One of the most interesting conferences on Process Analytical Chemistry is the annual IFPAC event which took place last week in Bethesda, Maryland, USA. It is a truly international forum, where close to half of the audience I met were from outside the USA. The conference also had a striking diversity of professions represented; process engineers, chemists, chemometricians, executives, and government regulators. This diversity indicates that the field is getting broader as more professional roles take an interest in process analytics and that the understanding of the potential in focusing on how processes and analytics can be developed to improve quality and efficiency grows.

As a spectroscopist, I was very interested in what developments in instrumentation have taken place in the process spectroscopy market space recently. Certainly, there has been an explosion of interest in using Raman spectroscopy in process control, judging by the number of Raman exhibitors present at IFPAC. Another trend I noticed is spectrometer miniaturization (which has been around for a while and still going strong). Both Viavi (their MicroNIR™ is featured in the picture above) and Spectral Engines displayed near infrared spectrometers that fit in the palm of your hand. Traditionally in spectroscopic process monitoring the sample is brought to the spectrometer via plumbing, or the electromagnetic radiation is brought to and from the process via fiber optics. These small spectrometers now allow bringing the spectrometer to the process. This means that rather than monitoring one place in a process it is now feasible to install and control spectrometers in multiple places in a process, giving a more accurate and clearer picture of the process. Yet, there is still demand for traditional spectroscopic process instruments, given the amount of near- and mid-infrared spectrometers on display.

Of course what ties this all together is the need for multivariate analysis (MVA) in process spectroscopy. Regardless of whether you are looking at near-infrared, Raman, or most any other type of process spectra, you are going to need quantitative and or classification models to make sense out of your data. Additionally, to monitor a process in real time you will need software that can take MVA models and apply them to multiple data streams to obtain an up to the minute accurate picture of a process. Several companies at IFPAC exhibited such software, including of course CAMO. In the end, I think the use of multiple spectrometers to monitor processes combined with MVA will give companies an even better and more accurate idea of their processes than is currently available.

Dr. Brian C. Smith, West Coast Business Development Manager, CAMO Software USA