Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1 Hydrothermal Upgrading of Algae Paste: Inorganics and Recycling 2 Potential in the Aqueous Phase 3 Bhavish Patel, Miao Guo, Chinglih Chong, Syazwani Hj Mat Sarudin, Klaus Hellgardt* 4 Imperial College London, Department of Chemical Engineering, Exhibition Road, South Kensington, London SW7 2AZ, UK. 5 *Corresponding Author – Email: [email protected] ; Tel: +44 (0)20 7594 5577 6 Abstract 7 Hydrothermal Liquefaction (HTL) for algal biomass conversion is a promising technology capable of 8 producing high yields of biocrude as well as partitioning even higher quantity of nutrients in the aqueous 9 phase. To assess the feasibility of utilising the aqueous phase, HTL of Nannochloropsis sp. was carried 10 out in the temperature range of 275 to 350°C Residence Times (RT) ranging between 5 and 60 minutes 11 3− 2− − + + The effect of reaction conditions on the NO− 3 , PO4 , SO4 , Cl , Na , and K ions as well as Chemical 12 Oxygen Demand (COD) and pH was investigated with view of recycling the aqueous phase for either 13 cultivation or energy generation via Anaerobic Digestion (AD), quantified via Lifecycle Assessment 14 (LCA). It addition to substantial nutrient partitioning at short RT, an increase in alkalinity to almost pH 15 10 and decrease in COD at longer RT was observed. The LCA investigation found reaction conditions 16 of 275°C/30 min and 350°C/10 min to be most suitable for nutrient and energy recovery but both 17 processing routes offer environmental benefit at all reaction conditions, however recycling for 18 cultivation has marginally better environmental credentials compared to AD. 19 20 Keywords: Hydrothermal Liquefaction (HTL), nutrient recycling, biocrude, LCA, Anaerobic Digestion 21 (AD), microalgae 22 23 24 25 26 1 27 Graphical Abstract 28 Algae Cultivation Algal Biorefinery Use phase Biofuel Sunlight Chemicals Harvesting Water CO2 Dewatering Algae Biomass Hydrothermal Liquefaction Land Biocrude Nutrient or Energy Energy Aqueous Phase Nutrients Anaerobic Digestion Algae Growth Bioenergy Nutrient Recycling 29 30 31 32 33 34 35 36 37 38 39 40 41 42 2 43 1.0 Introduction 44 It is well established that the nutrients required for algal cultivation incur significant costs and results 45 in significant environment burdens owing to the energy and resources employed during their 46 synthesis/production (Clarens et al. 2010). Nitrogen and Phosphorus based salts in particular are 47 required in large quantities to support algal growth (Clarens et al. 2010, Johnson et al. 2013 & Patel et 48 al. 2012), and when implemented at scale, these inorganics may well become a limiting factor, 49 especially considering the impending peak Phosphorus theory suggesting a shortage in coming decades 50 with no real substitute possible (Beardsley, 2011). Even if sea water were used for commercial 51 cultivation, the replenishment of nutrients, such as nitrogen-based ammonium salts, phosphorus and 52 sulphates would still be necessary (Chebil and Yamasaki, 1998). Demirbas (2010) estimated that for 53 commercial production of microalgae, approximately 8 to 15 tons of fertilizers are required per hectare 54 per year. The production of fertilizers is energy intensive and requires substantial resource inputs, which 55 is one of the bottlenecks for achieving a high economic return for mass cultivation of algae to date 56 (Clarens et al. 2009). Consequently, the preparation of growth medium also contributes considerably 57 to the high biofuel production cost (Molina Grima et al, 2003), particularly where a saline growth 58 medium is needed. Most of these nutrients are absorbed by algal cells to synthesise and maintain their 59 chemical composition as well as survive and therefore nutrient reclamation would involve direct 60 removal/extraction from algal biomass, ideally without additional treatment or affecting the product 61 pool. 62 Amongst various processing technologies for biomass transformation to fuel/chemicals, treatment of 63 wet algae paste under elevated temperature and pressure via Hydrothermal Liquefaction (HTL) is 64 considered to be a promising route, especially since HTL also makes the recovery and recycling of 65 inorganic nutrients possible (Patel et al. 2015 & Peterson et al. 2008). Nutrient recycling is especially 66 important because not all the input nutrients are consumed completely by algae and some end up in the 67 aqueous phase after HTL. Furthermore, during hydrothermal treatment the inorganics bound to the 68 biomass can be retrieved as it is not desirable to have these in the oil phase, thus alleviating special 69 treatment. Therefore, recycling the nutrient-rich and carbon-containing aqueous phase after HTL could 70 be a solution via nutrient reclamation for algal cultivation, which would help to reduce the input of 3 71 fertilizers; another potential use is aqueous stream recycling for energy generation via Anaerobic 72 Digestion (AD). 73 The concept of using HTL produced water for growth has already been demonstrated by Biller et al. 74 (2012) who recycled HTL (at 300 and 350°C) aqueous phase for cultivation of four algae species, 75 namely Chlorella vulgaris, Scenedesmus dimorphus, Spirulina platensis and Chlorogloeopsis fritschii. 76 This study indicated the feasibility of recycling the HTL aqueous phase based on the high concentration 77 of a range of nutrients required for algae growth, this has been confirmed by other studies which showed 78 successfully cultivated algae using various concentrations of HTL aqueous phase (Alba et al. 2013, 79 Biller et al. 2012 and Nelson et al. 2013). However, the effect of Residence Time (RT) on inorganics 80 concentration in the aqueous phase is unknown. Particularly, recent studies (Faeth et al. 2014, Patel & 81 Hellgardt, 2015 and Patel & Hellgardt, 2013) have suggested that HTL can be carried out at shorter RT 82 and as a result, it is necessary to observe the partitioning of inorganics with respect to RT and to evaluate 83 the feasibility to re-use the aqueous phase obtained at these reaction conditions. An alternative to 84 nutrient reclamation is to digest the carbon-containing aqueous phase and recover energy via AD. 85 However, deciding the most desirable method for HTL aqueous phase recovery should take into account 86 economic and environmental variables. To elucidate the knowledge gap, this study focuses on the 87 environmental aspects, where the holistic environmental impacts of two alternatives for aqueous phase 88 treatment are compared using an LCA approach. Economic evaluation will be explored in future 89 research. 90 This manuscript investigates the concentration of recoverable nutrients and the potential to re- 91 use/recycle the aqueous phase. Experiments were carried out using a batch reactor to investigate the 92 effects of processing temperatures (275, 300, 325 and 350°C) and residence time (5, 10, 15, 20, 30, 45 93 and 60 minutes) on the concentration of sodium, potassium, sulphates, phosphates, nitrates and trace 94 metals in the aqueous phase. In addition, the experimental data is fed into an LCA model to investigate 95 the environmental impacts for aqueous phase utilisation as cultivation medium and AD feedstock. To 96 our best knowledge, the analysis of inorganics with respect to RT of HTL reaction and subsequent 4 97 environmental assessment of recovered aqueous phase has not been addressed in any publically 98 available literatures. 99 2.0 Experimental 100 Freeze dried Nannochloropsis sp. algae was obtained from the University of Almeria. All chemicals 101 used for standards, sample preparation and extraction protocol were obtained from Sigma Aldrich Ltd. 102 (unless stated otherwise) and used as purchased (purity >98%). The batch reactors were constructed in- 103 house using Stainless Steel (316L) tubes and cap endings on either side, purchased from Swagelok®. 104 2.1 Reaction 105 The hydrothermal liquefaction of the microalgae was carried out in ½ inch outer diameter 316L stainless 106 steel batch reactors of volume 6 cm3 in an isothermal recirculating oven. To expose the reactors to 107 hydrothermal conditions and condition them, all fabricated reactors were filled with De-Ionised Water 108 (DIW) and placed in the oven for 3 hours at 380°C. In a typical run, a fresh stock supply of 10 wt.-% 109 algae paste was made and 3 gm of this stock solution was added to the reactor and the reactor sealed. 110 It was then placed in the oven at required temperature (275, 300, 325, 350°C) and removed at designated 111 Residence Time (RT) 5, 10, 15, 20, 30, 45, 60 min). The reactor was immediately quenched in ice and 112 left overnight to equilibrate. The mass of reactor (+ contents) before and after the reaction was measured 113 to ensure no leakage occurred during the reaction. 114 2.2 Product Extraction 115 The quenched batch reactor was slowly opened to release any produced gas and rinsed with 116 Dichloromethane (DCM) to extract the products. A spatula was used to remove char stuck to the reactor 117 walls. The product was left to settle for 2 hours to allow the organic rich DCM and aqueous phase to 118 separate. A 2ml aliquot of the top aqueous phase was taken for analysis and filtered through a 0.22µm 119 membrane filter. The remaining aqueous phase was stored in a fridge. After allowing the filtrate to 120 separate in a separating funnel, the lower DCM solubilised biocrude phase was filtered, separated and 121 run through anhydrous MgSO4 column to remove residual water. Moisture free biocrude was obtained 5 122 by evaporating DCM over a steady stream of Nitrogen at room temperature for 5 hours. The pH of the 123 extracted aqueous phase was also measured. 124 2.3 Feedstock Inorganic Composition 125 The initial inorganic (trace metals) composition was determined as follows. A sample of 0.15-0.30 gm 126 freeze dried algae was added to a 60ml Teflon vessel (Savillex), with 2ml of concentrated Nitric Acid 127 at 70°C for 4 hours followed by the addition of 5ml of concentrated Hydrochloric Acid and left to 128 evaporate for 3 hours. The dried sample was then subjected to addition of 1ml 70% Perchloric Acid at 129 100°C and left for 2 hours in an open vessel. The vessel was then tightly closed and kept at 130°C for 130 48 hours. After cooling adding 0.2ml 40% Hydrofluoric Acid the vessel was placed in an oven at 130°C 131 for 4 hours. The samples were then dried at 150°C prior to preparation for Inductively Coupled Plasma 132 Atomic Emission Spectroscopy (ICP-AES) analysis using 7% Nitric Acid. 133 2.4 Analysis 134 2.41 Feedstock Inorganic Composition 135 ICP-AES analysis was conducted on a Thermo Scientific iCap Duo 6500. The calibration samples were 136 prepared from single element stock solution in 5-10% Nitric Acid. Au and Rh at 5 ppm. was added 137 inline as internal standard to account for matrix effects. 138 2.42 Ash Content 139 The ash content of the initial algal biomass was determined according to the National Renewable Energy 140 Laboratories Laboratory Analytical Procedure (Sluiter et al. 2005). The algae was treated in a furnace 141 at 105°C for 12 min after which the temperature was increased at 10°C/min to 250°C and held for 30 142 min. A final temperature of 550°C and hold time of 180 min was applied at a ramp pate of 20°C/min. 143 Finally, after allowing the furnace to cool to 105°C, samples were removed and measured. 144 2.43 Trace Metal Composition 6 145 The concentrations of metal ions; sodium (Na+), potassium (K+), and nickel (Ni2+), were determined 146 using a ICP-OES (Perkin-Elmer Optima 4300 DV). Due to the detection limit of ICP-OES 147 (concentrations up to 30 ppm), the aqueous phase samples were diluted to give 0.5 % of its original 148 concentration (dilution factor of 200). Commercially available standard solutions were analysed to 149 deriving calibration curves. 150 2.44 Inorganic Salts Composition 151 The concentrations of anions nutrients, namely nitrate (NO3-), sulphate (SO42-) and phosphate (PO43-), 152 contained in the aqueous phase were determined using a 882 Compact IC Plus Conductivity Detector 153 Ion Exchange Chromatography (ICP - IEC) unit. A dilution factor of 40 was used to stay within the 154 detection range. 155 2.45 Chemical Oxygen Demand (COD) 156 The COD test is used to determine the amount of organics present in the aqueous phase which involves 157 two steps - oxidation of the samples using potassium dichromate (K2Cr2O7) digestion solution and COD 158 determination using Ultraviolet – visible (UV-VIS) spectrophotometer. The digestion solution was 159 prepared by adding 10.2 gm of dried (103°C for 2 hours) K2Cr2O7, to 167 ml of concentrated sulphuric 160 acid (H2SO4) and 33.3 gm of mercury sulphate (HgSO4) to 500 ml of distilled water. This digestion 161 solution was then diluted by adding distilled water to make up a 1000 ml solution. 0.6 ml of this solution 162 was added to 1 ml of aqueous sample in a COD tube which was followed by the addition of 0.4 ml of 163 concentrated H2SO4. The tube was capped and inverted several times to ensure a well-mixed solution. 164 The tubes were then heated at 150°C for 2 hours and left to cool overnight. The COD concentration was 165 determined by UV-VIS measurement where potassium hydrogen phthalate (KHP, C8H5KO4) was used 166 for calibration. 167 3.0 LCA methodology 168 Generally, two types of LCA have been distinguished: Attributional LCA (ALCA) and Consequential 169 LCA (CLCA) approaches. The former is focused on describing the environmentally relevant flows to 7 170 and from a life cycle and its subsystems, while the latter aims to study how environmentally relevant 171 flows change in response to possible decisions. In this study, an ALCA approach was applied to 172 examine two potential options for utilizing aqueous phase – direct nutrient recycling for algae growth 173 vs. energy recovery from AD treatment. As illustrated in Figure 1, nutrient recovery and AD units were 174 included in this LCA comparison study along with its appropriate product. The functional unit was 175 defined as ‘treating aqueous phase generated from production of per kg biocrude oil at biorefinery gate’. 176 A ‘system expansion’ approach was applied to demonstrate the potential environmental benefits of 177 nutrient or energy recovery. The surplus electrical and thermal energy generated from combustion of 178 biogas at AD unit was modelled as net co-products export after accounting for the energy requirement 179 for AD operation and assumed to substitute an equivalent amount of electricity and heat generation 180 from UK national grid. The ‘functional equivalent’ quantity of commonly applied inorganic N or P 181 fertilizers in the UK was allocated as an ‘avoided burden’ to nutrient recovery unit. Life cycle impact 182 assessment (LCIA) methodologies can be categorised as midpoint and endpoint-oriented approaches 183 which are also termed as ‘problem-oriented’ and ‘damage approach’ respectively. The former is chosen 184 along with environmental mechanisms between the life cycle inventory (LCI) results and endpoints 185 (ISO, 2000) and the latter is defined at the level of protection area (Finnveden et al., 2009). A midpoint 186 approach developed by the Centre of Environmental Science (CML) of Leiden University - CML 2 187 baseline 2000 - was applied as characterisation method in this study at LCIA stage where the evaluation 188 focused on six impact categories - abiotic depletion, global warming potential (GWP100), acidification, 189 eutrophication, ozone depletion (ODP) and photochemical oxidation (POCP). The LCA model was 190 implemented in Simapro 7.3 (PRe Consultants). 8 Electricity Anaerobic Digestion HTL Aqueous Phase Algal Biorefinery Heat N Nutrient Nutrient Recovery P Nutrient 191 192 193 194 Figure 1 - LCA system boundary defined for AD energy recovery and nutrient recycling from HTL aqueous phase 195 It should be noted that the gas production was ignored in this instance as it does not form part of the 196 study due to its minute volume and composition (mostly CO2). 197 4.0 Results 198 4.1 Initial Feedstock Composition 199 Table 1 shows the trace metal concentration present in the initial algal feedstock. Metals of interest 200 (underlined) were chosen for analysis in the processed aqueous fraction after HTL. The ash content of 201 the algae was found to be 11 wt.-%. The biocrude yields can be found in SI-3. 202 Table 1 –Concentration of metals present in untreated Nannochloropsis sp. biomass feedstock Metal Al Ba Ca Cr Cu Fe K Mg Mn Content (ppm) 3.7 5.8 3446.5 0.8 21.0 212.6 11608.2 3128.2 69.0 9 Mo N Na Ni P S Sr Zn < 7.10 72000 45289.1 < 3.55 10321.2 6024.3 43.2 32.7 203 204 205 206 4.2 Aqueous Phase: 207 4.21 Nitrates 208 Nitrates are derived from the nitrogen-containing substrates, mostly proteins present in the alga cell. 209 Ideally, it would be beneficial to reclaim nitrates in the aqueous phase so that produced biocrude 210 contains low nitrogen and also facilitates the reclamation of nitrogen rich aqueous phase. For the 211 purpose of elucidation of nitrogen partitioning during HTL, it is assumed that all the nitrogen in the 212 cells is converted to nitrates, which gives the initial (t0) concentration of nitrates to be approximately 213 37,000 ppm based on N content given in Table 1. The calculation can be found in SI-2. 214 Figure 2 shows that the concentration of nitrates decreases with increasing residence time and 215 temperature. For reactions carried out at 325°C and 350°C, significant changes in nitrates concentration 216 were observed for RT up to 15 minutes. For reactions carried out at 350°C, the most significant 217 decreases in nitrate concentration, approximately 84% and 61%, occurred between 5 to 10 minutes and 218 10 to 15 minutes, respectively. This suggests that most of the protein degradation of the components in 219 Nannochloropsis sp. that yields nitrate was completed in less than 15 minutes at high reaction 220 temperatures. The short protein degradation time is consistent with that observed in the oil phase by 221 Eboibi et al. (2015) and Patel and Hellgardt (2013). At reaction temperatures 275°C and 300°C, the 222 concentrations of nitrate decreases gradually from a RT of 5 to 60 minutes. This indicates that the 223 degradation reactions that produce nitrates were not completed in a short time period as is the case for 224 higher temperature, elucidating that severe conditions result in less nitrate in the aqueous phase. 10 80 275°C 300°C 325°C 350°C Concentration (ppm) 70 60 50 40 30 20 10 0 5 15 20 30 45 60 Reaction Time (min) 225 226 10 Figure 2 – Concentration of Nitrates in HTL processed aqueous phase 227 228 However, nitrogen could also be present as nitrite, ammonium and (dissolved) ammonia from 229 degradation of proteins (Lourenço et al.2004). The inverse relationship between nitrate concentrations 230 and RT as well as temperatures is the opposite to that of ammonia concentration. It was reported that in 231 general the concentration of ammonia increases with temperature and residence time. This is due to the 232 more favourable formation of ammonia from organic nitrogen at higher liquefaction temperatures 233 (Valdez et al. 2012). 234 4.22 Phosphates 235 The concentration of phosphate at t0, which is 2040ppm, was calculated using the same method as 236 nitrate (SI-2) and all the phosphorus was assumed to be converted to phosphates. There is a rapid 237 decrease within the first 5 minutes of the reaction after which the concentration stabilises as seen from 238 Figure 3. A substantial amount, about 39 to 51% (794 to 1040ppm) of the phosphorus in 239 Nannochloropsis sp., has partitioned into the aqueous phase. The rest of phosphorus could have ended 240 up in the bio-crude and solids (Valdez et al. 2012). The values reported in previous studies (Alba et al. 241 2013, Biller et al. 2012 and Valdez et al. 2012) vary significantly, which can be explained by the varying 242 RT and different analysis method. Biller et al. (2012) found the phosphorus concentration in the aqueous 11 243 phase decreases with increasing temperature given a fixed residence time and suggested that higher 244 HTL temperature is not favourable for phosphorus retention. However, based on the temperature and 245 RT investigated in this study, the data suggest this is not the case and the major change occurs within 246 the first 5 minutes. Concentration (ppm) 2500 275°C 300°C 325°C 350°C 2000 1500 1000 500 0 0 10 15 20 30 45 60 Reaction Time (min) 247 248 5 Figure 3 – Concentration of phosphates in HTL processed aqueous phase 249 250 4.23 Sulphate 251 The sulphate equivalent of the sulphur present in Nannochloropsis sp. is 2000ppm (SI-2). From Figure 252 4, the sulphate concentrations remained largely unchanged with temperature and residence time. 253 Sulphate ions once formed associate with the metal ions present, such as sodium and potassium, and 254 partition into the aqueous phase (UNIDO, 1980). The thermal stability of sodium sulphate is high and 255 depending on the heating rate, the decomposition starts only at temperature much higher (around 850°C) 256 than that used for HTL. Since magnesium ions are commonly present in seawater, the presence of 257 magnesium sulphate in the aqueous phase is also possible. The onset decomposition temperature of 258 magnesium sulphate was reported to be around 540°C (Ebert et al, 1997). Therefore, the high thermal 259 stabilities of the sulphate ions could be the reason why the sulphate concentration in the aqueous phase 260 does not change with temperature. Secondly, some sulphur based compounds also partition to the 12 261 organic phase. Valdez et al. (2011) found significant production of dimethyl disulphide in the DCM 262 extracted organic phase after 60 min of HTL at 350°C. Concentration (ppm) 2500 2000 1500 1000 275°C 300°C 325°C 350°C 500 0 0 5 10 15 20 30 263 Reaction Time (min) 264 Figure 4 – Concentration of Sulphates in HTL processed aqueous phase 45 60 265 266 4.24 Trace Metals 267 Sodium 268 The initial concentration (t0) of sodium ions in the microalgae was determined to be 47290 ppm from 269 the overall inorganics analysis. The concentration of sodium ions in the aqueous phase varies slightly 270 with reaction temperature and RT as observed in Figure 5, where the concentrations remain around 271 5000 to 6000 ppm. About 38 to 59 % of the sodium originally present in the microalgae had partitioned 272 into the aqueous phase while the rest of the sodium potentially ends up in remaining char and to a lesser 273 extent the biocrude but not the gaseous phase. The high concentration of sodium matched the high 274 concentration of chloride in the aqueous phase because the microalgae, Nannochloropsis sp., are marine 275 algae for which its cultivation media contain high concentration of the salt sodium chloride. The ratio 276 between the concentrations of sodium and chloride ions in our aqueous samples was found to be around 277 1.6. This is similar to the ratio of these two ions in average seawater of about 1.7 as reported by Castro 278 and Huber (2003). 13 12000 275°C 300°C 325°C 350°C Concentration (ppm) 10000 8000 6000 4000 2000 0 0 10 15 20 30 45 60 Reaction Time (min) 279 280 5 Figure 5 – Sodium concentration in HTL aqueous phase 281 282 Potassium 283 The initial concentration of potassium in the microalgae was 11608 ppm. Figure 6 depicts that the 284 concentrations of potassium ions vary slightly with increasing temperature and RT. The concentrations 285 were observed to remain at about 1500 ppm. The percentage of potassium ions originally present in the 286 microalgae that partitioned into the aqueous phase was found to be in the range of 33 to 48 % and the 287 concentration of potassium in our aqueous samples is higher than that of the 8-fold enriched f/2 culture 288 medium. However, the aqueous phase will be diluted before being recycled to reduce the concentration 289 of phosphates. This dilution in turn reduces the concentration of potassium, lower than that of the f/2 290 culture medium. Therefore, addition of potassium salts will be necessary prior to recycling the aqueous 291 phase into the culture medium, albeit at a lower quantity. 14 4000 275°C 300°C 325°C 350°C Concentration (ppm) 3500 3000 2500 2000 1500 1000 500 0 0 5 15 20 30 45 60 Reaction Time (min) 292 293 10 Figure 6 – Potassium concentration in HTL aqueous phase 294 295 4.25 pH 296 The pH of the aqueous aliquot obtained from the reaction mixture was measured instead of the diluted 297 samples. This ensures that an exact representation of pH change with respect to RT at different reaction 298 temperatures is obtained. 299 The pHs of the aqueous samples in Figure 7 shows an increasing trend with the pH at 5 minutes being 300 pH 7 for all temperatures except 350°C. This indicates that the aqueous samples become more basic as 301 HTL reaction was left to proceed for a longer period of time. This corresponds to a decrease in 302 concentration of nitrate ions, which form acidic solutions when dissolved in water. The increase in pH 303 of the aqueous phase is also due to the degradation of amino acids via decarboxylation and deamination 304 (Quitain et al. 2006) as well as the favourable formation of ammonia at higher RT and temperature. 305 The cause for the increase of pH could also be due to the presence of alkali metal halides and sodium 306 sulphate in the aqueous phase. At HTL temperatures, the halides and sodium sulphate act as the salts of 307 strong bases and weak acids (Katritzky et al, 1992). The pH found in this investigation is different from 308 that observed by Elliot et al. (2013) where the aqueous phase was almost neutral with high COD. An 309 explanation for this is the post reaction extraction procedure used in this investigation where the 310 organics were partitioned using DCM. Nevertheless, Rocha et al. (2003) proposed that the pH can be 15 311 lowered by adding CO2 to the aqueous product for cultivation purposes. Since microalgae use CO2 for 312 photosynthetic activity, the addition of CO2 will be beneficial to the microalgae as source of carbon. 313 Other methods for lowering the pH include the use of buffers or the addition of inorganic acids. 11 10 pH 9 8 275°C 300°C 325°C 350°C 7 6 5 15 20 30 45 60 Reaction Time (min) 314 315 10 Figure 7 – pH of HTL processed aqueous phase 316 317 4.26 COD 318 The COD in Figure 8 shows a general decreasing trend with increase in RT at each reaction 319 temperatures. This indicates that the amount of organics in the aqueous phase decreases with increasing 320 RT elucidating a general tendency of organics to partition into other phases such as the gaseous or 321 organic phase. This trend could also be used to explain the decrease in colour intensity of the aqueous 322 product fraction, which might be due to chlorophyll-based compound. The lower COD amounts at 323 increasing severity is expected because as the reaction is allowed to proceed further, there are multiple 324 reactions such as degradation, decomposition, (re)polymerisation of the newly formed reactants in the 325 aqueous, organic and char phase which in turn create new species, sometimes in a different phase. For 326 instance, Valdez et al. (2012) discovered that the yield of light crude, defined as Hexane soluble, and 327 thus composed of lower molecular weight compounds was observed to increase at greater severity. 16 328 Based on experimental data and calculation of reaction pathway, it was suggested that contribution from 329 the aqueous phase aids towards the light crude formation, even though the overall biocrude yield 330 remained almost constant. 331 332 333 Figure 8 – Experimentally determined COD of HTL processed aqueous phase 334 335 5.0 LCA results 336 5.1 Life cycle inventory analysis (LCI) 337 LCA inventories representing a reaction matrix for the nutrient and energy recovery potential of the 338 aqueous phase from HTL were developed using laboratory experimental data (e.g. biocrude yields, 339 COD, nitrate and phosphate concentrations) derived at different reaction temperature and RT. Statistics 340 on the UK country-level N and P fertilizer composition (Table 2 (IFA, 2014) )was adopted in this study 341 to award the nutrient recovery unit with an environmental ‘credit’ for the avoided fossil fuel 342 consumption and emissions for the production of an equivalent amount of N and P inorganic fertilizers 343 for algae growth. The production data for the N/P inorganic fertilisers were derived from Eco-invent 344 V2.2. The use of HTL aqueous phase as a growth medium at various dilutions was tested by Alba et al. 17 345 (2013) although additional nutrients are always required. Whereas, the production of biogas from HTL 346 aqueous phase is in its infancy; the possibility of performing AD on the aqueous phase has been 347 demonstrated in recent research (Tommaso et al., 2015). In this study, the theoretical biogas production 348 of an AD system was calculated based on 1g COD destruction equivalent to 0.395L CH4 at 35°C and 349 one atmosphere (Speece, 1996) and the assumed biogas composition of 35% CO2 and 65% CH4 (Guo 350 et al., 2013). The electrical and thermal power generated from the AD unit was estimated to be of 1.2 351 kWh/m3 of biogas and an assumed net calorific value for biogas of 21.48 MJ/m3 with 50% of the biogas 352 being converted to heat (Guo et al., 2013). The AD unit modelled was based on our published study 353 (Guo et al., 2013). The AD system modelled includes pre-treatment and two-stage digestion. In the first 354 hydrolysis stage with operation temperature of 57 ºC, mainly hydrolysis and acidification occur which 355 can be coincident with some degree of acetogenesis. In the digestion stage with operation temperature 356 of 37 ºC, the methanogenesis process dominates with certain degree of acetogenesis. The processes for 357 UK average electricity and heat generation from fossil resources (Table 3) were allocated to AD units 358 as energy substitution credits. The data for the fossil-sourced energy generation f were derived from 359 Eco-invent V2.2. 360 Table 2 - UK country-specific fertilizer compositions a UK N fertilizer as N (% N fertiliser applied) Ammonium nitrate 47.24% Ammonium phosphate 2.01% Ammonium sulphate 4.02% Calcium ammonium nitrate 10.05% NK compound fertiliser b 2.01% NPK compound fertiliser b 8.04% c Nitrogen solutions 11.06% Urea 15.58% 361 362 363 364 365 366 367 a. b. c. d. e. f. UK P fertilizer as P2O5 (% of P fertiliser applied) Ammonium phosphate 28.35% NPK compound fertiliser d 22.16% e other NP fertiliser 0.52% Other P straight fertiliser 1.03% f P K compound fertiliser 7.22% Single superphosphate 3.09% Triple superphosphate 37.63% Data derived from EU statistics (IFA, 2014) Assumed as ammonium nitrate Assumed as urea ammonium nitrate Assumed as diammonium phosphate Assumed as monoammonim phosphate Assumed as phosphate rock 368 18 369 Table 3 - Energy sources for electrical energy and heat generation in the UK a % Coal Oil Natural gas Biofuels Waste Nuclear Hydropower Geothermal Solar PV Solar Thermal Wind Tide 370 a. Electricity 39.63% 0.84% 27.51% 3.55% 1.14% 19.35% 2.27% 0.00% 0.33% 0.00% 5.38% 0.00% Heat 17.27% 2.96% 75.34% 2.30% 2.13% -------- Data derived from International Energy Agency (IEA, 2014) 371 372 5.2 LCIA profiles - environmental benefits of nutrient and energy recovery 373 Figure 9 shows the results for HTL with a RT of 5 and 10 min (normalised comparisons (%)) to 374 demonstrate the comparison between scenarios whereas the LCIA scores for each individual impact 375 category for all scenarios are given in SI-1. Nutrient recycling and energy recovery units for aqueous 376 phase treatment (at all HTL reaction conditions) deliver environmental ‘savings’ across all impact 377 categories (below the line) due to the environmental benefits from exported surplus energy and fertilizer 378 substitution. Nutrient recycling represents a more environmentally favourable option than energy 379 recovery. Such comparison results are driven by the P fertilizer substitution profiles (high PO43- 380 concentrations in aqueous phase), which contribute to over 90% of total environmental savings brought 381 by nutrient recycling. The environmental beneficial effects on abiotic depletion can be attributable to 382 the avoidance of fossil fuel inputs (natural gas, fossil oil and coal for P fertiliser manufacturing) due to 383 substitution of inorganic P fertilisers. In general, N, P fertiliser production are energy-intensive, not 384 only emitting greenhouse gases (GHG) CO2 and CH4 from fossil fuel combustion but also causing ODP 385 due to fossil fuel extraction, production and transport (e.g. CClBrF2 emission from natural gas 386 transportation, CClF3 released from crude oil production). Such ODP and GWP100 burdens can be 387 avoided by shifting from inorganic nutrient feeding to aqueous nutrient recycling for microalgae growth. 19 388 Besides, P fertiliser production processes (e.g. triple superphosphate, diammonium phosphate, single 389 superphosphate) demand phosphoric acid subsequently require sulphuric acid inputs (as essential inputs 390 to produce phosphoric acid) – these acid production processes evolve acidifying gases SOx, NOx and 391 cause aquatic eutrophication issues by releasing PO43- and P to water body. Thus nutrient recovery from 392 HTL aqueous phase offers a solution to avoid such environmental issues and brings substantial 393 environmental savings on acidification, eutrophication and POCP impact categories. 394 The HTL reaction matrix with various RT and temperature is investigated to identify the optimal 395 reaction to derive aqueous phase with maximised nutrient recovery efficiency. Figure 10 and Table SI- 396 1 present the environmental saving effects due to the environmental benefits from fertilizer substitution 397 brought by the nutrient recovery from aqueous phase. Via LCA comparison, aqueous phase acquired 398 from HTL reacted at 275 °C, 10 minutes or at 350 °C, 5 minutes represent superior options to other 399 reaction conditions, bringing higher environmental saving ‘credits’ across all impact categories (Figure 400 10 and SI- 1). ABD ACD ETP GWP ODP POCP -20 -40 -60 -80 -100 401 5 min 275 C - NR 275 C - ER 300 C - NR 300 C - ER 325 C - NR 325 C - ER 350 C - NR 350 C - ER 20 402 ABD ACD ETP GWP ODP POCP -20 -40 -60 -80 5 min -100 403 404 405 10 min 275 C - NR 275 C - ER 300 C - NR 300 C - ER 325 C - NR 325 C - ER 350 C - NR 350 C - ER Figure 9 - LCIA profiles of nutrient and energy recovery from HTL aqueous phase utilisation [ER = Energy Recovery; NR = Nutrient Recovery] (unit: 1kg biocrude; method: CML 2 baseline 2000). 406 21 300 C kg Sb equivalent 275 C 325 C 350 C -2.09E-04 -2.23E-04 -2.52E-04 -2.64E-04 -2.61E-04 -2.80E-04 -2.88E-04 -2.81E-04 -3.10E-04 -3.29E-04 -3.55E-04 -3.45E-04 -3.55E-04 -3.51E-04 -3.99E-04 -3.22E-04 -3.32E-04 -3.43E-04 -3.49E-04 -3.64E-04 -3.85E-04 -4.33E-04 ABD -5.01E-04 5 min 407 10 min 15 min kg CO2 equivalent 275 C -4.95E-04 20 min 300 C 30 min 45 min 325 C 350 C -3.18E-02 -3.35E-02 -3.80E-02 -3.99E-02 -4.41E-02 -3.91E-02 -4.22E-02 -4.23E-02 -4.78E-02 -5.03E-02 -5.24E-02 -5.47E-02 -5.31E-02 -5.33E-02 -4.84E-02 -4.99E-02 -5.14E-02 -5.24E-02 -5.47E-02 -5.85E-02 -6.14E-02 -6.50E-02 GWP100 GWP -7.61E-02 408 409 410 5 min 10 min 15 min -7.59E-02 20 min 30 min 45 min Figure 10 - Resource depletion and GWP100 profiles of nutrient recycling from aqueous phase (unit: 1kg biocrude; method: CML 2 baseline 2000). 411 412 6.0 Discussion 413 Several important equivalent features of the aqueous phase from HTL of Nannochloropsis sp. at 414 different RT were identified. The pH at longer RT (30 minutes onwards) was found to be higher than 415 that of the culture medium. Hence, it would require adjustment to bring it to the desired pH of 8.5, 416 potentially by addition of CO2, which has already been considered to be an environmentally favourable 417 and economically viable (based on carbon credits or taxation scheme) solution by using CO2 generated 418 from power plant waste streams (Rickman et al. 2013). The increase in pH reflects conversion of N to 22 419 ammonium, a key by-product of HTL. The phosphate and sulphate concentration do not follow a similar 420 trend resulting in substantial loss of these compounds once the reaction is initiated, as observed within 421 the first 5 minutes. The trace metals had a similar behaviour. The loss of these materials from the initial 422 concentration points towards their fractionation in the oil and significantly in the solid residue (char) 423 phase. For instance, calcium phosphate is known to form as solid during HTL (Elliot et al. 2013). 424 To put this into context, the phosphate concentration in the aqueous phase obtained from the HTL 425 experiment at 275°C and 20 minutes was used to calculate the dilution factor needed to produce f/2 426 medium (Table 4). In this case, the aqueous phase needs to be diluted by a factor of 26, reducing the 427 concentration (in ppm) of the other nutrient ions recoverable from the aqueous phase by the same factor. 428 The prospect of nutrient recycling remains enticing even though the nitrate, sulphate, sodium, and 429 chloride concentrations are small due to the full recovery of phosphates as a non-renewable resource 430 highly relying on mineral deposit is progressively depleting. By 2033 the worldwide demand will 431 outstrip supply as a consequence of global expanding population, which is accompanied by the global 432 food insecurity issues due to the dominancy of the P deposit in five nations (Mehta et al., 2014). 433 Table 4 – Dilution factor required for reuse of HTL processed aqueous phase as cultivation medium Components Concentration in culture medium (ppm) Phosphate 27.2 Concentration in aqueous phase before dilution (ppm) 700 Nitrate Sulphate Potassium Sodium Chloride 437.6 2701 390 10752 19345 36 1500 1400 5000 8000 Concentration in aqueous phase after dilution (ppm) 27.2 % recovered wrt. culture medium (after dilution) 100 1.39 57.7 53.1 200 307.69 0.32 2.1 13.6 1.86 1.59 434 435 Additionally, based on LCA analyses, it was found that either recycling for cultivation or processed 436 water for AD contributes to the environment positively. The added value or avoided burden exhibited 437 from aqueous phase utility contributes towards reducing the overall impacts arising from algal HTL. 438 The LCA profiles of HTL stage falls beyond the scope of current study but can be further explored in 23 439 future research. The statistical method based uncertainly and sensitivity analyses would be helpful to 440 understand the robustness of the LCA findings, which could be explored in future. It should be noted 441 that HTL carried out in this study presents an overall understanding of the inorganic concentration levels 442 at various reaction conditions. Recent studies (Patel and Hellgardt 2015, Faeth et al. 2014) have shown 443 that HTL at lower RT is also feasible and the mass fractioned into the aqueous phase can be as much as 444 60.-wt% of initial biomass content. Thus, complete characterisation of the aqueous phase from short 445 RT should be conducted and the processing options investigated to examine whether the added benefit 446 is practically possible and to what extent does the environmental benefit contribute to improving 447 environmental credentials for the overall process. 448 7.0 Conclusion 449 The recycling/re-use of the aqueous phase from the HTL of algal biomass offers an opportunity to 450 contribute towards the cost effectiveness and reduced environmental impact. Although environmental 451 benefits can be obtained for both investigated scenarios (HTL and AD), the outcome of integrating the 452 cultivation and biomass conversion process with the aforementioned is not known., The practicality of 453 using the aqueous phase for cultivation is a point of interest given the presence of some inhibitory 454 properties of the aqueous phase. The phosphate and potassium concentration are both severely above 455 that required in the growth medium and as such, dilution for cultivation will be necessary. The AD 456 scenario also appears to be appealing, although to a lesser extent given the lower environmental benefit 457 compared to cultivation for all impact categories. It should be noted that the cost of including an AD 458 unit far supersedes that of recycling for cultivations and thus there will certainly be an economic 459 incentive to prefer recycle for growth. Lastly, data from short RT is of greater importance given the 460 direction of current research as well as possibility to reclaim higher mass from short RT reactions. In 461 order to confidently decide and quantify the effect of integrating recycle/recovery processes of the 462 aqueous phase, a thorough study via inclusion of all product fraction is necessary to decipher whether 463 the trade-off between environmental and economic benefit exists and to what extent it adds value to 464 existing technique 24 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 References 1. Alba, Laura Garcia, Cristian Torri, Daniele Fabbri, Sascha RA Kersten, and Derk WF Wim Brilman. "Microalgae growth on the aqueous phase from hydrothermal liquefaction of the same microalgae." Chemical engineering journal 228 (2013): 214-223. 2. Beardsley, Timothy M. "Peak phosphorus." BioScience 61, no. 2 (2011): 91-91. 3. Biller, Patrick, Andrew B. Ross, S. C. Skill, A. Lea-Langton, B. Balasundaram, C. Hall, R. Riley, and C. A. Llewellyn. "Nutrient recycling of aqueous phase for microalgae cultivation from the hydrothermal liquefaction process." Algal Research 1, no. 1 (2012): 70-76. 4. Castro, P., and M. Huber. "Chemical and physical features of seawater and the world Ocean." Marine biology (2003): 48-71. 5. Chebil, Leila, and Shigehisa Yamasaki. "Improvement of a rotifer ecosystem culture to promote recycling marine microalga, nannochloropsis sp."Aquacultural engineering 17, no. 1 (1998): 1-10. 6. Clarens, Andres F., Eleazer P. Resurreccion, Mark A. White, and Lisa M. Colosi. "Environmental life cycle comparison of algae to other bioenergy feedstocks." Environmental science & technology 44, no. 5 (2010): 1813-1819. 7. Demirbas, Ayhan. "Use of algae as biofuel sources." Energy conversion and management 51, no. 12 (2010): 2738-2749. 8. Ebert, W. L., A. J. Bakel, and S. D. Rosine. Gas evolution during vitrification of sodium sulfate and silica. No. ANL/CTD/PP--86307. Argonne National Lab., IL (United States), 1997. 9. Eboibi, B. E., David M. Lewis, Peter J. Ashman, and Senthil Chinnasamy. "Influence of process conditions on pretreatment of microalgae for protein extraction and production of biocrude during hydrothermal liquefaction of pretreated Tetraselmis sp." RSC Advances 5, no. 26 (2015): 20193-20207. 10. Elliott, Douglas C., Todd R. Hart, Andrew J. Schmidt, Gary G. Neuenschwander, Leslie J. Rotness, Mariefel V. Olarte, Alan H. Zacher, Karl O. Albrecht, Richard T. Hallen, and Johnathan E. Holladay. "Process development for hydrothermal liquefaction of algae feedstocks in a continuous-flow reactor." Algal Research 2, no. 4 (2013): 445-454. 11. Faeth, Julia L., Peter J. Valdez, and Phillip E. Savage. "Fast hydrothermal liquefaction of Nannochloropsis sp. to produce biocrude." Energy & Fuels 27, no. 3 (2013): 1391-1398. 12. Finnveden, G., Hauschild, M.Z., Ekvall, T., Guinee, J., Heijungs, R., Hellweg, S., Koehler, A., Pennington, D., Suh, S. 2009. Recent developments in Life Cycle Assessment. Journal of Environmental Management, 91(1), 1-21 13. Grima, E. Molina, E-H. Belarbi, FG Acién Fernández, A. Robles Medina, and Yusuf Chisti. "Recovery of microalgal biomass and metabolites: process options and economics." Biotechnology advances 20, no. 7 (2003): 491-515. 14. Guo, M., D. C. Stuckey, and R. J. Murphy. "Is it possible to develop biopolymer production systems independent of fossil fuels? Case study in energy profiling of polyhydroxybutyratevalerate (PHBV)." Green Chemistry 15, no. 3 (2013): 706-717. 15. Guo, M., Stuckey, D.C., Murphy, R.J. 2013. Is it possible to develop biopolymer production systems independent of fossil fuels? Case study in energy profiling of polyhydroxybutyratevalerate (PHBV). Green Chemistry, 15(3), 706-717. 16. IEA. 2014. United Kingdom: Electricity and Heat for 2012 International Energy Agency. 17. IFA. 2014. IFA statistics, International Fertilizer Industry Association 25 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 18. ISO. 2000. ISO 14042 Environmental management -- Life cycle assessment--Life cycle impact assessment, (Ed.) I.O.f. Standardization, British Standards Institution, pp. 28. 19. Katritzky, Alan R., Steven M. Allin, and Michael Siskin. "Aquathermolysis: reactions of organic compounds with superheated water." Accounts of Chemical Research 29, no. 8 (1996): 399-406. 20. Lourenço, Sergio O., Elisabete Barbarino, Paris L. Lavín, Ursula M. Lanfer Marquez, and Elizabeth Aidar. "Distribution of intracellular nitrogen in marine microalgae: calculation of new nitrogen-to-protein conversion factors."European Journal of Phycology 39, no. 1 (2004): 17-32. 21. Mehta, C.M., Khunjar, W.O., Nguyen, V., Tait, S., Batstone, D.J. 2014. Technologies to Recover Nutrients from Waste Streams: A Critical Review. Critical Reviews in Environmental Science and Technology, 45(4), 385-427. 22. Nelson, Michael, Lian Zhu, Anne Thiel, Yan Wu, Mary Guan, Jeremy Minty, Henry Y. Wang, and Xiaoxia Nina Lin. "Microbial utilization of aqueous co-products from hydrothermal liquefaction of microalgae Nannochloropsis oculata." Bioresource technology 136 (2013): 522-528. 23. Patel, Bhavish, Miao Guo, Arash Izadpanah, Nilay Shah, and Klaus Hellgardt. "A review on hydrothermal pre-treatment technologies and environmental profiles of algal biomass processing." Bioresource technology 199 (2016): 288-299. 24. Patel, Bhavish, and Klaus Hellgardt. "Hydrothermal upgrading of algae paste: Application of 31 P‐NMR." Environmental Progress & Sustainable Energy 32, no. 4 (2013): 1002-1012. 25. Patel, Bhavish, and Klaus Hellgardt. "Hydrothermal upgrading of algae paste in a continuous flow reactor." Bioresource technology (2015). 26. Patel, Bhavish, Bojan Tamburic, Fessehaye W. Zemichael, Pongsathorn Dechatiwongse, and Klaus Hellgardt. "Algal biofuels: a credible prospective?."ISRN Renewable Energy 2012 (2012). 27. Patel, Bhavish, Pongsathorn Dechatiwongse, and Klaus Hellgardt. "Enzyme-Catalysed Processes in a Potential Algal Biorefinery." (2014). 28. Peterson, Andrew A., Frédéric Vogel, Russell P. Lachance, Morgan Fröling, Michael J. Antal Jr, and Jefferson W. Tester. "Thermochemical biofuel production in hydrothermal media: a review of sub-and supercritical water technologies." Energy & Environmental Science 1, no. 1 (2008): 32-65. 29. Pisarevsky, A. M., I. P. Polozova, and P. M. Hockridge. "Chemical oxygen demand." Russian journal of applied chemistry 78, no. 1 (2005): 101-107. 30. PRéConsultants, Simapro 7.0 Database Manual Methods library. 2004. 31. Rickman, M., Pellegrino, J., Hock, J., Shaw, S. and Freeman, B., 2013. Life-cycle and technoeconomic analysis of utility-connected algae systems.Algal Research, 2(1), pp.59-65. 32. Rocha, Jorge MS, Juan EC Garcia, and Marta HF Henriques. "Growth aspects of the marine microalga Nannochloropsis gaditana." Biomolecular engineering20, no. 4 (2003): 237-242. 33. Sluiter, A., R. Ruiz, C. Scarlata, J. Sluiter, and D. Templeton. "Determination of extractives in biomass." Laboratory Analytical Procedure (LAP) 1617 (2005). 34. Speece, R.E. 1996. Anaerobic Biotechnology for Industrial Wastewaters. Archae Press, Nashville, Tennessee, USA. 35. Speece, Richard E. "Anaerobic biotechnology for industrial wastewaters." InAnaerobic biotechnology for industrial wastewaters. 1996. 26 555 556 557 558 559 560 561 562 563 564 565 566 567 36. Tommaso, G., W.-T. Chen, P. Li, L. Schideman and Y. Zhang (2015). Chemical characterization and anaerobic biodegradability of hydrothermal liquefaction aqueous products from mixed-culture wastewater algae. Bioresource Technology , 178: 139-146. 37. United Nations Industrial Development Organization. Fertilizer Manual, United Nations, 1980 (http://pdf.usaid.gov/pdf_docs/pnaaj093.pdf) [ Accessed 26/06/2015, 00:15) 38. Valdez, Peter J., Jacob G. Dickinson, and Phillip E. Savage. "Characterization of product fractions from hydrothermal liquefaction of Nannochloropsis sp. and the influence of solvents." Energy & Fuels 25, no. 7 (2011): 3235-3243. 39. Valdez, Peter J., Michael C. Nelson, Henry Y. Wang, Xiaoxia Nina Lin, and Phillip E. Savage. "Hydrothermal liquefaction of Nannochloropsis sp.: Systematic study of process variables and analysis of the product fractions."biomass and bioenergy 46 (2012): 317-331. 27