Steam gasification kinetics of biochar at elevated pressures

Abstract

A waste product biomass sample was received and charred to produce the biochar sample. The char reactivity experiments were conducted in a high-pressure fixed bed reactor in the temperature range of 700–730 °C. The steam pressure was varied from 1 to 10 bar steam, and the CO and CO₂ products were measured and used to determine the specific reaction rate of biochar. The results showed that the reaction rate increased with conversion, temperature and steam partial pressure. The increase in steam partial pressure had a significant effect on the reaction rate up to 10 bar steam, where it was observed that the formation of CO₂ contributed more to the specific reaction rate than that of CO and that the selectivity of CO₂ increased over the steam pressure range. The use of these kinetic models also determined the activation energy, and the results were found to be consistent with the literature.

Highlights

• •Steam gasification kinetics of biochar at elevated pressures were investigated.
• •The biochar reactivity was found to be dependent on conversion, temperature and steam partial pressure.
• •The formation of CO₂ from biochar was found to be significantly higher than that of CO.
• •The overall reactivity of biochar was found to be at least 4 times that of a typical Highveld coal char.
• •Activation energies determined from the Langmuir-Hinshelwood and power law were 214 and 222 kJ mol^−1, respectively.

1. Introduction

Since the world's energy consumption has steadily risen over the past few decades, using energy resources effectively has become more critical than ever. Over the past 26 years, global total energy consumption has increased by 34 % [1]. According to Platchkov and Pollitt [2], a number of factors contribute to rising energy consumption, including the cost of power, the rate of economic production development, population expansion, climatic variations, and technological advancements.

In South Africa, about 77 % of electricity is generated by coal. While this is the primary use of coal, a significant amount is also utilized in coal-to-liquid (CTL) processes [3]. This CTL activity is used to produce synthetic fuels and other valuable chemicals. In general, coal is a primary energy source and is seen as a significant contributor to CO₂ emissions and particulate matter, substantially impacting the environment and human health [4]. The available coal resources and quality of coal in South Africa is also depleting [5], which has resulted in exploring alternative sources as fuel for energy supplies. These alternative sources include renewable energy sources, which are hydro-power, wind energy, geothermal energy, solar energy, and biomass energy [6]. Estimations show that coal and oil resources are used worldwide for energy consumption and supply [6]. However, the use of biofuels and waste for energy sources worldwide is 10–11 %, making it the most used renewable energy source [7]. The use of biomass as a renewable source of electricity has increased from 5.6 % in the year 2000 to 9.2 % in 2017, making it the 3rd highest renewable electricity source after wind and hydropower [7].

Biomass is an appealing alternative energy source since it can produce electricity, heat, and transportation fuels [8]. This energy resource has several advantages, including the potential to generate a variety of gaseous, liquid, and solid fuels which can be used for various applications [[8], [9], [10]]. The different types of biomass can be divided into four groups: waste, processing residues, locally collected feedstocks, and internationally traded feedstocks [9]. It can thus be comprehended why biomass is used for over 70 % of all renewable energy production [8].

One of the most attractive operations for utilising biomass is gasification, which is considered more environmentally friendly and produces higher efficiencies than combustion and pyrolysis [8]. Gasification is the thermo-chemical conversion of carbonaceous material into a gaseous product, like syngas, in an oxygen-lean environment [9]. The gasification process is carried out at elevated temperatures, atmospheric or elevated pressures and in the presence of a gasifying agent. These gasifying agents include air, oxygen, steam, or mixtures of these gases [11]. Some complex chemical reactions that can influence the operation of a biomass gasifier are pyrolysis, partial oxidation of pyrolysis products, gasification of produced char, tar formation and lower hydrocarbon conversion and water-gas shifting [9].

Despite all the advantages of biomass as a gasification feedstock, some drawbacks have been considered, such as high initial investment costs, food security and water management [10,12]. Biomass gasification produces char and tar/oils as by-products, whereas tar is generally undesirable because it results in blockages of pipes and filters in downstream equipment [12]. Another option is to focus more on applications which can use mixed feedstocks, such as the co-feeding of biomass and coal. The co-gasification of these new resources requires a comparison of the reactivity of biochar and coal.

Commercially, the gasification process in South Africa is performed in fixed-bed dry bottom (FBDB) gasifiers [13]. The gasification process is also considered very broad because the process includes drying, pyrolysis, gasification, and combustion, which all occur in a gasifier. The gasification term refers to the gasification zone inside the gasifier, which converts the residues produced by the pyrolysis step into syngas [14]. The main reason for investigating the gasification kinetics is that the gasification zone is the rate-limiting step in the process [15,16]. In the gasification zone, a gasifier agent or gas is introduced to the system [17]. For an FBDB gasifier, H₂O and O₂ are used as reagent gases in the industry; however, CO₂, steam, O₂-steam, CO₂-steam and O₂-steam-CO₂ are used for laboratory scale studies regarding reaction kinetics (Amed et al., 2016). Various studies have been conducted on the char gasification reaction with these gases; however, some noted that using steam resulted in faster reactions than when using CO₂ [[18], [19], [20]]. Ahmed and Gupta [15] found that the average reaction rate for woodchip char steam gasification was almost twice that of CO₂. The main constituents of biomass (cellulose and hemicellulose) results in biomass being more reactive than coals, leading to lower required gasification temperatures [21]. The biomass gasification industry prefers slight fixed bed updraft and downdraft technologies to generate heat and power. For the production of fuels and chemicals, large entrained flow gasifiers are favoured [10].

Various studies on the influence of steam partial pressure on the gasification kinetics of biochar at low pressures have been completed [15,[22], [23], [24], [25]]. This was achieved by changing the steam concentration at a constant system pressure or by changing the total system pressure at a constant steam concentration. The effect of steam partial pressure on the reaction rate has been found to be less prominent than that of temperature [23]. In a study by Ahmed and Gupta [15], an insignificant effect of steam partial pressure on woodchip-derived char gasification was found. The steam partial pressure was varied between 0.6 and 1.5 bar at 900 °C in a fixed bed reactor [15]. However, other authors observed a significant effect of steam partial pressure on the biochar reaction rate at low pressures [22, 23,25). Klose and Wölki [22] investigated the influence of steam partial pressure on the specific reaction rate of biochar. In their study, two different biochar samples were investigated. The reaction rate increased with an increase in steam partial pressure for both samples; however, the beech wood char has a more pronounced increase than the oil palm shell char. As the steam partial pressure increases, the influence on the oil palm shell char reaction rate becomes less significant. Tagutchou al [24]. found that if the partial pressure of steam is increased from 0.1 to 0.4 atm, the gasification rate is 3 times faster [24]. In comparison, Guizani al [25]. observed 2.5 times increase in the char reactivity of beech wood chips by increasing the steam concentration from 10 % to 30 % [25].

The investigation of the effect of steam gasification on biochar reaction rate at elevated pressure has not been widely studied. However, a study completed by Nandi and Onischak [26] on maple and jack pine evaluated the effect of steam partial pressure by changing the total system pressure at a constant steam concentration. The results from varying the partial pressure of steam between 0.85 and 10.85 bar. This study concluded that the effect of steam partial pressure on the gasification rate was significant for steam partial pressures higher than 3.95 bar [26]. The limited studies on biochar reactivity at elevated steam partial pressure led to an investigation into elevated steam gasification of other carbon materials and predictions of high steam partial pressure gasification of biochar. A study investigated the steam gasification of polyvinylchloride (PVC) coke at 1000 °C and a total system pressure of 10 bar while changing the concentration of steam. An increase in steam partial pressure from 0.55 bar to 4.3 bar increased the gasification rate [27]. Another carbon source to consider is coal char. Studies using coal char at elevated steam partial pressures have found that an increase in steam partial pressure results in an increase in reaction rate [18,28,29]. Sha al [29]. found that the reaction rate increased more slowly at steam pressures above 10 bar, which suggests a form of saturation. Another study reported this saturation effect at elevated pressures and found that this behaviour is affected by temperature [18]. The estimation of site occupancy can explain the saturation effect on the specific reaction rate. Site occupancy or surface coverage can express the desorption of intermediate surface complexes, and it has been reported that the desorption can affect the apparent reaction rate [28]. Roberts and Harris [28] reported that the site occupancy of Australian coal-derived char can be estimated by using an empirical method constructed from the reaction mechanisms, which are described by the Langmuir-Hinshelwood (LH) expressions and temperature programmed desorption (TPD). The study on coal derived char reported that the site occupancy increased with an increase in steam partial pressure from 1 to 30 bar for both LH and TPD methods used [28]. This indicates that the reactive surface of the char becomes more saturated with surface complexes at elevated steam partial pressures [18,28]. This suggests that the further increase of steam partial pressure will result in an unaffected specific reaction rate, and the saturation effect is observed. This can be predicted for biomass-derived char as well.

Coal remains a valuable resource in the generation of energy and transportation fuels; however, the world's coal resources are depleting, and its impact on the environment and human health raises concerns. Investigating alternative resources, such as biomass, has become more relevant. One of the most promising processes for biomass utilization is gasification. Commercially, the gasification process in South Africa is performed at elevated pressures and in large complex fixed-bed dry-bottom gasifiers, considered a fuel-flexible technology [10]. To successfully design such a gasifier, information on gasification kinetics is necessary. Most of the previous work on high-pressure gasification in South Africa was done on coal char; in South Africa, high-pressure biomass steam gasification kinetic research is lacking. Therefore, in this study, the steam gasification kinetics at elevated pressures is given attention to further understand the possible application of biomass as a gasification resource in the local industry. This study aims to determine the steam gasification reaction kinetics of biochar over a wide pressure range.

2. Materials and methods

2.1. Sample preparation and characterisation

Approximately 18 kg of the raw biomass sample was washed twice with tap water and then once with deionised water before air-drying the sample for three days. The raw biomass was then cone and quartered by continually dividing the raw biomass sample into 4 sub-samples to produce a representative sample of 670 g for characterisation. The remaining biomass produced the biochar in a three-phase Lenton horizontal furnace at 950 °C using a heating rate of 10 °C/min and a N₂ flow rate of 2 Nl/min.

The biomass sample shows a high cellulose content, while the hemicellulose and lignin contribute only a third of the biomass composition. The lignocellulosic composition of the raw biomass sample is similar to that found in other studies for hardwood samples [[30], [31], [32], [33]]. The ultimate and proximate analysis results for the raw biomass (RBM) and biochar (CBM) are presented in Table 1. The charring process of the biomass resulted in an increase in fixed carbon content and a decrease in the volatile matter. The results further indicate a decrease in oxygen and hydrogen content while carbon content is increased. This then leads to a significant decrease in H/C and O/C ratios for the biochar.

Table 1.

Characterisation results of parent coal and char sample.

Parameter	Biomass	Biochar
Fibre analysis (wt%)
Hemicellulose	11.1	ND
Cellulose	58.9	ND
Lignin	19.1	ND
Proximate analysis (wt%-adb)
Ash	1.0	2.4
Volatile matter	79.6	4.2
Fixed carbon	10.1	91.9
Moisture content	9.3	1.5
Ultimate analysis (wt%-daf)
Carbon	45.5	94.6
Hydrogen	5.4	0.2
Nitrogen	0.05	0.6
Total Sulphur	0.04	0.03
Oxygen	39.8 (by difference)	4.4 (by difference)
Gross calorific value (MJ/kg)
CV	18.1	30.4
Micropore surface area (m2/g-daf)
D-A method	ND	477

ND – Not determined.

2.2. Steam gasification experiments

A High-Pressure Fixed Bed Reactor (HPFBR), previously used by Gouws al [34]. for only CO₂ gasification measurements, was used in this investigation for measuring biochar gasification rates directly from associated chemical reactions in the atmosphere of steam. The experimental rig can be operated at temperatures varying from ambient to 1000 °C and up to 50 bar pressure. Fig. 1 shows a schematic presentation of the experimental setup, and a detailed description is presented elsewhere [34].

Fig. 1.

Experimental setup.

At the start of an experimental run, a biochar sample between 0.3 and 0.5 g is weighed and placed on top of the quartz filter disc inside the tubular reactor. After the reactor is reconnected to the rig, nitrogen is introduced to the system. The mass flow controller controls the nitrogen flow rate, which is set to the desired set point using the Flow View software on the computer. The heating wires before the mixing chamber are set to 105 (±2) ⁰C, and the steam generator temperature has a set point of 360 (±12) ⁰C. Before switching on the HPLC pump, the purge valve located downstream from the steam generator is opened. The water flow rate is then set using the HPLC pump user interface, and steam starts flowing from the purge line to ensure actual flow.

The furnace is set to the desired reaction temperature before each run. Once this temperature has been reached, the furnace is closed over the pipe reactor. The required reaction temperature at the sample bed is monitored by the thermocouple located inside the reactor. Nitrogen is used to flush the biochar sample to ensure an inert environment inside the reactor. After a flushing period of 30 min, the reactor is pressurised to the required pressure of that specific experiment. The desired pressure is set using the Flow View software on the computer, which communicates with the EPC to control the total system pressure. The ice bath, where the condensers are located, is filled with ice. Once the isobaric and isothermal conditions are reached, the steam is introduced into the system by closing the purge valve downstream from the steam generator. After that, the data logging is started on the computer. The ice level is monitored throughout each experiment to ensure that the water contained inside the product gas is condensed before flowing through downstream equipment. The steam gasification conditions used during the test work are given in Table 2. These conditions (relative low temperature) were specifically selected to operate the reaction in the reaction-controlled regime (absence of mass and heat transfer limitations). Prior to the experimental program was performed, separate experiments were performed, confirming that the reaction was performed in the reaction-controlled regime, e.g., the particle size was varied and at these relative low temperatures, and it was shown that the reaction rate was independent of particle size.

Table 2.

Experimental operating conditions.

Parameter	Gasification conditions
Sample mass	∼0.3 g
Particle size	−150 + 75 mm
Total flow rate (STP)	2 NL/min
Steam to N2 ratio	01:01
Reaction temperature	700-730 °C
Steam partial pressure	1–10 bar
Total system pressure	2–20 bar

After the gasification reaction, the logging software and steam pump are switched off. The depressurisation of the system follows this by bypassing the EPC and opening the purge valve downstream from the MFM. The water is removed from the condensers once the total system pressure is below 5 bar. Then, the system is further depressurised to a pressure below 1 bar. The reactor can now be removed from the furnace and cooled to room temperature before closing the nitrogen flow. Lastly, the reactor pipe is removed from the rig, and the remaining sample is collected from the reactor. Table 2 gives the experimental conditions used during the steam gasification experiment. The total pressure of the system was 2–20 bar.

After an experimental run, the raw data obtained is imported to Excel, where the fraction of CO and CO₂ are plotted over time. The fractions of CO and CO₂ were logged every 3 s, resulting in a visibly continuous line on the plot, although individual measurements were taken. To illustrate the data processing method, a high-pressure and complete conversion run was performed. This experiment was carried out at 720 °C and a total pressure of 20 bar with 50 % steam (10 bar steam partial pressure). After the concentrations of both products were obtained, the amount of carbon reacting with steam per second was determined using reaction stoichiometry and the product gas flow rate. Equation (1) is used to calculate the carbon conversion at each time interval:

$$r_c=(y_{CO}+y_{C{O}_2})\left(n_t\right)\left(M{M}_c\right)$$ (1)

Where $y_{CO}$ and $y_{C{O}_2}$ are the fractions of CO and CO₂, respectively,/is the total molar flow rate, and MW_c is the molecular weight of carbon. After that, the amount of carbon remaining in the sample is determined by calculating the amount of carbon reacted at each time interval. Equation (2) was used to calculate the carbon reacted after the time (t).

$$\Delta m_c\left(t\right)={\int }_0^t{r}_{c}dt\approx \frac{\Delta t}{2}[r_c\left(t_0\right)+2r_c\left(t_1\right)+\dots .+2r_c\left(t_{n-1}\right)+r_c\left(t_n\right)$$ (2)

Where $\Delta t$ equals 3 s, which is the time interval between each reading. Equation (3) was used to determine the total amount of carbon in the unconverted sample (M_c,0).

$$m_{c,0}=m_{s0}\left(1-x_{ash}\right)(x_c\left(daf\right)$$ (3)

M_s,0 is the initial sample mass, X_ash is the fraction of ash yield, and Xc(daf) is the carbon fraction on a dry ash-free basis obtained from the ultimate and proximate analysis. Lastly, the conversion is calculated on a dry ash-free basis using Equation 4.

$$X\left(t\right)=\frac{\Delta m_c\left(t\right)}{m_{c,0}}$$

After calculating the carbon conversion rate $\left(r_c\right)$, it is normalised with the amount of unreacted carbon in the biochar sample at the reaction time. This normalised reaction rate is termed as the specific reaction rate rate $\left(r_s\right)$,. Other authors used the same normalisation technique to describe the specific reaction rate of steam gasification [23,29]. The specific reaction rate is given by Equation (5).

$$r_s=\left(\frac{1}{1-X}\right)\left(\frac{dX}{dt}\right)$$ (5)

It is worth noting that the gas phase analysis was only performed with a CO and CO₂ analyser since the conversion rate of the material was of primary interest. Initial experiments with complete conversion were performed. It was shown that the carbon in the fuel was primarily (>95 %) converted to CO and CO₂, which was sufficient for determining the reaction rate.

3. Results and discussion

3.1. Steam gasification reactivity

3.1.1. Effect of conversion on specific reaction rate

Fig. 2 illustrates the specific reaction rate for conversion of 4 up to 50 % for different steam partial pressures (1–20 bar) at 720 °C. From Fig. 2, it is observed that the specific reaction rate increases with an increase in steam partial pressure as well as conversion. The increase in conversion has a more significant effect on the reactivity at elevated pressures (5–20 bar) than at lower pressures (1 and 2.5 bar). However, a form of saturation on the reaction rate is observed at elevated steam partial pressures, where the effect of partial pressure above 10 bar disappears. Previous studies of coal char steam gasification at steam partial pressures above 10 bar indicated no notable increase in specific reaction rate Muhlen al. [18], Roberts and Harris [29]; Sha al. [30]; and which agree with the results obtained here [18,29,30]. It has been reported that this is a result of an increase in the site occupation to almost complete coverage Roberts and Harris [29] reported that for coal steam gasification, an increase in steam partial pressure increased the quantity of the surface complexes, which again increased the reaction rate [29]. The insignificant change above 10 bar hereby implies that the active char sites are mostly covered and that a further increase from 10 bar steam partial pressure will have a less significant effect on the specific reaction rate.

Fig. 2.

Effect of conversion on the specific reaction rate at 720 °C.

Based on these findings, further discussions confined the experiments to partial steam pressures of 1–10 bar. The steam partial pressure for different temperatures (700, 710 and 730 °C) are varied between the 1–10 bar steam range and are shown in Fig. 2.

3.1.2. Effect of temperature on specific reaction rate

The effect of temperatures ranging from 700 to 730 °C on the specific reaction rate for different steam partial pressures are illustrated in Fig. 3. For the kinetics analysis, the specific reaction rate at 20 % conversion is further used to obtain a kinetic relation at different operating conditions. Over a steam partial pressure range of 1–10 bar, it was found that the specific reaction rate increased exponentially with an increase in temperature. This intense temperature dependence agrees with results found in literature where the temperature increase followed an Arrhenius type of increase [15,23,24,26]. These results were reported for different biomass feedstocks, which include sawdust-derived chars, i.e. Lopez al. [23], pine woodchips, Tagutchou al. [24], maple hardwood, and jack pine-derived chars (Nandi and Onischak, 1985). It was also shown in Fig. 3 that an increase in temperature resulted in a corresponding increase in steam partial pressure; for instance, at a steam partial pressure of 1 bar and temperature increase of 30 °C, the specific reaction rate increased by a factor of 2.5, while at a 10 bar steam partial pressure, it increased by a factor of 2.2. Similar results were obtained in a study on steam gasification of maritime pine woodchips, which reported that the reactivity increased by a factor of 9 with a temperature increase from 800 to 1000 °C [24].

Fig. 3.

Effect of temperature on specific reaction rate at 20 % conversion.

3.1.3. Effect of steam partial pressure on specific reaction rate

The effect of steam partial pressure on the specific reaction rate at 20 % conversion for 700 to 730 °C are displayed in Fig. 4a. It was observed in Fig. 4a that the specific reaction rate increases with an increase in steam partial pressure over the entire temperature range. Both Tagutchou al [24]. and Guizani al [25]. reported the same trend for maritime pine and beech wood chip steam gasification, which is in accordance with the results reported. Other studies have reported that faster char conversion is achieved by increasing the steam partial pressure up to 1.5 bar [15,35]. Fig. 4a shows that the increase in reaction rate between lower steam partial pressures is more significant than between higher steam partial pressures. The reaction rate increases by an average of 30 % between 5 and 7.5 bar steam, while the increase between 7.5 and 10 bar steam is, on average, c.a. 8 %. The start of the saturation effect on biochar reactivity at elevated steam partial pressures is seen here. Another investigator has reported a similar trend for the steam gasification of coal char [29].

Fig. 4.

(a) Effect of steam partial pressure on specific reaction rate, (b) Reaction order with steam partial pressure at different temperatures.

The natural logarithmic of the steam partial pressure and specific reaction rate at 20 % conversion were plotted to directly evaluate the observed reaction order with steam partial pressure. The results for the temperature range of 700–730 °C are shown in Fig. 4b. The experimental data shown in Fig. 4b follows a power law trend, while the difference in reaction orders throughout the partial pressure range remains similar. The observed reaction order decreased from 0.61 to 0.56 with a temperature increase from 700 to 730 °C. These results indicate that the reaction order is not significantly affected by temperature under Regime I (chemical reaction controlled) conditions. Nilsson al [36]. investigated the gasification kinetics of olive tree pruning-derived chars at 0.2–0.4 bar steam partial pressure and reported that the reaction order was affected by varying the temperature between 760 and 840 °C. However, this temperature range is higher than the 700–730 °C investigated, which might result in reaction conditions in Regime II.

3.1.3.1. Product distribution

The product formation of CO and CO₂ is investigated by comparing the rate of formation of both products to the specific reaction rate in Section 3.2.1 and the selectivity of the products in Section 3.2.2.

3.1.3.2. CO and CO₂ formation

The CO and CO₂ fractions are used to calculate the conversion rate of CO and CO₂. These conversion rates are compared to the rate of total carbon conversion (r_c). The results at 20 % total carbon conversion at different temperatures used in this study are shown in Fig. 5a-d. In Fig. 5a-d, it is seen that the CO₂ conversion rate contributes more to the total carbon conversion rate than that of CO. Smolinski al [37]. studied the steam gasification of energy crop biomass-derived chars at 700 °C and reported that biochar produced 3 times more CO₂ than CO, which confirms the higher CO₂ formation also observed from the results [37]. The difference in the CO₂ and CO conversion rates becomes larger with increased steam partial pressure. The increase in CO₂ can be attributed to increased adsorbed intermediates due to the increase in steam partial pressure. However, this occurs until a steam partial pressure of 7.5 bar due to the more negligible difference in carbon conversion rate between 7.5 and 10 bar steam. The water gas shift reaction can influence this ratio. Although it was shown that the homogenous conversion rate of this reaction is small, no further mechanistic conclusions were made from these observations.

Fig. 5.

Comparison of CO, CO₂ and total carbon conversion for a temperature of a) 700 °C, b) 710 °C, c) 720 °C and d) 730 °C.

3.1.3.3. Selectivity

In order to further investigate the favoured product formed from steam gasification, the selectivity of the CO and CO₂ are calculated. The results for selectivity at 20 % conversion at various steam partial pressures and temperatures are shown in Fig. 6 a) to d). The results from Fig. 6 suggest that the formation of CO₂ is favoured for the entire temperature range investigated. The selectivity of CO₂ at 10 bar steam increased from 0.83 to 0.90 with a temperature increase of 30 °C, while the CO selectivity decreased from 0.17 to 0.10. The influence of steam partial pressure on the selectivity is more considerable than that of temperature; however, the effect of temperature on the CO and CO₂ ratio becomes less significant with an increase in steam partial pressure. The selectivity of CO₂ increased with a factor of 1.3 over the investigated steam partial pressure range, while the CO selectivity decreased with a factor of 0.3. It is observed throughout the steam partial pressure range that the formation of CO₂ is larger than that of CO. This observation was found to be consistent with results in the literature regarding steam gasification of biochar [38,39]. In contrast, a study on steam gasification of elephant grass-derived biochar reported that the CO/CO₂ ratio increased at temperatures of 800–950 °C [40]. However, another study reported that the temperature increases from 750 to 900 °C increased CO₂ concentration and decreased CO concentration [39].

Fig. 6.

Observed selectivity of CO and CO for various steam partial pressures and temperatures of a) 700 °C, b) 710 °C, c) 720 °C and d) 730 °C.

Sattar al [38]. investigated the syngas formation from steam gasification of various biochar samples at 850 °C. They observed that the formation of CO₂ was larger than CO for both sewage sludge and wood pellet-derived chars (Satter et al., 2014). According to Sattar al. [38], the increase in CO₂ was attributed to two significant factors: the carbon content in the char decreases over time while the steam flow rate remains constant. This results in a higher steam to carbon (S/C) ratio, and it has been observed by other authors that higher S/C ratios can suppress the formation of CO [41]. The second factor is the carbon reduction in char, which ensures that the Boudouard reaction occurs less frequently, resulting in less CO formation [38]. However, a study on polyvinylchloride coke steam gasification at 1000 °C and steam partial pressure up to 4.3 bar reported that the CO₂ formation was attributed to the homogeneous gas phase water gas shift reaction [27]. This is also found to be the case in the results shown here.

3.1.3.4. Comparison with coal

The product formation from biochar and coal char are compared by investigating the selectivity of CO and CO₂ at a conversion of 20 %. The biochar and coal samples were gasified at 2.5 bar and 10 bar steam partial pressure, while the biochar was at a temperature of 730 °C and the coal at 740 °C. The results are displayed in Table 3. Table 3 shows that the formation of CO₂ is more significant for biochar at 2.5 bar and 10 bar steam partial pressure than that of coal. Similar results were reported by Loha al. [41], where rice husks, sugarcane bagasse, rice straw and groundnut shell steam gasification were compared to coal [41]. For coal char steam gasification at a partial pressure of 2.5 bar, the formation of CO₂ was below the detection limit of the gas analyser, which resulted in a selectivity of <0.01. The selectivity of CO₂ for both biochar and coal char increased with an increase in site coverage due to increased steam partial pressure or the heterogeneous water gas shift reaction. Although no direct mechanistic conclusions can be drawn from this, the product distribution from biochar differs significantly from that of coal char.

Table 3.

Observed selectivity of biochar and coal at 2.5 and 10 bar steam partial pressure.

Steam partial pressure (bar)	Selectivity to CO	Selectivity to CO2
Coal char
2.5	>0.99	<0.01
10	0.36	0.64
Biochar
2.5	0.22	0.78
10	0.11	0.89

3.1.3.5. Kinetic modeling and site occupancy

Another commonly used reaction mechanism for carbon gasification with steam is suggested by Ergun and Reif, which does not consider the formation of CO₂ [22,35]:

$$C_f+H_{2}O\rightleftharpoons \mathrm{C}\left(\mathrm{O}\right)+H_2\left(g\right)$$ (R-1)

$$C\left(O\right)\to CO\left(g\right)$$ (R-2)

Where C_f is the free active site, and C(O) is an adsorbed carbon-oxygen surface complex. This mechanism produces the simplest form of a Langmuir-Hinshelwood model to represent experimental data obtained.

The reaction mechanisms for carbon conversion with steam lead to the Langmuir-Hinshelwood rate expressions. This model describes the dependency of the reaction rate on the partial pressure and reaction temperature well [22]. The simplest form of the Langmuir-Hinshelwood is given in equation (6) [35].

$$r_s\frac{k_1{P}_{H_{2}O}}{1+\frac{K_{1.}}{K_3}{P}_{H_{2}O}}$$ (6)

Where K₁ is the rate constant for reaction 1, and K₃ is the rate constant for reaction 2, both of which follow the Arrhenius law. Equation (6) suggests that the H₂ formation has no significant effect on the reaction rate due to the low concentrations of H₂ formed [36]. The least-square method was used to determine the rate constants in the above-mentioned equation (7). The rate constants determined for different temperatures are displayed in Table 4.

$$\mathit{Min}={\sum }_{i=1}^n(r_{\mathit{exp},i}-r_{calc,i})²$$ (7)

Where $r_{\mathit{exp},i}$ is the experimental rate and $r_{calc,i}$ is the calculated rate.

Table 4.

Langmuir-Hinshelwood rate constants for different temperatures.

	700 °C	710 °C	720 °C	730 °C
$\left[C_t\right]k_1{\left(bars\right)}^{-1}$	1.7 × 10−5	2.1 × 10−5	3.2 × 10−5	3.6 × 10−5
$\frac{k_1}{k_3}\left({bar}^{-1}\right)$	0.14	0.15	0.16	0.13

K₁, K₂, and K₃ in equation (6) are dependent on temperature and can be expressed using the Arrhenius equation. Equation (8) gives the Arrhenius equation used in calculating the activation energy during the steam gasification.

$${\mathrm{K}}_{\mathrm{i}}={\mathrm{A}}_{\mathrm{i}}\exp \left(\frac{-{\mathrm{E}}_{\mathrm{i}}}{\text{RT}}\right)$$ (8)

Eq. (8) is linearlised to form Eq. (9):

$$In\left(K_i\right)=-\frac{E_a}{RT}+In\left(A_o\right)$$ (9)

The kinetic coefficient ki's pre-exponential factor and activation energy are denoted by the letters A_i and E_i. The results of steam gasification of biochar were used to calculate the values of K₁ and K₂. K₁ and K₂ values were computed using linear regression at a certain conversion temperature, based on the regression of r₂₀ vs T_H2O at each temperature. E_a is determined using the slope of In(K_i) against T⁻¹, using reactivity obtained at each temperature used in this study. The intercept of the plotted graph gives the exponential constant from the reactivity using Langmuir-Hinshelwood. The selection of x = 0.20 as the conversion reference state.

The Arrhenius plot of the LH rate constants are shown in Fig. 7a. The values from Table 4 and Fig. 7a show that the $\left[C_t\right]k_1$ rate constant increases with a temperature increase, while the ratio of $\frac{k_1}{k_3}$ does not significantly change. The rate constants shown in Table 4 were used, with a constant value of 0.15 for $\frac{k_1}{k_3}$, to determine the activation energy ($E_1)$, which is calculated as 214 ${kJmol}^{-1}$. The activation energy is in the range of values obtained from literature [22,23,26]. Activation energies of beech wood char and oil palm shell char were determined as 196 $kJ{mol}^{-1}$ and 299 $kJ{mol}^{-1}$ by Klose and Wölki [22], which indicate that the type of biomass used for steam gasification can also influence the activation energy. Nandi and Onischak [26] found activation energies of 170–177 $kJ{mol}^{-1}$ for high pressure steam gasification of maple and jack pine chars. In a study of switchgrass-derived char, the steam gasification kinetics were described using the LH model [42]. They used the Arrhenius equation to determine the activation energy and enthalpy and obtained values of 112.6 $kJ{mol}^{-1}$ and −37.3 $kJ{mol}^{-1}$ respectively [42]. However, these experiments were conducted at elevated temperatures of 1000–1150 °C, which suggests mass transfer limitations resulting in the relatively low value for the observed activation energy.

Fig. 7.

(a) Arrhenius plot of Langmuir-Hinshelwood rate constants; (b) Correlation between Langmuir-Hinshelwood model and experimental data with uncertainties given for the highest temperature.

The parameters in Table 4 were used to compare the LH model with the experimental data obtained at a conversion of 20 %. The results are displayed in Fig. 7b. Fig. 7b shows that the model shows an excellent correlation to the experimental values obtained. However, the LH model deviates more at 730 °C than the other temperatures. This is shown with the error estimation, where the maximum error of 4.9 % and the minimum error of 3.9 % is observed at 730 and 700 °C, respectively. The uncertainties shown in Fig. 7b are largest at low temperatures and pressures; this can result from the measurements being taken at the bottom range of the gas analyser. The experiments' uncertainty depends on steam partial pressure since the uncertainties decrease with increased steam partial pressure. Other authors investigating biomass gasification observed that the Langmuir-Hinshelwood model described the experimental data quite well for steam partial pressures of 0.1–1 bar [22,42]. Roberts and Harris [28] investigated a coal char at elevated steam partial pressures and observed that the Langmuir-Hinshelwood model is suitable for up to 30 bar steam pressures. These findings agree with the results shown in Fig. 7. The quality of fit (QOF) between the experimental and modelled values for the entire temperature range is an average of 91 %, which indicates that the model can accurately describe the experimental data over the selected pressure and temperature range.

3.1.3.6. Power law

The power law is a popular model for modelling the intrinsic reaction rates of char gasification. This nth-order kinetic model allows for the effect of the state of conversion. The power law is considered to be more accurate in representing the reaction rates at medium to high conversions [36].

The rate equation is described by:

$$r_s={k_s\left(P_{H_{2}O}\right)}^n$$ (10)

Where the rate constant is assumed to follow the Arrhenius equation shown below:

$${\mathrm{K}}_{\mathrm{s}}=\mathrm{A}\exp \left(\frac{-\mathrm{E}}{\text{RT}}\right)$$ (11)

The kinetic parameters for the power law were also determined with the least squared method, as was done for the Langmuir-Hinshelwood model above. The calculated parameters are shown in Table 5. The Arrhenius plot of the calculated PL rate constant is shown in Fig. 8a for the temperature range of 700–730 °C. The reaction rate constants from Table 5 were used to calculate the activation energy and frequency factor. The calculated activation energy and frequency factor were 222 (KJ mol-¹) and 1.8 x 10⁷ (gg⁻¹s⁻¹barⁿ), respectively. The activation energy, frequency factor values and the Arrhenius plot in Fig. 8a show that the rate is constant/depends on the temperature. At the same time, the reaction order is constant, with an average of 0.59. Barrio al [43]. found similar values for steam gasification of wood char at a steam partial pressure of 0.05–1 bar. They found that the reaction order for birch and beech chars were 0.57 and 0.51, respectively, which correlate well with values shown in Table 5 (Barrio et al., 2001). Other studies on biomass steam gasification reported reaction order values of 0.6–1.5 [35,36,44]. Using the nth-order reaction model, a study on the steam gasification of wood char reported activation energies of 211 and 237(KJ mol-¹) at 50 % conversion [43].

Table 5.

Parameters calculated for power law.

	700 °C	710 °C	720 °C	730 °C
$k_s\left(s^{-1}\right)$	1.8 × 10−5	2.2 × 10−5	3.4 × 10−5	3.9 × 10−5
$n$	0.60	0.58	0.58	0.60

Fig. 8.

(a) Arrhenius plot of PL parameters; (b) Correlation between the power law and experimental data with uncertainties given for the highest temperature.

The activation energy and frequency factor parameters were used to compare the power law model with the experimental data obtained at a conversion of 20 %. The results are shown in Fig. 8b. Fig. 8b shows that the predicted values from the power law provide an excellent correlation to the experimental values. However, the modelled values become less accurate at lower temperatures. The estimated error between the modelled and experimental values are 4.2 % and 3.8 % at 700 and 730 °C, respectively. The quality of fit (QOF) between the experimental and modelled values is 93 % for the entire temperature range. This suggests that the power law can accurately predict the experimental values throughout the specified temperature and pressure range. An insignificant difference in the QOF between the LH model and power law is observed. However, the LH model provides a more fundamental insight into the reaction of biochar and steam, resulting in the LH model being more appropriate for kinetic modelling.

4. Conclusions

Steam gasification kinetics of biochar at elevated pressures were investigated in this study. The biochar reactivity was found to be dependent on conversion, temperature, and steam partial pressure. An increase in all these factors increased the biochar specific reaction rate. However, the biochar reactivity was only significantly affected by steam partial pressure up to 10 bar steam. The formation of CO₂ from biochar was found to be significantly higher than that of CO. The overall reactivity of biochar was found to be at least 4 times that of a typical Highveld coal char, which could be partly attributed to the high initial internal surface area of the biochar.

CRediT authorship contribution statement

J. Schroeder: Writing – original draft, Methodology, Investigation, Formal analysis. H.W.J.P. Neomagus: Supervision. J.R. Bunt: Supervision, Funding acquisition. R.C. Everson: Supervision. R.C. Uwaoma: Writing – review & editing, Writing – original draft.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interestsJane Schroeder reports financial support was provided by Sasol Limited. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.