One-Step Synthesis and Sintering of Skutterudite CoSb 3 : Smart Houses Materials ?

Thermoelectric materials may be used in devices as thermoelectric “air conditioner” in a smart house for improved energy efficiency. Energy harvesting uses ambient energy to generate electricity. It provides potentially low-cost, maintenance-free, long-life equipment by reducing the need for batteries or power chords. Energy harvesting (EH) is also known as power harvesting or energy scavenging. EH is considered to give benefits related to environmental friendliness, safety, security, convenience and affordability. EH can be used for brand enhancing. Technically, it can be used to make new things possible depending on visionary engineering. The variety of thermoelectric (TE) materials that can be used in energy harvesting is quite large, and the optimal material for a given application depends mainly on the temperature range in which the material is to be used. In this work we study the development and characterization of thermoelectric materials which were prepared by two different method, which were: a) ball-milling followed by sintering and b) ball-milling, microwave synthesis, high energy planetary ball milling, and sintering. Finally, we study the thermoelectric properties, calculate the band gaps and the ZT for the thermoelectric materials.


Introduction
World energy crisis has triggered more attention to energy saving and energy conversion systems along with economic efficiency.The global population is anticipated to have grown to 10 billion before the end of 21 st century.The economic growth, increase in energy production, and global environmental protection represent a trilemma [1].From the first law of thermodynamics, energy can neither be created nor destroyed.However we still need to be conservative in usage because of the efficiency, we cannot make full use of the energy.Large amount of energy is wasted.So how can we minimize energy waste and how can we reuse the wasted energy becomes critical in energy saving.
The field of Thermoelectricity (TE) advanced rapidly in the 1950s when the basic science of TE materials became well established.TE materials are important for specific applications e.g.satellites and space missions, equipment, and medical applications, where reliability, and predictability are very important.Thermoelectricity, which involves the conversion of heat into electricity and via versa, can play an important role in minimizing our dependency on fossil fuels [2].Good TE materials must have a large dimensionless figure of merit which defined as = [I], where S is the Seebeck coefficient, σ is the electrical conductivity, k is the thermal conductivity ( = + , where ke is the electronic contribution and is the lattice contribution), and T is the absolute temperature.The ZT value of a TE material can be improved by either increasing the electrical power factor , decreasing the electrical conductivity k [3], or a combination of both.
Skutterudite compounds have attracted much attention in the last decade as promising TE materials [4].Materials with skutterudite crystal structure possess attractive transport properties and have a good potential for achieving ZT values substantially larger than innovative thermoelectric materials.The crystal structure is cubic with 32 atoms per unit cell (space group Im3) and contains two voids per unit cell.The void can be filled by several guest atoms.The filler atoms in the voids can play two roles.First, the filler atoms donate electrons into CoSb3 framework, making the material n type.Second, the filler atoms in the voids are loosely bound with the neighboring atoms and have Einstein-like vibration, which act to scatter phonons and significantly reduce the lattice thermal conductivity [5].
CoSb3 has been identified as a candidate for good thermoelectric materials because it has a large Seebeck coefficient and good electrical conductivity [6].The thermal conductivity of CoSb3 however is still high for being an efficient thermoelectric material.To lower the thermal conductivity and further to increase ZT value, great effort has been made in recent years.Filling the Sb-icosahedron voids by rare earth or other metallic atoms was usually used [7], which can optimize the electrical transport properties and significantly depresses the lattice thermal conductivity due to the rattling of these atoms.In addition, doping with Fe, Ni, Te and Pd atoms can also reduce the thermal conductivity due to the phonons scattering by the introduced point defects [8].There have been many reports on improvements on thermoelectric properties of skutterudites by elemental substitution.A peak of approximately 0.9 at 530 °C was reported for CoSb11.4Te0.6 made by a solid-state reaction method followed by a spark-plasma sintering [9] and a peak ZT of ~1 at 530 °C in CoSb2.625Ge0.125Te0.25 by melting-quenching-annealing method followed by spark-plasma sintering [10].
Thermoelectric materials may be used in devices as thermoelectric "air conditioner" in a smart house for improved energy efficiency.Energy harvesting uses ambient energy to generate electricity.It provides potentially low-cost, maintenance-free, long-life equipment by reducing the need for batteries or power chords.Energy harvesting (EH) is also known as power harvesting or energy scavenging.EH is considered to give benefits related to environmental friendliness, safety, security, convenience and affordability.EH can be used for brand enhancing.Technically, it can be used to make new things possible depending on visionary engineering.The variety of thermoelectric (TE) materials that can be used in energy harvesting is quite large, and the optimal material for a given application depends mainly on the temperature range in which the material is to be used.Although thermoelectric materials exhibit thermoelectric behaviour at all temperatures, their figure of merit (ZT) is quite strongly influenced by temperature, and the ZT value typically peaks in a certain temperature range.When evaluating the feasibility of thermoelectric materials for different applications, the material performance in the required temperature range, in addition to other factors like cost and availability, should therefore be taken into account.
Nanostructuring has led to significant improvement in the properties of thermoelectric materials.The main strategy has been to decrease the thermal conductivity via phonon scattering.Research is required to understand the interaction between thermal, electrical and entropy transport; controlling nanostructures that can be used in actual devices; and improvements in materials for soldering, ceramics, packaging, etc.
The main focus of the research is still on tellurides due to their outstanding properties.Some of the most interesting alternatives to make cheap and less toxic TE materials include Mg2Si, CoSb3, ZnSb, ZnO and other oxides.All these materials have been known about for a long time.More explorative work is required to find completely new materials.
In this work we study the development and characterization of thermoelectric materials which were prepared by two different method, which were: a) ball-milling followed by sintering and b) ballmilling, microwave synthesis, high energy planetary ball milling, and sintering.Finally, we study the thermoelectric properties, calculate the band gaps and the ZT for the thermoelectric materials.

Experimental Procedures
The samples were prepared with Co (99.8 purity, Sigma-Aldrich) and Sb (99.5 purity, Sigma-Aldrich) powders as sources, which were weighed according to the stoichiometric ratio of CoSb3.

First Method (Ball Milling)
The powders were uniformly mixed with stainless balls in a planetary mill (Fritsch PQ-N04) for 2 h, the mixtures were enveloped by graphite foil and were loaded into a cylindrical graphite die with diameter of 15 mm, where then sintered into bulk samples by a Compact RF Sintering System (AMEN Technologies).Table 1 shows analytically the temperature, the force, and the total time of the sintering.

Second Method (Microwaves)
Such as the first method, the powders were mixed with stainless balls in a planetary miller for 2 h, the mixtures were cold-pressed in pellets and sealed in evacuated quartz tubes.The sealed quartz tubes were placed into a crucible filled with Copper oxide powder (98% purity, Sigma-Aldrich) which act as microwave susceptor material.The whole set up was placed into a commercial microwave reactor with rotate plate (Panasonic NE-1856).The reactions were allowed to run at 900 W for 4 min and at 450 W for 10 min.Every sample allowed to cool before it was removed from the microwave.The samples were crushed using a percussion mortar and ground into small pieces.In order to obtain a fine powder, the small pieces were ball milled for 15 min (Spex-8000 M Mixer/Mill).The powders were enveloped by graphite foil and were loaded into a cylindrical graphite die with diameter of 15 mm, where then sintered into bulk samples by a Compact RF Sintering System (AMEN Technologies).Table 1 shows the temperature, the force, and the total time of the sintering.
For both methods the crystallographic phase of the samples was characterized by X-ray diffraction (XRD, Seifert 3003 TT) with Cu-Kα radiation (λ=1.54059Ǻ), and the microstructure was analyzed by scanning electron microscopy (SEM, Phenom ProX).The thermal conductivity of the samples (15mm diameter and 3mm thickness) is characterized using a four-probe, steady-state electrothermal measurement technique based on the ASTM D5470-06 standard, in the range (300-700) K. Every sample was cut into bar (2 x 2 x 12) mm 3 to measure the electrical conductivity and the Seebeck coefficient, which been achieved using a commercial equipment based on the ZEM-3 standard in the range 300 K to 700 K. Moreover the Figure 2 shows the XRD pattern of BM2 sample which displays the lowest RBragg residual factor for the main phase.The presence of SiO2 is due to a fragment of quartz which created during of its smashing in order to release the sample BM2.Table 2 summarizes the lattice parameters, the weight percentage of the phases and the residual factors RBragg, Rwp, Rp, and Rexp which obtained from Rietveld refinements of the synchrotron XRD data for the sample BM2.The refined lattice constants of CoSb3 match those reported in the literature [9] and only very small deviations are observed for the individual samples.3 .These images reveal that the sample have a uniform microstructure and surface morphology.However in some cases, as we can observe in the spot 1 of the Figure 2b, the scanning electron microscope shows that the Sb phase distributes in the form of clusters with micron dimension.The EDAX analysis of the samples which is shown in Table 3 reveals that the average composition is very close to the nominal ratio Sb/Co = 3:1.The electrical conductivity of the samples as a function of temperature is plotted in Figure 4. From the plot we observe that, the electrical conductivity of each sample increases with temperature over the measured range, indicating semiconductor behavior.In addition, we can observe that the second method gives us almost the same results for the five samples, something that we can't notice in the first method.This phenomenon may be largely due to the presence of Sb which we have as secondary phase in the first method, because it can lead to an electrical conductivity increase [11].In Figure 5, the Seebeck coefficient is plotted as a function of temperature from 300 K to 700 K. Initially almost all the samples (except from BM3) show n-type conduction with negative Seebeck coefficients.For the first method we can't have useful conclusions for the switching sign of Seebeck coefficient because of the existence of secondary phases in different percentages.However, in the second method the Seebeck coefficient changes to positive approximately at 540 K.The reason is that the hole mobility is much larger than the electron mobility in the temperature range higher than 540 K [12], at which intrinsic excitation commences.It has been reported that the intrinsic CoSb3 shows the p-type conductivity [13,14].In this study, the low purity of Co (99.8%) may be the reason for the appearance of n-type conductivity initially [15].From the temperature-dependent Seebeck coefficient measurement, the band gap Eg of various samples can be roughly estimated using the equation = 2

Results and Discussion
(II), where e is the elementary charge, Smax is the peak Seebeck coefficient, and Tmax is the temperature corresponding to the maximum Sebeck coefficient [16].Based on this equation the energy gaps of the samples were estimated to be in the range of 0.09-0.29 eV (table 4), which is consistent with the values reported in literature [17].The temperature dependence of thermal conductivity for the two methods is shown in Figure 6.It is clear that the thermal conductivity of the most samples increases with the increasing temperature.The increase of k in some samples is due to the increase of the grain size.Significant role play the grain boundaries which could scatter phonons and decreases the thermal conductivity.At high temperatures, acoustic phonon scattering is mostly responsible for the decrease in thermal conductivity with a T -1 dependence [18].The calculated dimensionless figure of merit ZT values are shown in Figure 7.As we can observe all the ZT are in a range of 0.02-0.10.The maximum value of ZT is 0.10 for the BM1 sample at 673 K.The combination of low thermal conductivity and the high Seebeck coefficient are the reasons which the sample BM1 has highest figure of merit.Moreover it's obviously that in the second method we have a minimum for all the samples approximately at 540 K.The reason is that at 540 K the Seebeck coefficient changes sign from negative to positive.Recommendations given include further material science research on materials structure, electrical properties and performance, as well as processing and manufacturing to overcome obstacles related to price and production up-scaling.The work requires multi-disciplinary activity and co-operation.The utilization of advanced modelling supports the required multi-disciplinary actions.All above techniques may develop new materials for energy harvesting and the uses in smart houses and key technologies in smart cities.

Figure 1 Figure 1 .
Figure1shows the XRD patterns of CoSb3 samples which prepared by the two different methods in a range of 2θ =10 o -90 o .It can be seen that all the samples are single-phase materials with a minute amount of Sb and CoSb2.

PreprintsFigure 3 .
Figure 3. SEM images of the cross-section of sample BM2.

Figure 5 .
Figure 5. Temperature dependence of Seebeck coefficient of samples, (a) first method, (b) second method.

Figure 6 .
Figure 6.Temperature dependence of thermal conductivity of samples, (a) first method, (b) second method.

Figure 7 .
Figure 7. Temperature dependence of ZT of samples, (a) first method, (b) second method

Table 2 .
Lattice parameters, weight percentage of phases Rwp, Rp, and Rexp of sample BM2.

Table 4 .
Band gaps of samples