Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.
We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.
The scientific world has urged us to improve the sustainability of our electricity base because now we need electricity for just about everything. The use of conventional electrical energy resources to accommodate the growing electricity demand is quite alarming. We are not only creating imbalance in nature but also increasing the levels of greenhouse gases and polluting emissions (e.g., power generation: carbon footprint; refrigeration: mercury, CFC, freon emission). Renewable energy sources (based on solar energy, wind energy, hydro energy, etc.) stand in contrast to conventional sources with respect to their diversity, abundance, and applicability on and around the planet, but above all, no greenhouse gas emissions throughout their life cycle. Fascinatingly, thermoelectric (TE) ceramic is a class of semiconducting material with huge scope in the future as a renewable energy source. The peculiarity about thermoelectric ceramic material is that the same material can be used for producing power (Seebeck effect) and for removing or adding heat (Peltier effect). This chapter highlights the peculiar characteristics of thermoelectric ceramics and their prominent applications in the current era of energy crisis.
Materials and Metallurgical Engineering Department, National Institute of Advanced Manufacturing Technology, Ranchi, India
*Address all correspondence to: vspoddar@niamt.ac.in
1. Introduction
Today we need electricity for just about everything. To accommodate the growing demand for electricity, the increase in the use of conventional and exhaustible electrical energy resources is quite alarming. This scenario is creating imbalance in nature by increasing the levels of greenhouse gases and polluting emissions (e.g.,power generation: carbon footprint; refrigeration: mercury, chlorofluorocarbon (CFC), freon emission) in our environment. And for this reason, the scientific world has urged us now to improve the sustainability of our electricity base.
Renewable energy sources (based on solar energy, wind energy, hydro energy, etc.) stand in contrast to the conventional sources with respect to their diversity, abundance, and applicability on and around the planet, but above all, no greenhouse gas emissions throughout their life cycle [1]. Fascinatingly, thermoelectric (TE) is a unique class of materials with huge scope in the future as a renewable energy source. The peculiarity about thermoelectric material is that the material can be used for producing power (Seebeck effect) as well as for removing or adding heat (Peltier effect) [2]. Of course, depending on the output requirement, i.e., power or heat, a slight modification in the circuit needs to be made. Therefore, thermoelectric may be used for both heating and cooling, thereby making it highly suitable for precise temperature control applications, e.g., thermocouple. Thermoelectricity is shown by various material families such as tellurides, chalcogenides, clathrates, skutterudites, half-Heusler alloys, silicides, oxides, and polymer composites [3, 4]. These materials have a large range of possible options with respect to composition and structure. Most of these materials are ceramic in nature because of large effective masses and under certain limitations based on two criteria: (1) bond characteristics; and (2) temperature. The bonding found in thermoelectric ceramics is a combination of ionic and covalent types with the dominance of ionic characteristics [5]. The operating temperature should be such that the charge carrier concentration is low and the corresponding activation energy is in the range of less than 0.08 eV. Thermoelectric ceramic material has been recently emerging, and there are very few reports on their performance. Thermoelectricity is also exhibited in all metals, but to a very low extent. Metals, although they have high carrier concentration and easy processing, promoting, thermoelectricity, but they have disadvantages such as sensitivity to impurities and oxidation, which degrade their thermoelectric behavior.
1.1 Thermodynamics of thermoelectric ceramics
According to the second law of thermodynamics, the energy input to a system is never completely converted to usable energy output. The difference between the two constitutes the unutilized heat, and it is, though in variable quantities, available in almost every system. Thermoelectric materials, used in power generation, scavenge the unutilized heat from the source and convert it to a usable electrical form of energy [6]. The unutilized heat may be of high temperature, medium temperature, or low temperature grade. That means heat is dumped, for example, from industrial processes like hot metallurgical processing (high temperature: greater than 600°C), from exhaust gases released by the combustion units (medium temperature: 250–600°C) or from hot surfaces of the indirectly involved processing units (low temperature: less than 250°C) into the environment, causing thermal pollution [7]. Surprisingly, around 66% of the unutilized heat in industries is available in the low-temperature range of around 66–230°C. Thermoelectric can work as a renewable energy source in low-temperature regimes or as a supplement to conventional energy sources in high-temperature regimes. In the medium temperature range, thermoelectric can work in either way depending on the system or the purpose for which it is used [8]. Therefore, thermoelectric comes under the category of sustainable energy-efficient technology with the additional advantages of simple configuration, no moving parts, no vibration, no noise, no pollution, and high stability [9]. However, the challenge in using a thermoelectric system is that for a particular temperature, it should be inclusive of all forms of heat source, be able to run continuously, have a long life, be low cost, and be light in weight [9, 10].
1.2 Ceramic as thermoelectric generator
A ceramic is suitable as a thermoelectric generator (TEG) when the supply of thermal energy produces many numbers of charge carriers due to the Edison effect [11]. Also, these generated charge carriers should move from the hot end of material to the cold end (that is, across the thermal gradient) due to the Seebeck effect with sufficient mobility. Under the effect of a thermal gradient, dc voltage develops across the length of the thermoelectric ceramic and a net current flows through the circuit by means of the charge carriers. At high temperatures, the charge carriers are moving, vibrating, and rotating with more energy compared to those at low temperatures. However, the transition of charge carriers from the hot end to the cold end will undergo scattering a gazillion times due to the atoms present in their path. This scattering is remarkable in ceramic due to its high effective mass and a cloud of polarization with a certain spatial diameter. The ceramic structure does not have enough space left for the movement of charge carriers without any interaction. During transition, charge carriers interact with atoms, which results in a change in direction of the charge carrier and/or generation of a photon or phonon. Thermoelectric ceramic transfers the bulk of the heat from the hot end to the cold end in the form of phonons rather than photons. However, most of the phonons are consumed in the generation and transfer of charge carriers from the hot end to the cold end of the thermoelectric. Thus, in thermoelectric, unlike metals, the number of phonons reaching the cold end is as low as possible.
1.3 Thermoelectric cooler
For a thermoelectric cooler (TEC), sometimes called a Peltier cooler, the material should respond to a low-voltage DC power input. The response is in the form of the transfer of heat from one side of the TEC to the other and is directly dependent upon the DC voltage. Depending on the direction of current, the side will get heated or cooled. With a change in the polarity of current, the hot side will now get cooled down and the cold side will get heated up. In other words, a thermoelectric cooler works as a semiconductor-based small heat pump. Consequently, it can be used for heating or cooling, although in practice the main application is cooling. However, there is a limit to the flow of current through a thermoelectric cooler because this current is prone to the generation of heat, which adds to the overall heat dissipation [2].
A change in DC voltage is reflected as a change in temperature, which can be amplified with the use of a proper TE material pair (p-type TE to n-type TE combination) in the circuit. One such ceramic material pair for this cooling application can be Ca3Co4O9 (p-type) and Ca0.95Sm0.05MnO3 (n-type) [12]. The controlled DC input to the thermoelectric cooler allows keeping control over the electron–hole pair combination rate in TEC. This merit is highly suitable for use of TEC in precise temperature control applications like aerospace (extremely demanding conditions), imaging technology (constant viscosity of ink), and temperature measuring units. Thermoelectric coolers also provide refrigeration and temperature control in electronic packages and medical instruments.
1.4 The figure-of-merit
Thermoelectric ceramic materials are a member of the advanced materials family, wherein tailoring of constituent elements, following a systematic synthesis route, and engineering of material structure is of crucial importance. The performance of TE is characterized by Figure-of-Merit expressed as Z with unit K−1. To make it dimensionless, Figure-of-Merit, Z is normally expressed as ZT, where T is the average temperature of hot side and cold side of the TE. The maximum efficiency of the energy conversion process in TEG and TEC at a given point in the TE material is determined by its ZT value and is given in Eq. (1).
ZT=S2σTkE1
where S, σ, k, and T are the Seebeck coefficients, (μV K−1), electrical conductivity (Ω−1 m−1), thermal conductivity (W m−1 K−1), and absolute working temperature (K), respectively [13]. From Eq. (1), it is observed that the TE material should behave like a glass to have low thermal conductivity (k), a metal to have high electrical conductivity (σ), and a semiconductor to have a high Seebeck coefficient (S) all at the same time [3]. In other words, a contradictory combination of material behavior is required in TE materials.
As shown in Figure 1, each of the performance parameters for the thermoelectric depends on the temperature of operation, T. To increase the ZT value, the Seebeck coefficient, S, and electrical conductivity, σ, should be high and the thermal conductivity, k, should be low. Though interrelated through classical physics, due to the contradictory requirements by each of these factors, improvement in one property leads to a trade-off in which the other properties get lowered. For example, if the electrical conductivity is increased by a high carrier concentration in the thermoelectric material, the Seebeck coefficient and thermal conductivity are decreased. It is to be noted that k is the total thermal conductivity inclusive of the lattice and electronic thermal conductivity components of the TE material [9]. It can be concluded that with the increase in temperature, both the numerator and denominator of Eq. (1) should show an increasing trend. But there is a limit to it. Thermoelectric ceramic is also subjected to a bipolar effect at high temperatures. This effect is predominant in doped TEs. The number of minority charge carriers increases and competes with the majority charge carriers. Now these carriers have opposite signs, and therefore they increase the Seebeck coefficient and electrical conductivity [14]. With minority charge carrier contribution, the Seebeck coefficient increases to a maximum and then decreases. This inter-relation of temperature, Seebeck coefficient, and band gap has been rightly put forth by Goldsmid-Sharp as band gap [15], as given by Eq. (2).
Figure 1.
Dependency schematic of Figure-of-Merit, ZT, Seebeck coefficient, S, electrical conductivity, σ, thermal conductivity, k, and absolute temperature of the performance evaluation, T of the TE material. The parameter indicated by the head of the arrow is strongly influenced by the parameter present at the other end.
p-type β-FeSi2
n-type β-FeSi2
Stoichiometry composition
Abbreviation
Stoichiometry composition
Abbreviation
Fe0.96Mn0.04Si2
Mn0.04
Fe0.99Co0.01Si2
Co0.01
Fe0.92Mn0.08Si2
Mn0.08
Fe0.98Co0.02Si2
Co0.02
Fe0.88Mn0.12Si2
Mn0.12
Fe0.95Co0.05Si2
Co0.05
Table 1.
Abbreviations used for the various compositions of Mn-doped p-type and Co-doped n-type thermoelectric (TE) β-iron disilicide legs [24].
Eg=2eSmaxTmaxE2
The ZT values of 1 are considered good, but a value of 3–4 makes the TE material compete with other mechanical devices available for the same purpose. Synthesis of higher ZT (ZT > 1) TE material is a challenge for their proper use in refrigeration and electrical power generation applications. For materials having similar thermal conductivities, the term power factor (PF = S2σ) is used to define the performance of TE [9]. To increase the Seebeck coefficient and electrical conductivity, valence band engineering (convergence) and conduction band engineering for TE materials are focused, respectively.
Thermoelectric effect is the phenomenon by which a temperature difference across a material sets up electric potential in it. It is exhibited by all metals, but TE materials are identified as a separate class of materials. This is because they are good at converting the temperature difference into voltage [9].
Thermoelectric effects result from the interaction of charge flow and heat flow across the length of a TE. Thermoelectric phenomena are known more specifically as the Seebeck effect (converting temperature to current), Peltier effect (converting current to temperature), and Thomson effect (conductor heating/cooling). All materials exhibit thermoelectric effects, but the name ‘thermoelectric materials’ is used to describe the materials that are good at converting heat to electricity and vice versa. It’s interesting to note that these are all old effects, but they are making a significant contribution to the recent green technologies.
2.1 Seebeck effect
A discussion of thermoelectric materials and devices starts with one of the most fundamental phenomena, the Seebeck effect. In 1821, Seebeck observed that if two dissimilar metals were joined together and the junctions were held at different temperatures (T and T + ∆T), then a voltage difference (∆V) proportional to the temperature difference (∆T) developed [8]. The ratio of the voltage developed to the temperature difference (∆V/∆T) is related to the intrinsic property of the material termed the Seebeck coefficient (S or α) or thermopower as described in Eq. (3). This developed voltage is used to drive a current through the device or a load resistance [6, 9].
S=∆V∆TE3
This effect can be understood by establishing a temperature gradient across a material. Energetic electrons flow from higher potential to a lower potential until an electric field is established to impede further flow of electrons. The Seebeck coefficient has the units of volts per degree Kelvin and is usually expressed in μV K−1. The type of charge carrier determines the sign of the Seebeck: S < 0 for electrons and S > 0 for holes in a semiconductor. S is very low for metals (only a few microvolts per degree Kelvin, e.g., Al: −0.2 μV K−1 at 100°C) and is much higher for semiconductors (typically a few hundred microvolts per degree Kelvin, e.g., Ge: −210 μV K−1 at 700°C) [16].
2.2 Peltier effect
A few years after the discovery of the Seebeck effect, Peltier observed that if an electrical current is passed through the junction of two dissimilar metals, heat is either absorbed or rejected at the junction, depending on the direction of the current. This phenomenon is known as the Peltier effect. This effect is attributed to the difference in Fermi energies of two materials. The Seebeck effect and the Peltier effect are related to each other through thermodynamics as per the following Kelvin Law mentioned as Eq. (4).
Π=STE4
Π is called the Peltier coefficient. The Peltier coefficient is simply the Seebeck coefficient times the absolute temperature [2].
The rate at which Peltier heat is liberated or rejected at the junction (Q̇p) is given by the following relation represented in Eq. (5).
Q̇p=SITE5
Here I is the current through the junction, and T is the temperature in degree Kelvin. Therefore, now using a low-temperature TE device, water can be frozen with the current in one direction and boiled with the current in the opposite direction [16].
2.3 Thomson effect
The Thomson effect was discovered in 1854 by the British physicist William Thomson (Lord Kelvin). In the Thomson effect, heat is absorbed or produced when current flows in a material with a temperature gradient. In many materials, the Seebeck coefficient is not constant with temperature, and so a spatial gradient in temperature can result in a gradient in the Seebeck coefficient (dS/dT). If a current is driven through this gradient, then a continuous version of the Peltier effect occurs. According to Thomson, the heat (H) is proportional to both the electric current (I) and the temperature gradient (∆T), and the proportionality constant, known as the Thomson coefficient, K, is related by thermodynamics to the Seebeck coefficient [9]. This can be presented as Eq. (6) and (7).
H=KI×ΔTE6
K=TdSdTE7
Though Seebeck and Peltier effects are reversible, it is rather more difficult to demonstrate experimentally the Peltier effect than the Seebeck effect because of the unintentional involvement of Joule heating. Thus, in practice, thermoelectric energy conversion is mostly expressed in terms of the Seebeck coefficient rather than the Peltier coefficient. Eq. (4) suggests that the Seebeck and Peltier coefficients are interdependent and that, if required, the Peltier coefficient can be evaluated from it.
3. Electrical conductivity in thermoelectric ceramics
When a temperature gradient is applied across the TE material, the charge carrier mobility is affected, as shown in Figure 2. The kinetic energy of the charge carriers at the hot end is greater than the charge carriers at the cold end. So, the number of charge carriers diffusing from the hot end (high energy) to the cold end (low energy) is higher compared to the charge carriers diffusing from the cold end to the hot end. This creates a charge gradient across the TE material corresponding to the temperature gradient.
Figure 2.
Electron gradient across an n-type TE material because of temperature gradient.
The most common methods for determining the electrical conductivity of TE material are the Hall Effect and the four-point probe method. The Hall Effect is measured by applying magnetic fields perpendicular to the surface of the sample and electrical currents along the edges of the sample while measuring the voltage across the sample. The applied magnetic field and current create a force that pushes mobile charges to one side of the material, creating an electric field [17]. Four-probe apparatus is one of the standard and most widely used apparatus for the measurement of resistivity of semiconductors. In this method, it is considered that the conductivity of material is uniform in the area of measurement [18].
If there is a minority carrier injection into the semiconductor by the current-carrying electrodes, most of the carriers recombine near electrodes so that their effect on conductivity is negligible. The surface on which the probes rest is flat with no surface leakage. The four probes used for resistivity measurement contact the material surface at points that lie in a straight line. The diameter of the contact between metallic probes and the semiconductor should be small compared to the distance between the probes. The boundary between the current-carrying electrodes and the bulk material is hemispherical and small in diameter. It is assumed that the size of the metal tip is infinitesimal and the sample thickness is greater than the distance between the probes; then the resistivity is determined according to Eq. (8).
ρ0=VI×2πsE8
where V is the potential difference between the inner probes in volts, I is current through the outer pair of probes in amperes and s is spacing between the probes in meters.
4. Thermal conductivity in thermoelectric ceramics
Thermal conductivity is the sum of electronic (kel) and lattice (klatt) thermal conductivity. In a material, the transport of heat across its length takes place through the movement of charge carriers (electrons and/or holes) and thermal carriers (phonons). In metals and degenerate semiconductors, the transport of heat is majorly through charge carriers, and it is well explained by Wiedeman-Franz-Lorentz Law, kel = LσT, which states that the electronic thermal conductivity is directly proportional to electrical conductivity, and the constant of proportionality is the Lorentz number, 2.45 x 10−8 W·Ω·K−2. In this law, it is considered that heat and electrical transport take place through charge carriers only [19]. However, for non-degenerate semiconductors like thermoelectric ceramics, a decrease in L takes place with an increase in temperature. This decrease in L is correlated with an increase in the value of the Seebeck coefficient (S). Using these L values, kel was determined with respect to temperature for all types of thermoelectric ceramics. In practice, the electronic contribution to the thermal conductivity is negligible, i.e., kel is very small as compared to klatt (kel < < klatt), and hence kel is neglected [19].
The concept of lattice thermal conductivity, klatt, was developed earlier by Born and Debye in 1913. Lattice thermal conductivity is contributed by the phonons with different wavelengths [10]. It is approximated by the classical kinetic theory, which treats phonons as gas particles, and following all the assumptions of the theory, klatt is given as Eq. (9).
klatt=13CvlvsE9
where Cv is the specific heat at constant volume, l is the mean free phonon path, and νs is the average velocity of sound. At very low temperatures (below 40 K), when lattice vibrations are negligible, Debye law is followed, which states that Cv is proportional to T3. Generated phonons are fewer in number and have long wavelengths. Above this temperature, the number of generated phonons increases, and their wavelength reduces. Also, they start interacting with each other (phonon–phonon scattering) as well as with the phase structure [20].
As the amount of heat absorbed by the TE material varies across the length extending from hot side to cold side, the phonons generated can exhibit a wide range of wavelengths. The defects or the particle size should match the mean free path of these phonons so as to scatter them. This requires proper engineering of the TE ceramic structure to ensure maximum heat-carrying phonons are utilized, which will reduce thermal conductivity [21]. This is one of the main design characteristics to be considered while developing the thermoelectric ceramic material of high ZT.
Ioffe et al. studied the physical problems of thermoelectricity and presented their study on the mobility of charge carriers and factors affecting the same. The mobility of charge carriers is limited by their scattering because of thermal vibrations and lattice defects. If the Debye temperature is less, then the amplitude is large, which will reduce the mobility [10].
The dependency of Z on temperature and mobility can be expressed roughly by the relation given in Eq. (10).
Z∼μklattT3/2E10
where Z depends on the type of impurity introduced in the system. The metallic impurity atoms act as donors irrespective of their valency, and non-metals act as acceptors [10].
If there is a temperature gradient in a solid body, the number of phonons moving from the hot side to the cold side is greater than the number moving in the opposite direction. As a result of collisions, the phonons can drag electrons along in their own direction of motion, and at the cold side of the specimen a negative charge will appear (similarly a positive charge will appear at the hot side) until the potential difference produced balances the effect of charge flow. This is the dominant phenomenon happening in the high-temperature thermoelectric ceramics, so that their electrical conductivity increases at higher temperatures.
In the early 1960s, a requirement for autonomous sources of electrical power arose from the exploration of space, the exploitation of the Earth’s resources, and advances in medical physics. But TEs have long been too inefficient to be cost effective in most of the applications. A. F. Ioffe initiated wide research in thermoelectric semiconductors. Along with coworkers, he demonstrated that the thermoelectric properties in bulk materials can be enhanced by tuning or doping techniques to vary the lattice thermal conductivity without affecting the electrical conductivity. The first commercial TE generator used an oil burning lamp as the heat source and ZnSb ceramic and constantan as TE materials. This TE generator developed in the twentieth century powered a radio. Similar work was carried out, and a focus was made on finding small-band gap semiconductors made from heavy elements [10, 22].
However, a revival of interest in thermoelectric began in the mid-1990s when theoretical predictions suggested that thermoelectric efficiency could be greatly enhanced through nanostructure engineering, which led to experimental efforts to demonstrate the proof-of-principle and high-efficiency materials [9]. In 1995, Slack et al. developed a TE material based on the “phonon glass and electron crystal (PGEC)” concept. The voids in the TE crystal structure are filled with heavy element atoms, which act as rattlers, increasing the number of phonon scattering centers, hence reducing the lattice thermal conductivity significantly [20]. The constituent atoms had small differences in the electronegativity and high charge mobility. The main characteristic of these TE materials is that they can carry the electric current as a crystal but behave as amorphous materials with respect to the lattice thermal conductivity. Such development was verified on TE materials like skutterudites, clathrates, ceramics, half-Heusler alloys, and β-Zn4Sb3 phases.
From Figure 3, it can be seen that in the low-temperature range, TE is used as solar TE—a renewable energy source—or it can be used along with photovoltaic (PV) to improve the PV efficiency. Also, Peltier coolers are included in this category to carry out refrigeration. In high-temperature applications, TE is employed for waste heat recovery. There are many kinds of TE materials, from semiconductors to ceramics, from oxides to organics, from bulk to superlattice, and from nanoparticles to nanowires. Each TE material is suitable for a particular temperature range. Higher the ZT, more is the suitability for that particular temperature. Figure 3 illustrates the mostly preferred TE material to be used at a particular temperature. For good efficiency, materials with high electrical conductivity, low thermal conductivity, and a high Seebeck coefficient are needed. The widespread commercial integration of these materials has been impeded for several reasons, the most crucial being the need for materials with high ZT that also have a long life, are low cost, and are light in weight. Some of the TE materials under study are tellurides, chalcogenides, clathrates, skutterudites, half-Heusler alloys, silicides, oxides, polymer composites, etc. Apart from the search for high ZT values, research on new TE materials now focuses on eco-friendly lead and/or tellurium-free materials with good mechanical and thermal stability [25]. Among thermoelectric materials with high efficiency and lacking the non-desired tellurium or lead, skutterudites (e.g., Ba0.08La0.05Yb0.04Co4Sb12) and silicides (e.g., Mg2Si, MnSi1.73, Fe0.92Mn0.08Si2) stand out with a promising future [26]. The research is also heading toward finding all-round user-friendly TE material with respect to cost, material availability, and scale-up viability.
Figure 3.
Highest ZT thermoelectric materials to be used at various temperatures for various applications [12, 16, 23, 24].
Various techniques for synthesis of TE bulk materials are (1) melting-casting method (2) Rapid Solidification Process (RSP) (3) Self-propagating high-temperature synthesis (SHS) process (4) generic powder metallurgy (PM) and Mechanical Alloying (MA) method. TE produced by conventional ingot technology generally has a rather coarse eutectic structure, which cannot be engaged in thermoelectricity [21]. In many cases, this first method, based on the classical melting process, does not yield a homogeneous TE bulk product directly because of the various transformation reactions (e.g., eutectic, peritectic, eutectoid, peritectoid, etc. depending on the equilibrium phase diagram) [27]. In the case of the SHS process, very low density is achieved, which reduces the phase formation tendency and the electrical conductivity of the TE material [21]. Rapid solidification followed by hot pressing is found to be an effective way to produce fine grain structures and improve the phase transformation speed and thermoelectricity properties. However, the fine grain sizes in these alloys grow rapidly up to about a few μm after hot pressing and impairing the TE properties [28]. PM and MA are found to be very good candidates for the demanding TE ceramic technology, but a lot of work has yet to be done to get rid of problems like inhomogeneity in phase formation, dispersion of dopants, etc. [29, 30]. To sum up, many techniques have been employed for the synthesis of TE material, but there is no such specific route that can lead us to the synthesis of this material on an industry level.
Establishment of proper electrical contact is essential to improve the TE power efficiency. The contact region is the only way to assist in the transfer of charge carriers from the TE material across the load resistance. Discontinuity in the contact region will reduce the TE performance [6]. The contacts used may be silver, aluminum or copper based depending upon the cost, shape and size of TE legs, temperature of application and the operating conditions. The interface between contact and TE material becomes crucial because it affects the electrical properties (transport) of the TE devices. This interface is called the junction. An ideal junction is the one where there are no defects formed at the interface. However, in practice most of the contacts have defects at the interface, which add to the contact resistance between the contact and the TE leg. The contact resistance depends on the material’s surface roughness, interstitial conditions, contact pressure, and temperature. The first three process parameters affect the contact area, whereas the fourth parameter, i.e., temperature, has a strong influence over the resistivity of the contact material [31]. For low-temperature TE, mostly Al is used for connections. In the case of intricate designs, physical vapor deposition of Al is carried out to create the connections. Cu is the preferred connector over Al in most of the applications, e.g., Bi2Te3 TE circuit. This is because of the better combination of electrical and thermal conductivities of Cu than Al. In high-temperature applications, Ag is used, as in the case of β-FeSi2 TE circuit [32, 33]. Here, the silver foil with 99.9% Ag was used for circuit formation with a resistivity at room temperature of about 1.59 μΩ-cm. The thermal conductivity of Ag varies with respect to temperature following the relation, y = −2.0714x2–0.4714x + 432.4, R2 = 0.9962 [33]. The hot side and the cold side of the TE need to be completely and uniformly layered by the silver paste so as to reduce the contact resistance. For this, conductive silver paste (particle size 10 nm) was used. The applied surfaces were then exposed to UV light for 45 min for curing. The silver foil of thickness 20 μm was properly cut to the size of the hot side and cold side. It was placed over the silver paste, and a pressure of around 4 MPa was applied, and then the TE leg was kept in the muffle furnace at 800°C for 1 h to ensure a perfect contact.
Evaluation of TE ceramic circuits with respect to the electrical performances can be made through either physical mode or mathematical mode. This study is required to find out the usable and worthy temperature regime and the energy efficiency of a particular TE ceramic. The discussion in this section, related to the thermoelectric power generator (TEG) and refrigeration circuits, is generic and can be applied to any thermoelectric systems with minor changes. It includes the Seebeck effect, the Peltier effect, a parameterization of the Joule heat, as well as all thermal and electrical resistances [2]. Geometry optimization and investigations of the influence of process parameters and load resistance, which affect the output power, are also discussed. The study is carried out stepwise in three stages: (1) TE single leg, (2) TE unicouple, and (3) TE module, and is applied to iron disilicide TE to cite an example.
8.1 TE single leg
The first stage includes the study of a single TE leg made up of p-type material. The p-type TE leg is connected electrically using silver film or silver wire as an element of TE generator (TEG) setup as shown in Figure 4(a). The two joints are held at different temperatures to have thermal gradient [35]. A similar study is applied to the n-type leg.
Figure 4.
Single thermoelectric leg made of p-type material under steady-state heat transfer conditions. The p-type leg is connected electrically using a silver connector. The two joints are held at different temperatures to have thermal gradient. A similar study can be applied to n-type leg. (a) Schematic showing physical setup of thermoelectric generator (TEG); (b) Thermal circuit representation of (a) under no load condition; (c) Thermal circuit in power generation mode [34].
The voltages generated by p-type and n-type legs are respectively evaluated using Eq. (3) and are given as:
∆Vp=SpTH−TCE11
∆Vn=SnTH−TCE12
where ∆Vp and ∆Vn are voltages generated, and Sp and Sn are Seebeck coefficients of p-type and n-type leg respectively. The sign of S is positive if the emf tends to drive an electric current from the hot side to the cold side.
The rate of heat flow through the TE leg is:
Qṗ=kpATH−TClE13
Qṅ=knATH−TClE14
where kp and kn are thermal conductivities of p-type and n-type legs, respectively, and A and l are cross-sectional area and length of each leg, respectively. The internal resistances are described as rinp=rp+2rc+2rAg and rinn=rn+2rc+2rAg for p-type and n-type circuits, respectively, as shown in Figure 4(b). They include TE resistance rp or rn; contact resistance at each side, rc and resistance of the contact material, here Ag, rAg. In TE, the hot side and cold side are having different temperatures, and rAg here, have to be calculated at the average temperature. The same circuit is closed in power generation mode using a variable load resistance RL (for refrigeration mode, dc voltage difference is applied instead of RL) as shown in Figure 4(c) and correspondingly an electric current Ip flows in the circuit [36]. The current Ip and In flowing through the p-type and n-type circuits, respectively, is given as:
Ip=∆Vprinp+RLE15
In=∆Vnrinn+RLE16
where rinp and rinn are the internal resistances for p-type and n-type legs, respectively, and all these factors are taken care of during generator circuit construction by following proper assembly procedure as mentioned in Section 6.3.
The energy conversion efficiency, η, of the TE leg is defined as the ratio of the power output (used or stored) to the total heat energy input. To evaluate the physical condition, let the circuit be divided into two halves, the upper hot side and the lower cold side. The power output is to the cold side and is represented as PL. The heat input, or in other words, the power input, is to the hot side and is represented as Q̇Hp .
Now let us determine the power output, PL, across load resistance RL as:
where PLp and PLn are the power outputs for p-type and n-type legs, respectively.
Power output and energy conversion efficiency are the primary parameters to characterize TE performance, and they should be as maximized as possible, and thus the condition for the same needs to be evaluated. For this purpose, the Maximum Power Transfer theorem is used for TE. From Figure 4(c), it can be predicted that the maximum power output depends on the value of the load resistance. To find the maximum power, differentiate Eq. (18) or Eq. (19) with respect to resistance RL and equate it to zero.
The condition for maximum power output, according to the Maximum Power Transfer theorem, is that the load resistance, RL, should be equal to the internal resistance, rinp or rinn. The power and efficiency calculations are carried out considering rin = RL, which is also termed the “matched condition” [37]. Applying this condition, the Eqs. (18) and (19) are now become, respectively:
PLp=Sp2TH−TC24rinpE21
PLn=Sn2TH−TC24rinnE22
The heat absorbed at the hot side or the power input consists of the heat getting absorbed due to flow of current (Peltier effect), rate of heat flow across the length (Seebeck effect), and heat transport boundary condition (Joules effect). An assumption made over here is that the convection and radiation at the overall surfaces of the legs are negligible. Each of these factors is related to Q̇Hp, as mentioned in Eqs. (23)–(24).
QHṗ=SpITH+kpATH−TCl−12I2rinpE23
QHṅ=SnITH+knATH−TCl−12I2rinnE24
The energy conversion efficiency of the p-type and n-type TE legs is respectively given as:
ηp=PLpQ̇HpE25
ηn=PLnQ̇HnE26
Rearranging the terms in Eqs. (21), (23) and (25) can be rewritten as;
ηp=Th−Tc4Th2+kpρpSp2−Th−Tc8E27
Similarly, for n-type leg Eq. (26) can be rewritten as;
ηn=Th−Tc4Th2+knρnSn2−Th−Tc8E28
where ρp=rinpAl and ρn=rinnAl are the resistivity of p-type and n-type legs, respectively. Direct energy conversion relies on the physical transport properties ρ, k, and S of the TE material, and their energy conversion efficiency can be written in terms of the Figure-of-Merit, Z.
The Eqs. (26–27), can be rewritten in terms of Z as;
ηp=2ZpTh−Tc8+Zp3Th+TcE29
ηn=2ZnTh−Tc8+Zn3Th+TcE30
where Zp and Zn are Figure-of-Merit for p-type and n-type legs, respectively. It can be seen from Eqs. (29–30) that the temperature difference across the TE leg and the TE Figure-of-Merit (Z) has a major impact on the energy conversion efficiency. The temperature difference between the hot side and the cold side is to be kept as high as possible to have maximum energy conversion efficiency [38]. This is strongly influenced by the thermoelectric generator (TEG) design and thus can be controlled by engineering the design parameters of a TEG. Figure-of-Merit, Z is the material property and is an important consideration while classifying any material as TE.
As shown in Figure 5, the direction of the current decides which TE side will get heated up or cooled down. If the mechanical setup of TE is made to be used for heating, then by just changing the direction of current, the same setup can be used for cooling. Normally, the Peltier or TE coolers are used till 200°C. The important criterion to evaluate the performance of a TE cooler is the Coefficient of Performance (COP). The COP is the ratio of heat absorbed at the cold side (QC) to the input power (QiP) of the TE leg.
Figure 5.
Direct current, I is passing through the TE circuit, which results in active cooling on one side of the TE. Thermal circuit of TE in refrigeration mode (a) p-type TE leg, (b) p-type TE leg, (c) TE unicouple.
COP=QCQiPE31
The heat dissipated by the heat sink, QH, is the sum of heat absorbed at the cold side and the input power.
QH=QC+QiPE32
From Eq. (1), the condition for maximum COP is that QiP is to be kept at minimal, i.e. minimum amount of heat to be dissipated by the heat sink. In other words, COP depends on the operating current, I, or the temperature difference (∆T = Th - Tc) [2].
The current I flowing through the cold side of TE absorbs Peltier heat at a rate of Q̇iP. Using Eqs. (4) and (5),
Q̇iP=ΠIE33
The current I flowing through the TE leg generates heat Q̇J due to the Joule heating [2].
Q̇J=lσAI2E34
This heat generated in the bulk distributes equally to both the hot and cold sides of the TE. Therefore, half of it will add to the heat absorbed at the cold side. The TE material within the TE leg will also contribute to the heat flow from the hot side to the cold side by conduction. As shown in Figure 5(a)–(b), this heat flow is given by the thermal conduction Eqs. (13)–(14) for p-type and n-type legs, respectively.
The net rate of heat absorption or the cooling rate Q̇net at the cold side of the p-type TE leg is given as;
Q̇net=Q̇iP−Q̇J2−Q̇PE35
The use of TE coolers is worth it if maximum cooling is obtained on the cold side. To determine the maximum cooling rate or the maximum heat pumping capacity, Eq. (35) is differentiated with respect to I. By doing this, the maximum current and the maximum cooling rate Q̇P,max are obtained as given in Eqs. (36) and (37).
Imax=AlΠσE36
Q̇P,max=Al12Π2σ−kPTH−TCE37
Under steady-state conditions, ∆T is constant and the net cooling rate at the junction becomes zero. From Eq. (36), the maximum temperature difference achievable is calculated as;
∆Tmax=Π2σ2kPE38
From Eqs. (4) and (38) it can be summarized that Figure-of-Merit, ZT can be determined from the temperature difference.
By following Eqs. (33)–(38), similar relations can be derived for n-type TE leg and for the TE unicouple, the thermal circuit of which is shown in Figure 5(c).
Based on the need or application, Peltier coolers come in various shapes, sizes, and performance categories. The shape and size of the TE majorly depends on the surface on which the TE cooler must be installed. Categorization based on the performance of the Peltier cooler is judged on the values of Q̇P,max, Imax and ∆Tmax. From Eq. (37), it can be inferred that there is a trade-off between the heat pumping capacity and the temperature difference. With a small ∆T value, a large amount of heat will be transferred in the Peltier circuit, and the COP is the maximum at the lowest value of ∆T. An increase in current increases ∆T, but simultaneously the Joule heat increases, and after a threshold value, the Joule heat supersedes the absorbed heat. This is known as thermal runaway and hence now the sole purpose of using the Peltier cooler has vanished. To get rid of the thermal runaway, multistage or cascaded TE (discussed in Section 7.1) cooler can be used.
As shown in Figure 6, it can be interpreted that not all the TEs are as good as both Peltier or TE coolers and thermoelectric generators. Some are good as Peltier coolers, and some are good for TEG applications. But there is a class of TE materials that are good in both types of application areas, i.e., refrigeration and power generation. To make many of the TEs multidimensional, suitable measures need to be taken from two perspectives: (1) Joule heat to be kept as minimum as possible; (2) no increment in lattice thermal conductivity due to increasing temperature. This can be achieved by suitable use of doping, nano structuration, band engineering and use of proper set of materials (e.g. use of oxide dielectric layer in between the layers of TE material) [3, 7, 15].
Figure 6.
Thermoelectric materials classification with respect to the power generation mode and refrigeration mode (Data was referred from Ref. [1, 11, 13, 21, 26, 38]).
One such example of TE leg synthesized and characterized is of Mn-doped p-type and Co-doped n-type thermoelectric (TE) β-iron disilicide ceramics [39]. These legs were fabricated for TEG using the powder metallurgy (PM) route. The reaction between iron and silicon to form TE β-iron disilicide is very sluggish [40], however, the PM route increases the kinetics of this reaction [30, 41]. The effects of variation in type and quantity of dopants (p-type dopant: Mn: stoichiometry Fe1-xMnxSi2 (x = 0.04–0.12) or n-type dopant: Co: stoichiometry Fe1-yCoySi2 (y = 0.01–0.05)) on thermoelectric properties were studied. The stoichiometric compositions studied in this research work were abbreviated as given in Table I to make the discussion more composed.
These prepared doped TE ceramic legs were studied for Seebeck coefficient, resistivity, thermal conductivity, and Figure-of-Merit with respect to temperature as shown in Figure 7(I)–(IV). The p-type leg and n-type leg with the highest Seebeck coefficient were Fe0.92Mn0.12Si2 and Fe0.95Co0.05Si2, respectively. This was due to the intrinsic p-type nature of β-FeSi2, which had a positive impact on the Seebeck coefficient of p-type compacts compared to n-type compacts. The maximum ZT values for Fe0.92Mn0.08Si2 p-type and Fe0.95Co0.05Si2 n-type compacts were 0.59 and 0.41, respectively, at 1150 K. Another important conclusion is that both p-type and n-type compacts had maximum ZT values at the same temperature, i.e., 1150 K. This indicates that for a TEG developed from these legs, the TE properties would be optimal at 1150 K and that they both will perform the best at this temperature [24].
Figure 7.
Temperature dependence of I. Seebeck coefficient, II. resistivity, III. thermal conductivity, IV. Figure-of-Merit of doped compacts (a) Mn0.04, (b) Mn0.08, (c) Mn0.12, n-type: (d) Co0.01, (e) Co0.02, (f) Co0.05 [24].
8.2 TE unicouple
A TE unicouple consists of one p-type and one n-type leg connected thermally parallel and electrically in series, as shown in Figure 8. The p-type and n-type legs are the two dissimilar TE materials that are connected using highly conducting film, silver here. The two joints are held at different temperatures. Assumptions here are:
being the same base material system, the temp distribution across p-type is same as that in n-type and
response from p-type and n-type legs to the change in temperature with respect to the physical properties is same.
Figure 8.
Single thermoelectric (TE) unicouple made of p-type and n-type legs and under steady-state heat transfer conditions. The p-type TE leg is connected electrically in series to the n-type TE leg using a silver connector. The two joints are held at different temperatures to have thermal gradient. (a) Schematic showing physical setup of TE unicouple; (b) Thermal circuit representation of TE unicouple in power generation mode [34].
The relations of Seebeck coefficient (S), thermal resistance (R), and electrical conductance (k) for a unicouple are given in Eqs. (39), (40), and (41) respectively.
S=Sp−SnE39
R=ρpLpAp+ρnLnAnE40
k=kpApLp+knAnLnE41
The voltage generated by p-type and n-type legs is respectively given as:
∆Vp=SpTH−TCE42
∆Vn=SnTH−TCE43
where ∆Vp and ∆Vn are voltages generated, and Sp and Sn are Seebeck coefficients of p-type and n-type legs, respectively.
Thus, the voltage generated, VT across TE unicouple, is determined from Eqs. (42) and (43) and is given as:
VT=∆Vp−∆Vn=Sp−SnTH−TCE44
The resulting ZT for the couple is given by Eq. (45).
ZT=Sp−Sn2Tρnkn1/2+ρpkp1/22E45
The rate of heat flow through the TE unicouple is thus:
Q̇=kp+knTH−TCAlE46
where kp and kn are thermal conductivities of p-type and n-type legs, respectively, and A and l are cross-sectional area and length of each leg, respectively. Consider that the thermal conductivities of p-type and n-type legs are nearly the same, represented as ks (ks = kp = kn), and so the rate of heat flow now becomes:
Q̇=2ksATH−TClE47
Now let us consider that the same circuit is closed using a variable load resistance RL and correspondingly an electric current I flows in the circuit. The current I flowing through the circuit is given as:
I=VTrin+RLE48
To evaluate the physical condition, let the circuit be divided into two halves, the upper hot side and the lower cold side. On the hot side and cold side, the rate of heat absorbed, Q̇H and the rate of heat dissipated, Q̇C are determined, respectively [36], as given in Eq. (48)–(49) as follows:
QḢ=Heat getting absorbeddueto flow of currentPeltier effect+Rate of heat flow across the lengthSeebeck effect−Heat transport boundary conditionJoules effect
QḢ=SITH+2ksATH−TCl−12I2rinE49
QĊ=Heat getting dissipateddueto flow of currentPeltier effect+Rate of heat flow across the lengthSeebeck effect+Heat transport boundary conditionJoules effect
QĊ=SITC+2ksATH−TCl+12I2rinE50
The power output, PL, is given as:
PL=QḢ−QĊ=SITH−TC−I2rinE51
Also, now let us determine the power output, PL, across load resistance RL.
The power of TE is proportional to the square of the temperature difference (∆T = Th - Tc), as given by Eq. (53). To increase the power output, the temperature difference needs to be increased [38]. This can be done by using a suitable cooling system. Also, from Eq. (20) and Eq. (53), maximum power output from the TE unicouple is possible only when the applied load resistance, RL, is equivalent to the internal resistance, rin.
Effective ZT determines the maximum energy conversion efficiency η, given in Eq. (55). The efficiency η of the TE unicouple is the ratio of power input to the load to the net heat flow rate. Here the heat flow is from the source to the heat sink, and so the net heat flow is positive.
η=PLQḢ=Th−TcTh1+ZT−11+ZT+TcThE55
The term (1 + ZT) 1/2 varies with the average temperature T. The important point that is concluded from efficiency η calculation is that η depends on two dimensionless quantities (Tc/Th) and ZT [42].Eq. (55) indicates that increasing efficiency requires both high ZT values and a large temperature gradient across the thermoelectric materials. A large temperature gradient can be established when the thermal conductivity is low. High thermal conductivity shorts the thermal circuit. The high Seebeck coefficient ensures a large potential/thermo voltage, the high electrical conductivity is needed to minimize the Joule heating effect, and the low thermal conductivity is needed to create a large temperature gradient [43].
8.3 TEG module
A TEG is made up of many legs of p-type and n-type TE materials that are connected electrically in series but thermally in parallel, as shown in Figure 9(a)–(b). However, their Carnot overall efficiency is very poor, typically in the range of 5%, which is very low compared to other renewable energy sources like solar energy and wind energy [23]. To obtain optimum performance, it is important to take care of several design features when making the module to be used in an application [44]. The TEG module consists of unicouples connected in a circuit as shown in Figure 10. They can be fabricated such that a pair or an unicouple is manufactured with the same set of process parameters as one unit (as shown in Figure 10(a)) or individual p-type and n-type legs are manufactured and later assembled on the TEG ceramic board using metal contacts (as shown in Figure 10(b)) [6]. In both cases, p-type and n-type TE regions of the unicouple or legs are electrically in series and thermally in parallel [16].
Figure 9.
Schematic of a thermoelectric generator module: (a) Various parts of the TE assembly; (b) Appearance of the complete TE assembly [34].
Figure 10.
Thermoelectric unicouple types: (a) p-type and n-type TE legs in physical contact; (b) p-type and n-type TE legs not in physical contact but are electrically in series via metal contact [6].
The design process of a TEG module involves energy balance calculations across the heat source, across the thermoelectric elements, and importantly, systematic analysis of the circuit (e.g., contact resistances, radiation effects, etc.) to improve output power values. These calculations certify the usefulness and importance of a dissipated thermal energy converter in the consumer market [45].
The energy balance can be expressed as:
Qsource=Qlegs+QlossE56
where, Qsource: heat absorbed from the heat source; Qlegs: heat transferred through TE legs; and Qloss: heat losses due to radiations from the heat source and heat sink, due to convection.
When properly installed, the thermoelectric device can run for thousands of hours with consistency. For competent installation, however, attention must be given to the mechanical details along with thermoelectric performance. Some important design features that need special consideration are [46]:
Compressive loading: The sample must be assembled in the circuit using compressive loading so that proper electrical connections can be established and the contact resistance can be minimized. It has been observed that 1.4 MPa pressure is enough to have proper electrical contact. Also, the p-type and n-type TE samples need to have the same height to ease the circuit formation process. The decision of compressive load depends on the thermal expansion coefficient of the p-type and n-type TE samples and the connectors. If it is different for each TE device part, then a spring of suitable tension must be used to accommodate the dissimilarity.
Heat source overhang: Since the heat source will cool near its edges, the edge of the heat source should extend well beyond the edge of the module. In a laboratory test setup, an overhang of at least 0.5 inch is necessary to avoid most of the cooling effects caused by the proximity of the edge of the heat source.
Heat distribution: To avoid nonuniform temperature distribution or formation of hot spots or cold spots at the heat source, it is helpful to place a high thermal conductivity plate like Al for low-temperature heating and SiC for high-temperature applications. This plate is to be placed between the heat source and TE device. Similarly, uniform temperature distribution is required in the heat sink, which can be achieved by using Al or Cu plate as a thermal spreader.
Area of contact: The power output is in DC form, and so area of contact is very important. The % area of contact of the conductor to the sample should be tending to 100% to have maximum power output.
The interface between the TE sample and the heat source and the interface between the sample and the heat sink are very critical components of any TE device. The surface of the device consists of the electrical conductors that join the n-type and p-type samples of the TE device. It is essential that these conductors are not electrically shorted together. Since most of the heat sources and heat sinks are electrically conducting, an insulating material must be placed at the interface.
Voltage detection is a very important aspect while determining TE module efficiency. This voltage detection is very sensitive to factors such as degree of homogeneity in the samples, position of the voltage probes, position of the thermocouples, method of contacting the sample, and precision in sample dimension determination [47].
The maximum power output, Wmax, for TEG module is determined with the condition rin = RL and is given as:
Wmax=∆T2xnAlxσS2E57
where n is the number of legs connected electrically in series.
Eq. (57) expresses the dependency of power output on temperature difference (first term), TE leg dimensions (second term), and the type of TE material (third term or the power factor) [48].
The thermal resistance θh between the heat source and the TE module is given by Eq. (58). The thermal resistance θc between the TE module and heat sink is given by Eq. (59). The thermal resistance of the TE module can be expressed as in Eq. (60). T1 and Ta are the heat source temperature and ambient temperature, respectively, as shown in Figure 10.
θh=T1−ThQh,netE58
θc=Tc−TaQc,netE59
θTEM=1mAlkp+knE60
Qh,net is the heat absorbed from the heat source by the TEG module, and Qc, net is the heat released from the cold side of the TEG module. Each unicouple is made of one p leg and one n leg, and m represents the number of couples present in the TE module. Using Newton’s law of heat transfer, Qh, net and Qc, net are determined (Eqs. (61) and (62)).
Qh,net=mSITh+Th−TcmθTEM−0.5I2rinE61
Qc,net=mSITc+Th−TcmθTEM+0.5I2rinE62
The electrical resistivity, ρs, and voltage generated, VT, by the TEG are given by Eq. (63) and Eq. (64) respectively.
r=2lmρsAE63
VT=mSp−SnTh−TcE64
The power of TE is proportional to the square of the temperature difference (∆T = Th - Tc), as given by Eq. (7). To increase the power output, this temperature difference can be increased by using thermal cascade or multistage TE. Depending on the design characteristics and TE material properties, two or more TEs are placed one above the other to form a thermal cascade and increase the ∆T. In such cases, the equivalent temperature difference, ∆Te, is considered as given in Eq. (65).
∆Te=∆T12+∆T22+∆T32+…E65
where ∆T1, ∆T2, ∆T3, …are the temperature differences across the cascade TE layers TE layer 1, TE layer 2, TE layer 3, … respectively.
The TEG module was designed from p-type and n-type legs of β-FeSi2, as shown in Figure 11(a). The electrical analogy for the TEG module is shown in Figure 11(b). On both the top and bottom surfaces of the TEG circuit, the SiC plate, alumina plate, and silver foil are represented as capacitors with their respective resistances. Across the TEG circuit, Th and Tc temperatures are maintained, which results in charge flow. The electrical circuit with gray background summarizes the circuit to determine the Seebeck coefficient. The actual laboratory-scale working model of the TEG is shown in Figure 11(c). Around 29 samples, 15 p-type and 14 n-type, were arranged alternately on a glass wool brick. Insight is the image of a red LED got lightened up by the power provided by the TEG module in the temperature range of around 650–870 K. To determine the voltage and hence power generation from the TEG module with respect to temperature, the red LED was replaced with a Keithley 81/2-digit multimeter and the experiment was repeated [34].
Figure 11.
TEG module (a) schematic representation of the TEG module, (b) electrical analogy of the TEG module, (c) working model of the TEG module [6].
The reported lifetime for TEG modules is 11–30 years, but the lifetime can be reduced if the modules are exposed to repeated hot side temperature changes. Repeated heating–cooling cycles may cause material deterioration of a TEG, so it is preferable to use TEGs in continuous heat flows [8].
9. Thermoelectric material applications in renewable
Photovoltaic-Thermoelectric Generator (PV-TE) is one of the emerging applications of TE in renewable energy resources. From last few decades, the usage of photovoltaic (PV) cells for electricity generation increased appreciably owing to its environmental sustainability, availability, and applicability. However, in the PV, most of the solar energy gets dissipated in the form of heat into the PV system. This waste heat causes an increase in temperature of the PV cell and degrades its efficiency and performance. In the current context, efforts are made to improve efficiency by utilizing this dissipated heat. One of the ways is to use low temperature thermoelectric (TE), based on the Seebeck effect, along with the PV cell. The overall efficiency with respect to temperature of hybrid Photovoltaic-Thermoelectric (PV-TE) is 0.6–1.91% more than PV alone [49].
Both PV and TE systems depend strongly on temperature, and thus, to reduce the temperature and improve the performance of the PV-TE hybrid system, an effective and affordable direct evaporative cooling (EC) system was used in this study. The mathematical model studied for performance evaluation was based on cascaded hybrid Photovoltaic-Thermoelectric-Direct Evaporative Cooling System (PV-TE-EC) combinations, i.e., PV-EC (without TE), PV-TE-EC, PV-2TE-EC (two cascaded layers of TE sandwiched between PV and EC), and PV-3TE-EC (three cascaded layers of TE sandwiched between PV and EC as shown in Figure 12) systems. The TE module placed adjacent to the lower surface of the PV is fabricated from p-type and n-type bismuth telluride ((ZT) ∼1). Various hybrid systems made of PV and TE assisted or not assisted with EC are studied, e.g., PV, PV-TE, PV-EC, PV-TE-EC, PV-2TE-EC, and PV-3TE-EC. These hybrid systems used PV, TE, and EC in parallel arrangements [49].
Figure 12.
Schematic representation of the structural diagram of the PV-3TE-EC system [49].
The individual and overall power output contributions from PV and TE in the hybrid systems with respect to the concentration ratio are summarized in Figure 13. As the concentration ratio increased, the overall power output of all the hybrid systems increased. An increase in solar concentration ratio increases the overall temperature of the hybrid systems. In addition, the temperature rise augments the evaporative cooling effect from EC and results in a higher equivalent temperature difference across TEs. This reduces the power output from PVs but increases the power output from TEs. This leads to an increase in the overall efficiency of PV-TE-EC, PV-2TE-EC, and PV-3TE-EC in the ascending order.
Figure 13.
Power output from the hybrid systems with respect to solar concentration ratios [49].
Another emerging application of TE in renewable energy solutions is Solar Thermoelectric Generator (STEG). STEG can be used to generate electricity from tapped solar energy, and therefore they are an attractive renewable energy solution for standalone power conversion [50]. In addition, STEG has been used in power generation for health monitoring systems, wireless sensors, space applications, and several other low power applications [51]. In this, solar concentrators are placed over the heat source to conduct the thermal energy to TE and to perform energy conversion analysis of the TE at various solar concentration ratios. The energy conversion efficiency was found to be around 15% and 23% for TE material with ZT 1 and 2, respectively. A modification to the setup of STEG was also tried in which the solar energy was concentrated on three-layered TE (first layer: Bismuth telluride, second layer: Lead telluride, third layer: silicon) using metamaterial. The Finned heat sink was used to increase the efficiency of conversion of solar energy to electricity to around 24%. However, the consideration over here was that all the energy falling on the metamaterial layer is reaching the TE top surface [2].
Thermoelectric ceramic material is a need of the future as it provides a means for harvesting waste heat energy and is also a clean form of energy. Many techniques have been employed for the synthesis of TE, a potential candidate for scavenging energy, but for many of the varieties of TEs, there is no such specific route that can lead us to the synthesis of this material on an industry level. Also, a unicouple or TEG or TEC needs to be formed of p-type and n-type thermoelectric material to achieve a higher value of Figure-of-Merit, Z.
Some TE materials contain toxic elements or less abundant elements, which narrow down their scope of application [52]. An ideal TE material should be a phonon glass and an electron crystal, which means that the materials should have a low lattice thermal conductivity just like glass and a high electrical conductivity just like in a crystal and at the same time should be environmentally friendly. Many types of TE have attracted attention in the last few years because of their technological potential in the fields of optoelectronics, light emitting diodes (LEDs), medical, aerospace, and defense applications due to the thermoelectric conversion properties, chemical stability, non-toxicity, and relatively low cost of production. There is a huge scope to do research in the choice of the element and its quantity that need to be added to get high ZT thermoelectric material.
References
1.Dey S, Sreenivasulu A, Veerendra GTN, Rao KV, Babu PSSA. Renewable energy present status and future potentials in India: An overview. Innovation and Green Development. 2022;1(1):100006. DOI: 10.1016/j.igd.2022.100006
2.Riffat SB, Xiaoli M. Thermoelectrics: A review of present and potential applications. Applied Thermal Engineering. 2000;23:913-935. DOI: 10.1016/s1359-4311(03)00012-7
3.Alam H, Ramakrishna S. A review on the enhancement of figure of merit from bulk to nano-thermoelectric materials. Nano Energy. 2013;2:190-212. DOI: 10.1016/j.nanoen.2012.10.005
4.Chao Han, Zhen Li, S. X. Dou, Recent progress in thermoelectric materials, Australian Institute for Innovative Materials (Chinese Science Bulletin), 2014;59(18):2073-2091. DOI: 10.1007/s11434-014-0237-2
5.Lange H. Electronic structure and interband optical properties of β-FeSi2. Thin Solid Films. 2001;381:171-175. DOI: 10.1016/S0040-6090(00)01739-9
6.V.S. Poddar, Influence of additives on transformation characteristics and thermoelectric behaviour of β-iron Disilicide [thesis]. India: Dept. of Meta and Matls Sci.: College of Engineering Pune, SPPU University; 2019.
7.Ohtaki M. Nanostructured oxide thermoelectric materials with enhanced phonon scattering. Oxide Thin Films, Multilayers, and Nanocomposites. 1st ed. Cham: Springer; 2015. pp. 109-122. DOI: 10.1007/978-3-319-14478-8_7
8.Johansson MT, Söderström M. Electricity generation from low-temperature industrial excess heat—An opportunity for the steel industry. Energy Efficiency. 2013;7(2):203-215. DOI: 10.1007/s12053-013-9218-6
9.Rowe DM. CRC Handbook of Thermoelectric. 1st ed. Boca Raton: CRC Press; 1994. pp. 1-20
10.Abram Ioffe LS. Stil'bans, physical problems of thermoelectricity. Reports on Progress in Physics. 2002;22:167-203. DOI: 088/0034-4885/22/1/306
11.Kishore RA, Priya S. A review on low-grade thermal energy harvesting: Materials, methods and devices. Materials (Basel). 2018;11(8):1433. DOI: 10.3390/ma11081433
12.Jérôme D, Pierre A, Simon W, Ndjaka J. Influences of thermal cycles on thermoelectric Uni-leg modules made by Ca3Co4O9/Ca0.95Sm0.05MnO3 oxides. Journal of Taibah University for Science. 2014;8(4):1-10. DOI: 10.1016/j.jtusci.2014.05.003
13.Poddar VS, Dhokey NB. Thermoelectric properties of iron Disilicide and manganese silicide: Synthesis and characterization. Transactions of the Indian Institute of Metals. 2019;72(10):2711-2719. DOI: 10.1007/s12666-019-01743-8
14.Kasap S. Principles of Electronic Materials and Devices. 3rd ed. New York: McGraw-Hill; 2005. pp. 373-443. DOI: 10.5555/1594045
15.Gibbs Z, Hyun-Sik K, Heng W, Jeff S. Band gap estimation from temperature dependent Seebeck measurement—Deviations from the 2e|S|maxTmax relation. Applied Physics Letters. 2015;106:022112. DOI: 10.1063/1.4905922
16.Tritt T. Thermoelectric materials, phenomena, and applications: A Bird’s eye view. Annual Review of Materials Research. 2011;41:433-448. DOI: 10.1557/mrs2006.44
17.Kim HJ, Skuz J, Park Y, King G, Choi S, Nagavalli A, System to Measure Thermal Conductivity and Seebeck Coefficient for Thermoelectrics, NASA STI program - Technical Memorandum, NASA/TM–2012-217791
18.Resistivity by Four Probe Method [Internet]. 2013. Available from: http://vlab.amrita.edu/?sub=1&brch=282&sim=1512&cnt=1 [Accessed: March 12, 2022]
19.Zhao L-D, Dravid VP, Kanatzidis MG. The panoscopic approach to high performance thermoelectrics. Energy and Environmetal Science. 2014;7:251-268. DOI: 10.1039/c3ee43099e
20.Sootsman J, Chung DY, Kanatzidis M. New and old concepts in thermoelectric materials. Angewandte Chemie International Edition. 2009;48:8616-8639. DOI: 10.1002/anie.200900598
21.Klochko NP, Klepikova KS, Zhadan DO, Petrushenko SI, Kopach VR, Khrypunov GS, et al. Structure, optical, electrical and thermoelectric properties of solution-processed Li-doped NiO films grown by SILAR. Materials Science in Semiconductor Processing. 2018;83:42-49. DOI: 10.1016/j.mssp.2018.04.010
22.Fedorov M. Thermoelectric silicides: Past present and future. Journal of Thermoelectricity. 2009;2:51-60
23.Zabek D, Morini F. Solid state generators and energy harvesters for waste heat recovery and thermal energy harvesting. Thermal Science and Engineering Progress. 2019;9:235-247. DOI: 10.1016/j.tsep.2018.11.011
24.Poddar VS, Dhokey NB. Evaluation of thermoelectric properties of doped β-iron Disilicide prepared by the powder metallurgy technique. Transactions of the Indian Institute of Metals. 2021;74:399-410. DOI: 10.1007/s12666-020-02167-5
25.He J, Sootsman JR, Girard SN, Zheng J-C, Wen J, Zhu Y, et al. On the origin of increased phonon scattering in nanostructured PbTe based thermoelectric materials. Journal of the American Chemical Society. 2012;132(25):8669-8675. DOI: 10.1021/ja1010948
26.Balout H, Boulet P, Record MC. Thermoelectric properties of Mg2Si thin films by computational approaches. The Journal of Physical Chemistry C. 2001;118(34):19635-19645. DOI: 10.1021/jp506323r
27.Salam L, Matthews R, Robertson H. Fabrication of iron disilicide (FeSi2) thermoelectric generator by the tape casting method. Materials and Design. 1999;20:223-228. DOI: 10.1016/S0261-3069(99)00031-X
28.Chen HY, Zhao XB, Lu YF, Mueller E, Mrotzek A. Microstructures and thermoelectric properties of Fe0.92Mn0.08Six alloys prepared by rapid solidification and hot pressing. Journal of Applied Physics. 2003;94(10):6621-6626. DOI: 10.1063/1.1622773
29.Suryanarayana C. Mechanical alloying and milling. Progress in Materials Science. 2001;46:119-184. DOI: 0079-6425/01/S
30.Poddar VS, Dhokey NB, Garbade RR, Butee SP, Deep Prakash RD, Purohit. Rapid production of iron disilicide thermoelectric material by hot press sintering route. Materials Science in Semiconductor Processing Journal. 2017;71:477-481. DOI: 10.1016/j.mssp.2017.06.027
31.Zhou Z, Uher C. Apparatus for Seebeck coefficient and electrical resistivity measurements of bulk thermoelectric materials at high temperature. Review of Scientific Instruments. 2005;76:023901-023901-5. DOI: 10.1063/1.1835631
32.Bergman TL, Lavine AS, Incropera FP, DeWitt DP. Introduction to Heat Transfer. 6th ed. Jefferson City: Wiley; 2011. p. 917. Available from: books.google.co.in/books?id=YBaNaLurTD4C
33.Cangel Y, Ghajar A. Heat and Mass Transfer- Fundamentals and Applications. 5th ed. New York: Mc-Graw Hill; 2015. p. 939
34.Poddar VS. Eco-friendly transition metal silicide for high temperature thermoelectricity. Journal of Materials Science: Materials in Electronics. 2023;34:1-11. DOI: 10.1007/s1085F164-023-11472-w
35.Skomedal G, Holmgren L, Hugh Middleton IS, Eremin GNI, Jaegle M, Tarantik K, et al. Design, assembly and characterization of silicide-based thermoelectric modules. Energy Conversion and Management. 2016;110:13-21. DOI: 10.1016/j.enconman.2015.11.068
36.Bonin R, Boero D, Chiaberge M, Tonoli A. Design and characterization of small thermoelectric generators for environmental monitoring devices. Energy Conversion and Management. 2013;73:340-349. DOI: 10.1016/j.enconman.2013.05.016
37.Zhou Z-G, Zhu D-S, Hong-Xia W, Zhang H-S. Modeling, experimental study on the heat transfer characteristics of thermoelectric generator. Journal of Thermal Science. 2013;22(1):48-54. DOI: 10.1007/s11630-013-0591-4
38.Liu Y, Xu L, Zhao C, Shao M, Hu B. Tuning the Seebeck effect in C60-based hybrid thermoelectric devices through temperature-dependent surface polarization and thermally-modulated interface dipoles. Physical Chemistry Chemical Physics. 2017;19(22):14793-14800. DOI: 10.1039/c7cp01736g
39.Katsuyama S, It M, Nagai H. Preparation of the composite thermoelectric materials with small particles dispersion by MA. Kona Powder and Particle Journal. 2004;22:186-194. DOI: 10.14356/kona.2004022
40.Vining C. Thermoelectric Properties of Silicides. 1st ed. California: CRC Press; 1995. pp. 1-18. DOI: 10.1201/9781420049718-23
41.Vaishali Poddar NB, Dhokey SP, Butee NB, Revade MM, Thombre RD, Purohit DP. Evolution of thermoelectric β – FeSi2 phase by Cryo milling and sintering. Transaction of Indian Institute of Metals. 2017;70(1):167-174. DOI: 10.1007/s12666-016-0873-0
42.Ware RM, McNeill DJ. Iron disilicide as a thermoelectric generator material. Journal IEE. 1964;111(1):518. DOI: 10.1049/piee.1964.0029
43.Gayner C, Kar KK. Recent advances in thermoelectric materials. Progress in Materials Science. 2016;83:330-382. DOI: 10.1016/j.pmatsci.2016.07.002
44.Sahin AZ, Yilbas BS. The thermoelement as thermoelectric power generator: Effect of leg geometry on the efficiency and power generation. Energy Conversion and Management. 2013;65:26-32. DOI: 10.1016/j.enconman.2012.07.020
45.Rowe DM, Min G. Design theory of thermoelectric modules for electrical power generation. IEEE Proceedings - Science, Measurement and Technology. 1996;143(06):351-356. DOI: 10.1049/ip-smt_19960714
46.Leavitt F, Elsner N, Bass J. Use, application and testing of hi-Z thermoelectric modules. In: Proceedings of the 15th International Conference on Thermoelectrics. Pasadena, CA, USA: ICT ‘96; 1996. pp. 378-382. DOI: 10.1109/ICT.1996.553508
47.Benedict TS, Angello SJ, Berman R, Adam G, Apker L, et al. Methods in experimental physics. In: Lark-Horovitz K, Johnson VA, editors. 1st ed. UK: Academic Press; 1959. pp. 1-143
48.Namir D, Beck J. On the significance of the thermoelectric figure of merit Z. Journal of Electronic Materials. 2012;39(9):1897-1901. DOI: 10.1007/s11664-009-1060-4
49.Poddar VS, Ranawade VA, Dhokey NB. Study of synergy between photovoltaic, thermoelectric and direct evaporative cooling system for improved performance. Renewable Energy. 2021;182:817-826. DOI: 10.1016/j.renene.2021.10.040
50.Li G, Shittu S, Ma X, Zhao X. Comparative analysis of thermoelectric legs optimum geometry between photovoltaic-thermoelectric and solar thermoelectric. Energy. 2019;171:599-610. DOI: 10.1016/j.energy.2019.01.057
51.Chen X, Zhou Z, Lin Y-H, Nan C. Thermoelectric thin films: Promising strategies and related mechanism on boosting energy conversion performance. Journal of Materiomics. 2020;6(3):494-512. DOI: 10.1016/j.jmat.2020.02.008
52.Wills BA, Napier-Munn T. Mineral Processing Technology. 7th ed. UK: Elsevier; 2006. p. 3. DOI: 10.1016/B978-0-7506-4450-1.X5000-0
Written By
Vaishali Poddar
Submitted: 09 September 2024Reviewed: 23 October 2024Published: 20 December 2024
Open access peer-reviewed chapter
