LWT - Food Science and Technology 62 (2015) 1112e1119

Contents lists available at ScienceDirect

LWT - Food Science and Technology

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

Physical quality of snacks and technological properties of

pre-gelatinized flours formulated with cassava starch and dehydrated
cassava bagasse as a function of extrusion variables
Fernanda Assumpça~o Fiorda a, Manoel Soares Soares Jr. a, Fla
vio Alves da Silva a,
Camila Melo Araújo de Moura , Maria Victoria Eiras Grossmann b, *
a

a s, Campus Samambaia, Rod. Goia

Escola de Agronomia e Engenharia de Alimentos, Universidade Federal de Goia ^nia/ Nova Veneza, KM 0, CP 131, Goia
^nia,
CEP 74690-900, GO, Brazil
b
Departamento de Ci^encia e Tecnologia de Alimentos, Universidade Estadual de Londrina, Rod. Celso Garcia, KM 380, CP 6001, Londrina,
CEP86055-900, PR, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: In cassava industrialization, starch is produced, but residues are generated, including bagasse, which has
Received 26 August 2014 high moisture content and is generally not availed, creating a serious environmental problem. The
Received in revised form objective of this work was to study the effects of moisture (12e20 g 100 g
1) and extrusion temperature
18 February 2015
(61.7e118.3  C) on the expansion and color of snacks, and on functional properties of pre-gelatinized
Accepted 23 February 2015
flours formulated with cassava starch and dehydrated cassava bagasse (70:30), to verify the possibility
Available online 4 March 2015
of extending the application of this co-product. Response surface methodology and analysis of variance
were applied. The moisture of the mixture and the extrusion temperature significantly affected physical
Keywords:
Manihot esculenta Crantz
properties of snacks and functional properties of pre-gelatinized flours. Snacks lighter and yellowish
Byproduct with larger expansion index and intermediate specific volume were obtained at the extrusion temper-
Expansion ature of 104.1  C and moisture of 16 g 100 g
1. Pre-gelatinized flour with high absorption rates, solubility
Oil absorption in water and oil absorption were obtained at extrusion temperatures of 90  C and moisture of
Water solubility 16 g 100 g
1. It is viable to use the dehydrated cassava bagasse and cassava starch (30:70) in the pro-
duction of snacks and pre-gelatinized flours.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction alternatives for using this product are necessary, including its
destination for human consumption.
Cassava starch (Manihot esculenta Crantz) producers generate Numerous studies using industrial residue from food processing
large amounts of residues that are harmful to the environment, have been conducted in order to develop new products. Thus, it
mainly the peel, wastewater and cassava bagasse. Cassava bagasse minimizes the environmental impact of these types of industries in
is the main solid residue produced by cassava starch industry and it the regions where they are located and add value to residues.
is thrown into water courses or left in ditches which over flows and Studies have been performed to develop technologies for using
carry much of the organic load towards them. The high moisture dehydrated cassava bagasse (CB) as a fiber enriching ingredient in
content, around 85 g 100 g
1, impairs the use of cassava bagasse by various food products such as breads, cakes, cookies and pasta
the starch industry, because there is a strong possibility of microbial (Fiorda, Soares, Silva, Grossmann, & Souto, 2013a, b, c; Shittu,
contamination of the product, causing undesirable fermentation. In Dixon, Awonorin, Sanni, & Maziya-Dixon, 2008). Another possible
this sense, for the effective utilization of cassava bagasse, drying application for CB is its mixture with cassava starch (CS) for pro-
simultaneously to production is essential. However, for the drying duction of snacks and pre-gelatinized flours (PGFs), applying the
process becomes economically interesting for entrepreneurs, extrusion technology.
The thermoplastic extrusion is a continuous process in which
the raw material is forced through a die or mold, under heating
conditions, pressure and molecular friction. Such conditions lead to
* Corresponding author. Tel.: þ55 43 33714080.
E-mail address: [email protected] (M.V.E. Grossmann). gelatinization of starch, denaturation of proteins and rupture of

http://dx.doi.org/10.1016/j.lwt.2015.02.030
0023-6438/© 2015 Elsevier Ltd. All rights reserved.
 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119 1113

hydrogen bonds, causing cooking, mixing, sterilization, drying, 2.3. Analysis

restructuring of the raw material and creating new textures and
shapes. 2.3.1. Chemical compositions
The physical quality of the extruded snacks is dependent on the The mixtures were evaluated with respect to their moisture, ash
raw materials used in its formulation (composition, particle size, and dietary fiber (total, soluble and insoluble) contents using
moisture), as well as on the processing variables (temperature, methods 925.09, 923.03 and 985.29, respectively, (AOAC, 1997); for
pressure, screw speed) and all the equipment variables, such as the protein using method 960.52 (AOAC, 1990) and lipids by method
mechanical conformation, screw configuration, type and di- 920.39C (AACC, 2000).
mensions of the die (Day & Swanson, 2013). On the other hand, the
extrusion conditions may cause hydrolysis or gelatinization of the
2.3.2. Structural analysis
starch molecule, which can modify its ability to interact with water
The volume of snacks was evaluated after extrusion and before
molecules, interfering with its hydration capacity and water solu-
drying process, by displacement of millet seeds, with 16 replicates
bility. These are important defining parameters for applications of
per experimental unit. The mass was evaluated in semi-analytical
pre-gelatinized flour as an ingredient (Altan, Mcarthy, & Maskan,
scale. The specific volume (SV) was calculated by the ratio be-
2009). The extrusion-cooking has been applied to raw materials
tween the average volume and mass of snacks. The expansion index
with different compositions (Alonso, Orúe, Zabalza, Grant, & Marzo,
(EI) was determined by the ratio between the diameter of the
2000; De Pilli, Carbone, Derossi, Fiore, Severini, 2008; De Pilli,
extrudate and the diameter of the output hole of the extruder
Derossi, Talja, Jouppila, Severini, 2011; De Pilli, Giuliani, Carbone,
(4 mm). A digital caliper was used for measuring the diameter of
Derossi, & Severini, 2005; De Pilli, Jouppila, et al., 2008, 2012;
the extrudate, and the arithmetic mean was calculated from 10
Ding, Ainsworth, Tucker, & Marson, 2005; Dust et al., 2004)
randomly chosen snacks per experiment.
becoming an interesting process for the alternative application of
dehydrated cassava bagasse.
The objective of this work was to study the effects of moisture 2.3.3. Instrumental color parameters
and extrusion temperature on the expansion and color of snacks, The snacks also were evaluated for instrumental parameters of
and on the technological properties of PGFs formulated with CS and color, according to the Cielab system obtaining the values L* (lu-
CB (70:30), to verify the possibility of extending the application of minosity), a* (red-green component) and b* (yellow-blue compo-
this co-product. nent) in colorimeter (Color Quest, XE, Reston, EUA). Observation
angle of 10 and standard illuminant D65 were fixed, correspond-
2. Material and methods ing to natural daylight. The color difference (DE*) between the CM
and PGF was also obtained (Eq. (1)). Analyses were performed in
2.1. Material triplicate.

Cassava starch (CS) and cassava bagasse (CB) samples were  1=2
donated by Fecularia Bela Vista Ltda., located in Bela Vista de Goias,
DE ¼ DL2 þ Da2 þ Db2 (1)
Goias, Brazil. CB was collected at the entrance of the storage silo, where, DL* ¼ Lo (CM luminosity)
Le (PGF luminosity); Da* ¼ ao
dried in an oven with air circulation at 55  C for 24 h, and ground in (CM a* value)
ae (PGF a* value); Db* ¼ bo (CM b* value)
be (PGF
a mill rotor equipped with a 30-mesh sieve. b* value).

2.2. Extrusion
2.3.4. Water absorption index (WAI), water solubility index (WSI)
and oil absorption index (OAI)
In the extrusion experiment it was used a crude mixture (CM) of
The PGF and CM were evaluated as for water absorption index
SC and CB (70:30), obtained in a Y-type homogenizer mixer (Tecnal,
(WAI) and water solubility index (WSI) according to the method of
TE 201/05, Piracicaba, Brazil) for 15 min. The mixture was stored in
Anderson, Conway, Pfeifer, and Griffin (1969). About 2.5 g (db) of
nylon bags in different moisture levels, according to the values set
the ground extrudate was suspended in 30 mL of water at room
in the experimental design, and kept under refrigeration, 24 h
temperature for 30 min gently stirred during this period, and then
before use.
centrifuged at 3000  g for 15 min. The supernatant was decanted
The thermoplastic extrusion was performed on equipment with
into an evaporating dish of known weight. The WSI is the weight of
single screw (Inbramaq, PQ-30, Ribeira ~o Preto, Brazil). The fixed
dry solids in the supernatant expressed as a percentage of the
processing parameters were: screw speed of 250 rpm (60 Hz),
original weight of sample. The WAI is the weight of gel obtained
circular die opening of 4 mm diameter, pre-die with 22 holes, screw
after removal of the supernatant per unit weight of original dry
of 300 mm length with compression ratio of 3:1, helical barrel
solids.
design, feed rate of 350 g min
1; and temperature in the first and
For the determination of oil absorption index (OAI) the same
second heating zones of 50  C and 57  C, respectively. The heating
WAI methodology adapted was used, but the water was replaced by
system is done by electrical resistors independents for each zone
soya oil. All analyses were performed in triplicate.
and the cooling is done with cold water. The temperature control
involves a programming system and thermocouples; when the
temperature increases above the programmed value, a solenoid 2.3.5. Scanning electron microscopy
valve is open, and begins pumping cold water until the temperature Micrographs of CM and fractured PGF samples (center point)
decrease, then closing. The variables processing parameters were with magnification of 1600 were performed in a scanning
the temperature of the third heating zone (61.7e118.3  C) and the electron microscope (FEI Quanta-200, Hillsboro, USA). The sam-
moisture of the material (12e20 g 100 g
1). After extrusion, the ples were completely dried at 105  C for 24 h, remaining in a
snacks were dried at 100  C for 15 min in an oven with air circu- desiccator until the time of its preparation. The time of deposition
lation, to reduce the moisture content to values around 3%. The of gold in the samples was 98 s, representing 15 nm layer of
snacks were stored in nylon bags. material.
1114 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119

2.4. Experimental design and statistical analysis in the formulation of foods with high total dietary fiber, such as
light and diet food.
The Central Composite Rotational design was used with 12 ex- The WAI value obtained in this study for CM (Table 1) is within
periments and 4 replicates at the central point (Box, Hunter & the range found by Trombini, Leonel, and Mischan (2013) that re-
Hunter, 2005). The coded values of the independent variables ported variations in WAI from 2.35 to 2.92 ggel g
1 dry matter, in CM of
were
1.41;
1; 0; 1 and 1.41, while the real temperature values of soy flour (10e30 g 100 g
1), CS (50e70 g 100 g
1) and CB (10e30 g
the third zone of the extruder ranged between 61,7 and 118,3  C 100 g
1) for the preparation of PGF. These authors found values of
and moisture between 12 and 20 g 100 g
1. WSI between 2.78 and 6.81 g 100 g
1, substantially higher than
Data from EI, SV, L*, chrome a* and b* of snacks and WAI, WSI those found for CM in this study, which is probably due to the high
and OAI of pre-gelatinized flours were evaluated by analysis of amount of protein from soybean meal, since most part of the soy
variance, with the construction of multiple regression models. proteins (globulins) are soluble in water. Oil absorption constitutes
Graphs of level curves for the visualization of the effect of inde- the ability of nonpolar sites of protein chains to imprison the tri-
pendent variables on the responses were constructed with the aid acylglycerides (Ravi & Sushelamma, 2005). Thus, the content and
of the software Statistica (Statsoft, Statistica 7.0, Tulsa, USA). From type of protein present in the flour determine the oil absorption
the significant mathematical models (p  0.05) and using the capacity of foods. The CM had low protein content (Table 1), and
function Response Desirability Profiling of the software Statistica, it thus low value of OAI.
was determined the most desirable snack, depending on the There was significant variation in all characteristics evaluated in
moisture and extrusion temperature used. The most desirable snacks, from 19% for b* to 163% for the color difference (DE).
snack was the one with higher values of EI, L* and croma b*, lower Therefore, all generated regression models were significant,
of DE*, and intermediate of SV. The optimization technique was explaining 77.3e96% of the responses. Also there was significant
based on the definition of a function of restricted desirability in the variation in all characteristics in PGFs, from 36% for WSI to 100.3%
range from 0 to 1, for which it was adopted as lower, middle and for OAI, and all generated regression models were significant,
upper limits at values of 0, 0.5 and 1.0 respectively for dependent explaining 88e98% of the responses (Tables 2 and 3). Linear effects,
variables studied. After determining the process conditions for quadratic effects and interactions of extrusion temperature and
obtaining the most desirable snack, a new extrusion was carried out moisture were also significant for all models except linear tem-
under these conditions for validating the mathematical models. perature (x1) for EI, chroma a* and WSI, quadratic temperature (x21)
for OAI, linear moisture (x2) for chroma b* WAI and WSI, and the
3. Results and discussion interaction temperature x moisture (x1x2) for EI and WAI.
Snacks EI ranged 75% * (1.22e2.13) (Table 2). The maximum
The CM showed intermediate luminosity, yellow color (sub- value of EI (2.0) was obtained when the sample was extruded with
stantial chroma b*) and low influence of the red spectrum (chroma lower moisture content (up to 16 g 100 g
1) and intermediate
a* near neutrality) (Table 1). Probably, due to the lack of red and temperature (76.6e103.4  C), while the lowest EI (below 0.8) was
presence of yellow pigments in the raw materials, or due to the ash obtained under conditions of higher moisture and temperature
and fiber contents present in the CB. Some studies also report the above 19.2 g 100 g
1 and 114.6  C, respectively (Fig. 1A). According
use of alternative flours and bran, rich in fiber, making food prod- to Ding et al. (2005), the increased amount of water during the
ucts darker than the traditionally formulated ones (Lacerda, Soares, extrusion may change the molecular structure of amylopectin,
Bassinello, Siqueira, & Koakuzu, 2009). acting as a plasticizer to starch material, reducing the viscosity and
The CM moisture and the ash content of 15 and 2.0 g 100 g
1, the dissipation of mechanical energy in the extruder. Additionally,
respectively (Table 1). Due to its high dietary fiber content the product becomes denser and bubble growth is compressed,
(23.5 ± 2.43 100 g
1 db), CM can be classified as an ingredient rich thereby reducing its expansion, a phenomenon also observed in the
in dietary fiber. Protein and lipid contents were low, according to present study. There is an optimum ratio of amylose to amylo-
the contents of these components observed in the CB and SC, pectin, which permits to obtain the maximum expansion of
respectively 0.14 and 1.97 g 100 g
1 for protein, and 1.56 and 2.35 g extrusion-cooked starchy products, indicating that both amylose
100 g
1 for lipids (Fiorda et al. 2013a, b, c). Thus, the CM can be used and amylopectin contribute to the expansion of starches. The
amylose and native lipids present in cereal starches may inhibit
swelling under particular conditions when amyloseelipid com-
Table 1
Color parameters (L*, a* e b*), chemical composition and functional plexes are likely to be formed (De Pilli, Jouppila, et al., 2008).
properties of the crude mixture (CM) of cassava starch and cassava The high fiber content present in CM (Table 1) probably influ-
bagasse (70:30). enced the expansion of the extrudates. Leonel, Souza, and Mischan
Parameters CMa (2010), in studies with extruded CS and orange fiber (90:10), with
temperatures between 40 and 90  C, moisture contents from 12.5
Luminosity (L*) 56.23 ± 0.55
Chroma a* 3.32 ± 0.55 to 19.5 g 100 g
1, obtained IE results ranging from 1.93 to 2.99,
Chroma b* 14.30 ± 0.55 which higher values were observed in intermediate extrusion
Moistureb 12.41 ± 0.22 temperatures and moisture levels.
Ashesb 0.71 ± 0.04 The SV variation among snacks was 115% (6.09e13.07) (Table 2).
Lipidsb 1.87 ± 0.01
Proteinb 0.60 ± 0.02
The highest SV values (above 14 mL g
1) were obtained when CM
Total dietary fiberb 23.5 ± 2.43 was extruded with moisture contents below 15.0 g 100 g
1 and
Soluble dietary fiberb 5.25 ± 0.82 temperatures above 97.5  C (Fig. 1B). SV is a measure of volumetric
Insoluble dietary fiberb 18.25 ± 1.53 expansion, which is the sum of the radial and axial expansions.
Water absorption índexc 2.83 ± 0.08
Launay and Lisch (1983) report the phenomenon of expansion as
Water solubility indexb 0.78 ± 0.05
Oil absorption indexd 0.64 ± 0.09 dependent on the viscoelastic properties of the molten mass.
a
Increasing the temperature, the molten viscosity is reduced and the
Mean value ± standard deviation.
b
g 100 g
1.
axial expansion increases. On the other hand, the lowest SV values
c
ggel g
1
dry matter.
(below 8 mL g
1) were found in intermediate conditions of mois-
d
gprecipitate g
1
dry matter. ture, from 14.3 to 18.5 g 100 g
1, and temperatures below 93.8  C.
 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119 1115

Table 2
Expansion and color of snacks and functional properties of pre-gelatinized flours as a function of temperature (x1) and moisture (x2).

Run x1 x2 Temperature Moisture (g 100 g
1) Snack properties Pre-gelatinized flour properties
( C)
EIa SVb L* a* b* DE WAIc WSId OAIe

1
1
1 70 13.1 1.54 6.47 37.03 5.02 16.37 19.40 7.61 64.66 5.42
2 1
1 110 13.1 1.74 12.07 30.24 4.04 17.05 26.15 5.92 61.83 5.85
3
1 1 70 18.8 1.29 7.46 25.54 4.75 15.16 30.74 7.77 69.70 3.12
4 1 1 110 18.8 1.23 7.34 37.44 6.13 16.81 19.17 5.60 58.93 4.27
5
1.41 0 61.7 16 1.78 7.94 38.57 6.29 17.03 18.11 7.50 61.26 3.94
6 1.41 0 118.3 16 1.25 13.07 42.71 6.04 17.04 14.09 6.77 64.43 5.48
7 0
1.41 90 12 2.13 12.87 34.34 4.64 16.29 22.02 5.76 60.55 5.41
8 0 1.41 90 20 1.22 8.56 26.05 4.67 16.63 30.30 5.75 58.30 2.92
9 0 0 90 16 1.82 6.09 44.49 7.43 17.93 12.98 9.19 73.97 4.75
10 0 0 90 16 2.00 7.11 45.89 7.68 17.74 11.88 8.48 72.02 4.84
11 0 0 90 16 1.94 7.65 46.15 7.73 18.05 11.67 10.48 79.29 4.94
12 0 0 90 16 1.86 7.24 44.19 7.14 17.62 13.06 10.36 79.09 4.86
a
Expansion index.
b
Specific volume (g mL
1).
c
Water absorption index (ggel g
1 dry matter).
d
Water solubility index (g 100 g
1).
e
Oil absorption index (gprecipitate g
1
dry matter).

Decreasing temperature, starch gelatinization is reduced and the were low, by 22 on a scale of 0 (black) to 100 (white), rendering
growth of the bubbles is compressed, resulting in a denser final darker samples (Fig. 1C). These same values were found in mois-
product with lower crispness (Ding et al., 2005). In studies with tures above 18.2 g 100 g
1 and extrusion temperature up to 82.5  C.
snacks of CS and CB, the SV of extrudates with different proportions However, at moisture and temperature in intermediate levels from
of CB (0.5e7.5 g 100 g
1) ranged from 1.5 to 5.6 g mL
1 (Camargo, 14.4 to 16.7 g 100 g
1 and from 109 to 80.6  C, respectively, less
Leonel & Mischan, 2008), which are lower than the results found in browning occurred in snacks, that is, they showed the highest L*
this study. According to these authors, high levels of fibers com- value (45), represented in the surface graph by the concave down,
bined with moisture content up to 18 g 100 g
1 reduce the SV of the indicating a point of maximum value of this component. Interme-
extruded products. However, in the present study, we used a much diate levels of moisture and temperature resulted in lighter prod-
higher amount of CB (30 g 100 g
1) and slightly higher moisture ucts. Increasing moisture reduces the residence time, providing
content (up to 100 g 20 g
1) and still higher values of SV were lower non-enzymatic browning of extruded products (Badrie &
obtained, probably due to the larger axial expansion caused by Mellowes, 1991). Vernaza, Chang, and Steel (2009), evaluating the
adding CB, resulting in thinner, but longer snacks. effect of moisture and extrusion temperature, on the color of
Luminosity values of snacks obtained after extrusion were breakfast cereal made from corn flour and commercial passion fruit
lower, from 17.9% for the experiment 11e54.8% for experiment 3 bran, observed L* values ranging from 44.2 to 68.3, higher than
compared with those for CM (Tables 1 and 2). That is, the snack those obtained in this study, while the increase in temperature with
became darker than the mixture of raw materials used in its higher moisture content produced lighter snacks, similar to what
formulation. Meanwhile, there was an increase of chroma a* after occurred in this study.
snacks processing, from 21.7% in experiment 2e132.8% in experi- The maximum value of a* (7.0) for snacks of SC and BC (70:30)
ment 11, and in chroma b* (6% in the experiment 3 and 26.2% in was obtained between 15.13 and 17.74 g 100 g
1 of moisture and
experiment 11), showing that the extrusion process slightly from 80.6 to 101.5  C (Fig. 1D). It was observed that after the snacks
intensified the red and yellow colors of the snacks. reached a maximum chroma a* value, and this value decreased
Under aggressive processing conditions, that is, temperature after 101.5  C and moisture of 17.7 g 100 g
1. In this sense, in in-
above 109  C and maximum moisture of 12.6 g 100 g
1, L* values termediate conditions of moisture and extrusion temperature, the
highest chroma a* values occurred. Trombini and Leonel (2010)
observed variation from 3.6 to 5.5 for chroma a* in extruded mix
Table 3 of SC, CB and soybean meal. The chroma a* was lower in the
Adjusted models and determination coefficient (R2) for expansion index, specific extruded products under low temperature (45e60  C) and speed
volume and instrumental color parameters (L*, a*, b* and DE*) of snacks, and water
(190e210 rpm) conditions. With higher content of CB in the mix
absorption index (WAI), water solubility index (WSI) and oil absorption index (OAI)
of pre-gelatinized flours formulated with cassava starch and cassava bagasse (25e30 g 100 g
1) and low speed (190e210 rpm) higher chroma a*
(70:30). values were obtained, within the range of chroma a* values
observed in this work.
Feature Adjusted modela R2
The maximum chroma b* value (17.5) of the snacks was ob-
Expansion y1 ¼ 1.906
0.26x2
0.23x21
0.15x22 0.79 tained between 14.9 and 17.2 g 100 g
1 of moisture and from 85.8 to
index
Specific y2 ¼ 7.029 þ 1.59x1
1.23x2 þ 1.17x21
1.43x1x2 þ 1.28x22 0.81
107.8  C. It was also observed that after reaching a maximum b*
volume value, there was a decrease after 107.8  C and 17.2 g 100 g
1 of
Luminosity y3 ¼ 45.189 þ 1.37x1
2.00x2
3.00x21 þ 4.67x1x2
8.25x22 0.96 moisture, slightly reducing the yellow intensity (Fig. 1E). The color
(L*) difference reflected the influence of extrusion variables on the color
Chroma a* y4 ¼ 7.497 þ 0.23x2
0.77x21 þ 0.60x1x2
1.53x22 0.95
characteristics of the flour in relation to their appearance before
Chroma b* y5 ¼ 17.835 þ 0.29x1
0.50x21
0.24x1x2
0.79x22 0.77
DE* y6 ¼ 12.391
1.31x1 þ 2.01x2 þ 2.54x21
4.60x1x2 þ 7.60x22 0.96 processing. Color difference (DE*) was lower in intermediate
WAI y7 ¼ 9.624
0.61x1
1.18x21
1.87x22 0.88 moisture and temperature extrusion. The lowest values of DE* (15)
WSI y8 ¼ 76.084
5.97x21
7.69x22 0.80 between the CM and FPG was observed from 15.0 to 16.1 g 100 g
1
OAI y9 ¼ 4.839 þ 0.47x1
0.97x2 þ 0.17x1x2
0.28x22 0.98 moisture and temperatures from 81.6 to 106.8  C, probably because
a
Effects in italics were not significant, but were held to improve the model fit. these levels reduced the occurrence of non-enzymatic browning
1116 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119

Fig. 1. Graphics of expansion index (A), specific volume (B), luminosity (L*) (C), chroma a* (D), chroma b* (E), oil absorption index (F), water absorption index (G) and water
solubility index (H) of snacks and pre-gelatinized flour formulated with cassava starch and bagasse (70:30) according to extrusion temperature ( C) and moisture (g 100 g
1).
 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119 1117

reactions. Above the mentioned limit, DE* value increased probably and higher chain polymers, and high availability of hydrophilic
by the intensified formation rate of polymers with variable struc- groups that water molecules bind to, resulting in high levels of WAI.
ture, such as caramel. During the caramelization reaction, ther- High WAI values can also be related to high moisture levels during
molysis leads to dehydration and generation of double bonds and the extrusion process combined with low shear rates. Because the
formation of rings with labile compounds that condense to form water acts as a lubricant in the extrusion medium, resulting in
polymers, giving the color and flavor of caramel. The intensity of reduction of friction of the screw and the inner wall of the extruder
browning can compromise the acceptance of the obtained extruded barrel on the starch molecules, thereby resulting in less degrada-
products and the results of color parameters seem to indicate that tion of the amylose and amylopectin and, consequently, in an in-
Maillard reaction occur during drying processing. In fact, samples crease in WAI. After the starch granules reach the maximum
showed an increase of L*, when compared to raw materials, as capacity of water absorption, WAI decreases with the start of the
expected when the Maillard reaction takes place (De Pilli, Giuliani, dextrinization.
Derossi, & Severini, 2014). It was also observed that the WAI of experimental FPGs
Snacks from SC and BC (70:30) with maximum expansion, decreased after 90  C with increasing temperature (Fig. 1G), prob-
specific volume, luminosity, chroma b* and lower color difference ably due to starch dextrinization. With increasing temperature and
could be obtained with moisture of 16 g 100 g
1 and extrusion moisture also occurred an increase in the degree of gelatinization,
temperature of 104.1  C (according to the desirability test). These starch fragmentation increased and, thus, decreased water ab-
conditions have not been tested in the applied experimental design, sorption as also observed in the study of Borba, Sarmento, and
but snacks from experiments 9, 10 and 11 (Fig. 2), obtained by Leonel (2005). In studies with snacks from CS and CB, water ab-
processing conditions (16 g 100 g
1 moisture and 90  C) closer to sorption rates ranging from 4.8 to 11.9ggel g
1
dry matter were observed,
those, were the ones with the best group of characteristics. in which treatments with higher amount of CB presented marked
The highest OAI value was observed under conditions of low effect on WAI increasing (Camargo et al., 2008), similar values to
moisture and high temperature (Fig. 1F). According to Alonso et al. those found in the present work. Highest WAI values are considered
(2000), the extrusion conditions may cause changes in protein desirable in bakery products and fast preparation (instant) meals as
molecules, resulting in denaturation, changing and increasing the noodles, puddings, soups and porridges. This is due to the fact of
amount of hydrophobic sites that react with lipids of soybean oil. PGF may allow the addition of more water to the dough, improving
OAI increased as the extrusion temperature was elevated. OAI its handling characteristics and preventing dryness of the product
values below 3gprecipitate g
1 dry matter were obtained at moistures during storage.
above 18 g 100 g
1 and up to 91.18  C of extrusion temperature. The The WSI value of the PGF was much higher than that observed in
maximum value of OAI (5.53gprecipitate g
1 dry matter) was obtained CM from 7.375% (experiment 8) to 10.065% (experiment 11)
above 100.6  C and moisture up to 14.7 g 100 g
1. OAI values ob- (Tables 1 and 2). This can be explained by the fact that the CM had
tained in the PGFs of this research were slightly higher than those no dextrinized starch, such as PGFs. It was observed that the
of PGF of passion fruit bran reported by Vernaza et al. (2009), from highest values of WSI of PGFs were obtained in intermediate con-
2.73 to 3.85gprecipitate g
1
dry matter. This discrepancy occurred probably ditions of moisture and temperature, and that it decreased with
because the content of insoluble dietary fiber present in CB increase in moisture content and temperature after reaching the
(Table 1) was higher when compared to passion fruit bran (49.16 g maximum value (75 g 100 g
1) that was verified between 15.2 and
100 g
1), indicating that insoluble dietary fiber affects this ratio. 17.4 g 100 g
1 of moisture and 82.9e97.9  C (Fig. 1H). The increase
Knowledge of the oil absorption rate (OAI) of the flour is important in moisture may have contributed to the reduction of friction,
for development of new food products and determination of stor- because water acts as a lubricant, leading to lower values of WSI.
age stability, especially with respect to the development of char- When extruding CS with orange fibers (90:10) using temperatures
acteristic flavor of oxidative rancidity. between 40 and 90  C and moisture content from 12.5 to 19.5 g
The highest WAI values were observed in intermediate moisture 100 g
1, Leonel et al. (2010) obtained WSI ranging from 37 8e84.9 g
and temperatures (Fig. 1G). The maximum WAI value (9ggel gdry 100 g
1, values close to those obtained in this work. FPGs with high

1
1
matter ) was obtained between 14.7 and 17.1 g 100 g of moisture values of WSI can be used in products that require lower temper-
and from 73.5 to 97.1  C extrusion temperature. Both low shear and atures to be cooked such as instant and jellified foods: soups,
low temperature extrusion for processing results in undamaged sauces, puddings, desserts, etc. (Trombini & Leonel, 2010).
Structural differences between particles primarily in the starch
granule before and after extrusion were observed. Before the starch
granules were intact with concavity and semi-spherical de-
pressions apparently due to the proximity between the granules
that lost their spherical shapes during formation. No starchy ma-
terials were also identified (probably fiber and protein), which were
adhered to each other or to starch granules, as shown in Fig. 3A.
After the extrusion process (Fig. 3B), there was disruption of the
starch granules of the material, indicating its gelatinization and
protein denaturation. The rupture of hydrogen bonds in these
constituents reflects changes in solubility, viscosity and levels of
amylose and amylopectin, increasing, in general, its digestibility
(Dust et al., 2004; Svihus, Uhlen & Harstad, 2005). The PGF
appeared as a compact, amorphous mass, which was not possible to
distinguish starch granules and not starchy material. It is also
observed that the extrudate had an irregular surface with smooth
and striated parts, many holes formed due to expansion of the
product at the extruder outlet as a function of the water vapor
Fig. 2. Snack of cassava starch and dehydrated cassava bagasse obtained at 90  C with released by the product and decompression occurred in the
16 g 100 g
1 moisture. material.
1118 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119

Fig. 3. Micrographs of (A) crude mixture and (B) pre-gelatinized flour of starch and cassava bagasse of central experiment.

4. Conclusion Day, L., & Swanson, B. G. (2013). Functionality of protein-fortified extrudates.

Comprehensive Reviews in Food Science and Food Safety, 12(5), 546e564.
De Pilli, T., Carbone, B. F., Derossi, A., Fiore, A. G., & Severini, C. (2008). Effects of
The moisture of the mixture and extrusion temperature signif- operating conditions on oil loss and structure of almond snacks. International
icantly interferes in the physical properties of snacks and pre- Journal of Food Science & Technology, 43(3), 430e439.
gelatinized flours. Snacks yellow and lighter with higher expan- De Pilli, T., Derossi, A., Talja, R. A., Jouppila, K., & Severini, C. (2011). Study of starch-
lipid complexes in model system and real food produced using extrusion-
sion and intermediate specific volume were obtained in interme- cooking technology. Innovative Food Science & Emerging Technologies, 12,
diate conditions of extrusion temperature (104.1  C) and moisture 610e616.
(16 g 100 g
1). Pre-gelatinized flours with high absorption rates and De Pilli, T., Derossi, A., Talja, R. A., Jouppila, K., & Severini, C. (2012). Starch-lipid
complex formation during extrusion-cooking of model system (rice starch and
solubility in water and high oil absorption were obtained at oleic acid) and real food (rice starch and pistachio nut flour). European Food
extrusion temperatures of 90  C and moisture of 16 g 100 g
1. It is Research and Technology, 2343(3), 517e525.
viable to use the cassava bagasse and cassava starch (30:70) in the De Pilli, T., Giuliani, R., Carbone, B. F., Derossi, A., & Severini, C. (2005). Study on
different emulsifiers to retain fatty fraction during extrusion of fatty flours.
production of snacks and pre-gelatinized flours, and the process of Cereal Chemistry, 82(5), 494e498.
extrusion-cooking applied to dehydrated cassava bagasse is an De Pilli, T., Giuliani, R., Derossi, A., & Severini, C. (2014). Development of Maillard
important technology for the alternative industrial use of this raw reaction in pasta made up of semolina and dried by microwaves. Italian Journal
of food Science, 26(2), 183e189.
material. De Pilli, T., Jouppila, K., Ikonen, J., Kansikas, J., Derossi, A., & Severini, C. (2008). Study
on formation of starch-lipid complexes during extrusion-cooking of almond
flour. Journal of Food Engineering, 87, 495e504.
Acknowledgments Ding, Q., Ainsworth, P., Tucker, G., & Marson, H. (2005). The effect of extrusion
conditions on the physicochemical properties and sensory characteristics of rice
based expanded snacks. Journal of Food Engineering, 66(5), 283e289.
The authors are grateful to CAPES for financial support and Dust, J. M., Gajda, A. M., Flickinger, E. A., Burkhalter, T. M., Merchen, N. R., &
scholarship, to FAPEG for financial support, and to FEBELA cassava Fahey, G. C., Jr. (2004). Extrusion conditions affect chemical composition and
industry as partnership. in vitro digestion of select food ingredients. Journal of Agricultural and Food
Chemistry, 52, 2989e2996.
Fiorda, F. A., Soares, M. S., Jr., Silva, F. A., Grossmann, M. V. E., & Souto, L. R. F. (2013a).
Farinha de bagaço de mandioca: aproveitamento de um subproduto e com-
References paraça ~o com a fe cula de mandioca. Pesquisa Agropecua ria Tropical, 43(4),
408e416.
AACC. American Association of Cereal Chemists. (2000). Approved methods of the Fiorda, F. A., Soares, M. S., Jr., Silva, F. A., Grossmann, M. V. E., & Souto, L. R. F. (2013b).
American Association of Cereal Chemists (10th ed.). Saint Paul: AACC Microestructure, texture and colour of gluten-free pasta made with amaranth
Internacional. flour, cassava starch and cassava bagasse. LWT e Food Science and Technology,
AOAC. Association of Official Analytical Chemists (United States of America). (1990). 54, 132e138.
Official methods of analysis (15th ed). Washington D.C.: AOAC. Fiorda, F. A., Soares, M. S., Jr., Silva, F. A., Grossmann, M. V. E., & Souto, L. R. F. (2013c).
AOAC. Association of Official Analytical Chemists (United States of America). (1997). Amaranth flour, cassava starch and cassava bagasse in the production of gluten-
Official methods of analysis of AOAC International (16th ed.). Gaithersburg: AOAC free pasta: technological and sensory aspects. International Journal of Food
International. Science & Technology, 48(1), 1977e1984.
Alonso, R., Orúe, E., Zabalza, M. J., Grant, G., & Marzo, F. (2000). Effects of extrusion Lacerda, D. B. C. L., Soares, M. S., Jr., Bassinello, P. Z., Siqueira, B. S., & Koakuzu, S. N.
cooking on structure and functional properties of pea and kidney bean prop- (2009). Qualidade de biscoitos elaborados com farelo de arroz extrusado em
erties. Journal of the Science of Food and Agriculture, 80, 397e493. substituiç~ ao 
a farinha de trigo e fe cula de mandioca. Archivos Latinoamericanos
Altan, A., Mcarthy, T. L., & Maskan, M. (2009). Effect of extrusion cooking on de Nutricion, 59(2), 199e205.
functional properties and in vitro starch digestibility of barley-based extru- Launay, B., & Lisch, J. M. (1983). Twin-screw extrusion cooking of starches: flow
dates from fruit and vegetable by-products. Journal of Food Science, 89, behavior of starch pastes, expansion and mechanical properties of extrudates.
1587e1592. Journal of Food Engineering, 2, 259e280.
Anderson, R. A., Conway, H. F., Pfeifer, V. F., & Griffin, E. L. (1969). Gelatinization of Leonel, M., Souza, L. B., & Mischan, M. M. (2010). Produç~ ao de snacks extrusados a
corn grift by roll and extrusion cook. Cereal Science Today, 14(1), 4e11. base de polvilho doce e fibra de laranja. Ci^ encia Rural, 40(6), 1411e1417.
Badrie, N., & Mellowes, W. A. (1991). Effect of extrusion variables on cassava Ravi, R., & Suselamma, N. S. (2005). Simultaneous optimization of a multy-response
extrudates. Journal of Food Science, 56, 1334e1337. system by desirability function analysis of boondi making: a case study. Journal
Borba, A. M., Sarmento, S. B. S., & Leonel, M. (2005). Efeito dos par^ ametros de of Food Science, 70, S539eS547.
extrusa ~o sobre as propriedades funcionais de extrusados da farinha de batata- Shittu, T. A., Dixon, A., Awonorin, S. O., Sanni, L. O., & Maziya-Dixon, B. (2008).
doce. Ci^encia e Tecnologia de Alimentos, 25(4), 835e843. Bread from composite cassavaewheat flour. II: effect of cassava genotype and
Box, E. P., Hunter, W. G., & Hunter, J. S. (2005). Statistics for experimenters design, nitrogen fertilizer on bread quality. Food Research International, 41(1),
innovation and discovery. Wiley. 569e578.
Camargo, K. F., Leonel, M., & Mischan, M. M. (2008). Produça ~o de biscoitos extru- Svihus, B., Uhlen, A. K., & Harstad, O. M. (2005). Effect of starch granule structure,
sados de polvilho azedo com fibras: efeito de para ^metros operacionais sobre as associated components and processing on nutritive value of cereal starch: a
propriedades físicas. Ci^encia e Tecnologia de Alimentos, 28(3), 586e591. review. Animal Feed Science and Technology, 122(3e4), 303e320.
 F.A. Fiorda et al. / LWT - Food Science and Technology 62 (2015) 1112e1119 1119

Trombini, F. R. M., & Leonel, M. (2010). Propriedades de pasta e termicas de misturas Vernaza, M. G., Chang, Y. K., & Steel, C. J. (2009). Desenvolvimento de cereal
instanta^neas de farinha de soja, fe cula e farelo de mandioca. Energia na Agri- matinal funcional orga^nico enriquecido com farelo de maracuj a: avaliaç~
ao
cultura, 25(4), 50e71. de propriedades tecnolo gicas. Brazilian Journal of Food Technology, 12,
Trombini, F. R. M., Leonel, M., & Mischan, M. M. (2013). Desenvolvimento de snacks 145e154.
extrusados a partir de misturas de farinha de soja, fe cula e farelo de mandioca.
Ci^
encia Rural, 43, 178e184.