9.2.1 Reaction Mechanism

HTL employs water at near-critical conditions as a reaction medium and solvent. At near-critical conditions, the solvent properties of liquid water change substantially (including changes in dielectric constant, density, diffusivity, polarity, viscosity, H-bonding, and H⁺ donor capabilities), transforming it from a polar, highly H-bonded solvent to exhibiting behaviour more typical of a nonpolar organic solvent. In this form, it can act as a solvent for a range of organic reactions and facilitate the breakdown of biomass to bio-crude [10] (Fig. 9.2).

The prevention of vapour formation at high pressure means the enthalpy associated with the phase change of water is largely reduced, giving vastly increased energy efficiencies for HTL processing over pyrolysis [12].

HTL is comprised of hundreds of simultaneous reactions, which are not well-characterized in the literature. In hot compressed liquid water (near the critical point of 374°C and 22.1 MPa), there are two competitive reactions: hydrolysis and repolymerization [13,14]. The former is more important and dominant at the early stages of the process, when the microalgae is decomposed and depolymerized into small compounds. These are highly reactive, and rapidly polymerize and form liquid bio-crude, gaseous, and solid products [15–17]. At higher reaction temperatures and longer reaction times, repolymerization, condensation, and decomposition of substances from different phases may occur. This results in an increase of the solid and gas yields and reduction of the bio-crude yield [18–20]. Dehydration reactions, leading to oxygen content reduction, also play a key role in improving the quality of the bio-crude.

9.3.1 Effect of Biomass Composition on Bio-Crude Oil Production

It has been previously shown that bio-crude yields from HTL of microalgae can be strongly correlated to the biomass composition. The lipid fraction of the biomass is much more readily converted into bio-crude than other biomass components, with a study by Biller and Ross reporting that conversion efficiencies of different algal components to bio-crude oil were in the order lipids > proteins > carbohydrates [15]. This explains the particularly high yields obtained from lipid-rich microalgae. However, as HTL utilizes all biomass components, a high lipid content is not essential for obtaining good bio-crude yields, in contrast to algal biodiesel, which relies entirely on lipids. This effect has been demonstrated by Yu et al., who reported a yield of 39.0% bio-crude oil from an algae strain containing only 0.1% lipids [21]. Utilizing the entire algal biomass for processing lifts the constraints on high-lipid algal strain selection, and can significantly reduce cost and energy requirements of cultivation by using faster-growing strains or algal communities.

Although the specifics of the effects of biomass composition remain unclear, a number of studies have examined the role of the different component fractions (eg, lipids, proteins, and carbohydrates) in bio-crude production via HTL. Most liquefaction processes under optimized conditions have resulted in bio-crude yields around 30–45% [15,22–25], regardless of algae strain [12], although, notably, Li et al. obtained yields of 55% for Nannochloropsis sp. under HTL at 260°C for 60 min and at 25% algal loading, and 83% for Chlorella sp. (220°C, 90 min, 25% algal load) [26]. The two algal species have very different compositions (Nannochloropsis contains protein, lipids, and carbohydrates in an approximate ratio of 52:14:22, whereas the same components in Chlorella have a ratio of 9:60:13), but in both cases, bio-crude yield was enhanced with respect to initial lipid content, further reinforcing the advantages of hydrothermal processing over lipid extraction for algal biodiesel [22].

Higher heating values (HHV) of 25–35 MJ/kg are typical for bio-crude, and higher lipid levels in the biomass appear to correspond to higher HHV, although the two properties are not linearly related. Li et al. found, under optimal conditions, Nannochloropsis and Chlorella bio-crude HHVs were 31.5 and 34.2 MJ/kg, respectively, with a maximum HHV of 37 MJ/kg. Although this constitutes a significant increase with respect to the starting biomass (22.4 and 32.3 MJ/kg initially), it still falls short of the energy content of mineral crude (41–48 MJ/kg) [27]. Interestingly, higher bio-crude yields do not necessarily correspond to better oil properties; Nannochloropsis processed at 350°C yielded 34% oil with an HHV of 38.1 MJ/kg, while 46% oil was obtained at a processing temperature of 310°C, with a much lower HHV of 27.7 MJ/kg [25] (Table 9.1).

The numerous simultaneous reactions occurring under HTL conditions lead to a bio-crude containing a diverse range of chemical compounds, the main constituents of which have been found to be C5–C16 cyclic nitrogen compounds, C15–C33 branched and unbranched hydrocarbons, branched oxygenates, aromatic compounds, and heterocycles [26]. Brown et al. carried out a detailed analysis of bio-crude produced by liquefaction of Nannochloropsis sp. Over 90 compounds were detected using GC–MS, although some lighter organics are likely to have been lost during sample extraction, and heavier compounds may not have been detected by GC. A brief overview of the major compounds is presented in Table 9.2.

Table 9.2

List of Major Bio-Crude Components Detected by GC–MS [28]

Major Compound	Structure
Indole
Methylindole
1-Pentadecene
Heptadecane
Isomers of 2-phytene
Myristic acid
Phytane
Palmitoleic acid
Palmitic acid
Oleic acid
Stearic acid
Cholest-4-ene
Cholest-5-ene
Cholesta-3,5,-diene
Cholesterol
Cholest-4-en-3-one
Cholest-4,6-diene-3-one

Elevated heteroatom (O and N) content with respect to mineral crude oil is typical of bio-crude oils [9,10,29]. Higher O and N levels give rise to undesirable fuel properties, such as high acidity and viscosity [24], and the increased diversity of chemical composition can negatively affect combustion performance, storage stability, and economic value [24,29]. The heteroatom content of the bio-crude can generally be attributed to the high O and N content of the starting biomass, although there is not always a direct correlation: Chlorella biomass containing 1.9% N was processed to bio-crude with a 0.3% N content, while Nannochloropsis (7.5% N initially) resulted in a bio-crude with 5.4% N. Chlorella bio-crude was also found to have a higher O content (11.5% compared to 9.5% for Nannochloropsis), although the O content of the biomass was 21.8%, compared to 29.1% for Nannochloropsis [26].

Additionally, the elevated N content of bio-crude of up to 11%, arising from the proteins in the starting biomass (compared to around 0.1, and seldom above 1%, for mineral crude) [30,31], limits its direct usability as a fuel, as it can lead to increased NOx emissions. Significantly, higher nitrogen levels may contribute to catalyst poisoning, making these bio-crude oils unsuitable for co-refining in existing refineries [12,31]. Hence, although bio-crude produced via HTL presents numerous advantages compared to lipid extraction, the product needs significant upgrading to obtain a suitable aviation fuel [24]. An alternative solution posed by Garcia Alba et al. suggests that proteins could be extracted prior to HTL processing and sold as a high-value co-product in a biorefinery paradigm [13]. However, this clearly has implications for bio-crude volumes, particularly when low-lipid algae such as Spirulina are used.

9.3.2 Effect of Microalgal Loading on Product Formation

One of the key benefits of wet biomass processing in comparison to lignocellulosic biomass or dry processing is that microalgae forms a slurry with water, which can be easily pumped and flowed through the reactor. Particularly for continuous HTL operation, determination of the optimal loading of microalgae in the slurry is crucial.

From the literature, many factors such as the scale of the reactor and economics need to be considered with the selection of the microalgal load used in HTL. Hence, there is no clear connection between the microalgal load (total solid fraction, TS) in the feed and the bio-crude yield. Studies of HTL of algae performed over a large range of microalgal loads (ranging from 4% to 50%) have apparently yielded contradictory results [6]. For example, Jena et al. concluded that an increase of the biomass fraction in the feed by 10% resulted in a jump of more than 20% in the bio-crude yield [32], though the opposite findings were reported by Garcia Alba et al. [13].

The optimum ratio of microalgae and water in the feed is presumably dependent on species and reactor design. A high microalgal loading would be economically beneficial, due to reduction of the water content of the feed, leading to a reduction in the energy required for heating the water, and a correspondingly higher volume efficiency and productivity of the HTL reactor [12]. However, there is expected to be an upper limit due to mass transfer limitations and the formation of blockages in continuous thermochemical conversion [33]. Continuous liquefaction studies have shown that while some practical operational issues of reactor blockage may occur at microalgal loads above 5 wt% in the feed [31], processing of microalgae slurries with microalgal loads of up to 35% (obtaining a bio-crude yield of up to 63.6%) are possible [34]. Another key consideration on the loading of microalgae is the cost of drying the original feedstock, which plays a significant role in dictating the most suitable loading achievable.

9.3.3 Effect of Holding Temperature on Microalgal Processing

The holding temperature has a remarkable influence on the yield and properties of HTL products. In general, an initial rise in the holding temperature triggers bio-crude production. After reaching a maximum value for the bio-crude yield, further increase in the holding temperature inhibits biomass liquefaction.

The optimization of holding temperature varies depending on microalgal species; however, in the majority of cases, compared to a bio-crude yield obtained at ca. 350°C, lower holding temperatures (~310–320°C) generated higher bio-crude yields. This effect was observed by Dote et al. [35], and subsequently by Minowa et al. [36], who observed that yields of bio-crude oil from the liquefaction of Dunaliella tertiolecta increased from 30.9% to 43.8% when increasing the holding temperature from 250°C to 300°C, but dropped to 42.6% when processing temperature was increased further to 340°C [36].

These findings are mirrored in the work of Gai et al. for each retention time examined, oil yields from Chlorella pyrenoidosa increased when holding temperature was increased from 260°C to 280°C, and proceeded to decline when reaction temperatures were brought to 320°C [37]. The authors suggested that with increasing temperatures, more secondary decomposition reactions may be triggered, resulting in the production of gases and char rather than oils. Regrettably, they were not able to satisfactorily confirm this using the gas yields obtained in their investigation, owing to the high experimental error.

In addition, the holding temperature has a strong influence on the bio-crude properties: increasing the temperature results in a decline in oxygen content and consequently increases the HHV of the oil [36,38]. The nitrogen content in turn appears to increase with holding temperature, likely due to the promotion of protein decomposition at high temperatures, although the opposite effect has been reported in other studies [21].

The length of time the reaction is held at temperature also has a significant effect on product formation. It is expected that short retention times should be sufficient to degrade the biomass effectively, since hydrolysis and decomposition reactions are expected to be rapid under supercritical conditions [39]. However, there is still a relationship between holding temperature and holding time when bio-crude yield is considered. In many cases, to attain a high bio-crude yield, higher maximum temperatures require shorter holding times. For example, Karagoz et al. determined that at low temperatures (150°C) long holding times favoured the conversion of sawdust to oil, with the opposite trend observed at higher temperatures (250–280°C) [40]. Yu et al. also found that while the overall conversion of biomass and gas yield increased at higher temperature [41], a 120-min hold time was needed at 280°C to produce a bio-crude yield of 39.4%, whereas at 370°C, only 5 min was needed to attain a bio-crude yield of 49.4% [13].

While the holding time has an effect on the bio-crude yield, due to the influence of other operational parameters such as temperature, algae strain, and algal loading, it is difficult to quantify differences across all published studies [42]. There is no consensus on the optimum holding time for continuous operation in the literature. However, as hold time has such a significant effect on the oil yield and composition, this needs to be carefully optimized for continuous HTL reactions, requiring more detailed research and analysis of kinetic data in this area.

9.3.4 Effect of Heating Rate on the Crude Bio-Oil

While the effect of heating rate on the bio-crude yield is also dependent on the algal strain, maximum temperature, and algal load, it has recently been suggested that rapid heating rates and shorter reaction times may result in comparable, if not better, bio-oil yields than slow temperature ramps with longer retention times. For example, bio-crude yields of up to 66% were obtained during the conversion of Nannochloropsis using a holding time of only one minute. This was achieved by placing batch reactors into a sand bath with a temperature significantly above the required set-point temperature, enabling very fast heating rates [43]. This would support conversion by continuous HTL reaction. However, other studies report that while higher heating rates are beneficial for bulk fragmentation of biomass and inhibit char formation, they do not have a great effect on the bio-crude yield [42].

9.3.5 Effect of Catalysts on the HTL of Microalgae

The effect of both homogeneous catalysts, such as organic acids and organic bases, and heterogeneous catalysts, such as metallic oxides, on the HTL of microalgae has been investigated. Homogeneous organic catalysts are desirable since they are inexpensive, do not require separation prior to further upgrading, and reduce the degree of coking [44]. In many cases, HTL experiments have been conducted in the presence of inorganic salts such as sodium carbonate and other water-soluble species [15,44,45]. However, while these catalysts appear to selectively promote the decarboxylation of carbohydrates, they have been found to have a detrimental effect on oil formation from lipids and proteins [15]. One study on the conversion of albumin suggests that the nitrogen partitioning to the oil may be reduced with the presence of sodium carbonate [35], but this observation has not been reported in other similar studies [44].

While homogeneous catalysts are difficult to recycle, heterogeneous catalysts allow post-HTL separation and high reaction selectivity. Liquefaction in the presence of heterogeneous metal catalysts has been reported to give increased liquefaction yields compared to catalyst-free conditions. Furthermore, the presence of the catalyst had a significant influence on the physical properties (eg, the colour and apparent viscosity) of the product oils [45]. The highest oil yields (57%) were achieved with supported Pd catalysts; however, Ni- and Ru-based catalysts produced bio-crudes with the lowest N content [45]. Pt, Ni, and CoMo catalysts were also reported to enhance the degree of deoxygenation, but with reduced bio-crude yields [23].

9.3.6 Effect of Initial Pressure and Additional Reductive Gas

For batch HTL operation, high pressure is required to ensure the water remains in the liquid phase, reduces the enthalpy of phase change of water, and improves the efficiency of the HTL process. Several investigations have introduced additional pressure by charging the reactor with a gas prior to reaction. For batch HTL operation, bio-crude yield was reportedly not affected by the initial pressure [41]. The gases most commonly used are either inert gases (eg, nitrogen, argon) or reducing gases such as CO and H₂. While reducing gases are added in order to reduce the oxygen content in the bio-crude, and stabilize the fragmented products of liquefaction, their hazardous nature, gas channelling, and maldistribution need to be carefully considered during the operation [42]. These problems can be overcome by using hydrogen donor solvents, such as tetralin and tetrahydrophenanthrene [42], which have been found to be more favourable than the parent gases in the hydrogenation of biomass fragments. Hydrogen donors (eg, tetralin) were reported by Wang et al. to be more effective than H₂ for increasing the bio-crude yield and reducing the level of oxygenates in the bio-crude, which typically corresponds to an increase in HHV and overall improvements in bio-crude quality [46].

9.5.1 Upgrading of Bio-Crude Oil Produced by the HTL of Microalgae

Only a few studies have investigated the upgrading of HTL oils from microalgae (Table 9.3), with the majority of these studies conducted in batch and achieving only partial nitrogen removal. While the choice of catalyst was found to impact the physical properties of the reaction products, it had little impact on the denitrogenation performance itself [66]. Instead, the degree of denitrogenation appears to be mainly dependent on the reaction temperature, which suggests that these reactions are thermally controlled [66]. Unfortunately, the high temperatures also result in a significant reduction in the hydrogen content of the oils, presumably due to an increase in the aromatic content of the reduction product above desirable levels. Potentially, the poor performance of the selected catalysts under these conditions could lead to insufficient hydrogen being provided for complete hydrogenation. It has also been suggested that the accumulation of ammonia may be limiting the complete reduction of nitrogen compounds [69]. Nonetheless, more recent studies have demonstrated the full catalytic upgrading of bio-crude directly to suitable hydrocarbon fuels. Research has focused on both in situ heterogeneous catalysis and hydrotreatment of the bio-crude after separation [70]. Duan and Savage reported on the use of a 5% Pd/C catalyst for hydrotreatment of algal bio-crude oil, finding that the bio-crude HHV from Nannochloropsis was significantly improved from approximately 37 MJ/kg to 44 MJ/kg, with corresponding improvements in the kinematic viscosity, and significant reductions in N, O, and S content. However, these results were achieved over long retention times of up to 4 h and 80 wt% catalyst loadings, and the yield was substantially decreased [71]. Other studies have centred around supported Pd catalysts, though more recently, Li and Savage reported the upgrading of algal crude oil produced by HTL using HZSM-5, generating a paraffin-like oil composed of >95 wt% C and H, which retained 80% of the energy content of the initial bio-crude and was suitable for use as a liquid fuel [65].

A number of studies have attempted to upgrade crude bio-oils produced by the liquefaction of Nannochloropsis sp. [66,67] or Chlorella sp. [8] in the presence of supercritical water. It was hoped that the reaction of water with hydrocarbons would generate additional hydrogen which, in turn, would promote the removal of heteroatoms from the bio-crude. However, while the presence of water did not appear to have a beneficial effect on the denitrogenation performance, it was found to increase the oxygen content of the reaction product [8]. The best results in the presence of supercritical water were obtained over Pt/C catalyst at 530°C with a reaction time of 6 h, with almost complete sulphur removal and a reduction in nitrogen content from 4.0% to 1.5% [66]. In contrast, upgrading of a liquefaction oil obtained from Nannochloropsis in the absence of water over an HZSM-5 catalyst, at 500°C, and with a reaction time of 4 h, resulted in a reduction in the nitrogen content from 5.3% to 1.6% [65].

Alternatively, Roussis et al. performed experiments to thermally treat HTL bio-crude oil over a temperature range of 350–450°C, with a residence time of 60 min [72]. While the total nitrogen content was not affected by the thermal treatment, the total oxygen content was significantly reduced from 5.7% in the bio-crude to 0.2% in the product oil treated at 400°C. Thermal treatment was also found to reduce the total acid content, as well as the boiling point range of the bio-crude oils, making them more volatile and less viscous. In addition, trace metals were partly removed from the bio-crude. Similar findings were also reported by Bai et al., who used a two-step process to hydrothermally treat the bio-crude oil [73].

These findings demonstrate that it is possible to process the HTL bio-crude in a similar way to heavy crude oil, continuously treating in a number of subsequent processes, including thermal treatment, hydrotreatment, and hydrocracking, to make fractions suitable for further processing towards production of different commercial products, such as gasoline, diesel, and jet fuel.

In a similar effort, promising results were achieved during the continuous hydrotreating of bio-oils produced by the HTL of various strains of Nannochloropsis over sulphided CoMo catalyst supported on fluorinated alumina [34]. Single-stage upgrading at 405°C with a space velocity of 0.20 h⁻¹ and a pressure of 14 MPa in excess hydrogen produced oils with a nitrogen content ranging from 0.07 wt% to 0.25 wt%. An even lower nitrogen content could be achieved by pretreating the oil at temperatures between 125°C and 170°C in the first quarter of the reactor, before conversion at 405°C. While this study demonstrated that almost complete denitrogenation of microalgal crude oils can be achieved under continuous conditions, further studies are required to verify these findings and optimize the reaction conditions.

9.7.1 Algae Cultivation and Waste Water Treatment

Significant value could be added by combining the microalgal production process with a secondary objective, such as waste water treatment, the recovery of metals from mining waste, or carbon sequestration from power plant effluents. Waste water treatment in particular could provide substantial quantities of algal biomass, without requiring the addition of costly nutrients. The environmental remediation industry has vastly expanded in recent years, with the remediation market in the United States alone generating an estimated $12.8 billion in 2010, of which waste water treatment represented just under 50% [75]. Wastewater treatment via algal remediation lends itself well to synergistic combination with biomass production, and numerous investigations into the cultivation of algae on industrial, municipal, and agricultural waste waters have already been carried out [76,77].

Global municipal waste water production amounts to approximately 300 billion m³, of which just over 50% is currently treated [78]. Assuming an average biomass yield of 1 g/L, and a liquefaction yield of 30%, complete conversion of the existing waste water treatment facilities could result in an annual bio-crude production of up to 90 billion L.

In addition, waste water treatment using microalgae has been proven to be highly effective in reducing the concentration of N and P pollutants, which are used as nutrients, in the effluent [79]. Compared to conventional methods such as chemical precipitation or the production of an activated sludge, forming waste products which are often disposed of by landfill, microalgal treatment provides a much more sustainable route as it allows the efficient recycling of these nutrients [77].

Metal recovery from mining waste could be another lucrative secondary function of microalgae cultivation. A number of studies have explored the concept of algal remediation, and although most studies to date have employed microalgae only as a nonliving adsorbent, remarkable metal concentrations were observed in the recovered biomass, which could potentially be retrieved from the liquefaction residue. For example, one gram of biomass produced from Oedogonium sp. adsorbed up to 145 mg of Pb(II) when exposed to a lead solution containing a metal concentration of 200 mg/L for 90 min, corresponding to a lead recovery of 35% [80]. Chromium absorption by a strain of the filamentous algae Spirogyra was somewhat lower, at only 14.7 mg/g, however the initial concentration of chromium in the treated solution was also substantially depleted, with a reduction of 5 mg/L observed [81]. More recent studies have also looked at the direct cultivation of algae on industrial waste water, with acid mine drainage representing another large-scale source of water pollution in need of continuous remediation [82]. The diatom Planothidium lanceolatum, for example, could be grown at respective Cd, Zn, and Cu concentrations of 0.1, 0.2, and 0.4 mg/L, without significant effects on photosynthesis [83]. Metal toxicity is a clear challenge, however, and must be addressed before living microalgae can be used for mining waste treatment [84].

9.7.2 The Biorefinery Concept and Nutrient Recovery

Within the biorefinery paradigm it is necessary to consider upstream sustainability factors, as well as final product quality. Despite the advantages conferred by HTL, cultivation of algal biomass is still a relatively energy-intensive process and requires high inputs of water, nutrients and CO₂ [85]. The HTL paradigm goes some way to solving those issues, though the process must be optimized to render the process economically feasible. As well as optimizing bio-crude yields, it is also vital to maximize the carbon efficiency, retain efficient water and nutrient recycling, and ensure an inexpensive source of CO₂. These issues can be at least partially addressed by encompassing nutrient recycling from the aqueous phase, as well as augmenting algal growth using the CO₂ from gaseous products [13,32,86]. This nutrient-cycling concept has been discussed in recent literature, and has been described as ‘Environment-Enhancing Energy’ or E²-Energy/E2E by Zhou et al. at the University of Illinois [87–89].

A reliable, low-cost water supply is critical to the success of biofuel production from microalgae to fulfil both economic and sustainability criteria [90]. Recirculating water is an obvious route to reducing overall water consumption. For nonthermochemical methods, this comes with a higher risk of infection and growth inhibition by bacteria, fungi, and viruses found in recycled water directly after biomass extraction. HTL confers the additional advantage of destroying biotic toxins (bacteria, viruses, and even prion proteins) in the bio-crude and aqueous phase [10], although nonliving and inorganic growth inhibitors remain.

The HTL route also concentrates trace mineral matter and nutrients, such as nitrogen, phosphorus, and potassium, as well as Fe, Ca, Mg, and polar organics in the aqueous phase [12,38,86]. This could present a route for simultaneous nutrient recycling and remediation of waste water from a synergistic process, which could not be discharged without further treatment.

Recovering phosphorus is crucial for continuing global agriculture; as well as being a dwindling resource in itself, phosphorus is extracted using fossil fuels, therefore peak oil and peak phosphorus may be intrinsically connected [91]. Apart from the impact on mineral resources, the vast quantities of nutrients needed for cultivation can severely affect the energy balance owing to the energy-intensive production [92]. As a result, nutrient provision for algal cultivation is a key sustainability concern [93], and nutrient recovery is a crucial step in making third-generation biofuel production viable [94].

In hydrothermal processing, high protein levels in the feedstock can lead to accumulation of light organics in the aqueous phase [26] (up to 50% of biomass carbon has been found to accumulate in process water) [86], leading to reduced bio-crude yields and poor carbon efficiency. Although this may be detrimental in some ways (in addition to carbon losses from the bio-crude, it has been noted that phenols present in the aqueous process water may inhibit algal growth) [95], some studies have found that elevated levels of total organic carbon (TOC) in algal growth media may actually supplement algal mixotrophic growth, leading to higher biomass yields [96].

Consequently, reusing HTL process water for algal growth may also improve carbon efficiency, with nutrient recovery from the aqueous phase simultaneously being used to treat process waste water, before release into waterways. Biller et al. carried out a comprehensive study in 2012 examining the impact on nutrient recycling potential. Studies on HTL of Chlorella vulgaris, Scenedesmus dimorphus, and Spirulina platensis found significantly higher nutrient levels in HTL process water than in standard growth media, 3N-BBM +V (Table 9.4) [86].

Table 9.4

Nutrient Content of HTL Process Water (Spirulina, 300°C) Compared to Standard Growth Media 3N-BBM +V [86]

Nutrient	HTL Process Water	3N-BBM+V
	Conc./ppm	Conc./ppm
TOC	15123	–
Total N	8136	124
${\mathrm{NH}}_4^{+}$	6295	–
${\mathrm{PO}}_4^{3-}$	2159	135
K	1506	63
Acetate	7131	–
${\mathrm{NO}}_3^{-}$	194	547
Phenols	98	–

The high levels of ammonium, phosphate, nitrates, and potassium ions are promising for potential nutrient recovery. However, elevated phenol levels could be problematic due to their inhibitory effects on algal growth [86], even at moderate concentrations (100–200 ppm) [97]. Additionally, high protein content in the starting biomass corresponds to elevated levels of nitrogen heterocycles (eg, pyrolidinones, piperidines, pyrroles, indoles) [15,23], thought to be produced via the Maillard reaction between amino acids and glucose/fructose [37]. This could, however, be beneficial if certain strains of algae are capable of utilizing organics in mixotrophic growth.

Several groups have already reported successful algal cultivation using HTL process water [32,84,86,89,98]. Jena et al. found that dilution was necessary to bring nutrient levels down to levels comparable to those in commercial growth media, but found that a 10-fold dilution was still too strong for algal growth to occur [32]. Consequently, Biller et al. performed tests using process water from the HTL of Spirulina at 300°C, diluted by a factor of 50, 100, and 400, for the cultivation of Spirulina, with algal growth measured using chlorophyll absorbance. The finding that no growth occurred at either 50× or 100× dilution (although at this point, nutrient levels are brought down significantly below typical concentrations encountered in growth media) suggests that the concentrations of phenols, fatty acids, and metals such as Ni were still above inhibitory levels. Finally, a 400× dilution was found to be effective, with Spirulina exhibiting very similar growth rates in diluted process water and 3N-BBM +V [86]. Notably, no growth was seen in pure distilled water, confirming the role of the process water as a nutrient source. For the algae Chlorogloeopsis, however, appreciable growth was seen at dilutions as low as 100×, while the cell counts at 400× dilution were actually increased by a third compared to those seen for standard growth media, suggesting that the organic carbon could supplement mixotrophic growth in this case. Growth does not seem to be inhibited by low concentrations of ${\mathrm{PO}}_4^{3-}$ (for Chlorogloeopsis at 200× dilution). The study demonstrates the potential of HTL process water to be used as a growth medium, although the optimum dilution was different for each strain. Additionally, Pham et al. demonstrated that recycling into algal cultivation systems, coupled with granular activated carbon treatment, was an effective means of removing organic toxins and significantly reducing HTL waste water cytotoxicity for environmental release [98]. Another recent investigation has demonstrated that HTL process water from liquefaction of Spirulina at 350°C is capable of supplementing growth of a strain of algae cultivated in acid mine drainage. The algae demonstrated appreciable growth over 15 days, and was able to utilize 75% of the phosphate provided by the HTL process water (1:100 dilution in synthetic acid mine drainage) [84].

These results suggest that there is significant potential to vastly reduce the amount of additional nutrients required for algal growth by using recycled HTL process water as a growth supplement. If the algal strain at the site is found to exhibit mixotrophic growth, water-soluble organics in the HTL aqueous phase could be incorporated (improving carbon efficiency and biomass yields). Alternatively, water-soluble organics could be extracted prior to process water recirculation and sold as high-value commodity chemicals.