A Comparative Study on Supervised Machine Learning Algorithms for Copper Recovery Quality Prediction in a Leaching Process.

The copper mining industry is increasingly using artificial intelligence methods to improve copper production processes. Recent studies reveal the use of algorithms, such as Artificial Neural Network, Support Vector Machine, and Random Forest, among others, to develop models for predicting product quality. Other studies compare the predictive models developed with these machine learning algorithms in the mining industry as a whole. However, not many copper mining studies published compare the results of machine learning techniques for copper recovery prediction. This study makes a detailed comparison between three models for predicting copper recovery by leaching, using four datasets resulting from mining operations in Northern Chile. The algorithms used for developing the models were Random Forest, Support Vector Machine, and Artificial Neural Network. To validate these models, four indicators or values of merit were used: accuracy (acc), precision (p), recall (r), and Matthew's correlation coefficient (mcc). This paper describes the dataset preparation and the refinement of the threshold values used for the predictive variable most influential on the class (the copper recovery). Results show both a precision over 98.50% and also the model with the best behavior between the predicted and the real values. Finally, the obtained models have the following mean values: acc = 0.943, p = 88.47, r = 0.995, and mcc = 0.232. These values are highly competitive when compared with those obtained in similar studies using other approaches in the context.

Released fraction of polychlorinated biphenyls from soil-biosolid system using a leaching procedure and its comparison with bioavailable fraction determined by wheat plant uptake.

The bioavailability of polychlorinated biphenyls (PCBs) in soils amended with biosolids was estimated using an aqueous leaching process of the compounds combined with rotating disk sorptive extraction (RDSE), and compared with bioavailability determined through of PCB absorption in wheat plants growing in the same soil-biosolid matrix. The matrices consisted of soil amended with biosolids at doses of 30, 90, and 200 Mg/ha, which increase concomitantly the organic matter content of the matrix. Considering that PCBs were natively absent in both the biosolids and soil used, the compounds were spiked in the biosolids and aged for 10 days. For each biosolid dose, the aqueous leaching profile was studied and equilibrium time was calculated to be 33 h. The leaching fractions determined by RDSE, considering total PCBs studied, were 12, 7, and 6% and the bioavailable fractions absorbed by the wheat root were found to be 0.5, 0.3, and 0.2% for 30, 90, and 200 Mg/ha doses, respectively. Both fractions leachable and bioavailable decrease with both increasing hydrophobicity of the compound (Kow) and increasing in the biosolid dose. It was found that both fractions (leaching and bioavailable) correlated according to the bivariate least squares regression, represented by a coefficient of correlation of 0.86. Therefore, the application of the chemical method involving a leaching procedure is an alternative to estimate the bioavailable fraction of PCBs in wheat plants in a simpler and in a shorter time.

Rotating disk sorbent extraction for pre-concentration of chromogenic organic compounds and direct determination by solid phase spectrophotometry.

A novel and very simple microextraction approach for pre-concentration and direct solid phase spectrophotometric measurement has been developed for the determination of chromogenic analytes. The model analyte to assess this approach was the chromophore malachite green (MG). The analyte was extracted from water samples onto a small rotating disk made of Teflon containing a sorbent phase of polydimethylsiloxane (PDMS) on one of its surfaces. We refer to the extraction procedure as rotating disk sorptive extraction (RDSE). After extraction, the sorbent phase with the concentrated analyte was separated from the Teflon disk and used directly for MG determination by solid phase spectrophotometry at 624 nm, without the necessity of a desorption step. Chemical and extraction variables such as concentration of sodium sulfate, pH, disk rotational velocity, extraction time, and temperature were studied in order to establish the best conditions for extraction. Under optimum conditions, the extraction of MG was carried out in 18 min and 90 min, for sample volumes of 100mL or 1000 mL, respectively. The detection limit, based on three times the standard deviation of the blank phase (3σ(b)), was 1.4 µg L⁻¹ and the repeatability, expressed as relative standard deviation (RSD), for 20 µg L⁻¹ MG was 8.1%. This study also applied the method to real samples, obtaining quantitative recovery (mean recovery of 99.3%). The PDMS phases could be reused after desorbing the MG into methanol for 3h. Replacement of the PDMS film onto the disk is very easy and low cost.

Polychlorinated biphenyl behavior in soils amended with biosolids.

Assessment of the mobility of polychlorinated biphenyls (PCBs) in soils, amended with biosolids at a rate of 30Mgha(-1), was performed using an incubation process and leaching columns. The incubation process was carried out for 0, 30, and 60d under field capacity conditions at 25 degrees C. The mobility of PCBs was assessed using solutions of 0.5molL(-1) CaCl(2) and 25mgL(-1) linear alkylbenzenes sulfonate (LAS). Ultrasound-assisted pressurized solvent extraction (US-PSE) was utilized for compound separation from the solid matrix. Compounds were determined by gas chromatography coupled to mass spectrometry. The biosolids, containing a background PCB concentration about 300microgkg(-1), were spiked with the analytes at 300mgkg(-1) to obtain a clearer determination of their behavior when the biosolid was mixed with soil. In biosolid-amended soils, an increase was observed in the extractability of PCBs with increasing incubation time, which may be attributed to organic matter breakdown. The leaching column study showed that CaCl(2) was unable to mobilize the PCBs from the biosolid to the soil, whereas LAS mobilized these compounds within the time scale implicit in the experiment (30d). The most mobilized congeners in the columns corresponded to those with the greatest molecular weight (hexa- and heptachlorinated), probably due to the higher hydrophobicity of these compounds. Results indicate that the presence of important concentrations of LAS in biosolids could mobilize PCBs from soil to the freatic level.

Rotating-disk sorptive extraction of nonylphenol from water samples.

In this study the sorption of nonylphenol was implemented on a rotating Teflon disk coated with a PDMS film on one of its surfaces. In this way, the disk, which has a high surface area, contacts only the liquid sample, which can be stirred at higher velocity than with the stir bar used in stir-bar sorptive extraction (SBSE), without damaging the phase while at the same time facilitating analyte mass transfer to the PDMS surface. We refer to the procedure as rotating-disk sorptive extraction (RDSE). Extraction variables such as disk rotational velocity, extraction time, and surface area of PDMS film were studied to establish the best conditions for extraction. With increasing rotational velocity, the amount of extracted analyte significantly increases because the stagnant layer concomitantly decreases. On the other hand, the extracted amount concomitantly increases with extraction time, reaching equilibrium at approximately 20 min, which can be reduced to 10 min when the surface area of PDMS increases from 1.74 to 6.97 cm(2). Precision of the method was determined by using the same disk (n=6) and different disks (n=3), showing relative standard deviations for the analyte of 3.7% and 10%, respectively. The detection limit of the method was 0.09 microg/L NP, defined at a signal to noise ratio of 3. The method was applied to a real sample, achieving quantitative recovery. The PDMS phase on the disk could be used for at least 50 experiments. In any case, replacement of the PDMS film on the disk is very easy and inexpensive, as compared to the commercial alternative SBSE.