Combining STEM and PiFM for Complementary Nanoscale Sample Analysis

Introduction

Scanning Transmission Electron Microscopy (STEM) is a premier imaging and analytical technique that combines the high spatial resolution of transmission electron microscopy with the precision of a focused electron probe. In STEM, a highly focused electron beam – often smaller than a single atom – is rastered across an ultrathin sample. Multiple detectors collect signals from transmitted and scattered electrons, as well as characteristic X-rays, enabling the acquisition of structural, elemental, and compositional information at sub-nanometer resolution.

This capability to image individual atomic columns, generate detailed elemental maps, and perform correlative nanoscale analyses has established STEM as a cornerstone tool in materials science, semiconductor research, and nanotechnology.

Despite its strengths, STEM has inherent limitations. The intense electron beam can damage sensitive samples, such as biological or soft materials, potentially altering surface morphology and limiting subsequent analyses. Moreover, STEM primarily provides elemental rather than molecular information, restricting its ability to fully characterize chemical composition.

Molecular Vista’s Photo-Induced Force Microscopy (PiFM) complements STEM by offering non-destructive chemical analysis with nanoscale resolution. PiFM uses a precisely tuned infrared (IR) laser to excite molecular vibrations in a defined region of the sample surface, while a sensitive atomic force microscope (AFM) probe measures the resulting photo-induced force along with its standard topography data. By operating in non-contact mode and leveraging the second mechanical resonance, PiFM minimizes tip-sample interactions, preserving sample integrity for follow-up analyses, including STEM. Additionally, the tip-enhanced field generated by the IR laser enables high-precision chemical characterization, many times with higher spatial resolution than the topographical image.

By combining STEM and PiFM, researchers can obtain a more complete understanding of sample structure, composition, and chemistry. The combination of STEM’s unparalleled structural and elemental imaging with PiFM’s molecular insight provides a powerful platform for comprehensive nanoscale analysis. For effective nanoscale correlative studies across different instruments, identifying the same region of measurement may be difficult. For STEM and PiFM, the use of the TEM grid provides the natural fiducial features to navigate to the same locations without any uncertainty.

Application 1: chitin nanocrystals

Chitin nanocrystals form twisted aggregates approximately 1 µm in length. These aggregates can be mounted on TEM grids and analyzed using STEM.

In the far-left image on figure one, we see a STEM image of a square in a TEM grid where the dotted blue square is zoomed in to produce the upper image second from the left. The blue square in that upper image is zoomed in again to produce the lower image second from the left with the target chitin aggregate identified from several chitin nanocrystals. When we place the same sample in our PiFM system, the same TEM grid can be identified via the integrated optical microscope (notice the AFM cantilever in the optical view) via the visible aggregates on the TEM grid (far-left) which correspond to the aggregates in the STEM image (second from the left). Further, when we collect our topography image from our PiFM instrument (second from the right), we can match the features between the zoomed-in STEM image as shown by the blue arrows pointing at the same features. This use of visible features associated with a TEM grid allows the two instruments to analyze the same location on the sample with both STEM and PiFM.

While STEM provides exceptional structural and topographical resolution, it has a notable limitation: it does not provide molecular information. From the STEM image alone, it is impossible to determine the chemical composition of the observed aggregates or to detect potential surface contaminants. In theory, the sample above could have samples other than chitin nanocrystals forming similar shapes. PiFM addresses this limitation by providing complementary molecular-level insight. Figure 2 shows the standard AFM topography of the targeted chitin nanocrystal from where four PiF-IR spectra were acquired from different regions of interest, highlighted with colored markers; the PiF-IR spectra associated with those regions are shown in Figure 3), revealing local chemical information. This combined approach allows researchers to correlate structural features with chemical composition, offering a more complete understanding of the sample.

PiF-IR spectra reveal distinct chemical signatures across different regions of the sample. Thick chitin area (green spectrum) exhibited a prominent amide I peak at 1660 cm⁻¹ with a shoulder at ~ 1630 cm⁻¹ along with a weaker amide II peak at ~1560 cm⁻¹ suggestive of α-chitin. The thinner, fiber-like structures (blue spectrum) display peaks at 1715, 1570, and 1440 cm⁻¹, clearly distinguishing these regions from the main chitin aggregate. Relatively clean regions (purple spectrum) of the substrate showed a 1715 cm⁻¹ peak along with an intensity increase near 1263 cm⁻¹, indicating contributions from the substrate itself. The red spectrum with a strong peak at 1107 cm⁻¹ likely corresponds to a contaminant. This spatially resolved molecular data highlights the chemical heterogeneity between chitin, fibers, the substrate, and localized contaminant, information that is not accessible via STEM alone. To further enhance visualization, PiFM images at specific IR wavelengths can be acquired to generate chemical maps as shown in Figure 4, where the images at 1660, 1440, 1263, and 1107 cm⁻¹ are highlighting the α-chitin, unknown fibrous molecules, substrate, and contaminants, respectively.

Application 2: TEM grid alignment of recrystallized ascorbic acid samples

In this application, we demonstrate correlative imaging using a recrystallized ascorbic acid sample mounted on a TEM grid. In Figure 5, on the far left, we show the TEM grid the ascorbic acid samples were deposited on. The second to the left image of figure 5 is a STEM image, showing the various grids within the TEM sample holder. The green square surrounding the sample labeled 1 is zoomed in to produce the image third from the left. The same grid is identified by the integrated optical microscope (far left image) in our PiFM instrument. A large AFM scan (in purple border, far-right bottom image) allows us to match the STEM and AFM features (bright features in STEM appear lower in AFM topography (darker)). Then we further zoom into the red square from that topography image to produce the topography image on the top right where we acquire multiple PiF-IR point spectra at selected locations indicated by the colored triangles (Figure 6 shows the corresponding spectra).

The point spectra revealed wavenumbers of interest at 1733, 1670, and 1144 cm⁻¹. PiFM images at those wavenumbers were highlighted in blue, green, and red, respectively and combined to produce a chemical map as shown in Figure 7. The combined PiFM image shows that ascorbic acid crystals are mostly highlighted in two distinct colors, red and green, while the lower-lying regions appear in blue and purple. Green-highlighted regions exhibit peaks at 1751, 1733, 1683, 1673, 1339, 1317, 1252, 1143, 1119, and 1022 cm⁻¹. In contrast, red-highlighted regions display a markedly stronger peak near 1143 cm⁻¹ compared to the green regions. Blue and purple regions show reduced intensity around 1683 – 1673 cm⁻¹, with a slightly enhanced response near 1733 cm⁻¹.

These spectral differences reflect variations in molecular orientation within the ascorbic acid crystals. The study of how PiF-IR peaks are associated with different molecular orientations is an ongoing investigation with our customer. These preliminary results demonstrate the power of correlative PiFM and STEM analysis for analyzing recrystallized ascorbic acid samples. By precisely aligning the same sample region across both techniques, we are able to directly link high-resolution structural information from STEM with molecular orientation data obtained via PiFM.

A word of warning…

In these examples, regions of interest were initially identified in STEM based on distinctive features across different grids. A similar complementary measurements can be performed on reviewing defects found on blank silicon wafers. In such cases, we use the coordinates of the defects provided by a defect inspection tool to navigate to the same defects. We have conducted such correlated studies on many defects analyzed by review SEMs and PiFM where PiFM measurements were taken before and after the SEM analysis. In many instances, we have seen that SEM interactions alter the topography and chemical composition of defects, particularly for features smaller than 20 nm in height and lateral extent. Therefore, for correlative workflows involving STEM, SEM, or other electron-based (and other destructive) techniques, PiFM should be performed first to be followed by subsequent electron-based analyses.

Conclusion

Integrating STEM and PiFM enables comprehensive nanoscale characterization by combining high-resolution structural and elemental imaging with molecular-level chemical insight. Correlative workflows allow precise alignment of the same sample regions, linking structural features with chemical composition and molecular orientation. Performing PiFM prior to STEM preserves the native state of sensitive samples, providing a reliable and complementary framework for detailed nanoscale studies.