Introduction

One of the largest contributors amongst greenhouse gas emissions is CO2, and the major source of the CO2 is the combustion of fossil fuels to supply energy. Carbon capture and storage technologies rely on the idea of CO2 sequestration from fossil fuel-based power plants and transportation of this captured CO2 to the storage sites. Amongst carbon capture and storage technologies, oxy-fuel combustion is one of the proven technologies which have less chemically complicated treatments. This technique is performed by using mixture of pure oxygen and a part of recycled flue gas instead of air as a combustion oxidiser. Nitrogen-free and oxygen-enriched atmosphere of oxy-fuel combustion increases the content of CO2 in the flue gas comparing the conventional combustion. For this reason, CO2 capturing and sequestration processes are easier and cost effective with the oxy-fuel combustion. The detailed description of oxy-fuel combustion technology can be found elsewhere [1, 2] and also in the previous research of oxy-combustion of Estonian oil shale [3].

Estonian energy sector is dominated by oil shale (OS) [4]. OS is a fine-grade sedimentary rock containing relatively large amounts of combustible organic matter (kerogen), which is well distributed in its mineral matrix [5]. Estonian oil shale is defined as highly heterogeneous fuel with complicated composition of organic and mineral matter. Its dry matter mainly consists of organic part, sandy-clay (terrigenous) minerals and carbonates [6]. During thermal decomposition of OS, a complex set of parallel reactions occurs in both organic and mineral parts of the fuel. Besides, mass and heat transfer in oxy-fuel combustion differs significantly from air firing because of the high CO2 partial pressure. The knowledge of the effect of CO2 atmosphere on the reactivity of OS and OS char is needed for evaluating application of oxy-combustion to OS. The study of reaction kinetics including pyrolysis and oxidation characteristics is crucial to understand the complex mechanism of OS oxy-fuel combustion.

There are a few studies on pyrolysis and kinetics of oxidation of OS and its char. Pyrolysis of OS using thermogravimetric analysis (TG) has been discussed in the literature from different regions: Colorado (Green River) [7], China [8], Jordan [9], Morocco [10], Turkey [11], etc. In general, kinetic calculations are based on non-isothermal and isothermal thermal analysis methods. Pyrolysis is regarded as occurring in two stages. In the first stage, OS is heated to form pyrobitumen; in the second stage, pyrobitumen decomposes to form shale oil, shale gas and shale char. Researchers have established that pyrobitumen is formed as an intermediate product when OS is heated [12]. According to the studies mentioned above, it was shown that formation of bitumen has specific temperature range depending on the OS. It was also reported that the heating rate has an important effect on pyrolysis. Effect of pyrolysis environment was also reported in several studies. Jaber and Probert [13] showed that CO2 as carrier gas led to a slightly greater mass loss during the devolatilisation stage of two Jordanian OS samples than N2. It was also reported that the carbon-dioxide/char reaction is more likely to occur at high temperatures. Fang-Fang et al. [14] studied Chinese Huadian OS under N2 and CO2 atmospheres. Much greater mass loss was reported at temperatures above 760 °C under CO2 due to the residual carbon reaction with CO2 forming CO. Residual carbon preparation in different heating rates by TG and its reactivity under isothermal conditions was carried out by Barkia et al. [15] with OS from Morocco. It was reported that the heating rate has an important effect on the amount of residual carbon obtained. Regarding OS residual carbon oxidation, there are a very limited number of studies in the literature. In [16], it was reported that the presence of minerals in OS can drastically alter the reactivity of residual char. It was also reported that removal of mineral matter with acid washing decreases the reactivity of residual carbon and the reactivity of residual carbon depends on the heating rate used for pyrolysis [17–19]. The effect of partial pressure of CO2 had very small impact on the reactivity of residual carbon [18]. However, the residual carbon prepared under fast heating rates was more reactive [17–19]. The effect of prior pyrolysis and heating rate on the combustion of OS semi-cokes was studied by Miao et al. [20]. The Coats-Redfern [21] method was employed to find the combustion-reaction kinetic parameters for OS semi-coke. It was reported that the activation energy and the frequency factor were much larger at high temperatures than they are at low temperatures.

So, TG has been widely used to study pyrolysis and oxidation kinetics of different OSs and their chars. However, there is very limited research conducted with Estonian oil shale. The knowledge about reaction kinetics including pyrolysis and oxidation characteristics of OS char is inadequate, and there is almost no research related to oxy-fuel conditions. So, more information is needed to understand the complex mechanism of Estonian oil shale oxy-fuel combustion.

In this study, non-isothermal and isothermal thermal analysis (TA) methods combined with FTIR spectroscopy have been applied to investigate pyrolysis of Estonian oil shale and combustion characteristics of its char. Kinetics of OS char oxidation has been analysed in different combustion atmospheres, and conversion-dependent apparent activation energies have been presented. The kinetic data reported here together with our previous research [3] are important for the future modelling of the complete conversion process of Estonian oil shale under oxy-fuel conditions.

Experimental

Materials

The OS sample used in the experiments is a common energetic OS from Estonia, and it was obtained from Narva Power Stations. The obtained OS sample was crushed with an alligator-type grinding machine. Mean sample was taken from the crushed material and ground in a big ball mill. Then, the sample was dried at 105 °C for several hours and ground additionally in a Retsch PM 100 grinding machine until the entire sample passed the 200-µm sieve. Characterisation of the OS sample is given in Table 1.

Table 1 Characterisation of the OS
Full size table

Methods

The Setaram Setsys Evo 1750 thermoanalyser coupled to a Nicolet 380 FTIR spectrometer was used in the experiments. Non-isothermal TG tests were performed at 10 K min−1 heating rate up to 1,000 °C with gas flow rate 30 mL min−1 using different atmospheres (Ar, CO2, Ar/O2 and CO2/O2). Standard 100-µL Pt crucibles were used, and the mass of samples was 20 ± 0.5 mg in most experiments and 30 ± 0.5 mg for the tests with gas analysis. To prepare residual carbon of OS, the sample was heated up to 450 °C in argon atmosphere. When the temperature was reached, the sample was maintained at 450 °C for 1 h. The temperature of the highest reaction rate from preliminary pyrolysis experiments was selected as 450 °C, and approximately in 1 h, the mass loss due to devolatilisation ceased.

Isothermal oxidation of the obtained residual carbon was performed at three different temperatures (450, 500 and 550 °C). For this, the samples were heated up to the selected temperature in argon, and after reaching to the desired temperature, Ar/21 % O2 mixture, CO2/30 % O2 or CO2/21 % O2 was introduced into the system.

5, 10 and 15 K min−1 heating rates were applied, and the AKTS Advanced Thermokinetics software [22] was used to calculate kinetic parameters of oxidation stage of OS char. For these experiments, the OS char was prepared in electric tube furnace under the same conditions to obtain bigger amount of char having uniform properties. FTIR measurements were recorded in the 400–4,000 cm−1 region with the resolution of 4 cm−1 taking an average of four scans.

Results and discussion

Pyrolysis in Ar and CO2

The main stages of combustion of solid fuel particles are drying, pyrolysis (devolatilisation) and char combustion. Pyrolysis of OS in Ar and CO2 and the OS char combustion stage were studied comparatively in Ar/O2 or CO2/O2 atmospheres, respectively, using non-isothermal and isothermal methods.

The pyrolysis behaviour (characteristic temperatures and mass loss) of the OS sample in Ar and CO2 was very similar up to 500 °C, which indicates that CO2 behaves like an inert gas and has the same influence as Ar on the organic part until the end of devolatilisation stage. The same nature of TG curves and the very close peak temperatures in Ar and CO2 atmospheres can be seen from Fig. 1a, b.

Fig. 1
figure 1

Mass loss (TG) and differential mass loss (DTG) curves of initial OS sample in pure CO2 and Ar

Full size image

The pyrolysis peak rate occurs at around 450 °C in both Ar and CO2 atmospheres. The pyrolysis reactivity is slightly higher in Ar atmosphere. This small difference can be explained by differences in heat capacity, diffusivities and radiative properties of the gas atmosphere. The effect of char gasification reactionFootnote 1

$$ {\text{C}} + {\text{CO}}_{ 2} \to 2 {\text{CO}}\quad\Delta {\text{H}}_{\text{r}}^{0} = + 1 7 2. 4 {\text{ kJ}} $$
(1)

is not visible in this temperature range, and gas analysis has also shown that the role of gasification reaction increases at higher temperatures—above 650 °C. There is no significant mass loss difference until the end of devolatilisation of organic part; furthermore, TG curves of OS sample are almost superposing in Ar and CO2 atmospheres. According to the steps of TG curves, two stages can be differentiated in general: pyrolysis in low temperature zone (300–520 °C) for both CO2 and Ar atmospheres and decomposition of mineral carbonates (magnesite, dolomite, calcite, siderite) at higher temperatures (above 650 °C). Decomposition of MgCO3 and CaCO3 is shifted apart in CO2 and proceeds with maximum rate at 814 and 920 °C (Fig. 1b), respectively.

Almost 27 % of the initial sample mass has been lost during thermal decomposition of organic part between 308 and 516 °C in Ar and between 328 and 513 °C in CO2 atmospheres. Following the decomposition of organic matter, mineral part decomposition occurs in one step and mass loss reaches almost 57 % level giving peak temperature at 771 °C for Ar atmosphere. For the decomposition of mineral part in CO2, there are two distinguishable mass loss steps. First step lies on a wide temperature region (650–890 °C) and results in almost the same amount of mass loss (30 %) like in Ar atmosphere. The final step has about 3 % additional mass loss in the temperature region of 889–956 °C.

The mass loss difference and slightly increased reaction rate in Ar in between 550 and 650 °C can be related to the processes in mineral part of OS. The effect of residual carbon and CO2 reaction in CO2 atmosphere (Eq. 1) can be assumed from the very small difference in the final mass loss and also from evolved gas analysis tests, which have shown increasing release of CO above 650 °C. However, the combination of different possible parallel reactions at higher temperatures is complex for such clear identification. Besides, the presence of carbonates can support CO2-related self-gasification of residual carbon in fuel particles at their decomposition temperatures even under 100 % Ar atmosphere. In addition, higher partial pressure of CO2 brings along more intensive decomposition at higher temperatures and can affect the pore structure, enhancing in-particle diffusion.

Char combustion in Ar/O2 and CO2/O2

Thermal behaviour of OS char was studied in CO2/O2 and Ar/O2 atmospheres using both non-isothermal and isothermal analysis methods.

It can be seen from non-isothermal data (Fig. 2) that higher O2 content shifts the characteristic temperatures of oxidation of OS char slightly towards lower values. There are no big differences in the total mass loss of the samples in three different atmospheres. Solely, CO2/21 % O2 has slightly higher mass loss as compared to CO2/30 % O2 and Ar/21 % O2. The presence of CO2 can affect the equilibrium in carbonates/sulphates system and have some small effect also on the reactions with kerogen and impurities. Reactions with pyrite, its intermediates and pyritic or sulphide sulphur can be associated with the peaks at around 500 °C.

Fig. 2
figure 2

TG, DTA (a) and DTG (b) curves of OS char in Ar/O2 and CO2/O2 mixtures

Full size image

Slow reaction rate in the beginning of process during isothermal heating (Fig. 3a, b) can be explained by continuing diffusion of heavier pyrolysis products and slow oxidation of char. High CO2 partial pressure in the char particles and possible blockage of pores by pyrolysis products restrict diffusion of oxygen into the particle. After the char carbon is consumed, oxygen concentration in the particle is increased and reactions with pyrite and related intermediates increase the reaction rate. Char carbon can be oxidised also by CO2 (Eq. 1). As the process is endothermic, char particle temperature can decrease. However, the results of gas analysis from non-isothermal experiments with OS have shown that release of CO in 100 % CO2 atmosphere increases notably at temperatures above 650 °C (Fig. 4a). So, the char gasification reaction does not have significant role at 450–550 °C in the case of OS.

Fig. 3
figure 3

TG and DTG curves of isothermal oxidation of OS char in 79 %Ar/21 % O2 (a) and 70 % CO2/30 % O2 (b) at different temperatures

Full size image
Fig. 4
figure 4

CO emission profile of OS in CO2 (a) and SO2 emission profiles (b–d) of OS in CO2/O2 mixture (b), OS char in CO2/O2 mixture (c) and OS in CO2 (d)

Full size image

In the reactions with pyrite, in the given temperature range, the first reaction is the formation of iron sulphide proceeds partly already during the pyrolysis at 450 °C, and then, oxidation of iron sulphide and pyrite takes place in the gas mixtures with oxygen. Thermodynamic calculations showed also relatively high probability of formation of iron sulphate in the excess of oxygen in this system.Footnote 2

$$ {\text{FeS}}_{ 2} \to {\text{FeS}} + {\text{S}}_{\text{pyritic}} \quad{+}7 8. 2 {\text{ kJ}} $$
(2)
$$ {\text{FeS}}_{ 2} + 3 {\text{O}}_{ 2} \to {\text{FeSO}}_{ 4} + {\text{SO}}_{ 2} \quad{-} 10 4 7. 4 {\text{ kJ}} $$
(3)
$$ {\text{FeS}} + 2 {\text{O}}_{ 2} \to {\text{FeSO}}_{ 4} \quad{-} 8 2 8. 9 {\text{ kJ}} $$
(4)

Formation of iron oxides (mainly Fe2O3, but also Fe3O4 and FeO) is also possible in the following reactions:

$$ 2 {\text{FeS}}_{ 2} + 5. 5 {\text{O}}_{ 2} \to {\text{Fe}}_{ 2} {\text{O}}_{ 3} + 4 {\text{SO}}_{ 2} \quad{-} 1 6 5 7.0{\text{ kJ}} $$
(5)
$$ 2 {\text{FeS}} + 3. 5 {\text{O}}_{ 2} \to {\text{Fe}}_{ 2} {\text{O}}_{ 3} + 2 {\text{SO}}_{ 2} \quad{-} 1 2 1 9. 8 {\text{ kJ}} $$
(6)
$$ 2 {\text{FeSO}}_{ 4} + {\text{CO}} \to {\text{Fe}}_{ 2} {\text{O}}_{ 3} + 2 {\text{SO}}_{ 2} + {\text{CO}}_{ 2} \quad{+} 1 5 4. 9 {\text{ kJ}} $$
(7)

The last reaction becomes thermodynamically possible at quite low temperatures already, and it is affected by CO2 partial pressure. These reactions are related to the release of SO2, and gas analysis has shown intensive emission of SO2 at these temperatures (Fig. 4). The first peak in emission profile (Fig. 4b) can be related to the oxidation of organic sulphur and the second to pyritic sulphur as it is repeated also on the emission profile of OS char (Fig. 4c) and corresponding to the peaks at around 500 °C in Fig. 2.

Siderite decomposes to give iron oxides, or in the presence of SO2 also FeSO4

$$ {\text{FeCO}}_{ 3} \to {\text{FeO}} + {\text{CO}}_{ 2} \quad{+} 7 4. 9 {\text{ kJ}} $$
(8)
$$ {\text{FeO}} + {\text{SO}}_{ 2} + 1/ 2 {\text{O}}_{ 2} \to {\text{FeSO}}_{ 4} \quad{-} 3 5 9. 9 {\text{ kJ}} $$
(9)

Different forms of sulphur left in the particle after pyrolysis stage and decomposition of pyrite (organic, sulphide, pyritic) can oxidise to give SO2. Besides, gas analysis of the products has shown the presence of traces of COS in several spectra [3, 24].

$$ {\text{S}} + {\text{O}}_{ 2} \to {\text{SO}}_{ 2} \quad{-}2 9 6. 8 {\text{ kJ}} $$
(10)
$$ {\text{S}} + {\text{CO}} \to {\text{COS}} \quad{-}3 1. 5 {\text{ kJ}} $$
(11)

As in the Ar/O2, mass losses are visibly different at different temperatures (Fig. 3a), and it can be assumed that half-decomposition of dolomite (part of MgCO3) has also a definite role in the process of char oxidation.

$$ {\text{MgCO}}_{ 3} \cdot {\text{CaCO}}_{ 3} \left( {\text{dolomite}} \right) \to {\text{MgO}} + {\text{CaCO}}_{ 3} + {\text{CO}}_{ 2} \quad{+} 1 2 4. 6 {\text{ kJ}} $$
(12)

In CO2/O2 at high partial pressure of CO2, the role of this reaction is minor and the mass loss at temperatures from 450 to 550 °C is almost the same (Fig. 3b). From non-isothermal data (Fig. 2), it can be seen that in CO2/O2, the maximum rate of MgCO3 decomposition is shifted by 30–35 °C towards higher temperatures as compared to Ar/O2 atmosphere.

Decomposition of CaCO3 takes place at higher temperatures and is notably dependent on CO2 partial pressure.

$$ {\text{CaCO}}_{ 3} \to {\text{CaO}} + {\text{CO}}_{ 2} \quad{+} 1 7 8. 2 {\text{ kJ}} $$
(13)

In Ar/O2 decomposition of dolomite, calcite and magnesite proceed in one step with maximum rate at about 782 °C (offset of DTG peak at 640 °C), but in CO2/O2 decomposition of CaCO3 takes place with maximum rate at 912–918 °C (Fig. 2).

During pyrolysis stage and at low partial pressures of O2, CaS can form

$$ 4 {\text{CaSO}}_{ 3} \to {\text{CaS}} + 3 {\text{CaSO}}_{ 4} \quad{-} 1 4 7. 2 {\text{ kJ}} $$
(14)
$$ {\text{CaSO}}_{ 4} + 4 {\text{CO}} \to {\text{CaS}} + 4 {\text{CO}}_{ 2} \quad{-} 1 80. 2 {\text{ kJ}} $$
(15)
$$ 2 {\text{CaO}} + 3 {\text{S}} \to 2 {\text{CaS}} + {\text{SO}}_{ 2} \quad{+} 8. 2 {\text{ kJ}} $$
(16)

As the partial pressure of oxygen increases, CaS is oxidised

$$ {\text{CaS}} + 2 {\text{O}}_{ 2}\, \to {\text{CaSO}}_{ 4} \quad{-} 9 5 1. 7 {\text{ kJ}} $$
(17)

SO2 formed in several reactions is bound by CaO (and partly also by MgO) or by direct sulphation of calcite.

$$ {\text{CaO}} + {\text{SO}}_{ 2} + 1/ 2 {\text{O}}_{ 2} \,\to {\text{CaSO}}_{ 4}\quad {-} 50 2. 4 {\text{ kJ}} $$
(18)
$$ {\text{CaO}} + {\text{SO}}_{ 2} \to {\text{CaSO}}_{ 3} \quad{-} 2 2 7. 7 {\text{ kJ}} $$
(19)
$$ 4 {\text{CaO}} + 4 {\text{SO}}_{ 2}\, \to {\text{CaS}}\, + 3 {\text{CaSO}}_{ 4} \quad{-} 10 5 7. 8 {\text{ kJ}} $$
(20)
$$ {\text{CaCO}}_{ 3} + {\text{SO}}_{ 2} + 1/ 2 {\text{O}}_{ 2}\, \to {\text{CaSO}}_{ 4} + {\text{CO}}_{ 2} \quad{-}3 2 4. 2 {\text{ kJ}} $$
(21)

However, the last reaction is enhanced at higher pressures [25]. Formation of mixed Ca, Mg sulphate is also possible [26].Footnote 3

$$ {\text{CaO}} + 3 {\text{MgO}} + 4 {\text{SO}}_{ 2} + 2 {\text{O}}_{ 2} \to {\text{CaMg}}_{ 3} \left( {{\text{SO}}_{ 4} } \right)_{ 4} $$
(22)

Part of free CaO reacts with SiO2 (and other sandy-clay components like Al2O3) forming different silicates (belite, CaO·Al2O3, etc.) [27]

$$ 2 {\text{CaO}} + {\text{SiO}}_{ 2} \to 2 {\text{CaO}} \cdot {\text{SiO}}_{ 2} (\beta - {\text{Ca}}_{ 2} {\text{SiO}}_{ 4} )\quad{-} 1 2 5. 4 {\text{ kJ}} $$
(23)

In the presence of CO, decomposition of sulphates can proceed well below 1,000 °C giving rise to additional emissions of SO2

$$ 4 {\text{FeSO}}_{ 4} \to 2 {\text{Fe}}_{ 2} {\text{O}}_{ 3} + 4 {\text{SO}}_{ 2} + {\text{O}}_{ 2} \quad{+} 8 7 5. 7 {\text{ kJ}} $$
(24)
$$ {\text{CaSO}}_{ 4} + {\text{CO}} \to {\text{CaO}} + {\text{SO}}_{ 2} + {\text{CO}}_{ 2} \quad{+} 2 1 9. 4 {\text{ kJ}} $$
(25)

The increased intensity in SO2 emission profile of OS in 100 % CO2 at temperatures above 900 °C can be related to these reactions (Fig. 4d).

So, the process of char oxidation followed by mineral decomposition involves a complex set of competitive endo- and exothermic reactions depending on temperature, oxidative/reductive properties of the gas phase and CO2 partial pressure inside the char particles resulting in the specific shape of DTG curves (Fig. 3a, b) as well as of SO2 emission profiles (Fig. 4).

Oxidation kinetics

The changes in activation energy (E/kJ mol−1) depending on the reaction progress α were determined from the experiments with 5, 10 and 15 K min−1 heating rates. Conversion α was calculated from the mass loss during oxidation stage. After the baseline correction, derivatives of the normalised TG signals were processed with the AKTS Advanced Thermokinetics software [22] enabling to obtain apparent conversion-dependent activation energies without assuming the form of f(α) function.

The equation describing the iso-conversional approach [28, 29] used in the software derived from the Friedman differential method [30] can be expressed as follows:

$$ \ln \;\left( {\frac{{\text{d}}\alpha }{{\text{d}}t}} \right) = \ln \;\left\{ {A\left( \alpha \right)\;f\left( \alpha \right)} \right\}\; - \;\frac{E\left( \alpha \right)}{RT} $$
(26)

The function dependent on the reaction model f(α) becomes a constant at each fixed conversion degree α in Eq. 26, and the relationship between the logarithm of the reaction rate dα/dt and 1/T is linear with the slope of E/R.

The relationship between the conversion-dependent activation energy and reaction progress for OS char in O2/Ar and O2/CO2 mixtures is shown in Fig. 5.

Fig. 5
figure 5

Activation energy (E, kJ mol−1) and pre-exponential factor ln {A(α)*f(α)/s−1} values for OS char oxidation in Ar/O2 and CO2/O2 atmospheres

Full size image

Kinetic calculations showed that the conversion-dependent activation energy of OS char was 110–200 kJ mol−1 in Ar/21 % O2, 90–125 kJ mol−1 in CO2/30 % O2 and 125–160 kJ mol−1 in CO2/21 % O2 in the range of α = 0.1–0.9. In comparison, activation energies for OS itself in these atmospheres were in the range of 120–100 kJ mol−1 in Ar/O2 and 35–45 kJ mol−1 in CO2/O2 in the range of a = 0.1–0.9, being higher at the beginning, then diminishing and increasing slightly for the final stage of oxidation. Kinetics of OS oxidation was discussed more in detail in [3]. The activation energies for the oxidation of OS residual carbon are lower at the beginning of oxidation, indicating more likely to diffusion limitation. Thereafter, there is an increase almost until 50 % of conversion. Following to this, a slight decrease and increase for the final stage can be seen. The initial and final stage activation energies were lower in CO2/30 % O2 as compared to Ar/O2. For the final stage, oxidation in Ar/21 % O2 shows higher activation energies than CO2/21 % O2 that can be related to the increased role of reactions in mineral part with the increase in temperature.

Simple kinetic modelling of char oxidation was carried out also on the basis of isothermal data. Several mechanism models, well organised lately in [31], were tested to find the best fit into

$$ g(\alpha ) = k\tau $$
(27)

where g(α) is the integral form of mechanism function, k—rate constant and τ—reaction duration. Conversion α was calculated as mass loss from the maximum mass loss at the respective temperature.

It was found that the highest regression coefficients (R 2 = 0.993–0.932) amongst all the models tested for Eq. 27 belonged to Avrami–Erofeev equation

$$ {\mathbf{A2}}:g(\alpha ) = [{-}{ \ln }( 1{-}\alpha )]^{ 1/ 2} $$
(28)

The calculated values of the rate constant at different temperatures for the model A2 are presented in Table 2. It can be seen that the rate constants for char oxidation in CO2/O2 are approximately 1.2–1.3 times higher as compared to Ar/O2 atmosphere. So, OS char combustion under oxy-fuel conditions should proceed with higher rate as compared to air combustion.

Table 2 Rate constant values for char oxidation in CO2/O2 and Ar/O2
Full size table

However, apparent activation energy could not be calculated in the temperature region studied due to non-Arrhenius behaviour (R 2 < 0.7) as can be assumed also from the shape of DTG curves (Figs. 3, 4).

Conclusions

Isothermal and non-isothermal TA methods were used to characterise OS pyrolysis and OS char oxidation in different atmospheres related to OS oxy-fuel combustion. The following conclusions can be drawn by analysing the experiment data and emission profiles of gaseous products from OS and OS char:

  • At OS pyrolysis, the influence of CO2 on the organic part during devolatilisation stage is highly similar to Ar. mass loss and characteristic peak temperatures do not differ in Ar or CO2. However, decomposition of carbonates in the mineral part of OS is notably influenced by CO2 partial pressure.

  • There is no notable effect of char carbon and CO2 reaction on the char oxidation stage. The results of gas analysis have shown that the release of CO from OS in 100 % CO2 atmosphere increases notably not until temperatures above 650–700 °C. Besides this, the combination of different possible reactions in this temperature region makes it complex to identify the role of char carbon and CO2 reaction.

  • The presence of inorganic minerals, such as pyrite, sulphides and sulphates, increases SO2 emissions, especially, at higher temperatures and low O2 concentrations. The presence of carbonates can support CO2-related self-gasification of residual carbon in fuel particles and increase CO formation at their decomposition temperatures, but high partial pressure of CO2 shifts the carbonate decomposition towards higher temperatures enabling to assume decreased CO2 emissions under oxy-fuel conditions.

  • The values of rate constants calculated for different temperatures from 450 to 550 °C indicate that OS char combustion should proceed under oxy-fuel conditions with higher rate as compared to air combustion.

  • Oxidation of OS char takes place with higher activation energies as compared to the initial OS sample. Char oxidation under oxy-fuel conditions (CO2/O2) occurs with lower activation energies as compared to Ar/O2, especially, at higher conversion levels. So, applying oxy-fuel combustion to OS should enhance the process as compared to air combustion.