Library based identification and characterisation of polymers with nano-FTIR and IR-sSNOM imaging

TL;DR

This study demonstrates that nano-FTIR and IR-sSNOM imaging can reliably identify and characterize polymers at the nanoscale with high resolution, enabling rapid library-based identification in under seven minutes per spectrum.

Contribution

It introduces a method for fast, library-based polymer identification using nano-FTIR and IR-sSNOM, showing comparable accuracy to traditional FTIR despite smaller sampled volumes.

Findings

Successful identification of all tested polymer samples within seven minutes.

Small spectral range (1700-1300 cm$^{-1}$) suffices for differentiating polymers.

Nano-FTIR spectra are comparable to macroscopic FTIR data in polymer analysis.

Abstract

AFM is a technique widely applied in the nanoscale characterisation of polymers and their surface properties. With nano-FTIR and IR-sSNOM imaging an optical dimension is added to this technique that allows for straightforward high resolution characterisation and spectroscopy of polymers. As the volume sampled by these near-field techniques depends mostly on the radius of the cantilever tip, typically 10 nm, it is orders of magnitude smaller than in conventional techniques. Nevertheless, comparability of nano-FTIR near-field spectra and data from macroscopic methods has been shown. Some relevant polymers such as polystyrene however, prove to be more difficult to detect than others. Furthermore, the small sampled volume suggests lower signal quality of nano-FTIR data and proof of its suitability for a reliable library search identification is lacking. To evaluate the techniques especially towards automatic and higher throughput identification of nanoscale polymers, for example in blends or environmental samples, we examined domain distributions in a PS-LDPE film and spectral responses of foils of the most relevant commercial polymers. We demonstrate the successful library search identification of all samples with nano-FTIR data measured in less than seven minutes/spectrum. We discuss aspects affecting the accuracy of the identification and show that already the small spectral range of 1700-1300 cm$^{-1}$ leads to similar success in differentiating between polymer types with near-field data as with conventional far-field FTIR spectroscopy.