Air drying of fucus vesiculosus seaweed and its effect on phytochemical characteristics of aqueous extracts

UNIVERSIDADE DE SANTIAGO DE COMPOSTELA ESCOLA TÉCNICA SUPERIOR DE ENXEÑARÍA Departamento de Enxeñaría Química AIR DRYING OF FUCUS VESICULOSUS SEAWEED AND ITS EFFECT ON PHYTOCHEMICAL CHARACTERISTICS OF AQUEOUS EXTRACTS Memoria presentada por: Santiago Gabriel Sexto Cancela Como Trabajo de Fin de Máster correspondiente al “Máster en Ingeniería Química y Bioprocesos” Santiago de Compostela, febrero de 2015 RAMÓN FELIPE MOREIRA MARTÍNEZ, Profesor Titular, y JORGE SINEIRO TORRES PÉREZ, Profesor Contratado Doctor, del Departamento de Enxeñaría Química de la Universidade de Santiago de Compostela, INFORMAN: Que la memoria titulada “Air drying of Fucus vesiculosus seaweed and its effect on phytochemical characteristics of aqueous extracts”, presentada por Santiago Gabriel Sexto Cancela para superar los créditos correspondientes al Proyecto de Fin de Máster correspondiente al Máster en Ingeniería Química y Bioprocesos, se realizó bajo nuestra supervisión en el Departamento de Enxeñaría Química de la Universidade de Santiago de Compostela y autorizan su presentación. Para que así conste a los efectos oportunos, expiden el presente informe en Santiago de Compostela, febrero de 2015. Los directores, El alumno R.F. Moreira Martínez J. Sineiro Torres S. G. Sexto Cancela Agradecimientos Al Grupo GI-1618 de investigación de Tecnologías para el Desarrollo de Bioproductos Industriales por permitirme formar parte de su equipo y desarrollar este Trabajo de Fin de Máster. A Ramón Moreira y Jorge Sineiro por la dirección y, sobre todo, la compañía y el apoyo aportados durante toda la duración del Trabajo de Fin de Máster. A Francisco Chenlo por la compañía, las correcciones y el buen criterio proporcionados. A Marivel Sánchez por su ayuda y paciencia en la introducción a las técnicas de laboratorio relacionadas con la extracción de compuestos bioactivos. A Santiago Arufe, excelente compañero y mejor amigo. Antón Taboada, gran amigo y compañero del Trabajo de fin de Máster. 1 Contents 1. Abstract.................................................................................................................. 4 2. Objectives .............................................................................................................. 6 3. Introduction ........................................................................................................... 7 3.1. Brown algae .................................................................................................... 9 3.1.1. Polysaccharides ....................................................................................... 9 3.1.2. Proteins .................................................................................................. 13 3.1.3. Lipids ..................................................................................................... 13 3.1.4. Minerals ................................................................................................. 13 3.1.5. Polyphenols ........................................................................................... 13 3.1.6. Carotenoids ............................................................................................ 15 3.1.7. Fucus vesiculosus .................................................................................. 16 3.2. Antioxidant activity ...................................................................................... 17 3.3. Drying fundamentals .................................................................................... 18 3.3.1. Moisture in foods ................................................................................... 20 3.3.2. Drying kinetics ...................................................................................... 21 3.4. Ultrasound assisted extraction ...................................................................... 22 3.4.1. Introduction ........................................................................................... 22 3.4.2. Principles of ultrasound assisted extraction .......................................... 23 4. Experimental........................................................................................................ 25 4.1. Materials ....................................................................................................... 25 4.1.1. Raw material .......................................................................................... 25 4.1.2. Reagents ................................................................................................ 25 4.2. Experimental techniques and equipment ...................................................... 25 2 4.2.1. Physical and chemical characterization of seaweed .............................. 25 4.2.2. Water sorption isotherms ....................................................................... 26 4.2.3. Determination of drying kinetics ........................................................... 29 4.2.4. Milling ................................................................................................... 34 4.2.5. Extracts obtention and characterization ................................................. 36 4.2.6. Statistical analysis ................................................................................. 39 5. Assayed systems .................................................................................................. 40 6. Results and discussion ......................................................................................... 41 6.1. Determination of water sorption isotherms .................................................. 41 6.2. Determination of drying kinetics .................................................................. 51 6.2.1. Deep-bed configuration drying kinetics ................................................ 51 6.2.2. Thin layer configuration drying kinetics ............................................... 57 6.2.3. Effect of load density on drying kinetics ............................................... 65 6.2.4. Colorimetric characterization ................................................................ 67 6.3. Physical characterization of milled seaweed ................................................ 68 6.3.1. Granulometric characterization ............................................................. 69 6.3.2. Colorimetric characterization ................................................................ 71 6.4. Extracts characterization .............................................................................. 78 6.4.1. Effect of changes in extraction operation conditions ............................ 78 6.4.2. Effect of drying temperature ................................................................. 81 6.4.3. Fractions screening ................................................................................ 86 7. Conclusions ......................................................................................................... 88 8. Bibliography ........................................................................................................ 90 9. Annex I: Calibration lines ................................................................................... 97 3 9.1. Total polyphenolic content ........................................................................... 97 9.2. Total carbohydrates content ......................................................................... 97 9.3. Uronic acids content ..................................................................................... 97 10. Annex II: Gantt diagram .................................................................................... 98 11. Annex III: List of figures ................................................................................... 99 12. Annex IV: List of tables .................................................................................. 101 4 1. Abstract The purpose of this study was to determine the effect of air drying at four different temperatures (35, 50, 60 and 75ºC) on drying kinetics at different configurations (deep-bed and thin layer) and properties of dried Fucus vesiculosus seaweed and the phytochemical constituents of its aqueous extracts. Water adsorption and desorption isotherms were also determined at four different temperatures (5, 25, 45 and 65ºC). Different models (GAB, BET, Caurie, Halsey and Oswin) were tested for water sorption isotherms modelling. Halsey model showed the best fit for all water sorption isotherms. In the case of drying kinetics modelling for deep-bed configuration, bibliographic models were tested (Newton, Henderson-Pabis, Logarithmic, Two-Term, Weibull, Page and Modified Page). Page and modified Page models were selected as the most adequate under statistical analysis criteria. Thin layer dyring kinetics of F. vesiculosus showed a constant rate period and falling rate period at all drying temperatures tested. Critical moisture content increased with drying temperature. For post-critical period, Fick’s second law equations were successfully applied and allowed the determination of effective coefficients of diffusion of water, which increased with drying temperature (from 95·10-12 up to 260·10-12 m2/s). This water effective diffusivities were estimated considering the shrinkage of samples. The shrinkage analysis indicated that real shrinkage was larger than ideal shrinkage (evaluated by means of water loss). Effect of different load densities (from 1.25 up to 14.88 kg/m2) was studied at 75ºC and a linear correlation between load density and drying time was achieved. Dried seaweed at 75ºC exhibited the largest colour differences. Seaweed powders previously dried at 35, 60 and 75ºC exhibited an appreciable yellow-tone coloration, whereas system dried at 50ºC presented a greenish coloration. F. vesiculosus extracts were obtained performing aqueous ultrasound assisted extractions. Two key extraction parameters (from 20 to 40 (g/g) liquid/solid ratio and contact time from 4 to 20 minutes) on the extraction yield of three bioactive compounds (polyphenols, carbohydrates and alginates) were studied for the system dried at 35ºC. The effect of drying temperature on antioxidant activity and the yield of bioactive compounds was studied by means of extractions performed at optimized L/S ratio of 30 (g/g) and contact Abstract 5 times of 4 minutes for all dried systems. Maximum phenolic content (1738 mg PHL/100 gdw) and antioxidant activity (measured by means of DPPH radical scavenging activity, 57%) were obtained for the system dried at 35ºC and declined significantly with drying temperature. Higher drying temperatures had a positive effect on alginate extraction yield (2069 mg GLU/100 gdw). Total phenolic content of extracts from different particle size fractions of dried F. vesiculosus powder was determined. Maximum phenolic content was attained at intermediate fractions 80-200 µm (2272 GLU/100 gdw). 6 2. Objectives The main objective of this Master Thesis ‘Air drying of Fucus vesiculosus seaweed: effect on phytochemical characteristics of aqueous extracts’ is to study the effect of convective air drying temperatures on phytochemical constituents of the aqueous extracts of F. vesiculosus seaweed. Specific objectives are summarised below: - Bibliographic review of marine algae characterization, drying conditions and extraction of their bioactive compounds. - Experience acquisition of experimental techniques and handling of different equipment to perform different operations such as drying, milling, sieving, colorimetry, ultrasound extractions and chemical analysis (antioxidant activity, carbohydrate content, polyphenols content and alginates content determination), among others. - Determination of water sorption isotherms of F. vesiculosus at different temperatures and modelling of experimental data. - Drying of F. vesiculosus at different temperatures. Obtention and modelling of experimental data. - Evaluation of shrinkage and colour changes during drying. - Granulometric and colorimetric characterization of F. vesiculosus powders previously dried at different temperatures. - Obtaining F. vesiculosus extracts by means of ultrasound assisted extraction and evaluation of its feasibility as an alternative to obtain F. vesiculosus bioactive compounds. - Phytochemical characterization of F. vesiculosus extracts previously dried at different temperatures. Introduction 13 3.1.2. Proteins Protein, peptide and amino acids content in brown algae is relatively low (5-24% d.w.) and it strongly varies depending on species and season (Gómez-Ordóñez 2013). 3.1.3. Lipids Lipids are present in low quantities in seaweeds (<5% d.w.) and are responsible of their low calories content. As with protein and polysaccharide content, lipid content depends on the season of the year and other environmental factors. Both, brown and red algae are a valuable source of omega-3 and omega-6, which are recognized for their health benefits (Gómez-Ordóñez 2013). 3.1.4. Minerals Seaweeds accumulate a high amount of minerals (8-40% d.w.) due to the wide variety of these compounds that can be found in marine environments. These minerals include sodium, calcium, magnesium, potassium, chloride, sulfates, phosphorous, fluoride, among others. Therefore, brown algae are a suitable as dietary supplement in foods. The mineral composition of seaweeds can vary depending on species and location. On the other hand, marine algae are good indicators of heavy metals bioaccumulation due to their capacity to hold these substances (Gómez-Ordóñez 2013). F. vesiculosus is rich in light elements such as Mg, K, Ca and Na, and in halogens (Br, I, Cl). Regarding toxic elements, Fucus reported to present high concentrations of As (over 300 ppm in the Baltic Sea) and low contents on Hg, Sb and Se. In addition, mineral content in this algae does not depend on time of the year but on salt concentration of the seawater (Truus et al. 2001). 3.1.5. Polyphenols Red and green algae contain little traces of polyphenols (<1% d.w.). Brown algae have higher fractions, with species like Ascophyllum and Fucus reaching up to 14% d.w. (Ordóñez 2013). Phenolics play a primary role as structural components of cell walls and may have secondary roles in signalling, defence or in responses to environmental stress. Polyphenols are compounds with an aromatic ring bearing one or more hydroxyl substituents and include simple phenols, coumarins, flavonoids, stilbenes, lignans, hydrolysable and condensed tannins, and phlorotannins (Figure 3-4). Phlorotannins have been reported to Introduction 14 show anti-inflammatory properties, therapeutic potential in arthritis treatment and antioxidant activities (Balboa et al. 2013). Figure 3-4. Classification of some phenolic compounds from natural sources (Kyung-tae 2012). Phlorotannins are formed by the polymerization of phloroglucinol (1,3,5trihydroxybenzene) monomer units and biosynthesized through the acetate-malonate pathway (polyketide pathway). They are highly hydrophilic components with a wide range of molecular sizes ranging between 126 Da and 650 kDa. Marine brown algae accumulate a variety of phloroglucinol-based polyphenols, as phlorotannins of low, intermediate and high molecular weight containing both, phenyl and phenoxy units. Depending on linkage, phlorotannins can be classified into four subclasses such as fuhalols and phlorethols (phlorotannins with an ether linkage), fucols (with a phenyl linkage), fucophloroethols (with an ether and phenyl linkage), and eckols (with a dibenzodioxin linkage). Figure 3-5 shows the structure of some phlorotannins found in marine brown algae: Introduction 15 Figure 3-5. Structure of phlorotannins from marine brown algae. (1) phloroglucinol, (2) eckol, (3) fucodiphloroethol, (4) 7-phloroeckol, (6) dieckol and (7) 6,6’-bieckol (Li et al. 2011). Several species of brown algae, such as Ecklonia cava, Ecklonia kurome, Fucus vesiculosus, Hizika fusiformis, and Sargassum ringgoldianum, have been found to possess a high content of phlorotannins, which is correlated with the antioxidant activity (Wang et al. 2012). 3.1.6. Carotenoids Carotenoids are a family of pigmented compounds which are synthesized by plants, algae, fungi and microorganisms, but not by animals. They are the most important pigments in nature that are responsible for various colors of different photosynthetic organisms. Carotenoids are related to the prevention of many human diseases including cardiovascular diseases, cancer and other chronic diseases (Ngo et al. 2011). Introduction 16 Although chlorophyll is also present in seaweeds, in the case of brown algae, xanthophyll pigment or fucoxanthin (Figure 3-6) is the main responsible for its color (Li et al. 2011). Figure 3-6. Structure of fucoxanthin, carotenoids found in Brown algae (Ngo et al. 2011). 3.1.7. Fucus vesiculosus Fucus vesiculosus (Figure 3-7) is commonly known as bladderwrack (gaelic). It is a lower limit intertidal seaweed species and it adheres to and grows on stones (Kim 2012). It is a dominant species of macro alga in the northern Atlantic Ocean and, ecologically, it is the most important seaweed, providing shelter and food for associated flora and fauna. It shows a high genetic variability between geographic locations which provides Fucus a high capacity to cope with environmental stress (temperature changes, changes in salinity, exposure to sunlight). This feature could be important when analysing and comparing with bibliography the results of the present work (Tatarenkov et al. 2007; Lesser 2011). Figure 3-7. Fucus vesiculosus underwater (Algae Base Database). Introduction 17 Fucus vesiculosus thallus has a strap-like geometry and forks towards the end. It has a thick midrib (holdfast) which attaches to subtract. Bladders or vesicles appear from the thallus, which helps the seaweed to stand erect underwater. In the tip of the thallus Fucus develops the receptacles, which host the conceptacles (fertile parts of the seaweed) (Hoek 1996). The average composition of Fucus vesiculosus is shown in Figure 3-8. Polysaccharide content may vary and reach its maximum value after spring (growth season) to generate enough biomass to survive winter. Other authors determine that F. vesiculosus contains up to 65% of dry weight in polysaccharides (Rioux et al. 2007). Figure 3-8. Average composition of Fucus vesiculosus in dry weight (Hahn et al. 2012). 3.2. Antioxidant activity Compounds present in F. vesiculosus that have reported to show antioxidant activity and the mechanisms involved are detailed below. Sulfated polysaccharides are present mostly as fucoidans in brown algae (Phaeophyceae), carrageenan in red algae (Rhodophyceae), and ulvan in green algae (Chlorophyceae). Fucoidans are responsible of the resistance to dryness of brown algae and contribute to the formation of a gel network (Kim 2012). There is a close connection between molecular weight of these molecules and their antioxidant activity. The lower the molecular weight, the higher the antioxidant activity. Low molecular weight molecules are able to easily diffuse through cell membranes to donate protons in a higher effective way. Besides, sulfated polysaccharides are important free-radical scavengers and antioxidants for the prevention of oxidative damage and sulfated groups provide anticoagulant activity (Ngo et al. 2011). Introduction 18 Phlorotannins are mostly extracted from marine brown algae and are responsible for antioxidant activities and have shown protective effects against hydrogen peroxideinduced cell damage. Phlorotannins act as free radical scavengers, reducing agents and metal chelators, inhibiting lipid oxidation. Antioxidant activity reported by carotenoids is due to its highly unsaturated nature, which leads to their own oxidation instead of other molecules. The antioxidant properties of carotenoids are based on their singlet oxygen quenching properties or their ability to trap free radicals. Antioxidant activity depends on the number of conjugated double bonds of the molecule and carotenoid end groups or the nature of substituents in carotenoids containing cyclic end groups (Ngo et al. 2011). 3.3. Drying fundamentals Drying is one of the studied processes worldwide as it accounts for about 10-25% of the total energy consumption in manufacturing processes. It represents an important step in the food processing industry. Almost every food product is dried at least once at one point of its preparation (Mujumdar 2006). The main reasons why food products are dried are detailed below (Brennan 1994): - Extended storage life: dried food products are less vulnerable to spoilage caused by bacteria, molds and insects. Activity of several microorganisms is inhibited in systems in which moisture is below certain levels. Additionally, enzymatic and oxidative reactions might not take place under low moisture conditions. - Quality enhancement: many favorable qualities and nutritional values of food may be enhanced by drying. Palatability and digestibility are improved. Drying changes color, flavor, appearance and texture of a food item. These features may affect the decision of the end user to buy or not a certain food (Figure 3-9). - Ease of handling: packing, handling and transportation of a dried product is considerably easier and cheaper because of weight and volume loss. - Further processing: food products are dried for improved milling, mixing or segregation. In addition, dry milling is far cheaper than wet milling. - Sanitation: during drying, insects and other microorganisms are destroyed. Introduction 19 Figure 3-9. Illustration of the impact of the colour of a particular natural food product upon ‘Like or dislike’ (Birren diagram modified from that summarised by Li (1998)) (Chen & Mujumdar 2009). Many foods and, more strictly seaweeds, are generally sundried, process that requires long periods of time. During recent years, an increase of production rates of marine algae requires the application of quicker industrial methods. The most frequently used industrial drying method in food industry is the convective air drying. This process is highly influenced by air temperature and material characteristics, as well as other parameters that must be controlled. For this reason, many researches in convective air drying of several seaweeds were carried out in the recent years: Vega-Gálvez et al. (2008) studied convective drying of Macrocystis pyrifera, Gupta et al. (2011) studied convective drying kinetics of Himanthalia elongate, Tello-Ireland et al. (2011) studied drying kinetics of Gracilaria chilensis and Fudholi et al. (2012b) studied drying kinetics of Gracilaria cangii. Along with the benefits remarked before, drying, and particularly hot air drying, might have some undesirable secondary effects. Under hot air, food products undergo several reactions that alter their physical (rehydration, color loss), chemical (browning reaction, lipid oxidation) and nutritional (vitamin and protein loss, and microbial survival). Seaweeds gained attention in the food industry for the antioxidant properties showed by some of their constituent components. In that matter, there are several researches carried out in the last ten years, which study the effect of varied convective drying conditions over antioxidant activity of many marine algae species (Tello-Ireland et al. 2011; Jiménez‐Escrig et al. 2001; Kuda et al. 2005a; Kuda et al. 2005b; Le Lann et al. 2008). Introduction 20 Convective air drying of algae must be carried out finding a compromise solution to several factors: loss of bioactive properties, energy consumption, time and customer oriented properties. 3.3.1. Moisture in foods Moisture is bonded to food, mainly, in two ways: to ionic groups, such as carboxyl groups and amino acids, and to hydrogen groups, such as hydroxyl and amides. In foods in which moisture is higher than 50% wet basis (kg water/100 kg food material), free water exists within the material pores and in intercellular spaces. The former is easier to remove than the latter and is evaporated in the first stages of drying. Equilibrium moisture content (Xe) of a food material is reached when its internal vapor pressure is in equilibrium with the outside vapor pressure. The different states of equilibrium at any temperature are usually represented by a Xe versus equilibrium related humidity, or water activity. This plot, generally has a sigmoid (S) shape due to variations in the way that water bonds to the solid. Figure 3-10 shows the typical water sorption isotherm types for food systems. A totally dried solid introduced in a slightly humid environment will strongly retain water molecules in a single layer. As the ambient humidity grows, the solid will trap water molecules in the monolayer until it is full. Once this happens, water will start filling the capillary pores of the material. This process is affected by temperature and that is why equilibrium states are always represented referencing to a temperature, on the so called isotherms. Normally, food materials show a hysteresis phenomenon during adsorption and desorption processes, which is bound to irreversible physical and chemical changes. The role of moisture in food drying and storage is expressed in terms of water activity (aw). Water activity is defined in a similar manner as the relative humidity is defined in moist air, that is, the ratio of vapor pressure to the saturated vapor pressure at same temperature. It measures the chemical activity of moisture in food during drying and storage. Oxidation is avoided if water activity is lower than 0.4 and growth inhibition of most microorganisms is achieved at aw lower than 0.7. At very low water activities, within the range 0.1-0.3, some enzymes may still be active, but their reaction rates are strongly diminished (Mujumdar 2006). Introduction 21 Figure 3-10. Typical isotherm shapes for food systems: type I isotherm (A), type II isotherm (B) and type III isotherm (C) (Bell & Labuza 2000). Drying affects physical and chemical properties of food materials. Convective drying can cause changes in appearance. Most food materials shrinkage while drying due to water removal and structural modifications. Besides, colour alterations can occur as a result of chemical reactions, colorants leaching or degradation while drying. Thereby, it is mandatory to study physical and chemical alterations of food materials along with drying kinetics. 3.3.2. Drying kinetics High moisture foods (>50% wet basis) show two different drying rate behaviours (Mujumdar 2006): - Constant drying rate period: during constant drying rate, moisture at the surface is eliminated and hence, the only limitation is that offered by the rate at which water can evaporate. When surface moisture is evaporated, drying rate starts to decrease. Moisture content at which this occurs is called critical moisture content, and there might be several for each food system, depending on drying conditions. - Falling rate drying period: after water from the surface is depleted, moisture from the interior of the food item is diffused to the surface for further elimination. The rate at which water diffuses through the material is delimited by a diffusion coefficient (m2). Its value will depend on the type of the material, its geometry and external conditions. Introduction 22 A typical drying curve for food materials with both drying rate periods, and its relation with moisture content is shown in Figure 3-11. Figure 3-11. Drying curve for high moisture material (Mujumdar 2006). 3.4. Ultrasound assisted extraction 3.4.1. Introduction The use of ultrasound technology is widely extended in the food industry. It has been implemented in several large-scale commercial applications such as emulsification, homogenization, extraction, crystallization, dewatering, low-temperature pasteurization, degassing, defoaming, activation and inactivation of enzymes, particle-size distribution and changing viscosity. In the recent years it has attracted the attention for its application for the extraction of natural products in a short time, which previously required, by conventional methods, many hours or days. The conventional solvent extraction within the bodies of plants and seeds is based on the right choice of solvent and the use of heat and agitation. The use of ultrasounds improves solvent penetration and disrupts cell walls, releasing its content. Ultrasound assisted extraction (UAE) usually reduces working times, increases yields and the quality of the extract. In recent years, there are several compounds that have been extracted by UAE, especially bioactive compounds in the food industry (Picó 2013). Experimental 29 ERMS=[1 N]∑(Xexp−Xcal)12 ⁄ N i=1 (Eq. 4-7) The values of the parameters of five all models were obtained using Microsoft Excel 2013 (Solver add-on) by means of non-linear programming in which ERMS is minimized. 4.2.3. Determination of drying kinetics 4.2.3.1. Determination of drying kinetics Drying experiments were carried out in a hot air convective dryer (Angelantoni, Challenge 250, Italy) at different temperatures (35, 50, 60 and 75ºC) keeping relative humidity (30%) and air velocity (2 m/s) constant in all experiments. All experiments were performed until moisture content was close to equilibrium. For all cases, experiments were carried out in two different configurations: deep-bed configuration and thin layer disposition. In the first case, seaweed samples were selected on the basis of having similar sizes (length: 15 cm, width: 3 cm) and arranged in a metallic mesh (45x45x7 cm3) to allow transversal flow with the same load density (14.88±0.08 kg/m2). In the thin layer disposition experiments, the height of the bed was reduced so the seaweed added was only enough to cover like a monolayer the whole mesh surface (45x45x1 cm3). In addition, experiments with different load densities (1.25, 2.48, 3.07 and 14.88 kg/m2) were carried out at 75ºC. The whole seaweed with the exception of the holdfast, which was cut with a thin blade, was employed in the drying experiments. In order to use fresh algae, a new batch of seaweed was collected for each drying temperature studied. All experiments were carried out at least in duplicate. Drying kinetics were determined by weight using a balance (Cobos D-6000-CS, ±0.1 g, España). Samples were withdrawn and weighed every few minutes in the first stages of drying and every hour towards the end until moisture content reached values close to 14% (d.b.). Drying rate (r) was measured as follows (Eq. 4-8): r=wt−1−wt t1−t−1 (Eq. 4-8) Where r (g/min) is the drying rate, w (g) is the mass of seaweed in the interval of time (t, min) considered. Experimental 30 4.2.3.2. Drying kinetics modelling Deep-bed and thin layer configurations were modelled separately. Thin layer was used to determine diffusional coefficients of water in seaweed, which can only be obtained if the water is mainly removed by diffusional drying. Drying curves were modelled by following the development of moisture ratio (MR, -) over time (t, min). MR was calculated as follows (Eq. 4-9): MR=Xt−Xeq X0−Xeq (Eq. 4-9) Where Xt is the moisture content (d.b.) at any time, X0 is the initial moisture content (d.b.) and Xeq is the equilibrium moisture content of the sample (d.b.). Equilibrium moisture contents at each drying temperature are obtained from the water sorption isotherms. The models selected for the adjustment of experimental data of F. vesiculosus in the case of deep-bed configuration are depicted below (Vega-Gálvez et al. 2008) (Fudholi et al., 2012a) (Tello-Ireland et al. 2011): Newton model (Eq. 4-10): MR=e−kt (Eq. 4-10) Where k is the drying rate constant (min). Logarithmic model (Eq. 4-11): MR=ae−kt+c (Eq. 4-11) Where k is the drying rate constant (min), and a (-) and c (-) are model constants. Henderson – Pabis model (Eq. 4-12): MR=ae−kt (Eq. 4-12) Where k is the drying rate constant (min) and a (-) is a model constant. Weibull model (Eq. 4-13): MR=e(−[tβ ⁄]α) (Eq. 4-13) Experimental 31 Where α (-) and β (min) are the shape and scale parameters, respectively. Two – Term model (Eq. 4-14): MR=ae(−k0t)+be(−k1t) (Eq. 4-14) Where k0 (min-1) and k1 (min-1) are drying rate constants, and a (-) and b (-) are model constants. Page (Eq. 4-15): MR=e−ktn (Eq. 4-15) Where k (min-n) is the drying rate constant and n (-) is a model parameter. Modified Page (Eq. 4-16): MR=e−(kt)n (Eq. 4-16) Where k (min-1) is the drying rate constant and n (-) is a model parameter. Thin layer configuration drying experimental data were fit d by means of equations based on Fick’s second law for different geometries, which goal is to determine the water effective diffusivity through the seaweed. The most relevant geometries are semi-infinite layer, sphere and semi-infinite cylinder. The latter is the one that better adjusted to the experimental data of the present work. The equations associated to the semi-infinite cylinder geometry and further elucidation are showed below. The application of the Fick’s second law derived equations is subject to two assumptions (Mujumdar 2006): Initially, the distribution of the moisture within the product is uniform. At t>0 surface reaches equilibrium moisture All resistances to water removal is inside the material. External resistances is negligible. There is no shrinkage of the material during drying. To determine the effective diffusion coefficient, two equations must be applied, (Eq. 4-17) for short-time diffusion coefficient and (Eq. 4-18) for long-time diffusion coefficient (Crank 1975): Experimental 32 Mt M∞=X−X0 Xe−X0=4 π1/2.(Deff.t r2)1/2−Deff.t r2−1 3.π1 2.(Deff.t r2)1 2+⋯ (Eq. 4-17) Mt M∞=1−∑4 αn 2.e−Deff.αn 2.t/r2 ∞ n=1 (Eq. 4-18) Where Mt (g) is the amount of water removed at time t (min), M∞ (g) is the total amount of water removed when equilibrium is achieved, Deff (m2/s) is the water effective coefficient of diffusion, r (m) is the radius considered for the cylinder, αn (-) are the roots of the first order Bessel function J0 (x) for each term n (-) of the equation. In the present work, three terms of the (Eq. 4-18) were considered enough to fit the experimental data, so the expanded equation is as shown in (Eq. 4-19): Mt M∞=1−(4 α12.e−D.α1 2.t/r2+4 α2 2.e−D.α2 2.t/r2+4 α3 2.e−D.α3 2.t/r2) (Eq. 4-19) Where α1, α2 and α3 are, respectively, 2.4048, 5.5201 and 8.6537. The application of the short-time or the long-time diffusion equations is subject to the following conditions (Table 4-3): Table 4-3. Conditions for the application of short or long-time Fick’s second law diffusion equations (Fito et al. 2001). Equation Fick Number Time Mt/M∞ Short-time Deff·t/r2 < 0.5-0.6 << r2/Deff 0-0.6 Long-time Deff·t/r2 > 0.5 >> r2/Deff 0.4-1 Both equations, short and long-time diffusion equations (Eq. 4-17, (Eq. 4-19), were applied by iterating water coefficient of diffusion until conditions cited in Table 4-3 were fulfilled. The goodness of the adjustment of each model was determined based on coefficient of determination (R2) and root mean square error (ERMS) (Eq. 4-7). The values of the parameters of diffusional models were obtained using Microsoft Excel 2013 (Solver add-on) by means of non-linear programming in which ERMS is minimized. Experimental 33 4.2.3.3. Colorimetry Surface colour of the samples was measured using a colorimeter (CR 400, Konika Minolta, Japan). Calibration of the colorimeter is done by measuring the colour parameters of a standardized white glossy ceramic tile. Colour was measured by means of CIELab coordinates (L*, a* and b*) (CIELab 2006). In addition, total colour difference (∆E*) was calculated at each drying time with fresh seaweed as control value (Eq. 4-20): ∆E∗=√(L∗−Lf∗)2+(a∗−af∗)2+(b∗−bf∗)2 (Eq. 4-20) Where L* is whiteness (L*=0) or brightness (L*=100), a* is redness (a*>0) or greenness (a*<0) and b* is yellowness (b*>0) or blueness (b*<0) (Figure 4-1). Figure 4-1. CIE-Lab colour expression system (Chen & Mujumdar 2009). Colour measurement was made separately on the receptacles/conceptacles and on the lamina of the seaweed. Each measure was made nine times (three triplicates on three different surface points). To determine the qualitative effect of ∆E values, Table 4-4 was employed: Experimental 34 Table 4-4. Colour difference evaluation (Li 1998). Colour change ∆E* Trace level difference 0 – 0.5 Slight difference 0.5 – 1.5 Noticeable difference 1.5 – 3.0 Appreciable difference 3.0 – 6.0 Large difference 6.0 – 12.0 Very obvious difference >12.0 4.2.3.4. Shrinkage Most materials shrink as drying takes place. These volume changes were measured by applying Archimedes’ principle of liquid displacement. Small seaweed samples (approximatively 0.5-1.5 g) were immersed in a probe filled with heptane at 25ºC at different drying times and the volume displaced was measured. From experimental shrinkage determination, a characteristic dimension (in this case, radius of semi-infinite cylinder) was calculated and then introduced into kinetic modelling. The variation of volume of the sample (V) is measured by means of variation in comparison to its initial volume (V0). Its plotting vs volume reduction due to water removal, measures how distant from ideal behaviour (volume loss is only due to water removal) the system is. Plotting of drying rate per unit of surface area (including effect of volume) against moisture was used to estimate the critical moisture content (Xc, d.b.) at different temperatures. Drying rate per unit of surface area was obtained applying (Eq. 4-21): Jw=Xt−1−Xt t1−t−1 a (Eq. 4-21) Where Jw (d.b..m-2) is the specific drying rate a (m2) is the area at the given interval. 4.2.4. Milling After drying, F. vesiculosus was left aerating for 1-2 days. In order to facilitate milling dried seaweed was then triturated (seaweed size lower than ~0.5 cm) in a laboratory blender (Waring Blender HGBTWT, USA). Then, it was milled in an ultra-centrifugal mill Experimental 35 (ZM200 Retsch GmbH, Germany) using a 500 µm internal sieve. Afterwards, milled seaweed was storage at 4ºC in vacuum sealed bags until further utilization. 4.2.4.1. Physical and chemical characterization 4.2.4.1.1. Water content The method used to determine the moisture content of the seaweed powder was the same as the one described in section 4.2.1.1 (Water content). 4.2.4.1.2. Granulometric characterization Granulometric characterization of milled seaweed was carried out using sieves (Cisa Cedaceria Industrial, Spain) with different standardized meshes (500, 250, 125, 80, 63 and 40 µm). Size distribution curves were obtained, and the mass mean diameter Dw, (Eq. 4-22), the volume mean diameter Dv, (Eq. 4-23) and the surface mean diameter Ds, (Eq. 4-24) were calculated for each drying temperature. Dw=∑xiDpi (Eq. 4-22) Dv=∑(xiDpi ⁄)−13 ⁄ (Eq. 4-23) Ds=1 ∑∆xiDpi ⁄ (Eq. 4-24) Where Dpi (µm) is the mean diameter for each fraction and xi (-) is the weight fraction. Mass mean diameter is important for milling characterization. Dv provides information about pore surface area and may be used when equivalent volume is the main concern (buoyancy, fluidized beds). Ds is used when equivalent surface exposure area is important: reactions, adsorption or, for this matter, extractions. 4.2.4.1.3. Colorimetric characterization Colorimetric characterization was carried out applying the method described in section 4.2.3.3. It was applied to all seaweeds, formerly dried at 35, 50, 60 and 75ºC. The colorimetric parameters were obtained for all fractions for each system. Experimental 36 In this case, the parameters used as control for the calculation of ∆E* of different fractions were the ones obtained for all the fractions mixed. 4.2.5. Extracts obtention and characterization 4.2.5.1. Ultrasound extraction The dried samples of algae were treated with ultrasound in order to facilitate the extraction of the components of interest. The equipment employed was an ultrasound homogenizer, model UIP – 1000 hdT (Hielscher, Germany) (Figure 4-2). It can operate in batch or in continuous flow. In this study, all experiments were carried out in batch because of the low amount of seaweed powder employed, and for a greater ease of use and control of operation conditions. Extractions took place using a 200 mL beaker. The operation temperature was controlled employing a cold water bath to avoid temperatures higher than 40ºC that could affect antioxidant activity. All extractions were performed using water as solvent, except for one, carried out with acetone/water (70:30 v/v) to obtain an extract as a reference point. All systems were rehydrated during 15 minutes before the extraction. Figure 4-2. Ultrasound transducer and generator model UIP – 1000 hdT (Hielscher). The equipment operates with a frequency of 20 kHz and irradiation power up to 1000 W. The latter can be regulated in the panel in the ultrasound generator. Experimental 37 Extracts were centrifuged at 12400 rpm for 15 minutes using a Sigma 2 – 15 centrifuge (SciQuip, United Kingdom). The filtered supernatant (0.25µm) of the extracts were used for physical and chemical analysis. 4.2.5.2. Physical and chemical characterization of extracts 4.2.5.2.1. Antioxidant activity Two main methods were used to measure antioxidant activity of the extracts: polyphenols content and DPPH scavenging activity. 4.2.5.2.1.1. Polyphenol content The quantitative determination of polyphenols was put into practice applying the Single and Rossi method (Singleton & Rossi 1965). It is a colorimetric method based on the change in absorbance (765 nm) of the Folin-Ciocalteau reagent when reacting with the hydroxyl groups of the polyphenolic substances. The absorbance measure is introduced in the calibration line obtained for a known phenolic compound. In this study, the compound of choice is the phloroglucinol. Hence, polyphenols content of the extracts will be expressed as phloroglucinol equivalents. 4.2.5.2.1.2. DPPH scavenging activity Antioxidant capacity of food systems is attributed to its capacity to inhibit the effect of different free radical substances. The DPPH assay proposed by Brand-Williams et al. (1995) measures the capacity of the system to react with a known free radical agent (2,2diphenyl-1-picrylhydrazyl, DPPH). In its radical form, DPPH absorbs at 515 nm, but upon reduction by an antioxidant (AH) or a radical species (R) the absorption disappears. As the reaction takes time to fully develop, for the determination of the DPPH scavenging activity absorbance is measured every 5 minutes until it reaches the stationary state. Scavenging activity is measured by means of (Eq. 4-25): Scavenging activity (%)=(𝑎𝑏𝑠𝑖𝑛𝑖𝑡𝑖𝑎𝑙−𝑎𝑏𝑠𝑓𝑖𝑛𝑎𝑙) 𝑎𝑏𝑠𝑖𝑛𝑖𝑡𝑖𝑎𝑙 (Eq. 4-25) Where absinitial (-) is absorbance at time 0 and absfinal (-) is the absorbance after one hour. Experimental 38 4.2.5.2.2. Carbohydrates Carbohydrates content of the extracts was determined using the Dubois, et al. (1956) method. This method employs sulfuric acid and phenol as reagents. In the presence of strong acids and heat, carbohydrates undergo a series of reactions which lead to the formation of furan derivatives such as furanaldehyde and hydroxymethyl furaldehyde. These compounds condense with themselves or with phenolic substances leading to the formation of dark colored compounds (Figure 4-3). Figure 4-3. Furan derivatives from (a) pentoses and hexuronic acids, (b) hexoses, (c) 6-deoxyhesoses, and (d) keto-hexoses, respectively (Brummer & Cui 2005) Furan derivatives from pentoses and hexoses exhibit peaks of light absorbance in the range of 480 – 490 nm (Figure 4-4) (Brummer & Cui, 2005) . Figure 4-4. Phenol–sulphuric acid assay absorbance maxima for hexoses and pentoses (Brummer & Cui 2005) For this work, the samples were evaluated measuring the absorbance read at 485 nm and glucose was used to draw the calibration curve. Hence, carbohydrates concentration is expressed as glucose equivalents. Figure 6-3. Experimental data of sorption isotherms along with GAB (Eq. 4-2) and Caurie (Eq. 4-4) models. 5ºC (♦), 25ºC (♦), 45ºC (♦) and 65ºC (♦). Adsorption adjustments are represented with straight lines: 5ºC (—), 50ºC (—), 45ºC (—) and 65ºC (—) (dashed lines represent desorption isotherms). 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Caurie model for water adsorption 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) GAB model for water adsorption 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) GAB model for water desorption 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Caurie model for water desorption Results and discussion 459 Figure 6-4. Experimental data of sorption isotherms along with Halsey (Eq. 4-5) and Oswin (Eq. 4-6) models. 5ºC (♦), 25ºC (♦), 45ºC (♦) and 65ºC (♦). Adsorption adjustments are represented with straight lines: 5ºC (—), 50ºC (—), 45ºC (—) and 65ºC (—) (dashed lines represent desorption isotherms). 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Halsey model for water adsorption 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Halsey model for water desorption 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Oswin model for water adsorption 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0 0.2 0.4 0.6 0.8 1 Xe(d.b.) aw(-) Oswin model for water desorption Results and discussion 46 Results and discussion 47 Models plotting allow a clearer appreciation of the isotherms crosspoint at high water activities. Oswin reported the lowest average R2 values and the highest average ERMS values, 0.985 and 0.444 respectively. In average, Halsey presented the best average R2 (0.99) and ERMS (0.040) coefficients. Caurie and GAB provide a good adjustment, but as it can be observed in Figure 6-3 and Figure 6-4 (and data provided by R2 and ERMS coefficients), said adjustment worsens at low temperatures. Halsey’s model properly adjusts type II and III isotherms and, hence, shows the best adjustment for lower temperatures (5 – 25ºC, type II isotherm) and higher temperatures (45 – 65ºC, type III isotherm). Halsey parameters have been found to follow linear (Eq. 6-1) and Arrhenius correlations (Eq. 6-2). The correlation of these parameters with temperature is shown in Figure 6-5. A,1𝐵 ⁄ =a+b·T (Eq. 6-1) ln(𝐴,1𝐵 ⁄ )=ln(𝐴0,1𝐵 ⁄0)− Ea R·T (Eq. 6-2) Where a (-) and b (-) are equation parameters, T (K) is the temperature, Ea is the energy of activation (J/mol) and R (8.314 J/K·mol) is the universal constant of gases. Constants of each correlation for both, adsorption and desorption Halsey model parameters, are shown in Table 6-2: Table 6-2. Linear and Arrhenius correlations parameters. Adsorption Desorption Linear (Eq. 6-1) a (d.b.·K) b (d.b.) R2 a (d.b.·K) b (d.b.) R2 A (d.b.·K) -284.091 1.108 0.991 -295.17 1.113 0.94 a (-) b (K-1) R2 a (-) b (K-1) R2 1/B (-) -4.04 0.018 0.96 -5.82 0.023 0.993 Arrhenius (Eq. 6-2) Ea (kJ/mol) ln(A0,1/B0) R2 Ea (kJ/mol) ln(A0,1/B0) R2 A 16.38 10.38 0.998 18.88 11.10 0.98 1/B 9.82 4.21 0.96 14.42 5.88 0.998 0.0 0.5 1.0 1.5 2.0 2.5 0 20 40 60 80 100 120 275 285 295 305 315 325 335 345 1/B (-) A (d.b.) T (K) Adsorption 0.0 0.5 1.0 1.5 2.0 2.5 0 20 40 60 80 100 120 275 285 295 305 315 325 335 345 1/B (-) A (d.b.) T (K) Desorption Figure 6-5. Linear and Arrhenius correlations of Halsey model parameters: A (♦, - - -) and 1/B (■,····). Upper figures represent linear correlation (Eq. 6-1) for adsorption and desorption parameters at different temperatures. Lower figures represent Arrhenius correlation (Eq. 6-2) at different temperatures. -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.0029 0.0031 0.0033 0.0035 0.0037 ln(1/B) ln(A) 1/T (K-1) Adsorption -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.0029 0.0031 0.0033 0.0035 0.0037 ln(1/B) ln(A) 1/T (K-1) Desorption Results and discussion 48 Results and discussion 49 Both correlations can be introduced into Halsey model, resulting in models shown in equations (Eq. 6-3) and (Eq. 6-4): Xeq=(−(aA+bA·T) Tln(aW))(a(1/B))+b(1/B)·T) (Eq. 6-3) Xeq=(−A0·e−Ea,AR·T ⁄ Tln(aW))1B ⁄0·e−Ea,(1/B)R·T ⁄ (Eq. 6-4) Figure 6-6 shows the representation of both models fitting for water adsorption and desorption experimental data. Linear correlation presents a better adjustment (average R2 of 0.990, average ERMS of 0.075) than Arrhenius correlation (average R2 of 0.980, average ERMS of 0.100) for all temperatures. Figure 6-6. Experimental data of sorption isotherms along with modified Halsey model (Eq. 6-3, (Eq. 6-4) : 5ºC (♦), 25ºC (♦), 45ºC (♦) and 65ºC (♦). Adsorption adjustments are represented with straight lines: 5ºC (—), 50ºC (—), 45ºC (—) and 65ºC (—) (dashed lines represent desorption isotherms). 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0 0.2 0.4 0.6 0.8 1.0 Xe(d.b.) aw(-) Halsey modified model for water adsorption (Arrhenius) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0 0.2 0.4 0.6 0.8 1.0 Xe(d.b.) aw(-) Halsey modified model for water desorption (Arrhenius) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0 0.2 0.4 0.6 0.8 1.0 Xe(d.b.) aw(-) Halsey modified model for water adsorption (linear) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0 0.2 0.4 0.6 0.8 1.0 Xe(d.b.) aw(-) Halsey modified model for water desorption (linear) Results and discussion 50 Results and discussion 51 6.2. Determination of drying kinetics As mentioned in chapter 4.2.3 (Determination of drying kinetics), experimental drying kinetics were determined using two configurations: deep-bed and thin layer configuration. Below, both configurations are treated separately. 6.2.1. Deep-bed configuration drying kinetics Experimental drying of F. vesiculosus for bed configuration was carried out at four different temperatures (35, 50, 60 and 75ºC). Evolution of moisture was measured by means of moisture ratio and can be seen in Figure 6-7. All cases reached a final moisture ratio lower than 0.04 (close to equilibrium moisture). Drying temperature effect could be observed on total drying time, which was reduced as the former was increased. Drying at 35ºC exhibited the highest drying time to achieve a MR of 0.04 with 1517 min, followed by 50ºC with 1395 min, and 60ºC and 75ºC required the lowest drying time with 1200 min. Figure 6-7. Experimental drying curves for Fucus vesiculosus (deep-bed configuration) at different drying temperatures: 35ºC (♦), 50ºC (♦), 60ºC (♦) and 75ºC (♦). Drying rate can be measured, as shown in Figure 6-8, as means of variation of mass over time. 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0200 400 600 800 1000 1200 1400 1600 MR (-) Time (min) Results and discussion 52 Figure 6-8. Drying rate curves for Fucus vesiculosus (deep-bed configuration) at different drying temperatures: 35ºC (♦), 50ºC (♦), 60ºC (♦) and 75ºC (♦). Drying rate curves for 60 and 75ºC display a clear induction stage at high moisture. This phase corresponds to the unsteady state heating period. Afterwards, both systems show a similar trend to that displayed by the 35 and 50ºC cases for the whole experiment. This subsequent stage is characterized by a falling rate period, in which the rate of moisture migration to the surface is lower than the rate of evaporation and, thus, originating a dried surface and an increase in its temperature. This stage lasts until the equilibrium moisture is reached (Hui 2008). Drying kinetics experimental data were fitted to the models mentioned in chapter 4.2.3.2 (Modelling of drying kinetics). Resulting parameters of the corresponding fittings are shown in Table 6-3. Two – term (Eq. 4-14), Page (Eq. 4-15) and modified Page (Eq. 4-16) presented the best adjustment (R2 equal to 0.999 and ERMS lower than 0.018). The former is a four parameter model and showed the best fit (ERMS<0.007). Regarding Page and modified Page, these two only need two parameters to be adjusted, so they are simpler. Besides, Page model is the most used model to fit drying kinetics data. Page model fit parameters are shown in Table 6-3. From the kinetic parameter k for each temperature, one can draw the same conclusion observed in Figure 6-7 but more clearly identified in Figure 6-8. There exists a gap between drying rate at 35-50 ºC and 60-75ºC. 0 5 10 15 20 25 0100 200 300 400 500 600 700 r (g/min) X (%, d.b.) Table 6-3. Drying models fitting parameters for each temperature (1) (Newton, Logarithmic, Henderson – Pabis models, Weibull and Two – term models). ϯ Model Newton (Eq. 4-10) Logarithmic (Eq. 4-11) Henderson – Pabis (Eq. 4-12) T (ºC) k·103(min-1) ERMS R2 a (-) k·103(min-1) c·103 (-) ERMS R2 a (-) k·103(min-1) ERMS R2 35 2.99±0.15 0.026 0.999 0.94±0.02 3.15±0.26 37.10±0.07 0.014 0.999 0.96±0.01 2.80±0.16 0.020 0.998 50 3.92±0.04 0.021 0.999 0.93±0.01 4.24±0.20 39.00±0.07 0.010 0.999 0.97±0.01 3.70±0.05 0.016 0.998 60 4.69±0.27 0.045 0.995 0.89±0.01 4.68±0.15 38.81±0.07 0.028 0.995 0.90±0.01 4.00±0.27 0.032 0.994 75 4.83±0.14 0.047 0.995 0.86±0.03 4.70±0.44 18.02±0.07 0.024 0.996 0.89±0.01 4.04±0.15 0.027 0.995 Model Weibull (Eq. 4-13) Two – term (Eq. 4-14) T (ºC) α (-) β (min) ERMS R2 a (-) k0·103(min-1) b (-) k1·103(min-1) ERMS R2 35 0.86±0.01 335.2±6.1 0.009 0.999 0.31±0.02 7.50±2.33 0.69±0.02 2.13±0.18 0.005 0.999 50 0.89±0.01 262.8±12.1 0.007 0.999 0.22±0.04 12.43±1.92 0.79±0.04 3.07±0.15 0.004 0.999 60 0.78±0.01 209.7±12.1 0.016 0.999 0.30±0.01 25.52±1.90 0.75±0.01 3.18±0.22 0.007 0.999 75 0.77±0.02 233.2±18.1 0.013 0.999 0.25±0.05 45.43±4.89 0.81±0.1 3.53±0.19 0.004 0.999 Model Page (Eq. 4-15) Modified Page (Eq. 4-16) T (ºC) k·103(min-n) n (-) ERMS R2 k·103(min-1) n (-) ERMS R2 35 6.63±0.52a 0.86±0.01a 0.009 0.999 2.98±0.16a 0.86±0.02a 0.009 0.999 50 6.94±0.21a 0.89±0.01a 0.007 0.999 3.83±0.03b 0.89±0.01a 0.007 0.999 60 14.74±0.13b 0.78±0.01b 0.016 0.999 4.58±0.27c 0.77±0.01b 0.018 0.999 75 15.78±0.12b 0.77±0.02b 0.013 0.999 4.67±0.19c 0.79±0.01b 0.015 0.999 ϯ: Data are presented as means of ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05 Results and discussion 53 Results and discussion 54 This is noticeable on the kinetic parameter k for the Page model, which is significantly different for both pair of temperatures. This gap is substantially more evident at earlier stages of drying. At this period the sample undergoes an unsteady heating phase and resistance to drying is focused on external convection. On later stages, drying rate curves seem to match their trends because resistance to water removal is now strong on diffusion of water through the food material. This event is studied more thoroughly within the monolayer configuration drying chapter. Page parameters for another macro algae species can be found in bibliography. Vega-Gálvez et al. (2008) adjusted drying experimental data of Macrocystis pyrifera (intertidal brown algae) to the modified Page model for similar temperatures and obtained the following kinetic parameters (k): of 6.44·10-3 min-n at 50ºC, 9.71·10-3 min-n at 60ºC and 13.74·10-3 min-n at 70ºC. These parameters are substantially higher than those observed for F. vesiculosus. Lemus et al. (2008) provides values of parameters for the convective drying of Gracilaria chilensis (intertidal red algae) for the modified Page model: 3.5·10-3 min-1 at 40ºC, 7.3 ·10-3 min-1 at 50ºC, 9.0·10-3 min-1 at 60ºC and 13.7·10-3 min-1 at 70ºC. These values are higher than those obtained for F. vesiculosus within the same range of temperatures. Fudholi et al. (2012b) obtained the Page parameters values for the convective drying of Eucheuma cottonii (subtidal red seaweed): 0.99 min-n (n = 0.83) at 40ºC, 1.00 min-n (n = 0.90) at 50ºC and 0.94 min-n (n = 1.03) at 60ºC. Values are considerably lower than those obtained for F. vesiculosus. All parameters values are within the range of those consulted in bibliography. However, when comparing them with those obtained in this work, it must be taken into account that different configurations and conditions of drying were applied. In Figure 6-9 and Figure 6-10, the drying experimental data is shown along with the models adjusted for each temperature. Results and discussion 61 Figure 6-15. Volume shrinkage vs moisture content at different temperatures: 35ºC (♦), 50ºC (♦), 60ºC (♦) and 75ºC (♦). Shrinkage modelling was introduced into the evaluation of drying rate curves (by means of moisture loss over time). The subsequent representation of the drying rate per unit of surface area curves is shown in Figure 6-16: Figure 6-16. Effect of shrinkage on drying rate curves for Fucus vesiculosus (thin layer configuration) at different drying temperatures: 35ºC (♦), 50ºC (♦), 60ºC (♦) and 75ºC (♦). Figure 6-16 shows that there is a constant drying rate and a falling rate period at each drying temperature. The moisture cross-point of both periods corresponds to the critical 0 10 20 30 40 50 60 70 80 90 100 0100 200 300 400 500 600 700 V/V0 (%) X (%, d.b.) 0 10 20 30 40 50 60 70 80 90 100 0100 200 300 400 500 600 700 Jw(X%, dry basis/min.m2) X (%, d.b.) Results and discussion 62 moisture content. This moisture content increases with drying temperature increases. In other words, at higher temperatures, the constant drying rate period is shortened. Table 6-5 collects the critical moisture content at each drying temperature for F. vesiculosus: Table 6-5. Critical moisture content (Xc) at different drying temperatures of F. vesiculosus. Drying temperature (ºC) XC (%, d.b.) 35 161 50 249 60 293 75 300 These values were used as initial moisture content for the determination of water effective coefficient of diffusion. From that point, it is considered that drying is governed by mass diffusion and, thereby, Fick’s second law equations may be applied. It is important to note that shrinkage of a single sample cannot be introduced into deep-bed configuration kinetics, since the behaviour is altered by moisture gradients throughout the bed. Below, modelling of the falling drying rate period is depicted. 6.2.2.2. Thin layer configuration modelling Drying kinetics of F. vesiculosus modelling under thin layer configuration and in the post-critical period was performed applying equations (Eq. 4-17) and (Eq. 4-19), and conditions presented in Table 4-3, to drying experimental data beyond critical moisture content for each temperature. The radius considered is the average radius for all systems when volume shrinkage is almost negligible (0.00175 m). Volume is considered constant at this sftage, as it is a requirement for the application of diffusional modelling equations. Table 6-6 shows the effective diffusional coefficient calculated for each temperature. Water coefficient of diffusion increases with temperature, doubling its value when temperature is raised from 35ºC to 75ºC. Results and discussion 63 Table 6-6. Estimation of water effective coefficient of diffusion for Fucus vesiculosus at different drying temperatures. ϯ Temperature (ºC) Deff·10-12 (m2/s) R2 ERMS 35 95.00±0.01a 0.995 0.082 50 153.75±0.02b 0.993 0.065 60 192.50±0.02c 0.993 0.034 75 260.03±0.08d 0.992 0.037 ϯ: Data are presented as means of ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05 It is hard to compare effective diffusion coefficient with bibliography, since its value strongly depends on the geometry chosen to model drying kinetics. A similar approach was made by Vega-Gálvez et al. (2008) for Gracilaria chilensis (2.76 – 22.41·10-9 m2/s) and Macrocystis pyrifera (5.56 – 10.22 ·10-9 m2/s) within a range of temperatures of 50 to 80ºC. In these cases, drying kinetics were adjusted to an infinite slab geometry. Taking into account that difference, it can be said that F. vesiculosus shows a considerable lower effective diffusion coefficient for all drying temperatures. Effective diffusivity coefficients were correlated making use of linear (Eq. 6-6) and Arrhenius relationships (Eq. 6-7): Deff=a+b·T (Eq. 6-6) ln(Deff)=ln(D0)−Ea R·T (Eq. 6-7) Where a (-) and b (-) are equation parameters, T (K) is the temperature, Ea is the energy of activation (J/mol) and R (8.314 J/K·mol) is the universal constant of gases. The linear correlation (Eq. 6-6) is displayed on Figure 6-17. On the other hand, Arrhenius adjustment (Eq. 6-7) is displayed on Figure 6-18. The parameters of both fittings are shown in Table 6-7: Table 6-7. Linear and Arrhenius fitting parameters of effective diffusivity coefficients at different drying temperatures. Linear (Eq. 6-6) a (m2/s) b (m2/s·K) R2 Deff (m2/s) -1·10-9 4·10-12 0.998 Arrhenius (Eq. 6-7) Ea (kJ/mol) ln(D0) R2 Deff (m2/s) 22.38 -14.3 0.992 Results and discussion 64 Figure 6-17. Fitting of water coefficient of diffusion with temperature by means of (Eq. 6-6) and parameters shown in Table 6-7. Figure 6-18. Fitting of water coefficient of diffusion with temperature by means of (Eq. 6-7) and parameters shown in Table 6-7. Energy of activation (Ea) can be obtained from the slope in the Arrhenius correlation (Figure 6-18), and is equal to 22.38 kJ mol-1, which is within the range 12.7 – 110 kJ mol-1 exhibited by other food materials (Zogzas et al. 1996) and other seaweeds like Macrocystis pyrifera, 19.87 kJ mol-1 or Gracilaria chilensis, 39.92 kJ mol-1 (Vega-Gálvez et al. 2008). 0.E+00 5.E-11 1.E-10 2.E-10 2.E-10 3.E-10 3.E-10 305 310 315 320 325 330 335 340 345 350 355 Deff (m2/s) Temperature (K) -23.2 -23.0 -22.8 -22.6 -22.4 -22.2 -22.0 -21.8 2.9E-03 2.9E-03 3.0E-03 3.0E-03 3.1E-03 3.1E-03 3.2E-03 3.2E-03 3.3E-03 3.3E-03 ln(Deff) 1/T (K-1) Results and discussion 65 6.2.3. Effect of load density on drying kinetics The effect of load density was studied at a drying temperature of 75ºC. Four load densities were considered: 14.88, 3.07, 2.48 and 1.25 (kg/m2). Systems required, respectively, 660, 116, 108 and 40 minutes to achieve a moisture ratio of 0.1. A linear correlation (R2 = 0.999) between drying time and load density could be established Figure 6-19. Figure 6-19. Effect of load density on drying time required to achieve a moisture ratio of 0.1. Page’s model was used to adjust experimental data. The parameters determined for said model are displayed in Table 6-8: Table 6-8. Effect of load density on Page model parameters. Load density (kg/m 2 ) k·10 3 (min-n) n(-) E RMS R2 1.25 61.40 0.89 0.024 0.995 2.48 25.89 0.95 0.001 0.999 3.07 14.54 1.07 0.001 0.999 14.88 14.91 0.79 0.013 0.999 In Figure 6-20, the experimental drying kinetic curves are represented along with the profiles obtained with Page’s model: 0 2 4 6 8 10 12 14 16 0 100 200 300 400 500 600 700 Load density (kg/m2) Time (min) Results and discussion 66 Figure 6-20. Effect of load density on drying curves (14.88 (♦), 3.07 (-), 2.48 (■) and 1.25 (▲) kg/m2) and fit of Page’s model (—). System dried with a load density of 14.88 kg/m2 shows a clear induction stage at high moisture content. In the case of the system dried with a load density of 3.07 kg/m2, a small induction phase is observed. It cannot be distinguishable when adjusted to Page’s model. When a water content of around 250 %, d.b. is reached, all systems converge. At that point (critical moisture), diffusion of water through the solid (bed in this case) becomes the limiting factor. This transition was better observed on the thin layer configuration and the subsequent modelling. In this case, the assumptions needed to adjust experimental data to diffusion models are not fulfilled, as moisture profiles are generated through the material depth during drying. The effect of load density on drying rate can also be appreciated on Figure 6-21. The main difference of drying rate with different load densities was observed at high moisture content, where free water removal is carried out during the initial stages of drying. Page’s model provided a good fit of the drying rate curves. 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0200 400 600 800 1000 1200 1400 MR (-) Time (min) Results and discussion 67 Figure 6-21. Effect of load density on drying rate curves (14.88 (♦), 3.07 (-), 2.48 (■) and 1.25 (▲) kg/m2) and fit of Page’s model (—). 6.2.4. Colorimetric characterization Table 6-9 shows the fresh and the final colour parameters of F. vesiculosus seaweed for all four drying temperatures studied: No big differences were appreciated between receptacles and lamina initial and final colour parameters. It can be observed that parameter L* (brightness of the product) increases significantly its value after processing at 75ºC for both, receptacles and lamina. This substantial change may be triggered by compounds expelled at high temperatures due to structural damage. Regarding parameter a* (redness – greenness), its diminution could be linked to alterations in algae pigments. Chromatic parameter b* (blueness – yellowness) increase at 75ºC in comparison to the final value at 35, 50 and 60ºC could be related to the reaction of reducing sugars and amino acids producing non-enzymatic browning reactions (Rahman 2006). Nevertheless, it is difficult to correlate colour parameters behaviour to a single process, since they represent the global properties of the materials, and they measure an average change between the degradation of starting materials and the formation of new coloured species (Landrum 2009). 0 5 10 15 20 25 0 100 200 300 400 500 600 700 r (g/min) X (%, d.b.) Results and discussion 68 Table 6-9. Colorimetric parameters of fresh and dried (at different temperatures) Fucus vesiculosus seaweed. ϯ Receptacles Temperature (ºC) L* a* b* ∆E Fresh seaweed 16.26±0.48 a 1.48±0.45 a 7.38±1.72a - 35 16.42±1.17 a 0.88±0.24 b 3.49±1.21b,c 3.78 ±0.11a 50 15.72±0.98 a 0.09±0.23 c 3.30±0.39b,c 4.36±0.44a,b 60 18.86±0.77 b 0.07±0.11 c 2.85±0.66b 5.01±0.33b 75 28.72±0.1 c 0.98±0.11 b 4.50±0.05c 12.79±0.01c Lamina Temperature (ºC) L* a* b* ∆E Fresh seaweed 18.62±2.11a 1.38±0.15a 6.77±0.71a - 35 20.04±0.92a 0.62±0.14b 1.54±0.45b 5.52±0.64a 50 18.66±1.35a 0.42±0.05c 1.74±0.97b,c 4.14±0.01b 60 18.95±0.19a -0.28±0.01d 2.67±0.18c 3.41±0.01a,b 75 26.72±0.37b 0.22±0.11e 3.84±0.54d 7.60±0.01c ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. According to the scale displayed in Table 4-4, systems processed at 35, 50 and 60ºC reveal an appreciable colour difference (∆E*) in comparison with fresh samples. Seaweed dried at 75ºC exhibits a large difference in colour change in the lamina, while it develops an obvious difference in its receptacles. The relationship between colour perception and water content is not clear. In this case, colour parameters may be affected as water is removed, due to pigments concentration, or their degradation. Furthermore, surface colour parameters may not be representative of the whole food material, as its temperature can reach higher values and that could stimulate the presence of Maillard reactions (Chen & Mujumdar 2009). 6.3. Physical characterization of milled seaweed After drying, algae was aerated during 1 or 2 days prior to milling. Seaweed powder had an average moisture content of 10.47±1.38 (%, d.b.). In this section, the measured physical properties of the milled seaweed are discussed. Results and discussion 69 6.3.1. Granulometric characterization Granulometric characterization, as stated in chapter 4.2.4.1.2, was achieved by screening with mesh sizes varying from 40 to 500 µm. Table 6-10 displays the fractions acquired for each system assayed: Table 6-10. Size distribution for seaweed powder formerly dried at different temperatures. ϯ Mass fractions (%) Dp (µm) Dpi (µm) FV35 FV50 FV60 FV75 >500 500 1.49±0.12 1.14±0.66 0.97±1.33 1.64±0.25 250 - 500 375 45.82±4.98 40.81±0.53 38.06±15.64 46.73±1.36 200 - 250 225 9.27±1.24 9.06±2.21 14.32±5.46 9.10±0.41 125 - 200 162.5 14.05±0.54 16.07±1.99 19.27±6.76 14.56±2.83 80 - 125 102.5 10.82±0.80 14.32±5.11 10.51±1.99 11.90±3.64 63 - 80 71.5 7.81±0.58 7.82±3.13 4.22±0.73 6.41±0.91 40 - 63 51.5 7.54±1.31 9.52±7.69 8.33±0.92 6.59±4.22 < 40 30 3.19±5.22 1.26±1.51 4.32±1.12 3.07±4.37 ϯ: Data are presented as means ± standard deviation. Table 6-10 reflects that, for all systems assayed, the highest fractions are present at a particle size of 375 µm (38.06 – 46.73%). The particle size with the second highest fraction was 162.5 µm (14.05 – 19.27%). The particle size that exhibited the lowest fraction for all systems was the corresponding to fines or lower than 40 µm (1.26 – 4.32% - particle size higher than 500 µm is considered residual, lower than 1.64%). Figure 6-22 shows the size particle distribution plotting. All systems exhibit similar distributions and three distinguishable peaks. The highest is observable at particle sizes of 350 to 400 µm for all systems. The other two were e found at smaller particle sizes (around 50 and 150 µm). Figure 6-23 the cumulative distribution of seaweed powder dried at different temperatures. A linear correlation is found for the cumulative particle size for all systems. Results and discussion 70 Figure 6-22. Particle size distribution for F. vesiculosus powder formerly dried at different temperatures: 35ºC (—), 50ºC (—), and 60 ºC (—) and 75ºC (—). Figure 6-23. Cumulative particle size distribution for F. vesiculosus powder formerly dried at different temperatures: 35ºC (—), 50ºC (—), and 60 ºC (—) and 75ºC (—). Table 6-11 shows the mass mean diameter (Dw), volume mean diameter (Dv) and the surface mean diameter (Ds) calculated employing, respectively, Eqs. (Eq. 4-22), (Eq. 4-23) and (Eq. 4-24) for each system. 0 5 10 15 20 25 30 35 40 45 50 050 100 150 200 250 300 350 400 450 500 Mass fraction (%) Particle size(mm) 0 10 20 30 40 50 60 70 80 90 100 050 100 150 200 250 300 350 400 450 500 Mass fraction (%) Particle size(mm) 0 5 10 15 20 25 30 35 40 45 50 Mixture <40 >40 >63 >80 >125 >200 >250 >500 b* DP(µm) 0 10 20 30 40 50 60 70 80 90 100 Mixture <40 >40 >63 >80 >125 >200 >250 >500 L* Dp (µm) -15 -13 -11 -9 -7 -5 -3 -1 Mixture <40 >40 >63 >80 >125 >200 >250 >500 a* Dp(µm) 0 5 10 15 20 25 30 35 40 45 50 Mixture <40 >40 >63 >80 >125 >200 >250 >500 ∆E* Dp (µm) Figure 6-26. Colour parameters for F. vesiculosus size fractions previously dried at: 35ºC (—), 50ºC (—), and 60 ºC (—) and 75ºC (—). Results and discussion 77 Results and discussion 78 6.4. Extracts characterization 6.4.1. Effect of changes in extraction operation conditions The effect of operation variables were studied on FV35 system. 6.4.1.1. Polyphenols content Total polyphenolic content (TP) is presented by means of phloroglucinol equivalents referred to sample (mg PHL/100 g dry sample, TPw) or to total solids content in the extract (mg PHL/100 g dry solids, TPs). The first condition studied was the liquid/solid ratio employed for the extraction (L/S) The rest of the conditions remained constant (residence time of 4 minutes and 70 of amplitude level) and the L/S was studied between a 20 and 40 L/S range. Table 6-13 shows the effect of L/S variations on TP: Table 6-13. Effect of the variation in L/S ratio on TP for FV35 system. ϯ L/S (g/g) TPw TPs 20 1007.47±45.61a 1433.18±64.88a 30 1571.21±75.53b 2940.44±141.35b 40 1771.52±18.96c 2377.29±25.45c ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. The highest yield is obtained with the system with a 40 L/S ratio (1771.52±18.9 TPw). A higher proportion of solids (30 L/S) results in a slightly lower TP content (1571.21±75.53 TPw). Further increase in L/S ratio reduces TP yield by ~43% (1007.47±45.61 TPw). This effect is a known problem when applying ultrasound extraction to dispersed phases. In a highly concentrated system, sound wave amplitude may be employed in breaking particle agglomerates (Kadam et al. 2013). A change in trend is observed when TP is referred to total solids content in the extract. The lowest TP within solids of the extract is again achieved at the lowest L/S ratio (1433.18±64.88 TPs). The highest is now reached by extraction carried out with a L/S ratio of 30 (2940.44±141.35 m TPs) System studied with a 40 L/S ratio exhibited a ~30% lower TP (2377.29±25.45 TPs). Lower L/S ratios may be favourable for an overall extraction of seaweed components. However, a higher L/S ratio appears to be more selective with polyphenols extraction. Results and discussion 79 The effect of changes in contact time on TP was also studied (Table 6-14): Table 6-14. Effect of contact time on TP content of extracts obtained with system FV35. ϯ Contact time (min) TPw TPs 4 1571.20±75.52a 2253.06±108.30a 12 1514.25±68.47a 2029.01±91.75a 20 1334.02±72.00b 1758.59±7.16b ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. Contact times within the range of 4 to 12 minutes reported little or no influence over TP, achieving 1571.20±75.52 and 1334.02±72.00 TPw for 4 and 12 minutes respectively. A slight reduction is observed when extraction time is increased to 20 minutes (1334.02±72.00 TPw). Due to degradation polyphenols fractions may be progressively reduced from 2253.06 to 1758.59 TPs when contact time is increased from 4 to 20 minutes, respectively. This is due to the higher extraction yields of other substances (i.e. carbohydrates) when applying longer residence times, or the degradation or polyphenols. Rodrigues et al. (2008) also reported no differences in polyphenols content on extracts of Ascophyllum nodosum when varying contact time. 6.4.1.2. Carbohydrates content Carbohydrates content of extracts, referred to dry sample (mg CHO/100 gdw, CHOw) or to total solids in the extract (mg CHO/100 gds, CHOs) Table 6-15 shows the effect of varying the L/S ratio of the extraction on carbohydrates content: Table 6-15. Effect of the variation in L/S ratio on carbohydrate content for FV35 system. ϯ L/S (g/g) CHOw CHOs 20 5423.70±344.02a 8572.77±543.76a 30 5152.44±259.59a 8217.62±414.01a 40 5204.15±881.63a 7759.66±478.89a ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. Carbohydrate content showed no significant differences between systems extracted with L/S ratios of 20, 30 or 40 (g/g). Results and discussion 80 The effect of extraction residence time is displayed on Table 6-16: Table 6-16. Effect of contact time on carbohydrate content of extracts. ϯ Contact time (min) CHOw CHOs 4 5152.44±259.59a 8217.61±414.01a 12 7438.67±815.22b 11300.32±911.74b 20 7095.70±168.47b 10427.19±247.57b ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. The highest extraction yield of carbohydrates is attained for contact times of 12 (7438.67±815.22 CHOw, 11300.32±911.74 CHOs) and 20 (7095.70±168.47 CHOw, 10427.19±247.57 CHOs). A decline in extraction yield of carbohydrates is observed when contact time is reduced to four minutes (5152.44±259.59 CHOw, 8217.61±414.01 CHOs). 6.4.1.3. Alginates Alginates content is measured by means of glucuronic acid equivalents (mg GLU/100 gdw, GLUw; mg GLU/100 gds, GLUs). The influence of L/S ratio was also studied over alginate content of extracts. Table 6-17 shows the results: Table 6-17. Effect of L/S ratio on alginate content of an extract obtained with FV35. ϯ L/S (g/g) GLUw GLUs 20 1102.36±157.33a 1742.40±248.68a 30 2069.58±106.82b 3300.77±170.36b 40 2527.50±435.26c 4141.82±649.00c ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. The extraction of alginates must be limited if it is carried out with a high concentration of solids. As low L/S ratios diminish the effect of ultrasound waves. Therefore, the highest glucuronic acid content was achieved with a L/S ratio of 40 (2527.50±435.26 GLUw; 4141.82±649.00 GLUs), followed by a L/S ratio of 30 (2069.58±106.82 GLUw; 3300.77±170.36 GLUs) and, the lowest was obtained for the system extracted with the highest solid content (1102.36±157.33 GLUw; 1742.40±248.68 GLUs). Results and discussion 81 6.4.2. Effect of drying temperature 6.4.2.1. Antioxidant activity To study the effect of drying temperature of F. vesiculosus on the antioxidant activity of the extracts, the extractions were carried out with a liquid/solid ratio (L/S) of 30, a residence time of 4 minutes and a rehydration time prior to extraction of 15 minutes. These conditions would provide the highest content in polyphenols, carbohydrates and alginates in the extracts. Table 6-18 displays the influence of drying temperature of algae on total polyphenol content of the extracts. TPrel is the relation between TPw content of the current system and seaweed formerly dried at 35ºC. Experiment was carried out in duplicate, with both systems (A and B) algae gathered and dried separately. Table 6-18. Total polyphenolic content of F. vesiculosus extracts (contact time of 4 minutes and L/S ratios of 30) formerly dried at different temperatures. ϯ System A System B T (ºC) TPw TPs TPrel TPw TPs TPrel 35 1571.21±75.53a 2940.44±141.35a 1.00 1738.46±93.72a 3514.21±189.46a 1.00 50 982.69±33.22b 2006.40±67.83b 0.63 1258.59±57.97b 2207.78±101.69b 0.72 60 943.38±65.76b 2056.22±143.34b 0.60 789.07±30.75c 1695.22±66.07c 0.45 75 847.50±73.36b 1670.23±144.57b 0.54 799.08±11.47c 1405.42±20.18c 0.46 ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. Both systems showed the same trend. TP decreases when drying temperature is raised, varying within a range of 847.50 - 1571.21 TPw for system A, and 799.08 - 1738.46 TPw for system B. The maximum TP was achieved for the extract made with seaweed dried at 35ºC (1571.21±75.53 TPw, system A; 1738.46±93.72 TPw, system B). Increasing drying temperature to 50ºC induces a reduction in TPw of 33±7% and, further increase to 60 or 75ºC, reduces it by 51±7%. Total polyphenolic content referred to total solids content in the extract shows the same trend and implies that differences in TP is attributed to the effect of drying temperature, and not to a difference in extraction yields. An extract (system A, FV35) was carried out with an acetone/water mixture (70/30 v/v) to achieve the highest extraction yield, replicating the rest of operation conditions. TP Results and discussion 82 attained was 11428.29±1123.83 TPw, which is within the interval of 8–13% of dry matter (reaching maximums close to winter) reported by Ragan & Jensen (1978), close to 10.5% as shown in Figure 3-8 or 16272 TPw, achieved by Díaz-Rubio et al. (2009), with an extraction with acetone/water and methanol/water mixtures. The maximum TP obtained employing only water as solvent in the ultrasound extraction accounts for 14.4±1% of TPw achieved with an acetone/water mixture. Acetone may contribute to a higher degradation of seaweed structure and, therefore, a higher release of these compounds. In addition, polyphenols exhibit a wide difference among their composition and structure and, as a result, in polarity. The use of water as only solvent allows the extraction of water soluble polyphenols, while the addition of acetone promotes the extraction of the non-polar fraction as well (López et al. 2011). DPPH radical scavenging activity of both systems is shown in Table 6-19: Table 6-19. Total DPPH radical scavenging activity after one hour for F. vesiculosus extracts (contact time of 4 minutes and L/S ratios of 30) formerly dried at different temperatures. System A System B T (ºC) Scavenging activity (%) Scavenging activity (%) 35 57.67±3.44a 41.68±1.24a 50 31.03±1.26b 27.51±1.11b 60 30.75±3.19b 19.49±0.94 b 75 26.02±1.68b 19.40±0.69b ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. Extracts obtained system FV35 exhibit the highest radical scavenging activity (57.67±3.44%, for system A; 41.68±1.24, for system B). As in the case of polyphenols, an increase in drying temperature results in a reduced radical scavenging activity. The lowest is that determined for the system FV75 (26.02±1.68%, for system A; 19.40±0.69, for system B). The evolution of radical scavenging activity over time is shown in Figure 6-27. Linear correlations (R2>0.98) were found between TPs and radical scavenging activity (Figure 6-27). The corresponding fittings are: the equation (Eq. 6-8) for system A and the equation (Eq. 6-9) for system B: % Radical scavenging activity=0.044·TPs−11.90 (Eq. 6-8) Results and discussion 83 % Radical scavenging activity=0.023·TPs+0.64 (Eq. 6-9) The correlation coefficients R2 for systems A and B were, respectively, 0.998 and 0.98. Total polyphenolic content and antioxidant activities are reduced when F. vesiculosus is dried at higher temperatures. Tello-Ireland et al. (2011) reported the loss of antioxidant activity when drying Gracilaria chilensis at high temperatures (70ºC). Gupta et al. (2011) observed a 30% decrease in TPC of Himanthalia elongata when dried at 40ºC for 24 hours in comparison with fresh seaweed. Polyphenols content is reduced as a combination of both, drying temperature and drying time. The reduction in polyphenols content and antioxidant activity at high drying temperatures may be due to several factors (Gupta et al. 2011; Tello-Ireland et al. 2011; Le Lann et al 2008): Release of phenolic compounds bound to cell wall during drying. Thermal degradation by oxidative enzymes. Phenolic compounds may rapidly degrade at drying temperatures above 40ºC. Binding of polyphenols to other substances (proteins) or alterations in their chemical structure which cannot be determined by methods employed. 0 10 20 30 40 50 60 70 80 90 100 010 20 30 40 50 60 (absinitial-absfinal)/absinitial) (%) Time (min) System A 0 10 20 30 40 50 60 70 80 90 100 010 20 30 40 50 60 (absinitial-absfinal)/absinitial (%) Time (min) System B 0 10 20 30 40 50 60 70 80 90 100 700 900 1100 1300 1500 1700 1900 Radical scavenging activity (%) TP (mg PHL/100 gdw) System A 0 10 20 30 40 50 60 70 80 90 100 700 900 1100 1300 1500 1700 1900 Radical scavenging activity (%) TP (mg PHL/100 gdw) System B Figure 6-27. Upper figures display radical scavenging activity over time for both systems formerly dried at different temperatures: 35ºC (—), 50ºC (—), and 60 ºC (—) and 75ºC (—). Lower figures show the correlation (Eq. 6-8) and (Eq. 6-9) between TPs and radical scavenging activity. Results and discussion 84 85 6.4.2.2. Carbohydrates content The effect of drying temperature was also evaluated over the carbohydrates content of extracts of F. vesiculosus (Table 6-20): Table 6-20. Effect of drying temperature on carbohydrates content of F. vesiculosus extracts (contact time of 4 minutes and L/S ratios of 30). ϯ T (ºC) CHOw CHOs 35 5152.44±259.58a 8217.61±414.01a 50 4591.55±89.88b 10282.56±510.86b 60 5192.44±523.57c 13485.25±1269.26c 75 5882.66±357.01c 12894.53±782.55c ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. Regarding carbohydrate yield referred to raw seaweed powder, carbohydrates content showed no significant differences when F. vesiculosus was dried at 35, 50, 60 or 75ºC (average value of 5204.22 CHOw) Carbohydrate content referred to total solids content increasingly varies within the range of 8217.61 (35ºC) and 12894.53 (75ºC) CHOs. Although extraction yields referred to raw seaweed powder are similar, there might be compounds already extracted or degraded during drying at 60 and 75ºC. These substances are not removed at lower drying temperatures and, hence, are extracted during sonication, increasing total solids content of the extract. An extraction with acetone/water (70/30 v/v) of FV35 was carried out in order to compare the yield of the carbohydrates extraction. The concentration of the carbohydrates in the acetone/water extraction was 11388.15±156.10 CHOs. System FV75 extract exhibited the highest yield in relation to the acetone/water extract (~52%). As most polysaccharides present in the seaweed are structural substances, a higher drying temperature (75ºC) might have contributed to a better subsequent extraction of carbohydrates. 6.4.2.3. Alginates content The alginate content of the extracts was determined for two systems: FV35 exhibited a lower content (2069.58±106.82 GLUw; 3300.77±170.36 GLUs) than FV 75 (3240.00±266.05 GLUw; 7101.93±583.17 GLUs). In contrast with antioxidant activity, alginate content in the extracts appears to be increased when seaweed drying operation is Results and discussion Results and discussion 86 carried out at high temperatures. Tello-Ireland et al. (2011) observed a similar behaviour with Gracilaria chilensis. Higher temperatures reduced the antioxidant activity of the seaweed but also increased the extraction yield of agar (a structural polysaccharide of red seaweeds). A higher drying temperature may allow an easier extraction of structural polysaccharides of algae. 6.4.3. Fractions screening Extracts of different F. vesiculosus size fractions (with the exception of lowest and higher particle diameter fractions) were carried out under the same experimental conditions of temperature screening. For both systems, seaweed powder initially dried at 35ºC was employed. TP values for both systems are displayed in Table 6-21: Table 6-21. Total polyphenolic content of different fractions of F. vesiculosus extracts (contact time of 4 minutes and L/S ratios of 30) formerly dried at different 35ºC. ϯ System A System B Dp (µm) TPw TPs TPw TPs <80 1221.70±55.82a 2.334±0.107a 915.92±157.13a 1390.58±238.57a 80-125 1672.24±12.54b 3.258±0.024b 2272.42±249.76b 3724.60±409.36b 125-200 1216.22±75.43a 2.412±0.150a 1706.85±162.86c 3094.63±162.86b >200 1041.61±26.29c 2.265±0.057a 829.40±81.70 a 1460.04±81.70a ϯ: Data are presented as means ± standard deviation. Data value of each parameter with different superscript letters in rows are significantly different, P ≤ 0.05. For both systems, the 80-125 and 125-200 µm fractions exhibited the highest TP (1672.24±12.54 TPw, for system A; 2272.42±249.76 TPw, for system B). In system A, <80 µm and >200 µm fractions showed a lower TP (1221.7±55.82 TPw for 40 – 63 µm fractions; 1216.22±75.43 TPw for 125 µm fraction). The fractions with lowest TP are those with the highest particle diameter (200 – 250 µm) with a total of 1041.61±26.29 mg PHL/100 gdw. Concerning system B the 125-200 µm fraction showed the second highest TP (1706.85±162.86 TPw). The biggest fraction (>200 µm) and the smallest (<80 µm) showed the lowest TP, with 829.40±81.70 and 915.92±157.13 TPw, respectively, without significant differences. Both systems presented the highest TP on the 80-125 and 125-200 µm fractions, while the smallest and the biggest fractions exhibited a reduction in TP. This could be attributed to the accumulation of inorganic substances, like salts or metals present in the Bibliography 93 doi:10.1016/j.jfca.2004.06.015. Kuda, T., Tsunekawa, M., Hishi, T., Araki, Y. 2005b. Antioxidant Properties of Dried `kayamo-Nori’, a Brown Alga Scytosiphon lomentaria (Scytosiphonales, Phaeophyceae). Food Chemistry 89 (4): 617–622. doi:10.1016/j.foodchem.2004.03.020. Landrum, J. T. 2009. Carotenoids: Physical, Chemical, and Biological Functions and Properties. CRC Press: Boca Raton, FL, USA. Le Lann, K., Jégou, C., Stiger-Pouvreau, V. 2008. Effect of different conditioning treatments on total phenolic content and antioxidant activities in two sargassacean species: comparison of the frondose Sargassum muticum (yendo) fensholt and the cylindrical Bifurcaria bifurcata R. Ross. Phycological Research 56 (4): 238–45. doi:10.1111/j.1440-1835.2008.00505.x. Lemus, R., Pérez, M., Andrés, A., Roco, T., Tello-Ireland, C. M., Vega-Gálvez, A. 2008. Kinetic Study of Dehydration and Desorption Isotherms of Red Alga Gracilaria. LWT - Food Science and Technology 41 (9): 1592–99. doi:10.1016/j.lwt.2007.10.011. Lesser, P. M. 2011. Advances in Marine Biology 58. Academic Press: London, England. Li, L. T. 1998. The Color of Food and Its Optical Properties. Physical Properties of Food, 324. Li, Y., Wijesekara, I., Li, Y., Kim, S. 2011. Phlorotannins as Bioactive Agents from Brown Algae. Process Biochemistry 46 (12): 2219–2124. doi:10.1016/j.procbio.2011.09.015. Lianfu, Z., Zelong, L. 2008. Optimization and Comparison of Ultrasound/microwave Assisted Extraction (UMAE) and Ultrasonic Assisted Extraction (UAE) of Lycopene from Tomatoes. Ultrasonics Sonochemistry 15 (5): 731–737. doi:10.1016/j.ultsonch.2007.12.001. López, A., Rico, M., Rivero, A., Suárez de Tangil, M. 2011. The Effects of Solvents on the Phenolic Contents and Antioxidant Activity of Stypocaulon scoparium Algae Extracts. Food Chemistry 125 (3): 1104–1109. doi:10.1016/j.foodchem.2010.09.101. Ma, Y., Chen, J., Liu, D., Ye, X. 2009. Simultaneous Extraction of Phenolic Compounds of Citrus Peel Extracts: Effect of Ultrasound. Ultrasonics Sonochemistry 16 (1): 57–62. doi:10.1016/j.ultsonch.2008.04.012. Moreira, R., Chenlo, F., Vázquez, M. J., Camean, P. 2005. Sorption Isotherms of Turnip Top Leaves and Stems in the Temperature Range from 298 to 328 K. Journal of Food Engineering. 71 (2): 193–99. Mujumdar, A. S. 2006. Handbook of Industrial Drying. CRC Press: Boca Raton, FL, USA. Bibliography 94 Ngo, D., Wijesekara, I., Vo, T., Ta, Q. V., Kim, S. 2011. Marine Food-Derived Functional Ingredients as Potential Antioxidants in the Food Industry: An Overview. Food Research International 44 (2): 523–29. doi:10.1016/j.foodres.2010.12.030. Oliver, M. J., O’Mahony, P., Wood, A. J. 1998. To Dryness and beyond - Preparation for the Dried State and Rehydration in Vegetative Desiccation - Tolerant Plants. Plant Growth Regulation. 1: 193–201. Oswin, C. R. 1946. The Kinetics of Package Life. III. The Isotherm. Journal of the Society of Chemical Industry 65 (12): 419–21. doi:10.1002/jctb.5000651216. Picó, Y. 2013. Ultrasound-Assisted Extraction for Food and Environmental Samples. TrAC Trends in Analytical Chemistry 43: 84–99. doi:10.1016/j.trac.2012.12.005. Polatoğlu, B., Beşe, V., Kaya, M., Aktaş, N. 2011. Moisture Adsorption Isotherms and Thermodynamics Properties of Sucuk (Turkish Dry-Fermented Sausage). Food and Bioproducts Processing. 89 (4): 449–456. doi:10.1016/j.fbp.2010.06.003. Ragan, M. A., Jensen, A. 1978. Quantitative Studies on Brown Algal Phenols. II. Seasonal Variation in Polyphenol Content of Ascophyllum nodosum (L.) Le Jol. and Fucus vesiculosus (L.). Journal of Experimental Marine Biology and Ecology 34 (3): 245–58. doi:http://dx.doi.org/10.1016/S0022-0981(78)80006-9. Rahman, M. S. 2006. Drying of Fish and Seafood. In Handbook of Industrial Drying 3rd Edition. CRC Press: Boca Raton, FL, USA. doi:10.1201/9781420017618.ch22. Rioux, L. E., Turgeon, S. L., Beaulieu, M. 2007. Characterization of Polysaccharides Extracted from Brown Seaweeds. Carbohydrate Polymers 69 (3): 530–537. doi:10.1016/j.carbpol.2007.01.009. Rodrigues, S., Pinto, G. S., Fernandes, F. N. 2008. Optimization of Ultrasound Extraction of Phenolic Compounds from Coconut (Cocos nucifera) Shell Powder by Response Surface Methodology. Ultrasonics Sonochemistry 15 (1): 95–100. doi:10.1016/j.ultsonch.2007.01.006. Shao, P., Chen, X., Sun, P. 2013. In Vitro Antioxidant and Antitumor Activities of Different Sulfated Polysaccharides Isolated from Three Algae. International Journal of Biological Macromolecules 62 (11): 155–161. doi:10.1016/j.ijbiomac.2013.08.023. Singleton, V. L., Rossi, J. A. 1965. Colorimetry of Total Phenolics with PhosphomolybdicPhosphotungstic Acid Reagents. American Journal of Enology and Viticulture 16 (3): 144–158. Bibliography 95 Soria, A. C., Villamiel, M. 2010. Effect of Ultrasound on the Technological Properties and Bioactivity of Food: A Review. Trends in Food Science & Technology 21 (7): 323– 331. doi:10.1016/j.tifs.2010.04.003. Stephen, M. A., Phillips, G. O., Williams, P. A. 2006. Food Polysaccharides and Their Applications. CRC Press: Boca Raton, FL, USA. Symons, G. E., Morey, B. 1941. The Effect of Drying Time on the Determination of Solids in Sewage and Sewage Sludges. Sewage Works Journal 13 (5): 936–939. Tatarenkov, A., Jönsson, R. B., Kautsky, L., Johannesson, K. 2007. Genetic Structure in Populations of Fucus vesiculosus (Phaeophyceae) Over Spatial Scales From 10 M To 800 Km. Journal of Phycology 43 (4): 675–685. doi:10.1111/j.1529-8817.2007.00369.x. Tello-Ireland, C., Lemus, R., Vega-Gálvez, A., López, J., Di Scala, K. 2011. Influence of Hot-Air Temperature on Drying Kinetics, Functional Properties, Colour, Phycobiliproteins, Antioxidant Capacity, Texture and Agar Yield of Alga Gracilaria chilensis. LWT - Food Science and Technology 44 (10): 2112–2118. doi:10.1016/j.lwt.2011.06.008.. Thuy, T. T. T., Ly, T., B. M.,Van, T. T. T., Quang, N. V., Tu, H. C., Zheng, Y., SeguinDevaux, C., Mi, B., Ai, U. 2015. Anti-HIV Activity of Fucoidans from Three Brown Seaweed Species. Carbohydrate Polymers 115 (1): 122–128. doi:10.1016/j.carbpol.2014.08.068. Timmermann, E. O. 2003. Multilayer Sorption Parameters: BET or GAB Values? Colloids and Surfaces A: Physicochemical and Engineering Aspects 220 (1-3): 235–260. doi:10.1016/S0927-7757(03)00059-1. Truus, K., Vaher, M., Taure, I. 2001. Algal Biomass from Fucus vesiculosus (Phaeophyceae): Investigation of the Mineral and Alginate Components. 1: 95–103. Van den Berg, C., Bruin, S. 1981. Water activity and its estimation in food systems: theoretical aspects. A Treatise on the Influence of Bound and Free Water on the Quality and Stability of Foods and Other Natural Products. pp 1-61. doi:http://dx.doi.org/10.1016/B978-0-12-591350-8.50007-3. Vega-Gálvez, A., Ayala-Aponte, A., Notte, E., De La Fuente, L., Lemus, R. 2008. Mathematical Modeling of Mass Transfer during Convective Dehydration of Brown Algae Macrocystis pyrifera. Drying Technology 26 (12): 1610–16. doi:10.1080/07373930802467532.. Wang, T., Jónsdóttir, R., Liu, H., Gu, L., Kristinsson, H. G., Raghavan, S., Olafsdóttir, G. 2012. Antioxidant Capacities of Phlorotannins Extracted from the Brown Algae Fucus vesiculosus. Journal of Agricultural and Food Chemistry 60 (23): 5874–5883. doi:10.1021/jf3003653. Bibliography 96 Wijesinghe, W., You-Jin J. 2012. Biological Activities and Potential Industrial Applications of Fucose Rich Sulfated Polysaccharides and Fucoidans Isolated from Brown Seaweeds: A Review. Carbohydrate Polymers 88 (1): 13–20. doi:10.1016/j.carbpol.2011.12.029. Wolf, W., Spiess, W. E. L., Jung, G. 1985. Standardization of Isotherm Measurements (CostProject 90 and 90 BIS). Properties of Water in Foods 90: 661-679. doi:10.1007/978-94-009-5103-7_40 Wrolstad, R. E., Decker, E. A., Schwartz, S. J., Sporns, P. 2005. Handbook of Food Analytical Chemistry, Water, Proteins, Enzymes, Lipids, and Carbohydrates. Handbook of Food Analytical Chemistry. John Wiley & Sons, New Jersey, USA. Zogzas, N. P., Maroulis, Z.B., Marinos-Kouris, D. 1996. Moisture Diffusivity Data Compilation in Foodstuffs. Drying Technology 14 (10): 2225–2253. doi:10.1080/07373939608917205. 97 9. Annex I: Calibration lines Calibration lines employed for the chemical characterization of extracts are shown below. 9.1. Total polyphenolic content The calibration line employed for the determination of total polyphenolic content is presented in (9-1): [mg PHLL ⁄ ]=[Abs]·114.94−5.03 (9-1) Where [mg PHL/L] is the TPC expressed as phloroglucinol equivalents and [Abs] is the absorbance (R2 of 0.992). This calibration line was applied within the range of 20-100 mg PHL/L. 9.2. Total carbohydrates content Carbohydrate content of the extracts, by means of glucose content, was determined using the following calibration line (9-2): [mg CHOL ⁄ ]=[Abs]·74.07+0.16 (9-2) Where [mg CHO/L] is the total carbohydrates content expressed as glucose equivalents and [Abs] is the absorbance (R2 of 0.992). This calibration line was applied within the range of 10-100 mg PHL/L. 9.3. Uronic acids content Alginates content is measured by means of glucuronic acids content (9-3): [mg GLUL ⁄ ]=[Abs]·208.33+4.00 (9-3) Where [mg GLU/L] is the uronic acids content expressed as glucuronic acid equivalents and [Abs] is the absorbance (R2 of 0.999). This calibration line was applied within the range of 20-330 mg GLU/L. *Each system assayed (FV35, FV50, FV60 and FV75) includes collection and drying with deep-bed and thin layer configurations. Extracts obtention include DPPH radical scavenging activity and total polyphenols content determination. Powder obtention includes its physical characterization. 98 10.Annex II: Gantt diagram 99 11. Annex III: List of figures FIGURE 3-1. HAWORTH PROJECTION OF ALGINATE.. .................................................................................................... 11 FIGURE 3-2. PROPOSED STRUCTURE OF FUCOIDANS FROM FUCUS VESICULOSUS ................................................................ 12 FIGURE 3-3. BIOLOGICAL PROPERTIES AND POTENTIAL INDUSTRIAL USES OF FUCOIDANS. .................................................... 12 FIGURE 3-4. CLASSIFICATION OF SOME PHENOLIC COMPOUNDS FROM NATURAL SOURCES. .................................................. 14 FIGURE 3-5. STRUCTURE OF PHLOROTANNINS FROM MARINE BROWN ALGAE.. .................................................................. 15 FIGURE 3-6. STRUCTURE OF FUCOXANTHIN, CAROTENOIDS FOUND IN BROWN ALGAE. ....................................................... 16 FIGURE 3-7. FUCUS VESICULOSUS UNDERWATER. ........................................................................................................ 16 FIGURE 3-8. AVERAGE COMPOSITION OF FUCUS VESICULOSUS IN DRY WEIGHT .................................................................. 17 FIGURE 3-9. ILLUSTRATION OF THE IMPACT OF THE COLOUR OF A PARTICULAR NATURAL FOOD PRODUCT UPON ‘LIKE OR DISLIKE’ 19 FIGURE 3-10. TYPICAL ISOTHERM SHAPES FOR FOOD SYSTEMS ....................................................................................... 21 FIGURE 3-11. DRYING CURVE FOR HIGH MOISTURE MATERIAL ....................................................................................... 22 FIGURE 3-12. CAVITATION FORMATION DURING SONICATION ........................................................................................ 23 FIGURE 3-13. ULTRASOUND ASSISTED EXTRACTION. .................................................................................................... 24 FIGURE 4-1. CIE-LAB COLOUR EXPRESSION SYSTEM ..................................................................................................... 33 FIGURE 4-2. ULTRASOUND TRANSDUCER AND GENERATOR MODEL UIP – 1000 HDT ......................................................... 36 FIGURE 4-3. FURAN DERIVATIVES ............................................................................................................................. 38 FIGURE 4-4. PHENOL–SULPHURIC ACID ASSAY ABSORBANCE MAXIMA FOR HEXOSES AND PENTOSES ...................................... 38 FIGURE 5-1. WORK FLOW DIAGRAM AND SYSTEMS ASSAYED. ........................................................................................ 40 FIGURE 6-1. EXPERIMENTAL DATA OF WATER ADSORPTION AND DESORPTION ISOTHERMS. .................................................. 42 FIGURE 6-2. EXPERIMENTAL DATA OF WATER SORPTION ISOTHERMS AND BET MODEL ADJUSTMENT ..................................... 44 FIGURE 6-3. EXPERIMENTAL DATA OF SORPTION ISOTHERMS ALONG WITH GAB AND CAURIE MODELS .................................. 45 FIGURE 6-4. EXPERIMENTAL DATA OF SORPTION ISOTHERMS ALONG WITH HALSEY AND OSWIN MODELS ............................... 46 FIGURE 6-5. LINEAR AND ARRHENIUS CORRELATIONS OF HALSEY MODEL PARAMETERSAT DIFFERENT TEMPERATURES. .............. 48 FIGURE 6-6. EXPERIMENTAL DATA OF SORPTION ISOTHERMS ALONG WITH MODIFIED HALSEY MODEL .................................... 50 FIGURE 6-7. EXPERIMENTAL DRYING CURVES FOR FUCUS VESICULOSUS (DEEP-BED CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES ........................................................................................................................................... 51 FIGURE 6-8. DRYING RATE CURVES FOR F.VESICULOSUS (DEEP-BED CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES ..... 52 FIGURE 6-9. EXPERIMENTAL DRYING CURVES FOR FUCUS VESICULOSUS (DEEP-BED CONFIGURATION) AND MODELS ADJUSTMENTS AT DIFFERENT DRYING TEMPERATURES ............................................................................................................. 55 FIGURE 6-10. EXPERIMENTAL DRYING CURVES FOR FUCUS VESICULOSUS (DEEP-BED CONFIGURATION) AND MODELS ADJUSTMENTS AT DIFFERENT DRYING TEMPERATURES ............................................................................................................. 56 FIGURE 6-11. DRYING RATE CURVES FOR FUCUS VESICULOSUS (DEEP CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES ... 57 FIGURE 6-12. EXPERIMENTAL DRYING CURVES FOR FUCUS VESICULOSUS (THIN LAYER CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES ........................................................................................................................................... 58 FIGURE 6-13. DRYING RATE CURVES FOR FUCUS VESICULOSUS (THIN LAYER CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES. .......................................................................................................................................... 59 Annex III: List of figures 100 FIGURE 6-14. VOLUMETRIC SHRINKAGE DURING CONVECTIVE DRYING OF FUCUS VESICULOSUS AT DIFFERENT TEMPERATURES .... 60 FIGURE 6-15. VOLUME SHRINKAGE VS MOISTURE CONTENT AT DIFFERENT TEMPERATURES ................................................. 61 FIGURE 6-16. EFFECT OF SHRINKAGE ON DRYING RATE CURVES FOR FUCUS VESICULOSUS (THIN LAYER CONFIGURATION) AT DIFFERENT DRYING TEMPERATURES ................................................................................................................. 61 FIGURE 6-17. FITTING OF WATER COEFFICIENT OF DIFFUSION WITH TEMPERATURE BY MEANS OF LINEAR CORRELATION ............ 64 FIGURE 6-18. FITTING OF WATER COEFFICIENT OF DIFFUSION WITH TEMPERATURE BY MEANS OF ARRHENIUS CORRELATION ...... 64 FIGURE 6-19. EFFECT OF LOAD DENSITY ON DRYING TIME REQUIRED TO ACHIEVE A MOISTURE RATIO OF 0.1. .......................... 65 FIGURE 6-20. EFFECT OF LOAD DENSITY ON DRYING CURVES ......................................................................................... 66 FIGURE 6-21. EFFECT OF LOAD DENSITY ON DRYING RATE CURVES .................................................................................. 67 FIGURE 6-22. PARTICLE SIZE DISTRIBUTION FOR F. VESICULOSUS POWDER FORMERLY DRIED AT DIFFERENT TEMPERATURES ....... 70 FIGURE 6-23. CUMULATIVE PARTICLE SIZE DISTRIBUTION FOR F. VESICULOSUS POWDER FORMERLY DRIED AT DIFFERENT TEMPERATURES ........................................................................................................................................... 70 FIGURE 6-24. SEAWEED POWDER MIXTURE FORMERLY DRIED AT DIFFERENT TEMPERATURES ............................................... 72 FIGURE 6-25. SEAWEED DURING INITIAL, LATE AND FINAL STAGES OF DRYING AT 75ºC. ...................................................... 75 FIGURE 6-26. COLOUR PARAMETERS FOR F. VESICULOSUS SIZE FRACTIONS PREVIOUSLY DRIED AT DIFFERENT TEMPERATURES ..... 77 FIGURE 6-27. RADICAL SCAVENGING ACTIVITY OVER TIME FOR SYSTEMS A AND B FORMERLY DRIED AT DIFFERENT TEMPERATURES AND LINEAR CORRELATION BETWEEN TPS AND RADICAL SCAVENGING ACTIVITY. ....................................................... 84 101 12. Annex IV: List of tables TABLE 3-1. CLASSIFICATION OF THE MOST IMPORTANT POLYSACCHARIDES FOR HUMAN CONSUMPTION PRESENT IN ALGAE .......... 9 TABLE 4-1. REAGENTS EMPLOYED DURING EXPERIMENTATION, APPLICATION AND SAFETY DATA SHEET. ................................. 25 TABLE 4-2. SALTS USED IN ACQUEOUS SATURATED SOLUTIONS AND THE WATER ACTIVITY GENERATED. .................................. 27 TABLE 4-3. CONDITIONS FOR THE APPLICATION OF SHORT OR LONG-TIME FICK’S SECOND LAW DIFFUSION EQUATIONS .............. 32 TABLE 4-4. COLOUR DIFFERENCE EVALUATION ............................................................................................................ 34 TABLE 6-1. MODEL PARAMETERS FOR ADSORPTION AND DESORPTION ISOTHERMS AT 5, 25, 45 AND 65ºC. .......................... 43 TABLE 6-2. LINEAR AND ARRHENIUS CORRELATIONS PARAMETERS. ................................................................................. 47 TABLE 6-3. DRYING MODELS FITTING PARAMETERS FOR EACH TEMPERATURE .................................................................... 53 TABLE 6-4. PARAMETERS FOR THE MODELLING OF PROGRESSION OF VOLUME WITH MOISTURE CONTENT AT DIFFERENT TEMPERATURES. .......................................................................................................................................... 60 TABLE 6-5. CRITICAL MOISTURE CONTENT (XC) AT DIFFERENT DRYING TEMPERATURES OF F. VESICULOSUS. ............................. 62 TABLE 6-6. ESTIMATION OF WATER EFFECTIVE COEFFICIENT OF DIFFUSION FOR FUCUS VESICULOSUS AT DIFFERENT DRYING TEMPERATURES. ......................................................................................................................................... 63 TABLE 6-7. LINEAR AND ARRHENIUS FITTING PARAMETERS OF EFFECTIVE DIFFUSIVITY COEFFICIENTS AT DIFFERENT DRYING TEMPERATURES. .......................................................................................................................................... 63 TABLE 6-8. EFFECT OF LOAD DENSITY ON PAGE MODEL PARAMETERS. ............................................................................. 65 TABLE 6-9. COLORIMETRIC PARAMETERS OF FRESH AND DRIED (AT DIFFERENT TEMPERATURES) FUCUS VESICULOSUS SEAWEED. 68 TABLE 6-10. SIZE DISTRIBUTION FOR SEAWEED POWDER FORMERLY DRIED AT DIFFERENT TEMPERATURES. ............................ 69 TABLE 6-11. MEAN DIAMETERS OF MILLED AND SIEVED F. VESICULOSUS PREVIOUSLY DRIED AT DIFFERENT TEMPERATURES. Ϯ .... 71 TABLE 6-12. COLOUR PARAMETERS OF SEAWEED POWDER FORMERLY DRIED AT 35, 50, 60 AND 75ºC AND MILLED WITH A MESH SIZE OF 500 µM AND THE CORRESPONDING SIZE FRACTIONS. ............................................................................... 73 TABLE 6-13. EFFECT OF THE VARIATION IN L/S RATIO ON TP FOR FV35 SYSTEM. ............................................................. 78 TABLE 6-14. EFFECT OF CONTACT TIME ON TP CONTENT OF EXTRACTS OBTAINED WITH SYSTEM FV35. ................................. 79 TABLE 6-15. EFFECT OF THE VARIATION IN L/S RATIO ON CARBOHYDRATE CONTENT FOR FV35 SYSTEM. ............................... 79 TABLE 6-16. EFFECT OF CONTACT TIME ON CARBOHYDRATE CONTENT OF EXTRACTS. ......................................................... 80 TABLE 6-17. EFFECT OF L/S RATIO ON ALGINATE CONTENT OF AN EXTRACT OBTAINED WITH FV35. ...................................... 80 TABLE 6-18. TOTAL POLYPHENOLIC CONTENT OF F. VESICULOSUS EXTRACTS (CONTACT TIME OF 4 MINUTES AND L/S RATIOS OF 30) FORMERLY DRIED AT DIFFERENT TEMPERATURES........................................................................................... 81 TABLE 6-19. TOTAL DPPH RADICAL SCAVENGING ACTIVITY AFTER ONE HOUR FOR F. VESICULOSUS EXTRACTS (CONTACT TIME OF 4 MINUTES AND L/S RATIOS OF 30) FORMERLY DRIED AT DIFFERENT TEMPERATURES. ............................................... 82 TABLE 6-20. EFFECT OF DRYING TEMPERATURE ON CARBOHYDRATES CONTENT OF F. VESICULOSUS EXTRACTS (CONTACT TIME OF 4 MINUTES AND L/S RATIOS OF 30). ................................................................................................................. 85 TABLE 6-21. TOTAL POLYPHENOLIC CONTENT OF DIFFERENT FRACTIONS OF F. VESICULOSUS EXTRACTS (CONTACT TIME OF 4 MINUTES AND L/S RATIOS OF 30) FORMERLY DRIED AT DIFFERENT 35ºC. ............................................................... 86