Comparison of Anaerobic Co-Digestion of Food Waste and Livestock Manure at Various Mixing Ratios under Mesophilic and Thermophilic Temperatures

Abstract

In this study, the optimum mixing ratio of food waste (FW) and livestock manure (LM) was investigated to improve the methane yield efficiency and prohibit the inhibition factors (organic loading rate and NH₄⁺) from inhibiting the anaerobic co-digestion of FW and LM under mesophilic and thermophilic conditions. The research involved the following: (I) the analysis of the characteristics of FW and LM, (II) the evaluation of the potential and toxicity of the anaerobic digestion of I have confirmed that there is no problem. FW and LM using the biochemical methane potential (BMP) and anaerobic toxicity assay (ATA) tests, (III) the evaluation of the anaerobic co-digestion of FW and LM using the BMP test, and (IV) the evaluation of the optimum mixing ratio using mathematical modeling. The characteristics of FW and LM were analyzed to evaluate the theoretical methane potential and inhibition factor. The BMP test was carried out to evaluate the concentration of the biodegradable organic matter, biogas production rate, and methane yield. The ATA test was carried out to evaluate the impact of the inhibition concentration. Ultimately, mathematical models, such as a first-order reaction and a modified Gompertz model, were implemented to evaluate the optimum mixing ratio for the anaerobic co-digestion of FW and LM. FW had a higher concentration of degradable organic matter than LM. The initial operational parameters of the anaerobic digestion were determined to be appropriate at an organic matter concentration of less than 2.5 g/L and a TN concentration of 2,000 mg/L. In conclusion, as a result of evaluation through mathematical models, it was determined that anaerobic microorganisms were more sensitive to inhibitory factors under the thermophilic condition than under the mesophilic condition, and the optimum mixing ratio of FW to LM was 5:1 (vol:vol) based on kinetic results (k: 0.080; B_u: 0.23 L CH₄/g VS_added; P: 100.84 mL; R_m: 10.23 mL/day; λ: 1.44 days).

1. Introduction

Anaerobic digestion has gained importance within the last three decades as an effective treatment process for different characteristics of organic waste because it can produce renewable biogas through biological treatment. Research on anaerobic digestion has recently accelerated regarding the effects of the process performance and stability, inhibition and toxicity, and optimization on anaerobic digestion [1,2,3]. Among organic wastes, food waste (FW) and livestock manure (LM) have received attention as raw materials for anaerobic digestion.

FW is not suitable as a traditional approach for waste treatment, such as incineration, landfills, and composting, because of its high moisture content and low calorific value. However, it is a highly appropriate substrate for anaerobic digestion because of its high concentrations of biodegradable and nutrient contents. The theoretical methane production rate of FW is estimated to be from about 0.4 to 0.5 L CH₄/g VS, which is 2~3 times that of sewage sludge [4]. Therefore, many systems for the anaerobic digestion of FW have been developed and applied. However, FW hinders the efficiency and stability of anaerobic digestion because of characteristics such as low pH and high oil, salt, and protein contents [5].

The anaerobic digestion of LM is effective for waste management, renewable energy generation, and reductions in pollution and greenhouse gas emissions [6]. However, several factors of LM, such as its high moisture, high lignocellulosic component, and ash contents and low organic load, hamper anaerobic digestion [7]. In addition, the inhibition of ammonia (NH₄⁺) of high N concentrations in LM can result in low methane production and a high volatile fatty acid (VFA) level in the effluent. In general, the anaerobic digestion of NH₄⁺ was shown to be inhibited at concentrations of 1500–3000 mg/L [8,9].

Despite these potential factors of organic waste for anaerobic digestion, mono-anaerobic digestion requires many technologies because of the characteristics of organic waste. Recently, anaerobic co-digestion has attracted attention because it provides an advantage in overcoming the inhibition factor of the mono-digestion by simultaneously digesting two or more substrates [9,10]. The anaerobic co-digestion of FW and LM can boost the system efficiency via the dilution of toxic and inhibitory compounds, the promotion of microbial diversity, and the balance of nutrients [9]. The anaerobic co-digestion efficiency can be increased through an appropriate C/N ratio of co-substrates of 20–30 for anaerobic digestion by mixing FW with a high organic content and LM with a high nitrogen content. However, the increase in nitrogen substances inhibits anaerobic digestion as the mixing ratio of the LM increases. Therefore, it is necessary to predict the mixing ratio of the LM by evaluating the NH₄⁺ inhibition concentration when FW is used as a substrate in anaerobic digestion. Many studies have been conducted for optimizing the mixing ratio of FW to LM [7]. Nevertheless, it is important to determine the appropriate mixing ratio for the stable anaerobic co-digestion of FW and LM based on the characteristics of wastes. Otherwise, unsuitable substrate mixing ratios can result in organic overloading, acidification, and system failure.

Many studies have evaluated the impacts of the temperature and influent substrate, as important parameters of efficiency and stability, on anaerobic digestion. These parameters affect the kinetics of biochemical reactions and the microbial community’s structure, particularly with complex substrates, such as FW and LM. These effects can slow down or interfere with the digestion process by disrupting the homeostatic equilibrium of microbes or inhibiting biochemical reactions and are generally shown as a decrease in biomethane because of the accumulation of inhibitors, such as VFA and NH₄⁺. In addition, the rates of reactions and the potential of inhibition can be increased with increasing anaerobic digestion temperatures [11]. However, few studies on the systematic effects of the anaerobic co-digestion efficiency and inhibitory factors depending on the mixing ratio and temperature conditions of the FW and LM have been conducted.

The main purpose of this study was to evaluate the optimum mixing ratio for the anaerobic co-digestion of FW and LM. In order to determine the optimum mixing ratio, the characteristics of the FW and LM were analyzed, and the potential and toxicity of the anaerobic digestion were determined using the biochemical methane potential (BMP) and anaerobic toxicity assay (ATA) tests. Lastly, the optimum mixing ratio of FW to LM was evaluated through mathematical modeling by applying a first-order reaction and a modified Gompertz model based on the results of the BMP and ATA tests.

2. Materials and Methods

2.1. Food Waste and Livestock Manure

FW was collected from an FW recovery facility at 10 kg/day for 7 days. The collected FW was ground and mixed using a household blender. LM was collected from a septic truck entering the sewage treatment plant. These wastes were collected in Chungju, South Korea, and their characteristics are shown in

Table 1. Characteristics of food waste and livestock manure.

Parameter	Unit	Value (Average ± S.D. a)
		FW	LM
Total COD	g/L	220.4 ± 48.2	54.0 ± 5.3
Soluble COD	g/L	115.8 ± 40.2	35.6 ± 4.2
TN	g/L	5.6 ± 1.8	4.2 ± 2.2
TP	g/L	2.0 ± 0.9	0.9 ± 0.7
NH4+	g/L	0.2 ± 0.1	2.8 ± 0.8
Na+	g/L	3.8 ± 0.06	0.6 ± 0.01
Cl−	g/L	2.6 ± 1.8	0.9 ± 0.6
TS	g/L	153.0 ± 29.3	20.2 ± 14.7
VS	g/L	142.9 ± 28.5	13.1 ± 11.8
VS/TS	%	93.4	64.9
pH	-	4.0	7.2
C	%	49.7	43.1
H	%	6.9	5.6
O	%	33.0	24.3
N	%	4.0	4.9
S	%	0.3	1.5

^a S.D.: standard deviation.

.

2.2. Experimental Setup

2.2.1. Biochemical Methane Potential (BMP) and Anaerobic Toxicity Assay (ATA) Tests

Figure 1 shows the detailed steps and processes of the BMP and ATA tests in the experiment. The first test evaluated the potentials of the anaerobic digestions of the FW, LM, and mixed waste, whereas the second evaluated the toxicities of the FW and LM during anaerobic digestion. The mixing ratio of FW to LM of step II was set based on the results of these tests. In step III, mathematical modeling was performed using the potential anaerobic digestion results of the mixed waste based on the results of step II to evaluate the optimum mixing ratio.

The BMP test has been used to characterize a wide variety of substrates by applying many operational factors, such as pH and the mixing and incubation times, based on the popular methodology of Owen et al. (1979) [12]. The principal rule of the BMP test is that the microorganisms must be acclimated to the pollutants among the many factors of the BMP test. Therefore, for the BMP test, the microorganisms were adapted to the pollutants, and the efficiency of the anaerobic digestion was evaluated for each input waste in this study. The anaerobic microorganisms, collected from a mesophilic anaerobic digester in a wastewater treatment plant in South Korea, were adapted by feeding FW into a 5.0 L flask. The adaptation conditions were an organic loading rate (OLR) of 1.6 g VS/L/d and a hydraulic retention time (HRT) of 30 days for a total period of 285 days and an OLR of 3.0 g VS/L/d for an HRT of 30 days for a total period of 65 days at mesophilic (38 °C) and thermophilic (55 °C) temperatures.

The ATA test evaluates the potential toxicity of a substance (Owen et al. 1979) [12]. The bottle preparation and operational procedures of this test are similar to those of the BMP test. The ATA test procedure involves adding increasing volumes of organic and toxic substances to a serum bottle. If there is toxicity in the substances, the initial rate of the gas production will be reduced in proportion to the volume of the substance added. The significant difference between the ATA and BMP tests is that ATA is interested in the initial rate of the gas generation, while in the BMP test, the total amount of the gas production is important. Therefore, in this study, the ATA test was carried out to evaluate the toxicities and concentrations of the inhibition factors, such as high concentrations of organic matter and NH₄⁺.

2.2.2. Evaluation of the Biodegradable Organic Matter’s Concentration (BMP Test)

The BMP test was carried out to evaluate the degradable organic matter concentrations of the FW and LM. The FW and LM were each injected at 1, 2, 4, and 6 mL, and the inocula were injected at 90 mL into a 125 mL serum bottle. More microorganisms were added compared to those in the organic waste to prevent the impairment of the anaerobic digestion efficiency because of the high concentrations of the inhibitors. However, because the chemical oxygen demand (COD) conversion is generally proportional to biomass x time, the relative content of the microorganisms affects the rate of the conversion but not the net ultimate value. The BMP tests were carried out at 38 and 55 °C in an incubator (DI-250, DH Science, Seoul, Republic of Korea). All the experiments were performed in duplicate for experimental accuracy. N₂ gas was purged to ensure anaerobic conditions for two minutes, and the serum bottle was sealed with a rubber stopper. Biogas production was measured to be the pressure difference between the inner pressure of the serum bottle and the atmospheric pressure using a manometer (Series 1221, Dwyer, Michigan City, IN, USA). Biogas production was measured every day until the amount of the gas production was observed. The degradable organic matter’s concentration was determined using Equation (1) as follows [8]:

$${De}_{COD}={\displaystyle \frac{\left(V_1-V_0\right)\times C_1}{S_1\times T_1}}$$

(1)

where De_COD is the degradable COD concentration, V₁ is the biogas volume of the organic waste measured with a syringe (mL), V₀ is the biogas volume of the organic waste at the previous sampling time (mL), C₁ is the methane content (%) at the sampling time, S₁ is the injection volume of the waste in the serum bottle (mL), and T₁ is the CH₄ production equivalent to 1 g of COD reduction (mL).

2.2.3. Anaerobic Toxicity Assay (ATA) Test

The ATA test was carried out to evaluate the initial toxicity of the FW and the inhibitory concentration of NH₄⁺ of the LM. The FW was injected at 5, 10, 15, and 20 mL, and the inoculum was injected at 90 mL into a 125 mL serum bottle to evaluate the initial toxicity of the organic substance. The NH₄⁺ was adjusted to 0, 2000, 3000, 4000, 5000, 6000, and 7000 mg NH₄⁺/L using NH₄HCO₃, and 90 mL of the inoculum was injected into a 125 mL serum bottle to evaluate the NH₄⁺ inhibition concentration. The pH in the serum bottle ranged from 8.1 to 8.5 at the start of the experiment. An amount (10 mL) of glucose, which easily decomposes during anaerobic digestion, was injected for the ATA test. The toxicity inhibitions of the FW and LM were evaluated by the initial biogas production rate and methane yield. The ATA tests were carried out at 38 and 55 °C, and the experiments were performed in duplicate. Biogas production was measured using the same method used in the BMP test. The amount of methane produced was determined using Equation (2) as follows [13]:

$$V_{CH4}=C_1\left(V_1+V_0\right)-C_0{V}_0$$

(2)

where V_CH4 is the volume of methane produced (mL), C₁ is the methane content (%) at the sampling time, C₀ is the methane content (%) at the previous sampling time, V₁ is the biogas volume measured with a syringe (mL), and V₀ is the gas-phase volume of the reactor (mL).

2.3. Analytical Methods

The determinations of the total chemical oxygen demand (TCOD), soluble chemical oxygen demand (SCOD), total nitrogen (TN), total phosphorus (TP), ammonium (NH₄⁺), total solid (TS), and volatile solid (VS) were carried out according to the standard methods [14]. The concentrations of sodium ions (Na⁺) and chlorine ions (Cl⁻) were analyzed using ion chromatography (Eco IC, Metrohm, Herisau, Switzerland). For Na⁺ ion analysis, a Metrosep C4-150/4.0 column was used, and for Cl⁻, a Metrosep A Supp 5-100/4.0 was used. The temperature of the column was set at 30 °C. The trace elements were determined using a Spectra AA (220FS, Varian Ltd., Palo Alto, CA, USA). The pH was measured using a pH meter (ORION STAR A221, Thermo Fisher Scientific Inc., Waltham, MA, USA).

The methane contents in the BMP and ATA tests were determined with gas chromatography (STAR 3400 CX, Varian, CO, USA) equipped with a capillary column (DB-5, 0.53 mm in diameter, 30 m in length, J&W Scientific, Folsom, CA, USA) and a flame ionization detector (FID). The injection volume of the biogas for methane measurements in the GC was 10 µL. The injector, detector, and column temperatures were 150, 200, and 29 °C, respectively. The initial oven temperature was maintained at 30 °C for 2 min and then increased by 10 °C/min to 50 °C. The nitrogen used as the carrier gas was delivered at a flow rate of 20 mL/min. The FW and LM were dried at 105 ± 5 °C for 24 h before determining the weight percentages of the chemical elements using an organic elemental analyzer (PerkinElmer 2400 Series II, Thermo Finnigan Ltd., Shelton, CT, USA). These results are given as average values and were determined in triplicate for the evaluation of the theoretical biogas production.

The equilibrium concentration between the NH₄⁺ and FA depends on the process’s pH, as given in Equation (3).

$${{\mathrm{N}\mathrm{H}}_4}^{+}+{\mathrm{O}\mathrm{H}}^{-}\leftrightarrow {\mathrm{N}\mathrm{H}}_3+{\mathrm{H}}_2\mathrm{O}$$

(3)

Therefore, the fraction of the FA relative to the total NH₄⁺ was calculated depending on the pH and temperature using Equation (4) as follows [15]:

$${\mathrm{N}\mathrm{H}}_3\left(Free\right)=TAN\times {\left(1+{\displaystyle \frac{{10}^{-pH}}{{10}^{-\left(0.09018+\frac{2729.92}{T\left(K\right)}\right)}}}\right)}^{-1}$$

(4)

2.4. Theoretical Methane Potential

The chemical compositions of the FW and LM were determined from the elemental analysis data. The theoretical methane potential was determined based on the stoichiometry of the degradation reaction using the Buswell equation, as shown in Equation (5) [16].

$$C_a{H}_b{O}_c{N}_d{S}_e+{\displaystyle \frac{4a-b-2c+3d+2e}{4}}{H}_{2}O\to {\displaystyle \frac{4a+b-2c-3d-2e}{8}}{CH}_4+{\displaystyle \frac{4a-b+2c-3d+2e}{8}}{CO}_2+{dNH}_3+{2H}_{2}S$$

(5)

where a, b, c, d, and e are the molecular amounts of carbon, hydrogen, oxygen, nitrogen, and sulfide, respectively. B_t (the theoretical methane potential, in terms of milliliters per VS content (mL CH₄/g VS_added) under standard conditions (0 °C and 1 atm)) was calculated using Equation (6) as follows:

$$B_t=22.4\times \left\{{\displaystyle \frac{\left[4a+b-2c-3d-2e/8\right]}{12a+b+16c+14d+32e}}\right\}$$

(6)

2.5. BMP Mathematical Modeling

The decomposition rate of the substrate can be represented by a differential kinetic equation. In this study, the decomposition rate and biodegradability of the substrate were evaluated using a first-order reaction model and a modified Gompertz model applied to the experimental BMP tests. The decomposition rate of the substrate by microorganisms can be expressed using a first-order reaction, as shown in Equations (7) and (8).

$${\displaystyle \frac{dS}{dt}}=-kS$$

(7)

$$\ln \left({\displaystyle \frac{S}{S_0}}\right)=-kt$$

(8)

where k is the kinetic constant, and S is the substrate concentration.

The concentration of the biodegradable VS in the reactor is related to the amount of gas generated and can be expressed as follows in Equation (9):

$${\displaystyle \frac{S}{S_0}}=(B_u-B)/B_u$$

(9)

where B is the cumulative methane production, and B_u is the ultimate methane production.

Therefore, the decomposition rate and biodegradability of the substrate, according to the reaction time, were determined using Equation (10). As a result, a high value of the kinetic constant can mean that the decomposition of the substrate is rapid.

$${\displaystyle \frac{\left(B_u-B\right)}{B_u}}=\mathrm{e}\mathrm{x}\mathrm{p}(-kt)$$

(10)

The modified Gompertz equation is represented by Equation (11) to describe the progress of the cumulative methane production [17].

$$M=P\times exp\left\{-exp\left[{\displaystyle \frac{R_m\times e}{P}}\left(\lambda -t\right)+1\right]\right\}$$

(11)

where M is the cumulative methane production (mL), e is exp (1), R_m is the maximum specific methane production rate (mL/d), P is the methane production potential (mL), and λ is the lag-phase time (days).

The best fit values of the kinetic constants M, R_m, and λ were determined with the aid of the solver command in Microsoft Excel.

3. Results and Discussion

3.1. Characteristics of the FW and LM

Table 1 shows the characteristics of the FW and LM. The FW contains a high concentration of organic matter, which was converted to an amount of biogas during anaerobic digestion. However, the characteristics of the FW in South Korea are easily affected by seasonal variation, which affects the biogas production and COD removal efficiency [18]. The LM in South Korea is treated with compost, liquid fertilizer, and purification. About 85% of the treated LM is treated with compost and liquid fertilizer, and 15% is purified and discharged into water bodies [19]. The LM containing high concentrations of nitrogen (e.g., NH₄⁺ and FA) may decrease the efficiency of the biogas production because of the inhibition of the NH₄⁺ during anaerobic digestion [3]. Therefore, it is necessary to understand the characteristics of the organic waste and evaluate the appropriate mixing ratio for the co-digestion of FW and LM.

The TCODcr and TN concentrations of the FW were 220.4 ± 48.2 and 5.6 ± 1.8 g/L, respectively, which were about 4 and 1.3 times higher than those of the LM. It was estimated that there was more biogas production from the FW than the LM. However, the NH₄⁺ concentration was 0.2 ± 0.1 g/L, which represented about 3% of the TN concentration, and it was estimated that the inhibition of the anaerobic digestion did not occur by NH₄⁺.

A high concentration of Na⁺ would lead to high osmotic pressure, resulting in the loss of the intracellular water of microorganisms [20]. Other researchers have reported that Na⁺ caused an inhibition of microorganism activity and a decrease in the degradation rates of VFAs. It has been reported that methanogenic activity and biogas production are inhibited in the concentration range from 5 to 10 g Na⁺/L [21]. However, it was determined in this study that there was no inhibition of anaerobic digestion, because the Na⁺ concentration was 3.8±0.06 g Na⁺/L.

The TCODcr concentration of the LM was lower than that of the FW, and the NH₄⁺ concentration was about 14 times higher, which represents about 93% of the TN concentration. Therefore, the anaerobic digestion of the LM could have been inhibited by NH₄⁺, lowering the efficiency of the biogas production. The NH₄⁺ concentration must be controlled because anaerobic microorganisms are affected by high concentrations of NH₄⁺.

Table 2. Characteristics of trace elements in food waste and livestock manure.

Trace Element	Value (mg/L, Average ± S.D. a)		Theoretical Range (mg/L as Ion)
	FW	LM
Mg	239.0 ± 0.75	78.0 ± 0.66	39
Fe	36.1 ± 0.63	8.5 ± 0.13	5.8
Co	0.03 ± 0.02	0.08 ± 0.01	2.5
Ni	0.17 ± 0.06	0.19 ± 0.06	0.13
Zn	6.71 ± 0.04	1.21 ± 0.08	0.25
B	2.60 ± 0.06	3.03 ± 0.04	0.09
Cu	0.36 ± 0.01	0.06 ± 0.04	0.19
Mn	0.06 ± 0.01	0.94 ± 0.06	0.15
Se	0.09 ± 0.03	0.04 ± 0.03	0.21
Mo	0.08 ± 0.02	0.02 ± 0.01	0.20
V	0.02 ± 0.01	0.02 ± 0.01	0.21

^a S.D.: standard deviation.

shows the characteristics of the trace elements in the FW and LM. These elements are important operational parameters for increasing the microbial activity in anaerobic digestion. The addition of trace elements increases the functions of enzymes and the metabolisms of anaerobic microorganisms. Many studies have determined that iron (Fe), nickel (Ni), cobalt (Co), and molybdenum (Mo) ions are essential for increasing the activity of methanogenic microorganisms [22]. In this study, the trace elements of Co²⁺, Se⁶⁺, Mo⁶⁺, and V⁵⁺ ions in the FW and LM were measured to be significantly low compared to those in the study by Speece et al. (2006) [23]. However, it was determined that these elements were sufficient in anaerobic microorganisms for the anaerobic digestion of organic substances, as the anaerobic microorganisms for the BMP and ATA tests have continuously adapted to the FW.

The theoretical methane yield was evaluated through the elemental analysis of the FW and LM, which can be used to estimate the methane yield for the appropriate mixing ratio of FW to LM. The theoretical methane yields of the FW and LM were 0.55 and 0.57 L CH₄/g VS, respectively. The carbon-to-nitrogen (C/N) ratios were 12.4 and 8.8, respectively [24]. Therefore, it is necessary to use an appropriate mixing ratio for the stable co-digestion of FW and LM.

3.2. Degradable COD Concentration and Methane Yield

The degradable COD concentration and methane yield were evaluated using BMP tests. The BMP test is a general method to determine the methane yield and biodegradability of organic waste and wastewater [25]. Although the generation of biogas and methane could be predicted by physical and chemical analyses of raw materials, the BMP test realistically measures anaerobic biodegradability, and it is important to analyze the concentrations of organic pollutants in wastewater that can be converted to anaerobic CH₄.

Table 3. The anaerobically degradable COD concentrations and methane yields of food waste and livestock manure.

Parameter	Unit	Value (Average ± S.D. a)
		FW		LM
		38 °C	55 °C	38 °C	55 °C
Degradable COD	g COD/L	190.7 ± 11.2	132.7 ± 9.6	29.2 ± 1.5	28.5 ±1.3
Ratio of degradable COD to TCOD concentrations in waste	%	86.7	60.3	54.2	52.7
Ultimate methane yield	L CH4/g COD	0.25 ± 0.02	0.28 ± 0.02	0.06 ± 0.01	0.09 ± 0.03
	L CH4/g VS	0.49 ± 0.03	0.46 ± 0.04	0.42 ± 0.06	0.38 ± 0.04
Theoretical methane yield	L CH4/g VS	0.55 ± 0.01	0.55 ± 0.01	0.57 ± 0.01	0.57 ± 0.01
Ratio of ultimate methane yield to theoretical methane yield in waste	%	89.1	83.6	73.7	66.7

^a S.D.: standard deviation.

shows the anaerobically degradable COD concentrations and methane yields of the FW and LM. These concentrations in the FW were 190.7 ± 11.2 and 132.7 ± 9.6 g COD/L, respectively, according to the temperature conditions, which represented 86.7 and 60.3% of the TCOD in the FW, respectively. This concentration at 38 °C was 58 g COD/L higher than that at 55 °C. The ultimate methane yields were 89.1 and 83.6% of the theoretical methane yields, respectively. The ultimate methane yield at 38 °C was 5.5% higher than that at 55 °C. The concentrations in the LM were 29.2 ± 1.5 and 28.5 ± 1.3 g COD/L, respectively, according to the temperature conditions, which represented 54.2 and 52.7% of the TCOD in the LM, respectively. The concentration at 38 °C was 0.7 g COD/L higher than that at 55 °C. The ultimate methane yields were 79.7 and 66.7% of the theoretical methane yields, respectively. The ultimate methane yield at 38 °C was 7.0% higher than that at 55 °C. Because anaerobic microbial activity is more sensitive at 55 than at 38 °C, in the case of the presence of an inhibitor factor, the methane yield efficiency at 38 °C was higher compared to that at 55 °C.

Other researchers have estimated that an increase in the temperature resulted in a reduction in the methane yield because NH₄⁺ is converted to NH₃, which microbes are more sensitive to, with increasing temperature [3]. In addition, lignocellulosic material is an essential component of LM and constitutes from about 40 to 60% of the manure. These materials are recalcitrant and only partly converted to biogas during anaerobic digestion, of which 20~30% of the volatile solids will be converted to biogas [26]. As a result, to stabilize the anaerobic digestion of the LM, it was estimated that it is necessary to achieve the co-digestion of FW to increase the degradable organic matter and dilute the inhibitory substances.

3.3. ATA Test

3.3.1. Initial Toxicity of Organic Matter Concentration in FW

Figure 2 and

Table 4. Comparison of the gas production rates and methane yields of food waste for various amounts of food waste input according to temperature conditions.

Volume of FW	Input COD Concentration	Value (Average ± S.D. a)
		Gas Production Rate for 10 Days		Ultimate Methane Yield
mL	g/L	mL/g VS/d		mL CH4/g CODadded		mL CH4/g VSadded
		38 °C	55 °C	38 °C	55 °C	38 °C	55 °C
5	1.3	124.8 ± 9.8	124.8 ± 11.3	318.2 ± 3.8	262.1 ± 4.4	568.3 ± 4.1	468.0 ± 1.8
10	2.5	51.7 ± 7.7	62.9 ± 7.8	211.8 ± 4.7	152.3 ± 3.7	378.2 ± 2.6	271.9 ± 1.1
15	3.8	22.0 ± 4.1	24.2 ± 5.5	28.9 ± 1.2	18.3 ± 2.4	51.6 ± 0.8	32.7 ± 0.6
20	5.0	17.3 ± 1.5	19.2 ± 2.4	24.2 ± 1.5	15.5 ± 2.1	43.2 ± 1.1	27.6 ± 0.6

^a S.D.: standard deviation.

show the biogas production, gas production rate, and methane yield for various amounts of FW input according to the temperature conditions. The ATA test was performed to determine the effect of high concentrations of organic matter on the initial anaerobic digestion. The gas production rate and methane yield decreased with the increase in the amount of the FW input. It was shown that the initial gas production rate decreased from 124.8 ± 9.8 to 17.3 ± 1.5 and 19.2 ± 2.4 mL/g VS/day, respectively, at 38 and 55 °C with an increase in the amount of the FW input from 5 to 20 mL. Therefore, the initial gas production rate decreased by 86.1 and 84.6%, respectively, confirming that the efficiency of the initial gas generation rate was low at 38 °C.

In the case of the methane yield, it was shown that the methane yield at 38 °C was decreased from 318.2 ± 3.9 mL CH₄/g COD_added and 568.3 ± 4.1 mL CH₄/g VS_added to 24.2 ± 1.5 mL CH₄/g COD_added and 43.2 ± 1.1 mL CH₄/g VS_added, respectively, and the methane yield at 55 °C was decreased from 262.1 ± 4.4 mL CH₄/g COD_added and 468.0 ± 1.8 mL CH₄/g VS_added to 15.5 ± 2.1 mL CH₄/g COD_added and 27.6 ± 0.6 mL CH₄/g VS_added, respectively, with increasing the input amount of the FW from 5 mL to 20 mL. Therefore, the methane yield was shown to decrease more than 92.4%. It was confirmed that the efficiency of the methane yield at 55 °C was lower than that at 38 °C, contrasting with the gas production rate results. It was determined that the initial gas production rate increased because of accelerated biochemical reactions at 55 °C. Many studies have determined that the biogas production and methane yield at 55 °C were higher compared to those at 38 °C because of the increase in the microbial activity and greater organic stability [11,27]. However, it was estimated in this study that the ultimate methane yield was decreased by inhibition factors, such as the accumulation of organic acids decomposed from organic matter and NH₄⁺ and the FA from nitrogen substances, during anaerobic digestion.

Complex organic matters, such as FW, are converted to various short-chain VFAs, such as acetic, propionic, lactic, and butyric acids. They are the dominant intermediates of methane production in anaerobic digestion. However, the accumulation of VFAs is considered as a major operational factor during anaerobic digestion. The accumulation of VFAs could occur because the reactor has a high OLR or an imbalance in the C/N ratio. These conditions decrease methane production because of the sharp drop in the pH [28].

Furthermore, other researchers have determined that the accumulated inhibitory substances by hydrolysis followed the acidification of the digester and inhibition of the methanogens in thermophilic anaerobic digestion compared to mesophilic conditions [29,30]. Therefore, it is considered as desirable to perform the anaerobic digestion of the FW under mesophilic temperature conditions because of the inhibitory factors being overcome and because of the high efficiency of the ultimate methane yield, even if the initial methane production rate was low.

3.3.2. The Effect of NH₄⁺ Inhibition on the Initial Biogas Production Rate

Figure 3 shows the initial biogas production rates at various TN concentrations in FW as a substrate according to the temperature conditions. The initial gas production rate decreased at both 38 and 55 °C with increasing the TN concentration. It was shown that the initial gas production rate at 55 °C was lower than that at 38 °C. It was also shown that the initial gas production rate decreased from 59.2 ± 10.1 to 4.2 ± 0.1 mL/day and from 57.8 ± 8.1 to 1.2 ± 0.1 mL/day, respectively, at 38 and 55 °C with an increase in the TN concentration. The initial gas production rates decreased on average by 33.1 and 43.5%, respectively, and the methane yields decreased on average by 50.1 and 72.1%, respectively, with a step-by-step increase in the TN concentration at 38 and 55 °C.

Ammonia nitrogen could be present in the form of NH₄⁺ or FA, which is dependent on the pH and temperature. Many studies have confirmed various inhibition ranges of NH₄⁺ or FA for anaerobic microbial activity [3,31].

FA has been reported to be the main inhibition factor because of its high penetrability through the bacterial cell membrane [32]. The FA that enters the cell is converted to ammonium, which leads to a change in the pH, absorbing protons in the process. Methanogen microbes allow for the hydrogen ions to diffuse in from the surroundings and efflux the potassium ions, which must then consume energy in proton rebalancing. Therefore, the inhibition of specific enzyme reactions occurs because of the weak microbial activity due to the lack of intracellular potassium ions [33,34]. Studies have confirmed the inhibition of anaerobic microbial activity because of the outflow of potassium ions upon increasing the ammonia concentration [35].

Table 5. Comparison of the gas production rates, methane yields, and NH₄⁺ and FA concentrations according to the TN concentration and pH and the temperature in the serum bottle.

TN Concentration	Value (Average ± S.D. a)						FA Concentration		pH in Serum Bottle
	Gas Production Rate for 6 Days		Ultimate Methane Yield
mg/L	mL/Day		mL CH4/g CODadded		mL CH4/g VSadded		mg/L		-
	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C
500	59.2 ± 10.1	57.8 ± 8.1	236.7 ± 7.7	240.6 ± 4.6	338.1 ± 6.1	343.7 ± 8.4	9.1	20.3	7.19	7.39
2000	46.8 ± 9.8	33.8 ± 7.6	228.3 ± 5.8	228.0 ± 4.8	326.1 ± 5.1	325.7 ± 3.1	235.9	528.9	7.45	7.59
3000	25.3 ± 8.8	22.5 ± 4.5	87.3 ± 6.1	100.8 ± 1.5	124.7 ± 4.8	144.0 ± 2.2	487.0	1032.8	7.27	7.46
4000	17.7 ± 4.5	15.1 ± 1.8	72.9 ± 0.5	88.2 ± 0.4	104.1 ± 3.1	126.0 ± 1.5	742.7	1535.9	7.42	7.45
5000	15.2 ± 1.1	11.6 ± 0.1	48.6 ± 0.1	95.1 ± 0.2	69.4 ± 0.8	135.9 ± 0.8	1040.9	2096.9	7.47	7.57
6000	5.5 ± 0.5	2.3 ± 0.1	8.7 ± 0.1	6.9 ± 0.1	12.4 ± 0.8	9.9 ± 0.5	1361.6	2684.4	7.55	7.72
7000	4.2 ± 0.1	1.2 ± 0.1	1.1 ± 0.1	0.8 ± 0.1	1.8 ± 0.2	1.2 ± 0.1	1996.4	3656.4	7.84	7.87

^a S.D.: standard deviation.

shows the gas production rates, methane yields, and NH₄⁺ and FA concentrations according to the TN concentration and pH value and temperature in the serum bottle. The FA concentration was increased from 9.1 mg/L to 1996.4 mg/L at 38 °C and from 20.3 mg/L to 3656.4 mg/L at 55 °C. McCarty and McKinney found out that an FA concentration of 150 mg/L was inhibitory to anaerobic digestion [36]. In another study, inhibition was confirmed at 3000 mg/L of total ammonia, with livestock manure as the substrate [37,38]. The gas production rate and methane yield were decreased rapidly at a TN concentration of 2000 mg/L in this study. The FA concentrations were 235.9 and 528.9 mg/L, respectively, at a TN concentration of 2000 mg/L at 38 °C and 55 °C. Furthermore, the FA concentrations were 487.0 and 1032.8 mg/L at a TN concentration of 3000 mg, according to the temperature condition, which increased more than two times compared to those at a TN concentration of 2000 mg/L. The methane yields at a TN concentration of 3000 mg/L were 87.3 ± 6.1 and 100.8 ± 1.5 mL CH₄/g COD_added and 124.7 ± 4.8 and 144.0 ± 2.2 mL CH₄/g VS_added. It showed a decrease more than 55.7% compared to the methane yields at a TN concentration of 2000 mg/L. This result was similar to the result in which the methane yield decreased by up to 65% at an FA concentration of 1000 mg/L with chicken manure slurry as the substrate. In addition, the methane yield at 55 °C was lower than that at 38 °C because of the FA inhibition [3]. These results confirmed that the total nitrogen concentration, digestion temperature, pH, etc. are important factors affecting the anaerobic digestion efficiency during the anaerobic digestion of wastes, such as FW and LM. Therefore, the nitrogen content must be controlled to not inhibit anaerobic microorganisms during anaerobic digestion.

3.4. Modeling Results of First-Order Reaction and Modified Gompertz Model

Figure 4 and

Table 6. Comparison of the methane yields and kinetic constants for mixture ratios of food waste to livestock manure according to temperature conditions.

Mixture Ratio	Correlation Coefficient		k		Value (Average ± S.D.a)
					Ultimate Methane Yield				Theoretical Methane Yield
FW:LM (Vol:Vol)	-		d−1		L CH4/g CODadded		L CH4/g VSadded		L CH4/g VSadded
	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C	-
10:1	0.9049	0.9732	0.082	0.074	0.14 ± 0.07	0.23 ± 0.06	0.28 ± 0.07	0.47 ± 0.07	0.56 ± 0.02
5:1	0.9711	0.9711	0.080	0.078	0.18 ± 0.07	0.25 ± 0.06	0.35 ± 0.07	0.50 ± 0.06	0.56 ± 0.01
2:1	0.9118	0.9118	0.074	0.073	0.12 ± 0.04	0.21 ± 0.04	0.24 ± 0.06	0.43 ± 0.04	0.54 ± 0.02
FW	0.9617	0.9875	0.067	0.063	0.20 ± 0.09	0.28 ± 0.07	0.32 ± 0.08	0.46 ± 0.05	0.55 ± 0.01
LM	0.9901	0.9550	0.068	0.073	0.05 ± 0.01	0.09 ± 0.02	0.19 ± 0.04	0.38 ± 0.04	0.57 ± 0.01

^a S.D.: standard deviation.

show the methane yields and kinetic constants for mixture ratios of FW and LM according to the temperature condition. The theoretical methane yields of the FW and LM were determined as 0.55 ± 0.01 and 0.57 ± 0.01 L CH₄/g VS, respectively. The ultimate methane yields showed the highest values, 0.35 ± 0.07 and 0.50 ± 0.06 L CH₄/g VS, which were 63% and 89% of the theoretical methane yield, at a mixture ratio of 5:1 of FW and LM, depending on temperature conditions. The increase in the microbial activity under thermophilic conditions may enhance the hydrolysis efficiency, which improves the methanogenic performance [11,27].

The ultimate methane yields of only FW or LM were lower than or similar to that of a mixture of FW and LM. It was estimated that the anaerobic digestion was inhibited by the initial high concentrations of the organic matter and salt in the FW and the NH₄⁺ and FA in the LM during the anaerobic digestion of the organic matter alone.

The kinetic constant showed the highest values of 0.082 and 0.080 at mixture ratios of 10:1 and 5:1 under mesophilic conditions. It was estimated that anaerobic microorganisms reacted more sensitively to inhibitory substances upon an increase in the mixing ratio of the FW under thermophilic compared to mesophilic conditions.

Table 7. Results of the modified Gompertz model analysis according to temperature conditions.

Mixture Ratio	Bu		P		Rm		λ
FW:LM (Vol:Vol)	L CH4/g VSadded		mL		mL/Day		Days
-	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C	38 °C	55 °C
10:1	0.20	0.30	89.41	78.62	8.74	10.74	2.07	1.98
5:1	0.23	0.33	100.84	95.43	10.23	11.35	1.44	1.10
2:1	0.20	0.27	80.44	68.48	8.64	9.14	0.89	0.56
FW	0.24	0.35	108.07	91.25	11.13	12.23	1.61	1.64
LM	0.15	0.14	63.61	28.25	6.20	4.35	1.81	0.43

shows the results of the modified Gompertz model analysis for various mixture ratios of FW to LM according to the temperature conditions. The ultimate methane potential (B_u), maximum methane production (P), and maximum methane production rate (R_m) showed the highest values at a mixture ratio of 5:1 in the case for mixing FW and LM. However, the lag-growth-phase time (λ) increased with an increase in the mixture ratio of the FW. This also confirms similar results in other experiments using the Gompertz model during the anaerobic co-digestion of FW and LM [18,39]. According to the results of the first-order reaction and modified Gompertz model fitting, combining FW and LM in the anaerobic digestion process has the potential to increase the methane production rate and methane yield.

4. Conclusions

The characteristics of FW and LM and the anaerobic digestion potential were analyzed to evaluate the anaerobic co-digestion of FW and LM according to temperature conditions. Biogas production was investigated at each inhibition factor concentration that could inhibit the anaerobic microorganisms. As a result, the degradable COD concentration and methane yield of the FW were 6.5 and 4.6 times higher and 1.1 and 1.2 times higher compared to those of the LM according to the temperature conditions, respectively. In addition, the methane yield at a TN concentration of 3000 mg/L decreased more than 55.7% compared to that at a TN concentration of 2000 mg/L. Therefore, the inhibition concentration of the NH₄⁺ was determined to be more than 2000 mg/L of TN during anaerobic digestion.

The efficiency of the anaerobic co-digestion of FW and LM was evaluated using first-order and modified Gompertz model fittings based on the amount of biogas production. As a result of this modeling, the decomposition rate of the substrate and the maximum specific methane production rate (R_m) showed maxima at 0.080 and 0.078 mL/day and at 10.23 and 11.35 mL/day, respectively, at a mixing ratio of 5:1. The anaerobic co-digestion of FW and LM can increase the efficiency of the biogas production amount and rate compared to the anaerobic digestions of FW and LM separately. In addition, the lag-phase time (λ) can be reduced by increasing the activity of anaerobic microorganisms through the dilution of inhibitors at an appropriate C/N ratio. As a result, the highest efficiency was achieved at a mixing ratio of FW to LM of 5:1. The highest methane yield compared to the theoretical methane yield and the maximum specific methane production rate were shown by the Gompertz model. Therefore, it was determined that the reduction in inhibitory substances and the improvement in methane production could be achieved through anaerobic co-digestion rather than the anaerobic digestions of FW and LM alone.