Computer modeling of thermal transport at the nanoscale

Resume : Many physical properties of metals can be understood in terms of the free electron model, as proven by the Wiedemann-Franz law. According to this model, electronic thermal conductivity can be inferred from the Boltzmann transport equation (BTE). However, the BTE does not perform well for some complex metals, such as Cu. Moreover, the BTE cannot clearly describe the origin of the thermal energy carried by electrons or how this energy is transported in metals. The charge distribution of conduction electrons in metals is known to reflect the electrostatic potential of the ion cores. Based on this premise, we develop a methodology for evaluating electronic thermal conductivity of metals by combining the free electron model and nonequilibrium ab initio molecular dynamics simulations. We confirm that the kinetic energy of thermally excited electrons originates from the energy of the spatial electrostatic potential oscillation, which is induced by the thermal motion of ion cores. This method directly predicts the electronic thermal conductivity of pure metals with a high degree of accuracy, without explicitly addressing any complicated scattering processes of free electrons. Our methodology offers a route to understand the physics of heat transfer by electrons at the atomistic level. The methodology can be further extended to the study of similar electron-involved problems in materials, such as electron-phonon coupling.

Resume : The lattice thermal expansion and conductivity in bulk Mo and W-based transition metal dichalcogenides are investigated by means of density functional and Boltzmann transport theory calculations. To this end, a recent van der Waals density functional (vdW-DF-CX) is employed, which is shown to yield excellent agreement with reference data for the structural parameters. The calculated in-plane thermal conductivity compares well with experimental room-temperature values, when phonon-phonon and isotopic scattering are included. To explain the behavior over the entire available temperature range one must, however, include additional (temperature independent) scattering mechanisms that limit the mean free path. Generally, the primary heat carrying modes have mean free paths of 1μm or more, which makes these materials very susceptible to structural defects. The conductivity of Mo- and W-based transition metal dichalcogenides is primarily determined by the chalcogenide species and increases in the order Te-Se-S. While for the tellurides and selenides the transition metal element has a negligible effect, the conductivity of WS2 is notably higher than for MoS2, which may be traced to the much larger phonon band gap of the former. Overall, the present work provides a consistent set of thermal conductivities that reveal chemical trends and constitute the basis for future investigations of thermal transport in van der Waals solids.

Resume : Organic semiconductors are attractive for applications as varied as electronics, optoelectronics and thermoelectricity. Despite its importance, thermal transport in organic based semiconductors have received little attention. From a computational point of view one of the main challenges is to correctly describe intramolecular and intermolecular interactions. The latter are dominated by van der Waals (vdW) forces and are crucial to reach a reliable description of the thermal transport properties in these systems. So far, the most accurate description of both interactions is provided by density functional theory (DFT) in combination with methods to account for vdW forces. In this talk we will present a detailed benchmark of different models to describe the vibrational properties, and hence thermal properties, of a set of organic semiconductors. The benchmark includes: DFT, the density functional tight binding method DFTB , and the COMPASS and CHARMM force field potentials. DFT and DFTB were performed including the pairwise and many-body vdW methods of Tkatchenko et al. The accuracy of the DFT-vdW calculations is validated against Raman spectra measurements of organic thin films. As an outcome, we show that DFTB offer the best tradeoff between accuracy and cost.

Resume : With ever-decreasing device downscaling, understanding thermal transport in nanoscale systems is a key technological issue[1]. In this context, we theoretically address the ultrafast cooling of metal nano-objects embedded in transparent environment as measured in time-resolved optical spectroscopy[2]. In the experiment a “pump” laser pulse impulsively heats a metal nano-object. The thermal relaxation is then accessed exploiting a time-delayed “probe” pulse, monitoring the temperature-dependent relative transmittivity variation. The modeling, based on the Finite-Element Method, couples the two physics involved in the experiment, namely thermal and optical problem. The system thermal dynamics is first computed in the frame of Fourier law and Kapitza-like thermal resistance[3]. The system electromagnetic extinction spectrum is then calculated at various delay times, thus accounting for the temperature-dependent variations of the system dielectric functions. Within this frame we numerically simulate the experiments performed on metal nano-spheres embedded in a liquid environement and metallic nanodisks patterned on a dielectric substrate. By tuning the Kapitza resistance we obtain a good agreement between the experimental and theoretical optical traces, allowing to estimate the Kapitza resistance at metal nano-object-environment interface. [1] Hartland, Chem.Rev. 111,3858(2011) [2] Stoll et al., J.Phys.Chem.C 119,12757(2015) [3] Banfi et al., Appl.Phys.Lett. 100,011902(2012)

Resume : Heat transfer can be drastically limited in non-uniform materials due to enhanced phonon scattering. The decrease of the thermal conductivity is related to the details of the microstructure. Understanding and modeling this relation would be useful in designing structures for applications such as efficient energy converters. We have examined mesoscopic structures consisting of two phases that differ in the thermal conductivity using the phonon Monte Carlo and Molecular Dynamics techniques. The results have been analyzed in terms of a thermal network model. Phenomenology has been developed extending existing Effective Medium Approximation models that interprets the MC results and provides physics insight. We have systematically explored the effects of the characteristic lengths and the dimensions of the nanoscale non-uniformity. These dependences are introduced as parameters in the phenomenological model. Transition from ballistic to diffusive transport signature in thermal transport is explicitly indicated. We discuss our results and we use the parametric modeling to provide generic conclusions.

Resume : Heat transport through a solid–solid junction: the interface as an autonomous thermodynamic system Riccardo Rurali-a, Luciano Colombo-ab, Xavier Cartoixa-c, Øivind Wilhelmsen-d, Thuat T Trinh-d, Dick Bedeaux-d, Signe Kjelstrup-d aInstitut de Cie`ncia de Materials de Barcelona (ICMAB–CSIC) Campus de Bellaterra, 08193 Bellaterra, Barcelona, Spain bDipartimento di Fisica, Universita` di Cagliari, Cittadella Universitaria, I-09042 Monserrato (Ca), Italy cDepartament d’Enginyeria Electro`nica, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain dDepartment of Chemistry, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway. E-mail: dick.bedeaux@chem.ntnu.no We performed computational experiments using nonequilibrium molecular dynamics simulations,1 showing that the interface between two solid materials can be described as an autonomous thermodynamic system.2-5 We verify local equilibrium and give support to the Gibbs description6 of the interface also away from global equilibrium. In doing so, we reconcile the common formulation of the thermal boundary resistance as the ratio between the temperature discontinuity at the interface and the heat flux7 with a more rigorous derivation from nonequilibrium thermodynamics.8,9 We also show that thermal boundary resistance of a junction between two pure solid materials can be regarded as an interface property, depending solely on the interface temperature, as implicitly assumed in some widely used continuum models, such as the acoustic mismatch model. Thermal rectification can be understood on the basis of different interface temperatures for the two flow directions. 1) R Rurali, L Colombo, X Cartoixa, Ø Wilhelmsen, TT Trinh, D Bedeaux, S Kjelstrup, Phys.Chem.Chem.Phys. 2016, 18, 13741 2) G. Bakker, Handbuch der Experimentalphysik, Akad. Verlag, Leipzig, Germany, 1928, vol. 6, ch. 10. 3) E. A. Guggenheim, Thermodynamics, North-Holland, Amsterdam, 1985. 4) A. Rosjorde, D. W. Fossmo, D. Bedeaux, S. Kjelstrup and B. Hafskjold, J. Colloid Interface Sci., 2000, 232, 178–185. 5) E. Johannessen and D. Bedeaux, Physica A, 2003, 330, 354–372. 6) J. W. Gibbs, The Scientific Papers of J. W. Gibbs, Dover, New York, 1961. 7) P. L. Kapitza, J. Phys., 1941, 4, 181–210. 8) D. Bedeaux, A. M. Albano and P. Mazur, Physica A, 1976, 82, 438–462. 9) S. Kjelstrup, D. Bedeaux, Non-Equilibrium Thermodynamics of Heterogeneous Systems, World Scientific, Singapore, 2008.

Resume : In ferroelectric materials external electric fields are today routinely used to write, move and erase consecutive regions of different electric polarization, ferroelectric domains, spatially separated by elastic interfaces known as domain walls (DWs). The lattice distortion at these coherent interfaces can induce some amount of scattering to the traveling phonons, a property that shapes them as candidate mobile elements to dynamically manipulate heat fluxes with electrical signals. Indeed, thermal conductivity modulation via DW engineering was demonstrated long time ago at low temperatures [Physica 52, 577 (1971)] and more recently over a broad temperature range, including room temperature [Appl. Phys. Lett. 102, 121903 (2013) & Nano Lett. 15, 1791 (2015)], thus boosting its technological impact. However, little is known about the mechanisms of phonon scattering at different ferroelectric DWs, an information difficult to obtain from experimental measures but necessary in order to rationally design DW thermal transport engineering devices. Therefore, the combined use of theory and numerical simulations becomes imperative. In this contribution we will present the first phonon transport calculations in multidomain ferroelectrics with atomistic precision. Our modeling is based on a second-principles model potential approach [J. Phys. Cond. Matt. 25, 305401 (2013)] and a non-equilibrium Green's function calculation extended to obtain the single-mode contribution of individual phonons to the total transmission function [Phys. Rev. B 91, 174302 (2015)]. We will demonstrate that thermal transport across 180º DWs formed in bulk PbTiO3 is sensitive to the polarization of the heat carrying phonons. The propagation of transverse phonons across the DWs is strongly suppressed by the lattice distortion in the transverse directions whereas longitudinal phonons propagate through samples with multiple DWs without feeling them. All in all 180º DWs behave as longitudinal phonon polarizers of high selectivity whose effect is robust against deviations from the ideal flat shape.

Resume : To extract the thermal conductivity of a micro or nanoscale system, experimental techniques typically require the determination of temperature between two points. Raman spectroscopy provides a contactless non-destructive method to achieve this in samples with challenging geometries, by simultaneously using a focused laser as a heat source and measuring probe. The experimental data of weighted temperature as a function of the power absorbed on the probed volume is subsequently compared to a heat conduction model to extract the thermal conductivity. In the case of bulk materials and thick layers, an analytical formulation based on a steady-state two dimensional heat conduction model is often used. For this derivation, the studied material is considered as a semi-infinite medium with strong light absorption at the surface. However, thermal characterization sometimes is needed for samples that may not be considered as semi-infinite volumes or that have significant light penetration depths at the used wavelength. Then, to employ such approximation in the experiment analysis may introduce considerable measurement errors. In this work, we implemented a finite element based numerical simulation of the heating of a bulk sample by means of a Gaussian laser beam. As a result, the validity ranges of the assumptions involved in the analytical heat conduction model used in Raman thermometry analysis are evaluated. By comparing our numerical model with the simplified analytical solution for the case of laser heating of a semi-infinite solid, we obtain the range of applicability of different conditions, such as sample dimensions and penetration depth of light of the studied material. We found that for a ratio of penetration depth to laser spot radius higher than 0.2 , the assumption of strong shallow absorption used in the simplified heat conduction model is no longer fulfilled, leading to an overestimation higher than 10% of the thermal conductivity. Similarly, we found that the ratio of spot radius to thickness of the studied specimen should be smaller than 0.1 to predict thermal conductivity values below 10% error using the simplified analytical approximation. Work is in progress to extend this study to thin film-on-substrate systems. The simulation here developed appears as a powerful tool for predicting the appropriate sample geometry and range of absorption to analyze Raman thermometry measurements by means of analytical formulas.

Resume : In this contribution, we will discuss two examples of heat transfer phase change at the nanoscale. The first issue concerns the formation and growth of nanobubbles generated by heated nanoparticles. Nanobubbles have been experimentally generated by laser heated colloidal particles. Understanding their production and growth turns out to have applications in cancer cell therapy [1, 2]. Despite their promising use, the fundamental description of the mechanisms at the origin of nanobubble generation is still missing. Here, we model water dynamics around heated gold nanoparticles [3, 4], using hydrodynamic phase field model. We discuss the major role played by ballistic thermal transport inside the nanobubble [4]. Our simulations demonstrate the existence of a flux threshold to generate long-lived nanobubbles, with relative large nanobubble size. The laser duration is found to have little effect on the maximum nanobubble size, but long laser pulses are preferrable to increase the nanobubble lifetime. The second example of heat transfer phase change discussed here concerns boiling of nanofluids. In this case, we have conducted molecular dynamics simulations to probe boiling at a gold/model nanofluid interface. The nanofluid model consists in massive particles suspended in a solution of liquid Argon atoms, with a typical volume fraction around 0.1. Our simulations show that the presence of nanoparticles may significantly shift the occurrence of boiling crisis, as compared to the pure fluid case. This shift is to a large extent explained by the decrease of the thermal boundary conductance induced by the presence of the nanoparticles. This decrease has been quantified as a function of the nanoparticle-solid and nanoparticle-fluid interactions. Moreover, a spectral analysis of the heat flowing at the interface allowed us to understand the physical origin of the decrease of thermal boundary conductance. The heat flux transmitted from gold to the suspended nanoparticles is rather low as compared to the energy flowing from gold to the suspending fluid. This is the result of the poor coupling between the internal modes of the nanoparticles and the vibrational modes characterizing gold. This study may help in designing fluids with enhanced thermal properties. [1] Lukianova-Hleb, E. and Hu, Y. and Latterini, L. and Tarpani, L. and Lee, S. and Drezek, R.A and Hafner, J.H. and Lapotko, D.O. ACS Nano, 4, pp. 20192123, (2010) [2] Siems, A. and Weber, S.A.L. and Boneberg, J. and Plech, A. New J. Phys., 13, pp. 043018, (2011) [3] Lombard, J. and Biben, T. and Merabia, S.,Phys. Rev. Lett., 112, pp. 106701, (2014) [4] Lombard, J. and Biben, T. and Merabia, S., Nanoscale 8, 14870-14876, (2016)

Resume : Using nano-technologies makes it possible to change in wide range electronic and optical characteristics of semi-conductor nano-crystals. The systems containing silicon nano-crystals in dielectric matrix have a perspective for creation of light-emitting systems compatible with IC-devices technology. Alloying structures of silicon nano crystals with ions of rare-earth metals makes it possible to realize a unique process of practically entire transmission of exiton energy on ions' internal degrees of freedom. For correct alloying such structures it is necessary to know distribution of temperature in process of heating because alloying itself takes place through the heating time. This is important essentially for changing electrical and optical properties of these structures. Semiconductor structures may be partially or fully opaque to radiation at the laser wavelength. Depending on the degree of transparency of different approaches will be useful for modeling the laser heat source. In addition, it must be remembered that all scales should be compared with the wavelength. Different approaches are required to describe the focused radiation for a relatively wide beam. If a material interacting with the incident beam, the geometrical features are comparable with the wavelength, it is necessary to consider further how the beam will interact with these fine structures. Fourier equations gives a possibility to calculate a heat distribution not only in objects-models, but in real studied structures for real experiment environment. It simplifies a process of the experimental planning and enlarges information value of the results obtained. Calculations realized in the given investigation make it possible to forecast a temperature distribution in silicon periodic structures in process of laser annealing. It gives a possibility for more precise alloying such structures that will change their optical and electrical properties in required range.

Resume : By means of state-of-the-art first principles calculations, we investigate the thermal properties of SiO2 as it undergoes a pressure induced ferroelastic structural phase transition at high pressure. The lattice thermal conductivity is obtained in a wide range of temperatures and pressures by solving the Boltzmann transport equation self-consistently. SiO2 shows a robust peak in the out-of-plane diagonal component of the thermal conductivity tensor close to the phase transition from the stishovite phase to the CaCl2-type phase. The lattice thermal conductivity is almost two orders of magnitude larger close to the critical pressure than at lower or higher pressures in both phases. We attribute the origin of this anomalously high thermal conductivity to a steep decrease in the isotopical scattering rates close to the critical pressure. We thus propose this effect to be exploited to create a pressure-controlled anisotropic thermal transistor.

Resume : Based on a recent theory of wave localization and the development of a mathematical tool (the localization landscape) we propose an approach that predicts the localization of thermal phonons in disordered lattices. An analogy between the Schrödinger equation and the classical equation of motion is introduced to demonstrate that localization of thermal energy in phonon systems arises from atomic disorder through which the localization landscape can be revealed. This approach, illustrated on disordered graphene, provides a powerful framework for engineering heat conduction in nanostructures using wave effects.

Resume : Two-dimensional phononic crystals (PnCs) can effectively control nanoscale heat conduction at low temperatures and typically consist of periodic arrays of either holes in a membrane (hole-based PnCs) or pillars on top of the membrane (pillar-based PnCs). In such periodic structures, phonon interference flattens phonon dispersion, which causes reduction in the phonon group velocity, modifications in the density of states (DOS), and overall change in thermal conductance. In this work, we simulate heat conduction in both types of PnCs within the purely coherent regime and systematically investigate how various geometrical parameters, lattices and materials affect the thermal conductance. We show that PnCs of both types suppress heat conduction when the period is sufficiently long, and on the contrary, enhance thermal conductance when the period is short (< 60 nm). In pillar-based PnCs, where local resonances play an important role, we show that the resonances increase the DOS and thus surprisingly contribute to enhancement rather than suppression of heat conduction. Thus, the overall suppression of thermal conductance in pillar-based PnCs appears despite (not due to) the local resonances and is caused by periodicity of the structure. The lowest thermal conductance is achieved when the pillars are short and made of materials with a low Poisson’s ratio, and when both pillars and holes have the highest possible radius-to-period ratio. In general, we show that pillar-based PnCs can match the efficiency of the conventional hole-based PnCs and have richer physics which opens new possibilities for heat conduction engineering. Finally, we propose hybrid hole/pillar-based PnCs. These hybrid PnCs combine both types of PnCs and suppresses heat conduction stronger than hole- or pillar-based PnCs, thus can potentially become the second generation of PnCs for heat conduction engineering.

Resume : The properties of ferroelectric materials have become an intensively studied topic, both theoretically [1] and experimentally [2], attracting the interest of many researchers. In this work, we study the thermal conductivity of the PbTiO3, and also the thermal boundary resistance [3] of the regions disturbed by its domain walls, by means of non-equilibrium molecular dynamics simulations. For these simulations we have modeled the atomic interactions with a potential obtained from first principles [4], accurate but computationally efficient. Our result of the thermal conductivity is in good agreement with previous reports obtained from first-principles [5] and allowed estimating for the first time the interface thermal resistance of a domain wall. [1] Liu Yong, Ni Li-Hong, Xu Gang, Song Chen-Lu, Han Gao-Rong, and Zheng Yao. Phase transition in PbTiO3 under pressure studied by the first-principles method. Physica B: Condensed Matter, 403(2122):3863 – 3866, 2008. [2] S. T. Davitadze, S. N. Kravchun, B. A. Strukov, B. M. Goltzman, V. V. Lemanov, and S. G. Shulman. Specific heat and thermal conductivity of BaTiO 3 polycrystalline thin films. Applied Physics Letters, 80(9), 2002. [3] R. Dettori, C. Melis, X. Cartoixà, R. Rurali & L. Colombo (2016) Thermal boundary resistance in semiconductors by non-equilibrium thermodynamics, Advances in Physics: X, 1:2, 246-261, DOI: 10.1080/23746149.2016.1175317 [4] Jacek C Wojdel, Patrick Hermet, Mathias P Ljungberg, Philippe Ghosez, and Jorge Íñiguez. First-principles model potentials for lattice-dynamical studies: general methodology and example of application to ferroic perovskite oxides. Journal of Physics: Condensed Matter, 25(30):305401, 2013. [5] Anindya Roy. Estimates of the thermal conductivity and the thermoelectric properties of PbTiO 3 from first principles. Phys. Rev. B, 93:100101, Mar 2016.

Resume : In this contribution, we present experimental evidence on the change of the phonon spectrum and vibrational properties of a bulk material through phonon hybridization mechanisms. The phonon spectrum in a finite material is strongly affected by the presence of free surfaces, which is the addition of a contribution from an essentially two-dimensional crystal. The phonon spectrum of a bulk material can hence be altered by an hybridization mechanism between confined phonon modes in nanostructures introduced on the surface of a bulk material and the underlying bulk phonon modes. We measured the heat capacities of bare and surface-structured silicon substrates originating from the same silicon wafer. Then, we deduced important features of the phonon spectra of the samples investigated through a rigorous analysis of the measured heat capacity curves. The results clearly showed that the shape and size of the nanostructures made on the surface of the bulk substrate have a strong effect on the phonon spectrum of the bulk material.

Resume : Although atomistic simulations and first principles computation of harmonic normal modes and anharmonic forces are nowadays widely used to solve Boltzmann equation for phonons and accurately calculate the lattice thermal conductivity, there is still a need for analytical approximate models leading to reliable solutions to heat transport problems in nanostructures, particularly at low temperatures. The utmost need for such approximate models stems from the fact that most of the atomistic simulation techniques do not account for quantum mechanics effects occurring at low temperatures, and first principles computation requires very small mesh size to account for the excitation of only long wavelength phonons. This paper presents significant advances in the analytical calculation of the low temperature lattice thermal conductivity in cylindrical nanowires. It shows that an accurate prediction of the thermal conductivity in cylindrical nanowires can be obtained at low temperatures when Boltzmann equation in cylindrical coordinates is solved. It also provides an approach to predict from a spatial-dependent Boltzmann equation the rate at which phonons are scattered by the nanowire boundary in the presence of intrinsic scattering mechanisms. The derived formalism clearly shows that the thermal conductivity of a cylindrical nanowire presents specific features that do not appear for the case of bulk materials. It shows that a contribution from an essentially two-dimensional crystal alters the temperature dependence of thermal conductivity and gives rise to distinct size effects on the thermal conductivity. The accuracy of the derived formalism is demonstrated clearly with reference to reported data regarding the size effect on the thermal conductivity of Si nanowire, the dramatic drop in the thermal conductivity for rough Si nanowires, and the deviation of the thermal conductivity of thin cylindrical nanowires from the Debye law at low temperatures.

Resume : Disordered or amorphous structure graphene is a type of defective graphene. Recently it has been shown that, disordered graphene can be synthesized by focused ion beam converting ordered hexagonal arrangement of graphene cells to stochastic and disordered polygonal ones. Strain is one of the external parameters that could be used in tuning the thermal and mechanical properties of graphene as well as other nanostructures. In this investigation, we explore the engineering of the thermal transport along amorphous graphene structures through employing the mechanical strains. It is found that the effect of strain on the thermal conductivity of disordered graphene is clearly less considerable as compared with the pristine structure.

Resume : We introduce a non-equilibrium molecular dynamics simulation approach, based on the generalized Langevin equation, to study vibrational energy relaxation in similar condition to pump-probe spectroscopy experiments. A colored noise thermostat is used to selectively excite a set of vibrational modes, leaving the other modes nearly unperturbed, to mimic the effect of a monochromatic laser pump. Energy relaxation is probed by analyzing the evolution of the system after excitation in the microcanonical ensemble, thus providing direct information about the energy redistribution paths at the molecular level and their time scale. The method is applied to hydrogen bonded molecular liquids, specifically deuterated methanol and water, which show a manifold energy dissipation dynamics due to strong directionality of H-bonds and complexity of their network. The colored thermostat is tuned to excite the stretching mode of the OD, or OH, bond involved in H-bonding. We evaluate how energy is transferred from the excited mode to other modes of the nearby molecules, providing a robust picture of energy relaxation at the molecular scale.

Resume : Ideal non-metallic thermoelectric materials should show low thermal conductivity ĸ along with low electrical resistivity ρ. Thus, strategies are needed to prevent the propagation of phonons mostly responsible for thermal conduction while only marginally impeding charge carrier diffusion. Defect engineering may provide tools to this aim, provided that an adequate understanding of the role played by morphological defects may guide the selection of scattering centers mostly effective at reducing ĸ with minor impacts on ρ . We present the results of a study on how silicon thermal conductivity is differently affected by multiple morphological defects. A blended approach has been adopted, encompassing simulations and experiments so as to cover a wide range of defect densities. We show that co-presence of morphological defects with very different characteristic sizes effectively reduces the thermal conductivity. We also point out that models with no spectral resolution are limited to predict and explain the dependency of ĸ upon the density of morphological defects. However, limits coming from the use of the spectral Matthiessen rule are put forward, especially in the range of very few nanovoids (NV), showing how a fully physically consistent picture of the thermal conductivity calls for non-local effects related to lattice disordering induced by NVs over a wider range of porosities.

Resume : Heat transfer at the meso/nano-scale represents an outstanding challenge, among the most relevant under an applicative standpoint [1]. In this context Thermal Boundary Resistance (TBR) plays a key role. Accessing the TBR between nanosized metals and insulating substrates remains an open issue and the prerequisite to enhance heat dissipation in next generation micro and nano-devices. Unfortunately theoretical predictions deviate from TBR values extracted from TRTR measurements, the go-to technique to inspect TBR at these junctions. The present work, based on multiscale modelling, reconciles the discrepancy for the paradigmatic case of the Al-Sapphire heterojunction. First, we calculate the TBR at the interface between a nanosized Al film and an Al2O3 substrate at an atomistic level. The thermal dynamics occurring in TBTR experiments is then modelled via macro-physics equations upon insertion of the materials parameters obtained from atomistic simulations. Electrons and phonons non-equilibrium and spatio-temporal temperatures inhomogeneities are found to persist up to the nanosecond time scale. The strategy adopted in the literature to extract the TBR from transient reflectivity traces is revised at the light of the present findings. The results are of relevance beyond the specific system, the physical picture being general and readily extendable to other heterojunctions. [1] D. G. Cahill et al., Appl. Phys. Rev.,2014, 1, 011305 [2] C. Caddeo et al., PRB 2017 (submitted)

Resume : In this work, we study thermal transport properties of single extended isolated chains of poly-3,4-ethylenedioxythiophene (PEDOT) with the aim of identifying possible anomalous regimes in the heat transport holding for one-dimensional systems. In particular, by investigating the PEDOT length-dependence thermal conductivity by means of the Approach to Equilibrium Molecular Dynamics (AEMD) method[1,2], we observe an ~L^β divergence (β=0.48) indicating that single PEDOT chains obeys to a super diffusive regime defined as an intermediate between a purely ballistic (β=1) and diffusive (β=0) regime[3]. The present results are further analyzed by means of Equilibrium Molecular Dynamics in which the time-dependence of the heat current auto-correlation function has been estimated for a PEDOT chain of 1000-monomers. The results show a ~t^(1-β) time decay giving rise, according to the Green-Kubo formula, to a corresponding ill-defined thermal conductivity[3]. The estimated value of the β exponent is found in perfect agreement with the one estimated with AEMD, further confirming the occurrence of a thermal transport super diffusive regime. Present findings are eventually validated by investigating the decay in time of a localized thermal excitation. [1] E. Lampin et al., J. Appl. Phys. 114, 033525 (2013) [2] C. Melis et al., Eur. Phys. J. B 87, 96 (2014). [3] S. Lepri et al., Lecture Notes in Physics 921, 1-37 (2016)

Resume : Thermoelectric devices are promising and environmentally friendly systems to recover energy from industrial waste heat or natural heat. Even if inorganic materials generally show high efficiencies in thermoelectric devices, these materials are typically expensive and are characterized by brittleness, which makes difficult their application in large areas. Furthermore, organic materials have peculiar advantages as thermoelectric materials, such as cost effectiveness, low thermal conductivity and high flexibility. For these reasons , conjugated polymers, which possess good electrical conductivity, have been actively investigated. Among different conducting polymers, poly(3,4-ethylenedioxythiophene) (PEDOT) and PEDOT: poly(styrenesulfonate) (PSS) are promising candidates due to their water-solubility and high electrical conductivity. Recently, a room temperature ZT value of 0.42 has been reported in PEDOT:PSS which was mainly attributed to the relatively high electrical conductivity and extremely low thermal conductivity k=0.34 W m-1 K-1. Other studies reported k values ranging from 0.34 up to 2.5 W m-1 K-1 consequently affecting the estimates of ZT. A possible explanation of such uncertainty in the k values can be related to the different experimental setups employed (e.g. solvent, temperature). This highlights the possible role played by morphologies in polymers thermal transport. In this work we elucidate this issue by demonstrating how morphology ultimately governs thermal transport properties in polymers. In particular, by means of the Approach to Equilibrium Molecular Dynamics[1], we estimate thermal conductivity of a PEDOT system. Crystalline, amorphous and a series of semi-crystalline structures in between are obtained by varying the degree of polymerization. A dramatic increase up to two orders of magnitude is observed from amorphous to purely crystalline PEDOT. Also the degree of polymerization is found to have a strong effect: by increasing the actual PEDOT chain length from 100 to 1000 monomers, we observe a corresponding one order of magnitude increase in thermal conductivity. [1] E. Lampin et al., J. Appl. Phys. 114, 033525 (2013). [2] C. Melis et al., Eur. Phys. J. B 87, 96 (2014).

Resume : A promising field which could highly benefit from the intrinsic features of eumalinin molecule is photovoltaics, where such a pigment could be used as photo-active layer in hybrid solar cells, being Si the inorganic counterpart. In these devices, the process of optical to electrical energy conversion takes place at the eumelanin/inorganic interface via the exciton dissociation followed by carrier diffusion and collection at the electrodes. The overall power conversion efficiency of the cell is likely affected by the presence of scattering sources (impurities and vibrational modes): in particular, vibrations could detrimentally impact the dissociative capabilities of the interface and reduce the mean free path of carriers in the organic layer. Hence, an analysis of the thermal properties of the eumelanin/inorganic interface is indeed needed for improving our fundamental understanding of the above processes. In our work we study the thermal transport properties of a silicon[100]/eumelanin interface. By non-equilibrium molecular dynamics techniques, we compute the thermal boundary resistance (TBR) at the interface. Due to the elusive nature of eumelanins and the amorphous character of the corresponding layer, we perform a statistical analysis including the effects of chemical disorder which characterizes our knowledge of eumelanin-like pigments to-date. In particular, we show the particular dependence of the TBR from the compositional structure of the organic layer and the morphological features of the silicon surface (curvature, roughness).

Resume : The newest members of 2D transition metal carbides and nitrides, so-called MXenes, recieve increasing attention due to their novel electronic and thermal properties. Depending on their prominent performances, MXene materials are promising materials for applications in various electronic devices. Previously, the electronic properties and the dynamical stability of some MXene structures have been investigated by different researchers. Only very few studies focus on thermoelectric performance of semiconductor MXene materials. In this study, we investigate the lattice thermal transport properties of semiconductor Ti2CO2 MXene crystal in different functionalized forms using Density Functional Theory and the Phonon Boltzmann Transport Theory. We obtain the thermal conductivity as 40.1 W/(mK) for Model 2 and 17.4 W/(mK) for Model 3 of Ti2CO2 at T=300 K. Because of its low thermal-conductivity, it would be a good candidate for next-generation thermoelectric applications.

Resume : We introduce and discuss a new class of structurally disordered phononic crystals that efficiently combine the behaviour of periodic and disordered structures. Phononic crystals have shown remarkable applications over the last two decades, while stealthy hyperuniform disordered structures (HUDS), formed by appropriately decorating a 2D stealthy hyperuniform point pattern, have shown remarkable behaviour as photon cavities and waveguides. Here, the underlying mechanisms for the formation of large phononic band gaps in HUDS will be thoroughly presented. As an example, we numerically investigate, through finite element calculations, the formation of band gaps in HUDS made of 500 lead cylinders in an epoxy matrix. In the 2D case (infinite height of the cylinders) the elastic wave for both in-plane and out-of-plane polarization exhibits large phononic band gaps, similar to the periodic ones. Therefore, very high-Q cavity modes (Q~10^14, neglecting dissipative losses) can be easily formed in these structures. We will also present, a new bottom-up approach available for HUDS, to form arbitrary shaped waveguides with 100% transmission through sharp bends, not possible in the periodic structures. Finally, we will discuss 3D elastic wave propagation through finite height slabs of HUDS and the existence of high-Q modes (Q~10^11, neglecting dissipative losses) and efficient waveguiding. Applications of these structures as MEMS, thermoelectric, integrated phonon circuits are anticipated.

Resume : The output characteristics of the semiconductor microwave devices are determined by a multiplicity of factors reflecting both structural and technological features of the heterostructure production. It is extremely important to choose optimal parameters defining the basic electrophysical characteristics of the structure - concentration and mobility of the charge carriers in the two-dimensional electron gas channels (2D electron gas). This work presents a new approach that allows solving the tasks of optimization of nanosized semiconductor heterostructures as the optimal control tasks. The problem of the optimal barrier layer doping of AlGaN/GaN heterostructures is considered.

Resume : Controlling and improvement of interfacial thermal conductance between graphene nanosheets plays a crucial role with respect to the synthesis of highly thermally conductive nanomaterials and devices[1]. This way, finding practical routes toward the enhancement of interfacial conductance without suppressing the thermal transport is currently among the most challenging issues. In this work, we suggest the chemical functionalization with organic molecules covalently attached to adjacent graphene sheets as a promising approach for the enhancement of thermal conductance between them. In particular, molecules like phenoxypentoxybenzene between graphene sheets were studied, in different possible experimental configurations, including strained,bent molecules and out of plane transmission. To provide physical insight concerning the heat transfer mechanism at interfaces, we conducted molecular mynamics (MD) simulations. To this aim, we employed non-equilibrium MD (NEMD) technique along with the COMPASS force-field for describing the interatomic interactions. Density Functional Tight Binding (DFTB) MD was also carried out using the DFTB+ code [3] to benchmark the LAMMPS NEMD results. Also, phonon properties of the bridging chain were studied via Green's functions formalism [5]. [1] Nature Communications 7 (2016) 11281 [2] Carbon 63 (2013) 460-470 [3] Computational Materials Science 47 (2009) 237-253 [4] http://www.dftb-plus.info/ [5] Numerical Heat Transfer, Part B, 51 (2007) 333-349

Resume : In this work we show a phenomenological alloy-like fit of the thermal conductivity of Ad1:Bd2 superlattices with d1 =/ d2. The presented method is a generalization of the Norbury rule of the summation of thermal resistivities in alloy compounds. Namely, we show that this approach can be also extended to describe the thermal properties of crystalline and ordered-system composed by two or more elements, and, has a potentially much wider application range. Using this approximation we estimate that the interface thermal resistance depends on the period and the ratio of materials that form the superlattice structure.

Resume : In the study of thermal transport phenomena at the nanoscale, it is usual to introduce a spectrum-averaged phonon mean free path (the gr ay-MFP approximation). However, recent reports have shown that the use of a single-averaged MFP may be inadequate. The analysis of the contribution of heat carriers with different MFP to the total thermal conductivity is key to understand the role that different physical phenomena play in the reduction of thermal conductivity in nanostructured materials. In the MFP reconstruction process, the election of the regularization parameter is of great importance to obtain an adequate suppression function that relates the bulk thermal conductivity and that of the nanostructure. In this work we study the criteria for choosing a value for the regularization parameter and the influence of this election in the final result of the MFP reconstruction. This process is then applied to calculate the thickness-dependant suppression function for thermal conductivity in Silicon membranes. This model is also used to study how different geometrical parameters affect the phonon MFP distribution in nanostructured Silicon membranes.

Resume : Designing materials with exceptionally low or high phonon conductance remains an open challenge, contrary to what have been achieved with electrons or photons. For instance, Fresnel coefficients allow to easily compute the angular-dependent reflection/transmission coefficients of light in multilayer systems. Analogous quantities for phonons, which are still to be found, would guide the design of interfacial materials with optimal conductance. In this work, we have investigated how to describe phonon interfacial transport from a modal perspective at the atomic scale. We have characterized how phonons propagate through multilayer materials by developing a modal version of the Atomistic Green's Function. These results give a detailed framework to reinterpret phonon transmission at interfaces in analogy with light propagation. This work provides a foundation to design interfacial materials with extreme values of thermal conductance for thermoelectricity and heat management.

Resume : Clathrates exhibit a very low thermal conductivity, which is a key factor for their very good thermoelectric properties and has been attributed to "phonon-glass" conduction behavior. Here, we present a computational analysis of the conduction mechanism using Ba8X16Y30 (X={Al,Ga}, Y={Si,Ge}) as model systems. Contributions to the thermal conductivity from both electrons as well as phonons are computed with Boltzmann transport theory using input from first-principles calculations. The calculations are in good agreement with experimental data and in particular reproduce the experimentally observed ordering of the lattice thermal conductivity between the different systems. We demonstrate that these rather non-intuitive trends can be traced to the presence (or lack thereof) of dispersed higher frequency optical phonon modes, which provide a surprisingly large contribution to the lattice thermal conductivity. The "phonon-glass" behavior manifests itself in our calculations in the form of very short lifetimes, which indicate that already at room temperature the majority of modes is overdamped. In terms of the electronic thermal conductivity, our results provide insight into the applicability of the Wiedemann-Franz law. While the latter is regularly used to separate lattice and electronic contributions to the thermal conductivity in clathrates. Depending on the choice of pre-factor our data show that one can incur an error of up to 50% for the electronic thermal conductivity.

Resume : Thermal boundary resistance is a critical quantity that controls heat transfer at the nanoscale which is primarily related to interfacial phonon scattering [2]. In microelectronics for instance, there is a strong need to know how energy can be exchanged at separation distance of few nanometers where heat is primarily exchanged by acoustic waves for sub-nanometric gaps [3]. In this communication, we combine lattice dynamics calculations and inputs from first principles ab initio simulations to predict phonon transmission at Si/Ge interface as a function of both phonon frequency and phonon wavevector. This technique allows us to determine the overall thermal transmission coefficient as a function of the scattering direction and frequency. Our results show that, the thermal energy transmission is highly anisotropic while thermal energy reflection is almost isotropic. In addition, we found the existence of a global critical angle of transmission beyond which almost no thermal energy is transmitted. This critical angle around 60 degree. is found to be almost independent on the interaction range between Si and Ge, the interfacial bonding strength, and the temperature above than 30 K. We interpret these results by carrying out a spectral and angular analysis of phonon transmission coefficient and differential thermal boundary conductance (TBC). Besides, our calculations allow us to predict heat transfer mediated by phonons across a nanometer vacuum scale gap and we compare the results with recent experiments [4]. The results are also interpreted by investigating the transmission across the vacuum gap. [1] A. Alkurdi and S. Merabia J. Phys.: Conf. Ser. (2016) [2] D. G. Cahill et al Appl. Phys. Rev. 1 011305 (2014). [3] V. Chiloyan, J. Garg, K. Esfarjani and G. Chen Nature Communications 6 6755 (2015) [4] K. Kloppstech, N. Könne, S.-A. Biehs, A. W. Rodriguez, L. Worbes, D. Hellmann, and A. Kittel. Nature Communications (2017).

Resume : In this work, we have investigated thermal properties of interfaces made of stacked Si and Ge using Molecular Dynamic simulations. The thermal conductance was calculated for several lattice orientations over a wide range of temperatures. The non-equilibrium method based on the temperature difference set between two materials as well as the equilibrium method which relies on the fluctuations of the heat flux are compared. The results of the two methods are discussed as well as their limitations.

Resume : The prediction of thermal transport at the nanoscale has been a question of intense debate in the last decade. Although several abinitio models have been able to predict bulk thermal conductivity of a large number of materials from calculated relaxation times, using the same magnitudes at the nanoscale seems to give wrong predictions. This disagreement may have two reasons: the presence of different orders of magnitude in the mean free paths of the phonons and the role of momentum conserving collisions like normal scattering. The Kinetic Collective Model (KCM) has been proposed recently to predict the thermal transport under the influence of the above characteristics. In this poster we present a new set of analytical equations based on KCM that can be easily used in design time through a finite elements program. The model includes memory and nonlocal effects to describe most of the phenomena observed at the nanoscale. Numerical results and comparison to experimental data are presented for thin films and nanowires. The results show that anomalies found on reduced size samples may have an hydrodynamic basis. Poiseuille flow and effective Fourier flow are shown to be the limit regimes of reduced samples depending on the dominance of normal scattering. In the light of KCM predictions we show that an hydrodynamic description is necessary to understand them. These results open the door to introduce new physics in finite elements programs that will be able to give better predictions for the current electronic devices, where its characteristic scales are of the same order as the hydrodynamic scales predicted by KCM equations.

Resume : Under electrical bias, small nanoparticles exhibit remarkable local heating effects. As a matter of fact, it is possible reaching temperatures of several hundreds of ºC with just a few microwatts of electrical power [1]. This so called self-heating effect has mostly been observed in small nanosystems, based on just one or a few 1D nanoparticles, preferently nanotubes or nanowires. In this work, we report on local Joule heating effects observed in large arrangements of 1D nanoparticles in which relatively high temperatures could still be achived with unexpectedly low power consumption figures [2]. Combining thermal micrographs with simulations, we found that the reason why such a low power consumption is also possible in large systems is the formation of random percolation networks for current that lead to extremely localized power dissipation spots (the "hot-spots") [3]. Our experiments and models also explain how these effects relate to the macroscopic resistance signal of the system, and how this electrical resistance can be used to monitor heating. Also, the results obtained with our test system (carbon nanofibers) were also confirmed with other nanosystems of reduced dimensionatily (metal oxide nanowires, reduced graphene oxide, CNTs, etc.). Furthermore, thumb rules to design efficient self-heated systems will be provided and discussed. [1] J.D. Prades, R. Jimenez-Diaz, F. Hernandez-Ramirez, S. Barth, A. Cirera, A. Romano-Rodriguez, S. Mathur, J.R. Morante, Ultralow power consumption gas sensors based on self-heated individual nanowires, Appl. Phys. Lett. 93 (2008) 123110. doi:10.1063/1.2988265. [2] O. Monereo, J.D. Prades, A. Cirera, Self-heating effects in large arrangements of randomly oriented carbon nanofibers: application to gas sensors, Sensors Actuators B Chem. 211 (2015) 489–497. doi:10.1016/j.snb.2015.01.095. [3] O. Monereo, S. Illera, A. Varea, M. Schmidt, T. Sauerwald, A. Schütze, A. Cirera, J.D. Prades, Localized self-heating in large arrays of 1D nanostructures, Nanoscale. 8 (2016) 5082–5088. doi:10.1039/C5NR07158E.

Resume : We explore numerically the morphological patterns of thermo-diffusive instabilities in combustion fronts with a continuum fuel source, within a range of Lewis numbers and ignition temperatures, focusing on the cellular regime. For this purpose, we generalize the recent model of Brailovsky et al. to include distinct process kinetics and reactant heterogeneity. The generalized model is derived analytically and validated with other established models in the limit of infinite Lewis number for zero-order and first-order kinetics. Cellular and dendritic instabilities are found at low Lewis numbers. These are studied using a dynamic adaptive mesh refinement technique that allows very large computational domains, thus allowing us to reduce finite-size effects that can affect or even preclude the emergence of these patterns. Our numerical linear stability analysis is consistent with the analytical results of Brailovsky et al. The distinct types of dynamics found in the vicinity of the critical Lewis number, ranging from steady-state cells to continued tip-splitting and cell-merging, are well described within the framework of thermo-diffusive instabilities and are consistent with previous numerical studies. These types of dynamics are classified as “quasi-linear” and characterized by low amplitude cells that may be strongly affected by the mode selection mechanism and growth prescribed by the linear theory. Below this range of Lewis number, highly non-linear effects become prominent and large amplitude, complex cellular and seaweed dendritic morphologies emerge. The cellular patterns simulated in this work are similar to those reported by Malchi et al. in experiments of flame propagation over a bed of nano-aluminum powder burning with a counter flowing oxidizer. These resemble the dendritic fingers observed in this study, in the limit of low-Lewis number. It is noteworthy that the physical dimension of our computational domain is roughly close to their experimental setup.