Symposium EN10—Advanced Materials for Thermal Energy Management and Harvesting

Solid-state thermal diodes are bottlenecked by poor diodicity while phase-change thermal diodes are constrained by either a gravitational dependence, a 1D configuration, or poor durability. Here, we invent a novel bridging-droplet thermal diode which bypasses all of these existing shortcomings. Our diode consists of two opposing copper plates separated by a micrometric-thick insulating gasket; one plate contains a superhydrophilic wick structure while the other contains a smooth hydrophobic coating. In the forward mode of operation, liquid water evaporates from the heated wicked plate and vapor condenses onto the hydrophobic plate. The large contact angle of the dropwise condensate enables droplet bridging across the gap to replenish the wicked evaporator, sustaining the phase-change heat transfer. Conversely, in the reverse mode, our diode is an excellent thermal insulator as the heat source is now on the back face of the hydrophobic plate. A lumped resistance thermal model is developed to quantify the overall heat transfer coefficient across our thermal diode. Results show that the overall heat transfer ratio (i.e. diodicity) can be boosted considerably through careful design of the wick structure and gasket thickness.

Boiling heat transfer is a highly effective but stochastic process where the working fluid undergoes vigorous liquid to vapor transition. Physics-based modeling of bubble dynamics during boiling is challenging due to the drastic changes in system parameters, such as nucleation, bubble morphology, temperature, pressure, and velocity. The vast amount of data from both experiments and numerical simulations have led to the use of artificial intelligence (AI)/deep learning (DL) to gain valuable physical understanding of complex phenomena occurring during natural and industrial processes, including reduced order modeling, turbulence closure model, flow control, and extraction of governing equations. Real-time analysis of bubble statistics in pool boiling is challenging at high heat fluxes where vigorous bubble coalescence is observed. Unsupervised deep learning approaches, such as principal component analysis (PCA), clustering, and autoencoders, have the advantage of minimal human intervention unlike supervised learning. The PCA is a versatile data-analysis tool to determine low-dimensional representations of high-dimensional images by extracting dominant variations and patterns in the raw data without any supervision. In nuclear reactors and other boiling heat transfer applications, early forecasting of the boiling crisis can aid in mitigating catastrophic damage of the heater surface. To predict the future states of the boiling phenomena, long short-term memory (LSTM) neural network is a very efficient tool to estimate time-series variations in the input data, such as bubble dynamics.
In the present study, PCA is used to extract hidden physics from boiling heat transfer processes using in-house pool boiling experimental images of the bubble dynamics. The principal component (PC) representations are made in the new low-dimensional coordinate system by determining the dominant correlations within the high-dimensional data. The first few dominating principal components are used to reconstruct the instantaneous bubble shape and size, which drastically reduces the dimensions of boiling images. The results show that the dominant frequency of the first PC increases with heat flux in the discrete bubble regime until it reaches a peak and then decreases with heat flux in the bubble interference and coalescence regime where the former is associated with the increase in bubble nucleation sites and the latter is associated with the bubble coalescence during pool boiling. Image segmentation is also performed to capture bubble statistics, such as size and numbers to verify the findings of PCA analysis. The temporal variations of the PCs are then used as training data for long short-term memory (LSTM) which predicts the future variations of the PCs. Future predictions of PCs are used to reconstruct the bubble images demonstrating the capability of deep learning (DL) model to predict bubble dynamics. Both steady state and transient pool boiling cases are studied using PCA-LSTM DL algorithm and trained DL model can predict bubble dynamics well.

By operating in a high-temperature, high-pressure environment, the efficiency of a thermodynamic cycle can be significantly improved. However, extreme operating conditions require material and design innovations on the classic heat exchanger component. Previous work has attempted to develop heat exchangers for extreme environments by employing superalloy materials in traditional architectures, leading to costly fabrication and limited power density improvement. In this work, we present the design of a multiscale silicon carbide heat exchanger capable of operating in extreme conditions (1300 °C, 250 bar). Enabled by ceramic co-extrusion and geometrically optimized for both high thermal performance and mechanical strength during steady state operation, our heat exchanger offers a scalable and effective solution to address traditional material limitations while offering promising efficiency and performance improvements compared to current state-of-the-art heat exchangers.

Making diffusion-resistant and thermally stable electrical and thermal contacts to the hot side of a thermocouple-based thermoelectric generator (TEG) is a major technological challenge. Such contacts are not needed in transverse thermoelectric devices, in which a thermal gradient in a single material with two cold end contacts induces a perpendicular voltage. Furthermore, the contact resistances of multi-thermocouple (thermopile) generators contribute irreversible losses, to the extent that the device figure of merit ZT of such a thermopile-based TEG is often much lower than the materials zT of the n-type and p-type legs. In a transverse TEG, only two voltage contacts are needed, reducing the irreversible losses in the contacts and bringing ZT much closer to the material transverse thermoelectric figure of merit, zxy¬T. Until now material zxyT values have always been too low for practical applications. When cut at a particular angle, single-crystals of Re4Si7 have a zxyT of 0.7±0.15 at 980 K. A combination of thermal and electronic measurements of multiple single crystals with different doping levels, density functional theory calculations, and analytical models establish that these large zxy¬T values come from anisotropic thermopowers that arise from thermal population of the highly anisotropic valence band and isotropic conduction band. A transverse Re4Si7 TEG was constructed with no contacts on the hot side, and the ZT derived from measurements of its thermal efficiency near 600 K fits the values of zxy¬T well. This proves experimentally that contact losses are minimized in this geometry as well. Hopefully, thus, transverse TEGs open a new approach to circumventing two of the major technological problems that stand in the way of making practical TEGs.

Semiconducting single walled carbon nanotubes (s-SWCNTs) have demonstrated promising performance in low temperature thermoelectric thin films. Like polymer-based organic electronic materials, charge carrier density and in-turn thermoelectric properties, can be tuned using molecular charge-transfer dopants. In this work, we build upon previous demonstrations of p-type doping by employing a series of novel molecular dopants based on dodecaborane (DDB) clusters. The unique chemical and electronic structure of these DDB clusters, compared to oxidants like triethyloxonium hexachloroantimonate (OA) and 2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane (F4TCNQ), increases the spatial separation between the negative DDB counter ion and the hole carriers injected into the nanotubes. Thereby increasing the mobility of the charge carriers and yielding an increase in the thermopower (Seebeck coefficient) for a given electrical conductivity. This enhanced transport ultimately increases the thermoelectric power factor to surpass the previous best-in-class for enriched s-SWCNT thin film networks.

Although great progress has been made in thermoelectric conversion, the performance of thermoelectric generators is still not good enough for large-scale practical application. The conventional thermoelectric generators utilize heat only from the temperature gradient through the Seebeck effect. Apart from the temperature gradient, the temperature related to waste heat always fluctuates. Here, I will report our work on developing hybrid ionic-electronic thermoelectric converters that can harvest generate electricity from both temperature gradient and temperature fluctuation. As a result, they can generate output power at least 4 times of the conventional thermoelectric generators.

Thermoelectric technology converts heat into electricity directly and is a promising source of clean electricity. Commercial thermoelectric modules have relied on Bi₂Te₃-based compounds because of their unparalleled thermoelectric properties at temperatures associated with low-grade heat up to 550 K. However, the scarcity of elemental Te greatly limits the applicability of such modules. Here we report the performance of thermoelectric modules assembled from Bi₂Te₃-substitute compounds, including p-type MgAgSb and n-type Mg₃(Sb,Bi)₂, by using a simple, versatile, and thus scalable processing routine. For a temperature difference of 250 K, whereas a single-stage module displayed a conversion efficiency of 6.5 ⁰/₀, a module using segmented n-type legs displayed a record efficiency of 7.0 ⁰/₀ that is comparable to the state-of-the-art Bi₂Te₃-based thermoelectric modules. Our work demonstrates the feasibility and scalability of high-performance thermoelectric modules based on sustainable elements for recovering low-grade heat.
Reference: P. Ying et al., Nature Communications 12, 1121 (2021)

The development of miniature (~ 0.1 cm² area) silicon integrated circuit (IC) sensors and networking devices for internet-of-things (IoT) and biomedical electronics has spurred the question of how to make such devices energy autonomous, i.e., how to reliably and sustainably energize such devices when they are embedded in remote environments that are often dark, have no access to wall plug power, and are not routinely accessible for battery replacement. Recently, interest has developed in small microelectronic thermoelectric generators (µTEGs) as one method to give IoT and biomedical devices energy autonomy wherever a reliable thermal gradient exists. Most current research on TE technology concentrates on developing new materials having a high TE figure-of-merit ZT = (S²σ/κ)T, where S, σ, and κ are the material’s thermopower, electrical conductivity, and thermal conductivity, and T is the mean operating temperature. Doped Si is known to have a poor ZT ~ 0.01 due to its high κ. The focus on complex materials having high ZT ~ 1 is because a TEG’s ideal thermodynamic efficiency increases with the ZT of the materials used to form the thermopile. However, high ZT materials can be expensive to synthesize, often contain toxic or non-earth-abundant elements, and are generally incompatible with Si IC processing, all of which can substantially increase the cost of integration with standard IC devices.

We report on small area (<< 1 mm²) µTEGs using doped Si and Si_0.97Ge_0.03 as the TE material, all fabricated on a standard industrial Si IC process line following the same protocols and process flows used to make commercial IC devices. These µTEGs can generate specific power densities (per unit area for heat flow per square of temperature difference, ΔT) greater than 80 µWcm^–2K^–2, which is comparable to or better than most existing bulk TEGs using high ZT materials. Moreover, these small area Si µTEGs can generate voltages exceeding 1.5 V with several µA of current using commonly encountered ΔTs ~ 20 to 25 °C. This voltage and current level is sufficient to properly energize some existing commercially available IoT ICs using only a warm copper rod as the energy source. Because these µTEGs can be directly integrated on-chip in the same process flow with the circuits they support, they provide one solution to energy autonomy at extremely low marginal cost.

These Si-based µTEGs build on an alternative approach to µTEG device design that emphasizes application of device physics and circuit engineering principles to optimize a µTEG’s generated power per unit area at any given ΔT, rather than narrowly focusing on thermodynamic efficiency. This strategy exploits the ability of industrial Si processing to fabricate thermopiles consisting of a very large number of TE elements in a small area, thereby producing a high total power and voltage density despite relatively low efficiency per TE element. Modern Si processing is also very good at controlling parasitic thermal and electrical resistances, thus minimizing extrinsic degradation of overall TEG performance. Experimental results on power and voltage generation as well as Peltier cooling will be presented. Physics-based models for optimizing performance will also be discussed at both the material level (e.g., effects of dopant concentration and small percentage Ge alloying) and the device design level (e.g., effects of parasitic electrical and thermal resistances and proper design of thermopile packing fraction).

The flexible thermoelectric generator (f-TEG) is a very promising technology for energy harvesting to enable self-powered wireless sensors and wearable devices, an area of exponential growth. Here, we present high-throughput and scalable additive manufacturing methods to fabricate f-TEGs using colloidal nanocrystals, and rapid screening and optimization of doping to achieve unprecedented thermoelectric performances. Flexible TE films with gradient doping concentrations were printed using a high-throughput aerosol jet printing method. A scanning probe method is applied to quickly identify the optimal doping concentrations. Highly scalable extrusion printing (or screen printing) method is applied to print relatively thick films to achieve high thermoelectric efficiency and enable practical device applications. Along with the nanoparticle ink printing, we developed innovative interface engineering and sintering method in order to improve the density and charge carrier mobility of the printed films.
The p-type and n-type printed films demonstrate peak ZTs of 1 and 0.8 near room temperature along with superior flexibility, which is among the highest reported ZT values in flexible thermoelectric materials. A flexible thermoelectric device fabricated using the printed films produces a high power density above 19 mW/cm² with 80 K temperature difference between the hot side and cold side. The high-throughput and scalable printing methods can efficient and low-cost manufacturing process to transform high-efficiency colloidal nanocrystals into high-performance and flexible TEG devices for broad range of applications.

Reducing phonon or lattice thermal conductivity is one of the most straightforward ways to improve thermoelectric performance of a material. Introducing phonon scattering easily lowers lattice thermal conductivity but in order to improve thermoelectric performance it must reduce electron mobility to much a lesser extent. While this has been quantified for some point defect scattering examples, it is rarely considered for other scattering mechanisms.
Interfaces such as grain boundaries and precipitates are well known to reduce thermal conductivity but their effect on mobility is poorly studied both experimentally and theoretically. Dirty high-angle grain boundaries are known to dramatically reduce mobility in Mg2Si and Mg3Sb2 making some samples unusable as thermoelectrics while clean, low angle grain boundaries can be as good as single crystals. There are many ways to engineer various types of clean, epitaxy-like interfaces in bulk materials.
It has recently become apparent that the low thermal conductivity in PbTe that enables zT ~ 2 is largely due to the lattice softening correlated to defects that produce internal strain rather than directly scattering phonons. New revelations about heat transport further indicate that heat is transported diffusively and not even by phonons in many good thermoelectric materials. The detailed understanding of nanoscale microstructure will usher in a new era of nanostructured thermoelectric materials.

Cooling plays an important role in the human society. It consumes significant amount of electricity and generates 8% of CO2 by human activities. Passive Daytime Radiative Cooling (PDRC) is an electricity-free method for cooling terrestrial entities, which has potential to remarkably reduce electricity consumption. In PDRC, a surface has a solar reflectance of nearly 1 to avoid solar heating, and high emittance close to 1 in the long wavelength infrared (LWIR) transparent window of the atmosphere (wavelength λ = 8- 13 mm) for radiating heat to the cold sky. This allows the surface to passively achieve sub-ambient cooling. In the talk I will present our recent studies on developing porous polymers with excellent PDRC performance. The high density of nano/micropores in polymer leads to efficient light scattering at the interface between the polymer and air, which effectively enhance solar reflectance and thermal emittance. High solar reflectance above 0.96 and high thermal emittance of 0.97 are achieved simultaneously. This talk will emphasize on 1) How to achieve high performance by simple and environmental benign aqueous processing, and 2) Modeling to understand optimal pore distribution in porous radiative cooling polymers.

Thermal conductivity measurements at elevated temperatures are complicated by the contribution of radiative heat transport. Accounting for the radiative contribution is difficult as it often depends on the configuration of the measurement apparatus and the dimensions of the sample. Furthermore, the Radiation Transfer Equation (RTE), which is essentially the Boltzmann transport equation for photons, is difficult to solve even for simple geometries. The laser-induced transient thermal grating (TTG) technique has an advantage in that heat transport takes place entirely within the sample, with no other parts of the apparatus involved, and the simplicity of the sinusoidal spatial temperature profile makes the heat transfer problem amenable to analytical treatment.

In this report, we present a rigorous analysis of the thermal grating decay in a non-scattering absorbing medium in the presence of both radiative transport which is described by the RTE and non-radiative (intrinsic) transport described by the conventional heat diffusion equation. We provide an analytical solution to the coupled equations and quantify the radiative contribution to the effective thermal conductivity. We analyze the limiting cases of diffusive and ballistic radiative transport and show that the radiative contribution is maximized in the intermediate case when the absorption length of thermal radiation is comparable to the heat transfer distance, namely, at a/L ~1/4, where a is the absorption length and L is the TTG period. We present calculations for specific materials focusing on molten salts that have been proposed as thermal energy transport and storage media. We also calculate the upper bound for the radiative contribution, which does not require the knowledge of the absorption coefficient.

We find that while the radiative contribution to the effective thermal conductivity may be significant for L in the millimeter or centimeter range, in a typical transient grating measurement with L ≤ 20 μm it will be in most cases negligible. Thus the TTG technique yields the intrinsic thermal conductivity unaffected by radiative transport. In addition to this practical conclusion, our analysis provides a Fourier-domain Green’s function that can be used to calculate the temperature field in the presence of radiative transport for an arbitrary initial temperature distribution or an arbitrary distribution of heat sources in an unbounded medium.

With industrialization and population growth, the rise in energy consumption has become a critical global issue. A significant portion of the energy in buildings is due to heating, ventilation, and air conditioning (HVAC). The major source of the consumption is the heat transfer to building envelopes. Recently, smart windows have been investigated extensively to reduce energy consumption while maintaining thermal comfort for indoor environment. Smart windows are glazed units coated with special materials to have dynamic control on the transmittance of solar irradiation into buildings under different ambient environments. In our study, a high-performance solar-thermal regulation is achieved through switchable transmittance window. Specifically, the surface turns reflective (lessen the cooling demand) and absorbing (lessen the heating demand) under different modes. The difference of solar transmittance (Δτ) is 0.59, accordingly. The device level testing shows a good ability of large range thermal regulation under the illuminance of 1 sun.

Thermally generated light is a fundamental feature of nature. Its ubiquity makes its control and harnessing both of intrinsic scientific interest and of great importance for energy and heat transfer applications. In this talk, we will discuss our results in nanophotonic approaches to controlling thermal emission, particularly using epsilon-near-zero (ENZ) materials. We will introduce gradient epsilon-near-zero materials that can support leaky electromagnetic modes that couple to free-space waves at fixed angles of incidence over a broad bandwidth. We will then present experimental demonstrations of broadband directional thermal emitters using a range of oxides, designed to operate over long-wave infrared wavelengths. We will furthermore demonstrate implementations of gradient ENZ using doped semiconductors that allow for arbitrary control of spectral bandwidths and directional response as a function of doping gradient, highlighting a new degree of freedom enabled by configuring ENZ response spatially in a layered photonic structure. Finally, we will also illustrate the benefits of directional and spectral control of thermal emission in the context of radiative cooling applications.

One of the many pathways to decreasing the impact of climate change is to increase the efficiency of current fossil fuel based systems. Even the best internal combustion engines suffer from thermal losses where up to 70% of energy is lost as exhaust heat. These losses can be offset to an extent by thermoelectric devices which convert waste heat into useful electrical energy via the Seebeck effect. The efficiency of thermoelectric materials is typically evaluated using the figure of merit (ZT), with some of the best performing materials reaching ZTs of over 2 at low temperatures.[1] However, current state-of-the-art thermoelectric materials contain rare or toxic materials, making them commercially unappealing.

The quest for high ZT materials at low temperatures is complicated by the interdependence of the properties that determine ZT, so maximising the factor is challenging. High Seebeck coefficient and low lattice thermal conductivity favour good thermoelectric performance. Generally, anions of differing mass also help to scatter phonons which decreases the lattice thermal conductivity meaning that mixed anion materials are particularly good candidates for the discovery of novel thermoelectrics.

Y₂Ti₂O₅S₂ has been studied as a potential battery anode, with both experimental and computational studies agreeing on the presence of fast Li-ion diffusion through the material.[2,3] The unique cation deficient Ruddlesden-Poppler structure opening type I Wadsley-Roth intercalation windows in the distorted perovskite-like TiO5S layer was found to govern the diffusion of species. The relative stability to Li, Na, K and Mg intercalation also points to the the dopability of the material, which is important for thermoelectric purposes.

In this study we use hybrid density functional theory to computationally predict the thermoelectric properties of Y₂Ti₂O₅S₂. Analysis of the electronic conductivity and Seebeck coefficient using the state-of-the-art AMSET code, which factors in realistic scattering processes, demonstrates that Y₂Ti₂O₅S₂ shows promise for low temperature thermoelectric applications.


[1] L. D. Zhao, S. H. Lo, Y. Zhang, H. Sun, G. Tan, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Nature, 2014, 508, 373–377.
[2] K. McColl and F. Corà, J. Mater. Chem. A, 2021, 9, 7068–7084.
[3] G. Hyett, O. J. Rutt, Z. A. Gál, S. G. Denis, M. A. Hayward and S. J. Clarke, J. Am. Chem. Soc., 2004, 126, 1980–1991.

Passive devices based on water absorption and evaporation offer a robust, cost-effective, off-grid and sustainable alternative for a wide variety of applications, ranging from personal thermal management to water treatment, filtration and sustainable cooling technologies. These devices do not require high quality energy inputs and, due to the absence of moving mechanical parts, require low maintenance, are cost-effective and usually made from off-the-shelf materials. Moreover, they are optimal for off-grid installations and, in general, promote the sustainable transition to the water-energy nexus. These devices generally exploit porous capillary media to overcome small hydraulic heads and supply the working fluids throughout the system without the need for active mechanical or electrical components. Clearly, maximizing the capillary properties of the porous materials used is crucial to enhance their overall performance: poor capillary properties leads to dry-out and would significantly limit the maximum achievable device size. Thus, neglecting the capillary properties of the porous materials employed would significantly hinder the productivity and scalability of the proposed technology.

This work proposes a paradigm shift in the design of passive devices by replacing the traditional porous materials with a rigid layer, etched with optimized grooves. The concept was tested on aluminum sheets, which were machined by femto-laser to promote wicking on an otherwise flat surface. The technique guarantees an optimal control of the resulting geometrical parameters of the channel, which, through its interplay with the material surface chemistry, allows fine-tuning of the resulting capillary properties. The wicking properties of the obtained sample were compared with those of other engineered porous layers, used as components of passive devices previously reported in the literature. The experimental results were interpreted by a novel theoretical model for mass transfer in open microchannels, which was then used to estimate the effects of the geometry and hydrophilicity of the grooves on their resulting capillary action. Finally, the perspective application of the proposed material in passive desalination devices was investigated both theoretically and experimentally. First, the validated capillary model was coupled with a lumped-parameters heat and mass transfer model for vapor transport in hydrophobic membranes, aiming to evaluate how the capillary properties of porous materials affect the optimal size and maximum productivity of such devices. Then, the salt-rejection capabilities and the durability of the proposed material were evaluated.

The developed theoretical model might serve as a tool to investigate the combined effects of capillary and geometric properties and working conditions on the design constraints of a passive devices component and identify its optimal configuration. Furthermore, the experimental investigation demonstrates the potential of the proposed material for novel components of passive devices for the energy and water sector, paving the way to a new class of rigid materials capable of significant scalability and, thus, their range of use.

Aerogels constitute one of the best thermal insulating materials, typically featuring remarkably low values of thermal conductivity due to their extremely high porosity above 95% comprised mostly of mesopores with an average size of approximately 10 nm, as well as considerably high values of specific surface area (> 500 m²/g), all properties that result in an outstanding option for applications related to energy saving purposes. However, two of their main disadvantages are associated to (a) a relatively high cost at industrial level, and (b) their brittleness and lack of good mechanical properties as a consequence. By leveraging the benefits of ion-exchange sodium silicate solution as silica source, as well as the use of surfactants, we report an ambient pressure drying silica aerogel nanocomposite material that yields superior thermal insulation and mechanical properties. Such combination of components and process parameters provide an exceptionally overall low-cost alternative for the production of aerogel-based, eco-friendly solutions that are suitable for energy efficient applications.

Latent thermal energy systems (LTES) store/discharge energy at nearly constant temperatures using melting/solidification of phase change materials (PCMs). However, energy barriers associated with the nucleation of a solid can cause the liquid state to remain in a metastable state below the equilibrium melting temperature, requiring additional undercooling to initiate solidification. Thus, the existence of undercooling tends to reduce reversibility and overall efficiency of LTES, and to introducing an element of stochastic variability in the solidification process that makes accurate prediction of the state of the system challenging. Among PCMs, metallic PCMs are of particular interest because of their high volumetric energy density and ability to rapidly absorb heat, which is critical in thermal management of electronics, optical, and photovoltaic systems. Gallium is a low melting metal (T_fusion = 30 °C) with a high volumetric energy density (400 MJ/m³) and high thermal conductivity (29.4 W/m°C), making it an ideal candidate for room temperature thermal management applications, where rapid absorption of heat is a key performance indicator. However, the application of Ga as a PCM is severely limited due to the large degree of undercooling, which can exceed 60 °C in small volumes (10 ml) as measured by DSC in aluminum containers at relatively rapid cooling rates (10 °C/min). Here, we exploit epitaxial relationships to identify several cubic carbide and nitride phases that reduce the undercooling of Ga. Among these, HfC and ZrN result in maximum reduction undercooling, with undercooling in the range of <20° C and <10° C for volumes of 10 ml. The observed relationship between lattice mismatch and measured undercooling conform to the relationship predicted by heterogeneous nucleation theory. Thus, this study supports parallel efforts in other classes of materials systems (organic compounds, inorganic salts and salt hydrates), which suggest that epitaxial relationships offer an efficient and valid approach for identify candidate nucleation agents, and predicting the activity of those phases across a broad class of phase change material systems. Among other applications, Ga and Ga based eutectics are of interest for thermal management in electronics packaging and photovoltaics. The carbides and nitrides used in this work are commonly used as coatings and diffusion barriers, thereby suggesting the usage of these carbides and nitrides as thin film barriers to protect the underlying components in addition to serving as nucleation agents for Ga.

The Heusler alloy Fe₂VAl shows great promise as an ecofriendly and low cost replacement to conventional low temperature (250-500 K) thermoelectric materials. Current thermoelectrics use toxic and expensive elements like Te and Sb, whereas Fe₂VAl offers a larger power factor at a lower cost and is non-toxic. Fe₂VAl also allows for mechanical robustness leading to formable thermoelectric modules than can wrap a concave surface rather than typical flat modules that are commercially made for cold temperature applications. The key issue with Fe₂VAl is it’s a relatively large thermal conductivity compared to its semiconductor competitors. The aim of this project is to examine multiple routes to reduce thermal conductivity by enhancing phonon scattering by:

1. Partial lattice substitutions/doping with heavier atomic mass elements (1D phonon scattering sites)
2. Microstructural engineering through anneal/quench treatments (3D phonon scattering)

Arc-melted Fe₂VAl alloy shows a thermal conductivity around 10 W/m*K measured at 300 K. Replacing some of the V with Sn or Ti has been shown to reduce the thermal conductivity by 20% and produce a good n-type or p-type material, respectively. Thermal conductivity, Seebeck coefficient and electrical conductivity have been measured for the dopant additions of 1-5 at.% for Sn, Ti, W, Ta, Ge and Si. Ge shows the most notable performance improvement with both a reduction in thermal conductivity and an enhancement in the Seebeck coefficient. However, Sn additions produce very similar performance improvements with the added benefit of better ductility.

Using water-quench treatments, we have also seen a significant drop in thermal conductivity in Ge-doped Fe₂VAl. X-ray diffraction (XRD) showed that annealing at 1273 K for 48 h produces an ordered L₂₁ phase. Differential scanning calorimetry + XRD have shown that a stable B2-phase is achievable if the material is held above 1393 K for 2 h. Once the material achieves the B2 phase, where the V and Al atoms are random in the lattice, the material is water-quenched to preserve the B2 phase as the bulk phase. Dark-field transmission electron microscopy using superlattice reflections showed the presence of numerous thermal anti-phase boundaries (APBs). The room temperature thermal conductivity of Fe₂VAl_0.9Ge_0.1 annealed at 1273 K for 48 h is 9.3 W/m*K whereas the thermal conductivity of the same alloy annealed at 1393 K for 2 h and water-quenched is 4.5 W/m*K mainly due the presence of APBs in the material. Various anneals have been used to produce changes in the density of the APB domains and the resulting effect on both the thermal and electrical conductivity has been determined. Overall, through a combination of partial lattice doping and microstructural phase engineering, the zT of the Fe₂VAl Heusler alloy has been increased from 0.07 to 0.18 at 300 K.

Thermoelectric energy conversion has a lot of attention because of its simple and convenient energy conversion of heat to electricity. In this work, thermoelectric nanocomposite with coating layer at the grain boundary was synthesized using chemical synthesis. Coated grain nanostructure has significant advantages for thermoelectricity because of energy filtering and effective coherent phonon scattering by coated grain boundary. Our newly developed phonon scattering model has proved phonon scattering by coated grain boundary is caused by interference of wave-like phonons. The coherent phonon scattering in coated grain boundaries is more powerful than particle-like phonon scattering by other kinds of impurities. The chemical synthesis strategy for coating process is conducted by cation exchange reaction of SnTe powder considering HSAB theory, thermodynamic factors. The thermoelectric figure of merit(zT) was 1.90 at 923 K for the CdTe/SnTe and 1.68 at 823 K for the eco-friendly CuInTe₂/SnTe nanocomposite.

Hall mobility measurements of samples with high carrier density is challenging due to the low Hall voltage. Although an increase of the drive current enhances the Hall voltage, over a certain level, Joule heating at the contacts results in unreliable measurements, due to changes in the resistivity around the contact and/or the generation of thermoelectric voltage at the contacts. Localised heating may even damage the sample.
Therefore, suppressing measurement noise is essential to obtain valid Hall mobility results from low resistivity samples, which can be achieved using modulation methods, varying periodically either the magnetic field or the drive current during the Hall measurement.
The Parallel dipole line (PDL) AC Hall measurement technique is a very practical way to realize pure harmonic magnetic field modulation in a very compact design using rotating permanent magnets . The frequency selective “lock-in” evaluation of the resulting sinusoidal Hall voltages significantly reduces measurement noise and makes it possible to determine the Hall-mobility in highly conductive materials, up to a certain limit.
However, in metals or in thermoelectric materials featuring very high free carrier density (over 1 E21 cm^-3) the signal-to-noise ratio is still not sufficient to get a valid result.
In this work, we combine magnetic field modulation with alternating polarity drive current, i.e. using a simultaneous two-parameter modulation at very different frequencies to gain further noise reduction.
The reliability of the bi-modulated PDL Hall method was investigated by comparing the results with the standard AC field PDL Hall measurement from several moderately conductive materials. Hall mobility results from Si and SiAs samples using the two methods were within 1% and were in good agreement to typical values in the literature.
Then, Sn doped NbCoSb thermoelectric samples were tested ,featuring very high carrier density, above the specification of the standard PDL Hall measurement.
While the standard PDL Hall measurement provided inconsistent and highly fluctuating results depending on the drive current as well, the bi-modulated method resulted in very stable Hall voltage reading and repeatable mobility values. The Hall mobility values measured from the highly conductive samples (1-3 E21 cm^-3 carrier density) are in the 2-11 cm²/V^-1s^-1 range in good agreement with results of material simulations.
Analysing the frequency dependence of the AC drive current using a series of resistors defined the applicability range of the drive current modulation which turned to be limited by the “RC” time constant of the sample resistance (R) and the cable capacitances (C). Since higher resistance samples can be tested reliably without alternating the drive current, these tests confirmed the applicability of bi-modulated PDL Hall method for the high carrier density range in interest.
In conclusion, the combination of the AC magnetic field and alternating drive current in PDL Hall systems provides an astonishing Hall voltage sensitivity which results in accurate Hall mobility results even in thermoelectric samples featuring carrier concentration above 1 E21 cm^-3.

The thermal properties of semiconductors following exposure to ion irradiation are of great interest as pertaining to the cooling of electronic devices, however the spatially-varying structural and compositional changes inherent to ion bombardment often make measurement difficult. We present an advance to the Time Domain Thermoreflectance (TDTR) technique, and a study in which we demonstrate the measurement of Silicon bombarded with Krypton ions, throughout the stopping region. We also use Transmission Electron Microscopy (TEM) to validate our findings and further explore the structural effects of bombardment.
TDTR is a pump-probe technique that measures the change in thermoreflectance on the surface of a sample as a function of pump-probe delay time from picoseconds to nanoseconds. This change in thermoreflectance is related to both the temporal temperature decay from impulse heating driven by the sub-picosecond pump pulse and the frequency-dependent temperature rise induced from the modulated pump pulse train. The thermal decay response is then fitted to a multi-layer model in order to extract the thermal properties of interest. Ordinarily, TDTR is well-suited to the measure thermal properties even beneath the surface of the sample, which in our case, allows for the measure of properties within the ion-stopping region. We modify the traditional analysis by discretizing a layer and instead fitting for a function of thermal conductivity. This allows the measure of the depth-dependent thermal conductivity due to ion bombardment, and we find a thermal conductivity profile with an excellent match to the predicted and observed damage profiles from SRIM and TEM. This is beneficial for thermal characterization of materials, and as a tool for profiling bombardment-induced damage.
Our technique, in combination with TEM, is able to show a measured thermal conductivity of as low as 2.34 W/m/K for regions of defected crystalline silicon containing 5-15nm amorphous pockets. We explain this measured thermal conductivity as being a product of both porosity / nanocomposite interface scattering effects, and greatly reduced thermal conductivity of the crystalline regions as a product of purely structural disorder. We also note that the measure of the thermal conductivity of the defected crystalline regions between the amorphous pockets is among the lowest ever measured.

Thermal management in Li-ion batteries is critical for their safety, reliability, and performance. I will discuss three aspects of our current work on the physical properties of solid electrolytes (SEs) for solid-state batteries. 1) Measurements of the thermal conductivity of air and moisture sensitive halide and sulfide SEs are enabled by depositing the Al transducer film for time-domain thermoreflectance (TDTR) measurements with a compact sputter deposition chamber mounted inside a glove box. Thermal conductivities of halide and sulfide SEs are in the range 0.45–0.70 W m⁻¹ K⁻¹, a factor of 2-3 smaller than the thermal conductivities of oxide SEs. The thermal conductivity has a glass-like temperature dependence that presumably arises from the strong atomic scale disorder that is necessary for high ionic conductivities. 2) Characterization of the Li composition of solid-electrolytes and electrodes is a ubiquitous challenge in battery research because analytical techniques based on core-level excitations are not applicable to Li. We recently implemented nuclear reaction analysis using a 1.8 MeV proton beam and spectrometry of 4-8 MeV alpha particles that are generated by the ⁷Li(p,α)⁴He reaction. This ion-beam analysis technique achieves a Li composition accuracy of a few percent across a depth of several microns with a depth resolution of 100 nm. 3) While knowledge of boundary conditions and thermal properties of materials is sufficient to calculate the temperature field within a battery, direct measurements of temperature fields in three-dimensions would be valuable in situations where the heat sources and thermal properties are not well known. As a first step in this direction, we have characterized the temperature dependence of the nuclear magnetic resonance linewidths of ⁷Li in common solid electrolytes. The linewidth has a strong temperature dependence when the hopping rate of the ⁷Li ion overlaps with the natural linewidth created by dipolar coupling and may provide an approach for 3D thermometry by magnetic resonance imaging.

The fundamental interactions among photons, electrons and phonons at interfaces dictate heat transport in a wide array of materials and interfaces used in energy conversion and storage technologies. Characterization of these electron and phonon thermal properties are commonly studied with pump-probe methods. In standard implementations of these techniques, visible light laser sources are used to excite and probe electronic transitions on the surface of samples, and thus relate the temporal decay of these electronic states to temperature and resulting material properties. At these photon energies, the nature of the probe response is electronic, in that electronic interband transitions dominate the optical response.

In this presentation, we demonstrate the use of sub-picosecond infrared laser pulses as a means of more direct resolving the thermal response of nonequilibrium carriers. The low energy nature of these laser pulses with respect to electronic interband transition energies in metals and semiconductors allows for direct probing of free electron responses in metals and semiconductors and resonating with coupled nonequilibrium polaritonic carriers in non-metals. We will show the ability to: i) probe the purely free electron response in Au and recover the predicted thermal electron-phonon coupling factor; ii) probe the time domain relaxation of phonons in semiconductors based on optical interactions with optical phonon modes; and iii) monitor the ultrafast thermal decay of coupled plasmon-polariton modes in doped CdO and phonon-poliarton modes in hexagonal boron nitride.

Tomko, J.A., Runnerstrom, E.L., Wang, Y.-S., Chu, W., Nolen, J.R., Olson, D.H., Kelley, K.P., Cleri, A., Nordlander, J., Caldwell, J.D., Prezhdo, O.V., Maria, J.-P., Hopkins, P.E., “Long-lived modulation of plasmonic absorption by ballistic thermal injection,” Nature Nanotechnology 16, 47-51 (2021).

Tomko, J.A., Kumar, S., Sundararaman, R., Hopkins, P.E., “Temperature dependenct electron-phonon coupling of Au resolved via lattice dynamics measured with sub-picosecond infrared pulses,” Journal of Applied Physics 118, 193104 (2021).

The increasing prevalence and efficiency of electric vehicles have continuously driven the need for high-energy-density lithium-ion battery packs. Within these battery packs, the exothermic reactions of the cell cause an increase in temperature and thus require efficient thermal management systems to prevent thermal runaway events and catastrophic failure. Current solutions include polymer PCMs, which absorb heat as they melt. Their low thermal conductivity limits the heat transfer rate through the PCM, making them inefficient for thermal runaway scenarios. We tackle this problem by incorporating aligned carbon fibers into a polymer PCM. Our material shows a five times increase in through-plane thermal conductivity and more efficient thermal energy absorption. We demonstrate this through thermal characterization, computational modeling, and simulated thermal runaway on a benchtop scale testbed.

The ability to predict, understand, and control thermal transport in materials and at interfaces remains a critical challenge and goal of nanoscale thermal transport research. In nanostructures where phonons are the primary thermal energy carriers, interfaces between dissimilar materials represent the dominant thermal resistance. An increased understanding of phonon-mediated transport across interfaces is critically needed, so that nanostructured devices can be more effectively designed and implemented.

Over the past several decades, the nanoscale heat transfer community has identified and investigated many factors that influence phonon transport across solid-solid interfaces. Experimental data, simulation results and theoretical analysis have all elucidated the collective effects of vibrational spectra, harmonic and anharmonic processes, defects, and mixing on the ability to transport heat across a solid-solid interface. As the community continues to study these phenomena many challenges arise. In the regime of simulations, it is difficult to develop a computational model that captures all the properties of a physical material and the interactions between materials. From the standpoint of the experimentalist, there is no perfect system or set of systems that allows factors influencing thermal transport to be decoupled. In all areas of thermal transport research, the challenge is either decoupling or understanding the interplay of complex effects that manifest differently in various materials and experimental/computational environments.

At the Nanoscale Energy Transport Laboratory at the University of Virginia, our research efforts combine computational and experimental techniques to model, measure, and predict phonon dynamics, and the resulting thermal properties, for a wide range of technologically relevant systems. We have approached this problem experimentally, measuring nanoscale systems with time-domain thermoreflectance (TDTR), and computationally, tracking atomic thermal motion in non-equilibrium molecular dynamics (NEMD) simulations, and investigating quantum mechanical effects using the non-equilibrium Green’s functions (NEGF) approach This work is complicated by the wide spectra of phonon mean free paths, ranging from a single atomic spacing to the size of the material system.

Here we review approaches for measuring, modeling and manipulating thermal boundary conductance across solid-solid interfaces in hopes of developing avenues to control the heat transport at the atomistic length scale. For example, we have demonstrated that insertion of a layer with intermediatevibrational properties of the two contacts joining at an interface can enhance the interfacial thermal conduction under certain conditions. By utilizing an exponentially mass-graded interface, even further increases in enhancement are possible, attributed to the strength of anharmonicity in addition to matching of vibrational spectra. The ultimate goal is to drive the discussion of thermal management from a design afterthought down to intentional design at the nanoscale where the heat is generated, completing the journey from TIM to transistor.

Thermal regulators that enable controlling temperature on demand is a key component in the development of solid-state thermal circuits. A highly efficient thermal regulator should operate near or above room temperature and offer a quick and large on/off switching ratio. Switching the thermal conductivity of a material can be achieved upon repeatable transformations in the electronic or phononic structure. However, reliable high-performance thermal regulators are still missing from the literature. A promising class of material candidates for dynamic thermal conductivity switching that have recently received considerable attention due to their fast, repeatable, and well-integrated trigger mechanism are ferroelectric perovskites such as lead zirconate titanate PbZr_1−xTi_xO₃ (PZT_x) and lead titanate PbTiO₃(PTO). Despite the large focus on investigating the thermal conductivity of PZT and PTO with varying conditions, their antiferroelectric end-member, lead zirconate PbZrO₃(PZO), has received much less attention. In this material, a sufficient electric field will transition the antiferroelectric phase to a ferroelectric phase where there is a volume expansion, reduction in the unit cell size from 8 formula units to 6, and the possibility of altering the populations of ferroelastic domains. Furthermore, there are ferroelastic domains within antiferroelectric PZO and they too may impact phonon scattering rates. In addition to AFE to FE phase transition, PZO undergoes another phase transition upon heating where the lattice structure goes from 8 formula units to 1, and is expected to reduce the phonon scattering rate and lead to higher thermal conductivities. In this study, we experimentally demonstrate how thermal conductivity of antiferroelectric PbZrO₃ can be bidirectionally switched by -10% and +15% upon applying electric field and thermal excitation, respectively. Our approach takes advantage of two separate phase transformation mechanisms in PbZrO₃ that influence the phonon scattering rate in different manners. This is the first experimental realization of a fast (<1 second) and high-performance thermal regulator with a net switching ratio of 25% from 1.2 to 1.5 W/m/K.

Thermoelectrics transform heat into electricity and vice versa, making them valuable tools for the generation and recycling of clean, renewable energy. While progress has been made increasing the efficiency of thermoelectrics, measured by the dimensionless figure of merit, ZT, the best systems often contain rare or toxic elements, such as PbTe and Bi₂Te₃, which reduces their sustainability. Furthermore, ZTs of n-type thermoelectrics have lagged behind those of their p-type counterparts. Conductive oxides have shown promise as potential thermoelectrics¹ due to their strong electrical conductivity and high temperature stability, particularly ZnO, which has reached a ZT of 0.65 at 1000 K.² As they are often hindered by their high thermal conductivity, nanostructuring has been extensively tested for overcoming this problem,³ but limits the electrical conductivity as well as the thermal conductivity, reducing the potential ZT increases.

Here we use hybrid density functional theory (DFT) to screen the anion mutated ZnO derivatives Zn₂NX (X = Cl, Br, I)⁴ for thermoelectric properties. We analyse their thermal and electronic transport beyond the constant relaxation time approximation (CRT) using Phono3py⁵ and AMSET⁶ respectively. We show that anion mutation is a highly effective means to reduce the lattice thermal conductivity, and analyse the combined effect of the various other changes this makes to the materials, comparing to each other, ZnO and thermoelectric materials in general. We then use these insights to assess the potential of the anion mutation method.

[1] Spooner, K. B. et al., Submitted. 2021.
[2] Ohtaki, M. et al., J. Electron. Mater. 2009, 38, 1234.
[3] Spooner, K. B. et al., J. Mater. Chem A 2020, 8, 11948.
[4] Liu, X. et al., J. Solid State Chem. 2013, 203, 31.
[5] Togo, A., Tanaka, I., Phys. Rev. B 2015, 91, 094306.
[6] Ganose, A. M. et al., Nat. Commun. 2021, 12, 2222.

Efficient harnessing and management of thermal energy attracts widespread interest in the context of water-energy nexus. Recently, hydrogels have shown great promises for water-based heat and mass transfer processes such as thermally driven water harvesting, desalination and thermal management of electronics due to the high capacity of water adsorption by hydrogel (could store water up to > 100 times of dry gel mass) and high osmotic pressure for salt rejection. Despite extensive studies on the high adsorption capacity, much less efforts have been paid to improve the kinetics of the heat and mass transfer processes, which are important performance metrics such as the water production yield in water harvesting from air, water flux in desalination and thermal conductance of electronic cooling. In this talk, we present the structural engineering of hydrogels for significantly enhanced kinetics. The general strategy is based on thin polyelectrolyte hydrogel coating on the surface of a metal scaffold with large effective surface area. Due to small thickness of the hydrogel coating and high thermal conductivity of the metallic scaffold, the heat and mass transfer rate are significantly improved, leading to high adsorption/desorption kinetics. As an example, porous aluminum foam was coated with ~ 100 μm thick hydrogel to provide 10-fold increase in the surface area, leading to a short adsorption time constant of < 15 mins and short desorption time constant of 10 mins. In an optimized adsorption/desorption cycle, we can obtain water yield > 20 kg_water kg_gel^-1 day^-1 and > 10 kg_water m^-2 day^-1 by using the thermo-responsive polyelectrolyte hydrogel. We also apply an aluminum fin heat sink coated with thin hydrogel for passive evaporative cooling of electronics. Taking advantage of the high latent heat of water during evaporation, small thermal resistance across the thin gel and high adsorption capacity of gel, we achieve high thermal conductance ~ 200 W m^-2 K^-1, high cooling power of > 1.5 kW m^-2 and long operational time of ~ 6 hours in the peak-hours. The gel can also be fully regenerated during the off-peak hours at night. Our studies show the potential of structural engineering of hydrogel-based materials to enhance heat and mass transfer kinetics.

PbTe, one of the most highly performing thermoelectric materials, and similar GeTe- and SnTe- based alloys have been used in the power sources for NASA space missions for several decades. The exceptional thermoelectric performance of PbTe and its alloys is partially attributed to the high valley degeneracy of its valence band edges, which leads to an enhnaced thermoelectric quality factor, and hence, maximum attainable efficiency. The high valley degeneracy of the valence bands is due to the convergence of multiple valence band maxima at low-symmetry points in the first Brillouin zone. In fact, prior studies using first-principles calculations suggest that as the valence band edges become more highly converged, the valence band Fermi surface undergoes a topological transition from a valence band edge described by a 3D density of states to a thread-like Fermi surface described by a 2D density of states. Transport modeling with the Boltzmann transport equation (BTE) predicts significant gains in thermoelectric performance upon this transition to lower-dimensional bands in bulk 3D materials. In the work presented here, we study the chemical origins of this favorable valence band behavior in PbTe by considering both numerical and analytical solutions to a tight-binding model with nearest and next-nearest neighbors. The tight-binding method offers the advantages of extremely short computational times and of employing a chemically intuitive basis comprised of recognizable atomic orbitals. We develop simple atomic orbital pictures and molecular orbital energy level diagrams for high symmetry k-points in the Brillouin zone to describe a tight-binding wavefunction corresponding to the valence band edge in PbTe. Furthermore, we find that in a theoretical tight-binding electronic structure of a model IV-VI rock salt compound where four different valence band extrema are effectively converged, a 1D Fermi surface can be obtained. Using the BTE, we predict further gains in thermoelectric performance from this 1D Fermi surface topology relative to the 2D and 3D cases. We use the results of this analysis to predict and to understand potential alloying strategies to further optimize the electronic structure of PbTe and its analogues. This work can be directly extended to the other IV-VI rock salt compounds (IV = Ge, Sn, Pb; VI = S, Se, Te), and the orbital chemistry framework applied here can be more generally applied to new materials systems in order to guide the understanding and discovery of materials with optimal electronic structures for thermoelectric performance.

The on-demand exothermic crystallization of stably supercooled phase change materials, which have long been used in the context of hand warmers and heat pads, shows promise as a component of engineered energy transduction cascades. These processes amplify a small nucleation trigger to release large quantities of thermal energy that can be used to activate downstream processes that respond to changes in temperature.

Our group has developed a strategy for controlling the spatiotemporal evolution of crystal growth and heat release in metastable phase change materials. In this work, we synthesize polymer networks from acrylamide monomers in supersaturated aqueous solutions of sodium acetate using ultraviolet light and a photoinitiator. Photomasks are used to create metastable hydrogels with patterned domains that remain unpolymerized. Upon nucleation with seed crystals of sodium acetate trihydrate, long needle-like crystals grow through unpolymerized regions; the growth front in these areas is up to seven times faster than in polymerized domains, where a polycrystalline composite forms. Crystal growth is accompanied by a wavelike temperature profile with a peak at the solid/liquid interface. In one set of conditions, the peak temperature during fast crystallization through unpolymerized solutions was 45 °C, whereas the slower crystal growth process through a polymer network peaked at 30 °C. These profiles are well-described by a one-dimensional analytical model; the discrepancy between peak temperatures can be fully explained by the velocity difference.

We demonstrate the utility of this technique by using it to pattern thermoresponsive processes, which can be induced to proceed in a wavelike fashion following the peak temperatures of the crystal growth fronts. We use the temperature difference between polymerized and unpolymerized regions to selectively activate processes that proceed at threshold temperatures above 30 °C, such as wax melting or hydrogel contraction, within predefined areas to form complex patterns such as text.

The ability to dynamically control heat flow in materials (i.e. thermal conductivity switches) would enable the development of more efficient energy systems, ranging from controlling heat flow in computer chips to allowing phase change materials to be more efficient in buildings. However, developing polymer systems that exhibit dynamic thermal conductivity changes remains a challenge due to the difficulty in using external stimulus to significantly alter a polymer’s ability to conduct heat. It has been previously shown that a mixture of polymers that interact via hydrogen bonding can form a network with a drastic increase in thermal conductivity compared to their pure components. One potential mechanism is to reversibly switch the strength of interactions between chains by using dynamic covalent crosslinks. It has been hypothesized that Diels-Alder reactions are one such way to dynamically control crosslinking and thereby the utilization of the Diels-Alder reaction could result in dynamic thermal switching. In this work, a series of precursor uncrosslinked statistical copolymers with varying mole fractions of furan functional groups were synthesized via RAFT polymerization. The precursor polymers were crosslinked using a reversible Diels-Alder cycloaddition to form dynamic covalent networks of varying crosslink densities. Two separate crosslinkers were used to investigate how the molecular weight and length of the crosslinker influence the thermal and mechanical properties of the network. It was found that the thermal conductivity was largely unaffected by the extent of crosslinking and the structure of the crosslinker counter to previous motivating studies. Additionally, the elastic modulus was measured and gives insight into why thermal conductivity did not significantly change. For this system, the Diels-Alder reaction was also found to be slow in the solid-state, possibly further limiting its utility as a thermal conductivity switch due to slow switching speeds.

Non-parabolic electronic structures oftentimes reinforce the thermoelectric performance of a material, due to a larger effective valley degeneracy. While materials with inverted bands sometimes exhibit high valley degeneracy, for example rock-salt SnSe, the beneficial feature is not guaranteed for all materials with such unique inverted electronic structure topology. We survey the electronic structures of rock-salt IV-VI compounds, which demonstrates that band inversion is a necessary, but not sufficient, condition for high valley degeneracy in inverted band compounds. Using a combination of k●p theory, tight binding theory, and Density Functional Theory calculations, we show that the degree to which the bands are inverted is a chemically-tunable parameter that is partly responsible for inducing multiple carrier pockets in IV-VI compounds.

Recent improvements in the efficiency of heat-to-electricity energy conversion in lead chalcogenide thermoelectrics involve reducing the thermal conductivity by incorporating large amounts of internal strain. The extent to which typical lead chalcogenide processing techniques (such as doping, ball milling, and densification) increase internal strain and dislocation density must be quantified to improve materials design. In this study, neutron powder diffraction is leveraged to evaluate the internal strain introduced by ball milling in doped and undoped powders. Doping with Na and/or Eu increases internal strain beyond ball milling alone with the greatest increase from combining the two dopants. Strain recovery occurs in each powder above 400 K but can be suppressed by co-doping – indicating a strong dopant-dislocation interaction in this system. Therefore, high temperature processing of PbTe powders should be avoided if high internal strain is desired. Low temperature densification and/or rapid pressing techniques may be key to maintaining internal strain in the final pressed pellet. This work provides key guidance for defect engineering to maximize internal strain and thermoelectric performance in PbTe thermoelectrics.

Over the last five years, atmospheric water harvesting has gained significant interest as a promising alternative to address the growing water scarcity problems across the globe. In this work, we synthesized a hygroscopic, organic-inorganic hybrid material and combined it with an efficient thermodynamics cycle to convert atmospheric water vapor to liquid water. We demonstrate the potential of the novel hygroscopic materials under extreme (arid) climate conditions: ambient temperature higher than 30°C and dew point lower than 5°C. The hybrid hydrogel material is obtained through the crosslinking reaction of polyelectrolytes with counter ions. We used sodium alginate and calcium alginate as the precursors to form the hygroscopic material into any desired geometries (spherical particles, rectangular structure, etc). The water vapor capture and subsequent release and conversion to liquid water is achieved through a co-located heat/mass transfer. During the water uptake step (adsorption), the stream of humid air flows through the sorbent, and the water vapor is stored in the hydrogel through a combination of endothermic and exothermic reactions. During the regeneration step, the air flow running through the device is inverted to allow a stream of hot air to contact the hydrogel. The hot air combined with high relative humidity allows for high dew points operations, such that spontaneous water condensation can take place at ambient temperatures. In addition, our experiments show that our setup can operate under dry climate conditions (ambient dew points lower than 5°C), and only use solar thermal energy to drive the regeneration cycle (water release). Finally, we demonstrate that the whole process can be scaled-up to enable the production of liter-scale freshwater on a day-night cycle.

Thermally localized solar evaporation provides a sustainable solution to address the water-energy nexus, which is promising for the applications in vapor generation, seawater desalination, wastewater treatment, and medical sterilization. However, salt accumulation has been identified as one of main practical challenges, which induces undesired fouling and reduces device lifetime. Here, we develop a novel confined water layer structure enabling a simultaneously highly efficient and salt rejecting solar evaporation. With high-fidelity modeling and experimental characterization of salt transport, we optimize the fluidic flow in the confined water layer. With the optimized flow field, our design is capable of salt transport without sacrificing energy efficiency. In addition, we further applied our design to the contactless solar evaporator and report a record-high efficiency. The fundamental understanding of salt transport shown in this work paves a new avenue toward the high-performance and reliable solar evaporation with low-cost and high material flexibility.

Water adsorption has attracted particular attention owing to its broad applications in atmospheric water harvesting, energy storage, and thermal management. Hygroscopic hydrogel has been recognized as a promising material due to the large total uptake, fast kinetics, and low material cost. Although high-performance water adsorption with hydrogel has been experimentally demonstrated by several recent works, the fundamental understanding of the adsorption kinetics remains elusive because of the complex transport mechanics inside the hydrogel. In this work, we develop a two-concentration model to predict the adsorption kinetics of hydrogel, where two coupled diffusion equations are introduced to describe the vapor and liquid transport, respectively. Taking the CaCl₂ embedded hydrogel as an example, we calculated the adsorption kinetics at various relative humidity and showed good agreement with experiments. More importantly, we show that liquid transport, driven by the gradient of water chemical potential, plays a significant role in the rapid kinetics of hydrogel. Liquid transport induces the redistribution of water in the polymer matrix the therefore leads to the volumetric expansion of the hydrogel. With a parametric study, we identified the key parameters responsible to the kinetics of adsorption, where an optimal initial porosity is predicted. Furthermore, the kinetics can be largely engineered by changing the elastic modulus of the hydrogel. This work provides the first modeling framework for the adsorption kinetics of hydrogel, which can serve as a design guideline in applications of the atmospheric water harvesting, energy conversion, and thermal management.

The levitation of water drops can be triggered by augmenting the temperature of the solid on which they sit. Above a threshold (typically 200°C on regular solids), water detaches from its substrate, owing to the presence of a continuous vapor film, which makes the liquid extremely mobile. We discuss how to decrease and increase the Leidenfrost point (LFP), which can be exploited in different applications: on the one hand, a smaller LFP allows us to make water mobile with a much smaller quantity of energy; on the other hand, a larger LFP makes it possible to maintain heat exchange at high temperature, a valuable property for heat pipes.

The ever increasing societal demand for clean energy, coupled with the impending global climate emergency has driven an influx of research into renewable energy. To attempt to mitigate the most catastrophic effects of climate change, it has been proposed that greenhouse gas emissions must be reduced by 20%, alongside a significant increase in the energy efficiency of existing processes. Wasted thermal energy is generated by processes across the energy sector, transportation, and the human body itself. Thermoelectric devices, can be used to convert thermal energy into a clean source of electricity, generating no emissions and without the need for moving parts. Importantly, these materials can provide a route to significantly increase the overall energy efficiency of existing process across a range of sectors via waste-heat harvesting.^[1]

The effectiveness of a thermoelectric material is measured using the dimensionless figure of merit ZT, with the world record set at 2.6 for single-crystal SnSe along the through-plane direction.^[2] Realising reasonable conversion efficiencies generally requires high electrical conductivity and low thermal conductivity, with the maximum ZT of a material often limited by the strong correlation between these properties. The parameters that contribute to the overall ZT of a material are also highly dependent on the concentration of charge carriers in the system, and therefore it is important to gain a comprehensive understanding of the defect chemistry of thermoelectric materials to maximise efficiency.

In this study, we predict the maximum theoretical ZT of LaZnOP, a mixed-anion compound which has promising thermoelectric properties, using density functional theory. The performance of oxide thermoelectrics generally lagging behind the efficiencies achieved by chalcogenide-based materials such as Bi₂Te₃, with poor carrier mobilities and high intrinsic thermal conductivities hampering progress.^[3] Mixed-anion materials have been looked to in an attempt to tackle these issues, with the more complex crystal structures reducing the lattice component of thermal conductivity. Many high-mobility mixed-anion compounds have also been reported on recently. The charge transport properties will be calculated using state-of-the-art methods based on hybrid-density functional theory, employing the AMSET approach to calculate electron-scattering rates going beyond the constant relaxation time approximation.^[4]

We show that LaZnOP has the potential to be an extremely promising earth-abundant thermoelectric material, with a ZT exceeding that of most oxide materials, provided it is dopable to 10¹⁹-10²⁰ carriers cm^-3.^[5] Electronic-structure and band-alignment calculations suggest p-type dopability, according to the doping limit rules, and predict high hole mobilities. A full study of the defect chemistry of LaZnOP was also performed using hybrid-DFT, to assess the doping potential of LaZnOP, and guide future synthesis.


1. L. E. Bell, Cooling, heating, generating power, and recovering waste heat with thermoelectric systems. Science 321, 1457–1461 (2008)
2. L-D, Zhao, S-H. Lo, Y. Zhang, H. Sun, G. Tan, C. Uher, C. Wolverton, V. P. Dravid and M. G. Kanatzidis, Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals. Nature 508, 373-377 (2014)
3. G. Tan, L-D. Zhao, and M. G. Kanatzidis, Rationally designing high-performance bulk thermoelectric materials, Chemical Reviews 116, 12123–12149 (2016)
4. Ganose, A. M.; Park, J.; Faghaninia, A.; Woods-Robinson, R.; Persson, K. A.; Jain, A. Efficient Calculation of Carrier Scattering Rates from First Principles. arXiv:2008.09734 [cond-mat, physics:physics] (2020)
5. Einhorn, M.; Williamson, B. A. D.; Scanlon, D. O. Computational prediction of the thermoelectric performance of LaZnOPn (Pn = P, As) J. Mater. Chem. A, 8, 7914–7924 (2020)

Flexible thermoelectric generators have the ability to harvest thermal energy from a variety of heat sources (e.g. human skin) and directly convert it into electricity to work as a sustainable power source for sensors and wearable devices. Due to the limited thermoelectric (TE) performance of conducting polymers and rigidity of inorganic materials (e.g., Bi₂Te₃ and Sb₂Te₃), it remains a challenge to prepare low-cost, highly flexible, and high-performance TE materials. An efficient thermoelectric generator is composed of materials with a high thermoelectric power factor S²σ (S is the Seebeck coefficient and σ is the electrical conductivity), and low thermal conductivity κ. Herein, we fabricated n-type silver selenide films on flexible porous nylon membrane using vacuum-assisted filtration, followed by photonic flash (i.e., intense pulsed light) sintering. In contrast to conventional thermal sintering, rapid and versatile photonic sintering using intense pulsed light offers great advantages as it can sinter the silver selenide films at elevated temperatures without overheating/damaging the underneath polymeric substrate. Moreover, silver selenide shows an excellent absorptivity throughout the ultraviolet-vis (UV-vis) spectra with perfect overlap with the wavelength range of the emitted light during flash sintering. Silver selenide with exceptional thermoelectric power factor of ~ 2200 μW/mK² can be achieved within an extremely short sintering period of 2 ms. The flexible silver selenide materials with high thermoelectric performance near ambient temperatures can be considered an ideal candidate for low-grade waste-heat-recovery applications. This work introduced a promising pathway to realize high-performance and flexible TE devices with a broad range of applications in harvesting and managing thermal energy.

(p × n)-Type transverse thermoelectrics were proposed as a potential thermoelectric material for Peltier cooling at cryogenic temperature and integrated thermoelectric devices. In transverse thermoelectric materials, highly anisotropic ambipolar conduction gives rise to an off-diagonal Seebeck coefficient, generating a voltage drop perpendicular to an applied temperature gradient [1]. Some anisotropic thermoelectric materials are found to have an ambipolar Seebeck effect, with a negative Seebeck coefficient along one direction and positive orthogonal Seebeck, hence the name (p × n)-type. However, standard experiments only measured Seebeck coefficients along specific crystal axes, which is, in principle, insufficient to map the full Seebeck tensor of a low symmetry crystal. And due to the lack of Onsager symmetry, the Seebeck tensor, unlike transport coefficients like electrical conductivity, can have up to 9 independent components for the lowest symmetry case (triclinic) [2]. It is also advantageous to use a single sample to measure all Seebeck components since multiple samples might have slightly different doping [3]. Here, we report results for an all-in-one measurement set-up for measuring the full Seebeck tensor on one sample with custom-made apparatus. Experiments on pre-characterized isotropic material confirm this apparatus can accurately determine the 9 Seebeck components. Anisotropic van der Pauw measurement can also be integrated into the apparatus to measure the full in-plane anisotropic resistivity tensor on the same sample, hence giving the anisotropic in-plane power factor [4]. Results of Seebeck tensor measurements in known anisotropic materials such as bismuth single crystals will also be presented.
The apparatus is an insulated aluminum sandwich-structure with contacts and thermocouples built in. First, two aluminum plates (25×25×4mm³) are machined with aligned thru-holes and tapped holes for the respective plates. Second, grooves for contacts and thermocouples are laser cut on the inner surface of machined aluminum blocks with accurate positioning. Then, the aluminum plates are anodized to create an electrically insulated surface. After that, indium contacts and thermocouples are placed into the grooves. Finally, samples (8×8×2mm³) are sandwiched between the two aluminum plates, with thermal epoxy/paste to ensure good thermal contact. A temperature gradient along 3 orthogonal directions can be achieved by placing the sandwiched blocks between the heater and heat sink in different orientations. The temperature gradients are measured by the built-in thermocouples while voltages along different directions are measured by the electrical contacts, thus giving the full Seebeck tensor. We used pre-characterized isotropic thermoelectric material PbTe as a reference sample to test the accuracy of the method. The diagonal Seebeck coefficients agree with results from the standard measurement method within 10%, and the error introduced in the 6 off-diagonal terms are smaller than 5% of the diagonal values.



References
1) Zhou, C. et al, Phys. Rev. Lett., 110, 227701 (2013); Tang, Y. et al, J. Electron. Mater., 44, 2095–2104 (2015); Shao, Q. et al, MRS Advances, 4, 491–497(2019).
2) R. E. Newnham, Properties of Materials: Anisotropy, Symmetry, Structure, (OUP Oxford, United Kingdom, 2005), pp. 234-240.
3) J.-J.Gu et al, Phys. Rev. B, 71.11, 113201 (2005); K.A. Borup et al, Energy Environ. Sci., 8, 423(2015).
4) Peng, L. et al, Phys. Rev. Lett., 120.8, 086801(2018).

We synthesized several derivatives of lead ruthenate pyrochlore (Pb_(2+x)Ru_(2-x)O_(6.5±z) ) by varying the Pb/Ru ratio, thereby introducing changes of lattice site occupancy, vacancies and valences. Reducing the number of Pb²⁺ ions create more vacancies in the A₂O′ sublattice and change the properties of the already existing oxygen vacancies, which are occupied by Pb 6s² electron lone pairs in pure lead ruthenate (Pb₂Ru₂O_6.5). Less than two Pb²⁺ ions mean less electrons for the vacancies. Decreasing the concentration of Ru⁴⁺ affects electrical conductivity, which is mainly governed by the RuO₆ backbone structure of ruthenate pyrochlores. We applied solid state synthesis and measured the thermoelectric properties of the ceramic compounds between 298 and 573 K. All compounds were isomorphic, like pure Pb₂Ru₂O_6.5. We analyzed our results for underlying scattering mechanisms of electrical (σ) and thermal conductivity (κ), the Seebeck coefficients (S) in terms of carrier concentrations, using existing quantum physical models. All ceramics were p-type and showed metal-like electrical conductivity and glass-like thermal conductivity. Therefore, the same scattering mechanisms were seen for all pyrochlores. Electrical conductivity σ(T) and electronic thermal conductivity κ_e(T) were governed by ‘electron impurity scattering’. The 3-phonon resistive process (Umklapp scattering) controlled the lattice thermal conductivity κ_L(T), which supports the electron-impurity scattering mechanism. In all compounds, the Seebeck coefficients were inversely proportional to the carrier concentration.

Alloy systems with multiple principal elements are considered attractive for thermoelectric materials because the mass and strain fluctuations introduced into the lattice can effectively scatter phonons and suppress the lattice thermal conductivity. However, both charge carriers and heat-carrying phonons are known to experience scattering due to alloying effects. We apply analytic transport models, based on perturbation and effective medium theories, to predict how alloy scattering will affect the thermal and electronic transport across the full compositional range of several pseudo-ternary and pseudo-quaternary alloy systems. For carrier mobility, the work of Makowski and Glicksman[1] provides a straightforward expression for alloy effects in a binary system, however, an extension to higher-order systems has not been presented. We develop a multicomponent extension to the alloy mobility model by applying computational thermodynamics techniques used to calculate excess Gibbs energy, namely the Redlich-Kister polynomial and Muggianu model. In the case of thermal transport, a recent reformulation of the Klemens alloy model[2] provides a straightforward route to compute the thermal conductivity of multicomponent alloys. From our calculations, we find that the thermal conductivity is lowest along the binary system with the largest mass contrast such that adding additional alloying elements is not necessarily beneficial from the standpoint of alloy scattering.

The alloy scattering analytic methods will predict smooth variations of transport properties with increasing disorder. However, through experiment, we can occasionally observe discontinuous changes in properties, resulting from abrupt transitions due to electronic band convergence or phase separation. In order to retain these experimental insights, we combine the physical models described above with the statistical Gaussian Process Regression (GPR) method. In this physics-informed GPR, the transport models are used to specify the prior means of the Gaussian process. Bayesian inference is then used to simultaneously update the prediction mean and variance using all the available experimental data. As a result, we can acquire both predicted values and their associated uncertainties, informed both by the physics-based functional form of alloy scattering and the full corpus of experimental data available.

References:
[1] Makowski, L., Glicksman, M. Disorder Scattering in Solid Solutions of III-V Semiconducting
Compounds. J. Phys. Chem. Solids. 34, 487-492, (1973)
[2] Gurunathan, R., Hanus, R., et al. Analytical Models of Phonon—Point-Defect Scattering. Physical Review Applied. 13 (034011) (2020)

Energy storage is arguably the most important single technology we have to develop in order to mitigate climate change. We have to store an overabundance of renewable energy, when the weather is favorable, and discharge it back to grid when there is demand. Commercially available technologies are unable to meet the future needs of the grid, and the most critical property is the cost per unit energy (CPE). Towards solving this problem, the Atomistic Simulation & Energy (ASE) group at MIT is developing an extremely low-cost (< $10/kWh) technology termed thermal batteries. This talk will review how the technology works and progress to date.

Energy-Efficient Adaptive Personal Thermal Management
Thermal homeostasis is one of the most vital factors for humans. Small deviations from thermal comfort zones can reduce work productivity, while sudden or large changes in thermal environments can lead to serious health issues such as cardiovascular diseases, influenza, and stroke. The indispensable nature of thermal comfort leads to enormous energy consumption for indoor temperature control, which constitutes 15% of the total US energy consumption and is responsible for considerable amounts of carbon footprint and climate change. To mitigate the issue and find the balance between thermal comfort and energy-efficient, personal thermal management has been demonstrated to be a powerful solution. By developing new wearable technologies with thermal engineering functionalities, we can provide the optimal local thermal environment to the users rather than the entire building. As our environments and physiological conditions change frequently, there is a pressing need to develop dynamic personal thermal management techniques that can essentially act as artificial thermoregulation. Nevertheless, the limited portable power source has again become a harsh constriction. There is still room to improve in terms of wearability, tunability, and energy efficiency.
In this talk, I will introduce our recent research progress of wearable varied emittance (WeaVE) devices that modulate the human body radiative heat transfer to accomplish thermoregulation with ultralow energy consumption. Compared with active generating/pumping heat, WeaVE devices are non-volatile and do not consume energy to maintain heating/cooling. I will explain three different versions of WeaVE devices, each of which with different application scenarios. (i) Electrochromic WeaVE vary the emissivity with only 0.5 V and are stretchable to fit the human body contour. (ii) Sweat-actuated WeaVE combines radiative and convective heat transfer to achieve large-range autonomous thermoregulation. (iii) Mechanical triple-mode WeaVE can vary between transmittance, emittance, and reflectance modes to expand from human skin to other objects with various emissivities.

Inverse opals have attracted significant interest over the last decade as a templated microporous architecture with great potential for high-performance phase-change heat transfer devices. However, commonly used fabrication methods for the initial opal template yield structures with limited permeability due to small achievable feature sizes and the presence of defects in the crystalline array, hindering the use and performance of inverse opals in thermo-fluidic applications. In this work, we show how slope self-assembly method is a promising approach for phase-change thermal management. We developed a model, which was validated by experiments, to achieve high throughput and controllable opal self-assembly method with the potential of overcoming limitations in previous fabrication methods. We demonstrated the ability to fabricate millimeter-scale polycrystalline opal monolayers with sphere sizes ranging from 500 nm up to 10 µm, while centimeter-scale opals, comparable with wafer-scale state-of-the-art demonstrations, were also possible with some sparse sphere stacking or small uncovered areas. Due to the large achievable pore sizes, which exceed by an order of magnitude the sizes possible with the vertical deposition method, these structures have the potential to yield inverse opals capable of withstanding heat fluxes up to two orders of magnitude higher than those made with other fabrication methods. In addition, due to the simplicity, high throughput, and large range of achievable feature sizes of the method, these structures can also be used in a broad range of applications such as molecule printing and high-performance electrochemical systems.

The topic of developing high-performance heat spreader has been around for a long time and there has been great advancement thanks to the community, but the demand for better heat spreader is getting larger than ever for thermal management of high-speed telecommunication and electric vehicles. In those application of important for sustainability and resilience, the large heat flux generated during the operation would overly elevate the temperature of devices, which can degrade the device and limit the performance. What is required is to effectively spread the condensed heat in three-dimensions using materials with (effectively) isotropically high thermal conductivity. In this talk, we will introduce two approaches; copper-diamond composite and graphite-copper composite. In both cases, the key is to bond materials with small thermal boundary resistance or large thermal boundary conductance (TBC).

As for the copper-diamond composite, an unconventional approach applying self-assembled monolayer (SAM) prior to the high-temperature sintering of copper/diamond composite was utilized to enhance the TBC between copper and diamond. The enhancement was first systematically confirmed on a model interface system consisting of deposited-copper-layer/SAM/diamond-substrate by detailed SAM morphology characterization and TBC measurements. TBC significantly depends on the SAM coverage and ordering, and the formation of high-quality SAM promoted the TBC. There, by systematically varying the end groups of SAM and the kind of substrate, some interesting insight was found. Although, TBC is typically positively correlated with interfacial adhesion at an interface, the experiments showed that a weak van der Waals interface can give a higher TBC than a strong covalently bonded interface. This occurs in a system with highly mismatched vibrational frequencies (copper/diamond) modified by SAM. Further analyses revealed that this is attributed to the trade-off between the bridging (of vibrational spectra) and binding effects [1]. Then, applying the knowledge to the copper diamond composite, the diamond particles were simultaneously functionalized by SAM with the condition giving the highest TBC in the model system and sintered together with the copper to fabricate isotropically high thermal conductivity composite [2].

For the graphite-copper composite, although graphite is a promising candidate owing to its high basal-plane thermal conductivity, its application has been restricted by its low c-axis thermal conductivity. Here, this issue is resolved by transforming graphite into an isotropic thermal conductor by building a structure that can effectively route heat in all three dimensions. We developed a double-decker structure with differently oriented graphite layers to realize high heat dissipation from a local heat source. The critical issue of bonding the graphite layers was overcome by a high-temperature process using copper as the binding layer. As a result, graphite/copper composite efficiently dissipates heat nearly isotropically and performs even better than the above copper-diamond composite [3]. The detailed experimental analysis and demonstration of the heat transfer will be presented at the talk.

[1] B. Xu, S. Hu, S.-W. Hung, C. Shao, H. Chandra, F.-R. Chen, T. Kodama, J. Shiomi, “Weaker bonding can give larger thermal conductance at highly mismatched interfaces”, Science Advances 7, eabf8197 (2021).
[2] B. Xu, S.-W. Hung, S. Hu, C. Shao, R. Guo, J. Choi, T. Kodama, F.-R. Chen, J. Shiomi, "Scalable monolayer-functionalized nanointerface for thermal conductivity enhancement in copper/diamond composite", Carbon, 175, 299-306 (2021).
[3] B. Xu, etal, (submitted).

Dynamic droplet events on heat transfer surfaces give rise to interesting and surprising behaviours that are relevant to advancing our understanding of thermal energy transfer across interfaces. Here, we discuss two particularly interesting cases; jumping droplet condensation on superhydrophobic surfaces and the impact of a volatile droplet on a superheated wetting surface.

In the first part of this talk we address the heat transfer performance of superhydrophobic surfaces limited by the low individual droplet growth rates associated with extreme droplet apparent advancing contact angles (θ_app → 180°) and the fundamental size limits of coalescence-induced droplet jumping. Our detailed condensation heat transfer modelling coupled with numerical simulations of binary (N = 2) and coordinated (N > 2) droplet coalescence, show that biphilic surfaces with smooth, low surface energy spots on a superhydrophobic background exhibit an unprecedented 10X higher jumping droplet condensation heat transfer coefficient when compared to homogenous superhydrophobic surfaces. Model predicted design optimization of the biphilic surface is validated against condensation experiments. Our findings clarify the role of droplet jumping dynamics and distribution densities by revealing optimum design guidelines for biphilic surface development for maximum condensation heat flux. Contrary to current understanding, we observe that spot wettability should not be optimized towards minimizing droplet nucleation energy barrier, rather it should minimize droplet adhesion while maximizing individual droplet growth rate.

We then describe the crucial impact thermal capillary waves and ambient gas rarefaction have on enhancing/limiting the jumping speeds of coalescing nanodroplets on low adhesion surfaces. By using high-fidelity non-equilibrium molecular dynamics simulations in conjunction with well-resolved volume-of-fluid continuum calculations, we are able to quantify the different dissipation mechanisms that govern nanodroplet jumping at length scales that are currently difficult to access experimentally. We find that interfacial thermal capillary waves contribute to a large statistical spread of nanodroplet jumping speeds. As the gas surrounding these liquid droplets is no longer in thermodynamic equilibrium, we also show how the reduced external drag leads to increased jumping speeds. This work demonstrates that, in the viscous-dominated regime, the Ohnesorge number and viscosity ratio between the two phases alone are not sufficient, but that the thermal fluctuation number (Th) and the Knudsen Number (Kn) are both needed to recover the relevant molecular physics at nanoscales. Our results and analysis suggest that these dimensionless parameters would be relevant for many other free-surface flow processes and applications that operate at the nanoscale.

In the second part of this talk, we present preliminary results on the Leidenfrost (critical heat flux) limit for dynamic droplet interactions with superheated surfaces. We identify a scientifically and practically relevant wetting behaviour in the Leidenfrost transition regime. We provide experimental observation of an inverted phase fingering instability demonstrating fractal-like wetting structures at the Leidenfrost transition indicating a rich interplay of fluid flow undergoing a change of phase and surface forces that is not currently well understood.

Ultrawide bandgap semiconductors made from high Al content AlGaN alloys and Ga₂O₃ have promise for future rf electronics and power switches. One of the key issues that arises from using these ternary alloys is the intrinsic low thermal conductivity of AlGaN and Ga₂O₃ and the low thermal boundary conductance at contacts with these alloys. This requires careful design of the device architecture and layout in order to yield effective heat dissipation pathways for AlGaN systems. In this talk, we will present modeling results which demonstrated specific regimes where cooling from the backside or topside of the ultrawide bandgap devices provide the most efficient pathway for heat dissipation. We will also present modeling and experimental results that demonstrate the effectiveness of the integration of high thermal conductivity dielectrics on the heat dissipation efficiency in these devices. Limitations on the integration of these layers through direct growth or bonding will be discussed. Finally, the impact of the reaction between metal contacts and UWBG materials like Ga₂O₃ will be discussed and its role on thermal transport from the devices analyzed through device modeling.

Power densification of electronics is greatly shaped by their co-design with emerging thermal management technologies. Heat spreaders represent a thermal routing technique implemented to reduce thermal resistance and limit electronic device temperature fluctuation. Here, we demonstrate the co-design of electronic systems through the monolithic integration of copper (Cu) directly on electronic devices to serve as heat spreaders and temperature stabilizers. We developed a fabrication recipe to electrically insulate the electronics with Parylene C and conformally coat them with Cu, demonstrating greater proximity to heat generating elements, elimination of thermal interface materials, and unprecedented cooling performance compared to existing cooling technologies. Our sustainable cooling method with 8 µm thick Parylene C can handle up to 600 V, providing exceptionally low device-to-ambient specific thermal resistances approaching 2.5 (cm2K)/W in quiescent air and 0.8 (cm2K)/W in quiescent water, with time constants of 110 s in air and 5.61 s in water. Our cooling technology has the potential to enable the continued realization of very compact and highly power dense electronics and reduction in the cooling energy consumption.

Passive Daytime Radiative cooling (PDRC) technology has the potential to significantly reduce our energy dependency for cooling buildings. It is an emerging, innovative green technology which uses the deep space as a cold sink to emit excess heat from an object through earth’s atmosphere mitigating the heat island effect in urban areas. Achieving an efficient PDRC cooler requires high emission through the spectral range (8-13 μm) where the atmosphere is transparent, while reflecting the incoming solar spectra^[1,2]. IR emissive polymers present a simple yet promising approach that enables controlling emission through molecular bonding^[3]. However, relatively thick polymer coatings (20-200µm) are required to achieve sufficient high emissivity. To realize high cooling performance, an emitter with low absorption of the downward heat radiation from the atmosphere is desired. Aforementioned spectral selectivity on the other hand, generally requires complex nano-engineered structures to achieve restricted emission in the spectral range of 8-13µm. We have recently demonstrated for the first time a PDRC structure based on a spectrally selective organopolysilazane planar emitter which is simple to manufacture ^[4].
Organopolysilazane are silicon-based polymers where alternating silicon and nitrogen atoms form the backbone of the polymeric structure along with methyl and vinyl group attached to it. Polysilazane (PSZ) barrier coatings have been studied widely for their chemical resistance, stability, corrosion protection properties and easy applicability. The exceptional high transmissivity in the UV-Vis region and flexibility of polysilazane based coatings enable versatile application of these coatings^[5,6]. We present here the optical properties and cooling performance of a simple dual layer spectrally selective PDRC structure with PSZ coating as a thermal emitter on a thin film silver reflector. Due to the high transparency of the polymer coating the structure allows for 97% reflection of the solar spectrum. We demonstrate the characteristic bond vibrations from the functional groups of the polymeric structure which coincides very well with the atmospheric transmittance window of 8-13µm resulting in the high emissivity of 0.9 in the 8-13µm wavelength range. The structure with a 5 µm thin coating can cool down to 6.8°C below ambient owing to a cooling power of 93.7 W/m². Based on long term outdoor performance tests and degradation tests under different environment, the changes in optical response and reliability of the PDRC structure was evaluated. We believe that polysilazane emitter coatings due to its simplicity and tenability will open new research possibilities for PDRC applications.
References-
[1] A. P. Raman, M. A. Anoma, L. Zhu, E. Rephaeli, S. Fan, Nature 2014, 515, 540.
[2] D. Zhao, A. Aili, Y. Zhai, S. Xu, G. Tan, X. Yin, R. Yang, Applied Physics Reviews 2019, 6, 21306.
[3] A. Aili, Z. Y. Wei, Y. Z. Chen, D. L. Zhao, R. G. Yang, X. B. Yin, Materials Today Physics 2019, 10, 100127.
[4] U. Banik, A. Agrawal, H. Meddeb, O. Sergeev, N. Reininghaus, M. Götz-Köhler, K. Gehrke, J. Stührenberg, M. Vehse, M. Sznajder, C. Agert, ACS applied materials & interfaces 2021, 13, 24130.
[5] U. Banik, K. Sasaki, N. Reininghaus, K. Gehrke, M. Vehse, M. Sznajder, T. Sproewitz, C. Agert, Solar Energy Materials and Solar Cells 2020, 209, 110456.
[6] U. Banik, N. Reininghaus, P. Seefeldt, M. Sznajder, K. Gehrke, M. Vehse, P. Spietz, T. Sproewitz, C. Agert, 2019 European Space Power Conference (ESPC) 2019.

The conventional thermoelectric (TE) cooling modules are optimized for refrigeration to pump heat from the cold end to a heat sink. To achieve the best energy conversion efficiency, it requires a combination of a large thermoelectric power factor and low thermal conductivity to reach a high figure of merit. Whereas active cooling aims at draining heat from the heat source to a heat sink, which makes the low thermal conductivity of the commercial TE cooling disadvantageous. The metallic thermoelectric modules, on the contrary, combine large thermoelectric power factor and large thermal conductivity, allowing a more efficient drain of heat via both active Peltier cooling and passive cooling than the passive or the TE refrigeration ones. The fact that the active Peltier cooling can be adjusted by applying current to the modules also provides a more controllable way of heat management. In this study, we show that a Cu-Ni alloy with both a high power factor and a reasonable thermal conductivity can be synthesized with cheap industrial level ingredients and via the simple ball-mill and hot-press process in the lab. A power factor of 42μW/(cm¹K²) can be reached at room temperature with the proper Cu-Ni composition and annealing process, which is surpassed the state-of-art study on Cu-Ni thermoelectric materials synthesized with expensive high purity elements. The effective thermal conductivity of the Cu-Ni alloy with proper composition can surpass the passive thermal conductivity of pure copper at raised temperature (>400K). The deposition parameters of Directed Energy Deposition (DED) have been optimized for the specific Cu-Ni composition. Dense samples with properties comparable with samples from the laboratory have been made and characterized. Our study shows the potential of Cu-Ni alloys as an active cooling material and the compatibility of Cu-Ni Alloy with additive manufacturing, which enables the TE active cooling modules to adopt the optimal geometry to minimize the interfacial thermal resistance between the heat source and the heat sink, as well as scale down for the heat management of integrated circuits.

Although much attention has been paid to the development of photothermal materials and designs that can convert solar irradiation into exploitable thermal energy, it remains some obstacles such as limited light absorbing band and narrow incident angle. This study proposes a black gold-coated hierarchical nanoturf (Au/h-nanoturf) membrane incorporated with randomly distributed high aspect ratio (AR)
nanostructures and micro-through holes. Thanks to structural advantages, this large area membrane exhibited good absorption of the broadband solar light spectrum. The h-nanoturf is further combined with a microcone array to enhance solar absorption extended to the near-infrared range as well as the omnidirectional incident direction of the light. The fundamental mechanism of the strong omnidirectional broadband absorption performance of the h-nanoturf was thoroughly analyzed with computational electrodynamics simulations. Consequently, we employed the Au/3D h-nanoturf with microscale hole (μ-hole) membrane to fabricate an advanced solar-vapor generator.

We present a theoretical study of the near-field radiative heat transfer between high-Tc superconducting thin films made of optimally doped YBa2Cu3O7−δ. The critical temperature for these ceramics is Tc = 93 K. We show that thin films significantly enhance the radiative heat transfer by efficiently coupling each interface’s optical modes. The heat transfer between thin films turns out to be up to one order of magnitude higher than that for bulk plates in normal and superconducting states. The change of the optical response above and below Tc also plays an important role since the superconducting phase transition leads to an abrupt suppression of the total heat flux. A detailed study of the different modes and the heat transfer is also presented, showing that for these materials, the mir infrared damping of the dielectric function determines to a significant extent role in the magnitude of the heat flux.

Achieving an efficient management of the heat dissipated in a variety of devices, is important for their proper functioning, and to prevent their uncontrolled failure. Research in the field of thermal regulation has mainly focused on the development of complex fluids through the dispersion of nanoparticles in different solvents (Keblinski et al., Materials Today, 8, 36, 2005) and on the manufacture of ‘defective’ solid materials, in which the thermal conductivity is limited by the introduction of impurities or interfaces at the atomic scale. However, these materials have several drawbacks: in the case of nanofluids, the high concentrations of dispersed particles needed to achieve a substantial change in thermal conductivity often result in rheological problems, preventing their application; in the case of ‘defective’ solids, the artificial impurities cannot be modified once the material has been manufactured, so that the modulation of thermal conductivity by an external stimulus is impossible. Thus, the development of efficient and tunable thermal regulators is one of the challenges in materials science.
‘Soft matter’ thermal devices based on liquids, gels or mesophases, are promising candidates in this direction. Shin et al. reported a reversible threefold change in the thermal conductivity of an azobenzene polymer induced by light (Shin et al., PNAS, 116, 8629, 2019).
Here we report the synthesis and thermal conductivity of 4,4’-dialkyloxy-3-methylazobenzene derivatives, with different alkyl chain lengths (between 3 and 10 carbon atoms). These systems show a rich thermal phase diagram, with different crystalline, anisotropic mesophases, and isotropic liquid configurations, as a function of temperature. The reversible changes in the alignment of the alkyl chains along these phases results in reversible variations of the thermal conductivity up to 26%. Moreover, the conformational transition between trans and cis azobenzene groups under UV-Vis illumination leads to reversible crystal-to-liquid transition, changing the thermal conductivity up to 40%.
We discuss the effect of chain length and temperature on the kinetics of the transformation, and on the stability of the different phases formed.
The large variety of light-responsive chemical groups, combined with the powerful strategies developed by organic and polymer chemists over the last decades, suggest that this could be a successful approach to develop effective thermal switches.

We analyze near-field thermophotovoltaic systems in terms of their open-circuit voltage (Voc). Unlike previous numerical results with fluctuational electrodynamics, we introduce a simple analytic model that captures the physics of near-field thermophotovoltaic systems and predicts their performance metrics in terms of both current and voltage. We show that (i) operating in the near-field naturally enhances radiative recombination, and that (ii) owing to photon recycling and minimal radiation leakage, near-field thermophotovoltaic systems can be much thinner compared to typical PV cells, thus reducing non-radiative losses. Combination of these effects leads to values of Voc that can approach the radiative limit.

The work function is the key surface property that determines how much energy is required for an electron to escape the surface of a material. This property is crucial for thermionic energy conversion, band alignment in heterostructures, and electron emission devices. The discovery of thermally stable, ultra-low work function materials (less than 1.5 eV) would allow thermionic conversion of heat (>1500 °C) directly to electricity with efficiencies exceeding 30% (typical thermionic and thermoelectric converters have efficiencies around 10%). In recent years, data-driven approaches based on high-throughput ab-initio calculations have emerged as a new paradigm to facilitate the search through vast chemical spaces for new materials with tuned properties or novel behavior. The rapid increase in available computational data structured in open source material databases such as Materials Project, AFLOW, and NOMAD has opened an avenue towards material discovery using data-mining and statistically driven machine learning approaches. However, most big material databases largely lack to report surface properties like the work function as each bulk material typically has dozens of distinct low-index crystalline surfaces and terminations. Here, we report a high-throughput workflow using density functional theory (DFT) to calculate the work functions of 29,270 surfaces (23,603 slab calculations) that we created from 2,492 bulk materials, including up to ternary compounds. On the low tail end of the work function distribution, we discover 162 surfaces with an ultra-low (<2.5 eV) work function. Based on this database we develop a physics-based approach to featurize surfaces and use supervised machine learning to predict the work function. We find that physical choice of features improves prediction performance far more than choice of model. Our random forest model achieves a mean absolute test error of 0.19 eV when predicting the work function, which is more than 4 times better than the baseline and comparable to the accuracy of DFT. This surrogate model enables rapid predictions of the work function (~10⁵ faster than DFT) across a vast chemical space and facilitates the discovery of material surfaces with extremely low work functions for energy conversion and electronic applications.

Thermoelectric (TE) materials grabbed much interest in various applications such as cooling, power generation, sensors, aircraft, etc. due to their ability to convert waste thermal energy into electric energy^1–3. The interdependency among thermal conductivity, electrical conductivity, and Seebeck coefficient acts as a barrier to enhance thermoelectric performance. However, thanks to nanotechnology, it is possible to break this interdependency by employing interface phonon scattering approach to enhance the TE efficiency. Introducing lattice mismatch concept at nanoscale can decrease the lattice thermal conductivity more effectively than electrical conductivity⁴. Also, Seebeck coefficient can increase due to the quantum confinement effect at nanoscale, without decreasing the electrical conductivity⁵.
The current work is the continuation of our previous work based on the synthesis and characterization of poly (3,4-ethylene dioxythiophene): polystyrene sulfonate (PEDOT: PSS) filled with different wt. fractions (0.001-0.05) of inorganic graphene oxide-titanium dioxide (GO-TiO₂). Here, we succeed to break the interdependency of thermoelectric parameters by enhancing electrical conductivity and Seebeck coefficient whilst deducing thermal conductivity as a function of filler wt. fraction.
The present work is more focused on in-depth analysis of the obtained experimental results by using theoretical predictions. Firstly, the electrical conductivities, the thermal conductivities, and the Seebeck coefficients of aforementioned composites were measured experimentally. Secondly, the experimentally obtained results were fitted by using Effective medium theory (EMT) and Generalized effective medium theory (GEMT) based on Landauer and Sonntag's equations. The obtained best-fit parameters demonstrate the individual TE properties of the matrix and the filler.
EMT and GEMT have been directly applied to understand the evolution of electrical and thermal conductivities of the composites as a function of filler wt. fraction. However, for the Seebeck coefficient, we have utilized the Sonntag’s extended model for degenerate semiconductors where the heat flux was replaced by the entropy flux. Therefore, the ratio of thermal conductivity to Seebeck coefficient is considered as a transport parameter and has been utilized to understand the evolution as a function of filler wt. fraction instead of Seebeck coefficient alone. The obtained best-fit parameters for the matrix and the filler were found to be in good agreement with the literature.

Keywords: Thermoelectricity, thermal conductivity, Electrical conductivity, Seebeck coefficient, Generalized effective medium theory.

Bibliography:
1. Samson, D., Otterpohl, T., Kluge, M., Schmid, U. & Becker, Th. Aircraft-Specific Thermoelectric Generator Module. J. Electron. Mater. 39, 2092–2095 (2010).
2. Kim, S., Park, S., Kim, S. & Rhi, S.-H. A Thermoelectric Generator Using Engine Coolant for Light-Duty Internal Combustion Engine-Powered Vehicles. J. Electron. Mater. 40, 812–816 (2011).
3. LeBlanc, S. Thermoelectric generators: Linking material properties and systems engineering for waste heat recovery applications. Sustain. Mater. Technol. 1–2, 26–35 (2014).
4. Ahmad, K., Wan, C. & Zong, P. Thermoelectric properties of BiSbTe/graphene nanocomposites. J. Mater. Sci. Mater. Electron. 30, 11923–11930 (2019).
5. Tavkhelidze, A. Large enhancement of the thermoelectric figure of merit in a ridged quantum well. Nanotechnology 20, 405401 (2009).

Advances in thermal and energy sciences bring transformational efficiency enhancements in energy, water, manufacturing processes, and electronics cooling by fundamentally manipulating energy processes across multiple length and time scales. Herein, the energy processes are naturally complex, and create dynamic changes in thermofluidic features, which are represented as bubbles or droplets. Furthermore, the improvements in the energy process critically depend on our ability to design and realize materials with optimal properties, adding another level of complexity. Unveiling energy processes integrated with materials design can be possible by collecting interpretable and rich physical descriptors from bubbles or droplets. These challenges for extracting physics from complex, dynamic, and fast-moving objects may be addressed using revolutionary intelligent machine vision. Therefore, one of my research interests is to build a data-driven framework based on live images by converging the recent advancements in artificial intelligence (AI)-enabled machine vision, data processing, and deep learning models, to push the boundaries of knowledge about thermofluidic physics.

In this talk, I will discuss AI-enabled innovations, data-centric thinking, and opportunities, by showcasing a few examples. 1) I will first introduce the modular framework that can produce high-throughput fast data acquisition, streamline automatic data processing, and create a spatiotemporal thermal mapping at extreme resolutions. 2) I will explore input space (e.g., material or surface parameters) by generating imaging data on dynamic bubble/droplet topologies along with heat transfer data where the experimental design and heat transfer performance will be fully connected with physical descriptions. Therefore, the rational design of material or surface parameters can engineer solid-liquid-vapor contact lines that result in favorable phase change curves. 3) Then, I will discuss avenues to design functional thermal architectures that are widely employed as high-surface conduits for thermofluidic applications.

Previous density functional theory work suggested that SrVO_3-δ is a promising candidate as electron emissive material for thermionic energy conversion because of the predicted ultralow work function (1.86eV). Additionally, A- and B-site doping in ABO₃ perovskites can increase power factors and decrease thermal conductivity compared to traditional oxide perovskites for thermoelectric energy conversion. To experimentally explore the potential of using SrVO_3-δ and its mixed perovskites for thermal energy management via thermionic or thermoelectric conversion, bulk polycrystalline SrVO_3-δ samples were synthesized via spark plasma sintering. The effects of Ba- and Ti- doping on the microstructure, electrical conductivity, and thermal stability of SrVO_3-δ were investigated. Results revealed that the microstructure consists of predominant SrVO_3-δ phase with minor SrTiO_3-δ phase, with grain size of approximately 15 μm. The thermionic emission behavior tested in an Ar plasma will be discussed as well.

The anomalous thermoelectric transport properties of iron-telluride (FeTe) near magneto-structural transition temperature (~70K) have been reported earlier with little insight. The coexistence of magnetism and quantum effects can be the probable origin of these anomalies, requiring detailed investigations of the magnetic, galvanomagnetic, and thermoelectric transport properties. The magnetic susceptibility shows a bicollinear stripe type spin ordering in the antiferromagnetic domain and short-range ferromagnetic ordering at the beginning of the paramagnetic domain. The galvanomagnetic properties reveal a multicarrier transport impacted by spin fluctuations. Further, phonon- and magnon-drag contributions are evaluated related to the origin of the anomalous thermoelectric properties. The small entropy contribution from the magnonic heat capacity supports the phonon-drag as the origin of a thermopower peak at ~50K, while the field-dependent reduction of the thermopower is associated with the field-dependent modification of the spin-phonon interaction. The field-dependent change of the paramagnetic thermopower is attributed to the spin-fluctuation suppression or the paramagnons. The sharp drop in thermal conductivity above ~70K is related to the structural change modifying the phonon modes. The knowledge gained from studying the spin and quantum effects on transport properties of FeTe is instrumental for designing high-performance spin-driven thermoelectrics.

A major goal of thermoelectric (TE) research is to find or develop materials with improved efficiencies. One strategy towards higher-performance TEs is band convergence, which seeks to align multiple electronic valleys within a narrow energy range. This approach can boost the power factor, a component of TE efficiency, compared to a single valley by simultaneously increasing the Seebeck coefficient and the electrical conductivity – two competing quantities. While band convergence has been demonstrated experimentally and is the source of many theoretical studies, rigorous scattering calculations indicate that the benefits of this strategy depend sensitively on the strength of inter-valley collision processes. Unlike intra-valley scattering with both initial and final states residing within the same valley, inter-valley processes scatter carriers from one valley to another. The latter can increase significantly when aligning multiple valleys, and potentially offset the gains from band convergence.

In this talk, we present results of a first-principles investigation to characterize, and to understand what controls, intra- and inter-valley electron-phonon scattering in TE materials. For this study we selected six materials, including three lead chalcogenides (PbS, PbSe, PbTe) and three half-Heuslers (ScNiBi, ScPbSb, ZrNiSn), which possess multiple equivalent conduction valleys. Density functional theory (DFT) was utilized to rigorously compute the electron-phonon scattering and TE characteristics. The intra- and inter-valley scattering rates are analyzed by separating the contributions originating from the electron dispersion and from the electron-phonon coupling, and by elucidating what controls each component. We find that increasing the energy of the zone-edge phonons is effective at reducing inter-valley scattering and enhancing the power factor by >50%. Finally, since DFT electron-phonon calculations can be very demanding, we propose a simple approach to estimate the inter-valley coupling strength, which is found to agree qualitatively with the rigorous results. These findings provide insights into what controls the scattering characteristics to help to guide our search and design of new TE materials with reduced inter-valley scattering and enhanced power factors.

This research was supported by NSERC (Discovery Grant No. RGPIN-2016-04881) with computational resources provided by Compute Canada.

Weyl semimetals combine both topological and semimetallic effects in their transport of fermions through both bulk and surface states, making them excellent candidates for transverse thermoelectric transport via the Nernst effect. The Nernst effect is a transverse thermoelectric effect in which an applied magnetic field and perpendicular temperature gradient produce a mutually perpendicular voltage. The applied field creates a skew force separating charge carriers. Contributions from both polarities of charge carriers add together, so two-carrier systems such as NbP and other Weyl semimetals are advantageous. Single crystalline NbP has shown an unprecedentedly large Nernst thermopower, exceeding 800 microV K-1 at 109 K and 9 T [1]. Bulk polycrystalline NbP maintains a large Nernst thermopower, although its magnitude is decreased by approximately a factor of 8 under similar circumstances [2]. Here, we explore both transverse and longitudinal thermoelectric and thermomagnetic transport as a function of grain size in bulk polycrystalline samples of NbP, where different annealing times result in different grain sizes. The polycrystalline samples show a significantly larger Seebeck coefficient in comparison to results in single-crystalline NbP, however this decreases with increasing grain size. The Nernst thermopower is much smaller when compared to single-crystalline NbP, however the Nernst effect is still very pronounced in polycrystalline NbP, even as the Nernst thermopower decreases with increasing grain size. Polycrystalline NbP also exhibits a large isothermal magneto-Seebeck effect, which was rigorously not the case in isothermal measurements of single-crystalline NbP. The polycrystalline samples’ deviation in transport from results on single-crystalline NbP are hypothesized to be due to increasing the number of grains and thus interfaces of the topological material, which can then cause topological states to potentially dominate transport. The field dependence of electrical resistivity, thermal conductivity, and the Hall effect in each sample have also been determined. Chemical analysis has been completed using ICP-OES in order to determine the effect of annealing on composition and ensure it remained consistent for each sample. Via microstructural characterization and thermoelectric transport measurements, we can begin to determine the effects of grain size on transport in Weyl semimetals.
[1] S. J. Watzman et al. Phys. Rev. B 97(16), 161404(R) (2018).
[2] C. Fu et al. Energy Environ. Sci. 11(10), 2813-2830 (2018).

Thermoelectric devices placed in the core of a nuclear reactor can generate electricity from thermal energy to enable self-powered sensors for in-situ monitoring of critical parameters of the nuclear fuel assembly, increasing power plant safety. The in-situ performance of a Thermoelectric Generator (TEG) has never been tested in the extreme environment of a nuclear reactor core. A TEG made from state-of-the-art nanostructured bulk half-Heusler materials was inserted into the core of a nuclear reactor and irradiated for 30 days to an exposure of 153 megawatt-days. The TEG electrical resistance, voltage, and power output were measured continuously while it generated electricity using radiation as a heat source. Despite a sharp drop in TEG power when operating at relatively lower temperature, the TEG showed a 1900% increase in power output due to in-situ annealing and a significant recovery of both the Seebeck coefficient and electrical conductivity when operating at increased temperature and reactor power. Atomic scale modeling indicates the irradiation provided sufficient energy to induce a partial phase transformation, converting the semiconducting half-Heusler structure into a metallic phase stable at low temperatures. However, with sufficient thermal energy, the structure reverts to the original non-irradiated phase and thermoelectric properties quickly recover. The same materials comprising the TEG were separately irradiated with alpha particles and scanning thermal microscopy was used to determine the influence of irradiation dose on thermal conductivity and Seebeck coefficient while microstructural characterization of the irradiated materials confirmed predictions from atomic scale modeling. These results indicate the thermoelectric materials experienced thermally activated healing during which irradiation-induced defects were annihilated and the material microstructure partially recovered along with the thermoelectric properties. These results suggest that with proper control over the TEG operating temperatures, the nanostructured half-Heusler TEG could produce power and operate indefinitely in the core of a nuclear reactor. This work creates opportunities for solid-state power generation utilizing high-efficiency nanostructured thermoelectric materials in the harsh environment of a nuclear reactor.

MoP (Molybdenum Phosphide) is a topological metal which has been shown to host a triply degenerate point in its electronic structure and a pair of Weyl points in its bulk electronic structure [1]. Its complex band structure makes it unique for electrical and thermal transport, with single-crystalline MoP previously demonstrating an extremely high electrical conductivity and the potential for hydrodynamic charge flow [2]. In this work, both longitudinal thermoelectric (Seebeck effect) and transverse thermomagnetic (Nernst effect) phenomena are investigated for the first time in polycrystalline samples of MoP over a range of temperatures (2 K – 400 K) and externally applied magnetic fields (-9 T to +9 T). Temperature and magnetic field dependent measurements of thermal conductivity, electrical resistivity, and the Hall effect are also investigated. This study is extended to grain-size dependence of the aforementioned physical properties by annealing different samples for varying amounts of time to promote grain growth, with average grain size confirmed via SEM.
Compared to single-crystalline MoP [2], thermal conductivity is found to be approximately 2 orders of magnitude smaller while electrical resistivity is 1-2 orders higher in polycrystalline MoP. The Seebeck coefficient is small, due to a very high carrier density as is the case with most metals, with a maximum value of ~4 μV K^-1 at 380 K. A small magneto-Seebeck effect is seen at temperatures below 200 K with the magnitude of the Seebeck coefficient decreasing in a magnetic field. At temperatures below 100 K, thermal conductivity has also been found to decrease in a magnetic field, with the effect most pronounced at the peak of thermal conductivity in the temperature range of 28-35 K. The magnetoresistance effect is large and increases with decreasing temperatures, showing a similar trend in comparison to single-crystalline MoP [2]. Moreover, a minimal Nernst effect is detected. Furthermore, the effect of grain size is clearly visible in these measured transport properties. Thermal conductivity increases significantly with increasing grain size, while the electrical resistivity reduced marginally with larger grain size. The magnitude of the Seebeck coefficient also increased in larger grain-sized samples. These changes can be attributed to the reduced number of grain boundaries in the material due to annealing and hence lesser obstruction to charge flow.
[1] B. Q. Lv et al. Nature 546, 627–631 (2017).
[2] N. Kumar et al. Nat. Commun. 10, 2475 (2019).

Thermal transport in porous structures is yet to be fully understood. For example, the influence of the microstructure’s heterogeneities in porous electrodes of Li ion batteries remain poorly addressed, in part due to challenges in accurate thermal measurement of porous samples. In this talk, we introduce an approach based on time-domain thermoreflectance (TDTR) to accurately measure the thermal conductivity of porous materials, by transferring a metal film to bridge the micro-pores to provide a continuous, flat, and smooth surface for TDTR measurements. We demonstrate our approach by measuring the cross-plane thermal conductivity (k_PnC) of a wide range of microscale SiO₂ phononic crystals, with pore diameters of 1 – 3.5 µm and porosity of 13 – 50 %. To achieve accurate TDTR measurements, we ensure that the metal transducer films are sufficiently stiff and thus establish good thermal contact with the necks of the phononic crystals. The good agreement between our k_PnC measurements and calculations of effective medium theory validates our approach. We also introduce a guideline that allows the design of the film transfer process onto the porous materials. This approach could be conveniently employed in future studies of heat transport in a wide range of porous materials.

The control of temperature in lithium-ion batteries (LIBs) is critically important for the performance, reliability, and safety of LIBs. To design proper thermal management strategies, a greater understanding of thermal effects in LIBs is needed. The reversible heat in LIBs due to entropy change is fundamentally important to understand the chemical reactions in LIBs. However, the reversible heat has not been directly measured previously due to the limited temperature resolution of applied thermometry. We develop an ultra-sensitive thermometry with a noise floor of 32 μK/sqrt(Hz) and a noise equivalent temperature resolution of ± 10 μK, which is several orders of magnitude higher than previous temperature measurements applied on LIBs in the literature. We observe the reversible heat absorption of a LIR2032 coin cell during charging with negligible irreversible heat generations at a charging rate of C/200. The entropy changes of the cell obtained from the directly measured reversible heat agree excellently with the entropy changes obtained by measuring temperature dependent open circuit voltage (OCV). This is the first verification of the way to obtain entropy changes by measuring temperature dependent OCV which is widely used in the literature, while the changes in mechanical stress with temperature could also possibly change OCV. Our work enables a new regime of studying thermal effects in batteries at charging rates orders of magnitude lower than previous works, and significantly contributes to fundamental understanding of the entropy changes of the chemical reactions in LIBs.

A lot of energy is lost as waste heat in the process of producing and consuming electricity. Utilizing a vast amount of heat energy is essential for realizing a sustainable society. Thermoelectric conversion technology is one of the solutions for utilizing waste heat effectively because it enables a direct conversion of heat into electricity without any moving parts. The higher manganese silicide of MnSi_γ is a promising thermoelectric material composed of less-toxic elements with a material figure of merit ZT reaching up to 1.0. We have previously reported that it is possible to reduce the thermal conductivity and improve ZT by co-doing Re and Ge to the matrix of MnSi_γ. In this study, silicide thermoelectric modules using p-type (MnRe) (SiGe)_γ were prepared and its power generation characteristics were evaluated.
An alloy powder of p-type (MnRe)(SiGe)_γ was prepared by a mechanical alloying method, and then sintered using a spark plasma sintering method. For the n type, a Mg₂Si-based material, which was developed by Yasunaga Co., was used. To join the p-type, n-type elements and Cu electrodes, the modules were sintered at 550 – 750°C with a pressure of 10 MPa.
We investigated the dependence of thermoelectric properties on the Re and Ge amount in (MnRe)(SiGe)_γ. The ratio between sublattices of MnSi_γ, γ, was clarified to be reduced with increasing Re and Ge amount. Both Seebeck coefficient and electrical resistivity decreased, and the power factor increased, owing to the generated carriers by the reduced γ. On the other hand, the lattice thermal conductivity decreased with increasing Re and Ge amount due to the mass-difference and strain-field-difference phonon scattering. As a result, ZT increased up to 0.70 at 500°C when the substitution amount of Re was 0.10 and that of Ge was 0.03. The power generation characteristics of the 1-pair and 9-pair thermoelectric modules composed of (MnRe)(SiGe)_γ and Mg₂Si were evaluated by changing the elements heights h = 1.0, 2.0, 3.0, 3.5 mm when the hot side and cold side are 500°C and 20°C, respectively. The output power density increased as the element height decreased and showed the similar tendency as the results of the power generation characteristic simulation. As the element height decreased, the temperature difference applied to the element decreased and the open voltage also decreased, but the electrical resistance value also decreased significantly. As a result, output power density took the maximum value with respect to the element height. The maximum output power density of 14.3 kW/m² at 500°C was obtained when h = 2.0 mm. On the other hand, module resistance analysis revealed that interfacial resistance was present in about 20 – 30% of the total due to non-uniform bonding between the thermoelectric elements and the electrode. To improve the binding condition, sintering temperature of the modules were varied in the temperature range of 550 – 750°C. As increasing sintering temperatures, the module resistance decreased due to the improvement of the bonding between the elements and electrode. At high sintering temperatures above 650°C, although the module resistance decreased, the open voltages of the modules decreased due to the diffusion of Cu from the electrode to the elements, and the output power also decreased. As a result, maximum output power density of 17 kW/m² was obtained at the sintering temperature of 600°C, which is the highest level in the reported thermoelectric module using MnSi_γ and Mg₂Si. The power generation simulation suggests that the maximum output power increases up to 22 kW/m² if the interfacial resistance in the module were completely removed.
This study is based on results obtained from the Future Pioneering Program “Research and Development of Thermal Management and Technology” commissioned by the New Energy and Industrial Technology Development Organization (NEDO). This work is also supported by TherMAT.

Thermoelectric materials directly convert thermal energy to electrical energy and vice versa. To improve the energy conversion efficiency, accurate characterization of the atomic structure is critical, as this defines the charge- and phonon-transport. Scanning transmission electron microscopy (STEM) is widely used for defect characterization in thermoelectric materials, as structural defects and lattice strain often play a critical role in phonon scattering. However, the recently emerging high entropy thermoelectric materials are challenging objects for this characterization due to their very small strain and atomic-scale elemental mixture.
Here we show that these challenges can be solved by careful sample preparation, microscope fine-tuning, and the application of new experimental methods. Three STEM-based techniques are applied: atomic resolution energy-dispersive X-ray spectroscopy (EDS), geometric phase analysis (GPA), and nanobeam diffraction analysis (NBED) to understand the elemental positions and subtle lattice distortion, which are important for understanding the structure-property relationship.
Our high entropy material has a composition of Pb_0.89Sb_0.012Sn_0.1Se_0.5Te_0.25S_0.25 and was compared to the low entropy PbSe control material. The atomic positions of Pb, Sn, Te, Se were found by atomic resolution EDS mapping. Pb and Sn occupy the ‘Na’ position, while Te and Se occupy the ‘Cl’ position of the NaCl crystal structure, demonstrating that the elements were mixed homogenously at the atomic level.[1]
GPA can quantify the local strain distribution. However, weak strain fields require careful attention since the inevitable sample drift and scan noise will induce artifacts. To optimize this, we first wait a long time to minimize the sample drift, then we adopt fast scanning and take a stack of 30 frames, and finally, we use the ‘Smart Align’ software plugin to align the stacks together. This can simultaneously reduce the artifacts from sample drift and from scan noise, to give a strain analysis with 0.3% accuracy. With this, we find that the strain distribution in our high entropy alloy is much wider than that of low entropy alloy. The histogram shows that the statistical distribution in the high entropy alloy is much wider along with the normal <002> and shear <220> directions than in the low entropy alloy. This is important information, as the normal and shear strains are directly linked to the lattice thermal conductivity.[1]
As GPA requires atomic resolution images, the observational area cannot be very large and NBED is used to overcome this shortage. We use a 4 nm parallel probe to scan a much larger area of 400 nm by 400 nm to obtain a series of electron diffraction patterns. With careful analysis, we are able to map the strain with an accuracy of 0.06%. The results of NBED indeed confirm that the strain distribution of the high entropy alloy is much wider than the low entropy alloy. However, the absolute values in the strain field by NBED are lower than what was obtained with GPA. This shows the strength of GPA at a very small scale: in this high entropy alloy, the local strain varies in areas smaller than 2 nm and the strain measured with NBED only gives an average of the probed 4 nm area. In the case of our high entropy alloy, a full picture of the strain fields will therefore only be obtained by a combination of the GPA and NBED results.[1]

The authors acknowledge support from the Singapore Ministry of Education (MOE), AcRF Tier 1 (R-284-000-179-133) grant.

Reference:
[1] B. Jiang, Y. Yu, J. Cui, X. Liu, L. Xie, J. Liao, Q. Zhang, Y. Huang, S. Ning, B. Jia, B. Zhu, S. Bai, L. Chen, S. J. Pennycook, J. He, High-entropy-stabilized chalcogenides with high thermoelectric performance. Science (80-. ). 371, 830–834 (2021).

Sustainable development goals (SDGs) have recently been considered across the world. Keeping requirements of sustainable and clean energy in mind, energy harvesting is one of the important technologies, since this technology can convert waste heat into electricity. The thermoelectric (TE) figure of merit ZT is expressed by σS²T/κ, where σ is electrical conductivity, S is Seebeck coefficient, T is absolute temperature and κ is thermal conductivity. For the practical applications, TE materials with ZT > 1 have been desired. On the other hand, over 80% of the waste heat generated in industry is in the ranges of 373−575 K.¹⁾ Therefore, sustainable TE materials consisting of earth abundant and relatively non-toxic elements whose ZT value is higher than 1 at such low temperature range have been desired for many years.
There are several methods to improve ZT of TE materials. Among them, the fabrication of TE materials by sintering chemically-synthesized nanoparticles (NPs) has attracted much attention, mainly because κ can be significantly reduced through phonon scattering at the interfaces of well-defined nanograins. Comparing to the conventional ball milling technique, the use of chemical-synthesis of NPs as building blocks has several advantages.^2,3) For example, the fabrication time required to obtain a pellet is shorter; the size of well-defined nanograins in the pellet can be easily controlled; various kinds of nanoinclusions can be introduced in the material just by mixing two or more different types of NPs; mesoscale defects can be created by controlling the size of domain of nanoinclusions; atomic defects can readily introduced in NPs. In other words, it offers unprecedented opportunities for the hierarchical defect engineering.
Previously, we synthesized Zn doped Cu₂SnS₃ NPs (Zn-CTS) with changing Zn content. The results showed that the pellet fabricated by either Cu₂Sn_0.85Zn_0.15S₃ or Cu₂Sn_0.95Zn_0.05S₃ NPs showed relatively high ZT value of 0.37 at 670 K than that of pristine CTS fabricated by the conventional method.⁴⁾ Basically, Zn incorporation causes hole doping, and thus, as Zn content increased, σ of Zn-CTS increased. However, when the material became metallic, κ increased according to the Wiedemann–Franz law. In order to decrease carrier thermal conductivity κ_car, Cu₂Sn_0.9Zn_0.1S₃ NPs, which has the lower κ, were mixed with Cu₂Sn_0.85Zn_0.15S₃ NPs, which has the higher σ, in the ratio of 1:9. Owing to the sea-island structure composed of Cu₂Sn_0.9Zn_0.1S₃ NPs (island) and Cu₂Sn_0.85Zn_0.15S₃ (sea), the pellet could maintain high σ while κ showed lower value than that of pure Cu₂Sn_0.85Zn_0.15S₃ because of the efficient phonon scattering at the interfaces between sea and islands. As a result, the highest ZT value of 0.64 was achieved at 670K.⁵⁾
As this example shows, there is plenty of room for improvement of ZT by the hierarchical defect engineering using NPs as building blocks. In this talk, we will show you our recent results about the hierarchical defect engineering for discussion.
References:
1) S. O. J. Long et al., Chem. Mater., 30, 456 (2018); 2) S. Fei et al., AIP Adv., 10, 075021 (2020); 3) A. O. Moghaddam et al., Iran J. Mater. Sci. Eng., 16, 20 (2019); 4) W. Zhou et al., Appl. Phys. Lett., 111, 263105 (2017).; 5) W. Zhou et al., ACS Appl. Nano Mater., 1, 4819 (2018).

N-type Mg₃Sb₂-based compounds, discovered by our group in 2016, are one of the potential candidates for low-cost and high-performance thermoelectric materials comparable to the commercial Bi₂Te₃ [1]. A Bi-rich side of N-type Mg₃Sb₂-Mg₃Bi₂ solid solution has recently attracted attention by its excellent low-temperature thermoelectric performance [2]. One possible, yet controversial, mechanism is the band convergence between the six-valley carrier pocket (CB₁ band) and the single-valley band at Γ-point (Γ band) with increasing the Mg₃Bi₂ content [3, 4]. This motivates us to perform DFT-based modelling of electronic structure and thermoelectric properties for Mg₃Sb₂-Mg₃Bi₂ solid solution.
Our analysis reveals that the band convergence does not occur even for Mg₃Bi₂ end, but ultra-light Γ band slightly above the CB₁ minima enhances its power factor. We find that, by increasing Bi content, the band effective mass of the Kane-type Γ band becomes one order of magnitude lighter than that of the parabolic CB₁ band, where the energy gap is nearly closed. Despite the single-valley and not converged character, the contribution of the ultra-light Γ band to the transport properties is as significant as the six-valley CB₁ band, leading to the enhanced power factor greater than 30μW/cmK² for Mg₃Bi₂ end. In this talk, we also compare our computational modeling with experimental results. Discovery of the ultralight Γ band in Mg₃Bi₂-based materials will open the possibility of ultra-light-mass band engineering to design high-performance thermoelectric materials.

[1]. H. Tamaki, H.K. Sato, and T. Kanno, Adv. Mater 28, 10182 (2016).
[2]. Zhijia Han, Zhigang Gui, Y. B. Zhu, Peng Qin, Bo-Ping Zhang, Wenqing Zhang, Li Huang, and Weishu Liu, Research, 2020, 1672051 (2020).
[3]. K. Imasato, S. D. Kang, G. J. Snyder, Energy Environ. Sci. 12, 965 (2019).
[4]. J. Zhang and B.B. Iversen, J. Appl. Phys. 126, 085104 (2019).

Due to the current growing energy issues, the importance of thermoelectric energy conversion has been increasing. Recently, tin selenide (SnSe) with a layered crystal structure has extensively been studied as it exhibits a high figure-of-merit ZT > 2 [1]. However, SnSe has a serious issue; i.e., the carrier polarity control is still difficult, which is critical to realize p-n homojunction devices. Pure SnSe is a native p-type semiconductor, and p-type conductivity is enhanced by impurity doping, e.g., by alkali ion-doping [2]. Bi and halogen ions have been employed for electron doping at the Sn and the Se sites, respectively [3-5]. However, the maximum carrier concentration has been limited to 10¹⁷ cm^-3 because of the low activation efficiency (less than 0.1%) of the Bi³⁺ donor. Antimony ion (Sb³⁺) doping at Sn²⁺ site is expected to produce n-type SnSe, and the Sb-doped SnSe is reported to show high thermoelectric efficiency with ZT = 1.1 even in practical polycrystalline form. However, there are contradicting reports on the charge polarity (n-type or p-type); i.e., some papers report n-type and the others p-type [6-8].
In this work, we have discovered an unusual phenomenon, the double inversion of the charge polarity, in (Sn_1-xSb_x)Se by varying only the Sb concentration (x), and elucidated the doping mechanism. Pure SnSe shows p-type conduction, while (Sn_1-xSb_x)Se shows n-type conduction for 0.005 < x < 0.05, and finally re-inverts to p-type conduction for x > 0.05. The substitution sites of Sb distributes for the Sn sites and the Se sites, and the major substitution site switches from the Se (Sb_Se) to the Sn site (Sb_Sn) with increasing x. This means the charge state of Sb changes from -3 to +3 with increasing x. We explain the mechanism of the double polarity inversion from the band alignment and the first-principles calculations; pure SnSe shows p-type conduction due to the formation of Sn vacancy, and (Sn_1-xSb_x)Se exhibits the apparent n-type behavior due to conduction through additional Sb_Se impurity band, and then finally re-inverts to weak p-type, where the Fermi level approaches the midgap level located between the additional Sb_Se band and conduction band minimum. The clarification of Sb doping mechanism would give a crucial guide to develop more sophisticated doping route for SnSe and high-performance energy-related devices.
References
[1] L.-D. Zhao et al., Nature 508, 373-377, (2014). [2] T.-R. Wei et al., J. Am. Chem. Soc. 138, 8875-8882 (2016). [3] A. T. Duong et al., Nat. Commun. 7, 13713 (2016). [4] Q. Zhang et al., Adv. Energy Mater. 5, 1500360 (2015). [5] S. Li et al., Adv. Sci.5, 1800598 (2018). [6] J. Umeda, J. Phys. Soc. Jpn. 16, 124-125 (1961). [7] S. Patel et al., J. Appl. Phys. 124, 215103 (2018). [8] X.-L. Shi et al., Adv. Energy Mater. 8, 1800775 (2018).

Thermoelectric power generators that can produce electrical energy from the temperature difference between the two ends are attracting attention as a next-generation clean power generation system because they can be applied to various places where heat sources exist and do not generate by-products during the power generation process. Recently, results of operating application systems such as simple acceleration sensors or electrocardiograms by attaching them to the human body have been published. As such, thermoelectric generators are easy to apply with a small volume of independent power source. It is also receiving attention as a power source for wearable devices that can be monitored at all times.

In this study, a flexible thermoelectric element capable of uniform contact with human arm was fabricated through a printing process-based thermoelectric element manufacturing method that applied a BiTe-based inorganic material and a flexible silver electrode on the sacrificial layer. In particular, by using Polyurethane foam as a polymer filler material to support the structure and lowering the thermal conductivity, it was possible to confirm more than twice the power compared to the case of using Poly(dimethylsiloxane) (PDMS). In addition, on a curved surface with a radius of curvature of 4 cm, the output was 5 times higher than that of a flat metal substrate, and through this, a method for effectively collecting heat from the human body was presented.

In the case of a cooling unit to maintain a temperature difference, a fin-type heatsink that is generally used has a relatively large volume and uses a hard material, so it is difficult to apply it to a flexible thermoelectric device. To solve this, UV curability hydrogel that can absorb water was used as a flexible and stretchable heat sink. Effective cooling properties were obtained from the evaporation of water contained in the hydrogel heatsink, and as a result of testing the performance of the manufactured thermoelectric generator system in the form of a band that can be worn on the human body, the power generation amount of more than 30 μW/cm² under the convection condition of 1.5 m/s was able to confirm. It was confirmed that the capacitor could be charged and the temperature sensor could be driven. As an eco-friendly independent power system technology, it is expected that it can contribute as an energy device for various wearable applications that require continuous measurement and communication, such as an attachable biometric sensor.

Metal halides have been intensively studied for their optoelectronic applications e.g. solar cells and light emitting deivces. Recently, their potential as thermoelectric materials has emerged where unique vibrational properties of metal halides -- such as localised phonons and ultra-low lattice thermal conductivity -- plays important roles. Up to date, the highest thermoelectric figure of merit ZT reported among metal halides is ~0.15, which is well short of the champion ZT of 2.6 achieved in SnSe. Cs₃Cu₂I₅ is an alternative metal halide with an low-dimensional crystal structure where [Cu₂I₅]³⁻ clusters are surrounded and isolated by Cs⁺ cations and there is no direct connectivity between neighboring anionic clusters [1]. Its unique optoelectronic properties related with the low-dimensional structure have been studied, but its charge and heat transport properties have not been indenified yet.

In this study, we assess the thermoelectric potential of Cs₃Cu₂I₅ combining state-of-art first-principles simulation techniques. Thermal transport properties were calculated using anharmonic lattice dynamics [2] where an ultra-low conductivity for the perfect crystal that is found. By employing an ab initio description of carrier scattering [3], we further investigate electron transport properties of the material. Efficient electron transport throughout the crytal is confirmed, which is attributed to an unexpectedly low effective mass for the electron in the low dimensional structure. As a result of imbalanced electron and heat transport, Cs₃Cu₂I₅ is predicted as an ideal thermoelectric semiconductor that exhibits the characteristics of "phonon-glass electron-crystal" with an accessible ZT of greater than 2 [4].

[1] T. Jun, K. Sim, S. Iimura, M. Sasase, H. Kamioka, J. Kim, and H. Hosono, Adv. Mater. 30, 1804547 (2018).
[2] A. Togo, L. Chaput, and I. Tanaka, Phys. Rev. B 91, 094306 (2015)
[3] A. M. Ganose, J. Park, A. Faghaninia, R. Woods-Robinson, K. A. Persson, and A. Jain, Nat. Commun. 12, 2222 (2021)
[4] Y.-K. Jung, I. T. Han, Y. C. Kim, and A. Walsh, npj Comput. Mater. 7, 51 (2021)

Phonon drag thermopower (S_g) has received great attention because total S is enhanced compared to the conventional carrier-diffusion thermopower (S_d) by S = S_d + S_g, and thus the energy conversion efficiency is enhanced [1]. S_g is expressed as S_g = (m^* v_s l_λ ) / (τ_e e T ) in the non-degenerated regime [2]. Therefore, the conventional approach to observe S_g is to utilize ballistic phonon with long mean free paths (l_λ) in high quality semiconductors [3] or large electron effective mass (m^*) in strongly correlated electron systems [4]. However, S_g is observed only in limited materials at low temperatures and its peak value is small. Here, we report an unusually large S_g enhancement in the thin-film heterostructure of a strongly correlated electron oxide LaNiO₃ (LNO).
We investigated the effects of layer thickness and surface termination on S of LNO epitaxial films grown on LaAlO₃ (LAO) substrates with the in-plane lattice mismatch of -1.5%. The 50 – 3 unit cells (u.c.) thick LNO epitaxial films with step & terrace surfaces were grown by pulsed laser deposition on (001) LAO single crystal. LAO-capped LNO films were fabricated by first depositing a 3 u.c. LNO film, followed by growing a 10 u.c. LAO capping layer. The LNO film grows coherently with strain up to thickness 50 u.c. on the LAO substrate. The pseudo-cubic lattice parameters for out-of-plane (c) and in-plane (a) are 3.88 Å and 3.78 Å, respectively. The c/a is 1.026, indicating the in-plane compressive strain in the LNO films. For the LAO-capped 3 u.c. LNO film, the tri-layer structure of LAO/LNO/LAO grows coherently with strain. By reducing the LNO layer thickness, the film showed the metal-insulator transition at a critical thickness of 5 u.c. On the other hand, the surface termination by 10 u.c. LAO capping layer changed the 3 u.c. LNO film more conductive again, which shows a metal-insulator transition at T = 77 K.
The phonon-drag S_g is not observed in the bulk LNO, whereas it appeared at T = 25-33 K for the LNO epitaxial films with compressive strain by the LAO substrate. By reducing LNO layer thickness down to 3 u.c. and subsequent LAO surface termination, S was enhanced 10 times from that of bulk LNO due to the large phonon drag S_g, and the S_g contribution to the total S spreads in much wider temperature range, reaching up to 220 K. We clarified that the S_g enhancement originates from a coupling of lattice vibrational phonons to the Ni 3d-electrons whose effective mass becomes larger in the strained ultra-thin LNO film and also from the strong electron-phonon scattering enhanced by the phonon leakage from the LAO substrate and the capping layer. The present results demonstrate a new approach to manipulate both electronic and phononic properties in atomic-scale controlled transition metal oxide (TMO) heterostructures for high-performance thermoelectrics.

References
1) J. Zhou et al., Proc. Natl. Acad. Sci. USA 112, 14777 (2015).
2) C. Herring, Phys. Rev. 96, 1163 (1954).
3) B. L. Gallagher et al., J. Phys.: Condens. Matter 2, 755 (1990).
4) A. Bentien et al., EPL 80, 17008 (2007).

Enhancement in the thermoelectric (TE) properties through carrier filtering in polycrystalline Sb₂Te₃ thin film is realized by incorporating size-selected Au nanoparticles grown by an integrated gas-phase synthesis method. Crystalline Singlet nanoparticles of Au having a narrow size distribution are embedded between two Sb₂Te₃ layers. Thermoelectric properties have been investigated by varying the concentration of nanoparticles, which reveal a significant increase in the Seebeck coefficient and power factor with a slight deterioration in electrical conductivity. Samples containing non-size selected Au particles with broad size distribution are also synthesized to understand the effect of non-uniformity in size distribution on TE properties and compared with the samples containing size-selected particles. In addition, scanning probe techniques were employed to understand the nature of the interface formed between Sb₂Te₃ and Au nanoparticles which reveals the formation of a Schottky barrier in the interface. The study provides an opportunity to manipulate the TE properties in Sb₂Te₃ thin films by constructing a metal-semiconductor heterostructure by controlling the concentration and randomness to achieve a high TE performance.

Significant interests in enhancing the efficiency of condenser designs and reducing water consumption have emerged for thermoelectric power plants. While substantial efforts have focused on methods to enhance heat transfer performance in condensers, ongoing challenges remain including maintaining robust non-wetting behavior for enhanced heat transfer enhancement during dropwise condensation. In our work, we present a new mode of condensation termed capillary-driven condensation. In this approach, a porous hydrophobic membrane atop a wicking structure on the condensing surface helps generate a capillary pressure gradient to drive liquid transport through the wick to an exit port for condensate removal. The wick both sets the thickness of the condensate-wick layer and increases its effective thermal conductivity, reducing the overall thermal resistance. Meanwhile, the hydrophobic porous membrane helps generate the capillary driving pressure via its microscale pores to transport the condensing fluid out of the wick and significantly increase the heat transfer performance. With well-controlled microfabricated structures, i.e., a silicon nitride membrane suspended on a micropillar wick, we show a successful proof-of-concept. The wick fills completely upon nucleation, and small droplets on top of the membrane are continually absorbed into the wick through the membrane pores. The visualizations suggest promising heat transfer enhancements and are corroborated by our developed models. Simultaneously, we have been pursuing scalable approaches to achieve capillary-driven condenser designs, including diffusion-bonded, hydrophobized hierarchical copper meshes, and electro-spun membrane covered metal wicks. This work suggests a promising path for incorporating capillary-based condenser designs to enhance the efficiency of power plants, thermal management strategies, and water desalination systems, among others.

Despite the widespread occurrence of frost, understanding and controlling condensation-frosting still poses several challenges. Control of condensation-frosting is relevant for a range of industries: In the energy, transportation or telecommunication sectors, frost constitutes serious hazards when it forms on critical components of machines and devices, causing them to fail.

Amongst anti-frosting and anti-icing surfaces, slippery lubricant-impregnated porous surfaces (SLIPS) are discussed due to frosting-retardant and ultra-low ice adhesion properties [1]. However, the understanding of its formation and growth − particularly on SLIPS − remains elusive. The reasons are multi-fold: Insights into frosting are hampered by poor optical contrast between water, frost and lubricant and by the non-equilibrium nature of frost formation, that involves multiple time and length scales.

To reach better insight, we have developed a novel setup that enables the in-situ monitoring of lubricant reorganization, condensation and freezing of water drops using Laser Scanning Confocal Microscopy. Such a setup descriminates between condensation and freezing of water, reorganization of lubricant accompanied by growth of dendrites. In particular, our setup provides spatially and temporally resolved data of the formation and growth of frost, accompanied by quantifiable lubricant reorganization.

As model system, we utilize surfaces with micro- and nanostructures. The coatings are lubricated with silicone oil. Condensation frosting starts with nucleation and growth of water droplets. The condensation and continued growth of droplets on lubricant impregnated surfaces is largely influenced by the structure geometry and wetting property of the lubricated solid surface [2].

Lowering the temperature results in droplet freezing and frost formation. Frosting results in lubricant depletion in the array. Although detrimental at first sight, we note that frost induced lubricant depletion is reversible when the frost is melted in situ. We determined the suction pressure induced by frost dendrites, which wicks lubricant into the frost structure [3].

Formation of frost dendrites result in a particularly harsh depletion mechanism. Notably lubricant depletion can be prevented. Therefore, the interstitial spacing of the lubricant impregnated surface needs to be kept lower than that found in frost dendrites (≥ 100 nm). To prove this concept, we used densely packed nanoparticles. The nano-interstices are kept between 10 to 50 nm. The capillary pressure keeping lubricants in the nano-interstices exceeds the capillary suction pressure by frost dendrites. We term this the design concept “capillary-balancing”. Notably, the facile and scalable coating shows passive anti-icing [4].

[1] Kreder, M.J., Alvarenga, J., Kim, P. & Aizenberg, J. Design of anti-icing surfaces: smooth, textured or slippery? Nat. Rev. Mater. 1, 15003 (2016).
[2] T. Kajiya, F. Schellenberger, P. Papadopoulos, D. Vollmer and H.-J. Butt: 3D imaging of water-drop condensation on hydrophobic and hydrophilic lubricant-impregnated surfaces, Scientific Reports, 2016, 6, 23687.
[3] L. Hauer, W.S.Y. Wong, V. Donadei, K.I. Hegner, L. Kondic, D. Vollmer: How frost forms and grows on lubricated micro- and nanostructured surfaces, ACS Nano, 2021, 15, 4658.
[4] W.S.Y. Wong, K.I. Hegner, V. Donadei, L. Hauer, A. Naga and D. Vollmer: Capillary balancing: designing frost-resistant lubricant-infused surfaces, Nano Lett, 2020, 20, 8508.

Boiling is an essential process in a variety of applications such as power plants, water purification, and steam generation. The boiling performance can be described by two major parameters: heat transfer coefficient (HTC) and critical heat flux (CHF), which dictate the efficiency and operational heat flux limit during the nucleate boiling, respectively. The previous study has shown that simultaneous enhancements of HTC and CHF values can be achieved with hemi-wicking surfaces consist of micropillar and microtube structures. There has been a trade-off, however, between the two parameters associated with the competition between nucleation site density and complete utilization of capillary wicking. In this work, we mitigate the trade-off by incorporating nanostructures on top of microtube and micropillar structures and demonstrate the extreme pool boiling performance for both HTC and CHF. Specifically, we controlled vapor nucleation for HTC enhancement with surface structures in millimeter, micrometer, and nanometer scales by the separation of microtube clusters, microtube cavities, and nanostructures, respectively. At the same time, the augmented capillary wicking on hierarchical structures achieved significant CHF enhancement. The visualization of bubble dynamics with a high-speed camera supports our findings that HTC and CHF enhancements can be achieved by multi-scale control of vapor nucleation on hemi-wicking surfaces. The extreme boiling performance achieved in this work is not only promising for various boiling applications but also providing important implications for fundamental aspects of boiling heat transfer mechanisms.

Fast and programmable transport of liquid droplets on a solid substrate is desirable in microfluidic devices, condensation in power plants, microchips cooling, and even water collection in the space. Past research has focused on designing substrates with asymmetric structures or surface modifications where droplets behaviors are passively controlled, or by applying external stimuli such as electrowetting where patterning of intricate electrodes is required. In this work, we demonstrate tunable and programmable droplet motion on a slippery liquid-infused porous surface (SLIPS) using light and photo-responsive surfactants. When illuminated by the light of appropriate wavelengths, these surfactants can reversibly change interfacial tensions in a multi-phase fluid system, which generates Marangoni flow that drives droplet motions. Experimental results demonstrate that a 70 μL static droplet can move 9 mm in one minute upon irradiation of a UV laser and the merging time of two droplets can be manipulated by adjusting light projection positions. We also model the photo-Marangoni flow inside the droplet and calculate the net driving force. The method demonstrated in this study serves as a simple and exciting new approach for dynamic manipulation of droplets for microfluidic, thermal and water harvesting devices.

Heterogeneous condensation on soft substrates benefits from higher nucleation density. However, compared to rigid substrates, condensate droplets on such surfaces suffer from viscoelastic dissipation at the three-phase contact line as they slide, inhibiting their removal. These droplets manifest as thermal resistance to condensation heat transfer between the cooled substrate and the ambient vapour. The presence of these pinned droplets also limits the available area for fresh nucleation sites as well, thus lowering overall heat transfer efficiency.

In this work, we explore the effects of infusing a lubricant into the bulk of polydimethylsiloxane (PDMS, Sylgard 184) on the material properties relevant to wetting phenomena. By varying the proportion of the base, curing agent and the lubricant, we identify the resulting amount of uncrosslinked chains by a swelling-and-deswelling process and their molecular distribution with gel permeation chromatography (GPC). Elasticity of the different substrates is measured with micro-indentation tests which we correlate with measured wetting ridge heights at the contact line. Viscoelastic dissipation at the contact line is characterised by measuring droplet sliding velocities. Cloaking of the water droplets is studied with laser scanning confocal microscopy followed by a transient measurement of the resulting change in droplet surface tension. We conclude the study with microscale observation of condensation for nucleation densities and droplet growth rate, and a dew collection experiment for overall performance improvement in condensate removal. We correlate the measured condensation characteristics with the material properties of the substrate.

We find that 5 wt. % of bulk lubricant infusion, despite resulting in a substrate of lower bulk elasticity, significantly reduces viscoelastic dissipation at the contact line as indicated by an increase in droplet sliding velocity. Additionally, the nucleation density is more than doubled compared to PDMS substrates without bulk lubricant infusion. We also observe that further increase above 5 wt. % bulk lubricant concentration has an insignificant effect on the condensation dynamics over the substrate. Moreover, there is a small reduction in droplet growth rate for lubricated substrates. This reduction can be attributed to the cloaking of droplets by uncrosslinked chains due to the presence of a lubricant layer on the substrate top surface. Overall, these factors combine to result in an increase of ~40% in dew water collection for lubricated substrates compared to non-lubricated ones.

The temperature dependence of experimental mobility is commonly used as a predictor of the dominant scattering mechanism in thermoelectric materials. Knowledge of the dominant scattering mechanism is employed to fit models of charge transport including deformation potentials and effective masses, and to obtain estimates of the optimal doping concentration and temperatures that maximise thermoelectric performance. However, if the scattering mechanism is determined incorrectly, this can lead to wildly inaccurate predictions of materials properties and optimal conditions, frustrating efforts to optimise devices. By reviewing the temperature dependence of 24,000 mobility calculations, I demonstrate that the temperature-dependence of mobility is not a reliable indicator of the dominant scattering mechanism. Furthermore, I reveal that many materials long considered to be dominated by deformation-potential scattering are instead controlled by polar optical phonons. I conclude with the pitfalls of predicting the major scattering type based on the experimental mobility trend alone.

Thermoelectric materials enable direct conversion of waste heat into electrical energy. The conversion efficiency is inversely proportional to the thermal conductivity, which is generally dominated by phonons in semiconductors. Zintl compounds AMg₂X₂ (A=Mg,Ca,Yb; X=Sb,Bi) constitute a class of new thermoelectric compounds with excellent thermoelectric performance in n-type Mg₃(Sb,Bi)₂ alloys, with zT values up to 1.6 reported so far. Mg₃Sb₂ exhibits very low lattice thermal conductivity (~1-1.5 W/m/K at 300K), comparable with PbTe and Bi₂Te₃ with only half of their mass densities. Meanwhile, contrary to common mass-trend expectations, replacing Mg with heavier Ca or Yb yields a threefold increase in k_L in CaMg₂X₂ or YbMg₂X₂. We report on neutron scattering and first-principles studies of the lattice dynamics of AMg₂X₂. Inelastic neutron scattering measurements founds a large phonon softening and flattening of low-energy transverse acoustic phonons in Mg₃Sb₂ and Mg₃Bi₂ compared to CaMg₂X₂ or YbMg₂X₂. Combined with simulations, we highlight the importance of a specific soft Mg-X chemical bond that suppresses phonon group velocities and drastically enlarges the scattering phase-space, enabling the threefold suppression in thermal conductivity. These results provide key insights for manipulating phonon scattering without the traditional reliance on heavy elements and disorder.

Reference: Ding, J., Lanigan-Atkins, T., Calderón-Cueva, M., Banerjee, A., Abernathy, D. L., Said, A., Zevalkink, A., Delaire, O. (2021). Soft anharmonic phonons and ultralow thermal conductivity in Mg3 (Sb, Bi) 2 thermoelectrics. Science Advances, 7(21), eabg1449.

Imperfections such as stacking faults and interfaces are recognized as critical factors in modifying thermal properties and heat transport in crystal materials by scattering phonon and changing vibrational structure. They play an even more decisive role when the material’s dimension is reduced to the nanoscale. Although diverse modelling tools have been deployed to understand the interaction between heat and imperfections in solids, the effect of crystal defects on thermal conductivity is being treated without considering the local change of phonon dispersion relation. In contrast, some studies have predicted the existence of extra vibrational modes due to dislocations or interfaces [1, 2]. However, the experimental evidence of those local modes is fundamentally elusive from either thermal conductivity measurements or conventional vibrational spectroscopies such as Raman and inelastic neutron scattering spectroscopies due to their inefficient spatial resolution [3]. Recently, the energy resolution of monochromated electron energy-loss spectroscopy (EELS) has been dramatically improved to a few millielectronvolts, which opens a new era of probing nanoscale feature of vibrational modes in materials using the aberration-corrected scanning transmission electron microscope (STEM) [3]. By balancing spatial, energy, and momentum resolutions, we can obtain a powerful space- and angle-resolved vibrational EELS method [4]. Here we show three cases to demonstrate how to utilize this emerging tool to reveal the local vibrational modes at (1) a stacking fault in cubic silicon carbide, (2) an interface in a Si–Ge heterojunction, and (3) an interface in a two-dimensional (2D) MoS₂-WSe₂ lateral heterojunction.

Stacking faults are common imperfections in SiC, and remarkably affect the bulk thermal conductivity. We acquired the angle-resolved vibrational spectra from both the stacking-fault-containing region and defect-free region. At the defect, the acoustic phonons at X point of the first Brillouin zone undergo an energy red shift of 3.8 meV and major intensity modulations. These striking features arise from the symmetry breaking and variation of interatomic force constants and are assigned to the local defect phonon modes. Both the reduced phonon energy and flattened dispersion curves of defect phonon modes help understand the defect-induced reduction of thermal conductivity [4]. In the second case, we employed a similar strategy to probe the interface phonon modes at a Si-Ge epitaxial interface, which is believed to facilitate the heat transport across interface [2]. The interfacial vibrational EEL spectrum contains extra vibrational signals at 48 meV (11.6 THz), which are absent in the phonon structure of either Si or Ge [5]. The exotic vibrational feature is also unveiled at the interface of monolayer MoS₂-WSe₂ heterojunctions [6]. Our work opens the door to investigating phonon propagation around crystal defects and provides guidance to the engineering of desired heat management for semiconductors and power electronic devices [7].

References:
[1] M. D. Li et al. Nano Lett. 17 (2017), p. 1587−1594.
[2] Y. Chalopin and S. Volz, Appl. Phys. Lett. 103 (2013), 051602.
[3] O. L. Krivanek et al., Nature 514 (2014), p. 209−212.
[4] X. X. Yan et al., Nature 589, (2021), p. 65–69.
[5] Z. Cheng, R. Y. Li, X. X. Yan et al., arXiv:2105.14415 (2021).
[6] X. Z. Tian, X. X. Yan, G. Varnavides et al., arXiv:2104.08978 (2021).
[7] This work was supported by the Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering (DE-SC0014430). The authors acknowledge the use of facilities and instrumentation at the UC Irvine Materials Research Institute (IMRI) supported in part by the National Science Foundation through the Materials Research Science and Engineering Center program (DMR-2011967).

Alternative concepts of electric mobility, miniaturization in the electronics industry and the quest for environmentally friendly air-conditioning and refrigeration technology have opened many doors for advancements in the area of new materials for thermal management. The electrocaloric effect can be an alternative technology in this regard. The electrocaloric effect is a phenomenon observed in dielectric materials, as they undergo an adiabatic temperature change or an isothermal entropy change upon the application/removal of an electric field. Ferroelectric relaxor polymers are attractive, as they show slim hysteresis loops close to room temperature, which is favorable for the electrocaloric performance.
In the present work, we measure the electrocaloric temperature change in Polyvinylidene fluoride – trifluoroethylene – chlorofluoroethylene P(VDF-TrFE-CFE). The solution casting method was employed to prepare free-standing polymer films of P(VDF-TrFE-CFE). The free-standing polymer films are about 20 µm thick. The sputter coating technique was used to deposit silver electrodes onto the free-standing film. The dielectric permittivity curves measured against temperature show typical broad frequency dependent maxima which is typical for relaxors. The P – E hysteresis loops were measured as a function of temperature. A slim hysteresis loop is observed confirming the relaxor behavior of P(VDF-TrFE-CFE). The direct electrocaloric effect was measured as a function of temperature and electric field using a custom-designed quasi-adiabatic calorimeter. A maximum electrocaloric temperature change of ~ 2 K is observed in 20 µm thick P(VDF-TrFE-CFE) films.

Over two-third of the world’s population will lack sufficient access to fresh water by 2025. As water scarcity increases, desalination plants are on the rise as over 98% of the earth’s water is brackish and sea. The current large-scale desalination plants mainly utilize thermal evaporation or reverse osmosis to increasing the supply of fresh water. However, these techniques are energy-intensive and require large infrastructures. Along this line of thought, solar desalination systems could be great candidate choice as a reliable device. Despite the great attempt to increase the evaporation efficiency of the solar desalination devices, the condensation part of the system is not yet optimized which limits the overall solar-to-water conversion efficiency. Here, we are presenting a passive, reliable, and highly efficient solar desalination system that does not require any energy consumption except one sun power. Our device consists of two parallel puros membranes which are mounted closely together that make our device efficient due to the minimal vapor diffusion resistance. Unlike other methods that use capillary force to transport the sea water, we are using hydrostatic pressure to flow the sea water into our device which is subsequently getting preheated using heat recycling. This study provides insights into improving the overall efficiency of solar desalination devices. Using scaling analysis, the evaporation and condensation flux has been modelled.

Vapor condensation is prevalent in various industrial and heat/mass transfer applications, such as power generation and conversion, water harvesting/desalination, and electronics thermal management. Dropwise condensation has been demonstrated to have better heat transfer performance due to faster condensate removal (i.e., droplets shedding), as compared to the traditional filmwise mode, which occurs on typical metal oxide condenser surfaces. The key to enhanced condensation is thin (< 100 nm) coatings with low contact angle hysteresis. Thin low surface energy coatings (e.g., self-assembled monolayers or SAMs) have been utilized to promote dropwise condensation especially on copper surfaces for over half a century. However, all thin coatings (polymer or SAM) degrade within a few hours during condensation of water vapor, and specifically trichlorosilane SAMs fail within 30 mins on metal surfaces. After coating failure/degradation, water vapor condensation transitions from the dropwise mode to the filmwise mode. In this work, we significantly enhance the durability of silane SAMs on copper surfaces via plasma modification of the substrate, and the preferential dropwise condensation mode is maintained for hundreds of hours when condensing water vapor. Low surface energy silane SAM coatings deposited on our plasma modified copper substrates demonstrate a low contact angle hysteresis (≈20°), and better bonding between the silane SAM and the copper substrate is achieved for substrates at different roughness levels. Additionally, we elucidated the degradation mechanism of silane SAMs on copper during water vapor condensation and revealed the required surface chemistry to develop robust silane SAM coatings on copper surfaces. Various nano-to-macroscale surface analysis techniques are utilized to characterize the coating, surface and interface properties. This work sheds light on the degradation mechanisms of low surface energy monolayer coatings on copper during water vapor condensation, and provides durable solutions for industrial applications with enhanced heat transfer performance.

The exploitation of electrocaloric effect (ECE) is an emerging field of research because it provides an energy-efficient and environment–friendly alternative to conventional cooling devices for a broad range of applications, such as on-chip and temperature regulation for sensors and electronic devices. Although ECE has been observed in a variety of ferroelectric ceramics, crystal and polymer and performance were not satisfactory. Recently, a ferroelectric thin film based on P (VDF-TrFE) has been demonstrated a high EC effect. Compared to ferroelectric ceramics, P (VDF-TrFE) has the advantage of low density, good toughness, high electrostriction rate, and environmental protection too. Currently, a number of the existing methods have been developed to characterize the electrocaloric effect in the Direct and Indirect methods. Direct methods, measure the change in EC temperature or heat exchanged in the material in the presence or removal of the electric field. However, the indirect method is based on Maxwell’s equation that allows obtaining the variation of entropy and variation of temperature. Detect the temperature variation in the thin film is difficult due to the rapid dissipation of heat in the substrate. In addition mass of the assembly (substrate +measuring device) is greater than that of the thin layer alone, and makes the heat variation is undetectable by direct and indirect method. Therefore, we need to study this material using instruments offering high resolution. Thermoreflectance is a photothermal technique whose high frequency of operation makes it possible to probe the material on small thicknesses. This method has a function based on temperature dependant reflectivity of a surface which is probed using the intensity variation of a reflected laser beam.
In this work, blend film of copolymer P (VDF-TrFE) and relaxor Ter polymer P (VDF-TrFE-CFE) in a different ratio. After optimization, the copolymer crystallinity was improved due to the coupling that was observed from the characterization. This optimized crystalline thin film was further characterized by recently developed direct measurement via thermoreflectance where the pump laser beam was replaced by an alternating electric field that is applied across our electrocaloric material to induce the electrocaloric effect (ECE). To apply this method, variation of deltaT as a function of applied field and frequency was studied in optimized blend film and achieved the delta T(K) >2 at room temperature.

Conjugated polymers need to be doped to increase charge carrier density and reach the electrical conductivity necessary for electronic and energy applications. While doping increases carrier density, Coulomb interactions between the dopant molecules and the localized carriers are poorly screened in organic semiconductors because of their low dielectric permittivity and thus causes an increase in energetic disorder as evident in the broadening and a heavy tail effect in the electronic density-of-states (DOS). We relate the dopant-induced energetic disorder to a reduction in the Seebeck coefficient while deep traps in the heavy tail cause a collapse in the conductivity. We show that increasing the dielectric permittivity (ε) of the polymer mitigates dopant-carrier Coulomb interactions resulting in a simultaneous increase in conductivity and Seebeck coefficient. We simulated electron hopping between localized sites with a modified Gaussian disorder model and included electrostatic interactions between carriers and clustered dopants. We iteratively solved the non-linear Pauli’s master equation to compute time-averaged occupational probabilities of the sites from which relevant transport quantities are calculated. Increasing ε from 3 to 12 nearly restores the intrinsic DOS, resulting in a large increase in the power factor. We validated this prediction experimentally in iodine-doped P3HT and P3HT blended with barium titanate nanoparticles. The addition of 2% w/w barium titanate nanoparticles increased conductivity and Seebeck across a broad range of doping, resulting in a four-fold increase in power factor. Beyond improving the thermoelectric performance, we note that most of the improvement we observed in the power factor comes from increases in the conductivity, particularly at low to medium doping concentrations, which is broadly useful in organic electronics. Our method of incorporating additives with a high dielectric constant precludes the need for synthetic modifications and thus can be applied to a wide range of polymers. The vapor doping method followed in our study avoids interfacial effects on the DOS that can arise from modulation doping using field-effect transistors and also wide angle X-ray scattering (WAXS) studies show that the overall film morphology is essentially maintained during the measurement over a wide range of carrier concentration. This synergetic computational and experimental study shows a clear path forward to mitigate dopant-induced energetic disorder and develop high-performance organic electronic materials with improved charge transport through dielectric screening.

For decades, ZrW₂O₈ has been known as a negative coefficient of thermal expansion (CTE) material. The negative CTE is probably due to its unique crystal structure. ZrW₂O₈ has been therefore considered as a promising candidate for thermal management applications, which can be applied to use as a CTE compensating component for controlling thermal expansion of various composites such as ceramics, polymers, and metals. On the other hand, thermal properties other than the CTE of ZrW₂O₈, such as thermal conductivity, are almost unknown. Here, we synthesized a highly dense bulk sample (%T.D. > 95%) of ZrW₂O₈ from a very fine powder (~ 31 nm) using a spark plasma sintering (SPS) technique. It was revealed that ZrW₂O₈ exhibited the lowest level of thermal conductivity among various oxides in a wide temperature range from 300 to 823 K. Furthermore, we investigated various physical properties related to such low thermal conductivity through the anharmonicity such as peculiar CTE, elastic constant, and Grüneisen parameter. Owing to its extremely low thermal conductivity, ZrW₂O₈ can be used as a thermal barrier material, which generally operates at high temperatures.

Thermal interface materials (TIMs) with high thermal conductivity, surface wettability, and fluidic characteristics are crucial for the thermal management of high-power electronics. Gallium-based liquid metal alloys (LMAs) have been investigated as TIM due to their relatively high thermal conductivity (~23 Wm^-1K^-1), low melting point (<10°C), and low processing temperature. To further increase the thermal conductivity of LMA, a few studies tried to incorporate metallic fillers into the liquid metal matrix. However, these attempts have been limited due to the oxidation of the LMA matrix, which leads to severe solidification of the fabricated composite at a high volume fraction of fillers. In this work, we incorporate copper nanoparticles (Cu NPs) into the eutectic gallium-indium-tin (GaInSn) LMA using an oxide-free ultrasonication-assisted particle internalization method. Developed GaInSn/Cu NPs composite provides over 180% enhancement of thermal conductivity (~65 Wm^-1K^-1) with only 4% volume fraction of fillers. The obtained thermal conductivity is higher than those of the previously-reported Ga-based composites at over 20% volume fraction of fillers. In the liquid metal droplet impacting test, the GaInSn/Cu NPs shows almost identical spreading diameter compared to the untreated GaInSn, which represents the maintained fluidity of the composite. Experimentally measured thermal property is demonstrated based on the nanoparticle clustering phenomenon, in which we microscopically visualize the formation of the copper clusters within the liquid metal matrix to calculate the prediction model. The calculated effective thermal conductivity is well-matched to the experimental data, indicating that the nanoparticle clusters provide additional heat pathways. In terms of surface wetting characteristics, the LMA shows a high-quality interface with silicon substrate due to the formation of ~10nm of Ga₂O₃ adhesion layer confirmed through X-ray photoelectron spectroscopy (XPS). In the heat dissipation performance test, GaInSn/Cu NPs presents over 20% lower hot-spot temperature compared to the one obtained with the thermal grease at high (>150 Wcm^-2) heat flux regime and also provides thermal stability over 230 cycles of acceleration test at high-temperature exposure (~100°C), suggesting applicability for high-power density semiconductor packaging applications.

Single crystalline pyroelectric materials offers new opportunities for efficient thermal energy harvesting utilizing temperature fluctuation. Recent works have demonstrated the pyroelectric generators utilizing various thermal energy such as body heat and latent heat of fluids. However, the sustainable and high performance pyroelectric thermal harvesting has not been realized yet. In this work, we introduce the pyroelectric generator utilizing the rapid temperature fluctuation during a droplet impact. The thin single-crystalline PMN-PT of 100 μm thickness was incorporated to the generator in order to improve the pyroelectric coefficient with reducing the thermal mass. To further enhance the temperature fluctuation during the drop impact, the top surface of the pyroelectric generator was coated by the silanized titanium oxide nanoparticles. Then a 14 μl water drop of 293K was impacted on the developed pyroelectric generator by varying the surface temperature from 313K to 353K. Based on the measured pyroelectric current and voltage values, the pyroelectric current prediction model was developed utilizing the correlation of the maximum spreading diameter during drop impact. The developed model accurately predicted the pyroelectric electricity generation under a wide range of experimental conditions. The maximum pyroelectric current and power density were measured to be 66 μA and 19.2 μW/cm², which surpasses previous pyroelectric thermal harvesting results. We believe this work will help to develop efficient pyroelectric thermal energy harvesters.

Erythritol (ET) has received increasing research attention for medium-temperature phase change solar-thermal energy storage. However, ET suffers from serious supercooling, low thermal conductivity, poor solar absorption, and leakage after melting, which lead to low solar-thermal energy harvesting and releasing efficiency. Herein, we simultaneously overcome these inherent shortcomings by impregnating ET within the surface-roughened hydrophilic copper foam, which is prepared by subsequent oxidization of commercial copper foam and thermal reduction under hydrogen and using stably dispersed crumpled graphene particles as the solar-thermal converter. Such oxidization-reduction-treated copper foam not only provides numerous heterogeneous nucleation sites and lowers the nucleation energy barrier for forming ET crystals during the solidification process, but also facilitates rapid charging and discharging along with three-dimensional heat conductive networks, and confines the melted ET. Compared with neat ET, the copper foam-templated composites have reduced the supercooling degree by ~60 °C and improved thermal conductivity by more than 5 times, which in turn lead to a large latent-heat releasing percentage of 85.8% and a high heat-extraction efficiency of 94.2%. We demonstrate that such composites can be used for high-efficiency, medium-temperature, form-stable, direct solar-thermal energy harvesting, and the harvested heat can be readily converted into electricity to power small devices.

Boiling heat transfer is a promising approach for an ever-growing need for higher cooling capacities. To heighten the operating limit of boiling heat transfer, critical heat flux (CHF) should be enhanced. CHF is the maximum heat flux of fully developed nucleate boiling, just before vapor film covers heated surfaces, which results in an abrupt increase of wall temperature. To delay CHF, supplying sufficient working fluid to the heated surface should be ensured, thus research into CHF enhancement had explored various surface modification techniques including nanoparticle coated surface. In general, nanoparticle coatings are known to enhance CHF and heat transfer coefficient (HTC) by efficiently dissipate heat into the working fluid and promoting bubble nucleation by providing porous nucleation sites. And graphene materials could be fitting candidates since they exhibit superb thermal conductivity and graphene platelet could form porous layers on the substrates which promote bubble nucleation. However, gathering an excessive amount of graphene platelet onto the surface could discourage liquid supply due to its hydrophobic nature. This overcrowding graphene problem could be modulated by combining graphene deposition with an appropriately micro-structured surface since the particle deposition takes place at the microlayer region. Here, we propose reduced graphene oxide (rGO)-coated micropillar structure with varying aspect ratios to promote nucleate boiling and save liquid supply paths. The rGO-coated micropillar surfaces were prepared by deep reactive ion etching (DRIE) and following nanofluid boiling step with rGO solution of 0.0005 wt%. Micropillar structures of different aspect ratios (m5, m10, m20) fabricated by DRIE are 4 μm in diameter and 5, 10, 20 μm in height, respectively. Fabricated surfaces were evaluated by a pool boiling setup with deionized(DI) water as a working fluid. The temperature of DI water was maintained at the saturated temperature at atmospheric pressure. The results showed that rGO coated surface shows smoother and faster ONB by providing suitable cavities which promote bubble nucleation for lower wall superheat. And wall superheat was lessened for all rGO-coated cases owing to the high thermal conductivity of rGO particles. For the morphologies of the rGO coating, porous rGO coating embraced the lateral side of micropillar when the aspect ratios of the micropillar arrays were relatively lower (m5, m10), which might result in blockage of water supply, while the rGO layer coated over the tip of the micropillar when the aspect ratio of the micropillar was sufficiently high (m20). This difference came from the location of the microlayer varies with the aspect ratio of the micropillar arrays during the fully developed nucleate boiling regime. Presented surfaces of plain, rGO-coated m5, m10, and m20 recorded 89 W/cm², 174 W/cm², 177 W/cm², and 256 W/cm² in CHF, respectively. The maximum heat transfer coefficients were recorded as 20.4 kW/m²K, 50.9 kW/m²K, 45.9 kW/m²K, and 74.3 kW/m²K, respectively. The significant improvement on CHF and HTC in the rGO-coated m20 case indicates that rGO coatings on the m20 were mainly placed over the tip, saving spaces between micropillars and emphasizes the importance of ensuring flow paths to enhance the nucleate boiling. This study will be helpful for enhancing boiling heat transfer for the usage in the field needing extreme cooling capacities such as data storage center, heat exchanger, and fusion reactor.

Comprehensive understanding on radiation resistance of nuclear fuels and cladding materials during normal reactor operation is very important for the construction of a new generation of safe and energy-efficient nuclear reactors. Particularly, the control of radiation-induced heat transport is a critical issue for nuclear reactor system design and validation.
Swift heavy ion (SHI) irradiation simulates nuclear fission products in terms of the structural damage caused to materials in the actual nuclear reactors [1]. At high energies, typical of SHIs, where the ion penetration depth is of the order of several microns, the projectile energy is deposited in the target through an inelastic process through which electronic energy loss dominates on nanoscale depths, leading to intense localized thermal spikes causing displacement of lattice atoms along nanoscale-thick ion tracks along the ion path in insulating materials [2].
It is well known that nitride ceramics demonstrate a high level of resistance to SHI radiation and no visible tracks have been observed in most ceramics except silicon nitride. This material is considered to be a promising material as inert matrix fuel host and therefore it is vital to understand the parameters of latent track formation and structural change. Several works have shown that the doping of materials and thermal conductivity play an important role in the formation of these tracks [3,4]. Therefore, the main goal of this study is to determine the impact of Al doping on thermal conductivity degradation due to the related nano-structural damage.
We used polycrystalline β-Si₃N₄ ceramic samples with different level of Al doped concentration ranging from 0.05 to 3%. The SHI irradiation was performed at the IC-100 FLNR JINR cyclotron facility in Dubna, Russia where the samples were irradiated by Bi, Xe, Kr and Ar ions at room temperature and fluences ranging from 10¹² to 10¹⁴ ions/cm². For structural analysis the samples were analyzed in both TEM and STEM modes and complimented with diffraction analysis using a JEOL ARM-200F TEM operating at 200 kV. Cross-plane thermal conductivity measurements were done using picosecond time domain thermoreflectance (TDTR) technique at different modulation frequencies to enable thermal probing at nanoscale depths. Both TEM analysis and thermal conductivity measurements revealed that the high concentration of Al doping lowers the resistance of β-Si₃N₄ to ion track formation resulting in amorphization and substantial decay in thermal conductivity. Depth-resolved TDTR measurements were used to validate models based on semi-analytical Klemens phonon transport and non-equilibrium molecular dynamics to analyze SHI-induced nanoscale damage.

References
[1] G.S. Was, Challenges to the use of ion irradiation for emulating reactor irradiation, J. Mater. Res. 30 (2015). https://doi.org/10.1557/jmr.2015.73.
[2] V.A. Skuratov, J. O’Connell, N.S. Kirilkin, J. Neethling, On the threshold of damage formation in aluminum oxide via electronic excitations, Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms. 326 (2014). https://doi.org/10.1016/j.nimb.2013.10.037.
[3] A. Janse Van Vuuren, V. Skuratov, A. Ibrayeva, M. Zdorovets, Microstructural effects of Al doping on Si3N4 irradiated with swift heavy ions, in: Acta Phys. Pol. A, 2019. https://doi.org/10.12693/APhysPolA.136.241.
[4] M. Toulemonde, W. Assmann, C. Dufour, A. Meftah, C. Trautmann, Nanometric transformation of the matter by short and intense electronic excitation: Experimental data versus inelastic thermal spike model, Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms. 277 (2012) 28–39. https://doi.org/10.1016/j.nimb.2011.12.045.

The progress of human civilization is highly dependent on the capability to harness and deploy energy and water. On one hand, owing to carrying plenty of kinetic and electrostatic energies, water is at the heart of many energy conversion and harvesting devices. On the other hand, the large latent heat of water also endows it as a ubiquitous flowing currency in energy transfer (heat transfer) devices. The close intertwining among water, energy, and heat becomes pronounced with the penetration of miniaturized devices in every corner of our society. Despite inherent coupling and diversity in varying energy processes, there lies in a common need, i.e., achieving high energy efficiency while low water consumption.
This simple need, however, encounters several challenges. First, the miniaturization of electrical devices driven by the Moore’s Law poses increased heat flux and difficulty in energy supply, especially for implantable application. Second, scientifically, harnessing the synergy between water, energy and heat to improve energy efficiency calls for the heterogeneous integration of different materials, interfaces, and systems, which poses the daunting complexities in formulating a basic framework, rational design, and scalable fabrication. Third, technically, recent breakthroughs have demonstrated the power of designing surfaces to mediate water transport for efficient thermal cooling and energy harvesting for sustained power supply, however, conventional approaches for surface design are handled separately, which are complicated and inefficient. Without proper design, surfaces are also susceptible to structural and functional instability and compromise overall performances.

Leveraging the dual nature of both water and surfaces as well as drawing inspirations from nature’s power in managing the flow of water, energy, heat through the elegant control of surface properties, in this talk, I will discuss an innovative surface engineering-based approach for transformative, hybrid energy management and exploitation (THEME) that results in synergistic effects and novel functionalities that go beyond conventional designs. Examples include how to efficiently promote phase change heat transfer such as condensation, boiling, in particular cooling down large-scale hot surfaces by several hundreds of degrees, how to achieve energy-free cooling and how to efficiently harvest water energy from moisture, encapsulated water and flowing water.

Nanostructures with carefully tailored properties can be used to manipulate the flow of light, heat and water, to enable high performing passive cooling through evaporation and radiation. I will present our recent progress.

The first example is cooling photovoltaic device through evaporative cooling. We demonstrate a monolithic tandem solar electricity-water generator that synergistically produces electricity and clean water by utilizing the full spectrum of solar irradiance. This design posseses two components: a top infrared-transparent photovoltaic device using above-band-gap photons, and a bottom solar water purifier using the below-band-gap photons. A well-designed water-proof thermal interconnecting layer (WTIL) makes these two components work synergistically: the bottom purifier serves as an evaporative cooler of the top solar cell to increase its efficiency, whereas the thermalization energy of the top cell is reutilized by the bottom purifier for producing more clean water. We experimentally demonstrate that a prototype hybrid tandem solar device with WTIL can generate electricity with a power output of 204 W m-2 and purify water at a rate of 0.80 kg m-2 h-1 under 1-sun illumination.

The second example is about passive cooling. Radiative cooling which sends heat to space through atmospheric transparency window without any energy consumption, is attracting significant attention. For radiative cooling to achieve high cooling performance, it is ideal to have a selective emitter, with an emissivity dominant in the atmospheric transparency window. However, so far scalable production of radiative cooling materials with selective emissivity has not been realized. Here we will present a hierarchical design for a selective thermal emitter to achieve high performing all-day radiative cooling. Moreover, it is revealed that this hierarchically designed selective thermal emitter shows significant advantage if being applied to alleviate Global Warming or to regulate temperature of the Earth-like planet.

Harvesting natural energy and water resources from the earth and beyond is essential for the mankind. Solar energy-water nexus is of paramount importance to achieve sustainable development goals of our world, where effective transport at the light-matter and fluid-solid interfaces is the key. In this talk, I will present recent advances in enhancing interfacial transport by tailoring optical and wetting properties of materials and structures via micro-3D printing and nanofabrication. A new class of spectrally selective nanocomposite solar absorbers is proposed to harness sunlight and minimize thermal emission. Functionalized micro and nanoporous membranes are applied to enhance liquid propagation and vapor generation, meanwhile new physical insights are obtained from infrared imaging and magnetic resonance analyzer. Latest development of micro-3D printing technologies enables the digital fabrication of biomimetic structures for high-performance transport devices in various energy and water applications. These innovations pave the way for accelerating the knowledge discovery of light-heat-fluid transport and developing cost-effective solutions to real-world grand challenges.

Two-thirds of humanity live under conditions where the net fresh water withdrawal is more than twice the natural water availability for at least 1 month per year, and half a billion people suffer from this water stress throughout the entire year. Atmospheric water vapor is ubiquitous and represents a promising alternative to address global clean water scarcity. Sustainably harvesting this resource requires energy neutrality, continuous production, and facility of use. In this talk I will discuss promising approaches from my research group addressing the above challenges, Refs. (1-3), involving the development of innovative materials and systems. These utilize radiative cooling to the outer space with optimized radiation shielding (1), or sorption intermittency guided by the time scales of the designed materials, combined with solar energy (2), to harvest atmospheric water efficiently and continuously (over 24 hours), beyond the capabilities of the state-of-the-art solar energy driven water harvesting systems. Water removal from such systems without expending additional energy is also addressed. This is enabled with a fully passive superhydrophobic condensate harvester, working with a coalescence-induced water removal mechanism from a novel sprayable superhydrophobic coating (3).
1. I. Haechler, H. Park, G. Schnoering, T. Gulich, M. Rohner, A. Tripathy, A. Milionis, T. M. Schutzius and D.Poulikakos, Exploiting radiative cooling for uninterrupted 24-hourwater harvesting from the atmosphere, Science Advances, in press.
2. H. Park, I. Haechler, G. Schnoering, M. D. Ponte, T. M. Schutzius, D. Poulikakos, Enhanced atmospheric water harvesting with sunlight activated sorption ratcheting, in review.
3. Matteo Donati et al., Advanced Materials Interfaces, vol. 8: no. 1, pp. 2001176 2020. DOI: 10.1002/admi.202001176.