Development and testing of mid-infrared sensors for in-line process monitoring in biotechnology

Monitoring yeast fermentations by nonlinear infrared technology and chemometrics—understanding process correlations and indirect predictions

“…In accordance with the literature (Bogomolov et al 2015), and with the Supplementary Fig.s S5 and S6 showing the direct comparison of cell-free and cell-containing samples and the PCA score plot of all data sets, respectively, the presence of cells did not affect the spectra significantly and hence the prediction results. This is also confirmed in Fig.…”

“…However, a new calibration procedure must be considered, since the spectra (baseline shape) acquired on-line differed significantly from the spectra acquired off-line due to the different ATR elements in use and the application of flow. Although the spectra acquired with the NLIR technology showed a slightly downgraded data quality compared to the standard FTIR instrument, as do previously described prototypes (Bogomolov et al 2015), models resulted in comparable predictability (compared to the standard FTIR) yielding RMSEP values of less than 15 % (excluding the models built for glycerol and acetate). Hence, the novel NLIR technology presented in this work can be seen as a valid alternative to the FTIR technology, besides being more suitable for an industrial production environment.…”

“…Within this work, the yeast performance parameters glucose, ethanol, glycerol and acetate, and additionally, ammonium and phosphates were modeled successfully, and importantly, applying novel NLIR technology. Due to their broad relevance in various fermentations, these parameters have been modeled by PLS based on IR data previously, mostly with similar performance results and limitations (Bogomolov et al 2015;Diewok et al 2004;Kornmann et. al.…”

“…Its stagnancy in industrial biotech applications, however with incremental progress, might be due to the high cost, the presence of fragile and sensitive moving parts making it sensitive towards mechanical vibrations, intrinsic thermal noise and the limited length of attenuated total reflectance (ATR) fibers for remote sampling. These challenges are addressed by recent technical advances in the MIR sensor technology, as for instance described in Bogomolov et al (2015) and besides, employed in this work (Barh et al 2017;Dam et al 2016). These novel technologies use simpler grating spectrometers, are free of any moving parts, reduce intrinsic detector noise and are significantly less expensive than traditional high-precision FTIR analyzers, typically employing the full spectral range (800 -4000 cm -1 ).…”

Quantitative analysis of total hydrocarbons and water in oil‐contaminated soils with attenuated total reflection infrared spectroscopy

“…This mixing scheme called “diagonal design” provides uncorrelated variability of both components and even coverage of the chosen concentration ranges (0%‐13% for both components) using the minimum number of samples . Both oil and water content models were built on the same set of 25 designed samples.…”