Mean oxalate content of loose-packed black tea infusions after different brewing time (2g/240 ml)/ All analytical data are the mean of triplicate measurements of three independent samples SEM/ A one-factor ANOVA and the post hoc test showed that there were significant differences between the oxalate content of different brewing times (P
Acidity Of Different Samples Of Tea Leaves.pdfl
Mean oxalate content of loose-packed black tea infusions after different dilutions (2g/240 ml)/ All analytical data are the mean of triplicate measurements of three independent samples SEM/The results of independent t-test showed that there were significant differences between the mean oxalate content of different dilutions (P
As shown in Fig. 1, the range of oxalate in Iranian consumed loose-packed black tea after different brewing times was from 4.4 to 6.3 mg/240 ml. The brewing of tea in hot water extracts that portion of total oxalate that is soluble, leaving the insoluble oxalate (mainly oxalate bound to calcium) in the tea leaves [25]. The oxalate contents of black tea samples from tea bags for 1 and 5 min infusions were 10.8 and 20.8 mg oxalate/250 ml, respectively, for one brand and 10.8 and 12.0 mg oxalate/250 ml for another brand [26]. By contrast, no significant increase in oxalate content of Chinese green teas was observed by longer brewing time (5 vs. 10 min) [27]. The results of present study revealed incremental increases in soluble oxalate with increased brewing times all the way up to 60 min although after the 15 min time point, there was a marked decrease in the magnitude of this increase per unit of time.
Some researchers have used nondestructive detection technology in agriculture5,6,7,8. This is especially applicable for the prediction of moisture content in withering leaves. In the study conducted by Liang et al.9, 15 features of withering leaves were used to establish a moisture detection model. The Rp and the root-mean-square error of prediction (RMSEP) were 0.9314 and 0.0411, respectively. Shen et al.10 proposed the use of Elman neural network to predict the moisture content using miniaturized near-infrared spectroscopy and a smartphone, and the results were satisfactory. The E-eye technology could predict the moisture content of withering leaves because the color and texture features change regularly with the decrease in moisture level. Spectral technology could evaluate the moisture content of withering leaves based on changes in the effective features, such as some bands corresponding to the OH-stretching overtone spectra. Although both of these technologies could obtain the information related to the moisture of the withering leaves from different aspects, the application of a single technology produced one sided results: they only collected sample information from one aspect and ignored other types of sample information, thus it was difficult to collect the overall information of tea samples. From a biochemical perspective, fresh tea leaves display slow enzymatic and non-enzymatic reactions during the withering process. Their moisture and chlorophyll content decrease gradually, and the opposite is true for the theaflavin content, thus leading to a significant color change. Remarkably, near infrared spectroscopy can predict the moisture level based on the internal information of withering leaves, such as the OH-stretching overtone spectra, while machine vision technology uses appearance features for this task11,12. However, the obtained information of these technologies is relatively independent. Hyperspectral imaging technology has been used to evaluate the moisture content of tea, especially during the withering process, because it can express the image and spectral in-formation at the same time. An et al.13 and Wei et al.14, and Dong et al.4 established satisfactory moisture prediction models for single and accumulative withering leaves, respectively, and realized the visualization of moisture distribution. However, these studies established a moisture evaluation model using spectra data, and did not involve any appearance information. Only a few studies using data fusion strategy have detected the moisture of tea leaves in withering processing. In a recent study2, the withering degree was evaluated successfully by the fusion of colorimetric sensing array (CSA), E-eye and NIRS information. Liu et al.15 accurately predicted the moisture content using machine vision and NIRS during green tea processing. The above studies showed that a data fusion strategy could yield better prediction results than any single technology; therefore, developing new fusion strategies involving different technologies is an imminent task for the evaluation of tea quality.
In order to accurately predict the moisture content of withering leaves, their spectral and appearance information were collected to establish the prediction model. These technologies express the information of tea samples from different aspects. Hence, it was necessary to further analyze these sets of data.
Figure 2 presents the NIR spectral features of withering leaves. In Fig. 2a, the spectral information of all samples are displayed, and these spectra contain 760 bands. However, the absorbance changes irregularly at the end of raw spectra because of the influence of noise. Therefore, the absorbance of the last 15 bands were eliminated. Subsequently, the raw spectra without noise bands were subjected to SNV preprocessing, and the resulting spectra were displayed in Fig. 2b. SVR models were established using SNV spectral data and the SNV spectral data without noise bands, and the performance of the established SVR models were shown in Table 1. The spectral data without noise bands showed better performance, indicating that these noise bands affected the accuracy of the established moisture detection model. According to the moisture levels in our samples, three withering degrees, such as including insufficient withering, moderate withering and excessive withering, were established, and the average spectral curve of the spectral information for these withering degrees were shown in Fig. 2c. In the figure, the absorbance intensity showed the same trend for different withering degrees. However, there were some differences in the spectral profiles of different withering degrees. For instance, the spectra of insufficient withering leaves showed higher absorbance than the spectra of other withering degrees, a phenomenon also demonstrated by a previous study2. In addition, the absorbance intensity near 1450 nm decreased over the withering time. As can be seen in Fig. 2, obvious absorption peaks appear near 965 nm, 1093 nm, 1176 nm and 1450 nm. The absorption peaks near 960 nm and 1093 nm correspond to the second free OH and bound OH-stretching overtone spectra for water29. Furthermore, the absorption peak near 1176 nm is attributable to the second CH-stretching overtone spectra for catechins30. In addition, the absorption peak around 1450 nm is related to the first OH-stretching overtone spectra for water31. The groups corresponding to these absorption peaks provide a theoretical basis for establishing the moisture detection model.
Concentrations of selected metals (Cu, Mn, Zn, Cd) in tea leaves were investigated. Samples included black, green, and other (red, white, yellow, and oolong) teas. They were purchased on a local market but they covered different countries of origin. Beverages like yerba mate, rooibos, and fruit teas were also included in the discussion. Metal determinations were performed using atomic absorption spectrometry. In black teas, Mn/Cd ratio was found to be significantly higher (48,091 35,436) vs. green (21,319 16,396) or other teas (15,692 8393), while Cd concentration was lower (31.4 18.3 μg/kg) vs. other teas 67.0 (67.0 24.4). Moreover, Zn/Cu and Cu/Cd ratios were, respectively, lower (1.1 0.2 vs. 2.2 0.5) and higher (1086 978 vs. 261 128) when comparing black teas with other teas. Intake of each metal from drinking tea was estimated based on the extraction levels reported by other authors. Contributions to recommended daily intake for Cu, Mn, and Zn were estimated based on the recommendations of international authorities. Except for manganese, tea is not a major dietary source of the studied elements. From the total number of 27 samples, three have shown exceeded cadmium level, according to local regulations.
Manganese concentration in the samples varied from 457 4 to 2210 35 mg/kg (mean SD 962 388 mg/kg). Similar results were published by Street et al. [1], where manganese concentration in 30 samples of different types of teas varied from 511 to 2220 mg/kg. The authors did not notice a major difference between manganese concentration in black and green teas (nor they did for other elements: iron, zinc, and copper).
Comparing black and green teas, Mn/Cd ratio was found to be significantly different between these two groups. When comparing black teas to the others, four parameters showed significant differences: Cd concentrations, Mn/Cd, Zn/Cu, and Cu/Cd ratios. Further studies, including more tea samples, are needed to establish if there is such a general trend for these groups of teas.
In black teas, Mn/Cd ratio was found to be significantly higher vs. green or other teas, while Cd concentration was lower vs. other teas. Moreover, Zn/Cu and Cu/Cd ratios were, respectively, lower and higher when comparing black teas with other teas. This differentiation can be caused by the fermentation process during black tea production. Our results partly agree with the reports of other researchers; however, some differences can be noticed. In particular, zinc content in black tea as well as cadmium content in black and green teas was found to be much lower than reported by other authors. Very high content of manganese in two samples of black teas from Kenya was observed. Tea is a major dietary source of manganese while the intake of other elements is negligible. In three samples, content of cadmium was found to be higher than allowed by regulations of the Health Minister of Poland. 2ff7e9595c
Comments