CO\(_\text {2}\) injection for enhanced oil recovery (CO\(_\text {2}\)-EOR) or for storage in depleted oil and gas reservoirs can be a means for disposing of anthropogenic CO\(_\text {2}\) emissions to mitigate climate change. Fluid flow and mixing of CO\(_\text {2}\) and hydrocarbons in such systems are governed by the underlying physics and thermodynamics. Gravity effects such as gravity override and convection are mechanisms that can alter fluid flow dynamics, impacting CO\(_\text {2}\) migration, oil production and eventual CO\(_\text {2}\) storage at the field scale. This study focuses on convection in a miscible setting caused by non-monotonicity in oil density when mixed with CO\(_\text {2}\), i.e., a maximum mixture density occurs at an intermediate CO\(_\text {2}\) concentration. We perform high-resolution simulations to quantify the convective behavior in a simple box system where gravity effects are isolated. We show that convection of CO\(_\text {2}\) in oil is dependent on whether CO\(_\text {2}\) originates from above or below the oil zone. From above, convection follows classic convective mixing but is accelerated by viscosity decrease with increasing CO\(_\text {2}\). From below, convection flows upward due to CO\(_\text {2}\) buoyancy, but is countered by downward convection due to the heavier mixture density. This convective system is significantly more complex and efficient than from above. We characterize the instabilities in both early- and late-time regimes and quantify mixing rates. For a 100 mD reservoir, convective fingers would be on the order of centimeters in width and mix over a meter length scale within days to a month, depending on the placement of CO\(_\text {2}\). The simulations are performed in non-dimensional form and thus can be rescaled to a different reservoir parameters. Our results give important insights into field-scale impacts of convective mixing and can guide future work in development of upscaled models and experimental design.

CO \（_ \ text {2} \）注入以提高采油率（CO \（_ \ text {2} \）- EOR）或存储在枯竭的油气藏中可以是处理人为CO \ （_ \ text {2} \）排放以缓解气候变化。此类系统中的流体流动以及CO \（_ text {2} \）和碳氢化合物的混合受基本物理和热力学支配。重力效应（例如重力超控和对流）是可以更改流体流动动力学，影响CO \（_ \ text {2} \）迁移，产油和最终CO \（_ \ text {2} \）的机制。以现场规模存储。这项研究的重点是在与CO \（_ \ text {2} \）混合时由油密度的非单调性引起的可混溶环境中的对流，即，最大混合密度出现在中间CO \（_ \ text {2 } \）集中。我们执行高分辨率模拟，以量化一个简单的盒子系统中的对流行为，其中重力效应被隔离。我们表明，油中CO \（_ \ text {2} \）的对流取决于CO \（_ \ text {2} \）来自油区的上方还是下方。从上面可以看出，对流遵循经典的对流混合，但是随着CO \（_ \ text {2} \）的增加，粘度降低而加速了对流。从下方，由于CO，对流向上流动\（_ text {2} \）浮力，但由于较重的混合物密度而被向下对流所抵消。该对流系统比上面的复杂得多，效率也更高。我们描述了早期和晚期两种情况下的不稳定性，并量化了混合速率。对于100 mD的储层，对流指状体的宽度约为厘米，并在几天至一个月内以米长的刻度进行混合，具体取决于CO \（_ \ text {2} \）的位置。模拟以无量纲形式执行，因此可以重新缩放为不同的储层参数。我们的结果为对流混合的田间规模影响提供了重要的见识，并可以指导开发高级模型和实验设计的未来工作。