Jez Willian Batista Braga

Development and analytical validation of a multivariate calibration method for determination of amoxicillin in suspension formulations by near infrared spectroscopy

This paper proposes a new method for determination of amoxicillin in pharmaceutical suspension formulations, based on transflectance near infrared (NIR) measurements and partial least squares (PLS) multivariate calibration. A complete methodology was implemented for developing the proposed method, including an experimental design, data preprocessing by using multiple scatter correction (MSC) and outlier detection based on high values of leverage, and X and Y residuals. The best PLS model was obtained with seven latent variables in the range from 40.0 to 65.0 mg mL(-1) of amoxicillin, providing a root mean square error of prediction (RMSEP) of 1.6 mg mL(-1). The method was validated in accordance with Brazilian and international guidelines, through the estimate of figures of merit, such as linearity, precision, accuracy, robustness, selectivity, analytical sensitivity, limits of detection and quantitation, and bias. The results for determinations in four commercial pharmaceutical formulations were in agreement with the official high performance liquid chromatographic (HPLC) method at the 99% confidence level. A pseudo-univariate calibration curve was also obtained based on the net analyte signal (NAS). The proposed chemometric method presented the advantages of rapidity, simplicity, low cost, and no use of solvents, compared to the principal alternative methods based on HPLC.

Near infrared spectroscopy (NIRS) as a potential tool for monitoring trade of similar woods: Discrimination of true mahogany, cedar, andiroba, and curupixá

Abstract Mahogany is one of the most valuable woods and was widely used until it was included in Appendix II of the Convention on International Trade in Endangered Species as endangered species. Mahogany wood sometimes is traded under different names. Also, some similar woods belonging to the Meliaceae family are traded as “mahogany” or as being of a “mahogany pattern”. To investigate the feasibility of the use of near infrared spectroscopy for wood discrimination, the mahogany (Swietenia macrophylla King.), andiroba or crabwood (Carapa guianensis Aubl.), cedar (Cedrela odorata L.), and curupixá (Micropholis melinoniana Pierre) woods were examined. Four discrimination models based on partial least squares-discriminant analysis were developed based on a calibration set composed of 88 samples and a test set with 44 samples. Each model corresponds to the discrimination of a wood species from the others. Optimization of the model was performed by means of the OPUS® software followed by statistical analysis software (Matlab®). The observed root mean square errors of predictions were 0.14, 0.09, 0.12, and 0.06 for discriminations of mahogany, cedar, andiroba, and curupixá, respectively. The separations of the species obtained based on the difference in the predicted values was at least 0.38. This makes it possible to perform safe discriminations with a very low probability of misclassifying a sample. This method can be considered accurate and fast.

Optimization and validation of a LIBS method for the determination of macro and micronutrients in sugar cane leaves.

Laser induced breakdown spectrometry (LIBS) was applied for the determination of macro (P, K, Ca, Mg) and micronutrients (B, Cu, Fe, Mn and Zn) in sugar cane leaves, which is one of the most economically important crops in Brazil. Operational conditions were previously optimized by a neuro-genetic approach, by using a laser Nd:YAG at 1064 nm with 110 mJ per pulse focused on a pellet surface prepared with ground plant samples. Emission intensities were measured after 2.0 μs delay time, with 4.5 μs integration time gate and 25 accumulated laser pulses. Measurements of LIBS spectra were based on triplicate and each replicate consisted of an average of ten spectra collected in different sites (craters) of the pellet. Quantitative determinations were carried out by using univariate calibration and chemometric methods, such as PLSR and iPLS. The calibration models were obtained by using 26 laboratory samples and the validation was carried out by using 15 test samples. For comparative purpose, these samples were also microwave-assisted digested and further analyzed by ICP OES. In general, most results obtained by LIBS did not differ significantly from ICP OES data by applying a t-test at 95% confidence level. Both LIBS multivariate and univariate calibration methods produced similar results, except for Fe where better results were achieved by the multivariate approach. Repeatability precision varied from 0.7 to 15% and 1.3 to 20% from measurements obtained by multivariate and univariate calibration, respectively. It is demonstrated that LIBS is a powerful tool for analysis of pellets of plant materials for determination of macro and micronutrients by choosing calibration and validation samples with similar matrix composition.

The use of near Infrared Spectroscopy to Identify solid wood Specimens of Swietenia Macrophylla0 (Cites Appendix II)

Near infrared spectroscopy (NIRS) has been shown effective as a tool for identifying Swietenia when tested as laboratory-processed powder, but testing such powdered wood is not readily adaptable to the fieldidentification of wood. This study explored the efficacy of a fiber optic NIRS scan of solid wood surfaces to separate Swietenia macrophylla King, Carapa guianensis Aubl., Cedrela odorata L., and Micropholis melinoniana Pierre. Transverse, radial, and tangential surfaces were scanned to determine if the surface from which data were collected influenced the spectra recorded. Surfaces were scanned before and after removing the oxidized surface layer of the blocks to test effects of exposure on the spectra. Partial least squares for discriminant analysis models were developed for each taxon separately, based on a calibration set composed of at least 67 samples and a test set with at least 45 samples. The anatomical surface scanned, but not the presence of an oxidized layer, influenced the spectra for each species, necessitating the comparison of the same planes of section. The discriminant models showed small errors for each species, indicating that reliable identifications can be made with NIRS of solid wood surfaces in these species.

Determination of vegetable oils and fats adulterants in diesel oil by high performance liquid chromatography and multivariate methods.

The current legislation requires the mandatory addition of biodiesel to all Brazilian road diesel oil A (pure diesel) marketed in the country and bans the addition of vegetable oils for this type of diesel. However, cases of irregular addition of vegetable oils directly to the diesel oil may occur, mainly due to the lower cost of these raw materials compared to the final product, biodiesel. In Brazil, the situation is even more critical once the country is one of the largest producers of oleaginous products in the world, especially soybean, and also it has an extensive road network dependent on diesel. Therefore, alternatives to control the quality of diesel have become increasingly necessary. This study proposes an analytical methodology for quality control of diesel with intention to identify and determine adulterations of oils and even fats of vegetable origin. This methodology is based on detection, identification and quantification of triacylglycerols on diesel (main constituents of vegetable oils and fats) by high performance liquid chromatography in reversed phase with UV detection at 205nm associated with multivariate methods. Six different types of oils and fats were studied (soybean, frying oil, corn, cotton, palm oil and babassu) and two methods were developed for data analysis. The first one, based on principal component analysis (PCA), nearest neighbor classification (KNN) and univariate regression, was used for samples adulterated with a single type of oil or fat. In the second method, partial least square regression (PLS) was used for the cases where the adulterants were mixtures of up to three types of oils or fats. In the first method, the techniques of PCA and KNN were correctly classified as 17 out of 18 validation samples on the type of oil or fat present. The concentrations estimated for adulterants showed good agreement with the reference values, with mean errors of prediction (RMSEP) ranging between 0.10 and 0.22% (v/v). The PLS method was efficient in the quantification of mixtures of up to three types of oils and fats, with RMSEP being obtained between 0.08 and 0.27% (v/v), mean precision between 0.07 and 0.32% (v/v) and minimum detectable concentration between 0.23 and 0.81% (v/v) depending on the type of oil or fat in the mixture determined.

Discrimination of whisky brands and counterfeit identification by UV–Vis spectroscopy and multivariate data analysis

The discrimination of whisky brands and counterfeit identification were performed by UV-Vis spectroscopy combined with partial least squares for discriminant analysis (PLS-DA). In the proposed method all spectra were obtained with no sample preparation. The discrimination models were built with the employment of seven whisky brands: Red Label, Black Label, White Horse, Chivas Regal (12years), Ballantines Finest, Old Parr and Natu Nobilis. The method was validated with an independent test set of authentic samples belonging to the seven selected brands and another eleven brands not included in the training samples. Furthermore, seventy-three counterfeit samples were also used to validate the method. Results showed correct classification rates for genuine and false samples over 98.6% and 93.1%, respectively, indicating that the method can be helpful for the forensic analysis of whisky samples.

Discrimination and quantification of cocaine and adulterants in seized drug samples by infrared spectroscopy and PLSR

Middle infrared spectroscopy and multivariate analysis have been applied for the development of methods to perform both quantitative and qualitative analysis of real drug samples seized by the Brazilian Police Federal (BPF). Currently, quantification of cocaine and determination of adulterants in seizures is performed using gas chromatography with flame ionization detection. However, this technique requires a relatively complex sample preparation, higher time of analysis, the destruction of sample and a high cost. In this context, this paper presents a simpler method to quantify cocaine and its major adulterants in seized materials. Out of 375 seizures, taken within a time frame of 2009-2013. A total of 1085 samples were analyzed of which 500 were selected for the calibration set and 585 for the validation set. Cocaine concentration in seized samples was determined by using middle infrared spectroscopy and partial least squares regression (PLSR), obtaining an average prediction error of 3.0% (w/w), precision of 2.0 and 11.8% (w/w) of minimum detectable cocaine concentration in a range varying from 24.2 to 99.9% (w/w). Results indicate that the developed method is able to discriminate between cocaine hydrochloride and free base samples, to quantify cocaine content as well as to estimate the concentration of main adulterants phenacetin, benzocaine, caffeine, lidocaine and aminopyrine.

Assessment of total phenols and extractives of mahogany wood by near infrared spectroscopy (NIRS)

Abstract The resistance to decomposition of mahogany wood (Sweitenia macrophylla King) ranges from high to moderate level. Wood extractives, mainly due to the presence of phenol compounds are related to the natural durability of wood. The technique of near infrared spectroscopy (NIRS) coupled with multivariate analysis has been applied to assess the extractives and phenols of 41 samples of mahogany in powder form. The hot water-soluble extractives were quantitatively determined, and total phenol content was measured with the Folin-Denis colourimetric reagent. Models were developed with the NIRS data for each of the two variables. The results indicated that NIRS can be a useful tool to a rapid evaluation of the extractive contents and total phenolic compounds of mahogany wood. The method was able to predict the interesting properties with errors lower than 10% (w/w) and had the capability of detecting samples that have a minimum concentration of 2.4% (w/w) of extractives and total phenolic compounds, respectively.

Quantification of cocaine hydrochloride in seized drug samples by infrared spectroscopy and PLSR

The determination of cocaine in drug samples is an important task for law enforcement agencies such as the Brazilian Federal Police (BFP). In this sense, this paper proposes a method based on infrared spectra obtained by attenuated total reflectance (ATR) and partial least squares regression (PLSR) to quantify cocaine hydrochloride in drug samples. The method was developed and validated with 275 actual samples of drugs seized by the BFP. The determination was performed between 35 to 99% (m/m) of cocaine in the drug samples. Results indicate that the method is able to directly analyze drug samples containing cocaine in its hydrochloride form without any sample preparation with average prediction errors of 3.00% (m/m), 1.50% (m/m) precision and 13% (m/m) of minimum detectable concentration.

Correlation of cocaine hydrochloride samples seized in Brazil based on determination of residual solvents: an innovative chemometric method for determination of linkage thresholds.

Cocaine sample correlation provides important information in the identification of traffic networks. However, available methods for estimating if samples are linked or not require the use of previous police investigation and forensic expert knowledge regarding the number of classes and provide thresholds that are both static and data set specific. In this paper, a novel unsupervised linkage threshold method (ULT) based on chemometric analysis is described and applied to the analysis of headspace gas chromatography mass spectrometry (HS-GC/MS) data of more than 250 real cocaine hydrochloride samples seized by Brazilian Federal Police. The method is capable of establishing linkage thresholds that do not require any prior information about the number of classes or distribution of the samples and can be dynamically updated as the data set changes. It is envisaged that the ULT method may also be applied to other forensic expertise areas where limited population knowledge is available and data sets are continually modified with the inflow of new information.