Tutorials
Author: Eleni Vomhof
Published: 27 Jul 2026 · Last updated: 27 Jul 2026
Confocal Raman microscopy is a non-destructive imaging technique for comprehensive analyses of organic polymer composition, distribution, and defects. This tutorial presents a complete step-by-step workflow for performing initial Raman measurements on polymer samples. It includes practical tips for choosing the region of interest (ROI), selecting the measurement type, and setting key acquisition parameters. It also covers essential data evaluation steps. By following this workflow, you can create high-resolution chemical maps of polymer samples, and extract Raman spectra of individual components with confidence.
All steps described can be carried out using Oxford Instruments’ witec360 Raman microscope equipped with the Hexalight spectrometer, providing high resolution, speed, and sensitivity.
The first step in a typical polymer measurement workflow is to define the goal of the experiment. Examples include:
Generally, minimal sample preparation is required for analysing polymers with Raman. Most polymer samples can be placed directly onto the microscope stage, or a small section can be cut out for investigation. Ensure the sample is mounted securely to prevent drift.
Depending on the measurement objective, additional preparation might involve cutting a representative piece of the polymer and fixing it on a sample holder. For investigation of the sample composition and distribution of the components, a piece large enough to contain all components, but small enough to fit easily into the microscope is ideal. A piece between a few mm2 and a few cm2 is usually sufficient. When studying defects or contaminations, the sample should include at least one.
Begin the polymer analysis by obtaining an overview image using the video mode of the Oxford Instruments witec360 Raman microscope. Depending on the microscope configuration, select from white-light illumination modes such as reflection, transmission, or darkfield. If multiple options are available, choose the mode that best highlights the ROI in the sample. A fast overview of the sample is gained with low magnification objectives like 5x or 10x. Then, use a high magnification objective like 50x or 100x for detailed imaging of the desired area (Figure 1). Define the region to be analysed with Raman using this image.
Figure 1: Examples of video images of a copolymer sample in different modes, recorded with a 50x objective: (A) reflection, (B) darkfield, (C) transmission mode.
Before starting the Raman measurement, adjust parameters in “Oscilloscope Mode”, i.e. using the live spectrum from the sample. To maximise Raman signal and spatial resolution, use an objective with the highest possible numerical aperture, for example 100x/0.9.
Start with a 532 nm (green) laser, which typically generates strong Raman signals. If the sample burns, melts, or exhibits strong fluorescence under 532 nm, switch to an alternative wavelength. Note that shorter wavelengths produce higher Raman signals and better spatial resolution, but increase the risk of sample damage.
Set the laser power to the highest value that does not damage the sample while maximising signal (as high as possible but as low as necessary).
Then adjust the integration time to optimise the balance between acquisition speed and signal-to-noise ratio (S/N). Avoid detector saturation to prevent non-linear effects.
After all parameters are defined, record several single Raman spectra from different positions to gain an initial understanding of the sample composition. Set the integration time as described for the “Oscilloscope Mode”, and apply ~5–20 accumulations to get Raman spectra with high S/N ratio.
For another quick evaluation, perform a “Line Scan” in any chosen direction. This reveals transitions between different sample components, or determines sample thickness and/or transparency (“Line Scan” along Z axis).
Finally, Raman imaging in the XY plane (area) or XZ plane (depth) reveals sample heterogeneity, the distribution of different components, contaminations, or defects. “Depth” scans show the layers in the sample, while 3D scans (image “stacks”) visualise the 3D distribution of different components. Adjust the pixel size for the Raman image depending on the desired spatial resolution and the size of the relevant sample features. Large-area, high-resolution Raman images may require overnight acquisition. The software estimates the measurement duration, allowing parameter optimisation also with respect to the available time.
All generated objects (e.g. video images, single Raman spectra, Raman images, and processed data) appear in the Project Manager and are stored together in one project file. This can be opened in the data evaluation software (PROJECT), and the data can easily be processed while a new measurement is running.
The first steps in Raman data analysis are cosmic ray removal (CRR) and background subtraction. The PROJECT software provides a guided workflow (wizard) through these steps. Several algorithms and preview options facilitate the correct selection of parameters for both operations to preserve the Raman peaks (Figure 2A, B).
Smoothing of the Raman spectra increases S/N ratio, but can distort Raman peaks (Figure 2C). Apply it only when other ways of improving S/N (e.g. increasing laser power or accumulation time) are not feasible.
Figure 2: Examples of excessive processing of the polymer Raman spectra. Arrows indicate where the Raman signals are significantly altered: (A) excessive CRR; (B) excessive background subtraction; (C) excessive smoothing.
For heterogeneous samples, use TrueComponent analysis to find all components in a Raman image. Component identification may be manual (assisted by the software) or fully automatic. Average all spectra of identical components to improve the S/N ratio.
For Raman spectra representing a mixture of several different chemicals, apply the demixing function to obtain pure spectra (Figure 3). Ensure that no part of any Raman spectrum becomes negative during demixing.
Figure 3: Example of Raman spectra demixing: (A) Raman spectra before demixing; (B) Raman spectra after demixing.
Identification of unknown sample components can be performed with TrueMatch using commercial, free, or custom-built spectral databases.
After the analysis, the obtained Raman image visualises the spatial distribution of the identified chemical components, revealing sample heterogeneity, composition, contaminations, and defects. In the presented example, the two polymers polystyrene (PS) and poly(methyl methacrylate) (PMMA) were identified by their Raman spectra, and their spatial distribution was visualised in a Raman image of the blend (Figure 4).
The full analysis presented in this tutorial can be done in less than a minute using the PROJECT software’s guided data processing.
Figure 4: High-resolution Raman image of a copolymer sample after data processing: (A) Raman image colour-coded according to the identified polymer components PMMA (red) and PS (blue), (B) corresponding Raman spectra with the same colour code.
Raman imaging is a reliable, non-destructive method for chemical analysis of polymer composition, distribution, defects, and contamination. This guide presents a workflow for performing accurate, high-quality Raman measurements of polymer samples: from defining measurement goals through sample preparation to data acquisition and analysis. Oxford Instruments’ advanced Raman systems are designed to simplify workflows, improve accuracy, and save time.
© Oxford Instruments 2026