Pyrodextrinization of yam (Dioscorea sp.) starch isolated from tubers grown in Brazil and physicochemical characterization of yellow pyrodextrins

Carbohydrate Polymers 242 (2020) 116382

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Pyrodextrinization of yam (Dioscorea sp.) starch isolated from tubers grown
in Brazil and physicochemical characterization of yellow pyrodextrins

T

Mighay Lovera (Dr.)a,b,1,*, George Meredite Cunha de Castro (M.S.)a,
Natalia da Rocha Pires (Dr.)c, Maria do Socorro Rocha Bastos (Dr.)d,
Márjory Lima Holanda-Araújo (Dr.)a, Alexander Laurentin (Ph.D.)e,
Renato de Azevedo Moreira (Dr.)b, Hermógenes David de Oliveira (Dr.)a
a

Department of Biochemistry and Molecular Biology, Federal University of Ceará, CEP 60440-900, Fortaleza, Ceará, Brazil
University of Fortaleza, Health Sciences Center, Av. Washington Soares, 1321 Edson Queiroz, CEP 60811-905, Fortaleza, Ceará, Brazil
Department of Organic and Inorganic Chemistry, Federal University of Ceará, CEP 60440-554, Fortaleza, Ceará, Brazil
d
Food Packaging Technology Laboratory, Embrapa Agroindústria Tropical, St. Dr. Sara Mesquita, 2270-Pici, CEP 60511-110, Fortaleza, Ceará, Brazil
e
Instituto de Biología Experimental, Facultad de Ciencias, Universidad Central de Venezuela, Apartado postal 47114, Caracas, 1041-A, Venezuela
b
c

A R T I C LE I N FO

A B S T R A C T

Keywords:
Pyroconversion
Response surface methodology
Digestibility
Available starch
Dietary fiber
Physicochemical properties

This study optimizes the pyrodextrinization of yam (Dioscorea sp.) starch isolated from tubers grown in Brazil to
produce a yellow pyrodextrin with the lowest enzymatically available starch (AS) content and color difference
(ΔE) index. At 140 °C (fixed heating temperature), the effects of acid concentration (0.65 − 2.99 g of HCl/kg of
starch) and incubation time (53 − 307 min) on the response variables were evaluated using a response surface
methodology. Some physicochemical characteristics were also determined on pyrodextrins. Both factors negatively affected the AS content, although positively influenced the ΔE (P < 0.05). The yellow pyrodextrin produced with 1.82 g/kg and heating for 307 min, presented physicochemical properties similar to the commercial
pyrodextrins from potato starch, with 46.6 % of AS, 24.5 of ΔE, high solubility and very low viscosity. The
pyrodextrinization caused a decrease of 30 − 54 % in AS content (P < 0.05), making these yam pyrodextrins a
promising material for water-soluble and very low viscous dietary fiber.

1. Introduction
The dietary changes towards an increase in fiber intake have been
associated with a reduction in the risk of developing chronic diseases
such as type-2 diabetes, cardiovascular disease and colorectal cancer
(Lockyer, Spiro, & Stanner, 2016). To fulfill the consumers' demand for
healthy processed foods, there is a growing interest in the food industry
in fortifying their products with dietary fiber and/or the physiological
analogs (Kapuśniak & Nebesny, 2017).
One of the analogous carbohydrates to dietary fiber is pyrodextrin.
Yellow pyrodextrins are heat-treated starches prepared under low
moisture content in an acidic environment (Laurentin, 2004). The

formation of indigestible fractions during the process requires the inclusion of mineral acid as catalyst when the pyrolysis is carried out at
temperatures below 200 °C and for relatively short periods (minutes to
several hours) (Kroh, Jalyschko, & Häseler, 1996; Laurentin, Cárdenas,
Ruales, Pérez, & Tovar, 2003; Wurzburg, 1986). The acidification is
commonly performed by spraying a diluted acid solution onto the dry
starch powder (Campechano-Carrera, Corona-Cruz, Chel-Guerrero, &
Betancur-Ancona, 2007; Laurentin et al., 2003; Luo et al., 2019), or by
adjusting pH of the starch slurry with a diluted acid solution, followed
by dehydration and pulverization before pyroconversion (Bai & Shi,
2016), or by distributing the acid in the form of gas (Lin, Lin, Zeng, Wu,
& Chang, 2018). A homogeneous distribution is of importance for

Abbreviations: ΔE, color difference; Adj-R2, adjusted R2; AS, available starch; CCRD, central composite rotational design; db, dry mass basis; DE, dextrose equivalent;
MW, molecular weight; Pyr, pyrodextrin; RSM, response surface methodology
⁎
Corresponding author.
E-mail addresses: (M. Lovera), (G.M.C.d. Castro), (N.d.R. Pires),
(M.d.S.R. Bastos), (M.L. Holanda-Araújo), (A. Laurentin),
(R.d.A. Moreira), (H.D.d. Oliveira).
1
Permanet address: Instituto de Biología Experimental, Facultad de Ciencias, Universidad Central de Venezuela, Apartado postal 47114, Caracas 1041-A,
Venezuela.
/>Received 29 October 2019; Received in revised form 24 April 2020; Accepted 26 April 2020
Available online 11 May 2020
0144-8617/ © 2020 Elsevier Ltd. All rights reserved.


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

nutritional characteristics, if sufficiently exploited, could generate interesting applications in both food and non-food industry (Amani,
Kamenan, Rolland-Sabaté, & Colonna, 2005; Otegbayo, Oguniyan, &
Akinwumi, 2014).
Response surface methodology (RSM) is a statistical method, used to
design experiments, build models, evaluate the effects of factors and
search optimum conditions of factors for desirable responses (Myers &
Montgomery, 1995). In RSM, a central composite rotational design
(CCRD) is useful to build a second-order polynomial model for the response variable without needing to use a complete three-level factorial
experiment, limiting the number of assays. When the experimental
space is unknown and non-preliminary experiment is performed, a
novel starch source could be tested using a CCRD, bringing additional
information about the relationship between the variables (dependent
and independent) selected.
Pyrodextrins from Dioscorea spp. have been studied only to a limited
extent, as well as the use of RSM to evaluate the effect of pyroconversion conditions on digestibility and physicochemical characteristics.
Therefore, this study aimed to optimize the pyroconversion of yam
(Dioscorea sp.) starch isolated from tubers grown in Brazil to produce a
yellow pyrodextrin with the lowest enzymatically available starch (AS)
content and color difference (ΔE) index, and dextrose equivalent (DE)
values < 10; and to evaluate the acid concentration and incubation
time effects on digestibility and some physicochemical properties of
pyrodextrins. This research provides a set of conditions for pyroconversion of yam starch and prepares yellow pyrodextrins with different in vitro digestibility and physicochemical properties, with potential applications in both food and non-food industry.

minimizes the undesired charring resulting from uneven catalysis
during the course of pyroconversion (Wurzburg, 1986).
The chemical reactions that occur during pyrodextrinization of
starch are complex and involve hydrolysis, transglucosydation, repolymerization, and oxidation (Bai & Shi, 2016; Wurzburg, 1986).
Transglucosydation reactions predominate in the formation of yellow
pyrodextrins and structural analyses have revealed the formation of

new glycosidic bonds, such as 1→2, 1→3 and 1→6 (Bai & Shi, 2016;
Laurentin, 2004; Luo et al., 2019; Okuma & Matsuda, 2002) in either αor β-anomers, with a simultaneous reduction of 1→4 linkages occurrence (Le Thanh-Blicharz, Blaszczak, Szwengiel, Paukszta, &
Lewandowicz, 2016). Also, an extensively branched structure but with
lower molecular size than starch molecule has been previously reported
(Bai & Shi, 2016; Han, Kang, Bai, Xue, & Shi, 2018).
The result of pyrodextrinization is a cold-water-soluble product,
with low or nil viscosity in solution, partially resistant to digestion or
indigestible, highly fermented in the colon and capable of affecting the
growth of probiotic bacteria, such as Lactobacillus and Bifidobacterium
strains, acting as prebiotics (Barczyńska, Śliżewska, Libudzisz,
Kapuśniak, & Kapuśniak, 2015; Laurentin & Edwards, 2004; Le ThanhBlicharz, Sip, Malcher, Prochaska, & Lewandowicz, 2015). In modified
starches by dextrinization, the pyroconversion has the advantage that
produces indigestible materials and fully soluble preparations that satisfy the condition to be a resistant starch, with the technological disadvantage that under pyrodextrinization a darker color is develop
compare to enzymatically-hydrolyzed and acid-modified starches
(maltodextrins and acidic hydrolysates, respectively), which requires
later processing of the final products.
Pyrodextrins are classified as a food additive [INS No. 1400] (FAO,
2001) and a food ingredient (EU Commission Directive, 2000). Products
based on pyrodextrins are found commercially, and they are industrially produced from corn, potato and wheat starches (LefrancMillot, 2008; Ohkuma, Hanno, Inada, Matsuda, & Katta, 1997). Because
their physicochemical, functional and nutritional properties, nonconventional starch sources such as lentil, sorghum, cocoyam, sagu,
cassava, and beans were also appropriated for preparing pyrodextrins
with high solubility, low viscosity and different ranges of indigestibility
(Campechano-Carrera et al., 2007; Laurentin et al., 2003), depending
on the pyroconversion condition and the origin of starch used.
Yam (Dioscorea spp.) starch is an excellent source of resistant starch
based on the low digestibility found in some species (Lovera, Pérez, &
Laurentin, 2017; Riley, Bahado-Singh, Wheatley, & Asemota, 2014) and
could be used as an alternative to other tuber and root starches, such as
potato or cassava, to produce pyrodextrins and potentials carbohydrates prebiotic. Recently, Luo et al. (2019) prepared pyrodextrins by
spraying the Chinese yam (Dioscorea opposita Thunb.) starch with a HCl

solution (0.01 M) in a final acid concentration of 0.07 ∼ 0.23 % (w/w)
on a dry starch basis, heating temperatures between 130 ∼ 170 °C,
and treatment times of 30 ∼ 120 min. These pyrodextrins presented
an enzyme-resistant fraction content in the range of 11.8 − 71.3 %, a
whiteness index of 80.7 − 34.5 % and a water-solubility between
14.3 − 98.4 %. These authors also obtained an enzyme-resistant dextrin produced by simultaneous pyroconversion and chemical modification of Chinese yam starch in the presence of an organic acid (citric
acid) as a modifying factor, which presented characteristics that make it
suitable for use in the soft drink industry as the soluble dietary fiber and
prebiotic in the beverages.
The yellow yam (D. cayennensis Lam. cultivar “inhame-da-Costa”)
together with the water yam (D. alata L. cultivar “inhame São Tomé”)
are the most widely cultivated species in the Northeast of Brazil.
Despite the wide distribution and its importance as a pharmacological
and food source, yams are referred to as an underutilized species
(Siqueira, 2011). Research in this field would further help the development of yams as a sustainable crop, as well as the processing of the
added value of their starches (Zhu, 2015). Dioscorea genus is a source of
native starches whose functional, physicochemical properties and

2. Materials and methods
2.1. Raw material and chemicals
Yam tubers (Dioscorea sp.) were purchased from the local market in
Fortaleza, Ceará State, grown in the Northeast of Brazil. Analytical
grade reagents and enzymes (EC: 3.2.1.1 and EC: 3.2.1.3) were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA).
2.2. Starch extraction and chemical characterization
The starch from yam tubers was extracted according to Pérez et al.
(2011) with minor variations. Briefly, one portion of the edible tuber
was blended in a Skymsen industrial blender (LAR-06MB, Santa Catarina, Brazil) for 2 min with twice their volume of distilled water and
sieving the collected homogenate with a nylon cloth. The grinding and
screening operation was repeated three more times. The resulting slurry
was centrifuged (Hitachi Koki, Himac CR-22 G III, Tokyo, Japan)

at 276 x g for 15 min, for easy separation of starch from the viscous
mucilage. The sediment was washed two times by suspension in distilled water and centrifugation, then was dried in a stove at 45 °C for
48 h. The starch was ground in a mini chopper (Philips Walita, Brazil),
screened through an 80-mesh sieve, and stored in a sealed plastic bag
inside a desiccator.
Native starch was analyzed for crude protein (N × 6.25 %), ash and
crude fat following methods described in AACC International (2003):
46–11A; 08–17 and 30–10, respectively. Amylose content was determined colorimetrically according to ISO (6647)-1 (2007) method.
The moisture content was determined for native starch and pyrodextrins by drying 2 g of a sample at 130 °C for 2 h. All analyses were
performed in triplicate.
2.3. Pyroconversion process
A CCRD 22 was performed with Statistica 10.0 software (StatSoft,
Inc.), using an α = ± 1.41 to generated 5 levels and 11 total assays (4
factorial points, 4 axial points and 3 repetitions in the central point, C).
2


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

starch, and using the Eq. (1):

Table 1
A CCRD 22 matrix for pyrodextrins production from yam starch.a.
Assays

Pyr-1
Pyr-2
Pyr-3

Pyr-4
Pyr-5
Pyr-6
Pyr-7
Pyr-8
Pyr-9 (C)
Pyr-10 (C)
Pyr-11 (C)
a

ΔE =

Independent variables real value (coded value)
Acid concentration (X1)
(g HCl/kg of starch, db)

Incubation time (X2)
(min)

0.99 (−1)
0.99 (−1)
2.65 (+1)
2.65 (+1)
0.65 (−α)
2.99 (+α)
1.82 (0)
1.82 (0)
1.82 (0)
1.82 (0)
1.82 (0)

90 (−1)
270 (+1)
90 (−1)
270 (+1)
180 (0)
180 (0)
53 (−α)
307 (+α)
180 (0)
180 (0)
180 (0)

(ΔL*)2 + (Δa*)2 + (Δb*)2

(1)

where ΔL*, Δa* and Δb* were the differences in the values of whiteness
(L*), redness-to-greenness (a*) and yellowness-to-blueness (b*), respectively.
2.4.3. Dextrose equivalent
The number of reducing groups on native starch and pyrodextrins
was determined as follows: 20 mL of such suspension (1 % w/v, db) was
gelatinized in a boiling water bath for 20 min. After cooled, the solution
was transferred into a 50 mL volumetric flask, made up to volume with
distilled water and mixed. The reducing sugars were measured according to the 3,5−dinitrosalisilic acid (DNS) method with 1 mL of
DNS reagent and an equal volume of gelatinized sample (Hostettler,
Borel, & Deuel, 1951). The DE was expressed as a percentage of the
reducing value of pure dextrose calculated on dry basis (Le ThanhBlicharz et al., 2016).

C, central point; Pyr, pyrodextrins.

The central point was a priori selected from the pyrodextrinization
standard condition tested by Laurentin et al. (2003) with a fixed
heating temperature of 140 °C, which reported a 52 % reduction
(P < 0.05) in the enzymatically AS content compare to the native
starch. The RSM was used to evaluate the isolated and combined effects
of acid concentration (X1) and incubation time (X2) on AS content, ΔE
index, and DE value of the pyrodextrins produced from native starch
(Table 1). The independent variables were expressed as a final acid
starch ratio (g of HCl/kg of starch, dry mass basis, db) and minutes.
The pyroconversion was performed according to Laurentin et al.
(2003) with modifications. Briefly, 22 g (db) of yam starch was placed
in a mortar and a specific volume (ranged from 0.18 to 0.82 mL) of
2.2 M HCl solution was sprayed on starch to a final acid starch ratio, as
showed in Table 1, mixed thoroughly and let sit overnight at room
temperature. Then, it was roasted in a stove at 140 °C for the corresponding time. The pyrodextrin was cooled to room temperature
for 30 min, ground, passed through 80-mesh sieve, and stored in a
closed plastic container inside a desiccator until use. The yield of pyrodextrinization was calculated as the mass ratio on dry basis of the
obtained pyrodextrin to native starch used and expressed as percentage.
AS content, ΔE index, and DE value of the pyrodextrins were determined (in triplicate), and the optimum pyrodextrinization condition
which produced the lowest AS content and ΔE index compared to the
native starch, and DE values < 10, was selected.

2.4.4. Water solubility
The water solubility was assayed for triplicate according to
Campechano-Carrera et al. (2007). Briefly, 40 mL of a 1 % (w/v, db)
starch suspension was prepared in a 50 mL centrifuge tube and placed
in a water bath at 25 °C for 30 min with agitation every 5 min. Then,
the solution was centrifuged at 2120 x g for 15 min, and 10 mL of supernatant was removed, placed in a constant-weight crucible and dried
in a stove at 120 °C for 4 h to obtain the weight of dissolved starch. The

water solubility was calculated as follows (Eq. (2)):

Water solubility= dry weight of dissolved starch/sample weight (db)×400 %
(2)
2.4.5. Rheological properties
The rheological properties were investigated on 40 % (w/v, db)
pyrodextrin solutions at 20 °C, except for treatment 9 and 10 (both
replicates of the central point), according to Le Thanh-Blicharz et al.
(2016) with modifications. After solubilized and before rheological
analysis, all pyrodextrin solutions were centrifuged at 2120 × g
for 15 min to remove particulate materials. The rheological behavior
of solutions was evaluated using a TA Instrument Rheometer (AR-550,
New Castle, USA) with cone and plate geometry (40 mm, 1° cone and
28 μm gap) at shear rate from 0 to 600 s−1. The rheological parameters
were calculated using the Herschel − Bulckley model employing the TA
Data Analysis software (TA instruments Inc., New Castle, USA), according to Eq. (3):

2.4. Digestibility and physicochemical characterization
2.4.1. Available starch
The AS content of native starch and pyrodextrins were assessed
following the multienzymatic protocol of Holm, Björck, Drews, and
Asp, (1986) with minor modifications. AS represents the digestible
fraction once the starch was gelatinized. A starch suspension (300 mg,
db dispersed in 20 mL of distilled water) was incubated for 20 min at
boiling temperature with 100 μL of a heat-stable α-amylase (200 U/
mL) from Bacillus licheniformis (A-3306: Sigma-Aldrich Co., USA) and an
aliquot was digested with amyloglucosidase from Aspergillus niger (A7095: Sigma-Aldrich Co., USA) at 60 °C for 30 min (3.5 U/mL). Released glucose was quantified with glucose oxidase/peroxidase colorimetric assay (Glicose Liquiform kit; Labtest Diagnóstica S. A, Brazil).
Standard calibration curve using glucose solution (0 to 0.8 mg/mL) was
constructed for the calculation of glucose released.

σ = σ0 + Ky n̊

(3)
−1

where σ (Pa), σ0 (Pa), K (mPa.s ), ẙ (s ), and n (dimensionless) were
the shear stress, the yield stress, the consistency coefficient, the shear
rate and the flow behavior index, respectively.
n

2.5. Statistical analysis
For the statistical analysis were applied Statistica 10.0 software
(StatSoft, Inc.). The response variables were fitted by a second-order
polynomial regression model, using a multiple regression analysis. The
quality of prediction and fitting of the polynomial model was evaluated
through the coefficient of determination R2, adjusted R2 (Adj-R2) and
analysis of variance (ANOVA). A lack-of-fit test was run and the adequate precision value was calculated for the three models to assist in
their validation (Germec, Ozcan, & Turhan, 2019; Noordin, Venkatesh,
Sharif, Elting, & Abdullah, 2004). To describe both isolated and combined effects of the independent variables of pyrodextrinization on the
three responses, 3D response surface plots were also developed. Data
were also analyzed using one-way ANOVA followed by Duncan's test as

2.4.2. Color parameters
Color parameters were measured according to Le Thanh-Blicharz
et al. (2016) by the reflection method on a Konica Minolta Chromameter (CR-410, Osaka, Japan) and the classification system of the
CIEL*a*b*. ΔE index was determined by comparison to the native
3


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

3.2. Yield and effects of pyroconversion process

Table 2
Chemical characterization of native starch isolated from
yam tubers grown in Brazil.a.
Component

Composition (%)

Moisture
Crude protein
Crude fat
Ash
Amylose
Available starch

13.8
0.71
0.24
0.21
9.33
96.6

±
±
±
±

±
±

The Table 3 presents the yield, the digestibility and some physicochemical characteristics of yellow pyrodextrins obtained by pyroconversion of yam starch. The yield of pyrodextrinization ranged from
97 to 99 % db, higher than reported by Falade and Ayetigbo (2015) for
acid hydrolyzed and acid modified yam starches. A relatively mild
treatment (< 1 g of HCl/kg of starch) such as Pyr-1 and Pyr-5 had no
impact on AS content (P > 0.05) but yielded markedly lighter products
(P < 0.05) than stronger treatment (< 2.5 g of HCl/kg of starch)
such as Pyr-3, Pyr-4 and Pyr-6, which were the less available pyrodextrins compared to native starch (P < 0.05). In general, a longer
heating time such as Pyr-8 led to a lower AS content than shorter
treatments such as Pyr-11 and Pyr-7, all with 1.82 g of HCl/kg of starch.
This tendency was also observed under condition of 0.99 g/kg, but interestingly, not with 2.65 g/kg. As expected, the yam starch was rapidly pyroconverted using a high acid concentration, and no significantly differences in AS content were found after increasing
incubation time from 90 min to 270 min (Pyr-3 and Pyr-4, respectively).
Luo et al. (2019) reported that in the first 60 min of pyroconversion, the
resistant fraction content in enzyme-resistant dextrins increases sharply
with the increasing the heating time, then, the increase slowly diminishes with longer treatment times.

0.2
0.28
0.04
0.02
0.18
2.1

a
Values (% db, except for moisture) are given as the
mean ± standard deviation of three replicates.

post hoc comparison of means and simple correlations (P < 0.05),

employing the same software.

3. Results and discussion
3.1. Starch extraction and chemical characterization
The chemical composition of yam starch extracted is showed in
Table 2. The moisture content (13.8 ± 0.2 %) was in the range reported for Dioscorea spp. and commercial starches (Jiang et al., 2012;
Swinkels, 1985), as were also the crude protein, ash and crude fat
contents (Amani et al., 2004; Nwokocha & Williams, 2011). The high
AS content reported reflect the presence of low levels of non-starch
components in this preparation (protein < 1 %, fat and ash < 0.3 %),
and it is indicative of adequate purity level. The total starch content in
yam starch (not determined) could be expected to be similar to the
recorded AS content because raw starches do not contain retrograded
fractions that would decrease AS content, as reported by Lovera et al.
(2017).
The apparent amylose content of yam starch was lower than those
reported for Dioscorea spp., which are in the range of 13.58 − 20.05 %
(amylopectin: 79.95 − 86.42 % vs. 90.67 %) (Jiang et al., 2012). The
amylose content recorded here was also quantified by the iodine
binding-colorimetric method but using non-defatted samples, and this
usually gives lower amylose contents than defatted samples, because
lipids can form inclusion complexes with amylose (Zhu, 2015). This low
amylose content was consistent with a high digestibility of yam starch,
as reported by Riley et al. (2014) for low-amylose yam starches.

3.2.1. Available starch
Pyrodextrins AS content was always lower (P < 0.05) than native
starch, with the exception of Pyr-1 and Pyr-5. Yam pyrodextrins produced were grouped into four categories based on their reduction in AS
content compared to native starch, an easier way to estimate the indigestible fraction reported by Laurentin et al. (2003). These groups
were: those treatments with the highest reduction in digestibility with

52 − 54 % (Pyr-3, Pyr-4, Pyr-6 & Pyr-8), the intermediary group with
44 − 48 % (Pyr-9, Pyr-10 & Pyr-11), followed by the group with a 30 %
decrease in AS (Pyr-2 & Pyr-7) and finally, those with the smallest fall in
AS content and statistically similar (P > 0.05) to native yam starch
(Pyr-1 & Pyr-5). Pyrodextrin AS content was negatively correlated with
ΔE index, DE value and water solubility of sample
(rΔE = −0.844; rDE = −0.850; rS = −0.798; P < 0.05; n = 33).
The acid concentration, incubation time and their interaction had
significant influence (P < 0.05) on AS content of pyrodextrins (Table
S1). The acid concentration had the largest influence on the AS content
of pyrodextrins, being linear and second order coefficients significant.
However, the acid concentration and incubation time linear terms negativity affected the AS content, although the interaction had a positive
effect on this variable (Fig. 1a). The mathematical model (R2 = 0.9861;

Table 3
Yield, digestibility and physicochemical characterization of yellow pyrodextrins produced from yam starch.1,2,3.
Pyrodextrins

Pyr-1
Pyr-2
Pyr-3
Pyr-4
Pyr-5
Pyr-6
Pyr-7
Pyr-8
Pyr-9 (C)
Pyr-10 (C)
Pyr-11 (C)
Native starch

2

Yield

AS

Fall in

(%, db)

(%, db)

AS (%)

97.4
98.3
97.5
97.5
97.2
97.1
97.0
98.0
97.9
97.7
98.5
−

95.4 ± 2.9a
67.4 ± 1.7b

44.1 ± 1.7d
44.2 ± 3.2d
86.2 ± 6.9a
44.2 ± 1.4d
67.6 ± 6.3b
46.6 ± 2.3d
49.8 ± 5.5c, d
53.4 ± 5.3c
54.2 ± 3.9c
96.6 ± 2.1a

1.2
30.2
54.4
54.3
10.7
54.2
30.0
51.7
48.4
44.7
43.9
−

Color parameters
L*
92.3
93.4
60.4
63.5

91.4
56.3
83.6
83.4
80.2
80.4
76.7
94.6

± 0.03c
± 0.04b
± 0.51h
± 0.51g
± 0.16c
± 0.78i
± 0.22d
± 0.46d
± 0.09e
± 0.15e
± 0.13f
± 0.62a

DE
a*

b*

ΔE

(%, db)

0.11 ± 0.00g
−0.28 ± 0.00h
5.3 ± 0.04c
5.8 ± 0.06b
0.49 ± 0.07e
6.0 ± 0.12a
−0.46 ± 0.07i
0.43 ± 0.04e
1.2 ± 0.02d
1.2 ± 0.03d
1.2 ± 0.06d
0.29 ± 0.07f

7.3 ± 0.07j
10.0 ± 0.03i
30.5 ± 0.24b
31.4 ± 0.18a
9.6 ± 0.27i
28.8 ± 0.32c
20.4 ± 0.23h
25.1 ± 0.06d
23.6 ± 0.11f
24.7 ± 0.24e
21.3 ± 0.12g
4.0 ± 0.46k

4.0 ± 0.0h
6.2 ± 0.0g
43.5 ± 0.2b

41.7 ± 0.3c
6.5 ± 0.2g
45.9 ± 0.5a
19.7 ± 0.3f
23.9 ± 0.2e
24.3 ± 0.1e
25.0 ± 0.3d
24.9 ± 0.1d
−

2.8 ± 0.0d
3.4 ± 0.0c
7.4 ± 0.1a
7.3 ± 0.1a
1.6 ± 0.0f
7.5 ± 0.2a
7.3 ± 0.1a
6.6 ± 0.1b
6.4 ± 0.1b
6.6 ± 0.1b
6.6 ± 0.2b
1.8 ± 0.1e

1
Values are means ± standard deviation of triplicate analysis. Means in columns not sharing same superscript letters are significantly different (P < 0.05; Duncan's
test).
2
Fall in AS content was calculated as (AS native – AS pyrodextrinized) × 100/AS native.
3
ΔE, color difference; AS, available starch; C, central point; db, dry mass basis; DE, dextrose equivalent; Pyr, pyrodextrins.

4


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

Fig. 1. Pareto chart of standardized effects of the acid concentration (X1) and incubation time (X2) on pyrodextrins characteristics: (a) AS content, (b) ΔE index and
(c) DE value.

Adj-R2 = 0.9721; P < 0.05) that illustrates the behavior of the AS
content, in terms of the coded independent variables is represented by
Eq. (4):

AS (%) =178.08–73.10 X1 +9.88 X12 –0.372 X2 +0.00034 X22 +0.094 X1 X2

color and were characterized by smaller values of the whiteness (L*)
and higher values of a* and b* parameters than native yam starch
(Table 3). The lightest and darkest materials were Pyr-5 and Pyr-6,
respectively. An inverse correlation between L* and parameters a* and
b* (ra* = −0.944; rb* = −0.872; P < 0.05; n = 33) and a positive relationship between a* and b* (r = 0.744; P < 0.05; n = 33) were
found. These correlations indicated a predominance of red and yellow
colors in the pyrodextrins and reinforce the idea of the progress of the
caramelization. The whiteness of pyrodextrins decreased with the increasing concentrations of HCl, which was consistent with the reduction
of the whiteness degree reported for pyrodextrins prepared from Chinese yam starch using the same acid (Luo et al., 2019).
The acid concentration, incubation time and their interaction had
significant influence (P < 0.05) on pyrodextrin ΔE index (Table S2).
Similar to AS response, significance analysis on coefficients of each
factor showed that the acid concentration had the largest influence on

the ΔE index of pyrodextrins, linear and second order coefficients were
significant. Incubation time had a smaller but significant effect on the
ΔE index, linear and quadratic coefficients were also significant. Linear
terms of acid concentration and incubation time positively affected the
ΔE index, although the interaction had a negative effect on this variable
(Fig. 1b). The final estimative regression model (R2 = 0.9776; AdjR2 = 0.9552; P < 0.05) for the ΔE index of pyrodextrins is represented by Eq. (5):

(4)

Starch pyroconversion produced a decrease in AS content as reported previously by different authors (Campechano-Carrera et al.,
2007; Laurentin et al., 2003). Laurentin et al. (2003) reported a
55 − 65 % decrease in digestibility for pyrodextrins from cassava
starch, using the same pyroconversion conditions as for Pyr-1, Pyr-2,
Pyr-3, Pyr-4 and the central point, although spraying a constant volume
(0.5 mL) of the acid solution onto the starch powder. These authors
reported that non-digestible fractions produced by pyroconversion
differ depending on the starch source. It was reported for yam starch
that a low amylose content could predispose the starch to acidic and
thermal resistance (Amani et al., 2005), this suggests that it would need
a more extreme pyroconversion condition to decrease its digestibility.
In general, the pyroconversion results in a substantial reduction of
digestibility relative to that of the native starch (Laurentin et al., 2003;
Le Thanh-Blicharz et al., 2016; Lin et al., 2018). Pyrodextrins prepared
using similar conditions as here have been structurally characterized
(Bai & Shi, 2016; Laurentin, 2004; Le Thanh-Blicharz et al., 2016; Luo
et al., 2019; Okuma & Matsuda, 2002), reporting the presence of 1→2,
1→3 and 1→6 glycosidic linkages in either α- or β-anomers, with a
reduction of 1→4 bonds occurrence. Similar glycosidic linkage patterns
are expected to be formed under the conditions tested (Table 1), independently of starch source used. The indigestibility of pyrodextrins
has been mainly attributed to the formation of these atypical linkages

during transglucosydation reactions (Siljeström, Björck, & Westerlund,
1989), because the retrograded starch is very low for both native and
modified samples, according to Laurentin et al. (2003).

ΔE =–19.81+18.43 X1 +1.02 X12 +0.100 X2 –0.00019 X22 -0.013 X1 X2

(5)

The color difference shows its direct dependence on pyrodextrinization conditions, where extreme conditions lead to more variation
in color, as it was reported elsewhere (Campechano-Carrera et al.,
2007; Laurentin et al., 2003; Lin et al., 2018; Terpstra, Woortman, and
Hopman, (2010) produced yellow pyrodextrins by heating potato
starch for 5 h at 165 °C (with ∼1.85 g of HCl/kg of starch) and they
reported the presence of significantly different aggregate-like structures
made of small starch fragments which were intensely colored, physically linked, and probably still susceptible to repolymerization and

3.2.2. Color parameters
The Fig. 2 presents the visual aspects of native yam starch and
pyrodextrins. All pyrodextrins presented a cream or yellowy-brownish
5


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

Fig. 2. Native yam starch and pyrodextrins visual aspects from CCRD 22 matrix.

desired as significant lack-of-fit indicates that there might be contributions in the regressor–response relationship that are not accounted
for by the model (Noordin et al., 2004). Nevertheless, the adequacy of

the models was also investigated calculating the signal to noise ratio or
adequate precision, which compares the range of the predicted values
at the design points to the average prediction error, and ratios greater
than 4 indicate adequate model discrimination (Noordin et al., 2004).
For the three responses AS, ΔE and DE, the adequate precision values
were 23.1, 18.7 and 23.4, respectively; indicating that there were
adequate signals. Therefore, the models for responses were reliable and
can be used to navigate the design space, as reported by Germec et al.
(2019). Therefore, the predictive models (Eqs 4–6) were successful to
precisely predict the AS, ΔE and DE in pyrodextrinization of yam starch.
Eqs. (4)–(6) are plotted in Fig. 3a–c, respectively as 3D response
surface plots. As can be observed from Fig. 3a, increasing the acid
concentration and incubation time caused a decrease in the AS content
of pyrodextrins. However, at the pyrodextrinization conditions used in
this study, the response reached a minimum value instead of reaching
the optimum value. The critic values to obtain a minimum AS content of
41.6 % were 3.18 g/kg of HCl and 109 min of heating. For ΔE index and
DE value (Fig. 3b & c, respectively) a high acid concentration caused an
increases in both responses, reaching a saddlepoint value instead of
optimum value. This behavior was more evident with response DE
value, with a predicted value of 7.5 % using 2.59 g/kg of HCl and
215 min of incubation time.
Pyr-3, Pyr-4, Pyr-6 and Pyr-8 were the less digestible and the
darkest pyrodextrins, with the exception of Pyr-8. According to these
results, the use of 1.82 g/kg acid concentration with 307 min of heating
time at 140 °C, appears to be an appropriate condition for preparing a
yam pyrodextrin with the lowest AS content and ΔE index. Even though
Pyr-8 had a slightly higher AS content (P > 0.05) than Pyr-3, Pyr-4,
and Pyr-6, it was selected as the optimum pyrodextrinization treatment
because produced a lighter pyrodextrin (Table 3 & Fig. 2). Pyr-8 uses

the highest heating time with an acid concentration not more than
2.0 g/kg and a temperature commonly used for pyroconversion (Table 1
& Fig. 2). In this work, the central point of CCRD design was Laurentin

transglucosydation. These changes would explain the low digestibility
found in Pyr-8 with similar pyroconversion conditions.
Pyrodextrin ΔE index was positively correlated with DE and water
solubility (rDE = 0.846; rS = 0.577; P < 0.05; n = 33), although
negatively correlated with AS content; similar results were found by Le
Thanh-Blicharz et al. (2016) for commercial potato pyrodextrins.
3.2.3. Dextrose equivalent
The pyroconversion process significantly increased DE values
(Table 3), with exception of the Pyr-5 (slightly lower than native starch;
P < 0.05), suggesting that this pyrodextrin was poorly hydrolyzed.
Pyrodextrins from corn and potato starches according to Ohkuma,
Hanno, Inada, Matsuda, & Katta, 1997 are preferably between 1 and 10
in DE, such as reported here. Nevertheless, the reducing sugar quantification was assayed on gelatinized samples, and since pyrodextrins
were not neutralized after pyroconversion, the possibility that digestion
had continued during gelatinization cannot be discarded.
Only acid concentration, but not the incubation time had significant
influence (P < 0.05) on DE value (Table S3 and Fig. 1c). Both linear
and quadratic increases on acid concentration affected the DE value.
The determination coefficients (R2=0.9784; Adj-R2=0.9639;
P < 0.05), exhibited an excellent experimental result which fits the
mathematical model proposed for DE value, represented by Eq. (6):

DE (%) =–3.09+8.51 X1 –1.64 X12 –0.0042 X2 +0.00001 X22

(6)

The decrease in DE values at higher acid concentrations was probably associated to thermal decomposition and/or oxidation process of
reducing end. Le Thanh-Blicharz et al. (2016) reported that degree of
oxidation is restrained and surprisingly independent of the type of
pyrodextrin. In this study, DE was positively correlated with ΔE index
and water solubility (rS = 0.863; P < 0.05; n = 33), although negatively correlated with AS content.
The coefficients of determination R2 and Adj-R2 for the three
quadratic models were higher than 0.95, which is desired. On the other
hand, except for ΔE index the probability value of the lack of fit test
were insignificant (Table S1, S2 and S3). Insignificant lack-of-fit is
6


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

Fig. 3. 3D response surface plot for pyrodextrins characteristics in function of acid concentration and incubation time. (a) AS content, (b) ΔE index and (c) DE value.

be compatible with beverage products.
The 40 % pyrodextrin solutions showed pseudoplastic flow behavior
index (Table 4 and Table S4); confirming the non-Newtonian flow of
concentrated solutions, whose viscosity decreases with increasing shear
rate. Le Thanh-Blicharz et al. (2016) reported a Newtonian behavior in
four commercial pyrodextrins even solubilized at 40 % (w/v). Differences between these studies suggest that yam pyrodextrin fluids could
be composed of intermediary and high-molecular weight (MW) substances instead of predominantly low-MW sugars.
In general, all pyrodextrins presented very low yield stress (which
indicate a smaller resistance to flow) and shear viscosity (Table 4 and
Table S4). The drop in viscosity is not a surprise as native starch is
rapidly broken down by acid and heat (Terpstra et al., 2010), even at
short reaction time of pyroconversion such in Pyr-7 (53 min).

Table 4 shows that pyrodextrins from yam starch did not differ
much from four commercial yellow pyrodextrins, obtained by dry
roasting of potato starch with an acid catalyst. Pyr-8 was the most similar to potato pyrodextrins. The almost nil viscosity found with a high
content of dry mass could be considered a better functional property for
specific applications, for example, to provide a non-viscous dietary fiber
to patients under enteral feeding. Even in cooked food preparations
could be used because after gelatinization pyrodextrins solutions remain with very low viscosity (data not showed). The shear behavior
also found is often required in cosmetic products like emulsions, suspensions and for encapsulating material. These features make these
pyrodextrins suitable for use in the beverage and food industry, as well
as, a promising material for the cosmetic, pharmaceutical and paper
industries.

et al. (2003) pyroconversion condition which also proved to be efficient
to produce yellow pyrodextrins with similar appearance and digestibility than Pyr-8, but in a shorter time, which is technologically more
advantageous.
3.2.4. Water solubility and rheological properties of pyrodextrins
All pyrodextrins presented a water solubility < 95 %, with the exception of Pyr-1 (58 ± 2 %) and Pyr-5 (32 ± 1 %), and were similar
to commercial yellow pyrodextrins from potato starch (Table 4). Pyroconversion considerably increases the solubility as a consequence of
the increase in low molecular linear fractions (Campechano-Carrera
et al., 2007; Luo et al., 2019). These water-soluble pyrodextrins would
Table 4
Comparative between in vitro digestibility and physicochemical properties of
potato (commercial) and yam pyrodextrins.1,2.
Parameter

Digestibility
(Fall in AS; %)
ΔE
L*
a*

b*
DE (%)
Water solubility (%)
Moisture (%)
Rheological parameters
Yield stress, σ0 (Pa)
Consistency coefficient, K (mPa.sn)
Flow behavior index, n
(dimensionless)
Standard error, S.E

Pyrodextrins
Potato1

Yam2

Pyr-8

40.6 − 53.9
(46 − 59 %)
15.8 − 25.4
82.7 − 87.4
−0.62 − 0.70
16.2 − 24.6
1.6 − 2.4
> 95
nd

44.1 − 67.6
(30 − 54 %)

6.2 − 45.9
56.3 − 93.4
−0.46 − 6.0
10.0 − 31.4
3.4 − 7.5
> 95
1.3 − 3.2

46.6
(52 %)
23.9
83.4
0.43
25.1
6.6
> 95
2.0

nd
80 − 240
nd

0.251 − 0.717
16 − 132
0.768 − 0.969

0.474
56
0.827

nd

1.5−9.0

2.0

4. Conclusions
To the best of the authors’ knowledge, this paper reports the first
documented pyrodextrinization of yam (Dioscorea sp.) starch isolated
from tubers grown in Brazil. Under the studied experimental conditions, pyroconversion resulted in 30 − 54 % decreases of AS content,
with changes in their physicochemical properties. The use of
1.82 g HCl/kg of starch acid concentration and ∼ 5 h of heating at

nd, not determined; Pyr, pyrodextrin.
1
Four yellow pyrodextrins produced from potato starch by WPPZ SA (Lubón,
Poland) as reported by Le Thanh-Blicharz et al. (2016).
2
The ranges exclude pyrodextrins with AS content similar to the native starch
(Pyr-1 and Pyr-5; Table 3).
7


Carbohydrate Polymers 242 (2020) 116382

M. Lovera, et al.

140 °C was appropriate for preparing a yellow pyrodextrin (Pyr-8) with
simultaneously low AS content and ΔE index, and similar physicochemical properties to the commercial pyrodextrins from potato starch,
which is a promising material for water-soluble and very low viscous

dietary fiber. Further research is needed to elucidate the molecular
structure of these pyrodextrins, as well as the nutritional effects as
potential dietary fiber and carbohydrate prebiotic.

FAO (2001). Modified starches. Compendium of food additive specifications. Addendum 9.
Joint FAO/WHO expert committee on food additives, 57th session, Rome. FAO food and
nutrition paper - 52 add. 9.
Germec, M., Ozcan, A., & Turhan, I. (2019). Bioconversion of wheat bran into high valueadded products and modelling of fermentations. Industrial Crops and Products, 139,
111565.
Han, X., Kang, J., Bai, Y., Xue, M., & Shi, Y. C. (2018). Structure of pyrodextrin in relation
to its retrogradation properties. Food Chemistry, 242, 169–173.
Holm, J., Björck, I., Drews, A., & Asp, N.-G. (1986). A rapid method for the analysis of
starch. Starch‐Stärke, 38, 224–226.
Hostettler, F., Borel, E., & Deuel, H. (1951). Über die reduktion der 3,5-dinitrosalicylsäure
durch Zucher [About the reduction of 3,5-dinitrosalicylic acid by sugars]. Helvetica
Chimica Acta, 34, 2132–2139.
ISO 6647-1 (2007). International Organization for Standardization. Rice – Determination of
amylose content – Part 1: Reference method.
Jiang, Q., Gao, W., Li, X., Xia, Y., Wang, H., Wu, S., et al. (2012). Characterizations of
starches isolated from five different Dioscorea L. species. Food Hydrocolloids, 29,
35–41.
Kapuśniak, K., & Nebesny, E. (2017). Enzyme-resistant dextrins from potato starch for
potential application in the beverage industry. Carbohydrate Polymers, 172, 152–158.
Kroh, L. W., Jalyschko, W., & Häseler, J. (1996). Non-volatile reaction products by heatinduced degradation of α-glucans. I: Analysis of oligomeric maltodextrins and anhydrosugars. Starch‐Stärke, 48, 426–433.
Laurentin, A. (2004). Starch pyrodextrins: In vitro fermentation and physiological effects. PhD
thesis. Glasgow University.
Laurentin, A., & Edwards, C. A. (2004). Differential fermentation of glucose based carbohydrates in vitro by human faecal bacteria: A study of pyrodextrinised starches
from different sources. European Journal of Nutrition, 43, 183–189.
Laurentin, A., Cárdenas, M., Ruales, J., Pérez, E., & Tovar, J. (2003). Preparation of indigestible pyrodextrins from different starch sources. Journal of Agricultural and Food
Chemistry, 51, 5510–5515.

Le Thanh-Blicharz, J., Blaszczak, W., Szwengiel, A., Paukszta, D., & Lewandowicz, G.
(2016). Molecular and supermolecular structure of commercial pyrodextrins. Journal
of Food Science, 81, 2135–2142.
Le Thanh-Blicharz, J., Sip, A., Malcher, A., Prochaska, K., & Lewandowicz, G. (2015). The
effect of surface activity of pyrodextrins on their assimilability by selected strains of
bacteria from genus Lactobacillus. Starch‐Stärke, 67, 267–275.
Lefranc-Millot, C. (2008). NUTRIOSE® 06: A useful soluble dietary fiber for added nutritional value. Nutrition Bulletin, 33, 234–239.
Lin, C.-L., Lin, J.-H., Zeng, H.-M., Wu, Y.-H., & Chang, Y.-H. (2018). Indigestible pyrodextrins prepared from corn starch in the presence of glacial acetic acid.
Carbohydrate Polymers, 188, 68–75.
Lockyer, A., Spiro, S., & Stanner, S. (2016). Dietary fiber and the prevention of chronic
disease–should health professionals be doing more to raise awareness? Nutrition
Bulletin, 41, 214–231.
Lovera, M., Pérez, E., & Laurentin, A. (2017). Digestibility of starches isolated from stem
and root tubers of arracacha, cassava, cush-cush yam, potato and taro. Carbohydrate
Polymers, 176, 50–55.
Luo, S. Z., Wu, X. Z., Xu, P. L., Pan, L. H., Zheng, Z., Cao, L. L., et al. (2019). Enzymeresistant dextrin from Chinese yam starch for potential application in beverage industry: Preparation, physicochemical properties and in vitro digestion. Current Topics
in Nutraceutical Research, 17, 140–147.
Myers, R. H., & Montgomery, D. C. (1995). Response surface methodology: Process and
product optimization using designed experiments. New York: John Wiley and Sons.
Noordin, M. Y., Venkatesh, V. C., Sharif, S., Elting, S., & Abdullah, A. (2004). Application
of response surface methodology in describing the performance of coated carbide
tools when turning AISI 1045 steel. Journal of Materials Processing Technology, 145,
46–58.
Nwokocha, L. M., & Williams, P. A. (2011). Comparative study of physicochemical
properties of breadfruit (Artocarpus altilis) and white yam starches. Carbohydrate
Polymers, 85, 294–302.
Ohkuma, K., Hanno, Y., Inada, K., Matsuda, I., & Katta, Y. (1997). Indigestible dextrin. U.S.
Patent 5.620.873.
Okuma, K., & Matsuda, I. (2002). Indigestible fraction of starch hydrolysates and their
determination method. Journal of Applied Glycoscience, 49, 479–485.

Otegbayo, B., Oguniyan, D., & Akinwumi, O. (2014). Physicochemical and functional
characterization of yam starch for potential industrial applications. Starch‐Stärke, 66,
235–250.
Pérez, E., Gilbert, O., Rolland-Sabaté, A., Jimenéz, Y., Sánchez, T., Giraldo, A., et al.
(2011). Physicochemical, functional and macromolecular properties of waxy yam
starches discovered from “Mapuey” (Dioscorea trifida) genotypes in the Venezuelan
Amazon. Journal of Agricultural and Food Chemistry, 59, 263–273.
Riley, C. K., Bahado-Singh, P. S., Wheatley, A. O. B., & Asemota, H. N. (2014).
Physicochemical properties of low-amylose yam (Dioscorea spp.) starches and its
impact on α-amylase degradation in vitro. International Journal of Nutrition and Food
Sciences, 3, 448–454.
Siljeström, M., Björck, I., & Westerlund, E. (1989). Transglycosidation reactions following
heat treatment of starch–effects on enzymatic digestibility. Starch‐Stärke, 41, 95–100.
Siqueira, M. V. B. M. (2011). Yam: A neglected and underutilized crop in Brazil.
Horticultura Brasileira, 29, 16–20.
Swinkels, J. J. M. (1985). Composition and properties of commercial native starches.
Starch-Stärke, 37, 1–5.
Terpstra, K. R., Woortman, A. J., & Hopman, J. C. (2010). Yellow dextrins: Evaluating
changes in structure and colour during processing. Starch‐Stärke, 62, 449–457.
Wurzburg, O. B. (1986). Converted starches. In O. B. Wurzburg (Ed.). Modified starches:
Properties and uses (pp. 17–40). Boca Raton, FL, USA: CRC Press.
Zhu, F. (2015). Isolation, composition, structure, properties, modifications, and uses of
yam starch. Comprehensive Reviews in Food Science and Food Safety, 14, 357–386.

CRediT authorship contribution statement
Mighay Lovera: Conceptualization, Methodology, Validation,
Formal analysis, Investigation, Data curation, Visualization, Writing original draft, Writing - review & editing, Project administration.
George Meredite Cunha de Castro: Methodology, Formal analysis.
Natalia da Rocha Pires: Investigation, Resources. Maria do Socorro
Rocha Bastos: Resources, Funding acquisition. Márjory Lima

Holanda-Araújo: Conceptualization, Resources, Visualization, Writing
- original draft. Alexander Laurentin: Conceptualization, Formal
analysis, Visualization, Writing - original draft, Writing - review &
editing. Renato de Azevedo Moreira: Conceptualization, Resources,
Supervision, Funding acquisition, Project administration. Hermógenes
David de Oliveira: Conceptualization, Resources, Writing - original
draft, Supervision, Funding acquisition, Project administration.
Declaration of Competing Interest
The authors declare no conflicts of interest.
Acknowledgements
The authors gratefully acknowledge support from the Brazilian
Federal Agency for Support and Evaluation of Graduate Education
(Coordenaỗóo de Aperfeiỗoamento de Pessoal de Nível Superior –
CAPES), and the PAEC OAS-GCUB (Partnerships Program for Education
and Training between the Organization of American States and the
Coimbra Group of Brazilian Universities) for the international scholarship awarded to the first author. The authors also thank to Embrapa
Agroindústria Tropical for the technical support. Lovera also acknowledges her leave of absence from Central University of Venezuela.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi: />References
AACC International (2003). Approved methods of analysis (11th ed.). St Paul, MN: AACC
International.
Amani, N. G. G., Buleon, A., Kamenan, A., & Colonna, P. (2004). Variability in starch
physicochemical and functional properties of yam (Dioscorea sp.) cultivated in Ivory
Coast. Journal of the Science of Food and Agriculture, 84, 2085–2096.
Amani, N. G. G., Kamenan, A., Rolland-Sabaté, A., & Colonna, P. (2005). Stability of yam
starch gels during processing. African Journal of Biotechnology, 4, 94–101.
Bai, Y., & Shi, Y.-C. (2016). Chemical structures in pyrodextrin determined by nuclear
magnetic resonance spectroscopy. Carbohydrate Polymers, 151, 426–433.
Barczyńska, R., Śliżewska, K., Libudzisz, Z., Kapuśniak, K., & Kapuśniak, J. (2015).

Prebiotic properties of potato starch dextrins. Postepy Higieny I Medycyny
Doswiadczalnej (Online), 69, 1031–1041.
Campechano-Carrera, E., Corona-Cruz, A., Chel-Guerrero, L., & Betancur-Ancona, D.
(2007). Effect of pyrodextrinization on available starch content of lima bean
(Phaseolus lunatus) and cowpea (Vigna unguiculata) starches. Food Hydrocolloid, 21,
472–479.
Commission Directive (2000). 2000/63/EC amending Directive 96/77/EC laying down
specific purity criteria on food additives other than colors and sweeteners. Official
Journal of the European Communities L, 277(43), 1–61 Available from: https://eur-lex.
europa.eu/legal-content/EN/TXT/?uri=OJ:L:2000:277:TOC. (Accessed 2019
June 26).
Falade, K. O., & Ayetigbo, O. E. (2015). Effects of annealing, acid hydrolysis and citric
acid modifications on physical and functional properties of starches from four yam
(Dioscorea spp.) cultivars. Food Hydrocolloids, 43, 529–539.

8