Pinewood sawdust was selected as typical lignocellulose material to compare the effect of hydrochar from lignocellulose, and nitrogen-containing biowaste on the anaerobic digestion process nitrogen-containing biowastes including food waste, digested sewage sludge, and Chlorella were selected as nitrogen-rich feedstock. The proximate analysis of the feedstock is shown in Table 1. For each run of hydrothermal carbonization, a mass ratio of 85% (feedstock/water, close to the moisture) was prepared for the feedstock, meaning approximately 170 g water and 30 g dried biowaste was added, which was close to the natural moisture content of the feedstock. The reactor was heated at a rate of 5°C/min, and a final residence temperature was set as 250°C. After hydrothermal carbonization, the reactor was cooled to room temperature, and the mixture was separated by vacuum filtration. The solid residue, food waste hydrochar (F-C), Chlorella hydrochar (C-C), sludge hydrochar (S-C), and pinewood hydrochar (P-C) with certain moisture were further dried in an oven at 105°C until a constant mass was reached. After grinding, all the hydrochar samples were used to pass through a 40-mesh (0.63 mm) sieve.

Hydrochar was used as an additive during anaerobic digestion, and the amount of hydrochar was set as 2.5 g/L. To mimic the anaerobic digestion, synthetic wastewater was used as the substrate. The COD concentration of synthetic wastewater was set at 5 g/L. The granular anaerobic sludge was used for inoculum with a volume of 40 mL. For comparison, the anaerobic digestion without hydrochar addition was set as the control group. Anaerobic digestion was conducted using an automatic methane potential test system-II (AMPTS II, Bioprocess Control AB, Lund, Sweden). The total working volume was 400 mL and with 200 mL headspace. The liquid samples were taken every day, and total organic carbon (TOC) and low VAFs concentrations were measured. Each bottle was fed with 400 μL nutrient stock solution, and each liter of nutrient stock solution contained 0.48 g CoCl₂⋅6H₂O, 0.1 g CuCl₂⋅2H₂O, 1.3 g ZnCl₂, 5 g MnCl₂⋅4H₂O, 2.5 g FeCl₃⋅6H₂O, 0.4 g NiSO₄⋅6H₂O, 0.1 g AlK (SO₄)₂⋅12H₂O, 0.25 g Na₂MoO₄⋅2H₂O.

The TOC of aqueous samples was measured by a TOC analyzer equipped with a Nondispersive infrared (NDIR) gas detector (TOC-V CPN, Shimadzu, Japan). The VFAs were analyzed through gas chromatography (Shimadzu Nexis-2030, Japan) which was equipped with a flame ionization detector (FID) and capillary column (SH-Stabilwax-DA). Phosphoric acid (30% wt.) was added to the samples before injecting to prevent the denaturation of some volatile acids. The nitrogen was used as the carrier gas with a flux of 4 mL/min. The temperature of the inject port and the detector were both maintained at 250˚C. Elemental analysis of the hydrochar was carried out using Elementar Vario EL, and the elemental composition of feedstock is shown in Table 2. A scanning electron microscope (JSM 7800F Prime) was used to investigate the surface morphology features. The Raman measurements were performed on a customized Raman microscope equipped with a Pixis-100BR CCD (Princeton Instruments, USA), an Acton SP-2500i spectrograph, and a He Ne laser (14 mW). The surface functional groups of hydrochar were investigated using a Fourier transform infrared spectroscopy (FTIR) (PerkinElmer Spectrum Two, USA).

The methane yield from anaerobic digestion with different hydrochar addition was presented in Fig. 1. As observed, the anaerobic digestion process was finished within approximately 40 h for all sets except for the F-C, and the methane yield of all groups reached a peak within 14∼18 h. The total methane was only 218 mL for the F-C group, about two-thirds of the control group, while the cumulative methane volume of the S-C and P-C groups was close to that of the control group. This difference suggested that F-C addition was not favored for the anaerobic digestion process, and inhibiting effect can be seen during the initial stage (13∼17 h). As for the C-C group, the total methane (about 345 mL) was higher than that of other groups, and a promotion effect on anaerobic digestion methanogenesis can be observed. There was a slight difference in methane production between the S-C, P-C group, and control groups. As a whole, with the same dose of hydrochar, methane volume was different, and it was speculated that F-C may not suitable for promoting methane production.

To explore the effect of hydrochar on the metabolism of the organics, the TOC removal efficiency and variation of VAFs concentrations were investigated (Fig. 2). It can be observed that the TOC removal rate of the control, S-C, C-C, and P-C group on day-1 reached up to approximately 80%, indicating that most of the organics in the reaction system were consumed within 24 h, while the TOC completely consumed with 48 h. However, the TOC removal rate of substrate for F-C was only 51% on day-1, and this value increased by about 13% on day-2, which indicated 26% of organics still remained. Based on the VAFs analysis, a high content of VFAs 1.4 g/L was noticed on day-1, and this value increased to1.6 g/L on day-2, so it can be speculated that the accumulation of VAFs aroused by F-C was the dominant reason for the low TOC removal. On the contrary, the addition of C-C showed the best result with the lowest VAFs on day-2. This comparison confirmed that F-C led to VFAs inhibition instead of a degradation promotion.

Acetate, propionate, butyrate, pentanoate, and ethanol were the main products during acidification (Table 3). A higher distribution of VFAs and ethanol with hydrochar addition was observed. In particular, the much higher acetate, ethanol, and butyrate concentration in C-C, S-C and P-C group than that in the control group was observed. Combined with the similar TOC removal in C-C, S-C, and P-C groups and the control group, hydrochar derived from sludge, Chlorella, and pine may facilitate the VFAs and ethanol production from glucose compared to the control group. A similar result was also confirmed by Bu et al. [23] that biochar addition was assumed to promote acidification via enhancing electron transfer efficiency, stimulating bacterial growth, and improving critical enzymatic activities. It should be noted that F-C had the lowest TOC removal, relatively lowest acetate concentration 0.285, and higher butyrate 0.868 g/L on day 1. Tang et al. proposed the multiple roles of biochar in anaerobic digestion, including pH buffering, functional microbes enrichment, inhibitors alleviating, and regulating the hydrolysis, acidification, and methanogenesis [9]. It could be speculated that the F-C addition accelerated the acidification since the highest VFAs production on the day 1 (Table 3), compared to others. However, it inhibited acetogenesis and methanogenesis, and the methane production confirmed this in F-C group (Fig. 1), which suggested the shortest lag phase but the lowest methane production. It should be noted that all the hydrochar addition enhanced ethanol production. The ethanol could stimulate DIET’s establishment by providing electrons and by enriching exoelectrogens and electrotrophic methanogens for co-digesting complex organic wastes [24].

Accumulated VFAs in F-C assisted anaerobic digestion was found on the second day, and the concentration of acetate and butyrate reached up to 0.267 and 1.161 g/L, respectively, which were higher than that of other groups (Table 4). The accumulation of butyrate indicated the inhibition of methane production, and it can be observed that butyrate accounted for about 70% of the total VFAs for the F-C group. Acetate in the P-C group was completely consumed, indicating that the lignocellulose-derived hydrochar was favored for the acetotrophic methanogenesis, while the concentration of butyrate (0.398 g/L) was higher than that of S-C and C-C group. The low conductivity of F-C may account for the poor ability to develop DIET and enhance anaerobic digestion. In addition, food waste is known as a suitable feedstock for bio-oil production via hydrothermal conversion due to its high protein and lipid content. The produced F-C may mix with a small amount of bio-crude oil, which contains nitrogen heterocyclic and phenolic substances [25], and they would cause toxicity to the acetogens and methanogens, which further results in the accumulation of organic acids [26].

The surface functional groups in this work were analyzed by FTIR since oxygen-containing groups like C=O and -OH benefit the development of DIET [22]. As observed in Fig. 3, the 3600∼3300 cm⁻¹ were assigned to overlapping peaks of hydroxyl and amino groups. This peak was more obvious for P-C, indicating that N-containing biowaste-derived hydrochar had limited oxygen groups [27]. All hydrochar consisted of distinct aliphatic methylene bands at 2925 and 2855 cm⁻¹, and it indicated the presence of aliphatic structures. The C=O stretching vibrational band of hemicellulose at 1730 cm⁻¹ suggested the presence of oxygen-containing functional groups on the surface of hydrochar [28]. This peak can be observed in all hydrochar except S-C, which may be attributed to the hemicellulose components, while sludge had a limited content of hemicellulose [29]. The peak at 1600 cm⁻¹ belonged to the aromatic C=C stretching in aromatic groups ascribed to the lignin content and this peak was more obvious in P-C than in other samples. The band around 1080 cm⁻¹ (1060–1260 cm⁻¹) was assigned to the C-O-C aliphatic ethers and C-O-C stretching in ether, and limited peaks in this range were found in S-C since ash dominated the main components in sludge instead of carbohydrate [30]. It can be concluded that all hydrochar preserved rich oxygen-containing groups like C=O and C-O, and it speculated that S-C and F-C had relatively low -OH due to the carbonization degree and ash content, respectively. Nevertheless, the nitrogen-containing components may not be the main reason for the difference in the electron transport of oxygen-containing functional groups, but the toxic substances contained in the hydrochar from F-C may account for this [31].

The surface morphological characteristics of all hydrochar samples are shown in Fig. 4. A large number of holes are generated on the surface of P-C, and the size of the particle is small. The microspheres are mainly formed by the carbonization of xylose and cellulose in pine [32]. Unlike lignocellulose from pine which undergoes the hydrolysis process, food waste can be easily carbonized and a large size (1∼3 μm) of hydrochar microsphere was formed. In contrast, few spherical particles and more irregular particles were observed for the S-C since its fixed carbon content was low. Thus, a porous structure on the surface of the hydrochar can be observed. However, algae hydrochar also had a large microsphere incorporated into the surface, which is different from that of F-C. No obvious relationship can be built between the surface morphological structure of hydrochars and methane production performance. It suggested that the main difference in methane production may not be attributed to hydrochar’s surface morphological structure but the involvement and conductivity of toxic substances of hydrochar.

The carbon content in hydrochar increased from 42.87%–52.45% to 64.87%–72.64% in C-C, F-C, and P-C, except for S-C due to the ash content (Table 5). Especially the carbon content for F-C reached up to 72.64%, which suggested that F-C had a high carbonization degree coupled with the low O/C and H/C ratio. Additionally, it can be clearly seen that P-C had the highest oxygen content (25.17%). The carbon content for F-C reached 72.64%. As for N, the ratio in hydrochar was reduced from 1.52%–7.54% to 1.25%–6.45%, and P-C remained the lowest N content.

Fig. 5 illustrates the Raman spectroscopy and the I_D/I_G value of hydrochar from different biowaste. The parameter of I_D/I_G can be used for the estimation of the graphitisation extent or graphite-like carbon structures in hydrochar [33]. As observed, C-C had the highest I_D/I_G value of 1.07, while the I_D/I_G value for S-C and F-C were 0.97 and 0.95, respectively. This suggested that C-C had the lowest degree of graphitization coupled with properly poor conductivity for the development of DIET, while the enhancement of biogas production may depend more on surface groups. The production of methane can also confirm this. However, P-C had the lowest I_D/I_G value (0.93). The high graphitization degree of P-C may be ascribed to the carbonization of components like cellulose and xylose by the formation of aromatic structures [34,35]. On the other hand, P-C was the only biowaste with low N content, and limited N can be incorporated into the matrix of carbon from hydrochar since it suggested that high temperature and Millard reaction can enhance the stable N forms formation like pyridine-N and quaternary-N, which can provide good conductivity for the hydrochar [36,37]. Based on this characteristic, it can be inferred that the DIET of pine-hydrochar in this work may rely more on the conductivity of body structure or physical structure rather than the surface oxygen-containing groups. Nevertheless, hydrochar from biowaste with rich nitrogen resources may depend more on the surface oxygen-containing groups due to the introduction of elemental N.