Enhanced Energy Recovery from Food Waste by Co-Production of Bioethanol and Biomethane Process

Abstract

The primary objective of this research is to study ways to increase the potential of energy production from food waste by co-production of bioethanol and biomethane. In the first step, the food waste was hydrolysed with an enzyme at different concentrations. By increasing the concentration of enzyme, the amount of reducing sugar produced increased, reaching a maximum amount of 0.49 g/g food waste. After 120 h of fermentation with Saccharomyces cerevisiae, nearly all reducing sugars in the hydrolysate were converted to ethanol, yielding 0.43–0.50 g ethanol/g reducing sugar, or 84.3–99.6% of theoretical yield. The solid residue from fermentation was subsequently subjected to anaerobic digestion, allowing the production of biomethane, which reached a maximum yield of 264.53 ± 2.3 mL/g VS. This results in a gross energy output of 9.57 GJ, which is considered a nearly 58% increase in total energy obtained, compared to ethanol production alone. This study shows that food waste is a raw material with high energy production potential that could be further developed into a promising energy source. Not only does this benefit energy production, but it also lowers the cost of food waste disposal, reduces greenhouse gas emissions, and is a sustainable energy production approach.

1. Introduction

With the outbreak of the COVID-19 virus since the beginning of 2020, the global economic growth rate has stalled and resulted in a sharp drop in energy demand. The International Energy Agency, IEA, forecasts that the severity of the epidemic will subside and economic growth will gradually recover in 2021, bringing back energy demand, which may increase by leaps and bounds to compensate for the contraction from the recent situation. In particular, despite the COVID-19 pandemic, demand for renewable energy continues to grow (approximately 3% in 2020) [1]. This implies that the world is becoming more conscious of the necessity of renewable energy use, and it pushes academics to investigate low-cost alternative and environmentally acceptable energy sources that contribute to sustainable development.

Bioethanol production from waste, such as organic fraction municipal waste, and agricultural waste, has consistently been one of the most popular alternative energy production pathways. In comparison to fossil fuels, bioethanol emits considerably lower greenhouse gases and thus receives widespread support as a vehicle fuel source. By mixing it in various proportions with gasoline, it transforms into gasohol, which can be used immediately in internal combustion engines without requiring further engine modifications.

Food waste (FW) is classified as a low-cost, high-potency second-generation feedstock due to its main constituents of biodegradable organic compounds (such as carbohydrates, proteins, and fats). It can be highly bioavailable for the production of various forms of bioenergy, such as biohydrogen [2,3], biogas [4,5], biodiesel [6,7,8], biobutanol [9], andbiohythane [10,11], as well as ethanol. Furthermore, converting FW into energy is a solution to the environmental crisis caused by the current amount of FW, which is steadily increasing as the economy and population grow. Globally, 931 million tons of FW were produced in 2019, with approximately 30% of food produced being discarded as waste [12]. The capacity to dispose of this FW is significantly less than the rate of production, resulting in a municipal waste overflow problem. FW contains a high level of moisture; therefore, it is difficult to transport, consumes more fuel when incinerated, and produces wastewater, odors, and greenhouse gases when disposed of in a landfill. As a result, eliminating FW is more challenging and more costly than other types of waste.

Many studies on the production of ethanol from FW and other organic waste have been conducted in a variety of areas. In particular, the effects of the composition of FW, effects of pretreatment prior to the fermentation process, types of enzymes, types of microorganisms used in ethanol production, and suitable operational conditions were investigated with the main goal of increasing ethanol productivity [13,14,15,16,17,18,19]. Even so, commercial production of bioethanol from second-generation feedstocks is being questioned for its cost-effectiveness in terms of both economic feasibility and energy efficiency. Previous studies have reported that the cost-effectiveness of ethanol production from second-generation biomass has a very low Energy Return on Investment (EROI) (approximately 0.8–1.6) compared to fossil fuels (18–45) [20]. As a result, there has been an increase in a number of studies conducted on the process of co-production in which multiple products are produced. In particular, the co-production process in this context refers to the production of other fuels or valuable substances as a by-product of ethanol production, which can be sold to increase revenue or used as fuel to reduce energy costs in ethanol production. As a result, the economic competitiveness of second-generation bioethanol production is enhanced [21]. In this regard, previous research has demonstrated the production of bioethanol coupled with biogas from red oak [22], sugarcane bagasse [23], wheat straw [24], corn stover [25], spruce wood [26], and switchgrass [27]. However, research on co-production processes that utilize FW as a feedstock remains limited.

This research examines the production of bioethanol and biomethane from FW, with an emphasis on a simple approach in which the smallest amounts of chemicals and energy are used. This began with a mechanical pretreatment of FW to reduce their size without the use of heat or chemicals. After hydrolysis with enzyme, the liquid fraction of the hydrolysate was separated to produce ethanol, while the solid fraction was subjected to anaerobic digestion to produce biogas (Scenario 1). The energy yield was then compared to that obtained by first producing ethanol from the entire fraction of hydrolysate and then biogas (Scenario 2). Eventually, a production scenario that achieved the highest productivity and gross energy output was suggested.

2. Materials and Methods

2.1. Raw Material

FW used in this research was collected from the cafeteria of King Mongkut’s University of Technology North Bangkok, Rayong campus. To mitigate variability due to differences in starting raw material, FW was collected continuously for two weeks, allowing for the use of a single lot of samples throughout the trial. After, non-biodegradable components such as fishbones, chicken bones, packaging fragments, and so forth were removed. The remainder of the sample was reduced in size using a food processor and then stored at −20 °C for further use. The total solids (TS) and volatile solids (VS) of FW were determined using a method proposed by Sluiter et al. (2008) [28].

2.2. Enzymatic Hydrolysis

In the enzymatic hydrolysis stage, FW was digested using α-amylase from Aspergillus oryzae with a specific activity of 30 U/mg (Sigma–Aldrich, St. Louis, MI, USA) in 500-mL Erlenmeyer flasks with 200 mL working volumes. The concentration of the original FW was 10% (w/v). To determine the optimal enzyme concentration and duration of hydrolysis that led to the highest reducing sugar (RS) content, the enzyme concentration was varied at 1, 3, 4, and 5% w/w (g enzyme/g dry FW), and hydrolysis was carried out for 1, 2, 3, 4, 5, and 9 h at 60 °C with a stirring rate of 150 rpm. As the pH of the FW was in the optimal range for enzymatic activity (4.0–6.5), pH adjustment was negligible.

2.3. Bioethanol and Biomethane Production

The main objective of this research is to study the energy production methods from FW that provide the highest gross energy output from bioethanol and biomethane production. Following the hydrolysis, the experiment is divided into two scenarios (Figure 1):

• Scenario 1: In the ethanol fermentation, only the liquid fractions from the hydrolysis were used, and the solid residues were separated for use in the production of biomethane.
• Scenario 2: The entire hydrolysate was used in ethanol fermentation, followed by the extraction of fermented solid residues for additional anaerobic digestion.

The process of producing bioethanol began with the preparation of yeast. Saccharomyces cerevisiae (commercial dry baker’s yeast purchased at a local store) was dissolved in sterile water at a concentration of 10 g/L and stored at 4 °C prior to use without cultivation [29]. The prepared solution was placed in 500-mL Erlenmeyer flasks (with working volumes of 200 mL), the pH was adjusted to 5.5, and S. cerevisiae yeast solution was added at the concentration of 5.0, 10.0, and 15.0% (v/v). Fermentation took place at 35 °C for 120 h with a stirring rate of 150 rpm. At the end of fermentation, the liquid was separated by filtration. The clear solution was then analysed for ethanol content by gas chromatography (Flame Ionization (FID), Bruker Scion 456-GC) following the method by Cutaia (1984) [30].

In the anaerobic digestion process to produce biomethane, the experiment was carried out in batches using 5 L volumetric flasks with working volumes of 2 L. The substrate was prepared by mixing the fermentation solid residues (FSR), or hydrolysis solid residues (HSR) with 500 mL DI water to a concentration of 20% (w/v). Then, inoculum (biogas digester sludge from swine farm wastewater) was added at a substrate to inoculum ratio of 1:3, followed by a 5-min purge with N₂ before closing the lid to allow anaerobic conditions. During the 80-day digestion at 37 °C, the produced biomethane was measured with a mass flow meter (F-111B-100-RAD-22-K, Bronkhorst, The Netherlands) and the biogas composition was determined using a biogas analyser (Biogas 5000, Geotech, England).

3. Results and Discussion

3.1. Food Waste Characteristic

The properties of the raw materials to be fed into the production process are essential prerequisites in determining the optimum conditions and estimating the expected yield. The composition of FW varies according to eating culture. Asian foods are similar in that they are primarily composed of starch. The FW used in the study was collected from university cafeterias, almost all of which sell local Asian food based on starchy ingredients such as rice and noodles, which contributed up to 51.03% of the total carbon content, 2.11% of nitrogen, and a very high moisture content of 89.01% (

Table 1. Characteristics of the raw material.

Component of FW	Fraction
Moisture content (%)	89.01 ± 0.61
Total solids 1, TS (%)	10.99 ± 0.61
Volatile solids 1, VS (%)	10.59 ± 0.58
Ash 1 (%)	0.40 ± 0.03
Total carbon 2 (%)	51.03 ± 0.75
Total nitrogen 2 (%)	2.11 ± 0.21

¹ Calculated on wet basis. ² Calculated on dry basis.

). This composition is similar to FW collected from canteens in Korea [31,32,33], China [34,35], and Japanese households [36], with total carbon content ranging between 40% and 54% (dry basis) and moisture content ranging between 70% and 90%. According to recent research, the amount of ethanol produced is proportional to the carbon content of the raw material. That is, the higher the carbon content of the raw material, the greater the productivity, and the higher the moisture content, the less water is required during the process [37]. Additionally, the FW collected for this study possessed a VS of 10.59% and an ash content of 0.4%. The VS and ash contents could be used to estimate the amount of biofuel to be produced. This high VS and low ash raw material demonstrate the presence of a significant amount of organic matter that microorganisms can consume during the biological process. Based on the initial composition, it was certain that the collected FW was a potential feedstock for biofuel production due to the abundant nutrients required by microbes to convert to valuable chemicals, particularly bioethanol and biogas.

3.2. Enzymatic Hydrolysis

Hydrolysis is a process that converts carbohydrates in raw materials into sugars, which are then used as food by microorganisms to produce ethanol in the following step. Thus, this first step is critical as it determines the overall efficiency of the biofuel production process. In past studies of bioethanol production from FW, thermochemical pretreatment with acid or base, followed by enzymatic hydrolysis, is often used to ensure the complete conversion of carbohydrates to sugars. However, research indicates that severe conditions are not always beneficial. For instance, high-temperature treatments cause sugars to undergo partial degradation [38] and the formation of microbial inhibitors can occur as a result of acid treatments [39]. Enzymatic hydrolysis is thus the most widely used method as it performs under mild conditions, does not produce inhibitors that interfere with the subsequent fermentation [40], and allows for fully biological processes. In this study, FW was hydrolysed with α-amylase without pretreatment other than size reduction. The amount of RS obtained from FW degradation by the α-amylase enzyme is depicted in Figure 2, which clearly indicates the increase in RS yield with the increasing enzyme dosage and the duration of the hydrolysis. After 9 h of hydrolysis, the highest RS yields were 17.90, 27.08, and 49.45 g/L, corresponding to 0.18, 0.27, and 0.49 g/g FW, when 1.0%, 3.0%, and 5.0% of enzyme was introduced, respectively. This range of RS yields was similar to that reported by Kim et al. (2011) [41], who obtained RS yields of 0.436 and 0.627 g/g TS from FW digestion with glucoamylase and carbohydrase, respectively, and by Han et al. (2020) [42], who obtained 0.784 g RS/g FW from waste hamburger hydrolysis with α-amylase. It is noted that the amount of RS produced rapidly increased during the initial stages of hydrolysis and gradually increased over time until it reached a stable state. This is consistent with the findings of Han et al. (2019) [43], who found a slight increase in the amount of RS obtained from waste cake after 80 min of hydrolysis with α-amylase. Likewise, a study conducted by Hong et al. (2011) [31] on the enzymatic hydrolysis of FW demonstrated that the glucose content remained stable after 10 h. Additionally, increasing the enzyme concentration resulted in an increase in glucose content, which reached its maximum at 600 g/kg FW when 4 mL of enzyme was introduced. However, in this study, the enzyme dose of 0.5 mL/100 g FW was chosen as the optimal condition for further processing. This was because increasing the enzyme dose from 0.5 mL to 1, 2, and 4 mL increased the amount of glucose obtained only slightly. In other words, almost all carbohydrate molecules had already been digested and thus increasing the amount of enzyme further from this point did not result in a significant increase in glucose. This is consistent with the findings of Kim et al. (2011) [41], who found that when the enzyme content was increased from 5 to 10%, the amount of glucose produced remained similar.

3.3. Ethanol Production from Food Waste Hydrolysate

The hydrolysate was processed in two ways after being hydrolysed by the enzyme. Scenario 1 divided FW hydrolysate into two fractions: HLF for ethanol fermentation and HSR for biomethane production via anaerobic digestion. In Scenario 2, the hydrolysate was completely fermented to produce ethanol. After the ethanol was separated, the remainder was used to produce biomethane. In both scenarios, the fermentation was set to take place under the same conditions. That was, at 35 °C for 120 h with various concentrations of S. cerevisiae. It was found from the experiment that the amount of ethanol produced from HLF (Scenario 1) ranged from 21.26–23.24 g/L (Figure 3), which only slightly increased as the yeast content increased from 5 to 10 and 15% (v/v). This is because most of the RS had already been converted to ethanol, considered based on the theory where 100 g of glucose can be converted to 51 g of ethanol and 49 g of CO₂ by a fermentation process [44]. The amount of ethanol produced from HLF at yeast concentrations of 5, 10, and 15% (v/v) resulted in ethanol yields of 0.43, 0.46, and 0.45 g/g RS, accounting for 84.3, 90.2, and 88.2% of theoretical yield, respectively. Unlike Scenario 2, the ethanol produced from whole hydrolysate increased from 20.77 to 24.77, and 26.23 g/L as yeast dosage increased from 5 to 10, and 15% (v/v), respectively. The reason why the ethanol content produced from whole hydrolysate was higher than that produced from HLF may be due to the greater amount of RS as there was no loss during the filtration process as in Scenario 1, as well as remaining sugars in the solid fraction. According to recent ethanol production studies, FW ethanol yields range between 0.4 and 0.5 g/g RS [41,45,46]. Moreno et al. (2021) [47] reported that the ethanol yield from unpretreated organic fraction municipal waste was highest at 80% of the theoretical yield. Kiran et al. (2015) [46] produced 0.5 g/g glucose from waste cake ethanol, accounting for 98% of the theoretical yield after 32 h-fermentation. These are consistent with Han et al. (2019) [43], where waste cake was used as a substrate in ethanol production, yielding as high as 1.13 g ethanol/g RS as the waste cake was readily biodegradable and contained other constituents that promote yeast activity. When the amount of bioethanol produced from FW is considered, this research produced ethanol with a maximum yield of 0.22 and 0.25 g ethanol/g dry FW from the liquid fraction (Scenario 1) and from whole hydrolysate (Scenario 2), respectively. At this point, it can be seen that the yields of ethanol produced from untreated FW are comparable to those produced from pretreated FW. Considering the costs that can be saved by eliminating pretreatment or detoxification procedures, such as costs of chemicals, energy, and investment, this approach can bring the overall cost of ethanol production in line with conventional processes.

3.4. Biomethane Production

The anaerobic digestion process was chosen for the production of biomethane from the solid residues from enzymatic hydrolysis and fermentation of FW. This is because the process has been shown to be one of the most efficient methods for extraction and conversion of remaining vital substances into energy. In this study, the anaerobic digestion process was carried out at 35 °C using sludge from an anaerobic digester from a swine farm as an inoculum. Figure 4 shows the cumulative methane content and production rate over an 80-day period. Methane produced from FSR (Scenario 2) was the highest, 264.53 ± 2.3 mL/g VS, followed by HSR (Scenario 1), 224.29 ± 1.8 mL/g VS and raw FW (RFW), 215 ± 3.2 mL/g VS. The greater biomethane production potential of this FSR may be due to the fact that FW had undergone two stages of degradation, enzymatic hydrolysis, and ethanol fermentation, leaving the remaining organic matter with higher solubility which was prompt to be converted to biomethane. Furthermore, considering the methane production rate, the FSR showed a relatively higher methane production rate than the HSR and RFW, especially during the first 30 days where more than 90% of the methane was produced. As seen in the graph of the FSR’s daily methane production rate, methane began to accumulate in the first week of the anaerobic digestion process at a rate of approximately 100–300 mL/d with no lag phase. This is in contrast to HSR and RFW, where methane production began in the second and fifth weeks, respectively. The results of this study are consistent with research by Wu et al. (2015) [44] reporting that pretreatment by fermenting FW to ethanol with alcohol active dry yeast at 35 °C for 24 h prior to anaerobic digestion resulted in a 71.7% increase in methane production yield. Due to the FW conversion to ethanol rather than volatile fatty acids (VFA) during the normal acidification process of anaerobic digestion, the pH of the system remained constant until the methanogens’ activity was inhibited. As a result, biogas production was increased and the lag phase was shortened. This is similar to research conducted by Zhao et al. (2016) [48] who studied the effect of ethanol pre-fermentation on pre-anaerobic digestion pretreatment and found that methane production was increased by 49.6% compared to untreated FW. Furthermore, Refai et al. (2014) [49] investigated the effects of various volatile fatty acids, including acetate and ethanol, on methane formation during anaerobic digestion processes. It was found that by adding acetate and ethanol to the system, methane formation rates increased by 35–126% as a result of the metabolic activity of aceticlastic methanogens being promoted. Similarly, Prasertsan et al. (2021) [50] reported that the addition of approximately 5% of ethanol to the palm oil mill effluent resulted in an increase in bioassimilation of the process. This not only enhanced the amount of methane produced by 2.7 times, but also improved COD removal efficiency.

3.5. Gross Energy Output

Bioethanol and biomethane co-production can increase the total energy yield of FW energy production through bioprocessing. The energy values obtained from the co-production in this study were calculated from the low heating value (LHV) of 26.7 MJ/kg and 35.8 MJ/m³, for bioethanol and biomethane, respectively [24]. Figure 5 illustrates the yield and gross energy output of 1 ton dry FW. In this study, a ton of FW could be converted to 0.49 tons of RS under optimal enzymatic hydrolysis conditions. When the liquid fraction was separated to produce bioethanol and the solid fraction was used to produce biomethane (Scenario 1), the yields were 282 L and 68 m³, respectively, corresponding to a gross energy output of 8.37 GJ, or 262 L of gasoline equivalent. Meanwhile, continuous production (Scenario 2) where all hydrolysate was processed into ethanol fermentation and then fermentation solid residues were used to produce biomethane produced higher total bioethanol and biomethane of 318 L and 80 m³, respectively, corresponding to gross energy output of 9.57 GJ, equivalent to 299 L of gasoline. The gross energy outputs produced in both scenarios are slightly higher than outputs reported in previous research by Karimi and Karimi (2018) [51], where co-production of ethanol and biogas from kitchen and garden wastes yielded a maximum gasoline equivalent of 162.1 L/ton waste. However, these gross energy outputs were comparable to those reported by Papa et al. (2015) [52] in which the energy recovery obtained from the co-production of bioethanol and biomethane from corn stover and switchgrass pretreated with mild ionic liquid was higher than that pretreated with pressurized hot water, ranging from 8.8–10.9 GJ/ton biomass. Similarly, Bondesson et al. (2013) [53] yielded total energy output of 9.2–9.8 GJ/ton of corn stover which was pretreated with 0.2% H₂SO₄. Additionally, when considering the production of ethanol or biomethane alone, the energy obtained from ethanol produced by Scenarios 1 and 2 was 5.94 and 6.70 GJ/ton, respectively. Meanwhile, biomethane production from food waste without ethanol production yielded 215 + 3.2 mL/g VS (Figure 4), equivalent to 2.34 GJ/ton of energy. This result clearly demonstrates the potential for bioethanol and biomethane co-production to increase gross energy output by approximately 1.4 times when compared to bioethanol alone and approximately 4 times when compared to biomethane production alone. This is consistent with a study by Moshi et al. (2015) [54], the co-production of ethanol and methane from cassava peels resulted in a 1.2–1.3-fold increase in energy yield compared to methane-only production, and a 3–4-fold increase compared to ethanol production alone. The results are also in line with the study by Wu et al., 2021 [55], who found that the co-production process of Pennisetum purpureum increased the energy recovery by 98.9% and 53.6% compared to ethanol production and biomethane alone, respectively. However, with the increase in gross energy output, it is still questionable whether the additional steps and associated extra capital and operating costs are worthwhile. In this respect, further research is required.

While co-production of bioethanol and biomethane from FW has resulted in high yields, the majority of research has still been carried out on a laboratory scale. As a result, economics, cost effectiveness, and life cycle assessment remain unexplored areas for researchers to investigate further. According to previous research, the price of ethanol produced from FW varies significantly. Sondhi et al. (2020) [56] determined the minimum ethanol selling price (MESP) of bioethanol produced from microwave-treated kitchen waste at a power level of 90 W for 30 min. The ethanol yield of 0.32 g/g biomass resulted in a MESP of approximately 0.14 USD/L, which is very low compared to the market price of 0.59 USD/L for ethanol. This is significantly different than the estimate of 0.64 USD/L of ethanol by Intan Shafinas Muhammad and Rosentrater (2020) [57]. Additionally, while FW is classified as a zero-cost raw material for energy production, there are hidden costs associated with management, collection, and sorting, which are major impediments to industrial waste utilization. As FW is typically disposed of as municipal waste, which contains both organic and inorganic materials, sorting this waste at the waste disposal site is nearly impossible. Proper management, which includes source sorting, collection planning, public awareness, and government promotion, are therefore critical factors in determining the future success of energy production from FW.

4. Conclusions

FW is primarily composed of carbohydrates, possessing the potential and suitability to be used as a raw material for biofuels. Without any pretreatment other than size reduction, FW in this study was hydrolysed with α-amylase to decompose the primary constituent carbohydrates into fermentable sugars. This, under optimal conditions, yielded the highest amount of RS of 0.49 g/g FW. The study was then divided into two scenarios: (1) the hydrolysis liquid fraction was used for ethanol production and the hydrolysis solid residues were used for biomethane production; and (2) the entire hydrolysate was used for ethanol production followed by biomethane production from fermentation solid residues. The study found that the ethanol production yields obtained from both scenarios were in the range of 0.43–0.5 g/g RS when fermented with S. cerevisiae for 120 h at 35 °C. The maximum ethanol content of 0.25 g/g dry FW was obtained from fermentation of the entire hydrolysate (Scenario 2). Additionally, the fermentation solid residues in Scenario 2 resulted in a greater and faster potential for biomethane production than the hydrolysis solid residues in Scenario 1. However, both Scenarios 1 and 2 show high potential for biofuel production from FW, with gross energy output of 8.37 and 9.57 GJ/ton dry FW, equivalent to 262 and 299 L of gasoline, respectively.

This study demonstrates that co-production of bioethanol and biomethane from food waste is an extremely efficient method of increasing gross energy output when compared to producing either product alone. Additionally, it is a truly sustainable and environmentally friendly method of energy production that reduces GHG emissions in two ways: by reducing food waste and by reducing the use of fossil fuels. The findings of this study serve as an important starting point for demonstrating the feasibility of converting food waste to energy, potentially paving the way for industrial scale production in the future. However, additional research on the economics, investment costs, operation costs, and energy consumption of the entire process is required. This includes the collection and sorting of food waste, as well as the sorting and purification of all products from the subsequent manufacturing process.