 C. M. Tang, J. S. Peng, X. L. Li, Z. J. Zhai, W. Bai, N. Jiang,
S. B. Umbarkar, Green Chem., 2013, 15, 1211–1217.
38 G. M. Lari, B. Puertolas, M. S. Frei, C. Mondelli and J. Perez-
Ramirez, ChemCatChem, 2016, 8, 1507–1514.
39 C. Yuan, H. Y. Liu, Z. K. Zhang, H. F. Lu, Q. L. Zhu and
Y. F. Chen, Chin. J. Catal., 2015, 36, 1861–1866.
40 J. F. Zhang, Y. L. Zhao, X. Z. Feng, M. Pan, J. Zhao, W. J. Ji and
C. T. Au, Catal. Sci. Technol., 2014, 4, 1376–1385.
41 H. J. Wang, D. H. Yu, P. Sun, J. Yan, Y. Wang and H. Huang,
Catal. Commun., 2008, 9, 1799–1803.
H. J. Gao and Y. W. Liao, Green Chem., 2015, 17, 1159–1166. 42 P. Sun, D. H. Yu, K. M. Fu, M. Y. Gu, Y. Wang, H. Huang and
0 S. Lomate, T. Bonnotte, S. Paul, F. Dumeignil and
B. Katryniok, J. Mol. Catal. A: Chem., 2013, 377, 123–128.
1 J. F. Zhang, J. P. Lin and P. L. Cen, Can. J. Chem. Eng., 2008,
H. H. Ying, Catal. Commun., 2009, 10, 1345–1349.
43 J. S. Peng, X. L. Li, C. M. Tang and W. Bai, Green Chem., 2014,
16, 108–111.
44 C. M. Tang, J. S. Peng, G. Fan, X. L. Li, X. L. Pu and W. Bai,
Catal. Commun., 2014, 43, 231–234.
45 C. M. Tang, J. S. Peng, X. L. Li, Z. J. Zhai, N. Jiang, W. Bai,
H. J. Gao and Y. W. Liao, RSC Adv., 2014, 4, 28875–28882.
46 Z. Guo, D. S. Theng, K. Y. Tang, L. L. Zhang, L. Huang,
A. Borgna and C. Wang, Phys. Chem. Chem. Phys., 2016, 18,
23746–23754.
86, 1047–1053.
2 J. F. Zhang, Y. L. Zhao, M. Pan, X. Z. Feng, W. J. Ji and
C. T. Au, ACS Catal., 2011, 1, 32–41.
3 G. C. Gunter, D. J. Miller and J. E. Jackson, J. Catal., 1994,
1
48, 252–260.
4 M. S. Tam, J. E. Jackson and D. J. Miller, Ind. Eng. Chem. Res.,
999, 38, 3873–3877.
1
5 X. L. Li, L. W. Sun, W. X. Zou, P. Cao, Z. Chen, C. M. Tang and 47 S. Lyu and T. F. Wang, RSC Adv., 2017, 7, 10278–10286.
L. Dong, ChemCatChem, DOI: 10.1002/cctc.201701332.
6 L. W. Sun, X. L. Li and C. M. Tang, Acta Phys.-Chim. Sin.,
48 A. Linsebigler, G. Q. Lu and J. T. Yates, J. Chem. Phys., 1995,
103, 9438–9443.
49 G. Q. Lu, A. Linsebigler and J. T. Yates, J. Chem. Phys., 1995,
102, 3005–3008.
2016, 32, 2327–2336.
7 M. E. Sad, L. F. Gonz ´a lez Pe n˜ a, C. L. Padr ´o and
C. R. Apestegu ´ı a, Catal. Today, DOI: 10.1016/ 50 Y. Matsuura, A. Onda, S. Ogo and K. Yanagisawa, Catal.
j.cattod.2017.03.024. Today, 2014, 226, 192–197.
8 C. M. Tang, Z. J. Zhai, X. L. Li, L. W. Sun and W. Bai, J. Catal., 51 M. Behrens, F. Studt, I. Kasatkin, S. Kuhl, M. Havecker,
2
2
3
3
3
3
2
015, 329, 206–217.
F. Abild-Pedersen, S. Zander, F. Girgsdies, P. Kurr,
B. L. Kniep, M. Tovar, R. W. Fischer, J. K. Norskov and
R. Schlogl, Science, 2012, 336, 893–897.
9 C. M. Tang, Z. J. Zhai, X. L. Li, L. W. Sun and W. Bai, J. Taiwan
Inst. Chem. Eng., 2016, 58, 97–106.
0 B. Katryniok, S. Paul and F. Dumeignil, Green Chem., 2010, 52 C. Li, B. Wang, Q. Q. Zhu and T. W. Tan, Appl. Catal., A, 2014,
2, 1910–1913. 487, 219–225.
1 T. J. Korstanje, H. Kleijn, J. Jastrzebski and R. Gebbink, 53 B. Wang, C. Li, Q. Q. Zhu and T. W. Tan, RSC Adv., 2014, 4,
Green Chem., 2013, 15, 982–988. 45679–45686.
2 X. L. Li, Z. J. Zhai, C. M. Tang, L. W. Sun, Y. Zhang and 54 M. S. Holm, S. Saravanamurugan and E. Taarning, Science,
1
W. Bai, RSC Adv., 2016, 6, 62252–62262.
2010, 328, 602–605.
3 R. Beerthuis, G. Rothenberg and N. R. Shiju, Green Chem.,
2015, 17, 1341–1361.
This journal is © The Royal Society of Chemistry 2017
RSC Adv., 2017, 7, 54696–54705 | 54705