Search

The increasing power of computer technology does not dispense with the need to extract meaningful information out of data sets of ever growing size, and indeed typically exacerbates the complexity of this task. To tackle this general problem, two methods have emerged, at chronologically different times, that are now commonly used in the scientific community: data mining and complex network theory. Not only do complex network analysis and data mining share the same general goal, that of extracting information from complex systems to ultimately create a new compact quantifiable representation, but they also often address similar problems too. In the face of that, a surprisingly low number of researchers turn out to resort to both methodologies. One may then be tempted to conclude that these two fields are either largely redundant or totally antithetic. The starting point of this review is that this state of affairs should be put down to contingent rather than conceptual differences, and that these two fields can in fact advantageously be used in a synergistic manner. An overview of both fields is first provided, some fundamental concepts of which are illustrated. A variety of contexts in which complex network theory and data mining have been used in a synergistic manner are then presented. Contexts in which the appropriate integration of complex network metrics can lead to improved classification rates with respect to classical data mining algorithms and, conversely, contexts in which data mining can be used to tackle important issues in complex network theory applications are illustrated. Finally, ways to achieve a tighter integration between complex networks and data mining, and open lines of research are discussed. [ABSTRACT FROM AUTHOR]

[Display omitted] • Cu (II) was reduced to Cu(I) on both pyrite and arsenopyrite surface. • Adsorption of IPETC was not accompanied with surface oxidised product removal. • Deprotonated N of IPETC coordination with As of enargite invoked for IPETC reaction with Cu-activated arsenopyrite. • Separation of arsenopyrite from pyrite achieved with 200 mg/l CuSO 4 and 300 mg/l IPETC. • Selectivity was lost when pH was dropped to values <11. To provide an understanding of the preferential interaction and flotation of Cu-activated arsenopyrite compared to pyrite, Cu species adsorbed, extent of surface oxidation, nature of bond and coordination between donor atoms of IPETC, and Cu and As acceptor atoms on mineral surfaces was investigated using X-ray photoelectrons spectroscopy (XPS) and Fourier-transform infrared spectroscopy (FTIR) analysis. Adsorption of Cu (II) and concurrent reduction to Cu(I) ions, which was paramount for the formation Cu-IPETC chelate complex was apparent on the surface of both pyrite and arsenopyrite from the XPS analysis. IPETC adsorption did not lead to removal of surface oxidation product on either pyrite or arsenopyrite surface, with extent of ferric hydroxide coverage on arsenopyrite far exceling that of pyrite. Adsorbed Cu(I) on the surface of pyrite was higher than arsenopyrite and was irreconcilable to their flotation performance. Bonding of S occurred with Cu(I), and As coordination with deprotonated N was suggested with the strength of the bond increasing at pH 11 using enargite as surrogate mineral. Roughing and single stage flotation of a refractory ore using 200 mg/l CuSO 4 and 300 mg/l IPETC at pH 11 gave an Au and As grade of 26.2 g/t and 1.35% respectively, at a mass pull of 6.9 wt% and 82.2% Au recovery. [ABSTRACT FROM AUTHOR]

Abstract: Century Mine, operated by Zinifex, is one of the largest zinc and lead deposits in Australia. The ore contains sphalerite and galena as valuable minerals, and carbonaceous material is present among the gangue minerals. The presence of carbonaceous matter is detrimental to the concentration of zinc and lead by flotation. Therefore, preflotation, which is flotation in the absence of collector and prior to galena flotation, is employed to remove some of the organic carbon, some of which is naturally hydrophobic, from the flotation feed. The main issues affecting the effectiveness of preflotation are: (i) low carbon recovery, (ii) low selectivity against zinc, and (iii) very high frother (methyl isobutyl carbinol, MIBC) consumption. The present study revealed that: (a) most of the MIBC added during flotation disappears from solution due to its high adsorption onto ore particles and especially on fine carbon enriched particles which have high surface area, (b) the addition of diesel oil can reduce MIBC consumption by 50% (in laboratory flotation tests) and (c) polypropylene glycol (PPG 192) showed lower affinity for carbon surfaces due to this reagent’s higher hydrophilic/lipophilic balance, producing stable froths at 50% of the current MIBC addition. [Copyright &y& Elsevier]

We have studied the degradation resulting from reprocessing of post-consumer HDPE in order to verify its re-stabilization needs. Cone-plate rheometry was used to provide information about the molar mass and the molar mass distribution after reprocessing and stabilizing the material. Infrared spectroscopy (FTIR), oxidation induction time (OIT), melt flow index (MFI) and cone-plate rheometry measurements were performed. Also, atomic absorption spectrophotometry tests were carried out to verify which catalytic residue was present in the recycled HDPE so as to correlate it to the degradation processes taking place in the polymer. The results showed that in this case the post-consumer HDPE displayed cross-linking when reprocessed, indicating the need for re-stabilization. They also showed that the use of 0.2% antioxidant was efficient in blocking these processes, and that the cone-plate rheometry technique was powerful in assessing degradative processes. [Copyright &y& Elsevier]

[Display omitted] • Differential oxidation of pyrite and arsenopyrite achieved with pulp aeration. • Recovery of pyrite and arsenopyrite was high when singularly treated. • Xanthate adsorption removed high FeOOH removal from arsenopyrite compared to pyrite. • Adsorption of xanthate on heavily oxidised arsenopyrite surface was slow. • Selectivity was lost at high xanthate concentration. Differences in the oxidation rate and extent of model pyrite and arsenopyrite minerals, augmented by pulp aeration, was investigated using dissolved oxygen demand tests and X-ray photoelectron spectroscopy (XPS) analysis. Flotation studies of single mineral admixture coupled with zeta potential measurements, UV–Vis spectroscopy analysis of collector concentration and ex-situ metal ions in solution after collector addition were ascertained to shed light on reasons for observed window for selective flotation. The highest selectivity was achieved with low collector concentration of 1 × 10-5 M PAX and selectivity index of 5.2 after 6 min flotation with collector concentration playing a critical role in attaining a good selectivity. Fast adsorption of xanthate and simultaneous oxidation to dixanthogen on slightly oxidised pyrite surface starved heavily oxidised arsenopyrite surface of the collector species, at low collector concentration, leading to its depression. The study was extended to the flotation of an auriferous refractory ore where similar observations were made. 62.7% rejection of pyrite was achieved in the staged flotation process. [ABSTRACT FROM AUTHOR]

• Lime plays a critical role in flotation by controlling pulp chemistry and collector adsorption. • Lime is the most economical and commonly used reagent for pH control in flotation. • Ca2+ ions in lime can enhance depression of minerals and flotation selectivity in a variety of ores. • For talc- and clay-rich ores, Ca2+ ions can be detrimental, and other pH modifiers are preferred. • Automated lime dosage control provides tangible advantages to operations. Lime (calcium oxide or calcium hydroxide) is arguably the most common reagent used in the extractive minerals processing industry. As an effective and economic pH modifier, lime is widely used in the process of selective flotation of the major sulphide minerals. This paper reviews: lime preparation factors that influence the reagent performance properties; mechanisms of lime interaction with collector chemicals and mineral surfaces; its role in promoting selective separation between different sulphide minerals; and the effect on pulp and froth properties. The practical aspects of lime dose locations and the most common sulphide flotation process applications are also discussed. [ABSTRACT FROM AUTHOR]

The present study investigates the effect of aeration and diethylenetriamine (DETA) on the selective depression of pyrite in a porphyry copper–gold ore, after regrinding (at grind sizes, d 80 = 38 and 8 μm) with respect to Au recovery and grade using oxygen demand tests, flotation, QEMSCAN, X-ray spectroscopy (XPS) and EDTA extraction analysis. It was found that pyrite depression increases after aeration and with decreasing grind size. This was observed to be due to the markedly higher oxygen consumption rate of pyrite at the 8 μm (kla = 0.10 min −1 ) than at the 38 μm grind size (kla = 0.02 min −1 ). The addition of DETA improved pyrite depression (9% with aeration only versus 39% with aeration + DETA) at the 38 μm grind size. Gold and copper flotation recovery followed pyrite recovery for the two grind sizes using XD5002 in the presence of air and DETA. The surface analysis (XPS and EDTA extraction) revealed that the significant pyrite depression at the 8 μm grind size was due to increased amount of surface iron oxides, oxy-hydroxides (Fe O/OH), sulphate species and increased liberation of mineral phases (QEMSCAN analysis), whilst the poorer pyrite depression at the 38 μm grind size was due to insufficient liberation of mineral phases and the persistence of activating Cu on the pyrite surface. The addition of DETA increased pyrite depression at the coarser grind size due to a significant reduction in Cu(I)S and increased Cu(II) O species, correlating with the flotation results of pyrite under this test condition. Two-stage copper and pyrite flotation, followed by Au cleaning after regrinding to 38 μm grind size, under high pH or aerated condition is proposed as the recommended route to optimise Au flotation. [ABSTRACT FROM AUTHOR]

Highlights: [•] DETA depressed xanthate-induced Cu-activated pyrite (Py) at pH 10 (p80=38μm). [•] Pyrite depression is enhanced when aeration and DETA are used in combination. [•] Depression of Py was due to complexation of Cu in solution or its adsorbed state. [•] Formation of Cu(I)–DETA complex on pyrite surface increased surface Cu oxidation. [Copyright &y& Elsevier]

The present study investigates the implications of gold mineralogy on its flotation recovery, in a typical porphyry copper–gold ore (1.7g/t Au feed grade). Also, flotation strategies to optimise gold recovery and grade without compromising copper recovery and grade were examined. The approach used in this study involves a combination of different mineralogical techniques such as quantitative X-ray diffraction (QXRD) and scanning electron microscopy (QEMSCAN). Based on the mineralogical information obtained, different collector regimes involving sodium iso-butyl xanthate (SIBX), N-butoxycarbonyl-n butyl thionocarbamate (XD5002), mercaptobenzothiazole (MBT) and dithiophosphate (DTP) at different grind sizes (p80, i.e. a nominal product size of 80wt% passing (a) 70µm and (b) 38µm screen) were tested to examine the flotation behaviour of gold in relation to copper recovery. Chalcopyrite and covellite are the principal copper mineral phases in the ore, with pyrite as the main sulphide gangue mineral. Quartz, Fe oxide and carbonates were identified as the main non-sulfide gangue minerals. The identified gold particles occurred as either liberated particles (∼10–22µm) or fine inclusions (∼1–3µm) in pyrite, biotite and sphalerite. Two-stage flotation (copper and pyrite rougher stages) at a p80 of 70µm indicated that high gold (95.2%) and copper (97.4%) recovery can be achieved, but the levels of pyrite in the concentrate are also very high (49.1%). At finer grind (p80=38µm) and without pyrite flotation stage, gold recovery decreased from 95.2% to 73%, but at an improved grade (5.3g/t) compared to 3.5g/t in the baseline. This approach improved gold grade significantly due to a drastic reduction in pyrite recovery. The flotation study indicates that the collector regime, SIBX/XD5002 (at 38µm feed size) gives the maximum gold recovery and grade while rejecting significant pyrite without compromising copper recovery and grade. Further studies are warranted to identify means of recovering Au from the copper flotation tail. [ABSTRACT FROM AUTHOR]

[Display omitted] • Minimum fluidization water rate increases with increasing bed particle size. • The relationships between bed-level and superficial water rate, and bed-level and superficial gas rate are more apparent at finer bed sizes compared to coarser bed sizes. • At constant bed masses, flotation performance is optimal at finer bed sizes due to the higher bed levels that can be attained. • After adjusting for bed level, flotation performance remains optimum at finer bed sizes due to low porosity at coarser bed sizes leading to poor bubble-dissipation. • This indicates a potential upper bed particle size limit, beyond which fluidization is unfeasible. This work investigates the effect of superficial water rate, superficial gas rate, and bed particle size on bed-level and coarse particle flotation performance. 5 kg of quartz in three size ranges (- 1180 + 850 µm, − 850 + 425 µm, − 425 + 250 µm) were used to study the effect of bed particle size on bed-level at varying superficial water rates and superficial gas rates. The influence of bed particle size on flotation performance was studied in two stages. One stage maintained constant bed masses, and another maintained constant bed-level, superficial water rate, and superficial gas rate. Results showed that bed-level increases with increasing superficial water rate after a minimum is attained but the rate of increase is limited by increasing bed particle size. Bed level decreases with an increasing superficial gas rate within a stable bed but this rate of decrease is also lowered at relatively coarser bed particle sizes. At constant bed masses, flotation performance tends to be better for the − 425 + 250 µm bed due to the high range of bed levels that can be attained. However, after adjusting for these parameter differences, we found that there may be an inherent upper bed particle size limit beyond which fluidization is unfeasible. [ABSTRACT FROM AUTHOR]

[Display omitted] • Bubble-load is optimized in bubble-particles aggregates of fine particles compared to coarser ones. • Coarser particles simulate the detachment behavior of finer particles at increased fluid SGs. • Increased critical detachment amplitudes at increased fluid SGs translate to better flotation performance for coarse particles in terms of recovery, rate, and maximum floatable size. • Increasing fluid SG beyond a limit increases the likelihood of detachment in the froth phase. • Particle weight in fluid is the major detachment force, and the effect of cell turbulence on coarse particle flotation performance is largely dependent on it. In this work, we investigate the influence of fluid density on bubble-particle detachment and coarse particle flotation performance. Detachment studies were carried out using an electro-acoustic apparatus in a dense liquid environment using a lithium heteropolytungstate (LST) mixture with subsequent flotation studies carried out in a micro flotation cell at varying fluid densities. Our results indicate that bubble load was inherently limited by particle surface area and weak particle–particle interactions at coarser sizes. Coarser particles tend to follow the detachment behavior of relatively finer particles at increased fluid density. We also found that flotation performance in terms of recovery, rate, and maximum floatable size can be enhanced by controlling fluid density within a stated limit beyond which detachment in the froth phase causes overall performance to drop. [ABSTRACT FROM AUTHOR]

GPR* – Gaussian Process Regression. [Display omitted] • The selected input variables define flotation copper recovery prediction performance. • Regularised Neighbourhood Component Analysis algorithm selects good features for predicting copper recovery. • Gaussian Process Regression model make good rougher copper recovery prediction using relevant input variables. • Among on-plant measured parameters, feed particle size is the most sensitive rougher flotation input variable. The emergence of advanced computational techniques provides opportunity in better modelling complex, nonlinear relationship of industrial processes. In this work, the performance of Gaussian Process Regression (GPR) models in predicting rougher copper recovery, following development with and without model-selected, relevant rougher flotation variables has been investigated. Randomly selected training and validation data sets as well as an independent testing data set (later than the original training and validation sets) were used to demonstrate the model predictive behaviour. The results showed that exponential GPR covariance function, modelled with selected relevant rougher flotation variables (throughput, feed particle size, xanthate dosage, frother dosage and froth depth), was the best model yielding correlation coefficient (r), root mean square error (RMSE), normalised root mean square error (NRMSE), variance accounted for (VAF) and performance index (PI) values of 0.99, 0.10, 0.01%, 99.99% and 1.97, respectively for training set. In the same manner, the validation set yielded 0.95, 0.40, 0.06%, 91.0%, and 1.75 whilst the testing set gave 0.96, 0.39, 0.07%, 92.10% and 1.78, confirming that exponential GPR algorithm can make good rougher copper recovery prediction given a set of useful rougher flotation variables. In all cases, the Regularised Neighbourhood Component Analysis algorithm performed best better in selecting useful flotation variables for rougher copper recovery prediction. Sobol's global sensitivity analysis indicated feed particle size as the most sensitive input variable recording first and total sobol indices values of ~63% and ~67%, respectively. Partial dependence plots revealed the functional relationship between the top two sensitive input variables (feed particle size and froth depth of tank cell 1) and predicted rougher copper recovery, suggesting their critical operating regime. [ABSTRACT FROM AUTHOR]

[Display omitted] • Bed permeability is limited at increased bed particle sizes. • Bed permeability was identical irrespective of apparent bed density. • Bed material properties influence bubble size range. • Less dense beds are more likely to elutriate into concentrate, limiting grade. • Recovery was identical irrespective of apparent bed density. This work investigated the influence of bed material density on fluidized bed flotation performance. Experiments were carried out in a HydroFloat cell to investigate the influence of bed material density on required bed masses and bed behaviour at a bed level of 15 cm and water rate of 1.19 cm/s. Bubble size image analysis was carried out under four different fluidization conditions: water only, water and frother (1 ppm), water, frother (1 ppm) and quartz, and water, frother (1 ppm) and aluminium oxide. Flotation experiments were undertaken to investigate the influence of the resulting bed density and characteristics on concentrate recovery and grade performance. The results confirm that there is an upper particle size limit on fluidization. Apparent bed density was higher for the denser bed material (aluminium oxide) even though bed behaviour remained identical with both being highly permeable. The product Cu grades for the quartz bed were lower (8.42 % and 15.82 %) compared to the alumina bed (23.81 % and 21.86 %). This was due to the elutriation of quartz into the concentrate with Si concentrations of 25.85 % and 15.69 % compared to 2.4 % in the feed. In contrast, Al concentrations were 2.44 % and 1.09 % in the concentrates compared to the 0.38 % in the feed. However, recovery performances for both bed materials were identical (39.72 % and 40.88 % at low collector concentrations and 98.93 % and 97.95 % after ethylenediaminetetraacetic acid (EDTA) cleaning and increased collector concentration for the quartz and aluminium oxide beds respectively). This indicates that apparent bed density may have only a slight influence on flotation recovery compared to other direct control variables such as fluidization water rate and bed height. [ABSTRACT FROM AUTHOR]

In most copper sulphide flotation processes, optimising recovery of trace elements such as gold or maximising the rejection of penalty elements such as uranium, arsenic and fluorine remains a major metallurgical challenge. This review looks at the existing and possible strategies of improving gold recovery in porphyry copper ores and rejection of penalty elements such as uranium, arsenic and fluorine in copper flotation concentrates. [ABSTRACT FROM AUTHOR]

Large carnivores are declining worldwide and few examples of successful reintroductions exist, because of their large home-ranges, low reproductive rates, and penchant for human–wildlife conflict that is the main cause of their decline. Moreover, few studies assess whether habitat suitability predicted before reintroduction, a critical evaluation step, matches post-reintroduction habitat selection. We examined habitat-related factors contributing to a successful brown bear ( Ursus arctos ) reintroduction in central Europe. Starting in 1999, 10 brown bears were translocated from Slovenia to Trentino in the Italian Alps, and this population has since grown by >10%/year. First, we estimated multi-scale resource selection functions (RSF) with GPS collar data and validated models with k-folds cross validation and external VHF data. Then, we used Kappa-statistics to compare our population-scale RSF with a habitat suitability model (HSM) developed to predict potential habitat before reintroduction. Lastly, we employed least-cost path (LCP) analyses integrating our within home-range scale RSF to define movement paths. Overall, the HSM predicted post-reintroduction habitat selection well in many areas, but bears used orchards and shrubs more, and mixed/conifer forests and pastures less than expected prior to reintroduction. Finally, we identified road crossings of predicted paths between preferred habitat patches. We found two potential crossings in the Adige Valley, likely the biggest constraint for the study population to expand eastward and impeding dispersal to/from the closest bear population (Dinaric–Pindos population). Increasing awareness for key brown bear habitats and corridors, especially in potential ecological traps within cultural landscapes, will be necessary for large carnivore conservation. [ABSTRACT FROM AUTHOR]

In the past years, network theory has successfully characterized the interaction among the constituents of a variety of complex systems, ranging from biological to technological, and social systems. However, up until recently, attention was almost exclusively given to networks in which all components were treated on equivalent footing, while neglecting all the extra information about the temporal- or context-related properties of the interactions under study. Only in the last years, taking advantage of the enhanced resolution in real data sets, network scientists have directed their interest to the multiplex character of real-world systems, and explicitly considered the time-varying and multilayer nature of networks. We offer here a comprehensive review on both structural and dynamical organization of graphs made of diverse relationships (layers) between its constituents, and cover several relevant issues, from a full redefinition of the basic structural measures, to understanding how the multilayer nature of the network affects processes and dynamics. [ABSTRACT FROM AUTHOR]