Raman spectroscopy for process monitoring

Karen Esmonde- White reveals fibre optic sampling probe considerations for in-process quantitative Raman spectroscopy

Integrating analytical tools to understand, monitor and control product manufacturing has many benefits, including improved process efficiency and ensuring consistent quality.

Raman spectroscopy is an important tool for laboratory, analytical and process applications. Successful Raman applications have been demonstrated at all scales, from at line in the laboratory to on-line in manufacturing in PAC and PAT environments. Kaiser has applied the measurement principles of Raman spectroscopy in a process or manufacturing environment for understanding, monitoring and controlling continuous processes or unit operations for 20 years.

Process, logistical and technological aspects need to be considered when integrating Raman spectroscopy in a process environment. One physical attribute that affects sampling of the process is optical scattering.

Here we describe technologic and process considerations in sampling probe design in response to optical scattering and how these factors affect in situ implementation of Raman spectroscopy.

Optical scattering arises from differences in refractive index. Immiscible phases, particulates or bubbles are sources of optical scattering and can cause the reaction medium to appear turbid. Multiple scattering of photons results in diffusion of the excitation laser. Kaiser has a line of proven process- compatible fibre optic probes to ensure that the measurement is representative of the sample. Fig.1 shows a variant of Raman spectroscopy that can be used in process measurements. Backscattered Raman spectroscopy is achieved using a single excitation and single collection fibre in the probe and typically samples a small volume close to the probe window. This probe configuration is primarily used as an immersion probe for in situ liquids measurements or to measure the surface of a solid. If a backscattered Raman probe is moved across the surface of a solid, then a wider area of the sample is measured. Wide area Raman can be used to measure compositional heterogeneity at the surface of a solid, and it can be employed to over-sample a surface to reduce the limits of detection or improve quantification.1, 2 Large volumetric Raman uses a large excitation spot and multiple collection fibres to achieve sampling in both the axial and lateral dimensions. Large volumetric Raman provides information on deeper layers in addition to the surface, which is useful for measuring a layered solid such as a tablet or capsule.

In situ Raman measurements

Raman spectroscopy is a robust and proven analytical technology for processes involving liquids, solids and turbid media. Applications of Kaiser Raman for liquids include primary API reaction monitoring, sealed microwave systems, continuous flow reactors, NeSSI platform devices, and small volume thermal reactors. In applications involving liquids, optical scattering arising from bubbles or immiscible phases can be affected by flow, temperature and mixing parameters. In liquids applications, turbidity is undesirable since it can attenuate the Raman signal but these effects are minimised in a small-volume backscattered Raman probe immersed into the reaction or flow. Additional benefits of using a small-volume immersion Raman probe are its compatibility with reactive or corrosive chemistries, in situ measurements, and real-time process knowledge and control.

Successful applications of small sampling volume probes include flowing streams, slip-streams  or directly in the reaction vessel. In-process Raman measurements involving solids or turbid media are established in pharmaceutical and polymer applications. Successful examples can be found in real-time release testing, tablet coating and in understating formation of spatially heterogeneous polymers, elastomers or composites. In solids and turbid media sampling, the effects of optical scattering are harnessed and representative sampling is achieved because multiple collection fibres collect signal from the surface and subsurfaces.

Raman hybrid sampling

Coupling multiple probes with different sampling volumes to a single analyser is a powerful tool because it provides a multi-scale examination of the process.

A Raman hybrid sampling approach improved understanding during polymerisation of a spatially heterogeneous material.3 High- impact polystyrene preparation involves multiple steps and the end product is a spatially heterogeneous graft copolymer. Micro-scale and macro-scale Raman measurements were collected during the polymerisation process. An immersion probe measured an approximate volume of 1μm3 and was used to measure styrene polymerisation on the micro-scale. A non-contact sampling probe measured an approximate volume of 100μm3 and was sensitive to changes macro-scale changes in the copolymer’s morphology arising from phase changes and formation of nodules. Raman can provide real-time process feedback on stirring rate, reactant addition rate or temperature in order to consistently achieve the target copolymer morphology. Embodiment of the principles described by Brun et al can be easily incorporated into a process environment using a Kaiser Raman hybrid system.

Raman spectroscopy is a valuable technique for process monitoring and control, with applications in reaction monitoring and solids manufacturing. Flexibility in fibre optic sampling probes expands the utility of Raman spectroscopy in order to meet current and future application needs. A hybrid Raman approach, using a micro-scale and a macro-scale probe, is powerful for understanding hierarchical properties of materials. Raman spectroscopy provides increased process knowledge that enables advanced process monitoring, improved product quality and real-time process control.

For more information, visit www.scientistlive.com/eurolab

Karen Esmonde-White is with Kaiser Optical Systems.

References: (1) Li, B.; Calvet, A.; Casamayou-Boucau, Y.; Morris, C.; Ryder, A. G. Anal. Chem. 2015, 87 (6), 3419–3428.
(2) Shin, K.; Chung, H. Analyst 2013, 138 (12), 3335–3346.
(3) Brun, N.; Chevrel, M.-C.; Falk, L.; Hoppe, S.; Durand, A.; Chapron, D.; Bourson, P. Chem. Eng. Technol. 2014, 37 (2), 275–282.

Recent Issues