Rapid classification of pharmaceutical ingredients with Raman spectroscopy using compressive detection strategy with PLS-DA multivariate filters.

Identifying pharmaceutical ingredients is a routine procedure required during industrial manufacturing. Here we show that a recently developed Raman compressive detection strategy can be employed to classify various widely used pharmaceutical materials using a hybrid supervised/unsupervised strategy in which only two ingredients are used for training and yet six other ingredients can also be distinguished. More specifically, our liquid crystal spatial light modulator (LC-SLM) based compressive detection instrument is trained using only the active ingredient, tadalafil, and the excipient, lactose, but is tested using these and various other excipients; microcrystalline cellulose, magnesium stearate, titanium (IV) oxide, talc, sodium lauryl sulfate and hydroxypropyl cellulose. Partial least squares discriminant analysis (PLS-DA) is used to generate the compressive detection filters necessary for fast chemical classification. Although the filters used in this study are trained on only lactose and tadalafil, we show that all the pharmaceutical ingredients mentioned above can be differentiated and classified using PLS-DA compressive detection filters with an accumulation time of 10ms per filter.

Rapid insight into heating-induced phase transformations in the solid state of the calcium salt of atorvastatin using multivariate data analysis.

PURPOSE: To investigate the heating-induced dehydration and melting behavior of the trihydrate phase of the calcium salt of atorvastatin. METHODS: Multivariate curve resolution (MCR) was used to decompose a variable-temperature synchrotron X-ray powder diffraction (VT-XRPD) data matrix into diffraction patterns and concentration profiles of pure drug phases. RESULTS: By means of the MCR-estimated diffraction patterns and concentration profiles, the trihydrate phase of the drug salt was found to dehydrate sequentially into two partially dehydrated hydrate structures upon heating from 25 to 110°C, with no associated breakage of the original crystal lattice. During heating from 110 to 140°C, the remaining water was lost from the solid drug salt, which instantly collapsed into a liquid crystalline phase. An isotropic melt was formed above 155°C. Thermogravimetric analysis, hot-stage polarized light microscopy, and hot-stage Raman spectroscopy combined with principal component analysis (PCA) was shown to provide consistent results. CONCLUSIONS: This study demonstrates that MCR combined with VT-XRPD is a powerful tool for rapid interpretation of complex dehydration behavior of drug hydrates, and it is also the first report on a liquid crystalline phase of the calcium salt of atorvastatin.

Analysis of counterfeit Cialis® tablets using Raman microscopy and multivariate curve resolution.

Counterfeit medicines have become a serious global problem. Consequently, analytical and pharmaceutical scientists have been active in developing and applying new methodologies to detect and analyze counterfeit medicines. Vibrational spectroscopy combined with chemometric methods is becoming a popular choice in this area of research. In this study, Raman microscopy was used to collect chemical images of counterfeit tadalafil tablets and multivariate curve resolution (MCR) was used to analyze the Raman spectra and reveal the identities of the excipients and active pharmaceutical ingredients (API) in each tablet. Resolved counterfeit tablet spectra were compared to the resolved genuine tablet spectra. Both similarities and dissimilarities were revealed by the analysis in terms of the identity of the excipients, the quantity of the API, and the distribution of the components. It was concluded that Raman microscopy combined with MCR is a powerful method to detect and analyze counterfeit tablets.

Kinetics of moisture-induced hydrolysis in powder blends stored at and below the deliquescence relative humidity: investigation of sucrose-citric acid mixtures.

Previous studies have shown that deliquescent organic compounds frequently exhibit chemical instability when stored in environmental conditions above their deliquescence relative humidity (RH). The goal of the current study was to investigate the effect of atmospheric moisture on the long-term chemical stability of crystalline sucrose-citric acid mixtures following storage at RHs at and below the mutual deliquescence relative humidity (MDRH). Interestingly, it was found that sucrose hydrolysis can occur below the MDRH of 64% and was observed for samples stored at 54% RH. However, hydrolysis was not seen for samples stored at 33 or 43% RH. The rate of sucrose hydrolysis could be modeled by taking into account the rate and extent of moisture uptake, which in turn was dependent on the composition of the powder and the storage RH. A reaction mechanism initiated by capillary condensation and involving additional deliquescence lowering by the degradation products formed as a result of sucrose hydrolysis (glucose and fructose) was proposed.

Determination of active pharmaceutical ingredients by heteroatom selective detection using inductively coupled plasma mass spectrometry with ultrasonic nebuilization and membrane desolvation sample introduction.

The combination of ultrasonic nebulization with membrane desolvation (USN-MD) is utilized to determine active pharmaceutical ingredients (API) by heteroatom inductively coupled mass spectroscopy (ICP-MS) detection. Ultrasonic nebulization provides efficient sampling while use of the membrane desolvator acts to reduce solvent-based interferences. This approach reduces interferences sufficiently so that a standard argon ICP-quadrupole MS can be utilized. Examined APIs and associated heteroatoms included: phosphomycin (P), amoxicillin (S), chlorpropamide (Cl), and ofloxacin (F). The optimum plasma r.f. powers for P, S, and Cl were in the 1000 to 1200 watts range. The high ionization energy of F required that the plasma be operated at 1500 W. The 16O2+ interference at mass 32 precluded determinations using the sulfur-32. The sulfur-34 (4.2% natural isotopic abundance), however, was relatively free of isobaric interferences. Interferences were relatively small at the mass 35 isotope of Cl, but increased with higher ICP r.f. powers. Overlaps were significant at the masses of monoisotopic species, fluorine-19 and phosphorus-31. Detection limits for P, S, Cl, and F of 2, 3, 90, and 3000 ng/mL, respectively, were generally lower than those produced with other quadrupole systems and comparable to or better than values published utilizing high-resolution instruments.

Effects of liquid chromatography mobile phases and buffer salts on phosphorus inductively coupled plasma atomic emission and mass spectrometries utilizing ultrasonic nebulization and membrane desolvation.

Atomic spectrometry, specifically inductively coupled plasma atomic emission spectrometry (ICP-AES) and mass spectrometry (ICP-MS) show promise for heteroatom-based detection of pharmaceutical compounds. The combination of ultrasonic nebulization (USN) with membrane desolvation (MD) greatly enhances detection limits with these approaches. Because pharmaceutical analyses often incorporate liquid chromatography, the study herein was performed to examine the effects of solvent composition on the analytical behaviors of these approaches. The target analyte was phosphorus, introduced as phosphomycin. AES response was examined at the 253.7 nm atom line and mass 31 ions were monitored for the MS experiments. With pure aqueous solutions, detection limits of 5 ppb (0.5 ng in 0.1 mL injection volumes) were obtained with ICP-MS. The ICP-AES detection limit was 150 ppb. Solvent compositions were varied from 0 to 80% organic (acetonitrile and methanol) with nine buffers at concentrations typically used in liquid chromatography. In general, solvents and buffers had statistically significant, albeit small, effects on ICP-AES sensitivities. A few exceptions occurred in cases where typical liquid chromatography buffer concentrations produced higher mass loadings on the plasma. Indications are that isocratic separations can be reliably performed. Within reasonable accuracy tolerances, it appears that gradient chromatography can be performed without the need for signal response normalization. Organic solvent and buffer effects were more significant with ICP-MS. Sensitivities varied significantly with different buffers and organic solvent content. In these cases, gradient chromatography will require careful analytical calibration as solvent and buffer content is varied. However, for most buffer and solvent combinations, signal and detection limits are only moderately affected. Isocratic separations and detection are feasible.