Characterization of Solid Fuel Chars recovered from Microwave Hydrothermal Carbonization of Human Biowaste

Microwave hydrothermal carbonization (M-HTC) is reported in this study as a viable sanitation technology that can reliably overcome the heterogeneous nature of human faecal biowaste (HBW) and realize its intrinsic energy value. Solid chars produced from the M-HTC process at 180°C and 200°C were characterized to further the understanding of the conversion pathways and their physicochemical, structural and energetic properties. The study revealed solid chars recovered were predominantly via a solid-solid conversion pathway. In terms of yield, more than 50% of solid chars (dry basis) can be recovered using 180°C as a benchmark. Additionally, the carbonized solid chars demonstrated enhanced carbon and energy properties following the M-HTC process: when compared to unprocessed HBW, the carbon content in the solid chars increased by up to 52%, while the carbon densification factor was greater than 1 in all recovered chars. The calorific values of the chars increased by up to 41.5%, yielding heating values that averaged 25MJ.kg-1. Thermogravimetric studies further revealed the solid fuel chars exhibited greater reactivity when compared with unprocessed HBW, due to improved porosity. This work strengthens the potential of the M-HTC sanitation technology for mitigating poor sanitation impacts while also recovering energy, which can complement domestic energy demands.


Introduction
Two of the key issues facing more than 2 billion people in developing countries are poor sanitation and energy scarcity. Having missed the Millennium Development Goals' targets by wide margins, achieving access to adequate and equitable sanitation, ending open defecation/untreated faecal wastewater and increasing the share of affordable, renewable clean energy by 2030 now constitute the key targets of the recently adopted Sustainable Development Goals 6 and 7 (1). Annually, an estimated one billion tons of faecal wastewater is generated (2) and as much as 90% of this is discharged untreated (3,4). Open defecation is still practised by almost one billion people, and about 2.4 billion people still lack access to improved sanitation (5). The consequences of poor sanitation are pervasive. Open defecation fields (which serve as breeding sites for insects, vectors/disease pathogens), odour nuisance, exposure to human faecal biowaste (HBW) during manual pit emptying and the indiscriminate disposal of this waste into surface water, open drains, near slums and other fragile settlements constitute serious public health and environmental risks. Faecal contamination of drinking water resources is primarily responsible for the high infant mortality rates due to waterborne diseases such as diarrhoea, which kills 700,000 children per year (6). Aside the environmental and health impacts, experts estimate that lack of access to sanitation cost the global economy US$223 billion in 2015 -based on an economic valuation of the costs associated with premature death and loss in productivity, the healthcare costs of the sick and time forgone due to lack of access to improved sanitation (7).
In common with poor sanitation, energy scarcity affects the least well off; an estimated 90% of people in developing economies lack access to reliable energy supplies (8). More than two billion people rely on firewood, charcoal and other related forest biomass to meet domestic energy demands such as cooking and heating (9). Over-collection of firewood and production of wood M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 5 rich renewable biomass resource of significant value, which can be beneficially exploited, rather than a taboo or problematic waste or expensive liability, as it is conventionally viewed across different cultures around the world. Essentially, HBW is rich in organic matter (more than 90%, dry basis), nutrients (nitrogen [N], phosphorus [P], potassium [K]) and thermal heating value (12). With a generation rate ranging between 120g and 530g.cap -1 day -1 of wet human faeces and 1-1.4L.cap -1 day -1 of urine (12,13), HBW represents a huge nutrient-rich, sustainable and underexploited resource. Hence, harvesting, recovering and reusing the intrinsic value in HBW using appropriate technologies is sensible -not only in terms of mitigating the impacts of sanitation, but also in contributing to the generation of clean and renewable energy. This aligns with the concepts of ecological sanitation, waste valorization, nutrient recycling/resource looping and the circular economy.
Microwave hydrothermal carbonization (M-HTC), a microwave-assisted thermochemical conversion process, represents a potential sanitation technology that can achieve this; i.e. it is viable to process/treat biowaste and recover valuable end products using this process (14)(15)(16).
The feasibility of M-HTC technology has been demonstrated to process/treat pure substrates (e.g. cellulose, fructose and glucose) (15,17,18) and more complex waste streams such as sewage sludge (14), seafood waste (19), pine sawdust (15), sugarcane bagasse (20), and other lignocellulose waste materials (16,21,22). Water, which constitutes up to 95% w/w in HBW, can interact with microwaves, with this electromagnetic interaction causing dielectric heating (12). The electromagnetic copulation of microwaves with water and other dipolar organics in HBW facilitates rapid and volumetric heating, which can promote novel and faster reaction pathways (12,15). This is the background theory behind the exploration of a microwave-assisted HTC to process HBW. Conventional HTC process can also be used to process biowaste. When compared, studies have reported that the M-HTC process promotes faster processing time, due to rapid volumetric heating; facilitates higher processing rates, due to relatively lower residence time; improves dewaterability and most importantly consumes less energy (about half the energy required for conventional HTC process) to convert biomass materials to valuable products (16,23,24). Hence, using the M-HTC process could potentially improve overall biowaste-processing efficiency. Essentially during any carbonization process, organic biomass is processed under subcritical water conditions at temperatures 160°C to 220°C, under autogenous pressure and in the absence of oxygen. These conditions mimic the natural coalification process, except that a series of simultaneously occurring reaction mechanisms is triggered by the process and these reduce reaction times significantly; they also facilitate the production of a carbonaceous coal-like end product (25,26). One of the key merits of the process, which makes it attractive in the context sanitation, is its capacity to utilize wet biomass (moisture content > 50%), thereby obviating the need, energy and cost associated with drying biowaste, as obtained during compositing and other thermochemical processes (25). Additionally, using wet HBW essentially makes it ideal for improving sanitation -i.e. for hygienic collection, containment and treatmentas these practices further minimize direct exposure and associated health risks. Preclusion of microbial culture, complete pathogen deactivation at high temperatures, short processing times, a smaller footprint and the absence of fugitive GHG emissions are other factors supporting the M-HTC proposition as a potential sanitation technology (12,25).
To date, studies involving real faecal waste as a test material using a thermochemical conversion process such as HTC, either for treatment purposes or other related scientific investigations, are very limited. Using faecal surrogates/mixtures is not ideal, as real faecal sludge is different in composition, water absorption properties, chemical and physical properties to such mixtures (12,13,27). The adoption of real faecal waste as a test material for the HTC process offers an opportunity to address knowledge gaps in this field. This work is related to that currently being undertaken to develop novel technology based sanitation facility capable of processing HBW to a safe form whilst at same time recover valuable products from it. The focus of this work is on solid fuel char recovery from HBW for potential energy use. Objectively, this work seeks to provide insights into the likely reaction pathways and conversion models associated with the recovery from of solid fuel char from HBW using the M-HTC process. Reporting of the physicochemical, structural, energy and combustion properties of the treated faecal material in this work aims to further address knowledge gaps on char characteristics lacking in the literature and provides useful information for the design, operation and optimization of an HTC-based sanitation facility.

HBW feedstock
HBW feedstock used in these study were: primary sewage sludge (SS); faecal sludge simulant (FSS); human faeces (HF) -without urine, flush water or sanitary products; and human faecal sludge (HFS) -including faeces, urine, flush water and tissue paper.
The primary sewage sludge (SS) used for the study was obtained from the primary sedimentation holding tank at Wanlip Sewage Treatment works, Leicester, UK. The SS derives from a catchment area serving a population of 0.5 million people, with mixed domestic and industrial effluent. The moisture content of SS (as received) falls between 95 and 96%.
The human faeces (HF) and human faecal sludge (HFS) used in this study were collected from anonymous donors. For the HF sampling, donors were instructed to separate urine from faeces at the point of defecation. Polythene bags provided to the donors were used to collect faeces.
After collection, the faecal specimens were examined to ensure no urine or sanitary products were present. Abnormal faeces were not used in the study. Abnormality was determined through a physical examination and evaluation against the Bristol Stool Chart (28). Only faecal types 3 and 4 were used. Types 1 and 2 are considered to be indicative of constipation, while type 5 is typically low in fibre, and types 6 and 7 are due to diarrhoea. These grouped types have abnormal properties -such as transit time in the human body, diameter and shape configuration, rheology, and chemical and biological compositions, which make them different from normal human faeces defecated daily. HFS samples were collected using a portable mobile toilet ('Porta Potti') placed in a designated toilet. The HFS samples were also subjected to same physical examination and evaluation against the Bristol Stool Chart, as with the HF samples. Both the HF and HFS samples were homogenized by maceration, and their resultant moisture content adjusted to about 95% (i.e. 5% total solids content) to mimic the faecal sludge characteristics typically found in onsite sanitation facilities (11,29,30).
Faecal sludge simulant (FSS) is an artificial faecal sludge prepared from a formulated recipe reported in a study (27) to replicate the chemical composition, water absorption capacity and rheology of real human faeces. After weighing each constituent of the FSS components shown in Table 1, the moisture content of the recipe was adjusted (by mixing uniformly with water) to about 96% (≈4% solids). This was to ensure consistent water-to-solid loading with the SS, HF and HFS samples.  and pore sizes (nm) of samples were determined using a single point BET nitrogen adsorption analysis on a Micrometrics Tristar Surface Area and Porosity Analyser. Before analysis, 0.2-0.3g of the representative samples were degassed in a vacuum. Analyses of the data and isotherms generated during the analysis were processed to determine the char specific surface area and pore sizes. Surface functionalities were examined using Fourier Transform Infrared (FTIR) spectrometry. The FTIR analysis of all samples was performed using a Shimadzu FTIR-8400S.
Samples were run using a Golden Gate diamond ATR (attenuated total reflectance) (Specac Ltd, UK) FTIR spectrometer accessory. Infrared spectra were collected within the 4000-600cm -1 regions, with a spectra resolution of 2cm -1 . To ensure accuracy during this analysis, the background emission spectrum of the infrared (IR) source was recorded and taken into account while collecting the emission spectrum of the IR source from the test material. Background emissions were automatically deducted from each sample emission spectra. Sixty-four (64) scans were collected for each sample.

Conversion of HBW during M-HTC based on organoleptic assessment
When assessed against the unprocessed biowastes, the smell, colour and texture of the end products recovered after the M-HTC processing of the four feedstocks were all distinctively transformed. The foul odour associated with unprocessed SS, HFS, HF and FSS was replaced by: a coffee-like smell for SS; an almond-like smell for HFS and HF; and a smell characteristic of burnt oil for FSS. The smell of the end products represents a significant improvement when compared to the foul odour of unprocessed biowastes. The colour of both the carbonized solids (and recovered liquor) after the M-HTC process changed from the brown colour associated with human faeces or sludges to a carbonaceous, coal-like colour. The dried solid chars also appeared denser, friable and could be easily pulverized. The texture of the solid chars further suggested they could be moulded into briquettes or pellets for use as solid fuel.

M A N U S C R I P T
A C C E P T E D ACCEPTED MANUSCRIPT

Mechanisms behind smell and colouration changes
HBW is composed of large macromolecular components: nitrogenous protein compounds, carbohydrates, lipids, minerals and bacterial debris in varying proportions (12,13,27). The eradication of the foul odour in carbonized biowastes is primarily due to the thermal hydrolysis of these organic macromolecular components during the M-HTC process. The hydrolytic process facilitates the solubilization of compounds causing foul odour in HBW, such as the nitrogenous benzopyrrole compounds -notably indole and skatole, hydrogen sulphide, the methyl sulphides and other sulphur-containing compounds (32,33,34); and renders them nonvolatile, dissolved and trapping them in the liquid phase (35). Other reactions associated with HTC, including aromatization, may also suppress odour (36).
Apart from caramelization reactions, a non-enzymatic browning effect observed on sugars in biowaste processed at elevated temperatures (37) that the type of smell that will be produced in end products depends largely, among other factors, on the type and nature of proteins present in the feedstock (41). This partly explains the differences in smell observed from the processed biowastes.
The sensory assessment of the smell and colour of the carbonized end products recovered after the M-HTC process showed them to be similar and consistent with thermochemical conversion/transformation processes of organic biomass reported in literature involving other conventional heating methods (35,42,43,44). In essence, these similarities in the organoleptic properties of smell and colour of the end products from the M-HTC of HBW with existing methods, and the characteristic colour and smell change of end products typical of HTC processes, provide preliminary evidence that suggests M-HTC converted HBW to carbonaceous materials.

Conversion of HBW during M-HTC based on SEM micrographs
Scanning electron microscope (SEM) imaging has been used to study potential conversion/formation pathways for char derived from different feedstocks, including glucose and cellulose (15,18), digested sewage (42) and lignocellulose biomass (45,46 mag.) of chars recovered from the four biowastes compared with their unprocessed samples.
The differences in microstructural morphology are discernible. The information obtained from the SEM micrographs suggests two possible M-HTC char formation pathways from HBW: direct solid-to-solid conversion and induced nucleation pathways.

Induced nucleation, polymerization of dissolved intermediates
The SEM micrograph of FSS chars looks entirely different from the others, as it reveals spherical, hollow-like features (see Figure 8), as reported previously (46,50,51,52). The formation of regularly shaped, hollow carbon microspheres from glucose and fructose solutions is well studied and known (53,54). Such microspheres are formed from induced nucleation and polymerization of dissolved soluble intermediates (mainly furfural compounds, including 5hydroxymethylfurfural , furfural and 5-methylfurfural) when glucose, for example, is heated to temperatures similar to HTC conditions (52,55). At 120°C to 140°C, fructose solution undergoes intermolecular dehydration to HMF (50). Another study that also reported the formation of carbon spheres from glucose under HTC conditions of 160°C-180°C, further mentioned that 5-HMF forms after glucose hydrolysis (51). These studies concluded that the HMF intermediates were susceptible to subsequent polymerization/polycondensation reactions, leading to the formation of hollow carbon microspheres. The FSS chars' formation appears to follow this mechanism, as FSS is a cellulose-based recipe (see Table 1). The thermal hydrolysis of cellulose during M-HTC forms reducing sugar monomers, including glucose, which can exist in isomerism with fructose (42,56). Intermolecular dehydration of these reducing sugars under HTC conditions leads to the formation of dissolved soluble intermediates, i.e. HMF, which subsequently undergo polymerization -characterized by further intermolecular water loss (dehydration) and forming spherical, hollow char particles similar to those shown in Figure 8.
In essence, both conversion pathways relate to the present study, as modelled in Figure 9. While the SEM micrographs of chars from SS, HFS and HF agree with the solid-to-solid conversion model (as no spherical particles were found), the FSS char micrographs agree with the secondary pathway. What is unknown is the extent to which the conversion pathways contribute to char formation, as the effect of the Maillard reactions (i.e. smell and colour changes due to reactions of reducing sugars monomers from the carbohydrate contents in HBW and cellulose in faecal simulant) are seen across all feedstocks. One study used cellulose as a HTC substrate and argued that both pathways occurred during conversion to char; however, it maintained that at certain temperature ranges <200°C, the solid-to-solid conversion route predominates (57). Although knowledge of char formation mechanisms from complex and heterogeneous substrates, such as human faecal biowaste, is still evolving, the SEM studies conducted here indicate the solid-tosolid conversion route to be the predominant pathway for HBW at the temperature used.

Porosity and surface area
In order to gain further insight into the structural changes caused by the M-HTC process, the surface area (m 2 .g -1 ) and pore sizes of chars obtained from the M-HTC of each feedstock run at 180°C and 200°C are shown in Table 3. From Table 3, the pore sizes -ranging from 9.6nm to 36nm -may be classified according to the IUPAC classification as Type 2 pore sizes, mesopores 2nm to 50nm (58); these are similar to the pore sizes of HTC chars of sunflower and walnut (59). The char pore sizes were consistent, with their surface areas ranging between 0.9m 2 .g -1 to 5m 2 .g -1 ; this was similar to the values reported for chars recovered from microwave pyrolysis of straw pellets and willow chips characterized under BET and mercury porosimetry (60).
Literature values for BET surface areas of most HTC chars (derived from conventional heating) for feedstocks such as apricot, sugar bagasse, willow, algal and sewage sludge range between 0.67m 2 .g -1 and 14.68m 2 .g -1 (61 -65); these are comparable to the values observed in the present study.
The surface area of the chars was generally greater, by more than 50% in most cases, than that of the feedstock. This can be attributed to tunnelling effects caused by heating and the mass transfer processes This may be attributed to the effect of dehydration during the M-HTC process (18). In addition, bands at 880cm -1 to 700cm -1 , corresponding to aromatic C-H bends, were observed in chars but not in unprocessed feedstock, indicating the presence of aromatic structure in chars (65).

21
These observations are similar to previous studies (71,75), which noted that the FTIR spectra of organic matter are usually qualitatively similar, but differ in the relative intensity of absorbance bands and specific bands. Such differences can provide clues to likely reaction pathways occurring during a conversion process. From this study, hydrolytic solubilization, dehydration and aromatization reactions can be said to have occurred during M-HTC processing of HBW.       Char yield from the HTC process is sensitive to the type, nature and compositional characteristics of the feedstock, including initial solid loading/moisture content and carbonization temperature, among other factors (25). This may explain the yield observed from all substrates, although slight differences in solid loading must be noted (see Table 2). The range of char (yield) recovered in the present study was similar to that reported in other studies (15,16,20), which all involved microwave as the heating source, although with different feedstocks and pre-treatments. More than 50% conversion efficiency of biowaste to solid chars from the M-HTC process is feasible using 180°C as a benchmark. Remarkably, M-HTC initiated changes in the elemental composition of chars recovered from unprocessed biowaste (compare Tables 2 and 8). Changes in elemental composition were observed to be dependent on carbonization

Yield and elemental properties of HBW chars
temperature. An increase in the carbon content of chars, and a corresponding decrease in the oxygen content, was observed for chars recovered at the carbonization temperatures investigated. Percentage carbon increment due to carbonization effects (see C EFF in Table 8 factors of the chars, which were greater than 1 in all cases, were comparable to the values in literature, which range between 1 and 1.8 (76,77). In essence, M-HTC increased the carbon content and carbon densities of the chars.

Molar ratios of H/C and O/C were estimated from elemental compositions, and were analysed with a
Van Krevelen diagram to further understand the reaction pathways involved during the M-HTC of each biowaste. Van Krevelen diagrams allow the delineation of reaction pathways: a line at a slight angle to the x-axis usually represents a decarboxylation pathway, while diagonally drawn lines usually denote a dehydration pathway, as shown in Figure 10.

Higher heating values and energetic properties of HBW chars
The calorific value, also known as the higher heating value (HHV), is an important characteristic of chars that enables the determination of key energy parameters -such as improved calorific value, energy densification and energy yield -for comparative assessment with unprocessed biowaste and conventional fuels.  Table 10). This suggests the use of these chars either as a stand-alone solid fuel for domestic energy consumption (e.g. for cooking and heating) or as one that can be cocombusted with other fuels of similar heating value.
Similar observations of increased heating value of biowastes after pyrolytic processing have been reported in many studies, with many substrates and heating sources (20,83,84). For example, reported calorific values of chars recovered from wastewater sludge range from 14.4 to 27.2MJ.kg -1 (65,79,85,85) and are comparable to the HHVs obtained for all chars recovered in the present study. Energy densification -as indicated by the energy enrichment factor (EEF -see Table 9) -of all chars recovered from all unprocessed biowastes ranged from 1.05 to 1.41. Similar energy densification ratios were reported for HTC chars produced from municipal solid waste (MSW) (1.01 to 1.41 [86]) and woodbased substrates (1.11 to 1.43 [83]). This is evidence that M-HTC appears to promote energy densification in chars. EEF was also observed to increase slightly with increasing temperature, with FSS recording the highest densification (1.41 at 200°C). Energy yield, which provides a means for assessing the energy recoverable from chars, ranged between 55% and 74% and decreased over the carbonization temperature investigated for all feedstocks -primarily due to the reducing char yield.     Table 11 summarizes key results of the thermogravimetric analyses (TGAs).
for SS, between 350°C and 360°C for both HF and HFS, while for FSS, the phase ended around 413°C. For chars, the temperature ranges associated with this phase were lower when compared with their unprocessed biowastes. This suggests chars were more easily thermally degraded, as the phase ended at less than 300°C for most chars. Differences in peak temperatures, which corresponded to maximum weight loss at this phase, were analysed -as indicated as PT a in Table   11. The final stage, i.e. burning in air (O 2 atmosphere) and ashing, began after completion of the second stage. Once again this phase ended at lower temperatures (see BT column in Table 11) when the chars were compared with their unprocessed biowastes.
Based on the TG-DTG analyses, M-HTC results in marked differences in the combustion behaviour of chars recovered from unprocessed biowaste. These differences were as follows: A distinctive DTG combustion profile and differences in the amount of starting material combusted.
An increase in carbonization temperature influenced the combustion behaviour of chars, as it tended to make chars more reactive during the decomposition phases.
Chars exhibited a greater reactivity-to-combustion profile along the TG temperatures In essence, the M-HTC processing of HBW does more than manage a faecal sludge/sanitation problem, i.e. it has the potential to yield products that are especially useful in energy applications.
This can propagate changing views on HBW as a sustainable resource and improve interest for integrating the technology with existing sanitation systems, or encourage design and development of sanitary facilities -either as stand-alone or mobile processing units -that will promote energy recovery from HBW.