Accepted Manuscript Biohythane production of post-hydrothermal liquefaction wastewater: A comparison of two-stage fermentation and catalytic hydrothermal gasification Buchun Si, Jamison Watson, Aersi Aierzhati, Libin Yang, Zhidan Liu, Yuanhui Zhang PII: DOI: Reference:
S0960-8524(18)31632-8 https://doi.org/10.1016/j.biortech.2018.11.095 BITE 20748
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
12 October 2018 25 November 2018 26 November 2018
Please cite this article as: Si, B., Watson, J., Aierzhati, A., Yang, L., Liu, Z., Zhang, Y., Biohythane production of post-hydrothermal liquefaction wastewater: A comparison of two-stage fermentation and catalytic hydrothermal gasification, Bioresource Technology (2018), doi: https://doi.org/10.1016/j.biortech.2018.11.095
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Biohythane production of post-hydrothermal liquefaction wastewater: A comparison of two-stage fermentation and catalytic hydrothermal gasification Buchun Si1,2, Jamison Watson2, Aersi Aierzhati2, Libin Yang3, Zhidan Liu1*, Yuanhui Zhang1,2*
1
Laboratory of Environment-Enhancing Energy (E2E), Key Laboratory of Agricultural Engineering in
Structure and Environment, Ministry of Agriculture, College of Water Resources and Civil Engineering, China
Agricultural University, Beijing 100083, China (E-mail:
[email protected])
2
Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign, Urbana,
IL, 61801, USA (E-mail:
[email protected])
3
State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and
Engineering, Tongji University, Shanghai 200092, China
*
Corresponding author. Fax:+86-10-6273-7329; Tel: +86-10-6273-7329. E-mail address:
[email protected] (Z. Liu);
*
Corresponding author. Fax:217-244-0323; Tel: 217-333-2693. E-mail address:
[email protected] (Y. Zhang)
1
Abstract Developing efficient methods to recover energy from post-hydrothermal liquefaction wastewater (PHW) is critical for scaling up hydrothermal liquefaction (HTL) technology. Here we evaluated two-stage fermentation (TF) and catalytic hydrothermal gasification (CHG) for biohythane production using PHW. A hydrogen yield of 29 mL∙g-1 COD and methane yield of 254 mL∙g-1 COD were achieved via TF. In comparison, a higher hydrogen yield (116 mL∙g-1 COD) and lower methane yield (65 mL∙g-1 COD) were achieved during CHG. Further, a techno-economic and sensitivity analysis was conducted. The capital cost and operating cost for TF varied with the different reactor systems. TF with high-rate reactors suggested its promising commercialized application as it had a lower minimum selling price (-0.71-2.59 USD per gallon of gasoline equivalent) compared with conventional fossil fuels under both the best and reference market conditions. Compared with TF, CHG was only likely to be profitable under the best case conditions. Keywords: Biohythane; Hydrothermal liquefaction; Post-hydrothermal liquefaction wastewater; Two-stage fermentation; Catalytic hydrothermal gasification
2
1 Introduction Hydrothermal liquefaction (HTL) can convert wet organic wastes into biocrude oil at temperatures between 280-370oC and pressures between 10-25 MPa (Toor et al., 2011). HTL has been reported as a promising conversion pathway for human feces. One study reported that HTL of human feces led to a biocrude oil yield of 34.44% and a heating value of 40.29 MJ∙ kg-1 (Lu et al., 2017). Furthermore, HTL is a relatively quick thermochemical process, as it can be finished within one hour, and it is also much more efficient than compost (over 15 days) (Anand and Apul, 2014) and anaerobic digestion (60 days) (Owamah et al., 2014). In addition, HTL avoids the safety risk caused by pathogens in human feces because of the high temperature adopted during the HTL process. However, high-strength post-hydrothermal liquefaction wastewater (PHW) (total organic carbon content reaching up to 3-80 g∙L-1) is produced as a by-product during HTL (Leng et al., 2018), which usually contains 35-40% of the carbon and 65-70% of the nitrogen in the feedstock (Yu et al., 2011). In addition, PHW contains a high concentration of hazardous and highly cytotoxic organics which would cause serious pollution if it is directly discharged into the environment (Pham et al., 2013; Si et al., 2018). Hence, developing efficient methods to recover energy from PHW is critical for scaling up HTL technology. Anaerobic fermentation is one of the most widely used conversion technologies that has been implemented worldwide (Zheng et al., 2014). In particular, utilizing
3
anaerobic digestion as a means of treating PHW while concomitantly producing methane has been extensively studied in recent years(Fernandez et al., 2018; Posmanik et al., 2017; Tommaso et al., 2015; Zheng et al., 2017; Zhou et al., 2015). However, anaerobic fermentation for methane production suffers from a long retention time (over 30 days) and low methane yield (123-169 mL∙g-1 chemical oxygen demand (COD) ), which is caused by the inhibition of toxic compounds in the PHW (Tommaso et al., 2015; Zheng et al., 2017; Zhou et al., 2015). In comparison to conventional one-stage fermentation for methane production, two-stage fermentation (TF) consists of separate hydrogen and methane production steps. In the first step, the complex substrates are hydrolyzed, leading to the production of organic acids and hydrogen. The produced organic acids are then converted into methane in the second step. The produced mixture of hydrogen and methane, which is referred to as hythane, has been acknowledged as one of the most important fuels for transitioning from a fossil fuel-based society to a terminal hydrogen-based society (Liu et al., 2013). TF not only can produce a cleaner biofuel, but also enhance the conversion efficiency (Schievano et al., 2014). Hydrogen production has proven to be able to degrade furfural and 5-hydroxymethyl furfural (5-HMF), which can be used as a detoxification step for methane production (Liu et al., 2015). Si et al., 2016 compared one-stage fermentation and TF using PHW from cornstalk. The detoxification of hydrogen production and an enhancement of energy recovery in TF were observed. Catalytic hydrothermal gasification (CHG) has also drawn increasing attention as 4
it offers advantages for treating high-moisture content feedstock (Sikarwar et al., 2017). CHG has been combined with HTL to improve the energy recovery and produce hythane from PHW, and the hydrogen and methane in the produced gas can reach up to 54.9% (vol.%) and 15.8% (vol.%), respectively (Cherad et al., 2016; Zhang et al., 2011). Recently, Watson et al., 2017 studied the CHG of PHW, and a hydrogen rich gas (56.3%, vol.%) was achieved. However, CHG operates at high temperatures (300-700oC) and incorporates transition metal catalysts, which leads to a high energy demand and large economic input (Cherad et al., 2016; Watson et al., 2017; Zhang et al., 2011). Previous studies have investigated the performance of biohythane production from PHW via TF and CHG individually, including PHW from human feces (Watson et al., 2017), cornstalk (Si et al., 2016), microalgae (Cherad et al., 2016) and municipality sludge (Zhang et al., 2011). However, there is a lack of information to compare these two methods and benchmark commercial applications. In this study, the effect of PHW concentration on TF, and the influence of temperature and retention time on CHG were studied. By doing that, a comparison of biohythane production from human feces PHW using TF and CHG, including gas production and techno-economic analyses, was conducted.
5
2 Methods and materials 2.1 HTL process and characteristics of PHW HTL of human feces was conducted at the University of Illinois at Urbana-Champaign. Human feces was collected from volunteers. The characteristics of the collected human feces is shown in Table 1. Stainless steel reactors (100 mL, Model 4593, Parr Instrument Co.) were used to perform HTL. HTL experiments were conducted at 280oC with a retention time of 60 min. The feedstock was placed directly into the reactor, and then the reactor was sealed and purged with nitrogen three times to replace the air in the reactor head-space. The initial pressure of the reactor was 100 psi, and the total solids content was 20%. The gas products from HTL reactions were collected using a gas bag. The HTL products were then separated using a glass fiber filter, and the PHW was defined as the water-soluble portion that passed through the filter. As shown in Table 1, the PHW had a COD concentration of 52606± 1577 mg∙ L-1, a total nitrogen concentration of 1160±28 mg∙ L-1, and an ammonia concentration of 592±88 mg∙ L-1. The GC-MS analysis of human feces PHW indicated that it had a similar distribution of organic compounds to that of the PHW from swine manure (Yang et al., 2018), which mostly consisted of acids and alcohols. Among these compounds, glycerol and acetic acid were the two main components, which had concentrations of 15880±1005 mg∙ L-1 and 7893±834 mg∙L-1, respectively.
6
2.2 TF process for biohythane production TF was conducted using 160 mL serum bottles at a temperature of 37oC using a water bath (American Scientific). At the beginning of hydrogen production, the pH was adjusted to 6.5 using NaHCO3 (Fisher Scientific). The inoculum was collected from the anaerobic digester at the Urbana Sanitary District (Urbana, Illinois, USA), and it had a total solid content of 2.7±0.1%. A 10 mL inoculum was added for hydrogen production. The inoculum was heat treated (100oC, 30min) by a hot plate to inhibit methanogenesis activity and harvest anaerobic spore-forming bacteria before use (Si et al., 2015). Hydrogen production was stopped when the gas production rate was less than 0.5% of the cumulative gas volume. The effluent of hydrogen production was used for methane production. A 20 mL inoculum without heat treatment was added to the bottles. The pH was adjusted to 6.5-7.0 via adding NaHCO3 (0.5g ∙g-1 COD) before methane production. Methane production was stopped when the gas production rate was less than 0.5% of the cumulative gas volume. The concentration of PHW was reported as one of main factors for its anaerobic fermentation, and a fermentation concentration higher than 4.452 g COD∙L-1 indicated the inhibition due to the increased toxicity (Zheng et al., 2017). Hence, the effect of the concentration of PHW on fermentation was investigated at four different fermentation concentrations in this study, including 10, 7, 4 and 1 g COD∙L-1. The gas volume was measured using a glass syringe, and the gas content was measured using a gas chromatograph (Shimadzu GC-780). Kinetic analysis of TF 7
was performed using the modified Gompertz model (1-1) (Tommaso et al., 2015):
Y = Ym exp[- exp(Rm × e Ym × (λ - t ) + 1)]
(1-1)
Where Y is the accumulative hydrogen/methane yield (mL∙g-1 COD); Ym is the maximum hydrogen/methane yield (mL∙g-1 COD); Rm is the maximum hydrogen/methane production rate (mL∙g-1 COD∙d-1); e is exp (1) =2.71828; λ is the lag phase (d); t is the fermentation time (d).
2.3 CHG process for biohythane production CHG reactions were carried out utilizing a batch reactor. The gasification reactor was 36 cm long, and had a total operating volume of 20 mL. The heating of the reactor was provided by a high temperature furnace (Thermo Scientific, Lindberg Blue M Muffle Furnace). A catalyst mixture of Raney Ni and Ru/AC was used, which has been reported to enhance hydrogen production (Watson et al., 2017). The Raney Ni to Ru/AC ratio was 0.5, and the catalyst to feedstock ratio was 0.1. The catalysts were purchased from Sigma-Aldrich and Fisher Scientific. Five mL of PHW was added into the reactor along with the catalyst. The reactor was purged three times with nitrogen to replace the headspace air. A reactor leakage test was conducted, and the reactor was then placed in the furnace. After the reaction, the reactor was cooled down to room temperature using a water bath. The produced gas was collected using a gas bag. The gas volume was measured using a syringe, and the gas composition was measured using a gas chromatograph (Shimadzu GC-780). The effect of temperature (350-500oC) 8
and retention time (15-90 min) on CHG was investigated, which are widely demonstrated to be the most important parameters during hydrothermal conversion (Cherad et al., 2016; Zhang et al., 2011).
2.4 Analytical methods Acids, ethanol, 5-HMF and glycerol in the liquid samples were quantified using a high performance liquid chromatograph (HPLC) (Shimadzu, Japan). The HPLC was equipped with a refractive index detector and an Aminex HPX-87H column (Bio-RAD, USA). The temperature of the column was set to 40oC. Five mM sulfuric acid was used as the mobile phase, and the flow rate was 0.5 mL∙min-1. The H2, CH4 and CO2 content (vol.%) were analyzed using a gas chromatograph (Shimadzu GC-780) which was equipped with a thermal conductivity detector. The column (Ohio Valley Specialty, ZW7547) was 5.5 m long and packed with silica gel. Nitrogen was used as the carrier gas. The organic compounds in the PHW were analyzed by a GC-MS (7890A, Agilent Technologies, USA). The procedure for GC-MS analysis was performed as previously described (Zheng et al., 2017). The data were interpreted by a Mass Spectral Database (NIST08) and W8N08 library (John Wiley & Sons, Inc.). The statistical significance was assessed using the SPSS 19.0 analytical software package (IBM SPSS Inc.). The Turkey test was used to assess the significance of differences between conditions at a p<0.05 probability level.
9
2.5 Techno-economic analyses Further quantifications regarding the technology and economic disparities between TF and CHG were conducted. Process modeling software (Aspen Plus V8) was utilized to build a process model based on the data collected in this study for each of the approaches. The experimental conditions which had the highest hydrogen yield were used for the technological and economic comparison of TF and CHG. These conditions were chosen based on the challenges associated with production, transportation and storage of hydrogen from external resources (Chaubey et al., 2013), and the fact that hydrogen is much cleaner than methane from an emissions standpoint. The HTL operational data was obtained from a DOE report (Jones et al., 2014). The model assumed PHW was the feedstock for the biohythane production process. The reaction conditions and biohythane production data of TF and CHG were deduced from the experimental results. TF using conventional reactors (continuous stirred tank reactor) and high-rate reactors (upflow anaerobic sludge blanket reactor) were considered, respectively. The operation of anaerobic high-rate reactor was based on a previous study (Si et al., 2016). A simplified overview of CHG and TF is shown in Fig. 1. A thermodynamic model of the non-random two-liquid model (NRTL) was used. CHG reactions were operated under 400oC and 200 bar using a yield reactor model. Implementation of a heat exchanger was integrated to recover the heat from the product stream (Fig. 1a). A shell and tube countercurrent heat exchanger with 83.5% efficiency was integrated to recover the heat from the product stream. CHG reactors with 10% 10
catalyst loading were simulated in the process. During TF (Fig. 1b, c), the pH of the influent and effluent (hydrogen fermentation) was adjusted by sodium salt. The produced hydrogen and methane were then mixed and purified. The ratio of hydrogen in biohythane would be adjusted to the suggested value (10–25% by volume) before it was offered as the fuel (Liu et al., 2013). Based on the simulation, the techno-economic analysis of three different scenarios, CHG, TF with conventional reactors and TF with high-rate reactors was conducted to estimate a reference point before process scale-up and commercialization. The total capital investment was calculated by Aspen Capital Cost Estimator V8.8. The yearly operating cost of the facility was performed with a service factor of 0.9. Fixed capital for equipment cost is the cost for process units, i.e. reactors, heaters, flashes, heat exchangers and pumps. The purification process was based on water scrubbing-regeneration, for which data was obtained from the literature (Sun et al., 2015). The credit of wastewater treatment was calculated based on the state-of-the-art treatment cost for an activated sludge-based modern municipal wastewater treatment plant (Li et al., 2014). The energy returns of these scenarios were determined by the ratio of energy input (heating, plumb, mixing and purification etc.) to energy output (biohythane). In particular, the energy input did not include the energy content of PHW. The minimum selling price (MSP) of biohythane was also calculated for each process. MSP was presented as USD per gallon of gasoline equivalent ($∙GGE-1) using energy densities of gasoline, and the calculation was based on a previous study (Jones et al., 2013). 11
3 Results and discussion 3.1 Biohythane production from PHW via TF TF of PHW was conducted with a fermentation concentration of 10, 7, 4 and 1 g COD∙L-1, respectively (Fig. 2). The hydrogen production finished within 12 days. The hydrogen content at 10 g COD∙L-1 and 7 g COD∙L-1 reached up to 25%. A hydrogen yield of 29.3 mL H2∙g-1 COD was achieved at a 7 g COD∙L-1 fermentation concentration, which was the highest value in the anaerobic fermentation test. Hydrogen production of the human feces PHW mainly came from the metabolic pathway of glycerol bioconversion (Sarma et al., 2016), with the highest hydrogen yield being 0.40 mol H2∙mol-1 glycerol. The concentration of glycerol decreased after hydrogen production, and the concentration of acetic acid, butyric acid and ethanol increased. The hydrogen yield of glycerol in this study was much lower than the theoretical value of glycerol (3 mol H2∙mol-1 glycerol) (Xia et al., 2016) and the highest value (2.44 mol H2∙mol-1 glycerol) achieved from batch experiments (Faber and Ferreira-Leitão, 2016). However, this value was still higher than several studies, which ranged from 0.13 to 0.39 mol H2∙mol-1 glycerol (Kumar et al., 2015; Poleto et al., 2016; Sittijunda and Reungsang, 2012). It is worth noting that PHW is more complicated than the substrates referred to in the above studies in which pure glycerol or crude glycerol was used. Inhibitors such as furfurals and phenols in PHW would lead to a decrease in the hydrogen yield as previously reported (Sharma and Melkania, 2017). In the second 12
stage, acids produced during hydrogen production were converted to methane. The methane production finished within 22 days, and the methane content in all groups was over 40%. The highest methane yield was achieved at 4 g COD∙L-1, which was 287.4 mL∙g-1 COD. The kinetic analysis based on the modified Gompertz model suggested there was inhibition in the hydrogen production of PHW (Table 2). The group at 10 g COD∙L-1 showed the longest lag phase, which reached up to 4.0 days. The inhibition may have resulted from the potential inhibitors in the PHW, including acids, furanics, phenolics and N-heterocyclic compounds. Wang et al., 2008 reported that the hydrogen yield and hydrogen production rate all tended to decrease with an increasing concentration of acids. Furanic and phenolic compounds have been reported to be inhibitors of microbial dark fermentation (Monlau et al., 2014). Although there is no research reporting the impact of N-heterocyclic compounds on hydrogen production so far, N-heterocyclic compounds are recognized as hazards for biological treatment (Pham et al., 2013; Yao et al., 2011). The decrease of the PHW concentration reduced the microbes’ adaption time and accelerated hydrogen production. However, increasing the organic loading from 1 to 7 g COD∙L-1 also significantly enhanced the hydrogen yield (p<0.05), and similar results were observed by a previous study (Si et al., 2015). The further increase of the fermented concentration (10 g COD∙L-1) resulted in a decreased hydrogen yield. This could be attributed to the inhibitors in PHW which lead to a metabolic shift from hydrogen-producing pathways to non-hydrogen-producing pathways (Si et al., 2016). 13
Hence, there is a balance point between inhibition and enhancement, and a maximum hydrogen production rate and maximum hydrogen yield was achieved at a concentration of 7 g COD∙L-1. A shorter lag phase was observed in the methane production step (<2.9 d) compared with that in the hydrogen production period, which may result from the potential detoxification effect during the hydrogen production step.
3.2 Biohythane production via CHG Fig. 3 depicts the influence of retention time (15, 30, 60, and 90 minutes) and temperature (350, 400, 450, and 500oC) on the hydrogen and methane production via CHG. As the retention time increased from 15 to 90 minutes, the hydrogen content initially increased and then decreased (Fig. 3a). It reached a maximum value at 30 minutes with a value of 38.6% and a minimum value of 24.2% at 90 minutes. The maximum hydrogen yield (116.2 ± 6.6 mL∙g-1 COD) was achieved at 60 min (Fig. 3a), which suggested a significant improvement compared to that under other retention times (p<0.05). The hydrogen content also initially increased and then decreased with an increase of the temperature (Fig. 3b). Lee and Ihm, 2010 also reported a similar trend for the gasification of molasses-fermented wastewater, in which they found that the hydrogen content first increased from 58.2% to 65.3% when the temperature increased from 625 to 685oC and then subsequently decreased to 62.2% at 685oC. This could be explained by the increased impact of the methanation reaction at higher reaction conditions which favors the production of methane at the expense of hydrogen 14
(Watson et al., 2018). Hydrogen production can thereby be maximized at a low temperature with a high retention time or a high temperature with a low retention time, respectively. The methane content increased from 10.2% to 32.1% with an increase of the retention time and temperature (Fig. 3c, d). This result was supported by a previous study which found that for the gasification of oily wastewater, as the temperature increased from 500 to 700oC, the methane content linearly increased from 19% to 29%, respectively (Zhiyong and Xiuyi, 2015). It can therefore be concluded that the methanation reaction begins to dominate at higher temperatures and retention times leading to the favorability of methane over hydrogen at higher reaction conditions. This resulted from the difference in the reaction enthalpy values between the methanation and water-gas shift reaction. Since the water gas shift reaction (-42.1 MJ/kg) favors lower temperatures than the methanation reaction (-221 kJ/mol), hydrogen is favored at lower temperatures in comparison to methane throughout the reaction severities in this study for supercritical water gasification (Watson et al., 2018).
3.3 Comparison of TF and CHG for biohythane production from PHW Compared with TF, the hydrogen yield in CHG was 3 times higher in this study. The comparison of anaerobic fermentation and CHG combined with previous studies is discussed in the Supplementary Information. The data confirmed the higher hydrogen yields (116-151 mL∙g-1 COD) and hydrogen content (38.0-71.1%) of CHG compared 15
with anaerobic fermentation. This indicates that CHG is a promising technology for hydrogen production using PHW. However, a lower methane yield (12.9-65.4 mL∙g-1 COD) and methane content (4.8-23.4%) was achieved in CHG than that of anaerobic fermentation. One of the most commonly-quoted advantages of anaerobic fermentation is that it operates at 37oC which is much milder than the conditions for CHG. However, anaerobic fermentation of PHW suffers from significant inhibition (Si et al., 2018), which resulted from the presence of toxic compounds in PHW (Tommaso et al., 2015). In addition, Posmanik et al., 2017 reported the presence of recalcitrants in PHW would lead to a lower energy recovery for anaerobic fermentation. Hence, high dilution rates (4.5-15) and long retention times (1-90 d) for anaerobic fermentation are usually adopted. Different methods were adopted to improve anaerobic fermentation of PHW, including increasing the dilution rate (Zhou et al., 2015), co-fermenting with swine manure (Fernandez et al., 2018) and glucose (Si et al., 2016), adding activated carbon, zeolite and polyurethane matrices (Zheng et al., 2017; Zhou et al., 2015), and using high-rate reactors which enrich the high concentration of microbes in the form of granules and biofilms(Si et al., 2016). Compared with TF, CHG is a much faster process which could finish in 30-60 min. In addition, biological inhibitors in PHW would not significantly affect the CHG reactions in the same way they would affect anaerobic digestion due to the elevated reaction temperature and pressure (Watson et al., 2018). Thus, neither dilution or detoxification of PHW was required for CHG processing. 16
The operational conditions varied significantly between TF and CHG. In order to offer a quantification reference for commercialized application, further techno-economic analyses for three scenarios: TF with conventional reactors, TF with high-rate reactors and CHG, were conducted. The technical assumptions utilized for the three scenarios are shown in the Supplementary Information. Fig. 4 shows the energy demand, generation and return in these three scenarios. The heater contributed to over 81% of the energy demand for all scenarios, reaching up to 91% in TF with the high-rate reactors. High dilution rates of PHW and long retention times during TF significantly increased the heating needed for the mass, thus increasing the total heating energy consumed. It seems that the higher temperature and pressure make CHG costly with regards to the heating energy. However, when a heat exchanger with 83.5% efficiency was used during CHG, CHG had a much lower energy demand than TF. The energy return for high-rate TF, conventional TF and CHG, were 111.4%, 110.2% and 80.9%, respectively. The results suggested that a positive energy balance for biohythane production using PHW was theoretically achievable by fermentation. Compared with conventional TF, the TF with high-rate reactors had a lower energy demand due to the savings in the agitator energy input. The total capital costs and operating costs for the three scenarios is presented in Fig. 5. The capital cost of TF with conventional reactors was the highest. Intuitively, anaerobic fermentation would have a lower equipment cost than hydrothermal gasification. However, our study showed that this was not applicable for PHW 17
conversion. The reactor volume of TF with conventional reactors significantly increased when compared with CHG, which contributed to a longer reaction time and additional water dilution. Hence, huge reactor investment for conventional TF was required (Fig. 5a). TF with high-rate reactors would reduce the reactor cost due to the efficient processing via enriching the high density of robust microbes. CHG had a much smaller reactor than that of TF. However, CHG had a higher capital cost compared with TF with high-rate reactors, due to the requirement of expensive resistant materials, heat exchangers and high-pressure pumps for reactors operating at supercritical conditions. The higher capital cost of conventional TF also resulted in a higher maintenance cost, leading to the operating cost of conventional TF being 2 times as much as both CHG and TF with high-rate reactors. CHG and TF with high-rate reactors had a similar operating cost at the simulated scale. TF with high-rate reactors had a higher heat cost and chemical cost than CHG due to the adjustment of pH and dilution of water needed. PHW contained a high concentration of organic acids which required an addition of alkali to ensure a stable pH. The back-mixing of the effluent for pH adjustment may be a possible method to reduce the addition of alkali (Zhou et al., 2013). However, that would need further evaluation which may lead to a decrease of hydrogen production (O-Thong et al., 2016) and inhibition caused by the accumulation of ammonia (Micolucci et al., 2014). The chemical cost for CHG was attributed to the incorporation of catalysts. A significant reduction of chemical cost for CHG may be expected when the catalysts can be recycled for reuse. 18
In this study, the factors which significantly contributed to the economic viability of the modeled scenarios were considered (Table 3), i.e. project lifetime, capacity factor, subsidies, discount rate, tax rate, operating conditions (dilution water and chemicals) and price of resources (electricity and natural gas) (Gerber Van Doren et al., 2017; Jones et al., 2014; Jung et al., 2013; Zhou et al., 2013). Based on the range of these parameters (worst and best case), a sensitivity analysis was conducted. The minimum selling price (MSP) of conventional TF, high-rate TF and CHG was compared with gasoline (Fig. 6). The MSP of gasoline was calculated based on the average price of gasoline in United States between 2013 and 2018 (Energy Information Administration, 2018). The MSP of conventional TF was found to be far from economically feasible due to its expensive capital and operating cost. Compared to the MSP of conventional TF, the reference MSP of CHG was much lower (5.96$∙ GGE-1). It was found that CHG was likely to be profitable under best-case conditions, which had a MSP of 3.59$∙GGE-1. TF with high-rate reactors had a lower MSP than conventional fossil fuels under both of the best and reference case conditions. In particular, a MSP of -0.71$∙GGE-1 was achieved at the best conditions which indicated its promising commercialized application. However, a MSP of 9.02$∙GGE-1, was obtained under worst conditions, which showed the economic feasibility of TF with high-rate reactors was largely dependent on market prices of resources (natural gas and electricity) and government policy support. In general, the sensitivity analysis suggested that the TF with high-rate reactors would be more suitable from an economic perspective. In 19
addition, there is space for further improvement of the economics for biohythane production. For example, the TF and CHG could be integrated with algae cultivation. The effluent from TF and CHG contained a high concentration of nitrogen and phosphorus which could be used as a substrate for algae cultivation. This has been proven by Yang et al., 2018, that over 82.7% of the ammonia and all of the phosphorus in fermented PHW can be utilized by algae. Onwudili et al., 2013 reported the potential for microalgae cultivation using the processed water from CHG. In addition, carbon dioxide from the upgrading step is relatively pure, which also could be used as a carbon source for algae and crop cultivation.
4 Conclusion TF and CHG were compared to produce biohythane from the PHW generated from human feces. A hydrogen yield of 29.3 mL∙g-1 COD for TF was achieved, and a methane yield of 254.3 mL∙g-1 COD was reached. Compared with TF, a higher hydrogen yield (116.2 mL∙g-1 COD) and hydrogen content (38.0%) was achieved for CHG. The techno-economic analysis based on the experimental data determined that TF with conventional reactors had a higher net energy return but higher cost than CHG. Further improvement of TF using anaerobic high-rate reactors has the potential of economically competing with petroleum products.
20
Appendix A. Supplementary data The metabolite changes during two-stage fermentation, summary of fermentation and hydrothermal gasification of PHW and technical assumptions for scenario of TFH, TFC and CHG can be found in the electronic supplementary file.
Acknowledgements This work was financially supported by National Key Research and Development Program of China (2016YFD0501402), the Bill & Melinda Gates Foundation (RTTC-C-R2-01-001), Beijing Youth Top-notch Talents Program (2015000026833ZK10) and International Postdoctoral Exchange Fellowship (20170086).
References 1.
Anand, C.K., Apul, D.S., 2014. Composting toilets as a sustainable alternative to urban sanitation- A review. Waste Manage., 34, 329-343.
2.
Chaubey, R., Sahu, S., James, O.O., Maity, S., 2013. A review on development of industrial processes and emerging techniques for production of hydrogen from renewable and sustainable sources. Renew. Sust. Energ. Rev., 23, 443-462.
3.
Chen, W., Zhang, Y., Zhang, J., Schideman, L., Yu, G., Zhang, P., Minarick, M., 2014. Co-liquefaction of swine manure and mixed-culture algal biomass from a wastewater treatment system to produce bio-crude oil . Appl. Energ., 209-216. 21
4.
Cherad, R., Onwudili, J.A., Biller, P., Williams, P.T., Ross, A.B., 2016. Hydrogen production from the catalytic supercritical water gasification of process water generated from hydrothermal liquefaction of microalgae. Fuel, 166, 24-28.
5.
Energy Information Administration, U.S., 2018. Gasoline and Diesel Fuel Update.
6.
Faber, M.D.O., Ferreira-Leitão, V.S., 2016. Optimization of biohydrogen yield produced by bacterial consortia using residual glycerin from biodiesel production. Bioresource Technol., 219, 365-370.
7.
Fernandez, S., Srinivas, K., Schmidt, A.J., Swita, M.S., Ahring, B.K., 2018. Anaerobic digestion of organic fraction from hydrothermal liquefied algae wastewater byproduct. Bioresource Technol., 247, 250-258.
8.
Gerber Van Doren, L., Posmanik, R., Bicalho, F.A., Tester, J.W., Sills, D.L., 2017. Prospects for energy recovery during hydrothermal and biological processing of waste biomass. Bioresource Technol., 225, 67-74.
9.
Jones, S., Meyer, P., Snowden-Swan, L., Padmaperuma, A., Tan, E., Dutta, A., Jacobson, J., Cafferty, K., 2013. Process Design and Economics for the Conversion of Lignocellulosic Biomass to Hydrocarbon Fuels. U.S. Department of Energy.
10. Jones, S., Zhu, Y., Anderson, D., Hallen, R., Elliott, D., Schmidt, A., Albrecht, K., Hart, T., Butcher, M., Drennan, C., Snowden-Swan, L., Davis, R., Kinchin, C., 2014. Process Design and Economics for the Conversion of Algal Biomass to 22
Hydrocarbons: Whole Algae Hydrothermal Liquefaction and Upgrading. U.S. Department of Energy. 11. Jung, K., Moon, C., Cho, S., Kim, S., Shin, H., Kim, D., 2013. Conversion of organic solid waste to hydrogen and methane by two-stage fermentation system with reuse of methane fermenter effluent as diluting water in hydrogen fermentation. Bioresource Technol., 139, 120-127. 12. Kumar, P., Sharma, R., Ray, S., Mehariya, S., Patel, S.K.S., Lee, J., Kalia, V.C., 2015. Dark fermentative bioconversion of glycerol to hydrogen by Bacillus thuringiensis. Bioresource Technol., 182, 383-388. 13. Lee, I., Ihm, S., 2010. Hydrogen Production by SCWG Treatment of Wastewater from Amino Acid Production Process. Ind. Eng. Chem. Res., 49, 10974-10980. 14. Leng, L., Li, J., Wen, Z., Zhou, W., 2018. Use of microalgae to recycle nutrients in aqueous phase derived from hydrothermal liquefaction process. Bioresource Technol., 256, 529-542. 15. Li, W., Yu, H., He, Z., 2014. Towards sustainable wastewater treatment by using microbial fuel cells-centered technologies. Energ. Environ. Sci., 7, 911-924. 16. Liu, Z., Zhang, C., Lu, Y., Wu, X., Wang, L., Wang, L., Han, B., Xing, X., 2013. States and challenges for high-value biohythane production from waste biomass by dark fermentation technology. Bioresource Technol., 135, 292-303. 17. Liu, Z., Zhang, C., Wang, L., He, J., Li, B., Zhang, Y., Xing, X., 2015. Effects of furan derivatives on biohydrogen fermentation from wet steam-exploded 23
cornstalk and its microbial community. Bioresource Technol., 175, 152-159. 18. Lu, J., Zhang, J., Zhu, Z., Zhang, Y., Zhao, Y., Li, R., Watson, J., Li, B., Liu, Z., 2017. Simultaneous production of biocrude oil and recovery of nutrients and metals from human feces via hydrothermal liquefaction. Energ. Convers. Manage., 134, 340-346. 19. Micolucci, F., Gottardo, M., Bolzonella, D., Pavan, P., 2014. Automatic process control for stable bio-hythane production in two-phase thermophilic anaerobic digestion of food waste. Int. J. Hydrogen Energ., 39, 17563-17572. 20. Monlau, F., Sambusiti, C., Barakat, A., Quéméneur, M., Trably, E., Steyer, J.P., Carrère, H., 2014. Do furanic and phenolic compounds of lignocellulosic and algae biomass hydrolyzate inhibit anaerobic mixed cultures? A comprehensive review. Biotechnol. Adv., 32, 934-951. 21. Onwudili, J.A., Lea-Langton, A.R., Ross, A.B., Williams, P.T., 2013. Catalytic hydrothermal gasification of algae for hydrogen production: Composition of reaction products and potential for nutrient recycling. Bioresource Technol., 127, 72-80. 22. O-Thong, S., Suksong, W., Promnuan, K., Thipmunee, M., Mamimin, C., Prasertsan, P., 2016. Two-stage thermophilic fermentation and mesophilic methanogenic process for biohythane production from palm oil mill effluent with methanogenic effluent recirculation for pH control. Int. J. Hydrogen Energ., 41, 21702-21712. 24
23. Owamah, H.I., Dahunsi, S.O., Oranusi, U.S., Alfa, M.I., 2014. Fertilizer and sanitary quality of digestate biofertilizer from the co-digestion of food waste and human excreta. Waste Manage., 34, 747-752. 24. Pham, M., Schideman, L., Scott, J., Rajagopalan, N., Plewa, M.J., 2013. Chemical and biological characterization of wastewater generated from hydrothermal liquefaction of Spirulina. Environ. Sci. Technol., 47, 2131-2138. 25. Poleto, L., Souza, P., Magrini, F.E., Beal, L.L., Rodrigues Torres, A.P., Paula De Sousa, M., Laurino, J.P., Paesi, S., 2016. Selection and identification of microorganisms present in the treatment of wastewater and activated sludge to produce biohydrogen from glycerol. Int. J. Hydrogen Energ., 41, 4374-4381. 26. Posmanik, R., Labatut, R.A., Kim, A.H., Usack, J.G., Tester, J.W., Angenent, L.T., 2017. Coupling hydrothermal liquefaction and anaerobic digestion for energy valorization from model biomass feedstocks. Bioresource Technol., 233, 134-143. 27. Sarma, S., Dubey, V.K., Moholkar, V.S., 2016. Kinetic and thermodynamic analysis (with statistical optimization) of hydrogen production from crude glycerol using Clostridium pasteurianum. Int. J. Hydrogen Energ., 41, 19972-19989. 28. Schievano, A., Tenca, A., Lonati, S., Manzini, E., Adani, F., 2014. Can two-stage instead of one-stage anaerobic digestion really increase energy recovery from biomass? Appl. Energ., 124, 335-342. 29. Sharma, P., Melkania, U., 2017. Impact of furan derivatives and phenolic 25
compounds on hydrogen production from organic fraction of municipal solid waste using co-culture of Enterobacter aerogenes
and E. coli. Bioresource
Technol., 239, 49-56. 30. Si, B., Li, J., Zhu, Z., Shen, M., Lu, J., Duan, N., Zhang, Y., Liao, Q., Huang, Y., Liu, Z., 2018. Inhibitors degradation and microbial response during continuous anaerobic conversion of hydrothermal liquefaction wastewater. Sci. Total Environ., 630, 1124-1132. 31. Si, B., Li, J., Zhu, Z., Zhang, Y., Lu, J., Shen, R., Zhang, C., Xing, X., Liu, Z., 2016. Continuous production of biohythane from hydrothermal liquefied cornstalk biomass via two-stage high-rate anaerobic reactors. Biotechnol. Biofuels, 9, 254-269. 32. Si, B., Liu, Z., Zhang, Y., Li, J., Xing, X., Li, B., Duan, N., Lu, H., 2015. Effect of reaction mode on biohydrogen production and its microbial diversity. Int. J. Hydrogen Energ., 40, 3191-3200. 33. Sikarwar, V.S., Zhao, M., Fennell, P.S., Shah, N., Anthony, E.J., 2017. Progress in biofuel production from gasification. Prog. Energ. Combust., 61, 189-248. 34. Sittijunda, S., Reungsang, A., 2012. Biohydrogen production from waste glycerol and sludge by anaerobic mixed cultures. Int. J. Hydrogen Energ., 37, 13789-13796. 35. Sun, Q., Li, H., Yan, J., Liu, L., Yu, Z., Yu, X., 2015. Selection of appropriate biogas upgrading technology-a review of biogas cleaning, upgrading and 26
utilisation. Renew. Sust. Energ. Rev., 51, 521-532. 36. Tommaso, G., Chen, W., Li, P., Schideman, L., Zhang, Y., 2015. Chemical characterization and anaerobic biodegradability of hydrothermal liquefaction aqueous products from mixed-culture wastewater algae. Bioresource Technol., 178, 139-146. 37. Toor, S.S., Rosendahl, L., Rudolf, A., 2011. Hydrothermal liquefaction of biomass: A review of subcritical water technologies. Energy, 36 2328-2342. 38. Wang, B., Wan, W., Wang, J., 2008. Inhibitory effect of ethanol, acetic acid, propionic acid and butyric acid on fermentative hydrogen production. Int. J. Hydrogen Energ., 33, 7013-7019. 39. Watson, J., Si, B., Li, H., Liu, Z., Zhang, Y., 2017. Influence of catalysts on hydrogen production from wastewater generated from the HTL of human feces via catalytic hydrothermal gasification. Int. J. Hydrogen Energ., 42, 20503-20511. 40. Watson, J., Zhang, Y., Si, B., Chen, W., de Souza, R., 2018. Gasification of biowaste: A critical review and outlooks. Renew. Sust. Energ. Rev., 83, 1-17. 41. Xia, A., Cheng, J., Murphy, J.D., 2016. Innovation in biological production and upgrading of methane and hydrogen for use as gaseous transport biofuel. Biotechnol. Adv., 34, 451-472. 42. Yang, L., Si, B., Tan, X., Chu, H., Zhou, X., Zhang, Y., Zhang, Y., Zhao, F., 2018. Integrated anaerobic digestion and algae cultivation for energy recovery and nutrient supply from post-hydrothermal liquefaction wastewater. Bioresource 27
Technol., 266, 349-356. 43. Yao, H., Ren, Y., Deng, X., Wei, C., 2011. Dual substrates biodegradation kinetics of m-cresol and pyridine by Lysinibacillus cresolivorans. J. Hazard. Mater., 186, 1136--1140. 44. Yu, G., Zhang, Y., Schideman, L., Funk, T., Wang, Z., 2011. Distributions of carbon and nitrogen in the products from hydrothermal liquefaction of low-lipid microalgae. Energ. Environ. Sci., 4, 4587-4595. 45. Zhang, L., Champagne, P., Charles Xu, C., 2011. Supercritical water gasification of an aqueous by-product from biomass hydrothermal liquefaction with novel Ru modified Ni catalysts. Bioresource Technol., 102, 8279-8287. 46. Zheng, M., Schideman, L.C., Tommaso, G., Chen, W., Zhou, Y., Nair, K., Qian, W., Zhang, Y., Wang, K., 2017. Anaerobic digestion of wastewater generated from the hydrothermal liquefaction of Spirulina:
Toxicity assessment and
minimization. Energ. Convers. Manage., 141, 420-428. 47. Zheng, Y., Zhao, J., Xu, F., Li, Y., 2014. Pretreatment of lignocellulosic biomass for enhanced biogas production. Prog. Energ. Combust., 42, 35-53. 48. Zhiyong, Y., Xiuyi, T., 2015. Hydrogen generation from oily wastewater via supercritical water gasification (SCWG). J. Ind. Eng. Chem., 23, 44-49. 49. Zhou, Y., Schideman, L., Yu, G., Zhang, Y., 2013. A synergistic combination of algal wastewater treatment and hydrothermal biofuel production maximized by nutrient and carbon recycling. Energ. Environ. Sci., 6, 3765-3779. 28
50. Zhou, Y., Schideman, L., Zheng, M., Martin-Ryals, A., Li, P., Tommaso, G., Zhang, Y., 2015. Anaerobic digestion of post-hydrothermal liquefaction wastewater for improved energy efficiency of hydrothermal bioenergy processes. Water Sci. Technol., 72, 2139-2147.
29
Figure captions: Fig. 1 Process diagrams: (a) catalytic hydrothermal gasification (CHG); (b) two-stage fermentation with conventional reactors(TFC); and (c) two-stage fermentation with high-rate reactors (TFH) for biohythane production using PHW. Fig. 2 Hydrogen production (a) and methane production (b) via two-stage fermentation (TF) using PHW. Fig. 3 The effect of retention time (a, c) and temperature (b, d) on the hydrogen yield, hydrogen content, methane yield and methane content production during catalytic hydrothermal gasification (CHG). Fig. 4 Power demands, overall balances and energy return for the three scenarios. Fig. 5 Total capital costs (a) and operating costs (b) for the three scenarios. Fig. 6 Minimum biohythane selling price (MSP) for the three scenarios, and for the three economic cases: reference, best and worst.
30
Table 1 Characteristics of the feedstock and PHW. Swine manure(Chen et al., 2014; Yang et al., 2018)
Human feces Feedstock a Crude Protein (%) Crude Fat (%) Ash (%) Carbohydrate b (%) Reaction condition PHW COD (mg∙L-1) Total Nitrogen (mg∙L-1) Ammonia (mg∙L-1) Formic acid(mg∙L-1) Acetic acid (mg∙L-1) Propionic acid (mg∙L-1) Butyric acid (mg∙L-1) Glycerol (mg∙L-1) Ethanol (mg∙L-1) 5-HMF (mg∙L-1) Acids c (%) Alcohols c % Cyclic Hydrocarbons c (%) Phenols c (%) N-heterocyclic c (%) Aldehydes c (%) Ketones c (%) Benezoic acid Derivatives c (%) a Based on dry weight; b Calculated based
35 24 16 25 280 oC, 60min
25 20 16 39 270 oC, 60min
52606± 1577 99400 1160±28 492 592±88 151 N.A.d N.A.d 7893±834 857 128±73 755 57±0.1 689 15880±1005 N.A.d d N.A. N.A.d 93±18 N.A.d 35.6 51.8 45.1 5.27 0.1 2.06 0.5 N.A.d 1.5 25.3 5.0 N.A.d 10.4 N.A.d d N.A. 4.0 on the difference between crude protein, crude fat and ash;
Based on the relative area from GC-MS; d N.A.=Not applicable
31
c
Table 2 The biogas yield (Ym), maximum production rate (Rm), lag phase (λ), and (R2) for two-stage fermentation (TF). COD Concentration
1 g∙L-1 4 g∙L-1 7 g∙L-1 10 g∙L-1
Hydrogen production Ym Rm λ (mL∙g-1 COD) (mL∙g-1 (d) COD-1∙d-1) 13.7 6.9 0.7 23.5 10.2 1.5 29.3 14.8 1.7 25.3 14.0 4.0
R2
0.992 0.984 0.997 0.993
32
Methane production Ym Rm λ (mL∙g-1 COD) (mL∙g-1 (d) COD-1∙d-1) 269.1 30.4 0.8 274.0 43.5 0.2 254.3 31.8 1.1 227.9 31.2 2.9
R2
0.882 0.987 0.991 0.994
Table 3 Parameters used for techno-economic analysis.
a
Project lifetime (year) Capacity factor (%)a Discount rate (%)a,b Tax rate (%)b Electricity price (USD∙kWh-1)a,b Natural gas price (USD∙kWh-1)a Subsidies for green fuel (USD∙L-1)a Dilution water reduction (%)c Chemicals reduction (%)d a
Reference 30 0.9 0.067 0.17 0.0556 0.01 0.2 100 0
Best 40 0.95 0.067 0.05 0.0556 0.01 0.7 100 50
Worst 20 0.85 0.15 0.35 0.2 0.2 0 0 0
Value from Gerber et al.(Gerber Van Doren et al., 2017); b Value from Jones et al.(Jones et al., 2014); c
Water reduction using municipal wastewater with low concentration organics (Zhou et al., 2013). d Back mixing of methane fermenter effluent reduce the alkali addition requirement(Jung et al., 2013).
Fig. 1 Process diagrams: (a) catalytic hydrothermal gasification (CHG); (b) two-stage fermentation with conventional reactors (TFC); and (c) two-stage fermentation with high-rate reactors (TFH) for biohythane production using PHW. Biohythane
Gas
a PHW Pump
Water Purification & adjustment Flash
Heater Heat Exchanger
CHG Reactors Hydrogen
Biohythane Methane
Chemicals
b
Purification & adjustment
PHW Pump
Mixer
Heater
Water 1st Phase Reactors
2nd Phase Reactors Biohythane
c
Hydrogen
Chemicals
Methane Water
PHW Pump
Mixer Heater 1st Phase Reactors
2nd Phase Reactors
Purification & adjustment
Fig. 2 Hydrogen production (a) and methane production (b) via two-stage fermentation (TF) using PHW.
Fig. 3. The effect of retention time (a, c) and temperature (b, d) on the hydrogen yield, hydrogen content, methane yield and methane content production during catalytic hydrothermal gasification (CHG).
Fig. 4 Power demands, overall balances and energy return for the three scenarios. (An energy return lower than 100% means that grid power is required, and an energy return higher than 100% means that excess power is produced and can be exported to the grid; CHG, catalytic hydrothermal gasification; TFH, two-stage fermentation with high-rate reactors, TFC, two-stage fermentation with conventional reactors).
Fig. 5 Total capital costs (a) and operating costs (b) for the three scenarios. (CHG, catalytic hydrothermal gasification; TFH, two-stage fermentation with high-rate reactors, TFC, two-stage fermentation with conventional reactors).
Fig.6 Minimum selling price (MSP) of biohythane for the three scenarios, and for the three economic cases: reference, best and worst. (CHG, catalytic hydrothermal gasification; TFH, two-stage fermentation with high-rate reactors; TFC, two-stage fermentation with conventional reactors; MSP presented as USD per gallon of gasoline equivalent).
Highlights
Biohythane production from post hydrothermal liquefaction aqueous was investigated Anaerobic fermentation and catalytic hydrothermal gasification were compared Techno-economic and sensitivity analyses based on experiments data were conducted Anaerobic fermentation showed a promising commercialized application
Graphic Abstract
Step1: Bench-scale tests
Step2: TEA analyses Hydrothermal gasification
Wastewater
Biohythane
Hydrothermal liquefaction Two-stage fermentation