Correlations between the physicochemical properties of hydrochar and specific components of waste lettuce: Influence of moisture, carbohydrates, proteins and lipids

Accepted Manuscript Correlations between the physicochemical properties of hydrochar and specific components of waste lettuce: Influence of moisture, carbohydrates, proteins and lipids Yang Li, Huan Liu, Kangxin Xiao, Xiang Liu, Hongyun Hu, Xian Li, Hong Yao PII: DOI: Reference:

S0960-8524(18)31498-6 https://doi.org/10.1016/j.biortech.2018.10.066 BITE 20630

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

10 September 2018 22 October 2018 23 October 2018

Please cite this article as: Li, Y., Liu, H., Xiao, K., Liu, X., Hu, H., Li, X., Yao, H., Correlations between the physicochemical properties of hydrochar and specific components of waste lettuce: Influence of moisture, carbohydrates, proteins and lipids, Bioresource Technology (2018), doi: https://doi.org/10.1016/j.biortech. 2018.10.066

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.

Correlations between the physicochemical properties of hydrochar and specific components of waste lettuce: Influence of moisture, carbohydrates, proteins and lipids Yang Lia, Huan Liua, b, Kangxin Xiaoa, Xiang Liua, Hongyun Hua, Xian Lia, Hong Yaoa a

State key Laboratory of Coal Combustion, School of Energy and Power Engineering, Huangzhong

University of Science and Technology, Wuhan, 430074, China b

Department of New Energy Science and Engineering, School of Energy and Power Engineering,

Huazhong University of Science and Technology, Wuhan 430074, China



Corresponding author Tel & Fax: +86-27-87542417

E-mail: [email protected] (A.Prof. Huan Liu) 1

Abstract This study aims to figure out the influence of moisture content and chemical constitution, i.e. carbohydrates, proteins and lipids, of waste lettuce on the physicochemical structure of hydrochar produced via hydrothermal carbonization. The experimental results showed that homogenized carbon material can be obtained by hydrothermal treatment, regardless of the moisture content of feedstock. During the hydrothermal carbonization process of waste lettuce, carbohydrates were the most active reactants contributing to hydrochar formation. Meanwhile, Maillard reaction between proteins and carbohydrates occurred, which promoted the aromatization of the organic intermediates and increased the relative content of nitrogenous heterocyclic functional groups on the surface of hydrochar from 10.7 to 18.7%. Different from these two constitution, lipids did not participate in the carbonization reaction, the main hydrolyzates of lipids were adsorbed to the surface of hydrochar, leading to an increase in the mass of solid products. Keywords: Waste lettuce; Hydrothermal carbonization; Moisture content; Feedstock constitution; Hydrochar

2

1. Introduction Food waste represents a largely underutilized organic fraction of municipal solid waste (MSW), accounting for about 40-60% of the total MSW in China (Li et al., 2013; Zhou et al., 2018). The disadvantages of high and variable moisture during storage and transportation restrict the direct utilization of food waste. In view of this situation, hydrothermal carbonization (HTC) as a novel, low-cost, environmentally friendly thermochemical conversion technique, with the advantage of skipping pre-drying step, was proved to be an effective means to convert food waste into energy and carbonaceous materials (Hu et al., 2010; Deng et al., 2016; Titirici et al., 2012). HTC is usually conducted at mild processing temperatures (180-250 oC) and the carbonization medium is water, under self-generated pressures (A Funke et al., 2010; Wu et al., 2017). The solid product obtained by HTC, referred to as hydrochar, with stable aromatic structure and abundant surface functional groups, has broad application prospects in the fields of solid fuels, soil improvers, adsorbents, electrode materials, etc. (Zhao et al., 2014; Mäkelä et al., 2017; Parshetti et al., 2014; Baccile et al., 2010). Hydrochar with different physical and chemical properties can be used in different fields, which are associated with variable chemical compositions of feedstocks (Sevilla and Fuertes, 2009; Wang et al., 2018; Sevilla and Fuertes, 2009). As noted by Berge et al. (2011) for hydrothermal carbonization of paper, rabbit food and sludge, the yield of char was 29.2%, 43.8%, and 47.1%, respectively, and char from the rabbit food has 3

the highest carbon content and aromaticity, whereas the sludge char has the lowest. Blankenship et al. (2017) used the cigarette butt as feedstock for hydrothermal carbonization, and found that the activated hydrochar had abundant pore structure and ultra-high surface area (4300 m2 g-1), which can be used as hydrogen storage materials. Heilmann et al. (2011) reported that abundant fatty acid was produced during the HTC of high lipid-containing microalgae. Obviously, variable components undergo complex reactions in the hydrothermal process, which affects the properties of the char product consequently. As for the food waste mentioned above, some are mainly composed of cellulose, hemicellulose and lignin, while others mainly consist of proteins, lipids and carbohydrates. How these dominant components affect properties of char products is a very significant but under-researched issue. Therefore, the authors have recently investigated the correlations between chemical constitution of cellulose, hemicellulose, lignin and the physicochemical structure of hydrochar by employing the orange peel as representative (Xiao et al., 2018). Some interesting results were found that more hydrochar, with higher specific surface area and more benzene rings, were produced with lignin increasing, while hemicellulose increasing, hydrochar had higher density of carbonaceous microspheres. The same problem also needs to be addressed to the high water-containing food waste, mainly composed of carbohydrates, proteins and lipids. Waste lettuce (WL) is a typical representative selected because it is a common food 4

waste, with ultrahigh content of moisture (~95%) (Bovi et al., 2016) as well as carbohydrate (60-66%), protein (18-26%) and lipids (3-6%) (McKeehen et al., 1996). In view of the above, this study aims to investigate the influence of moisture content and chemical constitution, i.e. carbohydrates, proteins and lipids, on properties of hydrochar produced via HTC of WL by controlling the content of the components of feedstock. The influence mechanism of components was further revealed by analysis of the difference in physical and chemical structure of hydrochar. This mechanism can serve as a guide to adjust the specific structure of hydrochar by regulating the composition of feedstock.

2. Material and methods 2.1 Materials and sample preparation

Waste lettuce (WL) were collected from a local market in Wuhan, China. The moisture content of WL (as received) was 95.3 wt%, measured with a moisture analyzer (Mettler Toledo HC-103). The chemical characteristics of the WL are given in Table 1. As Table 1 showed that raw waste lettuce had a high ash content and a low fixed carbon content. To obtain lettuce with different moisture content, raw waste lettuce were dried at 105 oC for several hours. The specific pretreatment methods and the moisture content of the samples after pretreatment are listed in Table 2. Then the dry WL were crushed into powder and screen-sieved (< 200 μm). Lactose, serum albumin and 5

triacylglycerides, purchased from Aladdin Reagent Co. Ltd., Shanghai, China, were used as the three components of carbohydrate, protein and lipids in WL.

2.2 HTC experiments

Samples were carbonized in a 250-mL stainless steel batch reactor (see Fig.1) with a heater, temperature controller and pressure gauge under autogeneous pressure and Ar atmosphere. In order to investigate the effect of moisture content on hydrochar, five groups of hydrothermal experiments of WL samples with various moisture content were performed. Five groups of WL samples with a dry basis mass of 10 g were added into reactor. Then deionized water was added to meet the solid-liquid mass ratio in Table 2. To investigate the impact of chemical composition on hydrochar, another set of experiments were performed. 10 g of dry WL, lactose, protein (serum albumin) and triacylglycerides (named as TAGs) were separately added into reactor as 4 blank groups. Afterwards, 5 g of lactose, protein and TAGs, separately mixed with 10 g of WL powder were added into reactor as experimental groups. The feedstock of experimental groups was named as WL-X (Lactose, Protein and TAGs). Deionized water was subsequently added to obtain a solid-liquid mass ratio of 1:8. The reactor was then heated using an electric furnace at an approximately rate of 4 °C/min to reach a final temperature (240 °C) and maintained for 120 min with an rotation speed of 200 rpm. In the above hydrothermal environment, the reactor pressure 6

is 4.0 MPa in all HTC experiments. Following this hold time, the reaction kettle was placed in cold water and cooled to room temperature (25 oC). The gas was collected into gas sampling bags. The mass of the gas were calculated by the difference in mass before and after the reactor was exhausted. The process liquid and solid products were separated by vacuum filtration with medium speed qualitative filter papers. All solid products were washed three times with deionized water and dried at 105 °C. Then solid and liquid products were weighed respectively. To control experimental error, each set of experiments was repeated twice. Hydrochar yield (HCY) was expressed as the following formula: HCY= MHC/MRS

(1)

where MHC is the dry basis mass of the solid product after hydrothermal treatment, and MRS is the dry basis mass of the raw sample. Similarly, gas yield and liquid yield were defined as the mass of gas product or liquid product produced per unit mass of dry raw samples. WL-X* represents the theoretical yield of WL-X hydrothermal products, which is calculated by the following equation. YP = YWL KWL + YX KX

(2)

where YP is the predicted yield data from linear superposition calculation, Y WL and YX are the yield data from WL and X respectively, KWL and KX are the mass proportion of WL and X in the corresponding WL-X experimental group separately.

7

2.3 Hydrochar characterization

To get the chemical characteristics of WL, proximate analysis were performed using a proximate analyzer (ELTRA, THERMOSTEP) on the basis of proximate analysis of solid biofuel GB/T 28731-2012. To obtain the proportion of elements in hydrochar, ultimate analyses were performed using an elementary analyzer (Elementar, Vario Micro cube) for CHNS. To reveal the physical structure of hydrochar, SEM images of the surface morphology were acquired on a field emission scanning electron microscope (FE-SEM, MIRA3 TESCAN, 15 kV). To investigate the differences in chemical properties of hydrochar, the surface chemical states of hydrochar were carried out using Xray photoelectron spectroscopy (XPS, Kratos, AXIS-ULTRA DLD-600W). Binding energies for the high-resolution spectra were calibrated by setting C 1s to 284.6 eV. The functional groups were investigated by Fourier transform infrared spectroscopy (FTIR, Bruker, VERTEX 70) in the 4000-400 cm-1 region with 100 scans. Before the determination, the samples were ground into fine particles to mix with KBr for the preparation of disks, with 1:200 to the sample-to-KBr mass ratio.

8

3. Results and discussion 3.1 Influence of moisture content on waste lettuce derived hydrochar

Table 3 showed that the results of ultimate analysis and product yields for hydrochar obtained from waste lettuce with various moisture content. After hydrothermal treatment of the waste lettuce, the oxygen content was significantly reduced from 48.7 to 25.0% and the carbon content was increased from 41.3 to 63.0%, as well as the H/C and the O/C ratios decreased significantly showed in the Van Krevelen diagram (see Fig. 2), which was similar to the atomic ratios of fuel coals. These results, mainly caused by the dehydration and decarboxylation, indicated that HTC can effectively improve the quality and utilization value of waste lettuce. It was not hard to find out that internal moisture content and S/L of raw materials had little effect on the hydrochar yields, which ranged between 26.9% and 28.4%. These results are in agreement with previous research (Sabio et al., 2016) on HTC of tomato-peel waste. Five groups of hydrochar exhibit similar functional group (OH, C=O, O-CH3, Aro CH) and elemental composition, shown in FTIR spectra and Table 3. Notably, the content of nitrogen in hydrochar was slightly reduced when the solid-liquid ratio was lowered (from 1:8 to1:20), and the possible reason was that more nitrogen-containing intermediates dissolved into the liquid phase rather than further forming char due to the increasing water contents (Volpe et al., 2018). Homogenized hydrochar with minor 9

differences in nitrogen content can be obtained by HTC of waste lettuce, regardless of moisture content and solid-liquid ratio.

3.2 Effect of chemical constitution on mass yield and physical structure of hydrochar

As shown in Fig. 3, the total recovery of three-phase products was (88-94%) was in a reasonable range. These lost parts resulted from a small fraction of volatile in liquid and residue in the reactor. There was no hydrochar formed during the HTC process of protein and TAGs singly. However, the hydrochar yield of lactose was up to 41% and the yield of WL-Lactose derived hydrochar was close to the theoretical value (WLLactose*), indicating that carbohydrates were the most active reactants contributing to char formation during HTC of WL. The char yields of WL-Protein and WL-TAGs were significantly higher than that of WL-Protein* and WL-TAGs*. With the addition of protein or TAGs, the significant increase in char yield possibly resulted from the interaction between WL and protein or TAGs during HTC. More information is needed to investigate the specific interaction path which leads to an increase in char yield when protein or TAGs added. SEM images revealed the surface morphology of hydrochar from different feedstock. WL-derived hydrochar exhibits an irregular block morphology. On the contrary, the hydrochar from lactose consists mainly of aggregates of microspheres with a diameter in the 0.5-2 μm range. Whereas，lactose or proteins were added, more 10

randomly dispersed squares with a size of about 2 μm appeared on the surface of the hydrochar. It should be noted that the surface of the hydrochar obtained from WL-TAGs seemed to be wrapped by a layer of oily material, exhibiting an agglomerate morphology. It was inferred that an insoluble intermediate product attached to the surface of hydrochar during HTC of WL-TAGs.

3.3 Impact of chemical constitution on chemical properties of hydrochar

Table 4 showed the information of hydrochar ultimate analysis. The O/C atomic ratio of hydrochar from WL-Lactose, WL-Protein and WL-TAGs was lower than that of WL derived hydrochar. As the Van Krevelen diagram showed, among the three experimental groups, hydrochar from WL-TAGs had the lowest O/C atomic ratio and the highest H/C atomic ratio, which suggested that hydrochar from WL-TAGs has a high heating value. WL-lactose hydrochar exhibited the similar O/C and H/C ratio to coal. Proteins were carbonized and incorporated in chars when waste lettuce were also present, which can be confirmed by that no solid product formed during HTC of proteins and the increase of the nitrogen content of hydrochar from WL-Protein. To further investigate the effects of the three organic components on the chemical composition of hydrochar, the analysis of functional groups was presented below. As FTIR spectra of the hydrochar samples shown, The bands at approximately 2920 and 3000–3600 cm-1 correspond to stretching vibrations of aliphatic CH and OH (hydroxyl 11

or carboxyl), respectively (Sevilla and Fuertes, 2009). The bands at 1700 and 1620 cm-1 (together with the band at 1513 cm-1) can be attributed to C=O (carbonyl, ester, or carboxyl) and C=C vibrations respectively, whereas the bands around 1460 cm-1 region correspond to C=C stretching in aromatic groups (Sevilla and Fuertes, 2009; Yang et al., 2007). The band at 500-650 cm-1 are assigned to aromatic CH out-of-plane bending vibrations (Sevilla et al., 2011). The position and intensity of each peak in the FTIR spectrum of hydrochar from WL-Lactose is similar to that of hydrochar from WL, indicating that the increasing carbohydrate content in WL has no significant effect on the functional structure of hydrochar. The peaks at 3350 cm-1, 2920 cm-1 and 1700 cm-1 from WL-Protein hydrochar showed stronger peak intensity than that from WL hydrochar due to the bands at 29003480 cm-1 and 1680-1750 cm-1 also corresponding to stretching vibrations of N-H and C-N (Titirici and Antonietti, 2010). These results were consistent with the ultimate analysis results in Table2, suggesting that more proteins had participated in carbonization during HTC with the protein content increasing in WL. In addition, the bands ascribed to aromatic C=C and aromatic CH vibrations from WL-Protein char shifted to higher peaks with comprehensive consideration of the peak intensity of WL char, which suggested that the increasing proteins content had positive influences on the formation of the aromatic structure. The peak of WL-TAGs hydrochar at 3350 cm-1 was almost disappeared, which 12

means that the hydroxyl group contents reduced significantly. The possible reason is that the added TAGs could promote the removal of hydroxyl groups by enhancing the dehydration reaction (intermolecular dehydration, aldol reaction, etc.) in the hydrothermal process. However, the intensity of peak at 1700 cm-1 (C=O) showed obvious increase, with the consideration of the appearance of the characteristic peak at 925 cm-1 (OH--H in acid dimer), indicating the increasing in carboxyl functional group (-COOH) content of WL-TAGs hydrochar. Moreover, a series of absorption peaks appearing at 1385-1100 cm-1 together with the characteristic peaks at approximately 720 cm-1 were specifically assigned to in-plane bending vibrations of CH2 in long-chain alkanes (fatty acids, esters, etc.) (Dean et al., 2010). Thus, it can be speculated that the aforementioned substance attached to the surface of WL-TAGs hydrochar (as shown in SEM image) is specifically fatty acid (hydrolysate of TAGs). The evolution of surface functional groups in hydrochar was detected using XPS. As the XPS C1s spectrum of hydrochar showed, it contains four signals attributed, respectively, the aliphatic/aromatic carbon group (CH X, C-C/C=C) (284.6 eV), hydroxyl/ amino groups (-C-OR/-C-NR) (285.8 eV), carbonyl/imine groups (>C=O/>C=N) (287.3 eV) and carboxylic groups, esters or lactones (-COOR) (289.0 eV) (Sevilla and Fuertes, 2009; Si et al., 2013; Chen et al., 2017). The normalized relative intensities of carbon functionalities were listed in Table 5. These results showed that there was an abundance of oxygen groups in the shells of hydrochar. It was noticed 13

that the oxygen groups exhibited no significant change in hydrochar from WL and WLLactose. However, after TAGs added, the hydrochar had a significantly higher CHx/CC/C=C content (82.1%) compared to that of the WL hydrochar (70.9%). This result agreed with the increase of aliphatic -CH2 in FTIR spectra, suggesting that fatty acid attached to the surface of WL-TAGs char. The relative content of C-O/C-N, C=O/C=N and O=C-O functional groups on the surface of WL-Protein hydrochar (19%, 10.5%, 4.7%) is slightly higher than that of WL hydrochar (17.8%, 7.8%, 3.5%), indicating that the content of oxygen and nitrogen functional groups on the surface of hydrochar was slightly increased due to the increasing protein content. In order to further clarify the specific structure of the nitrogen-containing functional group on the surface of chars, the XPS N1s spectrum of chars from WL and WL-Protein was analyzed. Three N1s signals were identified at 398.8 eV, 399.7 eV and 400.2 eV, which were attributed to pyridine-N, protein-N and pyrrole-N individually (Si et al., 2013; Wang et al., 2018). Table 6 showed the normalized relative contents of nitrogen functionalities. The WL-Protein hydrochar had a higher pyridine-N content (18.7%) compared to that of the WL hydrochar (10.7%). Furthermore, the protein-N percentage in hydrochar reduced from 45.3 to 37.2%. Therefore, it can be inferred that more protein-N to be converted into pyridine-N during the carbonization process of WL-protein. In addition, as shown in the photograph of hydrochar, the WL hydrochar were black solid powders while the WL-Protein hydrochar exhibit brown. According to 14

previous literature (Sara et al., 2001; Zhuang et al., 2017), in the subcritical water, the products of the typical Maillard reaction between amino compound and reducing sugar were brown nitrogen-containing heterocyclic polymer. Thus, it was reasonable to deduce that Maillard reaction occurred during HTC of WL-Protein, and heterocyclic Maillard reaction products with reactive substituent groups were able to participate in the formation of hydrochar, which confirmed the increase of pyridine-N content in XPS N1s spectrum. To further reveal the influence mechanism, HTC of the mixed pure substances (lactose, protein, TAGs) was performed. FTIR spectra of these chars showed that char from Lctose+TAGs possessed abundant chain CH2 and carboxyl group but little hydroxyl group, which consistent with the WL-TAGs hydrochar. Moreover, the normalized relative peak intensities of XPS (Table 5) showed the relative content of CHX, C-C/C=C of hydrochar from Lctose+TAGs reached up to 83.6%. These results further pointed out that fatty acid adsorbed onto the surface of hydrochar. The char from Lactose+Protein had superior relative intensities of aromatic C=C. In addition, the relative content of pyridine-N is 21.1%, which demonstrated that HTC of mixed lactose and protein generated abundant nitrogen-containing heterocyclic compounds (the typical Maillard reaction products). According to the above experimental results, the influence mechanism of raw components on the properties of hydrochar obtained from waste lettuce can be inferred 15

(see Fig. 4). Carbohydrates are the main source of hydrochar in the process of hydrothermal carbonization of waste lettuce, which undergo a series of hydrolysis, dehydration, cracking, decarboxylation, polymerization and aromatization reactions, finally forming hydrochar, in accordance with literature reports. Triacylglycerides did not participate in carbonization during the hydrothermal process, meanwhile the hydrolyzate of TAGs, i.e. fatty acid produced more hydrogen ions into the hydrothermal environment, which promoted the dehydration reaction leading to the decrease of hydroxyl content in hydrochar. However, fatty acid will be adsorbed on the surface of char formed by carbonization of carbohydrates, resulting in an increase in the mass of solid products. Maillard reaction between proteins and carbohydrates occurred during the HTC of waste lettuce, which enhanced the incorporation of nitrogen (pyridine-N and pyrrole-N) into the aromatic network of the hydrochar with the increase of protein content in waste lettuce. Hence, it is possible to increase the proportion of carbohydrates during HTC of food waste to obtain more stable hydrochar; increase the proportion of lipids to acquire hydrochar with high yield and potential application as a hydrophobic adsorbent; increase the proportion of proteins to gain hydrochar with more surface nitrogen heterocyclic structure for electrode materials.

16

4. Conclusion Correlations between moisture content, chemical constitution of waste lettuce and the physicochemical structure of its derived hydrochar were revealed in this study. Homogenized hydrochar can be obtained by hydrothermal treatment, without consideration of the moisture content of feedstock. Carbohydrates were the main contributors to the formation of hydrochar. Proteins participated in the formation of char only when carbohydrates were also present, Maillard reaction between proteins and carbohydrates effectively increased the content of aromaticity and nitrogenous heterocyclic functional groups. Lipids did not participate in the carbonization reaction and fatty acids were able to be adsorbed onto the surface of hydrochar.

Supplementary data Supplementary data associated with this article can be found in the online version.

Acknowledgments The authors are grateful to National Nature Science Foundation of China (No. 51661145010, 51506064) and Fundamental Research Funds for the Central Universities (HUST: 2017KFKJXX015) for the financial supports. They also express thanks to the Analytical and Testing Center of Huazhong University of Science and Technology for the testing. 17

References 1. A Funke, F.Z., 2010. Hydrothermal carbonization of biomass: A summary and discussion of chemical mechanisms for process engineering. Biofuels, Bioproducts and Biorefining. 4, 160-177. 2. Baccile, N., Antonietti, M., Titirici, M., 2010. One-Step Hydrothermal Synthesis of Nitrogen-Doped Nanocarbons: Albumine Directing the Carbonization of Glucose. ChemSusChem, 3(2), 246-253. 3. Berge, N.D., Ro, K.S., Mao, J., Flora, J.R.V., Chappell, M.A., Bae, S., 2011. Hydrothermal Carbonization of Municipal Waste Streams. Environ. Sci. Technol., 45(13), 5696-5703. 4. Blankenship, T.S., Mokaya, R., 2017. Cigarette butt-derived carbons have ultra-high surface area and unprecedented hydrogen storage capacity. Energ. Environ. Sci., 10(12), 2552-2562. 5. Bovi, G.G., Caleb, O.J., Linke, M., Rauh, C., Mahajan, P.V., 2016. Transpiration and moisture evolution in packaged fresh horticultural produce and the role of integrated mathematical models: A review. Biosyst. Eng., 150, 24-39. 6. Chen, N., Huang, Y., Hou, X., Ai, Z., Zhang, L., 2017. Photochemistry of Hydrochar: Reactive Oxygen Species Generation and Sulfadimidine Degradation. Environ. Sci. Technol., 51(19), 11278-11287. 7. Dean A., Sigee D., Estrada B, Pittman J. 2010. Using FTIR spectroscopy for rapid 18

determination of lipid accumulation in response to nitrogen limitation in freshwater microalgae. Bioresource Technol., 101, 4499-4507. 8. Deng, J., Li, M., Wang, Y., 2016. Biomass-derived carbon: synthesis and applications in energy storage and conversion. Green Chem., 18(18), 4824-4854. 9. Heilmann, S.M., Jader, L.R., Harned, L.A., Sadowsky, M.J., Schendel, F.J., Lefebvre, P.A., von Keitz, M.G., Valentas, K.J., 2011. Hydrothermal carbonization of microalgae II. Lipidsty acid, char, and algal nutrient products. Appl. Energ., 88(10), 3286-3290. 10. Hu, B., Wang, K., Wu, L., Yu, S., Antonietti, M., Titirici, M., 2010. Engineering Carbon Materials from the Hydrothermal Carbonization Process of Biomass. Adv. Mater., 22(7), 813-828. 11. Li, L., Diederick, R., Flora, J.R.V., Berge, N.D., 2013. Hydrothermal carbonization of food waste and associated packaging materials for energy source generation. Waste Manage. , 33(11), 2478-2492. 12. Mäkelä M, Kwong C., Broström M, Yoshikawa K, 2017. Hydrothermal treatment of grape marc for solid fuel applications. Energ Convers Manage, 145, 371-377. 13. McKeehen, J. D., Smart, D. J., Mackowiak, C. L., Wheeler, R. M., Nielsen, S. S., 1996. Effect of CO2 levels on nutrient content of lettuce and radish. Adv. Space. Res., 18, 85-92. 14. Parshetti, G.K., Chowdhury, S., Balasubramanian, R., 2014. Hydrothermal 19

conversion of urban food waste to chars for removal of textile dyes from contaminated waters. Bioresource Technol., 161, 310-319. 15. Sabio, E., Álvarez-Murillo, A., Román, S., Ledesma, B., 2016. Conversion of tomato-peel waste into solid fuel by hydrothermal carbonization: Influence of the processing variables. Waste Manage. , 47, 122-132. 16. Sara I.F.S.M., Wim M.F.J., Martinus A.J.S., 2011. A review of Maillard reaction in food and implications to kinetic modelling. Trends in Food Sci. Technol., 11, 364373. 17. Sevilla, M., Fuertes, A.B., 2009. The production of carbon materials by hydrothermal carbonization of cellulose. Carbon, 47(9), 2281-2289. 18. Sevilla, M., Fuertes, A.B., 2009. Chemical and Structural Properties of Carbonaceous Products Obtained by Hydrothermal Carbonization of Saccharides. Chemistry - A European Journal, 15(16), 4195-4203. 19. Sevilla, M., Maciá-Agulló, J.A., Fuertes, A.B., 2011. Hydrothermal carbonization of biomass as a route for the sequestration of CO2: Chemical and structural properties of the carbonized products. Biomass and Bioenergy, 35(7), 3152-3159. 20. Si, W., Zhou, J., Zhang, S., Li, S., Xing, W., Zhuo, S., 2013. Tunable N-doped or dual N, S-doped activated hydrothermal carbons derived from human hair and glucose for supercapacitor applications. Electrochim. Acta, 107, 397-405. 21. Titirici, M., Antonietti, M., 2010. Chemistry and materials options of sustainable 20

carbon materials made by hydrothermal carbonization. Chem. Soc. Rev., 39(1), 103116. 22. Titirici, M., White, R.J., Falco, C., Sevilla, M., 2012. Black perspectives for a green future: hydrothermal carbons for environment protection and energy storage. Energ. Environ. Sci., 5(5), 6796-6822. 23. Volpe, M., Goldfarb, J.L., Fiori, L., 2018. Hydrothermal carbonization of Opuntia ficus-indica cladodes: Role of process parameters on hydrochar properties. Bioresource Technol., 247, 310-318. 24. Wang, T., Zhai, Y., Zhu, Y., Li, C., Zeng, G., 2018. A review of the hydrothermal carbonization of biomass waste for hydrochar formation: Process conditions, fundamentals, and physicochemical properties. Renewable and Sustainable Energy Reviews, 90, 223-247. 25. Wang, T., Zhai, Y., Zhu, Y., Peng, C., Xu, B., Wang, T., Li, C., Zeng, G., 2018. Influence of temperature on nitrogen lipidse during hydrothermal carbonization of food waste. Bioresource Technol., 247, 182-189. 26. Wu, Q., Yu, S., Hao, N., Wells, T., Meng, X., Li, M., Pu, Y., Liu, S., Ragauskas, A.J., 2017. Characterization of products from hydrothermal carbonization of pine. Bioresource Technol., 244, 78-83. 27. Xiao, K., Liu, H., Li, Y., Yi, L., Zhang, X., Hu, H., Yao, H., 2018. Correlations between hydrochar properties and chemical constitution of orange peel waste during 21

hydrothermal carbonization. Bioresource Technol., 265, 432-436. 28. Yang, H., Yan, R., Chen, H., Lee, D.H., Zheng, C., 2007. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel, 86(12-13), 1781-1788. 29. Zhao, P., Shen, Y., Ge, S., Chen, Z., Yoshikawa, K., 2014. Clean solid biofuel production from high moisture content waste biomass employing hydrothermal treatment. Appl. Energ., 131, 345-367. 30. Zhou, Y., Engler, N., Nelles, M., 2018. Symbiotic relationship between hydrothermal carbonization technology and anaerobic digestion for food waste in China. Bioresource Technol., 260, 404-412. 31. Zhuang, X., Huang, Y., Song, Y., Zhan, H., Yin, X., Wu, C., 2017. The transformation pathways of nitrogen in sewage sludge during hydrothermal treatment. Bioresource Technol., 245, 463-470.

22

Figure captions Fig. 1. Schematic diagram of hydrothermal reactor Fig. 2. Van Krevelen diagram of raw WL and hydrochar from WL with different moisture and chemical components. Fig. 3. Yield distribution of three-phase hydrothermal products from each feedstock Fig. 4. Influence mechanism of feedstock constitution on hydrochar properties

23

Fig. 1. Schematic diagram of hydrothermal reactor

24

Fig. 2. Van Krevelen diagram of raw WL and hydrochar from WL with different moisture and chemical components.

25

Gas

Liquid

Solid

100

Yeild (wt.%)

80

60

40

20

0

L W

n * se n* se ein se tei ot tei cto to ro cto r a o c a P r P L L La -P LLLW W Samples WL W

TA

G

s TA LW

G

s* L W

A -T

G

s

Fig. 3. Yield distribution of three-phase hydrothermal products from each feedstock

26

Lipids

Proteins

Carbohydrates

hydrolysis

hydrolysis

hydrolysis

Glucose Fructose

Fatty acid

Amino acid

dehydration

polymerization

decarboxylation aromatization

promote

adsorbed

附着 … n

Maillard reaction

promote

O

OH

Hydrochar

Chemical reaction

Indirect effects

Interactive reaction

Surface structure

Physical effect

Fig. 4. Influence mechanism of feedstock constitution on hydrochar properties

27

Table list Table 1. Chemical characteristics of the waste lettuce on dry basis (wt.%) Table 2. Pretreatment and description of waste lettuce samples Table 3. Ultimate analysis and product yields of hydrochar obtained from waste lettuce samples with various moisture contents Table 4. Ultimate analysis of hydrochar from all samples on dry basis (wt.%) Table 5. Normalized peak intensities of XPS C1s spectra for the results of hydrochar Table 6. Normalized peak intensities of XPS N1s spectra for the results of hydrochar

28

Table 1. Chemical characteristics of the waste lettuce on dry basis (wt.%) Proximate analysis

a

Ultimate analysis

VM

FC

Ash

C

H

Oa

N

S

71.39

12.13

16.48

41.31

5.11

32.23

4.69

0.18

By difference

29

Table 2. Pretreatment and description of waste lettuce samples

a

Samples

Pretreatment

Moisture content (%)

S/La

WL-95.3-1:20

No pretreatment

95.3

1:20

WL-0-1:20

Dried in oven at 105 oC for 24 h

0

1:20

WL-0-1:8

Dried in oven at 105 oC for 24 h

0

1:8

WL-68.5-1:8

Dried in oven at 105 oC for 5 h

68.5

1:8

WL-40.6-1:8

Dried in oven at 105 oC for 6.5 h

40.6

1:8

S/L, the solid-liquid mass ratio, means the mass ratio of the dry base to the water

consisting of internal water and externally added deionized water.

30

Table 3. Ultimate analysis and product yields of hydrochar obtained from waste lettuce samples with various moisture contents Ultimate analysis (wt.%, dry basis)

HCY

Samples a

C

H

O

N

S

63.97

6.20

25.84

3.84

0.15

26.87

63.77

6.12

26.36

3.61

0.14

27.85

63.18

5.86

25.97

4.85

0.14

28.23

64.13

5.84

25.39

4.49

0.15

27.24

63.79

6.02

25.30

4.73

0.16

28.38

(%)

WL95.3-1:20 WL-01:20 WL-01:8 WL68.5-1:8 WL40.6-1:8 a

HCY, hydrochar yield.

31

Table 4. Ultimate analysis of hydrochar from all samples on dry basis (wt.%) Sample

C

H

O

N

S

H/C(at.)

O/C(at.)

WL

64.04

6.08

24.66

5.06

0.16

1.139

0.289

Lactose

68.35

4.34

27.25

0

0.06

0.762

0.299

WL-Lactose

67.03

5.24

23.42

4.13

0.18

0.938

0.262

WL-Protein

64.52

6.19

22.39

6.29

0.61

1.151

0.260

WL-TAGs

68.11

8.19

20.71

2.81

0.18

1.443

0.228

Lower

lower

Linear range

Higher

32

Table 5. Normalized peak intensities of XPS C1s spectra for the results of hydrochar Normalized relative intensities of carbon functionalities (%) Samples

CHX, CC/C=C

-C-OR/C-NR

>C=O/>C=N

-COOR

WL

70.9

17.8

7.8

3.5

WL-Lactose

70.2

17.1

7.7

5.0

WL-Protein

65.8

19.0

10.5

4.7

WL-TAGs

82.1

6.0

7.0

4.2

Lactose

65.1

18.0

8.9

8.0

Lactose+TAGs

83.6

8.7

4.7

3.0

Lactose+Protein

74.2

14.9

5.7

5.2

Lactose+Protein+TAGs

73.5

13.6

6.3

6.6

Lower

lower

Linear range

Higher

33

Table 6. Normalized peak intensities of XPS N1s spectra for the results of hydrochar Normalized relative intensities of nitrogen functionalities (%) Samples

Pyridine-N

Protein-N

Pyrrole-N

WL

10.7

45.3

44.3

WL-Protein

18.7

37.2

44.1

Lactose+Protein

21.1

62.3

16.6

Lactose+Protein+TAGs

15.3

62.8

21.8

Lower

Linear range

Higher

lower

34

35



The moisture content of feedstock showed no effects on hydrochar performances.



Carbohydrates were the main contributors to the formation of hydrochar.



Proteins were able to enhance nitrogenous heterocyclic functional groups.



Lipids did not participate in the carbonization reaction.



The main hydrolyzates of lipids were adsorbed to the surface of hydrochar.

36