Abstract Recent pyrolysis analytical methods such as Shale Play (Romero-Sarmiento et al. 2016), HAWK-PAM (Maende et al., 2017) and Incremental S1 (Beti et al., 2018) enable improved understanding of the quantity and quality of free petroleum present in rock samples. However, it is well established that the evaporation losses restrict the ability of an experiment to accurately quantify and qualify free petroleum in rock sample (Michael et al., 2013; Jiang et al., 2016; Li et al. 2020). This article presents a thermodynamic model to explain the change in hydrocarbons composition of rock samples residing in a laboratory as a function of time. The model involves multicomponent residue curves and vapor-liquid equilibrium calculations. The changes in hydrocarbon composition predicted using models were compared with experimental data obtained from thermal desorption-gas chromatography (TD-GC) analysis of rock samples. High correlation coefficients are observed between predicted compositions and the experimental data inferring a good model fidelity. Relatively higher deviations were also observed between experimental data and predicted composition with time. The proposed model is useful in complementing the experimental results and enhancing the fundamental understanding of hydrocarbon loss from rock samples. This model also has the potential to eliminate multiple experiments to estimate the compositions after hours of sample collections.

中文翻译：

---

岩石样品中碳氢化合物成分随时间的变化：热力学蒸发模型

摘要 最近的热解分析方法，如页岩作用（Romero-Sarmiento 等人，2016 年）、HAWK-PAM（Maende 等人，2017 年）和增量 S1（Beti 等人，2018 年）能够提高对岩石样品中存在游离石油。然而，众所周知，蒸发损失限制了实验准确量化和鉴定岩石样品中游离石油的能力（Michael 等人，2013 年；Jiang 等人，2016 年；Li 等人，2020 年）。本文提出了一个热力学模型来解释实验室中岩石样品的碳氢化合物成分随时间的变化。该模型涉及多组分残留曲线和汽液平衡计算。使用模型预测的碳氢化合物组成的变化与从岩石样品的热解吸气相色谱 (TD-GC) 分析获得的实验数据进行了比较。在预测的成分和实验数据之间观察到高相关系数，推断出良好的模型保真度。随着时间的推移，在实验数据和预测的组成之间也观察到了相对较高的偏差。所提出的模型有助于补充实验结果并增强对岩石样品中碳氢化合物损失的基本认识。该模型还有可能在数小时的样本收集后消除多次实验以估计成分。在预测的成分和实验数据之间观察到高相关系数，推断出良好的模型保真度。随着时间的推移，在实验数据和预测的组成之间也观察到了相对较高的偏差。所提出的模型有助于补充实验结果并增强对岩石样品中碳氢化合物损失的基本认识。该模型还有可能在数小时的样本收集后消除多次实验以估计成分。在预测的成分和实验数据之间观察到高相关系数，推断出良好的模型保真度。随着时间的推移，在实验数据和预测的组成之间也观察到了相对较高的偏差。所提出的模型有助于补充实验结果并增强对岩石样品中碳氢化合物损失的基本认识。该模型还有可能在数小时的样本收集后消除多次实验以估计成分。