1. Introduction
The main zirconium minerals in magmatic systems are baddeleyite (ZrO
2) and zircon (ZrSiO
4). They occur as common accessory phases in a wide range of compositions of igneous rocks [
1,
2,
3,
4,
5,
6]. When developing placer titanium-zircon deposits, the commercially important heavy minerals extracted from mineral sands are zircon (ZrSiO
4), rutile (TiO
2), ilmenite (FeTiO
3), and products of ilmenite alterations (including pseudorutile and leucoxene), as well as smaller amounts of the rare-earth minerals of monazite ([Ce, La, Th]PO
4), which determine their radioactivity [
1]. The value of zircon and rutile is higher than that of ilmenite [
1], but since the zircon content in most deposits is much higher than that of rutile, zircon tends to be the main economic factor in the development of a mineral sand deposit [
1]. Zircon is a tetragonal crystal, in which zirconium is isomorphically substituted for hafnium (zirconium content in zirconium concentrates is 1–4%, 2% on average). Uranium, thorium, and lanthanides are part of the phosphate mineral monazite, which is often present in deposits together with zircon [
7,
8,
9].
Zircon is chemically stable and thermally resistant to thermal shock due to its very low coefficient of thermal expansion (~4.1 × 10
−6 K
−1) and low thermal conductivity, ranging from 5.1 W/(m∙K) at room temperature to 3.5 W/(m∙K) at 1000 °C [
2,
7]. In addition, high-purity sintered zircon retains the flexing endurance of up to 1400 °C [
7,
10,
11,
12]. Having these characteristics, zircon is a good refractory material [
13]. Silica and zirconium dioxide obtained after zirconium enrichment are widely used in the production of heat-insulating materials, catalysts, pharmaceutical products, electronic gadgets, thin-film substrates, pigments, welding rod coatings, silencers, refractory products, and abrasives [
14,
15,
16,
17,
18,
19,
20]. Ceramics account for 50% of zircon supply, followed by refractories and foundries with approximately 30% of the demand, and the remaining 20% of the demand consists of zirconium, zirconium chemicals, and metal [
18,
21]. Zirconium can also be used to produce zirconium-based alloys in nuclear power reactors as a cladding material due to zircon’s high mechanical strength, corrosion-resistant properties, and a low thermal-neutron capture cross-section [
18]. Due to their chemical and biological neutrality, zirconia ceramics can be used in medicine as a material to manufacture prostheses [
22,
23,
24,
25,
26,
27,
28].
For the successful use of zircon or zirconium dioxide, interfering impurities must be removed as they can negatively affect performance, thereby reducing the value and limiting industrial applications. The main interfering impurities are titanium, iron, and thorium. As the concentrate contains uranium and thorium, it has a radioactivity, which depends on the concentration of uranium and thorium in zircon. In [
29,
30,
31,
32,
33,
34], the possibility of using zircon as an adsorbent for uranium was investigated. Currently, the existing main applications of zircon do not require removal of the radioactivity, provided that the radioactivity is 70 kBq/kg [
35].
Electrostatic and electromagnetic separation, as well as gravity (screw separators) and hydraulic methods (separation in hydrocyclones), are the most commonly used methods for the beneficiation and removal of impurities from rutile-zircon ores. At the same time, the use of pneumatic and flotation methods for beneficiation and primary clarification is also increasing [
36].
This paper investigates the possibility of supplementary enrichment and reduction of the zirconium concentrate obtained from the Obukhov Mining and Processing Plant by the methods of the gravitational and electrostatic separation in the Obukhov titanium-zirconium deposit.
2. Materials and Methods
The object of the study is zircon concentrate of the Obukhovsky GOK. Sampling was carried out in July 2021 in the form of zircon concentrate as a free-flowing pink-cream-colored sand with a rounded, elongated, and elliptical grain shape. The declared grain size class was 0.071 + 0.044 mm.
The granulometric and elemental composition of the original zircon concentrate of the Obukhovsky GOK is presented in
Section 3.1. Primary chemical analysis of the samples was carried out using a portable energy dispersive spectrometer Niton XL5. Furthermore, the chemical composition of the samples was investigated by means of an XRF-1800 wavelength dispersive X-ray fluorescence spectrometer. The mass content of the elements was recalculated relative to the mass content of oxides of the corresponding elements according to the equation:
where
is the molar mass of the element i;
is the index of the element i in the oxide;
is the index of oxygen in the oxide with the element i;
is the elemental mass content.
The samples were investigated using a Hitachi TM-3000 scanning electron microscope at an accelerating voltage of 15 kV under charge-off mode conditions (the electron gun was 5 × 102 Pa, the sample chamber was 30–50 Pa) equipped with a QUANTAX 70 attachment for elemental analysis. The approximation was ×500. The exposure time was 240 min.
The analyses were carried out using the equipment of the Tomsk Regional Core Shared Research Facilities Center of National Research Tomsk State University. The Center was supported by the Ministry of Science and Higher Education of the Russian Federation, grant no. 075-15-2021-693 (no. 13.RFC.21.0012).
For all the obtained powders, their activity was measured using a dosimeter-radiometer DRBP-03. When measuring, a container with a fixed area (in our case, a ceramic boat) was used to uniformly distribute 25 g of the measured substance over the entire volume. Before working with the fractions, the background was measured. The number of measurements of the background was 50; the number of measurements of the fraction was 30.
The actual density of the samples was determined using a pycnometer with a 50-mL capacity. Each part of the sample was poured into a clean, dried, and pre-weighed pycnometer, and then they were weighed together with the sample. Then, distilled water was added to the pycnometer so that the pycnometer was filled up to about 2/3 of its volume; the contents were stirred and placed at a slightly inclined position in a water bath. The content of the pycnometer was boiled for 15–20 min to remove air bubbles.
After air removal, the pycnometer was cooled to room temperature (23 °C), filled to the mark with distilled water, and weighed. Then, the pycnometer was emptied, washed, filled to the mark with distilled water, and weighed again.
The true density of zircon (ρ
t), g/cm
3, was calculated according to the equation:
where
m is the mass of the pycnometer with sand, g;
m1 is the mass of the empty pycnometer, g;
m2 is the mass of the pycnometer with distilled water, g;
m3 is the mass of the pycnometer with sand and distilled water after removing air bubbles, g;
ρw is the density of water, g/cm3.
To determine the bulk density of zircon in the standard unconsolidated state, the sample was poured using a scoop into a pre-weighed measuring cylinder at a height of 10 cm from the upper edge up to the cone formation over the top of the cylinder. The cone without sealing the sample was removed at the brim of the vessel with a metal ruler, after which the vessel with the sample was weighed:
where
m is the mass of the measuring vessel, kg;
m1 is the mass of the measuring vessel with the sample, kg;
V is the capacity of the vessel, m3.
Separation tests on the zircon concentrate in the centrifugal air classifier were carried out using the laboratory setup shown in
Figure 1.
The device for the centrifugal air classification of powders consists of a casing of the classifier (1), a collector (2) with nozzles (3), a rotor (4) with a drive (5) and disk elements that form a separation zone (6), and a batcher (7) with a neck (8). There is also an air distributor (9) with a shutter (10), the first outlet pipe and a flow meter (11), the second outlet pipe and a flow meter (12). The device also includes an outlet pipe of a classifier (13) connected to a cyclone (14), a filter (15), a high-pressure blower (16), a bunker for coarse fraction (17) and a bunker for fine fraction (18), and a pulsator (19).
In this device, a high-pressure fan creates an air stream that enters the air diffuser. Part of the air flow comes through the air supply tube to the dispenser, and the main flow moves through the tube to the collector. The initial powder together with the air flow enters the central channel in the middle of the separation zone. The preliminary classification of the initial powder occurs under the influence of centrifugal forces created by the rotor and oppositely directed forces of aerodynamic resistance of particles of the main swirling flow. As the particles pass through the separation zone, they are separated by size [
37].
For the zirconium concentrate separation, the airflow rate into the separation zone was 2000 Nm3/h. The concentrate flow rate was set to 20 kg/h. The diameter of the separator rotor was 200 mm.
At the first stage of separation, the initial zircon concentrate was loaded into the centrifugal classifier. At this stage the rotor speed was set to be 3000 rpm. After a complete separation of the concentrate from the bunker, used for the fines fraction, we extracted Fraction 1; the material from the bunker for the coarse fraction was reloaded into the batcher and the process was repeated again. Fraction 2 and Fraction 3 were also extracted at the same rotor speed (3000 rpm).
The rotor speed was then reduced to 980 rpm and the process was repeated. After such separation, Fraction 4 was obtained from the fines hopper. The material from the bunker for the coarse fraction was then separated again at a rotor speed of 600 rpm. The material in the bunker for the fine fraction was called Fraction 5, and the material in the bunker for the coarse fraction was called Fraction 6.
Magnetic separation was carried out using the laboratory magnetic separator EVS-10/5. This separator has the following features: an initial material throughput rate of up to 80 g/min, a maximum magnetic induction in the working area of 1.7 Tl, a roll diameter of 100 mm, a roll width of 50 mm, and a roll speed of 70 rpm. The initial material grain size was not more than 2 mm.
The magnetic separation of the zircon concentrate was performed at a feed rate of 5 g/min and the magnetic induction of 1.3 Tesla.
3. Results
3.1. Examination of the Initial Zircon Concentrate
The chemical composition of the initial zircon concentrate obtained from the Obukhov deposit is presented in
Table 1. The sample activity was 10.3 ± 0.6 kBq/kg. Such level of activity can pose problems for the sale and transport of the mentioned raw materials. For example, in Russia and Kazakhstan, there are restrictions on the transport of raw materials containing more than 10 kBq/kg of naturally occurring radionuclides.
The distribution of the elements in the initial zirconium concentrate and each obtained fraction after the air separation were determined by energy dispersive X-ray spectroscopy.
Figure 2 shows the data on the distribution of the elements in the initial zircon concentrate.
Figure 2 shows that titanium is not included in zircon. Iron forms agglomerates with titanium and chromium. Traces of hafnium, bromine, barium, and carbon were also found. The high content of iron, titanium, and chromium in the majority of fractions was noted. The particle size distribution in the initial zircon concentrate is shown in
Figure 3.
Figure 3 shows that the particles in the zircon concentrate are not uniformly distributed. The fraction of 40–100 microns prevails in the concentrate containing more than 83%. The average particle size of the initial zircon concentrate is about 70 μm.
3.2. Air Classification
Air separation was carried out on a zircon sample weighing 10 kg. As a result of the air separation, six fractions were obtained. The fractions were marked as they left the air-centrifugal separator. The yield of fractions and their physical characteristics are presented in
Table 2.
Table 2 shows that the centrifugal air separation allowed for separating the fractions by increasing the bulk and actual density. At the same time, the average particle size does not proportionally change. It is related to the separation of the raw materials not only by size, but also by density. The density of titanium minerals is much lower than that of zirconium minerals.
Figure 4 presents the data on the distribution of the elements in fraction 1 after the centrifugal air classification of the zircon concentrate.
Figure 4 shows that iron and chromium are uniformly distributed throughout the fraction due to the small size of the particles. Zirconium, titanium, and aluminum are contained in separate grains and are not included in agglomerates of other elements. Traces of monazite were not detected due to its low content. The particle size distribution in fraction 1 of the zircon concentrate is shown in
Figure 5.
According to the distribution, the main part of fraction 1 is particles of up to 20 microns. This indicates a good separation of the elements by size using a centrifugal air separator.
Figure 6 shows the data on the distribution of the elements in fraction 2 after centrifugal air classification of the zircon concentrate.
According to the results of the energy dispersion analysis (
Figure 6), iron and titanium are in the same agglomerate, which indicates their bonding in the form of ilmenite. Chromium is associated with iron in another grain and forms another phase with it. Bromine, carbon, and phosphorus are also detected and are not uniformly distributed. The detected phosphorus phase suggests the distribution of the monazite fraction. The distribution of phosphorus is identical to that of zirconium, which suggests that the monazite and zircon phases are connected. The average particle size of the fraction is about 45 µm. The largest particle size is 107 µm. The particle size distribution in fraction 2 of the zircon concentrate is shown in
Figure 7.
According to the distribution, the main part of fraction 2 is the particles of 2050 µm. In the fraction, there are also particles with sizes of more than 80 µm; large particles are represented by titanium minerals. The occurrence of large particles of titanium-containing minerals in fraction 2 is associated with their lower density relative to that of zircon.
Figure 8 shows the data on the distribution of the elements in fraction 3 after the centrifugal air classification of the zircon concentrate.
According to the distribution of the elements in fraction 3 (
Figure 8), iron is agglomerated with titanium and distributed over the entire surface. This suggests the impossibility of complete extraction of iron using the mechanical method. There are also agglomerates of titanium with a low iron content, probably formed by the TiO
2 phase (leucoxene-pseudo-rutile agglomerates). The distribution of phosphorus confirms the connectivity of monazite and zircon minerals. The average particle size of fraction 3 was 75 µm. The particle size distribution in fraction 3 of the zircon concentrate is presented in
Figure 9. The main part of fraction 3 consists of particles of 60–80 microns.
Figure 10 presents the data on the distribution of the elements in fraction 4 after the centrifugal air classification of the zircon concentrate.
The distribution of the elements in fraction 4 shown in
Figure 10 is similar to the distribution of the elements in fraction 3. The average particle size of the fraction is 75 μm. The particle size distribution in fraction 4 of the zircon concentrate is shown in
Figure 11. The distribution evidences that the main part of fraction 4 consists of particles of 60–90 μm. There are no particles less than 40 μm in the fraction.
Figure 12 shows the data on the distribution of the elements in fraction 5 after centrifugal air classification of the zircon concentrate.
The distribution of the elements in the grains of fraction 5, shown in
Figure 12, indicates that iron is agglomerated with titanium, barium, and aluminum. Manganese is uniformly distributed over the surface of the fraction. The average particle size is about 72 μm. The largest particle size is 112 μm. The particle size distribution in fraction 5 of the zircon concentrate is shown in
Figure 13.
The distribution demonstrates that the main part of fraction 5 consists of particles of 60–80 µm. The particles sizes over 100 μm are present in the fraction. Large particles are represented by titanium minerals, the absolute density of which is much less than that of zirconium minerals.
Figure 14 shows the data on the distribution of the elements in fraction 6 after the centrifugal air classification of the zircon concentrate.
Fraction 6 is characterized by the distribution of the elements shown in
Figure 14. Aluminum is included in the agglomerates of both titanium and zircon. Iron in this fraction is found to be weakly connected with titanium, indicating that titanium is mainly in the rutile form. The average particle size is about 79 μm. The largest particle size is 224 μm. The particle size distribution in fraction 6 of the zircon concentrate is shown in
Figure 15.
The distribution shows that the main part of fraction 6 is made up of particles of 60–80 microns. The particles sized over 200 µm are present in the fraction. Large particles are represented by titanium minerals, the absolute density of which is much less than that of zirconium minerals.
Table 3 shows the chemical composition of fractions 1–6.
Table 3 demonstrates that, at this stage, there is no clear relationship between the density of the fraction and its content of zirconium and titanium. However, the distribution of thorium, uranium, and rare earth elements is seen to depend on the density and particle size of the fraction. The content of these elements is higher the smaller the particle size of the fraction, and the higher its bulk density.
3.3. Magnetic Separation
The main impurities in the initial zircon concentrate and fractions after air separation are titanium, iron, and chromium. At the same time, magnetic minerals most often form a separate phase, so most of the impurities can be mechanically separated. It was decided to conduct magnetic separation. Magnetic separation was carried out using the laboratory magnetic separator EVS-10/5 “Mekhanobr-Tekhnika”. Experiments were conducted to determine the optimum process parameters. This was carried out by varying the magnetic induction value. For this purpose, magnetic separation was carried out at a feed speed equal to 5 g/min and currents in the winding of the electromagnetic system of 3, 6, 9, and 12 A. The experiments were carried out three times for each value of current; the average values of the experimental results are presented in
Table 4.
As a result of the studies, the value of magnetic induction in the separator’s working area was found to significantly affect the separation of the magnetic fraction. When conducting an additional magnetic separation at a magnetic induction of 1.7 T, a non-magnetic fraction of <1% was obtained, which indicates insignificant losses of the product.
The result of the experiments determined the fact that magnetic separation at a magnetic induction value of 1.3 Tesla was preferable. A feed rate of 5 g/min was selected.
In order to investigate the applicability of magnetic separation to the concentrate obtained from the Obukhovsky Mining and Processing Plant, the non-magnetic fraction obtained from the initial concentrate was additionally analyzed using the magnetic separator, including the separation parameters described above.
Figure 16 presents the data on the distribution of the elements in the non-magnetic part of the zircon concentrate.
After magnetic separation, shown in
Figure 16, there are small amounts of iron connected with titanium. The complete removal of iron is not achieved due to the increased loss of zircon with the increased magnetic separation time or magnetic flux. However, chromium is completely removed by the magnetic separation. Aluminum is detected, which is uniformly distributed throughout the concentrate. Traces of monazite in the form of thorium and phosphorus were not detected due to the low content. The particle size distribution in the non-magnetic fraction of the zircon concentrate is presented in
Figure 17. In the nonmagnetic part, the fraction of 60–100 microns prevails.
Then, magnetic separation of the fractions after the air separation was carried out. Consequently, after the air separation the concentrate is expected to be separated into the fractions with different physical characteristics, such as density, particle size, etc.
The yield of fraction 1 was too small to allow magnetic separation to be applied to this fraction. Furthermore, the separation was complicated by the extremely small particle size of the fraction.
The content in the initial concentrate of the magnetic part of fraction 2 is 0.56 wt.%; that of the non-magnetic part is 0.64 wt.%. The activity of the magnetic part of fraction 2 is 18.7 ± 0.6 kBq/kg; that of the non-magnetic part is 8.7 ± 0.6 kBq/kg.
The content in the initial concentrate of the magnetic part of fraction 3 is 2.45 wt.%; that of the non-magnetic part is 4.10 wt.%. The activity of the magnetic part of fraction 3 is 16.8 ± 0.6 kBq/kg; that of the non-magnetic part is 8.6 ± 0.6 kBq/kg.
The content in the initial concentrate of the magnetic part of fraction 4 is 14,21 wt.%; that of the non-magnetic part is 23,88 wt.%. The activity of the magnetic part of fraction 4 is 14.7 ± 0.6 kBq/kg; that of the non-magnetic part is 9.6 ± 0.6 kBq/kg.
The content in the initial concentrate of the magnetic part of fraction 5 is 12.21 wt.%; that of the non-magnetic part is 28.76 wt.%. The activity of the magnetic part of fraction 5 is 16.1 ± 0.6 kBq/kg; that of the non-magnetic part is 8.8 ± 0.6 kBq/kg.
The content in the initial concentrate of the magnetic part of fraction 6 is 2.92 wt.%; that of the non-magnetic part is 10.20 wt.%. The activity of the magnetic part of fraction 5 is 12.8 ± 0.6 kBk/kg; that of the non-magnetic part is 8.1 ± 0.6 kBk/kg.
The raised activity of the magnetic part of all the fractions is noted, reducing the activity of a concentrate part and receiving a product of low activity.
4. Discussion
Based on the analysis of the chemical composition and activity of the fractions, the fractions were grouped. As a result, three concentrates with different properties were obtained. The concentrates were obtained by the mechanical mixing of fractions with similar properties in proportions equal to their mass ratio in the initial concentrate. As a result, three products were obtained: a zircon concentrate of low activity; a titanium-containing fraction with low zirconium content, and a concentrate with considerable zirconium content and high activity. The zircon concentrate of low activity consists of the non-magnetic parts of fraction 2, fraction 4, fraction 5, and fraction 6. The choice of fractions is due to a higher zirconium content and low activity. The content of the fraction in the initial concentrate is 63,46 wt.%. The activity of the fraction is 5.8 ± 0.6 kBq/kg. The chemical composition of the zirconium concentrate of a low activity is presented in
Table 10.
As a result, a 44% reduction in activity was achieved. The content of zirconium and hafnium dioxide was 55.4 wt.%. In addition, the content of interfering elements such as iron, titanium, and chromium in the obtained concentrate was reduced.
The titanium-containing fraction with a low zirconium content consists of the magnetic parts of fraction 2, fraction 4, and fraction 6. The choice of fractions is conditioned by a low zirconium content and high activity. The content of the fraction in the initial concentrate is 17.77 wt.%. The activity of the fraction is 14.2 ± 0.6 kBk/kg. The chemical composition of the titanium-containing fraction is presented in
Table 11.
This fraction has the highest activity, probably due to the dissemination of thorium minerals (ThO2 content is increased by 80%) in ilmenite.
The concentrate with high zirconium content and high activity consists of fraction 3 and the magnetic part of fraction 5. The content of the fraction in the initial concentrate is 18.77 wt.%. The activity of the fraction is 12.8 ± 0.6 kBk/kg. The chemical composition of the concentrate is presented in
Table 12.
The concentrate with high zirconium content and high activity is not significantly different from the titanium-containing fraction. However, to reduce the loss of zirconium, we propose returning it to the screw separation stage for the additional separation of zircon.