1 Introduction

Thermochemical conversion of biomass (e.g., sugarcane bagasse, coconut shell, rice husk, banana leaves, etc.) has been a field of interest since the last decades [1,2,3,4], and now a variety of solid fuels are categorized in terms of chemical, physical, thermochemical and thermodynamic properties. On the other hand, the solid residues such as the sewage and algal sludges residues from water treatment plants, and raw human feces appear in the scenario of conversion technologies in need of characterization. These solids have a relatively large amount of organic content soaked in water, more than 75%. Even though the organic content may confer to them a biomass characteristic, the biosolids classification seems to be more appropriated [5]. However, to use these residues as an energy source, reducing the water content to the level required for thermochemical conversion technologies is still a challenge. Very few works have attempted to convert human feces [5,6,7,8], and the conversion has been reached by mixing feces with inert materials, e.g., sand [5], or with plastic materials [8]. Heretofore, the only successful case of converting raw feces has been made in a bench-scale fixed bed reactor [6].

Some authors explored the thermochemical characterization of simulated feces on smoldering combustion [9,10,11], and gasification [6, 12], but to reproduce feces is a very complex challenge considering its heterogeneous characteristics. Nevertheless, the results were very encouraging. Monhol and Martins [8] conducted combustion of real feces, showing that a high energy content could be recovered. Recently, Jurado et al. [13] developed preliminary work on a micro-combustor prototype to recovery energy from human feces. Tests were carried on surrogate feces and then applied on real feces. Yacob et al. [14] performed slow pyrolysis experiments on human feces, quantifying the char yield and gas evolution. Yet, some contradictory results are found in the literature. For instance, Onabanjo et al. [12] reported for real feces an average amount of 51% of ash, 17% of volatile matter and 32% of fixed carbon, and for simulant feces an average amount of 14% of ash, 86% of volatile and 0% of fixed carbon, both analysis on dry basis. Jurado et. al [13] reported no presence of fixed carbon neither for real nor simulant feces. Although these previous studies delivered large contributions in the field of waste management, it did not account for a detailed thermochemical characterization of real human feces as a solid fuel. Here, for the first time, the exothermic characteristic of human feces pyrolysis is experimentally presented. According to Roberts [15], the exothermic pyrolysis can release around 8–10\(\%\) of the heat content of the feedstock. A few works show experimental evidence on this subject [16]; however, the exothermic behavior is only revealed when separating the main constituents of wood—hemicellulose, cellulose and lignin.

Since the conversion of human feces is not fully understood and there is no conversion technology developed for such potential fuel, the gasification technology is still a good reference, even though it has been built for wood biomass. To suggest gasification as one way to recover energy from feces, a thermodynamic analysis is more suitable. The method of exergy analysis is an engineering tool applied to assess the performance of energy conversion processes and systems in light of the second law of thermodynamics. Exergy is defined as the maximum theoretical potential of a system energy amount that can be converted in work, regarding a reference environment [17]. Such potential can be split into two fractions, namely physical exergy and chemical exergy. Physical exergy is the work potential fraction due to pressure and/or temperature difference of the system and environment, while chemical exergy is the remaining fraction due to composition difference. Then, exergy may be interpreted as an indicator that put all energy quantities on the same basis for comparison, i.e., their respective potentials to do work. As exergy is destroyed along processes due to irreversibilities, instead of being conserved as energy is, the true thermodynamic inefficiencies of processes are revealed [18].

Several exergy analyses of gasification process have been done and related in the literature. Ptasinski et al. [19] performed a comparative study for the gasification exergy efficiency of different types of biofuels (wood, vegetable oil, sludge and manure) and benchmarked it against the gasification of coal. Pellegrini and De Oliveira [20] and Onabanjo et al. [12] performed an exergy analysis of sugarcane bagasse and human feces gasification, respectively. All these works were conducted under theoretical assumptions, applying a thermodynamic equilibrium model with an adiabatic reactor. As previously established by Gomez and Leckner [21] and Arnavat et al. [22], the equilibrium model overestimates the production of \(\hbox {H}_{2}\) and CO, underestimates \(\hbox {CO}_{2}\) and almost zero out the concentration of \(\hbox {CH}_{4}\), tars and char.

At present, the suitability of human feces as a feedstock for gasification is exclusively based on theoretical assumptions that occasionally lead to unrepresentative descriptions of the actual process. In this work, a thermodynamic assessment of human feces gasification is performed considering for the first time the inefficiencies of a downdraft gasifier. Thus, the real state of the thermochemical conversion of such residue is achieved for an existent technology. The thermochemical characterization from the sterilized material to the products streams is also presented.

2 Materials and methods

2.1 Raw material

The raw feces used in experiments were collected directly from the dry sanitary toilets installed at the Laboratory of Combustion and Combustible Matter from the Federal University of Espirito Santo, Brazil. Notable characteristics were strong odor and dark coloration. For security and hygiene procedures, the raw feces were sterilized inside ovens at \({105}\, {^{\circ }}{\hbox {C}}\), to prevent proliferation and action of pathogens. Later, an open-air drying treatment followed by mass measurements was conducted to leave the material with three moisture contents: 0%, 10% and 20% which according to the literature is the maximum acceptable value to conduct experiments in a downdraft gasifier [23].

2.2 Experimental device and protocol

Thermal analyses were performed using TA INSTRUMENTS model SDT Q600, where both thermogravimetric and differential thermogravimetric analyses were performed simultaneously. The samples were prepared with 3.0 mg, and the heating rate was set on \({10}\, {^{\circ }}{\hbox {C}}/\text {min}\) up to \({600}\, {^{\circ }}{\hbox {C}}\), whereas the flow rate of atmosphere gas was fixed at 100 mL/min. The experiments were repeated with an inert atmosphere (\(\hbox {N}_{2}\)) and oxidative (air) atmosphere. A non-standard proximate analysis combining TG and DSC was used to determine the fractions of fixed carbon, volatile matter and ash.

Inductively coupled plasma optical emission spectrometry (ICP OES) was used for chemical characterization, following standard ASTM 5373-93. Levels of carbon, hydrogen, phosphor and sulfur were also set. Nitrogen levels were determined by the method Kjeldahl 4500-Norg. The amount of ash was determined following Standard ASTM 3174-00.

Batch gasification process was conducted on the downdraft Biomass Gasifier ALL POWER PELLET GEK 20 kW. The full description of the device and gasification protocol can be referred in [24]. The gasifier was fed at the top with 300 g of the feedstock, following the zones of drying, pyrolysis, oxidation and reduction. The heat generated from the reaction char + \(\hbox {O}_{2}\) (oxidation zone) was responsible for providing conditions for pyrolysis and for drying the feces in the other zones. Since a downward flow was established, tar was cracked within the reduction zone, leaving the output gas tar-free. The airflow rate was fixed at 0.291 L/min and the pressure was 99.96 kPa. The temperature was monitored through the whole process by thermocouples located at the throat and the bottom, as shown in Fig. 1.

Fig. 1
figure 1

Experimental device used for batch gasification

Full size image

Gases produced were directed to the flare, while a part was conducted through a silicone hose connected to the gas analyzer TESTO 454 M/XL, capable of performing online measuring. When the flare was active, it was an indication that combustible gases were being produced. Before entering the analyzer, the gases passed through an intermediate dry filter to prevent penetration of particulates. The experiments were repeated for the three moisture samples to check the influence on the thermochemical conversion. Table 1 summarizes the parameters of gasification.

Table 1 Parameters for batch gasification of sterilized human feces
Full size table

2.3 Thermodynamic methodology

The classical approach presented in Kotas [18] is followed, where a mass, energy and exergy balance are given in detail and well explained. The computer software Engineering Equation Solver [25] was used to implement and perform the calculations. As the gasification is a batch process, 0.3 kg of fecal material and moisture values of 0%, 10% and 20% were used as an input for the analysis. Due to the local environmental conditions, the input air for the model was a binary mixture of \(\hbox {N}_{2}\) and \(\hbox {O}_{2}\) in wet conditions (80% of water at 99.96 kPa, Table 1). The equations that model the problem will be presented along with the results section, already including the experimental-based approach, i.e., feces characterization and the gasifier’s measurements. The thermodynamic methodology can be summarized into the three steps as follows:

Fig. 2
figure 2

Thermal analysis of human feces: a TG/DTG under air, b TG/DTG under \(\hbox {N}_{2}\), c DSC under air, and d DSC under \(\hbox {N}_{2}\). Heating rate at \({10}\, {^{\circ }}{\hbox {C}}\)/min

Full size image
Fig. 3
figure 3

DSC under \(\hbox {N}_{2}\) from a chicken and b pig feces, both presenting exothermic peaks during pyrolysis. Heating rate at \({10}\, {^{\circ }}{\hbox {C}}\)/min

Full size image
Fig. 4
figure 4

Van Krevelen chart for human feces (gray) and other solid fuels (black). Data from [1, 2, 32, 51, 53, 54, 56, 57]

Full size image

  • A mass balance is performed using data from the ultimate analysis, non-standard proximate analysis and from the syngas composition to balance the chemical equations.

  • The molar composition and the temperature of the syngas for each moisture content are used to perform the energy balance. Here, key parameters as the cold gas efficiency (CGE) and heating value are determined, also using data from the proximate and ultimate analysis.

  • At last, the exergy balance is performed to identify and quantify the inefficiencies related to the gasification process, using the results from mass and energy balance.

3 Results and discussion

3.1 Analysis of sterilized feces

3.1.1 Thermogravimetric analysis (TGA) and differential scanning calorimetry analysis (DSC)

From 30 to \({150}\, {^{\circ }}{\hbox {C}}\), TGA under air (Fig. 2a) reveals two peaks of mass loss, representing around \(17.3\%\). The first one is around \({50}\, {^{\circ }}{\hbox {C}}\) and can be related to the release of some volatile organic compounds (VOCs) typically found in feces, such as ethanoic, butanoic, pentanoic acids, benzaldehyde, ethanal, carbon disulfide, dimethyl disulfide, acetone, 2-butanone, 2, 3-butanedione, 6-methyl-5-hepten-2-one, indole and 4-methylphenol [26, 27]. The second peak, around \({80}\, {^{\circ }}{\hbox {C}}\), can be related to water evaporation. Both are characterized by an endothermic valley on DSC under air (Fig. 2c). From 150 to \({400}\, {^{\circ }}{\hbox {C}}\), approximately \(50.2\%\) of mass loss is observed. This mass loss is related to the major breakdown mechanism in feces: the overlap pyrolysis and homogeneous/heterogeneous reactions. Between this temperature range, DSC (Fig. 2c) shows only one exothermic peak, characterizing that the energy released is an average positive value from all reactions. Noteworthy, there is no consensus in the context of solid combustion, about the real contribution of the gas phase oxidation on the heat release [28, 29]. This issue will not be addressed in this work. From 400 to \({550}\, {^{\circ }}{\hbox {C}}\), almost 25.1% of mass loss is observed. This loss is related to the oxidation of char, confirmed by DSC, where a strong exothermic peak is noted. Furthermore, no mass loss was observed, remaining 7.4% of ash.

We now call the attention for pyrolysis, represented in Fig. 2b (TG/DTG) and Fig. 2d (DSC). From 30 to \({150}\, {^{\circ }}{\hbox {C}}\), following the same fashion of TG under air, a mass loss related to water evaporation and volatiles release is observed, while DSC also reveals only one and wide endothermic valley. From 150 to \({500}\, {^{\circ }}{\hbox {C}}\) almost 60% of mass loss occurs. As a first guess, it is related to overall pyrolysis forming char. Combining the remaining mass on both TG under air (about 7%) and TG under \(\hbox {N}_{2}\) (about 32%), the char amount is about 25%. Table 2 shows the proximate, ultimate analysis and ash composition of human feces and other solid fuels.

Table 2 Proximate and ultimate analysis and ash composition (wt%) of sterilized human feces and other solid residues from the literature
Full size table

Concerning Fig. 2d, a strong exothermic peak is observed. However, pyrolysis generally is noted for presenting endothermic reactions, since devolatilization normally consumes energy. The fact is that a large amount of the VOCs have oxygen on composition and as the compounds get released, these oxygen molecules turn the atmosphere oxidative. It is important to state that the DSC curve shows the overall balance of energy, meaning that during this stage, exothermic reactions released more energy than endothermic reactions from devolatilization consumed, characterizing the exothermicity of pyrolysis. Nevertheless, the exothermic energy and the oxygen released provided conditions only for partial char oxidation. Yang et al. [16] also observed an exothermic behavior during cellulose pyrolysis for temperatures above \({400}\, {^{\circ }}{\hbox {C}}\), trying to explain the competition between endothermic and exothermic reactions in wood pyrolysis.

To ascertain the result found, we also conducted DSC under \(\hbox {N}_{2}\) (Fig. 3) for chicken and pig feces. It is observed the same behavior of human feces, highlighting the influence of VOC during pyrolysis. After \({200}\, {^{\circ }}{\hbox {C}}\), an exothermic peak is dominant in both analysis, ratifying our previous observation.

By integrating DSC curves in time, it was possible to obtain the heat of reaction by the mass of samples for each peak. Performing a simple energy balance, the evaporation energy demand (210.5 J/g) could be covered by the exothermic pyrolysis energy (545.5 J/g), leading to a positive balance (335 J/g) of the overall pyrolysis process. It means that theoretically, for a chemical kinetics dominant phenomenon there is a given level of moisture that could be dried using the energy from pyrolysis (from 8 to 10% according to Roberts [15]). In terms of combustion, the overlapped pyrolysis and homogeneous/heterogeneous reactions had a positive heat of reaction of 865 J/g, while char oxidation reached 1562 J/g of sample.

3.1.2 Chemical characterization

Table 2 also shows the chemical composition of feces, trying to establish some comparison with categorized biomass. Also, biosolids are included since such residue has an important amount of organic matter, with a similar composition to human feces in terms of proximate and ultimate analysis. A few authors refer to human fecal material as a categorized biomass [6, 12, 14, 30], but the high levels of moisture content tend to approximate their classification as a biosolid. Indeed, while eliminating the water levels, the biomass category fits with more accuracy. The Van Krevelen chart is plotted (Fig. 4) and feces are located inside the biomass region, with atomic H:C ratio of 1.51 and O:C ratio of 0.80, closer to values from pine sawdust [1], sewage sludge [31] and peanut shell [32].

3.2 Products of gasification

3.2.1 Syngas analysis

During the gasification process, the composition of the permanent gases was established. Since the amount of the syngas composition is process dependent, the values presented here must be taken as estimated for the three moisture values. The species \(\hbox {O}_{2}\), \(\hbox {CO}\), \(\hbox {H}_{2}\), \(\hbox {NO}\) and \(\hbox {NOx}\) were analyzed using electrochemical sensors, \(\hbox {CO}_{2}\) levels were set using NDIR analyzer, and \(\hbox {CxHy}\) analysis used heated bead sensors. The results are expressed in vol. %.

Traces of \(\hbox {NO}\) and \(\hbox {NOx}\) were found, reaching maximum values of 100 ppm, regardless of the moisture levels. The total hydrocarbons \(\hbox {CxHy}\) were measured, but due to safety limitations of the device, it was not possible to conduct a continuous measurement. Nevertheless, independent of the moisture content, all samples reached at least 4% at the beginning of the gasification process. The actual concentration of hydrocarbons will be calculated further by a mass balance (Sect. 3.3.1).

Figure 5 presents only the permanent measurements of \(\hbox {O}_{2}\), \(\hbox {CO}_{2}\), \(\hbox {CO}\) and \(\hbox {H}_{2}\), together with the syngas temperature at the exit. The average molar fractions of the syngas components were calculated by integrating the curves in time intervals of: 704s to 2112s for 0% moisture, 670s to 2010s for 10% and 588s to 1764s for 20%. Note that these intervals are considered to be in the steady gasification since the temperature and \(\hbox {O}_{2}\) reached a plateau and the gas flaring was active.

The average composition of \(\hbox {H}_{2}\) decreased with the increase in moisture, from 6.56 to 3.91%, with low experimental uncertainty (0.62% average), while \(\hbox {CO}\) average composition slightly decreased, from 9.65 to 7.89%. However, the measurements uncertainties were higher for samples with 20% of moisture (6% maximum). \(\hbox {CO}_{2}\) composition was relatively unaffected by the moisture content, staying in the range of 6.23–7.60%, with very low uncertainties (0.45% average). Finally, \(\hbox {O}_{2}\) decreased from 3.39 to 1.86% with average uncertainty of 0.48%. The temperature decreased with the increase in moisture, from 833 to \({786}\, {^{\circ }}{\hbox {C}}\), with average standard deviation of \({33}\, {^{\circ }}{\hbox {C}}\). Table 3 summarizes the information above.

Table 3 Average volume fractions of the syngas components (% wb)
Full size table

Returning to the analysis of Fig. 5, focusing on the classical chemical reactions involving C\(_{(s)}\), CO, \(\hbox {CO}_{2}\), \(\hbox {H}_{2}\) and \(\hbox {CxHy}\) in gasification [33,34,35,36], the rate of formation and consumption of species provide inference on the pathway of syngas production. For example, the formation of \(\hbox {CO}\) and \(\hbox {H}_{2}\) is distributed along the whole process for moisture contents of 0% and 10%, while for 20% the massive formation of the same species takes place at the beginning (300s to 800s). Also, since \(\hbox {CO}_{2}\) curves follow the same fashion regardless of the moisture content, it suggests that the water-gas shift reaction (\(\hbox {CO} + \hbox {H}_{2}\hbox {O} \leftrightarrow \hbox {CO}_{2} + \hbox {H}_{2}\)) contributes significantly to the \(\hbox {CO}_{2}\) and \(\hbox {H}_{2}\) formation; however, \(\hbox {H}_{2}\) is consumed in another reaction, leading to the formation of hydrocarbon.

3.2.2 Ash composition

The ash formed had a sandy and whitish appearance, representing \(7.4\%\) of the feedstock initial mass. Although the ash had a sandy aspect, it easily turned into dust with simple hand friction. A small amount of sludge, around 8% of the ash mass, was also noted. It had harder consistency when compared to the ash. Table 2 shows the results for the main chemical components found in ash from feces and compares it with the ash from sewage sludge [31], forest residues [37] and wood waste [36], highlighting significant differences to the compounds formed.

Due to great levels of P, K, Ca and Mg, the ash generated from human feces gasification can be used as a fertilizer. Another possible destination for the product is the construction industry, where the ash may be used to produce magnesium potassium phosphate cements (MKPCs) [38], since great levels of MgO are observed. Oxides like \(\hbox {Al}_{2}\hbox {O}_{3}\), \(\hbox {SO}_{2}\), \(\hbox {Fe}_{2} \hbox {O}_{3}\), MnO, CuO, \(\hbox {SiO}_{2}\), \(\hbox {B}_{2} \hbox {O}_{3}\), PbO, CdO and HgO completed the composition, representing around 3%.

The clean appearance and the lack of odor from the ash were characteristics observed that reinforced the sanitation contribution of the gasification process. Figure 6 shows the before (raw human feces) and after (ash) gasification.

Fig. 5
figure 5

Evolution of average temperature and gases formed during the batch gasification of sterilized human feces of varying moisture content: 0% (a, b), 10% (c, d) and 20% (e, f)

Full size image
Fig. 6
figure 6

Transformation of human feces after sterilization to ash with sandish and white appearance, formed after the gasification process

Full size image
Fig. 7
figure 7

Temperature and molar composition of the syngas produced by batch gasification of human feces samples with 0, 10 and 20% of moisture

Full size image

3.3 Thermodynamic analysis

3.3.1 Mass balance

To perform a mass balance, Eq. 1 is taken into account for 1 kg of human feces. The stoichiometric coefficients are obtained from the proximate and ultimate analyses of the sample. Also, in the syngas, there is a small molar fraction of \(\hbox {NO}\) and \(\hbox {NOx}\), which is neglected in the analysis. The chemical reaction is also written considering 10% and 20% of moisture, Eqs. 2 and 3, respectively.

$$\begin{aligned}&{(40.85{\,}\hbox {C} + 73.51{\,}\hbox {H} + 24.39{\,}\hbox {O}^{+} 3.21 {\,}\hbox {N} + 0{\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH})}\,\\&\quad{+ 12.75 {\,}(\hbox {O}_{2} + 3.76{\,}\hbox {N}_{2}) } +\, {1.58{\,}\hbox {H}_{2}\hbox {O}{\,}\rightarrow {\,}} {10.43{\,}\hbox {CO}}\\&\quad{ + 7.09 {\,}\hbox {H}_{2} + 10.7{\,} \hbox {C}_{2.18}\hbox {H}_{2.18} + 7.07 {\,}\hbox {CO}_{2} + 19.57 {\,}\hbox {H}_{2}\hbox {O}} \nonumber \\&\quad +\, {49.57 {\,}\hbox {N}_{2} + 3.66 {\,}\hbox {O}_{2} + \hbox {ASH}} \end{aligned}$$
(1)
$$\begin{aligned}&(36.77{\,}\hbox {C} + 66.16{\,}\hbox {H} + 21.95{\,}\hbox {O} + 2.89 {\,}\hbox {N} + 5.55 {\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH}) \\&\quad + 9.93 {\,} (\hbox {O}_{2} + 3.76 {\,}\hbox {N}_{2}) + +\, 1.23 {\,}\hbox {H}_{2}\hbox {O} {\,}\rightarrow {\,}7.65 {\,}\hbox {CO} \\&\quad + 6.13{\,} \hbox {H}_{2} + 9.95 {\,}\hbox {C}_{2.21}\hbox {H}_{2.21} + 7.17 {\,}\hbox {CO}_{2} + 22.76{\,}\hbox {H}_{2}\hbox {O} \\&\quad + +\, 38.79 {\,}\hbox {N}_{2} + 1.92 {\,}\hbox {O}_{2} + \hbox {ASH} \end{aligned}$$
(2)
$$\begin{aligned}&(32.68 {\,}\hbox {C} + 58.81 {\,}\hbox {H} + 19.51 {\,}\hbox {O} + 2.57 {\,}\hbox {N} + 11.10 {\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH})\\& \quad + 9.20 {\,} (\hbox {O}_{2} + 3.76 {\,}\hbox {N}_{2}) +\, 1.14 {\,}\hbox {H}_{2}\hbox {O} {\,}\rightarrow {\,} 7.20 {\,}\hbox {CO}\\& \quad + 3.57 {\,}\hbox {H}_{2} + 9.04 {\,}\hbox {C}_{2.19}\hbox {H}_{2.19} + 5.69 {\,}\hbox {CO}_{2} + 28.18 {\,}\hbox {H}_{2}\hbox {O}\\& \quad + +\, 35.89 {\,}\hbox {N}_{2} + 1.70 {\,}\hbox {O}_{2} + \hbox {ASH} \end{aligned}$$
(3)

The stoichiometric coefficients of the air components and the remaining syngas components are obtained in molar basis by the atomic balance of C, H, O and N. Besides, it is used the psychrometric definition of relative humidity and the definition of air–fuel ratio. To account the total mass of air which enters the gasifier, the batch process interval is considered along with the properties of humid air, obtained from the EES Air \({\hbox {H}_{2}\hbox {O}}\) package.

From Eqs. 1, 2 and 3, it is noted that the respective H/C ratios for the syngas hydrocarbons are approximately equal to 1. Since one knows that light hydrocarbons are yielded from gasification processes, it is considered that the hydrocarbons behave as an average hydrocarbon modeled as acetylene, \(\hbox {C}_{2}\hbox {H}_{2}\). Table 3 shows the average molar fractions of the syngas components in wet basis. Note that, regardless of the moisture content, \({\hbox {H}_{2}/\hbox {CO}},\, {\hbox {CO}_{2}/\hbox {CO}}\) and \(\hbox {CxHy}/\hbox {CO}\) ratios are around 0.7, 0.8 and 1.2, respectively.

The equivalent ratio (ER) parameter indicates whether a fuel–oxidizer mixture is rich, lean or stoichiometric [39]. The calculation takes into account the air–fuel ratio for the complete combustion of the fecal materials (0%, 10% and 20% of moisture) and the theoretical amount of dry air (Eqs. 4, 5 and 6).

$$\begin{aligned}&(40.85 {\,}\hbox {C} + 73.51 {\,}\hbox {H} + 24.39 {\,}\hbox {O} + 3.21 {\,}\hbox {N} + 0 {\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH})\\ &\quad + 47.04 {\,}(\hbox {O}_{2} + 3.76 {\,}\hbox {N}_{2}) \rightarrow \, 40.85 {\,}\hbox {CO}_{2}\\ &\quad + 36.76 {\,}\hbox {H}_{2}\hbox {O} + 178.56 {\,}\hbox {N}_{2} + \hbox {ASH} \end{aligned}$$
(4)
$$\begin{aligned}&(36.77 {\,}\hbox {C} + 66.16 {\,}\hbox {H} + 21.95 {\,}\hbox {O} + 2.98 {\,}\hbox {N} + 5.55 {\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH})\\ &\quad + 42.33 {\,}(\hbox {O}_{2} + 3.76 {\,}\hbox {N}_{2}) \rightarrow \, 36.77 {\,}\hbox {CO}_{2}\\ &\quad + 38.63 {\,}\hbox {H}_{2}\hbox {O} + 160.70 {\,}\hbox {N}_{2} + \hbox {ASH} \end{aligned}$$
(5)
$$\begin{aligned}&(32.68 {\,}\hbox {C} + 58.81 {\,}\hbox {H} + 19.51 {\,}\hbox {O} + 2.57 {\,}\hbox {N} + 11.10 {\,}\hbox {H}_{2}\hbox {O} + \hbox {ASH})\\&\quad + 37.63{\,} (\hbox {O}_{2} + 3.76 {\,}\hbox {N}_{2}) \rightarrow {\,} 32.68 {\,}\hbox {CO}_{2}\\&\quad + 40.51 {\,}\hbox {H}_{2}\hbox {O} + 142.85 {\,}\hbox {N}_{2} + \hbox {ASH} \end{aligned}$$
(6)

Finally, to obtain the syngas mass (\(m_\mathrm{syn}\)), the mass balance equation (Eq. 7) is applied considering the mass of the feces from the batch process (\(m_\mathrm{fuel}\)), the mass of ash recovered at the end of the experiment (\(m_\mathrm{ash}\)) and the mass of air (\(m_\mathrm{air}\)) measured at the inlet.

$$\begin{aligned} m_\mathrm{fuel} + m_\mathrm{air} = m_\mathrm{ash} + m_\mathrm{syn} \end{aligned}$$
(7)

3.3.2 Energy balance

Figure 7 shows the molar composition (\(y_i\)) and the temperature of the syngas (\(T_\mathrm{syn}\)) for each moisture content, which are going to be used in the energy balance. The ash temperature (\(T_\mathrm{ash}\)) used in the model is the same as the syngas temperature, and it is treated as sand, modeled as an incompressible substance, with specific heat capacity (\(c_{p,\mathrm{ash}}\)) obtained from [40].

The lower heating value of the raw material (\(\mathrm{LHV}_\mathrm{fuel}\)) in wet basis is obtained from its higher heating value (\(\mathrm{HHV}_\mathrm{fuel}\)) in dry basis, proximate analysis and ultimate analysis (Table 2). Equation 8 is applied, where W is the moisture mass fraction and X is the hydrogen mass fraction in wet basis.

$$\begin{aligned} \mathrm{LHV}_\mathrm{fuel} = (1 - W)(\mathrm{HHV}_\mathrm{fuel} - 9\cdot X \cdot 2440) - (W \cdot 2440) \end{aligned}$$
(8)

Regarding the energy entering and leaving the control volume, it is necessary to make some considerations: The air enters the gasifier at the standard reference temperature, \(T_\mathrm{ref} = {25}\, {^{\circ }}{\hbox {C}}\); the syngas is modeled as an ideal gas mixture; specific enthalpies (\(h_i\)) and molar masses (\(M_i\)) of each syngas component i are retrieved from the EES Ideal Gases package. With that stated, the energy that enters the gasifier with the feedstock (\(H_\mathrm{fuel}\)) is given by Eq. 9. The energy that exits with syngas (\(H_\mathrm{syn}\)) and ash (\(H_\mathrm{ash}\)) is given by Eqs. 10 and 11, respectively. The energy balance of the process is given by Eq. 12, also considering the heat losses through the walls (\(Q_L\)).

$$\begin{aligned} H_\mathrm{fuel}= & {} m_\mathrm{fuel} \cdot \mathrm{LHV}_\mathrm{fuel} \end{aligned}$$
(9)
$$\begin{aligned} H_\mathrm{syn}= & {} \frac{m_\mathrm{syn}}{\sum y_{i} \cdot M_{i}} \cdot \sum y_{i} \cdot M_{i} \cdot [h_{i}(T_\mathrm{syn})-h_{i}(T_\mathrm{ref})] \end{aligned}$$
(10)
$$\begin{aligned} H_\mathrm{ash} & = m_\mathrm{ash} \cdot c_{p,\mathrm{ash}} \cdot (T_\mathrm{ash}-T_\mathrm{ref}) \end{aligned}$$
(11)
$$\begin{aligned} H_\mathrm{fuel}= & {} H_\mathrm{syn} + H_\mathrm{ash} + Q_{L} \end{aligned}$$
(12)

Therefore, the main performance indicator of gasifiers is computed by the cold gas efficiency (Eq. 13), which takes into account the lower heating value of the syngas (\(\mathrm{LHV}_\mathrm{syn}\)), given by Eq. 14, and the lower heating value of the fuel, calculated previously.

$$\begin{aligned} CGE= & {} \frac{\mathrm{LHV}_\mathrm{syn}}{\mathrm{LHV}_\mathrm{fuel}} \end{aligned}$$
(13)
$$\begin{aligned} \mathrm{LHV}_\mathrm{syn} & = \frac{\sum y_{i} \cdot M_{i} \cdot \mathrm{LHV}_{i}}{\sum y_{i} \cdot M_{i}} \end{aligned}$$
(14)

3.3.3 Exergy balance

The exergy analysis was carried out to determine the source, location and magnitude of inefficiencies related to the gasification process. Equation 15 gives the specific chemical exergy for a solid fuel (\(b_\mathrm{fuel}^\mathrm{CH}\)), where the term \(\phi _\mathrm{dry}\) is calculated from Eq. 16. The mass fractions ratios H/C, N/C and O/C are on wet basis.

$$\begin{aligned} b_\mathrm{fuel}^\mathrm{CH} & = (\mathrm{LHV}_\mathrm{fuel} + 2442\, W) \cdot \phi _\mathrm{dry} \end{aligned}$$
(15)
$$\begin{aligned} \phi _\mathrm{dry} & = \frac{1.0438 + 0.1882 (H/C) - 0.2509 (1 + 0.7256 H/C)+ 0.0383 (N/C)}{1 - 0.3035 (O/C)} \end{aligned}$$
(16)

The specific physical exergy of the syngas (\(b_\mathrm{syn}^\mathrm{PH}\)) is given by Eq. 17, considering, for the sake of simplicity, an average specific heat capacity of the syngas, \(c_{p,\mathrm{syn}}\) (Eq. 18). Equation 19 gives the specific chemical exergy of the syngas (\(b_\mathrm{syn}^\mathrm{CH}\)). The standard specific molar chemical exergy of each syngas component (\(\bar{b}_{i}^{\mathrm{CH},0}\)) was taken from [25].

$$\begin{aligned} b_\mathrm{syn}^\mathrm{PH} & = c_{p,\mathrm{syn}} \cdot (T_\mathrm{syn}-T_{0}) - T_{0} \cdot c_{p,\mathrm{syn}} \cdot ln \frac{T_\mathrm{syn}}{T_{0}} \end{aligned}$$
(17)
$$\begin{aligned} c_{p,\mathrm{syn}} & = \frac{H_\mathrm{syn}}{m_\mathrm{syn} \cdot (T_\mathrm{syn} - T_{0})} \end{aligned}$$
(18)
$$\begin{aligned} b_\mathrm{syn}^\mathrm{CH} & = \frac{\sum _{i}^{} y_{i} \cdot \bar{b}_{i}^{\mathrm{CH},0} + \bar{R}.T_{0} \cdot \sum _{i}^{} y_{i} \cdot ln y_{i} }{\sum _{i}^{} y_{i} \cdot M_{i}} \end{aligned}$$
(19)

To obtain the irreversibility of the gasification process, the exergy balance equation is applied (Eq. 20). The total irreversibility (\(I_{cv}\)) is the sum of the exergy destruction, due to the entropy generation, and the exergy loss associated with the ashes and the stray heat transfer.

$$\begin{aligned} I_{cv} = m_\mathrm{fuel} \cdot b_\mathrm{fuel}^\mathrm{CH} - m_\mathrm{syn} \cdot (b_\mathrm{syn}^\mathrm{PH} + b_\mathrm{syn}^\mathrm{CH}) \end{aligned}$$
(20)

Another two exergy-based indexes are used herein. The first one is the cold gas exergetic efficiency, \(\varepsilon _\mathrm{cold}\) (Eq. 21), which takes into account the syngas chemical exergy only. The other one is the hot gas exergetic efficiency, \(\varepsilon _\mathrm{hot}\) (Eq. 22), which takes into account both physical and chemical exergy of the syngas.

$$\begin{aligned} \varepsilon _\mathrm{cold} & = \frac{m_\mathrm{syn} \cdot b_\mathrm{syn}^\mathrm{CH} }{m_\mathrm{fuel} \cdot b_\mathrm{fuel}^\mathrm{CH}} \end{aligned}$$
(21)
$$\begin{aligned} \varepsilon _\mathrm{hot} & = \frac{m_\mathrm{syn} \cdot (b_\mathrm{syn}^\mathrm{PH} + b_\mathrm{syn}^\mathrm{CH})}{m_\mathrm{fuel} \cdot b_\mathrm{fuel}^\mathrm{CH}} \end{aligned}$$
(22)

Table 4 summarizes the calculation results from the whole thermodynamic analysis, considering the 0%, 10% and 20% moisture for human feces samples. While quantifying experimentally the heat losses during the gasification process, it was found a large amount of energy leaving the system, around 85%, regardless of the moisture content. For this reason, it is not desirable to establish a comparison with works developed under the assumptions of the adiabatic reactor and equilibrium modeling [12, 19, 41,42,43]. These assumptions lead to an inaccurate syngas yield, since the temperature levels become much higher than reality and sometimes unreal, carrying the Gibbs minimization process to overestimate \(\hbox {CO}\) and \(\hbox {H}_{2}\) and underestimate or even neglect \(\hbox {CxHy}\) concentration. Specifically for low-rank solid fuels, the ratio \(\mathrm{LHV}_\mathrm{syn}/\mathrm{LHV}_\mathrm{fuel}\) is unrealistic, because \(\mathrm{LHV}_\mathrm{syn}\) is overestimated.

Table 4 Calculation results from the exergy analysis, for samples with 0, 10 and 20% of moisture
Full size table

Regarding the irreversibilities (\(I_\mathrm{CV}\)), it is noted that it decreases with the increase in moisture content. The explanation lies in the amount of \(\hbox {O}_{2}\) that exits with the syngas (more than 3%, for 0% of moisture content—Fig. 7). The volatile compounds are responsible for adding \(\hbox {O}_{2}\) radicals during the process, leading to local oxidation reactions inside the pyrolysis zone, as suggested by the exothermic behavior of pyrolysis. The main drawback of \(\hbox {O}_{2}\) radicals released is to reduce the syngas quality and increase the exergy destruction. In this case, there is a direct relationship between the \(\hbox {O}_{2}\) concentration and the ER. Zainal et al. [23] showed that by increasing the ER (from 0.25 to 0.37), the \(\hbox {O}_{2}\) composition in the syngas is reduced (from 4.2 to 1.9 vol.%). Even so, when looking out the wood-dominant literature, e.g., refuse-derived fuel gasification, the opposite trend is observed [44].

Figure 8 displays the evaluation in terms of the physical and chemical exergy and the associated irreversibility, normalized by kg of fuel. Noteworthy, to the best of the authors, only [19] conducted a rigorous exergy analysis providing the data for comparison. While the irreversibility levels are close to other solid fuels, the physical exergy is higher than the chemical exergy, expressing the potential of energy that can be recovered from the streams of syngas, ash and walls. On the other hand, the chemical exergy is lower because the exergy destruction is accentuated due to unexpected chemical reactions, such as homogeneous/heterogeneous oxidations in the pyrolysis zone.

Fig. 8
figure 8

Comparison between chemical exergy, physical exergy and irreversibility from batch gasification process of human feces and other solid fuels

Full size image

Despite the low-rank syngas produced, \(\mathrm{LHV}_\mathrm{syn}\) around 6.5 MJ/kg of fuel and CGE in the range of 30–38\(\%\) (Table 4), it is still possible to use it in internal combustion engines due to favorable ratios of \(\hbox {H}_{2}/\hbox {CO}\), \(\hbox {CO}_{2}/\hbox {CO}\) and \(\hbox {CxHy}/\hbox {CO}\) [45] and due to the high exergy efficiency (around 85%), which carries into calculation the ratio of \(m_\mathrm{syn}\) and \(m_\mathrm{fuel}\). The solid and clean product is a thoroughly sanitized material that brings a negligible energetic impact on the process and great chemical exergy if used as a feedstock for a secondary process.

4 Conclusion and future perspective

The gasification of human feces showed some peculiarities along the whole process, from the raw material, with a strong odor and dark coloration, to an inert and cleaned material at the end. Yet, before a wide discussion, it is important to have a critical eye about some assumptions used in the literature to establish a fair comparison between the ideal and experimental-based gasification modeling.

A series of thermochemical characterization, from the raw material to the products of gasification, allowed to establish a factual thermodynamic assessment of the whole gasification process.

Regardless of the heterogeneous characteristics of the material, the average proximate analysis showed around \(25\%\) of fixed carbon and \(50\%\) of volatile matter. Part of this volatile matter, released at temperatures around \({150}\, {^{\circ }}{\hbox {C}}\), could be associated with the exothermic characteristic of feces pyrolysis. The ash composition showed that some important minerals are trapped in the form of oxide, conferring a valuable asset to the material. Unlike the equilibrium models, which boost \(\hbox {CO}\) and \(\hbox {H}_{2}\) composition, the online gas analysis revealed that \(\hbox {CO}\), \(\hbox {H}_{2}\), \(\hbox {CxHy}\) are effectively present, despite the low-rank syngas produced.

The outcome of this study is the critical relationship of moisture content of the feedstock and air, and the energy loss of the gasification process. The interplay suggests an unfavorable application in terms of energetic efficiency, varying from 30% (dry feces) to 38% (\(20\%\) moisture); however, when the environmental bias is added together with the exergy analysis (exergy efficiency around 85%, regardless of the moisture content), the process becomes more attractive due to the potential of energy recovery and the possible use of small-scale gasifiers as a sanitation tool for materials with the moisture content up to the limitation of the device.

Looking into the future, many research efforts can be done to address the gasification of fecal excreta. Two main challenges arise from the present work and must receive considerable attention in the years to come:

  • The concept of using gasification as on-site thermal technology combined with pit latrines and dry toilets. In low- and middle-income countries, the collection of human feces can be pointed as one of the most complex waste management problems. Therefore, the tanks of pit latrines and dry toilets [46], commons in these countries, must be part of the same system. The application seems favorable to the dry toilets since it separates the urine from feces.

  • The direct conversion of raw human feces using existent thermal technologies. Even though the high physical exergy of the process shows a possibility to bring the moisture content to acceptable levels [23, 39, 47, 48], the direct conversion dealing with high water content is still a challenge. In this context, the concept of waste heat recovery must be addressed [49].