Thermal insulation for quantum technologies

Electron microscope image of a silicon sample with nanocrystallites of different orientation (coloured), pores (black) and the pore network recognised by the image processing software (white grid lines).

Electron microscope image of a silicon sample with nanocrystallites of different orientation (coloured), pores (black) and the pore network recognised by the image processing software (white grid lines). © D. Kojda / HZB

New energy-efficient IT components often only operate stably at extremely low temperatures. Therefore, very good thermal insulation of such elements is crucial, which requires the development of materials with extremely low thermal conductivity. A team at HZB has now used a novel sintering process to produce nanoporous silicon aluminium samples in which pores and nanocrystallites impede the transport of heat and thus drastically reduce thermal conductivity. The researchers have developed a model for predicting the thermal conductivity, which was confirmed using experimental data on the microstructure of the samples and their thermal conductivity. Thus, for the first time, a method is available for the targeted development of complex porous materials with ultra-low thermal conductivity.


Thermal insulation is not only important for buildings, but also in quantum technologies. While insulation panels around a house keep the heat inside, quantum devices require insulation against heat from the outside world, as many quantum effects are only stable at low temperatures. What is needed are materials with extremely low thermal conductivity that are also compatible with the materials used in quantum technology.

Novel sintering process

A team led by Dr Klaus Habicht from HZB has now taken a big step forward in this direction. Using a novel sintering process, they produced samples of silicon and silicon aluminium that were compacted under pressure and an electric field for a few minutes at high temperature. Before that, further microstructures were added to the Si starting material using electrochemical etching processes, which further suppress heat transport. "Silicon is the ideal material here for many reasons, in particular it fits possible devices based on silicon qubits," Habicht points out.

Obstacles for phonon transport

This gave them a number of material samples with tiny pores, crystalline nanoparticles and so-called domain boundaries. Heat conduction works via vibrations of the crystal lattice, so called phonons. However, these phonons can only propagate if they do not encounter obstacles on which they are scattered. Pores as well as nanoparticles and domain boundaries with suitable distances and diameters can become such scattering centres and thus reduce heat conduction.

Separating the contributions

Using an elegant model, the scientists calculated the behaviour of the phonons and thus the thermal conductivity in different samples. Their microstructure was incorporated with parameters such as the size and spacing of pores and nanoparticles. "With this model, we can clearly separate the contributions of nanoparticles and pores to thermal conductivity," Habicht explains.

Microstructures evaluated in detail

The experimental results on microstructures and thermal conductivity in the individual samples confirm the new model. First author Danny Kojda determined the microstructures at the HZB's scanning electron microscope. Using special image analysis software, developed by Kojda for this purpose, he determined the size and number of nanoparticles and pores, as well as their spacing. The thermal conductivity as a function of temperature was carefully measured in all samples. The measured data matched the modelled results extremely well. This means that it is now possible to determine whether, in a sample with a given microstructure, it is mainly the pores or rather the nanocrystallites that are responsible for the suppression of heat conduction.

Materials design

"Understanding the basic transport processes helps us to produce and further develop customised materials with strongly reduced thermal conductivity in a targeted manner," says Kojda.


You might also be interested in

  • Calculating the "fingerprints" of molecules with artificial intelligence
    Science Highlight
    Calculating the "fingerprints" of molecules with artificial intelligence
    With conventional methods, it is extremely time-consuming to calculate the spectral fingerprint of larger molecules. But this is a prerequisite for correctly interpreting experimentally obtained data. Now, a team at HZB has achieved very good results in significantly less time using self-learning graphical neural networks.
  • Water distribution in the fuel cell made visible in 4D
    Science Highlight
    Water distribution in the fuel cell made visible in 4D
    Teams from Helmholtz-Zentrum Berlin (HZB) and University College London (UCL) have visualised the water distribution in a fuel cell in three dimensions and in real time for the first time by evaluating neutron data from the Berlin Experimental Reactor shut down in 2019. The analysis opens new possibilities for more efficient and thus more cost-effective fuel cells.
  • Magnetic nanoparticles in biological vehicles individually characterised
    Science Highlight
    Magnetic nanoparticles in biological vehicles individually characterised
    Magnetic nanostructures are promising tools for medical applications.  Incorporated into biological structures, they can be steered via external magnetic fields inside the body to release drugs or to destroy cancer cells. However, until now, only average information on the magnetic properties of those nanoparticles could be obtained, thus limiting their successful implementations in therapies. Now a team at HZB conceived and tested a new method to assess the characteristic parameters of every single magnetic nanoparticle.