Condensing solvent bitumen recovery processes are a promising alternative to steam-assisted processes in terms of greenhouse gas emissions and energy consumption reduction. However, the details of the underlying physical mechanisms of these processes (gravity drainage and mass transfer) are not fully understood, and predictive models are lacking. Three key parameters in these mechanisms are permeability, irreducible water saturation, and mechanical dispersion. In particular, mechanical dispersion can significantly impact mass transfer rates but has not been yet been measured for mass transfer of bitumen into a draining liquid solvent layer. The objectives of this study are to measure and model the impact of permeability and irreducible water saturation on a lab-scale representation of a condensing solvent process, and to assess mechanical dispersion. Mass transfer rates were measured with toluene injected over a bitumen layer in a Hele-Shaw cell filled with silica sands and glass beads with and without irreducible water saturation. With this apparatus, key variables in the gravity drainage process such as the drainage layer flow rate, inclination, and composition are measured. Hence, there is sufficient information to identify and model the process mechanisms. Toluene volumetric flow rates between 0.5 and 15 cm/min and permeabilities within the range of 47 to 254 D were considered. The high dilution (high flow rate) data were predicted with an analytical model. The model was previously derived from Fick’s second law of diffusion for the effective orthogonal mass transfer of bitumen into a drainage layer flowing according to Darcy’s law. The full range of data was matched with a numerical model. This model was based on convective mass transfer of bitumen into the drainage layer using a correlated convective mass transfer coefficient derived from effective molecular diffusivity and Darcy flow. Neither model required mechanical dispersion even in presence of irreducible water. A square root dependence of mass flux versus permeability was observed, consistent with molecular diffusion with no mechanical dispersion. The numerical method is a step towards predicting mass fluxes at reservoir permeabilities and condensing solvent processes at the field scale.

中文翻译：

---

填充 Hele-Shaw 池中液体甲苯在沥青上的重力排水：渗透性、束缚水饱和度和分散性的影响

就减少温室气体排放和能源消耗而言，冷凝溶剂沥青回收工艺是蒸汽辅助工艺的一种有前景的替代方案。然而，这些过程（重力排水和质量传递）的潜在物理机制的细节尚未完全了解，并且缺乏预测模型。这些机制中的三个关键参数是渗透率、束缚水饱和度和机械分散度。特别是，机械分散可以显着影响传质速率，但尚未测量沥青进入排出液体溶剂层的传质。本研究的目的是测量和模拟渗透率和束缚水饱和度对冷凝溶剂过程的实验室规模表示的影响，并评估机械分散。通过在 Hele-Shaw 池中的沥青层上注入甲苯来测量传质速率，该池填充有硅砂和玻璃珠，具有和不具有束缚水饱和度。利用该设备，可以测量重力排水过程中的关键变量，例如排水层流速、倾斜度和成分。因此，有足够的信息来识别和建模过程机制。考虑了 0.5 至 15 cm/min 之间的甲苯体积流速和 47 至 254 D 范围内的渗透率。使用分析模型预测高稀释度（高流速）数据。该模型先前源自菲克第二扩散定律，用于沥青有效正交质量传递到根据达西定律流动的排水层中。所有数据均与数值模型相匹配。该模型基于沥青进入排水层的对流传质，使用源自有效分子扩散率和达西流的相关对流传质系数。即使存在束缚水，这两种模型都不需要机械分散。观察到质量通量与渗透率的平方根依赖性，这与无机械分散的分子扩散一致。数值方法是预测油藏渗透率质量通量和油田规模冷凝溶剂过程的一步。