An improved balanced replicated sampling design for preliminary screening of the tailings ponds aiming at zero-waste valorization. A Romanian case study. (1st April 2023)

• Record Type: Journal Article
• Authors: Pencea, Ion; Turcu, Ramona-Nicoleta; Popescu-Argeș, Alina Cristina; Timiș, Andrei Lucian; Priceputu, Adrian; Ungureanu, Constantin; Matei, Elena; Nedelcu, Liviu; Petrescu, Mircea Ionuţ; Niculescu, Florentina
• Is Part Of: Journal of environmental management. Volume 331(2023)
• Publication Date: 2023-04-01
• Subjects: Tailings -- Recovery -- Sampling -- Balanced replicated design -- ANOVA -- Sampling and sub-sampling and analytical uncertainties; Environmental policy -- Periodicals; Environmental management -- Periodicals; Environment -- Periodicals; Ecology -- Periodicals; 363.705
• DOI: 10.1016/j.jenvman.2023.117260
• Languages: English
• ISSNs: 0301-4797

Abstract

Abstract: The scope of this study consists of setting up of an integrated cost-effective sampling & laboratory analyses procedure which delineates sampling, sub-sampling and analytical uncertainties in case of fine-grained extractive waste deposits. This procedure is designed to support the decision makers towards fine-grained waste deposits upcycling and land reclamation. This procedure consists of a balanced replicated sampling design (BRSD) coupled with a three split levels ANOVA data processing. The paper provides the readership with the mathematical backgrounds of the three split level ANOVA analysis (3L-ANOVA) and an Excel algorithm for its implementation. Also, the paper presents an example of implementation of the developed methods in the case of a Romanian iron ore tailings (IOT) old pond. The findings of the paper consist of: a) argues, based on OM, SEM-EDS, XRFS and XRD observations, that classical TOS is ineffective for fine-grained waste deposits; b) BRSD in conjunction with 3L-ANOVA analysis is the only approach fit for reliable characterization of the fine-grained stockpiles; c) sampling uncertainty is the critical factor of the uncertainty budget of the analyte concentration; d) Lilliefors approach is adequate for the hypothesis testing where or not the measurand is normal distributed; e) The outcomes of the BRDSD&3L-ANOVA investigations carried on Teliuc tailings, estimated at circa 5.5* 106 m3, consist mainly of mineral quantification at lot level i.e. quartz ∼54% (±7%), hematite ∼15% (±3%), calcite ∼11% (±3%), MgO 3% (±1%), Al2O3 9% (±2%). The concentrations of some CRMs like Ti, V, Ba, Y, W were found at ACE limits and their associated relative expanded uncertainties overpass 50%. Thus, the expanded uncertainties clearly depict the reliability of acquired data for the decision makers regarding waste valorization. f) The IOT into Teliuc can be upcycled as minerals for cement and ceramic industries as well as for geopolymer manufacture. Also, IOT can be downcycles as filler in road construction and mine closure. Finally, the Teliuc yard can be rehabilitated with zero-waste left behind. The data exactness provided by this procedure can be increased to any desirable level through increasing the number of collected items, but the cost of sampling and analyses increases proportionally. In such circumstances, the posted approach can be tailored at the stakeholder request as to safely underpin the decision to turn finegrained by-products into valuable secondary resources, facilitating a greater circularity of the mining industry. Highlights: Tailings valorization critically depends on the data reliability regarding pond content. The sampling, subsampling and analytical uncertainties quantify the data reliability. Balanced replicated sampling design coupled with three split levels ANOVA is proposed. Teliuc pond contains quartz (54%), iron oxide (17%), calcite (10%) and others minor ones. All minerals can be recycled as iron ore, additive in cement industry and geopolymers etc.