Non-destructive Quantitative Analysis For Tablets By Use Of PLS And Near-infrared Spectroscopy

With the fabulous development of modern sciences and technology, usingmodern chemical instruments, now analysts can rapidly and precisely acquire alarge amount of measured data. Then, the analysts have to face this problem:how to extract more meaningful chemical information from the measured data.Chemometrics is a brand-new branch of chemistry. It's the art of extractingmeaningful information from chemical data by the combination of mathematics,statistics and computers. Now the analysts can use this powerful tool to solvetheir problems.Near-infrared (NIR) spectroscopy is a technique whose time has arrived. Inthe past few years, it has been common to use NIR spectra for the qualitativeand quantitative analysis of complex mixtures. Analytical applications of NIRreflectance spectroscopy have grown very rapidly in response to the increasingavailability of instruments of high resolving power capable of recording wholeNIR spectra in an expeditious, precise manner. This, in addition to developmentsin chemometric techniques for signal processing and the minimal samplehandling involved in the NIR technique, has turned it into a powerful tool forcontrol analyses in a broad range of industrial scenarios.The NIR spectral region is generally defined as the wavelength range froml100-2500nm, although there is considerable variation in wavelength ranges ofthe different instrument types. The absorption bands in this region are due toovertones and combinations of the fundamental mid-IR molecular vibrationbands. The most prominent overtone bands observed in the NIR are thoserelated to O-H, C-H, and N-H groups. In addition to the simple overtones,combination band also occur. These usually involve O-H, C-H, or N-H stretch NIR diffuse reflectance spectroscopy has been widely accepted tive and quantitative measurement of organic species, al, such as UV plus one or more bending or rocking modes. Such bands are typically weak, their intensity decreasing with increasing overtone order. This, in addition to the large number of bands and extensive overlap occurring in this spectral region, originally limited the application of the NIR technique to analysis. More recently, computer-aided NIR reflectance instruments have used mathematical manipulations to retrieve qualitative and quantitative information from the complex band envelopes. for the qualitamineralconstituents, and moisture in a variety of solid substrates. The technique is rapid, is nondestructive, and in many instances requires little or no sample preparation. Near-infrared (NIR) spectroscopy has proved to be a powerful analytical tool for analyzing a wide variety of samples used in the agricultural, food, petrochemical, textile and pharmaceutical industries, especially the use of NIR spectroscopy for quantitative analysis of the pharmaceutical tests has experienced a significant increase during the two last decade. Currently, the quantitative analytical methods of pharmaceuticspectrophotometric, GC and HPLC methods, usually require dissolving the samples, separating them and determining their ingredients. However NIR technique is a rapid and nondestructive quantitative analytical technique of the samples with no need for reagents or solvents. Compared to conventionally analytical methods the NIR spectroscopy methods with unparalleled advantages not only attract pharmaceutical industry, but also draw more attention inresearch and exploitation areas. The PLS methods for spectra analyses are related to other multivariate calibration methods such as classical least-squares, this paper, we use cross-validation for selecting optimal numbers of loading Vectors of PLS. A predicted residual error sum of squares (PRESS) is calculated for each number of factors tested. The number of factors giving the first local minimal PRESS is considered as the optimal number of PCs. Once the calfactor analysis(FA) and principal component regression(PCR). In ibration eter with an ISR-3100 inof different dosages. The possibilities of using NIR spectroscopy is an English software. It has been translated into the Chinese during the using. model has been constructed, further samples can be analyzed directly and expeditiously. In this paper, the NIR diffuse reflectance spectra and first-derivative spectra are measured using a Shimadzu UV-3100 UV-VIS-NIR recording spectophotomtegrating sphere attachment. The objective of this paper is to study the combination of multivariate statistical techniques and NIR spectroscopy for qualitative and quantitative quality control for quality control of different powder dosages are investigated. We use multivariate statistical classification techniques to identify qualified, poor and false pharmaceuticals of different dosages from their NIR spectra successfully. In quantitative analysis, these dosages are analyzed noninvasively by NIR spectra combined with PLS method. The influence of the interval of wavelengths and the size of sample sets on the capacity of quantitative prediction of PLS method is discussed. The results are satisfactory. The PLS software used in this paper was programmed by Nicolet Corp. and it 83...