Drying ginger And Preserving 6-gingerol Content

by

LiZhuo Li

Supervisor: Dr Robert Driscoll

Co-supervisor: Dr George Srzednicki A thesis submitted in fulfilment of the requirements for the degree of Masters by Research

School of Chemical Engineering, Faculty of Engineering The University of New South Wales Sydney, Australia July 2017

ORIGINALITY STATEMENT

'I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.'

Signed

Date

Abstract

Ginger rhizome (Zingiber officinale Roscoe) is widely used as a spice or a folk medicine. 6-gingerol is the major bioactive component in fresh ginger and has numerous physiological effects. 6-gingerol is heat sensitive while cooking and drying will transform 6-gingerol to 6-shogaol. Therefore, 6-gingerol content is used to determine the quality of ginger after drying. A drying model called the Two layer model was tested for prediction of drying ginger and compared with a single layer model. In this study, two layer model was used to describe ginger drying process. 6-gingerol content was measured by using HPLC method. Several factors which could affect 6-gingerol content were reviewed and a 6-gingerol prediction model was established from the experimental data.

The results showed that the two layer drying model gave no significant improvement to describing the ginger drying process compared with the single layer model. Drying time and relative humidity (ranging from 10% to 40%) impacted 6-gingerol content, although drying temperature (ranging from 30°C to 60°C) had less effects on 6-gingerol content. It was found that 6-gingerol content was highly variable in fresh ginger, which making conclusions on models difficult.

Keywords: Ginger, 6-gingerol, thin layer drying, two layer model, 6-gingerol perdition Content Chapter 1 Literature Review: ...... 1 1.1 Ginger Introduction ...... 1 1.1.1 Background of Ginger...... 1 1.1.2 Diversity of Ginger ...... 1 1.1.3 Uses of Ginger and Relative Products ...... 2 1.1.4 Global Ginger Production ...... 3 1.1.5 Global Trade in Ginger ...... 3 1.2 Chemistry of Ginger...... 5 1.2.1 Composition of Ginger...... 5 1.2.2 Essential Oil ...... 5 1.2.3 Gingerol and Shogaol...... 6 1.2.4 Medicinal use of Ginger...... 7 1.2.5 Pharmacological use of Ginger ...... 9 1.3 Drying Theory ...... 10 1.3.1 Purpose of Drying ...... 10 1.3.2 Drying of Ginger ...... 10 1.3.3 Thin Layer Drying ...... 11 1.3.4 The Limitation of Existing Thin Layer Drying Models ...... 15 1.3.5 Two Layer Diffusion Model for Changed Drying Condition ...... 16 1.4 Separation and Identification Methods ...... 18 1.4.1 Chemical Separation Technology ...... 18 1.4.2 Chromatography...... 18 1.4.3 Chromatography-Mass Spectrometry Method ...... 19 1.4.4 High-Performance Liquid Chromatography Method...... 19 Chapter 2 Materials and Methods ...... 21 2.1 Raw materials, Chemicals and Equipment used ...... 21 2.2 Drying Preparation ...... 21 2.3 Drying Procedure ...... 22 2.4 Drying Conditions ...... 23 2.5 Extract Preparation...... 24 2.6 Instrumentation and Chromatographic Conditions ...... 24 2.7 Data Analysis Procedure ...... 25 Chapter 3 Results and Discussion ...... 26 3.1 Individual Model Fitting ...... 26 3.2 Global Model Fitting...... 27 3.3 HPLC Analysis of 6-gingerol Content and Predictive Modelling ...... 28 3.3.1 6-gingerol Content ...... 28 3.3.2 6-gingerol Predicting Model ...... 33 3.3.3 Humidity Effect on 6-gingerol Preservation ...... 35 3.4 Optimum Drying Condition ...... 36 Chapter 4 Conclusion ...... 38 Bibliography ...... 39 Appendix ...... 45 Acknowledgements

I am indebted to Dr Robert Driscoll and Dr George Srzednicki for their support during whole experiment and thesis writing period, to Ms Kitty Tang for her support in the HPLC analysis. Ginger supply supports by Mrs Pam Fielder from Buderim Ginger in Yandina Qld. Laboratory support by the School of Chemical Engineering are gratefully acknowledged. Chapter 1 Literature Review:

1.1 Ginger Introduction

1.1.1 Background of Ginger

Ginger is a kind of universal spice which is widely cultivated and used in the world, particularly in Asia for example China and India. Ginger is consumed in various forms. In Asia, it is normally used in cooking as a spice, while ginger drinks, ginger oil, as well as dried ginger products are sold on the market in Europe and America. Ginger root is used as a tonic to treat common ailments in India and ancient China. Ginger, also named as Zingiber Officinale Rosc, is classified as a member of the tropical and sub-tropical family Zingiberaceae, and originated in tropical rainforests in southern Asia before spreading to Mediterranean regions by the 1st century. In ancient Rome, ginger was a popular spice used to make delicacies (sweets) in medieval times. Throughout the history of global trade, ginger has been traded longer than most other spices. It was regarded as a costly herb for its medicinal merits and nutritional value in the ancient world, and in the14th century, it cost the same as a piece of livestock (Katie, P. 2011).

1.1.2 Diversity of Ginger

Over the long history of ginger trading around the world, ginger has been planted on almost all of the continents. Given different growing environment, ginger has developed into several species. Table 1 lists some important ginger species in the business market (Kizhakkayil et al, 2011). In business trading, the product ginger often being marked by the country where it is from, such as Chinese ginger, Indian ginger, Australian ginger and Jamaican ginger (Pakrashi & Pakrashi, 2003). However, ginger actually has an even more extensive cultivar diversity, so that even in a country, there could be dozens of cultivars. Generally, one cultivar often comes from a specific growing place, and so people have often named the growing place of the ginger before the cultivar to distinguish different cultivars (Ravindran & Babu, 2005).

1 Table 1. Some of the economically important Zingiber species (Kizhakkayil et al 2011). Species/subspecies Occurrence Use Z. officinale Roscoe Tropical countries, China, USA Spice, condiment, medicinal Z. officinale var. rubrum Malaysia Medicinal, spice Z. officinale var. rubra Malaysia Medicinal, spice Z. mioga Roscoe Japan Vegetable. Shoot and flower are edible Z. zerumbet (L.) Smith Tropical Asia Medicinal, ornamental Z. montanum (Koenig) Link ex Dietr India, Malaysia, SriLanka, Java Used in traditional medicine Z. clarkii King Sikkim Himalayas Ornamental Z. aromaticum Val Tropical Asia Ornamental, medicinal, flavouring Z. rubens Roxb Indo-Malaya Medicinal, ornamental Z. griffithii Baker Malaysia Ornamental Z. ottensii Valet South East Asia Medicinal, ornamental Z. corallinum Hance South East Asia Chinese medicine, ornamental Z. americanus Bl. South East Asia Medicinal, vegetable Z. argenteum (J. Mood and I. Theilade) Sarawak, Malaysia Ornamental

In China, the major superior ginger cultivars include LaiWu ginger, which includes LaiWu slice ginger and LaiWu big ginger, GuangDong fleshy ginger, which has two secondary species, Sparse-ringed big fleshy ginger and Dense-ringed delicate fleshy ginger, ZheJiang red-claw ginger, TongLing ginger, YuLin ginger, LaiFeng ginger, XingGuo ginger, FuJiang red bud ginger and ZunYi white ginger. In India, several species reported include Z. squarrosum Roxb. , Z. officinale Rosc. , Z. griffithii Baker. , Z. barbatum Wall. , Z. spectabile Griff. , Z. parishii Hook. , Z. pardocheilum Wall. , and Z. marginatum Roxb (Baker JG, 1882).

1.1.3 Uses of Ginger and Relative Products

Ginger has been used as a main ingredient in many products through the world. Fresh ginger roots are juicy with a mild taste and can be used as spices for sweet or salty food such as soup, meat, vegetable, seafood, pickle, drinks and cake. They can be used to make jam and sweets. Ginger essential oil is the major constitute for ginger beer and ginger ale. In Japan, pickled ginger is a traditional seasoning which is often served with sushi or sashimi. In India, ginger is a common ingredient in several recipes, for example curry.

2 Fresh ginger can be eaten directly. However, due to the strong pungent flavour of fresh ginger root, fresh ginger is normally dried, and used to produce ginger powder and ginger slices, which makes the spicy flavour weaker. In western countries, ginger powder can be used to make ginger teas, ginger bread, crackers and ginger biscuits.

1.1.4 Global Ginger Production

Ginger is cultivated worldwide, but the vast majority of ginger is grown and harvested in Asia. According to data from the Food and Agriculture Organization Statistics Division (FAOSTAT), Asia contributed 87.3% of the total ginger production between 2000 and 2014, while Africa was 12%. India ranked first with respect to ginger production from 2010 to 2013, with about 33% of the global ginger harvesting quantity, followed by China (about 20%) and Nepal (about 11%). Nigeria and Thailand were once ranked 4 and 5, before Indonesia (produced 232,669 tons) surpassed these two countries and nearly caught up Nepal with 235,033 tons in 2013.

1.1.5 Global Trade in Ginger

Various forms of ginger products are traded across the world, for example，fresh ginger, crushed ginger, ground ginger and ginger ale. Generally, China, India, Nigeria, Thailand and Netherland are the major ginger export countries. Japan, US, Germany, Saudi Arabia and United Kingdom are main ginger import nations. According to the data from The United Nations Commodity Trade Statistics Database (COMTRADE), in 2015, global ginger exporters sold US$668.1 million worth of non-crushed and non-ground ginger and US$ 78.7 million worth of crushed or ground ginger. Table 2 and Table 3 lists top 10 countries that sold highest US dollar value worth of ginger, while Table 4 and Table 5 shows top 10 countries that imported highest US dollar value worth of ginger. All data in Table 2, Table 3, Table 4 and Table 5 are from COMTRADE.

3 Table 2: Top 10 Top Non-Crushed and Non-Ground Ginger Exporters in 2015 Rank Country Value (US) 1. China $416.6 million 2. Netherlands $63.7 million 3. India $37.2 million 4. Nigeria $26.1 million 5. Thailand $25.1 million 6. Peru $22.1 million 7. Indonesia $17.5 million 8. Brazil $10.2 million 9. Germany $5.8 million 10. Singapore $3.83 million

Table 3: Top 10 Crushed or Ground Ginger Exporters in 2015 Rank Country Value (US) 1. China $31.4 million 2. India $11.4 million 3. Nigeria $7 million 4. Germany $5.4 million 5. Netherlands $3.9 million 6. United Kingdom $3 million 7. Vietnam $2.6 million 8. France $1.5 million 9. United States $1.2 million 10. Spain $940,000

Table 4: Top 10 Non-Crushed and Non-Ground Ginger Importers in 2015 Rank Country Value (US) 1. Japan $104.3 million 2. USA $104 million 3. Netherlands $68.1 million 4. Pakistan $63.8 million 5. Saudi Arabia $34.1 million 6. United Kingdom $29.6 million 7. Germany $28.4 million 8. Malaysia $23.8 million 9. Russian Federation $20.6 million 10. Canada $19.9 million

4 Table 5: Top 10 Crushed or Ground Ginger Importers in 2015 Rank Country Value (US) 1. Japan $13.4 million 2. Germany $11.4 million 3. USA $9.2 million 4. Malaysia $7.4 million 5. United Kingdom $7.2 million 6. Netherlands $4.1 million 7. France $2.4 million 8. Canada $1.9 million 9. Australia $1.4 million 10. Austria $1.3 million

1.2 Chemistry of Ginger

1.2.1 Composition of Ginger

Ginger contains a variety of constituents. Researchers have found more than one hundred compounds which can be classified into three groups: essential oils, gingerol, and diarylheptanoids. Essential oils are hydrophobic liquids, containing volatile chemical constituents. Distillation and extractions are the most common waies to isolate the essential oils. The major component of essential oils is the terpenoids, including monoterpenes and hemiterpenes. Most compounds from these two groups have a strong volatile aroma and biological activity, which are important ingredients in medicine, cosmetics, and food production. Gingerols are major pungent constituents of ginger which are made up by several different compounds. Gingerols have a 4-hydroxy-3-methoxyphenyl group in chemical structure, varying according to different aliphatic chains attached to the main group. Gingerols can be classified as gingerol, shogaol, gingerdione and gingerdiol.

1.2.2 Essential Oil

Among all of the compounds in ginger, essential oil plays an important role in improving consumers’ systemic system. The benefits of ginger essential oil include 5 offering a warm, spicy aroma which enhances feelings of vitality and promotes feeling of physical well-being, and helps improve body blood circulation. It is a frequent addition to blends for massage, arthritis and muscle aches and pains. Ginger oil is commonly used to soothe, comfort, and balance digestive discomfort. While the benefit of ginger essential oil goes further than just comfort mood. It has significantly antifungal effects against Fusarium oxysporum (Singh, Gurdip et al, 2005), Fusarium moniliforme (Singh, Gurdip et al, 2008) and effectively prevent nausea in general anaesthesia patients at high risk for post-operative nausea and vomiting (Geiger & James 2005).

1.2.3 Gingerol and Shogaol

Gingerols are biologically active constituent in fresh ginger which give ginger pungent principles. Under ambient environment, gingerols are normally found as yellow oil or low melting crystalline form solid. The content of gingerols varies significantly with ginger varieties and cultivating locations. Gingerols are thermally labile due to the presence of a β-hydroxy keto group in the structure, and produce corresponding shogaols via a dehydration reaction (Bhattarai, 2001). The dehydration process will be affected by environmental temperature, pH and reacting time. It is reported that raising the reaction temperature and extending time significantly improved the conversion of 6-gingerol to 6-shogaol (Kou, Xingran et al., 2017) Figure 1 presents the structure of gingerols and shogaols. Figure 1. Structure of major gingerol and shogaol.

6 Gingerols may make a contribution towards health effects and medical applications of ginger. Previous research has shown that 6-gingerol has antitumor promotional activity effects （Park, Kwang-Kyun, et al., 1998), the analgesic and anti-inflammatory effects (Young, Haw-Yaw et al., 2005), and 10-gingerol and 12-gingerol has antibacterial activity against periodontal bacteria (Park et al., 2008). 6-gingerol has been studied by more researches compared to other gingerols such as 8-gingerol and 10 gingerol. This is because fresh ginger contains the largest 6-gingerol proportion of the total gingerol amount, while 8-gingerol, 10-gingerol and other gingerols with different chain lengths account for a small amount.

1.2.4 Medicinal use of Ginger

Ginger has undergone considerable research compared with other herbal drugs. In many Asian countries, especially China, India and Japan, ginger is treated as one of the additives in traditional medicine, but not a necessary herb. In ancient time ginger was considered as a kind of herbal medicine with strong medicinal benefits, rather than for its cooking use. Even today in western countries, ginger is still one of the most popular components in pharmaceutical preparations (Xizhen et al., 2004).

Anti-oxidant activity

Gingerol related compounds and diarylheptanoids are considered as playing a major role in the antioxidant properties of ginger. Studies show that gingerol related compounds extract isolated by dichloromethane have a pronounced kinetic behaviour on 2, 2-diphenyl-1-picryl-hydrazyl-hydrate (DPPH) radical scavenging reaction and inhibiting oxidation of liposome induced by 2′-azobis (2-amidinopropane) hydrochloride (AAPH) (Masuda, et al, 2004)

Anti-inflammatory activity

Studies showed that methanol extract of ginger has an inhibition effect on rat paw and skin edema induced by carrageenan and 5-hydroxytryptamine (5-HT), but has no effects on rat skin edema induced by substance P (ARG-PRO-LYS-Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met-[NH.sub.2]) or bradykinin 7 (Arg-Pro-Pro- Gly-Phe-Ser-Pro-Phe-Arg). Report indicated that the inhibition mechanism might be due to antagonism of the serotonin receptor (Penna, et al., 2003).

Anti-microbial activity

Ginger extract have an effect on inhibiting microbial growth. From the studies, ginger hydroalcoholic extract had a dose-dependent anti-microbial activity against pseudomonas aeruginosa, Salmonella typhimurium, Escherichia coli and Candida albicans (Jagetia, Ganesh Chandra, et al., 2003). And Ginger ethanol extract showed a strong antifungal activity against a wide range of fungi, including C. albicans, T. mentagrophytes, and Rhizopus sp, which was not inhibited by berberine (a weak antimicrobial isolated from herb) and ketoconazole (a synthetic imidazole antifungal drug) (Ficker, Christine E., et al., 2003).

Anti-neoplastic Activity

Ginger ethanol extract was reported to significantly reduce the rate of ovarian cancer cell proliferation, inhibited nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) in ovarian cancer cells, and prevented growth of a tumour from secreting Interleukin 8 (a chemokine produced by macrophages) and vascular endothelial growth factor (VEGF) in ovarian cancer cells.

Anti-angiogenic Activity

It was reported that 6-gingerol has anti-angiogenic activity in vitro and in vivo. 6-gingerol was shown to inhibit proliferation of human body endothelial cells induced by both VEGF and basic fibroblast growth factor (bFGF) in vitro, which caused cell cycle arrest in growth phase (G1 phase). In vivo, 6-gingerol blocked capillary-like tube formation by endothelial cells in response to VEGF, and inhibited sprouting of endothelial cells in the rat aorta and mouse cornea neovascularisation in response to VEGF (Kim, Eok-Cheon et al., 2005).

Hypoglycaemic activity

Some researches demonstrated that the juice of ginger could prevent hyperglycaemia and hypoinsulinaemia induced by 5-HT (1mg/kg) in normglycaemic rats. On the other

8 hand, streptozotocin (STZ) - induced type I diabetes produced a significant increase in fasting glucose levels in rats and a decrease in insulin levels in rats, which can be prevented by the treatment of ginger juice (Akhani, Sanjay et al., 2004).

1.2.5 Pharmacological use of Ginger

In spite of the wide use of ginger from ancient time till today, the curing mechanism and health efficiency have not been well verified by adequate clinic trials. So far, a huge number of science test have been carried out to deepen public understanding of ginger. Some of the findings are in accord with the previously stated effects in ancient medicine, while some are introducing more potential positive effects on specific diseases. According to these studies published, several kinds of ingredients have shown good potential, including 6-gingerol, 8-gingerol, 10-gingerol, 6-shogaol, 6-paradol, 6-gingerdiol, gingerenone A, zingerone, diarylheptanoids, hexahydrocurcumin and its derivatives, as well as other components. Among these ingredients, gingerols and shogaols are considered as the most effective compounds for the medicinal properties of ginger since these compounds displayed an efficient antioxidant activity (Habsah, M., et al., 2000). Furthermore, ginger are reported to have benefits to improve treatments of disease. From animal tests, some conclusions have been obtained that ginger may be valuable against diabetic complications in human and have a positive effect on lowering blood pressure. Some studies showed that ginger would largely change the levels of prostaglandin E2 in the serum of rats. This indicated that ginger may could be used as an anti-inflammatory and anti-thrombotic medicine. Other researchers indicated that gingerols have potential against Helicobacter pylori (which can cause peptic ulcer disease, gastric and colon cancer). Ginger is also reported to inhibit prostaglandin synthesis as gingerols are very active in inhibiting both prostaglandins and leukotrienes in red blood cells. Fresh and dried ginger rhizomes are generally used for various clinical purposes in ancient Asian countries. Fresh ginger has been used as an antitussive, antiemetic or expectorant, as well as to induce perspiration and dispel colds, while dried ginger is normally used for treating vomiting, stomach-ache, diarrhea accompanied by cold

9 extremities and faint pulse (Kimura, Pancho & Tsuneki， 2005). Despite the pressure from the development of modern medicines, some traditional ginger-containing medicines still perform effectively and competitively (Pakrashi & Pakrashi, 2003).

1.3 Drying Theory

1.3.1 Purpose of Drying

Drying is a mass transfer process for removing water from a solid, semi-solid or liquid through evaporating. In the food industry, drying is commonly applied in order to to lower the water activity and prevent microorganisms from multiplying, improve food stability and extend shelf life. In addition, drying products may significantly lower the cost of transport, storage and packaging (Fellows, 2009), which makes dried products more competitive on the market. Food products may be dried for other reasons, for example controlling food texture (Driscoll, 2004) and making food easier for consumption (Singh & Heldman, 2001). Given different raw products have different physical characters, studying drying theory and drying methods is vital for food trade and manufacture. Till today, drying is still the most useful, common and economic method of food preservation (Chen & Mujumdar, 2008).

1.3.2 Drying of Ginger

Dried ginger and dried ginger products account for the largest amount of ginger consumption around the globe, as fresh ginger is mainly produced in Asian countries and dried ginger significantly lowers the cost of transporting and storing. Generally, dried ginger is produced from fresh mature ginger rhizomes whereas immature ginger rhizomes are processed to make preserved ginger, as the mature rhizome has stronger flavour and aroma. Fresh ginger is normally dried in the sun. However, in unfavourable weather when natural drying cannot reach the required drying potential, a mechanical drying system is necessary. Under this circumstance, researchers have studied how mechanical drying temperature, time and other factors could affect the quality of dried ginger. Richardson,

10 K. C. ’s (1966) study showed that 57℃ is the maximum temperature for drying ginger as higher temperature darkens the colour and makes products undesirable for consumption (Richardson, 1966). Studies also have reported that the moisture level of dried ginger must be between 8 and 10 percent to avoid infestation by storage pests. To give a smooth finish to the dried rhizome, peeled and cleaned raw rhizomes are soaked in water, and then in 2 percent limewater for 6 hours before drying (Kannan & Nair, 1965). Besides, a series of other ginger drying researches have been conducted in multiple fields, for example studying the effect of oven drying, microwave drying, and silica gel drying (use silica gel to absorb water molecule of sample in a bottle at room temperature) on the volatile components of ginger. The result showed that microwave and silica gel drying can maintain the taste and appearance of fresh ginger (Huang, Baokang et al., 2012). The quality and oil/oleoresin content of ginger was compared with four treatments: whole-unpeeled, split-unpeeled, whole-peeled and split-peeled and four drying methods: sun, solar, natural air and fire-heat drying. Analysis shows that the highest yield of oil/oleoresin contents was 2.0% from the whole-unpeeled samples dried by fire-heat drying (Yiljep, Fum & Ajisegiri, 2005). The effects of sun drying, solar tunnel drying and cabinet tray drying on the quality of ginger were also studied by Jayashree, E et al (2014). Results indicated that when drying whole ginger, sun drying and solar tunnel drying retained the highest essential oil (13.9 mg/g) and oleoresin content (45.2 mg/g) of dry ginger. (Ayashree, Visvanathan, & John Zachariah, 2014).

1.3.3 Thin Layer Drying

Thin layer drying is one of the most widely used drying processes for agricultural products. In the food industry, building a drying model to predict and control the quality of the product is essential. Thin layer drying model is the simplest method and the basics for in-depth study of drying process and the drying characteristics. Thin layer drying models are classified generally into three groups: (1) pure empirical equations, (2) theoretical equations, (3) semi-theoretical/semi-empirical equations (Ghazanfari et al. 2006b). While theoretical equations normally take into account only the internal

11 resistance to moisture transfer within products, the other two groups consider only external resistance to moisture transfer between product and drying air (Akpinar, 2006). Drying of agricultural products is a complex process, affected by many factors and a precise mathematical model to describe the entire drying process is unrealistic. Thus, in analysing the drying model, a combination of theoretical and empirical models are often used, where the empirical equation are visual and easy to apply while the theoretical equation provide physical reality. The main justification for the empirical approach is the satisfactory fit to the experimental data (Ghazanfari et al, 2006), but is limited to the data rang of the experiment. Empirical models are obtained after conducting large amounts of experiments, and normally they draw from a direct relationship between average moisture content and the drying time. In this case, the parameters in the model have no physical meaning, and the fundamental mechanisms of the drying process are neglected. Thus, they cannot offer a clear insight into the entire drying process. Equations such as the model of Wang and Singh show good agreement for the experiment only, but are then adopted by other authors (for example, Akpinar, 2006). The most widely used and investigated theoretical drying model is Fick’s second law of diffusion. So far, many food have been successfully studied in their drying process applying Fick’s second law for example rice (Ece and Cihan, 1993) and hazelnut (Demirtas, Cevdet, Teoman Ayhan, and Kamil Kaygusuz, 1998). Fick’s second law can be described as;

∂M ∂2M ∂2M ∂2M = D( + + ) (1) ∂t ∂X2 ∂y2 ∂z2 where M is the moisture content at a location within the object at time t, D is the diffusion coefficient (m2/s). The diffusion coefficient varies during the drying process in terms of drying temperature, moisture content, and changes of structure of materials.

However, in most drying cases, an average Deff is used to simplify analysis (Ghazanfari et al., 2006). Assuming that the resistance to moisture flow is uniformly distributed throughout the

12 interior of the homogeneous isotropic material, the diffusion coefficient D is independent of the local moisture content, and that the volume shrinkage is negligible, then we can write Fick’s second law as follow: ∂M ∂2M = D (2) ∂t ∂X2 Where M is the moisture content (kg water/kg dry solids), t is the time (s), x is the diffusion path (m), and D is the moisture dependent diffusivity (m2/s) (Akpinar, 2006). Crank (1979) reported that Eq.(2) can be used for different regularly shaped solids, such as rectangular, cylindrical and spherical, with the appropriate initial and boundary conditions. If a constant diffusion coefficient is given, Crank’s equation for describing slab geometry from Eq (2) can be expressed as: 2 2 8 ∞ 1 (2n+1) π Dt MR = ∑ exp⁡(− ) (3) π2 n=0 (2n+1)2 4L2 where MR is the fractional moisture ratio, L is the half thickness of the slice. Compared to pure empirical equations and theoretical equations, semi-theoretical equations offer a compromise between theory and ease of use (Fortes, Mauri, and Martin R. Okos, 1981). But these models are only valid within the drying temperature, relative humidity and moisture content range for which they were developed. They require small time compared to theoretical thin layer models and do not need assumptions of geometry of a typical food, its mass diffusivity and conductivity (Parry, J. L., 1985). Table 6 shows the some semi-theoretical thin layer drying models. Table 6 Semi-theoretical thin layer-drying models Model name Equations References n The Page model MR=exp (-kt ) Page (1949) cited in Bruce (1985) The modified Page model MR=exp (-kt)n Overhults et al. (1973) The Lewis model MR=exp (-kt) Bruce (1985) The Henderson and MR=Aexp (-kt) Henderson and Pabis (1961) Pabis model

The two-term model MR=Aexp (-k1t) +Bexp (-k2t) Henderson (1974)

One of the simplest semi-theoretical equations is the Lewis’ model, which is be 13 described as (Lewis, 1921): dM = −k(M − M ) (4) dt e where k is the drying constant (t-1), M is the moisture content (dry basis) at time t and

Me is the equilibrium moisture content (dry basis). The model expresses that drying rate is proportional to the difference between the moisture content of the product being dried and the equilibrium moisture content (Sawhney et al., 1999). Integrating Eq 4 it can be presented as:

M−Me MR = = Aexp(−kt) (5) M0−Me

Where A is dimensionless characteristic constant, and M0 is the initial moisture content of the material being dried. The two-compartment model is based on the first two terms of the general series solution to the analytical solution of Fick’s second law. This solution can be applied regardless of particle geometry and boundary conditions by changing the model constant to fit the data. The model still requires that diffusivity is constant (Madamba, Ponciano, Robert, and Ken, 1996). The two terms model can be written as:

MR = Aexp(−k1t) + Bexp(−k2t) (6)

Where A and B are dimensionless characteristic constants, k1 and k2 are the drying constant (t-1). All four model parameters are determined by regression of the data against the model of Eq 6, and then represent the best fit for the experimental range covered by the experiments. The four parameters are therefore dependent on the product being tested and the test conditions used. The Arrhenius equation can show the effects on the temperature to drying constant. Palipane, B., and Driscoll (1994) have reported this effect in detail and have concluded that irrespective of whether the moisture removal mechanism is desorption, vapour pressure, spreading pressure or by a vapour diffusion process or, more probably, a combination of the four, the drying rate k varies with temperature in the same manner as the Arrhenius-type relationship (Palipane, Keerthi and Robert, 1994):

14 h푖 k푖 = k0푖exp(− ) (7) RTa where ki is drying rate, k0i is constant and hi is activation energy (J), R is the gas constant and Ta is the product absolution temperature (K). Mohapatra et al (2005) has studied the thin layer drying characteristics of parboiled wheat and they found that Temperature dependence of the diffusivity was well documented by Arrhenius-type relationship (Mohapatra, Debabandya, and Srinivasa, 2005). A moisture sorption isotherm is the relationship between the moisture content of a product, its water activity and its temperature at equilibrium, and is an important concept in drying studies. There will be a unique desorption (drying) and absorption isotherm for each product. The desorption isotherm is generally used for determining the equilibrium moisture content (EMC). From Phoungchandang et al (2009)’s study, several different isotherm models have been concluded, including modified Henderson model, modified Oswin model, modified-Chung-Pfost model and modified Halsey model. From their report, modified Halsey model is the best in calculating⁡Me = 푓(RH,T), given the equation as (Phoungchandang, Nongsang, and Sanchai.2009):

− ln(RH) −1 ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡M = [ ]1.364812 (8) e exp(3.944330−0.002816푇) where Me is EMC, RH is relative humidity, T is temperature.

1.3.4 The Limitation of Existing Thin Layer Drying Models

Most thin layer drying models are established based on assumptions, which means that the model can only describe drying accurately if all assumptions are fulfilled. However, in real cases, it is hard to justify all the assumptions and the accuracy of the model is affected. One of the common assumptions is assuming drying conditions, for example temperature and relative humidity, are constant, which often is not true, especially in natural drying. Variable drying conditions may affect the model to some extent. Thus, developing new models under changing conditions is necessary.

15 1.3.5 Two Layer Diffusion Model for Changed Drying Condition

This new model is based on the diffusion theory, and was developed by Wu Zhe et al (2012)

Figure 2 Two layer samples in drying

Layer 2 Air flow

Layer 1

Figure 2 describes the mechanism of two layer drying. In this model, it is assumed that a product can be divided into two parts; an inner layer (layer 1) and an outer layer (layer 2). The arrows indicate moisture transfer within the two layers and with the environment. During the drying process, moisture in layer 1 moves towards layer 2. In layer 2 there are two moisture exchanges, one from layer 1 and the other evaporating to the environment. Note that this division into two layers is not dependent on the product conditions. To determine the boundary of two layers, a new parameter  is introduced in this new model. The definition of µ is the mass ratio of the solid weights of the two layers. It can be expressed as: ms2 μ = (9) ms1 where ms1 is solid weight of layer 1 and ms2 is solid weight of layer 2. The value of  was determined experimentally by regression of the data against the model. A larger value of  represents a thin or small product where surface evaporation

16 is dominant, but a small value of  represents a thick product where moisture diffusion dominates the drying rate. Based on mass balance:

mw1+mw2 M1ms1+M2ms2 Mavg = = (10) ms1+ms2 ms1+ms2

Where Mavg is average moisture content of the sample, mw1 is mass of water of layer 1. mw2 mass of water of layer 1. M1 is the moisture content of layer 1 and M2 is the moisture of layer 2. Therefore:

(M1+μM2) M = ⁡ (11) avg 1+μ where Mavg is average moisture content of the sample, M1 is the moisture content of layer 1 and M2 is the moisture of layer 2. The moisture diffusion between the layers and surface evaporation can be described as follow: For layer 1:

dM1 = −k μ(M − M ) (12) dt 1 1 2 For layer 2:

dM2 = −k (M − M ) (13) dt 1 2 1

dM2 ⁡ = −k (M − M ) (14) dt 2 2 e These indicate there is only one form of moisture movement in layer 1, but two forms in layer 2. Consisting Eq 13 and Eq 14, total moisture movement for layer 2 can be expressed as:

dM2 ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡ = −k (M − M ) − k (M − M ) (15) dt 1 2 1 2 2 e From Eq 10:

17 (M +μM ) d 1 2 dMavg 1+μ 1 dM1 μ dM2 ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡ = = + (16) dt dt 1+μ dt 1+μ dt

After some mathematic of deduction from the equations above, the total drying rate is:

dMavg k2μ(M2−Me) = − (17) dt 1+μ

1.4 Separation and Identification Methods

1.4.1 Chemical Separation Technology

Chemical separation is a technology that is used to convert a mixture of chemical substances into two or more distinct product mixtures (Wilson, Adlard, Edward， Cooke, Michael, et al., 2000), in which one or more compounds are enriched after the process. The purpose of the separation process is often for analysis, identification and quantification of one or more elements in the mixture. Currently, there are many chemical separation methods, for example distillation, crystallization, extraction, chromatography and electrochemical separation.

1.4.2 Chromatography

Chromatography is a basic lab technology to separate chemical mixture. Since the 20th century, chromatography has significantly developed and now is used in various area; for example the petrochemical industry, organic synthesis, environmental protection and even space exploration. Chromatography can be separated into gas-solid chromatography (GSC), gas-liquid chromatography (GLC), liquid-solid chromatography (LSC) and liquid-liquid chromatography (LLC), based on the choice of stationary phase and mobile phase. The technology can also be classified as adsorption, partition, ion, size exclusion and affinity chromatography, based on the different physicochemical properties used by each method.

18 1.4.3 Chromatography-Mass Spectrometry Method

Chromatography-mass spectrometry (GC-MS) is a combination of gas chromatography and mass spectrometry. This technology takes advantages of the efficient separation and quantification of chromatography, and the accurate identification of mass spectrometry, in order to precisely separate, identify and quantify one or more compounds from a mixture. GC-MS has been used in many different fields. For example Daferera and Dimitra (2000) have reported analysing essential oils from seven air-dried plant species by using GC-MS. It was found that Origanum majorana (marjoram) oil consisted of hydrocarbons (42.1%), alcohols (24.3%) and and phenols (14.2%). Ethers predominated in Rosmarinus officinalis (rosemary) and Salvia fruticosa (sage) essential oils, constituting 88.9 and 78.0%, respectively (Daferera, Dimitra J., Basil N. Ziogas, and Moschos G. Polissiou, 2000). For ginger studying, Nirmala Menon’s (2007) report shows GC-MS has been using for analysing the volatile oil from fresh and sun dried ginger. In the study, geranial (24.2%) and zingerone (14.2%) in fresh ginger were reported to decrease after drying. Compared to fresh ginger, dried ginger contains a greater hydrocarbon content in the oil and less oxygenated compounds (Nirmala Menon et al., 2007). Other researchers combined supercritical fluid extraction with GC-MS to analyse volatile constitute from both fresh and dried ginger. They identified 62 volatile ingredients in fresh ginger and concluded that the major effects of the drying process were a reduction in gingerol content, an increase in terpene hydrocarbons and the conversion of some monoterpene alcohols to their corresponding acetates.

1.4.4 High-Performance Liquid Chromatography Method

High-performance liquid chromatography (HPLC) is one analytical method used to separate, identify and quantify each compound in a mixture. In this method, a pump is used to pressurize liquid mobile phases containing sample injection in a specific column. As each component has a different retention time in the column and different absorption wavelength, ingredients could be identified and quantified by these specific measurements from the detector.

19 HPLC has been widely used in many areas, including natural science studying, medical research, product detecting as well as manufacturing (Gerber, Krummen, Potgeter, Roth, Siffrin, Spoendlin, 2004). In the food industry, HPLC method has been used in many studies, such as detecting gingerol from ginger (Hiserodt, Franzblau, and Rosen, 1998), isolating Cis‐Trans Carotene Isomers from fruits (Chandler, and Schwartz, 1987), and determining selenium content in nut (Kannamkumarath, Sasi et al., 2002). Chen et al (1986) have reported using HPLC to analyse pungent gingerol compounds of green (4-5 months) and dry (8-9 months) ginger extracted by acetone. The total gingerol content (6-, 8- and 10-gingerol) of green ginger varied from 0.65-0.88% (w/w) and that of drying ginger ranged from 1.10 – 1.56% (w/w). In another report, Schwertner and Rios (2007) utilized HPLC to analyse 6-gingerol, 6-shogaol, 8-gingerol and 10-gingerl in various ginger-containing dietary supplements, spices, teas, mints and beverages. They developed a HPLC method for analysing 6-, 8-, and 10-gingerol in ginger-containing dietary supplements, spices, food products and beverages. Compared with other separation methods, for example solid phase extraction (SPE), gas chromatography (GC) and thin layer chromatography (TLC), HPLC is extremely fast, sensitive and efficient. Because it applied a pump to push the mobile phase through the column instead of gravity to make separation occur, HPLC takes much less time than other methods. Generally each run can be completed between 10 to 25 minutes. The detecting result is highly reliable. Finally HPLC is largely automated, so basic HPLC operation and analysis can be performed with minimal training (Kazakevich and LoBrutto 2007). HPLC is versatile and precise when it comes to identifying and quantifying chemical compounds from a mixture. But an HPLC instrument is quite costly compared with other methods. The high price and need for large quantities of expensive organic solvents for operation makes HPLC analysis an uneconomic approach for some research institutions. Also despite being relatively easy to repeat an existing HPLC method, it can be quite complex to develop a new analysis method, with many problems such as different modules, columns and mobile phases (Proteins: Biochemistry and Biotechnology, Walsh, 2002). 20 Chapter 2 Materials and Methods

2.1 Raw materials, Chemicals and Equipment used

The raw fresh ginger of Australian origin were obtained from Buderim Ginger PTY LTD. 6-Gingerol (≫ 98% purity) standard was obtained from Sigma-Aldrich. Methanol was of HPLC grade (purchased from Burdick & Jackson). Water for HPLC analysis was purified with a Milli-Q water system. Agilent vials for HPLC with caps were used. Whatman filter paper: pore size 0.45 μm.

Figure 3 Schematic diagram of thin layer dryer

Figure 3 showed the thin layer dryer used in this study. Fresh air enters the system and is mixed with recycled air before being accelerated by a fan. After setting the drying conditions on the control panel, the temperature of the air was adjusted by a heater, and the air humidity was adjusted by addition of steam. Two sensors were installed in the drying chamber for measuring the air conditions. A balance was used to weight the sample mass and computers were used to record mass changes and air conditions during the drying process.

2.2 Drying Preparation

Fresh ginger roots were stored in a refrigerator at -18 °C. Frozen ginger was defrosted

21 at 4°C for 24h, then placed at ambient temperature for 12h before use. Defrosted ginger bulk was peeled and sliced to 5 mm thickness by using slicer. About 10g sample slices were weighted and placed in an aluminium container with lid for measuring initial moisture content. Three sets of sample were measured for duplicates. The container was placed into the oven at 105°C for 24h with the lids placed below the container. Initial moisture content of fresh sample was obtained by following equation:

m2−m1 ⁡M = × 100% (18) m3−m1 where M is moisture content (wet bulb). m1 is container weight, m2 is The total weight of container and samples after drying, m3 is the total weight of container and samples before drying.

2.3 Drying Procedure

1. The chamber of the dryer was closed and heating system and steam supply was switched on. Air conditions in the chamber must reached the set conditions before started drying. As air flow in the chamber could affected the accuracy of the balance installed in thin layer dryer, fresh sample was weighted on an individual balance before being dried for adjusting experiment data.

2. The fresh sample was loaded upon the tray as quickly as possible in order to reduce the influence of ambient environment to the air conditions in the chamber. Weight changes and drying conditions was monitored and recorded by a computer connected with the dryer. Drying conditions were changed when about half of the sample weight had been removed in changing conditions runs and the time point of change was marked.

3. After the weight of the sample became steady the drying should be considered finished. Dried sample was weighted on an individual balance and packed in a sealing bag then vacuumed the bag and labelled. About 10g of dried sample was weighted for

22 measuring moisture content of dried sample by using the same method as measuring initial moisture content. The rest sample was placed in to a desiccator for measuring 6-gingerol content.

2.4 Drying Conditions

Steam was supplied during the drying process to adjust the humidity in the dryer during drying. The exact treatments were as follow: Table 7 Summary of drying conditions

RUN Temperature Relative humidity (RH) (°C) (%) 1 40 30 2 50 20 3 60 10 4 60 10 50 20 5 60 10 50 30 6 60 10 50 40 7 60 10 40 30 8 60 10 30 40 9 50 30 40 30 10 50 20 60 10 11 40 30 50 20 From Run 4 to Run 11, the drying conditions changed from the point of 1st row to that of 2nd row when half weight of the sample had been removed

Changed drying conditions was applied in this study. The reason of this design is because drying temperature could significantly impact drying process, while in practice the undeniable changes, such as weather change, may affect the drying conditions in the dryer. Therefore, drying condition was manually changed for improving reliability and

23 flexibility of experiment.

2.5 Extract Preparation

1. Dried ginger slices were pulverized using a grinder and passed through a 40 mesh (0.42 mm) sieve before extraction. However, the fresh ginger was ground up with a mortar and pestle prior to extraction.

2. 1g of ginger powder or paste was dissolved in 25 mL HPLC-grade methanol and sonicated for 30 min. The mixtures were centrifuged at 10,000 rpm for 10 min and supernatant was filtered through Whatman filter paper. Then it was diluted with water until the final constitute is 10% methanol and 90% water. Extracts of ginger were transferred to an Agilent vials and capped. All the extracts were kept at 4 °C

3. For 6-Gingerol calibration curve, HPLC-grade methanol was exactly measured and transferred to 6-gingerol standard vial to produce a stock solution of 5.0 mg/mL. Serial standard dilutions were made from the stock solution by dilution with 10% methanol and 90% water. The working standards were prepared containing 5.0, 10.0, 20.0, 40.0, 60.0 and 80.0 μg/mL, respectively. All 6-gingerol standards were capped and stored at 4◦C until used.

2.6 Instrumentation and Chromatographic Conditions

The HPLC system used in this study was manufactured by Shimadzu, model Prominence LC-20AD. The separation of the extract was conducted in a C18 column (Xterra), 3.5 μm, and 2.1×150mm. Water (A) and methanol (B) constituted the mobile phase which was used for separation. The following linear gradient was used: 0-5 min, 50% B; 5-10 min, 50-60% B; 10-15min, 60% B; 15-25min, 60-80% B; 25-30 min, 80% B; 30-35min, 80-50% B; 35-50min, 50% B. The injection volume was 20 μL and the flow rate was 0.2 mL/min. The detection wavelength of the UV detector (0~1000 nm) was set at 281 nm and the column temperature was maintained at 30 °C.

24 2.7 Data Analysis Procedure

The parameters for the two layer diffusion model were fitted to the experimental data (see Eqs (9), (10), (11), (12), (13), (14), (15), (16) and (17)) by using the Excel 2010 Solver. Firstly each drying run was individually fitted to the model in order to test for the significance and dependence of the model parameters. Secondly, the model was fitted simultaneously to all of the drying runs, and this was called global fitting. The resulting model parameter values were recorded for both individual and global fitting (see tables 9 and 10). To analyse the 6-gingerol content, the average 6-gingerol content was obtained from three duplicates. Coefficient of variation (C.V.) was used to estimate the precision of the experiment. Analysis of variance (ANOVA) was processed using SPSS Statistics V22.0 to show the significance level of 6-gingerol between different drying treatment. To both the drying model and the gingerol prediction model, the quality of fit of the model was evaluated using the coefficient of determination (R2) and root mean square error (RMSE)

25 Chapter 3 Results and Discussion

3.1 Individual Model Fitting

Table 8 shows the coefficient of determination and RMSE for each run. The coefficient of determination was in the range of 0.993 to 0.999, while RMSE changed between 0.09 and 0.18. The results show that the model is a good moisture predictor for individual fitting.

Table 8 Summary of coefficient of determination and RMSE for each runs Run Drying conditions Coefficient of determination RMSE (d.b.) 1 40 °C 30% RH 0.996 0.09 2 50 °C 20% RH 0.996 0.14 3 60 °C 10% RH 0.993 0.19 4 60 °C 10% RH to 50 °C 20% RH 0.998 0.07 5 60 °C 10% RH to 50 °C 30% RH 0.997 0.09 6 60 °C 20% RH to 50 °C 40% RH 0.996 0.17 7 60 °C 10% RH to 40 °C 30% RH 0.995 0.18 8 60 °C 10% RH to 30 °C 40% RH 0.999 0.09 9 50 °C 30% RH to 40 °C 30% RH 0.998 0.17 10 50 °C 20% RH to 60 °C 10% RH 0.997 0.18 11 40 °C 30% RH to 50 °C 20% RH 0.999 0.05

Table 9 shows the model constants k10, k20, h1, h2 and μ for different drying conditions.

The value of k10 in all runs was less than k20 and the value of μ was in the range of 3.45 to 2.58E+08, effectively showing that the inner layer 1, the core, was not significantly contributing to drying. As μ is the ratio of solid mass of the outer layer (layer 2) to the inner layer (layer 1), a greater value of μ means that the outer layer dominated the inner layer in the sample. The high variation in predicted h1 makes it hard to draw conclusions on the drying temperature effects on the drying rate of layer 1. In most runs, h2 was larger than 500J, indicating that drying temperature could significantly affect the drying rate of layer 2. Overall, the results show that the product is adequately be modelled by layer 2 alone. However, the individual sets of constants have little significance and consistency over the wide range of drying conditions. So after fitting the model to individual runs, the model was then fitted to all of the runs at once (global

26 fitting).

Table 9 Summary of constants for each runs

Run k10 k20 h1（J） h2 (J) μ 1 0.2987 812 1836 3411 2460 2 0.0487 30568 1E-06 4556 40.46 3 0.0129 201 3.25E-09 2922 156 4 1E-07 0.0260 0.0640 0.0012 33.26 5 0.0055 0.1258 7.49E-08 561.45 369 6 0.0006 0.0160 0 0 3620 7 0 0.1343 18167 596 2.58E+08 8 0.4490 8.05 4.80E-08 1909 3.45 9 0.0042 0.0150 0.9735 1E-06 481 10 0 1.72E+10 18617 8790 2.58E+08 11 64.24 2.77E+10 3.4E+06 8891 2.44E+06

3.2 Global Model Fitting

Global model fitting gives an overall test of the new model’s performance. Table 10 shows the constants obtained by global model fitting. From the result, μ is large, which indicates moisture transfer in the outer layer was dominate while water diffusion in the inner layer was too small to be calculated. In addition, layer 2 contains the overwhelming majority of the sample weight, which means the drying process sample dried as a whole, rather than as two separate layers with diffusion. This conclusion was consistent with the results from individual model fitting, and showed that the two layer diffusion model was not necessary for describing fresh ginger drying in thin layer dryer in this case.

27 Table 10 Summary of global fitting of two layer model Two layer diffusion Model constants

k10 0.2814

k20 148

h1 (J) 2240

h2 (J) 2875 μ 2313

Coefficient of determination 0.9807 RMSE (d.b.) 0.31

Table 11 shows the constants after applying the single layer model. The results show that the drying rate in a single layer model is very similar to the outer layer drying rate, which dominated the drying in the two layer diffusion model. This result suggests drying was governed by convection rather than diffusion in the study. The reason for this may be due to the shape of the sample, which were thin slices. The thickness of the sample significantly affects the relative importance of diffusion to convection. Comparing the coefficient of determination and RMSE of two models, it is obvious that the two layers model does not get improved compared to the single term model, which means that the single model would be adequate for this study.

Table 11 Summary of global fitting of single term model

Single layer model constants

k30 147

h3 (J) 2874

Coefficient of determination 0.9807

RMSE (d.b.) 0.31

3.3 HPLC Analysis of 6-gingerol Content and Predictive Modelling

3.3.1 6-gingerol Content

6-gingerol standard calibration curve was measured and plotted in each HPLC test. Figure 4 and Figure 5 shows a calibration curve of 6-gingerol content and a 10

28 microgramme 6-gingerol per milliliter of extracts (ug/mL) standard profile measured by HPLC in one experiment. The retention time of pure 6-gingerol was about 23.0 min. From this a linear plot was obtained: Y=65551X+16043 (19) where Y is retention area, X is 6-gingerol content. The results of the regression equation analysis indicates a good linear relationship between 6-gingerol concentration and peak area. The 6-gingerol concentration for the extracted sample was calculated by using the standard calibration curve.

Figure 4 Standard Calibration Curve

6000000 y = 65551x + 16043 5000000 R² = 0.9995

4000000

3000000

2000000 Retention Area Retention 1000000

0 0 10 20 30 40 50 60 70 80 90 6-gingerol concentration(ug/ml)

Figure 5 HPLC profile of 10 훍g/mL standard 6-gingerol solution

29 Ret.time Area Height NTP(USP) HETP(USP) resolution(USP) Area% 1.503 11285 364 26 5727.37 -- 0.422 2.178 25624 4063 2233 67.187 0.993 0.959 2.42 175482 19249 1796 83.528 1.169 6.566 9.206 2086 164 13235 11.333 24.748 0.078 23.039 2369463 109591 27536 5.447 31.6 88.658 26.606 23566 968 27851 5.386 5.98 0.882 39.311 55882 2293 59975 2.501 19.856 2.091 40.518 5458 295 100066 1.499 2.091 0.204 41.557 3738 280 202353 0.741 2.357 0.14 Total 2672583 137266 100

From the experiments, 6-gingerol content of fresh ginger from different ginger root was 0.59 ± 0.06 microgramme 6-gingerol per milligram of ginger of dry weight (ug/mg), which indicated 6-gingerol content was highly variable in fresh ginger. In addition, after measured three fresh ginger slices from one ginger rhizome, 6-gingerol content was 0.58ug/mg, 0.61ug/mg and 0.59ug/mg respectively. This indicated that using gingerol content measured from a small part of ginger as initial 6-gingerol content of the whole ginger would affect the accuracy of modelling. However, since fresh ginger cannot be homogenized and pulverized before drying process, and exact initial gingerol content was necessary for further analysis. A method which can measured accurate gingerol content and does not affect ginger drying is required. To solve this problem, 10g ginger slice was respectively obtained from each fresh ginger sample, then all of the collected slices were homogenized together in order to get an average initial 6-gingerol content used for all runs. After HPLC measurement, average 6-gingerol content was 0.607, C.V. was 4.03%

Table 12 and Table 13 shows the summary for three constant drying conditions and multiple comparisons. From the ANOVA test, the 6-gingerol content from the three runs was significantly different from the fresh sample, and also different between each other. Run 2 showed a higher 6-gingerol content than run 3. This was expected since the drying temperature in run 2 was lower. This could be because the much longer drying time in run 2 affected 6-gingerol decreasing. Run 1 had a milder drying temperature compared with run 2 but a lower 6-gingerol was obtained, this could due to the highly 30 variation of initial 6- gingerol.

Table 12 Characteristics of samples by different treatments with fixed temperature and humidity Run Drying conditions Drying time 6-gingerol content* C.V. min ug/mg % 1 40 °C 30% RH 314 0.337 9.52 2 50 °C 20% RH 293 0.456 4.57 3 60 °C 10% RH 154 0.444 3.43 * Average of three experiments. Table 13 ANOVA test between constant drying condition runs Run P value 1 to 2 0.000 1 to 3 0.000 2 to 3 0.360 ANOVA, P=0.05

Table 14 and Table 15 shows the summary for eight changed drying conditions and multiple comparisons. In general, trend in 6-gingerol content was observed: higher drying temperature and longer drying time resulted in lower gingerol content, which corresponds with the study of Bhattarai et al (2001) that higher temperature results in rapid and faster dehydration of 6-gingerol and forming the degradation product [6]-shogaol. It is clear that there is no significant difference between Run 4 and Run 10, Run 5 and Run 9, Run 7 and Run 8. Run 4 and Run 10 was dried in similar conditions but run 10 was 72mins longer, which indicates drying time had little did less impact on 6-gingerol decreasing. There was a significant difference between runs 4, 5 and 6, which were dried at the same temperature but different relative humidity and drying time. This showed that relative humidity could has impact on gingerol decreasing. No significant difference between Run 7 and Run 8 while Run 8 was dried in a lower average temperature and longer drying time. The reason could be due to the combined impact of drying temperature and drying time. The 6-gingerol content of Runs 5 and 9 both slightly decreased from the initial gingerol

31 content. Run 5 was dried in a short time compared with other runs, which could lead to high gingerol preservation. Run 9 was dried with high relative humidity in the whole drying process (30%), which could be a reason that high relative humidity prevent gingerol from decreasing. Similar situation happened in Run 6, it was dried with a long time, high temperature and high relative humidity, gingerol content was higher than other runs with shorter drying time and lower temperature such as Runs 7 and 8.

Table 14 Characteristics of samples subjected to different treatments with changing conditions Run Drying conditions Drying time 6-gingerolcontent* C.V. min ug/mg % 4 60 °C 10% RH to 50 °C 20% RH 207 0.349 3.53 5 60 °C 10% RH to 50 °C 30% RH 133 0.578 4.79 6 60 °C 20% RH to 50 °C 40% RH 283 0.448 2.79 7 60 °C 10% RH to 40 °C 30% RH 253 0.401 2.47 8 60 °C 10% RH to 30 °C 40% RH 275 0.403 4.56 9 50 °C 30% RH to 40 °C 30% RH 231 0.588 4.40 10 50 °C 20% RH to 60 °C 10% RH 279 0.356 9.08 11 40 °C 30% RH to 50 °C 20% RH 282 0.497 0.98 * Average of three experiments.

Table 15 ANOVA test between changed drying conditions runs Run P value Run P value 4 to 5 0.000 6 to 8 0.000 4 to 6 0.000 6 to 9 0.000 4 to 7 0.000 6 to 10 0.000 4 to 8 0.000 6 to 11 0.000 4 to 9 0.000 7 to 8 0.863 4 to 10 0.519 7 to 9 0.000 4 to 11 0.000 7 to 10 0.000 5 to 6 0.000 7 to 11 0.000 5 to 7 0.000 8 to 9 0.000 5 to 8 0.000 8 to 10 0.000 5 to 9 0.713 8 to 11 0.000 5 to 10 0.000 9 to 10 0.000 5 to 11 0.000 9 to 11 0.000 6 to 7 0.0 00 10 to 11 0.000 ANOVA, P=0.05

Table 16 shows ANOVA test between constant drying conditions and changed drying conditions. Four groups shows no significant difference, which are Run 1 and Run 4, 32 Run 1 and Run 10, Run 2 and Run 6, Run 3 and Run 6. Run 1 and Run 4 both obtained a lower gingerol yield while Run 4 was dried at a higher temperature and lower humidity. The reason for this could because of the much longer drying time of run 1 (107 mins longer). Similar situation happened in Run 1 and Run 10, while Run 1 was dried 35mins longer than Run 10. Run 2 and Run 6 had a close drying time. However, Run 6 was dried in a higher temperature and humidity. This result again suggests that humidity could affect 6-gingerol content in drying process. No significant difference between Run 3 and Run 6 shows drying with a higher temperature and less time and lower relative humidity could obtained a similar gingerol content in the run with lower temperature and longer drying time and higher humidity, which indicates that the interactions between drying temperature, time and relative humidity could have an apparent impact on 6-gingerol content.

Table 16 ANOVA test between constant conditions and changed conditions Run P value Run P value 1 to 4 0.483 2 to 8 0.006 1 to 5 0.000 2 to 9 0.000 1 to 6 0.000 2 to 10 0.000 1 to 7 0.001 2 to 11 0.027 1 to 8 0.001 3 to 4 0.000 1 to 9 0.000 3 to 5 0.000 1 to 10 0.301 3 to 6 0.790 1 to 11 0.000 3 to 7 0.023 2 to 4 0.000 3 to 8 0.028 2 to 5 0.000 3 to 9 0.000 2 to 6 0.662 3 to 10 0.000 2 to 7 0.005 3 to 11 0.005 ANOVA, P=0.05

3.3.2 6-gingerol Predicting Model

Drying time and drying temperature was considered as the two main factors affecting gingerol content (two factors model). The prediction model function of 6-gingerol in this case was chosen to be:

33 G = G0 exp⁡(−kG0t) (20) where G is final 6-gingerol content, G0 is initial 6-gingerol content. kG0 is 6-gingerol rate constant, t is drying time. Combine Eq.(20) and Eq.(7), gives

h G = G exp⁡(−k exp⁡( )t) (21) 0 G0 RT

Table 17 Summary of 6-gingerol content by sample from each drying conditions Run Stage Drying time 6-gingerol content Model min ug/mg ug/mg 1 - 282 0.336 0.402 2 - 293 0.454 0.413 3 - 154 0.444 0.479 4 1 35 2 172 0.348 0.463 5 1 32 2 101 0.578 0.510 6 1 53 2 231 0.448 0.419 7 1 33 2 220 0.401 0.436 8 1 54 2 219 0.403 0.423 9 1 60 2 170 0.587 0.448 10 1 49 2 230 0.356 0.421 11 1 77 2 204 0.494 0.420 "-" means constant drying conditions. 1 and 2 means before changed drying conditions and after changed drying conditions.

Table 18 Summary of two factors predicting model Model Temperature and Time Model

kG0 0.0013 h 0.0001 RMSE 0.0713 Coefficient of determination 0.4994

Table 17 and Table 18 shows the results of model used to predicting 6-gingerol content and its constants and coefficients. h was small, showing that drying temperature had no

34 effect on 6-gingerol preservation. It can be assumed that except for drying temperature and drying time, more factors could impact transformation of gingerol. The results suggested that the model result was not well fitted with experimental data. The possible reason for this could be: (1) Initial 6- gingerol content was variable. Additionally, as mentioned earlier, the high variation in the initial 6-gingerol content in fresh ginger chunks may have been the reason result for the unexpected gingerol content, especially the high 6-gingerol content in Runs 5 and 9. Despite the fact that the average 6-gingerol content was measured so as to eliminate this effect, the initial gingerol content could be highly variable between each fresh chunk, impacting the result to a large extent. (2) Relative humidity may affect 6-gingerol transforming process From the data analysis mentioned earlier, a higher RH could slow the reducing of 6-gingerol, which might increase the deviation between experimental data and model results as an unexplained variance. Therefore, RH was considered as a probable third factor that might affect the gingerol preservation for further modelling study. It was not possible to study the kinetics of 6-gingerol depletion further as with limited samples, only the initial and final levels of gingerol could be measured. Kinetic measurement of 6-gingerol content during drying would be beneficial for future studies, which could improve the result. In further study, the kinetics of gingerol content could be studied individually.

3.3.3 Humidity Effect on 6-gingerol Preservation

A new model was developed which incorporated relative humidity RH as a factor. This results in a three factor model (temperature, time and relative humidity) which was then applied to predict 6-gingerol content. The modified model used was:

h G = G exp⁡(−(k + A ∗ RH) exp ( ) t) (22) 0 G0 RT where A is a constant, RH is relative humidity. Eq(22) is based on Eq (21), and is the simplest modification to this model for including the effect of humidity on 6-gingerol depletion.

35 Figure 6 compared the results of the two factors model and three factors model by plotting experimental data vs model prediction. From the charts, the three factors model gave a better fit with the experimental data. Table 19 showed the constants and coefficients of the three factors model. The results showed that drying temperature did not significantly affect 6-gingerol content, which was consistent with the two factors model. Coefficient of determination improved about 10% comparing with previous model, and RMSE decreased about 0.0035. The result indicates that RH had an effect on rate of 6-gingerol decrease, higher relative humidity will prevent 6-gingerol from reducing. Similar results were obtained by Arabhosseini et al (2007), the essential oil content and hue value of dried tarragon leaves seem to be affected by the RH value of drying air, higher relative humidity lead to less degradation of essential oil and hue value.

Figure 6 Comparison of two factors model (left) and three factors model (right)

Table 19 Summary three factors predicting model Model Tem. Time and RH Model

KG0 0.0019 h 0.0003 A -2.43E-05 RMSE 0.0675 Coefficient of determination 0.5923

3.4 Optimum Drying Condition

Combining single term drying model and three factors model for predicting gingerol 36 content, an optimum drying condition which can get maximum 6-gingerol yield can be concluded. Four requirements had been set: (1) The change of average moisture content between two continuous drying data points (1min interval) is less than 0.15, which could be considered the sample reached its steady mass and drying is finished. (2) For saving cost in industry operating, the drying time should be less than 300 minutes. (3) Drying temperature should be higher than 30°C and lower than 70°C . (4) Relative humidity during drying should be higher than 0% and lower than 40%. The Result suggested optimum drying condition of ginger is: air temperature at 38°C ;

RH at 40%, drying time 298mins. .

37 Chapter 4 Conclusion

1 .Drying ginger slices was adequately described by a single layer model. The two layer model showed no improvement for describing drying in this study. 2. From the model constants, drying temperature ranging from 30°C to 60°C had little impact on 6-gingerol degradation. 3. The prediction model of 6-gingerol suggested that relative humidity ranged from 10% to 40% contributed to 6-gingerol decrease. From the result, a higher relative humidity would reduce 6-gingerol degradation. 4. The initial 6-gingerol content of fresh ginger was highly variable between each fresh ginger root. In further studies, fresh ginger rhizome could be divided equally into two pieces. One could be dried as a whole or slices and the other could be used to measure initial 6-gingerol content. Therefore, for each run, a unique initial 6-gingerol content could be available rather than an average content.

38 Bibliography

[1] Katie, P. (2011, March 16). Know Your Spice: A Brief History of Ginger. Retrieved from http://www.motherearthliving.com [2] Kizhakkayil, Jaleel, and Bhaskaran Sasikumar. "Diversity, characterization and utilization of ginger: a review." Plant Genetic Resources 9.03 (2011): 464-477. [3]Pakrashi, S.C. & Pakrashi, A. (2003b). Cultivation. Pakrashi, S.C. & Pakrashi, A. (Ed) Ginger: A Versatile Healing Herb, New Delhi: Vedams eBooks Pvt. Ltd.; 5-13 [4]Ravindran, P.N. & Babu, K.N. (2005). Introduction. Ravindran, P.N. & Babu, K.N. (Ed) Ginger : the genus Zingiber, CRC Press; 1-14 [5] Baker JG (1882) Scitaminae. In: Hooker JD (ed.) The Flora of British India. vol. VI. Dehra Dun: Bishen Singh, Mahendrapal Singh, pp. 198–264 [6]Essential Oils for Beginners: The Guide to Get Started with Essential Oils and Aromatherapy:,Press, A., 2013,Callisto Media Incorporated [7]Singh, Gurdip, et al. "Studies on essential oils, Part 42: chemical, antifungal, antioxidant and sprout suppressant studies on ginger essential oil and its oleoresin." Flavour and Fragrance Journal 20.1 (2005): 1-6. [8]Singh, Gurdip, et al. "Chemistry, antioxidant and antimicrobial investigations on essential oil and oleoresins of Zingiber officinale." Food and Chemical Toxicology 46.10 (2008): 3295-3302. [9]Geiger, James L. "The essential oil of ginger, Zingiber officinale, and anaesthesia." International Journal of Aromatherapy 15.1 (2005): 7-14. [10] S. Bhattarai, V. H. Tran and C. C. Duke, J. Pharm. Sci., 90, 1658 (2001)

[11] Kou, Xingran, et al. "Efficient dehydration of 6-gingerol to 6-shogaol catalyzed by an acidic ionic liquid under ultrasound irradiation." Food Chemistry 215 (2017): 193-199. [12] Park, Kwang-Kyun, et al. "Inhibitory effects of [6]-gingerol, a major pungent principle of ginger, on phorbol ester-induced inflammation, epidermal ornithine decarboxylase activity and skin tumor promotion in ICR mice."Cancer letters 129.2

39 (1998): 139-144. [13] Young, Haw-Yaw, et al. "Analgesic and anti-inflammatory activities of [6]-gingerol." Journal of ethnopharmacology 96.1 (2005): 207-210. [14] Park, Miri, Jungdon Bae, and Dae‐Sil Lee. "Antibacterial activity of [10]‐ gingerol and [12] ‐ gingerol isolated from ginger rhizome against periodontal bacteria." Phytotherapy Research 22.11 (2008): 1446-1449. [15] Xizhen, Ai, Song Jinfeng, and Xu Xia. "6 Ginger Production in Southeast Asia." Ginger: The Genus Zingiber (2004): 241. [16] Masuda, Yuki, et al. "Antioxidant properties of gingerol related compounds from ginger." Biofactors 21.1‐4 (2004): 293-296. [17]Penna, S. C., et al. "Anti-inflammatory effect of the hydralcoholic extract of Zingiber officinale rhizomes on rat paw and skin edema." Phytomedicine 10.5 (2003): 381-385. [18] Jagetia, Ganesh Chandra, et al. "Influence of ginger rhizome (Zingiber officinale Rosc) on survival, glutathione and lipid peroxidation in mice after whole-body exposure to gamma radiation." Radiation research 160.5 (2003): 584-592. [19]Ficker, Christine E., et al. "Inhibition of human pathogenic fungi by ethnobotanically selected plant extracts." Mycoses 46.1‐2 (2003): 29-37. [20]Kim, Eok-Cheon, et al. "[6]-Gingerol, a pungent ingredient of ginger, inhibits angiogenesis in vitro and in vivo." Biochemical and biophysical research communications 335.2 (2005): 300-308. [21] Akhani, Sanjay P., Santosh L. Vishwakarma, and Ramesh K. Goyal. "Anti‐ diabetic activity of Zingiber officinale in streptozotocin‐induced type I diabetic rats." Journal of pharmacy and Pharmacology 56.1 (2004): 101-105. [22]Habsah, M., et al. "Screening of Zingiberaceae extracts for antimicrobial and antioxidant activities." Journal of ethnopharmacology 72.3 (2000): 403-410. [23]Kimura, I., Pancho, L.R. & Tsuneki, H. (2005). Pharmacology of Ginger. Ravindran, P.N. & Babu, K.N. (Ed) Ginger : the genus Zingiber, CRC Press; 469-488 [24]Pakrashi, S.C. & Pakrashi, A. (2003a). Introduction. Pakrashi, S.C. & Pakrashi, A. (Ed) Ginger: A Versatile Healing Herb, New Delhi: Vedams eBooks Pvt. Ltd.; 1-4 40 [25] "drying - definition of drying by the Free Online Dictionary, Thesaurus and Encyclopedia". Farlex. Retrieved 23 April 2011. [26] Fellows, P. J. (2009). Dehydration. Fellows, P. J. (Ed) Food processing technology: Principles and practice Third edition, Woodhead Publishing Limited and CRC Press LLC; 481-524 [27]Driscoll, R. (2004). Food Dehydration. Smith, J.S. & Hui, Y. H. (Ed) Food Processing: Principles and Applications, United Kingdom: Blackwell Publishing Ltd.; 31-44 [28] Singh, R. & Heldman, D.R. (2001). Dehydration. Singh, R. & Heldman, D.R. (Ed) Introduction to food engineering- 3rd ed., Academic Press; 557-590 [29] Chen, X.D. & Mujumdar, A.S. (2008). Preface. Chen, X.D. & Mujumdar, A.S. (Ed)Drying Technologies in Food Processing, United Kingdom: Blackwell Publishing Ltd.; xiii [30] Richardson, K. C. "Effect of dehydration temperature on quality of ginger."Food Technology 18 (1966): 95. [31] Kannan, K., and K. P. V. Nair. "Zingiber officinale (ginger) in Kerala." Madra Agric. J 52.4 (1965): 168-176. [32] Huang, Baokang, et al. "Effect of oven drying, microwave drying, and silica gel drying methods on the volatile components of ginger (Zingiber officinale Roscoe) by HS-SPME-GC-MS." Drying Technology 30.3 (2012): 248-255. [33] Yiljep, Y. D., G. A. Fumen, and ES A. Ajisegiri. "The effects of peeling, splitting and drying on ginger quality and oil/oleoresin content." Agricultural Engineering International: CIGR Journal (2005). [34] Jayashree, E., R. Visvanathan, and John Zachariah. "Quality of dry ginger (Zingiber officinale) by different drying methods." Journal of food science and technology 51.11 (2014): 3190-3198. [35]Ghazanfari, A., et al. (2006b) Thin-Layer Drying of Flax Fiber: I. Analysis of Modeling Using Fick's Second Law of Diffusion. Drying Technology, 24(12): 1631-1635. [36]Akpinar, E. K. (2006). Determination of suitable thin layer drying curve model for 41 some vegetables and fruits. Journal of Food Engineering. 73:75–84 [37] Ghazanfari, A., et al. (2006a). Thin-Layer Drying of Flax Fiber: II. Modeling Drying40 Process Using Semi-Theoretical and Empirical Models. Drying Technology. 24(12): 1637-1642Ghazanfari, [38] Akpinar, Ebru Kavak. "Determination of suitable thin layer drying curve model for some vegetables and fruits." Journal of food engineering 73.1 (2006): 75-84. [39] Ece, M. C., and A. Cihan. "A liquid diffusion model for drying rough rice." Transactions of the ASAE (USA) (1993). [40]Demirtas, Cevdet, Teoman Ayhan, and Kamil Kaygusuz. "Drying behaviour of hazelnuts." Journal of the Science of Food and Agriculture 76.4 (1998): 559-564. [41] Ghazanfari, A., et al. (2006b) Thin-Layer Drying of Flax Fiber: I. Analysis of Modeling Using Fick's Second Law of Diffusion. Drying Technology, 24(12): 1631-1635. [42]Akpinar, E. K. (2006). Determination of suitable thin layer drying curve model for some vegetables and fruits. Journal of Food Engineering. 73:75–84 [43] Crank, John. The mathematics of diffusion. Oxford university press, 1979. [44]Fortes, Mauri, and Martin R. Okos. "Non-equilibrium thermodynamics approach to heat and mass transfer in corn kernels." Transactions of the ASAE 24.3 (1981): 761-0769. [45] Parry, J. L. "Mathematical modelling and computer simulation of heat and mass transfer in agricultural grain drying: A review." Journal of Agricultural Engineering Research 32.1 (1985): 1-29. [46] Lewis, W. Km. "The Rate of Drying of Solid Materials." Industrial & Engineering Chemistry 13.5 (1921): 427-432. [47] Sawhney, R. L., et al. "Determination of drying constants and their dependence on drying air parameters for thin layer onion drying." Drying technology 17.1-2 (1999): 299-315. [48] Madamba, Ponciano S., Robert H. Driscoll, and Ken A. Buckle. "The thin-layer drying characteristics of garlic slices." Journal of food engineering 29.1 (1996): 75-97. [49] Palipane, Keerthi B., and Robert H. Driscoll. "The thin-layer drying characteristics 42 of macadamia in-shell nuts and kernels." Journal of food engineering 23.2 (1994): 129-144. [50] Phoungchandang, S., S. Nongsang, and P. Sanchai. "The development of ginger drying using tray drying, heat pump–dehumidified drying, and mixed-mode solar drying." Drying Technology 27.10 (2009): 1123-1131. [51] Mohapatra, Debabandya, and P. Srinivasa Rao. "A thin layer drying model of parboiled wheat." Journal of food engineering 66.4 (2005): 513-518. [52] Wilson, Ian D.; Adlard, Edward R.; Cooke, Michael; et al., eds. (2000). Encyclopedia of separation science. San Diego: Academic Press. ISBN 978-0-12-226770-3. [53] Daferera, Dimitra J., Basil N. Ziogas, and Moschos G. Polissiou. "GC-MS analysis of essential oils from some Greek aromatic plants and their fungitoxicity on Penicillium digitatum." Journal of Agricultural and Food Chemistry 48.6 (2000): 2576-2581. [54] Nirmala Menon, A., et al. "Effects of processing on the flavor compounds of Indian fresh ginger (Zingiber officinale Rosc.)." The Journal of essential oil research 19.2 (2007): 105-109. [55]Gerber, F.; Krummen, M.; Potgeter, H.; Roth, A.; Siffrin, C.; Spoendlin, C. (2004). "Practical aspects of fast reversed-phase high-performance liquid chromatography using 3μm particle packed columns and monolithic columns in pharmaceutical development and production working under current good manufacturing practice". Journal of Chromatography A 1036 (2): 127–133. doi:10.1016/j.chroma.2004.02.056. PMID 15146913. [56]Hiserodt, R. D., S. G. Franzblau, and R. T. Rosen. "Isolation of 6-, 8-, and 10-Gingerol from Ginger Rhizome by HPLC and Preliminary Evaluation of Inhibition of Mycobacterium a vium and Mycobacterium t uberculosis." Journal of Agricultural and food Chemistry 46.7 (1998): 2504-2508. [57]Chandler, L. A., and S. J. Schwartz. "HPLC Separation of Cis‐Trans Carotene Isomers in Fresh and Processed Fruits and Vegetables." Journal of Food Science 52.3 (1987): 669-672 [58]Kannamkumarath, Sasi S., et al. "HPLC–ICP–MS determination of selenium 43 distribution and speciation in different types of nut." Analytical and bioanalytical chemistry 373.6 (2002): 454-460.

[59] CHEN, CHU‐CHIN, MAY‐CHIEN KUO, and CHI‐ TANG HO. "High performance liquid chromatographic determination of pungent gingerol compounds of ginger (Zingiber officinale Roscoe)." Journal of food science51.5 (1986): 1364-1365. [60] Schwertner, Harvey A., and Deborah C. Rios. "High-performance liquid chromatographic analysis of 6-gingerol, 8-gingerol, 10-gingerol, and 6-shogaol in ginger-containing dietary supplements, spices, teas, and beverages." Journal of Chromatography B 856.1 (2007): 41-47. [61]HPLC for Pharmaceutical Scientists, Kazakevich, Y.V. and LoBrutto, R.,2007,Wiley [62]Proteins: Biochemistry and Biotechnology,Walsh, G.,2002,Wiley [63] Bhattarai, Sushila, Van Hoan Tran, and Colin Charles Duke. "The stability of gingerol and shogaol in aqueous solutions." Journal of Pharmaceutical Sciences 90.10 (2001): 1658-1664. [64] Arabhosseini, A., et al. "Long-term effects of drying conditions on the essential oil and color of tarragon leaves during storage." Journal of Food Engineering 79.2 (2007): 561-566.

44 Appendix I Summary of drying sample for each treatment Run Initial weight Final weight Initial moisture content Final moisture content (g) (g) % (d.b.) % (d.b.) 1 63.52 14.39 379.38 12.77 2 87.12 14.06 572.29 10.80 3 66.85 11.29 517.48 6.94 4 57.30 10.86 489.79 18.14 5 84.10 17.40 442.76 13.17 6 82.61 14.78 531.22 11.65 7 73.28 10.32 750.84 18.68 8 85.53 13.16 678.35 19.98 9 86.78 16.95 484.90 13.17 10 83.40 8.48 1009.98 13.06 11 91.71 17.36 473.38 11.70

II Graphs for individual model fitting

Figure 6 The graph of model fitting for Run 1

400

350 Experiment 300 Model 250

200

150

100 Moisture Content (%) Content Moisture 50

0 0 50 100 150 200 250 300 350 -50 Time (min)

Figure 7 The graph of model fitting for Run 2

45 700 Model 600 Experiment 500

400

300

200 Moisture Content (%) ContentMoisture

100

0 0 50 100 150Time (min)200 250 300 350

Figure 8 The graph of model fitting for Run 3

600

Experiment 500 Model 400

300

200

Moisture Content (%) Content Moisture 100

0 0 50 100 150 200 -100 Time (min)

Figure 9 The graph of model fitting for Run 4

46 600

Experiment 500 Model

400

300

200 Moisture Content (%) Content Moisture

100

0 0 50 100 150 200 250 Time (min)

Figure 10 The graph of model fitting for Run 5

500 Model 450 Experiment 400

350

300

250

200

Moisture Content (%) ContentMoisture 150

100

50

0 0 20 40 60 80 100 120 140 Time (min)

Figure 11 The graph of model fitting for Run 6

47 600

500 Experiment

Model 400

300

200 Moisture Content (%) Content Moisture

100

0 0 50 100 150 200 250 300 Time (min)

Figure 12 The graph of model fitting for Run 7

800

700 Experiment 600 Model 500

400

300

200 Moisture Content (%) Content Moisture

100

0 0 50 100 150 200 250 300 -100 Time (min)

48 Figure 13 The graph of model fitting for Run 8

800

700 Experiment

600 Model

500

400

300 Moisture Content (%) ContentMoisture 200

100

0 0 50 100 150 200 250 300 Time (min)

Figure 14 The graph of model fitting for Run 9

600

Experiment 500 Model 400

300

200

Moisture Content (%) Content Moisture 100

0 0 50 100 150 200 250 -100 Time (min)

49 Figure 15 The graph of model fitting for Run 10

1200

Experiment 1000 Model 800

600

400

Moisture Content (%) ContentMoisture 200

0 0 50 100 150 200 250 300 -200 Time (min)

Figure 16 The graph of model fitting for Run 11

500

450 Experiment

400 Model 350

300

250

200

Moisture Content (%) Content Moisture 150

100

50

0 0 50 100 150 200 250 300 Time (min)

50 Ⅲ Summary and profile of HPLC results

Treatment No. Mass Area 6-gingerol content g ug/mg 1 1.008 208933 0.601 2 1.033 214549 0.639 Fresh 3 0.984 198104 0.608 4 1.041 203572 0.580 1 1.007 741700 0.306 40°C 30% RH 2 1.051 830089 0.335 3 1.055 926919 0.370 1 1.004 1043159 0.434 50°C 20% RH 2 1.021 1182482 0.475 3 1.039 1113310 0.459 1 1.045 1069659 0.428 60°C 10% RH 2 1.085 1111036 0.445 3 1.076 1145068 0.458 1 1.065 655712 0.336 60°C 10% RH to 2 1.096 709544 0.360 50°C 20% RH 3 1.097 691473 0.352 1 0.983 1385976 0.610 60°C 10% RH to 2 1.092 1303407 0.559 50°C 30% RH 3 0.976 1280994 0.565 1 1.072 1045299 0.438 60°C 20% RH to 2 0.976 1077608 0.462 50°C 40% RH 3 0.964 947191 0.445 1 0.924 743855 0.391 60°C 10% RH to 2 1.004 806950 0.411 40°C 30% RH 3 1.065 764358 0.402 1 0.927 928028 0.424 60°C 10% RH to 2 1.047 905328 0.392 30°C 40% RH 3 0.971 870539 0.393 1 0.95 1271660 0.617 50°C 30% RH to 2 0.929 1202725 0.580 40°C 30% RH 3 0.982 1191103 0.567 1 1.007 668484 0.318 50°C 20% RH to 2 0.939 769067 0.376 60°C 10% RH 3 1.013 776777 0.372 1 0.925 1016462 0.497 40°C 30% RH to 2 1.009 1030102 0.492 50°C 20% RH 3 1.071 1063125 0.502

51 40°C, 30% RH 50°C, 20% RH

60°C, 10% RH 60°C 10% RH to 50°C 20% RH

60°C 10% RH to 50°C 30% RH 60°C 20% RH to 50°C 40% RH

60°C 10% RH to 40°C 30% RH 60°C 10% RH to 30°C 40% RH

52 50°C 30% RH to 40°C 30% RH 50°C 20% RH to 60°C 10% RH

40°C 30% RH to 50°C 20% RH

53