Spectral Superdiffusive Phonon Sensor – SPiDER-man
Spectral Superdiffusive Phonon Sensor
SPiDER-man is devoted to the development of a new instrumentation to probe ballistic-diffusive transfer in nanostructured semiconductors. Over the past decade, rapid improvement of experimental techniques devoted to nanomaterial characterization has arisen. Yet, if those approaches allow spectacular temporal or spatial resolution of heat transfer at nanoscales, they can scarcely describe the ballistic-diffusive regime that exists when length scale and energy carriers mean free path are similar.
SPSS - A new instrumentation for probing superdiffusive phonons
Three challenging tasks shall tackle several scientific issues, especially in the development and the calibration of the SPSS. Among them, there is the handling of a TDTR heterodyne device to work at frequencies up to 10 THz, where thermal wave propagation dominates, keeping low signal to noise ratio. Another critical point is to produce artificial Lévy materials mixing alloy and nano-inclusions which will be responsible of superdiffusive behavior with respect of the modelling pathways. Modelling issues are also possible bottlenecks to the achievement of the project as nanoscale heat transport simulation often imply to use large computational resources and lies on physics which is not fully understood for some crystalline materials and alloys. The latter points are considered in the present project and alternative paths are considered to overcome such issues.<br />In addition to the development of a new characterization device and models for nanoscale material thermal properties characterization, the SPiDER-man project aims to deeply investigate innovative nanomaterials that will be useful for several technologies like electronics, energy, sensors, etc. Examples of that could be improved cooling of electronic devices, thermoelectric materials with high figure of merit, development of ultrafast magnetic switching materials, etc.
First, the design of a new instrumentation to probe superdiffusive phonon transport as a Spectral Phonon Superdiffusive Sensor. The core of the project lies on this instrumentation, which will be a significant improvement to the classical Time Domain Thermoreflectance (TDTR) experimentation. In this device, implementation of several features allows investigating thermal transport on a broad frequency range from the kHz to tens of THz. This shall accurately capture the transition between ballistic and diffusive transport that are at the core of this proposal.
Secondly, the design and elaboration by MBE of tailored nanomaterials that exhibit superdiffusive behavior. The development of the SPSS device has to be in-line with the elaboration of model superdiffusive materials. We will produce crystalline and alloy thin layers that can incorporate nano-inclusions (based on Si, Ge, Mn and Sn). Modifying the phases (crystalline/alloy) and the size distribution of nano-inclusions, we will intent to tailor phonon mean free path at different scales to enhance the expected superdiffusion. At the end of the project, similarly to what have been in done in optics, the aim of the project is to propose a new family of artificial “Lévy materials” for phonons.
Finally, the development of theoretical tools valuable to understand and predict results of experiments on ballistic-diffusive regime with respect to Lévy framework. On one hand, output data of the SPSS apparatus will have to be post-treated to derive material properties (thermal conductivity, superdiffusive parameter, boundary thermal resistances, etc.). On the other hand, heat transport in Lévy materials will be considered in the frame of Monte Carlo modelling approach. This modelling will be useful to support the MBE material design giving. Output of this part of the project also lies on the predictive ability to design semiconductor devices with tailored properties for specific applications.
Concerning the experimental development of the SPSS based on Time-Domain Thermoreflectance laser setup. The accessible range of probing frequencies extends from low frequencies (below 1MHz) up to THz. Nevertheless, the metallic transducer which is used to cap the samples acts as a “thermal” low-pass filter. We studied various metallic transducers in order to find the optimum transducer. We have shown that an Al film (less than 100nm) is the best choice for probing phonons in a frequency range from 100kHz up to 10GHz. We have also investigated the effect of Au transducer. We have shown that, in that case, the SPSS captures mostly the electron dynamics from 1GHz up to 5THz. In the future, the instrument we have developed will be used to study ultrafast energy transfer between electrons and phonons. Additionally, we have developed a second innovative SPSS using continuous lasers. This new setup will (i) provide a solid comparison between the continuous (CW) and pulsed approaches and (ii) despite a narrower bandwidth (up to 200MHz), provide a better frequency resolution for studying superdiffusive thermal transport.
We worked on improving the 3 omega measurements which can provide a deeper understanding of phonon transport in nanostructured semiconductors. We have now a much better understanding of the contribution of thermal boundary resistances (TBR) between the different layers of the samples (thermometer/insulating layer).
We have applied Monte-carlo methods to study phonon ballistic-diffusive regimes. Starting from random motion of energy carriers, we have investigated ballistic-diffusive transitions in silicon and germanium materials by changing the phonon population, the mean-free path, etc. We have tried to link these calculations to statistical analysis based on Lévy “heavy-tail” distributions.
Outstanding results can be underlined during the first year of the project. First, seeking of an optimal metallic transducer has shown that it was possible to study electron dynamics to frequencies up to 5THz. While for phonon dynamics, the best material (despite a smaller upper frequency) is a 100nm thick aluminum transducer. A second striking result is the development of a broadband frequency domain thermoreflectance setup that is more “user-friendly” than the original proposed SPSS. Also, promising improvements of the 3 omega allows to precisely evaluate Thermal Boundary Resistance (TBR). This will allow to perform preliminary thermal property measurements on the Lévy model materials to evaluate. Finally, we successfully modify our Monte-Carlo approach to explore the ballistic-diffusive Lévy regime of thermal transport.
1.A. Zenji, J. M. Rampnoux, S. Grauby, S. Dilhaire, Ultimate-resolution thermal spectroscopy in time domain thermoreflectance (TDTR), Journal of Applied Physics, 128, 065106, 2020.
2.G. Pernot, A. Metjari, H. Chaynes, M. Weber, M. Isaiev, D. Lacroix, Frequency domain analysis of 3 omega scanning thermal microscope probe – application to tip/surface thermal interface measurements in vacuum environment, 129, 055105, 2021.
The main goals of SPiDER-man project are to develop a new instrumentation to probe ballistic-diffusive energy transfer in nanostructured semiconductors which can be captured by the frame of Lévy Walks theory (i.e. superdiffusive regime). Over the past decade, rapid improvement of experimental techniques devoted to nanomaterial characterization has arisen: Time-Domain Thermoreflectance (TDTR) and Scanning Thermal Microscopy are now well-established techniques to probe energy transport in nanostructured materials. Yet, if those approaches respectively allow spectacular temporal or spatial resolution of heat transfer at nanoscales, they can scarcely describe the ballistic-diffusive regime that exists when length scale and energy carriers mean free path are similar. This represents a longstanding problem in nanoscale thermal transport for several engineering applications. The scientific program of SPiDER-man project gather the expertise of three academic partners to address those issues through three main workpackages:
1.Design of a new instrumentation to probe superdiffusive phonon transport called Spectral Phonon Superdiffusion Sensor (SPSS); the core of the project lies on this new instrumentation, which will be a significant improvement to TDTR experimentation. In this device, implementation of several new technical features allows investigating thermal transport on a broad frequency range up to 10 THz. It will provide a complete description of the phonon transport properties as a function of frequency and temperature. This shall accurately highlight the transition between ballistic and diffusive effects that are at the core of this proposal.
2.Design and elaborate by MBE enhanced superdiffusive nanomaterials; development of the SPSS device has to be in-line with the elaboration of model materials that exhibit ballistic-diffusive behavior. We will use MBE and annealing post-treatment to produce crystalline and alloy thin layers that incorporate a tailored size distribution of nano-inclusions (5-50nm). We intent to alter phonon mean free path at different scales to enhance the expected superdiffusion. The aim of the project is to propose a new family of artificial “Lévy materials” for phonons based on silicon and/or germanium technologies that could be used in micro-devices heat management, thermoelectric, etc.
3.Develop all numerical tools valuable to understand and predict results of experiments on ballistic-diffusive regime; last objective of the project is to develop new metrology (model and simulation tools) to support the development of the two experimental parts of the project. First, output data of the SPSS apparatus will have to be post-treated to derive thermal conductivity, superdiffusive parameter, boundary thermal resistances, etc. The resulting data will be input for the development of Lévy walks model, crucial to understand ballistic-diffusive behavior of elaborated materials. Secondly, heat transport in Lévy materials will be considered in the frame of Monte Carlo modelling which has proved to be well-suited for the simulation of phonon transport in complex nanostructured materials. The ultimate output of this part of the project lies on the predictive ability to design semiconductor devices with tailored properties for specific applications.
These three challenging tasks shall tackle several scientific issues, especially in the development and the calibration of the SPSS. In addition to the development of a new characterization device and metrology for nanoscale material thermal properties characterization, the SPiDER-man project aims to deeply investigate innovative nanomaterials that will be useful for several technologies like electronics, energy, sensors, etc.
Monsieur Gilles Pernot (Laboratoire d'énergétique et de mécanique théorique et appliquée)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LOMA LABORATOIRE ONDES ET MATIERE D'AQUITAINE
INEEL Institut Néel - CNRS
LEMTA Laboratoire d'énergétique et de mécanique théorique et appliquée
Help of the ANR 457,686 euros
Beginning and duration of the scientific project: December 2018 - 42 Months