Electrical properties of ternary Bi-Ge-Sb and Al-Cu-Sb alloys

Milena Premovic^a,*, Duško Minić^a, Milan Kolarevic^b, Dragan Manasijevic^c, Dragana Živković^{c, †}, Aleksandar Djordjevic^a, Dusan Milisavljevic^a

^a University in Priština, Faculty of Technical Science, K.M. 7, 4000 Kos. Mitrovica, Serbia

^b University of Kragujevac, Faculty of Mechanical and Civil Engineering, Dositejeva 19, 36000 Kraljevo, Serbia

^c University of Belgrade, Technical Faculty, VJ 12, 19210 Bor, Serbia

*Corresponding author: milena.premovic@gmail.com

(^† Deceased 26^th November 2016)

1. INTRODUCTIONTOP

Antimony is an important metal in the world economy because of the applications in many industries, such as electronics industry, military industry, pharmaceutical industry, etc. (Vinhal et al., 2016). The largest producer of antimony is China with 114 Sb mines and approximately 90% of the world’s production (He et al., 2012; Sun et al., 2016). The other producers are Canada, Russia, Bolivia and South Africa.

In the electronics industry antimony is used for making semiconductor devices, such as infrared detectors and diodes (Serrano et al., 2016). Antimony compounds are used to make bullets (Johnson et al., 2005), flame-retardant materials, car brake liner lubricants, type metal (in printing presses), enamels, cable sheathing, catalysts for the production of polyethylene terephthalate (PET) and alloy additives glass and pottery (Bech et al., 2012; Cui et al., 2015; Macgregor et al., 2015). On the other hand antimony is used as a medicine in the treatment of Leishmaniasis and their potential usefulness in AIDS and cancer therapy (Pierart et al., 2015).

Beside wide application of antimony it is important to study different combination of antimony with other element. During the past decades so many combination and also different composition of binary systems based on Sb were studied (Liu et al., 2013; Gierlotka, 2014) and later ternary systems (Zobač et al., 2015; Guo et al., 2016).

These two ternary systems Al-Cu-Sb and Bi-Ge-Sb were investigated before by the same group of authors as this paper. The ternary Al-Cu-Sb system was previously investigated by Minić et al. (2013) who experimentally studied liquid surface, three vertical sections and two isothermal sections at 200 ºC and 400 ºC. Another ternary Bi-Ge-Sb system was investigated by Premovic et al. (2014). In the paper given by Premovic et al. (2014) are investigated liquid projection, invariant reactions, three vertical sections, as well as isothermal sections at 100 ºC and 500 ºC.

In this paper electrical properties and microstructure of the alloy samples from the ternary Al-Cu-Sb and Bi-Ge-Sb systems are presented. Beside mentioned properties results of SEM-EDS and XRD measurement are presented as well. Also micro hardness was determined for some alloys from ternary Bi-Ge-Sb system. This kind of paper or similar (Illescas et al., 2009; Verbeken et al., 2010) will give a better insight into properties of alloys which should contribute to further development of application area.

2. MATERIALS AND METHODSTOP

All investigated samples were prepared from high purity (99.999 at. %) element produced by Alfa Aesar (Germany) in an induction furnace under high-purity argon atmosphere. Given that Sb is highly volatile an additional amount of Sb (about 1 to 2 at. %) was added to compensate for the weight loss. Additional amount of antimony is highly recommended especially for high antimony systems. In general, the average loss of mass during melting of samples was about 1 at.%.

Microstructural analysis was carried out by TESCAN VEGA3 scanning electron microscope with energy dispersive spectroscopy (EDS) (Oxford Instruments X-act) and by light microscopy using (LOM) OLYMPUS GX41 inverted metallographic microscope. Samples were prepared by the classic metallographic procedure without etching. Polished samples were firstly subjected to EDS elemental mapping to check compositional homogeneity and possible segregation and then analyzed. Overall composition and compositions of coexisting phases was determined using EDS point and area analysis.

XRD patterns of the studied samples were recorded on a D2 PHASER powder diffractometer equipped with a dynamic scintillation detector and ceramic X-ray Cu tube (KFLCu-2K) in a 2θ range of 5º to 75º with a step size of 0.02º. The patterns were analyzed using Topas 4.2 software and ICDD databases PDF2 (2013).

As the next step, electrical conductivity and hardness measurements were performed. Electrical conductivity measurements were carried out using Foerster SIGMATEST 2.069 eddy current instrument. The microhardness of the phases present in the microstructure was determined using a Vickers microhardness tester Sinowon, model Vexus ZHV-1000V.

3. RESULTSTOP

The alloy samples from ternary Bi-Ge-Sb system were investigated using several different techniques. Thirteen ternary alloy samples were observed with light optical microscopy, same samples were subjected to the measurement of the electrical conductivity. Further each phase detected in microstructure is subjected for determination of the microhardness. Additionally two ternary samples are investigated with XRD and SEM-EDS techniques.

Eight alloy samples from ternary Al-Cu-Sb system were observed with light optical microscopy and on same samples electrical conductivity were determined. For better insight into microstructures two ternary alloys are investigated with SEM-EDS and XRD techniques.

In Table 1 are presented list of phases for both ternary system with their corresponding Pearson symbols.

Thirteen alloy samples for investigation of electrical conductivity were selected along three vertical sections Ge-BiSb, Sb-BiGe and Bi-GeSb for ternary Bi-Ge-Sb and eight alloy samples from two vertical sections Al-CuSb and Sb-AlCu for ternary Al-Cu-Sb. The obtained values of the electrical conductivity of the studied alloy samples are given in Table 2. Also in Table 2 are presented experimentally obtained values for five binary alloys and literature value for electrical conductivity of pure elements, which are taken from online database periodictable, Electrical Conductivity of the elements (Gray et al., 2013).

In comparison of the obtained experimental results of electrical conductivity given in Table 2, it is clearly visible that ternary Al-Cu-Sb alloys have a significantly higher value of the electrical conductivity than alloy from ternary Bi-Ge-Sb system.

Based on experimentally determined values of electrical conductivity given in Table 2 and using an appropriate mathematical model, electrical conductivity of all other alloys in ternary systems have been calculated.

To define a mathematical model for electric conductivity of alloys software package Design Expert v.9.0.3.1 were used. Out of possible canonical or Scheffe models (Cornell, 1990; Kolarević, 2004; Lazić, 2004) that meet the requirements of adequacy recommended is Special Quartic model, Eq. (1):

The Analysis of variance (ANOVA) implies that the model is significant. However, the diagnosis of the statistical properties of the assumed model found that the distribution of residuals is not normal and that it is necessary to transform the mathematical model in order to meet the conditions of normality. The Box-Cox diagnostics (Box and Draper, 2006; Myers et al., 2009) recommends a ”Square Root“ transformation for the variance stabilization.

The Model summary statistics for ”Square Root“ model transformation are suggested Special Quartic Mixture Model. Again using the Analysis of variance (ANOVA) sugested model was checked. Value such as F factor, p-value and R-squared are statistical value which are important in way of searching a best model. According to the statistic calculation the highest value of the R-squared, F factor and low value of p-value factor (p value should be lower then 0.0001) justified the model. Obtained value for F and R for the Special Quartic Mixture Model for ternary Al-Cu-Sb system are F=86.26 and R-Squared=0.931. Which are in comparison with F and R value for other models significantly higher and chosen model is justified. Same check was done for ternary Bi-Ge-Sb and obtained value for F=526.34 and R-Squared=0.9827 which also justified chosen Special Quartic Mixture Model.

For both ternary systems Al-Cu-Sb and Bi-Ge-Sb same mathematichal model were used and same procedure was applied.

The final equation of the predictive model for Bi-Ge-Sb system is given as Eq. (2):

The final equation of the predictive model for Al-Cu-Sb system is given as Eq. (3):

Based on defined models electrical conductivity were predicted for all other alloys in the ternary systems and iso-lines for electrical conductivity of alloys defined by Eq. (2) and Eq. (3) are shown in Fig. 1 (a and b) respectively.

Some of the obtained light optical micrographs are presented in Fig. 2. All alloy samples from ternary Bi-Ge-Sb system have same three (Bi)+(Sb)+(Ge) phases in structure.

From micrographs presented on Fig. 2, is clearly visible that all samples are from three phase region. For better insight into microstructure two additionally samples per ternary system are investigated with SEM-EDS and XRD techniques. Obtained results are presented in Table 3. Microstructure of other samples which are not presented on Fig. 2, are presented in Table 4. All microstructures of samples from ternary Al-Cu-Sb are given in Fig. 2.

Table 4. Measured Vickers microhardness of the phases in the ternary Bi-Ge-Sb system

Experimentally results of XRD are compared with literature data. Literature value for lattice parameters of solid solution (Bi) are a=b=4.546 Å and c=11.862 Å defined by Cucka and Barrett (1962). Lattice parameter for solid solution (Sb) are compared with data given by Barrett et al. (1963), a=b=3.301 Å and c=11.232 Å. Obtained result for (Bi) and (Sb) solid solution are in reasonable agreement with literature, small deviation is visible because of solubility Sb in Bi and conversely. Data for (Ge) are used from Swanson and Tatge (1953) were determined lattice parameter are a=b=c=5.658 Å. Three detected intermetallic phases AlSb, Al₄Cu₉ and Cu₂Sb from Al-Cu-Sb system are in good agreement with data given by Woolley and Smith (1958), Westman (1965), and Pearson (1985) respectively. In Fig. 3 are presented two SEM microstructure for samples A and B (see Table 3). On both micrographs are visible three phases region. Sample A rich with (Ge) formed structure with three solid solution (Bi) light phases, (Sb) grey and (Ge) darkest phase. On micrographs from LOM (see Fig. 2) is hard to notice difference between (Sb) and (Bi) phases and on SEM micrographs phases are clearly visible. Sample B belongs to three phases region (Sb)+AlSb+Cu₂Sb.

Alloy samples from ternary Bi-Ge-Sb are further investigated with micro-Vickers hardness test and results of test are shown in Table 4. Each phase was tested several times (from ten up to twenty) at different position and mean value of hardness are given in Table 4 with micrographs after testing.

Experimentally obtained hardness for solid solution (Ge) is in the range of values from 844.78 MN.m^-2 to 860.22 MN.m^-2. In microstructures is very difficult to distinguish solid solution (Bi) from solid solution (Sb) and hardness of this two phases is determined as a harness of (Bi)+(Sb) solid solution with value in the range from 64.992 MN.m^-2 to 110.244 MN.m^-2. This big difference in hardness is possible to justify by quantity of Bi and Sb in alloy. Alloys rich with Sb have a higher value of hardness than alloys rich with Bi.

4. CONCLUSIONSTOP

ACKNOWLEDGMENTSTOP

This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, under Project No. ON172037 and TR37020.

REFERENCESTOP