Mode

Kimberlite consists of a fine-grained carbonate-serpentine mass (26.4%) with olivine inclusions (together 32.7%), pyroxene (33.5%) and accessory minerals (7.4%).

#### Grain Shape

Olivine and pyroxene grains are often of irregular shape, the sphericity coefficient for olivine is 0.23, and pyroxene is 0.48 (Figure 15a).

**Figure 15.** (**a**) Distribution of olivine grains according to the sphericity coefficient and (**b**) the orientation of the longest axis in grade

#### Specific Surface

The specific surface for both olivine and pyroxene is large (40.17 and 56.10 mm2/mm<sup>3</sup> respectively), which again indicates a complex morphology of grains.

#### Orientation

Olivine grains are slightly orientated in the rock volume (Figure 15b).

#### Clustering

The pyroxene and olivine grains are distributed in the rock very unevenly, they form separate clusters of 1.161–1.595 mm in size respectively (Figure 16a), and single aggregates up to 15 mm in size as well (Figure 16b).

**Figure 16.** (**a**) Olivine and pyroxene grains in the volume of kimberlite and (**b**) their distribution according to the maximum diameter.

#### Porosity

A sample of kimberlite is characterized by a minimal porosity: the total porosity is 0.13% and the opened porosity is 0.02%. In general, the pores are of 15–50 μm in size and have irregular shape (sphericity factor is 0.66). The void density is 12.81 mm<sup>−</sup>3.

#### **4. Discussion**

Analysis of the data obtained in the study of three different rock types using CT and QMA allows to reveal interesting textural and structural features. Thus, for example, both methods give sufficiently close measurements for such a parameter like the sizes of grains and mineral aggregates, especially if they sharply differ from the surrounding minerals in X-ray density. In this regard, the maximum correspondence of the data (0.647 and 0.594 mm) is observed for mica grains enclosed in a quartz-feldspar matrix of granodiorite, which is particularly well observed in Figure 17.

**Figure 17.** Grain size distributions of mica in granodiorite.

Thus, the CT method was proven to be an efficient tool to analyse materials with high X-ray density such as ore minerals (native elements, sulphides and metal oxides) in a matrix consisting of low X-ray density materials such as silicates (quartz, feldspar, olivine, pyroxenes, amphiboles, etc.) or carbonate matrix. The CT often analyzes not individual grains of minerals, but their aggregates. So, there may be slightly overestimated mean diameters of grains and their aggregates, which occurred with copper minerals in copper ore (see Figure 18). The calculated sizes of these minerals according to the CT data exceed the QMA data (0.381 and 0.277 mm, respectively), since one grain is taken as the aggregate of grains of copper minerals.

**Figure 18.** Grain size distributions of copper minerals in copper ore.

Similar distortions can occur for other morphometric parameters, such as, for example, the specific surface area of grains (see Table 3). In such cases, tomographic studies must be accompanied by additional petrographic and electron microscopic studies, but in this case, one of the main advantage of CT—the speed of the method—is lost.


**Table 3.** Comparison of the specific surface of minerals obtained with QMA and CT.

Some differences in estimated parameters can also be explained by the peculiarity of sample preparation for CT and QMA: by the CT method the total volume of a sample is scanned while for QMA three orthogonally oriented sections are investigated. Accordingly, the sample volume is different in both cases. For example, several large phenocrysts of olivine in the very heterogeneous kimberlite sample influenced significantly the values of the average morphometric parameters of the whole sample (see Figure 19).

**Figure 19.** Grain size distributions of olivine in kimberlite.

#### **5. Conclusions**

In the paper, a comparative analysis of textural-structural rock characteristic using QMA and CT methods is carried out. The undoubted advantage of the CT method is the possibility of 2D and 3D data visualization with the help of specialized programs such as CTVox and CTan, which help to analyse not only the grain and agglomerate sizes, but also their distribution in the total rock volume and relationships with each other. Although, possibilities of CT are not enough for measuring separate grains while their similar density. In these cases, the QMA results are much more representative.

The sample preparation for each method differs as well. Thus, CT is a non-destructive method and it is possible to analyse hand specimens. Whereas QMA requires more difficult and lengthy preparations (three orthogonally orientated thin sections or polished sections should be prepared).

So, it can be concluded that it is possible to use CT as a quick simple nondestructive method, as well as for more serious and difficult measurements, it is better to use QMA method. The analysis of the obtained data indicates a quite good repeatability of measurements, and consequently, the possibility of using the CT method in addition to other methods such as QMA for the purpose of studying the behaviour of rocks and ores during their beneficiation, in particular for the better understanding the important process of comminution.

**Author Contributions:** Conceptualization, H.L.; methodology, O.P. and I.T.; software, A.D.; validation, A.D.; O.P.; formal analysis, I.T.; H.L.; investigation, A.D.; O.P.; data curation, A.D.; O.P.; writing—original draft preparation, A.D.; O.P.; writing—review and editing, I.T.; H.L.; visualization, A.D.; O.P.; supervision, I.T.; H.L.; project administration, I.T.; H.L.; funding acquisition, I.T.; H.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** The CT Study was performed as part of the Joint German-Russian Project No 20-55-12002 of the Russian Foundation for Basic Research.

**Conflicts of Interest:** The authors declare no conflict of interest.
