GModel
CATTOD-8823; No. of Pages7
ARTICLE IN PRESS
F. Auneau et al. / Catalysis Today xxx (2014) xxx–xxx
3
2.2. Trickle-bed reactor
and molar selectivities to the different products. The conversion
is defined as the percentage ratio of the amount of xylitol con-
verted to the amount of xylitol in the feed stream. The description
of the results by weight concentrations has the advantage to show
directly the production of products compared to the concentration
of converted xylitol, without taking into account the differences
in molecular weights of the different products. The selectivity to a
particular product is defined as the percentage ratio of the molar
concentration in this product to the total molar xylitol concentra-
tion converted, without accounting the number of carbons in the
molecules. Therein, if one considers that one mole of xylitol will
ideally yield one mol of EG and one mol of 1,2-PDO, the selectivity
to each product should be then 100%. It gives an indication of the
efficiency of the reaction in producing one desired product.
The hydrogenolysis reactions of aqueous solution of xylitol
(200 g L−1) were performed in a laboratory scale, high-pressure
fixed-bed reactor run in the trickle-bed mode with co-current
downflow of liquid feed and hydrogen. The system consisted of
a tube made of Hastelloy C (1 cm inner diameter and 15 cm in
length) that was heated by a tubular furnace. For each series of
experiments, the tube was packed with the catalyst located in the
isothermal portion of the reactor tube between two layers of inert
packing material (1 mm spherical glass or low surface area titania
particles) and two sintered stainless-steel filters at the inlet and
exit of the reactor. The temperature of the catalytic bed was moni-
tored and controlled via a thermocouple placed into a thermowell
inserted in the reactor tube. The liquid feed, constituted of a xyli-
tol aqueous solution with NaOH as basic promoter, was introduced
from a feed glass tank into the system by an HPLC pump through
the filter at the top of the column. The pressure and the flow rate
of the gas were controlled with a back-pressure controller. The
system was also equipped with a heat exchanger, a high-pressure
gas–liquid separator from which the liquid effluent was continu-
ously drained and collected in a flask with a liquid level control
device and the gas stream discharged.
Initially, after purging with Ar, the catalyst was re-activated
overnight under hydrogen gas flow at 200 ◦C under 40 bar. The tem-
perature, the hydrogen pressure and flow, and a water flow, were
set at the desired values. Upon entering the reactor, the aqueous
solution and hydrogen were mixed and heated in the inert packing
before contacting the catalyst. Once the system was stabilized at
these reaction conditions, the water flow was replaced by a flow of
the xylitol alkaline aqueous solution (10.2 mL h−1) in the reservoir.
The gas flow stream was fixed at 20 NL h−1, giving in these experi-
ments a high hydrogen to xylitol feed molar ratio > 60. Lower ratios
of 5–7 were used on the pilot plant, which did not change signifi-
cantly the selectivity. The solution was cooled in a condenser and
the gas was vented. The liquid sample at the exit was collected and
periodically withdrawn for off-line analysis.
3. Results and discussion
Hydrogenolysis reactions of alkaline solutions of xylitol
(20 wt.%, 200 g L−1) were conducted to investigate how the pres-
sure in the range 40–80 bar in different concentrations of NaOH,
influence xylitol conversion and product distribution. Two masses
of catalysts were packed in the reactor in two separate series of
peratures were found appropriate and are close to those used
recently for hydrogenolysis of sorbitol or xylitol over Ni-Re/C
[29,31] or carbon-supported Ru catalysts [33,34,37] or Ni and Pt
[35]. Sun et al. [37] observed that the activity and selectivity of
Ru/C in the presence of Ca(OH)2 as solid base depended largely on
the H2 pressure, temperature and amount of base in batch experi-
ments and the selectivities were governed by the relative rates of
the different competitive reactions.
3.1. Reaction at low conversion of xylitol
3.1.1. Effect of reaction conditions
2.3. Analytical procedures
Hydrogenolysis of a xylitol solution (20 wt.%, 200 g L−1) was per-
formed over an initial loading of a relatively low mass of 2 wt.%Ru/C
catalyst (1.15 g, 7.5 cm height in the reactor in the isothermal part
of the tube) between two layers of inert material. With a liq-
uid flow rate of 10.2 mL h−1, this corresponds to a space time of
1.13 gRu h gxylitol−1. After purging with nitrogen, the reactor was
heated to the desired temperature of 200 ◦C under H2 flow rate of
20 NL h−1. The reaction was performed at two temperatures (190 ◦C
and 200 ◦C) at different pressures (40–80 bar) and different base
concentrations (0.13–0.26 M NaOH, corresponding to molar ratios
of NaOH/xylitol of 0.1–0.2). Indeed, it has been shown that the reac-
tion rate depends on the amount of base and that the selectivity to
glycols can markedly be increased without decreasing polyol con-
version when an inorganic hydroxide base is added [15,21,24,33].
The alkaline nature of the mixture produced from the hydrogenoly-
pH, the volatility of all the organic acids (acetic, formic, etc.) by-
products is lowered, thus it is easier to separate them from the
light glycols (EG, PDO). Each condition was evaluated for at least
100 h to check the stability of the catalyst.
The main products of the hydrogenolysis reaction of xylitol are
EG, 1,2-PDO and GLY. Other hydrogenolysis products which may
be formed are tetritols (threitol and erythritol), butanediols, light
alcohols and some carboxylic acids. Reaction samples (dilution
by a factor of 20-25 before HPLC analysis) were analyzed using
a Shimadzu LC 20A HPLC connected to a RI detector and a UV
detector at 210 nm, using a Transgenomic Coregel 107H column
(L = 300 mm, Ø = 7.8 mm) heated at 65 ◦C. Elution was performed
by 0.6 mL min−1 0.01 N H2SO4 mobile phase. With such condi-
tions, in addition of xylitol and xylose, the C4-polyols (butanediol
isomers – BDOs, threitol, erythritol), C3-polyols (glycerol – GLY,
1,2-propanediol – 1,2-PDO), C2 polyol (ethyleneglycol – EG), C1–C3
alcohols (methanol, ethanol, 1- and 2-propanol), C1–C3-carboxylic
acids (formic, acetic, lactic, propionic acids) could be analyzed.
Calibration curves were established for the quantification of the
products.
The Total Organic Carbon (TOC) of the solutions was mea-
sured, using a TOC-VCSH analyzer, to check the material balance
and the possible formation of gaseous products (methane, ethane)
by cleavage in the experimental conditions used. A dilution fac-
tor of 400–500 of the samples was applied before the TOC analysis
to be within the concentrations of the calibration curves, so that
the results are obtained with 2 g L−1. The pH of the solution was
measured ex situ using a Meterlab PHM 240 pH-meter.
Fig. 1 shows the data while applying a sequence of different
operating conditions. Due to the complex reaction network with
several parallel and consecutive reactions, a wide product distribu-
tion was observed. It shows the concentration of xylitol converted
and the bar cumulative weight concentrations of main polyols pro-
duced in the samples periodically taken at the outlet of the reactor:
EG, 1,2-PDO, GLY, and butanediols (2,3-BDO + 1,3-BDO + 1,4-BDO)
and formic acid FA. Threitol was also detected in small amounts but
The results are expressed in terms of weight concentrations of
xylitol converted and in the formed products, xylitol conversions,
Please cite this article in press as: F. Auneau, et al., Exploring the reaction conditions for Ru/C catalyzed selective hydrogenolysis of xylitol alkaline
Auneau, Florian
