Logarithmic sensitivity ratio elucidates thermal transport physics in multivariate thermoreflectance experiments

Area of Science:

• Materials Science
• Condensed Matter Physics
• Nanotechnology

Background:

• Electronic devices increasingly use heterogeneous nanolayers with unknown thermal properties.
• Simultaneous fitting of multiple unknown thermal parameters in experiments like frequency-domain thermoreflectance (FDTR) can yield unreliable results due to parameter correlations.
• Existing methods for identifying these correlations are inadequate, hindering accurate and rapid characterization of multilayer thermal transport.

Purpose of the Study:

• To develop a reliable sensitivity-based method for identifying thermal transport physics and suitable parameters for multivariate nonlinear fits in FDTR experiments.
• To address the unreliability of current methods in characterizing multilayer thermal properties.

Main Methods:

• Mathematical derivation of a sensitivity-based method.
• Development of the logarithmic sensitivity ratio (LSR) analysis to evaluate parameter fitting feasibility and clarify thermal transport physics.
• Validation using Monte Carlo simulations and experimental FDTR data for fitting up to three unknown parameters.

Main Results:

• The proposed LSR analysis effectively evaluates the feasibility of fitting pairs of unknown thermal parameters.
• The method successfully clarifies the governing thermal transport physics in complex nanolayered structures.
• Demonstrated effectiveness and convenience in fitting up to three unknown parameters using simulations and experimental data.

Conclusions:

• The LSR analysis provides a reliable approach to overcome parameter correlation issues in FDTR experiments.
• This method enhances the accuracy and speed of thermal property characterization for multilayer stacks.
• The principle of LSR analysis is adaptable to other techniques requiring multivariate fits.