Hydrogenation of lactic acid to propylene glycol over a carbon-supported ruthenium catalyst

Article abstract of DOI:10.1016/j.molcata.2013.09.006

Catalytic hydrogenation of lactic acid to propylene glycol is performed in a high-pressure batch reactor over ruthenium on various carbon supports (i.e., VulcanXC-72, ketjen black, CNTs, CNFs, and graphite) prepared using the incipient wetness impregnation method. The crystallinity of the synthesized catalyst is investigated via X-ray diffraction, and the particle sizes are determined using transmission electron microscopy. The surface areas of the synthesized catalysts are analyzed using the BET method; the catalytic activity correlates remarkably with the BET surface area. The yield of propylene glycol increases with pressure, and the highest yield is achieved at 130 C. The catalytic activity is strongly dependent on the type of support. Among the catalysts tested, Ru on ketjen black shows the highest yield of propylene glycol.

Full text of DOI:10.1016/j.molcata.2013.09.006

Journal of Molecular Catalysis A: Chemical 380 (2013) 57–60
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Hydrogenation of lactic acid to propylene glycol over a
carbon-supported ruthenium catalyst
Hyuk Janga, Sung-Hwan Kima, Duwon Leea, Sang Eun Shima,b, Sung-Hyeon Baecka,∗
,
Beom Sik Kimb, Tae Sun Changb
a Department of Chemistry and Chemical Engineering, Inha University, Incheon, South Korea
b Korea Research Institute of Chemical Technology, 141 Gajeongro, Yuseong, Daejeon 305-600, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Catalytic hydrogenation of lactic acid to propylene glycol is performed in a high-pressure batch reactor
over ruthenium on various carbon supports (i.e., VulcanXC-72, ketjen black, CNTs, CNFs, and graphite)
prepared using the incipient wetness impregnation method. The crystallinity of the synthesized catalyst
is investigated via X-ray diffraction, and the particle sizes are determined using transmission electron
microscopy. The surface areas of the synthesized catalysts are analyzed using the BET method; the cat-
alytic activity correlates remarkably with the BET surface area. The yield of propylene glycol increases
with pressure, and the highest yield is achieved at 130 ◦C. The catalytic activity is strongly dependent on
the type of support. Among the catalysts tested, Ru on ketjen black shows the highest yield of propylene
glycol.
Received 2 September 2013
Accepted 8 September 2013
Available online 17 September 2013
Keywords:
Lactic acid
Propylene glycol
Hydrogenation
Ruthenium
© 2013 Elsevier B.V. All rights reserved.
Carbon support
1. Introduction
Miller et al. [13] investigated the production of propylene glycol
from lactic acid over various catalysts and reported a high yield of
Lactic acid is a very useful raw material for the food and medical
industries, and its consumption has increased steadily. The produc-
tion of lactic acid by fermentation of biomass is important to meet
the increasing demand. Via fermentation, lactic acid can be gener-
ated from renewable sources such as refined carbohydrates derived
from agricultural crops [1,2] and waste biomass [3]. Lactic acid is
known to be very reactive because it contains both hydroxyl and
carboxy functional groups. Accordingly, it can be converted into
a variety of chemicals through polymerization [3], condensation
[4–6], dehydration [7], esterification [8,9], and oxidation [10].
Propylene glycol is widely used in foods, consumer products,
and chemical applications and is produced industrially via the
hydration of propylene oxide [11]. Propylene glycol production
from lactic acid via a very good method was reported in the 1930s
(Fig. 1). However, in general the catalytic conversion of lactic acid to
propylene glycol requires high temperature and pressure. Dumesic
et al. reported the conversion of lactic acid to propylene glycol
over a silica-supported copper catalyst [12]. They proposed that
hydroxyl acryl species were formed by dissociative adsorption of
lactic acid; successive hydrogenation of the adsorbed species was
thought to comprise the main mechanism for the production of
propylene glycol.
propylene glycol using ruthenium supported on activated carbon
prepared via impregnation. According to their results, carbon is a
promising support because of its inertness and stability in aqueous
solution.
In the present work, ruthenium was supported on five differ-
ent types of carbon (i.e., VulcanXC-72, ketjen black (KNB), carbon
nanotubes (CNTs), carbon nanofibers (CNFs), and graphite) via the
incipient wetness impregnation method. The resultant supported
ruthenium catalysts were characterized by X-ray diffraction (XRD),
the Brunauer–Emmett–Teller (BET) method, and high-resolution
transmission electron microscopy (HR-TEM).
The catalytic activities of the catalyst towards the conversion
of lactic acid into propylene glycol were investigated in a liquid-
phase batch reactor, and the influence of the reaction temperature
and pressure on lactic acid conversion was also studied. The main
goal of this study is to investigate the effect of the support on the
catalytic activity of ruthenium.
2. Experimental
2.1. Preparation and characterization of supported ruthenium
catalyst
All carbon supports used in this study were obtained from
commercial sources: VulcanXC-72 from Carbot International, ket-
jen black from Mitsubishi Chemical, CNTs from Iljin Nanotech,
∗
Corresponding author. Tel.: +82 32 860 7474; fax: +82 32 860 8908.
E-mail address: shbaeck@inha.ac.kr (S.-H. Baeck).
1381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.09.006

58
H. Jang et al. / Journal of Molecular Catalysis A: Chemical 380 (2013) 57–60
Fig. 1. Conversion of lactic acid to propylene glycol.
CM-95 and CNFs from Carbon Nano-material Technology, and
graphite from Aldrich.
Prior to the impregnation of ruthenium, all carbon supports
were immersed overnight in a stirred mixture of nitric acid and
sulfuric acid (volume ratio of 3:1). After washing the resultant car-
bon in warm distilled water several times, the supports were placed
in an oven at 100 ◦C and left overnight.
The ruthenium catalysts supported on various carbons were
prepared using the incipient wetness impregnation method fol-
lowed by drying at 100 ◦C. The ruthenium loading was 5 wt% for
all the catalysts, and the dried powder was calcined at 500 ◦C for
4 h under flowing air. Prior to the reaction, all synthesized cata-
lysts were reduced under flowing hydrogen at 400 ◦C for 4 h. After
cooling, the catalysts were passivated in nitrogen for 30 min before
being exposed to air [14].
All the synthesized catalysts underwent detailed analyses. The
crystallinity of the synthesized samples was investigated using
X-ray diffraction analysis (Rigaku, D/max-2200). A transmission
electron microscope (JEOL/JEM2100F) was used to examine the sur-
face morphologies and ruthenium particle sizes. The surface areas
of the synthesized catalysts were determined using the BET method
(ASAP 2020), and both the pore volumes and average pore sizes
were calculated using the BJH model.
2.2. Hydrogenation of lactic acid
Liquid-phase catalytic hydrogenation of lactic acid to propylene
glycol was performed in a 150 mL batch reactor at various reaction
pressures (40–80 bar) and temperatures (110–170 ◦C).
Fig. 2. (a) X-ray diffraction pattern of 5 wt% Ru/CNTs, and (b) a TEM image of 5 wt%
A mixture of 1 M lactic acid solution (50 mL) and prepared cat-
alyst (0.5 g) was placed into the reactor and agitated at 700 rpm.
After purging the reactor with nitrogen for 5 min, the reactor was
heated under flowing hydrogen. When the temperature reached
the desired value, the hydrogen pressure was set to the specified
pressure (40–80 bar) and the reaction was left for 6 h. The prod-
ucts were analyzed using a gas chromatograph (HP 6890) equipped
with a flame-ionization detector. A free fatty-acid phase capillary
column was used for the separation of the reactants and products.
Ru/ketjen black.
the TEM images, are summarized in Table 1. Before impregnation
of ruthenium, all carbon supports were pretreated in a mixture of
nitric acid and sulfuric acid. Many types of functional groups, such
as hydroxyl groups, can form on the surface of carbon supports;
these functional groups play an important role in the fine dispersion
of ruthenium particles.
The BET surface areas, pore volumes, average pore sizes, and
average particle sizes of the various carbon-supported ruthenium
catalysts are listed in Table 1. Although the ruthenium particles
were observed to be finely dispersed on all the carbon supports,
the average particle size correlated remarkably with the surface
area of the carbon support. Ketjen-black – supported ruthenium
exhibited the highest surface area (1242.4 m2/g), highest pore vol-
ume (5.54 cm3/g), and smallest average ruthenium particle size
(1.6 nm) of the investigated catalysts. The surface area of ket-
jen black is reported to be 1423 m2/g, and this high surface area
was maintained after the impregnation of ruthenium. In con-
trast, graphite-supported ruthenium showed the lowest surface
area (17.2 m2/g) and the largest average ruthenium particle size
(10.4 nm), which is probably because of its high crystallinity and
dense structure. Overall, the supports with larger surface areas
exhibited smaller average ruthenium particle sizes. These results
can be explained by the presence of functional groups that form
on the surface of the carbon support during acid treatment. Larger
carbon surface areas result in the formation of more functional
3. Results and discussion
3.1. Characterization of supports and supported catalysts
X-ray diffraction analyses were performed on all samples; the
XRD pattern of the CNT-supported ruthenium catalyst after reduc-
tion under hydrogen flow is shown in Fig. 2(a). For all the catalysts,
two broad and strong peaks are evident at 2ꢀ = ∼25◦ and 44◦, which
are assigned to the (0 0 2) diffraction of graphite carbon and the
diffraction of the (1 0 1) planes of metallic ruthenium, respectively
[15]. This indicates that the ruthenium in each catalyst was reduced
to metallic ruthenium during thermal treatment under hydrogen
flow.
Fig. 2(b) shows a TEM image of Ru/ketjen black after reduction
under hydrogen flow. The Ru/ketjen black catalyst exhibited fairly
uniform dispersion of small ruthenium species (less than 3 nm).
Fine dispersion was also observed for the other carbon supports,
and the average particle sizes of each catalyst, as measured from

H. Jang et al. / Journal of Molecular Catalysis A: Chemical 380 (2013) 57–60
59
Table 1
BET surface areas, pore volumes, average pore sizes, and average Ru particle sizes of the synthesized catalysts.
Ru/Vulcan
Ru/KNB
Ru/CNTs
Ru/CNFs
Ru/graphite
Average Ru particle size (nm)*
Surface area (m2/g)
3.9
209.5
0.49
9.4
1.6
1242.4
5.54
3.5
225.5
1.26
8.3
118.0
0.18
5.9
10.4
17.2
0.05
11.6
Pore volume (cm3/g)
Average pore size (nm)
17.9
22.5
*
Average particle size was measured from TEM imasges.
Fig. 3. Lactic acid conversion over various carbon-supported ruthenium catalysts.
Reactant: 50 mL of 1 M lactic acid solution, reaction time = 2 h, temperature = 150 ◦C,
and hydrogen pressure = 50 bar.
Fig. 4. Catalytic activity of 5 wt% Ru/ketjen black with respect to time on-stream.
Reactant: 50 mL of 1 M lactic acid solution, temperature = 150 ◦C, and hydrogen pres-
sure = 50 bar.
groups causing the finer dispersion of ruthenium on the surface
[16].
3.2. Hydrogenation of lactic acid to propylene glycol
Catalytic hydrogenation of lactic acid was performed in a high-
pressure batch reactor. Fig. 3 shows a comparison of the lactic
acid conversions over ruthenium supported on various carbons
(i.e., VulcanXC-72, ketjen black, CNTs, CNFs, and graphite) after 2 h.
The reaction temperature was 150 ◦C, and the hydrogen pressure
was 50 bar for all catalysts except for graphite-supported ruthe-
nium. Ruthenium supported on ketjen black showed the highest
conversion (above 75%) of the catalysts tested. Interestingly, the
catalytic activity correlated remarkably with the BET surface area:
The higher the BET surface area, the higher the conversion of lac-
tic acid. However, the correlation between the BET surface area
and lactic acid conversion was not linear; for example, Ru/ketjen
black showed a 10% increase in lactic acid conversion over that
of Ru/Vulcan, but its BET surface is 6 times greater. These results
can be explained by the amount of ruthenium loaded. For all cata-
lysts, 5 wt% ruthenium was impregnated. If the loading amount is
increased, the difference in the catalytic activities of the two cat-
alysts would increase. The influence of the amount of ruthenium
loaded on ketjen black on lactic acid conversion is currently under
investigation.
The effect of the reaction time on catalytic activity (i.e., lactic
acid conversion, propylene glycol selectivity, and propylene glycol
yield) was investigated using the Ru/ketjen black catalyst (Fig. 4).
The reaction temperature was 150 ◦C, and the hydrogen pressure
was 50 bar. Lactic acid conversion increased with reaction time
because the reaction was carried out in a batch reactor. The pro-
pylene glycol selectivity decreased slightly with time on-stream,
implying that propylene glycol underwent further reactions with
additional time. The other main products were propionic acid and
2-hydroxy propionaldehyde. Neither propanol nor C1 or C2 com-
pounds were detected during the reaction. These results differ
Fig. 5. Catalytic activity of 5 wt% Ru/ketjen black with hydrogen pressure. Reactant:
50 mL of 1 M lactic acid solution, reaction time = 4 h, and temperature = 130 ◦C.
somewhat from those previously reported by Miller et al. [13]:
they investigated lactic acid conversion over various catalysts and
reported that, at temperatures below 170 ◦C, the only liquid product
was propylene glycol and the major byproducts were light hydro-
carbons (i.e., methane, ethane, and propane) [13]. Dumesic et al.
[12] reported the hydrogenation of lactic acid at 100 ◦C over a silica-
supported copper catalyst; the main byproducts of this reaction
were propionic acid and 2-hydroxy propionaldehyde and no light
C1 or C2 compounds were detected. The difference in product dis-
tribution can be explained by the different catalysts and reaction
conditions as lactic acid conversion is strongly dependent on the
catalyst and reaction conditions.
Fig. 5 shows the catalytic activity of 5 wt% Ru/ketjen black with
respect to reaction pressure after 4 h at 130 ◦C. Both the lactic
acid conversion and propylene glycol selectivity increased with

60
H. Jang et al. / Journal of Molecular Catalysis A: Chemical 380 (2013) 57–60
prepared using an incipient wetness impregnation method and
employed for the hydrogenation of lactic acid to propylene gly-
col in a high-pressure batch reactor. The ruthenium particle size
is strongly dependent on the BET surface area of the carbon sup-
port. Among the catalysts tested, ruthenium supported on ketjen
black exhibited the highest catalytic activity because of its fine
dispersion of ruthenium. The propylene glycol yield over ruthe-
nium on ketjen black increased with increasing hydrogen pressure;
the highest yield was achieved at 130 ◦C. It is concluded that
propylene glycol can be efficiently produced from a bio-based
chemical (i.e., lactic acid) using ruthenium supported on ketjen
black.
Acknowledgments
This research was performed as a cooperative project between
KRICT Own and was supported by the Korea Research Institute of
Chemical Technology (KRICT). This work was also supported by the
National Research Foundation of Korea (NRF), funded by the Korean
Government (MEST) (No. 3148213322).
Fig. 6. Catalytic activity of 5 wt% Ru/ketjen black with reaction temperature. Reac-
tion time = 2 h, Pressure = 50 bar.
reaction pressure. This is mainly because of the increased concen-
tration of hydrogen in aqueous solution with increasing reaction
pressure, which speeds up the reaction resulting in a higher pro-
pylene glycol yield. The optimal reaction pressure should primarily
be determined by equipment cost.
The influence of the reaction temperature on the catalytic
hydrogenation of lactic acid using 5 wt% Ru/ketjen black was inves-
tigated at 50 bar and is plotted in Fig. 6. In general, the conversion
of lactic acid increased with reaction temperature in the range of
110–170 ◦C. However, the propylene glycol selectivity reached a
maximum at 130 ◦C and then decreased with increasing tempera-
ture. The decreased selectivity at higher temperatures is attributed
to secondary reactions involving propylene glycol. Because the
hydrogenolysis of products competes with the hydrogenation of
lactic acid over supported ruthenium catalysts, the selection of
support and reaction conditions are very important.
References
[1] E.S. Lipinsky, R.G. Sinclair, Chem. Eng. Prog. 82 (1986) 26–32.
[2] J.H. Litchfield, Adv. Appl. Microbiol. 42 (1996) 45–95.
[3] P.R. Gruver, et al. US Patent 5. 5, 142,023 (1992).
[4] G.C. Gunter, D.J. Miller, J.E. Jackson, J. Catal. 148 (1994) 252–260.
[5] G.C. Gunter, R. Craciun, M.S. Tam, J.E. Jackson, D.J. Miller, Ind. Eng. Chem. Res.
34 (1995) 974–980.
[6] G.C. Gunter, R. Craciun, M.S. Tam, J.E. Jackson, D.J. Miller, J. Catal 164 (1996)
207–219.
[7] R. Sawicki, U.S. Patent 4, 729, 978 (1988).
[8] J.I. Choi, W.H. Hong, H.N. Chang, Int. J. Chem. Kinet. 28 (1996)
37–41.
[9] D. Stephane, W. Hughes, C. Fernand Thyrion, J. Chem. Technol. Biotechnol. 59
(1994) 149–156.
[10] M. Ai, K. Ohdan, Appl. Catal. A 165 (1997) 461–465.
[11] H.H. Szmant, Organic Building Blocks of the Chemical Industry, Academic Press,
New York, USA, 1989.
[12] R.D. Cortright, M. Sanchez-Castillo, J.A. Dumesic, Appl. Catal. A 39 (2002)
353–359.
[13] Z. Zhang, J.E. Jackson, D.J. Miller, Appl. Catal. A 219 (2001) 89–98.
[14] C. Liang, Z. Wei, Q. Xin, C. Li, Appl. Catal. A 208 (2001) 193–201.
[15] H. Gao, S. Liao, J. Zeng, Y. Xie, J. Power Sources 196 (2011) 54–61.
[16] E.B. Fox, H.R. Colon-Mercado, Int. J. Hydrogen Energy 35 (2010)
3280–3286.
4. Conclusions
Ruthenium catalysts supported on various carbons (i.e.,
VulcanXC-72, ketjen black, CNTs, CNFs, and graphite) were