Effect of temperature and particle size of three types of municipal solid wastes on biochemical methane potential Open Access

The conversion of biomass from solid waste into biofuels, such as methane, is an alternative energy source that contributes to the energy sector while also extending the lifespan of landfills by promoting a circular economy.

Anaerobic digestion (AD) has emerged as a method of recovering energy from municipal organic solid waste (MOSW), thus reducing its polluting potential. The biogas produced from organic solid waste is composed of approximately 60–70% CH₄ and 30–45% CO₂ (Cárdenas Cleves et al. 2016), and it can be reused for thermal, electric or fuel purposes (Fan et al. 2019). Despite this advantage, some factors can influence the AD process, limiting its benefits; therefore, better control should be considered. These factors include (i) substrate composition and concentration; (ii) macro and micronutrient dosing balance; (iii) pH, humidity, and alkalinity; (iv) the presence of toxic compounds; and (v) temperature and size of substrate particles (Gueri et al. 2017; Krause et al. 2018; Ohemeng-Ntiamoah & Datta 2019).

Temperature can affect the energy balance and substrate physicochemical properties, compromising the efficiency of the organic matter (OM) stabilization process and methane production (Mao et al. 2015; Kunz et al. 2019). Thermophilic processes (45–70 °C) exhibit greater efficiency in the degradation of OM and, consequently, in the production of biogas compared with mesophilic processes (20–45 °C) (Zábranská et al. 2002). Nonetheless, thermophilic processes with high organic loads generate higher amounts of volatile acids, and they are more likely to cause reactor failure due to the imbalance between acid production and consumption, inhibiting methanogenesis (Li et al. 2015). In general, AD is more stable in mesophilic than in thermophilic environments, consuming less energy and being less susceptible to shock loads or the addition of inhibitory materials (Gueri et al. 2017).

The mesophilic condition also favors the diversification of the microbial population (Kim et al. 2017) and increases the methane production potential (Pearse et al. 2018). The nature of the waste is essential when assessing the influence of temperature on the AD process.

In the case of particle size, its influence on the OM stabilization rate is because particle reduction results in a larger specific surface area of the substrate available to microorganisms. Thus, with higher biomass, the acetogenesis and methanogenesis steps are accelerated, consequently enhancing biogas production (Zhang & Banks 2013). As particle size decreases, the hydrolysis rate increases, facilitating the accumulation of organic acids, which in turn may inhibit, for instance, the methanogenic Archaea, resulting in reduced methane generation (Alcântara 2007; Zhao et al. 2012; Zhang & Banks 2013).

The prediction of methane generation is an important step in solid waste management (Angelidaki et al. 2009). The evaluation of the biochemical methane potential (BMP) aims to quantify the maximum generation of methane from residues, determine the anaerobic biodegradability of organic substrates, and define the most suitable substrate for the process (Cárdenas Cleves et al. 2016). Although thermophilic AD conditions are known to increase degradation rates (Gueri et al. 2017; Fan et al. 2018), most BMP studies (70%) have been performed under mesophilic conditions (Ohemeng-Ntiamoah & Datta 2019).

The size of the particles is also important since the smallest particles can facilitate biodegradation; however, at the same time, the size must also be large enough to prevent changes in sample properties.

According to Ohemeng-Ntiamoah & Datta (2019), some researchers have performed BMP tests in both mesophilic and thermophilic conditions but have kept other factors almost constant. Nonetheless, a concomitant variation of different experimental conditions in BMP tests may present an important contribution to the development and improvement of the degradation of organic wastes. In this context, this research aimed to evaluate methane production from MOSW using factorial design methodology to verify the effects of temperature and particle size on BMP.

The food waste (FW) sample was prepared according to Figueiras (2016), and the paper waste (PW) sample was prepared according to a gravimetric study by Silva (2014). The yard waste (YW) sample was obtained from a gravimetric study of the yard wastes of the university campus where the experiment was performed (Guerra 2020). The composition of each sample by weight in percentage: FW (24.2% potato, 4.4% onion, 2.6% carrot, 8.9% banana, 7.36% apple, 3.5% orange, 16.3% bread, 6.4% bean, 6.4% rice, 5.61% chicken, 4.6% beef, 2.7% lettuce, 2.6% tomato, and 4.3% cheese) according to Figueiras (2016); PW (1.41% white paper, 68.76% paperboard, and 29.83% mixed paper) according to Silva (2014) and YW (2.29% twigs, 46.64% leaves and 51.07% grass) according to Guerra (2020). The moisture values were 68, 15, and 42% for FW, PW, and YW, respectively.

For all waste types, the following procedure was performed: grinding in a domestic blender with enough distilled water to form a pasty mixture and allowing characterization according to the methodologies presented in Table 1. The proportion of water added per waste type for the characterization step was as follows: 400 mL for 200 g of FW, 1,000 mL for 100 g of PW, and 1,000 mL for 100 g of YW.

Table 1

Parameters and methods used to characterize the wastes

Characteristics .	Parameter .	Method .	Reference .
Physical	Moisture at 100–110 °C	–	Kiehl (1985)
	Series of solids	2540 B, C, D, E, F, G	APHA (2017)
Chemical	pH	4,500 – H+	APHA (2017)
	Electrical conductivity	2510 A	APHA (2017)
	Salinity	2520 A	APHA (2017)
	Alkalinity	2320 B	APHA (2017)
	Chemical oxygen demand (COD)	5220 C	APHA (2017), Gomes (1989)
Nutritional	Total Kjeldahl Nitrogen (TKN)	4,500 – Norg B	APHA (2017)
	Ammoniacal nitrogen	4,500 – NH3	APHA (2017)
	Phosphorus (P)	4,500 – P E	APHA (2017)
	Iron (Fe)	3,500 – Fe B	APHA (2017)
	Sulphate (⁠⁠)	4,500 –	APHA (2017)

Characteristics .	Parameter .	Method .	Reference .
Physical	Moisture at 100–110 °C	–	Kiehl (1985)
	Series of solids	2540 B, C, D, E, F, G	APHA (2017)
Chemical	pH	4,500 – H+	APHA (2017)
	Electrical conductivity	2510 A	APHA (2017)
	Salinity	2520 A	APHA (2017)
	Alkalinity	2320 B	APHA (2017)
	Chemical oxygen demand (COD)	5220 C	APHA (2017), Gomes (1989)
Nutritional	Total Kjeldahl Nitrogen (TKN)	4,500 – Norg B	APHA (2017)
	Ammoniacal nitrogen	4,500 – NH3	APHA (2017)
	Phosphorus (P)	4,500 – P E	APHA (2017)
	Iron (Fe)	3,500 – Fe B	APHA (2017)
	Sulphate (⁠⁠)	4,500 –	APHA (2017)

View Large

Table 2 presents the characterization of the waste classes. Thus, in the assembly of the reactors, as detailed in the next section, since the weight of each waste added to the reactors was the same (12.4 g of waste), the mass of the OM in terms of COD added to the reactors was different for each waste type: FW = 28 g O₂, PW = 6.3 g O₂, YW = 10,9 g O₂).

Table 2

Waste characterization

.	.	Waste type .
Parameter .		Food .	Paper .	Yard .
pH		5.8	7.9	6.1
Total alkalinity (mg CaCO3·g−1 waste)		5 ± 1	46 ± 2	9 ± 1
Salinity (‰)		2.6	0.2	0.9
Conductivity (mS·cm−1)		4.7	0.8	1.9
Moisture at 100–110 °C (%)		68 ± 0.1	15 ± 1	42 ± 5
Organic carbon (%)		16	43	26
COD (g O2·g−1 waste)		2.257	0.511	0.883
C/N ratioa		8	43	26
Solids	Total (mg·g−1)	179	564	351
	Volatile (mg·g−1)	176	487	327
	Volatile (%)	98	86	93
Iron (mg Fe·g−1)		3 ± 1	1.5 ± 0.0	13 ± 2
Sulphate (mg ·g−1)		0.02	0.10	0.20
Phosphorus (mg P-PO4·g−1)		0.5 ± 0.0	0.5 ± 0.2	1 ± 0.1
Nitrogen	N-TKN (mg N·g−1)	92	174	175
	N-NH3 (mg N·g−1)	2	0	0

.	.	Waste type .
Parameter .		Food .	Paper .	Yard .
pH		5.8	7.9	6.1
Total alkalinity (mg CaCO3·g−1 waste)		5 ± 1	46 ± 2	9 ± 1
Salinity (‰)		2.6	0.2	0.9
Conductivity (mS·cm−1)		4.7	0.8	1.9
Moisture at 100–110 °C (%)		68 ± 0.1	15 ± 1	42 ± 5
Organic carbon (%)		16	43	26
COD (g O2·g−1 waste)		2.257	0.511	0.883
C/N ratioa		8	43	26
Solids	Total (mg·g−1)	179	564	351
	Volatile (mg·g−1)	176	487	327
	Volatile (%)	98	86	93
Iron (mg Fe·g−1)		3 ± 1	1.5 ± 0.0	13 ± 2
Sulphate (mg ·g−1)		0.02	0.10	0.20
Phosphorus (mg P-PO4·g−1)		0.5 ± 0.0	0.5 ± 0.2	1 ± 0.1
Nitrogen	N-TKN (mg N·g−1)	92	174	175
	N-NH3 (mg N·g−1)	2	0	0

^aBased on the ratio COD/total nitrogen.

View Large

The anaerobic reactors consisted of 600 mL bottles with a useful volume, and they were operated in batches and under a static regime. In each reactor, 160 g of inoculum was added (∼1/3 of the reactor's useful volume), resulting in 28.7 g TVS·L⁻¹. The inoculum consisted of granular sludge from a brewery wastewater treatment plant adapted to high organic loads. Nevertheless, before adding the substrates (wastes), the sludge was acclimated to the study temperature for 15 days, during which the specific methanogenic activity of the present microorganisms was evaluated. Based on these data, the theoretical potential for methane production by the bacteria group was analyzed.

All reactors first received a volatile fatty acid (VFA) solution resulting in a concentration of 4.4 g COD·L⁻¹ (Florencio et al. 1993) and remained in a water bath at the established experimental temperatures for 15 days, aiming to acclimatize the microorganisms and determine the specific methanogenic activity of the sludge, according to Aquino et al. (2007). During this period, methane production was measured every 24 h. Afterward, 300 mL of the supernatant from each reactor was removed, and 300 mL of distilled water was added to 12.4 g of each organic waste. The alkalinity was adjusted using sodium bicarbonate (Trzcinski & Stuckey 2012). The pH stabilized between 7.0 and 7.2, and the reactors were immersed in water baths of 35, 45, and 55 °C. The assembled experiment is also shown in Figure 2. The biogas produced in the reactors passed through a 3% sodium hydroxide solution with a blue bromothymol indicator that scrubbed out CO₂.

Factorial planning allowed the observation of the individual effects of each parameter (temperature and particle size) and its possible interactions using mathematical tools. These tools were used both in the planning of the assays varying the parameters simultaneously and for comparing the results. Table 3 presents the factorial matrix used, in which the index ‘ − 1’ was attributed to the minimum temperature values and size, index ‘ + 1’ to the maximum values, and index ‘0’ to the central value.

The central point was performed in triplicate to calculate the experimental error and allow the observation of the statistical difference among the reactors fed with the three waste types, as proposed by Breitkreitz et al. (2014). The values of methane production with the assay of the central point did not enter the calculation of the effects of the parameters; they were used to verify whether the behavior was linear or quadratic in case the observed effect was significant.

Statistica software was used to compare methane production under the conditions tested and to verify whether variations in the parameters tested, particle size, and temperature, would influence methane production. Therefore, it would be possible to define the ideal conditions to achieve the optimum potential for methane production with this type of sludge and waste. For this purpose, the Spearman correlation test (p) was calculated. Values greater than p>0.05 indicate similarity between the values tested, whereas values lower than 0.05 indicate that the samples have different behaviors and are influenced by some parameter. The Spearman test compared the average methane production with the same particle sizes regardless of temperature to verify whether variations in waste sizes interfered with methane production and then compared the average methane production with the same temperatures irrespective of particle size to confirm whether temperature changes would alter methane production.

Table 4 shows the results of the methane production at the end of the 51-day experimental period, as a function of the different temperatures and particle sizes of the three wastes.

Table 4 presents the results of methane production and yield (in mL CH₄·g⁻¹ TVS) at the end of the 51 days. Methane yield was lower under mesophilic conditions (35 °C) and for both particle sizes of the tested wastes. The methane yields from PW and YW were lower compared to that of FW, possibly due to their high lignin content. Lignin is a complex molecule known to inhibit cellulolytic methanogenesis. Krause et al. (2018) reported methane yields ranging from 330 to 370 mL CH₄·g⁻¹ TVS for AD of office paper, 235 to 273 mL CH₄·g⁻¹ TVS for cardboard, and 46 to 61 mL CH₄·g⁻¹ VS under mesophilic conditions. The authors investigated different types of paper separately over 60 days. Castro et al. (2025) obtained average yields of 450 ± 50 mL CH₄·g⁻¹ VS under both mesophilic and thermophilic conditions, testing the co-digestion of FW with microalgae. Suphawatkon et al. (2024) achieved methane yields of 120 mL CH₄·g⁻¹ VS with 20% FW, but they observed VFA accumulation when the FW proportion exceeded 30%.

Despite the low methane production observed in our experiment, the yield was considered adequate because of the absence of additional costs associated with the non-use of chemical additives. Moreover, the experiment provided valuable insights into the individual behavior of each residue, which may allow for the formulation of mixtures that compensate for specific biochemical limitations, potentially leading to optimal yields at low cost.

The inoculum-to-substrate ratio was 1.4 for FW, 1.6 for PW, and 1.4 for YW, indicating a higher proportion of inoculum relative to the substrate. In AD, ratios below 1 are generally more concerning due to the increased risk of acidification.

In the case of the temperature range between 45 and 55 °C, the color change demonstrated a quadratic behavior of methane production for FW (Figure 5(a)) and PW (Figure 5(b)). Temperature interference was more significant in the mesophilic microbial community, with optimum activity up to 45 °C. Although no significant effects were observed, especially with particle size change, it was possible to notice that for the three waste types tested, the highest methane production occurred at 55 °C and with the smallest sizes tested, demonstrating that, in general, there was a trend toward reduction in methane production with the rise in particle size.

The effect of the interaction between the two parameters, particle size, and temperature, on methane production, was not statistically significant within the tested particle size range. However, methane production using larger particle sizes, compared to that of the smallest, was on average 2.2 times higher at 35 °C and 1.3 times lower at 55 °C for FW and PW. This behavior was not observed for YW. Nevertheless, the interaction with other parameters, such as pH and the C/N ratio, may have acted together and influenced the obtained results. Both are related to waste composition and to the solubilization of compounds that can alter the pH and inhibit or stimulate microbial activity.

The difference in the pH values may be related to the type of compound released during waste solubilization and AD. For FW and YW, the solubilized compounds were VFAs; therefore, the pH values decreased. However, the compounds were alkaline for PW, causing increased pH. Thus, for FW and YW in the reactors at 35 °C, the alkalinizing compounds might have been sufficient to neutralize the acids produced and avoid acidification. In reactors subjected to a temperature of 55 °C, hydrolysis and acidogenesis probably occurred at a higher speed, with a consequent increase in the production of acids and a pH reduction.

Figure 8(b) shows that, for the same particle size, by varying the temperature from 35 to 55 °C, there was a mean increase in OM stabilization of 306 ± 27 mL CH₄·g⁻¹ TVS·day⁻¹ for FW, 127 ± 75 mL CH₄·g⁻¹ TVS·day⁻¹ for PW and 199 ± 3 mL CH₄·g⁻¹ TVS·day⁻¹ for YW. At a temperature of 35 °C, the reduction in waste particles did not increase the specific methane production rate; the rate decreased 2 and 2.4 times for the reactors with FW and PW, respectively. At 55 °C, there was an increase in the rate of 1.2 times for FW and 1.6 times for PW, with a reduction of the particle from 2 to 0.6 mm and from 22 to 2.2 mm, respectively. For YW, at temperatures of 35 and 55 °C, there was a decrease in the methane production rate when the size was reduced.

The mean production of methane using larger particle sizes were 934 ± 978, 671 ± 979, and 736 ± 950 mL for FW (2 mm), PW (22 mm) and YW (22 mm), respectively. In the case of the smallest sizes, the volume produced was 998 ± 1,237, 691 ± 925, and 784 ± 1,030 mL for FW (0.6 mm), PW (2 mm), and YW (2 mm), respectively. The standard deviations were high since, for each particle size, the mean and the deviations were taken among the reactors with distinct temperatures and the same particle size. For distinct temperatures, methane production was significantly different, justifying the values of the deviations.

Zhang & Banks (2013) and Silvestre et al. (2015) also found no significant difference in methane production when they reduced the sizes of urban solid waste particles during the AD process. Krause et al. (2018) modified the granulometry of PW from 2 to 22 mm and concluded that the reduction in particle size did not impact methane production because of the small thickness of the paper. Some authors have also reported that the decrease in waste particle size may cause the accumulation of VFAs (Izumi et al. 2010), the increase in the apparent density of the waste, and the reduction in water retention and pH, inhibiting methanogenic Archaea (Zhao et al. 2012; Zhang & Banks 2013).

In this study, no statistically significant difference in methane production was observed between the tested particle sizes. However, an increasing trend in methane production was noted for FW and PW with larger particle sizes, within the tested range under mesophilic conditions. This may be attributed to a slower hydrolysis rate, which likely prevented the accumulation of intermediates to inhibitory levels of microbial activity. Under thermophilic conditions, smaller particles were degraded more rapidly, in line with the faster biochemical reaction rates of the microbial groups involved, resulting in slightly higher methane production compared with that of the larger particles. For YW, these patterns were not observed, likely due to the complexity of the compounds in this type of residue and the presence of inhibitory intermediates, such as phenolic compounds.

The mean production of methane at 35 °C was 183 ± 84, 224 ± 137, and 60 ± 6 mL at 35 °C and 1,701 ± 194 mL, 1,209 ± 320 and 1,461 ± 74 mL of methane at 55 °C for FW, PW and YW, respectively. Fernández-Rodríguez et al. (2016) reported that high temperatures increased the solubilization of solid waste and accelerated microbial enzymatic activity, increasing the speed of CH₄ production. Additionally, Streitwieser (2017) and Jiang et al. (2018) reported that the bacterial diversity involved in hydrolysis, acidogenesis, and acetogenesis was higher as the temperature rose in the following order: 45, 55, and 60 °C. Thus, the temperature of 55 °C may have favored a greater microbial diversity than at 35 °C, which may have promoted a higher volume of methane.

By identifying the species present at different temperatures, Watanabe et al. (2017) observed the predominance of Methanosaeta concilii at 37 °C and Methanosarcina thermophila at 55 °C. At mesophilic temperatures (35 °C), there was a greater diversity of methanogenic microorganisms, converting complex OM into methane without VFA accumulation.

Pongsopon et al. (2023) also observed a pH reduction when they operated an anaerobic digester fed only with FW, attributed to the accumulation of acetate, butyrate, and propionate. They also tested the co-digestion of FW with pretreated YW, but there was again a pH reduction. Nevertheless, the acetate concentration was lower than in the reactor fed exclusively with FW. Shrestha et al. (2023) verified low methane production when FW was used as the only feeding source. The authors attributed this to the fact that FW is highly biodegradable, and that hydrolysis occurred very fast, with an accumulation of VFA and NH₃, possibly toxic to methanogenic bacteria.

Conversely, Xu et al. (2022) affirmed that although FW presented high biodegradability, the low C/N ratio and the high protein concentration resulted in high NH₃ production, which could have been toxic to microorganisms. The authors suggested using co-digestion of FW and PW since PW would promote slow AD and a high C/N ratio, which could minimize ammonia production.

When the C/N ratio is not ideal, ammonia inhibition may occur, thereby reducing carbon usage by methanogenic microorganisms. The C/N ratio is usually high in FW, given the low concentration of nitrogenous compounds. According to Mu et al. (2020) and Panigrahi & Dubey (2019), low nitrogen concentrations favor hydrolysis and acidogenesis, resulting in the formation of VFAs and consequent pH reduction. PW and YW are rich in nitrogenous compounds, such as lignocelluloses, according to Mu et al. (2020), resulting in a lower C/N ratio in reactors with these waste types.

In the reactors with PW and YW, the quantity of Total Kjeldahl Nitrogen (TKN) (2.16 and 2.17 g, respectively) was twice that found in the reactor with FW (1.14 g). The mean C/N ratio was 5, 6, and 11 for the PW, YW, and FW reactors, respectively. The smaller the C/N ratio, the greater the amount of nitrogen. Therefore, since the reactors with PW had more nitrogen and a good degradation of the OM, they presented higher methane production, and there was no pH reduction due to ammonia formation. All reactors presented C/N ratio values lower than that recommended as ideal, between 20 and 30, according to Mao et al. (2015).

The rate of OM degradation is related to the methane production rate. Figure 8(b) presents the mean values of the specific rates for each waste type. According to Zhang & Banks (2013), particle size can increase the speed of the process of AD. Nonetheless, this behavior was observed only at 55 °C for the reactors with FW and PW. At a temperature of 35 °C, the reduction in particle size did not favor the specific methane production rate. For the reactor with YW, there was no difference in the specific methane production rate as a function of particle size.

The reactors with FW and PW presented the best specific speeds for the degradation of OM with the smallest particle sizes (0.6 and 2 mm, respectively) and at 55 °C. Mao et al. (2015), Campuzano & González-Martínez (2016), and Viriato et al. (2015) also concluded that the higher the temperature and the smaller the waste particle size, the better the OM stabilization and, consequently, CH₄ production.

The reactor with YW behaved differently from the other reactors at 55 °C. Comparing the results of reactors with particle sizes of 2 and 22 mm, the specific methane production rate was only 5 (35 °C) and 200 (55 °C) mL CH₄·mg⁻¹ TVS·day⁻¹ (Figure 8(b)). Particle size did not significantly influence the stabilization and final production of methane.

This behavior can be explained, since despite being a lignocellulosic waste like PW, YW is formed by a more complex structure of cellulose, hemicellulose, and lignin. Panigrahi et al. (2019) affirmed that a chemical pretreatment, such as electrolysis or co-digestion, may be necessary before the process of AD of YW is conducted. Thus, the processes that promote the break of the chemical bonds by hydrolysis had more of an impact on the efficiency of AD than simply reducing the waste particle size. Therefore, size reduction without breaking complex chemical bonds may not accelerate the process of YW digestion. However, size reduction, together with heating, can favor digestion.

All the waste types tested in this experiment demonstrated a high potential for methane production through AD. Despite minimal interference with the digestion process, methane generation was successfully achieved. Although the co-digestion of the tested wastes was not performed in this study, the results indicate that combining these residues may be a promising strategy to enhance methane production due to potential synergistic effects, as suggested by the individual biochemical performance observed.

This work aimed to verify the effects of particle size and temperature on the biochemical potential of methane production from municipal organic solid wastes. Under the studied conditions, the variation in waste particle size did not significantly influence methane production or the rate of OM stabilization. Nevertheless, in general, regardless of the waste type, the change in temperature from mesophilic to thermophilic promoted a significant increase in methane production (mean of 1,238 ± 361 mL) and the production rate (mean of 208 ± 82 mL CH₄·mg⁻¹SSV·day⁻¹).

The best conditions for CH₄ production were at a temperature of 55 °C and using the biggest particle sizes (2.0 mm for FW and 22 mm for PW and YW). Considering these findings, it is inferred that the size variation of the particles was not a determining factor for methane production; thus, it is more relevant to concentrate efforts and energy on a pretreatment process than to reduce the size of the waste.

Since the results indicated good individual biochemical performance, it is recommended that these residues be further combined. Due to potential synergistic effects, co-digestion can be a promising strategy to enhance methane production.

The authors would like to thank all the people from the Laboratory of Environmental Engineering (LEA-UFPE) for their analytical support, the Foundation for the Science and Technology of Pernambuco (FACEPE) and the Coordination for the Improvement of Personnel of Higher Education (CAPES) for financially supporting the research.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.