3.1. Powder theoretical density
The theoretical density of the milled andesite basalt powder was determined to be 2.63 g/cm3. The theoretical density value obtained in this study is in good agreement with the available literature data. Namely, the theoretical density of the andesite basalt originated from the "Donje Jarinje" deposit site was estimated to be in the range from 2.60 to 2.63 g/cm3 [6, 17].
3.2. Powder particle size distribution and morphology
The particle size distribution of the andesite basal powder obtained by milling of the basalt rock is shown in Fig. 2. As can be seen, the values obtained during the powder analysis can be summarized as follows: d(0.1) = 0.465 µm, d(0.5) = 1.917 µm, d(0.9) = 11.803 µm. Bimodal distribution of the powder particles size is evident in the range from 0.2 to 12 µm and from 100 to 300 µm. The volume fraction of powder particles with dimensions ranging from 0.2 to 12 µm is determined to be 93.3%, while the volume fraction of powder particles with dimensions ranging from 100 to 300 µm is estimated to be 6.7%. The observed bimodal particle size distribution enables good powder compaction with the usage of a minimal amount of binder.
The morphology of the andesite basalt powder obtained by dry milling for 30 min is presented in Fig. 3. The presence of angularly shaped particles with a rough surface is clearly visible in the SEM micrograph. Moreover, agglomeration of small powder particles of approximately 0.5 µm in size can also be observed in Fig. 3.
3.3. Relative density
A macrographic overview of the ceramic samples obtained after sintering of the andesite basalt powder compacts in the temperature range from 1040 to 1080°C and in the sintering time range from 30 to 240 min is given in Fig. 4. It can be observed that the initial black color of the andesite basalt rock is transformed after milling into the grey color of the obtained basalt powder which is additionally transformed during sintering in the air to the reddish-brown color of the ceramic samples. The observed color change is induced by the chemical composition alterations, i.e. the transition of FeO into Fe2O3 according to the following Eq. (3):
4FeO + O2 = 3Fe2O3 (3)
The final surfaces of the sintered samples under an oxidizing atmosphere are reddish-brown color suggesting the presence of Fe3+ ions and the formation of hematite (Fe2O3) [32–36].
Based on Fig. 4 it can be seen that with an increase of the sintering temperature a change in the reddish-brown color from lighter to darker shades of the sintered ceramic samples occurs. The significant color change is clearly visible when the sintering temperature was increased from 1040 to 1050°C. However, it must be underlined that the research results indicated that the observed color alterations do not influence the quality of the obtained ceramic samples.
The effect of sintering temperature on density of ceramic samples obtained after 60 min long sintering of the andesite basalt powder is presented in Fig. 5. As can be seen, the lowest relative density of 92.9% was measured in ceramic samples sintered at 1040°C. The relative density of the obtained ceramics increases with an increase in sintering temperature reaching the maximum value of 99.5% in sample sintered at 1060°C. Therefore, this temperature will be designated as optimal. Based on the available literature data [17, 36, 37] and results obtained in the present study, this is the highest relative density achieved during basalt sintering. As the sintering temperature further increases, the relative density of the obtained ceramic samples decreases. Namely, at sintering temperatures of 1080°C the relative density of 95.8% is achieved. Research results presented in Fig. 5 show that small sintering temperature alterations have a significant impact on relative density.
The decrease in density of samples sintered above 1060°C can be explained with the help of "liquid channels". In the localized surface area of the ceramic samples sintered at 1070 and 1080°C the appearance of so-called "liquid channels" was noticed. The similar phenomenon was observed at the andesite basalt surface originated from the same locality . This phenomenon was also reported for the olivine basalt and explained with the occurrence of partial melting of formed crystal phases between the solidus and liquidus temperatures . Moreover, previous research in this field showed that a higher silicon, sodium, and potassium contents in the basalt composition led to an increase of the glassy phase content in the microstructure, which reduced the material density and increased its acid resistance .
Polished surfaces of ceramic samples sintered at 1040, 1060, and 1080°C for 60 min are presented in Fig. 6. Shown LOM micrographs demonstrate that the reason for a decrease in the relative density of samples sintered above 1060°C is the appearance of surface craters as a result of melting and evaporation of the material, Fig. 6c. The size of the observed surface craters is up to 50 µm.
The effect of sintering time on the density of ceramic samples sintered at 1060°C is presented in Fig. 7. A decrease of the sintering duration from 60 min to 30 min, as well as an increase of the sintering time to 120, 180, and 240 min, did not lead to an increase of relative density. Namely, obtained results indicated that the relative density of the sintered samples decreased with an increase in the sintering duration. Therefore, the optimal sintering temperature and time were found to be 1060°C and 60 min, respectively.
The effect of sintering parameters on the shrinkage of high-density ceramic samples is presented in Fig. 8. The effect of sintering temperature on shrinkage is given in Fig. 8a which shows that the shrinkage increases with an increase in sintering temperature, reaching the maximum value of 12.81% in samples sintered at 1060°C. Further increase in sintering temperature causes a decrease of shrinkage reaching the value of 10.83% in samples sintered at 1080°C.
The difference between the maximum and minimum shrinkage value of 1.98% was observed during the sintering in the specified temperature range. If compare graphs given in Fig. 5 and Fig. 8a one can conclude that the effect of sintering temperature on relative density and shrinkage is quite similar. This is expecting knowing that the increase of relative density as well as the increase of shrinkage is affected by pore elimination.
The effect of sintering time on the shrinkage of high-density ceramic samples sintered at 1060°C is presented in Fig. 8b. It is evident that a decrease in sintering time to 30 min, as well as an increase in sintering time to 120, 180, and 240 min, did not contribute to an increase of the shrinkage. The lowest shrinkage of 12.12% was measured in samples sintered for 240 min. The negligible difference of only 0.69% between the maximum and minimum shrinkage value could be detected at the optimal sintering temperature.
3.5. Weight loss during sintering
The effect of sintering parameters on the weight loss of the obtained high-density ceramic samples is presented in Fig. 9. From the research results presented in Fig. 9a, it can be observed that the highest weight loss of 3.10 wt.% was detected during sintering at 1080°C, while the lowest weight loss of 2.81 wt.% was measured after sintering at 1050°C. The slight difference between the maximum and minimum weight loss value of 0.29 wt.% was observed during the high-density ceramics sintering in the specified temperature range, although the detected weight loss was increased with an increase of the sintering temperature to 1080°C.
The effect of sintering duration on the weight loss of high-density ceramic samples attained during sintering of the andesite basalt powder at 1060°C is presented in Fig. 9b. Presented research results indicate that in the sintering time range from 30 to 240 min during the processing at the optimal sintering temperature the highest weight loss of 3.01 wt.% was evident after sintering for 30 min, while the lowest weight loss of 2.87 wt.% was shown for the ceramic samples sintered for 120 min. Namely, the negligible difference of only 0.14 wt.% between the maximum and minimum weight loss value could be detected after the high-density ceramics sintering at the optimal sintering temperature.
It must be emphasized that the weight loss in the range from 3.19 to 3.77 wt.% was also recorded during the sintering of the basalt originated from the Arkhangelsk region in Russia where the detected weight change variation was confirmed to be within 0.60 wt.% range . Having this in mind it can be concluded that the lower weight loss was recorded during the sintering in the air of the andesite basalt originated from the deposit site "Donje Jarinje", which was used in this study, then during the sintering of the basalt from the deposit site located in the Arkhangelsk region, which was reported in the literature. Previously published research results indicate that during the basalt heat treatment the weight loss can be expected due to the hygroscopic moisture evaporation, the chemically bound water removal, burnout of the organic impurities, dissociation of the present carbonates and sulfides, and removal of the gaseous inclusions [36, 38]. Therefore, it can be presumed that the andesite basalt from the "Donje Jarinje" deposit site shows higher purity and can be considered as a high-quality raw material for the high-density ceramics production.
3.6. Phase composition
Phase analysis of the starting andesite basalt material was performed using X-ray powder diffraction analysis to reveal its main mineralogical composition. For that purpose, the andesite basalt rock was crushed and milled into fine powder form. The obtained andesite basalt mineralogical composition is presented in Fig. 10. Based on the shown XRD pattern andesine, as an intermediate member of plagioclase mineral series ((Na,Ca)Al(Si,Al)Si2O8), was identified as a dominant mineral of the basalt rock used for the presented research. Moreover, labradorite ((Ca,Na)Al(Al,Si)Si2O8), as another mineral from the plagioclase series which is richer in Ca than andesine, was also identified. Furthermore, small amounts of other minerals, such as olivine (Mg,Fe)2SiO4) and magnetite (Fe3O4), as well as the common rock-forming pyroxene minerals, such as augite ((Ca,Na)(Mg,Fe,Al,Ti)(Si,Al)2O6) and clinohyperstene ((Mg,Fe)SiO3), were additionally identified in the basalt rock sample.
XRD patterns of all sintered high-density ceramic samples obtained during the present study along with starting andensite basalt powder are presented in Fig. 11. Since the sintered ceramic samples are pressed in the form of a pastille they had to be placed in a sample holder for the solid sample analysis. This holder is made of polyethylene terephthalate glycol (PETG) by using the Ender-5 3D printer, Creality, China. Due to the usage of the PETG sample holder small intensity hump (broad peak) in the low angle region (2θ range from 10° to 20°) can be observed on the recorded XRD patterns, indicating the presence of amorphous phase ascribed to PETG.
Detail analysis of the obtained XRD patterns, given in Fig. 11, shows that there is no significant difference between the recorded diffractograms of the ceramic samples sintered under different conditions indicating the absence of the phase transformations during these treatments. The only difference that can be singled out is the presence of hematite (Fe2O3) in the patterns of the high-density ceramic samples which is absent in the XRD pattern of the starting andesite basalt powder. The appearance of hematite during sintering in the air was expected (please see Eq. 3) and the assumption on the transition from FeO to Fe2O3, which was made based on the coloration of the sintered samples, is in this way confirmed.
SEM micrograph of the polished surface of high-density ceramic sample obtained during sintering at 1060°C for 60 min is given in Fig. 12. Isolated closed spherically-shaped pores, with dimensions up to 2 µm, are visible after sintering under these optimized conditions (Fig. 12) in contrast to the distinctly irregularly-shaped pores with mostly negative curvatures and a large surface-to-volume ratio present in the ceramic sample obtained during sintering at 1040°C for 60 min (Fig. 6a). Namely, it can be concluded that with an increase of the sintering temperature the pores inevitably present in the ceramic structure will tend to change their shape, i.e. pores will tend toward cylindrical to spherical morphologies and positive curvature with the lowest surface-to-volume ratio, clearly visible in Fig. 12. SEM micrograph, shown in Fig. 12, also indicates the non-homogenous structure of the ceramic sample obtained during sintering at 1060°C for 60 min in the presence of different aggregates and a clear boundary in between.
The effect of sintering temperature on the hardness of ceramic samples attained during sintering of the andesite basalt powder for 60 min is presented in Fig. 13a. A minimal hardness value of 4.8 GPa was achieved at an initial sintering temperature of 1040°C. As the sintering temperature increases, the hardness increases and reaches a maximal value of 6.7 GPa during sintering at 1060°C. With a further increase in the sintering temperature, the hardness value begins slightly to decrease and for the ceramic sample obtained at 1080°C the hardness value reaches 6.1 GPa. By comparing the obtained density values (Fig. 5) and the measured hardness values (Fig. 13a) the expected conclusion that the hardness correlates with the density of the sintered samples can be drawn and accordingly with an increase of the sample relative density its hardness increases.
The effect of sintering time on the hardness of ceramic samples attained during sintering of the andesite basalt powder at 1060°C is presented in Fig. 13b. It is clear from Fig. 13b that a decrease in sintering time to 30 min, as well as an increase in sintering time to 120, 180, and 240 min, did not contribute to an increase of the hardness value. Namely, when the sintering duration increases from 60 to 240 min then the hardness decreases linearly due to a decrease in the sintered ceramic samples density. The maximal hardness value of 6.7 GPa was attained at the optimal sintering temperature during sintering for 60 min.
3.8. Fracture toughness
Тhe effect of sintering temperature on the fracture toughness of ceramic samples attained during sintering of the andesite basalt powder for 60 min is presented in Fig. 14a. Results obtained during the present study show that the fracture toughness value remains unchanged when the sintering temperature is increased from 1040 to 1060 °C and reaches a value of 2.2 MPa√m. However, a further increase of the sintering temperature to 1070 and 1080 °C induced a decrease of the fracture toughness to 1.8 MPa√m.
The effect of sintering time on the fracture toughness of ceramic samples attained during sintering of the andesite basalt powder at 1060 °C is shown in Fig. 14b. Presented results show that at the optimal sintering temperature the fracture toughness values vary only slightly in the narrow range from 2.0 to 2.2 MPa√m even though the sintering duration is changed during the experiments by 5 times (from 30 to 240 min). The maximal fracture toughness value of 2.2 MPa√m was reached after sintering for 60 min, while the minimum value of 2.0 MPa√m was reached after sintering for 240 min. From the results presented in Fig. 14b it is clear that a longer sintering time at the optimal sintering temperature of 1060 °C did not contribute to an increase in the fracture toughness, although this could be expected as this is the case with most ceramic materials.
Figure 15 shows the impression of a square-based diamond pyramid made at the surface of the high-density ceramic sample sintered under the optimal sintering conditions, i.e. at 1060 °C for 60 min, for which the highest relative density of 99.5% was obtained and cracks created at the corners of the impression made during the Vickers test method. It can be seen that the crack propagation is radial relative to the axis of the imprint and that formed cracks have not branched out into many smaller cracks.
If the maximal fracture toughness value of 2.20 MPa√m, obtained during this study, is compared with the fracture toughness values of some aluminosilicate rocks, such as granite (granite fracture toughness value is ranging from 0.97 to 1.15 MPa√m) , it can be concluded that during this study the optimization of the sintering procedure led to fabrication of the refractory high-density ceramics with the fracture toughness value two times higher than granite.