3.1 TGA results
Figure 1 shows the Thermogravimetric (TGA) and Derivative Thermogravimetric (DTG) curves for all three samples at three different heating rates of 5, 10, and 15°C/min. A very slow and initial weight loss of up to 180°C temperatures in PCB was observed because of the release of CO2, H2O, CH4, HBr (Lin & Chiang, 2014) and in the CS sample it was a result of moisture loss associated with depolymerization and vitrification transition (Leroy et al., 2010; Ranzi et al., 2008). The main pyrolysis process begins at around 180–200 ℃ temperature range and nearly at 577 ℃ temperature the devolatilization ceases to act for all three samples, which indicates the char generation and carbonization stage (Quan et al., 2013; Aboulkas & Harfi, 2008; Leroy et al., 2010; Chen et al., 2018). As can be seen in Table 1, CS has higher volatile material and lower ash content that favours the production of a large amount of pyrolysis oil upon condensation as it has an advantage of high reactivity which results in lesser yields of char and more condensable gases as oil. Moreover, CS tends to have high moisture content than PCB, the mixture will contribute to the reduction of water content in the pyrolysis oil, which consecutively helps to increase HHV.
It is also worth to be noted that the residue of PCB which is collected as char at the end of the process is relatively very high (50–60 %) while using PCB:CS (1:1) composition char was decreased to 16–27 %. Moreover, when compared to the degradation patterns of PCB and CS; the degradation profiles of mixed samples initiated at a lower temperature.
Table 1
Proximate and ultimate analysis of samples.
Sample
|
HCV Cal/gm
|
LOD (%)
|
Proximate Analysis
|
Ultimate Analysis
|
Ash (%)
|
Volatile Material (%)
|
Fixed C (%)
|
C (%)
|
H (%)
|
N (%)
|
S (%)
|
PCB
|
1415
|
14.05
|
43.95
|
41.87
|
0.13
|
26.73
|
2.39
|
1.93
|
0.19
|
CS
|
2319
|
23.93
|
5.49
|
70.51
|
0.07
|
32.52
|
4.12
|
2.15
|
0.16
|
PCB:CS (1:1)
|
2279
|
12
|
23.73
|
64.2
|
0.07
|
29.76
|
5.45
|
2.86
|
0.148
|
The appearance of the shoulder in the DTG curve may be due to the degradation of the hemicellulose and lignin in the CS sample and attributed to the decomposition of tetra-bromo-bisphenol-A in the PCB sample (Quan et al., 2012; Kim et al., 2013; Chen et al., 2018). Whereas, a continuous slow degradation in CS and mixture may correspond to the slow decomposition of lignin (Mailto et al., 2018; Dhaundiyal et al., 2018; Mankhand et al., 2012).
For the PCB sample, this slow degradation is due to the rupture of ether bonds in brominated resin into bisphenol A, propyl alcohol, tetra-bromo-bisphenol-A, and small phenolic molecules (Islam et al., 2018). The total weight loss at the end of 700 ⁰C were 42–50, 94–95, 73–84 (wt %) for PCB, CS and PCB:CS (1:1), respectively. The total weight loss in CS was higher due to higher volatile matter and lower ash content, as shown in Table 1. From the same table, one can also observe that there is a higher H/C ratio for PCB:CS (0.183) in comparison to PCB (0.089) and CS (0.126). This could be the reason for the higher calorific value.
3.3 Pyrolysis experimental results
As per TGA analysis, the optimum temperature for producing maximum oil was in the range of 300–500°C. Above 500°C the operation reduces liquid and char production and provides an increased amount of gas, several studies have explored the same effect (Madhu et al., 2018; Mankhand et al., 2012). This reduced liquid and char at higher temperatures and could be the reason of the secondary reaction for the liquid fraction of the volatiles and further decomposition of the char particles. However, lowering the temperature below 300°C may cause incomplete decomposition of the biomass (Quan et al., 2012).
As the pyrolysis reaction progresses while increasing temperature, vapours generated were cooled with chilled water in a condenser and collected as oil. Non- condensable gases were collected as gas product and the residue remaining after the completion of pyrolysis reactions were received as char. The yield of these three products were calculated based on feed by weight. The density of oil from three different samples were found to be 1.2, 0.98 and 1.02 gm/ml for PCB, CS and PCB:CS (1:1), respectively. The yield of different components with reference to the % fraction of generated oil, gas, and char is shown in Fig. 2 (b). As demonstrated, the amount of oil was increased significantly with co-pyrolysis of PCB:CS (1:1) to 27.5 wt % while 19.6 wt % was obtained for PCB.
PCB generated oil is dark brown in colour compared to CS alone. As shown in Fig. 2 (a), light colour of CS could be the reason of higher water content, which leads to low energy density, corrosivity, and chemical instability, also high oxygen content makes it difficult to blend with fossil fuels. Whereas, blending it with PCB helps lower water content. A number of studies evidenced that a sufficient amount of water can have a positive effect to use it as fuel, like reduced viscosity, reduced pollutant emissions, which contributes to a micro explosion of droplets in combustion and increased oxidation (Lu et al., 2009; Zhang et al., 2007). Furthermore, solid particles in CS oil makes phase separation difficult due to the presence of lignin which makes solids adsorbed with it and generates gummy tars whereas, during the co-pyrolysis event, radical interaction can enhance the development of a stable pyrolysis oil that avoids phase separation (Lu et al., 2009; Oasmaa & Czernik, 1999; Oasmaa et al., 2005; Brebu & Spiridon, 2012).
3.4 Liquid analysis
The collected oil was analysed in GC − MS and the chromatograms for all three samples are presented in Fig. 4. Various hydrocarbons present in the oil of three samples are shown on peaks of the spectrum. In the first 20 minutes of GC/MS spectra acquisition, more than 25 distinct aromatic compounds were discovered in each sample. The components were identified with an increase in retention time. The compounds that had an area % of more than 0.1 are discussed here. The area % of the peak for a compound was correlated with the percentage of that compound in the oil. The liquid analysis performed in a GC-MS showed that the main products of the pyrolysis of PCB were fragments of polycarbonate epoxy resin. Table 2 shows all the detected compounds in PCB, CS, and PCB:CS (1:1) listed from higher to the lower area %.
A large difference in product distribution can be observed for PCB, CS and PCB:CS (1:1) samples for phenol and phenolic compounds like, phenol, 2-methyl,phenol, 4-methyl,Phenol, phenol,3,4-dimethyl-, phenol,3-(1-methylethyl), Phenol, 2-ethyl, Phenol, 3,4,5-trimethyl-, Phenol, m-tert-butyl-, phenol, 3-ethyl and P-isopropenylphenol.
Phenol present in PCB by area % was 38.47 which is in accordance with literature. A number of studies have examined PCB oil components and different results found for phenol which varies from 10–40% due to compositional and pyrolysis parameter difference (Hall & Williams, 2007b)(Ng et al., 2014), whereas (Quan et al., 2010) has pyrolyzed PCB up to 600 ℃ and found phenol as a most dominant compound (58.58%).
Figure 5 (a) reveals that there is an increase in phenol from 38.47 % to 54.83 % when co-pyrolysis was done with PCB. Other phenolic compounds were present in PCB, CS and PCB: CS (1:1) as 22.47, 4.09 and 21.99 % area, respectively, so total of 60.94, 25.94, 76.82 % area Phenol and phenolic compounds are present in PCB, CS and PCB:CS (1:1) which also shows highest in co-pyrolysis mixture. This result is true in accordance that the addition of polycarbonate in the pyrolysis of biomass can increase the phenol compounds in the oil where bisphenol A present in PCB is a poly carbonate which has enhanced the amount of phenol during co-pyrolysis with CS (Brebu & Spiridon, 2012). It's also worth noting that the high amount of phenolic compounds in CS, may be originated from the decomposition of lignin, and can help to enhance phenolic concentration in PCB:CS (1:1).
Furthermore, Naphthalene and other polycyclic aromatic hydrocarbons (PAHs) has the highest concentration in PCB, like Naphthalene, 2-methyl, anthracene, naphthalene, 2-ethenyl, acenaphthylene, fluorene, fluoranthene, Pyrene, 4H-Cyclopenta[def]phenanthrene with others listed in Table 2, having total 13.92 % area covered, with carbon ranges from C10-C18. In contrast, CS sample contains aliphatic hydrocarbons only, ranging from C3-C8.
Whereas, Co-pyrolysis includes Naphthalene as well naphthalene,2-ethenyl, acenaphthylene, fluorene, anthracene and fluoranthene as major PAHs having 3.64 % area of total. As overall PCB compounds include hydrocarbons ranging from C6 to C12, Whereas CS and the mixture includes components with C6-C18 and C6-C25, respectively, with the highest molecular weight 453. 5 of 3-Bromobenzoic acid, octadecyl ester (C25H41BrO2) in the mixture, but with a very small amount. Heat deterioration is the consequence of a number of competing breakdown events which shows complex behaviour of the reaction mechanism.
Additionally, Tetrabromobisphenol-A (TBBA) is the most widely used fire retardants in PCB, and degradation of this compound in presence of oxygen release brominated dibenzodioxines and furans, but as evidenced by the GC/MS results listed in Table 2, pyrolysis of PCB products do not contain PBDD/Fs because the de-hydrogen or de-hydroxyl radicals is difficult to happen in this reductive environment (W. Liu et al., 2013) also, at higher temperatures, brominated materials undergo debromination and additional deterioration, inhibiting the production of PBDD/Fs.
Furthermore, at the retention time of 22.53, triphenyl phosphate was observed which is a phosphorous based fire retardant. The phosphate based flame retardants are mostly found in thermoplastics, hence their presence in PCB pyrolysis oil is most likely attributable to the pyrolysis of plastic components attached to the printed circuit boards (Hall & Williams, 2007a). Some authors presented TBBA degradation mechanism and release of brominated compounds in their research (Grause et al., 2008; W. J. Liu et al., 2013; Kim et al., 2013). Moreover, Furans are highly toxic components and they are present in PCB oil in the form of Benzofuran, 2-isopropenyl-3-methyl-, Dibenzofuran, and 2(5H)-furanone, 3-hydroxy-4,5-dimethyl- (total of 1.44 % area) but their quantity decreased to 0.78 % area with co-pyrolysis, see Table 2. This is also evidenced by some researchers (W. J. Liu et al., 2013; W. Liu et al., 2013; Lu et al., 2009) that lower temperatures of pyrolysis, say up to 500 ℃, the amount of furans decreases. In actual with co-pyrolysis, amount of Benzofuran in oil sample is relatively decreased but could not be diminished totally. Overall, 98.25 % area of the PCB pyrolysis oil, 99.47 % area of the CS pyrolysis oil, and 96.17 % area of the PCB:CS (1:1) pyrolysis oil in total were identified.