Effects of thermal pretreatment on the anaerobic digestion characteristics of fruit waste liquids

Owing to seasonal changes, a large amount of seasonal fruit wastes (fruit peelings, residues, and rotten fruits) are transported to the waste disposal station for treatment (Kalogiannis et al. 2024), accounting for more than half of the food waste (Zhu et al. 2023). Fruit waste has a high moisture content (Li et al. 2019), therefore, when fruit waste is transported to the waste disposal station, it is usually subjected to an extrusion and separation process, which produces a lot of waste liquid.

The fruit waste liquid separated from the waste disposal station also has a lot of soluble nutrients. At present, the waste liquid is generally discharged directly to the municipal sewage network, where the waste liquid will eventually be concentrated in the sewage treatment plant. But its water quality fluctuates greatly with the seasons, which not only results in a great waste of resources but also creates a greater burden on the subsequent sewage treatment (Zhang et al. 2023). Therefore, the resourcing and valuing of fruit waste residues and their waste liquid is a research field that has received much attention in recent years (Valta et al. 2017).

Fruit waste will produce waste liquid in the processes of screw conveying, three-phase separator, and crushing. Waste liquid treatment technologies include physical–chemical, chemical, and biological methods (Shah et al. 2024). Extraction is one of the commonly used physical–chemical methods (Ying et al. 2022), which can be used to remove harmful substances such as heavy metals in waste liquid. Chemical precipitation method is one of the commonly used chemical treatment methods (Benalia et al. 2022), which can effectively remove heavy metals from industrial wastewater. Benalia et al. (2022) used Na₂CO₃, Ca(OH)₂, and NaOH reagents to remove heavy metals from industrial wastewater using a chemical precipitation method, and the removal rate of copper and zinc reached more than 90%. Biological treatment methods include aerobic and anaerobic biological methods. Kang et al. (2023) used sauerkraut digestive waste liquid for aerobic composting, which improved the complexity and stability of bacterial community and increased soluble nutrients. Pastor et al. (2013) found that waste oil was more suitable as a co-substrate for the anaerobic digestion (AD) process than waste leachate.

AD is considered to have the most potential technology for the treatment of fruit waste, which cannot only avoid soil and air pollution, but also produce clean energy such as hydrogen and methane (Jin et al. 2021; Samoraj et al. 2022). The thermal pretreatment (abbreviated to pretreatment) of the substrate before AD can destroy its cell wall through high temperature or high pressure, and promote the dissolution of organic matter. Park et al. (2014) found that the soluble chemical oxygen demand (SCOD) of the system increased by about 53% when the treatment conditions were 120 °C for 60 min. Compared with solid fruit waste, when using fruit waste liquid for AD, small molecular liquid can effectively shorten the digestion period during AD and does not produce new residues after digestion (Lv et al. 2021; Pasalari et al. 2021).

The purpose of this study is to develop the potential of resource utilization from AD of fruit waste liquid. The gas production from fruit waste, its combined AD with municipal sludge, and the effect of thermal pretreatment on pH, volatile fatty acids (VFAs), and SCOD on the digestive system were studied. In addition, the experimental data were verified by the modified Gompertz model. This study provides new approach for the resource utilization of fruit waste.

The solid fruit waste (mainly composed of banana peel, pitaya peel, bagasse, apple core, pear core, and so on) used in this experiment was taken from the waste treatment project of Hubei Lufa Huanneng Technology Ltd (China). The sludge involved in digestion was the dewatering sludge of the municipal sewage treatment plant. The inoculated sludge was taken from a residual sludge digestion tank in Xianning City, Hubei Province (China). To reduce the effect of endogenous gases on the experiment, the inoculated sludge was allowed to stand for 5 days before the experiment until no gas was produced. The experimental materials and related parameters are presented in Table 1. 

An AD bottle of volume 500-mL was used as a reactor in the experiment. The digestion bottle has a sealing cover to maintain good sealing condition. A thermostatic water bath (HWS-26) was used to maintain warm conditions, with the temperature set at 35 ± 1 °C, and a hose was connected to the gas-collection bag to collect the produced gas. The air tightness of the system was carefully examined before starting the experiment.

The test items included total solids (TSs), VSs, pH, VFAs, SCOD, gas production, and gas components. The TS was determined by the oven-drying method, and dried at 105 °C in a drying oven (DZF-6050AB) for 5–6 h. The VS value was determined by the cauterization method in a muffle furnace (SX2-4-10A) at 550 °C for 5–6 h. The SCOD was determined by the potassium dichromate method using a SCOD analyser (COD-57). The pH is determined using a pH meter (PHS-3E). The VFAs and gas components were determined using a gas chromatograph (GC9790plus).

Table 2 presents the gas production and gas combustion state for each experimental group. For the unit VS gas production rate, group A2 (fruit waste liquid) has the highest gas production rate of 767.09 mL/gVS, which is 6.01 times that of group A5 with the lowest gas production rate. This may be because the liquid is more fully digested with the inoculated sludge. The gas production rate of the other three experimental groups from high to low were 720.18 mL/gVS in group A1, 319.87 mL/gVS in group A3, and 154.27 mL/gVS in group A4. It can be seen that the experimental groups (A4 and A5) with the addition of municipal sludge generally had a lower unit VS gas production rate, and the experimental group with the use of fruit waste liquid generally had a higher unit VS gas production rate. This also shows that the method of using municipal sludge as co-digestion substrate in traditional AD has no advantage in improving the gas production rate of fruit waste.

Table 2

Gas production and gas combustion state of each experimental group

Experimental group .	Unit VS gas production rate (mL/gVS) .	Digestion period (h) .	Gas combustion diagram .	Flame description .
A1	720.18	20.5		Light blue, jet flame about 4 cm, stable flame, the burning condition is general.
A2	767.09	20.5		Bright blue, jet flame about 4–5 cm, stable flame, the burning condition is good.
A3	319.87	44.5		Blue, jet flame about 5 cm, stable flame, good burning condition.
A4	154.27	44.5		Light blue, unstable when ignited, poor burning conditions.
A5	127.59	20.5		Light blue, unstable when ignited. Be extinguished in a short duration, and the combustion is poor.

Experimental group .	Unit VS gas production rate (mL/gVS) .	Digestion period (h) .	Gas combustion diagram .	Flame description .
A1	720.18	20.5		Light blue, jet flame about 4 cm, stable flame, the burning condition is general.
A2	767.09	20.5		Bright blue, jet flame about 4–5 cm, stable flame, the burning condition is good.
A3	319.87	44.5		Blue, jet flame about 5 cm, stable flame, good burning condition.
A4	154.27	44.5		Light blue, unstable when ignited, poor burning conditions.
A5	127.59	20.5		Light blue, unstable when ignited. Be extinguished in a short duration, and the combustion is poor.

View Large

Although the hydrogen unit VS yield and total hydrogen production of group A4 were the lowest in all experimental groups, the methane unit VS yield and total methane production of group A4 were 30.58 mL/gVS and 168.47 mL, respectively, which were much higher than those of the other groups. This is because the addition of municipal sludge as a mixture to fruit waste increases the initial pH of the AD system and provides a more suitable living environment for methanogenic bacteria. In addition, the fruit waste liquid in group A4 had better fluidity than the fruit waste slurry, which could be more fully mixed with the sludge to make the anaerobic process more thorough, thereby promoting methane production.

In each stage, the hydrogen content of group A4 was the lowest, but its methane production rate and methane content were always at the highest level. The methane production reached a maximum of 29.55% during the period 20.5–44.5 h. In addition, the methane content of group A5 was also higher than the remaining three groups, which was attributed to the unpretreated municipal sludge mixed in the digested substrate and its own high pH in groups A4 and A5. The addition of municipal sludge not only made the initial pH higher, but also enriched the bacterial strains of the system, and methanogenic bacteria became the dominant strains under this condition, so the methane content was higher, which was in line with the rationale of mixing municipal sludge into the traditional food waste AD system to increase the concentration of methane gas production.

Pretreatment of digested substrates can reduce pH, promote cell lysis, release more organic matter, and increase VFAs production (Zhen et al. 2017), which can be used by microorganisms as intermediate products to generate gases such as H₂, CH₄, and CO₂. As the digestion process proceeds, the SCOD content of the digestive system decreases due to the utilization of soluble organic matter by the microorganisms (hydrogen-producing bacteria, methanogenic bacteria) while the system produces hydrogen and methane (Ma et al. 2018).

From Table 3, it can be seen that the initial pH of the fruit waste slurry was 4.96, and after thermal pretreatment at 80 °C for 30 min, its pH was reduced to 4.59, which is due to the fact that in the pretreatment process, fruit waste was hydrolysed, producing weakly acidic organic matter (Yasser Farouk et al. 2020). At the end, the pH values of the experimental groups A1, A2, A3, A4, and A5 were 5.42, 5.37, 4.71, 6.49, and 5.69, respectively. The pH of group A2 was 0.05 lower than that of group A1, and the pH of group A3 was also lower than that of group A1. This may be because the pretreatment of groups A2 and A3 produced some VFAs that are more conducive to the direct utilization of microorganisms, hence the rate of hydrolysis of VFAs produced by the AD process was greater than the consumption rate, which resulted in a decrease of pH. It is inferred that thermal pretreatment can moderately reduce pH. The pH of group A4 was the highest at the end stage, probably because the fruit waste liquid of group A4 was less and the municipal sludge was still added to the digestive substrate. Hence, the initial pH of group A4 was larger, which created an advantageous condition for methanogens (the suitable survival condition for methanogens was pH 6.5–7.2 (Li et al. 2017)), therefore the unit VS methane production of group A4 was also the highest.

From Table 3, it can be seen that the VFAs content of the unpretreated fruit waste slurry was 39.24 mg/L, and the VFAs content increased by 5.79% after pretreatment. At the end of digestion, the VFAs in experimental groups A1, A2, A3, A4, and A5 were 91.89, 111.75, 161.64, 74.56, and 112.59 mg/L, respectively. Group A4 had the lowest content of VFAs and the largest pH, group A3 had the highest content of VFAs and the smallest pH, and there was not much difference in the content of VFAs in groups A1, A2, and A5, nor was there much difference in their pH. Moreover, an inverse relationship was indicated between the system's pH and the content of VFAs, which was in agreement with the results of Xiong et al. (2020).

As presented in Table 3, the SCOD content in the unpretreated fruit waste mixture was 58,900 mg/L, and the SCOD content was increased by 56.96% after thermal pretreatment, which was due to the fact that the cell wall of the fruit waste was destroyed by thermal pretreatment, and the insoluble organic matter therein was continuously transformed into soluble organic matter, which led to the increase of the SCOD content. At the end of digestion, the SCOD contents of experimental groups A1, A2, A3, A4, and A5 were 16,900, 19,350, 26,550, 13,550, and 27,450 mg/L, respectively. The SCOD content of groups A3 and A5 after digestion was significantly higher than that of the remaining three groups. This is because the fruit waste slurry can dissolve more organic matter after thermal pretreatment, and the fruit waste liquid is more fully in contact with microorganisms during the AD process, hence the consumption rate of organic matter in the digestive system is greater than the decomposition rate, and therefore the SCOD accumulation is lower.

T₉₀ is used to represent the time when the cumulative biogas production of the system reaches 90% of the total biogas production. Table 4 shows the dynamics parameters of Gompertz model and digestion time T₉₀. It can be seen from Table 4 that the T₉₀ of group A3 was the largest one, which was 21.12 h, the T₉₀ of group A5 was the smallest, which was 10.22 h, and the T₉₀ of groups A1, A2, and A4 was between A3 and A5, which was 11.52, 11.38, and 17.76 h, respectively. The T₉₀ of group A2 was 0.18 h lower than that of group A1, which indicated that thermal pretreatment shortens AD period of fruit and vegetable waste. The T₉₀ of group A2 was shorter than that of group A3 by 9.74 h. This may be caused by the smaller size and more mobility of the substrate molecules of group A2, whose contact with the inoculated sludge was more adequate and more favourable for microbial reactions. R_m reflects the maximum biogas production rate of the system. Group A2 had the largest R_m of 304.1 mL/h, which was 1.71 times higher than that of group A3 (177.5 mL/h), and group A1 (224.9 mL/h) also had a higher R_m than that of group A3, which reflected that the maximum biogas production rate of liquid was higher than that of slurry. The parameter λ denotes the delayed response time required for microorganisms to adapt to dynamic changes in the AD environment. The λ of group A2 was largest (5.1 h), which was 2.83 times that of group A3 (1.8 h), and the λ of groups A3 and A5 were both smaller than those of groups A1, A2, and A4, which may be due to the fact that the substrates used in groups A3 and A5 have a rough surface morphology that promotes microbial growth, and also facilitates rapid microbial adaptation to the environment (Shen et al. 2016).

Table 4

Dynamic parameters of the Gompertz model and digestion time T₉₀

Experimental group .	Pm (mL) .	Rm (mL/h) .	(h) .	R2 .	T90 (h) .
A1	1,494	224.9	4.3	0.9995	11.52
A2	1,587	304.1	5.1	0.9986	11.38
A3	2,851	177.5	1.8	0.9973	21.12
A4	855	74.0	4.0	0.9997	17.76
A5	1,143	183.6	2.7	0.9977	10.22

Experimental group .	Pm (mL) .	Rm (mL/h) .	(h) .	R2 .	T90 (h) .
A1	1,494	224.9	4.3	0.9995	11.52
A2	1,587	304.1	5.1	0.9986	11.38
A3	2,851	177.5	1.8	0.9973	21.12
A4	855	74.0	4.0	0.9997	17.76
A5	1,143	183.6	2.7	0.9977	10.22

View Large

• (1) The fruit waste liquid was more fully in contact with the inoculated sludge. The group A2 used the pretreated fruit waste liquid for AD, and it gained the highest total gas production rate of 767.09 mL/gVS, which was 6.01 times of group A5 with the lowest gas production rate. This indicates that the method of using municipal sludge as a co-digestion substrate in traditional AD has no effect on increasing the gas production rate of fruit waste liquid.

• (2) Thermal pretreatment of AD substrates produced weakly acidic organic matter through hydrolysis, which increased the VFAs content of the system and was able to reduce its pH. The pH of the system at the end of digestion showed an inverse relationship with the VFAs content.

• (3) Without the addition of municipal sludge, compared with the fruit waste slurry, the use of fruit waste liquid for AD can significantly shorten the AD period. This indicates that the fruit waste liquid can be decomposed by anaerobic microorganisms more quickly, thus speeding up the hydrolysis phase of AD.

• (4) The addition of municipal sludge to fruit waste liquid or slurry obviously reduced the gas production rate and gas composition of the unit VS. The gas production rate of unit VS in the experimental group with municipal sludge was generally low, while the content of methane in the gas fraction was higher.

This work has been carried out with the financial support of the Henan Provincial Science and Technology Research Project (232102321088).

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

The authors declare there is no conflict.