Heterogeneous catalytic hydrogenation of biobased levulinic and succinic acids in aqueous solutions

Article abstract of DOI:10.1002/cssc.201300608

Supported noble-metal catalysts (Ru, Pd or Pt) and the corresponding Re-promoted catalysts exhibit a high activity for the hydrogenation of biobased carboxylic acids. Levulinic acid and succinic acid are converted into the lactones or the diols depending on the nature of the catalyst and the reaction conditions. The highest selectivity to 1,4-pentanediol of 82 % is achieved at 140 °C in the presence of the 1.9 % Ru-3.6 % Re/C catalyst. Switching between diols and lactones: This article focuses mainly on the comparison of the reactivity of two biobased carboxylic acids, namely levulinic acid and succinic acid, in water in the presence of carbon-supported noble metals. The optimization of the reaction conditions allows to increase the yield and selectivity to diols (butanediol and pentanediol). Thus, it is possible to obtain pentanediol from levulinic acid under milder conditions compared to succinic acid. Copyright

Full text of DOI:10.1002/cssc.201300608

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201300608
Heterogeneous Catalytic Hydrogenation of Biobased
Levulinic and Succinic Acids in Aqueous Solutions
[
a, b]
[a]
[a]
[c]
Louis Corbel-Demailly,
Bao-Khanh Ly, Doan-Pham Minh, Benoit Tapin,
[c]
[c]
[b]
[b]
Catherine Especel, Florence Epron, Amandine Cabiac, Emmanuelle Guillon,
Michꢀle Besson, and Catherine Pinel*
[a]
[a]
Supported noble-metal catalysts (Ru, Pd or Pt) and the corre-
sponding Re-promoted catalysts exhibit a high activity for the
hydrogenation of biobased carboxylic acids. Levulinic acid and
succinic acid are converted into the lactones or the diols de-
pending on the nature of the catalyst and the reaction condi-
tions. The highest selectivity to 1,4-pentanediol of 82% is ach-
ieved at 1408C in the presence of the 1.9%Ru–3.6%Re/C
catalyst.
Introduction
Nowadays, a large range of acids are produced from biomass
either through chemical or biochemical transformations. The
production of levulinic acid (LevA) from lignocellulosic deriva-
tives under acidic conditions has been described since the
reaction conditions and the nature of the catalysts
[4–7]
(Scheme 1).
To avoid the use of undesirable stoichiometric hydride re-
ductants, the catalytic conversion of biobased carboxylic acids
using molecular hydrogen under pressure has received consid-
th
early years in the 20 century; however, large-scale applica-
tions were reported more re-
cently that involve a continuous
process based on the Biofine
[
1]
technology. On the other hand,
succinic acid (SucA) is now pro-
duced through an economically
viable process from renewable
sources by different compa-
[
2,3]
nies.
These acids are considered to
be platform molecules that can
be further transformed into
value-added products. The hy-
drogenation of these acids leads
to the formation of the corre-
sponding lactones, diols, or
cyclic ethers depending on the
Scheme 1. Catalytic hydrogenation of SucA and LevA.
erable attention. The hydrogenation of SucA has been de-
[
a] L. Corbel-Demailly, B.-K. Ly, Dr. D.-P. Minh, Dr. M. Besson, Dr. C. Pinel
IRCELYON
[8–13]
scribed using a large range of reaction conditions.
We re-
Institut de recherches sur la catalyse et l’environnement de Lyon
UMR 5256 CNRS-Universitꢀ Lyon1
cently reported efficient supported mono- or bimetallic cata-
lysts for the selective hydrogenation of SucA in the aqueous
2
Avenue Albert Einstein, 69626 Villeurbanne (France)
[8,9]
phase. In the presence of Pd- or Ru-supported monometallic
E-mail: catherine.pinel@ircelyon.univ-lyon1.fr
catalysts, g-butyrolactone (GBL) was the main product, whereas
in the presence of Pd–Re or Ru–Re catalysts 1,4-butanediol
(1,4-BDO) was formed predominantly. We have shown that the
Re/M ratio (M=Pd or Ru) and the nature of the support affect-
ed the reaction rate as well as the selectivity for the desired
product. However, only few reports on the hydrogenation of
LevA have been published. A few homogeneous and heteroge-
neous catalysts mainly for the formation of g-valerolactone
[
b] L. Corbel-Demailly, Dr. A. Cabiac, Dr. E. Guillon
IFPEN-Lyon, BP3
6
9360 Solaize (France)
[
c] B. Tapin, Dr. C. Especel, Dr. F. Epron
IC2MP
Institut de Chimie des Milieux et Matꢀriaux de Poitiers
UMR 7285 CNRS-Universitꢀ de Poitiers
4
rue Michel Brunet, 86022 Poitiers (France)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201300608.
[10–12]
(GVL) have been described.
The presence of a solid-acid
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2388

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
co-catalyst increased the reaction rate of the first step of the
reaction (formation of GVL, Scheme 1); an initial activation of
the carbonyl group for the hydrogenation to the intermediate
g-hydroxyvaleric acid was proposed, followed by the favorable
[11]
intramolecular esterification reaction to yield the lactone. Ru-
based complexes containing sulfonated phosphine ligands
have been reported to selectively hydrogenate LevA to GVL,
[
12]
even in the absence of a solvent. Wright and Palkovits com-
piled a list heterogeneous catalysts, which were developed to
perform this transformation under a large range of reaction
[
13]
conditions. More drastic conditions are needed to achieve
the syntheses of 1,4-pentanediol (1,4-PDO) or methyltetrahy-
drofurane (MTHF). Geilen et al. reported the selective hydroge-
nation of LevA into 1,4-PDO (95% yield, 18 h) at 1608C and
100 bar in the presence of the Ru-Triphos complex [Triphos=
1
,1,1-tris(diphenylphosphino-methyl)ethane, 0.1 mmol%] as
[
14]
homogeneous catalyst.
Addition of an acidic ionic liquid
shifted the reaction to the selective formation of MTHF (92%)
under identical reaction conditions. The selective hydrogenoly-
sis in ethanol at 2008C and 60 bar of the intermediate GVL
into 1,4-PDO has been reported to be efficiently performed in
[
15]
the presence of a Cu-ZrO catalyst (30 wt%).
2
Herein, the performances of mono- and bimetallic catalysts
in the hydrogenation of LevA are reported, with a special
focus on the selectivity for 1,4-PDO. The reactivity of LevA acid
will be compared with the behavior of SucA.
Results and Discussion
X-Ray diffraction (XRD) patterns of the Pd and Ru monometal-
lic catalysts showed no diffraction peak that could be attribut-
ed to the metallic phase, suggesting a small particle size (Fig-
ure 1b). TEM analysis confirmed the presence of only small Pd
or Ru particles, with the size distribution centered on 2 nm
Figure 1. a) N adsorption–desorption isotherms of commercial (gray) and
oxidized (black) carbon; b) diffractograms of active carbon-supported mono-
metallic catalysts. *=diffraction peaks attributed to Pt.
2
(
Figure 2). On the other hand, diffraction peaks attributed to Pt
reactions. Re–O interactions, evidenced through extended X-
ray absorption fine structure (EXAFS) spectroscopy, were re-
ported to be responsible for the variation in catalyst perform-
crystallites were observed by means of XRD, indicating that
larger particles were present on the support (d=5 nm accord-
[
16]
[17,18]
ing to the Debye–Scherrer equation, Figure 1b).
ances.
The evaluation of the catalytic performances of the mono-
metallic catalysts was carried out for both LevA and SucA hy-
drogenation. As an illustration, the evolution of LevA and GVL
concentrations in the presence of the different monometallic
catalysts is reported on Figure 3.
The bimetallic catalysts prepared from these monometallic
parent catalysts were evaluated under the same reaction con-
ditions. The conversion of LevA and the formation of GVL, in
the presence of the bimetallic catalysts during 2 h, are report-
ed in Figure 4.
The Ru catalyst (2.1% Ru/C) was the most efficient in the hy-
drogenation of LevA; after 10 min of reaction, complete con-
version was achieved. In contrast, under the same reaction
conditions, the Pd (2.7% Pd/C) and Pt (4.1% Pt/C) catalysts ex-
hibited a much lower activity and the conversion after 2 h
reached only 70% and 53% in the presence of Pd/C and Pt/C,
respectively. As expected, GVL was the main product formed in
the initial step of the hydrogenation reaction. In the case of
the Ru catalyst, this intermediate was further transformed
mainly to 1,4-PDO. However, the carbon balance reached only
Similar performances were observed for the mono- and bi-
metallic catalysts based on Ru, that is to say that the hydroge-
nation rate of LevA was very high and full conversion was ach-
ieved after 10 min. However, using the Pd- and Pt-based cata-
lysts, the deposition of Re significantly increased the reaction
rate compared to the monometallic parent catalyst and full
conversion of LevA was achieved after 2 and 7 h in the pres-
ence of Pd–Re/C and Pt–Re/C, respectively. In a second step,
regardless of the metallic system, GVL could be further hydro-
genated to a mixture of 1,4-PDO and MTHF (Scheme 1) togeth-
er with some minor byproducts, such as 2-butanol (2-BuOH)
and 2-pentanol (2-PeOH), which were identified by using GC–
MS (see the Supporting Information). At the beginning of the
reaction, the pH value increased rapidly from 2.6 to 3.3, corre-
8
0% due to the occurrence of a large amount of cracking reac-
[8]
tions as previously observed for the hydrogenation of SucA.
Addition of Re was shown to modify the performances of
the supported noble-metal catalysts in several hydrogenation
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2389

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Figure 2. Particle-size distributions determined from measurements of 200 particles of a) Ru-based catalysts (2.8%Ru–3.9%Re/C) and b) Pd-based catalysts
2%Pd–4%Re/C) and the respective TEM images.
(
Figure 4. Hydrogenation of LevA and formation of GVL in the presence of
bimetallic catalysts [RuRe (*); PtRe (~); PdRe (
)]. Evolution of main prod-
Figure 3. Hydrogenation of LevA in the presence of monometallic catalysts
2.1% Ru/C (*); 4.1% Pt/C (~); 2.7% Pd/C (
)]. Evolution of main products
&
[
&
ucts as a function of time: full line LevA, dotted line GVL. Reaction condi-
tions identical to Figure 3.
as a function of time: full line LevA, dotted line GVL. Reaction conditions: Le-
vA=430 mmolL (150 mL), Molar ratio LevA/Ru=2070, LevA/Pt=2050,
À1
LevA/Pd=1710, 1608C, 150 bar.
sponding to the transformation of LevA into GVL, and then
a further increase up to 4.7, reaching complete conversion of
GVL.
version of GVL was achieved after 24 h, yielding mainly 1,4-
PDO (55% selectivity, Figure 5a) together with 2-PeOH (8%),
MTHF (5%), and 2-BuOH (3%) in solution (Figure 5b). We con-
firmed that the hydrogenation of MTHF under similar condi-
tions yielded mainly 2-PeOH (5% after 24 h) together with
In the presence of 2.8%Ru–3.9%Re/C the highest GVL yield
(85%) was observed after 10 min (Figure 4), whereas total con-
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2390

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Figure 5. Hydrogenation of LevA: Evolution of the concentration of a,c,e) the main products and b,d,f) the by-products as a function of time (160 8C,
50 bar) in the presence of 2.8%Ru–3.9%Re/C (a,b), 3.3%Pt–4.1%Re/C (c,d), and 2.7%Pd-3.9%Re/C e), f). g) TOC concentration (expressed as C equivalent
molar concentration) as a function of the conversion of (LevA+GVL) [RuRe (*); PtRe (~); PdRe (
)]. For clarity, LevA concentration was not reported.
1
5
&
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2391

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
a small amount of 2-BuOH, whereas 1-PeOH was not detected.
Furthermore, a significant number of cracking reactions oc-
curred and only 65% of the initial total carbon loading could
be detected in the liquid phase at the end of the reaction
[from both HPLC and total organic carbon (TOC) analysis], indi-
cating the formation of a very large amount of gaseous prod-
ucts (Figure 5 g). On the other hand, in the presence of
3.3%Pt–4.1%Re/C, a very low transformation rate of GVL was
observed and less than 20% GVL conversion was achieved
after 24 h; low amounts of both 1,4-PDO and MTHF were
formed (Figure 5c and d). This low conversion can be partly at-
tributed to the larger particle size of Pt compared with Ru and
Pd. Using the 2.7%Pd–3.9%Re/C catalyst, significantly higher
conversions were achieved compared to the monometallic cat-
alyst: LevA completely disappeared after 1 h and subsequent
hydrogenation of the GVL intermediate was observed, yielding
mainly 1,4-PDO (Figure 5e) and a minor amount of MTHF (Fig-
ure 5 f). As mentioned for Ru–Re/C, the amount of TOC in the
solution decreased drastically, indicating that there was also
formation of gaseous products, but to a lesser extent in com-
parison to the monometallic catalyst (Figure 5g).
The efficiency of Ru–Re/C and Pd–Re/C catalysts were com-
pared for both LevA and SucA hydrogenation, and the more
active one (Ru–Re/C) was evaluated at a higher concentration
(15 wt%) (Figure 6). In a previous study on SucA hydrogena-
tion, it was observed that the selectivity to 1,4-BDO did not
[
9]
change for a range of SucA concentrations. In the presence
of Ru–Re/C and at 1608C, full conversion of LevA was attained
after 10 min, whereas SucA was only completely converted
after more than 40 h (Figure 6a and b). This result is in accord-
ance with the different reactivities of ketone and acid function-
Figure 6. Evolution of the concentration of the main products as a function
À1
of time (1608C, 150 bar, [LevA]
0
=[SucA]
0
=1400 mmolL ) in the presence
of RuRe/C for a) SucA hydrogenation and b) LevA hydrogenation.
[
19]
al groups towards hydrogenation (Scheme 2). In both cases,
the lactones (GVL or GBL) were subsequently hydrogenated
into the corresponding diols (1,4-PDO and 1,4-BDO, respective-
ly) as the main product. However, the reaction rates were dif-
ferent due to the fact that hydrogenation of SucA is consider-
ably slower. In the case of LevA, hydrogenation reactions of
LevA and GVL are successive reactions, whereas in the case of
SucA, SucA and GBL are reduced simultaneously. The maxi-
mum GVL yield (formed from LevA) was obtained after 30 min,
whereas the maximum GBL yield
(
formed from SucA) was only de-
tected after 25 h. Moreover, after
days of reaction, a 15% mass
2
balance loss was measured
when SucA was used as the sub-
strate, whereas a 20% loss was
detected in the case of LevA.
The maximum selectivity to 1,4-
BDO (from SucA) was 62%,
whereas 54% 1,4-PDO (from
LevA) selectivity was achieved.
Independent of the substrate,
the selectivity to the cyclic
ethers (THF or MTHF) remained
low (<10%).
Similar behaviors were ach-
ieved using the Pd–Re catalyst
for the hydrogenation of
a
5
wt% LevA or SucA solution
Scheme 2. Different reactivities of LevA and SucA during hydrogenation reaction.
(Figure 7a and b). The hydroge-
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2392

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
À1
Figure 7. Evolution of the concentration of the main products as a function of time (1608C, 150 bar, [LevA]
0
=[SucA]
0
=430 mmolL ) in the presence of
PdRe/C for a) SucA hydrogenation, b) LevA hydrogenation and for the by-products of c) SucA hydrogenation and d) LevA hydrogenation.
nation rate of LevA was significantly higher than that of SucA;
the corresponding lactones were transformed mainly to diols,
with less than 10% formation of cyclic ethers (Figure 7c and
d). Using this bimetallic catalyst, 66% selectivity for 1,4-BDO
and 15% carbon loss were observed during SucA hydrogena-
tion, whereas 1,4-PDO was formed with 69% selectivity at 16%
carbon loss during LevA hydrogenation. In both cases, the for-
mation of the corresponding primary monoalcohols (1-BuOH
and 1-PeOH) was observed but with a small selectivity (<5%).
These results clearly show that the differences observed be-
tween LevA and SucA are only related to the first hydrogena-
tion step and that the behavior of both lactones is similar.
To reduce the undesirable cracking products observed for
LevA, we carried out the reaction at lower temperatures. The
first rapid hydrogenation of LevA to GVL was not significantly
affected by the temperature (in the range from 1208 to 1608).
However, as expected, the lower the temperature, the slower
the second hydrogenation step. In the presence of Ru–Re/C,
GVL was completely hydrogenated after 22 h at 1608C (Fig-
ure 5a), whereas 48 h and 72 h were necessary to achieve
complete hydrogenation at 140 and 1208C, respectively. How-
ever, the selectivity to 1,4-PDO significantly increased from
products that derive from cracking of the substrate or the in-
termediates.
Similarly, the hydrogenation of LevA in the presence of Pd–
Re/C was carried out at 1408C. The hydrogenation rate was
significantly decreased, and even after 48 h 28% GVL was still
present in the solution.
Figure 8. Influence of temperature on the product distribution at 100% GVL
69% to up to 80% with a decrease of the temperature
À1
conversion. Reaction conditions: LevA=430 mmolL , 2.2%Ru–3.8%Re/C,
(Figure 8). Simultaneously, the selectivity to ether (MTHF) and
molar ratio LevA/Ru=2000, 150 bar. Reaction time: 1608C=22 h;
monoalcohols (1-PeOH, 2-PeOH, and 2-BuOH) decreased. This
was mainly attributed to a reduced amount of degradation
1408C=48 h; 1208C=72 h. Alcohols=1-PeOH+2-PeOH+2-BuOH [alcohols
&
); MTHF ( ); PDO ( )].
&
&
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2393

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
The influence of pressure on LevA hydrogenation was stud-
ied by performing the reaction at 120 bar (1408C in the pres-
ence of Ru–Re/C as catalyst, see the Supporting Information).
The hydrogenation rate of GVL was only slightly reduced and
full conversion was achieved after 54 h compared to 48 h at
150 bar. On the other hand, no significant influence on the se-
lectivity was observed (71% selectivity to 1,4-PDO).
We have previously demonstrated that in the hydrogenation
of SucA the Re/M ratio strongly affected the activity of the cat-
alysts, which was the main factor influencing the selectivity to
1,4-BDO. Different Ru–Re/C catalysts with various Re loadings
(1.8 to 4.6 wt%) were prepared and evaluated for the hydroge-
nation of LevA at 1408C. The properties of the Ru-based cata-
lysts for ketone hydrogenation were significantly affected by
the amount of loading of Re: the initial rate of the hydrogena-
tion of LevA to GVL decreased as the amount of Re was in-
creased (Figure 9a). On the other hand, the reaction rate of
the second step was much less affected by the Re/Ru ratio
(Figure 9b): the hydrogenation rate of GVL was only slightly
higher in the presence of the 1.9%Ru–1.8%Re/C catalyst com-
pared with 3.6% and 4.6%Re loadings. We observed that the
selectivity to MTHF was not affected by the Re amount, where-
as the selectivity to 1,4-PDO increased and the selectivity to
monoalcohols decreased (Table 1). The subsequent hydroge-
nolysis reactions seem to have been partly suppressed because
of the presence of Re. These results are in agreement with ob-
[8]
servations we reported previously for SucA hydrogenation.
As mentioned above, few studies on the hydrogenation of
LevA into 1,4-PDO have been published. The best catalytic re-
sults reported in literature are summarized and compared with
our best result in Table 2. All reports showed that drastic con-
ditions (pressure or temperature) are needed to obtain a good
selectivity to 1,4-PDO. The Ru–Re catalytic system can efficient-
ly and selectively hydrogenate LevA to 1,4-PDO in water.
Figure 9. Influence of the xRe% in the 1.9%Ru–x%Re/C catalysts [x=1.8
(c), 3.6 (b), and 4.6(a)] on the conversion of a) LevA (^) and b) GVL
À1
(
*) to PDO (
&
). Reaction conditions: [LevA]=430 mmolL , 1408C, 150 bar.
In parentheses: Re wt%.
Conclusions
Noble-metal (Ru, Pd, or Pt) supported catalysts and
the corresponding Re-promoted catalysts exhibit
a high activity for the hydrogenation of biobased car-
boxylic acids. Both LevA and SucA are converted into
the corresponding lactones or diols depending on
the nature of the catalyst and the reaction condi-
tions. Although the reactivity of LevA and SucA is sig-
nificantly different because of the nature of the first
functional group to be hydrogenated (carbonyl or
carboxyl group), the behavior of intermediate lac-
tones (GBL and GVL) is similar regardless of the cata-
lyst. A maximum selectivity to 1,4-PDO of 82% was
achieved at 1408C in the presence of 1.9%Ru–
Table 1. Influence of Re loading on the 1,4-PDO yield.
[
a]
[b]
Re content Molar ratio Yield [%]
Carbon balance
[wt%]
Re/Ru
1,4-PDO 2-BuOH 2-PeOH MTHF [%]
1
3
4
.8
.6
.3
0.5
1
1.3
72
82
80
2
1
1
9
5
5
3
3
3
88
90
93
[a] Yields are for full conversion of GVL (t=24–28 h). [b] Carbon balance determined
by TOC analysis.
Table 2. Comparison of the best 1,4-PDO yield in this work with results published in
the literature for LevA hydrogenation.
3
.6%Re/C as catalyst. Further studies are in progress
[a]
Catalyst
Ratio
P
H2
T
Solvent
t
YieldPDO
Ref.
to optimize the preparation of the bimetallic catalyst
and to reach the highest diol yield.
LevA/metal [bar] [8C]
[h] [%]
1
.9%Ru–3.6%Re/C 2290
150
100
60
140 water
160 neat
200 ethanol
28 82
18 95
this work
[14]
[15]
[
b]
Ru-Triphos
Cu/ZrO2
4000
[
c]
10
6
95
[a] Yields are for full GVL conversion. [b] Homogeneous catalyst. [c] Molar ratio is for
GVL/Cu.
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2394

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
methane, carbon monoxide, etc.) at the end of the reaction by
using a mGC–MS (see the Supporting Information).
Experimental Section
Catalysts preparation
2
À1
Active carbon (Acticarbon L3S, 900 m g ) was used as supporting
material after oxidation with 15% NaOCl for 24 h to introduce car-
Acknowledgements
[20]
boxylic groups.
The oxidation procedure did not significantly
Part of this work was supported by the French Research National
Agency within the program HCHAIB (B-K. Ly, PhD grant) and by
IFPEN (L. Corbel-Demailly, PhD grant). Mimoun Aouine is ac-
knowledged for TEM measurements. The Cost action CM 0903
(UBIOCHEM) is also acknowledged.
affect the textural properties of the support [Figure 1a, S
=
BET
2
À1
8
50 m g (BET=Brunauer–Emmett–Teller)].
The monometallic catalysts were prepared through cationic ex-
change at room temperature. To a suspension of the oxidized sup-
À1
port in NH OH (1 molL ), an aqueous solution of the metallic salt
4
(
[Ru(NH ) Cl ], [Pt(NH ) Cl ] or [Pd(NH ) Cl ] was added to obtain 2–
3 6 3 3 4 2 3 4 2
4
wt% metal and the suspension was stirred for 24 h. After filtra-
Keywords: biomass · carboxylic acids · diols · heterogeneous
catalysis · hydrogenation
tion, washing, and drying, the solid was reduced under H2 at
008C for 3 h and finally passivated in 1% O /N gas flow. The bi-
3
2
2
metallic catalysts were prepared by impregnation of the monome-
tallic Ru and Pd solids by using an aqueous solution of NH ReO ,
evaporation of water, drying under vacuum at 1008C, reduction at
[1] S. W. Fitzpatrick, US5608105, 1995.
4
4
[
2] J. B. McKinlay, C. Vieille, J. G. Zeikus, Appl. Microbiol. Biotechnol. 2007,
6, 727.
3] J. J. Beauprez, M. De Mey, W. K. Soetaert, Process Biochem. 2010, 45,
103.
7
4
508C for 3 h, and passivation (1% O /N flow). The nominal load-
2 2
[
ing was 4 wt% Re.
The catalysts were characterized by elemental analysis (inductively
coupled plasma optical emission spectrometry), XRD analysis by
1
[
[
4] C. Delhomme, D. Weuster-Botz, F. E. Kuhn, Green Chem. 2009, 11, 13.
5] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539.
using a Siemens Bruker D5005 diffractometer with CuK radiation,
and by transmission electron microscopy (TEM, JEOL 2010)
equipped with an energy dispersive X-ray detector (EDX).
a
[6] J. Zhang, S. Wu, B. Li, H. Zhang, ChemCatChem 2012, 4, 1230.
[7] I. T. Horvꢂth, H. Mehdi, V. Fꢂbos, L. Boda, L. T. Mika, Green Chem. 2008,
10, 238.
[
8] D. Pham Minh, M. Besson, C. Pinel, P. Fuertes, C. Petitjean, Top Catal.
010, 53, 1270.
2
Hydrogenation reaction
[9] B. K. Ly, D. Pham Minh, C. Pinel, M. Besson, B. Tapin, F. Epron, C. Especel,
Top Catal. 2012, 55, 466.
SucA and LevA were of reagent grade (Aldrich).
[10] H. Mehdi, V. Fꢂbos, R. Tuba, A. Bodor, L. T. Mika, I. T. Horvꢂth, Top. Catal.
2008, 48, 49.
[11] A. M. R. Galletti, C. Antonetti, V. D. Luise, M. Martinelli, Green Chem.
All experiments were conducted in a 300 mL Hastelloy Parr 4560
high pressure reactor equipped with an electrically heated jacket,
a turbine stirrer with a variable speed magnetic driver, and a liquid
sampling line. In a typical reaction, the reactor was loaded with
LevA (7.5 g), water (142 g), and catalyst (1 g). After purging with Ar,
the reactor was heated to the desired temperature (typically
2
012, 14, 688.
[
12] J. M. Tukacs, D. Kirꢂly, A. Strꢂdi, G. Novodarszki, Z. Eke, G. Dibꢃ, T. Kꢄgl,
L. T. Mika, Green Chem. 2012, 14, 2057.
[
[
13] W. R. H. Wright, R. Palkovits, ChemSusChem 2012, 5, 1657.
14] F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J. Klankermay-
er, W. Leitner, Angew. Chem. 2010, 122, 5642; Angew. Chem. Int. Ed.
1
608C), and a H pressure (150 bar) was used to initiate the reac-
2
tion. The liquid samples taken from the reactor at regular intervals
were analyzed by using both HPLC chromatography equipped
with UV and refractive index detectors (ICSep Coregel 107H
column, 0.005 N H SO as mobile phase at a flow rate of
2
010, 49, 5510.
[15] X.-L. Du, Q.-Y. Bi, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N. Fan, Green Chem. 2012,
14, 935.
[16] H. P. Klug, L. E. Alexander, X-ray Diffraction Procedures for Polycrystalline
and Amorphous Materials, 2nd ed., Wiley, New York, 1974.
17] S. Koso, I. Furikado, A. Shimao, T. Miyazawa, K. Kunimori, K. Tomishige,
Chem. Commun. 2009, 2035.
2
4
À1
0.5 mLmin ) and gas chromatography (FFAP column, 30 m,
.25 mm using He as carrier gas). Acids, lactones, diols (1,4-penta-
[
0
nediol and 1,4-butanediol), cyclic ethers, and by-products (propion-
ic, butyric, and acetic acids; n-butanol and n-propanol) could be
quantified. GC–MS analysis was used for identification of the differ-
ent products. TOC was also measured by using a Shimadzu TOC-
VCHS analyzer. Comparison of the measured TOC with the TOC cal-
culated from chromatography analysis verified the material balance
based on the initial concentration of the acid and yielded an esti-
mation of gaseous products formed by cracking. The products con-
tained in the gas phase were analyzed (butane, propane, ethane,
[
18] M. Chia, Y. J. Pagꢂn-Torres, D. Hibbitts, Q. Tan, H. N. Pham, A. K. Datye,
M. Neurock, R. J. Davis, J. A. Dumesic, J. Am. Chem. Soc. 2011, 133,
12675.
[
19] P. Rylander, Hydrogenation Methods, Academic Press, London, 1985.
[20] P. Korovchenko, C. Donze, P. Gallezot, M. Besson, Catal. Today 2007, 121,
13.
Received: June 26, 2013
Published online on August 23, 2013
ꢁ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 2388 – 2395 2395