Abstract A Population Balance model for the continuous leaching of a roasted zinc concentrate was developed. The Maximum Mixedness and the Segregated Flow approaches were chosen to model the flow patterns and mixing conditions in the reactors. In order to compare these approaches and validate the proposed model, leaching experiments were carried out in a pilot plant. For that, it was also determined the residence time distributions in this plant reactors. As a result, it was observed that the Segregated Flow did not represent the actual pilot plant operation, since the conditions in which the Segregated Flow is recommended (i.e. segregation in the micromixing of the slurry phase, low diffusivity, and high slurry viscosity) were not present. On the other hand, using the Maximum Mixedness resulted in an excellent agreement between the experimental and modeled data (mean absolute error of 0.5% for the zincite conversion). This indicates that the complex particulate phenomena that occur during the leaching, as well as the flow patterns in the continuous reactors were adequately accounted in the model. Therefore, a robust tool for the design and control of leaching plants is delivered, since it is able to predict the effect of operational parameters (e.g. particle size distribution, solid/liquid ratio, feed flow rate) on the reactors performance. This can lead to better decision-making and costs reduction, contributing also to the prediction of problems faced in the operation of these plants. In a direct application of the developed model, it was predicted that the pilot plant could operate with the doubled capacity.

中文翻译：

---

预测连续浸出反应器性能的群体平衡方法：使用焙烧锌精矿的中试工厂模型验证

摘要 建立了焙烧锌精矿连续浸出的种群平衡模型。选择最大混合度和分离流方法来模拟反应器中的流动模式和混合条件。为了比较这些方法并验证所提出的模型，在中试工厂中进行了浸出实验。为此，还确定了该工厂反应器中的停留时间分布。结果，观察到分离流并不代表实际的中试设备操作，因为推荐使用分离流的条件（即在浆料相的微混合中分离、低扩散率和高浆料粘度）是不存在。另一方面，使用最大混合度在实验数据和模型数据之间产生了极好的一致性（锌矿转化的平均绝对误差为 0.5%）。这表明模型充分考虑了浸出过程中发生的复杂颗粒现象以及连续反应器中的流动模式。因此，提供了一种用于设计和控制浸出设备的强大工具，因为它能够预测操作参数（例如粒度分布、固/液比、进料流速）对反应器性能的影响。这可以导致更好的决策和成本降低，也有助于预测这些工厂运营中面临的问题。在开发模型的直接应用中，