2.2. Kinetic Analysis by Modified Gompertz Model
A modified Gompertz model was employed to simulate methane production during the HSAD process, and the principal kinetic parameters were investigated by regression analysis. The simulation results are summarized in
Table 3. The simulated maximum cumulative methane yield (A) was similar to the experimental yield, indicating that the cumulative methane yield curve fitted well with the Gompertz model. The λ values representing lag phase time were almost zero when evaluating the effect of adding T20, T60, and the control, revealing that the hydrolysis stage proceeded smoothly for these samples. In contrast, when PEG 300 was added into HSAD, the λ values increased with the augmentation of the dose of PEG 300, which was coincident with the decrease in cumulative biogas production during the first five days and confirmed the negative effect of PEG 300 on the initial stage of HSAD. In contrast, Singh et al. [
33] found that the lag phase of biohydrogen decreased from 7–12 h to 1–3 h with PEG addition. Because of the different substrates and disparate microbial processes, the effect of PEG could be inconsistent. The maximum methane production rates (U) were enhanced when adding T20, T60, and PEG 300 into HSAD compared with the control, indicating that adding these non-ionic surfactants accelerated the hydrolysis rate of the substrate and improved digestion. When T60 was added into the substrate, the U value was much higher at the T60/grass ratio of 0.20 than at the other three ratios, which supported the previous result that T60 at a high dose apparently improved HSAD, whereas T20 showed enhancement in HSAD even at a relatively low concentration. Moreover, the R
2 obtained from the control was 0.807, which was much lower than those obtained from the samples with T20 or T60 added. This was probably attributed to the limitation of mass transfer in HSAD, causing the inferior fitting by the Gompertz model, while adding T20 or T60 obviously improved the HSAD process and showed the ideal growth curve of methane production. The R
2 values of the samples after adding PEG 300 at the ratios of 0.04 and 0.08 were also poor because of the impeding effect of PEG 300 on the initial stage of AD, while the other two samples with PEG 300/grass ratios of 0.16 and 0.20 were 0.995 and 0.988, respectively. The effect of PEG 300 on HSAD was the balance of enhancement and hindrance, and the enhancement took the dominant place when the PEG 300/grass ratios were 0.16 or 0.20.
2.4. Improvement in the Mechanism of High-Solid Anaerobic Digestion by Adding Surfactants
The substrate’s structure was analyzed using FTIR to gather additional evidence supporting the enhancement of HSAD through the addition of surfactants. Therefore, the samples with the highest biogas production (i.e., all substrates mixed with PEG 300, T60, or T20 at the ratio of 0.20) were chosen to compare with the untreated grass. Here, it should be clear that the samples analyzed were mixed well with surfactants before anaerobic digestion.
Figure 5 shows the FTIR spectra reflecting the structural variation in the functional group. There was an obvious broad absorption band at 3250~3500 cm
−1. Some researchers ascribed the broad peak to the stretching vibration of the –OH group [
35], whereas it was better to consider the peak as the stretching vibration of the –OH group extended by hydrogen bonding between molecules. Because hydrogen bonding extensively existed among lignin, hemicelluloses, and cellulose, the hydrogen bonding showed a broad absorption band while the stretching vibration of the –OH group was a relatively sharp peak. Therefore, the peak in the range of 3250~3500 cm
−1 was the result of the stretching vibration of the –OH group and hydrogen bonding. The intensity of this peak slightly decreased after T20 and T60 were added into the substrate, indicating the intrusion of Tween into the lignocellulose structure, causing the weakness of hydrogen bonding, which further improved the separation of cellulose from lignin and provided more accessibility for the microbes. This may be part of the reason for the enhancement caused by T20 and T60 in the hydrolysis process in HSAD. Nevertheless, the intensity of this peak was strong when the substrate was added with PEG 300. It was reported that the hydrophobic parts of lignin such as phenyl and –CH
2 and –CH
3 groups will interact with the –CH
2 groups in PEG, and there were hydrogen bonds between phenolic hydroxyls and the ether oxygens in PEG [
36]. These interactions and hydrogen bonding reinforced the absorption peak in the range of 3250~3500 cm
−1, indicating the bonding of PEG 300 to lignin. The consolidated lignocellulose structure may have impeded the access of anaerobic microbes and thus delayed the hydrolysis process, causing the hindrance of PEG 300 in the initial stage of HSAD.
The absorption peaks at 2850 cm
−1 and 2920 cm
−1 predominantly arose from symmetric and asymmetric stretching of –CH bonds [
37,
38]. The absorption became stronger when PEG 300 was added compared with the untreated grass and those added with Tween, which provided further proof of the interaction between lignocelluloses and PEG 300. The peaks at 1450 cm
−1 and 1580 cm
−1 were attributed to phenyl ring skeletal vibrations of lignin macro molecules [
39]. The small peak that appeared at 1245 cm
−1 was due to the in-plane bending vibration of –OH, probably in cellulose. The absorption band at 1050 cm
−1 was the stretching vibration of C-O and C-O-C, indicating the existence of pyran ring skeleton vibration. The peaks in the range of 1050~1580 cm
−1 exhibited great differences when the substrate was added with PEG 300 in contrast to the untreated grass and those added with Tween. The differences confirmed the interaction between the hydrophobic parts of lignin and PEG [
36].
The FTIR gave deep insight into the interaction between the substrate and surfactants in the chemical structure aspect and made it much easier to explain the effect of surfactants such as T20, T60, and PEG 300 on the substrate. Based on the biogas production, properties of digestate, kinetic analysis, and IR spectrum during HSAD, the mechanism of HSAD improvement by adding surfactants became clear, which is illustrated in
Figure 6.
First, the interactions between the surfactants and lignin are somewhat different for PEG 300 and Tween. Some PEG 300 is tied to lignin owing to the interactions between the -CH
2 groups in PEG and the hydrophobic parts of lignin such as phenyl and the –CH
2 and –CH
3 groups [
36], which was proven by the FTIR spectrum. Contrary to PEG, Tween could promote the hydrolysis stage by modifying lignin surface properties, such as changing its hydrophobicity, hydrogen bonding ability, and surface charges [
19]; thus, the biodegradable fractions are easier to remove from the cages of lignocellulose, which was verified by the high ammonia nitrogen concentration in the digestate. Although PEG and Tween both contain the –(CH
2CH
2O)– group, the reason for the combination of lignin with PEG instead of with Tween probably is the steric hindrance that prohibits Tween from adsorbing onto lignin. As
Figure 6 shows, Tween intervenes in the middle of lignin and cellulose and weakens the interaction, especially the hydrogen bond between them, while PEG goes deep into the lignin structure and ties close to it. This consolidated structure likely hinders microbial access, resulting in the prolonged lag phase observed during digestion with PEG 300 addition. In contrast, the addition of Tween releases more biodegradable fractions, accelerating biogas production in the initial stage of HSAD.
Despite the hindrance caused by PEG 300 in the initial stage of HSAD, the overall biogas production was significantly increased by adding surfactants like PEG 300, T20, and T60. This is because these surfactants can easily penetrate the interior of lignocellulose, where cellulose is closely bound to hemicellulose and lignin. Surfactants help to extract the hydrophobic part from lignocelluloses by forming emulsions so that the biodegradable fraction is enhanced and becomes available to microbes [
29,
40]. Another effect of surfactants is that the hydrophobic molecules are surrounded by surfactants in the center of the micelle, which provides active sites for the reaction between the hydrophobic molecules and enzymes [
41]. In addition, adding surfactants reduces the surface tension and viscosity of the substrate [
42]; consequently, hydrophobic compounds move easily into microbe cells, and enzymes transfer well from cells to the solid substrate or liquid media.
2.5. Economic Assessment
An economic analysis was carried out to investigate the feasibility of adding surfactants such as Tween 20, Tween 60, and PEG 300 to improve HSAD performance. The analyzed scenarios adopted the ratio of surfactants/grass of 0.08, 0.16, and 0.20 for T20 and 0.20 for both T60 and PEG 300, at which the methane production was above 140 mL/g TS. The methane yield of grass without adding surfactant was 51.9 mL/g TS. Therefore, the extra methane produced per gram of grass after adding surfactants could be calculated. Assuming all the methane could be converted to electricity without heat loss, the profit was obtained by multiplying the methane yield by its calorific value and electricity price. The price of the additional methane was calculated based on previous reports, which indicate that the calorific value of methane is 11 kWh/Nm³ and the price of electrical energy is 0.15 EUR/kWh [
43]. The price of surfactant was obtained by interviewing a local company. Thus, the ratio of cost to profit was obtained, as shown in
Table 4. It is important to note that the ratio was merely a simplified index of economic analysis, omitting many real conditions such as fiscal subsidy, maintenance, depreciation of devices, transportation, construction cost, and so on.
As
Table 4 shows, the
of the scenario with T20 at 0.08 was 0.85, which was lower than 1.00, indicating that the practice of adding T20 at the ratio of 0.08 had the potential to make a profit. The other scenarios were above 1.00, resulting from the high adding ratio of surfactants, indicating the cost of chemical reagents was much higher than the profit generated by the increased biogas production. Based on these results, adding T60 or PEG 300 into HSAD was not economically feasible. However, it was still meaningful to add these surfactants to enhance HSAD, which provided an option for the exploration of improving mass transfer in HSAD. It also pointed toward a direction to find more effective and cheap surfactants to enhance HSAD.