3.1. Stage 1—Laboratory-Scale Research
According to the available literature data, it is possible to monitor the effectiveness of the pretreatment of organic substrates prior to the AD process by evaluating the changes in the concentration of organic matter in the dissolved phase [
69]. It has been shown that the concentrations of dissolved COD and TOC often correlate strongly with the final technological effects of AD, namely the efficiency of biogas and CH
4 production [
70]. Based on these indicators, optimisation procedures and empirical equations have also been developed that allow reliable estimation of AD energy efficiency after the initial pre-treatment of substrates with different properties and origins [
71]. An example of this type of research is the use of solidified CO
2 in the disintegration process of aerobic granular sludge (AGS) prior to AD, where a successive increase in COD in the supernatant was observed with increasing doses of solidified CO
2 from 0.1 to 0.5 in the supernatant from 334±15 to 450 ±13 mgO
2/L. For raw sludge, the COD concentration in the overlying liquid was 152±14 mgO
2/L [
33]. Other studies focused on the use of ultrasonic disintegration for the pretreatment of excessive sewage sludge [
72]. They showed that the COD concentration in the supernatant reached 4537.5 mg/L after 30 minutes of ultrasonic exposure. In the raw sludge, the COD content in the overlying liquid was 227.5±27.5 mg/L [
72]. Park et. al (2013) [
73], on the other hand, used ultrasonic disintegration to process microalgal biomass. The concentration of dissolved COD increased with the amount of energy applied. Energy doses of 50, 100 and 200 J/mL were used, resulting in COD concentrations of around 1250, 1350 and 1500 mg/L, respectively. With non-disintegrated biomass, the concentration of dissolved COD was around 750 mg/L [
73].
Laboratory-scale experiments (S1) also documented the effects of the HC process on the changes in organic matter content characterised by COD and TOC in the dissolved phase. This process was evaluated by monitoring the concentrations of these indicators and by assessing the degree of solubilisation, i.e. the efficiency of the transfer of organic matter from the solid phase to the solution, expressed as a percentage. For the raw substrate that was not subjected to the HC process (V0), the COD and TOC contents were 9.31±1.02 gO2/L and 2.91±0.18 g/L respectively. In the variants V1 - V3, where the pre-treatment time was between 2 and 6 minutes, there was no significant effect of HC on the changes in the concentration of dissolved COD. The COD concentrations ranged from 9.97±0.94 gO2/L (V1) to 11.98±1.02 gO2/L (V3), with the degree of solubilisation increasing significantly from 3.7±0.4% (V1) to 15.2±1.2% (V3). Monitoring of the TOC content showed that the use of 6 and 8 minutes HC ensured a comparable, significant increase of this indicator in the dissolved phase to values of 4.08±0.21 g/L (V3) and 4.37±0.20 g/L (V3). L (V4) and the degree of solubilisation 19.0±1.2% and 23.7±1.7%.
Figure 7.
Changes in the concentrations of dissolved COD and TOC (A) and the degree of solubilization of these indicators (B) as a function of the experimental variant in S1.
Figure 7.
Changes in the concentrations of dissolved COD and TOC (A) and the degree of solubilization of these indicators (B) as a function of the experimental variant in S1.
In V5, a significant increase in the concentration and solubility of both COD and TOC was observed after 10 minutes of HC. The COD content was 16.36±1.34 gO
2/L at a solubility level of 40.0±2.3%, while the TOC content was 5.13±0.29 g/L with 36.0±2.1% solubilisation. The increase in HC time had no significant effect on the observed concentrations of organic compounds in the dissolved phase. For COD, the range was between 17.02±2.07 gO
2/L and 17.93±1.74 gO
2/L and for TOC between 5.52±0.14 g/L and 6.11±0.48g/L. The degree of solubilisation in V7 for COD increased to 48.9±2.6%, while for TOC it was 51.9±4.4%. Lower results were obtained by Mancuso et. al (2021) [
43], who investigated the influence of the HC time of the sewage sludge on the concentration of dissolved COD and the degree of solubilisation. The applied HC times of 0, 1, 2, 4 and 8 hours resulted in dissolved COD concentrations of 244, 992, 1719, 2693 and 4578 mg/L and corresponding degrees of solubilisation of 0%, 1.7%, 6.5%, 10.8%, 19.2% [
43]. An increase in the degree of solubilisation as a result of pretreatment was also achieved by Lee and Han (2013) [
74] by disintegrating activated sewage sludge with HC. After 1 minute, a degree of solubilisation of 11.52% was achieved. After 5 minutes it was about 14%, after 10 minutes about 16.5% and after 15 minutes about 21%. After 20 minutes, however, it reached a value of 23.67% [
74]. Other trends were found in other studies by Mancuso et al. (2019) [
75], in which the use of 1 hour of HC WAS at a pressure of 2 bar resulted in a degree of dissolution of only 1.4% and an increase in dissolved COD from 228 to 617 mg/L. However, higher decay values can easily be achieved by changing the HC operating conditions such as the inlet pressure and the HC treatment time [
76]. This was confirmed in the study by Langone et al. (2018) [
64], in which one hour of HC treatment of cattle slurry led to an increase in COD in the liquid by 2040 mg/L at an inlet pressure of 6.0. A further increase in inlet pressure led to a further increase in COD values of 3340 and 5500 mg/L at inlet pressures of 7.0 and 8.0 bar, respectively. The degree of disintegration increased by 5.8, 8.9 and 15.8% after HC treatment at pressures of 6.0, 7.0 and 8.0 bar, respectively [
64].
In HC, cavitation bubbles grow on a solid surface (cavitation scoops) and then collapse under the influence of shear flows [
21]. A low energy input and a short pre-treatment time only promote the deagglomeration of flocs and not the release of soluble material. Increasing the energy input promotes the disruption of the cell wall and a high solubilisation of the organic material. However, low energy levels lead to smaller particles, while higher energy input produces larger particles due to re-flocculation, resulting in structures that are resistant to anaerobic degradation [
77]. HC is mainly used to disrupt and destroy the structures of cell walls and membranes, e.g. in bacteria [
74] and microalgae [
78], less frequently to destroy complex structures of ligocellulosic biomass [
79]. The substrate used in the study, consisting of milking plant effluent and cattle slurry, was unusual and had not been treated with HC before. According to Szaja et al. (2022) [
80], the results of HC treatment can be influenced by many factors, such as the type of cavitation reactor, operating inlet pressures, cavitation number, temperature, pH and initial pollutant concentration. In addition, the physico-chemical properties of the liquid used, including its vapour pressure, viscosity and surface tension, play an important role [
81].
In the further part of the experimental work carried out in S1, respirometric tests were performed to evaluate the direct effects of pretreatment with HC on the effectiveness of the AD process. The focus was on determining the possible total yield of biogas and CH
4 and estimating the kinetics of the methane fermentation process. Respirometric testing is a common method of evaluating the effectiveness of the AD process. This is due to many advantages, including stable and repeatable conditions, thanks to which it is possible to reliably compare and evaluate the results obtained, the ability to simulate different process conditions, which allows easy optimisation of technological parameters, reliability as a measurement tool, ease of use and the possibility of evaluating the influence of inhibiting factors on biogas production [
82]. They are also frequently used to evaluate the efficiency of the pre-treatment process. An example of this is the work of Cydzik-Kwiatkowska et al. (2022) [
83], in which the effect of ultrasound on the AD of aerobic granulated sludge in 1 L glass bioreactors (OxiTop system) was investigated. After 0.5, 4.0 and 8.0 minutes of digestion, the biogas yield was 400 mL/gVS, 420 mL/gVS and 455 mL/gVS respectively. In the case of raw sludge, 375 mL/gVS of biogas was obtained [
83]. Another example is the work of Kim et al. (2023) [
84], which investigated the effects of alkaline thermal hydrolysis of cattle slurry on AD carried out in 250 mL serum bottles. The highest CH
4 production of 227.0±11.0 mL/gVS was measured at 160°C with 2% (dry weight) NaOH addition, compared to AD of cattle manure without disintegration, where 182.2±2.5 mL/gVS was observed [
84]. In other studies [
68], the influence of ultrasonic disintegration of granulated microalgae-bacteria sediment on AD performed in respirometric batch reactors (AMPTS II, BPC Instruments AB, Lund, Sweden) was determined. Studies have shown a positive effect of sonication on the kinetics of the anaerobic process and CH
4 production. The highest production was achieved in the variants where sonication lasted 150 to 200 s and was between 534±16 and 561±17 mL/gVS. CH
4 production from the non-disintegrated substrate was 329±20 mL/gVS [
68].
In the control experiment (V0), the biogas yield was 140±20 mL/gCOD with an observed AD rate (r) of 42.6±2.9 mL/day on average and a rate constant (k) of 0.30 1/day. The CH4 content was 62.4±1.4%, resulting in 87±10 mLCH4/gCOD. Similar AD efficiencies were observed in V1 and V2, with the HC time used ranging from 2 to 4 minutes. The production of biogas and CH4 in V1 was 142±23 mL/gCOD and 88±11 mLCH4/gCOD, while in V2 it was 161±16 mL/gCOD and 100±9 mLCH4/gCOD. A significant increase in the efficiency of the AD process and an increase in the values characterising the reaction kinetics were observed in V3. The amount of biogas obtained after 6 minutes of HC was 209±31 mL/gCOD and the amount of methane was 130±13 mLCH4/gCOD. The indicators characterising the kinetics of the biogas production process were r - 84.0±9.1 mL/day and k - 0.40 1/day.
The expansion of the pretreatment process with HC led to a successive increase in AD efficiency. The highest technological effects in terms of biogas and CH
4 production were found in V5, where the tested HC time was 10 minutes. In this variant, 376±14 mL/gCOD biogas (r - 183.5±4.3 mL/day, k - 0.50 1/day) and 233±5 mLCH
4/gCOD (r - 111.8±2.7 mL/day, k - 0.48 1/day) were achieved. Increasing the HC time in V6 and V7 had no significant effect on AD efficiency, which is characterised by the production of biogas, CH
4 and the observed process kinetics. The biogas quantities obtained were in a narrow range from 369±12 mL/gCOD to 362±21 mL/gCOD and CH
4 from 233±6 mLCH
4/gCOD to 228±9 mLCH
4/gCOD. The study confirmed the significant influence of the HC used on the increase in biogas production and the values of the indicators characterising the kinetics of the process; the pre-treatment process used had no significant influence on the CH
4 content in the biogas. The CH
4 concentration was within narrow limits from 62.0±1.9% in V3 to 63.6±3.2% in V4. Detailed data on the efficiency and kinetics of the AD process in the subsequently implemented variants of S1 are shown in
Table 3 and
Figure 8.
The research results of Lee and Han (2013) [
74] on the effects of the disintegration of activated sewage sludge using HC on AD also show the dependence of the cumulative biogas yield on the degree of sludge disintegration. By using HC, a biogas production of 51.28 mL/gCOD was achieved, with a CH
4 content of 56.99%. The sludge that did not undergo any pretreatment had a biogas production of 45.52 mL/gCOD and a CH
4 content of 54.07% [
74]. Fardinpoor et al. (2022) [
85] investigated the effect of HC on the AD of the cyanobacteria
D. tharense. It was shown that it is possible to obtain 290.6 mLCH
4/gVS, which meant an increase in CH
4 production of up to 20.3%, as the CH
4 production of untreated
D. tharense was 241.5 mLCH
4/gVS [
85].
In the study by Langone et al. (2018) [
64], HC was used to disintegrate cattle slurry prior to AD. At an inlet pressure of 6.0, the biogas production was 476 mL/gVS and the CH
4 concentration was 62.6%. A further increase in inlet pressure led to an increase in biogas yield to 486 mL/gVS and 491 mL/gVS mg/L at inlet pressures of 7.0 and 8.0 bar, respectively. Conversely, the CH
4 concentration at a pressure of 7.0 was 62.7%. However, at a pressure of 8.0, a decrease in CH
4 content to 61.4% was observed [
64]. The reduced CH
4 production after pretreatment, which was also observed in this study at V7, could be related to the high concentration of bioavailable substances that can negatively affect the methanisation process [
86]. This phenomenon may also be related to the removal of organic material, leading to a net decrease in organic material available for CH
4 production [
87] and the formation of resistant/toxic compounds [
88]. In particular, inhibitors can be formed either by the pretreatment of biomass, which produces soluble phenolic compounds and soluble hemicellulose compounds such as furfural and hydroxymethylfurfural (HMF) [
87], or by the degradation of carbohydrates and amino acids into melanoidins that are difficult or impossible to degradation, the so-called Maillard reaction [
64].
The estimated energy balance based on the laboratory test results, presented in
Table 4, showed that the highest similar value of gross energy obtained from biogas was found in V5 and V6, 5.33 Wh and 5.34 Wh respectively. For the control variant V0 without pre-treatment, the value was only 2.00 Wh. After taking into account the energy requirement for the HC process, it was found that none of the technological variants tested had a higher net energy gain than V0, which was 2.0 Wh. The highest net energy value among the variants with HC was found in V5, where the pre-treatment process lasted 10 minutes. It was 1.65 Wh. In the other cases, the net energy yield ranged from 0.07 Wh (V7) to 1.28 (V1). Considering that the energy balance for AD research conducted in lab-scale batch reactors provides very estimated and approximate data. In the next stage (S2), further optimisation work was planned on a technical scale, with the initial pre-treatment time being that used in S1V5.
Fardinpoor et al. (2022) [
85] also determined the energy efficiency of the HC process in combination with AD. The energy production (CH
4) from raw
D. tharense via AD was 9 MJ/kgTS, the energy recovery was 41.4%. The use of HC increased the energy production to 10.8 MJ/kgTS. This shows that an energy recovery of 49.8% can be achieved under the conditions of maximum CH
4 production [
85]. However, it may not be possible to produce a large amount of CH
4 from all strains because their internal properties, such as composition and cell wall, are completely different [
89]. The energy consumption of the HC pretreatment process was measured to be 7.24 KWh/kg
D. tharense for maximum CH
4 production. The sum of the energy consumption for microalgae, AD and HC production was 8.5 KWh/kg cyanobacteria. The energy assessment showed that the energy gain did not compensate for the energy input required for HC pretreatment [
85]. The energy efficiency of the HC process was also the subject of research by Langone et al. (2018) [
64], who disintegrated cattle slurry prior to AD. When the inlet pressure was increased from 6.0 to 8.0 bar, the energy consumption in the HC system increased from 1905 to 2872 kJ/kgVS, corresponding to 0.53 and 0.80 kWh/kgVS, respectively. All experiments with HC treatment showed a positive net energy content. However, at pressures of 6.0, 7.0 and 8.0 bar, the net energy content of HC decreased by 17, 20 and 25% compared to the untreated sample, reaching values of 2.43, 2.36 and 2.20 kWh/kgVS, respectively. This is due to the fact that the energy consumed for HC treatment was greater than the energy generated by the additional CH
4 volume due to slurry dissolution. Nevertheless, the energy balance of this study did not take into account the benefits of HC treatment, which could lead to a more favourable energy balance [
63]. HC treatment improves the rheological properties of the substrate, reducing the energy consumption for mixing and pumping in the AD process and allowing preheating of the substrates entering the digester. In addition, many factors, including the low environmental impact of the HC process, the efficiency of nitrogen removal and the reduction of pumping problems and mixer failures, have led to the HC system being recognised as a valuable pretreatment system [
90].
The experimental work carried out in S1 confirmed the fact, repeatedly confirmed in the literature, that there are relationships between the concentration of dissolved organic matter in the substrate and AD efficiency [
91,
92]. This is confirmed by the work of Lu et al. (2018) [
91], who used alkaline-ultrasound pretreatment of WAS in AD. Linear Pearson correlations confirmed the relationship between the absorbed dose and the concentration of dissolved organic matter [
91]. Other studies by the authors concerned the use of ultrasonic disintegration for the pretreatment of acid whey [
92]. The results of the correlation analysis of the duration of the UD with the concentration of dissolved organic compounds showed strong, statistically significant correlations for all tested technological variants, which differed in the absorbed dose applied to the substrate. The coefficients of determination calculated for the correlations UD/COD and UD/TOC reached R
2 = 0.8908 and R
2 = 0.7851 respectively [
92].
In our own investigations, which we carried out in S1, both the increase in the COD and TOC concentration in the dissolved phase and the degree of solubilisation were strongly positively correlated with the absorbed dose of HC introduced into the pretreated substrate (
Figure 9). The coefficient of determination R
2 was 0.9723 for the COD concentration and 0.9824 for the TOC. An even stronger correlation was found for the relationship between the degree of COD and TOC dissolution and the absorbed dose of HC introduced. The R
2 values were 0.9855 and 0.9909 respectively.
High values of the coefficient of determination R
2 were also characteristic for the correlations between the tested HC variant and the efficiency of biogas and CH
4 production and amounted to 0.9109 and 0.9054, respectively (
Figure 10A). Strong positive correlations regarding the effect of HC on the production of gaseous metabolites of anaerobic bacteria had a direct influence on the gross energy obtained (
Figure 10B). In this case, the strong relationship between HC time and gross energy value was characterised by the R
2 coefficient of 0.9054. However, no strong relationship was found between the absorbed energy dose introduced into the substrate and the percentage of CH
4 in the biogas (
Figure 10A). Therefore, the R
2 value of the indicator was 0.4056. When investigating the effects of the HC variants used on the net energy obtained, a weak negative coefficient was found in the laboratory tests, for which the R
2 value was 0.3801 (
Figure 10B).
Other researchers have also observed relationships between similar parameters. Park et al. (2013) [
73], who used ultrasonic disintegration for the treatment of
C. vulgaris biomass, found that cumulative CH
4 production showed a strong relationship with ultrasound dose. Ultrasound doses below 50 J/mL did not improve CH
4 production. However, at doses higher than 50 J/mL, gas production was proportional to energy input [
73]. In the study by Lu et al. (2018) [
91] on the influence of alkali-ultrasonic-based WAS pretreatment on AD, Pearson's linear correlations confirmed the relationships between dissolved organic matter and biogas production [
91].
In other studies [
33], very strong positive correlations were found between COD concentrations and the amount of biogas and CH
4 with increasing doses of solidified CO
2 in the range of 0.1 to 0.3 used for the conditioning of AGS. The coefficients of determination were R
2 = 0.8755 and R
2 = 0.8881, respectively. Very strong positive correlations were also found between the mass ratio of solidified CO
2 to AGS and the net energy gain, with a coefficient of determination R
2 of 0.9643 [
33]. In the study by Zupanc et al. (2023) [
66], which aimed to increase CH
4 production from WAS as a result of HC disintegration, Pearson correlations also confirmed positive or negative relationships between physicochemical changes and CH
4 production [
66]. In the authors' study [
92] on the use of ultrasonic disintegration for the pretreatment of acid whey. Very strong positive correlations were found between COD and TOC concentrations and CH
4 production efficiency, as shown by the coefficient of determination R
2 = 0.9995 and R
2 = 0.9746, respectively. A strong positive correlation was also observed between the duration of UD and gross energy gain (R
2 = 0.8986). However, a negative correlation was found between the duration of the UD and the increase in net energy with a coefficient of determination of R
2 = 0.6629, although this was not statistically significant [
92].