LWT - Food Science and Technology 128 (2020) 109465

Contents lists available at ScienceDirect

LWT - Food Science and Technology

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

Comparative study of physicochemical and functional properties of pan and T

microwave cooked underutilized millets (proso and little)
Simmi Ranjan Kumara, Muhammad Bilal Sadiqb, Anil Kumar Anala,∗
a
Food Engineering and Bioprocess Technology, Department of Food, Agriculture and Bioresources, Asian Institute of Technology, Pathum Thani, 12120, Thailand
b
School of Life Sciences, Forman Christian College (A Chartered University), Lahore, 54600, Pakistan

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

Keywords: Physiochemical properties such as colour, water absorption index (WAI), water solubility index (WSI) bulk
Underutilized millets density and the functional properties such as water absorption capacity (WAC), oil absorption capacity, least
Microstructure gelation concentration, foaming capacity, foaming stability, pasting properties, in vitro starch digestibility and
Functional properties microstructure of underutilized millet (proso and little) flours were evaluated. The millets flour exhibited
In vitro digestibility
considerable amounts of carbohydrates (78.37–81.69 g/100 g) and proteins (4.48–10.32 g/100 g). Microwave
cooked little millet (MWLM) (10.32 g/100 g), and pan-cooked proso millet (PCPM) (9.07 g/100 g) flour con-
tained a significantly higher amount of proteins than the raw millets. WAI (4.48 g/g) and WSI(4.17 g/g) of
MWLM were significantly higher than the other flours. The functional properties of MWPM such as WAC (3.23 g/
g) and OAC (0.90 g/g) were higher than the other flour samples. The microstructure of flour samples revealed
that the starch granules of raw flours were smooth, oval and granular structure whereas cooked flours have
deshaped starch granules. The under-utilized millets can further be used as an essential source of nutrients and
functional ingredients in various food formulations, including for functional foods.

1. Introduction Millets are traditionally consumed as a food which is prepared by

various processing methods such as boiling, roasting and pressure
Millets are small-seeded grasses from the family Poaceae, cultivated cooking (Annor, Tyl, Marcone, Ragaee, & Marti, 2017). There are many
on marginal drylands in subtropical and tropical regions. Millets are factors which affect the characteristics of cooked grain such as the pre-
rich in carbohydrates, protein, fat, dietary fibres, minerals, flavonoids processing applied, size of the grain, water/grain ratio, cooking time,
and some of the vitamins (Chandrasekara & Shahidi, 2010). Most of cooking method and temperature (Dias-Martins et al., 2019). Micro-
these millets are used as a staple food, and as one of the main in- wave is an electro-heat technique which converts electrical energy into
gredients in the preparation of various traditional foods such as por- thermal energy on a frequency range of 300 MHz to 300 GHz, and food
ridges, idli, dosa, bread, chakli, papads etc (Sarita & Singh, 2016). Proso application are in the range of 915 MHz and 2.45 GHz (Gavahian et al.,
millet (Panicum miliaceum L.), and little millet (Panicum sumatrense) 2019). Due to the elctro-heat technique effect, the microwave method is
both are widely cultivated throughout Asia, Australia, North America, a more efficient energy saver and short process than conventional such
Europe, and Africa (Yang et al., 2018). Little millet contains proteins as open pan boiling, which uses an indirect heat transfer mechanism
(9.80–12.49 g/100 g), fat (2.87–5.09 g/100 g), ash (0.98–4.78 g/ (Ekezie, Sun, Han, & Cheng, 2017). Microwave cooking is considered as
100 g), crude fibre (0.49–8.72 g/100 g), carbohydrates (62.25–76.59 g/ a quick household cooking method as compared to conventional pan
100 g) with various other essential minerals and phytochemicals and cooking. Liu, Zheng, Wang, and Chen (2019) reported that microwave
does not contain gluten (Tiwari, Tiwari, & Tripathi, 2018). Proso millet cooking resulted in a least loss of vitamins in rice as compared to high
contains proteins (9.4–9.9 g/100 g), fat (1.2–3.8 g/100 g), ash pressure and conventional pan cooking techniques. In comparison to
(0.6–3.3 g/100 g) and carbohydrates (70.0–74.0 g/100 g) with various conventional cooking techniques, microwave facilitates to achieve high
other essential minerals and phytochemicals (Devisetti, Yadahally, & temperature in short time by rapid vibrations of polar molecules, uni-
Bhattacharya, 2014). Although millets are nutritionally superior, lack form heating, high heating rates and safe handling (Guo, Sun, Cheng, &
of refined and processed millets limit their extensive use and accept- Han, 2017).
ability. This study aimed to explain the effects of different cooking

∗
Corresponding author. Department of Food, Agriculture and Bioresources, Asian Institute of Technology, Pathum Thani, 12120, Thailand.
E-mail addresses: [email protected], [email protected] (A.K. Anal).

https://doi.org/10.1016/j.lwt.2020.109465
Received 28 September 2019; Received in revised form 25 November 2019; Accepted 18 April 2020
Available online 28 April 2020
0023-6438/ © 2020 Elsevier Ltd. All rights reserved.
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

protocols, including open cooking, and microwave cooking on little and 2.4. Physical properties of flours
proso millet. The in vitro digestibility physiochemical and functional
properties of proso millet and little millet were also evaluated after pan 2.4.1. Color characteristics
and microwave cooking. None of the reports has yet been reported in The colour of the samples was analyzed by Hunter Lab
details on the variations in physicochemical and functional properties Spectrocolorimeter (Model TC-P III-A, Tokyo Denshoku Co., Ltd.,
of underutilized millets after treating with different cooking methods. Japan) by following the method of Medhe, Jain, and Anal (2019). The
colorimeter was calibrated by using Hunter lab color standard white
plate (L* = 93.33, a* = −0.91 and b* = 1.46) and black plate. Sample
2. Materials and methods (10 g) flour was kept in the glass container of the instrument and placed
over the slit of the equipment. CIE lab system was used to measure the
Proso millet (BR-7) and little millet (OLM-203) were purchased color parameters, where L* (L* = null means black and L* = 100
from the local market of Bihar, India, harvested in May 2017. Samples means white), a* (−a* = greenness and + a* = redness) and b*
were physically examined to ensure disease-free and stored in the cold (−b* = blueness and + b* = yellowness). The total color difference
(5 °C) temperature until further use. All chemicals and reagents used (ΔE) was calculated by using the following equation 1
were analytical grades.
ΔE = √ (L∗ − Li∗)2 + (a∗ − ai∗)2 + (b∗ − bi∗)2 (1)

2.1. Microwave and pan cooking

2.4.2. 2.4.2 Water absorption index (WAI) and water solubility index
Millet samples were cooked by microwave and open pan cooking to (WSI)
find the optimum cooking time by following Dias-Martins et al. (2019), Water absorption index (WAI) and water solubility index (WSI) of
the millets were considered cooked when atleast 95% of the grains were all flours were determined following the method of Maninder, Sandhu,
not opaque. For pan cooking, grains were added after the water tem- and Singh (2007) with slight modification. The samples (0.83 g) was
perature reached 98 °C with a ratio of 1:7 grain/water respectively. The dispersed in 10 mL of distilled water and cooked at 90 °C for 15 min in a
pan cooking processes were carried out for optimum cooking time water bath. The cooked flour paste was cooled at room temperature
(predetermined); 5 min and 36 min for little millet and proso millet (25 °C) and centrifuged (model EBA 8S, Hettich, Germany) at 3000 rpm
respectively, under the same temperature profiles (98 °C ± 2 °C). For for 10 min. The supernatant was transferred into an evaporating dish,
microwave cooking millet samples were added to water with a 1:7 and the weight of sediment was taken. The sediment was dried at
grain/water respectively at 25 °C. The microwave cooking processes 110 °C and weighed again. Water absorption index and water solubility
were carried at 600 W for 6 min and 30 min (predetermined optimum index were calculated following equations (2) and (3), respectively:
microwave cooking times) for little millet and proso millet respectively. weight of sediment
Samples were cooked inside the microwave (Sharp-R-3801, Japan) by WAI (g/g) =
weight of flour sample (2)
placing microwave-safe borosil bowl (diameter = 20.5 cm) at the
centre. After microwave and conventional pan cooking, the samples
Weight of dissolved in supernatant
were dried overnight at 50 °C in a hot air oven followed by milling and WSI (g/100g) = × 100
weight of flour sample (3)
finally passed through a 70 mesh sieve (USA Standard Testing Sieve No.
70). The samples were stored at 10 °C for further analysis.

2.4.3. 2.4.3 Bulk density

The bulk density of flours was determined by the method described
2.2. Chemical composition of flours
by Medhe et al. (2019) with slight modification. Samples were gently
filled into 10 mL graduated cylinders. The bottom of each cylinder was
Flour samples of raw and processed proso millet and little millet
gently tapped several times until there was no further diminution of the
were estimated for their moisture, ash, fat and protein contents by using
sample level after filling up to the 10 mL mark. Bulk density was cal-
the standard methods of analysis given by Association of Official
culated as the weight of the sample per unit volume of sample (g/mL).
Analytical Chemists (https://www.sciencedirect.com/science/article/
pii/S0023643812001065Association of Official Analytical Chemists,
2005). Carbohydrate content was calculated by subtracting the sum of 2.5. Functional properties of flours
the moisture, fat, protein and ash contents from 100 as outlined in
Association of Official Analytical Chemists (2005). Gross energy was 2.5.1. Water absorption capacity and oil absorption capacity
determined using bomb calorimeter. Water absorption and oil absorption capacity of flour samples were
determined following the method described by Chandra, Singh, and
Kumari (2015) with slight modification. For water absorption, sample
2.3. In vitro starch digestibility (0.83 g) was dispersed in 10 mL of distilled water and placed in cen-
trifuge tubes. The dispersion was further stirred occasionally, held for
In vitro starch digestibility (IVSD) of millet flours were evaluated by 30 min, followed by centrifugation (model EBA 8S, Hettich, Germany)
the method of Rathod and Annapure (2017) with slight modifications. at 3000 rpm for 25 min. The supernatant was poured in petriplate, and
Sample (50 mg/mL of 0.2 M phosphate buffer, pH 6.9) was mixed with the excess water was removed from centrifuge tubes. The sediment was
0.5 mL of pancreatic amylase (35.7 U/mg) suspension (1.42 mg/mL of placed in a hot air oven at 50 °C for 25 min, and the sample was re-
0.2 M phosphate buffer, pH 6.9) followed by incubation at 25 °C for 2 h. weighed. For the determination of oil absorption capacity, sample
3, 5-dini-trosalicylic acid (2 mL) was further added in the mixture, (0.5 g) was mixed with 6 mL of soybean oil and centrifuged at 3000 rpm
maintaining the volume to 25 mL by adding distilled water and boiled for 25 min. After centrifugation, separated oil was removed with the
for 5 min. The mixture solution was filtered. The optical density of the help of pipette, and the tubes were inverted for 25 min to drain the oil
solution was measured at 550 nm by using (UNICAM UV/Vis spectro- before reweighing. The water and oil absorption capacities were ex-
photometer, UK). Maltose was used as a reference standard, and IVSD pressed as a gram of water or oil bound per gram of the sample on a dry
was expressed as mg of maltose released per gram of sample on a dry basis. Water absorption capacity and oil absorption capacity were cal-
weight basis. culated following equations (4) and (5), respectively:

2
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

WAC (g/g) 2.7. Thermal properties

(Weight of centrifuge tube after drying − weight of centrifuge
Thermal properties of flour samples were determined by using dif-
tube) − weight of sample ferential scanning calorimeter (DSC) (SHIMADZU/DSC-60A Plus) ac-
=
weight of sample cording to the method described by Falade and Christopher (2015) with
(4) slight modifications. The flour sample (1 mg) was weighed into pierced
DSC aluminium pans and distilled water was added to make the flour:
OAC (g/g) = water ratio of 1:3. The pans were hermetically sealed, and samples were
weight of centrifuge tube after drawing oil − (centrifuge tube left to stand for an hour at 25 °C for moisture equilibration. The sealed
pans were heated from 20 °C to 130 °C under nitrogen gas at a heating
weight + weight of sample )
rate of 10 °C min−1 to gelatinize the flour samples. An empty alumi-
weight of sample
nium pan was used as a reference, and the calorimeter was calibrated
(5) with indium. From the DSC thermograms, the onset temperature (To),
peak temperature (Tp), conclusion temperature (Tc) and enthalpy of
2.5.2. Least gelation concentration of flour samples gelatinisation (ΔHG) were determined.
Least gelation concentration of samples was determined by fol-
lowing the method of Kaushal, Kumar, and Sharma (2012) with slight 2.8. FTIR of millet flours
modification. The sample (5 mL) prepared in different concentrations of
(2, 4, 6, 8, 10, 12, 14, 16, 18 and 20 g/100 mL) was placed in test tubes Infrared spectra (500–4000 cm−1) of the raw and cooked millet
and heated for an hour at 98 °C in water bath followed by rapid cooling flours were obtained by attenuated total reflectance (ATR) Fourier
under running tap water. The tubes were further cooled at 4 °C for 2 h. transform infrared (FT-IR) spectrometer (Nicolet Avatar 36). Infrared
Least gelation concentration was expressed as the concentration above spectra of raw and cooked millet samples were observed by following
which the sample did not fall or slip when the test tubes were inverted. Nasrin et al. (2015). Powdered millet samples were used for direct
measurement of the spectra in the range of 500–4000 cm−1 wave-
2.5.3. Foaming capacity and foaming stability numbers.
Foaming properties of flours were determined by the method de-
scribed by Maninder et al. (2007) with slight modification. The sample 2.9. Scanning electron micrographs of millet flours
(1.5 g) was dispersed in 50 mL of distilled water and homogenised,
using the homogeniser (Servodyne, Model 50000-25) at 960 rpm for Microstructure images of millet flours were observed by scanning
3 min. The blended mixture was further transferred into a graduated electron microscope (SEM) (Hitachi SU8230, Japan), as described by
cylinder, and the homogeniser cup was rinsed with 10 mL of distilled Ramashia, Gwata, Meddows-Taylor, Anyasi, and Jideani (2018). The
water and added to the graduated cylinder. The volume was recorded sample particles were sprayed with gold over carbon tape splattered,
before and after whipping. Foaming capacity (FC) and foaming stability and then examinations were observed at an accelerated voltage of
(FS) were calculated by equation (6): 5.000 kV using SEM.

Increase in volume after homogenization − initial volume

FC = 2.10. Statistical analysis
Initial Volume (6)

FS = Foam volume changes in the graduated cylinder recorded at The data were subjected to analysis of variance (ANOVA) by using
an interval of 20, 40, 60 and 120 min of storage. SPSS version 23 (SPSS, IBM, Chicago USA) significant differences
(p < 0.05) among mean observations were evaluated by Tukey's HSD
2.6. Pasting properties of millet flours test.

Pasting properties of flours were tested according to the method of 3. Results and discussion
Nasrin, Noomhorm, and Anal (2015) by using rapid visco analyzer
(RVA) (Model 4, Newport Scientific Pvt., Ltd. Australia). Sample (2.5 g) 3.1. Nutritional composition of millet flours
was placed into the canister and mixed thoroughly with 25 mL of dis-
tilled water. The flour suspensions was heated at 50 °C for 1 min, and The compositions of different flour samples are presented in
then the temperature was slowly increased to 95 °C for 3.2 min and Table 1. The carbohydrate content varied from 78.37 to 81.69 g/100 g
finally, the temperature was again decreased to 50 °C. All the flour among different flours. The increasing order of carbohydrate content
samples were mixed and homogenised at 960 rpm throughout the test, was RLM (78.37 g/100g) < RPM (78.57 g/100 g) < PCPM (78.91 g/
and the total test time was 13 min. 100 g) < MWLM (79.40 g/100 g) < PCLM (81.57 g/

Table 1
Nutritional analysis of millets after cooking treatments.
Sample Ash (g/100 g) Moisture (g/100 g) Fat (g/100 g) Protein (g/100 g) Carbohydrate (g/100 g) Energy (Cal/g) Starch digestibility
%

RPM 1.25 ± 0.08bc 8.71 ± 0.01a 3.24 ± 0.12c 8.28 ± 1.12b 78.51 ± 1.15a 4037 57.18 ± 1.00c
PCPM 1.5 ± 0.21c 8.13 ± 0.04b 2.3 ± 0.11b 9.07 ± 0.71bc 78.91 ± 0.91a 4067 42.20 ± 0.28b
MWPM 1.32 ± 0.13bc 8.93 ± 0.05b 3.05 ± 0.34c 4.48 ± 0.50a 81.69 ± 0.41b 4293.2 43.73 ± 1.08bc
RLM 1.9 ± 0.06d 8.65 ± 0.02c 1.91 ± 0.41ab 9.16 ± 0.61bc 78.37 ± 0.28a 4004 41.96 ± 0.22b
PCPM 0.51 ± 0.04a 7.27 ± 0.17c 1.56 ± 0.24a 9.07 ± 0.71bc 81.57 ± 0.69b 4116 29.71 ± 2.71a
MWLM 0.45 ± 0.08a 8.02 ± 0.14d 1.79 ± 0.03ab 10.32 ± 0.72c 79.40 ± 0.52a 4049 30.09 ± 0.66a

Each value is a mean of triplicates ± SD of triplicates. Means with no common letters within a column significantly differ (p < 0.05). RPM = raw proso millet,
PCPM = pan-cooked proso millet, MWPM = microwave cooked proso millet, RLM = raw little millet, PCLM = pan-cooked little millet, MWLM = microwave
cooked little millet.

3
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

100 g) < MWPM (81.69 g/100 g. The highest amount of ash content The colour of flours depends on the presence of flavonoids, antho-
was observed in RLM (1.9 g/100 g), and lowest in MWLM (0.45 g/ cyanins and tannin (Medhe et al., 2019). Sandhu and Siroha (2017)
100 g). reported L* (75.5–83.5), a* (0.07–0.96) and b* (9.2–12) respectively in
Devisetti et al. (2014) also reported similar contents, including the six species of pearl millet. After cooking treatments the increase in
carbohydrate (70% w/w) and ash (1.3 g/100 g) in raw brown porso lightness of millet flours was due to protein denaturation and attach-
millet. Wani, Hamid, Hamdani, Gani, and Ashwar (2017) observed the ment of fat droplets to the denatured protein matrix (Kahyaoglu &
increased amount of carbohydrate content after cooking. The high ash Kaya, 2006; Sharanagat et al., 2019). After cooking the increase in “a”
and high carbohydrate content of the flour indicated that these millets (redness) and “b” (yellowness) values of flours indicated the formation
could serve as a significant source of minerals and energy for consumers of brown pigments in the Maillard reactions and thermal oxidation of
(Kaushal et al., 2012). polyphenols (Bagheri, Kashaninejad, Ziaiifar, & Aalami, 2016).
The protein content was in the range of 4.48–9.16 g/100 g for dif-
ferent flours. The protein contents of RLM, PCLM, RPM and PCPM were 3.3.2. 3.3.2 Water absorption index (WAI) and water solubility index
not significantly different (p > 0.05), whereas protein content of (WSI)
MWPM (4.48 g/100g) was significantly lower than other millets. The WAI of different kinds of flours was in the range of 4.48 g/g to
MWLM (10.32 g/100 g) showed a significantly high amount of protein 3.68 g/g, whereas the highest value was observed for MWLM (4.48 g/g)
content, and the variation in the protein contents of cooked and raw and lowest value for PCPM (3.68 g/g). The WAI values of RPM, RLM,
millets might be due to variation in the nitrogen content after exposure PCLM and MWPM were not significantly different (p > 0.05).WAI is
to different cooking techniques (Wani et al., 2017). interconnected with hydrophilicity and gelation capacity of biomacro-
The fat content varied from 1.56 to 3.24 g/100 g and followed the molecules, such as starch and protein in flour (Maninder et al., 2007).
increasing order as: PCLM (1.56 g/100 g) < MWLM (1.79 g/ Dias-Martins et al. (2019) reported WAI and WSI in pan-cooked pearl
100 g) < RLM (1.91 g/100 g) < PCPM (2.3 g/100 g) < MWPM millet after conventional cooking treatment as 3.74 g/g and 3.45 g/
(3.05 g/100 g) < RPM (3.24 g/100 g). The moisture content varied 100 g, respectively. Furthermore, there was no significant effect of
from 7.27 to 8.65 g/100 g among different flour samples. The cooking cooking on WAI and WSI of pearl millet.
treatments decreased the moisture content in millet flours due to re- WSI value was in the range of 1.41 g/100 g–4.17 g/100 g for dif-
moval of water content in response to heat treatment (Wani et al., ferent flour samples, whereas the highest value was observed in MWLM
2017). (4.17 g/100 g) and the lowest value observed in RLM (1.41 g/100 g).
Devisetti et al. (2014) reported protein content of 14.8 g/100 g, fat
content of 4.3 g/100 g in raw brown proso millet. Saleh, Zhang, Chen, 3.3.3. 3.3.3 Bulk density
and Shen (2013) reported protein content of 9.7 g/100 g, fat content of Significant differences were observed among the bulk densities of
5.2 g/100 g in raw little millet. The nutritional value of millet flours the flours (Table 2). The bulk density of flours was varied from 0.72 g/
varies due to difference in geographical regions, varieties and growth mL to 0.85 g/mL, whereas the highest value and lowest value were
conditions (Maninder et al., 2007). The energy value in flour samples obtained for RLM (0.85 g/mL) and PCPM (0.72 g/mL) flour respec-
ranged from 4004 to 4293.2 cal/g. Saleh et al. (2013) also reported tively. The bulk densities of pan and microwave cooked proso millet
gross energy of 329 kcal/100g in little millet. flours were not significantly different, whereas the bulk density of RLM
and RPM was significantly different (p < 0.05).
3.2. In vitro starch digestibility of millet flours The bulk density of several millet flours depends on grain moisture
content (Subramanian & Viswanathan, 2007). The higher bulk density
Starch is a predominant component of cereals classified as rapid of millet flour suggests having denser in structure than other flours (Du,
digestive starch, slow digestive starch and resistant starch. Table 1 Jiang, Yu, X. and Jane, 2014). Flour with low bulk density is used in the
summarises the starch digestibility of flours. The starch digestibility preparation of weaning food formulations (Devisetti et al., 2014).
ranged from 57% to 29%. In RPM and RLM, digestibility of starch ob- Higher bulk density depends on the presence of lipids which act as
served higher comparare to cook one. adhevises in agglomeration of carbohydrate and protein molecules
In vitro digestibility of starch depends on many factors such as (Joshi, Liu, & Sathe, 2015).
amylose content, degree of crystallinity, amylose -lipid complexes and
the molecular structure of amylopectin (Ying et al., 2017tibility found a 3.4. Functional properties of flours
decrease in parboiled millets can be related to the formation of more
undigested or resistant starch. Annor et al. (2017) also observed finger 3.4.1. Water and oil absorption capacity
millet starches are resistance to digestive enzymes because of rigid The WACs of the flours were ranged from 1.50 to 3.23 g/g, whereas
starch granule structure compared to rice. Digestibility of millet starch the WAC of MWLM (3.23 g/g) flour was the highest, and RLM (1.50 g/
depends upon many factors such as starch morphology, amylose/ g) exhibited the lowest WAC (Table 3). The WAC of raw proso and little
amylopectin ratio, lipids, proteins, fibre and presence of antinutrients. millet were found statistically similar, whereas pan and microwave
cooked proso and little millet showed significantly different (p < 0.05)
3.3. Physical properties of flours from other flours.
The water absorption capacity (WAC) plays a vital role in food
3.3.1. Color preparation because it affects other functional and sensory properties of
Hunter colour values (L*, a*,b*) of different flours are presented in food products (Kaushal et al., 2012). Partial denaturation or dissocia-
Table 2. The lightness “L” values of RPM, RLM, PCPM, PCLM, MWPM tion of proteins and gelatinisation of carbohydrates after heat treatment
and MWLM were observed as 79.37, 78.18, 81.46, 90.35, 80.73 and increases the binding sites in roasted flours compared to raw samples
89.81 respectively. PCLM flour was more white (high value of L*) (Wani et al., 2017). The polar amino acids residues of the proteins have
whereas RLM flour showed the lowest L* value. The L* value of all the an affinity for water molecules and causing differences in WACs of
samples was significantly different from each other. The value of “a*” is different flours. The composition of carbohydrate being hydrophilic in
significantly different in millet flours. Kaushal et al. (2012) reported the majority is another factor affecting the WAC of the flours. Additionally,
colour attributes; L* (94.37), a* (−0.02) and b* (4.22) respectively in WAC is a vital property of protein and other components of flours in
rice. The total color difference values (ΔE) were significantly higher various foods, e.g. soups, dough, custards, and baked products because
(P < 0.05) for little millet as compared to compared to proso millet these are assumed to imbibe water without dissolution of protein which
after cooking treatments. provides body, thickening and viscosity (Devisetti et al., 2014).

4
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

Table 2
Physical properties of millet flours.
Parameter RPM PCPM MWPM RLM PCLM MWLM

b d c a f
L* 79.37 ± 0.43 81.46 ± 0.10 80.73 ± 0.25 78.18 ± 0.14 90.35 ± 0.50 89.81 ± 0.18e
A* 0.93 ± 0.02a 1.42 ± 0.03c 1.86 ± 0.03d 1.36 ± 0.13c 0.16 ± 0.02a 0.21 ± 0.01a
B* 10.78 ± 0.22a 17.40 ± 0.02e 18.95 ± 0.19f 13.12 ± 0.04b 14.90 ± 0.10c 15.45 ± 0.04d
ΔE – 6.96 ± 0.29a 8.33 ± 0.44b – 12.36 ± 0.11c 11.92 ± 0.28c
WAI (g/g) 3.99 ± 0.23a 3.68 ± 0.07a 3.90 ± 0.05a 3.99 ± 0.22a 4.35 ± 0.83a 4.48 ± 0.09a
WSI (g/100g) 1.42 ± 0.59a 2.61 ± 0.24ab 2.93 ± 0.48ab 1.41 ± 1.02a 3.57 ± 0.24b 4.17 ± 1.00b
BD (g/ml) 0.83 ± 0.01c 0.72 ± 0.02a 0.72 ± 0.01a 0.85 ± 0.01d 0.74 ± 0.02c 0.73 ± 0.02ab

L*, a*,b* are colour values, ΔE indicates the total color difference. WAI: Water absorption index, WSI: Water solubility index, BD: Bulk density. Each value is a mean
of triplicates ± SD of triplicates. Means with no common letters within a row significantly differ (p < 0.05). RPM = raw proso millet, PCPM = pan-cooked proso
millet, MWPM = microwave cooked proso millet, RLM = raw little millet, PCLM = pan-cooked little millet, MWLM = microwave cooked little millet. Where, -
indicates not estimated.

The OAC of different flours was in the range of 0.63 g/g to 0.92 g/g. produced by proteins, that maintains the air bubbles in suspensions and
The higher OAC was observed for MWLM (0.92 g/g) and the lowest for reduce the rate of coalescence, whereas, the presence of carbohydrates
RPM and RLM (0.63 g/g). The OAC of RPM and RLM millets were increase the viscosity of suspending medium (Devisetti et al., 2014).
significantly different (p < 0.05) from other samples, whereas PCPM, Therefore, stable foams are formed due to lower surface tension and
MWPM, PCLM and MWLM were significantly similar (p < 0.05). After high viscosity at interface. Du, Jiang, Yu, and Jane (2014) reported a
cooking OAC increased because of the variations of non-polar sides of protein content of 13.6 g/100 g and presence of FS by proso millet. In
proteins subunit generated, which possibly bind with the hydrocarbon current study FS was not observed in proso millet, which might be due
side chains of oil (Wani et al., 2017). The high OAC indicates that the the difference in varieties of proso millet and lower protein content.
flours are suitable for enhancing the flavour and mouthfeel while using
in food preparations. Additionally, due to these properties, flours can be 3.4.3. Least gelation concentration
used as a functional ingredient in food such as whipped toppings, The least gelation concentration (LGC) of different flours was in the
sausages, chiffon deserts and sponge cakes (Kaushal et al., 2012). range of 6–12% (Table 4). RLM and RPM formed gel at lower con-
Oil absorption capacity (OAC) is the capacity of the flour protein to centration (6–8% w/v) whereas, other flours formed a gel at a higher
bind fat by capillary attraction physically (Kaushal et al., 2012). Oil concentration (12% w/v). PCPM and PCLM formed a gel relatively at a
absorption gives specifically physical entrapment of oil within the lower concentration. Gel formation depends mainly on swelling and
protein isolates, and non-covalent bonds such as hydrophobic, elec- hydration of predominatly amorphous region of starch and starch
trostatic and hydrogen bondings are the forces involved in lipid-protein granules. The lower value of LGC indicates the better gelation ability of
interaction reported by Falade and Christopher (2015). The binding the protein ingredient and the swelling ability of the flour (Chandra
capacity between oil and water depends on the intrinsic factor such as et al., 2015). The gel strength depends on the intragranular binding
amino acid composition, the conformation of protein and their hydro- force of swollen starch granules due to heat, moisture treatment and
phobicity (Kaushal et al., 2012). annealing of amylose and amylopectin (Devisetti et al., 2014).

3.4.2. Foaming capacity and foaming stability 3.5. Pasting properties

The highest foaming capacity was observed in RPM and lowest in
PCLM and MWLM (Table 3). The foaming capacity were ranged from Table 5 summarises the pasting properties of different flour samples.
0.20 mL/100 mL–0.18 mL/100 mL. It was observed that the foaming Highest peak temperature was observed in MWPM (94 °C) and lowest in
capacity decreased after cooking. However, it was not significantly RLM (53 °C). The peak temperature of little millet flours was similar,
different among cooked and control samples. Wani et al. (2017) ob- whereas for proso millet flours it was found significantly different. The
served a decrease in foaming capacity in sweet chestnut after roasting. peak viscosities of different flours were ranged from 2152 RVU to 2165
Denaturation of several proteins occurred while cooking, which re- RVU. Final viscosities and setbacks in the flour samples ranged from
duced the protein solubility that is expected the cause of lower foaming 2125 RVU to 2145 RVU and 0.82 RVU to 4.75 RVU, respectively. The
capacity. Foaming stability was not observed in any flour samples. FC breakdown of flour samples ranged from 21.63 RUV to 33.14 RUV. The
and FS depend on the composition of carbohydrates and proteins pre- breakdowns observed in RLM, PCLM and MWLM were significantly
sent in the sample (Bhat, Wani, Hamdani, Gani, & Masoodi, 2016). The similar (p < 0.05) whereas RPM, PCPM and MWPM flours exhibited
stability and the formation of foams are dependent on interfacial film significantly different breakdowns.

Table 3
Functional properties of millet flours.
Parameter RPM PCPM MWPM RLM PCLM MWLM

WAC (g/g) 1.96 ± 0.01a 2.64 ± 0.40b 3.23 ± 0.01b 1.52 ± 0.15a 2.8 ± 0.10b 2.8 ± 0.45b
OAC (g/g) 0.63 ± 0.01a 0.87 ± 0.05b 0.90 ± 0.10b 0.68 ± 0.01a 0.85 ± 0.05b 0.92 ± 0.05b
FC (ml/100 ml) 0.20 ± 0.02a 0.19 ± 0.01a 0.18 ± 0.01a 0.19 ± 0.01a 0.18 ± 0.02a 0.18 ± 0.03a
FS(ml/100 ml) -
– – – -
–
20 min
40 min -
– – – - -

60 min -
– – – - -

120 min – – – – -
–

WAC: Water absorption capacity, OAC: Oil absorption capacity, FC: Foaming capacity, FS: Foaming stability. Each value is a mean of triplicates ± SD of triplicates.
Means with no common letters within a row significantly differ (p < 0.05). RPM = raw proso millet, PCPM = pan-cooked proso millet, MWPM = microwave
cooked proso millet, RLM = raw little millet, PCLM = pan-cooked little millet, MWLM = microwave cooked little millet.

5
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

Table 4 et al., 2007). Wani et al. (2017) observed higher gelatinisation tem-
Least gelation concentration of flours. peratures in pan-roasted and microwave chestnut flours due to the
Concentrations (% w/v) RPM PCPM MWPM RLM PCLM MWLM- exposure of high temperatures while roasting. Conclusion temperature
ranged from 80 to 123.91 °C whereas the highest one was observed in
2 – – – – – – MWLM and the lowest one in RPM. Wani et al. (2017) also reported in
4 – – – – – –
pan-roasted and microwave chestnut onset temperature
6 – – – + – –
8 + + – + + –
(93.40–92.98 °C), peak temperature (106.19–103.96 °C), conclusion
10 + + + + + + + temperature (121.26–123.94 °C) and enthalpy (12.63–11.94 J/g) re-
12 + + + + + + + spectively.
14 + + + + + + + + +
16 + + ++ + + + + + ++
18 + + ++ ++ + + + + ++ 3.7. FTIR of millet flours
20 + + ++ ++ + + + + ++
FT-IR spectra of RLM, PCLM, MWLM, RPM, PCPM and MWPM
- = No gelation, + = gel, + + = firm gel RPM = raw proso millet, millet illustrated are in Figs. 1 and 2. The peaks were observed in the
PCPM = pan-cooked proso millet, MWPM = microwave cooked proso millet,
range of specific spectral regions in raw, pan-cooked and microwave
RLM = raw little millet, PCLM = pan-cooked little millet,
samples. The major peaks were observed at 3427, 1654, 1541, 1379,
MWLM = microwave cooked little millet.
1245, 1159, 766 and 707 cm −1. These peaks showed the occurrence of
hydroxyl (-OH), amine groups (–NH) and carbonyl group (= C = O)
Pasting property plays an essential role in the selection of food
bonds, respectively (Wani et al., 2017). There was a difference observed
thickener and binder. Pasting properties dependent on the rigidity of
in raw millet flours and cooked millet flours. PCLM and MWLM millet
starch granules that are affecting granule swelling potential and
flours had a difference in carbohydrate region (1200- 900 cm−1)
leaching amount of amylose in the solution (Medhe et al., 2019). Fur-
compare to RLM but did not show any variation in protein region
thermore, the increase in pasting temperature was observed with in-
(1700-1600 and 1570 -1534 cm−1). There were not much difference
crease in heating treatment time. This prolonged heating treatment
observed in raw and cooked proso millet in carbohydrate region (1200-
might influence the structure and arrangement of starch molecules (Qu,
900 cm−1) and in protein, region (1700-1600 and 1570 -1534 cm−1).
Wang, Liu, Wang, & Liu, 2017). Low breakdown of flour, indicates that
Spectral changes in amide I and II regions are related to the protein
flour exhibits good paste stability and strong shearing (Du, Jiang, Yu, X.
backbone conformation due to high temperature, high pressure.
and Jane, 2014). Yang et al. (2018) reported pasting properties of non-
waxy cooked proso millet (pasting temperature, peak viscosity, through
viscosity, breakdown, setback, and final viscosity) as 79.93 °C, 2111.00, 3.8. Scanning electron micrographs of millet flours
1112.50, 998.50, 3263.50, 4376 cP respectively.
Fig. 3 illustrates the morphological micrographs of different flour
samples. The flours of RPM and RLM were granular, spherical, oval and
3.6. Thermal properties smooth close together with no gaps. Some granules were damaged and
irregular in shape. Ramashia et al. (2018) reported that the whole
Whole flours are composite systems of the starch molecule, fibres, finger, millet flour starch granules had a different shape like oval,
proteins, lipids and other components that can overall affect the heat polygonal with a smooth surface. After cooking significantly changes
capacity of the flour (Chávez-Murillo, Veyna-Torres, Cavazos-Tamez, de occurred in starch, granules are entirely damaged. Compared with raw
la Rosa-Millán, & Serna-Saldívar, 2018). The onset temperature ranged samples, the cooked one showed many branching structures that con-
from 63.56 to 108.56 °C for flour samples, as shown in Table 5 and Figs. nected to form a network in little millet but in proso millet not affected
S1–S6 (supplementary material). The highest onset temperature was much after cooking. Shrestha, Sadiq, and Anal (2018) also observed
observed in PCPM and lowest in RPM flour. The increase in To found changes in starch such as irregular shapes and an amorphous mass of
because of the transformation of inter-crystalline amorphous form and cohesive structure in culled banana starch after heat treatment.
structural changes of the starch granules (Sharanagat et al., 2019). The
peak gelatinisation temperature ranged from 72.68 to 118.69 °C, 4. Conclusion
whereas the highest in MWLM, and the lowest one was observed in
RPM. The gealatization temperatures found different among the flours In this study, differences have been observed among the aesthetic
due to size, form and internal distribution of starch granules (Maninder properties, pasting properties, FTIR, and in vitro digestibility of millets

Table 5
Pasting properties of millet flours.
Parameters RPM PCPM MWPM RLM PCLM MWLM

ab c ab ab a
Peak viscosity (RVU) 2153.64 ± 0.79 2165.02 ± 2.89 2156 ± 1.01 2154.38 ± 1.2 2152 ± 1.37 2157 ± 0.41b
Trough viscosity (RVU) 2131.67 ± 1.09b 2134.36 ± 3.78b 2124.63 ± 2.60a 2123.21 ± 0.55a 2134.72 ± 2.00b 2129.74 ± 1.05b
Break down (RVU) 21.63 ± 0.47a 24.36 ± 0.47b 31.58 ± 0.36c 31.74 ± 1.37c 33.14 ± 0.58c 31.70 ± 0.93c
Final viscosity (RVU) 2125.83 ± 0.79a 2139.56 ± 0.92c 2150.47 ± 1.27e 2129.03 ± 1.96b 2145.44 ± 1.41d 2131.20 ± 0.54b
Set back (RVU) 4.75 ± 0.28e 1.27 ± 0.02b 1.56 ± 0.01c 3.57 ± 0.01d 0.82 ± 0.03a 0.84 ± 0.02a
Pasting temp (oC) 67.15 ± 1.63b 73.83 ± 0.62c 94.65 ± 0.39d 53.97 ± 1.14a 54.43 ± 1.07a 55.73 ± 1.21a
Thermal properties
TO (°C) 64.56 108.56 82.37 71.26 97.73 99.5
TP (°C) 72.68 112.8 102.72 78.44 117.53 118.69
TC (°C) 80.09 123.91 124.37 82.28 123.71 123.91
H (J/g) 0.98 0.24 3.27 1.69 4.65 5.34
TC – TO (°C) 15.53 7.83 21.65 11.02 8.24 5.22

To: onset gelatinisation temperature, Tp: peak gelatinisation, Tc: conclusion, H: enthalpy.. Each value is a mean of triplicates ± SD of triplicates. Means with no
common letters within a row significantly differ (p < 0.05). RPM = raw proso millet, PCPM = pan-cooked proso millet, MWPM = microwave cooked proso millet,
RLM = raw little millet, PCLM = pan-cooked little millet, MWLM = microwave cooked little millet. RVU = rapid visco units.

6
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

Fig. 1. FTIR spectra of raw, pan cooked and microwaved cooked little millet.

Fig. 2. FTIR spectra of raw, pan cooked and microwave cooked proso millet.

Fig. 3. Scanning electron microscopic structures of RPM = raw proso millet, PCPM = pan-cooked proso millet, MWPM = microwave cooked proso millet,
RLM = raw little millet, PCLM = pan-cooked little millet, MWLM = microwave cooked little millet.

7
S.R. Kumar, et al. LWT - Food Science and Technology 128 (2020) 109465

pan and microwave cooked millets. From Nutritional aspects, the car- properties of whole legume flour. LWT - Food Sci. Technol. 55(1), 308–313.
bohydrate composition and calorific value have also been increased by Ekezie, F. G. C., Sun, D. W., Han, Z., & Cheng, J. H. (2017). Microwave-assisted food
processing technologies for enhancing product quality and process efficiency: A re-
both cooking methods. Both methods of cooking do not have any effect view of recent developments. Trends in Food Science & Technology, 67, 58–69.
on physiochemical properties such as WAI, but it has been observed Falade, K. O., & Christopher, A. S. (2015). Physical, functional, pasting and thermal
that WSI has increased. Increase in WAC and OAC has enhanced the properties of flours and starches of six Nigerian rice cultivars. Food Hydrocolloids, 44,
478–490.
functional properties, which can be further used in the food industries Gavahian, M., Chu, Y. H., & Farahnaky, A. (2019). Effects of ohmic and microwave
for the formulation of healthy gluten-free products. However, the pre- cooking on textural softening and physical properties of rice. Journal of Food
sent study reports in-depth analysis of nutritional, physio-chemical and Engineering, 243, 114–124.
Guo, Q., Sun, D. W., Cheng, J. H., & Han, Z. (2017). Microwave processing techniques and
functional properties of pan and microwave cooked millets (Proso and their recent applications in the food industry. Trends in Food Science & Technology, 67,
little), these under-utilized millets have great potential to be used in the 236–247.
food industry for the formulation of new products or as a replacement Joshi, A. U., Liu, C., & Sathe, S. K. (2015). Functional properties of select seed flours. LWT
- Food Sci. Technol. 60(1), 325–331.
in food products due to their improved nutritional physiochemical and
Kaushal, P., Kumar, V., & Sharma, H. K. (2012). Comparative study of physicochemical,
functional properties. functional, antinutritional and pasting properties of taro (Colocasia esculenta), rice
(Oryza sativa) flour, pigeon pea (Cajanus cajan) flour and their blends. LWT - Food Sci.
Funding Technol. 48(1), 59–68.
Liu, K., Zheng, J., Wang, X., & Chen, F. (2019). Effects of household cooking processes on
mineral, vitamin B, and phytic acid contents and mineral bioaccessibility in rice. Food
This research did not receive any specific grant from funding Chemistry, 280, 59–64.
agencies in the public, commercial, or not-for-profit sectors. Maninder, K., Sandhu, K. S., & Singh, N. (2007). Comparative study of the functional,
thermal and pasting properties of flours from different field pea (Pisum sativum L.)
and pigeon pea (Cajanus cajan L.) cultivars. Food Chemistry, 104(1), 259–267.
CRediT authorship contribution statement Medhe, S., Jain, S., & Anal, A. K. (2019). Effects of sprouting and cooking processes on
physicochemical and functional properties of moth bean (Vigna aconitifolia) seed and
flour. Journal of Food Science and Technology, 56(4), 2115–2125.
Simmi Ranjan Kumar: Data curation, Formal analysis. Nasrin, T. A. A., Noomhorm, A., & Anal, A. K. (2015). Physico-chemical characterization
Muhammad Bilal Sadiq: Formal analysis, Conceptualization, Writing - of culled plantain pulp starch, peel starch, and flour. International Journal of Food
original draft. Anil Kumar Anal: Formal analysis, Writing - original Properties, 18(1), 165–177.
Qu, C., Wang, H., Liu, S., Wang, F., & Liu, C. (2017). Effects of microwave heating of
draft, Conceptualization. wheat on its functional properties and accelerated storage. Journal of Food Science and
Technology, 54(11), 3699–3706.
Appendix A. Supplementary data Ramashia, S. E., Gwata, E. T., Meddows-Taylor, S., Anyasi, T. A., & Jideani, A. I. O.
(2018). Some physical and functional properties of finger millet (Eleusine coracana)
obtained in sub-Saharan Africa. Food Research International, 104, 110–118.
Supplementary data to this article can be found online at https:// Rathod, R. P., & Annapure, U. S. (2017). Physicochemical properties, protein and starch
doi.org/10.1016/j.lwt.2020.109465. digestibility of lentil based noodle prepared by using extrusion processing. LWT, 80,
121–130.
Saleh, A. S., Zhang, Q., Chen, J., & Shen, Q. (2013). Millet grains: Nutritional quality,
References processing, and potential health benefits. Comprehensive Reviews in Food Science and
Food Safety, 12(3), 281–295.
Annor, G. A., Tyl, C., Marcone, M., Ragaee, S., & Marti, A. (2017). Why do millets have Sandhu, K. S., & Siroha, A. K. (2017). Relationships between physicochemical, thermal,
slower starch and protein digestibility than other cereals? Trends in Food Science & rheological and in vitro digestibility properties of starches from pearl millet cultivars.
Technology, 66, 73–83. LWT - Food Sci. Technol. 83, 213–224.
Association of Official Analytical Chemists (2005). Official methods of analysis of the as- Sarita, E. S., & Singh, E. (2016). Potential of millets: Nutrients composition and health
sociation of analytical Chemists. Association of Official Analytical Chemists. benefits. J. Scient. Innovat. Res. 5(2), 46–50.
Bagheri, H., Kashaninejad, M., Ziaiifar, A. M., & Aalami, M. (2016). Novel hybridized Sharanagat, V. S., Suhag, R., Anand, P., Deswal, G., Kumar, R., Chaudhary, A., et al.
infrared-hot air method for roasting of peanut kernels. Innovative Food Science & (2019). Physico-functional, thermo-pasting and antioxidant properties of microwave
Emerging Technologies, 37, 106–114. roasted sorghum [Sorghum bicolor (L.) Moench]. Journal of Cereal Science, 85,
Bhat, N. A., Wani, I. A., Hamdani, A. M., Gani, A., & Masoodi, F. A. (2016). 111–119.
Physicochemical properties of whole wheat flour as affected by gamma irradiation. Shrestha, S., Sadiq, M. B., & Anal, A. K. (2018). Culled banana resistant starch-soy protein
LWT - Food Sci. Technol. 71, 175–183. isolate conjugate based emulsion enriched with astaxanthin to enhance its stability.
Chandrasekara, A., & Shahidi, F. (2010). Content of insoluble bound phenolics in millets International Journal of Biological Macromolecules, 120, 449–459.
and their contribution to antioxidant capacity. Journal of Agricultural and Food Subramanian, S., & Viswanathan, R. (2007). Bulk density and friction coefficients of se-
Chemistry, 58(11), 6706–6714. lected minor millet grains and flours. Journal of Food Engineering, 81(1), 118–126.
Chandra, S., Singh, S., & Kumari, D. (2015). Evaluation of functional properties of Tiwari, N., Tiwari, S., & Tripathi, N. (2018). Genetic characterization of Indian little
composite flours and sensorial attributes of composite flour biscuits. Journal of Food millet (Panicum sumatrense) genotypes using random amplified polymorphic DNA
Science and Technology, 52(6), 3681–3688. markers. Agric. Nat. Resour. 52(4), 347–353.
Chávez-Murillo, C. E., Veyna-Torres, J. I., Cavazos-Tamez, L. M., de la Rosa-Millán, J., & Wani, I. A., Hamid, H., Hamdani, A. M., Gani, A., & Ashwar, B. A. (2017). Physico-che-
Serna-Saldívar, S. O. (2018). Physicochemical characteristics, ATR-FTIR molecular mical, rheological and antioxidant properties of sweet chestnut (Castanea sativa
interactions and in vitro starch and protein digestion of thermally treated whole pulse Mill.) as affected by pan and microwave roasting. Journal of Advanced Research, 8(4),
flours. Food Research International, 105, 371–383. 399–405.
Devisetti, R., Yadahally, S. N., & Bhattacharya, S. (2014). Nutrients and antinutrients in Yang, Q., Zhang, P., Qu, Y., Gao, X., Liang, J., Yang, P., et al. (2018). Comparison of
foxtail and proso millet milled fractions: Evaluation of their flour functionality. LWT - physicochemical properties and cooking edibility of waxy and non-waxy proso millet
Food Sci. Technol. 59(2), 889–895. (Panicum miliaceum L.). Food Chemistry, 257, 271–278.
Dias-Martins, A. M., Cappato, L. P., da Costa Mattos, M., Rodrigues, F. N., Pacheco, S., & Ying, D., Hlaing, M. M., Lerisson, J., Pitts, K., Cheng, L., Sanguansri, L., et al. (2017).
Carvalho, C. W. (2019). Impacts of ohmic heating on decorticated and whole pearl Physical properties and FTIR analysis of rice-oat flour and maize-oat flour based
millet grains compared to open-pan cooking. Journal of Cereal Science, 85, 120–129. extruded food products containing olive pomace. Food Research International, 100,
Du, S. K., Jiang, H., Yu, X., & Jane, J. L. (2014). Physicochemical and functional 665–673.