Search

Improving the control of heat flow at the nanoscale is essential for promoting its applications in many fields, such as energy conversion, thermal informatics, and communication technologies. Here we perform a systematic study on the synergistic effect of screw dislocations and surface resonators on thermal transport of Si nanowires and the corresponding mechanisms based on molecular dynamics simulations. We uncover that screw dislocations reduce the thermal conductivity by enhancing the anharmonicity of nanowires due to the nonhomogeneous stress field. For resonant structures, we demonstrate that the suppression of relaxation time is the main mechanism for thermal conductivity reduction. The suppression of relaxation time by more than two orders of magnitude below 4 THz dramatically reduces the resonant structure thermal conductivity, while the previously proposed group velocity reduction mechanism can only impede phonon transport beyond 4 THz slightly. By comparing the mechanisms produced by dislocations and resonators, we find that the resonators have a stronger effect over screw dislocations in impeding the phonon transport at low frequencies while it becomes opposite at high frequencies. As a result, they can be combined together to manipulate phonon transport synergistically at all frequencies. Our findings not only provide insights into the mechanisms of thermal conductivity engineering by screw dislocations and surface resonators, but they also illustrate a paradigm for ultralow thermal conductivity design through the tailoring of the entire frequency range of phonon transport.

Glasses usually represent the lower limit for the thermal conductivity of solids, but a fundamental understanding of lattice heat transport in amorphous materials can provide design rules to beat such a limit. Here we investigate the role of mass disorder in glasses by studying amorphous silicon-germanium alloy (a-Si1-xGex) over the full range of atomic concentration from x=0 to x=1, using molecular dynamics and the quasiharmonic Green-Kubo lattice dynamics formalism. We find that the thermal conductivity of a-Si1-xGex as a function of x exhibits a smoother U shape than in crystalline mass-disordered alloys. The main contribution to the initial drop of thermal conductivity at low Ge concentration stems from the localization of otherwise extended modes that make up the lowest 8% of the population by frequency. Contributions from intermediate frequency modes are decreased more gradually with increasing Ge to reach a broad minimum thermal conductivity between concentrations of Ge from x=0.25 to 0.75. Modal analysis unravels the correlations among localization, line broadening, and the contribution to thermal transport of modes within different frequency ranges.

Understanding heat transport in semiconductors and insulators is of fundamental importance because of its technological impact in electronics and renewable energy harvesting and conversion. Anharmonic lattice dynamics provides a powerful framework for the description of heat transport at the nanoscale. One of the advantages of this method is that it naturally includes quantum effects due to atoms vibrations, which are needed to compute the thermal properties of semiconductors widely used in nanotechnology, like silicon and carbon, even at room temperature. While the heat transport picture substantially differs between amorphous and crystalline semiconductors from a microscopic standpoint, a unified approach to simulate both crystals and glasses has been devised. Here, we introduce a unified workflow, which implements both the Boltzmann Transport equation and the quasi-harmonic Green-Kubo methods. We discuss how the theory can be optimized to exploit modern parallel architectures, and how it is implemented in κALD o: a versatile and scalable open-source software to compute phonon transport in solids. This approach is applied to crystalline and partially disordered silicon-based systems, including bulk silicon and clathrates, and on silicon-germanium alloy clathrates with largely reduced thermal conductivity.

Anisotropic phonon transport is observed along different lattice directions in two-dimensional (2D) and layered materials. However, this effect vanishes in homogeneous, covalently bonded films, as the thickness is increased beyond few atomic layers. Here we establish a fundamental mechanism to induce anisotropic phonon transport in quasi-2D materials with in-plane isotropic symmetry. The anisotropy is engendered by the resonant modes of surface nanostructures that hybridize with membrane modes. Using atomistic lattice dynamics and classical molecular dynamics we demonstrate that the thermal conductivity of silicon membranes with surface nanofins is larger by approximately 100% parallel to the fins than in the perpendicular direction. The primary advantage of these configurations is that they would be technologically viable for implementation to existing and innovative material architectures. We anticipate that our results will open up alternative research directions to control phonons of technology-enabling nanomaterials for a broad range of applications, including thermal management in three-dimensional interconnected nanoelectronics, thermoelectric conversion to IR sensing.

Germanium manganese compounds exhibit a variety of stable and metastable phases with different stoichiometries. These materials entail interesting electronic, magnetic, and thermal properties both in their bulk form and as heterostructures. Here, we develop and validate a transferable machine learning potential, based on the high-dimensional neural network formalism, to enable the study of Mn xGe y materials over a wide range of compositions. We show that a neural network potential fitted on a minimal training set reproduces successfully the structural and vibrational properties and the thermal conductivity of systems with different local chemical environments, and it can be used to predict phononic effects in nanoscale heterostructures.

The thermal conductivity of GeTe crystalline nanowires has been computed by means of non-equilibrium molecular dynamics simulations employing a machine learning interatomic potential. This material is of interest for application in phase change non-volatile memories. The resulting lattice thermal conductivity of an ultrathin nanowire (7.3 nm diameter) of 1.57 W m-1 K-1 is sizably lower than the corresponding bulk value of 3.15 W m-1 K-1 obtained within the same framework. The analysis of the phonon dispersion relations and lifetimes reveals that the lower thermal conductivity in the nanowire is mostly due to a reduction in the phonon group velocities. We further predict the presence of a minimum in the lattice thermal conductivity for thicker nanowires.

Thermoelectric devices convert temperature gradients into electrical power and vice versa, thus enabling energy scavenging from waste heat, sensing, and cooling. Yet, many of these attractive applications are hindered by the limited efficiency of thermoelectric materials, especially in the low-temperature regime. This review provides a summary of the recent advances in the design of new efficient silicon-based thermoelectric materials through nanostructuring, alloying, and chemical optimization, emphasizing the contribution from theory and atomistic modeling.

The possibility of tuning the vibrational properties and the thermal conductivity of layered van der Waals materials either chemically or mechanically paves the way to significant advances in nanoscale heat management. Using first-principles calculations we investigate the modulation of heat transport in MoS2 by lithium intercalation and cross-plane strain. We find that both the in-plane and cross-plane thermal conductivity (, ) of MoS2 are extremely sensitive to both strain and electrochemical intercalation. Combining lithium intercalation and strain, the in-plane and cross-plane thermal conductivity can be tuned over one and two orders of magnitude, respectively. Furthermore, since and respond in different ways to intercalation and strain, the thermal conductivity anisotropy can be modulated by two orders of magnitude. The underlying mechanisms for such large tunability of the anisotropic thermal conductivity of MoS2 are explored by computing and analyzing the dispersion relations, group velocities, relaxation times and mean free paths of phonons. Since both intercalation and strain can be applied reversibly, their stark effect on thermal conductivity can be exploited to design novel phononic devices, as well as for thermal management in MoS2-based electronic and optoelectronic systems.

Step junctions are often present in layered materials, i.e. where single-layer regions meet multilayer regions, yet their effect on thermal transport is not understood to date. Here, we measure heat flow across graphene junctions (GJs) from monolayer-to-bilayer graphene, as well as bilayer to four-layer graphene for the first time, in both heat flow directions. The thermal conductance of the monolayer-bilayer GJ ranges from ∼0.5 to 9.1 × 108 W m-2 K-1 between 50 K to 300 K. Atomistic simulations of such a GJ device reveal that graphene layers are relatively decoupled, and the low thermal conductance of the device is determined by the resistance between the two distinct graphene layers. In these conditions the junction plays a negligible effect. To prove that the decoupling between layers controls thermal transport in the junction, the heat flow in both directions was measured, showing no evidence of thermal asymmetry or rectification, within experimental error bars. For large-area graphene applications, this signifies that small bilayer (or multilayer) islands have little or no contribution to overall thermal transport.

The potential application of monolayer MS2 (M = Mo, W) as thermoelectric material has been widely studied since the first report of successful fabrication. However, their performances are hindered by the considerable band gap and the large lattice thermal conductivity in the pristine 2H phase. Recent discoveries of polymorphism in MS2s provide new opportunities for materials engineering. In this work, phonon and electron transport properties of both 2H and 1T′ phases were investigated by first-principle calculations. It is found that upon the phase transition from 2H to 1T′ in MS2, the electron transport is greatly enhanced, while the lattice thermal conductivity is reduced by several times. These features lead to a significant enhancement of power factor by one order of magnitude in MoS2 and by three times in WS2. Meanwhile, the figure of merit can reach up to 0.33 for 1T′–MoS2 and 0.68 for 1T′–WS2 at low temperature. These findings indicate that monolayer MS2 in the 1T′ phase can be promising materials for thermoelectric devices application. Meanwhile, this work demonstrates that phase engineering techniques can bring in one important control parameter in materials design.

We characterized the structural and dynamical properties of the quasi-liquid layer (QLL) at the surface of ice by molecular dynamics simulations with a thermodynamically consistent water model. Our simulations show that for three low-index ice surfaces, only the outermost molecular layer presents short-range and midrange disorder and is diffusive. The onset temperature for normal diffusion is much higher than the glass temperature of supercooled water, although the diffusivity of the QLL is higher than that of bulk water at the corresponding temperature. The underlying subsurface layers impose an ordered template, which produces a regular patterning of the ice/water interface at any temperature and is responsible for the major differences between QLL and bulk water, especially for what concerns the dynamics and the midrange structure of the hydrogen-bonded network. Our work highlights the need for a holistic approach to the characterization of QLL, as a single experimental technique may probe only one specific feature, missing part of the complexity of this fascinating system.

Halogenated aromatic molecules assemble on surfaces forming both hydrogen and halogen bonds. Even though these systems have been intensively studied on flat metal surfaces, high-index vicinal surfaces remain challenging, as they may induce complex adsorbate structures. The adsorption of 2,6-dibromoanthraquinone (2,6-DBAQ) on flat and stepped gold surfaces is studied by means of van der Waals corrected density functional theory. Equilibrium geometries and corresponding adsorption energies are systematically investigated for various different adsorption configurations. It is shown that bridge sites and step edges are the preferred adsorption sites for single molecules on flat and stepped surfaces, respectively. The role of van der Waals interactions, halogen bonds and hydrogen bonds are explored for a monolayer coverage of 2,6-DBAQ molecules, revealing that molecular flexibility and intermolecular interactions stabilize two-dimensional networks on both flat and stepped surfaces. Our results provide a rationale for experimental observation of molecular carpeting on high-index vicinal surfaces of transition metals.

Thermal decomposition is a promising route for the synthesis of highly monodisperse magnetite nanoparticles. However, the apparent simplicity of the synthesis is counterbalanced by the complex interplay of the reagents with the reaction variables that determine the final particle size and dispersity. Here, we present a combined experimental and theoretical study on the influence of the heating rate on crystal growth, size, and monodispersity of iron oxide nanoparticles. We synthesized monodisperse nanoparticles with sizes varying from 6.3 to 27 nm simply by controlling the heating rate of the reaction. The nanoparticles show size-dependent superparamagnetic behavior. Using numerical calculations based on the classical nucleation theory and growth model, we identified the relative time scales associated with the heating rate and precursor-to-monomer (growth species) conversion rate as a decisive factor influencing the final size and dispersity of the nanoparticles.

Steps at high-index metallic surfaces display higher chemical reactivity than close-packed surfaces and may give rise to selective adsorption and partial dissociation of water. Inspired by differential desorption experiments, we have studied the adsorption and dissociation of water clusters and one-dimensional wires on Pt(211) by density functional theory and molecular dynamics simulations. These calculations reveal that water at the step edges of Pt(211) adsorbs more weakly than at Pt(221), but partial dissociation of adsorbed water clusters is energetically competitive. We observe that the one-dimensional structure proposed experimentally can be realized only by partially dissociated water wires. In addition, weaker adsorption allows the formation of structures in which a number of water molecules detach from the step and form weak hydrogen bonds with the terrace. This study is further extended to the energetics of small water clusters on (211) surfaces of Ir, Rh, and Pd.

A detailed understanding of the relation between microscopic structure and phonon propagation at the nanoscale is essential to design materials with desired phononic and thermal properties. Here we uncover a new mechanism of phonon interaction in surface oxidized membranes, i.e., native oxide layers interact with phonons in ultrathin silicon membranes through local resonances. The local resonances reduce the low frequency phonon group velocities and shorten their mean free path. This effect opens up a new strategy for ultralow thermal conductivity design as it complements the scattering mechanism which scatters higher frequency modes effectively. The combination of native oxide layer and alloying with germanium in concentration as small as 5% reduces the thermal conductivity of silicon membranes to 100 times lower than the bulk. In addition, the resonance mechanism produced by native oxide surface layers is particularly effective for thermal conductivity reduction even at very low temperatures, at which only low frequency modes are populated.

Two-dimensional materials have unusual phonon spectra due to the presence of flexural (out-of-plane) modes. Although molecular dynamics simulations have been extensively used to study heat transport in such materials, conventional formalisms treat the phonon dynamics isotropically. Here, we decompose the microscopic heat current in atomistic simulations into in-plane and out-of-plane components, corresponding to in-plane and out-of-plane phonon dynamics, respectively. This decomposition allows for direct computation of the corresponding thermal conductivity components in two-dimensional materials. We apply this decomposition to study heat transport in suspended graphene, using both equilibrium and nonequilibrium molecular dynamics simulations. We show that the flexural component is responsible for about two-thirds of the total thermal conductivity in unstrained graphene, and the acoustic flexural component is responsible for the logarithmic divergence of the conductivity when a sufficiently large tensile strain is applied.

Molecular junctions are promising candidates for thermoelectric devices due to the potential to tune the electronic and thermal transport properties. However, a high figure of merit is hard to achieve, without reducing the phononic contribution to thermal conductance. Here, we propose a strategy to suppress phonon transport in graphene-based molecular junctions preserving high electronic power factor, using nonbonded π-stacked systems. Using first-principles calculations, we find that the thermal conductance of π-stacked systems can be reduced by about 95%, compared with that of a covalently bonded molecular junction. Phonon transmission of π-stacked systems is largely attenuated in the whole frequency range, and the remaining transmission occurs mainly below 5 THz, where out-of-plane channels dominate. The figure of merit (ZT) of the π-stacked molecular junction is dramatically enhanced because of the very low phononic thermal conductance, leaving room for further optimization of the electronic properties. (Figure Presented).

In adaptive resolution simulations the same system is concurrently modeled with different resolution in different subdomains of the simulation box, thereby enabling an accurate description in a small but relevant region, while the rest is treated with a computationally parsimonious model. In this framework, electrostatic interaction, whose accurate treatment is a crucial aspect in the realistic modeling of soft matter and biological systems, represents a particularly acute problem due to the intrinsic long-range nature of Coulomb potential. In the present work we propose and validate the usage of a short-range modification of Coulomb potential, the Damped shifted force (DSF) model, in the context of the Hamiltonian adaptive resolution simulation (H-AdResS) scheme. This approach, which is here validated on bulk water, ensures a reliable reproduction of the structural and dynamical properties of the liquid, and enables a seamless embedding in the H-AdResS framework. The resulting dual-resolution setup is implemented in the LAMMPS simulation package, and its customized version employed in the present work is made publicly available.

Silicon nanostructures with reduced dimensionality, such as nanowires, membranes, and thin films, are promising thermoelectric materials, as they exhibit considerably reduced thermal conductivity. Here, we utilize density functional theory and Boltzmann transport equation to compute the electronic properties of ultra-thin crystalline silicon membranes with thickness between 1 and 12 nm. We predict that an optimal thickness of ∼7 nm maximizes the thermoelectric figure of merit of membranes with native oxide surface layers. Further thinning of the membranes, although attainable in experiments, reduces the electrical conductivity and worsens the thermoelectric efficiency.

Elemental two-dimensional (2D) materials exhibit intriguing heat transport and phononic properties. Here we have investigated the lattice thermal conductivity of newly proposed arsenene, the 2D honeycomb structure of arsenic, using ab initio calculations. Solving the Boltzmann transport equation for phonons, we predict a highly anisotropic thermal conductivity of 30.4 and 7.8 W/mK along the zigzag and armchair directions, respectively, at room temperature. Our calculations reveal that phonons with mean free paths between 20 nm and 1μm provide the main contribution to the large thermal conductivity in the zigzag direction; mean free paths of phonons contributing to heat transport in the armchair directions range between 20 and 100 nm. The obtained anisotropic thermal conductivity and feasibility of synthesis, in addition to high electron mobility reported elsewhere, make arsenene a promising material for nanoelectronic applications and thermal management.

Adsorption of aromatic molecules on surfaces is widely studied due to applications in molecular electronics. In this work, the adsorption of iodobenzene molecules on the Pt(111) surface has been studied using density functional theory. Iodobenzene molecules, with various numbers of iodine atoms, have two non-dissociative adsorption minima. One structure exhibits chemisorption between the ring and the surface (short-range) and the other structure exhibits chemisorption between the iodine ions and the surface (long-range). Both structures have a strong van der Waals interaction with the surface. In general, the adsorption energy increases as the number of iodine atoms increases. The dissociated structure of monoiodobenzene was investigated, and the dissociation barrier and the barrier between the short- and long-range states were compared.

Understanding and controlling vibrations in condensed matter is emerging as an essential necessity both at fundamental level and for the development of a broad variety of technological applications. Intelligent design of the band structure and transport properties of phonons at the nanoscale and of their interactions with electrons and photons impact the efficiency of nanoelectronic systems and thermoelectric materials, permit the exploration of quantum phenomena with micro- and nanoscale resonators, and provide new tools for spectroscopy and imaging. In this colloquium we assess the state of the art of nanophononics, describing the recent achievements and the open challenges in nanoscale heat transport, coherent phonon generation and exploitation, and in nano- and optomechanics. We also underline the links among the diverse communities involved in the study of nanoscale phonons, pointing out the common goals and opportunities.

The discovery of nanostructures and the development of growth and fabrication techniques of one- and two-dimensional materials provide the possibility to probe experimentally heat transport in low-dimensional systems. Nevertheless measuring the thermal conductivity of these systems is extremely challenging and subject to large uncertainties, thus hindering the chance for a direct comparison between experiments and statistical physics models. Atomistic simulations of realistic nanostructures provide the ideal bridge between abstract models and experiments. After briefly introducing the state of the art of heat transport measurement in nanostructures, and numerical techniques to simulate realistic systems at atomistic level, we review the contribution of lattice dynamics and molecular dynamics simulation to understanding nanoscale thermal transport in systems with reduced dimensionality. We focus on the effect of dimensionality in determining the phononic properties of carbon and semiconducting nanostructures, specifically considering the cases of carbon nanotubes, graphene and of silicon nanowires and ultra-thin membranes, underlying analogies and differences with abstract lattice models.

Understanding and controlling heat transport in molecular junctions would provide new routes to design nanoscale coupled electronic and phononic devices. Using first-principles full quantum calculations, we tune thermal conductance of a molecular junction by mechanically compressing and extending a short alkane chain connected to graphene leads. We find that the thermal conductance of the compressed junction drops by half in comparison to the extended junction, making it possible to turn on and off the heat current. The low conductance of the off state does not vary by further approaching the leads and stems from the suppression of the transmission of the in-plane transverse and longitudinal channels. Furthermore, we show that misalignment of the leads does not reduce the conductance ratio. These results also contribute to the general understanding of thermal transport in molecular junctions.

Crystallisation of liquid solutions is of uttermost importance in a wide variety of processes in materials, atmospheric and food science. Depending on the type and concentration of solutes the freezing point shifts, thus allowing control on the thermodynamics of complex fluids. Here we investigate the basic principles of solute-induced freezing point depression by computing the melting temperature of a Lennard-Jones fluid with low concentrations of solutes, by means of equilibrium molecular dynamics simulations. The effect of solvophilic and weakly solvophobic solutes at low concentrations is analysed, scanning systematically the size and the concentration. We identify the range of parameters that produce deviations from the linear dependence of the freezing point on the molal concentration of solutes, expected for ideal solutions. Our simulations allow us also to link the shifts in coexistence temperature to the microscopic structure of the solutions.

We derive expressions of interatomic force and heat current for many-body potentials such as the Tersoff, the Brenner, and the Stillinger-Weber potential used extensively in molecular dynamics simulations of covalently bonded materials. Although these potentials have a many-body nature, a pairwise force expression that follows Newton's third law can be found without referring to any partition of the potential. Based on this force formula, a stress applicable for periodic systems can be unambiguously defined. The force formula can then be used to derive the heat current formulas using a natural potential partitioning. Our heat current formulation is found to be equivalent to most of the seemingly different heat current formulas used in the literature, but to deviate from the stress-based formula derived from two-body potential. We validate our formulation numerically on various systems described by the Tersoff potential, namely three-dimensional silicon and diamond, two-dimensional graphene, and quasi-one-dimensional carbon nanotube. The effects of cell size and production time used in the simulation are examined.

The presence of oligopeptides often has a drastic impact on the nucleation and crystal growth of calcium-containing minerals. Molecular simulation of complexes of peptides with Ca2+ ions can give mechanistic insight into how the pre-alignment of ions from solution by additives may steer the crystallization of calcium rich minerals into different polymorphs. We exploit advanced molecular modeling techniques to sample and analyze the complex multidimensional configurational space of aqueous solutions of oligo-glutamate and Ca2+ ions. We find that glutamate oligomers induce the formation of patterns in the associated Ca2+ ions in solution, which can serve as templates for the growth of different calcium oxalate pseudopolymorphs - depending on the peptide length, in excellent agreement with experimental observations. Glutamate decamers prestructure the Ca2+ ions in remarkable correspondence with typical distances in calcium oxalate dihydrate. These are distinct from those in calcium oxalate trihydrate which typically grows in the absence of peptide or in the presence of shorter glutamate oligomers such as pentamers.

Nanostructuring provides a viable route to improve the thermoelectric performance of materials, even of those that in bulk form have very low figure of merit. This strategy would potentially enable the fabrication of thermoelectric devices based on silicon, the cheapest, most integrable and easiest to dope Earth-abundant semiconductor. A drastic reduction of the thermal conductivity, which would lead to a proportional enhancement of the figure of merit, was observed for silicon low-dimensional nanostructures, such as nanowires and ultra-thin membranes. Here we provide a detailed analysis of the phononic properties of the latter, and we show that dimensionality reduction alone is not sufficient to hinder heat transport to a great extent. In turn, the presence of surface roughness at the nanoscale reduces the thermal conductivity of sub-10 nm membranes up to 10 times with respect to bulk.

Phonon dispersion relations and electron-phonon coupling of hole-doped trigonal GeTe have been computed by density functional perturbation theory. This compound is a prototypical phase change material of interest for applications in phase change non-volatile memories. The calculations allowed us to estimate the electron-phonon contribution to the thermal boundary resistance at the interface between the crystalline and amorphous phases present in the device. The lattice contribution to the thermal boundary resistance has been computed by non-equilibrium molecular dynamics simulations with an interatomic potential based on a neural network scheme. We find that the electron-phonon term contributes to the thermal boundary resistance to an extent which is strongly dependent on the concentration and mobility of the holes. Further, for measured values of the holes concentration and electrical conductivity, the electron-phonon term is larger than the contribution from the lattice. It is also shown that the presence of Ge vacancies, responsible for the p-type degenerate character of the semiconductor, strongly affects the lattice thermal conductivity of the crystal.

Understanding the structure and chemical reactivity of water adsorbed at metallic surfaces is very important in many processes such as catalysis, corrosion, and electrochemistry. Using density functional theory calculations, we investigate the adsorption and dissociation of water clusters on flat and stepped surfaces of several transition metals: Rh, Ir, Pd, and Pt. We find that water binds preferentially to the step edges than to terrace sites, thus linear clusters or one-dimensional water wires can be isolated by differential desorption. The clusters formed at the step are stabilized by the cooperative effect of chemical bonds with the metal and hydrogen bonding. The enhanced reactivity of the step edges and the cooperative effect of hydrogen bonding improve the chances of partial dissociation of water clusters. We assess the correlations between adsorption and dissociation energies, observing that they are increased on stepped surfaces. We present a detailed interpretation of water dissociation by analyzing changes in the electronic structure of both water and metals. The identification of trends in the energetics of water dissociation at transition metals is expected to aid the design of better materials for catalysis and fuel cells, where the density of steps at surfaces would be a relevant additional parameter.

Carbon atoms can form structures in one, two and three dimensions due to their unique chemical versatility. In terms of thermal conductivity, carbon polymorphs cover a wide range from very low values with amorphous carbon to very high values with diamond, carbon nanotubes and graphene. Schwarzites are a class of three-dimensional fully covalent sp2-bonded carbon polymorphs, with the same local chemical environment as graphene and carbon nanotubes, but negative Gaussian curvature. We calculate the thermal conductivity of a (10,0) carbon nanotube, graphene and two schwarzites with different curvature, by molecular dynamics simulations based on the Tersoff empirical potential. We find that schwarzites present a thermal conductivity two orders of magnitude smaller than nanotubes and graphene. The reason for such large difference is explained by anharmonic lattice dynamics calculations, which show that phonon group velocities and mean free paths are much smaller in schwarzites than in nanotubes and graphene. Their reduced thermal conductivity, in addition to tunable electronic properties, indicate that schwarzites could pave the way towards all-carbon thermoelectric technology with high conversion efficiency. © IOP Publishing and Deutsche Physikalische Gesellschaft.