Oxidative Pyrolysis for Variable Heating Output with Wood Pellets

Abstract

A carbon-negative heating system can be realized by pyrolyzing wood pellets, burning the product gas, and storing the produced biochar. Oxidative pyrolysis simplifies the reactor design by replacing an external heat supply with internal oxidation driven by a sub-stoichiometric “primary” air supply. Previous studies have only examined the influence of primary air supply on biochar yield and heating power in a continuous pyrolysis reactor within a limited fuel–air spectrum. In this work, an oxidative pyrolysis reactor, with a nominal heating power of 15 kW, was investigated with the aim to vary the useful heat output and biochar yield over a wide range and still produce biochar of the highest quality in accordance with the EBC (European Biochar Certificate) guidelines. This study demonstrated that within an air flux range of 0.03–0.14 kg/(m²s), there is a linear relationship between air flux and both wood flux and useful heat, resulting in a power output range of 4–30 kW. The useful heat output could be varied by a factor of three in less than 15 min, verifying concept feasibility as a central heating system to meet the variable heating demands of both single and multi-household applications. The biochar yield was observed to range from 12% to 24% of the incoming wood mass flow, meeting the EBC Feed Plus quality standards at all conditions. Depending on the operating point, up to 40% of the biomass’s heating value is stored in the biochar.

1. Introduction

Given the increasing CO₂ concentration in the atmosphere, “negative emissions” technologies removing CO₂ from the atmosphere will become a focus of climate policy [1]. One of the available negative emission technologies is biochar from sustainable wood resources, which stores the CO₂ that plants extract from the atmosphere in a permanent form [2]. Biochar is a safe and feasible way to store CO₂ before it even enters the atmosphere [3]. Biochar is also a stable form of carbon and can, e.g., be used in agriculture by being applied to the soil, where it can contribute to humus formation in the long term as a soil conditioner and increase the water retention capacity of the soil [3,4]. During biochar production, heat can be generated from the combustion of the product gas, making biochar heating systems net energy suppliers. The startup Pyronet GmbH is developing these systems, which can be used for private, commercial, and agricultural applications [5,6].

Biochar can be produced from a wide range of biomass waste streams via pyrolysis. Pyrolysis is an endothermic process in which biomass is oxidized by the previously bound oxygen in the absence of air at a temperature between 450 °C and 700 °C. This process produces biochar, pyrolysis oil, and pyrolysis gas, with the product distribution dependent on the process conditions. Up to 50% of the carbon content of the original biomass can be stored in the biochar [7].

A pyrolysis reactor requires a heat supply to maintain high temperatures for endothermic reactions. One option is to use an external heat supply for allothermic pyrolysis. This external heat transfer becomes increasingly difficult as the reactor size increases. The increasing heating demand exceeds the possible supply through heat input due to the volumetric heating requirement versus surface heat supply [8].

Another option is autothermal pyrolysis, also known as oxidative pyrolysis, which is classified as an advanced pyrolysis process and has only been studied to a limited extent [9]. Using a reactor with a fixed bed and a co-current flow of biomass fuel and air, the pyrolysis process can be operated without an external heat supply [8,10,11,12,13,14,15,16]. This leads to fewer components and, overall, a simpler reactor design. The air supply can vary with an air excess ratio (lambda) of 0.1–0.3 for a controlled and stable process. The following sections present findings from previous studies on oxidative pyrolysis.

In an experimental study, the EEI (Escuola Ingenieria Industrial) research team at the University of Vigo, Spain, investigated aspects of the pyrolysis front as a function of the air supply in a fixed-bed co-current reactor using the TLUD (Top Lit Up Draft) principle. With a low air supply, in this case below an air excess ratio of 0.5, the reactions are limited only by the oxygen supplied, as shown in Figure 1. In this regime, the conversion rate of the supplied biomass to biochar and product gas is mass-transfer-limited and is described by a linear relationship with respect to the oxygen mass flow flux. At a certain air supply, the reactions are no longer mass-transport-limited, and not all oxygen is consumed, leading to residual oxygen reaching the produced biochar, which is further oxidized to product gas. This process must occur slowly as the overall conversion rate does not significantly increase in this regime. At stoichiometric combustion (lambda 1), the regime changes, and the conversion rate decreases, as excess air cools the process [10].

Further investigations showed that the pyrolysis front thickness is of the same order of magnitude as the particle size of the initial biomass. With a stationary and sufficiently high air supply for a stationary pyrolysis front, the pyrolysis front thickness is almost independent of the air supply. For wood pellets, the thickness is approximately 2 cm. The investigation showed that with a very low air supply, the smoldering layer increases, the pyrolysis front does not reach a stationary state, and, therefore, the thickness of the front increases [11].

The research was continued by the CIRAD research team from Montpellier, France, on a “downdraft continuous fixed bed reactor”. The reactor is operated with oxidative pyrolysis and has a wood turnover of 3.3 kg/h. The oxidation zone (pyrolysis front) and the pyrolysis products were analyzed. In this study, it was shown again on a specific reactor that the wood flux increased almost linearly with the increase in air flux in a low range of 0.014–0.029 kg/(m²s). Temperature measurements distributed over a height in the pyrolysis reactor showed that the pyrolysis front forms a surface and is orientated at a 90° angle to the air supply [14]. The lambda value for the three examined air fluxes with wood pellets remained constant, while it slightly increased with wood chips as the air supply increased. The biochar yield was not dependent on lambda. The biochar yield decreased as the pyrolysis temperature increased due to the higher air supply [14].

The research group RE-CORD (Renewable Energy Consortium for Research and Development) from Scarperia e San Piero in Italy carried out similar tests on a ten-fold larger co-current fixed-bed pyrolysis reactor with wood chips. It was confirmed that oxidative pyrolysis is scalable up to a wood turnover of 36.5 kg/h. Furthermore, the entire system was balanced, and it was determined that of the calorific value, 36% (69 kW) remains stored in the biochar and 52% (103 kW) is the thermal energy output [15].

The research group at the Institute of Thermal Engineering at the HES-SO (University of Applied Sciences and Arts Western Switzerland) investigated oxidative pyrolysis with wood chips and wood pellets at low air flows with a range of 0.008–0.017 kg/(m²s). It was shown that the pyrolysis front speed is inversely proportional to the bulk density of the biomass. For wood chips, the speed was 30 cm/h at a bulk density of 244 kg/m³ [16].

A recent study by Gao et al. has demonstrated the effect of an atmosphere containing 11% oxygen and 89% nitrogen compared to a 100% nitrogen atmosphere on the properties of biochar. The process was slow pyrolysis with a heating rate of 20 °C/min and a residence time at a final temperature of 30 min. Oxygen reduced the carbon content while oxidizing hydrogen species on the biochar surface. This resulted in a higher C/H ratio and a lower C/O ratio. An oxygen-rich environment could promote carbonization reactions, thereby enhancing crystallinity. Another influence of oxygen was observed in increasing the hydrophilicity of biochar, indicating that oxidation improved its ability to retain water [17].

In previous studies, the primary air supply in a continuous pyrolysis reactor has only been investigated in a very limited range and without focusing on meeting a variable heating demand. Our group has already conducted in-depth research in the field of biochar in recent years, investigating the influence of primary air on the pyrolysis process. In the current study, an oxidative pyrolysis reactor, with a nominal heating power of 15 kW, is investigated with the aim to vary the useful heat output and biochar yield over a wide range and still produce biochar of the highest quality in accordance with the EBC (European Biochar Certificate) guidelines. The parameters for operation with maximum flexibility are determined for a continuous, fixed-bed downdraft pyrolysis reactor intended for use as a heating system and are compared with measurements reported in the literature.

2. Oxidative Pyrolysis

To gain a better understanding of the process of generating heat and biochar without an external heat source, we explain the mechanisms of controlled oxidative pyrolysis.

While oxidative pyrolysis is not a conventional pyrolysis process, it has similarities with slow pyrolysis. The temperature is in a range of 500–800 °C; the residence time in all known setups is at least 10 min, and the heating rate is slow (max 10 °C/s) but not constant because of the acceleration at higher temperatures.

The presence of a moving pyrolysis front, in which pyrolysis reactions propagate through a fixed bed, is a characteristic of oxidative pyrolysis. Through a self-sustained heat supply via oxidation, the propagation of this front depends only on the primary air mass flux, as long as the process is oxygen limited. For the controlled movement of the pyrolysis front and high biochar quality, the primary air is directed from the fresh biomass to the pyrolysis front, known as a co-current reactor. (Co-current means that the mass flows of biomass and air come from the same direction.) Another variant is a counter-current reactor, where the primary air must first pass through the hot biochar, leading to lower biochar yields and about ten times slower pyrolysis front speed [18,19]. This study examined a co-current fixed-bed reactor.

2.1. Terms

To help in understanding the following discussion, we define important terms here. The pyrolysis temperature is the average value of the highest temperature measurement in the pyrolysis reactor over a stationary period of useful output. It is assumed that the biochar was exposed to this temperature over a longer period.

The air flux is the reactor cross-sectional area-specific primary air mass flow that is supplied to the pyrolysis zone.

The wood flux is the reactor cross-sectional area-specific pellet mass flow that is introduced into the pyrolysis reactor.

The speed of the pyrolysis front (v pyr) describes the relative speed in relation to the fresh pellets.

Lambda pyrolysis describes the molar ratio of the supplied primary air to the supplied pellets in relation to the stoichiometric amount of air required for its complete combustion.

The biochar yield is determined by the discharged biochar mass in relation to the input pellet mass over a stationary period.

The useful heat describes the sensible heat power output that is transferred to the heating water after the heat exchanger of the combustion chamber. The flow and return temperatures and the volume flow of the heating water are used to determine this.

2.2. Pyrolysis Front

In the considered control volume, as shown in Figure 2, all mass flows are visible (incoming and outgoing). The primary air and biomass mass flows are incoming, while the pyrolysis gas and biochar mass flows are outgoing. The pyrolysis front moves toward the primary air, with its speed determined almost solely by the primary air mass flux, as all oxygen reacts through partial oxidation.

At the highest temperature, oxidation reactions with oxygen from the primary air take place simultaneously with the gasification reactions. This results in increased exothermic oxidation reactions of, for example, CO to CO₂ or H₂ to H₂O, which raises the temperature of the preceding unreacted biomass by heat transfer, represented by the heat flux $\dot{\mathrm{Q}}$. In this preceding layer, the temperature is sufficient for endothermic gasification reactions and increased production of CO and H₂. The greater the temperature, the more long-chain molecules transition into the gas phase and can react with each other and with oxygen from the primary air. In the first layer, biomass drying occurs. The subdivision of the layers is not distinct, and they can overlap [18].

For a better understanding, an attempt was made to qualitatively illustrate the progression of the pyrolysis front using a two-layer pellet model, as shown in Figure 3. The process was divided into seven steps. At the start, the pellet is outside of the area of influence of the pyrolysis front.

As soon as heat is transferred from the pyrolysis front to the pellet, the outer layer begins to dry, and H₂O evaporates. At 150 °C, irreversible changes in the structure of the biomass begin by breaking up the hydrogen and carbon bonds [19]. Between 150 and 220 °C, volatile molecules and molecular fragments are released and transition into the gas phase [20]. This process closely resembles torrefaction. The temperatures at this stage are still too low for oxidative reactions with air oxygen.

Pyrolytic decomposition starts from a temperature of approximately 220 °C, at which more molecular fragments are released, which are further decomposed, and CO, H₂, CH₄, and C_xH_x are produced in endothermic reactions [21]. These do not yet react with the air oxygen due to the temperature being below 300 °C. Between 300 and 500 °C, the gas formation rate increases significantly, and more CO, H₂, and CH₄ are produced; in this temperature range, oxidation reactions occasionally take place [8,20,21].

Above 500 °C, the reaction rate increases, and all of the air oxygen is consumed at the pyrolysis front. There is no further oxygen available for gasification of the formed biochar. Depending on the oxygen supply, a temperature between 500 and 800 °C is established due to exothermic oxidation reactions. Part of the released heat is transferred to the preceding pellets and drives the pyrolysis front. The main part of the released heat is contained in the pyrolysis gas. The oxygen supply determines how much heat is released, and this heat determines the reaction rate of the endothermic reactions. The conditions described here correspond to those of “slow pyrolysis”, with relatively low heat rates and enough time for the intermediates released in the pyrolysis reactions to further react to form CO, H₂, and CH₄.

3. Laboratory Test Facility and Methodology

The core element of the experimental setup, as shown in Figure 4, is the pyrolysis reactor with the “fixed bed downdraft” principle and autothermal/oxidative pyrolysis. Product gas flows from the pyrolysis reactor into the combustion chamber, and hot biochar is discharged at the bottom. The product gas is further oxidized in the combustion chamber. The useful energy of the process is contained in the heating value and sensible heat of the exhaust gas as well as the sensible heat of the biochar.

The pyrolysis reactor is equipped with several temperature sensors (T5, T6, and T7) for process monitoring and control. Additionally, the primary and secondary air volume flows, the mass loss in the biomass feed container (fuel mass flow), and biochar quantity are measured. The pyrolysis reactor is insulated with a 100 mm thick ceramic fiber. Additional parameters of the pyrolysis reactor are displayed in

Table 1. Overview of the parameters of the pyrolysis reactor of the laboratory test facility.

Designation	Unit	Value
Reactor cross-sectional area	m2	0.026
Reactor height	m	0.37
Reactor volume	m3	0.0079
Thermal insulation thickness	m	0.1
Distance between temperature measurements (T5, T6, and T7)	m	0.06
Biomass input range	kg/h db	2.3–9.6
Primary air flux range	m3/h	2–11
Temperature range	°C	500–800

. Wood pellets certified by the ENplus^® (Austria) A1 Standard were used as fuel in this study and fed by a screw conveyor from a container. The water content of the pellets is 8.5%, and the calorific value dry is 18.5 MJ/kg. The hot biochar is discharged from the reactor at the bottom. The sensible heat from the hot biochar as it cools down can be utilized as useful heat. The combustion chamber is surrounded by a shell and a tube heat exchanger through which the exhaust gas flows. Into the combustion chamber, a secondary air flux, with a range of 7–62 m³/h, is introduced to oxidize the product gas completely. The generated useful heat is measured by the temperature increase in the water circuit at the system’s inlet and outlet and the water volume flow.

3.1. Course of the Measurement

In nominal operations, the pyrolysis front moves upward at a speed dependent on the reaction rate and biochar yield. These two variables are, in turn, dependent on the oxygen supply within the mass-transport-limited regime. Due to the volume shrinkage during the conversion of the pellets to biochar (blue), the pellet level in the reactor continuously decreases. The empty volume in the reactor is monitored by a level sensor and regularly refilled with fresh pellets (green), ensuring continuous reactor filling. The steps described are shown in Figure 5.

The process described would cause the pyrolysis front to move upward through the reactor.

As shown in Figure 6, to maintain the pyrolysis front at a fixed position, biochar is discharged (purple) when the temperature at the measurement point T6 exceeds 400 °C until the temperature drops again. The temperature drops because the pyrolysis front sinks in the reactor due to the reduced biochar volume. Pellets are refilled (green) in the empty volume at the top of the reactor during biochar discharge. Due to the volume shrinkage during the pyrolysis of the pellets to biochar, the volume of the produced biochar is smaller than the volume of the pellets consumed.

In a stationary state, the discharged biochar corresponds to the yield relative to the introduced pellets, assuming that the bulk density of the biochar remains constant. However, the bulk density of the biochar in the pyrolysis reactor before discharge is a variable that cannot be measured.

The course of a measurement using three temperatures in the pyrolysis reactor is shown in Figure 7.

The pyrolysis front is held at the point of T6, and the biochar discharge is controlled. T5 is located below the pyrolysis front in the produced biochar, and T7 is located above the pyrolysis front in the still cold pellets.

3.2. Calculations for Characterization

In Figure 8, the system used to calculate the mass balance is shown and

Table 2. Symbols used in calculations with the corresponding unit.

Designation	Unit	Symbol
Mass flow of biochar	kg/s	${\dot{m}}_{BC}$
Mass flow of dry biomass	kg/s	${\dot{m}}_{BM}$
Mass flow of exhaust gas	kg/s	${\dot{m}}_{EG}$
Mass flow of primary air	kg/s	${\dot{m}}_{PA}$
Mass flow of secondary air	kg/s	${\dot{m}}_{SA}$
Difference in mass over time	kg	$∆m$
Timeframe	s	$∆t$
Volume flow of primary air	m3/s	${\dot{V}}_{PA}$
Density of air (primary and secondary)	kg/m3	${\rho }_A$

describes the symbols used in the mass and energy balances. It is evident that three mass flows enter, and two mass flows exit the system boundaries. For the balance, it is assumed that the process is stationary.

The formula for the mass balance is as follows:

$${\displaystyle \frac{d{m}_{\mathit{KV}}}{dt}}={\dot{m}}_{BM}+{\dot{m}}_{PA}+{\dot{m}}_{SA}-{\dot{m}}_{BC}-{\dot{m}}_{EG}=0,$$

(1)

For comparability, the mass balance is based on dry biomass (100%).

To determine the mass flow of the biomass fed in, the weight loss of the daily pellet container over time was measured and converted to dry mass flows. Pellets are continuously transported from this container into the pyrolysis reactor using a screw conveyor.

$${\dot{m}}_{BM}=∆m÷∆t={\displaystyle \frac{\left(m_{t2}-m_{t1}\right)}{\left(t_2-t_1\right)}},$$

(2)

The volume flow of primary and secondary air was measured directly with a hot-wire anemometer.

$${\dot{m}}_{PA}={\dot{V}}_{PA}\times {\rho }_A,$$

(3)

The mass flow of the produced biochar was determined over the measurement duration by weighing the produced biochar mass.

$${\dot{m}}_{BC}={\displaystyle \frac{∆m_{BC}}{\left(t_2-t_1\right)}},$$

(4)

The combustion gas mass flow was determined using the mass balance. Due to tars in the exhaust gas at some operation points, the volume flow could not be measured directly.

$${\dot{m}}_{EG}={\dot{m}}_{BM}+{\dot{m}}_{PA}+{\dot{m}}_{SA}-{\dot{m}}_{BC},$$

(5)

Figure 9 shows the system used to perform the energy balance.

The calorific values of the biomass ($H_{u,BM}$) and the produced biochar ($H_{u,BC}$), along with the useful heat (${\dot{Q}}_{Use}$), were measured to evaluate the overall energy efficiency and performance of the pyrolysis process. The reference temperature of the environment ($T_{Env}$) is 25 °C so that the calorific values can be used. The difference from an ideal balance under stationary conditions is the loss of the exhaust gas (${\dot{Q}}_{EG}$) and the loss of heat via the housing of the system (${\dot{Q}}_{Loss}$). Complete combustion is assumed in the combustion chamber.

The energy balance is as follows:

$${\displaystyle \frac{d{E}_{KV}}{dt}}={\dot{m}}_{BM} H_{u,BM}-{\dot{m}}_{BC} H_{u,BC}-{\dot{Q}}_{EG}-{\dot{Q}}_{Loss}-{\dot{Q}}_{Use}=0,$$

(6)

The energy for pyrolysis (which is provided by the internal oxidation reactions) can be calculated using the following formula, considering the heating of the biomass and the reaction enthalpy [8]:

$${\dot{Q}}_{Pyr}={\dot{m}}_{BM} \left(c_{p,BM} \left(T_{Pyr}-T_{Env}\right)+∆h_{pyr}\right),$$

(7)

The reaction enthalpy of pyrolysis ($∆h_{pyr}$) is determined using a differential thermal analysis (DTA). This has been documented in various publications. A linear relationship dependent on the biochar yield ($Y_{BC}$) was found as follows [8]:

$$∆h_{pyr}\left[{\displaystyle \frac{kJ}{kg}}\right]=553-3142 Y_{BC},$$

(8)

The heat losses through the exhaust gas depend on the exhaust gas temperature ($T_{EG}$) with respect to the heat capacity of the exhaust gas ($c_{p,EG}$).

$${\dot{Q}}_{EG}={\dot{m}}_{EG} c_{p,EG} \left(T_{EG}-T_{Env}\right),$$

(9)

For comparability, the energy balance is based on dry biomass (100%).

The velocity of the pyrolysis front (${\nu }_{PF}$) was calculated using the relationship between the measured wood flux (${\varphi }_w$) and wood bulk density (${\rho }_w$) [14].

$${\nu }_{PF}={\displaystyle \frac{{\varphi }_w}{{\rho }_w}}$$

(10)

4. Results

To meet the variable heating demands of both single and multi-household applications, a central heating system must work at a broad range of operating points. Three different operating points from the pyrolysis heating system are characterized in

Table 3. Overview of the parameters of characterization (* accuracy is ± 2%).

Designation	Unit	Partial Load	Nominal Load	Maximal Load
Air flux	kg/(m2s)	0.027	0.075	0.144
Pyrolysis temperature	℃	520	724	809
Wood flux	kg/(m2s) db	0.025	0.052	0.102
Biomass input	kg/h db	2.3	4.9	9.6
Velocity pyrolysis front	cm/h	16	33	65
Lambda pyrolysis	-	0.20	0.26	0.25
Biochar yield *	kg/kg db	23.7%	14.2%	12.4%
Useful heat	kW	4.1	15.5	29.5
	kW/m2	158	596	1132

to provide an overview of the range of key figures at a nominal load (100%), partial load (30%), and maximal load (200%). The air flux was varied in the range of 0.027–0.144 kg/m²/s. At an air flux of 0.144 kg/m²/s, the limit of the used suction fan was reached.

The air flux drives the oxidation reactions at the pyrolysis front, and the exothermic reactions determine the pyrolysis temperature at this front. Consequently, the pyrolysis temperature dictates the rate at which fresh biomass converts to product gas and biochar, thereby defining the wood flux. Figure 10 shows the influence of air supply in the pyrolysis reactor on the temperature and the influence of the pyrolysis temperature on the wood flux.

Although only three air flux conditions were measured, it appears that the influence of air flux on the pyrolysis temperature decreases as the flux is increased. The increased air supply has a partially cooling effect due to the mass of cold air. While a higher temperature in the pyrolysis reactor increases the heat transfer to the environment, the influence of the air flux is reduced.

Higher pyrolysis temperature positively affects the amount of converted wood mass. It is assumed that the speed of the pyrolysis front toward fresh pellets increases with higher temperatures. As long as the reactions are oxygen limited, wood conversion is directly dependent on the air supply, as shown in Figure 11.

The CIRAD measurements showed that with a low air flux and an optimized reactor (well-insulated, preheated air supply) for oxidative pyrolysis, the wood flux is linearly dependent on the air flux. In the RE-CORD measurement, a similar characteristic of the chips was measured in a ten-fold larger reactor than in CIRAD but with a similar design.

The EEI measurements of the pellets showed, however, that a maximum for the wood flux can be reached at an air flux of 0.3 kg/(m²s). This maximum depends on the reactor type and design. EEI employed a basic reactor design without insulation, leading to increased heat loss that had to be compensated for by oxidation reactions. Consequently, these measurements required a higher air flux to achieve a wood flux comparable to that of other reactors. The area of the transition from the oxygen-limited to the reaction-limited regime is marked in red in Figure 11. It is not yet clear where this transition is for other reactor types.

The measurements of this study at a maximum load showed a wood flux above 0.1 kg/(m²s), and the curve did not flatten. This suggests that the maximum for stable oxidative pyrolysis is higher. Once the maximum is reached, reactions are no longer limited by the oxygen supplied but by the reaction kinetics.

It is assumed that at the maximum, the regime shifts from predominantly pyrolysis to biochar gasification, significantly reducing the biochar yield. This determines the modulation limit for the pyrolysis reactor. The exact point of this limit is not yet determined. Attempts to determine the maximum power in this study still produced biochar, but the measured yield is inaccurate, so no absolute statements can be made.

The influence of pyrolysis temperature on biochar yield is shown in Figure 12, where we compare our results and those of CIRAD [14] to values in the literature for non-oxidative pyrolysis with an external heat supply [7]. (Only data for temperatures above 400 °C are shown, where biochar is formed with a high carbon content.) The literature values (measurements and calculations) show a decrease in biochar yield with the increasing pyrolysis temperature, as more molecules are released and transition to the gas phase.

The biochar yield measurements by IBRE and CIRAD are slightly lower than the literature values. Adding oxygen in the pyrolysis reactor for oxidative pyrolysis produces more pyrolysis gas, oxidizes more biochar, and produces less biochar than conventional pyrolysis with an external heat supply [8,12].

The biochar yields of CIRAD are slightly higher than those of this study with comparable temperatures. This is caused by the different setups with the following impact factors: reactor geometry, air preheating, residence time, and thermal insulation.

4.1. Mass and Energy Balance

The mass balance is reported in Figure 13.

Approximately twice as much oxygen is required at a nominal load as compared to a partial load. This leads to more than three times the useful heat output. The higher primary oxygen mass flux leads to a higher pyrolysis temperature, which results in a reduced biochar yield. The energy balance associated with the mass balance is shown in Figure 14.

The energy requirement for oxidative pyrolysis to maintain the pyrolysis temperature was calculated using Equation (7). This energy is not lost but stored in the hot biochar and pyrolysis gas. The energy for pyrolysis was found to be between 6 and 12% of the heating value contained in the dry biomass. As a comparison, 10% was estimated in the literature, and RE-CORD indicated 10% for a 50 kW reactor [15,18].

By reducing the power by reducing the air flux, more energy from the biomass is stored in the biochar, and less energy is transferred as useful heat in the heat exchanger. At a nominal load, 11.5 MJ/kg of useful heat, or 62% of the wood heating value, is produced per kilogram of pellets. At a partial load, 6.4 MJ/kg of useful heat, or 35% of the wood heating value, is produced per kilogram of pellets. This means that at a partial load, about 56% less useful heat is produced than at a nominal load. The relationship between the air supply for oxidative pyrolysis and the resulting specific useful energy is shown in Figure 15.

The relationship between the supplied air flow and the specific useful heat, determined by the cooling of the exhaust gas and the biochar, is linear in the investigated range. This shows that a heater with oxidative pyrolysis can be modulated over a wide range solely by adjusting the air supply.

Figure 16 illustrates the response time of the useful heat output to a sudden change in air supply. In this instance, the useful heat output tripled and stabilized within 13 min, increasing from 4 kW to 12 kW by augmenting the air supply in the pyrolysis reactor. Other measurements indicated that the response time for decreasing the useful heat output is comparable. The impact on biochar quality will now be addressed.

4.2. Biochar Quality and Properties

Table 4. Composition of biochar produced at different operation points. (db = dry basis, n.m. = not measurable, below detection limit).

Designation	Unit	Input Biomass	Partial Load	Nominal Load
Pyrolysis Temperature	°C	-	520	724
Water Content	kg/kg	8.49%	3.40%	4.30%
Ash Content	kg/kg db	0.30%	1.30%	2.20%
C	kg/kg db	49.80%	92.80%	95.40%
H	kg/kg db	6.10%	1.70%	1.30%
O	kg/kg db	44.10%	4.30%	1.60%
H/C	molar	1.47	0.22	0.16
O/C	molar	0.672	0.035	0.013
Sum 16 EPA-PAH	mg/kg	-	1.6	n.m.

shows the composition of the initial biomass and the biochar produced with a nominal load of 15 kW and a partial load of 4 kW.

At a partial load, the temperature was the lowest at 520 °C, resulting in the lowest pyrolysis temperature of the samples studied. This leads to a lower degree of carbonization, indicated by the carbon content, and a PAH (Polycyclic Aromatic Hydrocarbons) contamination. The measured PAH contamination of 1.6 mg/kg is very low and well within the limits set by the EBC guidelines for Feed Plus, which is 6 mg/kg. The hydrogen content is very similar in the samples and has a low temperature-dependency in the studied range. The original pellets had a hydrogen content of 8.49%. This could indicate that H₂ and long-chain CH compounds are already produced at lower temperatures. Due to the differences in oxygen content and the decrease at higher temperatures, it is assumed that a higher temperature than 520 °C produces more CO. This CO is further oxidized by the supplied air and is, in turn, a reason for the temperature increase. The ash content indicates the degree of concentration of the biochar and the yield. We started with an ash content of 0.3% in the wood pellets. At a pyrolysis temperature of 520 °C, the ash content is 1.3%, and at a pyrolysis temperature of 724 °C, the ash content rises to 2.2%. The increase in ash content is due to concentration in the biochar.

As visible in Figure 17, the higher the pyrolysis temperature, the higher the concentration of carbon in the biochar. (Some measurement points resulted from a slightly different reactor design, which had a comparable residence time at the referenced temperature.) As a reference, there is a graph for various solid fuels, and the red rectangle shows the range of the whole diagram [22].

It is interesting to note that at a temperature above approximately 700 °C, no major changes in the ratios are evident. This is an indication that carbon, oxygen, and hydrogen are transferred more evenly into the gas phase above this temperature. The biochar produced at a temperature of 700 °C is very similar in composition to anthracite, which has an O/C ratio under 0.03 mol/mol and an H/C ratio under 0.35 mol/mol.

The characteristics of the pyrolysis reactor and biochar are summarized in

Table 5. Biochar calorific value and potentially captured carbon.

Designation	Unit	Partial Load	Nominal Load
Pyrolysis Temperature	°C	520	724
Useful Power	kW/m2	165	621
Calorific value	MJ/kg db	33.4	34.2
Biochar Yield	kg/kg db	23.7%	14.2%
C in Biochar	% db	44%	27%

. The calorific value of biochar is very similar at the temperatures analyzed, as the carbon and hydrogen content are also very similar. Largely depending on the yield, almost twice as much carbon of the initial biomass is stored in the biochar at 520 °C than at 724 °C. An analysis of the stability of biochar in the soil showed that the decomposition rate decreases with the higher pyrolysis temperature, but the difference between 500 °C and 700 °C is not significant [3]. Biochar produced at a partial load can capture up to 44% of the carbon from the original biomass.

5. Discussion

Further investigations into oxidative pyrolysis have provided data on the performance of such heating systems with wood pellets. In oxidative pyrolysis, thermal energy is provided at the pyrolysis front through exothermic oxidation reactions with primary air. This study demonstrated the influence of air mass flux on the process of oxidative pyrolysis and the produced biochar. In the air flux range of 0.027 to 0.144 kg/m²/s, the air excess ratio (lambda) in the pyrolysis reactor is nearly constant with a range of 0.2–0.26. Maintaining a constant lambda while increasing the air flux results in an increased movement velocity of the pyrolysis front. The pyrolysis front, confined to a small region within the pyrolysis reactor, accelerates when the air flux is increased as long as the reactions are limited by the oxygen supply. This leads to an almost linearly dependent wood flux in the range of 0.025 to 0.102 kg/m²/s, as is the useful power of the investigated pyrolysis heater.

The evaluated design of a pyrolysis reactor produces biochar, which meets the quality requirements of the EBC (European Biochar Certificate) Feed Plus limits for all conditions. During operations at a specific load setting, the pyrolysis temperature (T5 in Figure 7) remained stable, resulting in consistent biochar quality. It was shown that with a constant pyrolysis front area and a pyrolysis temperature range of 520 °C to 724 °C, the biochar mass yield ranges from 14% to 24%. At a nominal load of 621 kW/m², 62% of the pellet heating value was available as useful heat from combustion of the product gas, and 26% was stored in the biochar. At a partial load with 165 kW/m², the proportion of heat losses increases, with 35% converted to useful heat and 43% stored in the biochar. Regarding the limit of oxidative pyrolysis and the transition to gasification at higher air flux rates, it was shown that at an air flux of 0.144 kg/m²/s and a specific useful power of 1132 kW/m², oxidative pyrolysis is stable and achieves a biochar yield of 12% ± 2%. Uncertainty remains regarding biochar yield at higher temperatures and the limit of oxidative pyrolysis and the oxygen-limited regime, as shown by EEI [10].

The influence of the input biomass characteristics on biochar quality was analyzed for wood pellets. It is expected that the quality of biochar from wood chips will be similar depending on the pyrolysis temperature; this must be verified by further measurements. For alternatives to wood, it must be determined whether oxidative pyrolysis is stable and what the influence on the biochar is. Initial investigations have already been carried out by HES-SO with LADR (lignocelulosic anareobic digestion residues). LADR has an ash content of 20%, while wood chips have an ash content of 0.5%. These investigations showed that the yield and composition at 600 °C are similar to those of wood chips. There were differences in the carbon content, which was only around 50% for LADR, and in the tar content in the pyrolysis gas, which was 30% higher for LADR [16].

The advantage of oxidative pyrolysis compared to allothermal pyrolysis is that the process depends almost exclusively on one variable, the primary air, which can be increased or reduced quickly. The pyrolysis front limits the pyrolysis process to a specific cross-sectional area, making it significantly easier to scale with primary air than by heat transfer through the reactor outer wall, as is common in allothermal pyrolysis [8]. By dividing the heating system into two stages, a pyrolysis reactor and a combustion chamber, the solid phase of the biomass remains in the pyrolysis reactor, and only the gas phase enters the combustion chamber, resulting in low particle emissions. One potential issue is the relatively high front temperature in oxidative pyrolysis of up to 700 °C, which may impact performance when using alternative biomasses with high ash contents and low ash melting points, such as grain waste products.

A heating system driven by oxidative pyrolysis offers flexibility in applications. The broad range of variable heating output allows for a choice between more heat and lower biochar yield or lower heat and higher biochar yield. However, the disadvantage of oxidative pyrolysis compared to allothermal pyrolysis is the reduced biochar yield due to partial oxidation for heat generation, as shown in Figure 12.

The precise boundaries of the stable oxidative pyrolysis range, along with the variability in wood flux and biochar yield, remain as areas requiring further investigation. Further studies are essential to fully understand these parameters and optimize the process for different biomass types and operating conditions. A potential application could involve utilizing a single reactor capable of operating in oxidative pyrolysis mode during periods of moderate heat demand and switching to full gasification for short durations when high heat demand is required. Additionally, the influence of air flux and pyrolysis temperature on the lambda value in the pyrolysis reactor is not yet fully understood and should be a part of future research.