Plant Foods Hum Nutr (2009) 64:6267 DOI 10.

1007/s11130-008-0103-y

ORIGINAL PAPER

Effect of the Temperature on the Spray Drying of Roselle Extracts (Hibiscus sabdariffa L.)
Salvador Gonzalez-Palomares & Mirna Estarrn-Espinosa & Juan Florencio Gmez-Leyva & Isaac Andrade-Gonzlez

Published online: 10 December 2008 # Springer Science + Business Media, LLC 2008

Abstract The effect of the drying temperature on the volatile components and sensory acceptance of the Roselle (Hibiscus sabdariffa) extract in powder was investigated. The Roselle extraction was carried out by maceration with 7 L of 30% ethanol (v/v), 560 g of fresh Roselle calyces for 168 h. The Roselle extracts were spray dried at different temperatures 150, 160, 170, 180, 190, 200 and 210 C, giving different outlet values about yield and final moisture. The volatile compounds in Roselle extract and dried samples were performed using needles of solid phase microextraction (SPME) and gas chromatography/mass spectrometry (GC/MS HP-5890). Twenty volatile compounds were identified in the extracts among them terpenoids, esters, hydrocarbons and aldehydes. Fourteen volatile compounds were identified in the powder sample, but only ten were present in the Roselle extract. This indicates that some compounds were lost and some others were generated due to a degradation process. An acceptS. Gonzalez-Palomares : J. F. Gmez-Leyva : I. Andrade-Gonzlez (*) Graduate and Research Studies Department, Instituto Tecnolgico de Tlajomulco, Km. 10 Carr. San Miguel Cuyutln, CP 45640 Tlajomulco de Ziga, Jal., Mxico e-mail: isaacag2001@yahoo.com.mx S. Gonzalez-Palomares e-mail: chava1142@yahoo.com.mx J. F. Gmez-Leyva e-mail: jfgleyva@hotmail.com M. Estarrn-Espinosa Centro de Investigacin y Asistencia en Tecnologa y Diseo del Estado de Jalisco, A.C. Av., Normalistas No. 800. Colinas de la Normal, CP 44270 Guadalajara, Jal., Mxico e-mail: mestarron@ciatej.net.mx

ability sensory analysis showed that the best powder sample was the Roselle extract dehydrated using temperature between 190 C and 200 C (p <0.05). There was not statistically significant difference in the pH of Roselle extracts ranging from 3.4 to 3.9. It was concluded that the spray drying temperature of the Roselle extracts has an effect on the volatile compounds losses. Keywords Roselle powder . Sensory analysis . Spray dryer . Volatile components

Introduction Roselle calyx (Hibiscus sabdariffa) is the plant part of greatest interest because the calyx is utilized in the processing of fruit preserves, jellies and jams for its rich content in pectin, ascorbic acid and anthocyanin color [1 3]. Today, the Roselle calyces are utilized as a good source of natural food colorants by their high natural pigment content [46]. In addition, Hibiscus sabdariffa anthocyanins have been reported to possess antioxidative, antitumor and anticarcinogenic activity [79]. Nevertheless, Hibiscus sabdariffa anthocyanins are unstable during the heat treatments, therefore studies on drying process conditions are requested in order to achieve the best stability of the final product [10]. The obtaining of Roselle powder through spray drying is an important alternative method for the use of the calyces [1113]. The dehydrated products then can be added easily to other foods to improve their storage, transportation and shelf-life. This classifies the spray drying method as the best alternative to obtain colorants and natural powder flavoring [1416]. Many studies on Roselle (Hibiscus sabdariffa) powder have been focused on the factors that are influencing the reconstitution characteristics

Plant Foods Hum Nutr (2009) 64:6267

63

or the effects of some additives and processes on characteristics of agglomeration and granulation, in addition of anthocyanin concentrates and the processing conditions, and physicochemical characteristics of the powder produced [12]. Although the spray drying is a fast process, changes on the spray drying temperature can affect the concentration of volatile compounds which would generate the loss of some sensorial characteristics [1720]. In spite of the knowledge that the volatile compounds show low stability at the spray drying temperatures, only scare studies have been reported on this issue for volatiles in Roselle [6]. To diminish some degradation effects of compounds and lost of volatile products is very common to use encapsulation agents [21, 22]. For this reason, the extraction of volatile substances by micro extraction in a solid phase (SPME), the quantitative and qualitative analysis by gas chromatography and mass spectrometry (GC-MS) as well as sensorial assays are useful tools in the evaluation of the effect of certain process variables about the quality of the dehydrated products [2326]. The objective of this work was to evaluate the effect of the spray drying temperatures on the Roselle extracts (Hibiscus sabdariffa L.), the retention of the volatile compounds and the sensorial acceptance of the reconstituted spray dried samples.

for the spray drying process in all the experiments. The outlet temperature was fixed in 80 C and the atomization was kept at 37,788g. A flexible plastic tube was inserted inside one container and connected to a variable flow peristaltic pump. The flexible outlet tube from the pump was directly connected to the liquid fed inlet of the atomizer. Seven different spray drying temperatures were used; 150, 160, 170, 180, 190, 200 and 210 C. The Roselle powder samples obtained by spray drying were weighed and packed in their respective amber-coloured glass flasks of 100 mL. These were stored in containers with silica for future analysis. Determination of Moisture in the Spray Dried Samples of Roselle In the samples of spray dried Roselle the percentage of moisture was measured by the method of the vacuum stove according to the AOAC-934.06 [27]. Six grams of each sample were placed in an aluminum dish and introduced in a vacuum stove with a temperature of 40 C until the weight remained constant. Reconstitution of Roselle Powder The reconstituted samples were carried out by placing a portion of Roselle powder in 300 mL of distilled water. A magnetic stirrer was used for mixture homogenization. The standardization of the mixture was carried out in all the samples until reaching the same concentration of soluble solids of the liquid extract (22%) [12]. The reconstituted samples were stored in amber-colored glass flasks and refrigerated until their analysis. Volatile Analysis of Roselle Extracts by SPME-GC-MS The extraction of volatile compounds from the matrix of the Roselle extracts was carried out by solid-phase microextraction (SPME). Twenty milliliters of Roselle extract were placed in a 40 mL amber headspace vial. The vial was fitted with a PTFE-faced silicone septum and put in a thermostatted bath. A fiber coated with 50/30 mm of divinylbenzene/carboxen on polydimethylsiloxane (DVB/ Carboxen/PDMS, Supelco, Bellefonte, PA) exposed for 30 min at 60 C in the headspace of the vial was then immediately inserted into the injection port of the gas chromatograph for 5 min at 240 C [28]. A HewlettPackard 5890 Series II gas chromatograph with flame ionization detector (FID) coupled to 5972 MSD quadrupole mass spectrometer was used. The HP5890 chromatograph was equipped with HP-1 (50 m0.2 mm ID0.33 m film thickness) a capillary column. Operating conditions for GC/ MS were: the oven temperature was programmed at 40 C

Materials and Methods Roselle Extracts Samples Fresh Roselle calyces (with 79% humidity) were collected from an experimental field in Michoacan, Mexico during October and November 2006. The fresh calyces were washed for quick dipping over water distilled. After the calyces were crushed manually and deposited in a closed container with 7 L of 30% methanol (v/v), all under constant conditions of static maceration. The content was left in maceration for 168 h with an occasional shaking to increase the extraction capacity at room temperature. After, the macerated extracts were filtered through a 150 m mesh. A Bchi rotavapor was used to concentrate the samples and to separate all the ethanol at 40 C. The characteristics of the roselle extracts were 22% of soluble solids and pH of 3.4. The roselle extracts were kept under refrigeration at 4 C. An ATAGO refractometer was used to determine the percentage of soluble solids. An ORION potentiometer was used to measure the pH. Spray Drying A NIRO spray dryer (with the capacity to evaporate 40 kg of water per hour) with a wheel rotating atomizer was used

64

Plant Foods Hum Nutr (2009) 64:6267

for 5 min then ramped at 5 C/min to 240 C, helium with a flow rate of 0.8 mL/min, injector port and detector temperatures were 220 and 260 C. All the samples were run at least in triplicates. Volatile compounds were identified by comparison of their retention indices tentatively only by mass spectra library Wiley 275L. Quantitation was based in area percent corresponding to the identified components in the Total Ion Chromatogram (TIC). Sensorial Analysis All the powder samples were reconstituted at 12% of soluble solids and slightly sweetened with sugar. The control was the original extract. All the reconstituted samples were evaluated for sensorial acceptability by pairing a preference test. The panel participating in the acceptance evaluation included 50 volunteers (50% females 1845 years), who were students and staff members of the Technological Institute of the West Mexico, Jalisco Campus. These panelists were randomly selected based on availability, interest and regular consumption of Roselle drinks. The reconstituted samples were chilled at 5 C for 24 h prior to sensory evaluation. Each panelist received the reconstituted sample in a random order, and a glass of water for rinsing and crackers for consumption between samples were supplied. Each panelist was served with 10 mL of chilled reconstituted samples and control in 30 mL clear plastic containers, which were randomly coded. In this paired preference test, panelists were asked to make a forced choice between the drinks [2, 4, 29]. This test was used as a measurement of relative preference of one Roselle drink over another. Statistical Analysis An Analysis of Variance (ANOVA) using the Statistical Analysis System Software (SAS) and Duncan test at (p >0.05), was performed with the data obtained from GCMS. Also, analysis of significant minimum difference (DMS) was made, among the Roselle powders to determine differences based on the pick area of each volatile component. The peak areas identified by GC-MS were used as variables. The paired preference data were analyzed by binomial distribution [23].

Results & Discussion Spray Drying of Roselle Extracts The processing conditions for all the spray drying experiments were kept at the outlet temperature (80 C) and the atomizer rate (26,000 rpm). In all experiments, Roselle powder showed a noticeable tendency to stick to internal stainless steel surfaces of the drying chamber especially at higher inlet temperature because this temperature increase the feed flow rate [12]. This is probably due to the nature of the soluble solids. The adherence of powder to walls of the drying chamber is a commonly recognized effect in spray drying of solutions containing sugars and solids easy to agglomerate [30]. The weight, moisture and pH of seven Roselle powder samples obtained by spray drying at different temperature are shown in Table 1. It is observed that the pH of the powder does not change with different temperature treatment; the name of current Roselle variety and the cited one should be mentioned [1, 12]. Volatile Compounds Identified in the Roselle Extract by SPME and GC-MS In the Roselle extract, 20 volatile compounds were identified. In general, there were terpene components, esters, aldehydes and phenolic derivatives. However, only the terpene compounds were similar to another works [6]. These compounds were limonene, linalool and alphaTerpineol. This has two significant reasons: first, the variety of the Roselle is different; this indicates that each variety can have different compositions. The second is that the isolation method of volatile compounds used in this work did not require boiling temperatures, only required an appropriate material of the fiber (DVB/Carboxen/PDMS) and a temperature of 60 C. This implies that the degradation problem in the samples of powder Roselle extracts were not gendered by the SPME analysis method. Volatile Compounds of the Roselle Powder The Roselle powder was rehydrated to the same soluble solid concentration than the liquid extract to determine the volatile compounds. In these samples 14 volatile com-

Table 1 Weight (grams), moisture percentage and pH of the Roselle powder dried at different temperatures Treatment Spray drying temperature (C) Weight (g) Humidity (%) pH T1 150 73 5.0 3.48 T2 160 74 4.0 3.48 T3 170 74 4.0 3.44 T4 180 78 4.0 3.44 T5 190 80 3.0 3.44 T6 200 75 3.0 3.48 T7 210 74 3.1 3.48

Plant Foods Hum Nutr (2009) 64:6267

65

pounds were identified. Ten of the 14 compounds identified are the same as the ones identified in the liquid extract, although there was a decrease in their concentration. Four different compounds were also generated, derived of the degradation of sugar and fatty acids that were present in the liquid Roselle extract [6]. In Table 2 the compounds of the Roselle powder are shown and the chromatograms are illustrated in Fig. 1. The compounds generated by chemical degradation of the samples during the spray drying were the furfural, cislinalool oxide, furanic linalool oxide Z and E, and the eugenol. It was observed that the concentration of the furfural increased because the drying temperature propitiated its formation by degradation of sugars. The cis-linalool oxide and the furanic linalool oxide Z and E were the product of the chemical degradation of the fatty acids. The eugenol is a compound resulting from degradation of the phenolic compounds as a consequence of the spray drying temperature of the Roselle samples. This also was reported in drying Roselle calyces [6]. In Fig. 2 is shown the upward behavior of the furfural, cis-linalool oxide, furanic linalool oxide Z

and E and eugenol compounds due to the obtained the changes by the spray drying temperature. Duncans test (p <0.05) determined that exist significant difference among the seven treatments realized to obtain Roselle powder at different spray drying temperatures. The Duncans test was based on the percentage area of the volatile compounds. Only the compounds the alphaterpinolene, ethyl hexadecanoate, and decanal compounds were present in all experiments without significant difference, as shown in Table 2. According to the analysis of the least significant difference (LSD) carried out among the Roselle powder samples, the Roselle extract dehydrated at 190 C presented the biggest concentration of volatile compounds and the least concentrations of degraded compounds that the other samples obtained from different spray drying temperatures, as shown in Table 2. This result proved the hypothesis that the drying temperature had an effect on the retention of volatiles compounds. In addition, it is important to acknowledge that significant differences exist among the compounds identified in the liquid extract as the compounds identified in the Roselle powder. This

Table 2 Volatiles identified in the Roselle powder Spray drying temperature (C) Area (%) T150 150 C Volatile compounds retained in the powder Terpenoids p-Cymene 0.10 ab Limonene 1.70 ab Alpha-Terpinolene 0.10 a Linalool 0.10 b Esters Ethyl hexadecanoate 1.70 a Ethyl linoleate 0.53 b Ethyl linoleolate 0.10 c Aldehydes Benzaldehyde 0.07 c Decanal 0.18 a Divers 4-Ethylguaiacol 0.06 b Volatile compounds generated by chemical degradation Sugar derivatives Furfural 0.30 d Phenolic derivatives Eugenol 0.09 c Fatty acid derivatives cis-Linalool oxide 0.76 b Furanic linalool oxide Z and E 0.20 c T160 160 C T170 170 C T180 180 C T190 190 C T200 200 C T210 210 C

0.14 1.60 0.09 0.09

ab ab a b

0.12 1.80 0.04 0.05

ab ab a b

0.13 1.50 0.06 0.06

ab ab a b

0.20 4.00 0.07 0.33

a a a a

0.12 1.00 0.10 0.25

ab b a ab

0.07 0.05 0.08 0.10

b b a b

1.70 a 0.50 b 0.13 c 0.09 c 0.16 a 0.06 b

1.72 a 0.47 b 0.13 c 0.35 b 0.17 a 0.07 b

1.68 a 0.71 ab 0.19 b 0.31 b 0.18 a 0.07 b

1.69 a 0.87 a 0.28 a 0.55 a 0.18 a 0.13 a

1.68 a 0.83 a 0.11 c 0.40 ab 0.16 a 0.06 b

1.67 a 0.45 b 0.10 c 0.05 c 0.16 a 0.02 c

2.51 c 0.10 c 0.87 b 0.29 b

2.69 c 0.10 c 0.85 b 0.35 b

3.80 b 0.15 b 1.00 ab 0.40 b

4.00 ab 0.20 b 1.07 ab 0.51 ab

4.28 a 0.35 a 1.55 a 0.69 a

4.44 a 0.30 a 1.40 a 0.50 ab

Different letters mean statistically significant difference using Duncan (p <0.05). Multiple range tests with base in their retention

66 Fig. 1 Chromatogram of volatile constituents in Roselle powder dried at 190 C. The time is in minutes

Plant Foods Hum Nutr (2009) 64:6267

indicates that it is difficult to obtain a powder exactly similar to the liquid extract of Roselle. Sensorial Analysis According to the sensorial analysis, more than 76% of the panelists showed that preference for the sample T190 than corresponded to 190 C drying outlet temperature, although the liquid extract has 92% of preference in the same conditions (T0). In comparison to the other treatments (different temperatures), this had the lowest percentage of
5 4.5 4 3.5

preference: T150=16%, T160=40%, T170=52%, T180= 52%, T200=12%, and T210=12%. Figure 3 shows the percentage over the flavor preference among all the samples analyzed, where the highest acceptability was for the liquid extract followed by the T190 sample with statistically significant differences (p <0.05). Therefore, it is clear than there is significant difference on acceptance among the Roselle powders and the original liquid extract according to the preference tests. The major components detected by chromatographic analysis played an important role.

Conclusions The inlet air temperature of 190 C used in the spray the drying of Roselle extract resulted in the best powder in terms of composition regarding the other treatments, since it contained the highest concentration of ten volatile com-

Area (%)

3 2.5 2 1.5

100 90 80

Acceptance (%)

1 0.5 0 150 160 170 180 190 200 210

70 60 50 40 30 20 10 0 T0 T150 T160 T170 T180 T190 T200 T210

Spray Drying Inlet Temperature (C)

Furfural Furanic linalool oxide Z and E Cis-linalool oxide Eugenol

Treatments at different temperatures (C)

Fig. 2 Effect of spray drying temperature on the concentration of degradation compounds present in Roselle powder

Fig. 3 Results of the paired preference tests of the Roselle reconstituted powder. T0 is Roselle liquid original extract

Plant Foods Hum Nutr (2009) 64:6267

67 12. Al-Kahtani AA, Hassan BH (1990) Spray drying of roselle (Hibiscus sabdariffa) extract. J Food Sci 55:10731078 doi:10.1111/ j.1365-2621.1990.tb01601.x 13. Andrade I, Flores H (2004) Optimization of spray drying of roselle extracts (Hibiscus sabdariffa L.). Drying Proceedings of the 14th International Drying Symposium. Sao Paulo, Brazil. A: 597604 14. Brenner J (1983) The essence of spray-dried flavors. Perf Flav 40:886888 15. Huang L, Passos ML, Kumar K, Mujumdar AS (2004) A threedimensional simulation of a spray dryer with a rotary atomizer. Drying. Proceedings of the 14th International Drying Symposium. Sao Paulo, Brazil. A: 319325 16. Re M, Messias LS, Schettini H (2004) The influence of the liquid properties and the atomizing conditions on the physical characteristics of the spray-dried ferrous sulfate microparticles. Drying. Proceedings of the 14th International Drying Symposium. Sao Paulo, Brazil. B: 11741181 17. King CJ, Kieckbush TG, Greenwald CG (1984) Food quality factors in spray drying. Adv Dry 3:322327 18. King CJ (1990) Spray drying food liquids and the retention of volatiles. Chem Eng Prog 54:3339 19. Futura T (2000) Estimating retention of emulsified flavor in a single droplet during drying. Dry Technol 2(3):11251130 20. Scarpellino R, Soukup RJ (1993) Key flavors from heat reactions of food ingredients. In: Acree TE, Teranishi R (eds) Flavour science. Sensible principles and techniques. American Chemical Society, Washington, DC 21. Beristain CI, Vernon-Carter EJ (1995) Studies on the interaction of arabic (Acacia senegal) and mesquite (Prosopis juliflora) gum as emulsion stabilizing agent. Dry Technol 13:455461 doi:10.1080/ 07373939508916965 22. R M (1998) Microencapsulation by spray drying. Dry Technol 16:11951236 doi:10.1080/07373939808917460 23. Pedrero FDL, Pangborn RM (1989) Evaluacin sensorial de los alimentosMtodos analticos. Alhambra mexicana S.A. de C.V., Mxico, D. F. 24. Zhang ZM, Zeng DD, Li GK (2008) Study of the violatile composition of tomato during storage by a combination sampling method coupled with gas chromatography/mass spectrometry. J Sci Food Agric 88:116124 25. Acree TE (1993) Bioassay in flavor research. In: Acree TE, Teranishi R (eds) Flavor science: sensible principles and techniques. American Chemical Society, Washington, DC, pp 122 26. Spanier AM, Miller JA (1993) Roles of proteins and peptides in meat flavor. In: Food flavor and safety. Molecular analysis and design. ACS Symp. Series 528, Washington, DC 27. AOAC (1996) Official methods of analysis, 18th edn. Association of Official Analytical Chemists, USA 28. Ortiz MA, Dorantes L, Galndez J, Guzmn RI (2003) Effect of different extraction methods on fatty acids, volatile compounds, and physical and chemical properties of avocado (Persea americana Mill.) oil. J Agric Food Chem 51:2216 2221 doi:10.1021/jf0207934 29. Shahidi F, Rubin L, Souza L (1986) Meat flavor volatiles: a review of the composition, techniques of analysis and sensory evaluation. Food Sci Nutr 24:141243 30. Brennan JG, Herrera J, Jowitt R (1971) A study of some factors affecting the spray drying of concentrated orange juice, on a laboratory scale. J Food Technol 6:295

pounds also showed in the original liquid extract, and the peak area of the degraded compounds is smaller than other samples treated at higher temperature. This result was rehearsed for panelists in the preference test, choosing the T190 sample as the best after the original extract, with an acceptability of 76%. This work suggests that the inlet air temperature of spray drying has a significant effect on the volatile compounds concentration of the Roselle liquid extract and therefore in its acceptance.
Acknowledgement We thank their support in this investigation to the ITA-JAL, DGETA, CoSNET and CIATEJ.

References
1. Mouning P, Badrie N (2006) Roselle/sorrel (Hibiscus sabdariffa L.) wines with varying calyx puree and total soluble solids: sensory acceptance, quantitative descriptive and physicochemical analysis. J Foodserv 17:102110 doi:10.1111/j.1745-4506.2006. 00028.x 2. Pouget MP, Vennat B, Pourrat A (1990) Identification of anthocyanins of Hibiscus sabdariffa. Food Technol 23:101102 3. DHeureux C, Badrie N (2004) Consumer acceptance and physicochemical quality of processed red sorrel/roselle (Hibiscus sabdariffa L.) sauces from enzymatic extracted calyces. Food Serv Technol 4:141148 doi:10.1111/j.1471-5740.2004.00100.x 4. Esselen WB, Sammy GM (1975) Applications for roselle as red color food colorant. Food Prod Develop 9:3740 5. Jirovetz L, Jger W, Remberg G, Espinosa-Gonzlez J, Mora R, Woidich A, Nikiforov A (1992) Analysis of the volatiles in the seed oil of Hibiscus sabdariffa (Malvaceae) by means of GC-MS and GC-FTIR. J Agric Food Chem 40:11861187 doi:10.1021/ jf00019a021 6. Chen SH, Huang TC, Ho CT, Tsai PJ (1998) Extraction, analysis, and study on the volatiles in roselle tea. J Agric Food Chem 46:11011105 doi:10.1021/jf970720y 7. Carvajal O, Waliszewski SM, Barradas DM, Orta Z, Hayward PM, Nolasco C, Angulo O, Sanchez R, Infanzon RM, Trujillo PRL (2005) The consumption of Hibiscus sabdariffa dried calyx ethanolic extract reduced lipid profile in rats. Plant Foods Hum Nutr 60:153159 doi:10.1007/s11130-005-9023-x 8. Akindahunsi AA, Olaleye MT (2003) Toxicological investigation of aqueous-methanolic extract of the calyces of Hibiscus sabdariffa L. J Ethnopharmacol 89:161164 doi:10.1016/S03788741(03)00276-9 9. Tsai P, McIntosh J, Pearce P, Camden B, Jordan B (2004) Anthocyanin and antioxidant capacity in roselle (Hibiscus sabdariffa L.) extract. Food Res Int 35:351356 doi:10.1016/ S0963-9969(01)00129-6 10. Chen HH, Tsai PJ, Chen SH, Su YM, Chung Ch, Huang TCh (2005) Grey relational analysis of dried roselle (Hibiscus sabdariffa L.). J Food Process Preserv 29:228245 11. Rosenberg M, Talmon Y, Kopelman IJ (1988) The microstructure of spray-dried microcapsules. Food Microstruct 7:1523