Food Research International 44 (2011) 1174–1181 Contents lists available at ScienceDirect Food Research International j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f o o d r e s Characterisation and performance assessment of guava (Psidium guajava L.) microencapsulates obtained by spray-drying Coralia Osorio ⁎, Diana P. Forero, José G. Carriazo Departamento de Química, Universidad Nacional de Colombia, AA 14490 Bogotá, Colombia a r t i c l e i n f o Article history: Received 24 July 2010 Accepted 2 September 2010 Keywords: Guava Psidium guajava Microencapsulates Spray-drying Aroma volatiles a b s t r a c t The aqueous extract of pink-fleshed guava fruit was encapsulated by spray-drying with maltodextrin (MD), arabic gum (AG), and their mixtures. The use of AG improves the fluidity during the drying process but produces an undesirable residual taste and decreases the thermal stability in the final microencapsulated powders. Retention of some aroma-active guava volatiles in the powders was confirmed by using HS-SPME– GC–MS analyses. The sensory analyses performed by two sets of non-trained panellists (adults and children) allow to select the most promising powders (MD and AGMD-1). They were physicochemically characterised and subjected to thermal (TGA and DSC) and morphologic (SEM) analyses. The successful production of spherical microencapsulates was also confirmed. From a storage stability study at two relative humidity (RH) conditions (74% and 94%), a strong influence of this parameter in the structure stability of microencapsulates and aroma release was found. The solids developed in this study represent an innovative and natural processed product from guava fruits which can be incorporated into different food products due to their sensory properties. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction Guava is a known tropical fruit characterised by a low content of carbohydrates, fats, proteins, and high vitamin C (more than 100 mg/ 100 g fruit) and fibre content (2.8–5.5 g/100 g fruit) (Pérez Gutiérrez, Mitchell, & Vargas Solis, 2008). In addition to its nutritious properties, this fruit is very appetizing due to its sensory (flavour and colour) properties (Steinhaus, Sinuco, Polster, Osorio, & Schieberle, 2008, 2009; Mercadante, Steck, & Pfander, 1999; González, Osorio, MeléndezMartínez, González-Miret, & Heredia, 2011). In spite of the fact that there are a variety of guava processed products, such as, marmalades, jellies, juices, and soft drinks, this fruit is usually consumed in the fresh stage. However, the exportation of fresh guavas from producing countries has been restricted because this fruit is highly perishable and susceptible to tropical fruit fly attack. Nowadays the interest of the food industry in natural flavour- and colour-enriched additives has been increased significantly due to the demand of consumers for reducing the use of synthetic food additives with potential short- and long-term health risks. Although there is no direct evidence, people usually associate artificial food additive consumption with the occurrence of cancer and reproductive problems in adults, as well as, behavioural problems in children (Brannen, Davidson, Salminen, & Thorngate, 2002). In another way, fruit demand around the world has increased due to results obtained ⁎ Corresponding author. Tel.: + 57 1 3165000x14472; fax: + 57 1 3165220. E-mail address: cosorior@unal.edu.co (C. Osorio). 0963-9969/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2010.09.007 from epidemiological and nutritional studies, which have shown an apparent relationship between fruit and vegetable high-consumption with a decreased incidence of degenerative diseases (Terry, Terry, & Wolk, 2001). Additionally, most of the tropical fruits possess intense colours and flavours which make them excellent candidates as a source of new and diverse natural additives. Therefore an interesting alternative is to transfer those molecules responsible for sensory and/ or biofunctional properties to a solid phase which will be able to enhance their stability and control their release in food matrices. In this way, microencapsulation is a process in which tiny particles of an active principle or droplets are surrounded by a coating, or embedded in a homogeneous or heterogeneous matrix to give small capsules (Gharsallaoui, Roudaut, Chambin, Voilley, & Saurel, 2007). In this form, the active principle is protected from external conditions and its release is controlled. Among different methods to obtain microcapsules, spray-drying is the most used in food industry. This is a unit operation by which a liquid sample (solution, emulsion or suspension) is atomized in a hot gas current to instantaneously obtain a powder. Spray-drying encapsulation has been used in the food industry since the late 1950s to provide flavour oils with some protection against degradation/oxidation and also to convert liquids to powders. At present, this technology is well established, rather inexpensive and straightforward (Gouin, 2004). Recently, we have focused our investigations (Osorio, Franco, Castaño, González-Miret, & Heredia, 2007; Osorio et al., 2010) toward a development of new and sensory-enriched processed products from tropical fruits in order to generate promising alternatives for producers, which let to reduce the loss of the production by C. Osorio et al. / Food Research International 44 (2011) 1174–1181 postharvest handling, and give to consumers new reliable food additives. A few attempts to produce powders from guava have been published. Muralikrishna, Nanjundswamy, and Siddappa (1969) reported that the maltodextrin serves as carrier and facilitate the drying. Chopda and Barrett (2001) used clarified guava juice to make powders by freeze-drying, spray-drying and tunnel-drying; as result, the freeze-dried product exhibited superior sensory quality but the spray-dried product was more stable and economically favourable. In a previous study (Forero, Morales, & Osorio, 2010), the sensory properties of four guava extracts obtained by aqueous homogenization, osmodehydration with ethanol or sucrose, and soxhlet extraction with ethanol, were compared. It was found that aqueous extract exhibited the highest sensory similarity (colour and flavour) with the fresh fruit, and also the major vitamin C content. Thus, the objective of this work was to obtain flavour-enriched microencapsulates from guava fruits by using the spray-drying technique and evaluate their physical and chemical properties, as well as, their stability under different humidity conditions. The longterm purpose of this work is to develop natural additives with addedvalue and higher shelf-time than the fresh fruit. 2. Materials and methods 2.1. General The colour of the guava pureé, aqueous extract, and guava microcapsules was determined by using a Hunterlab ColorQuest® XE d/8° colorimeter (1.00 in. diameter light pass, nominal standardization); and CIE L*a*b* coordinates (D65, 10°) were obtained. For the solids, the colour was determined by reflectance. All of the measurements were performed in triplicate. The colour parameters, chroma (Cab * ) and hue (hab) were calculated according to the following equations (Meléndez-Martínez, Vicario & Heredia, 2003):  Cab = h

 2 i1  2 2 a + b
  hab = arctan b =a : Vitamin C content was determined in the guava pureé, aqueous extract, and guava microencapsulates following the procedure described by Gökmen, Kahraman, Demir, and Acar (2000). The complete separation of ascorbic acid (AA) and dehydroascorbic acid (DHAA) was performed in a Shimadzu Prominence HPLC equipment with an RP-18 LiChrosorb column (250 × 4 mm, 5 μm i.d.) using 0.2 M KH2PO4 (pH adjusted to 2.4 with H3PO4) as the mobile phase with a flow rate of 0.5 mL/min. The amount of vitamin C (μg/mL of juice) was determined as the sum of the AA determined in the original sample (relative to the external standard ascorbic acid (Merck) in concentrations from 10 to 50 μg/mL) and the AA after DHAA reduction using dithiothreitol (DTT, 1 mg/mL, 2 h in darkness) as reducing agent. Powder moisture content was determined gravimetrically by drying in an oven at 105 °C until constant weight. The water activity (aw) of guava purée and microencapsulates was measured in a hygrometer HygroPalm AW1, at 20 °C using 1 g of each solid. pH and solid soluble content of guava purée, aqueous extract, and guava microencapsulates were measured in a Schott CG820 pH-meter and an Abbe Atago 8682 refractometer, respectively. 2.2. Sample preparation Ripe guava fruits (Regional roja variety) were purchased at a Puente Nacional (Santander, Colombia) commercial crop. Pureé made from fully ripe fruits were characterised by their pH, solid soluble content, acidity content (expressed as grams of citric acid per 100 g of 1175 fruit), and colour parameters. The composition of the fruits was determined following the procedure published by AOAC (2006). Before processing, the mature fruits were dipped into sodium hypochlorite solution to achieve an acceptable sanitary status and then processed without previous freezing. Then, whole fruit (pulp, peel and seeds) was homogenized in a blender with distilled water in a ratio 1:2 (w/w, fruit/water), then the extract was filtered through an analytical filter paper to eliminate the suspended solids. This ratio was established because diluted extracts did not show good sensory properties, and more concentrated ones were difficult to handle during the feeding of spray-drier. The aqueous extract so-obtained was characterised by pH, solid soluble content and colour. 2.3. Microencapsulation by spray-drying The aqueous extracts of guava were separately combined in a ratio of 1/1 (w/w) with four different carrier agents. The spray-drying process was performed in a laboratory scale spray-drier LabPlant SD-06 (Huddersfield, England), with a 0.5 mm diameter nozzle and a main spray chamber of 1110× 825×600 mm. Four encapsulating agents were used; for each case, 1 L of 38 °Brix feed-mixture was prepared and kept under magnetic stirring at 20 °C until homogeneity. The mixtures were separately spray-dried with an air flow rate of 100 m3/h and a compressor air pressure of 4 bar. The feed flow rate was 485 mL/h, and the inlet and outlet air temperatures were 200±2 °C and 100±4 °C, respectively. Thus, depending on the carrier agent, four guava microencapsulates were obtained: MD (corn maltodextrin DE 19-20), AG (arabic gum), AGMD-1 (arabic gum/maltodextrin 1:5 w/w), and AGMD2 (arabic gum/maltodextrin 1:10 w/w). 2.4. Sensory and volatile analyses The acceptance of each microencapsulate based on their sensory properties was evaluated in a sensory assay by two non-trained panels, using a non-structured scale. The first set of people was 30 adults that did not have any previous experience in sensory analysis; the second panel was constituted by 35 children between 5 and 6 years old. Selection procedure was only based on their interest and willingness to participate in this study. A pair of samples (approximately 1 g each) was randomly served to the panellists in white cups coded with random numbers. The panellists were instructed to indicate their preference for one of the samples. The volatile compounds released from the headspace of guava aqueous extract and four microencapsulates were analyzed by HSSPME (Carasek & Pawliszyn, 2006). A portion of each sample, 5 mL of guava extract or 2.5 g of powder mixed with 5 mL of water was equilibrated during 1 h in a 20 mL sealed vial at 40 °C. The headspace was collected on a CAR/PDMS fibre (70 μm thickness, Supeltex) during 2 h, and then directly injected (desorption time was set at 5 min.) into an HP 5890 series II gas chromatograph equipped with an FID and operated in splitless mode. A FFAP fused silica column (J&W Scientific, 30 m × 0.32 mm i.d., 0.25 μm film thickness) was used. The column oven was programmed from 50 (after 4 min) to 250 °C at 4 °C/ min and the final temperature was held for 5 min; the injector temperature was maintained at 250 °C; carrier gas was 1.0 mL of He/ min; and make up gas was nitrogen at 30 mL/min flow rate. GC–MS analyses were performed in a gas chromatograph Shimadzu GC17A coupled to a selective mass detector QP5050, using the same conditions used in GC analyses. Linear retention indices were calculated according to the Kovats method using a mixture of normal paraffin C6–C28 as external references. Mass spectral identification was completed by comparing spectra with commercial mass spectral databases WILEY and EPA/NIH and by comparison with authentic reference standards (Barrios & Morales, 2005). 1176 C. Osorio et al. / Food Research International 44 (2011) 1174–1181 2.5. Microencapsulate characterisation The two guava microencapsulates that exhibited the highest score in sensory analyses, MD and AGMD-1, were selected for their physical, structural and thermal characterisation. 2.5.1. Bulk density and fluidity Bulk density of powders was measured by weighing 260 g of sample and placing it into a 250 mL graduated cylinder. Bulk density was calculated by dividing the mass of powder by the volume filled in the cylinder. The fluidity was determined by weighing 150 g of each solid and placing it in a plastic funnel over a vibrator (VEB Koombinat, thyv2) and registering the time spent by the powder to be deposited. 2.5.2. Particle morphology Morphology of the microcapsules was evaluated using a scanning electron microscope FEI QUANTA 200 (operating at 30 kV), coating them by gold sputtering before their examination. Particle size determination was carried out by measuring the diameter of each of 260 particles localised into a selected area (ca. 80%) of SEM images. 2.5.3. Thermal analysis Thermogravimetric measurements (TGA) were carried out with a Thermogravimetric Analyzer TGA 2050 (TA Instruments, accuracy±0.1%; resolution 0.2 μg) coupled with thermal analyzer Dupont 990. The powders were also analyzed by using a Differential Scanning Calorimeter (DSC 2910, TA Instruments) and the TA Instruments software (Universal V2.5H). The apparatus was calibrated with high-purity indium. The experiments were performed under a nitrogen flow. Samples (2 mg) were heated from 0 to 400 °C in aluminium crucibles with a linear heating rate of 10 °C/min and using an empty aluminium crucible as a reference material. 2.6. Powder storage The guava microencapsulates, MD and AGMD-1 were stored at controlled temperature (18 °C), humidity, and absence of light for 30 days. Samples of 1.5 g of each powder were spread in a thin layer in Petri dishes (2 cm diameter). The experimental design was a 22 × 2 × 15 complete factorial in a random design. The independent variables considered were the humidity (74% and 94%) and time of storage. The samples were placed on sealed desiccators containing 200 mL of saturated sodium chloride and potassium nitrate solutions to obtain constant humidity values of 74% ± 2% and 94% ± 2%, respectively (Rockland, 1960). During equilibration, the humidity was measured using a thermo-hygrometer (Extech Instruments). The dependent variable was the water activity (aw), determined by triplicate removing samples after each measurement. 2.7. Statistical analysis All measurements were done in triplicate and analyzed by variance and regression analyses and averages were compared using Tukey's test with a probability of p ≤ 0.05. 3. Results and discussion 3.1. Sensory and volatile analyses It is remarkable that colour parameters of guava purée and aqueous extract are quite similar (Table 1), and also that vitamin C remains in the aqueous extract. The flavour of the aqueous extract was intense and resembled the aroma of the fresh fruit. HS-SPME/GC–MS analyses of the guava extract were performed and the results compared with those of the aroma released by guava microencapsulates dissolved in water (Table 2). These Table 1 Composition of guava (Psidium guajava, var. Regional roja) purée and aqueous extract. Property Guava purée Aqueous extract Moisture content (% wet basis)a Aw °Brix pH Proteins (%) Lipids (%) Crude fibre (%) Pectin (%) Carbohydrates (%) Minerals (%) Ash (%) Acidity (% citric acid) Vitamin C Colour parameters L* a* b* C*ab hab 87.3 ± 1.2 0.985 ± 0.008 8.5 ± 0.1 4.25 ± 0.11 0.8 ± 0.1 0.0 6.4 ± 0.1 1.52 ± 0.01 11.8 ± 0.1 0.5 ± 0.1 0.50 ± 0.10 0.48 ± 0.01 118.7 ± 0.1b 59.41 ± 0.01 12.78 ± 0.01 21.31 ± 0.05 24.85 ± 0.03 59.05 ± 0.02 – – 3.90 ± 0.01 4.12 ± 0.02 – – – – – – – – 1036 ± 6c 49.50 ± 0.09 11.02 ± 0.01 21.39 ± 0.08 24.06 ± 0.08 62.75 ± 0.06 All data are the mean of triplicate measurements ± standard deviation, pb 0.0001; – = not determined. a g of water/100 g of sample. b mg of ascorbic acid/100 g of fruit. c μg of ascorbic acid/mL. analyses revealed that the major volatile constituents detected in the guava extract through HS-SPME were the C6-aliphatic compounds: hexanal, (Z)-3-hexenal and (E)-2-hexenal. Among identified volatile compounds, ethyl butanoate, hexanal, (Z)-3-hexenal, 4-methoxy-2,5dimethyl-3(2H)-furanone, methyl benzoate, ethyl benzoate, 2- and 3methyl butanoic acid, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, cinnamyl acetate, and cinnamyl alcohol were identified as aroma-active compounds in fresh guava (Steinhaus, Sinuco, Polster, Osorio, & Schieberle, 2008). Some of these compounds, such as hexanal, 4-methoxy-2,5-dimethyl-3 (2H)-furanone, ethyl benzoate, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, cinnamyl acetate and cinammyl alcohol were effectively transferred to the guava microencapsulates and released after dissolution in water. It could be suggested that the presence of furaneol and methoxy furaneol enhanced the sweet note in the guava microencapsulates. It is important to note that the (Z)-3-hexenal predominant in the aqueous extract, disappeared in microencapsulates by effect of the spray-drying. To this regard, Steinhaus et al. (2009) demonstrated that the concentration of this compound is very susceptible during the guava's processing and depend on tissue disruption. Noteworthy, (E)-2-hexenal and (E)-2-hexenol increased their amount after microencapsulation suggesting that the oxidation via lipoxygenase (Fisher & Scott, 1997) could be activated during the spray-drying process. Two isomers of valencene were detected only in microencapsulates suggesting that these compounds could be generated by the effect of the heating during the microencapsulation process. These sesquiterpenes are characterised as citrus-derived odorants. Interestingly, no volatiles were detected in the headspace of microencapsulates without dissolution in water, which evidenced that these compounds were encapsulated instead of adsorbed. The four powders were evaluated by two sets of non-trained consumers in order to choose those with the better sensory attributes. These results are summarized in Fig. 1, and showed that microencapsulates AGMD-1 and AGMD-2 were the most accepted by adults and MD and AGMD-1 by children. The sensory characteristic of different microencapsulates influenced the consumer perceptions of freshness and sweetness. Notably, the taste of MD was described as “too sweet” by some adult consumers, in contrast, the panel of children preferred this sweet taste. The AG powder was the less accepted in both panels because of a residual off-flavour which was described as rubber-like. Taken collectively these results, MD and AGMD-1 microencapsulates were selected as promising materials for food industry and for that reason were submitted to further characterisation. C. Osorio et al. / Food Research International 44 (2011) 1174–1181 Table 2 Volatile compounds detected by HS-SPME/GC–MS in guava microencapsulates in comparison with the guava extract. Sample compound Guava RI (FFAP) extract MDa AGa AGMD-1a AGMD-2a IDb Ethyl butanoate Hexanal 2-Pentanol (Z)-3-Hexenal (E)-2-Hexenal Hexyl acetate 3-Hydroxy-2butanone (Z)-3-Hexenyl acetate Hexanol (Z)-3-Hexenol (E)-2-Hexenol 3-Octanol 4-Methoxy-2,5dimethyl-3(2H)furanone Methyl benzoate Ethyl benzoate 2- and 3-Methyl butanoic acid Valencene (isomer I) Valencene (Isomer II) Hexanoic acid Benzyl alcohol Phenyl propyl acetate Methyl cinnamate 4-Hydroxy-2,5dimethyl-3(2H)furanone Cinnamyl acetate Cinnamyl alcohol Cinnamic acid 1026 1074 1115 1145 1220 1269 1280 + ++++ + +++ ++ + + − ++ + − ++ − − − − − − − − − − ++ − − ++ +++ − − − − − +++ − +++ A A B A B A A 1323 + ++ ++ ++ ++ A 1349 1383 1388 1409 1587 + + + − + − − ++ ++ ++ − ++ +++ − ++ − − ++ ++ ++ − − ++ ++ ++ A A B B A 1610 1640 1673 + + + − ++ − − ++ + − ++ − − ++ + A A B 1711 − ++ ++ + ++ C 1717 − ++ ++ + ++ C 1828 1844 1929 + + + + + − − + − + + − + + − A A B 1949 2026 + + + + − + − + + + A A 2145 2286 2451 + + + + + + − + + − − − − − + A A B 1177 Table 3 Physicochemical characterisation of guava microencapsulates. Sample parameter MD AGMD-1 Moisture content (%) Aw pHa Bulk density (g/cm3) Fluidity (g/min) Carbohydrates (%) Protein (%) Ash (%) Vitamin C Colour parameters 4.1 ± 0.1 0.199 ± 0.009 4.28 ± 0.00 0.23 4.12 92.2 3.3 0.3 39.8 ± 0.6b 87.57 ± 0.02 7.53 ± 0.04 1.22 ± 0.01 4.1 ± 0.1 0.218 ± 0.003 4.11 ± 0.01 0.34 6.68 91.9 3.3 0.6 20.0 ± 0.6b 88.65 ± 0.09 6.53 ± 0.02 2.42 ± 0.03 L* a* b* – = not determined. MD = maltodextrin, GMD-1 = arabic gum/maltodextrin 1:5 w/w. a 0.5 mg solid in 20 mL water. b mg ascorbic acid/100 g solid. the characteristic pink colour of the guava flesh. The powder encapsulated with maltodextrin exhibited a minor bulk density than that obtained with a mixture of maltodextrin and arabic gum. According to these low values (Tonon, Brabet, & Hubinger, 2010), the powders were quite bulky. Regarding fluidity, MD and AGMD-1 powders showed different values in the range of 4–7, which allowed to classify them as powders of easy and uniform flux (Table 3). Arabic gum Relative abundance: + b 5%, ++ 5–10%, +++ 10–30 %, ++++ 30–60, − = not detected. MD = maltodextrin, AG = arabic gum, AGMD-1 = arabic gum/maltodextrin 1:5 w/w, AGMD-2 = arabic gum/maltodextrin 1:10 w/w. a 2.5 g of solid was dissolved in 5 mL of water. b A, mass spectrum and Kovats index were in agreement with those of standards (Barrios & Morales, 2005); B, mass spectrum and Kovats index were in agreement with literature data (Kondjoyan & Berdagué, 1996); and C, tentatively identified by only the mass spectrum. 3.2. Microencapsulate characterisation Microencapsulates MD and AGMD-1 selected because of their sensory properties, exhibited a measurable content of vitamin C. The colour parameters of these powders were located in the first quadrant (+a*, +b*), corresponding to red and yellow and was associated to (a) Maltodextrin 60 % of acceptance 50 40 30 20 10 0 children MD AG adults AGMD-1 Microenca psulates (b) AGMD-2 Fig. 1. Results of sensory analyses of guava microencapsulates. Fig. 2. SEM images of carrier agents, (a) arabic gum, and (b) maltodextrin. 1178 C. Osorio et al. / Food Research International 44 (2011) 1174–1181 & Kunzek, 2009; Pereira, Carmello-Guerreiro, & Hubinger, 2009). So, it is very important that MD sample surpasses the 100 °C without loss of mass, because this is a typical temperature for sterilisation processes in food industries. On the other hand, DSC profiles for these samples are very complex and involve a first process of glass transition below 50 °C (Iijima, Nakamura, Hatakeyama, & Hatakeyama, 2000; EinhornStoll & Kunzek, 2009) that indicates a likely re-organisation of pectin polymeric-chains contained in guava (1.52 g/100 g fresh fruit), since this curve-decline could not be observed in the DSC profile of pure maltodextrin (Osorio et al., 2010). Also, a continuous and broad endothermic peak centred on 150 °C (but showing a shoulder at 210 °C) was observed in both cases (MD and AGMD-1 samples) due to the melting point of pectin contained in guava (Iijima et al., 2000; Pereira et al., 2009). A strong endothermic peak centred on 240 °C, detected for the MD sample, clearly can be assigned to the melting point of maltodextrin (Osorio et al., 2010). This strong signal could not be observed (but a shoulder is) in the AGMD-1 sample due to the blending of maltodextrin with arabic gum in the encapsulating system. In sum, the addition of arabic gum to the microencapsulating agent modifies the thermal stability of microencapsulates. The pH values were between 4.11 and 4.28 and water activity values in the range of 0.199 and 0.218, with the MD powder having the less value. These values are acceptable to assure the stability of powders and are significantly minor than those of the fruit. Scanning electron microscopy (SEM) analyses show an effective change on the morphological characteristics of encapsulating-agent particles as a consequence of the spray-drying process to encapsulate the guava extract. Fig. 2 shows very irregular particles in form and size for both arabic gum and maltodextrin. Moreover, the spray-drying process led to the formation of spherical particles with smaller size than those of encapsulating agents (Figs. 3 and 4). A wide range of particle size was observed for both MD and AGMD-1 samples, with a higher fraction of particles centred on 3 and 5 μm as revealed in particle size distributions (Figs. 3d and 4d). However, more regular spherical particles were observed on AGMD-1 than MD sample, which indicates a favourable effect of arabic gum on the microcapsule formation. In general, SEM analysis reveals a successful production of spherical microencapsulates of MD and AGMD-1 samples. Fig. 5 shows the TGA-DSC profiles of MD and AGMD-1 samples, revealing similar and complex curves for both materials. TGA curves indicate that the MD sample has a higher thermogravimetric stability than the AGMD-1 sample, which shows a higher loss of mass percentage (initially 15% between 80 and 120 °C) along the temperature changes. This initial loss of mass around 100 °C can be related with the water (hydration) loss of arabic gum (Mothé & Rao, 2000). The mass of the MD sample is very stable until 160 °C, but a pronounced loss of mass was observed after 200 °C for both MD and AGMD-1 samples. Loss of mass in these materials along this temperature range can be understood as a possible degradation or thermal decomposition of one or more components (polysaccharides) and their subsequent volatilization (Mothé & Rao, 2000; Einhorn-Stoll (a) The effect of storage of guava microencapsulates at different relative humidity (RH) conditions was evaluated for a future quality control and food application of these powders. It is known that the release rates of microencapsulated volatile compounds mainly depend on the composition of the encapsulating agent (Rosenberg, Kopelman, & Talmon, 1990) and the relative humidity (Yoshii et al., 2001), which can be considered as a critical factor. The water activity (aw) was measured during 30 days (Fig. 6). In general, the aw value (b) MD MD MD Particle size distribution: MD Number of particles (c) 3.3. Storage stability test 100 90 80 70 60 50 40 30 20 10 0 (d) 260 particles 3 5 7 8 12 13 17 20 22 25 33 42 47 Particle size (µm) Fig. 3. SEM analysis of guava microencapsulates, (a) and (b) micrographs of MD samples, (c) amplified image of a spherical particle, and (d) particle size distribution. C. Osorio et al. / Food Research International 44 (2011) 1174–1181 AGMD-1 (a) AGMD-1 (b) AGMD-1 Particle size distribution of AGMD-1 Number of particles (c) 1179 100 90 80 70 60 50 40 30 20 10 0 (d) (260 particles) 3 5 6 7 8 10 12 13 14 15 16 17 20 23 30 35 Particle size (µm) Fig. 4. SEM analysis of guava microencapsulates, (a) and (b) micrographs of AGMD-1 microencapsulates, (c) amplified image of a spherical particle, and (d) particle size distribution. (a) (b) 0,0 0,0 100 100 20 Weight (%) -2,0 Heat flow (mW) -1,5 40 -1,0 60 -1,5 40 -2,0 20 -2,5 -2,5 MD sample AGMD-1 sample 0 0 100 200 300 0 -3,0 400 0 100 200 300 -3,0 400 Temperature (°C) Temperature (°C) (c) 100 MD 80 Weight (%) Weight (%) -1,0 60 80 AGMD-1 60 40 20 0 0 100 200 300 400 Temperature (C) Fig. 5. TGA-DSC curves for the (a) MD (b) AGMD-1 guava microencapsulates, and (c) a TGA profile comparison for both samples. Heat flow (mW) -0,5 -0,5 80 1180 C. Osorio et al. / Food Research International 44 (2011) 1174–1181 tended to be constant after day 10. There were negligible differences between the two solids; however, significant differences between different RHs were evidenced. The aw value increased in the range of 0.15 to 0.50 at 74% RH and 0.25 to 0.85 at 94% RH. These results indicate that the amount of arabic gum in AGMD-1 does not modify the storage stability (at a fixed RH) of this solid in comparison with the MD powder. These findings suggest that the release of volatile compounds increases with increasing RH of the environment (Soottitantawat et al., 2004). 4. Conclusions The microencapsulation of aqueous guava extract by spray-drying with different carrier agents was performed. The retention of aroma volatiles of guava was confirmed as well as their release by the effect of water on microencapsulated powders. SEM observations verified the production of spherical microencapsulates, and thermal analyses revealed a higher thermal stability of the MD solid. Moreover, it was found that the high-content of pectin strongly influences the thermal behaviour of obtained microencapsulates, and the arabic gum addition decreased the thermal stability of solid AGMD-1. The increase of relative humidity causes a significant loss of structure and posterior volatile release. Therefore, the shelf-life of these products highly depends on humidity during storage. In this way, the microencapsulation is a useful alternative to preserve the sensory and biofunctional properties of guava in processed products. The obtained powders could be used in food industry as innovative commercial products with very good sensory properties. (a) 0,90 Water activity (aw) 0,80 0,70 0,60 0,50 0,40 0,30 MD 74% RH 0,20 MD 94% RH 0,10 0,00 0 10 20 30 Time (days) (b) 0,90 Water activty (aw) 0,80 0,70 0,60 0,50 0,40 0,30 AGMD-1 74% RH 0,20 AGMD-1 94% RH 0,10 0,00 0 10 20 30 Time (days) Fig. 6. Storage stability of guava microencapsulates (a) MD and (b) AGMD-1 at 18 °C and different relative humidity (RH) conditions. Acknowledgments Financial support from Ministerio de Agricultura y Desarrollo Rural de Colombia, ASOHOFRUCOL and Frutar S.A. is gratefully acknowledged. References AOAC (2006). Official methods of analysis of AOAC International, 18th ed. Gaithersburg, USA: Association of Official Analytical Chemists. Barrios, J. C., & Morales, A. L. (2005). Tabla de índices de Kovats. In C. Duque, & A. L. Morales (Eds.), El Aroma Frutal de Colombia (pp. 335−341). Bogotá (Colombia): Universidad Nacional de Colombia. Brannen, A. L., Davidson, P. M., Salminen, S., & Thorngate, J. H. (2002). Food additives (pp. 6−9)., 2d. edition. New York: Marcel Dekker, Inc. Carasek, E., & Pawliszyn (2006). Screening of tropical fruit volatile compounds using solid-phase microextraction (SPME) fibers and internally cooled SPME fiber. Journal of Agricultural and Food Chemistry, 54(23), 8688−8696. Chopda, C. A., & Barrett, D. M. (2001). Optimization of guava juice and powder production. Journal of Food Processing and Preservation, 25(6), 411−430. Einhorn-Stoll, U., & Kunzek, H. (2009). The influence of the storage conditions heat and humidity on conformation, state transitions and degradation behavior of dried pectins. Food Hydrocolloids, 23(3), 856−866. Fisher, C., & Scott, T. R. (1997). Food flavours biology and chemistry (pp. 27−28). Cambridge, UK: The Royal Society of Chemistry. Forero, D. P., Morales, A. L., & Osorio, C. (2010). Microencapsulación del color de la guayaba. In A. L. Morales, L. M. Melgarejo (Eds.), Desarrollo de productos funcionales promisorios a partir de la guayaba (Psidium guajava L.) para el fortalecimiento de la cadena productiva (pp 141–151). Bogotá (Colombia): Ed. Panamericana. Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Saurel, R. (2007). Applications of spray-drying in microencapsulation of food ingredients: an overview. Food Research International, 40(9), 1107−1121. Gökmen, V., Kahraman, N., Demir, N., & Acar, J. (2000). Enzimatically validated liquid chromatographic method for the determination of ascorbic and dehydroascorbic acids in fruit and vegetables. Journal of Chromatography A, 881(1–2), 309−316. González, I. A., Osorio, C., Meléndez-Martínez, A. J., González-Miret, M. L., & Heredia, F. J. (2011). Application of tristiumuls colorimetry to evaluate color changes during the ripening of Colombian guava (Psidium guajava L.) varieties with different carotenoid pattern. International Journal of Food Sciences and Nutrition, 46(4), 840−848. Gouin, S. (2004). Microencapsulation: Industrial appraisal of existing technologies and trends. Trends in Food Science & Technology, 15(7–8), 330−347. Iijima, M., Nakamura, K., Hatakeyama, T., & Hatakeyama, H. (2000). Phase transition of pectin with sorbed water. Carbohydrate Polymers, 41(1), 101−106. Kondjoyan, N., & Berdagué, J. L. (1996). A compilation of relative retention indices for the analysis of aromatic compounds. Inra de Theix, France: Laborsatorie Flaveur, Station de Recherches sur la Viande. Meléndez-Martínez, A. J., Vicario, I. M., & Heredia, F. J. (2003). Application of tristimulus colorimetrý to estimate the carotenoids content in ultrafrozen orange juices. Journal of Agricultural and Food Chemistry, 51(25), 7266−7270. Mercadante, A. Z., Steck, A., & Pfander, H. (1999). Carotenoids from guava (Psidium guajava L.): Isolation and structure elucidation. Journal of Agricultural and Food Chemistry, 47(1), 145−151. Mothé, C. G., & Rao, M. A. (2000). Thermal behavior of gum arabic in comparison with cashew gum. Thermochimica Acta, 357–358, 9−13. Muralikrishna, M., Nanjundswamy, A. M., & Siddappa, G. S. (1969). Guava powder preparation, packaging, and storage studies. Journal of Food Science and Technology, 6, 93−98. Osorio, C., Franco, M. S., Castaño, M. P., González-Miret, M. L., & Heredia, F. J. (2007). Colour and flavour changes during osmotic dehydration of fruits. Innovative Food Science Emerging Technology, 8(3), 353−359. Osorio, C., Acevedo, B., Hillebrand, S., Carriazo, J., Winterhalter, P., & Morales, A. L. (2010). Microencapsulation by spray-drying of anthocyanin pigments from corozo (Bactris guineensis) fruit. Journal of Agricultural and Food Chemistry, 58(11), 6977−6985. Pereira, L. M., Carmello-Guerreiro, S. M., & Hubinger, M. D. (2009). Microscopic features, mechanical and thermal properties of osmotically dehydrated guavas. LWT-Food Science and Technology, 42(1), 378−384. Pérez Gutiérrez, R. M., Mitchell, S., & Vargas Solis, R. (2008). Psidium guajava: a review of its traditional uses, phytochemistry and pharmacology. Journal of Ethnopharmacology, 117(1), 1−27. Rockland, L. B. (1960). Saturated salt solution for static control of relative humidity between 5 and 50 °C. Analytical Chemistry, 32(10), 1375−1376. Rosenberg, M., Kopelman, I. J., & Talmon, Y. (1990). Factors affecting retention in spraydrying microencapsulation of volatile materials. Journal of Agricultural and Food Chemistry, 38(6), 1288−1294. Soottitantawat, A., Yoshii, H., Furuta, T., Ohgawara, M., Forssell, P., Partanen, R., Poutanen, K., & Linko, P. (2004). Effect of water activity on the release characteristics and oxidative stability of D-limonene encapsulated by spray-drying. Journal of Agricultural and Food Chemistry, 52(5), 1269−1276. Steinhaus, M., Sinuco, D., Polster, C., Osorio, C., & Schieberle, P. (2008). Characterization of the aroma-active compounds in pink guava (Psidium guajava L.) by application of the aroma extract dilution analysis. Journal of Agricultural and Food Chemistry, 56(11), 4120−4127. C. Osorio et al. / Food Research International 44 (2011) 1174–1181 Steinhaus, M., Sinuco, D., Polster, C., Osorio, C., & Schieberle, P. (2009). Characterization of the key aroma compounds in pink guava (Psidium guajava L.) by means of aroma re-engineering experiments and omission tests. Journal of Agricultural and Food Chemistry, 57(7), 2882−2888. Terry, P., Terry, J. B., & Wolk, A. (2001). Fruit and vegetable consumption in the prevention of cancer: an update. Journal of Internal Medicine, 250(4), 280−290. Tonon, R. V., Brabet, C., & Hubinger, M. D. (2010). Anthocyanin stability and antioxidant activity of spray-dried acai (Euterpe oleraceae Mart.) juice produced with different carrier agents. Food Research International, 43(3), 907−914. 1181 Yoshii, H., Soottitantawat, A., Liu, X. -D., Atarashi, T., Furuta, T., Aishima, S., Ohgawara, M., & Linko, P. (2001). Flavor release from spray-dried maltodextrin/gum arabic or soy matrices as a function of storage relative humidity. Innovative Food Science & Emerging Technologies, 2(1), 55−61.