Symposium EE7 : Mechanics of Energy Storage and Conversion---Batteries, Thermoelectrics and Fuel Cells

Lithium ion batteries have been used as a key component in portable electronic devices, and more importantly, they may offer a possible near-term solution for environment-friendly transportation and energy storage for renewable energies sources, such as solar and wind. In this talk, I will present an introduction to the combined application of first principles methods based on density functional theory (DFT), statistic mechanics, experiment and machine learning, etc, followed by an overview of the computation work aimed at designing better electrode materials. Specifically, we show how each relevant property is related to the structural component in the material that is computable, and we benchmark the computation results with experimental observations. Finally, we present some of the key challenges faced by researchers in the field.

There is a strong need in developing stretchable batteries that can accommodate stretchable or irregularly shaped applications including medical implants, wearable devices and stretchable electronics. There has been a fair amount of work exploring the development and performance of stretchable electrodes, but very little for stretchable electrolytes. Solid polymer electrolytes are ideal candidates for creating fully stretchable lithium ion batteries because of their mechanical and electrochemical stability, thin-film manufacturability and enhanced safety. However, the characteristics of ion conductivity of polymer electrolytes during tensile deformation are not well understood.

We have investigated the effects of tensile strain on the ion conductivity of thin-film polyethylene oxide (PEO) through an in situ study. The results of this investigation demonstrate that both in-plane and through-plane ion conductivities of PEO undergo steady and linear growth with respect to the tensile strain. This increasing trend in conductivity infers that the structural changes induced in the polymer electrolyte results in altered and improved ion conduction. The coefficients of strain-dependent ion conductivity enhancement (CSDICE) for in-plane and through-plane conduction were found to be 28.5 and 27.2, respectively.

We hypothesize that the stretching and aligning of the amorphous polymer chains decreases the degree of tortuosity in the polymer, allowing for faster, and less obstructed ion transport. Semi-crystalline PEO consists of a crystalline phase and an amorphous phase. The amorphous phase is generally present along the edges of the crystallites where the polymer chains are disordered, twisted and entangled and tie one crystallite to another. The semi-crystalline conformation of PEO can be seen using polarization light microscopy, which confirms the growth and extension of the amorphous regions as the chains stretch and disentangle due to tensile strain.

Recently we have demonstrated that the Rashba effect that arises from the atomic spin-orbital coupling and inversion asymmetry can lead to a dimensionality reduction in the electronic density of states (DOS) [Phys. Rev. B 90, 195210 (2014) and Appl. Phys. Lett. 105, 202115 (2014)], therefore results in higher thermopower and better thermoelectric (TE) performance. A case in point is BiTeI, a three-dimensional material with a giant Rashba effect.

In this talk, we show that intercalation of small amounts of Cu dopants can substantially alter the equilibria of defect reactions, effectively mediate the donor-acceptor compensation, and tune the defect concentration in the carrier conductive network. Consequently, the potential fluctuations responsible for electron scattering are reduced and the carrier mobility in BiTeI can be enhanced by a factor of two to three. Cu-intercalation in BiTeI gives rise to higher power factors, slightly lower lattice thermal conductivities, and consequently improved TE figure of merit. These results demonstrate that defect equilibria mediated by selective doping in complex TE materials could be an effective approach to carrier mobility and TE performance optimization.

Solid oxide fuel cells are receiving increasing attention as a promising energy conversion technology. Not only do they convert chemical energy to electrical energy with high efficiency, but also produce much lower carbon dioxide emission. Electrolyte which is typically an ionically conducting but electrically insulating solid ceramic, is a major component of solid oxide fuel cells. Doped ceria, as compared to yttria-stabilized zirconia, shows higher ionic conductivities at lower temperatures, which potentially can reduce the fuel cells operating temperature to the range of 500-700 C. In this work, we use various scanning probe microscopy (SPM) techniques, including electrochemical strain microscopy (ESM) to study the nanoscale electrochemistry of doped ceria at elevated temperatures. SPM techniques allow us to correlate the electrochemistry with the surface topography, differentiating the behaviors of grain boundaries from the grains. Previous ESM experiments on doped ceria directly show that diffused space charge region in ceria does exist. Here we focus on extensively studying the space charge region by applying a series of DC pulses and then measuring the strain response after the DC pulses are removed, referred to relaxation study. Relaxation studies are performed locally point by point at different temperatures. With such systematic experiments, the nanoscale ion and electron activities can be probed.

Development of new anion-conducting polymeric materials has been an active area due to anion exchange membranes application in alkaline fuel cells, which demonstrated significant advantages of higher cell efficiencies and low cost. The current challenge is that high conductivity is often accompaied by significant increase in water uptake, leading to uncontrolled dimensional swelling or even disintegration of the membrane. We have developed a unique design of cellulose nanocrystals-based membrane with superior dimensional stability and great water uptake. Cellulose nanocrystals, produced by acid or base hydrolysis of cellulose-rich sources, are nanosized particulates with exceptional mechanical properties. The nanocellulose hydrogels are known to exhibit high water absorption while maintaining excellent dimentional stability. In this study, cellulose nanocrystals were incorporated with different commercially available polymeric binder systems to prepare electrolyte membrane for alkaline fuel cells. The influence of cellulose nanocrystals content, binder formulations, and temperature in water absorption, swelling, and hydroxide conductivity was systematically studied. The resulting membrane exhibited improved dimentional stability (< 10% swelling) and great water uptake (>100%) compared to the polymer binders. The presence of cellulose nanocrystals did not compromise the hydroxide conductivity of the polymer binder system. Compared to complex polymer synthesis route, the approach reported here is a facile and renewable strategy to prepared solid electrolyte with great dimensional stability and high hydroxide conductivity, thus opening up a new perspective on developing solid electrolyte for alkaline fuel cells.

Vertically aligned catalysts comprised of platinum-nickel thin films on nickel nanorods (designated as Pt-Ni@Ni-NR) with varying ratios of Pt to Ni in the thin film were prepared by magnetron sputtering and evaluated for their oxygen reduction reaction (ORR) activity. A glancing angle deposition (GLAD) technique was used to fabricate the Ni nanorods (NRs) and a small angle deposition (SAD) technique for growth of a thin conformal coating of Pt-Ni on the Ni-NRs. The Pt-Ni@Ni-NR structures were deposited on glassy carbon for evaluation of their ORR activity in aqueous acidic electrolyte using the rotating disk electrode technique. The Pt-Ni@Ni-NR catalysts showed superior area-specific and mass activities for ORR compared to Pt-Ni alloy nanorod catalysts prepared using the GLAD technique and compared to conventional high surface area Pt and Pt-Ni alloy nanoparticle catalysts.

The progress of fuel cell technologies lies in the development of a low-cost and efficient catalyst that can undergo oxygen reduction reaction (ORR). In our previous studies, we demonstrated that by using a nature materials Vitamin B12, Co-corrin, could replace platinum as a electrocatalyst in fuel cells, and that could lead to a new generation of emission free, low cost hydrogen fuel cells. However, ORR follows a four- electron pathway that makes it difficult for us to probe catalyst using conventional means. Furthermore, the ex-situ experiments performed an interesting results but are far from explaining why certain reaction pathway proceeded. Our recent study demonstrates the elucidate the mechanism of ORR of the electrocatalyst by using X-ray techniques under operating conditions ("in operando”). In the mechanism of ORR, in-situ K edge and L edge X-ray absorption spectroscopy combined with electrochemical impedance spectroscopy, give us powerful evidences to prove the mechanism of ORR that relies on the O2 connection with electrocatayst.

Zirconia ceramics cover a lot of high temperature applications because of its high temperature stability, chemical resistance and electrical and ionic conducting properties. It is employed for example also as high temperature solid oxide fuel cell (SOFC) or as oxygen sensor material.
The final ceramic parts are usually produced by sintering of a green body containing some amounts of organic binder. The organic binder burns off at around 200 to 400°C, followed by the sintering process above 1000°C. The burn-out and sintering parameters are influencing the quality of the ceramics.
Thermal Analysis methods belong to the classical investigations. With thermogravimetry (TG) coupled to evolved gas analysis methods (EGA) the binder amount and the evolved gases can be determined. With differential scanning calorimetry (DSC) the energetic effects and the specific heat can be analyzed.
The sintering and density change of the zirconia green body is traditionally studied by dilatometry. With the laser flash method the temperature diffusivity and the thermal conductivity of the sintered and the raw material can be determined.
This contribution will show the binder burnout and sintering results of a zirconia green body investigated by TG-DSC-EGA methods, dilatometry and laser flash analysis (LFA).

The United States generates over 80% of its primary energy from fossil fuels and rejects more than 60% of primary energy as heat. Harvesting waste heat has the potential to generate over 200 gigawatt of power, but traditional heat recovery methods, e.g. thermoelectrics and mechanical heat cycles, either require megawatt scales or lack efficiency. We present a new class of electrochemical heat engines for direct conversion of heat to electricity in continuous operation. Our design allows for scaling between kilowatt and megawatt range with independent modular optimization of heat transport, electrical transport, and effective Seebeck coefficient of the system. Specifically, we will present a laboratory demonstration using solid-oxide fuel cells, and system-level modeling focusing on efficiency at the maximum power point and scaling behavior. We show that power densities over 100 mW/cm² are achievable for high-temperature solid-oxide systems with efficiencies over 30% of Carnot at maximum power, and power densities over 10 mW/cm² are achievable in sub-100 °C systems harvesting low-grade heat. Finally, we will discuss directions for future materials research and mechanical systems optimization.

Direct methanol fuel cells (DMFC) typically use low concentrations of methanol in water as fuel so as to reduce methanol cross-over and consequent losses in efficiency. Here we have explored using storing relatively large amounts of methanol in polymer gels and allowing slow diffusion from the gels provide methanol to the DMFC. Gels containing 70-80 weight percent methanol would be able to serve the fuel cell longer than a 1-2 M solution of methanol in water. After finding that many polymeric hydrogels were not compatible with pure methanol, we discovered that gels could be made from 2-hydroxylethyl methacrylate (HEMA) cross-linked with ethylene glycol dimethacrylate (EGDMA) in methanol (70wt%). HEMA-EGDMA gels were synthesized using EGDMA concentrations of 1 mol%, 5 mol%, 10 mol%, 20 mol% and 25 mol% (relative to HEMA). The monomers in methanol were polymerized with AIBN (0.05 mol%) at 65 °C. Gelation times ranged between 8 hours (with 25 mol% EGDMA) and 18 hours (for 1mol% EGDMA). The gels were translucent with 10 mol % or less EGDMA and opaque with more than 10 mol% cross-linker. Methanol release rates from the gels in water, determined using in situ IR spectroscopy, ranged between 5-10 hours depending on the degree of crosslinking. Release rates were slowed over four-fold by coating the gels with a polysulfone membrane precipitated on the surface of the cylindrical gels. The gels were then used to provide methanol to power a DMFC fuel cell.

In order to better understand the capacity fade in Sn-based anodes a systematic post-mortem transmission and scanning electron microscopy study was performed after electrochemical cycling of SnS/C anodes. After eighty cycles the capacity had dropped to 350mAh/g and significant morphological changes had occurred. A unique observation is that the SnS particle size had increased after cycling suggesting coarsening; although the crystal size decreased after cycling they had significantly agglomerated. At the same time severe fracture was observed between the interface of the SnS particles with the C matrix and micro-cracks in the C were visible. These observations are in agreement with theoretical predictions regarding mechanical modeling of fracture in nanocomposite anodes.

Nanomaterilas have received extensive attention as electrode materials for lithium-ion batteries due to their superior electrochemical performances compared to their bulk counterparts. Such improvements benefit from not only the fast kinetics on the nanoscale, but also their distinct thermodynamics, which may lead to different (de)lithiation pathways from those in bulk materials. Unfortunately, revealing the reaction mechanism inside nanomaterials poses significant challenges on the in-situ characterization techniques that require both high spatial and temporal resolutions. Here, by conducting in-situ transmission electron microscopy observation, we show a unique multiple-stripe lithiation mechanism in SnO₂ nanowires, and provide direct evidence that lithiation of anatase TiO₂, previously long believed to follow a two-phase reaction path, switches to a single-phase one when the crystal size goes down to ~20 nm, therefore greatly improving the reaction rate. Such successful visualization of (de)lithiation pathways can provide important insights into the fundamental understanding of the strain accommodation and the particle-size-dependent performances of electrode materials.

Controlling the mechanics of materials has become critically important to improve the performance of many energy applications ranging from batteries, thermoelectrics, fuel cells, to thermal management. A clear and straightforward in-situ diagnostic method during the device normal operation process is highly desirable [1] and can elaborate the fundamental materials dynamics for superior performance. In this talk, we will present our recent progress in exploring the dynamic lattice and ionic mechanics using ultrafast optical spectroscopy and multi-scale modeling [2]. We use our recently developed approach to investigate two different mechanic regimes: the coherent lattice behaviors (i.e. phonons) in solid-state energy materials (e.g., thermoelectrics), and the ionic behaviors in electrochemical materials (e.g., battery electrodes). We will present these spectral mapping and in-situ mechanics results to show that such development enables a new fundamental understanding from the nanoscale. The impact of this study to rational design and improve energy efficiency in the future will also be discussed.


Reference:

[1] S. Chu and A. Majumdar, “Opportunities and Challenges for a sustainable energy future,” Nature 488, 294-303 (2012).

[2] Y. Hu, L. Zeng, A. Minnich, M. S. Dresselhaus, and G. Chen, “Spectral mapping of thermal conductivity through nanoscale ballistic transport,” Nature Nanotechnology 10, 701-706 (2015).

In this study, various scanning probe microscopy techniques, including biased-AFM, conductive-AFM, Electrochemical Strain Microscopy (ESM), bi-model AFM and AM-FM techniques, are used to in-situ characterize the changes in topography, ion-diffusion, conductivity, composition and elastic properties of Li-rich layered oxide cathode materials (Li_1.2Mn_0.54Ni_0.13Co_0.13O₂), in the form of both nanoparticles and thin film electrode. The diffusion coefficient of the material is also derived from the band-excitation ESM measurement, and the results are comparable with the macroscopic electrochemical measurements. The effects grain boundary on Li-ion diffusion are studied. The results show that the biased-AFM technique can be used to simulate the changes of the materials during the charge/discharge cycles. The results also confirm that the SPM techniques are ideal tools to study the changes of various properties of electrode materials during the battery performance at nano- to micro-scales. These techniques can also be used to in-situ characterize the electrochemical performances of other energy storage materials.

Variation of ionic concentration is typically associated with significant change in volume of the host compound. This strong coupling between ion concentration and lattice strain (Vegard strain) is exploited in the electrochemical strain microsopy (ESM) to probe the ionic transport. In this technique, a combination of AC and DC biases is applied to the scanning probe to locally induce redistribution of ions resulting in periodic deformation of the electrode surface beneath the tip. This deformation is measured by the same tip revealing the ionic transport activity with nanometer resolution providing detailed insight to the electrode material operation.

In this study, a thermodynamics-based model of the coupled concentration vitiation and mechanics involved in the formation of the ESM signal is implemented into the Comsol finite element platform. The ESM tip imposes electric field and local kinetics condition at the surface of the sample and the semi-infinite space beneath the tip imposes infinite boundary conditions. This finite element model is employed to investigate some aspects of the ESM probing technique, with realistic conditions applied through the tip.

We have recently demonstrated a strong correlation between electrochemical state and acoustic response of closed batteries. A physical truth that underlies all closed electrochemical energy systems is that they are, by design, reactors which redistribute density (mass within a given volume) as a function of charged passed in the ideal case. Our previous efforts have shown quantitative correlations between acoustic response and the state of charge and state of health of lithium ion and alkaline cells of various geometries

In this presented we explore specific physical changes that can be confirmed with this method, including phase changes, layer evolution, and electrolyte migration to name a few. Through the study the interplay between geometry and chemistry within electrochemical cell, we believe we can extract layer-by-layer changes in elastic modulus and density (both gradual and sudden) through modeling and experiments. These changes should then be relatable to phase changes. By examining “quasi-equilibrium” (very low current) as compared to “non-equilibrium” (very high current) cases in operando and in situ, we can explore mechanical and chemical changes simultaneously in complex cells, with models and experiments that have not been explored previously.

While acoustic methods are commonplace, within the battery field previous efforts have only measured gross changes (e.g. post delamination events) or specific failure modes of certain chemistries (e.g. acoustic emissions during cracking). Our proposed methodology provides insight into both state of charge and state of health, allowing battery companies and end users to track degradation and evolution of materials within cells well before failure, at the least providing a tool for predictive control of deposition parameters in full scale reactors. The results of the proposed research will allow companies to build models which link mechanical and electrochemical properties for full cells, and will be of use from materials processing. Thus far we have demonstrated never-before-studied correlations for batteries that have been sold in the billions.

Lithium-ion batteries are commonly used to power small electrical devices and are of increasing importance for grid scale applications where they can buffer the fluctuating energy supply and demand in the electrical grid. For these applications, the long term reliability is of very high importance. The repeated insertion and removal of lithium into/from the electrodes leads to inevitable mechanical stresses that interact with the electrochemical processes and affect the reliability of a battery. In order to investigate such mechanical effects, we have used in situ substrate curvature experiments and scanning electron microscopy. In this presentation, the mechanically induced degradation of a positive electrode will be exemplified using LiMn₂O₄. Here severe mechanical damage and capacity loss was found after 1000 cycles. The evolution of the damage as investigated by SEM will be reported. The degradation mechanism can be inferred from the combination of SEM and mechanical stress data. Mechanical effects can be particularly strong in high capacity materials for negative electrodes. Here, the mechanical stress data provided by substrate curvature measurements not only helps in assessing reliability but also can be used in conjunction with the electrochemical data to study reaction pathways. In this context, the lithium induced crystallization and amorphization of germanium electrodes were investigated using mechanical stress as a probe for the underlying electrochemical processes. In silicon and germanium film electrodes, mechanical stresses and plastic flow cause strong changes in electrode morphology. We will show how mechanical constraints can suppress such processes and improve the reliability of high capacity electrodes.

Battery systems are often difficult to characterize due to the numerous processes and reactions occurring inside them. Recently, mechanical measurements such as strain have been used to understand material evolution during battery usage, resulting in improved understanding of underlying physics in battery systems and better prediction of phenomenon such as lithium plating and capacity fade. However, many battery systems and models currently rely on voltage and current for characterizing materials and predicting aging, and the wealth of data available from expansion is underutilized as a tool for studying batteries. Here, we use a model battery with graphite and LCO electrodes to show that the derivatives of expansion reveal the same information as the voltage data, namely the locations of electrode stage transitions. Additionally, we show that these transitions are visible in the expansion data at practical charge rates significantly higher than those necessary to view transitions in the voltage data. We also provide a theoretical justification for the relationship between voltage and expansion. Until now, battery characterization has depended almost entirely on voltage and current. With the use of expansion, particularly the derivatives of expansion, we demonstrate a potentially more useful tool for characterizing batteries during operation and for use in future models that aim to completely describe battery systems.

Nanostructured materials are commonly in proposed next generation battery electrodes. The small size can accommodate the large volumetric changes that occur during charge-discharge cycles and prevent pulverization. Here we seek to understand how the nanosize in well-defined arrays impacts the morphological changes and electrochemical cycling performance using in operando x-ray measurements. For a model system, block copolymer templated ordered mesoporous nickel-cobalt metal oxide films were synthesized. These materials served as model electrodes to investigate structural changes or distortions during the insertion and desertion of lithium ions. At the mesoscale, grazing incident small angle X-ray scattering (GISAXS) is used to probe the ordered nanostructure, while simultaneously at the atomic scale, grazing incident X-ray diffraction (GIXD) probes the crystal lattice of the porous framework. These changes are subsequently correlated with its long-term cycle stability. Simply varying the template/precursor ratio or molecular weight of block copolymer templates can effectively control the pore size, wall thickness and transport path for Li ion. From GISAXS, both the structure and form factor are determined, which provides both the mesostructure and nanoparticle size/shape. This technique also provides some insights in the formation of solid electrolyte interphase during the initial discharge cycle. Combined these studies provide improved basic understanding of structure-property relations for porous Li ion battery electrodes and potentially can provide new engineering solutions for high performance batteries.

Traditional Li-ion battery models assume solid solution behavior of the intercalation host materials, but some common electrodes, such as LiFePO4 (LFP) and graphite, exhibit multiple stable lithium concentrations. When coherent phase separation occurs in single crystal nanoparticles, elasticity plays a major role in the dynamics. In LFP, striped phase patterns arise at low currents, which are suppressed at high discharge rates by strain-enhanced activation overpotential. The nucleation barrier for phase separation scales with the particle’s volume-to-surface ratio, due to the competition between the misfit strain energy (scaling with volume) and surface phase growth (scaling with area), triggered by lithium wetting or de-wetting on different crystal facets. These theoretical predictions and some recent experimental validations will be discussed, as well as analogous chemical phase separation phenomena in metal catalyst nanoparticles and patchy polymer colloids.

The development of high-energy storage devices is one of top most important research areas in recent years and rechargeable batteries are anticipated to be the primary sources of power for modern-day requirements. Lithium (Li) ion battery is one such rechargeable batteries that has been investigated because of their high energy density, no memory effect, reasonable life cycle, and one of the best energy-to-weight ratios and has applications in portable electronic devices, satellites, and potentially electric vehicles. Silicon is an attractive anode material being closely scrutinized for use in Li-ion batteries because of its highest-known theoretical charge capacity of 4,200 mAh/g. However, the development of Si-anode Li-ion batteries has lagged behind because of their large mechanical deformation, i.e., up to 400% volumetric change, during electrochemical reactions, which results in fracture, pulverization and early capacity fading. In other words, this coupled mechanics (e.g., volumetric change) and electrochemistry problem is the bottleneck on the development of Si anode Li-ion batteries. Therefore, a fundamental understanding of this coupled behavior of mechanics and electrochemistry will not only advance our knowledge on the failure of Si under lithiation, but also provide a basis to resolve this bottleneck in the development of the promising Si-anode Li-ion batteries.

In this presentation, we will report a systematic study by direct measurement of Li-Si composition using Auger Electron Spectroscopy (AES), micro-nano scale observation of Si expansion using Focus Ion Beam (FIB), and ex-situ measurement of young’s modulus and hardness by nanoindentation during a-Si film lithiation.

Slow Magnesium transport kinetics and large morphological transformations associated with charge-discharge steps currently limit energy densities and lifetimes of Mg²⁺ ion batteries. Among possible Magnesium intercalation materials, Sn is of great interest, having a high theoretical capacity, a relatively small associated volume expansion, and a low Mg²⁺ diffusion activation energy. However, pure Sn micropowder films electrodes have shown rapid capacity fading, prompting a search for alternative systems.
Here we use scanning transmission electron microscopy, quantitative energy-dispersive X-ray spectroscopy elemental mapping, and density functional theory (DFT) modeling to reveal how the changes in chemical composition of SnSb nanostructured anodes and the corresponding morphology and crystal structure transformations emerge from Mg intercalation and extraction during battery operation.
During the initial magnesiation stage, Mg intercalates the pristine 300-500 nm β-SnSb particles from the surface inward by occupying and migrating through interstitial lattice sites, inducing lattice expansion and crack formation. Once approximately 3/8 of the local interstitial sites are filled, Mg²⁺ begin to replace Sn atoms, thus forming Sb-rich (i.e., Sn-poor) grains. As the concentration of ejected Sn atoms grows, magnesiated Sn nanoparticles form near the surface of the parent SnSb particle. The dramatic changes in local lattice volume during this process cause the parent particle to break down into a stable network of compositionally segregated 33± 20 nm subparticles: Mg_1.2Sn and Mg_1.36Sn_0.2Sb_0.8.
During the demagnesiation stage, nearly all the Mg is extracted from the Mg_1.2Sn particles leaving pure Sn domains behind. In contrast, the Sb-rich domains retain most of it Mg atoms, where they occupy Sn lattice sites. Subsequent charge-discharge cycles involve minimal morphological and Sn-Sb compositional transformations with the Sn-rich domains being primarily responsible for reversible Mg storage and the Sb-rich domains being largely inactive.
Thus, not only the parent SnSb particles mechanically degrade during the battery conditioning stage but also a significant amount of Sb and Mg is sacrificed, which ultimately limits the achievable specific capacity of the SnSb anode system. Based on these results, we propose that small (<40 nm) pure-Sn nanoparticles would provide superior reversible storage capacity, Mg²⁺ transport kinetics, C-rate behavior, and lifetime without the need for electrochemical conditioning or sacrificial Sb and Mg.

L. R. Parent, Y. Cheng, P. V. Sushko, Y. Shao, J. Liu, C.-M. Wang, N. D. Browning, Nano Letters 15, 1177 (2015).
Y. Cheng, Y. Shao, L. R. Parent, M. L. Sushko, G. Li, P. V. Sushko, N. D. Browning, C.-M. Wang, J. Liu, Advanced Materials (in press).

Silicon and nickel-tin Li-ion batteries supported by a Ni inverse opal scaffold undergo large volume change during cyclic (dis)charging, which causes mismatch stresses and strains between the active anode (amorphous Si and Ni₃Sn₂) and the mechanically constraining Ni scaffold. These (de)lithiation-induced mismatch strains and stresses are modeled via sequentially coupled diffusion- and stress-based finite element (FE) analysis, which takes the mechanical contact between the two phases into account, as well as the complex geometry and material properties of the inverse opal anode system. During lithiation, compressive strains are developed in the Ni scaffold as the active layer expands. A rapid recovery of these lithiation-induced mismatch strains occurs during subsequent delithiation, though full recovery is not achieved. Strain histories upon multiple (de)lithiation cycles vary depending on the mechanical contact conditions employed between the two phases. The numerically predicted strains are compared with experimental strain data collected in operando using X-ray diffraction. The simulated strain histories agree with the measured data, demonstrating that predicting mechanical performance and eventual degradation may be possible via numerical modeling. The (de)lithiation-induced deformation behavior is very similar in both anode systems, even though the magnitude of mismatch strain in the silicon anode is larger due to the larger (de)lithiation-induced contraction/expansion.

We have recently developed a novel class of mechanical energy harvesters based on electrochemical reactions. The device utilizes the stress-composition coupling in electrochemically active materials, such as partially Li-alloyed Si or Ge. We demonstrated that mechanical bending induces different stress states in two identical partially Li-alloyed Si electrodes, which drives Li+ migration and generates electricity. Such electrochemically driven mechanical energy harvesting possesses certain advantages over other types. Compared to piezoelectric or triboelectric generators, our prototype demonstrates low internal impedance (~300Ω as opposed to ~100MΩ in the other two types) and continuous electric current for 3 seconds. We show from our modelling effort that the 3 seconds timescale in electric current is kinetically limited by Li diffusion inside the electrode material; this timescale is in principle tunable by engineering the diffusion length of Li inside the electrode. We demonstrate that this timescale tunability can be achieved by designing porous electrodes or thicker electrode films. Based on these findings, we will discuss strategies to develop an energy harvester optimized for specifically targeted low-frequency motion, such as human walking.

Low frequency (<10 Hz) mechanical loading presents challenges for traditional energy harvesting materials, which can require resonant and higher frequencies to meet minimal operational needs. In this work we propose a new class of systems for mechanical energy harvesting based on “off-the-shelf” Li-ion batteries. Li-ion batteries are known to exhibit changes in their electrochemical state due to applied mechanical loading and by taking advantage of this “piezoelectrochemical” phenomena, it is possible to construct a thermodynamic process that can harvest mechanical stress at low frequencies. In this presentation, we show that materials exhibiting this behavior are expected to exhibit orders of magnitude higher energy density per mechanical loading cycle than conventional alternatives such as piezoelectrics. This higher energy density is a result of the high charge density associated with electrochemical processes compared with electrostatic processes. We discuss how the piezoelectrochemical effect can be exploited to harvest mechanical energy and experimentally demonstrate the working principle using a commercial pouch cell configured to harvest mechanical energy.

Energy harvesting from small ambient vibrations (e.g., wind, earthquake, rain, body movement, etc.) is a new pathway for production of green and renewable energy. Piezoelectric nanogenerator (PNG) is an energy harvesting device that converts ubiquitous vibrations into a form of electrical signal based on the energy conversion by nano-structured piezoelectric materials.¹ Harvesting energy from our surroundings is a good choice to meet the energy needs without causing unexpected environmental issues. Among various energy sources, mechanical energy is universally available, environmental friendly and reliable energy sources in our daily life, which is accompanying us regardless of the weather or temperature conditions as for solar and thermoelectric energy.² In addition, the choice of material is another critical issue to meet the criteria of environmental viability. Poly(vinylidene fluoride) [PVDF,-(CH₂-CF₂)-_n] and its co-polymers are classified as good ferroelectric bio-polymer materials those also exhibit electroactive response, including piezo, pyro- and ferro-electric effects.^3,4,5 In the presence of electroactive β-phase, PVDF promises wide range of applications such as in biomedicine, energy generation and storage, sensors and actuators, separator in power cell, proton exchange membranes, smart scaffolds, and many others.^2,3 However, most of the applications based on this polymer are impeded by its low piezoelectric coefficients (d_ij). In this context, polymer electret based NGs with high d_ij are the suitable choice for next generation flexible electronics.
In this work, a self-poled piezoelectretic nanogenerator (PNG) based on novel hybrid ferroelectretic polymer (PVDF) nanocomposite film composed of β-crystalline phase and porous structure is demonstrated. This film possess large piezoelectric charge coefficient, d₃₃ 〉-200 pC/N and large remnant polarization, P_r 〉 30 μC/cm2 with square shaped hysteresis loop. The PNG delivers sufficient amount of power that easily drives several consumer electronics such as, more than 50 LEDs, digital wrist watch, calculator etc. without any subsidiary batteries rather under gentle touch of human finger. This self-powered system is highly applicable as a strain sensor that can be used in wireless transport monitoring system where the input of the traffic can be sensed and electrical signal is being generated.
Acknowledgment: This work was financially supported by a grant from the Science and Engineering Research Board (SERB/1759/2014− 15), Government of India.
References:
[1] Z. L. Wang, J. Song, Science 2006, 312, 242.
[2] Y. Hu, Z. L. Wang, Nano Energy 2015, 14, 3.
[3] A. J. Lovinger, Science 1983, 220, 1115.
[4] S. K. Ghosh, M. M. Alam, D. Mandal, RSC Adv. 2014, 4, 41886.
[5] S. K. Ghosh, T. K. Sinha, B. Mahanty, D. Mandal, Energy Technol. 2015, DOI: 10.1002/ente.201500167.

A rapid development in portable electronics has raised urgent and challenging requirements for their sustainable power sources, which are now mostly using batteries. Currently the utilization of batteries leads to an inevitable design trade-off dilemma between mobility (size & weight) and sustainability (lifetime). A fundamental solution to this challenge is to develop technologies that can constantly convert ambient energy into electricity and get the battery continuously charged to ensure its sustainable operation. Human biomechanical energy is a promising energy source for that and has huge potential applications. Recently, to effective harvest ambient mechanical energy, we invent triboelectric nanogenerator (TENG) by incorporating contact electrification and electrostatic induction, and it has been demonstrated as a promising technology with numerous advantages. However, the output of a TENG is in the form of pulsed AC signal with variable frequency, so an energy storage unit is necessary to serve as a “reservoir” for collecting the generated charges and outputting them in a regulated manner. Nevertheless, a direct integration of a TENG with an energy storage unit has shown extremely low charging efficiency (Here, we perform both theoretical and experimental study on the TENG working principle, structural optimization, charging characteristics, and system-level integration technique. Finally, we develop the first mW-level TENG self-charging power unit, which consists of a multilayered triboelectric nanogenerator, a power management circuit to convert random AC energy to DC electricity, and an energy storage device. A multilayered triboelectric nanogenerator is designed and optimized to harvest human biomechanical energy. Through optimization, its short-circuit transferred charges and open-circuit voltage can reach 3 μC and 700 V, respectively. A power management circuit is designed to solve the impedance mismatch problem, which can achieve 60% total efficiency, about two orders improvement compared to direct charging. With palm tapping as the only energy source, this power unit provides a continuous DC electricity of 1.044 mW (7.34 Wm^-3). This self-charging unit can be universally applied as a standard “infinite-lifetime” power source for continuously driving numerous electronics, such as thermometers, electrocardiograph system, pedometers, watches, scientific calculators, and wireless communication system. The demos presented have covered all of the fundamental parts of mobile systems, including sensors, microcontrollers, memories, arithmetic logic units, displays, and even wireless transmitters, which show immediate and broad applications of this system in personal sensors and internet of things. [1]
Reference:
1. S. Niu, X. Wang, F. Yi, Y. Zhou, Z. L. Wang, Nat. Commun., DOI: 10.1038/ncomms9975

High efficiency multijunction solar cells in concentrator photovoltaic (CPV) modules are becoming an increasingly cost effective option in utility scale power generation. In order to compete with tradition silicon-based installations, CPV modules need to guarantee similar operational lifetimes of greater than 25 years. The adhesion and mechanical reliability of optical elements in CPV modules pose a unique materials challenge resulting from increased UV irradiance and enhanced temperature cycling associated with the concentrated solar flux. Understanding the underlying mechanisms of defect evolution and materials degradation is critical for commercialization of CPV technology. A survey of CPV module manufacturers revealed the most critical interface for reliability is the attachment of the secondary optical element to the multijunction cell. We developed fracture mechanics based metrologies to characterize the adhesion energy of the silicone encapsulant and the adjacent surfaces. Further, we developed the capability to age specimens in an outdoor concentrator in excess of 1100x the AM1.5 direct irradiance and in an indoor environmental chambers with broadband UV irradiation with controlled temperature and humidity. We observed a sharp increase in adhesion energy followed by a gradual decrease in adhesion energy as a result of both outdoor concentrator exposure and indoor UV exposure. We characterized changes in mechanical properties and chemical structures to understand the fundamental connection between mechanical strength and defect evolution in the silicone encapsulant. We developed physics based models to explain the change in adhesion and to predict operational lifetimes of the materials and interfaces. In addition, understanding degradation of materials under concentrated solar flux can offer unique insights into degradation modes in other encapsulants used in flat-panel PV.

Thermoelectric energy harvesting is the process of converting low grade heat (<150 °C) into electricity. One of the approaches for thermoelectric energy harvesting is the application of thermoelectric materials. It is demonstrated that conducting polymers can perform as a thermoelectric material^[1-3]. In fact by controlling the polymerization process and chemistry, it is possible to control the crystallinity, electronic band structure, charge carrier density and mobility and morphology of the polymer^{[1, 4-8]} and consequently walk towards development of a phonon-glass electron-crystal ideal structure and enhance the thermoelectric performance of polymers^{[1, 3, 9]}. Following these strategies, it is found that that by using a polymeric additives based on block copolymers it is possible to enhance the mobility of charge carriers in a conducting polymer film by 10-100 times. This indicates the possibility of reducing the number of charge carriers without a major loss of electrical conductivity. Therefore by reducing electronic thermal conductivity, the thermoelectric properties of the materials (Seebeck Coefficient and Figure of Merit) are expected to enhance significantly ^[10-11]. In addition to using polymeric additives, other strategies for reducing the electronic and lattice thermal conductivity of the conducting polymers and enhance their thermoelectric performance are explored and reported in this presentation.
References
[1] O. Bubnova, et al., Nat Mater 2014, 13, 190-194.
[2] J. Luo, et al., Journal of Materials Chemistry A 2013, 1, 7576-7583.
[3] O. Bubnova, et al., Nat Mater 2011, 10, 429-433.
[4] R. Brooke, et al., European Polymer Journal 2014, 51, 28-36.
[5] P. Hojati-Talemi, et al., Chemistry of Materials 2013, 25, 1837-1841.
[6] P. Hojati-Talemi, et al., ACS Applied Materials & Interfaces 2013, 5, 11654-11660.
[7] M. Mueller, et al., Polymer 2012, 53, 2146-2151.
[8] M. V. Fabretto, et al., Chemistry of Materials 2012, 24, 3998-4003.
[9] Z. U. Khan, et al., Journal of Materials Chemistry C 2015.
[10] G. J. Snyder, et al., Nat Mater 2008, 7, 105-114.
[11] P. Sun, et al., Nat Commun 2015, 6.

Among a variety of thermoelectric materials, half Heusler NiMSn (M = Ti, Zr, Hf) and full Heusler Fe₂VAl semiconducting alloys have been attracted considerable attention over last years. The interest to these materials has been conditioned by the cheapness of the constituting chemical elements, easy fabrication method and attractive thermoelectric properties. Other representatives of the semiconducting Heusler alloys, Fe₂TiSi and Fe₂TiSn, have enjoyed recently growing interest. Results of first principles calculations [1] have indicated that a large Seebeck coefficient, up to – 300 µV/K, can be expected in Fe₂TiSn_1-xSi_x alloys. On the other hand, experimental investigations [2] have shown that physical properties of Fe₂TiSn significantly depend on both atomic disorder and stoichiometry. Although a secondary phase is presented in bulk samples of Fe₂TiSi, the single Heusler phase can be stabilized in thin films of Fe₂TiSi [3]. The mentioned above facts imply that precipitations of the secondary phase in Fe₂TiSi and the physical properties of Fe₂TiSn can be controlled by preparation method and/or thermal treatment. In order to check this we have studied Fe₂TiZ (Z = Si, Sn) alloys prepared by mechanical alloying.
Ingots of stoichiometric Fe₂TiSi and Fe₂TiSn were prepared by induction melting in a protective Ar atmosphere. The ingots were solution treated at 1073 K for 24h, slowly cooled down to room temperature and crushed into small pieces. The pieces were milled in argon atmosphere by a Fritsch Pulverisette-5 planetary ball miller with the ball-to-powder weight ratio was 20:1 Crystal structure of the prepared powder was analyzed by X-ray diffraction (XRD) and transmission electron microscopy (TEM) by Rigaku Ultima IV diffractometer (Co Kα radiation) and JEM 1400, respectively. Results of these studies revealed that Fe₂TiSn consists of the single phase while a secondary phase was detected in the powder of Fe₂TiSi. The obtained powders of Fe₂TiSi and Fe₂TiSn were consolidated by Spark Plasma Sintering system (Labox 650, Sinter Land, Japan). Preliminary measurements of electrical resistivity ρ, Seebeck coefficient S and thermal conductivity k revealed that the mechanically alloyed Fe₂TiSn has larger ρ and S and significantly smaller k as compared to the polycrystalline sample [2]. In the mechanically alloyed Fe₂TiSi sample, temperature dependencies of ρ, S, and k turned out to be similar to those observed in Fe₂TiSn. Details of these measurements shall be given in the presentation.

References
[1] S. Yabuuchi et al., APEX 6 (2013) 025504.
[2] C.S. Lue and Y.-K. Kuo, J. Appl. Phys. 96 (2004) 2681.
[3] M. Meinert et al., Phys. Rev. B 90 (2014) 085127.

The fabrication and assembly of solid oxide fuel cells (SOFCs) - including both support and functional layers- remains one of the primary challenges preventing the widespread adoption of SOFCs as an energy conversion technology. SOFC structures produced using traditional manufacturing techniques are inefficient- the structural components contribute most to the weight of the fuel cell device as a whole- precluding their practical use in mobile applications such as transportation. We present an efficient and highly scalable multi-material process for fabricating SOFC constructs using a combination of room-temperature, liquid extrusion-based 3D-printing and dip-coating of particle-laden, liquid-based 3D-inks. 3D-printing is used to sequentially deposit anode and cathode functional layer materials, nickel oxide-yttria stabilized zirconia (NiO-YSZ) and lanthanum strontium manganite (LSM), respectively, without the need to alter printing parameters, allowing unprecedented control over gas channel geometries. Depositing layers thinner than 100 μm using 3D-printing is impractical, so the same inks designed for 3D-printing are repurposed for the production of mechanically robust, controllably thick, multi-material films via dip-coating to be used as YSZ electrolytes and strontium lanthanum titanate (SLT)/LSM interconnect bilayers. The inks used for both 3D-printing and dip-coating are synthesized through simple, room-temperature mixing of a combination of organic solvents, a biomedical elastomer binder (~10-40 vol.%) and powders of interest (~60-90 vol.%). Vol.% powder controls shrinkage and porosity during firing; tailoring the powder vol.% for each ink is vital to preventing warping and cracking during cell co-firing and to ensure optimal performance of each component. Due to identical solvent and polymer content between dip-coated and 3D-printed components, layers fuse seamlessly with one another, meaning that dip-coated electrolyte and interconnect films can be placed manually and will instantly fuse to preceding and successive 3D-printed layers, ensuring excellent inter-layer/material contact and integration, mitigating the risk of delamination prior to firing. Fully assembled fuel cell structures are co-fired in air at 1250°C for 4 hours. The microstructural and electrochemical characteristics of fired cells are analyzed, and compared with cells produced entirely using tape-casting techniques. We demonstrate that this highly scalable technique is useful for fabricating monolithic planar SOFCs of various sizes without the need for cumbersome support materials. This new, simple and highly versatile technique will form the foundation for improved SOFC manufacturing and function, as well as enable the exploration of complex, non-traditional SOFC designs not compatible with established fabrication processes, in order to increase their adoption and integration across industries thus improving their prospect for providing reliable, carbon-free power.

Proton conducting solid oxide fuels cells (p-SOFC) have many promising features that can improve next-generation fuel cell technologies. Rather than operating at the high temperatures (~800°C) required for traditional SOFCs, p-SOFCs are able to operate at an intermediate temperature (~500°C) which will help to reduce the degradation associated with high-temperature fuel cells and allow for the use of cheaper materials in stack manufacturing. Proton conducting solid oxide fuel cells are able to operate at lower temperatures because of the promising proton conductivity of materials such as yttrium-doped barium zirconate (BYZ). The achieved proton conductivity of BYZ is comparable to the oxygen anion conductivity of the commonly used SOFC electrolyte, yttria-stabilized zirconium (YSZ), however this proton conductivity of BYZ is attained at a much lower temperature. While BYZ is chosen as the electrolyte material for p-SOFCs, a nickel-BYZ cermet is chosen as the anode layer and lanthanum strontium cobalt ferrite (LSCF) has been chosen as the cathode material. While materials such as BYZ have been shown to have high proton conductivity, there have been difficulties with the sintering steps required for BYZ formation. Additionally, the sintering steps associated with traditional SOFC manufacturing are prohibitively expensive for wide scale manufacturing. To avoid these issues, this work will examine the production of the full p-SOFC cell (Ni-BYZ|BYZ|LSCF) using Reactive Spray Deposition Technology (RSDT). RSDT is a one-step flame-based deposition process which eliminates the need for the sintering processes associated with other manufacturing methods. As a result, using RSDT is a cost-effective manufacturing method for producing p-SOFCs. This work focuses on defining the deposition parameters for RSDT necessary to produce a trilayer cell as well as reports the initial electrochemical data that has been obtained.

Thermogalvanic cells exploit the entropy changes inherent in redox reactions to couple a flow of ions to a flow of heat. While these systems have been studied in the past in the context of thermal energy harvesting, few chemistries have been identified with the requisite low activation energies, high entropy changes, and high overall reversibility for effectively coupling thermal and electrical energy flows. However, recent developments in the redox flow battery community have opened up many avenues for improvement in thermogalvanics, especially in the nearly-unstudied but complementary field of electrochemical refrigeration, which offers a potentially environmentally and climate-friendly means of climate control.

In this presentation, we develop the thermogalvanic concept from basic principles, derive the fundamental relationship between cold-side activation energy and overall entropy change required to yield a cooling effect, provide an accounting of the dominant loss mechanisms encountered in these systems, and review a number of proposed chemistries and system configurations, including charge-cycling, static electrolyte, and flow systems with and without membrane separators. We present a matrix of candidate redox couples and describe the criteria for their utilization in thermogalvanic systems, as well as favorable operating regimes for each in terms of temperature, partial pressure of gaseous reactants, electrolyte composition, and catalyst and support materials.

Finally, we present a novel 4-electrode vanadium-bromide flow cell for electrochemical refrigeration. The 4-electrode cell is projected to produce a cooling power up to 40mW/cm^2 of electrode area with a cooling COP of 0.9, based on a model accounting for activation losses (based on Butler-Volmer kinetics and measured exchange current and transfer coefficient values), concentration polarization, and thermal losses. This model is described in full, and power curves are derived for varying temperature, pressure, pump power, and electrolyte concentration and composition. Additionally, we present the results from preliminary electrochemical and thermal testing in a 2-electrode static-electrolyte cell. This system is characterized via potentiostatic and galvanostatic techniques, and its transport characteristics are established via impedance spectroscopy. Finally, measured cooling power is compared to a model of the 2-electrode cell, and used to validate the model of the full 4-electrode electrochemical refrigerator.

Thermionic transport holds the promise to provide high-performance energy conversion devices for solid-state cooling or waste heat recovery. Two dimensional (2D) layered materials are potential candidates for solid-state thermionics due to their weak thermal conduction in the cross-plane direction and their tunable band gap via changing the number of layers. In this work, we study the thermionic transport across graphene/phosphorene/graphene van der Waals heterostructures in contact with gold electrodes by using first principles DFT calculations and real space Green’s function formalism. We provide an efficient approximation to accurately estimate the band gap of phosphorene in cross-plane transport calculations. We show that for monolayer phosphorene in the heterostructure, quantum tunneling dominates the transport. By adding more phosphorene layers, one can switch from tunneling dominated transport to thermionic dominated transport, resulting in transporting more heat per charge carrier, thus, enhancing the cooling coefficient of performance. Moreover, varying the number of layers can change the barrier heights and doping characteristics of phosphorene. We identified upper and lower limits of the thermionic coefficient of performance for the proposed device and showed that the proposed or similar devices have the potential to perform cooling with higher efficiency.

High-temperature thermoelectric materials offer enormous potential for the recovery of waste heat from various industrial sources. Currently, the best thermoelectric materials that operate at >500C are skutterudites which contain toxic and scarce elements such as arsenic or antimony. Transition metal oxides (TMOs) based on ABO3-type perovskites have recently gained much attention as viable alternatives because they contain cheap, plentiful and non-toxic elements, e.g. SrTiO3 and CaMnO3. As thermoelectrics they have many favourable intrinsic properties such as low thermal conductivity and large Seebeck coefficient, S, however they are often poor electrical conductors. The electrical conductivity is commonly improved in these materials by aliovalent doping and/or inclusion of defects on the A- and/or B-sites, however the resulting increased carrier concentration causes a dramatic reduction in S due to the direct competition between the underlying mechanisms. Here we propose a new route to improved electrical conductivity in these materials through synthesis of Ag/TMO composites. Our results suggest that silver inclusions improve grain-boundary conduction, rather than simply increasing carrier concentration or mobility. This different mechanism offers the possibility of improving electrical conductivity without detrimentally affecting S and therefore the overall thermoelectric performance. We also show the large discrepancy between thermoelectric properties of TMOs in 2- and 4-terminal geometry can be resolved with silver inclusions via a mechanism that is not solely dependent on improved contact resistance. As current thermoelectric generators operate in 2-terminal geometry, Ag/TMO composites provide an important step towards delivering these materials in real devices.

The efficiency of a thermoelectric device depends on its material figure of merit (ZT). PbTe is a suitable thermoelectric material applied in the temperature range 300-500 ^oC. Recently, researchers usually improve ZT through nano-composite methods, including nano-grain, micro/nano mixing, nano-inclusion and nano-precipitation method. The highest ZT record of N-type PbTe is 1.55. In this study, N-type PbTe (PbTe_1-xI_x) was synthesized by employing nano-precipitation method. PbTe_1-xI_x, bulk was fabricated through hot pressed sintering of nano-precipitated PbTe_1-xI_x powders. The effects of powder size and hot pressing time on thermoelectric properties were explored and optimized. The electrical conductivity was enhanced around 8% through fine tuning powder size, where optimize grain boundaries. With a further optimization of hot pressing time, the Seebeck coefficient enhancement and electrical conductivity reduction are around 20% and 8%, respectively. As a consequence, ZT=1.85 of PbTe_1-xI_x was achieved through two optimizations. The micro-structure of PbTe_1-xI_x was observed by high resolution transmission electron microscopy. The numerous PbI₂ nano-precipitations spread inside the PbTe grains homogeneously, where typical size are around 1-3 nm.

In order to fulfill the current need of alternative ways of environmental-friendly energy production and overcome from the crisis of present energy requirements, lot of efforts have been given to thermoelectric research in the past few decades. Since most of the high performance state-of-the-art thermoelectric materials contain toxic and rare elements, development of non-toxic, earth abundant and efficient alternatives are very important. In the present work, we report high thermoelectric performance in undoped hot-pressed In₄Se_3-x (x = 0.05), which does not contain any non-toxic and rare element. The stoichiometric imbalance caused by the selenium deficiency resulted in precipitation of indium secondary phase in In₄Se₃. The precipitated Indium rich domains at the grain surfaces/boundaries found to redistribute inside the thermally grown larger grains and their junctions after the heat treatment. The redistributed secondary phase of indium caused enhanced phonon scattering, due to which very low values of thermal conductivity were observed in In₄Se_3-x. Consequently, high conversion efficiency was achieved in terms of maximum figure of merit of value 1.13 at 723K along the hot-press direction.

Abstract Body: Thermoelectric (TE) materials show that a temperature difference induces an electric potential or vice versa. These phenomena are known as the Seebeck effect, Peltier effect, and Thomson effect. In many effects, The Seebeck effect is converting temperature difference into electric power, so that TE can be used for the application of electrical power generator or sensor. In small dimension, TE properties show the different trend comparing to bulk. In previous researches, TE is mainly focused on 1D or bulk. In this study, we focus on the 2D TE devices. Quasi-2D Si sheets are fabricated on silicon-on-insulator (SOI) substrate with different doping profile, length and width of dimension splits. Especially, such a 2D TE device has an advantage of CMOS compatibility. SOI active layer thickness is about 50nm, 2D Si channel’s width and length are in the range of 10μ~40μm to 10μm interval. The resistance of platinum metal line and voltage difference according to temperature were confirmed by using the semiconductors parameter analyzer (Keithley 4200). Using an electrometer (Keithley 6517a) & a sourcemeter (Keithley 2400) give the bias to heat a part of heater. Seebeck coefficient was measured for different doping conditions of 1E14, 5E14, 1E15, and 5E15 cm-2. When the doping concentration increases above 5E15 cm-2, the Seebeck coefficient is saturated. In general, the Seebeck coefficient in semiconductor is in the range of hundreds. Comparing to semiconductor, metal’s Seebeck coefficient is tens of order. The reason why the Seebeck coefficient has saturated value is heavily doped (above 5E15 cm-2) semiconductor operates metallically. Also, the simulation of heat distribution and the Seebeck coefficient using the COMSOL was carried out comparing to the measurement data. To investigate the effect of device dimension, the Seebeck coefficient is extracted for different channel width and length. At width modulation, the Seebeck coefficient decreases by increasing of width, whereas difference of the Seebeck coefficient is lower when doping concentration is higher. In conclusion, we show the TE effect in 2D Si devices comparing to different parameters. It has been shown that the Seebeck effect by doping concentration and dimension through heat distribution simulation in 2D Si TE device.

Zintl compounds are potential candidates for efficient thermoelectric materials, because they are typically small band gap semiconductors. In addition, such compounds allow fine tuning of carrier concentration by chemical doping for the optimization of thermoelectric performance. Herein, such tunability is demonstrated in Mg₃Sb₂- based Zintl compound via. Zn²⁺ doping at Mg²⁺ site of the anionic framework (Mg₂Sb₂)^2-, in the series Mg_3-xZn_xSb₂ (0 ≤ x ≤ 0.1). The materials have been successfully synthesized by spark plasma sintering (SPS) technique. The X-ray diffraction (XRD) analysis confirms single solid solution phase of Mg_3-xZn_xSb₂ (0 ≤ x ≤ 0.1). The thermoelectric properties are characterized by Seebeck coefficient, electrical conductivity, and thermal conductivity measurements from 323 K to 773 K. The isoelectronic Zn substitution at Mg site presents the controlled variation in the carrier concentration for optimizing high power factor and reduced thermal conductivity. These results lead to a substantial increase in ZT of 0.37 at 773 K for a composition with x=0.10 which is ~ 42 % higher than undoped Mg₃Sb₂. The electronic transport data of Mg_3-xZn_xSb₂ (0 ≤ x ≤ 0.1) compound are analyzed using a single parabolic band model predicting that Mg_2.9Zn_0.1Sb₂ exhibits near-optimal carrier concentration for high ZT. The electronic structure of transport properties of these disordered Mg_3-xZn_xSb₂ (0 ≤ x ≤ 0.1) are also studied by density functional theory and the results obtained are in good agreement with experimental results. The low cost, lightness and non-toxicity of the constituents elements make these materials ideal for mid-temperature thermoelectric applications.
*Corresponding author. E-mail address: [email protected], [email protected] (DKM)

An ideal thermoelectric material combines a high Seebeck coefficient and electrical conductivity, along with a low thermal conductivity, all of which depend on interrelated material properties, such as electronic band structure, Fermi level, and atomic arrangement. In the last few years, much effort has been paid to Cu-Sb-Se and Cu-Sb-S (CAS) based compounds, especially the Cu₃SbS_3.25 (tetrahedrite) compound. Cu₃Sb(Se/S)₃ and CuSb(Se/S)₂ possess anomalously low lattice thermal conductivity compared to Cu₃Sb(Se/S)₄. In-depth studies suggest that s² lone pair orbital electrons are a key factor to achieve minimum lattice thermal conductivity in chalcogenide compounds that contain a nominally trivalent group V_A element. The four main crystalline phases of sulfur based CAS compounds are Cu₃SbS₄ (fematinite), CuSbS₂ (chalcostibite), Cu₃SbS₃ (skinnerite), and tetrahedrite. The tetragonally coordinated fematinite exhibits the highest thermal conductivity while skinnerite has the lowest value at room temperature as a result of its complex crystal structure and the s² lone pair electrons on the Sb site, which is very similar to the bonding arrangement in tetrahedrite. Skinnerite has better electrical conductivity compared with fematinite and chalcostibite, although it is not high enough for a good TE material. Disadvantageously, the Cu₃SbS₃ has at least three temperature-dependent polymorphs, and the skinnerite is only persistent between 263 K and 395 K. Thus, stabilising the crystal structure and increasing the electrical conductivity of Cu₃SbS₃ are the two main objectives in order to enhance its TE properties
In this work, trivalent transition element iron without lone-pair electrons was incorporated into Cu₃SbS₃ compound. Experimental result showed that small amount of iron is capable of pushing the crystal structure from monoclinic to cubic. Simultaneously, the electrical conductivity is significantly improved without a significant negative influence on favourable low thermal conductivity. The changes in the bonding environment by substituted atoms and their influences on the crystal structure and transport properties have been studied and correlated.

Oxide thermoelectric materials, which are highly durable at high temperature in air, non-toxic, low cost with minimal environmental impact, are apparently promising for recuperation of decentralized waste heat energy at the temperature range of > 400 °C, where all the non-oxide candidate materials will eventually be oxidized under aerobic conditions. Although strongly ionic characters of oxide materials has been regarded as an inherent disadvantage leading to low carrier mobility and high lattice thermal conductivity, it has been revealed that such disadvantages are not always the case with all oxides. Recent reports on reduced ferroelectric oxides [1] and cage-like structure oxides [2] are convincing that the simple picture of ionic compounds no longer holds for these oxides.
In this paper, some new aspects in metal oxides that show unconventionally enhanced phonon scattering will be highlighted in terms of their transport properties and crystal structures. In particular, β-pyrochlore oxides ABB’O₆ (A = K, Rb, Cs) show lower lattice thermal conductivity with decreasing the mass and size of the A-site alkali cations, clearly evidencing that the larger size mismatch between the A-site cations and the surrounding oversized cage framework enhances the “rattling” motion of the A-site cations and thereby efficiently shortens the phonon mean free path more for the smaller A-site cations. As a consequence, the thermal conductivity of the oxide with the smallest A cation, KBB’O₆, was revealed to be virtually the same with its theoretical minimum, κ_min [3].

[1] S. Lee, J. A. Bock, S. Trolier-McKinstry, C. A. Randall, J. Eur. Ceram. Soc., 32, 3971 (2012).
[2] M. Ohtaki, S. Miyaishi, J. Electron. Mater., 42, 1299 (2013).
[3] C. Dames, G. Chen, J. Appl. Phys., 95, 682 (2004).

The lead- free SnTe has the two valence bands, including light hole and heavy hole bands, which can contribute to the hole density of states. The similar band structure with PbTe suggests it has the potential to be good thermoelectric properties. Indium dopant can modify the band structure by resonant level, increase the effective mass and then enlarge Seebeck coefficient and power factor. By the solid state reaction, the In doped SnTe alloys have been synthesized. 0.75% and 1% samples can get the high power factor and the higher figure of merit zT ~0.9 was obtained in 1% In doped sample. So the optimized amount of In was 1%. However, the thermal conductivity was about two times higher than that of PbTe, so in-situ nanostructure should be done for reducing it and enhancing TE properties. Therefore, we introduced nanostructures by inducing Se substitution to Te site. This substituted Se became clustered nanostructures which could reduce lattice thermal conductivity by phonon scattering. We clarified the sizes and distribution of nanostructures by electron microscope analysis. Finally, the higher figure of merit could be addressed by further reduction of thermal conductivity.

As one important research direction, the properties of graphene can be effectively tailored by fabricating periodic nanopores across graphene, which is called graphene antidote lattices (GALs). GALs can be designed to introduce a structure-dependent bandgap in semimetal graphene, as well as dramatic changes in its thermal, optical, and magnetic properties. This enables wide applications of GALs in electronic/optoelectronic devices, magnetics/spintronics, and waveguides. In addition, GALs also provide unique opportunities to achieve a high thermoelectric figure of merit (ZT) for power generation and refrigeration in graphene-based devices. Here ZT is defined as ZT=S²σT/κ, where S, σ, κ, and T represent Seebeck coefficient, electrical conductivity, thermal conductivity, and absolute temperature, respectively. In GALs, a high ZT is anticipated with strongly suppressed phonon transport (lower lattice part of κ) and concurrently improved electrical properties (higher power factor S²σ).

For phonon transport, periodically porous GALs may form a phononic crystal [1,2], whose κ can be significantly reduced by the phonon band structure modification due to coherent phonon processes in a periodic structure. Under the pursuit of a high ZT, such coherent phononic effects have been proposed for porous Si films [3] though the amorphization of pore edges may also play an important role in the observed κ reduction [4]. Compared with Si films, two-dimensional GALs eliminate the uncertainties from phonon boundary scattering on the top and bottom film surfaces and are more ideal in studying the influence of coherent phonon processes and amorphous pore edges [1]. Different from Si films, particular attention will also be paid to the antidot-pattern chirality and the pore-edge atomic configuration (e.g. zigzag or armchair), both of which are critical to the opening of an electronic bandgap within GALs [5]. In this work, the impact of these two factors on the obtained phonon dispersion will be further explored using first-principles methods.
For electron transport, the modified electronic band structure [6] will also be further computed for the investigated cases and be incorporated into the modeling based on the Boltzmann transport equation. The consideration will further include the scattering of charge carriers by a potential field formed on the pore edges due to trapped charges, which can be adjusted to tune the power factor. The work presented here is critical to the fundamental understanding of different transport processes within periodically nanoporous structures.


References:
1. J-F Robillard et al., Chin. J. of Phy. 49, 448 (2011)
2. A. Sgouros et al., J. Appl. Phys. 112, 094307 (2012).
3. J.-K. Yu et al., Nature Nanotechn. 5, 718 (2010).
4. Y. He et al., ACS Nano 5, 1839-1844 (2011).
5. X. Liu et al., Small 9, 1405-1410 (2013).
6. F. Ouyang et al., ACS Nano 5, 4023 (2011).

Although ZnO has been investigated for past several decades, studies on the thermoelectric properties of ZnO films are still limited. In this study, ZnO thin films were prepared by atomic layer deposition (ALD) at relatively low temperature of 200°C using diethylzinc and either water or ozone as oxidant. Extraordinary opposite trends of temperature dependent electrical conductivity (σ) and Seebeck coefficient (S) were observed under Ar/H₂ reducing ambient over a temperature range of 300-787 K. In particular, water based ZnO films (ZnO-H) show initially metallic behavior and a decrease in carrier concentration after annealing. In contrast, ozone based ZnO films (ZnO-O) show initially non-degenerate semiconducting behavior with increasing carrier concentration after annealing. X-ray photoelectron spectroscopy shows similar level of oxygen vacancy concentration in ZnO-O and ZnO-H. This suggests that the decrease in carrier concentration observed in ZnO-H after annealing may be attributed to the spontaneous formation of zinc vacancies. Low temperature photoluminescence confirmed enhanced green emission near 540nm on annealed ZnO-H which is suggested as oxygen vacancy to zinc vacancy transition. The highest thermoelectric power factor (σS²T) of 0.45 W m^-1 K^-1 is obtained at 787 K for ZnO-O, which is nearly double the previously reported values in literature, and even comparable with Al-doped ZnO thin films. This work highlights the importance of the oxidant used in depositing the ALD films, and it opens a possibility of greater improvements of thermoelectric performance of the most cost-effective ZnO thin films.

Thermoelectric materials can convert heat into electricity and serve as a heat pumping media by proving an electrical power. In this study, a unique electrical stressing assisted thermal treatment is applied on the printed thermoelectrics using a specially designed sintering system. Both Seebeck coefficient and electrical resistivity are measured for power factor optimization of printed thermoelectrics. The dependence of annealing temperature and applied electric current density on the variations of carrier concentration and carrier mobility of printed thermoelectrics are also investigated. The results show that the electrically stressed sample with a current density of 300 A/cm² has better thermoelectric properties than the thermally annealed ones. The best thermoelectric properties achieved so far are 1.82*10^-3 W/m×K for N-type material and1.82*10^-3 W/m×K for P-type material, respectively. It is concluded that electrical stressing indeed can effectively enhance the thermoelectric properties of printed thermoelectrics.

Because of the generally lower activation energy associated with proton conduction in oxides compared to oxygen ion conduction, protonic ceramic fuel cells (PCFCs) should be able to operate at lower temperatures than solid oxide fuel cells (250° to 550°C versus ≥600°C) on hydrogen and hydrocarbon fuels if fabrication challenges and suitable cathodes can be developed. The fabrication of typical ceramic-based fuel cells, such solid oxide fuel cells (SOFCs) or protonic ceramic fuel cells (PCFCs) is a complex, expensive, and time-intensive process involving multiple powder synthesis, coating, and firing steps. Here, we present a novel approach to fabricate the complete sandwich structure of a PCFC (porous anode, dense electrolyte, and porous cathode bone) directly from raw precursor oxides or carbonates in a single firing step. Our novel approach, which leverages the recently discovered solid-state reactive sintering process, greatly simplifies ceramic fuel cell fabrication and lowers costs, while also yielding cells with superior performance. We show that proper choice of the solid-state reactive sintering additives enables conversion of the precursor oxides to the desired phase-pure perovskite compositions and promotes full densification of the electrolyte layer while ensuring that the anode and cathode structures remain highly porous. We also developed a mixed proton, oxygen ion, and electron hole conducting PCFC-compatible cathode material, BaCo_0.4Fe_0.4Zr_0.1Y_0.1O_3-d (BCFZY0.1), that greatly improved oxygen reduction reaction kinetics at intermediate to low temperatures. We demonstrated high performance from five different types of PCFC button cells without degradation after 1400 hours. Power densities as high as 455 milliwatts per square centimeter at 500°C on H₂ and 142 milliwatts per square centimeter on CH₄ were achieved, and operation was possible even at 350°C.

Solid oxide fuel cells (SOFCs) are energy conversion devices that produce electricity by electrochemically reacting a fuel and an oxidant across an oxide electrolyte. Mixed ionic and electronic conducting (MIEC) ceramic oxides are commonly used as cathode materials in SOFCs because of their high catalytic activity for the oxygen reduction reaction. The high ionic conductivity makes MIEC cathodes have more active reaction area and better catalysis. Perovskite oxide ceramics La_0.6Sr_0.4CoO_3-δ(LSC) has received much attention as SOFCs cathode based on its high ionic conductivity and catalytic activity. In order to promote LSC practical application, the electrochemical and electrical properties of LSC have been improved further by doping elements. And doping influence on oxygen diffusion of LSC has been discussed.
Ba and Fe doped cathode materials La_0.6-xBa_xSr_0.4Co_1-yFe_yO_3-δ(LSBCF) have been synthesized by glycine nitrate process. It has been found that the dopant element has an effect on the extent of combustion. The oxygen atmosphere is expected to affect the results of oxygen diffusion. Oxygen diffusion is measured using electrical conductivity relaxation (ECR). The ECR results obtained at different beginning partial pressure and end partial pressure are presented to investigate the mechanism of oxygen diffusion. Besides, the effect of element doping on electrochemical and electrical properties has been discussed.

Recently, thin-film thermoelectric devices based on organic materials, polymers, and organic–inorganic hybrids have been widely investigated because of large area, low cost, and flexibility in their forms. Among various materials, poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) is one of the good candidates for thermoelectric devices owing to the high electrical conductivity over 1000 S/cm. Thus, the devices with PEDOT:PSS generally exhibit a good power factor (PF) or a thermoelectric figure of merit (ZT) among organic materials. One approach to increase the thermoelectric performance of PEDOT:PSS is to increase the electrical conductivity of the thin films by H₂SO₄ treatment as reported previously. However, due to the high acidity of H₂SO₄, the treatment is hardly applicable plastic-based substrates. Also, dropping the acid solution generally results in bad surface morphology. In this work, we introduce a new H₂SO₄ treatment method via vaporization, showing a good electrical conductivity and smooth surface in H₂SO₄-treated PEDOT:PSS thin films. By using this technique, we can also demonstrate flexible thermoelectric devices fabricated on plastic substrates.

Development and deployment of thermoelectric materials with enhanced figure of merit (ZT) is a challenging issue for modern and future energy conversion and recovery technology. Zintl phase Mg₃Sb₂-based thermoelectric materials have recently been considered as a potential candidate, offering desirable features of characteristic phonon glass electron crystal, intrinsically small-band gap, complex structures and relatively low thermal conductivity for efficient thermoelectric application. We have recently investigated the effect of isoelectronic Bi^3- substitution on Sb^3- site and Pb^4- substitution in the anionic frame work of Mg₃Sb₂ for the optimization of ZT. Herein, we report an unprecedented success in achieving an astoundingly large p-type thermopower of +721 μV K⁻¹ at room temperature by substituting of Se^2- on Sb^3- sites in Mg₃Sb_2-xSe_x (0 ≤ x ≤ 0.05). Single phase p-type Mg₃Sb_2-xSe_x (0 ≤ x ≤ 0.05) compounds, with high purity of Mg, Sb and Se powders as starting materials, have been prepared employing spark plasma sintering (SPS) reaction route. Electronic and Thermal transport data of Mg₃Sb_2-xSe_x (0 ≤ x ≤ 0.05) alloys have been analyzed and compared to pristine Mg₃Sb₂. We observe that Se^2- substitutions on Sb^3- sites in Mg₃Sb_2-xSe_x (0 ≤ x ≤ 0.05) significantly decreases the hole carrier concentration resulting in increased thermopower. The enhanced high thermopower at near room temperature establishes that an appropriately controlled doping can further improve the ZT in this class of Zintl phase materials for room temperature thermoelectrics applications.
*Corresponding Author. Email Address : [email protected], [email protected]

Most of the oxide thermoelectric materials reported so far exhibit low thermoelectric performance due to low electrical conductivity which has been mainly jeopardized by very low carrier mobility. BiCuSeO oxyselenides based materials particularly Cu deficient alloys have been shown as promising thermoelectric materials with ZT ~ 0.81 at 923 K [1] for BiCu_0.975SeO and is considered as robust candidates for energy conversion applications. In this work, Te doping on Se site in BiCu_0.975Se_1-yTe_yO (0 ≤ y ≤ 1) was performed to further optimize ZT via increasing the electrical conductivity and decreasing the thermal conductivity while keeping little variation in the Seebeck coefficient. All the samples were prepared employing combined approach of mechanical milling and furnace reaction followed by conventional hot pressing technique. The reduction in thermal conductivity is attributed to mass fluctuation resulting due to heavy Te doping while increase in electrical conductivity is attributed to the band gap minimization and increasing carrier mobility. Microstructural details of samples of BiCu_0.975Se_1-yTe_yO (0 ≤ y ≤ 1) employing high resolution transmission electron microscopy (HRTEM) are investigated to correlate the observed thermoelectric properties.

Reference :
[1] Liu, Yong, et al. "Remarkable enhancement in thermoelectric performance of BiCuSeO by deficiencies." Journal of the American Chemical Society133.50 (2011): 20112-20115.

*Corresponding author. E-mail address: [email protected], [email protected] (DKM)

We synthesize and characterize layered oxychalcogenides with the goal of discovering novel thermoelectric oxides with relatively high ZT. Layered oxychalcogenides, BiOCuSe, consists of alternating [Bi₂O₂]²⁺ oxide layers and [Cu₂Se₂]^2- layers stacked along the c-axis. To date, efforts on modifying the electrical transport properties of BiOCuSe to improve its thermoelectric performance have been primarily focused on doping at the bismuth site or copper site. Here, we investigate the effects of S doping at the oxygen site on the thermoelectric properties of BiOCuSe. The prepared S-doped BiOCuSe are characterized by X-ray diffraction (XRD), differential thermal analysis-thermogravimetry (DTA-TG), and X-ray photoelectron spectroscopy (XPS). Electrical resistivity (ρ), Hall coefficient (RH), Seebeck coefficient (S), and thermal conductivity (k) measurements are performed on S-doped oxyselenide samples.

Piezoelectric nanogenerators (PENGs) harvest energy originated from various physical movements in our living environment. Various PENGs using different piezoelectric materials and device structure have been developed to date. Among them, PENGs consisting of ferroelectric nanomaterials and polymer, have demonstrated as simple, robust, and suitable way to achieve high performance flexible PENGs. However, this composite structure was not suitable for ZnO-based PENG fabrication due to difficult polarization domain alignment of ZnO in the composite structure.
In this work, we developed ZnO-based composite PENGs by employing ferroelectric phase transformed ZnO nanosheets (NSs). ZnO were doped with vanadium (V) by substituting Zn atoms during hydrothermal synthesis. Due to the large difference in ionic radii between V (0.54 Å) and Zn (0.74 Å), the ferroelectric phase can be induced in V-doped ZnO NSs. Thus, polarization domain in ZnO can be aligned even in the composite structure by externally applying electric field. The PENGs fabricated with V-doped ZnO NSs and PDMS composite demonstrated significantly enhanced output power generation after poling process, attributed by the polarization alignment of the V-doped ZnO NSs. The PENG with an active area 4 cm × 4 cm generated output voltage of ~32 V and output current of ~6.2 mA, respectively, which was two orders of magnitude higher output voltage and current by comparison to the PENG based on intrinsic ZnO. Furthermore, the device assured the superior durability and reproducibility, measured over a month, thanks to the PDMS polymer encapsulation for the NWs. The approach introduced in our work is simple, effective, and suitable for large-scale flexible high performance ZnO-based PENGs.

Triboelectric generators (TEGs) have attracted great interest as a viable route to realize self-powered electronic devices thanks to their simple fabrication process, low cost, and especially high output power generation compared to other energy harvesting devices. Until now, most efforts to improve the output power of triboelectric generators (TEGs) have mainly focused on developing surface patterning methods to increase friction surface area. These approaches, however, have been hindered by the limited material choices, complicate processes, and easy wear-out. Thus, it will be noteworthy to find alternative solution, addressing these issues as well as improving output performance of TEGs.
Contact electrification in TEGs is determined by triboelectric charging sequence difference between two contacting material surfaces. To this end, we developed chemical surface functionalization method to modify the triboelectric charging sequence. Specifically, polyethylene terephthalates (PETs) were coated with poly-L-lysine (PLL) solution and trichloro (1H,1H,2H,2H-perfluorooctyl) silane (FOTS) using solution-coating and vapor deposition, respectively. After surface functionalization, electron affinity of two surfaces are differentiated, where one surface is positively charged by the NH₂⁺ group in PLL and other PET surface is coated with fluorine that has high electron affinity. The TEGs output using two functionalized surfaces exhibited V_open-circuit (V_oc) of ~330 V and J_{short-circuit} (J_sc) of ~270 mA/m² under contact motion at an applied force of ~0.5 MPa. In addition, the surface functionalized TEGs demonstrated the superior mechanical durability and reproducibility without performance degradation, measured over ~7200 cycles for a month, thanks to the strong covalent bond between hydroxyl onto O₂ plasma treated PET surface and coating molecules. The approach introduced here is a facile, effective, and cost-competitive route to fabricate high performance TEGs.

Energy harvesting using triboelectric generators has gained great attentions due to high output power generation and simple fabrication method. Since its first introduction, several approaches have been tried to further extend its performance limit. Among them, physically enlarging friction surface area is a distinct way to enhance the output power generation. To date, various surface patterning methods using micro sized template, anodized aluminum oxide (AAO) patterns, and block copolymer self-assembly have been adopted for triboelectric generators.
Here, we report the performance of triboelectric generators with chemically functionalized nanoimprint surface patterns. Our results show that the performance of the TENG can be greatly enhanced by enlarging surface area using nanoimprint. Specifically, the generator with the smallest line width and pitch (a 200 nm line width and 200 nm pitch) exhibits voltage and current up to ~ 400 V and ~ 48 μA cm^-2 at applied force of 0.3 MPa, which is 2-fold higher values than those from flat surfaces. Besides, nanoimprinted line patterns are also chemically functionalized with poly (diallyldimethylammonium chloride) (PDDA) to maximize the performance of triboelectric generator, generating voltage and current up to ~ 480 V and 60 μA cm^-2, respectively.

Zinc oxide (ZnO) based piezoelectric nanogenerators (PENGs) have been widely studied due to its non-toxicity and abundant in nature. However, performance of ZnO PENG has been limited by piezoelectric potential screening effect, originated from large excess electron concentration in ZnO. Several methods have been introduced to alleviate this problem. One effective approach was to form p-n junction with ZnO and p-type materials. To date, various p-type materials have been adopted for p-n junction formation. However, ZnO homojunction formation has been rarely investigated even if it is more suitable for PENGs.
Here, we report the performance of ZnO p-n homojunction PENG with phosphorus doped ZnO (p-ZnO:P) and ZnO on a ITO/Ag/ITO (IAI) coated PET substrate. Our results show that ZnO p-n homojunction PENG demonstrates the output voltage and current up to ~ 24 V and ~ 6 μA, respectively, at the applied force of 0.5MPa, which is 6-fold higher than that of PENG only with ZnO layer. Besides, optimal thickness ratio between p-ZnO and ZnO has been systemically investigated to maximize the performance of ZnO p-n homojunctin PENGs.

Nafion/SiO₂ composite membranes were prepared via sol-gle method tetraethylorthosilicate (TEOS) in ethanol using three different nafion membranes (115, 117 and 212, in order of decrescent thickness) in order to control the silicate nanoparticles concentrarion on the membrane. The composite membranes were characterized using electrochemical methods, X-ray diffraction, FTIR spectroscopy and SEM anallysis. XRD show that the thickness of the polymeric membrane have an influence on the crystallinity of the SiO_2. It was observed that lower SiO₂ crystalinity was obtained as the membrane thickness was increased. The filler concentration showed an important mechanical modifiaction of the membrane, as higher the concentration of SiO₂, the membrane expancion was reduced what indicates that the inorganic filler material induces affinity to water and it assist the proton transportation across the electrolyte membrane, even under low relative humidity conditions. The membrane characteristics were studied in a hydrogen/oxygen proton exchange membrane fuel cell prototype, were the humidity generated under operation was measured by electrochemical analysis. FTIR and SEM analysis indicated an interesting chemical interaction between polymer sulfonate groups and the nanoparticles surface what induces diffusion or nanoparticles onto the polymeric matrix. These results indicate that the addition of the filler to the original nafion membrane is a favorable strategy to increase the efficiency of the fuel cell and to enlarge its life-time.

Molybdenum Disulfide (MoS₂) have attracted great attention because of their novel properties such as the presence of a direct bandgap, good field-effect transistor characteristics, large spin-orbit splitting, intense photoluminescence, catalytic properties, magnetism, superconductivity, ferroelectricity and several other properties with potential applications in electronics, optoelectronics, energy devices and spintronics.
Hydrothermal based facile synthesis process was developed to synthesize molybdenum disulfide (MoS₂) nanospheres by hydrothermal method at 200 °C, in which sodium molybdate, thiourea and polyethylene glycol (PEG-1000) were used as starting materials. As-prepared MoS₂ was then annealed in vacuum at 60 °C for 12 hours. The effect of hydrothermal time on MoS₂ nanospheres morphology and its electrochemical properties as supercapacitor has been studied. XRD and TEM studies showed that the MoS₂ prepared in shorter hydrothermal dwell time was a mixture of amorphous and monocrystalline particles, and there was an evolution of crystallinity of the nanostructures for particles synthesized at 6 hours, 12 hours and 24 hours. MoS₂ working electrodes were prepared by mixing 80 wt% of MoS₂ sample,10 wt% acetylene black and 10 wt% polyvinylidene fluoride (PVDF). Electrochemical properties of MoS₂ were evaluated using cyclic voltammetry (CV), electrochemical impedance spectroscopy, and galvanostatic charge−discharge studies in 3 M KOH medium. The specific capacitance of nanostructured MoS₂ was observed to be between 122 F/g and 198 F/g at different scan rates along with excellent stability for long cycle time.
It is demonstrated that the obtained MoS₂ nanospheres with three-dimensional architecture has excellent electrochemical performances as anode materials for super capacitor applications. The formation mechanism of the nanospheres and their impact on the specific capacitance were discussed in detail.

Many of the intrinsic properties of magnetic ferrite particles are primarily determined by their shape, size, and chemical composition. Hence, exploring a feasible/controlled synthetic approach for ferrite synthesis with adjustable morphologies and components is highly desirable. Furthermore, search for an alternative energy source with features of high energy density, miniaturization and portability is much in demand. Supercapacitors are one such emerging alternative with power storage capacity in the range 20–2000 F/g. Cationic surfactant, CTAB, have been used as templating micelle molecule to synthesize mesoporous materials to control morphology. The SrFe₁₀Al₂O₁₉ ferrites were prepared via wet chemical method. Stoichiometric amount of analytical grade nitrate salts of Fe, Sr and Al were dissolved in water solution (Fe^{3 +}/Sr^{2 +} mole ratio = 11) and was dropped into a NaOH under constant stirring for 20 min. Different Wt.% CTAB was added according to the weight of nitrate salts. The obtained precipitates were filtered, washed, dried and calcined at 900 °C (10 h). XRD patterns of samples prepared using CTAB were single phase SrFe₁₀Al₂O₁₉ powder without any a-Fe₂O₃. The SEM analysis show that the sample prepared without CTAB are hexagonal plates and rods, while samples prepared with CTAB show grain growth with plate like morphology of size ~ 650 nm (6 Wt.% CTAB). In the presence of CTAB, owing to selective adsorption of CTAB to a certain crystal plane, the smooth reaction interface leads to the formation of hexagonal platelet-like particles. The RT demagnetization curves show 17.46% enhancement in Ms value for CTAB sample (66.41 emu/g, 9% CTAB). In presence of CTAB, single phase highly crystalline SrFe₁₀Al₂O₁₉ was obtained with a remarkable enhancement in Ms. The electrochemical measurements were performed using standard three electrode system. The working electrode was prepared by mixing sample, acetylene black and PVDF in 80:10:10 Wt.% ratio in N-methyl pyrrolidinone. The mixed slurry was pasted onto nickel foam. Cyclic voltammograms (with 3M KOH electrolyte) were recorded in the potential window of 0 to 0.6 V. The anodic/cathodic peaks were observed and may be related to the formation of redox couple Fe⁺²/Fe⁺³ during charge transfer reaction [[x]]. The specific capacitance at 1mA for 1% CTAB was highest (92 F/g) while lowest (37 F/g) for 9 Wt.% CTAB. Furthermore, highest energy density was observed for 1% CTAB (0.96 Wh/kg) and power density of (368 W/kg) was observed for 3 Wt.% CTAB sample. These values are superior to that of oxide (NiO, WO₃ etc.) electrodes. It is thus shown that the samples prepared with CTAB exhibits greatly enhanced specific capacitance compared with samples prepared without CTAB at the identical measurement conditions. High pore volume is important to provide rich sites that can absorb ions and accelerate electron transfer or decrease electric resistance loss.

The incorporation of charge storage functionality into fuel cell electrodes opens new opportunities for the sustained operation of a cell during periods of refueling and intermittent fuel interruption. Charge storage in an electrode may also reduce the response time in these devices, lowering the barriers for integration with other energy conversion and utilization platforms. The multiple and easily accessible oxidation states of vanadium make vanadium oxide (VO_x) an attractive candidate as a charge storage material to be integrated into solid oxide fuel cell (SOFC) anodes. Thin film yttria-stabilized zirconia (YSZ) cells that generate power for several minutes after H₂ fuel supply interruption have been demonstrated using thin film VO_x anodes, suggesting considerable potential for the emerging concept of a SOFC with built-in charge storage. In this work we explore the fabrication, electrochemical properties, and response to thermal stresses of a thin film YSZ fuel cell structure supported by a YSZ/VO_x composite anode. The YSZ/VO_x anodes were produced in the shape of 0.5-inch diameter (1-mm thick) pellets by mechanical milling and pressing of commercially available V₂O₅ and YSZ powders, followed by annealing at 650°C for 24 h in air. Reduction of the pellets in 5% H₂ atmosphere at 450°C for 6 h produced anodes with good electrical conductivity as determined by electrochemical impedance spectroscopy (EIS) measurements. Optical microscopy image analysis is used to probe the details of the conductive network of the anodes, including average grain size of interconnected vanadium-rich domains and porosity of the pellets. X-ray diffraction (XRD) indicates that the anodes contain metallic vanadium and several phases of vanadium oxide, including VO, V₂O₃, and V₂O₅. Pulsed laser deposition (PLD) is then employed to deposit YSZ thin films on a polished face of the anodes. For this purpose, the focused beam of a KrF excimer laser (248 nm) is used to ablate 8-mol.% YSZ targets at a laser fluences of 1.5 J/cm² in an 80 mTorr oxygen environment. The temperature of the YSZ/VO_x substrate was varied within the 500-650°C range for different depositions to maximize the chance of obtaining an effective YSZ electrolyte layer on the anodes. The YSZ thickness was targeted at 200 nm for all films based on PLD deposition rate calibrations. The deposition of cubic YSZ films on the anodes is confirmed on all samples by grazing incidence XRD measurements. XRD and EIS data obtained after Pt contact deposition on YSZ, and the thermal cycling of the structure, are used to correlate strain effects caused by the thermal mismatch of VO_x and YSZ and the overall electrical response of the structure up to 670°C, which is suitable for a thin-film YSZ cell. We present a comparison of our results with measurements in equivalent reference structures utilizing conventional YSZ/Ni cermets as the anode material.

We have measured the thermopower and resistance of Bi and BiSb nanowires for compositions that range from the semimetal phase (pure Bi) to the topological insulator phase (TI) for Sb atomic concentrations up to 0.17 and for nanowires as small as 75 nm. Our motivation is that bulk BiSb alloys with high concentration of Sb, that is with an atomic ratio of Sb in Bi of 0.08 and more, display an inverted band spectrum. In alloys with an inverted spectrum, the insulating phase is a TI, which is a novel state of quantum matter. It has been predicted that surface states of topological insulators have large thermopower and also high mobilities. In contrast bismuth is not a topological insulator. Because of the large surface to volume ratio in nanowires, electronic transport in nanowires is mediated by surface states and therefore nanowires are a platform for testing these properties. We are also interested in the semimetal -semiconductor transition SMSC) induced by the size quantization and elastic deformation because these mechanisms change the relative positions of the valence and conduction bands in nanowires of these materials according to their effective mass. It is well known that quantum confinement decreases the bulklike carrier density in Bi nanowires with a critical diameter of 70 nm. In our investigation of BiSb nanowires, we have determined the SMSC critical diameter for a number of compositions. For this research, we have developed a method to produce long single crystalline Bi and BiSb alloy nanowires in a glass capillary via casting. Nanowires of this type are well protected by a glass cover of the external environment and open a unique opportunity to investigate the quantum transport properties. The diameters of the nanowires can be as small as 75 nm. We have observed new effects such as negative magnetoresistance in transverse magnetic field at low temperatures. The temperature dependence of thermopower is non-monotonic, changing from n-type to p-type and exhibiting a significant dependence on the diameter. Bi and bismuth-antimony nanowires have applications in nanoelectronics, spintronics and thermoelectricity. Our fabrication method for single crystal nanowires in glass cover affords works with wires with a wide range of diameters and allows compositional, structural, and electronic tunability. We gratefully acknowledge financial support by the STCU-project 5986. TEH research is supported by NSF PREM 1205608, NSF STC 1231319, and the Boeing Co.

Aerogels are now very popular with their well-known high porosity, large interconnected open pores, large internal surface area and low densities. One potential application of metallic aerogels is acting as self-supported catalysts in fuel cells, which has been regarded as attractive energy conversion devices with high energy efficiency and environmental-benign advantages. Herein, we report a facile approach of obtaining AuPtPd metallic aerogels employing citrate as stabilizing agent and ascorbic acid (AA) as reducing agent for the first time. The formation of hydrogels would only take 4 or 5 hours at room temperature, which is relatively time-saving and energy-saving. The electrochemical results showed excellent electrocatalytic activity and durability for ORR (mass activity of AuPtPd reached 0.175 A/mg in 0.1 M HClO_4，which is about 2.5 times higher than commercial Pt/C). The outstanding electrochemical performance is attributed to their large surface area, high porosity, eliminated carbon support corrosion, high robustness and synergetic effect between different elements. The rationally designed multi-metallic aerogels are not only beneficial to enhance the electrochemical catalytic performances in fuel cells, but also paves the way to explore the novel porous metal nanostructures with wide applications.

Miniaturized self-contained enzymatic biofuel cells with high cell performance possess the ability to enable a new generation of minimally invasive implantable medical devices in vivo studies. Integration of the high surface area nanomaterials such as carbon nanotubes (CNTs) and graphene could be one of the effective solutions to further improve performance of current biofuel cells. Herein we present a novel method for fabricating micro-biofuel cells based on three-dimensional carbon micropillar arrays coated with reduced graphene oxide (rGO) and CNTs composite. The fabrication process of this system combines top-down carbon microelectromechanical systems (C-MEMS) technique to fabricate the 3D micropillar arrays platform and bottom-up electrophoretic deposition (EPD) to deposit the reduced rGO/CNTs/enzyme onto the electrode surface. In addition, on the basis of our prototype design of an EBFC chip, the modeling utilizing finite element analysis based on this micro biofuel cells has been conducted. Two modules from COMSOL Multiphysics have been applied to obtain the maximum theoretical cell performance: 1) diffusion module to incorporate the mass transport and enzymatic kinetics; 2) conductive module to integrate concentration and potential. Most of the parameters have been either extracted from experiment results. The details on fabricating the graphene/CNT based 3D micro biofuel cells, evaluating the experimental biofuel cell performance and comparing the cell efficiency between experimental and modeling study will be presented in this talk.

Consumption of non-renewable fossil fuels, coupled with the impact of greenhouse gas emissions, are driving increasing interest in technologies that can improve the sustainability of the transportation sector. Multi-valent batteries show promise as an alternative, higher energy density storage device. In particular, magnesium-based batteries are an attractive chemistry due to their apparent resistance to dendrite formation on the anode, as well as the potential for low cost due to the high natural abundance of magnesium. For a Mg battery to be viable, magnesium must efficiently plate (during charging) and strip (during discharge) from the anode during cycling. Employing first-principles calculations, we characterize the efficiency of these processes on several low-energy magnesium surfaces by evaluating so-called “thermodynamic overpotentials.” These data are compared to similar calculations on prototype surfaces for a lithium anode. Limiting potentials of the two metallic anodes are discussed in light of expected cycling efficiencies.

A first time method for the synthesis of continuous nanochains by employing electrospinning and post processes are reported with theoretic support. High aspect ratio electrospun PAN nanofibers were stabilized in air at a specific heating rate followed by functionalization in aqueous KMnO₄ solution. The composite membrane was calcined in air in order to remove polymer skeleton along with reduction of KMnO₄ into MnO₂. The highly crystalline and phase pure birnessite type MnO₂ nanochains were characterized by different microscopic and spectroscopic techniques. Electrochemical studies of these nanochains were carried out using three electrode and two electrode set up with 0.5 M Na₂SO₄ aqueous electrolyte. A possible mechanism for the formation of nanochains was also explained.

The performance of the state-of-the-art energy storage materials in Li- and Na-ion batteries is dependent upon the reversibility of the ion intercalation and de-intercalation processes. Such processes often are accompanied by phase transformations for which strain and defects can play critical roles in determining its true reversibility. This talk will give an overview on the challenges and opportunities in the field of electrochemical-mechanics. I hope to demonstrate that with the advances in multi-scale modeling and operando imaging, we can now effectively probe the roles of strain and defects in energy storage materials. Such understanding is crucial in formulating new strategies for designing next generation energy storage materials with superior cycle life and safety.

The mechanics of lithium-ion-battery electrodes is important for their behaviour and handling during cell production and especially for the long-life performance of lithium-ion-battery cells. The mechanics of differently produced electrodes of different materials was measured via nanoindentation and tensile tests. It can be shown that besides the material recipe every single process step contributes to the coating structure and, thus, to the mechanical properties. The dispersion step changes especially the morphology of the carbon black agglomerates, the drying step influences the distribution of the binder within the electrodes and the calendering step reduces significantly the pore sizes. For example an intensive dry dispersing of carbon black reduces the total deformation energy of an electrode and the calendaring decreases significantly the plastic deformation energy while the elastic deformation energy stays almost constant. By simulating the mechanical behaviour of the electrodes using the discrete element method the effect of structural and material properties on the mechanical electrode behaviour can be shown. By applying the nanoindentation test to physical and simulated electrodes the effect of electrode aging at different temperatures and different C-rates on the mechanical parameters at different State of Charges (80 %, 90 %, 100 %) were investigated. The measurements show the elasticity of the anode samples decrease with increasing cell age, depending on the C-rate and temperature. Higher temperatures lead to faster chemical side reactions like the decomposition of the electrolyte generating products which close the pores in the composite, consequently decreasing the elasticity of the particle arrangement.

Nowadays, of growing importance are rechargeable batteries for example the single-ion batteries including Ni-MH, magnesium ion or lithium ion batteries, which utilizes at least one intercalation electrode with the guest cations (proton, magnesium ion or lithium ion or the rechargeable alkaline MnO₂/Zn batteries. Here we report a new rechargeable aqueous zinc ion battery, which utilizes the rapidly reversible intercalation of a Zn²⁺ guest cation into the positive manganese dioxide (MnO₂) tunnels and the fast electrochemical deposition/dissolution of the negative zinc in a mild aqueous electrolyte containing Zn²⁺ ions. As it utilizes the intercalation of zinc ions into MnO₂ tunnels, this zinc ion battery not only could provide the capacity higher than the alkaline Zn/MnO₂ cells, but, for the most important, due to the reversibly rapid intercalation/deintercaltion rate of Zn²⁺ ions, this zinc ion battery could also deliver a power density and excellent columbic efficiency comparable to the state-of-the-art supercapacitors. We found that multivalent ions could be stored in nano-MnO₂ thus we designed a rechargeable zinc ion battery based on a zinc anode and a MnO₂ cathode in aqueous solution electrolyte.

In this talk we study the role and influence of mechanical stresses on the electrical double layer. On both sides of the interface between an electrode and an electrolyte an electrical double layer develops. Up to now the charge distribution on the electrolytic layer is described by a Poisson-Nernst-Planck model (PNP). However, PNP models only are appropriate for dilute electrolytes and for small electric potentials. Moreover, a model revision within thermodynamics reveals that electrochemistry has overlooked the coupling of the conservation laws for mass and momentum [1-3]. It will be shown the new improved PNP model is capable to predict experimental data qualitatively and quantitatively as well [4,5].

The essential improvement of the new PNP model relies on the incorporation of the momentum balance equation describing the impact of electromagnetic forces on matter. In the boundary layer the electric potential is steeply decreasing within a few nanometers whereby a strong electric field is created. The corresponding force directly acts on the local matter and produces large mechanical stresses. This mechanism will be discussed as a further cause of degeneration for both the electrode material and the solid-electrolyte.

[1] W. Dreyer, C. Guhlke, R. Müller; Overcoming the shortcomings of the Nernst-Planck model; Phys. Chem. Chem. Phys., 15 (2013) pp. 7075-7086.
[2] W. Dreyer, C. Guhlke, M. Landstorfer; A mixture theory of electrolytes containing solvation effects; Electrochem. Commun., 43 (2014) pp. 75-78.
[3] W. Dreyer, C. Guhlke, M. Landstorfer; Modeling of electrochemical double layers in thermodynamic non-equilibrium, Phys. Chem. Chem. Phys., 17 (2015) pp. 27176-27194.
[4] G. Valette; Double layer on silver single-crystal electrodes in contact with electrolytes having anions which present a slight specific adsorption: Part I. the (110) face. J. Electroanal. Chem., 122 (1981) pp. 285-297.
[5] W. Dreyer, C. Guhlke, M. Landstorfer; Theory and structure of the metal/electrolyte interface incorporating adsorption and solvation effects, WIAS Preprint 2058, (2014)

Due to their good expected safety properties, lithium ion batteries based on solid electrolytes provide an attractive alternative to conventional liquid systems. Whereas the ionic conductivity of solid electrolytes has become compatible with the one of liquid electrolytes, the power density of all solid state batteries is still considerable smaller than batteries based on liquid electrolytes. Two potential sources, both related to contact properties of solid electrolytes can be identified. At interfaces between electrolyte and electrode space-charge layers are formed, which give rise to very high electric fields, considerable mechanical stress and strongly increased interfacial resistances. In addition, contact defects in composite electrodes obstruct ionic pathways and decrease the accessible reaction surfaces. Understanding in detail the mechanisms and dynamics of space-charge-layer formation and the influence of contact defects on the overall performance of all solid state batteries will be an important step towards improved solid-state systems with more favorable properties.
In this presentation a new thermodynamically consistent model for the static and dynamics of solid electrolyte is described, which is based on first principles only. Boundary-layer like structures at interfaces are the result of our theory and do not have to be augmented a posteriori as, e.g., in Gouy-Chapman theory for fluids. Analytical and numerical solutions for a solid electrolyte in contact with electrodes are evaluated to study the parametric dependencies of the cation and potential distributions as well as the stress distribution under the influence of varying electrode potentials and varying interface properties. We observe that there are fundamental differences between space-charge layers in fluid and in solid electrolytes. In addition microstructure resolved simulations for composite electrodes are performed to study the influence of contact defects on the detailed charging dynamics and the power density of all solid state batteries.

Dielectric capacitors, in which dielectric materials form the core component, are electrostatic charge storage devices commonly adopted in electronics and electrical power systems. Among the current energy storage technologies, dielectric capacitors feature the highest rate capability and hence possess the greatest power density. The demand for high-temperature dielectric materials toward dielectric capacitors has been driven by the rise of many high power applications such as electric automobiles and pulsed power technologies where the power electronics are exposed to elevated temperature conditions. Polymer dielectrics enjoy inherent advantages of scalability, lightweight, flexibility and the ability to be shaped into intricate configurations in addition to their higher breakdown strengths and greater reliability, compared to their inorganic counterpart, but have low operation temperatures and inferior energy and power densities at high temperatures. On the other hand, switching to inorganic dielectric materials would certainly compromise the intriguing properties exclusively present in polymer dielectrics that are in demand for advanced electronics and power systems. Here we report a novel polymer-based dielectric material by design. At elevated temperatures oriented toward applications such as electric vehicles and pulsed power technologies, these dielectric materials function reliably with unprecedented dielectric and capacitive properties, e.g., a high energy density that is twice as much as that of the state-of-the-art polymer-based dielectric and is even comparable to that of some ceramic dielectrics, and a power density that is significantly greater than that of the best commercial polymer dielectric. The outstanding high-temperature performance implies that these polymer-based dielectrics may become competitive with that of ceramic dielectrics under harsh environmental conditions.

For portable electronics and electric vehicles, high-energy-density capacitors are being actively investigated. Recent efforts focus on nanoscale fabrication and interface engineering with synergistic effects and outstanding properties. In such capacitors, spatial- and time-resolved energy storage and release are still a mystery. Here we explore time-resolved scanning probe microscopy (SPM) to directly image nanoscale dynamics of the energy storage and release in the dielectric capacitors. We investigate P(VDF-CTFE), which has highest energy density among dielectric polymers, create interface by loading a small amount of monodisperse nanoparticles, and manipulate its polar nanoregions to achieve energy density as high as possible. Due to high sensitivity of SPM, both fast and slow energy release processes are identified in the nanocomposite capacitors; moreover, the slow process can be divided into two processes, indicating existence of various energy decay pathways. The spatial map of the first slow decay process shows that the size of polar nanoregions is tailored down by the cross-linking. Most importantly, static and dynamic reconstructions of the slow energy release process reveal the contribution of interphase between the nanoparticles and polymer matrix. Time-resolved SPM technique is attractive for study of energy utilization at the nanoscale level in capacitors, batteries, and fuel cells.

Carbon nanomaterials have proven to be well suited for integration into renewable energy technologies. Supercapacitors, which combine the high energy density of batteries with the power delivery of capacitors, are one class of devices which have benefitted greatly from the incorporation of nanomaterials. In an effort to improve the specific capacitance of these devices, we have produced binder-free electrodes composed of vertically aligned, helically coiled carbon nanotube arrays grown on stainless steel current collectors and compared them to vertically aligned arrays which are not coiled. In this talk, I will detail the results of our experiments involving the coiled arrays. By virtue of their helicity, the coiled nanotubes provide a greater surface area than their straight counterparts, thus improving the specific capacitance. The effects of electrode treatment in an argon plasma are also explored. It is thought that the coiled nanotubes will be more susceptible to plasma induced defects due to the surface being more strained than the surface of their straight counterparts. The plasma treatment may also help to reduce amorphous surface deposits and induce charged defects which alter the fermi level of the nanotubes, helping to overcome the quantum capacitance which is comparable to the electric double layer capacitance in carbon nanomaterials.

Over the last 20 years, electromechanical scanning probe microscopies has emerged as a powerful tool for probing broad range of nanoscale phenomena from ferroelectric domains and polarization switching to ionic transport and reactivity in energy storage and conversion materials. In these techniques, SPM tip concentrate electric field at the tip surface junction, and associated strain due to piezoelectric or electrochemical polarization, ionic motion, or electrochemical reaction are detected by SPM electronics. While the detected hysteretic responses are ubiquitous for multiple material classes, their quantitative measurements and unambiguous interpretation remains complex. In this presentation, I will summarize recent advances in quantitative SPM measurements and existing approaches for information processing in SPM. I will further introduce the approach for full information capture in SPM based on recording and complete analysis of data stream from photodetector. This general-mode (G-Mode) SPM is illustrated for classical SPM modes such as intermittent contact mode SPM, as well as electrochemical strain microscopy and spectroscopy (ESM) and Kelvin probe microscopy. The analysis of the information contact allows deducing in which cases classical signal processing allows unbiased representation of the tip-surface interactions and which it incurs significant information loss. The approaches for full mapping on frequency responses providing complete view of tip-surface interactions are discussed. Finally, I will discuss possible models for strain generation in ionic materials and pathways for their separation based on time- and voltage dependences of response.
This research is supported by and performed at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, BES DOE.

Li-ion batteries are a system that features strong coupling between mechanical stresses and electrochemical reactions. Prior studies on stresses in electrodes are focused on single particles or mono-phase materials. The kinetics and morphology of composite electrodes regulated by the mechanical interactions are much less exploited. We integrate a continuum theory of coupled diffusion and large elasto-plastic deformation into a finite element program. Such a computational tool enables us to explore the intimate coupling between the lithiation kinetics and stresses in three-dimensional electrodes that are composed of multiple components. We find that Li profiles and stress states in multiple particles constrained by a matrix are significantly different from that in a free-standing configuration. The mechanical interactions regulate Li chemical potential in Si nanowires and transform the anisotropic deformation to an isotropic behavior and vice versa. The modeling is in good agreement with a recent experimental report.

The electrodes of electrochemical cells such as solid oxide fuel cells (SOFC) or lithium ion batteries (LIB) are composed of granular materials in order to shorten diffusion paths. Typically, particles of an additional phase are added to ensure sufficient electric and/or ionic transport. The effective transport properties of this phase are governed by the morphology of the assembly and, when present, by the mechanical load applied to the electrode.
In this presentation we will present a particle based modeling approach. First, random assemblies of spherical particles, monosized or according to a given size distribution, are generated such that a certain packing factor is obeyed. This is followed by a step of either so-called numerical sintering (SOFC) or calendering (LIB). In the latter case, the calendering step is simulated by uniaxial compression via the discrete element method. The third step is then to identify percolated clusters of the conducting phase under consideration. Finally, a pair of particles in contact is replaced by a resistor according to a fit law and the effective conductivity is then calculated solving Kirchhoff's laws.
For SOFC, we investigate the range of volume fraction of ion and electron conduction particles such that both phases are percolated for varying ratio of particle radius of the two phases. In particular, we calculate the critical coordination number for the percolation threshold for various scenarios. Also the three phase boundary and the effective conductivity are studied.
For LIB, special emphasis of the work is on the effect of mechanical forces on the effective electronic conductivity of the carbon black phase. Mechanical load is induced by the calendering step during production on the one side and by swelling of active material particles during lithium initercalation on the other side. We show the effect of these processes on the connectivity of the particle network and compute the effective conductivity.

Battery aging and state of health degradation are important problems for Li-ion batteries yet are difficult to measure accurately. In this work, we present a method that links the measured stress state of a pouch cell to its state of health (SOH). We conduct long-term cycling tests on lithium-ion cells, and verify this method under various cycling rates, depths of discharge, and temperatures. The SOH of the batteries are determined intermittently by measuring the capacity at C/10 rate, while the stress is measured by placing an amplified load cell in series with a pouch cell under a designed fixture. The results indicate a linear relationship between measured stress and SOH under normal cycling conditions in agreement with our previous models. By varying the range of cycling, we further verify that this linear relationship can be extended to various charging/discharging profiles, which more accurately simulates real-life battery use. This method not only relates the battery aging to mechanical change of the battery but we will discuss how it also serves as an indicator for the change of the degradation mechanism, which can be used to anticipate premature and catastrophic cell failure.

The prevention of capacity loss after electrochemical cycling is of paramount importance to the development of Lithium-ion batteries, especially for the application in the electric vehicle industry. The objective of this research is to investigate C-rate dependent diffusion-induced-stresses in a continuum system by adapting the thermal stress analysis approach. LiFePO4 is selected as the model system in this study since it is one of the promising cathode materials used for the electric vehicle application. Three different concentration dependencies are incorporated in the finite element analyses: orthotropic elastic constants, volume expansion coefficients, and lithium diffusivities. Six different particle orientations are considered in calculating the average mechanical properties. Our simulation results show that the effect of concentration dependency on mechanical properties and lithium diffusivities cannot be neglected for the mechanical stress prediction. The results of lithium concentration profiles, total strain energies, normal stresses and shear stresses at different C-rates (1C, 2C, 6C and 10C) are compared and discussed. A higher maximum stress at a higher C-rate due to a steeper lithium concentration gradient suggests that particles inside the material may undergo fractures faster and lead to the capacity loss at higher C-rates.

Recently, several novel concepts have been proposed to enhance the performance of thermoelectric (TE) materials and dozens of new TE compounds with high ZTs have been reported. Among them, great interests have been recalled in copper-based TE materials because some unique characters in this family were revealed bestowing them high thermoelectric figure of merits. For example, the liquid-like thermal transport character was found in super-ionic conductors such as copper chalcogenides (Cu_2-δX, where X could be S, Se, or Te), in which the greatly depressed lattice thermal conductivity and tunable electrical transport contributed the large ZTs in this family. The integrated computation-aided optimization of electrical transport also supplies an effective way to explore high performance TE compounds diamond-like structure. The greatly enhanced thermoelectric properties during phase transition by critical scattering is another new mechanism for tuning TE transport. The physical mechanisms behind these abnormal thermoelectric properties raised great interests in thermoelectric basic research. For the practical application as TE materials, great challenges are still remained, such as the poor service behaviour of these copper-based compounds because of the significant migration of copper ions under electric field and high temperatures. Some recent progress and discussions on the copper ion migration and its modulation will be also presented in this talk.

Scanning probe microscope (SPM) is a versatile instrument for material characterization. Soon after invention of SPM, researchers tried to probe thermal properties using scanning thermal microscopy (SThM) which is essentially a SPM-based technique. SThM is now used for thermometry, data storage, and measuring thermophysical properties, such as thermal conductivity and specific heat. One of the main difficulties in SThM is the long time-constant for reaching the equilibrium which is associated with the length-scale of the system. Therefore, fast and accurate probing of thermal properties requires minimal heating area. Recently developed thermal probes can produce a substantial amount of heat at a tip radius of tens of nanometers and hence, offer higher measurement accuracy.

Despite all the advances in the production of thermal probes and measurement techniques, quantitative measurements of thermal properties at nano-level still remains challenging. Moreover, advances in nano-camposite materials require higher resolution probing of thermal properties. Nano-composite thermoelectric (TE) materials, for example, showed much higher conversion efficiency by reducing local thermal conductivity and increasing the Seebeck coefficient due to low dimensionality. Therefore, characterizing phonon and electron transport with high temporal and spatial resolution is critical for understanding TE material behavior.


Our research is toward developing experimental and numerical tools capable of measuring the electron and phonon transport phenomena in TE. The thermal conductivity is characterized through using 3ω-method. In the 3ω-method, under an alternating joule heating and due to linear proportionality of the probe electrical resistance to temperature, the third harmonic component of the probe voltage is also linearly dependent on temperature. The third harmonic response of the probe voltage can be used to characterize local thermal properties which is independent of the ambient and probe temperature. Thus, the uncertainty in the 3ω-method measurement is significantly lower than other methods, resulting in an increased range of thermal conductivity measurements. In order to analyze the dynamics of the system, we perform 3D finite element analysis (FEA) to compute the steady periodic solutions in the frequency domain. Using the FEA as an inverse calculation tool, the experimental parameters can be calibrated and optimized for an enhanced measurement of thermal properties such as thermal conductivity. This will further lead to quantitatively measure the thermal conductivity in the SThM experiments.

Our approach will provide thorough understanding of interfacial and functional gradient electron and phonon scattering mechanisms at sub-30 nm spatial resolution. Moreover, it will allow for higher efficiency TE material advances by connecting inhomogeneous distribution of local properties with structural heterogeneity.

As commercial interest in flexible power-conversion devices increases, the demand is ever growing for high-performance alternatives to brittle inorganic thermoelectric materials. As an alternative, we propose a rationally designed graphene/polymer/inorganic nanocrystal free-standing paper with high-TE performance, high-flexibility, and mechanical/chemical durability. The ternary hybrid system of graphene/polymer/inorganic nanocrystal has two heterojunctions to induce energy filtering effect, thus increasing the electrical conductivity without a major decrease in the thermopower. The ternary hybrid shows a power factor of 143 μW m^-1 K^-1 that is 1–2 orders of magnitude higher than those of single- or binary-component materials, and it shows the maximum dimensionless figure-of-merit of 0.2 at 300 K. We believe that the strategy proposed here to improve the performance of flexible TE materials by introducing more heterojunctions and optimizing carrier transfer at those junctions shows great potential for the preparation of flexible/or wearable power-conversion devices.

Recent advancement in flexible and wearable electronic devices has made flexible and light-weight power delivery systems in urgent demand. Among all the energy conversion systems for power delivery, devices that convert low-grade waste heat from human body or other power consuming systems to electricity are under intensive research. For accomplishing compact and silent energy conversion without mechanically moving parts, Seebeck effect, Soret effect and/or change of chemical potential at different temperatures can be employed. However, generation of voltage in a material by creating a temperature difference has been mainly achieved with heavy, brittle, and toxic inorganic semiconductors. Additionally, systems exploiting the Soret and chemical potential effects are typically implemented in fluid. These drawbacks prompt researchers to develop systems based on organic counterparts as an alternative for more viable and efficient energy conversion. Though recent researches on organic p-type thermoelectric composites showed feasibility of increasing electrical conductivity while keeping thermopower and thermal conductivity relatively constant, n-type organic materials have been seldom reported and their performances is not satisfactory with regards to practical applications.
Here, we demonstrate that hybrids of carbon nanotubes (CNTs) and poly(3,4-ethylenedioxythiophene) (PEDOT) treated by tetrakis(dimethylamino)ethylene (TDAE) can generate a large n-type voltage in response to temperature difference. We aim to increase the thermopower of PEDOT by using chemical reduction of PEDOT with TDAE. In addition, electrical conductivity of PEDOT is further improved by using CNTs as additives or main ingredients without sacrificing mechanical flexibility and light weight. With hybrids of CNTs and PEDOT, we exploit the high mobility of CNTs for improving electrical conductivity as well as TDAE treatment for controlling the concentrations of electrons and ions for obtaining high thermopower. The thermal conductivity is kept at a relatively low level so as to achieve overall figure-of-merit improvement.
Our PEDOT/CNT composites yielded a remarkable increase in thermopower with TDAE treatment without significantly sacrificing electrical conductivity owing to the high electronic mobility of CNTs, resulting in a large power factor, 1050 µW/m-K². The high thermopower could be attributed to the largely reduced concentration of major electronic carriers by the TDAE treatment as well as the chemical potential difference of TDAE ions created by temperature difference. The figure-of-merit at room temperature is estimated to be up to ~0.4, which is superior to p-type organic materials. We expect that our methodology and high performance material provide an effective new route for developing organic composite with desired electronic and thermal transport properties.

Thermoelectric devices are typically rigid and their performance and reliability could be compromised by thermal stress. There is also a growing need to develop mechanically flexible thermoelectric devices to be integrated into wearable electronics. By using inherently soft polymers, hybrid organic/inorganic composites are considered a promising candidate for flexible thermoelectrics. In addition, the composites also provide an opportunity for enhanced thermoelectric performance by taking the advantages of low thermal conductivity and high Seebeck coefficient from the organic and inorganic components, respectively. Experimentally, however, enhanced thermoelectric performance in hybrid composites has not been observed, primarily due to inefficient carrier transport across the organic/inorganic interfaces. In particular, if the composites are energetically heterogeneous, the more conductive component dominantly determines the thermoelectric performance of the entire composites and the less conductive component behaves as a resistor, preventing higher power factor from the hybrid thermoelectric materials compared to their single-component counterparts.

Here we demonstrate a strategy for interface engineering to improve the charge carrier transport from a conducting polymer, i.e., poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS) to nanostructures of a prototypical inorganic thermoelectric material, i.e., Bi_0.5Sb_1.5Te₃ (BST). Conduction among PEDOT:PSS is mediated by a phonon-assisted hopping process that requires an activation energy due to the delocalized carriers within the backbones of PEDOT. Therefore, forming uniform dispersions of BST nanostructures in the polymer matrix minimizes the discrepancy in the activation energy by reducing the hopping distance for transport of both PEDOT-to-PEDOT and PEDOT-to-BST, yielding enhanced contribution of BST to carrier transport for higher Seebeck coefficient. We achieved homogeneous dispersion of BST in the PEDOT matrix and consequently efficient interfacial charge transport by adopting a non-ionic surfactant and controlling the micelle concentration, and observed simultaneous increase in electrical conductivity and Seebeck coefficient. Therefore, these hybrid BST/polymer composites are promising for flexible and high-performance thermoelectrics.

Heat is the natural by-product of energy conversion processes. Of the 4.25 × 10²⁰ J of energy the United States consumes every year, more than 60% is wasted in the form of heat. Therefore, waste heat recovery is a crucial step to improve the energy generation and utilization efficiency. Thermoelectric (TE) materials, which can directly convert rejected or waste heat into usable electric power, has been extensively developed for this issue. Thick-film-based devices have advantages over conventional TE module because of its compact size. By shrinking the size of thermoelectric devices, it not only allows the device to operate under smaller temperature gradients, but by so doing it expands its capability to handle a wider range of thermal and power management microelectronic systems. The combination of electrochemical deposition of compound semiconductors (metal tellurides) with standard integrated circuit technique makes fabrication of thermoelectric microdevices possible.

In our work, we demonstrated a rapid deposition process (180 µm/h) in alkaline solution to synthesize dense thick PbTe film (>50 µm) for thermoelectric applications. Compared to the commonly used acidic baths with deposition rate about 4 µm/h, our electrodeposition process enhanced the film growth rate 48 folds higher. The electrodeposition mechanism of PbTe was examined. The thermoelectric property of thick PbTe films were optimized by introducing energy barriers from crystal grain boundary and tellurium nanoinclusion, which was achieved by tailoring composition and crystal structures of the thick PbTe films. Additionally, control over the composition and crystal structures was realized by tuning the electrodeposition conditions, such as the ratio of [TeO₃^2-], [Pb²⁺], [EDTA^4-]/[Pb²⁺], solution pH, and applied potential, as well as post-annealing process.

Thermoelectric devices have seen increased performances in the past decade through various techniques to improve the material’s figure of merit ZT. With these improved material performances, the challenge lies to translate them into high performing and reliable devices to enable widespread applications. Materials used in creating thermoelectric devices must consider thermal, electrical, and mechanical performances.

Managing thermal stresses at the interfaces is usually one of the most important challenges for thermoelectric generators (TEGs), which is especially true for devices working at intermediate or high temperatures. The mismatch of coefficient of thermal expansion (CTE) between the metal contact and the thermoelectric materials can result in significant thermal stresses at the interfaces and a high risk of device failure.

In this study, a new method is presented to achieve optimal CTE match and minimum stresses using metal layers directed bonded to a ceramic. The TEG based on nanostructure half-Heusler materials is selected for this study due to its high thermoelectric performances at intermediate and high temperatures.

By optimizing the design of the directly bonded metals, the effective CTE of the metal contact can be modified to closely match the thermoelectric materials despite of significant mismatches between the intrinsic metal CTE the TE materials CTE. The thermal stresses show significant reduction compared with the baseline TE device without adopting the approach.

This work provides a promising approach to fabricate high performing and mechanically robust thermoelectric devices using a wide range of thermoelectric materials.

In recent past, considerable attention and development have been made in green energy harvesting technologies due to restriction on CO₂ emission. Thermoelectric devices are explored as one of possible solution to utilize waste heat generated from industrial process, household usage, heat engines, and solar conversion.
Studies on thermoelectric devices are mainly focused on to the solid-state devices including p and n semiconductor materials. Thermogalvanic cells constitute another family of thermoelectric device. Thermogalvanic cells are attractive due to their simple manufacture, architecture, and the absence of toxic metals.
Thermogalvanic cell is an electrochemical device which allows direct conversion of thermal energy to electrical energy. It consists of an electrolyte with redox couple and two identical electrodes at a different temperatures. Most studied thermogalvanic cell is electrolyte containing aqueous 0.4 M ferri/ferrocyanide redox couple with platinum electrodes. In this system temperature difference creates a difference in the redox potential of the ferri/ferrocyanide electrolyte at the platinum electrodes allowing to generate power.
In order to make thermogalvanic cell commercially available, the electrode material has to be replaced with a low cost and efficient material. Conducting polymers are promising candidates for electrodes because they intrinsically transport both electronic and ionic charge carriers. Furthermore, conducting polymers can be synthesized from solution with higher conductivity, stability and porosity. We show that the power generated with Poly(3,4-ethylenedioxythiophene)-tosylate (PEDOT:Tos) electrodes increases with the polymer thickness and reaches values similar to a flat platinum electrode.

Magnesium silicide and stannide solid solution is a promising thermoelectric material for intermediate temperature range due to high performance as well as non-toxicity and abundance. Large number of electronic structure valley degeneracy combined with low electron effective mass makes its power factor as high as that of bismuth telluride, one of the best thermoelectric materials. Moreover the low thermal conductivity, which is beneficial for driving a high voltage, is achieved by atomic scale alloy scattering between randomly distributed silicon (Si) and tin (Sn) atoms. However we recently found that this ideal solid solution, which is crucial for obtaining high valley degeneracy and maximized alloy scattering, is meta-stable material due to miscibility gap inducing the segregation into Si-rich and Sn-rich phases. Although the overall composition between Si and Sn in a material system is 4:6, known to be the best, the microscopic phase segregation would reduce the valley degeneracy and alloy scattering so that its thermoelectric figure of merit (ZT) is decreased significantly both electrically and thermally. In order to provide a solution to this shortcoming, we first investigated this aspect systematically. By synthesizing the material with different time of post annealing we observed the qualitative trend that thermal conductivity increases with post annealing time. In addition, the experimental observation was qualitatively analyzed using the calculated thermal conductivity based on modified Callaway model so as to reveal the quantitative population of segregation in a sample. As a key strategy to compensate the deterioration due to segregation we embedded TiO₂ nano particle and obtained the suppressed thermal conductivity followed by higher ZT ~1.1. Theoretical prediction using our model revealed the highest ZT as much as 1.8, when the sample retains the ideal solid solution without any segregation. Lastly the novel strategy and future direction for the prevention of phase segregation will be given.

Due to new research efforts in the field of thermoelectrics with a focus on size effects, there is a growing need for measurement setups dedicated to samples with small geometrical dimensions like thin films and nanowires with considerably different physical properties than bulk material. The characterization of these samples is important to learn more about their structure and conduction mechanism but also important for technical applications e.g. in the semiconductor industry.
We report on the development of a new system to simultaneous measure the electrical and thermal conductivity, the Seebeck Coefficient and the Hall Constant of a thin film sample in the temperature range from liquid nitrogen up to 350°C. Due to the nearly simultaneous measurement at only one sample, errors caused by different sample compositions, different sample geometries (thickness) and different heat profiles can be avoided.
The system consists of two main parts, a structured Si-wafer and a suitable measurement setup. To measure the el. conductivity and the Hall constant, the wafer owns a structure with four electrodes to use either the Van-der-Pauw or Hallbar method. For the Seebeck measurement an additional temperature gradient can be applied and two of the four electrodes are used to measure the thermovoltage between them. The temperatures for the Seebeck calculation are measured with resistance thermometers, also deposited on the chip. The thermal conductivity can be measured in plane and cross plane using the Völklein Method [1], doing a steady state or transient measurement. Therefore a small heating/sensing stripe is deposited on a very thin nitride membrane. On the backside of this membrane the sample has to be deposited. To get a correct result, the measurement has to be done under vacuum (no heat transfer by convection) in a thermally stabilized and controlled chamber (to avoid heat transfer by radiation, the ambient temperature has to be identical with the chip temperature).
In order to meet these requirements a suitable vacuum chamber with sample holder and necessary ports has been designed. The sample holder can be cooled with liquid nitrogen and heated by joule heating on both ends to create either a constant temperature or a defined temperature gradient. To measure the Hall constant, the chamber is put between two spools of an electro magnet to apply a variable magnetic field with a maximum of +/-1 T.

Thermoelectric (TE) materials are of significant technological interest as they can convert waste heat into electrical power. There is a need for new high-performance TE materials that are inexpensive, non-toxic and earth abundant due to the world’s demands for green energy generation. Copper containing sulphides are potentially low cost and environmentally friendly TE materials. In the last decade, some of them have been reported as promising TE materials, e.g. Cu₁₂Sb₄S₁₃ (Tetrahedrite) and Cu₂₆V₂M₆S₃₂ (Colusite), while most of them are still unexplored. In this work, we use high-throughput screening technique to aid the development of new TE sulphides. Firstly, crystal structural features are used as ‘descriptors’ to screen of the 1600+ Cu/S containing entries in the International Crystal Structure Database (ICSD). As a result, 84 phases are identified as potential high performance TE materials. Secondly, some of the identified compounds, which have never been investigated as TE materials, were fabricated as single-phase ceramics using ball milling mechanical reaction and spark plasma sintering (SPS). Transport coefficients measurements indicate that they are promising TE materials, which verifies our screening criterion. Thirdly, the crystal structural relationships between the identified 84 phases were analyzed and then used to guide the design of new TE sulphides. Based on the analysis and first principles calculations, two approaches to design new TE sulphides were discussed: 1) High-pressure synthesis of metastable phase; 2) In-situ 2-phase composites with coherent interface.

Thermal transport in silicon nanowires has captured the attention of theoretical scientists for understanding phonon transport at the nanoscale, and the thermoelectric figure-of-merit reported in rough nanowires has inspired engineers to develop cost-effective waste heat recovery systems. Thermoelectric power generators composed of silicon target high-temperature applications due to improved efficiency beyond 550 K. However, there have been no studies of thermal transport in silicon nanowires beyond room temperature. Moreover, high-temperature thermal transport measurements enable studies of unanswered questions regarding the impact of surface boundaries and varying mode contributions as the highest vibrational modes are activated (the Debye temperature of silicon is 645 K). Here we develop a technique to investigate thermal transport in individual nanowires up to 700 K. We use high-purity silicon nanowires and thermal transport models to assess the unique features of high-temperature phonons including the optical mode contribution. We verify temperature-dependent thermal transport in rough silicon nanowires using established synthesis techniques, which allow investigation of surface roughness effects with respect to vertical and lateral roughness scales. The rough nanowires show a significant thermal conductivty reduction throughout the temperature range (20 K – 700 K), demonstrating a potential for efficient power generation. The high temperature measurements can not only extend the available knowledge in phonon transport to a new temperature regime but can also provide new insight into unresolved phenomena from the past research that only explored up to room temperature. The metrology opens up opportunities to study high-temperature phenomena in nanomaterials including one- and two-dimensional structures. New findings may make significant advances in high-temperature relevant energy conversion systems such as thermoelectric power generators. The estimation based on the thermal conductivity of rough nanowires predicts a commercially-competitive thermoelectric efficiency at 700 K, which adds a credential to developing cost-effective waste heat recovery systems using silicon nanostructures.

Small molecules with functional groups can show different electron affinity and binding behavior on nanomaterial surface, which, in principle, could be used to alternate the electrical transport in nanocomposites by changing carrier type, density, and mobility. These small molecules can also serve for scattering the phonons to reduce the thermal conductivity. In this presentation, we show an effective way to significantly improve the thermoelectric properties in nanocomposites by several types of surface molecules, which suggests that the surface-bound small molecules could serve as a beneficial component to build high-performance nanocomposite thermoelectric devices operating at low temperature even without sintering.

In recent years, hierarchical structures have been intensively studied as an effective approach to tailor the electron and phonon transport inside a bulk material and thus achieve high thermoelectric (TE) performance [1-3]. In physics, the effectiveness of a TE material is evaluated by its dimensionless figure of merit (ZT), defined as ZT=S²σT/k, where S, σ , k, and T represent Seebeck coefficient, electrical conductivity, thermal conductivity, and absolute temperature, respectively. Here the thermal conductivity k can be split into two parts, the lattice (phonon) contribution and the electronic contribution. With atomic to micro-scale structures in a bulk material, the lattice thermal conductivity can be effectively suppressed across the whole phonon spectrum, while maintaining or somewhat enhancing the power factor S²σ. This leads to enhanced ZT in various bulk materials.

Despite many encouraging experimental results, accurate lattice thermal conductivity predictions are still challenging for a bulk material with hierarchical structures. Earlier studies by either numerically solving the Boltzmann transport equation or phonon Monte Carlo (MC) simulations mainly adopted a frequency-independent phonon mean free path (MFP). Further incorporating frequency-dependent phonon MFPs are complicated and computationally expensive for MC simulations, which hinders the potential ZT enhancement by optimizing the hierarchical structures in materials design. In this aspect, effective medium formulation (EMF) has been proven as a powerful tool for accurate and effective predictions of the lattice thermal conductivities, as demonstrated for nanoparticle-embedded bulk materials in frequency-independent analysis [4] and polycrystals for general cases [5,6]. However, a general EMF to consider hierarchical structures is still lacking, particularly when and frequency-dependent phonon MFPs and phonon transmissivity across nanostructured interfaces are considered. In this work, we extend the existing EMF to widely studied SiGe nanocomposites and validate our modeling with frequency-dependent phonon MC simulations. The applications of this EMF in other TE materials will be discussed. This EMF will greatly facilitate the research for TE and other relevant fields, where hierarchical structures can be introduced to effectively tune the phonon transport inside a material.

1. A. J. Minnich et al., Energy Environ. Sci. 2 (5), 466 (2009).
2. B. Poudel et al., Science 320 (5876), 634 (2008).
3. K. Beswas et al., Nature 489, 414 (2012).
4. A. Minnich and G. Chen, Appl. Phys. Lett. 91, 07305 (2007).
5. Q. Hao, J. Appl. Phys. 111 (1), 014307 (2012).
6. Q. Hao, J. Appl. Phys. 116, 034305 (2014).

Coherent scattering of phonons by sub-micron feature size Phononic Crystals (PnCs) at room temperature is studied in this work. Different silicon/air-holes PnCs with different unit cell size with the same critical dimensions are fabricated and the in-plane thermal conductivity reduction is measured. Calloway-Holland formulation with band structures of silicon and PnCs is used to explain the reduction in the thermal conductivity. Initially, the most conservative classical porosity models are employed to correct the Calloway-Holland formulation with assumption of only incoherent scattering of phonons. It is revealed that the assumption of all incoherent scattering of phonons does not suffice to explain the measured reduction in the measured thermal conductivity. Comparison between the experiment and the Calloway-Holland formulation suggests that a fraction of phonons with longer mean free paths undergoes the coherent scattering. This study has future applications in thermoelectric for reduction of thermal conductivity and enhancing ZT figure of merit.

We elucidate thermal conductivity along the screw dislocation line, which represents a transport direction inaccessible to classical theories. By using equilibrium and non-equilibrium molecular dynamics simulations, and the atomistic Green function method, we uncover a Burgers vector dependent thermal conductivity reduction in Si, SiC, PbSe, and SiGe nanowires^1-3. The effect is uncorrelated with the classical theory of Klemens⁴. The influence of dislocations on thermal transport originates in the highly deformed core region, which represents a significant source of anharmonic phonon-phonon scattering. High strain reduces the phonon relaxation time, especially in the longitudinal acoustic branches, and creates an effective internal thermal resistance around the dislocation axis. The effect can be distinguished from the thermal transport reduction caused by the nanowire surface imperfections and vacancies. Our results have implications for designing materials useful for high-temperature electronics and thermoelectric applications.

Thermal management in nanostructures for energy applications has been identified as matter of critical importance to further advance the performance of these applications beyond the state-of-the-art limit. Phonon engineering in nanostructured semiconductors has been shown to improve the efficiency of thermoelectric systems, by reducing the thermal conductivity of the crystalline materials while preserving their electronic properties. When the system size reaches nanoscale, the phonon transport in the material is greatly affected by the surface nanoscale character [1]. The concept of nanophononic metamaterial (NPM) was introduced recently [2] to engender the emergence of unique subwavelength properties at the nanoscale with the inclusion of local surface resonators, and thereby, to affect the nanoscale thermal transport. Ultrathin silicon thin membranes are chosen as the foundation material for the NPM and the resonators are modeled as a periodic array of nanoscale pillars that extrude off the surface of the membrane.

We carried out a systematic investigation of the impact of band structure hybridization on phonon scattering and consequentially, on phonon transport in locally resonant silicon-based NPMs. We used classical molecular dynamics (MD) under equilibrium and non-equilibrium conditions as well as an approach based on the Boltzmann transport equation with the relaxation time approximation to investigate the nature of phononic thermal transport in nanopatterned silicon membranes with thicknesses of the order of 10 nm and below. The phonon relaxation times are calculated using anharmonic lattice dynamics using Fermi’s Golden Rule and taking into account the contribution of three-phonon processes. We find that the presence of the local surface resonators has a significant effect on the phonon dispersion and has a direct consequence of suppression of group velocities of phonons in the nanopatterned silicon membranes. However, the effect of the resonators is less prominent on the phonon relaxation times in these membranes. We also probed the imprint of the nanoscale characters of the resonators, namely the geometry and material composition on the local resonances triggered by them. We completed the investigation by relating the nanoscale resonant character with the phonon scattering, and consequently, the phonon transport in the locally resonant silicon membrane NPMs.

[1] S. Neogi et al, “Tuning Thermal Transport in Ultrathin Silicon Mem- branes by Surface Nanoscale Engineering.” ACS nano, 9(4), 3820-3828 (2015).
[2] B. L. Davis and M. I. Hussein, Physical review letters 112, 055505 (2014).

Acknowledgment: This project is funded by the program FP7-ENERGY-2012-1-2STAGE under contract number 309150.