Edenir Rodrigues Pereira-Filho

Rapid detection and quantification of milk adulteration using infrared microspectroscopy and chemometrics analysis

The application of attenuated total reflectance mid-infrared microspectroscopy (MIR-microspectroscopy) was evaluated as a rapid method for detection and quantification of milk adulteration. Milk samples were purchased from local grocery stores (Columbus, OH, USA) and spiked at different concentrations of whey, hydrogen peroxide, synthetic urine, urea and synthetic milk. Samples were place on a 192-well microarray slide, air-dried and spectra were collected by using MIR-microspectroscopy. Pattern recognition analysis by Soft Independent Modeling of Class Analogy (SIMCA) showed tight and well-separated clusters allowing discrimination of control samples from adulterated milk. Partial Least Squares Regression (PLSR) showed standard error of prediction (SEP) ~2.33, 0.06, 0.41, 0.30 and 0.014 g/L for estimation of levels of adulteration with whey, synthetic milk, synthetic urine, urea and hydrogen peroxide, respectively. Results showed that MIR-microspectroscopy can provide an alternative methodology to the dairy industry for screening potential fraudulent practice for economic adulteration of cows milk.

Evaluation of biodiesel-diesel blends quality using 1H NMR and chemometrics.

In this work, the use of (1)H NMR spectroscopy and statistical approach to the evaluation of biodiesel-diesel blends quality is described. Forty-six mixtures of oil-diesel, biodiesel-diesel, and oil-biodiesel-diesel were analyzed by (1)H NMR and such data were employed to design four predictive models. Thirty-six mixtures were used in the calibration set and the others in the validation. The PCR and PLS models were evaluated through statistical parameters. Briefly, PLS and PCR models were suitable for the prediction of biodiesel and oil concentration in mineral diesel. Specially, in higher concentration the predicted values were quite similar to the real ones. This fact was evidenced by the low relative errors of high concentrated samples; this means that the prediction of low concentrated samples will probably show high deviation. Therefore, (1)H NMR-PLS and (1)H NMR-PCR methods are fairly useful for the quality control of biodiesel-diesel blends, particularly they are suitable for prediction of concentrations greater than 2%.

Simultaneous sample digestion and determination of Cd, Cu and Pb in biological samples using thermospray flame furnace atomic absorption spectrometry (TS-FF-AAS) with slurry sample introduction

In a Ni tube placed in an oxidizing air/acetylene AAS flame simultaneous digestion/oxidation of biological samples and powerful trace determinations were performed. A ceramic capillary was used for sample introduction. This capillary acted as a flame-heated thermospray nozzle (TS-FF-AAS). Slurry samples, prepared in 0.3 mol L−1 HNO3, were transported by a new 0.6 MPa (84 psi) flow injection system through the hot capillary into the flame tube furnace. A nitric acid vapour environment exists inside the tube at temperatures above 1000 °C oxidising the samples. The element traces could then be measured by AAS. For these investigations, 300 µL slurry samples (0.1 to 1.0% w/v) from various biological materials, such as pig kidney (CRM 186), bovine liver (SRM 1577), spinach (SRM 1570), orchard leaves (SRM 1571) citrus leaves (SRM 1572), tomato leaves (SRM 1573) and tomato puree were used. Reference values were additionally obtained by electrothermal atomic absorption spectrometry (ETAAS). The standard deviation (n = 10) was between 3 and 5% using the same slurry and about 10% when the whole process was repeated on subsequent days. The detection limits for Cd, Cu and Pb were found to be 0.5, 4.3 and 3.5 µg g−1, respectively. A t-test showed no differences between the found and certified values at a 95% confidence level. In all experiments the calibration could be performed with aqueous standard solutions.

Twelve different types of data normalization for the proposition of classification, univariate and multivariate regression models for the direct analyses of alloys by laser-induced breakdown spectroscopy (LIBS)

This study applies laser-induced breakdown spectroscopy (LIBS) for the direct analysis of 80 metal samples (alloys and steel) for multivariate and univariate regression models, aiming at the determination of 10 analytes (Al, Cr, Cu, Fe, Mn, Mo, Ni, Ti, V and Zn). To optimize the LIBS system, the Doehlert design was used for energy, delay time and spot size adjustment for all samples and analytes. Twelve normalization modes were used to reduce the interference matrix and to improve the calibration models, with error values ranging from 0.27% (Mn) to 14% (Cu and Ni). Models without normalization presented two- to five-fold higher errors. In addition to quantification, classification models (KNN, SIMCA and PLS-DA) were also proposed for sample differentiation. Multivariate and univariate models presented similar performance, and among the classification models, KNN presented the best results, with an accuracy of 100%.

Fingerprinting of anthocyanins from grapes produced in Brazil using HPLC–DAD–MS and exploratory analysis by principal component analysis

The anthocyanin profile was studied for 11 types of grapes of different varieties and geographical origins, including 10 wine grape varieties (Vitis vinifera) and one hybrid variety. Twenty anthocyanins were identified by means of the absorbance spectrum and fragmentation pattern by tandem mass spectrometry. The multivariate method of principal component analysis (PCA) was used to evaluate differences amongst the cultivars. The results show the hybrid grape separated from a cluster represented by viniferas grapes, mainly due to anthocyanin diglucosides. Within this group, it was possible to discriminate the different wine grapes. Clusters discriminating according to geographical origin were not observed.

Application of hand-held and portable infrared spectrometers in bovine milk analysis.

A simple and fast method for the detection and quantification of milk adulteration was developed using portable and hand-held infrared (IR) spectrometers. Milk samples were purchased from local supermarkets (Columbus, OH, USA) and spiked with tap water, whey, hydrogen peroxide, synthetic urine, urea, and synthetic milk in different concentrations. Spectral data were collected using mid-infrared (MIR) and near-infrared (NIR) spectrometers. Soft independent modeling of class analogy (SIMCA) classification models exhibited tight and well-separated clusters allowing the discrimination of control from adulterated milk samples. Partial least-squares regression (PLSR) was used to estimate adulteration levels, and results showed high coefficients of determination (R(2)) and low standard errors of prediction (SEP). Classification and quantification models indicated that the tested MIR systems were superior to NIR systems in monitoring milk adulteration. This method can be potentially used as an alternative to traditional methods due to their simplicity, sensitivity, low energy cost, and portability.

Recent advances on determination of milk adulterants

Milk adulteration is a current fraudulent practice to mask the quality parameters (e.g. protein and fat content) and increase the product shelf life. Milk adulteration includes addition of toxic substances, such as formaldehyde, hydrogen peroxide, hypochlorite, dichromate, salicylic acid, melamine, and urea. In order to assure the food safety and avoid health risks to consumers, novel analytical procedures have been proposed for detection of these adulterants. The innovations encompass sample pretreatment and improved detection and data processing, including chemometric tools. This review focuses on critical evaluation of analytical approaches for assay of milk adulteration, with emphasis on applications published after 2010. Alternatives for fast, environmentally friendly and in-situ detection of milk adulterants are highlighted.

Digital image analysis – an alternative tool for monitoring milk authenticity

This paper proposes the use of digital imaging as an alternative method for the identification and quantification of milk adulteration. Bovine milk samples were furnished by a local milk company (Sao Carlos, SP, Brazil) and spiked with tap water, whey, hydrogen peroxide, synthetic urine and synthetic milk at different levels of adulteration. Digital images were obtained by a flatbed scanner and the means of ten color descriptors (red, green, blue, hue, saturation, value, luminosity and the three relative colors, i.e. relative red, relative blue and relative green) were used to evaluate the information from the images. Classification models exhibited tight and well-separated clusters allowing the discrimination of control from adulterated milk samples. Regression models were developed by using partial least squares (PLS) and principal component regression (PCR) and results showed that the method could be applied to estimate adulteration levels in milk with high coefficients of correlation (r) and low standard error of prediction (SEP). This method can be potentially used as an alternative to the traditional methods due the simplicity, sensitivity, low cost and portability allowing in situ analysis.

Analysis of the polymeric fractions of scrap from mobile phones using laser-induced breakdown spectroscopy: Chemometric applications for better data interpretation

Because of their short life span and high production and consumption rates, mobile phones are one of the contributors to WEEE (waste electrical and electronic equipment) growth in many countries. If incorrectly managed, the hazardous materials used in the assembly of these devices can pollute the environment and pose dangers for workers involved in the recycling of these materials. In this study, 144 polymer fragments originating from 50 broken or obsolete mobile phones were analyzed via laser-induced breakdown spectroscopy (LIBS) without previous treatment. The coated polymers were mainly characterized by the presence of Ag, whereas the uncoated polymers were related to the presence of Al, K, Na, Si and Ti. Classification models were proposed using black and white polymers separately in order to identify the manufacturer and origin using KNN (K-nearest neighbor), SIMCA (Soft Independent Modeling of Class Analogy) and PLS-DA (Partial Least Squares for Discriminant Analysis). For the black polymers the percentage of correct predictions was, in average, 58% taking into consideration the models for manufacturer and origin identification. In the case of white polymers, the percentage of correct predictions ranged from 72.8% (PLS-DA) to 100% (KNN).

Performance evaluation of collision-reaction interface and internal standardization in quadrupole ICP-MS measurements.

The combined use of internal standardization with collision and reaction interface in an inductively coupled plasma quadrupole mass spectrometer (ICP-QMS-CRI) was evaluated. The behavior of several elements (Ba, Cd, Co, Cr, Cu, Li, Mn, Mo, Pb, Sb, V and Zn) was studied by introducing H(2) or He through the skimmer and sampler cones of the CRI device and by using In, Rh and Sc as internal standards. Certified reference material of trace elements in water (NIST 1643e) was used to evaluate the performance of the method. A vinegar sample and mixed food diet standard reference material were directly introduced into the equipment as complex matrices for Cr and V determinations. Improvements in accuracy and precision were attained combining both strategies. The introduction of H(2) through the skimmer cone was the best way to overcome polyatomic ions formation and to improve SBR and BEC values for several elements. The use of Sc as internal standard improved the performance of ICP-QMS-CRI.