Alleviation of Organic Load Inhibition and Enhancement of Caproate Biosynthesis via Fe₃O₄ Addition in Anaerobic Fermentation of Food Waste

Abstract

The conversion of food waste into caproate via anaerobic chain elongation has gained increasing attention. However, limitations such as reliance on external electron donors, low carbon conversion efficiency under high loads, and unclear microbial mechanisms hinder its application. Fe₃O₄ reportedly can act as an electron shuttle and mitigate product inhibition during anaerobic digestion of sludge. Thus, Fe₃O₄ addition could overcome the challenges from high loads under certain conditions. In this study, the experiments were conducted under batch and semi-continuous conditions. This study investigated the effects of organic loads on the hydrolysis, acidification, and chain elongation of fermentation. Furthermore, the influences of Fe₃O₄ on caproate production and microbial profile under varying substrate-to-inoculation ratios and dosages were examined. The key results harvested from the semi-continuous trial indicate that high organic loads severely inhibited caproate production. And in batch tests, at an F/M ratio of 1:2, increasing Fe₃O₄ dosage evidently enhanced caproate production by promoting lactate conversion to butyrate and carbon chain elongation. At an F/M ratio of 6:1, maximum caproate yield reached 0.45 g COD/g COD at Fe₃O₄ of 2.0 g/L. High organic load reduced the abundance of butyrate-producing bacteria (Latilactobacillus and Stenotrophomonas). Nevertheless, the addition of Fe₃O₄ increased the abundance of butyrate-producing and caproate-producing bacteria (Caproiciproducens). In conclusion, Fe₃O₄ at an optimal dosage evidently enhanced caproate production under high organic loads by stimulating microbial electron transport and enriching relevant microorganisms.

1. Introduction

According to the Food and Agriculture Organization (FAO), approximately 1.3 billion tons of food are lost annually in the global food supply chain, posing significant risks to public health and environmental safety [1,2]. Effective and sustainable food waste management methods, including landfilling, composting, incineration, and anaerobic fermentation, are critical to mitigate the risks. However, landfilling and composting release greenhouse gases and/or require significant land resources, while incineration is rather intensive in energy consumption and could cause air pollution [3,4]. Seemingly, traditional disposal methods impose substantial environmental and economic burdens. Due to its high biodegradability, food waste is highly suitable for anaerobic fermentation, which enables the production of value-added resource products. This approach aligns with the principles of the circular economy by generating renewable chemicals and fuels and reducing reliance on fossil fuels. Anaerobic fermentation, influenced by pH, temperature, organic loading rate, and other factors [5,6], yields various products, including biohydrogen [7] and biogas [8], ethanol, lactate, short-chain fatty acids (SCFAs), and medium-chain fatty acids (MCFAs) [9], for example, caproate, which has broad commercial applications [10,11]. Furthermore, the large-scale application of caproate production derived from the fermentation of food residues has been practiced by several enterprises, such as ChainCraft B.V. in the Netherlands [12].

The production of caproate by anaerobic fermentation commonly involves three main processes, namely, hydrolysis, acidification, and chain elongation processes [13]. The chain elongation pathway includes reverse β-oxidation (RBO) [14] and fatty acid biosynthesis (FAB) pathways [15]. Both pathways require an electron donor (ED) and an electron acceptor (EA). Electron donors are defined as reducing compounds that provide energy, reducing equivalents nicotinamide-adenine dinucleotide (NADH), and acetyl-CoA in chain elongation [16]. Electron acceptors are compounds that provide a carbon skeleton to accept two carbon atoms, mainly SCFAs, such as acetate, propionate, and butyrate. Both RBO and FAB pathways are cyclic reactions, in which two carbon atoms are added to the starting molecule in each cycle, and subsequently fatty acids with 6–12 carbon atoms can be produced by repeating the cycle process [17].

Commonly, many microorganisms are involved in the fermentation process of organic waste such as food waste. Under different organic loading rates (OLR) [6], the microbial communities developed in the process are also diverse, and the distribution and concentration of the metabolites formed (like lactate, acetate, propionate, butyrate, valerate, etc.) are also different. The previous study [18] demonstrated that a high organic loading rate affected the distribution and composition of electron acceptors in ethanol-driven chain elongation. Under high organic loading conditions, the ratio of acetate to butyrate was relatively low, and in the niche, microorganisms tended to convert ethanol into butyrate rather than caproate. Moreover, at higher substrate concentrations with an ethanol-to-acid ratio of 2:1, the production of caproate was most favorable [19]. This study primarily investigated the distribution of electron donors and electron acceptors in the lactate-based chain elongation process during in-situ fermentation and their influences on caproate production under different conditions. It is also dedicated to finding the optimal ratio of electron donors to electron acceptors for caproate production, aiming to enhance the yield of caproate. Several reagents reportedly can be used to adjust the ratio of electron donors to electron acceptors, such as biochar [20], Fe₃O₄, and zero-valent iron (ZVI) [21].

Fe₃O₄ has gained significant attention for its role in enhancing substrate degradation and improving electron transfer in anaerobic processes [22]. As an electron shuttle, Fe₃O₄, just as ZVI, presumably is conducive to facilitating caproate production by alleviating product inhibition and promoting electron flow in fermentation processes. Furthermore, Fe₃O₄ contributed to the redox cycle between Fe²⁺ and Fe³⁺, enhancing electron transfer [23]. The in-situ generation of ferrous ions by Fe₃O₄ can stimulate the activity of key enzymes such as pyruvate-ferredoxin oxidoreductase and electron-transfer flavoprotein involved in metabolic processes, enhancing overall system performance [24,25]. Compared to ZVI, Fe₃O₄ exhibits a higher oxidation–reduction potential (ORP), which preferentially enriches dissimilatory iron-reducing bacteria (DIRB) [23,25]. This microbial enrichment could facilitate more efficient substrate hydrolysis and acidification, further promoting anaerobic digestion and chain elongation processes. Furthermore, [23] stated that the addition of 5.0 g/L of Fe₃O₄ significantly increased caproate production from waste-activated sludge from 6063.7 mg COD/L to 9614.3 mg COD/L, compared with that of the control group, and shortened the fermentation time from 24 days to 9 days.

The studies [6,26] indicated that high organic loading can cause shock to the fermentation system, leading to microbial instability and consequently inhibiting acid production. As food waste contains high levels of organic matter, the high organic content, in particular with a high load to a fermentation process, may lead to microbial disturbance, which is not conducive to caproic acid production. Thus, it is of substantial necessity to develop a strategy to cope with the microbial shock caused by high organic loading. The study [23] stated that Fe₃O₄ effectively promoted the production of caproic acid and the abundance of caproate-producing microorganisms in anaerobic fermentation of activated sludge. To date, there are few studies on the effects of Fe₃O₄ on caproate production from food residues and the relative mechanisms involved in fermentation processes, including the chain elongation process, especially with the mechanisms in electron transfer. Hence, this present study aimed to understand the effects of Fe₃O₄ on caproate production and the fermentation process treating food waste under high organic loading and decipher the mechanisms aforementioned, consequently offering an effective approach to tackle the challenges faced by industry in in-situ caproate production under high loads and with the need for external electron donors.

Commonly, in a fermentation process with a complex substrate such as food waste without sterilization, the microbial profile is quite diverse, and the process also has complex spectra of products and relatively low carbon flux, which substantially limits the practical application of this fermentation process. Thus, there is a great necessity in properly regulating the process to obtain a high abundance of relevant microbes and a high yield of caproate as well as high carbon flux, in particular under high organic loads. Fe₃O₄ reportedly could enhance the electron transfer during anaerobic digestion of waste-activated sludge and obtain high effective carbon flux [23]. However, the effects of Fe₃O₄ on caproate production from food waste and the mechanisms involved in the fermentation process are still not well understood. Particularly, the role of Fe₃O₄ in electron transfer and its effects on the key microorganisms need to be further deciphered. Therefore, this study thoroughly explored the effects and mechanism of Fe₃O₄ in in-situ fermenter for caproate production from food waste under batch and semi-continuous conditions. Overall, this study provides effective strategies and mechanisms for enhancing the yield of in-situ caproate production in a fermenter fed with food waste.

2. Materials and Methods

2.1. Materials

2.1.1. Reagents

The specific components of the trace element solution include 1.50 g/L nitrilotriacetic acid, 3.00 g/L MgSO₄·7H₂O, 0.50 g/L MnSO₄·H₂O, 1.00 g/L NaCl, 0.10 g/L FeSO₄·7H₂O, 0.18 g/L CoSO₄·7H₂O, 0.10 g/L CaCl₂·2H₂O, 0.18 g/L ZnSO₄·7H₂O, 0.01 g/L CuSO₄·5H₂O, 0.02 g/L KAl(SO₄)₂·12H₂O, 0.01 g/L H₃BO₃, 0.01 g/L Na₂MoO₄·2H₂O, 0.03 g/L NiCl₂·6H₂O, 0.30 mg/L Na₂SeO₃·5H₂O, and 0.40 mg/L Na₂WO₄·2H₂O. All reagents were of analytical grade and purchased from China National Pharmaceutical Group (Shanghai, China).

Ferroferric oxide (Fe₃O₄, analytical grade), sodium hydroxide (analytical grade), and hydrochloric acid (analytical grade) were also all purchased from China National Pharmaceutical Group.

2.1.2. Substrate

Food waste used in this study consisted of 41.8% staple food (21.8% rice and 20% noodles), 34.6% vegetables (12% cabbage, 5.6% carrots, 5% onions, and 12% potatoes), 9.2% protein (4.6% eggs and 4.6% tofu), 3% mushrooms, 8.2% pork, and 3.2% condiments. These ingredients were blended in proportion, steamed for 0.5 h, crushed, and then homogenized with a crusher. The prepared food waste was stored in a refrigerator at −20 °C, thawed at 4 °C, and kept at 4 °C for no more than 15 days before use.

2.1.3. Setup and Operation

The total volume of the semi-continuous anaerobic fermentation reactor was 4.5 L, and the effective volume was 3.0 L. The pH of the system was maintained at 5.5 ± 0.2 by online pH monitoring and adding 6 M NaOH and 6 M HCl. The temperature was maintained at 37.0 ± 0.1 °C by a water bath. During the inoculation of the reactor, the mixture of food waste (100 mL), inoculum (30.0 g, taken from a full-scale digester treating food waste), and fermentation medium (100 mL) was added to the reactor, and then deionized water was added to reach a volume of 3.0 L. Afterwards, nitrogen gas was injected for 30 min to purge oxygen.

During the operation period of the reactor, 100 mL of food waste was fed daily, and pH was maintained at 5.5. The reactor was operated for 144 days and was divided into four stages. The organic load of the fermenter increased from 2.71 g COD/L/d to 4.5 g COD/L/d in

Table 1. Semi-continuous reactor operational stages.

Period (d)	OLR (g COD/L/d)
Stage 1 (0~20)	2.71
Stage 2 (21~63)	2.71
Stage 3 (64~108)	4.50
Stage 4 (109~144)	2.56

Note: In Stage 2, the ORL was maintained as that in Stage 1 to obtain relatively stable acid production before further elevating OLR.

.

2.1.4. Batch Test

Eight groups of batch tests with Fe₃O₄ dosage of 0, 2.0, 5.0, and 10.0 g/L at the food-to-microorganism (F/M) ratio of 1:2 and 6:1 were carried out, respectively, to explore the influence of Fe₃O₄ addition on caproate production from food waste. Serum bottles with a working volume of 300 mL were employed. The test of each group was carried out in triplicate. Then, the serum bottles were placed in an incubator shaker at 120 rpm, and the temperature was maintained at 37 ± 1 °C. The pH of the bottles was daily adjusted to 5.5 with sodium hydroxide (6 M) and hydrochloric acid (6 M).

2.2. Methods

2.2.1. Analytical Methods

Volatile fatty acids (VFAs), including SCFAs and caproate, were determined by gas chromatography equipped with a flame ionization detector (FID) and a fused quartz capillary column (PEG-20M, Replete technology, Chengdu, China) (GC-2010, Shimadzu, Kyoto, Japan). Lactate was determined using a high-performance liquid chromatograph (Ultimate 3000, Thermo Fisher, Waltham, MA, USA) equipped with a Shim-pack GIST C18 AQ column (5 μm, 4.6 mm × 250 mm) and a diode array detector. Fe²⁺ and Fe³⁺ were determined by phenanthroline spectrophotometry and the analytical method of electron transport system (ETS) activity from the study [23].

2.2.2. Calculations

Hydrolysis and acidification efficiencies were calculated according to total chemical oxygen demand (TCOD), soluble chemical oxygen demand (SCOD), COD_VFAs (equivalent COD of VFAs), and COD_La (equivalent COD of lactate) of samples, respectively. All calculations were determined according to the following equations:

$$\mathrm{H}\mathrm{y}\mathrm{d}\mathrm{r}\mathrm{o}\mathrm{l}\mathrm{y}\mathrm{s}\mathrm{i}\mathrm{s} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \left(\mathrm{\%}\right) = SCOD/TCOD\times 100$$

(1)

$$\mathrm{A}\mathrm{c}\mathrm{i}\mathrm{d}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{e}\mathrm{f}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{y} \left(\mathrm{\%}\right) = ({COD}_{VFAs}+{COD}_{La})/SCOD\times 100$$

(2)

$$\mathrm{S}\mathrm{p}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{f}\mathrm{i}\mathrm{c}\mathrm{i}\mathrm{t}\mathrm{y} \left(\mathrm{\%}\right) = C_c/{\sum }_i^n{C}_i\times 100$$

(3)

where C_c is the concentration of a particular product and C_i is the concentration of individual carboxylic acids, including lactate, acetate, propionate, n-butyrate, iso-butyrate, n-valerate, isovalerate, and n-caproate.

The ratio of electron donors to electron acceptors (ED/EA) for the chain elongation process was also defined according to the following equation. C_La is the concentration of lactate, and C_(Ac+Bu) is the concentration of the sum of acetate and butyrate.

$$\mathrm{E}\mathrm{D}/\mathrm{E}\mathrm{A}=C_{La}/C_{(Ac+Bu)}$$

(4)

2.2.3. Statistical Analysis Methods

Analysis of variance (ANOVA) was employed to statistically examine whether there were significant differences or not in the effects of additions of Fe₃O₄ on SCFA and caproate production from food waste during fermentation in the batch tests.

3. Results and Discussion

3.1. Effects of Organic Load on Anaerobic Fermentation

3.1.1. Effects of Organic Load on Substrate Hydrolysis

Commonly, hydrolysis is the rate-limiting step in the anaerobic fermentation process of complex organic wastes and plays essential roles in substrate utilization during fermentation [27]. The changes of TCOD and SCOD during anaerobic fermentation can indicate the hydrolysis degree. In Stage 1 (Figure 1), SCOD and TCOD concentrations increased rapidly, reaching 36.4 g/L and 54.4 g/L, respectively, on Day 20, and the hydrolysis efficiency was 67% (Figure 1), indicating relatively efficient hydrolysis in the start-up. The hydrolysis efficiency of the reactor slightly decreased with an average of 62 ± 4% (Figure 1), accompanied by the elevated OLR in Stage 2. The observation aligned well with the study by [28]. In Stage 3, TCOD also presented a large fluctuation and increased from 97.2 g/L on Day 64 to 118.2 g/L on Day 118. Hydrolysis efficiency also fluctuated, and its overall trend gradually decreased with an average of 39.5 ± 8.0%, which was significantly lower than that of Stage 2. In Stage 3, the SCOD of the reactor did not significantly increase, which led to the decrease in the hydrolysis efficiency. Increasing organic load evidently inhibited hydrolysis of the fermenter. In Stage 4, the OLR was lowered to 2.56 g COD/L/d, and therefore, the TCOD substantially decreased. However, the SCOD was comparable with that of Stage 3, and thus the hydrolysis efficiency gradually increased in Stage 4. Thus, hydrolysis efficiency could be gradually restored by lowering the organic load rate of the process. Overall, the increase in organic load enhanced the availability of substrates, but the SCOD overall decreased, which might be attributed to the fact that the excessive organic acids inhibited microbial activities and subsequently decreased the hydrolysis efficiency of organic matter in the fermenter [28].

3.1.2. Effects of Organic Load on Acidification

In Stage 1, the decrease in acetate and n-butyrate concentrations (Figure 2a) was observed, which could be attributed to carbon chain elongation towards caproate. The result was well agreed with the increase in n-caproate. However, a delay in the increase in caproate was also observed, which is consistent with the study by [29]. In this stage, the carbon chain elongation towards odd fatty acids with odd carbon was not noticeably observed, which was probably also beneficial to caproate production [30].

In Stage 2, the concentration of n-caproate was relatively stable, compared with that of other stages, and maintained at an average of 10.21 g COD/L. The trend of n-butyrate was comparable to that of n-caproate (Figure 2a), whereas the concentration of acetate kept declining and finally stabilized at 1.87 ± 0.34 g COD/L. The results implied that acetate might be mainly involved in carbon chain elongation towards butyrate. The concentration of lactate gradually increased to 8.17 g COD/L in Stage 2 and further accumulated in the whole process, which indicated that the production rate of lactate was greater than its consumption rate. In Stage 3, accompanying the further increase in OLR to 4.5 g COD/L/d, the concentrations of n-caproate and butyrate presented a continuous decline, whereas acetate remained at a low level. However, lactate gradually built up (Figure 2a), reaching 13.62 g COD/L on Day 108, finally becoming the dominant product in the fermenter and consequently resulting in low availability of electron acceptors such as acetate and butyrate for the chain elongation process. The increase in organic load evidently led to more carbon flux towards lactate [6], which could be related to the evolution of the microbes in the fermenter. Reportedly, Lactobacillus, the main lactate producer, has a stronger ability to resist environmental changes such as lower pH and high organic loads [29].

Figure 2b showed the change in acidification efficiency during fermentation. The average acidification efficiencies in Stages 2, 3, and 4 were 42.72 ± 4.53%, 34.87 ± 4.03%, and 33.95 ± 2.98%, respectively. The higher acidification efficiency was in Stage 2, and the further increase in organic load led to a decrease in the acidification rate of the fermenter. Under high organic load, the buildup of lactate occurred, which seemingly also presented inhibitory effects on the production of other acids and also influenced the relative abundance of the relevant microbes [31].

The yield of fatty acids (2 to 6 carbons) produced in each stage (i.e., product specificity) is shown in Figure 3. In Stage 1, n-butyrate, n-caproate, and acetate were the main products, and in Stage 2, among the fatty acids, the fraction of caproate increased from 34.2% to 37.7%, whereas the proportions of n-butyrate and acetate decreased from 38.6% to 14.7% and 25.3% to 8.9%, respectively. Nevertheless, with the increasing OLR in Stage 3, the main product became lactate, accounting for 77.9% in the acids, whereas butyrate accounted for 9.4%. Thus, the ED/EA ratio was severely unbalanced, substantially hindering the chain elongation. Carbohydrates can be metabolized into lactate, and in the RBO pathway, the production of butyrate and caproate was also based on the formation of lactate [32]. The decrease in butyrate might be due to the undermined pathway from lactate to butyrate. The proportion of caproate in Stage 4 increased from 6.6% to 35%. In Stage 4, after lowering the OLR, the increase in butyrate proportion and the decrease in lactate proportion indicated that the carbon chain elongation seemed to be restored. The results evidently confirmed that organic load presented an important impact on the carbon chain elongation process. Excessive organic load presumably led to the imbalance of the concentration of electron donor and electron acceptor and substantially hindered the process. Therefore, more efforts and investigations should focus on regulating the production of electron donors and electron acceptors to ensure that the sufficiently balanced ED/EA ratio promotes more carbon flowing to caproate so as to greatly improve its yield.

To update, the results in this study indicated that the most suitable OLR for the fermenter was approximately around 2.71 g COD/L. Nevertheless, a high load substantially inhibited the hydrolysis, acidification, and chain elongation processes within the fermenter. This is likely due to the high organic load suppressing the activity of certain microorganisms, leading to an imbalance of ED/EA within the fermenter. The relative abundance of the microbial community in the fermenter during the operation indicated that the increase in Leuconostoc seemingly was related to the gradual accumulation of lactate, and the abundance of Latilactobacillus decreased under the high organic load as shown in Section 3.3.1). Therefore, under high-load conditions, an excess of lactate and a deficiency of butyrate took place, consequently undermining the chain elongation process towards caproate.

3.2. Effects of Fe₃O₄ on Caproate Production

Reportedly, Fe₃O₄ provides a certain reduction potential during the transformation of Fe²⁺ to Fe³⁺, which would be conducive to balancing or enhancing electron transfer during anaerobic processes [23]. To decipher the mechanisms involved in the process and tackle the issue of excessive lactate accumulation and insufficient butyrate in the fermenter under high load conditions, the present study explored the effects and mechanisms of different dosages of Fe₃O₄ under varying organic loads. The conversion of lactate to butyrate is a reductive process. Therefore, Fe₃O₄ may facilitate the conversion of lactate to butyrate, thereby restoring the chain elongation process under high load conditions and enhancing the utilization of organic matter and effective carbon flux under high load conditions.

3.2.1. Effect of Fe₃O₄ Addition on Profile and Yields of Fatty Acids

In the batch tests of an F/M ratio of 1:2 (i.e., at lower organic load), the total VFA concentration reached a maximum of 20.90 g COD/L at 2.0 g/L of Fe₃O₄, which increased by 0.93 g COD/L (Figure 4a), compared with the group without Fe₃O₄ addition. With the addition of Fe₃O₄, the selectivity of caproate merely increased by 4.1%, 2.3%, and 5.3%, respectively, compared with that of the control group (29.6%). The yield of caproate was only 0.084 g COD/g COD, 0.10 g COD/g COD, 0.093 g COD/g COD, and 0.099 g COD/g COD with the increasing Fe₃O₄ dosage, respectively. The addition of Fe₃O₄ apparently slightly promoted the yield of caproate. ANOVA analyses of the data from the control and experimental groups further confirmed and indicated that the addition of Fe₃O₄ at proper dosage evidently enhanced SCFA production (p < 0.05) but did not significantly promote caproate production (p > 0.05) at the F/M ratio of 1:2, which might be attributed to the deficiency of sufficient electron donor for further chain elongation.

However, at a higher organic load (i.e., at an F/M ratio of 6:1), it is statistically confirmed that the addition of Fe₃O₄ evidently enhanced both the VFA (p = 0.005) and caproate production (p = 0.001), and with the step-wise increase in Fe₃O₄ concentration, the total VFA production gradually increased by 6.82 g COD/L, 8.86 g COD/L, and 11.33 g COD/L (Figure 4a), respectively. The maximum selectivity of caproate increased by 22.8% at Fe₃O₄ of 2.0 g/L, compared with the control group (Figure 4b). Nevertheless, with the further elevation in Fe₃O₄ dosage, the yield of caproate increased only by 6.1% at 5.0 g/L of Fe₃O₄, and the yield even decreased at 10.0 g/L of Fe₃O₄. The maximum yield of caproate reached 0.45 g COD/g COD at Fe₃O₄ of 2.0 g/L. Compared with the control group without Fe₃O₄, the yield increased by 125%. The undermined yield of caproate at 10.0 g/L Fe₃O₄ might be attributed to the deficiency of electron donors, that is, the relatively rapid conversion of lactate in the fermenter. The addition of Fe₃O₄ seemingly promoted the conversion of lactate to butyrate, leading to the deficiency of lactate for the subsequent chain elongation process. With the addition of 2.0 g/L Fe₃O₄, caproate concentration at the F/M ratio of 6:1 increased by 24.37 g COD/L, compared with that at the F/M ratio of 1:2, and the yield of caproate increased by 34.2%. The increment indicated that the addition of Fe₃O₄ was presumably conducive to promoting electron transfer and subsequently enhancing the yield of caproate.

3.2.2. Mechanisms of Fe₃O₄ Involved in Electron Shuttle

Figure 5a showed that the electron transfer capability index of the electron transfer system (ETS) evidently increased by 7.1%, 209.6%, and 258.3% with the elevated Fe₃O₄ dosage, respectively, compared with the control group without Fe₃O₄. Furthermore, when the concentration of Fe₃O₄ increased from 0.0 g/L to 5.0 g/L, the concentration of soluble Fe²⁺ increased from 9.97 mg/L to 230.30 mg/L, and the concentration of Fe³⁺ increased from 14.64 mg/L to 450.00 mg/L (Figure 5b). The electron transfer ability of the system noticeably enhanced with the elevated dosage of Fe₃O₄.

The increase in Fe³⁺/Fe²⁺ suggested that Fe₃O₄ presumably provided electrons to the process. Reportedly, Fe₃O₄ can lead to changes in the distribution of products within bioprocesses [23,25]. To explore whether Fe₃O₄ was related to product distribution, this study subsequently examined the changes in the ED/EA ratio in different groups of batch fermenters.

At the F/M ratio of 1:2 (Figure 6), compared with the group without Fe₃O₄ (ED/EA = 7.04), the groups with Fe₃O₄ presented lower ED/EA ratios, namely, 5.24, 5.50, and 5.19, respectively. The results suggested that Fe₃O₄ likely promoted the conversion of lactate or increased the production of electron acceptors. At the F/M ratio of 6:1, subsequently, the caproate production rate remained almost unchanged, and the average of ED/EA ratios was lower in the groups with added Fe₃O₄, compared to those without added Fe₃O₄. In batch fermenters with high organic loads, the addition of Fe₃O₄ evidently promoted the conversion of lactate into electron acceptors, resulting in a lower ED/EA ratio in the fermenters. The results were consistent with the increased abundance of butyrate-producing bacteria, which were depicted in Section 3.3.2, further confirming that the addition of Fe₃O₄ promoted the conversion of lactate to butyrate.

Figure 6 showed that the ED/EA ratio was significantly higher at an F/M ratio of 1:2 compared with that at an F/M ratio of 6:1. In a semi-continuous reactor without Fe₃O₄, under high-load conditions, a higher ED/EA ratio was observed, which might undermine the chain elongation process and further decrease caproate production. The results implied that the addition of Fe₃O₄ in fermenters with high-load conditions could greatly promote the utilization and conversion of lactate, thereby further promoting caproate production. An appropriate ED/EA ratio was conducive to caproate production. The addition of Fe₃O₄ promoted the conversion of lactate into butyrate and other electron acceptors. Therefore, more electron acceptors were available for the chain elongation process, subsequently resulting in a balanced ED/EA ratio in the fermenters and consequently promoting caproate yield.

In contrast to the materials often reported in the enhancement of MCFA production, such as biochar [20], ZVI [21], and Fe₂O₃ [25], the addition of Fe₃O₄ in this study not only substantially elevated the caproate production but also effectively shortened fermentation time. Furthermore, under high load conditions with the addition of 2.0 g/L of Fe₃O₄, the caproate concentration increased from 16.74 g COD/L to 31.42 g COD/L, and the caproate yield increased from 0.24 g COD/g COD to 0.45 g COD/g COD, increasing by 87.7%. Moreover, the caproate production rate basically was stabilized within 16 days.

3.3. Microbial Community in Batch Fermentation and Continuous Operation of Fermenter

3.3.1. Microbial Diversity in Continuous Operation of Fermenter

Figure 7a,b presented the distribution of the bacterial community at the phylum and genus levels, respectively. The microbial community on Day 2 of the fermenter was significantly different from that by the end of the fermentation. Clostridium_sensu_stricto_11 and Lactococcus were the dominant bacteria on the second day, with relative abundances of 51.41% and 41.20%, respectively. The dominant strains of microorganisms were significantly different from those in the initial period. During fermentation periods (45 d, 90 d, and 140 d) under different organic loading rates, the dominant strains were Caproiciproducens, and their relative abundances were 40.78%, 83.67%, and 68.61%, respectively. Caproiciproducens is a typical genus of caproate-producing bacteria, which can produce caproate by metabolizing sugar and lactate [33]. However, Caproiciproducens in Stage 2 (OLR is 2.71 g COD/L/d) were evidently abundant, and the relative abundance of Latilactobacillus reached 36.32%, which also contained Lactococcus (1.45%), Leuconostoc (3.97%), and Stenotrophomonas (2.36%). All microbes are commonly reported to be able to degrade carbohydrates to produce lactate [34,35]. In addition, it has been reported that Lactobacillus and Stenotrophomonas can increase the content of butyrate in the fermentation process [35,36]. In Stage 3, although the relative abundance of Caproiciproducens increased, the production of caproate strikingly decreased. The increase in the abundance of Leuconostoc seemingly was related to the gradual accumulation of lactate, and the abundance of Latilactobacillus and Caproiciproducens that are related to butyrate production decreased. Although a large number of electron donors were present in the fermenter, the deficiency of butyrate took place, which was a key intermediate for the chain elongation towards caproate. With the step-wise decrease in organic load (Stage 4), the abundance of key bacteria related to lactate and butyrate production gradually elevated, which evidently promoted caproate production. Therefore, it is reasonable to presume that the co-existence of Caproiciproducens, Lactobacillus, Lactococcus, and Leuconococcus played an important role in the carbon chain elongation process, in which Lactobacillus, Lactococcus, Leuconococcus, and Stenotrophomonas seemingly supplied electron donors and acceptors (lactate and butyrate) to Caproiciproducens to biosynthesize caproate.

3.3.2. Microbial Diversity in Batch Fermenters with Added Fe₃O₄

The microbial diversity of the batch fermenters with Fe₃O₄ addition was analyzed by the end of fermentation. The results are summarized in

Table 2. Microbial diversity index table.

Fe3O4 Dosage (g/L)	Coverage (%)	Ace	Chao	Shannon	Simpson
0	99.97	85.52	83.27	1.57	0.33
2.0	99.97	92.34	92.6	1.74	0.32
5.0	99.96	101.47	101.00	1.63	0.32
10.0	99.96	113.05	97	1.54	0.35

. In groups with different concentrations of Fe₃O₄, Ace and Chao indices gradually increased with the increase in Fe₃O₄ addition, indicating that the addition of Fe₃O₄ seemingly enhanced community richness and diversity.

Figure 8 depicted the distribution of microorganisms in the groups with different Fe₃O₄ dosages at the genus level. Noticeably, the four groups presented similar microbial communities. These included Lactococcus, Latilactobacillus, and Lactobacillus; butyrate-producing bacteria (Latilactobacillus and Stenotrophomonas); and carbon chain elongation bacteria (Caproiciproducens and Clostridium_sensu_stricto_11). Fe₃O₄ did not cause significant changes in microbial structure, and there was a significant increase in the abundance of certain functional microorganisms. Therefore, Caproiciproducens still dominated the microbial community, compared with the control group (84%). A higher relative abundance of 91%, 90%, and 92% was observed in groups with Fe₃O₄ (2.0, 5.0, and 10.0 g/L), which was also consistent with differences in caproate concentrations at the end of fermentation. Moreover, more abundant Latilactobacillus and Leuconostoc were noticeable in groups supplemented with Fe₃O₄ (2.0, 5.0, and 10.0 g/L). These results implied that accompanied by Fe₃O₄ addition, the relative abundances of both acid-producing microorganisms and caproate-producing microorganisms increased, thus promoting caproate production.

4. Conclusions

The results of this study indicate that Fe₃O₄ addition evidently enhanced the yield of caproate derived from food waste and was conducive to mitigating the inhibitory effects of high organic loads on the chain elongation process. Compared to the control group, the maximum yield of caproate increased by 22.8% at an F/M ratio of 6:1 and 2.0 g/L of Fe₃O₄. Although the addition of Fe₃O₄ did not significantly change the overall profile of the microbial community, the abundance of certain key microorganisms substantially increased. In particular, the carbon chain elongation functional microbes such as Caproiciproducens reached 91%, 90%, and 92% in the groups of Fe₃O₄ of 2.0 g/L, 5.0 g/L, and 10.0 g/L, respectively. In addition, the high abundance of Latilactobacillus and Leuconostoc in the Fe₃O₄ group was evidently related to the enhancement of caproate yield in the in-situ fermenter. Nevertheless, due to the rapid promotion of the conversion of lactate to butyrate by Fe₃O₄ at high concentrations, a deficiency of lactate took place and was responsible for the poor chain elongation process towards caproate. The results also confirmed that lactate can be used as an electron donor for chain elongation to generate electron acceptors (i.e., butyrate), thus ensuring proper balance of the ratio of ED to EA in both butyrate and caproate production processes is of great importance to consequently obtain a high yield of caproate-derived food waste during fermentation. The current study provided insights into optimizing anaerobic fermentation processes for enhanced caproate yield, which has implications for sustainable food waste management and resource recovery.

5. Patents

There was a patent about this study.

Title: Enhancing in-situ fermentation of food waste for caproic acid biosynthesis via Fe₃O₄ addition). Application Time: 10 December 2024.