Apostol Todorovski , Nikola Geskovski , Marina Petreska , Eric Deconinck , Hervé Rebière , Natalija Nakov , Katerina Brezovska , Jelena Acevska
{"title":"筛选振动光谱技术同时定量粉末混合物中维生素 B1、B6 和 B12 的能力","authors":"Apostol Todorovski , Nikola Geskovski , Marina Petreska , Eric Deconinck , Hervé Rebière , Natalija Nakov , Katerina Brezovska , Jelena Acevska","doi":"10.1016/j.vibspec.2024.103711","DOIUrl":null,"url":null,"abstract":"<div><p>The rising possibilities of vibrational spectroscopy coupled with multivariate modeling, enables many applications of these well-established analytical techniques, easy-to-use for rapid and non-destructive analysis. Owing to the lack of need for hazardous solvents or reagents in sample preparation, FTIR, NIR and Raman spectroscopy belong to the group of environmentally friendly techniques, known as “green analytical techniques”. Such characteristics re-introduce them into the pharmaceutical industry, especially as part of process-analytical-technology (PAT) tools for both qualitative and quantitative analysis.</p><p>This research aimed to assess the potential of ATR-FTIR, NIR and Raman as a PAT for the quantification of Vitamins B1 (thiamine), B6 (pyridoxine) and B12 (cyanocobalamin) in a powder blend, using a partial least square (PLS) regression model, as well as to monitor the powder blend homogeneity.</p><p>Results from each of the vibrational spectroscopy techniques were compared with those obtained by HPLC, which served as the reference analytical technique for quantifying the active substances in complex matrices using multivariate analysis. The developed PLS models were evaluated for their ability to quantify each of the active substances in the powder blend. The ATR-FTIR, NIR and Raman spectrum segments, revealed by the Variable Importance for Projection (VIP) plot for each model, respectively, were linked to the band assignment of each active ingredient.</p><p>The statistical indicators of the models (interpretation rate (R2X and R2Y), predictive ability (Q2 and RMSEP) and accuracy (RMSEE)) demonstrated their suitability for in-process estimation of vitamin B1, B6 and B12 contents in powder blends. Estimating B12 content proved more challenging, likely due to sampling issues related to its low content in the powder blend. Nevertheless, the observed variability in the models aligns with the variability in results obtained by HPLC, indicating the lack of blend homogeneity. Therefore, the quantification models could be considered for a further upgrade as PAT for monitoring the powder blend homogeneity during the manufacturing of vitamin blends.</p></div>","PeriodicalId":23656,"journal":{"name":"Vibrational Spectroscopy","volume":"133 ","pages":"Article 103711"},"PeriodicalIF":2.7000,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Screening the capability of vibrational spectroscopic techniques for simultaneous quantification of vitamins B1, B6, and B12 in a powder blend\",\"authors\":\"Apostol Todorovski , Nikola Geskovski , Marina Petreska , Eric Deconinck , Hervé Rebière , Natalija Nakov , Katerina Brezovska , Jelena Acevska\",\"doi\":\"10.1016/j.vibspec.2024.103711\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>The rising possibilities of vibrational spectroscopy coupled with multivariate modeling, enables many applications of these well-established analytical techniques, easy-to-use for rapid and non-destructive analysis. Owing to the lack of need for hazardous solvents or reagents in sample preparation, FTIR, NIR and Raman spectroscopy belong to the group of environmentally friendly techniques, known as “green analytical techniques”. Such characteristics re-introduce them into the pharmaceutical industry, especially as part of process-analytical-technology (PAT) tools for both qualitative and quantitative analysis.</p><p>This research aimed to assess the potential of ATR-FTIR, NIR and Raman as a PAT for the quantification of Vitamins B1 (thiamine), B6 (pyridoxine) and B12 (cyanocobalamin) in a powder blend, using a partial least square (PLS) regression model, as well as to monitor the powder blend homogeneity.</p><p>Results from each of the vibrational spectroscopy techniques were compared with those obtained by HPLC, which served as the reference analytical technique for quantifying the active substances in complex matrices using multivariate analysis. The developed PLS models were evaluated for their ability to quantify each of the active substances in the powder blend. The ATR-FTIR, NIR and Raman spectrum segments, revealed by the Variable Importance for Projection (VIP) plot for each model, respectively, were linked to the band assignment of each active ingredient.</p><p>The statistical indicators of the models (interpretation rate (R2X and R2Y), predictive ability (Q2 and RMSEP) and accuracy (RMSEE)) demonstrated their suitability for in-process estimation of vitamin B1, B6 and B12 contents in powder blends. Estimating B12 content proved more challenging, likely due to sampling issues related to its low content in the powder blend. Nevertheless, the observed variability in the models aligns with the variability in results obtained by HPLC, indicating the lack of blend homogeneity. Therefore, the quantification models could be considered for a further upgrade as PAT for monitoring the powder blend homogeneity during the manufacturing of vitamin blends.</p></div>\",\"PeriodicalId\":23656,\"journal\":{\"name\":\"Vibrational Spectroscopy\",\"volume\":\"133 \",\"pages\":\"Article 103711\"},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2024-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Vibrational Spectroscopy\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S092420312400064X\",\"RegionNum\":3,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"CHEMISTRY, ANALYTICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Vibrational Spectroscopy","FirstCategoryId":"92","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S092420312400064X","RegionNum":3,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, ANALYTICAL","Score":null,"Total":0}
Screening the capability of vibrational spectroscopic techniques for simultaneous quantification of vitamins B1, B6, and B12 in a powder blend
The rising possibilities of vibrational spectroscopy coupled with multivariate modeling, enables many applications of these well-established analytical techniques, easy-to-use for rapid and non-destructive analysis. Owing to the lack of need for hazardous solvents or reagents in sample preparation, FTIR, NIR and Raman spectroscopy belong to the group of environmentally friendly techniques, known as “green analytical techniques”. Such characteristics re-introduce them into the pharmaceutical industry, especially as part of process-analytical-technology (PAT) tools for both qualitative and quantitative analysis.
This research aimed to assess the potential of ATR-FTIR, NIR and Raman as a PAT for the quantification of Vitamins B1 (thiamine), B6 (pyridoxine) and B12 (cyanocobalamin) in a powder blend, using a partial least square (PLS) regression model, as well as to monitor the powder blend homogeneity.
Results from each of the vibrational spectroscopy techniques were compared with those obtained by HPLC, which served as the reference analytical technique for quantifying the active substances in complex matrices using multivariate analysis. The developed PLS models were evaluated for their ability to quantify each of the active substances in the powder blend. The ATR-FTIR, NIR and Raman spectrum segments, revealed by the Variable Importance for Projection (VIP) plot for each model, respectively, were linked to the band assignment of each active ingredient.
The statistical indicators of the models (interpretation rate (R2X and R2Y), predictive ability (Q2 and RMSEP) and accuracy (RMSEE)) demonstrated their suitability for in-process estimation of vitamin B1, B6 and B12 contents in powder blends. Estimating B12 content proved more challenging, likely due to sampling issues related to its low content in the powder blend. Nevertheless, the observed variability in the models aligns with the variability in results obtained by HPLC, indicating the lack of blend homogeneity. Therefore, the quantification models could be considered for a further upgrade as PAT for monitoring the powder blend homogeneity during the manufacturing of vitamin blends.
期刊介绍:
Vibrational Spectroscopy provides a vehicle for the publication of original research that focuses on vibrational spectroscopy. This covers infrared, near-infrared and Raman spectroscopies and publishes papers dealing with developments in applications, theory, techniques and instrumentation.
The topics covered by the journal include:
Sampling techniques,
Vibrational spectroscopy coupled with separation techniques,
Instrumentation (Fourier transform, conventional and laser based),
Data manipulation,
Spectra-structure correlation and group frequencies.
The application areas covered include:
Analytical chemistry,
Bio-organic and bio-inorganic chemistry,
Organic chemistry,
Inorganic chemistry,
Catalysis,
Environmental science,
Industrial chemistry,
Materials science,
Physical chemistry,
Polymer science,
Process control,
Specialized problem solving.