Kinetic Analysis of Carpathian Source Rock Pyrolysis Under Dynamic Conditions

Abstract

The research presented in the article was undertaken in order to better investigate the generation potential of the Oligocene Menilite Formation due to its importance as source rocks within the Outer Carpathian Basin. The non-isothermal decomposition of the selected Carpathian source rock was studied to determine the kinetic parameters of the pyrolysis process. The kinetic parameters of bulk rock and separated kerogen were determined using the model-free Kissinger, Kissinger–Akahira–Sunose (KAS), and Friedman methods. The pyrolysis process exhibits a complex reaction mechanism. The obtained apparent activation energy (Ea) and pre-exponential factor (A) values depend on the extent of conversion, suggesting that the process involves multiple reaction steps. This dependence is very similar when calculated using both isoconversional methods, Friedman and KAS; however, the calculated values of the kinetic parameters differ depending on the method used. It was found that the activation energy of kerogen is lower than that of bulk rock, and the reaction maximum was shifted to higher temperatures. This shift is attributed to the presence of clay minerals in the rock. The values of average activation energy and the pre-exponential factor found in this study are relatively high, possibly due to the nature of the short-chain organic matter contained in the source rock.

1. Introduction

The oil deposits in the Carpathians and the foreland have a long history, as this is the region of the oldest world oil mining. Currently, oil exploitation in the Carpathians is carried out on about 20 deposits from boreholes, the depth of which reaches 3600 m. Annual production is about 33 thousand tons. The current production of hydrocarbons in the Carpathians is lower than it was 100 years ago, and many fields are already exhausted. However, there are prospects for discovering new accumulations of oil and gas thanks to continuously advancing exploration technologies. High hopes are associated with the Oligocene Menilite Formation, although it is characterized by diverse thermal maturity in different locations and tectonic units. This variability may result from differences in sedimentary environments that existed in the Oligocene within the Outer Carpathian Basin [1,2].

The objective of petroleum source rock evaluation is to determine the petroleum generation potential of the rock and its relationship to neighboring formations. Oil and gas source rocks contain highly transformed organic matter of high molecular weight, known as kerogen. The structure of kerogen determines the kinetic parameters of its decomposition reaction, which are essential for modeling hydrocarbon generation processes [3].

The determination of reliable kerogen kinetic parameters—activation energy (Ea) and the Arrhenius constant (A)—is one of the fundamental tasks of organic geochemistry. These parameters reflect the structure of organic matter as well as the synsedimentary and postdiagenetic processes it has undergone [4,5]. Isothermal and non-isothermal pyrolysis methods are widely used to determine parameters describing the kinetic decomposition of kerogen, contributing to a better understanding of organic matter transformation processes responsible for the formation of hydrocarbons.

Source rocks are primarily fine-grained sedimentary rocks (siltstones, mudstones, and shales) containing variable amounts of transformed organic matter, along with minerals such as quartz, clay minerals, and carbonates, as well as feldspars, mica, and sulfides (mainly pyrite). The mineral composition of source rocks is an important parameter due to its influence on interactions between minerals and reservoir fluids, as well as on the adsorption and desorption processes of kerogen. Inorganic components affect kerogen conversion, leading to the release of gaseous products [6].

Karabakan and Yürüm [7] state that silicates have an inhibitory effect, whereas carbonates catalyze the pyrolysis reaction in rock samples. Additionally, mineralogical data are used in the initial stages of the recovery process, geological modeling of the deposit, and efforts to accelerate hydrocarbon recovery [8].

Thermal methods, including thermogravimetry (TG) and differential scanning calorimetry (DSC), have significant analytical potential, providing information on the mineral composition of rocks (by detecting thermoreactive components) and the organic matter present in samples. These methods are widely used to examine organic matter in oil shale and to determine kerogen in source rocks [9,10,11,12,13].

In most cases, kerogen is extracted from rock samples before analysis [14], but some studies have also attempted to analyze bulk rocks directly [15,16]. In the case of shale rocks, the vast majority of research has focused on oil shale [14,15,17,18,19,20], while relatively little research has been devoted to gas shale and source rocks [5,21].

The experimentally determined values of the kinetic parameters of a solid-state process must be interpreted as effective values, as standard experimental techniques (TG, DTA, and DSC), as well as more sophisticated methods, generally do not allow for the isolation of elementary reactions [22]. Instead, they provide a global measure of the rate or extent of a process, which typically involves multiple steps with different activation energies.

Thermal analysis is widely applied for the evaluation of kinetic parameters in the pyrolysis process; however, extrapolation to larger-scale devices cannot be performed perfectly [15]. On the other hand, it is highly useful from a fundamental perspective and for comparing different samples. It has been reported that non-isothermal methods are generally more accurate than isothermal ones. Many authors have observed that thermogravimetric studies conducted under isothermal conditions, where reaction rates are determined at constant temperatures, involve certain inaccuracies [19,20]. The non-isothermal method is considered more accurate, mainly due to its shorter experimental duration and the fewer difficulties encountered (e.g., the initial heating-up period in isothermal methods) [19].

The main objective of this study was to compare the selected non-isothermal methods of kinetic analysis applied to the decomposition of source rock and kerogen. The obtained kinetic parameters of Menilite source rock could be useful for modeling the conversion of source rock organic matter into fossil fuels and for predicting its maturity state.

2. Experimental

2.1. Materials and Methods

A sample of shale from the Oligocene Menilite Formation was taken from the outcrop near Gorlice (Silesian Unit of Flysh Carpathians) (Figure 1). Kerogen was separated from the source rock with the use of the conventional method of HCl&HF treatment. The Rock-Eval data for the source rock and kerogen sample are provided in

Table 1. Rock-Eval pyrolysis results for analyzed samples.

Samples	Tmax	S1	S2	S3	PI	PC	RC	TOC	HI	OI	MINC
Bulk rock	407	1.65	55.45	2.14	0.03	4.86	8.31	13.17	421	16	0.25
Kerogen	401	5.39	261.76	12.77	0.02	23.88	31.56	55.44	472	23	1.67

T_max—maximal hydrocarbon generation temperature [°C]; S₁—free hydrocarbons below 300 °C [mg HC/g rock]; S₂—hydrocarbons generated at 300–650 °C [mg HC/g rock]; S₃—the quantity of CO₂ produced during pyrolysis [mg CO₂/g rock]; PI = S₁/(S₁ + S₂)—generation index; PC—pyrolytic carbon content [wt.%]; RC—residual carbon content [wt.%]; TOC—total organic carbon [wt.%]; HI—Hydrogen Index [mg HC/g TOC]; OI—Oxygen Index [mg CO₂/g TOC]; MINC—Mineral Inorganic Carbon [wt.%].

.

Before the TG experiment, the rock sample was ground and sieved to a size of less than 0.08 mm to eliminate the effects of heat and mass transfer in rock particles on the kinetic parameters. The pyrolysis experiments on the source rock samples and separated kerogen samples were performed using the NETZSCH simultaneous thermal analysis STA 449 F3 Jupiter^® equipment, produced by NETZSCH-Gerätebau GmbH, Selb, Germany.

Before the analysis, the instrument was calibrated using indium as a reference temperature standard. A baseline correction for the buoyancy effect on the TG curve was applied. Prior to the measurements, the system was vacuumized twice to eliminate oxygen. The samples were placed in an alumina (Al₂O₃) crucible, with approximately 20 mg used for bulk rock and about 10 mg for kerogen.

All the experiments were performed under an inert nitrogen atmosphere with a gas flow rate of 50 mL min⁻¹. The tests were conducted under non-isothermal conditions over a temperature range of 40–600 °C. The applied heating rates were 2, 5, 10, 15, and 20 °C min⁻¹. The NETZSCH Kinetics Neo^® software (version 2.5.3) was used to analyze the thermo-chemical process data.

2.2. Kinetic Calculations

The kinetic data can be calculated by different methods, including model-free and model-based methods. According to reaction rate theory [22], the kinetic equation of the decomposition of solid matter can be written as follows:

$${\displaystyle \frac{d\alpha }{dt}}=k{\left(1-\alpha \right)}^n$$

(1)

where k is the reaction rate constant, given by the Arrhenius equation: k = $\mathrm{A}·{\mathrm{e}}^{-{\displaystyle \frac{\mathrm{E}}{\mathrm{RT}}}}$, and n is reaction order. Assuming the first-order reaction, the decomposition rate may be written as follows:

$${\displaystyle \frac{d\alpha }{dt}}=A·e^{-{\displaystyle \frac{Ea}{RT}}}\left(1-\alpha \right)$$

(2)

where A—frequency factor [min⁻¹]; E_a—activation energy [kJ mol⁻¹]; R—universal gas constant (8.3145 J·(mol K)⁻¹); T—temperature (K); t—time (min); and α—the ratio of actual mass loss to total mass loss at a given stage of the reaction. The ratio α is the extent of conversion, or fraction of the material pyrolyzed, defined by the expression [22]:

$$\alpha ={\displaystyle \frac{W_t}{W_0}}$$

(3)

where W_t is the weight loss quantity after time “t”, and W₀ is the total weight loss quantity after the complete pyrolysis of the sample.

When the reaction (1) is first order and occurs under isothermal conditions then activation energies (E_a) and frequency factors can be obtained from a plot of the natural logarithm of the reaction rate (ln k) versus the inverse of the absolute temperature (1/T), where k is the reaction rate (mass/time) and T is the temperature [K]. However, activation energies and frequency factors may also be found using non-isothermal experiments so long as the temperature varies at a constant rate. For non-isothermal kinetic parameters, with a constant heating rate β = dT/dt, the following equation is obtained:

$${\displaystyle \frac{d\alpha }{dT}}={\displaystyle \frac{A}{\beta }}·e^{-{\displaystyle \frac{E}{RT}}} \left(1-\alpha \right)$$

(4)

Modifications of these equations by various researchers have been aimed at determining the kinetic parameters of the processes involved. In this study, three methods were used to evaluate the kinetic parameters: (1) the Kissinger method, (2) the Kissinger–Akahira–Sunose (KAS) method, and (3) the Friedman method. These are model-free analysis methods, which allow for the determination of the activation energy (E_a) and the pre-exponential factor (A) without assuming a specific kinetic model or requiring prior knowledge of the reaction type. Model-free methods are widely used in the study of kerogen, source rocks, oil shales, and related materials [18,23].

2.3. Kissinger Method

Kissinger’s method is an example of a non-isoconversional method. Kissinger [24] found an approximate solution for the Arrhenius equation under non-isothermal conditions. In his solution, Kissinger shows that the temperature that corresponds with the maximum reaction rate (T_p) shifts with the heating rate (β) in accordance with the activation energy (E_a) and frequency factor (A) as follows:

$$\ln \left({\displaystyle \frac{\beta }{{T_P}^2}}\right)=\ln {\displaystyle \frac{AR}{E_a}}-{\displaystyle \frac{E_a}{R{T}_P}}$$

(5)

Because of its easy use, the Kissinger method is extensively used; however, it has some limitations, of which one is that this method produces a single value of the activation energy for any process, regardless of its actual kinetic complexity [25].

2.4. Kissinger–Akahira–Sunose Method

The KAS kinetic method (Equation (6)) is an integral isoconversional method, where the activation energy is a function of the conversion degree of a chemical reaction, and can be applied without any assumption concerning the kinetic model.

$$\ln \left({\displaystyle \frac{\beta }{T^2}}\right)=\ln {\displaystyle \frac{AR}{E_{a}g\left(\alpha \right)}}-{\displaystyle \frac{E_a}{RT}}$$

(6)

This method is suitable for multiple-step reactions without parallel reaction steps. The advantage is that each reaction point is evaluated. It is, however, suitable only for dynamic measurements.

2.5. Friedman Method

The Friedman analysis is the most common differential isoconversional method, where the measurements are analyzed for multiple conversion levels. Activation energy is determined without the assumption of reaction type by using points with the same conversion from the measurements with different heating rates or different isothermal conditions. This method is suitable for dynamic and isothermal measurements, moreover for multiple-step reactions without parallel reaction steps. In this analysis, each reaction point is evaluated [26]. This method is based on the equation:

$$\ln \left(\beta {\displaystyle \frac{d\alpha }{dT}}\right)=\ln \left[\mathrm{f}\left(\mathsf{\alpha }\right)·\mathrm{A}\right]-{\displaystyle \frac{E_a}{RT}}$$

(7)

3. Results

3.1. TG Analysis

Based on TG/DTG (Differential Thermogravimetry) curves, the decomposition of the rock sample is a two-stage process. The first mass loss occurs between 150 and 200 °C and is associated with the dehydration of clay minerals or the presence of free residual hydrocarbons (represented by the S₁ peak in Rock-Eval pyrolysis). The second mass loss, occurring between 350 and 540 °C, is attributed to the release of hydrocarbons as a result of organic matter pyrolysis.

When analyzing the decomposition of bulk rock samples at different heating rates, it can be observed that the mass loss between 250 and 600 °C is not highly variable, remaining within a range of 13–14%. The general trend indicates that higher heating rates result in lower total weight loss due to sample decomposition. At a heating rate of 2 °C min⁻¹, decomposition occurs as a continuous process, and the DTG curve shows no visible inflection. As the heating rate increases, the peaks become more distinct, and the temperature at which the maximum decomposition rate (T_p) occurs shifts to higher values: from approximately 420 °C at a heating rate of 5 °C min⁻¹ to 434 °C at a heating rate of 20 °C min⁻¹.

Mass loss during kerogen sample decomposition varies depending on the heating rate, ranging from 35% to 45%. The general trend is similar to that observed for bulk rock—the higher the heating rate, the lower the total weight loss due to sample decomposition. As in the bulk rock experiment, the temperature at which the maximum decomposition rate (T_p) occurs shifts to higher values with increasing heating rate, from 408 °C at a heating rate of 5 °C min⁻¹ to 429 °C at 20 °C min⁻¹ (Figure 2).

The main mass loss during decomposition occurs between 300 and 530 °C, but a slight DTG inflection can also be observed in the range of 200–280 °C. The general decomposition mechanism of kerogen in oil shale, as discussed in the literature (e.g.,: [15,27]), is typically divided into the following stages:

• Pyrolysis of free bitumens or kerogen decomposition into pyrolytic bitumen, causing low mass loss, occurring in the temperature range of 185–350 °C.
• Bitumen decomposition into oil, gas, and char, occurring up to 600 °C, producing rapid and significant mass loss.

In the experiments on source rock kerogen, this decomposition scheme is consistent, as evidenced by the Gram–Schmidt curve, which illustrates the amount of gas evolved during kerogen degradation (Figure 3). It is also noteworthy that the temperature of the maximum reaction rate (T_p), determined from the DTG curve, is lower than the temperature at which the highest gas evolution occurs. In the case presented in Figure 3, at a maximum heating rate of 20 °C min⁻¹, these temperatures are 429.6 °C and 457.3 °C, respectively.

3.2. Kinetic Parameters of Pyrolysis

3.2.1. Kissinger Method

The peak reaction rate temperatures (T_p) were estimated based on the inflection points of the DTG curves (Figure 2) for heating rates of 5, 10, 15, and 20 °C/min. The temperatures (recalculated to Kelvin) and corresponding heating rates (β) for bulk rock and kerogen were plotted according to the Kissinger equation (Equation (5)) (Figure 4). The slope and intercept of the linear regression allow for the determination of the activation energy (E_a) and frequency factor (A): slope = −E_a/R, and intercept = ln (AR/E_a), where gas constant R = 0.008314 kJ·(mol K)⁻¹.

For the rock sample, the activation energy is 293 kJ/mol and the frequency factor is 1.11 × 10²⁰ s⁻¹. For kerogen, these parameters are: 248 kJ/mol and 6.19 × 10¹⁶ s⁻¹ respectively (

Table 2. Kinetic parameters of bulk rock and kerogen based on Kissinger kinetic equation.

Heating Rate β [K/min]	Tp [K]	Linear Regression	Linear Correlation r2	Activation Energy Ea [kJ/mol]	Frequency Factor A [s−1]
rock
5	692.5	y = 39.78 − 35.32x	0.8396	293	1.11 × 1020
10	694.4
15	705.1
20	707.1
kerogen
5	681.0	y = 32.46 − 29.89x	0.9996	248	6.19 × 1016
10	691.2
15	697.4
20	702.3

). The kerogen analyses show a strong linear correlation (r² > 0.99) when plotted as ln(β/Tp²) versus 1000/T_p. The correlation is much weaker for bulk rock samples (r² > 0.83). This phenomenon confirms the observations made based on the TG analysis, suggesting that the decomposition of the rock is a multi-stage process, while the pyrolysis of kerogen is closer to one-stage reaction. The disadvantage of this method is that only one data point for each heating rate is obtained.

3.2.2. KAS and Friedman Methods

For analysis by the KAS and Friedman methods, the kinetic parameters were calculated in the temperature range from 250 to 450 °C (523–723 K), as in this range, the main weight loss indicates the decomposition reaction (Figure 2).

Figure 5 shows the KAS and Friedman plots for the model-free analysis of bulk rock and kerogen under dynamic conditions. These plots are presented within the conversion range of 0.2 to 0.9.

The KAS analysis depicts a graph of Log(β/T²) vs. inverse temperature for points at the same conversion level. The linear fit curves are shown for each conversion. The Friedman method, on the other hand, presents a graph of Log(dα/dt) vs. inverse temperature for points at the same conversion level. The activation energy and pre-exponential factor are determined from the slope and intercept of these lines.

From the plots, it can be observed that within the considered conversion range, the isoconversional lines are not parallel, and their slopes change as the process progresses. This indicates variations in the kinetic parameters during the pyrolysis process. For bulk rock, at conversion levels of 0.02 and 0.05, the isoconversional lines exhibit a horizontal or even reversed slope compared to other lines. This suggests that the kinetic parameters are unstable in the initial stages of pyrolysis. For kerogen, a similar pattern is observed in the conversion range of 0.2 to 0.9, regardless of the method used.

At the final stage of the process (conversion > 0.8), the regression lines have a steep slope, which corresponds to high apparent activation rates (Figure 6). Figure 6 illustrates the variation in activation energy (E_a) with the extent of conversion (α), calculated using both isoconversional methods. It is evident from this figure that the pyrolysis activation energy follows a similar trend in both the Friedman and KAS methods. The obtained E_a (and A) values depend on the extent of conversion, suggesting that the process follows a multi-step reaction mechanism.

In the initial stage of conversion (0.1 < α < 0.2), the activation energy for both bulk rock and kerogen decreases. However, beyond a conversion level of 0.2, it increases to over 1000 kJ/mol. A reasonable conversion range for considering the obtained kinetic parameters is 0.2–0.8. The apparent activation energies and pre-exponential factor values at specific conversion levels (α) are presented in

Table 3. Apparent activation energies and pre-exponential factor using KAS and Friedman for bulk rock decomposition.

Conversion α	Kissinger–Akahira–Sunose		Friedman
	Ea (kJ/mol)	Log A (Log(1/s))	Ea (kJ/mol)	Log A (Log(1/s))
0.1	376.908	50.829	336.286	46.186
0.2	157.600	28.665	152.583	27.369
0.3	134.782	25.464	134.759	24.678
0.4	173.360	27.873	187.041	28.224
0.5	261.628	34.305	297.796	36.355
0.6	358.818	41.296	392.651	43.101
0.7	479.445	49.912	554.653	54.795
0.8	740.899	68.872	945.262	83.197
0.9	1627.533	133.359	1919.576	153.787

(for bulk rock) and

Table 4. Apparent activation energies and pre-exponential factor using KAS and Friedman for kerogen decomposition.

Conversion α	Kissinger–Akahira–Sunose		Friedman
	Ea (kJ/mol)	Log A (Log(1/s))	Ea (kJ/mol)	Log A (Log(1/s))
0.1	188.141	32.455	153.073	28.53
0.2	129.731	25.711	126.802	24.76
0.3	170.258	28.294	200.169	30.05
0.4	236.220	33.021	274.241	35.32
0.5	302.393	37.678	331.881	39.22
0.6	352.165	40.967	386.303	42.82
0.7	434.835	46.724	489.899	50.10
0.8	578.686	56.933	670.648	62.97
0.9	1017.825	88.636	1255.099	105.20

(for kerogen).

4. Discussion

The kinetic analysis of source rock pyrolysis is a challenging task, as the decomposition of this rock is a highly complex reaction—even more complex than that of oil shales. Source rock is a mixture of kerogen and a wide range of minerals, undergoing numerous reactions both in parallel and in series [12,18]. Therefore, the kinetic parameters obtained from bulk rock decomposition, based on TG data, represent the overall reaction rather than individual reactions. Consequently, the activation energy should be regarded as apparent activation energy [15].

The mass loss observed in the TG curves for the non-isothermal analysis varies with different heating rates, particularly for the kerogen sample. The general trend is as follows: the higher the heating rate, the smaller the total weight loss. The difference between the extreme values of mass loss is quite significant (2.34%), which contradicts the conclusions of some authors regarding oil shale (e.g., [15]), who claim that overall mass loss during pyrolysis has little correlation with the heating rate.

The average values of kinetic parameters (for a conversion rate α between 0.2 and 0.8) obtained using the Kissinger, Kissinger–Akahira–Sunose (KAS), and Friedman methods are presented in

Table 5. The average values of activation energy (E_a), pre-exponential factor (A), and coefficient of determination (r²) calculated by the used methods.

Sample	Kissinger			KAS			Friedman
	Ea (kJ/mol)	Log A (Log(1/s))	r2	Ea (kJ/mol)	Log A (Log(1/s))	r2	Ea (kJ/mol)	Log A (Log(1/s))	r2
Source rock	293	20.05	0.8396	302	37.42	1.0000	345	39.84	0.9907
Kerogen	248	16.79	0.9996	305	37.79	0.9764	342	39.88	0.9991

. The apparent activation energy calculated using the Kissinger method is lower than the average E_a obtained using the KAS method. The highest values of apparent activation energy were determined using the Friedman method.

A strong linear correlation (r² = 1.00) for source rock analysis is achieved using the KAS method, while for the kerogen analysis, the best correlation (r² > 0.99) is obtained using the Kissinger and Friedman methods.

The differences in results obtained using the various methods, particularly when considering the average activation energy, are significant. Therefore, they should be compared with the findings of other researchers investigating the kinetics of oil shale or source rock reactions related to oil and/or gas generation.

Table 6. Comparison of kinetic parameters for kerogen.

Sample	Kerogen Type	HI (mg HC/g TOC)	Tmax (°C)	Ea (kJ/mol)	Log A (Log(1/s))	References
Green River oil shale	I	no data	no data	201	32.41	[29]
Huadian oil shale	I	670	442	231	36.79	[30]
Central Congo oil shale	I	746	438	212	33.41	[31]
Mongolia oil shale	II	no data	no data	379	89.80	[18]
Iwonicz Zdrój Menilite shale	II	601	431	219	13.37	[32]
Gorlice Menilite shale	II	472	401	305	37.79	This study

presents the selected kinetic parameter data from the literature for kerogen Type I and Type II. The sample examined in this study represents kerogen Type II, as evidenced in a previous study (sample 2 and 2 K in [28]). For the comparison presented in Table 6, the mean values obtained using the KAS method for kerogen were selected.

We can compare the results for different kerogen types (I and II), as numerous literature sources show that the type of kerogen has no significant influence on the parameters of reaction kinetics. Tegelaar and Noble [33] found (from 71 samples) that type II kerogens exhibit similar variations in kinetic behavior to type I kerogens.

As seen from the examples shown in Table 6, oil shales from different regions have different activation energies, which are related to variations in organic and mineral compositions [30]. The higher the volatile content, the more easily the kerogen reacts, or the more prone it is to react [18,34]. Wood [5] observed that the activation energy of most type I, type II, and type III kerogens tends to range from E = 175 kJ/mol to E = 265 kJ/mol. Furthermore, Wood [5], summarizing the trend of a strong positive correlation between E_a and ln A—the so-called compensation effect—noted activation energies ranging from E = 100 kJ/mol to 300 kJ/mol, which covers a broader spectrum of hydrocarbon reactions.

Based on the data contained in Table 6, it can be concluded that the values for the average activation energy and pre-exponential factor found in this study are relatively high but lower than those for Mongolian oil shale, which represents kerogen type II. Menilite shale from Iwonicz Zdrój, also a source rock from the Oligocene Menilite Formation of the Silesian Unit in the Polish Carpathians, has lower values for kinetic parameters [32] (E_a = 219 kJ/mol; A = 8.496 × 10¹⁶ h⁻¹). In the study, the lowest values of kinetic parameters, which are also the closest to those obtained in the work by Lewan et al. [32], were obtained using the Kissinger method.

The Kissinger method, however, provides a single activation energy value for the entire process, regardless of its actual kinetic complexity, making it less informative. In contrast, the implemented isoconversional methods are more informative, as they allow tracking of changes in kinetic parameters with varying temperatures and conversion rates. The activation energy is not stable across the entire conversion range, and the results of the KAS and Friedman methods clearly demonstrate this variability.

In this study, using a sample of source rock and kerogen from the Menilite layers, the results obtained using the Friedman method appear to be overstated. This is due to the inclusion of kinetic parameter values at high conversion rates (e.g., α = 0.8), which significantly overestimate the average values (Table 3 and Table 4). When considering only the conversion range of 0.2–0.8, the KAS method yields lower mean values, which seem more realistic—especially when compared to the results obtained using the non-isoconversional Kissinger method.

When comparing the results obtained from all three methods, it is also worth noting that the single activation energy value from the Kissinger method correlates with the kinetic parameter values obtained from the other methods at a specific degree of conversion. For the bulk rock sample, the results from the non-isoconversional Kissinger method are close to those obtained from the isoconversional methods at a conversion level of approximately α ≈ 0.5. In the case of kerogen, these values are similar at a conversion level of α ≈ 0.4.

At this point, it is also worth reiterating the reasons for the differences in reaction kinetic parameters among various rocks and kerogens. These parameters vary significantly and depend on numerous factors, making it highly complex to establish governing rules for these relationships. This complexity is why these issues remain the subject of extensive research in the literature.

It should be noted that the obtained activation energy values largely depend on the test method used (Rock-Eval, open-pyrolysis, thermogravimetry, etc.). As mentioned earlier, the mineral composition of the rock is a crucial factor. Many authors highlight the presence of carbonates, which play a significant role in oil shale pyrolysis [13,18,22,33]. The influence of clay minerals on rock decomposition kinetics has also been clearly indicated [34,35,36]. However, in many studies, including those cited in Table 6, there is no available data on the mineral composition of the samples.

The impact of the rock matrix can be of two types: (1) the presence of some minerals accelerates pyrolysis (e.g., carbonates, certain clay minerals) and (2) others delay pyrolysis (e.g., specific types of clay minerals). In the case of the experimental results, it can be observed that the activation energy for kerogen is usually slightly lower than that of the bulk rock (based on the Kissinger and Friedman methods). Additionally, the T_p of kerogen pyrolysis on DTG curves occurs at lower temperatures. This suggests that the mineral components of the source rock, particularly the selected clay minerals, have a delaying effect on pyrolysis. Their presence was confirmed by the TG analysis.

The kinetics of kerogen degradation depends on its composition, although, as mentioned above, it does not correlate with the genetic type of kerogen. Table 6 compares the available data on the Hydrogen Index (HI) and Tmax parameters for the cited samples. As seen in these examples, T_max does not appear to correlate with the activation energy value. However, in the case of HI, a clear trend is observed: higher HI values correspond to lower activation energy.

On the other hand, as stated by Tegelaar and Noble [33], kerogen typing based on the Hydrogen Index (HI) alone is insufficient to predict the kinetic parameters of its decomposition. For samples with HI values greater than approximately 400, there is no clear correlation between HI and molecular fingerprint [33]. The same authors demonstrated that the kinetics of hydrocarbon generation depend on the molecular structure of kerogen. For example, sulfur-rich kerogens exhibit lower activation energies due to the weaker bond strength of S-S and S-O bonds. The cleavage of these weaker heteroatomic bonds contrasts with the cleavage of C-C bonds, which requires higher activation energies.

According to Tegelaar and Noble [33], two primary factors control the molecular structure of kerogens and, consequently, their kinetic parameters: (1) the types and mixtures of biomacromolecules selectively preserved during diagenesis and (2) the organic sulfur content, which typically leads to lower decomposition temperatures for kerogen.

The molecular composition of the Menilite shale sample, which is the subject of this research, was previously examined in earlier studies [28,37] using Py-GC/FID analysis.

Table 7. Molecular characteristics of organic matter using pyrolysis gas-chromatography (Py-GC) [37].

Process	Pyrolysis 300–650 °C			Combustion 650–1000 °C
Hydrocarbon Fraction	C1–C9	C9–C15	C15+	C1–C9	C9–C15	C15+
Sample	Fraction Share [%]			Fraction Share [%]
Bulk rock	53.54	24.00	22.46	100.00	0	0
Kerogen	40.96	27.48	31.56	93.63	0	6.37

and Figure 7 present the results of these studies.

The analysis was conducted in two temperature ranges: 300–650 °C under inert gas conditions, and 650–1000 °C under oxidizing gas conditions. The products obtained in the Py-GC/FID steps were grouped into C₁–C₉, C₉–C₁₅, and C₁₅₊ hydrocarbon fractions. The composition of pyrolysis products in the 300–650 °C range is dominated by light, short-chain C₁–C₉ hydrocarbons.

Based on previous studies [33,34,38], it is generally considered that a higher proportion of long-chain hydrocarbons corresponds to lower activation energy. In the case of the studied samples, the presence of short-chain hydrocarbons explains the high activation energies observed.

5. Conclusions

The thermogravimetric analysis of the pyrolysis process of Carpathian source rock and kerogen was performed under an inert nitrogen atmosphere at five different heating rates. The kinetic parameters of pyrolysis were studied comparatively using model-free methods: Kissinger, Kissinger–Akahira–Sunose (KAS), and Friedman.

The decomposition reaction of bulk source rock can be divided into two distinct stages. The first stage, occurring between 150 and 200 °C, is associated with the dehydration of clay minerals, while the second stage, occurring between 350 and 540 °C, corresponds to the release of hydrocarbons as a result of organic matter pyrolysis.

Similarly, kerogen decomposition also occurs in two stages. The first stage results in a low mass loss within the temperature range of 200–280 °C, while the second stage, within 300–530 °C, exhibits the main mass loss effect. The total mass loss during kerogen decomposition varies with different heating rates, following the general trend: the higher the heating rate, the smaller the total mass loss. The temperature corresponding to the maximum reaction rate, as determined from the DTG curve, is approximately 30 °C lower than the temperature at which maximum gas evolution occurs in the kerogen sample.

The activation energy calculated using the Kissinger method is 293 kJ/mol for bulk rock and 248 kJ/mol for kerogen. The kinetic parameters determined experimentally by this method represent effective values for the pyrolysis process but do not account for individual steps with different activation energies.

The mean activation energy (Ea) values calculated using the Kissinger–Akahira–Sunose (KAS) isoconversional method are 302 kJ/mol for rock and 305 kJ/mol for kerogen. In contrast, the Friedman differential isoconversional method yields the E_a mean values of 345 kJ/mol for rock and 342 kJ/mol for kerogen.

The overall pyrolysis process exhibits a complex reaction mechanism. The apparent activation energy (E_a) and pre-exponential factor (A) values depend on the extent of conversion, suggesting that the process is a multi-step reaction. This dependence is consistent between the Friedman and KAS methods, though the exact kinetic parameter values differ between them.

The average activation energy and pre-exponential factor obtained in this study are relatively high, possibly due to the short-chain organic matter present in the source rock. A comparison of reaction kinetic parameters for bulk rock and kerogen indicates that kerogen has a lower activation energy than bulk rock, along with a shift in the reaction maximum to higher temperatures. This shift is attributed to the presence of clay minerals in the rock, as confirmed by thermogravimetric analysis.