UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Transferability of near-infrared spectra for the identification of pharmaceuticals

Yoon, Weng Li; (2000) Transferability of near-infrared spectra for the identification of pharmaceuticals. Doctoral thesis (Ph.D), UCL (University College London). Green open access

[thumbnail of Transferability_of_near-infrar.pdf] Text
Transferability_of_near-infrar.pdf

Download (11MB)

Abstract

The application of near-infrared spectroscopy (NIRS) in the pharmaceutical industry is facing continued growth due to its speed and ease of use, non-destructive nature and minimal need for sample preparation. The Medicines Control Agency and pharmaceutical industry, however, have expressed their concern about the transferability of the technique. The construction of spectral libraries can be a costly and time-consuming process and therefore the ability to transfer libraries between instruments would offer tremendous savings. However, it is generally recognized that calibrations for quantification purposes are not always directly transferable between instruments. As yet, it is still to be established if libraries of spectra for identification purposes are transferable. Initially, the transferability of spectra for solvents was examined on eight different instrumental setups which included both grating and Fourier-Transform based spectrophotometers. Careful selection of wavelength range and mathematical pretreatment were found to be essential parameters for the ability to transfer spectra. The success of this part of the project was particularly marked by the outcome of an inter-laboratory trial conducted based on the above spectral library. An identification rate of almost 100% was achieved for spectra measured from standard modules. The investigation was then extended to solid pharmaceutical excipients. Effects of sample presentation wee first examined. Four important parameters were identified: cup diameter, sample thickness, cup material and packing method, which can significantly affect identification algorithms i.e. Correlation in Wavelength Space (< 0.95) and Maximum Wavelength Distance (> 3.0). Stray light and variation in the reference standards were important sources of errors to be considered. However, their effects can be corrected by measuring the spectra to a common reference standard. This work has demonstrated that libraries of spectra for purposes can be transferred between different instruments/laboratories providing factors such as spectral wavelength range, mathematical pre-treatment and sample presentation are carefully controlled.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Transferability of near-infrared spectra for the identification of pharmaceuticals
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Thesis digitised by ProQuest.
URI: https://discovery.ucl.ac.uk/id/eprint/10113557
Downloads since deposit
115Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item