Identifying tailings layers susceptible to static liquefaction, particularly for contractive soils, is a fundamental question considered by geotechnical engineers involved with the design, construction and operations of tailings storage facilities (TSFs). To determine the likelihood of static liquefaction for any given TSF, usually, geotechnical engineers aim to understand the in situ state of the tailings using empirical correlations and/or critical state soil mechanics. The state parameter approach has been recognised as an efficient tool to assess the liquefaction potential of tailings. Cone penetration testing (CPT) has been demonstrated to play an essential role in determining the in situ state parameter. However, the size and weight of the CPT rigs may limit their access to soft tailings surfaces. Furthermore, the logistical requirements to mobilise a CPT rig can be cumbersome in remote areas. The Panda dynamic cone penetrometer (Panda DCP) is a lightweight variable energy tool that has been used throughout Europe, particularly in France, for compaction control of engineered fills. Panda DCP applications have also been reported in Chile for compaction control of upstream construction of TSFs. The Panda DCP can rapidly be deployed for testing, is quick to carry out a test, can be undertaken by one person and can access areas inaccessible by low-pressure CPT rigs. The paper presents the results of a trial program that aimed to estimate state characteristics of tailings using the Panda DCP. The trial program was undertaken at a TSF in Queensland, Australia. Firstly, this study compared Panda DCP dynamic cone resistance, qd against CPT cone tip resistance, qc. Secondly, this study estimated in situ state of the tailings with the Panda DCP. This was done using two methods referred to in this paper as Method-1 and Method-2, with the results compared with CPT-based approach for determining in situ state of tailings. The results in this paper indicate that the Panda DCP could become an interesting alternative to screen and estimate the state of contractive and dense tailings.

A thin, distinctly complex layer was identified within the foundations of a tailings storage facility located in northern Australia. The strength of the layer in comparison to surrounding layers meant that a potential preferential slope instability plane was present in the foundation, and thus the layer became a key target for additional investigation. A site investigation was conducted in 2022 whereby cone penetration testing, vane shear testing, seismic and dissipation testing were conducted targeting this layer within the foundation. The results of the in situ testing indicated a layer exhibiting a cone tip resistance of less than one megapascal. The layer was classified as clay-like contractive and clay-like sensitive using conventional CPTu techniques. CPT correlations indicated that the material behaved in a contractive manner. However, the pore water pressure response to CPTu testing was partially drained, often with negative pore pressure. Vane shear testing in the layer was challenging, often resulting in broken vanes due to a ‘high strength’ response. Four mini-block samples were obtained utilising the mini-block sampling apparatus based on the Sherbrooke sampler technique to obtain high-quality undisturbed samples. CT scanning was completed on these samples to assess the internal structure before testing. Further XRD and electronic microscope analysis were completed. Laboratory testing included index, triaxial, oedometer and simple shear testing. The case study presented herein follows the process of identification of a complex geotechnical unit in a tropical residual soil, obtaining high-quality mini-block samples at depth from the field, analysis of the sample disturbance, and the subsequent laboratory testing and analysis. Results are shown to enable a holistic evaluation. The aim is to demonstrate the role mini-block sampling can play in the site characterisation of complex layers while highlighting potential pitfalls and difficulties encountered throughout the process.

CPTu (piezocone) testing is routinely used as part of modern site investigations in tailings storage facilities. The test is well-established and is celebrated as a convenient and rapid means to establish parameters used to analyse stability, settlement, potential for liquefaction, etc. The test was developed originally for natural soils. Many correlations exist for determining the various parameters, based on the three fundamental test outputs: cone resistance (qc or qt); sleeve friction (fs); and pore pressure (u). An inconvenient truth however is that in extremely soft or very loose (ooze-like) materials, typical in many tailings situations, even the best quality conventional CPTu equipment cannot reliably or repeatably measure and report extremely small values of fs. There are reasons for this, related to the physical design of CPTu sleeve systems, almost guaranteeing that this problem will exist. The in situ testing industry to some extent deals with this inconvenient truth as an ‘elephant in the room’; a topic not talked about much, if at all, but rather agonisingly hanging around in the shadows in the background of CPT testing. Of course treating it this way means that parameters determined from the well-established correlations, and other parameters that might be directly measured, are compromised. In 2021 two of the authors, a testing practitioner and an equipment manufacturer, developed a new CPTu cone that has been proven to overcome this problem. This required a significant design paradigm shift that has evolved into a new innovative design that works. The new CPTu cone was described by the authors at the CPT’22 Conference in Bologna (McConnell and Wassenaar, 2022). Now it has been in use for more than a year and some meaningful comparative tailings testing data is available that demonstrates the efficacy of the new device, provided (100 per cent) that it is maintained and calibrated carefully. The new CPT cone is commercially available to all; it is not a secret in-house tool.

Slurry tailings pumped to tailings storage facilities usually consist of 75 per cent of process water. The process water, usually hazardous if released to the ambient environment, will undergo one of the following processes: (1) evaporation; (2) forming overland flow and later merging in a decant pond; (3) mixing with settled tailings as porewater; and (4) infiltrating to the underlying tailings layer by gravity drainage and eventually forming seepage at the toe of the tailings dam. The conventional settling test may be able to quantify the ratio of overland flow to porewater storage, while incapable of estimating the gravity drainage as it has a sealed base. A purpose-made settling column allowing bottom drainage was constructed to investigate the outgoing process water during and after settling test. An outlet was designed at the bottom of the column to allow the drainage of process water while intercepting solids by placing 2-cm-thick settled tailings at the bottom with the particle size the same as the testing tailings samples. The test was started by depositing well-mixed tailings slurries into the column. During the experiment, the bottom drainage, and elevations of water surface and solid surface were monitored. The experiment was terminated when all the decant process water above was drained from the bottom outlet. After the initial settling, fresh water was poured above the settled tailings in the column to simulate rainfall-induced seepage. The outflow rate, EC, pH and total dissolved solids of the bottom drainage were monitored over time to quantify the leaching of chemicals/solids from the settled tailings. This paper reports the testing results of sandy and slime iron tailings.

Accurate estimation of the residual (or liquefied) undrained shear strength of tailings is crucial to the stability assessment of tailings storage facilities (TSFs). Several approaches are available for the estimation of the residual undrained shear strength of tailings, with these approaches generally including cone penetration testing with pore pressure measurement (CPTu) and/or laboratory testing on tailings samples retrieved from within the TSF. One such approach, proposed in the literature and adopted in tailings engineering practice, assumes that the residual undrained shear strength is equal to the CPTu friction sleeve resistance, measured during standard rate CPTu with a standard CPTu cone, henceforth referred as the ‘friction sleeve’ method.
This paper presents a review of the literature relating to inferring the residual undrained shear strength of soils from CPTu friction sleeve resistance, including the theoretical considerations forming the basis for the method as well as the supporting empirical evidence. A case study from a TSF located in the Western Australian Goldfields-Esperance region is also presented by the authors. The literature review and case study indicate that the friction sleeve may be able to provide an estimate of the residual undrained shear strength of fine-grained soils and tailings, however, the accuracy of this estimate may not be sufficient for TSF design purposes.
The friction sleeve method is also compared to other CPTu based residual undrained shear strength estimation techniques, which primarily utilise CPTu cone resistance. This paper highlights that, for specific applications, the friction sleeve method provides similar estimates to accepted industry standard practice CPTu based approaches, eg Jefferies and Been (2016) state parameter-based semi-empirical method. Lastly, some considerations on the uniqueness of residual undrained shear strengths are presented, which should be acknowledged when undertaking stability assessments of TSFs.

With the increasing number of tragic failures of tailings storage facilities (TSFs), the reliable determination of remoulded strength for stability analyses has become essential. However, the determination of remoulded strength for contractive strain softening materials includes many uncertainties. Considering the difficulty of recovering undisturbed samples of contractive fine grained soils, and limitations associated with laboratory testing, different correlations have been developed for the determination of remoulded strengths. One widely used correlation for fine grained soils is between the sleeve friction from cone penetration tests and the remoulded strength of fine grained soils. In an attempt to re-evaluate that correlation, a wide range of in situ tests and sampling was carried out in a coal tailings deposit to enable comparisons of inferred remoulded strength from a variety of techniques. In particular, both conventional friction sleeves and an innovative new 3 MPa CPT were used to assess the performance of the new sleeve system in the context of remoulded strength estimation. Vane shear tests (VST) at a range of rotation rates were then used to compare to the friction sleeve measurements and critically assess their performance in the measurement of remoulded strength. Finally, as reliable VST measurements of remoulded strength require undrained conditions to be maintained, an assessment of the potential for drainage to have occurred in these tests is made and only remoulded values deemed to be undrained were used in comparisons to the sleeve measurements.

To improve the settling, consolidation, and desiccation efficiency of bauxite residue in tailings storage facilities, a series of physical and chemical treatments were proposed before their disposal, including:
• whether or not to add fly ash and flocculent to thicken the residue
• whether or not to use caustic to reduce the pH
• pH neutralisation by either acid or sea water
• whether or not to dewater the sample through filtration and repulping.
Six types of residue samples subjected to various processing methods were identified, including one as Control. To identify the best combinations of these treatments, a series of laboratory and field tests were carried out on bauxite residue samples that underwent different sequences of treatments. The proposed tests include:
• Settling tests, particle size distribution tests, consolidation tests, and Atterberg limit tests to identify the impact of these treatments on the geotechnical parameters of tailings.
• Basin desiccation tests to identify how these treatments influence the desiccation behaviour of the tailings.
This paper discusses the outcome of the range of tests to date.