Ultrasonic Disintegration to Improve Anaerobic Digestion of Microalgae with Hard Cell Walls—Scenedesmus sp. and Pinnularia sp.

Microalgae are considered to be very promising feedstocks for biomethane production. It has been shown that the structure of microalgal cell walls can be highly detrimental to the anaerobic digestibility of biomass. Therefore, there is a real need to seek ways to eliminate this problem. The aim of the present study was to assess the effect of ultrasonic disintegration of Scenedesmus sp. and Pinnularia sp. microalgal biomass on the performance and energy efficiency of anaerobic digestion. The pretreatment was successful in significantly increasing dissolved COD and TOC in the system. The highest CH4 yields were noted for Scenedesmus sp. sonicated for 150 s and 200 s, which produced 309 ± 13 cm3/gVS and 313 ± 15 cm3/gVS, respectively. The 50 s group performed the best in terms of net energy efficiency at 1.909 ± 0.20 Wh/gVS. Considerably poorer performance was noted for Pinnularia sp., with biomass yields and net energy gains peaking at CH4 250 ± 21 cm3/gVS and 0.943 ± 0.22 Wh/gVS, respectively. Notably, the latter value was inferior to even the non-pretreated biomass (which generated 1.394 ± 0.19 Wh/gVS).


Introduction
The current pollutant emission limits necessitate measures such as development and widespread deployment of renewable energy sources [1]. Development of biomass energy systems could do much to contribute to this goal [2,3]. Mismanagement of traditional energy crops can lead to a negative energy balance, reduced global food supply, and significant inflation of food prices. This necessitates the pursuit of competitive biomass sources [4,5]. From the standpoint of environmental policy, circular economy, and bioeconomy, the best strategy moving forward is to recycle and reuse municipal, agricultural, and industrial waste [6][7][8]. Another feasible alternative may be to harness microalgae, especially given their ability to rapidly build biomass, accumulate high-energy products in their cells, digest waste materials, and grow on land unsuitable for other purposes [9,10].
Microalgae-to-heat processes span a very wide range of methods: from simple combustion, through more advanced thermochemical methods (such as gasification and pyrolysis), to biodiesel generation and biotechnological ethanol/methane/biohydrogen harvesting [11]. Numerous studies have argued that anaerobic digestion (AD) is the most promising method for producing energy from algae, both in terms of performance and cost-effectiveness [12][13][14]. Algae-to-methane processes have been shown to be cost-effective and comparable to cellular lipid extraction in terms of harvested energy (in the form of biodiesel) [15]. In addition to high-energy biogas, the process also adds value in the form The aim of the study was to determine the applicability of ultrasounds for the disintegration of Scenedesmus sp. and Pinnularia sp. microalgal biomass prior to anaerobic digestion (AD), as well as to assess the impact of this pre-treatment on biogas productivity and composition. We verified the effect of the ultrasound dose on the anaerobic degradation process, biogas production rate, and the potential for achieving net energy gains.

Changes in Dissolved Organics
The effectiveness of pre-treatment methods is often evaluated by monitoring the changes in dissolved organics levels [56]. Various metrics can be used, depending on the organic feedstock used, including levels of volatile solids, genetic material, lipids, glucose and biodegradable organic compounds (expressed as biochemical oxygen demand (BOD)) [57,58]. Chemical oxygen demand (COD) and total organic carbon (TOC) are some of the most universal and commonly used parameters [59]. The AD process and its performance are partly predicated on ensuring efficient transfer of organic substances from the biomass to the dissolved phase [60].
Our study showed a significant increase in the levels of monitored dissolved organic compounds. S1 showed significant increases in COD and TOC in variants S1V1 to S1V3. S1V1 (Scenedesmus sp., no DU) had a COD of 64 ± 7 mgO 2 /dm 3 and TOC of 47 ± 6 mg/dm 3 ( Table 1). The values for S1V3 were 471 ± 19 mgO 2 /dm 3 and 388 ± 31 mg/dm 3 , respectively ( Table 1). The gains were less dramatic in the subsequent variants. In S1V5, the COD was 512 ± 31 mgO 2 /dm 3 , whereas TOC was 441 ± 35 mg/dm 3 (Table 1). UD exposure time was found to strongly and positively correlate with dissolved COD (R 2 = 0.8305) and TOC (R 2 = 0.8670) levels ( Figure 1a). S2 (Pinnularia sp.) experienced a consistent, statistically significant rise in levels of dissolved organic compounds across all of the tested UD intensities. COD increased from 59 ± 4 mgO 2 /dm 3 in S2V1 to 360 ± 12 mgO 2 /dm 3 in S2V5 (Table 1). TOC ranged from 50 ± 2 mg/dm 3 to 263 ± 18 mg/dm 3 ( Table 1). The concentrations of dissolved organic compounds were much lower than in S1. UD exposure time in S2 was found to correlate very strongly and positively with dissolved COD (R 2 = 0.9581) and TOC (R 2 = 0.9803) levels ( Figure 1b).  [61] have also found a significant increase in the levels of dissolved organic compounds after treating microalgal biomass (a mixture of Chlorella sp. and Scenedesmus sp.) with UD. The ultrasonic pretreatment was applied for 30, 90 and 180 s at 130 W. The initial soluble COD in the non-pretreated microalgal biomass was 770 mg/dm 3 , rising to 973 mg/dm 3 after 30 s disintegration, 1639 mg/dm 3 after 90 s, and 2282 mg/dm 3 after 180 s [61]. Similarly, Gruber-Brunhumer et al. (2015) [62] demonstrated increases in soluble COD after applying ultrasound pretreatment to Acutodesmus obliquus. Ultrasonication led to 538.0 ± 59.6 g/kgTS, compared to the 383.5 ± 4.4 g/kgTS in the non-pretreated control [62]. Tensions appear in the biomass subjected to UD, which cause numerous changes in the cell structure. They are the effect of ultrasonic pressure, forces related to the change in viscosity, the phenomenon of moving a biological object in the medium and the increase in temperature [63]. UDs cause twisting, rotation or spinning of macromolecules with asymmetric shapes [64]. These phenomena cause a change in the charge of the cell surface, a change in the permeability of the cell membrane, rupture, disintegration and fragmentation of the cell membrane [65]. This allows organic compounds to transfer to the dissolved phase and thus be more available to anaerobic bacteria in the AD process [66].

Biogas and Methane Production
The present study showed that UD had a positive effect on anaerobic digestion of microalgal biomass, providing significantly better biogas and methane production performance during digestion of Scenedesmus sp. Specific biogas yield in S1V1 was 371 ± 21 cm 3 /gVS with a CH 4 fraction of 49.2 ± 2.4%, which corresponds to CH 4 productivity of 183 ± 25 cm 3 /gVS. Significant improvements in anaerobic digestion performance were observed up to S1V3. Extending UD duration to 100 s boosted the CH 4 yields to 284 ± 11 cm 3 /gVS. The highest AD performance for Scenedesmus sp. was obtained in S1V4 and S1V5 (150 s and 200 s UD, respectively), with no statistically significant differences in CH 4 production between the two (S1V4-309 ± 13 cm 3 /gVS, S1V5-313 ± 15 cm 3 /gVS). The CH 4 fraction in the biogas was similar across all of the UD variants, falling within the range of 53.2 ± 1.7% (S1V3) to 54.9 ± 0.9 (S1V4). A very strong positive correlation (R 2 = 0.9126) and a strong positive correlation (R 2 = 0.8718) were found between UD exposure time and biogas/methane production ( Figure 2a).
On the other hand, 50 s UD applied to Pinnularia sp. did not significantly affect anaerobic digestion performance in terms of biogas yield and composition. S2V1 yielded 312 ± 14 cm 3 /gVS biogas containing 48.8 ± 3.0% CH 4 , whereas S2V2 produced 317 ± 23 cm 3 /gVS and 48.2 ± 2.1% CH 4 ( Table 2). Variant S2V3 and S2V4 showed incremental increases in biogas yield at 384 ± 31 cm 3 /gVS and 479 ± 17 cm 3 /gVS, respectively ( Table 2). Increasing UD exposure time to 200 s produced no further significant gains in AD performance. The UD treatment also did not produce any statistically significant changes in the CH 4 fraction in the biogas, which varied between 48.8 ± 3.0% in S2V1 to 50.9 ± 1.6% in S2V5 (Table 2). UD exposure time was found to strongly and positively correlate with biogas (R 2 = 0.9212) and methane (R 2 = 0.9293) production ( Figure 2b).   [62] produced findings similar to our own, noting an increase in methane outputs from AD of A. obliquus from 191 m 3 /tCOD in the control to 292 m 3 /tCOD after UD, which translates to a 51% increase in methane yields. A positive effect of UD on microalgal anaerobic digestion performance was also demonstrated by   [67], who optimized their UD experiment by screening different power levels (100 W-180 W) and different disintegration periods (0 min-120 min). Biogas production for the UD-treated mixed microalgal biomass peaked at 185.9 cm 3 /g COD biogas, whereas the non-pretreated control produced only 17 cm 3 /g COD [67]. In contrast, Cho et al. (2013) [61] obtained only slight uptick in methane yields after subjecting Chlorella sp. and Scenedesmus sp. to UD. After 30 s, 90 s and 180 s UD, the yields rose to 356 cm 3 /gVS, 368 cm 3 /gVS and 385 cm 3 /gVS methane, respectively (the non-disintegrated control produced 336 cm3/gVS) [61]. Caporgno et al. (2016) [68] also found that methane yields from Phaeodactylum tricornutum were unaffected by increasing UD energy levels. Energy inputs of 21 MJ/kgTS, 36 MJ/kgTS and 52 MJ/kgTS prior to AD resulted in yields of 287 ± 11 cm 3 /gVS, 284 ± 9 cm 3 /gVS and 285 ± 4 cm 3 /gVS, respectively-a mere 10% in-crease in methane production compared with non-treated microalgae (258 ± 12 cm 3 /gVS). The authors see this as further proof that the refractory nature of the organic fraction in Phaeodactylum tricornutum is the main obstacle for methane production since the pretreatment destroys microalgae cells, but does not increase degradability [68]. Biomass composition is the most important determinant of methane production in anaerobic digestion. Methane yields from algal biomass correlate mainly with the cellular lipid content [69].

Energy Balance
Calculating from the obtained methane yields and the calorific value of CH 4 (9.17 Wh/dm 3 ) shows that variants S1V4 and S1V5 performed the best in terms of gross energy production at 2.834 ± 0.12 Wh/gVS and 2.870 ± 0.14 Wh/gVS (Table 3). In contrast, S1V1 (no UD pretreatment) generated only 1.678 ± 0.23 Wh/gVS ( Table 3). The energy consumed by the UD biomass pre-treatment was proportional to its duration, starting from 0.420 Wh/gVS and peaking at 1.680 Wh/gVS (Table 3). When factoring in the energy inputs, S1V2 and S1V3 are shown to have significantly higher net energy values: 1.909 ± 0.20 Wh/gVS and 1.764 ± 0.10 Wh/gVS, respectively ( Table 3). The net energy gain for other variants was significantly lower than the control, varying between 1.574 ± 0.12 Wh/gVS and 1.190 Wh/gVS (Table 3). A strong positive correlation (R 2 = 0.8718) was found between the DU exposure time and gross energy gain. Conversely, a moderate negative correlation (R 2 = 0.5837) was found between the DU exposure time and net energy gain (Figure 3a). S2 (Pinnularia sp.) failed to produce a net energy balance in any of its UD variants, whereas the control showed a net energy gain of 1.394 ± 0.19 Wh/gVS, the UD variants yielded between 0.943 ± 0.22 Wh/gVS (S2V2) and 0.453 ± 0.21 Wh/gVS (S2V5) ( Table 3). A very strong positive correlation (R 2 = 0.9293) was found between the DU exposure time and gross energy gain. Conversely, a strong negative correlation (R 2 = 0.8870) was found between the DU exposure time and net energy gain (Figure 3b). Studies have shown that a positive energy balance in relation to the control sample, where UD (V1) was not used, was obtained only when Scenedesmus sp. biomass was tested in S1V2 and S1V3. It was, respectively, 0.231 ± 0.02 Wh and 0.086 ± 0.07 Wh (Table 3). In the remaining variants, the net energy gain differential was negative. However, it must be emphasized that the results obtained on a laboratory scale can only be the initial basis for further research. More reliable data used for reliable LCA, LCC and in-depth energy and economic balance analysis can be obtained in tests conducted in conditions close to the real one. Pilot-scale studies reduce the differences and uncertainties associated with the scale-up process. Energy efficiency is a major factor in viability assessments, especially for large-scale processes [70]. Cho et al. (2013) [61] investigated net energy balance as part of their study and found ultrasonic pre-treatment of microalgal biomass to be net energy negative. UD exposure times of 30 s, 90 s, and 180 s resulted in net energy production of −26.4 kJ/gVS, −104.0 kJ/gVS, and −220.4 kJ/gVS, respectively, calculated from energy inputs of 39 kJ/gVS, 117 kJ/gVS, and 234 kJ/gVS, respectively, and outputs of 12.6 kJ/gVS, 13.0 kJ/gVS, and 13.6 kJ/gVS, respectively [61].   [67] found that maximum solubilization of COD was achieved after applying −2542.55 kWh ultrasonic energy per ton biomass. The energy yield from the methane was calculated to be 333.38 kWh/ton biomass, meaning that the net energy balance was −2209.17 kWh/ton biomass [67]. Biomass density seems to be a key factor in determining the energy balance of the process. Due to the specific energy applied, which is inversely proportional to the initial solid concentration, it is possible to reach a positive energy balance by increasing the solids in the system. This means that the harvested microalgal biomass should not only be thickened, but also dewatered before ultrasound pre-treatment [71].

Experimental Design
This study on the effect of ultrasonic disintegration (UD) on the performance of microalgal biomass AD was separated into two stages, each focusing on a different species of microalgae: Scenedesmus sp. in stage 1 (S1) and Pinnularia sp. in stage 2 (S2). Each stage was subdivided into five experimental variants (V1-V5) with different rates of ultrasonic energy applied to the feedstock. The energy dose was adjusted by modifying the biomass retention time in the sonification zone. The experimental design, UD exposure time, and energy input values are presented in Table 4.

Materials
Scenedesmus sp. (UTEX 1589) and Pinnularia sp. (UTEX LB FD462)-S1 and S2, respectively-were grown in tap water at 20 ± 1 • C and illuminated with a 3500 lux fluorescent lamp under a 12 h light/12 h dark regime. Ambient air fed via a diffuser system (with a capacity of 150 dm 3 /h) served as the source of carbon dioxide. S1 used Bold's basal medium (BBM) containing (dm −1 ): KH 2  The cultured microalgal biomass was separated using a vacuum membrane filtration kit (MDS 1, Whatman), which included a 50 mm MCE (mixed cellulose ester) filter insert (5.0 µm porosity). The microalgal biomass was separated from the culture medium in the filter compartment by applying 0.5 atm suction with a vacuum pump (Mobil 20). The membrane separation process produced thickened biomass containing an average 97 ± 1% water. The biomass profiles are given in Table 2. 500 cm 3 of the thickened biomass was fed into the ultrasonic disintegrator then into the digestion respirometers. The anaerobic sludge inoculum was sourced from the digesters of the wastewater treatment plant in Olsztyn ( Table 5). The digester operational parameters were: organic load rate 2.5 kg VS/m 3 ·d, hydraulic retention time 20 d, and temperature 35 • C.

Experimental Set-Up
The microalgae biomass was subjected to UD with an UP 400S ultrasonic disintegrator (Hielscher Ultrasonics GmbH, Teltow, Germany) running at 400 W power and 24 kHz frequency. In successive variants, the ultrasound exposure time was incrementally increased while keeping the volume of feedstock constant (Table 4). After the UD, the anaerobic digestibility of the biomass was tested. Methane output was measured using the volumetric gas production method in batch respirometric reactors (AMPTS II, BPC Instruments AB, Lund, Sweden). The produced biogas was purified by an alkaline scrubbing solution (3M NaOH) to absorb CO 2 and other non-methane gases in an ex situ absorption unit. The digestion process was run at 37 ± 1 • C. The bioreactors were equipped with a vertical stirrer operating for 30 s every 10 min at 100 rpm. The active volume of the respirometers was 200 cm 3 . Initial organic load rate (OLR) was 5.0 gVS/dm 3 . The quantities of feedstock injected into the respirometers are given in Table 6. The ensure anaerobic conditions in the respirometers prior to the measurements, the system was purged with 150 dm 3 /h nitrogen for 5 min. Measurements were taken over a period of 40 d. A gas production report was software-logged once a day, using a program that generates results for a normalized gas volume (standard atmospheric pressure of 101.3 kPa at 0 • C and zero humidity). The readouts continued until the available organic compounds were completely decomposed. Successive biogas volume readouts were automatically compared against each other. The measurement was concluded when ten consecutive gas volume measurements were within 1% of each other. Endogenous biogas generated by anaerobic sludge was excluded from the calculation.

Analytical Measurements
Contents of dry matter, dry organic matter, and dry mineral matter were determined gravimetrically. Biomass samples desiccated at 105 • C were assayed for TC, TOC, and TN. The analysis was performed using a Thermo Flash 2000 organic elemental molecule analyzer (Thermo Scientific, Waltham, MA, USA). TP was determined colorimetrically in ammonium metavanadate (V) and ammonium molybdate after prior mineralization in a mixture of sulfuric (VI) and chloric (VII) acids at 390 nm using a DR 2800 spectrophotometer (Hach-Lange GmbH, Düsseldorf, Germany). Total protein was calculated by multiplying the value of TN by the protein conversion factor of 6.25. Reducing sugars were determined colorimetrically with an anthrone reagent at 600 nm using a DR 2800 spectrophotometer (Hach-Lange GmbH, Düsseldorf, Germany). Lipids were quantified using the Soxhlet method with a Büchi extraction apparatus (B-811, Büchi AG, Flawil, Switzerland). The pH determination procedure was as follows: 10 g of the homogenized air-dried sample was weighed out to a 100 mL beaker, after which 50 mL of distilled water was added, and sample pH was measured with a calibrated apparatus. The dissolved chemical oxygen demand (COD) was determined using a DR 5000 spectrophotometer with a HT 200 s mineralizer (Hach-Lange GmbH, Düsseldorf, Germany). Dissolved TOC was quantified by means of a TOC-L analyzer TOC-L (Shimadzu, Kyoto, Japan). Methane in the biogas was assayed with a GC Agillent 7890 A gas chromatograph (Santa Clara, CA, USA).

Calculation Methods
The specific energy input (E in ) was calculated using Equation The energy output (E out ) generated from methane production was calculated using Equation (2):  (3):

Statistical Analysis
The experiments were conducted in four repetitions. The statistical analysis of experimental results was conducted using STATISTICA 13.1 PL (StatSoft, Inc., Tulsa, OK, USA). One-way analysis of variance (ANOVA) was used to determine differences between variables. Significant differences between the variables were determined via Tukey's HSD. Results were considered significant at p = 0.05.

Conclusions
Pre-treating microalgal biomass with UD was found to significantly impact the performance of the anaerobic digestion (AD) process, directly affecting the levels of COD and TOC in the liquid phase. Dissolved organics significantly rose after Scenedesmus sp. was pretreated with UD, with a eight-fold increase for COD and almost a 9.5-fold increase for TOC. However, UD-treated Pinnularia sp. performed much worse.
The present study shows that UD has a positive effect on anaerobic digestion of microalgal biomass, providing significantly better performance with regard to Scenedesmus sp. digestion. The highest CH 4 yields were noted for the 150 s and 200 s UD exposure times, as these variants produced 309 ± 13 cm 3 /gVS and 313 ± 15 cm 3 /gVS, respectively. The best performing Pinnularia sp. variant yielded 250 ± 21 cm 3 CH 4 /gVS.
The 50 s UD/Scenedesmus sp. group performed the best in terms of net energy gain at 1.909 ± 0.20 Wh/gVS. The energy efficiency of the AD process was not significantly improved by extending the disintegration time, with results similar to or lower than the control group. The digestion of UD-treated Pinnularia sp. failed to perform better than the non-pretreated biomass in terms of energy yield.