Quantitative imaging methods for heterogeneous multi-component films

The drying of multi-component dispersions is a common phenomenon in a variety of everyday applications, including coatings, inks, processed foods, and cosmetics. As the solvent evaporates, the different components may spontaneously segregate laterally and/or in depth, which can significantly impact the macroscopic properties of the dried film. To obtain a quantitative understanding of these processes, high-resolution analysis of segregation patterns is crucial. Yet, current state-of-the-art methods are limited to transparent, non-deformable labeled colloids, limiting their applicability. In this study, we employ three techniques that do not require customized samples, as their imaging contrast relies on intrinsic variations in the chemical nature of the constituent species: confocal Raman microscopy, cross-sectional Raman microscopy, and a combination of scanning electron microscopy and energy dispersive X-ray analysis (SEM–EDX). For broad accessibility, we offer a thorough guide to our experimental steps and data analysis methods. We benchmark the capabilities on a film that dries homogeneously at room temperature but exhibits distinct segregation features at elevated temperature, notably self-stratification, i.e., autonomous layer formation, due to a colloidal size mismatch. Confocal Raman microscopy offers a direct means to visualize structures in three dimensions without pre-treatment, its accuracy diminishes deeper within the film, making cross-sectional Raman imaging and SEM–EDX better options. The latter is the most elaborate method, yet we show that it can reveal the most subtle and small-scale microseparation of the two components in the lateral direction. This comparative study assists researchers in choosing and applying the most suitable technique to quantify structure formation in dried multi-component films.

1 Supporting videos Video S1 3D volume view of the integrated confocal Raman signals of the film dried at 20 • C, rotated over the z axis.The latex, silica and the silicon substrate signals are shown in red, blue and green respectively.Note that these colours are overlaid, leading to mixed colours.The scales are the same as in Fig. S1a.At the bottom, the silica signal seems to increase drastically, but this is an artefact related to the oxidation of the silicon substrate.This video shows a 3D representation of the data presented in Fig. 3a-e.

Video S2
3D volume view of the integrated confocal Raman signals of the film dried at 70 • C, rotated over the z axis.The latex, silica and the silicon substrate signals are shown in red, blue and green respectively.Note that these colours are overlaid, leading to mixed colours.The scales are the same as in Fig. S1b.This video shows a 3D representation of the data presented in Fig. 3f-j.

Video S3
3D volume view of ϕ SiO2 of the film dried at 20 • C, rotated over the z axis.These values are derived from the CRM data in Video S1.The colour coding scale and the scale bars are the same as in Fig. S2a.At the bottom, ϕ SiO2 seems to increase drastically, but this is an artefact related to the oxidation of the silicon substrate.This video shows a 3D representation of the data presented in Fig 4b-e.

Video S4
3D volume view of ϕ SiO2 of the film dried at 70 • C, rotated over the z axis.These values are derived from the CRM data in Video S2.The colour coding scale and the scale bars are the same as in Fig. S2b.This video shows a 3D representation of the data presented in Fig. 4f-i.

Fig
Fig. S1.3D volume view of the integrated confocal Raman signals for the sample dried at (a) 20 • C and (b) 70 • C. The latex, silica and the silicon substrate signals are shown in red, blue and green respectively.At the bottom of (a), the silica signal seems to drastically increase, but this is an artefact related to the oxidation of the silicon substrate.For both films, we are able to capture the Raman signals in x y an z throughout the whole depth, and identify clear differences between a homogeneous and heterogeneous composition distribution.The signal is progressively more distorted in the deeper layers, due to scattering by refractive index mismatches, this can be resolved by cross-section Raman microscopy as shown in Fig.5.For the 3D views from different angels, see Video S1 and Video S2.This figure shows a 3D representation of the data presented in Fig.3.

Fig. S3 .
Fig. S3.Typical EDX elemental analysis spectra of different areas in the cross-section of the film dried at 70 • C: (a) close to the top, (b) just below the top, (c) in the middle, and (d) at the bottom.The EDX spectra correspond to the areas indicated by white squares in Fig. S4i-l for a (upper) and b (lower) and figure S4m-p for c (upper) and d (lower).The insets show the corresponding weight fractions of the detected elements.Traces of Al (mounting substrate), Na, K and Ir (sputter-coated element) are excluded from the final analysis of w Si in Fig. 6.The silicon and oxygen signals can be found at the same positions, revealing the silica nanoparticles (which are mainly composed out of Si and O).The carbon signal mainly represents the latex phase.