Semiconductor performance is closely linked to material structure. Strain, defects, doping and variations in crystallinity can all influence how a device behaves, making techniques that connect structure with optical performance particularly valuable.
In this Edinburgh Instruments webinar, Dr Matthew Berry and Angela Flack explore advanced Raman spectroscopy, PL and multimodal microscopy through semiconductor case studies spanning silicon, silicon carbide, gallium nitride LEDs and wafer-scale analysis.
Raman spectroscopy offers high chemical and structural specificity, enabling changes in crystal structure, strain, crystallinity and film thickness to be investigated. PL provides complementary information on semiconductor band gaps and radiative efficiency. Bringing both techniques together enables structural and optoelectronic information to be collected from the same region of a sample, providing a more complete picture of material performance.
In silicon, simultaneous Raman and PL imaging was combined with principal component analysis (PCA) to identify a defect exhibiting both a shift in the Raman phonon mode and reduced PL. This correlated structural variation, potentially associated with strain or doping, with a reduction in radiative efficiency.
For silicon carbide, confocal Raman microscopy takes defect characterisation into three dimensions. By rejecting out-of-focus light and optically sectioning the sample, Raman maps can reveal both the chemical composition and spatial extent of features beneath the surface.
In the webinar case study, Raman spectroscopy differentiated the expected 4H-SiC from a 6H-SiC polytype inclusion. 3D Raman imaging showed that the inclusion formed a pipe-like feature extending more than 3 µm into the wafer, demonstrating how Raman microscopy can provide both chemical and morphological information on semiconductor defects.
Multimodal spectroscopy can also directly correlate material structure with device performance. In a gallium nitride LED, Raman mapping tracked variations in the strain-sensitive E2 high phonon mode, while PL mapping provided a corresponding view of optical performance.
Overlaying the datasets revealed that regions of higher strain corresponded with lower PL intensity. Raman therefore identified where the crystal lattice was distorted, while PL showed how efficiently those regions emitted light, establishing a direct structure-performance relationship across the device.
The same Raman and PL approaches can be scaled from individual semiconductor structures to complete wafers. With the Edinburgh Instruments RMS1000 Multimodal Confocal Microscope, wafer mapping can automate spectral acquisition across the surface and assess parameters such as Raman or PL peak position, intensity and linewidth against user-defined tolerances.
This provides a route to assessing semiconductor uniformity and identifying regions that fall outside required specifications, extending multimodal spectroscopy from research and development towards wafer-level quality control.
Watch the full webinar to see Dr Matthew Berry and Angela Flack explain how Raman spectroscopy, PL, 3D Raman imaging, and wafer mapping can uncover defects, investigate strain and connect semiconductor structure with optoelectronic performance, from individual microscopic features through to entire wafers.
