Received: 7 April 2016 | Revised: 2 August 2016 | Accepted: 4 August 2016

DOI 10.1111/jfpe.12470

ORIGINAL ARTICLE

Physical properties of spray dryed Stenocereus griseus pitaya

juice powder

ndez-Lagunas1 | Juan Rodríguez-Ramírez1 |

Karina A. García-Lucas1 | Lilia L. Me
Osvaldo H. Campanella2 | Bhavesh K. Patel2 | Luis Gerardo Barriada-Bernal3

1

cnico Nacional, CIIDIR-IPN-
Instituto Polite
Oaxaca, Hornos No. 1003, Santa Cruz
Abstract
Xoxocotl
an, Oaxaca 71230, Me
xico The aim of this work was to analyze the effect of spray drying conditions on the physical proper-
2
Whistler Carbohydrate Research Center ties of pitaya juice powder using a 23 factorial design. Independent variables were inlet air
and Agricultural and Biological Engineering, temperature (150 and 180°C), feed flow rate (1.5 and 2 L/hr), and the amount of encapsulants (60
Purdue University, West Lafayette, Indiana,
and 90%). The spray process was performed on pilot-scale spray dryer and maltodextrin/pectin
USA
3 combination was used as encapsulating material. Moisture content, yield, color, water adsorption,
Consejo Nacional de Ciencia y Tecnología,
Hornos No. 1003, Santa Cruz Xoxocotl
an, glass transition temperature, and morphology of pitaya juice powders were analyzed. Critical con-
Oaxaca 71230, Me
xico ditions of storage, at which the glass transition occurs were determined. Experimental results
Correspondence showed that all the individual factors had a significant effect on moisture content and color, how-

ndez-Lagunas, Instituto Polite
Lilia L. Me
c-
ever, yield was only affected by feed flow rate and percentage of encapsulants. Lower drying
nico Nacional, CIIDIR-IPN-Oaxaca, Hornos
No. 1003, Santa Cruz Xoxocotl
an, Oaxaca temperatures (150°C) favored higher yield of 59% and privileged red-violet color retention during
71230, Me
xico. spray drying. The GAB and Gordon-Taylor models gave good fit for the experimental data of
Email: mendezll@hotmail.com
adsorption isotherms and glass transition temperatures, respectively.
Funding Information
The authors wish to thank the Instituto Practical Applications

cnico Nacional-Mexico for the
Polite In this work betalains with compounds including nutraceutical properties, which can be used as a
financial support provided through the SIP
dye, were encapsulated. The knowledge of the properties of natural products, with potential either
Project: 20130767, CONACyT (grant
268716), BEIFI-IPN and COFFA-IPN. as an additive or as a colorant, is of great significance for applications in both the food and phar-
maceutical industry. The efficiency, color, morphology of powders, and properties influencing
hygroscopicity, collapse, compression, and agglomeration were determined. These factors control
the processability, handling properties, and the stability of pitaya powders.

KEYWORDS
adsorption isotherms, color, glass transition, morphology, pitaya, spray drying

1 | INTRODUCTION increased due to the high consumer demand for natural products, mainly
because of the artificial colorants that have been associated with toxicity
Pitaya (Stenocereus griseus) is a fruit produced by the columnar cactus, and carcinogenicity in humans (Boo et al., 2012). A lot of techniques are
native to Mexico. It is round-oval in shape and the color of its flesh varies used for microencapsulation of colorants, yet the most efficient, cost-
(violet, red, orange, yellow, and white) depending on the variety. The effective and advantageous for heat sensitive products is spray drying
color of the flesh is caused by the presence of betalains, natural water- (Cai & Corke, 2000; Pitalua, Jimenez, Vernon-Carter, & Beristain, 2010).
soluble pigments. Betalains exhibit a wide range of desirable biological Spray drying is a process in which a liquid (solution, emulsion or suspen-
activities, including antioxidant, anti-inflammatory, hepatoprotective and sion) is atomized into a stream of hot (air or nitrogen) gas to obtain a pow-
anti-cancer properties (Georgiev et al., 2010). Betalains are thermolabile, der instantly (Gharsallaoui, Roudaut, Chambin, Voilley, & Saurel, 2007).
hence susceptible to degradation during processing. However, they are a This process is suitable for heat sensitive products because it offers rela-
promising source of natural colorant and bioactive compounds. The food tively low temperatures and short residence times (5–100 s) (Barbosa-
industry’s interest in the use of natural colorants has most recently C
anovas & Vega-Mercado, 1996; Filkov
a, Huang, & Mujumdar, 2006).

Journal of Food Process Engineering 2016; 1-9 wileyonlinelibrary.com/journal/jfpe V

C 2016 Wiley Periodicals, Inc. | 1
2 | GARCÍA-LUCAS ET AL.

FIGURE 1 Pitaya fresh sliced in halves (A) and pitaya spray-dried powders (B)

During microencapsulation liquid droplets, solid particles or ture content, yield, color, adsorption isotherm, glass transition
gas compounds are entrapped into thin films of a microencapsulat- temperature, and morphology of pitaya juice powders.
ing agent forming a core. The retention of these cores is governed
by their chemical functionality, solubility, polarity, and volatility 2 | MATERIALS AND METHODS
(Gharsallaoui et al., 2007). Microencapsulation in spray-drying
involves these steps; preparation of the dispersion, homogeniza- 2.1 | Raw materials
tion of the dispersion, atomization of the infeed emulsion, and
Pitayas (Stenocereus griseus) were obtained from Puebla, Mexico. Flaw-
dehydration (Shahidi & Han, 1993). In this process, although the
less, mature, and same color peel was selected from various fruits (Fig-
evaporation phenomenon is predominant, convection, diffusion of
ure 1A). The fruits were kept frozen until used. To prepare the juice,
species and changes in interactions also occur between molecules
the spines and peel were eliminated and the pulp, manually crushed,
due to an increased concentration (Gharsallaoui et al., 2007; Mas-
was passed through 60 mesh sieves. Fresh juice had 10.2 ± 0.2°Brix
ters, 1991).
and a total of solids of 87.55 ± 2.33 g/L.
Powder from fruit juice has a sticky behavior attributable to the
low glass transition temperature (Tg) of sugars of low molecular weight
as fructose (5°C), glucose (31°C), sucrose (62°C) and organic acids (cit- 2.2 | Preparation of suspensions
ric acid, malic acid or tartaric acid), which constitute over 90% of the The suspensions were prepared with pitaya juice and a mix of malto-
solids in fruit juices (Bhandari, Datta, Crooks, Howes, & Rigby, 1997; dextrin (13 DE, Cytecsa) and pectin (Cytecsa). The mixture of encapsu-
Fazaeli, Emam-Djomeh, Ashtari, & Omid, 2012). Encapsulants of high lants contained a 40:20 and 60:30 ratios of maltodextrin and pectin,
molecular weight should be used to dry sticky juices to increase (Tg) respectively pitaya juice was added at two levels, at 60% and 90% of
and ensure efficient drying. the total solids of juice.
Encapsulated process has been successfully applied for pig- The suspensions were homogenized for 5 min and submitted to
ments stability in plant foods (Fang & Bhandari, 2011; Goula & Ada- magnetic agitation at 400 rpm for an hour.
mopoulos, 2005b; Grabowski, Truong, & Daubert, 2006;
Ravichandran et al., 2014). The spray drying process produces a
2.3 | Spray drying
high-quality final product with good sensory and nutritional charac-
teristics. The physicochemical properties of spray-dried powders A co-current pilot-scale spray dryer (Niro, Copenhagen, Denmark)
mainly depend on the operating parameters (inlet air temperature, equipped with a rotary atomizer (TS-Minor, M02/A) was used for all
feed flow rate, atomization speed), as well as the type and concen- samples. Distilled water at room temperature (25ºC) was used for
tration of encapsulants (Phisut, 2012). Studying the effect of operat- the start-up and shut down operations. The suspensions were fed
ing parameters on the physical properties of powder helps identify into the dryer using a peristaltic pump (Watson-Marlow 505S/RL).
the optimum spray drying conditions, thus, these must be used to The atomization speed remained constant at (28,000 rpm). The pow-
obtained good microencapsulation efficiency (Chegini & Ghobadian, ders (collected from cyclone and manually sweeping walls of the
2007; Kandansamy & Somasundaram, 2012). drying chamber) were deposited in polyethylene bags and vacuum
The objective of this work was to evaluate the effect of spray dry- sealed. The bags were stored in a desiccator containing silica gel
ing operation parameters and the amount of encapsulants in the mois- until later use.
GARCÍA-LUCAS ET AL. | 3

2.4 | Physical properties of pitaya juice powder stant related to monolayer sorption heat, K is a constant related to mul-
tilayer sorption heat and aw is water activity.
2.4.1 | Moisture content
The solid surface area was determined using the following equa-
Moisture content was determined using the gravimetric technique tion 6 (Cassini, Marczak, & Noren~a, 2006).
described in the A.O.A.C. 930.15 method (1984); moisture loss was  
1
expressed as percentage dry base (db). SA ¼ Xm   NA  AH2 O ¼ 3:53103 Xm (6)
PMH2 O

2.4.2 | Yield Where SA is the solid surface area (m2/g of solid), Xm is the mono-
The powder yield was determined by the following relationship 1: layer moisture content (g H2O/g dry matter), PMH2O is the molecular
weight of water (18 g/mol), NA Avogadro’s number (6 3 1023 mole-
Wp 2Xwb  Wp
y¼  100 (1) cules/mol), and AH2O is the area of a water molecule (10.6 310220 m2).
Vs  ST

Where y is yield (%); Wp is powder weight (g); Xwb is the moisture 2.4.5 | Glass transition temperature
content of the powders on a wet base (wb); Vs is the volume of sample The powders were equilibrated in desiccators with different saturated
fed into the dryer (L); and ST is the total solids (g dry matter/L), deter- salt solutions (LiCl, MgCl2.6H2O, K2CO3 and Mg(NO3)2) to maintain aw
mined by Equation 2 in accordance with the A.O.A.C (1984) rule. between 0.11 and 0.53 at 25°C. The glass transition temperature (Tg)
W2 2W1 was determined with a differential scanning calorimeter model Q2000
ST ¼ (2)
V (TA Instruments, New Castle, DE, USA). The samples were placed in

Where W2 is the weight of the capsule with dry residue (g); W1 is aluminum pans and hermetically sealed. An empty pan was used as a

the capsule weight (g); and V is the sample volume (cm ). 3 reference. The sample was kept at 240°C for 10 min and then heated
at a rate of 2°C/min at 120°C with modulation amplitude of ±1.272°C
2.4.3 | Color and a period of 60 s. Tg was calculated using the Software Universal

The color of the powder was determined using a MiniScan EZ colorim- Analysis 2000 version 4.5A (TA Instruments, New Castle, DE, USA).

eter model (HunterLab, Reston, VA, USA) with D65 as an illuminant, a Experimental Tg– xw (g H2O/g product) data was fitted to the Gordon-

10° observation angle as a reference system, and the results were Taylor model (Eq. 7) (Gordon & Taylor, 1952). The parameters for this

recorded on a CIELAB scale (L*, a*, and b*). L* represents the lightness model were estimated using the MS Excel Solver algorithm (Microsoft,

of the sample ranging from 0 (black) to 100 (white), coordinate a* rep- Redmond, WA, USA).

resents red color (1) or green (2), and coordinate b* represents yellow ð12xw Þ:Tgs 1k:xw :Tgw
Tg ¼ (7)
color (1) or blue (2). The hue angle (h) and chroma (C*) were calculated ð12xw Þ1k:xw
according to Eqns. 3 and 4. The hue angle values vary from 0° (pure where Tg, Tgs, Tgw are the glass transition temperature (°C) of the pow-
red color), 90° (pure yellow color), 180° (pure green color) to 270° ders, solids and water respectively, xw is the mass fraction of water (g
(pure blue color) (Sant’Anna, Gurak, Marczak, & Tessaro, 2013). The H2O/g product) and k is a constant model. Tgw = 2135°C (Johari, Hall-
measurements were made in triplicate. brucker, & Mayer, 1987).
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C ¼ a2 1b2 (3) 2.4.6 | Storage stability
 
b The water activity and Tg value were predicted by the GAB and
h ¼ arctan  (4)
a
Gordon-Taylor models correspondingly, and were plotted in a single
graph. By analyzing this figure, it is possible to obtain the critical values
2.4.4 | Adsorption isotherms for the water activity (awc) and water content (Xc) (Moraga, Martíinez-
Adsorption isotherms were determined using VTI-SA1, a dynamic Navarrete, & Chiralt, 2004).
vapor sorption analyzer instrument (TA Instruments, New Castle, DE,
2.4.7 | Particle morphology
USA) with water activity (aw) ranging from 0.05 to 0.80 at 25°C. Exper-
imental data was fitted to the GAB (Guggenheim-Anderson-Deboer) Particle morphology was evaluated using a Nova NanoSEM200 (FEI

model (Eq. 5). The parameters for this model were estimated with the Company, Dawson Creek Drive Hillsboro, Oregon, USA) scanning elec-

MS Excel Solver algorithm (Microsoft, Redmond, WA, USA). The initial tron microscope (SEM). The samples were covered with platinum. The
coating was carried out in a sputter coater 208 HR (Cressington Scien-
values of the parameters introduced in the software were delimited for
tific Instruments, England, UK). The SEM was operated at 5kV with
their physical significance (Quirijns, Van Boxtel, Van Loon, & Van,
magnifications of 12003.
2005a).

Xm  C  K  aw 2.4.8 | Statistical analysis

Xeq ¼ (5)
ð12K  aw Þ  ð12K  aw 1C  K  aw Þ
A full factorial design 23 (Table 1) was used to evaluate the effect of
where Xeq is the equilibrium moisture content (g H2O/g dry matter), Xm the following operating parameters: inlet air temperature, Ti (150–180°
is the monolayer moisture content (g H2O/g dry matter), C is a con- C); feed flow rate, Fe (1.5–2.0 L/hr); and percentage of encapsulants,
4 | GARCÍA-LUCAS ET AL.

TA BL E 1 Factorial design of drying, moisture content and yield of lower moisture content (Ferrari, Germer, & De Aguirre, 2012; Quek,
powders Chok, & Swedlund, 2007; Tonon et al., 2008). With an Em/p increment,
Treatment Ti (°C) Fe (L/hr) Em/p (%) MC (%,db) y (%) quantity of present solids in feeding increases, reducing the quantity of
P1 150 1.5 60 3.47 ± 0.17 59.41 ± 0.97 water to evaporate (Abadio, Domingues, Borges, & Oliveira, 2004;

P2 180 1.5 60 2.98 ± 0.33 62.43 ± 0.19 Fazaeli et al., 2012; Kha, Nguyen, & Roach, 2010). On the other hand,

P3 150 2.0 60 3.98 ± 0.31 59.94 ± 2.80

the decrease in Fe increments the time of contact between the particle
and drying air causing a greater driving force for water evaporation
P4 180 2.0 60 3.44 ± 0.21 61.98 ± 0.48
(Tonon, Brabet, & Hubinger, 2008). These results are in agreement
P5 150 1.5 90 3.33 ± 0.39 67.32 ± 2.35
with those reported by Kurozawa, Morassi, Park, and Hubinger (2009)
P6 180 1.5 90 2.74 ± 0.16 64.04 ± 0.55
and Yunus, Lee, and Idham (2011).
P7 150 2.0 90 3.42 ± 0.31 64.21 ± 1.04
P8 180 2.0 90 2.95 ± 0.37 58.99 ± 1.26

Abbreviation: Ti, inlet air temperature; Fe, feed flow rate; Em/p, percent- 3.2 | Effect of drying conditions on yield
age of encapsulants; y, yield; MC, moisture content; MC and y values
are means of three replications. In all the treatments, the powder yield was efficient (>59%; Table 1).
Bhandari et al. (1997) proposed that efficient drying recovers more

Em/p (60 and 90%) as factors and yield, moisture content, color, and than 50% of the product in relation to the fed volume. The powder

glass transition temperature as the dependent variables. yield takes into account so much of the samples gathered in the collec-

The statistical analysis of data was carried out to analyze the var- tor vessel (Figure 1B) like samples collected at the end of the process

iance (ANOVA) using the general linear models (GLM) approach of on the sweeping walls of the drying chamber.

NCSS software (Hintze, J., Kaysville, Utah, USA) to estimate the signifi- Statistically, a significant effect was found (a = 0.05) of Fe and Em/p

cance (a = 0.05) of the main effects and their interaction. as individual factors and as a combination Ti & Em/p and Fe & Em/p on

All experiments were carried out in triplicate. The results were yield (Table 3). The major yield was obtained by treatment P5, which

expressed as mean value ± standard deviation (SD). combines Ti and Fe low and Em/p high. A low level in Ti decreases the
adherence of powders on the wall of the drying chamber. This is
because there are minimum changes in structure and thermoplastic
3 | RESULTS AND DISCUSSION condition of the material (Chegini & Ghobadian, 2007; Goula & Ada-
mopoulos, 2005a). Several phenomena have been suggested to explain
3.1 | Effect of drying conditions on moisture content
the decrease of the adhesion of powders. Chegini and Ghobadian
The moisture content (MC) of the pitaya juice powders is less than (2007) suggest that higher the temperature, the droplets are drier
3.5% db (Table 1), which assures the stability of the product. Low level when they hit the wall, reducing adhesion, while Goula and Adamopou-
of moisture content decreases the rates of chemical, biochemical, and los (2005a) propose the phenomenon of melting of some components
microbial reactions. of the powder and the cohesion wall.
The moisture content of powders was significantly affected The results suggest that Ti (150°C) is sufficient to dry the drop
(a = 0.05) by Ti, Fe, and Em/p (Table 3) and decrease with the increment before reaching the wall of the dryer. This process is favored with a
in Ti and/or Em/p and with the decrement in Fe. This can be attributed slow feed flow rate and a high concentration of maltodextrin. Better
to the increase in Ti, this generates a greater temperature gradient aptitude for drying has been found in encapsulants with higher melting
between the atomized feed and the drying air, resulting in a greater temperature, like maltodextrin 13DE. Low dextrose equivalents (DE)
driving force for water evaporation and thus, producing powders with maltodextrins give higher collapse temperatures than those with higher

TA BL E 2 Color parameters of pitaya juice powders

Treat Ti (ºC) Fe (L/hr) Em/p (%) L* a* b* h (º) C*

P1 150 1.5 60 39.52 ± 0.19 44.19 ± 0.34 11.31 ± 0.13 14.36 ± 0.06 45.61 ± 0.36
P2 180 1.5 60 42.35 ± 0.85 41.70 ± 1.27 20.61 ± 2.85 26.29 ± 3.84 46.60 ± 0.13
P3 150 2.0 60 39.24 ± 0.70 44.56 ± 0.17 10.24 ± 0.38 12.94 ± 0.48 45.73 ± 0.17
P4 180 2.0 60 40.79 ± 0.36 43.50 ± 0.17 15.24 ± 0.60 19.31 ± 0.77 46.10 ± 0.06
P5 150 1.5 90 41.05 ± 1.64 42.68 ± 0.49 11.49 ± 0.12 15.07 ± 0.32 44.20 ± 0.44
P6 180 1.5 90 41.35 ± 0.46 42.17 ± 0.61 14.94 ± 0.39 19.50 ± 0.21 44.74 ± 0.71
P7 150 2.0 90 39.38 ± 0.36 43.26 ± 0.19 11.28 ± 0.09 14.61 ± 0.05 44.71 ± 0.20
P8 180 2.0 90 39.70 ± 1.38 43.05 ± 0.47 12.10 ± 0.78 15.70 ± 1.12 44.72 ± 0.26

Color values are means of three replications; L* is lightness, a* is the degree of redness or greenness, b* is the degree of yellowness or blueness, C* is
indicates the amount of color, h is indicative of a hue shift.
GARCÍA-LUCAS ET AL. | 5

TA BL E 3 Sum of squares from analysis of variance of pitaya juice centage of encapsulants resulted in a minor quantity of adsorbed
powders properties water, probably because the balance of hydrophilic/hydrophobic active
Color parameters sites is modified (Gabas et al., 2007). The parameters obtained for
Source of
variation MC Yield La
Hue C a model of GAB is showed in Table 4. The model presented a good fit to

Ti 2.1745a 2.9843 18.8627a 425.8421a 2.7552a the experimental data (R2 > 0.99). The value of monolayer (Xm) indicates

Fe 0.022a 16.3823a 19.9821a 120.1118a 0.0080 the water quantity that is strongly adsorbed to specific sites in surface
and be linked to the feed stability (Gabas et al., 2007). A minor avail-
Ti3Fe 0.0034 2.1389 1.1876 59.6525a 0.9747a
ability of active sites for water adsorption was observed in samples
Em/p 1.0318a 29.2411a 0.1355 48.3005a 24.1117a
produced with high percentage of encapsulants. The values in C (=1)
Ti3Em/p 0.0006 45.9345a 10.5750a 122.6561a 0.4880
and K (>0.9) can mean that the water molecules of the monolayer and
Fe3Em/p 0.2252 17.0363a 1.6688 12.9065a 0.5808a
multilayer do not have great difference and that the properties of the
Ti3Fe3Em/p 0.0165 0.2328 1.2904 3.7130a 0.0070
water molecules of the multilayer are similar to those of the liquid
Significant term at a = 0.05.
a
water (Quirijns, Van Boxtel, Van Loon, & Van Traten, 2005b). The val-
ues in SA (Table 4) suggest a major available area for hydrolytic bonds
DE maltodextrins, usually phenomena related to viscous flow of the and are major than reported for food products (100–250 m2/g solid;
dried powder or stickiness. Cassini et al., 2006) probably for the presence of high molecular weight
Additionally, when Fe is in low level, a scanty quantity of fluid is compounds.
atomized in the drying chamber, increasing the transfer of mass and
heat, proving an efficient drying (Chegini and Ghobadian, 2007). A high 3.5 | Glass transition temperature
level in the proportion of encapsulants increases solids substances in
Table 5 shows the glass transition temperature (Tg) of powders equili-
the feeding and improves the yield. A similar behavior was observed by
brated to different aw. The Tg decreases with an increase in aw due to

n-Martínez, Me
Papadakis, Gardeli, and Tzia (2006), Leo
ndez-Lagunas,
the plasticizing effect of the water. Similar tendency was also reported
€
and Rodríguez-Ramírez (2010) and Şahin, Torun, and Ozdemir (2011)
in previous studies (Goula & Adamopoulos, 2010; Goula, Karapantsios,
in spray drying materials.

rdoba, &
Achilias, & Adamopoulos, 2008; Mosquera, Moraga, De Co
Martínez-Navarrete, 2011). Powders with 90% of encapsulants and
3.3 | Effect of drying conditions on color
aw = 0.11 presented a value in Tg major than those produced with only
The color parameters of powders are showed in Table 2. L* was a 60%. The increment of molecular weight increases the Tg of the final
affected significantly (a = 0.05) by Ti, Fe and by combination Ti & Em/p product. Roos and Karel (1991) reported that the values in Tg of a
while h was affected significantly by Ti, Fe, Em/p and by combinations Ti binary system (maltodextrin/sucrose) increase with an increase in the
& Fe, Ti & Em/p, Fe & Em/p and Ti & Fe & Em/p. In C* the main effects Ti, concentration of maltodextrin, nevertheless, the increment is not linear
Em/p and combination Ti & Fe and Fe & Em/p showed significant effects and depends widely on the concentration of every component. Gener-
(a = 0.05) (Table 3). The value of L*, h, and C* increased with high Ti, ally, the Tg of the powders obtained in this study is lower than those
low Fe and Em/p, possibly because on these conditions of drying, there obtained by Lee, Wu, and Siow (2013) and Tze et al. (2012) in red pit-
was a major heat transfer and a minor protection of the encapsulants, aya juice powders (Hylocereus polyrhizus). The differences in variety
favoring the reactions of degradation in the betalains. These operation species, composition of juices and the percentage of encapsulants used
conditions result in a less red powder and a clearer color. The degrada- explain the dissimilarity of the properties. Experimental data of Tg and
tion reactions of betalains by thermal effect as decarboxylation, hydro-
lysis and dehydrogenation, produce a reduction of color (Herbach,
Stintzing, & Carle, 2004; Herbach, Stintzing, & Carle, 2006). The sam-
ples P3 and P7 showed the lower value of h indicating they are redder
and that operation conditions are more favorable to avoid the degrada-
tion of betalains.

3.4 | Adsorption isotherms

Adsorption isotherms of powders are showed in Figure 2. All the iso-
therms present a behavior type III.
This type of curve was also observed at powder of pineapple, kiwi,
acai and opuntia mucilage (Gabas, Telis, Sobral, & Telis-Romero, 2007;

n-Martínez et al., 2010; Tonon et al., 2009; Wang, Zhang, & Chen,
Leo
2008). Powders with 90% of encapsulants (P5–P8) are less hygroscopic
than powders produced with a 60% (P1–P4). An increase of the per- FIGURE 2 Adsorption isotherms of pitaya juice powders
6 | GARCÍA-LUCAS ET AL.

TA BL E 4 Estimated GAB parameters for pitaya juice powders

Treatment
Parameters P1 P2 P3 P4 P5 P6 P7 P8
Xm (g H2O/g d.m.) 0.1558 0.1514 0.1583 0.1548 0.1497 0.1416 0.1459 0.1406
C 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000
K 0.9253 0.9447 0.9280 0.9225 0.9193 0.9274 0.9257 0.9321
SA (m2/g solid) 550.57 534.78 559.25 546.95 528.94 500.38 515.59 496.93
R2 0.9975 0.9953 0.9974 0.9969 0.9969 0.9964 0.9967 0.9969
2
Xm is the monolayer moisture content; C and K are constants of GAB equation; SA is the solid surface area; R is the coefficient of determination.

xw (g H2O/g product) were adjusted satisfactorily (R2 > 0.95) to the are hollow with a thick wall. According to Nijdam and Langrish (2006)
Gordon-Taylor model. The estimated parameters appear in Table 6 and the formation of hollow particles with smooth surface is carried out to
the curves fitted in the Figure 3. The parameter k was founded high temperatures of drying. This behavior has been attributed to the
between 2.73 and 4.26. Similar values were obtained in juice powders formation of a vacuole (a vapor bubble) and by a positive pressure cre-
of acai, grapefruit and borojo (Mosquera et al., 2011; Telis & Martínez- ated inside the particle. When the drying temperature is sufficiently
Navarrete, 2010; Tonon et al., 2009). This parameter controls the high, moisture is evaporated quickly and the skin becomes dry and
degree of curvature of Tg dependence on water content (in a binary hard, causing that the hollow particle cannot deflate when vapor con-
system) and can be related to the strength of the interaction between denses within the vacuole, while the particle moves into cooler regions
the system components (Gordon & Taylor, 1952). Due to an increase of the dryer. On the other hand, the agglomeration or join of individual
of free volume between molecules of polymer structure, as a direct particles is associated to electrostatic effects and forces of Van der
consequence, there is an increase of moisture content in the Tg Waals and can be reduced by modifying drying conditions (Walton &
increment. Mumford, 1999). The formation of hollow particles is frequent in mate-
As the moisture content increases the Tg increment, causing an rials, which may gelatinize, as those containing polysaccharides of high
increase of free volume between the molecules of the polymer struc- molecular weight and starch (Sloth et al., 2009). The pitaya juice con-
ture. The water molecules penetrate between the intermolecular tains mucilages and sweeteners that they have still not been identified,
spaces of the polymeric chains of the material, increasing the average nevertheless, studies on the composition of mucilage cells from Opun-
distance between these chains which then favors the molecular mobil- tia ficus indica (a member of the cactaceae family) revealed that the
ity and changes the material from an amorphous to a rubbery state mucilage is a high molecular weight polysaccharide that contains a
(Abiad, Carvajal, & Campanella, 2009). molecular structure of up to 30,000 different sugars (McGarvie &
Parolis, 1981). A high concentration of maltodextrin influences signifi-
cantly the particle diameter experimenting considerable inflation.
3.6 | Particle morphology
Flow properties (e.g., viscosity) are affected by particle shape and
Figure 4 shows the powders morphology of the treatments P3 and P7. morphology, as well as by the agglomeration degree. Walton and
These particles present a spherical shape, a smooth surface, an irregular Mumford (1999) suggest that spherical particles flow freely, as they
distribution of particle size and a light agglomeration. Some particles lack surface roughness and do not form agglomerates. In such case, the

TA BL E 6 Estimated Gordon-Taylor parameters for pitaya juice

TA BL E 5 Glass transition temperature (Tg) of powders conditioned powders
at different aw
Parameters
Water activity (aw) Treatment Tgs (ºC) k R2
Treatment 0.11 0.33 0.43 0.53
P1 34.33 3.28 0.989
P1 32.24 ± 0.05 3.19 ± 0.33 213.60 ± 0.45 220.80 ± 1.01
P2 29.89 2.73 0.994
P2 29.22 ± 0.24 4.14 ± 1.07 211.28 ± 0.71 219.24 ± 0.55
P3 33.59 3.62 0.997
P3 31.53 ± 1.32 22.39 ± 0.39 216.33 ± 0.28 226.22 ± 0.54
P4 35.74 3.53 0.994
P4 32.68 ± 0.03 3.93 ± 1.17 212.45 ± 0.78 225.59 ± 0.28
P5 39.57 3.66 0.992
P5 38.26 ± 0.36 1.85 ± 0.96 28.92 ± 0.71 219.67 ± 1.50
P6 41.28 3.69 0.964
P6 41.23 ± 1.04 1.03 ± 0.67 26.87 ± 1.22 215.37 ± 0.17
P7 37.01 3.90 0.995
P7 N/Aa 2.62 ± 0.67 213.54 ± 0.72 225.28 ± 0.76
P8 44.18 4.26 0.995
P8 42.22 ± 0.52 3.57 ± 1.27 29.79 ± 1.15 221.80 ± 1.17
Tgs is the glass transition temperature of solids; k is a constant of Gor-
a
Not available. Tg values are means of three replications. don and Taylor equation; R2 is the coefficient of determination.
GARCÍA-LUCAS ET AL. | 7

TA BL E 7 Water activity (awc) and water content critical (Xc)

Treatment awc Xc (g H2O/gdm)

P1 0.189 0.033
P2 0.192 0.034
P3 0.178 0.031
P4 0.201 0.035
P5 0.220 0.038
P6 0.241 0.041
P7 0.211 0.036
P8 0.240 0.041

ture above the Tg can increase (Maltini, Torreggiani, Venir, & Bertolo,
2003).
FIGURE 3 Tg vs xw of pitaya powders
This also results in significant changes in the spray drying behavior
and storage stability. The glass transition temperature of solids, like
particle sphericity and minimal formation of agglomerates will reduce
skim milk, show the stickiness and caking zone at about 10°C or higher

ndez Lagu-
resistance to the flow (García-Cruz, Rodríguez-Ramírez, Me
above the Tg (Roos, 2002). It must then maintain the product at a tem-
nas, & Medina-Torres, 2013).
perature below Tg, in a glassy state to avoid change of state during
storage. Changes in the relative humidity of the air in contact with
3.7 | Storage stability
powders induce changes in water content and as a result, Tg also, lead-
The stability of spray-dried powders during storage is strongly associ- ing the material to a very viscous glass state or to more liquid-like “rub-
ated with moisture content, water activity and glass transition tempera- bery” amorphous structure. This change occurs at a critical water
ture (Tg) of the material (Roos & Karel, 1991). Food is a multi-domain content/water activity that is the value when the glass transition tem-
and multi-component system. The range of glass transition tempera- perature of the product is equal to the room temperature. Water con-
tures of agro-industrial products depends mainly on the composition tent and moisture content values are essential to evaluate
(Recondo, Elizalde, & Buera, 2006). specifications for storage and packing of food powders (Roos, 1995).
The fruit juice powders are composed mainly of sugars and organic Table 7 shows critical values to a storage temperature of 25°C. Below
acids and may contain soluble hydrocolloids and suspended particles. these conditions the powders can be stored maintaining their stability.
This is why juice powders have a Tg in a temperature range; and even Above this critical point, powders become sticky and the crystallization
the same product can have numerous state transitions (Hui, 2006; of amorphous compounds could take place. Structure powder under-
Quiroz, 2007). goes: stickiness, collapse, crispness, caking and agglomeration (Kuro-
Below Tg, compounds such as sugars and organic acids, are struc- zawa et al., 2009; Moraga & Martínez-Navarrete, 2006; Roos, 1995;
turally stable in the metastable solid “glassy” state and critical tempera- Tonon et al., 2009).

F I G U R E 4 SEM images of the pitaya juice powder particles for: (A) P3:150°C, 2.0 L/hr, 60% of encapsulants and (B) P7: 150°C, 2.0 L/hr,
90 % of encapsulants.
8 | GARCÍA-LUCAS ET AL.

4 | CONCLUSION Fazaeli, M., Emam-Djomeh, Z., Ashtari, A. K., & Omid, M. (2012). Effect
of spray drying conditions and feed composition on the physical
properties of black mulberry juice powder. Food and Bioproducts Proc-
The drying parameters had a significant effect on the yield, moisture
essing, 90, 667–675.
content and color of the powders, overall, pitaya juice powders with
Ferrari, C. C., Germer, S. P. M., & De Aguirre, J. M. (2012). Effects of
the treatment P5 (150°C, 1.5 L/hr, 90% of encapsulants) and P7 (150° spray-drying conditions on the physicochemical properties of black-
C, 2.0 L/hr, 90% of encapsulants) have better physical properties; red- berry powder. Drying Technology, 30, 154–163.
violet color retention, high yield, high glass transition temperature and Filkova
, I., Huang, L. X., & Mujumdar, A. S. (2006). Industrial spray drying
low water adsorption. A high proportion of encapsulants increases the systems. In A. S. Mujumdar (Ed.), Handbook of industrial drying (pp.
215–255). Boca Rato
n, FL: CRC Press - Taylor & Francis Group.
Tg and decreases lightly the aw and superficial area. The microstructural
Gabas, A. L., Telis, V. R. N., Sobral, P. J. A., & Telis-Romero, J. (2007).
analyses showed that the use of maltodextrin/pectin as an encapsulat-
Effect of maltodextrin and arabic gum in water vapor sorption ther-
ing mixture results in homogenous, well-formed particles; which are modynamic properties of vacuum dried pineapple pulp powder. Jour-
recommended in spray drying microencapsulation products. The pow- nal of Food Engineering, 82, 246–252.
ders analyzed in this research represent an interesting food ingredient, García-Cruz, E. E., Rodríguez-Ramírez, J., Me
ndez Lagunas, L. L., &
which can be incorporated into functional foods as a colorant. Medina-Torres, L. (2013). Rheological and physical properties of
spray-dried mucilage obtained from Hylocereus undatus cladodes. Car-
bohydrate Polymers, 91(1), 394–402.
ACKNOWLEDG MENTS Georgiev, V. G., Weber, J., Kneschke, E. M., Denev, P. N., Bley, T., &
Pavlov, A. I. (2010). Antioxidant activity and phenolic content of

rrez, from Centro de
The authors thank PhD Hugo Martínez-Gutie
betalain extracts from intact plants and hairy root cultures of the red
Nanociencias y Micro y Nanotecnologías (CNMN-IPN) for his assis- beetroot Beta vulgaris cv. Detroit Dark Red. Plant Foods for Human
tance in SEM images. Nutrition, 65, 105–111.
Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Saurel, R. (2007).
Applications of spray-drying in microencapsulation of food ingredients:
CONFLIC T OF I NTE R ES T An overview. Food Research International, 40, 1107–1121.
The authors declare no conflict of interest. Gordon, M., & Taylor, J. S. (1952). Ideal copolymers and the second-
order transitions of synthetic rubbers. I. Non-crystalline copolymers.
Journal of Applied Chemistry, 22, 493–500.
RE FE RE NCE S
Goula, A. M., & Adamopoulos, K. G. (2005a). Spray drying of tomato
A.O.A.C. (1984). Official methods of analysis of the association of official ana- pulp in dehumidified air: I. The effect on product recovery. Journal of
lytical chemists. Arlington, VA: Association of Official Analytical Chemists. Food Engineering, 66, 25–34.
Abadio, F. D. B., Domingues, A. M., Borges, S. V., & Oliveira, V. M. Goula, A. M., & Adamopoulos, K. G. (2005b). Stability of lycopene during
(2004). Physical properties of powdered pineapple (Ananas comosus) spray drying of tomato pulp. LWT-Food Science and Technology, 38,
juice–Effect of malt dextrin concentration and atomization speed. 479–487.
Journal of Food Engineering, 64, 285–287.
Goula, A. M., & Adamopoulos, K. G. (2010). A new technique for spray
Abiad, M. G., Carvajal, M. T., & Campanella, O. H. (2009). A review on drying orange juice concentrate. Innovative Food Science and Emerging
methods and theories to describe the glass transition phenomenon: Technologies, 11, 342–351.
Applications in food and pharmaceutical products. Food Engineering Goula, A. M., Karapantsios, T. D., Achilias, D. S., & Adamopoulos, K.
Reviews, 1, 105–132. G. (2008). Water sorption isotherms and glass transition tempera-
Barbosa-C
anovas, G. V., & Vega-Mercado, H. (1996). Dehydration of ture of spray dried tomato pulp. Journal of Food Engineering, 85,
Foods (pp. 185–228). New York: Chapman & Hall. 73–83.
Bhandari, B. R., Datta, N., Crooks, R., Howes, T., & Rigby, S. (1997). A Grabowski, J. A., Truong, V. D., & Daubert, C. R. (2006). Spray-drying of
semi-empirical approach to optimize the quantity of drying aids amylase hydrolyzed sweetpotato puree and physicochemical proper-
required to spray dry sugar-rich foods. Drying Technology, 15, 2509– ties of powder. Journal of Food Science, 71, E209–E217.
2525. Herbach, K. M., Stintzing, F. C., & Carle, R. (2004). Impact of thermal
Boo, H. O., Hwang, S. J., Bae, C. S., Park, S. H., Heo, B. G., & Gorinstein, treatment on color and pigment pattern of red beet (Beta vulgaris L.)
S. (2012). Extraction and characterization of some natural plant pig- preparations. Journal of Food Science, 69, C491–C498.
ments. Industrial Crops and Products, 40, 129–135. Herbach, K. M., Stintzing, F. C., & Carle, R. (2006). Betalain stability and
Cai, Y., & Corke, H. (2000). Production and properties of spray-dried degradation_structural and chromatic aspects. Journal of Food Science,
amaranthus betacyanin pigments. Journal of Food Science, 65(6), 71, R41–R50.
1248–1252. Hui, Y. H. (2006). Handbook of food science, technology and engineering
~a, C. P. Z. (2006). Water adsorp-
Cassini, A. S., Marczak, L. D. F., & Noren
n, FL: CRC Press - Taylor & Francis Group.
(Vol. 1). Boca Rato
tion isotherms of texturized soy protein. Journal of Food Engineering, Johari, G. P., Hallbrucker, A., & Mayer, E. (1987). The glass–liquid transi-
77, 194–199. tion of hyperquenched water. Nature, 330, 552–553.
Chegini, G. R., & Ghobadian, B. (2007). Spray dryer parameters for fruit Kandansamy, K., & Somasundaram, P. D. (2012). Microencapsulation of
juice drying. World Journal of Agricultural Sciences, 3, 230–236. colors by spray drying-a review. International Journal of Food Engi-
Fang, Z., & Bhandari, B. (2011). Effect of spray drying and storage on neering, 8, 1–15.
the stability of bayberry polyphenols. Food Chemistry, 129, 1139– Kha, T. C., Nguyen, M. H., & Roach, P. D. (2010). Effects of spray drying
1147. conditions on the physicochemical and antioxidant properties of the
GARCÍA-LUCAS ET AL. | 9

Gac (Momordica cochinchinensis) fruit aril powder. Journal of Food Ravichandran, K., Palaniraj, R., Saw, N. M., Gabr, A. M. M., Ahmed, A. R.,
Engineering, 98, 385–392. Knorr, D., & Smetanska, I. (2014). Effects of different encapsulation
Kurozawa, L. E., Morassi, A. G., Park, K. J., & Hubinger, M. D. (2009). agents and drying process on stability of betalains extract. Journal of
Spray drying of protein hydrolysate of chicken breast meat. In Fourth Food Science and Technology, 51, 2216–2221.
InterAmerican Drying Conference Montreal, Canada. Recondo, M. P., Elizalde, B. E., & Buera, M. P. (2006). Modeling tempera-
Lee, K. H., Wu, T. Y., & Siow, L. F. (2013). Spray drying of red (Hylocer- ture dependence of honey viscosity and of related supersaturated
eus polyrhizus) and white (Hylocereus undatus) dragon fruit juices: model carbohydrate systems. Journal of Food Engineering, 77, 126–
Physicochemical and antioxidant properties of the powder. Interna- 134.
tional Journal of Food Science & Technology, 48, 2391–2399. Roos, Y. (1995). Water activity and glass transition temperature: How do

n-Martínez, F. M., Me
Leo
ndez-Lagunas, L. L., & Rodríguez-Ramírez, J. they complement and how do they differ. In G. Barbosa-C
anovas
(2010). Spray drying of nopal mucilage (Opuntia ficus-indica): Effects (Ed.), Food preservation by moisture control: Fundamentals and applica-
on powder properties and characterization. Carbohydrate Polymers, tions (pp. 29–46). Lancaster, PA: Technomic Publishers.
81, 864–870. Roos, Y., & Karel, M. (1991). Phase transitions of mixtures of amorphous
Maltini, E., Torreggiani, D., Venir, E., & Bertolo, G. (2003). Water activity polysaccharides and sugars. Biotechnology Progress, 7, 49–53.
and preservation of plant foods. Food Chemistry, 82, 79–86. Roos, Y. H. (2002). Importance of glass transition and water activity to
Masters, K. (1991). Spray drying handbook (5th ed.). New York: Longman spray drying and stability of dairy powders. Le Lait, 82, 475–484.
Scientific & Technical. €
Şahin-Nadeem, H., Torun, M., & Ozdemir, F. (2011). Spray drying of the
Mcgarvie, D., & Parolis, H. (1981). The acid-labile, peripheral chains of mountain tea (Sideritis stricta) water extract by using different hydro-
the mucilage of Opuntia ficus-indica. Carbohydrate Research, 94, 57– colloid carriers. Food Science and Technology, 44, 1626–1635.
65. Sant’anna, V., Gurak, P. D., Marczak, L. D. F., & Tessaro, I. C. (2013).
Moraga, G., Martíinez-Navarrete, N., & Chiralt, A. (2004). Water sorption Tracking bioactive compounds with colour changes in foods – A
isotherms and glass transition in strawberries: Influence of pretreat- review. Dyes Pigments, 98, 601–608.
ment. Journal of Food Engineering, 62, 315–321. Shahidi, F., & Han, X. Q. (1993). Encapsulation of food ingredients. Criti-
Moraga, G. N., & Martínez-Navarrete, A. C. (2006). Water sorption iso- cal Reviews in Food Science and Nutrition, 33, 501–547.
therms and phase transitions in kiwifruit. Journal of Food Engineering, Sloth, J., Jorgensen, K., Bach, P., Jensen, A. D., Kiil, S., & Dam-Johansen,
72, 147–156. K. (2009). Spray drying of suspensions for pharma and bio products:
Mosquera, L. H., Moraga, G., De Co
rdoba, P. F., & Martínez-Navarrete, Drying kinetics and morphology. Industrial & Engineering Chemistry
N. (2011). Water content–water activity–glass transition temperature Research, 48, 3657–3664.
relationships of spray-dried borojo as related to changes in color and Telis, V. R. N., & Martínez-Navarrete, N. (2010). Application of compres-
mechanical properties. Food Biophysics, 6, 397–406. sion test in analysis of mechanical and color changes in grapefruit
Nijdam, J. J., & Langrish, T. A. G. (2006). The effect of surface composi- juice powder as related to glass transition and water activity. LWT -
tion on the functional properties of milk powders. Journal of Food Food Science and Technlogy, 43, 744–751.
Engineering, 77, 919–925. Tonon, R. V., Baroni, A. F., Brabet, C., Gilbert, O., Pallet, D., & Hubinger,
Papadakis, S. E., Gardeli, C., & Tzia, C. (2006). Spray drying of raisin juice M. D. (2009). Water sorption and glass transition temperature of
concentrate. Drying Technology, 24, 173–180. spray dried açai (Euterpe oleracea Mart.) juice. Journal of Food Engi-
Phisut, N. (2012). Spray drying technique of fruit juice powder: Some neering, 94, 215–221.
factors influencing the properties of product. International Food Tonon, R. V., Brabet, C., & Hubinger, M. D. (2008). Influence of process
Research Journal, 19, 1297–1306. conditions on the physicochemical properties of açai (Euterpe olera-
Pitalua, E., Jimenez, M., Vernon-Carter, E. J., & Beristain, C. I. (2010). Anti- ceae Mart.) powder produced by spray drying. Journal of Food Engi-
oxidative activity of microcapsules with beetroot juice using gum Ara- neering, 88, 411–418.
bic as wall material. Food and Bioproducts Processing, 88, 253–258. Tze, N. L., Han, C. P., Yusof, Y. A., Ling, C. N., Talib, R. A., Taip, F. S., &
Quek, S. Y., Chok, N. K., & Swedlund, P. (2007). The physicochemical Aziz, M. G. (2012). Physicochemical and nutritional properties of
properties of spray-dried watermelon powders. Chemical Engineering spray-dried pitaya fruit powder as natural colorant. Food Science and
and Processing, 46, 386–392. Biotechnology, 21, 675–682.
Quirijns, E. J., Van Boxtel, A. J. B., Van Loon, W. K. P., & Van, S. G. (2005a). Walton, D. E., & Mumford, C. J. (1999). Spray dried products—Character-
An improved experimental and regression methodology for sorption ization of particle morphology. Chemical Engineering Research and
isotherms. Journal of the Science of Food and Agriculture, 85, 175–185. Design, 77, 21–38.
Quirijns, E. J., Van Boxtel, A. J. B., Van Loon, W. K. P., & Van Traten, G. Wang, H., Zhang, S., & Chen, G. (2008). Glass transition and state dia-
(2005b). Sorption isotherms, GAB parameters and isosteric heat of gram for fresh and freeze-dried Chinese gooseberry. Journal of Food
sorption. Journal of the Science of Food and Agriculture, 85, 1805–1814. Engineering, 84, 307–312.
Quiroz, L. O. (2007). El Modelo Extendido de Zwanzig y la Teoría de Adam- Yunus, M. A. C., Lee, C. Y., & Idham, Z. (2011). Effects of variables on
Gibbs en el Plegamiento y Desnaturalizacio
n de Proteínas. Universidad the production of red-fleshed pitaya powder using response surface
Auto
noma Metropolitana: Me
xico. methodology. Jurnal Teknologi, 56, 15–29.