3.2 Catalytic mechanism analysis
As known, the char gasification mainly involves the adsorption and desorption process of C-O complex[28]: the active site Cf on the char reacts with CO2 to form CO and the intermediate C(O) containing O on the surface, and then the CO separates from the intermediate to produce the active site of char. Therefore, the gasification reactivity of char is directly related to the active carbon sites.
The catalytic mechanism of the char gasification mainly includes oxygen transfer theory, electrochemical theory and intermediate theory. According to the theory of oxygen transfer[8], the catalyst acts as the intermediate of oxygen storage and transfer in carbon gasification. According to the electrochemical theory [29], the catalyst transfers the charge of carbon skeleton on coal surface through partial bonding with C atom, changes the electron cloud distribution of carbon skeleton on char surface, weakens the binding strength of C-C bond, enhances the binding force of C-O bond, and makes the surface of char have more reactive sites. According to the intermediate theory [30], the catalyst forms an electron donor receptor (EDA) complex with aromatic carbon in carbon. EDA has strong catalytic activity and can catalyse the reactions of C-H2, C-H2O and C-CO2. However, due to the complexity of biomass components and carbon structure, so far, there is lacking of understanding of the catalytic mechanism of the biomass addition on the char gasification.
The characteristics of carbon structure, such as molecular structure, crystal structure, active sites concentration and surface area, have a direct impact on the active carbon sites.
In order to reveal the mechanism of biomass addition on char gasification, the mineral components in the co-pyrolysis char ash were analyzed, as summarised in Table 2. The obtained results showed that SiO2 and Al2O3 were the key components in the ash samples. The two oxides accounted for > 70% of the total composition. It was also found that the proportions of K2O significantly increased with the increasing of biomass proportion. In contrast, the proportions of CaO, MgO, and Na2O only increased slightly, while the amounts of SiO2, Al2O3, Fe2O3, and TiO2 decreased with increasing proportion of biomass. These outcomes suggested that the selected biomass was rich in potassium, which could act as a catalyst in the gasification of co-pyrolysis char.
Several researchers previously reported that the presence of potassium in biomass was conducive to increasing the amount of active sites on the carbon surface during gasification [31, 32]. Thus, the morphologies of the co-pyrolysis chars were charactered by the SEM firstly, as shown in Fig. 3. It can be found that the surface of the coal char was compact without obvious pore structure, and only a few small particles adhere to the surface. After mixing 5% WS into co-pyrolysis process, the surface becomes rough, and the number of small particles on the surface increases obviously, and a few small pore structures can be found. With increasing WS proportion to 10%, the original small grooves on the surface of char begin to expand and form a large number of lamellar structures, and the pore structure has developed significantly regardless of size or depth, and most of the small particles adhere to the lamellar structures. When WS proportion increases to 15%, the lamellar structures deepen, the pore structure continues to increase, and the small particles attached to the lamellar structures decrease gradually. While, the gully structure disappeared in the co-pyrolysis char with 20% WS, which developed to larger pore structure, and there were a few pore structures with smaller pore size around the macropore. During the co-pyrolysis of biomass and coal, alkali metals begin to adsorb in the initial pores of char, and then gradually diffuse into the inner carbon matrix. Alkali metals are embedded in the graphite layer of coke carbon matrix to form alkali metal graphite intercalation compounds, such as KC6 and KC60 [33]. In this process, the internal stress of carbon matrix increases, resulting in the volume expansion of char. In addition, the reaction of alkali metals with other minerals in char will also cause the volume expansion of minerals, and then change the structure of char. In the gasification process, the large number of layered structures make the char structure loose, which makes it easier to react with CO2.
The XRF of the ash and the SEM results of the char together illustrated that the presence of alkali metals (especially potassium) in biomass plays an important role in the development of the porous structure [34, 35], and led to an increase the amount of active sites on the carbon surface per unit weight [31, 32]. The reaction between alkali metals and other minerals in char will resulted in the expansion of mineral volume, which is conducive to the development of pore structure [36]. Moreover, alkali metals in the biomass will adsorbed on the surface of the char first and then gradually diffused into the internal carbon structure and embedded in the graphite layer to form alkali metal-graphite intercalation compounds which also will resulted in the expansion of pore structure of char[33].
The results of previous studies also demonstrated that alkali metals significantly affected the carbon crystallite characteristics, including the crystallisation height (Lc), microcrystalline size (La), and crystal layer spacing (d002)[21, 34]. The results of the XRD analysis the prepared chars are presented in Fig. 4. The minor peaks in the patterns were caused by the presence of inorganic minerals in coal and biomass, while the sharp peaks at approximately 27° was attributed to the existence of SiO2. Furthermore, the (002) band at ~ 25° and the (100) band in the vicinity of graphite at ~ 43° [37] corresponded to graphite-like structures (i.e., crystalline carbon). The appearance of γ peaks at approximately 20° suggested the existence of aliphatic side chains attached at the periphery of the carbon crystallites [38]. The obtained outcomes implied that the structures of the carbon crystallites were between graphite and the amorphous state.
Figure 5 illustrates the XRD patterns after smoothing and processing by Peakfit 4.2. It was found that the intensity of the (002) and (100) bands decreased with an increase in the biomass proportion, which indicated that biomass doping could inhibit the graphitisation of char during pyrolysis. The results also demonstrated that the intensity of the γ peak increased with an increase in the biomass proportion. It was speculated that this was caused by the inhibition of the breaking of the side chains in the aliphatic moieties by the catalyst components in the biomass during pyrolysis. It was also hypothesised that there were more aliphatic chains in raw biomass than in raw coal.
The calculated La, Lc, and d002 values are shown in Fig. 6. As it can be seen, the values of La and Lc decreased with an increase of the biomass proportion. In contrast, the values of d002 exhibited a reverse trend. These observations indicated that the order of aromatic moieties in the sample was more irregular in space, this is due to as the increasing of biomass addition, the content of alkali metals in the co-pyrolysis char increased, especially potassium, which means more alkali metals atoms penetrate into the carbon structure and occurs intercalation reaction, which affects the carbon microcrystalline structure of char disintegrate and makes the carbon structure become looser. in a word, the addition of biomass makes the microcrystalline structure of carbon develop to reverse graphitisation during co-pyrolysis. The ash XRD patterns of the co-pyrolysis char was shown in Fig. 7, the alkali metals in char samples mainly exist as KAlSiO4 and NaAlSi3O8, when the biomass content increases, the peak of KAlSiO4 becomes sharper, which indicates that K in biomass reacts with other ash contents to form KAlSiO4.
To further verify the XRD results of the chars, we conducted Raman spectroscopy analysis to evaluate the chemical features of highly disordered carbon in the co-pyrolysis chars. We averaged multiple Raman spectra of each sample to account for the heterogeneity of the char particles. The Raman spectra obtained in the range from 800 to 2000 cm− 1 are illustrated in Fig. 8.
The spectra were deconvoluted into four Lorentzian bands (designated as the G, D1, D2, and D4 bands) and one Gaussian band (labelled as the D3 band). The D1 band at ~ 1350 cm− 1 was attributed to the vibration between the aromatic rings and aromatic moieties with no less than six rings. It belonged to the A1g vibration of the amorphous hexagonal irregular lattice structure, and represented disorder in the carbon structure [39]. The D2 band was detected at the shoulder band of the G band, which corresponded to the vibrational mode of the disordered graphite lattice. It was attributed to the E2g mode of the symmetric graphite lattice [40]. On the other hand, the D3 band observed at ~ 1500 cm− 1 was ascribed to the amorphous sp2-bonded forms of carbon [41, 42]. In addition, the D4 band at ~ 1250 cm− 1 was attributed to amorphous mixing of the sp2–sp3-bonded forms of carbon. Lastly, the G band was the graphitic band. Subsequently, the band area ratios (ID1 + ID3 + ID4)/IG and IG/Iall were calculated to indicate the reacting sites and the extent of graphitisation in the co-pyrolysis chars. The variation of the band area ratios (ID1 + ID3 + ID4)/IG and IG/Iall with the biomass proportion is shown in Fig. 9. The results implied that (ID1 + ID3 + ID4)/IG increased with increasing biomass proportion, while IG/Iall exhibited a reverse trend. This suggested that the graphitic microcrystallite size decreased. Importantly, the Raman spectroscopy analysis was consistent with the XRD evaluation.
3.2 Kinetic analysis
The char-CO2 gasification mainly involves the adsorption and desorption process of C-O complex: CO2 in the atmosphere diffuses to the surface of particles and adsorbs on these surfaces, and then reacts with carbon; the reaction product desorbs and diffuses from the surface to the outside of the particles, which means all of the chemical reaction rate, internal diffusion rate and external diffusion rate may become the limiting-step in the char gasification. Therefore, it is necessary to analyse the limiting-step before studying the char gasification kinetics. Figure 10 shows that the relationship between ln(R) and 1/T at different gasification conversion (X) presents a good linear relationship, with R2 ≥ 0.95, which indicated that the investigated gasification reactions were Regime I and under chemical reaction control.
Random pore model (RPM) has been proposed as the optimal model to define the char gasification kinetics under chemical reaction control, which can address the char structure evolution during gasification. The gasification rate and carbon conversion curves were first defined by RPM (Fig. 11 (a–c)). As it can be seen, the gasification rate initially increased and then decreased with carbon conversion. Moreover, the maximum gasification rate was noted at a carbon conversion of 0.25–0.4. However, RPM was not a suitable model for the prediction of char gasification behaviours. This is due to the synergistic effects of the interactions between carbon and minerals as well as between minerals themselves during char gasification.
Thus, considering the assumptions of RPM and basing on the specific characteristics of the obtained experimental data, an extended RPM model is proposed in this work. Meanwhile the following assumptions are made in this model:
-
The char particles are porous, and the pores are cylindrical holes with uneven diameters, and the CO2 gasification predominantly occurs on the inner surface of the char particles;
-
The variation in the reaction area is a result of the interaction between the pore structures and the consumption of carbon active sites;
-
In the initial stage of gasification, CO2 rapidly reacts with the amorphous carbon structures, leading to a gradual decrease in the amorphous carbon content, the subsequent main reaction between CO2 and the aromatic carbon structure is very slow;
-
The interactions between carbon and minerals as well as between minerals themselves affect the pore structures of chars, the interactions between carbon and minerals also affect the carbon crystallite characteristics, and these interactions is related to gasification conversion.
Based on the conducted calculations, it was determined that the influence of the interactions between carbon and minerals as well as between minerals themselves on the entire gasification process can be expressed by the following equation (Eq. 3):
An extend random pore model (eRPM) will be obtained by putting the Eq. 3 into RPM, as shown in Eq. 4.
where, α indicates a dimensionless dimension related to the mineral composition and gasification temperature, ψ is the porous structural parameter.
Thus, a semi-empirical formula (G(x)) was introduced to the original RPM, which was named eRPM (Eq. 3). The corresponding fitting results are shown in Fig. 12 (a–c). Notably, significantly better fitting results were obtained employing eRPM than RPM. In addition, the experimental data was consistent with the theoretical data calculated by eRPM. Importantly, as shown in Table 3, most of the R2 values were high (R2 ≥ 0.99). It was also found that the k values increased with increasing biomass composition at a given temperature, which confirmed the catalytic effect of biomass. The calculated α values ranged between 0.69 and 1.70. Furthermore, the ψ values for gasification of the prepared chars were in the range of 4.5–26.16. The above calculations confirmed that eRPM proposed in the present study could accurately describe the kinetic behaviour of char gasification. Moreover, it was demonstrated that the synergistic effects of the interactions between chars and minerals as well as between minerals themselves on the gasification reaction were valid.
Table 3
Kinetic parameters and R2 values calculated by eRPM.
Temperature
|
Parameter
|
Sample
|
0% bio
|
5% bio
|
10% bio
|
15% bio
|
20% bio
|
950 oC
|
R2
|
0.97
|
0.99
|
0.99
|
0.99
|
0.99
|
k
|
0.0024
|
0.0039
|
0.0063
|
0.0087
|
0.011
|
α
|
1.21
|
1.23
|
1.28
|
1.67
|
1.70
|
ψ
|
24.14
|
26.16
|
14.43
|
17.33
|
22.01
|
1050 oC
|
R2
|
0.99
|
0.99
|
0.99
|
0.99
|
0.99
|
k
|
0.026
|
0.028
|
0.033
|
0.045
|
0.056
|
α
|
0.92
|
0.95
|
1.04
|
1.22
|
1.34
|
ψ
|
15.43
|
14.73
|
14.19
|
11.17
|
10.02
|
1150 oC
|
R2
|
0.99
|
0.99
|
0.99
|
0.99
|
0.99
|
k
|
0.094
|
0.11
|
0.12
|
0.14
|
0.15
|
α
|
0.96
|
0.96
|
1.00
|
0.86
|
0.69
|
ψ
|
8.09
|
6.75
|
5.97
|
5.13
|
4.51
|
The above calculations confirmed that eRPM proposed in the present study could accurately describe the kinetic behaviour of co-gasification. Moreover, it was demonstrated that the synergistic effects of the interactions between chars and minerals as well as between minerals themselves on the gasification reaction were valid. Hence, G(x) in Eq. 3 could be used to predict the synergistic influence of the above interactions on the entire co-gasification process. The relationship between these effects and the gasification conversion is demonstrated in Fig. 13 (a–c). Furthermore, Fig. 13(a–c) shows that the catalytic effect of minerals on co-gasification gradually decreased during gasification. This was caused by the gradual consumption of the carbon matrix as well as by the strengthening of the interactions between minerals, which resulted in the deactivation of the catalyst [9]. The data shown in the above figures also indicated that the catalytic effects of the catalyst component on co-gasification decreased with increasing biomass proportion when the temperature was below 1050°C. We speculated that this was caused by the inhibition of the carbon lattice growth by the catalyst component in biomass and by the enhancement of the production of volatile matter, which in turn resulted in an increased amount of reactive sites in co-pyrolysis char. In addition, as shown by the XRD and Raman spectroscopy analyses, the amount of reactive sites increased with an increase in the biomass proportion [43]. In contrast, the effect of the catalyst on the amount of reactive sites was smaller. Figure 13c demonstrates that when the temperature was raised to 1150°C and the biomass increased to 15%, the interaction between the catalyst and char, particularly the graphite-like carbon matrix, was enhanced, and the catalytic effect of the catalyst was improved.
The Arrhenius law was applied to determine the activation energy (Ea) and pre-exponential factor (k0) values according to the following equation (Eq. 5):
where, T is the absolute isothermal gasification temperature and R indicates the universal gas constant. The k values obtained from eRPM were used to calculate the values of Ea and k0. The Arrhenius plots (lnk vs. 1/T) acquired by linearising plots obtained by Eq. 5 are shown in Fig. 14. In addition, the calculated Ea and k0 values are summarised in Table 4. It was found that the Ea values of chars decreased from 266.5 to 190.8 kJ/mol with an increase in the biomass composition from 0 to 20%. The decrease of the Ea values further confirmed the catalytic activity of the biomass.
Table 4
The Ea, k0, and R2 values for the char gasification reaction based on eRPM.
Sample
|
Ea (kJ/mol)
|
lnk0
|
R2
|
0% bio
|
266.5
|
20.30
|
0.97
|
5% bio
|
240.2
|
18.14
|
0.99
|
10% bio
|
216.4
|
16.23
|
0.99
|
15% bio
|
198.9
|
14.88
|
0.99
|
20% bio
|
190.8
|
14.32
|
0.98
|
The char gasification is almost simultaneous with the reduction reaction of oxygen carriers, while the carbon crystallite and structural characteristics of the chars exhibit direct effects on the gasification reactivity. Studying the relationship between gasification reactivity and char microstructure can help to further understand the char gasification behaviours in the iG-CLC process. The diagram of the relationships between the char parameters (i.e., SBET, La, Lc, (ID1 + ID3 + ID4)/IG, and IG/Iall) and the activation energies is illustrated in Fig. 15. The results indicated that most of the parameters (i.e., SBET, Lc, (ID1 + ID3 + ID4)/IG and IG/Iall) displayed linear correlations with the activation energies (R2 ≥ 0.95). On the other hand, 2nd order polynomial correlation was noted between the La parameter and the activation energies. Based on the values of R2, the crystallite characteristics, and therefore aromaticity, had a significant impact on the activation energies[44]. These observations indicated that the crystallite characteristics governed the char gasification rates during experiments conducted in a thermogravimetric analyser. Additionally, the structural characteristics of char also had an important effect on the reactivity of gasification of co-pyrolysis char. Overall, we determined that the reactivity of gasification of co-pyrolysis char composed of coal and biomass could be predicted based on characteristic char parameters, such as SBET, La, Lc, (ID1 + ID3 + ID4)/IG, and IG/Iall.