On the role of the atmosphere in the catalytic glycerol transformation over iridium-based catalysts

Article abstract of DOI:10.1016/j.catcom.2011.09.011

Here, we report the transformation of glycerol alkaline solutions in the presence of supported iridium catalysts under either reductive or inert atmosphere at 453 K. It was shown that under H₂, the expected 1,2-PDO issued from hydrogenolysis of the triol was the main product while under He, lactic acid was formed predominantly via Cannizzaro rearrangement.

Full text of DOI:10.1016/j.catcom.2011.09.011

Catalysis Communications 16 (2011) 144–149
Contents lists available at SciVerse ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
On the role of the atmosphere in the catalytic glycerol transformation over
iridium-based catalysts
Florian Auneau a,b, Sébastien Noël , Guillaume Aubert , Michèle Besson ,
a
a
a
a
a,
Laurent Djakovitch , Catherine Pinel ⁎
Institut de Recherches sur la Catalyse et l'Environnement de Lyon, IRCELYON, UMR5256, Université de Lyon, Université Lyon I, CNRS, 2, avenue Albert Einstein,
a
F-69626 Villeurbanne Cedex, France
b
Laboratoire de chimie, UMR 5182 Ecole Normale Supérieure de Lyon, CNRS, Université Lyon, 46 allée d'Italie, F-69364 Lyon Cedex 07, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 July 2011
Received in revised form 5 September 2011
Accepted 12 September 2011
Available online 18 September 2011
Here, we report the transformation of glycerol alkaline solutions in the presence of supported iridium cata-
lysts under either reductive or inert atmosphere at 453 K. It was shown that under H , the expected 1,2-
2
PDO issued from hydrogenolysis of the triol was the main product while under He, lactic acid was formed
predominantly via Cannizzaro rearrangement.
©
2011 Elsevier B.V. All rights reserved.
Keywords:
Iridium catalyst
Glycerol
Lactic acid
1,2-propanediol
Hydrogenolysis
1
. Introduction
the addition of acid and alkaline promoters as well as the nature of
the gas atmosphere. Depending on these parameters, dramatic
changes of selectivity were achieved from a 100% yield to 1,2-PDO
up to the formation of 1,3-PDO or LA.
Recently, several reports dealt with the partial dehydroxylation of
glycerol under inert gas. D'Hondt et al. described the formation of 1,2-
PDO (54%) at 503 K under inert atmosphere in the presence of a
Pt/NaY catalyst after 15 h [21]. Roy et al. reported the use of a 1:1 ad-
Glycerol is a feedstock available in large amount since it is a by-
product in the process of bio-diesel production by transesterification
of vegetable oils [1,2]. The selective transformation of glycerol repre-
sents one of the most attractive routes as large range of valuable com-
pounds such as acrolein, dihydroxyacetone, glyceric acid, lactic acid,
1
,2-propanediol or 1,3-propanediol can be produced [3–5]. Among
all potential routes, the reaction of glycerol with hydrogen in the
presence of homogeneous or heterogeneous metallic catalysts
received a growing attention leading mainly to 1,2-propanediol
2 3 2 3
mixture of Ru/Al O and Pt/Al O catalysts to perform the dehydrox-
ylation of glycerol to 1,2-PDO without hydrogen addition; a 24% yield
of 1,2-propanediol was achieved at 493 K after 6 h [22]. It was sug-
gested that hydrogen was produced via liquid steam reforming
[23,24] but the exact mechanism is still questionable. Independently,
the teams of Lercher et al. [25] and Fierro et al. [26] studied the reac-
tivity of glycerol without added hydrogen in the presence of bifunc-
(
1,2-PDO) as a major commodity chemical. This product finds some
uses as antifreezing, lubricant, industrial solvent or bulk chemical in-
termediate for the production of polyester materials [6]. Depending
on the reaction conditions, the production of some other compounds
was described such as 1,3-propanediol (1,3-PDO) used in polymeric
applications [7–12] or ethylene glycol (EG) issued from C―C hydro-
genolysis [13–15]. Lactic acid (LA), a central feedstock for chemical
industry, applied either in food industry or as a renewable building
block for sustainable chemistry in particular in biopolymer synthesis,
was also mentioned in few reports [16–20].
2 3
tional catalysts such as Pt/γAl O or Pt/ASA (amorphous silico-
alumina).
As mentioned previously, the presence of acid or base additives
may influence dramatically the product distribution in the final solu-
tion. The reaction rate was significantly enhanced in the presence of
basic medium and in that case several reports described the concom-
itant formation of lactic acid together with 1,2-PDO and ethylene gly-
col as the main products. Montassier et al. described the conversion of
glycerol under hydrogen in the presence of Cu Raney® at different pH
Not only the nature of the catalyst is of large importance for the
conversion rate of glycerol but also the reaction conditions such as
[16]. They showed that under alkaline conditions (NaOH 1 M), lactic
acid was the main product (85% initial selectivity) while it was not
observed under neutral conditions. Casale and Gomez patented the
⁎
Corresponding author. Fax: +33 4 7244 5300.
E-mail address: catherine.pinel@ircelyon.univ-lyon1.fr (C. Pinel).
1566-7367/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2011.09.011

F. Auneau et al. / Catalysis Communications 16 (2011) 144–149
145
use of ruthenium catalyst modified with sulphur at 513 K in the pres-
ence of NaOH to yield after 2 h, 75% of 1,2-PDO and 13% of LA as the
main constituents [17]. The team of Davis reported the presence of
lactic acid during hydrogenolysis of glycerol in the presence of bases
on 2.94° at 50 kV and 35 mA. Transmission electronic spectroscopy
images were obtained using a JEOL 2010 LaB6 microscope operating
at 200 kV.
(
CaO or NaOH) over carbon- supported Ru, Pt, Pt-Ru and Au-Ru cata-
2.2. Catalytic reaction and analytical method
lysts [18,19]. However, the selectivity to lactic acid decreased with
the conversion to reach 35–38% at complete conversion. Liu et al. in-
Initial screening of catalysts was conducted in a Slurry Phase Reac-
tor 16 (AMTEC), designed to conduct several experiments in parallel,
for 12 h using 6 mL of a sodium hydroxide aqueous solution of 5 wt.%
glycerol, in the presence of 30 mg catalyst at 1000 rpm. However this
system did not allow to perform sampling and in order to establish
the kinetic profile, the reaction was performed in a 200 mL stainless
steel autoclave equipped with a graphite-stabilized Teflon® container
and designed to sample the liquid phase. 100 mL of 5 wt.% glycerol in
NaOH 1 M were introduced in the vessel with 500 mg catalyst and the
reactor was flushed with helium and heated at 453 K. When the tem-
perature was reached, the atmosphere was adjusted to 30 bar and the
stirring (1500 rpm) started corresponding to time zero of the reac-
vestigated TiO
alkaline (NaOH) conditions; they obtained a 45% yield of LA using
the Au-Pt/TiO catalyst [27]. To our knowledge, this is the best yield
2
-supported Au, Pt or Au-Pt catalysts under oxidative
2
ever achieved for the transformation of glycerol into LA using hetero-
geneous metallic catalysis. Non-catalyzed hydrothermal conversion
of glycerol into lactic acid was also reported [28,29].
Few reports mentioned the use of iridium-based catalysts in this re-
2
action [7,16]. According to Montassier's work, Ir/SiO exhibited a lower
activity than the corresponding Rh and Ru-based catalysts but it selec-
tively hydrogenolyzed C―O bonds without affecting C―C ones. In this
report, the activities and selectivities of glycerol conversion were inves-
tigated in the presence of iridium catalysts under inert or reductive
atmosphere and the results were compared with rhodium-based cata-
lysts [16,30,31]. We have shown that while 1,2-propanediol is selective-
ly produced under hydrogen pressure, a high selectivity towards lactic
acid can be achieved under helium (Scheme 1). The key role of the
nature of the atmosphere in the selectivity of the reaction has been
highlighted.
tion. For reaction under hydrogen, the reactor was purged with H
2
and the pressure was adjusted to 50 bar at the desired temperature
since this pressure was used in previous study [12]. Samples of the re-
action medium were taken out regularly and analyzed by HPLC on a
2 4
CarboSep 107H column (0.5 mL/min 0.005 N H SO , T=313 K). 1,3-
PDO, 1,2-PDO, ethylene glycol, 1-propanol, 2-propanol, ethanol,
methanol, acetol, lactic acid, formic acid and acetic acid were ana-
lyzed. GC-MS analysis confirmed the identification of lactic acid and
1,2-propanediol. The Total Organic Carbon (TOC) of the aqueous solu-
tion was measured using a Shimadzu TOC-5050A analyzer. The com-
parison of the measured TOC and the calculated TOC obtained by
chromatographic analysis allowed to verify the carbon balance
based upon glycerol consumed and products formed in the aqueous
phase. Also, the difference between the TOC measured and the initial
TOC concentration gave an estimation of carbon-containing gaseous
products formed during the reaction. It was checked that similar re-
sults were achieved in both reactors.
The gas phase was also collected in a gas bag at the end of the re-
action and analyzed using a GC-MS (Agilent Technologies, 5975 C)
equipped with Alumina, Poraplot U and 5 Å-Molecular sieve columns
and thermal conductivity detectors. Backflush injectors were used for
Poraplot U and 5 Å-Molecular sieve columns.
2
. Experimental
2
.1. Catalyst preparation and characterization
Iridium catalysts supported on carbon were prepared by cationic
3 5 2
exchange using [Ir(NH ) Cl]Cl (Alfa Aesar) as metal precursor. MAST
carbon (mesoporous synthetic carbon activated at 1123 K under CO
2
2
−1
atmosphere; SBET=1220 m g ) was first oxidized by treatment
with sodium hypochlorite (5% active chlorine) for 20 h at room temper-
ature. The reaction was quenched by the addition of a 1.25 N hydrochlo-
ric acid solution. The support was then filtered and washed with
demineralized water until complete elimination of chloride ions from
the support. 4.95 g of oxidized support were dispersed in 120 mL of
1
M ammonium hydroxide under nitrogen bubbling. 300 mg of the irid-
ium salt (to get 1 wt.%Ir) dissolved in 60 mL of 1 M ammonium hydrox-
ide was added dropwise at room temperature. After stirring under
nitrogen atmosphere for 24 h, the suspension was filtered; the catalyst
was washed with water until neutralization, and dried overnight at
The conversion (Conv(%)) of glycerol was defined as the number
of mol of glycerol consumed with respect to the initial number of
mol of glycerol. The selectivity in a given product i was calculated
according to the following equation:
3
33 K under vacuum. The solid was reduced according to the following
protocol: 2.5 g of solid were introduced in a reduction cell which was
purged with argon before the hydrogen flow was adjusted at 12 L/h.
The catalyst precursor was heated at 2 K/min up to 573 K and this tem-
perature was maintained for 3 h. After cooling down slowly to room
temperature, the cell was then flushed with argon. Iridium loading
was determined by ICP analysis (Ir%=0.6–0.8 wt.% depending on the
batch).
t
t
Xi ꢀ nCi
Seli ð%Þ ¼
ꢀ 100
t
gly
X
ꢀ 3
t
X ꢀ nCi
t
i
i
Yields were similarly expressed as: Y ð%Þ ¼
ꢀ 100. Or,
0
gly
X
ꢀ 3
t
Sel ð%Þ
t
i
i
t
i
t
Y ð%Þ ¼ Convð%Þ ꢀ
of reacted glycerol at reaction time t, respectively, Xgly initial amount
of glycerol, and nC
where X
and Xgly: mol of product i and mol
1
00
0
For comparison, a commercial 5 wt% Ir/CaCO
m g ) catalyst was reduced following the same protocol as for the
3
(Alfa Aesar, SBET
=
2
−1
8
i
standing for the number of carbon atoms of product
Ir/C catalyst. The effective loading was confirmed by ICP analysis.
Powder X-ray diffraction (XRD) analysis were performed with a
Bruker D8A25 Advance diffractometer (λ=1.54184 Å) using a one
dimensional multistrip fast detector (LynxEye) with 191 channels
i.
To evaluate the leaching of metallic species in solution, the reac-
tion medium was filtered and the iridium amount was determined
by ICP/AES analysis. It was shown that there was no leaching of irid-
ium at the end of the reaction whatever the support.
OH
OH
O
OH
[
Ir]
3
. Results and discussion
HO
OH
HO
+
He or H2
HO
3.1. Characterization of iridium catalysts
1
,2-PDO
LA
3
XRD of 0.6% Ir/C and 5% Ir/CaCO catalysts did not show any dif-
Scheme 1. Main products formed during catalytic transformation of glycerol in the
presence of supported iridium catalysts.
fraction peak characteristic of the iridium phase, indicating that the

146
F. Auneau et al. / Catalysis Communications 16 (2011) 144–149
Ir crystallites were probably very small. Fig. 1 shows typical TEM im-
ages of homemade 0.6% Ir/C and of commercial 5% Ir/CaCO catalysts.
hydrogen or helium pressure. Conversions and yields achieved after
8–12 h reaction time are summarized in Table 1.
3
The samples consisted of relatively small iridium particles (2–3 nm)
homogeneously distributed throughout the support whatever its
nature.
For comparison, entries 1–2 show the glycerol conversion in the
presence of a 0.7 wt.% Rh/C, under H or He atmosphere respectively.
2
1,2-PDO and LA were the main products of the reaction. Other com-
pounds such as formic acid (FA), acetic acid (AA), 1,3-propanediol
(
(
1,3-PDO), ethylene glycol (EG), acetol, ethanol (EtOH) and methanol
MeOH) were also detected but in negligible amounts. The LA yield
3
.2. Influence of temperature
was significantly higher under inert atmosphere (ca 25%) compared
to the reaction under hydrogen (ca 5%), whereas the 1,2-PDO
The catalytic transformation of glycerol in the presence of Ir/C was
studied in the 15 mL parallel reactors, at different temperatures
under either hydrogen or helium atmosphere (Fig. 2a and b, respec-
tively). The reaction of a 5% glycerol solution was carried out in the
presence of 0.8% Ir/C catalyst, in 1 M NaOH (initial pH ca 13.0).
Under hydrogen pressure, after 12 h reaction, less than 10% conver-
sion was achieved for temperatures lower than 413 K whereas a con-
version higher than 95% was observed at 473 K. Up to 453 K, the mass
balance was nearly equilibrated only 10% being missing at 473 K. The
best selectivity (64%) towards the main product (1,2-PDO) was
obtained at 453 K. Under helium pressure, whatever the temperature,
conversions were higher than under hydrogen but the selectivity
changed in favor of LA (vide infra). The formation of acid was correlat-
ed with the decrease of the final pH of the solution (from 13 up to
yield was slightly lower under He pressure than under H
4 and 9%, respectively). Similar tendencies were previously reported
in the presence of Cu Raney® catalyst [16]. The performances of the
.6 wt.% Ir/C catalyst were evaluated at different pH (entries 3–6). It
is clear that the catalyst was not efficient under acidic conditions
entry 3) since there was no conversion after 12 h. In water (neutral
2
pressure
(
0
(
conditions), very low conversion (b5%) occurred whatever the nature
of the atmosphere leading to traces of 1,2-PDO (entries 4–5). Addi-
tionally, some acetol was detected under helium (entry 5). On the
other hand, significantly higher conversions were achieved in basic
conditions. Under H
2
atmosphere, the conversion reached 22%
and 51% after 12 h in the presence of 0.1 M or after 8h in the presence
of 1 M NaOH, respectively (entries 6, 8). A more pronounced behavior
was observed under inert atmosphere; the conversions reached 26%
1
2.6). Up to 413 K, all the products formed during the reaction were
analyzed so there was no difference between conversion and yields
of products. Above this temperature, some of them were not analyzed
by HPLC (up to 30% at 473 K). However, TOC analysis revealed that
the carbon mass balance in solution was higher than 90%, suggesting
that the unidentified products could be ethers obtained by condensa-
tion of glycerol [32].
Further studies were performed at 453 K, the conversion and
selectivity being both significant at this temperature. Moreover, it
was verified that under these reaction conditions, glycerol was stable
in the absence of supported metallic catalyst.
(
after 12 h) and 76% (after 8h) using 0.1 or 1 M NaOH solution, re-
spectively (entries 7, 9). When a molar solution of NaOH was used,
the conversions measured in the presence of iridium-based catalyst
were more important than with rhodium catalyst. The conversions
under hydrogen or helium atmosphere reached respectively 51%
and 76% in the presence of Ir/C (Table 1, entries 8, 9) versus 22% and
a 1
00
13,1
13,0
12,9
12,8
12,7
12,6
LA
1,2-PDO
Others
pHf
Conversion
3
.3. Evaluation of reaction parameters
8
6
4
2
0
0
0
0
0
The influence of other reaction parameters was screened in the
two types of reactors at 453 K with 5% glycerol solution under either
373
393
413
433
453
473
temperature (K)
b 1
00
13.1
13.0
12.9
12.8
12.7
12.6
LA
Others
pHf
1,2-PDO
Conversion
8
6
4
2
0
0
0
0
0
373
393
413
433
453
473
temperature (K)
Fig. 2. Influence of the temperature on glycerol conversion and products distribution
under a) H (50 bar) and b) He (30 bar) atmosphere. 6 mL of 5 wt.% glycerol in
NaOH 1 M, 30 mg 0.8% Ir/C, t=12 h.
2
3
Fig. 1. TEM images of supported iridium catalysts. a) 0.6% Ir/C; b) 5% Ir/CaCO .

F. Auneau et al. / Catalysis Communications 16 (2011) 144–149
147
Table 1
Catalytic transformation of glycerol in the presence of carbon supported metallic.
Entry
Catalyst
Gas
Solvent
pHfinal
Conv
%)
Mass
Yield (%)a
1,2-PDO
Gas
phase
(
balance
LA
Othersc
b
(%)
1d
Rh/C
Rh/C
Ir/C
Ir/C
Ir/C
Ir/C
Ir/C
Ir/C
Ir/C
H
He
H
H
He
H
He
H
He
NaOH 1 M
NaOH 1 M
12.9
12.8
1.7
5.6
4.9
12.1
8.3
12.8
12.9
22
55
0
b 5
b 5
22
26
51
76
100
94
–
9
4
5
25
–
3
12
–
H
2
He, H
–
2
d
2
3
4
5
6
7
8
2
e
2
H
3
H
2
H
2
PO
O
O
4
0.1 M
–
e
2
–
1
1
10
5
37
7
ε
ε
10
15
11
37
ε
nd
nd
nd
nd
H
2
He, H
2
e
–
2
e
2
NaOH 0.1 M
NaOH 0.1 M
NaOH 1 M
NaOH 1 M
100
100
99
94
ε
e
3
2
6
d
2
d
9
nd : not determined.
a
Yields determined by HPLC.
Mass balance was calculated on the basis of TOC analysis.
Others: EG, 1,3-PDO, formic acid, acetic acid, acetol, ethanol, methanol.
b
c
d
e
Reaction conditions: 100 mL 5 wt.% GLY in NaOH 1 M (pH
mL 5 wt.% GLY in solvent , 30 mg catalyst, t=12 h.
i
=13.0), 500 mg catalyst (0.7%Rh/C or 0.6%Ir/C), T=453 K, t=8 h, PH2 =50 bar, PHe =30 bar.
6
5
5% in the presence of Rh/C (Table 1, entries 1, 2). It also appears that
the nature of atmosphere had a strong influence on the selectivity to
,2-PDO and LA which were completely reversed using the iridium
catalyst just by changing the nature of the gas atmosphere. Under
, the formation of 1,2-PDO was promoted up to 37% yield (72% se-
detected. Surprisingly, LA was the major product whatever the nature
of the gas phase, and after 48 h high LA yields were reached under
1
both H
2
(67%) and He (75%).
The initial rates activities determined for the Ir/C catalyst are 18.1
−
1
−1
H
2
2
and 8.5 mol min .molIr under He and H respectively (Table 2, entries
−
1
−1
lectivity) (entry 8) [6,33]. In contrast under He pressure, high selec-
tivity to LA was achieved, up to 37% yield at 76% glycerol conversion
3, 4) and lower than 0.5 mol min
catalyst (entries 1, 2).
3
molIr when Ir/CaCO was used as
(
49% selectivity) at the expense of 1,2-PDO (9% selectivity, entry 9).
It is noteworthy that under hydrogen, the 1,2-PDO/LA relative ratio
To highlight the differences between both catalysts, in terms of
product distributions, the selectivities in 1,2-PDO and LA were com-
pared at similar glycerol conversion. At ca 20% conversion, the highest
LA selectivity (86%) was achieved using 5% Ir/CaCO catalyst in the ab-
3
sence of hydrogen (entry 1). As mentioned before, even under
hydrogen atmosphere, LA was also the major product with a selectivity
was significantly affected by the temperature (Fig. 2a). Below 413 K, LA
was produced predominantly, but with low glycerol conversion (b10%).
Above this temperature, 1,2-PDO was the main product. Whatever the
temperature, the discrepancy on mass balance between TOC analysis
and product analysis, was less than 10%. Under helium atmosphere,
the selectivity in LA reached 50–60% whatever the temperature
of 48% in the presence of 5% Ir/CaCO
Ir/C catalyst, the selectivity to LA reached 51% and 29% under helium
and hydrogen atmosphere, respectively. The CaCO support was conse-
quently found to promote a high LA selectivity, while the carbon sup-
port favored the formation of 1,2-PDO. Using Ir/CaCO catalyst, an
3
(entry 2). In the presence of 0.6%
(Fig. 2b). However, it was shown that the amount of by-products in-
3
creased steadily from 393 K up to 473 K at the expense of 1,2-PDO.
A continuous increase of the pressure within the reactor was ob-
served during the reaction under He atmosphere indicating the for-
mation of gaseous species as reported in the presence of Pt-based
catalyst [21]. However, as mentioned previously, 94% of initial carbon
was analyzed in the liquid phase by TOC analysis. Analysis of the gas
phase carried out at the end of the reaction have shown only traces of
CO and ethylene together with hydrogen as the main gas, which for-
mation is consistent with a dehydrogenation step (Table 1, entry 9).
At the light of these results we focused the study on the use of irid-
ium catalyst, and the influence of the nature of the support. The reac-
tion was carried out in the presence of the commercial 5 wt.%
3
outstanding LA yield of 75% was obtained at 96% glycerol conversion
under He atmosphere after 48 h (Fig. 3b), the selectivity being as high
as 78%. The 1,2-PDO yield was remarkably low (ca 5%). Under H , high
2
selectivity into LA was also achieved together with the formation of
ca 12% of 1,2-PDO (Fig. 3a). Whatever the support, no iridium was
detected in solution after the reaction.
3.4. Mechanistic considerations
One should consider that the presence of 1,2-PDO in significant
concentrations after the reaction, even under inert gas pressure, of-
fers a convenient proof of the ability of these metals to carry out the
deoxygenation reaction in the absence of added hydrogen. Then, we
3 2
Ir/CaCO under H and He atmosphere (Fig. 3).
As previously, LA and 1,2-PDO were the main products of the reac-
tion, and only traces of acetol, formic acid, ethylene glycol were
3
Fig. 3. Evolution of the concentration of glycerol (♦), lactic acid (▲) and 1,2-PDO (■) with the time in the presence of Ir/CaCO (a) under hydrogen, (b) under helium.

148
F. Auneau et al. / Catalysis Communications 16 (2011) 144–149
Table 2
Influence of the atmosphere on the conversion of glycerol in alkaline conditions.
Table 3
2
Catalytic transformation of 1,2-PDO and LA in the presence of Ir/C under He or H .
Entry Catalyst Gas Initial activity
mol min− molIr
Time Conv. LA Y(Sel) 1,2-PDO,
Entry
Substrate
Gas
Conv
(%)
pH
f
Mass
balance
Yield (%)
1
−1
(
)
(h)
(%)
(%)
Y(Sel) (%)
LA
Others a
(%)
1
2
3
4
Ir/CaCO
Ir/CaCO
Ir/C
3
3
He 0.4
6
6
1
2
21
23
25
21
18 (86)
11 (48)
13 (51)
6 (29)
1(3)
2 (8)
6 (22)
14 (67)
H
2
0.2
He 18.1
8.5
1
2
3
1,2-PDO
1,2-PDO
LA
H
2
13
78
0
13.0
13.0
12.9
100
93
–
3
18
–
3
9
ε
He
He/H
2
Ir/C
H
2
Reaction conditions: T=453 K,
glycerol solution 1 M NaOH, 500 mg catalyst.
P
H2 =50 bar,
PHe =30 bar, pH
i
=13.0; 100 mL 5%
6 mL 5 wt.% substrate in NaOH 1 M, 30 mg catalyst, T=453 K, t=12 h, pH
i
=13.0
a
Others: EG, 1,3-PDO, formic acid, acetic acid, acetol, ethanol, methanol.
examined the evolution of LA and 1,2-PDO selectivity as a function of
glycerol conversion in the presence of 0.6% Ir/C under hydrogen or
helium pressure (Fig. 4) at 453 K.
LA and 1,2-PDO were simultaneously formed at different levels
depending on the nature of the gas. Under hydrogen (Fig. 4a), glycer-
ol yielded mainly 1,2-PDO (more than 65%) while LA was the minor
product (b30%). The distribution between LA and 1,2-PDO evolved
smoothly all over the conversion of glycerol with a slight increase of
the propanediol selectivity and small diminution of LA selectivity
their unstability under the reaction conditions as observed from addi-
tional experiments (i.e. by treating these compounds in alkaline
media) or to their strong adsorption on the metallic surface. Acetol
is also unstable in such alkaline conditions. Under reductive atmo-
sphere, the hydrogenation step is favored yielding 1,2-PDO as the
main product. Under inert atmosphere, the second step is slow and
the competitive irreversible Cannizaro reaction catalyzed by base is
favored, leading to the formation of lactic acid mainly. Besides,
some polyethers could be formed concurrently. These unanalysed
oligomers could account for the differences observed between the
conversions and the sum of the yields.
Moreover, 1,2-PDO is also reactive under the reaction conditions.
It may be suggested that under helium, 1,2-PDO could be reversiblely
dehydrogenated on iridium catalyst, first to acetol, then to pyruvalde-
hyde that is further converted to lactate under alkaline conditions
according to the proposed mechanism, this route being inhibited
under hydrogen atmosphere as mentioned for Cu-catalyzed reactions
(Scheme 2) [16]. The production of oligomers is also potential.
(
less than 10% difference). At 85% conversion, the selectivity in 1,2-
PDO reached 76% after 24 h. On the other hand, under He (Fig. 4b),
,2-PDO selectivity decreased regularly from ca 20% to less than 5%
1
at 96% conversion whereas the selectivity to LA just slightly increased.
We also evaluated the conversion of 1,2-PDO and LA under both
reaction conditions (Table 3). After 12 h, in the presence of hydrogen,
1
3% of 1,2-PDO were converted into a mixture of LA and formic acid
as the main products (entry 1). Under inert gas, the conversion of
,2-PDO was much higher and reached 78% (entry 2), this correlates
with the decrease of this product under helium (Fig. 4b). Lactic acid
18%) and formic acid were the main products together with a large
1
3
When using the Ir/CaCO catalyst, the reaction rate is much lower
(
whatever the atmosphere and so the Cannizaro reaction prevails lead-
ing to the formation of lactic acid regardless of the nature of the gas. In
addition the two supports have entirely different texture characteris-
tics. The mesoporous carbon has a high BET surface area (1220 m²
amount of unknown products. This indicates that 1,2-PDO was reac-
tive under reaction conditions. Contrarly to 1,2-PDO, lactic acid was
stable under the reaction conditions and no transformation occurred
−
1
−1
either under He or H
2
atmosphere (entry 3). These results are in ac-
g
3
) while the one of the CaCO support is as low as 8 m² g . It en-
cordance with the observed decrease of 1,2-PDO selectivity with the
time.
tails that the reactivity of glycerol and the intermediates of the reac-
tion on iridium catalysts are deeply affected by the nature of the
support.
Based on these data and as proposed in the literature, the first step in
alkaline conditions is thought to be the dehydrogenation of glycerol
into glyceraldehyde on a metallic surface. The dehydrogenation of glyc-
erol to dihydroxyacetone is unlikely, because it has never been ob-
served. Besides, its formation is a dead-end, as it cannot be dehydrated.
Under the conditions used, no reforming reaction occurs since al-
most no carbon-containing species were detected in the gas phase
and all the organic carbon is recovered in the aqueous phase as sug-
gested from TOC analysis. Subsequent dehydration leads to the for-
mation of the enol in equilibrium with pyruvaldehyde that can
either be hydrogenated over the metal into 1,2-PDO or converted
via a Cannizzaro reaction to LA, that reaction being also favored in
the presence of a metal [34] (Scheme 2).
4. Conclusion
In conclusion, we have shown that iridium catalysts efficiently
promoted the conversion of glycerol either to 1,2-PDO or to lactic
acid under hydrogen or helium atmosphere, respectively. The use of
3
CaCO as support also enabled to favor LA formation (77% selectivity)
but with moderate activity (up to 75% LA yield in 48 h) while 1,2-PDO
was produced with up to 76% selectivity in the presence of the more
active 0.6% Ir/C catalyst.
Further works are in progress to improve either LA or 1,2-PDO se-
lectivity and to increase reaction rate. In addition, to get a better esti-
None of the key aldehydes (i.e. glyceraldehyde, pyruvaldehyde)
was detected during the reaction. This can be attributed either to
3
mation of the influence of the support, CaCO supported catalysts
with lower iridium loading are under development as well as
a 8
b
0
80
PDO
LA
7
6
5
4
3
2
1
0
0
0
0
0
0
0
70
60
50
40
30
20
10
0
PDO
LA
0
0
20
40
60
80
100
0
20
40
60
80
100
Conversion (%)
Conversion (%)
2
Fig. 4. Lactic acid and 1,2-propanediol selectivities as a function of glycerol conversion under a) H or b) He atmosphere.

F. Auneau et al. / Catalysis Communications 16 (2011) 144–149
149
H2
H2
[7] Y. Nakagawa, Y. Shinmi, S. Koso, K. Tomishige, Journal of Catalysis 272 (2010)
191.
OH
OH
O
OH-
O
OH
OH
OH
[
8] T. Kurosaka, H. Maruyama, I. Naribayashi, Y. Sasaki, Catalysis Communications 9
(2008) 1360.
OH
HO
,2-propanediol
OH
metal
metal
[9] A. Alhanash, E. Kozhevnikova, I. Kozhevnikov, Catalysis Letters 120 (2008) 307.
[10] T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomishige, Journal of Catalysis 240
(2006) 213.
11] J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel, C. Rosier, Green Chemistry
6 (2004) 359.
1
H O
2
glyceraldehyde
[
metal
H2
[
12] M. Besson, L. Djakovitch, P. Gallezot, C. Pinel, A. Salameh, M. Vospernik, Chemical
Industries (Boca Raton, FL, United States) 2009, 123(Catalysis of Organic Reactions),
O
OH
O
metal
O
Cannizzaro
313–318.
O
HO
[
13] C. Montassier, J.C. Ménézo, L.C. Hoang, C. Renaud, J. Barbier, Journal of Molecular
Catalysis 70 (1991) 99.
NaO
sodium lactate
reaction
-
OH
pyruvaldehyde H2
acetol
[14] Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catalysis Communications 6
2005) 645.
(
[
[
15] D. Lahr, B.H. Shanks, Journal of Catalysis 232 (2005) 386.
16] C. Montassier, D. Giraud, J. Barbier, J.P. Boitiaux, Bulletin de la Societe Chimique de
France (1988) 148.
Scheme 2. Potential routes for the transformation of glycerol into 1,2-propanediol and
lactic acid.
[
[
[
17] B. Casale, A.M. Gomez, European Patent 92830357.7 (07.07.92).
18] E.P. Maris, R.J. Davis, Journal of Catalysis 249 (2007) 328.
19] E.P. Maris, W.C. Ketchie, M. Murayama, R.J. Davis, Journal of Catalysis 251 (2007)
different supports. To deeper understand the elemental mechanistic
steps, density functional theory (DFT) calculations are under way to
bring new knowledges from a theoretical approach of the problem.
281.
[
20] Y. Fan, C. Zhou, X. Zhu, Catalysis Reviews 51 (2009) 293.
[21] E. D'Hondt, S. Van de Vyver, B.F. Sels, P.A. Jacobs, Chemical Communications
2008) 6011.
(
[
[
[
22] D. Roy, B. Subramaniam, R.V. Chaudhari, Catalysis Today 156 (2010) 31.
23] J.W. Shabaker, G.W. Huber, J.A. Dumesic, Journal of Catalysis 222 (2004) 180.
24] R.R. Soares, D.A. Simonetti, J.A. Dumesic, Angewandte Chemie, International
Edition 45 (2006) 3982.
25] A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A.A. Lemonidou, J.A. Lercher, Journal of
Catalysis 269 (2010) 411.
26] I. Gandarias, P.L. Arias, J. Requies, M.B. Güemez, J.L.G. Fierro, Applied Catalysis B:
Environmental 97 (2010) 248.
27] Y. Shen, S. Zhang, H. Li, Y. Ren, H. Liu, Chemistry A European Journal 16 (2010)
Acknowledgements
We are indebted to EU (Topcombi Contract No: NMP2-CT2005-
[
[
[
[
[
5
15792) for financial support. We thank Johnson Matthey for a
gift of catalyst (Ir/CaCO ). FA acknowledges the French Minister of
Research and Education for a PhD grant. Mimoun Aouine, Françoise
Bosselet and Chantal Lorentz are acknowledged for their assistance
3
7368.
(
TEM, XRD and gas phase analysis respectively). Carine Michel,
28] H. Kishida, F. Jin, Z. Zhou, T. Moriya, H. Enomoto, Chemistry Letters 34 (2005)
1560.
29] C.A. Ramirez-Lopez, J.R. Ochoa-Gomez, M. Fernandez-Santos, O. Gomez-Jimenez-
Aberasturi, A. Alonso-Vicario, J. Torrecilla-Soria, Industrial and Engineering
Chemistry Research 49 (2010) 6270.
Françoise Delbecq and Philippe Sautet from ENS Lyon and Pierre
Gallezot are acknowledged for helpful discussions.
References
[30] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Green
Chemistry 9 (2007) 582.
[
[
[
[
1] M. Pagliaro, M. Rossi, The Future of Glycerol, RSC Green Chemistry Book Series,
RSC Publishing, 2008.
2] T.S. Viinikainen, R.S. Karinen, A.O.I. Krause, Catalysis for Renewables, Wiley-VCH,
Weinheim, 2007, pp. 209–222.
3] B. Sels, E. D'Hondt, P. Jacobs, Catalysis for Renewables, Wiley-VCH, Weinheim,
[31] Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Applied Catalysis B:
Environmental 94 (2010) 318.
[32] J. Barrault, Y. Pouilloux, J.M. Clacens, C. Vanhove, S. Bancquart, Catalysis Today 75
(2002) 177.
[33] M. Balaraju, V. Rekha, P.S. Sai Prasad, B.L.A. Prabhavathi Devi, R.B.N. Prasad, N. Lingaiah,
Applied Catalysis A: General 354 (2009) 82.
[34] J. Cook, P. Maitlis, Journal of the Chemical Society, Chemical Communications
(1981) 924.
2
007, pp. 223–256.
4] C.-H. Zhou, J.N. Beltramini, Y.-X. Fan, C.Q. Lu, Chemical Society Reviews 3 (7)
2008) 527.
(
[
[
5] A. Behr, J. Eilting, K. Irawadi, J. Lenschinski, F. Lindner, Green Chemistry 10 (2008) 13.
6] M.A. Dasari, P.-P. Kiatsimkul, W.R. Sutterlin, G.J. Suppes, Applied Catalysis A:
General 281 (2005) 225.