Ryan Gosselin

A hyperspectral imaging sensor for on-line quality control of extruded polymer composite products

Abstract This study examines the ability of chemometrics methods, namely multivariate image analysis (MIA) and Grey Level Co-occurrence Matrix analysis (GLCM), to extract meaningful information from visible and near-infrared spectral images of extruded wood/plastic composite materials for predicting spatio-temporal variations in their properties. The samples were produced under varying process and feed conditions according to designed experiments. Mechanical properties of the samples were measured using standard analytical methods both during steady-state and dynamic transition periods. A Bootstrap-PLS regression technique was first used for selecting the spectral bands (i.e. wavelengths) that were the most highly correlated with the material properties. In a second step, a more parsimonious PLS regression model was built between the spectral and textural features extracted from the lower dimensional spectral images and the corresponding quality properties of each sample. The imaging sensor was able to simultaneously monitor 7 properties in both steady-state operation and during transitions.

Using multiple Process Analytical Technology probes to monitor multivitamin blends in a tableting feed frame

As Process Analytical Technology (PAT) implementation grows in the pharmaceutical industry, more studies are being performed to evaluate its suitability in new applications and processes within the manufacturing chain. As the last step in tablet production, the compression stage represents a critical phase that ensures product quality. In-line control put in place at this stage has the potential to detect powder blends that are out of specification limits and, thus, help to improve product quality. The objectives of the present project are to quantify the composition of a commercial 31-component multivitamin powder blend in real time on an industrial feed frame, using 3 different PAT tools: light-induced fluorescence spectroscopy, near infrared spectroscopy and red, green and blue color imaging. To do so, the concentrations of 5 components (Beta-Carotene, Riboflavin, Ferrous Fumarate, Ginseng and Ascorbic Acid) were alternately changed and monitored with one or many probes. Transition periods between batches served to quantify different powder flow dynamics with sequential composition step changes. The results showed that 4 out of 5 components, each present in commercially-relevant concentrations, could be monitored by one or more tools. Flow dynamics were measured and found to vary significantly in different powder blends.

Development of a multivariate light-induced fluorescence (LIF) PAT tool for in-line quantitative analysis of pharmaceutical granules in a V-blender

Process analytical technologies (PAT) enable process insight, process control and real-time testing. Light-induced fluorescence (LIF) spectroscopy is especially well suited for low-concentration ingredients as, in many cases, it is the most sensitive probe of the in-line PAT toolbox. This study is aimed at verifying the applicability of a multivariate LIF analyzer to monitor granulated powder blends in industrial settings. Its targets are to: (1) evaluate the critical parameters of powders to obtain robust, precise and accurate concentration predictions and (2) assess technology performance for in-line monitoring of blending operations. Varying dye properties, moisture levels and particle sizes have been shown to have the most significant impact on fluorescence emission. Reliable quantitative models can be obtained by controlling and/or mitigating these factors.

Monitoring the concentration of flowing pharmaceutical powders in a tableting feed frame

Abstract The use of process analytical technology (PAT) tools is increasing steadily in the pharmaceutical industry. Such tools are now located throughout the process. When producing tablets, the tableting step itself may be the ideal moment to assess final product composition. Being the last unit operation in tablet production where the elements are still free flowing, it is relatively straightforward to ascertain the composition of the blend in real time. However, a single probe cannot be expected to monitor the composition of every component of a multicomponent blend. In this study, three PAT tools (light-induced fluorescence spectroscopy, near-infrared spectroscopy and color (RGB) imaging) simultaneously checked the composition of powder blends flowing through the feeding unit (feed frame) of a tablet press. The results demonstrate the potential of these tools in monitoring changes in the concentration of a multicomponent mixture in real time, providing users with means to both scrutinize the process and better understand phenomena occurring inside the feed frame.

Predicting the dissolution behavior of pharmaceutical tablets with NIR chemical imaging.

Near infrared chemical imaging (NIRCI) is a common analytical non-destructive technique for the analysis of pharmaceutical tablets. This powerful process analytical technology provides opportunity to chemically understand the sample, and also to determine spatial distribution and size of ingredients in a tablet. NIRCI has been used to link disintegrant, excipients and active pharmaceutical ingredient (API) to tablet dissolution, as disintegrants play an important role in tablet disintegration, resulting in API dissolution. This article describes a specific methodology to predict API dissolution based on disintegrant chemical information obtained with NIRCI. First, several tablet batches with different disintegrant characteristics were produced. Then, NIRCI was successfully used to provide chemical images of pharmaceutical tablets. A PLS regression model successfully predicted dissolution profiles. These results show that NIRCI is a robust and versatile technique for measuring disintegrant properties in tablet formulations and predicting their effects on dissolution profiles. Thus, NIRCI could routinely complement and eventually replace dissolution testing by monitoring a critical material attribute: disintegrant content.

Optimization of a landfill gas collection shutdown based on an adapted first-order decay model

LandGEMs equation was reformulated to include two types of refuse, fast decaying refuse (FDR) and slow decaying refuse (SDR), whose fractions and key modeling parameters k and L0 were optimized independently for three periods in the life of the Montreal-CESM landfill. Three scenarios were analyzed and compared to actual biogas collection data: (1) Two-Variable Scenario, where k and L0 were optimized for a single type of refuse; (2) Six-Variable Scenario, where three sets of k and L0 were optimized for the three periods and for a single type of refuse; and (3) Seven-Variable Scenario, whereby optimization was performed for two sets of k and L0, one associated with FDR and the second with SDR, and for the fraction of FDR during each of the three periods. Results showed that the lowest error from the error minimization technique was obtained with the Six-Variable Scenario. However, this scenarios estimation of gas generation was found to be rather unlikely. The Seven-Variable Scenario, which allowed for considerations about changes in landfilling trends, offered a more reliable prediction tool for landfill gas generation and optimal shutdown time of the biogas collection system, when the minimum technological threshold would be attained. The methodology could potentially be applied mutatis mutandis to other landfills, by considering their specific waste disposal and gas collection histories.

Classifying Pharmaceutical Capsules Through X-Ray Image Analysis Based on the Agglomeration of Their Contents

PurposePharmaceutical gelatin shell capsules are commonly used to deliver powdered solid dosage forms. The dissolution behavior of these products depends on a number of factors, including, notably, powder distribution within capsules. Powder agglomeration occurring during the filling and sealing stages may lead to regulatory compliance issues. Although not a typical procedure, X-ray images of the final capsules may be inspected manually to determine the presence of such agglomerates.MethodsThe objective of this study was to develop a screening system, based on X-ray image analysis, capable of distinguishing capsules containing agglomerated powder from those containing unagglomerated powder. To do so, capsules were produced under various conditions to induce differing levels of agglomeration. Samples were first classed according to expert opinion before being tested to calibrate chemometric soft independent modeling by class analogy. As capsules were cylindrical and visually opaque, each was imaged multiple times to determine the robustness of the method to uneven powder agglomeration inside the capsules.Results and ConclusionsThe results show that X-ray imaging can automatically detect and classify powder agglomerates within pharmaceutical capsules, thus reducing reliance on subjective human inspection while increasing the online potential of capsule imaging.

Developing a quality by design approach to model tablet dissolution testing: an industrial case study

Abstract This study applied the concept of Quality by Design (QbD) to tablet dissolution. Its goal was to propose a quality control strategy to model dissolution testing of solid oral dose products according to International Conference on Harmonization guidelines. The methodology involved the following three steps: (1) a risk analysis to identify the material- and process-related parameters impacting the critical quality attributes of dissolution testing, (2) an experimental design to evaluate the influence of design factors (attributes and parameters selected by risk analysis) on dissolution testing, and (3) an investigation of the relationship between design factors and dissolution profiles. Results show that (a) in the case studied, the two parameters impacting dissolution kinetics are active pharmaceutical ingredient particle size distributions and tablet hardness and (b) these two parameters could be monitored with PAT tools to predict dissolution profiles. Moreover, based on the results obtained, modeling dissolution is possible. The practicality and effectiveness of the QbD approach were demonstrated through this industrial case study. Implementing such an approach systematically in industrial pharmaceutical production would reduce the need for tablet dissolution testing.

Specificity of process analytical tools in the monitoring of multicomponent pharmaceutical powders

Abstract The application of Process Analytical Technologies in pharmaceutical manufacturing has been the subject of many studies. Active pharmaceutical ingredient monitoring in real time throughout the manufacturing process is commonly the target of many such implementations. The tools in place must be sensitive to, and selective of, the parameter(s) to be monitored, i.e. in the case of component quantification, they must respond to the component in question and be robust against all others. In this study, four different ingredients (riboflavin, ferrous fumarate, ginseng, and ascorbic acid) in a multi-component blend were monitored by three different tools (near infrared spectroscopy, laser-induced fluorescence and red-green-blue camera) using a full factorial design. The goal was to develop efficient and robust concentration-reading/prediction models able to assess and monitor component interference. Despite relatively high complexity of the blend studied, the three tools demonstrated reasonable specificity for the tracked ingredients (and showed advantages when combined), taking into account larger acceptance criteria typical of dietary products. In certain cases, some interference might lead to biased predictions, highlighting the importance of good calibration. The tools tested and the methodology proposed has divulged their potential in monitoring these components, despite the complexity of the 31-component blend.

A Hierarchical Multivariate Curve Resolution Methodology To Identify and Map Compounds in Spectral Images

The use of spectroscopic methods, such as near-infrared or Raman, for quality control applications combined with the constant search for finer details leads to the acquisition of increasingly complex data sets. This should not prevent the user from characterizing a sample by identifying and mapping its chemical compounds. Multivariate data analysis methods make it possible to obtain qualitative and quantitative information from such data sets. However, samples containing a large (and/or unknown) number of species, segregated trace compounds (present in few pixels), low signal-to-noise ratios (SNR), and often insufficient spatial resolutions still represent significant hurdles for the analyst.