1. Catalyst Activity and the Ways to Improve
A great deal of work has been carried out to improve the activity of M1 catalysts. Optimization of synthesis procedures, doping with other elements, the introduction of other phases, etc., have been employed.
1.1. Doping Method
The introduction of new elements as promoters in the catalyst system seems to be a useful way to modulate the catalytic performance. For MoVTeNb multi-component oxides, a lot of work has been investigated in elemental doping methods, including Cu, Fe, Sm, Sn, Bi, Co, Cr, Ce, Bi, In, Ni, Mn, Ga, K, Bi, and Ge, etc. [
112,
113,
114,
115,
116], and some of these catalysts were evaluated in the ODHE.
Ishchenko et al. [
114,
116] investigated the effect of Bi and K doped M1 MoVTeNbO
x catalysts on the catalytic performance of ODHE to ethylene. The introduced K and Bi are well dispersed on the M1 phase and become part of the M1 phase without the presence of any corresponding compound. These two additives show different effects on the M1 phase, including morphology, structure, valence state of surface V and Mo elements, and surface abundance of Te, giving opposite results on the catalyst activity. The additive of K into the M1 phase promotes the growth of the particles along the (001) direction, while the additive of Bi into the M1 phase does not affect the morphology of the M1 phase but modifies the structural features with an increased proportion of accessible (001) surfaces. The doped K decreases the active surface and abundance for both V
5+ active sites and Te on the surface, while Bi-doped samples show a different trend with a positive effect on the catalytic performance. Furthermore, the effect of Bi concentration in the MoVNbTeO
x M1 phase was investigated for selective ODHE. A strong correlation between the concentration of Bi and properties of the prepared catalysts was observed. An optimized concentration of Bi can have the modified MoVNbTeO
x catalysts with the best catalytic performance, ethylene selectivity, and stability. The doped Bi can constrain the Te segregated from the hexagonal channel to prevent the loss of Te so as to improve the stability of the M1 phase during ODHE.
Yun et al. [
115] prepared MoVTeNb(Ce)O catalysts using Ce as a promoter doped into M1 in the framework. The doping of Ce promotes the reducibility of the M1 phase and increases the amounts of active sites and lattice oxygen species available for ethane conversion. The improved properties significantly enhanced catalytic performance and maintained excellent ethylene selectivity. Lazareva et al. [
117] used Nd, Mn, Ga, and Ge as promoters to modify the catalytic properties of MoVNbTeO
x catalysts, which seems to be inserted into the orthorhombic structure. The physical-chemical properties of MoVNbTeO
x were affected, including the content of the M1 phase, surface compositions, and acidities. At small amounts of additives, the catalytic performance was enhanced during the ODHE process. The introduced elements were well dispersed on the M1 phase, and the valence of all elements and the composition of the surface M1 catalyst were comparable to those of the unmodified sample. Moreover, the ethylene over-oxidation at high ethane conversion was suppressed by the reduced amount of acid sites, which positively influenced the M1 catalysts with a high intrinsic ethylene selectivity.
1.2. Post Treatment Method
Chu et al. [
21] investigated the catalytic performance of the MoVTeNbO
x catalyst in ODHE, which was prepared from the same precursor slurry by hydrothermal synthesis with different post-treatments. The different purification processes will affect the tellurium contents and V
5+ concentration in the catalysts. They proposed that the catalytic performance is directly related to the amount of V
5+. While the formation of reduced Te blocks the active site, which is considered to be the main reason for catalyst deactivation. Therefore, lower Te content will facilitate the formation of stable catalysts and reduces the risk of Te aggregation. The hydrogen peroxide treatment will increase the V
5+ concentration and decrease the Te content, which can simultaneously improve the catalytic activity and stability of the catalysts in the ODHE process. However, under severe operating conditions, the phase-pure M1 catalyst with low Te content can also be significantly deactivated due to the conversion of the active phase to the inactive phases. Although the surface morphology of the oxalic acid-treated M1 phase was improved with a significant reduction in particle size and an increase in surface area, the loss of vanadium during the treatment reduced its catalytic activity.
The non-thermal plasma consists of energetic components that can modify the surface by a physical or chemical interaction of the active species in the gas phase with the solid surface. Chen et al. [
19] used oxygen plasma with strong oxidization properties to modify the surface of the M1 phase, which can change the valence state of surface vanadium sites to enhance the abundance of active sites. The oxygen plasma can efficiently increase the V
5+/V
4+ ratio on the catalyst’s surface and maintain the structure. The catalytic performance of M1 showed a good correlation with the concentrations of V
5+, which proves the feasibility of this method. Qian et al. [
30] also applied this method to modify the M1-CeO
2 nano-composite catalysts, showing a good catalytic performance during the ODHE process.
1.3. Introduction of Diluters/Promoters
Low activity and high catalyst cost are the main issues limiting the application of M1 catalysts for ODHE. Although a number of oxides have been applied to the ODHE process, none of them possess high ethylene selectivity activity except for MoVO
x, V
2O
5, and NiO-based catalysts [
23,
110,
118,
119]. However, some oxides exhibit unique redox properties, oxygen storage capacity, and high specific surface area and are often introduced as new phases to improve the catalytic and physicochemical properties of the original catalyst. As for the ODHE process over M1 MoVNbTeO
x catalysts, much work has been done to introduce oxides phase with poor activity in the ODHE into the MoV-based system to improve the M1 catalytic performance.
Nguyen et al. [
70] prepared silica diluted MoVTe(Sb)NbO catalysts by using a slurry method to optimize the catalytic performance during the ODHE process. They investigated the influence of the addition of silica in the slurry solution and the heat treatment in nitrogen after the dissolution of the M2 phase. However, the results showed opposite trends. The introduction of SiO
2 only improved the conversion of ethane without modifying the selectivity of ethylene. In contrast, the re-heat treatment reduced the catalytic performance and improved the selectivity of ethylene. These changes in catalytic properties can be attributed to the same factors, the degree of aggregation and sintering of the M1 phase and the distribution of small pores in the catalyst. Chu et al. [
20] used two methods (sol-gel method and physical mixing method) to combine the CeO
2 with the M1 phase and form an M1-CeO
2 nano-composite catalyst. The introduced CeO
2 oxide enhances the catalytic activity of M1 in the ODHE process. Compared with physically mixed M1-CeO
2 nanocomposites, M1-CeO
2 prepared by sol-gel method exhibited smaller CeO
2 particle size, well dispersed on the surface of the M1 phase, and showed good catalytic properties. In the ODHE process, the introduced CeO
2 phase enhances the redox properties of M1 phase, and at the same time, the introduced CeO
2 also enhances the valence state of V, which is related to the particle size of CeO
2. Dang et al. [
23] have systematically synthesized a series of M1-CeO
2 nano-composite catalysts by sol-gel method and determined the optimal CeO
2 loading of 30 wt.%. The results show that the introduced CeO
2 phase can promote the amounts of active sites, which is one of the major factors affecting the catalytic performance. In addition, the presence of Ce
4+ promotes re-oxidation of vanadium sites, which leads to an increase in the turn-over frequency (TOF). Chen et al. [
24] used MnO
x oxide as a promoter in combination with phase-pure M1 MoVNbTeO
x oxide for ODHE. The introduced manganese oxide was used as an oxygen promoter to allow M1 to be oxidized by gas-phase oxygen at a lower temperature during the preparation. In the ODHE process at 400 °C, the promoted M1 catalyst achieved more than 20% catalytic performance based on ethane conversion. Due to the high surface area and redox properties of anatase TiO
2, Dang et al. [
25,
28] used TiO
2 as a promoter to improve the catalytic performance of the M1 phase during the ODHE process. M1-TiO
2 nano-composites were prepared by the physical mixing method and the sol-gel method. The results showed that M1-TiO
2 prepare by sol-gel method shows a smaller TiO
2 particle size and well dispersed on the M1 surface, which presented an excellent activity for ODHE. In addition, M1-TiO
2 prepared by sol-gel method was optimized with an optimal TiO
2 content of 40 wt.%. Although the abundance of V
5+ sites content was enhanced to a certain extent, the total amount of active sites normalized to per mass was decreased after the introduction of TiO
2. The improvement in catalytic performance was mainly due to the introduction of TiO
2 to enhance the reduction/re-oxidation rate of lattice oxygen species in the catalyst.
Due to the disadvantages of Te-containing catalysts, which are toxic and harmful to the environment and tend to volatilize during the ODHE reaction, Zenkovets et al. [
120] used Ce-doped MoVSbNbO
x catalyst with SiO
2 as diluter to obtain the MoVSbNbCeO
x-SiO
2 catalyst by spray drying method in aqueous solution and calcined in He flow. The intergrowth between the M1 and M2 phases in the MoVSbNbCeOx-SiO
2 catalyst forms the interphase boundary of the highly active catalyst. The prepared catalysts can obtain up to 74% ethylene yield, which is the highest ethylene yield reported in the literature for Sb-containing catalysts. López-Medina et al. [
121] used alumina as support to deposit the active MoVNbTeO
x phase. Compared to their bulk form, the stable nanoscale active MoVNbTeO
x can expose a higher active surface and exhibit economic advantages. In addition, the alumina carrier offers better mechanical resistance and easier control of catalyst pellets formation.
1.4. Design and Optimization of Synthesis Procedure
Melzer et al. [
84] improved the hydrothermal synthesis method by using insoluble metal oxides as raw materials and organic additives, which can generate M1 phases with high surface area, showing higher activity compared to conventionally prepared catalysts. With the help of complexing agents that control the activity of ionic intermediates, the concentration of metal cations and polyoxometalate clusters is controlled at a concentration suitable for crystallization to avoid the formation of amorphous mixed oxides. The excellent catalytic performance can be attributed to the formation of M1 crystals with highly corrugated sidewalls, exposing a large number of active sites. The proposed synthesis method uses inexpensive and abundant metal oxide reactants and the simplicity of one-batch synthesis, enabling the synthesis can be scale-up directly.
López Nieto’s group [
122,
123] developed a new reflux synthesis method for the preparation of MoVTeNbO
x metal oxide catalysts. The synthesis procedure and parameters were optimized to obtain catalysts that can compete with conventional preparation methods. The ramp of the synthesis temperature is an important parameter that affects the vanadium content of the precipitate in favor of the formation of a pseudoamorphous Mo-V-Te-Nb oxometallate. The optimized synthesis parameters can produce a MoVTeNbO
x catalyst with a smaller size, which significantly enhances the catalytic behavior. Yu’s group [
124,
125,
126] investigated the effect of high pressure and temperature on the preparation of MoVNbTeO
x catalysts in a hydrothermal synthesis method. The catalytic performance of the oxidative conversion of propane/propylene to acrylic acid reaction was investigated. Li et al. [
126] synthesized the MoVTeNbO
x mixed metal oxides in sub/super-critical conditions by a stainless tube-reactor. The synthesized oxides with various phase compositions and morphologies can be directly used to convert propylene to acrylic acid without calcination. The physical-chemical properties are affected by the synthesis temperatures. High temperatures are beneficial for a mixed phase. Moreover, they also have prepared a series of MoVTeNbO
x-mixed metal oxides by hydrothermal synthesis under different pressure and temperatures [
124].
As a complex catalytic system, the introduction of high-throughput equipment can greatly accelerate the development and optimization process of the MoVNbTeO
x catalyst [
112,
127,
128,
129]. Mestl et al. [
112] used combinatorial and high-throughput methods to investigate the MoVTeNb based catalysts for the oxidation of propane to acrylic acid. The M1 catalysts containing additional promoters (Mn, Ni, W, In, Cu, Sb, Fe, Sm, Sn, Bi, Co, and Cr) were investigated. Moreover, the use of different ratios of citric acid as a structure-directing agent in the synthesis has also been investigated. During the optimization process, a five-generation catalyst was designed by an optimization platform consisting of artificial neural networks and a holographic optimization algorithm. Zhu et al. [
129] optimized the MoVNb mixed oxides catalysts for the ODHE process based on a commercial high-throughput reactor system. They screened about 75 catalysts with different Nb loading and Mo/ratio, and a composition range were established based on the high-throughput results.
2. Reactor Development and Industrial-Scale Tests for ODHE
The exothermic properties of the ODHE process tend to cause hot spots in the reactor, which would lead to temperature runaway in the reactor so as to negatively affect the product selectivity distribution and stable reactor operation. Over the years of research, some attempts have been carried out to apply ethane oxidative dehydrogenation on an industrial scale. Dalian Institute of Chemical Physics, Chinese Academy of Sciences and Wison Engineering have been devoted to the industrial development of ODHE to ethylene technology, and the related technology passed the pilot evaluation in 2021, which marks that the catalytic ODHE to ethylene is ready for industrial and commercial implementation [
130].
2.1. Lab-Scale Reactor Design
Nguyen et al. [
131] have synthesized a structured MoVTeNbO
x supported on pre-oxidized SiC foam by dip coating from slurry containing the precursors as catalysts for the ODHE and the ammoxidation of propane. However, the structured catalysts showed lower activity and ethylene selectivity compared to the powdered form due to the lower content of the M1 phase and the presence of other active but non-selective phases in the structured catalyst.
As a superior support for structured catalysts in industry implementations, cordierite has good chemical inertness, high thermal stability, and low thermal expansion coefficient. Chen et al. [
27] used cordierite monolith as a carrier to prepare an M1@Coridierite structured catalyst for the strong exothermic properties of the ODHE process. A CeO
2 layer was pre-coated on the cordierite to improve the catalytic performance of the structured catalyst. The M1@CeO
2@Monolith layered structure prepared by the two-step procedure shows excellent superior catalytic performance and stability for the ODHE reaction. Due to the constrained contact between the gas phase and the CeO
2 interlayer, the formation of CO
x is inhibited, thus maintaining the selectivity of ethylene. In addition, the structured M1@CeO
2@Monolith catalyst shows comparable ethylene selectivity to the powdered form diluted by 10 times silicon carbide, and the temperature runaway due to the highly exothermic reaction has been avoided. Yan et al. [
31] also used a foam SiC with a high thermal conductivity as a structuring support to coat a crystalline M1 powder catalyst onto this carrier and investigated methods to enhance the stability of the coating layer. The M1@foam SiC structured catalyst shows an excellent heat transfer and catalytic performance, which can eliminate the generation of hot spots and maintain the ethylene selectivity as well. Proper calcination treatment is beneficial to achieve good coating layer adhesion without loss of activity, while the addition of a binding agent, or a “stabilize the coating first, then activate the catalyst strategy, can achieve a robust coating and fair catalytic performance.
Chu et al. [
22] applied the microreactor to the strongly exothermic ethane oxidative dehydrogenation reaction. The hot effect during the oxidative dehydrogenation of ethane was investigated in a lab-scale fixed reactor (inner diameter of 8 mm) with different diluter-catalyst ratios. Hot spots in the lab-scale fixed-bed reactor can be eliminated when the ratio of SiC to catalyst is up to 10 times. At a SiC/catalyst ratio of 1/4, the hot spot temperature in the lab-scale fixed bed reactor is about 100 °C higher than that of the reactor environment and causes a significant decrease in ethylene selectivity and catalyst stability. In that work, they used PVA solution as a binder to prepare a stable and active phase-pure M1 catalyst layer on a metal-ceramic composite substrate by dip-coating method and inset into a micro-reactor to avoid the generation of hot spot. Compared with the microchannel reactor, the traditional fixed-bed reactor needs about 5 times its volume to achieve the same reactor productivity. Lin et al. [
132] also utilized the merits of the large heat transfer surface area and short heat transfer distance of the micro-reactor to achieve better-controlled operation conditions, especially the reaction temperature during propane ammoxidation over MoVNbTeO
x catalyst. Compared with the temperature gradient in a conventional fixed bed tubular reactor (43.2 °C), the gradient in a microchannel reactor can be controlled at less than 0.5 °C. The strongly exothermic propane ammoxidation reaction can be easily and precisely operated in a microchannel reactor under much harsher reaction conditions to achieve higher productivity and selectivity.
2.2. Industrial-Scale Reactor Design
Che-Galicia et al. [
133] regressed a kinetic model in lab-scale experiments and then coupled it with an industrial-scale reactor transport parameters obtained from other independent experiments. Industrial-scale wall-cooled packed bed reactor models illustrate the importance of transfer parameters for the simulation of this reactor. The coolant will affect the temperature distributions and formation of hot spots in the large-scale packed reactors, which will cause the damage of M1 MoVNbTeO
x catalysts.
Chen et al. [
134] have developed an experimentally based kinetic model over M1 catalyst and applied it for comparison between autothermal and multi-tube reactors and demonstrated a feasible autothermal reactor design for ODHE. The results show that the autothermal reactor configuration is more favorable for the high exothermic ODHE process. The highly exothermic properties of the ODHE process makes it hard to be operated steadily in a multi-tubular reactor. Although the diluents for the reactants and catalyst bed will help to reduce the exothermic intensity, this is achieved at the cost of reduced capacity. The designed autothermal reactor with cold feed allows for near-complete conversion of oxygen in the outlet and reduces the size of the reactor by a factor of 2–10. Fazlinezhad et al. [
135] investigated the effect of removing water treatment on the ODHE process over MoVTeNbO
x catalyst in the fixed-bed reactor. After water removal, the reactor hotspot and temperature were reduced, with the hotspot dropping from 500 °C to 460 °C, making the operation more controllable. Besides, the ethylene was enhanced after the removal of water. Moreover, a 20-bed mode membrane-like reactor was also investigated to remove water more sufficiently, which resulted in more than 94% ethylene selectivity. Therefore, a membrane reactor with intermediate water removal could be an appropriate option for the ODHE process.
Baroi et al. [
136] systematically evaluated the impact of feedstock composition on the operating cost, profitability, and process safety of ODHE based on M1 catalysts. The whole process was evaluated under different operating conditions, where no kinetic considerations were involved, catalytic performance was obtained from references, and the reactor was assumed as a black box. Based on the simulations, they proposed a staged oxygen feed process to minimize nitrogen and vapor in the gas stream, and the use of membrane separators made the process more profitable and safer. Moreover, they suggested that the use of CO
2 as an oxidant is expected to be enhanced.
2.3. Selective Oxidation of Ethane over M1 Catalyst via a CL-ODH Process
Considering the separation and safety features, the conversion of ethane to ethylene by a chemical looping process has attracted the attention of researchers in the last few years [
137,
138,
139,
140], which also can be combined with a fluidized bed reactor [
141]. Mishanin et al. [
142] used an MoVNbTeO
x catalyst to convert ethane to ethylene in a cyclic mode, which can also be considered as a chemical cyclic conversion process. The feed is alternated between ethane and air, and the ethane is converted by the lattice oxygen species of the M1 phase. The lattice oxygen species available for ethane conversion is increased with the temperature. Luongo et al. [
143] developed a cyclic chemical-looping-based ODHE process by combining a NaNO
3-modified perovskite Sr
0.8Ca
0.2FeO
3−δ as the oxygen carrier with the highly selective M1 catalysts to convert ethane. The Sr
0.8Ca
0.2FeO
3−δ has a low ethylene selectivity in the process of ethane feed only; however, it can release oxygen in the reduction process. Therefore, they used Sr
0.8Ca
0.2FeO
3−δ as an oxygen carrier and utilized NaNO
3 to suppress the CO
x formation during the CL-ODH process. The modified oxygen carrier (which can release oxygen during CL-ODH) is then combined with an excellent M1 catalyst to achieve highly selective ethane conversion under uncoupled oxygen conditions. Compared with the conventional ODHE over M1 phase and CL-ODH process with only perovskite Sr
0.8Ca
0.2FeO
3−δ oxygen carrier, the combination of highly selective M1 phase with high oxygen carrier ability perovskite Sr
0.8Ca
0.2FeO
3−δ allows this process being operated under the oxygen uncoupled condition and transfer the problem of ethylene selectivity for oxygen storage materials under CL-ODHE conditions into the inhibition of CO
x generation. Therefore, it brings new insights into the development of CL-ODH catalysts, i.e., catalysts for ethane oxidative dehydrogenation reactions and oxygen carriers can be optimized independently, thus reducing the difficulty of catalyst development.
This entry is adapted from the peer-reviewed paper 10.3390/catal13010204