Pyrolysis of domestic sewage sludge: influence of operational conditions on the product yields using factorial design.

This study aims to investigate a sustainable method for sewage sludge (SS) safe disposal and reuse. The study involved exploring the optimum parameters of thermal treatment of SS by pyrolysis to produce biochar. Based on the analysis of the full factorial design, the effects of pyrolysis conditions: temperature, heating rate, and isothermal time on pyrolysis product yields were evaluated. The average yield of biochar was significantly reduced when the pyrolysis temperature was increased from 300 to 500 °C, while the average yields of bio-oil (BO) and non-condensable gases (NCGs) were increased. The yield of biochar was nearly the same when the heating rate was increased from 5 to 35 °C/min, while the yield of BO was increased and the yield of NCGs was decreased. The average yields of biochar and NCGs were reduced when the isothermal time was increased from 45 to 120 min, while the yield of BO was slightly increased. Factorial design methodology revealed all potential interactions between the variables of the pyrolysis process of SS. To predict pyrolysis product yields, first-order regression models were developed based on the effects' magnitude of the process parameters and their interactions. The models were agreed to the experimental data.

Introduction

Sewage sludge (SS) is becoming a global critical environmental problem since it may contain hazardous substances of heavy metals (Wahi et al., 2006; Xie et al., 2014), microplastics (Carr et al., 2016), and organic pollutants of xenobiotic nature (Luo et al., 2014; Petrie et al., 2015). In Jordan SS quantities are estimated to reach about 180.7 tons/day by 2022, this amount is equivalent to 65,955.5 tons/year. The natural decomposition of the accumulated sludge in Jordan resulted in emission of about 105 ton of CO2 per year (Aljbour et al., 2021a). The current management of SS in Jordan lacks any treatment method except dewatering and air drying. The SS in 32 wastewater treatment plants (WWTP) out of 34 is stored at the same plants, then transported to the local landfills (Breulmann et al., 2019). This practice can cause serious environmental and economic problems, as well as an adverse effect on public health and safety (Al-Hamaiedeh, 2010; Aljbour et al., 2021a, 2021b). The high level of pathogens and high possibly concentration of heavy metals in SS, besides the high generation rates, necessitates proper management of this waste (Agrafioti et al., 2013). Therefore, to overcome the above problems it is important to improve the management of SS in Jordan. This might be achieved by introducing new sustainable methods for SS treatment, which assist the reuse and/or safe disposal of SS. A promising idea from both ecological and economical perspectives is the transformation of SS into a carbonaceous final product material namely biochar (Breulmann et al., 2015). This can be achieved by thermal treatment of SS using pyrolysis. The thermal upgrade of biomass can be divided into three main methods: pyrolysis, carbonization, and torrefaction (Poudel et al., 2015; Wilk and Magdziarz, 2017; Chen et al., 2017). The produced biochar can be utlized in a wide range of applications, such as soil amendment and remediation (Beiyuan et al., 2020; O'Connor et al., 2018; Shen et al., 2018; Kumar et al., 2021), sewage treatment (Palansooriya et al., 2020; Wu et al., 2020; Zhang et al., 2020), contaminated air treatment (Klasson et al., 2014; Wang et al., 2020a, 2020b) and climate change mitigation (Dissanayake et al., 2019; Igalavithana et al., 2019).

Pyrolysis is an endothermic reaction that embraces the thermal cracking of the organic material at elevated temperatures in an inert atmosphere (Shabangu et al., 2014). The main end-products of pyrolysis are biochar as a solid product, bio-oil (BO) as a condensable vapor, and gas as a non-condensable product which basically consists of CO2, CO, H2, and CH4 (Aljbour, 2018; Aljeradat et al., 2021). BO is typically used as an upgraded and refined fuel (Bridgwater et al., 2007). Biochar production is desirable from the perspective of nutrient recycling and can be used to enhance fertility of agricultural soils (Mohawesh et al., 2018, 2021).

Recently novel pyrolysis methods such as microwave-assisted pyrolysis (Kong et al., 2019; Li et al., 2016), co-pyrolysis and wet pyrolysis (Zhou et al., 2019a, 2019b), have been extensively investigated. Moreover, new trends in biochar pyrolytic production with focus on the effects of feedstock and pyrolysis conditions on biochar physicochemical properties have been emerged (Kumar et al., 2020). All of these investigations focused on the quality of biochar as subjected to different operating condtions, namely: temperature, isothermal time, and heating rate. These operating conditions are major process parameters believed to influence the performance of the pyrolysis process (Aljeradat et al., 2021). Moreover, the particle size of the used feedstock is a process parameter that may affect the char decomposition (cracking) step and the formation of the long-chain hydrocarbon gases. Similar conclusion was reached while producing the pyrolytic ash of the Oil Shale (El-Hasan, 2018; El-Hasan et al., 2021). Classically, investgations with respect to the effect of operating condtions on pyrolysis are conducted following the one-variable-at-a-time (OVAT) approach. The OVAT approach involves testing process parameters, or factors, one at a time instead of multiple factors simultaneously. This approach is not favored as it might miss revealing possible interaction effects among process factors. In addition, the OVAT approach can miss optimal settings of factors (Aliedeh et al., 2021). Design of Experiments (DOE) approach is applied in this research to overcome the above mentioned drawbacks of the OVAT approach. DOE is a type of controlled experimentation in which the process factors (independent variables or inputs) are varied at different levels to see how they affect a response variable (dependent variable or output) (Aljbour, 2019). A full factorial DOE design is one of the methods for planning and executing a set of experiments to determine the effects of process input levels on process outputs. DOE is beneficial to figure out what process input levels will optimize the outputs and what combination of process factors should be used to get the best results.

Therefore, this study aims to explore the optimum pyrolysis process parameters using SS as feedstock to achieve a sustainable, economical and environmentally viable management of SS. A full factorial design methodology will be followed to identify the most influential process factors on the pyrolysis product yields. In addition, this study aims to reveal the presence or absence of any possible interaction that affects the pyrolysis operating conditions.

Materials and methods

Material

Samples of domestic SS were collected from the drying beds at Mutah wastewater treatment plant (MWWTP) in Jordan. SS samples were filled into a container (60 cm × 40 cm), dried at 105 °C for 24 h, and ground to a size less than 5 mm in diameter.

Experimental apparatus

Pyrolysis experiments were conducted in a stainless steel tubular reactor (50 mm ID). The reactor was heated externally via a split type tube furnace (CARBOLITE, Type VST 12/300, England) controlled by a PID controller. Pyrolytic products were subjected downstream the reactor to a BO capturing system, NCG collection, and metering system. BO capturing system consisted of two glass bottles immersed in an ice and salt bath to condense the BO. The NCGs were passed through a silicon tube and collected in sampling bags.

Experimental procedure

The experiments were performed under reduced air conditions. The air was withdrawn from the system by suction. A 100 g of SS are fed to the reactor then heated from room temperature to the desired final pyrolysis temperature at different heating rates. The isothermal time refers to the operational time under isothermal conditions, once the final reaction temperature was reached, the isothermal conditions were maintained for (45 or 120) min. The char yield was determined by weighting the biochar after completing the pyrolysis process. BO yield was determined by weighting the amount of oil collected in the bottles and tubings. The non-condensable gas yield was determined by mass balance. System cleaning was carried out by washing with propanol.

Samples characterization

SS samples were tested for moisture, ash, volatile matter (VM), and fixed carbon (FC) contents. Moisture content was determined based on ASTM D2216-98 (ASTM D2216-98, 2005). The samples were dried in a muffle furnace at 105 °C for 24 h and were kept in airtight cans to perform the chemical analyses. The VM and ash contents were determined according to ASTM D5142 standard method (ASTM D5142, 2009). After drying, the sample was combusted in a muffle furnace at 950 °C for 7 min to determine the VM content. The ash content was determined after the combustion at 750 °C for 6 h. The FC content was calculated as follows:

The heat of combustion of SS was determined as per the procedure of ASTM D240-19 (ASTM D240-19, 2019).

For the determination of the heavy metal concentrations, the samples were dried for 24 h at 105 °C in the oven. The samples were manually milled using a mortar grinder, then 0.2 g of samples were digested using a model CEM Mars 6™ micro digestion unit according to the analysis method mentioned in (Malwina, 2019). The concentrations of Boron (B), Lead (Pb), Potassium (K), Copper (Cu), Nickel (Ni), Manganese (Mg), Zinc (Zn), Chromium (Cr), Cadmium (Cd), Molybdenum (Mo), Selenium (Se), Arsenic (As), Sodium (Na) were determined by using inductively coupled plasma optical emission spectrometry (ICP-OES, Optima5300, Perkin Elmer, USA), while Mercury (Hg) was determined by using Cold Vapor Atomic Absorption Spectroscopy (CVAAS).

The pH value was measured according to the standard method 4500-H + B. The Electrical conductivity (EC) wa measured according to the standard method 2510 B. The surface area of SS (AMB) was determinded via the methylene blue adsorption method (Hang and Brindley, 1970).

Full factorial design analysis

A 23 type full factorial design approach was followed to study the effects of three process parameters, namely: temperature (Temp), heating rate (HR), and isothermal time (Time). Table1 shows the levels of process parameters employed in this study.

Table 1

Levels of process parameters employed in this study.

	Low level (-1)	High level (+1)
Temp	300 °C	500 °C
HR	5 °C/min	35 °C/min
Time	45 min	120 min

For each run, three process responses were determined, namely, yield of biochar (Ybiochar), yield of BO (YBO), and yield of NCGs (YNCG). The experimental runs and the process responses for the full factorial design methodology employed in this study are shown in Table 2.

Table 2

The experimental runs and responses for a 23 type full factorial design.

	Temp	HR	time	Ybiochar (wt%)	YBO (wt%)	YNCG (wt%)
Experimental runs for a 23 type full factorial design	-1	-1	-1	90.39	7.08	2.53
	+1	-1	-1	54.53	18.62	26.85
	-1	+1	-1	90.59	7.91	1.50
	+1	+1	-1	56.16	27.76	16.08
	-1	-1	+1	80.00	9.04	10.96
	+1	-1	+1	52.61	19.78	27.61
	-1	+1	+1	80.44	9.8	9.76
	+1	+1	+1	52.81	30.12	17.07
Experimental runs for model validation	0	-1	-1	68.16	15.87	15.87
	0	1	-1	74.47	11.72	11.72
	0	-1	1	65.08	16.6	16.6
	0	1	1	66.09	14.92	14.92
	-1	-1	-1.8	98.46	0	0
	-1	1	-1.8	98.59	0	0
	-1	-1	-0.6	88.43	8.82	8.82

The statistical package (Minitab 17) was utilized to estimate the means of effects and interactions. In addition, the software was utilized to construct the main effect and interaction plots.

Results and discussion

SS characterization

The results of fixed proximate analysis for SS are given in Table 3. The MC of SS is 8.64 wt%, the ash, FC, and VM contents are 41.63, 2.77, and 55.60 wt%, respectively (dry basis). These values were slightly higher than those reported in the literature (Pulka et al., 2019). The difference might be referred to the application of different wastewater treatment processes, as well as the conditions of the SS.

Table 3

Proximate analysis and physical properties of SS.

	VM (wt%, dry basis)	Ash (wt%, dry basis)	FC (wt%, dry basis)
	55.60	41.63	2.77
MC (wt%)	8.64
pH (-)	7.03
EC (μS/cm)	6320
AMB (m2/g)	45
Calorific value (MJ/kg)	12.69

The used SS in this study was considered as type I in terms of heavy metals concentration. The results shown in Table 4 show high concentrations of K, Na, and Zn in the dry SS.

Table 4

Heavy metals concentrations in SS (ppm).

B	Pb	T-(K)	Hg	Cu	Ni	Mn	Zn	Cr	Cd	Mo	Se	As	Na
12.79	10.29	2192.2	<1.00	129.9	32.25	135.3	1127.5	26.47	<5.0	50.98	29.46	<10	2073.5

Effects of process parameters on pyrolysis products yields

The effects of temperature, heating rates, and isothermal time on the pyrolysis product yields are evaluated based on the analysis of the full factorial design. Figure 1 shows the main effects plot of the process parameters on the biochar yield. Figure 2 shows the interaction plot of the process parameters on the biochar yield.

Figure 1

Main effect plot for the process parameters on the biochar yield (%).

Figure 2

The interaction plot of the process parameters on the biochar yield (%).

The results indicate that temperature has a strong negative impact on the yield of biochar. Increasing the pyrolysis temperature from 300 to 500 °C significantly reduced the average yield of biochar from 85.4 to 54.0 wt%. The negative impact of temperature on the yield of biochar is free of any interaction effect with the heating rate or the isothermal time. The temperature is inversely affecting the yield of the biochar irrespective of the level of heating rate or the isothermal time. The decrease in biochar yield is related to dehydration and volatilization reactions which take place during the pyrolysis process. In addition, the release of volatile matter at high temperatures will decrease the biochar yield. This trend in results is consistent with those reported by Tarelho et al. (2019).

The results also indicate that the heating rate does not affect the yield of biochar. Increasing the heating rate from 5 to 35 °C/min possessed almost the same yield of biochar (70 wt%). The isothermal time slightly and inversely affected the yield of biochar. Increasing the isothermal time from 45 to 120 min reduced the average yield of biochar from 73.0 to 66.4 wt%. Table 5 shows the magnitude of effects of process parameters on the product yields.

Table 5

The magnitude of effects of process parameters on the product yields (%).

	Effect of Temp	Effect of HR	Effect of Time
Ybiochar (%)	-31.3	0.62	-6.45
YBO (%)	15.6	5.3	1.8
YNCG (%)	15.7	-5.9	4.6

The results of the magnitude of effects of process parameters on the biochar yield indicate that temperature has an effect with a magnitude 52.2 times higher than that of the heating rate on biochar yield. In addition, the temperature has an effect with a magnitude of 4.9 times higher than that of the isothermal time on biochar yield.

Figure 3 shows the main effects plot of the process parameters on the BO yield. Figure 4 shows the interaction plot of the process parameters on the BO yield.

Figure 3

Main effect plot for the process parameters on the BO yield (%).

Figure 4

The interaction plot of the process parameters on the BO yield (%).

The results indicate that temperature has a strong positive impact on the yield of BO. Increasing the pyrolysis temperature from 300 to 500 °C significantly increased the average yield of BO from 8.5 to 24.1 wt%. The positive impact of temperature on the yield of BO is free of any interaction effect with the isothermal time. The temperature is affecting the yield of the BO irrespective of the level of the isothermal time. However, an interaction effect exists between temperature and the heating rate. The results indicate that temperature affects the yield of BO at a heating rate of 35 °C/min more strongly than at a heating rate of 5 °C/min. The impact of increasing temperature on BO yield can be ascribed to char decomposition into liquid fraction.

The results also indicate that the heating rate has a positive effect on the yield of BO. Increasing the heating rate from 5 to 35 °C/min increased the average yield of BO from 13.6 to 18.9 wt%.

Higher heating rates speed up the breaking of volatiles, resulting in more condensable vapor release and reduce char yield (Montoya et al., 2015).

The isothermal time slightly affected the yield of BO. Increasing the isothermal time from 45 to 120 min possessed an average yield of BO at 16.3 wt%. The interaction plot indicates the absence of an interaction effect between heating rate and the isothermal time on the yield of BO.

The results indicate that temperature has a magnitude of effect 2.9 times higher than the heating rate on BO yield. In addition, the temperature has a magnitude of effect 8.7 times higher than that of the isothermal time on BO yield.

Figure 5 shows the main effects plot of the process parameters on the NCGs yield. Figure 6 shows the interaction plot of the process parameters on the NCGs yield.

Figure 5

Main effect plot for the process parameters on the NCG yield (%).

Figure 6

The interaction plot of the process parameters on the NCG yield (%).

The results indicate that temperature has a strong positive impact on the yield of NCGs. Increasing the pyrolysis temperature from 300 to 500 °C significantly increased the average yield of NCGs from 6.2 to 21.9 wt%. The positive impact of temperature on the yield of NCG is accompanied by interaction effects with the heating rate and isothermal time. The temperature is affecting the yield of NCG at a heating rate of 5 °C/min more strongly than at a heating rate of 35 °C/min. Similarly, the temperature affects the yield of NCGs at an isothermal time of 45 min more strongly than at an isothermal time of 120 min. Increasing the temperature will subject the liquid fraction product to undergo further cracking reactions to produce more NCGs.

The results also indicate that the heating rate has a negative effect on the yield of NCGs. Increasing the heating rate from 5 to 35 °C/min decreased the average yield of NCG from 17.0 to 11.1 wt%. The isothermal time slightly affected the yield of NCGs. Increasing the isothermal time from 45 to 120 min increased the average yield of NCG from 11.7 to 16.4 wt%. The interaction plot indicates the absence of an interaction effect between heating rate and the isothermal time on the yield of NCGs.

The results indicate that temperature has a magnitude of effect 2.7 times higher than that of the heating rate on NCG yield. In addition, the temperature has a magnitude of effect 3.4 times higher than the isothermal time on NCG yield.

Prediction of products yields

A first-order model was used to mathematically describe the pyrolysis product yield as a function of the influential factors according to Eqs. (2), (3), and (4). The regression equations in un-coded units are:

The regression models contained all the interaction terms. To validate the regression modes, new experimental data at various operation conditions were carried out (Table 2).

The predicted values of biochar, BO, and NCGs yields as compared to their experimental values are shown in Figures 7, 8, and 9, respectively.

Figure 7

Biochar yield prediction (Eq. (2)) vs. experimental data.

Figure 8

BO yield prediction (Eq. (3)) vs. experimental data.

Figure 9

NCGs yield prediction (Eq. (4)) vs. experimental data.

The developed regression models adequately predicted the biochar, BO, and NCGs yields. The calculated average absolute errors between the predicted and experimental data were 2.1, 4.4 and 2.9 for biochar, BO, and NCGs yields respectively.

Conclusions

This study reports on the feasibility of pyrolysis of sewage sludge in Jordan. Sludge amounts have increasingly been accumaulating as the Country strives to reach universal coverage of sanitation services and wastewater treatment. Acceptable and sustainable sludge management practices are still lacking and pyrolysis stems up as a possible sustainable option. Operational variables of pyrolysis, nalmely temperature, heating rate, and isothermal time, were evaluated based on full factorial analysis.

Increasing the pyrolysis temperature from 300 to 500 °C significantly reduced the average yield of biochar 85.4 to 54.0 wt%, increased the average yields of BO 8.5 to 24.1 wt% and and increased the NCGs yield from 6.2 to 21.9 wt%.

Increasing the heating rate from 5 to 35 °C/min possessed almost the same yield of biochar at ∼ 70 wt%, increased the yield of BO from 13.6 to 18.9 wt%, and decreased the yield of NCGs from 17.0 to 11.1 wt%.

Increasing the isothermal time from 45 to 120 min reduced the average yields of biochar from 73.0 to 66.4 wt%, and possessed almost the same yield of BO at ∼ 16.3 wt % and increased the NCGs yield from 11.7 to 16.4 wt%.

The factorial design methodology was able to reveal all the possible interactions among the process variables. First-order regression models were developed based on the magnitude of effects of the process parameters and their interaction effects to predict the pyrolysis product yields. The models fitted the experimental data very well.

Declarations

Author contribution statement

Tuqa Al-Mrayat: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Husam Al-Hamaiedeh & Tayel El-Hasan: Analyzed and interpreted the data; Wrote the paper.

Salah H. Aljbour: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Ziad Al-Ghzawi & Osama Mohawesh: Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This work was supported by UNICEF (WASH 2108), Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) and the Deanship for Scientific Research Support fund, Mutah University, Jordan.

Data availability statement

The authors do not have permission to share data.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.