**3. Results and discussion**

,

#### **3.1. Proof of the concept of the fluidized bed TGA**

The proof of the concept of the fluidized bed TGA was carried out with the thermal decomposition of the calcium hydroxide. The results for the fluidized bed and conventional TGAs are shown in **Figure 3**. For the conventional TGA, three different amounts of calcium hydroxide (10, 25, and 140 mg) were tested, while 4 g of calcium hydroxide were used in the FB-TGA.

**Figure 3.** Saddle reactor.

For the first temperature interval, the results with 25 mg from the conventional TGA were similar to the ones for the FB-TGA with 4 g of calcium hydroxide. The conventional TGA results for 10 and 25 mg are similar, but they are different from those obtained for 140 mg. Two different parts can be distinguished in **Figure 4**. For 10 and 25 mg, the first part can be defined from 370°C to 470°C, whereas this first part is from 395°C to 565°C for 140 mg [1].

The weight loss measurement and the quantity of gas produced showed general agreement,

decomposition: comparison between conventional and fluidized bed TGAs. Reproduced from

Innovative Microreactors for Low-grade Feedstock Gasification

http://dx.doi.org/10.5772/intechopen.74602

245

Due to a 170–209oC temperature gradient, the obtained values of some activation energies are lower or higher than what was found in the literature. The obtained activation energy for the CO shift reaction was 195.0 kJ/mol. This value is 20% higher than the one in the literature. The obtained activation energies for coal pyrolysis reactions were 19–21% lower than the ones obtained in the literature for similar coals [2]. These new results confirm that there was a measurement error of the temperature in certain past studies. Such measurement error means that for the past studies, the reaction chamber temperature was not the one that is used to find kinetic parameters. Thus, there is a temperature gradient between the measured temperature by the thermocouple of the conventional thermogravimetric analyzer and the real one of the reaction. Such increase of 19–21% of the activation energy is the equivalent of 185–209°C temperature gradient of similar relatively exothermic reactions [2, 4, 15]. Finally, the results obtained were as expected and confirmed that the FB-TGA provides reliable kinetic parameters. More discussion and results are shown in our previ-

The second experiment that was carried out on the FB-TGA is about the catalytic ash free coal gasification. In this work, the effect of the catalyst on ash free coal gasification is included. A comparison of the total product gas yield and the weight loss is represented in **Figure 5**.

for both coal pyrolysis and gasification results.

**3.3. Catalytic ash-free coal gasification in a fluidized bed TGA**

ous work [3, 4].

**Figure 4.** Ca(OH)2

reference [1].

For the first part, the heat transfer limitation and/or the temperature gradient are responsible for the difference between the results obtained for 10, 25, and 140 mg. The intraparticle and the external diffusions of the produced water through a small layer of CaO that was formed around the Ca(OH)2 became the rate-controlling step of the thermal decomposition during the second step [1].

Only for the first step, the results for 25 mg of Ca(OH)2 in the conventional TGA agree with those for 4 g in the fluidized bed TGA. Indeed, the thermal decomposition of Ca(OH)2 in the FB-TGA was carried out in one stage, from 360°C to 540°C. Thus, a better heat transfer and mass transfer of water vapor was obtained with the use of the FB-TGA and no diffusion control was observed.

#### **3.2. Coal pyrolysis and gasification in the fluidized bed TGA**

The experiments of coal pyrolysis and gasification were carried out in the fluidized bed TGA. These experiments were used to derive novel kinetic parameters from the fluidized bed TGA.

**Figure 4.** Ca(OH)2 decomposition: comparison between conventional and fluidized bed TGAs. Reproduced from reference [1].

The weight loss measurement and the quantity of gas produced showed general agreement, for both coal pyrolysis and gasification results.

Due to a 170–209oC temperature gradient, the obtained values of some activation energies are lower or higher than what was found in the literature. The obtained activation energy for the CO shift reaction was 195.0 kJ/mol. This value is 20% higher than the one in the literature. The obtained activation energies for coal pyrolysis reactions were 19–21% lower than the ones obtained in the literature for similar coals [2]. These new results confirm that there was a measurement error of the temperature in certain past studies. Such measurement error means that for the past studies, the reaction chamber temperature was not the one that is used to find kinetic parameters. Thus, there is a temperature gradient between the measured temperature by the thermocouple of the conventional thermogravimetric analyzer and the real one of the reaction. Such increase of 19–21% of the activation energy is the equivalent of 185–209°C temperature gradient of similar relatively exothermic reactions [2, 4, 15]. Finally, the results obtained were as expected and confirmed that the FB-TGA provides reliable kinetic parameters. More discussion and results are shown in our previous work [3, 4].

#### **3.3. Catalytic ash-free coal gasification in a fluidized bed TGA**

For the first temperature interval, the results with 25 mg from the conventional TGA were similar to the ones for the FB-TGA with 4 g of calcium hydroxide. The conventional TGA results for 10 and 25 mg are similar, but they are different from those obtained for 140 mg. Two different parts can be distinguished in **Figure 4**. For 10 and 25 mg, the first part can be defined from 370°C to 470°C, whereas this first part is from 395°C to 565°C for 140 mg [1].

For the first part, the heat transfer limitation and/or the temperature gradient are responsible for the difference between the results obtained for 10, 25, and 140 mg. The intraparticle and the external diffusions of the produced water through a small layer of CaO that was formed

Only for the first step, the results for 25 mg of Ca(OH)2 in the conventional TGA agree with those for 4 g in the fluidized bed TGA. Indeed, the thermal decomposition of Ca(OH)2 in the FB-TGA was carried out in one stage, from 360°C to 540°C. Thus, a better heat transfer and mass transfer of water vapor was obtained with the use of the FB-TGA and no diffusion control was observed.

The experiments of coal pyrolysis and gasification were carried out in the fluidized bed TGA. These experiments were used to derive novel kinetic parameters from the fluidized bed TGA.

**3.2. Coal pyrolysis and gasification in the fluidized bed TGA**

became the rate-controlling step of the thermal decomposition during

around the Ca(OH)2

the second step [1].

**Figure 3.** Saddle reactor.

244 Gasification for Low-grade Feedstock

The second experiment that was carried out on the FB-TGA is about the catalytic ash free coal gasification. In this work, the effect of the catalyst on ash free coal gasification is included. A comparison of the total product gas yield and the weight loss is represented in **Figure 5**.

The two experimental results are in global agreement, and the slight difference is due to the produced tar from ash coal experiments.

gasification had the lowest carbon conversion result. Thus, coal beneficiation has a negative affect on carbon conversion. Nevertheless, at 700°C, there is an increase of carbon conversion by 15.3 and 52.6%, for coal and ash free gasification. These values increased to 44.5 and

Innovative Microreactors for Low-grade Feedstock Gasification

http://dx.doi.org/10.5772/intechopen.74602

247

Moreover, novel kinetic parameters are obtained form the FB-TGA experimental results for the reactions of partial oxidation, gas-water shift, and methane reforming. For char gasification, the results are similar to those found in literature [16]. For the gas-water shift and methane reforming reactions, the obtained activation energies were 56.5 and 77 kJ/mol, respectively. Compared to the values found in literature, these results are significantly lower. The lowest activation energy for the water-gas shift reaction was 70 kJ/mol [2, 17–19], which is 1.24 times

The lowest activation energy found in literature for the methane reforming reaction is 85 kJ/

The values of the activation energies of the CO shift and the methane reforming reactions decreased by 56% and 33%, respectively, by using the catalyst. This catalyst was applied on other reactions in the literature and the activation energy was decreased by 43 to 75% [2, 20].

Farag and his co-authors have employed the obtained experimental data from the developed MW-TGA, following the method presented in Section 2.3.3, to carry out a kinetic study based on the lumped approach. The established model in their work takes into consideration the chemical composition of the oil and aqueous products that are obtained from the microwave pyrolysis of lignin. As shown later, it considers that the virgin material converts into seven main products: remaining solid, phenolics, aromatic single ring and non-phenols (ASR-Non-Ph), aliphatics, heavy molecular weight compounds (HMWCs), water, and noncondensable gas, as shown later. The kinetic model demonstrated in Eq. (1) was used to estimate the kinetic parameters, activation energy, pre-exponential factor, and reaction order

mol [2, 18], which is 10% higher than the one obtained with the fluidized bed TGA.

69.1% at 750°C [2].

\_\_\_ *dS*

*dt* <sup>=</sup> <sup>−</sup>*kos*

(*e* −*Ea* \_\_\_\_*s*

− *koa* (*e* −*Ea* \_\_\_\_*a*

*RT* ) (*<sup>S</sup>* <sup>−</sup> *<sup>S</sup>*∞)*ns* <sup>=</sup> <sup>−</sup>*koPh*

*RT* ) (*<sup>S</sup>* <sup>−</sup> *<sup>S</sup>*∞)*na* <sup>−</sup> *koHMWC*

higher than the one obtained in the fluidized bed TGA.

**3.4. Microwave thermogravimetric analyzer**

More results and discussions are given in our previous article [2].

associated with every reaction toward producing these seven products.

(*e* <sup>−</sup>*Ea* \_\_\_\_*Ph*

(*e* <sup>−</sup>*Ea* \_\_\_\_\_ *HMWC*

*RT* ) (*<sup>S</sup>* <sup>−</sup> *<sup>S</sup>*∞)*nPh* <sup>−</sup> *koASR*−*Non*−*Ph*

*RT* ) (*<sup>S</sup>* <sup>−</sup> *<sup>S</sup>*∞)*<sup>n</sup>HMWC* <sup>−</sup> *kow*

(*e* <sup>−</sup>*Ea* \_\_\_\_\_\_\_ *ASR*−*Non*−*Ph*

(*e* <sup>−</sup>*Ea* \_\_\_\_*<sup>w</sup>* *RT* ) (*S* − *S*∞)*<sup>n</sup>ASR*−*Non*−*Ph*

(*e* −*Ea* \_\_\_\_*g*

*RT* ) (*S* − *S*∞)*ng* (1)

*RT* ) (*<sup>S</sup>* <sup>−</sup> *<sup>S</sup>*∞)*nw* <sup>−</sup> *kog*

A comparison of the carbon conversion results of coal, ash free coal and catalyst with ash free coal is illustrated in **Figure 6**. CatAFC, AFC, and coal, stand for the catalyst with ash free coal, ash free coal and coal gasification, respectively. The lowest carbon conversion is obtained from ash free coal for temperatures lower than 730°C, after which the coal

**Figure 5.** Catalytic ash free coal gasification in a fluidized bed TGA. Reproduced from reference [2].

**Figure 6.** Temperature effect on carbon conversion. Reproduced from reference [2].

gasification had the lowest carbon conversion result. Thus, coal beneficiation has a negative affect on carbon conversion. Nevertheless, at 700°C, there is an increase of carbon conversion by 15.3 and 52.6%, for coal and ash free gasification. These values increased to 44.5 and 69.1% at 750°C [2].

Moreover, novel kinetic parameters are obtained form the FB-TGA experimental results for the reactions of partial oxidation, gas-water shift, and methane reforming. For char gasification, the results are similar to those found in literature [16]. For the gas-water shift and methane reforming reactions, the obtained activation energies were 56.5 and 77 kJ/mol, respectively. Compared to the values found in literature, these results are significantly lower. The lowest activation energy for the water-gas shift reaction was 70 kJ/mol [2, 17–19], which is 1.24 times higher than the one obtained in the fluidized bed TGA.

The lowest activation energy found in literature for the methane reforming reaction is 85 kJ/ mol [2, 18], which is 10% higher than the one obtained with the fluidized bed TGA.

The values of the activation energies of the CO shift and the methane reforming reactions decreased by 56% and 33%, respectively, by using the catalyst. This catalyst was applied on other reactions in the literature and the activation energy was decreased by 43 to 75% [2, 20]. More results and discussions are given in our previous article [2].

#### **3.4. Microwave thermogravimetric analyzer**

The two experimental results are in global agreement, and the slight difference is due to the

A comparison of the carbon conversion results of coal, ash free coal and catalyst with ash free coal is illustrated in **Figure 6**. CatAFC, AFC, and coal, stand for the catalyst with ash free coal, ash free coal and coal gasification, respectively. The lowest carbon conversion is obtained from ash free coal for temperatures lower than 730°C, after which the coal

**Figure 5.** Catalytic ash free coal gasification in a fluidized bed TGA. Reproduced from reference [2].

**Figure 6.** Temperature effect on carbon conversion. Reproduced from reference [2].

produced tar from ash coal experiments.

246 Gasification for Low-grade Feedstock

Farag and his co-authors have employed the obtained experimental data from the developed MW-TGA, following the method presented in Section 2.3.3, to carry out a kinetic study based on the lumped approach. The established model in their work takes into consideration the chemical composition of the oil and aqueous products that are obtained from the microwave pyrolysis of lignin. As shown later, it considers that the virgin material converts into seven main products: remaining solid, phenolics, aromatic single ring and non-phenols (ASR-Non-Ph), aliphatics, heavy molecular weight compounds (HMWCs), water, and noncondensable gas, as shown later. The kinetic model demonstrated in Eq. (1) was used to estimate the kinetic parameters, activation energy, pre-exponential factor, and reaction order associated with every reaction toward producing these seven products.

In Eq. (1), ko is the pre-exponential factor [time−1], Ea is the apparent activation energy [J/mol K], T is the reaction temperature [K], and R is the universal gas constant [J/mol K]. The subscripts s, ph, ASR-Non-ph, a, HMWC, g, and w refer to solid, phenolic, aromatic single ring and non-phenolic, aliphatic, heavy molecular weight compound, gas and water products, respectively. **Figure 7**

demonstrates the experimental and predicted yield of these products, except water, which can be calculated by subtracting the summation of these products from unity. **Table 2** shows that the estimated kinetic parameters of each single reaction lead to the production of each of these products. For the full details regarding how these parameters were determined, kindly

Farag et al. believe that up to 725 K the condensable gas yield is slightly lower than that of the non-condensable gas, which could be the result of the swift split in lignin-side aliphatic hydroxyl groups [10]. Beyond 725 K, the total liquid yield continues increasing because of the decomposition of strong chemical bonds in the lignin network. Based on the estimated kinetic parameters, the reaction rate of the liquid products is lower than that of the solid product. The authors have claimed that the low secondary reactions under these conditions might be the reason for this result. The non-condensable gas product is mostly produced from the cracking of lignin-side chains, and the liquid product is produced from the breakdown of bonds between lignin aromatic rings. Therefore, the estimated activation energy of the non-condensable gas is lower than that of all oil products. Since the structure of the decomposed material's network is totally poly-aromatics, the reaction rate to produce phenolics and HMWC groups is much higher than that of ASR-Non-Ph. Accordingly, the estimated activation energy of aliphatics is greater than that of the other groups. The authors also think about the impact of microwave heating to decrease the probability of a secondary reaction when producing

In the scientific literature, an apparent contradiction to the interpretation of the influence of electromagnetic irradiation on reaction kinetics has been documented. Wang et al. 2013, Li et al. [26], Fukushima et al. 2013, Adnadjevic and Jovanovic [25], Sun et al. [22], Jovanović 2012, and a few other research groups believe that the reaction activation energy decreases under microwave irradiation [21–28]. On the other hand, Mazo et al. [28] and Yadav and

well known that the wavelength of microwaves is significantly longer than the intermolecular distance of the target, which ideally doubles the impact of MWH on the activation energy. However, this does not reject the probability of producing some intermediates that could

**Product ko [min−1] Ea [kJ/mol] n** Remaining solid 7 19 1 Water 9 27 1 Phenolics 21 38 1 HMWC 22 35 1 ASR-Non-Ph 1 40 1 Aliphatics 20 47 1 Condensable gas 22 29 1 Non-condensable gas 6 22 1

is the same in both cases, MWH and CH [13, 29–31]. It is

Innovative Microreactors for Low-grade Feedstock Gasification

http://dx.doi.org/10.5772/intechopen.74602

249

refer to [10].

aliphatics.

Borkar [27] have reported that *Ea*

**Table 2.** The estimated kinetic parameters of the Farag et al. [5] model.

**Figure 7.** Experimental and predicted yields of lignin pyrolysis products. Reprinted with a permission form [5].

demonstrates the experimental and predicted yield of these products, except water, which can be calculated by subtracting the summation of these products from unity. **Table 2** shows that the estimated kinetic parameters of each single reaction lead to the production of each of these products. For the full details regarding how these parameters were determined, kindly refer to [10].

In Eq. (1), ko

248 Gasification for Low-grade Feedstock

is the pre-exponential factor [time−1], Ea

is the reaction temperature [K], and R is the universal gas constant [J/mol K]. The subscripts s, ph, ASR-Non-ph, a, HMWC, g, and w refer to solid, phenolic, aromatic single ring and non-phenolic, aliphatic, heavy molecular weight compound, gas and water products, respectively. **Figure 7**

**Figure 7.** Experimental and predicted yields of lignin pyrolysis products. Reprinted with a permission form [5].

is the apparent activation energy [J/mol K], T

Farag et al. believe that up to 725 K the condensable gas yield is slightly lower than that of the non-condensable gas, which could be the result of the swift split in lignin-side aliphatic hydroxyl groups [10]. Beyond 725 K, the total liquid yield continues increasing because of the decomposition of strong chemical bonds in the lignin network. Based on the estimated kinetic parameters, the reaction rate of the liquid products is lower than that of the solid product. The authors have claimed that the low secondary reactions under these conditions might be the reason for this result. The non-condensable gas product is mostly produced from the cracking of lignin-side chains, and the liquid product is produced from the breakdown of bonds between lignin aromatic rings. Therefore, the estimated activation energy of the non-condensable gas is lower than that of all oil products. Since the structure of the decomposed material's network is totally poly-aromatics, the reaction rate to produce phenolics and HMWC groups is much higher than that of ASR-Non-Ph. Accordingly, the estimated activation energy of aliphatics is greater than that of the other groups. The authors also think about the impact of microwave heating to decrease the probability of a secondary reaction when producing aliphatics.

In the scientific literature, an apparent contradiction to the interpretation of the influence of electromagnetic irradiation on reaction kinetics has been documented. Wang et al. 2013, Li et al. [26], Fukushima et al. 2013, Adnadjevic and Jovanovic [25], Sun et al. [22], Jovanović 2012, and a few other research groups believe that the reaction activation energy decreases under microwave irradiation [21–28]. On the other hand, Mazo et al. [28] and Yadav and Borkar [27] have reported that *Ea* is the same in both cases, MWH and CH [13, 29–31]. It is well known that the wavelength of microwaves is significantly longer than the intermolecular distance of the target, which ideally doubles the impact of MWH on the activation energy. However, this does not reject the probability of producing some intermediates that could


**Table 2.** The estimated kinetic parameters of the Farag et al. [5] model.

by 56 and 33%, respectively. A novel microwave-TGA equipped with an innovated air thermometer and a product manifold was built. The developed system was used to predict the product yield and the bio-oil composition from pyrolysis of kraft lignin using a lumping approach. The experimental data were employed to estimate the kinetic parameters of every reaction to produce the solid, water, alipahtics, phenols, aromatic with a single ring aromatic

Innovative Microreactors for Low-grade Feedstock Gasification

http://dx.doi.org/10.5772/intechopen.74602

251

At a temperature below 725 K, the yield of the gas product was higher than that of the liquid products. This results from the swift split of the lignin chains that mainly produce water and/ or gas products. The estimated kinetic parameters showed that the rate of thermal cracking of lignin is higher than that of the liquid product, which points out that the possibility of secondary reactions is low. The formation of the ASR-Non-pH and aliphatics families is lower than that of the HMWC, which could originate from the complex structure of the lining.

A saddle reactor is developed, and good mixing was observed for 20–30 g of silica sand and

\*

1 Department of Chemical Engineering, Process Engineering Advanced Research Lab

[1] Samih S, Chaouki J. Development of a fluidized bed thermogravimetric analyzer (in

[2] Samih S, Chaouki J. Catalytic ash free coal gasification in a fluidized bed thermogravi-

[3] Samih S. Développement d'un analyseur thermogravimétrique à lit fluidisé : application à la gazéification catalytique du charbon. PhD Thesis, École Polytechnique de Montréal.

[4] Samih S, Chaouki J. Coal Pyrolysis and Gasification in a Fluidized Bed TGA. The

[5] Process Engineering Advanced Research Lab (PEARL). Available: http://pearl.polymtl.ca/

non-phenol, heavy molecular weight compounds, and gas product.

speed mixing of 15–30 RPM.

Said Samih1,2, Sherif Farag1,3,4 and Jamal Chaouki1

4 RMTech for Environmental Solutions Inc., Canada

English). Aiche Journal. Jan 2015;**61**:84-89

Canadian Journal of Chemical Engineering. 2018

\*Address all correspondence to: jamal.chaouki@polymtl.ca

(PEARL), Ecole Polytechnique de Montreal, Montréal, QC, Canada 2 AFMERICA TECHNOLOGY Inc., Corot, Verdun (QC), Canada

3 Faculty of Engineering at El-Mattaria, Helwan University, Cairo, Egypt

metric analyzer. Powder Technology. Jul 1, 2017;**316**:551-559

**Author details**

**References**

2016

**Figure 8.** Effect of the speed mixing on the temperature profile in the saddle reactor.

have different behaviors than that of the starting material which would impact the activation energy. In such cases, the dielectric properties of the produced intermediates should be measured and investigated to understand the potential of having hot spots that affect the reaction kinetics locally [14, 32, 33]. Therefore, further investigations are required to discover the fact behind the reported effects on the activation energy and other reaction kinetic parameters.

#### **3.5 Saddle reactor**

As explained previously, the saddle reactor is mainly used to avoid the dilution of the product gas. Indeed, the solid sample can be very well mixed in the saddle reaction chamber without using a gas mixing agent. To prove this concept, different masses of silica sand—from 20 to 30 g—were mixed and heated up to 350ᵒ C in the reactor. The results shown in **Figure 8** are for different mixing speeds, ranging from 15 to 30 RPM. The similar temperature profiles confirm the good mixing in the developed reactor.
