The Joint Center for Energy Storage Research (JCESR) pursues high performance, low cost beyond lithium ion electricity storage that will transform transportation and the electricity grid. JCESR will leave three legacies:
bull; a library of fundamental knowledge of the materials and phenomena of energy storage at the atomic and molecular level
bull; two prototypes, one for the grid and one for transportation, that, when scaled to manufacturing will be able to deliver five times the energy density at one-fifth the cost
bull; a new paradigm for battery R&D that combines discovery science, battery design, research prototyping and manufacturing collaboration in a single highly interactive organization, accelerates the pace of discovery and innovation and significantly shortens the time from discovery to commercialization.
An introduction to JCESR&’s vision, mission and legacies will be followed by research highlights illustrating its advances in fundamental science and the promising pathways to transformational battery designs and prototypes.
This work was supported as part of the Joint Center for Energy Storage Research, an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences

One of the ongoing challenges in the field of Atomic Force Microscopy (AFM) has been quantitative measurement of the motion of the cantilever probe. This is certainly the case for electromechanical force microscopies such as piezoresponse force microscopy (PFM) and electrochemical strain microscopy (ESM) where accurate quantification of the piezo-electric sensitivities (d33, d15, etchellip;) has been a longstanding goal. Conventional PFM systems almost exclusively use a “beam bounce” optical beam deflection system (OBD)[1] where a laser is focused on the back of the cantilever and the angle of the reflected light is used to deduce the cantilever tip motion. However, non-desirable buckling and torsion of the cantilever may be misinterpreted as cantilever tip motion. This is a shortcoming of the OBD method which measures the angle of the cantilever, rather than the displacement of the tip.
Here we describe results from a metrological AFM, based on a commercial Cypher S AFM that combines highly sensitive electromechanical imaging and spectroscopies with a separate quantitative Laser Doppler Vibrometer (LDV) system that allows accurate (NIST traceable) measurements of the displacement and velocity of the cantilever tip. The separately motorized LDV spot position is independent of the conventional OBD laser spot. This opens up a great number of opportunities for independently measuring cantilever dynamics during typical AFM imaging.
Although this presentation is focused on electromechanical measurements, this instrument allows a host of quantitative measurements to be performed including measuring a variety of in-situ AFM cantilever oscillation modes. In addition, parameters such as the optical lever sensitivity and spring constant can readily be measured without the cantilever touching the surface. We will present metrological data on a variety of ferroelectric materials and electrochemical materials. We will also discuss various noise advantages, disadvantages and potential applications of this novel microscope.
[1] S. Alexander et al., J. Appl. Phys., 65, 1 (1989).
[2] OBD PPLN reference here http://arxiv.org/abs/1409.0133

In this study, a promising cathode material: layered lithium (Li)-rich oxide material Li₂MnO₃ - LiMO₂ (M = Ni, Co, Mn), with the energy density (280mAh/g) of approximately twice of that of the commercial cathode materials, is characterized by using various Scanning Probe Microscopy techniques, including Bimodal Dual AC Imaging, Band Excitation Electrochemical Strain Microscopy (BE-ESM), and Biased AFM. In these SPM techniques, BE-ESM allows the high frequency periodic bias to be applied on the sample surface of the electrochemically active materials. The bias will induce the local periodic oscillatory displacement caused by the Li-ions redistribution within the material, and the surface deformation caused by the Li-ion re-distribution is measured and defined as electrochemical strain. Biased AFM can apply a DC positive/negative point-bias in contact mode, and hence to study the influences of positive and negative biases on the surface deformation. This process can be used to study the effects of charging and discharging processes to the cathode material. Finally, bimodal Dual AC Imaging technique is used to study the composition changes due to the Li-ions redistribution under the electrical field. With the capability of the in-situ characterizing the topography, surface deformation (electrochemical strain), ionic movement as well as the corresponding changes of the composition and properties in layered Li-rich cathode films, this work will provide the fundamental understanding of the structure-property-functionality relationships at the nano- to micro-scales for the cathode materials used in the Li-ion rechargeable batteries.

In order to better understand stress gradient leading to mechanical failure of Lithium manganese (III, IV) oxide (LiMn₂O₄) particles, the ionic diffusion parameter needs to be measured. The common values range over several orders of magnitude in the literature, from 10^-17 to 10^-12 m²/s [1], and are obtained from non-local techniques. Electrochemical Strain Microscopy (ESM) time spectroscopy can measure the evolution of material's mechanical strain caused by lithium ions as a function of time. Typically, the ESM strain response is measured by applying an ac excitation voltage with frequency of several hundreds of kHz. Traditional impedance spectroscopy as well as models of ESM developed by Morozovska et al. (2014) [2] exhibited that lithium diffusion has an insignificant response in the high frequency regime for both cases . The former probes other non-local electronic effects in this range, such as capacitive effects. In the latter case, displacements in the order of 0.1 pm were simulated. At this length scale, continuous mechanics used in the model does not hold. Besides, it is much below the instrumentation detection limit. However an ESM signal originating from a mechanical strain induced by lithium ions can be experimentally detected. After applying positive (negative) DC pulses, a well known transition phenomenon occurs as the ESM signal increases (decreases). While the transition phenomenon is lead by the diffusivity of lithium in the crystal, the ESM strain itself is induced by a different mechanism. In the present research, commercial LiMn₂O₄ cathode was characterized by means of ESM. A thermodynamical Finite Element Model was developed in order to fit experimental data for different DC pulses and different lithium concentrations. It simulates the quasi-static strain and the diffusion of lithium ions induced by the DC pulse. The ESM (AC) signal is then derived from the results. The measured strain could fit analytical model previously developed by Morozovska et al. The unitless ESM signal can be fitted on experimental measurements. A better range for the diffusivity of LiMn₂O₄ can be extracted. Besides, the different behaviors observed experimentally can be explained by the particle geometry as well as a concentration dependent diffusion parameter. ESM and the model can be extended to virtually any lithium host.
References:
[1] Myounggu Park, Xiangchun Zhang, Myoungdo Chung, Gregory B. Less, Ann Marie Sastry, A review of conduction phenomena in Li-ion batteries, Journal of Power Sources 195 (2010) 7904-7929
[2] Alexander Tselev, Anna N Morozovska, Alexei Udod, Eugene A Eliseev, Sergei V Kalinin, Self-consistent modeling of electrochemical strain microscopy of solid electrolytes, Nanotechnology 25 (2014) 445701

Since Scanning Probe Microscopy (SPM) emerged as a powerful tool for probing materials from meso- to atomic level, a multitude of methods have been developed to study local mechanical, electromechanical, and electrochemical properties. However there remains a challenge in extracting high sensitivity, high veracity, and quantitative information from SPM measurements. In atomic force microscopy (AFM), operating on cantilever resonances maximizes sensitivity, but often at the expense of quantitativeness due to viscoelastic variations across material surfaces and non-linearities inherent to mechanics of the tip-surface junction. Band Excitation has emerged as a robust method enabling reliable on-resonance measurements by continuously measuring the cantilever transfer function. In Band Excitation SPM (BE-SPM) the system is excited through a digitally synthesized signal waveform with a finite spectral density in a band (or bands) centered on a resonance peak(s), as opposed to the single sine wave used in classical SPM. (1-3) The response is detected via a photodetector and is Fourier transformed. The ratio of the Fourier transform of the response to the excitation signals yields the transfer function of the system. The frequency dependence of the response can be analyzed to yield parameters including the amplitude, resonant frequency, and Q-factor thus yielding multiple decoupled channels of information on local linear and non-linear viscoelastic properties. Furthermore, BE combined with complex bias waveforms can be used as a sensitive probe to map properties including local ferroelectric and polarization properties (SS-PFM) (3-4), electrochemical and ionic activity (ESM)(5-6), and surface potential (contact-KPFM) with 10&’s nm resolution thus providing a multifaceted view of functionalities critical to understanding energy materials. Discussed in this presentation are the application of BE combined with complex probing and analysis methods to reveal the nanoscale behavior of ionic activity in energy materials.
1. Jesse S, Kalinin SV, Proksch R, Baddorf AP, Rodriguez BJ. 2007. The band excitation method in scanning probe microscopy for rapid mapping of energy dissipation on the nanoscale. Nanotechnology 18:435503
2. Jesse S, Kalinin SV. 2011. Band excitation in scanning probe microscopy: sines of change. J. Phys. D 44:464006
3. Kalinin SV, Rodriguez BJ, Jesse S, Proksch R. 2007. A biased view of the nanoworld: electromechanical imaging. R&D Mag. 49:34-36
4. Jesse S, et. al., Direct imaging of the spatial and energy distribution of nucleation centres in ferroelectric materials, Nature Materials, 7 (2008)
5. Kumar A, Ciucci F, Morozovska AN, Kalinin SV, Jesse S, Measuring oxygen reduction/evolution reactions on the nanoscale, Nature Chemistry, 3 (2011)
6. Arruda TM, Kumar A, Kalinin SV, Jesse S. Mapping Irreversible Electrochemical Processes on the Nanoscale: Ionic Phenomena in Li Ion Conductive Glass Ceramics, Nano Letters, 11 (2011)

Active particles are the crucial fundamental building blocks for lithium ion batteries (LIBs). They are blended with binders and conductive additives to compose the electrodes for LIBs. Having a comprehensive understanding of the responsive mechanisms of these particles is essential to overcome the current limitations of LIBs and to develop new LIBs with further improved performance. However, rarely the behaviors of an isolated single electrode particle, especially when they are in nanoscale size, have been experimentally characterized, such as the volume evolution, stiffness variation, compositional change and stress redistribution. The multifrequency SPM and conductive AFM have brought such characterizations to reality.
In this work, the isolated particles of lithium excess/rich layered cathode material Li(Li_0.2Mn_0.54Ni_0.13Co_0.13)O₂ were studied using various Scanning Probe Microscopy techniques, in comparison to the particles of its two composing phases Li₂MnO₃ and LiNi_1/3Co_1/3Mn_1/3O₂ (NCM). The Li-rich layer-structured systems show promising high capacity (>230 mAh g^-1), but suffered from dramatic capacity loss in the first charge/discharge cycle. The results presented in this work shed new light to the understandings of the material behaviors of the layered structure oxide cathode particles. Gradually, increasing DC biases will cause Li ions redistribution and eventually phase transformation. During this process, the particle morphology changes, accompanied with local stiffness variations as well as the stress redistribution. All of these observations are made at the nanoscale. The differences in electrochemical reaction mechanisms are also revealed from the current profiles between Li-rich and NCM particles. These results may complement the findings from conventional techniques for LIBs characterization (e.g. charge/discharge curves, TEM, XPS, etc.).

Thin film electrodes, especially those grown as oriented single crystals, can be utilized as model systems to correlate electrochemical processes with structural changes on an atomic scale. In this work we investigated microstructures of LiCoO₂ (LCO) films deposited epitaxially on SrTiO₃ (STO) substrates of different orientations by pulse laser deposition. The STO substrates were covered with a 50-nm-thick epitaxial conductive SrRuO3 (SRO) layer acting as a current collector. The films were electrochemically tested with a half-cell, with Li metal serving as counter and reference electrode. The LCO films were electrochemically cycled and hold at different potentials. Thin cross-sectional lamellas were prepared from the samples at different cycling stages by focused ion beam and investigated by analytical TEM/STEM, including electron energy-loss spectroscopy. The samples at initially charged state demonstrate that the LCO {104} facets are significantly more active in comparison to the {001} facets; near the (104) surfaces the changes in LCO crystal structure and Co oxidation state changes were observed. With increasing charging potential, up to 4.2 V vs. Li/Li⁺, the entire LCO film had altered to a new structure. The reaction primarily focus at the defects areas, where electrolyte gradually penetrates the LCO film. The observed microstructural changes are discussed and linked to the electrochemical reactions.

Charged solid electrodes immersed in electrolytes are found in many systems, such as in electrochemical cells, batteries, bio- and pH-sensors and actuators, where the electromechanical properties of electrodes play very important role in the performance and reliability of the systems. Although great progress has been achieved in the experimental investigation of the charge-deformation coupling behavior, the theoretical study of such behavior at the atomic/electronic level is still very much limited. The present work conducted Joint First-Principles/Continuum (JFPC) calculations to study the charge-induced deformation of monolayer MoS₂ and graphene in different electrolytes. In the JFPC calculations, the electrode of MoS₂ or graphene was analyzed by Density Functional Theory (DFT) based calculations and the electric field in the electrolyte was determined by solving the modified Poisson-Boltzmann equation, where the same electrostatic potential is involved in the DFT and the modified Poisson-Boltzmann equation. The JFPC calculations show that the monolayer thickness and the in-plane dimension of the monolayer MoS₂ decreases and increases, respectively, and almost linearly with negative or positive excess charges. With the calculated data, we calculate the charge volume expansion coefficient, which is 0.081/(absolute e) under negative charging and 0.003/(absolute e) under positive charging, indicating that the negative charge volume expansion coefficient is about 27 times as high as the positive charge volume expansion coefficient. For the graphene electrode, the C-C bond length monotonically changes when graphene is negatively charged. When graphene is positively charged, the C-C bond length varies nonlinearly with charge, decreasing first and then increasing, exhibiting the minimal value at excess charge of about 0.009 (absolute e)/C-atom. In addition, charged monolayer MoS₂ exhibits the in-plane charge expansion coefficient of 0.096/(absolute e), which is much higher than that of 0.033/(absolute e) in the graphene. This result implies that MoS₂ monolayer might have a much better prospect than graphene in the application of nano-actuators.

Lithium plating is an undesirable failure mechanism in lithium-ion battery cells that occurs when lithium metal forms on the surface of the negative electrode instead of undergoing the desired insertion reaction. Lithium plating causes capacity loss due to further side reactions between the newly formed lithium metal and electrolyte, and in the worst case scenarios causes catastrophic failure by creating internal battery shorts. In this presentation we provide experimental and computational results that demonstrate a linkage between lithium plating in lithium-ion cells and mechanical stress, with higher levels of stack stress leading to higher degrees of lithium plating. We attribute this link to local separator deformation, which creates non-uniform ion transport within the cell and leads to local hot spots of high utilization. This assertion is supported by results from experiments in which local lithium plating is shown to occur in coin cells with locally deformed separators. We then conduct two-dimensional electrochemical finite element simulations of coin cells containing mechanically deformed separators to demonstrate how non-uniform mechanical phenomena can result in localized regions of high-utilization, resulting in potentials below the plating potential of lithium. The simulation results agree well with experiment, and suggest possible mitigation strategies for designing safer battery cells.

There is currently a growing demand for low-cost, high-performance electrochemical energy storage solutions to consumer electronics, vehicle electrification and stationary power management. The successful development and deployment of such solutions necessitate a fundamental understanding of the mechanical properties of electrochemical materials, as well as the intricate coupling between the electro-chemo-mechanical processes in these materials. In this talk, I will present an integrated experimental and computational investigation of deformation and fracture in silicon and germanium electrodes for their use in advanced lithium-ion batteries. Novel in-situ experiments using high-resolution transmission electron microscopy (HRTEM) and nanoindentation were designed to study the two-phase lithiation, stress generation and mechanical failure in lithiated electrodes. Concurrently, continuum and atomistic models were developed to reveal the mechanistic origins of these phenomena. The fundamental findings of our work provide important insights for the multiphysics modeling and design of next-generation lithium-ion batteries.

High capacity lithium ion battery electrodes may experience large volume changes caused by lithiation and de-lithiation. Electrode failure, in the form of fracture or decrepitation, can occur as a result of repeated volume changes. In this presentation, we will discuss our recent work on understanding the evolution of concentration, stress, and strain energy in electrodes of spherical, cylindrical, and thin film geometries. We show that a dimensionless parameter, the electrochemical Biot number, can be used to characterize stress and strain energy evolution, including maximum stress and strain energy, in electrodes. We will also discuss several strategies, such as artificial solid-electrolyte interphases and self-healing electrodes based on reversible liquid-solid phase transformation, to enhance the performance and durability of lithium ion battery batteries.

An isothermal, physics-based, dual lithium-ion insertion cell sandwich model was used for simulating the galvanostatic charge performance of a graphite (Li_xC₆)/ liquid electrolyte/ Li_y(Ni_aCo_bMn_c)O₂ cell at room temperature at various current densities [i].The cell capacity vs. voltage data from simulations were validated with experimental results. The cell was charged only to 67% and 57% of the designed capacity at 2C and 3C constant-current rates (without a subsequent constant-voltage step) respectively due to corresponding overpotential losses of 279 and 414 mV. Lithium plating at the negative electrode was shown to be thermodynamically feasible during galvanostatic charging at 2C rate and above [i]. If lithium is plated unevenly and irreversibly, it will grow as a dendrite during repeated cycling. This can lead to both capacity loss and safety concerns. Sluggish charge transfer kinetics at low temperatures further accelerates lithium plating. Therefore recent work to understand this phenomenon in detail to overcome the hurdles to fast charging of advance electric vehicle batteries will be presented at the meeting.
Acknowledgement: Andy Drews and Ted Miller are acknowledged for their support.
References:
[i]. R. Chandrasekaran, J. Power Sources, 271, 2014, 622-632.

The battery separator is a porous polymer membrane used to create a physical barrier between lithium-ion battery electrodes. Compressive deformation of this membrane during lithium-ion battery operation has been linked in previous work to increased impedance, accelerated chemical degradation, and in the worst cases catastrophic failure through internal shorting of the electrodes. Thus knowledge of the compressive mechanical properties are important for understanding and predicting battery degradation due to separator deformation. In this presentation we present results from mechanical testing of battery separators under compression, which generally differ from tensile measurements due to the anisotropic processing of the separator during manufacturing. The compressive measurements show the existence of multiple mechanical phenomena that must be considered during battery operation: a reduction in mechanical properties due to swelling of the polymer material by electrolyte solvent, and a rate-dependence of the mechanical properties due to viscoelasticity and poroelastic phenomena arising from fluid ejection from the pores during compression. We present a model that incorporates these different physical phenomena, which we use to shed insight on the effects of parameters such as strain rate, fluid viscosity, and polymer swelling. This understanding of mechanical properties provides insight into how separator deformation and pore closure occurs in lithium-ion cells, resulting in accelerated degradation in both power capability and capacity.

Alloy-based anode materials such as SnO2 are promising candidates for increasing the capacity and energy density of lithium ion batteries, which require higher volumetric and gravimetric energy densities to be successfully deployed into many emerging applications, including transportation. While SnO2 based anodes offer higher capacity and energy density in new lithium-ion batteries systems (up to 780 mAh/g), these anodes suffer from dramatic volume changes during both the first cycle and subsequent (de)lithiations (up to 100%), which limit their lifetime. Nanostructured anodes represent one possible route for reversibly accommodating large volume changes, but their design is hindered by a lack of experimentally measured, quantitative strain data for SnO2-based 3D structured anodes, especially after the first cycle. Additionally, the unclear lithiation mechanism in nanostructured SnO2 anodes after the first cycle has greatly hindered in operando strain measurement to date. Using synchrotron-based X-ray diffraction techniques at the Advanced Photon Source, lattice strains in SnO2-coated interconnected commercial Cu foams were measured in operando in order to deduce the mismatch stresses and strains evolution during charging and discharging in the deposited anode material thin film and the Cu foam. Since the SnO2/ Sn and Li2O anode matrix forms a strong bond with the Cu foam, the elastic strains measured in the Cu are similar to those present in the anode material, and thus the stress and strain state in the SnO2 anode can be indirectly measured. These inverse opal anodes were cycled with different (dis)charge rates in order to explore the sensitivity of strain evolution to the cycling rate. Strains measured in the copper foam were directly correlated with the electrochemical cycling of the anode. These strains are discussed in terms of elasto-plastic deformation of the foam, cracking of the active material matrix and other potential stress relief mechanisms.

The energy densities of Li-ion batteries are governed by the amount of lithium ions that can be reversibly intercalated into and extracted from electrodes, which is governed by thermodynamics, while the rates of charging and discharging are determined by kinetics of electrochemistry and transport of lithium ions into and out of electrodes. Clearly, intercalation and extraction of lithium ions in electrodes is critical for high performance Li-ion batteries, which are rather complicated involving coupled electrochemistry and mechanics.
In this study, thermodynamics of intercalating ions in the electrode is governed by coupled mechanics and electrochemistry. In the chemo-mechanical potential of ions, the effect of entropy, enthalpy and mechanical potential of intercalation are considered. The mechanical potential is related to the hydrostatic stress in the electrode, generated by the eigenstrain due to intercalation, which requires mechanical equilibrium equation to be solved. Applying the Fick&’s law and mass balance relation, the ion transfer equation can be found. Arrhenius kinetics is used to model the electrode-electrolyte surface reaction rate and chemo-mechanical potential difference between the electrode surface and electrolyte are used as the activation energy of the reaction. The described coupled physics are implemented in COMSOL multiphysics and the results of simulation for different electrode conditions are presented. The results predict that electrode experiencing less hydrostatic stress at the surface during intercalation shows higher capacity.

Meanwhile, it is generally accepted that mechanical properties and processes influence performance and degradation of electrochemical storage systems such as lithium ion batteries (LIB). In this talk, we consider the interaction of mechanics and diffusion in individual electrode particles as well as the effect of collective mechanical processes in granular electrodes on effective properties.
Most host materials for electrochemical energy storage show phase changes upon intercalation. By phase-field modeling we study the effect of phase changes on the generation of mechanical stresses in electrode particles. Coupling the Cahn-Hilliard-equation to small and finite strain mechanics shows the tremendous stresses due to the strain mismatch caused by phase segregation. Imposing symmetries of the geometry on the solution is shown to possibly exclude states of minimum energy, which, for example, questions the so-called core shell scenario. A phase-field approach to fracture is also used to simulate mechanical particle damage during charging and discharging cycles.
Mechanical effects are also present in the interaction of particles in a granular electrode. First, by calendering, electrodes of lithium ion batteries are densified and, second, the expansion of active material particles leads to changes of interparticle forces. Here, we focus on the effect of these processes on the effective electronic conductivity of cathodes in LIB, where carbon black is added to connect active material particles to the current collector. Discrete Element Modeling (DEM) allows accounting for processes typical of granular materials such as changes in the configuration of electrodes particles. The effective conductivity is calculated by a resistor network approach. In particular, we discuss aspects such as size ratio and volume fraction of percolation probability and conductivity.
Coworkers on these activities are:
Huttin, Magalie (KIT); Ott, Julia (KIT); Walk, Ann-Christin (KIT); Santoki, Jay (IITM); Klinsmann, Markus (Bosch); Rosato, Daniele (Bosch); McMeeking, Robert M. (UCSB); Gan, Yixiang (USYD)

During battery operation, Li flows into and out of electrode particles, causing microstructural changes and deformation-induced degradation. A variety of models have been proposed to interpret these mechanical and microstructural changes, but they lack of proper input materials&’ properties and direct experimental supports. To address this challenge, we developed “atomically informed” battery mechanics models by integrating density functional theory (DFT) predicted mechanical properties into continuum models to predict diffusion induced deformation and stress evolution and make direct comparisons with experiments. Based on DFT calculations, we found many electrode materials change their elastic properties, fracture energies, interfacial properties, and bonding natures upon lithiation.
This talk will focus on computation-aided design of mechanically robust Si composite electrodes, in which the interfacial properties become very critical. Various interfaces, such as Si/Cu, Si/C, and Si/coating interface properties are predicted from density functional theory (DFT) and ab-initio molecular dynamics (AIMD) calculations. For example, it was found that Li tends to segregate at the Si/Cu interface, but not at the Si/amorphous-C interface; and the adhesion at the Si/sp2-bonded carbon interface is intrinsically weak, but can be improved by functioning groups. These predicted properties serve as inputs for continuum level of modeling in order to compare directly with experiments and give morphology design guidelines.

LiFePO₄ (LFP) is a most promising cathode material for lithium ion batteries (LIBs) due to its high theoretical capacity (170 mAh g-1), flat voltage plateau, long cycle life, abundant material supply, and excellent safety. However, it suffers from low electronic conductivity, which results in sacrificed capacity and low rate performance of the cathode. Conventionally, introducing conducting additives would help to improve the electronic conductivity of the LFP-based cathode. On the other hand, binders are required to construct a workable cell. Finding efficient additives that enhances the electronic conductivity of LFP, while at the same time minimizing the content of both additives and binders, become a tempting goal as it would lead to high performance cathode with high packing density. In the present work, we demonstrate a binder free composite electrode of LiFePO₄/reduced graphene oxide (rGO) with ultrahigh LiFePO₄ mass ratio (91.5 wt% of LiFePO₄) using electrophoresis. No additional additives (other than rGO) is introduced. The quasi-spherical LiFePO₄ particles are uniformly connected to and/or wrapped by three-dimensional networks of rGO nanosheets, with intimate contact formed between the two. Enhanced capacity is achieved in the electrophoretic composite cathode, when compared to either the conventional one or composite cathode formed by mechanically mixing LiFePO₄ and rGO. The present methodology is simple and does not disturb the active material growth process. It can be generally applied to a variety of active material systems for both cathode and anode applications in Li-ion batteries. This work is supported by General Research Funding of the National Natural Science Foundation of China/ Research Grants Council Joint Research Scheme under project No. N_CUHK448/13.

The measurement of state of health (SOH) and state of charge (SOC) is critical for safe and efficient operation of battery systems. However, the real-time determination of SOH and SOC is a major technological challenge, as both of these parameters require a controlled discharge to measure correctly. Conventionally researched approaches to estimating these parameters rely on complex and computationally intense physical models to relate measurable parameters (e.g. voltage and current) to the desired quantities SOH and SOC. In this presentation we propose the use of mechanical measurements (i.e. stress and strain) of battery cells to determine the SOH and SOC of lithium-ion batteries. This method of SOH/SOC determination is simple and straightforward, a distinct advantage compared to model-based estimation methods. We present data from long-term aging studies of commercial lithium-ion pouch cells, which show a linear relationship between stack stress and SOH. This linear relationship is found to hold over a range of cycling conditions, including cycling at elevated temperature as well as calendar aging. We investigate the mechanisms for the stress-SOH relationship through nondestructive analysis of cycling data and show that a linear stress-SOH relationship is expected for cells which age through a surface film growth mechanism . We also discuss the relationship between stack stress and SOC, which follows from the well-known expansion of lithium-ion electrode materials with changing lithium concentration. Using mechanical measurements to determine SOC offers the advantage of higher sensitivity to SOC than voltage measurements, especially in two-phase material systems with flat discharge curves.

Surface passivation has become a routine strategy of design to mitigate the chemomechanical degradation of high-capacity electrodes by regulating the electrochemical process of lithiation and managing the associated deformation dynamics. Oxides are the prevalent materials used for surface coating. Lithiation of SiO₂ leads to drastic changes in its electro-chemo-mechanical properties, from an electronic insulator and a brittle material in its pure form to a conductor and a material sustainable of large deformation in the lithiated form. We synthesized SiO₂ coated SiC nanowires that allow us to focus on the lithiation behavior of the sub-10 nm SiO₂ thin coating. We systematically investigate the structural evolution, the electronic conduction and ionic transport properties, and the deformation pattern of lithiated SiO₂ through coordinated in-situ transmission electron microcopy experiments, first-principles computation, and continuum theories. We observe the stress-mediated reaction that induces inhomogeneous growth of SiO₂. The results provide fundamental perspectives on the chemomechanical behaviors of oxides used in the surface coating of Li-ion technologies.

The prevalence of flexible electronics, evident by the ubiquitous touch-screens, roll-up displays, wearable sensors and implantable medical devices, has brought attention to the development of high performance and safer flexible energy storage devices. A major milestone in the development of the all-solid-state flexible LIB is the replacement of the traditional organic liquid electrolyte with high performance solid electrolyte, namely, ceramic/glass or polymer.
In this study, we have developed a high performance flexible Li ion battery based on a solid nanocomposite polymer electrolyte (1% graphene oxide particles in polyethylene oxide host) exhibiting a capacity of 0.13 mAh cm^-2 and excellent cycling stability over 100 charge/discharge cycles. Improvement in ion conductivity is observed with the addition of only 1 wt.% graphene oxide filler to the polymer electrolyte demonstrated by complex impedance spectra. The flexible LIB displays a relatively high operating voltage of 4.9 V compared to that of conventional batteries based on liquid electrolyte. Furthermore, the energy density of the fabricated flexible LIB is measured to be 4.8 mWh cm^-3 at room temperature which is within the range of reported energy densities of thin film LIBs (1-10 mWh cm^-3). Overall, the plastic laminated flexible LIB exhibits robust mechanical flexibility (over 100 bending cycles) and good electrochemical performance in both flat and bent positions.

We demonstrate experimental and computational results on lithium dendrite reduction on the post-charge lithium electrode surface. We demonstrate that increasing the cell temperature up to 60°C reduces the dendrites length as low as 30%. The quantified reduction has been demonstrated in optically accessible Li-metal cells and the verification has been done by means of reactive force field (reaxFF) molecular dynamics simulations. Detailed QM computations reveal that higher post-charge temperatures provides energy barrier, for creating vacancies in the dendrite bulk structure as well as surface diffusion as major mechanisms for annealing microstructures.

Layered electrodes, which are composed of active layers and current collectors, are
typical structures in lithium ion batteries. During charging and discharging, the lithiation
induced deformation of active layer is restricted by the current collector, which leads to
induction of stresses in the electrodes. To achieve a better understanding of structural design
of electrodes, theoretical studies were performed to investigate the diffusion of lithium ions
and diffusion induced stresses (DIS) during the charging and discharging of layered
electrodes, such as plate and cylindrical electrodes. The results obtained are described below.
Firstly, based on the evolution of diffusion induced stresses obtained from the elastic
study of plate and cylindrical layered electrodes, a current collector is best designed as soft
and flexible as possible to decrease the stress in the active material. In addition, the layers of a
cylindrical electrode should be as thin as possible to reduce the stress inhomogeneity induced
by curvature. Moreover, an optimized charge operation of galvanostatic first followed by
potentiostatic is also proposed based on the results obtained. Secondly, it has been found that
by minimizing the elastic energy, the induced stresses would promote the Li-ion diffusion
from the region of high compressive stresses to that of low stresses. Finally, it has been found
to be beneficial to allow a thin current collector to yield plastically in charge/discharge cycles.
The benefits include both increase of capacity as more space can be provided to contain active
materials and enhancement of electrochemical stability due to significant reduction of stresses
in the active layer. Hence, some designing insights are provided on both structure and charge
operations.

Lithium ion batteries are the most widely used energy storage device in both portable electronics and electric vehicles due to its high energy density. However, there are still remaining issues unsolved. To further improve the Li-ion battery performance, it&’s important to understand the lithium ion intercalation/de-intercalation process upon cycling. The development of a new scanning probe microscopy technique (SPM), electrochemical strain microscopy (ESM), in which a combination of AC bias and DC bias is applied to the sample surface, allows us to evaluate mobile ionic diffusivity and electrochemical reactivity on the nanoscale.
In this work, we will first discuss about using a series of strain-based SPM techniques to distinguish different microscopic mechanisms including electromechanical coupling, spontaneous polarization, induced dipole moment and electrostatics. Then we will demonstrate the localized probing of lithium ion transport property in a lithiated carbon anode using ESM. Pristine carbon anode is also studied as comparison. Also, we&’ll investigate the lithium ion transport through the solid electrolyte interface (SEI) layer, which is formed on the carbon anode surface upon cycling.

Lithium sulfur batteries are attractive due to their high theoretical energy density and reasonable kinetics. Despite the success of trapping soluble polysulfides in a matrix with high surface area, spatial control of solid state sulfur and lithium sulfide species deposition as a critical aspect has not been demonstrated. Herein, we show a clear visual evidence that these solid species deposit preferentially onto tin-doped indium oxide instead of carbon during electrochemical charge/discharge of soluble polysulfides. To incorporate this concept of spatial control into more practical battery electrodes, we further prepared carbon nanofibers with tin-doped indium oxide nanoparticles decorating the surface as hybrid three-dimensional electrodes to maximize the number of deposition sites. With 12.5 µl of 5M Li₂S₈ as the catholyte and a rate of C/5, we can reach the theoretical limit of Li₂S₈ capacity ~1470 mAh g^-1 (sulfur weight) under the loading of hybrid electrode only at 4.3 mg cm^-2.

For the powerful rechargeable battery system, lithium sulfur(Li/S) batteries can deliver and exceptionally high theoretical specific capacity of 1672mAhg-1 and an energy density of 2500 Whkg-1 with the low cost environment friendly sulfur as the cathode material. Although the potential use of sulfur as a cathode material has long been discovered, several severe drawbacks have hindered the realization of Li/S batteries. One limitation is the insulating nature of sulfur with a very low conductivity of 5*10-30 Scm-1, which results in low utilization of sulfur. Another well -known problem is associated with the easy dissolution of polysulfides, the intermediate products formed during the electrochemical reaction, in organic electrolytes. The dissolved polysulfieds “shuttle” between the electrodes, leading to the low coulombic efficiency and deposition of a highly resistive layer on the surface of electrodes. These detrimental issues result in unsatisfactory electrochemical performance with rapid fading of capacity. Also, the lithium sulfur batteries have suffered from the volume expansion during the charge/discharge the cyclic reactions. In this research, we adopted metal organic frameworks (MOFs) for the template for blocking the volume expansion of sulfur. The MOFs have a lot of large surface area because of micro-pores in the frameworks which are guide for alleviate the volume expansion of sulfur. However, the MOFs don&’t have high conductivity as a non-treated condition. For giving the conductivity in the MOFS, we carbonizated MOFs as a various temperature. In addition we comparing the MOF-867 and UiO-67, which have a same pore size and structure but a nitrogen was doped in the MOF-867&’s linker. The lone pairs of electron in the nitrogen can affect to the battery cyclic ability and capacity performance. The sulfur infiltrated in the micro-pores can diffuse out from the MOFs because there are no capping layer for restraining the polysulfide dissolution. For completely blocking the polysulfide, we used a conductive polymer, PEDOT:PSS which don&’t have chemical reaction in the charge/discharge in the Li/S batteries. In conclusion, we studied that the MOFs can be a template for blocking the volume expansion of sulfur and how to affect the capacity of batteries as doping of the nitrogen. Also, the conductive polymer was adopted to the blocking layer of MOF/S composites.

With the expanding need for large electrical energy storage systems in connection with renewable energy sources, flow batteries, have been enormously considered due to their high flexibility in upgrade and low cost associated with scale-up. Of all the flow batteries, vanadium redox flow battery (VRFB) use of same element in both half-cell solutions that overcomes the inherent issue of cross contamination by diffusion of different ions across the membrane. Together with the absence of any toxic emissions, the vanadium redox flow battery has demonstrated its uniqueness in terms of safety and long life cycle. Typical charge-discharge reactions of a VRFB involve two vanadium redox couples, V(II)/V(III) and V(IV)/V(V), in the negative and positive half-cells, respectively. In a fashion similar to most batteries, electrons are transferred between the two electrodes through the external circuit during the charge and discharge processes. In a VRFB, the ion exchange membrane is a key component as an ionic conductor and separator: it not only provides an ionic conduction pathway between the two electrolytes but also prevents mixing of the negative and positive electrolytes. The crossover of ions through the membrane, with the diffusion of vanadium ions from one half-cell to the other due to the concentration gradients between the two electrolytes, will result in self-discharge and thus the loss of the chemical energy.
In this study, the hybrid membrane of inorganic materials with perfluorinated organic membrane were fabricated and characterized in terms of ionic conductivity and permeability. The ionic conductivity was measured with four point probe method and the permeability was measured with UV spectroscopy. The hybrid membrane exhibited similar ionic conductivity with perfluorinated organic membrane and 30% lower permeability than perfluorinated organic membrane. With performances of VRFBs, it is proposed that the energy efficiencies of VRFBs are compromised with membrane conductivity and permeability. The columbic efficiencies are more contributed by membrane permeability. The permeability properties are more dominant in low current density and the ionic conductivity is more effective in high current ranges. In order to obtain higher performances of VRFBs, the membrane design for selectivity should be considered according to the operation conditions.

In this study, we fabricated three-dimensional NiCrAl metal foam for current collector of positive electrode and demonstrated a high electrochemical performance of Li-ion batteries. The effect of the electrode thickness and the pore size were investigated. Small thickness and small pore size electrode showed a higher charge/discharge capacity, less capacity reduction, and lower charge transfer resistance than the electrode with a large thickness and large pore size. Moreover, the anodic peak in cyclic voltammetry curve showed a high current and low voltage because of a quick redox reaction. Furthermore, it indicated that many triple points (electrolyte + active material + current collector) contributed to charge-transfer reaction and short Li-ion diffusion length.

In an effort to minimize cost and environmental impact in the production of LiFePO₄ (LFP) cathode material for lithium-ion batteries, an ionothermal synthesis has been studied. Through this method, LFP synthesis occurs with an ionic liquid medium serving as a solvent and—due to characteristics of ionic liquid functional groups and anion/cation interaction—a structural directing agent in the formation of LFP material. In this approach, 1-ethyl-3-methylimidazolium trifluoromethanesulfonate ([EMI][TfS]) had been used as a liquid-phase ionic medium. Our methods allow for a low-temperature reaction capable of producing large quantities of material in open environment conditions. Synthesized material had been characterized by scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and x-ray diffraction (XRD). Here is reported an analysis on the morphology, crystallinity, and electrochemical performance of synthesized LFP particles.

Exfoliated graphene oxide (GO) and polysiloxane were blended and pyrolyzed to synthesize freestanding SiOC-graphene composite papers (~10 µm thick). The structural and chemical characterization of the composite prepared with varying polymer concentrations were carried out using electron microscopy, XRD, and FT-infrared spectroscopy. High resolution microscopy images shows layer by layer stacking of GO sheets and an increase in interlayer spacing was observed by X-ray analysis. FTIR peaks at 3400 cm-1 (O-H), 1720 cm-1 (C=O), 1600 cm-1 (graphene), 3056 cm-1 (Si-CH=CH2) and 1034 cm-1 (Si-O-Si) confirmed the successful functionalization of SiOC with GO. Thermo-gravimetric analysis showed enhanced thermodynamic stability of the composite paper up to at least 700 °C in flowing air. The SiOC/Graphene composite paper anodes showed stable electrochemical capacity of approx. 500 mAh/g which was twice that of free standing graphene anodes. The average columbic efficiency (second cycle onwards) was observed to be approx. 97%.

We have demonstrate the synthesis of SiOC-CNT/G composite layered paper with rGO as the structural and electrically conductive support matrix. Microscopic and chemical characterizations were performed to study the morphology and composition of the free-standing paper. Further these papers when utilized as anodes to study its electrochemical performance at varying current densities (C-rate). The SiOC-CNT/G composite anodes showed stable charge capacity of 600 mAh/g at 100 mA/g. Randomly arranged papers had higher specific capacity at low current densities but when the current density was increased to 2400 mA/g the layered paper had higher specific capacity. Post cycling characterization reveled the integrity of free-standing paper was structurally intact even after cycling at high current density.

In the anode-supported solid oxide fuel cells (SOFCs), Ni-3YSZ cermet provides structural support and necessary porosity for anode operation at significantly reduced material cost as compared to Ni-8YSZ cermet. However, on long-term operation of the SOFC, Nickel-3mol % Yttria Stabilized Zirconia (Ni-3YSZ) anode support suffers degradation in strength and structural properties at severe anode operating conditions. In this study, the effect of humidity and temperature on the mechanical strength, micro-structure and phase changes have been studied. The strength and structural properties of Ni-3YSZ and Ni-8YSZ were evaluated after 100, 500 and 1000 hrs of degradation at 800, 850 and 900 ^oC and 20% - 30 % humidity in hydrogen environment. The stabilizing effect of additives (Al₂O₃ and TiO₂) in Ni-3YSZ cermet was also studied. It was found that at 900 ^oC and 30% humidity, the strength of Ni-3YSZ decreased up to 70% after 1000 hrs operation while Ni-8YSZ remained stable with only 20% decrease in strength. Addition of 3% Al₂O₃ in Ni-3YSZ resulted in improved strength on long-term degradation while TiO₂ addition greatly reduced the strength of Ni-3YSZ cermet. These results will be discussed in this talk.

A microfluidic fuel cell is a particular kind of micro fuel cells that operates without a proton exchange membrane by transporting the aqueous fuel, oxidant and/or electrolyte streams in a microchannel through different inlets under laminar flow condition. Through the contact of the fuel and oxidant with separate catalyst-covered electrodes, both the electron-losing reaction at the anode and the electron-gaining reaction at the cathode can take place independently on the electrode surface. The half-cell reactions for both anode and cathode, incorporating the diffusive ion transportation between the electrodes assisted by the supporting electrolyte in the reactants, accomplishes the overall electrochemical reaction in such microfluidic fuel cells, thus generating electricity. This study intends to fabricate an air-breathing direct formic acid microfluidic fuel cell employing soft-lithography process and laser machining to investigate the anode catalyst ratio on the cell performance at various operating conditions. Both cell performance and flow visualization will be performed with various formic acid concentrations at various volumetric flow rates to characterize the relationship between cell output and bubble behavior. An air-breathing direct formic acid microfluidic fuel cell, which had a self-made anode electrode of 10 mg/cm² Pd loading and 6 mg/cm² Nafion content, was fabricated and tested. The microfluidic fuel cell was achieved by bonding a PDMS microchannel that was fabricated by soft-lithography process and a PMMA sheet that was machined by CO₂ laser for obtaining 50 through holes of 0.5 mm in diameter. Formic acid of 0.3 M, 0.5 M and 1.0 M mixed with 0.5-M H₂SO₄ as supporting electrolyte was used as the fuel. Instead of acting with an aqueous oxidant, the oxidant in this study was air supplied by breathing through a porous gas diffusion electrode. The air-breathing microfluidic fuel cell having a 1.5-mm-wide and 0.5-mm-deep microchannel was tested at volumetric flow rates ranging from 0.1 mL/min to 0.7 mL/min. The self-prepared catalyst ink was brushed onto a carbon paper (Toray TGP-H-090) and then baked on a hot plate to obtain an anode electrode with catalyst loading and Nafion content of approximately 10 mg/cm² and 6 mg/cm², respectively. The anode catalyst ratio, which is defined as the weight ratio between Pd catalyst and carbon carrier Vulcan XC-72, was 75%, 85% and 95%. the As regards the cathode, a commercially available gas diffusion electrode containing catalyst loading of 2 mg/cm² Pt black supported by Vulcan XC-72 was used. The results showed that the cell having a 75% Pd/C anode had the worst cell output because numerous deep cracks on the anode surface caused the CO₂ bubbles to stick on the electrode. Contrarily, the cell having an 85% Pd/C anode had the best cell output because the electrode had an optimal parameter which caused the bubbles to periodically generate and detach from the electrode surface.

Developing a reliable sealant or sealing system remains one of the top priorities in research on planar solid oxide fuel cell (SOFC) technology. Many recent studies on SOFC glass-based sealants have focused on self-healing glass seals, such as those made of compliant alkali-containing silicate glass.
The objective of this study was to investigate the effect of fillers on the strengthening of alkali/alkaline-earth glass sealants after sealing and during operation at elevated temperatures. This was accomplished by adding Al₂O₃ powder as a filler to the alkali-containing borosilicate sealing glass to strengthen glass network structure. In alkali aluminosilicate glass, the trivalent aluminum ions act as network formers when the alumina content is lower than that of the alkali oxide. In this study, glass-based sealants, both with and without an Al₂O₃ nanopowder or a 8YSZ nanopowder added as fillers, were prepared. The effects of filler types on the strength, viscosities, electrical conductivities, and infrared spectroscopy of the sealants were investigated. It was concluded that strengthening of glass network structure was possible by the addition of Al₂O₃ nano-filler.

Among the most appealing features of the solid oxide fuel cells (SOFC) the efficiency to convert chemical energy into electric energy as well as the low environmental impact which have these electrochemical devices are remarkable. Because of that SOFCs are a relevant alternative to generate clean energy; however there are a lot of improvements to do in them, as lowering the operation temperature and production cost. Despite the cogeneration processes to rise the SOFCs efficiency, research on materials to be used as electrolytes or electrodes is the way to get both: lower operation temperatures and better efficiencies. In this work were studied two effects on the electronic properties of the Sr₂Fe_1+xMo_1-xO₆ double perovskites: the owed to the Fe/Mo ratio and the one from the nanopores grown direction in them. Those effects must modify the ionic diffusion routes and are proposed on the basis of the calculated electronic properties. The Density Functional Theory calculations were made on the Generalized Gradient Approximation, using the Perdew-Burke-Ernzerhof (PBE) functional. Hydrogen and oxygen ions were located at high symmetry points in the perovskites to calculate the system energies, from which diffusion routes were proposed.
Acknowledgments: This work was partially supported by the multidisciplinary projects 2014-1640 y 2014-1641 from SIP-Instituto Politécnico Nacional, PAPIIT-IN100313 from Universidad Nacional Autoacute;noma de México and 131589 from Consejo Nacional de Ciencia y Tecnología.

Many energy storage systems rely on the intercalation of a small solute into the interstitial sites of a host matrix. One of the simplest systems to study is the intercalation of hydrogen atoms in metals, and particularly palladium, as hydrogen uptake tends to occur near ambient temperature and at low hydrogen pressures. While studies on the palladium-hydrogen system have often focused on collections of nanoparticles, it is often difficult to deconvolve intrinsic properties of individual particles from averaging effects.
In this presentation, we investigate hydrogen loading of individual palladium nanoparticles between 13 and 29 nm. The particles are prepared in two different syntheses by 1) an ascorbic acid-mediated reduction of an aqueous solution of CTAB and hydrogen tetrachloropalladate and 2) sodium ascorbate-mediated reduction of an aqueous solution of CTAB and sodium tetrachloropalladate. Our syntheses yield both single-crystalline cubic nanoparticles and multiply twinned, icosahedral nanoparticles, which we compare with our single-particle studies. We use environmental scanning transmission electron microscopy combined with electron energy loss spectroscopy to monitor the bulk plasma frequency of palladium nanoparticles at varying hydrogen partial pressures. Importantly, the bulk plasma frequency provides a direct measure of the particle&’s complex permittivity. Multivariate curve resolution then allows determination of particle phase at each pressure, allowing us to construct hydrogen loading isotherms.
We first consider single-crystalline Pd nanoparticles. Isotherms indicate that these particles undergo an abrupt phase change between a hydrogen-poor α phase and a hydrogen-rich β phase upon an increase in the hydrogen partial pressure [1]. We compare these results with the isotherms of individual icosahedral particles, investigating the impact of defects on hydrogen absorption. Unlike single-crystalline cubes, icosahedra absorb hydrogen gradually because defects alter the chemical potential of hydrogen in the system. As hydrogen enters the metal, it expands the lattice, giving rise to a long-ranged H-H interaction. Defects reduce the elastic coupling of the system, effectively screening of the elastic H-H interaction.
By modeling the chemical potential we successfully reproduce the experimental isotherms of single-crystalline and icosahedral nanoparticles, showing how hydrogen can be used to probe defect content at the single particle level.
[1] A. Baldi, T. C. Narayan, A. L. Koh and J. A. Dionne, Nature Materials AOP (2014)

Thermoelectric materials are capable of converting heat directly into electricity and vice versa based on Seebeck effect and Peltier effect, and thus have been widely pursued for applications in waste heat recovery and solid state thermal management. In this work, we develop asymptotic homogenization to analyze the effective behaviors of three-dimensional thermoelectric composites. Asymptotic homogenization theory is first developed from the governing equation of thermoelectricity. The separation of length scales allows us to construct a set of solutions similar to Green&’s function method, with which the governing equations on the unit cell are formulated, and appropriate interfacial continuity conditions and boundary conditions are derived. The unit cell problem is further solved numerically by finite element method, and the homogenized governing equations are then developed for thermoelectric composites with the effective thermoelectric properties calculated. It is discovered that the homogenized thermoelectric equations are significantly different from those of homogeneous materials. The analysis provides considerable insight into the effective behavior of thermoelectric composites, and can help guide the design and optimization for high performance thermoelectric materials.

Bi₂Te₃ semiconductor has been widely used as thermoelectric material. On the contrary, silicon has been considered as the impropriate material due to high thermal conductivity property. However, recent research revealed the possibility of silicon as thermoelectric material by incorporating nanotechnology. One-dimensional (nanostructured) silicon nanowire can dramatically reduce the phonon propagation through the nanowire while maintaining the electron/hole propagation property.
In this work, top-down silicon manufacturing process is adopted to implement the n-/p-leg included silicon thermoelectric device. The 50 nm width n- and p-type silicon nanowires (SiNWs) are manufactured by incorporating conventional KrF photolithography with O₂ plasma PR ashing technique. Phosphorus and BF₂ ion implantation methods were applied for the doping of n- and p-type nanowires on 8-inch silicon wafer. For the evaluation of the Seebeck coefficients of the SiNWs, heaters and temperature sensors embedded test pattern is fabricated. For the elimination of electrical and thermal contact resistance issues, the SiNWs, heaters and temperature sensors are fabricated monolithically using a CMOS process. Electric and thermoelectric properties of 50 nm n- and p-type SiNWs are investigated with the variation of doping concentration from 1.0×10²⁰ to 2.5×10²¹ cm^-3. The optimized maximum power factor values are 1.59 and 2.43 mW#8729;m^-1 K^-2 for n- and p-type SiNWs with the doping concentration of 4×10²⁰ cm^-3. For the higher doping concentration over 4.0×10²⁰ cm^-3, electrical resistivity increases and Seebeck coefficient decreases rapidly, due to the imperfection of crystal structure. For the lower doping concentration below 4.0×10²⁰ cm^-3, the increased resistivity gives dominant impact on the power factor.

The non-toxic and light-weight (~2.0 g/cm³) material magnesium silicide (Mg₂Si) is one of the dominant candidates for a thermoelectric (TE) material that could enable devices to operate at temperatures ranging from 500 ~ 900 K. An important aspect of Mg₂Si is its capability for being doped in order to modify its electrical conductivity, thermal conductivity and durability at elevated operating temperatures. Some features of Mg₂Si, such as its lightweight, the relative abundance of its constituent elements, and its non-toxicity, can broaden its application in the field of thermoelectric devices and encourage the development of a possible system for waste heat recovery. In order to realize an Mg₂Si TE power generator, information about its mechanical properties such as strength, elastic moduli and hardness are required, in addition to a study of its thermoelectric capabilities. The current status for Mg₂Si is aimed toward TE module fabrication and appropriate system integration techniques for automotive applications. The basic mechanical properties of Mg₂Si, such as Young&’s modulus, bending strength and fracture toughness, are needed in order to design TE modules and TEG systems.
As a promising donor impurity in Mg₂Si, antimony (Sb ) has been dominantly used in our previous experiments, enabling in practical terms an impeccable power factor of ~3.0x10^-3 W/mK² and a ZT value of ~0.78 at 873 K. Sb-doping also provides satisfactory thermal durability, showing low and stable electrical resistivity characteristics at 873 K in atmospheric conditions for up to 1000 h. To modify thermoelectric properties of Mg₂Si doped solely with Sb, an isoelectric impurity of zinc (Zn), which is expected to predominantly substitute in place of Mg, was incorporated as a supplementary dopant in Mg₂Si. The concentrations of co-doped Sb and Zn were varied from 0.5 to 1.0 at%, respectively, and both elements were introduced into a congruent melt of Mg₂Si using the all molten synthesis method. The resultant polycrystalline Mg₂Si was pulverized and then sintered using a “Plasma Activated Sintering” (PAS) technique. Almost all of the co-doped specimens typically had no cracks and a relative density of 98% or more, indicating the good sinterability and the reproducibility.
We examined ultrasonic tests, nano-indentation, indentation fracture test methods on sintered Mg₂Si specimens. We report here the mechanical properties, such as Young&’s modulus, hardness, fracture toughness with reference to the TE characteristics of the Seebeck coefficient, electrical conductivity, and thermal conductivity. The observed values for the Young&’s modulus, hardness, fracture toughness were 140 GPa, 700, and ~1.2MPam^(1/2), respectively. For the thermoelectric properties, the highest power factor and ZT value were 3.4x10^-3 W/mK² at 723 K and 0.95 at 873 K, respectively, when the content of doped Sb and Zn were 0.5 at% each.

With the rapid growth of portable electronics and sensor networks, mobile and sustainable energy sources for these devices becomes indispensable in modern society. Among all of the energy sources, the universally available mechanical energy has become a plausible solution. Recently, triboelectric nanogenerators (TENGs) based on contact electrification and electrostatic induction emerge as a promising mechanical energy harvesting technique because of their unique figure of merits, including high output power density, ultra-high energy conversion efficiency, low weight, cost-effective materials, and high adaptability design to different applications. However, because of the inherent uncontrollable and unstable characteristics of the environmental mechanical energy source, the converted electrical energy from TENG is thus unstable and hardly utilized to directly power electronic devices. Thus, understanding the integration performance of TENGs with an energy storage unit is critical for designing a practical energy harvesting system. However, until now there are still many mysteries about the charging behavior of TENGs and a in-depth study is completely necessary.
In this work, the characteristics of utilizing a TENG to charge a capacitor are thoroughly studied. First, the easiest unidirectional charging is analyzed. An optimum load capacitance that equals to the TENG inherent capacitance is observed to maximize the energy storage. Then a much more complicated multiple-cycle charging process is analyzed to show the unique TENG charging characteristics. Through rigorous mathematical derivation, TENGs also have the saturation charging behavior, which is completely analogous to utilizing a DC voltage source with an internal resistance to charge a load capacitor. The TENG with larger short circuit transferred charges and smaller inherent TENG capacitance can charge a load capacitor to a higher saturation voltage under unlimited charging cycles. Besides, the TENG with larger inherent capacitance can provide a smaller charging time constant. An optimum load capacitance that matches the TENG&’s impedance is observed for the maximum stored energy as well. However, different from unidirectional charging, this optimum load capacitance is linearly proportional to the charging cycle numbers and the inherent capacitance of the TENG. Finally, corresponding experiments were performed to further validate the above theoretical prediction to show its application in guiding experiments. This work clearly shows the unique TENG charging characteristics, which can serve as an important guidance to design an integrated TENG energy harvesting system for practical application. [1]
Reference:
1. S. Niu, Y. Liu, Y. S. Zhou, S. Wang, L. Lin, Z. L. Wang. IEEE Trans. Electron Devices. Accepted.

Two-dimensional (2D) semiconducting materials, such as phosphorene, MoS₂, et al. hold great promise for many important applications, such as in nanoelectronics, molecular and bio-sensors, thermoelectric conversion and solar energy harvesting. To fully explore their functionalities and potentials, band-gap engineering is often required. It is now recognized that these 2D materials are able to subject to large tensile strains, which can cause significant changes in electronic, optical, magnetic and thermal properties, which in turn can greatly widen the range of applications of 2D materials. In this talk, we report our research work on the strain engineering of two members of 2D semiconducting materials, that is, phosphorene and MoS2, to tune their electronic, magnetic and thermal properties using first-principles calculations.
For phosphorene, we observe a strong anisotropy of phononic properties in phosphorene. Through phonon dispersion analysis, we find that phosphorene is structurally stable under large uniaxial and biaxial tensile strains. We also uncover several “hidden” directions along which small-momemtum phonons are found to be “frozen” with strain and possess the smallest degree of anharmonicity. We also find a significant crystallographic orientation dependence of thermal conductance, and the thermal conductance anisotropy with the orientation can be tuned by applying strain.
For MoS₂ sheet, we find that strain can significantly change the electronic band gap and cause a transition from a direct band semiconductor to an indirect band semiconductor. In addition, we also find that strain can also change the band gap and the electron and hole effective masses of MoS₂ nanotubes. Furthermore, strain can also cause a large change in magnetic moment for MoS₂ nanoribbons.
The present work demonstrates a practical route to tune the electronic properties of 2D semiconducting materials by strain engineering, which may be useful for applications in nanoelectronic devices and energy conversion.

Understanding the phase behaviors of electrodes during operation is an essential step toward improving the electrochemical performance of Li-ion batteries. In this presentation, we present a novel statistical approach to investigate the physical origin of the interface alignment between Li-rich and Li-lean phases in phase-separating electrode particles in terms of interfacial coherency states, which are altered upon incorporation of structural defects. Using Khachaturyan-Shatalov microelasticity theory and phase-field simulations, we explore the entire space of possible degrees and types of coherency states, and identify all possible phase boundary orientations and the probabilities of their occurrences employing Li_XFePO₄ as an example. The results are able to rationalize the wide variability in available experimental characterizations, which has hitherto been poorly understood. We further show that by analyzing the statistical measures of the densities of interfacial coherency states for representative phase boundaries, we can gain detailed insight for the propensity and nature of phase boundary orientation variation during cycling. Key implications for understanding the thermodynamics and kinetics of phase boundary formation in Li_XFePO₄ are discussed.

Interfaces are ubiquitous and important in battery electrode as they occur in regions of material intergrowth, in compositional gradients and also at interfaces with protective coatings. In this work, we study the Li intercalation voltage behavior across the interface of olivine FePO₄-MPO₄ (M=Co, Ti, Mn). These olivine-structured interfaces provide a representative model to help us further understand the physics of Li intercalation across material interfaces with different intercalation voltages. We find that as one moves across the interface from high to low Li intercalation voltage material the voltage stays constant in the higher voltage material and decays approximately linearly in the lower voltage region. This effect is shown to be due to electron transfer from the intercalated Li across the interface. The existence of the higher voltage material and the interface can significantly enhance the Li intercalation voltage in the lower voltage region within a 1-2 nm scale. Implications of this result for cathode structures will be discussed.

As the application of electrical devices is growing up, importance of using more efficient energy storages for high energy density applications, such as hybrid cars, mobile electric devices and so on, is being evident. Rechargeable lithium-sulfur (Li/S) batteries are a promising candidate for next generation of energy storages since sulfur as an alternative material for cathode has been successfully exhibited highest theoretical specific energy density after oxygen among all periodic table elements (~2600 Wh/kg). In spite of all advantages, polysulfide (Li₂S_x) shuttling process has a hug effect on the performance of these batteries, results in very low recyclability, high capacity fading, and adversity of analysis in modeling. The primary goal of this work is to propose a facile mathematical model to predict capacity as a function of cycle number by considering shuttling effect for different nano-structured cathodes in Li/S battery systems. A structure-based sulfur-confinement factor is quantitatively determined for each cycle and in a stationary numerical model we propose an expression for accumulation of solid polysulfide deposited on the anode surface as solid electrolyte interphase (SEI) layer. Capacity fading during cycling is mainly attributed to the concentration of reduced polysulfide on the anode surface and is expressed as a function of sulfur-confinement factor as well as shuttling constant. In addition, high plateau polysulfide dynamic in electrolyte during charge and discharge process is studied in detail and electrochemical performance of the battery is modeled and compared with experimental data of some of the most well-known cathode structures. An ionic/electric resistivity model for anode is also built to study resistivity enhancement during cycling performance, due to creation of SEI layer. The discrepancy between the model and the collected data, for most of the analyzed structures is in the order of 10% for the first 15 cycles and less than 5% for the next cycles.

For electric vehicles being widely accepted, the driving distance achieved from a single charge must be increased. To increase the driving distance, batteries need to be reconfigured with materials that have high capacities. Li ion batteries are the most extensively used for long-term energy storage, portable electronics and transportation, though chemists are exploring all kinds of new batteries. As all the other traditional batteries, Li ion battery has two electrodes, which host the Li ion insertion/extraction reversibly. Thus the larger capacity density, the more volume expansion, which causes irreversible electrode materials fracture. Novel materials such as silicon (Si) are currently desirable replacements for graphitic carbon based systems in commercial use because of Si&’s improved gravimetric power density. However, Si anodes suffer from capacity and power fade, which make them still inapplicable for electric vehicles batteries. We developed a fundamental understanding of the Si anode failure, which allows rational development of tailored battery electrode with minimal capacity decay during cycling.
In present work, we combine chemical/electrochemical experiments and finite element simulation (FEM) to study the chemical, electrochemical and mechanical response of single crystal Si electrode through multiple charge-discharge cycles. From battery experiment, we record the model Si surface charging-discharging curves and characterize the surface chemical signal by FTIR and XPS techniques. The evolution of surface morphology (e.g. roughening, micro-crack initiation and propagation) is inspected by scanning electron microscope (SEM) to determine electrode failure mechanism. In FEM, equivalent, transient thermal strain field is applied to represent the volumetric expansion/contraction in electrode induced by diffusion of Li ion through charge-discharge cycles. Electrode deformation is tracked and compared with experimental observation; possible crack initiation location and propagation direction are obtained by stress analysis, and correlated to failure modes observed.
The importance of this research is to provide in-depth understanding on the failure modes of Si electrode and provide further guidance for future process and design optimization (e.g. surface modification, additive selection, electrolyte recipe) to eliminate or minimize the electrode failure/degradation in Li-ion battery system. And such model electrode investigation methodology can be applied to other ion-host energy storage materials.

Chemo-mechanical models have been extensively used in the stress analysis and device design in lithium-ion batteries. In the present study, we generalize the conventional model to take into account the anisotropy of electrode materials, the strain-dependence of diffusivity and the surface energy, respectively. First, we propose a theoretical model to study the diffusion-induced stress in the transversely isotropic electrodes during lithiation. It is indicated that the radial and tangential stresses depend slightly on the Young&’s modulus ratio and lithiation expansion coefficient ratio of axial direction to isotropic plane, while the axial stress depends strongly on the two ratios. Then, based on our first principles results that reveal the strong influence of strain on the diffusivity in graphene-decorated electrode materials, a coupling model is established and demonstrates that the strain-dependent diffusivity greatly affects diffusion-induced stresses within the electrode particles. Finally, we develop a nonlinear chemo-mechanical equilibrium model for battery materials considering the surface elastic energy. It is shown that surface stress has a significant effect on the facture behaviors of electrodes. The above studies put the related work in the literature to a higher level and are instrumental for engineering design and practical application of lithium-ion batteries.

Lithium ion batteries are the most widely used energy storage device in both portable electronics and electric vehicles due to its high energy density. However, there are still remaining issues unsolved. To further improve the Li-ion battery performance, it&’s important to understand the lithium ion intercalation/de-intercalation process upon cycling. The development of a new scanning probe microscopy technique (SPM), electrochemical strain microscopy (ESM), in which a combination of AC bias and DC bias is applied to the sample surface, allows us to evaluate mobile ionic diffusivity and electrochemical reactivity on the nanoscale.
In this work, we will first discuss about using a series of strain-based SPM techniques to distinguish different microscopic mechanisms including electromechanical coupling, spontaneous polarization, induced dipole moment and electrostatics. Then we will demonstrate the localized probing of lithium ion transport property in a lithiated carbon anode using ESM. Pristine carbon anode is also studied as comparison. Also, we&’ll investigate the lithium ion transport through the solid electrolyte interface (SEI) layer, which is formed on the carbon anode surface upon cycling.

Thermoelectric materials are capable of converting heat directly into electricity and vice versa based on Seebeck effect and Peltier effect, and thus have been widely pursued for applications in waste heat recovery and solid state thermal management. In this work, we develop asymptotic homogenization to analyze the effective behaviors of three-dimensional thermoelectric composites. Asymptotic homogenization theory is first developed from the governing equation of thermoelectricity. The separation of length scales allows us to construct a set of solutions similar to Green&’s function method, with which the governing equations on the unit cell are formulated, and appropriate interfacial continuity conditions and boundary conditions are derived. The unit cell problem is further solved numerically by finite element method, and the homogenized governing equations are then developed for thermoelectric composites with the effective thermoelectric properties calculated. It is discovered that the homogenized thermoelectric equations are significantly different from those of homogeneous materials. The analysis provides considerable insight into the effective behavior of thermoelectric composites, and can help guide the design and optimization for high performance thermoelectric materials.

The ability to control mechanical wave propagation has significant implications for energy conversion and storage devices. In particular, this includes the high frequency mechanical vibrations that contribute to heat conduction. Materials containing local mechanical resonances, or “locally-resonant acoustic metamaterials”, have been shown capable of drastically affecting mechanical wave propagation. The majority of experimental studies on such materials have involved mechanical waves with sub-megahertz frequencies. In this talk, I will discuss our recent work, in which we studied the propagation of surface acoustic waves (SAWs) in a metamaterial composed of a self-assembled monolayer of microspheres adhered to the surface of an elastic substrate. The laser-induced transient grating technique is used to generate and detect SAWs, and to characterize their dispersion. The measured dispersion exhibits “avoided crossing” phenomena, as a result of the hybridization of the SAWs with the contact resonance of the microspheres at frequencies in the hundreds of megahertz. The measured dispersion is matched by an analytical model, where the contact resonance is the only fitting parameter. The measured contact resonance is compared with estimates based on nonlinear adhesive contact models. This study may lead to a new class of self-assembled, locally-resonant metamaterials designed to affect even higher frequency mechanical waves relevant to energy conversion and storage applications.

One of the key challenges in the area of solar cells is to develop new mechanisms for improving their absorption efficiency, which is well below their thermodynamic limit. While multi-junction solar cells are approaching the maximum possible efficiencies, quantum dot-based solar cells are evolving as a promising alternative. The key feature that is used to define their basic behavior is size, which restricts the spatial expansion of the electronic wavefunctions, thereby, producing confinement of electrons and allowing absorption of photon energy. There are many materials (such as CdSe, PbS, GaAs) that show good confinement characteristics, but they are usually expensive. SiGe quantum dots, which are cheaper and Si-technology compatible, form dots that are much larger (on the order of 50 nm) in size compared to their single crystalline counterparts (which are on the order of 5 nm). Consequently, their confinement ability have usually been disregarded. Here, using a parallel multiscale computational framework that combines density functional theory, k.p method and the finite element calculations, takes into account the effects of all components of the strain tensor and effective mass of the electrons and holes. Results show that heterogeneous distribution of an array of alloy quantum dots, mediated by non-uniform mismatch strain at the interface between the quantum dots and the substrate leads to substantial confinement of electrons. It shows a new pathway for enhancing controllability of the absorption characeteristics of QD solar-cells solely from a mechanics perspective. The influence of near-neighbor interactions of the QDs are also discussed.

Polymer-electrolyte fuel cells (PEFCs) are energy-conversion devices with great potential to supply clean electrical power for transportation and stationary applications. PEFCs are required to perform over long operational times without the failure of cell components, in particular, the polymer-electrolyte membrane (PEM). The role of the PEM is to provide a conductive pathway for protons (anode-to-cathode) while blocking reactant crossover. In PEFCs, chemical decomposition of the membrane due to attack by hydroxide radicals formed during operation is accompanied by the physical defects that arise due to mechanical stresses caused by cell assembly and operating environment. Combination of chemical/mechanical degradation leads to increased crossover of reactant gases and reduced power loss. Due to the complex interaction between the chemical and mechanical failure modes in a PEFC, it is imperative to understand the relationship between different failure modes and the dominating mechanisms that control PEM lifetime. In this talk, we investigate the effect of compression on chemical degradation of perfluorosulfonic-acid (PFSA) ionomers, the most widely used PEM in PEFCs. Chemical decomposition of the PEM increases with increasing levels of compression. Moreover, the mechanical response of the membrane also changes upon chemical degradation. These results con#64257;rm and provide insight into the synergistic nature of chemical/mechanical failure in PEMs where the two mechanisms are strongly coupled in a detrimental manner. Compression and degradation of the membrane also change its nanostructure, consistent with the changes in membrane's mechanical properties and chemical decomposition rate. It is likely that the deformation energy accumulated in the membrane due to mechanical loads accelerates the PEM chemical decomposition due to weakened attack sites. The #64257;ndings here have implications on understanding real-world operation wherein multiple chemical and mechanical stressors co-exist and interact synergistically.

In this work it is intended to do a comparative analysis between bismuth nanotubes (BiNTs) and carbon nanotubes (CNTs) and to reveal which influence these inclusions have on the thermoelectric properties of their matrix material. Therefore we want to create three different types of Bi_1#8209;xSb_x - nanoalloys and compare their resulting thermoelectric transport parameters against each other. First it focusses on the synthesis of BiNTs and the inclusion of these in Bi_1#8209;xSb_x - nanoalloys, the corresponding matrix material, as well as the thermoelectric characterization of the alloy-nanotube composites. The necessary BiNTs were synthetized by a transformation of a β - BiI precursor with n - BuLi solution into tubular bismuth structures. Moreover CNTs, purchased from Bayer Materials, will get included in the Bi_1#8209;xSb_x - nanoalloys as well and undergoing a thermoelectric characterization also. The corresponding nanoalloys, acting as a matrix material, were produced by ball milling. To do so the bulk powder was milled under argon atmosphere for 20 hours at 450 rpm and with a ball-to-powder ratio of 7.5:1. Including different quantities of nanotubes in those alloys, resulting in three series: an alloy series without inclusions and two series with BiNT's contents of 3 and 5 wt.% and CNTs with 0.5 wt.%. This leads to the possibility to compare three different species of alloys against each other.

Presented here are Silicon (Si) nanowire network based thermoelectric devices fabricated on flexible copper foil substrates. Nanowire networks provide a flexible and scalable material platform for heat recovery. Thin layers of electrically conductive titanium nitride deposited on flexible copper substrates served as barrier layers for the growth of Si nanowire networks. After the growth of the Si nanowire network, a secondary Si growth was used to form a continuous top layer that prevented electrical and thermal shorting. Semiconductor nanowires were directly in contact with top and bottom metallic electrodes which allowed multiple nanowire networks to be stacked up into thick thermoelectric devices. In this study, single, double, and quadruple stacked devices were fabricated and tested. Total thermoelectric power output increased as the number of nanowire networks stacked increased, for a given temperature gradient per stack. We demonstrate the concept of multiple nanowire networks stacked into thermoelectric devices in series is scalable making it possible to generate incrementally higher power by goingfrom two-layer to four-layer devices thereby supplying a larger temperature gradient, increasing power production by 27%. We present and assess the concept of a multi-stage thermoelectric device design that is laterally and vertically scalable, using semiconducting nanowires, and removing the need for both n-type and p-type semiconductors within the device.

Manganese dioxide (MnO₂), and in particular Bi- and Ba-doped MnO₂, is promising as an electrode material for low-cost, high-capacity rechargeable alkaline batteries. Due to the changes in density that occur along the reaction pathway from MnO₂ to MnOOH to Mn(OH)₂ and back, the electrodes can experience up to ~25% volume swelling/contraction during cycling. As such, one of the limitations to achieving practically-relevant cycle lifetimes with MnO₂-based electrodes is the loss of electrical contact between regions of the active material and current collector, particularly if a macro-porous separation membrane is used. In addition, a dissolution-reprecipitation reaction occurs during the second electron discharge step, which is another source of active material loss. In the literature, while compression is commonly employed when evaluating long-term cycling performance, to the best of our knowledge the exact role played by compression in the performance of MnO₂ electrodes has not yet been studied.
We present on our work towards gaining a fundamental understanding of the relationship between mechanical compression, storage capacity, as well as capacity retention over the course of the cycling life. We use cellophane-wrapped separators to limit the loss of active material to the fibrous membranes. Using flat, flexible piezoresistors, we measure the force applied to the electrodes during cell assembly and find that while some compression is beneficial to cycling performance. However, we observe that with increasing compression force, there is a decrease in the maximum specific capacity that is achieved as well as an increase in the number of cycles during the break-in period. Nevertheless, cycling stability does seem to benefit from the increased compression. The piezoresistors are also used to monitor the forces experienced within the cell during cycling. In sum, these insights are valuable and crucial to achieving further improvements in performance through optimized electrode and cell design.

Solar-selective absorbers (SSAs) are used in a variety of solar thermal energy systems (e.g. concentrating solar power, solar thermoelectric generators, solar thermal heat engines, solar hot water systems etc.) to absorb solar radiation from the sun and convert it into thermal energy, while minimizing heat leakage out of the system by thermal radiation. The goal of SSA is to exhibit high absorptance in the solar wavelength regime (280 nm to 2500 nm), and high reflectance, therefore low emittance, in the IR region while at the temperatures of interest. High-temperature SSAs are critical for reducing the cost of electricity from photothermal devices and from all receivers used in concentrated solar power. In order to address the issues of efficiency and durability of solar-selective coatings for power tower applications, we have developed a multilayer SSA coating structure for high-temperature applications. The multilayer coatings were fabricated by reactive co-sputtering (similar to the process used for commercial receiver tubes). The performance of the SSA coatings was evaluated in the 500-1000 °C temperature range, and both their structural and optical properties were studied. The SSAs have proven to be stable after testing through multiple furnace cycles to 1000 °C for a 1-hour duration in different controlled environments. They exhibited low infrared emissivity and high solar absorptance of asymp; 95%. On-sun testing of the SSAs was performed at NREL&’s high-flux solar furnace (HFSF) with successful cycling of an SSA sample to 1000 °C. When compared with samples coated with colloidal graphite, the samples with multilayer SSAs require less optical power to achieve the same temperature, demonstrating greater optical-to-heat-flux efficiencies.

Filled skutterudites are the most potential thermoelectric (TE) materials for TE applications, particularly in automotives, because they combine excellent TE properties in the temperature range needed with excellent mechanical properties, the starting material is cheap and abundant and they can be produced easily and fast and already in large scale. This paper will outline preparation and optimisation of Sb-substituted p-type and multifilled n-type skutterudites with ZT values approaching 1.3 and 1.8 with thermal-electric conversion efficiencies eta; > 14% and eta; > 17%, respectively. Severe plastic deformation via high-pressure torsion due to introduced defects and cracks as well as grain refinement can even top these ZT-values. In addition a review on mechanical properties essential for TE device engineering will be given.

The homogeneity region at 450°C and solidus temperatures have been defined for novel quaternary filled skutterudites Ba_yNi₄(Sb_1-xSn_x)₁₂ by means of Electron Probe Microanalysis (EPMA), X-ray Powder Diffraction (XPD) and Differential Thermal Analyses (DTA). For two selected samples, Ba_0.73Ni₄Sb_8.1Sn_3.9 and Ba_0.95Ni₄Sb_6.1Sn_5.9, temperature dependent single crystal X-ray structure analyses (at 100 K, 200 K, 300 K) revealed the thermal expansion coefficients, Einstein and Debye temperatures and the Debye Waller factors. These Atom Displacement Parameters (ADP&’s) indicate rather insignificant differences between framework and filler atoms, suggesting a strong coupling of the filler atoms with the framework.
Physical property measurements (i.e. temperature dependent specific heat, electrical resistivity, Seebeck coefficient and thermal conductivity) compare unfilled Ni₄Sb_8.2Sn_3.8 with the Ba-filled samples, Ba_0.42Ni₄Sb_8.2Sn_3.8 and Ba_0.92Ni₄Sb_6.7Sn_5.3. Small Sommerfeld constants from specific heat data (γ<15 mJ/molK) are consistent for all samples with low carrier concentrations (n~4x10²⁰/cmsup3;) derived from Hall data (n-type from negative Sebeck coefficients). Interestingly Ba-fillers and rising Sn-contents increase the Debye temperatures significantly. Electrical resistivities reveal a crossover from metallic to semiconducting behaviour, which is discussed in terms of a temperature-dependent carrier concentration employing a simple model for a rectangular density of states with the Fermi energy slightly below a narrow gap. In contrast to common features in filled skutterudites, a rising Ba content in Ba_yNi₄(Sb_1-xSn_x)₁₂enhances thermal and lattice thermal conductivity.
Finally, mechanical properties of Ba_yNi₄(Sb_1-xSn_x)₁₂ (elastic moduli, thermal expansion, hardness) will be compared with the general behaviour of skutterudites.

Thermally-activated (or molten salt) batteries utilize salts as the electrolyte. A main advantage of these types of batteries is that there is no self-discharge prior to activation, which significantly increases the shelf life compared to conventional electrochemical batteries. Since the electrolyte is solid at room temperature, a heat source (usually an internal pyrotechnic) is required to melt the electrolyte and activate the battery stack. During activation, these batteries commonly undergo a stack relaxation caused by the thinning of the separator layer. An open question is what becomes of the melted electrolyte during this activation process. Infiltration of electrolyte into the cathode and anode reduces the internal impedance of the battery; however, excess electrolyte could cause collapse or shorting of the cells. Thus, an improved fundamental understanding of the electrolyte transport during activation is needed to enhance the performance of future thermally-activated batteries.
In this work, we performed a series of fundamental studies using bromine as a tracer element to track the mobility of the electrolyte within single cells. Since the electrolyte is solid at room temperature, we were able to perform electron probe microanalysis (EPMA) to quantify the transport of the bromine tracers within the separator and electrodes of the battery. By utilizing a series of single cells frozen at various activated times, we elucidated the complex process of electrolyte transport from the separator into the electrodes. Collectively, these studies offer a step towards understanding and predicting ionic permeability through the various components of molten salt batteries.
Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy&’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

Multicaloric materials, which couple the material&’s thermal response to applied electrical, magnetic, and elastic fields, hold promise for technological applications, such as refrigeration or energy reclamation from waste heat. In this presentation our work focused on understanding the electrocaloric effect in BaTiO₃ and KNbO₃, and the elastocaloric effect in PbTiO₃ is presented. The impact of applying elastic strain, in the form of epitaxial loading, on the electrocaloric response is also presented. The adiabatic temperature change, ΔT, is calculated using a first-principles based effective Hamiltonian method that is studied using a molecular dynamics simulation. Both an “indirect” method, which uses Maxwell relations, and a “direct” method, which pairs a canonical to microcanonical simulation, are used to predict ΔT. Simultaneously applying compressive epitaxial strain and electric fields to BaTiO₃ results in the emergence of a monoclinic phase just below the Curie temperature. This new polymorph that is due to a coupling of elastic and electric fields causes the local polarization to rotate. From a technological perspective this is highly important because having multiple polymorphs near the Curie temperature results in a broadening of the temperature range over which the electrocaloric effect can be experienced. From a fundamental science perspective, the co-existence of polymorphs provides a model system for studying phase transformations.

Silicon is a promising anode material for lithium-ion batteries due to its enormous theoretical energy density. The practical viability of silicon electrodes, however, has been limited by fracture during electrochemical cycling. We have performed a number of experiments to examine the mechanical behavior of amorphous silicon electrodes of lithium-ion batteries. In particular, we have measured the fracture energy of lithiated silicon thin-film electrodes as a function of lithium concentration. The fracture energy is found to be similar to that of pure silicon and essentially independent of the concentration of lithium. Thus, while lithiated silicon can flow plastically, it appears to fracture in a brittle manner. We have also varied the rate of lithiation of amorphous silicon thin films while simultaneously measuring stresses. Increasing the rate of lithiation resulted in a corresponding increase in the flow stress. These observations indicate that rate-sensitive plasticity occurs in a-Li_xSi electrodes at room temperature and at charging rates typically used in lithium-ion batteries. Using a simple mechanical model, we have extracted material parameters from our experiments, finding a good fit to a power law relationship between the plastic strain rate and the stress. This rate-sensitivity provides insight into the unusual ability of a-Li_xSi to flow plastically while fracturing in a brittle manner. Moreover, the results have direct ramifications concerning the rate-capabilities of silicon electrodes: faster charging rates (i.e., strain rates) result in larger stresses and hence larger driving forces for fracture.

We have developed a robotic instrument that can measure the electrochemical impedance of hundreds of thin film microelectrodes in automated fashion. By measuring electrodes with systematically varied area, thickness, surface decoration, and composition, it is possible, in principle, to probe reaction pathways, decouple bulk and surface properties, and rapidly screen hundreds of chemical compositions to discover trends and identify new high activity materials. Here we introduce the capabilities of this new instrument by using it to measure geometrically graded microdot electrodes of the solid-oxide fuel-cell (SOFC) cathode material (La_0.8Sr_0.2)_0.95MnO_3-δ (LSM). We collect A.C. impedance spectra from several hundred microdots with diameters ranging from 100 to 500thinsp;mu;m and thicknesses from 30 to 300thinsp;nm over the temperature and oxygen partial pressure ranges of 700 to 800thinsp;°C and 3.2 × 10^-4 to 1thinsp;atm, respectively. Automated data analysis using a physically motivated equivalent circuit model yields physical parameters for each dot at each measurement condition. The LSM surface reaction resistance and bulk ionic resistance both exhibited a power law dependence on dot diameter with an exponent close to -2, indicative of a surface reaction pathway that encompasses the entirety of the dot surface. The slight deviation from -2 is attributed to local cooling of the sample by the microprobe tip, which slightly increases the resistances for smaller diameter microelectrodes. A surprising increase in surface reaction resistance with microelectrode thickness was observed, tentatively assigned to a decrease in grain boundary density as LSM grain size increases with increasing film thickness. The results set the stage for exploration of a wide range of gradient types, from composition to growth temperature to catalyst coating, while the use of impedance spectroscopy implies that a broad range of properties, from ionic conductivity to material nonstoichiometry, can be extracted.

DMFC, Direct Methanol Fuel Cell Technology, can be used for fabrication of methanol sensor. A fundamental limitation in DMFC technology is methanol crossover. In this process methanol diffuses from the anode through the electrolyte to the cathode, where it reacts directly with the oxygen and produces no electrical current from the cell. Poisoning of the cathode catalysts is also another major problem. We have used a passive mode design protocol using COMSOL Multiphysics to seek solutions for these problems. The design and simulation would involve optimization of various parameters, in the construction of the cell. This would optimized the overall power density and hence the sensitivity of the sensor by the modification of various parameters like the area of the working electrodes and, separation distance and the electrode-electrolyte interface.
A passive mode design, of about a cm area, using various parametric functions, and interfacing Darcy&’s law of fluidic flow through a porous medium, under specific pressure and temperature, was applied.
The designing involves the construction of gas diffusion layers using carbon cloth for anode and cathode with various parametric variations. Nafion membrane was selected as proton exchange membrane for the construction with different interface structure to analyze the sensor&’s performance. Platinum and various alloy catalysts like Pt-Ru, Pt-Fe, Pt-Sn and Pt-Mo was chosen as the working catalysts.
There is contributes to the loss of cell voltage potential, due to cross over as any methanol that is present in the cathode will be oxidized. The change in the overlap length and porosity optimizes the power density of the cell and hence the sensitivity. The obtained anode polarization curves indicate the drop in cell potential due to methanol crossover from anode to cathode due to oxidation occurred at cathode. This determines a strong anode activation control that was reflected in the overall polarization curve. The terminal voltage of the cell is deconvoluted into the anode and cathode polarizations: the actual E_cell = E_cathode-E_anode.
Alternatively, the anode polarization can be measured in the driven mode and the cathode curve is calculated from the above equation. In the driven mode, hydrogen is fed to the cathode that acts as both counter and reference electrode. Besides the strong activation control at the anode, the effect of the mixed potential on the cathode polarization curve is clearly observed. The parametric functions of the cell were optimized to obtain good order of sensitivity. A cell was fabricated and compared with the simulation results with good convergence of the results.

The thermomechanical reliability of low temperature polymer proton exchange membranes (PEM&’s) in hydrogen fuel cells is one of the biggest challenges in improving fuel cell reliability. During operation, the PEM is highly compressed between the flow channels of the polar plates while being exposed to a wide range of temperature and water content. The reliability of the membrane is controlled by several factors, including the molecular structure, the operation conditions (e.g. relative humidity, temperature), and possible contamination of the membrane. Of particular importance, however, the role of mechanical constraint on the thermomechanical properties, including the tearing resistance, is lacking but highly relevant because membranes are always highly constrained in the fuel cell geometry.
In our study, we used several classes of polymer membranes including perfluorosulfonic acid polymer (PFSA), styrenic block copolymer, brominated poly(phenylene oxide) (BPPO), along with other representative membrane materials such as polyethylene and polystyrene. We employed a suite of characterization techniques to assess salient mechanical properties, such as stiffness, strength, tearing resistance, and stress. We used a free tearing test to examine the effect of temperature and water content on local plasticity during tearing. In addition, we used a constrained tearing test to investigate the effect of mechanical constraint on the tearing resistance. Furthermore, we exploit a substrate curvature test to study the mechanical effect of water content and temperature on the membrane stresses that develop when confined. We report on the role of molecular structure on thermomechancial properties and particularly discuss the effects of constraint on tearing resistance and film stresses. This study provides insight on the role of mechanical constraint related to design of the fuel cell on reliability.

Solid oxide fuel cells (SOFCs) generate electricity with high electrochemical efficiency and low emission output, but the high required operating temperatures impose a number of challenges related to material performance, compatibility, and durability. For example, the anode reduction-oxidation (redox) reaction that occurs when the fuel supply is interrupted or the system is shut down leads to rapid oxidation of the metallic nickel phase, inducing large volumetric expansion strains. These redox strains can lead to detrimental micro-cracking or, potentially, catastrophic brittle failure of the entire cell. A better understanding of the complex chemical and structural evolution processes that occur during anode oxidation and their roles in overall mechanical and electrochemical degradation is therefore fundamental to designing more robust and efficient future generation SOFCs. In this talk, an expanded phase field model for SOFC anodes undergoing redox reactions will be presented. The model describes microstructural evolution associated with the Ni to NiO transformation, and in particular incorporates a large deformation elasticity framework suitable for describing the large transformation strains involved. Advective swelling of the metallic phase leads to transmission of stresses into the supporting ceramic oxide phase, and subsequent plasticity of the overall anode structure is described via a coupled phase field model for brittle fracture. Redox fracture resistance will be examined as a function of anode material properties and under various microstructural conditions, with the aim of identifying optimal qualities for SOFC durability. This work is supported by the Energy Frontier Research Center on Science Based Nano-Structure Design and Synthesis of Heterogeneous Functional Materials for Energy Systems funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (award DE-SC0001061).

Polymer electrolyte fuel cell (PEFC) was fabricated using highly stretchable Ag nanowires percolation network as current collector in the previous study. It showed comparable electrochemical performance to normal PEFC. Interestingly, the performance increased with the increase of bending. It was found that internal compressive stress normal to membrane-electrode assemply (MEA) generated by bending a polydimethylsiloxane (PDMS) endplate was a main factor of the decrease of ohmic and faradaic resistances, thus increasing the performance of flexible PEFC. In this study, setup for flexible PEFC was specially designed in order to measure the generated compressive stress and resulting performance quantitatively. As a result, difference between the compressive stresses firstly calculated from finite element method and measured directly from force sensor in this setup was not more than 50 % of original values. Tendencies of variations of ohmic and faradaic resistances were also same as the previous result that both resistances decrease exponentially as compressive stress increases. In addition, test setup for compressing the flexible fuel cell was specially designed again in order to explore the thickness effect of MEA and corresponding stress distribution effect. It showed almost no performance variation, meaning that quantitative value of compressive stress and resulting electrochemical performance was successfully measured.

Salinity gradient is an enormous source of clean energy. A process for such an energy conversion is being discussed A process for potential generation from an ionic concentration gradient produced in single and multicell assembly is presented. The ionic gradient is created using a fuel cell type cell with a microporous ion exchange membrane, both anionic (AEM) and Cationic (CEM). Various salinity gradients, Salt : Fresh, from 100: 0 to 16000: 0 was established using NaCl solution, in the electrode chambers. A potential of 20 mV/cm to 25 mV/cm can be realized at ambient temperatures and pressures for a bipolar AEM/CEM cell. The performance can be optimized for various static and dynamic flow rates of the saline and fresh water. The cell performance can also be optimized for Membrane Electrode System (MES) morphology. A multicell assembly is been assembled and the results presented for dynamic load characteristics. The thermodynamic, electrochemical and electrical efficiency is being discussed for various gradients and flow rates. The relation with number of valance electrons/ ion and the potential generated is being discussed for various dynamic condition of salinity. The higher the salinity gradient the larger is the potential generated. This is limited by the membrane characteristics. There exists a monotonic relation between the number of valence electron/ion/unit time and the potential generated up to about 16000 concentration. The membrane characteristics have been studied for optimal ion crossover for various gradients and flow. The graph between ln (gradient) versus Voltage provides insights this process. This provides a very cost effective and clean process of energy conversion.

With the ability to store the energy of sunlight at room temperature and release it later in the form of heat, solar thermal fuel provides an emission free and renewable approach for solar energy conversion and storage. However, finding an efficient and cost-effective material for this approach remains a challenge after over forty years of development. Recently, we have developed a high-throughput approach using ab initio simulations to tackle this materials design problem. Using our approach, we have not only discovered materials as high-energy density STF candidates, but also physical principles for future STF materials design. When building our high-throughput simulation architecture, we have discovered empirical approximations that could allow us to speed up our screening process significantly. We have validated our approximations with higher accuracy simulation methods.

Solid oxide fuel cell performance suffers from a multitude of degradation mechanisms, which include coarsening and oxidation of the anode&’s metallic phase. In this study, we explore the anodes&’ microstructural stability and susceptibility to these processes by considering modifications in key features. A continuum, phase-field model is employed to capture the behavior of the metallic phase and anode structure during evolution. We find considerable differences in evolution behavior across various representative anode configurations and a spectrum of material properties. Multiple methods of analysis, such as interface curvature distributions and cluster contiguity, are used to characterize the system morphology and track variations in performance-critical features. This insight potentially opens avenues for solid oxide fuel cell development and optimization. This work is supported by the Energy Frontier Research Center on Science Based Nano-Structure Design and Synthesis of Heterogeneous Functional Materials for Energy Systems funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences (award DE-SC0001061).

A novel anode with coking-free on-cell reformer is demonstrated to develop the direct methane fueled solid oxide fuel cells (SOFCs) at low temperature. Nickel based anodes for SOFCs have provided the potential for the direct utilization of methane, which is a crucial building block for sustainable energy generation systems. Although many efforts have been devoted to utilize the methane fuel without an external reformer, employment of nickel catalysts give rise to the degradation of the cell via carbon coking as well during either conventional steam reforming or electrochemical oxidation. Herein, we suggest a coking-free on-cell reformer working by the partial oxidation (POX) reaction. Utilization of POX reactions can prevent the carbon coking fundamentally due to the presence of ample oxygen, while the stable operation required the excellent oxygen storage capacity of reforming layer. To elucidate the influence of chemical reforming reactions on stable SOFC operation, undoped ceria was employed to the reforming layer support, on which the electrochemical layer of Ni-gadolia doped ceria (GDC) was deposited by dip-coating methods. The gas conversion rate of the on-cell reformer exhibited the excellent methane utilization by gas chromatography (GC, Acme 6100, Younglin, Korea), which increased as the electrochemical reaction rate was accelerated. Electrochemical impedance spectroscopy (EIS) showed that the methane oxidation through the POX on-cell reformer was activated when it was compared to the electrochemical oxidation of methane. Ex situ analyses including elemental analyses and conductive atomic force microscopy (c-AFM) exhibited that there occured no nickel oxidation and carbon coking during long-term operations, which allows for the successful operation of direct methane fueled SOFCs.

Using material-sensitive and conductive atomic force microscopy (AFM) the nanostructure and conductivity of Nafion^® ionic conducting perfluorinated sulfonated ionomer (PFSA) was investigated at cross sections. At equilibrium conditions, isolated ionically conductive regions were found with a size of several ten nanometers. The apparent low adhesion at these areas was explained by the formation of a local electrochemical double layer, and they were assigned as large water-filled regions that include numerous ionic domains. Under an applied voltage the formation of a connected ionically conductive network in voltage direction was observed, induced by the forced current and concurrent water flow through the sample.
In pristine samples the existence of large crystalline lamellae with spherulitic structure was observed and analyzed, in Nafion^® as well as in Aquivion^® PFSA.
The phase separation of the ionomer was also analyzed using the capacitive current distribution. In Aquivion^® PFSA, larger connected water-rich ionic areas were found than in Nafion^®. The time dependent change of a cross section was investigated to monitor the formation of a new skin layer.
Material-sensitive imaging was also applied to whole slices of MEAs and catalytic layers. Electrodes for fuel cells are complex porous compounds with nanosized platinum catalyst particles supported on conducting mesoporous carbon and PFSA. With material-sensitive and conductive atomic force microscopy the identification of the different components can be achieved, especially in combination with measurements of the local current using a conductive AFM-tip. The detection of the local surface potential across the MEA, biased, under current flow, or in hydrogen atmosphere, was applied to investigate degraded samples.

Thin film nanocrystalline silicon prepared by chemical-vapor deposition is an established material used in multijunction amorphous silicon solar cells. Its potential in low cost and scalable thermoelectric applications depends on the possibility to reduce grain sizes to nanometers while at the same time maintaining its high crystalline to amorphous ratio. In this work, we show that by varying the hydrogen dilution of silane gas flow during deposition, we can reduce average grain sizes to a few nanometers while still maintaining ~90% crystallinity of the material. Annealing at 600°C improves crystalline content with only a small increase of the grain sizes. The values of thermal conductivity, measured from 85 K to room temperature as function of hydrogen dilution ratio from full amorphous to nanocrystalline silicon, remain at a level that is typical for amorphous silicon.
* Work supported by the Office of Naval Research

In recent decades, thermoelectric (TE) materials have been of great interests for potential application as power generation devices. Although TE-based technologies are promising due to being emission free and having a compact size, their applicability for broad commercial deployment is limited by their relatively poor operating efficiency, characterized by the dimensionless ZT factor. Finding good TE for the high-temperature regime, T>900 K is especially challenging. The Si_xGe_(1-x) compound has been considered because single crystals are known to have a good electrical conductivity and can be easily doped p- or n-type. Unfortunately, due to their high thermal conductivity and relatively low Seebeck coefficient, the typical ZT for these compounds is between 0.6 and 1.
There is evidence that many of the boron-rich borides, which are constructed from B₁₂ icosahedra have a good Seebeck coefficient. The borides are thermally stable at high temperatures and have a relatively large band gap, both of which are prerequisite for a good thermoelectric material. Here first-principles methods are applied to examine the AlLiB₁₄ compound as a potential high-temperature TE material from first-principles methods. Boltzmann transport theory predicts that at temperatures near 1000 K the Seebeck coefficient will be greater than 200 mV/K for carrier concentrations around 1'10²⁰ cm^-3. The TE figure of merit is approximated using an elasticity based expression for the thermal conductivity and is found to be around 0.33'10^-3 T.
Recent work has shown that doping this compound with Y results in an increased density of states at the band edge, increasing the thermoelectric response. We examine the role of Y, in the AlYB₁₄ crystal, on the thermoelectric properties and bonding. The transport properties as a function of temperature and carrier concentration are reported.

Translating the revolutionary potential of thermoelectric materials for transformative solid-state refrigeration requires bulk materials with a room-temperature thermoelectric figure of merit ZT_300K >1.5. Recently, we showed that sub-atomic-percent sulfur doping of pnictogen chalcogenides can result in 25-250% increases in ZT_300K.^1,2 Here, we report the remarkable trend of sulfur-doping-induced increase in both the Seebeck coefficient (α) and the electrical conductivity (σ) in nanostructured macro-assemblies of stoichiometric bismuth telluride. We obtained the highest σ = 2.5×10⁴ #8486;^-1m^-1 and α = -230 µVK^-1 at 750 ppm S doping. We also show that trace chorine can confound the effect of sulfur doping even though both dopants are electron donors. In fact, 400 to 1000 ppm Cl results in electron concentrations 1×10¹⁹ le; n le; 1×10²⁰ cm^-3 that are 10-fold higher than that obtained with a similar S doping level without Cl, indicating that Cl is a stronger donor. Extended X-ray absorption fine structure analyses reveal that sulfur substitutes both Bi at 6c and Te at 3a sites while Cl occupies the 3a interstitial sites, providing different mechanisms of altering the electronic band structure, and hence thermoelectric properties. Based upon these results, and first-principles calculations of the electronic structure, we describe the doping-induced α enhancement mechanism and presage a strategy to realize substantial increases in α, and hence ZT_300K, in Bi₂Te₃-based materials through sub-atomic-percent dopants.
References: Mehta, R. J.; Zhang, Y. L.; Karthik, C.; Singh, B.; Siegel, R. W.; Borca-Tasciuc, T.; Ramanath, G. Nature Materials 2012, 11, 233; Mehta, R. J.; Zhang, Y. L.; Zhu, H.; Parker, D. S.; Belley, M.; Singh, D. J.; Ramprasad, R.; Borca-Tasciuc, T.; Ramanath, G. Nano Letters 2012, 12, 4523.
.

Energy transport across heterostructure interfaces is of fundamental interests to many energy-related applications. One example can be thermionic cooling, in which electrons absorb heat to pass a potential barrier at the interface and deposit heat on the other side of the barrier. Within a nanocomposite (e.g. Si nanoparticles dispersed within a Ge host), this interfacial energy barrier can also be engineered to benefit the overall thermoelectric performance of the nanocomposite.¹ In this aspect, electron Monte Carlo (MC) simulations have been carried out to understand the temperature distribution with a dc current passing through an InGaAs-InGaAsP interface.^2,3 Although sufficient detail is given for electron transport, the critical phonon transmission/reflection across the interface (resulting in an interfacial thermal resistance) is not considered and conventional Fourier&’s law is often used to solve for the temperature profile. In this work, such issue is addressed by coupling the electron and phonon MC simulations to provide more accurate temperature predictions. The concept is demonstrated by simulating the local heating and cooling across a Si/Ge interface with a potential barrier formed due to charges trapped on the interface. This provides important guidance for the future improvement of SiGe nanocomposites by engineering the Si/Ge interfaces. It also reaches beyond existing work on coupled electron and phonon MC simulations to study energy transport along a one-dimensional 10-nm Si channel.⁴
References:
1. A. Minnich et al., Energy & Environmental Science 2, 466-479 (2009).
2. A. Stephen et al., Journal of Applied Physics 114, 043717 (2013).
3. M. Zebarjadi et al., Physical Review B 74, 195331 (2006).
4. Y. Kamakura et al., International Conference on Simulation of Semiconductor Processes and Devices, 89-92 (IEEE, 2010).

GaN and its alloys are promising candidates for high temperature thermoelectric (TE) materials due to their high Seebeck coefficient, high thermal and mechanical stability, and large band-gap. Moreover, these materials can overcome the toxicity concern of current Te-based TE materials, such as Bi₂Te₃ and PbTe. These materials have recently shown a higher Seebeck coefficient than that of SiGe in high temperature region because their large bandgap characteristic eliminates the bipolar conduction. In this study, we report the room temperature thermoelectric properties of Mg doped GaN, grown by metalorganic chemical vapor deposition (MOCVD) on sapphire substrate with various carrier concentrations. Undoped and n-type GaN are also incorporated with p-type GaN films to make comparison. The structural, optical, electrical, and thermal properties of the samples were examined by X-ray diffraction, photoluminescence, van der Pauw hall-effect, and thermal gradient methods, respectively. The minimum Seebeck coefficient of 710µV/K at room temperature of Mg: GaN were observed, which further indicated their potential TE applications.

Thermoelectric (TE) generators have very important applications, such as emerging automotive waste heat recovery and cooling applications. However, reliable transport properties characterization techniques are needed in order to scale-up module production and thermoelectric generator design. DOE round-robin testing found that literature values for figure of merit (ZT) are sometimes not reproducible in part for the lack of standardization of transport properties measurements. In Sandia National Laboratories, we have been optimizing transport properties measurements techniques of TE materials and modules. We have been using commercial and custom-built instruments to analyze the performance of TE materials and modules. We developed a reliable procedure to measure thermal conductivity, seebeck coefficient and resistivity of TE materials to calculate the ZT as function of temperature. We use NIST standards to validate our procedures and measure multiple samples of each specific material to establish consistency. Using these developed thermoelectric capabilities, we studied transport properties of Bi₂Te₃ based alloys thermal aged up to 2 years. Parallel with analytical and microscopy studies, we correlated transport properties changes with chemical/structural changes. Also, we have developed a resistance mapping setup to measure the contact resistance of Au contacts on TE materials and TE modules as a whole in a non-destructive way. Recently, a major effort has been conducted to investigate chemical/structural stability and output power of TE materials under representative conditions (temperature gradient and current flow). The development of novel but reliable characterization techniques has been fundamental to better understand TE materials as function of aging time, temperature and environmental conditions. All this information is crucial for the development of long-term reliable and stable TE modules for multiple applications.
(SAND2014-19035 A) Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy&’s National Nuclear Security Administration under contract DE-AC04-94AL85000.

N-type Mg2Si and p-type HMS (high manganese silicide) are two of the most promising candidates for thermoelectric power generators (TEG) that operate with high temperature heat sources at around 700K to 900K. These materials have several promising features, such as the abundance of their constituent elements, their non-toxicity, and the fact that they are light weight and have the capability of being used in TEGs.
The combination of these materials is expected to provide exceptional performance, and they are suitable for the conventional P-structure TE module. To realize P-structure TE modules with these materials, the thermally induced internal stresses during fabrication and operation should be considered in order to obtain the expected performance and reliability of the devices. Furthermore, the modules will be exposed to severe mechanical loading especially while operating in automotive applications. Because of these mechanical issues, the mechanical properties of these materials need to be well understood in order to properly design the module structure. For this purpose, the mechanical properties of these thermoelectric materials are examined in this paper.
Specimens of each of these silicides were prepared by powder processing and then sintered by a spark plasma sintering method. The fundamental mechanical properties such as Young&’s modulus, bending strength, and fracture toughness were measured using various testing methods.
In addition, for the aiming toward TE module fabrication and appropriate system integration techniques for automotive applications, We evaluated on both n-type Mg₂Si and the p-type HMS material of mechanical properties not only at room temperature but also after the heat treatment at high temperature around 873K.
Details of the other mechanical properties measured are given for both n-type Mg₂Si and the p-type HMS material.

As a lead-free material, GeTe has drawn a growing attention in thermoelectrics, and the figure of merit ZT close to unity was previously obtained via traditional doping/alloying, largely owing to the hole carrier concentration tuning. In this contribution, we show that a remarkably high ZT of ~1.9 can be achieved at 773K in Ge_0.87Pb_0.13Te upon the introduction of 3mol% Bi₂Te₃. Bismuth Telluride promotes the solubility of PbTe in the GeTe matrix thus leading to a significantly reduced thermal conductivity. At the same time, it enhances the thermopower by activating a much higher fraction of charge transport from the highly degenerate Σ valence band, as evidenced by Density Functional Theory (DFT) calculations. These mechanisms are incorporated and discussed in a 3-band (L+Σ+C) model, and found to well explain the experimental results. The detailed microstructure (including rhombohedral twin structures) analysis in Ge_0.87Pb_0.13Te+3mol%Bi₂Te₃ is carried out using transmission electron microscopy (TEM) and crystallographic group theory. The complex microstructure explains the reduced lattice thermal conductivity, and electrical conductivity as well. These findings create new insights that pave the way for further improvements of the thermoelectric performance of GeTe-based as well other types of thermoelectric materials.

The efficiency of a thermoelectric device depends on its material&’s figure-of-merit (ZT) and the practical realization of an efficient thermoelectric device is currently limited by its low ZT. Bulk Cu₂Se is known to exhibit a high ZT ~ 1.5 owing to its liquid-like behaviour. In the present work, we report an enhancement of ZT in Cu₂Se by employing nanostructuring approach. A ZT ~ 2.1 at 973 K was realized in nanostructured Cu₂Se, which was synthesized employing spark plasma sintering of mechanically alloyed elemental nanopowders. Significant enhancement in thermoelectric performance is mainly due to the drastic reduction in the lattice thermal conductivity which has been attributed to the enhanced phonon scattering by nanoscale defects as well as dense nanograin boundaries. Further, high resolution transmission electron microscopy of nanostructured Cu₂Se samples was carried out to study the nanograin size morphology and to correlate it with the observed reduction in the lattice thermal conductivity.

Thermoelectric materials research has been focused on the intrinsic and artificial low-dimensional structures. Here we demonstrate the low dimensionality increases power factor in CeTe_2-xSb_x single crystals. The band structures of CeTe₂ show the 2-dimensional (2D) Fermi surface nesting behavior as well as a 3-dimensional (3D) electron Fermi surface hindering the perfect charge density wave (CDW) gap opening. By hole doping with the substitution of Sb at the Te-site, the 3D-like Fermi surface disappears and the 2D perfect CDW gap opening maximizes the power factor up to x = 0.1. The thermoelectric properties of Sn-doped polycrystalline compounds of CeSe_2-xSn_x (x = 0.0, 0.1, and 0.2) have shown that the temperature-dependent electrical resistivity ρ(T) exhibits abnormal increase of ρ(T) with decreasing temperature. From the fitting of the hopping mechanism, the non-semiconducting behavior of ρ(T) may be associated with small polaron hopping. The significant decrease of Seebeck coefficient S(T) at low temperature corresponds to the freeze-out temperature of Umklapp scattering. By comparing with S(T) and κ(T), the decreases of Seebeck coefficients are caused by freeze-out of the electron-phonon Umklapp scattering process. We argue that the strong electron-phonon coupling decreases thermal conductivity by phonon softening.

Solid-state thermoelectric (TE) devices have the ability to directly convert heat into electricity for power generation. However, the potential impact of TE technology for waste heat recovery is now hindered by the heavy usage of toxic, rare, and expensive (e.g. Te) elements. Furthermore, ultra-high-temperature (>1200 K) TE materials are still lacking due to the poor thermal stability and degraded TE performance of existing materials. For large-scale applications, it is thus of urgent need to broadly search for cost-effective TE materials for high-temperature energy harvesting. In physics, the performance of TE materials is evaluated by their TE figure of merit (ZT), defined as ZT = S²σT/k, where S, σ, k, and T represent Seebeck coefficient, electrical conductivity, thermal conductivity, and absolute temperature, respectively. In previous computation-drive material search, challenges exist in accurately and effectively predicting the mean free paths (MFPs) of electrons and phonons for transport property predictions. However, this need can now be eliminated when phonon MFPs are largely restricted by nanostructures embedded within a bulk material and the nanostructure sizes further decrease to electron MFPs. This concept is aligned with the current trend of using nanostructured bulk materials to enhance ZTs of general materials. Material search based on electrical properties with nanograin-restricted electron MFPs is shown in Ref. 1. In this work, the overall ZT is further predicted with this new strategy to evaluate the TE performance of various nanostructured oxide compounds and alloys. Other opportunities, such as electronic band structure engineering with alloy compositions, will also be discussed as part of the high-throughput material search.
References:
1. Wang, S., Wang, Z., Setyawan, W., Mingo, N. & Curtarolo, S. Physical Review X1, 021012 (2011).

Maintaining sustainable society, it is necessary to develop the renewable energies with dispersion type power supply system. Thermoelectric (TE) power generation from waste heat is attracted as co-generation system combined with heat and electrical power. TE property of Na_xCoO₂ (NCO, p-type) gives many impacts to develop oxide TE materials. In oxide material, low thermal conductivity (κ) can be found in comparison with metallic material, and high stability is also kept in high temperature. Nowadays, particularly, n-type oxide material is also promised to be investigated for high dimension less figure of merit (ZT=S²σT/κ, where S is Seebeck coefficient, and σ is electrical conductivity). Although partly reduced Na layer involves in NCO, κ is possible to be reduced with conductive CoO₂ layered crystal structure. Nb-doped SrTiO₃ (STO, n-type) also show prominent TE properties, on the other hand κ of STO is slightly high value with cubic structure. In addition to NCO, Funahashi et al. shows typical S and σ in Bi₂Sr₂Co₂O_δ whisker with low κ from 1D structure. Non-stoichiometric tungsten bronze type oxide, such as Sr_1-xBa_xNb₂O₆ (SBN) and Ba₂NaNb₅O₁₅ (BNN, preference orientation: 88.2%) are possible to adopt for thermoelectric oxide material.
In experimental, BNN was prepared by BaCO₃, NaCl and Nb₂O₅ powders at 1100-1300^oC for 12h by solid state reaction. Needle-like (Ba, Na)Nb₂O₆ (N-BNN) particles were synthesized at 900-1100^oC for 4h in closed crucible. Mixture of BNN and N-BNN was made by tape casting for in-plane orientation, and was sintered at 1340^oC for 12h. The oriented BNN was partly reduced at 1050^oC-1100^oC for 2-5h in CO atmosphere. Temperature dependence of S and that of σ for reduced BNN were measured from 20^oC to 300^oC by conventional two probe method and four-terminal method, respectively. The curve fitting was carried out for S(T) data with Gasumyants eq. The ab-initio band calculation and transport calculation were carried out using Wien2K and BoltzTrap software, respectively.
In presentation, we will report about band edges, band filling degree and total effective band width estimated from band calculation, transport calculation, curve fitting for S(T) data with Gasumyants eq., and κ of non-stoichiometric BNN utilizing Harman method for in-situZT and κ comparison with results of transport calculation.

Thermoelectric (TE) materials with high ZT is necessary for practical applications of thermoelectric modulus not only power generation, but also cooling applications. Their thermoelectric properties such as Seebeck coefficient, electrical conductivity, resistivity and thermal conductivity measurements are unavoidable for evaluation of ZT. For this reason reliable, accurate, and consistent thermoelectric measurements are significant characterizations. Indeed, the measurement time for these parameters is highly significant due to the temperature effect on the materials during the measurement at high-temperature. There are two primary techniques used to measure the relative Seebeck coefficient (S): the integral and the differential methods. The differential method, which comprises the majority of high temperature S characterization: steady-state (dc), quasi-steady-state (qdc), and transient (ac) assuming the observation time scale are the best ways to measure the S. We developed measurement modes for fast measurement of S: “continues-step and oscillation”. We compare these different modes of measurements with others to measure and calculate the S, evaluate the accuracy of the results, and show modes to enhance the accuracy. Using these modes, S can be measured only in 2 h. Temperature gradient is created from 5 to 10 K according to the temperature set. Due to the short oscillation frequency, data can be obtained at each minute that corresponds to 10 K. With this advantage, the time effect on S data can also be investigated. Data for S is obtained at each temperature gradient occur during heating and cooling cycles. In oscillation mode, when the specific equilibrium points that occur in the range that of <0.06 K, the system does also the electrical resistivity (R) measurement during heating and cooling cycles. Because of the Peltier effect and unwanted thermoelectric voltages coming from wiring, direct current is applied up to 100 ms along with current reversal and axial four-point technique is applied for R measurements. We furthermore point out experimental and data analysis parameters to evaluate the reliability of the obtained result. The shown analysis can be used to find and minimize errors in the S and R measurements and therefore increase the reliability of the measured material properties. The measurement uncertainty is estimated for S and R measurement is less than 3 and 1 %. Niobium rod and platinum wire purchased from GoodFellow are used to ensure the accuracy and reliability of the system. As a result, these measurement modes have unique properties that shorten the measurement time, facilitate the data analysis and enhance the accuracy.

There is a huge desire to increase operation speeds in modern integrated circuits as they get smaller and more compact. Heat generation in such a submicron devices is a key factor limiting their performances. As a solution, thermoelectric cooling in heterostructures can address heat dissipation issue in submicron devices. Performance of single barrier heterostructures depends strongly on several parameters including barrier height, barrier width and thermal conductivity of barrier. Superlattice structures have been known to have the lowest thermal conductivities reported for crystalline materials. Low thermal conductivity is beneficial for thermoelectric cooling as it reduces the heat flow from hot end to cold junction. Moreover the band offset between the barrier and base material can be easily tuned by changing the superlattice period. By optimizing the conduction band offset (barrier height), it is possible to control the Joule heating and also optimize the amount of heat absorbed due to Peltier cooling. We investigate the feasibility of using PbSe/PbSnSe superlattice in heterostructures using Monte Carlo simulations. The effect of different parameters such as barrier height, barrier width and superlattice thermal conductivity on thermoelectric cooling of such structures will be presented.

Thermoelectric (TE) materials have gained increasing interest by the scientific community, due to their capability of harvesting the waste heat into useful electrical power. Efficiency of TE materials is known as dimensionless figure of merit (ZT). To date, efficiency of TE materials and design of the devices are the major challenges for the researchers to enhance the overall performance of the TE systems. Silicide based materials are known to be the best candidates for TE systems due to their nontoxicity, high earth abundance and low production cost. Furthermore, they have exhibited promising TE properties including high electrical conductivity and low thermal conductivity, which improved the overall performance, ZT. Nanostructured magnesium silicide (Mg₂Si) as n-type and higher manganese silicide (HMS), represented by MnSi_x (x=1.71-1.75), are promising p-type materials for TE energy harvesting systems in the middle-high temperature range (600-900 K).[1] Both type of materials were prepared by ball milling followed by sintering using spark plasma sintering (SPS) technique. SPS process parameters like temperature, heating rate, holding time and applied pressure were investigated for fine tuning of the microstructure and the crystal phase purity. All compacted samples have shown the compaction density up to 99%. X-ray diffraction (XRD) analysis was utilized to investigate the crystalline phase of compacted samples; Scanning Electron Microscopy (SEM) coupled with Energy Dispersive Spectroscopy (EDS) was used to evaluate the detailed microstructural and chemical composition respectively. Temperature dependent TE characteristics of n- and p type silicides were measured over temperature range of 300 -900 K.

The high demand for converting natural and waste heat into electrical energy makes thermoelectric generators very attractive energy resources. As compared to inorganic thermoelectric materials, organic thermoelectric materials, which are flexible, ecofriendly, highly elastic, and cost-effective, are promising candidates for thermoelectric generators. However, low electrical conductivity and power factor of organic thermoelectric materials have been obstacles for realization of practical device applications. In this study, we propose a new strategy for enhancing the electrical conductivity and power factor of organic thermoelectric materials by fabricating multilayer structure using two different conducting polymers- Poly (3,4-ethylendioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) and polyaniline doped with camphorsulfonic acid (PANI-CSA). In the PEDOT:PSS/PANI-CSA multilayer structure, we found that the PANI-CSA helps stacking of PEDOT:PSS multilayers and enhances the conductivity as well as power factor in comparison with PEDOT:PSS thin films.
At first, PANI-CSA film was deposited on the glass substrate followed by annealing at 50°C for 1 hr in N₂ ambience. PEDOT:PSS film was then deposited on top of the PANI-CSA film and then baked at 120°C for 15 min in N₂ atmosphere. The sample was immersed in DMSO for 20 min at 60°C in air. The above process was repeated from 1 to 5 times for fabricating multilayer structures, which are labeled as n(PEDOT:PSS/PANI) with a repetition number n. Thickness of both PANI-CSA and PEDOT:PSS component layer is around 20 nm, which was confirmed using transmission electron microscopy. We found that the electrical conductivity and the power factor were increased from 1.73×10³ S/cm and 2.18 mWm^-1K^-2 for 1(PEDOT:PSS/PANI) layer to 4.44×10³ S/cm and 21.59 mWm^-1K^-2, for 5(PEDOT:PSS/PANI) layers, respectively. Our study suggests that fabrication of multilayers with two different conducting polymers is a promising technique to enhance the thermoelectric properties of organic materials.

The growing market of portable/wearable electronic devices has promoted the increasing demand for high-performance flexible and renewable energy sources. Thermoelectric technology is one of most exciting renewable and clean energy techniques, which has the unique ability to generate electricity directly from untapped waste heat from various sources such as computers, automobiles or even human bodies. Therefore, the pursuit of high-performance flexible thermoelectric generators has become one the most promising strategies to produce portable/wearable renewable energy sources. In this work, we report a novel highly-flexible and ultrathin thermoelectric fabric based on self-assembled tellurium nanorod/PVDF composites. The 10 mu;m-thick free-standing thermoelectric fabrics are composed of a 1 mu;m layer of self-assembled tellurium nanorod networks and 9 mu;m layer of PVDF, which show a high room temperature Seebeck coefficient of 288 µV/K and electrical conductivity of 3750 S/m. Physical characteristics of the thermoelectric fabrics are favorable due to their excellent flexibility and durability. The fabrication process could be easily scaled up to industry-production level. This work opens a new avenue to fabricate highly-flexible and lightweight sustainable energy sources that could be compatible with portable/wearable electronic devices.

Thermoelectric generators (TEG) consist of many thermocouples (pair of n- and p-type materials) which are switched electrically in series and thermally in parallel. During operation the materials are exposed to huge temperature gradients of sometimes several hundred Kelvin per centimetre and an electrical field. These factors can cause thermal degradation (Ludwig-Soret effect) and electromigration. Hitherto, influences on the thermoelectric properties have rarely been investigated, although module properties can significantly deteriorate, especially on a long-term scale. The applicability of new materials under specific thermal and electrical conditions can only be predicted knowing possible degradation effects. Therefore, a better understanding of the occurrent transport processes is essential for ensuring a successful development of future TEGs.
In the ongoing research project we intend to investigate segregation effects in thermoelectric materials caused by temperature gradients. Sodium cobaltate (Na_xCoO₂) as a good sodium ion conductor is indicating a high tendency to segregation. Highly oriented thin films of Na_xCoO₂ are produced via pulsed laser deposition (PLD) and characterized using different characterization methods (XRD, TEM, etc.). By applying temperature gradients parallel to the layered structure of the samples, enrichment as well as depletion of sodium ions along the gradient can be observed.
Identifying materials whose thermoelectric properties diminish according to segregation effects as well as materials which show long-term stability against temperature gradients and electric fields is crucial for a purposeful development and improvement of promising systems.

CaMnO₃ is a n-type thermoelectric material used for power generation from waste energy. Efficiency of thermoelectric materials is governed by ZT, dimensionless figure of merit. ZT = α²σT/κ,where α= seebeck coefficient, σ = electrical conductivity, κ= thermal conductivity & T= temperature. ZT>1 is needed for practical applications. Recently Zhao et al, reported ZT = 2.6 for SnSe single crystal at 923K along b-axis. But for oxide thermoelectric materials figure of merit is 1.2 . Oxide materials are non toxic and stable at high temperature and hence advantageous over non oxide materials.
CaMnO₃ adopts orthorhombic crystal structure with space group P n m a and shows high seebeck coefficient of -651mu;V/K, electrical resistivity 0.303Omega;cm, 0.01 ZT at 300K [1].It has been shown that the figure of merit ZT of CaMnO₃ could be improved by chemical doping in either one of Ca and Mn sites or both of these sites. Some of the typical dopants in the Ca-site are Yb,Ho,Tb,Bi,Y,Gd,Nd and Dy and in Mn-site Ru and Nb are doped [2]. Terbium doping in Ca-site has been shown increase in electrical conductivity and decrease in thermal conductivity but decrease seebeck coefficient. Similar behavior has seen when Niobium doped in Mn-site. In this work, we have carried out co-doping Tb in the Ca site and Nb in the Mn site of the CaMnO₃ compound.
We synthesized CaMnO₃, Ca_0.9Tb_0.1Mn_0.98Nb_0.02 by solid state method. Single phase compounds are confirmed by X-ray powder diffraction studies.CaMnO₃ is a perovskite structure and P n m a space group with lattice parameter a asymp;b asymp;a_pradic;2 & c asymp; 2a_p.Lattice parameters of these compounds were refined by FullProf software. The morphological structure studied by scanning electron microscopy. Electrical resistivity and seebeck coefficient of bar shaped samples were measured by ULVAC ZEM-3 .Resistivity shows 0.89 #8486; at 300K for parent compound and co-doped sample shows reduction in resistivity value. Seebeck values show electron as a major carrier of -76 µV/K and increases to -173 µV/K at 300K . Thermal conductivity of pellets measured by Anter corporation ltd. From the above studies power factor, figure of merit calculated and will be discussed in the presentation.
References:
1. Bocher et al., Inorganic Chem 47.,8077-8085(2008)
2. G.Xu et al. Solid state Ionics171, 147-151 (2004).

Solid state thermoelectric generators are clean and reliable sources of energy harvesting systems due to their ability to generate electricity directly from waste heat. Zintl compounds e.g., Yb₁₄AlSb₁₁, Ca₅Al₂Sb₆ or YbZn₂Sb₂, etc., have emerged as a promising class of thermoelectric materials. They are typically small band gap semiconductors possessing complex crystal structures. Among bulk materials, Zintl compounds exhibit some of the lowest reported lattice thermal conductivities. Recently, Yb₉Mn_4.2Sb₉ was also reported to show extremely low lattice thermal conductivity due to its complex and highly disordered crystal structure. Relative to Yb₉Mn_4.2Sb₉, both Yb₉Zn_4+xSb₉ and the Yb₉Mn_4.2-xZn_xSb₉ solid solution display an increase in the carrier mobility mainly due to a reduction of spin dependent scattering. However, the majority carrier concentration of Yb₉Zn_4.2Sb₉ is still too high to achieve the highest thermoelectric figure of merit predicted by a simple effective band mass model, it has higher quality factor than that of Yb₉Mn_4.2Sb₉. In this study the carrier concentrations were tuned in A₉Zn₄Sb₉ (A = Ca, Yb, Eu) systems by several different methods to achieve higher thermoelectric efficiency.