Modeling and model performance evaluation of sewage sludge gasification in fluidized-bed gasifiers using Aspen Plus

ABSTRACT

A model was developed to simulate the sewage sludge gasification in an atmospheric fluidised bed gasifier using Aspen Plus. The model here presented was based on the Gibbs free energy minimisation and the restricted equilibrium method was used to calibrate it against previously published experimental data obtained in a lab-scale gasification plant. A sensitivity analysis of the model was carried out by modifying parameters such as the temperature, equivalence ratio (ER) and the steam-to-biomass ratio. The modeled results were in good agreement with the experimental data (especially when air was used as gasifying agent) and reproduced satisfactorily the experimental trends found for the gas composition, the carbon conversion (Xc) and the cold gas efficiency (CGE) under different gasification conditions. Operating at higher temperatures increased the production of H₂ and CO, as well as the Xc and the CGE. The increase in ER produced higher Xc, yet the CGE experienced slight changes due to a decrease in the lower heating value of the resulting syngas, as well as the oxidation of combustible gases. The use of air+steam as gasifying agent increased the H₂ content of the produced gases but decreased the accuracy of the model.

Implications: Gasification is an available alternative to produce energy as well as several raw materials from sewage sludge. The syngas obtained from this technology totally depends on the type of gasifier and the operation conditions, which can be optimized with the help of models. In this work, a relatively simple model was built using ASPEN PLUS. Despite its simplicity, the outputs of the model are in good agreement with experimental results what makes its use interesting for assessing scaling-up possibilities from lab-scale to pilot-scale gasification processes.

Introduction

Sewage sludge is the residue, liquid or semiliquid form that is generated from the treatment of domestic and industrial wastewater (Dogru, Midilli, and Howarth 2002). The chemical composition of sewage sludge is determined by the origin of the raw wastewater, and its production is influenced by population growth and legislation standards (Urban Waste Water Treatment Directive 91/271/EEC [UWWTD] 1991). Although specific types of sludge can occasionally be used as a fertilizer, sludge is normally considered a waste that requires appropriate management or further treatment (Sewage Sludge Directive 86/278/EEC [SSD] 1986). The most common disposal solutions for sewage sludge are landfilling or incineration, which are restricted by the European legislation (Council Directive 99/31/CE [CD] 1999 ) and often have a negative public perception associated with their potential environmental risk.

An alternative to these conventional methods is the gasification of sewage sludge. This technology presents advantages such as the destruction of pathogenic bacteria and sludge volume reduction, as well as the additional benefits of energy recovery and lower-cost atmospheric emission controls (Seggiani et al. 2012). Gasification is a thermal process that consists of the conversion of the organic contents of sludge into combustible gas and other substances, in the presence of a gasifying agent at high temperatures under a reducing atmosphere (Fytili and Zabaniotou 2008). The produced gas (syngas) is mainly composed of carbon monoxide (CO), hydrogen (H₂), carbon dioxide (CO₂), water vapor, methane, and light hydrocarbons. In addition to the syngas, a solid carbonaceous residue (char) is produced along with a gaseous mixture of condensable organic compounds (tars). Syngas is a high-value product of the sludge gasification process, as it can be burnt in gas engines and turbines for energy production, used as a raw material in methanol synthesis, or used in the production of synthetic fuels through the Fischer-Tropsch process (Ciferno and Marano 2002; Holmgren et al. 2014; Princiotta 2009; Speight 2015).

Although gasification can be an autothermic process, most of the studies at laboratory and pilot scales have been developed allothermically by modifying independently the temperature, the air-to-biomass ratio, and the steam-to-biomass ratio (Campoy et al. 2008). In addition to the above-mentioned parameters, the gasifying agent and gasifier chosen are also critical aspects that influence the process and condition the quality of the obtained syngas. The most commonly used gasifying agents are air, oxygen, steam, or a mixture of these (Campoy et al. 2010; Meng et al. 2011). Regarding the type of gasifier, the most common reactors are fixed, moving, and fluidized-bed reactors (Beenackers 1999; Warnecke 2000). Among these, fluidized-bed gasifiers have a uniform temperature distribution in the gasification zone and elevated heat transfer rates, which are typical of the circulation patterns of gas-solid systems within the bed (McKendry 2002). Another important advantage of these gasifiers is their higher potential to be scaled up to medium and large scales, overcoming the limitations often found in smaller-scale, fixed-bed gasifiers (Kurkela, Nieminen, and Simell 2004).

The selection of equipment and operation conditions of a gasification system for an optimal performance needs to take into account the complex interplay between all the factors that condition the process. In this sense, models are useful tools for designing, optimizing, and assessing scaling-up possibilities for the gasification processes (Gómez-Barea and Leckner 2010). However, when dealing with fluidized-bed gasifiers, modeling can be difficult due to the complexity of the fluidization hydrodynamics and the complex nature of the physical and chemical phenomena occurring inside a gasifier with these characteristics (Damartzis, Michailos, and Zabaniotou 2012).

Three main types of modeling approaches are used for describing sludge gasification: (i) computational fluid dynamic (CFD) models, (ii) kinetic models, and (iii) equilibrium models (Champion et al. 2014; Gambarotta, Morini, and Zubani 2017). CFD models can simulate several physical phenomena, but they require very detailed information (geometry, materials, boundary conditions) and high computational resources. Kinetic models are appropriate for modeling the influence of reactor design and process parameters, but they are substantially more complex in their conceptual modeling (Petersen and Werther 2005a). For instance, Petersen and Werther (2005b) developed a kinetic model for sludge gasification in a circulating fluidized-bed gasifier that included the hydrodynamics of the circulating bed and a comprehensive set of gasification reactions. However, this model also required carrying out axial measurements of temperature and gas concentration to develop the kinetic rate expressions.

Equilibrium models, on the contrary, are simple, easy to program, and provide a good approximation of the gasification process within a reactor (Mostoufi, Cui, and Chaouki 2001). These models are classified as stoichiometric and nonstoichiometric depending on the whether they are based on equilibrium constants or on the minimization of the Gibbs free energy (Gambarotta, Morini, and Zubani 2017). Ultimately, the choice of model depends mainly on the objectives and the experimental information available (Gómez-Barea and Leckner 2010).

Nonstoichiometric equilibrium models are the most common approach for the description of the performance of a fluidized-bed gasifier. As an example, Ramzan et al. (2011) developed a nonstoichiometric equilibrium model for a biomass gasification process, and the model was validated using experimental results from hybrid biomass gasifier. In this paper, a comprehensive description of a nonstoichiometric equilibrium model that describes the performance of a laboratory-scale fluidized-bed gasifier is presented. The model is based on Aspen Plus and has been built to predict the output of gasification products from sludge under different operating conditions and has been validated against experimental data from the actual pilot-plant gasifier (De Andrés, Narros, and Rodríguez 2011). Although there are several examples of Aspen Plus being used for modeling biomass gasification processes (Baruah and Baruah 2014; Damartzis, Michailos, and Zabaniotou 2012; Doherty, Reynolds, and Kennedy 2013; Nikoo and Mahinpey 2008; Ramzan et al. 2011; Zheng, Kaliyan, and Morey 2013), scientific literature on its application in sewage sludge gasification is still scarce (Lumley et al. 2014; Ptasinski, Hamelinck, and Kerkhof 2002). Ultimately, the development of such a model is essential for finding optimal operation conditions that could inform the potential up-scaling of the gasifier to produce syngas from sewage sludge from several wastewater treatment facilities of the local water supplier (Canal de Isabel II) of the region of Madrid.

Sludge gasification model

The nonstoichiometric equilibrium model of this work is based on the fundamental chemical equilibrium problem, which is formulated as follows: given temperature T, pressure P, and molar amounts of elements i, find the equilibrium composition n that minimizes the total Gibbs free energy (G) of the system (Leal, Kulik, and Kosakowski 2016; Sreejith, Arun, and Muraleedharan 2012):

(1) $\underset{n_i}{min}\frac{G}{RT}=\sum _{i=1}^N{n}_i\frac{\mathrm{\Delta }{G}_{f,i}^0}{RT}+ln{n}_i\sum n_i$(1)

where ΔG°_f,i is the standard formation free Gibbs energy of species i and R is the ideal gas constant. The calculation of the composition of syngas requires finding the values of n_i that minimize the objective function G/RT. The equilibrium problem (eq 1) is a nonlinear constrained minimization problem, which requires not only the minimization of the Gibbs free energy but also satisfying the inequality constraint, which guarantees that the species have non-negative molar amounts (Gambarotta, Morini, and Zubani 2017; Wenger, Farouk, and Wittle 1996).

Modeling approach

The gasification model was constructed in Aspen Plus as a combination of different blocks and unit operations, aiming to reproduce as accurately as possible the experimental results from the laboratory-scale plant gasifier that is used for treating sewage sludge from the local wastewater treatment facilities of Madrid (De Andrés, Narros, and Rodríguez 2011). Once the model had been configured, analyses of the sensitivity of the yield (Y_gas), composition, and lower heating value (LHV_gas) of syngas, as well as the carbon conversion (X_C) and cold gas efficiency (CGE), to operation variables such as temperature, equivalence ratio (ER), and steam-to-biomass ratio (S/B) were carried out.

Modeling assumptions

In order to maintain a balance between the robustness of the model, its operability, and the available information, the following assumptions were made:

• Steady-state and isothermal process

• Zero-dimensional and kinetic-free model

• Instantaneous devolatilization of the biomass at the entrance of the reactor (Nikoo and Mahinpey 2008)

• The main components of the syngas are H₂, CO, CO₂, methane (CH₄), and water (H₂O)

• All the nitrogen and sulfur in the biomass react to give ammonia (NH₃) and hydrogen sulfide (H₂S) (De Jong, Andries, and Hein 1999; Schuster et al. 2001)

• Tars are not modeled but are considered as a nonequilibrium component (Panopoulos et al. 2006)

Chemical reactions

The homogeneous and heterogeneous chemical reactions that are considered to occur in the gasification process of the unit are shown in Table 1. The hydrogen and carbon combustion reactions (R-1, R-5), as well as the water-gas shift and methanation reactions (R-3, R-7), are all exothermal and ideally provide the system with the energy needed. On the other hand, the steam-reforming, Boudouard, and water-gas reactions are endothermal (R-2, R-6, R-8), and their effect on the gasification products becomes more evident at high temperatures (González et al. 2008).

Table 1. Reactions considered in the model.

Reaction	Reaction type	Reaction number
Homogeneous reactions
H2 + 0.5O2 → H2O	Hydrogen combustion	R-1
CH4 + H2O → CO + 3H2	Steam reforming	R-2
CO + H2O → CO2 + H2	Water-gas shift	R-3
0.5N2 + 1.5H2 → NH3	NH3 formation	R-4
Heterogeneous reactions
C + O2 → CO2	Carbon combustion	R-5
CO2 + C → 2CO	Boudouard	R-6
C + 2H2 → CH4	Methanation	R-7
C + H2O → H2 + CO	Water-gas	R-8
H2 + S → H2S	H2S formation	R-9

Model assembly and configuration

The model was developed by reproducing the gasification process with the software Aspen Plus, coupling and combining different modules to represent the parts of the plant (the resulting flow sheet is displayed in Figure 1). The gasifier was represented by multiple blocks that correspond to the different phases of the gasification process, namely, the feed drying, the decomposition of the sludge, and the gasification itself (Table 2).

Table 2. Description of the Aspen Plus reactor blocks considered in the model.

Aspen Plus ID	Block ID	Description
RYIELD	DRIER	Dries the biomass eliminating the moisture
	MASSPROC	Converts nonconventional biomass into conventional components
RSTOIC	NS	Reacts nitrogen and sulfur in the fuel to form NH3 and H2S
	NONEQ	Reacts part of the carbon in the fuel to form tar
RGIBBS	GASIFIER	Minimizes Gibbs free energy to reach chemical equilibrium and calculate outlet composition
	GASIF2	Restricted equilibrium to calculate syngas composition
SEP	WSEP	Separates the moisture from the fuel and sends it to the gasifier
	ELEMSEP	Separates the reacted NH3 and H2S as well as the unreacted carbon and sends them to the final outlet
	B2	Separates tar from the rest of the reaction stream
	ASHSEP	Separates ashes from the outlet of the gasifier
	SYNSEP	Separates syngas from moisture and solids
HEATER	HEATER	Heats the stream to the same temperature as that of the outlet of the gasifier by means of calculator block
	COOLER	Simulates the gas cooling to the required clean-up temperature of 298 K
MIXER	MIX	Mixes the different outlets of the blocks to reproduce a single product stream

The sewage sludge was specified as a nonconventional component and the HCOALGEN and DCOALIGT algorithms were used as enthalpy and density functions, together with the LHV and the proximate and ultimate analyses of the sludge.

The drying (removal of the residual moisture) and the decomposition of the sewage sludge into elemental constituents (C, O, H, N, S, ash) occurred in RYIELD blocks (DRIER and MASSPROC, respectively). The MASSPROC block allows the user to specify the component distribution of the outlet stream, based on the proximate and ultimate analysis of the sludge. After drying and decomposition, the moisture was sent to the GASIFIER, whereas the rest of the elements were fed to a RSTOIC block (NS). The NS was configured with chemical reactions of known stoichiometry (R-4, R-9) that can be set to a determined reaction extent or conversion. In this work, this block simulated the total conversion of nitrogen and sulfur to NH₃ and H₂S (De Jong, Andries, and Hein 1999; Schuster et al. 2001), which were then separated from the main stream by a SEP block (ELEMSEP).

The ELEMSEP block also diverts 10% of the carbon content in the feedstock, which is the quantity of unreacted C found in the sewage sludge according to the determinations carried out in De Andrés, Narros, and Rodríguez (2011). Another RSTOIC block (NONEQ) was included to consider the formation of tars. This block was not intended to estimate tar production but to exclude from the gasification process the quantity of carbon being part of tar that did not become syngas. In order to set up the production of tar in this module, temperature-dependent tar production functions from De Andrés, Narros, and Rodríguez (2011) were used. In addition, the amount of carbon in the tar was set by a calculator block considering that naphthalene is the major component of tar (Panopoulos et al. 2006). The SEP block B2 separates tar from the rest of the reaction stream.

The modeling of the gasification process was carried out with a RGIBBS reactor block, which uses the minimization of Gibbs free energy to reach phase and chemical equilibrium. Heating demands for this module are provided by the upstream drying and decomposition processes (Ramzan et al. 2011). A SEP block (ASHSEP) simulated the ash removal from the gasifier.

Subsequently, the gaseous stream (GASES1) entered another RGIBBS reactor block, where restricted equilibrium was applied to adjust the gas composition to match the experimental data. A temperature profile was specified for each considered reaction in order to set a realistic syngas composition (Gumz 1950).

The stream of syngas then entered a mixer (MIX), together with the stream of hydrogen sulfide and ammonia (SUAM), tar (TAR), and unreacted carbon (INERTC), to simulate the final outputs from the gasifier.

Finally, the cleaning and cooling of the syngas were operated by a SEP block (SYNSEP) combined with a COOLER block. The SYNSEP simulated the removal of particulate matter from the syngas, and the COOLER lowered the temperature of the syngas to less than 100 °C, which simulates the cooling of the gases in the train of isopropanol condensers (De Andrés, Narros, and Rodríguez 2011) and the necessary cooling when using the syngas as fuel for internal combustion engines (Gómez-Barea et al. 2013). The stream “SOOTDIRT” was then composed of the remaining unreacted carbon, moisture, and tar.

Model validation

The validation of the gasification model was done against experimental results from the laboratory-scale plant under the following operation conditions (test number 5 in De Andrés, Narros, and Rodríguez [2011]):

• Temperature: 800 °C

• Equivalence ratio (ER): 0.3

• Steam-to-biomass ratio (S/B): 0

• Sewage sludge input rate (Q_sludge): 1.44 g/min

• Air volume rate (Q_air) = 0.00128 Nm³/min

The validation rationale of this work was based on the relative error (ε_r), which is the ratio between the absolute error (i.e., the difference between the modeled, X_m, and experimental, X_e, values) and the experimental value (eq 2). Relative errors are useful for determining model over- or underestimations based on whether the relative error is positive or negative, respectively.

(2) ${\epsilon }_{\mathrm{r}}=\left({\mathrm{X}}_{\mathrm{m}}--{\mathrm{X}}_{\mathrm{e}}\right)/{\mathrm{X}}_{\mathrm{e}}\times 100$(2)

The variables that were compared are (i) the syngas stream dry-basis composition in terms of hydrogen (H₂), nitrogen (N₂), methane (CH₄), carbon monoxide (CO), and carbon dioxide (CO₂); (ii) the total syngas yield (Y_gas); (iii) its lower heating value (LHV_gas); (iv) the total carbon conversion (X_C); and (v) the cold gas efficiency (CGE). Using the performance of similar gasification systems from literature as a benchmark, the outputs of the model were considered acceptable if the relative errors for each of the variables were within ±15% (Wang and Yan 2009; Xiao et al. 2009).

Sensitivity analysis

Once the model had been validated, the sensitivity of the model to different operating conditions was evaluated through the study of the change in the performance of the gasification process against the variation of the operation variables stated in section Sludge gasification model. Model validation

The operation values of the selected operation variables for the sensitivity analysis were set according to De Andrés, Narros, and Rodríguez (2011) and are as follows:

• Temperature: 750, 800, and 850 °C.

• Equivalence ratio (ER): 0.2, 0.3, and 0.4.

• Steam-to-biomass ratio (S/B): For tests with air, 0; for tests with air and steam, 0.5.

This sensitivity analysis provides an indication of those variables that have a more significant influence on the desired yields and composition of the output streams of the system. In addition, the sensitivity analysis will be useful for selecting adequate operation conditions of a gasification system, which can then be used in plant up-scaling exercises.

Experimental

Materials

The dried sludge samples were supplied by a sludge thermal drying plant managed by the local water company of Madrid (Spain). The results of the proximate and ultimate analyses of the sludge samples are shown in Table 1S (supplementary material).

Laboratory-scale plant

The plant on which this model is based mainly consists of a stainless steel fluidized-bed gasifier followed by a freeboard, both of which are electrically heated. A cyclone and a micronic filter are located downstream of the freeboard (inside a hot box) to remove any particles from the syngas. A diagram of the laboratory-scale plant is shown in Figure 1S (supplementary material).

A comprehensive explanation of the experimental procedure, tar collection, and quantification system, as well as how syngas production and composition was determined, can be found in De Andrés, Narros, and Rodríguez (2011).

Results and discussion

Model validation

As previously mentioned in the section Sludge gasification model. Model validation, the gasification model of this work was validated against experimental data collected from the laboratory-scale sludge gasification plant described in De Andrés, Narros, and Rodríguez (2011), operating with air (ER = 0.3) at 800 °C. As shown in Table 3, the model was in very good agreement with the experimental results (the percentage error was always below 12%), particularly as regards the syngas composition.

Table 3. Model validation: Comparison of experimental (De Andrés, Narros, and Rodríguez 2011) and model results.

Parameter	Experimental	Model	% Error
Syngas composition (% v/v)
H2	10.41	10.41	−0.01
CO	7.97	8.03	0.72
CO2	14.10	14.08	−0.19
CH4	3.01	3.24	7.45
N2	60.81	58.18	−4.33
LHVgas (MJ/Nm^3)	3.33	3.72	11.95
CGE (%)	37.25	38.90	4.43

Sensitivity analysis

Sensitivity analyses were performed to evaluate the model response to changes in temperature, ER, and SB. The following sections discuss the behavior of the parameters of interest with respect to the operation variables.

Influence of temperature

Figure 2 shows the influence of the gasification temperature on the gas composition, Y_tar and X_C, in tests with air and ER = 0.3. Within a temperature range of 750–850 °C, the model corresponded with the experimental data of gas composition, except for the CO content, which was underpredicted at 750 °C temperature and overpredicted at 850 °C. In both modeled and experimental results, the concentrations of H₂ and CO in the syngas increased with higher temperatures, whereas the concentration of CO₂ decreased. The concentration of methane showed a maximum around 850 °C.

Figure 1. Comprehensive Aspen Plus flow sheet for the fluidized-bed gasification process.

Figure 2. Effect of the gasification temperature on (a, b, c, d) syngas composition, (e) gas production (Y_gas), and (f) carbon conversion (X_C).

The gas production increased with temperature due to a more intense volatilization of the sludge and conversion of the char (Campoy et al., 2009). The highest deviations of the modeled results from the experimental values were found for carbon conversion. In De Andrés, Narros, and Rodríguez (2011), the ethane (C₂H₆) and ethylene (C₂H₄) contents of the syngas were included in the estimation of X_C. However, the model has not considered the C_nH_m gases, which would partially explain these differences. If C_nH_m gases are excluded from the X_C calculations in De Andrés, Narros, and Rodríguez (2011) (Figure 2f), the data fit between modeled and experimental results improved. At 850 °C, the higher CO content (Y_gas and X_C) modeled together with the lower CO₂ content modeled may be associated with the excessive promotion of the Boudouard reaction by the model.

Influence of the equivalence ratio

The effect of the ER in the experimental and modeled results is shown in Figure 3. The represented figures correspond to tests with air and at 800 °C. It can be observed that the modeled results were in good agreement with the experimental data. Figure 3 shows that H₂, CH₄, and CO contents decreased when ER increased. The oxidation reactions were stimulated with the increase in ER, which led to lower H₂, CH₄, and CO productions and to a slightly higher CO₂ content (De Andrés, Narros, and Rodríguez 2011). With regards to the Y_gas and X_C, the effect of the ER was similar to that of the temperature, that is, higher ER increased the value of these two parameters.

Figure 3. Effect of the equivalence ratio (ER) on (a, b, c, d) syngas composition, (e) gas production (Y_gas), and (f) carbon conversion (X_C).

Effect of air + steam as gasifying agent

Figure 4 shows the comparison of modeled and experimental results when a mixture of air + steam was used as gasifying agent (S/B = 0.5). As explained in De Andrés, Narros, and Rodríguez (2011), the addition of steam increased hydrogen production due to the steam reforming, water-gas shift, and water-gas reactions (R-2, R-3, R-8).

Figure 4. Comparison of modeled and experimental results with air + steam as gasifying agent (S/B = 0.5): (a, b, c, d) syngas composition; (e) gas production (Y_gas); (f) carbon conversion (X_C).

The trends obtained by the model followed experimental data satisfactorily, although with less accuracy than in tests with air. It can be explained by the fact that using air + steam decreased the residence time of the gases inside the reactor, preventing the ongoing reactions in the gasifier from reaching the chemical equilibrium assumed by the model. According to Figure 4, the maximum quantitative differences were found for CO (at lower temperatures) and for H₂ content (at higher temperatures), which are the main reagents and products of the chemical reactions promoted by the presence of the steam in the gasifier. In relative terms, the highest error was found for CH₄ (94%). Lumley et al. (2014) obtained similar results, with relative errors of 112% and 78% for modeled CH₄ production for sewage sludge gasification with air and steam, respectively. According to these authors, thermodynamic equilibrium methods tend to overpredict CH₄, yet they remain the best modeling choice for interrogating system designs at an early stage.

Despite the observed differences in gas composition, the modeled Y_gas and X_C fitted reasonably well with the experimental values (Figure 4e, f, respectively).

Energetic evaluation

The experimental and modeled trends of the LHV_gas and the CGE at different temperatures and ERs were compared (Figures 5 and 6, respectively). According to the results obtained, both energetic parameters are more sensitive to variations of the temperature than to variations of ER. As it can be seen in Figure 5a, the modeled LHV_gas at different temperatures agreed with the experimental data. The relative errors in modeled results were 15.2% for LHV_gas and 18.0% for CGE. The operation at higher temperatures increased the production of combustible gases (H₂, CO, and CH₄) and therefore increased the LHV_gas. At the same time, as shown in Figure 2e, these higher operation temperatures produced an increase of Y_gas and therefore improved the CGE (Figure 5b).

Figure 5. Modeled and experimental results under different temperatures: (a) lower heating value (LHV_gas); (b) cold gas efficiency (CGE).

Figure 6. Modeled and experimental results under different equivalence ratios: (a) lower heating value (LHV_gas); (b) cold gas efficiency (CGE).

As for the LHV_gas and the CGE obtained at different ERs, it can be observed in Figure 6 that both modeled parameters were in good agreement with the experimental data. The relative errors in modeled results were 4.2% for LHV_gas and 5.6% for CGE. The ER had the opposite effect on the LHV_gas than the temperature, that is, higher ER decreased the LHV_gas because of the oxidation of part of the combustible gases present in the syngas (Figure 6a). The changes in the ER had a negligible effect on the CGE (Figure 6b), which can be explained by the fact that when the ER increased the Y_gas rose but the LHV_gas decreased.

Conclusion

Despite the relative simplicity of the model developed with Aspen Plus, its outputs were generally in good agreement with the experimental results used for model validation and sensitivity analysis.

According to the modeled results with air, increasing the gasification temperature above 850 °C may not produce large benefits regarding the syngas lower heating value (LHV_gas), carbon conversion (X_C), and cold gas efficiency (CGE) but would increase the H₂ and CO contents of the produced gas.

The best data fit between modeled and experimental results was found when the temperature was kept at 800 °C and the equivalence ratio (ER) varied between 0.2 and 0.4. Higher ER increased both the gas production (Y_gas) and the X_C but had a negligible effect on the CGE because of the decrease of the LHV_gas. Considering the information gathered, the optimal gasification conditions could be set at 850 °C and ER of 0.3.

The addition of steam decreased the model accuracy (specially for modeled gas composition), which can be explained by the fact that using air + steam decreased the residence time of the gases inside the reactor, preventing the ongoing reactions in the gasifier from reaching the chemical equilibrium assumed by the model. Despite this, the modeled Y_gas and X_C fitted reasonably well with the experimental values.

With these results in mind and despite their limitations, the thermodynamic equilibrium models may be considered as an interesting modeling choice for interrogating system designs at an early stage.

Abbreviations and acronyms

daf	=	dry and ash-free
ER	=	equivalence ratio, defined as the ratio between the flow rate of air introduced into the gasifier and the stoichiometric flow rate of air required for complete combustion of the sludge
LHVgas	=	lower heating value of the produced gas, MJ/Nm3, dry basis
SB	=	steam-to-biomass ratio, defined as the flow rate of steam fed to the reactor divided by the flow rate of sludge (daf)
Nm3	=	cubic meter, normal conditions (0 °C, 101 kPa)
Ygas	=	gas yield, Nm3 dry gas/kg sludge, daf
XC	=	carbon conversion, weight of carbon in the produced gas divided by weight of carbon in the sludge introduced in the gasifier
CGE	=	cold gas efficiency = LHV of gas divided by the LHV of sludge

Supplemental data

Supplemental data for this paper can be accessed on the publisher’s website.

Additional information

Funding

This work was carried out as part of the research project entitled “Optimization of Municipal Waste Management” (reference number CTQ2013-48280-C3-2-R), granted by Spain’s Ministry of Economics and Competitiveness.

Notes on contributors

Juan Manuel de Andrés

Juan Manuel de Andrés, Ph.D., is an assistant lecturer at the Universidad Politécnica de Madrid (UPM).

Michel Vedrenne

Michel Vedrenne, Ph.D., is a principal technical consultant at Ricardo Energy & Environment in London.

Matteo Brambilla

Matteo Brambilla is Part-time research fellow at the Universidad Politécnica de Madrid (UPM).

Encarnación Rodríguez

María Encarnación Rodríguez, Ph.D., is a senior scientist and the head of the UPM research group on environmental technologies and industrial resources.