This “**Mineralogy in Mining**” blog series started with a general overview of the value of mineralogical monitoring for an efficient ore beneficiation. During the last decades high-grade iron ore deposits, particularly in Western Australia, required hardly any downstream processing.

Iron ores were simply crushed and shipped as lump ore for iron making. However, with decreasing ore grades in existing high-grade deposits and exploration of new lower-grade deposits, the need for additional beneficiation steps is on its way to becoming the new standard for the iron ore industry. Depending on the mineralogy of the ore, in some cases, crushing and simple screening is sufficient, in other cases more complex concentration processes (such as washing, magnetic separation or even flotation) may be required.

Typical minerals in lateritic iron ore deposits are goethite, hematite and impurities such as kaolinite, other clay minerals, carbonates or silicates. The amount of the impurities defines if:

- ore can be directly shipped (DSO),
- ore requires beneficiation to meet the required specifications or
- ore quality is so low that it is not excavated at all.

In the first blog about the “Value of mineralogical monitoring”, we concluded that X-ray diffraction (XRD) is a fast, versatile and accurate tool for mineralogical analysis, which can be easily implemented in process environments and mine operations. In the following case study, we discuss the added value of XRD for the mineralogical analysis of lateritic iron ore.

#### Accurate mineralogical analyses of iron ore

For this case study, seven samples of lateritic iron ore were analyzed. All samples were prepared as pressed pellets and were measured on Aeris Minerals tabletop diffractometer with a scan time of 10 minutes, followed by an automatic quantitative phase analysis.

Figure 1 shows an example of full-pattern XRD analysis of one iron ore sample.

Any XRD pattern is a set of diffraction peaks of different intensities, located at certain diffraction angles (2q), specific to a certain mineralogical phase. Peak positions enable identification of existing phases. The relative intensities of each mineral contribution to the XRD pattern allows to quantify the relative amount of each mineral present using the full-pattern Rietveld method [1].

In Figure 1, the lateritic iron ore sample primarily consists of goethite, with only 28% of hematite, 1.4% of quartz, and minor amounts of carbonates and clays. Furthermore, this sample contains over 12% of amorphous phase.

Comparing the XRD result shown in Figure 1 with the rest of the samples (Figure 2), it is clearly visible that the amount of clay minerals, quartz and carbonates varies from sample to sample. Other phases, like rutile and magnetite are present as well. There is a clear correlation between the amount of crystalline goethite and amorphous content. Based on such a detailed mineralogical analysis, the different ore grades can be mixed to and optimal blend and further downstream processing can be adapted and optimized to save costs.

#### Fast and efficient ore monitoring using modern statistical methods

Accurate quantitative mineralogical analyses is not the only added value of XRD. Using known stoichiometries of the minerals present, the Fe^{2+} content can be directly calculated from the XRD pattern, eliminating the need for time-consuming wet chemistry tests. Furthermore, a data set representing all variations of a mining operation or during processing can be used to build a statistical model (Principle Component Analysis, PCA) for clustering different iron ore grades based on XRD raw data using the software package HighScore Plus. Partial Least Square Regression, PLSR [2,3,4] can be used to predict process parameters for iron ore beneficiation directly from the XRD raw data, avoiding time and cost intense chemical and physical tests. An existing PLSR model of can be applied for fast definition of ore grades for new unknown samples.

An example for a statistical model (PCA analysis) of different iron ore grades is shown in Figure 3 bottom, [5]. The samples are characterized by a varying hematite/goethite ratio, clay and quartz content (Figure 3, top).

The PCA analysis of all XRD pattern shows four 4 clusters. Additionally, the total Fe content is plotted (size of the individual spheres representing the individual samples).

We clustered the patterns based on similarities and combined the cluster model with the quantitative mineralogical analyses results (Figure 3, bottom). Each small solid sphere corresponds to an XRD measurement of one sample, size of the sphere represents the amount of total iron in a corresponding sample. Samples with similar mineralogy are grouped in a cluster, shown by a semi-transparent sphere. Our statistical model consists of four clusters, characterized by different mineralogical content and hence different ore grade: green for low grade (low hematite), red for medium grade, blue and yellow for high grade, with yellow showing the sample with the highest hematite content.

To use this model for easy assessment of ore grade, an XRD pattern for a new sample should be measured and inserted into the model. It will be automatically assigned to one of the clusters, corresponding to a certain ore grade. Phase analyses can be done as well; however, is not required.

#### On-line control of clay content in copper ore

In the introduction blog, we discussed the advantages of near-infrared (NIR) spectroscopy for online mineralogy monitoring. Not all minerals, commonly present in iron ore are spectrally active, however, such minerals like goethite, clays are spectrally active, and can therefore be easily identified and quantified using an on-line NIR over-the-belt analyzer. Real-time monitoring of goethite content can be used for e.g. effective ore blending. On-line control of the content of soft minerals in the run-of-mine will improve efficiency of beneficiation processes, prevent possible equipment blockage and other common issues associated with the presence of large quantities of soft minerals in the ore.

To summarize, with the decreasing iron ore grade, the chemical analyses alone is no longer sufficient for cost-effective mine operation. Mineralogy of iron ore deposits, directly influencing the outcome of any beneficiation step, must be considered. X-ray diffraction (XRD) is an indispensable tool for fast, accurate and tailored mineralogical analyses. XRD can be used for the quantitative assessment of full mineralogical composition at every step of ore-to-metal process. New statistical methods (e.g. clustering, partial least square regression) opened up new possibilities for efficient use of large data sets, enabling quick and easy assessment of the ore grade, deviations in the process, as well as extraction of relevant process parameters directly from the diffraction data, eliminating the need for additional time-consuming, costly tests.

To learn more about this topic, please watch our **recorded webinar **or check the following scientific papers.

__Paine, M., König, U. & Staples, E. (2011)____König, U., Gobbo, L. & Macchiarola, K. (2011)____König, U. (2013)__

Stay tuned for our next blog on the topic, where we will discuss the added value of XRD for process control in sinter production.

**References:**

- [1] H.M. Rietveld (1969): A profile refinement method for nuclear and magnetic structures, J. Appl. Cryst., 2, 65 – 71.
- [2] H. Lohninger (1999): Teach Me Data Analysis, Springer-Verlag, Berlin-New York-Tokyo, ISBN 3-540-14743-8.
- [3] G.N. Lance, W.T. Williams (1966): A general theory of classification sorting strategies 1., Hierarchical systems, Comp. J., 9, 373 – 380.
- [4] S. de Jong (1993): Simpls, An alternative approach to partial least square regression, Chemometrics Intell. Lab. Syst., 18(3), 251-263.
- [5] Paine, M., König, U. & Staples, E. (2011): Application of rapid X-ray diffraction (XRD) and cluster analysis to grade control of iron ores. Proceedings 10
^{th}International Congress for Applied Mineralogy (ICAM), 1-5 August 2011, Trondheim, Norway, 495-501. - [6] König, U., Gobbo, L. & Macchiarola, K. (2011): Using X-Ray Diffraction for Grade Control and Minimizing Environmental Impact in Iron and Steel Industries. Proceedings IRON ORE conference / Perth, WA, July 2011, 49-56.
- [7] König, U. (2013): Application of X-ray diffraction to iron ores – Potential implications for grade control and downstream processing. Proceedings IRON ORE conference / Perth, WA, July 2013, 275-280.