|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 DEVELOPMENT AND EVALUATION OF A NATURAL CONVECTION SOLAR DRYER FOR MANGO IN RURAL HAITIAN COMMUNITIES By DREW FRANK SCHIAVONE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Drew Schiavone
3 To my wife, Jessica, who lovingly supported and encouraged me throughout the long hours of work and preparation involved with this study
4 ACKNOWLEDGMENTS I would like to thank my advisor Dr. Teixeira as well as my committee members, Dr. Bucklin and Dr. Sargent, for their support and guidance throughout my research. I would also like to thank the technicians whose work allowed for the construction of the prototype system. The assistance, suggestions and engineering work that James Rummel and Steve Feagle conducted w ere instrumental in accomplishing this study. Furthermore, I am thankful for the support of WINNER through the UF/IFA S Office of International Programs i n allowing this project to be undertaken. Finally, I would like to thank all my family and friends who supported me during this work with their presence and encouragement. I could not have made it through this degree program without the support of everyone mentioned here and the blessings lavished on me by Jesus Christ, who is my Lord and Savior and the F oundation and P erfector of my faith. None of this would have ever been possible without the Author of Salvation.
5 TABLE OF CONTENTS P age ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 8 LIST OF FIGURES .......................................................................................................... 9 LIST OF ABBREVIATIONS ........................................................................................... 13 ABSTRACT ................................................................................................................... 16 CHAPTER 1 INTRODUCTION .................................................................................................... 18 2 O BJECTIVES ......................................................................................................... 22 Task 1: Mango Fruit ................................................................................................ 22 Task 2: Dryer Design .............................................................................................. 22 3 LITERATURE REVIEW OF SOLAR DRYERS ....................................................... 23 Background ............................................................................................................. 23 Post Harvest Losses ......................................................................................... 23 Biological Degradation ...................................................................................... 24 Drying Rev iew .................................................................................................. 25 Traditional Drying Methods ............................................................................... 30 Industrial Drying Methods ................................................................................. 32 Solar Drying Methods ....................................................................................... 33 Solar Dryers ............................................................................................................ 36 Active and Passive Mode ................................................................................. 36 Direct Mode Dryers .......................................................................................... 38 Indirect Mode Dryers ........................................................................................ 41 LargeScale Dryers .......................................................................................... 51 Case Studies .......................................................................................................... 53 Direct Solar Dryer Designs ............................................................................... 55 Indirect Cabinet Designs (Passive) .................................................................. 57 Indirect Cabinet Designs (Active) ..................................................................... 59 Supplemental Heat Dryers ............................................................................... 63 Desiccant Integrated Dryers ............................................................................. 70 Indirect Tunnel Dryers ...................................................................................... 71 LargeScale Solar Dryers ................................................................................. 74 Construction ............................................................................................................ 74 Evaluation ............................................................................................................... 75 Product Evaluation ........................................................................................... 76
6 Dryer Parameter Evaluation ............................................................................. 79 Performance Evaluation ................................................................................... 81 Evaluation Procedure ....................................................................................... 84 4 MATERIALS AND METHODS .............................................................................. 143 Task 1: Sorption Isotherm ..................................................................................... 143 Task 2: Dryer Design ............................................................................................ 148 Design Features and Considerations ............................................................. 149 Mathematical Procedure ................................................................................. 149 Construction .......................................................................................................... 156 Absorber ......................................................................................................... 156 Cabinet ........................................................................................................... 158 Dryer Operation .................................................................................................... 159 General Overview ........................................................................................... 159 No Load .......................................................................................................... 159 Load ............................................................................................................... 160 Evaluation ....................................................................................................... 162 5 RESULTS AND DISCUS SION ............................................................................. 168 Sorption Isotherm ................................................................................................. 168 Solar Dryer Design ................................................................................................ 169 Dryer Operation .................................................................................................... 169 No Load .......................................................................................................... 169 Load Experiments .......................................................................................... 172 Evaluation ............................................................................................................. 176 Dryer Efficiency ..................................................................................................... 176 Product Qualit y ..................................................................................................... 176 6 CONCLUSION ...................................................................................................... 189 7 FUTURE DEVELOPMENTS ................................................................................. 191 APPENDIX A D RYING PARAMETERS ...................................................................................... 193 B STANDARD OPERATING PROCEDURES .......................................................... 194 Product Preparation .............................................................................................. 194 Dryer Operation .................................................................................................... 194 Maintenance/Cleaning .......................................................................................... 195 C OPERATIN G SCHEDULE .................................................................................... 196 LIST OF REFERENCES ............................................................................................. 197
7 BIOGRAPHICAL SKETCH .......................................................................................... 205
8 LIST OF TABLES Table P age 3 1 Estimated postharvest losses of fresh produce in developed and developing countries ........................................................................................... 94 3 2 Moisture absorption capability ............................................................................ 95 3 3 Commonly used materials for solar dryers in develo ping countries .................. 142 3 4 Optimum tilt angles for solar collectors ............................................................. 142 4 1 Design parameter input for natural convection solar dryer ............................... 165 5 1 Estimated parameters of sorption isotherm models of mango slices ................ 179 5 2 Design parameter output for natural convection solar dryer ............................. 179 A 1 Mango drying parameters and storage conditions ............................................ 193
9 LIST OF FIGURES Figure P age 3 1 Drying curve ....................................................................................................... 94 3 2 Psychrometric chart showing a drying process ................................................... 95 3 3 Tr aditional open air sun drying ........................................................................... 96 3 4 Working principle of direct solar dryer ................................................................ 96 3 5 Working principle of indirect, active solar dryer .................................................. 97 3 6 General classi fication of solar dryers .................................................................. 98 3 7 Working principle of direct type, solar dryer ........................................................ 99 3 8 Typical direct type, tent solar dryer ..................................................................... 99 3 9 Typical direct type, seesaw dryer ..................................................................... 100 3 10 Typical direct type, box dryer ............................................................................ 100 3 11 Typic al indirect type cabinet dryer .................................................................... 101 3 12 Airflow principles in assorted solar collectors ................................................... 101 3 13 Typical indirect type, tunnel dryer ..................................................................... 102 3 14 Typical large scale, greenhouse solar dryer ..................................................... 102 3 15 Roof i ntegrated solar drying system ................................................................. 103 3 16 Drying cabinet and solar collector of in house dryer ......................................... 103 3 17 Solar dryer categorization overview .................................................................. 104 3 18 Schematic of direct, box dryer ......................................................................... 105 3 19 Box dryer with H2O heater ................................................................................ 106 3 20 Box dryer with limited tracking .......................................................................... 107 3 21 D etailed schematic of box dyer ......................................................................... 108 3 22 Box dr yer with variable inclination .................................................................... 109 3 23 Box dryer. A) Pictorial. B) Side view schematic ................................................ 110
10 3 24 Simple cabinet dryer ......................................................................................... 111 3 25 Cabinet dryer schematic ................................................................................... 112 3 26 Cabinet dryer w ith multiple solar collectors ....................................................... 113 3 27 Reverse absor ber cabinet dryer, RACD ........................................................... 114 3 28 Cabinet dryer wit h top absorber ........................................................................ 115 3 29 Simple active cabinet dryer, schematic ............................................................ 116 3 30 Active cabinet dryer, schematic ........................................................................ 117 3 31 Active cabinet dryer with absorber mesh, schematic ........................................ 118 3 32 Acti ve cabinet dryer with piping ........................................................................ 119 3 33 Rotary co lumn cylindrical dryer, RCCD ............................................................ 120 3 34 Largescale ca binet dryer with heater array ...................................................... 121 3 35 Cabinet dryer with thermal, grave l storage ....................................................... 122 3 36 Cabinet dryer with thermal, rock storage .......................................................... 123 3 37 Cabinet dry er with thermal PCM storage ........................................................ 124 3 38 Cabinet dryer with thermal, silica gel storage ................................................... 125 3 39 Cabinet dryer with intercha ngeable thermal, sand storage ............................... 126 3 40 Cabinet dryer with auxiliary heater ................................................................... 127 3 41 Cabinet dryer wi th auxiliary heating channel .................................................... 128 3 42 Cabinet dryer with h eating elements, schematic .............................................. 129 3 43 Cabinet dryer wi th exhaust recirculation ........................................................... 130 3 44 Indirect c abin et dryer with biomass burner ....................................................... 131 3 45 Direct c abi net dryer with biomass burner ........................................................... 132 3 46 Cabinet dryer with biomas s burner and thermal storage .................................... 133 3 47 Forced c abinet dr yer with integrated desiccant ................................................ 134 3 48 Cabinet dryer with integrated desiccant ............................................................ 135
11 3 49 Cabinet dryer with int egrated desiccant and mirror .......................................... 136 3 50 Simple t unnel dryer schematic .......................................................................... 137 3 51 Tunnel dryer schematic .................................................................................... 138 3 52 T unnel dryer with biomass burner .................................................................... 139 3 53 Greenhouse dryer schematics .......................................................................... 140 3 54 Greenhouse dryer ............................................................................................. 141 4 1 Natural convection solar dryer featuring a solar collector and separate drying cabinet with five trays ....................................................................................... 164 4 2 Schematic of solar collector, cross section ....................................................... 166 4 3 Solar convection dryer on UF campus in Gainesville, FL ................................. 166 4 4 Trays loaded within the cabinet section of the solar convection dryer .............. 167 5 1 Sorption isotherm for mango at 71 F (22 C) ........................................................ 178 5 2 Solar collector schematic .................................................................................. 180 5 3 Drying cabinet with chimney, shutter and trays (schematic) ............................. 180 5 4 Incident solar radiation recorded at weather station ......................................... 181 5 5 Temperatures recorded within the solar dryer ................................................. 182 5 6 Cabinet temperatures based on average temperatures between all trays ........ 183 5 7 Relative humidity as recorded near chimney exhaust ...................................... 183 5 8 Exhaust air velocity recorded at chimney exhaust ............................................ 184 5 9 Temperatures within the solar dryer at various locations .................................. 184 5 10 Temperatures of incoming air in the plenum and rock bed ............................... 185 5 11 Temperatures recorded at the exhaust outlet ................................................... 185 5 12 Relative humidity of ambient air as recorded at weather station and in chimney/exhaust ............................................................................................... 186 5 13 Moisture content of mango slices in solar dryer over 2 full days of sunlight with batch mode of operation ............................................................................ 186
12 5 14 Moisture content of mango slices in solar dryer over 2 full days of sunlight with continuous mode of operation ................................................................... 187 5 15 Photograph of solar dried mangoes compared with commercially available mango slices ..................................................................................................... 188
13 LIST OF ABBREVIATION S A aperture area of dryer cross sectional flow area Ac solar collector area Acs crosssectional, tray area aw water activity B surface heat constant C constant discharge coefficient Cp specific heat of air D diffusion coefficient d slice thickness E total useful energy G mass flow rate of air per unit of collector area g gravitational acceleration H height (head) h vertical dista nce between trays height of chimney has absolute humidity of the air entering dryer at the point of adiabatic saturation hfinal final enthalpy hi absolute humidity of air entering the drying chamber hinitial initial enthalpy I total solar energy incident upon plane of collector per unit time per unit area k1, kc rate constant of constant rate period k2, kE rate constant of falling rate period L latent heat of vaporization of water length of sample
14 MC final moisture concentration of constant rate period ME final moisture concentration of falling rate period MEQ equilibrium moisture content Mf final moisture content (wb) Mi initial moisture content (wb) initial moisture concentration MO monolayer moisture M(t) absorbed moisture concentration ma airflow rate mdr average drying rate mfruit weight of whole fruit mp weight of sliced fruit mw mass of water to evaporate Nfruit number of whole fruit Ntrays number of trays n numbered day of year PATM ambient air pressure Pi internal air pressure Q chimney effect (flow rate) Tatm, TATM ambient air temperature TAVG average temperature between To and Ti Tda temperature of drying air Tmax maximum allowable temperature To absorber outlet temperature, average ambient temperature Ti a bsorber inlet temperature, average in ternal temperature t, td drying time (sunshine hours)
15 tC drying time of constant rate period tE drying time of falling rate period vwind wind speed V volumetric air flow rate Va volumetric airflow rate W weight of water evaporated from the product Wf final humidity ratio Wi initial humidity ratio collector angle/slope angle of declination air pressure difference COLLECTOR collector efficiency PICK -UP pickup efficiency SYSTEM system efficiency air density of air latitude atm ambient relative humidity f equilibrium relative humidity
16 A bstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DEVELOPMENT AND EVALUATION OF A NATURAL CONVECTION SOLAR DRY ER FOR MANGO IN RURAL HAITIAN COMMUNITIES By Drew Frank Schiavone December 2011 Chair: Arthur Teixeira Major: Agricultural and Biological Engineering A natural convection solar dryer capable of producing d ried mango slices in rural communities of Haiti was designed and evaluated. Studies of both the mango fruit and the mathematical design of the drying system were undertaken to develop an adequate and effective system capable of preserving mango fruit slices. The design, construction and operation of this newly developed dryer is described in this report based on a variety of factors including the physical properties of mangoes, Haitian environmental conditions and the local production capabilities of small Haitian villages. Moisture equilibrium data for desorption of water from mango slices were used for the development of a sorption isotherm to determine target levels of moisture content and water activity. A mathematical procedure for solar dryer design was adopted and modified for the development of this dryer and schematics were created based on the results. The dryer consists of two primary components; the solar collector which produces thermal energy and the drying chamber which houses five trays loaded with product. Fabrication of the dryer was carried out using locally available material such as plywood for the frame, a corrugated aluminum absorber, polycarbonate glazing, rock wool insulation and wire mesh for trays. The 18.4 ft2 solar collector was designed
17 to allow flow on both sides of the absorber. A 0.67 x 0.67 x 2.2 ft3 chimney was built into the upper portion of the 2 x 2 x 3.125 ft3 cabinet to dissipate heat and moisture with a thin steel shutter for airflow regulation. Additionally, a 2 ft3 thermal rock bed was integrated into the bottom of the cabinet to provide heat during inclement weather and during the night. Evaluation of the solar dryer in Gainesville, Florida found temperatures inside the cabinet significantly elevated compared to env ironmental air with temperature increases of up to 5 8.3 F (32.4 C) and 72.1 F (40.0 C) depending on airflow and loading. Loading tests conducted with an average of 21.6lbs (9.80kg) of fresh mango slices resulted in effective drying within two days from a moisture content of 84% (wb) down to approximately 9.4% (wb) and 11.1% (wb) for batch and continuous modes of operation respectively. The collector efficiency, drying efficiency and system efficiency were 29. 5 %, 10. 8 % and 33.9% respectively. These results indicated sufficient drying and preservation of mango slices within two full days of sunlight. The quality of solar dried product was competitive with commercially available mango slices.
18 CHAPTER 1 INTRODUCTION Many rural areas in developing nations suffer substantial losses of vital agricultural products. In fact, it has been reported that Haiti loses between 20 to 40% of their mango harvest each year (Castaeda et al., 2011; Lush, 2010). While there are several contributing factors which account for this loss, spoilage due to insufficient postharvest management plays a key role particularly in provincial communities. Mechanical injury of the fruit actually begins with inadequate and limited harvesting practices and continues throughout transportation and sorting processes (Medlicott, 2001; Yahia, 1999). Additionally, the fruit is quickly lost to fungal and microbial degradation due to the high humi dity levels associated with tropical environments (Dauthy, 1995; Enebe and Ezekoye, 2006). Limited resources and a complete lack of electricity in many rural locations, leaves farmers with few options to preserve their crops. While education and improvement of proper postharvest practices can be of benefit, processing operations are needed in close proximity to the harvesting areas in order to minimize losses associated with spoilage (Singh, 1994). While sophisticated cooling and mechanizeddrying methods have high rates of performance (Buteau 2009; Chua and Chou, 2003), they are unfeasible due to the energy requirements which are unavailable in many rural communities. In contrast, traditional openair and smoke drying preservation methods result in contam ination by animals, insects, dust, and microorganisms due to the extensive environmental exposure times of the products (Bakeka and Bilgen, 2008; Bhandari et al., 2005; Kandpal et al., 2006). Developing countries with humid environments and limited resources like Haiti, pose unique
19 challenges in quickly and efficiently removing moisture from agricultural crops to reach adequate preser vation standards by drying. For these reasons, the use of a natural convection type, solar dryer is proposed for rural communities of Haiti. The operation of solar dryers is dependent entirely on solar energy, which is a widely available resource in tropi cal communities. In principle, air is heated by solar radiation and naturally circulated by pressure gradients which promote vertical airflow. Consequently, these dryers require no electrical or mechanical components because the natural convection driving force is based only on temperature difference or changes in air density. Thus, product quality can be improved while reducing wasted produce and minimizing the use of traditional fuels. For thes e reasons solar dryers are often considered more effective than sun drying with operating costs generally lower than mechanized dryers (Chen et al., 2009). The ease of construction of previously reported designs has indicated the potential for adoption into small, rural communities where financial and material resources are limited (Chen et al., 2009). Studies further show that prolonged product shelf life is achieved while significantly reducing both the product volume and weight. By establishing these conditions, packaging, storing and transportation costs are effectively minimized (Chaudhri et al., 2009). The controlled environment of solar dryers also ensures that food remains unaffected by water intrusion, convective heat losses and contamination by foreign particles thereby reducing the likelihood of fungal and microbial growth. Additionally, these advantages benefit the end user by limiting the work needed for protecting the crop from these threats. Moreover, solar drying results in improved
20 quality which enhances the product marketability. Thus, the application of this drying process allows for improved financial opportunities for farmers compared with traditional drying methods. To this end, the application of solar dryers has been shown to be practical, economical and environmentally responsible in the conser vation of agricultural products (Buchinger and Weiss, 2002). However, significant amounts of mango fruit continue to spoil under mango groves in remote areas of Haiti despite the potential of solar drying technology. This may be attributed to the development of inappropriate solar dryer designs, elevated construction costs or the inaccessibility of resources needed for construction (Akoy et al., 2006). To promote drying processes in these developing communities, an inexpensive, natural convection solar dryer capable of producing dried mango slices was proposed for development. An ideal region for implementation of this system has been identif ied as the mountainous, Saut d Eau region of Haiti which is a major mango harvesting area outside of the Haitian power grid. Considering the limited resources in rural locations, the widely available energy source of solar radiation has significant potential for implementation. This resource can easily undergo conversion into low grade heat for evaporation of moisture from agricultural products. In this manner, moisture content can be reduced until deterioration is effectively slowed. Air flow must also be established by the natural convection principle to remove heated water vapor and help prevent accumulation of emerging vapors on the product surface. The application of drying systems in developing communities can greatly reduce post harvest losses of agricultural commodities and significantly contribute to food availability in these areas. Additionally, this technology has potential to generate local
21 employment opportunities depending on the size and scale of operation. Increased revenue is also possible as otherwise spoiled product is dried and either sold domestically or exported. This is a significant realization consi dering that dried mango is currently the highest value mango product according to a recent assessment of the Haitian mango industry (Castaeda et al., 2011). All of these factors must be recognized in order to meet the needs as well as the limitations of d eveloping communities.
22 CHAPTER 2 OBJECTIVES The overall objective of this project was to design, construct, and make operational a natural convection type, solar dryer for use in the tropical weather conditions of rural Haitian communities. To accomplish this task, studies of both the mango fruit and the drying system were undertaken to develop an adequate preservation system. Each of these project components was comprised of integral steps, procedures, and considerations which required meeting the following specific directives Task 1 : Mango Fruit The objective of Task 1 was to construct a sorption isotherm for mango fruit slices from which the target values of moisture content and water activity could be determined. These target values were to be determined by analysis of commercially available, dried mango slices serving as a standard. Equilibrium moisture data was also collected for fresh mango. An appropriate sorption isotherm model describing the equilibrium moisture data was then determined. Task 2: Dryer Design The objective of Task 2 was to design, fabricate and operate a natural convection, solar dryer, and evaluate its operating performance in the dehydration of mango fruit slices.
23 CHAPTER 3 LITERATURE REVIEW OF SOLAR DRYERS Background Post Harvest Losses Compared t o industrial i zed nat ions su c h as t he United Sta t es, many rural areas and deve l oping nat ions suf f er su b sta n tial pos t harvest losses of v i tal agricultural p rodu ct s. In fa c t po s t h arvest losses in deve l opi n g coun tries are estimated o ver a w i de r ange, genera l l y in t he o rder of 4 0% but in c ertain a reas t h ese losses may ex c eed 5 0% (National Academy of S ci ences, 1 9 7 8). Whi l e si g n if i ca n t pos t harvest losses do ex i st in deve l oped co unt ri e s, t h e p r i m ary differen c e in d eve l oping c ommu n it i es is t hat more of t h e prod u c t is l ost before reac h ing re t ail sites as sh own in T a ble 3 1 H i gh m o isture c ont ent p r od u c ts, such as fr u it a n d ve g eta b l e s, are q uickly lost t o fungal and m i cr obial d e g radat ion in t hese developing c ommu n it i e s d ue t o the lack o f a p prop riate p r eservation a n d st o rage systems. This d egradation beg ins shor t ly af t er h arvesting and is great l y increased in areas with hi g h l e v e l s o f h um i dit y T h e development of proc ess i ng fa c i l ities in c lose pr o x i m i ty to h arvesting areas is c r u c ial t o reduc ing t hese losses. Howe v er, l i m ited res o ur c es and a co mp l ete lack o f elec t ricity in many rural areas, leave farmers w ith f ew o p tions to preserve t heir cr ops. Ac c ording t o Bhan d ari et al. ( 2 005), the pri m ary requ i r ement of agricultural producers is to secure a surplus of fresh produce from being wasted through spoilage, so that it c an be p reserved for extended p er i ods of time. T his al l ows for co nsumpt i on of t h e p r eserved pr o duct in offseas o ns as we l l as inc r eas ing t he marke t ability o f t he p rodu c t which serves to u pl i ft t h e lo c al ec o nomy. Add i tionally t h is agr i cu ltural pr o duc tion stimu l us sta n ds to assist national d eve l opm ent. However, ac c ording t o A m ir et al.
24 ( 1 9 9 1), agricultural c rops must meet hi g h qual i ty sta n dards in or d er t o b e made ava i lab l e in t he world market. O t h erw i se, t he p rice wi l l dec l ine result i ng in low p rofits for t h e exp o rting c o unt ry and t h e p r odu cing farmer (A m ir et al 1 9 9 1). F o r t h ese reason s, a great ch al l enge is posed in effic i ently and appr o priately p reser v ing agr icultural goo d s. Biological Degradation A general u nders t andi n g of t h e biologi cal p ro c esses which l e ad to sp oi l age is necessary for t h e c o nsideration, d eve l opment a n d i m plement ation of a p prop riate p r eservation syst ems t h at w i ll effec t i v e l y s low det er i ora t ion and ma i n t ain pr o duc t q u al i ty. It sho u ld be no t ed t hat in addition to quant itative losses, qual i ta t i v e losses c ont ribute to t he o vera l l l oss of agr i cu ltural goods during pos t harvest proc esses. T herefore a b ri e f d escription of t hese p r o c esses is pr o v i ded here. As mentioned prev i ou s l y h i g h moisture c ont ent prod u ct s q uickly sp oil as m i cr oor gan i sms t hri v e in su c h env i ronment s. Add i tiona l l y deteriorat ion is k n own to b e ca u sed b y a ra n ge of biol og i cal pro c esses including me chanical inju r i e s, physio l o g ical disorders, p at h ologi cal b reakdown, respira t ion rate, e t hyl e ne p rodu ct ion a n d a ction, sp r o u t ing a n d r o o t in g water stress and r a tes of comp o sit i onal c h anges which are asso ciated w i th q ual i ta t i v e c harac t e ristics su c h as n u t riti v e value, fla v or, c olor a nd tex t ure (Kader, 2 0 0 5 ). The a ct u al rate of d eteriorat ion is dependent u pon environmental c ond it i ons su c h as air velocity, sa n itation pr o ce d ures, t emperatur e, atmos p heric c omposition and relative humi d ity (Kader, 2 0 0 5). To c o u n t eract t his susc eptibi l ity t o deteriorat ion, C haudh ri et al ( 2 0 0 9 ) suggests t hat harvested prod u c ts sh ould o p t i m al l y be preserved, so l d, o r p r o c essed as quickly as p ossib l e af t er harvesting. T o overc o me t h is prob l e m, drying p ro c ess e s
25 h ave been w i dely i mp l e mented as p rerequis i te for t h e storage of agr i cu ltural pr o duc ts. D ry i ng i m prov es t he p reserved shelf l i fe a n d si gn i fican t ly reduc es t he p roduct volu m e and wei g ht while mi n imiz i ng pa c kag i ng, st o rage and tra n spor t ation c os t s ( C haudh ri et al. 2 0 0 9). T hus, was t ages can eas i ly b e p r evented by a p ply i ng pr o per dryi n g pr o ce d ures. Drying Review Ef f ec t i v e d r y i ng of agr i cu ltural goods is essenti a lly t he evapo ration of moisture f r om t he c r op, t h us reduc ing t he p r o du c t s moisture c o n tent so t h a t d eteriorat ion no longer o cc urs. E a ch agricultural c r op has different exper i mental l y det er m ined levels of moistu r e c ont ent t hat is co nsidered safe for preservation (see Append i x f o r specific information pertaining to the safe moisture levels of mango fruit) In a n ormal drying process, the product is placed in an environment in which supplied heat evaporates moisture from the product and air flow then removes the water vapor. Without an air current, emerging vapors will accumulate on the product surface whic h hinders further transport of moisture from within the fruit. The amount of heat needed for water removal from the product is equal to the latent heat of vaporization of water (Arata and Sharma, 1991). It follows that elevated temperatures, along with inc reased airflow rates and reduced levels of relative humidity, permit higher rates of food drying. These conditions apply as long as the diffusion rate is not the rate controlling process for the removal of water vapor. However, at lower moisture content, a irflow rate is much less important than temperature. In this case, higher temperatures lead to higher rates of diffusion within the product which corresponds to the falling rate portion of the drying curve as shown in Figure 31.
26 During the drying process, water at the products surface evaporates first. As additional heat is absorbed by the product, water then begins to migrate from within a crops interior to the product surface. The ease of this migration is dependent on characteristics such as the diffu sion rate which is affected by the porosity of the substance and the surface area exposed to the environment. It should be note d that only the physically held water is removed during drying while the chemically bound water remains in the product. Drying co n t inues unt il rea c hing a point where t he moist u re vap o r p ressure within t h e p r oduct is equivalent t o t he p ressure o f a t mospheric moisture. T his c orrespo nds to an equi l ibr i um st ate in which moisture a b sorp t ion a n d desorpt ion o c cu r at t h e same ra t e and is known as t he eq u i l ibr i u m moisture c ont ent. Drying kinetics are often presented by measuring the average product moisture content as a function of time. This relationship is known as a drying rate which is shown in Figure 3 1. During the initial drying stages, excess moisture on the product surface results in a rapid rate of moisture removal. Subsequent drying of the material depends on the product dependent rate at which internal water migrates to the product surface via diffusion (Chen et al., 2009). Figure 3 1 shows multiple rates at which biological products undergo the drying process. The first period is described by Arata and Sharma (2009) as the constant rate of drying where moisture on the product surface is evaporated. A vapor pressure gradient is established which is greatly influenced by the air temperature and air flow rate.
27 Vapor pressure is essentially the partial pressure of water vapor in a known volume of air and is a function of the humidity ratio which describes the relationship between the mass of water vapor that is actually present in moist air and the mass of dry air. Saturated vapor pressure is the maximum vapor pressure which is a function of temperature. The vapor pressure deficit essentially describes the difference between act ual vapor pressure and saturation vapor pressure evaluated at the same temperature which serves as a good indicator of the evaporative capacity of the air. As water is evaporated by the heated air, less potential is established as evaporative cooling effec ts which introduce additional pressure drops. Th e rate of evaporation during the constant rate period is determined by the humidity ratio, ambient temperature and air circulation (Arata and Sharma, 2009). The second portion of the drying curve, in which th e moisture removal rate decreases, is known as the falling rate period. This is a result of the diffusion rate which is slower than evaporation at the surface of the product. T he n at u re a n d a t tribu t es of t he drying c urve are analogous f o r b o t h hygrosco pic and non hygrosc o pic material u n t il all o f t he u nbo und moisture within t he ma t er i al is removed. B e yond t h is point, a fr a ct ion o f t he b o u nd water is a c t u al l y removed fr o m hygros c opic material. Figure 32 depicts a simple psychrometric chart that demonstrates a drying process. In this example, ambient air is assumed to have a dry bulb temperature corresponding to the point labeled as 1 in Figure 32. As this ambient air is heated, the humidity ratio remains cons tant until reaching the heated temperature described by the point labeled as 2. In this process, the relative humidity is reduced. If the heated air is then used to remove moisture from agricultural products until equilibrium is reached, the
28 tem perature of the drying air will be reduced to the point labeled as 3. During this drying process, the enthalpy remains constant. The final humidity ratio can then be assessed at point 3. The difference in the final and initial humidity ratio allows for the determination of the amount of water removed from the product. T he moisture loss occurring in the constant rate period can be described by Equation 31 while the moisture loss occurring in the falling rate period can be expressed by Equation 32 ( ) = exp ( 1tC) (3 1) ( ) = exp ( 2tE) (3 2) The variable M (t) refers to the absorbed moisture concentration which is a function of time. Mi is the initial moisture concentration, MC is the final moisture concentration directly following the constant rate period and ME is the final moisture concentration after the falling rate period. The variables, tC and tE are the drying times for the constant rate and falling rate periods, respectively. The rate constants, k1 and k2 are defined by the general expression described in Equation 33. =2 4 2 (3 3 ) The variable D refers to the unique moisture diffusion coefficient that depends on which drying rate period is being investigated. The variable L refers to length of the sample. Safe storage of biological materials is particularly difficult in e nv i ronment s with hi g h re l ative hum id i ty. This challenge becomes even more difficult if t he required e q ui l ibr i u m moisture co n t ent of the material is sufficiently low. For t h is reason, a
29 si g nif i ca n t di l e m ma in dev e lop i ng c ommu n it i es, par t icular l y in t ropical environments, lies in t h e removal of moisture f r om agricultural p rod u ct s t o rea c h adequat e preservation states in a timely manner. If the drying process is too slow, growth of microorganisms will occur due to the conditions associated with high ambient temperature and relative humidity. The primary objective of a dryer system is to provide increased temperatures to the product that are higher than the ambient conditions. By doing so, the vapor pressure of moisture within a product is sufficiently raised, while the relative humidity is low ered. However, vapor pressure is only a function of the humidity ratio as it does not change with temperature. The saturation vapor pressure that drives the drying process is what actually changes. This expression is essentially defined as the difference b etween vapor pressure and saturated vapor pressure. This condition ensures sufficiently low equilibrium moisture content by increasing the moisture carrying capacity of the air since heated air is able to retain larger quantities of moisture than cool, ambient air. The manner in which temperature and humidity affect the potential moisture absorption of air is shown in Table 3 2. The application of such dryer systems in developing countries can greatly reduce post harvest losses of agricultural commodities and significantly improve the availability of food in these areas. However, appropriate systems must b e c onsidered t hat meet the needs a s w e l l as t he limita t ions o f t he communi ty. A brief out l i ne o f d ry i ng pra c tices is mentioned here in o rder t o demons t rate t h e n ecessity o f an a p p r opriate level of te c hno lo g y f o r d eve l oping nat ions. After ident i fication, t his technology w i l l be elabora t ed o n and rev i e w ed in t h is repor t T h is broad o verv i ew of
30 dryer so phistica t ion lev e ls rang es fr o m si m p l e tra d it i onal p ra c tices to a d va n ce d industrial prac tices. Traditional Drying Methods O p enair sun drying is o ne o f t he most c ommon and l o ng pra c ticed method s o f f o od preservation among a lar g e nu mber o f c oun tries d u e to its si m pl i city and t h e abu ndanc e o f so lar irrad i anc e (Bha n dari et al 2 0 0 5 ; Chen et al., 2 0 0 9 ; K a ndpal et al., 2 0 0 6 ; Sreekumar et al., 2 008). In t h is pro c ess, t h e f o od p r odu ct is spread into thin layers on t h e groun d, on mats, o r o n t rays as is shown in Fi g ure 3 3 The uneven c r op surfa c e is t hereby exp o sed t o sh ort wavel e ngth solar energy wh i ch is c onver t ed in t o t h ermal energy af t er b e i ng abs o rbed. How e ver, only a por t ion o f t h is energy i s a c t u al l y abs o rbed by t he p rodu ct whi l e t he r ema i n ing ra d iation is re f l e ct ed. Addit i ona l l y air flow a c ross t h e p r oduct su rfa c e results in c onvect i ve heat loss and can introduce moisture. T he p rod u c ts are ra r e l y pre t reated and must be t urned frequent ly in o rder to sufficiently dry. O p enair sun drying h as d i m inished over t i m e as t h e limita t ions asso ciated w i th t his pr o cess are fur t her rec o gni zed. In fa c t, it is wide l y a c knowled ged t hat th i s met h od is unhyg i enic sin c e t he c rops are eas i l y c ont am i nat ed (E n ebe a n d E zekoye, 2 0 0 6) by at mospheric dus t pol l u t ion, intrusion by a n i m als, infestation by b irds or insects, and animal d ropp in g s t hat all lead t o infestation by f u ng i b a cteria, and ot her m i cr oor gan i sms. Ad d it i ona l l y t his meth o d is known to b e labor and t i m e intensive as c rops need to b e covered at n i g ht a n d dur i ng inclem e nt wea t her. Fur t her, the c rops must b e co n t inua l ly wa t ch ed t o prevent co n t am i nat ion fr o m animal s a n d birds. Addit i onal l y, t h e p r odu ct s must b e t u rned frequent ly t o at t ain adeq u ate dryin g
31 M o reover, t h is proc ess requ ir e s sign i f icant land area a n d prolongs t he d ry i ng period which may res u lt in d eteriorat ion of c rop qual it y. Due to t h e hygros c opic pr o per t i e s o f agricultur a l pro d u c ts, c rops exposed to env i ronment al c o nditions sta n d the risk o f b e i ng rewet t ed, p a r tic u larly at n i g ht when amb i ent temperat ure is l owered and humid i ty increases. In t hese situa t ions, remo i sten i ng effe c ts occur by condensat ion o r b y vapor diffus i on induced via osmo t ic f o r ces. T his poses si g nif i c ant p roblems in humid, tr o pical regi ons where cer t ain c rops are grown and p ro c essed d u ring rainy seas o ns (Bu c hin g er an d We i ss, 2 0 0 2 ). Non uniform and insuff i cient drying also l e ads to spoilage and d eteriorat ion of t h e c rop during st o rage. Additionally, many prod u c ts are kn own to exhib i t discoloring when exp o s ed t o direct U V r adiation. Fur t hermore, d ir e ct su n exposure during hi g h temperat ure days is kn o wn t o contribut e t o case h ardening, in which hard shel l s form o n t h e ex t er i or surfa c es of agr i cu ltural pr o duc ts (Chen et al., 2 0 0 9). T his shell su bsequent ly t raps moisture inside and t h us extends t h e p r oduc t s expos u re t o moisture. All of these issues result in the deterioration of food quality in terms of a loss of nutritional val ue, adverse enzymatic reactions, loss of germination and an overall deterioration of the product (Bhandari et al., 2005). Under these conditions, studies have shown that agricultural losses can actually rise as high as 40 60% of the total harvest production (Chaudhri et al., 2009). For these reasons, it has b een determined t hat openair sun drying d oes not fulf i ll t he in t ernat ional q ual i ty sta n dards and t h erefore it ca n not b e so ld in t he in t ernat ional market (Chen e t a l 2 0 0 9).
32 Smoke d r yi ng is ano t h er t raditional meth o d of preservation used in t r opical c o unt ri e s. In t his pro c ess, heat i s genera t ed b y biomass c ombus t ion o f s u c h material s as timber, co co n u t shel l s, rice h u sks and o t her agr i cu ltural wastes (A m ir et al 1 9 9 1 ). A l t h ough t h is te c hnique is wea t her independent a n d preserves c rops to an ex t e n t, it also c o n taminates t h e prod u c t w i th co mbustion res i dues and t hus severely di m in i shes t he p rodu ct q ual i ty. It has been n o ted t h at t he q ual i ty of su ch p ro d u c ts are n ot good enough to a t t ract t he market and henc e are t o b e c ons u med locally o nly (Bhand ari et al 2 0 0 5 ). Industrial Drying Methods In indus t rialized reg i ons, t raditional dryi n g meth o ds have been replaced by me chanical dryers t h at exhib i t faster drying ra t es, require less l and, a n d p rovide hi g her qua l ity pr o duc t. T hese industria l ized areas emp l oy a dvanc ed d ry i n g equipm e nt su ch as st eam dryers, infrared, fluid i zed bed, spou ted bed, drum dryers a n d freeze d ryers t o p ro c ess c ommercial p rod u ct s. Howe v er, t h is equipm e nt is expens i v e and energy intens i ve since it requ ir e s rel a tive l y s i g ni f icant amo u n t s o f energy in t he f o rm of e l ec t ricity or f u el t o operat e. While c ompanies t hat generat e su b sta n tial revenues c an reasonab ly aff o rd t h is te c hno lo g y, most sma l l scale organizations or co mmun i ties t hat are direc t ly invol v ed w ith t he f a rms are u nable t o afford imp l ementat ion o f t hese t e chno lo g i e s (C h en et al., 2 0 0 9). O nly l arge plantat ions o r co mm e r cial es t ab l ish m ents find t hese t ec h nolog i es e c ono m i cally v iabl e in deve l opi n g coun tries (Bu c hinge r a n d We i ss, 2 0 0 2 ). Add i tiona l l y many rural areas in dev e lop i ng c oun tries have l i m ited res o ur c es. Cons t ru c tion suppli e s may b e l imited and energy sour c es su c h as fossil fuels and
33 e l ec t ricity may be unrel i able or t o tally absent S t udies have sh own t h at even sma l l, sim p le o il fired bat ch d ryers are not app l i ca b le f o r r ural farmers in t hese regions (Bu c hing e r a n d W e iss, 2 0 0 2). Hence, t here is a need t o identify an intermedi ate practical d ry i ng t ec h nology t h at can eas i ly b e implemented in dev e lop i ng r e g i ons to ensure fo o d supply t o a growing pop ulation. Fur t hermore, a p prop riate d r y i ng t ec h nology c an enable farmers t o prod u c e high qua l i ty, market a ble goods (Bu c hin g er a n d We i ss, 2 0 0 2 ). Solar Drying Methods Solar d ry i ng has been des c ribed as a pot ential decen tralized t hermal a p pl i ca t ion o f so lar energy par t icularly in deve l o p ing c oun tries ( C hen et al 2 0 0 9 ). In fa c t t h e u se o f so lar t h ermal syst ems has been shown to be practical, ec o n omical and env i r onment al l y resp o nsi b l e in c onserving agr i cu ltural pr o duc ts (Bu c hin g er an d We i ss, 2 0 0 2 ). Mo reover, solar heat ing syst ems are capable of i m provi n g pr o du c t qual i ty whi l e redu cing was t ed p rodu ce and m i ni m iz i n g t h e use of tra d it i onal f u e l s. T he j u stif i ca t ion f o r so lar dryers is t h a t t h ey may be m ore effec t i v e t han sun dryin g but have lower operat ing c os t s t han me chanized d ri e rs (C hen et al 2 0 0 9). T h e app l i ca t ion o f equ ip m ent for col l ec t ing so lar radi a tion serves to d iffer e n t iate s o lar drying from t raditional meth o ds su c h as openair sun dryin g As not ed b y Arata and S harma ( 1 9 9 1), t he superiority of so lar drying h as already b een es t abl ish e d over o pen air sun d ry i ng t hrou gh many s o lar drying st udies. In solar d ry i ng o p erations, a por t ion o f sh or t wave solar ra d iation is first received by a n absor b er at it t rave l s t hro u gh a transpar ent co ver. T he r adiation is su bsequent ly c onver t ed in t o low g rade heat after strik i ng an opaq ue wal l Since long w a ve l e n gth
34 radiation is unable t o travel bac k a c ross the tra n sparent c over, t he h eat b ec o mes t r apped within the d r yer as shown in Fi g ure 3 4 In t his way, radiative energy is harnessed for drying a p pl i ca t ions as op p osed to direc t ly exp o sing t he pro d u c t to the env i ronment By enc los i ng t he p rodu ct inside a con t rol l ed e n v i ron m ent, t he f o od is l ess l i ke l y t o be co n t am i nat ed by anima l s, birds, insects and dust t hereby re d u c ing t he li k eliho o d of fungal and m i cr obial growth. T his also l i m its t he in t rusion of water in poo r weather co nditions and redu c es t he d ir e ct co nvec t i v e losses t o t h e ambient env i ronment T h ese i m provements direc t ly b enef it t he e n d user b y l i m it i ng t he w o rk t hat must b e d one to prot e c t the c r o p from t hese t hreat s a n d results in a h i g her qual i ty pro d u c t. Add i tiona l l y solar drying res u lts in q uicker d ry i ng ra t es b y a c hie ving h i g her tem p erat u res, lower humid i ty, a n d increased air move m ent (Bu c hing er an d We i ss, 2 0 0 2 ). H e n c e, fo o ds c an be d ri e d over shor t er p er i ods of time which increases t he efficiency of t he p r o cess by al l owing le s s t i m e f o r sp oi l a g e t o oc cu r. M o re co mp l ete, u niform drying i s also possible which al l ows for bet t er p reserved quality and longer st o rage po t ential. In c reased p rod u ct t hrou ghp u t is a l so possible with increased drying ra t es. Nutritional val u es o f t he p roducts are also be t ter pre s erved by drying f o od items in short t i m es and opt i m al temperat ures. Addit i on a l l y, research h as sh own t h at so lar drying c an i m prove t h e q u al i ty of a p rodu ct in regards to co lor, fla v or and appearance w hich enhanc es t he pro d uct s market a bi l ity and c o nsequent ly al l ows for i m proved financi a l opp o rt u nities for farmers (Chen et al., 2 0 0 9). A l t h ough solar dryers are ca p ital
35 i n t ensi v e, it h a s b een d iscussed t h at t h e u n it co st of so lar drying is expec ted to b e a sma l l fra ction of t h e sel l ing p rice o f t he p rodu ct (Ka n d pal and Kumar, 2 0 0 5 ). Fur t hermore, n umerous st u dies have n o ted the financ i al at t ra c tiveness of solar dryers to t h e operat ors as fa b rication is si mp l e and c ommercial f u els are su bs t itu t ed ( Kan d pal et al 2 0 0 6 ). Several additional aspe c ts must also be co nsidered when c omparing solar dryi n g to c onvent ional d ehydrat ion proc esses. F or instanc e, solar dryers must be a b le t o p rovide t h e equ i v al e nt p erformanc e o f a co nventional p r o c ess in t erms of capac ity, labor inp u t, pro d u c t qual i ty, a n d re l iab i l ity. Howe ver, t h e perf o rmance of solar dryers is sti l l l ar g ely dependent o n weather co n ditions bec ause t he h eat required to remove moist u re is o ften generated by so lar energy only. W e at h er c o nditions also si g nif i ca n tly in f luence t h e c apac ity of t he p rodu ct t hat can be d ri e d w i t h in a g i ven t i m e period. T h e d r y i ng t i m e is sh o rt under sunny co nditio n s a n d is extended during adverse wea t her. Cons i der ing t his dependen ce on weather, t h e u t i l ization o f so lar energy as t he o nly energy so ur c e is rec o mm e nded for sma l l sca l e d r yers where t he r isk o f large qua n tities of spoi l a ge in incl e ment weather is l ow. It is rec o mm e nded t h at lar g e sca l e d r yers used for co mm e r cial p urp o ses, are equ ipped with bac kup hea t ers t o ens u re d ry i ng d u ring in c l e m ent weather (Bu c hing er a n d We i ss, 2 0 0 2 ). T h e d r y i ng behavior o f agricultur a l cr o ps is a l so dependent o n the p roduct size and sh a pe, initial moisture c ont ent, final moisture c ont en t bulk density, la y er t hickness, me chanical o r c hem i cal pre t reatment t urning intervals, product t emperature, t e mperatu r e and h um i dity o f t he d ry i ng air, and t h e air ve l o c ity. T o
36 i m prove t h ese c ond it i ons, d ryers must b e designed t o ensure t h at drying air flows t h rough t h e ch amber whi l e c ont ac t ing as much of t h e pro d u c t surf a ce as p ossib l e. The pro d uct su rfa c e area is inc r eased b y t hinl y sl icing t he f o od before plac e ment on drying ra c ks t hat al l ow for max i mum c o n t a ct between t h e h eated air and p r odu ct As t he h e ated air flows o ver t he trays, the pr o duct becomes loaded w i th moistu r e. The moist air is t hen exhaus t ed f r om t he d ryer whi l e fresh air is subsequen tly drawn in. Solar d ryers are generally clas s if i ed a c co rding to the manner in which solar heat is appl i ed and used during the dryin g process Generally d ryers are categorized into two broad groups; act i v e d r yers ( c onven t iona l ) and passi ve d r yers (na t ural circ u lation) (Sreekumar et al 2 0 0 8). Ac t i v e d r yers are systems which induce for c ed air c irculation, while p assi v e d r yers o n ly ma k e use of t h e n a t u ral co nvec t ion p r incip l e which generates mov e ment of t h e heat ed air. T h e f o l l owing s ec t ion o f t h is rev i ew ser v es t o identify and e l abor a te on t h ese t wo t ypes o f so lar dryers. Additional l y, a c lass i fication sch eme for solar dryers is deve l oped and t h e distin c tions made are des cribed in fu r t h er d etail with ac t ual exampl es o f drying systems. Solar Dryers Active and Passive Mode Pass i v e so lar dryers are also c al l ed n at u ral cir c ulation o r n at u ral co nve c tion systems. T heir operat ion de p ends co mp l etely on solar energy. A ir in t h ese systems is h eated and nat ural l y c irculated by pressure gradi ents establis h ed b y wind a n d temperat ureinduc ed b uoyanc y f o r ces. As a result, t h ese dryers d o not re q uire e l ec t rical or mechanical c ompone n t s su ch as fa n s o r b lowers bec ause t h e nat ural c o nvec t ion d r i ving f o r c e is based on temperat ure diff e ren c e o r c hanges in air d ensity. T he
37 d iffer e n c e in spe c if i c we i ght between t h e ambient air and t h e d r y i ng air pr o motes a ver t ical air flow independent of e l ec t rical supply. S t udies have sh own t h at in genera l solar dryers can b e easi l y m a intained and c ons t ru c ted fr o m inexpensive, local l y av a i l a b le ma t er i a ls. Consequent l y it h a s b een d etermined t hat so lar dryers are appr o priate for sma l l farms where f inanci a l and mater i al resour c es are l i m ited (Chen et al., 2 0 0 9). Likew i s e, it h a s b een sh o wn t h at the nat ural co nvec t i o n solar dryer has pot ential for i m plem ent a tion in t h e t ropic a n d sub t r o pic co mmun i ties of t he d eve l oping world. In fa c t, S harma et al. (1 9 9 5 ) sh owed t h a t th ese dryers were suitable at h ouseho ld le ve l s f o r d ryi n g sma l l bat ch es o f h i g h moisture c ont ent prod u c e, despite lim itations due to t h e de p endenc e on t emperature difference and pressure drop of air as it is for c ed t h r ough t h e c r op. A primary c onc ern of t h e perf o rmance o f so lar dryers is t hat t h e airf l ow in t h ese systems is n ot sufficient in penet rating higher c rop bulks (Bu c hin g er an d We i ss, 2 0 0 2 ). Furthermore, t h e air flow c omes to a standsti l l during ni g ht a n d adverse weat h er c ond it i ons. T hese l i m itations e l evate t he r isk o f p rodu ct deteriora t ion fr o m mold at ta c k a n d in c reased en z ymatic reac tions. T herefore, it has b een c o n cluded t hat su c cessf u l use o f n a t u ral co n vect ion d ryers is res t ricted t o t h e d ry i ng of sma l l bat ch loads in a r eas with hi g h inso l ation ( A m i r et al., 1 9 9 1). A l ter n ativ e l y, for c ed c onvec tion solar dryers have been intro d u c ed in order t o ma i n t ain c o n t inuous venti l ation and air flow. S ince t hese d r yers u t i l ize solar ener gy as we l l as mo t orized f a ns or blowers for air cir c ulation, t hey general l y have t h e adv ant ages of hi g h re l ia b ility and efficiency. O n t h e ot her hand, t he r equire m ent of e l ec t ricity for fa n s o r b l owers, l i m its im p l e mentat ion o f t hese devices since
38 elec t ricity is non existent in many rural areas. Even when electricity is available the incomes of potential energy consumers a r e o ften t oo low to make application feas i bl e In t his c ase, t h e co st of elec t ricity must b e balanced w i th an improved syst em performanc e su c h as greater d ry i ng c apac ity, reduc ed drying time, and improved p r o duc t q ual i ty. Ac c ording t o Mrema et al. ( 1 9 8 7), n a t u ral co nve c tion dryers t ypical l y exhib i t overal l dryi n g effici e n c i e s o f a b out 1 0 1 5% whi l e f o r ced c onvec tion d ryers are genera l l y 2 0 3 0%. Fi g ure 3 5 s h ows a sc hematic of t he main c ompo n ents of an a ct i ve s o lar fo o d dryer. Active dryers su c h as t hese, u se fa n s o r blowers t o move solar hea t ed air f rom so lar c o l l ec t ors to d ry i ng c hambers. H i gh moisture co n t ent produ ct s such as fr u its a n d ve g eta b l e s are of t en pro c essed with ac t i v e d r yers. Direct Mode Dryers Dryer co nfi g ura t ions c an be f u r t h er d iffer e n t iated in t o sub classes o f in t egral t ype (direc t ) a n d distributed type ( i ndirec t ) d r yers depend ing o n whet h er t he p rodu c t is exposed direc t ly t o so lar radiation or dried in t he sh ade. Integral t ype (direc t ) d r yers c onsist of a sing l e d r y i ng u n it, of t en with t h e solar c ol l ec t or forming t he r oof o r wall o f t he c hamber W i th d ir e ct mo d e, t h e prod uct itself serves as t h e abs o rber. T he h eat t ransfer is affec t ed not only by co nvec t ion b u t also by ra d iation acco r ding t o the albedo o f t he p roduc t s s u rfa c e (B u ch in g er a n d We i ss ( 2 0 0 2). In contrast, d istributedt ype (indirect) dryers o fte n co nsist o f t wo separat e units. A solar c o l l ec t or first h eats the air which is then f o r ced t h r ough t h e p r oduct in a separ ate drying c hamber. Fi g ure 3 6 d epicts a general o ve r v i e w o f d ryer t ypes based o n these distinc t ions.
39 Direct so lar dryers h old t he p roduct within an enc losu r e t hat is o f ten shiel ded w i th a thin, tra n sparent c o ver formed out o f p last i c or g l ass. T h e d r y i ng c hamber itself is essen t i a l l y an ins u lated, re c ta n gu l ar b ox t h at al l ows for airf l ow t h rough sma l l holes in t he t op a nd bo t t om. Perforated t rays are used to hold t h e pro d u c t as t he air f lows t h rough both t h e t rays and the prod u ct Several d ir e ct d ryer desi g ns rev i e w ed in t h is report are t h e t ent d ryer, t h e seesaw dryer a n d the b ox dryer. A typical d ir e ct d ryer (b o x type) is shown below in Figure 3 7 T h ese direct d ryers are desi g ned so t h at heat is not o n ly generat ed by solar r adiation a bsor p tion on t h e sur f aces o f the drying c hamber, b ut also on t he prod u c t itse l f. Direct p assi v e d r yers have b een shown to successf u l l y dry sma l l ba t ch es o f h i g h moisture c ont ent p rodu ce su ch as b anana, c arro t s fren c h beans, mango, pineapple, a n d pot a t o (Ja yaraman et al., 2 000). Howe ver, when u sing int e gral (direc t ) mode of dryin g is should be n o t ed, t h at su nl i ght may affect c ertain essen t ial c ompon ents in t h e p r oduct. In f ac t direct exposure to sunl i ght of t en r esults in discolora t ion, v i tamin loss a n d undesirable t emperature rise s in t he t hin, t o p layer of t h e prod u c t (Sreekumar et al., 2008 ). Due to these li m itat i ons of t h e b ulk dept h, su c h dry e rs n eed fre q uent c r op t u rning to at t ain u niform drying and lar g e sur f ace areas to sp read the p roduct. T herefore, if grounds are sc ar c e, indirect mo d e t ype o f d ryers are p referred f o r d ry i ng larger quant it i es. A lso, moisture evapo rated fr o m t he f o od may cond ense o n the inside of t h e abs o rber cover, thus r educ ing t h e t ransmittiv i ty. Tent so lar dryers, as sh own in Fi g ure 3 8 are inexpensi v e and simp l e in c ons t ruction. T h ese units cons i st of a plastic sheet c overing a frame t hat is fa b ricated
40 with woo d en poles. Black plastic sheeting is preferred on t h e side o p posite of sunl i ght exposure to a b sorb more heat W i t h in t h e frame, a ra c k is situa t ed to ho l d t h e f o od. H owever, st u dies have sh o wn t hat d ry i ng t i m es are hardly i m proved co mpared with openair sun drying ( B u c hin g er a n d We i ss, 2 0 0 2 ). Rather, t he p ri m ary pur p ose of te n t dryers may o nly serve to p ro t e c t the p rodu ct f r om p o t ential co n taminants su c h as dus t rain, or predat ors where was t age is o t herwise h i g h. T ent dryers are also able t o be s t o red when n o t in use a n d t h ey are typical l y used to d ry c r ops w i th low dens i ty and por o sity. T h e t raditional seesaw dryer co nsists o f a s tiff, re c ta n g ular frame su p por t ed a b out an ax i s. T h e suppo rt is des i gned t o al l ow ti l ting of t h e frame t o tr a c k su n l i ght t h rough o ut t he day. T h e p r oduct is en c losed in t he f r ame on a n umber o f mesh trays whi c h al l ows vertical air c irculation a nd subsequen tly promot es evapor a tion. Corrugated iron sheet is of t en used w i th w o oden suppo rts t o a b sorb hea t T hese surfac es are painted black for i m proved heat a b sor p tion. Additional l y t h ermal insula t ion c an be integrated into t h e frame with t h e use of w o od fiber, po l ystyrene, c orrugat ed c ardbo ard o r other insu l ating m ateria l T he p rodu c t is placed on removab l e t rays t hat are p ositioned above t h e co rrugated iron in either co n t inuous rows or with spa c e b e tween t hem, which al l ows for i m proved hea t ing o f air. A typical seesaw d ryer is sh o wn in Fi g u r e 3 9 A greenhouse effect is observed with the inclusion of a transparent plastic sheet above the trays. Air circulation is driven by the natural convection principle while fresh air enters the lower end of the drying chamber and esca p es at t h e u p per end. A i r c irculation is improved with a wider air o u t l e t opening c ompared to th e air inl e t opening.
41 T his a l lows for a gradu a l w i deni n g of t h e c r oss se c tional area of t h e fr a me which improves convection. T h e b o x type solar dryer has been extensivel y im p l e m ented in small sc ale f o od drying p ro c esses. T h e general desi g n is co mpr i sed o f a woo d en b ox w i th an incl i ned, t ransparen t l i d made of glaz i ng material. T he in t ernal c ol l ec t or wa l ls are pain t ed b lack t o absorb incom i ng radiation and t h e p r od u c t is held on mesh trays. Holes in t he b o t t o m and f r ont of t he d ryer frame allow air to ent er t he d ryer whi l e heat er air is exhaus t ed f r om ven t s lo c ated at t he upper end o f t he b ack wal l F i gure 3 10 i l lustra t es t he elem e n t al featu r es and d es i gn of a st andard so lar box dryer. T h ese types of dryers are c apable of ac h i e v i ng higher temperat ures a nd co nsequent ly sh or t er d ry i ng t i m es t han tent d ryers. Howe ver, drying ra t es are still relat i vely low and pro d u c ts o f ten exhib i t discolora t ion. Due to t h e small drying c apac ity of t h ese dryers, the i r u se is generally limited t o domestic use (B h andari et al 2 0 0 5). Indirect Mode Dryers Ind i rect so lar dryers are generally less co mpact t han d ir e ct d ryers, b u t o f ten exhib i t i m proved effic i en cy (Chen et al., 2 0 0 9) and are ca p able of drying larger quan tities of fo o d prod uct (B u ch in g er a n d We i ss, 2 0 0 2 ). The pri m ary distin c tion w i th indirect d ryers is t hat u nl i ke direct d ryers, t h e product is not exposed directly to solar radiation. This minimizes the possibility of decomposition such as discoloration, surface cracking and inadequate internal drying, which arise from direct radiation exposure. The primary indirect dryer designs investigated in this review are known as cabinet and tunnel dryers.
42 T h e so lar co l l ec t ors in t h ese distribu t ed syst ems must be app ropriately p o sit i oned in or d er t o op t i m ize t h e c ol l ec t ion o f so lar energy. T his requires t he d etermination of a suitable incl i nat ion as greater amo u n t s o f so lar energy are gat h ered with t h e sur f ace of the co l l ec t or p ositioned perpendic u lar l y t o t h e su n l i ght. A t i l ti n g of t h e so lar co l l e ct or also assists in air flow via t h e nat ural c o nvec t ion p r incip l e in which warmer, less dense air r i ses t hrou gh t he syst em. The solar cabinet dryer is considered more sophisticated as compared to typical box dryers as they generally consist of two separate components: a collector which h eats air w i th so lar radiation an d ; a drying c hamber t hat h ouses trays or sh e l ves of prod uct. T h e p r incip l e o f o pera t ion is sim i lar to t hat o f th e b o x d r yer where amb i ent air is drawn in as h ot, moist air is expel l e d from ven t s at t he h e i ght of t he dryer. A sta n dard solar c abinet dryer is sh o wn in Fi g ure 3 1 1 Whi l e t he h i g her co mp l ex i ty of desi g n res u lts in a re l atively c os t l i er o p t ion, t h ese dryers are s ti l l co nsidered suitable f o r sma l l scale, income generat ing c ommu n it i es (Bha n dari et al., 2 005). T h ese dryers are d es i gned t o operat e with solar radiation serv i ng as t h e main energy sou r ce, altho u gh bac kup hea t ers are used when ra d iation is inadequat e due t o p oor wea t her con d it i ons and during t he n i g ht so t hat co n tinuo u s d ry i ng is made p ossib l e Radiation first passes t hrou gh t he tra n sparent cover of these dryers, where it is absor b ed b y the interior surfaces of the solar collector. T he h eat generated results in an increased temperat ure of t h e sur r ou n ding a i r. Natural c onvec tion ca u ses t h e hea t ed air to rise and is t h us for c ed t hro u gh t h e d r y i ng t rays where moisture is c ol l ec t ed. T h e moist air t h en exits t h r ough vents loca t ed at t h e t op o f t h e d r yer which redu c es intern a l
43 ca b inet p ressure. Consequent l y amb i ent air is c o n t inua l ly d rawn into t h e d r yer. T he airflow c an be r e gu l ated b y vary i n g t h e o u t l e t vent size. A i rfl o w in a nat ural c onvec tion system is est a bl i shed by t he so lar hea t ed air b ec o m i ng l i ghter or l e ss d ense t han t h e ambient air. A sma l l pressure diff e renc e is t hus cr eated b y t he density gradient which draws air t h rough t h e co l l ec t or, d ry i ng c hamber, and cr op. T h is effect in c reases with greater hei g h t s b etween t h e inl e t and bed, as we l l as t he o u tlet and bed. However, t he effect on an in c reased hei g ht o f t he o u t l e t is l ess t han t hat of an in c reased h e i ght of bed be cause t he air is c ooled a s it passes t h rough t h e be d (B u ch in g er a n d We i ss, 2 0 0 2 ). T h e moist air is then d ischarged t hro u gh air vents or a c h i m ney lo c ated a b ove t he d ry i ng c hamber. T h e o p t i m ization o f c hi m ney height for na t ural c onve c tion solar dryers is discussed by Irtwa n ge and Adebayo ( 2 0 0 9). T h e c abinet c ompon ent of t hese d r yers is essential l y a lar g e wo o den or metal box which is pro p er l y insulated t o m i nimize h eat loss. Wa t er resist a nt c ladd i ng is of t en u sed in t h e c on s tr u ct ion of th ese dryers. Internal r u nners are fitt ed inside t he c abinet to support t h e t rays of fo o d b e i ng pr o cessed. T h e d r y i ng t rays sl i de o n t h ese ru n ners so t hey can easi l y be remo v ed f o r loading, unloading and cleaning. A general r u le o f t humb is th a t a one m 2 tray area is needed t o lay o ut 1 0 kg of fresh produce (Buc hinge r a n d We i ss, 2 0 0 2 ). T h e basic co mponent s of a so lar air co l l ec t or are a co v er, absor b er, air passage and insulation. As solar radiation is tra n sm i t t ed through t he cover, t h e a b sorber is heated, which in t u rn heat s t he air in t h e air p a ssage. Wh i le air h as a re l ativ e ly low h eat ca p acity co mpared with water, t h ese solar air c o l l ec t ors are
44 p referred as t h ey require less t ec h nical equipment t h an water b ased c ol l ec t or systems and w i ll n ot malfunction when small l eaks exist (B u chi nger and Weiss, 2 0 0 2 ). The result o f t his co mpromi se is t hat h i g her volum e fl o w ra t es must b e at t ained with air co l l ec t ors. A lar g e va r i e ty of flat p late c ol l ec t ors which are uti l ized in agri c ultural drying proc esses have b een rev i e w ed b y Sh o ve (1 9 7 7 ). The basic c ol l ec t or type is t he b are p late which is co mpr i sed o f an air c h amber bet ween insulation with t h e u p permost su rf a ce ac t ing as t h e a b sorber plate. Preference is g i ven t o bare plate c ol l ec t ors as t hey are eas i ly in c orp o rated into t he r o ofs of st orage b u i l din g s. A l ter n ativ e l y, co vered plate co l l ec t ors exhib i t i m prov e d co l l ec t ion efficiency but result in hi g her c o st a n d c o mp l ex i ty. T his sop h ist i ca t ion a n d hi g her c o st arises from t he a d dition a n d u ti l ization o f t ranslucent c o vers abo ve t he a b sorber plates. M o re so phistica t ed d es i gns have a lso been devel oped including t he flow o n bot h sides abs o rber in which an air ch annel is formed between two p lates o f metal (Bu c hing er a n d We i ss, 2 0 0 2 ). The upper sides of t h e plates are c o a ted black and a g l ass c over is mou n ted above. Sus p ended plate c ol l ec t ors are known to exhibit h i g her l e ve ls of effic i ency t han bo t h the bare and co vered plate col l ec t ors, b u t require a more co mp l ex fabric a tion as air is al l owed to flow o n bot h sides of t he p late. F i g ure 3 1 2 sh o ws several air flow principl e s o f so lar co l l ec t ors Add i tiona l l y at t ention has been g i ven t o t he in t egration o f sidewa l l c o l l ec t ors into dryer wal l s. Howe ver, t h ese desi g ns are o ften expensive and are u sable for only t wo or three seasons ( C hen et al 2 0 0 9 ). P l astic fi l m solar c ol l ec t ors have also b een p resented by Keener et al. ( 1 9 7 7) and Chau et al. (1 9 8 0 ).
45 A solar c hi m ney increases t h e b u oyanc y f o r c e within the dryer and t h us p r ovides a h i g her vel o c ity of air cu rren t T h is results in an in c reased r a te of moistu r e removal. Howe ver, it h a s b een sh o wn t h at t he imp l ementat ion o f a ch i m ney o nly bec o mes useful when t h e inc o m i ng air is heat ed in excess of 1 0 3 0 C (B u ch in g er a n d We i ss, 2 0 0 2 ). O t h erw i se t he chi m ney ma k es no sign i f icant improvem e nt u nless it efficiently rai s es air t emperature by servi n g as a s o lar co l l ec t or. It sho u ld be n o ted t h a t passi ve d r yers ach i e ve only mi n i m a l pressure difference per unit of c hi m ney hei ght even when hi g h density d iffe r en ces are a t tained (Bu c hinge r a n d We i ss, 2 0 0 2 ). In contrast, forced c onvect i on systems o perat e at m u ch hi g her mag n itudes of pressure differences. The chimney effect is essentially the movement of air into and out of the chimney and is driven by buoyancy forces resulting from temperature and moisture differences. Greater thermal differences and chimney heights result in greater buoyancy forces and an increased chimney effect. Thus, t he draft flow rate that is induced by the chimney effect involves large temperature differences between heated air and ambient air This term is defined by Equation 3 4, where Q is the chimney effect (flow rate), C is the discharge coefficient, A is the cross sectional flow area, g is the gravitational acceleration, h is the height of the chimney, Ti is the average internal temperature and To is the ambient air temperature. = 2 (3 4)
46 A l t h ough t h e repo rted p erformanc e o f n at u ral co nve ct ion c abinet dryers is a c ce p ta b l e there are many inherent co ns t raints. T he main l i m i ta t ion o f t hese syst ems is i nadequ ate air flow w h ich reduc es t he d ry i ng ra t e and p o or moist air removal which results in cr op spoi l age. T hese c ond it i ons oc cu r when t h e flowing air b ec o mes nearly sat u rated to t h e ex t ent t hat t h e temperat ure is nearly equ al to t h a t of the amb i ent air. T his results in co nsiderabl y small b uoyanc y differences, a n d as a result, low air flow ra t es ( M umba, 1 9 9 6). On t he o t h er end of the spe c tr u m, except iona l ly h i g h internal temperat ures c an result in overheating of t he p r o du c t. In fa c t, t emperatures as h i g h as 7 0 1 0 0C may b e reac h ed with t h ese dryers which are ex c essi ve le v e ls for most prod u ct s (B u ch in g er a n d We i ss, 2 0 0 2 ). An extensive range o f d esign i m prov e ments have been su g g ested in respon se t o t h ese l i m itations. O ne su ch d es i gn enhan ces venti l ation w i th t he in t roduction of wind p o wered r o tary vanes instal l ed o n t h e t op of c hi m neys (Bu c hinge r a n d We i ss, 2 0 0 2 ). Dampers a re u sed to co n t rol t h e temperat ure and air flow ra t es o f t hese d r yers. However, t h ese systems have been shown to be essen t ia l ly ineffect u al between w i nd peaks a n d exhib i ted co mp l e t e inac t i v ity dur ing a b eyances in t h e wind. Hence, t h is dryer design is l imited to u se in areas w i th relative l y high, sust ained winds. T h e inc o rpor a tion of f o r ced c onvec tion c ompon ents has also been g i ven c onsiderable at t ention in i m provi n g t h e t emperature and flow ra t e c o n t rol of solar c abinet dryers. In t hese hyb r id solar c abinet dryers, opt i m um air flow c an be p rovided in t he d ryer t h rough o ut t h e d r y i ng pr o cess to co n tr o l temperat ure and moisture independ ent of weather co nditions. Fur t hermore, t he b ulk dept h is l ess restrict ed.
47 Hence, t he c apa city and the re l iab i l ity of th e se dryers are in c reased c onsiderably co mpared to n at u ral co nvec t ion d r yers. F or t hese reas o ns, well d es i gned and exe c u t ed, f o r cedc o nvec t ion d r yers are c onsidered t o be mo r e effecti v e and exhibit a hi g her l e ve l of co n t rol t h an t he n at u ral circulation type. In fa c t, it h as b een sh o wn t h at dry i ng t i m es c an be r educ ed b y up to t h ree t i m es whi l e t he r equired area f o r t he col l ec t or can be r edu c ed b y up to 5 0 % wi t h t h e use of f o r c ed co nvec t ion (B u ch in g er a n d We i ss, 2 0 0 2 ). If fol l ows t h at for c edco nvec t ion d r yers are a b le t o pro c ess t h e same amo u nt o f p rodu c t as n at u ral co nvec t ion dryers t hat h ave co l l ec t or areas that are six t i m es larg e r (B u ch in g e r a n d We i ss, 2 0 0 2 ). Cons i der ing t hese c ond it i ons, it is clear t h a t air flow ra t e is c ru c ial t o the overall syst em performanc e. T oo hi g h an air flow c ons u mes excessive fan power and too low o f a flow ra t e ca u ses p o or t h ermal p erformanc e. Addi t iona l l y, t h e effe c t of leakage s in c reases with t h e air flow r a te (Bu c hin g er a n d We i ss, 2 0 0 2 ). In genera l t h e p r essure d rop sho u ld also b e low t o keep the nec essary electrical p ower f or t h e fa n s as low as p ossib l e. When fans are imp l e m ented t o i m prove cir c ulation, t h e design of t h e d r yer requires a sl i ght modif i ca t ion since t h e c hi m ney may no longer be necessary (M r ema et al 1 9 8 7 ). In fa c t fo r ced cir culation c abinet dryers with o ut ch i m neys have been sh own to si g nif i ca n tly i mprove t he r ate of dryin g which min i m izes t h e c hanc es o f cr op damage due t o irre g ular dryi n g ( M umba, 1 9 9 6). E ven s o the establishm ent of fi x edspeed air f low si g n if i ca n tly red u ces t h e d r yer perf o rmance compared with more sop h ist i ca t ed, c o n t rol l ed air sp eed syst ems which use elec t ronic co n t rol l ers ( M umba,
48 1996 ). Howe v er, th ese electrical so ur c es are e i t h er u navai l able or unaff o rdable for small sc ale fa r me r s in t he d eve l oping world. For t h is reason, ph o t o voltaic (PV) cel l s h ave re c e i ved c onsiderable at t ention as an energy su p ply for fans. In fa c t, t he p erforman c e o f PV d ri v en syst ems exhibi ts an advant age of dependab i l ity over gr i d driv e n systems in so me dev e l opi n g communiti e s (Bu c h in g er a n d We i ss, 2 0 0 2 ). In t h ese PVp o wered systems, fans are d ir e ct ly c oup l e d to s olar pan e l s which results in si m ple and reliable syst e ms t hat operat e with o ut the integration of sop h ist i ca t ed a cc u mu l at o rs o r load co n t r ol l ers. A fluc t uat ing air flow rate is th u s es t abl i shed as c hanges in so lar radiation result in ac c e l eration or de ce l eration of the fan. Whi l e t his system h a s t he a d vantage o f a si m ple tem p erat u re c ont rol, t he co n t rol of air flow rate is l os t T h us a compromise must be made in desi g n co mp l e x ity and t he a b i l ity t o c ont rol air flow. Howe ver, t h e c ost o f so lar drying syst ems greatly increases with t h e integration of PV c e l ls. T his si g ni f ican t ly limits t h e implementat ion o f PVd ri v en sys tems in deve l opi n g communiti e s as t h e investment c apital of t h e d r yer bec omes t oo h i g h. T herefore, it is w i dely a c ce p ted t h at when gr i d power is availabl e it p r ovides a c heaper e l ec t ricity so u r ce (Bu c hin g er an d We i ss, 2 0 0 2 ). T h e abil i ty to c ont inua l ly p ro c ess c rops is importan t, howev er, a sign i f icant d isad vant age of solar dryers is t hat t heir use is l i m i ted during inc l ement we a t h er a n d t h e d ry i ng t i m e is c onsequ ently extende d In addition to l i m i ted t h roughpu t, so lar dry e rs c an also result in dec reased p roduct q ual i ty. A range of i m provem ents have been pro p osed and test e d for t h is pur p ose. In fa c t, f o cu s h as b een a p pl i ed in
49 addressing t he low heat t ransfer co effic i ent between co nventional air c ol l ec t ors and the flow i ng air stream. I m prov ements have been made by adding fin s mak i ng t he a b sorber V c orrugat ed, or by roughing t h e sur f ace of t h e a b sorber (Bu c hin g er an d W e i ss, 2 0 0 2 ). Add i tionally, a b sorbers have been equipped with t h ermal st orage c ompo n ents su c h as rock b eds, wa t e r desicc a nt, o r c o n c r ete. T hese st o rage compon ents c ol l ect a n d st o re h eat whi l e so lar radiation is ac t ing o n t h e syst em. T his heat is t hen diss i pat ed in t o t h e d r yer dur ing p oor wea t her c ond it i ons. G o swami ( 1 9 8 6) rec o mm e nded a st orage volum e o f 0 .15 t o 0 .35m 3 Anot h er met h od of ensuring adequat e heat abs o rp t i o n is the integration o f b acku p h eating c o mponent s. Add i tional heat is usef u l beca u se warmer air c an absorb more moisture a n d it helps t o raise the prod u c t t emperature which i m proves water m i grat ion t o t h e p rodu c t surfa c e. A g ricultural wastes su c h as p ee l s, husks and sh e l ls can b e used in combustion proc esses, b u t as Buching e r a n d We i ss (2 0 0 2 ) noted, b i omass, p artic u lar l y fuel wo o d, is t he most co mmon so ur c e of energy i n rural areas of deve l opi n g coun tries. However, in many c urrent implement a tions, t he f u el w o od is burned ineff i cient l y. T hus, t he d eve l opment o f simp l e and affo rdable c ombus t ion syst ems is necessary to c o mp l ement solar drying te c hno lo g i es when exposure to ra d iation is min i m a l. Howe ver, i m prov ements in solar d r y i n g systems inc r e a se t he c ost and the co mp l ex i ty of t h e desi g n. As a res u lt, t h e use of t hese more sop h ist i ca t e d solar dryers bec omes l i m ited in d eve l opi n g c o unt ri e s with t h e increased dependenc e on i m por t ed c ommercial c ompon ents and material s (Bakeka and Bi l gen, 2 0 0 8 ). Due to
50 t h ese constr a ints, a nd t h e subsequent co mprom i se in maintaini n g t h e small ca p acity of the dryers, t hey are l i m i ted to small sc ale o pera t ions. T h e so lar t u nnel dryer is effec t i ve l y an intermediat e stage d ryer in t erms of its so phistica t ion. W i th t he in t egration o f f o r c ed c onvec tion p ro c esses, t he t unnel dryer is dependent o n e l ec t rical power; t h erefore it o nly suited f o r in t ermed i ate sized farms o r small c oop eratives where e l e ct ricity is ava i lab l e or where investment ca p ital a l lows for i m p l e mentat i o n of alterna t i v e energy so u r ces. T he f o r ced c o nvec t ion con t rol in t he t u nnel dryer increases t he d r y i ng ra t e and results in a hi g her qual i ty prod u c t th an is ac h i e ved in t raditional, openair method s. In fa c t, so me st u dies have sh o wn t hat co mpared with tra d it i onal met h ods, h i g h moisture c ont ent fr u it c an be d ri e d in half t he t i m e (Bu c hin g er an d We i ss, 2 0 0 2 ). T h e major co mponent s of the t unnel dryer are a solar co l l ec t or and a d ryer co mpartment Add i tiona l l y t u nnel dryers are equ ipped with airf l ow s ystems which cir culate air with t h e use of fans powered by a PV pan e l a generator or c ent r al u t i l ity. T he p rincip l e mode of operat ion f o r t hese d r yers is si m i lar to t hat o f c abinet dryers. T u nnel dryers essentia l ly u se blowe r s o r f a ns to for c e air into t h e so lar c o l l ec t or where so lar radiation is used to raise t he temperat ure. T he air c ont inues flow i ng t hro u gh t he fo o d drying c ompar t ment where moisture is removed. Some t unnel dryer desi g ns incor p orat e t he u se of gas powered heat ing u nits t o promot e d r yi ng even d uring i n c lement wea t her. T he d ry i ng c hambers a r e of t en a c cessed by removing t he c overing manua l l y al t ho u gh some d es i gns are equipped with hand c r anks. T h e p r odu c t is spread on mesh which is suspended ac r oss t h e length of t h e dryer c ompar t ment. A typical t unnel dryer is depic t ed in Fi g ure 3 13
51 T u nnel dryers can either be cons t ructed as a permanent installation on t o p o f a f oun dat ion o r in a p or t able desi g n for mobi l ity depend i n g on t h e lo c al n eeds a n d cir cumstanc es o f t he target co mmun i ty ( C hen et al 2 0 0 9 ). The advant age of f o r c ed cir culation t unnel dryers is th a t air t h roughpu t can be altered b y t he sp eed o f t h e fa n s d epending on t h e amo u nt o f so lar radiation ava i lable. U n iform dryi n g is also est a bl i shed w i t h out t he n eed f o r turning t h e c rop (Bu c hing er an d We i ss, 2 0 0 2 ). These dryers c an eas i l y be adapt ed t o t he lo c al cl i mate and manufac t u r i ng demands of spe cif i c co unt ri e s, h owever t he integration o f f o r c ed convec t ion fa n s a n d gas p o wered bac k up heat ing results in hi g her inv estment Large Scale Dryers Solar d ryers of more sop h ist i ca t ed d es i gns will only be brief l y m entioned in t h is report as t h ese el abor a te desi g ns do not fall i n t o t h e co n t ext of smal l s cale syst ems for use in deve l opi n g c ommuniti e s T h ese c omplex solar drying systems requ ire s i gni f icant finan c ial in v estment a n d co nsiderab l y intensive c o ns t ruction p ro c esses. Additional l y, many of t h ese desi g ns incor p ora t e t he u se o f me c hanized syst ems, hi g h l y depen dent o n e l ec t ricity which is of t en lim ited o r a b sent in rural areas of devel oping nat ions. The sma l l scale syst ems des c ribed ea r l i er in t his re v i e w are c onsidered b et t er suited for fa r m l evel o r c o operat i v e use in rural c ommuniti e s o f developing c oun tries. Hen c e, t h ese more sop h ist i ca t ed systems, which include greenho u se dryers and in house dryers, sho u ld only be co nsidered in large scale, c o mm e r cial a p p l ications of deve l oping c oun tries. T h e basic desi g n of nat ural c irculation, solar greenhouse dryers c onsists of drying ra c ks made of w i re mesh spread ac r oss woo d en b eams. T hese ra c ks are
52 arranged in paral l e l rows with spa c e bet ween al l o w ing easy ac c ess t o t h e produ ct. T h ese ra c ks are l o ca t ed u nder fixed, slanted g l ass r o of s which al l ow solar radiat ion o ver t he p rodu c t and fr a me where it is absor b ed b y black c oat ed inter n al wa l l s. Rid g ed c aps are f o rmed o ver t he r oof s to p rovide ex i t vents for air which is of t en regulated by sh u t ters. A typical greenhou se dryer as d escribed here is sh o wn in Fi g ure 3 14. Anot h er t ype o f largescale, solar drying system is t he i n h o use d ryer which has si g nif i ca n t c o ns t ruction re q uire m ents, c ompl i ca t ed o pera t ion p r o cesses and re l atively h i g h inves t ment co mpared to p rev i ously discussed solar dryers. T hese in house d r yers c onsist of ro o f integrated solar c ol l ec t ors, drying bins and e l ec t ric mo t ors operat ed with ax i al flow fans. A t y pical in house dryer is sh o wn in Fi g ure 3 1 5 Aux i l i a ry heat so ur c es su c h as LPG gas burne rs h ave been integrated in some d es i gns (Sm i ta b hindu et al 2 0 0 8). T h ese dryers are c apable of a cco mmodating lar g e nu m bers o f st acked trays as sh own in Fi g ure 3 1 6 and t h erefore have greater process i ng loads t h an small scale dryers. T he exper i m ent a l performanc es o f so me in house dryers h ave also demonstrated significant reductions in drying times compared with open air sun drying. Add i tiona l l y t h e d r i e d pro d u c ts o f t hese systems have been determined t o be o f h i g her qual i ty. F ur t hermore, u niform air t emperature and prod u c t moisture co n t ent c an b e a c quired with e l ec t ric b lower integration (Janjai a et al., 2 0 0 8). As was discu ssed, so lar drying t ec h nology pr o vi des an a t tra c tive opt ion f or fo o d p reservation pur p oses t h at is clean, h yg i enic a n d establishes sanit a ry c ond it i ons t h at meet nat ional a n d internat ional st andar d s ( C hen et al 2 0 0 9 ). T h is i s a c c ompl i shed w ith either zero or l i m i ted energy use w i th a c tive and passi ve mode dryers respe c tively.
53 Ho w ever, solar drying is more t han just a subs t itu t ion for fossil f u e l s, but is a technology based proc ess f o r pro d u c ing d ri e d materia l s o f t he r equired q u al i ty. It also saves time, o c cu pies l e ss area, improves pr o duc t qual i ty, makes t he p ro c ess more effic i ent and prot ec t s t h e environment (Chen et al., 2 009). T h e c apac ity of a solar dryer ma i nly depends on t h e c r op itse l f a n d t h e sha p e. On t he o ne h and, it sh o uld be sma l l enough to ens u re t he p ro d u c t h as a dequat e t i m e f o r p reparat ion su c h as washing a n d sl i ci n g. O n t h e o t h er h and, it sh ould b e big enough to enable t he u ser t o generate in c ome. Additional l y, solar dryer systems infl u en ce t he marke t ing c apac ity and in c ome generating p o ten t ia l since a hi g her price can b e o b tained for prod u ct s of i m proved qual i t y T h erefore, t he o p por t unities pr o v i ded t h rough t h e development of low cost and local l y manufac t u red solar dryers, o ffer an a u spicious o p tion t h a t promises to sign i f ican t ly re d u c e losses asso ciated w i th p ost h arvest degradation. For t h ese reason s, effo r ts have been made over t h e la s t t h ree dec ades to d eve l op, d es i gn, and c o ns t ruct so lar dryers. In fa c t, so lar drying t ec h nology is cu rren t ly u sed t hrou gho u t t h e w o rld t o dry a w i de r ange of fo o d prod ucts. T h e f o l l owing d iscussion o f t his report serves t o o u t l i ne a n d des c ribe t h e di vers i ty of solar dryers cu rren t ly in u se. Case Studies A w i de r ange of small scale dryer des i g ns have been su g g ested for adop tion and i m pl e mentat ion in d eve l o p ing c ommunit i es depen ding on t h e t ype o f ma t er i als t h at are lo c al l y ava i lab l e as well as t h e mode of heat t ransfer emp l oyed by t he dryer. T h e use of solar dryers has in fa c t b een i m pl e mented in t h e d r y i ng app l i ca t ion o f various prod u c ts
54 in or d er to meet the needs o f smal l scale farmers fr o m arou nd t h e world. T herefore, a co mprehensi v e re v i e w o f t he d ist i n c tive desi g ns, p rinciples of operat ion, d e tails o f fa b rication, a n d d ry i ng c harac t er i stics of prev i ously repor ted solar dryers is present ed h ere. T h e p r opo sed so lar dryers revie w ed in t h is report were se l ec t ed f o r d iscussion b eca u se o f t h e i r low in i ti a l ca p ital co sts. Additionally, t hese syst ems were det erm i ned to meet re l atively sim p le fabricat ion req u ir e ments of t en with t h e use of u n c onv entional, local materia l s Fur t hermore, t hese systems are co nsidered easy t o operat e as they re q uire no sop h ist i ca t ed me chanical o r elec t rical c o mponent s. T h us, t h e designs discussed here a r e easi l y maintai n ed a n d require o n ly sim p le replacement of par t s d uring repairs. F ur t hermore, c o n sideration was given t o identify dryers which effec t i v e l y pr o mote i m pr o ved drying k inetics a n d result in a h i g her qual i ty prod u c t than is a c hie v ed via openair, sun dryin g Hence, t his rev i ew identi f i e s a n d provides brief surve y s o f so lar dryers t hat sa t isfy t h ese c riteri a. To p resent this information, a systematic appr o ach for solar dryer c lass i fication is pro p osed here. In t his c ont ext, t hree generic grou pin g s o f so lar dryers h ave been identified; simp l e o r d ir e ct so lar dryers, co nventional o r indirect so lar dryers a n d lar g e scale d r yers. S ever a l types of si m ple (di r ec t ) dryers are out l i ned here in c lud i ng t he b ox, t ent a nd se esaw dryers. Ho w ever, t h ese dryers are mentioned here o nly briefly since many of t h ese desig n s are co nsidered t o be relati v e l y ineff i cient and result in a d egraded pr o duc t q ual i ty co mpared with the c onvent iona l ind i rect syst ems. For t h is reason, emphasis is placed on the co nvention a l dryers, par t icular l y c abinet dryers sin ce t h ese desi g ns represent a reasonab le c ompromise bet ween
55 so p histica t ion a n d effic i en cy. T he c o nventional d r yers elabora t ed in t h is rev i ew are t h e c abinet dryer a n d t h e tunnel dryer. Largesca le systems are briefly rev i ewed in t his repor t ; however, t he h i g h in v estment sop h ist i ca t ed d es i gn, and increased power demand of t h ese desi g ns do not fall i n t o t h e sc ope of smal l sc al e co operat i v e use. An overv i e w o f d ryer c ategorization for t h is discussion is dep i ct ed in Fi g ure 3 1 7 Direct Solar Dryer Designs So d ha et al. ( 1 9 8 5) d es i gned and p ropo sed t he u se o f a so lar box dryer as shown in Fi g ure 3 1 8 T h e results show t h a t hi g h moisture c ont ent fr u it such as mango fl e sh, with a t hickness o f 1cm a n d an in i ti a l mo i st u re c ont ent of a p proximately 9 5% (wb) c ould be dried t o 1 3 % (wb) in only 12 h ours o f su nl i ght exposure. T herefore, it was det erm i ned from t his st u d y t hat t h e use of box t ype dryers c ould b e effec t i v e in d omestic appl i ca t ions f o r p ro c ess i ng high moisture co n t ent p r odu ct s su ch as f r uits a nd ve g eta b les. T he o veral l ef f ici e n c y of t he b ox type dryer was i m proved co mpared t o t h e efficiency of openair sun drying whi l e t he q ual i ty of t h e p r oduct w as also bet t er p reserved. Pande a n d T h anvi (19 9 1) eva l uat ed a solar dryer in co n jun c tion w i th a water h eater as shown in Figure 319 In t his sense, t h e syst em t h ey devel oped was able t o be u sed for e i t h er d ry i ng high moisture co n t ent prod u c e or for water hea t ing p urp o ses ex c lus i ve l y. E xper i ments sh owed t h at b etween 1 0 1 5kg o f h i g h moisture c o n t ent p rodu c ts cou l d be d ri e d in only 3 5 days. It w a s su rm i sed t hat t he d ryer co uld pr o cess appr o x i mately 5 00 kg o f p rodu ct ann ual l y. Mwithhiga and K i go ( 2 0 0 6) deve l oped and eva l ua t ed a solar b o x d r yer which was designed with l i m ited, so lar radiation tra c ki ng abi l it y as shown in Figure 320 T h e
56 so lar absor b er p late was fabricat ed f r om mi l d steel with a tra n sparent c over o f p oly v inyl c hloride (P V C ) T h e t ra c ki ng c apabilities a l lowed t he syst em to b e adjus t ed in 1 5 increments to fol l ow t h e sun througho ut t he d ay. E va l ua t ion o f t he d ryer perf o rmance was c ond u c ted with e i t h er o n e t hree, f i ve or nine adjus t ments of t h e a n g l e made each d ay with e i t h er no load c ond it i ons or loaded w i th c offee beans. T he d r y i ng c hamber was fou n d to reac h 7 0.4C which al l owed for co ffee bean s t o be lowered from a n in i tial mo i st u re c ont ent of 5 4 .8% (wb) to a p proximately 13 % (wb) in o nly 2 days. T h ese results ind i ca t e an i m provement over o penair sun d ry i ng which requires appr o x i mately 5 7 days. How e ver, it was determined t hat n o si g nif i ca n t i m provement in drying d ura t ion was o bserved with t h e so lar tr a cking abi l ity of t h e d r yer. A domestic solar dryer was designed, co ns t ru c ted, a n d c hara c terized by E n ebe a n d Ezekoye (2 0 0 6 ) f o r t he d ry i ng of pepper and gro u ndnu ts during h i g h humid i ty and low temperat ure periods in N i g er i a T h e d r yer was fabricat ed f r om lo c al l y avai l a b le material in N i g e r ia su c h as h ardwoo d, Perspex gl ass f o r gla z in g a n g l e iron for ske l et o n, sc rews, w i re mesh and ply w ood as shown in Figure 32 1 T h e effe c tive collector area was 1.255m 2 and was tilted 22.9 while the volume of air needed for drying was .0.24m 3 The optimal temperature of the dryer was determined to be 67C with a relative humidity of 43%, corresponding to an ambient temperature of 31C. Experimentation resulted in the drying of pepper to a moisture content of 56.2% (wb) and groundnuts to 40.53% (wb) in 8 days and 5 days respectively. The average collector efficiency was estimated to be 10% while the dryer efficiency was found to be significantly improved at 22%.
57 T h is stu d y exhib i ted t h e advant age of pr o cessing at low t emperatures which effec t i v e l y prevented c r acking and su bsequent exp o sure to fungal, b a cterial a n d insect infestation. T he p roduct was also sh own to maintain its nat ural co l or. T he p o t e n t ial f o r large scale, co mm e r ci a l use o f t he d ryer by enlarg i n g t h e c ol l ec t or area a n d increasing t he n u m ber of trays was also n o ted. Sin g h et al. ( 2 0 0 6) desi g ned a multi shel f sol ar d ryer w hich co nsisted of three p erforat ed trays w i th a d jus t ment ca p abi l it i es t o al l ow var i a b le in c l i nat i on for different seasons as shown in Figure 322. Intermediate h eating between t h e t rays al l owed for uniform d r yi ng and experiments d etermi n ed that a max i mum stagnat ion temperat ure of 1 0 0C was established during t he d ry i ng of 1kg p o wder f o rms of chil i gar l ic, g i n ger, mango, co riander, onion, and fen u greek leaves. T he u se o f t his dryer was fou n d su itable for domestic l e ve l purp o ses u n der hyg i enic c ond it i ons. Sin g h et al. ( 2 0 0 4) also desi g ned and eva l uat ed a por t a ble solar b o x d r yer with a multi she l f desi g n w i th d ry i n g air interme d iate l y heat ed b etween t h e t rays as shown in Figure 32 3 This d es i gn pr o moted uniform d r yi ng among t he t rays and experimentat ion det erm i ned a maxi m um s t agnation temperat ure of 7 5 C was established during dr y ing o f fe n ugreek l eaves which w ere effe c tive l y pre s erved t o a sh e l f l i fe greater t h an one year. T he d ryer was fou n d t o b e e c o n omically viab l e and c ould p o t ential l y enable far m ers in remote places t o increase t he value o f t h e i r p rod u ce. Indirect Cabinet Designs (Passive) A si m p l e so lar dryer was des i gned and c ons t ructed by Bolaji a n d Olalusi ( 2 0 0 8) co nsisting o f a black painted, alum i num p late heat a b sorber mount ed o n a wel l seasoned wood fr a me as shown in Figure 32 4 F oam ma t er i al w i th t hermal
58 c ond u c tiv i ty of .04 3 W / mK was placed in the spa c e bet ween t he inner and o u t er box and a mesh sc reen was placed half way bet ween t he transparent, glass co ver and the absor b er p late t o p rovide improved air hea t ing b y abs o rbing addition a l solar rad i ation. E xperi m ent a tion fou n d hou rly variation of the internal cabinet temperature to be significantly elevated compared with ambient air and was actually observed to attain an excess of 74% after several hours. The drying rat e and collector efficiency for drying yam chips in this study were 0 .62 kg / hr and 57.5% ) respectively. The results of this evaluation demonstrated sufficie ncy in quickly drying food products to adequately preserved moisture levels. E z ekwe (1 9 8 1 ) evaluated a t ypical ca b inet d ryer modif i ed with a w o oden plenum t h at served to gui de inc omi n g air. An extended c hi m n e y was also inc o rpor a ted in t h is dryer to in c rease t he n a t u ral cir c ulation. T he r esulting ra t e o f d ry i ng was fou n d to b e app roximatel y f i ve t i m es g r eater t h an openair sun dryin g O t hieno et al. (1 9 8 1 ) d eve l oped an ind i rect so lar dryer which c onsists of a sing l e g lazed air hea t er with effec t i v e area of 1m 2 f o r t he a b sorber. A d ry i n g chamber equipped with a c hi m ney was co ns t ru c ted fr o m har d b oard and was connec ted to t h e air heat er. T h e air hea t er was modif i ed t o ac c ommodat e several l ayers of absor b er mesh by in c reasing t he width o f t he air gap. T ests co nducted w i th 9 0kg o f wet maize resulted in drying fr o m 2 0 % t o 12% (wb) within 3 d ays. Pangav hane et al. ( 2 0 0 2) prop osed, d es i gned and evaluated t h e use of a solar dryer w i th an alum in u m foil a b sorber, glass co ver and a d ry i ng c hamber made of GI sheet as shown in Figure 32 5 Gra p es were su cc essful l y dried in 4 d ays c ompared
59 w i th t raditional meth o ds such as openair sun drying which resulted in 7 days of drying which c orrespo nds to a r eduction in d ry i ng t i m e o f 4 3%. Li et al. (2 0 0 6 ) investi g ated a solar dryer as shown in Figure 3 2 6 which was designed w i th an effe c t i v e c ol l ec t or area o f 6 m 2 f o r t he d ry i ng of salted greengag e s During exper i m ent a tion, f u l l y w e t pro d uct was iso l ated fr o m sem i dried pr o duc t to improve efficiency. Res u lts indica t ed t hat effecti ve d r y i ng of salted greenga ges was redu ced t o 15 days c ompared w i th 4 8 days n e cessary for openair su n dryin g Goyal and Tiwari ( 1 9 9 9) prop osed and a n al y zed a mo d if i ed, indirect syst em referred to as t h e reverse absor b er c abinet dryer (RA C D) in which a d ownward fa c ing a b sorber and a c y l indr i cal reflec t or were placed below t he d ry i ng c hamber. A si n g l e w ire mesh t ray held t he p rodu c t in t h e d ry i ng c hamber and t h e glass absor b er c over was incl i ned at an angl e of 4 5 to receive maximum ra d iation. T he cyl i ndrical re f l e ct or redirected t he so lar radiation t o w a rd t h e abs o rber where inc o m i ng air was hea t ed as shown in Figure 3 2 7 Indirect Cabinet Designs (Active) Sreekumar et al. (2 0 0 8 ) developed a nd eva l ua t ed an efficient ca b inet d ryer with a p roduct load arrangement beneath t h e abs o rber plate which p revents discolora t ion by avoid i ng direct so lar irrad i ation as shown in Figure 32 8 W h i l e t his des i gn is si milar t o integrated b o x d r yers, it is mentioned here in t his report d ue to t he essen t ial separ ation of the hea t ing a n d drying component s. Two axial flow f ans were used to ac c e l erate t h e d r y i ng ra t e, a co l l ec t or g l azing was incli n ed f o r maxi m um abs o rp t ion o f so lar radiation, and six p erfora t ed trays were u sed f o r ma t er i al l oading. T ests resulted in a t emperature of 9 7 .2C at t ained by t h e abs o rber p late u n der no load co nditi o ns,
60 which correspo n ded to a maxi m um d r yer temperat ure of 7 8 .1 C It was determined t hat within 6 h ours, 4 kg of bit t er gou r d co uld be d ri e d from a n initial moisture content of 9 5 % to 5% (wb) w i t h out t he loss of p roduct c olor, whereas openair sun drying re q uired 1 1 hou rs. Chen et al. (2007) evaluated the performance of a forcedflow cabinet dryer for preserving banana chips. The banana chips were reduced from a moisture content of 7585% (wb) to approximately 7 8% (wb) in 5 sunlight days with a thermal efficiency of the dryer reported to be 30.86%. Better quality product was achieved compared with product evaluated with open air sun drying Shorter drying periods were also accomplis hed. Tiris et al. ( 1 9 9 5) desi g ned and eva l uat ed a solar dryer as shown in Figure 329 using c hi l i peppers, sweet peppers, sultana grapes and green b eans. T h e results of t h is st u dy indica t ed t hat t h e d r yer c o uld si g nif i ca n tly reduc e t he d urat ion o f d ry i ng and essen t ia l ly p rovide a h i g her qual i ty pro d u c t t h an o penair sun d ry i ng. T h e t hermal efficiency of t h e d r y i ng c hamber a n d t h e s olar air hea t er were discussed in relation to t he physical parameters of t h e d r yer as we l l as t he r e sulting p rodu ct q ual i ty at different air flow ra t es. T h e results ind i ca t ed t hat overall d ry i ng perf o rmance was inc r eased with hi g her flow ra t es and t he drying syst em was f o und to h ave a t h ermal efficiency in t h e ra n ge of 0.3 t o 0 .8. M o hanraj and Chandrasekar ( 2 0 0 8) d eve l oped a f o r c e d c onvec tion solar dryer as shown in Figure 33 0 which cons i sted of a 2 5 ti l ted, flat plate c ol l ec t or of area 2m 2 c onn e ct ed to a d ry i ng c hamber f o r d ry i ng c opra. A c ent r ifu g al fan was used t o f or c e air bet w een t he glass cover and the a b sorber below To ensure air cir c ulation was uniform
61 a c ross t h e abs o rbe r a di v ergent e n try was f o rmed f o r the so lar air heat er. It was determined t hat in 8 2 hours c opra cou l d be d ri e d from a n initial moistu r e c o n tent o f 5 1.8% t o 9.7% and 7.8% (wb) at the t op and b o t t om o f t he d ryer respe c tive l y, wh i le t he t hermal efficiency was determined t o be 2 4%. Mumba (1 9 9 6 ) designed and evaluated t h e perf o rma n ce a so lar c a binet dryer as shown in Figure 3 3 1 which was equipped with a PVpowered DC fan t h at perm i t t ed p assi ve c ont rol of t he temperature of drying air. T h e so lar co l l ec t or wa s ti l ted 1 5 and wa s co mpr i sed o f b lackened sisal absor b er mesh which help ed t o i m prove heat t ransfer. Add i tiona l l y, t h ree t ranspare n t c overs o f a Tedlar/Teflon c ombination al l owed for hi g h shor t wave tra n sm i t t anc e and re d u c ed h eat loss from the u pper s urfa c e. Fur t hermore, t ransparen t insulation material ( T IM) was placed between t h e abs o rber and t he c overs t o assist in raising t he t emperature. D ry wood sh av i ngs were used for t h e air hea t er wall insulat i on and all su rfa c es were painted black to increase t h e heat gain. It was determined t hat t he d ry i ng air t emperature of t his system h a d an upper l i m i t of appr o x i mately 6 0 C w h ich effec t i v e l y prevent ed o verh e ating and c ra c ki ng of t h e grain. F rom performanc e t esting, it was determined t hat t h e syst e m c ould d ry a batch of a b out 9 0kg maize from 3 3 .3% (db) moisture c ont e n t t o below 2 0 % (d b ) in only a day. T h e t hermal efficiency was 7 7% and t he opt i m um ra t io of PV t o so lar hea t er area was det erm i ned to b e 0.22 t o a c hie v e a co s t effec t i v e d r yer desi g n. T h is dryer resulted in a bet t er p reserved pr o duc t co mpared with openair sun dryin g t h erefore, t h is dryer was determined t o be su itable for
62 appl i ca t ion in r u ral co mmun i ties where e l ec t ricity and f u el is unavai l a b le o r t he c ost is t o o hi g h. Al Juamily et al. (2007) developed a solar dryer system as shown in Figure 3 3 2 which was comprised of a drying cabinet, solar collector and an air blower connected with piping for processing high moisture content produce such as f ruits and vegetables. The absorbers of the solar collector were V corrugated and had glass covers providing an effective area of 2.4m 2 Results indicated that apricots were reduced from an intial moisture content of 80% to 13% (wb) within 1.5 days, grapes from 80% to 18% (wb) in 2.5 days, and beans from 65% to 18% (wb) in only 1 day. It was determined tha t the primary factor affecting the drying rate is the temperature of the drying air whereas the airflow variation was discovered t o be negligible since the relative humidity of exhausted air was only between 25 and 30%. Sarsilmaz et al. (2000) investigated the use of a rotary column cylindrical dryer (RCCD) with a fan and rotor system to rotate product trays within the cabinet as sho wn in Figure 33 3 The optimal airflow and speed of rotation for processing apricots was determined in order to reduce drying times and maintain uniform and hygienic conditions. It was found that 23kg trays of apricots were effectively dried to a moisture content of 25% (wb) in half the time needed to dry them compared to openair sun drying. The products were also found to be more attractive in color with uniform quality among trays. However, this design is limited in use due to the additional power demand of the rotary system. Pawar et al. (1995) developed and tested a largescale solar dryer as shown in Figure 3 3 4 that was loaded with coriander powder for performance
63 analysis. The drying system consists of 3 cabinets and an array of 40 solar collectors with an integrated blower to establish air flow. This dryer was shown to be feasible and results indicated that substantial amounts of fuel c ould be saved compared with larger, industrial systems. Additionally, the dried product was found to be untainted a nd was actually dried quicker than with open air sun drying. As a result, the use of forcedconvection solar dryers was found to be suitable for use in the food industries of developing communities. Supplemental Heat D ryers A laboratory scale, passive drye r as shown in Figure 33 5 was developed by Irtwange and Adebayo (2009) and was comprised of a solar collector, a drying chamber made of wood and a thermal storage unit. The collector was constructed from corrugated zinc roofing sheet and covered with a g lass sheet for glazing while the chimney was fabricated from metal tube and corrugated roofing sheet. The thermal storage unit consists of black painted gravel for better heat absorption and trays were made of stainless steel wire to avoid rusting. The dry er was evaluated using 10 kg of maize which took 4 days to dry from an initial moisture content of 32.8% to 13.1% (wb). Compared with open air sun drying which took 8 days to dry to 13.4% (wb), this dryer system clearly resulted in a faster drying rate. It was noted that the dryer could be enlarged for community level, cooperative use and for further improvement; the airflow rate could be increased. Ayensu (1997) developed an inexpensive, solar dryer that operates at low temperature and is considered simple to operate. The dryer consists of a thermal, rock storage component and a drying chamber equipped with a chimney as shown in Figure
64 3 3 6 Only locally available materials such as wood and scrap metal were used in the fabrication of the dryer. Chick en wire mesh was used to form a double layer food bed. Plywood access doors were overlapped to prevent air leakage and a glazing was integrated on top of the drying chamber to provide additional heat. It was found that the solar collector transferred approximately 118 W / m 2 of thermal power to the drying air, which allowed heating of the air to 45C at 40% relative humidity from ambient conditions of 32C and a relative humidity of 80%. Consequently, various crops (cassava, groundnuts, okra, etc ) were dried to moisture content levels below 14% (wb) which enabled preservation in excess of one year without biological degradation. Enibe (2002) designed a passive air heating system consisting of a heat storage system integrated into the solar collector as shown i n Figure 3 3 7 The storage system was comprised of distinct modules of phase change material (PCM) that were equally spaced across the absorber plate and served as parallel air heating channels. Testing of the system was conducted over environmental temper ature ranging from 1941C while global irradiation was measured between 4.919.9 MJ / m 2 Results indicated successful application for crop drying. Ezeike (1986) developed a modular, solar dryer with a dehumidification chamber and a triple pass air collector as shown in Figure 338. The dehumidification chamber is essentially a frame containing perforated trays that are loaded with silica gel. The collector surface was equipped with baffles to assist in uniformly distributing the heated air. The cabinet was m odified with wall collectors to provide additional heat gain. Through experimentation, the outlet temperatures were found to be between 90 to 101C with wind speeds of up to 3.5 m / s The dryer efficiency was determined to be
65 between 7381% for the drying of rice paddy from 25.93% to 5.31% (wb) in 10 hours and yam slices from 64.90% to 10.66% (wb) in 31 hours. An openair sun drying control resulted in significantly increased drying time and lower quality product. El Sebaii et al. (2002) developed and evaluat ed a natural convection dryer as shown in Figure 3 39 which consisted of a drying cabinet connected to a flat plate air heater that was designed to allow interchanging of storage material to improve performance. Testing was performed with and without sand as a storage material for an assortment of vegetables and fruits such as apples, figs and grapes. The storage material was found to reduce the drying process by 12 hours in some cases with 10kg of product loaded. Mohamed et al. (2008) investigated the dryi ng kinetics of edible red algae using an indirect, forcedconvection solar dryer as shown in Figure 3 4 0 The dryer was equipped with a circulation fan and an auxiliary heater to gain greater control over the drying parameters. Experiments were conducted at a range of temperatures and some variation in relative humidity and airflow rate was introduced. Most drying was observed within falling rate period and it was concluded that the temperature of the drying air was the primary influence on the drying kine tics. Zomorodian et al. (2007) introduced an active solar dryer with a semi continuous, timer assisted discharging system as shown in Figure 34 1 Multiple solar air heaters, tilted at 45, were employed in this design to provide a total area of 12m 2 The system also consisted of an air distribution system and an auxiliary heating channel. A timer controlled, electrical rotary discharging valve wa s installed at the bottom of the dryer.
66 Experiments were conducted with rough rice kernels in order to investig ate the effect of both the rate of airflow and the discharging time interval on the drying rate. The capacity of the dryer was approximately 132kg of rough rice with 27% (db) initial moisture content, which required 3 hours to achieve a moisture content of 13% (db) The overall dryer efficiency was 21.24% and the average temperature of the drying air was determined to be 55C. Furthermore, it was found that the auxiliary heating channel used only a fraction of the solar energy (68%). A small scale dryer consisting of a drying, heating and plenum chamber was realized by Singh (1994) as shown in Figure 3 4 2 Aluminum sheets, mineral wool insulation and galvanized iron sheet form ed the walls of the drying chamber which contained 20 aluminum framed, nylon mesh trays stacked in two rows. To assist is uniform distribution of air; wire mesh wa s incorporated between the drying chambers and plenum. This mesh also prevent ed foreign contaminants from entering the system. Mild steel sheet ing wa s used to form the heating chamber which housed 16 heating elements controlled by thermostat. Additionally, an electric motor was used to drive a blowing unit and thus establish control of the airflow. Testing was carried out while maintaining a cons tant air intake of 0 .33m 3 / s Under this condition, it took 11 hours to lower the moisture content from 92.5% to 8.69% (wb), 14 hours from 92.65% to 8% (wb), and 12 hours from 90% to 7.5% (wb) respectively for 50kg batches of cabbage, cauliflower and onion slices. The overall drying efficiencies for processing these crops were 28.21%, 30.83% and 29.51% respectively.
67 Sarsavadia (2007) developed a forced convection dryer as shown in Figure 3 4 3 for drying onions with recirculation of exhaust air under controlled rates of airflow and air temperatures. It was determined that the energy needed for drying was reduced as air temperature increased. Conversely, higher flow rates caused an increase in th e energy needed for drying. The total energy required to dry onions from 86% (wb) to 7% (wb) without recirculation, ranged from 23.548 to 62.117 MJ / kg water The contribution of energy from the electrical heater, the solar air heater and the blower were found to range from 41.0% to 66.9%, 24.5% to 44.5% and 8.6% to 16.3% respectively. Chaudhri et al. (2009) designed a solar cabinet dryer for processing onion flakes using a packed rock, thermal storage bed to supply heat and recirculated exhaust to dry the product. The absorber was created with corrugated galvanized iron ( GI) sheet painted black with a cover made of UV stabilized plastic sheet. It was determined that onion flakes could be dried from 85.5% to a moisture content between 9.56% and 6.76% (db) for cases with and without recirculation respectively, in under 7 hours. These results indicate that under recirculation conditions, less moisture is removed. Therefore it was concluded that the drying potential was reduced regardless of the temperature when exhaust air was recirculated. The efficiency of the dryer was determined to be 19% and 23% for the cases of with and without recirculation respectively. Furthermore, it was noted that a 74% increase in thermal efficiency was observed in the case without r ecirculation. The dried onion quality was also found to be of higher quality in terms of color for the without recirculation test. From these results, it was concluded that recirculation procedures may only be feasible for processing products with low mois ture content or with the introduction of desiccated material.
68 A solar biomass cabinet dryer as shown in Figure 34 4 has been developed by Bhandari et al. (2005),which consists of two solar collectors, a drying chamber, a heat exchanger, a chimney, a flue gas outlet and a mild steel biomass stove that was installed adjacent to the collector system. The solar collector wa s comprised of a corrugated aluminum absorber plate, a glass window glazing, and a collector box formed with glass wool sandwiched between G I sheets. Nine stainless steel, wire mesh trays we re supported by a wooden drying cabinet over which the chimney wa s formed from GI sheet. The roof of the drying chamber was equipped with an additional solar collector to provide supplemental heating. All s urfaces in contact with the heat air were painted black to increase heat absorption. Stove heated flue gas heats the incoming, ambient air as it passes through the heat exchanger. The flue gas then exits from an outlet pipe, while the heated, ambient air passes into the drying chamber, is forced through the product and exits out the chimney. The heat exchanger wa s fabricated from aluminum sheet and wa s provided with fins to improve heat transfer, while the flu e gas outlet was formed from GI sheet. Rubber gaskets were used on all the joining components of this system to ensure airtight joints. The effective drying time of 10kg of cauliflower was determined to be 10 hours over the span of two days with a dryer ef ficiency of 16.32%. Therefore, this dryer was found to be efficient and technically feasible when compared to similar dryer types with reported dryer efficiency of only 9%. The dried product was also found to be higher quality with this dryer compared to t raditional techniques.
69 Prasad et al. (2006) designed and evaluated a hybrid, biomass dryer as shown in Figure 34 5 that could generate continuous heated airflow between 55 and 60C for processing turmeric rhizomes. The quality of dried products was evaluat ed with both water boiling and slicing pre treatments. The products of both pre treatment methods were found to be similar in respect to physical characteristics and required only 1.5 days to dry. The efficiency of the dryer was determined to be 28.57%., while open air sun drying took 11 days to dry and resulted in a lower quality product. Madhlopa and Ngwalo (2007) developed an indirect solar dryer equipped with a biomass heater and a solar collector with a thermal storage system as shown in Figure 3 4 6 The biomass burner was designed with a small, rectangular duct leading to a flue gas chimney for exhaust, while a conventional chimney provided exhaust above the drying chamber. The collector was equipped with a flat absorber made of black painted concret e. Additionally, an integrated rock pile was incorporated to serve as thermal storage. Results indicated that the thermal rock pile was able to effectively store heat from the biomass burner and absorbed solar energy. Drying of 20kg batches of fresh pineapple was successfully conducted under exclusively solar operation, biomass only operation, and a hybrid mode of operation. It was observed that drying proceeded during inclement weather only with use of the biomass heater. The pineapple slices were reduced from a moisture content of 66.9% to about 11% (db) with the biomass burner mode of operation. During the last day of processing, the average efficiencies of moisture pickup were recorded as 15%, 11% and 13% for the exclusively solar, biomass only and hybri d modes
70 respectively. Based on these results, it was determined that high moisture content products such as pineapples could be adequately preserved by the dryer. Desiccant Integrated Dryers Shanmugam and Natarajan (2006) developed a forced convection solar dryer as shown in Figure 3 47 which was comprised of a flat plate solar absorber and a desiccant component holding 75kg of CaCl 2 based desiccant which is composed of bentonite, CaCl 2 vermi culite and cement mixed in the ratio of 6:1:2:1. Experiments were carried out under hot, humid conditions for the preservation of green peas at various airflow rates and the results indicated that equilibrium moisture content could be reached between 14 to 22 hours depending on the airflow. A pickup efficiency of 63% was observed while the performance of the system was found to be satisfactory for uniform desiccant drying. Thoruwa et al. (1996) developed and evaluated a solar cabinet dryer as shown in Figur e 348 which was capable of holding 32.5kg of bentonite (CaCl 2 ) desiccant enclosed in individual 250g bags. A maximum of 14.6kg of water was expected to be absorbed by the desiccant as determined with preliminary experiments. To achieve the optimal solar c ollection, the desiccant bed and Tedlar glazing panel were inclined at 15. This angle also assisted in the prevention of rain leakage. Experiments were conducted without load in order to access the dryers ability to deliver dehumidified air which could subsequently be used in external applications. While exposed to solar radiation, a small, PV powered fan drew air through an entry port where it passed through the desiccant bed before being
71 exhausted into the ambient air. During periods of no solar radiation exposure, this valve was closed, thus forcing the dried air over the food product. Shanmugam and Natarajan (2007) also developed and evaluated an indirect forcedconvection cabinet dryer as shown in Figure 349 which was equipped with a desiccant system for processing pineapple slices and green peas. The integration of a reflective mirror on the desiccant bed was found to significantly increase the drying potential. In fact, a 10C temperature rise was achieved with the addition of the mirror. Thi s temperature rise resulted in a reduction of the drying time for green peas by 2 hours and pineapple slices by 4 hours. Compared to both openair sun drying and other desiccant integrated processes, this system exhibi ted higher relative pick up and average thermal efficiencies. Furthermore, the drying rate wa s comparatively quick and the products from this dryer were observed to be uniformly dried. Product quality was good in terms of both color and minimal microbial decay. Furthermore, continuous operatio n of the dryer in excess of one year was possible as the desiccant material remained stable. Indirect Tunnel Dryers A forced convection solar tunnel dryer as shown in Figure 350 was developed by Hossain and Bala (2007) and was evaluated under tropical conditions in processing hot red and green chilies. Red chili was reduced from a moisture content of 85% to 5% (db) in only 20 hours whereas open air sun drying was found to achieve 9 % (db) in 32 hours. Green chili was also reduced from an initial moisture content of 7 6% to 6% (db) in 22 hours whereas openair sun drying was found to achieve 10% in 35 hours. Thus, the use of the solar tunnel dryer following blanching pre treatment, was found to reduce
72 drying tim e considerably. Additionally, t his proc ess w as fo und to result in better quality color and pungency compared with products processed with open air sun drying. Bala et al. (2003) also investigated a solar tunnel dryer as for processing pineapple in Bangladesh. The dryer wa s comprised of a flat plate collector with a clear plastic cover and a tunnel connected in series. A solar module was used for operation of two DC fans to supply hot air. The loading capacity of the dryer was estimated between 120150kg of pineapple and experiments resulted in greatl y reduced drying times compared with openair sun drying. Datta et al. (1998) developed a thermal performance model for a tunnel dryer to assist in evaluating the drying behavior of high moisture content products. In essence, this study entailed the development of a transient, one dimensional model for a tunnel dryer while the performance for natural convection mode was estimated. From the model, it was determined that large quantities of barley could be dried to equilibrium moisture content in the system w ithin only two days of operation. The proposed tunnel dryer wa s comprised of a solar collector and a connected drying chamber as shown in Figure 3 5 1 An absorber was fabricated from black polyester fabric with UV stabilized polycarbonate sheet forming a c over. The performance analysis of this system was presented in terms of the air collector as well as the drying chamber. It was determined that approximately 1941 kg of barley could be processed from 34% moisture content to between 1014% (wb) within only two days. The potential adaption of this system to drying processes of other crops in various locations was noted.
73 Amir et al. (1991) modified a multi purpose solar tunnel dryer as shown in Figure 3 5 2 for locations with tropical weather conditions. This m odified dryer was comprised of a solar collector, a centrifugal blower and a tunnel chamber built on a wooden structure which effectively prevented water intrusion. The dryer was also equipped with a biomass furnace and heat exchanger for operation during inclement weather. The absorber was fabricated from a thermostable, polyester fabric covered with UV stabilized transparent foil. Additionally, palm fiber insulation was used to reduce heat loss and plastic foil was installed to minimize the leakage of water and heated air. Sealing foil and bamboo mats on the chamber floor, eased evaporation of water especially during the first drying stage. Investigations of the system performance demonstrated a considerable reduction in drying time and product quality was improved overall. More specifically, results indicated that the drying time of cocoa, coconut and coffee was reduced by up to 40% compared with open air sun drying. Furthermore, the products were found to meet both national and international market standards as well as the moisture content needed for storage. It was also determined that raising the ambient temperature from 30 C to 70 C in this system was possible at peak conditions. Furthermore, the heat storing capacity of the concrete foundation was f ound to contribute to a temperature rise during the night. Further discussion was made regarding the potential adaptation of this system to different farm sizes and cooperative use areas. It was also suggested that alternative energy sources should be investigated in order to become independent from electricity.
74 Large Scale Solar Dryers Smaller natural circulation greenhouse dryers have also been developed, constructed and tested by Koyuncu (2006). These dryers wer e comprised of corrosionresistant plastic mesh, black coated absorbers and polyethylene sheet on a frame of black coated metal bars as shown in Figure 35 3 To determine effects on airflow, the dryers were tested both without chimneys and with galvanized iron chimneys. It was determined that natur al circulation greenhouse dryers exhibited higher efficiencies than open air sun drying by a factor of 2.5 for the drying of pepper. Additionally, it was found that the black coated absorber and chimney improved the dryer performance. Forced convection gre enhouse dryers as shown in Figure 35 4 have also been developed and evaluated by Condori et al. (2001). The primary advantage of these dryers is that almost continuous production can be established since fresh product can be introduced as dried product is removed from the opposite end of the dryer. Additionally, these dryers result in lower labor costs than some simple systems since they are partly mechanized. The ease of conventional heater installation was noted which could assist in maintaining a constant production rate with significant reduction in energy consumption compared with other sophisticated dryers. Construction Selection of an appropriate solar dryer for a given food product is dependent on product characteristics, quality requirements and relevant economic factors (Kandpal et al., 2006). Additionally, consideration should be given to material selection which dir ectly impacts the quality of the product. Attention is given here to identify adequate
75 evaluation procedures of the product, dryer, and performance parameters that provide an overview of the necessary components of analysis which help in the selection of a ppropriate dryer systems. A synopsis for evaluation procedure is then outlined from the preparation of dryer, product and instrumentation; to the actual experimentation. When selecting materials for dryer fabrication in developing countries, consideration must be given to the available resources, including equipment or tools, as well as the level of craftsmanship employable. Additionally, all materials should be resistant to heat, light and dampness to improve the lifetime of the dryer. Specifically, collectors and drying chambers should be watertight while absorber surfaces should be resistant to heat and moisture. Synthetic material used for covers or glazing must be resi stant to UV light and high temperatures. Commonly used materials (as reported by studies reviewed in this report) are listed in Table 3 3. Evaluation Sufficient testing of dryer systems is necessary to evaluate technical performance and establishes a basi s for comparison to other dryer designs. Such analysis can assist in the selection of appropriate dryer designs given specific conditions which must be met. Furthermore, adequate evaluation can provide an indication of a dryers performance in conditions d ifferent from those tested in. Leon et al. (2002) proposed a comprehensive procedure for the evaluation of solar dryer performance which provided methodology and test conditions in order to develop a standard practice of analysis. The procedure provides an explicit evaluation of solar dryer performance while facilitating in the comparison of different solar food dryers. The parameters reviewed here provide a basis for comparison to different dryers and can assist in dryer selection to meet specific needs. A nalysis of the product quality
76 involves evaluation of rehydration, sensory elements and chemical procedures. Evaluation of dryer parameters should include consideration of dimensioning and sizing, solar collector, construction requirements, drying temperat ure, relative humidity and the airflow rate. Performance parameters include drying time, efficiency, maximum drying temperature and cost associated factors which are all detailed further in this review. Product Evaluation As mentioned earlier in this review, the physical properties of a product may be affected by drying and result in changes in size, shape, color, and texture while chemical and enzymatic conversions take place. For instance, adverse color changes occur in dried mango slices as a result of e nzymatic browning. This discoloration is often prevented with careful control of temperature and moisture parameters particularly in more sophisticated drying systems. However, simple solar drying systems do not necessarily provide a significant level of c ontrol for these parameters. For this reason, sulfites and other preservatives are frequently employed to control the enzymatic browning of mango slices. Similarly, citric acid can be supplied from fresh lemons where commercially available preservatives ar e unavailable. T he assessment of dried product quality is necessary to establish a basis of comparison between drying systems. However, as Kader (2005) noted, it is far more challenging to measure qualitative losses than quantitative. Analysis of quality characteristics such as consumer acceptability, edibility, caloric and nutritive value are often neglected in studies due to the challenge in developing and understanding adequate evaluation parameters.
77 While reduction of quantitative losses is a higher pr iority than qualitative losses in developing countries, the opposite is true in developed countries where consumer dissatisfaction with produce quality results in a greater percentage of the total postharvest losses (Kader, 2005). However, by reducing qua litative losses through better implementation of preservation processes, a higher quantitative level can be maintained. Thus the quantitative parameters worth mentioning here are rehydration, sensory evaluation and chemical tests. A products ability to regain the original volume by soaking in water, serves as a good indicator of product quality since some dried products are consumed after rehydration. The common index used to express rehydration is the rehydration capacity, which is the ratio between the product weight after and before rehydration. The dried product will be graded best if it approaches the original fresh product volume. It has been generally observed that reducing the drying time and pretreating the product using additives like salt, sugar and glycerol improves the rehydration quality of dried fruits and vegetables (Leon et al., 2002). Rehydration tests can also serve to indicate the damage inflicted to the product caused by drying and pretreatment. In fact, Khedkar and Roy (1990) observed a higher rehydration ratio in cabinet dried raw mango slices compared to sundried slices, and attributed this to less rupture of cells during cabinet drying (36.4%) than sun drying (67.3%). Although flavor loss of dried products is often due to v olatile losses, chemical reactions such as oxidation and browning also contribute considerably. The size, shape, uniformity and absence of defects are all important in assessing product quality.
78 While the evaluation of these parameters is relatively straightforward, evaluation of color, aroma and taste are more difficult. The change of color can influence a consumers perception and may affect other attributes such as the flavor of the product. Measurement of product color implies either visual matching pai red against standard colors or expression in terms of numerical dimensions using hue, saturation and lightness. Devices such as tintometers and spectrophotometers are relatively expensive, while visual matching against standard colors is more practical. Th e Munsell color order system displays a collection of colored chips arranged with scales and thus provides a method of precisely specifying colors. Evaluation of color is particularly important in drying fruit slices as many products experience enzymatic c olor changes. Mangoes for example, undergo enzymatic browning which could be measured using some of these techniques. This information could be used to assess the need of preservatives such as citric acid or lemon juice which help in color retention. Color assessment could also help to estimate when drying of the fruit slice is complete. Samples from the rehydration test may be salted to taste, and cooked in steam or in a microwave oven. Once the best cooking time for the product is established, it should b e used for all samples of that particular product so that texture may be judged comparatively. Quality can be subsequently attributed on a specified grading scale. Quality taste parameters such as sweetness can be evaluated with the use of hydrometers, osc illating Utubes or refractometers which are calibrated to read in Brix which refers to the sugar content in a solution. Nutritional attributes affected by
79 dehydration include chemical parameters such as ash and sugar content, ascorbic acid or Vitamin C, carotene content, and acidity content. Vitamins A and C are destroyed by heat but are better preserved with sulfite treatments. However this leads to thiamin degradation while the blanching of vegetables results in some vitamin and thiamin loss. High ash content indicates dust contamination while low sugar and Vitamin C content corresponds with degraded product quality due to high temperatures (Leon et al., 2002). The nutritive values, ash content and acidity can be determined by simple chemical analysis with the use of a standard index. Dryer Parameter Evaluation T h e t o t al area o f t he so lar co l l ec t or is an impor t ant cons i derat ion in t he es t i m ation of drying effic i en cy. In m i xedmode drye r s, it sh o uld include t h e co l l ec t or area a n d t he sp read area o f t h e p r oduct receiv i n g direct so lar radiation. Apart from t he size a n d physical c harac teristics of the co l l ec t o r t h e system is largely depend ent on t h e angle of incl i nat ion with t h e su n In fact, t h e highest y i e l d is at t ained w i th t he c ol l ec t or oriented perpendic u lar t o t h e su n a nd as a general r u l e ; t h e o p t i m um angle of ti l t is equal t o t h e degree o f latitu d e o f t he site (Leon et al 2 0 0 2 ). A l ter n ative l y, Ad e goke and Bolaji ( 2 0 0 0) rec o mm e nded an incl i nat ion o f 1 0 more t h an the lo c a l geo g raphic a l l atit u de whi l e E nebe and E z ekoye (20 0 6 ) p r o p osed t h at t he opt i m um c ol l ec t or slope c ould be de t erm i ned by Equation 35 = + (3 5 ) The variable represents the latitude at which evaluation is conducted and the angle of declination, is calculated with Equation 36 = 23. 45sin [ 0 9863 ( 284 + ) ] (3 6 )
80 The variable n refers to the numbered day of the year Regardless of the method employed, a minimum angle of 15 should be maintained to assist the thermosiphon effect and to ensure adequate water runoff and air circ ulation (Buchinger and Weiss, 2002). Table 3 4 provides optimum tilt angles by latitude and season. A chimney can be designed with consideration given to a proposed buoyancy force of air within the solar dryer. Air pressur e generally increases with the establishment of greater density gradients through increased chimney heights according to Equation 3 7. The variable P refers to the pressure, g is the gravitational acceleration, h1 is the base height, h2 is the peak height and h is the difference in height between h1 and h2. Additionally, the density of the air at h1 and h2 are given as 1 2, respectively, where each is a function of temperature. = 1 1 2 2= ( 1 2) (3 7) Furthermore, consideration must be given to actual construction parameters and limitations that could restrict or prohibit the implementation of a specific dryer. For instance, the ease of construction indicates the feasibility of building a dryer based on the availability of materials, manpower and technical fabrication skills. These issues extend into the operation and maintenance as all of these conditions will be required for upkeep of the system. Available floor space should also be considered especially in hilly terrain where flat land must be located for installation. Attention should also be give n to the safety and reliability of the dr yer. The drying rate will be increased by raising the drying air temperature in two ways. First, this increases the ability of drying air to hold moisture. Secondly, the
81 heated air will heat the product, increasing its vapor pressure which will drive the moisture to the surface faster. However, there is a limit to raising the temperature of air in a dryer. The thermal sensitivity of food products limits the operation of dryers at significantly high temperatures as mentioned earlier in this report. A high drying air temperature could also result in more heat loss by conduction and radiation from both the collector and drying cabinet resulting in overall reduction in system efficiency. The humidity of the drying air is also crucial to the drying process. The ability of air to hold more moisture can be increased by either dehumidifying or heating the air as outlined earlier in this report. Recirculation of exhaust air also helps by utilizing the thermal energy of exit air as was detailed earli er Airflow is another parameter which influences the drying process by minimizing conduction and radiation losses through limitation of temperat ure rise. In natural circulation systems, airflow is primarily determined by the temperature rise in the collector. Higher flow may be used at the beginning of drying and lower flow when drying enters the fallingrate period (Leon et al., 2002). Furthermor e, drying efficiency drops at significantly high airflow, since contact time with food is lowered. Alternatively insufficient air flow can result in slow moisture removal. Performance Evaluation One important parameter regarding the evaluation of dryer performance is the drying time. However, it may in fact be challenging to monitor the drying time closely and thus stop the process when the product reaches the same final moisture content value in all the trials or drying systems. Instead, a single final moisture value can be evaluated by analyzing the drying curve for each dryer. An estimation of the drying time
82 can then be made for the individual dryers to arrive at that specific moisture content. Typically, this value can be taken as 15% (wb) for most fruits and vegetables. Furthermore, drying times may be considerably reduced by decreasing the product dimensions through slicing or with the addition of a small amount of sulfite while chemical pret reatments increase the drying rate (Leon et al., 2002). Drying system efficiency takes into account the weight of moisture evaporated from the product and the energy input to the drying system during the drying time. However, the efficiency of a dryer is o ften reported individually as collector efficiency, pickup efficiency, and drying efficiency. Collector efficiency is a common measure of collector performance generally ranging between 40 to 60% for flat plate collectors and is expressed by Equation 38 ( Ahmad et al., 1996) COLLECTOR=G Cp T I =G Cp( To Ti) I (3 8 ) The variable G is the mass flow rate of air per unit collector area, Cp is the specific heat of air evaluated at the average temperature within the solar collector, To is the temperature of air at the outlet of the absorber, Ti is the temperature at the absorber inlet, and I is the total solar energy incident upon the plane of the collector per unit time per unit area. T h e pick u p effic i en cy det erm i nes t he d ry i ng airs ability t o remove moisture a nd is expressed by Equation 39 ( Tiris et al 1995). PICK UP=ho hihas hi =W Vt ( has hi) (3 9 ) The variables ho, hi and has are the absolute humidities of air leaving the drying chamber, entering the drying chamber and entering the dryer at the point of adiabatic saturation, respectively. W is defined as the weight of water evaporated from the
83 evaluated at the average temperature of air within the dryer and V is the volumetric air flow rate. Drying efficiency indicates the overall thermal performance of the system, including the dryer efficiency as well as the collector efficiency. This is essentially a measure of how effective the use of solar radiation is to the drying system. For natural convection dryers, the expression is defined by Equation 310 (Leon et al., 2002). SYSTEM=WL IA (3 10 ) The variable L is the latent heat of vaporization of water at the exit air temperature and A is the aperture area of the solar collector. Temperatures in the range of 50 60C are recommended for drying temperaturesensitive products like fruits and vegetables. While temperatures up to 65C may be used initially, they should be lowered during drying to avoid quality degradation. Additionally, reports have shown that temperatures in excess of 55C for at least the last hour of the drying period may degrade the quality (Leon et al., 2002). It is generall y recommended that the maximum drying temperature, under no load conditions, be used for dryer performance analysis since a consistent measure is unattainable under loaded condition due to product vari ability. Financial Evaluation A financial analysis of s olar dryers generally includes the cost or upfront investment in the dryer, the operating cost including maintenance, and the payback of the dryer. Since dryers are capital intensive, their use and implementation are only feasible if the operating cost can be balanced against fuel savings for which methods have been reported by Kandpal and Kumar (2005). While the fixed investment and
84 operation costs can be determined in a straightforward manner, the payback period is the measure of time required to recoup the total investment. Among many factors, the payback period also takes into account, the comparative product yield, the rate of drying, the cost benefits of improved quality and the price of the final product. Additionally, the time for loading and unloading is an important consideration in the appropriate design of dryers because this may result in extended periods of time and labor that can significantly reduce throughput. In fact, some studies have noted that the loading and unloading of products is important in commercial dryers due to possibilities of contamination, cost of labor, and handling convenience (Leon et al., 2002). Evaluation Procedure Produc t preparat ion may invol v e p r e drying p ro c esses, su ch as washin g peel i ng, c u t t in g sl i cing, c ori ng, p itting, trimm i ng, cu t t in g o r ch opp in g ; a n d p re t r eatment met h ods, such as blanching, sulfatin g salt i ng, al ka l i ne d i pping, hea t in g co oking, f r eezing o r thawing. For instance, fresh mango fruit is often washed, peeled and cut before undergoing chemical pretreatment for drying. Pre treatment with ethylene action inhibitors and 1methylcyclopropene is known to delay softening and browning in fresh mango fruit while maintaining improved appearance and textural quality. However, more common pretreatments such as dipping in 0.5 1.0% ascorbic acid, citric acid (lemon juice), Lcysteine or N acetylcysteine solution for 3 minutes is often employed due to the effectiveness and simplicity of these preservative treatments. Product information, namely variety, kind or breed, maturity, and pretreatment, should also be noted. Notes and details on the predrying processes and pre treatment
85 methods are essential for maintaining c onsistency among the different products used in different dryers. Pre heating the fruit slices by blanching or another method can also help to reduce the energy consumed by the drying system and minimize the drying time. All such product preparation should be documented to allow for proper analysis and comparison among studies. Preparation of the dryer involves ensuring proper functioning of the dryer. Cover glazing must be thoroughly cleaned, and black coating for the solar collector should be checked and repainted if necessary. The dryer test site should be free from shadows during the testing period. Physical features of the dryer such as collector area, collector tilt, tray area and number of layers also need to be recorded. Loading density can be estimated by Equation 3 1 1 (Leon et al., 2002 ). = ( ) ( 2) (3 1 1 ) Instrumentation and measuring equipment on the dryer and at the test site generally include temperature and relative humidity sensors, pyranometers, anemometers and dataloggers among other devices. The following measurements are briefly described since they are essential in evaluating dryer performance. Common methods and devices utilized in each are mentioned when relevant and available. Global solar radiation on solar collectors is measured using pyranometers which are commercially available at a wide range of prices depending on the precision demanded. Simple, dome solar meters are more commonly implemented as they are easy to use, and once calibrated, retain their accuracy for long periods (Leon et al., 2002). Furthermore, the wind speed or airflow rate can be measured with a hot wi re anemometer.
86 Temperature and relative humidity can be measured using appropriate sensors and recorded in a datalogger as closely as possible although typically every five minutes depending on memory constraints (Leon et al., 2002). Sensors should be installed at the air inlet, exit and several internal locations of the dryer. In cases where a data logger is not available, these data may be recorded manually every 30 min or in one hour intervals by a thermometer or air probe and a hygrometer, respectiv ely (Leon et al., 2002). By measuring the temperature and relative humidity over time, the minimum, maximum and average of these values can be determined as well as the duration of air temperature above ambient. The moisture content of the products can be determined with fixed surface probes inside the dryer or by testing a sample selected at regular intervals, such as on an hourly basis as suggested by Ranganna (1986). Samples should represent the average moisture content of the whole lot, and therefore care should be taken in making this selection. Special attention must be given in sample selection particularly if trays are interchanged during drying. Additionally, the weight of fresh product loaded in the dryer must be noted as well as the weight of the dried product. This allows for determination of the moisture content of the fruit slices. The moisture content is essentially the amount of moisture in a product and is expressed as a percentage. In other words, moisture content is the ratio of the mass of water contained in a product over the total mass of the food sample. However, moisture content is often expressed in dry weight as the mass of water divided by the mass of dry matter.
87 Common methods for measuring moisture content are either by direct te chniques, such as drying, distillation or extraction; or by indirect techniques, such as spectroscopic measurements which do not actually remove water from the sample. Generally, the standard and easiest procedure employed in moisture content determination of a food sample is referred to as the gravimetric oven method. In this procedure, the initial weight of a sample is first determined. The sample is then placed in an oven where the product is periodically weighed while drying progresses. The change in weight will continue to be monitored until no further weight loss is observed. There are also a wide range of moisture meters that are available for determining the moisture content. These devices quickly determine the moisture content by rapidly evaporating moisture from the sample using infrared radiation. Once equilibrium is reached in a drying process, the weight of the sample remains constant as no moisture in the food sample is lost or gained over time. At this point, the moisture content remains const ant and is referred to as the equilibrium moisture content (EMC). The corresponding air relative humidity (RH) at which equilibrium is established is known as the equilibrium relative humidity (ERH). The degree of water availability within a food product i s known as the water activity (aw) which is actually defined as the vapor pressure of water within a product divided by the vapor pressure of pure water. Thus, water activity is essentially the measure of the state of water in foods This term is commonly used in discussing topics of food safety and quality. When a food product reaches equilibrium in respect to the atmosphere surrounding it, the water activity becomes equal to the relative humidity of the surrounding air.
88 The water activity scale extends f rom a completely dry state at 0 up to pure water at 1.0. While most microorganisms strive at water activities above 0.8, most microorganisms are unable to grow at levels below 0.6; however they can still survive at these reduced levels. If added to a suitable medium or rehydrated, these microorganism may be resuscitated and start to grow again. Thus, drying techniques rely on lowering the water activity in order to minimize microbial growth rates and other adverse chemical reactions. The desired level of w ater activity must be selected to be sufficiently low in order to ensure spoilage reactions are unsupported by excess moisture. However, the precise level of water activity must also be sufficiently high in order to maintain a soft, flexible texture for the dried fruit slice. Therefore, by lowering the water activity to adequate levels, the product quality is maintained while greatly improving the shelf life through prevention of deteriorative reactions. Water activity can be measured with a variety of methods including the use of desiccating jars which is of ten referred to as the isopiestic method. This approach to measuring water activity is frequently employed as it is rather inexpensive and straightforward to perform. In this method, a food product is brought into equilibrium with a closed atmosphere of known constant relative humidity which is established with either the use of calibrated saturated salt solution or a reference material of known moisture sorption isotherm This approach requires knowledge of the initial moisture content and observations of the moisture loss over time. In this process, the weight of the sample will change as moisture is lost until equilibrium is reached with the relative
89 humidity of the surrounding air. Once equilibrium has been reached the water activity of the product wil l be equal to the controlled constant relative humidity. Water activity can also be measured with the use of a water activity meter or hygrometer. The most commonly used instrument for these measurements is known as a chilled mirror dew point hygrometer w hich determines vapor pressure based on fundamental thermodynamic principles. These instruments are accurate, precise, fast and simple to use. Dew point evaluation of air vapor pressure works on the basic principle that air can be cooled without changing t he water content until it becomes saturated. Dew point temperature essentiall y refers to the point at which air becomes saturated. In practice, this temperature is determined by assessing the precise point at which condensation begins to form on a chilled mirror. These devices are generally composed of a mirror, optical sensor, fan and infrared thermometer. Typical dew point instruments allow an agricultural product to come into equilibrium with the sealed headspace surrounding it. As a thermoelectric cooler controls the temperature of the mirror, a thermometer measures the temperature at which condensation begins to form. This condensate is detected with the use of an optical reflectance sensor. The temperature of the product is also measured simultaneously using an infrared thermometer. Both temperatures are then used to calculate the water activity using an integrated algorithm. Measurement of water activity with these devices generally takes less than 5 to 10 minutes whereas the isopiestic techniques coul d require multiple weeks depending on the initial moisture content of the sample and calibrated solution.
90 The relationship between moisture content and water activity of an agricultural product is important This relationship is often described in referenc e to moisture sorption isotherm s. Sorption isotherms essentially represent the correlation between moisture content and water activity using a graphical plot. These graphical representations are generally expressed at a single, constant temperature over a range of water activities for a specific food material. The data points for moisture content and the corresponding water activities, generally fall on a smooth, sigmoidal curve for most agricultural products. The unique curve is considered a signature phys ical property for each specific food material at a given temperature. These tools illustrate the steady state amount of water held by the agricultural products as a function of water activity. In order to establish the critical moisture contents for agric ultural products, the sorption isotherm must be adequately developed and understood for that particular product. This also helps to predict potential changes in food stability. Using an isotherm, you can predict how conditions such as high humidity will af fect your products and determine the most stable point for your food product. The shape of a sorption isotherm also helps to predict the quality and physical characteristics of an agricultural material as a function of water activity. To obtain a sorption isotherm, it is necessary to gather at least several data points relating moisture content to water activity. Thus, each point should be d efined by the moisture content evaluated at a specific water activity This data should be collected over a range of specified water activities to provide enough information to predict the specific shape of the curve. These data points are then fitted to sorption
91 isotherm models such as the GAB and BET models which have both been reported to be suitable for high sugar content fruits (Falade & Aworh, 2004) such as mangoes. While no isotherm model reported in literature is valid over the entire water activity range of 0 to 1, the GAB model is widely used with an approximate range of 0.10 up to 0.90, while the BET isotherm generally holds from 0.05 to 0.45 (Shyam et al., 2001). This signature physical property of agricultural samples serves to specify the moisture content needed in order to reach adequate preservation standards in which microbial spoilage and deteriorative c hemical reactions are prohibited. Thus, the sorption isotherm helps determine the amount of moisture needed to be removed in the drying process. Summary of Solar Dryer Review Agricultural spoilage in developing countries is known to be significantly high particularly in moist, tropical regions. The various causes associated with product degradation have been reviewed in this report and a wide range of solutions have been described. Traditional drying processes such as openair sun drying and smoke drying a ctually result in lower quality product due to microbial and fungal infestation while the implementation of sophisticated mechanized dryers in developing countries is scarce as finances and resources are limited. For this reason, small scale solar dryers a re considered and elaborated on in this review. Solar dryers essentially harness sunlight within enclosed environments thus protecting the product and minimizing the use of traditional energy sources. A systematic classification scheme was developed in order to establish clear distinctions in the design of solar dryers. Generally, solar dryers are considered either
92 passive or active based on the utilization of forced convection. While passive systems ar e often easier to construct and relatively cheaper, they commonly exhibit lower quality product or decreased efficiency. Further distinction can be made considering the mode of solar radiation exposure. Direct dryers often result in downgraded product qual ity as direct exposure to radiation can hurt the product in a variety of ways. Indirect dryers are often slower but increase the quality substantially. Compromises must be made here to ensure adequate product quality and dryer efficiency while minimizing t he complexity and investment in the dryer. The wide selection of solar dryers reviewed here demonstrates the ongoing efforts to improve these drying systems. Modifications to solar dryers such as chimneys, ventilators, absorber improvements, thermal storage, and biomass burners, among many others are discussed in this review. While some of these modifications are found ineffective such as the wind ventilator, others show marked improvement. Some designs increase the solar absorption efficiency while others ensure heating continues during lulls in solar activity. Materials for fabrication of solar dryers have also been reviewed here and a comprehensive list was created for the primary components. Additionally, discussion was presented on commonly evaluated pa rameters of solar dryers as well as parameters that are recommended for future evaluations which aim to improve characterization of dryers and allow for adequate comparison of different systems. Analysis procedures were briefly discussed with consideration given to instrumentation, the frequency of recording measurements and other useful guidelines. This review
93 should thus serve to sufficiently guide individuals in selecting appropriate dryer designs for given conditions and subsequently evaluating the perf ormance of the system.
94 Table 3 1. Estimated postharvest l osses of f resh produce in d eveloped and developing c ountries Locations Developed a Countries Developing g Countries Range (%) Mean (%) Range (%) Mean (%) From production to retail sites 2 23 12 5 50 22 At retail, foodservice, and consumer sites 5 30 20 2 20 10 Cumulative total 7 53 32 7 70 32 Kader, A. A. (2005). Increasing food availability by reducing postharvest losses of fresh produce. International Postharvest Symposium 5 Verona, Italy. Figure 31. Drying c urve
95 Table 3 2 Moisture absorption c apability Initial relative humidity Moisture absorption capability (grams of H20 per m3 of air) [g/m3] Not heated Heated to 40C Heated to 60C 40% 4.3 9.2 16.3 60% 1.4 8.2 15.6 80% 0 7.1 14.9 (Buchinger and Weiss, 2002) Buchinger, J., & Weiss, W. (2002). Solar Drying. Austrian Development Cooperation: Institute for Sustainable Technologies Figure 32. Psychrometric chart showing a drying process. A) Overview. B) Enlarged area of interest.
96 Figure 33 Traditional o pen air sun d rying [Reprinted with permission from Li et al., 2006 (Page 840, Figure 2 ).] Figure 34 Working p rinciple of d irect solar d ryer
97 Figure 35 Working principle of i ndirect, active solar dryer
98 Figure 36 General classification of solar dryers [Reprinted with permission from Ekechukwu and Norton, 1999 (Page 620, Figure 2).]
99 Figure 37 Working p rinciple of direct type, s olar d ryer Figure 38 Typical direct type, tent solar d ryer
100 Figure 39 Typical direct type, seesaw dryer Figure 310. Typical d irect type, box d ryer A B
101 Figure 311. Typical i ndirect type c abinet d ryer [Reprinted with permission from Jairaj et al., 2009 (Page 1703, Figure 10 ).] Figure 312. Airflow p rinciples in assorted solar c ollectors
102 Figure 313. Typical i ndirect type, tunnel d ryer [Reprinted with permission from Jairaj et al., 2009 (Page 1705, Figure 18).] Figure 31 4 Typical large scale, g reenhouse solar d ryer
103 Figure 315. Roof i ntegrated solar drying s ystem [Reprinted with permission from Janjai et al., 200 8 (Page 93, Figure 1 ).] Figure 316. Drying c abinet and solar collector of in h ouse d ryer A) Operation of dryer with loaded trays. B) Roof integrated solar collectors. [Reprinted with permission from Smitabhindu et al., 2008 (Page 1 525, Figure 2 ).] A B
104 Figure 31 7 Solar dryer categorization overview
105 Figure 31 8 Schematic of d irect, box d ryer A) Front. B) Side. C) Rear. [Reprinted with permission from Sodha et al., 1985 (Page 265 Figure 2 ).] A B C
106 Figure 319. Box d ryer with H2O h eater [Reprinted with permission from Pande and Thanvi, 1991 (Page 584, Figure 1 ).]
107 Figure 320. B ox d ryer with l imited tracking. A) Pictorial. B) Cross section schematic [Reprinted with permission from Mwithiga and Kigo, 2006 (Page 248 Figure 1 ).] A B
108 Figure 321. Detailed s chematic of box d yer [Reprinted with permission from Enebe and Ezekoye, 2006 (Page 187, Figure 2 ).]
109 Figure 322. Box dryer with variable i nclination [Reprinted with permission from Singh et al., 2006 (Page 1803, Figure 3 ).]
110 Figure 323. Box dryer. A) Pictorial. B) Side view s chematic [Reprinted with permission from Singh et al., 2004 (Page 756757 Figure 1 and 2 ).] A B
111 Figure 324. Simple cabinet dryer. A) Sectional view. B) Isometric v iew schematic [Reprinted with permission from Bolaji and Olalusi, 2008 (Page 229, Figure 1 and 2).] A B
112 Figure 32 5 Cabinet dryer schematic [Reprinted with permission from Pangavhane et al., 2002 (Page 582 Figure 1 ).]
113 Figure 32 6 Cabinet d ryer with multiple solar collectors A) Cross sectional schematic. B) Pictorial v iew [Reprinted with permission from Li et al., 2006 (Page 839 840, Figure 1 and 2).] A B B C
114 Figure 32 7 Reverse absorber cabinet dryer, RACD [Reprinted with permission from Goyal and Tiwari, 1999 (Page 387, Figure 1 ).]
115 Figure 328. Cabinet dryer with top absorber A) Side schematic. B) Photograph. [Reprinted with permission from Sreekumar et al., 2008 (Page 13891390, Figure 3 and 4).] A B
116 Figure 329. Simple active cabinet d ryer schematic [Reprinted with permission from Tiris et al., 1995 (Page 206, Figure 1 ).]
117 Figure 33 0 Active cabinet dryer, sc hematic [Reprinted with permission from Mohanraj and Chandrasekar, 2008 (Page 605, Figure 1 ).]
118 Figure 33 1 Active cabinet dryer with absorber mesh, s chematic [Reprinted with permission from Mumba, 1996 (Page 616, Figure 1 ).]
119 Figure 33 2 Active cabinet d ryer with p iping [Reprinted with permission from Al Juamily et al., 2007 (Page 166, Figure 1 ).]
120 Figure 33 3 R otary c olumn cylindrical d ryer, RCCD [Reprinted with permission from Sarsilmaz et al. 2000 (Page 119 Figure 1 ).]
121 Figure 33 4 Largescale cabinet dryer with h eater a rray A) Dryer schematic. B) Heater array. [Reprinted with permission from Pawar et al., 1995 (Page 10881090, Figure 2 and 5).] A B
122 Figure 33 5 Cabinet d ryer with t hermal gravel s torage [Reprinted with permission from Adebayo and Irtwange, 2009 (Page 45, Figure 1 ).]
123 Figure 33 6 Cabinet d ryer with t hermal rock s torage. A) Mode of operation from side view. B) Schematic of dryer. [Reprinted with permission from Ayensu, 1997 (Page 123 Figure 2 and 3 ).] A B
124 Figure 337. Cabinet d ryer with t hermal PCM storage. A) Photograph of drying system. B) Schematic of drying system. [Reprinted with permission from Enibe, 1997 (Page 7172 Figure 1 and 2 ).] A B
125 Figure 338. Cabinet dryer with t hermal, silica g el storage [Reprinted with permission from Ezeike, 1986 (Page 3 Figure 1 ).]
126 Figure 339. Cabinet d ryer with i nterchangeable t hermal sand s torage. A) Schematic of drying system. B) Cross section of solar collector. [Reprinted with permission from El Sabaii et al., 2002 (Page 2254, Figure 1 ).] A B
127 Figure 34 0 Cabinet d ryer with auxiliary h eater A) Schematic. B) Photograph. [Reprinted with permission from Mohamed et al., 2008 (Page 942, Figure 1 ).] A B
128 Figure 34 1 Cabinet d ryer with a uxiliary heating channel A) Solar collector array. B) Drying system. [Reprinted with permission from Zomorodian, 2007 (Page 131 132, Figure 1 and 2).] A B
129 Figure 342. Cabinet dryer with heating elements, schematic [Reprinted with permission from Singh, 1994 (Page 21 Figure 1 ).]
130 Figure 343. Cabinet dryer with exhaust recirculation. A) Pictorial. B) Schematic view. [Reprinted with permission from Sarsavadia, 2007 (Page 25322533, Figure 1 and 2).] A B
131 Figure 344. Indirect c abinet dryer with biomass burner. A) Schematic of dryer. B) Photograph of dryer. [Reprinted with permission from Bhandari et al., 2005 (Page 2627, Figure 1 and 2).] A B
132 Figure 34 5 Direct c abinet d ryer with b iomass b urner [Reprinted with permission from Prasad, 2006 (Page 498 Figure 1 ).] C
133 Figure 34 6 Cabinet d ryer with b iomass b urner and thermal s torage A) Photograph of dryer. B) Schematic of dryer. [Reprinted with permission from Madhlopa and Ngwalo, 2007 (Page 451453, Figure 1 and 3).] A B
134 Figure 347. Forced c abinet d ryer with integrated desiccant [Reprinted with permission from Shanmugam and Natarajan, 2006 (Page 1242 Figure 1 ).]
135 Figure 348. Cabinet d ryer with integrated desiccant [Reprinted with permission from Thoruwa et al. 1996 (Page 688 Figure 1 ).]
136 Figure 349. Cabinet dryer with integrated desiccant and mirror A) Photograph of dryer B) S chematic of dryer. [Reprinted with permission from Shanmugman and Natarajan, 2007 (Page 15451546, Figure 1 and 2).] B A
137 Figure 35 0 Simple t unnel d ryer schematic [Reprinted with permission from Hossain and Bala, 2007 (Page 87, Figure 1 ).]
138 Figure 35 1 Tunnel d ryer schematic [Reprinted with permission from Datta et al., 1998 (Page 1167, Figure 1 ).]
139 Figure 352. Tunnel dryer with b iomass b urner [Reprinted with permission from Amir et al., 1991 (Page 169, Figure 1 ).]
140 Figure 35 3 Greenhouse d ryer schematics [Reprinted with permission from Koyuncu, 2006 (Page 10581059, Figure 2 and 4).]
141 Figure 35 4 Greenhouse d ryer A) Face view. B) Plant view. C) Side view. [Reprinted with permission from Condori et al., 2001 (Page 448449, Figure 1 3 ).] A B C
142 Table 3 3. Commonly used materials for solar dryers in d eveloping countries PART OPTIONS Support Barrels, Concrete, Loam, Metal, M ortar, Slate, S tone Frame Concrete, Loam, Metal S heet ( aluminum, corrugated iron/steel, Galvanized I ron, Mild S teel ) S late, Stone, Wood (Plywood) Insulation Building Insulation, Coconut Fibers, Cork, Corrugated Cardboard, Expanded Polystyrene, Flax, Hay, Leaves, Palm Fibers, Mineral Wool (Ro ckwool, G lasswool), Polyurethane, Rice Husk, Sawdust, Straw, Styrofoam, Wood ( P lywood), Wood Fiber, Wood Shavings Glazing (cover) Glass, FEP (Suntek), Plastic Sheets ( LD PE, LLD PE, PTFE, PVC, PEEVA, PMMA, Polycarbonate, Plexiglas), PET (Novolux), Polyester (Mylar), PVF (Tedlar) Power Source Diesel/Petroleum Generator, PV cells, Utility Power Trays Aluminum Frames, Bamboo Lattice, Fishing Nets, Mesh (Chicken Wire, Stainless Steel, Galvanized Steel, Nylon), Wire Netting Absorber Aluminum Sheet, Corrugated GI Sheet, Galvanized Iron/Steel, Black Polyester Fa bric Thermal Storage Gravel, P hase C hange M aterial (PCM), Rock, Sand, Silica Gel Note: Materials listed here are summarized from previously reported studies as discussed in this literature review Table 3 4. Optimum t ilt a ngles for solar c ollectors Latitude Best collector tilt i n: [degree] June Orientation Sept./March Orientation December Orientation 50 N 26.5 S 50 S 73.5 S 40 N 16.5 S 40 S 63.5 S 30 N 6.5 S 30 S 53.5 S 20 N 3.5 N 20 S 43.5 S 15 N 8.5 N 15 S 38.5 S 10 N 13.5 N 10 S 33.5 S Equator = 0 23.5 N 0 23.5 S 10 S 33.5 N 10 N 13.5 S 15 S 38.5 N 15 N 8.5 S 20 S 43.5 N 20 N 3.5 S 30 S 53.5 N 30 N 6.5 N 40 S 63.5 N 40 N 16.5 N 50 S 73.5 N 50 N 26.5 N Buchinger, J., & Weiss, W. (2002). Solar Drying. Austrian Development Cooperation: Institute for Sustainable Technologies.
143 CHAPTER 4 MATERIALS AND METHOD S Task 1: Sorption Isotherm Since the production of dried mango slices is largely dependent on the properties of the fruit, analysis of fresh mango was necessary. Preliminary experiments were conducted with ripe mangoes to determine appropriate slice thickness and to develop a thermodynamic tool known as a sorption isotherm. The sorption isotherm helps to determine the extent of drying required to obtain a stable product and allows for estimation of mois ture content if a samples water activity is known. Thus, work was undertaken in this task to gain an understanding of the moisture content and corresponding water activity of the fruit as both of these properties influence product quality, stability and s usceptibility to microbial spoilage (Bolin, 1980) Moisture content is the quantity of water in the product while water activity is essentially the measure of the state of water in the mango. As such, water activity must be lowered indirectly by decreasing the moisture content through a drying process. Accomplishing this task helps to minimize microbial growth rates and prevent adverse chemical reactions from occurring within the fruit. The development of the sorption isotherm provided an understanding of t he demand placed on the drying system to produce dried mango slices with increased shelf life and improved quality. Commercially available, dried mango slices of an unknown variety were obtained from a local market, sealed to prevent significant moisture exchange with the environment, and analyzed shortly thereafter. The moisture content of these dried slices was determined by standard oven drying practices at 122F (50C) for 96 hours. The samples were weighed several times daily and equilibrium conditio ns were considered
144 to be reached after three consecutive readings were within 1mg. Water activity was determined using an AquaLab water activity meter (Decagon Series 3) as this method provided a relatively quick assessment of the state of water. This inf ormation served as a reference and standard of the quality, moisture content and corresponding water activity that should be achieved in drying fresh mango with the solar dryer. Experiments were conducted in triplicate as recommended by Akanbi el al. (2006 ) and Falade et al. (2004). Fresh, ripe mangoes of the Tommy Atkins variety were also obtained from a local market. The fruit was manually inspected to determine the general stage of ripeness. In this tactile assessment, the firmness of each fruit was manually inspected to ensure adequate ripeness. Fruit that was found to be exceptionally soft or overly firm were excluded from these experiments as the fruit was overly ripe or under ripe, respectively. The mangoes were first prepared to afford consistentl y uniform slices in an effort to determine adequate thickness for drying. Several unique thicknesses were evaluated to not only increase the drying rate, but also to improve product manageability, preserve compositional integrity and maintain sensory quali ty parameters such as texture and visual appearance. The thickness of dried mango slices has previously been evaluated and reported over a wide range from 3mm to 15mm, although inconsistent results are seen between studies (Brett et al., 1996; Madamba et al., 2002; Pott et al., 2005). Thus, slice thickness was measured with the use of a Vernier Caliper over this same range of values to identify an appropriate thickness for the mango slices. Drying of these variable
145 thickness slices was carried out with standard oven drying practices at 122F (50C) for 96 hours. Fresh mango fruit was also washed, peeled and manually cut with sharp knives into thin, 3mm thick slices. Care was taken i n selecting fruit that was firm to the touch which indicated a sufficient sta ge and progression of ripeness while avoiding fruit that exhibited significantly soft flesh T actile assessm ents found that the softer fruits were indicative of overly ripe mango pulp which resulted in increased moisture content. In contrast, exceptionally firm fruit indicated under ripe fruit with poor flavor and decreased moisture content. Thus, t he data collected in this study was assumed to be reliable given the consistent methods of tactile assessment which provided a reasonable indication of the stage of ripeness. Furthermore, triplicate experiments contributed to the reliability of this data which was used to form the sorption isotherm. Evaluation of unripe or overly ripe fruit was avoided as the quality of these fruit was c onsidered unsuitable and unnecessary for drying. Had experiments been conducted with these outlying ripeness fruits, the sorption isotherm would certainly be affected since the moisture content changed based on the stage of ripeness. Isopiestic drying proc edures, as detailed by Shyam et al. (2001), were used for measuring water activity by storing the sliced samples in closed chambers of known relative humidity. In this method, samples are allowed to reach equilibrium with the controlled atmosphere surround ing them, which is analogous to the water activity of the sample. Triplicate samples of fruit slices were placed on trays located in five distinct desiccating jars. Each jar contained a unique saturated salt solution that was selected to provide adequate s creening of air relative humidity. The salts in each jar were dried and
146 calibrated prior to use. Equilibrium moisture contents were determined using MgCl2 (32.8%), NaBr (57.6%), NaCl (75.3%), and KBr (81.8%) Several drops of formalin were added to cotton pads which were placed i n each desiccator to prevent mold growth, particularly at high aw as recommended by Falade and Aworh (2004). The effect of ambient temperature variability on sorption isotherms is considered insignificant in most engineering work ac cording to Henderson and Perry (1976). Therefore, the desiccators were left exposed to the ambient temperature of the lab which was approximately 71 F (22 C) for the duration of the experiment, despite the limited control over small temperature fluctuations. Potential shifts in the sorption isotherm resulting from these temperature deviations were considered negligible for the purpose of designing this solar dryer. Triplicate samples were also left exposed to and evaluated at the relative humidity of the lab. The temperature and relative humidity of the lab were monitored and recorded with an Onset Temperature and Relative Humidity Data Logger (HOBO H08 003 02) on an hourly basis. The average relative humidity of the lab was found to be 51% over the course of the experiment. Each sample was weighed once daily and equilibrium conditions were considered to be reached when three consecutive measurements gave identical readings within 1mg as suggested by Mohamed et al. (2005). After reaching equilibrium condit ions, samples were removed from the jars to undergo standard oven drying procedures for 96 hours at 122F (50C). The remaining moisture was removed from the fruit through this process which allowed for the determination of equilibrium moisture content. The results of this experiment were used
147 to form a graphical plot showing the relationship between moisture content and water activity. The dry matter of analogous fresh fruit was concurrently obtained by oven drying at 122F (50C) until achieving constant weight after three consecutive measurements within 1mg.This process allowed for the determination of the initial moisture content of the fresh mangoes. These experiments were also performed in triplicate. Experimentally derived equilibrium moisture data were then fitted to both the BET and GAB sorption isotherm models that were detailed earlier in the literature review and are expressed in Equation 4 1 and Equation 4 2 respectively, as both have been report ed to be suitable for high sugar content fruits (Falade & A worh, 2004). While no isotherm model reported in literature is valid over the entire water activity range of 0 to 1, the GAB model is widely used with a n approximate range of 0.10 up to 0.90, whil e the BET isotherm generally holds from 0.05 to 0.45 (Shyam et al., 2001). The parameters of the models were calculated using a developers software package (Water Analyzer Series, Version 97.4) and the best fit model was determined using the least squares method. MEQ=awAB ( 1 aw) ( 1 + aw( B 1 ) ) ( 4 1 ) MEQ=awABC ( 1 C aw) ( 1 C aw+ BC aw) ( 4 2) In Equation 4 1 and Equation 42, A is the monolayer moisture (MO), B is a surface heat constant, and C is a constant. Thus, the BET model is essentially a special case o f the GAB model where the constant, C, is equal to a value of one. The development of this signature physical property of the mangoes served to specify the moisture content needed in order to reach the target water activity of the
148 commercially available, dried mango reference. Using this information, the sorption isotherm helped determine the amount of moisture needed to be removed in the drying process. Thus, this tool served as an important parameter in determining the adequate and appropriate sizing and design of the solar drying system. Task 2: Dryer Design The results of these experiments on mango fruit were used to aid in the design of a solar dryer based on procedures detailed in previously reported literature. Design criteria were identified and cons idered for implementation only after conducting an extensive literature review of solar dryers. The general design of this solar dryer was considered to be a natural convection solar dryer with a thermal rock bed as described earlier in the literature revi ew. Specific design parameters were determined based on a series of engineering calculations that were modified from a procedure proposed by Ampratwum (1998) for natural convection solar dryers. This adapted model takes into account a variety of factors including the physical properties of mangoes, the characteristics of previously reported solar systems in literature, Hai tian environmental conditions, and the local production capabilities of small Haitian villages. These properties served as input for the proposed model and provided a preliminary standard for which the actual design was expected to meet and/or exceed in te rms of efficiency. Specific characteristics of the solar dryer were then determined from the mathematical calculations and schematics were developed based on this information. Fabrication of the solar dryer was then carried out based on the schematics. Con sideration was given to construction materials that were inexpensive and potentially
149 available in developing communities. Of course, alternative materials are expected to replace the current design depending on material availability in specific Haitian com munities. Experiments were then conducted to evaluate the performance of the solar dryer. These assessments were performed with and without loading of produce to evaluate the operation and efficiency of the dryer. Design Features and Considerations The proposed dryer features an indirect, solar heating component and a separate drying chamber which houses five trays of product as shown in Figure 41. The justification for having two distinct components is that the product is spared from issues associated with direct exposure to solar radiation. Under such conditions, solar radiation is known to increase susceptibility to discoloration and casehardening which results in declined product quality. While there is an extensive list of solar absorbers in the l iterature, this system was designed with simple and affordable fabrication practices in mind. Since building materials and skilled labor are often limited in developing countries attempts were made to keep this design from becoming overly sophisticated. M athematical Procedure Specific design criteria were determined from engineering calculations after considering the general design of the dryer. Physical characteristics of the m ango fruit were employed in these calculations as the design needed to meet the demands placed on it by the product. Weather conditions were also taken into account since the natural convection principle employed in this solar dryer depends completely on weather for the creation of a driving force. Assumptions were also made in estim ating feasible harvesting rates of small villages.
150 The actual performance of the newly developed dryer was expected to meet and/or exceed the model estimated efficiency as the mathematical approach was considered to be rather conservative. A summary of th e parameters considered in this analytical procedure are summarized in Table 4 1. Mango Properties As detailed earlier in the report, the moisture content of mangoes was determined experimentally by gravimetric oven analysis. The initial moisture content of firm, ripe mangoes was found to range between 72.4% (wb) and 83.8% (wb). Thus, an average moisture content of 78.1% (wb) for fresh mangoes was assumed for this mathematical procedure. In contrast, the moisture content required in the final dried fruit w as 11.9% (wb) as determined from measurements taken on commercially available dried mangoes. A range of thicknesses was evaluated with thinner slices (7mm or less) exhibiting both cracking and poor manageability throughout the drying process. This degraded product quality resulted in a loss of compositional integrity as the mango flesh was torn, and in some cases, dried product was too thin for handling and recovery. Thus, a thicker slice of 10mm was accepted as a reasonable thickness in terms of maintaining adequate drying rates and improving both the quality and manageability of the dried slices. A single mango weighs approximately 1 lb (0.45 kg) and this assertion was confirmed by weighing whole, fresh mangoes. The maximum allowable temperature for high sugar content fruit was taken as 150F (65.56C) according to Sodha et al. (1987) and was substantiated by visual inspection of slices dried at several temperatures.
151 Temperatures in excess of 150F (65.6C) resulted in significant brown discoloration of th e mango flesh. Local production capabilities of Haitian villages were also considered such as the duration of the harvesting period. A reasonable estimate of the potential harvesting rate was also made. A five month harvesting season (Castaeda et al., 20 11) was accounted for with a value of 30 fruit considered to be harvested, prepared and dried in each batch. This was assumed to be a reasonable estimate for small communities based on the use of a single solar drying system. A twoday period is assumed here as high moisture content fruit generally require extended time in excess of a single day to dry (Al Juamily et al., 2007; Bhandari et al., 2008; Chen et al., 2007; Datta et al., 1998; Prasad et al., 2006). Extension of the drying time any further than t his two day assumption was avoided as a slower drying process could allow for increased growth of microorganisms. The identification of these parameters allowed for the determination of a suitable loading densit y as well as other physical specifications of the dryer which were resolved by means of mathematical and thermodynamic calculations Dryer Parameters The loading density or batch size of the dryer was assumed by the weight difference between whole and sliced mango fruit. The results of this analysis found that approximately 19 lbs (8.62 kg) of whole, ripe mango yielded only 10 lbs (4.5kg) of sliced fruit. This difference in weight can be attributed to the large seed in each mango as well as the stem and peel that accounts for over 47% of the total weight. As a result, it was estimated that about 26.3 lb (11.9 kg) of freshly sliced mangoes could be processed
152 with this system based on the assumption that 30 whole, ripe cull mangoes could be easily available from a single days harvest. It was also assu med in this model that the minimal temperatures ranging between 131140F (5560C) that are needed for drying high sugar content fruit, as recorded by Sodha et al. (1987), could be attained by the proposed dryer. While temperatures in excess of 149F (65 C) are shown to be effective in the initial stages of drying, Leon et al. (2002) recommends that this value be lowered later in drying to avoid quality degradation. For these reasons, a temperature of 140F (60C) was considered to be a reasonable and cons ervative estimate for this model. This assumption was further validated by inspection of the preserved quality of ovendried slices at this temperature as well as numerous studies which demonstrate temperature gains in solar dryers of at least 54F (30C) above ambient conditions (Amir et al., 1991; Enebe & Ezekoye, 2006; Mumba, 1996; Mwithiga & Kigo, 2006; Prasad et al., 2006; Singh et al., 2004; Sreekumar et al., 2008; Zomorodian et al., 2007). An estimate of 26% was also assumed for the collector efficie ncy which is an even more conservative estimate than the Ampratwum (1998) model. This proposed value is based on the lower limit of efficiency as reported in a review conducted by Jairaj et al. (2009). However, the actual efficiency of the collector was expected to exceed this conservative estimate. The cross sectional area of the drying chamber directly corresponds to the area of each tray. Although Buchinger and Weiss (2002) proposed a cross sectional area with 10kg of product for every square meter of tray, a more conservative spatial arrangement of onethird of this estimate was assumed for application in the
153 mathematical procedure. The height of the thermal rock storage was determined by dividing the minimal volume of 0.15 m3, as recommended by Goswa m i (1986) and detailed by Adebayo and Irtwange (2009), by the cross sectiona l area. Furthermore, a vertical distance of three inches between the trays was maintained to allow clearance between the loaded trays and to permit easy access to each tray. An overall design height was set as 6.56 ft (2 meters) as this height was expected to create enough density gradient to establish airflow within the dryer without creating a larger, unmanageable system. Of course, greater heights would increase this gradient, but this comes at the expense of additional material, labor and maintenance as sociated with a taller system. The number of trays also serves as an input to these mathematical calculations and as such, a total of five trays were proposed for this dryer in order to increase the product throughput. The vertical orientation of successiv e trays was expected to increase the pressure gradient as vertical height was extended accordingly. Establishing greater vertical height was thus given more consideration than simply enlarging the tray area to hold more product. It is assumed that negligib le pressure resistance would result from the inhibited air flow through the trays. Weather Conditions Environmental conditions such as ambient temperature, relative humidity, solar radiation, and solar inclination were also determined through climatic data or reasonable assumptions were made since these properties directly impact the sizing and design characteristics of the system (Enebe & Ezekoye, 2006) These data allowed for the determination of physical parameters of the dryer such as collector area, co llector tilt and tray area among other attributes. This information was critical as the natural -
154 convection process depends completely on thermodynamic air properties which are established by weather conditions. While data for specific rural locations withi n Haiti w ere unavailable, weather data were taken primarily from the capital, Port au Prince for the purpose of this study. The average monthly temperature of Port au Prince, Haiti was gathered from AccuWeather Statistical Weather Data for the harvesting period spanning five months from April to August (Castaeda et al., 2011). The average daily temperature was found to be 82.9F (28.3C) and the density of air was referenced at this temperature. Similarly, the average monthly relative humidity was found t o be 48%, the average monthly sunlight hours per day was determined as 8.71 hours, and the average monthly wind speed was found to be 31 The average, annual solar radiation level of Port Au Prince, Haiti was adopted from the Solar and Wind Energy Resou rce Assessment program (SWERA, Direct Normal Solar Radiation) as 6 -2/day while th e solar Saut dEau, Haiti was taken as 18490 N. Engineering and Thermodynamic Calculations The initial humidity ratio and enthalpy values were determined with p sychrometrics from ambient air temperature and the ambient relative humidity. The final enthalpy was then determined from the initial humidity ratio and the maximum allowable temperature. Equilibrium relative humidity was determined from the final moisture content using the isotherm model developed from isopiestic drying techniques of fresh mango slices. The final humidity ratio was determined from psychrometrics using the final relative humidity and enthalpy.
155 The mass of water to be evaporat ed, mw, was determined by Equation 4 3 (Ampratwum, 1998). mw=mp( Mi Mf) 100 Mf ( 4 3) The variable mp is the loading density of sliced fruit, Mi is the initial moisture content, and Mf is the final moisture content. It is important to note that mp refe rs to the mass of the sliced fruit which is significantly lower than the weight of whole fruit. This reduction in mass is expected to be a result of deseeding and peeling of the fruit. The average drying rate, mdr, was determined by Equation 4 4 mdr=mwtd (4 4 ) The variable td i s the drying time assuming a two day period of sunlight exposure. The mass airflow rate, ma, was determined from Equation 4 5 (Sodha et al., 1987). ma=mdrWf Wi ( 4 5) The variables Wf and Wi are the final and initi al humidity ratios, respectively. The volumetric airflow rate, Va, was estimated using Equation 46. Va=maair ( 4 6) The variable air is the density of air. Total useful energy, E, was calculated by Equation 4 7 (Sodha et al., 1987) E = ma( hf hi) td ( 4 7) The variable hf inal is the enthalpy of drying air while hi nitial is the enthalpy of ambient air.
156 The solar collection area, Ac, of the absorber was calculated with Eq uation 4 8 (Sodha et al., 1987). Ac=E I (4 8) The variable I is The difference in air pressure, P, across the bed was estimated by Equation 49 given by Jindal and Gunasekaran (1982). P = C ( Pi PATM) gh = 0 00308 g ( Ti Tatm) H ( 4 9) The variable g is the acceleration due from gravity, Tda is the assumed temperature of drying air, Tatm is the ambient atmospheric temperature and H is the assumed height of the dryer system. Construction Suitable materials were selected and specified for fabrication of this s olar dryer with consideration given to affordable and readily available materials specifically in rural Haitian communities. The solar dryer was comprised of two primary components; a solar absorber and a drying cabinet. The two components were designed to be detachable for easier maintenance and mobility as needed. A single, hinged door was installed on the rear of the cabinet to provide access to the trays. Absorber The framework of the absorber was constructed primarily from half inch plywood with a c orr ugated aluminum sheet serving as a solar absorbing component.. A rigid layer of rock wool, mineral slab insulates the bottom layer of the absorbing panel to effectively maintain elevated temperatures within the absorber system. An air gap of approximately one inch was left above the insulation material and was sandwiched on
157 the top by the suspended a luminum sheet. Another air gap was formed between this metal absorber and the upmost layer of polycarbonate glazing. Thus, the absorber was designed to allow fl ow on both sides of the metal panel as shown in Figure 42. The corrugated aluminum was oriented to allow airflow through the channels formed by the corrugation. The translucent glazing material made of thin wall polycarbonate allows radiation to enter the absorber system where opaque surfaces convert the radiation into low grade heat. All interior and exterior surfaces of the absorber were painted matte black in order to improve heat collection. Additionally, the suspended, corrugatedaluminum sheet was or iented parallel to the airflow in order to enhance the heat gain while taking care to minimize airflow inhibition or pressure reduction. A screen was placed over the inlet of the absorber to prevent intrusion of insects and other debris. The solar absorber was inclined at an approximate angle of 28 which is 10 more than the loca l geographical latitude of Saut dEau, Haiti as recommended by Bolaji and Olalusi (2008). This angle was suggested to allow for maximum absorption of solar radiation with a fixed or stationary absorber. This same angle was employed during evaluation of the solar dryer in Gainesville, Florida despite the differences in latitude. The integration of solar tracking and tilting mechanisms was avoided as these designs were expected to inc rease design sophistication and would consequently increase investment cost. Besides, solar tracking systems have shown negligible improvements over stationary designs particularly in tropical communities as discussed by Mwithiga and Kigo (2006).
158 Cabinet The 2x2 ft2 (0.61x0.61m2) cabinet, which was constructed primarily of plywood, houses five wire mesh trays in which mango slices are intended to be spread on. Each tray is constructed of a wooden frame encompassing a steel, expanded metal grating. Although steel is known to rust, particularly in moist environments, it was selected here to reduce investment and material costs. The trays are designed to be removable for easier cleaning and loading potential. Additionally, the trays can be rotated and exchanged with other positions to ensure uniform drying between each tray. The front end of each tray is designed with a wider frame to assist in the distribution of incoming heated air. T he chimney, which was designed to extend 26 inches above the cabinet, allow s for a greater pressure gradient to be established through the dryer as the warm air rises and exits through the top of the system. A simple shutter was built into the chimney to provide some control of internal cabinet temperatures and wind speed. The shutter is essentially a rigid, steel panel which slides into the chimney to restrict the airflow. In the closed position, elevated temperatures were expected as airflow was cut off. Additionally, a thermal rock storage component was installed in the plenum area at the base of the cabinet and was sized according to the lower volume limit of 5.3 ft3 (0.15 m3) as recommended by Goswami (1986) and detailed by Adebayo and Irtwange (2009). This storage area was designed to collect heat during sunlight exposure and subsequently dissipate heat at night, during cloud cover or in inclement weather. The expectation was to prevent rewetting of the mango fruit by providing a continuous heat source. A latch was installed to provide access to the thermal rock bed. The drying
159 cabinet, trays, chimney and thermal rock bed were all painted matte black to help maintain elevated temperatures within the dryer. Dryer Operation General Overview The operation of the newly developed solar dryer was evaluated in Gainesville, FL during the summer months of July and August. The system was installed on University of Florida grounds, adjacent to a weather station operated by the Agricultural and Biological Engineering department. The local weather conditions were expected to closely resemble those seen during the harvesting season of Haiti. To investigate the thermal performance of the dryer, preliminary tests were conducted with no load in the cabinet. Further experi ments were conducted while loaded with mango slices. Dryer operation was evaluated with respect to both the condition of the drying air and the dried mangoes. Figure 4 3 is a photograph of the solar dryer in operation, situated on the experimental site. No Load Evaluation of the solar dryer was first carried out with no product loaded in the cabinet. Triplicate experiments were conducted with the shutter half closed and with the shutter fully open to evaluate operation with and without shutter restriction. These experiments were carried out for two full sunlight days (39 hours) starting at 6:00am on the first day and ending at 9:00PM on the second day. Ambient air conditions were recorded from the adjacent weather station since the weather is known to direc tly influence the dryer performance. Solar radiation, temperature, relative humidity, wind speed and wind direction were all recorded from
160 this weather station every 15 minutes. Temperatures within the solar dryer were measured with T type, polyvinyl therm ocouples and recorded to a datalogger with an accuracy of 1.8F (1.0C) every five minutes as recommended by Leon et al. (2002). The thermocouples were installed at the air inlet, exit, thermal rock bed and at each tray level inside the dryer. Relative h umidity was recorded at the chimney exhaust every five minutes with an Onset Temperature and Relative Humidity Data Logger (HOBO H08 003 02). Wind speed was manually recorded at the chimney exhaust on an hourly basis using a Reed ThermoAnemometer (LM 8000). Load Fresh, ripe mangoes, of the Tommy Atkins variety, were obtained from a local market and were briefly cleaned in a bath containing 15ppm bleach. The fruit was manually inspected to determine the general stage of ripeness. In this tactile assessmen t, the firmness of each fruit was manually inspected to ensure adequate ripeness. Fruit that was found to be exceptionally soft or overly firm were excluded from these experiments as the fruit was overly ripe or under ripe, respectively. The mangoes were m anually cut and peeled to afford thin slices of fruit with an approximate range of 5 to 10mm thick. To prevent discoloration, 0.1% citri c acid was used as an antioxidant. This preservative helps to prevent adverse color changes resulting from enzymatic reactions. In this process, fruit slices were dipped in the citric acid solution for approximately 3 minutes. Alternatively the use of lemon juice is proposed where commercially available preservatives are unavailable. The total weight of product was then recorded prior to loading. Random, analogous samples were selected to determine the initial moisture content by standard oven drying procedures.
1 61 An average of 4.2 lbs (1.9 kg) of sliced mango w as distributed and spread evenly to each tray as shown in Figure 44. Special care was given to avoid overlapping of fruit slices in an effort to achieve uniform drying of all slices. Loading of the drying cabinet was carried out at 6:00am and experiments were conducted for two full sunlight days (39hours). Based on preliminary investigations, uneven drying was observed between product loaded on upper trays compared with product on lower trays. As was expected, the drying air became saturated with evaporated moisture from the lower trays, thus preventing adequate dryi ng at the upper trays. Localized temperatures within the drying cabinet were also observed to drop as a result of evaporative cooling. As moisture was evaporated from the mango slices, heat energy was lost to the water vapor in the local vicinity of the mango slices. For this reason, a continuous mode of operation was also evaluated in the operation of this solar dryer. In this process, the lowest tray was removed when the product was found to be adequately dry. The remaining trays were shifted downward in the cabinet and fresh product was added to the highest tray position. This allowed for continuous operation of the dryer rather than processing as a single batch. Trays of product were considered adequately dry by examining random samples which were select ed and marked for ongoing evaluation of moisture content. Four tagged fruit slices were distributed to each tray and were removed hourly as suggested by Ranganna (1986) for weight readings. The moisture content of these slices was determined from the weight measurements. Product spread on the same tray was considered adequately dry when the moisture content of the tagged samples dropped below 11.9% (wb). Alternatively, tactile assessments or visual inspections can be
162 performed on the fruits slices to determ ine the stage of drying. While manual evaluation of the product texture can easily be assessed for adequate dryness, softness and flexibility; color evaluation is slightly more involved. As the mango slices dry, some color change is observed from yellow to orange. Therefore, a standard color chart must be created or comparison must be made to similar dried product in order to assess the stage of drying based on color. Evaluation The efficiency of the solar dryer was evaluated in terms of collector efficienc y, pickup efficiency and drying efficiency as recommended by Leon et al. (2002). Collector efficiency is the measure of how effectively the energy available in the solar radiation is transferred to the flowing air within the system (Tiris et al., 1995). This parameter was determined by assuming steady state conditions using Equation 4 10 (Ahmad et al., 1996). COLLECTOR=G Cp T I =G Cp( To Ti) I ( 4 10) The variable G is the mass flow rate of air per unit collector area, Cp is the specific heat of air evaluated at the average temperature within the solar collector, To is the temperature of air at the outlet of the absorber, Ti is the temperature at the absorber inlet, and I is the total solar energy incident upon the plane of the collector per unit time per unit area. Pick up efficiency is a practical evaluation of the amount of moisture evaporated from the product. This parameter essentially measures the effectiveness of the heated air to absorb the evaporated moisture. Thus a comparison is made between the actual absorbance of moisture to the capacity of moisture absorbance by the heated air.
163 Eq uation 411 was used to evaluate the pick up efficiency according to Tiris et al. (1995). PICK UP=ho hihas hi =W Vt ( has hi) ( 4 11) The variables ho, hi and has are the absolute humidities of air leaving the drying chamber, entering the drying chamber and entering the dryer at the point of adiabatic saturation, respectively. W is defined as the weight of water evaporated from the dryer and V is the volumetric air flow rate. The system efficiency of a solar dryer is the measure of how effectively the solar energy input is used in drying the product (Leon et al., 2002).This term is defined by Eq uation 412. SYSTEM=WL IA ( 4 12) The variable L is the latent heat of vaporization of water at the exit air temperature and A is the aperture area of the solar collector. The water activity of the s olar dried mangoes was determined using an AquaLab water activity meter (Decagon Series 3) and verified with the newly developed sorption isotherm In this manner, the quality of the solar dried mango was compared with the commercially available, dried mangoes. Sensory quality parameters of the solar dried mango slices were visually evaluated in terms of color and appearance to determine if quality standards were achieved in comparison to commercially available product.
164 Figure 41. Natural convection solar dryer f eaturing a solar collector and separate drying c abinet with five trays
165 Table 4 1. Design parameter input for natural convection solar dryer SI IP Mango Properties Symbol Value Units Value Units Number of Fresh Fruit N fruit 30 30 Weight of Fresh Fruit m fruit 13.61 kg 30 lb Weight of Sliced Fruit m p 7.16 kg 15.79 lb Slice Thickness d 10 mm 0.39 in Initial Moisture Content (WB) M i 78.1 % 78.1 % Final Moisture Content (WB) M f 11.9 % 11.9 % Maximum Allowable Temp T max 65.6 C 150 F Dryer Parameters Symbol Value Units Value Units Number of Trays N trays 5 5 Vertical Distance Btw Trays h 7.63 cm 3.00 in Height (head) H 2.00 m 6.56 ft Cross sectional Area A cs 0.14 m 2 1.54 ft 2 Depth of Thermal Storage h thermal 26.77 cm 8.78 in Temp of Drying Air T da 60 C 140 F Collector Efficiency 26 % 26 % Weather/Climate Symbol Value Units Value Units Collector Angle 29 29 Ambient Air Temperature T atm 28.3 C 83 F Ambient Relative Humidity atm 48.0 % 48 % Drying Time (sunshine hours) t d 8.7 hours 8.7 hours Wind Speed v wind 3 m / s 9.8 ft / s Incident Solar Radiation I 6 kWh / m 2 day 1902 Btu / ft 2 day Air Density (@ T atm ) p air 1.17 kg / m 3 0.07 lb / ft 3
166 Figure 4 2. Schematic of solar collector, cross section Figure 43. Solar convection dryer on UF c ampus in Gainesville, FL
167 Figure 44. Trays loaded within the cabinet section of the solar convection dryer
168 CHAPTER 5 RESULTS AND DISCUSSI ON Sorption Isotherm Sorption data were fitted into both the GAB and BET isotherm models. The parameters of both models are shown in Table 5 1 The GAB model provided the best fit by exhibiting the highest correlation coefficient of 0.995. This sorption isotherm model is shown in Figure 5 1 with the associated data as determined by isopiestic drying experiments. The water activity of commercially available, driedmango slices was found to be 0.56 with an initial moisture content of 11.9% (wb) which lies considerably close to the isotherm cur ve. The average temperature in the lab in which these experiments were conducted was determined to be 71 F (22C) with an average deviation of 0.7F (0.4C). These minimal temperature fluctuations were considered negligible. While the general shape of s orption isotherms is sigmoidal for most agricultural products (De Jung et al., 1996) this isotherm exhibits a type III isotherm pattern (Labuza and Altunakar, 2008) as is common with high sugar content fruit (Ayranci and Dogantan, 1990; Falade and Aworh, 2004; Tsami et al., 1990). This shape is due to changes in constitution, dimensions and phase transformation of sugars which occur as the mango slices experience moisture loss from within the fruit (Falade and Aworh, 2004). This isotherm pattern is similar to a previously developed model ( Rangel Marrn et al., 2011) which was conducted with liquefied mango pulp at several temperatures ranging from 59F (15C) to 95F (35C). However, the current model, which was evaluated at an intermediate temperature, exhibits a small shift to the left of this previously reported model. This shift results in slightly higher values for moisture
169 content in the current model than the values based on the previously reported model. Minor inconsistencies between the current model and the previously reported model may be attributed to the different physical states of mango fruit in which the studies were conducted. Solar Dryer Design The results of the engineering calculations are shown in Table 5 2 along with references corresponding to the applied methodology as discussed earlier in this report. The absorber was designed to have a surface area of 18.4 ft2 (1.71m2) as determined from the calculations and shown in Figure 5 2 The 2x2 ft2, crosssectional cabinet which houses five w ire mesh trays is shown in Figure 5 3 Dryer Operation No Load The incident solar radiation, as recorded from the weather station, is shown in Fig ure 54 for two full days of sunlight exposure. Triplicate experiments were conducted for each procedure to pr ovide the average radiation levels as shown. The solar radiation levels were generally similar amongst separate trials except during experiments conducted with the shutter in place. During this experiment with shutter restricted airflow, the observed solar radiation was considerably lower particularly on the first day of testing. Since the operation of this drying system is completely dependent on solar energy, the reduced level of radiation seen in this trial certainly had effects on the outcome of the study as discussed further in this report. Daily solar radiation levels generally ranged from 50 up to 800 2 with an average of 319.7 2 during sunlight hours. Peak solar radiation occurred between 11:30am and 2:30pm each day with
170 values of 893.0, 831.0 and 870.5 2 for shutter restricted, un restricted and loaded experiments, respectively. Temperatures, as recorded without product loaded in the cabinet, are shown in Figure 55. As seen in both cases, with and without air flow restriction, the temperatures within the solar dryer were significantly elevated from the environmental temperature. For the case with unrestricted airflow the plenum temperature reached an optimum of 156.3F (69.0C) which is 62.5F (34.7C) in excess of environmental conditions which had a high of 93.8F (34.3C). In the experiments conducted with shutter restriction, temperatures were able to reach a high of 162.9F (72.7C) on the second day which is approximately 72.1F (40.0C) in excess of the environmental temperature of 90.8F (32.7C). However, the temperatures on the first day of this experiment were substantially lower as the solar radiation observed on the first day was drastically reduced by cloud cover. This indicates the strong correlation of solar radiation to the attainable temperatures within the solar dryer. The plenum temperatures were generally the highest as the plenum area directly follows the solar absorber. As heated air flowed through the cabinet there was some heat loss resulting in temperature reductions further in the cabinet. This is evident from the slightly decreased temperatures recorded at each tray level as well as the exhaust/outlet. The rock bed exhibited the lowest recorded temperatures; however, this thermal storage unit maintained temperatures elevated above the ambient levels until 7:00am on the second day. This indicated effective heat storage and dissipation through the night although i t generally took 6 to 7 hours for the rock bed to reach the heated temperature each morning which is a somewhat slower temperature rise than other
171 regions within the dryer. This delayed warming is a result of the slower heat adsorption into the rocks. The temperatures recorded at all five tray levels were averaged together to provide general, internal cabinet temperatures as shown in Figure 56 Contrary to what was expected, the case with unrestricted airflow exhibited the highest temperatures within the cabinet while the shutter restricted case was considerably lower particularly on the first day. The average cabinet temperatures were found to be 98.9 F (37.1 C) and 110 F (43.7 C) for the case of with and without shutter restriction during sunlight hours, respectively. Again this was due to the relatively low solar radiation observed during experiments that were conducted with restricted airflow. The temperatures for the case with a loaded solar dryer were also significantly lower from the noload experiments. This was due to the heat lost to the mango slices which corresponded to lo wer temperatures in the cabinet. Relative humidity, as recorded near the chimney exhaust, is shown in Figure 57 As was expected, the shutter trapped the moisture inside the cabinet which resulted in elevated relative humidity. In the case with no shutter to restrict the airflow, the moisture was able to escape from the cabinet resulting in lower relative humidity within the dryer. In fact, without the shutter, relative humidity was lowered to an average of 35.1% during sunlight hours. This is much lower than the average relative humidity of 46.2% resulting from shutter restricted experiments. However, both with and without the shutter, relative humidity was effectively decreased from ambient relative humidity by at least 18.8% during sunlight hours. These results indicate effective transport and exhaust of moisture from within the dryer.
172 The exhaust air velocity recorded at the chimney exhaust is shown in Figure 58 As was expected, the exhaust air velocity was slightly reduced with the shutter somewhat closed Removing the shutter allowed for increased airflow exhausting from the chimney. Average, u nrestricted air flow velocity was determined to be 1.4 mph during sunlight hours, while restricted air flow velocity was only 1.0 mph on average during sunlight hours. The exhausted air velocity in loaded experiments was found to be 1.1 mph on average during sunlight hours. Although unexpected, this preserved rate of exhaust indicates negligible airflow restriction through the loaded trays. Load Experiments The temperatures, as recorded while the cabinet was loaded with product, are shown in Figure 5 9 The plenum temperature was elevated to an optimum of 151.8F (66.6C) which is still 58.3F (32.5C) in excess of the environmental temperature which had a high of 93.5F (34.1C). The internal cabinet temperatures were somewhat lower which is expected from the heat loss to the cabinet as well as the mango slices. The average cabinet temperature during sunlight hours was determined to be 98.5 F (37.0 C). Figure 5 10 shows the plenum and rock bed temperatures for the cases of loaded cabinet and noload experiments. As both the plenum and rock bed are upstream from the cabinet, the attainable temperatures were unaffected by the presence of mango slices in the cabi net. The temperatures in these two regions directly correspond to the efficiency of the absorber which does not change based on the loading of the dryer with product.
173 Figure 1511 shows the temperatures recorded at the exhaust outlet for the instances of no loading and loaded with mango slices. The temperature in the exhaust was greatly reduced by loading product into the cabinet particularly on the first day. This reduction in tem perature is due to the heat lost to the mango slices. The larger discrepancy during Day 1 is a consequence of the increased level of relative humidity observed during the initial stages of drying. In general, exhaust temperatures were reduced by approximat ely 30 F (16.7 C) as compared with the incoming heated air delivered into the plenum area. The relative humidity, as recorded in the exhaust of the loaded cabinet, is shown in Figure 512 The moisture from the mango slices placed additional loading on the dryer resulting in increased relative humidity than what was observed in the non loaded experiments. In fact, the average relative humidity recorded in the loaded experiments was found to be 61.6% during sunlight hours. This is at least 15.4% higher than the average relative humidities recorded in nonloaded experiments. This additional moisture from the product prohibited significant removal of relative humidity from within the cabinet. It is also noted that the relative humidity on the second day of the loaded experiment was slightly reduced from the first day as a consequence of the excess moisture that was present on the product surface on Day 1. Figure 513 shows the moisture content for mango slices when the dryer was operated in the batch mode. The i nitial moisture content of the slices was approximately 84% (wb) which was slightly elevated as a result of the citric acid solution which added some moisture to the product. A total weight of 21.4 lb (9.71 kg) of sliced mango was distributed among the fiv e trays and t he product was dried over a twoday period.
174 Drying rates were somewhat slow at the initial stages of drying as solar radiation was low in the morning and the mango slices had to be heated before water began to evaporate. It would be expected that preheating of the mango slices would improve the initial drying rate of the product. The bottom tray exhibited the fastest drying rate as expected since evaporated moisture from this tray accumulated on the upper trays. Thus, the upper trays exhibited slower drying rates as they were exposed to the evaporated moisture from below. By the end of the second day, product on all trays had reached a moisture content of approximately 9.4% (wb) on average, with a range of as low as 6.4% (wb) on the bottom tr ay and up to 11.1% (wb) on the top tray. These values of moisture content are rather low even compared with the commercially available, dried mango reference which was generally 11.9% (wb). By bringing the moisture content down to such low levels, heat energy was essentially wasted as the product does not require drying to this extent. Furthermore, product quality was degraded as the mango slices lost significant amounts of moisture. The final weight of dried slices was determined to be 3.4 lbs (1.53 kg) wh ich is somewhat lower than the expected value due to partially unrecoverable product on the trays. This corresponds to a total of 18.0 lbs (8.18 kg) of water removed from the mango slices. Experiments conducted with continuous loading of the solar dryer were carried out by replacing finished product with fresh product as needed. The results for moisture content, under the continuous mode of operation, are shown in Figure 514 The initial moisture content of the slices was found to be approximately 85% (wb) after citric acid treatment. A total of 21.9 lb (9.95 kg) of sliced mango was initially distributed among the
175 five trays with an average of 4.4 lb (1.99 kg) per tray. When the product on a single tray reached approximately 12% (wb), it was removed. Fresh product was then added to sustain a continuous drying process. By noon on the second day, the first tray had reached a moisture content of 11.1% (wb) which demonstrated the improved drying rate of the bottom tray compared with the upper trays. This tray w as removed and fresh product was rotated into the cabinet. In this manner, three additional trays were able to begin drying on the second day. The moisture content of these new trays was lowered by up to 20% (wb) before sunset. The final weight of the initially loaded product was determined to be 3.5 lbs (1.61 kg) which is somewhat lower than the expected value due to partially unrecoverable product on the trays. This corresponds to a total of 18.4 lbs (8.34 kg) of water removed from the initial load of mango slices. Continuous loading allowed for drying of more product than what was accomplished with batch loading. In fact, the weekly productivity of the solar dryer was determined to be 83.2 lbs (37.7kg) of fresh mango for the continuous mode of operati on while only 74.9 lbs (34.0 kg) of fresh mango could be dried with batch mode of operation. This corresponds to a weekly production of 24.5 lbs (11.1 kg) of dried slices with continuous mode of operation while only 23.8 lbs (10.8 kg) are produced with bat ch mode of operation. In both the batch and continuous modes of operation, initial rates of drying progressed rapidly. This is a result of the rapid evaporation of excess moisture on the product surface as well as the high temperatures reached within the c abinet. Subsequent drying of the mango slices was dependent on the rate at which internal
176 water migrated to the product surface via diffusion (Chen et al., 2009). Furthermore, as solar radiation levels dropped at nightfall, the drying rate slowed where it generally remained constant through the night. Although slight increases in moisture content were observed for the bottom tray through the night, the small rise of approximately 7.4% (wb) was considered negligible. Drying rates improved once again early on the second day. Evaluation Dryer Efficiency The performance of the solar dryer was evaluated in terms of several efficiency parameters as described earlier using the thermodynamic properties shown in Table 5 3 The efficiency of the solar collector was determined to be 29.5% which is significantly improved over the assumption that was made during mathematical analysis and development of this solar dryer. This improvement was evidenced by the fact that temperatures in excess of the analytical estimated values were achieved in the actual experiments. The pick up efficiency was found to be only 10.8% which indicates a slightly diminished effectiveness of moisture absorbance by the heated air. However, the drying efficiency was found to be 33.9% which indicates a rather effective use of solar radiation in drying the product. Product Quality The average moisture content of the solar dried mangoes between batch and continuous modes of operation was found to be 10.25% (wb) using standard oven drying procedure. Thi s moisture content corresponds to a water activity of approximately 0.5 6 as determined by analysis using the sorption isotherm. This water
177 activity was verified with an AquaLab water activity meter (Decagon Series 3) as 0.55 which is lower than the water activity of the commercially available, driedmango slices. Color and appearance were also preserved in the solar dried slices as determined by visual inspection of the fruit. Figure 515 shows a photograph of the solar dried mango sl ices compared to commercially available mango slices. The solar dryer was able to dry at least 21.4 lbs (9.71kg) of fresh mango slices every two days. This drying rate is significantly increased from the estimated loading capacity of 15.8lbs (7.16kg) used in the mathematical design calculations. This corresponds to a 135% increase in processing capacity. Thus, the actual performance of the solar dryer exceeded the conservative analytical model that was used to design the system. Considering this improved performance over the mathematical design, the actual weather conditions should also be noted in relation to the design parameters. The environmental conditions observed during solar drying experiments were generally downgraded compared to the values assumed in the mathematical design calculations. While the average environmental temperature of 81.3 F (27.4 C) during sunlight hours was relatively close to the temperature input used in the mathematical calculations, the actual relative humidity was increased. The average relative humidity observed during field operations was 22% higher than the estimated value of 43% which placed a much larger strain on the dryer. Additionally, solar radiation was decreased from the estimated value by about 28.2% as the dryer was designed for the higher solar radiation levels typical of tropical regions such as Haiti. It was concluded that the conservative, mathematical calculations used in the design and development of this solar dryer were effective.
178 Figure 5 1 Sorption isotherm for mango at 71 F (22 C) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1MC (DB)awSorption Isotherm Equilibrium Moisture Data GAB Model BET Model
179 Table 5 1 Estimated parameters of sorption isotherm models of mango s lices Parameters BET Model GAB Model A 0.1448 0.4203 B 0.5732 0.1117 C 1.000 0.945 R 2 0.949 0.995 Table 5 2 Design parameter output for natural conve ction solar d ryer Parameter Symbol Value Units Value Units Initial Humidity Ratio W i 0.01 kg w/kg da 0.01 lb w/lb da Initial Enthalpy h initial 58.07 kJ /kg da 32.62 Btu /lb da Final Enthalpy h final 96.31 kJ /kg da 49.05 Btu /lb da Equilibrium RH f 51.00 % 51.00 % Final Humidity Ratio W f 0.02 kg w/kg da 0.02 lb w/lb da Mass H 2 0 to Evaporate m w 5.38 kg 11.86 lb Average Drying Rate m dr 0.31 kg w/ hr 0.68 lb w/ hr Airflow Rate m a 28.80 kg da/ hr 63.45 lb da/ hr Volumetric Airflow Rate V a 24.65 m 3 / hr 869.85 ft 3 / hr Total Useful Energy E 9595 kJ 9078 Btu Solar Collector Area A c 1.71 m 2 18.36 ft 2 Air Pressure Difference P 1.91 Pa 37.09 psi
180 Figure 5 2 Solar collector s chematic Figure 5 3 Drying cabinet with c himney, shutter and trays (s chematic )
181 Figure 5 4. Incident solar radiation recorded at weather s tation 0 100 200 300 400 500 600 700 800 900 1000 0 5 10 15 20 25 30 35 40 45I (W/m2)HoursSolar Radiation No Load (Restricted) No Load (Unrestricted) Load 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time
182 Figure 5 5 Temperatures recorded within the solar dryer A) With no shutter. B) W ith Shutter 60 80 100 120 140 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursTemperatures (No Shutter) Ambient Tray 1 Tray 2 Tray 3 Tray 4 Tray 5 Rockbed Plenum OutletA 60 80 100 120 140 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursTemperatures (Shutter) Ambient Tray 1 Tray 2 Tray 3 Tray 4 Tray 5 Rockbed Plenum OutletB 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time Time
183 Figure 5 6 Cabinet temperatures based on average t emperatures bet ween all t rays Figure 5 7 Relative humidity as recorded near chimney exhaust 60 80 100 120 140 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursCabinet Temperature Ambient No Load (Unrestricted) No Load (Restricted) Load 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 30 35 40 45RH ( %)HoursRelative Humidity at Outlet Ambient No Load (Restricted) No Load (Unrestricted) 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time Time
184 Figure 5 8 Exhaust air velocity recorded at chimney exhaust Figure 5 9 Temperatures within the solar dryer at various l ocations 0 0.5 1 1.5 2 2.5 3 3.5 0 5 10 15 20 25 30 35 40 45V (mph)HoursExhaust Air Velocity at Outlet No Load (Unrestricted) No Load (Restricted) Load 60 70 80 90 100 110 120 130 140 150 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursTemperatures (Loaded) Ambient Tray 1 Tray 2 Tray 3 Tray 4 Tray 5 Rockbed Plenum Outlet 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9: 00PM 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time Time
185 Figure 5 1 0 Temperatures of incoming air in the plenum and rock b ed Figure 5 11. Temperatures recorded at the exhaust outlet 60 80 100 120 140 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursIntake Temperatures Rockbed (Load) Plenum (Load) Rockbed (No Load) Plenum (No Load) 60 80 100 120 140 160 0 5 10 15 20 25 30 35 40 45T ( F)HoursOutlet Temperature Load No Load 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time Time
186 Figure 5 1 2 Relative humidity of ambient air as recorded at w eather station and in chimney/exhaust Figure 5 13. Moisture c ontent of m a ngo slices in solar dryer over 2 full days of sunlight with batch mode of operation 0 20 40 60 80 100 120 0 5 10 15 20 25 30 35 40 45V (mph)HoursRelative Humidity Ambient Outlet 0 10 20 30 40 50 60 70 80 90 0 5 10 15 20 25 30 35 40 45Moisture Content (wb)HoursMoisture Content (Batch Loading) TRAY 1 TRAY 2 TRAY 3 TRAY 4 TRAY 5 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time Time
187 Figure 5 14. Moisture content of mango slices in solar dryer over 2 full days of sunlight with continuous mode of operation Table 5 3 : Efficiency p arameters VARIABLE DEFINITION VALUE UNITS G Mass flow rate of air per unit collector area 0.720 kg m 2 s Cp Specific heat of air (@ T AVG ) 1.005 kJ kgK T o Absorber Outlet Temperature 40.41 T i Inlet Temperature (Average Ambient Temp) 27.40 C T AVG Average Temperature between T o and T i 33.91 C I Total solar energy incident upon plane of collector per unit time per unit area 319.73 W m 2 W Weight of water evaporated from the product 8.26 kg Density of air (@ T AVG ) 1.151 kg m 3 V Volumetric Air Flow Rate 0.026 m 3 s t Drying Time 43020 s has absolute humidity of the air entering the dryer at the point of adiabatic saturation 0.022 % h i absolute humidity of air entering the drying chamber 0.082 % L Latent heat of vaporization of water at exit air temperature 2257 kJ kg A Aperture Area of the Dryer 1.719 m 2 The averaged values in this table were taken for sunlight hours only, thus omitting nighttime 0 10 20 30 40 50 60 70 80 90 0 5 10 15 20 25 30 35 40 45Moisture Content (wb)HoursMoisture Content (Continuous Loading) TRAY 1 TRAY 2 TRAY 3 TRAY 4 TRAY 5 TRAY 6 TRAY 7 TRAY 8 12:00AM 5:00AM 10:00AM 3:00PM 8:00PM 1:00AM 6:00AM 11:00AM 4:00PM 9:00PM Time
188 Figure 5 1 5 Pho tograph of solar dried mangoes compared with commercially available m ango slices
189 CHAPTER 6 CONCLUSION An inexpensive, natural convect ion solar dryer capable of producing dried mango slices has been developed to reduce spoilage of fresh fruit in rural communities of Haiti. Experiments conducted with fresh mango fruit found the average initial moisture content to be 78.1% (wb) while commercially available dried mango was 11.9% (wb). Sorption isotherms of mango slices followed a type III isotherm, which is characteristic of high sugar content products. Of both models tested, the GAB model resulted in the best fit of experimental data with a correlation coefficient of 0.995. A conservative analytical procedure was used in the design and development of this indirect mode solar dryer. From this mathematical procedure, a 2.2 ft ( 0.67 m) high chimney was designed for exhaust of moist air with a solar absorber designed to be 18.36ft2 (1.71m2) at a 28 inclination for maximum absorption of solar radiation in Haiti. Evaluation of the solar dryer showed that under noload conditions optimum temperatures of 162.9F (72.7C) and 156.3F (69.0C) were attained for trials with and without airflow restriction respectively. The shutter demonstrated effective control of internal cabinet temperatures and relative humidity within the dryer. Tests performed with batch and continuous modes of operation, found t hat an average of 21.6 lbs (9.80 kg) of mango slices could be dried from approximately 84% (wb) down to an average of 10.25% (wb) in only 39 hours with optimum temperatures of 151.8F (66.6C). From these experiments, the collector, drying and system effic iency were determined to be 29.46%, 10.77% and 33.93% respectively. Furthermore, the thermal rock bed was found to effectively store and dissipate heat during periods of limited solar radiation. Solar dried mango slices were found to have an average water activity of 0.56
190 using the newly developed isotherm model. This indicates an effective level of preservation as compared with commercially available dried mangoes. Furthermore, color and texture were also preserved in these solar dried mango slices. Thus, this inexpensive dryer was found to be efficient and technically feasible for producing dried mango slices with the utilization of a widely available natural energy resource. Additionally, it stands to uplift socioeconomy through potential employment opportunities and income generation by producing the highest value mango product while minimizing postharvest loss associated with fungal and bacterial infestation and wastage.
191 CHAPTER 7 FUTURE DEVELOPMENTS Several suggestions for the improvement of the current solar dryer design are noted here. These potential developments are detailed and discussed in this document in order to provide a basis for continuing the improvement of the current prototype. These pr oposed modifications are intended to improve the system in terms of the drying effectiveness, efficiency and throughput. Through evaluation of the solar dryer, it was determined that an average of 21.6 lbs (9.80 kg) of mango slices could be dried i n only 39 hours. A primary method of increasing the throughput is by scale up of the current prototype. In this way, a larger system can be developed to process more fruit in the same time period. It is also expected that the drying time could be reduced by angli ng the drying trays. This would potentially allow more surface area contact with the drying air resulting in improved drying rates. Of course the drying rate can also be improved with the introduction of forced air, but this comes at the expense of a fan, a photovoltaic power source and other necessary electrical components. The solar collector can also be modified to evaluate more effective designs. For instance, mesh can be incorporated to absorb additional heat or different material compositions can be explored to achieve improved solar collection. Additionally, the use of a supplemental heat component can be investigated. One option to explore is a biomass burner which could be fueled partially with mango peels. Although this option would require additional construction material to create an organic combustion area and a system to capture the heat, it does make use of mango peels which seem to have little use in the rural communities intended for the implementation of this solar dryer.
192 Further use of the mango peel byproduct can also be investigated to determine any uses such as oil extraction. Furthermore, the use of a desiccant can be explored to determine the feasibility of removing excess moisture from the air. The solar dryer can also be modified to incorporate a direct radiation component in the top portion of the drying cabinet. This could be beneficial particularly if the solar dryer is used with different products which are able tolerate direct radiation.
193 APPENDIX A DRYING PARAMETERS Table A 1. Mango drying parameters and storage c onditions Buchinger, J., & Weiss, W. (2002). Solar Drying. Austrian Development Cooperation: Institute for Sustainable Technologies.
194 APPENDIX B STANDARD OPERATING PROCEDURES Product Preparation Care should be taken in selecting fresh mango fruit which is moderately firm to the touch indicating a sufficient stage of ripening. Significantl y soft fruit should be avoided as this is indicative of overly ripe fruit while excessively firm fruit indicates fruit that is not fully ripe. The whole m angoes should be thoroughly washed, peeled and cut into thin slices of approximately 5 to 10 mm. If any internal browning, bruising or other adverse quality characteristi cs are observed, the use of that fruit is not recommended. The mango slices should be dipped in warm water containing 0.1ppm citric acid or lemon juice for 3 minutes. Allow fresh mango slices to drain for several minutes before spreading on trays. Dryer Op eration Product should be spread evenly on the trays while ensuring no overlapping of fruit. For initial operation of the solar dryer, all trays are to be added at sunrise (7:00am) as indicated in the schedule presented further in this document. Proper operation of the solar dryer should allow the bottom tray to be completed midway (2:00pm) through the first day. The bottom tray should be removed at this point and finished product recovered from the tray. Remaining trays should be shifted downward in the cabinet, and a new tray with fresh product can be added to the top position. Subsequent drying of trays should generally occur every 3 hours during sunlight exposure; although new trays should never be loaded within 2 hours of sunset (8:00pm) as drying wi ll not progress before nightfall. Product can remain in the solar dryer during inclement weather and at night
195 time. If actual drying does not complete as scheduled, adjust the loading density with either less or more fresh product. Maintenance/Cleaning Dr ying trays should be washed and cleaned between use. All solid particulate should be removed and any remaining residues should preferably be cleaned with moist, soapy water. The trays should be allowed to dry before loading with product. This process shoul d be repeated each time a tray is removed during operation of the solar dryer The solar collector should also be cleaned each morning to remove moisture and debris.
196 APPENDIX C OPERATING SCHEDULE TIME OF DAY NOTES Sunrise (7:00am) Initial loading of the fresh mango slices Midday (2:00pm) Remove bottom tray as drying finishes Shift rema ining trays downward in cabinet Add new tray with fresh product to top position Every 3 Hours (5:00pm) Next tray removed from bottom as drying finishes Add new tray with fresh product to the top position Sunset (8:00pm) Trays/product can remain in solar dryer over night (Do Not Add Fresh Product Within 2 Hours of Sunset) Sunrise ( 7:00am) Clear any debris/moisture from solar collector Check cabinet/product to ensure no problems exist Every 3 Hours (10:00am, 1:00pm) (4:00pm, 7:00pm) Remove bottom tray as drying finishes Shift remaining trays downward in cabinet Add new tray with fresh product to top position
197 LIST OF REFERENCES Adebayo, S. & Irtwange, S. V. (2009). Development and performance of a laboratory scale passive solar grain dryer in a tropical environment. Journal of Agricultural Extension and Rural Development 1 (2), 139 152. Adegoke, C. O. & Bolaji, B. O. ( 2000 ) Performance evaluation of solar operated thermosyphon hot water system in Akure. Int ernational Journal of Engin eering Technology, 2(1) 35 40. Ahmad, A Saini, J S ., & Varma, H K. (1996). Thermohydraulic performance of packedbed solar air heaters. Energy Conversation Management 37( 2 ) 205 214. Akanbi, C H Adeyemi, R S & Ojo, A. (2006). Drying c haracteristics and sorption isotherm of tomato slices. Journal of Food Engineering, 73, 157163 Akoy, El A. O. M Ismail M A ., Ahmed, El F. A ., & Luecke, W. (2006). Design and construction of a solar dryer for mango s lices. Conference of International Research on Food Security, Natural Resource Management and Rural Development (pp 17), University of Bonn, October 11 13. Al Juamily, K E J. Khalifa, A J. N & Yassen, T A. (2007). Testing of the performance of a fruit and vegetable solar drying system in Iraq. Desalination, 209, 163 170 Amir, E. J., Djaya, C., Esper, A., Grandegger, K., & Mhlbauer, W. (1991). Development of a multi purpose solar tunnel dryer for use in humid tropics. Renewable Energy 1 167176. Ampratwum, D B. (1998). Design of solar dryer for date. Agricultural Mechanization in Asia, Africa and Latin America, 29( 3 ), 59 62. Arata, A. & Sharma, V. K. (1991). Performance evaluation of solar assisted dryers for low temperature drying applications I. plants description. Renewable Energy 1 (5), 729735 Ayensu, A. (1997). Dehydration of food crops using a solar dryer with convective heat flow. Solar En ergy 59, 121 126 Ayranci E Ayranci, G., & Dogantan Z (1990) Moisture sorption isotherms of dried apricot, fig and raisin at 20 C and 36 C. Journal of Food Science, 55(6), 15911593. Bakeka, B. J. D., & Bilgen, E. (2008). Solar Collector systems to prov ide hot air in rural applications. Renewable Energy 33, 1461 1468.
198 Bala B K Mondol M R A Biswas B K Chowdury B L D & Janjai S. (2003). Solar drying of pineapple using solar tunnel drier. Renew able Energy, 28, 183 1 90. Bhandari, R., Gewali, M. B., & Joshi, C. B. (2005). Performance evaluation of a hybrid solar biomass cabinet dryer. European Journal of Scientific Research, 4( 1 ), 2 5 33. Bolaji, B. K. & Olalusi, A. P. (2008). Performance evaluation of a mixedmode solar dryer. A ssumption University Journal of Technology 11(4), 225231 Bolin, H R. (1980). Relation of moisture to water activity in prunes and raisi ns. Journal of Food Science, 45( 5 ), 1190 1192 Brett, A ., Cox, D R Simmons, R ., & Anstee,G. (1996). Producing so lar dried fruit and vegetables for micro and small scale rural enterprise development. Handbook 3: Practical Aspect of Processing, Natural Resour ces Institute, Chatham, UK Buchinger, J., & Weiss, W. (2002). Solar Drying. Austrian Development Cooperation: Institute for Sustainable Technologies. Buteau, J. M. (2009). Mango Exports. JMB S. A.: Premium Quality Export Produce. Castaeda N P Rodriguez F ., & Lundy M (2011). Assessment of Haitian Mango Value Chain: A participatory assessment of mango chain actors in southern Haiti. Catholic Relief Services (CRS), United States Conference of Catholic Bishops. Chau, K. V., Baird, C. D., & Bagnall, L. O. (1980). Performance of a plastic suspended screen solar air heater. Journal of Agricultural Engineering Resources 25, 231 238. Chaudhri, S., Kothari, S ., & Panwar, N. L. (2009). Performance evaluation of exhaust air recirculation system of mixed mode solar dryer for drying onion flakes. International Journal of Renewable Energy Technology 1 (1), 2941 Ch en C R Sharma A & Lam H X. (2007). Experimental thermal performance studies of a forced flow solar dryer. SOLARIS, Third International Conference Ne w Delhi, India, 1, 293297. Chen, C. R., Sharma, A., & Vu Lan, N. (2009). Solar energy drying systems: A review. Renewable and Sustainable Energy Reviews 13, 11851210. Chua, K. J., & Chou, S. K. (2003). Low cost drying methods for developing countries. Trends in Food Science & Technology 14, 519 528.
199 Condori, M., Echazu, R., & Saravia, L. (2 001). Solar drying of sweet pepper and garlic using the tunnel greenhouse drier. Renewable Energy 22, 447460. Datta, G., Garg, H. P., & Kumar R. (1998). Simulation model of the thermal performance of a natural convectiontype solar tunnel dryer. Intern ational Journal of Energy Resources 22, 11651177. Dauthy, M. E. (1995). Fruit and Vegetable Processing. FAO Agricultural Services Bulletin No. 119. De Jung, G I W Van den Berg, C., & Kokelaar A J. (1996) Water vapour sorption behavior of original and defatted wheat gluten. International Journal of Food Science Technology 31 519 526. Ekechukwu, O. V. & Norton, B. (1999). Review of solar energy drying systems II: An overview of solar drying technology. Energy Conversion and Management, 40(6), 615655. El Sebaii, A A Aboul Enein, S Ramadan, M R I & El Gohary, H G. (2002). Experimental investigation of an indirect type natural convection solar dryer. Energy Conversion and Management, 43, 22512266. Enebe, O. M. & Ezekoye, B. A. (2006). Development and performance of modified integrated passive solar grain dryer. The Pacific Journal of Science and Technology 7( 2 ), 185 190 Enibe, S O. (2002). Performance of a natural circulation solar air heating system with phase change material energy storage. Renewable Energy, 27, 69 86. Ezeike, G O I. (1986). Development and performance of a triplepass solar collector and dryer sys tem. Energy in Agriculture, 5 ,1 20 Ezekwe, C I. (1981). Crop drying with solar air heaters in tropical Nigeria. ISE S Sol ar World Forum, Brighton, UK, 997 1005. Falade, K O & Aworh, O C. (2004). Adsorption isotherms of osmooven dried African star apple ( Chrysophyllum albidum ) and African mango ( Irvingia gabonensis ) slices. European Food Resource Technology, 218, 278 283 Falade, K O Olukini, I & Adegoke, G O. (2004). Adsorption isotherm and heat of sorption of osmotically pretreated and air dried pineapple slices. European Food Resource Technology 218, 540 543. Goswami, Y. (1986). Alternative energy in ag riculture CRC Press Inc orporated, Florida, 2 83 102
200 Goyal, R. K., & Tiwari, G. N. (1999). Performance of a reverse flat plate absorber cabinet dryer: a new concept. Energy Conversion Management 40, 385392. Henderson, S M ., & Perry, R L. (1976). Agricultural process engineering (3rd ed. ). Westport, Connecticut: The AVI Publishing Company, Inc. Hossain M A & Bala B K. (2007). Drying of hot chilli using solar tunnel drier. Solar Energy, 81, 85 92. Jairaj, K. S., Singh, S. P., & Srikant, K. (2009). A Review of solar dryers developed for grape drying. Solar Energy 83, 169 8 1712. Janjai, S., Srisittipokakuna, N., & Bala, B. K. (2008). Experimental and modeling performances of a roof integrated solar drying system for dryi ng herbs and spices. Energy, 33, 91 93. Jayaraman K. S., Das Gupta D. K., & Babu Rao N. (2000). Solar drying of vegetables. Developments in Drying: Food dehydration, 1, 179 186 Jindal, V K & Gunasekaran, S. (1982). Estimating air flow and drying rate due to natural convection in solar rice dryers. Renewable Energy Review 4 (2), 1 9 Kader, A. A. (2005). Increasing food availability by reducing postharvest losses of fresh produce. International Postharvest Symposium 5, Verona, Italy. Kandpal, T. C. & Kumar, A. (2005). Solar drying and CO2 emissions mitigation: potential for selected cash crops in India. Solar Energy 78 321 329 Kandpal, T. C., Kumar, A., & Purohit, P. (2006). Solar drying vs. open sun drying: A framework for financial evaluation. Solar Energy 80, 15681579. Keener H. M., Sabbah, M. A., Meyer, G. E., & Roller, W. L. (1977). Plastic film solar collectors for grain drying. Proceedings of the Solar Grain Drying Conference, UrbanaChampaign, IL, 56 77 Khedkar D M & Roy S K. (1990). Histological evidence for the reconstitutional property of dried/dehydrated raw mango slices. J ournal of Food Sci ence Tech nology, 25, 47 Koyuncu, T. (2006). An investigation on the performance improvement of greenhousetype agricultur al dryers. Renewable Energy, 31( 7 ), 10551071. Labuza, T. P. & Altunakar, L. (2008) Water activity prediction and moisture sorption isotherms, in water activity in f oods: fundamentals and a pplications Oxford, UK: Blackwell Publ ishing Ltd.
201 Leon, M A Kuman, S & Bhattacharya, S C. (2002). A comprehensive procedure for performance evaluation of solar food dryers. Renewable and Sustainable Energy Reviews, 6, 367 393. Li, Z., Zhong, H Tang, R Liu, T., Gao, W & Zhang, Y. (2006). Experimental investigation on solar drying of salted greengages. Renewable Energy, 31, 837847. Lush, T. (2010). Mangoes, Tasty and Popular, Offer Hope for Haiti. The Free Library/Associated Press August 2010. Online Service. Available from: http://www.mnn.com/your home/organic farminggardening/stories/mangoes tasty and popular offer hope for haiti. Madamba, P S & Lopez, R I. (2002). Optimization of the osmotic dehydration of mango ( Mangifera indica L. ) slices. Drying Technology 20( 6 ) 1227 1 242 Madhlopa, A ., & Ngwalo, G. (2007). Solar dryer with thermal storage and biomass b ackup heater. Solar Energy, 81, 449 462 M edlicott, A. (2001). Postharvest improvement program for the Haitian mango industry. USAID/Haiti Hillside Agricultural Program March 2001. Online Service. Available from: http://pdf.usaid.gov/pdf_docs/PNACL742.pdf. Mohamed, L. A Kouhila, M., Jamali, A Lahsasni, S & Mahrouz, M. (2005). Moisture sorption isotherms and heat of sorption of bitter orange leaves ( Citrus aurantium ). Journal of Food Engineering, 67, 491498 Mohamed, L. A Kane, C S E Kouhila, M Jamali, A Mahrouz, M & Kechaou, N. (2008). Thin layer modeling of Gelidium sesquipedale solar drying process. Energy Conversion and Management, 49, 940 9 46. Mohanraj, M ., & Chandrasekar, P. (2008). Drying of copra in a forced convection solar dri er. Biosystems Engineering, 99, 604 607. Mrema, G. C., Bren ndorfer, B., Kennedy, L., Oswin Bateman, C. O., Trim, D. S., & Wereko Brobby, C (1987). Solar Dryers the ir Role in Post Harvest Processing. Commonwealth Science Council London, UK Mumba, J. (1996). Design and development of a solar grain dryer incorporating photovoltaic powered air circulation. Energy Conservation Management 37, 615 621. Mwithiga, G., & Kigo, S. N. (2006). Performance of a solar dryer with limited sun tracking capability. Journal of Food Engineering, 74, 247 252.
202 National Academy of Sciences. (1978). Postharvest Food Losses in Developing Countries. National Academy of Sciences Washingto n D. C., 202 Othieno, H., Grainger, W., & Twidell, J. W. (1981). Application of small scale solar crop driers to maize drying in Kenya. Proceeding of the 2nd Conference on Energy for Rural and Island Communities London, UK : Pergamon Press 377386 Pande, P. C. & Thanvi, K. P. (1991). Design and development of a solar dryer cum water heater. Energy Conversion Management 31, 419424 Pangavhane, D R Sawhney, R L & Sarsavadia, P N. (2002). Design, development and performance testing of a new natural convection solar dryer. Energy, 27, 579590 Pawar, R S Takwale, M G & Bhide, V G. (1995). Solar drying of custard powder. Energy Conversion Management, 36,10851096. Pott, I. Neidhart, S M hlbauer, W & Carle, R. (2005). Quality improvem ent of nonsulphited mango slices by drying at high temperatures. Innovative Food Science Emerging Technology 6, 412419 Prasad, J Vijay, V K Tiwari, G. N & Sorayan, V P S. (2006). Study on performance evaluation of hybrid drier for turmeric ( Curcuma longa L. ) drying at village scale. Journal of Food Engineering, 75 497 502. Ranganna, S. (1986). Handbook of analysis and quality control for fruits and vegetable products. New Delhi: Central Food Technological Research Institute, Tata: McGraw Hi ll Pub lishing Co mpany Rangel Marrn, M Welti Chanes, J CrdovaQuiroz, A V Cern Bretn, J G CernBretn, R M & Anguebes Franseschi F. (2011). Estimation of moisture sorption isotherms of Mango Pulp Freezedried. International Journal of Biology and Biomedical Engineering, 5 (1),18 23. Sarsavadia, P N. (2007). Development of a solar assisted dryer and evaluation of energy requirement for the dry ing of onion. Renewable Energy 32, 25292547. Sarsilmaz, C Yildiz, C & Pehlivan, D. (2000). Drying of apricots in a rotary column cylindrical dryer (RCCD) supported with sol ar energy. Renewable Energy, 21 117127 Shanmugam, V ., & Natarajan, E. (2006). Experimental investigation of forced convection and desiccant integrated sol ar dryer. Renewable Energy, 31, 12391251.
203 Shanmugam V ., & Natarajan, E. (2007). Experimental study of regenerative desiccant integrated solar dryer with and without reflective mirror. A pplied Thermal Engineering, 27, 15431551. Sharma, V K Colangelo, A & Spagna, G. (1995). Experimental investigation of different solar dryers suitable for fruit and vegetabl e drying. Renewable Energy, 6 413424. Shove, G. C. (1977). Solar energy collectors for drying grain. Proceedings of t he Solar Grain Drying Conferenc. UrbanaChampaign, Illinois 34 38. Shyam, S S Rahman, M S ., & Labuza, T P. (2001). Measurement of Water Activity Using Isopiestic Method. Current Protocols in Food Analytical Chemistry New York: John Wiley & Sons, A2.3.1A2.3.10. Singh, K. K. (1994). Development of a small capacity dryer for vegetables. Journal of Food Engineering, 21, 1930. Singh P. P., Singh, S., & Dhaliwal, S. S. (2004). Multi shelf portable s olar dryer. Renewable Energy 29, 753765. Singh P. P., Singh, S., & Dhaliwal, S. S. (2006). Multi shelf domestic solar dryer. Energy Conversion Management, 47, 17991815. Smitabhindu, R., Janjai, S., & Chankong, V. (2008). Optimization of a solar assisted drying system for drying bananas. Renewable Energy, 33, 1523 1531 Sodha, M S Bansal, N K Kumar, A Bansal, P K ., & Malik, M A. (1987). Solar crop drying, Vol. I and II. Boca Raton, Florida: CPR Press Sodha, M. S., Dang, A., Bansal, P. K., & Sharma S. B. (1985). An analytical and experimental study of open sun drying and a cabinet type drier. Energy Conversion Management 25, 263271 Sreekumar A Manikantan P E & Vijayakumar K P. (2008). Performance of indirect solar ca binet dryer. Energy Conversion Management 49, 13881395. Thoruwa, T F N Smith, J E Grant, A D & Johnstone, C M. (1996). Developments in solar drying using forced ventilation and solar regenerated desiccant materials Renewable Energy, 9, 68668 9 Tiris, C Tiris, M ., & Dincer, I. (1995). Investigation of the thermal efficiencies of a solar dryer. Energy Conservation Management 36( 3 ), 205 212. Tsami E Marinos Kovris D & Maroulis Z B (1990) Journal of Food Science, 55, 1594.
204 Yahia, E. M. (1999). Postharvest Handling of Mango: Technical Report. Agricultural Technology Utilization and Transfer/RONCO. August 1998. Online Service. Available from: http://www.manarasoft.com/pdf/P85.pdf. Zomorodian, A Zare, D & Ghasemkhani, H. (2007). Opt imization and evaluation of semi continuous solar dryer for cereals (Rice, etc). Desalination, 209 129 135
205 BIOGRAPHICAL SKETCH Drew Schiavone was born in Decatur, Georgia but he spent most of his youth in St. Augustine, Florida where he graduated with honors from Bartram Trail High School in 2003. While completing his undergraduate c urriculum Drew became interested in resource availability in a global context. At this time, h e joined Engineers Without Borders to assist international communities with engineering applications. He earned his B.S. in a gricultural and biological e ngineering in 2008; graduating cum laude, and spent several years working before returning to the ABE Graduate Program at UF. Drew held several research positions at the University of Florida during his graduate program, including his position as a biological technician at the Interdisciplinary Center of Biotechnology Research and as a postharvest lab technician in the Horticultural Sciences Department. Drew has made an effort to improve the handling practices of fresh commodities in order to ensure higher quality products are available to a wider community. Through his academic and professional endeavors Drew hopes to minimize postharvest losses, improve process efficiency and enhance product quality with practical, economical and environmentally responsible practices. Upon completion of his M.S. program, Drew plans to pursue a Ph.D. program in agricultural engi neering. Drew has been married to Jessica M. Schiavone for almost two years. They are expecting their first daughter in January 2012.