Microwave Pre-Treatment and Blending of Biomass Pellets for Sustainable Use of Local Energy Resources in Energy Production

Abstract

In the present study analysis of co-firing microwave (MW) pre-treated biomass pellets of different origins (wood and wheat straw), with raw pellets (wood, straw, and peat), to control and improve thermochemical conversion of biomass blends and achieve a sustainable use of local energy resources in energy production has been carried out. Effects of MW pre-treatment regimes and composition of blends were studied experimentally using measurements of the weight loss of blends, the yield of volatiles, flame temperature, total heat output from the device, and composition of products. It was found that co-firing MW pre-treated and raw biomass pellets promotes synergistic interaction between components of blends by increasing mass loss rate, the intensity of which depends on the proximate composition of pellets, MW pre-treatment regime and mass fraction of pre-treated pellets in the blend. The most effective synergistic interaction was found when co-firing pre-treated straw or wood pellets with raw peat, which increased the yield of combustible volatiles and heat output from the device as well as improved the composition of emissions. The least effective synergistic interaction was observed when co-firing pre-treated straw with raw wood pellets. Main factors that influenced the thermal and chemical conversion of MW pre-treated blends are discussed considering the effects of MW pre-treatment on the structural changes, elemental and chemical composition, and heating value of pre-treated pellets.

1. Introduction

In accordance with requirements set in the context of climate change by the 2030 Agenda for Sustainable Development and the Paris Agreement [1,2] and a mounting push from the United Nations to increase renewable energy sources (RES) quotas and move to greener technologies [1,2,3,4], progress is to be made towards accessibility and improvement of sustainable energy production, clean and safe fuels, and technologies in district heating. To achieve sustainable growth and limit climate change, a transition from conventional fossil fuels (coal, oil, and natural gas) to renewable energy recourses (RES) with reduced carbon-neutral emissions in district heating is needed to provide innovative solutions [4,5].

Renewable energy technologies should aim to cover a significant portion of electricity demand and reduce GHG emissions, increasing the possibilities for regional energy supply, which are not associated with environmental pollutions from fossil fuels, long-term hazardous waste storage, or high impact accident risks (e.g., mismanagement or disaster, natural or otherwise) linked to nuclear power [6]. Biomass is among the alternatives to fossil fuels that provides sustainable, low emission energy production [7,8]. While it does not offer energy production with no associated greenhouse gas (GHG) emissions, it does provide carbon-neutral CO₂ emissions that have been locked into a biomass that is GHG renewable in the short-term; for example, wood or straw. The use of biomass for energy production is limited, and according to goals of the energy framework [9], research must be carried out to increase bioenergy production, gradually replacing fossil fuels with biofuels as well as using blends of different origin biofuels [10,11,12,13].

Lignocellulosic biomass, such as crops, wood, and agricultural and forestry residues, is a major biomass resource and has been recognized as a sustainable feedstock to produce green bioenergy. The main reason for its limited use is its high diversity of physical and chemical properties, high moisture content, and low heating value in comparison to fossil fuels. Therefore, technologies must be developed to convert biomass feedstocks into more applicable fuel forms, producing biogas, bioethanol, or solid biofuels with improved and controllable characteristics as compared with raw biomass feedstocks.

The main objective of this study is to provide more efficient use of solid biofuels and improve their characteristics. To achieve this, convective heating pre-treatment/torrefaction of lignocellulosic biomass at 200 °C to 300 °C in an inert or non-oxidative environment was carried out. This process leads to weakening of the cell-wall structure, thus increasing energy density and improving grindability as well as hydrophobicity [14,15,16,17]. More effective torrefaction can be carried out using microwave (MW) pre-treatment of biomass at a frequency of 2.45 GHz [18,19,20]. MW heating in many ways can have advantages over conventional heating. First, solid products of MW pre-treatment can have higher carbon contents and heating values, and higher specific surface areas, with higher gas and solid yields but lower liquid yields than products of conventional pyrolysis [19,20,21]. Second, MW pre-treatment is a faster and less energy-consuming process for biofuel production that enhances cracking reactions [22]. MW-induced changes in the main characteristics depend on a pre-treatment regime, which allows to provide controllable changes in the main flame characteristics, heat energy production, and composition of products [23,24].

The results of preliminary research suggest that MW pre-treated biomass pellets can be used to produce selectively activated biomass blends by mixing them with different types of raw biomass pellets, providing faster and more complete thermochemical conversion [25]. Moreover, compared to untreated blends [26,27], a positive synergistic effect of MW pre-treatment on the thermochemical conversion of blends has been observed, that depend on numerous factors [28,29,30,31] and requires further investigation.

In this study, three types of biomass pellets—wood, wheat straw, and peat—were used. The main motivation of using these types of biomasses was their regional availability since all used pellets are commercially produced in Latvia. More importantly, used wood and wheat straw pellets were made from agricultural and industrial residues which makes them good potential candidates for MW pre-treated additives.

In this research, assessment of the main factors that influence thermal decomposition of MW pre-treated biomass pellets blends available in Latvia was carried out. The influence of synergistic effects on thermochemical conversion of the blends and optimal pre-treatment regimes and composition of the blends were evaluated. Thus, the results of this research will offer opportunities for wider use of different origin biomass pellets for sustainable energy production in district heating devices.

2. Experimental Equipment and Methods

Locally commercially available softwood, wheat straw, and peat pellets with a preliminary defined elemental composition (

Table 1. Proximate composition and higher heating value (HHV) of commercial wood, wheat straw, and peat pellets (on dry basis).

Biomass Type	C, %	H, %	N, %	O, %	HHV, MJ·kg−1	Ash, %	Moisture, %
Wood pellets	50.59	5.45	0.17	43.47	19.94	0.3	7.14
Wheat straw pellets	46.43	5.79	0.59	41.50	18.41	3.7	10.2
Peat pellets	52.81	5.20	1.17	37.48	20.86	3.4	8.9

) were used to produce MW pre-treated blends: straw* + straw, straw* + wood, straw* + peat, wood* + straw, wood* + peat (where “*” denotes “MW pre-treated”). The pre-treatment temperature of pellets and the mass fraction of pre-treated pellets were varied in the blends.

Wood, wheat straw, and peat pellets all have different chemical compositions [23] with the highest content of hemicelluloses in wheat straw pellets (28.1%), highest contents of cellulose (43.2%) and lignin (28.83%) in wood pellets, and lowest content of hemicellulose (13–22%), cellulose (4.78%), and lignin (7.12%) in peat pellets. In addition, wood, wheat straw, and peat pellets all have different contents of volatile matter in their biomass [32], with the highest content of volatile matter in wood biomass (77–79%), and lower contents of volatiles in wheat straw (67.8–74.8%) and peat (57.8–67.6%). To produce MW pre-treated biomass blends, pre-treatment of pellets was carried out at frequency 2.45 GHz using an originally developed MW reactor with preliminary convective heating of pellets (300 g) in an argon atmosphere up to the maximum temperature (T = 473 K, 548 K), which was followed by isothermal MW heating for 20 min [21]. The effect of MW pre-treatment on the element composition of pellets was estimated providing measurements according to standard LVS EN 15104:2011, and HHV was calculated based on elemental analysis data.

In Figure 1, the effect of MW pre-treatment on biomass composition is shown. MW pre-treatment of wood and wheat straw pellets enhances the carbonization of pellets, increasing the relative content of carbon (C_MW/C_raw) and heating value (HHV_MW/HHV_raw) of the pellets, while decreasing the relative content of oxygen in pellets (O_MW/O_raw). In addition, MW pre-treatment of wood pellets promotes an increase up to a maximum value of the relative content of hydrogen in pellets (H_MW/H_raw), gradually decreasing the hydrogen content in wheat straw pellets, while to the minimum value decreasing the relative nitrogen content in pellets (N_MW/N_raw), if compared with raw wood or wheat straw pellets.

In addition to changes in elemental composition and HHV, MW pre-treatment of pellets results in structural changes in pellets, increasing specific surface area and porosity [23]. The influence of MW-induced structural changes and changes in elemental composition of pellets on thermochemical conversion of selectively pre-treated blends was estimated, providing simultaneous measurements of the weight loss of blends during their thermal decomposition, flame temperature, heat output from the device, and composition of the products. The experimental study was provided using the universal batch-size device, which was developed for experimental study of the thermochemical conversion of different origin biomass pellets and their mixtures [25].

The experimental device consisted of a gasifier (1) and the combustion reactor (3) with water cooled walls outputting on average 5 kW of heat [25] (Figure 2). Biomass blends in the device were introduced from top of the reactor. The thermal decomposition/gasification of pellets was initiated providing an additional heat supply at the top of the blend (2) using propane flame flow. During thermal decomposition of pellets, a primary air supply of average rate 0.82 g/s was introduced from below, while combustion of volatiles and char was supported by secondary swirling air supply above the blend (7) at average rate 1.23 g/s. The flame temperature and composition of the products were measured using orifices (4, 5) designed for the injection of diagnostic tools in the flame flow.

Estimation of the weight loss rate of pellets during thermal decomposition was performed using continuous measurements of the height of biomass layer in the gasifier with a moving rod and a pointer, which allowed to control the variation of the biomass height in a gasifier (dL/dt, cm/s) with an accuracy of ±0.5%. Thus, accuracy of weight loss rate dm/dt, g/s was ±2%. The yield of volatiles was measured using a gas sampling probe in the outlet of the gasifier. Then composition of probe material was measured using a Testo 350 XL gas analyser.

Pt/Pt/Rh thermocouples connected to the Pico data logger were used for dynamic measurements of the flame temperature with an accuracy of ±5%. Measurements of the flame and outlet products composition—the mass fraction of volatiles (CO, H₂), the volume fraction of the main product (CO₂), the mass fraction of NO_x, the combustion efficiency, and the air excess ratio (α)—were made using a gas sampling probe and a Testo 350 XL gas analyzer. According to instrument specification, the CO₂ volume fraction was measured with an accuracy of ±1%, and CO, H₂, and NO_x with an accuracy of about ±0.5% for the mass fraction.

The calorimetric measurements of the cooling water flow were provided to estimate heat output from the device and produced heat energy mass density of burned pellets. They involve joint measurements of the cooling water mass flowrate, which was measured with an accuracy of ±2.5% as well as the water flow temperature at the inlet and outlet of each section with an accuracy of ±1% using sensors AD 560 and Data Translation DT9805 data acquisition module [23]. The local measurements of the axial (u) and tangential (w) flow velocity components by varying a distance from the flow centerline (r/R = 0) were made using a Pitot tube and a Testo 435 flowmeter, providing continuous online data monitoring with an accuracy of ±1%.

3. Results

3.1. Thermal Decomposition of Different Origin Microwave Pre-Treated Biomass Blends

At the given configuration of the experimental device, the thermochemical conversion of biomass blends is governed by thermal decomposition/gasification of the biomass (1) promoting the yield of both condensable and non-condensable fractions of pellets. The main steps of thermal decomposition include biomass drying, the yield of volatiles, partial oxidation of combustible volatiles (2,3), char oxidation (4,5), and endothermic Boudouard, water vapor reactions (6,7):

$$C_x{H}_y{O}_z+heat\to aC{O}_2+b{H}_{2}O+cC{H}_4+dCO+e{H}_2+char+tar$$

(1)

$$CO+0.5O_2\to C{O}_2\hspace{1em}\hspace{1em}∆H=-283 \mathrm{kJ}/\mathrm{mol}$$

(2)

$$H_2+0.5O_2\to H_{2}O\hspace{1em}\hspace{1em}∆H=-242 \mathrm{kJ}/\mathrm{mol}$$

(3)

$$C+0.5O_2\to CO\hspace{1em}\hspace{1em}\hspace{1em}∆H=-111 \mathrm{kJ}/\mathrm{mol}$$

(4)

$$C+O_2\to C{O}_2\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}∆H=-393.5 \mathrm{kJ}/\mathrm{mol}$$

(5)

$$C+C{O}_2\to 2CO\hspace{1em}\hspace{1em}\hspace{1em}∆H=+159.9 \mathrm{kJ}/\mathrm{mol}$$

(6)

$$C+H_{2}O\leftrightarrow CO+H_2\hspace{1em}\hspace{1em}∆H=+118.5 \mathrm{kJ}/\mathrm{mol}$$

(7)

In kinetic study of biomass blend thermal decomposition it is important to assess the contribution of each component in the process due to the different composition of blends. If the overall process is developing uniformly throughout the pellets, the weight loss of blend during the thermal decomposition can be approximately described using first-order reaction model [31]:

$$dm/dt=k\left(1-m\right)$$

(8)

$$m=\left(m_0-m_t\right)/\left(m_0-m_f\right)$$

(9)

$$k=A·exp\left(-E/RT\right)$$

(10)

where $m$ is the mass conversion, $k$ is the reaction rate constant, $m_0$ is the original mass of the test sample, $m_t$ is the mass of the test sample at time $t$, $m_f$ is the final mass at the end of thermal decomposition, $E$ is the activation energy, $T$ is the temperature, $A$ is the pre-exponential factor, and $R$ is the universal gas constant.

A more detailed approach is to use models with distributed activation energy [33,34], where the thermal decomposition process can be described as many independent, parallel rate processes assuming them as irreversible first-order reactions. Considering the mass balance of species i, reactions for volatile production read:

$$d{V}_i/dt=k_i\left(V_i^t-V_i\right)$$

(11)

$$V_i^t-V_i=V_i^t·exp{\left(A_i·exp\left(-\frac{E}{RT}\right)\right)}_i$$

(12)

where $V_i^t$ is the final quantity of volatile matter for the generic species, $i$, and k_i is the rate constant of the reaction expressed according to the Arrhenius law.

To successfully use this model, kinetic parameters of reactions (1–7) are necessary. Estimation of kinetic parameters from experimental data of biomass thermochemical conversion practically is not possible and data from the literature are necessary.

In a recent study, the weight loss rate and the yield of volatiles were studied and compared using the most widely available bioenergy resources in Latvia (wood, agriculture residues, and peat) and producing the selectively activated blends of commercially available pellets (straw* + straw, straw* + wood, straw* + peat, wood* + straw, wood* + peat) (Figure 3a–e).

Results of kinetic studies of the weight loss of blends during their thermal decomposition/gasification have shown that for all blends MW pre-treatment of pellets has a significant role on the kinetics of the weight loss rate. In Figure 3 it has been shown that even a relatively small mass fraction of pre-treated pellets in the blend (30%) can vary the weight loss rate.

Comparing average values of the weight loss rate of biomass blends of different MW pre-treatment temperatures and different mass fraction of pellets (Figure 4), it can be concluded that MW pre-treatment of pellets promotes a higher weight loss rate and devolatilization of the blend than predicted using a simple additivity approach (13) [25]. Such deviation from purely linear behavior suggests synergy mechanism between pre-treated and raw biomass pellets in the blends.

$$\frac{dm}{dt}=C_{\ast }{\left(\frac{dm}{dt}\right)}_{\ast }+C_0{\left(\frac{dm}{dt}\right)}_0$$

(13)

where: $\frac{dm}{dt}$, g/s—weight loss rate of the blend, ${\left(\frac{dm}{dt}\right)}_{\ast }$, g/s—weight loss rate of MW pre-treated wheat straw or wood pellets, ${\left(\frac{dm}{dt}\right)}_0$, g/s—weight loss rate of raw peat pellets, $C_{\ast }$—mass fraction of pre-treated pellets, and $C_0$—mass fraction of raw pellets in the blend.

The results in Figure 4 allow to conclude that for the most of selectively activated blends a positive synergistic effect on the weight loss rates is achieved. However, a slightly negative synergistic effect is observed for the blend of raw wood with pre-treated wheat straw pellets.

Comparing the effect of changes in MW pre-treatment temperature of pellets on the kinetics of the weight loss rate of blends it can be concluded that they are significantly different (Figure 3a–e) and can be related to both-diversity of the main characteristics of the blend components (Table 1, Figure 1) and MW-induced structural changes. It leads to variations of combustible volatile yield and a balance between the rates of reactions (1–7), which is confirmed by measurements of combustible volatiles yields at the outlet of the gasification reactor (Figure 5a–f) and supported with measured yields of CO₂ and CO for the different MW pre-treatment temperatures in studied biomass blends (Figure 6a–e).

Providing analysis of the main processes, which promote changes in the weight loss rates of the blends and the yield of volatiles during thermal decomposition/gasification of selectively MW pre-treated blends, it should be noticed that MW-induced structural changes with increased reactivity of pellets and changes in elemental composition with carbonization [23,24] of pellets enhance the development of gasification reaction (4), increasing the weight loss rate of blends, the yield of CO and the temperature in gasification zone. Increasing the temperature in the gasification zone enhances endothermic drying in raw biomass, thus promoting the yield of H₂O and additional weight loss in the blend defined by the moisture content and mass fraction of raw pellets in the blend.

Drying is followed by the enhanced thermal decomposition of blend components by increasing of the yields of volatiles (CO₂, CO, and H₂) and the weight loss. In accordance with data [35], polymer-like hemicelluloses decompose, increasing the temperature to 540–560 K and promoting the enhanced yield of CO₂, cellulose decomposes at temperatures above 580–600 K with the enhanced yield of CO, while lignin decomposes at temperatures between 430–1100 K with the enhanced yield of H₂. The enhanced yields of H₂O and CO₂ during drying and thermal decomposition of hemicelluloses are the main factors, which promote an endothermic carbon and water steam reaction (7) increasing yields of CO and H₂ as well as the Boudouard reaction (6), which has high importance in the process of biomass gasification [19]. These reactions are responsible for the additional yield of CO and the weight loss of the blends.

This is confirmed by analyzing results of the FTIR measurements (Figure 7), where the yields of CO₂ and CO at different stages of thermal decomposition selectively pre-treated blends are shown. Two major stages can be distinguished. In the primary stage (e.g., Figure 7a, t < 1200 s and Figure 7b, t < 2000) enhanced gasification promotes correlating increase in yields of CO and CO₂. In the second or end stage (e.g., Figure 7a, t > 1200 s and Figure 7b, t > 2000) the char conversion occurs and development of the Boudouard reaction promotes an increase in the CO yield with correlating decrease in the CO₂ yield.

The balance between the exothermic gasification of biomass blend (4) and endothermic processes (6, 7) changes the weight loss rates of the blends (Figure 3a–e) and the average values of the yields of volatiles (Figure 5 and Figure 6a–e) depending on pre-treatment conditions and composition of blends, thus varying the thermal and chemical interaction between the components and synergistic effects of interactions (Figure 4).

In cases of a positive synergetic effect, e.g., when pre-treated wheat straw pellets are co-fired with raw straw, wood, or peat pellets, the exothermal gasification of pre-treated pellets acts as a heat source, enhancing drying and thermal decomposition of raw pellets. It leads to a faster weight loss of the blend (Figure 3a) and the yield CO₂ and by varying relation between the yields of CO₂ and CO. In case of low pre-treatment temperature of wheat straw pellets (T = 473 K) the balance between exothermic and endothermic reactions (1, 4, 6, 7) is achieved, when the mass fraction of pre-treated straw pellets in the blend reaches 30% which leads to increase in CO₂ yield to the maximum value and limits the yield of CO. As follows from Figure 5a and Figure 6a, the yield of CO₂ tends to decrease with the pre-treated wheat straw pellets increase in the blend above 30–45%. It is caused by the heat release during exothermal gasification of the pellets (4) exceeding the amount of heat required to induce endothermic reactions (6, 7). As a result, the rate of reactions (6, 7) tends to increase, increasing the yields of combustible volatiles (CO, H₂) while decreasing the relation between the yields of CO₂ and CO.

The more effective increase in the yield of combustible volatiles (CO, H₂) up to the maximum value with correlating decrease to minimum value the yield of CO₂ is observed by increasing the pre-treatment temperature to T = 548 K (Figure 5a and Figure 6a). It leads to an increase in the reactivity of pre-treated pellets and heat release during gasification of pre-treated pellets (4), thus enhancing the development of reactions (6, 7), responsible for the enhanced yield of combustible volatiles (CO, H₂).

For the blend of raw wood pellets with pre-treated wheat straw pellets replacing a part of raw wood with pre-treated additives of wheat straw pellets results in changes in the heating value of the blend and contents of hemicelluloses, cellulose, and lignin in the blend, increasing content of hemicellulose, responsible for the higher yield of CO₂, while decreasing contents of cellulose and lignin in the blend, responsible for the yields of CO and H₂. Like changes in the yield of CO₂ for the blends of pre-treated wheat straw with raw wheat straw pellets, at low temperature pre-treatment of wheat straw pellets (T = 473 K) the balance between endothermic and exothermic reactions is achieved when the mass fraction of pre-treated straw pellets in the blend with raw wood pellets increases and reaches 30–45%, increasing up to the maximum value the yield of CO₂, while limiting the yields of CO and H₂. The more effective the increase in the yields of combustible volatiles (CO, H₂) up to the maximum value with a correlating decrease to the minimum value of the yield of CO₂ can be achieved by increasing the pre-treatment temperature to T = 548 K (Figure 5 and Figure 6), when increasing reactivity of pre-treated pellets, which enhances gasification of the blend (4) and correlates with enhanced development of endothermic reactions (6, 7). During the thermal decomposition of pre-treated wheat straw blend with raw wood pellets a difference in chemical composition of lignocellulosic wood and wheat straw pellets influences the synergetic effect of interaction on the weight loss and the yields of combustible volatiles. According to [22], the thermo-chemical conversion of biomass decomposition of cellulose leads to a positive synergetic effect, whereas decomposition of hemicellulose and lignin can result in both positive and negative synergistic effects depending on the blend composition and temperature. A similar trend is observed when only partially (15%) adding pre-treated wheat straw to raw wood pellets (Figure 4).

A faster thermal decomposition of the blends (Figure 3c) with a positive synergetic effect (Figure 4) is also observed when mixing pre-treated wheat straw with raw peat pellets. Pre-treated pellets promote the evident changes in chemical composition of the blend, increasing contents of hemicellulose, cellulose, and lignin in the blend compared to raw peat pellets. A low pre-treatment (T = 473 K) temperature of wheat straw enhanced the reactivity of the pellets, promoting faster thermal decomposition of the blend and increasing the heat release during the blend gasification. In addition, increasing the content of hemicelluloses in the blend enhances the yield of CO₂, increasing to a maximum value the ratio of CO₂/CO. Thus, the balance between the reactions (1, 4, 6, 7) at low mass fraction of pre-treated wheat straw (15%) is achieved. Increasing the pre-treatment temperature to 548 K of wheat straw pellets enhanced carbonization and HHV leads to an enhanced gasification of the blend with correlating decrease in the relation CO₂/CO, which suggests the development of endothermic reactions (6, 7).

Faster decomposition of the blends (Figure 4d,e) with a positive synergetic effect is observed also during the thermal decomposition of pre-treated wood with raw wheat straw and peat pellets (Figure 3). Replacing raw straw or peat with pre-treated wood pellets promotes changes in chemical composition of blends and HHV of the blends. Analogous to pre-treated wheat straw blends, the most pronounced increase in the positive synergetic effect is observed for the blend of pre-treated wood pellets with peat. In this case, additives of pre-treated wood pellets promote an increase in hemicelluloses in the blend, thus enhancing volatilization of the blend and development of endothermic processes (6, 7) responsible for the enhanced yields of CO and H₂, and decreasing the relation between the yields of CO₂ and CO (Figure 5d,e and Figure 6d,e).

3.2. Thermochemical Conversion of Microwave Pre-Treated Biomass Blends of Different Origin

MW-induced changes in thermal decomposition/gasification of pre-treated pellets are responsible for thermochemical conversion of fuel blends, including processes of flaming combustion of volatiles, char conversion, and heat energy production (2–5). At the given pilot device configuration (Figure 2) the development of combustion dynamics at biomass thermochemical conversion is influenced by mixing of the axial flow of volatiles with secondary air swirl determining the formation of typical flame temperature and flow velocity profiles close to the inlet of the combustor [36]. Experimental temperature and velocity profiles of raw straw combustion at height L/D = 1.5 are compared with numerically calculated results using 2D axisymmetric model [37] and presented in Figure 8.

In Figure 8a temperature profiles indicate higher flame temperature in the central part of the cylindrical reactor, then decreasing closer to the outer wall. While profiles qualitatively agree, the estimated temperature by numerical model is higher in the central part, but lower in the near-wall region, while the average temperature is similar in both cases. In Figure 8b, tangential velocity profiles are compared. Moreover, in this case qualitative agreement is obtained. Order of magnitude of tangential velocity is c.a. 1 m/s closer to the wall. The biggest differences of experimental and numerical results are found in the central part, where the numerical model predicts the gradual decay of tangential velocity towards the symmetry axis.

In Figure 8c, the largest differences from the numerical to experimental results are observed. It should be noted that in zone r/R = 0.6…0.8 it was not possible to measure negative axial velocities that are caused by recirculation of flow. First, numerical results predict negative axial flow near the wall, but in the experiment a strong positive flow is observed. In both cases, positive velocity is observed in the central part of reactor; however, the numerical model significantly overestimates it. It should be noted that the numerically obtained axial velocity profile corelates with the temperature profile. It suggests that the experimentally observed positive axial velocity in the near wall could contribute to the smaller temperature gradient from the center to near wall region.

The differences of the current numerical model compared to the experimental data in Figure 8 suggests that a higher order reaction model could be necessary to model the process more precisely. Moreover, the influence of the 2D axi-symmetric geometry could contribute to differences of velocity distribution. Due to the transient nature of the process, a time-dependent study should be also considered. Further development of the numerical model is ongoing to obtain better agreement between experimental and numerical results.

The kinetics of the flame temperature as well the total amount of produced heat power (Figure 9a–e) correlate with kinetics of the weight loss rate of blends shown in (Figure 3a–e). The analysis of such a correlation suggests that MW pre-treated additives enhance the axial flow of combustible volatiles, thus promoting faster ignition and transition to self-sustaining combustion conditions. In addition, the radiative heat transfer from the reaction zone, which supports the thermal decomposition of blends, is responsible for the formation of the interrelated processes of thermal decomposition and thermochemical conversion of the blends.

As follows from Figure 9a–e, the kinetics of the heat production is importantly dependent on the chemical and elemental composition of the blend components, the changes in which depend on the MW pre-treatment temperature and mass fraction of pre-treated pellets in the blends. It leads to variation in the rate of the main reactions (1–7) developing during the thermochemical conversion of the blends. Comparing kinetic profiles of the thermal decomposition and heat power during thermochemical conversion of the selectively MW pre-treated blends, one can see that the more pronounced changes in produced heat power with faster and higher heat production can be obtained, when the MW pre-treated wheat straw or wood pellets are blended with raw peat (Figure 9c,e), since additives of pre-treated wheat straw enhance devolatilization of raw peat pellets.

As follows from Figure 5a,b, the average yield of the combustible volatiles importantly depend on blend components, thus producing different flux of combustible volatiles in the inlet of the reactor. At constant primary and secondary air supply rates in the device the changes in the axial flow of combustible volatiles correlate with changes in the air to fuel supply by varying the average value of the air excess ratio at the inlet of the combustor, combustion conditions in the reaction zone, and burnout of volatiles. As follows from Figure 10a–c, the air excess ratio at the inlet of the combustor tends to decrease with the increase in the fraction of pre-treated pellets leading to nearly stoichiometric combustion conditions. As a result, the average values of the flame temperature increase up to the maximum value and tend to decrease with the increase in the mass fraction of pre-treated pellets in the blend and the heat output from the device. In addition, the dependence of the flame temperature and produced heat on the mass fraction of pre-treated pellets confirm the presence of synergistic effects of interaction between the components.

It should be noted that at the fixed distance from the inlet of the reactor (L/D = 3) a higher average value of the flame temperature with a lower amount of produced heat is observed pre-treated wheat straw with raw wheat straw pellets blends (Figure 10b,c), when flaming combustion of volatiles produced during thermal decomposition of hemicelluloses dominates. The average flame temperature tends to decrease with the increase in the average value of produced heat and heat loss from the flame reaction zone.

The lower the value of average temperature, the higher the value of the produced heat observed for blends of pre-treated wood with peat. This suggests that increasing the fraction and MW pre-treatment temperature of pre-treated wood pellets in the blend enhances transition from flaming combustion to char conversion, which can be related to the increase in the fixed carbon content and HHV (Figure 1). As a result, the flame length and measured flame temperature decreases.

The MW-induced changes in the thermochemical conversion of blends influence the total amount of produced heat energy in the device (Figure 11a–f), which is predominately determined by the MW-induced changes in the balance between the reactions (1–7) during thermal decomposition of the blends, changes in the yield of combustible volatiles pellets, and relation between the yields of CO₂ and CO (Figure 5 and Figure 6a–e). As follows from Figure 11a–f, the total amount of produced heat energy predominately tends to increase with the pre-treatment temperature and the mass fraction of MW pre-treated additives in the blends up to 30–45%. This suggests of MW pre-treated additives can be used to complete thermochemical conversion of the blends. A slight decrease in produced heat is observed at low MW pre-treatment temperature (T = 473 K), when the relatively low HHV, reactivity of pre-treated pellets [23], and intensive endothermic drying of raw pellets are limiting factors for the yield of combustible volatiles (Figure 5a–f) and thermochemical conversion.

3.3. The Composition of Emissions of Blends with MW Pre-Treated Biomass Pellets

As discussed above, the yield of combustible volatiles and CO₂ depend on the diversity of carbon contents in pellets and MW-induced changes, varying structure of pellets and their reactivity rate of reactions (1–7). Additionally, the formation of carbon-neutral CO₂ emissions during thermochemical conversion of the blends depends on the inlet conditions of the reactor, the burnout of volatiles (2, 3), and the temperature of the reaction zone. As follows from Figure 12, for all blends, increasing the mass fraction of pre-treated additives in the blend and pre-treatment temperature correlates with an increase in the volume fraction of CO₂ in the products, confirming that MW pre-treated additives enhance thermochemical conversion of the blends. By analogy with the MW-induced changes in the yield of combustible volatiles (Figure 5 and Figure 6), the most effective increase in the CO₂ volume fraction in the products can be obtained for the blend of pre-treated pellets with peat (Figure 12). The less effective increase in CO₂ for the blend of pre-treated straw with wood, the thermochemical conversion of which enhances flaming combustion of volatiles, produced the higher average amount of heat downstream of the combustor (Figure 11).

The diversity of elemental composition of pellets with difference of nitrogen contents in raw pellets (Table 1) and their changes during MW pre-treatment (Figure 1), as well as MW-induced changes in the inlet conditions and the flame temperature, also influence the formation of NO_x emissions, primarily in the form of NO. In the combustion of biomass blends there are two main mechanisms of NO formation—thermal NO and fuel NO. According to the Zeldovich mechanism [38], intensive production of thermal NO_x occurs at temperatures above 1500 K and are more pronounced when the mixture is on the fuel-lean side. At temperatures below 1000 K, the formation of thermal NO is limited and can be neglected. Fuel NO_x results from oxidation of nitrogen contained in biomass [39].

During the combustion of biomass blends, the oxidation of fuel-bound nitrogen is the dominant mechanism of forming NO_x and depends on the nitrogen and oxygen content in the blend components (Figure 13). As follows from the Figure 13a, the blending of pre-treated wheat straw with raw wheat straw or wood pellets promotes an increase in nitrogen content in blends, while blending pre-treated wheat straw with raw peat pellets leads to the nitrogen content decrease. A similar decrease in the nitrogen content in the blend is observed for the blends of pre-treated wood with raw wheat straw of peat pellets. Comparing changes in nitrogen contents in blends with changes in the yields of NO_x suggest (Figure 13b) that the formation of NO_x emissions at the thermochemical conversion of the blends is slightly influenced by the synergistic effects on NO_x formation and the mass fraction of NO_x emissions in the products cannot be approximated with linear dependence on the mass fraction of pre-treated pellets in the blend, assuming the development of interactions, which can influence the NO_x formation and require additional research.

4. Conclusions

From the experimental study of the effects of MW pre-treatment on the thermal decomposition of selectively pre-treated blends, by varying the MW pre-treatment temperature and the mass fraction of pre-treated pellets in blends, the following conclusions can be made:

The MW pre-treatment of wheat straw and wood pellets causes changes in the structure and reactivity, elemental, and chemical composition of pellets, that depend on the pre-treatment temperature. Measurements of the weight loss rate indicate the presence of synergistic effects between pre-treated and raw pellets in blends up to 25–30% (Figure 3 and Figure 4). It is confirmed that increased reactivity and volatilization in pre-treated pellets enhances the thermal decomposition of raw pellets by varying the yields of combustible volatiles (Figure 5). The most pronounced positive synergistic effect on the weight loss rate is observed for blends of pre-treated wheat straw or wood pellets with raw peat, where enhanced yield of combustible volatiles from pre-treated additives promote faster and enhanced thermal decomposition of raw peat. The least synergetic effect is observed when blending pre-treated wheat straw with wood pellets.

The formation of positive synergetic effects that influence flame temperature, heat output from the device, and produced heat energy per mass of burned pellets is observed especially at higher MW pre-treatment temperature of lignocellulosic pellets (T = 548 K). The most pronounced increase in the flame temperature (4.5–7.5%) is observed for blends of pre-treated and raw wheat straw pellets, when increasing the mass fraction of pre-treated pellets in the blend in a range from 30% to 60% (Figure 10).

As a result of the synergistic effects of MW pre-treatment on the weight-loss rate and the yield of combustible volatiles, the most pronounced increase in the heat output from the device and produced heat energy is observed for the blends of pre-treated wheat straw or wood pellets with raw peat (Figure 10 and Figure 11). For blends of pre-treated wood with raw peat, the maximum value of synergistic effect on the heat output is observed with a relatively low mass fraction (15%) of pre-treated wood in the blend. Increasing the MW pre-treatment temperature from 473 K to 548 K, a synergistic effect on the heat output is increased from 9.4% to 11.9%.

There is an increase in the volume fraction of carbon-neutral CO₂ emissions in the products when pre-treated pellets of wheat straw or wood are blended with raw pellets (Figure 12). The most significant increase in the carbon-neutral CO₂ emissions can be achieved for the blends of pre-treated wood with peat.

The positive synergistic effects of the interaction between the components also influence the formation of polluting NO_x emissions. Formation of these emissions can be predominantly related to the mechanism of fuel NO_x formation, which depends on the nitrogen and oxygen contents in pre-treated and raw pellets. Blending pre-treated wheat straw pellets that have higher nitrogen content with raw wood pellets promotes enhanced formation of NO_x emissions. However, formation of NO_x emissions is decreased in the cofiring of pre-treated wheat straw or wood with raw peat (Figure 13).

Considering the effect of MW pre-treatment of biomass pellets on the thermochemical conversion of different origin biomass blends it can be concluded that additives of MW pre-treated biomass pellets in blends with raw pellets can be used to control and improve the thermal decomposition of blends as well as the formation, ignition, and combustion of volatiles and composition of emissions, thus providing wider and sustainable use of different origin regional biomass resources.