Co-combustion characteristics and kinetics of meager coal and spent cathode carbon block by TG-MS analysis

2.1 Materials

The raw materials were meager coal and typical SCCB. The meager coal was purchased from Taiyuan, Shanxi province, China. The typical SCCB used for 5 years was taken from an aluminum reduction cell in Shandong Weiqiao Aluminum Power Co., Ltd. The meager coal and separated SCCBs were pulverized, dried, and then ground evenly with a grinding pestle. After sieving, the powder smaller than 0.125 mm was obtained. Different qualities of meager coal and SCCB were placed in the experimental mixer (Siehe Industry, SHDH) for 30 min to fully mix, then a fully mixed sample was obtained and stored in a glass desiccator. The two random samplings of the mixture were consistent, which ensured the uniformity of the sample. Specifically, the mass proportions of SCCB in the mixture were 0, 1, 3, 5, and 100% respectively, and the prepared sample was named SCCB-X (X represented the proportion of SCCB while 100-X represented the proportion of meager coal). Considering the corrosion of furnace in actual operating conditions, and the literature (Mikša et al, 2003) has also shown that the content of SCCB is controlled about 4.5% in the current collaborative treatment experiment of SCCB in rotary kiln. So the upper limit of mixing ratio of SCCB was set as 5%. The proximate analysis and ultimate analysis of raw materials are shown in Tables 1 and 2 respectively. Specifically, the content of moisture, volatile matter, ash, and fixed carbon was determined in accordance with American Society of Testing Materials (ASTM) standards. The ultimate analysis of samples was carried out with a CHNS analyzer. The higher heating value (HHV) was evaluated by a digital bomb calorimeter.

2.2 Combustion experiment and characterization

5 ± 0.1 mg of the sample was put into a platinum crucible in an air atmosphere for the combustion experiment. The reaction temperature was raised from 40 to 1000 °C, the flow rate of air was 50 mL/min, and five different heating rates (10, 20, 30, 40, and 50 °C/min) were conducted in this study. During the heating process, the thermogravimetric analyzer (TG/DTG, Swiss, Mettler-Toledo TGA/DSC1) was used to record the mass loss of the experimental sample with temperature in real-time to obtain the TG curve (shows the mass loss) and DTG curve (illustrates mass loss rate). The above two curves have been used to evaluate the combustion parameters that reflect the thermal behavior (Kök, 2005), such as ignition temperature ( $T_i$ ), burn-out temperature ( $T_f$ ), maximum combustion rate ( $V_p$ ), maximum combustion rate temperature ( $T_p$ ), average combustion rate ( $V_{\mathit{mean}}$ ). And the comprehensive combustion index ( $S$ ) reflecting the combustion performance and flammability of the fuel and the combustion stability index ( $\mathit{CSI}$ ) reflecting the stability of the combustion process can be calculated by Eq. (1) and Eq.(2), respectively (Li et al., 2019a; Wang et al., 2020c).

In addition, a high-resolution tandem mass spectrometer (TG-MS, Germany, NETZSCH STA449C-QMS403) was performed to collect spectrum of different samples at the same time during the thermogravimetric test at a heating rate of 20 °C/min. The scanning electron microscope (SEM, Japan, JEM-2010) equipped with an energy dispersive X-ray spectrometer (EDS, USA, EDAX Appllo XL) was used to observe the morphological characteristics of molten ash. The Fourier transform infrared spectrometer (FTIR, USA, Nicolet iS50) was conducted to determine the surface properties of meager coal and SCCB, respectively. All experiments were carried out at least twice under the same conditions to ensure the repeatability and accuracy of the experimental results.

2.3 Interactive calculation and analysis

The deviation between theoretical TG and SCCB-X combustion test values was used to evaluate the synergy between the two raw materials (Jayaraman et al., 2017; Kök, 2007; Wang et al., 2020b). Assuming that meager coal and SCCB have no relevance during combustion, Eq. (3) is used to calculate the theoretical value. The deviation value (%) is calculated by Eq. (4) and used to evaluate the degree of interaction (Huang et al., 2017; Skodras et al., 2007).

Specifically, $Y_1$ and $Y_2$ represent the true value of TG of meager coal and SCCB respectively, x₁ and x₂ are corresponding mass fraction, $Y_{\mathit{theo}}$ represents the theoretical TG value, and $Y_{exp}$ represents the real mass loss.

2.4 Kinetics models

Generally, the calculation of various kinetic parameters is based on Arrhenius law (Coats and Redfern, 1964). This method has been widely used to solve the kinetic parameters under combustion conditions. The reaction kinetic equation can be expressed as:

Where $\alpha$ means reaction conversion rate, %; t means time, s; T means thermodynamic temperature, K; $k\left(T\right)$ means chemical reaction rate constant; $f\left(\alpha \right)$ means depend on the mechanical function of the fuel combustion process.

According to the Arrhenius equation:

Substitute Eq. (7) into Eq. (5) to get:

When the heating rate remains $\beta \left(\beta =\frac{\mathit{dT}}{\mathit{dt}}\right)$ unchanged, Eq. (8) can be transformed into an equation containing:

Integrate both sides of Eq. (9) at the same time, and get:

The DAEM and OFW models were used to solve the kinetic parameters in this paper.

The DAEM model equation is as follow (Miura and Maki, 1998):

The OFW model equation is described in Eq. (12) (Flynn and Wall,1966; Ozawa, 1965):

Under different conversion rates, $ln\left(\frac{\beta }{T^2}\right)$ in Eq. (11) and $ln\beta$ in Eq. (12) are both proportional to $\frac{1}{T}$ . A straight line can be obtained by fitting, and the activation energy at the corresponding conversion rate can be obtained from the slope of the straight line.

3.1 Analysis of physicochemical properties

As shown in the proximate and ultimate analysis of Tables 1 and 2, although SCCB's HHV is lower than that of meager coal, it still maintains a high value (16.73 MJ kg⁻¹), so it is a potential fuel candidate. Its low volatility and high ash content make it harder to burn than meager coal. The higher content of C and H in meager coal determines its higher HHV, and higher content of O in SCCB shows better thermochemical reaction performance. In addition, SCCB contains excessive amounts of F, and the emission of pollutants in the combustion process is worthy of attention. Therefore, the SCCB ratio should be kept at a low value during mixed combustion.

In order to further study the surface chemical properties of the raw materials, FTIR tests were carried out. As shown in Fig. 1, both SCCB-0 and SCCB-100 have the tensile vibration of hydroxyl in the range of 3710–3610 cm⁻¹, and the former has higher intensity, which indicates that meager coal is rich in alcohols and phenols. In addition, the presence of primary amine groups on the peak surface near 3500–3400 cm⁻¹ in SCCB, while the meager coal around 3000–2850 cm⁻¹ corresponds to the aliphatic CH stretched to alkanes, which means that the coal riches in alkanes, which is consistent with the data in Table 1. Meanwhile, meager coal contains a small amount of olefins (1630–1590 cm⁻¹), the strong absorption peak of meager coal at around ∼1000 cm⁻¹ is caused by Si-O-Si vibration. While the strong peak of SCCB at 1020 cm⁻¹ is attributed to the tensile vibration of oxygen-containing groups (C-O-C), which means that SCCB has higher chemical activity. The peaks at 900–700 cm⁻¹ of the two are caused by the deformation vibration of CH outside the aromatic ring (Jiang et al., 2019a; Wang et al., 2020a). In addition, SCCB-0 and SCCB-5 have absorption peaks at ∼500 cm⁻¹, 578 cm⁻¹ due to the torsional vibration of Si-O.

Close

Fig. 1 FTIR spectrum of the samples. (a) SCCB-0; (b) SCCB-100.

Export to PPT

3.2 Combustion characteristics analysis

3.2.1

3.2.1 Combustion parameter analysis

SCCB-0, SCCB-5 and SCCB-100 were put at a heating rate of 30 °C/min for the combustion experiment, and TG-DTG curves of the combustion process are shown in Fig. 2. The entire combustion processes of three samples are basically similar and can be divided into three stages: dehydration and drying, combustion of volatile matter and carbon, decomposition of residue. The specific combustion characteristics of different samples are shown in Table 3. When the temperature is lower than 300 °C, the mass losses of all samples are small due to the evaporation of water, and the main mass losses of SCCB-0 and SCCB-100 occur at 540–720 and 570–790 °C, respectively (Table S1). Due to the higher volatile content, SCCB-0 has the lower $T_i$ (548.76 °C), and SCCB-100 also begins to burn at 574.56 °C. As a result of a low mixing ratio, and the $T_i$ deviation of SCCB-5 is within 3.8 °C, this fully demonstrates the rationality of this ratio. The residual amount of ash of SCCB-0 is 22.52%, and which of SCCB-100 is 54.20%. The result is consistent with the corresponding ash content in Table 1. At the same time, the $T_f$ of SCCB-100 (790.92 °C) is much higher than that of SCCB-0 (718.65 °C), which shows that ash can inhibit the combustion process. In addition, the $V_p$ of SCCB-5 (3.38%/min) is higher than that of SCCB-0 (3.17%/min), which fully proves the combustion-supporting effect of oxygen groups (Mureddu et al., 2018; Wang et al., 2020a).

Close

Fig. 2 TG-DTG curve of meager coal, SCCB and their mixtures at a heating rate of 30 °C/min. a. SCCB-0, b. SCCB-5, c. SCCB-100.

Export to PPT

Table 3 Characteristic parameters obtained for samples with different mixing ratios at a heating rate of 50 °C/min.

Sample	Vp/% min−1	Vmean/% min−1	Tp/°C	Ti/°C	Tf/°C	S/×10−7	CSI/×10−7
SCCB-1	1.26	11.81	654.58	558.25	819.41	0.58	86.42
SCCB-3	1.38	12.86	655.97	561.19	798.21	0.71	103.75
SCCB-5	1.53	13.50	654.11	556.21	780.37	0.86	122.71

Fig. 2 shows that all samples have only one obvious mass loss peak in the whole temperature range (The TG-DTG curves of SCCB-1 and SCCB-3 are listed in Fig. S1). This shows that in the three samples, there is no clear boundary between the combustion of volatile matter and fixed carbon, which is consistent with the literature report (Niu et al., 2009; Wang et al.,2016). And then suitable combustion conditions and how to simultaneously exert the combustion characteristics of meager coal and SCCB are needed to be further discussed.

3.2.2

3.2.2 Effect of heating rate

Previous studies prove that the heating rate is one of the most important factors for the improvement of combustion property (Wang et al.,2016; Yang et al., 2014; Buratti et al., 2015). As shown in Fig. 3, with the increase of the heating rate, the temperature range experienced by the sample burning becomes wider, but the mass loss starting temperature, mass loss and residue rate of three samples change non-obviously under different heating rates. As the heating rate rises, all samples burn out at a higher temperature. Simultaneously, the increase in heating rate decreases the maximum mass loss rate, while the corresponding characteristic temperature rises. It should be noted that the SCCB enters the second stage at a higher temperature than the first two samples, which is determined by the physicochemical properties of the first part of the discussion. In addition, the temperature difference in the mass loss peak width also increases significantly. On the one hand, the rapid increase in temperature promotes molecular collisions and intensifies the combustion reaction. On the other hand, as the heating rate increases, both the ignition temperature and burnout temperature gradually increase as well (Fig.S2). The shortening of heat transfer time causes the temperature difference between the inner and outer surfaces of the sample particles to be too large (Zou et al., 2014), which is the phenomenon of thermal hysteresis (Yang et al., 2014; Fang et al.,2019; Gai et al., 2015; Sanchez-Silva et al., 2012). The meager coal has a higher volatile content, and the emission of the volatile gas produced creates favorable conditions for the subsequent combustion of fixed carbon. As a result, the increase of the heating rate affected both the gas diffusion and the chemical reaction rate, while indirectly affected the thermal decomposition process inside the meager coal and SCCB (Niu et al., 2009). Within a certain range, increasing the heating rate was beneficial to the combustion of meager coal and SCCB. In fact, this conclusion was consistent with the actual application of pulverized coal boilers.

Close

Fig. 3 TG-DTG curves of meager coal, SCCB and their mixtures at different heating rates. (a ∼ b) SCCB-0;(c ∼ d).SCCB-5;(e ∼ f)SCCB-100.

Export to PPT

3.2.3

3.2.3 Effect of mixing ratio

In order to analyze the co-combustion characteristics more clearly, the influence of different mixing ratios was studied at a heating temperature of 50 °C/min. Considering the actual tolerance of the pulverized coal boiler, three different mixing ratios (1, 3, 5%) were selected. Fig. 4 indicates that the increase in the mixing ratio of SCCB has a certain impact on the co-burning process. With the increase of the mixing ratio of SCCB, the mass loss ratio decreases (Fig.S3) while the maximum and average mass loss rate of sample combustion show a increasing trend. At this heating rate, as the mixing ratio increases, the ignition temperature gradually boosts, but the average and maximum combustion rates gradually increase as well. When the proportion of SCCB is 5%, the combustion characteristic index achieves the highest value (0.86 × 10⁻⁷) and the combustion process is the most stable (Table 3), which shows that this suitable mixing ratio achieves a balance of combustion characteristics and reactivity, and also proves the practical feasibility of mixing treatment. Related interaction investigation of co-combustion will be discussed in depth below.

Close

Fig. 4 TG-DTG curves of meager coal, SCCB and their mixtures at a heating rate of 50 °C/min. a. TG curves; b. DTG curves.

Export to PPT

3.2.4

3.2.4 Analysis of combustion gas release characteristics

During the thermogravimetric test, the gas products released by the combustion of meager coal, SCCB and mixed samples were recorded by a mass spectrometer using quadruple mass spectrometry. It is generally believed that gaseous products are mainly composed of light volatiles, such as H₂ and H₂O, carbon oxides (CO and CO₂), hydrocarbons (CH₄), alcohols (CH₂OH), nitrogen compounds (NO or NO₂), aromatic compounds (C₆H₆), sulfur Compound (COS) composition and SO₂), fluoride (HF), etc (Jayaraman et al., 2020). Fig. 5 shows the 3D spectrum of the gas released with temperature changes, the monitoring molecular mass range is 10–58, and the temperature range is 40 1000℃. The main observations are the initial release of CH₄, the CO produced by incomplete combustion, and the later release of CO₂.

Close

Fig. 5 3D mass spectra of SCCB-0 and SCCB-100.

Export to PPT

As shown in Fig. 6, the pre-combustion stage is the precipitation of moisture and the volatilization of light alkanes, a small amount of sulfur compounds, especially COS, will be released during the decomposition stage, and SO₂ will be produced in the subsequent combustion stage. The latter can be attributed to the fact that part of the sulfur remaining in the coke may be oxidized into gas. In addition, the production of light condensable gases such as alcohols (CH₂OH) and benzene aromatics (C₆ H₆) is much lower. The gaseous pollutant release of SCCB-0 mainly occurs at 300–700 °C, while the main pollutant release temperature of SCCB-100 is 500–800 °C. The release temperature range of gaseous pollutants in SCCB-5 is similar to that in SCCB-0, which is consistent with the precipitation of volatile substances and the combustion process of carbon. More importantly, the volatilization intensity of fluoride in SCCB-100 is not as high as that of SCCB-0, which is because the fluorine in the SCCB existed in the solid form of NaF, Na₃AlO₃, CaF₂, AlF₃ (Li et al., 2020), and their boiling points are higher than 1000 °C. This also shows that the co-processing temperature of meager coal and SCCB is higher than 1000 °C, and the use of the pulverized coal boiler is reasonable.

Close

Fig. 6 Mass spectrum of the gas release of three samples. (a ∼ b) SCCB-0;(c ∼ d) SCCB-5; (e ∼ f) SCCB-100.

Export to PPT

3.2.5

3.2.5 Interaction analysis

In order to deeply study the interaction between meager coal and SCCB in the combustion process, theoretical and experimental TG deviation calculations were carried out (Huang et al., 2017; Skodras et al., 2007). As shown in Fig. 7, in the initial stage of combustion, the deviations of the three samples are all negative, indicating that SCCB inhibits the escape of meager coal volatiles during mixed combustion, and its internal pore structure has a certain degree of adsorption for organic gases. With the increase of temperature, the deviation value of SCCB-1 is still negative, while as the mixing ratio increases, the interaction gradually becomes positive. In the high-temperature zone (>500 °C), the porous structure of SCCB and the trapped volatiles promote the accumulation and gradual transfer of heat. Meanwhile, the oxygen-containing groups improve the chemical reactivity, so the combustion effect is better when the mixing ratio is relatively large. In the final stage of combustion (>900 °C), the interaction is negative when the blending ratio is 5%, indicating that the higher ash content in SCCB affects the burnout of the mixture. In addition, the deviation values of SCCB-3 and SCCB-5 show similar trends, and both reach the maximum value near 800 °C, which means that the proper blending ratio promotes co-combustion, this is consistent with the above conclusion.

Close

Fig. 7 Interaction analysis. (a ∼ c) The experimental and calculated TG curves of the SCCB-1, SCCB-3, SCCB-5 samples. (d) Deviation curve of the SCCB-1, SCCB-3, SCCB-5 samples.

Export to PPT

3.3 Morphology analysis of ash

The morphology and composition of the ash after combustion shown in Fig. 8 are both importmant indicators used to reflect the combustion performance of the fuel (Altun et al, 2002). The ash of SCCB-0 has a flaky layered structure, and which of SCCB-100 is a granular structure. The regular flaky structure in Fig. 8a is caused by the high volatile content and low ash content of SCCB-0, and it also implies that meager coal is an amorphous structure with defective benzene ring as the unit. For SCCB-100 (Fig. 8b), the residue contains a certain amount of fluoride, which indicates that part of the F element is converted into HF after combustion and the rest remains in the residue as solid fluoride. Due to the highly ordered nature of SCCB, the slagging is severe, and this situation has been significantly improved after mixed combustion (Fig. 8c). The microporous structure of meager coal can partially concentrate heat to promote the ignition of SCCB, while the higher O content in the appropriate ratio of SCCB promotes the reactivity of co-combustion. The C content of SCCB-5 after combustion is greatly reduced than that of meager coal, which confirms that SCCB contributes to combustion of meager coal. At the same time, the content of Na, Al, Si, S and other elements is also approximative or reduced (Table S2), which means that the existing pulverized coal boiler technology can realize the disposal of solid residue. In summary, the above results once again prove the feasibility of co-combustion.

Close

Fig. 8 SEM images and EDS results of the burnt ash of three samples. a. SCCB-0; b. SCCB-100; c. SCCB-5.

Export to PPT

3.4 Kinetics analysis

Activation energy is the minimum energy to induce a molecule to become an activated molecule. The smaller the activation energy, the easier it is to become an activated molecule. Kinetics was measured using both OFW and DAEM methods. Accordingly (Chen et al., 2013; Kök, 2005; Wang et al., 2020c; Yang et al., 2014), all conversion ranges (0.2 ≤ α ≤ 0.8) have been used to investigate the effect of the different oxygen concentrations on the kinetic parameters for all the considered fuels, determining the variation of activation energy with the conversion.

Based on the DAEM model, when $\alpha$ is constant, the values of $ln\left(\frac{\beta }{T^2}\right)$ versus $\frac{1}{T}$ , obtained at different $\alpha$ values, have been correlated by a straight line, whose slope is associated with the activation energy (Fig. 9, Fig.S4).

Close

Fig. 9 Activation energy calculation in combustion modes by DAEM method. a. SCCB-0; b. SCCB-5; c. SCCB-100.

Export to PPT

According to OFW model, when $\alpha$ is constant, the values of $ln\beta$ versus $\frac{1}{T}$ , obtained at different α values, have been correlated by a straight line, whose slope is associated with the activation energy (Fig. 10, Fig.S5).

Close

Fig. 10 Activation energy calculation in combustion modes by OFW method. a. SCCB-0; b. SCCB-5; c. SCCB-100.

Export to PPT

It can be seen from Tables 4 and S3, except for SCCB-1 (0.6295–0.8319), the regression coefficient (R²) of other samples in the DAEM model is between 0.8811 and 0.9671, indicating that the accuracy of the results is within the acceptable range. Similar to the former, except for SCCB-1 (0.7888–0.8774), the R² of other samples in OFW model is acceptable, ranging from 0.8791 and 0.9826. For mixed samples, it is difficult to achieve uniformity when the mixing ratio is relatively low. This also shows that a suitable mixing ratio is the key to ensuring stable operation in the combustion process.

Taking meager coal as an example, when the conversion rate is 0.2, the activation energy calculated by the DAEM method is 40.07 kJ mol⁻¹. As the conversion rate increases, the corresponding activation energy decreases gradually (when the conversion rate is 0.8, the activation energy is 24.53 kJ mol⁻¹), indicating that the combustion of meager coal is complicated, which is consistent with the change trend of kinetic data obtained by studying bituminous coal and lignite (Zhou et al., 2012; Yang et al., 2014). SCCB has similar characteristics to meager coal. The activation energy of the mixed sample with relatively low SCCB content has little change, and its effect on the combustion process and the thermal efficiency of the co-fired boiler is weak compared with meager coal. In addition, under the same conversion rate, the activation energy calculated by OFW method is higher than that by DAEM method. The unexpected result of the higher activation energy in the different samples can be reasonably attributed to the energy required for the beginning of volatile degradation. These findings further support the idea that no simple correlation exists between the activation energy and volatile matter. For all coal and SCCB samples, the significant decrease of the activation energy could be associated with an intense burning of volatile matter. The activation energy of the SCCB calculated by the equal conversion rate method is greater than that of the meager coal, which is because the SCCB had a higher degree of graphitization making it more difficult to burn out. This is consistent with the results shown in Fig. 2 and 8.

3.5 Economic comparison

The economic comparison among this work, published in literature and common disposal methods is listed in Table 5. At present, the common disposal method of SCCB is landfill or incineration after pretreatment by relevant enterprises (Li et al., 2019b), and the cost of this method is generally $300–400 per ton. There are also companies that use hydrogen or pyrolysis to dispose of SCCB, which costs $100–200 per ton. At the same time, it has also been reported (Zhao et al., 2021) that methods such as batch high temperature treatment can achieve a profit of $162.50 per ton due to produced fluorides. The method mentioned in this work can save 0.75 tons of meager coal for disposal of each ton SCCB and at the same time generate 2400 kWh of electricity. The only cost happens in the crushing stage, it is expected to achieve a positive income of $ 269.7–279.7, which has obvious advantages compared with other methods.