From glycerol to lactic acid under inert conditions in the presence of platinum-based catalysts: The influence of support

Article abstract of DOI:10.1016/j.cattod.2014.09.034

In this work, it was shown that glycerol (Gly) can be effectively converted to lactic acid (LA) under inert atmosphere using a Pt/ZrO₂ catalyst. Starting from pure glycerol, at 180 °C and under a He pressure of 30 bar, we were able to achieve up to 80% yield of LA after a reaction time of 8 h. The catalysts performance of Pt/TiO₂, Pt/C and Pt/ZrO₂ were compared showing that using Pt/ZrO₂ high conversion and stable LA selectivity were achieved during all the process. Further, using Pt/ZrO₂ the LA selectivity was less sensitive to the nature of the reaction atmosphere while using either H₂ or He. While using crude Gly (85% purity), a lower reaction rate was obtained in the presence of Pt/ZrO₂, however high selectivity to LA (～80%) was maintained.

Full text of DOI:10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
Catalysis Today xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
From glycerol to lactic acid under inert conditions in the presence
of platinum-based catalysts: The influence of support
Jamal Ftouni, Nicolas Villandier, Florian Auneau, Michèle Besson,
Laurent Djakovitch, Catherine Pinel∗
IRCELYON UMR 5256, Université Lyon 1, CNRS, 2 avenue Albert Einstein, 69626 Villeurbanne, France
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work, it was shown that glycerol (Gly) can be effectively converted to lactic acid (LA) under inert
atmosphere using a Pt/ZrO2 catalyst. Starting from pure glycerol, at 180 ◦C and under a He pressure of
30 bar, we were able to achieve up to 80% yield of LA after a reaction time of 8 h. The catalysts performance
of Pt/TiO2, Pt/C and Pt/ZrO2 were compared showing that using Pt/ZrO2 high conversion and stable LA
selectivity were achieved during all the process. Further, using Pt/ZrO2 the LA selectivity was less sensitive
to the nature of the reaction atmosphere while using either H2 or He. While using crude Gly (85% purity),
a lower reaction rate was obtained in the presence of Pt/ZrO2, however high selectivity to LA (∼80%) was
maintained.
Received 24 March 2014
Received in revised form 1 September 2014
Accepted 16 September 2014
Available online xxx
Keywords:
Lactic acid
Zirconia
Platinum catalyst
Glycerol
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
sutures. Further, the corresponding racemic PLA is amorphous with
poor mechanical and thermal properties. Moreover, racemic lactic
In the last two decades, due to its large availability, glycerol
(GLY) has become an important renewable feedstock for chemi-
cals. The chemical structure of glycerol, with three hydroxyl groups,
allows the preparation of a large bench of products by various selec-
tive catalytic transformations (e.g. etherification, esterification,
oxidation, and hydrogenolysis reactions) [1–4]. Among them, the
hydrogenolysis reaction has received a growing attention, since it
allows the synthesis of high valuable products (e.g. 1,2-propanediol
(1,2-PDO) or 1,3-propanediol (1,3-PDO)) [5,6].
In our group, we carried out this reaction in the presence of
supported noble metal catalysts (Rh or Ir) under a wide range of
pH (7–10). Under alkaline conditions not only 1,2-PDO was formed
but also a significant amount of lactic acid (LA) appeared [7–10].
The catalytic transformation of glycerol into LA has received
increasing interest since lactic acid is the most widely occurring
hydroxycarboxylic acid with an annual worldwide production of
120,000 tonnes [11]. Most of the applications concern the use of
optically pure (or enriched) lactic acid such as in the food industry
(as an acidulant, as a preservative or even as an inhibitor of bacterial
spoilage), or in the production of polylactic acid (PLA), a biodegrad-
able thermoplastic polymer with medical uses such as implants or
acid may serve as a precursor of even higher added-value prod-
ucts such as pyruvic or acrylic acids [12]. The use of ethyl lactate
as a green and non-toxic (approved as a food additive by FDA [13])
solvent has also been reported. This solvent is miscible with water
and also with some organic solvents, allowing to adjust the polarity
of the medium by proper mixing with a co-solvent. Several routes
are described to achieve selective transformation of glycerol into
LA. The fermentation approach allowed preparing chiral lactic acid
under microaerobic conditions. Recent development of metaboli-
cally engineering strain of Escherichia coli yielded to the production
of 75% of d-lactic acid at a 2.78 g L−1 h−1 productivity [14]. Aerobic
catalytic oxidation, in the presence of supported noble metals was
also reported [15,16].
The use of platinum-based catalysts in the selective glycerol
hydrogenolysis into 1,2-PDO has been described in some papers
[4]. Under neutral conditions, Yuan et al. obtained a 92% conversion
after 22 h using a 3 bar of H2 at 220 ◦C over a Pt catalyst supported
on hydrotalcite (HTL) [17]. The selectivity of this process was quasi-
exclusively to 1,2-PDO with more than 90%; small traces of ethylene
glycol (EG) and propanol (PO) were also detected. Similarly, Pen-
dem et al. reported the use of a Pt/HTL catalyst under 45 bar N2
at 250 ◦C. Their results showed a 98% conversion after 3 h with a
1,2-PDO selectivity of 70%. This selectivity was improved to 75% at
lower conversion (78%) under 35 bar of N2 at 225 ◦C [18]. Moreover,
catalytic conversion of GLY to LA has also been studied using a Pt/C
∗
Corresponding author.
E-mail address: catherine.pinel@ircelyon.univ-lyon1.fr (C. Pinel).
http://dx.doi.org/10.1016/j.cattod.2014.09.034
0920-5861/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
2
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
H3C
OH
OH
OH
HO
OH
1,2-PDO
OH
GLY
OH
H3C
O
LA
Scheme 1. Conversion of glycerol (GLY) to propylene glycol (1,2-PDO), and lactic
acid (LA).
Fig. 1. XRD diffractograms of the supports and the corresponding Pt catalysts. The
vertical lines correspond to the reference metallic platinum, ଝ: Monoclinic zirconia,
ꢀ: Tetragonal zirconia.
catalyst. Maris et al. achieved, after 5 h, a 92% glycerol conversion at
200 ◦C under a 40 bar H2 pressure in the presence of a base (NaOH).
The selectivity was pointed to the formation of LA and 1,2-PDO as
the main products. Replacing NaOH by CaO led to a full glycerol con-
version within 5 h with an improved selectivity to LA up to 58% [19].
Checa et al. obtained a full glycerol conversion with 50% selectivity
to LA after 12 h while using Pt/ZnO catalyst under alkaline condi-
tions at 20 bar H2. Under the same experimental conditions, but
replacing H2 by He, the selectivity to LA was moderately improved
to reach 60% [20]. Using Pt/CaCO3, Ten Dam et al. obtained a low
glycerol conversion of 25% after 18 h at 200 ◦C and under a 40 bar H2
[18]. The selectivity to 1,2-PDO and LA reached 55% and 15%, respec-
tively. In the presence of boric acid, the conversion was increased up
to 55% and the selectivity to 1,2-PDO diminished to 20% in favour of
LA (60%) [21]. This effect was attributed to the formation of borate
ester under alkaline conditions.
2.2. Catalyst preparation and characterization
The support was dispersed in water and the required amount
of platinum (H2PtCl6, Alfa Aesar, theoretical amount 1 wt% for ZrO2
and 3 wt% for C), was added dropwise while stirring under nitrogen
flow. After 5 h of impregnation, the slurry was cooled down in an
ice bath and a 37 wt% solution of formaldehyde was added drop-
wise, followed by a 30 wt% KOH solution. After stirring overnight
under a nitrogen flow, the suspension was filtered and the solid
was washed with water up to elimination of all traces of base (pH
test) and chlorine (AgNO3 test). Finally, the catalyst was dried at
100 ◦C overnight [26].
The ICP analyses were carried out using an ICP-OES Activa
(Jobin Yvon) apparatus. Powder X-ray diffraction (XRD) analyses
were performed with a Bruker D8 Advance A25 diffractometer
Lately, Jin et al. described a new bimetallic catalyst based on Cu
and Pd particles supported on graphene. Under their experimen-
tal conditions (NaOH:GLY = 1.1; T = 140 ◦C and PN2 = 15 bar), a 56%
glycerol conversion with a high LA selectivity of 88% was achieved
after 16 h [22].
˚
(ꢀ = 1.54184 A) using a one dimensional multi-strip fast detector
(LynxEye) with 191 channels on 2.94◦ at 50 kV and 35 mA. The
crystallographic phases of zirconia (monoclinic and tetragonal)
were determined using ‘Highscore plus V3’ software (Table 1). The
diffractograms of the supports and of the corresponding supported
platinum catalysts were similar (Fig. 1). The absence of the Pt peaks
in the XRD diffractograms suggested that the average diameter of
Pt particles was small.
In this work, we report the formation of racemic lactic acid
(LA) from glycerol (GLY) in the presence of platinum-based cat-
alysts under inert conditions (Scheme 1). In previous works, we
highlighted the influence of both: (i) the used atmosphere (H2
or He), and (ii) the active catalytic phase (Ir, Rh) on both the
reaction rate and the selectivity of the reaction [7–10,20]. Zir-
conia (ZrO2) has shown a good stability in water, even at high
temperature and under elevated pressure [23–25]. For this rea-
son, we have focused our work using this support. The catalytic
performance of Pt/ZrO2 catalysts has been compared to those of
some other supported catalysts. Furthermore, the influence of some
other experimental conditions (e.g. nature of atmosphere, purity
of glycerol) on the reaction rate and the LA selectivity has been
investigated.
Transmission electron microscopy (TEM) images were obtained
using a JEOL 2010 LaB6 microscope operating at 200 kV. TEM anal-
ysis confirmed the results of XRD; representative TEM pictures of
supported platinum catalysts show the presence of well dispersed
small platinum particles, with an average diameter below the 5 nm
(Fig. 2); nevertheless some aggregates of Pt were observed on each
sample.
2.3. Catalytic reaction and analytical methods
Most of the experiments were carried out with commercial
pure glycerol (Aldrich G7893, purity > 99.5%). “Crude” glycerol
2. Experimental
2.1. Supports
Table 1
Summary of the supports used for the preparation of the platinum catalysts.
Two commercial ZrO2 supports with a specific surface area
Support
Supplier
Pt loading (%)
SBET (m2 g−1
)
Ratio tetrago-
nal/monoclinic
phases
of 92 and 50 m2 g−1, were provided by MEL Chemicals (ZrO2
)
M
and NORPRO (ZrO2N), respectively. A TiO2 support (Degussa
P25), with a SBET = 50 m2 g−1 and a carbon support from MAST
(40–100 ␮m), with a SBET = 1265 m2 g−1 were also used in this
study.
M
ZrO2
ZrO2
MELCAT
NORPRO
DEGUSSA
MAST
1.2
1
0.9
2.7
92
50
50
25/75
5/95
–
N
TiO2
Carbon
1265
–
The main characteristics of the supports are reported in Table 1.
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
3
Fig. 2. TEM micrographs of supported platinum particles supported on (a) ZrO2, (b) C and (c) TiO2.
was provided by Novance (purity ca 85%). The catalytic reactions
were performed in a 200 mL stainless steel autoclave equipped
with a graphite-stabilized Teflon® container. After introduction
of 100 mL of 5 wt% GLY in [1 M] NaOH aqueous solution (molar
ratio NaOH/GLY = 1.8 in standard conditions; NaOH/GLY = 1.1 while
using crude glycerol), the catalyst was added to achieve the desired
Pt/GLY ratio, the reactor was flushed three times with He and then
heated up to 180 ◦C. When the desired temperature was reached,
the pressure was adjusted (30 bar of He in standard conditions)
and the stirring (1500 rpm) was started setting the t0 of the reac-
tion. Samples of the reaction medium were taken out regularly,
quenched with H2SO4 (0.5 M) and analysed by HPLC (Shimadzu) on
a CarboSep 107H column (0.5 mL min−1 of 0.005 N H2SO4, T = 40 ◦C)
using both refractive index (RI) and ultra-violet (UV) detectors. The
total organic carbon measurements (TOC) were performed using
a Shimadzu TOC-5050A to determine the carbon mass balance.
The samples were diluted 1/200 before analysis, and the carbon
concentration was determined within 5% accuracy.
the tetragonal to monoclinic transformation of ZrO2 phases was
already reported in the presence of water or water vapour even at
relatively low temperature (starting at 65 ◦C) [27]. Further, since
significant amount of monoclinic phase was observed we focused
only on the use of the stable monoclinic zirconia as support.
As far as the selectivity was concerned, regardless the nature
of the support, LA was formed predominantly (Table 2). The other
observed products were propanediols and formic acid as well as
very low amounts of ethylene glycol, ethanol or acetic acid (≤1%).
The mass balance determined by the total organic carbon analysis of
the solutions was >95%, indicating that no gaseous compound was
produced, and this was confirmed by gas analysis of the atmosphere
after 24 h reaction (for selected reactions).
The least active Pt/TiO2 catalyst yielded the highest initial selec-
tivity to LA (>80%) but at low conversion (15%). However for longer
reaction time, we observed a slight decrease in the selectivity and
after 24 h, LA yield reached 63% (76% GLY conversion). As a general
matter, regardless the nature of the supports the selectivity to LA
after 24 h was in the range 75–80% (GLY conversion > 75%) (Table 2).
However, the evolution of the selectivity to LA was affected
by the type of the support (Table 2; Fig. 4). In the case of carbon
supported catalyst (Pt/C), a moderate selectivity of 60% to LA was
observed during the first hours of the reaction, and then it increased
progressively to reach 80% after 24 h (Fig. 4a). On the other side,
when using zirconia supported catalyst (Pt/ZrO2), high selectivity
to LA in the range of 74–84% was achieved from the beginning of
the reaction and it remained almost constant all over the course
of the reaction time (Fig. 4b). In both cases, stable selectivity to LA
was observed up to ca 90–95% GLY conversion, while at nearly full
glycerol conversion it increased at the expense of 1,2-PDO.
The differences in the selectivity observed between the two sup-
ported catalysts (Pt/C and Pt/ZrO2) were mainly attributed to the
simultaneous formation of 1,2-PDO at various levels. In the pres-
ence of Pt/C, the initial selectivity in 1,2-PDO reached 15% before
decreasing to 7% at full GLY conversion (Fig. 4a). We have shown
independently, that under the tested reaction conditions, 1,2-PDO
can be transformed to LA. On the other hand, in the presence of
Pt/ZrO2M, 1,2-PDO was produced at negligible level (<5%) and con-
sequently it did not affect significantly the LA selectivity over the
reaction.
In some case, the gas phase was collected in a gas bag and
analysed using an Agilent 5975C GC/MSD equipped with Alumina,
˚
Poraplot U and 5 A-Molecular sieve columns and thermal conduc-
tivity detectors. Backflush injectors were used for Poraplot U and
˚
5 A-Molecular sieve columns.
3. Results and discussion
3.1. Influence of the nature of the support
The influence of the supports’ nature on glycerol conversion was
studied over the four Pt catalysts supported on C, TiO2 and both
ZrO2. The evolution of the conversion as a function of time achieved
at 180 ◦C under 30 bar of He is reported in Fig. 3.
In the presence of Pt/TiO2 catalyst, the lowest reaction rate was
observed since using this catalyst only 50% GLY conversion was
achieved after 8 h corresponding to a calculated initial reaction
−1
rate of 0.9 mol h
g
Pt
−1. On the other hand, catalysts prepared on
ZrO2 or C supports exhibited a much higher activity since almost
a full GLY conversion was achieved after 8 h, corresponding−t1o
,
−1
−1
−1
and 2.7 mol h g
an initial reaction rate of 1.9 mol h
g
Pt
Pt
respectively (calculated from the slope of the curves; Fig. 3).
As far as the zirconia supports are concerned, they were both
mainly composed of monoclinic phase with minor amount of
tetragonal phase (5–25%); this resulted in similar evolution of
the glycerol concentration within the experimental errors. We
attempted to prepare and evaluate platinum catalysts supported
on tetragonal zirconia. However, this support was not stable under
our experimental conditions and using water as a solvent. Indeed,
3.2. Influence of the nature of the atmosphere and the pressure
level
The nature of the atmosphere could have a significant influence
on the reaction rate and the selectivity to LA [8–10]. This effect
was evaluated using Pt/ZrO2N catalyst and compared to the results
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
4
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
Fig. 3. Evolution of glycerol conversion as a function of time using different supported platinum catalysts. (Reaction conditions: 5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8;
Pt/GLY = 2000; PHe = 30 bar.)
Table 2
Catalytic transformation of glycerol in the presence of different supported catalysts. (Reaction conditions: 5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8; Pt/GLY = 2000; PHe = 30 bar.).
Catalyst
t (h)
GLY conv (%)a
Mass balanceb
Yield (%)a
LA (Selc)
Other productsd
2
24
51
95
38 (74)
80 (84)
N
Pt/ZrO2
Pt/ZrO2
Pt/TiO2
>95%
>95%
>95%
>95%
1,2-PDO = 4;
FAe = 6; 1,3-PDO = 2
2
24
49
94
38 (77)
84 (84)
M
1,2-PDO = 4;
FAe = 7; 1,3-PDO = 2
2
24
17
76
15 (87)
63 (83)
1,2-PDO = 6;
FAe = 3; 1,3-PDO = 1
2
24
62
100
38 (62)
80 (80)
Pt/C
1,2-PDO = 7;
FAe = 3; 1,3-PDO = 3
a
Determined by HPLC.
Determined by TOC analyses after 24 h.
Selectivity.
Some other products have been also detected in small amount (<1%; e.g. ethylene glycol, acetic acid and ethanol).
FA = formic acid.
b
c
d
e
achieved in the presence of the Pt/C catalyst. Regardless the nature
of the support, higher conversions were observed under inert He
atmosphere compared to the reductive one. In the presence of
Pt/ZrO2N, after 5 h reaction, 74–82% and 40–50% GLY conversions
were achieved under He and H2 atmosphere, respectively (Fig. 5b).
On the other hand, in the presence of Pt/C, 91% and 59% conver-
sions were obtained after 5 h under inert and reductive atmosphere,
respectively (Fig. 5a).
The respective yields of the two main products were compared
after 5 h reaction (Fig. 6). Under inert atmosphere, irrespective of
the catalyst, 60% LA yields were achieved. But as specified previ-
ously a larger amount of 1,2-PDO was produced in the presence
of Pt/C catalyst (15% vs 5–7% in the presence of ZrO2N). On the
other hand, under reductive atmosphere, LA was always the main
product but to a lesser extent (<35%) and as it was observed under
inert atmosphere, the presence of Pt/C favoured the formation of
Fig. 4. Selectivities to LA and 1,2-PDO in the presence of (a) Pt/C and (b) Pt/ZrO2M under the same experimental conditions as a function of GLY conversion and time. (Reaction
conditions: 5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8; GLY/Pt = 2000; PHe = 30 bar.)
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
5
Fig. 5. Conversion of glycerol in presence of (a) Pt/C and (b) Pt/ZrO2N catalysts, with different He or H2 pressures. (Reaction conditions: 5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8;
GLY/Pt = 2000.)
In the presence of Pt/C catalyst, slightly different behaviour was
observed depending on the nature of the atmosphere (Fig. 7b).
Under He atmosphere, the initial selectivities to LA and PDO
reached 60% and 20%, respectively. During the progress of the reac-
tion, the selectivity to LA progressively increased from 60% to 80%
(+20%) while the selectivity to PDO decreased to 8%. Under H2 atmo-
sphere, the initial selectivity to LA was lower than 50% to reach 56%
after 24 h (+6%). In parallel, the selectivity to PDO was in the range
30–40% during the reaction. In that case, some minor by-products
were not analysed at low glycerol conversion.
At complete conversion of glycerol, the nature of the used atmo-
sphere has a stronger impact on the yield in LA and PDO in the
N
presence of Pt/C than in the presence of Pt/ZrO2 (Fig. 7a and b).
While using Pt/ZrO2N, the LA yields reached 78 and 65% after 24 h
under He and H2 atmospheres, respectively. Under identical reac-
tion conditions, in the presence of Pt/C, LA yield reached 80 and 55%
under He and H2, respectively.
Fig. 6. Influence of the atmosphere on the conversion of glycerol and the yields
N
in 1,2-PDO and LA in the presence of Pt/C and Pt/ZrO2 catalysts, with different
He or H2 pressures. (Reaction conditions: 5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8;
GLY/Pt = 2000, t = 5 h.)
In the presence of Pt/ZrO2N, the influence of partial pressure
of gas was evaluated, taking into account that at the considered
temperature (180 ◦C), the inherent partial pressure of water was
15 bar. Under He atmosphere, no significant influence of the pres-
sure on the reaction rate or the selectivities to LA and PDO was
observed, the results being similar within the experimental errors
(80% GLY conversion after 5 h and 60% selectivity to LA; Fig. 6).
On the other hand, under reductive atmosphere, the conversion
slightly decreased when higher pressure was used: after 5 h, 40%
and 51% glycerol conversions were achieved under 50 and 30 bar of
hydrogen, respectively. The influence of the hydrogen pressure on
the reaction rate could be due to the fact that a dehydrogenation
reaction was proposed as the first step (vide infra). Regardless the
1,2-PDO (21 and 7% under H2 and He atmosphere, respectively;
Fig. 6).
Moreover, the selectivity to the products was affected by the
nature of the atmosphere (Fig. 7a and b). Over Pt/ZrO2N, what-
ever the nature of the atmosphere, the selectivities to LA and PDO
were almost constant during the progress of the reaction (Fig. 7a).
Under He atmosphere, 75–80% selectivity to LA was achieved
together with selectivity to PDO < 10%. On the other hand, under
H2 atmosphere, the selectivity to LA decreased to 65–70%, while
the selectivity to PDO increased up to ∼25%.
N
Fig. 7. Evolution of selectivity to LA and PDO in the presence of (a) Pt/ZrO2 and (b) Pt/C under different atmospheres (He or H2) as a function of time. (Reaction conditions:
5 wt% GLY; T = 180 ◦C; NaOH/GLY = 1.8; GLY/Pt = 2000; P = 30 bar of He or 50 bar of H2.)
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
6
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
O
OH
+H2
HO
HO
[metal]
O
[metal]
acetol
1,2-PDO
-H2
-H2
OH
[metal] +H2
Route 1
-H2O
OH
O
OH
OH
OH
enol 3
-H2
O
O
[metal]
HO
OH
[HO-]
O
OH
O
[HO-]
[metal]
Cannizzaro
reaction
OH
LA
pyruvaldehyde
glyceraldehyde
glycerol
H2O
[metal]
+H2
OH
Route 2
OH
OH
O
OH
-H2O
-H2
O
O
O
HO
OH
[HO-]
[metal]
OH
OH
OH
pyruvic acid
OH
glyceric acid
enol
Scheme 2. Proposed reaction pathway for glycerol conversion into lactic acid (LA) and 1,2-propanediol (1,2-PDO) starting with a dehydrogenation step.
pressure, the corresponding selectivity to LA reached 65–67% at ca
50% conversion.
To summarize, the influence of the support in the present
reaction is not clear. The main change concerns the LA/1,2-PDO
selectivity under H2 pressure while the difference is considerably
lower under inert gas (Fig. 6). The evaluation of Pzc of zirconia and
carbon has shown that the zirconia support is slightly more acidic
than the carbon one (Pzc = 4.9 and 7.2, respectively). However, this
parameter should have low impact since the reaction was carried
out in alkaline conditions and regardless the nature of the support,
the pH decreased from ∼13 to ∼12 at complete conversion. Further
works are under progress to characterize the catalysts in order to
understand the role of the support.
We have previously reported the influence of the nature of the
atmosphere in the presence of Ir- and Rh-based catalysts for glyc-
erol transformation [8,10]. With these metals, reverse selectivities
were achieved when changing from inert to reductive atmosphere:
namely, LA was the main product under He while 1,2-PDO was pre-
Fig. 8. Evolution of glycerol conversion and selectivities to LA and PDO as a
function of time using pure and crude glycerol. (Reaction conditions: 5 wt% GLY;
NaOH/GLY = 1.1; GLY/Pt = 2000; T = 180 ◦C; PHe = 30 bar.)
dominant under H2. In this study, we observed that in the presence
of Pt-supported catalysts, LA was the main product irrespective of
the used atmosphere. This could be due to the dehydrogenation
of its lower price. The presence of impurities such as salts, ash,
properties of Pt-based catalysts [26]. Furthermore, according to DFT
methanol and water could affect the catalyst activity as previously
calculation, it was assessed that the dehydrogenation to glycer-
reported in the 1,2-PDO synthesis [31]. As a preliminary study, we
aldehyde was the first step of the reaction [8] (Scheme 2). Then,
compared the performances of the catalyst using pure and crude
two routes could be involved: either dehydration to enol, then
glycerol. Since the target is to evaluate the catalyst under the most
pyruvaldehyde, followed by rearrangement via Cannizzaro reac-
scalable conditions, we performed the reaction using lower base
quantity. We have shown previously, that in the presence of Rh-
tion (route 1); or second dehydrogenation to glyceric acid (via the
formation of the gem diol) followed by dehydration and hydrogena-
based catalyst, the reaction rate was affected by the pH of the
tion (route 2). At the present time, we are not able to discriminate
solution [8]. The same evolution was observed in the presence of
between these different routes. Nevertheless, in some of our exper-
Pt-based catalyst: using NaOH/GLY ratio of 1.1, 32% glycerol con-
iments, traces of glyceric acid were detected exclusively at the
version with a 75% LA selectivity was obtained within 2 h (Fig. 8);
beginning of the catalytic test. These traces supported the route
while in the presence of NaOH/GLY = 1.8, the glycerol conversion
2; moreover, this route was also discussed by other researches
and LA selectivity reached 51% and 74%, respectively (Table 2). The
[5,28]. Furthermore, it was established that the Cannizzaro reac-
higher reaction rate was attributed to the increase of the alkaline
tion should not be favoured when hydrogen atoms in ␣-position
contribution of the reaction.
to carbonyl are present [29]. On the other hand, Lux and Sieben-
hoffer recently reported that Ca(OH)2 effectively catalysed lactic
The evolution of the reaction using “pure” and “crude” glycerol
N
sources in the presence of Pt/ZrO2 catalyst is shown in Fig. 8.
acid from dihydroxyacetone [30]. Accordingly, we suggest that the
Under the same standard experimental conditions (5 wt% GLY;
reaction occurred mainly through route 2 but further works need
NaOH/GLY = 1.1; pH = 12.7; GLY/Pt = 2000; T = 180 ◦C; PHe = 30 bar),
to be performed to establish the exact route.
“pure glycerol” gave a full conversion after 24 h with an initial reac-
−1
tion rate of 1.4 mol h
glycerol” resulted in lower initial reaction rate (0.8 mol h
g
−1. On the other hand, the use of “crude
Pt
−1
−1
)
3.3. Influence of the glycerol purity
g
Pt
giving a 70% conversion within 24 h. In both cases, similar selectiv-
ities of ca 80% to LA were achieved at the end of the process (after
24 h).
The catalytic transformation of crude glycerol, a by-product
from the biodiesel process represents an economic asset because
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034

G Model
CATTOD-9313; No. of Pages7
ARTICLE IN PRESS
J. Ftouni et al. / Catalysis Today xxx (2015) xxx–xxx
7
Table 3
Composition of glycerol sources.
Glycerol quality
Composition (%)
Inorganic composition (ppm)
Na
P
S
Fe
Pure (Aldrich)
Crude (Novance)
>99.5 Gly
4
<2
32
nd
39
<2
7
85% Gly11%H2O<1% monoester<0.1% MeOH4% ash
6.4 × 103
The analysis of the crude glycerol has shown that the mixture
is composed of 85% glycerol, 11% water and 4% ash. The presence
of residual monoester (<1%) was also observed. Furthermore, the
dark colour of the crude glycerol indicates the presence of other
compounds at low level. As far as the mineral content was con-
cerned, the crude glycerol contained mainly Na as impurity issued
from the transesterification of vegetable oils (Table 3). Since NaCl
was previously reported to act as a poison in glycerol hydrogenol-
ysis [29], this salt was added to the pure glycerol in a NaCl/GLY = 1
molar ratio.
After 24 h reaction, no effect of the addition of NaCl was
evidenced either on the glycerol conversion or on the selectivity
to LA. This indicates that the lower activity observed when using
crude glycerol could not be attributed to the presence of NaCl.
Other impurities will be tested to determine the compounds caus-
ing the decrease of the activity in the catalytic transformation of
‘crude glycerol’. Decolourization of crude glycerol solution will also
be carried out to evaluate the influence of the presence of these
by-products on the reaction rate.
References
[1] M. Pagliaro, M. Rossi, The Future of Glycerol: New Uses of a Versatile Raw
Material, 2008.
[2] C.-H. Zhou, J.N. Beltramini, Y.-X. Fan, G.Q. Lu, Chem. Soc. Rev. 37 (2008)
527–549.
[3] A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Green Chem. 10 (2008)
13–30.
[4] M.
1827.
[5] Y. Nakagawa, K. Tomishige, Catal. Sci. Technol. 1 (2011) 179–190.
Besson,
P.
Gallezot,
C.
Pinel,
Chem.
Rev.
114
(2014)
[6] F. Auneau, N. Villandier, L. Djakovitch, C. Pinel, in: C. Evangelisti, F. Zaccheria
(Eds.), Recent Advances and New Trends in Catalysis, ISTM Editions, Milano,
2012, pp. 103–139.
[7] J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel, C. Rosier, Green
Chem. 6 (2004) 359–361.
[8] F. Auneau, C. Michel, F. Delbecq, C. Pinel, P. Sautet, Chem. Eur. J. 17 (2011)
14288–14299.
[9] F. Auneau, L.S. Arani, M. Besson, L. Djakovitch, C. Michel, F. Delbecq, P. Sautet,
C. Pinel, Top. Catal. 55 (2012) 474–479.
[10] F. Auneau, S. Noël, G. Aubert, M. Besson, L. Djakovitch, C. Pinel, Catal. Commun.
16 (2011) 144–149.
[11] R. Datta, M. Henry, J. Chem. Technol. Biotechnol. 81 (2006) 1119–1129.
[12] P. Mäki-Arvela, I.L. Simakova, T. Salmi, D.Y. Murzin, Chem. Rev. 114 (2014)
1909.
[13] J.S. Bennett, K.L. Charles, M.R. Miner, C.F. Heuberger, E.J. Spina, M.F. Bartels, T.
Foreman, Green Chem. 11 (2009) 166–168.
[14] X.Z. Chen, K.M. Tian, D.D. Niu, W. Shen, G. Algasan, S. Singh, Z.X. Wang, Green
Chem. 16 (2014) 342–348.
4. Conclusion
This work presents an efficient alternative route for the prepa-
ration of lactic acid. High yields of LA (∼80%) at total conversion
were obtained starting from pure glycerol at 180 ◦C under inert con-
ditions using platinum supported catalysts. Among the supports
evaluated (C, TiO2, ZrO2), ZrO2 appeared to be the most efficient for
this process. Investigation of the experimental conditions (nature
and pressure of the gaseous atmosphere, purity of glycerol) showed
that major differences were observed at short reaction times (<10 h)
mainly on the selectivity (especially the LA), while after 24 h these
differences were mainly negligible. Under the same conditions,
despite high selectivity, crude glycerol was converted to LA in a
useful 52% yield at 70% conversion after 24 h. Current researches
focus on improving the crude glycerol transformation by evaluat-
ing the role of the various impurities present in the mixture on the
catalytic activity.
[15] Y. Shen, S. Zhang, H. Li, Y. Ren, H. Liu, Chem. Eur. J. 16 (2010) 7368–7371.
[16] R.K. Pazhavelikkakath Purushothaman, J. van Haveeren, D.S. van Es, O.
Melian-Cabrera, J.D. Meeldijk, H.J. Heeres, Appl. Catal.
92–100.
B 147 (2014)
[17] Z. Yuan, P. Wu, J. Gao, X. Lu, Z. Hou, X. Zheng, Catal. Lett. 130 (2009) 261–265.
[18] C. Pendem, P. Gupta, N. Chaudhary, S. Singh, J. Kumar, T. Sasaki, A. Datta, R. Bal,
Green Chem. 14 (2012) 3107–3113.
[19] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
[20] M. Checa, F. Auneau, J. Hidalgo-Carrillo, A. Marinas, J.M. Marinas, C. Pinel, F.J.
Urbano, Catal. Today 196 (2012) 91–100.
[21] J. ten Dam, F. Kapteijn, K. Djanashvili, U. Hanefeld, Catal. Commun. 13 (2011)
1–5.
[22] X. Jin, L. Dang, J. Lohrman, B. Subramaniam, S. Ren, R.V. Chaudhari, ACS Nano 7
(2013) 1309–1316.
[23] M. Besson, P. Gallezot, Catal. Today 81 (2003) 547–559.
[24] J.C. Serrano-Ruiz, J. Luettich, A. Sepulveda-Escribano, F. Rodirguer-Reinoso, J.
Catal. 241 (2006) 45.
[25] E.I. Gürbüz, E.L. Kunkes, J.A. Dumesic, Appl. Catal. B 94 (2010) 134.
[26] P. Korovchenko, C. Donze, P. Gallezot, M. Besson, Catal. Today 121 (2007)
13–21.
[27] X. Guo, T. Schober, J. Am. Ceram. Soc. 87 (2004) 746.
[28] J. ten Dam, U. Hanefeld, ChemSusChem 4 (2011) 1017–1034.
[29] J. March, Advanced Organic Chemistry: Reactions, Mechanisms and Structure,
second ed., 1977.
[30] S. Lux, M. Siebenhofer, Catal. Sci. Technol. 3 (2013) 1380–1385.
[31] C.-W. Chiu, M.A. Dasari, W.R. Sutterlin, G.J. Suppes, Ind. Eng. Chem. Res. 45
(2006) 791–795.
Acknowledgements
This work is part of the ANR (French Research Agency) “GALAC”
(ANR-10-CD2I-011). We would like to thank NOVANCE for provid-
ing us the crude glycerol as well as the corresponding analysis.
UBIOCHEM COST Action CM 0903 is acknowledged.
Please cite this article in press as: J. Ftouni, et al., From glycerol to lactic acid under inert conditions in the presence of platinum-based
catalysts: The influence of support, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2014.09.034