Non-linear viscoelastic behavior of cooked white, brown, and germinated brown Thai jasmine rice by large deformation relaxation test

ABSTRACT

Stress relaxation tests at high strain were conducted on scoops of cooked white, brown, and germinated brown Thai jasmine rice using a King Mongkut’s Institute of Technology Ladkrabang test rig. The diameter of the scoop was 35 mm and the height was 10 mm. Non-linear modeling, consisting of four relaxation models, was applied to the data obtained for each type of rice. The modeling methods included Peleg and Normand’s; Yadav, Roopa, and Bhattacharya’s; Jaya and Durance’s; and Myhan, Markowski, and Daszkiewicz’s. The cooked white rice showed greater tenderness compared to the others. The toughness of the three types of cooked rice was not found to be different. The Myhan et al. model was the most accurate in describing the non-linear viscoelastic behavior of all types of cooked rice. The cooked brown rice showed the highest initial decay rate, but the lowest relaxation, lowest elasticity, and greatest viscosity. In contrast, the cooked white rice had opposite characteristics.

KEYWORDS:

• White rice
• Brown rice
• Germinated brown rice
• Thai jasmine rice
• Cooked rice
• Non-linear viscoelastic

Introduction

Kao Dok Mali 105 (KDML105) is a well-known Thai fragrant rice (Oryza sativa, L.) variety produced in Thailand and has been exported throughout the world. It is included in “Khao Hom Mali” in Thai, in which “Khao” means “rice,” “Hom” means “fragrant,” and “Mali” means “jasmine.” Another variety of “Khao Hom Mali” is RD15. Internationally “Khao Hom Mali” is “Thai jasmine rice.” The common uses of the rice are milled rice, brown rice, and germinated brown rice. In 2015, Thailand exported 1.340 and 0.015 million tons of Thai jasmine and brown rice, with a value of 1200 and 16 million USD, respectively.^[1] Although germinated brown rice is not currently exported abroad, it is known to be highly nourishing. It has been supported by the Department of Industrial Promotion as a product with added economic value.^[2] In recent of years, Vietnam and Cambodia have made significant headways into the jasmine rice market.^[3] In 2015, Vietnam and Cambodia have exported approximately 1.2^[4] and 0.5^[5] million tons of jasmine rice, respectively.

Brown rice and germinated brown rice consists of the seed coat, embryo, and endosperm, which collectively contain many useful nutrients. In milled rice, the seed coat is polished off leaving the endosperm whose main constituents are carbohydrates. In germination, during malting, biochemical processes result in the carbohydrates being converted into oligosaccharides and reduced sugars.^[6–9] The composition of the grains affects its texture and eating quality. Factors that affect the texture quality of cooked rice include: rice variety, amylose content, pasting properties, protein content,^[10,11] post-harvest practice, drying condition, moisture content, temperature, and storage period.^[12] The texture of cooked rice also depends on the cooking method and the water to rice ratio.^[13] For aged rice of the KDML105 variety, the pasting properties of rice flour were determined by a rapid visco analyzer.^[14]

Milled or white rice dominates the consumer market due to its soft texture and fragrance after being cooked. Brown rice is nutritionally superior to milled rice in terms of protein, dietary fiber, vitamin B, minerals, and even functional antioxidants, all of which are mostly found in the external parts like the hull and the bran.^[15] Due to increasing health concerns in society, the health benefits of brown rice have received the attention of consumers, and the demand for brown rice has led consumption to increase greatly.^[16] The germination process activates residual enzymes in brown rice, inducing the formation of various metabolic components having bioactive functions. Thus, germinated brown rice has also gained significant attention during the last decade in terms of enhancing the eating quality and potential health benefits.^[17] Germinated brown rice has become popular among health-conscious consumers, such as diabetics, hypercholesterolemics, and hypertensive patients, for the most part due to the presence of bioactive compounds.^[17]

The texture of cooked Thai jasmine rice and its derived products have not been reported in the literature. Texture profile analysis has been shown as an effective method for evaluating the quality of meats under different modes; however, the consistency are not always satisfactory.^[18] The stress-relaxation test was applied on meat product and it proved that, in general, the stress-relaxation test shows improved consistency.^[18] In addition, the stress relaxation test can be used as a tool to systematically study food functionality, with examples including cheese,^[19] sago-wheat mixtures gel,^[20] and mango gels.^[21]

The viscoelasticity of food can be analyzed by rheological models. There is a simple test for obtaining the rheological model, i.e., stress relaxation test, where the constant strain is applied on the specimen. The test is divided into two categories, low strain (usually less than 5%) for linear viscoelastic behavior and high strain (more than 5%) for non-linear viscoelastic behavior. The well-known generalized Maxwell model explains the former. There have been several research articles that have reported on the generalized Maxwell model of several foods, such as cassava dough,^[22] dried persimmons,^[23] moth bean flour dough,^[24] date fruits,^[25] dates at khalal and rutab stages of maturity,^[26] Dutch-type cheese,^[27] muscle tissues from gilthead sea bream,^[28] and composite dough with wheat and mesquite (Prosopis spp) flours.^[29] However, food generally exhibits non-linear viscoelastic behavior when subjected to high strain due to their processing methods for human consumption. Examples include dough,^[24] tofu and gellan gum gels,^[30] alginate gels,^[31] and processed pork and poultry products.^[32] No studies on cooked white (milled), brown, or germinated brown Thai jasmine rice have previously been reported. Therefore, the objective of this research was to investigate the texture properties using the compression test and to apply and compare four different non-linear stress relaxation models (i.e., Peleg and Normand,^[33] Yadav, Roopa, and Bhattacharya,^[34] Jaya and Durance,^[35] and Myhan, Markowski, Daszkiewicz, Zapotoczny, and Sadowski^[32]) to characterize the viscoelastic behavior of cooked Thai jasmine white, brown, and germinated brown rice.

Materials and methods

Rice samples and cooked rice preparation

The white, brown, and germinated brown rice samples of Oryza sativa L., Thai jasmine rice, were purchased from a local department store in Bangkok, Thailand. Home electronic rice cookers (RC-10 MM, Toshiba, Thailand) were used to cook 200 g of each type of rice sample using different water to rice ratios as recommended by rice producers, i.e., 2.5:1 for brown rice; 1.6:1 for germinated brown rice; and 1:1 for milled rice. The recommended water-to-rice ratio were used to obtain a typical texture of cooked rice consumed by consumers. After the rice was fully cooked, and the rice cooker automatically switched to the warm mode, the pot of cooked rice was carefully put upside down on a screen and covered with a plastic lid and left for 30 min. The outer surface of the rice block was scraped out and only the middle portion was randomly sampled and put into plastic cup consisting of approximately 5 g. In total, 9 cups per sample were generated. The cooked rice samples were then subjected to the stress relaxation test. The rice samples were also measured for moisture content by the gravimetric method using a hot air oven at 105°C. Samples were heated until a constant weight was observed. Five replicates were performed for each sample.

Instrumental texture test

King Mongkut’s Institute of Technology Ladkrabang (KMITL) texture test rig

The KMITL texture test rig was designed for food texture testing as shown in Fig. 1. It consisted of two aluminum parts, including the cylindrical compression probe with 30 mm in diameter and a cylindrical hole with 35 mm in diameter and 20 mm in height. This configuration represented the size of the sample of a normal quantity of food put into a typical mouth while the compression area is approximately constant and equal to the cross section area of the probe.

Figure 1. KMITL texture test rig.

Simple compression and stress relaxation tests

The simple compression and stress relaxation tests were performed using a hold distance (HLDD) test which was performed using a texture analyzer (HD Plus, Texture Analyzer, Stable Micro Systems, UK) with a 50-kg load cell. The 5 g of cooked rice sample in KMITL test rig was compressed up to a 20 N maximum force using a deformation speed of 5 mm/s. This corresponded to the simple compression test. Then the probe was kept at this position, and the relaxation test was performed. In relaxation test, the plate was kept at the maximum force for 120 s relaxation time, and the decline in force was recorded with time using Exponent version 6,1,5,0 (Stable Micro Systems, UK). The pre- and post-test speed was 5 mm/s. From the tests, the typical graph of force and time was obtained. In the simple compression test, the texture parameters including strain at maximum force, initial firmness, average firmness, and toughness were calculated using the same software. The initial firmness (N/mm) and average firmness (N/mm) was calculated from the ratio of force and corresponding deformation at 2 N and at maximum force (20 N), respectively. The toughness (N mm) was determined as the area under curve from origin to the maximum force point. In addition, from force-time curve, the relaxation ratio at 120 s was calculated by dividing the relaxation force at that time by the maximum applied force.

Non-linear viscoelastic modeling

Peleg and Normand^[33] explained the viscoelastic property of materials with the linear Eq. (1).

(1)

where is relaxation time. is the decay parameter which can be the normalized force, stress or modulus of elasticity. In our case we used the normalized force.

(2)

is the force at a relaxation of zero and is the force at time t. and are constants which do not depend on the relaxation time. is the degree of solidity. When increases the elasticity of sample increases accordingly. At infinity, the sample is an ideal elastic body and there is no relaxation. When it is one, the sample is an ideal liquid which is totally relaxed. is the initial decay rate and is the hypothetical asymptotic level of normalized relaxation parameter.

Jaya and Durance^[35] developed a relaxation model as follows:

(3)

, is the level of stress that decays during the relaxation, and , is the rate of relaxation curve toward the asymptotic residual level, both of which are constants. When is zero, the sample behave as an ideal elastic body and if is one, the sample is an ideal liquid. A low value of means the stress relaxation is slow to reach residual stress. Yadav et al.^[34] proposed the following relaxation model:

(4)

where is the residual force and and are two empirical constants. is equivalent to when is zero and when is large, will approach . The represents the rate of exponential decay. Myhan et al.^[32] modified the stress relaxation model of Peleg and Normand.^[33] When stress applied area is constant, the equation becomes as follows:

(5)

The non-linear viscoelastic modeling was conducted using the Gauss-Newton algorithm in the R statistical package.^[36] The constants of optimum models from Eqs. (1, 3, 4, and 5) were obtained with the minimum residual standard error (RSE). The coefficient of determination (R²) was also calculated.

Statistical analysis

The mean and standard deviation associated with the nine replicates were calculated for the strain at maximum force, initial firmness, average firmness, toughness, and relaxation ratio. The significance of differences between the treatment means were determined using Duncan’s multiple range test (DMRT)^[37] at a confidence level of 95% (p < 0.05).

Results and discussion

The moisture content of the white rice, brown rice, and germinated brown Thai jasmine rice samples were 54.69 ± 0.97, 70.68 ± 1.51, and 56.86 ± 1.43 %wb, respectively. Figure 2 shows the average force-time curves obtained using the simple compression test and relaxation test of cooked white rice, brown rice and germinated brown Thai Jasmine rice. Because each curve was averaged from nine samples, and the maximum force (20 N) was reached at different time, the average value was slightly lower than 20 N. The curves associated with the simple compression test, where the strain was increased, did not show any difference among the different rice samples. Therefore the sections of the curves associated with simple compression where difference was apparent is shown in Fig. 3. From the figure, it can be clearly seen that the white rice was the softest (smallest slope) followed by brown rice and finally the hardest, the germinated brown rice. The texture characteristics extracted from the simple compression test were quantified and are shown in Table 1. shows that the strain applied at maximum force to samples, initial firmness, average firmness, toughness and relaxation ratio of the three types of cooked Thai Jasmine rice. It should be noted that at the same maximum applied force (20 N), the strain associated with white rice was the highest, followed by brown rice and germinated brown rice. This indicated that cooked white rice is softest and cooked germinated brown rice was the hardest. As a result of amylolytic action, germination causes a decrease in the amount of amylose in brown rice.^[17] This is one of the key factors for determining the cooked rice texture. Juliano et al.^[38] showed that hardness was positively correlated with amylose content. In addition, Moongngarm^[39] reported that smooth and densely packed starch granules changed to rough and eroded shapes with germination. This information would indicate that the cooked germinated brown rice should be softer than the cooked brown rice. However, this is opposite to our result where it is found that the cooked germinated brown rice was harder than the cooked brown rice. This could be due to the effect of the water to rice ratio (2.5:1 for brown rice and 1.6:1 for germinated brown rice) used for cooking in this experiment. By adjusting the water to rice ratio, the effect of the malting process in the germinated brown rice production on the texture properties of the rice measured by simple compression test became non-significant. The initial firmness implies the stiffness of the samples.^[40] The white rice had the lowest stiffness compared to the others. The stiffness of brown rice and germinated brown rice was not different. The average firmness is indicative of the tenderness of the cooked rice. The lower average firmness the more tender the rice will be. The cooked white rice was significantly more tender than the others. This might be because the white rice was milled where the seed coat was removed, while the brown rice and germinated brown rice were still covered with the hard seed coat. Brown rice has three tissue layers (endosperm layer, aleurone layer, and cuticular layer) while white rice merely has the endosperm layer.^[41] Comparison between brown rice and white rice indicates that the presence of the bran layer (aleurone layer and cuticular layer) significantly (p < 0.05) increased the hardness and decreased adhesiveness of rice.^[41] The toughness of the cooked rice corresponds to the energy used for deforming the rice. The degree of cohesion between rice grains and its matrices could be the reason. There was no difference in the toughness among these three types of cooked rice. The relaxation ratio indicates the viscoelastic behavior of the material. The lower the relaxation ratio, the higher the elasticity and the lower the viscosity will be.^[42] An ideal elastic body has a zero relaxation ratio,^[43] and the stress cannot be relaxed. The white rice had a significant lower relaxation ratio than brown rice and germinated brown rice, which indicated that it was more elastic and less viscous.

Table 1. Instrumental texture characteristics of cooked Thai jasmine rice.

Thai jasmine rice samples	Strain at maximum force	Initial firmness (N/mm)	Average firmness (N/mm)	Toughness (N mm)	Relaxation ratio
White rice	0.605 ± 0.025^a	0.44 ± 0.07^b	2.41 ± 0.19^b	29.84 ± 1.62^a	0.20 ± 0.01^b
Brown rice	0.573 ± 0.023^b	0.53 ± 0.09^a	2.75 ± 0.28^a	28.94 ± 1.78^a	0.23 ± 0.01^a
Germinated brown rice	0.536 ± 0.029^c	0.61 ± 0.11^a	3.00 ± 0.35^a	28.18 ± 2.03^a	0.23 ± 0.01^a

Mean ± standard deviation are given in the table.

The same letters in the same column mean that the mean values are not significantly different at a confidence level of 95% (p < 0.05).

Figure 2. Average force-time curves obtained from the simple compression test and relaxation test on cooked Thai jasmine rice.

Figure 3. Average force-time curves obtained from the simple compression test of cooked Thai jasmine rice.

The relaxation parameters of the four models associated with the three different types of Thai jasmine rice are shown in Table 2. The Myhan, Markowski, Daszkiewicz, Zapotoczny and Sadowski^[32] model were the most accurate model, with the highest coefficient of determination (R²), 0.948–0.955, and the lowest RSE, 0.203–0.229. Figures 4, 5, and 6 shows a typical experimental result and stress-relaxation curves fitted for cooked white rice, brown rice, and germinated brown rice, respectively, for the four different models, i.e., Peleg and Normand,^[33] Myhan, Markowski, Daszkiewicz, Zapotoczny, and Sadowski,^[32] Yadav, Roopa, and Bhattacharya,^[34] and Jaya and Durance.^[35] It can be seen that for all types of rice, the regression lines of models of Peleg and Normand,^[33] Myhan, Markowski, Daszkiewicz, Zapotoczny, and Sadowski,^[32] and Jaya and Durance^[35] were close to each other with similar R² and RSE.

Table 2. Relaxation parameters of four different non-linear viscoelastic models for cooked Thai jasmine rice samples.

Thai jasmine rice sample	Model type	k1	k2	n	RSE	R^Citation2	A	α	Fe	RSE	R^Citation2		a	b	RSE	R^Citation2
White rice	Eq. (1)	2.05 ± 0.12^a	1.25 ± 0.02^b		0.233 ± 0.007	0.951 ± 0.004
	Eq. (2)	1.81 ± 0.13^a	1.24 ± 0.02^b	0.91 ± 0.04^a	0.229 ± 0.008	0.953 ± 0.004
	Eq. (3)						6.30 ± 0.32^a	0.10 ± 0.01^b	4.41 ± 0.23^b	0.381 ± 0.010	0.870 ± 0.006
	Eq. (4)												0.49 ± 0.03^b	0.61 ± 0.04^c	0.233 ± 0.007	0.951 ± 0.004
Brown rice	Eq. (1)	1.64 ± 0.18^c	1.30 ± 0.02^a		0.208 ± 0.005	0.946 ± 0.003
	Eq. (2)	1.47 ± 0.15^b	1.29 ± 0.02^a	0.91 ± 0.04^a	0.203 ± 0.004	0.948 ± 0.004
	Eq. (3)						5.63 ± 0.14^b	0.12 ± 0.01^a	4.87 ± 0.22^a	0.339 ± 0.006	0.855 ± 0.006
	Eq. (4)												0.62 ± 0.07^a	0.80 ± 0.08^a	0.208 ± 0.005	0.946 ± 0.003
Germinated brown rice	Eq. (1)	1.88 ± 0.18^b	1.30 ± 0.02^a		0.228 ± 0.011	0.947 ± 0.003
	Eq. (2)	1.52 ± 0.16^b	1.28 ± 0.01^a	0.83 ± 0.07^b	0.210 ± 0.011	0.955 ± 0.006
	Eq. (3)						5.63 ± 0.30^b	0.10 ± 0.01^b	4.95 ± 0.21^a	0.362 ± 0.012	0.866 ± 0.007
	Eq. (4)												0.54 ± 0.06^b	0.70 ± 0.07^b	0.228 ± 0.011	0.947 ± 0.003

Equation (1) .

Equation (2)

Equation (3)

Equation (4) when .

R²: coefficient of determination; RSE: residual standard error.

Mean values in a column of the same equation followed by different letters (a–c) differ significantly (p < 0.05).

Figure 4. A typical relaxation curve fitted for cooked white rice by four different models of Eq. (1) ; Eq. (2) ; Eq. (3) , and Eq. (4) when .

Figure 5. A typical relaxation curve fitted for cooked brown rice by four different models of Eq. (1) , Eq. (2) , Eq. (3) , and Eq. (4) when .

Figure 6. A typical relaxation curve fitted for cooked germinated brown rice by four different models of Eq. (1) , Eq. (2) , Eq. (3) , and Eq. (4) when .

The value associated with the white rice was the highest ( was lowest) followed by germinated brown rice and brown rice, respectively. This indicated that the initial decay rate to reach the asymptotic residual stress was slowest for white rice and fastest for brown rice. The value of brown rice and germinated brown rice was not significantly different, yet both of which were higher than that of white rice. This indicated that the asymptotic residual stress of white rice was lower than those of brown rice and germinated brown rice and therefore the degree of elasticity was higher and viscous properties was lower. According to Nicoleti et al.,^[23] the equilibrium or residual stress, i.e., the amount of initial stress that remained unrelaxed, could be associated to the ability of the material to store some fraction of the applied energy. The of Myhan, Markowski, Daszkiewicz, Zapotoczny, and Sadowski^[32] model of all type of rice was closed to one which indicated the effect of was less than . The n of white rice and brown rice was not significantly different and higher than that of germinated brown rice which indicated that the former took less time to reach residual stress. However, the should be the main parameter to indicate the initial decay rate. Because the constant was the reciprocal of , it was clear that the brown rice had the highest rate of initial decay. The higher the constant value, the higher the residual stress was. The value of brown rice was the highest followed by germinated brown rice and white rice. Though the model of Yadav, Roopa, and Bhattacharya^[34] showed the lowest accuracy (lowest R² and highest RSE), its constant (, and ) indicated the viscoelastic behavior of the rice samples corresponded to those constants of the more accurate models.

Cooked rice, including the germinated brown rice, brown rice and white rice are gaining popularity for consumers. Cooked rice is subjected to high strain during production. This study elucidated the non-linear viscoelastic behavior of cooked white, brown, and germinated brown Thai jasmine rice through the use of a large deformation relaxation test. The elasticity and viscosity and other important texture properties of rice have been reported and compared. Accurate non-linear viscoelastic models have been developed. The model constants calculated could be used for interpreting the textural behavior within quality and production control disciplines. In the rice mill, rice quality improvement factory and germinated brown rice production factory, the texture profile analysis and normal compression test are the convention methods used for cooked rice quality control and phenomenon checking before sales. However, the consistency of the test is not always satisfactory and the stress-relaxation test showed improved consistency.^[18] Therefore, the stress-relaxation test should be introduced in a production environment. Using this method, the stress-relaxation modeling and fitting can be performed and the model coefficients, which indicate the viscous and elastic phenomena, initial decay rate and residual force or stress of the cooked rice can be obtained. This information could support the production of more consistent, reliable products. In addition, the KMITL test rig designed to obtain the constant area during texture test proved its utility in terms of the relaxation test of cooked rice. It is expected that it could be applied to other food in the future.

Conclusions

Using a simple compression test, it can be concluded that cooked white Thai jasmine rice was most tender when compared to cooked brown rice and germinated brown rice. There was no difference in the toughness among these three types of cooked rice. It seemed that the effect of the malting process in the germinated brown rice production affects the texture properties. Differences in results between the different samples obtained using the simple compression diminished by adjusting the water to rice ratio in the cooking process. Non-linear regression modeling proved that the Myhan, Markowski, Daszkiewicz, Zapotoczny, and Sadowski^[32] method could provide the best fitted and most accurate model for all types of rice used in this study. Using the stress relaxation test, it was confirmed that cooked brown rice displayed the highest initial decay rate, but lowest relaxation. This suggests it has the highest ability to store a fraction of the applied energy, due to it also having the lowest elasticity and highest viscosity. In contrast, the cooked white rice had the opposite characteristics. In addition, the KMITL texture test rig described in this study proved satisfactory for use in the stress relaxation test where the compression contact area was kept constant. The information gained from this study could prove useful for rice product processing industry, consumers, and researchers.

Acknowledgments

The authors would like to thank the Curriculum of Agricultural Engineering, Department of Mechanical Engineering, Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok, Thailand for of the use of the instruments used in this work.