Evaluation of the Effects of Hydrothermal Treatment on Rice Flour and Its Related Starch

Abstract

The main aim of this research was to compare the effects of hydrothermal treatment on rice flour and its related rice starch. The treatment was performed at 120°C for 3 and 5 h. Scanning electron microscopy results showed that the proteins of hydrothermaled rice flour were denatured and formed clusters and the granules of hydrothermaled rice starch became aggregated and had an irregular surface. The treatment reduced water solubility and water absorption and decreased peak viscosity while increased pasting temperature. It increased the final visosity of modified rice flour while reduced the final viscosity of modified rice starch. Following hydrothermal treatment, the hardness and elasticity of the gels increased. The cohesiveness of rice flour gels decreased while that of the rice starch gels remained unchanged. This study showed how the hydrothermal treatment can have different effects on rice flour and its related rice starch. The effects of hydrothermal treatment on rice flour were stronger than rice starch. Increasing the treatment time from 3 to 5 h was more effective on rice starch.

Keywords:

• Rice flour
• Rice starch
• Hydrothermal treatment

Introduction

Hydrothermal treatment (HT) or heat-moisture treatment is a common physical method for starch modification in order to induce new functional properties without using any chemicals and hence, is considered as environmentally friendly and safe for human consumption. HT starches can be used in frozen and canned foods, noodles, fried batter foods, pasta, bread,^[1] and biodegradable films.^[1–4] HT starches can also act as resistant starch and increase the level of slowly digested starch.^[5]

HT involves incubation of starch at above gelatinization temperature (80–120°C) at limited water (<35%) for a time period in the range of 15 min to 16 h.^[6] Under the conditions applied, starch undergoes some structural alternations that affect its physicochemical properties. Many studies have been performed to show the effects of various conditions of HT on different starches including tuber and root starches, cereal starches, legume starches, and starches extracted from less common sources such as pihão and African yam.^[6–10] These studies have documented that the HT causes significant changes in water absorption, swelling, starch crystallinity, gelatinization properties, enzyme susceptibility, and viscosity. The extent of these changes depends on the origin of starch and the conditions of the treatment.

Comparing to other starches, rice starch (RS) is not widely used in different products because of the high value of milled rice as food. However, its unique characteristics including small size of the granules, being gluten-free, non-allergenic, and having a wide range of amylose content are attractive aspects of RS in the food industry. Therefore, determination of the effects of different modifications including HT on RS is of great interest and can increase its applications.^[11] In many parts of the world, rice is milled and polished to obtain a white polished rice. The milling and polishing processes produce an average of about 14% of broken grains. This represents a financial problem for the rice industry since the price of the broken grains is about one-fifth of the whole grain.^[12] Therefore, finding economical methods to increase the applications of broken grains and convert them to value added products are of great value. One of the products obtained from broken rice grains is rice flour (RF). It is generally of lower price compared to the RS and many other starches. RF consists of mainly starch (85–90%), and other components including proteins (5–8%), lipids (0.5–1.2%), and ash (0.3–0.8%). One of the main applications of RF is in production of baby foods, gluten-free foods, rice drinks, noodles, and desserts. It is possible to enhance the applications of RF using HT. In contrast to RS, there is still a lack of knowledge to address the effects of HT on RF. HT can affect the starch constituent of the RF as well as the non-starch components (mainly proteins and lipids) and hence, the overall effects of the treatment should be different from that of RS, which requires further investigations. The effects of HT of the flour obtained from foxtail and proso millet have been studies.^[13,14] It has been found that starch and flour had some different characteristics after modifications. The positive effects of HT RF on noodle quality have been reported.^[3] The main objective of this study was to determine the physical changes of RF that occur after HT compared to those of RS. For this purpose, microstructure, water interactions, pasting properties, and gel textural characteristics of native and HT samples were determined.

Materials and Methods

Production of RF

Broken rice grains obtained from short rice grain with commercial name “Taroum” was purchased from local market. The grains were cleaned by hand to remove impurities, washed, and soaked in water for 3 h at an ambient temperature, then milled using a laboratory rotary mill (Retsch GmbH 5657 HAAN, Germany). The flour was dried overnight at 45°C, then sieved to obtain an average particle size of 270 µm, and packed in a sealable glass jar and kept refrigerated for further experiments.

Extraction of RS

To produce RS, the alkaline extraction method was used.^[15] Broken rice grains (1 kg) were soaked in a sodium hydroxide solution (0.3%, 10 L) at 25°C for 24 h. Then the grains were recovered, washed, and ground using a blender. The obtained slurry was again suspended in a sodium hydroxide solution (0.3%, 10 L) and left stirring over a magnet stirrer for 20 min. The suspension was left to settle for 6 h and then the supernatant was discarded. The process was repeated until the protein test (Biuret) of the supernatant showed negative result. The slurry was suspended in deionoized water and HCl (0.5 M) was added gradually to adjust the pH to 7.0. Then it was passed through a sieve with opening of 35 µm, left to settle for 6 h and the clear supernatant was decanted. The sediment starch was dried at 40°C for 72 h. Starch extraction yield (%) was determined by measuring the weight of dry starch obtained from 100 g broken rice grains.^[16]

Chemical Composition of RF and RS

Moisture, protein, lipid, and ash content of the RF and RS were determined according to the Approved Methods of the American Association of Cereal Chemists (AACC; 2000), methods 44-15A, 39-11, 35-20, and 08-01, respectively. Total carbohydrate content was obtained by subtracting the sum of the moisture, protein, lipid, and ash content from 100.^[17]

HT of the Starch and Flour

The amount of water required to adjust the moisture content of the samples to 20% was calculated from the difference between sample moisture content and the target moisture content. Then the water was sprayed on the samples, mixed well, and stored in a sealed glass container at 4°C for 4 days for equilibrium. The HT treatment was conducted at 120°C for 3 or 5 h. The modified samples were dried at 40°C to a moisture content of 12.2 ± 0.8% and then sieved to obtain an average particle size of 270 µm. The samples were stored in sealed glass jars at ambient temperature for further experiments.^[18] The modified RF obtained after 3 and 5 h treatment time were named “HTRF3” and “HTRF5,” respectively, and the modified RS obtained after 3 and 5 h treatment times were named “HTRS3” and “HTRS,” respectively.

Scanning Electron Microscopy

To study the morphological features of the samples, a tiny amount of each sample was fixed on an aluminum stub using double-sided adhesive tape and sputter-coated with gold. Then it was observed at an accelerating voltage of 20 kV under a scanning electron microscope (Model 5526, Cambridge, UK).

Determination of Starch Water Solubility and Water Absorption

Sample (starch or flour) was first dispersed in distilled water (3% w/w, starch in water) and heated up to 95°C for 30 min in a water bath (W350B Fater Electric, Tehran, Iran). Then it was centrifuged (RC-S Sorvall, New York, USA) at 700 g for 15 min. The supernatant containing solubilized starch was collected, dried at 120°C to a constant weight and the residue was weighed. The amount of components in the supernatant under the described condition (solubility) was determined according to the Eq. (1).^[19]

(1) $\mathrm{S}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\left(\mathrm{\%}\right)\hspace{0.17em}=\frac{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{d}\mathrm{s}\mathrm{u}\mathrm{p}\mathrm{e}\mathrm{r}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}}{\mathrm{D}\mathrm{r}\mathrm{y}\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{r}\mathrm{c}\mathrm{h}}\times 100$(1)

To determine the water absorption, samples were prepared in the same method as described above. After centrifugation, the pellet was collected and weighed and then dried at 120°C to a constant weight. Water absorption was calculated using the Eq. (2).^[19]

(2) $\mathrm{W}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{a}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\mathrm{g}/\mathrm{g}\right)\hspace{0.17em}=\frac{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{w}\mathrm{e}\mathrm{t}\mathrm{p}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{t}-\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{d}\mathrm{r}\mathrm{y}\mathrm{p}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{t}}{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{d}\mathrm{r}\mathrm{y}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{d}\mathrm{o}\mathrm{f}\mathrm{t}\mathrm{h}\mathrm{e}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}}\times 100$(2)

Pasting Properties

A rapid visco analyser (RVA Starch Master 2, Perten, Australia) was used to determine the pasting properties of the samples. For RS, HTRS3, and HTRS5 suspensions of 3.5 g (starch, dry weigh basis) in 25 mL distilled water and for RF, HTRF3, and HTRF5 suspensions of 4.0 g (flour, dry weight basis) in 25 mL of distilled water were made in an aluminum canister and placed in the RVA. The stirring speed was 960 rpm for the first 10 s and then 160 rpm for the rest of the trial. The samples were heated at 50°C for 2 min, then the temperature raised to 95ºC at the rate of 2ºC/s, maintained at 95ºC for 3 min, and then cooled to 50ºC at the same rate and remained for 5 min at 50°C. RVA parameters including pasting temperature (the temperature at which gelatinization begins), peak viscosity (maximum viscosity during heating), trough viscosity (minimum viscosity during heating), final viscosity (viscosity after cooling to 50°C), and setback viscosity (final viscosity–trough viscosity) were obtained from the RVA curve.^[19]

Instrumental Textural Properties

A suspension of each sample in distilled water (15%, w/w) was made in a test tube. The tubes were sealed and heated at 95°C for 30 min during which the sample was mixed every 5 min over a vortex mixer for 30 s. The hot paste was poured into a cylindrical plastic mold with a dimension of 10 × 10 mm and stored for 24 h at 4°C before the measurement. Texture of the gels were measured using a texture analyzer (TA Plus, Stable MicroSystems, Godalming, Surrey, England) by performing a two bite compression test, at a pretest speed of 5.0 mm/s, test speed of 2.0 mm/s, post-test speed 5.0 mm/s, time interval of 10 s between the two compressions and a strain deformation of 25% using a cylindrical plunger with a diameter of 20 mm. From the resulting force/deformation curves, the textural parameters including hardness, cohesiveness, and elasticity were calculated.^[20]

Statistical Analysis

The experiments were carried out using an entirely random experimental design with at least three repetitions. Mean and standard deviation were calculated using Microsoft Excel 2007 (XP Edition, Microsoft Corporation, USA). For the statistical analysis of the results the program SPSS (Version 16.0) was used. Analysis of variance (ANOVA) and comparison of means with Duncan’s test were conducted at a significance level of p < 0.05.

Results and discussion

The starch extraction yield in this study was 60% which is close to the values reported previously.^[21] Slightly higher or lower starch extraction yield (48.6–70%) as reported in the literature are related to differences in the experimental methods and rice variety.^[15] Determination of the chemical composition of the samples (Table 1) shows that the RF contained 9.60% moisture, 7.76% protein, 0.89% lipid, 0.74% ash, and 81.01% total carbohydrate content. The RS contained 9.36% moisture, 0.68% protein, 0.48% lipid, 0.30% ash, and 89.18% total carbohydrate content. These values are similar to the values reported for RF and RS.^[4,22] The moisture content of the RF and RS were similar; however, RF had significantly (p < 0.05) higher protein, lipid, and ash content, but lower carbohydrate content than RS.

TABLE 1 Chemical composition of rice flour (RF) and rice starch (RS) extracted from the same flour

Chemical composition (%)	Rice flour	Rice starch
Moisture	9.60 ± 0.40^a	9.36 ± 0.20^a
Protein	7.76 ± 0.05^a	0.68 ± 0.01^b
Lipid	0.89 ± 0.09^a	0.48 ± 0.17^b
Ash	0.74 ± 0.01^a	0.30 ± 0.01^b
Total carbohydrate content	81.01 ± 0.20^b	89.18 ± 0.17^a

Values are the average of triplicates ± standard deviation. Averages followed by different superscripts in the same row are statistically different (p < 0.05).

Microstructure of the Samples

Determination of the microstructure of the samples can help interpretation of some physicochemical properties of the samples. The micrographs of the samples are depicted in Fig. 1. In the rice grain endosperm, compound starch granules surrounded by protein bodies are present. The starch granules are round to polygonal and have very small sizes (less than 10 µm). However, in the RF, the compound structure of the granules is mostly dissociated during dry milling of the grains and as observed in Fig. 1A, the individual granules were randomly dispersed between the endosperm particles including protein bodies and other constituent of the endosperm. Most of the protein bodies were in the form of agglomerates rather than single bodies. Within the protein bodies, glutelin, and prolamin, the main storage proteins in the rice endosperm, are accumulated.^[23] These observations are in agreement with previous findings.^[24,25] However, the images of the HT flours were different from native RF (Fig. 1B and 1C). The starch granules appeared as clusters and the denatured protein bodies were spread over and adhered to the exterior of the starch granules. These changes were enhanced with increasing the treatment time. Similar changes have been reported for HT millet flour and it has been indicated that these changes are related to interactions between starch granules and the non-starch compositions which occur during dry heating process.^[14]

FIGURE 1 Micrographs of A: rice flour (RF); B: hydrothermaled rice flour for 3 h (HTRF3); C: hydrothermaled rice flour for 5 h (HTRF5); D: rice starch (RS); E: hydrothermaled rice starch for 3 h (HTRS3); and F: hydrothermaled rice starch for 5 h (HTRS5). Bars on the images are 20 µm. S: starch granules; P: protein.

The micrographs of native and modified RS revealed that the granules became more aggregated and irregular on the surface compared to the native starch (Fig. 1D, 1E, and 1F). These changes increased with increasing treatment time. Similar changes were reported for HT RS.^[26]

Water Solubility

Determination of water solubilization of the samples (Fig. 2) showed that the RS had higher water solubility than RF. During starch gelatinization the granules absorb water and swell and amylose molecules leach out of the granules accounting for the soluble materials. In the case of flour, the presence of water insoluble materials such as proteins and lipids and their interaction with amylose molecules can prevent them from diffusing out of the granules and thus reduce the solubility. Similarly, previous results have shown that among different HT starches, wheat starch which had higher lipid and protein contents showed lower water solubility.^[7]

FIGURE 2 Water solubility of the rice flour (RF), hydrothermaled rice flour for 3 h (HTRF3), hydrothermaled rice flour for 5 h (HTRF5), rice starch (RS), hydrothermaled rice starch for 3 h (HTRS3), and hydrothermaled rice starch for 5 h (HTRS5). Values are the average of triplicates ± standard deviation. Bars with different letters are statistically different (p < 0.05).

A significant decrease in the water solubility of both samples was observed after HT. Similar results were found for HT RS and some other cereal starches.^[6–9] The formation of amylose-lipid complex within the starch granules, an internal re-arrangement of the starch granules that provides higher interactions between starch functional groups and the formation of more ordered amylopectin clusters are the main reasons for solubility reduction.^[27] Morphological changes of the starch granules as a result of HT can also affect water solubility. For the modified flour it was found that the lipids and proteins were adhered to the surface of the granules (see Fig. 1), which can hinder amylose from diffusing out of the granules. Increasing the treatment time from 3 to 5 h had no significant effect on the RF solubility, while it reduced the water solubility of the RS significantly. The difference between the water solubility of the samples may be related to their differences in chemical composition. Comparing to the native RF, reductions of 35 and 38% was observed for HTRF3 and HTRF5, respectively. However, HTRS3 and HTRS5 had less reduction in solubility compared to the modified RF.

Water Absorption

Water absorption of the samples is presented in Fig. 3. During heating of starch in excess water, the starch granules absorb water and swell. Water absorption has a positive correlation with swelling capacity of the granules. The starch molecules are the main components to absorb water during heating. However, their interactions with other molecules such as lipids and proteins which are present in higher amount in RF, particularly on the exterior of the granules, may hinder water absorption of the starch molecules resulting in had lower water absorption of RF.

FIGURE 3 Water absorption of the rice flour (RF), hydrothermaled rice flour for 3 h (HTRF3), hydrothermaled rice flour for 5 h (HTRF5), rice starch (RS), hydrothermaled rice starch for 3 h (HTRS3), and hydrothermaled rice starch for 5 h (HTRS5). Values are the average of triplicates ± standard deviation. Bars with different letters are statistically different (p < 0.05).

The HT significantly reduced the water absorption of the samples. It has been indicated that the HT treatment causes internal re-arrangement of the starch granules which increases interactions between the functional groups of starch, the formation of amylose-lipid complexes within the granules and increase in the crystalline regions of the granules. These changes have negative effects on water absorption and swelling of the sample following HT. In addition, exterior changes of the granules during HT treatment as observed for modified RF and RS (see Fig. 1) may affect water absorption of the samples. Reduction in the swelling power of some cereal and tuber starches has been reported.^[6] The water absorption of the RS reduced significantly with increasing the HT time while the water absorption of the RF remained unchanged, which can be related to the different composition of these two samples. The water absorption of the HTRF3 and HTRF5 had a reduction of 19 and 21% compared to the RF. However, HT caused a greater reduction in water absorption of RS (27 and 39% for HTRS3 and HTRS5, respectively).

Pasting Properties

In the RVA “peak viscosity” is related to the maximum water absorption and swelling of the granules and is affected by starch botanic source, chemical composition, and presence of non-starch materials such as lipids and proteins.^[19] In this study (Table 2) the peak viscosity of the RF was significantly lower than that of RS. The lower peak viscosity of the RF can be related to the lower water absorption and solubility of the RF compared with the RS as explained before. The results showed that the HT reduced peak viscosity of both samples. Reduction in the peak viscosity of the HT samples is related to the structural changes of starch and restricted water absorption and swelling.^[3] The peak viscosity values of the modifed RF samples were lower than those of modified RS. This can be related to more limited water absorption and swelling of the modified RF which further reduced with increasing treatment time. The peak viscosity of the HTRF3 and HTRF5 had reductions of 5.52 and 7.30% compared to the native RF. However, greater reductions in the peak viscosity of the HTRS3 and HTRS5 (59.97 and 66.16%) were observed.

TABLE 2 Pasting properties of rice flour (RF), hydrothermaled rice flour for 3 h (HTRF3), hydrothermaled rice flour for 5 h (HTRF5), rice starch (RS), hydrothermaled rice starch for 3 h (HTRS3) and hydrothermaled rice starch for 5 h (HTRS5)

RVA parameters	RF	HTRF3	HTRF5	RS	HTRS3	HTRS5
Peak vis^1 (cP)	5106^b	4820^c	4733^d	7873^a	3151^e	2664^f
Pasting temp °C)	78.1^c	78.4^c	84.4^a	77.7^d	78.9^c	80.4^b
Trough vis (cP)	2342^b	2305^b	2162^c	2759^a	2334^b	1996^d
Final vis (cP)	5156^f	5852^d	6013^c	7094^a	6769^b	5459^e
Setback (cP)	4751^a	4464^b	3297^e	2396^f	3517^d	4017^c

Values are the average of triplicates ± standard deviation. Averages followed by different superscript in the same row are statistically different (p < 0.05).

¹Vis: viscosity.

The higher pasting temperature of RF than RS indicates that the starting point of starch gelatinization^[28] occurred at a higher temperature for the RF. The higher protein and lipid contents of RF can increase gelatinization temperature and hence, enhance its pasting temperature.

HT increased the pasting temperature of both samples. The increase in the pasting temperature indicates that more forces and crosslinkages were present in the HT samples which require a higher temperature to dessociate and the samples were strengthened by HT. The modified RF had higher pasting temperature than their RS counterparts. Re-inforcement of the granules as a result of HT, reduction in the water uptake due to the presence of more lipid and proteins in the RF and possibly the morphological changes of starch granules and proteins (see Fig. 1) can elevate the pasting temeparture of the modifed RF. Increasing the HT time from 3 to 5 h enhaced pasting temperature of both samples. Therefore, an increase of 0.32 and 7.99% for HTRF3 and HTRF5 and 1.60 and 3.47% for HTRS3 and HTRS5 were observed.

The results (Table 2) showed that the RS had higher trough viscosity than the RF since the starch molecules can form more interactions with water resulting in higher viscosity than RF. However, in the RF, proteins and lipids hinder starch and water interactions and reduce the trough viscosity. With increasing the crosslinkages between starch molecules as a result of HT treatment, starch–water interactions decrease resulting in reduced trough viscosity. However, the HTRF showed higher trough viscosity compared to the HTRS. Increasing the treatment time further reduced the trough viscosity. Trough viscosity of the HTRF3 and HTRF5 had a reduction of 1.57 and 7.68% for HTRF3 and HTRF5 and 15.38 and 27.64% for HTRS3 and HTRS5.

Setback and final viscosities are indicative of the retrogradation tendency of amylose.^[28,29] Based on the results in Table 2, the RF had lower final viscosity and setback than RS. The presence of higher protein and lipid in the RF can prevent a strong starch gel network formation resulting in lower final viscosity.

It was observed that HT increased setback and final visocity of the RF but reduced those of RS. Using similar condition for HT, it was found that the final viscosity of cassava and Pinhão starches increased while that of RS reduced.^[9] HT significantly reduced the setback of the RS by 6.05% and 30.61% for HTRS3 and HTRS5, respectively. However, significant increases of 46.75% and 67.59% in the setback of HTRF3 and HTRF5 were observed. Similar changes in the pasting properties of different starches have been reported.^[7,29,30]

The increase in final viscosity is an indication of more ordered structure and possibly more interactions between the moleules in the flour (starch, protein, and lipid). In addition, amylose content has a considerable effect on the final viscosity and it has been found that the starches with higher amylose content showed lower final viscosity after HT, while those of low amylose content exhibited higher final viscosity.^[26] Therefore, differences in the final visosity of the RF and RS may be related to their differences in chemical compositon and amylose content, solubility, degree of order during cooling, and gel integrity. Increasing the treatment time had significant effect on the final viscosity of the samples. Reductions of 4.57 and 23.04% in the final viscosity of HTRS3 and HTRS5 were observed, while increases of 13.49 and 16.63% in the final viscosity of HTRF3 and HTRF5 were obtained.

Textural Properties

During gelatinization, hydrogen bonds between starch molecules are broken and replaced by hydrogen bonds with water. However, on cooling, starch gel is formed as water is trapped by crosslinked starch molecules. Starch and amylose content, volume, and deformation of the granules and interactions between the continuous and dispersed phases have great impacts on the textural properties of starch gels.^[31] Table 3 shows textural properties of the samples before and after modification. The RS gels were firmer, more elastic and cohesive compared to the RF gels. The differences between the textural properties of RF and RS gels can be related to the higher starch content, water absorption, and swelling capacity of RS. The lipids and proteins in the RF can also delay a strong gel network formation and reduce cohesiveness and elasticity of the RF gel since they can interact with starch molecules, reduce water uptake, and obstacle the interactions between starch molecules and water.

TABLE 3 Textural properties of rice flour (RF), hydrothermaled rice flour for 3 h (HTRF3), hydrothermaled rice flour for 5 h (HTRF5), rice starch (RS), hydrothermaled rice starch for 3 h (HTRS3), and hydrothermaled rice starch for 5 h (HTRS5)

Parameters	RF	HTRF3	HTRF5	RS	HTRS3	HTRS5
Hardness (kg)	0.031 ± 0.002^e	0.133 ± 0.007^b	0.148 ± 0.008^a	0.041 ± 0.007^d	0.079 ± 0.006^c	0.071 ± 0.014^c
Elasticity	0.855 ± 0.031^c	0.946 ± 0.007^b	0.954 ± 0.028^ab	0.911 ± 0.032^b	0.945 ± 0.016^b	0.972 ± 0.013^a
Cohesiveness	0.759 ± 0.011^b	0.704 ± 0.014^c	0.707 ± 0.013^c	0.880 ± 0.022^a	0.887 ± 0.028^a	0.908 ± 0.020^a

Values are the average of triplicates ± standard deviation. Averages followed by different superscript in the same row are statistically different (p < 0.05).

Following HT, an increase in gel hardness and elasticity of both samples were observed. The cohesiveness of the RF decreased while the cohesiveness of the RS remained unchanged after the treatment. Increase in gel hardness has been reported for HT millet flour and starch (produced at 100°C for 16 h),^[14] cassava (produced at 100 or 120°C for 2 h)^[9] and RS (produced at 90–110°C for 3–7 h).^[32] On the contrary, a reduction in the gel hardness of RS after HT (produced at 100 or 120°C for 2 h) has been found.^[9] The differences in the findings from previous studies can be attributed to the different HT conditions and starch origin.

The changes in the textural properties of the samples following HT can be related to the increase in the crystalline regions of the granules, increase in the crosslinkages between starch molecules, reduction in the granules water absorption and swelling, formation of amylose-lipid and amylose-protein complexes, particularly in the RF.^[32] Possible changes in the RF proteins as a result of HT may also affect gel texture that needs further investigation. The results also showed no clear pattern for textural parameters of the samples with increasing the treatment time from 3 h to 5 h. Reduction in the cohesiveness of the modified RF can be related to the presence of more crystalline regions due to the formation of amylose-lipid and amylose-protein complexes which can reduce the unoiformity of the gel and hence reduce the overall cohesiveness. The HT condition used in this study had no significant effect on the cohesiveness of RS. Changes in the textural properties of the samples following HT (ΔE1 and ΔE2) were higher for HTRF than HTRS. In addition, the effect of HT was more obvious on gel hardness than elasticity and cohesiveness of both samples. Increasing the treatment time from 3 to 5 h slightly affected the textural properties of the samples.

Conclusion

While the main component of RF is starch, RF is substantially cheaper than RF, and hence, there is a growing interest to use RF instead of RS where possible. It was known that the HT can change the physicochemical properties of RS, but little information is available to show such effects on RF. This study showed how HT under identical condition can affect the physicochemical properties of RF and the starch extracted from the same flour. It was found that RF exhibited different behaviors compared to its related starch. Increasing the HT time from 3 to 5 h enhanced the effects of the treatment particularly on RS. The modified RF had lower water solubility, water absorption, and peak viscosity than the modified RS. However, the modified RF showed higher pasting temperature, gel hardness, and elasticity compared to RS. The higher protein and lipid contents of the RF compared to the RS and the morphological changes of starch granules caused by HT can be the main reasons for such differences which require further investigations.