Co-hydrothermal carbonization of microalgae and digested sewage sludge: Assessing the impact of mixing ratios on the composition of primary and secondary char

The role of microalgae cultivation in wastewater treatment and reclamation has been studied extensively, as has the potential utility of the resulting algal biomass. Most methods for processing such biomass generate solid residues that must be properly managed to comply with current sustainable resource utilization requirements. Hydrothermal carbonization (HTC) can be used to process both individual wet feedstocks and mixed feedstocks (i.e., co-HTC). Here, we investigate co-HTC using microalgae and digested sewage sludge as feedstocks. The objectives were to (i) study the material ’ s partitioning into solid and liquid products, and (ii) characterize the products ’ physicochemical properties. Co-HTC experiments were conducted at 180 – 250 ◦ C using mixed micro-algae/sewage sludge feedstocks with the proportion of sewage sludge ranging from 0 to 100 %. Analyses of the hydrochar composition and the formation and composition of secondary char revealed that the content of carbonized material in the product decreased as the proportion of sewage sludge in the feedstock increased under fixed carbonization conditions. The properties of the hydrochars and the partitioning of material between the liquid phase and the hydrochar correlated linearly with the proportion of microalgae in mixed feedstocks, indicating that adding sewage sludge to microalgae had weak or non-existent synergistic effects on co-HTC outcomes. However, the proportion of sewage sludge in the feedstock did affect the secondary char. For example, adding sewage sludge reduced the abundance of carboxylic acids and ketones as well as the concentrations of higher molecular weight cholesterols. Such changes may alter the viable applications of the hydrochar.


Introduction
The cultivation of microalgal biomass in nutrient-rich wastewater containing phosphorous and nitrogen has been studied extensively in recent decades as a more resource-efficient alternative to cultivation in freshwater (Gentili, 2014;Zhou et al., 2014).The potential applications of the resulting microalgal biomass depend strongly on its accumulation of micropollutants from the wastewater, which may limit the use of the extractables (e.g.nutrients and platform chemicals) from the biomass as well as any byproducts, and residual streams (Gentili and Fick, 2017).The utilization of residual streams in particular is important for resource-efficiency because it increases circularity.The main residual stream from microalgae processing consists of the extracted microalgae but low quality and/or low purity batches with high contents of contaminants, such as pharmaceuticals and heavy metals, are unsuitable for resource recovery (Lage et al., 2018) and are therefore seen as a separate residual stream.
Thermochemical conversion is a technique that can be used to process both microalgae and residual microalgae.The high moisture content of microalgae leads to high energy consumption during dry thermochemical conversion processes such as incineration and pyrolysis, making processing techniques that can tolerate wet feedstocks more attractive.Hydrothermal carbonization (HTC) is a comparatively mild alternative thermochemical conversion process performed at temperatures of 150-250 • C at the vapor saturation pressure of water (Wang et al., 2019).HTC transforms the feedstock into a solid product (hydrochar), an aqueous fraction rich in organic compounds, and a small amount of gas -mainly CO 2 (Libra et al., 2011).Some studies have reported the fundamental aspects on hydrochars from HTC-treated microalgae (Benavente et al., 2022;Castro et al., 2021;Ekpo et al., 2016a) while others have shown that their properties may lend themselves to fertilizer (Chu et al., 2021) and solid fuel applications (Lee et al., 2019).However, the use of microalgae as the sole feedstock for industrial-scale hydrochar production could present challenges relating to costs of feedstock production and seasonal availability.One way to secure feedstock availability, reduce costs, and potentially exploit synergistic chemical effects would be to co-process microalgae with readily available and low-demand side streams such as sewage sludge.
Sewage sludge is a wastewater treatment by-product that is generated in large quantities (Song et al., 2019).It has a complex chemical composition and may contain a wide range of organic and inorganic contaminants (e.g., heavy metals and pharmaceuticals) as well as microbes (Peng et al., 2017;Wang et al., 2021).If not properly managed, sewage sludge can be hazardous to the environment, animals, and humans (Peng et al., 2017).Like microalgae, it has a high moisture content that complicates thermochemical processing.Co-HTC of microalgae and sewage sludge could be performed at wastewater treatment plants that produce both waste streams, thereby valorizing them while also minimizing transport-related costs and environmental impacts.
Previous studies on co-HTC using different lignocellulosic residues such as agricultural and yard waste together with non-cellulosic residues such as sludge, manure, and food waste have shown that combining different feedstocks in this way can have positive synergistic effects on certain outcomes such as hydrochar yield, carbon retention, organics retention and fuel properties (Ma et al., 2019;Wang et al., 2022;Zhang et al., 2020;Zhang et al., 2017).
During HTC, the organic fraction of the feedstock undergoes a series of carbonization reactions that first produce gaseous products (formed via decarboxylation and demethanation) and soluble products in the liquid phase (formed via hydrolysis, dehydration, and depolymerization) (Funke and Ziegler, 2010).Some of the water-soluble components then undergo various condensation, re-polymerization, Maillard, and Mannich reactions (that form N-heterocyclic compounds) and precipitate as a solid phase known as secondary hydrochar that is soluble in organic solvents.Subsequent solid-solid reactions in the hydrochar lead to aromatization and the formation of the insoluble primary hydrochar, which contains heterocyclic structures (Funke and Ziegler, 2010;Sevilla and Fuertes, 2009;Titirici et al., 2007).The physicochemical properties of the hydrochar depend on the composition of the secondary hydrochar and determine the range of applications for which it is suitable.Secondary hydrochar consists mostly of organic acids, phenols, ketones, aldehydes and furfurals (Benavente et al., 2022;Lucian et al., 2018)), while the primary hydrochar determines the main structure and morphology of the product (Knežević et al., 2010).The secondary hydrochar can be extracted with organic solvents and its formation is promoted by high temperatures, high feedstock loads, and extended reaction times (Benavente et al., 2022;Lucian et al., 2018).
In previous studies we examined the composition of primary and secondary hydrochars derived from microalgal feedstocks at different HTC temperatures and varying reaction times (Benavente et al., 2022), and the potential for using mineral and organic acids to recover phosphorus from hydrochars derived from digested sewage sludge (Pérez et al., 2022(Pérez et al., , 2021)).However, only one published study on the co-HTC of microalgae with sewage sludge has investigated the effect of combining primary (i.e.not digested) sludge and microalgae on the solid fuel properties of the resulting hydrochar (Lee et al., 2019), which were found to be comparable to low-grade coal.Because co-HTC of sewage sludge with other biomass fractions has shown positive synergetic effects, it would be desirable to investigate the characteristics and potential applications of hydrochar formed by co-HTC of sewage sludge/microalgae mixtures, which could be a powerful tool for valorizing these feedstocks.Both feedstocks could be collected from the same wastewater treatment plant, so adding a valorization step would be efficient in terms of both costs and resources.This work therefore investigates the impact of mixing ratios on the composition and yield of hydrochar formed by co-HTC of microalgae and sewage sludge.The specific objectives were to (i) determine how the proportion of sewage sludge in mixed microalgae-sewage sludge feedstocks affects the co-HTC reaction in terms of solid (i.e., primary and secondary hydrochar plus the ash fraction) and liquid yields at different carbonization temperatures, (ii) determine the physicochemical properties of the resulting hydrochars, and (iii) characterize the chemical composition of the secondary hydrochar.The results presented here clarify the composition of hydrochars formed by co-HTC of microalgae and sewage sludge, which may facilitate future assessments of the potential applications of these waste streams.

Microalgae polyculture
A microalgal polyculture provided by the Swedish University of Agricultural Sciences (SLU; Umeå, Sweden) was cultivated in 6 m 3 outdoor ponds with untreated wastewater from a local municipal wastewater treatment plant (WWTP) (Vakin, Umeå, Sweden), using flue gases from a nearby combined heat and power plant as a CO 2 source.Further details of the cultivation system are available in an earlier publication (Lage et al., 2021).Each microalgae batch consisted of pooled microalgae samples centrifuged at 5000 rpm to form a paste with a moisture content of 85 % that was kept in a freezer at − 18 • C until needed.

Sewage sludge
Anaerobically digested and mechanically dewatered sewage sludge was collected in the spring of 2021 from the local WWTP (Vakin, Umeå, Sweden).Sewage sludge samples were packed in plastic bags and stored at − 18 • C until use.The moisture content of the sewage sludge was 72 %.

Hydrothermal carbonization experiments
HTC processing was conducted in a 1 L stirred stainless steel reactor (Büchiglasuster, Switzerland) at temperatures of 180, 215, and 250 • C. Before HTC processing, the microalgae-sewage sludge feedstocks were stirred for 16 h to ensure homogenization.Then, the feedstock was heated at 5 • C/min until the specified temperature was reached.All batches were processed for 2 h under autogenous pressure and stirred at 1000 rpm.
For each run, the reactor was loaded with a 600 g slurry of ultrapure water (Millipore, MA, USA) and microalgae, sewage sludge, or a microalgae-sewage sludge mixture with a total solid load of 11 %.The proportion of sewage sludge in the slurries was varied from 0 to 100 %.In total, 15 hydrochar samples were produced.The samples were assigned systematic designations of the form XX%-YYY, where XX denotes the content of sewage sludge (as a percentage), and YYY is the processing temperature ( • C) (Table 1).

Table 1
Experimental matrix outlining the process temperatures and microalgae (MA)sewage sludge (SS) proportions.
After processing, the reactor was cooled at 5 • C/min using an internal water-cooling system and then depressurized by releasing the gas phase.The hydrochar was separated from the processed slurry by centrifugation at 3600 rpm for 10 min at 20 • C. The HTC liquor was filtered through cellulose quantitative filter paper (pore size: 2-3 µm), and immediately frozen at -18 • C. The hydrochar was washed three times with ultrapure water by shaking and centrifugation before being dried overnight at 105 • C, sieved to a particle size of < 0.5 mm, and stored in air-tight bottles.

HTC liquor characterization
The concentrations of cationic inorganic elements (Al, Fe, Ca, Mg, P, Na, and K) in the HTC liquors were determined spectroscopically by ICP-OES in dilute HNO 3 using an Optima 200 DV instrument (PerkinElmer, Überlingen, Germany).The concentrations of heavy metals (Mn, Co, Ni, Cd, Zn, As, Pb, and Cu) were determined by ICP-MS in dilute HNO 3 using an Agilent Triple Quadrupole ICP-MS operating in single quadrupole mode.

Proximate and ultimate analysis
Proximate analysis of microalgae, sewage sludge, and hydrochars was conducted using a thermogravimetric analysis (TGA) method for measuring the contents of moisture, volatile matter (VM), fixed carbon (FC), and ash that was developed by (Saldarriaga et al., 2015) and adapted by (Benavente et al., 2022).The proximate analysis was performed in a Q5000IR thermogravimetric analyzer (TA Instruments).TGA was also used to determine the composition of the extracted hydrochars (i.e., primary chars) formed at 250 • C. The DTG profiles obtained were used to conduct a qualitative compositional analysis of the hydrochars and assess differences in composition and material fractions.Elemental composition (CHNS) was determined using a Thermo Finningan Flash 1112 Series Elemental Microanalyzer.Oxygen contents were determined by subtracting the masses of ash and CHNS from the total dry sample mass.

Inorganic elements in solids
The concentrations of inorganic elements in the microalgae, sewage sludge, and hydrochars were determined by ICP-OES and ICP-MS, as described above.Solid samples were submitted to microwave digestion with aqua regia according to the EPA 3051A procedure and each liquid digestate was filtered through a 0.45 µm syringe filter before analysis.

Characterization of secondary char by GC-MS
To separate the extractable fraction (i.e., secondary char) from the insoluble hydrochar structure (i.e., the primary char), pristine hydrochars were extracted with dichloromethane (DCM) (analytical grade, Merck KGaA, Germany) based on the procedure described in our previous study (Benavente et al., 2022).Briefly, 0.5 g pristine hydrochar was mixed with 10 mL of DCM in a glass centrifuge tube, shaken in an orbital rotator for 1 h at 20 • C, and then centrifuged at 3500 rpm for 10 min.The residual solid hydrochar (i.e., the extracted hydrochar) was then washed three times with 4 mL of fresh DCM using the same shaking and centrifugation procedure.All of the DCM fractions for each sample (i.e., the first extract and all three washes) were then combined after filtration through 0.45 µm PTFE syringe filters and evaporated to dryness.The dry extracts (i.e., the isolated secondary char) were weighed to calculate the secondary char yield and then dissolved in 1.5 mL DCM for quantitative GC-MS analysis to identify major components.The extracted hydrochars (i.e., primary chars) were dried at 65 • C and stored in airtight containers.The primary char yield was calculated by subtracting the yields of secondary char and ash from the total solid yield.
GC-MS analysis of the extractable fraction (i.e., the secondary char) was performed using an Agilent 6890 GC coupled to an Agilent 5975 inert MSD (Agilent Technologies Inc.) with a ZB-5MS capillary column (30 m, 0.25 mm i.d., 0.25 µm film thickness; Zebron, Phenomenex) using unit mass resolution and scan mode over the mass range m/z 50 to 550.Pretreatment of non-target data was done using MATLAB custom scripts (MATLAB R2017b, Mathworks, Natick, MA) as described previously (Jonsson et al., 2005).Each identified substance was assigned to a compound group associated with a feedstock fraction (fatty acids, carbohydrates, lignin, or proteins) known to promote its formation.The GC-MS analysis of the extractives was mainly qualitative, but semiquantitative estimates of each compound's abundance were obtained based on peak heights (Spokas et al., 2011) and the abundance of each compound group was expressed by comparing its peak area to that of the chromatogram as a whole.

HTC performance: Solid and liquid yields
The moisture content of the hydrochar decreased from 34.4 to 40.2 % to 16.4-27.8% as the HTC process temperature increased from 180 to 250 • C (Fig. 1), due to more severe carbonization and therefore, higher hydrophobicity of the hydrochar.Moreover, the moisture content of the hydrochar increased with the proportion of sewage sludge in the feedstock.The overall hydrochar yields (on dry basis, db) were subdivided into yields of primary hydrochar (non-extractable fraction), secondary hydrochar (extractable fraction), and ash (Figs. 1 and 2).As expected, the overall hydrochar yields decreased as the process temperature increased, irrespective of the feedstock composition (Fig. 1).This was likely due to increased feedstock decomposition into HTC liquor and gas products (Zhai et al., 2017).The proportion of sewage sludge had a larger influence on the overall hydrochar yield than the HTC process temperature but its strongest effect was on the ash yield.The proximate analysis (Table 2) revealed that sewage sludge contained 2.6 times more ash-forming elements than microalgae while having only 70 % of its carbon content.Accordingly, the ash yields for hydrochars formed from pure microalgae were just 12.1-21.9% while those for hydrochars formed from pure sewage sludge at the same temperature were 37.1-37.5 %.When comparing hydrochars formed from mixtures with the same sewage sludge content, the ash yield did not vary appreciably with the process temperature, fluctuating by less than 2 %.
Increasing the proportion of sewage sludge in the feedstock mix while keeping the HTC process temperature constant also increased the primary char yield.At 180 • C, the primary char yield (ash-free) obtained using pure microalgae (36 %) was 16 % lower than that for pure sewage sludge (52 %; see Fig. 2-B).Excluding the ash fraction, the total hydrochar yield for pure microalgae (44 %) was 12 % lower than that for pure sewage sludge (56 %; see Fig. 2-A).This increase in primary char yield with the content of sewage sludge in the feedstock became more pronounced at higher process temperatures and suggests that the presence of sewage sludge reduces the rate at which the solid product is degraded and carbonized.Accordingly, the moisture content of the hydrochar also increased with the proportion of sewage sludge in the feedstock.These results imply that the organic fraction of sewage sludge is less reactive than that of microalgae under the same carbonization conditions, probably due to the greater stabilization of sewage sludge during anaerobic digestion in the WWTP.It also suggests that the high inorganic content of sewage sludge has little or no catalytic effect on the degradation processes.As a result, the amount of HTC liquor decreased as the proportion of sewage sludge increased at all three tested temperatures (Fig. 2-D) because the formation of degradation products soluble in the liquid phase was reduced.This trend became more pronounced at higher temperatures (Fig. 2-D), suggesting that the soluble degradation products had a higher tendency to undergo further reactions to form products that precipitated into the solid phase as the temperature increased.This hypothesis is consistent with the observation that secondary char yields increased with the process temperature and the proportion of sewage sludge in the feedstock (Fig. 2-C).The literature shows that the amount of secondary char is directly dependent on the type of feedstock.For example, a recent study by Ischia et al. (2023) showed that the extraction of cellulose-derived hydrochars produced at 190, 220 and 250 • C with different combinations of solvents (acetone, ethyl acetate, and methanol) led to the highest secondary char yields at 220 • C, which was explained by the conversion of cellulose to cyclic/aromatic components (Ischia et al., 2023).However, for pure sewage sludge, the secondary char yield increased from 2.4 % at 180 • C to 6.4 % at 250 • C (Fig. 1), corresponding to a 6.7 % increase on an ashfree basis.Therefore, the degradation of sewage sludge at higher temperatures produces more soluble compounds in the HTC-liquor that  become available for formation of secondary char.This is consistent with the finding that HTC liquor recovery declined as the feedstock's sewage sludge content increased at 250 • C (see Fig. 2-D).At this temperature, the amount of HTC-liquor differs by 12 % between the pure microalgae (0.84 L/kg slurry) and the pure sewage sludge (0.74 L/kg slurry).
The proportion of sewage sludge in the feedstock varied linearly with the moisture content of the hydrochar and the yield and mass of HTC liquor (Figs. 1 and 2), suggesting that the effects of mixing microalgae with sewage sludge under the studied HTC conditions are primarily additive and that if synergistic effects exist, they are at best very limited.

Proximate and ultimate analyses
Because of the solubilization and decomposition reactions that occur during HTC, the volatile matter (VM) content of the hydrochars was lower than that of the microalgae and sewage sludge feedstocks (Table 2).Raising the processing temperature increased the content of fixed carbon (FC) in the hydrochars; the FC content of hydrochar obtained by processing pure microalgae at 250 • C was 20.3 %, while that of the feedstock was 15.5 %.This indicates that increasing the process temperature caused a higher degree of carbonization and VM removal, which could be beneficial if the hydrochars were used as solid fuels because it would help maintain a more stable flame when they were burned (Zhang et al., 2021).The sewage sludge had a lower initial VM content than the microalgae and consequently exhibited less VM removal after HTC.Therefore, increasing the proportion of sewage sludge in the feedstock reduced the FC content of the hydrochar at a given HTC processing temperature.For example, the FC content of hydrochars obtained by processing feedstocks containing 25 % and 75 % sewage sludge at 250 • C were 17.0 % and 13.7 %, respectively.These results are similar to those reported by (Lee et al., 2019) from co-HTC of primary sludge and microalgae.This is probably due to the different starting compositions of the raw microalgae and sewage sludge (Table 2) but could also indicate that sewage sludge undergoes less severe carbonization than microalgae, in accordance with previous observations.However, the fuel ratio for mixed microalgae-sewage sludge hydrochars at their respective processing temperature were consistent, suggesting that the behavior of the co-hydrochars in combustion processes would be similar, regardless of the proportion of sewage sludge in the mix (Table 2).
The carbon (C) content of the hydrochars ranged from 25.7 to 49.7 % (Table 2), with those formed at higher temperatures having higher C contents.Pure microalgae hydrochars had the highest C contents (49.7 % at 250 • C, compared to 28.3 % for hydrochar formed from pure sewage sludge).In addition, the C content decreased as the proportion of sewage sludge in the feedstock increased due to the higher concentration of ash-forming elements in sewage sludge (Table 2).To better understand the changes in elemental composition during HTC and to assess the degree of coalification of the products, Van Krevelen diagrams were plotted showing the molar H/C and O/C ratios of the feedstocks and hydrochars (Fig. 3) (Reza et al., 2014).The hydrochar H/C and O/C ratios exhibit a linear downward trend towards the origin of the plot, suggesting that dehydration was the main mechanism of carbonization during co-HTC processing.As proposed by (Xu et al., 2019), dehydration is associated with degradation of carbohydrates via hydroxyl group elimination, suggesting that the changes observed in these samples may be driven by their carbohydrate content.Conversely, proteins can only participate in hydrochar formation through Maillard reactions that occur in the presence of carbohydrates (Li et al., 2019).The carbohydrate content of sewage sludge is typically 8-15 % db (Wang et al., 2019), whereas that of microalgae is around 26 % db (Benavente et al., 2022).Together with the stabilization of the sewage sludge before HTC processing, this might explain why the sludge exhibited comparatively low reactivity.
The nitrogen (N) content of the hydrochars was lower than those of the pure microalgae and sewage sludge feedstocks, and decreased as the HTC processing temperature increased (Table 2).In addition, the hydrochar N content decreased as the sewage sludge content of the feedstock increased: for hydrochars formed at 250 • C, the N content was 3.9 % when using pure microalgae as the feedstock but only 2.3 % when using pure sewage sludge (Table 2).Nitrogen transfer from the hydrochar to the HTC liquor can occur via solubilization of inorganic ammonium salts and hydrolysis of proteins (Kruse et al., 2016) and amino acids into soluble organic compounds (Sumida H and Sumida H, 2013).The transfer rate is thought to depend on the types of N species present in the feedstock (e.g., salts or organic compounds such as proteins and amino acids), their oxidation state and chemical speciation (e. g., inorganic N may be present as ammonium, nitrate, or nitrite) (Kruse et al., 2016), and the pH (Ekpo et al., 2016a, Ekpo et al., 2016b).

Thermogravimetric analysis of hydrochars
The TG and DTG analyses showed that microalgae and sewage sludge volatilized in 3 and 4 stages, respectively (Fig. 4).Microalgae exhibited substantially greater weight loss than sewage sludge, which can be attributed to its higher initial content of volatile matter (Table 2).Stage 1 (<250 • C, peak 1, Fig. 4) corresponds to the release of weakly bonded  water molecules and desorption of highly volatile substances.Changes in lipid structure, thermal unfolding of proteins, and the initial steps of carbohydrate decomposition also occur during this stage (Pane et al., 2001).The peaks between 250 and 450 • C, corresponding to stages 2 and 3 of volatilization for microalgae (peaks 2 and 4 in Fig. 4) and stages 2 to 4 for sewage sludge (peaks 2, 3, and 4 in Fig. 4) are associated with decomposition of proteins, lipids, and carbohydrates, respectively (Naqvi et al., 2018;Pane et al., 2001).They may also reflect desorption of higher-molecular weight substances (Benavente et al., 2022).As shown in Fig. 4, the sewage sludge profile features one peak in the temperature range associated with protein decomposition (peak 3) that is absent in the microalgae profile.The presence of this peak may be due to differences in the chemical structure and properties of the organic and inorganic components present in microalgae and sewage sludge.For instance, literature reports indicate that sewage sludge contains macromolecular organic nitrogen originating from the proteins of microorganisms and debris (Sumida H and Sumida H, 2013) as well as dead bacteria (Shao et al., 2008;Tian et al., 2014).
For pure microalgae, increasing the HTC temperature reduced and delayed the release of VM during stage 2. In addition, there was a mass gain in stage 3 that was associated with the deposition of high molecular weight compounds (peak 4), in accordance with our previous observations (Benavente et al., 2022).These changes are consistent with the decomposition of the feedstock into liquid and gas products followed by repolymerization and aromatization reactions that cause the deposition of high molecular weight extractives onto the solid matrix and the formation of more stable chemical structures.These processes are all favored by high carbonization temperatures (Zhang et al., 2021), as are the trends observed in Fig. 4.
Weight loss peaks 2, 3 and 4 in the sewage sludge hydrochar profiles became less pronounced as the carbonization temperature increased, which is consistent with more extensive solubilization and/or degradation of sewage sludge components under severely carbonizing conditions (Fig. 4).On the other hand, the carbonization of sewage sludge at 215 and 250 • C also led to a gain of mass in stage 4 (peak 4) that increased slightly with the HTC temperature (Fig. 4).As discussed for microalgae, this mass gain is associated with the deposition of high-M w substances onto the primary hydrochar.Additionally, pure sewage sludge processed at 250 • C exhibited an extra peak at 600 • C (peak 5 in Fig. 4).Similar peaks have been reported previously (Zhang et al., 2021;Zheng et al., 2019) and appear to be characteristic of sewage sludge hydrochars produced at temperatures above 200 • C.
To compare the composition and properties of primary and secondary hydrochars, TGA was performed on the pristine hydrochar formed by HTC processing at 250 • C (the un-extracted hydrochar) and the extracted hydrochar, i.e., the solid material remaining after extracting the pristine hydrochar with DCM (the primary hydrochar).At temperatures of 200-300 • C, the differences between the TGA profiles of the pristine and extracted hydrochars increased with the proportion of microalgae in the feedstock (Fig. 5).This is consistent with the expectation that microalgae have higher contents of extractable lowmolecular weight compounds than sewage sludge (Stage 1, <250 • C, Fig. 5).Conversely, the peak for extractable volatile matter lost at 400-500 • C (stage 4, Fig. 5), which corresponds to high-M w compounds, became more pronounced as the sewage sludge content of the feedstock increased.Therefore, although the total quantity of extractives generated during HTC processing of microalgae exceeds that obtained during HTC of sewage sludge, adding sewage sludge to the feedstock mix shifts the distribution of extractives towards compounds of higher molecular weight.
Although the same compound groups were identified in both microalgae and sewage sludge hydrochars, there were interesting compositional differences between the extracts (Fig. 6 and S2).
Combining sewage sludge and microalgae in the feedstock increased the concentration of higher-M w extractives, as suggested in section 3.2.2.For example, the abundance of cholesterols, which have the highest molecular weights among the identified organic compounds, increased from 9 to 12 % to 52-69 % of the total peak area as the proportion of sewage sludge increased from 0 to 100 %.Conversely, the relative abundances of carboxylic acids and ketones with lower M w decreased substantially (by around 80 %) when comparing hydrochars obtained from pure microalgae and pure sewage sludge (Fig. 6 and S2).The main secondary char precursors were fatty acids, with carboxylic acids and cholesterols comprising over 70 % of the total organic matter.This was presumably due to the relatively high fatty acid content of microalgae.Additionally, sewage sludge hydrochar formed at 250 • C was rich in phenols (Figure S2-D), possibly reflecting the presence of lignin or other high-phenolic substances in the sewage sludge feedstock.
The analysis of the extractives revealed that increasing the sewage sludge content of the feedstock reduced the peak area associated with Nheterocyclic compounds (Figures S2-E, F) formed by Maillard reactions between carbohydrates and proteins and Mannich reactions between carbohydrate and lignin (Zhang et al., 2021).This may be because such N-heterocyclic compounds are further carbonized and integrated into the primary char structure, and therefore contribute to peak 5 in the DTG profiles of sewage sludge hydrochars (Figs. 4 and 5).
The addition of SS to MA had an impact on the composition of the secondary char and, thus, on its potential applications.In a previous study, (Benavente et al., 2022) reported that some organic compounds (i.e., organic acids, aldehydes, ketones and furans, benzene and its derivatives, and phenolic compounds) present in hydrochar may have toxic effects.Another study reported that carboxylic acids, furans and polyaromatic hydrocarbons present in hydrochars (regardless of biomass) may be phytotoxic or leach into groundwater (Karatas et al., 2022).In addition, (Ischia et al., 2023) reported that the extractable fraction of cellulose hydrochars accounted for > 50 % of carboxylic acids.In our study, the hydrochars toxicity seemed to decrease with %SS as a matter of the lower amounts of carboxylic acids formed due to the lower amounts of fatty acids contained in the feedstock (Fig. 6).Therefore, it is clear that the amount of extracted compounds highly depends on type of feedstock (lignocellulosic or non-lignocellulosic waste), and that the removal of these types of toxic compounds is highly environmentally relevant (Benavente et al., 2022;Ischia et al., 2023).In fact, a recent study demonstrated that post-treatment of microalgae-derived hydrochar significantly reduced the phytotoxic organic compounds by DCM washing (Sun et al., 2023).

Distribution of inorganics
Analyses of the inorganic element content of the hydrochars and HTC liquors generated from microalgae and sewage sludge (Tables S1 and S2) showed that increasing the reaction temperature and the sewage sludge content in the feedstock increased the concentrations of major inorganic elements (i.e., Al, Fe, Ca, Mg and P) in the hydrochars.This behavior was observed for all hydrochars and can be attributed to the enrichment of inorganics in the hydrochar due to the release of intracellular water and solubilization of organics in the liquid phase during HTC (Shi et al., 2019).Interestingly, however, the Na and K contents of hydrochar formed from pure microalgae at 250 • C (0.6 mg/kg and 0.2 mg/kg, respectively) were lower than those in the starting feedstock (6.1 mg/kg and 1.9 mg/kg, respectively).Similar results are found in (Smith et al., 2016) showing a significant reduction of Na and K after HTC of sewage sludge and microalgae.Regarding the co-HTC, it appeared that Na and K partitioning was not significantly affected by the carbonization temperature because the concentrations of these elements in the microalgaesewage sludge hydrochars did not vary appreciably with either the carbonization temperature or the sewage sludge content of the feedstock (Tables S1-S2).
Understanding the distribution of heavy metals between hydrochar and HTC liquor is important because their presence may disqualify hydrochars from use as soil amendments and fertilizers.In this study, the heavy metals were mainly retained in the hydrochars, with only small amounts (<0.42 ug/mL) being present in the HTC liquors (Tables S1-S2).In general, the highest heavy metal concentrations were observed in hydrochars derived from feedstocks with high contents of sewage sludge that were processed at 250 • C.However, there were some exceptionsfor example, the concentrations of Cd, Pb, and Cu were higher in microalgae hydrochars formed at 250 • C (4.1, 37.2 and 706.9 mg/kg, respectively) than in sewage sludge hydrochars formed at the same temperature (1.6, 18.9 and 318.2 mg/kg, respectively).Overall, hydrochars obtained from feedstocks with higher proportions of microalgae contained more heavy metals than those generated from feedstocks with higher proportions of sewage sludge (Table S1).The reasons why the heavy metals are greatly enriched in the hydrochars could be that: (1) the heavy metals in this study have high boiling point and low water solubility, therefore minimize the transformation into liquid and/or gaseous products, and (2) the decomposition of the organic matter contributing to the yield loss of hydrochar results into cumulative effect of heavy metals, as suggested by (Zhang et al., 2021) in the co-HTC of sewage sludge and banana stalk, where similar results were reported.

Conclusions
In this study on co-HTC of microalgae and sewage sludge, the rates of degradation and carbonization of the solid feedstock components at high carbonization temperatures were reduced by adding sewage sludge.However, at carbonization temperatures higher than 180 • C the rate of secondary char formation increased with the proportion of sewage sludge in the feedstock because highly soluble compounds were deposited onto the primary char structure, and possibly because heterocyclic structures were integrated into the solid carbon matrix.Nonetheless, despite testing a wide range of microalgae-sewage sludge ratios (25-75 %), the observed effects of adding sewage sludge to microalgae were exclusively additive.The hydrochars contained carboxylic acids originating from microalgae, cholesterols from sewage sludge, and heavy metals from both feedstocks, all of which could limit the potential applications of the hydrochars.In particular, the release of environmentally harmful or toxic organic acids, aldehydes, ketones, and benzene derivatives from microalgae-derived hydrochars should be measured under realistic and relevant environmental conditions before using such hydrochars as soil amendments or fertilizers.Hydrochars derived from sewage sludge had lower concentrations of potentially toxic organic compounds and relatively high contents of phosphorus, potentially making them more suitable for environmental applications if all of the relevant environmental safety guidelines and related requirements can be satisfied.
Given the high ash content of sewage sludge and the high retention of C and FC in microalgae, microalgae-derived hydrochars may be more suitable for combustion applications than hydrochars formed from sewage sludge.However, the low production of microalgae would limit the feasibility of using such hydrochars as the sole fuel for a thermal plant.This problem could be overcome through co-HTC of microalgae with sewage sludge, which might also have beneficial effects on the fuel properties of the hydrochar.Therefore, further studies are needed to identify optimal strategies for post-treatment and/or recovery of target compounds.

Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Stina Jansson reports financial support was provided by Vakin.

Fig. 1 .
Fig. 1.Solid yields and moisture contents of hydrochars formed by co-HTC of microalgae and sewage sludge at 180, 215, and 250 • C. Note that the hydrochar yield (%, db) is expressed as the sum of solids recovered in the form of ash, primary char (non-extractable), and secondary char (extractable).

Fig. 2 .
Fig. 2. Partitioning of the HTC products between the solid and liquid phases based on dry ash-free (daf) percentage yields of the organic fractions of microalgae, sewage sludge, and microalgae-sewage sludge hydrochars after co-HTC at 180, 215 and 250 • .

Fig. 3 .Fig. 4 .
Fig. 3. Van Krevelen diagram showing the composition of feedstocks and hydrochars formed by HTC of microalgae (MA) and sewage sludge (SS) individually and in combinations with different mixing ratios.Background.adapted from Levine et al., 2013

Table 2
Proximate and elemental analysis results for the sewage sludge (SS) and microalgae (MA) feedstocks and hydrochars generated by HTC of microalgae and sewage sludge individually and combined in different mixing ratios.