This recommendation is an output of a small-scale round-robin test involving three different laboratories from TG2 of the RILEM TC 295-FBB: “Fingerprinting bituminous binders using physico-chemical analysis” concerning the use of ¹H-NMR for fingerprinting of bituminous binders. It demonstrates the full capabilities of ¹H-NMR as a robust characterisation tool for complex organic materials, like bituminous binders, to examine their molecular composition in a reproducible way and with the best possible detail. This recommendation documents the key factors in sample preparation and the sensitivity of data post-processing steps. It concludes with best practices and a case study examining the effect of laboratory ageing on two bituminous binders. Overall, it highlights the potential, to the broader scientific community, of another efficient chemometric tool for bituminous binders.

Attenuated Total Reflection Fourier Transform Infrared spectroscopy has become a popular spectroscopic technique in bituminous binder analysis. However, comparable results are not obtainable yet due to differences in devices, measurement routines, sample preparation procedures, and spectral evaluation. Thus, the Task Group 1 of the RILEM TC 295-FBB: “Fingerprinting bituminous binders using physicochemical analysis” focuses on bringing this method towards pre-standardization. This study evaluates the reproducibility and consistency from round robin test, where 21 participating laboratories performed six different preparation techniques on three different binders in an unaged, short-term, and long-term aged state. A total of 6461 spectra were recorded and evaluated for their mean, standard deviation and coefficient of variation (CV) in the spectral region between 1800 and 600 cm⁻¹. The results show that the solid sample preparation methods provide excellent reproducibility, with a coefficient of variation below 2%. Only the solvent method showed a higher coefficient of variation at 7.18%. Outliers with a high CV were detected and categorized into two groups: one where only one of the four samples differed and the other where all 16 spectra showed slight scattering in the overall absorption. The consistency of the method is significantly influenced by the accuracy of sample preparation, which is crucial for minimizing differences in slope, baseline, and noise in the spectra. These findings show the excellent reproducibility of these sample preparation methods and will be further examined to establish universal indices for evaluating effects such as ageing, bringing the method closer towards standardization.