Models are developed for the transport of Li ions in the electrolyte of lithium ion batteries, their diffusion through storage electrode particles, and their kinetics through the surface of the particles between the electrolyte and the particles. As a consequence of the Li ion intercalating in the storage particles, their lattice swells, leading to elastic stress when the concentration of Li ions in the particles is not uniform. The models of transport are based on standard concepts for multi-component diffusion in liquids and solids, but are not restricted to dilute solutions, or to small changes in the concentration of the diffusing species. In addition, phase changes are permitted during mass transport as the concentration of lithium varies from the almost depleted state of the storage particle to one where the material is saturated with its ions. The elastic swelling and shrinkage may involve very large dilatations, which can be allowed for in the formulation of the model. Thus, the models can be suitable for storage particle, where the amount of Li can vary by large amounts depending on the state of charge, for staging as observed in the storage process in graphite, for the enormous swelling that takes place when silicon is used for storage, and for electrolytes in which the concentration of Li ions is high. The model is used to compute the processes of charging and discharging the battery to assess the parameters that influence the development of stress in the storage particles, and to deduce the likelihood of fracture of the storage particle material. The objective is to assess designs of porous electrode microstructures that permit rapid charging and discharging, but obviate the likelihood of fracture and other mechanical damage that limit the performance and reliability of the battery.

This presentation will discuss applications of the phase-field method to microstructural processes during Li-plating and Li-insertion/intercalation into or extraction from electrodes in Li-ion batteries. The focus will be on Li_xFePO₄, one of the most-studied cathode materials in Li-ion batteries. The thermodynamics of the FePO₄-LiFePO₄ two-phase system and the effect of coherent stress on the miscibility gap and two-phase morphology will be discussed. A three-dimensional phase field model for modeling the morphological evolution during the intercalation/extraction of Li-ions into a host electrode will be described. It incorporates the effects of anisotropic diffusional mobility of Li-ions in the electrode host lattice, flux of Li-ions across the electrode/electrolyte interface, and coherency strains arising from the lattice parameter mismatch between the lithiated and unlithiated phases. Implementation of spectral methods to solving the systems of equations under non-periodic boundary conditions will be presented. The microstructural features obtained from the simulations are compared with available experimental observations.

A PNP-type model for the transport of Li-ion resulting from diffusion and electro-migration is studied. The re-distribution of ion results in an evolving eigen-strain that further induces a displacement over the sample. The displacement on the top surface is proposed to be linearly scaled to the ion concentration, which can be measured by electrochemical strain microscopy (ESM). A bias of DC and AC using scanning probe microscopy can be implemented to probe the local ion concentration (n) and the diffusion coefficient (D) representing the thermodynamic and kinetic characteristics of Li-ion. We illustrate a proposed method to measure ion concentration and diffusivity by means of an ESM experiment and mathematical argument with simulation.

The effect of reversible electrochemical reaction on Li-ion diffusion and stress in a cylindrical Li-ion battery is studied. The volumetric change created by the diffusion of Li-ion and formation of reversible reaction product would generate the diffusion-reaction-induced stress in the electrode. The general relation among concentration of Li-ion, reversible reaction product, and mechanical stress is derived, and the numerical solutions of the concentration, stress and reaction product fields are obtained. The forward reaction with a high forward reaction rate can retard Li-ion diffusion and increase the compressive stress at the surface of the active material for potentiostatic charging, which may lead to premature failure of electrode structure; but reduce the compressive stress in the active material for galvanostatic charging. While the backward reaction has little effect on the distribution of Li-ion and stress. As charging goes on Li-ion diffusion and electrochemical reaction with extremely high forward reaction rate would enter an equilibrium state that the distributions of Li-ion concentration and corresponding stress are stable. The higher the forward reaction rate is, the sooner the equilibrium state comes. The effect of reversible electrochemical reaction on Li-ion diffusion and stress can be neglected if the forward reaction rate is relatively low.

The mechanical degradation of electrodes caused by lithiation and delithiation is one of the main factors responsible for the short cycle life of lithium-based batteries employing high capacity electrodes. Recent experiments have revealed a number of interesting phenomena, such as size-dependent delamination of patterned silicon thin film electrodes from a current collector during lithiation and delithiation cycling, which cannot be satisfactorily explained by existing theories in the literature. Here we discuss a combined series of recent theoretical, computational and experimental studies aimed to clarify the mechanisms of fracture, delamination and ratcheting in thin film Si electrodes on substrates during lithiation/delithiation cycles. We show that the observed delamination size effect can be rationalized by modeling thin film delamination in the presence of large scale interfacial sliding. A method is proposed to deduce the critical size for delamination based on the critical conditions for the nucleation and growth of edge or center cracks at the film-substrate interface under plane strain or axisymmetric conditions. We derive the critical film thickness for fracture as a function of both the fracture toughness of the film and the sliding resistance of the interface. Our analysis indicates that a slippery interface due to lithiation could significantly decrease the critical thickness for fracture. It is further shown that ratcheting can occur as soon as one allows the yield stress of Si and/or the friction strength of the interface to vary from lithiation to delithiation half-cycles, and that this important failure mode can be avoided by reducing the lateral size of the islands below a critical length scale.

Due to the emergence of effects such as the growth of the so-called solid electrolyte interface (SEI), the loss of contact of particles to conductive pathways and the complete disintegration of the electrode, fracture of active electrode particles is a crucial mechanism leading to capacity fade and power loss in commercial lithium ion batteries. The appearance of fracture and cracks in active particles is commonly ascribed to mechanical stresses evolving from heterogeneous swelling and shrinkage of the material when lithium is inserted or extracted.
In our work, we approach the problem of fracture in active particles by combining a coupled model for mechanical stressing and transport of lithium ions with a phase field description of an evolving crack. While the mechanics of the particle is described by a linear, elastic, isotropic constitutive law accounting for swelling and shrinkage, the diffusion equation for lithium transport is derived from fundamental thermodynamics and is valid beyond the dilute concentration limit. Furthermore, the generation of mechanical stress due to heterogeneous lithium concentration is not only simulated, but is also taken into account in the driving force for lithium diffusion. To determine the kinetics of crack evolution, the phase field model is coupled to the equations for stress by a Griffith type energy minimization principle, where the relevant energy is compromised of the surface or fracture energy associated with any crack surface that has been created and the elastic potential energy due to the presence of stress.
This novel approach allows us to simultaneously study the evolution of the lithium concentration together with the initiation and growth of a crack in an arbitrary geometry, in two and three dimensions, and without presuming a specific crack path. This also enables us to compare standard two-dimensional assumptions, such as plane strain or stress, to the fully three-dimensional situation. We investigate how the formation of cracks depends on the geometric features of the particle and the electrochemical boundary conditions applied at the particle surface, e.g. the lithium flux through the surface.

We quantified dendrite growth in an optically accessible Li-metal coin cell battery charged under applied temperature gradients normal to the electrodes. We found that average dendrite lengths decrease by ca. 30% upon increasing T-gradients 3.5 times.

Lithium-ion secondary batteries are commonly used to power many consumer devices such as handheld phones, laptops, portable music players, and even electric vehicles. One of the key properties that influence the performance of lithium-ion batteries is the ionic conductivity of the electrolyte (i.e., the movement of Li ions from one electrode to another). This is dependent on the speed at which Li ions diffuse across the cell and related to the solvation structure of the Li ions. The choice of the electrolyte can greatly impact both solvation and diffusivity of Li ions. In this work, we use first principles molecular dynamics to examine the solvation and diffusion of lithium ions in several bulk organic electrolytes. We find that differences in the local environment throughout the liquid can lead to solvation of Li ions by either carbonyl or ether oxygen atoms. In addition, we examine the differences in solvation of associated and dissociated Li(PF₆) salts, showing that the bulky PF₆ group blocks complete solvation of Li⁺ by solvent oxygen atoms. Finally, we calculate Li diffusion coefficients in each electrolyte, finding slightly larger diffusivities in a linear carbonate such as ethyl methyl carbonate (EMC) compared to a cyclic carbonate like ethylene carbonate (EC). Results from this work can be used to design new bulk electrolytes that will improve the performance of current Li-ion batteries.

The proliferation of portable electronics, hybrid electric vehicles (HEVs), and large scale energy storage has drawn a lot of attentions toward the development of new generation-Lithium ion batteries (LIBs) as the most prevailing power sources. Among all potential anode material for LIBs, Silicon (Si), being able to host a large amount of Lithium (Li)-each silicon atom can host up to 4.4 lithium atoms, is the most promising candidate. It is well-known that migration of sharp interface between the crystalline silicon and amorphous LixSi, called reaction front, controls the kinetics of lithiation. Recently, self-limiting lithiation of Si-nanowires (SiNWs) anodes has been observed using in situ transmission electron microscopy (TEM) [1]. However, the kinetics of Li interface reaction and the evolution of diffusion-induced stress resulting in the retardation of lithiation process remain unclear. In this study, systematic Molecular Dynamics (MD) simulation has been performed using ReaxFF reactive potential to investigate the atomistic mechanisms governing the retardation effect of the lithiation-induced stress upon lithiation. Moreover, the ledge flow process producing the amorphous LixSi through layer-by-layer peeling of the {111} Silicon facets has been modeled throughout this simulation. The simulation provides a comprehensive picture about the orientation-dependent mobility of the sharp-interface, and the underlying physics of the self-limiting dynamic lithiation in Si-NW anodes.
[1] Liu, Xiao Hua, et al. "Self-Limiting Lithiation in Silicon Nanowires." ACS nano 7.2 (2013): 1495-1503.

In the expanding world of small scale energy harvesting, the ability to combine thermal and mechanical harvesting is growing ever more important. We have demonstrated the feasibility of using ZnO nanowires grown using chemical vapor deposition to harvest both mechanical and low-quality thermal energy in simple, scalable devices. Nanowires were transferred to a flexible Kapton thin film by dry transfer. This method yielded well-aligned nanowires on the receiving substrate. Following dry-transfer, the Kapton film was evaporated with patterned gold electrodes using microfabrication techniques. The circuit was designed as long, parallel electrode arrays perpendicular to the nanowire axis. While the ZnO nanowires were of good quality, the internal resistance of the wires necessitated measuring the open circuit voltage (OCV) rather than a direct output power. Mechanical harvesting was demonstrated using a periodic application of force, modeling the motion of the human body. Tapping the device from the top of with a wood stick, for example yielded an OCV of 0.2 - 4 V, which is in an ideal range for device applications. To demonstrate thermal harvesting from low quality heat sources, commercially available Nitinol (Ni-Ti alloy) thin film, a phase transition material with transition temperature of ~ 50 C, was attached to the nanowire piezoelectric device. The whole device assembly was bent at room temperature. Upon heating above 50 C, Nitinol slip restored to its original flat shape, which yielded an output voltage/power of nearly 1 V.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract No. DE-AC52-07NA27344. LLNL-ABS-645293

This work studies the influence of temperature on the output performance of triboelectric generators (TEGs). PTFE film and aluminum foil are used as the contact materials for the TEG. A high temperature system and a low temperature system are used to conduct measurements of TEG output voltage and current from 300 K to 500 K and from 77 K to 300 K respectively. Dependence of output performance on temperature is subsequently obtained by statistically analyzing the data at each temperature point. Several LEDs connected in series are successfully lit up by TEG as the sole power source at temperatures as low as 77 K and as high as 500 K. The results of this study indicates that the output performance of TEG tends to degrade with increasing temperature and confirms the applicability of TEG from 77 K to 500 K, spanning a range of 423 K.

The fundamental science and applicable technology for harvesting environmental energy are not only essential in realizing the self-powered systems, but also tremendously helpful in meeting the rapid-growing world-wide energy consumptions. Mechanical energy is one of the most universally-existing, diversely-presenting, but usually-wasted energies in the natural environment, which has attracted a lot of research efforts in developing the harvesting technologies.
The triboelectric effect is a universal phenomenon that can generate electrostatic charges from mechanical contact. Here, a new type of technology—triboelectric nanogenerators (TENGs)—has been developed to efficiently convert mechanical energy into electricity, based on the coupling of triboelectrification and electrostatic induction. In order to develop TENGs of different structural designs for harvesting various types of mechanical energies existing in the natural environment, we established three basic modes/mechanisms of TENGs that serve as the basis for most of the TENG structures: (1) vertical contact-separation mode, in which two triboelectric-charged layers (with an electrode deposited at the back) periodically separate in vertical-to-plane direction [1]; (2) lateral sliding mode, in which the charge separation is achieved along the in-plane direction through the relative sliding of two triboelectric layers (with an electrode deposited at the back) [2]; (3) freestanding-triboelectric-layer based mode, which has two stationary electrodes and one freestanding triboelectric layer that moves in between under the guidance of external mechanical energy [3]. For every mode, we made both theoretical studies and experimental demonstrations to illustrate the basic working principles and the output characteristics. Also, the key factors and parameters determining the performance were identified for each case, which provides the importance guidance for the future design and optimization of device structures. Moreover, the utilized triboelectric materials and surfaces were purposely modified for the improvement of the static charge density. All of the three modes of TENGs are shown to be effective electrical sources that are capable of generating a voltage over hundreds of volts and a power density larger than 10 W/m2, and also instantaneously driving hundreds of electronic devices (such as LEDs). The generated electricity can also be stored in the energy storage units such as Li-ion batteries and capacitors, for the purpose of driving personal electronics, like cell phones. The establishment of the three basic modes opens the path of the research on triboelectric nanogenerators for applications in self-powered systems for personal electronics, environmental monitoring and even large-scale power.
[1] Wang, S. H.; Lin, L.; Wang, Z. L. Nano Lett. 2012, 12, 6339-6346.
[2] Wang, S. H.; Wang, Z. L. et al. Nano Lett. 2013, 13, 2226-2233.
[3] Wang, S. H.; Wang, Z. L. et al. Adv. Mater. Under review.

Energy harvesting from ubiquitous ambient vibrations has enormous potential for small power applications including but not limited to wireless sensors, portable electronics, and medical implants. Piezoelectric energy harvesting devices constitute the simplest means of scavenging power directly from ambient vibrations through the direct conversion of mechanical energy into electrical energy. In particular, nano-piezoelectric energy harvesting devices sensitive to small vibrations can be incorporated into small-scale devices, which is attractive in light of the increasing demand for flexible, wearable and implantable electronics.
Ceramics such as lead zirconium titanate and semiconductors such as zinc oxide are currently the most widely used piezoelectric energy harvesting materials. However, our work focuses on a different class of piezoelectric material, namely ferroelectric polymers, such as poly-vinylidene fluoride (PVDF) and its co-polymers. These are potentially superior energy harvesting materials due to their flexibility, lightweight, ease of fabrication and potentially low cost, as well as being lead-free and bio-compatible. Piezoelectric nanowires of polyvinylidenefluoride-co-trifluoro-ethylene [P(VDF-TrFE)] having diameter 200 nm and length 60 µm were grown within nanoporous anodized aluminium oxide (AAO) membranes using a simple, scalable, low-cost template wetting technique. Detailed characterization using scanning electron microscopy, differential scanning calorimetry, Fourier transform infrared spectroscopy and X-ray diffraction has shown that the nanowires grown are in the ferroelectric phase, without the need of poling. Energy harvesting devices comprising nanoporous AAO membranes filled with aligned P(VDF-TrFE) nanowires were shown to have an output voltage of 3 V and an output current of 5.5 nA when impacted by an aluminium arm oscillating at a frequency of 5 Hz with an amplitude of 1 mm. The energy harvesting performance of the P(VDF-TrFE) nanowires has been enhanced by fabricating devices comprising nanowires freed from the AAO membranes and laterally aligned on substrates with lithographically-patterned inter-digitated electrodes. Strategies to further improve the electromechanical coupling and thus the energy harvesting performance of the P(VDF-TrFE) nanowires will be discussed.

The modern world is largely defined by electricity. Everything that hallmarks the high-tech era, from advanced illumination, through smart appliances to portable and even wearable electronics, depends for their appearance and development on electricity. Although electricity-generating technologies have been developed for some two hundred years, people have never ceased to explore new methods of power generation in order to address the rapidly rising demand on the electric energy in the modern society.
In this work, we report a unique planar-structured approach called triboelectric generator (TEG) that converts mechanical energy into electricity through harnessing triboelectrification between two rotary surfaces. The periodically changing triboelectric potential generates induced alternating currents formed by free electrons. Enabled by a design of two radial-arrayed fine electrodes that are complementary on the same plane, the TEG completely solves the problem of low output current that has limited the use of triboelectrification in electricity generation. Operating at a rotation rate of 3000 revolutions per minute (rpm), a TEG having a diameter of 10 cm could produce an open-circuit voltage of ~850 V and a short-circuit current of ~3 mA at a frequency of 3 KHz. Under a matched load of 0.5 MOmega;, an average effective power of over 1.5 W could be delivered to an external load at an efficiency of over 25 %. The TEG could directly drive multiple types of light bulbs as a power supply and may potentially charge portable electronics with the assistance of a power management circuit. Given its compelling advantages including small volume, light weight, low cost, and proven scalability, not only is the TEG suited to harvest mechanical energy for self-powered electronics, but also it can be potentially applied for energy generation in large scale.

Interfacial debonding of photovoltaic (PV) encapsulants exposed to hostile application environments is not well understood. In particular, the effects of ultraviolet (UV) light and the environment on the debond kinetics of the highly strained material at a propagating crack tip remains unexplored. The damage of solar environments on PV module materials has traditionally been studied by exposing them to "stress parameters" such as elevated temperatures, moist environments and large doses of UV light. After exposure, degradation in the material has been quantified by measuring changes in selected properties such as color, stiffness and chemical composition. However, the mechanical stress and its importance in determining the kinetics of interface adhesive and cohesive cracking has been rarely investigated but remains critically important. The understanding obtained from improved control of multiple "stress parameters" and their effects on the evolution of defects forms the basis for life-time prediction and the design of accelerated tests.
Using a recently developed quantitative mechanics technique, we report the effect of indoor aging on the debond energy of a photovoltaic ethylene-co-vinyl acetate (EVA) encapsulant. Employing a debond-kinetics characterization method with in-situ UV, we report debond growth rates of the EVA encapsulant in terms of the applied mechanical loads in an environment of controlled temperature and relative humidity. We employ a load relaxation technique that allows debonding rates as low as 1 nm/s to be accurately quantified and related to the rupture of molecular bonds at the crack tip. The effect of moisture, temperature and UV exposure on debond growth acceleration is demonstrated. The debond energy of the encapsulant decreased dramatically from 2000 J/m2 at 20 °C to less than 500 J/m2 at 60 °C. The debond growth rate increased up to 1000-fold with small changes of temperature (10 °C). To elucidate the degradation processes leading to envrionmental debonding, the kinetics of the debond growth process are interpreted using a recently developed viscoelastic fracture-mechanics model, including the effects of moisture and UV light. The molecular bond rupture processes investigated underlie the principal causes of degradation in PV materials exposed to terrestrial environments and can be exploited to acquire, not only a fundamental understanding of damage formation and progression, but also to develop accelerated testing techniques and make long term reliability predictions.

The next generation of high-end rechargeable batteries will still rely on lithium-ion host materials. Later, post-lithium-ion systems, at first Li/S, are expected to enter the market. Independently of the technology, understanding the fundamental electrochemical, structural, and mechanical properties of battery materials and the interactions of these materials with their environment will be the key to further improvements in energy density, safety, and life time of batteries. Our approach to answer the related scientific questions starts with the development of various in situ methods for use mainly in the field of nonaqueous solid-state electrochemistry. Then, the physical and electrochemical properties of host materials and electrochemical interfaces are investigated in situ.
When information related to the bulk of a battery material is needed, X-ray diffraction (XRD) is the method of choice. The electrode materials normally accommodate (insert, intercalate) variable quantities of lithium ions. As a rule, the lithium insertion / de-insertion into / from the electroactive material results in lattice changes of the material with associated volume changes inducing strain, which can be followed by analyzing the X-ray diffractograms. X-ray powder diffraction experiments performed at a standard laboratory diffractometer are sufficient to answer most of the scientific questions, however, when the time factor is critical a synchrotron based XRD is advantageous. Furthermore, in contrast to XRD, neutrons are much more sensitive to lithium. Therefore, the combination of results from synchrotron based in situ X-ray diffraction methods and in situ neutron diffraction is essential for deep understanding of the reactions of battery materials. For in situ neutron diffraction experiments a sophisticated electrochemical cell was constructed allowing quantitative analysis using the Rietveld method. In addition, in situ dilatometry on battery electrodes provides valuable information on their periodic expansion and contraction.
In the talk, the experimental background with focus on the construction of the in situ cells will be given followed by discussion of selected results on battery electrodes.

The functionality of energy storage and generation systems like Li-ion batteries or fuel cells is not only based on but also limited by the flow of ions through the device. To understand device limitations and to draw a roadmap to optimize device properties, the ionic flow has to be studied on relevant length scales of grain sizes, structural defects, and local inhomogeneities, i.e. over tens of nanometers. Knowledge of the interplay between the ionic flow, material properties, and microstructure can be used to optimize the device properties, for example to maximize energy density, increase charging/discharging rates, and improve cycling life for Li-ion batteries for applications in electric vehicles and aerospace. Existing solid-state electrochemical methods are limited to a spatial resolution of 10 micron meter or greater, well above the characteristic size of grains and sub-granular defects, and are based on the measurement of current. Because of that, these established techniques cannot be scaled down easily to the nanoscale. However, ionic flows in solid state systems also generates measurable strains which can reach hundreds of percent in the case of Si anodes for Li-ion batteries. By developing Electrochemical Strain Microscopy (ESM), we utilize the correlation between ionic concentration and strain to investigate ionic transport on the nanoscale using Scanning Probe Microscopy.
Here, we present how ESM can be used to measure ionic transport properties in Li-ion batteries and electrochemical supercapacitors. Both systems exhibit very different origins for strain during the electrochemical reactions which will be discussed in detail. For the first, we will show how to use ESM to extract spatially resolved maps of the activation energy and diffusivity of Li-ions in thin film cathodes for Li-ion batteries. We also explore how the ionic transport is influenced by the electrode microstructure such as grain orientation, step edges, and mechanically induced defects. Additionally, the application of ESM to study ionic transport across solid-solid interfaces will be presented and discussed. For the second, we will explore how the transport kinetics of ions from an ionic liquid into a porous carbon electrode is limited by the pore size and effectively determines the rate limitation of the capacitance.
Support was provided by the U.S. Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division through the Office of Science Early Career Research Program, by the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy, and by the Fluid Interface Reactions, Structures and Transport (FIRST) Center, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences.

The coupling between electrochemical and mechanical behavior is of great importance for battery electrodes, particularly in light of the large volume changes associated with intercalation. Here, we introduce a method for in-situ analysis of micro-mechanical properties during electrochemical cycling of electrode materials. An electrochemical cell is integrated into a Nanoindenter (MTS G200) which allows the real time measurement of the electrode contraction/expansion and the change in elastic constants during dis-/charge and the determination of the mechanical properties like yield stress and toughness at different states of charge. The setup is suitable for studying the local response of bulk crystals, single particles or granular electrodes during electrochemical cycling in an ionic liquid or organic electrolyte.
Different material systems for anode and cathode materials are under investigation. First successful tests have been carried out on graphite materials such as HOPG (single crystal) and MCMB (particle). The observed volume expansion is consistent with the literature and it can be shown that the Li intercalation reversibly changes the elastic behavior and can be attributed to changes in the microstructure. A detailed analysis of the changes in elastic response of these highly anisotropic materials during lithiation will be presented.
In a further step, the setup is also used for probing the coupling between mechanical stress and the chemical potential of Li. In theory it is possible to shift the chemical potential using mechanical stress by as much as 10-100mV/GPa depending on the material system. For anode materials that are exposed to high stresses during the Li reaction and have working potentials just above the Li-plating potential (e.g. Si), the stress induced shift of the chemical potential becomes an important effect and can have a profound influence on battery performance. We present results on graphite and silicon electrode materials.

Rechargeable lithium ion batteries are currently still irreplaceable power sources for a variety of electronic devices. The batteries generally require fast charging and discharging rate, high energy density, long cycle life, and many other properties. The performance assessment of the Li-ion battery is primarily based on the overall behavior of the sintered bulk cathode or anode material. However, the fundamental components, i.e. the individual particles in these materials, have rarely been studied independently due to many technical difficulties, such as fixation and probing technique. In this work, the responses of a single particle under a controlled electric field were observed by the advanced Scanning Probe Microscopy technique. Two types of the cathode particles Li(Li0.2Mn0.54Ni0.13Co0.13)O2 and LiNi1/3Co1/3Mn1/3O2, whose size typically ranges from 100 to 200 nm, were characterized. In this case, the Li-ion diffusion and O2 evolution of the smallest unit in the Li-ion battery, i.e., the isolated single particle, can be observed based on the changes of particle volume and surface area under gradually increased/decreased DC bias. High-resolution three dimensional particle images were observed with tapping-mode Atomic Force Microscopy. These results greatly benefit to our understanding of the various mechanisms in the development of the Li-ion batteries.

Electrochemistry coupled with mechanics dictates the microstructural evolution and service life of many materials in the energy industry, and underlies problems such as stress-corrosion cracking and battery cyclability. While atomistic and first-principles modeling is adept at looking at the finer details of energetics and microstructural evolution, it often needs help from experiments to identity the key performance-limiting processes. The creation of a nanoscale electrochemical and mechanical testing platform [Science 330 (2010) 1515; Nano Letters 11 (2011) 4535; ACS Nano 6 (2012) 9425; Nature Nano. 8 (2013) 277] inside a transmission electron microscope (TEM) enables direct observations of the electrochemical lithiation and delithiation of the nanowires. SnO2, ZnO, Si, Ge, graphene and carbon nanotube anodes and LiFePO4 nanowire cathode have been tested. Lithium embrittlement is found to be a persistent issue. These in situ experiments complement our modeling efforts, and together they provide insight into how materials degrade in service due to combined electrochemical-mechanical actions.

The tremendous volume expansion and physical deformation of silicon as it alloys with lithium underscores the importance of understanding the mechanical properties and stability of lithiated silicon as a prospective material for lithium-ion battery anodes. Furthermore, the constant charging and discharging nature of battery electrodes implies inherent time constraints on the various physical processes occurring during alloying and de-alloying of the host material. In this work we explore time-dependent deformation mechanisms in fully-lithiated silicon nanowires using in situ mechanical testing methods. Creep testing was performed by applying constant force loading to lithium-silicon nanowires inside a scanning electron microscope. Elongation measurements were used to calculate the strain-rate and clear indications of quasi-viscous deformation behavior are present. Measurements of the wire size and microstructure are also used to gain insight into the atomic diffusion pathways which govern the creep behavior in these lithium-silicon alloy materials. Implications regarding the implementation, design and construction of silicon-based electrode materials will be discussed.

The formation of dendrites, i.e. deposits of metallic lithium on the surface of graphite anodes, is a major safety concern for lithium ion batteries (LIBs). In January 2013, two Boeing 787 Dreamliners experienced LIB failures that led to a worldwide grounding of all 787s.[1] Although the root cause for failure has not yet been officially identified, the incidents have been linked to lithium dendrite formation.[2] However, the mechanisms and the conditions of dendrite growth are still not understood. Theoretical efforts are underway [3,4] and experimental studies have already established general conditions for lithium plating. For example, optical microscopy experiments have been conducted that allow tracking lithium plating in real time.[5] However, these studies typically require tailored cells and specific electrode geometries for optical access, and therefore do not fully mimic the conditions in a real battery.
Here we present tomographic analyses visualizing and quantifying the morphology of lithium dendrites occurring in commercial LIBs. We focus on separators and link the three-dimensional shape of dendrites and their spatial location within the separator to electrochemical testing conditions and cell temperature. Insights gained by this approach can (1) provide the basis for understanding the fundamental mechanisms governing dendrite growth, (2) allow identification of safe operating conditions based on direct assessment of the root cause of battery failure, and (3) guide the selection and development of separators that can appropriately handle dendrite induced short-circuits.
[1] Interim Factual Report: Boeing 787-8, JA829J Battery Fire; Case Number: DCA13IA037, National Transportation Safety Board, Office of Aviation Safety, Washington DC, USA, 2013.
[2] A. Heller, The G. S. Yuasa-Boeing 787 Li-ion Battery: Test It at a Low Temperature and Keep It Warm in Flight, Interfaces 2013, 22, 35.
[3] R. Akolkar, Modeling dendrite growth during lithium electrodeposition at sub-ambient temperature, J. Power Sources 2014, 246, 84.
[4] D. R. Ely, R. E. García, Heterogeneous Nucleation and Growth of Lithium Electrodeposits on Negative Electrodes, J. Electrochem. Soc. 2013, 160, A662-A668.
[5] S. J. Harris, A. Timmons, D. E. Baker, C. Monroe, Direct in situ measurements of Li transport in Li-ion battery negative electrodes, Chem. Phys. Lett. 2010, 485, 265-274.

Lithium-ion batteries (LIBs) are the most widely used energy storage device for electric portable applications and power sources due to their outstanding features, such as non-memory effect, long cycle life, small volume, few self-discharging and light weight. However the capacity decay, power fading and the increase of the impedance during cycling still need to be understood. During the operation cycles of LIBs, lithium ions are repeatedly intercalated from cathode into anode upon charge and de-intercalated reversibly from anode into cathode during discharge. It is therefore necessary to study the transport mechanisms of Li-ions, further to understand the aging mechanism and hence contribute to the performance enhancement of LIBs. The newly-developed Electrochemical Strain Microscopy (ESM) technique allows the high frequency periodic bias to be applied on the sample surface of the electrochemically active materials, and the bias will induce the local periodic oscillatory displacement caused by the Li-ions transport and redistribution within the material. The local surface displacement, which is defined as electrochemical strain, can be detected by a highly sensitive photodetector in ESM. Therefore, ESM has been emerging as a powerful technique to investigate the Li-ion preferred transport paths and the local electrochemical reaction mechanisms associated with Li-ion concentration changes. In this study, a promising cathode system, lithium-rich layered oxide material Li2MnO3-LiMO2 (M = Ni, Co, Mn), with high capacity (280mAh/g) which is approximately twice of that of the commercial cathode materials, is studied by using ESM technique. The results indicate that transport phenomenon of the Li-ions has a direct relationship with surface topography variations. As the applied bias increases, two types of deformation are observed, which are closely, related to the different Li-ion transport mechanisms (migration induced by electric potential difference and diffusion induced by concentration gradient) in the Li-rich cathode thin film. In addition, the grain expansion in cathode is found to be the results of pure Li-ions diffusion.

The scale-up of lithium-ion cells to capacities of 5-100 Ah is critical for their successful implementation in vehicle and grid energy storage systems. However, complex current pathways in large cells cause an uneven spatial distribution of reacted species, and ultimately an under-utilization of active materials. As a result, when scaling to larger capacities the effective energy density of each cell diminishes. Furthermore, uneven current distribution compromises cell safety by causing localized overcharge and overdischarge hotspots. Recent modeling efforts have provided insight towards this spatial variation but such efforts should be validated by experimental techniques.
In this work we focus on measuring the spatially resolved state of charge in a commercial 8 Ah lithium iron phosphate cell. For this purpose, energy-dispersive x-ray diffraction experiments are carried out by probing the cell in situ with a high energy synchrotron white beam. Nine different locations are probed over the course of discharge and the localized state of charge is determined by measuring the relative amounts of lithiated and non-lithiated phases. The data show a variation of +/- 15% in state of charge across the cell and a faster rate of discharge through the center of the cell and towards the positive tab.

The non-aqueous lithium-oxygen battery is one of a host of emerging opportunities available for enhanced energy storage [1]. Unlike a conventional battery where the reagents are contained within the cell, the lithium-oxygen cell uses dioxygen from the atmosphere to electrochemically form the discharge product lithium peroxide. Degrees of reversible oxidation and formation of lithium peroxide has been demonstrated in a number of non-aqueous electrolyte classes, mostly notably in dimethysulfoxide based electrolytes [2], thus making the lithium-oxygen cell a potential energy storage device.
This talk will present our groups recent results of the electrochemistry of dioxygen in non-aqueous electrolytes, of which particular electrolytes could have practical application within a lithium-oxygen cell. Discussion will touch upon how the electrochemistry can be related to electrode substrate and will be presented with in situ spectroscopic studies that identify intermediate and surface species during the oxygen reduction reaction.
[1] P.G. Bruce, S. Freunberger, L.J. Hardwick, J.-M. Tarascon, Nature Mater. (2012) 11 19
[2] Z. Peng, S.A. Freunberger, Y. Chen, P.G. Bruce, Science, (2012) 337 563

Metal oxides are considered as potential future anode materials for lithium ion batteries due to their much higher theoretical capacities compared to currently commercial available anode, graphite (372 mAh g-1). Tin oxide (SnO2) and iron oxide (Fe3O4) were chosen due to the abilities to reversibly form alloy with lithium and react with lithium through a process known as conversion reaction, respectively. However, for both SnO2 and Fe3O4, huge volume changes occur during the process of lithium insertion and removal, which will induce a breakdown in electrical contact between adjacent active particles, and eventually a large drop in capacity over charge-discharge cycles. To alleviate such problem, different forms of carbon were introduced to metal oxides, forming composite structures. These carbons, including graphene (G), were able to improve the electrochemical performance of metal oxide anode materials due to their high conductivity, good lithium permeability, and flexibility to hold the structure integrity. Besides, specially designed 3D structures, such as hollow and porous beads, provided void space for volume change during lithium insertion/removal, thus enhancing the cycling performances of these metal oxide anodes.

We demonstrated the synthesis of carbon thin film with ordered nanostructure via nanoimprint lithography starting from thermal plastic polymer polyvinylpyrrolidone (PVP). After the pattern is transferred to PVP thin film during nanoimprinting process, a three-step thermal treatment is performed to fabricate carbon thin film with pattern. Feature size of the pattern ranges from 250 nm up to 2 um with various shapes including 1D line, 2D hole and pillar. Characterized with Raman spectroscopy, the carbon thin film shows a high degree of disorder with a small in-plane crystallite size of about 1.346 nm calculated using the peak intensity of D band and G band of sp2 carbon. Those imprinted carbon thin film exhibits good electrical conductivity. When used as anode of lithium ion battery, it shows an enhanced cyclic performance compared to those carbon films without patterns. Further study with Electrochemical Strain Microscope (ESM) helps to understand the enhanced Li ion capacity locally on nanoscale.

From mechanics perspective, the biggest impediment to improving reliability of Li-ion batteries is the large film stress induced in electrodes upon Li-ion intercalation. Inherent stress fluctuations within electrodes during regular battery life lead to crack initiation and propagation, eventually leading to capacity degradation. In this study, we investigate a strategy to restore conductance across newly formed anode cracks; thus preventing capacity degradation in Li-ion batteries. Within the anode matrix, we incorporate microcapsules with a conductive core (carbon black suspended in solvent). Upon mechanical trigger, microcapsules crack and deliver their core in damage zone. We designed a custom battery cell to observe healing in-situ besides other set ups to assess release, delivery, and functionality of the conductive core outside of the battery. The core is shown to be an effective healing agent outside of the battery. In the custom battery cell, we accelerate crack formation by creating mechanically vulnerably location on a conventional graphite base battery anode, and compare self healing electrodes to control electrodes based on the trend of battery capacity and color variations due to Li-ion intercalation.

“Electrochemical shock”—the electrochemical cycling-induced fracture of materials—contributes to impedance growth and performance degradation in ion-intercalation batteries, such as lithium-ion. Using a combination of micromechanical models and acoustic emission experiments, the mechanisms of electrochemical shock are identified, classified, and modeled in targeted model systems with different composition and microstructure. A particular emphasis is placed on mechanical degradation occurring in the first electrochemical cycle. Three distinct mechanisms of electrochemical shock are identified, and a fracture mechanics failure criterion is derived for each mechanism. In a given material system, crystal symmetry and phase-behavior determine the active mechanisms. A surprising result is that electrochemical shock in commercial lithium-storage materials occurs by mechanisms that are insensitive to the electrochemical cycling rate. This fundamental understanding of electrochemical shock leads naturally to practical design criteria for battery materials and microstructures that improve performance and energy storage efficiency. These microstructure and crystal chemical design criteria are demonstrated experimentally for spinel materials such as LiMn₂O₄ and LiMn_1.5Ni_0.5O₄. A case study of LiMn_1.5Ni_0.5O₄ is presented, in which small changes in composition that have negligible impact on electrochemical properties induce a significant change in phase behavior that allow electrochemical shock at relevant electrochemical cycling rates to be avoided. While lithium-storage materials are used as model systems for experimental study, the physical phenomena are common to other ion-intercalation systems, including sodium- and magnesium-storage compounds.

We demonstrate that nanometer-scale TiN coatings deposited by atomic layer deposition (ALD), and to a lesser extent by magnetron sputtering, will significantly improve the electrochemical cycling performance of silicon nanowire lithium-ion battery (LIB) anodes. A 5 nm thick ALD coating resulted in optimum cycling capacity retention (55% vs. 30% for the bare nanowire baseline, after 100 cycles) and coulombic efficiency (98% vs. 95%, at 50 cycles), also more than doubling the high rate capacity retention (e.g. 740 mAh/g vs. 330 mAh/g, at 5C). We employed a variety of advanced analytical techniques such as electron energy loss spectroscopy (EELS TEM), focused ion beam analysis (FIB) and x-ray photoelectron spectroscopy (XPS) to elucidate the origin of these effects. The conformal 5 nm TiN remains sufficiently intact to limit the growth of the solid electrolyte interphase (SEI), which in turn both improves the overall coulombic efficiency and reduces the life-ending delamination of the nanowire assemblies from the underlying current collector. Our findings should provide a broadly applicable coating design methodology that will improve the performance of any nanostructured LIB anodes where SEI growth is detrimental.

Lithium-ion batteries have revolutionized portable electronics, and will be key to electrifying transport vehicles and delivering renewable electricity. Amorphous silicon (a-Si) is being intensively studied as a high-capacity anode material for next-generation lithium-ion batteries. Its lithiation has been widely thought to occur through a single-phase mechanism with gentle Li profiles, thus offering a significant potential for mitigating pulverization and capacity fade. Here, we discover a surprising two-phase process of electrochemical lithiation in a-Si by using in situ transmission electron microscopy. The lithiation occurs by the movement of a sharp phase boundary between the a-Si reactant and an amorphous LixSi (a-LixSi, x ~ 2.5) product. Such a striking amorphous-amorphous interface exists until the remaining a-Si is consumed. Then a second step of lithiation sets in without a visible interface, resulting in the final product of a-LixSi (x ~ 3.75). We show that the two-phase lithiation can be the fundamental mechanism underpinning the anomalous morphological change of microfabricated a-Si electrodes, i.e., from a disk shape to a dome shape. Our results represent a significant step toward the understanding of the electrochemically-driven reaction and degradation in amorphous materials, which is critical to the development of microstructurally stable electrodes for high-performance lithium-ion batteries.

Silicon has been an attractive candidate for use as the anode material in Li ion batteries. This is because the storage capacity of Si for lithium insertion is over an order of magnitude greater than that of conventional anode materials like graphite. Bulk and thin film Si anodes suffer severe mechanical degradation and loss of capacity when subjected to several charge/discharge cycles because of the large volume change associated with lithiation into Si. Nanostructured Si electrodes, such as nanowires and inverse opal lattices, have greatly improved mechanical characteristics due to increased ductility in nano-sized Si and the ability of pores/free spaces to absorb Si volume change during lithiation cycles.
We present a 3D silicon nano-lattice whose structural geometry is optimized for mechanical toughness during large volume changes. Principles from solid cellular mechanics were used to determine connectivity and angles between lattice members to minimize global volume change and mechanical stress under lithiation. 2-photon laser lithography was used to write a 3D polymer scaffolds in the shape of the optimized lattice structure in which beams are ~300-700 nm wide and 1 to 5 microns in length. Cu and then amorphous Si is deposited onto the polymer lattice using RF magnetron sputtering. Si layer thickness is limited to hundreds of nanometers in order to take advantage of the enhanced ductility in nanoscale Si.
In-situ scanning electron microscopy was used to directly observe volume change, mechanical deformation and failure in the Si anode during cycles of lithiation. A lithium electrode was mounted on a telescoping bar on the wall of the SEM chamber. Electrochemical testing was performed by contacting the lithium electrode, electrolyte and Si nano-architected electrode (located on the SEM sample stage) and applying a bias voltage bias across the electrodes. In-situ SEM observations show that mechanical reliability of Si electrodes can be controlled and improved by constructing them as nano-architected structures.

Design and fabrication of reliable electrodes subject to repeated deformation is one of the most important challenges in flexible energy applications since mechanical and electrical properties of devices degrade gradually because of fatigue damage. Here, we introduce two unique nanostructure designs for flexible thin film electrodes on a polymer substrate to suppress fatigue crack damage during cyclic bending. First, a fatigue damage-free flexible metal electrode was developed. To suppress crack nucleation, we introduced a novel nanostructured fatigue damage-free copper electrode on flexible substrate by creating 2-D nanohole arrays. By Implementing this fatigue damage free cu electrode, we successfully demonstrated a flexible Li ion Batteries. Secondly, we developed a polymer nanofiber/TiO2 nanoparticle composite photo-electrode with high bendability by a spray-assisted electro-spinning method. The composite film structure similar to that of a fiber-reinforced composite is used as the photo-electrode in plastic dye-sensitized solar cells (DSCs). Compared to conventional DSCs, composite-based DSCs show outstanding bending stability because the polymer nanofibers prevent delamination of the electrode by relieving the external stress and effectively retarding crack generation and propagation.

The zinc antimonide has been of interest for years in the search for efficient and low-cost thermoelectric materials. Of primary interest has been the β-Zn₄Sb₃ phase which shows a thermoelectric figure of merit, zT, in excess of 1 in proper temperature ranges. This phase is relatively environmental friendly and earth abundant elements, continues to lead vibrant research. Especially, compound Zn_xSb_y is one of the most efficient thermoelectric materials known in high temperature. High temperature thermoelectric materials are of interest for applications as power generators by using waste heat from flue gas. For waste heat recovery applications such as vehicle exhaust and power plant heat, it is very important to find materials with enhanced power factors at above 150 °C. We investigated the Zn_xSb_y thin films deposited by radio frequency (RF) magnetron sputtering for applications in high temperature thermoelectrics. We maximized thermoelectric properties of Zn_xSb_y thin films by adjusting RF power, deposition time, substrate temperature and so forth. At 321 °C, the measured resistivity and Seebeck coefficient values of the Zn_xSb_y film were 17.2 mu;Omega; m and 146 mu;V K^-1, respectively, yielding a power factor value of about 1.39 mW m^-1 K^-2, which is consistent or little above with the reported results for Zn₄Sb₃ single crystals at 400 °C. We also studied extensively the relationships between phase transformation and thermoelectric properties of Zn_xSb_y thin films by using high temperature X-ray diffraction (XRD) analysis, as a function of operation temperature. According to high temperature XRD analysis, the thermoelectric performanceZn_xSb_y thin films depend primarily on a phase distribution in thin film such as Zn₄Sb₃ and ZnSb, etc. As a result, the highest zT of the Zn_xSb_y film were estimated to be 1.26 at 321°C. This value is conservatively estimated since the thin films should have a much reduced thermal conductivity due to their small crystallite sizes and thus the zT value would be higher. Moreover, the Field-emission scanning electron microscopy (FE-SEM), transmission electron microscopy (TEM), atomic absorption Spectrophotometer (AAS), and specific resistance measurements showed that Zn_xSb_y thin films is useful candidate for high temperature thermoelectric applications.

Recently, some of the layered MAX phases (Mn+1AXn systems, where n = 1, 2, or 3, "M" is an early transition metal, "A" is A group elements, mostly groups 13 and 14 elements, and "X" is carbon and/or nitrogen) have been exfoliated into two-dimensional single- and multi-layers, so-called MXenes [1,2]. Bulk MAX phases are metallic systems. However, our calculations exhibited that upon appropriate functionalization by F, OH, and O groups, some of the exfoliated MAX phases become narrow-band-gap semiconductors, and therefore, might be potentially good thermoelectric materials [3]. Here, on the basis of a set of first-principles band structure calculations combined with Boltzmann transport theory [4,5], we have studied the transport coefficients (Seebeck, electrical conductivity, and power factor) of various functionalized MXenes. Our calculations predict that a number of MXenes may exhibit high thermoelectric performance.
[1] M. Nagubi et al., Adv. Mater. 23, 4248 (2011).
[2] M. Nagubi et al., ACS Nano 6, 1322 (2012).
[3] M. Khazaei et al., Adv. Funct. Mater. 23, 2185 (2013).
[4] G. K. H. Madsen and D. J. Singh, Comput. Phys. Commun 175, 67 (2006).
[5] J. Yang et al., Adv. Funct. Mater. 18, 2880 (2008).

The materials class of skutterudites is one of the promising thermoelectric materials due to its decent electronic properties and cage-like structural feature that can be filled with guest atoms. In this study, first-principles calculations have been performed in order to investigate electronic band structures and related transport properties of pnictogen-substitued skutterudites filled with alkaline-earth elements (MxCo4X6Te6 where M=Ca, Sr, or Ba, X=Ge or Sn, and x=0.5 or 1). Electronic transport properties related to thermoelectricity, including the Seebeck coefficient and the electrical conductivity, are computed by using the Boltzmann transport formalism within the constant-relaxation-time-approximation. The results are compared against the corresponding properties of the unfilled pnictogen-substitued ternary skutterudites (CoX1.5Te1.5) to identify the effects of filling, based on which the potential of filled pnictogen-substituted skutterudites for thermoelectric applications is evaluated. The possible changes in the ionic character of the interatomic bonding, which was suspected to be an important scattering source in unfilled pnictogen-substitued ternary skutterudites, are probed by analyzing the projected density of states, in order to identify a path for potential improvement of the thermoelectric performance of pnictogen-substitued skutterudites.

Supercapacitors are one of the most promising electrochemical energy storage systems due to rapid charge/discharge, cycle stability, high power density, and applications in many fields from electric devices. However, because low energy density have biggest drawback of the supercapacitors, it is difficult to use of the lithium-ion battery levels. So, the role of the electrode is important to enhance the energy density. The importance of the electrode in supercapacitors has been already reported in the latest literature and it has been applied using many electrode carbon materials such as CNT, graphene. Recently, biomaterials have attracted considerable attention for their applications from environment to electric field. Of biomaterials, a new cellulose source, bacterial celluloses (BC), has emerged. BC is the name given to cellulose that is obtained from bacteria through process, such as biosynthesis from of various microorganisms, enzymatic in vitro synthesis, or even chemosynthesis from a glucose derivative. These materials are not only frequently used as model substances for further research on the structures and reactivity of cellulose but also used to develop new materials because of their specific nano-structures. Nanostructured design of electrode materials is important for electrode as supercapacitors to achieve high performances. Carbon materials obtained from carbonized BC maintain the nano-structure which made efficient diffusion of electrolyte ions and conducting pathway of the electrode. However, the carbonized BC has macroporous structure and deficiency of micropores, which are advantageous in electrical double layer capacitance. This can be improved by activation process using KOH. Also, nitrogen doping on the surface of the carbonized BC can induce increasing capacitance owing to both EDLC-based capacitance and pseudocapacitive effects.
In this study, the nitrogen-enriched hierarchical porous carbons from bacterial cellulose were prepared, and their electrochemical performances for supercapacitors were investigated.

Two-dimensional materials and their functionalized structures have shown interesting promise for thermoelectric applications. For example, by reducing the dimensionality of the structure, a new route for substantially enhancing the figure of merit in graphene has been recently suggested, suppressing the thermal conductivity while at the same time enhancing the power factor. In this work, we explore the potential of stacked two-dimensional materials for thermoelectric applications. These heterostructures, which consist of a combination of different isolated layers, give rise to unusual properties due to the relatively weak Van der Waals interaction between the layers. Because the individual layers intact with strong covalent bonds, they maintain their dimensionality even after stacking. Results will be presented for the power factor and thermal conductivity computed using a combination of quantum mechanical and classical simulation approaches, for a range of stacked heterostructures.

To reduce our dependence on fossil fuels and to reduce greenhouse gas emissions, thermoelectric(TE) technology, which can convert waste heat directly into electricity, is one of the more reliable technologies available. Magnesium silicide (Mg2Si) has been recognized as a promising environmentally-benign thermoelectric material operating in the temperature range from 600K to 900K due to several attractive features, such as its lightweight, the abundance of its consistent elements, and its non-toxicity. Focusing on the appropriate systems for integration into automotive applications, TE power generation modules should be resilient against thermal stresses and vibration during heat-cycling conditions. In order to realize a practical TE power generation module, greater understanding of the mechanical properties of appropriate TE materials is essential to enable device design of structures for TE power generators.
Currently, Sb-doped Mg2Si is one candidate that offers practical durability and useful ZT values at elevated temperatures. However, it has been recognized that it has poor sintering reproducibility due to the undesirable inclusion of cracks. In order to improve the scalability and reproducibility of Sb-doped Mg2Si during sintering to sufficient levels, metallic binders such as Al, Cu, Ni, and Zn have been incorporated during the sintering process, resulting in a successful increase in the productivity of sintered pellets,accompanied by improved stability and enhancement of their TE properties.
In order to design a TE module, appropriate mechanical properties are required for the corresponding TE materials. Thus, the purpose of this report is to understand the mechanical properties of sintered Sb-doped Mg2Sithat can be made using our present fabrication methods. We examined ultrasonic tests, nano-indentation tests,3- or 4-point bending tests, and compression tests on sintered pellets of Sb-doped specimens with or without metallic binders. We evaluated the mechanical properties, such as Young&’s modulus, hardness, bending strength, fracture toughness, compressive strength in order to characterizethe currently available Mg2Si. When we think about mass-production-scale fabrication process for Sb-doped Mg2Si TE chips, good spatial homogeneity of the sintered pellets in terms of TE and mechanical properties is taken as read. Thus, we would like to discuss the effects of the incorporation of metallic binders during sintering in terms of local or spatial variations in those mechanical properties that are correlated with the TE properties. Since the TE materials are expected to be utilized in atmospheres at elevated temperatures, additional mechanical properties such as aging functions were also investigated.

The direct conversion of waste heat to electricity is a large incentive to find viable thermoelectric (TE) materials and efforts worldwide are intensifying [1]. The challenge is that TE materials must meet the environmental friendliness and economical requirements in addition to achieving high TE efficiency. Therefore, designing TE device from non-toxic and earth-abundant materials should be a key strategy. Recently, we have discovered that the mineral chalcopyrite, chemically expressed as CuFeS₂, exhibits good TE properties by carrier doping [2]. The electron doped sample Zn_0.03Cu_0.97FeS₂ and Cu_0.97Fe_1.03S₂ show large negative Seebeck coefficient and relatively low electrical resistivity, resulting in a high power factor of above 1 mW/K²m around room temperature. However, the compounds exhibit high thermal conductivity of 5 W/Km. In addition, the compound is only stable at temperature below 600 K [3].
To overcome these problems, we have focused on Cu₉Fe₉S₁₆, known as mooihoekite. The crystal unit cell of Cu₉Fe₉S₁₆ is twice as large as that of CuFeS₂, involving 11 crystalographically inequivalent sites. The complex crystal structure will have a great effect in reducing lattice thermal conductivity [4]. We have synthesized Cu₉Fe₉S₁₆ samples and measured the properties. We observed a large negative Seebeck coefficient S = -150 mu;V/K with a low electrical resistivity ρ= 6 mOmega;cm at 400 K. The thermal conductivity around room temperature is significantly reduced to 2.0 W/Km. Furthermore, the sample is stable up to 800 K, sustainable at much higher temperatures than CuFeS₂.
[1] Thermoelectric Nanomaterials, Materials Design and Application, ed. K. Koumoto and T. Mori, (Springer, Heidelberg, 2013).
[2] N. Tsujii and T. Mori, Appl. Phys. Exp. 6 (2013) 043001. (Spotlights paper)
[3] N. Tsujii, T. Mori and Y. Isoda, J. Elec. Mater., submitted.
[4] E. S. Toberer et al., Chem. Mater. 22 (2010) 624.

Abstract: We have studied the thermoelectric properties of Pb-doped bulk samples prepared by high -pressure and high-temperature (HPHT) method. Seebeck coefficient, electrical resistivity, power factor were performed at room temperature. As our expected, the Seebeck coefficient has been increased with an increase of the synthetic pressure .This improvement is mainly ascribed to the abnormal increase in hole mobility, which is realized by a decreased degree of disorder with the introduction of Pb. The electrical resistivity increases with the increase of synthetic pressure, however that is not apparent yet. The results indicate that HPHT technique may be helpful for optimizing electrical and thermal transports in relatively independent way. In addition, the HPHT method has been shortened the processing time from several days to half an hour. The present study provides a new platform to prepare β-Zn4Sb3 with enhancing thermoelectric properties.
Key words: β-Zn4Sb3, HPHT, Pb-doped,Thermoelectric Performance

Converting waste heat into useful energy has been an attractive alternative in this energy-hungry world today. However, thermoelectric devices have been saddled with low efficiency or high cost material which limits its current potential in the commercial market. One class of emerging materials in the field of thermoelectrics is germanium antimony tellurides that are known for their application in phase-change memory devices. Recent report has shown that GeTe-rich germanium antimony tellurides have high thermoelectric performance. In this study, we tuned the vacancy in the alloy system by changing the germanium composition. Consequently, this influences the electronic properties of the alloy and improves the overall figure of merit, ZT, value from 0.7 to 1.1. Furthermore, quenching the alloy after annealing created more disorder in the crystal structure which influences the thermal conductivity of the material. In effect, this increases the ZT value from 1.1 to 1.48, which is one of the highest reported ZT values for this material.

NaxCoO2 is one of promising thermoelectric materials and its single crystal is reported that could have a high ZT larger than 1 at 800K. Compare to the single crystal, polycrystalline NaxCoO2 is much easier to fabricate, but has poor thermoelectric performance mainly due to a lower electrical conductivity. Texturing is one of the popular approaches to improve the electrical conductivity of anisotropic polycrystalline materials. In this research, highly textured NaxCoO2 ceramics is successfully manufactured through a novel technique which is based on sol-gel combustion synthesis and electric-field induced kinetic-demixing followed by a high temperature hot press process. The anisotropicity of the textured NaxCoO2 ceramics is observed using scanning electron microscopy and quantified through X-ray diffraction. Then direction dependent Seebeck coefficients, electrical and thermal transport properties of the sample are measured from room temperature to about 1000K.

Thermoelectric (TE) energy converters that make use of such alloys have attracted attention as they exhibit numerous interesting features such as solid-state operation, zero-emissions, vast scalability, low maintenance, and a long operating lifetime. Owing to the superior thermoelectric characteristics of such materials at temperatures close to room temperature, a number of techniques for fabricating high quality thin films, such as metal organic chemical vapor deposition (MOCVD), physical vapor deposition (PVD), molecular beam epitaxy (MBE), and electrodeposition have been well established. Among them, electrodeposition have been attracting attention because the techniques has many advantages, including its low costs, rapid deposition rate, and the ability to control the microstructures and crystallinities of the deposited materials by adjusting the parameters of the electrodeposition process. In this study, we investigated the effect of a surfactant on the electrodeposition of Sb2Te3 and Bi2Te3 films, and on their morphologies and structural, mechanical, electrical, and TE properties. On the addition of the cationic surfactant cetyltrimethylammonium bromide (CTAB), the electrodeposited films showed improvements in their thermoelectric property due to the better morphology and Te nanoparticle formation. In addition, dopant was used to further improve the TE property of electrodeposits. The relationship between the concentration of dopant and the TE properties was studied.

Thermoelectric materials are promising due to its capability of converting heat directly into electricity and vice versa, and can be used for both waste heat recovery and thermal management. In this work, we developed a homogenization method to study the effective behavior of thermoelectric composites with periodic microstructure. Unit cell problem is established first from asymptotic analysis, which is then solved numerically by finite element method. The effective thermoelectric properties are calculated, and the corresponding conversion efficiency is analyzed. It is discovered that the homogenized thermoelectric equations are significantly different from those of homogeneous materials. More importantly, higher conversion efficiency than those of the constituent phases is demonstrated, and the condition for improved conversion efficiency is identified. The analysis provides considerable insight into the effective behavior of thermoelectric composites, and it can be used to guide the design and optimization of high efficiency thermoelectric materials.

Kesterite material Cu2ZnSnS4 (CZTS) possesses promising characteristic to be a conventional absorber for thin-film solar cell due to its optimal band gap of 1.45-1.51 eV and high optical absorption coefficient (>104 cm-1). Recently, its versatility has been extended to thermoelectric energy harvesting. Comparing with many other thermoelectric materials composing rare or toxic element, CZTS is non-toxic, earth abundant, and low-cost. However, most currently reported CZTS or similar materials used for thermoelectric are synthesized from solid state reaction which requires long term high-temperature annealing process, or based on nanoparticles technique. However, it&’s well known that this quaternary nanoparticle is normally capped by insulating organic ligands which will block the carrier transport. To remove this organic impurity, it always comes to the price of highly toxic reducing agent like hydrazine or high-temperature annealing process, and even after annealing, the remaining carbon footprint and cracks due to the high volume loss will still harm the device performance.
Here, we&’re reporting a simple low-temperature technique to synthesis porous CZTS framework. In this method, the organic residues can be easily removed without involving any toxic or aggressive procedure. From the characterization of the thermoelectric properties of this material, porous CZTS is demonstrated to have excellent potential for thermoelectric applications.

Thermoelectric transport is usually considered in the context of thermoelectric generators and Peltier modules. These devices are typically large and operate relatively low thermal gradients. At small scales, the impact of thermoelectric transport becomes more significant. It is possible to achieve thermal gradients in the order of 10K/nm in self-heated micro- / nano- structures. Extreme thermal gradients and elevated temperatures give rise to a significant thermoelectric transport mechanism due to generation transport and recombination of minority carriers.
We have first observed this phenomenon in n++ doped silicon micro-wire bridges which were self-heated with application of short-duration high-amplitude voltage pulses. These wires, lithographically patterned to be perfectly symmetric and suspended in air, consistently showed melting at one end without melting the other. The asymmetry induced in the thermal profile of some of these wires were quite significant and the polarity was the opposite of what would be expected in conductors due to convective-electronic heat flow in the direction of the majority charge carriers.
In our analysis, we have identified that heat absorbtion by thermal generation of free carriers at the hot parts of the microwire and release of this heat down-stream for the minority carriers via recombination results in very significant thermal transport [1]. At sufficiently high-temperatures, this generation-transport-recombination (GTR) process dominates over electronic-convective heat flow in the opposite direction. Hence, in contrast to metals, highly doped semiconductors have a distinctly different high-temperature thermoelectric behavior.
[1] G Bakan, N Khan, H Silva, A Gokirmak “High-temperature thermoelectric transport at small scales: Thermal generation, transport and recombination of minority carriers”, Nature: Scientific reports 3
http://www.nature.com/srep/2013/130923/srep02724/full/srep02724.html

At the heart of a lithium-ion battery is a process that couples mechanics and chemistry. The electrodes in a lithium-ion battery are hosts of lithium. When the battery is charged and discharged, the electrodes absorb and desorb lithium, inducing inelastic flow and possibly fracture. Lithiation-induced fracture not only occurs in commercial lithium-ion batteries, but is also a bottleneck in developing future lithium-ion batteries. This talk describes a theory of concurrent diffusion, reaction, and flow, and relates the theory to recent experimental observations of high-capacity electrodes. The theory places driving forces for flow and reaction on the same footing, giving chemomechanical yield condition and chemomechanical flow rule. This combination of mechanics and chemistry generalizes the theory of plasticity and describes a large family of phenomena: reactive flow in solids. The theory leads to predictions of mass transport, deformation and fracture in electrodes caused by cyclic charge and discharge. The predictions of the theory are related to experimental observations.

The life of lithium-ion batteries is related to the mechanical expansion and contraction of the active materials along with solvent decomposition at the active material surfaces—lithium-ion batteries would not work if a protective shell did not cover the electroactive core of the positive and negative electrode materials. Exposure of the active core leads to cell degradation. Under what conditions are the protective (outer shell) materials compromised? In addition to reviewing literature that is relevant to answering this question, we derive governing equations and provide numerical and analytic results in this work to render a qualitative response to this question, and we identify open questions that limit the quantitative application of modeling results to these systems.
Frequently, host materials in lithium-ion batteries expand upon solute addition, and the outer shell plays a protective role and does not expand as much. Under such circumstances, the stress jump at the core-shell interface is maximum when the particle is fully saturated with solute (fully lithiated), as the core is fully expanded, and the shell is stretched over the core. Analytic solutions for this maximum-stress condition provide insight for the rational design and materials selection of robust core-shell structures. Hence, our overall approach is to employ numerical calculations for a conventional core-shell system, and we utilize the analytic solutions for the maximum-stress condition for an infinitely thin SEI stretched over a lithiated graphite core and over a lithiated Si core. There is a lack of knowledge of key material properties that are needed for quantitative mathematical modeling of many core-shell systems, particularly for SEI materials; we employed predictive modeling methods based on density functional theory to estimate needed parameter and property values.

We have measured the fracture energy of lithiated silicon thin-film electrodes as a function of lithium concentration. To this end, we have constructed an electrochemical cell capable of testing multiple thin-film electrodes in parallel. The stress in the electrodes is measured during electrochemical cycling by the substrate curvature technique. The electrodes are disconnected one by one after delithiating to various states of charge, i.e., to various concentrations of lithium. The electrodes are then examined by optical microscopy to determine when cracks first form. All of the observed cracks appear brittle in nature. The fracture energy of lithiated silicon is found to be similar to that of pure silicon and essentially independent of the concentration of lithium. Thus, lithiated silicon demonstrates a unique ability to flow plastically and fracture in a brittle manner.

Silicon is a promising material for the next generation lithium-ion anode materials due to its high energy density and abundant supply. However, wider use of silicon-based anodes as viable electrodes is hindered by challenges involving rapid performance degradation due to large volumetric changes in the film during cycling, which generate high stresses in the film that drive structural failure via fracture. Attempts at using novel architectures such as micro/nano-wires, patterned micro/nano-structures and thin films to improve the structural integrity of silicon-based anodes have had moderate success, but what still remains elusive is the ability to systematically optimize the designs of such architectures to minimize and even eliminate the likelihood of mechanical degradation. For this to occur, it is necessary to make quantifiable measurements of the stresses in the material that induce fracture and the material properties that resist it. Our previous work reporting in situ stress measurements during electrochemical cycling of thin film silicon anodes using multi-beam optical sensing (MOS) was an important milestone towards measuring the driving force behind fracture, and experimental and theoretical measurements of the biaxial moduli of lithiated silicon during cycling was another step towards determining the material properties at various states of charge. In this talk, we demonstrate a method to experimentally measure the fracture energy, G, of lithiated silicon as a function of lithium concentration, and preliminary results show that G varies between 9.8 and 35 J/m^2 with increasing lithium concentration. Topics regarding further refinement of experimental results will also be discussed.

During the operation of lithium-ion batteries, lithium ions diffuse from the cathode to the anode when the battery is charged, and from the anode to the cathode when the battery is used as a power source and discharged. The intercalation of ions into and out of electrodes induces cyclic volumetric expansion and contraction of the electrodes, leading to power and capacity fade via changes in the electrode porosity and electrode fracture respectively. The severity of the electrode degradation is related to the magnitude of the volumetric expansion of the active material. Graphite, which is used in commercial electrodes, expands 10% by volume when it is fully lithiated, leading to internal stresses small enough to allow many hundreds of cycles before significant fracture occurs. In contrast, high capacity anode materials, such as silicon and tin, expand 200-300% by volume and can experience complete capacity loss after only a few cycles.
To examine this interesting relationship between electrode mechanics and battery performance, we utilize digital image correlation (DIC) to measure the strains that develop in lithium-ion battery electrodes during electrochemical cycling. A custom battery cell that allows optical access to the electrodes is charged and discharged by galvanostatic cycling (constant current) and cyclic voltammetry (voltage sweep). Both graphite-based composite electrodes (similar to current commercial electrodes) and high-capacity silicon-based composite electrodes are studied. We find that changes in strain with respect to voltage (dε/dV) correspond directly with changes in capacity (dQ/dV) during lithiation and delithiation. Moreover, the polymer binder, conductive additives, and cycling rate have a significant effect on the electrochemical behavior of the electrode, specifically irreversible capacity, and on the magnitude of the electrochemically induced strains and non-recoverable deformation that occur during cycling.

Two important parameters governing battery performance are the charge rate and the charge capacity. The charging rate is proportional to and ultimately limited by the surface area of a material, while the charge capacity is limited by the amount, or volume, of material. By creating three-dimensional architected surfaces that increase the surface area without reducing the volume, it should be possible to greatly improve the charging rate without affecting the charge capacity. One such three-dimensional structure that people are working on is the inverse opal. An inverse opal is a structure that is formed by infiltrating the pores of a packing of spheres, then dissolving the original structure to create a negative, or inverse, packing. These structures are commonly made on nano- and micrometer length scales, using a colloidal polymer microsphere self-assembly technique in combination with electroplating. These structures are an ideal candidate for high-performance batteries due to their high surface area to volume ratio; their highly porous nature allows for enhanced infiltration of the electrolyte into the electrode, thereby decreasing electrolyte diffusion lengths.
In order to utilize inverse opals as high charge rate battery materials, a better understanding of the mechanics of the inverse opal nanostructure and its dependence on relative density must be developed. In this work, we utilize a technique called nanoindentation to characterize the mechanical properties of nickel inverse opal structures that have been fabricated using an electroplating technique and etched to have relative densities ranging from 9.4% to 19.5%. From the nanoindentation data, we extract the hardness and modulus of these structures, which range from 50MPa to 450MPa and 3GPa to 23 GPa respectively. In combination with nanoindentation, simple geometric relations were used to derive an analytic formulation for the relative density of the structure as a function of the diameter of the pore (D) and the spacing between the pores (A). Using this relation in combination with the nanoindentation data, we are able to show that the strength and modulus of the inverse opal structures follow a power law scaling with relative density to the 2.5 and 2.3 respectively. This matches well with the Ashby scaling laws for a bending-dominated structure. Additionally, we show that at low densities, the inverse opal structures become significantly weaker due to imperfections that arise during their fabrication, which affects the structure&’s ability to withstand local stress. This work gives a simple cohesive picture by which we can better understand the mechanical properties of inverse opal structures. By being able to accurately describe the mechanical behavior and strength of these structures, we will be able to better adjust the parameters of fabrication and further the development of high-performance power sources.

Lithiation and delithiation processes in Li ion battery electrodes lead to significant volume changes. In addition to creating internal stresses in the active electrode materials, chemically induced stresses can substantially alter the stability of the solid electrolyte interphase (SEI). It is difficult to probe the mechanical response of the SEI directly in complex electrode microstructures that consist of powdered active components and other constituents. However, thin films provide an opportunity to investigate fundamental processes more directly. This approach has been used to investigate SEI formation on both Si and Carbon electrodes. To accomplish this, we employed in situ stress, in situ AFM measurements, conventional in situ electrochemistry, and ex situ surface characterization with TEM, XPS, and SIMS. These experiments allowed us to investigate SEI behavior in different electrolytes and with different cycling conditions. Most relevant were the growth stresses and physical properties of the SEI obtained during this study. Significant differences between Si and graphitic carbon were observed in SEI growth and passivation mechanism. Both the electrolyte composition and the formation potential had significant effects on the SEI formation. The results from these experiments and corresponding models also suggest that stresses can be engineered during SEI formation, to enhance the stability of these critical passivation layers.

Ion-conductive polymers, or ionomers, are commonly used in most electrochemical energy conversion and storage devices including, but not limited to, polymer-electrolyte fuel cells, redox flow batteries, and artificial solar-fuel generators. The system performance and durability are strongly related to the transport and mechanical properties of the ionomer membranes which are correlated through its chemical structure and morphology. Moreover, most membranes used in energy devices are subjected to mechanical loads resulting from either device constraints or operational stresses. Thus, it is of great importance to understand the role of mechanics in the structure/functionality of ion-conductive polymers influenced by a number of phenomena such as heat treatment, ageing, compression and constraints, and interfacial confinement. Although seemingly unrelated, they are actually fundamentally related by the balance between the chemical and mechanical energies in the material. In this talk, we will demonstrate how mechanics can be used to provide a perspective for elucidating ionomers&’ properties as well as their behavior in devices across critical length scales. Specifically, we will focus on sulfonated ionomers which are one of the most widely studied solid-electrolyte materials due to their ability to provide good ion and solvent transport functionalities along with mechanical stability. First, it will be shown how environmental and operational conditions that improve the transport properties of bulk ionomers influence their mechanical properties. Transport and mechanical properties of ion-conductive materials are correlated via the nanostructure as governed by the chemical-mechanical balance. This balance is also altered during device operation due to ageing and degradation. Our findings, however, suggest that some of the degradation mechanisms affect the chemical-mechanical energies similar to the heat treatments (e.g., annealing), since they both increase the crystallinity and mechanical properties and therefore limit the transport properties. Lastly, we will discuss the ionomers as thin films at the solid interfaces and show the critical role of mechanical properties or energies in the confinement-driven changes in structure and properties of thin films. Thus, it is possible to enhance our state of understanding of ion-conductive polymers in electrochemical devices by studying the role of mechanics in the structure/property relationship of materials from nano- to continuum-scales.
Acknowledgement
This work was funded by the Assistant Secretary for Energy Efficiency and Renewable Energy, Fuel Cell Technologies Office, of the U. S. Department of Energy under contract number DE-AC02-05CH11231.

Although much research in thermoelectrics is centered around the material figure of merit zT, the device figure of merit ZT is often hugely determined by mechanical and thermal interfaces to the brittle thermoelectric material. Diffusion, sublimation, electro- and thermo-migration limit the lifetime and efficiency of commercial thermoelectric devices. The study of diffusion barriers, contact resistance and mitigation of thermomechanical stresses is rare considering the potential impact to devices. Here I will describe the issues with traditional methods for soldering and diffusion bonding Bi2Te3 and PbTe based thermoelectric materials for building modules; and recent microstructural analysis of these and related methods for variety of materials. Finally I will introduce some new concepts for thermoelectrics modules that utilize strong functionally graded designs, Thomson cooling and phase entropy thermoelectrics that will make control and understanding of thermo-mechanical considerations even more important to implement these novel devices.

In the paper, we present a study on the mechanical properties of a p-type half-Heusler thermoelectric material, using nanoindentation AFM methods. We provide a single crystal target for these measurements to be correlated with Electron Backscatter Diffraction methods for analysis in terms of dislocation behavior in response to the Berkovich indenter tip. EBSD analysis confirms the crystallographic influence on observed hardness and modulus variations.

The unique heat conduction phenomena of nanostructures is attributed to the convergence between material length scales and mean free paths of quantized lattice vibrations (phonons), enabling promising next-generation thermal transistors, thermal barriers, and thermoelectrics. Apart from size, elastic strain is also a potent modifier and is known to drastically affect heat transport when introduced locally via dislocations, impurities, or vacancies. However, the effect of strain has not been experimentally addressed due to the difficulty of isolating a single vacancy or dislocation or of simultaneously performing mechanical and thermal measurements at high strains.
We present a novel non-contact approach for measuring thermal conductivity of a single suspended nanowire subjected to varied strain levels, which we name Raman piezothermography. We apply uniaxial tensile stress to individual silicon nanowires under a confocal µ-Raman microscope. Raman maps across the length and width of the nanowire at several different laser intensities enable the deconvolution of the effects of stress and temperature on the Raman spectrum. Fits of the temperature profiles along the length of the nanowire combined with calculations of the shape-dependent laser power absorption yield measurements of thermal conductivity which we are able to correct for contact resistance and air conduction. Representative results of thermal conductivity as a function of stress for Si nanowires will be shown and discussed in the context of changes to phonon dispersion predicted by atomistic simulation. The role of strain hetereogeneity controlled either via external loading or by introducing lattice defects on phonon behavior will also be addressed.

For waste heat recovery applications, thermoelectric materials will be subjected to stresses arising from thermal transients and thermal gradients in addition to stresses due to mechanical vibration. Thus mechanical integrity, including fracture toughness is an important concern for the viability of thermoelectric devices. The thermoelectrics literature includes a large number of papers on nanocomposites in which the nanophase scatters phonons, thus improving the dimensionless figure of merit, ZT, by reducing thermal conductivity. In contrast, the literature is very limited on the topic of improving the fracture toughness via nanoparticle/nanoplatelets. This talk will consider mechanics-based experimental literature and theoretical models that deal with enhancing fracture toughness and evaluate them in terms of specific thermoelectric systems, including chalcogenides and skutterudites.

Thermoelectric materials are technically important for energy harvesting and conversion technology. Good thermoelectric materials must have high Seebeck coefficient, good electrical conductivity and low thermal conductivity. Incorporating nanoparticles into a thermoelectric compound matrix can effectively reduce thermal conductivity to a greater degree than electrical conductivity for ZT enhancement. Our recent studies revealed that nano-SiC dispersed Bi2Te3 polycrystalline materials show not only enhanced thermoelectric performance but also better mechanical strength and fracture toughness, which is practically favorable for device fabrication. By using such a kind of strengthened and fine-grained Bi2Te3-based materials prepared by spark plasma sintering (SPS), miniaturized thermoelectric (TE) modules were fabricated by combining mechanical cutting and photolithograph processes, whose power generation and cooling performances were confirmed.

With about two-thirds of all used energy being lost as waste heat there is a compelling need for higher-efficiency thermoelectric materials. PbSe and PbTe have gained importance as materials used in the mid-temperature (400-900 K) power generation range. The increase of the efficiency of thermoelectric performance is critically related to a lowering of their thermal conductivity (K_L). By means of state-of-art equilibrium molecular dynamics (EMD), we extensively investigated heat transport of several nano-structured PbSe/PbTe models, from alloys to complex nano-composites. Layered composites show a reduction of K_L of about 40% with respect to the bulk. The insertion of PbSe(Te) nano-particles into PbTe(Se) matrix reduces K_L up to 55 % while in alloys the reduction exceed 60 % in all temperature ranges. Further, the coexistence of nano-inclusions of different sizes and compositions is an additional means of controlling thermal transport properties as a function of nanocrystal spacing, which is capable of exceeding alloyed materials. This qualifies engineered nano-composites as outstanding thermoelectrics for the energy era. High-performance computational techniques prove their capabilities towards more efficient and performing materials.

In recent years, enhancements in the performance of III-V thermoelectrics have been demonstrated by embedding rare earth-arsenide nanoparticles in a semiconducting matrix. Prior work achieved a thermoelectric figure-of-merit (ZT) of 1.3 for n-type InGaAlAs doped with ErAs particles [1]. Above the solubility limit, ErAs precipicates out of III-As films and forms coherent interfaces with the underlying matrix. These particles improve ZT by contributing charge carriers as well as forming scattering centers for mid to long wavelength phonons. This raises electrical conductivity and reduces thermal conductivity [2]. However, there has not yet been a comprehensive comparison between the effects of traditional dopants (e.g. Si) and rare-earth dopants. In this study, we compare the thermoelectric properties of InGaAs with Si doping to Er, Sc, and Gd.
Thin films of InGaAs doped with either Si, Er, Sc, or Gd were grown by molecular beam epitaxy lattice matched to semi insulating InP (001) substrates. Samples of 0.01% to 1% rare earth-arsenide concentration were grown to explore doping effects below and above the solubility limit. This range also captures both the peak power factor and ZT for these structures. The electrical conductivity and Seebeck coefficient were measured as a function of temperature until 500K, when intrinsic substrate conduction becomes the dominant conduction mechanism. Thermal conductivity was measured at room temperature by time domain thermoreflectance.
Thermal conductivity results show no significant difference between choice of rare earth or Si doping until a carrier concentration of ~10^18 cm-3. Above 10^18 cm-3 Si concentration, the Weidemann Franz law leads to rapidly increasing thermal conductivity. Hall measurements show that as the rare earth-arsenide used gets closer to the lattice matched condition, more electrically active carriers are introduced into the matrix. Power factor scales with doping efficiency, and maximum power factor among rare earths is seen in Gd doped InGaAs. However, Si has a significantly higher doping efficiency, leading to higher room temperature power factors in Si doped InGaAs. Measurements up to 500K show that although Si doped InGaAs has a higher power factor, its conductivity decreases with temperature much more rapidly than rare earth doped InGaAs. Current trends suggest that substrate removal may show a crossover point where rare earth doped InGaAs becomes a more efficient thermoelectric due to energy enhanced conduction from nanoparticles.
[1] J. M. O. Zide, et. al., J. Appl. Phys. 108, 123702 (2010).
[2] W. Kim, et. al., Phys. Rev. Lett. 96, 1 (2006).

Thermoelectric materials, capable of scavenging electric power from sources of waste heat, are promising choices for solving the all-world energy crisis. Persistent efforts to enhance their efficiency (figure of merit, ZT) focus on increasing the electrical conductivity (σ) by carrier concentration tuning, enhancing the Seebeck coefficient (S) by band structure engineering, and lowering the thermal conductivity (κ) by atomic/nano/meso structures. However, the interdependence among S, σ and κ complicates efforts in developing strategies for improving a material's ZT. Up to now, no one can achieve an obvious concurrent improvement of electrical transport (high σ and S) and phone scattering (low κ) in one thermoelectric system.
In this contribution, inspiringly, we achieve an excellent thermoelectric performance (ZTasymp;2.3, 923 K) in K doped PbQ-PbQ&’(Q, Q&’=Te, S, Se) bulks by the contribution of carrier concentration tuning, band structure engineering and hierarchical architecture, hand-in-hand, among which, K plays series of significant roles. The system can be turned p-type by monovalent substitution of K on the Pb sub-lattice to enhance the σ. The K alloying PbQ/PbQ&’ modifies band structure to increase the Q&’ and suppress the bipolar effect to some extent. More importantly, K alloying strongly influences the density and morphology of nanoscaled precipitates in phase separation regions (PbQ and PbQ&’) at different levels. At 2.5% of K, cubical Nano-precipitates with {100} interfaces exhibit widely, especially in PbQ&’ regions. Such anisotropy morphology should be attributed by the segregation of K at interfaces between precipitates and matrix. The coherent interfaces between regular-shaped nano-precipitates and the matrix cannot disturb electron transport, but scatter phonons effectively. Thus these nanoscaled regular-shaped precipitates combining with high density of grain/phase boundaries in mesoscale, and dislocation and strain along interfaces in atomic scale construct a hierarchical architecture to allow wide-range phonon scattering effectively. Hopefully, this synthetic way to concurrently improve electron transport and phone scattering highlights a realistic prospect of wide thermoelectric application with high ZT (above 3).

Perovskite mixed conductors have long been explored as SOFC electrode materials due to their high surface activity for exchange of lattice oxygen, and high rates of bulk ionic transport. However, it has proven difficult to quantitatively separate the roles of kinetics, transport, and microstructure in determining overall kinetics.
Recently workers have employed thin films or other idealized microstructures to isolate kinetics from other effects. While these measurements have proven quite useful, their interpretation has remained somewhat challenging - in many cases generating as many questions as they have answered.
This talk will focus on the role that material thermodynamics plays in the measured kinetics, and how thermodynamic properties of the film may vary from one material to another based on its exact composition and/or processing history. The talk will focus on past and recent measurements of the nonlinear O2 exchange rate law on thin films of La1-xSrxCoO3-δ (LSC), as well as porous LSC and La1-xCaxFeO3-δ (LCF), and the modeling of that rate law in terms of underlying elementary kinetics.

Mechanical properties, e.g. Young&’s modulus, shear modulus, fracture strength, and non-elastic deformation were measured for SOFC materials at elevated temperatures under controlled gas atmospheres. Young&’s moduli of stabilized zirconia by a resonance method showed minimum at intermediate temperatures range where the internal friction showed peeks suggesting local reorientation of oxygen vacancies. Similar softening observed with cathode materials (La,Sr)(Co,Fe)O3 was attributed to anelastic deformation of the rhombohedral lattice. A large jump of Young&’s modulus was found with increasing temperature across the phase transition from rhombohedral to cubic. Nickel cermet anodes showed ductility at temperatures above 400C in reducing atmospheres. Transient creep followed by persistent creep was formulated as a function of stress and time. The observed mechanical properties were incorporated into the calculation of stress and deformation of SOFC. Oxygen potential profile was first calculated in the whole cell not only in gaseous phase but also inside the solid phase using mass and electron transport parameters which were taken from literatures and from our own experiments. Then, the stress and strain were calculated based on the thermal and chemical expansion coefficients. Microstructures of the porous electrodes were taken into consideration using homogenization technique. In order to verify the results of the calculation, experiments were carried out using single cells or stacks. Since SOFC under development has various types of cell/stack configurations, special simulation cells of different types were provided. Acoustic emission technique was employed to detect the initiation and propagation of cracks during operation.

Solid oxide fuel cells have emerged as a promising electrochemical conversion technology due to its high efficiency, stability, and low cost. A major component of these fuel cells is its electrolyte, a solid ion-conducting ceramic, which must be ionically conductive but electrically insulating. During operation, chemical expansion inside the electrolyte can lead to cracking and other adverse effects. In this work, electrochemical strain microscopy, a nanoscale probing technique in which an AC bias is applied to the surface of the electrolyte material, is used to evaluate induced strains on samarium doped ceria, and the corresponding electrochemistry on the surface is evaluated. Due to the typically high operating temperatures of solid oxide fuel cells, electrochemical strain microscopy was performed at varying temperatures, up to 250°C. We find that the induced strain increases proportionally with increased temperature, particularly at grain boundaries due to the space-charge effect. Furthermore, large DC biases were applied to the electrolyte at individual points, and the resulting strain relaxation after removal of the bias was measured at varying temperatures. Here, we find that environmental conditions can lead to dramatic differences in the relaxation behavior. Determination of the mechanisms behind such changes can lead to increased fundamental understanding of the chemical expansion behavior of ceramic electrolytes.

Thin films are generating increasing interest in electrochemical applications such as solid oxide fuel cells, gas sensors, and microbatteries. Recent observations indicate that defect thermodynamics in thin films lead to behavior which differs from analogous bulk materials. One important factor here is strong coupling between defect incorporation and molar volume of the solid, often referred to as chemical expansion. This effect is particularly relevant in ceria electrolytes for solid oxide fuel cells, where variations in the oxygen stoichiometry can lead to large stresses in thin films, in response to constraints from the underlying substrate. To measure these stresses, in situ wafer curvature measurements were conducted on Pr doped ceria films during oxidation-reduction cycling, over a range of temperatures and oxygen partial pressures. Corresponding strains were measured with high-temperature x-ray diffraction, under the same conditions. The stress state at a given temperature and oxygen potential was varied by using films deposited on different substrates to induce different thermal expansion mismatch stresses. Measurements on films with different grain sizes were also used to provide information about grain boundary effects. The coupling of these measurements allows interpretation of both elastic and inelastic properties. These, in turn, are compared with previously measured defect formation energetics, based largely on detailed dilatometry data for bulk ceria. Deviations between thin film and bulk behavior are discussed in terms of defect formation mechanisms, and the impact of elastic strain energies on the relevant thermodynamics.

Metal-supported solid oxide fuel cells offer potential advantages over conventional ceramic-supported cells such as reduced material cost and improved mechanical robustness. Mechanical properties of the porous metal support at high temperature need to be assessed and optimized for ensuring fail-safe design. This work investigates creep properties of a porous ferritic stainless steel at high temperature in hydrogen gas atmosphere by numerical modeling and experimental characterization. Creep rates of dense and porous metal supports are characterized by creep tests for stress ranging from 3 MPa to 15 MPa at 973 K. Creep rate of the porous metal support is computed by finite element homogenization of three-dimensional microstructural reconstructions from X-ray tomography and the measured creep rate of the solid material. A good agreement between computed and measured creep rate of the porous metal support is obtained. This suggests that the effect of porosity on creep rate can be captured with numerical microstructural modeling.

Li-ion batteries (LIBs) are the most-commonly used power source for portable electronics due to the advantages such as high voltage, non-memory effect, small volume, less self-discharging, light weight and so on. In LIBs, the function of the electrolyte is to provide Li-ions transport paths between the electrodes; also, it must be able to maintain the electrochemical stability during the charge/discharge processes. As the Li-ions transport properties and stability of electrolyte play important roles in the capacity, charge/discharge rate and cycling stability of the LIBs, it is therefore important to study the details of Li-ion transportation and the electrochemical reaction mechanisms of the electrolyte materials. Most importantly, the increased safety concerns in LIBs need to develop new, safe and reliable electrolyte materials, and solid electrolyte materials are therefore the obvious choices. Recent works have been demonstrated that the cubic garnet framework oxide Li7La3Zr2O12 (LLZO) has high ionic conductivity of about 2x10-4 S/cm and good thermal and chemical stabilities and can achieve the high density (96%) and conductivity (7.4x10-4 S/cm) by optimization of the sintering parameters. In this study, we have applied the newly-developed Electrochemical Strain Microscopy (ESM) technique to conduct in-situ study of the ionic transportation, volume deformation and surface strain of the solid electrolyte Li6.75La3Zr1.75Ta0.25O12 (LLZTO). ESM technique allows the high frequency periodic bias to be applied on the sample surface of the solid electrolyte, and the bias will induce the local periodic oscillatory displacement due to the Li-ions transportation and redistribution within the material. The local surface displacement, which is defined as electrochemical strain, can be detected by a highly sensitive photodetector in ESM. This study can provide detail understanding of the Li-ion transport mechanisms and electrochemical reaction mechanisms at nanoscale within the solid electrolyte material, accordingly benefitting the Li-ionic conductivity enhancement of electrolyte and the electrochemical performance improvement of LIBs.

Intensively growing Li-ion battery research area place high demands in the establishing of the long- and short-range structure of the electrodes after or during Li intercalation. Low crystallinity of the electrode materials and weak scattering of the X-rays by Li atoms make these systems crystallographically challenging. To understand the processes occurring in tin pnictide electrodes we have synthesized ternary Li-Sn-Pn (Pn = P, As) compounds. We hypothesize that the formation of Li_1-xSn_2+xPn₂ is the first stage of Li intercalation into Sn₄Pn₃. To fully characterize the long-range and local structure were applied a combination of powder and single crystal X-ray diffraction, ⁷Li NMR and ¹¹⁹Sn Mössbauer spectroscopy, as well as high resolution TEM. Observed local ordering of Sn and Li atoms was further probed by quantum-chemical calculations which indicate strong anisotropy of the properties of Li_1-xSn_2+xPn₂. Investigation of the resistivity and thermal conductivity performed on the large single crystals confirmed anisotropic transport properties of Li_1-xSn_2+xPn₂.

Tremendous thermal energy exists in low-grade heat sources (<100 °C), so high-efficiency and low-cost thermal energy harvesting system is highly demanded to utilize them. Current technologies, such as thermoelectric device, still have a low efficiency to satisfy commercial application. As an alternative choice, thermogalvanic cell, the electrochemical counterpart of thermoelectric device has benefit of higher electrochemical Seebeck coefficient low thermal conductivity, and material abundance. However, its efficiency is incomparable to thermoelectric system due to the low ionic conductivity. Another approach to utilize the high electrochemical Seeback coefficient is the thermally regenerative electrochemical cycle (TREC) aimed for harvesting high temperature thermal energy (>500 °C) half a century ago. In this approach, the electrochemical cell is discharged at T1 and recharged at a different temperature T2. The electrochemical Seeback effect induces lower charging voltage at T2 than the discharging voltage at T1, and thus net electricity is generated as the difference between the discharged and charged energy. This thermal-cycle-like design could lead high efficiency by decoupling electrical conduction and thermal conduction. However, large polarization of electrode materials of TREC has suppressed low temperature thermal energy (<100 °C). Here we demonstrate a new effective TREC to utilize low-grade heat by employing proper materials. The demonstrated system achieved very high efficiency at low temperature range. We believe that the demonstrated system is anticipated to be a good alternative choice of low cost and high efficiency thermal energy harvesting system from various low heat sources such as heat engine system, solar thermal, geothermal, and body heat.

Bismuth telluride based compounds have been considered as the best thermoelectric materials for waste heat recycling applications at low temperature regime. Many researchers have attempted to improve thermoelectric properties of bismuth telluride through microstructure engineering and compositional modulation. Among a variety of engineering approaches, nano-structured polycrystalline bismuth telluride appears to be the mainstream route in recent studies because it possesses low thermal conductivity and good mechanical properties. This study intends to investigate the interactions between process induced crystal defects and silver dopants in Bi-Se-Te polycrystalline systems. Different amount of nanosized silver powders were added during milling of Bi-Se-Te compounds. By increasing the Ag composition from 0.2 wt% to 3 wt%, the electron concentration decreases from 5.5×10^19 /cm^3 to 1.1×10^19 /cm^3, implying that Ag induced point defects act as electron acceptors. Because the solubility of silver in bismuth telluride is less than 1 wt%, excess silver may segregate at the grain boundaries. Our preliminary results reveal that the Ag-rich precipitates may reduce the thermal conductivity. The dependence of thermoelectric transport properties on doping efficiency and segregation of Ag in Bi-Se-Te compounds will be the subject of interests.

The deposition of transition metal layers on silicon and their reaction with substrate are important issues in semiconductor device technology. The interface between metal and semiconductor determines the device performance. The 3d transition metal monosilicides such as FeSi, CoSi, MnSi and CrSi have attracted much attention because they are easily formed in the interface between transition metal and Si. On the other hand, the Mn4Si7 compound is well known a pseudo-direct band gap semiconductor (0.42 ~ 0.98 eV) with a fundamental gap increasing linearly with the compression along c- or a-axis. We have grown Mn thin films on Si (111) substrates at 600 oC using MBE, resulting in the formation of Mn4Si7. In order to investigate the correlation between magnetization and charge carrier transport, we performed magnetoresistance and Hall resistance measurements by using a physical property measurement system. Interestingly, we observed the Seebeck coefficient of -565 mu;V/K and electrical resistivity of 2.26 mOmega; cm in Mn4Si7 films grown on Si substrate, resulting in the power factor of 14 mW/K2m.

Thermoelectric materials, which convert heat into electrical energy, may provide a solution to increased energy efficiency through waste heat recovery. To date, most commercial thermoelectric materials are too inefficient to be a viable option for most waste heat applications. This research proposes to investigate the fabrication and characterization of nanostructured InP to increase the performance of existing thermoelectric technology.
In order to improve thermoelectric material efficiency, either the lattice thermal conductivity must be lowered or the thermoelectric power factor must be increased. This research will focus on the latter by modifying the density of states of the semiconductor material and studying the effect of quantum confinement on the material&’s thermoelectric properties. Using focused ion beam milling, nanostructured cantilevers are fabricated from single crystal InP wafers. An all-around Al2O3 gate dielectric and Pt electrode are deposited to create a depletion region along the outer core of the cantilever, thus creating an inner conductive core. Both the electrical conductivity and the Seebeck coefficient are then measured as a function of confinement by varying the gate voltage. Measurements are also taken at various temperatures. This technique can be applied to a variety of material systems to investigate the effects of confinement on their thermoelectric properties.

The recent focus in energy harvesting using thermolectric devices and thermal management in nanostructures has motivated the interest towards understanding the role of phononic thermal transport in these nanoscale materials. A detailed understanding of the role and behavior of phonons in confined structures can lead us to the design of nanostructured materials with tailored thermal transport properties. Control of thermal conductivity based on phonon engineering in Earth abundant and cheap materials has already shown to be a promising path to viable thermoelectric (TE) devices with good performances [1]. One way to obtain thermoelectric systems with improved efficiency is to engineer nanostructured semiconductors, so as to reduce the thermal conductivity of the crystalline materials while preserving their electronic properties. Recent experiments have reported a strong reduction in the group velocities of phonons, thereby leading to a strong reduction in the thermal conductivity in sub-10 nm free-standing Si membranes [2].
Our study is driven towards understanding the nature of phononic thermal transport in nanostructured silicon membranes. We use harmonic lattice dynamics (HLD) and classical molecular dynamics (MD) to compute the phonon transport properties in silicon membranes, with thickness up to ~20 nm. We show that dimensionality reduction has a significant effect on the vibrational properties of the membranes and leads to a 4-fold reduction in the thermal conductivity of the membranes. Combining dimensional reduction with surface modification, by means of pattern formation and surface oxidation, we obtain a reduction in the thermal conductivity of the membranes to a factor of 20 with respect to the bulk, implying a 20-fold enhancement of the thermoelectric figure of merit at room temperature. Such figures make nanostructured silicon membranes viable materials for thermoelectric units.
[1] M. S. Dresselhaus et al, “New directions for low-dimensional thermoelectric materials,” Adv. Mater., 22, 3970 (2010).
[2] J. Cuffe et al, “Phonons in slow motion: dispersion relations in ultrathin Si membranes.,” Nano Lett., 12, 3569-3573, (2012).
Acknowledgment: This project is funded by the program FP7-ENERGY-2012-1-2STAGE under contract number 309150.

Lower dimensional structures can show a significant modification of transport properties due to quantization of the carrier energy in one or more directions. 20 years ago Hicks and Dresselhaus predicted that these effects could enhance the thermoelectric properties of 2D and 1D structures¹. Additionally an increased phonon boundary scattering has been predicted and found in structures with sizes comparable to the mean free path. This has led to an increase in the research on thermal transport in nanowires, and a significant decrease of the thermal conductivity was found with respect to the bulk values in e.g. silicon nanowires^2,3. In this work, we report on our investigations of the thermoelectric properties of single semiconductor nanowires.
The nanowires are grown using the Vapor Liquid Solid (VLS) method in a Metal Organic Vapor Phase Epitaxy (MOVPE) reactor. This gives complete freedom over wire length, crystal symmetry, diameter, radial and axial composition etc^4,5. In this way, the dependence of the phonon heat transfer on these parameters can be studied systematically.
The experimental study of the thermoelectric properties of the nanowires is done using suspended SiNx membranes with implemented heaters as previously used by Li Shi et. al.⁶. Two 20x20 pads or thermally isolated from the environment by suspending them using 450µm long SiNx beams. By placing a nanowire over gap between two pads, the wire conducts the heat generated by joule heating on one membrane to the other side. With platinum meanders (for heating and thermometry) on both platforms, heat conduction of the nanowire is investigated. Electrodes on both pads enable electrical characterization of single wires to study all thermoelectric properties on a single semiconductor nanowire.
References
1 L.D. Hicks, and M.S. Dresselhaus, Phys. Rev. B., 47, 16631 (1993)
2 A.I. Hochbaum, R. Chen, R.D. Delgado, W. Liang, E.C. Garnett, M. Najarian, A. Majumdar, and P. Yang, Nature 451, 163-167 (2008)
3 A.I. Boukai, Y. Bunimovich, J. Tahir-Kheli, J. Yu, W.A. Goddard III, and J.R Heath, Nature 451, 168-171 (2008)
4 S.R. Plissard, D.R. Slapak, M.A. Verheijen, M. Hocevar, G.W.G. Immink, I. van Weperen, S. Nadj-Perge, S.M. Frolov, L.P. Kouwenhoven, and E.P.A.M Bakkers, Nano Lett 12, 1794-1798 (2012)
5 R.E. Algra, M.A. Verheijen, M.T. Borgström, L Feiner, G. Immink, W.J.P van Enckevort, E. Vlieg, and E.P.A.M. Bakkers, Nature 456, 369-372 (2008)
6 L. Shi, D. Li, C. Yu, W. Jang, D. Kim, Z. Yao, P. Kim, and A. Majumdar, J. Heat Transfer 125, 881 (2003)

Most energy conversion devices today are based on the thermodynamic cycle, in which around two-thirds of energy is wasted in the form of heat. Thermoelectric devices can be used to recover this waste heat, although they have limited conversion efficiency or, in other words, a low thermoelectric figure of merit, zT so far. In this talk, we systematically investigate three different routes of synthesizing 2% Na-doped PbTe- and demonstrate a zT of ~ 2.0 at 773 K in one of the methods, which is the highest reported zT at this temperature. We found that the grain sizes and precipitate sizes varied significantly among different synthesis methods, which directly affected the thermoelectric properties of the materials based on our theoretical analysis. Also, these various morphologies led to different mechanical hardness of the materials; the material with the highest zT, i.e., zT ~ 2.0 at 773 K, exhibited a factor of around two times higher mechanical hardness than the other two materials. Overall, the combination of high zT and high mechanical hardness of the sample should enable its use in improved devices for recovering waste heat.

Bulk nanostructuring of silicon has been proven to be an effective method of increasing the thermoelectric figure merit, leading to increases of up to 250% over single crystal Si (at 1275 K). This large increase is due in part to an order of magnitude reduction in the lattice thermal conductivity while maintaining relatively high carrier mobility, thus making nanostructured Si a contender for high temperature thermoelectric applications. While significant gains have been achieved, further improvements in the thermal-to-electric conversion efficiency are necessary for large scale applications such as high grade waste heat recovery. Modeling suggests that the figure of merit can be further enhanced via a nanocomposite approach through the use of small (3-5 nm) inclusions. The inclusions are predicted to not only further reduce the lattice thermal conductivity but also enhance the Seebeck coefficient via carrier injection. In this presentation, an overview of the nanostructured Si work that has been done at Jet Propulsion Laboratory will be presented and discussed along with new strategies and methods to try to improve the thermoelectric figure of merit of Si at 1275 K.

The misfit layered calcium cobalt oxide Ca3Co4O9 is considered among the best of oxide thermoelectric materials for high temperature applications, because of its high Seebeck coefficient coupled with relatively high electrical conductivity. Ca3Co4O9 is often represented as [Ca2CoO3]xCoO2 (x ~ 0.62) consists of alternative stack of rock-salt type Ca2CoO3 layer, serving as blocking layer for phonon, and CdI2-type CoO2 layer, serving as conducting channel for charge carriers, along the c-axis. Such layered structure demands the growth of highly textured and epitaxial thin film for better thermoelectric performance. In spite of the fact that extensive studies have been performed on Ca3Co4O9 in bulk, investigations on thin films are limited. In this study, we report an approach for the growth of epitaxial misfit layered cobalt oxide Ca3Co4O9 on Al2O3(0001) substrate by co-sputtering from Ca and Co-targets by reactive rf-magnetron sputtering in presence of 1.5 % O2 followed by post-deposition annealing. X-ray diffraction analyses reveal as-deposited films to be of mixture of CaO and CoOx phases. After annealing at 975 K in O2-atmosphere highly crystalline phase of c-axis-oriented Ca3Co4O9 films are obtained. The epitaxial nature of the films is determined from the X-ray diffraction pole figure analyses. For the systematic study several samples with different relative concentration of Ca and Co have been prepared. The variation of elemental ratio Ca/Co is found to influence the growth of the films during the annealing process. The samples with Ca/Co ~ 0.75 are found to be nearly phase pure and highly textured resulting in their relatively high electrical conductivity. On the other hand, many spurious phases along with Ca3Co4O9-phase are found in Ca and Co-rich samples resulting in their lower electrical conductivity. The room temperature value of electrical resistivity and Seebeck coefficient of the best sample is measured to be 6.44 × 10-3 mOhm-cm and 118 microV/K, respectively. Electrical conductivity of the same sample is not found to decrease as a function of temperature until 550 K, which results in high power factor (above 2 × 10-4 W/mK2) throughout the temperature range of operation. The best value of power factor is calculated to be 3 × 10-4 W/mK2 at 625 K. The structural properties of all the films have been studied by transmission electron microscopic (TEM) analyses and the structure-property relationship of them has been established.

Understanding the role of interfaces in thermal transport across nanoscale superlattices is important to the design of thermoelectric materials and devices with improved efficiency and for building efficient thermal management technologies for nano- and optoelectronic devices. Nitride metal/semiconductor superlattices are not only promising for achieving highly efficient thermoelectric devices, they also serve as a model system where fundamental physical processes involving thermal conduction can be tested and understood. Here we present a detailed analysis of thermal conduction in lattice-matched high-quality TiN/(Al,Sc)N metal/dielectric superlattices.
Epitaxial and coherent TiN/Al0.72Sc0.28N superlattices with period thickness (a) ranging from 0.8-30 nm are deposited on (001) MgO substrates by reactive dc magnetron sputtering. High-resolution XRD analysis along with reciprocal space x-ray mapping indicate that the superlattices are pseudomorphic and grow with 002 orientation. The X-ray reflectivity (XRR) analysis suggest that the superlattice interfaces are extremely sharp with interface roughness of the order of one to two atomic layers. High resolution transmission electron microscopy (TEM) along with HAADF-STEM indicates cube-on-cube epitaxial superlattice growth with a very low density of extended defects.
The time domain thermo reflectance (TDTR) measurements are used to characterize thermal transport in the superlattices. The thermal conductivity undergoes a minimum at a period thickness of 4nm corresponding to 4.5 W/m-K. The corresponding thermal interface conductance are the highest for any known interface with G > 2 GW/m2-K for a<4nm at room temperature. The measured conductance increases linearly as a function of temperature for a=10nm, and 20nm and exceeds predictions of the diffuse mismatch model (DMM) by 5-6 fold even when full phonon dispersion relations are considered. An electron-phonon coupling model within the DMM framework is developed to explain the high thermal conductance at the interface.

Classical understanding of otherwise simple systems often breaks down when taken to the nanoscale. For example, theory predicts that a “negative temperature gradient” can be established utilizing coherent reflection and interference of thermal carriers via boundary scattering and ballistic transport. While this spontaneous flow of heat from a cold region to a warmer region would typically be seen as a violation of the second law of thermodynamics, manipulation of sample geometry and carrier excitation conditions can cause reflection of thermal energy and create a negative temperature gradient without and thermodynamic contradictions. In this work, we experimentally demonstrate the ability to create a reflected energy flux that opposes the temperature gradient from a thermal source by applying a frequency dependent temperature gradient in a system with physical dimensions that limit thermal wave penetration from the source.
Time-domain thermoreflectance experiments on micrometer thick, free standing silicon with an aluminum thermal transducer allows us to determine the various cross plane thermal conductivities of the silicon films of different thicknesses; when the silicon film is thin enough, the thermal penetration depth is greater than the thickness of the silicon. We show that the thermal conductivity decreases with an increasing modulation frequency, indicating that as thermal penetration depth exceeds sample thickness, the generated thermal wave is reflected off the back boundary of the silicon. Since the thickness of the sample is less than the mean free path of part of the phonon spectrum in silicon, phonon emission from the backside interface causes a thermal flux of phonons that opposes the flux from the front of the film. The resulting reduced net flux leads to a lower thermal conductivity in the thin silicon film. Although we show a “negative” thermal flux, this is not in violation of the second law of thermodynamics when discussed in terms of boundary induced ballistic energy transport from silicon phonons. These findings are also used to explain some recent experimental data in related literature.

A discussion of synthesis and characterization of bulk nanocrystalline silicon with grain sizes of around 20 nm and thermal conductivities as small as 100 mW/cmK at room temperature, will be presented. Nanostructured materials have great potential for thermoelectric applications because of the reduction in thermal conductivity due to phonon scattering at grain boundaries [1] and silicon is a well-understood, cheap, earth-abundant material. Other silicon nanostructures, such as nanowires [2], are being investigated as viable thermoelectric materials. We have used, for the first time, the combination of a non-thermal plasma process for the synthesis of silicon nanocrystals with hot pressing to produce bulk nanostructured silicon samples. The non-thermal plasma synthetic route has been proposed for the production of photo-luminescent silicon quantum dots with narrow size distribution (3+/-0.5 nm) [3]. The same reactor has been scaled up to produce silicon nanocrystals at a rate of hundreds of milligrams per hour. Silicon powder with sizes between 5 nm and 15 nm has been produced using either silane (SiH4) or silicon tetrachloride (SiCl4), which are low-cost silicon precursors. Results have shown surface termination of the non-thermal plasma synthesized particle, i.e. H or Cl, play a role in densification kinetics. Hot pressing is a high pressure, high temperature process that allows for the production of samples with bulk like densities while limiting grain growth. In this study we have produced bulk (12 mm diameter, 2-4 mm in thickness) samples of nanocrystalline silicon with relative densities exceeding 90%. Characterization by XRD and TEM confirms that grain sizes are around 20 nm. The effects of surface termination of nano-silicon on grain growth and grain boundary conditions will be extensively discussed.
1. Dresselhaus, M.S., et al., Advanced Materials, 2007. 19(8): p. 1043-1053.
2. Hochbaum, A.I., et al., Nature, 2008. 451(7175): p. 163-U5.
3. Mangolini, L., et al., Nano Letters, 2005. 5(4): p. 655-659.

We report the first room temperature experimental observation of Peltier cooling in nonlinear thermoelectric microrefrigerators. Nonlinear (current-dependent) Peltier coefficient of low-doped n-type InGaAs and its associated Peltier cooling are extracted from four-point probe electrical and thermal imaging characterization techniques.
The idea of utilizing current-dependent peltier coefficient for cooling application was first proposed by Mona Zebarjadi, et. al. in [1]. At high electric fields the linear relation between Peltier cooling/heating with current is no longer valid. Peltier coefficient depends nonlinearly on current and rises significantly as current increases. Nonlinear Peltier coefficient was calculated using Monte Carlo simulation. It is indicated that this coefficient is proportional to the effective mass and inversely proportional to the square of carrier concentration of a semiconductor [1]. In an effort to measure the Peltier coefficient and Peltier cooling in nonlinear thermoelectrics, a set of nonlinear microrefrigerators are designed and fabricated. A 5µm thick low-doped n-type InGaAs was grown by the molecular beam epitaxy. A highly-doped InGaAs contact layer was grown on top of the active layer for low metal/semiconductor contact resistance. Devices of various sizes ranging from 10x10µm2 to 150x150µm2 were fabricated on the wafer by creating 0.8 um deep square mesas of different sizes using the inductive-coupled plasma etch. The devices act as Metal-Semiconductor-Metal structure due to presence of two back to back Schottky diodes. The I-V characteristic of the devices measured by four-point probe method and exhibited asymmetric forward and reverse behavior.
Thermal imaging technique is employed to obtain the temperature profile of the devices under forward and reverse biasing conditions at different current densities ranging from 100A/cm2 to 1500A/cm2. A hybrid analytical-numerical model based on the full heat balance equation in the structure is developed to extract the thermoelectric material properties. Using the proposed model, current-dependent Peltier coefficient of the film is extracted. Temperature changes due to Peltier cooling/heating at different interfaces and Joule heating in the device are obtained. We estimate 7-8 degrees Peltier cooling at a current density of about 1200A/cm2. Due to excessive Joule heating, the overall cooling performance of the device is not significant. We discuss design requirements for a nonlinear thermoelectric device to achieve large cooling.
[1] Zebarjadi M, Esfarjani K, Shakouri A, " Nonlinear Peltier Effect in Semiconductors". App. Phys. Lett. 912785154 (2007).

In this work, we propose a nanoscale three dimensional (3D) Si phononic crystal (PnC) with spherical pores, which can reduce the thermal conductivity of bulk Si by a factor up to 10,000 times at room temperature. Thermal conductivity of Si PnCs depends on the porosity, for example, the thermal conductivity of Si PnCs with porosity 50% is 300 times smaller than that of bulk Si. The phonon participation ratio spectra demonstrate that more phonons are localized as the porosity increases. The thermal conductivity is insensitive to the temperature changes from room temperature to 1100 K. The extreme-low thermal conductivity could lead to a larger value of ZT than unity as the periodic structure affects very little the electric conductivity.

Thermoelectric devices directly convert heat to electrical energy and are relevant for power generation and energy conservation. The efficiency of a thermoelectric devices is determined by the thermoelectric material's dimensionless figure of merit, ZT = (α2σ/(κL+κe))T, where α is the Seebeck coefficient, σ is the electrical conductivity, κL is the lattice thermal conductivity, and κe is the electronic thermal conductivity). The κe is directly related to the σ through the Wiedemann-Franz law: κe = LTσ, where L is the Lorenz number. High-ZT thermoelectric materials require a large α, high σ and low κL. One strategy for developing high-ZT thermoelectric materials is based on the phonon glass-electron crystal (PGEC) concept proposed by Slack [1]. A PGEC material acts as a glass with respect to phonon scattering (low κL) and as a crystal with respect to electron scattering (high α2σ).
Reduced molybdenum chalcogenides are promising candidates as PEGC thermoelectric materials, because of their large unit cell, heavy constituent atom masses, low electronegativity differences between the constituent atoms and large carrier mobilities. They form covalent structures with low coordination numbers for the constituent atoms and so can incorporate atoms in relatively large voids formed totally or partially filled up by heavy atoms, which show high atomic displacement parameters.
A high ZT was achieved in the AgxMo9Se11 (3.4

We propose a new method for accurately calculating electrical transport properties of a lightly-doped thermoelectric material from density functional theory (DFT) calculations, based on experimental data for the corresponding undoped material and density functional theory results for both undoped and doped materials. We employ this approach because hybrid DFT calculations are prohibitive for the large supercells required to model low dopant concentrations comparable to those achieved experimentally for high-performing thermoelectrics. Using zinc antimonide as our base material, we find that the electrical transport properties calculated with DFT and Boltzmann transport theory exhibit the same trends with changes in chemical potential as those computed with hybrid DFT, and propose a fitting algorithm to quantify this trend. We confirm the validity of this approach in reproducing the experimental trends in electrical conductivity and Seebeck coefficient versus temperature for Bi-doped β-Zn4Sb3. We then screen various transition metal cation dopants, including copper and nickel, and find that a Cu dopant concentration of 2.56% in Zn39Sb30 exhibited a 14% increase in the thermoelectric power factor for temperatures between 300-400 K. We thus propose that transition metal dopants may significantly improve the thermoelectric performance of the host material, compared to heavy and/or rare-earth dopants.

Thermoelectric properties on a continuous compositional design space involving cobalt-containing spinels and nanoscale inclusions of a binary oxide are estimated using a novel combination of density functional theory (DFT), cluster expansions, Markov Chain Monte Carlo (MCMC) and a quantum non-equilibrium Green's function (NEGF) transport model. DFT (with GGA+U exchange-correlation) is used to determine ground-state properties of the electronic structure at a variety of different compositions and configurations of cations over a continuous solid solution. Uncertainty in these calculations is represented by using various values for the Hubbard U. Cluster expansions for the total energy along with various properties of the electronic structure are then calibrated to the set of DFT data using a Bayesian formalism. Simultaneously, as a part of the same MCMC process used to calibrate the expansion coefficients, samples from the thermodynamic ensemble are generated. A transformation of the resulting joint distribution over the space of lattice configurations and expansion coefficients leads to a joint distribution for the ensemble averages of properties required for the transport calculation as a function of temperature and composition. The two-dimensional NEGF transport model, which considers electron-electrion, electron-phonon and phonon-phonon interactions, is then used to produce estimates of thermoelectric properties as a function of temperature and composition along with inclusion size and volume fraction. The compositionally and microstructurally dependent results thus obtained for the thermoelectric figure of merit also contain the propagated uncertainty arising from the DFT calculations and cluster expansion representation.

Automotive applications of thermoelectric generators (TEGs) for the conversion of the waste heat of combustion engines into electricity is a most timely issue. Among the manifold of “intermetallic” clathrates, hitherto three series of clathrate compounds have shown interesting thermoelectric properties: type I compounds such as EA8M16Ge30 and EA8MxGe46-x-y#127;y (EA=earth alkaline metal, M=transition element, #127; stands for a vacancy) and type VIII clathrates EA8Ga16{Ge,Sn}46. The presentation will summarize the systematic investigations of clathrate formation backed by DFT calculations (phase equilibria in isothermal sections, isopleths, liquidus projections), clathrate structures, bonding and structure-property relation in multicomponent type I clathrates including also an overview on mechanical properties. For the temperature region of 300 to 850 K the quinary Ba8NixZnyGe46-x-y-zSnz synthesized in our laboratory exhibits hitherto the highest average ZT-value achieved for an n-type clathrate I in polycrystalline bulk samples.

Advanced functionalized materials become progressively vital to our everyday life and indispensable to industrial engineering, energy supply and sustainability, healthcare, environment as well as life sciences. The development of novel materials with tailored macroscopic properties and optimized performance calls for a comprehensive understanding of chemical and physical principles underlying the material properties which can be only accomplished at the microscopic level through careful experimental, simulation, and theoretical studies. Neutron scattering experiments - probing the structure and dynamics on a nanometer and Tera-Hertz scales - and ab initio calculations - sampling microscopic characteristics by approximating the electronic structure - are the tools of choice to tackle this challenge.
The fundamental obstacle for a comprehensive study of advanced functionalized materials is their easy formation in polycrystalline form. Consequently, any deeper understanding of the complex physics requires dedicated computation of powder averaged signals. Moreover material functionality is determined and modified by the thermodynamic working conditions and dedicated simulations of pressure and temperature effects are required for a deeper understanding of the dynamics.
We present two examples of materials considered as high performance thermoelectric compounds, namely Fe-Sb- and Co-Sb-based skutterudites, and ZnSb/Zn4Sb3. We discuss the potential microscopic mechanisms of breaking the thermal transport in those classes of thermoelectric compounds. These mechanisms differ strongly as they originate on one hand from strong phonon anharmonicities and mass transport on the other. Our study is based on high-resolution elastic and inelastic neutron scattering experiments and the combination of density functional calculations comprising lattice dynamics and molecular dynamics techniques.

Thanks to their capability of decoupling electrical and thermal transport, nanostructured materials are crucial for engineering high-efficiency thermoelectric devices. Recent studies have shown a remarkably low thermal conductivity in Silicon nanostructures, including nanowires, thin films and nanoporous materials [1,2,3]. At the nanoscale, heat transport no longer obeys to Fourier&’s law and a model including non-equilibrium phonons is necessary. Phonon classical effects can be effectively described by the so called phonon suppression function, which represents the suppression of phonon transport with a given mean free path, caused by the scattering with nanostructure boundaries. We introduce a model based on the Boltzmann Transport Equation that solves for heat transport directly in terms of the local phonon suppression function. This formalism has several advantages. First, it requires in input only the bulk thermal accumulation function, which is a material property that can be directly obtained by experiments [4]. Second, it can be easily embedded in a multiscale model, by seamlessly incorporating the ballistic and diffusive regimes. Furthermore, it turns out to be much more computatationally convenient than the frequency-dependent approaches. We apply the developed method to nanoporous Silicon and find good agreement with experiments. The presented method could have great potential for the prediction of thermal transport of unexplored shapes of nanomaterials, and provide a better understanding of nanoscale heat transport.
[1] J.-H. Lee et al. NanoLetters 8, 3750 (2008).
[2] J-H. Lee et al. APL 91, 223110, (2007).
[3] Jen-Kan Yu et al. Nat. Nanotech, 5, 718 (2010).
[4] A. J. Minnich. Physical Review Letters, 109, 205901 (2012)

Thermoelectrics are materials of top priority in view of waste heat conversion into valuable energy. In practice however, they applicative use is still limited by their ZT figure of merit. While certain compositions, also including doping and nano-structuring can do for better thermoelectric compounds, a more efficient and promising paradigm of controlling phonon scattering is through a complete understanding of material morphology, and its consequences on thermal conductivity while length scales vary. Nanostructures are an effective way of scattering phonons with short mean free paths. However, better candidates are grains and domain boundaries, for they implement a natural way of size re-scaling over different lengths. This offers a hierarchical systematics for an effective, spectrum-covering phonon scattering. We present an integral computational approach to grain engineering on PbSe materials, using B1-B2 polymorphic transformations as a source of material structuring via domain formation. We show that “real” materials are in principle already tailored for improved thermoelectric properties. We understand our state of art computational setup, based on advanced molecular dynamics and heat flux calculations, as a design tool for thermoelectric materials.

Crystalline C₆₀ is an interesting candidate for thermoelectric (TE) applications due to its extremely low thermal conductivity (~0.4 W/mK) as well as potentially high electrical conductivity. Many theoretical and experimental studies have demonstrated that intercalation of alkali-metal atoms in crystalline C₆₀ leads to metallic behavior, resulting in enhanced electrical conductivity by 3 orders of magnitude. In such alkali-metal doped C₆₀ solids, the electronic properties can be broadly tailored depending on the dopants (alkali metals, alkaline earth metals, or combinations) and doping concentration. In this work, we investigate the thermoelectric transport properties of alkali-metal doped C₆₀ at 300K based on a combination of classical and quantum mechanical calculations. We study the effects of doping on the electronic transport in C₆₀ as a function of both dopant and doping concentration. Our calculations demonstrate that smaller lattice constants generate higher power factors (S²σ) due to a larger overlap between the C₆₀ energy states. Our calculations show that a decrease in the lattice constant strengthens the interaction between the C₆₀ units, making the half-filled t_1u band more dispersive, the electrical conductivity larger, and consequently power factor higher in spite of a smaller Seebeck coefficient arising from the smaller density of states at the Fermi level. In addition, varying the doping concentration gives rise to a change of lattice geometry (e.g., fcc-A₃C₆₀, bct-A₄C₆₀, bcc-A₆C₆₀) and electronic structure as well, suggesting another route to controlling the power factor. By optimizing the type and concentration of doping in crystalline C₆₀ in order to maximize the power factor, combined with its low thermal conductivity, our results suggest that alkali-metal doped C₆₀ could be an efficient thermoelectric material.

Nanocrystalline diamond foils manufactured by hot-filament chemical vapour deposition (HFCVD) are a very promising thermoelectric material.
Unlike microcrystalline diamond nanocrystalline diamond is a thermal insulator, because the phonons, which are responsible for the heat transmission are heavily scattered at the grain boundaries of the nanodiamond grains. Thus the heat transmission is clearly constricted, which leads to a heat conductivity of < 10 W/mK compared to about 2000 W/mK for microcrystalline diamond. Diamond is usually electrically isolating and thus actually useless as thermoelectric material. By doping with boron (p-type) a electrical conductivity of 40000 #8486;-1m-1 can be reached. With a seebeck-coefficient up to 300 µV/K (depending on doping and temperature) a ZT-value of up to 2-3 seems possible. This is a very good value compared to other thermoelectric materials like Si80Ge20-Alloys, which can be used at higher temperature. In addition to that diamond is the only material where the phonon-drag effect, which leads to Seebeck-coefficient of additional 2.500 µV/K occurs at temperatures above 0°C and thus can be used technically.
Another advantage is the chemical stability of diamond up to 500 °C when exposed to air and up to 1000 °C under vaccum or inert gas. Unlike other thermoelectric materials no rare and partially toxic raw materials are needed for the production of CVD nanodiamond. Apart from electric power only hydrogen, methane and boron for doping are necessary for the manufacturing process, which allows a good calculation of the production costs. Diamond can be bonded by high temperature brazing (> 850°C) leading to a high strength of the soldering joint and outstanding electrical conductivity, which allows the manufacturing of a diamond based thermoelectric generator. This generator already proves the applicability of doped nanocrystalline diamond foils as thermoelectric material.

Thermoelectric (TE) conversion of waste heat into useful electricity demands optimised thermal and electrical transport properties of the leg material over a wide temperature range. In order to gain a high figure of merit ZT as well as high thermal-electric conversion efficiency, various conditions of the starting material of p-type DD0.60Fe3CoSb12 (DD stands for didymium) were studied. After a rather fast reaction-melting technique, the bulk was crashed and sieved with various strainers in order to get particles below the respective mesh sizes, followed by ball-milling at three different conditions. The dependence of the thermoelectric properties (after hot pressing) on the nano-sized particles, grains, as well as crystallites of the powders was investigated. Optimised conditions resulted in a ZT ~ 1.3 at 775 K and an efficiency of h > 13% (300 - 850 K). In addition a review will be given on mechanical properties (elastic moduli, hardness, fracture toughness and plastic deformation) essential for TE device engineering.

Bismuth telluride based materials have been widely used as thermoelectric refrigerators and generators due to their good thermoelectric properties. In this study, p-type Bi-Sb-Te ingots with preferred growth direction were fabricated by zone-melting process. Two different types of samples with surface normal in the direction parallel and perpendicular to the long-axis of ingot, respectively, were cut from the ingot. The effect of crystallographic anisotropy on thermoelectric properties of Bi-Sb-Te compounds was investigated. In addition, acceptor-like impurities were thermally driven from one side of sample to form a graded doping profile in the Bi-Sb-Te compound. The graded doping profile would modify the band structure and lead to a gradient of carrier concentration in Bi-Sb-Te compounds. It is found that the graded Bi-Sb-Te sample possesses a greatly enhanced Seebeck coefficient when a thermal gradient is applied in the same direction of carrier concentration gradient. However, the Seebeck coefficient will be suppressed or change the sign when the thermal gradient is against the carrier concentration gradient. The effect of crystallographic texture on dopant diffusion profile and in turn thermoelectric transport properties will be discussed.

Solid-state thermoelectric (TE) devices have the ability to directly convert heat into electricity (power generation) or instead to use electricity to drive a heat flow (refrigeration). In physics, the effectiveness of a TE material is evaluated by its dimensionless figure of merit (ZT), defined as ZT=S^2*σT/κ, where S, σ, κ, and T represent Seebeck coefficient, electrical conductivity, thermal conductivity, and absolute temperature, respectively. Here κ can be further split into the lattice part (κL) and electronic part (κE). Under the pursuit of a high ZT in crystalline thermoelectric materials, nanostructured interfaces are introduced in bulk materials to reduce κL while maintaining S^2*σ [1]. Along a different line, a high ZT can also be achieved by largely improving S^2*σ of highly disordered glasses which exhibit the theoretical minimum κL of its composition, known as the Einstein&’s limit [2]. This κL minimum is reached when the phonon mean free paths decrease to half of the phonon wavelength and thus invalidate the wave description of lattice vibrations. In principle, however, the highly disordered atomic structure would also largely suppress the electron transport and thus lower σ as well. In addition to searching for effective doping to increase the carrier concentration, thermal annealing is also performed to improve the charge carrier mobility and thus σ. In particular, σ can be dramatically improved by forming nanocrystals during annealing. The optimized nanocrystal size is larger than electron mean free paths (~10 nm or less) to conserve σ but lower than phonon mean free paths (10-100 nm) to maintain a low κL. This may reach beyond nanostructured bulk materials hot pressed from nanoparticles, in which ~10 nm grain sizes are challenging to be achieved due to strong grain growth during hot press. As the first attempt, Te-based glasses (kL of 0.1-0.3 W/m K at room temperature [3, 4]) are annealed with in-situ TE property measurements to avoid overgrowth of nanocrystals. The in-situ measurement presented here can be of importance to general TE materials for both ZT improvement via annealing and understanding their property degradation during high-temperature operations, the latter of which is directly related to the lifetime of a TE device.
References:
1 Poudel et al, Science 320, 634-638 (2008).
2 A. Einstein, Ann. Phys. 35, 679 (1911).
3 S.-N. Zhang et al, J. Non-Cryst. Sol. 355, 79-83 (2009).
4 Lucas et al, J. Mater. Chem. A 1, 8917-8925 (2013).