Abstract
Anaerobic digestion (AD) is generally considered to be an economic and environmentally friendly technology for treating waste activated sludge, but has some limitations, such as the time it takes for the sludge to be digested and also the ineffectiveness of degrading the solids. Various pre-treatment technologies have been suggested to overcome these limitations and to improve the biogas production rate by enhancing the hydrolysis of organic matter. This paper studies the use of hydrothermal pre-treatment (HTP) for a food waste and sewage sludge mixture (FW–SS mixture) as pre-treatment of co-digestion. The results of the capillary suction time, time to filter, and particle size decreased with increasing HTP temperature. These results of the assessment that was conducted in this study confirm that the HTP process indeed modifies the physical properties of the FW–SS mixture to enhance the solubilization of organic solids. A maximum increase in biogas production of 50% is achieved with a HTP temperature of 140oC. These findings show that to achieve high conversion efficiency, an accurately designed pre-treatment step must be included in the overall AD process for wastewater treatment.
Keywords
Introduction
The generation of food waste (FW) and sewage sludge (SS) has been increased with the rapid development of the industry, economy, and living standards in South Korea. In 2016, the generation of FW and SS was about 14,000 tons/day and 3,654,000 tons/year, respectively. Until now, most of these organic wastes have been treated through landfill, composting, and incineration. These treatment processes, however, cause secondary problems such as land pollution, groundwater contamination, and carbon emission.
Anaerobic digestion (AD) can be a solution for treating organic waste as it can stabilize sludge, reduce solids, and produce biogas, even though the process has the limitations of long retention time and low overall degradation efficiency of the organic matter. As most of the organics present in FW and SS are slowly biodegradable, the rate-limiting step in FW and SS digestion is the hydrolysis of organic matter (Park and Kim, 2015a; Park et al., 2016). Thus, various technologies have been developed to improve the biodegradability of complex wastes, through the application of different pre-treatment methods (Park et al., 2016). Many pre-treatment methods, including the biological, mechanical, chemical, and thermal methods, have been investigated (Carrere et al., 2009; Ge et al., 2010; Li et al., 2012; Liu et al., 2012; Toreci et al., 2009). Pre-treatment improves the overall digestion process’s velocity, efficiency, and sludge reduction, thereby reducing the anaerobic digester retention time and increasing the methane production rates (Braguglia et al., 2012; Erden and Filibeli, 2010 ; Mottet et al., 2009).
Among the thermochemical pre-treatment methods, hydrothermal pre-treatment (HTP) is an effective technique for achieving organic matter degradation (Park and Kim, 2015b). This allows the complex organic molecules (e.g., carbohydrates, lipids, proteins, and nucleic acids) to be released from the solids and be broken down. These hydrolysates can then be utilized by the extracellular enzymes produced by the anaerobic microorganisms, leading to improved AD (Wang et al., 2009). The efficiency of the HTP process prior to AD has prompted many to develop a relevant technology (Wang et al., 2010; Zheng et al., 1998). In HTP, a combination of high-temperature and high-pressure subcritical water (180oC < T < 373oC) is used to efficiently achieve hydrolysis, solubilization, and solid destruction (Mursito et al., 2010), in which the reaction to HTP commonly starts at approximately 180oC (Funke and Ziegler, 2010). Such hydrothermal methods are advantageous in that they do not require chemical catalysis, thus reducing the associated costs and being more environment-friendly.
In this study, HTP was applied prior to the AD of organic solid wastes from FW and SS to improve the biogas production and digestion ratio. The effect of the pre-treatment temperature (80–180oC) on the characteristics of organic matter was determined, and the efficiency of biogas production was evaluated, through the biochemical methane potential (BMP) test.
Materials and methods
FW and SS
The FW was collected from the FW recovery facility in Yongin, South Korea, which processes 315 tons of FW per day. The SS, on the other hand, was collected from the Ii-San Municipal wastewater treatment plant in Goyang, South Korea, which processes 305 tons of FW per day. Therefore, in the FW–SS mixture, FW and SS were mixed at a weight ratio 1:1. The characteristics of FW and SS are shown in Table 1.
Characteristics of food waste (FW) and sewage sludge (SS).
COD: chemical oxygen demand; ND: not detected; SD: standard deviation; TN: total nitrogen; TP: total phosphorus; TS: total solid; TVFAs: total volatile fatty acids; VS: volatile solid.
Experiment set-up
HTP process
The HTP experiments were performed using a 500 mL laboratory-scale reactor. The reactor consisted of a reactor body, a heater, and a steam condenser, and was operated under N2 gas. For all the experiments, 200 mL feedstock was loaded into the reactor, with equal amounts of water. The HTP operating temperatures and pressures ranged from 80 to 180oC and from 0.8 to 1.8 MPa, respectively, and the reaction time was 30 minutes. As the pressure in the reactor increases, the cell wall breaking effect is more increased. It was mixed using an agitator rotating at 200 revolutions per minute for the complete mixing of the FW–SS mixture and smoothly increasing temperature. After the completion of the hydrothermal reaction, the residual steam was discharged from the reactor, and the reaction products were removed.
BMP test
The anaerobic methane production was assessed using a batch anaerobic reactor under mesophilic conditions (38oC). A control with only the seed sludge without the FW–SS mixture which mixed FW and SS pre-treated under various temperatures (80, 100, 120, 140, 160, and 180oC) was prepared for the estimation of the produced biogas from only the substrate. Inocula were collected from a farm-scale biogas plant (Anseong, South Korea) digesting FW. The characteristics of the inocula are shown in Table 2. The inoculum samples collected from the anaerobic digester were sieved at 2 mm, and before the start of the BMP assay, 2 L inoculum samples were pre-incubated in a 5 L anaerobic batch reactor under mesophilic conditions (38oC) for 40 days, to deplete the residual biodegradable organic matter present therein. Thereafter, pre-incubated digestate was prepared as an inoculum for the BMP assay. The substrate-to-inoculum ratios of all the BMP assay cultures were equal to 0.3 g-VSsubstrate/g-VSinoculum, and the BMP assays were carried out at an 80 mL working volume, using a 160 mL serum bottle. Methanogenic bacteria are extremely sensitive to pH fluctuations and prefer an around 7.0 pH as the growth rate of methanogens (Park et al., 2016). Based on the study of Park et al., at a lower pH of 6.7 and a high pH of 8.5, the methane yield (MY) was much lower (Park et al., 2016); therefore, NaHCO3 of 6 g/L in the bottle was injected to prevent pH decrease (Park et al., 2016). The head space of the serum bottle was filled with N2 gas and was sealed with a butyl rubber stopper. Three BMP bottles for each sample were incubated up to 48 days for all the pre-treatment schemes in this study. The methane production was corrected for the standard temperature and pressure (STP), and the BMP was determined by the unit of the volatile solid (VS) content of waste added to the vial. To describe the progress of the cumulative methane production, the modified Gompertz equation (equation (1)) was used to fit the cumulative methane production data, as follows (Costa et al., 2012)
where M is the cumulative methane production (mL), e is exp (1), Rm is the maximum specific methane production rate (mL/d), P is the methane production potential (mL), and λ is the lag phase time (days).
Characteristics of the anaerobic inocula.
SD: standard deviation; TKN: total Kjeldahl nitrogen; TS: total solid; VS: volatile solid.
The kinetics constants M, Rm, and λ were determined using the non-linear regression approach for the best fittings with the aid of the solver command in Microsoft Excel (Park et al., 2016). The model parameters in equation (1) were estimated using least-square estimation. This equation was utilized by researchers to study the cumulative methane production in biogas production. Lay et al. (1996) used this equation to study bacteria growth.
Analysis methods
The total solid (TS), VS, pH, soluble chemical oxygen demand (SCOD), total chemical oxygen demand (TCOD), total nitrogen (TN), ammonium nitrogen (NH4+-N), total phosphorus (TP), and alkalinity were determined according to the standard methods (American Public Health Association, 2005). The pH was measured using a pH meter (Cyberscan pH 20). The particle size was analyzed using a Matersizer 2000 (Malvern Instruments, England). The filterability of the pre-treated mixed FW and SS samples with the HTP temperature was assessed by measuring the capillary suction time (CST) using an improvised CST apparatus (Model 319 Multipurpose CST, Triton Electronics Ltd.) with a single-radius test head and CST paper (size: 7×9 cm; quantity: ~200; Triton Electronics Ltd.). The time to filter (TTF) test can be used to evaluate the sludge dewater ability (Abelleira et al., 2012); according to the standard method 2710H, this test involves placing a sludge sample in a Büchner funnel with a paper support filter (Whatman Filter No. 1001-090, 11 μm), applying a vacuum/pressure pump, and measuring the time required to filter 50% of the original sample. The solubilization chemical oxygen demand (COD) was determined using equation (2)
where TCOD0 and SCOD0 are the initial TCOD and SCOD, respectively, and SCODT is the SCOD after the pre-treatment at each temperature.
The volatile suspended solids (VSS) reduction rate was calculated using equation (3)
where VSS0 is the initial VSS of the FW–SS mixture, and VSST is the VSS after the thermal pre-treatment of the mixture at each condition.
The FW–SS mixture and its products were dried at 105±5oC for 24 hours, after which they were evaluated for theoretical biogas production using a PerkinElmer 2400 Series II CHONS organic elemental analyzer (Thermo Finnigan Ltd., USA) to determine the weight percentages of the chemical elements. The results were the averages of the three experiments performed in triplicate. The chemical compositions of the FW and SS were analyzed using an energy-dispersive spectrometry system for SEM (TEAMTMEDS, EDAX Inc., USA). The total gas production by the BMP assay was measured daily for the first five days and every two to three days thereafter, through the displacement of an acidified brine solution in a burette, and the volume of the displaced solution after correcting to atmospheric pressure was recorded (Willems et al., 1996). The methane content was measured via gas chromatography (Varian Model STAR 3400CX, USA; carrier gas, N2; injector temperature, 150oC; column temperature, 29oC; flame ionization detector (FID) temperature, 200oC). Also, the volatile fatty acid (VFA) concentration was measured via gas chromatography (Agilent 7890a FID, USA; carrier gas, N2; injector temperature, 252oC; column temperature, 145oC; FID temperature, 250oC).
Theoretical methane potential
The chemical compositions of the substrates were determined from the elemental analysis data, and the theoretical MY (Bth) was estimated based on the stoichiometry of the degradation reaction using the Buswell formula (equation (4)), as follows (Boyle, 1976)
where a, b, c, d, and e are the molecular amounts of carbon, hydrogen, oxygen, nitrogen, and sulfide, respectively.
The Bu (ultimate methane potential) in terms of normal cubic meters per VS content (L-CH4/kg-VSadded) under the standard conditions (0oC, 1 atm) was calculated using equation (5)
Results and discussion
Characteristics of liquid products with an HTP temperature
The HTP carried out in this study was found to have improved the solubilization and to have increased the response surface of the FW–SS mixture. Table 3 shows the effects of various HTP temperatures on the characteristics of the FW–SS mixture. The VS and SCOD of the FW–SS mixture increased from 64,000 and 30,600 mg/L at 80oC to 69,000 and 69,900 mg/L at 180oC, respectively, and was found to have increased with increasing HTP reaction temperature. The changes in the FW–SS mixture solubilization were used to describe the transfer from particular fractions of the sludge. HTP reactions as a pre-treatment method are solubilization and the cell destruction of the sludge and biomass. The decrease in raw material (untreated) VS can be explained by the fact that the VS was pyrolyzed at higher temperatures using the HTP reaction.
Effects of the various hydrothermal pre-treatment temperatures on the characteristics of the food waste and sewage sludge mixture.
±: standard deviation; SCOD: soluble chemical oxygen demand; TCOD: total chemical oxygen demand; TS: total solid; TSS: total suspended solids; VS: volatile solid; VSS: volatile suspended solids.
Figure 1 shows the solubilization COD (solubilization rate) and SCOD concentration according to the reaction temperature. The COD solubilization rate and SCOD concentration increased with increasing temperatures. It was concluded that the SCOD concentration and solubilization rate increased as the particulate solids changed into dissolved substances with increasing HTP temperatures.

Enhancement of the chemical oxygen demand solubilization and soluble chemical oxygen demand concentration by hydrothermal pre-treatment temperature.
Figure 2 shows the NH4+-N concentration and the NH4+-N/T-N fraction according to the reaction temperature. As the HTP temperature increased, the nitrogen in the FW and the solid matter of the SS were dissolved, thus increasing the concentration of nitrogen and simultaneously increaseing the concentration of NH4+-N. The concentration of NH4+-N before HTP was 650 mg/L, but it increased by approximately 6.5 times to 4200 mg/L at the HTP temperature of 180oC. Wang et al. (2009) and Park et al. (2016) reported that the biogas production and organic removal decreased during AD of thermally hydrolyzed wasted sludge with a high concentration of ammonium ion (3.0–9.0 g NH4+-N-N/L) because 60–90% of methanogenesis were inhibited by the high-strength ammonium ion concentration. This factor also influenced the results of the biogas production measurements in this study.

Variation of NH4+-N/T-N with the hydrothermal pre-treatment temperature.
Figure 3 shows the total volatile fatty acid (VFA) concentration according to the reaction temperature. The total VFA concentration increased with increasing reaction temperatures. In AD, VFAs are the most important stage for biomethane production through a stable AD just before the production of biomethane. An increase in VFA concentration means that the amount of biomethane that can be produced is also increased. In other words, an increase in the VFA concentration with increasing solubilization temperatures may have a positive effect on AD. The total VFA concentrations at the reaction temperatures of 140oC and 160oC were the highest at 12,600 mg-COD/L or more, which was about 20 times higher compared with before the thermal-hydrolysis of the FW and SS. The total VFA concentration at 180oC decreased by 2000 mg-COD/L compared with those at 140oC and 160oC due to the volatilization of the VFAs caused by high temperatures. Therefore, the optimal thermal-hydrolysis temperatures based on the total VFA concentration were determined to be 140oC and 160oC.

Total volatile fatty acids of the food waste and sewage sludge mixture with the hydrothermal pre-treatment temperature.
Changes in the physical properties of the FW–SS mixture
The HTP led to alterations of the physical properties of the FW–SS mixture. The filterability of the FW–SS mixture was evaluated via CST and TTF analyses, which were conducted as assessments of the efficacy of HTP in breaking up flocks of sludge to increase the solubilization of organic matter (Jin et al., 2005).
Figure 4 shows the CST and TTF according to the reaction temperature, and the dewaterability can be evaluated through the CST and TTF measurement methods. FW and SS have an 80% or higher water content, but their physical properties are close to those of solids, giving them low dewaterability. The thermal-hydrolysis reaction, however, causes the physical properties of FW and SS to change to those of the slurry state, which increases the dewaterability. An increase in dewaterability means that the solid material is changed into the slurry state, indicating that the particulate material has been converted into a soluble material through the thermal-hydrolysis reaction. In other words, the solubilization is known through CST and TTF to progress. The CST of the raw FW–SS mixture was 450.2 seconds; after HTP, however, the CST of the FW–SS mixture decreased to 200.5, 105.8, 55.4, 60.0 and 40.4 seconds for the HTP at 80oC, 100oC, 120oC, 140oC, 160oC, and 180oC, respectively. In addition, after HTP, the TTF of the FW–SS mixture decreased from 210.1 seconds at the raw FW–SS mixture to 35.6 seconds and 21.5 seconds at 160oC and 180oC, respectively. As the reaction temperature increases, the CST and TTF reaction time is shortened because it destroys the cell walls of the SS and changes the bound water of sludge—which physically makes dewatering difficult—to free water. The CST and TTF results clearly indicate that the physical structure of the sludge was altered by HTP.

Variation of the capillary suction time and time to filter of the food waste and sewage sludge mixture by hydrothermal pre-treatment temperature.
Figure 5 shows the average particle size of the FW–SS mixture with the HTP temperature. The initial average particle size of the FW–SS mixture was over 800 μm, but after HTP, it decreased to 400, 350, 310, 250, 210, and 120 μm; and after the HTP at 80oC, 100oC, 120oC, 140oC, 160oC and 180oC, respectively. The waste is thought to consist of many particles that flock into larger particles in the FW–SS mixture; thus, these larger particles are likely to be broken down and converted to smaller molecules and particles within the sludge upon pre-treatment (Jin et al., 2005). The results of the CST, TTF, and particle size assessments in this study confirmed that the HTP process does indeed modify the physical properties of the FW–SS mixture, to enhance the solubilization of organic solids.

Average particle size of the food waste and sewage sludge mixture with the hydrothermal pre-treatment temperature.
Effect of HTP on co-digestion
The HTP process may enhance the solubilization and anaerobic biodegradability of the FW–SS mixture, and pre-treatment is thought to disrupt the frock and wall of the FW–SS mixture’s structure by damaging the physical structure of the organic solids; therefore, the methane production and rates can be increasing in AD (Mottet et al., 2009). The ultimate MY (UMY) values (Figure 6) showed a threshold value in the increase of methane production. Indeed, it increased with the thermal pre-treatment temperature until 140oC, from 0.444 L CH4/g VSadded for the untreated sludge to 0.294 L CH4/g VSadded for the sludge pre-treated at 140oC. Thus, among the BMP-tested temperatures, the optimum HTP temperature was found to be 140oC. Moreover, at 160 and 180oC, although large solubilization of particulate organic matter occurred, the sludge MY biodegradability was almost the same as the raw sludge biodegradability (0.42 and 0.36 L CH4/g VSadded, respectively). It was determined that at 160oC and 180oC, the carbohydrate in the soluble phase reacted with the other components to form the product slowly or to make it hardly biodegradable (Park and Kim, 2015b). These results are in agreement with those obtained by Stuckey and McCarty (1984) and Mottet et al. (2009). Kim et al. (2014) suggested the presence of a changed chemical structure due to the “burnt sugar” reaction and the Maillard reactions for a high pre-treatment temperature via the Fourier transform infrared spectroscopy spectrum. The brown color of the soluble phase of the sludge treated at 160oC and 180oC confirmed the presence of new compounds such as the Amadori compounds and melanoidins, which are recalcitrant to AD. The initial pH values of all the samples were 7.0–7.4, and the final pH values were 7.2–7.4. In other words, the pH did not decrease. This is probably because the reaction between CO2 and ammonia produces HCO3− according to equation (6) (Duan et al., 2012). In addition, it was determined that sufficient NaHCO3 was injected:
Table 4 shows the results of the modified Gompertz model analysis with the HTP temperature. The lag growth phase time (λ) estimated from the methane production date via optimization using the modified Gompertz equation considerably decreased with the HTP until 140oC, from 0.12 days for the untreated sludge to 2.55 days for the sludge pre-treated at 140oC. The shorter lag growth phase time means that the AD time can be shortened. The UMY (Bu), maximum methane production (P), and maximum methane production rate (Rm) increased until the HTP temperature was at 140oC. Howevewr, except for the parameter lag growth phase time (λ), all the parameter values decreased at the HTP temperature above 140oC in the modified Gompertz models. It was determined that the pre-treated FW–SS mixture at the HTP temperatures of 160oC and 180oC had a very high ammonium concentration (above 3.5 g NH4+-N/L) (Table 3). It has been reported that the ammonia inhibition is within the 1.5–3.0 NH4+-N/L range at a pH higher than 7.4 whereas ammonia can be claimed to be toxic irrespective of the pH at higher than 4.0 g NH4+-N/L (Park et al., 2016). In addition, the presence of the “burnt sugar” reaction and the Maillard reactions at the HTP temperatures of 160oC and 180oC, respectively, was determined. Therefore, the optimal HTP temperature of mixed FW and SS for AD was found to be 140oC for the following modified Gompertz model analysis.
Results of the modified Gompertz model analysis.
Bu: ultimate methane yield; P: maximum methane production; Rm: maximum methane production rate; λ: lag growth phase time.

Comparison of the theoretical methane yield and ultimate methane yields as well as the biodegradability of the food waste and sewage sludge mixture samples before and after hydrothermal pre-treatment (theoretical methane yield (TMY) (L CH4/kg VSadded), ultimate methane yield (UMY) (L CH4/kg VSadded), and biodegradability (%) (UMY/TMY×100, where UMY is the cumulative methane yield and TMY is the theoretical methane yield).
Conclusions
In this study, HTP was evaluated for its enhancement effect on the solubilization, physical properties, and co-digestion of an FW–SS mixture waste. HTP was shown to increase the SCOD, solubilization COD and VFAs. The CST, TTF, and particle size decreased with the increasing HTP temperature of the FW–SS mixture waste. These results clearly indicate that the physical structure of the FW–SS mixture was altered by HTP. MY increased with HTP until 140oC, from 0.444 L CH4/g VSadded for the untreated FW–SS mixture to 0.294 L CH4/g VSadded for the FW–SS mixture pre-treated at 140oC. In addition, based on the results of the modified Gompertz model analysis, except for the parameter lag growth phase time all the parameter values increased with the HTP temperature until 140oC. Therefore, the optimum operating temperature for the HTP of the FW–SS mixture waste was shown to be 140oC. Furthermore, it is recommended that HTP pretreatment be added during AD of the FW–SS mixture waste.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a grant [IFIP-B113506-01] from the development of plant program funded by the Ministry of Land, Infrastructure and Transport of the Korean Government. Also, this work was supported by the Korea Institute of Energy Technology Evaluation and Planning, and the Ministry of Trade, Industry & Energy of the Republic of Korea [No. 20172020108940].
