Journal of Environmental Chemical Engineering 10 (2022) 107540

Contents lists available at ScienceDirect

Journal of Environmental Chemical Engineering

journal homepage: www.elsevier.com/locate/jece

Hydrothermal co-carbonization of industrial biowastes with lignite toward

modified hydrochar production: Synergistic effects and
structural characteristics
Hao Zhan a, d, *, Shihui Zhang b, Yanpei Song c, Guozhang Chang e, Xinming Wang d,
Zhiyong Zeng a, **
a
School of Energy Science and Engineering, Central South University, Changsha 410083, People’s Republic of China
b
Hunan Rice Research Institute, Hunan Academy of Agricultural Sciences, Changsha 410125, People’s Republic of China
c
Yellow River Engineering Consulting Co., Ltd., Zhengzhou 450003, People’s Republic of China
d
State Key Laboratory of Organic Geochemistry, Guangdong Key Laboratory of Environmental Protection and Resources Utilization, Guangzhou Institute of
Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, People’s Republic of China
e
State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering, Ningxia University, Yinchuan 750021, People’s Republic of China

A R T I C L E I N F O A B S T R A C T

Editor: Apostolos Giannis Attempts to prepare high-grade and clean solid fuels from biowastes and low-rank coals are essential for pro­
tecting the environment and utilizing the energy from low-grade resources. The improvement of hydrothermal
Keywords: carbonization (HTC) and co-carbonization (co-HTC) on the upgrading and denitrogenation capabilities of in­
Hydrothermal co-carbonization (co-HTC) dustrial biowastes with low-rank coal as lignite (LC) was evaluated at 120–300 ◦ C. The results demonstrated that
Upgrading
coupled upgrading and denitrogenation occurred with a more intense effect on industrial biowastes than LC
Denitrogenation
during the HTC process. For co-HTC in the prevailing hydrolysis (180 ◦ C) or polymerization (240 ◦ C) stage, an
Industrial biowastes
Lignite coal optimal mixing ratio of industrial biowaste/LC of 1:1–3:1 could result in an enhanced experimental calorific
Synergy value, energy recovery efficiency, nitrogen removal efficiency, and weakened experimental nitrogen content,
demonstrating the significant positive synergies on both upgrading and denitrogenation capabilities. The cor­
responding synergistic coefficients were maximized at 7% and − 23% for the calorific value and nitrogen content
of hydrochars, respectively. The combined analyses of XPS, 13C NMR, and FTIR could provide evidence that these
synergies of co-HTC were intrinsically associated with LC: (1) its more stable carbon and nitrogen functionalities;
(2) promotion of its components or reaction sites on prevailing hydrolysis and polymerization reactions at
corresponding temperature stages.

1. Introduction energy have been shifting to low-rank coal such as lignite because of the
large consumption of high-rank coal [4]. Therefore, the co-firing of in­
Today, biowaste co-firing in existing coal-fired power plants is dustrial biowaste and lignite is predicted to be a promising alternative
globally applied because it is a helpful way to reduce the emissions of solution for future heat and power generation. Despite the advantages of
sulfur dioxide (SO2) and carbon dioxide (CO2) due to its low sulfur co-firing in energy conservation and emission reduction, there are still
content and carbon neutrality, and it enables waste disposal via efficient some similar unfavorable features restricting the large-scale application
energy recovery [1,2]. With the sustainable eco-civilization centering on of most industrial biowastes and lignite, such as their high levels of
industrialization development, environmental protection, and fossil-fuel inherent moisture, high oxygen content, low energy density, and
utilization in China, several new trends in the energy utilization of abundant fuel-bound nitrogen [5,6]. Given this situation, suitable pre­
biowaste and coal have emerged and received increasing attention: (1) treatment techniques targeting upgrading and denitrogenation for the
industrial biowastes have shown great potential as solid fuels due to production of modified solid fuel are indispensable for the pretreatment
their abundant organic matter [3]; (2) coal deposits aiming to supply of both industrial biowaste and lignite before their firing or co-firing.

* Corresponding author at: School of Energy Science and Engineering, Central South University, Changsha 410083, People’s Republic of China.
** Corresponding author.
E-mail addresses: isaaczhanhao@163.com (H. Zhan), zengzhiyong@csu.edu.cn (Z. Zeng).

https://doi.org/10.1016/j.jece.2022.107540
Received 20 November 2021; Received in revised form 21 February 2022; Accepted 8 March 2022
Available online 10 March 2022
2213-3437/© 2022 Elsevier Ltd. All rights reserved.
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

Hydrothermal carbonization (HTC) occurs at a moderate tempera­ of industrial biowaste and lignite.
ture under autogenous pressure in subcritical water, and has been As a laboratory study, the co-HTC of two different feedstocks to
extensively confirmed as an effective method of upgrading waste produce solid fuels has been gaining more attention in recent years.
biomass [7,8] and low-rank coal [9]. Continuous studies have elucidated Several studies centering on nitrogen-rich biowastes and conventional
the reactions and pathways of upgrading low-grade fuels through HTC. lignocellulosic biomass investigating the influence of co-HTC on
The upgrading of industrial biowastes by HTC mainly involves two hydrochar properties, which are characterized by the interactions be­
processes: dewatering and coalification. Between them, coalification is tween two feedstocks, have been reported [30–33]. Zhang et al. [30] and
regarded as the more critical reaction, as it consists of two opposite but He et al. [32] found that a significant synergistic enhancement occurred
sequential reactions that occur during HTC: hydrolysis/degradation and in the increased upgrading indices, including hydrochar yield, organic
condensation/aromatization. The former cleaves the bonds of functional retention, carbon retention, and calorific value, in the co-HTC of sewage
groups (ester, ether, carboxyl, and hydroxyl) in biomacromolecules, sludge with lignocellulosic biomass at a designated mixing ratio. This
while the latter would facilitate the formation of stable coal-like struc­ synergy of the upgrading properties was also validated by Lang et al.
tures [10]. According to several studies [11–13], HTC was expressed as a [31], who replaced sewage sludge with swine manure. Zheng et al. [33]
process simulating natural coalification, and it succeeded in converting carried out the co-HTC of two nitrogen-rich biowastes and indicated that
industrial biowastes to high-quality biofuels with excellent combustion the mass and energy densities of hydrochar from sewage sludge rose to a
behaviors. In our previous studies [3,14], it was identified that higher level with the addition of food waste. In regard to coal and bio­
carbon-containing structures in industrial biowastes could be trans­ waste, Nonaka et al. [5] proposed that polymerization was synergisti­
formed from low-energy bonds (-C-H/-O/=O) to high-energy bonds cally promoted during co-HTC of low-rank coal and biomass, acquiring
(aromatic -C-C/=C) via hydrolysis, carbonization and polymerization an upgraded and hydrophobic solid product. Saba et al. [28] evaluated
during HTC, producing more energy-efficient hydrochars. The removal the synergistic effect of miscanthus biomass on bituminous coal during
of oxygen-containing functional groups (hydroxyl, carboxyl, carbonyl, their co-HTC and observed that hydrochar had lower ash/sulfur contents
and methoxyl) and some aliphatic volatiles (methylene and methyl) and higher energy density. Furthermore, Kim et al. [34] and Song et al.
from low-rank lignite is believed to be the key to its upgrading [15,16]. [35] investigated the co-HTC of sewage sludge with peat and lignite
The reduction of these hydrophilic groups could efficiently decrease the coals, respectively. The synergistic improvements in fuel properties such
equilibrium moisture content [17] and inhibit the moisture as calorific value, coalification degree, and fuel ratio were similarly
re-adsorption performance [18], together with the enhancement of obtained under expected operational conditions. As revealed by these
organic carbon species and the loss of some inorganic matter [19], thus studies on biomass, biowastes, and coals, it is evident that the co-HTC
improving the lignite quality. technique has potential and advantages in producing more favorable
In addition, HTC has been reported for the removal or stabilization of solid fuels by upgrading organic species and removing harmful ele­
nitrogen functionalities in industrial biowastes [20–22]. Some conclu­ ments, which are becoming recognized by scholars as means of evalu­
sions were obtained regarding the evolution pathways of nitrogen ating the effective utilization of moist biowaste and/or coal blends.
functionalities in raw feedstocks: (1) unstable functionalities were However, previous studies on industrial biowaste and lignite had some
released into the liquid phase in the form of inorganic nitrogen ions obvious limitations.
(such as NH4+) and soluble organic nitrogen species via hydrolysis or
deamination reactions; (2) stable functionalities were converted into (1) Although denitrogenation pretreatment is vital for the clean en­
heterocyclic-N bound in the solid phase via condensation and cycliza­ ergy utilization of nitrogen-rich feedstocks, there have been no
tion reactions. Mo et al. [23] proposed that high-temperature HTC could studies on both the upgrading and denitrogenation properties
cause a partial transformation of nitrogen functional groups from lignite during the co-HTC process. These factors have been mentioned by
into the liquid phase, resulting in coal fuels that are cleaner than their few scholars [30,32,33] from the perspective of nitrogen immo­
precursors. According to studies by Cen’s group [24,25], HTC pre­ bilization during the co-HTC of biowaste/biomass blends, and the
treatment was able to alter the content and type of nitrogen function­ synergistic denitrogenation behaviors were also neglected.
ality in lignite, limiting the release of typical NOx precursors during the (2) The majority of studies have merely focused on sewage sludge,
subsequent thermal transformation process. Consequently, HTC pre­ which is different from other nitrogen-rich industrial biowastes
treatment was believed to be conducive to NOx emission reduction of with high ash content, low organic matter, and complex nitrogen
low-rank coal combustion. functionality [14,36]. Sample diversity on industrial biowastes
All the studies mentioned above have proven that HTC has the ability should be further broadened for their better energy utilization.
to produce modified solid fuels from either industrial biowaste or lignite (3) Evolution of carbon and nitrogen functionalities coexisted during
by upgrading low-grade fuels and promoting denitrogenation. Hydro­ the HTC process and pursued their respective distinctive path­
thermal co-carbonization (Co-HTC) treatment is suitable for the prepa­ ways at different temperature ranges for both industrial biowaste
ration of two feedstocks, such as industrial biowaste and lignite, to [14,22] and lignite [15,25]. In other words, the corresponding
produce a mixed modified hydrochar for the firing process instead of co- upgrading and denitrogenation characteristics determined by the
firing. However, the following challenges remain to be solved: (1) the evolution pathways might be associated with hydrothermal
nonuniformity of co-firing caused by the typical differences in density temperature. For the co-HTC of industrial biowastes and lignite,
and particle size [26] and (2) the poor energy efficiency of co-firing due the relevant synergistic effects and the properties of the corre­
to the sharply different calorific values of the feedstocks [27]. Further­ sponding hydrochars under specific operational conditions are
more, co-HTC has exhibited some additional advantages for biowaste still unclear, which needs to be further elucidated.
and coal over conventional HTC. The dehydration and decarboxylation
reactions can create acidic conditions during the HTC of biowaste, Given this situation, the objective of this work was to elucidate the
facilitating the removal of harmful elements while upgrading coal [28]. synergistic effects of industrial biowaste and lignite on both the
The metal elements or porous structures of coal provide potential upgrading and denitrogenation properties during co-HTC. The structural
condensation sites [29], possibly promoting the characteristics of the corresponding hydrochars were also examined to
condensation-polymerization reactions that determine hydrochar for­ acquire cleaner, more efficient solid fuels from biowaste/coal blends. A
mation during the upgrading of biowaste. Evidently, these coupling lignite coal (LC) from typical Chinese coal-producing area of ‘Energy
interactions indicate the formation of better modified hydrochar when Golden Delta’ was employed. Two representatives of industrial bio­
the co-HTC of biowaste and coal is used, suggesting the feasibility of wastes, Chuanxiong herbal waste (CHW) and penicillin mycelial waste
co-HTC as an effective pretreatment for the expected energy utilization (PMW), were reasonably selected due to the following reasons [3,13,

2
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

14]. (1) massive initial moisture but substantial organic matter (dry difference (FC=100-VM-Ash; O=100-C-H-N-S-Ash). (3) Calorific value
basis); (2) high nitrogen content with big gradient; (3) distinct (higher heating value, HHV) was determined using the reliable fitting
biochemical components (lignocellulose and non-lignocellulose types). formula proposed by literature [37] (HHV=0.3491⋅C+1.1783⋅H+
The key purposes of this study were to: (1) identify the characteristics of 0.1005⋅S-0.1034⋅O-0.0151⋅N-0.0211⋅Ash). (4) Ash components were
upgrading and denitrogenation in the HTC of each feedstock to ascertain analyzed by an AxiosmAX-PETRO XRF spectrometer (PANalytical B.V.,
the optimal and decisive influencing factors; (2) explore the effect of the Netherlands). (5) Crude fiber content was determined based on GB/T
mixing ratio on the co-HTC variables involving process variables (en­ 6434–2006; contents of protein and lipid were measured according to
ergy recovery efficiency, nitrogen removal efficiency) and state vari­ the Kjeldahl method (GB 5009.5–2016) and Soxhlet extraction method
ables (higher heating value, nitrogen content) in different characteristic (GB 5009.6–2016), respectively; carbohydrate content was calculated
temperature ranges to explicate the synergistic upgrading and deni­ by difference (Carbohydrate=100-Protein-Lipid-Ash). In addition,
trogenation behaviors; and (3) compare the chemical structures of cor­ proximate, ultimate and calorific value analyses of hydrochars were
responding hydrochar, especially in terms of their carbon and nitrogen determined using the same testing methods for feedstocks.
functionalities via the combination of XPS, solid-state 13C NMR, and XPS analysis was used to characterize the surface chemistry (carbon
FTIR analytic techniques to further evaluate the synergistic effects of and nitrogen functionalities) of each raw feedstock and their respective
co-HTC. hydrochars. The C 1 s and N 1 s spectra were recorded by an ESCALAB
250Xi XPS spectrometer (Thermo VG Scientific, USA) equipped with a
2. Experimental monochromatic Al (Kα) X-ray source (150 W, 1486.68 eV) operating at a
30 eV pass energy, a 0.1 step size, a 500 µm diameter spot size and a 90◦
2.1. Materials electron take-off angle. All spectra were evaluated using Advantage
software based on the following principles: (1) binding energy calibra­
The feedstocks (CHW, PMW, and LC) employed in this work were tion using principal C 1 s peak (284.8 eV); (2) peak fitting using a
collected from local factories in Guangdong, Hebei, and Jilin provinces 1.40–1.65 eV FWHM, a 30% Gaussian-Lorentzian function and Shirley-
in China. CHW and PMW are two pharmaceutical process residues, and type background subtraction; and (3) curve resolution using specific
LC is local low-rank coal. All samples were first dried at 105 ◦ C for at types and binding energies for carbon and nitrogen functionalities, as
least 24 h, after which time they were milled into a fine powder, filtered shown in Fig. S1. The detailed XPS analysis can be found in previous
through a 60 mesh sieve, redried to ensure that the moisture-free sam­ studies [3,36].
ples had the same particle size, and then stored in a dry vacuum desic­ FTIR analysis was introduced to characterize the organic functional
cator for further analyses and experiments. groups of raw feedstocks and corresponding hydrochars under target
operational conditions. A Tensor 27 FTIR spectrometer (Bruker, Ger­
2.2. Experimental procedures many) was used for each characterization trial in which a mixture of
sample and KBr (1:100 w/w) were ground and pressed into the pellet,
HTC and co-HTC experiments were both performed in a 250 mL and immediately measured in absorbance mode ranging from 4000
tightly-sealed stainless-steel autoclave reactor with stirring capabilities cm− 1 to 400 cm− 1 with a resolution of 4 cm− 1. Prior to each measure­
(SLM250, Senlang Co. Ltd., China). During each HTC experiment, ment, the background line was established by taking KBr as the refer­
feedstock (10 g) and deionized water (100 mL) were loaded into the ence, which was subtracted from the spectra of the target sample.
reactor, which was then sparged with an inert gas (99.999% argon) for Solid-state 13C-CP-MAS NMR analysis was used to evaluate the car­
at least 10 min to remove the air. After that, the reactor was heated to bon chemistry of the solid samples (raw feedstocks and targeting
the desired reaction temperature and the temperature was maintained hydrochars). The tests were performed on an AVANCE III 300 spec­
for a fixed residence time (30 min). The stirring rate was fixed at 300 trometer (Bruker, Germany) at a resonance frequency of 75 MHz using a
rpm throughout the reaction process to ensure product homogeneity. 4 mm MAS probe with a spinning rate of 6500 Hz. The 13C spectra were
Specifically, for the HTC experiment, the reaction temperature ranged recorded under the following conditions: contact time of 2 ms, recycle
from 120◦ to 300◦ C with a 30 ◦ C interval. For the co-HTC procedure, the delay of 5 s, and scan number of 5120. The 13C chemical shifts were
industrial biowaste-to-lignite coal ratios were set to 3:1, 1:1, and 1:3, externally referenced to the carbon tert-butyl signal of solid adamantane
and the samples were labeled CHW:LC and PMW:LC. The reaction at 38.48 ppm. The curve-fitting of each spectrum was processed using
temperature was selected according to the experimental HTC results. Gaussian PeakFit of Origin 2018 software. According to previous studies
When the reaction was complete, the reactor was rapidly cooled to [3,14,15], the spectra could be assigned to six main regions: (1)
ambient temperature with a fan. The solid-liquid mixture was separated unsubstituted saturated alkyl C (alk, 0–43 ppm); (2) alkyl C substituted
by vacuum filtration with a Buchner funnel through 0.45 µm PTFE filter by O and N atoms (OCH3/NCH, 43–60 ppm); (3) carbohydrate C (O-alk,
paper. The solid product (hydrochar) was transferred, washed with 60–110 ppm), including alkyl C singly bonded to one O atom (60–93
deionized water, and then oven-dried at 105 ◦ C until a constant weight ppm) and alkyl C bonded to two O atoms (93–110 ppm); (4) aromatic C
was reached. Subsequently, the hydrochar was weighed and milled into (Ar, 110–160 ppm), including aromatic C-H (Ar-H, 110–125 ppm), ar­
a fine powder of < 180 µm for storage before further physical and omatic C-C (Ar-C, 125–145 ppm) and aromatic C-O (Ar-O, 145–160
chemical characterization. If necessary, the liquid product was collected ppm); (5) carboxyl, ester and amide C (COO/N-C– – O, 160–190 ppm);
and stored in a refrigerator below − 20 ◦ C until analysis. To minimize and (6) aldehyde and ketone C (C– – O, 190–220 ppm).
the experimental error, each run was performed in triplicate so the
average and standard deviation could be taken. 2.4. Formula and calculation

2.3. Physical and chemical characterization There are four important indicators for evaluating the upgrading and
denitrogenation properties during HTC: the energy densification ratio
Basic fuel properties of feedstocks were determined according to the (EDR) and energy recovery efficiency (ERE) for the former, and the ni­
testing methods described in our previous studies [14,36]. (1) Elemental trogen retention ratio (NRR) and nitrogen removal efficiency (NRE) for
compositions were determined using a Vario EL cube CHNS analyzer the latter. These indicators can be calculated using Eqs. (1–4).
(Elementar Co., Ltd., Germany). (2) Two proximate components, vola­ /
EDR = HHVhydrochar HHVfeedstock (1)
tile matter (VM) and Ash, were measured in a conventional muffle
furnace according to ASTM D3175, D3174, E872 and D1102, respec­ /
NRR = Nhydrochar Nfeedstock (2)
tively; both the FC (fixed carbon) and O contents were calculated by

3
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

ERE = EDR∙ωhydrochar (3) 3. Results and discussion

NRE = 100-NRR∙ωhydrochar (4) 3.1. Basic fuel properties of raw feedstocks

where HHV and N are the high heating value and nitrogen content of Prior to discussion, the fuel properties of raw feedstock samples were
the corresponding solid sample, respectively. ωhydrochar is the hydrochar initially evaluated, which was presented in Table 1. As shown, CHW and
yield which was determined by Eq. (5). PMW are both suitable representatives of N-rich industrial biowastes
/ based on the following physicochemical properties: (1) their abundant
ωhydrochar = Weighthydrochar Weightfeedstock ∙100% (5)
organic matter of 80–90 wt%, including 40–45% C, 6–7% H, and
Carbon retention (CR) and organic retention (OR) were introduced to 30–40% O, ensured an excellent calorific value (HHV) of 18–20 MJ/kg
improve the understanding of hydrochar production, and their calcu­ on a dry basis (equal to forest residues), suggesting their better energy
lation is given in Eqs. (6–7). utilization; (2) their higher nitrogen content with great gradients and
/ distinctive biochemical and ash components covered the diversity of N-
CR = Chydrochar Cfeedstock ∙ωhydrochar (6)
rich industrial biowastes. In addition, the LC contained moderate vola­
/ tile matter (~50 wt%) and less ash (~ 6 wt%) and had a higher calorific
OR = (100-Ashhydrochar ) (100-Ashfeedstock )∙ωhydrochar (7) value (~24 MJ/kg), indicating that it was a lump of superior lignite coal.

where C and Ash are the carbon and ash contents of the corresponding
solid sample, respectively. 3.2. Upgrading and denitrogenation properties during the HTC process
To explore the synergistic effects of the co-HTC treatment, compar­
isons between the experimental value (EV) and the calculated value 3.2.1. Changes of upgrading and denitrogenation indicators during the HTC
(CV) were performed; these comparisons focused on the process vari­ process
ables of co-HTC (ERE and NRE) and the state variables of hydrochar Temperature is a main determinant of reaction intensity, element
(HHV and nitrogen content). Based on the assumption that no in­ transformation, and functional group evolution during the HTC process
teractions occur in the co-HTC of two different feedstocks, the CV was [38]. Herein, we first discussed the variation of each indicator relating to
determined by the physical mixture of hydrochars from two feedstocks upgrading and denitrogenation with the hydrothermal temperature for
at a target mixing ratio. Taking the mixing ratio of 3:1 (CHW: LC) as an HTC of industrial biowaste and lignite, as illustrated in Fig. 1.
example, the CV of each variable was calculated as follows. It was observed that the upgrading and denitrogenation indicators all
For state values (HHV, nitrogen content), the calculation is shown in present significant distinctions between industrial biowaste and low-
Eq. (8). rank coal. For LC, both the ERE and NRE showed a slight variation
CVstate = 0.75∙statehydrochar, CHW + 0.25∙statehydrochar, LC (8) with increasing temperature, remaining at approximately 85% and 10%,
respectively, when 300 ◦ C was reached. This phenomenon also corre­
For process values (ERE, NRE), the calculation was performed using sponds to the small decreases in ωhydrochar, CR, and OR (Fig. 1e). The
their respective equations mentioned above. The CV of the involved above observation can be explained by the fact that efficient hydro­
state variables were first confirmed according to Eq. (8). For example, thermal treatment of coals requires a higher temperature due to their
the ERE was calculated with Eq. (9): strong hydrothermal stability, which was reported by Mursito et al.
/
CVERE = CVHHVhydrochar CVHHVfeedstock ∙CVωhydrochar (9) [39], who observed that higher temperatures (> 250 ◦ C) could promote
the efficient hydrothermal conversion of coal. It is evident that HTC
CVHHVhydrochar = 0.75∙HHVhydrochar, CHW + 0.25∙HHVhydrochar, LC (9.1) treatment (120–300 ◦ C) led to limited energy loss and minor nitrogen
removal for the LC employed in this work. As the temperature increased,
CVHHVfeedstock = 0.75∙HHVfeedstock, CHW + 0.25∙HHVfeedstock, LC (9.2) a slight increase in the EDR was observed, as shown in Fig. 1c, which
may be caused by the loss of oxygen-containing groups [15,18]; the NRR
CVωhydrochar = 0.75∙ωhydrochar, CHW + 0.25∙ωhydrochar, LC (9.3) value initially showed a negligible decrease, and then it increased
slightly after 210 ◦ C due to the reduction of nitrogen-containing,
To quantitively explicate the synergistic effect, another parameter carboxyl or hydroxyl groups in different temperature ranges (Fig. 1d)
called the synergistic coefficient (SC) was defined as Eq. (10). [16]. Therefore, it was confirmed that HTC treatment below 300 ◦ C had
/ a finite effect on upgrading and denitrogenating the LC samples in this
SCstate or process = (EVstate or process − CVstate or process ) CVstate or process ∙100% (10)
work.
Regarding industrial biowastes, a significant decrease in the ERE and

Table 1
Properties of raw industrial biowastes and lignite coal.
Sample Proximate analysisdb (wt%) HHV Ultimate analysisdb (wt%) Biochemical component analysisdb (wt%)

VM FC Ash (MJ/kg) C H N S O Protein Carbohydrate Lipid Crude fiber

CHW 73.19 11.26 17.68 6.12 2.17 38.35 11.88 74.13 2.73 13.70
15.55 42.10 0.00
PMW 79.72 8.16 19.77 6.66 8.14 32.53 26.25 57.74 7.85 1.91
12.12 44.22 0.29
LC 52.23 6.25 24.39 5.26 0.88 26.76 n.a. n.a. n.a. n.a.
41.52 60.31 0.54
Sample Ash component analysis (%, expressed in form of oxides)
SiO2 P2O5 K2 O Al2O3 MgO Fe2O3 TiO2 SO3 Na2O
CaO
CHW 35.98 10.17 13.42 10.66 9.49 8.80 4.86 0.64 4.73 0.38
PMW 0.70 36.14 23.80 0.22 0.75 28.49 0.65 0.02 7.94 0.61
LC 28.71 0.34 1.01 12.60 2.16 21.21 15.42 0.65 17.12 0.11

db-dry basis; n.a.-not applicable.

4
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

Fig. 1. Changes in different variables vs. the temperature during the HTC process of each feedstock: (a-d): ERE, NRE, EDR, NRR; (e): auxiliary variables- ωhydrochar,
OR, CR.

an obvious increase in the NRE were observed in Fig. 1a and Fig. 1b for revealed in a previous study [14]. The distinctive ash components like
two samples at elevated temperatures from 120 ◦ C to 300 ◦ C, respec­ calcium, ferrum or magnesium in the two samples (Table 1) have some
tively. The tread of each curve can be characterized into two stages with effects on these two properties [7].
the dividing point between 180 and 210 ◦ C, exhibiting a more dramatic
rate at low temperatures than at high temperatures. According to pre­ 3.2.2. Evolution of carbon functionalities during the HTC process
vious studies [10,38], two series of opposite reactions occur during the To further explain the upgrading properties obtained above, carbon
HTC of biomass: (1) hydrolysis, dehydration, decarboxylation, and functionalities in feedstocks and corresponding hydrochar products
deamination; and (2) aromatization, cyclization, and condensation were identified by the XPS technique, and the C 1 s spectra were ac­
polymerization. The reactions in the first series break the linkage bonds quired and accurately peak-fitted in Fig. S2. By integrating the peak area
in biomacromolecule groups and convert some monomers or oligomer in each XPS spectrum, changes in relative content of carbon function­
fragments into the gas/liquid phases, leading to the loss of organic alities in solid phase during the HTC were depicted in Fig. 2.
substances in the solid phase and serving as the main pathway at low As shown in Fig. 2, for carbon functionalities, nonoxygen-containing
temperatures. The reactions in the second series achieve the agglomer­ groups (C– – C/C-C/C-H) account for a higher proportion than oxygen-
ation and integration of active intermediates by strong solid-solid or containing groups for samples, especially for LC. With increasing tem­
solid-liquid interactions to form secondary coal-like structures, which is perature, the predominant C– – C/C-C/C-H decreases first and then in­
favorable for superior energy-density solid fuels and more dominant at creases, showing more significance for industrial biowastes than lignite.
high temperatures. Due to the competitiveness of two opposite reaction This phenomenon is ascribed to two varying pathways of C– – C/C-C/C-H
series in different temperature ranges, the evidential results for HTC on considering the overlapping XPS peaks [3,40]: (1) the depolymerization
upgrading and denitrogenating industrial biowastes can be summarized of aliphatic C-C/C-H via hydrolysis; and (2) the formation of aromatic
as follows: (1) HTC is an energy-loss process (increased ERE in Fig. 1a) C-C/C– – C via condensation polymerization. The former is more domi­
consistent with other decreased auxiliary indicators in Fig. 1e, though it nant at low temperatures (leading to the resultant decrease), which is
can be used to upgrade solid products at high temperatures (enhanced largely associated with the energy loss of HTC. In contrast, the latter
EDR in Fig. 1c). The EDR decreased as the temperature increased from becomes more intense when the temperature goes up high (resulting in a
270◦ to 300 ◦ C probably due to some prevailing hydrothermal lique­ constant increase), which corresponds to the upgrading capability of
faction reactions at overly high temperatures, suggesting the importance HTC due to the formation of aromatic C-C/C– – C bonds, which are in a
of suitable HTC temperature on biowaste upgrading. (2) HTC is also an higher energy state. In addition, C-(O, N) groups, which are present in
efficient nitrogen-removal process (increased NRE in Fig. 1b) but results alcohol, ether, phenol, and protein species, are the main
in uncertainty regarding the relative content of nitrogen retained in oxygen-containing carbon functionalities, which exhibit a more obvious
hydrochar (irregular NRR in Fig. 1d), depending on the balance of two decline for two industrial biowastes than LC during the HTC process. The
reaction series in relation to nitrogen-containing groups [7]. Comparing other two oxygen-containing carbon functionalities (C– – O/O-C-O,
the two samples, PMW has higher values in both the ERE and NRE than O–– C-O) were observed at a low fraction and with a visible reduction
CHW, leading to a lower EDR and NRR, which also coincides with the with increasing temperature. It was inferred that the energy loss of HTC
corresponding values of ωhydrochar, CR, and OR. This finding is supported is caused by a combined decrease in oxygen-containing carbon func­
by evidence that PMW (non-lignocellulose type) contains more unstable tionalities with a stronger effect of C-(O, N) than C– – O, leading to a
biochemical components than CHW (lignocellulose type), which was reduction in the O/C ratio in the solid phase. This finding is mostly

5
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

Fig. 2. Comparisons of fraction of each carbon functionality in the solid phase (feedstocks and corresponding hydrochars).

consistent with those of previous studies [40,41]. In general, it can be feedstocks, as is a small amount of inorganic-N (N-IN). In the corre­
concluded that the transformation of nonoxygen-containing carbon sponding hydrochars formed in the temperature range of 120–300 ◦ C,
functionalities and the removal of oxygen-containing carbon function­ the number of N-A groups is dramatically lower, and N-IN vanishes
alities with HTC progression are mainly responsible for the upgrading of rapidly; additionally, different heterocyclic-N like pyrrolic (N-5), pyr­
solid products and the energy loss during HTC, respectively, which is idinic (N-6), and quaternary (N-Q) groups emerge and increase
more intense for industrial biowastes than LC. observably. The disappearance of N-IN is attributed to its dissolution or
hydrolysis into the hydrolysate in the form of NH4+ and NO3- ions with
3.2.3. Evolution of nitrogen functionalities during the HTC process the help of subcritical H2O [42]; the sustained reduction of N-A would
To further explicate the above denitrogenation properties, nitrogen be explained by two mechanisms: (1) labile N-A is hydrolyzed and
functionalities in feedstocks and corresponding hydrochar products degraded into low-molecular weight organic nitrogen in the liquid
were identified by the XPS technique, and the N 1 s spectra were ac­ phase, such as free amino acids and peptide fragments, simultaneously
quired and accurately peak-fitted in Fig. S3. By integrating the peak area producing inorganic NH4+ by deamination reactions [13,20]; (2) stable
in each XPS spectrum, changes in relative content of nitrogen func­ N-A is converted to N-heterocyclic compounds in the solid phase by
tionalities in solid phase during the HTC were depicted in Fig. 3. direct cyclization or dimerization reactions [21,43]. It is easy to
Regarding nitrogen functionalities, a distinction in categories is conclude that the denitrogenation of the HTC process originates from
observed between industrial biowaste and lignite. N-A, referring to the dissolution of N-IN and the deamination of N-A. The above two re­
amide, amino, and amine groups, is predominant in industrial biowaste action routes are predominant at low temperatures, leading to a

Fig. 3. Comparisons of fraction of each nitrogen functionality in the solid phase (feedstocks and corresponding hydrochars).

6
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

dramatic increase in the NRE below 180–210 ◦ C, as shown in Fig. 1b. their fractions. A lower N-6 than N-Q suggests that nitrogen at the edges
When the hydrothermal temperature is high, free nitrogen intermediates of the graphene layer is less significant. In addition, the change in N-X
(amino acids, peptide fragments) in the liquid phase probably react with (increase first and then decrease) with temperature probably depends on
reducing sugars via Maillard reactions, forming liquid heterocyclic-N, the balance between oxidization and repositioning effects in relation to
which would be further polymerized to solid heterocyclic-N (N-5, N-6, edge-N [25]. By comparing the changes in nitrogen content and func­
and N-Q) by cyclization or ring-condensation reactions [42,44]. tionality during HTC of lignite, it can be concluded that the intrinsic
Consequently, high temperature is favorable for the formation of an denitrogenation of hydrothermal modification (< 300 ◦ C) on lignite
aromatic structure with nitrogen incorporation via solid-solid reaction mainly lies in nitrogen stabilization instead of nitrogen removal in the
and liquid-solid transformation, immobilizing more stable nitrogen in solid phase.
hydrochars. This result corresponds to the slower increase of NRE shown In general, upgrading and denitrogenation during the conventional
in Fig. 1b and the more remarkable increasing fraction of N-heterocyclic HTC process (< 300 ◦ C) have these following characteristics: (1)
structures in Fig. 3 for two industrial biowastes within the higher tem­ simultaneously coupled behavior; (2) distinctive indicator tread and
perature range. In addition, compared to PMW, CHW presents a weaker intrinsic motivation identified by hydrothermal temperature; and (3) a
denitrogenation effect involving a lower NRE (Fig. 1b), higher NRR remarkable effect on industrial biowastes while extremely finite influ­
(Fig. 1d), and higher N-heterocyclic fraction (Fig. 3a), which is probably ence on LC. Due to these features, 180 and 240 ◦ C were selected as the
because CHW contains more stable N-A types than PMW [22,36]. representative temperatures for the prevailing hydrothermal hydrolysis
By comparison, the LC sample only contains N-heterocyclic groups and polymerization stages of industrial biowastes, respectively. By
involving N-5 (main), N-6, N-Q, and (oxide-N) N-X, and presents a focusing on their co-HTC with LC at two typical temperatures, upgrading
steady decrease in the N-5 fraction as well as a mild increase in the N-6 and denitrogenation behaviors were further separately investigated in
and N-Q fractions as the HTC temperature is raised. It is commonly the following two sections.
accepted that the HTC of lignite coal is an upgrading process at a mo­
lecular level and is associated with the rupture of functional groups, the
condensation of macromolecules, and interactions with the aqueous 3.3. Synergistic upgrading behavior during the co-HTC process
system. According to previous studies [25,43,45], the evolution of
coal-nitrogen functionalities during the HTC included (1) ring expansion First, HHV was employed as the state variable and ERE was
of pyrrole (N-5) and hydroxyl loss of pyridone (N-5) to form N-6; (2) employed as the process variable to indicate the upgrading behavior
condensation of macromolecules to the aromatic graphene layer that during the co-HTC. With the determination of two values (EV, CV) in
altered the nitrogen conformations located at different positions of the relation to HHV and ERE via Eqs. (3, 8, 9), their changes with the mixing
layer: ‘edge-N′ (N-6), ‘valley-N′ (N-Q) and ‘center-N′ (N-Q); and (3) ratio of industrial biowaste/LC during co-HTC at two representative
possible oxidation of edge-N to form N-X because the edge position was temperatures are illustrated in Fig. 4.
more susceptible to the hydrothermal environment. Notably, the con­ It was similarly observed that both the HHV of hydrochar and the
stant decrease in the N-5 fraction shown in Fig. 3c is attributed to its ERE of the co-HTC increase constantly when the mixing ratio of indus­
direct and indirect conversion to N-6 and N-Q, leading to an increase in trial biowaste/LC was decreased at the two temperatures. This result
indicates that co-HTC of biomass with low-rank coal is an appropriate

Fig. 4. Effect of mixing ratio on upgrading variables: (a, b)-HHV; (c, d)-ERE.

7
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

option for obtaining solid biofuels with high energy density and simul­ co-HTC is mainly associated with two factors related to LC: (1) good
taneously avoiding energy loss. When co-HTC is performed at any thermal stability; (2) favorable enhancement of hydrolysis and poly­
mixing ratio, a high-temperature scenario (240 ◦ C) can maintain a merization of industrial biowastes.
higher experimental HHV while presenting a close ERE than a low-
temperature scenario (180 ◦ C). This observation suggests that the
presence of a certain proportion of LC can effectively compensate for the 3.4. Synergistic denitrogenation behavior during the co-HTC process
increasing energy loss of extensive co-HTC; moreover, hydrochar
products with higher energy quality are also ensured. Notably, low-rank In addition, nitrogen content as the state variable and NRE as the
coal was proven to be an excellent candidate for improving the fuel process variable were selected to evaluate the denitrogenation behavior
performance of high-moisture biomass via co-HTC, coinciding with during the co-HTC process. With the confirmation of two values (EV,
some views of previous studies [5,35]. Moreover, it is interesting to note CV) in relation to nitrogen content and NRE via Eqs. (4, 8, 9), their
that the EVs of the two upgrading indicators acquired from co-HTC of variations with the mixing ratio of industrial biowaste/LC during co-
the industrial biowaste/LC blend were all higher than the corresponding HTC at two representative temperatures are depicted in Fig. 5.
CVs at the two temperatures. Taking the HHV as an example, when the As seen in Fig. 5, both the nitrogen content of hydrochar and the NRE
mixing ratio was 3:1, the EVs increased by 1.4–1.5 and 0.9–1.1 MJ/kg at of the co-HTC decreased gradually with increasing LC proportion at the
180 and 240 ◦ C compared to the CVs, respectively. The optimal mixing two temperatures. Compared to industrial biowastes, the nitrogen con­
ratio was observed in the range of 3:1–1:1 with corresponding syner­ tent of the LC was quite lower (Table 1), and the structures were more
gistic coefficients of 5–7% (180 ◦ C) and 3–5% (240 ◦ C). It was therefore stable (Fig. 3). Accordingly, the gradual addition of LC caused a signif­
confirmed that the co-HTC of industrial biowastes with LC had a sig­ icant decline in nitrogen content and increased the proportion of stable
nificant positive synergistic effect on enhancing the quality of hydrochar nitrogen functionalities in blending feedstocks, whose HTC can certainly
with less energy loss in all temperature ranges. This result may be due to result in a decrease in both the NRE and nitrogen content of hydrochar.
the additional interactions between specific components of two feed­ Interestingly, at any mixing ratio for the two temperatures, the nitrogen
stocks with different reactivities and thermal stabilities [30]. In the content had obviously lower EVs than CVs, while the NRE showed the
prevailing hydrolysis stage (180 ◦ C), LC seldom participates in the re­ opposite result. The combined findings demonstrate that co-HTC of in­
action; in such cases, it is possible that stable organic species and cata­ dustrial biowastes and LC is capable of synergistically denitrogenating a
lytic inorganic elements in LC can compensate for the energy loss of solid product, and this synergy is also found to have an enhancement
hydrochar and facilitate the dehydration of macromolecules in bio­ effect, exhibiting an optimal mixing ratio ranging from 3:1–1:1 with
wastes, respectively [46]. In the prevailing polymerization stage maximal synergistic coefficients of − 23% (nitrogen content) and 67%
(240 ◦ C), in addition to its excellent energy preservation capacity, LC (NRE). Based on the nitrogen evolution that occurs during conventional
can provide potential reaction sites to promote the aggregation and HTC, which was discussed in Section 3.2.1, it is understandable that the
condensation of biowaste-hydrolyzed intermediates, resulting in denitrogenation of industrial biowastes depends on the dominance be­
enhanced polymerization to attain a higher HHV and ERE [5,28]. Hence, tween nitrogen removal (dissolution, hydrolysis, and deamination re­
we can infer that the positive synergistic upgrading performance of actions) and nitrogen incorporation (Maillard, dimerization, and ring-
condensation reactions) relating to those nitrogen functionalities; the

Fig. 5. Effect of mixing ratio on denitrogenation variables: (a, b)-nitrogen content; (c, d)-NRE.

8
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

denitrogenation of LC manifests as the main nitrogen stabilization in carbon functionalities among different solid samples are compared in
accompanying very weak nitrogen loss. In addition, the detected nitro­ Fig. 6.
gen content in LC is much lower than that in industrial biowastes, – C/C-C/C-H and C-(O, N) are two main types of
As shown in Fig. 6, C–
especially in PMW. These combined results indicate that the positive carbon functionalities in both feedstocks and hydrochar at any operating
synergistic denitrogenation effect of co-HTC is probably ascribed to the condition. Co-HTC hydrochars possess an upward C– – C/C-C/C-H frac­
promotion of nitrogen removal pathways or the inhibition of nitrogen tion and a downward C-(O, N) fraction when the added amount of LC is
immobilization routes, which is caused by the interactions of the bio­ increased at the two temperatures. The aforementioned experimental
waste/lignite blend. As previous studies demonstrated [7,42], alkaline C–– C/C-C/C-H and C-(O, N) fractions are higher and lower than their
and alkaline earth metal ions could accelerate HTC denitrogenation corresponding calculated values (blue and red dotted lines), respec­
through enhanced hydrolysis and deamination reactions. In addition, tively. These structural characteristics can further demonstrate the
Zheng et al. [33] also reported that the elevated proportion of sewage capability of the co-HTC of industrial biowastes with LC to improve their
sludge/food waste during co-HTC would inhibit the Maillard reaction by fuel quality, as the enhancement of C– – C/C-C/C-H and the reduction of
weakened carbonization and polymerization reactions. Similarly, the C-(O, N) are typical structural indicators for the upgrading capability
addition of LC would achieve two favorable effects in the system. First, [14,40,41].
the increase in helpful metal ions such as Ca2+ and Fe3+ may promote Additionally, solid-state 13C NMR analysis was employed as the
the hydrolysis of solid-N into the liquid phase (Table 1); second, the corresponding supplementary demonstration. Based on the 13C NMR
organic macromolecules in feedstocks present a more stable tread, spectra in Fig. S5, the proportion of each carbon-containing structure in
which prevents their degradation into soluble small monomers such as the solid phase was calculated and is presented in Table 2.
reducing sugars, inevitably weakening the Maillard reaction during the As shown in Table 2, essential evidence for the evolution of carbon
secondary polymerization stage. In addition, it is also implied that LC functionalities in relation to upgrading capability can be observed,
exhibits a more significant nitrogen stabilization effect with the co-HTC coinciding with the XPS results. From raw feedstock to hydrochar at
of CHW due to their similar nitrogen contents. Therefore, as Fig. 5 180 ◦ C, the dominant carbon change for industrial biowastes is the
shows, the co-HTC of CHW with LC can largely narrow the difference in decline of oxygen- or nitrogen-containing carbons with the sequence of
nitrogen content and NRE between the two temperatures, and present a COO/N-C– – O > O-Alk, accompanying the main increase of Alk carbon.
more obvious effect on the NRE at each mixing ratio, in contrast to the This finding demonstrates that hydrolysis reactions (dehydration,
co-HTC of PMW. decarboxylation, and deamination) are more prevalent than polymeri­
zation reactions (aromatization, condensation) during the moderate
HTC of industrial biowastes. When undergoing extreme HTC, proportion
3.5. Structural characteristics of modified hydrochar
of the O-Alk carbon shows a significant drop higher than that of the
COO/N-C– – O carbon; additionally, much Ar carbon is formed (~ 40%)
3.5.1. Structural characteristics of solid-carbon from the co-HTC
and the proportion of Alk carbon increases constantly. These structural
The structural characteristics of the solid phase (feedstock and
carbon changes reveal that aromatization and polymerization reactions
hydrochar product) in terms of carbon and nitrogen functionalities were
are dominant and coexist with hydrolysis-related reactions during se­
essential to further evaluate the above positive synergies of co-HTC. For
vere HTC. Compared to raw LC, hydrochar at 180 ◦ C shows no
one thing, based on the detailed XPS identification (Fig. S4), the changes

Fig. 6. Comparisons of carbon functionalities in solid samples (raw feedstocks, targeting hydrochars).

9
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

Table 2
Proportion (%) of carbon-containing structures (13C NMR) in raw feedstocks and targeting hydrochars.
Samples Alk (unsubstituted OCH3/NCH (alkyl C substituted O-Alk Ar (aromatic COO/N-C– –O (carboxyl, C–
–O (aldehyde and
saturated alkyl C) by O and N atoms) (carbohydrate C) C) ester and amide C) ketone C)

CHW-raw 8.72 4.84 69.26 6.38 10.80 /

PMW-raw 24.62 6.27 46.34 4.51 19.16 3.10
LC-raw 38.31 5.66 1.37 50.48 4.18 /
CHW-180 11.77 5.88 65.88 10.15 6.32 /
PMW-180 26.19 5.58 45.26 6.78 14.77 1.42
LC-180 40.96 5.28 1.22 50.68 1.86 /
CHW-240 23.52 2.01 29.72 40.77 1.86 2.12
PMW-240 37.05 4.45 8.47 41.13 8.90 /
LC-240 36.48 4.29 0.86 56.75 1.62 /
(CHW: 35.94 5.26 26.20 30.57 2.03 /
LC=1:1)−
180
(CHW: 32.86 3.29 12.36 51.49 / /
LC=1:1)−
240
(PMW: 37.42 5.62 18.30 28.90 9.76 /
LC=1:1)−
180
(PMW: 36.36 3.70 5.38 51.00 3.56 /
LC=1:1)−
240

distinction in the carbon fraction, supporting its good thermal stability close to the average value of two individual HTC hydrochars at mixing
at low HTC temperatures; hydrochar at 240 ◦ C exhibits a mild increase ratio of 1:1, implying the capability of stable LC carbon component on
in Ar carbon at the expense of losing some COO and Alk carbons, sug­ the energy compensation of co-HTC hydrochar. (2) The significant
gesting its excellent energy preservation capacity at high HTC temper­ reduction in O-Alk and the obvious increase in Alk mean the excessive
atures. Accordingly, as for co-HTC (taking 50% LC addition for hydrolysis of carbohydrates with the formation of some free alkyl
example), the carbon atoms in low-temperature hydrochar present structures, and the corresponding enhanced values (compared to theo­
typical characteristics. (1) Ar carbon increases to a significant fraction retical value) can demonstrate that the positive synergistic upgrading

Fig. 7. Comparisons of nitrogen functionalities in solid samples (raw feedstocks, targeting hydrochars).

10
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

performance at low temperatures is mainly attributed to the enhanced thermal stability of carbon and nitrogen functionalities in LC. In addi­
hydrolysis of carbohydrates by active components in LC. Concerning tion, FTIR spectra of industrial biowastes and their typical derived
high-temperature co-HTC hydrochars, in addition to the sustained hydrochars can also provide supplementary evidence on the conversion
decrease in O-Alk carbon, the increase in the Ar carbon fraction (almost of N-A types, as shown in Fig. 8. Compared to the feedstocks, hydrochars
51% higher than the theoretical value) is the principal characteristic, formed by HTC have weakened peaks at 3090–3180 cm− 1 (NH2 stretch)
suggesting that the aromatization reaction promoted by the interactions and 1540–1570 cm− 1 (N-H in the plane) and enhanced peaks at
of two feedstocks is responsible for the positive synergy in upgrading 1320 cm− 1 (aromatic C-N stretch) and 750–850 cm− 1 (N-H out of
indicators at high temperatures. plane). These two opposite peak categories are linked with aliphatic and
aromatic N-A, respectively [20,40,43]. This result can explain the hy­
3.5.2. Structural characteristics of solid-nitrogen from the co-HTC drothermal transformation of N-A types via deamination and dimer­
For another, based on the detailed XPS identification (Fig. S6), the ization reactions. However, those peaks relating to aliphatic and
changes in nitrogen functionalities among different solid samples were aromatic N-A in co-HTC hydrochars (e.g., mixing ratio of 1:1) weaken
compared in Fig. 7. remarkably, further demonstrating that co-HTC can promote the con­
As illustrated in Fig. 7, with increasing LC proportion, a significant version of solid N-A types, which probably tend toward liquid-N species,
reduction in N-A and an increase in some heterocyclic-N, such as N-5 achieving the denitrogenation of hydrochar products.
and N-6, are the main features of co-HTC hydrochars at the two tem­
peratures, especially at high temperature. This observation verifies that 4. Conclusions
a higher LC ratio and temperature facilitate nitrogen functionalities with
stronger thermal stability in co-HTC hydrochars. As confirmed in Sec­ Conventional HTC had a coupled upgrading and denitrogenation
tion 3.2.3, LC contains much less nitrogen and more stable nitrogen capacity to produce high-grade and low nitrogen-containing hydrochars
functionalities than industrial biowastes. Notably, the LC in the blend is from industrial biowastes and LC, exhibiting a much more intense effect
likely to be a nitrogen stabilizer with insufficient nitrogen contribution. on industrial biowastes than on LC. Additionally, each indicator refer­
When co-HTC occurs at low temperatures, hydrochar contains abundant ring to upgrading (ERE, EDR) and denitrogenation (NRE, NRR) were
N-A and some heterocyclic-N. It can be confirmed that these divided into two temperature stages at 180 and 210 ◦ C, which were
heterocyclic-N types basically originate from LC based on the calcula­ attributed to the competitive evolution pathways of corresponding
tion of nitrogen contribution. Accordingly, the prevailing pathway at carbon and nitrogen functionalities controlled by hydrolysis or poly­
this stage is the enhanced hydrolysis conversion of unstable N-A and N- merization reactions. When the co-HTC was in the prevailing hydrolysis
IN into the liquid phase, which is responsible for the positive synergy of (180 ◦ C) or polymerization (240 ◦ C) stage, an optimal addition of LC
denitrogenation indicators at low temperatures. When co-HTC is per­ (25–50%) resulted in an increase in the experimental HHV, ERE, NRE
formed at high temperatures, hydrochar still contains a certain portion and the decline in the experimental nitrogen content compared to their
of N-A in addition to the dominant heterocyclic-N, suggesting that the calculated values, presenting significant positive synergistic effects of
conversion of stable N-A into heterocyclic-N is inevitable but con­ co-HTC on both the upgrading and denitrogenation capacities. The
strained to some extent. In addition, nitrogen incorporation via the maximal synergistic coefficients were observed at 7% and − 23% for the
Maillard reaction of liquid-N is also inhibited due to the increase in more HHV and nitrogen content of hydrochars, respectively. Structural ana­
stable carbon macromolecules in the system. Hence, restrained nitrogen lyses of feedstocks and hydrochars indicated that the intrinsic motiva­
immobilization is the critical factor for the positive synergistic deni­ tions for these synergies were probably due to two factors: (1) good
trogenating capability at high temperatures. It is evident that the above thermal stability of carbon/nitrogen functionalities in LC; and (2)
two synergistic denitrogenation mechanisms are associated with the favorable hydrolysis/polymerization reactions promoted by special LC

Fig. 8. FTIR spectra of industrial biowastes and their derived hydrochars (HTC, Co-HTC).

11
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

components or reaction sites. These findings suggested that co-HTC is a [13] D.C. Ma, G.Y. Zhang, P.T. Zhao, C. Areeprasert, Y.F. Shen, K. Yoshikawa, et al.,
Hydrothermal treatment of antibiotic mycelial dreg: more understanding from fuel
potential strategy for preparing enhanced solid fuels with higher energy-
characteristics, Chem. Eng. J. 273 (2015) 147–155.
density and lower nitrogen content from industrial biowaste and low- [14] X.Z. Zhuang, H. Zhan, Y.P. Song, C. He, Y.Q. Huang, X.L. Yin, et al., Insights into
rank coal, especially from wet nitrogen-rich industrial biowastes. the evolution of chemical structures in lignocellulose and non-lignocellulose
biowastes during hydrothermal carbonization (HTC), Fuel 236 (2019) 960–974.
[15] J.H. Wu, J.Z. Liu, X. Zhang, Z.H. Wang, J.H. Zhou, K.F. Cen, Chemical and
CRediT authorship contribution statement structural changes in XiMeng lignite and its carbon migration during hydrothermal
dewatering, Fuel 148 (2015) 139–144.
[16] H. Ullah, G.J. Liu, B. Yousaf, M.U. Ali, Q. Abbas, C.C. Zhou, et al., Hydrothermal
Hao Zhan: Conceptualization, Methodology, Writing – original dewatering of low-rank coals: Influence on the properties and combustion
draft, Visualization, Investigation, Funding acquisition. Shihui Zhang: characteristics of the solid products, Energy 158 (2018) 1192–1203.
Writing – review & editing, Validation, Supervision. Yanpei Song: [17] Y.X. Zhang, J.J. Wu, Y. Wang, Z.Y. Miao, C.D. Si, X.L. Shang, et al., Effect of
hydrothermal dewatering on the physico-chemical structure and surface properties
Software, Resources, Formal analysis, Data curation. Guozhang Chang:
of Shengli lignite, Fuel 164 (2016) 128–133.
Project administration, Funding acquisition, Resources. Xinming [18] X.F. Feng, C. Zhang, P. Tan, X.P. Zhang, Q.Y. Fang, G. Chen, Experimental study of
Wang: Writing – review & editing, Project administration, Funding the physicochemical structure and moisture readsorption characteristics of
Zhaotong lignite after hydrothermal and thermal upgrading, Fuel 185 (2016)
acquisition. Zhiyong Zeng: Supervision, Project administration.
112–121.
[19] J.J. Liao, Y. Fei, M. Marshall, A.L. Chaffee, L.P. Chang, Hydrothermal dewatering
of a Chinese lignite and properties of the solid products, Fuel 180 (2016) 473–480.
Declaration of Competing Interest [20] T.F. Wang, Y.B. Zhai, Y. Zhu, C. Peng, B.B. Xu, T. Wang, et al., Influence of
temperature on nitrogen fate during hydrothermal carbonization of food waste,
Bioresour. Technol. 247 (2018) 182–189.
The authors declare that they have no known competing financial [21] X.Z. Zhuang, Y.Q. Huang, Y.P. Song, H. Zhan, X.L. Yin, C.Z. Wu, The
interests or personal relationships that could have appeared to influence transformation pathways of nitrogen in sewage sludge during hydrothermal
the work reported in this paper. treatment, Bioresour. Technol. 245 (2017) 463–470.
[22] X.Z. Zhuang, H. Zhan, Y.Q. Huang, Y.P. Song, X.L. Yin, C.Z. Wu, Denitrification and
desulphurization of industrial biowastes via hydrothermal modification, Bioresour.
Acknowledgments Technol. 254 (2018) 121–129.
[23] Q. Mo, J.J. Liao, L.P. Chang, A.L. Chaffee, W.R. Bao, Transformation behaviors of
C, H, O, N and S in lignite during hydrothermal dewatering process, Fuel 236
The authors would like to acknowledge the National Natural Science (2019) 228–235.
Foundation of China (Grant No. 51906247, 52074348, 41961144029), [24] Q. Li, Z.H. Wang, Z.M. Lin, Y. He, K. Zhang, R. Whiddon, et al., Influences of
the Guangdong Basic and Applied Basic Research Foundation (Grant No. hydrothermal modification on nitrogen thermal conversion of low-rank coals,
Energy Fuel 30 (2016) 8125–8133.
2020A1515011336), the Science and Technology Program of Guangz­ [25] Z.H. Wang, Q. Li, Z.M. Lin, R. Whiddon, K.Z. Qiu, M. Kuang, et al., Transformation
hou (Grant No. 202002030421) and the Foundation of State Key Lab­ of nitrogen and sulphur impurities during hydrothermal upgrading of low quality
oratory of High-efficiency Utilization of Coal and Green Chemical coals, Fuel 164 (2016) 254–261.
[26] M. Zulfiqar, B. Moghtaderi, T.F. Wall, Flow properties of biomass and coal blends,
Engineering (Grant No. 2019-KF-20) for financial support to this work. Fuel Process Technol. 87 (2006) 281–288.
[27] S. De, M. Assadi, Impact of cofiring biomass with coal in power plants - a techno-
economic assessment, Biomass Bioenergy 33 (2009) 283–293.
Appendix A. Supporting information
[28] A. Saba, P. Saha, M.T. Reza, Co-Hydrothermal carbonization of coal-biomass blend:
Influence of temperature on solid fuel properties, Fuel Process Technol. 167 (2017)
Supplementary data associated with this article can be found in the 711–720.
online version at doi:10.1016/j.jece.2022.107540. [29] J. Mumme, M.M. Titirici, A. Pfeiffer, U. Luder, M.T. Reza, O. Masek, Hydrothermal
carbonization of digestate in the presence of zeolite: process efficiency and
composite properties, Acs Sustain. Chem. Eng. 3 (2015) 2967–2974.
References [30] X.J. Zhang, L. Zhang, A.M. Li, Hydrothermal co-carbonization of sewage sludge
and pinewood sawdust for nutrient-rich hydrochar production: Synergistic effects
and products characterization, J. Environ. Manag. 201 (2017) 52–62.
[1] T.Y. Mun, T.Z. Tumsa, U. Lee, W. Yang, Performance evaluation of co-firing various
[31] Q.Q. Lang, Y.C. Guo, Q.F. Zheng, Z.G. Liu, C. Gai, Co-hydrothermal carbonization
kinds of biomass with low rank coals in a 500 MWe coal-fired power plant, Energy
of lignocellulosic biomass and swine manure: Hydrochar properties and heavy
115 (2016) 954–962.
metal transformation behavior, Bioresour. Technol. 266 (2018) 242–248.
[2] S. Onenc, S. Retschitzegger, N. Evic, N. Kienzl, J. Yanik, Characteristics and
[32] C. He, Z. Zhang, C.F. Ge, W. Liu, Y.T. Tang, X.Z. Zhuang, et al., Synergistic effect of
synergistic effects of co-combustion of carbonaceous wastes with coal, Waste
hydrothermal co-carbonization of sewage sludge with fruit and agricultural wastes
Manag. 71 (2018) 192–199.
on hydrochar fuel quality and combustion behavior, Waste Manag. 100 (2019)
[3] X.Z. Zhuang, H. Zhan, Y.Q. Huang, Y.P. Song, X.L. Yin, C.Z. Wu, Conversion of
171–181.
industrial biowastes to clean solid fuels via hydrothermal carbonization (HTC):
[33] C.P. Zheng, X.Q. Ma, Z.L. Yao, X.F. Chen, The properties and combustion behaviors
Upgrading mechanism in relation to coalification process and combustion
of hydrochars derived from co-hydrothermal carbonization of sewage sludge and
behavior, Bioresour. Technol. 267 (2018) 17–29.
food waste, Bioresour. Technol. (2019) 285.
[4] Z.H. Rao, Y.M. Zhao, C.L. Huang, C.L. Duan, J.F. He, Recent developments in
[34] D. Kim, S. Park, K.Y. Park, Upgrading the fuel properties of sludge and low rank
drying and dewatering for low rank coals, Prog. Energ. Combust. 46 (2015) 1–11.
coal mixed fuel through hydrothermal carbonization, Energy 141 (2017) 598–602.
[5] M. Nonaka, T. Hirajima, K. Sasaki, Upgrading of low rank coal and woody biomass
[35] Y.P. Song, H. Zhan, X.Z. Zhuang, X.L. Yin, C.Z. Wu, Synergistic characteristics and
mixture by hydrothermal treatment, Fuel 90 (2011) 2578–2584.
capabilities of co-hydrothermal carbonization of sewage sludge/lignite mixtures,
[6] J.T. Wei, M. Wang, F.C. Wang, X.D. Song, G.S. Yu, Y.R. Liu, et al., A review on
Energy Fuel 33 (2019) 8735–8745.
reactivity characteristics and synergy behavior of biomass and coal Co-gasification,
[36] H. Zhan, X.L. Yin, Y.Q. Huang, H.Y. Yuan, C.Z. Wu, NOx precursors evolving during
Int. J. Hydrog. Energy 46 (2021) 17116–17132.
rapid pyrolysis of lignocellulosic industrial biomass wastes, Fuel 207 (2017)
[7] P.T. Zhao, Y.F. Shen, S.F. Ge, Z.Q. Chen, K. Yoshikawa, Clean solid biofuel
438–448.
production from high moisture content waste biomass employing hydrothermal
[37] S.A. Channiwala, P.P. Parikh, A unified correlation for estimating HHV of solid,
treatment, Appl. Energy 131 (2014) 345–367.
liquid and gaseous fuels, Fuel 81 (2002) 1051–1063.
[8] J.T. Wei, Q.H. Guo, X.D. Song, L. Ding, A. Mosqueda, Y.R. Liu, et al., Effect of
[38] T.F. Wang, Y.B. Zhai, Y. Zhu, C.T. Li, G.M. Zeng, A review of the hydrothermal
hydrothermal carbonization temperature on reactivity and synergy of co-
carbonization of biomass waste for hydrochar formation: Process conditions,
gasification of biomass hydrochar and coal, Appl. Therm. Eng. (2021) 183.
fundamentals, and physicochemical properties, Renew. Sust. Energ. Rev. 90 (2018)
[9] J.D. Yu, C.Y. Jiang, Q.Q. Guan, J.J. Gu, P. Ning, R.R. Miao, et al., Conversion of
223–247.
low-grade coals in sub-and supercritical water: A review, Fuel 217 (2018)
[39] A.T. Mursito, T. Hirajima, K. Sasaki, Upgrading and dewatering of raw tropical
275–284.
peat by hydrothermal treatment, Fuel 89 (2010) 635–641.
[10] A. Funke, F. Ziegler, Hydrothermal carbonization of biomass: a summary and
[40] C. He, J. Zhao, Y.H. Yang, J.Y. Wang, Multiscale characteristics dynamics of
discussion of chemical mechanisms for process engineering, Biofuel Bioprod.
hydrochar from hydrothermal conversion of sewage sludge under sub- and near-
Biorefin. 4 (2010) 160–177.
critical water, Bioresour. Technol. 211 (2016) 486–493.
[11] C. He, A. Giannis, J.Y. Wang, Conversion of sewage sludge to clean solid fuel using
[41] C. Peng, Y.B. Zhai, Y. Zhu, T.F. Wang, B.B. Xu, T. Wang, et al., Investigation of the
hydrothermal carbonization: Hydrochar fuel characteristics and combustion
structure and reaction pathway of char obtained from sewage sludge with biomass
behavior, Appl. Energy 111 (2013) 257–266.
wastes, using hydrothermal treatment, J. Clean. Prod. 166 (2017) 114–123.
[12] P.T. Zhao, Y.F. Shen, S.F. Ge, K. Yoshikawa, Energy recycling from sewage sludge
[42] C. He, K. Wang, Y.H. Yang, P.N. Amaniampong, J.Y. Wang, Effective nitrogen
by producing solid biofuel with hydrothermal carbonization, Energy Convers.
removal and recovery from dewatered sewage sludge using a novel integrated
Manag. 78 (2014) 815–821.

12
H. Zhan et al. Journal of Environmental Chemical Engineering 10 (2022) 107540

system of accelerated hydrothermal deamination and air stripping, Environ. Sci. [45] J.R. Pels, F. Kapteijn, J.A. Moulijn, Q. Zhu, K.M. Thomas, Evolution of nitrogen
Technol. 49 (2015) 6872–6880. functionalities in carbonaceous materials during pyrolysis, Carbon 33 (1995)
[43] H. Zhan, X.Z. Zhuang, Y.P. Song, X.L. Yin, C.Z. Wu, Insights into the evolution of 1641–1653.
Fuel-N to NOx precursors during pyrolysis of N-rich nonlignocellulosic biomass, [46] J. Mumme, M.-M. Titirici, A. Pfeiffer, U. Lüder, M.T. Reza, O. Mašek, Hydrothermal
Appl. Energy 219 (2018) 20–33. carbonization of digestate in the presence of zeolite: process efficiency and
[44] C. Falco, M. Sevilla, R.J. White, R. Rothe, M.M. Titirici, Renewable nitrogen-doped composite properties, ACS Sustain. Chem. Eng. 3 (2015) 2967–2974.
hydrothermal carbons derived from microalgae, Chemsuschem 5 (2012)
1834–1840.

13