Zn-HZSM-5 zeolite catalystwas synthesized in-situ by hydrothermal method^[22]. According to a certain ratio, TPAOH, sodium hydroxide and silica sol (mass fraction of 30%) mixed in deionized water to get Solution A. Sodium aluminate and different quality of zinc nitrate were added into deionized water to get solution B. Under stirring, solution B was slowly added dropwise to solution A. After the dropwise addition was completed, pH of the mixed solution was adjusted to 10.5 using sodium hydroxide solution of certain concentration or dilute sulfuric acid solution. The mixed solution was stirred at room temperature for 2 h, then moved into reactor. The prepared gel was crystallized statically at 413 K for 24 h. The solid product was filtered and washed with deionized water until the pH of the washing water reached 7-8 and then dried at 393 K for 12 h. The final product was calcinated at 823 K for 5 h. A certain amount of ammonium chloride solution with a mass fraction of 10% was arranged, the ammonium chloride solution mixed with the original powder of ZSM-5 zeolite with different mass contents of Zn in a mass ratio of 10 : 1 and stirred in water bath at 353 K for 5 h. The slurry was washed and filtered. Then the operation was repeated twice. Finally, the residue was dried and roasted to get HZSM-5 zeolite catalyst containing different content of skeleton Zn.Through XRF characterization for Zn-HZSM-5 zeolite catalyst, Zn content ranged from mass fraction of 0.90% to 3.69%. The synthesized Zn-HZSM-5 catalyst was compressed and sieved to evaluate activity of MTG. In order to compare with Zn-HZSM-5 zeolite catalyst, ZSM-5 zeolite was synthesized by hydrothermal method without adding zinc nitrate. Under the same ratio of silica to alumina, HZSM-5 zeolite catalyst was obtained.

Crystallization of zeolite catalysts were characterized by X-ray powder diffraction analysis (XRD) using Japan Rigaku company D/max-2500 instrument with Cu radiation(y = 0.154 056 nm). Working voltage was 40 kV and current was 40 mA. Scan speed was 5° · min^-1, scanning step was 0.02° , scanning angle was 5° -45° .

Specific surface area, pore volume and pore structure of zeolite catalyst were tested by nitrogen-physisorption characterization using the United States Quadrasorb SI-KR.

NH₃-TPD characterization was carried out to determin acid type and acidity of the zeolite catalyst using the United States Mieromeritics company Autochem II 2920.

The micro-fixed bed reactor was used to test catalyst activity. A certain amount of Zn-HZSM-5 zeolite catalystwas put into the reaction tube constant temperature section, the rest of the reaction tube filled with quartz sand. Under nitrogen atmosphere and 1 MPa, the reactor temperature was raised to a certain temperature. Then, under the conditions of activation, Zn-HZSM-5 zeolite catalyst was activated. After that, methanol was driven into the preheater through a parallel flow pump to be gasified and reacted on Zn-HZSM-5 zeolite catalyst. High-temperature hydrocarbon products were first condensed and then subjected to gas-liquid separation. Gas phase product and collected liquid product were then analyzed by Agilent GC 7890B gas chromatography with hydrogen flame ionization detector using PONA software. Yield and selectivity of each hydrocarbon was calculated based on carbon balance. Gasoline component contained all C₅-C₁₁ hydrocarbons.

Effect of internal diffusion is eliminated by changing particle size of Zn-HZSM-5 zeolite catalyst while keeping all other reaction conditions constant. Under the same reaction temperature, reaction pressure, space time and catalyst loading, only particle size of Zn-HZSM-5 zeolite catalyst changes. Methanol conversion is stable when particle size of zeolite catalyst decreases to a certain value. When particle diameter is less than or equal to this, influence of internal diffusion has been eliminated. Effect of external diffusion is eliminated by changing the amount of Zn-HZSM-5 zeolite catalyst in reactor tube and feed of raw materials to ensure that their ratio does not change. When the linear velocity of gas stream is greater than or equal to a certain value, conversion of raw material is stable, then influence of external diffusion is considered to be excluded.

It was found that catalyticperformance was the best when Zn content was 2.54%. Therefore, Zn-HZSM-5 zeolite (Zn mass content of 2.54%) catalyst was chosen for MTG intrinsic kinetics study. Under the reaction conditions of pressure 1 MPa, temperature (673-723) K, space time (0.11-0.41) h and airflow velocity 0.10m· s^-1, gas methanol concentration of 9.38%, the orthogonal experiment of methanol to gasoline was carried out.

ZSM-5 zeolite catalyst has characteristic diffraction peaks at diffraction angles of 8.0° , 8.9° , 23.10° , 23.30° , 23.90° and 24.4° (JCPDS00-037-0359). XRD patterns of samples are showed in Figure 1. We can see from Figure 1 that five zeolite catalysts synthesized have the same crystalline structure as ZSM-5 zeolite and no other miscellaneous crystals appear. According to literature^{[23, 24]}, the intensity of diffraction peak at diffraction angle of 22.0° -25.0° can represent the crystallinity of ZSM-5 zeolite catalyst.In figure 1, HZSM-5 zeolite catalyst has the highest crystallinity and Zn-HZSM-5 zeolite catalyst with Zn mass fraction of 3.69% has the least poor crystallinity. With the increase of Zn content in Zn-HZSM-5 zeolite catalysts, the crystallinity showed a decreasing trend due to Zn element into the zeolite framework during in-situ synthesis. The characteristic diffraction peaks of ZnO do not appear in figure 1, indicating that the distribution of Zn atoms in the zeolite catalyst framework synthesized in-situ is very good.

Figure 1

	Figure Option ViewDownloadNew Window
	Figure 1 XRD patterns of samples

Pore structurecharacterize by N₂ isothermal adsorption-desorption is showed in Table 1. As shown in Table 1, specific surface area, pore volume and average pore diameter of the HZSM-5 zeolite catalyst are larger than that of Zn-HZSM-5 zeolite catalyst. With the increase of Zn content, specific surface area, pore volume and average pore diameter of Zn-HZSM-5 zeolite catalysts decrease, probably because the atomic radius of Zn is larger than Si and Al. When Zn enters into the framework of zeolite through in-situ hydrothermal synthesis, the pore volume and pore size of zeolite decrease^[24].

Table 1

Table 1

Table 1 Pore structures of samples Samples SBET/m2· g-1 Vp/cm3· g-1 Average pore diameter/nm HZSM-5 zeolite 384 0.32 0.41 0.90wt%Zn-HZSM-5 zeolite 367 0.28 0.33 1.38wt%Zn-HZSM-5 zeolite 364 0.25 0.30 2.54wt%Zn-HZSM-5 zeolite 360 0.22 0.28 3.69wt%Zn-HZSM-5 zeolite 352 0.19 0.24

Table 1 Pore structures of samples

NH₃-TPD characterization results are showed in figure 2. Weak acid has little effect on MTG reaction while strong acid have great influence. According to literature^{[25, 26]}, the desorption peak appearing in (423-573) K attributes to weak acid, the desorption peak appearing in (573-723) K attributes to medium strong acid, the desorption peak appearing in (723-823) K attributes to strong acid. As shown in Figure 2, Zn-HZSM-5 zeolite catalyst with Zn mass content of 3.69% has the lowest strong acid content and the highest weak acid content. HZSM-5 zeolite has the lowest weak acid content and the highest strongest acid content.With the increase of Zn loading in Zn-HZSM-5, the desorption peaks at low-temperature and high-temperature of Zn-HZSM-5 zeolites shift to left, and the contents of weak acids in Zn-HZSM-5 zeolites increase. At the same time, decreasing of the strong acid content indicates that the skeleton of Zn atoms on the zeolite catalyst surface acid ratio of three kinds of acid play a regulatory role^[27].

Figure 2

	Figure Option ViewDownloadNew Window
	Figure 2 NH3-TPD patterns of samples

When linear velocity of airflow is greater than a certain value, methanol conversion remains unchanged, and the impact of external diffusion can be ruled out. Linear velocity is determined by amount of zeolite catalyst and feed of methanol. The external diffusion experiments were carried out under reaction pressure of 1 MPa, reaction temperature of 673 K and space time of 0.21 h. The concentration of methanol in the operating gas is 9.38%.

Effect oflinear velocity on methanol conversion is showed in Figure 3. As shown in Figure 3, when air stream linear velocity is gradually increased from the minimum value to 0.10 m· s^-1, methanol conversion gradually increased. However, when air flow linear velocity is more than 0.10 m· s^-1, methanol conversion keeps constant. Therefore, when linear velocity is greater than or equal to 0.10 m· s^-1, the influence of external diffusion has been eliminated.

Figure 3

	Figure Option ViewDownloadNew Window
	Figure 3 Effect of linear velocity on methanol conversion

When zeolite catalyst particle size is smaller than a certain value, methanol conversion is independent of catalyst particle size, indicating that effect of internal diffusion is excluded. Continue to increase zeolite catalyst diameter, methanol conversion gradually decreases. The internal diffusion experiments were carried out under reaction pressure of 1 MPa, reaction temperature of 673 K, space time of 0.21 h, Zn-HZSM-5 zeolite catalyst loading of 1 g, average Zn-HZSM-5 zeolite size of 0.21 mm, 0.27 mm, 0.30 mm, 0.50 mm, 0.71 mm, 1.1 mm.

Effect of particle size ofZn-HZSM-5 zeolit on methanol conversion is showed in Figure 4. As shown in Figure 4, when Zn-HZSM-5 zeolite catalyst particle size is smaller than 0.30 mm, methanol conversion keeps constant, indicating that internal diffusion has been ruled out. Continue to increase Zn-HZSM-5 zeolite catalyst average particle size to lager than 0.30 mm, methanol conversion decreases. Therefore, when Zn-HZSM-5 zeolite catalyst particle size is or smaller than 0.30 mm, the inner diffusion has been eliminated.

Figure 4

	Figure Option ViewDownloadNew Window
	Figure 4 Effect of particle size on methanol conversion

Among factors affectingMTG reaction over Zn-HZSM-5 zeolite catalysts, temperature and space time are the main influencing factors. Therefore, in order to scientifically and efficiently study the influence of the two factors, an orthogonal experiment was used to design the experiment scheme. The intrinsic kinetics of MTG reaction over Zn-HZSM-5 zeolite catalyst (mass fraction of Zn 2.54%) was investigated free of internal and external diffusion. The orthogonal experimental was designed as reaction pressure of 1 MPa, reaction temperature of 673 K, 698 K, 723 K, space time of 0.11 h, 0.13 h, 0.16 h, 0.21 h, 0.33 h, 0.41 h. Results are showed in Table 2.

Table 2

Table 2

Table 2 Kinetic data of methanol to gasoline Temperature/K Space time/h Component content/wt% Oxygenated compounds① Light olefin② Gasoline component③ 673 0.41 0 15.46 66.85 0.33 0.68 16.19 66.03 0.21 2.86 18.58 63.69 0.16 4.31 20.19 62.26 0.13 5.65 21.71 60.99 0.11 6.55 22.94 58.81 698 0.41 0 16.92 64.68 0.33 0.60 17.83 63.91 0.21 2.34 20.07 61.43 0.16 3.51 21.39 59.78 0.13 4.37 22.44 58.04 0.11 5.27 23.23 56.75 723 0.41 0 18.36 63.09 0.33 0.48 19.04 62.53 0.21 1.81 21.32 60.29 0.16 2.62 23.03 58.36 0.13 3.36 24.56 56.53 0.11 3.94 25.81 54.96 ①methanol and dimethyl ether; ②ethylene, propylene and butylenes; ③hydrocarbons between C5 and C11

Table 2 Kinetic data of methanol to gasoline

In 1979, based on the carbene mechanism, Chen and Reagan established a lumped kinetic model which was used by later researchers to study the intrinsic kinetics of MTG reaction. Therefore, the kinetics of MTG reaction over Zn-HZSM-5 zeolite catalyst (Zn mass fraction of 2.54%) was also studied by the kinetic model established by Chen and Reagan. The three-lumped kinetic model is as follows:

(1) A (oxygenate) reacts to form C (lighter olefins). The reaction in this step is first order reaction. According to the Chen and Reagan kinetic models, methanol and dimethyl ether are oxygenate components and ethylene, propylene and butene are light olefins.

A$\xrightarrow{{k_{i}}}$C(1)

(2) The reaction of A with C to form C is an autocatalytic reaction.

A+C$\xrightarrow{{k_{i}}}$C(Autocatalytic step)(2)

(3)C can generate G (gasoline component) through hydrogen transfer, oligomerization, aromatization, etc. The reaction is also a first order reaction. Gasoline component contains C₅-C₁₁ hydrocarbons.

C$\xrightarrow{{k_{i}}}$G(3)

k₁ is the reaction rate constant for oxygenates to light olefins; k₂ is the reaction rate constan for oxygenates and light olefins to light olefins; k₃ is the reaction rate constan for light olefins to gasoline.

The above three reaction rate constants and temperature are in line with the Arrhenius equation. The mass conservation equation of X_iis:

dXidW/F0=Ri(4)

Where X_i represents the mass content of corresponding component, Rirepresents the formation rate of the corresponding component, g· gcat-1· h^-1; W represents the zeolite catalyst mass, g; F₀ represents the mass flow of raw material, g· h^-1. The formation rate R_i of each component is expressed by the following equation:

RA=-k1XA-k2XAXC(5)

RC=k1XA-k3XC(6)

RG=k3XC(7)

The pre-exponential factork_0i and activation energy E_ai has given the initial value. According to Arrhenius equation, the reaction rate constant k_i of each component is:

ki=k0ie-EaiRT(i=1, 2, 3)(8)

In the formula, k_i represents the reaction rate constant of each component; k_0i represents the pre-exponential factor of each component; E_ai represents the apparent activation energy; T represents temperature.

Through the fourth-order Runge-Kutta method and the least-squares regressionof kinetic constants, OF value of the target residual function is minimized.

OF =∑i=1n∑j=1mXi, j-Xical, j2(9)

In the formula, Xi, jis the experimental value of component i under the reaction condition j, Xical, jis the lumped model calculated value.

The target residual function OF value indicates fitting result of experimental value to calculated value of the lumped model. The smaller OF value is, the better the fitting result is. According to literature, OF value is generally [0.270× 10^-2, 0.275× 10^-3], OF value for Zn-HZSM-5 zeolite catalyst kinematic target residual function is set to 0.412× 10^-3.

In the case ofOF value is 0.412× 10^-3, through fourth-order Runge-Kutta method and least squares regression of kinetic constants, simultaneous equations (5)-(7) and mass conservation equation (4) are solved. The initial values of pre-exponential factor k₀₁-k₀₃and activation energy E_a1- Ea3are given, and the search interval of exponential factor k0iand activation energy E_ai are set as [200, 1× 10¹³] and [150, 2× 10⁶]. By calculating the pre-exponential factor of each reaction component, the reaction activation energy and the correlation coefficient R² are shown in Table 3.

Table 3

Table 3

Table 3 Three lumped kinetic model constant regression results Component k0i Eai/J· mol-1 R2 Oxygenate 1.154× 1012 0.976× 105 0.993 Light olefin 0.687× 1012 1.052× 105 0.995 Gasoline 1.739× 107 0.847× 105 0.994

Table 3 Three lumped kinetic model constant regression results

Bring numerical value of k_0i and E_ai into ki expression (8), then:

Formation rate of light olefin from oxygenate k₁:

k1=1.154×1012exp(-97600/RT)(10)

Formation rate of light olefin from oxygenate and the light olefin k₂:

k2=0.687×1012exp(-105200/RT)(11)

Formation rate ofgasoline from lower olefin k₃:

k3=1.739×107exp(-84700/RT)(12)

k₁ is the largest, k₂ is slightly less, both of them are in the same magnitude. k₃ is the smallest. According to k_i, it can be concluded that it is very easy for oxygenate to light olefins on Zn-HZSM-5 zeolite. The formation rate of light olefin components is faster than that of gasoline components. The calculated activation energy E_ai of each reaction is in the same magnitude. The correlation coefficients R² of the three components are greater than 0.993, indicating that the experimental values of each component are in good agreement with the simulated values. The lumped kinetic model established by Chen and Reagan can well describe the reaction kinetics of MTG reaction over Zn-HZSM-5 zeolite catalysts.

Whetherthree-lumped kinetic model used is fit for reaction kinetics of MTG reaction over Zn-HZSM-5 zeolite catalyst is investigated. Comparison between experimental and simulated values based on the kinetic model established by Chen and Reagan is showed in Figure 5. As shown in Figure 5, at different temperatures and space time, experimental values of X_A (oxygenate), X_C (low carbon olefin) and X_G (gasoline component) are in good agreement with simulated results of the lumped model, which shows that the three-lumped model established by Chen and Reagan is very suitable for the reaction kinetics of MTG reaction on Zn-HZSM-5 zeolite.

The experimental values of X_A and X_C slowly decrease with the increase of space time at 673 K, 698 K and 723 K. The experimental value of X_G increases slowly with the increase of space time. Simulated and experimental values are completely consistent in changing trend, which shows that the simulation and the experimental data obtained by the three-lumped model established by Chen and Reagan are in good agreement.

Figure 5

	Figure Option ViewDownloadNew Window
	Figure 5 Comparison between calculated results and experimental data ■— XA (oxygenate) experimental value; □— XA (oxygenate) calculated value; ●— XC (light olefin) experimental value; ○— XC (light olefins) calculated value; ▲— XG (gasoline component) experimental value; ▽— XG (gasoline component) calculated value