Laura Mateo-Vivaracho

Analysis, occurrence, and potential sensory significance of five polyfunctional mercaptans in white wines.

A previously developed analytical method has been improved, validated and adapted for the analysis of 2-furfurylthiol (FFT), 4-methyl-4-mercapto-2-pentanone (MP), 3-mercaptohexyl acetate (MHA), 3-mercaptohexanol (MH) and benzylmercaptan (BM) in 136 white wines from different parts of the world. The overall uncertainty of the determinations was found to be around 20%, which was considered satisfactory given the low levels at which these compounds are found. The levels ranged from the method detection limits (0.5; 0.6; 2.0; 8.0 and 0.5 ng/L for FFT, MP, MHA, MH and BM, respectively) to 225; 87.9; 591; 7255 and 131 ng/L, which implies that nearly all of them can reach more than 100 Odor Units in some wines. The levels are significantly linked to both the grape variety (with the exception of FFT) and to the origin (in the case of Sauvignon Blanc samples), however, the range of variation within groups are so large that clear clusters could not be observed. Different sensory tests carried out on white wine models showed that all these compounds, even at low concentration, play an outstanding role on the aroma of wine, contributing to fruity, fresh and green notes. In some wines they are at concentrations high enough to act as genuine impact compounds.

Improved solid-phase extraction procedure for the isolation and in-sorbent pentafluorobenzyl alkylation of polyfunctional mercaptans. Optimized procedure and analytical applications.

A fast method for the determination of aroma-powerful polyfunctional thiols at ng L(-1) level has been developed. Mercaptans are selectively retained in a 50mg solid-phase extraction cartridge and derivatization takes place in the same cartridge at room temperature (25 degrees C) in 20 min by adding small amounts of pentafluorobenzyl bromide (PFBBr) and a strong alkali: 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU). The corresponding derivatives are further eluted and determined by gas chromatography-negative ion mass spectrometry (GC-NCI-MS). Isolation, derivatization, clean-up, elution and calibration conditions were examined. Carrying the reaction in the cartridge makes it possible to use water and non-polar reagents simultaneously, to avoid large volumes of toxic solvent, and to eliminate the excess of reagent. This was last accomplished by the reaction with mercaptoglycerol and further rinsing with a hydromethanolic solution. The method makes it possible to simultaneously determine 2-methyl-3-furanthiol (MF), 2-furfurylthiol (2-furanmethanethiol) (FFT), 4-mercapto-4-methyl-2-pentanone (MP) (as its methoxime), 3-mercaptohexylacetate (MHA) and 3-mercaptohexanol (MH). Absolute limits of detection were 0.2 (MF), 0.1 (FFT), 0.1 (MP), 0.3 (MHA) and 2 (MH) ng L(-1). Repeatability (1%<RSD<20%) and linearity (0.978<R(2)<0.999) were satisfactory. Problems with matrix effects were solved by the use of deuterated analogues for MP, MHA and MH and by avoiding the oxidation of analytes and standards via addition of cysteine and EDTA. The different aspects of the method optimization are discussed.

Comparison of extraction techniques and mass spectrometric ionization modes in the analysis of wine volatile carbonyls.

This work presents a comparative study of the analytical characteristics of two methods for the analysis of carbonyl compounds in wine, both based on the derivatization with O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride (PFBHA). In the first method derivatives are formed in the solid phase extraction (SPE) cartridge in which the analytes have been previously isolated, while in the second method derivatives are formed in a solid phase microextraction (SPME) fibre saturated with vapors of the reagent and exposed to the sample headspace. In both cases detection has been carried out by electron impact (EI) or negative chemical ionization (NCI) mass spectrometry. The possibility of determining haloanisols simultaneously has been also considered. The method based on SPE presents, in general, better analytical properties than the SPME one. Although linearity was satisfactory for both methods (R(2)>0.99), repeatability of the SPE method (RSD<10%) was better than that obtained with SPME (9% < RSD < 20%). Detection limits obtained with EI are better for the SPE method except for trihaloanisols, while with NCI detection limits for both strategies are comparable, although the SPME strategy presents worse results for ketones and methional. Detection limits are always lower with NCI, being the improvement most notable for SPME. Recovery experiments show that in the case of SPE, uncertainties are lower than 12% in all cases, while with the SPME method the imprecision plus the existence of matrix effects make the global uncertainty to be higher than 15%.

Automated and quantitative headspace in-tube extraction for the accurate determination of highly volatile compounds from wines and beers

An automatic headspace in-tube extraction (ITEX) method for the accurate determination of acetaldehyde, ethyl acetate, diacetyl and other volatile compounds from wine and beer has been developed and validated. Method accuracy is based on the nearly quantitative transference of volatile compounds from the sample to the ITEX trap. For achieving that goal most methodological aspects and parameters have been carefully examined. The vial and sample sizes and the trapping materials were found to be critical due to the pernicious saturation effects of ethanol. Small 2 mL vials containing very small amounts of sample (20 μL of 1:10 diluted sample) and a trap filled with 22 mg of Bond Elut ENV resins could guarantee a complete trapping of sample vapors. The complete extraction requires 100 × 0.5 mL pumping strokes at 60 °C and takes 24 min. Analytes are further desorbed at 240 °C into the GC injector under a 1:5 split ratio. The proportion of analytes finally transferred to the trap ranged from 85 to 99%. The validation of the method showed satisfactory figures of merit. Determination coefficients were better than 0.995 in all cases and good repeatability was also obtained (better than 7% in all cases). Reproducibility was better than 8.3% except for acetaldehyde (13.1%). Detection limits were below the odor detection thresholds of these target compounds in wine and beer and well below the normal ranges of occurrence. Recoveries were not significantly different to 100%, except in the case of acetaldehyde. In such a case it could be determined that the method is not able to break some of the adducts that this compound forms with sulfites. However, such problem was avoided after incubating the sample with glyoxal. The method can constitute a general and reliable alternative for the analysis of very volatile compounds in other difficult matrixes.

Selective preconcentration of volatile mercaptans in small SPE cartridges: quantitative determination of trace odor-active polyfunctional mercaptans in wine.

A general procedure for the selective preconcentration and purification of mercaptans has been developed. Mercaptans are strongly retained in a small (20 mg) SPE cartridge containing p-hydroxymercurybenzoate. The cartridge can then be rinsed with relatively high volumes of polar (water/methanol mixtures) and non-polar (pentane or pentane/ether mixtures) rinsing solutions to remove nearly all volatile compounds lacking a thiol functionality. Retained analytes are further eluted with a small volume of an organic solvent containing 1,4-dithioerythritol. Some basic aspects of the strategy, such as the retention of p-hydroxymercurybenzoate in the sorbent and its stability versus different rinsing and eluting systems, have been studied in depth. Light sulfur compounds contained in water or wine, including mercaptans such as methanethiol or thioethers, such as diethyl sulfide, can be quantitatively extracted, although only mercaptans can be quantitatively recovered if a polar rinsing is applied. The strategy has been applied to the GC-MS quantitative determination of some trace polyfunctional mercaptans that are key aromas in wine, such as 2-methyl-3-furanthiol, 2-furfurylthiol, 4-mercapto-4-methyl-2-pentanone, 3-mercaptohexyl acetate or 3-mercaptohexanol. The developed method reaches detection limits in the ng/L range and has a satisfactory analytical behavior, being quite simple and fast.

Fast and fully automated analytical method for the screening of residues of aziridine and 2-chloroethylamine in pharmaceutical active principles

A simple, fast and fully automated method for the screening of aziridine (AZD) and 2-chloroethylamine (CEA) in active pharmaceutical ingredients (API) has been developed. The method is based on the in-fiber derivatization of the amines extracted from the sample headspace (previously dissolved or suspended in alkaline water) with 2,3,4,5,6-pentafluorobenzoyl chloride (PFBCl) previously adsorbed in the PDMS/DVB solid phase microextraction (SPME) fiber. The derivatives formed are further desorbed and analyzed in a gas chromatograph with negative ion chemical ionization mass spectrometry (GC-NCI-MS) using methane as reagent gas. The different operational parameters of the procedure have been optimized to get highest sensitivity. The validation of the method, however, revealed a poor repeatability, particularly evident in water-soluble APIs (RSD>20% for AZD). In spite of that, the low detection limits (1-3ngg(-1) for AZD and CEA), speed (44min total analysis time) and automation make that this method can be satisfactorily used as screening tool to accept or reject API batches attending to their volatile amine content and a critical specified value derived from the 1.5μg/day Threshold of Toxicological Concern (TTC) and maxima daily dosages. This was shown by analyzing seventy-five fluvoxamine maleate samples containing known levels of AZD and CEA (between 0.05 and 1.05μgg(-1)) in intermediate reproducibility conditions to get reliable estimations of precision and linearity. From these data, acceptance, rejection and non-conclusive areas of response are defined for both analytes at different confidence and replication levels using normal statistics. The method was satisfactorily applied to real fluvoxamine maleate samples.

Optimisation and validation of a taste dilution analysis to characterise wine taste

Taste dilution analysis (TDA) has been applied to the characterisation of tastants in wine. The wines were concentrated under vacuum at 20°C and with a stream of N2 to obtain a completely dry extract without aroma compounds. The dry extracts were redissolved in water and injected in a semi-preparative C18 HPLC column. A thorough study to determine reproducibility and maximum load in the HPLC system was carried out. The effluent was separated in fractions, which were collected and concentrated to dryness under the same conditions that wine extracts, and then, again dissolved in bottled water. Different dilutions of the fractions were tasted by a sensory panel to assess the intensity of basic tastes and sensations in the mouth. Fractions were also submitted to HPLC-MS analysis to determine the identity of the tastants.