Energy Harvesting Society Poster Abstracts

Presenter:

Title of Poster:

Abstract:

2D heterolayered nanohybrids of uniformly-stacked transition metal dichalcogenide (TMD)-transition metal oxide (TMO) monolayers can be synthesized by an electrostatically-driven self-assembly of exfoliated nanosheets with charge-compensating cations. The obtained MoS2-MnO2 materials show well-ordered structure composed of restacked TMD and TMO monolayers. Moreover, the hybridization with MnO2 nanosheet is quite effective in enhancing the electrocatalyst activity of MoS2 for hydrogen evolution reaction (HER), which is attributable to the provision of more reactive MoS2 edge sites and the enhancement of charge transfer upon the incorporation of MnO2 nanosheet. The universal advantage of intercalative hybridization between TMD and TMO nanosheets is further confirmed by the remarkable enhancement of the electrode performance of other TMD-TMO hybrids.  This study highlights that the intervention of TMO nanosheet in restacked TMD nanosheets can provide an effective way to improve many functionalities of TMD materials.

Highly anisotropic 2D nanosheets of inorganic solids attract a great deal of research activity because of their unique merits as useful precursors for exploring novel efficient photocatalysts with energy and environmental applicabilities.  Several conductive 2D nanosheets such as reduced graphene oxide (rGO), RuO2, and MoS2 can be used as efficient hybridization matrices to improve the photocatalytic activity of semiconductor nanocrystal.  The hybridized conductive nanosheet can play a role of electron reservoir, photosensitizer, cocatalyst, and electron pathway.  To understand the relative efficacies of hybridization with these 2D NSs, several nanohybrids of CdS-RuO2, CdS-MoS2, and CdS-rGO are synthesized by electrostatic self-assembly of these exfoliated NSs with CdS quantum dots.  Among the present nanohybrids, the CdS-RuO2 nanohybrid exhibits the highest photocatalytic H2 production, underscoring the excellent role of the conductive RuO2 NS as a hybridization matrix in exploring visible light active photocatalyst.  According to photoluminescence spectroscopy, the hybridization with RuO2 NS leads to more prominent depression of electron-hole recombination than do those with the other NSs, indicating a stronger electronic coupling between CdS and RuO2 NS.  Of noteworthy is that the hybridization with RuO2 NS is more effective not only in increasing the visible light absorptivity of CdS but also in providing the visible light photocatalytic activity for wide band gap semiconductor TiO2 than those with MoS2 and rGO NSs, underscoring the better role of RuO2 NS as a visible light sensitizer.  The higher electrical conductivities of RuO2 and rGO NS than MoS2 NS strongly suggest the superior role of the formers as an electron transport pathway.  All the present experimental findings clearly demonstrate that the exfoliated RuO2 NS can act as the most effective hybridization matrix in exploring novel efficient visible light active photocatalysts, which is mainly attributable to its excellent function as electron reservoir and visible light sensitizer.

An Efficient Reconfigurable Power Management for a Multi-Beam Non-Resonant Mechanical Harvester

Miao Meng1, Shad Roundy2, Susan Trolier-McKinstry1, Mehdi Kiani1
1The Pennsylvania State University
2The University of Utah

An efficient, reconfigurable, switching power-management interface for a multi-beam (6 beams in this work) non-resonant mechanical harvester is designed and developed. The proposed reconfigurable power-management structure can significantly improve the harvested energy and efficiency, particularly when the mechanical harvester output has voltage-amplitude and frequency uncertainties, which is the case with the non-resonant mechanical harvester. A single diode in parallel with each beam is added to double the voltage across each beam by providing a DC voltage shift.  The power management seamlessly switches to voltage mode as an efficient half-wave voltage doubler for large input voltages (i.e. larger than the super-capacitor voltage, VCs). For small voltages lower than VCs (as low as 200 mV), the power management seamlessly reconfigures itself into current mode as an intermediate-inductor structure. In current mode, each beam is left open circuit until its voltage reaches its peak, at which an inductor is connected to the beam to transfer its energy to the inductor in a short time. Then, the inductor energy is transferred to the super-capacitor using a diode. The power management is reliable against any frequency variation of the incoming signal, because a peak detector is used to detect the peak voltage across each beam.
The proposed power management was designed and developed with commercial-off-the-shelf (COTS) components. The proof-of-concept prototype achieved an average efficiency of >60% for different voltage amplitudes. This power management can harvest energy from 6 beams simultaneously and is designed in a modular fashion such that it can easily be extended to even more beams. We have already developed an application-specific integrated circuit (ASIC) based on the proposed reconfigurable power-management structure which will be characterized and integrated with a 6-beam mechanical-harvester prototype in our future efforts.

In the last few decades, healthcare has shifted from a reactive stance to a more proactive one. This means that people are actively monitoring and maintaining their health rather than responding to a health crisis after the fact. Information on a person's state of health can be extracted from on-body sensor networks. Energy harvesting and wireless power transfer can be used to power these networks in order to motivate continuous use by minimizing user intervention (i.e. charging or changing batteries). System modeling can help us determine the required size of energy harvesters and storage elements (such as a supercapacitor) and bounds on the power and energy used by sensors. In this research, a framework is presented on how to use the data from a human behavioral database to perform realistic simulations of wearable self-powered systems. The Consolidated Human Activity Database (CHAD) is used to build typical day profiles as the input to the system. CHAD is a database containing information on human activities and locations throughout the day compiled from dozens of separate studies and contains over 54,000 days of data. The data can be filtered based on demographics, making it ideal for real-world usage analysis of wearable self-powered health systems. The activity and location information is used to estimate the power generated by different types of energy harvesters and simulate the model. This is not only a useful bench-marking and evaluation tool, but can also be for system design and optimization.

Poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT: PSS) is recently used in biosensing and bioenginnering application. As fabrication of self-healing conductive materials, many method has been used PEDOT: PSS with other chemicals which is used to acquire high self-healing property and electrical conductivity, but most process used not suitable materials in vivo environment. Here we report the fabrication of nonmetal free standing film using protein which has self-healing property with small amount of deionized water. The film was made mixing PEDOT: PSS and squid ring teeth (SRT) protein, which has unique self-healing properties and high elastic modulus in both dry and wet condition. 40 wt. % of SRT is achieved about 80% recovery of the conductivity after self-healing. We confirmed that SRT protein performed as a plasticizer and changed its microscale crystallinity, then 11 wt. % pristine sample showed high conductivity than control sample. In addition, mixing SRT protein with PEDOT: PSS showed high water stability than pristine sample. From this results, SRT protein showed new approach to electronics field and these result showed the possibility of the application for new bio-compost electrode.

This work describes the utilization of cold sintering to fabricate high output voltage, crack-free piezoelectric energy harvesters. The cold sintering process (CSP) was used to fabricate dense lead zirconate titanate (PZT) bimorph thick films (originally prepared by tape casting) on metal-based foils. 2 mol% Nb doped PZT 52/48 powder with an average diameter of ~ 100 nm is first mixed with 0.5 mol/L Pb(NO3)2 solution (with a weight ratio of 1:10 Pb(NO3)2 : PZT powder). The mixture is then dried and dispersed in the organic binder and plasticizer (QPAC®40, methyl-ethyl-ketone and Santicizer®160) to form a slurry. The slurry is tape cast and laminated onto both sides of LaNiO3/HfO2/Ni and HfO2/Ag foils and burned out at 275 °C to remove the organics. Afterwards, the samples are steam wetted by water at 80 °C for 1 hr and cold sintered under a uniaxial pressure of 500 MPa for 3 hours at 300 °C to densify the PZT layers. A post-annealing procedure in a box furnace at 700 °C (for Ni foil) and 900 °C (for Ag foil) for 3 hours in air was then applied to the cold sintered sample. For comparison, a tape cast sample without the steaming and cold sintering procedures were also annealed under the same condition.  The SEM images indicate cold sintered PZT thick films (layer thickness ~ 10 µm) have closed porosity after 700 °C or 900 °C annealing while presenting a significantly lower surface crack density than the samples annealed at 900 °C without cold sintering. The films on Ni foil show a saturated electric hysteresis loop with a coercive field ~ 90 kV/cm and a remanent polarization ~ 22.5 µC/cm2. However, possibly due to the residual lead oxides, a room temperature dielectric constant ~ 208 with a loss tangent ~ 3% at 20 Hz were observed. And the e31,f of the films on Ni foil is low (~-2.6 C/m2). On-going work, including electrical characterization of the cold sintered PZT thick films on Ag foil and the optimization of the film properties on Ni foil will be presented.

In the past few years, electromechanical strategies have been developed to reduce the energy cost of walking by using exoskeletons and suspended a backpack. Similar mechanisms have been developed to harvest energy from joint rotation and center of mass motion. However, there is not yet a common ground where these technologies meet. In other words, there is a lack of devices that reduce human metabolic cost while producing enough power to charge mobile gadgets. Here we show that forces on the body can be reduced when storing energy on a spring and then later releasing at an optimal time where it will decrease the burden of harvesting energy when walking. We call this system a decoupled harvester since a person will only feel the stiffness and small damping on a spring when compared to the high damping of a generator. A decoupled harvester was built to harvest energy from body center of motion and ankle dorsiflexion. Results show that the decoupled harvester produces about 65% of electrical energy while reducing the forces on the body by 52% when compared to a traditional rack and pinion harvester. Overall, this mechanism has the potential to be used in everyday apparel shoes, backpacks, etc. where one could seamlessly charge their own mobile devices when walking.

Deepa Singh, Ravi Anant Kishore, Prashant Kumar, Mohan Sanghadasa and Shashank Priya.

Energy harvesting possesses great potential for self-powered low power electronics. For instance, Energy harvesting from mechanical fluctuations is vastly investigated technique because it can be harnessed from fluctuations such as machines and human motion.  Among all mechanical to electrical energy conversion mechanisms, piezoelectric energy harvesters are the most sought out harvesters because of their simple integration and high electromechanical efficiency.  Here, we propose an array of spiral-shaped microelectromechanical (MEMS) systems energy harvester based on stainless steel substrates. Spiral cantilever design enhances effective length of piezoelectric film within the limited size while arraying of the spirals provides high power density. High quality poly(vinylidene fluoride-Triflouroethylene) P(VDF-TrFE) films exhibiting 9 µC/cm2 remnant polarization and 60 pm/V piezoelectric constant (d33) are drop casted over spiral substrate. Efficient micromachining of PVDF-TrFE has been done by CO2 laser ablation. A small spiral structure of 2.5X2.5 mm generates the output voltage of 4mV with a displacement of 0.5mm at 1 Hz frequency.  Peak voltage increases up to 25mV at 10 Hz.

Downsizing the metal particles in the non-noble metal based catalysts plays an important role in improving the catalytic performance. Here, we developed a novel strategy for preparing Ni-SAPO-11 hydroisomerization catalyst via an in-situ solid synthesis method. It only involved the grinding of nickel source with amorphous precursors of SAPO-11 before crystallization. Compared with the conventional catalysts preparation method, the in-situ solid synthesis method was much simpler, avoided the use of solvent and reduced the waste production. Through this method, both ultra-small NiO particles (i.e., diameter of 2-4 nm) strongly interacting with support and improved acidity were obtained simultaneously. Apart from NiO, nickel spinel and structural nickel also existed in Ni-SAPO-11, while metallic nickel originated from the reduction of NiO was the main active sites for (de)hydrogenation reaction. The obtained Ni-SAPO-11 catalyst exhibited comparable n-hexane conversion (71.2%) and iso-hexane yield (66.7%) to Pt/SAPO-11 catalyst in n-hexane hydroisomerization. The reaction over Ni-SAPO-11(s) followed the consecutive route to increase the iso-hexane selectivity, which was attributed to the large number of metal sites. Further improving metal-acid balance, the iso-hexane yield increased to 76.0% which was the highest value over zeolites based hydroisomerization catalysts to our knowledge. The outstanding catalytic properties of the obtained catalyst created a potential alternative to noble metal catalysts and thus a new approach of catalyst preparation has been developed.

The foot in the human body is the best part to make mechanical energy into electrical energy. Due to the gravity, a lot of force is transmitted to the food and walking is a big part of our daily life. We made a piezoelectric energy harvester (PEH) which consist of lead-zirconate-titanate (PZT) to convert energy generated from human to electrical energy and insulted under the insole of shoes. This harvester makes 3.7 V, 17 μA and 52 μW at a resistive matching point, 250 kΩ. The durability test was carried out using pushing test to see if the PEH located under the insole show sustaining power generation without destruction. The PEH has tested 1,000,000 cycles and it did not show the voltage decline. we confirmed that the PEH could be applied to shoes. The power generated from PEH was used to turn on a LED circuit that consists of a Mosfet device. When a user wearing the shoes walk, the LED of shoes turns on. This technology is able to save much energy than the previous system that uses a pressure sensor to check the input signal and a microcontroller unit (MCU)

Power electronics technology aims at efficient energy conversion. Solid-state Power-Semiconductor-Devices, PSD, are used as power switches and power rectifiers, and are at the heart of the transformative technology. The engineering science behind the physics of PSD performance is critical to optimize and control PSD operation. In addition to the engineering science is the issue of semiconductor materials appropriate for power electronics. While silicon is the workhorse of digital technology,  wide band-gap semiconductor materials are superior to silicon in power applications.. Wide band-gap materials presents issues of crystal growth, interfacial conditions, and processing .

Synthesis of La2NiO4+d platelets for energy conversion

Z. Zhao [a], R. Hinterding [a], C. Zhang [b], A. Feldhoff [a]

[a] Institute of Physical Chemistry and Electrochemistry, Leibniz University Hannover, Callinstraße 3A, D-30167 Hannover, Germany
[b] Institute for Mineralogy, Leibniz University Hannover, Callinstraße 3, D-30167 Hannover, Germany

Mixed ionic-electronic conducting (MIEC) Ruddlesden-Popper type oxides are promising candidate materials for energy conversion devices. Their excellent anisotropic electronic conductivity, oxygen-ion conductivity and small thermal expansion coefficient, give prospective to applications as cathodes in solid oxide fuel cells (SOFCs) and in thermoelectric generators (TEGs). However, it is challenging to control the crystal morphology and grain orientation, which restricts their predominance. In this work, molten-flux synthesis was applied to the Ruddlesden-Popper type oxide, La2NiO4+d, with the aim to produce it in platelet-like morphology with plate diameters of more than 10 microns. 
Melt of the hydroxide NaOH was used to dissolve precursors, which were either coarse mixtures of the oxides La2O3 and NiO from a classical solid state route or ultrafine mixtures of La2CO5 and NiO from an innovative sol-gel process [1]. To estimate optimum synthesis temperature for the sol-gel process, reaction sequence was monitored by in-situ X-ray diffraction (XRD). Purity and composition of products were investigated by ex-situ XRD and wave-length dispersive X-ray spectroscopy (WDXS). A maximum size of ca. 30 microns was observed by scanning electron microscope (SEM) for platelets at a thickness of ca. 2 microns (i.e. aspect ratio of 15). By varying several synthesis parameters, both size and aspect ratio of platelets could be varied over a wide range. Guidelines for obtaining La2NiO4+d platelets with size and aspect ratio being favorable for energy conversion are deduced from the physical chemistry behind the synthesis.

Reference
[1] A. Feldhoff, M. Arnold, J. Martynczuk, Th.M. Gesing, H. Wang, The sol-gel synthesis of perovskites by an EDTA/citrate complexing method involves nanoscale solid state reactions, Solid State Sciences 10 (2008) 689.

Udara Saparamadu1, Bed Poudel1, Carter Dettor1, Shashank Priya1
1 Department of Material Science and Engineering, Pennsylvania State University, University Park, PA 16802, USA

Half-Heusler thermoelectric materials has attracted much attention due to its thermal stability even at high temperatures, mechanical strength, high power factor (PF) and Figure-of-merit (ZT). To achieve good module performance, electrical and thermal contact resistances play critical roles besides materials ZT. In this work, we demonstrate device fabrication process to achieve low electrical contact resistance between electrode and thermoelectric material interface via high temperature compatible brazing technique. Poor contacts can cause Joule heating at the contact which can significantly degrade the power output from the design prediction. Ideally, the contact resistivity should be <1 µΩ cm2 where many material systems are far from achieving this value. Using a home-made scanning probe measurement system, the contact resistance was measured and specific resistivity at each interface was calculated. We were able to achieve a very low contact resistivity of ~1 µΩ cm2 for the n-leg and ~2 µΩ cm2 for the p-leg resulting in a power output close to theoretically predicted value. This device fabrication technique may lead to a potential breakthrough in high temperature power generation technology for various applications such as waste heat recovery vehicle engine exhaust and other industrial processes.

Healthcare has seen big advances in recent years as new technologies have emerged. Wearable sensor systems have enabled the possibility of continuous, long term personal health monitoring for the diagnosis, tracking and prevention of disease. In-time interventions that promote healthy behaviors or help manage pre-existing conditions have improved outcomes while reducing cost. Nevertheless, these devices have limitations that have prevented them from comprehensive acceptance in clinical settings. To ensure user adoption, challenges such as functionality, data management, form factor and battery life need to be addressed. A promising alternative for making these devices more reliable, invisible and easy to use is to implant them inside the body. However, small batteries create new risks given their possible chemical side effects. Furthermore, limited battery life necessitates frequent surgical procedures that are risky and expensive. Being able to harness energy from the body could solve these issues and enable continuous monitoring. Recent work on energy harvesting has focused on thermal and inertial energies. Kinetic energy in the fluids of the body is an unexplored energy source. In addition, if the harvester could be used as the sensor itself, the system integration can be improved. In this project, we explored energy harvesting for airflow monitoring in the trachea using piezoelectric cantilevers. In order to create a realistic setting, a 3D-printed trachea created from scans of a cadaver’s trachea was connected to a lung simulator. The piezoelectric cantilever was inserted through a small incision in the 3D printed trachea and placed normal to the flow. To monitor the output of the piezoelectric cantilever, the leads were connected to a rectifier circuit and capacitor which stored the energy for later use. To avoid loading the harvester while collecting data, a data acquisition (DAQ) system from National Instruments with an input impedance higher than 10 GΩ was used. Experiments show that breathing conditions for an adult at rest are able to generate 0.5 uJ of energy in less than 30 minutes. The results demonstrate the feasibility of a more complex self-powered sensor system that could measure detailed breath information. Furthermore, a prototype of a self-powered sensor system that transmits intermittent RF pulses in correlation with the airflow rate was developed. The pulses were detected by a software defined radio (SDR) connected to a computer which keeps the log of the RF transmission events.

The United States Geological Survey (USGS) has collected and shared water related data for 16,446 different locations throughout the United States. This represents roughly one data collection station for nearly every 200 miles of stream, which translates to an average of only 12 stations along the entire length of the Mississippi River. Furthermore, the existing stations are primarily located on large, economically significant rivers, leading to even lower data availability on significant but smaller waterways. To better characterize the nation’s waterways will require the deployment of a large number of sensor stations, many in hard to reach locations without power and network connectivity. The scale out the number of waterway monitoring sites also exacerbates the need of autonomy and robustness of sensor stations to remove the need for human intervention, especially in terms of energy management, as regular battery replacement is impractical or even infeasible. Energy harvesting for self-powered stations would therefore be critical to achieving denser sensing in waterways.

This work investigates energy harvesting and power management challenges and opportunities to enable self-powered sensor stations in waterways. The currently most common solution to self-powered sensor stations is the use of solar power and large batteries, keeping the system energy consumption well below the average harvested power and using most of the stored energy on batteries only during exceptional harvesting droughts. This design choice prioritizes the robustness of those stations to power outages in expense of the quantity and quality of data collected. For example, if the sensor station duty cycle is set to long idle periods, the energy consumption will go down but the sampling frequency will also be reduced. In our study, we analyze how the power consumption profile should be set to maximize data quantity and quality in face of harvested power fluctuations and battery storage limitations while avoiding power outages. 

Another limitation of the solar energy harvesting design is the drastic efficiency reduction of solar panels when deployed in places with low light incidence, such as dense forests or canyons. To increase the harvestable energy levels in such places, we investigated the potential use of small hydro turbines as a complementary energy source. Simulations were done based on commercially available hydro-turbine information and historical data from the USGS covering a variety of stream flow velocity conditions including strong seasonality variations and drought periods. Simulations results show that, for some rivers, the harvester would spend long periods of time without harvesting any power due to a minimum flow velocity required to turn voltage regulators on. The reduction of this minimum flow velocity threshold on hydro-turbines would enable sensor stations to harvest energy even during periods of low flow velocity. To evaluate the actual power harvesting capabilities of a commercial hydro turbine, we have conducted a field test deploying a small hydro-turbine and measuring the harvested power output. The preliminary results show power generation up to 0.2 watts in a small stream. Further experiments in laboratory testbench showed that the harvested energy is lower than claimed by the manufacturer, due to added USB port regulation electronics, suggesting that it requires hardware adaptation to be efficiently used on long term deployments.

CSP (cold sintering  process) is the process where inorganic powders are densified in the presence of 
a transient liquid phase at a phase fraction typically between 1 -10 vol%. 
During CSP, the liquid phase becomes the medium for mass transport. 
Recently, a broad number of ceramic compositions that can undergo a cold sintering were reported by Randall et al: 
binary, ternary, and quaternary compositions from oxide, bromide, chloride, fluoride, phosphate, and carbonate chemistries. 
In most cases, temperature and pressure typically span respectively the range between 25°C - 300°C and 1 atm - 500 MPa. 
Besides energy-efficiency, the interest in and utility of CSP is amplified by an expanded ability to integrate across conventional material classes, 
providing novel methods to assemble composites and join materials. In fact, it creates opportunities to combine materials that previously would chemically react, 
decompose, or volatilize, thus allowing the development of unique bulk materials, multilayers, and thick films. 
The constituent materials can encompass inorganics, nanomaterials, biomaterials, polymers, quantum dots, 2-D materials, liquid crystals, phosphors, Metal-OrganicFrameworks (MOFs), etc., 
and create new multifunctionality within new composite and device designs.
The opportunity to create new materials by CSP can potentially impact numerous technology areas in general and electroceramics in particular (i.e. solar cells, thermoelectrics, flexible electronics,).

The thermoelectric performance of polycrystalline Ca3Co4O9 (CCO) has been reported over the last two decades with emerging clean energy technologies. However, most of researches on CCO have not considered the anisotropy of this material, which hinders accurate thermoelectric characterization for practical applications. Here, highly textured CCO/Ag nanoinclusions composites have been prepared by spark plasma sintering (SPS) techniques and systematically investigated with regard to their microstructure and anisotropic thermoelectric properties for the first time. We measured thermoelectric properties for the CCO/x wt.% Ag (x=0, 1, 3, 5) composites along both vertical and horizontal direction with respect to the SPS pressure axis. Two step SPS method was also utilized to reorient small crystals for improving the electrical conductivity along ab planes. The results show that the addition of Ag nanoinclusions leads to the increase in electrical conductivity together with thermoelectric power factor values along both directions due to the key role of Ag in improving electrical connections between grains. The peak ZT value for the CCO/3 wt.% Ag composites measured along both vertical and horizontal directions is found to be 0.14 and 0.06 at 640°C, respectively, and the peak ZT value after two step SPS for the CCO/3 wt.% Ag composites measured along both vertical and horizontal directions is found to be 0.10 and 0.04 at 640°C indicating the preference of layered structure materials along the vertical direction to the press direction for practical thermoelectric applications.

Half-Heusler (hH) alloys are a promising thermoelectric material for energy harvesting applications in waste heat recovery system due to their exceptional thermal and mechanical stability and excellent electrical transport properties. However, a relatively high thermal conductivity (7-10 W/m⸱K) of the materials has been a primary challenge in producing a good thermoelectric performance for a practical use. This work has investigated a remarkably high thermoelectric performance in n-type hH alloy by adding metallic nanoparticles and incorporating them as nanoinclusions. The metallic nanoparticles play a role in not only an electron injection for higher electrical conductivity but also an energy filtering effect for relatively higher Seebeck coefficient in a given electrical concentration, resulting in outstanding improvement on power factor. In addition, we observed that multiple phases are formed in a diverse length scale from nano- to micron-meter, helping to scatter more extensive mean free path of heat-carrying phonons. Moreover, intrinsic properties of the metal nanoparticles, such as lower specific heat capacity and large acoustic impedance difference with hH matrix, can contribute to lower thermal conductivity. As the result, we achieved extraordinary thermoelectric performance of zTmax ~ 1.4 at 773K and zTavg ~ 1.2 in the medium temperature range (573-973K) from n-type (Hf0.6Zr0.4)NiSn0.99Sb0.01 compound with 5 wt% of the nanoinclusion.

Development of new architectures for thermoelectric generators (TEGs) is crucial for optimal and efficient operation over a wide temperature range, from 200°C to 700°C. In addition, a comprehensive mathematical model that considers heat transfer and thermal losses needs to be developed to model and analyze the geometrical effects on the characteristics of TEG. The model should also account for the thermal and electrical contact resistances. In this study, an analytical and numerical model for a segmented TEG was developed which was validated using single couple devices. The model of the TEG was used to investigate steady state behavior as a function of geometric parameters as well as operating conditions. The performances under different geometric parameters was analyzed where the thickness of layers and area ratio of TEG’s legs were varied. The analyses also quantify the effect of thermal losses such as conductive and radiative heat losses on performance of TEG using effective properties. Comparative studies revealed that the thermal losses of TEG predominantly increases heat transfer at the hot substrate and subsequently increases the power output, while the increase in heat input directly affects the efficiency of TEG. The results of parametric study showed that the efficiency increased as P-type and N-type leg area approaches a certain optimal ratio and each layer of the segmented and cascaded TEG operates within optimal temperature range. In addition, the effective properties and compatibility factor of the device accounted for the improvement of performance with optimized geometric architectures.

Entropy ist an essential quantity in thermoelectric energy conversion but traditionally not considered accordingly [1]. A thermoelectric generator uses entropy current to drive an electrical current. Depending on the sign of the Seebeck coefficient α of the thermoelectric material, which ist the entropy transferred per electrical charge transferred, both currents are parallel or anti-parallel. The electrical serial connection of n-type (α < 0) and p-type (α > 0) semiconducting legs with their contacts being alternatingly hot and cold (i.e. thermal parallel connection), allows constructing thermoelectric modules, which can make use of the entropy current running from “hot” to “cold” for energy harvesting [1]. The maximum energy harvested is determined by the difference of thermal energy (heat) entering the system at high temperature and leaving at lower temperature together with entropy. Using a direct entropic approach [2,3], a basic transport equation is derived, which contains fluxes of entropy and electric charge, gradients of classical thermodynamic potentials (i.e temperature and electrical potential or more precisely the electrochemical potential per charge) and a thermoelectric material tensor. Comprehensibility is much improved over the traditional Onsager–de Groot–Callen model, which describes thermoelectricity in the framework of the so-called “thermodynamics of irreversible processes” [4,5], and uses rather abstract generalized forces instead of gradients of thermodynamic potentials and a kinetic matrix instead of a thermoelectric material tensor. The material-specific tensor is composed only of the isothermal electric conductivity, the Seebeck coefficient and the entropy conductivity. Entropy is the key to an understanding of the thermoelectric phenomenon and the underlying energy conversion process. 

In this poster, we report the fabrication of two different classes of high energy density electrochemical capacitors that include a 3.5V ionic liquid based EDLC device and a 4.5V lithium ion capacitor. Both these capacitors were enabled by the synthesis of high purity bimodal porous carbon electrodes derived from pyrolysis and activation of polyfurfuryl alcohol. We show that the controlled porosity helps to improve both the rate capability as well as increase the specific capacitance of the electrode. Energy density as high as 160 Wh/Kg was achieved using lithium ion capacitors while the ionic liquid based EDLC showed an energy density of 74 Wh/Kg. The power densities of both the devices could exceed 10 KW/Kg.