Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as catalyst

Article abstract of DOI:10.1016/j.cattod.2016.02.015

The oxidation of glycerol into lactic acid using activated carbon with Pd or Pt as heterogeneous catalysts was study. Several reaction parameters were evaluated: catalyst loading, catalyst weight, NaOH/glycerol molar ratio, temperature and reaction time. The textural and structural properties of the carbon supported catalysts (fresh and spent) were evaluated by N₂ adsorption-desorption isotherms at ?196 °C, XRD, SEM, EDX, TEM and XPS. It was possible to achieve a glycerol conversion of about 99% with selectivity to lactic acid of 68% and 74% using 10% Pd/C and 5% Pt/C as catalysts, respectively, at 230 °C of reaction temperature. Moreover, Pd/C and Pt/C catalysts were used in five reaction cycles without significant loss of activity, highlighting the great stability of the prepared catalysts.

Full text of DOI:10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Conversion of glycerol into lactic acid using Pd or Pt supported on
carbon as catalyst
a
a
a
b
M Rosiene A. Arcanjo , Ivanildo J. Silva Jr. , Enrique Rodríguez-Castellón ,
b
a,∗
Antonia Infantes-Molina , Rodrigo S. Vieira
Grupo de Pesquisas em Separa c¸ ões por Adsor c¸ ao, Departamento de Engenharia Química, Universidade Federal do Ceará, Fortaleza, CE 60455-760, Brazil
Departamento de Química Inorgánica, Facultad de Ciencias, Universidad de Málaga, Málaga, 29071, Spain
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
The oxidation of glycerol into lactic acid using activated carbon with Pd or Pt as heterogeneous catalysts
was study. Several reaction parameters were evaluated: catalyst loading, catalyst weight, NaOH/glycerol
molar ratio, temperature and reaction time. The textural and structural properties of the carbon supported
Received 14 December 2015
Received in revised form 20 February 2016
Accepted 23 February 2016
Available online xxx
◦
catalysts (fresh and spent) were evaluated by N2 adsorption-desorption isotherms at −196 C, XRD, SEM,
EDX, TEM and XPS. It was possible to achieve a glycerol conversion of about 99% with selectivity to
◦
lactic acid of 68% and 74% using 10% Pd/C and 5% Pt/C as catalysts, respectively, at 230 C of reaction
Keywords:
Glycerol conversion
Acid lactic
Pd and Pt metals
Activated carbon
temperature. Moreover, Pd/C and Pt/C catalysts were used in five reaction cycles without significant loss
of activity, highlighting the great stability of the prepared catalysts.
©
2016 Elsevier B.V. All rights reserved.
1
. Introduction
market of 329,000 t in 2015 and recent projections announced an
estimated production of 830,000 t of poly (lactic acid) (PLA) in 2020
[4].
Nowadays, biodiesel production is on a large scale, mainly, due
to the government incentives and the development of renewables
sources for fuels production. In the transesterification reaction,
about, 10% in weight of raw material is converted into glycerin.
That is why biodiesel industries are facing a surplus of this main
by-product, glycerol, representing 10% (v/v) of the final ester. Due
to the government incentives to increase carbon and oil energy
independence and meet the growing energy demand, the biodiesel
market is expected to reach 37 billion liters in 2016. This production
will result in 4 billion liters of crude glycerol, generating a surplus
volume that the glycerol market cannot absorb [1,2]. In this way,
some studies have been conducted to make the glycerol industry
more profitable. Various catalytic conversion processes have been
proposed for the transformation of glycerol into chemicals with
high added value, such as 1,3 propanediol, 1,2 propanediol, succinic
acid, polyesters of lactic acid and polyglycerols [3].
Lactic acid is conventionally used as an acidulant and/or a con-
servative in the food industry and it has also been used as a raw
material for the production of pharmaceuticals, cosmetics, tex-
tiles, leather, and biodegradable polymer (Poly (lactic acid)). The
poly (lactic acid) has wide applications in a variety of fields due to
their biodegradability and biocompatibility. This organic acid can
be produced by chemical synthesis or by bioprocesses. However,
the last one is the predominant method for lactic acid produc-
tion and presents several disadvantages, such as large amount of
water costs, low reaction rate and high cost due to their separation
and purification in downstream processes. The development of a
chemical synthesis route for the production of PLA has stimulated
great interest among researchers because of both, the low cost and
environmental impact [5].
Nowadays, some researchers have been progressed in the con-
version of glycerol into lactic acid by chemical catalysis [6]. Kishida
et al. [7], reported that glycerol can be converted into lactic acid in
The potential of lactic acid (LA) and its importance as a product
in the chemical platform is clearly visible in the chemical market.
The LA world production increased from 40,000 t/year in the 1990s
to 260,000 t/year in 2008, and these rates are expected to increase
substantially. The Plastics Today magazine (2011) predicted an LA
◦
an alkaline medium and hydrothermal conditions (at 300 C), with
NaOH as homogeneous catalyst. Shen et al. [8], reported the effect of
different alkali and alkaline earth hydroxides in glycerol hydrother-
◦
mal conversion into lactic acid at 300 C, showing that the alkali
metal hydroxides were more effective catalysts than alkaline earth
metal hydroxides. Ramirez-Lopez et al. [9] studied the use of high
glycerol concentrations and equivalent amounts of NaOH, at 280 C,
∗ Corresponding author.
E-mail addresses: rodrigo@gpsa.ufc.br, rodrigo@gpsa.ufc.br (R.S. Vieira).
◦
http://dx.doi.org/10.1016/j.cattod.2016.02.015
0
920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
2
obtaining a yield of 84.5% to LA. This study showed that a high
basic concentration to catalyze the conversion using a high glyc-
erol concentration is required. However, due to the high alkalinity
and elevated temperature, it is necessary to use specific reactors,
to avoid high corrosiveness and relevant concerns on an industrial
scale.
According to Chen et al. [10], the use of solid catalysts includes
advantages such as, low corrosivity, reuse of the catalyst, low
temperature reaction and easily of separation of catalyst/product,
thereby reducing chemical wastes.
[21]. To this end, 1 g of pretreated carbon was added to a solution
containing the platinum precursor desired concentration under
stirring at 50 C and reduced with formaldehyde (37%) for 1 h at the
same temperature. The obtained catalyst was recovered by vacuum
filtration, followed by washing with Ultrapure water and drying
at room temperature. The catalysts were denoted as xPt/C-Fresh
or Used, for fresh and spent catalysts, respectively, where x is the
weight percent of platinum (1, 2.5 and 5).
◦
2.3. Characterization
The conversion of glycerol using mainly supported metal
nanoparticles has been extensively studied to obtain other acids
The X-ray powder diffraction (XRD) powder patterns of the Pd/C
and Pt/C catalysts were recorded on a diffractometer (PANalytical
modelo EMPYREAN) using Cu K␣ radiation (␭ = 1.54056 Å), scan-
(
acetic acid, glycolic acid, glyceric acid, formic acid) and have shown
that lactic acid can be generated from glycerol using heterogeneous
catalysis [11]. Some researchers have reported the formation of
lactic acid by using supported catalysts such as: Ru/C, Pt/C, Au/C,
PtRu/C and AuRu/C [12,13]; Au/TiO Pt/TiO , Pd/TiO and AuPt/TiO
◦
ning from 10 to 80 (2␪). The phases were identified according to
Joint Committee on Powder Diffraction Standards (JCPDS) database.
Scanning electron microscopy (SEM) was performed on a scan-
ning electron microscope (INSPECT 50) operated at an acceleration
voltage of 20 kV to characterize the morphology of Pd/C and Pt/C
particles. Previously, the samples were deposited on an aluminum
sample holder and sputtered with gold and then SEM measure-
ments were conducted.
Transmission Electron Microscope (TEM) images were obtained
using a Philips CM 200 Supertwin-174 DX4 microscope operated at
an acceleration voltage of 200 kV to characterize the morphologies
and the crystal structures of the Pd and Pt nanoparticles supported
on carbon. Samples were dispersed in ethanol and a drop of the
dispersion was put on a Cu grid (300 mesh).
2
2
2
2
[
14]; Cu/SiO2 e CuO/Al O3 [15]; Pt/CaCO3 [16]; Rh/C e Ir/C [17];
2
and Au/n-CeO , Pt/n-CeO2 and AuPt/n-CeO2 [18], and Pd/C [19].
2
In this case, the use of supported catalysts can improve the cat-
alytic performance, tailoring not only the catalytic activity but also
the selectivity for the target molecule. In this context, the aim of
this study was to investigate the catalytic performance of Pd/C and
Pt/C in glycerol conversion into lactic acid. The process conditions
such as catalyst loading, catalyst weight, NaOH/glycerol molar ratio,
temperature and reaction time were systematically studied. The
catalysts were characterized by a variety of experimental tech-
niques in order to better understand the factors governing the
catalytic activity of the studied samples.
The textural properties were experimentally determined by N2
◦
physisorption at −196 C using an Micromeritics ASAP 2020 ana-
◦
lyzer. The samples were previously degassed at 200 C at a vacuum
of 10 bar. The pore diameter was calculated according to the for-
mula: 4 V/A.
2
. Experimental
−
5
2.1. Materials
X-ray photoelectron spectroscopy (XPS) analyses were per-
formed with a spectrometer Physical Electronics 5700 using a
Mg-K␣ source (1253.6 eV) (model 04–548 Dual Anode X-ray
Source). The X-ray source was run at a power of 300 W (10 keV and
The chemicals used, palladium(II) chloride (PdCl -Sigma-
2
Aldrich), chloroplatinic acid hexahydrate (H PtCl ·6H O-Sigma-
2
6
2
Aldrich), sodium hydroxide (NaOH-Vetec-Brazil), sulfuric acid
H SO -Synth), hydrochloric acid (HCl-Synth), nitric acid (HNO -
3
0 mA). All spectra were obtained using a 720 m diameter analysis
(
2
4
3
◦
area. The specimens were analyzed at an angle of 45 to the surface
plane. The X-ray source was located at 54 relative to the analyzer
axis at 5.10
Synth), sodium carbonate (Na CO - Dynamic Contemporary
2
3
◦
Chemistry), formaldehyde (Vetec-Brazil), glycerol (Vetec-Brazil),
lactic acid (Sigma-Aldrich) and activated carbon (Vetec-Brazil)
were of analytical grade. Ultrapure water (Milli-Q System, Milli-
pore) was used in all experiments.
−
10
Torr of vacuum.
The palladium and platinum contents in the catalysts were
determined by inductively coupled plasma optical emission spec-
trometer (ICP-OES) (Thermo Fischer Scientific, iCAP 6000 model).
The standard calibration was carried out using Pd and Pt aqueous
2.2. Catalyst preparation
solution, prepared in HNO solution 2% (v/v), in concentrations of
3
−
1
0
.1, 0.5, 1, 5 e 10 mg L . The reaction samples, after one reaction
Palladium catalyst supported on activated carbon was prepared
cycle, were dissolved in deionized water. The wavelength used to
determine Pd and Pt was 340.4 and 214.4 nm, respectively. The
samples were measured five times and the results showed are the
average.
using PdCl as metal precursor, according to the method developed
2
by Kubota et al. [20]. Thus, in a typical synthesis, the activated
carbon sample was pretreated with a HNO solution (10% v/w –
3
−
1
◦
5
mL g ) at 80 C under stirring for 3 h in order to obtain a func-
tionalized surface. The palladium catalysts containing 2.5, 5 and
0 wt% of total metallic charge were prepared with solutions con-
taining PdCl , where 12 mL of 1 M HCl were added under stirring.
2.4. Catalytic tests
1
The reactions were carried out in a 300 mL capacity stainless
steel Parr reactor equipped with a mechanical stirrer (700 rpm).
In a typical experiment, several catalyst weights (0.2, 0.4, 0.6 and
0.8 g) were employed. 100 mL aqueous solution of glycerol and an
aqueous solution of NaOH (NaOH: glycerol molar ratio = 0.75, 1.1
and 1.25) were charged into the reactor. The reaction was car-
2
1
.0 g of the pretreated carbon was added to the prior solution and
◦
stirred for 20 h at 30 C. The solids were recovered by filtration,
mixed with a Na CO solution, maintained under stirring for 3 h
2
3
◦
at 30 C and reduced with formaldehyde (37%) under stirring for
h at 60 C. The obtained catalyst was recovered by vacuum fil-
◦
2
◦
tration, followed by washing with Ultrapure water and drying at
room temperature for 16 h. The catalysts were denoted as xPd/C-
Fresh or Used, for fresh and spent catalysts, respectively, where x
is the weight percent of metal (2.5, 5 and 10).
ried out at 200 or 230 C and the reaction time was 4 h. After that
time, the catalyst was separated by vacuum filtration. The reaction
parameters were investigated: metal loading (%), catalyst mass (g),
NaOH:glycerol molar ratio, temperature and reaction time.
The reaction products were analyzed by comparison with stan-
dard samples, using a HPLC Shimadzu equipment using a RI
Platinum catalysts supported on carbon were prepared using
H PtCl ·6H O, following the methodology adapted by Liang et al.
2
6
2
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
3
mol of desired product × 100
AC
%
%
SLA
=
(2)
5
5
1
1
% Pt/C-Fresh
molGlyin
% Pt/C-Used
Y = %
X × %S
(3)
0% Pd/C-Fresh
0% Pd/C-Used
LA
100
3. Results and discussion
3.1. The X-ray powder diffraction (XRD) analysis
Pd (111)
Fig. 1 shows the X-ray diffraction patterns of the carbon support
C), as-synthesized, and the fresh and used catalysts (Pd/C-Fresh,
(
Pd/C-Used, Pt/C-Fresh, Pt/C-Used). An important parameter in the
structural characterization of coals is the interlayer spacing of the
stacks (d002), which reflects the distance between the graphene
layers in the stacks of the coal. According to Takagi et al. [22], the
value of d002 is related to the perfection in the stacking structure
periodicity, a decrease in this value (d002 for the graphite structure
is 0.335 nm), is indicative of a more ordered structure.
Pt (111)
C (002)
◦
The broad band located between 2␪ = 20–30 is characteristic
2
0
30
40
Theta (º)
50
60
of amorphous materials. In both diffractograms (activated carbon
samples and Pd/C, Pt/C fresh and used catalysts) can be visual-
2
◦
ized a peak at 2␪ = ∼24.1 related to the diffraction plane (0 0 2)
Fig. 1. X-ray diffraction patterns of support (AC), fresh catalysts (Pd/C-Fresh, Pt/C-
Fresh) and used catalysts (Pd/C-Used, Pt/C-Used).
of the hexagonal graphitic carbon according the JCPDS card no.
7
5–1621. This result suggests that the material has a low degree of
crystallinity with hexagonal graphite belonging to P6 /mmc space
3
◦
group with broad peaks at 2␪ = 20–30 (0 0 2). These reflections are
likely associated with sp -bonding carbon lacking in periodicity in
detector. The system was equipped with a Biorad Aminex HPX-87H
2
◦
−1
column operating at 60 C and 5 mmol L of H SO4 solution were
2
the C-axis, which is consistent with graphite [23].
−
1
used as eluent. Samples were eluted at a flow rate of 0.6 mL min
and the injection volume was 20 ␮L. Concentrations were deter-
mined using calibration curves obtained by injecting standard
solutions of known concentrations (0–18 g L of glycerol and lactic
From Fig. 1 it is noted that the X-ray diffractograms of fresh
catalysts (Pd/C and Pt/C) are similar to the X-ray diffraction of acti-
vated carbon. It could be seen that neither the acid modification
nor the introduction of palladium or platinum metals changed the
structure of the activated carbon. The diffractogram correspond-
ing to 10% Pd/C-Fresh did not show diffraction peaks related to the
−1
acid).
The glycerol conversion (%X), selectivity to lactic acid (%SLA) and
lactic acid yield (%Y) were calculated according to the following
equations:
◦
tetragonal phase PdO [24] and Pd cubic [25]. This suggests that
the particles of palladium should be well dispersed on the sup-
port surface, with a low particle size or in low concentration to
be detected by X-ray diffraction [26]. For the 10% Pd/C-Used cat-
%
X = (
molGlyin − moGlyout) × 100
(1)
molGly
◦
in
alyst a significant peak was observed, at 2␪ = 40.1 (1 1 1) (JCPDS
Fig. 2. SEM images: Activated Carbon (A), Pd/C-Fresh (B), Pd/C-Used (C), Pt/C-Fresh (D), Pt/C-Used (E) at magnification of 3000×.
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
4
Fig. 3. TEM images: Activated Carbon (A1–3), Pd/C-Fresh (B1–3), Pd/C-Used (C1–3), Pt/C-Fresh (D1–3), Pt/C-Used (E1–3) at different magnifications.
card no. 62-2867), characteristic of metallic palladium. The Pt peak
or with a low degree of crystallinity. A different XRD pattern was
◦
◦
(
2␪ = ∼39 ) was also absent in the XRD of 5% Pt/C-Fresh, suggesting
observed for the used catalyst, showing a peak at 2␪∼39 , assigned
that the Pt species were uniformly dispersed as minuscule grains
to the (1 1 1) reflection of platinum (JCPDS Card no. 04-0802). It is
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
5
Fig. 4. N2 adsorption-desorption isotherms corresponding to 10% Pd/C (a) and 5% Pt/C (b) before and after the catalytic reaction.
Table 1
◦
a diffraction peak (2 theta = 40.1 ), related to the reflection plane
Textural properties of the support, fresh and spent catalysts: BET Surface Area (SBET),
pore volume (Vp) and average pore diameter (Dp).
◦
of Pd . However, this sintering phenomenum was not considered
significant since there was no influence on the catalytic activity for
the studied material. The EDX analyses confirmed the presence of
Pd with approximately 4.96% ± 0.857% in weight on the surface of
the support. This value is approximately about 50% of the exper-
imental loading (10% Pd/C). The particle size distribution shown,
without considering the agglomerates observed in Fig. 3 C2 and 3,
that the particle size slightly increased.
SBET (m2
g
−1
)
VP (cm3
g
−1
)
DP (nm)a
Sample
Activated Carbon
779
758
534
388
166
0.60
0.58
0.43
0.36
0.24
3.0
3.0
3.2
5.1
5.5
10% Pd/C-Fresh
10% Pd/C-Used
5% Pt/C-Fresh
5% Pt/C-Used
a
Dp: 4 V/A.
If Pt catalyst is considered, very small particles were observed in
the catalysts 5% Pt/C-Fresh and Used. Barrabés et al. [27] observed
that Pt particles were not visible for 0.5% Pt/CeO2 prepared by the
impregnation method, suggesting that the metal particles could be
smaller than 1 nm. The particle size distribution revealed a great
proportion of particles in the range 0.5–1.5 nm. After the catalytic
process (Fig. 3E1–3), the presence of larger particles was observed,
and the particle size distribution was broader and moved to higher
values. However, the sintering process was less evident than in
the case of Pd sample. These data could explain the stability pre-
sented by this sample (see below), used in five successive reaction
cycles. On the other hand, the amount of Pt calculated by EDX anal-
ysis, was about 3.8% (± 0.635%) in weight on the support surface.
This amount is approximately about 60% of the experimental load
interesting to note that these results agree with TEM observations,
which suggest that nanoscopic particles embedded in an amor-
phous matrix can clearly be observed for the used catalyst (See
below).
3.2. Microscopic analyses
Fig. 2 shows the SEM micrographs corresponding to 10% Pd/C
catalyst (Fig. 2B and C) and 5% Pt/C catalyst (Fig. 2D and E) with
magnifications of 3000 times, before and after reaction. The acti-
vated carbon particles have regular tubular porosity, probably from
the vascular structure due to carbon source (Fig. 2A). There was not
change in the carbon structure after Pd or Pt deposition (Fig. 2B and
D) probably due to the mild deposition conditions used.
(5% Pt/C). According to Marques et al. [19] this difference about
metal loading can be explained due to the low reactivity of acti-
vated carbon surface, despite the acid wash procedure carried out
briefly.
It is noteworthy that the spent catalysts (Fig. 2C and E), despite
◦
the high reaction temperature (200 and 230 C), the carbon struc-
ture remained stable after the reaction, probably due the chemical
and physical activated carbon resistance.
TEM images can directly give us information about the dimen-
sion and dispersion of the metallic nanoparticles. TEM micrographs
for activated carbon, Pd/C and Pt/C catalysts are shown in Fig. 3.
Dark particles uniformly dispersed on the carbon support can be
observed, which can be considered embedded metal particles. The
results indicate that Pd nanoparticles are well dispersed, with a
narrow size distribution in the range of 0.5–2.5 nm as observed
in the corresponding histogram, concluding that the methodol-
ogy used for the catalyst synthesis of Pd/C led to the presence
of homogeneously distributed Pd nanoparticles. This dispersion
is in good agreement with the conclusion drawn from the XRD
results for 10% Pd/C-Fresh, where the diffraction peaks of metallic
Pd were not observed. In the case of the used sample, 10% Pd/C-Used
3.3. Textural properties
The textural properties of the prepared samples were eval-
uated from N2 adsorption-desorption isotherms at −196 C. The
◦
isotherms are represented in Fig. 4. The isotherms obtained are
intermediate between type II and IV (according to the IUPAC clas-
sification) with the absence of the plateau at high pressures. The
hysteresis loop is of type H3 characteristic of mesoporous acti-
vated carbon. Moreover, both samples show a great adsorption at
low relative pressures, indicating that micropores are also present.
This fact is much more important for Pd/C sample, and indicating
that carbon support is less affected by Pd incorporation than by
Pt. After testing, both samples suffer a decrease of the volume at
low pressures and therefore indicating that micropores are those
affected by noble metal incorporation. The hysteresis loop hardly
changed, indicating that micropores are those affected by the metal
(Fig. 3C1–3), it is also observed the existence of some Pd particles
agglomerated, indicating some sintering. These results are again in
agreement with the XRD analysis, which showed the presence of
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
6
Pd 3d core level spectra
Pt 4f core level spectra
7
1.5
336.0
Used
Spent
337.7
Spent
Used
7
2.4
Fresh
Fresh
7
8
77 76 75 74 73 72 71 70
Binding Energy (eV)
3
46 344 342 340 338 336 334
Binding Energy (eV)
(a)
(b)
Fig. 5. (a) Pd 3d core level spectra for samples Pd/C-Fresh and Used. (b) Pt 4f core level spectra for sample Pt/C-Fresh and Used.
Fig. 6. Proposed reaction mechanism of Glycerol conversion [18].
deposition-sintering in accordance to the low particle size observed
for these samples.
Fig. 5b shows the Pt 4f XPS spectra for the 5% Pt/C-Fresh and Used
samples. The decomposition of the spectra gives rise to one doublet
in both cases: 4f7/2 component-solid line and 4f5/2 component-
dotted line. The Pt 4f7/2 component is located at 72.4 eV for the
Table 1 shows the values of the textural properties before and
after use. Comparing the values of area BET, it is noted a reduc-
tion of the catalysts compared to the bare support (Pt/C-Fresh
approximately 43% and Used approximately 21.3%) and (Pd/C-Fresh
approximately 97.3% and Used approximately 68.5%), caused by
clogged micropores by metal deposition. Moreover, a decrease is
also noticed after testing. This reduction is accompanied with a
reduction in the pore volume, indicative of the accumulation of
reactants and products after the reaction process inside the pores.
2+
fresh sample and at 71.5 eV for the used one and attributed to Pt
◦
and Pt , respectively, suggesting that the Pt species were present
as Pt2+ in the fresh sample and reduced during the course of the
reaction [25,32].
4. Catalytics tests
3.4. X-ray photoelectron spectroscopy
As stated before, the prepared catalysts were studied in the
In order to know the chemical states of the palladium and
reaction of conversion of glycerol into lactic acid. Fig. 6 shows
the possible reaction mechanism proposed by Purushothamana
et al. [18]. The reaction mechanism involves the initial oxidative
dehydrogenation of glycerol to glyceraldehyde. In basic condi-
tions, glyceraldehyde is in equilibrium with dihydroxyacetone. It is
well known that trioses like glyceraldehyde and dihydroxyacetone
undergo rearrangement into lactic acid under alkaline conditions.
An optimal catalyst for the conversion of glycerol into lactic acid
therefore should have a strong oxidative dehydrogenation capacity
under mild conditions and at the same time being a highly ineffi-
cient oxidation catalyst for the conversion of glyceraldehyde into
glyceric acid and subsequent oxidation products [18].
platinum on 10% Pd/C and 5% Pt/C catalysts, the surfaces were char-
acterized by XPS analysis (Fig. 5a and b). The Pd 3d signal is formed
by two doublets Pd 3d5/2 and Pd 3d3/2 (Fig. 5a). The Pd 3d5/2 signal
is composed of two contributions at 336.0 and 337.7 eV assigned
to Pd and Pd , respectively [28–30]. These are the typical valence
states for Pd and the peaks values are comparable with the liter-
ature. According to the deconvoluted profiles for fresh and used
Pd catalysts, shown in Fig. 5, Pd would be present in two differ-
ent oxidation states. A fraction of the metal would be as PdCl2
◦
2+
(
337–338 eV) [31]. This percentage is very similar in the fresh and
used catalyst, which indicates that the presence of palladium chlo-
ride is related to the precursor used in the catalyst manufacture.
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
7
1
00
100
(b)
(a)
8
6
4
2
0
0
0
0
0
80
60
40
20
0
2
.5
5.0
7.5
10.0
0
1
2
3
4
5
Palladium loading (%)
Platinum loading (%)
Fig. 7. Effect of metallic loading on glycerol conversion% (᭿), selectivity% to LA ( ) and Yield% of LA ( ). (a) Effect of palladium loading (0.2 g of catalyst) and (b) Effect of
−
1
−1
◦
platinum loading. (0.5 g of catalyst). Reaction conditions: 100 mL of 0.5 mol L glycerol solution, 0.55 mol L NaOH, 230 C, 3 h.
(
a)
(b)
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
0
.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Amount of catalyst 10% Pd/C (g)
Amount of catalyst 5% Pt/C (g)
Fig. 8. Effect of catalyst amount (g) on glycerol conversion% (᭿), selectivity% to LA (ᮀ) and Yield% of LA ( ). (a) 10% Pd/C catalyst and (b) 5% Pt/C catalyst. Reaction conditions:
−
1
−
1
◦
1
00 mL of 0.5 mol L glycerol solution, 0.55 mol L NaOH, 230 C, 3 h.
1
00
100
80
60
40
20
0
(
a)
(b)
8
6
4
2
0
0
0
0
0
0
.7
0.8
0.9
1.0
1.1
1.2
1.3
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Molar ratio NaOH:Glycerol
Molar ratio NaOH:Glycerol
Fig. 9. Effect of NaOH/glycerol molar ratio on glycerol conversion% (᭿), selectivity% to LA ( ) and yield% of LA ( ). (a) 10% Pd/C (0.2 g of catalyst) and (b) 5% Pt/C (0.4 g of
−
1
◦
catalyst). Reaction conditions: 100 mL of 0.5 mol L glycerol solution, 230 C, 3 h.
4
.1. Effect of Pd and Pt loading
of glycerol showed a gradual increase with increasing the metallic
loading on the support. The results showed that at the highest load-
ing of Pd and Pt the highest conversions were attained, 93.8% and
The effect of metal loading on the support (activated carbon)
is shown in Fig. 7a (palladium) and b (platinum). The conversion
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
8
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
100
100
8
6
4
0
0
0
8
6
4
2
0
0
0
0
0
20
0
0
30 60 90 120 150 180 210 240 270
Time (min)
0
30 60 90 120 150 180 210 240 270
Time (min)
◦
◦
Fig. 10. Effect of reaction temperature 200 C ( ), 230 C (᭿) and reaction time on: (a) Glycerol conversion and (b) selectivity to LA for 10% Pd/C catalyst. Reaction conditions:
1
−
1
−
1
00 mL of 0.5 mol L glycerol solution, 0.55 mol L of NaOH, 0.2 g of catalyst.
9
6
9.2%, respectively. Meanwhile, the selectivity to lactic acid was
1.9% for 10% Pd/C and 74.9% for 5% Pt/C one.
Therefore, to study other experimental parameters, 10% Pd and
% Pt containing catalysts were employed.
lactic acid conversion, concluding that alkali-metal hydroxides
were more effective than alkaline-earth-metal hydroxides. A glyc-
◦
erol conversion of ∼90% was obtained at 300 C with NaOH or KOH
5
as an homogeneous catalyst.
Ramirez-Lopez et al. [9] reported a glycerol conversion of ∼98%
◦
achieved at 280 C with a 1.1 NaOH/glycerol molar ratio, but using
4.2. Effect of the catalyst amount (g)—10% Pd/C and 5% Pt/C
a higher glycerol concentration of 2.5 M.
In this study, the performance presented by 10% Pd/C and 5% Pt/C
catalysts are promising since high conversion values were attained
but at lower reaction temperature and lower glycerol concentration
The effect of catalyst amount is shown in Fig. 8a (10% Pd/C), and
b (5% Pt/C). It is seen that by increasing the amount of 10% Pd/C
catalyst from 0.2 to 0.8 g, the conversion (93.9–99.6%) and selectiv-
ity to LA (62.6–62.6%) did not change significantly. A similar trend
was observed for the yield (58.8–62.4%). In the case of 5% Pt/C cat-
alyst, a similar behavior was observed. By increasing the amount
of catalyst from 0.4 to 0.8 g, it did not provoke a significant change
in the conversion values (99.8–99.4%). However the selectivity and
yield of lactic acid decreased, 62.9–54.2% and 57.1–53.9, respec-
tively. Therefore, 0.2 g was the amount of Pd catalyst and 0.4 g that
Pt one, for the following experiments.
(0.5 M) than the values found in literature. In a previous study on
the synthesis of lactic acid from glycerol by using 5% Pd/C, it was
obtained 95% of glycerol conversion with 1.1 NaOH/glycerol molar
ratio [19].
High selectivity values were also observed with the
NaOH/glycerol ratio equal to1.1, 68% (10% Pd/C) and 74% (5%
Pt/C). Increasing the molar ratio from 1.1 to 1.25, there was a
decrease in the selectivity to lactic acid (LA). According to Marques
et al. [19] this behavior indicates that by increasing the molar ratio
NaOH/glycerol it is favored the reactions involving the formation
of other lactic acid products or the degradation of lactic acid
preformed.
Moreover, the presented results are promising since it is
claimed in literature, the development of active catalysts capable
of operating under less corrosive conditions, such as lower OH
concentration as well as lower reaction temperatures to minimize
the severe corrosion of stainless steel reactors, making the process
suitable for use in industrial production.
4
.3. Effect of molar ratio NaOH/glycerol (Pd/C and Pt/C)
Dusselier et al. [33] explain that generally the reaction mecha-
nism for the conversion of glycerol into lactic acid formally requires
an oxidation and a dehydration/rehydration step (as presented
in the reaction mechanism in Fig. 6). The oxidation may be per-
formed by a dehydrogenation under reductive or inert conditions,
by oxidation with oxidants or electrochemically. Dehydration usu-
ally requires acidic conditions, but many glycerol reactions have
been reported in alkaline media.
According Roy et al. [15] the base accelerates lactic acid for-
mation due to two factors (1) favoring the transformation of
pyruvaldehyde to lactic acid, and (2) shifting the equilibrium
toward lactic acid formation (by reducing the lactic acid concen-
tration as a result of Na-lactate formation and lactate formation
reduces the effective base concentration in the reaction mixture).
The effect of NaOH/glycerol molar ratio is shown in Fig. 9a
4.4. Effect of the reaction time and temperature and catalyst
stability
The influence of temperature on the conversion and yield of the
lactic acid was studied at two temperatures (200 and 230 C) for
4 h with a NaOH/glycerol = 1.1, with 10% Pd/C and 5% Pt/C catalysts
(Figs. 10 and 11).
◦
(
10% Pd/C) and b (5% Pt/C). Using a molar ratio of 0.75 the val-
The temperature is expected to affect the selectivity of the
reaction, since the selectivity to lactic acid is improved at high
temperatures as showed in Figs. 10 and 11, for both catalysts.
Ftouni et al. [34] studied the conversion of glycerol to lactic acid
ues of glycerol conversion were of 86% and 90%, using 10% Pd/C
and 5% Pt/C catalysts, respectively. By increasing the ratio until
1
.25, the conversions achieved were higher and approximately
the same for both samples, 93%. However, the highest glycerol
conversion was obtained for a molar ratio of 1.1, 96% and 99%
for 10% Pd/C and 5% Pt/C catalysts, respectively. According to
Shen et al. [8], that investigated the effects of different bases as,
alkali metals and alkaline-earth metals on hydrothermal glycerol to
using Pt/ZrO catalyst, however also detected the presence of other
2
reaction products such as formic acid, 1.2-PDO, 1.3-PDO, ethylene
◦
glycol, acetic acid and ethanol in small amounts, at 180 C and under
the He pressure of 30 bars. In the present study, the possible pres-
ence of the products listed in the previous study was not detected,
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
9
100
100
8
6
4
0
0
0
8
6
4
2
0
0
0
0
0
20
0
0
30 60 90 120 150 180 210 240 270
Time (min)
0
30 60 90 120 150 180 210 240 270
Time (min)
◦
−1
◦
Fig. 11. Effect of reaction temperature 200 C ( ), 230 C (᭿) and reaction time on: Glycerol conversion and selectivity to LA for 5% Pt/C catalyst. Reaction conditions: 100 mL
−
1
of 0.5 mol L glycerol solution, 0.55 mol L of NaOH, 0.4 g of catalyst.
(
a)
(b)
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
1
2
3
4
5
1
2
3
4
5
Cycle number
Cycle number
−
1
◦
Fig. 12. Effect of repeated use of catalyst on reaction performance. Reaction conditions: (a) 100 mL of 0.5 mol L glycerol solution, 0.2 g of catalyst 10% Pd/C, 230 C, 3 h. (b)
1
−
1
◦
00 mL of 0.5 mol L glycerol solution, 0.4 g of catalyst 5% Pt/C, 230 C, 3 h. Conversion% (᭿), selectivity% (ᮀ) and Yield% ( ).
◦
since that temperature used was 230 C, which probably was not
favorable to the production of other compounds in addition to lac-
tic acid. Ramirez-Lopez et al. showed that elevated temperatures
acid selectivity, no significant differences in five consecutive cycles
(Fig. 12).
ICP analyses were carried out to determine the Pd and Pt amount
content in the fluid phase, after glycerol reaction. These analy-
ses were performed in order to investigate if the leaching process
occurred. The Pd and Pt content in fluid phase after the glycerol
◦
(
>277 C) are required in the hydrothermal process for converting
glycerol to lactic acid. However, the decomposition of lactic acid
and pyruvaldehyde is significant at this temperature, negatively
affecting the selectivity to lactic acid.
◦
(230 C, 3 h) was 0.822 ± 0.0003 mg/L (which corresponds to 0.05%
With regard to the reaction time, both samples, at both reaction
temperatures studied, showed an increased conversion and selec-
tivity to LA. According to Perosa and Tundo [35], such behavior is
related to the disposition of the interfacial area of the catalyst in
the reaction medium, directly affecting the overall mass transfer
coefficient of the system and consequently the reaction rate. The
glycerol consumption is also accelerated by increasing the temper-
ature, since this is an endothermic reaction and therefore favored
at high reaction temperatures, because under mild conditions there
is a difficulty to remove the OH groups of glycerol and oxidize the
glycerol to glyceraldehyde (or dihydroxyacetone) [19] in the first
reaction step according to reaction mechanism suggested in Fig. 6.
After the completion of the reaction, the catalysts could be easily
recovered by filtration followed by washing with water and reused
for subsequent runs. The recovered catalysts showed an efficient
recycling ability without giving any change in the reaction time
and the yield of the product. Suggesting an efficient stability of
the catalytic structure with values of glycerol conversion and lactic
(w/w) of Pd charged in the catalyst) and 0.117 ± 0.0007 mg/L (cor-
responds to 0.002% (w/w) of Pt charged in the catalyst).
5. Conclusions
Palladium and platinum-based catalysts supported on activated
carbon are efficient catalysts for the conversion of glycerol into
lactic acid in aqueous solution in the presence of a base and at a tem-
◦
perature of 230 C. The catalytic results have revealed that Pd and Pt
content and reaction conditions have significant effects on the cat-
alytic performance; meanwhile the catalyst weight only influenced
the selectivity to LA in the case of Pt catalysts. Thus only 0.2 g of
Pd/C and 0.4 g of Pt/C were enough to achieve high level of Glycerol
conversion and selectivity to LA. Thus, LA selectivity of up to 68%
and 74% were obtained under the optimum conditions for 10% Pd/C
and 5% Pt/C, respectively, with glycerol conversions of about 99%.
The reusability of the prepared samples was also high since they
hardly lost activity after 5 cycles. Compared to the hydrothermal
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015

G Model
CATTOD-10045; No. of Pages10
ARTICLE IN PRESS
M.R.A. Arcanjo et al. / Catalysis Today xxx (2016) xxx–xxx
1
0
process, this catalytic process requires of lower temperature, lower
glycerol/alkali ratios and no need of oxygen or hydrogen addition.
[16] J. ten Dam, F. Kapteijn, K. Djanashvili, U. Hanefeld, Catal. Commun. 13 (2011)
–5.
1
[
17] F. Auneau, S. Noël, G. Aubert, M. Besson, L. Djakovitch, C. Pinel, Catal.
Commun. 16 (2011) 144–149.
Acknowledgements
[18] R.K.P. Purushothamana, J. van Haveren, D.S. van Es, I. Melián-Cabrera, J.D.
Meeldijk, H.J. Heeres, Appl. Catal. B: Environ. 147 (2014) 92–100.
[
19] F.L. Marques, A.C. Oliveira, J. Mendes Filho, E. Rodríguez-Castellón, C.L.
Cavalcante Jr., R.S. Vieira, Fuel Process. Technol. 138 (2015) 228–235.
The authors wish to acknowledge financial support from
CAPES/Funcap (E1-0079-0004301), CAPES/DGU International col-
laboration grant Brazil/Spain, Project P12-RNM (Junta de Andalucía,
Spain) and MINECO project CTQ2015-68951-C3-3R and FEDER
funds and Central Analítica-UFC/CT-INFRA/MCTI-SISNANO/Pró-
Equipamentos.
[20] T. Kubota, H. Ogawa, Y. Okamoto, T. Misaki, T. Sugimura, Appl. Catal. A: Gen.
37–438 (2012) 18–23.
21] D. Liang, J. Gao, J. Wang, P. Chen, Z. Hou, X. Zheng, Catal. Commun. 10 (2009)
586–1590.
4
[
1
[22] H. Takagi, K. Maruyama, N. Yoshisawa, Y. Yamada, Y. Sato, Fuel 83 (2004)
2427–2433.
[
23] J.C.S. Araujo, C.B.A. Sousa, A.C. Oliveira, F.N.A. Freire, A.P. Ayala, A.C. Oliveira,
Appl. Catal. A: Gen. 377 (2010) 55–63.
References
[24] F.Z. Hamida, J. Barbier Jr., S. Labruquere, D. Duprez, Appl. Catal. B: Environ. 29
2001) 195–205.
(
[
[
1] R. Moita, A. Freches, P.C. Lemos, Water Res. 58 (2014) 9–20.
2] C. Phithakrotchanakoon, V. Champreda, S. Aiba, K. Pootanakit, S.
Tanapongpipat, J. Polym. Environ. 23 (2015) 38–44.
3] R.L. Manfro, M.M.V.M. Souza, Catal. Lett. 144 (2014) 867–877.
4] G. Dreschke, M. Probst, A. Walter, T. Pümpel, J. Walde, H. Insam, Bioresour.
Technol. 176 (2015) 47–55.
[25] Y. Li, S. Chen, J. Xu, H. Zhang, Y. Zhao, Y. Wang, Z. Liu, Clean Soil Air Water 42
(2014) 1140–1144.
[26] W.H. Cassinelli, L.S.F. Feio, J.C.S. Araújo, C.E. Hori, F.B. Noronha, C.M.P.
Marques, J.M.C. Bueno, Catal. Lett. 120 (2008) 86–94.
[
[
[27] N. Barrabés, K. FD ottinger, A. Dafinov, F. Medina, G. Rupprechter, J. Llorca,
Appl. Catal. B: Environ. 87 (2009) 84–91.
[
5] Y. Feng, H. Yin, A. Wang, D. Gao, X. Zhu, L. Shen, Appl. Catal. A: Gen. 482
[28] M. Chatterjee, T. Ishizaka, H. Kawanami, J. Colloid Interface Sci. 420 (2014)
15–26.
(
2014) 49–60.
[
[
6] L. Liu, X.P. Ye, Fuel Process Technol. 137 (2015) 55–65.
7] H. Kishida, F. Jin, Z. Zhou, T. Moryia, H. Enomoto, Chem. Lett. 34 (2005)
[29] A. Infantes-Molina, J. Mérida-Robles, E. Rodríguez-Castellón, J.L.G. Fierro, A.
Jiménez-López, Appl. Catal. B: Environ. 73 (2007) 180–192.
[30] E. Vilarrasa-García, A. Infantes-Molina, R. Moreno-Tost, E.
Rodríguez-Castellón, A. Jiménez-López, C.L. Cavalcante Jr., D.C.S. Azevedo,
Energy Fuels 24 (2010) 3436–3442.
[31] D. Briggs, M.P. Seah, Practical Surface Analysis-Auger and X-ray Photoelectron
Spectroscopy, 2nd ed., Wiley Interscience, 1990.
[32] N.R. Elezovic, V.R. Radmilovic, J. Kovac, B.M. Babic, Lj M. Gajic-Krstajice, N.V.
Krstajic, RSC Adv. 5 (2015) 15923–15929.
[33] M. Dusselier, P.V. Wouwe, A. Dewaele, E. Makshina, B.F. Sels, Energy Environ.
Sci. 6 (2013) 1415–1442.
[34] J. Ftouni, N. Villandier, F. Auneau, M. Besson, L. Djakovitch, C. Pinel, Catal.
Today 257 (2015) 267–273.
1
560–1561.
[
[
8] Z. Shen, F. Jin, Y. Zhang, B. Wu, A. Kishita, K. Tohji, H. Kishida, Ind. Eng. Chem.
Res. 48 (2009) 8920–8925.
9] C.A. Ramirez-Lopez, J.R. Ochoa-Gómez, M. Fernández-Santos, O.
Goméz-Jiménez-Aberasturi, A. Alonso-Vicario, J. Torrecilla-Soria, Ind. Eng.
Chem. Res. 49 (2010) 6270–6278.
[
[
10] L. Chen, S. Ren, S.P. Ye, Fuel Process. Technol. 120 (2014) 40–47.
11] P. Lakshmanan, P.P. Upare, N.T. Le, Y.K. Hwang, D.W. Hwang, U.-H. Lee, H.R.
Kim, J.-S. Chang, Appl. Catal. A: Gen. 468 (2013) 260–268.
[
[
[
[
12] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
13] E.P. Maris, W.C. Ketchie, M. Murayama, R.J. Davis, J. Catal. 251 (2007) 281–294.
14] Y. Shen, S. Zhang, H. Li, Y. Ren, H. Liu, Chem. Eur. J. 16 (2010) 7368–7371.
15] D. Roy, B. Subramaniam, R.V. Chaudhari, ACS Catal. 1 (2011) 548–551.
[35] A. Perosa, P. Tundo, Ind. Eng. Chem. Res. 44 (2005) 8535–8537.
Please cite this article in press as: M.R.A. Arcanjo, et al., Conversion of glycerol into lactic acid using Pd or Pt supported on carbon as
catalyst, Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.02.015