Spectroscopic on-line monitoring of reactions in dispersed medium: chemometric challenges

Emulsion and suspension polymerizations are important industrial processes for polymer production. The end-user properties of polymers depend strongly on how the polymerization reactions proceed in time (i.e. a batch or semicontinuous, rate of reagents feeding, etc.). In other words, these reactions are process dependent, which makes the successful process control a key point to ensure high-quality products. In several process control strategies the on-line monitoring of reaction performance is required. Due to the multiphase nature of the emulsion and suspension processes, there is a lack of sensors to perform successful on-line monitoring. Near infrared and Raman spectroscopies have been pointed out as useful approaches for monitoring emulsion and suspension polymerizations and several applications have been described. In such instance, the chemometric approach on relating near infrared and Raman spectra to polymer properties is widely used and has proven to be useful. Nevertheless, the multiphase nature of emulsion and suspension polymerizations also represents a challenge for the chemometric approach based on multivariate calibration models and demands the development of new methods. In this work, a set novel results is presented from the monitoring of 15 batch emulsion reactions that show the chemometric challenge to be faced on development of new methods for successful monitoring of processes taken under dispersed medium. In order to discuss these results, several chemometric approaches were revised. It is shown that Raman and NIR spectroscopic techniques are suitable for on-line monitoring of monomer concentration and polymer content during the polymerizations, as well as medium heterogeneity properties, i.e. average particle size. It is also shown that Hotteling and Q statistics, widely used in chemometrics, might fail in monitoring these reactions, while an approach based on principal curves is able to overcome such restriction.     

Three‐way analysis of a designed compost experiment using near‐infrared spectroscopy and laboratory measurements

Composts are complex organic systems that undergo batch fermentation processes. Traditional monitoring of such processes is usually based on measuring important chemical (physical) laboratory parameters but the common trend includes using more rapid and non‐destructive methods like near‐infrared (NIR) spectroscopy. A lab‐scale designed (simplex mixture) experiment with nine compost batches, including three repeated centre point batches, was monitored over 5 weeks by NIR spectroscopy (900–1700 nm) and by wet chemical and physical measurements: pH, energy content, moisture content, NH₃/NH and temperature. The data were organized in three‐way data arrays and different three‐way methods were used for analysis: (1) PARAFAC, (2) Tucker3 and (3) PARAFAC2. The present paper stresses the advantages and the possibilities of three‐way methods compared to traditional two‐way analysis methods such as principal component analysis (PCA). Two‐way methods have a tendency to mix variables and produce, from a parsimony point of view, more complex models which are hard to interpret. The results from the three‐way methods reproduced the mixture triangle, gave common time profiles (PARAFAC and Tucker3) for all compost batches and rate constants (half‐lives) could be calculated: 6.9 days for the PARAFAC loadings from the chemical/physical parameters and between 6 and 10 days for the PARAFAC loadings from the NIR data. PARAFAC2 includes the possibility of getting individual time profiles for each compost batch. The results show that chemical/physical data and the NIR data give similar interpretations. The conclusion is that three‐way methods can be used to monitor composts batches over time. Copyright © 2011 John Wiley & Sons, Ltd.     

The use of NIR as a multi-parametric in situ monitoring technique in filamentous fermentation systems

The use of Fourier transform near infrared (FT-NIR) spectroscopy for simultaneous determination of multiple properties in an active pharmaceutical ingredient (API) fermentation process is described, together with procedures for developing accurate NIR calibrations with a performance independent of scale and the specific bioreactor used. Measurements were made in situ, by insertion of transflection probes into pilot and industrial bioreactors providing direct contact with the fermentation culture media. The ultimate goal was to establish methods for real time process monitoring aimed at enhanced process supervision, fault detection diagnosis and control of bioreactors. The in situ acquired spectra were related to lab results of samples taken from the reactors during the course of the manufacturing process. Suitable spectral wavenumber regions were selected and calibration models based on partial least squares (PLS) were developed. The root mean square errors of prediction for API content, viscosity, nitrogen source and carbon source concentration were all within acceptable ranges as compared to the off-line lab measurements, respectively, 0.03% (w/w), 150 cp, 0.01% (w/w), and 0.4% (w/w).     

Petroleum resins adsorption onto quartz sand: near infrared (NIR) spectroscopy study

In this paper we have tried to evaluate adsorption parameters of petroleum resins. Near infrared (NIR) spectroscopy is applied for resins bulk concentration evaluation during adsorption process. NIR experimental scheme and parameters are provided. NIR spectra range of 9000-13,000 cm(-1) is chosen. Quartz sand (0.2-0.8 mm fraction) is used as adsorbent; benzene is used as solvent. Different approaches of "NIR spectra-resins concentration" calibration model building are discussed. Partial least squares (PLS) regression method is used. Langmuir model is chosen for experimental data fitting. Combined usage of kinetic and isothermic data gives us ability to evaluate the maximal adsorbed mass density, the equilibrium constant of adsorption, and the rate constants of adsorption (and desorption). The rate constants of resins adsorption and desorption are found to be concentration independent.     

Interpolation and extrapolation problems of multivariate regression in analytical chemistry: benchmarking the robustness on near-infrared (NIR) spectroscopy data

Modern analytical chemistry of industrial products is in need of rapid, robust, and cheap analytical methods to continuously monitor product quality parameters. For this reason, spectroscopic methods are often used to control the quality of industrial products in an on-line/in-line regime. Vibrational spectroscopy, including mid-infrared (MIR), Raman, and near-infrared (NIR), is one of the best ways to obtain information about the chemical structures and the quality coefficients of multicomponent mixtures. Together with chemometric algorithms and multivariate data analysis (MDA) methods, which were especially created for the analysis of complicated, noisy, and overlapping signals, NIR spectroscopy shows great results in terms of its accuracy, including classical prediction error, RMSEP. However, it is unclear whether the combined NIR + MDA methods are capable of dealing with much more complex interpolation or extrapolation problems that are inevitably present in real-world applications. In the current study, we try to make a rather general comparison of linear, such as partial least squares or projection to latent structures (PLS); "quasi-nonlinear", such as the polynomial version of PLS (Poly-PLS); and intrinsically non-linear, such as artificial neural networks (ANNs), support vector regression (SVR), and least-squares support vector machines (LS-SVM/LSSVM), regression methods in terms of their robustness. As a measure of robustness, we will try to estimate their accuracy when solving interpolation and extrapolation problems. Petroleum and biofuel (biodiesel) systems were chosen as representative examples of real-world samples. Six very different chemical systems that differed in complexity, composition, structure, and properties were studied; these systems were gasoline, ethanol-gasoline biofuel, diesel fuel, aromatic solutions of petroleum macromolecules, petroleum resins in benzene, and biodiesel. Eighteen different sample sets were used in total. General conclusions are made about the applicability of ANN- and SVM-based regression tools in the modern analytical chemistry. The effectiveness of different multivariate algorithms is different when going from classical accuracy to robustness. Neural networks, which are capable of producing very accurate results with respect to classical RMSEP, are not able to solve interpolation problems or, especially, extrapolation problems. The chemometric methods that are based on the support vector machine (SVM) ideology are capable of solving both classical regression and interpolation/extrapolation tasks.     

Quantitative monitoring of an activated sludge reactor using on-line UV-visible and near-infrared spectroscopy

The performance of an activated sludge reactor can be significantly enhanced through use of continuous and real-time process-state monitoring, which avoids the need to sample for off-line analysis and to use chemicals. Despite the complexity associated with wastewater treatment systems, spectroscopic methods coupled with chemometric tools have been shown to be powerful tools for bioprocess monitoring and control. Once implemented and optimized, these methods are fast, nondestructive, user friendly, and most importantly, they can be implemented in situ, permitting rapid inference of the process state at any moment. In this work, UV-visible and NIR spectroscopy were used to monitor an activated sludge reactor using in situ immersion probes connected to the respective analyzers by optical fibers. During the monitoring period, disturbances to the biological system were induced to test the ability of each spectroscopic method to detect the changes in the system. Calibration models based on partial least squares (PLS) regression were developed for three key process parameters, namely chemical oxygen demand (COD), nitrate concentration (N-NO₃⁻), and total suspended solids (TSS). For NIR, the best results were achieved for TSS, with a relative error of 14.1% and a correlation coefficient of 0.91. The UV-visible technique gave similar results for the three parameters: an error of ~25% and correlation coefficients of ~0.82 for COD and TSS and 0.87 for N-NO₃⁻. The results obtained demonstrate that both techniques are suitable for consideration as alternative methods for monitoring and controlling wastewater treatment processes, presenting clear advantages when compared with the reference methods for wastewater treatment process qualification.     

Near infrared (NIR) spectroscopy for in-line monitoring of polymer extrusion processes

In recent years, near infrared (NIR) spectroscopy has become an analytical tool frequently used in many chemical production processes. In particular, on-line measurements are of interest to increase process stability and to document constant product quality. Application to polymer processing e.g. polymer extrusion, could even increase product quality. Interesting parameters are composition of the processed polymer, moisture, or reaction status in reactive extrusion. For this issue a transmission sensor was developed for application of NIR spectroscopy to extrusion processes. This sensor includes fibre optic probes and a measuring cell to be adapted to various extruders for in-line measurements. In contrast to infrared sensors, it only uses optical quartz components. Extrusion processes at temperatures up to 300 degrees C and pressures up to 37 MPa have been investigated. Application of multivariate data analysis (e.g. partial least squares, PLS) demonstrated the performance of the system with respect to process monitoring: in the case of polymer blending, deviations between predicted and actual polymer composition were quite low (in the range of +/-0.25%). So the complete system is suitable for harsh industrial environments and could lead to improved polymer extrusion processes.     

Solubility Parameters Based on IR and NIR Spectra: I. Correlation to Polar Solutes and Binary Systems

IR and NIR spectra were correlated to Hildebrand and Hansen solubility parameters through use of multivariate data analysis. PLS‐1 models were developed and used to predict solubility parameters for solvents, crude oils, and SARA fractions. PLS regression showed potential for good correlation of the solubility parameters with IR and NIR spectra. Principal component analysis of IR spectra showed that crude oils are grouped according to their relative contents of heavy components such as asphaltenes. PCA of IR spectra for SARA fractions resulted in obvious groupings of the respective fractions. Prediction of solubility parameters from IR spectra of polymers, crude oils, and SARA fractions gave values that are comparable to literature values. This study indicates that correlation of solubility parameters with IR and NIR spectra is possible. In turn, it may be possible to develop models that can predict the polarities of crude oils and crude oil fractions such as resins and asphaltenes.     

Qualitative and quantitative analysis of oxytetracycline by near-infrared spectroscopy

Near-infrared (NIR) spectroscopy, in combination with chemometrics, enable the analysis of raw materials without time-consuming sample preparation methods. The aim of our work was to estimate critical parameters in the analytical specification of oxytetracycline, and consequently the development of a method for quantification and qualification of these parameters by NIR spectroscopy. A Karl Fischer (K.F.) titration to determine the water content, a colorimetric assay method, and Fourier transform-infrared (FT-IR) spectroscopy to identify the oxytetracycline base, were used as reference methods, respectively. Multivariate calibration was performed on NIR spectral data using principal component analysis (PCA), partial least-squares (PLS 1) and principal component regression (PCR) chemometric methods. Multivariate calibration models for NIR spectroscopy have been developed. Using PCA and the Soft Independent Modelling of Class Analogy (SIMCA) approach, we established the cluster model for the determination of sample identity. PLS 1 and PCR regression methods were applied to develop the calibration models for the determination of water content and the assay of the oxytetracycline base. Comparing the PLS and PCR regression methods we found out that the PLS is better established by NIR, especially as the spectroscopic data (NIR spectra) are highly collinear and there are many wavelengths due to non-selective wavelengths. The calibration models for NIR spectroscopy are convenient alternatives to the colorimetric method and to the K.F. method, as well as to FT-IR spectroscopy, in the routine control of incoming material.     

The effect of different gloss levels on in-line monitoring of the thickness of printed layers by NIR spectroscopy

Near-infrared (NIR) reflection spectroscopy was used for monitoring the thickness or rather the coating weight of thin printed layers of transparent oil-based offset printing varnishes in a range from 0.5 to 5 g?m^?2. Quantitative analysis of the spectral data was carried out with partial least squares regression. Surface properties such as the gloss were found to strongly affect the prediction of the coating weight. This influence was minimized by the development of calibration models, which contained spectra of layers with a broad range of gloss levels. The prediction error of these models was in the order of 0.12 to 0.16 g?m^?2. In-line measurements were carried out at a sheet-fed offset printing press in order to test the performance of the models under real process control conditions. Varnishes were applied to paper at printing speeds of 90 or 180 m?min^?1. A close correlation between the predictions from in-line NIR spectra and the reference data from gravimetry was observed regardless of the specific degree of gloss of the layers (errors between 0.15 and 0.17 g?m^?2). The results clearly prove the efficiency of NIR reflection spectroscopy for quantitative investigations on thin layers in fast processes such as printing and demonstrate its analytical potential for quality and process control.

Thermorheological studies on polymeric blends

Two petroleum-derived aromatic hydrocarbon resins (HRs) were blended (1:1) with expanded polystyrene (EPS) waste and small amounts (up to 10 mass%) of poly(vinyl chloride) (PVC) to increase both the lustrous carbon (LC) yield and softening point of the blends without any deterioration of their rheological characteristics. The blends were prepared and tested for LC content, softening points, shear stress and apparent viscosity to check their applicability as LC precursors under industrial conditions. The properties of polystyrene compositions with bitumen fractions depend primarily on composition and viscosity of oil fraction. Additional modification by poly(vinyl chloride) improves the blends’ properties, like bright coal content, softening point and viscosity, and opens new possibilities of plastics’ wastes utilization.     

Comparing near infrared and Raman spectroscopy for on-line monitoring of emulsion copolymerization reactions

In this work, Raman and Near InfraRed (NIR) spectroscopies are evaluated for the monitoring of different semicontinuous emulsion homo- and co-polymerization reactions. Important process variables, namely monomer concentrations and average particle sizes, were monitored by both techniques under realistic conditions that would be found in an industrial environment (e.g. low signal/noise ratio, probe placed in the reaction medium). Results suggest that Raman and NIR are suitable for on-line monitoring of emulsion polymerization reactions and that the success of their application is mainly related to representative calibration models used for the estimation of the properties of interest.     

In‐Line Monitoring of Vinyl Chloride Suspension Polymerization with Near Infrared Spectroscopy, 2 ‐ Design of an Advanced Control Strategy

Near infrared spectroscopy (NIRS) may constitute a powerful tool for in‐line monitoring of morphological properties of PVC particles in suspension polymerizations. It is shown that dynamic trajectories of morphological properties, as predicted with NIR‐based calibration models, change smoothly along the batch; thus, these trajectories can be used as references for process monitoring and control. It is also shown that modification of operation variables during the batch leads to modification of the final morphological properties of the powder. This indicates that the morphology of PVC grains can be manipulated along the batch and that advanced NIR‐based control procedures can be implemented for control of the morphological properties of PVC resins, as illustrated through simulation.

     

In-line monitoring of extraction process of scutellarein from Erigeron breviscapus (vant.) Hand-Mazz based on qualitative and quantitative uses of near-infrared spectroscopy

The application of near-infrared (NIR) spectroscopy for in-line monitoring of extraction process of scutellarein from Erigeron breviscapus (vant.) Hand-Mazz was investigated. For NIR measurements, two fiber optic probes designed to transmit NIR radiation through a 2 mm pathlength flow cell were utilized to collect spectra in real-time. High performance liquid chromatography (HPLC) was used as a reference method to determine scutellarein in extract solution. Partial least squares regression (PLSR) calibration model of Savitzky-Golay smoothing NIR spectra in the 5450-10,000 cm(-1) region gave satisfactory predictive results for scutellarein. The results showed that the correlation coefficients of calibration and cross validation were 0.9967 and 0.9811, respectively, and the root mean square error of calibration and cross validation were 0.044 and 0.105, respectively. Furthermore, both the moving block standard deviation (MBSD) method and conformity test were used to identify the end point of extraction process, providing real-time data and instant feedback about the extraction course. The results obtained in this study indicated that the NIR spectroscopy technique provides an efficient and environmentally friendly approach for fast determination of scutellarein and end point control of extraction process.     

Using chemometric methods and NIR spectrophotometry in the textile industry

A quantitative and qualitative technique for identification of textiles, moisture measurements, textile coatings and process control was developed, using near infra-red spectroscopy (NIR) in combination with chemometric methods. These applications demonstrate by the use of computer assisted data processing the possibility of identifying textile fibers, not only for quality control but also for online textile recycling processes. In this study, seven various textile fibers (cotton, polyester, viscose, silk, wool, polyacrylonitrile, acetate) were used and all combinations of two factor blends were qualitatively identified using NIR spectroscopy and the chemometric PLS2 method for the calibration.

A quantitative analysis of textile moisture can also be performed with this technique. Water content above 50% does not deliver good results for a calibration set to determine the dampness of fibers. But to measure residual moisture from ≈0.05 up to 50%, the NIR technique is particularly good. Furthermore, the examination shows that the NIR method and chemometric methods can be used in quality- and product-control during the industrial production of upholstery fabrics. With this technique it will be possible to identify nylon flocks and to measure the residual moisture of the flocks and fabric, too.     

Multivariate NIR spectroscopy models for moisture, ash and calorific content in biofuels using bi-orthogonal partial least squares regression

The multitude of biofuels in use and their widely different characteristics stress the need for improved characterisation of their chemical and physical properties. Industrial use of biofuels further demands rapid characterisation methods suitable for on-line measurements. The single most important property in biofuels is the calorific value. This is influenced by moisture and ash content as well as the chemical composition of the dry biomass. Near infrared (NIR) spectroscopy and bi-orthogonal partial least squares (BPLS) regression were used to model moisture and ash content as well as gross calorific value in ground samples of stem and branches wood. Samples from 16 individual trees of Norway spruce were artificially moistened into five classes (10, 20, 30, 40 and 50%). Three different models for decomposition of the spectral variation into structure and noise were applied. In total 16 BPLS models were used, all of which showed high accuracy in prediction for a test set and they explained 95.4-99.8% of the reference variable variation. The models for moisture content were spanned by the O-H and C-H overtones, i.e. between water and organic matter. The models for ash content appeared to be based on interactions in carbon chains. For calorific value the models was spanned by C-H stretching, by O-H stretching and bending and by combinations of O-H and C-O stretching. Also -C=C- bonds contributed in the prediction of calorific value. This study illustrates the possibility of using the NIR technique in combination with multivariate calibration to predict economically important properties of biofuels and to interpret models. This concept may also be applied for on-line prediction in processes to standardize biofuels or in biofuelled plants for process monitoring.     

Application of near-infrared spectroscopy in the sugar industry for the detection of betaine

One problem in industrial molasses desugarization is the lack of a fast analytical method for process control. At the moment, control of the chromatographic production process is achieved by detecting refractive index and conductivity. However, since elution of some components takes place only in a narrowly defined time frame, the data gained are insufficient for effective online product quantification. Near-infrared (NIR) spectroscopy was applied to this process by development of a simple method for detection of betaine. Compared to chemometric models currently used, the developed method demonstrates the advantage of requiring only a small calibration set. Additionally, it can easily be transferred to other processes without further re-calibration. Based on the NIR spectrum of betaine, a characteristic peak in the spectrum could be assigned to the molasses compound betaine. A calibration was developed by using dissolved betaine in pure water. Afterwards, the calibration was tested for samples from a molasses desugarization process. The method was than successfully transferred to a complete chromatographic cycle of the industrial molasses desugarization process.     

Multivariate near infrared spectroscopy models for predicting the iodine value, CFPP, kinematic viscosity at 40 degrees C and density at 15 degrees C of biodiesel

Biodiesel is one of the main alternatives to fossil diesel. It is a non-toxic renewable resource, which leads to lower emissions of polluting gases. In fact, European governments are targeting the incorporation of 20% of biofuels in the fossil fuels until 2020.Chemically, biodiesel is a mixture of fatty acid methyl esters, derived from vegetable oils or animal fats, which is usually produced by a transesterification reaction, where the oils or fats react with an alcohol, in the presence of a catalyst. The European Standard (EN 14214) establishes 25 parameters that have to be analysed to certify biodiesel quality and the analytical methods that should be used to determine those properties.This work reports the use of near infrared (NIR) spectroscopy to determine some important biodiesel properties: the iodine value, the cold filter plugging point, the kinematic viscosity at 40 °C and the density at 15 °C. Principal component analysis was used to perform a qualitative analysis of the spectra and partial least squares regression to develop the calibration models between analytical and spectral data. The results support that NIR spectroscopy, in combination with multivariate calibration, is a promising technique applied to biodiesel quality control, in both laboratory and industrial-scale samples.     

On-line monitoring of the acrylate conversion in UV photopolymerization by near-infrared reflection spectroscopy

Near-infrared (NIR) reflection spectroscopy was used to determine the conversion in acrylate coatings after UV photopolymerization in order to test it as a method for process control in UV curing. A probe head was developed which is adapted to the specific requirements of UV curing and which is linked to a photodiode array spectrometer by an optical fiber. Reflection spectra from thin acrylate layers which were taken in intervals down to the millisecond range have shown an excellent signal-to noise ratio. Quantitative conversion data show good correlation with results from independent reference methods (FTIR, HPLC). Following thesebasic investigations, it was demonstrated that NIR reflection spectroscopy can be used for on-line monitoring of the acrylate conversion in thin coatings. Some examples of such investigations in pilot scale are presented.     

Simultaneous determination of total polyphenols and caffeine contents of green tea by near-infrared reflectance spectroscopy

This paper indicates the possibility to use near infrared (NIR) spectroscopy as a rapid method to predict quantitatively the content of caffeine and total polyphenols in green tea. A partial least squares (PLS) algorithm is used to perform the calibration. To decide upon the number of PLS factors included in the PLS model, the model is chosen according to the lowest root mean square error of cross-validation (RMSECV) in training. The correlation coefficient R between the NIR predicted and the reference results for the test set is used as an evaluation parameter for the models. The result showed that the correlation coefficients of the prediction models were R = 0.9688 for the caffeine and R = 0.9299 for total polyphenols. The study demonstrates that NIR spectroscopy technology with multivariate calibration analysis can be successfully applied as a rapid method to determine the valid ingredients of tea to control industrial processes.