P. Liu, et al.
MolecularCatalysis490(2020)110951
previously achieved by noble metal catalysts [10,11]. Poliakoff et al.
a two-reactor system, one reactor being 240 °C and another being
300 °C. Zhu et al. [11] obtained a 97 % yield of 2MTHF under atmo-
spheric pressure using copper phyllosilicate and Pd/SiO2 by two-stage
packing in one reactor. Despite many potential applications of 2MTHF,
researches focusing on the conversion of furfural into 2MTHF are still
insufficient, leaving many important aspects unrevealed. Also, the
performance and mechanism of non-precious metal catalysts in the
conversion of furfural into 2MTHF have not been examined in the lit-
erature yet. According to our previous work [14], Co-based catalysts
methylfuran, while Ni-based catalysts showed high reactivity for furan
ring hydrogenation. Therefore, it is anticipated that furfural could be
transformed to 2MTHF with high yield if Co and Ni catalysts were
utilized in sequence, achieving an efficient transformation of furfural
into 2MTHF without non-precious metals.
2.2. Sample characterization studies
Powder x-ray diffraction (XRD) patterns of all samples were ob-
tained on a Shimadazu XRD-7000 powder diffractometer after the re-
duction in pure H2 of 20 mL/min at 500 °C for 3 h. Temperature pro-
grammed reduction (TPR) of all samples were detected by a CAM200
quadrupole mass spectrometer (InProcess Instruments). Typically,
100 mg of samples were loaded in a quartz tube put in the furnace. All
catalysts were heated from room temperature to 800 °C with a heat rate
of 5 °C/min in a flow of 10 vol% H2/Ar (30 mL/min). Total acid sites
were measured by temperature-programmed desorption of ammonia
(NH3-TPD) using the mass spectrometer. Pyridine adsorbed Fourier-
transform infrared spectroscopy (FTIR) were used to characterized
Brønsted and Lewis acidity.
2.3. Hydrogenation/hydrodeoxygenation reactions
In this work, the general strategy of transforming furfural into
2MTHF is to load two different types of catalysts into a single reactor
with the uniform temperature. The first stage of catalyst would aim at
converting furfural into 2MF, while the second stage of catalyst aiming
at converting 2MF to 2MTHF. In this respect, the reaction of furfural
hydrodeoxygenation to produce 2MF, and the subsequent 2MF hydro-
genation over various catalysts will be studied separately. Cobalt and
nickel catalysts supported on various oxides were prepared and loaded
in the reactor to achieve high 2MTHF yield. To further explore the
reaction mechanism in the reactor, furfuryl alcohol (FOL), 2-methyl-
furan (2MF), and tetrahydrofurfuryl alcohol (THFOL) were used as re-
actants and tested. By analyzing the distribution of products, a feasible
strategy with a high yield of 2-methyltetrahydrofuran was proposed.
Hydrogenation/hydrodeoxygenation reactions were carried out in a
fixed-bed tubular reactor. Catalyst powders were loaded into the re-
actor tube between two layers of quartz wool and pre-reduced in pure
hydrogen at 500 °C for 3 h with a flow rate of 42 mL/min. For the two-
stage packing, a layer of quartz wool was placed between two stages.
Reactants including furfural, furfuryl alcohol, 2-methylfuran and tet-
rahydrofurfuryl alcohol were pumped into the reactor using a HPLC
pump at a flow rate of 0.03 mL/min. The liquid products of first 2 h
were discarded and then collected every 50 min. All samples were
analyzed by Agilent 7890B gas chromatography equipped with a HP-5
capillary column (30 m × 0.32 mm × 0.23 μm) and flame ionization
detector (FID). The compositions of liquid products were calibrated by
the internal standard method using decane as the inner standard. The
carbon balance of every sample was calculated. The conversion and
product selectivity were calculated using the following formulas:
2. Experimental section
Mole of raw material in product
Mole of raw material in the feed
2.1. Catalyst synthesis
Mole of specific in product
CeO2 was prepared through the precipitation of cerium (III) nitrate
hexahydrate (Innochem, 99.5 %). The precursor was dissolved in
deionized water to form a 0.1 M solution and then the ammonium hy-
droxide aqueous solution was added dropwise to the solution with
continuous stirring. The resulting precipitate was washed with deio-
nized water and then dried overnight in an oven at 100 °C. The pre-
cursor was calcined in air at 500 °C for 4 h with a ramp rate of 5 °C/min.
ZrO2 were synthesized in a similar manner, using zirconium (IV) oxy-
nitrate hydrate (Innochem, 99.9 %) as the precursor, and calcined
under the same temperature profile. γ-Al2O3 was obtained by calcina-
tion of pseudo-boehmite at 550 °C for 4 h with a ramp rate of 2 °C/min
in air. TiO2 was prepared through the precipitation of titanium tetra-
chloride (Innochem, 99 %). Deionized water was added into the tita-
nium tetrachloride solution to form a 0.1 M solution and then the 0.1 M
urea solution was added dropwise to the solution with continuous
stirring until the solution pH reached 6. The resulting precipitate was
aged at room temperature for 30 min and then washed with deionized
water and dried overnight in an oven at 120 °C. The powder was then
calcined at 400 °C for 4 h with a ramp rate of 2 °C/min to obtain m-
TiO2. Commercial MgO (Acros, 98 %) and SiO2 gel (Qingdao Haiyang,
99.9 %, 100–200 mesh) were used as the support.
Mole of all products
3. Results and discussion
3.1. Catalysts characterization
A series of Ni-based and Co-based catalysts supported on various
support were synthesized. X-ray diffraction results of all samples after
3 h of H2 reduction at 500 °C are shown in Fig. 1. For nickel-containing
catalysts, peaks at 44.5° and 51.6°, corresponding to the metallic Ni
(111) and (200) planes, respectively, were observed [16]. Cobalt sam-
ples exhibited characteristic diffraction peaks at 44.2° and 51.5°, cor-
responding to the metallic Co (111) and (200) crystal planes, respec-
tively [17]. Characteristic diffraction peaks of CeO2 at 33.1°, 47.6°, and
56.4° indicated the synthesized CeO2 displayed the expected cubic
structure [18]. For Co/ZrO2, peaks at 31.5°, 34.1°, 35.3°, 50.1°, and
60.1° suggested that the synthesized ZrO2 is monoclinic (m-ZrO2),
consistent with previous research [19]. For Ni/CeO2, Co/CeO2, and Co/
MgO, diffraction peaks of the support (CeO2 or MgO) are much stronger
than diffraction peaks of the cobalt and nickel peaks, so that diffraction
peaks of the metallic nickel and cobalt particles were almost invisible.
For Co/Al2O3, the peak at 46.4° was also detected, which could be at-
tributed to the irreducible cobalt aluminate [20]. Based on the XRD
duction at 500 °C was sufficient to yield metallic particles, so the 500 °C
reduction was conducted before the catalysts reactivity tests. Metal
dispersions of all catalysts were calculated using Scherrer’s equation
All catalysts were synthesized using incipient wetness impregnation
method. The certain amount of cobalt (II) nitrate hexahydrate
(Sinopharm, 99 %) was dissolved in deionized water to form a 0.86 M
Co(NO3)2 aqueous solution. The oxide support was then added. The
solution of Ni precursor was prepared similarly using nickel (II) nitrate
hexahydrate (Sinopharm, 99 %). After impregnation at room tem-
perature for 6 h, all catalysts were dried in an oven at 90 °C overnight.
Dried catalysts were calcined in air at 120 °C for 2 h and then calcined
at 400 °C for 3 h both with a ramp rate of 2 °C/min.
The reducibility of all catalysts was evaluated using temperature
2