Heat Transfer and Kinetics Study of Moroccan Oil Shale Pyrolysis Process

Zhan Chenyu; Ma Yue; Yue Changtao; Li Shuyuan; Tang Xun; Wen Hua

(1. China University of Petroleum, Beijing 102249;2. Beijing Guodian Longyuan Environmental Protection Engineering Co., Ltd., Beijing 102249)

Abstract: The basic properties of Moroccan oil shale were analyzed in this paper. Pyrolysis experiments at different heating rates were carried out by thermogravimetry. The results show that the pyrolysis process of Moroccan oil shale can be divided into three steps. A shrink-nuclear model that considers the internal heat transfer of particles was established,which can better simulate the pyrolysis process of oil shale. The pyrolysis kinetic parameters of Moroccan oil shale were calculated based on heat transfer data and basic physical parameters. The results show that the apparent activation energy of the reaction is around 120 kJ/mol, and the apparent frequency factor is around 2×1014 mol/(s•m2).

Key words: kinetics; oil shale; pyrolysis; shrink-nuclear model

1 Introduction

Oil shale pyrolysis kinetics can be studied under isothermal and non-isothermal conditions. Since the nonisothermal method for measuring kinetic data requires a short test time, the initial temperature rise error that cannot be ignored by the isothermal method can be avoided, so most studies are carried out under nonisothermal conditions. The kinetic parameters of oil shale pyrolysis were determined by a variety of methods,including the integral method, the differential method,the Coats Redfern method, the Friedman method, the maximum pyrolysis rate model, the distributed activation energy model, and the parallel first-order reaction model.Most studies have shown that the kinetics of pyrolysis reaction of oil shale and kerogen can be described by the first-order reaction rate equation[1-4]. In addition, some studies have considered that the oil shale first forms pyrolysis asphalt and residual carbon, when the oil shale is thermally decomposed, and then the pyrolysis asphalt is further decomposed. Oil shale pyrolysis uses pyrolysis bitumen as an intermediate product, and its reaction kinetics can be described by the parallel first-order reactions[5-13].

In recent years, many scholars at home and abroad have used various new models and mathematical methods to carry out new pyrolysis kinetics studies on the pyrolysis process of oil shale. EC Moine, et al.[14]studied the pyrolysis kinetics of oil shale in the Rif region of Morocco under non-isothermal conditions. The temperature range was between 325 °C and 600 °C. The pyrolysis process of oil shale was successfully divided into three discrete processes by the Fraser-Suzuki algorithm for asymmetric functions. Maaten, et al.[15]applied the non-isothermal thermogravimetric analysis to study the pyrolysis kinetics of oil shale in the United States, China, and Estonia.The TGA curve clearly shows that the pyrolysis process of all oil shale samples is independent of its origin, and the pyrolysis process mainly occurs in the temperature range of 300―500 °C. Moine, et al.[16]used the multistage parallel reaction method and the Fraser-Suzuki distribution method to fit the non-isothermal TG/DTG curves of oil shale in the Timahdit region of Morocco.The pyrolysis process of oil shale is carried out in two separate reactions. Wang Qing, et al.[17]used the FLASHCHAIN model based on oil shale structure to simulate the formation process of pyrolysis products.This model was used to simulate the oil shale pyrolysis process, which could provide a theoretical support for oil shale development and comprehensive utilization in the future.

At present, although the shrink-nuclear model already has a theoretical basis, the pyrolysis process of this model for oil shale is rarely reported. The Galoter furnaces and the Fushun furnaces currently used are all granular or lumpy, but the experimental materials in their laboratory are powders with a particle size of below 0.072 mm. The data obtained by the thermogravimetric measurement ignores the effect of heat transfer. The adopted shrinknuclear model takes into account the influence of heat transfer inside the particles, and the experiment is also conducted with particles, which can better simulate the pyrolysis process of oil shale and make it closer to the actual situation. Therefore, this paper uses the shrinknuclear model to simulate the pyrolysis process of oil shale, and compares the predicted values of the model with the experimental results to verify the applicability of the model to oil shale.

2 Experimental

The oil shale samples used in this experiment were collected from the Tarfaya region of Morocco. The diameter of samples used in thermogravimetric analysis and heat transfer experiments is 30 mm. A small hole with a diameter of 1.5 mm (therm. diameter) and a sample radius is drilled in the center of the sample with a twist drill. In order to avoid cracking of the oil shale sample during the experiment, certain measures can be taken to fix it if necessary.

The spherical oil shale particle samples with a diameter of 30 mm were subjected to the thermogravimetric weight loss test at a heating rate of 2 °C/min, 5 °C/min,and 10 °C/min, respectively. The pyrolysis experiment was carried out at the same heating rate, and was finally heated to a final temperature of 550 °C. According to the characteristics of oil shale pyrolysis stage, the temperature range of 300―550 °C was selected as the key temperature region for studying the pyrolysis reaction.The thermogravimetric data were used to calculate the relationship between conversion rate and temperature at different heating rates.

3 Kinetic Model

The internal temperature distribution equation of oil shale particles is introduced[18], as shown below.

where,Tis the temperature, K;T0is the initial temperature, K;βis the heating rate, K/s;tis the heating time, s;R0is the sample radius, m;λpis the averaged coefficient of heat conductivity of oil shale and air, J/(s·m·K);ris the radius of oil shale particle, m; ΔHis the absorbed heat during pyrolysis, J/kg;W0is the mass loss per unit volume during oil shale pyrolysis process, kg/m3;ρsis the density of oil shale, kg/m3;Cpis the averaged heat absorption capacity of oil shale and air, kJ/(kg·K);Tsis the heating temperature around surface, K;Tcis the central temperature of reactor, K.

At a constant rate of temperature rise, the pyrolysis reaction time is relatively long, which means thattis relatively large, so the termin the above formula (2) is negligible. Also the term is calculated to give the following result:

So equation (2) is simplified as follows:

During the pyrolysis process of granular oil shale, the mass is gradually reduced due to the volatilization of oil and gas, and the particle volume is decreasing at the same time, while the temperature is transmitted from the outside to the inside. The final internal and external temperature is nearly close to the same value. Therefore, the pyrolysis kinetics can adopt the shrink-nuclear model[19]. As shown in Figure 1, the gas-solidification reaction is carried out on the surface of the solid. As the reaction continues,the reaction surface gradually advances inward, and the unreacted core shrinks. The shrink-nuclear model is generally applied to a gas-solid phase reaction system, in which the chemical reaction rate is much higher than the diffusion rate of the reaction gas in the core and the solid reactant is dense.

Figure 1 Pyrolysis process of spherical particles in shrinknuclear model

The shrink-nuclear model is used to explain the pyrolysis process of oil shale. The particle size of the spherical oil shale is unchanged, and the pyrolysis reaction firstly proceeds from the surface of the spherical particles. When the reaction progresses, the ash layer gradually forms, and the size of the unreacted nucleus shrinks with the reaction process until it completely disappears.

The expression for converting the thermogravimetric data into the change of the mass loss of pyrolysis with temperature is:

where,XAis the weight loss conversion rate of sample pyrolysis;ω0is the original sample weight, g;ωtis the weight of the sample at timet, g; andω∞is the weight of residual sample after complete pyrolysis, g.

The organic matter in oil shale is mainly composed of five elements: C, H, O, N, and S. The main components are C and H, while the contents of O, N, and S are few. The mass conversion rate and the molar conversion rate during the pyrolysis process of oil shale will be offset into the formula. They only represent numerical values, which are dimensionless, and would not affect the simulation of the pyrolysis process of oil shale. Therefore, the mass conversion rate and the molar conversion rate during oil shale pyrolysis can be approximately considered to be equal.

The shrink-nuclear model for oil shale pyrolysis can be considered as the first-order reaction of chemical reaction control, and the pyrolysis rate equation based on the shrinkage area can be written as:

where,rcis the unreacted nuclear radius, m;NAis the amount of substance of organic matter in the sample,mol;kS0is the reaction rate constant, m/s;CAis the molar concentration of sample, mol/m3;CA0is the initial molar concentration of the sample, mol/m3; andksis the reaction rate constant, mol/(s·m2).

Since the reaction rate constant and the initial molar concentration of the sample are constant, let:

then there are:

It is known thatNA=ρAV, you can get:

where,ρAis the density of pyrolyzable organic matter in oil shale, mol/m3; andVis the sample volume, m3.

Substituting equation (9) into (8), you can get:

That is:

into equation (11) gets:

in whichAis the apparent frequency factor, mol/(s·m2);andEis the apparent activation energy, J/mol.

Taking the logarithm of the two sides of the formula (13),you can get:

4 Results and Discussion

The thermal properties of oil shale are shown in Table 1.These parameters are obtained by referring to relevant literature, including the heat transfer coefficient, the specific heat capacity of oil shale, and the convective heat transfer coefficient[18,20-21].

Table 1 Analog parameter settings

The heat transfer results of 30-mm oil shale particles heated at a heating rate of 10 °C/min are shown in Figure 2. The measured temperature is consistent with the temperature fitted by the temperature distribution equation, so the temperature distribution equation is suitable for calculating the temperature of the pyrolysis process.Figure 3 is a graph showing the thermal weight loss of Moroccan oil shale measured at different heating rates.

Figure 2 Temperature curve of 30 mm oil shale particles at a heating rate of 10 °C/min

Figure 3 Thermal weight loss curves of Moroccan oil shale obtained at different heating rates

It can be seen from Figure 3 that the trend of three curves is roughly the same, showing two distinct weight loss peaks. One peak appeared before 300 °C and one peak appeared between 300 °C and 550 °C. This is consistent with the previous research results[15]. The pyrolysis process of oil shale goes through two stages. The first stage proceeds from room temperature to 300 °C, which is mainly the drying stage of oil shale. At this stage, the main manifestation is the water loss of oil shale, when the interlayer water and crystal water of clay minerals and surface water are lost at this stage, which is accompanied by partial weak bond fracture. The second stage of the oil shale pyrolysis process occurs between 300 °C and 550°C, which is also an important stage of oil shale pyrolysis.With the important reactions such as the decomposition of long-chain alkanes, the cleavage of carbon-carbon bonds in the side chain of the aromatic nucleus, and the cleavage of bridge bonds, the second weight loss peak appears at the position of 400 — 550 °C. Since this temperature range mainly belongs to an important stage of oil and gas production, the decomposition of oil and gas products from kerogen causes the occurrence of the second weight loss peak. After the reaction reaches 550 °C, the weight loss has almost stopped, indicating that the organic matter in the oil shale has basically reacted completely and the pyrolysis process is over.

The relationship between the conversion rate and the temperature at different heating rates is shown in Figure 4.

Figure 4 Relationship between pyrolysis conversion rate and temperature at different heating rates

It can be seen from Figure 4 that although the heating rate is not the same, the conversion rate of the oil shale pyrolysis process is consistent with the temperature change trend. The conversion rate of pyrolysis reaction changed little before 400°C, and the conversion rate of pyrolysis reaction in the 400―550 °C stage changed obviously.After 550 °C, the pyrolysis reaction was complete, and the conversion rate became stable, which basically reached the maximum conversion rate. During the pyrolysis of oil shale, the heating rate has little effect on the relationship between the conversion rate and the temperature.

As for the data in Figure 5, the conversion rate is derived from the temperature to obtain the function value dx/dT of the conversion rate and the temperature. It is known that the heating rate is dT/dt, and multiplying the two can obtain a function value dx/dtof conversion rate and time.Figure 5 is a graph showing the relationship between dx/dtand temperature of Moroccan oil shale pyrolysis at different heating rates. It can be clearly seen from Figure 5 that as the temperature rises, the change trend of dx/dtat different heating rates is about the same. They are all gentle at first, then rise to a maximum value, and finally decrease to zero. The higher the heating rate of the oil shale pyrolysis process, the larger the maximum dx/dtof the oil shale pyrolysis, and the temperature point at which the maximum occurs has obviously shifted to the high temperature region.

Figure 5 Relationship between dx/dt and temperature of pyrolysis of Moroccan oil shale at different heating rates

Since most of today’s oil shale retorting technology uses bulk oil shale samples, only studying the intrinsic kinetics of oil shale pyrolysis reaction cannot meet the current industrial demand for oil shale. In order to better exploit and utilize the block oil shale, the influence of the internal temperature distribution is also considered in the calculation of the kinetics. The derived model equation of oil shale pyrolysis process is shown in Equation 14. The results of the mass loss rate of pyrolysis and temperature, and the internal temperature distribution equation of oil shale particles were used to calculate the kinetic parameters. The apparent activation energy E and the frequency factor A can be obtained by nonlinear least squares method using the MATLAB calculation software.The kinetic parameters are shown in Table 2.

Table 2 Calculation results of the shrink-nuclear model of the Moroccan oil shale pyrolysis process

It can be seen from Table 2 that the heating rate does not have a significant effect on the kinetic parameters for pyrolysis of oil shale. The apparent activation energy of the pyrolysis reaction of Moroccan oil shale is around 125 kJ/mol, the apparent frequency factor is around 2.5 mol/(s•m2),and R2is above 0.98. This model can better describe the pyrolysis process of Moroccan oil shale particles.

5 Conclusions

A shrink-nuclear model considering the internal heat transfer of particles was developed. The shrink-nuclear model was used to simulate the pyrolysis process of oil shale. The kinetic parameters of oil shale pyrolysis were calculated. The main conclusions are as follows:

(1) According to the weight loss diagram of pyrolysis process, the pyrolysis process of oil shale is mainly divided into two stages. The first stage proceeds from room temperature to 300 °C, which is mainly the drying stage of oil shale. The second stage occurs between 300 °C and 550 °C, which is the most important stage of oil shale pyrolysis to produce oil and gas.

(2) The results of the shrink-nuclear model show that the apparent activation energy of the Moroccan oil shale pyrolysis reaction is around 120 kJ/mol, the apparent frequency factor is around 2×1014mol/(s•m2), and theR2is greater than 0.98.

Acknowledgements:The authors would like to thank the Science and Technology Department of Beijing Guodian Longyuan Environmental Protection Engineering Co., Ltd., for the key technology research project of using large-scale coalfired boilers to treat sludge (No. KH-2018-06).