Anaerobic co-digestion of canteen food waste and cow dung

ABSTRACT

There is growing global awareness and concern regarding the generation of significant amounts of food waste (FW) due to the issues related to its disposal. Despite being a tested solution for waste management, anaerobic digestion (AD) of FW has not yet been sufficiently explored. In this research, the impact of different mixing ratios of food waste (FW) and cow dung (CD), as well as temperature, on the production of biogas, methane content, and the bio-methane potential (BMP) of the substrates was examined. FW and CD mixed at ratios 0:100, 25:75, 50:50, 75:25, and 100:0 (w/w) were batch incubated at 35 ± 1°C and 55 ± 1°C, using a 2 × 5 full factorial experimental design. The obtained results showed that the temperature and substrate mixing ratios significantly (p < .05) affected biogas yield, methane content, and BMP. Highest biogas yield and BMP were 7151.67 ± 11.55 mL and 401.88 ± 1.98 mLCH4/g VS from AD of FW:CD (75:25) at 55°C. The lowest biogas yield and BMP were 3291.67 ± 81.45 mL and 328.28 ± 4.26 mLCH4/g VS from mono-digestion of FW at 35°C. Overall, co-digestion of FW and CD produced higher biogas yield and BMP than mono-digestion of the substrates at both test temperatures.

KEYWORDS:

• Anaerobic digestion
• biochemical methane potential
• co-digestion
• food waste
• waste management

Introduction

An alarming rise in the production of FW, primarily in institutional (school canteens), industrial (food-processing plants), residential, and commercial sectors (restaurants), has been attributed to population growth and global economic development (Morone et al. 2019). According to Melikoglu, Lin, and Webb (2013), FW production will rise by 44% globally between 2013 and 2025. A study by De Lange and Nahman (2015) estimated that the total amount of FW generated in South Africa was 12.6 × 10⁶ tonnes per year, of which 90% is disposed at landfills. As a result of the large volumes of FW being generated, municipal services are under increasing pressure to use alternative disposal methods (Sebola, Tesfagiorgis, and Muzenda 2014). Food waste decomposes to form methane which contributes to global warming and leachate contaminates groundwater (Costa, Alfaia, and Campos 2019). Converting FW into energy and bio-fertilizer through AD has great potential to reduce the negative impacts of FW disposal in landfills (Capson-Tojo et al. 2016).

Anaerobic digestion is a widely accepted technique used to handle organic waste, sewage sludge, animal manure, and agricultural by-products (Hoang et al. 2022). Anaerobic digestion of FW is increasingly considered viable for recovering energy and nutrients from FW due to its higher organic matter compared to sewage and animal manure (Franqueto, da Silva, and Konig 2020). However, solely digesting food waste (FW) in anaerobic digesters can lead to instability due to the production of ammonia and volatile fatty acids (VFAs), which have inhibitory effects on methanogenic activity (Karki et al. 2021). Co-digestion of FW is considered an effective approach to overcome challenges linked with the sole digestion of FW. When FW is co-digested with other substrates, it results in increased alkalinity, enhanced process stability, reduced formation of inhibitory substances like volatile fatty acids (VFAs), and higher methane yield (Morales-Polo et al. 2018). To improve the C/N ratio and achieve the stabilization of AD, researchers have utilized the strategy of co-digesting food waste (FW) with other substrates such as maize husks (Owamah and Izinyon 2015). Yong et al. (2015) co-digested FW with straw.

Factors that affect the AD process include reactor design, operational conditions, and substrate characteristics (Náthia-Neves et al. 2018). To achieve the highest biogas production, it is crucial to consistently monitor operational variables like temperature, organic loading rate, mixing, and retention time, ensuring that they remain within the optimal ranges (Lohani and Havukainen 2018). Among the several operational conditions, temperature is the most important parameter since it influences the activity rate, process stability, and microbial activity (Gaby, Zamanzadeh, and Horn 2017). Substrate characteristics such as volatile solids content, C/N, chemical oxygen demand, and substances such as ammonia, hydrogen sulfide, and heavy metals either enhance or inhibit the AD process (Jingura and Kamusoko 2017). Temperature and substrate characteristics are critical parameters which significantly influence AD performance.

Microbial communities involved in AD function within distinct temperature preferences, namely psychrophilic (10–20°C), mesophilic (20–45°C), and thermophilic (50–65°C) conditions (Náthia-Neves et al. 2018). Thermophilic temperatures favor faster biochemical reactions, higher death rate of pathogens and increased solubility compared to psychrophilic and mesophilic temperatures (Gebreeyessus and Jenicek 2016). However, the major drawbacks of thermophilic temperatures are the higher energy requirement and process instability (Liu et al. 2022). Mesophilic temperatures are more commonly used for AD than thermophilic temperatures because of higher process stability, lesser sensitivity to environmental changes and lower energy cost (Tufaner and Avşar 2016).

A study by Morales-Polo et al. (2018) established that the substrate properties affect biogas production and microbial activity. Furthermore, the substrate properties have an impact on methane content, biodegradability, and the rate at which degradation occurs (Nwokolo et al. 2020). Various pre-treatment methods, such as mechanical, thermal, chemical, and biological techniques, have been employed to enhance the properties of substrates (Kondusamy and Kalamdhad 2014). Oladejo et al. (2020) reported that the physico-chemical characteristics of FW can be improved by co-digestion with other substrates.

Although several papers reported on co-digestion of FW and CD, different parameters were investigated, combined, and individually. Bi et al.’s (2020) study examined the impact of hydraulic retention time (HRT) on co-digestion of FW and CD. Wang et al. (2020) investigated the impact of TS on an anaerobic co-digestion of pig manure and FW. Iqbal et al. (2014) looked at the effect of addition of sodium hydroxide to AD of FW and CD. However, few studies focused on the combined impact of substrate ratio and temperature on co-digestion of FW and CD. Furthermore, insufficient knowledge about the source-specific physico-chemical properties and energy potential of FW limits its use as a substrate for AD in South Africa. The aim of this paper was to characterize FW from a typical South African university food canteen and evaluate the effect of mixing FW and CD on biogas yield, methane content and BMP at different temperatures.

Materials and methods

Sample collection and preparation

Food waste was obtained from the University of Venda canteen after meals for five consecutive days. The FW was manually sorted to remove inorganic materials and mixed on a clean surface. To improve homogeneity, the FW was mashed and mixed using a kitchen blender. Cow dung was sourced from the University of Venda’s Farm. Inorganic materials such as pebbles were removed from CD. The prepared FW and CD samples were stored at 4°C prior to use as substrates.

The initial microbial population for the AD experiment was provided by anaerobic sludge. The anaerobic sludge used in the study was obtained from the laboratory AD setup, where CD was incubated at temperatures of 35°C and 55°C. The anaerobic sludge was degassed at test temperatures till insignificant biogas was produced. Anaerobic sludge was stored at ambient temperature before AD.

Experimental design

A 2 × 5 factorial experimental design was used to assess the influence of substrate mixing ratios at 35°C and 55°C on the yield of biogas, methane content and BMP. The FD:CD substrate mixing ratio use were used 100:0, 75:25, 50:50, 25:75, and 0:100 (w/w) with 0:100 (w/w) being the control treatment.

Statistical analysis

The results of a 2 × 5 factorial design experiment were subjected to statistical analysis using the Minitab 19 (Minitab, UK) statistical software using the generalized linear model. The effects of substrate mixing ratios and temperature on the biogas yield, methane content, and BMP were investigated using analysis of variance (ANOVA) at a 5% level of significance. Where a significant result was obtained through ANOVA, the Fisher’s least significant difference test was employed to compare means, with a significance level set at 5%

Experimental procedure

The AD batch experiments were conducted using 500 mL volume Duran glass vessels (reactor) coupled to the 500 mL eudiometer tube. According to the VDI 4630 methodology, the reactors were loaded with the substrate sample and inoculum at a 2:1 inoculum/substrate ratio (ISR) (VDI 4630, 2016). Deionized water was added to achieve 70% of the capacity of the reactor vessel. After the introduction of the inoculum and substrate into the reactor or eudiometer set-up, nitrogen gas was used to purge the headspace for 3 min to establish anaerobic conditions. A blank essay (inoculum only) was subjected to AD to determine its background biogas volume. Biogas generated during the blank assay was deducted from the biogas generated from the substrate assay at every sampling time. A positive control assay of CD mixed with inoculum at ISR = 2 was incubated at 35°C and 55°C using a water bath. The reactors were shaken by hand twice a day to enhance homogeneity and to prevent the formation of substances such as VFAs. The measurement of biogas production was conducted at 24 h intervals by reading the displacement of the level of the barrier solution in the eudiometer tube. Acidified water was used as a barrier solution to reduce the dissolution of carbon dioxide. The biogas produced was recorded at atmospheric pressure according to ISO standards (ISO/DIS 14,853, 1999). The experiment was concluded when the biogas volume generated continuously for three consecutive days was below 0.5% of the total accumulated biogas amount.

Analytical methods

Standard methods were employed to determine the moisture content (MC), total solids (TS), and volatile solids (VS) contents of the samples (APHA 2017). A pH meter (Thermo Fisher Scientific Orion, South Africa) was used to determine the pH of the samples. A portable biogas analyzer (GeoTech Biogas 5000, UK) was used to measure methane content.

Results and discussion

The physico-chemical properties of the substrates used in this study are displayed in Table 1. FW:CD (100:0) had the highest MC (80.51%) and FW:CD (25:75, w/w) had the least MC (71.99%). Moisture content increased with increase of FW proportion in the co-digestion mixture. This can be attributed to the high MC in FW. According to James et al. (2006), 68–80% MC is required for optimum biogas production. Substrates used in this study had sufficient MC favorable for AD. FW:CD (100:0, w/w) had the highest VS/TS% (97.60%), indicating the high biodegradability of FW, and cow manure had the lowest VS/TS% (77.31%). Zhai et al. (2015) reported comparable characteristics for kitchen waste and CD. According to Liu et al. (2022), biogas production experiences significant inhibition when the pH drops below 6.0 or exceeds 8.5. The pH values of the substrates (Table 1) fell within the range of 6.3 to 7.93, which is conducive to the AD process.

Table 1. Physico-chemical characteristics of substrates used in this study.

	Anaerobic	FW/CD	FW/CD	FW/CD	FW/CD	FW/CD
Parameter	sludge	0:100	25:75	50:50	75:25	100:0
MC (%)	95.40 ± 0.001	76.07 ± 0.002	71.99 ± 0.02	76.67 ± 0.02	77.31 ± 0.01	80.51 ± 0.01
TS (%)	4.60 ± 0.01	23.93 ± 0.02	28.00 ± 0.02	23.33 ± 0.02	22.70 ± 0.01	19.49 ± 0.01
VS/TS (%)	71.23 ± 0.02	77.31 ± 0.01	80.32 ± 0.04	87.79 ± 0.01	90.38 ± 0.02	97.60 ± 004
pH	7.44 ± 0.02	6.94 ± 0.01	7.7 ± 0.02	6.77 ± 0.03	7.93 ± 0.02	6.3 ± 0.02

MC = Methane content, TS = Total solids, VS = Volatile Solids, FW: CD = Food waste/Cow dung mixing ratio (w/w). Values given are means of triplicates plus standard deviation.

The study results show that higher biogas yields were produced at 55°C than 35°C (Table 2). Further, for all FW:CD mixing ratios, the HRT for 55°C was 35 days while for 35°C was 40 days. The lower HRT at 55°C can be attributed to faster rate of biochemical reactions at thermophilic temperatures relative to mesophilic temperatures. Yamashiro et al. (2013) also found that dairy CD and FW co-digested at thermophilic temperature (55°C) yielded a larger biogas volume than at mesophilic temperature (37°C). In the study conducted by Pax, Muzenda, and Lekgoba (2020), the co-digestion of sewage sludge/bio-waste mixtures under thermophilic conditions was observed to increase biogas yields by approximately 45% to 50% compared to mesophilic co-digestion. As reported by Nie et al. (2021), the growth rate of methanogenic microorganisms is more rapid at thermophilic temperatures compared to mesophilic temperatures, thereby accelerating the AD. The present study indicates that 55°C is optimum temperature for anaerobic co-digestion of FW and CD.

Table 2. Effect of FW:CD mixing ratio and temperature on biogas yield, biogas methane content and BMP.

SR − (FW: CD) (w/w)	T (℃)	Total biogas yield (mL)	Average biogas methane content (%)	BMP (mLCH4/g VS)
0:100	55	5301.67 ± 62.51	55.52 ± 0.21	328.278 ± 4.265
25:75	55	6865.00 ± 137.48	53.09 ± 0.36	324.238 ± 7.005
50:50	55	6256.33 ± 63.11	59.02 ± 0.03	36.638 ± 3.831
75:25	55	7151.67 ± 11.55	57.64 ± 0.36	401.887 ± 1.981
100:0	55	5901.67 ± 75.22	58.68 ± 0.07	354.156 ± 3.333
0:100 (control)	35	3786.67 ± 128.49	57.82 ± 0.81	211.563 ± 4.303
25:75	35	4308.33 ± 74.89	63.14 ± 0.07	248.066 ± 2.935
50:50	35	4816.67 ± 53.92	62.46 ± 0.46	293.736 ± 5.422
75:25	35	4903.33 ± 38.84	64.76 ± 0.38	301.879 ± 2.07
100:0	35	3291.67 ± 81.44	60.92 ± 0.28	236.686 ± 1.207

FW: CD = Food waste/cow dung mixing ratio (w/w), T℃ = test temperature, SR = Substrate mixing ratio, FW: CD (0:100) = control treatment, BMP = biochemical methane potential. The values presented in the table are mean of triplicates plus the standard deviation.

Figure 1 shows the daily variations of the biogas yield and accumulation for the experimental conditions used in this study. Figure 1a shows that FW:CD ratios of 100:0, 50:50, and 75:25 (w/w) reached peak biogas yields on day 1. FW:CD (25:75, w/w) and FW:CD (100:0, w/w) reached peaks on day 15 and 19, respectively. The high peak in biogas yield on day 1 is due to strong biodegradability of FW (Franqueto, da Silva, and Konig 2020). The delay in reaching peaks in FW (25:75, w/w) and FW:CD (0:100, w/w) can be attributed to the existence of less bio-degradable materials like cellulose and lignin in CD (Vanegas and Bartlett 2013). It was observed that no biogas was produced after day 25 from FW:CD (100:0, w/w) (Figure 1a) which can be due to the accumulation of process inhibitors such as VFAs (Gaby, Zamanzadeh, and Horn 2017). Kim, Kim, and Yun reported that AD of FW is affected by inhibitory compounds such as ammonia and VFAs.

Figure 1. Daily variations of biogas yield (a) and accumulation (b) at 35°C and daily variations of biogas yield (c) and accumulation (d) at 55°C.

The highest total biogas yield was obtained for mixing ratio of FW:CD of 75:25 (w/w) for both 55°C (7151.67 ± 11.55 mL) and 35°C (4903.33 ± 38.84 mL) as shown in Table 2. Mixing CD and FW improves nutrient balance and process stability. The findings are consistent with those of Aragaw and Gessesse (2013), who observed that the co-digestion of a mixture consisting of 75% kitchen waste and 25% CD produced the highest biogas yield. The total biogas yield for mono-digestion of FW and CD were 5301.67 ± 62.51 mL and 5901.67 ± 75.22 mL, and 3786.67 ± 128.49 mL and 3291.67 ± 81.44 mL, for 55°C and 35°C, respectively. The lower yields of biogas from mono-substrates can be attributed to poor nutrient balance and formation of possible reaction inhibitors. The overall biogas yield in this study was enhanced by adding more FW to the co-digestion mixture. The improvement in biogas yield can be due to enhanced C/N ratio by co-digestion (Baek et al. 2020). The findings suggested that co-digestion can increase biogas yield.

Figure 2 shows the daily variation of biogas methane content for experimental conditions used in this study. Figure 2a shows that methane content increased gradually from day 1 to about day 15 and then fluctuated between 45 − 65%. This is due to the multiplication of methanogens which increase production of biogas. At 55°C, the methane content increased steadily from day 1 to day 10 and thereafter fluctuated between 40% and 65%. The average methane content range at 35°C was 57.82–64.76% which is higher relative to 53.09–59.02% average methane content at 55°C (Table 2). Comparable methane content ranges of 55% to 65% were recorded by Malik, Mohan, and Annachhatre (2020) for the co-digestion of FW and CD. The average biogas methane content of co-digestion mixtures was higher than of mono-substrates. The findings of this study align with the research conducted by Oladejo et al. (2020), which demonstrated that the anaerobic digestion (AD) of food waste (FW) and pig manure yielded the highest methane content (64.6%), whereas the sole digestion of FW resulted in the lowest methane content (54.0%).

Figure 2. Daily variations of biogas methane content (a) at 35°C and (b) at 55°C.

The highest BMP (401.89 ± 1.98 mLCH₄/g VS) was produced from FW:CD mixing ratio of 75:25 (w/w) at 55°C and the lowest BMP (211.56 ± 4.30 mL CH₄/g VS) was produced from mono-digestion of food waste at 35°C. All co-digestion mixtures had greater BMP values than pure CD and FW. Superior BMP values were obtained at 55°C relative to 35°C. The findings of this study are consistent with the results reported by Rattanapan et al. (2019), who investigated the co-digestion of residential wastewater and FW. They demonstrated higher biomethane potential (BMP) values at 55°C compared to 35°C. The BMP of FW:CD (0:100, w/w) at 35°C and 55°C was higher than that of FW:CD (100:0, w/w), which were 211.56 mL CH₄/g VS at 35°C and 328 mL CH₄/g VS at 55°C, respectively, at 236.69 mL CH₄/g VS and 354.16 mL CH₄/g VS. This is due to VFA formation in AD of FW. The results obtained in this study for the BMP of mono-digested CD are comparable to those of Pax, Muzenda, and Lekgoba (2020) who obtained BMP of 302, 228, and 241 L/kg VS for the AD of fine and coarse and unscreened cow dung, respectively. The findings of this study demonstrate that the BMP can be enhanced by co-digestion.

Conclusions

The objective of this study was to determine the effect of co-digestion and temperature on AD of FW and CD. Both co-digestion and temperature significantly affected biogas yield and BMP of the substrates. Co-digested of FW and CD had higher biogas production compared to the individual substrates. The findings of this study indicated that an increase in the proportion of FW in the co-digestion mixture resulted in an increase in biogas yield. The highest biogas yield (7151.67 ± 11.55 mL) and BMP (401.89 ± 1.98 mLCH₄/g VS) were obtained from FW and CD co-digested at a mixing ratio of 75:25 (w/w) at 55°C. Higher biogas yield and BMP of the substrates were obtained at 55°C than at 35°C. Nevertheless, additional research is required to explore the overall effect of elevated temperatures on AD.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The funding for this research work was provided by various projects, including DST-ERAfrica, UNIDO, SANEDI, and CDM projects.

Notes on contributors

Prosper Mhlanga

Prosper Mhlanga is a master’s student in Agricultural Mechanization at the University of Venda. He graduated from Chinhoyi University of Technology (Zimbabwe) with a Bachelor of Engineering honors degree in Fuels and Energy Engineering. His research interests include bioenergy, waste management, and climate action.

Moses Okoth Marenya

Moses Okoth Marenya holds the position of Senior Lecturer within the Department of Agricultural and Rural Engineering at the University of Venda. He graduated with a PhD in Agricultural Engineering from the University of Pretoria in South Africa. His research focuses on various fields, including food engineering, operational systems simulation, and soil dynamics.

Nikita Tawanda Tavengwa

Nikita Tawanda Tavengwa is a senior lecturer in the Department of Chemistry at the University of Venda. He graduated from the University of Witwatersrand in South Africa with a PhD in analytical chemistry. His areas of interest in research include chromatographic analysis, analytical method development, and trace element analysis.

David Tinarwo

David Tinarwo is a senior lecturer in the Department of Physics at the University of Venda. He holds a PhD in Electrical Engineering from the University of Kassel (Germany). His research interests include wind, solar, bioenergy, and renewable energy for sustainable rural development.