Aqueous Phase Recycling in Hydrothermal Liquefaction: Mechanisms, Feedstock Interactions, and Sustainable Biorefinery Integration

Abstract

The aqueous phase (AP) produced during hydrothermal liquefaction (HTL) contains high organic loads and a chemically complex mixture of dissolved intermediates, posing significant environmental management challenges. Aqueous phase recycling (APR) has emerged as a strategy to enhance bio-crude yield, improve energy recovery, and reduce freshwater consumption by reintroducing reactive water-soluble species into subsequent cycles. However, repeated recycling can lead to the accumulation of N-containing compounds and phenolics, potentially diminishing bio-crude quality and heating value through secondary polymerization and condensation reactions. Simultaneously, the carbon and nutrient-rich character of AP presents opportunities for valorization via anaerobic digestion, microalgae cultivation, and supercritical water gasification. Despite growing interest, APR-HTL research remains feedstock-specific, and a systematic understanding of AP compositional evolution across multiple recycling cycles is limited. This review synthesizes recent progress, highlighting mechanistic linkages between AP composition, bio-crude performance, and integrated biorefinery strategies.

1. Introduction

Renewable energy sources, especially biomass, have gained increasing attention in recent decades as sustainable alternatives to conventional fuels, which aggravate environmental degradation and contribute to global warming [1,2]. Among thermochemical conversion methods, hydrothermal liquefaction (HTL) stands out as a promising approach for transforming both wet and dry biomass into energy-dense bio-crude at moderate temperatures (270–400 °C) and high pressures (10–35 MPa) [3,4]. During HTL, studies report that approximately 10–35% of biomass carbon is transferred to the aqueous phase (AP) [5,6]. This resultant AP requires additional energy- and cost-intensive treatment before safe disposal [7,8]. A techno-economic analysis by Zhu et al. [9] indicates that over 90% of waste disposal costs are attributed to AP treatment, positioning AP as the primary cost driver and underscoring the importance of minimizing organic losses to reduce costs effectively [10].

In the past decade, researchers have explored reusing AP by recirculating it within the HTL system as a water substitute, noting beneficial effects on bio-crude yield and energy recovery. This approach is known as AP recirculation (APR) [11,12,13,14,15,16]. The first documented APR-HTL experiment was carried out in 2014 at the Pacific Northwest National Laboratory (PNNL) using a continuous bench-scale reactor [17]. This study reported improvements in higher heating value (HHV) and carbon yield, along with reduced freshwater consumption [17]. Furthermore, the expense of heating the batch slurry can be diminished by employing a hot stream of recycled AP [18,19]. Over the years, APR has been tested with a wide range of feedstocks, including lignocellulosic biomass (e.g., barley straw, rice straw, wheat straw, corn stalk, desert shrubs, and aspen wood) [11,14,16,20,21,22], It has also been applied to protein-rich feedstocks (e.g., Chlorella vulgaris, Spirulina, and DDGS) [13,15,23,24,25,26,27], and even to petroleum-derived plastics such as polyethylene terephthalate (PET) [28]. While most APR studies have been conducted at the laboratory scale, exceptions include the work of Pedersen et al. [16], Biller et al. [13], and Klemmer et al. [29], who carried out their experiments on continuous-HTL setups. Irrespective of process mode, all investigations indicated increased bio-crude yields and energy recovery with APR, although no consistent findings were observed in the variation of HHV.

Though APR can maximize bio-crude yield, it may also lead to the accumulation of inhibitory compounds (e.g., N-containing species and phenolics), which could reduce bio-crude yield and HHV through polymerization reactions [30]. Depending on AP composition, these organics can be valorized through alternative routes via anaerobic digestion (biogas production) [30,31,32,33,34,35], biomass cultivation (particularly microalgae) [36,37,38,39,40,41,42], and supercritical water gasification (SCWG) [43,44,45,46]. Previous review papers, such as Watson et al. [8], reviewed APR in HTL using a relatively narrow range of feedstocks, since many potential feedstocks had not yet been evaluated for APR at the time. Leng et al. [47] published a study on APR in hydrothermal carbonization and HTL, but predominantly inclined towards carbonization. SundarRajan et al. [5] summarized the potential applications of AP HTL and briefly discussed its composition.

In the existing reviews, the APR-HTL strategy is generally regarded as a secondary operational route rather than as an integral or preferred process configuration. Moreover, prior reviews have concentrated on a limited spectrum of feedstocks and have predominantly emphasized application-oriented pathways, without adequately addressing the fundamental reaction mechanisms governing HTL system behavior. Consequently, the evolution of AP composition during multicycle recycling, especially the formation and transformation of reactive intermediates and recalcitrant compounds, and its effects on the stabilization or decline of bio-crude yield have rarely been interpreted mechanistically.

Addressing these gaps, this review aims to provide a comprehensive and critical synthesis of APR-HTL, positioning APR as a key determinant of process sustainability rather than a peripheral practice. Specifically, the review aims to (i) examine the compositional evolution of the AP across diverse lignocellulosic and protein-rich feedstocks during repeated recycling cycles; (ii) elucidate the mechanistic roles of major aqueous constituents, identifying beneficial and recalcitrant compounds in governing bio-crude yield, quality, and energy recovery; (iii) evaluate the integration of APR with downstream valorization pathways. This review integrates experimental and system-level insights to clarify the role of APR-enabled HTL within sustainable biorefinery frameworks, supporting resource-efficient, circular, and environmentally sustainable biomass conversion.

2. Characteristics of AP-HTL

During HTL, a variety of organic compounds are produced through different reactions such as hydrolysis, decarboxylation, dehydration, deamination, polymerization, and the Maillard reaction. The term “hydrothermal” refers to the conversion of biomass in the presence of a water medium. Initially, the biomass is converted into water organics, and those compounds can later be transformed into the bio-crude phase. Among biomass components, bio-crude formation is predominantly governed by lipids and proteins, while carbohydrates contribute to a lesser extent. In contrast, lignin markedly promotes char formation. Additionally, ash and high-molecular-weight cyclic compounds, particularly those derived from biomass with high inorganic content, such as sewage sludge, tend to increase solid residue formation [48]. The distribution of chemical components of different feedstocks used in APR-HTL is illustrated in Figure 1. Studies indicate that carbohydrates and proteins are readily liquefied, converting into organic acids, N-heterocyclic compounds, and amines, which are predominant among other organic compounds [49].

An extensive overview of the characteristics of the AP generated from diverse feedstocks in HTL-APR studies is presented in Table S1 (Supplementary Materials). Research reports that lower temperatures and shorter reaction times (RTs) typically lead to higher total organic carbon (TOC) due to partial depolymerization, thereby shifting more organic material into the AP, which consequently reduces bio-crude yield [1,6]. However, oxygenates like phenols, furfurals, and N-containing compounds, including fatty acid amides, can be partitioned into the bio-crude phase at elevated temperatures [50]. Reddy et al. [51] concluded that water-soluble compounds can be repolymerized at high temperatures and become part of the bio-crude. Apart from low-molecular-weight acids, a wide range of phenolics were also detected in AP-HTL, particularly from lignocellulosic and food waste biomasses [11], yielding a higher proportion of organics in AP. Conversely, algal biomass contained lower phenols. Madsen et al. characterized AP derived from different model compounds and reported a TOC trend (highest to lowest) of protein > carbohydrate > lipid [52].

The nitrogen in the AP-HTL originates from cleavage of peptides in protein-rich substances like microalgae; on the other hand, carbohydrates and lignin are mostly dominated by compounds derived from C, H, and O [19]. Similarly, various inorganic species, including phosphate, ammonia, sodium, potassium, calcium, aluminum, and iron, have been reported, particularly in high ash-containing biomass such as sewage sludge [12]. The existence of these compounds depends on the biochemical composition of the feedstock, the conditions under which it is grown, and the process parameters [53]. A study reported that AP-HTL of algae cultivated under saline conditions contained higher concentrations of Mg, Ca, Na, and K compared to that derived from algae grown in freshwater [54]. Zinc is commonly detected in manure-derived AP-HTL streams because it is routinely added to animal feed as a growth supplement and is subsequently excreted in dung and urine. Cang et al. [55] examined the feed and dung of thirty-one farms and noticed higher concentrations of Zn, Cu, Cd, Pb, and Cr in manures. Moreover, Conti et al. [29] have indicated that the AP-HTL of lignocellulosic biomass cultivated with untreated wastewater and domestic sewage includes 3 mg/L of Zn, attributable to the growth substrate. The key point is that the measured biomass composition rarely sums to exactly 100% because certain extractives (e.g., soluble sugars, organic acids, and other minor organics) were excluded from the analysis. As these constituents are solubilized during use of the Van Soest Detergent Fiber Method, these constituents cannot be captured, and they are designated as unquantified fractions, as illustrated in Figure 1.

Figure 1. Chemical composition of feedstocks used for APR in HTL. Data extracted from Zhu et al. [11], Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Seehar et al. [20], Yulin Hu et al. [56], Damizia et al. [57], Caprio et al. [58], Leng et al. [59], Yulin Hu et al. [23], Taghipour et al. [60], R. Tercero et al. [15], Haitao Chen et al. [24], Parsa et al. [61], Xinfei Chen et al. [62], Deniele et al. [25], Qian et al. [26], Hong et al. [63], Biller et al. [13], Shah et al. [12], Rajan et al. [64], Song et al. [65], Zhang et al. [66], and Kohansal et al. [27].

Figure 1. Chemical composition of feedstocks used for APR in HTL. Data extracted from Zhu et al. [11], Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Seehar et al. [20], Yulin Hu et al. [56], Damizia et al. [57], Caprio et al. [58], Leng et al. [59], Yulin Hu et al. [23], Taghipour et al. [60], R. Tercero et al. [15], Haitao Chen et al. [24], Parsa et al. [61], Xinfei Chen et al. [62], Deniele et al. [25], Qian et al. [26], Hong et al. [63], Biller et al. [13], Shah et al. [12], Rajan et al. [64], Song et al. [65], Zhang et al. [66], and Kohansal et al. [27].

3. APR in Hydrothermal Liquefaction

3.1. Effect of APR on HTL of Lignocellulose Feedstocks

All lignocellulose feedstocks, except for rice straw, exhibited an increase in yield through APR (Figure 2). Most of the improvement in bio-crude was observed during the initial cycles, after which the enhancement declined [12,13,15,24]. As APR is predominantly governed by reaction kinetics rather than thermodynamic limitations. Initially, recirculation enriches the system with reactive intermediates that facilitate secondary reactions, thereby enhancing bio-crude yield. However, prolonged APR promotes the transformation of these intermediates into recalcitrant compounds (Table 1), diminishing overall reactivity and slowing bio-crude formation. Ultimately, a yield plateau is established when bio-crude production equilibrates with carbon retention in unreactive forms, reflecting a kinetically constrained maximum conversion (Figure 3). However, Jensen et al. [18] recommended four to five rounds of APR to achieve a steady state. Here, C0 is the bio-crude yield of the baseline experiment, while Cn-max is the maximum bio-crude yield obtained at the nth cycle (Cn). The primary catalyst (K₂CO₃) was employed, which is effective in promoting the hydrolysis of cellulose and hemicellulose while simultaneously inhibiting char formation [67]. Similar results were previously documented by Chen et al. [24] utilizing three biomass types, α-cellulose, lignin, and Spirulina, observing a contrary trend with lignin, leading to a 6% reduction in bio-crude yield over APR.

Figure 2. Effect of APR on lignocellulose feedstocks. Data extracted from Zhu et al. [11], Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Seehar et al. [20], Damizia et al. [57], Caprio et al. [58], Leng et al. [59], R. Tercero et al. [15], Haitao Chen et al. [24], Xinfei Chen et al. [62], Deniele et al. [25], Qian et al. [26], Rajan et al. [64], Shah et al. [12], Song et al. [65], Parsa et al. [61], Zhang et al. [66], Yulin Hu et al. [23], Taghipour et al. [60], Hong et al. [63], Biller et al. [13], Klemmer et al. [29], and Kohansal et al. [27].

Figure 2. Effect of APR on lignocellulose feedstocks. Data extracted from Zhu et al. [11], Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Seehar et al. [20], Damizia et al. [57], Caprio et al. [58], Leng et al. [59], R. Tercero et al. [15], Haitao Chen et al. [24], Xinfei Chen et al. [62], Deniele et al. [25], Qian et al. [26], Rajan et al. [64], Shah et al. [12], Song et al. [65], Parsa et al. [61], Zhang et al. [66], Yulin Hu et al. [23], Taghipour et al. [60], Hong et al. [63], Biller et al. [13], Klemmer et al. [29], and Kohansal et al. [27].

There may be additional potential factors contributing to the lower yield from rice straw: (1) Reaction without alkali catalyst, as pH dropped from 3.9 to 1.3 from baseline to third cycle, which might be due to the polymerization of 5-hydroxymethyl furfural under acidic conditions (forming biochar) [68]. (2) Notably, rice straw and corn stover contain comparable lignin contents (~25%) and were liquefied under non-catalytic conditions, yet they exhibited contrasting product distributions. This divergence indicates that different structural features of lignin could play a key role in governing HTL reaction pathways. Rice straw lignin is often guaiacyl-rich, resulting in a higher fraction of condensation-prone aromatic sites and a greater prevalence of C–C linkages (and fewer readily cleavable β-aryl ether linkages) compared to hardwood lignin. Consequently, HTL tends to favor repolymerization and char formation rather than the production of stable bio-crude [69]. (3) Moreover, rice straw is typically enriched in ash-forming inorganics (e.g., K/Cl/Si species), which can further promote phenolic condensation reactions [70]. This can be supported by the higher presence of K⁺ and CL⁻ in AP of rice straw as detected by Harishankar et al. [21], with both increasing from a few hundred mg/L to 2800 mg/L and 25,000 mg/L, respectively, after three rounds of APR. Aspen wood showed a higher yield with APR at elevated temperatures, whereas the positive impact of reduced red mud was observed [16]. Earlier, Motavaf et al. [71] investigated CaO, Al₂O₃, and SiO₂ catalysts for food waste HTL and reported that SiO₂ yielded higher bio-crude yield and HHV than the non-catalytic run.

Figure 3. Conceptual relationship shows how APR influences reactive fragment formation, recalcitrant accumulation, and the resulting plateau in net bio-crude yield.

Figure 3. Conceptual relationship shows how APR influences reactive fragment formation, recalcitrant accumulation, and the resulting plateau in net bio-crude yield.

Since variation in bio-crude yield during APR is closely linked to the organic composition of the AP, it is therefore crucial to study AP compositional changes throughout the APR. Regardless of biomass composition, TOC showed a sharp rise during APR and then stabilized after a few cycles, following the same trend as observed in bio-crude yield (Figure 4). This implies that TOC is a precursor and primary contributor to the increase in bio-crude productivity. TN exhibited a similar increasing pattern due to the accumulation of ammonia/ammonium-nitrogen (NH₃/NH₄⁺) [12,27,54,60]. It was observed that higher temperatures (400 °C) shifted a higher amount of carbon to AP, which can be transferred to the system via APR, resulting in a substantial increase of (+12%) in bio-crude from aspen wood without disturbing/reducing HHV at continuous pilot-scale HTL [16]. Jensen et al. further promoted the concept of supercritical HTL and termed this phenomenon “Hydrofaction” [18].

Most studies used GC-MS at 250–350 °C to qualitatively identify the compounds. AP of lignocellulosic biomass yielded mainly acids regardless of reaction conditions. The dominant compounds detected in AP during APR are listed in Table S1. Acetic acid was the predominant compound in the AP, formed primarily through the decarboxylation of polysaccharides and the deamination of amino acids. During APR, acids decompose cellulose and hemicellulose, yielding intermediates that dehydrate to create bio-crude. It has been experimentally proven that model components (glucose and fructose) were enhanced under an acidic atmosphere [72]. The compound composition of AP before and after the recirculation from corn stalk (4.74% lignin) was evaluated by Yin et al. [22], who found that the increasing trend of acids, N-containing compounds, esters, and alcohols, while phenols and ketones decreased drastically after three rounds of APR (Table S1), indicating that bio-crude was formed by the conversion of phenols and ketones from AP to bio-crude. Qualitative analysis is adequate for chemical identification; however, quantitative analysis substantiates the actual concentrations of the detected compounds. Biller et al. studied DDGS and observed an increasing tendency with APR in acids (~1700–7250 µg/mL) and N-heterocyclic compounds (Pyrazine–methyl ~1850–3250 µg/mL). Chen et al. [24] also found larger peaks of 2-Pyrrolidinone (0.05–4.04%) and Benzeneethanamine (4.27–8.53%) in Spirulina platensis, indicating the enrichment of N-containing compounds. Rice straw produced a higher biochar yield (64%) under APR conditions. The major compounds identified in the rice straw-derived bio-crude were dimethyl phenols (~11%), p-cresol (~10%), and guaiacol (~11%) [21]. These compounds form through dealkylation of lignin-derived guaiacol to catechol intermediates, which further hydrolyze into phenolics retained in the biochar phase [73,74].

Figure 4. Effect of APR on AP characteristics. Data extracted from Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Leng et al. [59], Deniele et al. [25], Qian et al. [26], Rajan et al. [64], Shah et al. [12], Song et al. [65], Hong et al. [63], Biller et al. [13], Klemmer et al. [29], and Kohansal et al. [27].

Figure 4. Effect of APR on AP characteristics. Data extracted from Pedersen et al. [16], Harishankar et al. [21], Li et al. [14], Yin et al. [22], Leng et al. [59], Deniele et al. [25], Qian et al. [26], Rajan et al. [64], Shah et al. [12], Song et al. [65], Hong et al. [63], Biller et al. [13], Klemmer et al. [29], and Kohansal et al. [27].

Collectively, APR in lignocellulosic feedstocks affects both bio-crude yield and char formation; however, excessive char accumulation may offset the benefits of APR. GC-MS analysis indicated that the organic acids facilitate breakdown by stimulating hydrolysis reactions, hence enhancing the conversion of feedstock into the bio-crude phase [14,75,76]. Yet a significant portion of the cellulose and lignin may persist in the biochar phase even after HTL. Thus, the synergistic effect of APR is evident in lignocellulose feedstocks in terms of bio-crude and biochar yield. The schematic diagram of reaction mechanisms of APR with lignocellulose feedstock is illustrated in Figure 5a, summarized from [11,47,68,77]. This can be confirmed by Müller and Vogel [78], who liquefied glucose with high feed concentration, yielding higher bio-crude and biochar simultaneously.

To unlock the mystery of acetic acid, researchers [12,14,15,23,24] tried to test the effect of acetic acid on bio-crude properties by varying the concentration of acetic acid and comparing the results with baseline (water) and APR experiments (Table 2). The results showed that acetic acid increased the bio-crude yield but at the expense of HHV, as it promotes cracking and decarboxylation, leading to higher oxygenated products and reduced carbon retention compared to a water medium.

Table 1. Recalcitrant compounds that could suppress the bio-crude yield via APR.

Compounds Type/ Examples Reported or Implied in HTL Chemistry	Suppressing Bio-Crude Yield (HTL Mechanism)	Carbon Ends Up in the Product	Ref.
Phenolic monomers/oligomers (e.g., phenol/cresols/guaiacol-type derivatives), lignin-derived aromatic fragments	Secondary recombination/repolymerization of unstable fragments forms solid residue/biochar-like material, reducing bio-crude recovery	Solid residue/ biochar	[11,21,57,79,80]
Maillard-derived Heterocyclic/aromatic N products (Maillard pathway products; stable N-containing structures)	Maillard chemistry forms stable N-heteroaromatic structures (thermally persistent), which limit clean bio-crude formation and can shift products toward heavier/recalcitrant fractions	Heavier bio-crude fraction and/or solids; also, AP organics	[12,15,81,82,83]
Water-soluble low-molecular-weight oxygenates (small acids/oxygenated organics captured in the AP analyses; exact composition varies with feed)	These compounds are highly water-soluble under HTL conditions and represent direct carbon loss to the AP (recalcitrant-to-bio-crude partitioning)	AP	[11,13,24,57]
Nitrogenous organic compounds (NOCs): Pyridine-like species, indole-like species, other NOCs	Many NOCs are thermally stable and/or remain water-soluble, so nitrogen and carbon remain in non-bio-crude phases; these are recalcitrant in the sense that they persist rather than converting cleanly into bio-crude	AP (major), sometimes heavy bio-crude	[54,84]
High-ash/alkali or alkaline carbonates/mineral matter/metal chlorides for protein-rich biomass, as well as lignin	High ash retards bio-crude formation and worsens bio-crude quality, i.e., it shifts reaction pathways away from bio-crude formation	For protein-rich feedstock: Less bio-crude; more non-bio-crude products For lignin: More solid residue; less bio-crude	[18,22,27,85,86]
Protein-AP + (α-cellulose/lignin) Antagonistic mechanism	Antagonistic mechanism: APR causes inhibition in a mixed-model-component system	APR inhibited bio-crude generation, indicating antagonistic chemistry	[24]

Table 1. Recalcitrant compounds that could suppress the bio-crude yield via APR.

Compounds Type/ Examples Reported or Implied in HTL Chemistry	Suppressing Bio-Crude Yield (HTL Mechanism)	Carbon Ends Up in the Product	Ref.
Phenolic monomers/oligomers (e.g., phenol/cresols/guaiacol-type derivatives), lignin-derived aromatic fragments	Secondary recombination/repolymerization of unstable fragments forms solid residue/biochar-like material, reducing bio-crude recovery	Solid residue/ biochar	[11,21,57,79,80]
Maillard-derived Heterocyclic/aromatic N products (Maillard pathway products; stable N-containing structures)	Maillard chemistry forms stable N-heteroaromatic structures (thermally persistent), which limit clean bio-crude formation and can shift products toward heavier/recalcitrant fractions	Heavier bio-crude fraction and/or solids; also, AP organics	[12,15,81,82,83]
Water-soluble low-molecular-weight oxygenates (small acids/oxygenated organics captured in the AP analyses; exact composition varies with feed)	These compounds are highly water-soluble under HTL conditions and represent direct carbon loss to the AP (recalcitrant-to-bio-crude partitioning)	AP	[11,13,24,57]
Nitrogenous organic compounds (NOCs): Pyridine-like species, indole-like species, other NOCs	Many NOCs are thermally stable and/or remain water-soluble, so nitrogen and carbon remain in non-bio-crude phases; these are recalcitrant in the sense that they persist rather than converting cleanly into bio-crude	AP (major), sometimes heavy bio-crude	[54,84]
High-ash/alkali or alkaline carbonates/mineral matter/metal chlorides for protein-rich biomass, as well as lignin	High ash retards bio-crude formation and worsens bio-crude quality, i.e., it shifts reaction pathways away from bio-crude formation	For protein-rich feedstock: Less bio-crude; more non-bio-crude products For lignin: More solid residue; less bio-crude	[18,22,27,85,86]
Protein-AP + (α-cellulose/lignin) Antagonistic mechanism	Antagonistic mechanism: APR causes inhibition in a mixed-model-component system	APR inhibited bio-crude generation, indicating antagonistic chemistry	[24]

3.2. Effect of APR on HTL of Protein-Rich Feedstocks

Protein-rich feedstocks exhibited a greater increase in bio-crude yield under APR compared to lignocellulosic feedstocks. This enhancement is attributed to their higher protein and lipid contents. HTL of individual model components further demonstrated that bio-crude production follows the hierarchy: lipids > proteins > carbohydrates [87,88,89]. In most cases, protein-rich feedstocks have been processed without adding a catalyst, as alkaline catalysts are detrimental to yield, while acids reduce the HHV [89]. The enhanced bio-crude yield from protein-rich biomass is attributed to two key reaction pathways: (1) the Maillard reaction, which occurs between amino acids and polysaccharides; and (2) acylation–condensation reactions, involving amino acids and amines reacting with fatty acids and esters [47]. Recently, Shah et al. explored the detailed reaction pathways of APR with lignocellulose and protein-rich feedstocks [53]. The intermediates derived from Maillard and acylation reactions, especially N-heterocyclic compounds and amides, favor distribution into the bio-crude phase over the AP [90], resulting in higher enrichment of nitrogen in the bio-crude [82]. The increasing trend in the peak area of N-heterocyclic compounds in bio-crude was detected via APR. For instance, Pyridine (0.08–0.21%), methyl Pyrazine (0.56–1.02%), and 1-acetyl-Pyrrolidine (~1.40–2.40%) were detected in Chlorella vulgaris [23]. Shah et al. reported the presence of 4-Piperidinone, 2,2,6,6-tetramethyl (4–18%) in the bio-crude derived from sewage sludge [12]. Biller et al. quantitatively verified increasing concentrations of Pyrazine 2,5-dimethyl and 2-Pyrrolidone at ~1000–4000 µg/mL and ~1950–11,000 µg/mL, respectively [13]. Shah et al. [12] also observed higher lipophilic amides, specifically long-chain amides, including 9-Octadecenamide-Z (~9.70–17.0%), and Dodecanamide (~5.0–8.0%). Since protein-rich biomass contains a moderate quantity of carbohydrates that generate aldehydes (essential for hydro-char or biochar). Numerous studies revealed that adding N-containing compounds to cellulose biomass shifts organics toward bio-crude via condensation [24,47,91]. Chen et al. further confirmed this by performing APR of cellulose with AP derived from Spirulina platensis, reporting an enhanced bio-crude yield and lower solids due to the incorporation of N-containing compounds during APR [24]. While lignocellulosic APR shows a synergistic effect, protein-rich biomass APR exhibits an inverse relationship between bio-crude and biochar formation (Figure 2). The schematic diagram of reaction pathways is presented in Figure 5b, summarized from [24,47,91,92].

3.3. Effect of Catalyst on Protein-Rich Feedstock via APR

Several investigations have documented the APR of protein-rich biomass both with and without a catalyst [13,23,27,60,63]. Hu et al. [23] observed that Na₂CO₃ and HCOOH diminished bio-crude production in baseline tests; however, both catalysts enhanced bio-crude yield when used in APR by +35% and +14%, respectively, which is far higher than without catalytic APR (+9%). The enhancement in yield with Na₂CO₃ is associated with saturated molecules like Phytol (28%, Table S1), which can generate bio-crude products (isophytol, phytene, phytane) [93]. Formic acid improves yield through Maillard reactions by serving as a hydrogen donor and converting N-heterocyclic compounds into bio-crude [94]. Nitrogen intermediates under acidic circumstances enhance Maillard and acylation reactions, increasing bio-crude yield and energy recovery [12]. Klemmer and Biller [13,29] investigated DDGS with K₂CO₃ in a continuous HTL system, attaining a yield of 53–55%, the highest among all proteinaceous biomasses. Kohansal et al. [27] evaluated alkali catalysts using Biopulp at 350 °C, ranking them as follows: K₂CO₃ > Na₂CO₃ > KOH > NaOH. Taghipour et al. [60] investigated Spirulina utilizing Na₂CO₃ and Pt/Al₂O₃, including APR. Both catalysts enhanced bio-crude yield, with Na₂CO₃ producing sodium formate [95] and Pt/Al₂O₃ facilitating decarboxylation [96,97]. APR significantly enhanced yields, particularly in non-catalytic processes (+7.9%). Hong et al. [63] observed minimal APR increase with Na₂CO₃ on Penicillin residue attributable to the exceedingly low carbohydrate content (1.66%). Among all biomasses, DDGS-K₂CO₃ attained a 13% yield enhancement, approximately 1.3 times the baseline yield [13]. On the other hand, non-catalytic APR augmented Ch. vulgaris bio-crude by 29%, almost double the reference yield (13%) [15]. Scenedesmus abundans [64] and sewage sludge [12] showed increases of +5% and +13% in yield, respectively, demonstrating the effectiveness of APR for proteinaceous biomass.

Table 2. Effect of acetic acid on bio-crude yield and HHV.

Feedstock	Concentration of Acetic Acid (ACA)	Yield with ACA (%)	Yield with Water (%)	Yield with APR (%)	HHV–ACA (MJ/kg)	HHV–Water (MJ/kg)	HH–APR (MJ/kg)	Ref.
Desert shrub	pH: 2.7	27.00	25	30	26.5	27.1	25.4	[14]
Sewage sludge	2.5% of total slurry	26.85	25	38	34.48	35.22	33.82	[12]
Ch. vulgaris	6 g/L	15.51	14.3	42.3	33.6	34.8	32.3	[15]
Ch. vulgaris	3% of total slurry	31.45	29.39	38	NA	33.87	31.37	[23]
Ch. vulgaris	5% of total slurry	41.00	29.39	38	32.69	33.87	31.37	[23]
Sp. plantensis	5% of total slurry	35.51	30.07	40	NA	34.67	30.15	[24]

NA: Not available.

Table 2. Effect of acetic acid on bio-crude yield and HHV.

Feedstock	Concentration of Acetic Acid (ACA)	Yield with ACA (%)	Yield with Water (%)	Yield with APR (%)	HHV–ACA (MJ/kg)	HHV–Water (MJ/kg)	HH–APR (MJ/kg)	Ref.
Desert shrub	pH: 2.7	27.00	25	30	26.5	27.1	25.4	[14]
Sewage sludge	2.5% of total slurry	26.85	25	38	34.48	35.22	33.82	[12]
Ch. vulgaris	6 g/L	15.51	14.3	42.3	33.6	34.8	32.3	[15]
Ch. vulgaris	3% of total slurry	31.45	29.39	38	NA	33.87	31.37	[23]
Ch. vulgaris	5% of total slurry	41.00	29.39	38	32.69	33.87	31.37	[23]
Sp. plantensis	5% of total slurry	35.51	30.07	40	NA	34.67	30.15	[24]

NA: Not available.

3.4. Effect of APR on Bio-Crude Quality

It has been demonstrated that APR affects not only the productivity of bio-crude but also its quality. The majority of studies [12,14,15,23,26,59,60] showed a declining trend in carbon content and HHV, correlating with higher levels of nitrogenated and oxygenated compounds, as depicted in Figure 6. The reduced carbon may also result from increased gas formation via decarboxylation, converting a significant portion of carbon into gases and light hydrocarbons [15].

On the other hand, Hu et al. found that changing feedstock solid content resulted in a significantly lower O content (15.5%) than a water mixture (22.1%) [56]. The same pattern was detected in blackcurrant, DDGS, sewage sludge, and penicillin residue, as listed in

Table 3. Change in nitrogen and oxygen contents in bio-crude via APR.

Feedstock	Temp. (°C)/RT (min)/Catalyst/Cycles	N in C0 by Value (%)	Increase in N by (%)	O in C0 by Value (%)	Decrease in O by (%)	Ref.
Blackcurrant pomace	310/10/no cat./5	3.00	13.33	15.60	−12.82	[25]
DDGS	350/20/no cat./9	6.40	29.69	12.90	−31.78	[13]
Sewage sludge	350/15/no cat./8	4.67	17.77	11.40	−13.86	[12]
Biopulp	350/15/K2CO3/4	3.82	26.70	8.39	−26.10	[98]
Penicillin residue	280/180/no cat./3	6.39	24.88	10.01	−48.95	[63]

. The upward trend of nitrogen with APR is unavoidable due to consecutive APR, particularly with increased nitrogen enrichment through the intake of proteinaceous biomass [12,13,63]. Under non-catalyst conditions, in Ch. vulgaris, N increased by approximately two times, from 4.7 to 8.9% (+4.2%), as observed by Ramos–Tercero et al. [15]. For Sp. platensis, N increased from 9.8 to 15.4% (+5.6%) [24]. Other studies reported the same findings for penicillin residue [63], sewage sludge [12], and Biopulp [27]. To maintain APR sustainability, particularly for protein-rich feedstocks, it is essential to minimize nitrogen transfer into bio-crude via AP pre-treatment. Kohansal et al. [98] reduced AP nitrogen by evaporating light organics at ~70–80 °C before APR of the remaining AP concentrate, but this also resulted in simultaneous losses of TOC along with TN. Hence, nitrogen control is challenging and may be approached via feedstock composition management, biomass pre-treatment (N/ash removal) before HTL, use of suitable reaction solvents, and development or tailored modification of catalysts matched to the biomass and operating conditions; otherwise, downstream hydrotreatment is required to remove heteroatoms (N/O), adding cost and introducing risks such as catalyst poisoning and deactivation.

Figure 6. Effect of APR on carbon, nitrogen, and HHV of bio-crude. Data extracted from Barly straw [11], Rice straw [21], Wheat straw [20], Desert shrub [14], Corn stalk [22], Soybean straw (SS) [59], Chlorella powder (CL) [59], Chlorella vulgaris—Hu et al. [23], Spirulina [60], Chlorella vulgaris—R. Tercero et al. [15], Spirulina platensis [24], Chlorella pyrenoidosa [62], Blackcurrant pomace [25], Phycocyanin [26], Penicillin residue [63], DDGS—Biller et al. [13], Sewage sludge—Shah et al. [12], Scenedesmus abundans [64], Sewage sludge—Song et al. [65], Gracilaria gracilis [29], DDGS—Klemmer et al. [29], Biopulp [27], and SS + CL (ratio:1:1) [59].

Figure 6. Effect of APR on carbon, nitrogen, and HHV of bio-crude. Data extracted from Barly straw [11], Rice straw [21], Wheat straw [20], Desert shrub [14], Corn stalk [22], Soybean straw (SS) [59], Chlorella powder (CL) [59], Chlorella vulgaris—Hu et al. [23], Spirulina [60], Chlorella vulgaris—R. Tercero et al. [15], Spirulina platensis [24], Chlorella pyrenoidosa [62], Blackcurrant pomace [25], Phycocyanin [26], Penicillin residue [63], DDGS—Biller et al. [13], Sewage sludge—Shah et al. [12], Scenedesmus abundans [64], Sewage sludge—Song et al. [65], Gracilaria gracilis [29], DDGS—Klemmer et al. [29], Biopulp [27], and SS + CL (ratio:1:1) [59].

Despite the trade-offs in bio-crude quality, it has been demonstrated that APR improves the energy recovery (ER) of the HTL process. Across all biomasses, ER generally increased by 3–68%, except rice straw, which showed a decrease in ER: minus (−43%). The highest ER values, 82% and 98%, were observed for wheat straw and DDGS, respectively. In addition, Chlorella vulgaris–Na₂CO₃ exhibited a significant increase (change in ER) by 62%. The schematic diagram of the effect of APR in HTL is illustrated in Figure 7.

Catalyst can potentially affect the quality of bio-crude. Overall, catalyst K₂CO₃ demonstrated no substantial enhancements in HHV, which is instead decreased by reducing carbon content for both feedstock types except corn stover (+14.35 MJ/kg). Biller et al. [13] observed a lower concentration of DDGS-K₂CO₃ as compared to the non-catalytic run. However, oxygen concentration increased from 14% to 27%, resulting in a lower HHV. The alkali catalyst diminished the nitrogen content in bio-crude by APR, in contrast to non-catalytic bio-crude. K₂CO₃ enhances the conversion of N-organic compounds into ammonium via deamination, while suppressing Maillard reactions that would otherwise form N-heterocyclic compounds in the bio-crude [99]. Klemmer et al. [29] supported the outcomes of Biller et al., observing a reduction in HHV. Formic acid increased N content in bio-crude derived from Chlorella vulgaris [23], directing a greater proportion of N-heterocyclic compounds into the bio-crude phase instead of AP [23]. In biopulp [27], an alkali catalyst promoted the formation of more deoxygenated hydrocarbons, resulting in higher HHV. The Pt/Al₂O₃ demonstrated reduced nitrogen in both runs (baseline and APR) compared to APR without a catalyst [60]. This shows that denitrogenation of nitrogen compounds occurs efficiently via adsorption on heterogeneous catalyst sites.

4. Integrated Routes for Recycling of AP-HTL

4.1. AP-HTL Recycling in Anaerobic Digestion

Anaerobic digestion (AD) is a cost-effective technology. The anaerobic microorganisms are known to be more resistive towards organic compounds present in AP-HTL [100]. Zhu et al. [101] concluded through techno-economic analysis that AD requires lower cost and energy for AP-HTL compared to catalytic hydrothermal gasification. Figure 8 shows the overall methane range from 21 to 313 mL/g COD; however, the increase in methane yield is directly linked to the removal performance of COD. Posmanik et al. [39] found that increasing the concentration of NaCl (higher than 10 g/L) inhibits microbial activity in AD. In some scenarios, the degradation rate is also becoming slower due to the synergistic effect of various inhibitors. This can be confirmed by the study of Pham et al. [84], who performed HTL of Spirulina and used AP for AD, finding that nine major organic compounds were less cytotoxic than their synergistic effect (used in mixed form). The organic compounds exhibited the cytotoxicity in descending order as follows: 3-dimethylaminophenol > 2,2,6,6-tetramethyl-4-piperidone > 2,6-dimethyl-3-pyridinol > 2-picoline > pyridine > 1-methyl-2-pyrrolidinone > D-valerolactam > 2-pyrrolidinone > E-caprolactam. The synergistic effect of pyridine and phenol was reported in another study [78] as the pyridine degradation rate declined with increasing phenol concentrations, beyond 400 mg/L [102].

Kim et al. also reported the suppressive effect of phenol on pyridine degradation [103]. Earlier, Si et al. [34] presented the fate and distribution of organics, including volatile fatty acids, N-containing compounds, and cyclic compounds, during AD using AP-HTL, as illustrated in Figure 9. This shows that the volatile fatty acids can readily be digested via routes of methanogenesis and acetogenesis. Conversely, in the case of N-containing and cyclic compounds, acetogenesis is hindered by toxicity and high molecular mass of associated compounds [31]. The change in volatile fatty acids during pre- and post-AD treatment is provided in Table 4.

It was observed that granular-activated carbon (GAC) increased methane output by 13.4–300% and decreased the lag phase by 30.4–46.8%. Additives include struvite, polyurethane matrices (PM), and zeolite, which enhanced methane output up to 250%, 28.7%, and 5.1%, respectively, compared to the controlled media. Ozone pre-treatment enhanced methane output by 109.4%, while it marginally extended the lag period, indicating that macromolecules were transformed into low-molecular-weight chemicals that microorganisms cannot efficiently consume [34].

Figure 8. Effect of AP-HTL recycling on methane yield through AD. Data extracted from Zheng et al. [31], Shanmugam et al. [32], Zhou et al. [104], Si et al. [105], Fernandez et al. [33], Chen et al. [106], Si et al. [34], Chen et al. [107], and Aktas et al. [108].

Figure 8. Effect of AP-HTL recycling on methane yield through AD. Data extracted from Zheng et al. [31], Shanmugam et al. [32], Zhou et al. [104], Si et al. [105], Fernandez et al. [33], Chen et al. [106], Si et al. [34], Chen et al. [107], and Aktas et al. [108].

Table 4. Organic acids before and after AD treatment.

Concentration (mg/L)	Formic Acid ↓	Acetic Acid ↓	Propanoic Acid ↓	Butyric Acid ↓	Valeric Acid ↓	Lactic Acid ↓	Ref.
AP–HTL	NA	369	613	1346	ND	NA	[31]
AD effluent (Zeolite)	NA	55.8	496	748	1188	NA	[31]
AD effluent (GAC)	NA	68.8	423	418	4085	NA	[31]
AD effluent (PM)	NA	149	490	688	1261	NA	[31]
AP–HTL	830	2350	960	170	NA	121	[107]
AD effluent (Mesophilic)	ND	80	ND	ND	NA	ND	[107]
AD effluent (Thermophilic	430	240	ND	ND	NA	560	[107]

ND: Not detected. NA: Not available.

Table 4. Organic acids before and after AD treatment.

Concentration (mg/L)	Formic Acid ↓	Acetic Acid ↓	Propanoic Acid ↓	Butyric Acid ↓	Valeric Acid ↓	Lactic Acid ↓	Ref.
AP–HTL	NA	369	613	1346	ND	NA	[31]
AD effluent (Zeolite)	NA	55.8	496	748	1188	NA	[31]
AD effluent (GAC)	NA	68.8	423	418	4085	NA	[31]
AD effluent (PM)	NA	149	490	688	1261	NA	[31]
AP–HTL	830	2350	960	170	NA	121	[107]
AD effluent (Mesophilic)	ND	80	ND	ND	NA	ND	[107]
AD effluent (Thermophilic	430	240	ND	ND	NA	560	[107]

ND: Not detected. NA: Not available.

Nowadays, hythane (a mixture of hydrogen and methane) production has received considerable attention due to its characteristics as a cleaner fuel, which is produced through a two-stage dark fermentation process. For fermentation, monosaccharides from carbohydrates of AP-HTL are considered the main substrate for hydrogen production. Once hydrogen is produced, the resultant alcohols and volatile fatty acids could be used as feed for methane formation [8]. In contrast with the single-stage fermentation of AD, the two-stage dark fermentation technique enhanced the theoretical energy recovery by 10 to 12%. In the first stage of two-stage dark fermentation, along with hydrogen production, microbes break down inhibitors, which decrease the lag time for the second stage (process for methane formation). This has been validated by Liu et al. [82], who found that detoxification of furfural and 5-HMF is achieved along with hydrogen production well before the step of methane production.

Si et al. [109] performed a study on two-stage fermentation by converting AP-HTL of human feces into methane and hydrogen, at 254 mL/g COD and 29 mL/g COD, respectively. The lower magnitude of hydrogen was primarily due to the existence of phenols and furfurals (inhibitors). Almost one-third of hythane production was carried out in bench mode, which might be due to complexities in reactor integration, which requires further investigation. Azarmania et al. [35] carried out AD with co-digestion involving municipal sludge and dewatering concentrate for a prolonged period of 420 days; the digesters were fed with variable concentrations of COD in AP-HTL from 0.0% to 43.3%. Results revealed that co-digestors with AP-HTL having COD of 12, 23.1, and 43.3% were inhibited after 42, 29, and 15 days, respectively. The co-digestion with lower concentrations of COD, at 1.21% and 2.74%, displayed inhibition for 180 days. The best conditions identified were HTL with 4.89% COD, which provided the highest methane (about 300 mL CH₄/g VS feed) and maximum daily biogas production (approximately 330 mL/day) before inhibition commenced at around day 95. The COD removal efficiency was approximately 62% before inhibition and around 57% following recovery, suggesting partial adaptation of thermophilic cultures. This suggests that thermophilic cultures can adapt to inhibitors in AP-HTL to a limited degree. Higher concentrations of N-heterocyclic compounds (2-ethylpyrazine, 2-ethyl-6-methylpyrazine, etc.) in AP-HTL were identified as the cause of inhibition in co-digesters supplied, along with a greater AP-HTL contribution to the overall COD load.

Li et al. [110] evaluated continuous AD of AP HTL from Chlorella in up-flow bed reactors. Raw AP HTL caused methane inhibition at 20%, whereas zeolite pre-treatment significantly improved performance, achieving stable digestion at 60%. Methane accounted for 17.3% of the total carbon, and the abundance of functional genes aligned with digester performance. Macedo et al. [111] studied co-digestion of manure with AP HTL from wheat straw manure and sewage sludge. Straw manure AP HTL yielded 43 to 49% COD conversion to methane (17.8 g per COD), while sewage sludge AP HTL achieved 43% (12.8 g per COD) but showed complete inhibition at 17 g per COD due to nitrogen-containing heterocyclic compounds. Although hydrolytic bacteria remained resilient, methanogens were suppressed, indicating that sludge-derived AP HTL requires further treatment to mitigate inhibition.

As discussed, N-heterocyclic, ammonia, and phenols were found to be toxic for AD and responsible for inhibition. Wang et al. [112] treated municipal sludge–derived AP–HTL was processed by removing phenols (biochar detoxification) and ammonia (struvite precipitation), achieving a methane yield of 225 mL per g COD. The highest yield occurred at pH 7 with the removal of both inhibitors, exceeding ammonia-only treatment by 90% and phenols-only treatment by sevenfold, while also increasing methanogen abundance. Zhu et al. [113] evaluated conductive additives during AD of microalgae-derived AP HTL. Magnetite increased methane production by 77%, biochar enhanced degradation of nitrogen-containing organics by over 80%, and granular activated carbon effectively removed inhibitory compounds, highlighting the potential of conductive materials for improved long-term AD performance. Kulikova et al. [114] examined the composition and toxicity of AP HTL from primary and secondary sludges, reporting TOC of 4.2 to 9.6 g/L, COD of 7.9 to 14.0 g/L, and BOD₅ of 6.0 to 8.1 g/L. Noticeable toxicity was observed, with DR50 values of 64.7 to 142.2 for Artemia salina and 44.9 to 81.7 for Paramecium caudatum. However, aerobic treatment using industrial sludge (2.7 g/L biomass, 1:10 dilution) achieved 67 to 95% COD removal within 18 h, indicating good biodegradability.

Figure 9. Fate of organics during AP-recycling via AD.

Figure 9. Fate of organics during AP-recycling via AD.

4.2. Bioelectrochemical Systems

Bioelectrochemical methods are favorable for oxidizing organics into electrons via microorganisms, and those electrons can be used to produce energy and other valuable products. The oxidation of organic compounds is carried out through two types of cells: (1) microbial electrolysis cells and (2) microbial fuel cells.

A microbial electrolysis cell (MEC) is a system in which electrons are produced by anaerobic bacterial consortia through the degradation of waste. These electrons are shifted to the cathode for proton reduction to produce H₂. MECs are favorable due to lower energy consumption and are highly recommended for wastewater treatment containing a higher concentration of heavy metals [115,116]. Shen et al. [117] found that recalcitrant compounds (diethyl phthalate and dimethyl phthalate) in AP-HTL derived from corn stalk were removed with ~79% and 95% efficiency, respectively, by MEC, on account of producing hydrogen at a rate of 3.92 mL/L/day. In another study, the same author in [118] employed MEC using AP-HTL of swine manure, showing a H₂ production rate of 168 mL/L/day, higher than the previous study.

A microbial fuel cell (MFC) is a bioelectrochemical process in which organic matter is exploited for the generation of an electric current using metabolic activities. MFCs require less energy and do not demand a controlled distribution system, treat wastewater containing volatile substances (organic), and generate solids in minimal quantities [119,120]. Liu et al. [121] synthesized a carbon-based nanotube fixed-bed microbial cell to treat AP-HTL of corn stalk. The removal rate of 80% of organics was achieved with 2.41 g/L/day. Some disadvantages of bioelectrochemical systems, including longer start-up times, scale-up challenges, and the need for low-cost electrodes, still require attention [119,122].

4.3. AP-HTL Recycling in Microalgae Cultivation

Currently, microalgae species are considered to be the ideal source for the production of biofuel, as they grow in marshlands and cannot be compared with feed crops or other foods. Thus, the production of microalgae is at higher levels, and they require expensive nutrients (N, P, K) [123]. Thus, the generation of microalgae from nutrient-rich wastewater has received great attention over the last decade. In this context, various researchers performed experiments to investigate the factors enhancing the algal growth from wastewater generated from different sources, such as glycerol waste [124], biomass hydrolysate (lipid-free) [125], seaweed residues (agar-free) [126], wastewater from the oil industry [127], wastewater containing petroleum and diesel contaminants [128], etc. The very first nutrient recycling from HTL for algae cultivation was carried out by Jena et al. [40], providing the first demonstration of nutrient recycling from HTL for microalgae cultivation, reporting ~51% biomass productivity, which exceeded that obtained with a synthetic medium. Recent studies on AP-HTL–based microalgae cultivation are summarized in Figure 10, indicating that biomass production is largely governed by operating conditions. These conditions encompass HTL process parameters as well as the cultivation strategies and protocols employed for microalgae growth. The studies from Du et al. and McGinn et al. [129,130] highlighted that the yield from AP-HTL was higher than that of the conventional or synthetic media.

During the cultivation, the inorganic nitrogen in the form of ammonia was absorbed by microalgae, because algal cells favorably utilize NH₄⁺–N. Ammonia-nitrogen can be simply converted to amino acids more easily than nitrate [131]. Similarly, microalgae consume phosphorus (orthophosphates), as they enhance the growth of DNA, RNA, phospholipids, and ATP [132]. Like nitrogen, the extensive removal of phosphorus was observed, which is related to the N/P ratio of the AP. Because the considerable range for the N/P ratio for algal cultivation is from 5:1 to 12:1, if the ratio exceeds the mentioned limit, it could result in phosphorus starvation [130]. Metals such as Ca, Mg, Cu, Cd, and Zn are commonly present in sewage sludge and manure. These metals may lead to the formation of insoluble precipitates. In some cases, phosphorus is adsorbed onto hydroxides of Fe and Al and contributes to the solid phase, which can subsequently be recovered for use in fertilizers. Most of the nitrogen in AP is derived from N-containing compounds and ammonia. Though N and P are essential nutrients, there are some studies, such as Alba et al., reported that some essential nutrients other than N, P, and C inhibited the growth during the cultivation of microalgae using AP [133].

This observation is supported by Edmundson et al. [134], who experienced the enhanced growth rate (79 to 91%) of Chlorella sorokiniana. During HTL, macro- or micronutrients are being lost, so it is very important to ensure that AP-HTL contains sufficient macro (sodium, potassium, calcium, magnesium, chlorine, and sulfur, etc.) and micro (Cu, Fe, Co, and Mo). For example, Co supports vitamin B12 synthesis, while Mg promotes photosynthesis and chlorophyll formation [135]. Across different microalgae species, Chlorella species were widely investigated using AP-HTL, because these species possess a mixotrophic nature, which could assimilate a variety of organic compounds from AP, depending upon the degree of dilution and their tolerance [8]. Biller et al. [41], conducted growth trials using different microalgae species, such as Chlorella vulgaris, Scenedesmus dimorphus, and Chlorogloeopsis fritschii, and observed that Chlorella was stable towards inhibitory compounds and provided higher biomass growth at lower dilutions (200–400×), which is favored by Chen et al. [57]. Godwin et al. [136] also appreciated the findings and stated that Chlorella sorokiniana showed better tolerance towards recycled AP-HTL (concentration: 2%), while biomass from other species had a limited effect compared to its own. However, when polycultures of two to six species were used with recycled AP-HTL (<10%), they demonstrated resilience and significantly higher yields than monocultures. The same author, Godwin et al., in another study [137] recommended that it is obligatory to adjust polyculture combinations for balancing nutrient (N and P) utilization for enhanced bio-crude production over other monocultures.

Zhou [138] indicated that diverse HTL conditions resulted in the generation of several types of inhibitors (heterocyclic compounds). Conversely, Tommaso et al. [139] found that elevating the temperature from 260 to 320 °C resulted in an increase in phenol concentration (0.21–8.44%), which is recognized for its toxicity to microalgae. The results demonstrate that different HTL process variables significantly affect the efficacy of algal biomass output cultured with AP-HTL. Moreover, diluting AP-HTL to reduce inhibition from recalcitrant chemicals remains a major challenge, as it requires freshwater, making it economically unfeasible.

Consequently, the concept of utilizing low-strength wastewater has arisen as an alternative to freshwater dilution [140]. The variations in chemical characteristics of low-strength wastewater due to its inherent circumstances provide a challenge. The integration of adsorbents such as activated carbon (AC) [141] and zeolite [142] as a pre-treatment step has emerged as a promising method to mitigate the adverse effects of growth inhibitors on microalgae. However, these costly adsorbers are impractical for commercial use and clearly necessitate a replacement to address these challenges. The biochar derived from HTL biomass can serve as an economical adsorbent; however, it requires additional research and cost assessment.

Other Microbes Cultivation

Microbes can detoxify AP-HTL during the heterotrophic mode. Earlier, Nelson [68,143] has reported the growth of multiple types of microbes, such as Pseudomonas putida (P. putida), Escherichia coli (E. coli) (both are bacteria), and Saccharomyces cerevisiae (fungus) on AP-HTL derived from Nannochloropsis oculata. Results showed that P. putida and E. coli exhibited substantial growth at 20% v/v AP-HTL but ceased growth at 50%, indicating that bacterial tolerance to AP-HTL is considerably higher than that reported for microalgal cultivation. Conversely, Rhodococci have been documented to associate aromatic catabolism with lipid anabolism [144]. This route has been proven viable by He et al. [145] utilizing R. opacus, R. jostii, VanA⁻, and their co-culture to transform algal and woody APs into lipids. Their findings indicated that the Rhodococci co-culture could detoxify approximately one-third of the organic chemicals in AP by-products while enhancing lipid accumulation in cellular biomass.

In recent years, Romero et al. [42] performed a study on Chlorella vulgaris NIES 227 batch transfers and turbidostat mode; both approaches offset the toxicity and restore the health of microalgae. In adaptation, a higher addition of AP-HTL was observed from 1/600 to 1/200 without inhibition. Leme et al. [146] studied mycoremediation (bioremediation that uses fungi to clean up or detoxify contaminants) of AP-HTL with white-rot fungus T. versicolor to improve the availability of inorganic nitrogen (NO₃⁻-N and NH₃/NH₄⁺-N) for prospective use in hydroponic systems. The AP-HTL with 5% dilution did not show visible fungal growth. However, laccase activity validated fungal appearance, resulting in a significant rise in NO₃⁻-N and NH₃/NH₄⁺-N 17-fold and 18-fold, respectively, after passage of three days, which was noted as the optimal duration for treatment. Incorporation of nitrifying bacteria doubled NO₃⁻-N; whereas COD was reduced by 51.33% after 24 h, before rising due to the emergence of fungal by-products. Cantero et al. [147] reviewed nutrient recovery from AP-HTL, emphasizing that two-step chemical–biological processes and protein-rich feedstocks improve nitrogen bioavailability. However, further optimization of treatment scale, duration, and economic feasibility is essential for practical valorization.

Figure 10. Effect of AP-HTL recycling on algae cultivation. Data extracted from Jena et al. [40], Biller et al. [41], Zhou et al. [140], Hognon et al. [148], Erkelens et al. [141], Du et al. [130], Velasquez et al. [149], Barreiro et al. [37], Chen et al. [150], and McGinn et al. [129].

Figure 10. Effect of AP-HTL recycling on algae cultivation. Data extracted from Jena et al. [40], Biller et al. [41], Zhou et al. [140], Hognon et al. [148], Erkelens et al. [141], Du et al. [130], Velasquez et al. [149], Barreiro et al. [37], Chen et al. [150], and McGinn et al. [129].

4.4. AP-HTL Recycling in Gasification

Hydrothermal gasification is also a considerable route for the valorization of AP-HTL. It exploits the supercritical conditions (above 374 °C) for converting the biomass into hydrogen and other gases on account of the degradation of TOC, COD, and other various pollutants [8]. Researchers mostly prefer catalytic hydrothermal gasification (CHG) when complete degradation of biomass is required. Since AP-HTL possesses lower solids, most of the studies reported higher production of hydrogen in the absence of a catalyst, as depicted in Figure 11. However, the addition of a catalyst improved process efficiency and hydrogen yield with a CO₂ reduction. The low organic load (TOC: 3.3 g/L) demonstrated process efficiency of 35%; this could be associated with higher accumulation of oxygen and N-containing compounds or a lower concentration of acids and amides in AP-HTL [44].

Cherad et al. [45] used different organic loads of AP derived from HTL of Chlorella for gasification, and revealed that with the lowering of the concentration of AP-HTL from 11 to 2 g/L, exhibited higher yields (7-fold) of higher concentration of hydrogen. The results show that substantial hydrogen production requires reducing the organic load unless a catalyst is used. It also demonstrated that elevated temperature promotes the production of H₂ and CH₄ by reducing CO and CO₂, which could be correlated to the conversion of CO and CO₂ into methane at higher temperatures. These findings are in compliance with observations noticed by Watson et al. [151] and Duan et al. [44]. Both studies stated the significant change in the values of ammonia, TOC, and CH₃COO⁻ owing to the different biochemical composition of various macro/microalgae.

It can be concluded that microalgae-based AP-HTL with a higher value of TOC around 24 g/L led to lower H₂ than macroalgae-based AP-HTL containing TOC of 9.8 g/L. Apart from this, feedstocks containing higher acetate concentration produce more methane compared to hydrogen, which can be linked to decarboxylation of acetic acid and hydrogenation of CO₂/CO at higher temperatures, whereas adding a catalyst can increase the H₂ production but contribute to extra cost. The report of the U.S. Department of Energy stated that the catalyst-based gasification makes up 44% of the total cost, excluding feedstock [34]. Watson et al. [151], adding Raney-Ni and NaOH during gasification of human feces-based AP-HTL produced hydrogen yields of 41% and 47%, respectively, which are both higher than the control.

This is also confirmed by Cherad et al. [45], where a catalyst was employed with an organic load of AP-HTL at 11 g/L, resulting in ~2.3-fold higher hydrogen than gasification without a catalyst. The integration of HTL with gasification effectively diminished or eradicated toxic contaminants while enhancing energy yield from AP-HTL. It has been proposed that the hydrogen generated from gasification can be utilized for the hydrotreatment of the bio-crude. So far, only reductions in acetate and TOC have been observed during gasification of the AP-HTL, while ammonium, nitrate, and potassium remained largely unchanged. These nutrients are retained in the post-gasification wastewater, making it suitable for further use in algal cultivation [45].

Barreiro et al. [37] utilized AP obtained from Scenedesmus almeriensis and Nannochloropsis for SCWG and found a reduction in the load of TOC, while increasing the amount of nitrogen available as ammonium. Additionally, nitrogen-containing organic compounds from HTL-AP were degraded to ammonium during SCWG, resulting in a higher fraction of total nitrogen in the form of ammonium in AP-SCWG compared to AP-HTL-. However, the TN value showed a slight reduction in the nitrogen content of AP-SCWG. This nitrogen might be in the small amounts of bio-crude formed. The same trend was observed by Li et al. [152] and Duan et al. [44] (

Table 5. Change in characteristics of AP-HTL after SCWG or CHG.

Feedstock–SCWG–Temp. (°C)–Author	TOC (g/L)	ΔTOC ↓ (g/L)	TN (g/L)	ΔTN ↓ (g/L)	NH4+-N (g/L)	ΔNH4+-N ↑ (g/L)	PO43−-P (g/L)	ΔPO43−-P ↓↑ (g/L)	Ref.
S. almeriensis–SCWG–450–Barreiro et al.	12.57	−3.38	5.30	-0.85	3.51	0.72	0.11	0.07	[37]
Nannochloropsis–SCWG–450–Barreiro et al.	13.26	−5.59	4.90	−0.40	4.51	0.73	3.39	−0.95	[37]
Chlorella sp. and Scenedesmus sp.–-CHG–350–Ru/C–Li et al.	10.00	−9.75	9.60	−1.60	7.52	0.35	0.75	−0.41	[152]
A. pyrenoidosa-SCWG–600–Duan et al.	23.78	−20.06	NA						[44]
A. platensis-SCWG–600–Duan et al.	30.04	−21.21	NA						[44]
S. limacinum-SCWG–600–Duan et al.	21.95	−20.72	NA						[44]
N. occeanica–SCWG–600–Duan et al.	19.70	−18.25	NA						[44]

NA: Not available. CHG: Catalytic hydrothermal gasification. Δ = AP values obtained after (SCWG/CHG) minus the AP values of HTL.

). Barreiro et al. [37] also evaluated the changes in organic compounds and minerals, who noticed increased concentrations of acetic acid and phenols in the SCWG, indicating that SCWG partially degrades organic molecules, releasing simpler compounds that were previously bound in HTL-AP. Glycerol decreased during SCWG due to its degradation under high temperature and pressure, resulting in the formation of simpler compounds (

Table 6. Change in compound composition and minerals of AP-HTL after SCWG, Barreiro et al. Ref. [37].

Compounds and Minerals (mg/L)	S. almeriensis–SCWG–450	Nannochloropsis–SCWG–450
Acetic acid ↑	1254 (+798)	742 (+542)
Glycerol	1755 (−1070)	2784 (change: NA)
Phenols ↑	8 (+15)	36.2 (+14)
K ↑	1276 (+15) Rec. (101%)	1428 (+137) Rec. (109%)
Na ↓↑	450 (−35) Rec. (92%)	3050 (+337) Rec. (111%)
Ca ↓	5.3 (−0.2) Rec. (96%)	4.8 (−2.6) Rec. (46%)
Mg ↓	6.3 (−3.9) Rec. (38%)	0.4 (−0.1) Rec. (75%)

NA: Not available. Rec.: Recovery.

). As for minerals, SCWG showed high recovery of K and Na, which are valuable for algae cultivation. However, Ca and Mg exhibited slightly low recovery. This highlights the efficiency of SCWG in preserving certain nutrients while altering the organic composition of the by-products. The comprehensive chemical composition of AP–HTL feedstock is summarized in Tables S1–S4 (Supplementary Materials) for HTL, AD, microalgae cultivation, and SCWG, respectively.

4.5. Aqueous Phase Reforming

Aqueous phase reforming (AP-Reforming) is also another route for producing hydrogen at low temperatures, which facilitates the conversion of organics, particularly short-chain acids, alcohols, and sugars into hydrogen-rich gas at moderate temperature and pressures (lower than gasification) using a variety of catalysts. Recently, many researchers have tried to exploit the organic potential of AP-HTL using AP-reforming. For instance, Arlt et al. [153] utilized multiple components with a significant amount of lignin-containing AP-HTL and liquefied them into two different facilities lab scale (30 to 60 mL/h) and a pilot scale (44 L/h) to optimize the reaction conditions. Response surface methodology revealed that temperature was the most dominant variable to model the conversion of reactants. The stable conditions were achieved after running continuous experiments for 100 h. The highest hydrogen yield of 58% and carbon conversion of 54% were recorded at 273 °C. Fraia et al. [154] used lignin-rich stream after ethanol production from lignocellulosics into bio-crude production in continuous HTL, and they produced H₂ by AP-reforming. The catalysts (Na₂CO₃ and NaOH) were also added in HTL runs. NaOH produced a higher bio-crude yield 27%; however, AP-HTL with a higher concentration of phenolics was obtained, which were removed by liquid–liquid extraction using butyl acetate, preserving the AP-reforming catalyst stability. The hydrogen produced up to 146 mmol/L-AP. The mass balance revealed that AP-reforming could fulfill 46% of H₂ requirements for the hydrotreatment of bio-crude, thus significantly improving the suitability of the process. Tito et al. [155] performed a techno-economic analysis of combined HTL and AP-reforming, using corn stover and a lignin-rich stream post-ethanol production (CS). They concluded that the minimal fuel selling prices (MFSP) for the bio-crude, at a 0% internal rate of return, were €1.23 per kg for the lignin-rich stream and €1.27 per kg for corn stover. The heat exchangers constituted most of the fixed capital investment, although power and feedstock significantly influenced the operational expenses. The execution of AP-reforming was especially advantageous with CS, yielding 107% of the hydrogen necessary for bio-crude enhancement. In this instance, AP-reforming successfully diminished the hydrogen production cost to 1.5 €/kg, rendering it a competitive technique relative to traditional electrolysis.

5. Challenges and Recommendations for the Integrated AP-HTL Biorefinery Approach

Recycling of AP-HTL has demonstrated positive correlations with bio-crude yield and energy recovery. However, recycling leads to a significant increase in the COD and TN of AP-HTL. This may present serious challenges for subsequent disposal, reduce its effectiveness for algal cultivation, contribute to environmental pollution if discharged, and negatively affect the long-term storage stability of this wastewater. Consequently, while the recycling of AP-HTL is straightforward, economical, and easily scalable, it intensifies the issues related to the toxicity of AP-HTL. Anaerobic fermentation of AP-HTL holds significant potential for commercial application, being considered a well-established technology. The inadequate conversion of nitrogen-containing organics indicates that the remaining effluent requires further post-treatment through alternative valorization methods. Moreover, the pre-treatment of AP-HTL, the enrichment of functional microbes, and bio-augmentation must be prioritized to improve the anaerobic fermentation of AP-HTL. The cultivation of biomass utilizing AP-HTL as a culture-medium results in substantial nutrient recovery and represents a cost-effective method for fertilizer production. The inhibition and low efficiency of organic utilization restrict the use of AP-HTL with a high concentration of organics as a culture medium. Furthermore, the commercial utilization of the generated biomass encounters obstacles due to the environmental hazards associated with heavy metal accumulation and the extensive land area needed. Gasification is characterized by significant energy requirements and entails substantial capital and operational expenses related to catalyst integration and rigorous reactor parameters. Gasification must address challenges related to low efficiency, inadequate yields, catalyst deactivation, and gas separation/purification before it can be considered a mainstream treatment technique for AP-HTL. For better understanding,

Table 7. The advantages, challenges, and operational issues of existing AP-HTL recycling systems.

Approach and Product	Advantages	Major Challenges	Operational Issues
HTL (Bio-crude)	Higher carbon recovery Promotes secondary oil-forming reactions Reduces freshwater demand Improves process efficiency Improves slurry pumpability	Recalcitrant compound accumulation Ammonia buildup Increasing molecular complexity Yield plateau after repeated cycles	Recirculation ratio control Salt and inorganic concentration Reactor fouling risk Increased heat duty
Anaerobic Digestion (Methane)	Commercially mature Net energy recovery Reduces wastewater burden Integrates with existing digesters	Phenolic inhibition Free ammonia toxicity Limited biodegradability of refractory organics	Organic loading sensitivity Volatile fatty acid accumulation Requires dilution or co-digestion Slow microbial acclimation
Biomass Cultivation (Algae/Biomass)	Enables nitrogen and phosphorus recycling Reduces fertilizer demand Supports circular fuel systems Potential carbon dioxide coupling	Toxic organics suppress growth Light attenuation from dark AP Composition variability	Dilution management Culture instability risk Nitrogen speciation control Contamination prevention
Supercritical Water Gasification (Syngas/H2-rich gas)	Converts recalcitrant organics Very high carbon conversion Maximizes energy recovery Compatible with thermochemical platforms	High capital intensity Severe conditions Corrosion and salt precipitation Catalyst deactivation	Reactor plugging risk Requires corrosion-resistant alloys Strong heat integration Feed filtration often necessary
Bioelectrochemical Systems (H2/Electricity)	Treats inhibitory wastewater Low sludge production High theoretical efficiency Effective polishing stage	Expensive electrode materials Membrane fouling Sensitive electroactive biofilms	Conductivity control Hydraulic retention time optimization Internal resistance management Electrode maintenance
Wet Air Oxidation (Short-chain acids, N, NH4+, etc.)	Effective destruction of toxic organics Major chemical oxygen demand reduction Improves downstream biodegradability More energy-efficient than full incineration	High temperature and pressure requirements Corrosion concerns Possible formation of refractory oxidation by-products	Oxygen transfer efficiency Materials selection critical Slurry handling challenges Heat recovery needed for economic viability
Chemical Separation (Values added chemicals)	Generates additional revenue streams Reduces downstream toxicity Improves biorefinery economics Enables cascade utilization	Complex multicomponent chemistry Selectivity limitations Solvent or membrane cost Additional processing steps	Solvent regeneration energy demand Membrane fouling Adsorbent saturation Emulsion handling

outlines the advantages, challenges, and operational issues associated with these technologies.

Bioelectrochemical systems have proven effective at converting recalcitrant compounds such as phenols, furan derivatives, and N-heterocycles. In addition, these systems can generate useful outputs, including energy carriers, gases, and electricity. Future research should focus on enabling commercial-scale deployment by improving process efficiency and reducing overall costs.

The separation of value-added compounds from AP-HTL represents a promising strategy to improve the economic feasibility of HTL. Enhanced optimization of this method demonstrates possibilities for establishing mass market sales channels for solvents, acids, and other chemical precursors. Current research, however, is limited to AP-HTL with a less heterogeneous composition, such as the extraction of phenols and organic acids from AP-HTL generated from lignocellulose. The complex composition of AP-HTL has yet to be examined. Jensen et al. [18] demonstrated that approximately 5–15% of the AP should be removed before recirculation to prevent chloride buildup, which increased by more than 100% after six recirculation cycles when woody biomass was used as the feedstock. Therefore, although AP-HTL recirculation cannot completely eliminate the need for downstream wastewater treatment technologies (e.g., AD or gasification), it can substantially reduce the volume of wastewater requiring treatment.

Given these challenges, no single strategy is universally suitable for AP HTL recycling. Rather than prioritizing APR within the HTL framework, process selection should be guided by the chemical composition of the obtained AP, as illustrated in Figure 12. Acid-enriched AP, particularly rich in short-chain organic acids derived from carbohydrate hydrolysis, can be recycled to the HTL reactor to enhance secondary oil formation or directed to AD for methane production. However, repeated recycling poses risks of acid accumulation, reactor corrosion, char formation, and methanogen inhibition, requiring strict control of recycle ratios, pH buffering, and residence times. Additionally, AP contains high levels of total ammonia-nitrogen, occasionally exceeding the anaerobic consortia’s inhibitory threshold of 3000 mg/L, which can cause free ammonia toxicity and accumulation of fatty acids [114,156]. Nitrogen-rich AP, dominated by ammonia and soluble nitrogenous compounds from protein decomposition, offers potential for nutrient recovery via struvite precipitation, electrostripping, or wet oxidation. Challenges include ammonia toxicity, inhibition of downstream biological systems, and deterioration of bio-crude quality during recycling; thus, nitrogen-selective recovery is recommended over bulk reuse. Phenolic AP, containing lignin-derived aromatics, is resistant to biodegradation and toxic to microbes. Thermochemical or electrochemical treatments, such as wet air oxidation, electrochemical oxidation, or gasification, are necessary to degrade phenolics and recover energy, while conventional biological recycling is impractical [157]. Mineral-rich AP, high in salts, phosphates, and trace metals, can support microalgae cultivation and nutrient recovery, but high salinity and heavy metal accumulation pose microbial inhibition and environmental risks, making controlled nutrient extraction or high-severity thermochemical treatment preferable.

Across all categories, integrating AP-HTL recycling into a circular framework that links AD, algae cultivation, HTL, gasification, and hydrogen-assisted bio-crude upgrading can maximize resource recovery, reduce toxicity, and enhance process sustainability. Overall, a categorization-based approach enables targeted process design, balancing resource valorization with operational stability and environmental compliance. Building on this concept, optimizing HTL performance through the valorization of AP enables the linkage of plausible pathways with commercially available technologies, thereby establishing the paradigm of a sustainable biorefinery. As illustrated in Figure 13, the most preferred option is recirculation to the HTL unit through route 1. Once the maximum bio-crude yield has been achieved, AP-HTL can either be recycled through an alternative pathway or integrated through route 2 and directed to anaerobic digestion or gasification. Both processes generate valuable energy carriers, methane from anaerobic digestion and hydrogen-rich gas from gasification, while their by-products, digestate and effluent, can serve as nutrient sources for algae cultivation. AP-HTL can also be directly used for algae cultivation after nutrient recovery, such as struvite or ammonia extraction through electrostripping. The cultivated algae can subsequently be reused as a feedstock for HTL, thereby creating a closed-loop system. Furthermore, methane produced from anaerobic digestion can supply heat for the HTL process, while hydrogen-rich gas can meet the hydrogen requirement for hydrotreating.

Earlier, Shanmugam et al. [158] investigated adsorption dynamics of pyrolysis-derived biochar with AP–HTL, and they found comparable characteristics to granular activated carbon (GAC); hence, biochar may function as a viable alternative to GAC. Elliot et al. [159] suggested that the AP obtained from the gasification of AP–HTL can serve as a medium for algal cultivation. Whereas Jones et al. [160] asserted that H₂ from gasification can be supplied for the upgrading of bio-crude.

Some life cycle assessment (LCA) and techno-economic analysis (TEA) studies substantiated the value of AP-recycling [9,101,161], reporting that the direct AP-recycling within the HTL unit often decreases the capital expenditure and the MFSP [101]. Conversely, it increases natural gas consumption and poses risks of heteroatom and salt formation, hence elevating upgrading difficulties. Karka et al. [161] calculated CO₂-equivalent emissions of around 0.3 to 2.5 kg/kg of fuel mixture through LCA, indicating that AD management and its integration with other technologies can modify global warming potential (GWP) by a factor of two to four. As per Moser et al. [162], the GWP of manure-derived fuel is 1.18 kg CO₂ equivalent per kilogram, indicating approximately a 70 percent reduction in greenhouse gas emissions compared to conventional jet fuel. Snowden-Swan et al. [155] at PNNL estimated the minimum fuel selling price to be as low as 3.46 USD per gasoline gallon equivalent under optimal conditions. Furthermore, Zhu et al. [101], Alamo et al. [163], and Zoppi et al. [164] conducted LCA and TEA and promoted the integrated approach of AP-HTL.

Future HTL research should move from yield-focused optimization to a mechanism-driven, system-level approach that explicitly links AP evolution with bio-crude yield, quality, and process stability. Standardized APR protocols and feedstock-specific recycling strategies are required to improve reproducibility and enable meaningful comparison across studies. Nitrogen accumulation and recalcitrant AP compounds should be treated as design variables through selective mitigation, staged recycling, or catalytic control. Integrated valorization pathways should be co-designed with HTL operating conditions and supported by long-term continuous studies together with integrated TEA and LCA to enable translation from laboratory research to industrially viable and sustainable biorefinery systems.