Abstract Mineral processing produces large volumes of dilute tailings that are flocculated and thickened, returning water to the process. The drive towards more sustainable processing has seen seawater considered to replace fresh water usage and an increased focus on how flocculation conditions can influence the fractal-like aggregates formed. The fragility of such structures and the likelihood that they change with time makes their study challenging. A synthetic tailings slurry of quartz and kaolin in seawater was flocculated at different mixing times and flocculant dosages, with aggregate size monitored over time by focused beam reflectance measurement (FBRM) and hindered settling rates determined after set reaction times. A model relating hindered settling rate to size as estimated by FBRM was used to derive an aggregate fractal dimension (Df) at each time. With flocculation at typical tailings solids concentrations rapid, aggregate breakage will normally dominate after very short reaction times. Under mild mixing conditions, settling rates are higher than expected from sizes measured in real time due to aggregate growth continuing after mixing ceases, leading to inflated Df estimates. An apparent decline in Df under more intense mixing was consistent with an increased presence of high aspect ratio (low effective porosity) aggregate fragments, supported by in situ aggregate imaging at different reaction times. In contrast, the observed trends for a reduction in Df with mixing intensity and an increase with dosage are both thought to represent the expected aggregate structure changes for the synthetic tailings. While estimating aggregate structural properties by relating FBRM sizes to settling rates can offer useful insights, it is important to recognise the aggregation state observed by FBRM under mixing applied to induce flocculation may differ from that which exists during hindered settling. Values of Df derived in this manner must therefore be viewed with caution.

中文翻译：

---

尾矿聚集体在海水中絮凝结构的时间演化

摘要 矿物加工会产生大量的稀尾矿，这些尾矿经过絮凝和浓缩，使水返回到流程中。朝着更可持续的加工发展的趋势已经看到海水被考虑取代淡水的使用，并且越来越关注絮凝条件如何影响形成的分形聚集体。这种结构的脆弱性以及它们随时间变化的可能性使他们的研究具有挑战性。石英和高岭土在海水中的合成尾矿浆在不同的混合时间和絮凝剂剂量下絮凝，通过聚焦光束反射测量 (FBRM) 随时间监测聚集体尺寸，并在设定的反应时间后确定受阻沉降速率。使用 FBRM 估计的将受阻沉降速率与尺寸相关联的模型来推导每次的总分形维数 (Df)。由于在典型的尾矿固体浓度下絮凝速度很快，因此在非常短的反应时间后，聚集体破裂通常会占主导地位。在温和的混合条件下，由于混合停止后聚合体继续增长，沉降率高于实时测量的尺寸的预期，导致 Df 估计值过高。在更强烈的混合下 Df 的明显下降与高纵横比（低有效孔隙率）聚集体碎片的增加一致，这得到了不同反应时间的原位聚集体成像的支持。相比之下，观察到的 Df 随混合强度降低和随剂量增加的趋势都被认为代表了合成尾矿的预期骨料结构变化。虽然通过将 FBRM 大小与沉降速率相关联来估计聚集体结构特性可以提供有用的见解，但重要的是要认识到 FBRM 在用于诱导絮凝的混合下观察到的聚集状态可能与受阻沉降过程中存在的聚集状态不同。因此，必须谨慎看待以这种方式导出的 Df 值。重要的是要认识到在用于诱导絮凝的混合条件下 FBRM 观察到的聚集状态可能不同于在受阻沉降期间存在的聚集状态。因此，必须谨慎看待以这种方式导出的 Df 值。重要的是要认识到在用于诱导絮凝的混合条件下 FBRM 观察到的聚集状态可能不同于在受阻沉降期间存在的聚集状态。因此，必须谨慎看待以这种方式导出的 Df 值。