Elemental characterization of Japanese green tea leaves and tea infusion residue by neutron-induced prompt and delayed gamma-ray analysis

2.1 Sample and standard preparation

This pilot study is the first trial of the application of PGA to tea samples. In this study, six brands of commercially available green teas were purchased from local markets for analysis. The tea brands are Ito-en (hoji cha), Ito-en (ryoku cha), Okukuji (Ibaraki), Shizuoka, Uji, and Yame. Ito-en (hoji cha) is a roasted (brown) tea and the rest are normal green teas. Among the six brands, the last four brands are cultivated across the Japanese Islands shown in Fig. 1 (numbering serially 3–6), encompassing the area from the northern tea producing area, Ibaraki, to the southern tea producing area, Fukuoka (Yame) of Japan. Based on the price and proportion of sprigs (small pieces of tea plants with leaves) to tea leaves, the first two brands can be considered as low quality tea and rest of the four brands as high quality tea. Two sets of samples were prepared; one set was used without infusion and another set was used for infusion of the tea leaves. For non-infusion-samples, tea leaves were dried at 50°C overnight and ground into powder. Powder samples were used for PGA and INAA. For PGA, about 200 mg of each powder sample was made into pellets (10 mm Φ) and packed into FEP (fluorinated ethylene propylene) film bags. For INAA, about 50 mg of each sample was packed into double polyethylene film bags and heat-sealed. For preparing the after-infusion-samples, 2 gm of each sample was poured into 200 mL of boiling deionized water. The resulting infusion was stirred with a glass rod for about 30 s to ensure proper wetting, covered and steeped for 5.5 more minutes, yielding a total of 6 min infusion. For the preparation of Japanese green tea, it is generally infused with hot water for around 6 min. Therefore, totally 6 min infusion time was set for this experiment. Leached tea leaves were separated from hot water and evaporated to dryness in an oven and ground into powder for PGA and INAA. Analytical grade chemical reagents oxalic acid di-hydrate, B, NH₄Cl and S were heat-sealed into FEP film bags and used as comparators for the determination of H, B, Cl and S, respectively, for PGA. For preparing B standard, B standard solution was evaporated onto a filter paper and then heat-sealed into FEP film bag. For INAA, the National Institute for Environmental Studies (NIES) certified reference material No-7 (Tea leaves) and the Geological Survey of Japan (GSJ) certified reference material JB-1 were used as standard reference materials.

Close

Figure 1

Production areas of the tea leaves analyzed in this study are shown in map of Japan. Nos. 3, 4, 5 and 6 represent the production areas of the tea samples Okukuji, Shizuoka, Uji and Yame, respectively.

Export to PPT

2.2 Sample irradiation and counting

3.1 PGA of H, B, Cl and S

Hydrogen, B, Cl and S were successfully determined in the tea samples by PGA. Due to the nondestructive nature and easy sample preparation in PGA, these volatile elements can be determined reliably by avoiding contamination and loss during sample preparation, which may occur in destructive analysis. In PGA, because of the Doppler broadening of the 477.6 keV gamma-ray peak of B emitted by the reaction of ¹⁰B(n,αγ)⁷Li, several coexisting elements may interfere on the B peak. The major interfering elements, confirmed by irradiating several possible reagent standards, are Na (472 keV), Si (476 keV), Cl (468 keV), Mn (476 keV), Co (484 keV) and Ni (483 keV). Although all the above mentioned elements show spectral interference to B, only 472 keV peak of Na and 468 keV peak of Cl show serious interference on the B peak as Na and Cl contents are relatively higher compared to other interfering elements in the tea samples. Assuming that the B peak is symmetric with respect to the peak center corresponding to 477.6 keV, and that no overlapping with the Na peak occurs in the high-energy half side of the B peak, the net area of the B peak was calculated by doubling the net count of the half peak at the high-energy side. This method of spectrum analysis avoids the interference of Na and Cl on the B peak at the low-energy half side of the tea samples. The average H content of the tea leaves determined by PGA was 6.67%. Although the matrix H may cause a serious problem of sensitivity enhancements in biological samples based on sample geometry and matrix H contents in the thermal neutron beam PGA (Mackey et al., 1992), such an effect becomes rather small for PGA with the guided neutron beam PGA, especially for the PGA system at JAEA (Yonezawa and Wood, 1995; Islam et al., 2011b). Sulfur content in the samples was determined by using the 840.9 keV prompt gamma-ray peak. Chlorine content in tea samples could be determined by both PGA and INAA. There was good consistency between the Cl data obtained by both PGA and INAA. In this study, Cl content determined by PGA is reported.

3.2 Elemental composition of the tea leaves

The elemental compositions of the tea samples were determined by the comparison method using chemical reagent standards for PGA and a reference sample, NIES-7 for INAA as reference standards. The quality of the data obtained was evaluated by measuring elemental contents of NIES-7 relative to those of chemical reagents and GSJ-JB-1. The analytical results for single measurement and prompt and delayed gamma-ray energies used for elemental determinations are given in Table 1. The deviations of the measured values of this study from certified ones are also given in Table 1. Uncertainties associated with the determined values are due to counting statistics (1σ). Since counting statistics mainly control the total uncertainty in NAA, it was reported with concentration values. However, in addition to this reported uncertainty, there was commonly 3% uncertainty (estimated in our analysis) was also associated with the reported values which included sample and standard preparation, irradiation, positioning in the detector, pulse-pileup losses and peak integration. It is observed that all the determined values in this study are in good agreement with certified or literature values except for Sc (Table 1). The Sc value of NIES-7 is the only indicative value. As Sc is one of such elements having high sensitivity in INAA, we are confident with our lower value than the NIES value.

The elemental compositions of the green tea leaves are presented in Table 2. As shown in this table, a total of 18 elements were determined by both PGA and INAA, among which H, K, Ca, S, Mg, Cl, Mn and Al contents range from 74400 to 304 μg/g. These elements were reported to be present in relatively large amounts in green and black teas of different countries (Gallaher et al., 2006; Natesan and Ranganathan, 1990; Giulian et al., 2007; Salahinejad and Aflaki, 2010). Iron, Na, B, Zn and Rb are present in the range of 58 to 3 μg/g, while La, Co, Cs and Sc contents are below 0.408 μg/g. Although the tea samples analyzed in this study cover wide ranges of tea producing zones of Japan and the quality of leaves, elemental content are surprisingly similar, giving a credit for developing the general discussion in terms of elemental content in tea leaves produced in Japan. Among the six tea samples, Ito-en (hoji cha) and Ito-en (ryoku cha) are relatively low-priced compared with other four samples and show apparently lower contents of K and Zn than those for high quality teas (Table 2). On the contrary, these low- priced teas contain higher contents of the elements such as Ca, Mg and Al (Table 2). Considering that these low quality teas contain a relatively higher proportion of sprigs than that in high quality tea, some content of Ca, Mg and Al in low-priced teas may be contributed by the sprigs of tea plants (Natesan and Ranganathan, 1990).

3.3 Extraction of the elements in hot water infusion

In general, mineral availability in the tea infusion depends on the amount of minerals present in the tea leaves, solubility of the elements and time allowed for the infusion process. In this study, tea infusion residues were also analyzed to evaluate the extraction efficiencies of the elements by infusion. Fig. 2 shows the relative percentages of each element removed from the tea samples for high quality and low quality teas with a 6 min hot water infusion. Average elemental contents of each element in the tea leaves before and after infusion are used for extraction efficiency calculation. Uncertainties associated with extraction efficiencies in Fig. 2 are propagated errors due to counting statistics. Although Sc content was determined in tea samples before infusion, it could not be determined after infusion because of the very low concentration in the infusion residue. As shown in Fig. 2, extraction efficiencies for low quality and high quality teas are similar within the uncertainty except for B and Zn. Among the 16 elements concerned, Cl, Br, K, Rb, Co, Cs and Na were observed to be effectively infused in hot water and especially, Cl, Br and K were mostly carried away with water. Based on the extraction efficiency of the elements in infusion, analyte elements are sometimes classified into three groups (Matsuura et al., 2001). According to such criteria, the studied elements are categorized as follows;

1. Highly extractable elements (>55%):

Close

Figure 2

Extraction efficiencies of the elements for low quality and high quality tea samples with a 6 min hot water infusion process. Propagated uncertainties shown are due to counting statistics.

Export to PPT

1. Moderately extractable elements (20–55%):

1. Poorly extractable elements (<20%):

Although Japanese green teas contain high content of K and high extraction efficiency in infusion, this study suggests that seven cups (250 mL each cup) of infusion still would be required to provide 10% of the Daily Value (2000 mg/day) (FDA, 1999). It should be noted that deionized water was used in this study. Since tea infusion is generally not prepared by deionized water, sources of water have the potential to increase the mineral contents of the tea infusion. The contents of Na and K in tap water may be of interest, particularly if a person is drinking multiples cups of tea per day. Considering the nutritional importance, special attention is given to Mn and Mg availability in tea infusion. Previous study (Natesan and Ranganathan, 1990) reported that extraction efficiencies of Mg and Mn in green tea infusion are around 30%, in this study, however, their extraction rates are below 20% as shown in Fig. 2. The amount of an element extracted into the tea infusion depends principally on how strongly the element is bound to the matrix and whether it is soluble in the solution employed (Costa et al., 2002). Future research on the extraction efficiency may extend to examine the dependence of the solubility of minerals on pH and extraction time to better explain the wide variability in extraction efficiencies seen in this study. Aluminum content in tea infusion is of great concern for renal failure patients and for its relation to Alzheimer’s disease (McLachlan, 1995). Although Japanese green tea, especially low quality tea, contains relatively high contents of Al, no special attention needs to be paid in consideration of its poor extraction efficiency in infusion.