Investigation on the Heat-to-Power Generation Efficiency of Thermoelectric Generators (TEGs) by Harvesting Waste Heat from a Combustion Engine for Energy Storage

,


Introduction
Nowadays, the increasing demand for energy requirements and increase in energy prices is a prime challenge to create energy security in the increasingly competitive global market by adopting stringent/stagnate emission regulations.The automobile industry and power plants are the largest emissions of waste heat to the atmosphere, leading to global warming and change in the climate.These industries' heat energy wasted on the environment is identified as low-grade heat.In recent days, many researchers have been focusing their attention on the recovery of this waste heat.The recovered heat can be utilized for many applications like preheating the air, preheating water, turbocharging, desalination, and other purposes.In internal combustion engines (IC engines), the maximum efficiency is around 25-30% which indicates that the remaining 70-75% of the energy is gone as heat by engine coolant and exhaust gas [1].Hence, there is an excellent potential to reuse the exhaust waste heat.
This low-grade wasted heat energy can be straightly transformed into electric energy by means of a TEG and minor modifications to the waste heat source.The TEG is a device used to transfer thermal energy into electric energy based on the Seebeck effect [2].Energy generation using TEG has moveless parts and free to vibration and has no frictional loss with very lightweight and durable technology [3].For generating electric power, the TEG is fixed between this source and sink (i.e., hot and cold side).Due to the difference in temperature between the source and sink, heat will flow through the TE module.This module converts the heat energy into electricity till the temperature difference is maintained [4,5].
Many researchers tried to investigate the use of TEG in various heat sources such as gasoline engines, automobile exhaust systems, engine radiators, industrial driers, biomass power plants, solar systems, fuel cells, and domestic and industrial chimneys.Famous car manufacturing companies such as Ford, Honda, and BMW are focusing their research on the WHR in the exhaust of automobiles [6][7][8][9].Orr et al. [10] designed and fabricated an industrial drier-type heat source with TEG, producing 6.03 W electric power with a conversion efficiency of 1.43%.Singh et al. [11] analyzed the stored heat in the solar pond and extracted it by using a heat pipe arrangement and power generation of TEG.In this analysis, 3.2 W power was produced from 16 TEGs.
Date et al. [12] showed a power generation of 3.02 V at the difference in temperature of 75 °C in the TEG by conducting experiments on solar water heating combined with a thermoelectric system.Zhao et al. [13] established a mixture structure of a direct carbon fuel cell joined with a TEG and a regenerator.This structure could generate 50% more equivalent extreme power than a normal direct carbon fuel cell system.Nuwayhid et al. [14,15] also found a potential for WHR in a domestic wood stove.In this setup, a maximum of 4.2 W power was produced by a single TEG model.Remeli et al. [16,17] designed and modeled an experimental setup by combining the heat pipes and bismuth telluridebased TEGs.This model produced 10.39 W of electric power against the recovered waste heat of 1.345 kW.
Some researchers focused on the optimization of the TEG model, optimization of geometric parameters of thermocouples, optimization of the design, use of new material for TEG, and different structure of TEG.Espinosa et al. [18] identified and used engineering equation solver (EES) software to implement the thermoelectric model based on a finite difference method.The parameters of TEG like size, connection, and material proportion influenced the output power of TEGs.Huang et al. [19] designed a concentric cylindrical TEG system with an annular TE module instead of a typical square-shaped TE module.The simulations were performed using the CFD software to find the performance of the new proposed design of the TEG system.The performance was compared between the gas-inside method and the water-inside method.The higher output power of 29.8 W was gained in the water-inside method, whereas only 4.8 W of power was gained by the gas-inside method.
Ragupathi et al. [20] investigated and analyzed the parameter optimization in the performance of TEGs by using the Taguchi method.The heat input, temperature difference, and TEG material were taken as process variables at three levels.The higher energy conversion efficiency of 2.45% was achieved at 90 W of heat input, TEG made by Bi 2 Te 3 material, and 75 °C of temperature difference.Gou et al. [21] and Wu et al. [22] did theoretical and experimental analyses on the low-temperature WHR by using TEG in the industrial sector.It was found that to increase the performance of TEG, there should be an increase in the heat sink surface area and the cold side heat transfer rate.Liang et al. [23] compared and concluded that double-stage TEG had more output power and also more conversion efficiency when compared to single-stage TEG.They also showed that the number of thermoelectric increases; then, there is an increase in output power and heat absorption.
Eddine et al. [24] found a greater TEG power output of about 10-30% while the waste gases of engine exhaust were utilized as a heating medium compared with hot air.They analytically investigated and showed that the better power output was owed to the increase in the heat transfer coefficient of convection between the engine exhaust gas and the outer surface of TEG.The engine's exhaust gas has a higher C p value than hot air, and the existence of pulsations in the exhaust gas was identified as a key parameter in this investigation.Jang and Tsai [25] optimized the spacing in the TEG module and corresponding spreader thickness to maximize the power density.These two parameters were highly related to the heat transfer coefficient of waste gas and less dependent on the difference in temperature between the maximum and minimum temperature sides.Chen and Lin [26] compared and simulated the performance of TEGs with material properties like the Seebeck coefficient, thermal conductivity, and electrical resistivity.It was found that at different temperature levels, the TEG performance changed.In further analysis, the output power of TEGs was varied if the TEG properties are seeming to be dependent on temperature.Ahmed et al. [27] obtained the results from their study and concluded that the output power of TEG and conversion efficiency was increased by increasing the heat input rate.This improvement in TEG performance is due to the increase in temperature difference between the higher and lower temperature sides of TEG.
Zoui et al. [28] developed a test bench setup to evaluate the capability of the module to produce electrical power from hot water against forced ambient air.16 mW output power was produced at 600 mV open-circuit voltage with a temperature difference of 60 °C.
Zhao et al. [29] developed a heat pipe thermoelectric generator and analyzed the heat transfer characteristics and temperature consistency of the generator corresponding to the thermal power of the heat pipe evaporator and the number of modules used in the condenser.With an increase in 2 International Journal of Energy Research the heating power, a decrease in the thermal resistance of the heat pipe was observed with higher heat-to-electric conversion efficiency.Jame et al. [30] investigated enhancing the conversion efficiency of thermoelectric generators and to absorb the unsteady state heat.Their work shows the role of nanostructured graphite (G) and graphene oxide (GO) decorated phase change material (PCM) of D-mannitol connected with the hot side of the thermoelectric module on the improvement of conversion efficiency and TEG power output.
Ji et al. [31] proposed a simulation model and a design procedure for a solar thermoelectric generator.Various geometrical parameters, such as the height, the fill ratio, the ratio of the cross-section area of n-type material over ptype material of thermoelectric module, and the solar concentration ratio, were investigated under different working conditions.In this analysis, an L 27 (3 3 ) orthogonal array was used to evaluate the design parameters returning the maximum output power.The optimal design parameter set achieved a power output of 5.47 W compared with the power output of 1.95 W from the initial design.
This research is focused on electricity generation from the waste heat produced by the IC engine, which deals with fluid-to-solid heat transfer.Finned copper rod and TEGs are utilized for electricity generation.Here, finned copper rods were used for transferring the heat, and TEGs were utilized to produce electric power due to the difference in temperature across its terminals.A heat exchanger was fabricated by utilizing an absorber rod made of copper to absorb heat from the exhaust gas.During the experiment, the higher temperature side of the TEG was connected to the hot side of the absorber plate, and the lower temperature side of the TEG was joined to the cold side of the absorber plate.The connection was made in such a way that the higher and lower sides should have more temperature difference which may lead to generating more voltage and current.The experiments were conducted by changing the load on the engine, which is directly proportional to the temperature of exhaust gas.The load on the IC engine was varied in the range of 0% to 100% in the steps of 25%.Also, the rate of heat recovery and power generation and the conversion efficiency of different types of TEGs at different heat exchanger zones were analyzed with the variation in the engine load.

Methodology
2.1.Thermoelectric Generator.The thermoelectric (TE) phenomenon is a solid-state energy conversion that can translate thermal energy to electrical energy and vice versa with the help of thermoelectric materials.Thermoelectric devices produce an electric voltage when there is a difference in temperature on each side.For many years, TEGs have been used as a trustworthy source of power production.
A TE device contains n-type and p-type semiconducting materials joined in series for electric and parallel for thermal.TEGs make use of the Seebeck effect, which produces a voltage while one side of the TEG is kept at a higher temperature than the other side owing to the casual thermal motion of charge carriers, which causes the electric current to flow when the circuit is closed.The working mechanism of TEG is shown in Figure 1.
The p-type elements are made as positive charge carriers (holes) and doped to get the positive Seebeck coefficient, whereas the n-type elements are made as negative charge carriers (electrons) and doped to get the negative Seebeck coefficient.When a p-type element is connected electrically to the n-type element, then the movable holes in the p-type element "see" the movable electrons in the n-type element and travel just to the further side of the connection.For each hole that travels into the n-type element, an electron from the n-type element travels into the p-type element.
TEG modules are intended to deliver the power of DC to an extensive variety of devices with good reliability.TEGs use unique structure techniques to survive an increased quantity of thermal cycles and 200 °C nonstop hot side operations.TEGs are used in various industries like automobile, health care, oil, gas, lubricants, mining, and telecommunications.They are also utilized in space cooling systems, remote  The performance of TE material is described by "Z" which is known as the factor of merit, and the figure of merit is defined by the product ZT under a given operating temperature.
where T = ðT hs + T cs Þ/2(T hs is the temperature on the hot side, and T cs is the temperature on the cold side), S is the Seebeck coefficient, σ is the electrical conductivity, andƙ is the thermal conductivity.
The power produced by the TEG can be found and followed as where I is the generated current and V is the generated voltage.
The rate at which heat was transferred to the TEG can be calculated as follows.
Q in is calculated by using the following formula: where ƙ is the thermal conductivity, A is the area of heat transfer, and ∂T/∂x is the temperature gradient.
The energy conversion efficiency of TEG is obtained by where P TEG is the electric power generated by TEG and Q in is the rate at which heat is transferred in the TEG.The power generated by TEG was calculated by measuring the voltage and current at the two terminals of TEG.For this purpose, a multimeter was used to collect the data.

Experimental Setup.
The layout of the experimental setup is shown in Figure 2. A single-cylinder diesel engine of the Kirloskar make, water-cooled, four strokes, with 87.5 mm of cylinder diameter, 110 mm of stroke length, at a rated power of 8 hp, and speed of 1800 rpm was used for the experiment.The technical specifications of the combustion engine are given in Table 1.An electrical dynamometer     was attached to the engine for varying the load.The waste heat from the exhaust of the combustion engine was trapped to generate electricity by installing TEG.For this purpose, four different TEGs were chosen to make a comparative analysis of the effectiveness of individual TEGs and to obtain optimum conditions at which the maximum amount of waste heat can be trapped to generate useful energy.Copper fins were fabricated and connected to big size copper absorber plates with the help of a copper rod to transfer waste heat from the exhaust of the combustion engine.In this investigation, eight numbers of copper absorber blocks were used in which the TEGs were placed to convert heat to electricity.The C11000 grade of copper was used for making the finned copper rod.It is having a thermal conductivity of 300 W/m-K and a specific heat capacity of 385 J/kg-K.The detailed specifications of the copper block, copper rod, and fins are given in Table 2. Figure 3 illustrates the photograph of ducts used for the experiment.The exhaust gas supply ducts were made up of two-layered mild steel sheets, which were insulated using glass wool as an insulation packing material between two layers of steel sheets.The insulated ducts may minimize the unaccounted heat loss from the setup, which may provide consistency in the data collection and analysis of the results.Provisions were made in the duct to insert the finned copper rods.The finned copper rod and absorber plates were fabricated, as shown in Figure 4.The specifications of the copper block, duct, and glass wool are listed in Table 2.The fins were provided over the copper rod to increase the contact surface for heat transfer from the exhaust gas of the IC engine.The TEG was inserted between the two copper blocks to get temperature differences to generate electricity.
The inlet duct, exhaust duct, finned copper rod, copper absorber plates, and TEGs were fixed on the experimental setup to recover the waste heat.In this, the hot exhaust duct has four numbers of finned copper rods, and the cold inlet duct has four numbers of finned copper rods (shown in Figure 5).The finned copper rods in the exhaust duct are heated by the exhaust gas, which comes from the IC engine.The finned copper rods in the inlet duct are cooled by the inlet air going to the IC engine (shown in Figure 6).Here, these two finned copper rods have various temperatures; hence, a temperature difference occurs.A flow control valve was used to limit the flow of hot exhaust gas to the waste heat recovery setup, which was predetermined to operate in the safe operating range of the TEGs.With the increase in the load, the exhaust gas temperature was observed to elevate, so by partially closing the flow control valve, the quantity of exhaust gas supply to the experimental setup was limited as per the requirement.However, when the engine was operated at zero load or low load conditions, the flow control valve was widely opened.This flow control valve was manually operated during each set of experiments.
The TEG can be utilized to produce electricity by using this temperature difference so that it was inserted between the absorber plates of these two finned copper rods.One set of finned copper rod consists of a TEG sandwiched between the copper absorber plates.Hence, four TEGs were    During the experiment, the heat exchanger was divided into four zones.To create the temperature gradient, one of the copper blocks was fitted to the higher temperature side of the TEG, which gets heat from the hot exhaust gas through copper rod and fins.Another copper block was fixed to the lower temperature side of the TEG, which is joined to the intake manifold of the engine to remain cool with the help of intake air.The four zones are named as zone 1, zone 2, zone 3, and zone 4, as shown in Figure 2. The gap between each zone is fixed as 150 mm.Four different types of TEGs, namely, aluminium oxide (Al 2 O 3 ), bismuth telluride (Bi 2 Te 3 ), lead telluride (PbTe 3 ), and silicon germanium (SiGe), were taken for analysis on waste heat recovery, power generation, and conversion efficiency concerning variation in engine load.The temperature of the exhaust gas increases while the load on the engine increases.In general, the hot exhaust gas temperature will be in the range of 50 °C to 350 °C.A partial amount of hot exhaust gas was only allowed to pass inside the heat exchanger duct.A flow control valve was used to limit the amount of hot exhaust gas to maintain the temperature with respect to load variation.The remaining gas was passed into the atmosphere.The TEGs can withstand the maximum temperature of 138 °C at the hot side.Therefore, the temperature of 135 °C is selected for the maximum load of the engine.When the load on the engine decreases, the exhaust gas temperature also decreases.So, the reduction in exhaust gas temperature is fixed as 10 °C for the remaining load variation on the engine.The temperature of this exhaust gas passed through the outlet duct is fixed as 95 °C, 105 °C, 115 °C, 125 °C, and 135 °C for the engine loads of 0%, 25%, 50%, 75%, and 100%, respectively.

Experiment
Procedure.The hot exhaust gas from the IC engine was allowed to go through the exhaust duct, and the air from the atmosphere was passed through the inlet duct to the engine.The experiments were carried out by changing the engine load starts from 0% to 100% with a step increase of 25%.The load of the engine was varied by using an electrical dynamometer.It consists of a stator which fitted a number of electromagnets and a rotor disc made of copper or steel and coupled to the output shaft of the engine.When the rotor rotates, eddy currents are produced in the stator due to magnetic flux set up by the passage of field current  7 International Journal of Energy Research in the electromagnets.The load is controlled by regulating the current in the electromagnets.During the experiments, the steady state was attained after 10 minutes of continuous engine running.The experimental observations were taken during the stable state conditions of each load.During the experiment, four heat exchanger zones were fitted with the same type of TEG, and the observations were found at various load settings of the engine.The same procedure was repeated for other TEG materials.
The following readings were taken during the experiments of each load: (i) The temperature of cold air at entry and exit of all zones (i.e., at zone 1, zone 2, zone 3, and zone 4) was measured by using an infrared thermometer.The technical details of the infrared thermometer are given in Table 3 (ii) The temperature of hot exhaust waste gas at entry and exit of all zones (i.e., at zone 1, zone 2, zone 3, and zone 4) by using an infrared thermometer (iii) Generated voltage and current by using multimeter (the technical details of multimeter are given in Table 4)

Results and Discussion
3.1.Analysis of Temperature Variation.The variation of temperature with the variation of engine load for different zones and TEGs is shown in Figure 7. From the figure, it is observed that, for the case of all TEGs, zone 2 gives a high temperature compared to zone 1 because zone 1 is near the entry of flue gas which leads to excess carbon deposit and reduces the rate of heat transfer.Apart from this, zone 2 gets added advantage of a high heat transfer rate from the turbulence behavior of the flue gas, which is inhabited by the fins of zone 1. Subsequently, zone 3 and zone 4 show a reduction in heat transfer compared to zone 2. This is due to the subsequent decline in flue gas temperature due to the absorption of heat when it passes through zone 1 and zone 2.
From the figure, it is also observed that Bi 2 Te 3 gives Bi 2 Te 3 at 75% operating load of the engine gives an increase in temperature of about 4.2%, 1.7%, and 3.4% in comparison to Al 2 O 3 , PbTe 3 , and SiGe.It is also observed that the cold side TEG temperature increases linearly from entry to exit as well as with respect to the increase in load.This is due to heat conduction from the TEG interface.

Analysis of Heat Recovery
Rate.The heat recovery rate for the TEGs Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe with respect to variation in engine load at different zones is exposed in Figure 8. From the figure, it is noticed that the heat recovery rate grows with respect to a rise in load on the engine for the case of all TEGs.This is owing to an increase in flue gas temperature at the high operating load on the engine.From the figure, it is also noticed that the rate of heat recovery for the TEGs is maximum at zone 2. This is due to the high turbulence behavior of flue gas near zone 2 induced by the fins of zone 1. Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe give an increase in heat recovery rate of 1.8%, 3.4%, 1.2%, and 0.6% at zone 2 of 75% operating load, in comparison to 100% working load, respectively.Apart from this, it is also observed that Bi 2 Te 3 gives an overall increase in heat recovery rate in comparison to Al 2 O 3 , PbTe 3 , and SiGe throughout the load spectrum and operating conditions with respect to different zones.This is due to low resistivity, good electrical conductivity, and good Seebeck coefficient of Bi 2 Te 3 compared to Al 2 O 3 , PbTe 3 , and SiGe.The heat recovery rate for Bi 2 Te 3 at zone 2 and 75% operating load is higher by about 5.8%, 10.8%, and 17.9% and 4.6%, 7.2%, and 14.1% in comparison to Al 2 O 3 , PbTe 3 , and SiGe at 75% and 100% of operating load, respectively.

Analysis of TEG Power
Generation.The power generated by the TEGs Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe with respect to variation in engine load at different zones is indicated in Figure 9. From the figure, it is identified that the rate of power generation by TEG increases with a rise in engine load for all the zones.This is owed to the rise in temperature of the flue gas at high operating load conditions of the engine.Apart from this, during high operating load conditions, engine consumes more intake air which subsequently reduces the temperature on the cold side of TEG in each zone.This cooling effect results in to increase in the temperature difference (ΔT) between the hot side (source) and cold side (sink) of TEG.As per the operating conceptualization of TEG higher, the value of ΔT leads to increased power generation owing to the enhanced Seebeck effect.The maximum power generation by Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe was found to be 6.35 W, 7.18 W, 6.41 W, and 5.93 W at zone 2 while the engine was operated at 75% of load, respectively.This enhanced power production by the TEGs at zone 2 indicates that a maximum amount of heat transfer is happening to zone 2 compared to other zones.This is due to effective heat transfer induced by the turbulence behavior of flue gas adjacent to the fins of zone 2. Among all the TEGs, Bi 2 Te 3 was found to generate maximum power irrespective of the engine operating load.13.8%, 11.9%, and 21.04% of enhancement in power generation were observed for Bi 2 Te 3 at 75% of engine load at zone 2 in comparison to Al 2 O 3 , PbTe 3 , and SiGe.This is due to the material characteristics of Bi 2 Te 3 , which induce a more significant number of free electrons and holes to generate and move toward the terminals that leads to a higher amount of power generation.

Analysis of Conversion Efficiency of TEG.
The efficiency of energy conversion from heat to electricity for the TEGs Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe with respect to variation in engine load at different zones is shown in Figure 10.From the figure, it was found that Al 2 O 3 , Bi 2 Te 3 , PbTe 3 , and SiGe gave an increase in conversion efficiency of 2.6%, 3.8%, 2.7%, 11 International Journal of Energy Research and 2.8% at zone 2 when the operating load was 75%, in comparison to full load, respectively.Apart from this, it is also observed that Bi 2 Te 3 gives an overall increase in conversion efficiency in comparison to Al 2 O 3 , PbTe 3 , and SiGe throughout the load spectrum and operating conditions with respect to different zones.The conversion efficiency for Bi 2 Te 3 at zone 2 and 75% operating load is higher by about 2.6%, 3.8%, and 9.0% and 5.1%, 6.4%, and 11.5% in comparison to Al 2 O 3 , PbTe 3 , and SiGe at 75% and 100% of operating load, respectively.Bi 2 Te 3 gives the highest conversion efficiency compared to other TEGs irrespective of the working zones.At zone 2, Bi 2 Te 3 indicates maximum energy conversion efficiency of 7.8%, which is 2.63%, 4%, and 9.86% higher than that of Al 2 O 3 , PbTe 3 , and SiGe, respectively.This increase in energy conversion efficiency for Bi 2 Te 3 is due to the characteristics of the semiconductor, which has more free electrons to conduct current.It also has a unique atomic structure which allows its conductivity to be controlled by electric current, electromagnet, and other influential parameters because it carries negative electrons, a type of crystal in the category of doped semiconductor.It is also observed from the figure that SiGe gives the lowest conversion efficiency in comparison to Al 2 O 3 , Bi 2 Te 3 , and PbTe 3 .This may be due to the availability of less quantity of free electrons in the semiconductor, which may not be induced by the supplied temperature set in this experiment (135 °C).

Conclusion
Experimental investigations have found a novel perception of a completely passive and separate cogeneration structure built on the WHR technique and power generation system by using finned copper rods and TEGs.Experiments were conducted to find the rate of heat recovered, power generation, and conversion efficiency of the different TEGs at different zones of a heat exchanger by varying the engine load.During the experimental trials, the Bi 2 Te 3 generated a maximum of 7.18 W of electricity while the engine was operated at the 75% load condition.The heat recovery rate for Bi 2 Te 3 at zone 2 is higher by about 5.8%, 10.8%, and 17.9% and 4.6%, 7.2%, and 14.1% in comparison to Al 2 O 3 , PbTe 3 , and SiGe at 75% and 100% of operating load, respectively.Zone 2 Bi 2 Te 3 indicates maximum energy conversion efficiency of 7.8%, which is 2.63%, 4%, and 9.86% higher than that of Al 2 O 3 , PbTe 3 , and SiGe, respectively.This WHR and power generation method is an active technique for forthcoming improvement of energy harvesting systems, and this technique can be employed in industries, chimneys, and automotive vehicle exhaust ducts.Apart from this, the present technique can be used in solar power plants to generate electricity from hot solar panels instead of using phase-changing materials.

Figure 2 :
Figure 2: The layout of the experimental setup.

Figure 6 :
Figure 6: Waste heat recovery setup fitted with the combustion engine.

Table 1 :
Specifications of the combustion engine.

Table 2 :
Specifications of the copper block, duct, and glass wool.

Table 3 :
Specifications of infrared thermometer.