J. Cent. South Univ. Technol. (2008) 15: 127-131 

DOI: 10.1007/s11771-008-0025-8 

Kinetic model on coke oven gas with steam reforming

ZHANG Jia-yuan(张家元), ZHOU Jie-min(周孑民), YAN Hong-jie(闫红杰)

(School of Energy Science and Engineering, Central South University, Changsha 410083, China)

Abstract: The effects of factors such as the molar ratio of H₂O to CH₄ (n(H₂O)/n(CH₄)), methane conversion temperature and time on methane conversion rate were investigated to build kinetic model for reforming of coke-oven gas with steam. The results of experiments show that the optimal conditions for methane conversion are that the molar ratio of H₂O to CH₄ varies from 1.1 to 1.3 and the conversion temperature varies from 1 223 to 1 273 K. The methane conversion rate is more than 95% when the molar ratio of H₂O to CH₄ is 1.2, the conversion temperature is above 1 223 K and the conversion time is longer than 0.75 s. Kinetic model of methane conversion was proposed. All results demonstrate that the calculated values by the kinetic model accord with the experimental data well, and the error is less than 1.5%.

Key words: coke oven gas; steam reforming; kinetic model; conversion rate

1 Introduction

The coke oven gas is a by-product from coking plants during the production of blast furnace coke. Besides CO and H₂, these product gases always contain large amount of hydrocarbons like methane, benzene and naphthalene. The actuality of application technique about the coke oven gas behind productivity leads to that a large amount of coke oven gas cannot be utilized sufficiently, resulting in environmental pollution and resource wasting. With the development of the coke industry, the issue is acute increasingly^[1-2]. Reforming coke oven gas into synthesis gas, which can be used as metallurgical reduction gas or raw material for the production of chemicals, such as methanol or ammonia etc, is a promising technology^[3-5]. With regard to the use of the product gases, hydrocarbons should be converted^[1,6]. Steam-reforming is effective to reform coke-oven gas^[7], but the kinetic behavior needs further study.

In this paper, to develop a process of reforming coke-oven gas, the conversion of methane in the presence of H₂O was studied, and the kinetic model of CH₄ conversion was proposed.

2 Experimental

The scheme of experimental system is shown in Fig.1. The experiments were made in a laboratory-scaled apparatus at 1.4×10⁴ Pa and 773-1 323 K. The compositions of the coke oven gas for the experiments are listed in Table 1.

Fig.1 Scheme of experimental system

Table l Compositions of raw coke oven gas for experiments (molar fraction, %)

A tubular electric flow reactor made of quartz was used to heat and maintain the furnace at the required temperature. The alumina balls were put at the bottom of the furnace to keep the evenly-distributed flow field. Temperature of the furnace can be controlled in the range of 773-1 323 K. Real temperatures were measured by platinum-rhodium thermocouple that can be shifted in a small quartz tube with 5 mm in diameter. During the experiments, the coke oven gas from storage flowed into the reactor at the fixed speed of 18 m³/h. Water was injected into an electrothermal evaporator by a high precision syringe pump to from steam of about 573 K, and then the steam was blown into the furnace with coke oven gas in proportion. The desired molar ratio was adjusted by controlling the temperatures in the saturator^[8-9]. The flow rate and composition of the coke oven gas were kept constant during the experiments, the volume of the reactor(namely conversion time) was changed through adjusting the depth of the outlet. The composition of product was analyzed and recorded by gas analysis recorder.

3 Results and discussion

During the experiments, the factors including the molar ratio of H₂O to CH₄, conversion temperature and time in the reactor on methane conversion rate were investigated. The conversion temperature varied from 773 to 1 323 K, and the molar ratio of H₂O to CH₄ varies in the range of 0.8-1.4. Conversion time was changed by adjusting the depth of outlet. The results of experiments are shown in Figs.2-4.

Fig.2 Effect of temperature and conversion time on conversion rate of CH₄: 1—823 K; 2—923 K; 3—1 023 K; 4—1 123 K;   5—1 173 K; 6—1 223 K; 7—1 273 K

Fig.3 Conversion rate of CH₄ at different temperatures for 0.6 s

Fig.4 Effect of molar ratio of H₂O to CH₄ ratio on conversion rate and molar fraction of H₂+CO at 1 223 K for 0.6 s

Fig.2 shows that the methane conversion increases with decreasing temperature gradually. It can be seen from Fig.3 when conversion time is longer than 0.6 s and the temperature is above 1 223 K, the conversion reaction approaches equilibrium state and the methane conversion rate approaches the maximum value.

Fig.4 shows that the methane conversion rate increases with decreasing the molar ratio of H₂O to CH₄ in a rang of 0.8-1.4, the methane conversion rate is above 87%. When the molar ratio of H₂O to CH₄ is in a range of 1.4-1.6, the methane conversion ratio decreases as increasing the molar ratio of H₂O to CH₄. It is known that the gas conversion rate depends on its conversion time, viz., the contact time among reactant molecules. When the flow rate of coke oven gas is fixed, the velocity of the admixture is faster as the molar ratio of H₂O to CH₄ increases, so the conversion time of the admixture in reactor is reduced, resulting in less methane conversion.

To sum up, the optimal conditions for methane conversion are that the molar ratio of H₂O to CH₄ varies in a range of 1.1-1.3 and the conversion temperature varies in a range of 1 223-1 273 K. The methane conversion rate is more than 95% when the molar ratio of H₂O to CH₄ ratio is 1.2, the conversion temperature is above 1 223 K and the conversion time is longer than 0.75 s.

4 Model of kinetic equation

4.1 Integral conversion equation

The experimental reactor can approximately considered as a ideal piston integrate reactor. Reactant gets through the reactor at constant volumetic flow rate u and is converted continuously. The composition of outlet gas is the integration result of the whole reactor^[10]. The scheme of integral reactor is shown in Fig.5.

Fig.5 Scheme of integrate reactor

The original concentration of reactant A is assumed as c_A0, and A flows into reactor at the flow rate of u, the conversion rate of A is x_A before entering micro bulk dV, and x_A+dx_A when departing. When the conversion process reaches steady state, according to the principle of mass conservation there is a relationship as follows^[11]:

                             (1)

and

                          (2)

Then the dynamical integral equation of the conversion is:

                               (3)

where  V is the volume of reactor, m³; u is the flow rate of reactant, m³/s; c_A0 is the original molar concentration of A, mol/m³; x_A is the conversion rate of A, %; r_A is the reaction rate of A, mol/(m³?s); t is the conversion time, s.

4.2 Reaction rate equation

In the reactor, if ignoring the soot and nitrogen, there exist 5 components: CH₄, CO₂, CO, H₂O and H₂, and three elements (C, H and O). So the system has two independent reactions as follows^[12-13]:

CH₄+H₂O=CO+3H₂                          (4a)

CH₄+CO₂=2CO+2H₂                         (4b)

The reaction rate of methane are calculated by the following equations:

The total reaction rate of methane is:

          (5)

where  r is the total reaction rate of CH₄, mol/(m³?s); r₁ and r₂ are the reaction rates of 4(a) and 4(b), respectively, mol/(m^3?s); K₁ and K₂ are the constant of reaction rate of (4a) and(4b), respectively; c_t(CH₄), c_t(CO₂) and c_t(H₂O) are the instantaneous concentration of CH₄, CO₂ and H₂O at time t, respectively, mol/m³; α, β, θ and γ are conversion indexes.

The original concentrations of CH₄, CO₂ and H₂O are set as c₀(CH₄), c₀(CO₂), c₀(H₂O).

If the conversion amount of CH₄ at a certain time t is Y, then the concentration of CH₄, H₂O and CO₂ are respectively:

                     (6)

where  m and n are the coefficients of H₂O and CO₂ respectively, mY+nY=Y, namely, m+n=1, m and n can be obtained by experiment.

Substituting c_t(CH₄), c_t(CO₂) and c_t(H₂O) into Eqn.(5), then

     (7)

4.3 Constant (K) of reaction rate

The assumed values of α, β, θ and γ are given in Table 2^[14-15], then the reasonable values are confirmed by the experimental data.

Table 2 Supposed values of α, β, θ and γ

First, the values of α, β, θ and γ are supposed  as (1,1,1,1 ), then

If   and x=Y/c₀(CH₄), then    

      (8)

Substituting Eqn.(8) into Eqn.(3), then

         (9)

Integrating Eqn.(9), and supposing

  (10)

then

                     (11)

In the present work, the conversion rate x of CH₄ was measured and calculated at different conversion times and conversion temperatures. The relationship between t and ln(1-x) is plotted in Fig.6, indicating that the relationship between t and ln (1-x) is approximately linear. By regressing the line equation, the intercept and slope of the line are ln(1-ax) and -b. Then we can get the values of a and b, and furthermore get the corresponding values of K₁ and K₂. The results are shown in Table 3.

Fig.6 Relationship between t and ln(1-x)

Table 3 Values of a, b and K₁, K₂ to No.1 in Table 2

The values of K₁, K₂ to No.2 and No.3 can also be calculated by the same process when α, β, θ and γ are 1, 0, 1, 1 and 1, 1, 1, 0. The results are shown in Table 4.

Table 4 Values of K₁ and K₂ to No.2 and No.3

4.4 Conversion indexes α, β, θ and γ

According to Arrhenius theory, the relationship between K and T are^[16]

K=Ae^-E/(RT)                                 (12)

and

ln K=ln A-E/(RT)                            (13)

where  A is the apparent frequency factor; E is the apparent activation energy, J/mol, is a constant; R is the ideal gas constant.

The relationship between ln K and 1/T should be linear, by which the reasonability of the above assumption could be proved. The relationship curves between ln K and 1/T is plotted based on data in Table 3 and Table 4, as shown in Fig.7.

The results show that the relationship of lnK₁₁ and lnK₁₂ with 1/T are reasonable, but lnK₂₁ and lnK₂₂ with 1/T are not resonable, so the values of α, β, θ and γ should be 1, 1, 1, 1.

Fig.7 Relationship between lnK and 1/T

4.5 Kinetic equation

The kinetic parameters can be obtained as follows:

the apparent frequency factor: A₁=4.56×10⁹, A₂=   8.06×10⁸; the apparent activation energy: E₁=21 373.4 J/mol; E₂=20 843.7 J/mol.

The kinetic equation of conversion rate for CH₄ is as follows:

    (14)

5 Model of methane conversion rate

The model of methane conversion rate can be obtained as

                                 (15)

For example, when the reaction temperature is    1 223 K, the model of methane conversion rate is

     (16)

To verify the accuracy of the model, the comparison between the experimental data and calculated value is shown in Fig.8. The results demonstrate that the calculated value by the models is in good accordance with the experimental value, and the error is less than 1.5%.

Fig.8 Comparison of calculated and experimental data

6 Conclusions

1) The experiments of reforming coke oven gas into synthesis gas with steam are performed, the effects of the factors such as the molar ratio of H₂O and CH₄, methane conversion temperature and time in the reactor on methane conversion rate are investigated.

2) The kinetic data of the varied reaction conditions and the optimal dynamic conditions for methane conversion are obtained, and the dynamic model of methane conversion is proposed and verified. All results demonstrate that the calculated values by the dynamic models accord with the experiment ones.

References

[1] WANG Zhao-ping. Application of the heat-exchanged reformer in the methanol production from coke oven gas[J]. Coal Chemical Industry, 2005, 119(4): 62-63. (in Chinese)

[2] ZHENG Wen-hua, ZHANG Xing-zhu. Present situation and prospect on application of coke oven gas[J]. Fuel and Chemical Processes, 2004, 4(35): 1-3. (in Chinese)

[3] YE D P, AGNEW J B, ZHANG D K. Gasification of a south Australian low-rank coal with carbon dioxide and steam: Kinetics and reactivity studies[J]. Fuel, 1998, 77: 1209-1219.

[4] MACKENZIE R C. Differential thermal analysis: Fundamental aspect[M]. London: Academic Press, 1970.

[5] STARINK M J. The determination of activation energy from linear heating rate experiments: A comparison of the accuracy of iso-conversion methods[J]. Thermochimica Acta, 2003, 404: 163-176.

[6] EDWIGES S E, GANG Y, TIM M. A simple kinetic analysis to determine the intrinsic reactivity of coal chars[J]. Fuel, 2005, 84: 1920-1925.

[7] ZHOU Zhi-jie. Char gasification kinetics using non-isothermal TGA[J]. Journal of China Coal Society, 2006, 31: 219-222.

[8] YOUNG S, HYUN S R, YOUNG S B. A highly active catalyst Ni/CeZrO₂-A1₂O₃, for on-site H₂ generation by steam methane reforming: Pretreatment effect[J]. International Journal of Hydrogen Energy, 2003, 28: 1387-1392.

[9] YASUYUKI M, TOSHIE N. Steam reforming of methane over nickel catalysts at low reaction temperature[J]. Appl Catal, 2004, 258: 107-114.

[10] PASCALINE L B M，CUONG P H. Ni/SiC: A stable and active catalyst for catalytic partial oxidation of methane[J]. Catal Today, 2004, 46: 91-92.

[11] ANDREAS J. Catalytic upgrading of tarry fuel gases: A kinetic study with model components[J]. Chemical Engineering and Processing, 1996, 35: 487-494.

[12] OUYANG Zhao-bin. Experimental study of synthesis gas production by coal and natural gas co-conversion process[J]. Fuel Processing Technology, 2006, 87: 599–604.

[13] BHAT R N, SACHTLER W M H. Potential of zeolite supported rhodium catalysts of the CO₂ reforming of CH₄[J]. Appl Catal A: General, 1997, 150: 279-283.

[14] ANTONINHO V, NEFTALI L V C. Role of Vanadium in Ni/A1₂O₃ catalyst for carbon dioxide reforming of methane[J]. Appl Catal, 2003, 255: 34-37.

[15] SZEKLY J, EVANS J W, SOHN H Y. Gas-solid reaction[M]. London: Academic Press, 1976.

(Edited by YANG Hua)

Foundation item: Project(291054) supported by Postdoctoral Fund of China

Received date: 2007-06-11; Accepted date: 2007-08-27

Corresponding author: ZHANG Jia-yuan, PhD, Associate professor; Tel: +86-731-8876111; E-mail: zjyzhq@mail.csu.edu.cn