Journal Pre-proof

A comprehensive investigation of hydrothermal carbonization: Energy potential of

hydrochar derived from Virginia mallow

Maciej Śliz, Małgorzata Wilk

PII: S0960-1481(20)30662-5
DOI: https://doi.org/10.1016/j.renene.2020.04.124
Reference: RENE 13453

To appear in: Renewable Energy

Received Date: 5 March 2020

Revised Date: 22 April 2020
Accepted Date: 24 April 2020

Please cite this article as: Śliz M, Wilk Mał, A comprehensive investigation of hydrothermal
carbonization: Energy potential of hydrochar derived from Virginia mallow, Renewable Energy (2020),
doi: https://doi.org/10.1016/j.renene.2020.04.124.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.

© 2020 Published by Elsevier Ltd.

 1 A comprehensive investigation of hydrothermal carbonization:
2 Energy potential of hydrochar derived from Virginia mallow
3 Maciej Śliz1, Małgorzata Wilk2

4 AGH University of Science and Technology, 30 Mickiewicza Av., 30-059 Kraków, Poland,
1
5 e-mail: [email protected] – corresponding author
2
6 e-mail: [email protected]
7
8 Abstract
9 In this study, the effects of temperature, residence time and biomass to water ratio on
10 hydrothermal carbonization of Virginia mallow were conducted. The main goal of this
11 research was to investigate hydrochars derived from Virginia mallow in order to assess their
12 fuel properties. Proximate, ultimate and high heating value analyses were used to establish the
13 chemical energy content of raw biomass and hydrochars. TGA, SEM, BET and FTIR were
14 performed to discuss the combustion characteristics and changes in the surface of the material.
15 Temperature was found to be the main factor influencing the conversion of Virginia mallow
16 during the HTC process. It was confirmed by the properties of hydrochar derived at 220°C for
17 5 min which was as carbonaceous (51.8% carbon content) and reactive as that obtained at
18 200°C for 60 min (52.5% carbon content). A sample pretreated at 200°C for 90 minutes had
19 the highest fixed carbon value (16.29%) when compared to raw feedstock (5.63%). In the
20 resulting reduction of mass found in the hydrochar, an increase in the energy densification
21 ratio as well as better combustion stability were also observed. Detailed analysis of the results
22 confirmed that the HTC is a promising method for producing energy-dense solid biofuel.

23 Keywords
24 hydrothermal carbonization, HTC, Virginia mallow, Sida hermaphrodita, TGA, FTIR

1
25 1. Introduction
26 Nowadays, more and more biomass is used for energy purposes. It is connected with the
27 constant growth of demand for heat and electricity in industry, commerce, transport and
28 households [1]. According to the European Union each member state is obligated under law to
29 expand energy production from renewable resources. As a result of various support schemes,
30 biomass is commonly used to meet Renewable Energy Directive regulations [2,3], and will
31 probably be further used to fulfil the expectations of the Revised Renewable Energy Directive.
32 It sets a goal for a share of at least 32 % of energy from renewable sources in the EU’s gross
33 final consumption of energy by 2030 [4]. The main origins of biomass for combustion and
34 co-combustion are forestry, agriculture, waste from biological origin and energy crops. All of
35 these sources have their drawbacks: mostly high moisture content; therefore low calorific
36 value, biodegradability and poor grindability [5,6]. In recent years, attention has more often
37 been paid to biomass upgrading and conversion processes. Torrefaction, pyrolysis, gasification
38 and hydrothermal carbonization (HTC) are good examples of such pretreatment methods [7–
39 15]. Taking into account the fact that raw biomass can consist of 90% water, HTC is
40 considered to be one of the most promising enrichment procedures. In contrast to the
41 aforementioned techniques, HTC does not require pre-drying of the material. It is a process
42 conducted in an aquatic environment with temperatures ranging from 150°C to 300°C.
43 Applied pressure is often autogenic and can range from 2MPa to 10MPa [16]. Recently, other
44 studies have considered using specific values of pressure to enhance the degradation of treated
45 material [17]. Most of the tested setups also have a built-in stirrer to enable reactions to take
46 place throughout the whole volume of the reactor. Due to an elevated temperature and
47 pressure hydrolysis, dehydration, decarboxylation, polymeric condensation and aromatization
48 occur [18]. Those reactions do not run successively, but rather mesh with each other [19]. The
49 mechanism of hydrothermal carbonization is widely studied and even though it is complicated
50 to describe everything which is happening in the reactor at any given moment there are trials
51 to predict the properties of the products [20–22]. That has led to the creation of three different
52 products in three states of matter. The gas product takes a few percent of input mass and it
53 consists mainly of CO2. The ratio of liquid and solid product depends on the initial ratio of
54 treated material and water. Usually, the mass of solid phase decreases because of the release of
55 water and carbon dioxide as well as leaching of some compounds into liquid. As a result, the
56 molar ratios of H/C and O/C become lower giving values closer to those of coal. Hydrochar is
57 mostly used as a fuel because of its enhanced properties, but other applications have been
58 studied. Considering their highly developed surface area, after further activation they can be
59 used as adsorbents or material for capacitors. Post reaction liquid has to be purified and can be
60 utilised as fertilizer or can be returned to the beginning of the process and employed as input
61 water.
62 Scientists have widely investigated different kinds of biomass and wastes. Kambo et al.
63 studied lignocellulosic material (Miscanthus giganteus) within a 190-260°C temperature range
64 and used short periods of time from 5 to 30 minutes [5,6]. Pine was used by Q. Wu et al. and
65 investigated at 180-240°C for 2, 4 and 6 hours [23]. Sevilla et al. examined cellulose between
66 200 and 250°C with residence times of 2 and 4 hours [8]. Simsir et al. performed a
67 comparative study of different saccharides and wood chips at 200°C but with reaction times
68 ranging from 6 to 48 hours [9]. Interest in lignin-cellulosic biomass originates from the fact
69 that those polymers, together with hemicellulose, decompose at these reactions temperatures
70 when the pressure is elevated. However, other types of biomass are also of interest. Oliveira
71 et al. performed experiments with agricultural residue [10]. Mäkelä et al. hydrothermally
72 treated food processing waste (180-260°C for 30 minutes) [24]. Similarly, Ruksathamcharoen

2
 73 et al. used empty fruit bunch pellets in the 10m3 reactor, inserting saturated steam at 2MPa
74 pressure for 30 minutes and with temperatures ranging from 150-180°C [17]. Other possible
75 applications include different types of sewage, eg. municipals or sludge [25–29]. HTC enables
76 easier water release from waste and handles problems of biodegradation and disinfection as
77 well as homogenisation for transport, storage and energy utilisation purposes. Wilk et al. have
78 hydrothermally treated Miscanthus giganteus, other lignocellulosic biomass, and sewage
79 sludges. The application of various temperatures between 180-220°C and reaction times from
80 1 to 12 hours clearly showed the greater effect produced by the process temperature in
81 comparison to the residence time. In the case of sewage sludge they even suggested that a time
82 frame exceeding 4 hours can decrease the value of the solid product [11,30]. Magdziarz et al.
83 have previously examined Virginia mallow under HTC conditions using 220°C for 4 hours.
84 Their main focus was to perform pyrolysis of hydrochars to prove the advantages of combined
85 pretreatment methods [31].
86 All the researchers above have come to similar conclusions. It is known that hydrothermal
87 treatment produces energy-dense solid fuel with properties comparable to brown coal. The
88 resulting hydrochars are more homogeneous, they have better milling properties and they
89 show a hydrophobic character. It is worth underlining that the most important advantage of
90 HTC is that treated material can be wet and it does not require drying. Another advantage of
91 this process is flexibility, it handles organic and inorganic waste very well. However, a certain
92 drawback is the need for additional treatment for the liquid phase before it can be further
93 utilised. Despite numerous publications there are still some unanswered questions.
94 In this study, the hydrothermal carbonization of Virginia mallow, an energy crop, was
95 conducted. The main goal of this research is to investigate the effects of temperature and
96 reaction time on energy and mass yields of hydrochars. Furthermore, how their
97 physicochemical properties change in regard of various working conditions and how it affects
98 the profitability of this pretreatment method. In order to establish the energy balance, the
99 usage of electricity during the HTC process was controlled. Proximate and ultimate analysis,
100 together with high heating value measurements were used to establish the chemical energy
101 content of biomass and hydrochars. TGA, SEM and FTIR were performed with the aim of
102 gaining further insight into the hydrothermal mechanism and changes in the surface of the
103 material and its combustion characteristics. Taking into consideration the fact that it is already
104 being grown in Poland it is important to look for new ways of utilising such a promising plant.
105
106 2. Materials and methods

107 2.1. Material

108 Virginia mallow is an example of one of the most popular energy crops in Poland [32].
109 Therefore, this was the reason it was selected for this study. The plantations of this perennial
110 plant in the Polish climate can last up to 20 years [33]. The biggest advantages of Virginia
111 mallow are its rather low requirements for soil and an optimal climate. In view of those facts,
112 it can be used for the reclamation of degraded areas (post-mine dumps, communal dumps) and
113 can be cultivated on so-called wastelands. Depending on conditions, between 6 to 17 tons of
114 dry biomass can be obtained from each hectare per year with an average calorific value of 17-
115 18 MJ/kg [14,33,34]. The biomass comes from domestic crops. It was firstly shredded but
116 required further chopping and milling. This was conducted by using knife mill LN-100
117 Testchem available at the Centre of Energy, AGH UST. The prepared material consisted of
118 particles smaller than 1mm. Its physical and chemical characteristics can be found in Table 1.

3
119 2.2. Hydrothermal carbonization
120 Hydrothermal carbonization (HTC) was conducted in a stainless steel (316) ZipperClave®
121 Stirred Reactor from The Parker Autoclave Engineers. It is equipped with a MagneDrive
122 Agitator and cooling coil. It can withstand temperatures up to 232°C and pressure up to 15.1
123 MPa. The reactor volume is 1000 ml. A control panel is used to set the stirrer speed and
124 temperature. A scheme of the apparatus can be found in [11].
125 For every experiment, 25 g of Virginia mallow was immersed in 200 ml of boiled water and
126 left for at least 12h for the material to be completely soaked. Before inserting into the reactor,
127 the mixture was weighed and the water volume was raised to 375 ml to achieve 1:15 biomass
128 to water mass ratio. In two cases a higher amount of water was used to achieve a 1:20 mass
129 ratio. To ensure repeatable composition of feedstock, the mixture was prepared using dried
130 biomass and distilled water in suitable proportions in order to prevent the biodegradable
131 process in raw biomass and its negative impact on the results. The experiments were carried
132 out at three different temperatures: 180°C, 200°C and 220°C. Each of them were held for 60
133 minutes. Moreover, further testing was conducted at 200°C for 30 and 90 minutes,
134 respectively. Additionally, to emphasize residence time importance, one sample was held for 5
135 minutes at a temperature of 220°C and another sample (1:20 mass ratio and 200°C) was held
136 for 240 minutes. Usually heating and cooling periods took about an hour each. After the
137 reactor reached an ambient temperature, it could be opened. Then the mixture was removed
138 and separated by vacuum filtration to dewater the solid products. Afterwards, the hydrochars
139 were inserted into the dryer for 24h at 105°C and after drying they were stored in sealed
140 containers for further analysis.
141
142 2.3. Analytical methods for solid materials

143 2.3.1. Ultimate analysis

144 The ultimate analysis was employed a few times to put a valuation on carbon, hydrogen
145 and nitrogen content in samples prior to and after hydrothermal carbonization. Analysis
146 was made using the Elemental Analyser Truspec LECO CHN628.
147
148 2.3.2. Proximate analysis
149 The proximate analysis was performed according to European Standards. EN ISO 18134-2
150 was used to determine the moisture content. 2g of the sample was held in the dryer at
151 105±2°C for at least 4h to achieve constant mass. EN ISO 18122 was used to measure the
152 ash content. In the first step, 2g of sample was heated for 30 minutes in a muffle furnace
153 up to 250°C and held there for 1h. After that, the sample was further heated for 30 minutes
154 up to 550°C and held for another 2h. The volatile matter content was determined as
155 described in EN ISO 18123. 1g of biomass was inserted in the furnace at 900°C for 7
156 minutes.
157
158 2.3.3. Thermogravimetric analysis
159 Thermogravimetric analysis (TGA) was conducted for four different samples to observe
160 the influence of the temperature of the carbonization process on the thermal stability of
161 materials. Biomass was heated from 20 up to 700°C with a 10°C/min heating rate in the
162 atmosphere of air. Weight loss, temperature and heat effects were recorded in order to
163 create a thermogravimetric (TG) measurement, the first derivative of thermogravimetric
164 (DTG) and differential scanning calorimetry (DSC) curves.

4
165 2.3.4. Scanning electron microscope analysis
166 Scanning electron microscope analysis (SEM) was used to determine changes in the
167 structure and surface of hydrochars after carbonization. For research purposes,
168 a FEI Inspect S50 microscope was used. The analysis was conducted in a low vacuum
169 with detection of backscattered electrons.
170
171 2.3.5. Fourier-transform infrared spectroscopy analysis
172 Fourier infrared spectroscopy (FTIR) measurements were made in order to observe
173 changes in the functional groups after hydrothermal carbonization under different
174 conditions. Analysis was conducted with a Bruker Optics-Vertex70V Fourier’s
175 spectrometer. Samples for testing were prepared using the KBr pellet method. Absorption
176 spectra were recorded at 128 scans and 4 cm-1 resolution in the 4000-400 cm-1 range.
177
178 2.3.6. Structural surface analysis
179 To investigate changes in the specific surface area raw, Virginia mallow and one of the
180 hydrochars was analysed using the multipoint BET adsorption method (ASAP 2010,
181 Micromeritics Inst.). Samples were degassed for 12h at 100°C prior to testing, as an
182 adsorbate nitrogen was used and adsorption took place at 77 K. Additionally, for the same
183 samples, pore size distribution was investigated by means of the mercury porosimetry
184 method via Poremaster 60 porosimeter (Quantachrome Inst.).
185
186 2.3.7. High heating value determination
187 High heating values (HHV) of samples were determined using two methods:
188 computational and experimental. The computational method used a Friedl’s equation [35]
189 based on ultimate analysis results. The other was conducted using a calorimeter LECO
190 AC500.
191

192 3. Results and discussion

193 There were two main products of hydrothermal carbonization. The first one being solid
194 hydrochars which achieved a visibly darker, browner colour. Particle sizes were smaller and
195 the scent was similar to roasted coffee. The second one was a liquid which also had a dark
196 brown, almost black colour. After separation of these two phases, the liquid fraction was clear
197 but not transparent due to its dye. The pH was measured by indicator paper and recorded
198 within a range of 3 to 4. For easy differentiation of samples, a system of symbols based on the
199 conditions of HTC was introduced:
200
_ : _ _
201 where:
202 – biomass symbol, (Virginia Malow)
203 : – biomass to water mass ratio (eg. 25g biomass : 375g water => 1:15)
204 – temperature of hydrothermal carbonization (eg. 200°C => 200)
205 – residence time in reactor given in minutes
206
207 One of the liquid samples, namely VM_1:20_200_240, was analyzed by liquid
208 chromatography to determine its composition. Besides water the following compounds, acetic
209 aldehyde (0.08%), acetic acid (0.31%) and furfural (0.12%) were found. Two other

5
210 compounds were detected, but due to the small amount (below 0.01%) they were
211 unidentifiable. Further results of the hydrochars analysis are presented below.

212 3.1. Proximate and ultimate analysis

213 The atomic composition of samples is presented in Table 1. The influence of both reaction
214 time and temperature are visible in the changes in carbon content. For VM_1:15_220_60, it
215 reached the highest value of 54.5% in comparison to 44.9% for raw Virginia mallow, which is
216 an increase of 21.4%. Setting the temperature at 220°C gave better results than prolonging the
217 reaction time: the carbon content for VM_1:15_200_90 was 52.6% and for
218 VM_1:20_200_240, it was 53.3%. On the contrary, the contents of hydrogen and oxygen
219 decreased with a rise of temperature or reaction time. Oxygen content dropped from 46.78%
220 for raw biomass to 38.39% for VM_1:15_220_60, and the hydrogen content for these samples
221 were 6.02% and 5.67%, respectively. This, of course, has resulted in molar ratios of O/C and
222 H/C which are presented in Figure 1. Similar trends were reported earlier [6,36].
223 Hydrothermal carbonization also has an impact on ash, volatile matter and the moisture
224 content of hydrochars. A decrease in ash and volatile matter content implies that some of the
225 compounds were transferred to a liquid or gaseous phase. The lowest ash content, equal to
226 0.26%, was achieved for VM_1:20_200_240 (1.97% for raw Virginia mallow) and the most
227 significant change in volatile matter content was for VM_1:20_200_60 sample with value of
228 78.02% in comparison to 85.50% input feedstock. All these properties resulted in higher Fixed
229 Carbon (FC) parameter. Its value changed from 5.63% up to 16.29% for VM_1:15_200_90.
230 The same sample has also the highest fuel ratio, which may represent combustibility of solid
231 fuel [28,36]. Even though the hydrochars were dried at 105°C after every hydrothermal
232 experiment, further handling of samples showed around 4 to 5% moisture content. Other
233 important factors include mass yield and energy densification ratio. Together, with a rise of
234 severity under HTC conditions mass yield decreased reaching a value as low as 55.96% for
235 VM_1:15_220_60, but at the same time, the mentioned ratio was the highest at 1.22.

6
236 Table 1. Fuel properties of raw Virginia mallow and hydrochars
1:15 1:20
Raw
180°C 200°C 220°C 200°C
VM
60 min 30 min 60 min 90 min 5 min 60 min 60 min 240 min

Cdb (%) 44.9 49.7 51.8 52.5 52.6 51.8 54.5 51.9 53.3
db
H (%) 6.02 5.77 5.81 5.78 5.79 5.77 5.67 5.82 5.99
Odb (%) 46.78 42.72 40.96 40.21 40.18 40.84 38.39 39.44 40.13
db
N (%) 0.33 0.41 0.39 0.46 0.39 0.38 0.39 0.58 0.32
Molar ratio O/C 0.78 0.64 0.59 0.57 0.57 0.59 0.53 0.57 0.56
Molar ratio H/C 1.61 1.39 1.35 1.32 1.32 1.34 1.25 1.35 1.35

Ashdb (%) 1.97 1.40 1.04 1.05 1.04 1.21 1.05 2.26 0.26
db
Volatile matter (%) 85.50 83.84 80.71 80.19 78.45 80.71 78.34 78.02 79.00
ad
Moisture (%) 6.90 5.01 4.22 4.34 4.22 3.95 4.57 4.24 4.72
Fixed Carbon (%) 5.63 9.75 14.03 14.42 16.29 14.13 16.04 15.48 16.02
Fuel ratio 0.07 0.12 0.17 0.18 0.21 0.18 0.20 0.20 0.20

Mass yield (%) - 65.88 61.00 60.90 60.36 59.20 55.96 62.85 70.00
Energy yield (%) - 72.94 70.64 71.58 71.05 68.50 68.32 73.08 84.00
Energy densification
- 1.11 1.16 1.17 1.18 1.16 1.22 1.16 1.18
ratio

HHV (kJ/kg) 17 798 19 707 20 612 20 912 20 951 20 593 21 729 20 685 20 950
LHV (kJ/kg) 16 482 18 447 19 345 19 652 19 689 19 335 20 494 19 416 19 670
Chemical energy (kJ) 445.0 324.6 314.3 318.5 316.2 304.8 304.0 325.2 366.6
Electrical energy (kJ) - 2700 2700 3240 3780 2618 3690 3240 6480
237 db – dry basis, ad – air dried
238 By difference: O% = 100% - C% - H% - N% - Ash%
239 By difference: Fixed Carbon% = 100% - Volatile matter% - Ash% - Moisture%
240 Fuel ratio = Fixed carbon / Volatile matter
241 Mass yield = mass of hydrochar / mass of feedstock
242 Energy densification ratio = HHV of hydrochar / HHV of feedstock
243 Energy yield = Mass yield × Energy densification ratio
244 Chemical energy = mass of hydrochar × HHV of hydrochar
245 Electrical energy is calculated as total electrical energy consumption for heating up to and holding reaction temperature
246
247

7
248 Molar ratios of oxygen and hydrogen to carbon can be found on Van Krevelen’s diagram
249 which is presented in Figure 1. Reactions which occurs during hydrothermal carbonization
250 lead to changes in those ratios of hydrochars. Dehydration and decarboxylation lower the
251 hydrogen and oxygen content in the subsequent products. As a result, the properties of
252 biomass are changed toward the direction of coal. The presented hydrochars had parameters
253 close to that of lignite coal (H/C = 0.8 – 1.3 and O/C = 0.2 – 0.38) [31]. Fuel with low H/C
254 and O/C ratios is favourable because of the reduced energy loss, smoke and water vapour
255 during the combustion process [28,37]. As can be seen in the diagram, the most significant
256 factor is the temperature used in the process – sample VM_1:15_220_60 has the lowest values
257 of O/C and H/C ratios, 0.53 and 1.2, respectively. Comparison with previous authors’ studies
258 on hydrothermally treated Virginia mallow (1:20, 220°C, 240 minutes) showed that
259 temperature and residence time have significant impact on hydrochars. The higher the
260 temperature and longer residence time, the lower the molar ratios of H/C and O/C [31].

261
262 Figure 1. Van Krevelen’s diagram

263 3.2. Thermogravimetric analysis

264 Curves received from thermogravimetric analysis (TGA) of hydrochars and biomass, which
265 are presented in Figure 2, allowed for investigation of the combustion process. The ignition
266 temperature (Ti) was determined by using the intersection method, and burnout temperature
267 (Tf) was detected at the point where mass loss rate decreased to below 1%/min [38,39]. Based
268 on the collected data, combustibility indexes were calculated according to equations given by
269 Mureddu et al. [40]. Ignition index (Di), burnout index (Df), combustion index (S) and
270 combustion stability index (Hf) informs us how easily combustion occurs and how it occurs
271 with an increase in temperature. Two stages of this process could be distinguished. The first
272 one starts after volatile matters begin to be released and ignition occurs around 300°C. In the

8
273 second stage, char is being combusted and this ends between 450 and 500°C. After that no
274 further change in mass occurs.

275

276 Figure 2. Results of thermogravimetric analysis for a) raw Virginia mallow, b) VM_1:15_180_60,
277 c) VM_1:15_200_60, d) VM_1:15_220_60, e) comparison of TG curves, f) comparison of DSC
278 curves

9
279 Table 2. The combustion characteristics parameters of raw Virginia mallow and selected hydrochars
1:15
Raw
180°C 200°C 220°C
VM
60 min 60 min 60 min

Ti (°C) 294 313 310 306

Tf (°C) 455 493 496 511
T1 (°C) 322 326 324 321
DTG1 (%/min) -16.99 -26.95 -22.32 -19.28
T2 (°C) 428 459 456 466
DTG2 (%/min) -3.93 -3.61 -3.83 -3.65
DTGmean (%/min) -1.47 -1.48 -1.48 -1.47
3 -2
Di (%/min ·10 ) 2.14 3.10 2.62 2.32
4 -4
Df (%/min ·10 ) 4.67 6.51 5.46 4.66
2 2 3 -7
S (% /(min ·°C ) ·10 ) 6.41 8.26 6.93 5.92
2 4
Hf (%/(min·°C ) ·10 ) 1.56 2.27 1.91 1.69
280 DTGmean – mean combustion rate
281 DTG1 – maximum combustion rate
282 Di = DTG1 / (t1 · ti), t1 – corresponding time for DTG1, ti – ignition time
283 Df = DTG1 / (t1 · tf · t0.5), tf – burnout time, t0.5 - time range of DTG/DTG1 = 0.5
284 S = (DTG1 · DTGmean) / (Ti2 · Tf)
285 Hf = T1 · ln(t0.5 / DTGmean)
286
287 As can be seen in Table 2, after hydrothermal carbonization there was a change in the
288 combustion parameters. Ignition temperature rose from 294°C for untreated Virginia mallow
289 to 313°C for a sample originally from 180°C. It is connected to a decrease in volatile matter
290 content after the HTC process. On the other hand, a decrease in this temperature with an
291 increase in the reaction temperature was an unexpected correlation, which may be connected
292 to uncertainty in the determination method as the difference does not exceed more than a few
293 degrees Celsius. The burnout temperature moved from 455°C to 511°C for the most severe
294 conditions. With T1 (temperature of the first DTG peak) being almost unchanged it meant that
295 the second stage of combustion was elongated. This is consistent with proximate and ultimate
296 analyses as the fixed carbon content increased. Hydrothermal carbonization also affected the
297 combustibility indexes. An increase in its values point to easier combustion and, overall, better
298 combustion properties. The lowest numbers received for VM_1:15_220_60 suggests that a
299 higher amount of hemicellulose and cellulose decomposed [28].

10
300 3.3. Scanning electron microscope analysis
301 Results of the scanning electron microscope (SEM) are shown in Figures 3 and 4. A fiftyfold
302 magnitude does not allow us to clearly distinguish samples. Many small particles visible in
303 Figure 3d can probably be associated with the highest temperature of thermal treatment and
304 thus the highest degradation of the sample. However, in Figure 3a, where untreated biomass is
305 presented, such small particles are also visible.

a) b)

c) d)

306
307 Figure 3. SEM results of analyzed samples at x50 magnitude:
308 a) raw Virginia mallow, b) VM_1:15_180_60, c) VM_1:15_200_60, d) VM_1:15_220_60

11
309 At two hundred fold magnitude it is easier to spot the changes. On the surface of the
310 hydrochar obtained at 180°C, there are regular, cuboid holes. From analysis of other images it
311 is inferred that with an increase of temperature, these holes become more irregular and
312 longitudinal. These changes can be connected with the course of reaction and the creation of
313 products after hydrothermal carbonization. Release of the gaseous phase may lead to a
314 degradation of the surface. In considering that, it is possible to achieve a higher specific
315 surface area, which is also one of the desired results of HTC.

a) b)

c) d)

316 Figure 4. SEM results of analyzed samples at x200 magnitude:

317 a) raw Virginia mallow, b) VM_1:15_180_60, c) VM_1:15_200_60, d) VM_1:15_220_60

12
318 3.4. Fourier-transform infrared spectroscopy analysis

319
320 Figure 5. FTIR spectra for Virginia mallow a) as received and hydrothermally treated at
321 b) 180°C 1:15 1h, c) 200°C 1:15 1h, d) 220°C 1:15 1h, e) 200°C 1:20 1h, f) 200°C 1:20 4h
322
323 Fourier-transform infrared spectroscopy analysis (FTIR) spectra can be found in Figure 5 and
324 Table 3. Bands between 2950-2850 cm-1 indicate a methylene group and relay information
325 about the organic origin of the sample. Similarly, bands in the range of 800-900 cm-1 indicate
326 C-H in the plane bend. Comparison of the methylene band around 1470 cm-1 and the methyl
327 band at 1380 cm-1 leads to the conclusion that firstly the branched chain becomes a more
328 linear structure after hydrothermal carbonization. There are clearly visible aromatic ring-
329 related vibrations between 1500 cm-1 and 1600 cm-1. After carbonization these bands move
330 slightly away from each other, which may suggest a change in the substituents on the ring.
331 Also, barely visible, are bands above 2100 cm-1 pointing to the existence of a C≡C bond, but
332 signals within a range of 610 cm-1 to 680 cm-1 are associated with an alkyne C-H bend. The
333 broad band between 3200 cm-1 and 3400 cm-1 clearly shows the existence of a hydroxyl group
334 in the compound. Bands around 1260-1350 cm-1 also indicate an OH in-plane bend. Esters are
335 with a C-O-C bond resulting in absorption signals at 1070-1240 cm-1. An aldehyde group is
336 associated with bands at 1740 cm-1 and 2700-2800 cm-1. Bands at 1030-1070 and exactly at
337 446, point toward a skeleton vibration of alkanes. Altogether, vibrations from 900 to 1250
338 indicate polysaccharides. Taking into account the origin of the samples, it is clear that those
339 polysaccharides are lignin, cellulose and hemicellulose in the structure [41,42].

13
340 Table 3. Indicators bands their vibration and corresponding functional groups.
Functional group or
Location wavenumber (cm-1) Vibration
component
Bonded and non-bonded
3400–3200 O–H stretch
hydroxyl groups and water
2950–2850 C–H stretch aliphatic methylene group
2800–2700 H–C=O: C–H stretch aldehydes
>2100 –C≡C– stretch alkynes
1740–1700 C=O stretch aldehydes
1600–1500 C=C stretch Aromatics
1485–1445 =CH2 stretch methylene
1380–1370 –CH3 bend methyl
1350–1260 O–H in plane bend Hydroxy compounds
1240–1070 C–O–C stretching Esters
1070–1030, 446 –C–C– alkanes
1250–900 C–O–C Polysaccharides
900–800 C-H in-plane bend aromatics
680–610 C–H bend alkyne
341

342 3.5. Structural surface analysis

343 The specific surface area of raw Virginia mallow determined by the BET method was
344 0.945 m2/g and it increased to 3.689 m2/g for sample VM_1:20_200_240. However,
345 even if the value is four times higher, it is still a relatively low result. Pore analysis
346 showed that both samples consisted of pores with diameters from 0.5 μm to 50 μm,
347 but a higher pore volume was detected for hydrochar. The porosimeter calculated the
348 contribution of detected macropores to surface area as 2.25 m2/g, which indicates that
349 there were smaller pores. Unfortunately, they were below the detection limit of the
350 applied method.

14
351 3.6. Energy balance
352 Measurements of electricity usage for conducting each experiment were evaluated by an
353 Electricity consumption meter type DT21, DPM Solid Polska. Determined high heating values
354 were used to calculate chemical energy carried in the received hydrochars. Mass and energy
355 yields are presented in Figure 6. Mass yields below 70% (55.96% for VM_1:15_220_60)
356 contribute to low energy yields. A trend can be seen that with an increase in temperature, both
357 yields decrease in contrast to the energy densification ratio and high heating value which are
358 the highest for hydrochar from 220°C – 1.22 and 21 729 kJ/kg, respectively in contrast to
359 untreated Virginia mallow with HHV equal to 17 798 kJ/kg.

360
361 Figure 6. Mass yield, energy yield and energy densification ratio for received hydrochars.
362
363 Consumption of electrical energy is shown in Figure 7. It can be clearly seen that the input of
364 energy is much higher than the output that could be received. In the worst scenario, the usage
365 of energy at 200°C maintained over a 4 hour period is 17 times higher than the energetic effect
366 which can be achieved with combustion of the hydrochar. The profitability of the laboratory
367 setup can be questioned but it ensures easy preparation of hydrochars from different materials
368 under diverse conditions.

15
369
370 Figure 7. Chemical energy bound in samples, electrical energy used to prepare hydrochars and
371 high heating value of examined samples.

372 4. Conclusions
373 The detailed physical and chemical characteristics of hydrochars derived from Virginia
374 mallow compared to a raw sample has been carried out. The presented results outline the
375 valuable combustible properties of Virginia mallow hydrochars. The insightful study of TGA,
376 proximate and ultimate analysis, has confirmed their carbonaceous nature and ability to
377 combust. The multivariate tests were also supported by SEM and FTIR results which
378 underlined the highest impact of temperature mainly responsible for the significant changes in
379 structure caused by the decomposition of hemicellulose and cellulose. The hydrochars derived
380 at 220°C for 5 min exhibited comparable combustible properties as those obtained at 200°C
381 for 60 min. The results confirmed that the solid biofuel produced by HTC can successfully be
382 used in energy production. The data suggests consideration of other important outcomes of
383 hydrothermal treatment such as a reduction of mass, an increase in energy densification ratio
384 or better combustion stability.

385 Acknowledgment
386 This work was supported by the Polish Ministry of Science and Higher Education [grant AGH
387 no 11.11.110.663]. The authors gratefully acknowledge EKOPROD Ltd. the proprietor of the
388 HTC apparatus used in the presented study. A major part of the experimental work was done
389 during the realization of the MSc thesis “Energy analysis of the biomass upgrading process”
390 defended in 2018 under supervision of Małgorzata Wilk, PhD.

16
391 References
392 [1] U.S. Energy Information Administration, Monthly Energy Review May 2019, 2019.
393 https://doi.org/10.1044/leader.ppl.24052019.24.
394 [2] M. Banja, R. Sikkema, M. Jégard, V. Motola, J.F. Dallemand, Biomass for energy in the EU – The
395 support framework, Energy Policy. 131 (2019) 215–228. https://doi.org/10.1016/j.enpol.2019.04.038.
396 [3] European Parliament, Directive 2009/28/EC of the European Parliament and of the Council on the
397 promotion of the use of energy from renewable sources and amending and subsequently repealing
398 Directives 2001/77/EC and 2003/30/EC, Off. J. Eur. Union. 52 (2009) 16–62.
399 [4] European Parliament, Directive (EU) 2018/2001 of the European Parliament and of the Council on the
400 promotion of the use of energy from renewable sources, Off. J. Eur. Union. 2018 (2018) 82–209.
401 [5] H.S. Kambo, A. Dutta, Comparative evaluation of torrefaction and hydrothermal carbonization of
402 lignocellulosic biomass for the production of solid biofuel, Energy Convers. Manag. 105 (2015) 746–
403 755. https://doi.org/10.1016/j.enconman.2015.08.031.
404 [6] H.S. Kambo, A. Dutta, Strength, storage, and combustion characteristics of densified lignocellulosic
405 biomass produced via torrefaction and hydrothermal carbonization, Appl. Energy. 135 (2014) 182–191.
406 https://doi.org/10.1016/j.apenergy.2014.08.094.
407 [7] M. Mihajlović, J. Petrović, S. Maletić, M.K. Isakovski, M. Stojanović, Z. Lopičić, S. Trifunović,
408 Hydrothermal carbonization of Miscanthus × giganteus: Structural and fuel properties of hydrochars and
409 organic profile with the ecotoxicological assessment of the liquid phase, Energy Convers. Manag. 159
410 (2018) 254–263. https://doi.org/10.1016/j.enconman.2018.01.003.
411 [8] M. Sevilla, A.B. Fuertes, The production of carbon materials by hydrothermal carbonization of cellulose,
412 Carbon N. Y. 47 (2009) 2281–2289. https://doi.org/10.1016/j.carbon.2009.04.026.
413 [9] H. Simsir, N. Eltugral, S. Karagoz, Hydrothermal carbonization for the preparation of hydrochars from
414 glucose, cellulose, chitin, chitosan and wood chips via low-temperature and their characterization,
415 Bioresour. Technol. 246 (2017) 82–87. https://doi.org/10.1016/j.biortech.2017.07.018.
416 [10] I. Oliveira, D. Blöhse, H.G. Ramke, Hydrothermal carbonization of agricultural residues, Bioresour.
417 Technol. 142 (2013) 138–146. https://doi.org/10.1016/j.biortech.2013.04.125.
418 [11] M. Wilk, A. Magdziarz, Hydrothermal carbonization, torrefaction and slow pyrolysis of Miscanthus
419 giganteus, Energy. 140 (2017) 1292–1304. https://doi.org/10.1016/j.energy.2017.03.031.
420 [12] J. Fang, L. Zhan, Y.S. Ok, B. Gao, Minireview of potential applications of hydrochar derived from
421 hydrothermal carbonization of biomass, J. Ind. Eng. Chem. 57 (2018) 15–21.
422 https://doi.org/10.1016/j.jiec.2017.08.026.
423 [13] M. Sieradzka, N. Gao, C. Quan, A. Mlonka-Mędrala, A. Magdziarz, Biomass thermochemical
424 conversion via pyrolysis with integrated CO2 capture, Energies. 13 (2020) 1050.
425 https://doi.org/10.3390/en13051050.
426 [14] S. Szwaja, A. Magdziarz, M. Zajemska, A. Poskart, A torrefaction of Sida hermaphrodita to improve
427 fuel properties. Advanced analysis of torrefied products, Renew. Energy. 141 (2019) 894–902.
428 https://doi.org/10.1016/j.renene.2019.04.055.
429 [15] A. Mlonka-Mędrala, A. Magdziarz, T. Dziok, M. Sieradzka, W. Nowak, Laboratory studies on the
430 influence of biomass particle size on pyrolysis and combustion using TG GC/MS, Fuel. 252 (2019) 635–
431 645. https://doi.org/10.1016/j.fuel.2019.04.091.
432 [16] L. Wang, Y. Chang, A. Li, Hydrothermal carbonization for energy-efficient processing of sewage
433 sludge: A review, Renew. Sustain. Energy Rev. 108 (2019) 423–440.
434 https://doi.org/10.1016/j.rser.2019.04.011.
435 [17] S. Ruksathamcharoen, T. Chuenyam, P. Stratong-on, H. Hosoda, T. Sesillia, K. Yoshikawa, Effects of
436 hydrothermal treatment and pelletizing temperature on physical properties of empty fruit bunch pellets,
437 Energy Procedia. 158 (2019) 681–687. https://doi.org/10.1016/j.egypro.2019.01.184.
438 [18] A. Funke, F. Ziegler, Hydrothermal carbonization of biomass: A summary and discussion of chemical
439 mechanisms for process engineering, Biofuels, Bioprod. Biorefining. 4 (2010) 160–177.
440 https://doi.org/10.1002/bbb.198.

17
441 [19] T. Wang, Y. Zhai, Y. Zhu, C. Li, G. Zeng, A review of the hydrothermal carbonization of biomass waste
442 for hydrochar formation: Process conditions, fundamentals, and physicochemical properties, Renew.
443 Sustain. Energy Rev. 90 (2018) 223–247. https://doi.org/10.1016/j.rser.2018.03.071.
444 [20] S. Román, B. Ledesma, A. Álvarez, C. Coronella, S.V. Qaramaleki, Suitability of hydrothermal
445 carbonization to convert water hyacinth to added-value products, Renew. Energy. 146 (2020) 1649–
446 1658. https://doi.org/10.1016/j.renene.2019.07.157.
447 [21] A. Gallifuoco, L. Taglieri, A.A. Papa, Hydrothermal carbonization of waste biomass to fuel : A novel
448 technique for analyzing experimental data, Renew. Energy. (2019).
449 https://doi.org/10.1016/j.renene.2019.10.121.
450 [22] L. Li, J.R.V. Flora, N.D. Berge, Predictions of energy recovery from hydrochar generated from the
451 hydrothermal carbonization of organic wastes, Renew. Energy. 145 (2020) 1883–1889.
452 https://doi.org/10.1016/j.renene.2019.07.103.
453 [23] Q. Wu, S. Yu, N. Hao, T. Wells, X. Meng, M. Li, Y. Pu, S. Liu, A.J. Ragauskas, Characterization of
454 products from hydrothermal carbonization of pine, Bioresour. Technol. 244 (2017) 78–83.
455 https://doi.org/10.1016/j.biortech.2017.07.138.
456 [24] M. Mäkelä, C.W. Kwong, M. Broström, K. Yoshikawa, C. Kwong Wai, M. Broström, K. Yoshikawa,
457 Hydrothermal treatment of grape marc for solid fuel applications, Energy Convers. Manag. 145 (2017)
458 371–377. https://doi.org/10.1016/j.enconman.2017.05.015.
459 [25] N. Gao, Z. Li, C. Quan, N. Miskolczi, A. Egedy, A new method combining hydrothermal carbonization
460 and mechanical compression in-situ for sewage sludge dewatering : Bench-scale verification, J. Anal.
461 Appl. Pyrolysis. 139 (2019) 187–195. https://doi.org/10.1016/j.jaap.2019.02.003.
462 [26] B. Lokahita, M. Aziz, K. Yoshikawa, F. Takahashi, Energy and resource recovery from Tetra Pak waste
463 using hydrothermal treatment, Appl. Energy. 207 (2017) 107–113.
464 https://doi.org/10.1016/j.apenergy.2017.05.141.
465 [27] T.M. Ismail, K. Yoshikawa, H. Sherif, M.A. El-salam, Hydrothermal treatment of municipal solid waste
466 into coal in a commercial Plant : Numerical assessment of process parameters, Appl. Energy. 250 (2019)
467 653–664. https://doi.org/10.1016/j.apenergy.2019.05.042.
468 [28] Y. Lin, X. Ma, X. Peng, S. Hu, Z. Yu, S. Fang, Effect of hydrothermal carbonization temperature on
469 combustion behavior of hydrochar fuel from paper sludge, Appl. Therm. Eng. 91 (2015) 574–582.
470 https://doi.org/10.1016/j.applthermaleng.2015.08.064.
471 [29] M. Liu, A. Cardoso, L. Zhao, Y. Duan, Effect of the Amount of Sludge on Physicochemical Properties
472 and Chemical Structure of Low-rank Coal under Hydrothermal Conditions, Asia-Pacific J. Chem. Eng.
473 12 (2017) 755–764. https://doi.org/10.1002/apj.2115.
474 [30] M.M. Wilk, A. Magdziarz, K. Jayaraman, M. Szymańska-Chargot, I. Gökalp, Hydrothermal
475 carbonization characteristics of sewage sludge and lignocellulosic biomass . A comparative study,
476 Biomass and Bioenergy. 120 (2019) 166–175. https://doi.org/10.1016/j.biombioe.2018.11.016.
477 [31] A. Magdziarz, M. Wilk, M. Wądrzyk, Pyrolysis of hydrochar derived from biomass – Experimental
478 investigation, Fuel. 267 (2020) 117246. https://doi.org/10.1016/j.fuel.2020.117246.
479 [32] A. Jezierska-Thöle, R. Rudnicki, M. Kluba, Development of energy crops cultivation for biomass
480 production in Poland, Renew. Sustain. Energy Rev. 62 (2016) 534–545.
481 https://doi.org/10.1016/j.rser.2016.05.024.
482 [33] Z. Pszczółkowska, Agata; Romanowska-Duda, Zdzisława; Pszczółkowski, Wiktor; Grzesik,
483 Mieczysław; Wysokińska, Biomass Production of Selected Energy Plants: Economic Analysis and
484 Logistic Strategies, Comp. Econ. Res. 15 (2012) 77–103. https://doi.org/10.2478/v10103-012-0018-6.
485 [34] G. Šiaudinis, A. Jasinskas, E. Šarauskis, D. Steponavičius, D. Karčauskiene, I. Liaudanskiene, The
486 assessment of Virginia mallow (Sida hermaphrodita Rusby) and cup plant (Silphium perfoliatum L.)
487 productivity, physico-mechanical properties and energy expenses, Energy. 93 (2015) 606–612.
488 https://doi.org/10.1016/j.energy.2015.09.065.
489 [35] A. Friedl, E. Padouvas, H. Rotter, K. Varmuza, Prediction of heating values of biomass fuel from
490 elemental composition, Anal. Chim. Acta. 544 (2005) 191–198.
491 https://doi.org/10.1016/j.aca.2005.01.041.
492 [36] O.O.D. Afolabi, M. Sohail, Y.-L. Cheng, Optimisation and characterisation of hydrochar production

18
493 from spent coffee grounds by hydrothermal carbonisation, Renew. Energy. 147 (2020) 1380–1391.
494 https://doi.org/10.1016/j.renene.2019.09.098.
495 [37] Z. Liu, A. Quek, S. Kent Hoekman, R. Balasubramanian, Production of solid biochar fuel from waste
496 biomass by hydrothermal carbonization, Fuel. 103 (2013) 943–949.
497 https://doi.org/10.1016/j.fuel.2012.07.069.
498 [38] G.K. Parshetti, S. Kent Hoekman, R. Balasubramanian, Chemical, structural and combustion
499 characteristics of carbonaceous products obtained by hydrothermal carbonization of palm empty fruit
500 bunches, Bioresour. Technol. 135 (2013) 683–689. https://doi.org/10.1016/j.biortech.2012.09.042.
501 [39] J.J. Lu, W.H. Chen, Investigation on the ignition and burnout temperatures of bamboo and sugarcane
502 bagasse by thermogravimetric analysis, Appl. Energy. 160 (2015) 49–57.
503 https://doi.org/10.1016/j.apenergy.2015.09.026.
504 [40] M. Mureddu, F. Dessì, A. Orsini, F. Ferrara, A. Pettinau, Air- and oxygen-blown characterization of coal
505 and biomass by thermogravimetric analysis, Fuel. 212 (2018) 626–637.
506 https://doi.org/10.1016/j.fuel.2017.10.005.
507 [41] J. Coates, Interpretation of Infrared Spectra, A Practical Approach, Encycl. Anal. Chem. 36 (2006)
508 10815–10837. https://doi.org/10.1002/9780470027318.a5606.
509 [42] M. Asadieraghi, W.M.A. Wan Daud, Characterization of lignocellulosic biomass thermal degradation
510 and physiochemical structure: Effects of demineralization by diverse acid solutions, Energy Convers.
511 Manag. 82 (2014) 71–82. https://doi.org/10.1016/j.enconman.2014.03.007.
512

19
• HTC of Virginia mallow at 180, 200, 220°C and 5, 30, 60, 90, 240 min were studied.
• Impact of residence time, water to biomass ratio and temperature was discussed.
• Upgrading of hydrochars, energy and mass yields were determined.
• Fuel combustion performances by TGA and combustibility indexes were illustrated.
• SEM and FTIR results underlined the strong impact of temperature.
Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: