Synthesis of bio-based methacrylic acid from biomass-derived itaconic acid over barium hexa-aluminate catalyst by selective decarboxylation reaction

Article abstract of DOI:10.1016/j.mcat.2019.110520

An environmentally-benign, efficient and inexpensive high-surface-area barium hexa-aluminate (BaAl₁₂O₁₉, BHA) was developed as a catalyst for the decarboxylation of the biomass-derived itaconic acid (IA) to bio-based methacrylic acid (MAA). A maximal 50% final yield of MAA with a high product selectivity was obtained under relatively mild synthesis reaction conditions (250 °C; 20 bar N₂). The reported selective MAA production was elevated, operating process characteristics were significantly less harsh, and no depleting critical raw materials were utilized when paralleled to the procedures with alkaline mineral bases, noble metal-containing heterogeneous catalysis systems and unrenewable feed resources (e.g. isobutene), applied previously. It was found that the doping of palladium on BHA support (Pd@BHA) did not improve MAA productivity. The effect of the time (25–300 min), temperature (175–275 °C), pressure (10–40 bar), reacting substrate concentration (0.10–0.19 mol L^–1), metallic oxide mass (0.5–3.0 g) and type on IA conversion, MAA content MAA content and rates was determined, examining also recyclability. BHA catalyst was characterized with various structural techniques, such as energy-dispersive X-ray spectroscopy (EDS), X-ray powder diffraction (XRD), CO₂ temperature-programmed desorption (TPD), scanning electron microscopy (SEM) and N₂ physisorption.

Full text of DOI:10.1016/j.mcat.2019.110520

MolecularCatalysis476(2019)110520
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Synthesis of bio-based methacrylic acid from biomass-derived itaconic acid
over barium hexa-aluminate catalyst by selective decarboxylation reaction
Ashish Bohre⁎, Uroš Novak, Miha Grilc, Blaž Likozar⁎
Department of Catalysis and Chemical Reaction Engineering, National Institute of Chemistry, Hajdrihova 19, 1000 Ljubljana, Slovenia
A R T I C L E I N F O
A B S T R A C T
Keywords:
An environmentally-benign, efficient and inexpensive high-surface-area barium hexa-aluminate (BaAl12O19,
Biomass
Monomer
Heterogeneous catalysis
Methacrylic acid
Itaconic acid
BHA) was developed as a catalyst for the decarboxylation of the biomass-derived itaconic acid (IA) to bio-based
methacrylic acid (MAA). A maximal 50% final yield of MAA with a high product selectivity was obtained under
relatively mild synthesis reaction conditions (250 °C; 20 bar N2). The reported selective MAA production was
elevated, operating process characteristics were significantly less harsh, and no depleting critical raw materials
were utilized when paralleled to the procedures with alkaline mineral bases, noble metal-containing hetero-
geneous catalysis systems and unrenewable feed resources (e.g. isobutene), applied previously. It was found that
the doping of palladium on BHA support (Pd@BHA) did not improve MAA productivity. The effect of the time
(25–300 min), temperature (175–275 °C), pressure (10–40 bar), reacting substrate concentration (0.10–0.19 mol
L
–1), metallic oxide mass (0.5–3.0 g) and type on IA conversion, MAA content MAA content and rates was
determined, examining also recyclability. BHA catalyst was characterized with various structural techniques,
such as energy-dispersive X-ray spectroscopy (EDS), X-ray powder diffraction (XRD), CO2 temperature-pro-
grammed desorption (TPD), scanning electron microscopy (SEM) and N2 physisorption.
Introduction
biomass-derived 2-methyl-1,3-propanediol have been attempted [6].
Alternative methodology proposes the oxidative dehydrogenation of
Methacrylic acid (MAA) and methyl methacrylate (MMA) are in-
dustrially important monomers, widely used in organic glass (poly-
methyl methacrylate), acrylic fibers, plastics, paint, and clustering
agent production [1]. The current industrial manufacture of MAA and
MMA is based in an acetone-cyanohydrin (ACH) process [2]. This un-
sustainable method relies on the use of expensive and extremely toxic
feedstocks alongside corrosive concentrated acids. Besides the use of
harmful substrates, low atom economy, poor product selectivity and the
net emission of greenhouse gases are also drawbacks associated with
the industrial process, not to mention that the production is exclusively
based on a non-renewable fossil-based resource [3]. To overcome the
aforementioned limitations, Mitsubishi Gas Chemicals developed an
improved ACH process; however, the short catalyst lifetimes and
parasitic reactions issues remain to be solved [4]. Later on, Mitsubishi
Rayon developed an alternative two-step process to produce MMA via
the oxidation of isobutylene to MAA and its further esterification to
MMA [5].
isobutyric acid (IBA) to MAA, being this prepared by lignocellulosic
biomass fermentation [7]. However, the similar IBA and MAA boiling
points (155 °C vs 162 °C) make this separation difficult [8,9]. In addi-
tion to the existing catalytic routes, MAA fermentative routes from re-
newable carbohydrates have also been envisaged using non-naturally
occurring microbial organisms [10,11]. Other bio-based routes to MAA
can also be accomplished using citric acid, itaconic acid, ethylene,
methanol and carbon monoxide as raw materials. These can easily be
obtained from biomass such as Lucite’s Alpha and Evonik’s AVENEER
processes [12–14].
Decarboxylation is an important step, widely applied in chemical
industry to obtain valuable products such as butanone, C9-ketone, C17
hydrocarbon, anisol, benzyl cyanide and 2-methylbutyronitrile via the
reaction of levulinic acid, 6-amyl-a-pyrone, stearic acid, ortho-anisic
acid, isoleucine and phenyl alanine, respectively [15–17]. One such
process involved the decarboxylation of biomass derived itaconic acid
to MAA. Itaconic acid is an important platform obtained by plant bio-
mass fermentation and can be a potential substitute for petroleum-de-
rived chemicals such as acetone cyanohydrin, maleic acid or acrylic
acid [18,19]. To the best of our knowledge, only three research articles
Considering the MAA high potential, both academic and industrial
researchers have attempted to derive MAA from bio-based feedstocks.
In this sense, oxidative bioconversion and catalytic dehydration of
⁎ Corresponding authors.
E-mail addresses: ashish.bohre@ki.si (A. Bohre), blaz.likozar@ki.si (B. Likozar).
https://doi.org/10.1016/j.mcat.2019.110520
Received 18 February 2019; Received in revised form 16 July 2019; Accepted 17 July 2019
2468-8231/©2019ElsevierB.V.Allrightsreserved.

A. Bohre, et al.
MolecularCatalysis476(2019)110520
Table 1
Experimental
A comparison of existing routes to bio-based MAA with the present work.
Preparation of catalyst
Entrya Catalyst
T [°C] time
P [bar] MAA
yield [%]
Ref.
BHA catalyst was prepared through co-precipitation method.
Typically, 90.03 g of Al(NO3)39H2O and 5.23 g of Ba(NO3)2 were dis-
solved in deionized water at 60 °C. Subsequently, 40 g of carbon black
were added and the mixture was vigorously stirred overnight to get a
slurry precursor. Likewise, 96.09 g of (NH4)2CO3 was dissolved in
200 mL deionized water and stirred for 1 h at 60 °C. Afterwards, BHA
slurry precursor was poured into an (NH4)2CO3 solution and the mix-
ture was vigorously stirred for 4 h to obtain a gel. The obtained gel was
filtered, washed with deionized water and vacuum dried at 110 °C
overnight. The resultant powder was calcined in Ar at 1250 °C for 5 h,
followed by calcination at 900 °C in air for 12 h to completely remove
the carbon, leading to a high surface area BHA catalyst.
Pd/BHA catalyst was prepared by a deposition–precipitation (DP)
procedure. In a typical procedure, 1 g of BHA support was added to
20 mL aqueous solution containing 250 mg of Pd(NO3)2. The pH of the
solution was adjusted to 9 by adding 0.2 M NH4OH under vigorous
stirring. The mixture was then stirred for 1 h at 60 °C. After cooling to
room temperature, the solid was recovered by filtration and washed
with distilled water. The mixture was then placed in a vacuum oven and
allowed to dry overnight at 40 °C. The dried material was then trans-
ferred to a Schwartz-type drying tube and reduced in a H2 flow at
350 °C for 3 h. The Pd/BHA catalyst was subsequently cooled to room
temperature under flowing N2. g-C3N4 catalyst was prepared from urea
precursor by adopting the reported method [40].
1
2
3
Pt/Al2O3 and NaOH
250
250
225
1 h
38
68
23
40
[20]
[21]
[22]
Hydrotalcite
Ruthenium carbonyl
propionate
15 min 34
1.5 h
3 h
28
20
4
a
BHA
250
50
This
work
Reaction conditions: Reactor mode (batch), feedstock (IA), solvent (water).
have been published on catalytic decarboxylation of IA to MAA. The
first report was published by Notre’s group in 2014 [20], showing a
good MAA yield. However, expensive Pt/Al2O3 catalyst and corrosive
alkaline base were required to decarboxylate the IA (Table 1, entry 1).
In order to avoid the use of homogenous alkali solution and expensive
noble metals, Pirmoradi’s group explored hydrotalcite as a solid catalyst
for MAA synthesis. Unfortunately, only 23% yield of MAA was achieved
(Table 1, entry 2) [21]. Recently, ruthenium carbonyl propionate
complex was used as a homogenous catalyst for the decarboxylation of
IA (Table 1, entry 3). Homogeneous catalysts have a general superior
selectivity and specificity but their separation from the reaction is
cumbersome [22]. It was reported that Pt/Al2O3, hydrotalcite and ru-
thenium carbonyl propionate complex based catalysts have a positive
stabilisation effect on the itaconate monoanion and allow the dec-
arboxylation to MAA to take place with a better selectivity. In com-
pared to the reported catalysts the excellent catalytic activity of barium
hexaaluminate is attributed to high surface area and moderate basicity
that promotes the decarboxylation of IA.
The as-synthesized BHA catalyst was characterized by Powder X-ray
diffraction, Temperature programmed desorption, scanning electron
microscope (SEM) and transmission electron microscope (TEM). The
detailed information is available in supporting information.
Hexaaluminates are an attractive class of hexagonal aluminate
materials widely used as catalysts or as a catalyst support for high
temperature applications such as natural gas combustion, carbon di-
oxide reforming of methane, partial oxidation of hydrocarbon and ni-
trous oxide decomposition [23]. The high catalytic activity and re-
markable thermal stability of hexaaluminates are associated with their
peculiar layered structure, consisting of alternatively stacked spinel
Catalytic decarboxylation
Decarboxylation reactions were performed in a high pressure batch
reactor (250 mL, Amar Equipment Pvt. Ltd., India) equipped with a
thermocouple, pressure gauge, rupture disk, gas and liquid sampling
line. In a typical experiment, 2 g of itaconic acid was dissolved in
150 mL of deionized water. The solution was loaded into the autoclave
reactor with a catalyst (1 g) and sealed. The reactor was pressurized
with N2 to 10 bar and vented three times to remove any residual
oxygen. Finally, the reactor was pressurized to 20 bars with N2, stirred
by the Rushton turbine (600 min–1) and heated up to 250 °C. It had been
previously determined that under the stirring speed of 600 min−1 the
reaction was conducted in the kinetic regime, since using higher stirring
rates did not increase the conversion. Most of the experiments were
performed under kinetic regime. We have taken the samples with re-
spect to time as seen in the Fig. S1. However the selectivity of MAA has
not improved at low IA conversion.The final temperature was reached
in 45 min and the autogenic pressure reached 52 bar, which remained
constant during the experiment. The recorded reaction time started
when the temperature reached the desired set-point. Once the reaction
was completed, the reactor was cooled-down to the room temperature
(20 °C). After collecting the final gas and liquid sample, pressure from
the headspace of the reactor was released. The crude reaction mixture
was analysed by UHPLC (liquid phase) and micro-GC (gas phase).
blocks of γ-Al2O3 and mirror planes in which large cations (La3+, Ba2+
,
Sr2+ etc.) are located [24]. The hexaaluminates crystal structure is
dependent on the atomic radius and charge density of the large cations
in the mirror planes. For instance, barium hexaaluminate (BHA) has
assigned the β-Al2O3 structure, whereas lanthanum hexaaluminate be-
longs to the magnetoplumbite structure [25]. Hexaaluminate pre-
cursors are typically prepared by co-precipitation or sol-gel method.
The materials crystallization at high temperature (> 1200 °C) results in
low surface area 5-15 m2 g−1, leading to a lower amount of active sites
[26]. To overcome such limitation, alternative methods have been
proposed, for instances coupling of a sol–gel process in reverse micro-
emulsions and solid-state reaction followed by two steps of ball milling.
However, the aforementioned routes are hardly applicable for in-
dustrial implementation due to the high-energy-consumption, con-
straints of scalability, economic, and safety issues.
Inspired by the pioneer work of Santiago et al. [27], we have pre-
pared a high surface area (96.0 m2 g–1) BHA catalyst through carbon
templating route. The excellent catalytic activity of BHA was tested for
the first time for high temperature decarboxylation reactions under
subcritical water conditions. The work here described has several ad-
vantages over the previously published works [20–22]. Most notably,
this work brings improvements in the use of heterogeneous catalyst by
avoiding corrosive liquid base additive and precious metals like Pt and
Ru. The BHA catalyst showed moderate activity and good recyclability
without the doping of expensive transition metals or using homogenous
catalyst (Table 1, entry 1 and entry 3).
MAA extraction and purification procedure
After reaction has occurred, the catalyst was separated by cen-
trifugation and the resultant liquid was collected in a round bottom
flask. The residual MAA in the aqueous solution was extracted by Et2O
(3 × 50 mL), and the combined organic layers were dried over MgSO4
while Et2O was removed in vacuo. Lastly, the purified MAA was ob-
tained after the vacuum distillation (20 mbar) at 40 °C for more than
2

A. Bohre, et al.
MolecularCatalysis476(2019)110520
Fig. 1. (a) XRD pattern, (b) Nitrogen adsorption-desorption isotherm, (c) pore size distribution, and (d) CO2-TPD profile of the BHA catalyst.
2 h. The isolated MAA was analyzed by NMR spectroscopy. Both 1H and
13C NMR spectra of the liquid product revealed the isolation of a pure
MAA (Fig. S2-S3).
the temperature programmed desorption of CO2. Results showed four
desorption peaks with different basic strength distributed on the surface
of BHA catalyst (Fig. 1d). The profile of BHA catalyst exhibited a large
amount of desorbed CO2 at 350 °C, implying the presence of low and
medium basic sites. These could be associated with the surface of HO−
and Al3+-O2- Lewis acid-base pairs. The desorption peak at 600 °C; can
be ascribed to the strong base site type related to a low-coordination
surface of O2- anion [31]. The total amount of basic sites in BHA cat-
alyst was 3.05 μmol g–1. The SEM image of BHA catalyst is shown in
Fig. 2a, where it can be seen irregular aggregates. Regarding the BHA
catalyst TEM images (Fig. 2 b–d), two distinctive morphologies can be
clearly seen, namely one with big crystallites and the other with small
agglomerates. The big crystallites seem to grow in the form and shape
of platelets. Small agglomerates show features in the form of needles or
3D worm-like. Selected area electron diffraction (SAED) pattern de-
monstrated the monocrystalline characteristic of the catalyst (Fig. 2c,
inset). According to the EDX maps, Ba is well incorporated into the big
crystallites (Fig. S6).
MAA: 1H NMR (CDCl3, 400 MHz): δ = 1.98 (dd, 3 H), 5.71 (m, 1 H),
6.28 (m. 1 H), 11.60 ppm (br s, 1 H); 13C NMR (CDCl3): δ = 17.84,
127.87, 135.75, 173.19 ppm.
Results and discussion
Structural and physicochemical characterization
The powder XRD pattern of BHA catalyst (Fig. 1a) showed diffrac-
tion peaks at 19.6°, 28.3°, 343° 40.1°, 45.1°, 57.7°, and 69.6° corre-
sponding to the (103), (105), (110), (114), (205), (302), and (220)
planes of BHA (JCPDS card, file no. 00-026-0135, BaAl12O19 phase),
respectively [28]. The strong and sharp diffraction peaks indicate that a
relatively pure BHA crystalline phase was obtained after calcination at
1250 °C. In the XRD pattern of Pd/BHA catalyst (Fig. S4) the Pd peaks at
40.1°, 46.6°, and 68.1 were attributed to the (111), (200), and (220)
planes of the face centered cubic structure of Pd (JCPDS card, file no.
46–1043), respectively. The X‐ray diffraction pattern of g-C3N4 (Fig. S5)
displays one strong diffraction peak at 27.4° corresponds to (0 0 2)
plane which depicts the interlayer stacking of conjugated aromatic
systems of graphene related materials (g-C3N4) which is in consistent
with the literature (JCPDS card, file no. 87-1526).
The nitrogen adsorption-desorption isotherms and BJH (Barrett-
Joyner-Halenda) pore size distribution curves of BHA catalyst are
shown in Fig. 1b–c. It can be observed that the catalyst demonstrated
type-IV curve with hysteresis loop lying at a relative pressure (p/p°) of
0.70-0.89, indicating the presence of uniform cylindrical mesoporous
channels [29], On the other hand, the BJH pore size distribution curve
of BHA catalyst exhibits pore size distribution in the macroporous and
mesoporous regions. It was also observed that BHA catalyst possessed a
high BET specific surface area and a pore volume of 96.0 m2 g–1 and
0.46 cm3 g–1, respectively. These are a result of the carbon black addi-
tion as a hard templating agent, which prevents the particles agglom-
eration during the high temperature calcination process [30].
Selective decarboxylation of itaconic acid
In order to determine the influence of the reaction temperature on
the BHA catalyst performance, the experiments were conducted in the
temperature range of 175–275 °C (Table 2, entry 1–5). At low tem-
peratures (175–200 °C), the IA conversion is incomplete and the se-
lectivity towards MAA is low. Nevertheless, upon increasing the reac-
tion temperature from 200 to 250 °C, the MAA selectivity was increased
from 19 to 30% with a complete IA conversion. By further increasing
the reaction temperature to 275 °C, the MAA selectivity decreased once
again to 27%. This might be due to the side reactions (dehydration and
decarboxylation) that take place at high temperature (Scheme 1) [32].
The catalytic activity was further investigated by tuning the reac-
tion pressure in the second set of experiments, while temperature was
kept constant at the previous optimized value (Table 2, Entry 6–8). As
expected, the catalytic reaction performed at low pressure led to a poor
MAA selectivity. Yet, when the reaction was performed at 20 bar N2
pressure, the decarboxylation rate increased and reached a MAA se-
lectivity maximum of 50%. Furthermore, when the pressure was in-
creased (> 30 bar) under these conditions, the MAA selectivity
Since basicity plays a crucial role in the decarboxylation reactions,
the basicity strength of the prepared BHA catalyst was determined by
3

A. Bohre, et al.
MolecularCatalysis476(2019)110520
Fig. 2. (a) SEM image, (b–d) TEM images, and SAED pattern (inset) of the BHA catalyst.
decreased and the formation of side products predominated, indicating
that a higher pressure is not favorable in this case. Notably, under these
reaction conditions, a 4 bar increase of the internal pressure was ob-
served probably caused by the formation of gaseous by-product (pro-
pene) due to the subsequent decarboxylation step (Scheme 1, Fig. S7)
[33]. Our initial experiment that used a 0.15 M itaconic acid resulted in
complete conversion after 3 h at 250 °C (Table 2, entry 7). The MAA
selectivity reached 50%, and the rest of the products consisted of
degradation products. If we decreased the substrate concentration to
0.10 M, the level of MAA selectivity decreased slightly to 45% (Table 2,
entry 9). If we increased the substrate concentration to 0.20 M, a lower
MAA selectivity (32%) was obtained and yield of degradation products
increased to 45% (Table 2, entry 10).
In the third set of experiments, we have varied reaction time, and
catalyst dosages. Unfortunately, it did not improve the MAA selectivity
(Table 2, entry 11–14). The screening experiments suggested that a
Table 2
Screening of reaction conditions for decarboxylation of itaconic acid with BHA catalyst.
Entrya
T [°C]
P [bar]
Conversion [%]
Selectivity [mol %]
MAA
Isomersb
2-HIBA
AA
Othersc
1
2
3
4
5
6
7
8
175
200
225
250
275
250
250
250
250
250
250
250
250
250
40
40
40
40
40
10
20
30
20
20
20
20
20
20
80
87
92
15
19
25
30
27
24
50
45
45
32
44
42
47
49
31
23
6
2
–
3
–
–
–
–
–
–
–
6
9
11
5
19
20
23
14
12
21
3
2
1
1
1
35
32
37
43
56
44
29
33
36
45
35
34
25
24
100
100
100
100
100
100
100
100
100
100
100
8
18
19
18
22
20
23
23
23
9d
10e
11f
12g
13h
14i
1
5
4
–
–
d–eIA = 0.10 M & 0.20 M.
f–gCatalyst = 2 g & 3 g.
h–iTime (4 h & 5 h). Values were determined by HPLC and micro-GC. IA is itaconic acid, MAA is methacrylic acid, AA is acetic acid and 2-HIBA is 2-hydroxyisobutyric
acid.
a
Reaction conditions : IA = 0.15 M, solvent (water), time (3 h), BHA catalyst = 1 g.
isomers = mesaconic acid + citraconic acid.
Others = pyruvic acid, crotonic acid, acetone, propene and carbon dioxide.
b
c
4

A. Bohre, et al.
MolecularCatalysis476(2019)110520
Scheme 1. Proposed catalytic mechanistic pathways to form methacrylic acid and by-products on BHA catalyst.
reaction temperature of 250 °C, 20 bar pressure, 1 g of BHA catalyst and
reaction time of 3 h offered the highest MAA yield (50%) and low by-
product formation within the tested window of process parameters.
With the optimized decarboxylation reaction conditions in hand, the
next investigation was conducted on a series of as-prepared catalysts
(Table 3). For the initial proof of concept, we have started our experi-
ment with a homogenous NaOH catalyst. It was found that the con-
version of IA was high but the selectivity towards the MAA was low
(Table 3, entry 1). This agrees with the literature data that shows that
the higher concentration of OH− in the reaction mixture quenches the
decarboxylation reaction, leading to a lower yield of the desired pro-
duct [32]. It has also been reported in literature that the addition of
NaOH as a co-catalyst is able to improve the decarboxylation product
yield by suppressing the hydration of double bonds [34].
However, in case of BHA catalyst, the NaOH addition impairs the
catalytic activity (Table 3, entry 2), presumably due to the adsorption of
by-products on the surface of BHA catalyst that could block the active
sites [35]. Afterwards, we tested the catalytic activity of graphitic
carbon nitride (g-C3N4) for the decarboxylation of MAA. g-C3N4 catalyst
was of great interest owing to its Lewis-base character, layered struc-
ture and number of surface defects, which can be useful in the dec-
arboxylation reaction [36,37]. Unfortunately, g-C3N4 catalyst gets so-
lubilized during the reaction under our reaction conditions resulting in
a low MAA yield (Table 3, entry 3, Fig. S8).
of Pd metal (10 wt %) in the BHA catalyst matrix did not improve the
MAA yield (Table 3, entry 7) probably due to the low surface area of
Pd/BHA catalyst (Fig. S9). It is believed that the high BHA catalytic
activity compared to the other tested catalysts is associated with its
moderate basicity and high surface area [38,39]. Moreover, the large
cations (Ba2+, Ca2+, Sr2+) stabilized in the matrix of hexaaluminates
are able to maintain a relative activity and superior stability under
subcritical water conditions [31]. These large cations provide the ap-
propriate basic capacity that could also improve the catalytic activity of
BHA. It was evident from the solvent variation experiment that water is
a necessary component in our reaction medium, as it mitigates the
formation of anhydrides as evident by the solvent variation experiments
(Table S2) [23].
To demonstrate the recyclability of the BHA catalyst, the reaction
was carried out in four consecutive repetitive cycles. As shown in Fig. 3,
the MAA yield decreased from 50% (1st cycle) to 37% (4th cycle). This
could also be partly attributed to the loss of the catalyst mass (7%)
during the recovery, which can also accounts for the variation in cat-
alytic activity. The XRD and CO2-TPD analysis of the used BHA catalyst
confirmed that the reaction conditions have not induced the structural
and textural modifications (Fig. S10-S11).
Reaction mechanism
There are some reports that Pd containing materials are robust
catalyst for the decarboxylation of itaconic acid reactions [20,34]. In
light on these studies, we have designed Pd catalyst supported on BHA
for the itaconic acid decarboxylation. It was observed that the addition
The proposed catalytic mechanistic pathways relating to formation
of MAA and other by-products are provided in Scheme 1. The formation
of MAA is that IA firstly isomerized to mesaconic acid and citraconic
acid mainly over the weak acid sites on the catalyst. In the second step
abstraction of proton from IA was occurred on the surface of the BHA
catalyst.
Table 3
The carbonium ion generated during this process was decarbox-
ylates to MAA by leaving the β-carboxyl group. The second decarbox-
ylation of MAA produced propene that we have identified in the gas
phase analysis (Fig. S7, Scheme 1, reaction A). We have not detected 2-
hydroxyisobutyric acid in the reaction mixture that confirmed that the
hydration of MAA is not possible during the reaction. Interestingly, the
formation of by-products was started from the beginning by following
parallel pathways.
Comparison of the activity of BHA with different catalysts.
Entrya
Catalyst
Conversion [%]
Yield of MAA (%)
1
2
3
4
5
NaOHb
BHA/NaOH
g-C3N4
Pd/BHA
BHA
87
95
79
100
100
10
32
20
46
50
a
Under hydrothermal conditions in reaction B, citramalic acid was
produced by the hydration of any of the three itaconic acid isomers.
However, we have not detected the citramalic acid in our reaction
Reaction conditions : IA = 0.15 M, solvent (water), time (3 h), temperature
250 °C, pressure 20 bar, catalyst = 1 g.
b
NaOH = 2 equivalents.
5

A. Bohre, et al.
MolecularCatalysis476(2019)110520
Fig. 3. Recyclability of the BHA catalyst. Reaction conditions: Itaconic acid (2 g), water (150 mL), catalyst (1 g), initial pressure (20 bar).
mixture. Citramalic acid is not stable and further decarboxylate to 2-
hydroxyisobutyric acid (2-HIBA) that subsequently produced different
degradation products such as pyruvic acid, acetic acid, acetone and
acetaldehyde. Propene can also be produced by following pathway C.
References
[1] M.J.D. Mahboub, J.L. Dubois, F. Cavani, M. Rostamizadeh, G.S. Patience, Catalysis
for the synthesis of methacrylic acid and methyl methacrylate, Chem. Soc. Rev. 47
(2018) 7703–7738.
[2] K. Nagai, New developments in the production of methyl methacrylate, Appl. Catal.
A 221 (2001) 367–377.
Conclusion
[3] Y. Hu, G. Jiang, G. Xu, X. Mu, Hydrogenolysis of lignin model compounds into
aromatics with bimetallic Ru-Ni supported onto nitrogen-doped activated carbon
catalyst, Mol. Catal. 445 (2018) 316–326.
[4] Y.L. Cao, L. Wang, L.L. Zhou, B.H. Xu, Y.Y. Diao, S.J. Zhang, A modified heteropoly
acid catalyst with cetyltrimethylammonium bromide for methacrolein to me-
thacrylic acid, J. Ind. Eng. Chem. 65 (2018) 254–263.
[5] L. Zhou, L. Wang, S. Zhang, R. Yan, Y. Diao, Effect of vanadyl species in Keggin-type
heteropoly catalysts in selective oxidation of methacrolein to methacrylic acid, J.
Catal. 329 (2015) 431–440.
[6] S.H. Pyo, T. Dishisha, S. Dayankac, J. Gerelsaikhan, S. Lundmark, N. Rehnberg,
R. Hatti-Kaul, A new route for the synthesis of methacrylic acid from 2-methyl-1, 3-
propanediol by integrating biotransformation and catalytic dehydration, Green
Chem. 14 (2012) 1942–1948.
[7] K. Zhang, A.P. Woodruff, M. Xiong, J. Zhou, Y.K. Dhande, A synthetic metabolic
pathway for production of the platform chemical isobutyric acid, ChemSusChem 4
(2011) 1068–1070.
[8] A. Morschbacker, Bio-ethanol based ethylene, J. Macromol. Sci. Polymer Rev. 49
(2009) 79–84.
[9] B.N.M. van Leeuwen, A.M. van der Wulp, I. Duijnstee, A.J.A. van Maris,
A.J.J. Straathof, Fermentative production of isobutene, Appl. Microbiol. Biotechnol.
93 (2012) 1377–1387.
[10] T. Nagasawa, T. Nakamura, H. Yamada, Production of acrylic acid and methacrylic
acid using Rhodococcus rhodochrous J1 nitrilase, Appl. Microbiol. Biotechnol. 34
(3) (1990) 322–324.
We have developed a new system for the highly selective dec-
arboxylation of IA to MAA using BHA catalyst. A number of important
parameters have been studied, and it was found that the reaction
temperature and pressure had a considerable effect on IA conversion
and MAA selectivity. Under optimal conditions, a complete conversion
of IA and a MAA yield of 50% were obtained after 3 h at 250 °C. Our
catalytic system has demonstrated two important advantages: (a) it did
not require a homogeneous alkaline base or loading of expensive noble
metals and (b) the catalyst shows high catalytic activity under relatively
mild reaction conditions.
Extensive solid-state characterization using spectroscopic techni-
ques revealed that a high BHA catalytic activity associated with its
moderate basicity and high surface area. To the best of our knowledge,
this is the first report of this method on the selective and high yield
production of MAA from biomass-derived itaconic acid over a hetero-
geneous catalyst under mild reaction conditions in absence of corrosive
mineral base. The current systems leave room for optimization, and
especially the decarboxylation of biomass-derived substrates to MAA at
low temperature is one of the most appealing starting-points to com-
mercialize this process. We are currently developing this chemistry
further through catalyst and process design with specific application
towards scale-up.
[11] A.P. Burgard, M.J. Burk, R.E. Osterhout, P. Pharkya, Microorganisms for the pro-
duction of methacrylic acid, US8241877, 2012.
[12] S. Ziemian, I.A. York, A Process for the Production of Ethylenically Unsaturated
Carboxylic Acids or Esters and a Catalyst Therefor, EP1176366A2 (2005).
[13] L. Zhoua, L. Wanga, Y. Diao, R. Yana, S. Zhang, Cesium salts supported heteropoly
acid for oxidation of methacrolein to methacrylic acid, Mol. Catal. 433 (2017)
153–161.
[14] L. Zhoua, L. Wang, Yunli Cao, Y. Diao, R. Yan, S. Zhang, The states and effects of
copper in Keggin-type heteropolyoxometalate catalysts on oxidation of methacro-
lein to methacrylic acid, Mol.Catal. 438 (2017) 47–54.
Acknowledgements
[15] P. Tuhin, D. Maiti, Decarboxylation as the key step in C-C bond-forming, Chem. Eur.
The authors gratefully acknowledge the financial support of the
Slovenian Research Agency (ARRS) through the Program P2-0152. The
work was partially carried out within the RDI project Cel. Cycle:
Potential of biomass for the development of advanced materials and
bio-based products», co-financed by the Republic of Slovenia, Ministry
of Education, Science and Sport and European Union under the
European Regional Development Fund, 2016-2020.
J. 23 (2017) 7382–7401.
[16] M.I. Alam, S. Gupta, A. Bohre, E. Ahmad, T.S. Khan, B. Saha, M.A. Haider,
Development of 6-amyl-α-pyrone as a potential biomass-derived platform molecule,
Green Chem. 18 (2016) 6431–6435.
[17] A.V. Ignatchenko, J.P. McSally, M.D. Bishop, J. Zweigle, Ab initio study of the
mechanism of carboxylic acids cross-ketonization on monoclinic zirconia via con-
densation to beta-keto acids followed by decarboxylation, Mol. Catal. 441 (2017)
35–62.
[18] J.L. Ramos, Z. Udaondo, B. Fernández, C. Molina, A. Daddaoua, A. Segura,
E. Duque, First‐and second-generation biochemicals from sugars: biosynthesis of
itaconic acid, Microb. Biotechnol. 9 (2016) 8–10.
[19] S.T. Ahmed, N.G.H. Leferink, N.S. Scrutton, Chemo-enzymatic routes towards the
synthesis of bio-based monomers and polymers, Mol. Catal. 467 (2019) 95–110.
[20] J.L. Nôtre, S.W. Dijk, J.V. Haveren, E.L. Scott, J.P. Sanders, Synthesis of bio‐based
methacrylic acid by decarboxylation of itaconic acid and citric acid catalysed by
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.110520.
6

A. Bohre, et al.
MolecularCatalysis476(2019)110520
solid transition‐metal catalysts, ChemSusChem 7 (2014) 2712–2720.
and catalytic performance, Catal. Sci. Technol. 6 (2016) 1984–2004.
[21] M. Pirmoradi, J.R. Kastner, Synthesis of methacrylic acid by catalytic decarbox-
ylation and dehydration of carboxylic acids using a solid base and subcritical water,
ACS sustain. Chem. Eng. 5 (2016) 1517–1527.
[22] J.C. Lansing, R.E. Murray, B.R. Moser, Biobased methacrylic acid via selective
catalytic decarboxylation of itaconic acid, ACS Sustain. Chem. Eng. 5 (2017)
3132–3140.
[23] N. Iyi, S. Takekawa, S. Kimura, Crystal chemistry of hexaaluminates: β-alumina and
magnetoplumbite structures, J. Solid State Chem. 83 (1989) 8–19.
[24] Y. Zhang, X. Wang, Y. Zhu, X. Liu, T. Zhang, Thermal evolution crystal structure and
Fe crystallographic sites in LaFexAl12–xO19 hexaaluminates, J. Phys. Chem. A 118
(2014) 10792–10804.
[25] G. Groppi, M. Bellotto, C. Cristiani, P. Forzatti, P. Villa, Thermal evolution crystal
structure and cation valence of Mn in substituted Ba-β-Al2O3 prepared via copre-
cipitation in aqueous medium, J. Mater. Sci. 34 (1999) 2609–2620.
[26] P. Forzatti, L. Lietti, Catalyst deactivation, Catal. Today 52 (1999) 165–181.
[27] M. Santiago, J.C. Groen, J. Pérez-Ramírez, Carbon-templated hexaaluminates with
enhanced surface area and catalytic performance, J. Catal. 257 (2008) 152–162.
[28] L. Liotta, G. Deganello, D. Sannino, M. Gaudino, P. Ciambelli, S. Gialanella,
Influence of barium and cerium oxides on alumina supported Pd catalysts for hy-
drocarbon combustion, Appl. Catal. A 229 (2002) 217–227.
[32] M. Carlsson, C. Habenicht, L.C. Kam, M.J.J. Antal, N. Bian, R.J. Cunningham,
M.J. Jones, Study of the sequential conversion of citric to itaconic to methacrylic
acid in near-critical and supercritical water, Ind. Eng. Chem. Res. 33 (1994)
1989–1996.
[33] A. Bohre, B. Hočevar, M. Grilc, B. Likozar, Selective catalytic decarboxylation of
biomass-derived carboxylic acids to bio-based methacrylic acid over hexaaluminate
catalysts, Appl. Catal. B Environ. 256 (2019) 117889.
[34] J. Verduyckt, D.E. De Vos, Highly selective one-step dehydration, decarboxylation
and hydrogenation of citric acid to methylsuccinic acid, Chem. Sci. 8 (2017)
2616–2620.
[35] E.M. Johansson, K.J. Danielsson, E. Pocoroba, E.D. Haralson, S.G. Jaras, Catalytic
combustion of gasified biomass over hexaaluminate catalysts: influence of palla-
dium loading and ageing, Appl. Catal. A 182 (1999) 199–208.
[36] J. Zhu, P. Xiao, H. Li, S.A. Carabineiro, Graphitic carbon nitride: synthesis, prop-
erties, and applications in catalysis, ACS Appl. Mater. Interfaces 6 (2014)
16449–16465.
[37] N.S. Kus, Organic reactions in subcritical and supercritical water, Tetrahedron 4
(2012) 949–958.
[38] J.G. Park, A. Cormack, Crystal/Defect structures and phase stability in Ba hex-
aaluminates, J. Solid State Chem. 121 (1996) 278–290.
[29] M. Manikandana, M. Prabu, S.K.A. Kumar, P. Sangeetha, R. Vijayaraghavan, Tuning
the basicity of Cu-based mixed oxide catalysts towards the efficient conversion of
glycerol to glycerol carbonate, Mol. Catal. 460 (2018) 53–62.
[30] S. Pyena, E. Honga, Mi Shin, Y.-W. Suh, C.-H. Shin, Acidity of co-precipitated SiO2-
ZrO2 mixed oxides in the acid-catalyzed dehydrations of iso-propanol and formic
acid, Mol. Catal. 448 (2018) 71–77.
[39] S.Q. Wang, Z.W. Wang, L.C. Yang, Jl. Dong, C.Q. Chi, D.N. Sui, Y.Z. Wang, J.G. Ren,
M.Y. Hung, Y.Y. Jiang, A novel efficient route for preparation of chiral β-hydro-
xycarboxylic acid: asymmetric hydration of unsaturated carboxylic acids catalyzed
by heterobimetallic complex wool–palladium–cobalt, J. Mol. Catal. 264 (2007)
60–65.
[40] S. Martha, A. Nashim, K. Pardia, Facile synthesis of highly active g-C3N4 for effecent
hydrogen production under visible light, J. Mater. Chem. A 1 (2013) 7816–7824.
[31] M. Tian, X. Wang, T. Zhang, Hexaaluminates: a review of the structure, synthesis
7