Extension of a dynamic headspace multi-volatile method to milliliter injection volumes with full sample evaporation: Application to green tea.

An extension of multi-volatile method (MVM) technology using the combination of a standard dynamic headspace (DHS) configuration, and a modified DHS configuration incorporating an additional vacuum module, was developed for milliliter injection volume of aqueous sample with full sample evaporation. A prior step involved investigation of water management by weighing of the water residue in the adsorbent trap. The extended MVM for 1 mL aqueous sample consists of five different DHS method parameter sets including choice of the replaceable adsorbent trap. An initial two DHS sampling sets at 25°C with the standard DHS configuration using a carbon-based adsorbent trap target very volatile solutes with high vapor pressure (>10 kPa) and volatile solutes with moderate vapor pressure (1-10 kPa). Subsequent three DHS sampling sets at 80°C with the modified DHS configuration using a Tenax TA trap target solutes with low vapor pressure (<1 kPa) and/or hydrophilic characteristics. After the five sequential DHS samplings using the same HS vial, the five traps are sequentially desorbed with thermal desorption in reverse order of the DHS sampling and the desorbed compounds are trapped and concentrated in a programmed temperature vaporizing (PTV) inlet and subsequently analyzed in a single GC-MS run. Recoveries of 21 test aroma compounds in 1 mL water for each separate DHS sampling and the combined MVM procedure were evaluated as a function of vapor pressure in the range of 0.000088-120 kPa. The MVM procedure provided high recoveries (>88%) for 17 test aroma compounds and moderate recoveries (44-71%) for 4 test compounds. The method showed good linearity (r(2)>0.9913) and high sensitivity (limit of detection: 0.1-0.5 ng mL(-1)) even with MS scan mode. The improved sensitivity of the method was demonstrated with analysis of a wide variety of aroma compounds in brewed green tea. Compared to the original 100 µL MVM procedure, this extension to 1 mL MVM allowed detection of nearly twice the number of aroma compounds, including 18 potent aroma compounds from top-note to base-note (e.g. 2,3-butanedione, coumarin, furaneol, guaiacol, cis-3-hexenol, linalool, maltol, methional, 3-methyl butanal, 2,3,5-trimethyl pyrazine, and vanillin). Sensitivity for 23 compounds improved by a factor of 3.4-15 under 1 mL MVM conditions.

Full evaporation dynamic headspace and gas chromatography-mass spectrometry for uniform enrichment of odor compounds in aqueous samples.

A method for analysis of a wide range of odor compounds in aqueous samples at sub-ng mL⁻¹ to µg mL⁻¹ levels was developed by full evaporation dynamic headspace (FEDHS) and gas chromatography-mass spectrometry (GC-MS). Compared to conventional DHS and headspace solid phase microextraction (HS-SPME), FEDHS provides more uniform enrichment over the entire polarity range for odor compounds in aqueous samples. FEDHS at 80°C using 3 L of purge gas allows complete vaporization of 100 µL of an aqueous sample, and trapping and drying it in an adsorbent packed tube, while providing high recoveries (85-103%) of the 18 model odor compounds (water solubility at 25°C: log0.54-5.65 mg L⁻¹, vapor pressure at 25°C: 0.011-3.2 mm Hg) and leaving most of the low volatile matrix behind. The FEDHS-GC-MS method showed good linearity (r²>0.9909) and high sensitivity (limit of detection: 0.21-5.2 ng mL⁻¹) for the model compounds even with the scan mode in the conventional MS. The feasibility and benefit of the method was demonstrated with analyses of key odor compounds including hydrophilic and less volatile characteristics in beverages (whiskey and green tea). In a single malt whiskey sample, phenolic compounds including vanillin could be determined in the range of 0.92-5.1 µg mL⁻¹ (RSD<7.4%, n=6). For a Japanese green tea sample, 48 compounds including 19 potent odorants were positively identified from only 100 µL of sample. Heat-induced artifact formation for potent odorants was also examined and the proposed method does not affect the additional formation of thermally generated compounds. Eighteen compounds including 12 potent odorants (e.g. coumarin, furaneol, indole, maltol, and pyrazine congeners) were determined in the range of 0.21-110 ng mL⁻¹ (RSD<10%, n=6).

Determination of mechanism of flock sediment formation in tea beverages.

The mechanism of sediment formation during the storage of green tea beverage was investigated. Green tea extract was separated by Diaion HP-20 column chromatography, and a sediment-formation test was performed. Results showed that at least one compound of the substance causing flock sediment was contained in each of the HP-20 nonadsorbed and adsorbed fractions. From the following fractionations and structure analyses, the substance in the HP-20 adsorbed fraction was determined to be 1-O-galloyl-4,6-O-(S)-hexahydroxydiphenoyl-beta-D-glucose (strictinin), which is one of the ellagitannins. Strictinin was hydrolyzed to ellagic acid by heat-sterilization processes such as retort sterilization or the ultra-high temperature processing used during the manufacturing of tea beverages. Ellagic acid combined with proteins in the HP-20 nonadsorbed fraction to form an irreversible sediment of green tea beverage; ellagic acid and proteins were confirmed to be present in that sediment. The HP-20 adsorbed fraction contained little strictinin and formed hardly any sediment, suggesting that control of the strictinin content is significant in avoiding sediment formation during the manufacturing process of tea beverages.