W. Mao et al.
Applied Catalysis A, General 623 (2021) 118247
different metal molar ratios (NMA-2 (5:3:2), NMA-3 (4:4:2), NMA-4
(2:4:4), NMA-5 (8:0:2) and NMA-6 (8:2:0)) were prepared. As shown
mesoporous interparticle voids. In addition, as shown in Table 2 and
Fig. 2B, presence of oxalic acid ligand increases the molecular weight
and metal spacing. Accordingly, pore size of the calcined sample was
mainly distributed in the mesopore range, so that the sample has a larger
specific surface area, pore size and pore size volume.
◦
in Table 1, when the catalytic performance was compared at 140 C,
NMA-2 is the most effective and obtains the highest FOL yield (97.5 %)
and selectivity (99.0 %). This clearly shows that NMA-2 is the optimal
catalyst for the production of FOL from CTH. Therefore, five NMA cat-
alysts with different molar ratios were selected for characterization and
analysis.
In order to further confirm the surface morphology and particle size
distribution of the catalyst, the NMA-2 sample was observed by TEM
(Fig. 3). As shown in Fig. 3a, particle size distribution range of the NMA-
2 catalyst is 6ꢀ 20 nm, and the average particle size is 11.2 nm, which
was consistent with the calculation result of the Scherrer formula. As
shown in Fig. 3b, the NMA-2 catalyst has a large number of pores, which
corresponds to relatively large pore volume and higher specific surface
area, as measured by BET.
Fig. 1 represents XRD curves of the calcined samples. As shown in the
figure, XRD reflection of the calcined sample was set at 37.4, 44.7 and
63.3◦ attributed to nickel oxide [JCPDS.No. 65-5745] and Ni-MgO
[JCPDS.No.34-0410], which are overlapped each other due to its
highly dispersed nature. The reflections belonging to the Mg(Al)O
magnesite phase similar to those of Ni-containing catalysts were
observed, which is signed as NiO or Mg(Al)O [JCPDS.No.4-829] at 44.7
and 63.3◦. In series of NMA catalysts samples, the peak of NiO increases
slightly with the increase of the content of Ni, and decreases slightly
with the doping of other metals (Mg, Al), this clearly indicates that
dispersion of Ni decreases when the loading increases. The high
Fig. 4 represent the CO2-TPD and NH3-TPD curves of NMA catalysts
with different molar ratios. A large number of acidic and basic sites are
found in the NMA catalysts. Futhermore, Fig. 4A and Table 3 revealed
that the addition of Mg makes catalyst have a strong base site (439.3 ◦C),
as well as, Fig. 4B and Table 4 show that the addition of Al maked
catalyst have a large amount of acid (13.81 mmol/g) at the medium
strong acid site (272.6 ◦C). Thus, Al can greatly improve the acidity of
catalyst. Moreover, the mixing of Ni-Mg-Al makes the surface acidity
and basicity of catalyst tend to be stable. Consequently, acid-base ratio
of catalyst can be adjusted by changing the molar ratio of different
metals. In short, combined with BET analysis, NMA-2 has a larger sur-
face area, pore volume and pore size, so it can provide more acid-base
sites and active sites, thereby having a better acid-base ratio to
improve the activity of the catalyst. From the catalytic performance of
Table 1 and Fig. 8b, it could be seen that NMA-2 is the best catalyst with
an acid-base ratio (0.73:1). Hence, the experimental and characteriza-
tion data could show that the optimal acid-base ratio (0.73:1) and metal
site (Ni) can achieve multi-centers collaborative catalysis.
+
dispersion of NiO is mainly due to the uniform distribution of Ni2
under the action of the oxalate ligand. It is reported that the aluminum
phase can interact strongly with NiO [38]. In summary, the
well-dispersed NiO phase is formed and can maintain strong interaction
with other metals (Mg, Al). In addition to the corresponding features for
Ni-MgO, the XRD reflections at 37.4, 44.7 and 63.3◦ belong to the cor-
responding characteristics of NiO (1 1 1), (2 0 0) and (2 2 0)
[PDF#65-5745], respectively. According to the Sherrer formula, the
diffraction of NiO (2 0 0) is about 11.2 nm, and the average crystallite
size of catalyst particles was calculated to be roughly consistent with the
TEM results.
The structural and textural data of NMA catalysts with different
metal molar ratios are given in Table 2. Compared with tri-metal mixed
oxides, bimetallic mixed oxides have larger specific surface area,
average particle size and pore volume, which may be attributed to the
less agglomeration caused by the interaction between different metals.
Moreover, the surface area of the tri-metal mixed oxides was almost
unchanged. Among them, the most active NMA-2 has a larger surface
area of approximately 129.7 m2/g, a larger pore size (9.47 nm), and a
larger pore volume (0.28 cc/g), these could be beneficial to the pro-
ceeding of the hydrogen transfer reactions. As shown in Fig. 2A, in the
low pressure region (P/P0: 0~0.25), the interaction between the catalyst
and N2 reveal weak, and all NMA catalysts show type IV isotherms with
H3 hysteresis loop, which was ascribed to the presence of irregular
The surface chemistry of NMA catalysts with different molar ratios
was evaluated by X-ray photoelectron spectroscopy (XPS) (Table 5 and
Fig. 5). As far as the molar ratio of Ni/Mg/Al was concerned, there was
no big gap between the surface value and theoretical value (Table 5).
Except for the bimetallic catalysts (NMA-5, NMA-6), the surface molar
ratios of trimetallic catalysts (NMA-1, NMA-2, NMA-3) were closer to
expectation, which might be due to the interaction of metals that make
the dispersion of catalyst more uniform. The molar ratio (3:1) of bime-
tallic catalysts (NMA-5, NMA-6) surface was lower than expected (4:1),
which might be due to the higher Ni content compared with other
metals. It was also possible that compared with Al3+ and Mg2+, Ni2+
species are more likely to migrate to the surface. In addition, it could be
inferred that the high content of nickel ions (Ni2+) strengthened the
synergistic effect of metal bonds and acid-base sites.
In Fig. 5, the Ni 2p spectrum of prepared NMA catalysts showed two
strong peaks at 854.4 and 872.4 eV, corresponding to 2p3/2 and 2p1/2
spin-orbit lines, respectively. It could be observed that a satellite peak at
860.4 eV. These peaks refer to the ionic form of nickel (Ni2+), most likely
in the form of NiO and Ni(OH)2 phases [39,40]. In addition, the Ni 2p3/2
spectrum also shows a peak at 853.7 eV, repesenting Ni+, respectively.
The co-existence of Ni2+ and Ni+ favored the generation of oxygen va-
cancies. The conversion of Ni2+ to Ni+ is the main reason for the for-
mation of oxygen vacancies, as shown below: 2Ni2+ + O2ꢀ →2Ni+ +□
(□ represents oxygen vacancy) [41]. In addition, Mg 1 s and Al 2p
showed two strong peaks at 1301.3 and 70.4 eV, respectively. These
peaks refer to the ionic forms of magnesium (Mg2+) and aluminum
(Al3+), they are most likely in the form of MgO and Al2O3 phases.
Therefore, the results were consistent with XRD data.
As summarized in Table 5, the specific gravity of surface oxygen of
NMA catalysts with different molar ratios exceeds 50 %. This indicated
that oxygen plays an important role in the performance of the NMA
catalyst activity. The O 1s spectra of NMA catalysts with different molar
ratios all contain two peaks around 529.6 and 531.2 eV, which is
attributed to lattice oxygen (O lattice), and adsorption Oxygen (O defect
)
Fig. 1. XRD patterns of (a) NMA-1, (b) NMA-2, (c) NMA-3, (d) NMA-5 and (e)
related to oxygen vacancies, respectively [42]. Actively adsorbed
NMA-6.
3