Translate this page into:
Direct ICP-OES multielement analysis of infused black and green teas and chemical fractionation of selected essential and non-essential elements prior to evaluation of their bioavailability and classification of teas by pattern recognition
⁎Corresponding author. pawel.pohl@pwr.edu.pl (Pawel Pohl)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Peer review under responsibility of King Saud University.
Abstract
Operationally defined chemical fractionation of Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn in infusions of loose leaf and bagged black and green teas was carried out. It was based on tandem-column solid phase extraction using reverse-phase and strong cation-exchange extraction tubes connected in series. Concentrations of elements in resultant infusions as well as column effluents and eluates were determined with the aid of inductively coupled plasma optical emission spectrometry and a newly developed simplified sample preparation procedure. It was established that information on separated chemical fractions, particularly the fraction of residual species of elements, had better discrimination power than information about total concentrations of these elements. Differences in chemical fractionation patterns of elements in infusions of analyzed four types of tea led to simple classification of different black and green teas by principal component analysis and linear discriminant analysis.
Keywords
Infusions
Black tea
Green tea
Elements
Solid phase extraction
Chemical fractionation
1 Introduction
Next to water, black and green teas (BTs and GTs) are the most widely consumed non-alcoholic beverages worldwide. When made teas are brewed in hot water, elements included in leaf material and bound to its matrix components are differentially extracted into infusions, making resultant beverages a reliable source of essential major, minor and trace elements, i.e., Ca, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, Zn, in human diet (Karak et al., 2017a). The inorganic composition of infusions of BTs and GTs may not include only essential elements but also those that are non-essential and toxic like As, Cd and Pb (Milani et al., 2016). Considering habitual drinking of tea, it is accepted that most of elements present in infusions of BTs and GTs have certain nutritious effects and health implications, but excessive exposure to non-essential and toxic trace elements can impose hazard on well-being (Karak et al., 2017a). Therefore, it is not surprising that tea infusions and their element composition are of great concern to consumers and producers of tea; they are also an interesting subject of research for food analysts and chemists. Indeed, element analysis of tea infusions, carried out using common atomic and mass spectrometry methods, including flame atomic absorption spectrometry (FAAS), graphite furnace atomic absorption spectrometry (GFAAS), inductively coupled plasma optical emission spectrometry (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS), is an important part of quality and safety control of this beverage. As it is demonstrated in Table 1S, including information on element analysis of infusions of BTs and GTs reported in relevant literature from 2010 to 2017, tea beverages are usually directly analyzed without any prior treatment (Dambiec et al., 2013; Jeszka-Skowron et al., 2015; Karak et al., 2017b; Konieczynski et al., 2017; Li et al., 2015; Ozdemir et al., 2014; Pekal et al., 2013; Polechonska et al., 2015) or after simplified sample preparation that is typically limited to acidification with HNO3 (Gomez-Nieto et al., 2017; Milani et al., 2015, 2016; Paz-Rodriguez et al., 2015; Salahinejad and Aflaki, 2010; Szymczycha-Madeja et al., 2016). More protracted sample preparation procedures, involving decomposition of the matrix of tea infusions at high temperatures, are also executed (Brzezicka-Cirocka et al., 2016, 2017; Schulzki et al., 2017).
Results of element analysis of infusions of BTs and GTs carried out with methods described above are commonly considered to judge the nutrition value and intoxication threat related to customary drinking of one to several cups of tea per day and intake of essential and non-essential elements. In this case, percentages of elements leached from made teas into infusions are evaluated (Brzezicka-Cirocka et al., 2016, 2017; Dambiec et al., 2013; Jeszka-Skowron et al., 2015; Li et al., 2015; Milani et al., 2016; Pekal et al., 2013; Pekal and Pyrzynska, 2013; Polechonska et al., 2015; Salahinejad and Aflaki, 2010; Schulzki et al., 2017; Szymczycha-Madeja et al., 2016) in addition to assessment of coverages of respective recommended daily intakes (RDIs) by dietary intake of tea (Brzezicka-Cirocka et al., 2016, 2017; Dambiec et al., 2013; Karak et al., 2017b; Milani et al., 2016; Pekal and Pyrzynska, 2013; Polechonska et al., 2015; Salahinejad and Aflaki, 2010). Hazardous quotients of non-essential and acutely toxic elements are also evaluated to verify whether or not regular consumption of tea infusions results in exceeding acceptable daily intakes (ADIs) and causes risk of poisoning via oral exposure with such elements like As, Al, Cd, Co, Cr, Cu, Hg, Ni and Pb (Brzezicka-Cirocka et al., 2016, 2017; Karak et al., 2017b; Li et al., 2015; Milani et al., 2016; Ozdemir et al., 2014; Pekal and Pyrzynska, 2013; Polechonska et al., 2015; Salahinejad and Aflaki, 2010; Schulzki et al., 2017).
In both mentioned cases, it is always assumed that elements released from made BTs and GTs during brewing and present in resultant infusions are 100% bioavailable and bioaccesible. This assumption is however unjustified and could not be considered to assess actual uptakes of elements from tea beverages. Bioavailability of elements from infusions of BTs and GTs and their toxicity (in case of hazardous and toxic elements) to organisms depend not only on their total concentrations but primarily on existing physicochemical forms and their contributions to total contents (Pekal et al., 2013). Accordingly, essential and non-essential elements can exist as simple ionic forms or bound forms with polyphenolic species and other macromolecular compounds (Pekal et al., 2013; Pekal and Pyrzynska, 2013). Therefore, due to complex organic composition, tea infusions can contain various complexes of elements with low, moderate and high molecular weight (LMW, MMW and HMW) biologands present in the tea matrix, i.e., water soluble polyphenols like catechins, theaflavanis and thearubigens, that willingly bind ions of elements through hydroxyl, carboxylate and phenolate O-donor groups (Cairns et al., 1996; Matsura et al., 2001). Presence of elements species of different charge, hydrophobicity, stability and molecular size in infusions of BTs and GTs affect bioavailability of these elements and their real uptake by the body (Cairns et al., 1996; Matsura et al., 2001; Pekal et al., 2013). Therefore, speciation analysis of essential and non-essential elements in tea infusions can help in determining their bioavaliablity from this beverage. Unfortunately, knowledge about type and distribution of speciation forms of elements in infusions of BTs and GTs, although critically important to predict their bioavailability and eventual bioaccessibility in the gastrointestinal digestion track, is not firmly established in literature. In overwhelming majority cases it is limited to fractionation analysis of Al by means of solid phase extraction (SPE) with chelating, ion exchange and adsorbing resins (Chen et al., 2003; Erdemoglu et al., 2000; Ruszczynska et al., 2004), size exclusion liquid chromatography (SE-LC) (Kralj et al., 2005), anion-exchange fast protein liquid chromatography (AE-FPLC) (Kralj et al., 2005), cation-exchange liquid chromatography (CE-LC) (Street et al., 2007) or reverse-phase liquid chromatography (RP-LC) (Cairns et al., 1996). In addition, single elements have been occasionally speciated using the SPE approach, e.g., Al, Cu, Fe, Mn, Ni and Zn with the aid of adsorbing resins (Mucchino and Musci, 2014), or Mn and Zn (Pohl and Prusisz, 2007a), Fe and Zn (Pohl and Prusisz, 2007b) and Ca and Mg (Pohl and Prusisz, 2006) with the aid of ion-exchange and adsorbing resins, or the LC approach, e.g., Ni by AE-LC (Scancar et al., 2013).
Total concentrations of elements determined in infusions of BTs and GTs can also be applied for classification and pattern recognition purposes. In this case, principal content analysis (PCA) is used to interpret the data on element analysis and classify different tea samples on the basis of total concentrations of selected elements measured in tea infusions. Unfortunately, it is difficult to achieve clear separation between compared samples of BTs and GTs even when total concentrations of a relatively high number of elements are used to build data matrices, i.e., 10 (Milani et al., 2016), 8 (Paz-Rodriguez et al., 2015), 13 (Szymczycha-Madeja et al., 2016). Clear separation between BTs and GTs can usually be achieved with PCA when information about the total content of elements is accompanied by information about the content of other components, i.e., selected phenolic compounds, flavonoids and antioxidant activity (Konieczynski et al., 2017), or when the number of measured elements is larger (Diniz et al., 2015). Unfortunately, collection of such data needs additional equipment, methods and time required for analysis.
Bearing in mind all mentioned difficulties in assessing bioavailability of elements from infusions of BTs and GTs as well as simply and univocally classifying and discriminating made teas on the basis of total concentrations of elements determined in their infusions, two research hypotheses were picked up and studied in the present work. The first one assumed that infusions of BTs and GTs significantly differ in terms of type and contribution of certain classes of speciation forms of elements, which likely result from inherent differences in processes involved in production of both tea varieties. The second research hypothesis assumed that determination of concentrations of distinct speciation forms of elements in infusions of BTs and GTs would enable to indirectly estimate bioavailability of elements from these two beverages and find the difference between them. This would help in univocally differentiating teas due to their variety by using PCA and other pattern recognition methods. Since direct speciation analysis of elements in tea infusions seemed to be difficult, it was decided that a better approach to provide useful information on classes of speciation forms of elements in infusions of BTs and GTs would be their fractionation and partitioning due to hydrophobicity and charge carried out by tandem-column SPE. This methodology was presumed to give an operationally defined and reproducible measure of chemical fractions of elements in infusions of analyzed BTs and GTs in a quite accessible and fast way. Very recently, a similar approach has been found very useful in bringing important information about fractionation forms of Cu and Fe in wines and their role in spoilage and aging processes (Kontoudakis et al., 2017).
Therefore, a tandem-column SPE procedure, in which non-polar reverse-phase Discovery DSC-18 and strong cation-exchange Discovery DSC-SCX SPE tubes connected in series, with subsequent determination of concentrations of Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn in separated column effluents and eluates by ICP-OES was applied to evaluate unique chemical fractionation patterns of these elements in infusions of studied BTs and GTs. Contributions of operationally defined chemical fractions, i.e., hydrophobic (HF), cationic (CF) and residual (RF), of Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn and their concentrations in these fractions were used to rationally predict and compare bioavailability of elements from infusions of both tea varieties, i.e., BTs and GTs, in addition to unambiguous differentiation and categorization of analyzed samples, divided into four groups due to their type, i.e., leaf and bagged black and green teas (LBTs, BBTs, LGTs and BGTs), by PCA and linear discriminant analysis (LDA).
2 Experimental
2.1 Apparatus
An Agilent bench-top optical emission spectrometer of inductively coupled Ar plasma (Ar-ICP-OES), model 720, was used to measure concentrations of studied essential (Ca, Cu, Fe, Mg, Mn, Zn) and non-essential (Al, Ba, Ni, Sr) elements in all samples, including infusions of BTs and GTs, as well as effluents and eluates resulting from tandem-column SPE chemical fractionation. The spectrometer was equipped with a high-resolution Echelle-type polychromator and a VistaChip II solid-state charge-transfer device detector for acquisition of radiation emitted by Ar-ICP. For sample introduction, an Agilent OneNeb concentric nebulizer mounted into a single-pass glass cyclonic spray chamber was used. Working parameters used to operate the spectrometer were as follows: forward power of 1.2 kW, plasma gas of 15.0 L min−1, auxiliary gas of 1.5 L min−1, nebulizing gas of 0.75 L min−1, sample/standard solution uptake of 0.75 mL min−1 and waste drainage of 1.5 mL min−1. In each measurement cycle, instrument stabilization and solution uptake delay were 15 and 30 s, respectively. Intensities of analytical lines selected for measurements were read 3 times within integration time of 1 s. The following analytical lines (I and II denote atomic and ionic emission lines, respectively), free from spectral interferences and with high signal to noise ratios, were selected: Al I 396.2 nm, Ba II 455.4 nm, Ca I 422.7 nm, Cu I 324.8 nm, Fe I 248.3 nm, Mg I 285.2 nm, Mn II 257.6 nm, Ni II 231.6 nm, Sr II 407.8 nm and Zn I 213.8 nm. Concentrations of elements were quantified using seven-point external calibration curves within the concentration range of 0.01–10 mg L−1.
2.2 Reagents and materials
Emsure® ACS premium grade solutions of concentrated HNO3 (65.0%) and HCl (37.0–38.0%) were purchased from Merck Millipore (Poland). A Merck Millipore CertiPur® multi-element stock (1000 mg L−1) ICP standard solution IV was used to prepare standard solutions for calibration. De-ionized water was used throughout. For tandem-column SPE, 6-mL Discovery DSC-18 and Discovery DSC-SCX propylene extraction tubes, filled with 500 mg of silica gel based material with polymerically bonded octadecyl and benzene sulfonic acid functional groups, respectively, were applied to retain elements species from solutions via hydrophobic interactions and cation-exchange.
2.3 Preparation of tea infusions
Commercially available and the most representative in Poland and other European Union countries BTs and GTs were selected for this study i.e., five loose leaf and five bagged black teas (LBTs, BBTs) as well as five loose leaf and five bagged green teas (LGTs, BGTs). Infusions of BTs and GTs were prepared following recommendations of tea producers. To be able to compare results of element analysis of studied BTs and GTs with those reported in related literature, de-ionized water was used for their brewing. In case of BTs, 100 mL of water was boiled in a clean laboratory electric cattle with a temperature controller and applied to brew 1.0 g-portions of LBTs and BBTs (in the latter case, the content of infusions bags was used) for 3 min. Infusions were prepared in 250-mL glass beakers covered with watch glasses. Afterwards, they were separated from grounds by decantation through pre-cleaned 390-grade quantitative filter papers (Munktell & Filtrak, Germany). Infusions of GTs were prepared in the same way as for BTs with the only difference being that water used for brewing was hot (85 °C).
2.4 Chemical fractionation by tandem-column solid phase extraction
Before use, Discovery DSC-18 and DSC-SCX SPE tubes were appropriately pre-conditioned. Beds of DSC-18 SPE tubes were rinsed at first with 5 mL of methanol and then washed with 20 mL of de-ionized water. Beds of DSC-SCX SPE tubes were rinsed with 10 mL of a 2 mol L−1 HCl solution and then with 25 mL of de-ionized water. In both cases, pre-conditioning solvents were passed through SPE tubes at 1.0 mL min−1 using a peristaltic pump.
According to previous research (Kontoudakis et al., 2017; Pohl and Prusisz, 2007a), the order of SPE tubes connected in series, i.e., DSC-18 followed by DSC-SCX, enabled to retain hydrophobic HMW and MMW species of elements on the 1st column and simple ions of elements along with their stable cationic complexes with LMW species and labile complexes on the 2nd column. It was experimentally verified (by passing respective solutions of selected elements and their mixtures with citric acid, pH 4.5, 5.0 and 5.5 in reference to infusions of BTs and pH 6.5, 7.0 and 7.5 in reference to infusions of GTs) that DSC-18 SPE tubes did not retain any simple ions of elements and their complexes with LMW ligands.
To partition elements species into three operationally defined chemical fractions, i.e., HF, CF and RF, 20 mL of prepared infusions of BTs and GTs were passed at first through DSC-18 SPE tubes. 10-mL portions of resultant effluents were saved for element analysis by ICP-OES to assess the content of elements species not retained by these SPE tubes. Remaining 10-mL portions of these effluents were immediately passed through DSC-SCX SPE tubes. Effluents of the latter SPE tubes were also saved prior to ICP-OES analysis to assess the content of elements species not retained by both SPE tubes. Finally, elements species retained by DSC-SCX tubes were eluted using 5.0 mL of a 2.0 mol L−1 HCl solution followed by rinsing with 5.0 mL of de-ionized water. Resulting eluates were saved for element analysis by ICP-OES on the content of retained cationic species of elements.
Total concentrations of studied elements in infusions of BTs and GTs (CT) as well as effluents of both SPE tubes (CEf, I for DSC-18 and CEf, II for DSC-SCX) and eluates (CEl, II) for DSC-SCX SPE tubes enabled to calculate percentage contributions of their separated chemical fractions in the following way: HF (%) = 100% × (CT–CEf, I)/CT; CF (%) = 100% × 2 × CEl, II/CT; and RF (%) = 100% × CEf, II/CT.
2.5 Sample preparation before element analysis
Total concentrations of elements in prepared infusions of BTs and GTs as well as effluents and eluates obtained after treatment of these infusions with the tandem-column SPE procedure were determined using a simplified method of analysis, developed in this study, that was solely limited to acidification of respective sample solutions (except for eluates) with concentrated HNO3 to reach its final concentration of 0.50 mol L−1. Reliability (in reference to accuracy of determined concentrations of elements) of results achieved with this simplified sample preparation procedure followed by ICP-OES analysis of resultant sample solutions was assessed in two ways. Closed-vessel, microwave-assisted wet digestion in concentrated HNO3 (including initial evaporation of 10 mL of infusions to ∼1 mL, digestion of resultant aliquots in 2 mL of HNO3, dilution of residues to 20 mL with de-ionized water) was used as a reference sample preparation procedure prior to element analysis of resultant sample solutions by ICP-OES versus external calibration with standard solutions acidified with HNO3 to ∼1.4 mol L−1. Two teas were selected, i.e., LBT1 and LGT1, and used to verify accuracy of total content analysis of infusions of BTs and GTs. In addition, systematic spike-and-recovery experiments were carried out in case of mentioned infusions of LBT1 and LGT2 as well as effluents and eluates achieved for these infusions after their initial treatment with the tandem-column SPE procedure. In case of ICP-OES measurements of studied elements, infusions and effluents of SPE tubes were undiluted and acidified with HNO3 to 0.50 mol L−1; standard solutions used for external calibration contained HNO3 at the same concentration. In case of eluates containing ∼1 mol L−1 HCl, they were also measured undiluted; hence, standard solutions used for external calibration were acidified with HCl at 1.0 mol L−1.
3 Results and discussion
3.1 Validity of results obtained with developed total content analysis method
Total concentrations of Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn in infusions of analyzed BTs and GTs are given in Tables 2S and 3S, respectively. Considering values of relative standard deviation (RSD) for measurements of three identically prepared samples (n = 3), precision of results of total content analysis of tea infusions obtained with the simplified sample preparation procedure proposed here was very good (see Table 1 with ranges of RSDs evaluated for total concentrations of elements in infusions of all analyzed samples of both tea varieties). In overwhelming cases, precision was better than 7%. In few cases, higher RSDs (8–13%) were noted for low concentrated elements, i.e., Ba in GTs, Cu in BTs, Fe in BTs and GTs, Ni in BTs and GTs.
| RSD (n = 3), % | ||||
|---|---|---|---|---|
| Total content analysis | Chemical fractionation analysis | |||
| BTs | GTs | BTs | GTs | |
| Al | 0.2–2.8 | 0.2–1.8 | 0.1–5.1 | 0.1–7.7 |
| Ba | 0.4–3.1 | 0.4–9.7 | 0.1–6.1 | 0.2–5.8 |
| Ca | 0.6–4.6 | 0.5–1.6 | 0.1–5.6 | 0.2–3.5 |
| Cu | 0.2–10.8 | 0.4–2.1 | 0.1–8.4 | 0.1–3.5 |
| Fe | 0.9–12.9 | 0.4–10.2 | 0.1–6.2 | 0.2–9.6 |
| Mg | 0.4–2.3 | 0.4–3.0 | 0.1–3.8 | 0.1–3.3 |
| Mn | 0.5–3.6 | 0.5–2.5 | 0.1–4.1 | 0.1–4.7 |
| Ni | 0.4–8.0 | 0.6–12.9 | 0.2–5.1 | 0.1–7.5 |
| Sr | 0.3–2.5 | 0.3–1.7 | 0.1–5.0 | 0.1–4.5 |
| Zn | 0.6–3.0 | 0.3–7.4 | 0.1–5.9 | 0.1–6.8 |
Results for all samples within a given group of tea variety.
In a similar way, precision was evaluated for measurements of concentrations of studied elements in chemical fractions separated using the tandem-column SPE procedure. RSD values of determined concentrations of elements were mostly within the range of 0.1–7%, pointing very good precision of these measurements (see Table 1 with ranges of RSDs evaluated for concentrations of elements in chemical fractions separated in infusions of all analyzed samples of both tea varieties). In some cases, higher RSDs were obtained, particularly in case of very low concentrations of elements present in separated chemical fractions, i.e., 8–10% for Cu in BTs, and Al, Fe and Ni in GTs.
Accuracy of determination of total concentrations of elements in infusions of BTs and GTs was verified in two different ways. By comparison of results of element analysis of infusions of LBT1 and LGT1 achieved with the aid of their prior close-vessel wet acid digestion (the reference sample preparation procedure) and acidification (the proposed simplified sample preparation procedure), it was established that there was no statistically significant differences between concentrations of elements measured in samples solutions prepared by both procedures. The t-Student test with Cochran-Cox adjustment (p = 0.05, n = 6) was used for that purpose due to differences in variance of results achieved for both compared procedures (Jackson, 2017). In addition, spike-and-recovery experiments were carried out. The simplified sample preparation procedure was used to prepare unspiked and spiked infusions of LBT1 and LGT1 prior to ICP-OES measurements. After analysis (see Table 2), it was found that recoveries of elements added to these infusions were within 96–103% (LBT1) and 97–106% (LGT1).
| Recoveries (n = 3), % | Sum of contributions of chemical fractions, % | |||||
|---|---|---|---|---|---|---|
| Total content analysis | Chemical fractionation analysis | |||||
| LBT1 | LGT1 | LBT1 | LGT1 | BTs | GTs | |
| Al | 96.3 ± 2.1 | 97.8 ± 3.3 | 99.8–102.6 | 103.1–105.5 | 100.2–106.6 | 99.0–107.7 |
| Ba | 103.1 ± 2.8 | 99.2 ± 1.0 | 97.6–99.6 | 101.3–102.6 | 95.3–105.2 | 97.9–106.9 |
| Ca | 99.0 ± 1.2 | 98.3 ± 1.8 | 104.4–106.2 | 102.8–106.0 | 92.7–106.7 | 98.2–112.6 |
| Cu | 99.2 ± 2.3 | 100.1 ± 0.7 | 94.1–96.3 | 95.5–98.1 | 98.1–106.7 | 99.0–106.0 |
| Fe | 98.4 ± 2.2 | 106.0 ± 2.6 | 102.7–106.3 | 97.8–99.4 | 95.3–110.1 | 93.7–109.8 |
| Mg | 102.2 ± 2.5 | 101.9 ± 2.7 | 96.2–98.2 | 98.2–99.4 | 95.6–102.0 | 97.9–105.9 |
| Mn | 100.7 ± 0.9 | 98.0 ± 2.2 | 97.4–99.2 | 100.3–102.3 | 99.5–106.5 | 97.3–107.6 |
| Ni | 102.6 ± 3.0 | 97.1 ± 1.8 | 94.6–97.4 | 99.5–101.4 | 97.5–107.5 | 95.2–106.9 |
| Sr | 98.9 ± 1.7 | 101.7 ± 2.5 | 101.4–103.4 | 95.0–98.6 | 94.5–105.4 | 97.4–108.6 |
| Zn | 99.1 ± 1.1 | 98.8 ± 3.1 | 101.7–104.9 | 103.3–107.1 | 96.4–104.3 | 96.4–102.9 |
In case of accuracy of measurements of elements in chemical fractions separated with the aid of the tandem-column SPE procedure, recoveries of elements added to effluents (for DSC-18 and DSC-SCX SPE tubes) and eluates (for DSC-SCX SPE tubes) distinguished for infusions of LBT1 and LGT1 were also quantitative and within the following ranges: 95–106% for LBT1 and 95–107% for LGT1. In addition, sums of percentage contributions of all chemical fractions separated in infusions of given BTs and GTs (see contributions of the HF, the CF and the RF for infusions of all BTs and GTs given in Tables 2S and 3S, respectively) were considered. Considering all samples within each variety, i.e., BTs and GTs, it was established that sums of percentage contributions of the HF, the CF and the RF were quantitative and within 93–110% (for BTs) and 94–113% (for GTs).
Both assessed figures of merit indicated validity of results obtained with the developed simplified sample preparation procedure prior to determination of concentrations of elements in infusions of BTs and GTs as well as effluents and eluates resulted from their treatment by the tandem-column SPE procedure followed by ICP-OES detection. The proposed sample preparation procedure, including only acidification of infused BTs and GTs with HNO3, certainly shortened analysis time and minimized any risk related to contamination of samples or loss of elements due to vigorousness of the digestion process and associated with it high temperature treatment. It is worth mentioning that although simplified sample preparation procedures of tea infusions prior to their spectrometric element analysis were reported so far in related literature, they have unfortunately been rarely validated (Dambiec et al., 2013; Jeszka-Skowron et al., 2015; Karak et al., 2017b; Konieczynski et al., 2017; Li et al., 2015; Memic et al., 2014; Milani et al., 2016; Ozdemir et al., 2014; Pekal et al., 2013; Pekal and Pyrzynska, 2013; Polechonska et al., 2015; Salahinejad and Aflaki, 2010). Just in few cases, spike-and-recovery studies were made to directly show if proposed sample preparation and detection brought reliable results (Diniz et al., 2015; Ghoochani et al., 2015; Gomez-Nieto et al., 2017; Paz-Rodriguez et al., 2015; Shekoohiyan et al., 2012; Szymczycha-Madeja et al., 2016). Unfortunately, comparison of results obtained using a simplified sample preparation procedure with those obtained with a reference procedure, e.g., microwave-assisted wet digestion in concentrated oxidizing reagents aimed at destructing the organic matrix of tea infusions and releasing elements from their bound forms, has been very seldom reported in related literature (Milani et al., 2015; Szymczycha-Madeja et al., 2016), although it is more appropriate than evaluation of recoveries of elements from spiked tea infusions.
Here, both comparison of results obtained with the simplified sample preparation procedure with those obtained with the reference sample preparation procedure (microwave-assisted wet acid digestion) in addition to the spike-and-recovery study were carried out. Their outcomes indicated high metrological quality of results of total content analysis with the simplified sample preparation procedure and chemical fractionation analysis with the tandem-column SPE procedure proposed in the present work.
3.2 Chemical fractionation of elements in infusions of black and green teas
Average total concentrations of Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn determined for a given (out of four) group of tea infusions, i.e., LBTs, BBTs, LGTs and BGTs, are gathered in Table 3. There are also included average percentage shares of chemical fractions of elements species separated with the aid of the tandem-column SPE procedure, i.e., the HF, the CF and the RF. In addition, respective standard deviations (SDs) for these both average measures are given, directly pointing magnitude of variance in each group of studied teas. To the best of our knowledge, these results are a complete set of profiles of chemical fractions of essential (Ca, Cu, Fe, Mg, Mn, Ni and Zn) and non-essential (Al, Ba, Sr) elements in infusions of BTs and GTs that has never before been reported in related literature. A closer look at this data revealed that established chemical fractionation patterns of elements in BTs and GTs, i.e., percentage contributions of given chemical fractions and their mutual proportions, differed to a great extend. Differences observed for studied elements between both tea varieties were likely present due to dissimilar processing and composition of the organic matrix of BTs and GTs and binding capacity of elements by different bioligands present in both tea varieties. It was presumed that these differences could be useful for the purpose of elucidation of bioavailability and bioaccessibility of elements from tea infusions as well as classification and grouping of teas by chemometric methods.
| LBTs | BBTs | LGTs | BGTs | Fa | p | Differencesb | |
|---|---|---|---|---|---|---|---|
| Total concentration (CT), μg mL−1 | |||||||
| Al | 1.78 ± 0.70 | 4.61 ± 0.78 | 1.74 ± 0.48 | 4.79 ± 1.98 | 17.669 | 0.001 | LBT-BBT, LBT-BGT, BBT-LGT, LGT-BGT |
| Ba | 0.0641 ± 0.0961 | 0.0416 ± 0.0276 | 0.0239 ± 0.0054 | 0.0373 ± 0.0090 | 2.910 | 0.103 | |
| Ca | 5.80 ± 0.94 | 5.46 ± 3.42 | 6.28 ± 3.60 | 4.38 ± 0.58 | 2.659 | 0.121 | |
| Cu | 0.0611 ± 0.0153 | 0.0674 ± 0.0111 | 0.0492 ± 0.0190 | 0.0600 ± 0.0064 | 1.068 | 0.414 | |
| Fe | 0.0106 ± 0.0043 | 0.0194 ± 0.0046 | 0.0326 ± 0.0203 | 0.0289 ± 0.0069 | 9.052 | 0.005 | LBT-LGT |
| Mg | 4.44 ± 1.14 | 6.93 ± 1.00 | 5.18 ± 1.52 | 5.65 ± 0.85 | 4.121 | 0.044 | LBT-BBT |
| Mn | 0.583 ± 0.256 | 1.71 ± 0.58 | 1.14 ± 0.59 | 2.19 ± 0.54 | 13.008 | 0.002 | LBT-BBT, LBT-BGT, LGT-BGT |
| Ni | 0.0195 ± 0.0038 | 0.0244 ± 0.0074 | 0.0362 ± 0.0083 | 0.0515 ± 0.0108 | 14.072 | 0.001 | LBT-LGT, LBT-BGT, BBT-BGT, LGT-BGT |
| Sr | 0.0284 ± 0.0044 | 0.0320 ± 0.0223 | 0.0372 ± 0.0085 | 0.0299 ± 0.0039 | 1.196 | 0.370 | |
| Zn | 0.0813 ± 0.0123 | 0.0872 ± 0.0057 | 0.0803 ± 0.0192 | 0.104 ± 0.016 | 2.097 | 0.180 | |
| Contribution of the hydrophobic fraction (HF), % | |||||||
| Al | 6.06 ± 1.15 | 5.28 ± 2.97 | 17.96 ± 7.35 | 19.02 ± 3.24 | 23.818 | 0.000 | LBT-LGT, LBT-BGT, BBT-LGT, BBT-BGT |
| Ba | 26.96 ± 2.43 | 19.56 ± 4.45 | 10.52 ± 2.65 | 14.62 ± 4.13 | 7.946 | 0.007 | LBT-LGT, LBT-BGT |
| Ca | 18.76 ± 11.02 | 23.92 ± 5.14 | 15.94 ± 2.14 | 18.00 ± 2.19 | 3.092 | 0.087 | |
| Cu | 26.66 ± 9.35 | 21.92 ± 8.53 | 8.60 ± 3.78 | 13.34 ± 3.32 | 6.598 | 0.014 | LBT-LGT, LBT-BGT, BBT-LGT |
| Fe | 19.04 ± 6.25 | 11.54 ± 4.62 | 51.86 ± 10.03 | 55.92 ± 1.23 | 160.855 | 0.000 | LBT-LGT, LBT-BGT, BBT-LBT, BBT-BBT |
| Mg | 7.58 ± 2.14 | 4.62 ± 1.71 | 7.46 ± 4.67 | 15.78 ± 2.00 | 26.736 | 0.000 | LBT-BGT, BBT-BGT, LGT-BGT |
| Mn | 18.18 ± 1.33 | 15.66 ± 4.53 | 11.34 ± 3.19 | 14.78 ± 0.68 | 9.644 | 0.006 | LBT-LGT |
| Ni | 6.96 ± 1.48 | 10.06 ± 3.61 | 1.36 ± 0.30 | 5.56 ± 2.57 | 30.156 | 0.000 | LBT-LGT, BBT-LGT, BBT-BGT |
| Sr | 18.26 ± 7.24 | 13.86 ± 2.94 | 15.22 ± 2.39 | 18.20 ± 2.40 | 2.268 | 0.152 | |
| Zn | 13.62 ± 1.85 | 13.36 ± 4.89 | 6.28 ± 1.81 | 6.90 ± 2.05 | 1.973 | 0.191 | BBT-LGT |
| Contribution of the cationic fraction (CF), % | |||||||
| Al | 46.90 ± 3.06 | 55.38 ± 4.60 | 38.54 ± 1.11 | 47.34 ± 4.68 | 28.424 | 0.000 | LBT-BBT, LBT-LGT, BBT-LGT, BBT-BGT, LGT-BGT |
| Ba | 72.34 ± 5.06 | 77.76 ± 6.15 | 88.86 ± 3.30 | 85.16 ± 2.72 | 12.544 | 0.002 | LBT-LGT, LBT-BGT, BBT-LGT |
| Ca | 76.58 ± 14.12 | 70.30 ± 8.27 | 81.18 ± 5.59 | 86.10 ± 1.72 | 6.179 | 0.021 | BBT-BGT |
| Cu | 72.28 ± 10.08 | 75.02 ± 7.82 | 89.34 ± 4.90 | 84.90 ± 2.44 | 5.614 | 0.023 | LBT-LGT, LBT-BGT, BBT-LGT |
| Fe | 68.66 ± 7.89 | 65.24 ± 1.65 | 7.96 ± 2.76 | 10.16 ± 4.22 | 581.403 | 0.000 | LBT-LGT, LBT-BGT, BBT-LGT, BBT-BGT |
| Mg | 86.38 ± 2.08 | 91.70 ± 3.53 | 93.54 ± 1.54 | 86.94 ± 3.47 | 12.908 | 0.002 | LBT-BBT, LBT-LGT, LGT-BGT |
| Mn | 83.50 ± 2.20 | 86.56 ± 4.38 | 89.58 ± 5.31 | 88.44 ± 2.69 | 3.729 | 0.057 | |
| Ni | 71.76 ± 1.87 | 74.76 ± 5.04 | 95.80 ± 1.83 | 95.44 ± 2.65 | 150.380 | 0.000 | LBT-LGT, LBT-BGT, BBT-LGT, BBT-BGT |
| Sr | 81.92 ± 6.46 | 83.52 ± 2.88 | 79.32 ± 5.69 | 85.66 ± 2.98 | 1.591 | 0.262 | |
| Zn | 85.88 ± 3.86 | 82.36 ± 5.26 | 94.02 ± 1.52 | 90.88 ± 5.07 | 0.164 | 0.918 | |
| Contribution of the residual fraction (RF), % | |||||||
| Al | 50.84 ± 3.84 | 43.00 ± 3.88 | 49.30 ± 6.09 | 37.76 ± 4.16 | 8.821 | 0.005 | LBT-BGT, LGT-BGT |
| Ba | 1.88 ± 2.69 | 1.64 ± 0.59 | 4.24 ± 1.86 | 4.02 ± 1.40 | 5.378 | 0.027 | |
| Ca | 2.20 ± 1.71 | 3.70 ± 3.04 | 5.24 ± 2.55 | 2.32 ± 0.56 | 2.031 | 0.194 | |
| Cu | 2.38 ± 2.02 | 4.06 ± 2.71 | 3.62 ± 2.78 | 4.18 ± 0.87 | 0.975 | 0.452 | |
| Fe | 17.40 ± 2.99 | 23.54 ± 3.33 | 41.68 ± 9.73 | 34.74 ± 6.88 | 14.468 | 0.001 | LBT-LGT, LBT-BGT, BBT-LGT |
| Mg | 4.12 ± 2.08 | 2.34 ± 0.78 | 0.62 ± 0.32 | 0.20 ± 0.00 | –c | ||
| Mn | 0.12 ± 0.04 | 0.12 ± 0.04 | 0.14 ± 0.09 | 0.10 ± 0.00 | –c | ||
| Ni | 26.12 ± 2.67 | 16.24 ± 4.99 | 1.56 ± 0.84 | 1.30 ± 0.42 | 131.368 | 0.000 | LBT-BBT, LBT-LGT, LBT-BGT, BBT-LGT, BBT-BGT |
| Sr | 0.52 ± 0.40 | 1.20 ± 1.28 | 7.18 ± 3.21 | 2.16 ± 0.62 | 12.364 | 0.002 | LBT-LGT, BBT-LGT, LGT-BGT |
| Zn | 2.08 ± 1.15 | 4.04 ± 2.58 | 1.28 ± 0.41 | 2.88 ± 1.90 | –c | ||
At the outset, both tea varieties, i.e., BTs and GTs, and four groups of studied made teas, i.e., LBTs, BBTs, LGTs and BGTs, were compared in reference to total concentrations of elements determined in their infusions and contributions of chemical fractions of these elements separated by SPE. Two-side one-way analysis of variance (ANOVA) was used for that, enabling to assess significant differences between within-group variance and between-group variance. Due to violation of assumption about homogeneity of variance in data of compared groups of teas, F-values were calculated using the Welch test (Jackson, 2017). All calculated F-values along with test significance measures (p values) are given in Table 4. In addition, the single-step multiple range test, i.e., the post hoc Tukey honest significant difference (HSD) test (Jackson, 2017), was used to indicate, which groups of teas statistically differed from each other due to determined total concentrations of elements in infusions and percentage contributions of chemical fractions of these elements distinguished in them by tandem-column SPE, e.g., LBTs-BBTs, LBTs-LGTs, LBTs-BGTs, BBTs-BGTs, BBTs-LGTs, or LGTs-BGTs. Considering total concentrations of elements, percentage contributions of separated chemical fractions (HF, CF, RF) of these elements and the number of possible differences for a single element, it appeared that the highest discrimination power of studied four groups of teas had Ni (16 differences out of 24 possible), Al (15) and Fe (12). Considering all ten elements and the number of possible differences for a given measure (the total concentration or the percentage contribution of the chemical fraction), it was established that studied groups of teas mostly differed due to percentage contributions of the CF (23 differences out of possible 60) and the HF (21). The number of differences in case of total concentrations of elements and percentage contributions of their RF was similar and equaled to 13.
| CT | CHF | CCF | CRF | ||
|---|---|---|---|---|---|
| % of variance | PC1 | 27.5 | 36.4 | 26.3 | 33.8 |
| PC2 | 24.3 | 22.8 | 23.7 | 23.0 | |
| Cumulative variance, % | 51.8 | 59.2 | 50.0 | 56.8 | |
| Eigenvalues | PC1 | 2.75 | 3.64 | 2.632 | 3.385 |
| PC2 | 2.43 | 2.28 | 2.375 | 2.306 | |
| Component weights (only > 0.20) | PC1 | 0.49 (Al), 0.21 (Fe), 0.43 (Mg), 0.52 (Mn), 0.44 (Ni) | 0.45 (Al), −0.37 (Cu), 0.43 (Fe), 0.42 (Mg), 0.21 (Mn), −0.42 (Zn) | 0.54 (Al), 0.49 (Mg), 0.50 (Mn), 0.25 (Ni), 0.24 (Sr), −0.22 (Zn) | 0.34 (Ca), 0.41 (Fe), −0.42 (Mg), −0.43 (Ni), 0.50 (Sr), −0.26 (Zn) |
| PC2 | −0.20 (Al), 0.55 (Ca), −0.38 (Cu), 0.52 (Sr), 0.38 (Zn) | 0.29 (Ba), 0.56 (Ca), −0.22 (Mn), 0.39 (Ni), 0.60 (Sr) | −0.22 (Al), 0.54 (Ca), −0.23 (Cu), −0.32 (Fe), 0.22 (Ni), 0.49 (Sr), 0.45 (Zn) | 0.46 (Al), −0.22 (Ba), 0.31 (Cu), −0.22 (Mg), 0.58 (Mn), −0.26 (Ni), 0.37 (Zn) | |
| Contribution of variables to PCs (only > 10%), % | PC1 | 24 (Al), 18 (Mg), 27 (Mn), 19 (Ni) | 20 (Al), 14 (Cu), 19 (Fe), 18 (Mg), 18 (Mn) | 29 (Al), 24 (Mg), 25 (Mn) | 17 (Fe), 20 (Mg), 20 (Ni), 20 (Sr) |
| PC2 | 30 (Ca), 14 (Cu), 27 (Sr), 14 (Zn) | 32 (Ca), 15 (Ni), 36 (Sr) | 30 (Ca), 10 (Fe), 24 (Sr), 20 (Zn) | 19 (Al), 12 (Cu), 29 (Mn), 14 (Zn) | |
| No separation between samples of BTs and samples of GTS | Clear separation between samples of BTs and samples of GTs | No separation between samples of BTs and samples of GTs | Clear separation between samples of all four groups of BTs and GTs, i.e., LBTs, BBTs, LGTs and BGTs |
PC1, PC2. Principle components. LBTs Leaf black teas. BBTs Bagged black teas. LGTs Leaf green teas. BGTs Black green teas.
Considering average percentage contributions of chemical fractions of elements evaluated for both studied tea varieties, i.e., BTs and GTs, it was found that chemical fractionation patterns of certain elements in these teas was quite different. The percentage contribution of the HF of Al, Fe and Mg determined in infusions of GTs was higher by about 3-, 4- and 2-fold, respectively, than this assessed for these elements in infusions of BTs. It meant that infusions of GTs likely contained more water-soluble hydrophobic complexes of Al, Fe and Mg than in case of infusions of BTs. It was supposed that it happened because both varieties of teas considerably differed due to chemical composition. Accordingly, the level of tannins in GTs is lower than this in BTs and less complex polyphenolic compounds are present in their matrix, i.e., catechins and other flavonoids (myricetin, quercetin, kaempferol) in addition to LMW organic acids (Ramdani et al., 2013). These compounds could bind Al, Fe and Mg ions to a higher degree while resultant species were likely soluble in water in much higher quantity (Dambiec et al., 2013). In case of BTs, due to the fermentation process, more complex and condensed structures of polyphenolic compounds, including theaflavins and thearubigins, are usually present in their matrix. Therefore, they could strongly bind and retain ions of elements, making them insoluble during brewing of this tea variety (Ramdani et al., 2013). In case of the RF, the percentage contribution of this fraction found for Ba, Fe, Ni and Sr in infusions of GTs was also higher by 2-, 2-, 12- and 5-fold, respectively, as compared to this established for these elements in infusions of BTs. The reason for this could be a higher content of organic acids in GTs as compared to BTs that could complex mentioned ions of elements, contributing in this way to presence of their LMW anionic and/or neutral complexes in infusions of this tea variety and in the RF as separated by tandem-column SPE. Considering the CF, it was found that percentage contributions of this fraction determined in infusions of BTs and GTs were similar in case of several elements, i.e., Ca (73% and 84% on average for infusions of BTs and GTs, respectively), Mg (89% and 90% on average for infusions of BTs and GTs, respectively), Mn (85% and 89% on average for infusions of BTs and GTs, respectively), Ni and Sr (83% and 82% on average for infusions of BTs and GTs, respectively), and Zn (84% and 92% for infusions of BTs and GTs, respectively). High percentage contributions of this fraction in BTs and GTs pointed out that these elements were highly bioavailable and bioaccessible from infusions of both tea varieties, likely including simple ions of elements, their LMW cationic complexes and labile species (Ozdemir and Gucer, 1998; Odegard and Lund, 1997; Mucchino and Musci, 2014; Pohl and Prusisz, 2007a,b). Percentage contributions of the CF of Al and Fe were significantly lower, i.e., 51% and 43% on average for Al in case of infusions of BTs and GTs, respectively, and 67% and 9% on average for Fe in case of infusions of BTs and GTs, respectively. For these two elements, relatively high percentage contributions of the HF could additionally point out that their bioavailability from infusions of both tea varieties was lower.
General conclusions about possible availability and percentage shares of operationally defined chemical fractions of selected elements from infusions of BTs were consistent with those reported in other research works (see Table 4S). Slight differences in results were likely due to different sorbents used for column SPE or application of a batch process, e.g., as in case of work by Mucchino and Musci (2014). In the latter work, SPE of elements species was proceeded for 24 h and therefore, percentage contributions of the HF of Al and Fe could be overestimated, i.e., 31% and 64% for Al and Fe, respectively, as compared to those evaluated in the present work, i.e., 6% and 15%, respectively. Similar comparison of results in case of infusions of GTs was limited because there was no data available on speciation of elements in infusions of this tea variety.
3.3 Chemometric data mining by principal component analysis and linear discriminat analysis
For the purpose of data exploration and analysis of correlations and similarities between BTs, as represented by LBTs and BBTs, and GTs, as represented by LGTs and BGTs, PCA was used at first to the data matrix with total concentrations of ten elements (Al, Ba, Ca, Cu, Fe, Mg, Mn, Ni, Sr and Zn) determined in infusions of these teas. Pearson (n) type PCA was applied and it was based on the correlation matrix and standardized data to avoid inflating the impact of variables with high variances on outcomes of this multivariable analysis. Results of PCA are given in Table 4 with variance explained by two principle components (PCs), i.e., PC1 and PC2, eigenvalues reflecting quality of projection of the ten-dimensional data matrix to the two-dimensional matrix of PC1 and PC2, component weights representing partial correlation between variables and both PCs and contributions of variables to these PCs. Unfortunately, PC1 and PC2 accounted together only for 52% of total variance when total concentrations of elements were used. No clear separation between four types of studied teas, i.e., LBTs, BBTs, LGTs and BGTs, and even between two varieties, i.e., BTs and GTs, was achieved (see Fig. 1a).
Therefore, considering better discrimination between both tea varieties due to information on chemical fractionation patterns of elements assessed in infusions of BTs and GTs, concentrations of studied elements determined in the HF, the CF and the RF were used as variables for PCA in the next turn. Results of these multivariable analysis are given in Table 4. It was established that indeed, the use of concentrations of elements in the HF and the RF could explain up to 59% and 57% of total variance in data for all studied samples, respectively, but it was also possible to clearly separate samples of BTs from GTs (Fig. 1b with PCA based on concentrations of elements in the HF) or even samples of LBTs from BBTs, LGTs and BGTs (Fig. 1d with PCA based on concentrations of elements in the RF). These results pointed out that information about chemical fractionation patterns of elements, particularly percentage shares and concentrations of elements in the HF and the RF, in infusions of BTs and GTs and their different types, i.e., LBTs, BBTs, LGTs and BGTs, unquestionably led to much better distinction and grouping of tea samples due to variety (black and green) and type (leaf and bagged) than usually determined and applied total concentrations of these elements.
At the end, linear discrimination analysis (LDA) was applied to evaluate ability of information about chemical fractionation of elements in infusions of BTs and GTs to categorize and classify their different samples due to variety and type of tea. As before, total concentrations of elements and their concentrations in separated chemical fractions were used to build respective data matrices. The backward selection algorithm was used to select variables that were responsible for categorization of all samples into four groups, i.e., LBTs, BBTs, LGTs and BGTs. Results of this analysis are given in Table 5 and relate to eigenvalues for three discrimination functions (DFs), i.e., DF1, DF2 and DF3, percentages of variance explained by DFs and significance measures (p values) of models along with elements that were statistically significant discriminators of samples between four groups. As can be seen from Table 5, sensitivity of classification was within 95–100%, meaning that 19 or 20 out of 20 samples were correctly classified into four different groups using models based on total concentrations of elements or their concentrations in separated chemical fractions. The number of useful predictors enabling classification of samples was 3 (case of total concentrations of elements and concentrations of elements in the HF) or 5 (concentrations of elements in the CF or the RF).
| CT | CHF | CCF | CRF | ||
|---|---|---|---|---|---|
| Eigenvalues | DF1 | 7.00 | 10.23 | 11.71 | 46.27 |
| DF2 | 2.00 | 3.20 | 4.85 | 4.95 | |
| DF3 | 0.14 | 0.03 | 0.73 | 0.57 | |
| % of variance | DF1 | 76.6 | 76.0 | 67.7 | 89.4 |
| DF2 | 21.9 | 23.8 | 28.0 | 9.6 | |
| DF3 | 1.5 | 0.2 | 4.2 | 1.0 | |
| Cumulative variance, % | 100.0 | 100.0 | 99.9 | 100.0 | |
| p value | DF1 | 0.000 | 0.000 | 0.000 | 0.000 |
| DF2 | 0.001 | 0.000 | 0.000 | 0.000 | |
| DF3 | 0.175 | 0.489 | 0.047 | 0.089 | |
| Statistically significant variables | Al, Fe, Ni | Al, Fe, Ni | Al, Mg, Mn, Ni, Zn | Al, Mg, Ni, Sr, Zn | |
| Correctly classified | 95% (19 out of 20) | 95% (19 out of 20) | 100% (20 out of 20) | 95% (19 out of 20) |
Backward selection algorithm was used. F-to-enter: 4.0, F-to-remove: 4.0.
Graphical representation of LDA, showing separated four groups of samples (LBTs, BBTs, LGTs and BGTs) along with centroids for each group, is given in Fig. 1S. It can be seen that likewise for PCA, the best separation of four groups of teas was obtained considering concentrations of elements in the RF. In this case, centroids were far from each other. A similar effect was also obtained when using concentrations of elements in the CF for building the data matrix.
4 Conclusions
Results of this work indicate that elements such as Ca, Mg, Mn, Ni, Sr and Zn are highly bioavailable from infusions of black and green teas because average percentage shares of the cationic fraction of these elements were 85 ± 5%. Bioavailability of Al and Fe (in reference to percentage shares of the cationic fraction of these elements) was lower due to relatively high contributions of other fractions, i.e., hydrophobic and residual species.
Analysis of variance, principle component analysis and linear discriminant analysis of data obtained for analyzed samples of two tea varieties (black and green teas), additionally divided into four groups due to type (leaf black, bagged black, leaf green, bagged green), clearly pointed out that chemical fractionation of elements in infusions of these teas brought new useful and important information. This information can be useful for univocal differentiation and classification of teas due to variety and type. Collected data on chemical fractionation patterns of studied elements, including percentage shares of hydrophobic, cationic and residual fractions of elements and their concentrations in these fractions, in infusions of different black and green teas are found to be much more useful then total concentrations of elements, overwhelmingly determined and used for evaluation of the nutritional status of infused black and green teas and assessment of differences between these two tea varieties for the purpose of their classification and discrimination.
Acknowledgements
The work was financed by a statutory activity subsidy from the Polish Ministry of Science and Higher Education for the Faculty of Chemistry Wroclaw University of Technology. It was also undertaken in the course of the research project 2013/09/B/NZ9/00122 funded by the National Science Centre.
References
- Evaluation of macro- and microelement levels in black tea in view of its geographical origin. Biol. Trace Elem. Res.. 2017;176:429-441.
- [Google Scholar]
- Monitoring of essential and heavy metals in green tea from different geographical origins. Environ. Monit. Assess.. 2016;188:183. (11 pp)
- [Google Scholar]
- Directly coupled high performance liquid chromatography – inductively coupled plasma-mass spectrometry for the determination of organometallic species in tea. Microchem. J.. 1996;54:88-110.
- [Google Scholar]
- Cation exchange microcolumn combined with fluorination-assisted electrothermal vaporization ICP-AES for speciation of aluminium in tea leaves. At. Spectrosc.. 2003;24:93-97.
- [Google Scholar]
- Levels of essential and non-essential elements in black teas commercialized in Poland and their transfer to tea infusion. J. Food Comp. Anal.. 2013;31:62-68.
- [Google Scholar]
- Simplified tea classification based on reduced chemical composition profile via successive projections algorithm linear discriminant analysis (SPA-LDA) J. Food Comp. Anal.. 2015;39:103-110.
- [Google Scholar]
- Speciation of aluminium in tea infusions by ion-exchange resins and flame AAS detection. Anal. Chim. Acta. 2000;411:81-89.
- [Google Scholar]
- Determination of aluminium and zinc in infusion tea cultivated in north of Iran. J. Environ. Health Sci. Eng.. 2015;13:49. (6 pp)
- [Google Scholar]
- Determination of essential elements in beverages, herbal infusions and dietary supplements using a straightforward sequential approach based on flame atomic absorption spectrometry. Food Chem.. 2017;219:69-75.
- [Google Scholar]
- Statistics Plain and Simple (fourth ed.). Boston: Cengage Learning; 2017.
- Determination of antioxidant activity, rutin, quercetin, phenolic acids and trace elements in tea infusions: influence of citric acid addition on extraction of metals. J. Food Comp. Anal.. 2015;40:70-77.
- [Google Scholar]
- Micronutrients (B Co, Cu, Fe, Mn, Mo, and Zn) content in made teas (Camellia sinensis L.) and tea infusion with health prospects: a critical review. Crit. Rev. Food Sci. Nutr.. 2017;57:2996-12034.
- [Google Scholar]
- Comparative assessment of copper, iron, and zinc contents in selected Indian (Assam) and South African (Thohoyandou) tea (Camellia sinensis L.) samples and their infusions: a quest for health risks to consumers. Biol. Trace Elem. Res.. 2017;175:475-487.
- [Google Scholar]
- Comparison of infusions from black and green teas (Camellia sinensis L. Kuntze) and Erva-mate (Ilex paraguariensis A. St.-Hil.) based on the content of essential elements, secondary metabolites, and antioxidant activity. Food Anal. Methods. 2017;10:3063-3070.
- [Google Scholar]
- The impact of wine components on fractionation of Cu and Fe in model wine systems: macromolecules, phenolic and sulfur compounds. Food Res. Int.. 2017;98:95-102.
- [Google Scholar]
- Speciation of aluminium in tea infusions by use of SEC and FPLC with ICP-OES and ES-MS-MS detection. Anal. Bioanal. Chem.. 2005;383:467-475.
- [Google Scholar]
- A comparison of the potential health risk of aluminium and heavy metals in tea leaves and tea infusion of commercially available green tea in Jiangxi. China. Environ. Monit. Assess.. 2015;187:228. (12 pp)
- [Google Scholar]
- Multielement determination and speciation of major-to-trace elements in black tea leaves by ICP-AES and ICP-MS with the aid of size exclusion chromatography. Anal. Sci.. 2001;17:391-398.
- [Google Scholar]
- Comparison of different digestion methods of green and black tea at the Sarajevo market for the determination of the heavy metal content. J. Food Measure. Character.. 2014;8:149-154.
- [Google Scholar]
- Trace elements in Camellia sinensis marketed in southeastern Brazil: extraction from tea leaves to beverages and dietary exposure. LWT – Food Sci. Technol.. 2016;68:491-498.
- [Google Scholar]
- Evaluation of direct analysis for trace elements in tea and herbal beverages by ICP-MS. J. Braz. Chem. Soc.. 2015;26:1211-1217.
- [Google Scholar]
- Extraction of Al, Cu, Fe, Mn, Ni and Zn-polyphenol complexes from black tea infusions by Amberlite resins. J. Sci. Food Agric.. 2014;94:2234-2238.
- [Google Scholar]
- Multi-element speciation of tea infusion using cation-exchange separation and size-exclusion chromatography in combination with inductively coupled plasma mass spectrometry. J. Anal. At. Spectrom.. 1997;12:403-408.
- [Google Scholar]
- Determination of the effect of lemon addition upon element concentration in tea. Instrum. Sci. Technol.. 2014;42:153-160.
- [Google Scholar]
- Speciation of manganese in tea leaves and tea infusions. Food Chem.. 1998;61:313-317.
- [Google Scholar]
- Application of high resolution-continuum source flame atomic absorption spectrometry (HR-CS FAAS): determination of trace elements in tea and tisanes. Food Chem.. 2015;170:492-500.
- [Google Scholar]
- Trace metals and flavonoids in different types of tea. Food Sci. Biotechnol.. 2013;22:925-930.
- [Google Scholar]
- Availability of some elements from different type of teas. Nat. Prod. J.. 2013;3:292-295.
- [Google Scholar]
- Fractionation analysis of manganese and zinc in tea infusions by two-column solid phase extraction and flame atomic absorption spectrometry. Food Chem.. 2007;102:1415-1424.
- [Google Scholar]
- Simple and versatile operational fractionation of Fe and Zn in dietary products by solid phase extraction on ion exchange resins. Talanta. 2007;71:411-418.
- [Google Scholar]
- Fractionation of calcium and magnesium in honeys, juices and tea infusions by ion exchange and flame atomic absorption spectrometry. Talanta. 2006;69:1227-1233.
- [Google Scholar]
- Concentrations and solubility of selected trace metals in leaf and bagged black teas commercialized in Poland. J. Food Drug Anal.. 2015;23:486-492.
- [Google Scholar]
- Chemical composition, plant secondary metabolites, and minerals of green and black teas and the effect of different tea-to-water ratios during their extraction on the composition of their spent leaves as potential additives for ruminants. J. Agric. Food Chem.. 2013;61:4961-4967.
- [Google Scholar]
- On the use of solid phase extraction for the fractionation of aluminium species. Chem. Anal. (Warsaw). 2004;49:19-28.
- [Google Scholar]
- Toxic and essential elements content of black tea leaves and their tea infusions consumed in Iran. Biol. Trace Elem. Res.. 2010;134:109-117.
- [Google Scholar]
- Ni speciation in tea infusion by monolithic chromatography-ICP-MS and Q-TOF-MS. Anal. Bioanal. Chem.. 2013;405:2041-2051.
- [Google Scholar]
- Transition rates of selected metals determined in various types of teas (Camellia sinensis L. Kuntze) and herbal/fruit infusions. Food Chem.. 2017;215:22-30.
- [Google Scholar]
- Determination of lead, cadmium and arsenic in infusion tea cultivated in north of Iran. Iran. J. Environ. Health Sci. Eng.. 2012;9:37. (6 pp)
- [Google Scholar]
- Total content and speciation of aluminium in tea leaves and tea infusions. Food Chem.. 2007;104:1662-1669.
- [Google Scholar]
- Comparison and validation of different alternative sample preparation procedures of tea infusions prior to their multi-element analysis by FAAS and ICP OES. Food Anal. Methods. 2016;9:1398-1411.
- [Google Scholar]
Appendix A
Supplementary material
Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.arabjc.2018.02.013.
Appendix A
Supplementary material
Supplementary data 1
Supplementary data 1