LWT - Food Science and Technology 96 (2018) 612–619

Contents lists available at ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Effect of multi-flash drying and microwave vacuum drying on the T

microstructure and texture of pumpkin slices
Ricardo Lemos Monteiroa, Jade Varaschim Linka, Giustino Tribuzib, Bruno A.M. Carciofia,
João Borges Laurindoa,∗
a
Department of Chemical and Food Engineering, Federal University of Santa Catarina, EQA/CTC/UFSC, 88040-900, Florianópolis, SC, Brazil
b
Department of Food Science and Technology, Federal University of Santa Catarina, CCA/UFSC, 88034-001, Florianópolis, SC, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: The influence of different drying methods on the microstructure, porosity, and texture of pumpkins slices was
Drying assessed, aiming to find the best process conditions. Pumpkins (Cucurbita moschata) were selected, peeled, sliced,
Microwave blanched and dehydrated by different methods: i) microwave multi-flash drying (MWMFD), ii) microwave va-
Vacuum cuum drying (MWVD), iii) conductive multi-flash drying (KMFD), iv) freeze-drying (FD), and v) air-drying (AD).
Multi-flash
Micrographs of dried samples showed the compacted structure of AD samples and the spongy structure of FD
Crispy
samples. Otherwise, large pores and a highly porous structure were observed in samples dried by KMFD, MWVD,
and MWMV. Porosity MWMFD samples were 2.6 times higher than those observed in AD pumpkins and slightly
higher than the observed in FD samples. Puncture test results from KMFD, MWVD and MWMFD samples showed
force oscillation that is typical of crispy products, which was not observed from AD and FD samples. Moreover,
KMFD, MWVD, and MWMFD samples showed low color changes. Thus, microwave under vacuum and both
multi-flash drying processes result in dried-and-crisp pumpkins, with high porosity and big pores. Therefore,
these drying processes are suitable for the production of dried-and-crisp pumpkins.

1. Introduction a preservation technology for increasing its shelf life, such as drying,
canning or freezing. In particular, pumpkin dehydration is an inter-
Pumpkin is one of the most cultivated vegetables in the world; this esting alternative to reduce postharvest losses of this fruits. It is well
product presents high content of vitamins, carotenoids, fibers, and known that the reduction of moisture and water activity of foods de-
minerals (Seremet, Botez, Nistor, Andronoiu, & Mocanu, 2016; Yang, creases microbial growth, enzymatic activity, and chemical reactions
Zhao, & Lv, 2007). Moreover, pumpkin has received considerable at- rates. Drying extend food shelf life, reduces transportation and storage
tention in recent years, especially pumpkin seed oil, because it has costs due to reduced product weight and volume, and adds value to
positive effects on the strengthening of the immune system (Bardaa vegetables, developing a new product.
et al., 2016). Pumpkin seeds protect against many diseases, such as Different drying methods can be used to produce dehydrated fruits
hypertension, carcinogenic diseases, and diabetes. Besides, these seeds and vegetables, e.g., air-drying, vacuum drying, freeze-drying, micro-
have antibacterial, antioxidant and anti-inflammatory effects (Bardaa wave drying, among others. The selection of the most appropriate
et al., 2016; Boaduo, Katerere, Eloff, & Naidoo, 2014; Hammar, Carson, process depends mainly on the food to be dehydrated and the desired
& Riley, 1999; Jian, Du, Lee, & Binns, 2005; Nawirska-Olszańska, Kita, characteristics for the final product.
Biesiada, Sokół-Łętowska, & Kucharska, 2013; Procida, Stancher, Freeze-drying is a suitable dehydration technique for sensible foods
Cateni, & Zacchigna, 2013; Rabrenovic, Dimic, Novakovic, Tesevic, & because it preserves the sensorial, nutritional and structural char-
Basic, 2014; Zuhair, Abd El-Fattah, & El-Sayed, 2000). Pumpkin seed oil acteristics of foods (Ibarz & Barbosa-Canovas, 2003, pp. 604–613; Ratti,
has high added value and has been produced in high scale in many 2008). However, the high equipment costs, energy needs, and the long
countries, due to its high demand and low cost of the pumpkin fruit. A processing times associated to freeze-drying limits its use for product
large amount of pumpkin pulp represents an interesting opportunity for with high added value (Louka & Allaf, 2002).
the development of novel products to increase industrial profitability. The development of novel drying processes that result in dried
Pumpkin pulp presents high moisture content (> 80 g/100 g) and needs products with desirable properties has been increasing in recent years

∗
Corresponding author.
E-mail address: [email protected] (J.B. Laurindo).

https://doi.org/10.1016/j.lwt.2018.06.023
Received 1 February 2018; Received in revised form 17 May 2018; Accepted 10 June 2018
0023-6438/ © 2018 Elsevier Ltd. All rights reserved.
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

Fig. 1. a) Patterns of temperature (black) and pressure (grey) variation during MWMFD; b) Pattern of power variation during MWMFD. Magnifications were inserted
for better visualization of the beginning of the drying process.

(Jangam, Law, & Mujumdar, 2010). A brief discussion of two of them Changes in the physical food properties (apparent density, porosity,
are presented in the follow. texture, color) during drying are extremely important because they
Microwave drying under vacuum has been considered a suitable influence product quality and consumer opinion (Madiouli, Sghaier,
process to produce crispy products with high nutritional and sensory Lecomte, & Sammouda, 2012).
quality. The costs of the equipments are affordable and the drying times The objective of this study was to assess the porosity, apparent
are much shorter than the observed for freeze-drying (Drouzas & density, texture, color, and microstructure of pumpkins dried by multi-
Schubert, 1996; Gunasekaran, 1999; Krokida & Maroulis, 1999; Lin, flash drying processes, using microwaves (MWMFD) or heat conduction
Durance, & Scaman, 1998; Mousa & Farid, 2002; Zhang, Tang, (KMFD) during the fruits heating step. The physical properties of the
Mujumdar, & Wang, 2006). dried pumpkins are compared with the properties of pumpkin slices
A class of drying methods named multi-flash drying (MFD) has been dried by hot air (AD), freeze-drying (FD) and microwave vacuum
reported as suitable for the production of dried-and-crisp fruits and drying (MWVD).
vegetables (Link, Tribuzi, & Laurindo, 2017a, 2017b; Monteiro,
Carciofi, & Laurindo, 2016; Porciuncula, Segura, & Laurindo, 2016;
Zotarelli, Porciuncula, & Laurindo, 2012). The method is based on the 2. Materials and methods
application of successive vacuum-pulse cycles to a material inside a
vacuum chamber. The product is heated at atmospheric pressure using 2.1. Pumpkins drying processes
heated plates, hot air, or microwaves until a desired temperature. Then,
a sudden pressure reduction (vacuum pulse) is applied, resulting in Fresh pumpkins (Cucurbita moschata var. Menina Brasileira) sam-
rapid water evaporation (flash evaporation) and product cooling ples were washed and peeled. Afterwards, their cylindrical part
(adiabatic process, because the product provides the latent heat of va- (“pumpkin neck”) were cut into slices (half-moon shape with
porization), causing both dehydration and texturization. 5.9 ± 0.9 mm thickness, 30.0 ± 3.3 mm radius and 60.02 ± 5.0 mm
diameter), blanched (97 °C ± 2 °C - 3 min), and cooled

613
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

(11 °C ± 3 °C–2 min) in distilled water. The resulting samples were 2.1.4. Freeze-drying
dried by the different drying processes (as described below). For all the Freeze-drying (FD) of pumpkin samples was performed using a la-
experiments, moisture content and water activity were determined for boratory freeze-dryer (Liotop, Model - L101, Brazil), adapted to record
fresh and dried samples. Samples moistures were determined in a va- the temporal evolution of the sample mass during the drying process.
cuum oven at 70 °C (TECNAL, Model TE-395, Brazil) (A.O.A.C., 2005). The details of the adapted freeze-drier was described by Tribuzi and
The water activity (aw) of each sample was determined with a hygro- Laurindo (2014). The experiments were performed with approximately
meter (Aqualab, Model - Series 3, Decagon Devices Inc. Pullman, USA). 90 g of pumpkin samples, which were frozen for approximately 3 h at
All the drying experiments were performed in triplicate. Temperature −60 °C ± 1 °C and immediately freeze-dried at pressure of 0.02 mbar.
was measured in the center of the samples by T-type thermocouples,
while the chamber pressure was measured by a transducer (Warme, 2.1.5. Air-drying
Model - WTP4010, Brazil). Details of the drying experiments are de- Air-drying (AD) of pumpkin samples was performed in a convective
scribed by Monteiro, Link, Tribuzi, Carciofi, and Laurindo (2018). oven (TECNAL, Model - TE 394/2, Brazil), adapted with the same
weighing system used in the freeze dryer to determine the drying
2.1.1. Microwave multi-flash drying curves. The drying air was at 60 °C, relative humidity of approximately
Microwave multi-flash drying (MWMFD) was performed in a do- 22% (Thermohygrometer, Model - Testo 610, Germany), and velocity of
mestic microwave oven (Electrolux, Model - MEX55, Brazil) adapted to approximately 0.6 m/s (portable anemometer, Model - Text 425,
work with vacuum and its rotary system that allows the vacuum Germany). Approximately 100 g of sample was used for each experi-
chamber rotate with the turntable. The chamber rotation results in a ment.
more homogeneous absorption of microwaves by the food during
drying (Monteiro, Carciofi, Marsaioli, & Laurindo, 2015; Monteiro 2.2. Characterization of pumpkin samples
et al., 2016). Approximately 100 g of fresh samples were disposed on a
circular polypropylene tray inserted into the vacuum chamber. Initially, The samples dehydrated by the different processes were character-
under atmospheric pressure, the pumpkins samples were heated at ized to assess the influence of each drying process on their physical
1000 W power to approximately 60 °C. Then, the microwave generator proprieties, i.e., apparent density, porosity, microstructure, texture and
was turned off, and the chamber was suddenly decompressed (vacuum color.
pulse) until 5 kPa (duration of 11 s). Then, the atmospheric pressure
was reestablished (duration of 3 s) to start a new cycle of heating-va- 2.2.1. Apparent density
cuum-pulse. This heating-vacuum pulse cycle was repeated three times The apparent density (ρap) of pumpkin was determined by the ratio
(the heating time steps were 27 s, 23 s and 15 s). After the application of of the sample mass (ms) and its apparent volume (Vap) (Equation (1)):
these cycles of heating-vacuum pulse, the samples were kept under
ms
vacuum (3 kPa - 5 kPa), with the application of the microwave field ρap =
until the end of drying process. Temperature, pressure and power pat- Vap (1)
terns during the MWMFD process are sketched in Fig. 1. Three different The Vap was determined indirectly by the measurement of the
power levels (1000 W - 7 min, 300 W - 10 min and 200 W - 60 min), buoyant force in n-heptane (Lozano, Rotstein, & Urbicain, 1980). The
were used during the microwave vacuum drying step, targeting high measurement of the buoyant force was performed by hanging a metallic
drying rates, but avoiding samples overheating. This power modulation support with the sample in the bottom of an electronic balance (Marte,
was determined by preliminary tests (data not shown). The evolution of Model AY220, Philippines).
the sample weight during drying was determined from a destructive
approach, in which a new fresh sample was used for determining each
2.2.2. Porosity
experimental point. It was done to avoid the increase of the power
Samples porosity (ε) was calculated from Vap and true volume (Vt),
density during drying, because of the decrease of the sample weight
using Equation (2):
after each withdraw.
V
ε = ⎜⎛1− t ⎟⎞ .100
2.1.2. Microwave vacuum drying Vap ⎠ (2)
⎝
Microwave vacuum drying (MWVD) of pumpkin samples was per-
formed by applying a pressure of 3–5 kPa and continuous microwave The Vt of the samples was determined using an air pycnometer, as
heating (1000 W for 7 min, 300 W for 10 min and 200 W for 60 min). proposed by Sereno, Silva, and Mayor (2007). This pycnometer mea-
sure the pressure change of compressed air filling a constant volume
2.1.3. Conductive multi-flash drying chamber when it expands into a second constant volume chamber
Conductive multi-flash drying (KMFD) process was performed using containing the solid sample. From such pressure change and the two
a 100 L drying chamber (Ethik Technology, Model-440-DE, Brazil) chamber volumes, the true volume of the sample is determined.
connected to a vacuum pump (DVP, Vacuum Technology, Model -
LC.305, Italy), as described by Porciuncula et al. (2016). Inside the 2.2.3. Mechanical properties
drying chamber, pumpkin samples were placed on plates heated by Mechanical resistance of pumpkin samples was determined from
electrical resistances connected to a PID (proportional–integral–der- penetration tests, and considered as an instrumental evaluation of their
ivative) controller, which kept the plates temperature around 90 °C. texture. The mechanical essays were performed with a texturometer
Approximately 1000 g of pumpkin samples were used in each drying (Stable Micro System, Model - TA-XT2, UK), using a cylindrical probe
experiment. The samples were initially heated up to 60 °C, at atmo- (2 mm diameter), with strain of 70% of the original sample thickness.
spheric pressure. After the heating step, a vacuum pulse (P ≈ 5 kPa) The essays were performed in quintuplicate, with perforations in three
was applied to the chamber. Five minutes after the vacuum pulse ap- different regions each sample. The force oscillations during the pene-
plication, the atmospheric pressure was reestablished in the chamber tration test was related to potentially sample crispness (Laurindo &
and a new vacuum pulse-heating cycle was applied. Three heating-va- Peleg, 2007, 2008). Strain-force data were smoothed by a non-para-
cuum-pulse cycles were applied, followed by vacuum drying (184 min metric fit based on a second-order polynomial approximation (Savitzky
and pressure of 3–5 kPa) of the samples until the end of the experiment. & Golay, 1964). The normalized residuals were calculated from the
More details of the equipment and the process can be found in Link difference between the recorded values and those calculated from the
et al. (2017a). non-parametric fit (Laurindo & Peleg, 2007, 2008). The strain-force

614
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

curves, the smoothed curves, and the residues curve, the number of 3.2. Scanning electron microscopy and mechanical properties of dried
peaks (irregular), as well as their amplitude, can be calculated and pumpkins
correlated to the crispness of samples dried by different methods
(Monteiro et al., 2016). Appearance and texture of foods are the attributes that most influ-
Data processing was performed with Matlab® software (Math Works ence consumer choice. Therefore, scanning electron microscopy and
Inc. Model - R2013a, USA), using the “sgolayfilt” function to apply the mechanical texture determination of dried pumpkins are suitable tools
Savitzky and Golay smoothing routine. As the smoothing procedure was to evaluate these attributes. Fig. 3 shows scanning electron microscopy
performed to take the overall shape of the force-strain curve, the re- (SEM) images of manual fractures of pumpkins dried by AD, FD, KMFD,
sidual pattern does not depend on the quality of the fit (Laurindo & MWMFD and MWVD, with magnification of 30× and 100×. Samples
Peleg, 2008). A second-order polynomial function and a sliding frame dehydrated by KMFD, MWMFD and MWVD presented porous structures
of 149 experimental points were chosen in the Matlab “sgolayfilt” with large and irregular pores. On the other hand, pumpkins dried by
function, for smoothing all the force-strain curves. AD presented important shrinkage that formed a compact structure,
while FD samples showed preserved cellular integrity, with uniform
small pores, forming a spongy structure.
2.2.4. Scanning electron microscopy From SEM images is evident that multi-flash drying (MFD) pumpkin
Dried samples were manually fractured and placed on aluminum samples showed a higher expansion of the porous matrix. The heating
supports with the aid of double-sided carbon tape and coated with gold vacuum-pulse cycles caused important changes in the structure of fruits
in a metallizer (LEICA EM, Model - SCD500, Germany). Images were and vegetables, because of the fast vapor expansion during the vacuum
captured (magnification of 30× and 100×) using a scanning electron pulse (Monteiro et al., 2016). Flash evaporation results in forces that
microscopy (SEM; JEOL, model - JSM 6390LV, Japan). cause expansion or rupture of the solid matrix, whose intensity depends
on the operational conditions, such as sample temperature (before the
vacuum pulse), pressure (initial and final) and decompression rate
2.2.5. Color analyses
(Link et al., 2017a; Louka & Allaf, 2002; Monteiro et al., 2016;
A Minolta Chroma Meter handheld colorimeter (Konica Minolta,
Porciuncula et al., 2016; Zotarelli et al., 2012). In MWVD and MWMFD
Model - CR-400, Japan), adjusted to operate with a D65 lighting and
dried samples, the porous structure is highly expanded due to the vo-
10° observation angle, was used to determine the color parameters of
lumetric heating caused by microwave heating and low pressure, which
fresh, blanched and dried samples. The color was measured on the
forces the vapor from the inside to the surface of the pumpkin, causing
samples surface and was expressed in the CIELAB system, L* (black/
this expansion (Monteiro et al., 2015, 2016; Zhang et al., 2006).
white), a* (red/green) and b* (yellow/blue), as defined by CIE
Fig. 4 shows the force-strain curves resulted from the puncture tests
(Commission Internationale de L'Èclairage). The total difference of color
performed with samples dried by AD, FD, KMFD, MWMFD and MWVD
was evaluated by the parameter ΔE*, using as reference the blanched
processes. The crispness of dried products is related to the irregularity
pumpkins (L0, a0 and b0), calculated according to Equation (3):
of the force-strain curves that is associated to the number and intensity
of the force peaks (Laurindo & Peleg, 2007, 2008; Monteiro et al.,
ΔE* = (L*−L 0)2 + (a *−a 0)2 + (b*− b0)2 (3) 2016). The force-strain curves for KMFD, MWMFD and MWVD samples
showed this irregular pattern. AD samples presented the typical beha-
vior of compact and rigid structure, represented by a continuous in-
2.2.6. Statistical analyses crease of the penetration force during the probe penetration. At strain
The results of moisture content, water activity, apparent density, of approximately 60% a rupture was observed, characterized by an
porosity, and color were analyzed statistically with Statistica 7.0 pro- abrupt drop in the penetration force. FD samples presented curves
gram (StatSoft, Tulsa, USA), using analysis of variance (ANOVA) and characteristic of spongy products (porous sample with small and uni-
Tukey test at 95% confidence level (p ≤ 0.05). form pores), with irregular and small force oscillations.
Experimental force-strain curves of each penetration test were
smoothed and the normalized residuals were calculated by the differ-
3. Results and discussion ence between the recorded force and the force calculated from the
smoothed curve (Fig. 5). As the smoothed curves represent the com-
3.1. Drying curves prehensive behavior of the strain-force (Gonzalez Martinez, Corradini,
& Peleg, 2003; Laurindo & Peleg, 2007, 2008), the residues represent
Pumpkin samples used in this study had initial moisture of the local force oscillation, resulted from successive structure breaks.
6.136 ± 1.269 g/g (dry basis, db) and water activity of 0.994 ± 0.003 The number of force peaks in the residues curve, and their amplitude,
(mean ± standard deviation). After the blanching process, the can be correlated to the crispness of the samples dried by each drying
moisture content increased to 6.931 ± 0.464 g/g (db), while the aw process. From all these mechanical essays, it was observed that
was 0.994 ± 0.003. MWMFD process resulted in dried pumpkins with a more fragile and
Fig. 2 presents representative drying curves of pumpkin samples crispier structure, with higher number of peaks and higher amplitude of
dried by the different methods, while Table 1 shows the mean values of these peaks, followed by MWVD and KMFD samples. These results
the final moisture content (Xdbfinal), aw and the drying time (tf) of confirm those reported by reported by Monteiro et al. (2016), about the
pumpkins dried by AD, FD, KMFD, MWMFD and MWVD. Dried pump- drying of banana slices.
kins showed Xdbfinal (db) lower than 0.022 ± 0.010 g/g and aw lower
than 0.438 ± 0.021. The differences among the drying times shows the 3.3. Apparent density and porosity
effectiveness of microwave heating in the drying process studied. The
processes that use vacuum and microwave as the heating method The mean values of the apparent density (ρap) and the porosity (ε) of
(MWMFD and MWVD) presented lower drying times than the observed fresh, blanched, and dried pumpkin samples are given in Table 2. The
for AD, FD, and KMFD. The same behavior was reported by Monteiro results show that these physical properties are influenced by the drying
et al. (2016) for the drying of banana slices. The drying times of method, as already reported by Krokida and Maroulis (1997), for ba-
pumpkin slices with the MWMFD and MWVD processes were approxi- nana, apple, carrot and potato submitted to air drying, vacuum drying,
mately 13 times less than those observed for AD and FD, and 3 times freeze-drying, microwave drying, and osmotic dehydration.
less when compared to the KMFD process. Fresh pumpkins presented apparent density of 1.0104 g/cm and

615
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

Fig. 2. Temporal evolution of the moisture (dry

basis) of the pumpkin samples during the dif-
ferent drying processes. AD (‒‒‒), FD (‒‒‒),
KMFD (●), MWMFD ( × ) and MWVD (▲). Solid
lines were used to ease the processes compar-
ison. Magnifications of MWMFD, MWVD and
KMFD drying curves were inserted for better vi-
sualization.

Table 1 conditions (relative humidity, temperature, and process time), as well

Drying times (tf), mean moisture content ± standard deviation (Xdbfinal) and as the moisture content at the end of the drying process influence the
water activities ± standard deviation (aw) of dried pumpkins. shrinking and apparent density of dried products (Garcia et al., 2007;
Drying condition Xdbfinal (g/g, db) aw tf (h) Krokida & Maroulis, 1997).
Samples dried by MFD and MWVD also suffer matrix shrinkage, as
a bc
MWMFD 0.021 ± 0.006 0.360 ± 0.015 1.32 shown in Fig. 3, but it is partially compensated by the forces related to
MWVD 0.022 ± 0.010a 0.342 ± 0.049b 1.28
the vapor expansion inside the product during the vacuum pulses,
KMFD 0.014 ± 0.002a 0.385 ± 0.010c 3.50
FD 0.017 ± 0.005a 0.145 ± 0.026a 16.67 which tends to expand the porous matrix (Monteiro et al., 2016; Segura,
AD 0.021 ± 0.005a 0.438 ± 0.021d 16.33 Badillo, & Alves-Filho, 2014).
MWMFD samples presented porosity 2.6 and 1.6 times higher than
Different letters represent a significant difference (p < 0.05) between the porosities of samples dried by AD and KMFD, and a little higher
averages, determined by the Tukey test. than the porosities of FD and MWVD samples. However, big differences
between pore sizes and structures of the samples produced by the dif-
porosity of 4.2%, after blanching. A small water absorption by the ferent drying methods is observed from the micrographs presented in
samples was observed, with a resulting decrease of the porosity to 1.5%. Fig. 3.
The apparent density of fresh samples is comparable to the values re- The low apparent densities of FD, MWVD and MWMFD samples are
ported by Garcia, Mauro, and Kimura (2007) for the same pumpkin explained by the highly porous structure formed during drying. Freeze-
species. They reported values ranging from 0.9988 to 1.0165 g/cm. drying preserves the structure and the large number of pores present in
AD samples presented an increase of the apparent density, ρap, while the raw sample (Krokida & Maroulis, 1997; Ratti, 2001). On the con-
pumpkins dried by FD, KMFD, MWMFD and MWVD presented a re- trary, MWMFD do not preserve the porous structure. The formation of
duction of this property. The increase of ρap, caused by air drying big pores during each vacuum pulse changes completely the porous
processes, was also reported by the literature for banana, carrot, potato, space, as showed in Fig. 3. On the other hand, the big pores observed in
and pumpkin (Garcia et al., 2007; Krokida & Maroulis, 1997). In the MWVD samples are explained by the volumetric heating caused by the
present study it was observed that the apparent density of AD pumpkin microwaves applied under vacuum, which causes fast vapor produc-
increased 41.6%. Garcia et al. (2007) showed increases of 24–30%. This tion. The force of the vapor flow from inside to the material surface
behavior is related to the well-known drying shrinking, caused by the causes the structure deformation, forming a new porous space.
stresses generated by the increase of the capillary pressure during (Monteiro et al., 2016; Porciuncula et al., 2016; Zotarelli et al., 2012).
drying. Shrinking reduces the material volume, resulting in more
compact porous structures (Liu, Chen, Liu, & Wan, 2012). The drying

Fig. 3. Scanning electron micrographs (SEM) of fractures of dried pumpkins (magnifications of 30× and 100×).

616
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

Fig. 4. Strain-force curves of pumpkins dehydrated by AD (- - -), FD (‒‒‒), KMFD (- - -), MWMFD (‒‒‒) and MWVD (·····).

Fig. 5. Amplitude of force oscillation from representative puncture tests results. Top line: Strain-force data (●) and non-parametric fit (−); Bottom line: residues
curves.

Table 2 fruits and vegetables that influence consumer choice. Processed foods
Apparent density (ρap) and porosity (ε) for fresh, bleached and dried pumpkins, that maintain some properties of the fresh material are well accepted by
mean ± (standard deviation). the consumers (Esehaghbeygi, Pirnazari, & Sadeghi, 2014).
Drying condition ρap (g/cm3) ε (%) From images of Fig. 6 (qualitative) and the parameter L* (bright-
ness), on can observe a browning of the blanched samples in compar-
Fresh 1.0104d ( ± 0.0124) 4.2a ( ± 1.2) ison to the fresh samples. Samples dried by AD and KMFD were darker
Bleached 1.0192d ( ± 0.0075) 1.5a ( ± 0.6)
than the blanched pumpkins, as given by the lower values of the
MWMFD 0.3098a ( ± 0.0074) 69.1d ( ± 0.6)
MWVD 0.2982a ( ± 0.0224) 67.4d ( ± 3.3) parameter L*. AD samples showed higher browning, due to the higher
KMFD 0.6425c ( ± 0.0470) 44.5c ( ± 7.0) temperatures and air exposition for long periods. FD, MWVD and
FD 0.5489b ( ± 0.0324) 63.7d ( ± 2.6) MWMFD samples showed higher values of the parameter L*, in com-
AD 1.4308e ( ± 0.0580) 26.2b ( ± 4.7) parison to the blanched pumpkins. However, freeze-drying resulted in
the highest value of L*, in agreement with the correspondent image
Different letters represent a significant difference (p < 0.05) between
averages, determined by the Tukey test.
showed in Fig. 6. High values of L* were also reported by Guiné and
Barroca (2012) for freeze-dried pumpkin, and by Monteiro et al. (2015)
3.4. Color analyses for freeze-dried carrots.
Only pumpkin samples dried by microwave (MWVD and MWMFD)
Fig. 6 presents pictures of fresh, blanched and dried pumpkin showed higher values of b*, in comparison to the blanched samples,
samples, while Table 3 shows the mean value ( ± standard deviation) of while all the dried samples showed reduced values of the parameter a*.
the color parameters (L*, a*, b* and ΔE *) of samples dried by the Alibas (2007) reported similar results for pumpkins dried by micro-
different drying methods. Color is another important parameter of dried wave, air-drying and air-drying combined with microwave drying.

617
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

Fig. 6. Photograph of pumpkins fresh, bleached and dried

by different methods of drying. (1) Fresh, (2) bleached, (3)
AD, (4) FD, (5) KMFD, (6) MWMFD (7) MWVD.

Table 3
Color parameters (L*, a*, b*, ΔE*) (mean ± standard error) of pumpkins fresh, bleached and dried by different drying methods.
Pumpkins L* a* b* ΔE*

Fresh 61.54 ± 0.83e

36.07 ± 0.47 c
56.84 ± 0.70e
–
Blanched 48.32 ± 0.24c 26.19 ± 0.55b 40.34 ± 0.76c –
MWMFD 54.90 ± 1.37d 25.56 ± 1.78b 45.19 ± 1.82d 8.51 ± 1.30a
MWVD 56.48 ± 2.03d 25.49 ± 2.01b 46.95 ± 2.08d 10.77 ± 2.52ab
KMFD 43.62 ± 1.73b 25.46 ± 1.34b 34.19 ± 2.25b 7.92 ± 2.64a
FD 76.03 ± 2.29f 17.06 ± 2.29a 31.82 ± 3.36ab 30.51 ± 3.69c
AD 38.53 ± 0.90a 24.20 ± 1.05b 29.80 ± 0.78a 14.56 ± 1.21b

Different letters represent a significant difference (p < 0.05) between averages, determined by the Tukey test.

KMFD, MWMFD and MWVD processes showed lower ΔE* values, Drouzas, A. E., & Schubert, H. (1996). Microwave application in vacuum drying of fruits.
indicating smaller color changes during drying. Moreover, ΔE* values of Journal of Food Engineering, 28, 203–209.
Esehaghbeygi, A., Pirnazari, K., & Sadeghi, M. (2014). Quality assessment of electro-
the samples are not different (p < 0.05). Thus, KMFD, MWMFD and hydrodynamic and microwave dehydrated banana slices. Lebensmittel-Wissenschaft
MWVD can produce dried pumpkins with uniform color, without und -Technologie- Food Science and Technology, 55, 565–571.
burned regions. Garcia, C. C., Mauro, M. A., & Kimura, M. (2007). Kinetics of osmotic dehydration and air-
drying of pumpkins (Cucurbita moschata). Journal of Food Engineering, 82, 284–291.
Gonzalez Martinez, C., Corradini, M. G., & Peleg, M. (2003). Effect of moisture on the
4. Conclusions mechanical properties of pork rind (Chicharrón). Food Science and Technology
International, 9, 249–255.
Guiné, R. P., & Barroca, M. J. (2012). Effect of drying treatments on texture and color of
Microwave vacuum drying and microwave multi-flash drying result vegetables (pumpkin and green pepper). Food and Bioproducts Processing, 90, 58–63.
in dried pumpkins with big pores in a highly porous structure, char- Gunasekaran, S. (1999). Pulsed microwave-vacuum drying of food materials. Drying
acteristic of crispy products. Pumpkins dehydrated by microwave Technology, 17, 395–412.
Hammar, K. A., Carson, C. F., & Riley, R. V. (1999). Antimicrobial activity of essential oils
multi-flash drying are crispier than those dried by hot air, conductive
and other plant extracts. Journal of Applied Microbiology, 86, 985–990.
multi-flash drying, microwave vacuum drying, and freeze drying. The Ibarz, A., & Barbosa-Canovas, B. (2003). Unit operations in food engineering. CRC Press LLC.
porosity of pumpkins from both microwave drying processes are com- Jangam, S. V., Law, C. L., & Mujumdar, A. S. (2010). Drying of foods, vegetables and fruits
parable to freeze dried samples, and higher than the observed from (v.1). Singapur: Universidad Nacional de Singapur.
Jian, L., Du, C. J., Lee, A. H., & Binns, C. W. (2005). Do dietary lycopene and other
samples produced by the other drying processes (KMFD and AD). carotenoids protect against prostate cancer? International Journal of Cancer, 113,
Moreover, all the multi-flash drying processes results in dried samples 1010–1014.
with low color changes, while air drying and freeze drying resulted in Krokida, M. K., & Maroulis, Z. B. (1997). Effect of drying method on shrinkage and
porosity. Drying Technology, 15, 2441–2458.
dried pumpkin with high color change. Both microwave-based pro- Krokida, M. K., & Maroulis, Z. B. (1999). Effect of microwave drying on some quality
cesses are suitable to produce high quality and crispy pumpkins with properties of dehydrated products. Drying Technology, 17, 449–466.
moisture content like freeze-dried pumpkins, in only 1.3 h. Therefore, Laurindo, J. B., & Peleg, M. (2007). Mechanical measurements in puffed rice cakes.
Journal of Texture Studies, 38, 619–634.
these drying methods have great potential to produce dried-and-crispy Laurindo, J. B., & Peleg, M. (2008). Mechanical characterization of shredded wheat.
fruits and vegetables. The conductive multi-flash drying process is also Journal of Texture Studies, 39, 444–459.
a suitable process to produce dried fruits with similar characteristics, Lin, T. M., Durance, T. D., & Scaman, C. H. (1998). Characterization of vacuum micro-
wave, air and freeze dried carrot slices. Food Research International, 31, 111–117.
but with higher drying times. The process choice will depend on the Link, J. V., Tribuzi, G., & Laurindo, J. B. (2017a). Conductive multi-flash drying of mango
industry's investment capacity. slices: Vacuum pulse conditions on drying rate and product properties. Journal of
Food Processing and Preservation, 42, e13440 1–10.
Link, J. V., Tribuzi, G., & Laurindo, J. B. (2017b). Improving quality of dried fruits: A
Acknowledgements
comparison between conductive multi-flash and traditional drying methods. LWT-
Food Science and Technology, 84, 717–725.
Authors thanks to CNPq-Brazil and CAPES-Brazil for research Liu, G., Chen, J., Liu, M., & Wan, X. (2012). Shrinkage, porosity and density behaviour
funding. Authors also are thankful to the microscopy center of the during convective drying of bio-porous material. Procedia Engineering, 31, 634–640.
Louka, N., & Allaf, K. (2002). New process for texturizing partially dehydrated biological
Federal University of Santa Catarina (LCME-UFSC). products using controlled sudden decompression to the vacuum: Application on po-
tatoes. Journal of Food Science, 67, 3033–3038.
References Lozano, J. E., Rotstein, E., & Urbicain, M. J. (1980). Total porosity and open-pore porosity
in the drying of fruits. Journal of Food Science, 45, 1403–1407.
Madiouli, J., Sghaier, J., Lecomte, D., & Sammouda, H. (2012). Determination of porosity
A.O.A.C. - Association of Official Analytical Chemists (2005). Official methods of analysis change from shrinkage curves during drying of food material. Food and Bioproducts
(18th ed.). Gaithersburg, Maryland: A.O.A.C. International. Processing, 90, 43–51.
Alibas, I. (2007). Microwave, air and combined microwave-air-drying parameters of Monteiro, R. L., Carciofi, B. A. M., & Laurindo, J. B. (2016). A microwave multi-flash
pumpkin slices. Lebensmittel-Wissenschaft und -Technologie- Food Science and drying process for producing crispy bananas. Journal of Food Engineering, 178, 1–11.
Technology, 40, 1445–1451. Monteiro, R. L., Carciofi, B. A. M., Marsaioli, A., Jr., & Laurindo, J. B. (2015). How to
Bardaa, S., Halima, N. B., Aloui, F., Mansour, R. B., Jabeur, H., Bouaziz, M., et al. (2016). make a microwave vacuum dryer with turntable. Journal of Food Engineering, 166,
Oil from pumpkin (Cucurbita pepo L.) seeds: Evaluation of its functional properties on 276–284.
wound healing in rats. Lipids in Health and Disease, 15, 73. Monteiro, R. L., Link, J. V., Tribuzi, G., Carciofi, B. A., & Laurindo, J. B. (2018).
Boaduo, N. K., Katerere, D., Eloff, J. N., & Naidoo, V. (2014). Evaluation of six plant Microwave vacuum drying and multi-flash drying of pumpkin slices. Journal of Food
species used traditionally in the treatment and control of diabetes mellitus in South Engineering, 232, 1–10.
Africa using in vitro methods. Pharmaceutical Biology, 52, 756–761. Mousa, N., & Farid, M. (2002). Microwave vacuum drying of banana slices. Drying

618
R.L. Monteiro et al. LWT - Food Science and Technology 96 (2018) 612–619

Technology, 20, 2055–2066. explanation from capillary pressure data. Drying Technology, 32, 1692–1698.
Nawirska-Olszańska, A., Kita, A., Biesiada, A., Sokół-Łętowska, A., & Kucharska, A. Z. Seremet, L., Botez, E., Nistor, O. V., Andronoiu, D. G., & Mocanu, G. D. (2016). Effect of
(2013). Characteristics of antioxidant activity and composition of pumpkin seed oils different drying methods on moisture ratio and rehydration of pumpkin slices. Food
in 12 cultivars. Food Chemistry, 139, 155–161. Chemistry, 195, 104–109.
Porciuncula, B. D. A., Segura, L. A., & Laurindo, J. B. (2016). Processes for controlling the Sereno, A. M., Silva, M. A., & Mayor, L. (2007). Determination of particle density and
structure and texture of dehydrated banana. Drying Technology, 34, 167–176. porosity in foods and porous materials with high moisture content. International
Procida, G., Stancher, B., Cateni, F., & Zacchigna, M. (2013). Chemical composition and Journal of Food Properties, 10, 455–469.
functional characterization of commercial pumpkin seed oil. Journal of the Science of Tribuzi, G., & Laurindo, J. B. (2014). How to adapt a lab-scale freeze dryer for assessing
Food and Agriculture, 93, 1035–1041. dehydrating curves at different heating conditions. Drying Technology, 32,
Rabrenovic, B. B., Dimic, E. B., Novakovic, M. M., Tesevic, V. V., & Basic, Z. N. (2014). 1119–1124.
The most important bioactive components of cold pressed oil from different pumpkin Yang, X., Zhao, Y., & Lv, Y. (2007). Chemical composition and antioxidant activity of an
(Cucurbita pepo L.) seeds. Lebensmittel-Wissenschaft und -Technologie- Food Science and acidic polysaccharide extracted from Cucurbita moschata Duchesne ex Poiret. Journal
Technology, 55, 521–527. of Agricultural and Food Chemistry, 55, 4684–4690.
Ratti, C. (2001). Hot air and freeze-drying of high-value foods: A review. Journal of Food Zhang, M., Tang, J., Mujumdar, A. S., & Wang, S. (2006). Trends in microwave-related
Engineering, 49, 311–319. drying of fruits and vegetables. Trends in Food Science & Technology, 17, 524–534.
Ratti, C. (2008). Freeze and vacuum drying of foods. In X. Chen, & A. Mujumdar (Eds.). Zotarelli, M. F., Porciuncula, B. D. A., & Laurindo, J. B. (2012). A convective multi-flash
Drying technologies in food processing (pp. 225–243). Blackwell Publishing. drying process for producing dehydrated crispy fruits. Journal of Food Engineering,
Savitzky, A., & Golay, M. J. E. (1964). Smoothing and differentiation of data by simplified 108, 523–531.
least squares procedures. Analytical Chemistry, 36, 1627–1639. Zuhair, H. A., Abd El-Fattah, A. A., & El-Sayed, M. I. (2000). Pumpkin-seed oil modulates
Segura, L. A., Badillo, G. M., & Alves-Filho, O. (2014). Microstructural changes of apples the effect of felodipine and captopril in spontaneously hypertensive rats.
(Granny Smith) during drying: Visual microstructural changes and possible Pharmacological Research, 41, 555–563.

619