Hydrothermal carbonization of sewage sludge: Multi- response optimization of hydrochar production and CO2- assisted gasi cation performance


 The harmful effects of improper sewage sludge (SS) treatment on the environment inspire the search for more benign sludge processing techniques such as hydrothermal carbonization (HTC); the abundant organic matter in SS is used for energy recovery. Herein, response surface methodology (RSM) was used to optimize the HTC-based preparation of SS hydrochar and its gasification performance. Specifically, the hydrochar yield, higher heating value (HHV), and gasification activity index were selected as optimization goals, while carbonization temperature (160–260°C), residence time (30–150 min), and acetic acid concentration (0–1.5 M) were selected as factors influencing the HTC process and CO2-assisted gasification performance. Carbonization temperature was the dominant parameter determining hydrochar yield, HHV, and gasification activity. The hydrochar yield (82.69%) and calorific value (7820.99 kJ kg−1) were maximized under comparatively mild conditions (160°C, 30 min, and 0.07 M acetic acid), whereas the gasification activity index (0.288 s−1) was maximized under harsher conditions (211.34°C, 88.16 min, and 1.58 M acetic acid). The obtained results help to guide the HTC of SS intended for gasification, thus promoting the development of this promising waste-to-energy technology, and may facilitate the design and further optimization of thermochemical SS conversion.


Introduction
According to recent estimates, the amount of domestic and industrial wastewater generated worldwide each year is of the order of billions of tons and is expected to increase because of population growth and living standard improvement (Mateo-Sagasta et al. 2015; Bora et al. 2020), which in turn will cause serious pollution (Meng et al. 2016). Compared with the traditional methods of wastewater SS disposal (e.g., land lling, incineration, and anaerobic digestion), which suffer from land shortage, secondary pollution, and long processing duration, the thermochemical conversion of SS by gasi cation offers the advantages of decreased pollutant emission and shorter processing duration, thus holding great promise (Zheng et al. 2020; Chen et al. 2020).
However, the gasi cation-based conversion of SS to valuable energy and fuels is hindered by its high moisture content. Currently, at least 50 million tons of SS with a moisture content of 80% are produced within the European Union annually (Kelessidis et al. 2012), and a similar value (40 million tons of SS with a moisture content of 80%) has been estimated for the United States and China (Venkatesan et al. 2015). Since most of the energy invested and released during the conventional heat treatment process is consumed to remove SS moisture, this has inspired the search for innovative and sustainable SS pre-drying technologies.
The main advantage of Hydrothermal carbonization (HTC) over other thermal conversion technologies, such as pyrolysis, gasi cation and incineration, is its ability to convert wet sludge to hydrochar with relatively high yields without preliminary dewatering and drying, which, consequently, requires less energy. Therefore, HTC has received much attention as a pretreatment technology allowing the effective conversion of wet biomass into solid fuel (hydrochar) (Yu et al. 2018;He et al. 2018), is a pressurized thermal conversion process that is achieved at a mild temperature (160-300°C) and self-generated pressure for various retention time (Sharma et al. 2020); it involves ve principal reactions, namely hydrolysis, dehydration, decarboxylation, polymerization, and aromatization (Zhao et al. 2014; Reza et al. 2014). The feasibility of employing HTC to prepare solid fuel from biomass waste has therefore been extensively investigated (Park et al. 2018; Peng et al. 2016). Park et al. showed that the HTC of algal biomass affords hydrochar with improved carbon content, carbon recovery, energy recovery, and C/O and C/H atomic ratios (Park et al. 2018). Some researchers have observed a signi cant enhancement in SS dewaterability after HTC Escala et al. 2012;Kim et al. 2014; Wang et al. 2015 spheres formed at all temperatures tested in the catalytic runs, no such spheres were produced in non-catalytic runs (Evcil et al. 2020). Given that the acids produced during hydrothermal biomass processing catalyze HTC (Stemann et al. 2013), we herein considered the in uence of the most abundant of these acids, namely acetic acid.
Numerous works show that HTC improves the fuel properties of sludge and provides raw materials more suitable for further thermochemical conversion (Park et  Therefore, in this study, we probed the gasi cation of SS hydrochar in an atmosphere of CO 2 . The recent years have witnessed a surge of interest in the optimization of HTC conditions using the response surface methodology (RSM). A response surface model describing the relationship between the given factors and the response values is constructed by using appropriate functions, and the optimal process parameters are obtained by analyzing these relationships and dealing with multivariable problems (Franceschini et al. 2008  ). These results indicate that RSM is a well-established and widely used technique for studying the joint interactions between independent (input) variables in the reaction process and is particularly useful for process optimization.
Despite the great potential of HTC and subsequent hydrochar gasi cation, information on statistically optimized conditions for realizing best-quality hydrochar for energy applications is still scarce. Herein, RSM is used to establish the optimal HTC conditions for maximizing hydrochar yield and HHV as well as to determine the HTC conditions best suited to obtain hydrochar with maximal gasi cation activity.

Materials
Dewatered SS with a moisture content of ~80 wt%, collected from a wastewater treatment plant in Jilin City, Jilin Province, China, was stored at 4 °C and used as the raw material for HTC. For characterization, SS was dried at 105 °C for 24 h, ground into ne powder, and sealed in a dry glass bottle for subsequent analysis. Table 1 lists the primary properties of raw SS. The high content volatiles were the main source of heat released during sludge thermochemical conversion, and the ash components were mainly SiO 2 , Al 2 O 3 , and Fe 2 O 3 .

HTC experiments
HTC experiments were carried out in a 0.5-L stirred batch reactor (stainless steel 316 L, GCF-type, Dalian Controlled Plant, China). In each experiment, the reactor was charged with a mixture of SS (20 g) and deionized water (200 mL). In experiments involving acid catalysis, SS (20 g) was mixed with acetic acid solutions (200 mL) of different concentrations. The reactor was sealed, purged with N 2 to remove residual air, heated to the predetermined temperature (160, 210, or 260 °C) using an electric heater, and held at this temperature for a certain time (30, 90, or 150 min). The reaction pressure (that of water alone at the respective temperature) ranged from 1.6 to 4.7 MPa. After the reaction, the slurry samples were collected and separated into ltrates and hydrochar by ltration. After 24-h drying at 60 °C, hydrochar was ground into ne powder and stored in an enclosed plastic pipe until analysis. Hydrochar samples were denoted as "HC-A-B-C," where A is the HTC temperature, B is the HTC residence time, and C is the acetic acid concentration. Two indexes (hydrochar mass yield (H y ) and HHV) were selected as dependent variables representing responses to HTC condition variation.
H y was estimated using Eq. (1), while HHV was estimated by bomb calorimetry (SDAC6000, Hunan Sundy Science and Technology Co., Ltd., China). The nitrogen, hydrogen, and carbon contents of hydrochar were determined by combustion at 950 °C employing an automatic elemental analyzer (EA3000, Euro Vector S.P.A., Italy).

CO 2 -assisted gasi cation experiments
Hydrochar gasi cation was carried out in a micro-uidized bed reaction analyzer coupled with a mass spectrometer (MFBRA-MS). The assembly mainly comprised a gas supply system, a gasi cation system, and an online gas monitoring and analysis system (Fig. 1).
Gasi cation was performed as follows. (1) The reactor was heated to 800 °C. (2) Hydrochar (10 mg) was placed into the feeding pipe. (3) The gasi cation agent (99.999 vol% CO 2 ) was supplied to the reactor at a ow rate of 0.5 L min −1 to stabilize the gas baseline. (4) The pulse was turned on through the solenoid valve, and the sample was instantly sprayed into the reaction area. The gas was analyzed using an online mass spectrometer. (5) Procedures (2)-(4) were repeated until all experiments were completed.
The carbon conversion of hydrochar gasi cation (X) was calculated as (2) where t is the reaction time (s), t 0 is the initial reaction time (s), t d is the end reaction time, ψ i is the volume fraction of component i in the produced gas (%), and q v is the ow rate (mol min −1 ).
To establish a relationship between HTC conditions and hydrochar gasi cation reactivity, we used RSM to optimize this reactivity and thus obtain optimal HTC conditions. The reaction index R 0.9 , employed to quantitatively characterize the overall gasi cation reactivity of hydrochar, was determined as where t X = 0.9 represents the gasi cation time (min) required for a carbon conversion of 0.9.

Experimental design for process optimization
RSM is a method of optimizing experimental conditions, which is suitable for solving the related problems of nonlinear data processing. By means of regression tting and response surface drawing, the predicted optimal response value and corresponding experimental conditions can be found out.
Experiments were designed using the Box-Behnken method, a typical RSM technique that is usually used to optimize response-affecting process parameters.
To determine the optimum HTC conditions for hydrochar production, we investigated the effects of three factors (reaction temperature, residence time, and acetic acid concentration) and optimized them within the ranges of 160-260 °C, 30-150 min, and 0-3.0 M, respectively, to maximize the yield and HHV of SS hydrochar. The relationship between HTC conditions and gasi cation performance was investigated using the gasi cation activity index as the optimization goal under different HTC conditions.
The experimental design was carried out using Design-Expert.V8.0.6.1 data analysis software (Stat-Ease, Inc). The optimization conditions and objectives of HTC and gasi cation processes are presented in Tables 2 and 3.   In the experimental process, only 17 groups of experimental conditions were selected, which could not intuitively obtain the optimal conditions. Through response surface analysis, the functional relationship between response target and factors can be established, and the response surface diagram can be used to display the functional relationship, and the optimal reaction condition can be obtained.   In the above and subsequent equations, A, B, and C stand for HTC temperature (°C), residence time (min), and acetic acid concentration (M), respectively.

Response surface analysis of hydrochar yield
Hydrochar yield was negatively correlated with all three of these parameters, with the largest and smallest effects observed for temperature and residence time, respectively (Fig. 3). In particular, hydrochar yield decreased by 15, 10, and 5% as the reaction temperature, acetic acid concentration, and residence time increased from 160 to 260°C, 0 to 3 M, and 30 to 150 min, respectively. ). The largest decrease in hydrochar yield was observed when the reaction temperature increased from 210 to 260°C. According to Reza et al., this decrease was primarily due to the signi cant breakdown of hemicellulose and lignin in this temperature range and the large number of cellulose decomposition reactions above 210°C (Tou q et al. 2016). Alternatively, the negative correlation between temperature and hydrochar yield was ascribed to the facts that (i) SS contains numerous nitrogen-containing compounds such as proteins, which, together with sugar hydrolysates, are dissolved in the aqueous phase ) and (ii) hydrochar is di cult to produce by the thermal decomposition of proteins at ~250°C. Temperature also had a larger effect on mass yield than organic acids. Acetic acid is the main organic acid produced during HTC through the hydrolysis and dehydration of straight-chain polymers such as cellulose and hemicellulose or simple monomers in the presence of subcritical water . The presence of organic acids increases the concentration of protons or hydroxide ions, thus increasing the total ion concentration and promoting decarboxylation and dehydration (Reza et al. 2015). Acids also enhance dehydration, which is the main carbonization mechanism, and thus signi cantly decrease the oxygen content of hydrochar. It has been reported that hydrochar yield can be increased by decreasing the reaction time (Saetea et al. 2013), which is in line with our results. This behavior was ascribed to the depolymerization of biomacromolecules in SS and the further degradation of the resultant intermediates via cleavage, dehydration, decarboxylation, and deamination at long residence times.

Results of process optimization
The optimization of HTC conditions is important for hydrochar recovery, especially from the perspective of process development. Ideally, maximal hydrochar yield and HHV should be obtained under mild conditions. The limiting conditions and results of hydrochar HHV and yield optimization are listed in Tables 7 and 8.  According to the prediction results, optimal conditions corresponded to an HTC temperature of 160°C, a residence time of 30 min, and an acetic acid concentration of 0.07 M, with the yield and HHV of hydrochar obtained under these conditions equaling 82.69% and 7820.99 kJ kg −1 , respectively. In the replication experiment conducted under these conditions, the hydrochar yield and HHV were determined as 83.07% and 7794.25 kJ kg −1 , respectively, deviating from the predicted values by 0.46 and 0.34%, respectively. The good agreement between simulated and experimental results con rmed the high reliability of our prediction.
The above results indicate that SS hydrochar can be used as a fuel for combustion, pyrolysis, and gasi cation. Among these applications, combustion is relatively simple but may produce gaseous pollutants. Therefore, the next section compares the release of gaseous pollutants in different combustion atmospheres.

Release of gaseous pollutants in different combustion atmospheres
The above analysis results show that SS hydrochar holds great promise as a combustion fuel. However, as gaseous pollutants are inevitably produced during conversion, alternatives featuring lower pollutant emissions are highly sought after. This section explores the release of nitrogen-and sulfur-containing gases during hydrochar combustion in O 2 (99.999 vol%) and CO 2 (99.999 vol%) atmospheres. As shown in Fig. 5, when hydrochar was burned in an O 2 atmosphere, the peaks of sulfur-and nitrogen-containing gases were more intense than those observed in a CO 2 atmosphere. In particular, almost no peak of sulfur-containing gas was observed in the latter case. Therefore, the gasi cation of hydrochar in a CO 2 atmosphere seems to be a promising thermal conversion method. The next section focuses on the gasi cation characteristics of hydrochar.
3.4 Analysis of CO 2 gasi cation characteristics 3.4.1 CO 2 emission due to combustion of residual char obtained in different gasi cation atmospheres Figure 6 shows the release of CO 2 upon the combustion of residual char obtained after the completion of hydrochar gasi cation in Ar and CO 2 atmospheres, revealing that more CO 2 (and hence, a lower carbon conversion) was obtained in the former case. Therefore, the CO 2 atmosphere was concluded to be more suitable for gasi cation reactions than inert (e.g., Ar) atmospheres. Figure 8. Effects of (a) temperature, (b) residence time, and (c) acetic acid concentration on the release of gaseous products during CO 2 -assisted gasi cation.

Analysis of gaseous product evolution
With increasing HTC temperature and residence time, the amount of released gas decreased.
HTC temperature had the largest in uence on gas release and was negatively correlated with the contents of CO, CH 4 , and H 2 (Fig. 8a) Figure 9 shows the effect of time on conversion during CO 2 -assisted hydrochar gasi cation, revealing that compared with raw SS, hydrochar required less time to reach a conversion of unity, i.e., the gasi cation activity of sludge increased after HTC. Moon et al. (Moon et al. 2015) showing that the rise in the lignin content of SS after HTC helped to increase the production of methane during gasi cation and allowed this gasi cation to be completed in a shorter time. At 160°C and 260°C, the conversion of hydrochar was lower than that of raw SS, but at the intermediate temperature (210°C), the conversion is higher. Moreover, hydrochar gasi cation reactivity increased with increasing residence time and acetic acid concentration.

Analysis of the gasi cation activity optimization of hydrochar
The ANOVA results for the regression model used to predict the gasi cation activity index are shown in Table 9.
The p-value of the employed model was less than 0.05, and the model lack-of-t (p > 0.05) was insigni cant. Figure 10 compares the predicted results with the experimental results, and the adj.R 2 is 0.993, indicating that the difference between the two is small and the tting degree is high.In the working condition 160-90-1.5, the value of R 0.9 is the largest, so it is distributed in the upper right corner. The quadratic regression equation used to calculate this index is given below.    The results optimization of activity index listed in Tables 10 and 11. According to the predicted results, the highest gasi cation activity index (0.29 s −1 ) was obtained at an HTC temperature of 211.34°C, a residence time of 88.16 min, and an acetic acid concentration of 1.58 M. When a replication experiment was conducted under the above conditions, R 0.9 was obtained as 0.28 s −1 , deviating from the predicted value by 2.19%, which was within the error range. Compared with those required to obtain high-quality hydrochar (160°C, 30 min, 0.07 M) as solid fuel in Section 3.2.3, the optimized conditions for hydrochar gasi cation (211.34°C, 88.16 min, 1.58 M) were more severe. Milder HTC conditions allowed hydrochar to retain more carbon and some soluble organic matter and thus maximize the HHV and yield, whereas harsher conditions favored hydrochar pore development and alkali metal retention, thus promoting catalytic gasi cation.

Conclusions
The main objective of this work is to convert the sludge into a usable bioenergy feedstock by hydrothermal carbonization. The main results are as follows.
(1) With increasing HTC temperature, residence time, and acetic acid concentration, the carbon, hydrogen, and nitrogen contents of hydrochar decreased. The highest hydrochar yield and HHV were obtained under mild conditions.
(2) The conditions affording maximal HHV and yield corresponded to 160°C, 30 min, and 0.07 M acetic acid. Under these conditions, the hydrochar yield and HHV were predicted to equal 82.69% and 7820.99 kJ kg −1 , respectively. These values deviated from the experimentally determined ones by 0.34 and 0.46%, respectively, i.e., were within the acceptable error range.
(3) HTC conditions strongly affected hydrochar gasi cation performance. As HTC conditions became more severe, the gasi cation activity index rst increased and then decreased, i.e., was maximized under moderate conditions. The RSM-predicted optimal HTC conditions corresponded to 211.34°C, 88.16 min, and 1.58 M acetic acid. Under these conditions, the hydrochar gasi cation activity index equaled 0.29 s −1 .
The results of this study provide a detailed observation of SS utilization by HTC coupled with gasi cation and providing referential information for the design, optimization, and even upscaling of thermochemical conversion processes. However, under the expected optimization conditions, extensive experiments on SS should be the future work. In addition, a comprehensive technical and economic analysis of SS HTC process is needed to prove its scalability.

Declarations
Authors' contributions Chenchen Zhao and Shuai Guo had the idea for the article; Dandan Xu, Shuai Guo, and Xin Guo performed the literature search and data analysis; and Shuai Guo, Dandan Xu, Xin Guo, and Xingcan Li Effects of HTC conditions on the conversion of raw sludge and sludge hydrochar during gasi cation.

Figure 10
Comparison between actual and predicted indexes of gasi cation reactivity.