Moisture determination in an active pharmaceutical ingredient (API) by near infrared spectroscopy

The development of computers, chemometric software, and fiber optics has increased the interest of the pharmaceutical industry in Near-infrared spectroscopy.;During many years the traditional method to determine water in a sample has been the Karl Fischer titration. Although the Karl Fischer titration method is adaptable to many types of samples it requires an equilibration time in addition to the analyses time, and requires the use of toxic reagents. Near infrared spectroscopy can be used as alternative to the Karl Fischer assay. Near infrared spectroscopy is advantageous for the moisture determination because it is rapid, requires little sample pretreatment and is nondestructive.;In this work, the moisture content of an active pharmaceutical ingredient (API) was determined in the final step of its production. Two diffuse reflectance accessories were used for the collection of the near infrared radiation. The moisture determination in an API was performed with these methods. The multicalibration methods (PLS and MLR) were used to create the calibration model. Diffuse reflectance using the fiber optics probe and Partial Least. Square (PLS) gave better results for the calibration model. The method was validated using samples from different lots. The linearity, accuracy, precision, ruggedness, and robustness of the model were also tested to evaluate the method. Although the analyses were carried out near the limit of quantification of the instrument the results obtained are very similar to those obtained with the currently used Karl Fischer method.