# Value of Rapid Mineralogical Monitoring of Copper Ores

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Materials

#### 2.2. Method

^{1D}Medipix3 detector (Malvern Panalytical B.V., Almelo, The Netherland) with an active length of 5.54 °2θ allows a scan acquisition time of a few minutes, and the results are immediately analyzed by automatic Rietveld routines installed in the Minerals Edition of Aeris.

#### 2.3. Analysis

## 3. Results

#### 3.1. Minerals Present in the Blends, Tailings and Concentrates

#### 3.2. Details on Chalcopyrite and Tetrahedrite Monitoring

_{2}, including its most intense peak at 2θ = 34.2°. The latter peak is due to the diffraction of Kα = 1.79 Å X-rays and chalcopyrite atomic planes with distance d = 3.03 Å (space group I-42d, a = b = 5.290 Å and c = 10.426 Å, miller indexes h = 1, k = 1 and l = 2). With a background count of 1680 counts and a peak intensity of 572 counts, the chalcopyrite diffraction peak has a signal-to-noise ratio SNR ≈ 14.4, which passes the criteria of limit of detection (SNR)

_{LOD}= 3 and limit of quantification (SNR)

_{LOQ}= 10 [13]. One advantage of Rietveld refinement over classical straight line calibration method is the proper intensity extractions of overlapping peaks, such as chalcopyrite peak at 2θ = 34.2° and gypsum peak at 2θ = 34.0°, see Figure 3.

#### 3.3. Rietveld Refinements and Trends

^{−0.5}] [13]. A SNR > 10 is the minimum prerequisite to accurately quantify a mineral in a mixture. The accuracy of the Rietveld results depends also on (1) particles statistic, (2) degree of overlap with peaks from other phases, (3) preferential orientation, (4) overgrinding and (5) microabsorption.

^{2}values of 0.99, 0.98 and 0.96, and slope values of 0.99, 1.05 and 0.94. The automatic Rietveld analysis gave accurate results with precision that depends on the mineral species and their concentrations. For instance, the precision of zinc calculated from Rietveld greatly decreases when the zinc content is below 0.2 wt%, which corresponds to a sphalerite concentration of 0.3 wt%, Figure 7. The relatively higher LOD of sphalerite is due to its strong overlap with pyrite diffraction peaks.

#### 3.4. Statistical Methods—Principal Component Analysis (PCA): Clustering of Copper Concentrates

_{9}S

_{5}and covellite CuS.

#### 3.5. Statistical Methods—Partial Least Squares Regression (PLSR): Predicting Antimony in Copper Concentrates

## 4. Discussion

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Gleisberg, D. Mineral Commodity Summaries; Government Printing Office: District of Columbia, WA, USA, 2021; Volume 3. [CrossRef]
- Leja, J. Surface Chemistry of Froth Flotation; Springer: Boston, MA, USA, 1981. [Google Scholar] [CrossRef]
- Rietveld, H.M. Line Profiles of Neutron Powder-Diffraction Peaks for Structure Refinement. Acta Crystallogr.
**1967**, 22, 151–152. [Google Scholar] [CrossRef] - Paine, M.; König, U.; Staples, E. Application of Rapid X-Ray Diffraction (Xrd) and Cluster Analysis to Grade Control of Iron Ores. In Proceedings of the 10th International Congress for Applied Mineralogy (ICAM); Springer: Berlin/Heidelberg, Germany, 2012; pp. 495–501. [Google Scholar] [CrossRef]
- König, U.; Norberg, N.; Gobbo, L. From Iron Ore to Iron Sinter—Process Control Using X-Ray Diffraction (XRD). In Anais dos Seminários de Redução, Minério de Ferro e Aglomeração; Editora Blucher: São Paulo, Brazil, 2017; pp. 146–153. [Google Scholar] [CrossRef] [Green Version]
- König, U.; Degen, T.; Norberg, N. PLSR as a New XRD Method for Downstream Processing of Ores:—Case Study: Fe2+ Determination in Iron Ore Sinter. In Powder Diffraction; Cambridge University Press: Cambridge, UK, 2014; Volume 29, pp. S78–S83. [Google Scholar] [CrossRef]
- König, U.; Gobbo, L.; Reiss, C. Quantitative XRD for Ore, Sinter, and Slag Characterization in the Steel Industry. In Proceedings of the 10th International Congress for Applied Mineralogy (ICAM); Springer: Berlin/Heidelberg, Germany, 2012; pp. 385–393. [Google Scholar] [CrossRef]
- Galluccio, S.; Pöllmann, H. Quantifications of Cements Composed of OPC, Calcined Clay, Pozzolanes and Limestone. In RILEM Bookseries; Springer: Berlin/Heidelberg, Germany, 2020; Volume 25, pp. 425–442. [Google Scholar] [CrossRef]
- Tornos, F. Environment of Formation and Styles of Volcanogenic Massive Sulfides: The Iberian Pyrite Belt. Ore Geol. Rev.
**2006**, 28, 259–307. [Google Scholar] [CrossRef] - Degen, T.; Sadki, M.; Bron, E.; König, U.; Nénert, G. The High Score Suite. In Powder Diffraction; Cambridge University Press: Cambridge, UK, 2014; Volume 29, pp. S13–S18. [Google Scholar] [CrossRef] [Green Version]
- Gates-Rector, S.; Blanton, T. The Powder Diffraction File: A Quality Materials Characterization Database. Powder Diffr.
**2019**, 34, 352–360. [Google Scholar] [CrossRef] [Green Version] - Rietveld, H.M. A Profile Refinement Method for Nuclear and Magnetic Structures. J. Appl. Crystallogr.
**1969**, 2, 65–71. [Google Scholar] [CrossRef] - Shrivastava, A.; Gupta, V. Methods for the Determination of Limit of Detection and Limit of Quantitation of the Analytical Methods. Chronicles Young Sci.
**2011**, 2, 21. [Google Scholar] [CrossRef] - De Jong, S. SIMPLS: An Alternative Approach to Partial Least Squares Regression. Chemom. Intell. Lab. Syst.
**1993**, 18, 251–263. [Google Scholar] [CrossRef] - Lance, G.N.; Williams, W.T. A General Theory of Classificatory Sorting Strategies: 1. Hierarchical Systems. Comput. J.
**1967**, 9, 373–380. [Google Scholar] [CrossRef] [Green Version] - König, U. Process Monitoring for Iron Ore Pelletizing—XRD in Combination with PLSR. In Iron Ore; AUSIMM: Perth, Australia, 2019; pp. 121–126. [Google Scholar]

**Figure 1.**XRD scan of a flotation feed. The diffraction pattern of this feed is the representative scan as defined by the cluster analysis.

**Figure 2.**Characteristic XRD scans of the ore blend used as feed for the flotation cell, tailings and concentrates. XRD scans of tailings and concentrates are here vertically shifted by 3000 and 6000 counts for clarity.

**Figure 3.**XRD pattern of a tailing sample with 0.16 wt% of copper in the range 26–38 °2θ. Upper curve represents measured data, lower curve in blue represents Rietveld contribution of 0.45 wt% chalcopyrite. Vertical bars are peak positions of crystalline phases.

**Figure 4.**XRD pattern of ore concentrate in the range 26–38 °2θ. Upper curve represents measured data, lower curve in gray represents Rietveld contribution of 1.6 wt% of tetrahedrite. Vertical bars are peak positions of crystalline phases.

**Figure 5.**Mineralogical variability of ore blends for the major gangue minerals as determined by XRD and Rietveld refinement.

**Figure 6.**Mineralogical variability of ore blends for minor mineralogical phases, as determined by XRD and Rietveld refinement.

**Figure 7.**Agreement between chemical compositions obtained from ICP and Rietveld refinement. Zero values obtained by Rietveld are set to 0.001 for graphical purpose.

**Figure 8.**Eigenvalues plot explaining how the data variance is described by the most relevant principal components (PCs).

**Figure 9.**Principal Components Analysis plot of XRD scans of copper concentrates. Each scan is represented by a sphere with a radius proportional to its pyrite content. The four clusters and the one outlier are visualized in different colors. *** representative XRD scan of each cluster, + most different XRD scans within a cluster.

**Figure 10.**PLSR plot of predicted values and reference values of antimony concentration in copper concentrates. The plot contains both training and validation data of the model.

**Figure 11.**PLS regression coefficients for antimony concentration in concentrates. The coefficients identify the parts of the diffractogram contributing most to the PLSR regression model.

**Table 1.**Average chemical composition of the available samples. #Samples represents the number of XRD samples analyzed per each material type.

Material Type | #Samples | Cu (%) | S (%) | Zn (%) | Pb (ppm) | Fe (%) | As (ppm) | Sb (ppm) | Bi (ppm) |
---|---|---|---|---|---|---|---|---|---|

Blends | 87 | 0.46 | 5.11 | 0.11 | 157 | 12.3 | 190 | 58 | 29 |

Tailings | 88 | 0.07 | 4.58 | 0.06 | 115 | 11.98 | 178 | 17 | 26 |

Concentrates | 90 | 20.33 | 35.90 | 3.26 | 2800 | 30.36 | 1260 | 2443 | 230 |

**Table 2.**Average weight percentages of major minerals as determined by Rietveld refinements. The mineral variability within the same material type is reported as the absolute standard deviation (1σ) in between brackets.

Material | Agreements Index | Minerals Quantification (wt%) | ||||||||
---|---|---|---|---|---|---|---|---|---|---|

Cu and Zn Bearing Minerals | Gangue | |||||||||

Rwp | GOF | Chalcopyrite | Sphalerite | Tetrahedrite | Pyrite | Quartz | Chamosite | Muscovite | Siderite | |

Blends | 7 (1) | 3.4 (0.4) | 1.1 (0.3) | 0.1 (0.1) | 0.0 (0.0) | 5 (2) | 53 (4) | 30 (5) | 6 (2) | 0.8 (0.6) |

Tailings | 7 (1) | 3.5 (0.5) | 0.2 (0.1) | 0.0 (0.0) | 0.0 (0.0) | 5 (2) | 53 (5) | 31 (6) | 6 (2) | 0.8 (0.3) |

Concent. | 3.5 (0.4) | 2.2 (0.2) | 55 (8) | 5.1 (1.6) | 0.9 (0.8) | 24 (9) | 4 (2) | 4 (2) | 0.2 (0.2) | 0.9 (0.3) |

**Table 3.**Average weight percentages of selected minerals of different clusters of copper concentrate, as determined by Rietveld refinements. The mineral variability within the same cluster is reported as the standard variation in between brackets. #Samples represents the number of samples included in each cluster.

Cluster Color | #Samples | Cluster Features | Minerals Quantification [wt%] | |||
---|---|---|---|---|---|---|

Chalcopyrite | Pyrite | Quartz | Chamosite | |||

Blue Cluster | 9 | High CuFeS_{2}/Low FeS_{2} | 60 (7) | 12 (2) | 10 (2) | 6 (1) |

Gray Cluster | 53 | High CuFeS_{2}/Medium FeS_{2} | 63 (4) | 20 (4) | 4 (1) | 2 (1) |

Green Cluster. | 25 | Medium CuFeS_{2}/Medium FeS_{2} | 50 (5) | 33 (7) | 4 (1) | 1 (1) |

Brown Cluster | 2 | Low CuFeS_{2}/High FeS_{2} | 40 (13) | 45 (14) | 4 (1) | 2 (1) |

Black Outlier | 1 | Secondary copper minerals | 35 | 45 | 3 | 1 |

**Table 4.**2theta angles in the diffraction pattern of copper concentrates with the highest regression coefficient for the PLSR model of antimony, and diffraction peaks of the corresponding phase.

°2θ | Phase Assignment | H K L Miller Indexes | I [%] from 01-074-3633 |
---|---|---|---|

18.9 | Tetrahedrite | 0 2 0 | 2.5 |

28.2 | Tetrahedrite | 0 2 2 | 7.1 |

34.7 | Tetrahedrite | 2 2 2 | 100 |

40.4 | Tetrahedrite | 0 4 0 | 21.2 |

42.9 | Tetrahedrite | 0 3 3 | 7.7 |

45.4 | Tetrahedrite | 0 4 2 | 1.6 |

50.0 | Tetrahedrite | 2 4 2 | 1.6 |

52.2 | Tetrahedrite | 1 4 3 | 6.1 |

58.4 | Tetrahedrite | 0 4 4 | 39.8 |

64.2 | Tetrahedrite | 1 6 1 | 6.3 |

69.7 | Tetrahedrite | 2 6 2 | 20.3 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Pernechele, M.; López, Á.; Davoise, D.; Maestre, M.; König, U.; Norberg, N.
Value of Rapid Mineralogical Monitoring of Copper Ores. *Minerals* **2021**, *11*, 1142.
https://doi.org/10.3390/min11101142

**AMA Style**

Pernechele M, López Á, Davoise D, Maestre M, König U, Norberg N.
Value of Rapid Mineralogical Monitoring of Copper Ores. *Minerals*. 2021; 11(10):1142.
https://doi.org/10.3390/min11101142

**Chicago/Turabian Style**

Pernechele, Matteo, Ángel López, Diego Davoise, María Maestre, Uwe König, and Nicholas Norberg.
2021. "Value of Rapid Mineralogical Monitoring of Copper Ores" *Minerals* 11, no. 10: 1142.
https://doi.org/10.3390/min11101142