Physicochemical and Sensory (Aroma and Colour) Characterisation of a Non-centrifugal Cane Sugar (“panela”) Beverage

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Elsevier in Food Chemistry on August 2017, available at: https://doi.org/10.1016/j.foodchem.2017.01.134” 1 Physicochemical and sensory (aroma and colour) characterisation of a 1 non-centrifugal cane sugar (“panela”) beverage 2 3 Juliana María García,1 Paulo César Narváez,1 Francisco José Heredia,2 Álvaro 4 Orjuela,1* and Coralia Osorio3* 5 6 1 Grupo de Procesos Químicos y Bioquímicos, Departamento de Ingeniería Química y 7 Ambiental, Universidad Nacional de Colombia, 111321 Bogotá, Colombia. 8 2 Laboratory of Food Colour & Quality, University of Seville, 41012 Sevilla, Spain. 9 3 Departamento de Química, Universidad Nacional de Colombia-Sede Bogotá, AA 14490, 10 Bogotá, Colombia. 11 12 Abbreviated running header: Odour active volatiles and colour in Colombian non- 13 centrifugal cane sugar beverage from three geographical sources 14 15 Correspondence to: Coralia Osorio, tel.: +57 1 3165000, ext. 14452; fax: +57 1 3165220; e- 16 mail: [email protected]; and Alvaro Orjuela, tel: +57-1-3165000, ext. 14303; e-mail: 17 [email protected]. 18 19 20 21 22 2 Abstract 23 24 Non-centrifugal cane sugar (NCS), also called “panela”, is a high carbohydrate-content 25 food obtained by boil evaporation of the sugar cane juice. This study was undertaken to 26 characterise the physicochemical properties and sensory characteristics of the panela 27 beverage at two different concentrations. Evaluation of pH, °Brix, and colour (tristimulus 28 colorimetry) was carried out in all panela drink samples. In order to characterise the odour- 29 active volatiles of the beverage, a simultaneous steam distillation-solvent extraction (SDE) 30 method was applied using a mixture of diethyl ether-pentane (1:1, w/w) as solvent. The 31 Aroma Extract Dilution Analysis (AEDA) revealed the presence of six odour-active 32 compounds, being 2-methyl pyrazine and furfural the key aroma compounds. By using 33 PCA (Principal Component Analysis), there were no found any sensory and 34 physicochemical difference regarding the geographical origin of samples. 35 36 Keywords: odour-active volatiles; simultaneous steam distillation-solvent extraction 37 (SDE); Saccharum officinarum L derivative; non-centrifugal cane sugar (NCS); jaggery. 38 39 Chemical compounds studied in this article: 2-Methylpyrazine (PubChem CID: 7976), 40 2,5-Dimethylpirazine (PubChem CID: 31252), Furfural (PubChem CID: 7362), Propanoic 41 acid (PubChem CID: 1032), 2-Methylpropanoic acid (PubChem CID: 6590), 2- 42 Furanmethanol (PubChem CID: 7361). 43 44 45 3 1. Introduction 46 47 Due to the increasing concern on the negative impacts of excessive refined sugars 48 intake, the global demand for sugar substitutes is continuously expanding. Recent estimates 49 indicate that the sugar substitutes market is around 11.5 billion dollars and it is expected to 50 grow up to 14 billion by 2019 (Markets and markets, 2015). Among the different 51 commercial substitutes, consumers’ preference is centered on those from natural origin 52 because they are considered healthier and safer than processed ones. An important natural 53 sweetener in developing countries is the non-centrifugal cane sugar (NCS), a solid 54 unrefined product obtained from the processing of sugarcane juice (FAO, 1994). 55 Depending on the country of origin, it is traditionally known with different names such as 56 “panela”, jaggery, gur, “piloncillo”, “mascavo”, or “chancaca”. 57 In general, NCS is produced from the extracted, cleaned, and clarified sugar cane 58 juice, by open evaporation of its water content up to 88-94 ºBrix, obtaining a hot and 59 caramel-like product that solidifies at 18 °C. NCS exhibits a high sucrose content (84.49 ± 60 5.83 %) and some additional constituents as reducing sugars (glucose and fructose, 7.33 ± 61 2.81 %), minerals (K 531.26 mg/100 g, Ca 102.62 mg/100 g, P 57.54 mg /100 g), vitamins 62 (vitamin E 55.65 mg/100g, vitamin C 4.23 mg/100 g), organic acids, amino acids, among 63 others (Jaffé, 2015; Guerra & Mujica, 2010). Phenolic compounds with antioxidant activity 64 have been also identified in this sugarcane derivative (Jaffé, 2015; Duarte-Almeida, 65 Salatino, Genovese & Lajolo, 2011). 66 According with recent reports, world production of NCS is around 172 million tons 67 per year (Foreign Agricultural Service, USDA, 2015), and it is dominated by India (~50%), 68 4 Colombia (~17%), and Pakistan (10%). In Colombia, together with coffee, a NCS hot 69 infusion (in water) is one of the most important local drinks. This beverage is commonly 70 known as “agua de panela” (i. e. water of panela) and it is used either as source of energy 71 or as a warming up infusion. It is also consumed as a cold refreshing beverage traditionally 72 mixed with limejuice. Because it is a high caloric solution with significant content of salts 73 and vitamins, it is used as rehydrating beverage, and to relief for cold and flu symptoms. 74 Although this traditional beverage represents an important cultural heritage in 75 Colombia, almost no studies have been done to characterise its physicochemical and 76 sensory properties. According with the only study available in open literature, around thirty 77 volatile compounds were identified by GC-MS in samples treated by Soxhlet extraction 78 with dichloromethane. The main volatile compounds were furans, furanones, 2- 79 acetylpyrrole, and 5-(hydroxy-methyl) furfural; these components are produced by the 80 thermal degradation of carbohydrates via caramelisation or Maillard reactions during the 81 drying of sugarcane juice (Payet et al., 2005). 82 Worldwide, there is an increasing interest for developing and consuming new added 83 value naturally-based and artisanal-manufactured foods and drinks with protected 84 designation of origin. Thus, the aim of this work was to carry out a physicochemical and 85 sensory (aroma and color) characterization of “panela” beverage. This was prepared with 86 NCS samples obtained from three different geographical regions of Colombia at different 87 concentrations, allowing to assess the influence of those variables in the properties of the 88 beverage. The medium-term purpose of this research is to use the obtained data as 89 comparing quality parameters during the development of new processed products from 90 sugarcane juice for the food industry. 91 5 2. Materials and methods 92 93 2.1. “Panela” samples 94 95 Solid “panela” blocks of ca. 500g, properly labeled with nutritional facts and 96 geographical origin, were purchased at local stores, and used in experiments without further 97 purification. The samples originated from three different geographical regions with 98 different edaphoclimatic characteristic in Colombia, which are the states of major 99 production in Colombia: Santander (northeast), Cundinamarca (center), and Valle del 100 Cauca (southwest). 101 102 2.2. Chemicals 103 104 Diethyl ether, n-pentane, and anhydrous sodium sulfate were acquired from Merck 105 (Darmstadt, Germany). All solvents were freshly distilled prior to use. An n-Alkane mix 106 (C8-C26) was purchased to Laboratory Dr.Ehrenstorfer GmbH, Augsburg, Germany. Pure 107 reference standards of 2-methyl pyrazine, 2,5-dimethyl pyrazine, furfural, propanoic acid, 108 2-methyl propanoic acid, and 2-furan methanol were were generously supplied by 109 DISAROMAS S.A. (Bogotá, Colombia). -Nonalactone (Sigma Chem. Co., St. Louis, MO, 110 USA) was used as internal standard. 111 112 2.3. Physicochemical characterisation 113 6 114 “Panela” samples were used as purchased to prepare the hot beverage following the 115 traditional method at two different concentrations (25 % and 60 % w/w). In each case, the 116 required weight of “panela” was put into a glass beaker and dissolved in 1000 mL boiling 117 water (362 K under local pressure). The solution was maintained under continuous stirring 118 and heating until complete dissolution. Once a homogeneous solution was obtained, heating 119 and stirring were stopped, and the beverage was let cool to room temperature. Residual 120 insoluble solids were removed by filtration before characterization. Further measurements 121 were performed in triplicate to report the mean and standard deviation. 122 123 2.3.1. Total soluble solid content and pH 124 Total soluble solids were determined by using an Atago refractometer (HRS-500, 125 Tokyo, Japan) and the results were expressed as °Brix. The pH of the beverage was 126 determined by using a 370 pHmeter (Jenway, London, UK). 127 128 2.3.2. Colour measurement 129 Colour of different “panela” beverage was determined by using a Cary 5000 UV-Vis- 130 NIR spectrophotometer (Varian, Victoria, Australia) with a 10 mm path length quartz cell. 131 Visible absorption spectrum of each sample was recorded at wavelengths between 380 and 132 770 nm. CIELAB parameters (L*, a*, b*) were determined by using the 133 CromaLab® software (Heredia, Alvarez, González-Miret & Ramírez, 2004). All 134 measurements were performed in triplicate to report the mean ± standard deviation. The 135 7 colour parameters, chroma (Cab*) and hue (hab), were calculated according to the following 136 equations (Meléndez-Martínez, Vicario & Heredia, 2003): 137     1 22 2 ab C a b        (1) 138 arctan ab b ha      (2) 139 2.4. Isolation of volatile extract 140 141 A Likens–Nickerson type SDE (simultaneous steam distillation-solvent extraction) 142 apparatus (Likens & Nickerson, 1964) was used to extract volatile compounds of the 143 different samples. For the analysis, 800 mL of each sample were loaded in a 2 L round 144 bottom flask, and the extraction was carried out with 500 mL of n-pentane-diethyl ether 145 (1:1, v/v) during 2 h. Each extract was dried over Na2SO4 and concentrated to 1.0 mL using 146 a Vigreux column. 147 148 2.5. Analyses of odour-active volatiles by GC–O (Gas Chromatography–Olfactometry) 149 and Aroma Extract Dilution Analysis (AEDA) 150 151 GC-O analyses were performed in a gas chromatograph HP 5890 Series II (Hewlett- 152 Packard, USA) equipped with an FID and operated in split mode (1:10, injected volume, 1 153 L). The injection port was set at 230 °C and Helium was used as carrier gas at 1.0 154 mL/min. Two capillary columns, DB-FFAP and DB-5 (each 30 m × 0.32 mm i.d., 0.25 μm 155 film thickness; J&W Scientific, Chromatographie-Handel Müller, Fridolfing, Germany, and 156 8 Restek, USA, respectively) were used for the volatile analyses. The column oven was 157 programmed from 40 °C (after 4 min) to 180 °C at 6 °C/min, then at 12 °C/min until 230 158 °C for the DB-FFAP and 300 °C for the DB-5, and finally held them for 10 min at the 159 maximum temperatures. The end of the capillaries were connected to a deactivated Y- 160 shaped glass splitter (Chromatographie Handel Mueller, Fridolfing, Germany), which 161 divides the effluent into two equal parts, one for FID (230 °C) and the other for heated 162 sniffing port (200 °C) by using deactivated fused silica capillaries of the same length (50 163 cm x 0.32 mm i.d.). Sniffing port consisted of a self-made elbow-shaped aluminium tube 164 (80 x 5 mm i.d). 165 Three panelists located the odour-active zones of the SDE extracts by GC-O. Then, the 166 aroma extract was stepwise diluted to obtain 2n dilutions, and each solution was analysed 167 by GC-O. The odour activity of each compound, expressed as flavor dilution factor (FD), 168 was determined as the greatest dilution at which that compound was still detected by 169 comparing all of the runs (Grosch, 1994). The FD factors obtained by three panelists were 170 averaged. 171 172 2.6. Gas Chromatography–Mass Spectrometry (GC–MS) 173 174 GC-MS (EIMS) analyses were carried out on a GC Agilent 7890B gas 175 chromatograph equipped with an ITQ 900 mass selective detector (Agilent Technologies 176 Inc. Wilmington, DE, USA). MS data were recorded between 40-400 u, with an electron 177 energy of 70 eV and processed by Mass Hunter software. Chromatographic conditions were 178 15 References 313 314 Duarte-Almeida, J. M., Salatino, A., Genovese, M. I., & Lajolo, F. M. (2011). Phenolic 315 composition and antioxidant activity of culms and sugarcane (Saccharum officinarum 316 L.) products. Food Chemistry, 125, 660-664. 317 FAO, Food and Agriculture Organization of the United Nations (1994). Definition and 318 classification of commodities. 3. Sugar crops and sweeteners and derived products. 319 Retrieved from http://www.fao.org/es/faodef/faodefe.htm. 320 Fisher, C., & Scott, T. R. (1997). Food Flavours, Biology and Chemistry. Cambridge, UK : 321 The Royal Society of Chemistry. 322 Foreign Agricultural Service, USDA. (2015). 323 https://apps.fas.usda.gov/psdonline/circulars/sugar.pdf. Consulted on line february 2015. 324 Grosch, W. (1994). Determination of potent odourants in foods by aroma extract dilution 325 analysis (AEDA) and calculation of odour activity values (OAVs). Flavour and 326 Fragrance Journal, 9, 147-158. 327 Guerra, M. J., & Mujica, M. V. (2010). Physical and chemical properties of granulated cane 328 sugar "panelas". Ciência e Tecnologia de Alimentos, 30, 1-9. 329 Heredia, F.J., Álvarez, C., González-Miret, M.L., & Ramírez, A. (2004). Cromalab®, 330 análisis de color. Registro General de la Propiedad Intelectual SE-1052-04. Seville, 331 Spain. 332 IOFI Working Group on Methods of Analysis (2011). Guidelines for the quantitative gas 333 chromatography of volatile flavouring substances, from the Working Group on Methods 334 16 of Analysis of the International Organization of the Flavor Industry (IOFI). Flavour and 335 Fragance Journal, 26, 297-299. 336 Jaffe, W. R. (2015). Nutritional and functional components of non centrifugal cane sugar: 337 A compilation of the data from the analytical literature. Journal of Food Composition 338 and Analysis, 43, 194-202. 339 Leffingwell & Associates. (2008). Odor Detection Thresholds and References. 340 http://www.leffingwell.com/odorthre.htm. March 2015. 341 Likens, S. T., & Nickerson, G. B. (1964). Detection of certain hop oil constituents in 342 brewing products. Journal of the American Society of Brewing Chemists, 11, 5–13 343 Markets and markets. Sugar Substitutes Market. (2015). Available at: 344 http://www.marketsandmarkets.com/Market-Reports/sugar-substitute-market-1134.html. 345 Meléndez-Martínez, A. J., Vicario, I. M., & Heredia, F. J. (2003). Application of 346 tristimulus colorimetrý to estimate the carotenoids content in ultrafrozen orange juices. 347 Journal of Agricultural and Food Chemistry, 51, 7266-7270. 348 Payet, B., Singh, A.S.C., & Smadja, J. (2005). Assessment of antioxidant activity of cane 349 brown sugars by ABTS and DPPH radical scavenging assays: determination of their 350 polyphenolic and volatile constituents. Journal of Agricultural and Food Chemistry, 53, 351 10074–10079. 352 Pripdeevech, P., & Machan, T. (2011). Fingerprint of volatile flavour constituents and 353 antioxidant activities of teas from Thailand. Food Chemistry, 125, 797-802. 354 Schieberle, P. (1995). Recent developments in methods for analysis of volatile flavour 355 compounds and their precursor. In G. Gaonkar (Ed.), Characterization of food: 356 Emerging methods (pp. 403–431). Amsterdam, The Netherlands: Elsevier. 357 17 Verschueren, K. (2001) Handbook of Environmental Data on Organic Chemicals. Volumes 358 1-2. 4th ed. John Wiley & Sons. New York, NY p. 1185. 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 18 Figure captions 381 382 Fig. 1. Localisation on the (a*, b*) diagram of colour parameters of panela beverage 383 samples from different geographical origins, obtained at two concentrations, 25% w/w and 384 60% w/w. Cundinamar ca (M1-60, M1-25), Santander (M2-60, M2-25), and 385 Valle del Cauca (M3-60, M3-25). 386 Fig. 2. GC analyses on FFAP column of the volatile compounds from panela beverages 387 (60% w/w) obtained by SDE from, A) Cundinamarca’s sample, B) Santander´s sample, and 388 C) Valle´s sample. Numbers correspond to Table 2. 389 Fig. 3. Principal component analysis (PCA) (A) loading plot, (B) score plot for 390 physicochemical parameters measured in panela beverage from different geographical 391 origins (C = Cundinamarca, S = Santander, and V= Valle del Cauca) at two concentrations 392 (25 = 25% w/w, 60 = 60% w/w), B). PH= pH, Brix= °Brix, Ca = a*, Cb = b* and CL = L*; 393 the concentration of each compound shown in Table 2 correspond to C1 = 2- 394 methylpyrazine, C2 = 2,5-dimethylpyrazine, C3= furfural, C4 = propanoic acid, C5 = 2- 395 methylpropanoic acid, and C6 = 2-furanmethanol. 396 397 398 399 400 401 402 403 404 19 Figure 1. 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 20 Figure 2 420 421 21 Figure 3. 422 A. Loading plot 423 Projection of the variables on the factor-plane ( 1 x 2) Active PH Brix CL Ca Cb C1 C2 C3 C4 C5 C6 -1.0 -0.5 0.0 0.5 1.0 Factor 1 : 54.68% -1.0 -0.5 0.0 0.5 1.0 Factor 2 : 31.41% PH Brix CL Ca Cb C1 C2 C3 C4 C5 C6 424 B. Score 425 Projection of the cases on the factor-plane ( 1 x 2) Active C25 C60 S25 S60 V25 V60 -5 -4 -3 -2 -1 0 1 2 3 4 5 Factor 1: 54.68% -4 -3 -2 -1 0 1 2 3 4 5 Factor 2: 31.41% C25 C60 S25 S60 V25 V60 426 22 Table 1 Physicochemical characterisation of panela beverage at different concentration and from different geographical regions a All data are the mean of three measurements ± standard deviation. Equal letters in a file means that there are no significant differences based on Tukey test (p < 0.05). Sample origin Propertya Cundinamarca (C) Santander (S) Valle del Cauca (V) Concentration (w/w) 25% 60% 25% 60% 25% 60% Total soluble solids (°Brix) 24.4 ± 0.5a 39.6 ± 5.5b 17.2 ± 1.1c 40.9 ± 3.4b 22.9 ± 1.2a 37.5 ± 5.4b pH 5.51 ± 0.11a 5.40 ± 0.01a 5.80 ± 0.03b 5.58 ± 0.05a 5.82 ± 0.05b 5.75 ± 0.22ab Colour parameters L* 71.6 ± 4.5a 73.0 ± 2.2a 89.8 ± 1.5b 85.6 ± 2.8bc 87.4 ± 0.8c 77.7 ± 2.9a a* 2.7 ± 1.6a 3.0 ± 0.2b -2.1 ± 0.2c -1.5 ± 0.3d -0.4 ± 0.2e 1.1 ± 0.4a b* 36.1 ± 6.5a 42.3 ± 0.7ac 24.3 ± 3.2b 25.8 ± 3.0b 22.3 ± 2.4b 33.2 ± 1.2ac C*ab 36.2 ± 6.6a 42.4 ± 0.7ac 24.4 ± 3.2b 25.9 ± 3.0b 22.3 ± 2.4b 33.2 ± 1.2ac hab 85.9 ± 2.0a 86.0 ± 0.4a 94.9± 1.0b 93.4 ± 0.7b 91.1 ± 0.7c 88.1 ± 0.6a 23 Table 2 Odour-active volatiles detected in panela beverage SDE extracts from different geographical regions Noa Odorantb Odour description c RI FDd Odour Threshold (µg/Kg water) Conc ± SD (% CV) (µg/Kg panela)e OAVf C S V C S V FFAP DB-5 1 2-Methylpyrazine Nutty, panela-like 1266 801 4096 60g 494.9 ± 2.1 (0.4%) 637.2 ± 0.1 (0.0%) 96.1 ± 0.9 (0.9%) 8 11 2 2 2,5-Dimethylpyrazine Burnt, toasted 1320 889 1024 800g 267.2 ± 0.0 (0.1%) 241.9 ± 0.1(0.0%) 247.2 ± 0.9 (0.3%) <1 <1 <1 3 Furfural Sweet, panela-like 1457 804 64 3000h 266.4 ± 0.1 (0.0%) 138.5 ± 0.1(0.0%) 302.1 ± 0.4 (0.1%) <1 <1 <1 4 Propanoic acid rancid 1535 714 512 20000g 109.8 ± 1.5 (1.4%) 126.3 ± 0.0 (0.0%) 89.4 ± 1.4 (1.5%) <1 <1 <1 5 2-Methylpropanoic acid Cooked, rancid 1570 807 64 8100g 139.3 ± 0.0 (0.0%) 184.8 ± 0.3(0.2%) 46.1 ± 1.0 (2.2%) <1 <1 <1 6 2-Furanmethanol Honey 1660 866 256 8000g 81.8 ± 1.9 (2.3%) 162.1 ± 0.9 (0.5%) 27.3 ± 0.7 (2.7%) <1 <1 <1 a Number were consecutively assigned as elution order in FFAP column (Fig. 2). b Odorants were RI (Retention index), odour quality, and mass spectra compared with reference compounds. c Odour quality as perceived at the sniffing port during GC-O. d FD factor = flavour dilution factor e All data are the mean of three measurements ± standard deviation. In all of the cases, correlation coefficient was higher than 0.995 and response. Geographical origin: C = Cundinamarca, S = Santander, V = Valle del Cauca, at 60% w/w concentration. f OAV = odour activity value, concentration divided by odour threshold. This parameter was calculated only for 60% (w/w) beverages. g Leffingwell et al., 2008. h Verschueren (2001).