Radiative transfer of energy at the nanometre length scale is of great importance to a variety of technologies including heat-assisted magnetic recording, near-field thermophotovoltaics and lithography. Although experimental advances have enabled elucidation of near-field radiative heat transfer in gaps as small as 20-30 nanometres (refs 4-6), quantitative analysis in the extreme near field (less than 10 nanometres) has been greatly limited by experimental challenges. Moreover, the results of pioneering measurements differed from theoretical predictions by orders of magnitude. Here we use custom-fabricated scanning probes with embedded thermocouples, in conjunction with new microdevices capable of periodic temperature modulation, to measure radiative heat transfer down to gaps as small as two nanometres. For our experiments we deposited suitably chosen metal or dielectric layers on the scanning probes and microdevices, enabling direct study of extreme near-field radiation between silica-silica, silicon nitride-silicon nitride and gold-gold surfaces to reveal marked, gap-size-dependent enhancements of radiative heat transfer. Furthermore, our state-of-the-art calculations of radiative heat transfer, performed within the theoretical framework of fluctuational electrodynamics, are in excellent agreement with our experimental results, providing unambiguous evidence that confirms the validity of this theory for modelling radiative heat transfer in gaps as small as a few nanometres. This work lays the foundations required for the rational design of novel technologies that leverage nanoscale radiative heat transfer.

Molecular junctions hold significant promise for efficient and high-power-output thermoelectric energy conversion. Recent experiments have probed the thermoelectric properties of molecular junctions. However, electrostatic control of thermoelectric properties via a gate electrode has not been possible due to technical challenges in creating temperature differentials in three-terminal devices. Here, we show that extremely large temperature gradients (exceeding 1 × 10(9) K m(-1)) can be established in nanoscale gaps bridged by molecules, while simultaneously controlling their electronic structure via a gate electrode. Using this platform, we study prototypical Au-biphenyl-4,4'-dithiol-Au and Au-fullerene-Au junctions to demonstrate that the Seebeck coefficient and the electrical conductance of molecular junctions can be simultaneously increased by electrostatic control. Moreover, from our studies of fullerene junctions, we show that thermoelectric properties can be significantly enhanced when the dominant transport orbital is located close to the chemical potential (Fermi level) of the electrodes. These results illustrate the intimate relationship between the thermoelectric properties and charge transmission characteristics of molecular junctions and should enable systematic exploration of the recent computational predictions that promise extremely efficient thermoelectric energy conversion in molecular junctions.

Atomic and single-molecule junctions represent the ultimate limit to the miniaturization of electrical circuits 1 . They are also ideal platforms to test quantum transport theories that are required to describe charge and energy transfer in novel functional nanodevices.Recent work has successfully probed electric and thermoelectric phenomena 2-8 in atomicscale junctions. However, heat dissipation and transport in atomic-scale devices remain poorly characterized due to experimental challenges. Here, using custom-fabricated scanning probes with integrated nanoscale thermocouples, we show that heat dissipation in the electrodes of molecular junctions, whose transmission characteristics are strongly dependent on energy, is asymmetric, i.e. unequal and dependent on both the bias polarity and the identity of majority charge carriers (electrons vs. holes). In contrast, atomic junctions whose transmission characteristics show weak energy dependence do not exhibit appreciable asymmetry. Our results unambiguously relate the electronic transmission characteristics of atomic-scale junctions to their heat dissipation properties establishing a framework for understanding heat dissipation in a range of mesoscopic systems where transport is elastic. We anticipate that the techniques established here will enable the study of Peltier effects at the atomic scale, a field that has been barely explored experimentally despite interesting theoretical predictions 9-11 . Furthermore, the experimental advances described here are also expected to enable the study of heat transport in atomic and molecular junctions-an important and challenging scientific and technological goal that has remained elusive 12,13 .Charge transport is always accompanied by heat dissipation (Joule heating). This process is well understood at the macroscale where the power dissipation (heat dissipated per unit time) is volumetric and is given by j 2 ρ, where j is the magnitude of the current density and ρ is the In this work, we overcome this challenging experimental hurdle by leveraging custom- Seebeck coefficient of the thermocouple. We note that R P and S TC were experimentally determined to be 72800 ± 500 K/W and 16.3 ± 0.2 µV/K, respectively (see SI).We began our experimental studies, at room temperature, by trapping single molecules of 1,4-benezenediisonitrile (BDNC, see Fig. 1c) between the Au electrodes of the NTISTP and the substrate using a break junction technique 5,20 . We first obtained electrical conductance versus displacement traces by monitoring the electrical current under an applied bias while the NTISTPsubstrate separation was systematically varied. In order to probe heat dissipation we created stable Au-BDNC-Au junctions with a conductance that is within 10% of the most probable low-bias conductance 20 . We studied heat dissipation in 100 distinct Au-BDNC-Au junctions, at each bias, to obtain the average temperature rise (ΔT TC, Avg ) and the time-averaged power dissipation in the NTISTP (Q P, Avg ) for both positive and negative biases. Here,...

Understanding energy dissipation at the nanoscale requires the ability to probe temperature fields with nanometer resolution. Here, we describe an ultra-high vacuum (UHV)-based scanning thermal microscope (SThM) technique that is capable of quantitatively mapping temperature fields with ∼15 mK temperature resolution and ∼10 nm spatial resolution. In this technique, a custom fabricated atomic force microscope (AFM) cantilever, with a nanoscale Au-Cr thermocouple integrated into the tip of the probe, is used to measure temperature fields of surfaces. Operation in an UHV environment eliminates parasitic heat transport between the tip and the sample enabling quantitative measurement of temperature fields on metal and dielectric surfaces with nanoscale resolution. We demonstrate the capabilities of this technique by directly imaging thermal fields in the vicinity of a 200 nm wide, self-heated, Pt line. Our measurements are in excellent agreement with computational results-unambiguously demonstrating the quantitative capabilities of the technique. UHV-SThM techniques will play an important role in the study of energy dissipation in nanometer-sized electronic and photonic devices and the study of phonon and electron transport at the nanoscale.

Because of its high spatial resolution, scanning thermal microscopy (SThM) has been developed quite actively and applied in such diverse areas as microelectronics, optoelectronics, polymers, and carbon nanotubes for more than a decade since the 1990s. However, despite its long history and diverse areas of application, surprisingly, no quantitative profiling method has been established yet. This is mostly due to the nonlocal nature of measurement by conventional SThM: the signal measured by SThM is induced not only from the local heat flux through the tip-sample thermal contact but also (and mostly) from the heat flux through the air gap between the sample and the SThM probe. In this study, a rigorous but simple and practical theory for quantitative SThM for local measurement is established and verified experimentally using high-performance SThM probes. The development of quantitative SThM will make possible new breakthroughs in diverse fields of nanothermal science and engineering.

Motivated by recent experiments, we present here a detailed theoretical analysis of the joule heating in current-carrying single-molecule junctions. By combining the Landauer approach for quantum transport with ab initio calculations, we show how the heating in the electrodes of a molecular junction is determined by its electronic structure. In particular, we show that in general heat is not equally dissipated in both electrodes of the junction and it depends on the bias polarity (or equivalently on the current direction). These heating asymmetries are intimately related to the thermopower of the junction as both these quantities are governed by very similar principles. We illustrate these ideas by analyzing single-molecule junctions based on benzene derivatives with different anchoring groups. The close relation between heat dissipation and 7

Quantitative studies of nanoscale heat dissipation (Joule heating) are essential for advancing nano-science and technology. Joule heating is widely expected to play a critical role in accelerating electromigration induced device failure. However, limitations in quantitatively probing temperature fields—with nanoscale resolution—have hindered elucidation of the role of Joule heating in electromigration. In this work, we use ultra-high vacuum scanning thermal microscopy to directly quantify thermal fields in nanowires during electromigration. Our results unambiguously illustrate that electromigration begins at temperatures significantly lower than the melting temperature of gold. Further, we show that during electromigration voids predominantly accumulate at the cathode resulting in both local hot spots and asymmetric temperature distributions. These results provide novel insights into the microscopic details of hot spot evolution during electromigration and are expected to guide the design of reliable nanoscale functional devices.

Although scanning thermal microscope has shown the highest spatial resolution in local temperature and thermophysical property measurement, its usefulness has been severely limited due to difficulties in quantitative measurement. We propose a double scan technique that measures temperature only from the heat transfer through the tip-sample contact by the subtraction of the signal due to the heat transfer through the air. A rigorous theoretical model for this technique is derived. The effectiveness of the double scan technique in quantitative temperature measurement is demonstrated experimentally.