الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 195 |  |  |
| | | from | 195 | | |
Abstract
Herbal products are of great interest nowadays as they are expected to be safer with less side effects and somehow more effective than synthetic drugs. Herbal products are complex multi-component mixtures that possess high risk of adulteration and contamination, so their quality control is very crucial for both human health and finance.
Due to high number of variables affecting the quality of herbal products such as geographical origin and factors affecting plant growth like climatic conditions, age of the plant and time of collection, so multivariate analysis is the best tool to find patterns and relationships between these several variables simultenously.
Green tea is one of the most important herbal products that is used nowadays in different biological and non-biological aspects. Green tea prices vary depending on the quality. Many quality control methods were developed before, but all had a lot of disadvantages. A new rapid easy method of quality control of green tea is highly needed to build a profile of green tea according to its geographical origin, which is a highly quality-influencing factor which comprises a lot of variables such as harvesting time, proper collection and proper drying at correct temperature, latitude and altitude, water supply, soil and nutrients.
Our objective was to investigate the ability of using modern chemometric approaches for quality authentication of green tea as a raw herbal material or finished products (herbal tea bag) in the market. Green tea samples were collected from different native countries (China, India, Kenya & Sri Lanka) and purchased from different markets reaching a total of forty-three samples. Different spectroscopic and chromatographic methods were applied along with unsupervised chemometric techniques such as PCA and HCA to study the optimized method or model for sample classification and green tea discrimination. The resulted model then was validated using supervised chemometric technique such as SIMCA.
107
Chapter I
Spectroscopic Analysis as a Chemometric Model for the Identification and Authentication of Green Tea:
1- Ultra Violet Spectroscopy as a Chemometric Model for the Identification and Authentication of Green Tea:
UV spectroscopy in combination with PCA and HCA was used to discriminate between thirty-one genuine green tea samples collected from different geographical origins. Only one sample was considered to be an outlier, which is SF “Fannings from Sri Lanka”; due to its fine condition compared to all other samples, which indicated that the particle size may affect the quality of green tea. The rest of samples were clustered into two main clusters; the first one contains the majority of Chinese samples, the second one contains all Indian and Ceylon samples along with the majority of Kenyan samples and four (out of eleven) Chinese ones (CD, CX, CM & CL). Since Ceylon green tea is superior in quality than Chinese one, so an explanation of these four samples was made depending on their quality, which was related to their geographical origin.
Validation of the built UV model was achieved using supervised chemometric techniques using twelve commercial samples collected from Egyptian and Chinese markets. The results have confirmed the results obtained from PCA score plot.
UV spectroscopy is considered an easy, rapid and non-destructive method, which has also the advantage of being available in many laboratories, which makes it a convenient technique to be used.
2- Visible Spectroscopy as a Chemometric Model for the Identification and Authentication of Green Tea:
Visible spectroscopy in combination with PCA and HCA was used to discriminate between thirty-one genuine green tea samples collected from different geographical origins. Only one sample was considered to be an outlier, which is SF “Fannings from Sri Lanka”; due to its fine condition compared to all other samples, which indicated that the particle size may affect the quality of green tea as in UV measurements.
108
Both PCA score plot and clustering dendogram obtained could not segregate samples according to their geographical origins.
3- Fourier Transformer Infrared Spectroscopy as a Chemometric Model for the Identification and Authentication of Green Tea:
Fourier Transformer Infrared spectroscopy in combination with PCA and HCA was used to discriminate between thirty-one genuine green tea samples collected from different geographical origins after applying Standard Normal Variate (SNV) transformation method on the data matrix.
Both PCA score plot and clustering dendogram failed to give any explanation regarding the segregation of genuine green tea samples from different geographical origins. However, it was observed that SF “Fannings sample from Sri Lanka” was not detected as an outlier; as all the genuine green tea samples were ground to approximately the same condition (powdered form) as SF.
4- Nuclear Magnetic Resonance Spectrometry as a Chemometric Model for the Identification and Authentication of Green Tea:
Nuclear Magnetic Resonance Spectrometry in combination with PCA was used to discriminate between thirty-one genuine green tea samples collected from different geographical origins. Most of Kenyan samples were clustered together, while most of South Asian (Indian and Ceylon) samples were clustered together along with CD, CX and CL, which have a possible explanation as mentioned before in UV measurements. In addition, CY can be observed near the cluster of South Asian samples, which can be explained by its geographical location.
Chapter II
Chromatographic Analysis as a Chemometric Model for the Identification and Authentication of Green Tea:
1- High-Performance Thin-Layer chromatography as a Chemometric Model for the Identification and Authentication of Green Tea:
High-Performance Thin-Layer chromatography in combination with PCA and HCA was used to discriminate between thirty-one genuine green tea samples collected from different
109
geographical origins. PCA score plot of the peak area percentage of chromatographic data matrix of the thirty-one genuine green tea samples collected from different geographical origins scanned at 450 nm revealed that samples are clustered into four main clusters; according to their geographical origins.
Validation of the built HPTLC model was achieved using supervised chemometric techniques using twelve commercial samples collected from Egyptian and Chinese markets. The results have confirmed the results obtained from PCA score plot and the caffeine content in each genuine green tea sample was quantified.
2- High-Performance Liquid chromatography as a Chemometric Model for the Identification and Authentication of Green Tea:
PCA score plot was constructed using peak area for the thirty-one genuine green tea samples collected from different geographical origins at 63 retention times. It was not able to segregate samples according to their geographical origins.
Validation of developed HPLC method was done according to ICH using caffeine as standard. The calibration curve of caffeine exhibited good linearity (r2 > 0.999) within test ranges. The LOD was 11.46236 μg/ml, while LOQ was 34.73442 μg/ml. In addition, the developed analytical method had good accuracy with overall recovery 98.61 ± 5.418892%. Intra-day and Inter-day variability (%) were 0.226575 and 4.453251 respectively.
Chapter III
Biological Assessment of Different Green Tea (Camellia sinensis) Samples from Different Geographical Origins:
Antioxidant activity of genuine green tea samples collected from different geographical origins was measured using DPPH method. The half maximal inhibitory concentration (IC50) of each sample was calculated and a bar chart was drawn. It was observed that not all samples are the same in their antioxidant capacity. Therefore, it could not give any evidence about the antioxidant activity of each sample in relation to its geographical origin