Surface roughness of preparations for backscattered electron-scanning electron microscopy: the image differences and...

Researchers have long used backscattered electron (BSE) scanning electron microscopy to study mineralization patterns in biological hard tissues. However, surface topography can interfere with the interpretation of compositional signals. To minimize this interference, samples are typically embedded in PMMA and polished or micromilled to create a flat surface. The extent of residual topography after these processes has not been fully characterized. Prior research has shown that surface roughness can affect BSE signal contrast, but the specific impact of different preparation methods remains unclear. This gap motivated a study to quantify the surface roughness of PMMA-embedded samples after various finishing steps. Understanding the relationship between preparation techniques and resulting surface topography is essential for improving image accuracy. The study aimed to determine how much residual topography remains after each preparation step. By comparing these findings to BSE imaging results, the researchers hoped to clarify the influence of surface roughness on image contrast. This work addresses a critical need in the field of electron microscopy for biological tissues.

The goal of this study was to assess the residual surface roughness of PMMA-embedded biological samples after different preparation techniques. Specifically, the researchers wanted to compare the effects of polishing and micromilling on the final surface topography. They hypothesized that the choice of preparation method could significantly influence the BSE signal contrast observed in SEM imaging. By quantifying the peak-trough relief after each step, the study aimed to identify the most suitable method for minimizing topographic artifacts. The researchers also sought to understand how these surface features might contribute to misinterpretations of mineralization patterns in bone. Using confocal reflection microscopy, they mapped the surface topography at each stage of preparation. These data were then used in Monte Carlo simulations to model the expected BSE signal variations. The ultimate purpose was to determine which preparation method best preserves the true compositional contrast in BSE imaging.

The study used a human rib sample embedded in PMMA as the test material. The sample was prepared using a sequence of steps: polishing on graded abrasives, followed by pre-milling and ultramilling. After each step, the block face was imaged using a confocal reflection microscope to record surface topography. These topographic data were then input into a Monte Carlo simulation to model the surface interface. The simulation calculated the expected variations in BSE signal for each preparation technique. The researchers compared the simulated results with experimental data obtained from standard SEM imaging conditions. This approach allowed them to assess how surface roughness affects BSE contrast. The study focused on quantifying peak-trough relief and undulation in the prepared surfaces. By correlating these measurements with the simulated and experimental BSE signals, the researchers evaluated the effectiveness of each preparation method.

Micromilling produced block faces with a peak-trough relief of approximately 80 nm, while hand polishing left occasional scratches 1.5 microns deep and a general undulation of 150-250 nm. Monte Carlo simulations predicted that micromilled surfaces would introduce an additional 5% contrast in BSE imaging. In contrast, hand polished samples were expected to show more than 10% additional contrast. These findings were supported by the experimental data from SEM imaging under standard conditions. The higher contrast in hand polished samples was attributed to the collagen orientation-related relief that develops during polishing. This relief may lead to the incorrect assumption that lamellation in bone is due to changes in mineralization levels. The simulations and experimental results were closely aligned, confirming the impact of surface roughness on BSE signal. The study demonstrated that micromilling is superior to hand polishing for minimizing topographic artifacts in BSE imaging. These results provide a quantitative basis for selecting the most appropriate sample preparation method.

The study concluded that micromilling is the preferred method for preparing PMMA-embedded bone samples for BSE imaging. This technique results in a smoother surface with less residual topography compared to hand polishing. The researchers found that micromilled surfaces produce a 5% increase in BSE contrast, while hand polished samples show more than 10% additional contrast. These differences are significant enough to affect the interpretation of mineralization patterns in bone. The study also highlighted that the collagen orientation-related relief introduced during hand polishing may be responsible for the mistaken belief that lamellation in bone is linked to changes in mineralization. The Monte Carlo simulations and experimental data were consistent, validating the impact of surface roughness on BSE signal. The findings suggest that micromilling should be used to avoid topographic artifacts in BSE imaging of bone. The authors emphasized the importance of selecting the right preparation method to ensure accurate compositional contrast in SEM imaging.