3.1. Rock-Eval data

The FID signals show bimodal characteristics (S1 and S2) at all three heating rates. Although the values of S₁ and S₂ differ between heating rates, the hydrocarbon potential is not affected. The mean values of the hydrocarbon potential obtained from the three Rock-Eval pyrolysis experiments at different heating rates were 20.70 mg g⁻¹ and 2.24 mg g⁻¹, for the unextracted sample and the extracted sample, respectively; the difference (18.46 mg g⁻¹) corresponds to the content of soluble organic matter. This is like the abundance of chloroform bitumen ‘A’ of 1.91%.

It should be noted that the S₂ peak of the whole source rock sample does not completely overlap with that of the extracted sample (Fig. 1). The S₂ of the extracted sample was almost half that of the whole source rock sample, indicating that half of the S₂ was composed of soluble organic matter, rather than kerogen pyrolytic hydrocarbon.

Figure 1 Flame ionisation detection signal per milligram sample. (a) Real-time FID signal vs temperature at 25°C min⁻¹. (b) Cumulative FID signal vs temperature at 25°C min⁻¹. (c) Real-time FID signal vs temperature at 15°C min⁻¹. (d) Cumulative FID signal vs temperature at 15°C min⁻¹. (e) Real-time FID signal vs temperature at 5°C min⁻¹. (f) Cumulative FID signal vs temperature at 5°C min⁻¹.

In addition, a relatively small peak was observed between S₁ and S₂, which is like that observed by Li et al. (Reference Li, Chen, Ma, Cao, Li and Jiang2018) and termed S_1b. The S_1b response of trapped hydrocarbons lies in the default temperature range of S₂ for a normal programmed heating procedure. A small S₁ peak remained for the extracted sample, which may be attributed to some isolated hydrocarbons that could not be extracted due to lack of contact with the solvent during the extraction process.

3.2. Evaporation kinetic model

The pyrolysis of kerogen and source rock rich in soluble hydrocarbons can be described using chemical kinetic models. Therefore, the thermal evaporation of soluble hydrocarbons can also be described by chemical kinetics (Nezhad & Hami, Reference Nezhad and Hami2016). A series of independent and parallel first-order chemical reaction kinetics is often used to describe the thermal degradation of source rock. The thermal evaporation of free and adsorbed hydrocarbons in source rocks can also be described by an n-parallel first-order reaction model. Let x be the cumulative conversion rate of the hydrocarbon evaporation in the total evaporable hydrocarbons:

$$\displaystyle{{{\rm d}x} \over {{\rm d}t}} = \mathop \sum \limits_{i = 1}^m a_iA_i{\rm exp}\left({-\displaystyle{{E_i} \over {RT}}} \right)x_i$$

where a_i is the relative contribution, A_i and E_i are the pre-exponential and activation energies of the ith compositional group, respectively, x_i is the cumulative conversion rate of hydrocarbon evaporation in the ith compositional group, R is the gas constant, T is the absolute temperature and m denotes the number of evaporable hydrocarbon groups.

3.3. Conversion rate of hydrocarbon evaporation

The FID signal has a positive linear correlation with the number of hydrocarbons in the source rocks (Romero-Sarmiento 2019). Therefore, the conversion rate can be expressed by the following formula:

$$x_t = \displaystyle{{( {( S_{w, t}/M_w) -( S_{ex, t}/M_{ex}) } ) } \over {( {( S_{w, c}/M_w) -( S_{ex, c}/M_{ex}) } ) }}$$

where x_t is the evaporation rate at time t, S is the FID signal, and M is the weight of the sample for Rock-Eval pyrolysis, while S _w,t and S _w,c are the cumulative FID signals of the whole source rock at time t and at completion, respectively, and the subscripts w and ex represent the whole source rock sample and extracted sample, respectively. The whole source rock sample and extracted sample pyrolysis were heated using the same pyrolysis procedure. The calculated FID signal and evaporation conversion rate of the extractable crude oil are shown in Fig. 2.

Figure 2 (a) The relationship between the calculated flame ionisation detection signal of extractable crude oil and pyrolysis temperature. (b) The evaporate conversion rate of extractable crude oil.

3.4. Kinetic parameters

The cumulative evaporation rates at different times and temperatures can be calculated using the methods described in the previous sub-section. Kinetics 2000 software was used to calculate the kinetic parameters for the evaporation kinetics of extractable free and adsorbed hydrocarbons. The values of the pre-exponential and discrete distribution of activation energies could be obtained based on the time, temperature and evaporation rate input data from the three heating rates (Fig. 3 and Table 1). The pre-exponential term was 1.0889 × 10¹⁰ s⁻¹. The calculated results have a high degree of fit with the experimental results.

Figure 3 The calculated evaporate conversion rate of extractable crude oil compared with measured results. (a) Fit effect diagram at 25°C min⁻¹. (b) Fit effect diagram at 15°C min⁻¹. (c) Fit effect diagram at 5°C min⁻¹. (d) Activation energy distribution using pre-exponential 1.0889 × 10¹⁰ s⁻¹ for thermal evaporation of extractable crude oil.

Table 1 Evaporation kinetics of extractable hydrocarbon in source.

The activation energy (Ea) for hydrocarbon evaporation was in the range of 27–47 kcal mol⁻¹ in a bimodal distribution, with a primary peak at 27 kcal mol⁻¹ and a secondary peak at 39 kcal mol⁻¹. The secondary peak of Ea represents some large molecules that are difficult to vaporise, and these compounds may be adsorbed on or be (?) mutually soluble with kerogen.