A Comparison of Selected Quality Attributes of Flours: Effects of Dry and Wet Grinding Methods

Abstract

Studies were conducted on the selected quality attributes of rice flour obtained by dry and wet grinding methods. Raw and parboiled rice have been dry ground at a moisture content of 12%, while wet grinding is performed at 30%. The functional properties of ground samples and doughs made out of the flours were characterized. The damaged starch contents in the dry ground rice flour were 21.5 and 22.6% for raw and parboiled samples, respectively; whereas, these were only 8.7 and 9.5% for wet ground samples. The dry ground rice flour formed sticky dough having high stickiness values but it was less cohesive in nature compared to those of wet ground samples. The doughs made from wet ground rice were more adhesive than the dry ground samples; the former was suitable for forming extrusion processing for preparing various convenience foods like pasta and noodles.

Keywords:

• Particle size
• Damaged starch
• Viscoamylograph
• Stickiness

INTRODUCTION

Rice is a prominent cereal in the diet of nearly half of the world population. Although, it is normally cooked in the form of discrete grains, nearly 10% of it is converted into flour for the preparation of various foods particularly snacks. Generally, products from cereal and legume based batters or doughs are an important part of the human diet in Southeast Asia, the Middle East, and Africa. Rice flour is used to produce several traditional Oriental foods, such as rice noodles and rice cakes, and fermented products. The utilization of rice flour has been broadened in developed countries and the market for rice flour is growing. Jomduang and Mohamed[1] studied the effect of raw materials, milling methods, and particle size on the product characteristics of traditional Thai rice-based snack foods. Nishita and Bean[2] investigated the differences in physicochemical and functional properties of the rice flour obtained by different grinding techniques. Effect of particle size on viscoamylographic behavior of rice flour in the preparation of vermicelli was studied by Hemavathy and Bhat.[3] These researchers suggested that the flours with particle sizes between 138 and 165 μm are suitable for making vermicelli with good textural attributes. For wet grinding, Solanki et al.[4] reported that starch damage is low for blackgram and raw rice, whereas parboiled rice samples exhibit high values. Still data on the quality characteristics of rice flour and dough in relation to their product characteristics are scanty. Besides, the differences due to variety and particle size also contribute to the behavior of rice flour, and in turn, the product characteristics. Therefore, consideration of the type of grinding equipment employed to prepare rice flour seems important from the point of formulation of products and process standardization. There is a need to know the quality attributes of rice flour prepared by dry and wet grinding methods, which are usually followed for the preparation of some of the extrusion formed traditional deep fat fried snacks.

A moistened blend of rice and pulse flour, along with other additives, is shaped in a forming extruder to obtain thin or thick strands, which on subsequent frying provides a crisp snack with attractive taste. In some products, parboiled rice is used while the use of raw rice is common in most of these products. A critical step in the making of dough for these traditional snack foods is wet grinding of rice. Dry grinding is a mere size reduction operation whereas wet grinding involves both physical and chemical changes.[5] During wet grinding of cereals, the protein matrix holding the starch granules is destroyed releasing starch from the protein network. It is largely unknown which grinding method (dry or wet) is suitable for a particular product making. The usefulness of such understanding lies in product development and improvement, and selection of appropriate processing steps. The objective of the present work was to study the effect of dry and wet grinding methods on some physicochemical properties of ground samples, and dough characteristics of rice flour in relation to fried product quality.

MATERIALS AND METHODS

Materials

Paddy (IR 64 variety) was procured from National Seeds Corporation, Mysore, India. A portion of paddy was parboiled by soaking in hot water (80°C) for 5 h, followed by steaming at ambient pressure for 5 min to obtain mild parboiled paddy after subsequent drying. Raw and parboiled rice samples were obtained by shelling and milling the raw and parboiled paddy, respectively, using a laboratory model rubber roller sheller (Satake Corporation, Tokyo, Japan), and a laboratory model rice polisher (McGill, Houston, USA), respectively. Blackgram (Phaseolus mungo) dhal (split cotyledons) was procured from the local market. Both raw and parboiled rice were pulverized separately in a laboratory model vertical disc mill (Sarovar, Mysore, India) to obtain flour that pass through a 60 mesh (aperture size 250 μm) British Standard (BS) sieve.

Raw and parboiled rice were soaked separately in water (1:5) at room temperature for 8 h. Excess water was drained off and the samples were ground in a stone mill (Shanti Wet Grinder, Coimbatore, Tamil Nadu, India). A pre-decided quantity of water was added gradually during grinding to facilitate grinding and to obtain thick and smooth dough (53% moisture content, wet basis).

Blackgram flour was obtained by roasting dehusked split blackgram dhal at 120°C for 10 min in a rotary grain roaster until it developed the characteristic toasted aroma. The roasted dhals were then ground in the above mentioned disc mill to pass through a 85 mesh BS sieve (aperture size 180 μm). The proximate composition, as determined using AOAC[6] methods on triplicate samples, is shown in Table 1.

Table 1 Proximate composition of raw rice, parboiled rice, and blackgram flour

Parameter (% db)	Raw rice	Parboiled rice	Blackgram
Moisture	11.7 ± 0.2	11.1 ± 0.4	6.5 ± 0.2
Protein ^a	5.8 ± 0.1	6.2 ± 0.2	23.8 ± 0.7
Fat	0.7 ± 0.1	0.8 ± 0.1	1.3 ± 0.1
Ash	0.4 ± 0.1	0.4 ± 0.1	3.3 ± 0.1
Carbohydrate (by difference)	81.4	81.5	65.1
Amylose equivalent ^b	28.6	28.6	—
^aN × 6.25; and ^b generally called total amylose or amylose content.

Sample Preparation

The wet ground rice samples were treated with an excess quantity of 80% aqueous ethanol and centrifuged (∼1100 g) to remove the supernatant. The residue obtained was spread as a thin layer and was dried in shade at the room temperature (about 28°C) for 12 h; this prevented agglomeration and lump formation during drying. The dried samples were converted into fine powder as mentioned by Solanki et al.,[4] and the same was used for further analysis.

Particle Size Analysis

Particle size distribution (Table 2) of rice flours (raw and parboiled with both wet and dry grinding methods) was determined in duplicates using a vibratory sieve shaker with a set of sieves comprising 60 (250 μm), 72 (210 μm), 85 (180 μm), 100 (150 μm), 120 (125 μm), 170 (90 μm), 200 (75 μm), and 240 (63 μm) mesh BS sieves. Mass mean diameter was then calculated for the flours.[7] For wet ground samples, the process of treating with alcohol followed by drying (as mentioned earlier) was used prior to sieving.

Table 2 Sieve analysis of rice flour and content of damaged starch

Sample	Grinding method	Different fractions of flour (%) retained on mesh aperture (μm)									Average particle size (μ)	Damaged starch (%)
		210	180	150	125	105	90	75	63	53
Raw rice	Wet	2.0	5.0	25.0	10.8	20.1	8.0	8.1	8.1	13.0	121.1 ± 1.2^a	9.5 ± 0.2^b
	Dry	—	0.2	1.4	4.2	20.1	5.3	17.5	16.1	35.2	82.7 ± 1.9^b	22.6 ± 0.3^d
Parboiled rice	Wet	3.0	7.0	26.0	11.5	24.0	4.0	7.0	6.5	11.0	127.4 ± 0.9^d	8.7 ± 0.3^a
	Dry	0.2	0.3	11.9	13.5	30.0	6.9	12.6	7.8	16.9	106.0 ± 2.2^c	21.5 ± 0.2^c

Damaged Starch

The damaged starch content of wet and dry ground flours was determined according to AACC method #76–30A[8] by determining the content of reducing sugars in terms of ferricyanide reduced after hydrolysis by fungal amylase enzyme.

Dough Preparation

Doughs were prepared by using dry and wet ground rice flour (raw and parboiled) and also their blend containing 20% toasted blackgram flour by adding appropriate quantity of water to get the desired moisture content (44–53%). The dough was mixed for 2 min in a Hobart mixer at a low speed to obtain a homogeneous sample. Cylindrical (35 mm in diameter and 20 mm in height) samples were prepared as mentioned by Bhattacharya et al.[9] for rheological testing and the process of dough preparation was replicated thrice.

Rheology of Dough

The cylindrical dough samples, as prepared earlier, were compressed up to a strain of 0.5 for two times successively at a crosshead speed of 20 mm min⁻¹ employing a Universal texture testing machine (Model # 5R, Lloyd, UK) having a full load scale of 100 N. The software provided by the manufacturer was used to calculate different textural parameters like hardness (peak force during first compression), cohesiveness (ratio of the positive areas of second and first compression curves), and adhesiveness (area of the curve below the datum line at the end of first compression cycle). The stickiness of the dough in degrees was determined as mentioned earlier by Bhattacharya and Narasimha.[10] The reported values are the mean of three observations.

Viscoamylograph

Viscography of dry and wet ground (raw and parboiled) flours was carried out by following the method of Hallick and Kelly[3,11] by employing a Brabender visco-amylograph using a dispersion containing 10% flour solids (dry basis). The sample was heated up to 95°C at the rate of 1.5°C/min, held at 95°C for 20 min and cooled at 1.5°C/min to 50°C. Duplicate runs were made for each sample.

Frying of Sample

Dough strands were made using rice and blackgram flours (4:1) with 2% salt and water followed by shaping in a laboratory model pasta press (forming single screw extruder) with a 10 mm diameter star shaped dies, and finally fried in refined groundnut oil at 170°C for 5 min. Hardness of fried sample was determined by employing the same texture measuring system using a three-point bending set up. The span length was 40 mm, while the crosshead speed was 50 mm min⁻¹. The peak force obtained at the point of failure was taken as hardness of the sample. Five samples were tested each time.

Microstructure

The samples used for examination under a scanning electron microscope were prepared according to the method mentioned by Bhat and Bhattacharya.[12] The samples were viewed at different magnifications of 100X, 1kX and 5kX, and the representative photomicrographs were obtained. All examinations were made at an accelerating voltage of 15 kV using a scanning electron microscope (Model #435VP, Leo Electron Microscopy Ltd., Cambridge, UK).

RESULTS AND DISCUSSION

Damaged Starch and Particle Size

Starch damage for raw rice was markedly higher for dry ground samples compared to wet grinding (22.6 and 9.5%, respectively) (Table 2). In the case of dry grinding employing a disc mill, shear force is a dominating factor that induces rapid reduction in particle sizes. The mechanical conversion of larger pieces into smaller ones results in high fracturing in the starch granules leading to high levels of damage. The finer is the ground particle, the higher is the starch damage. This is probably due to the fact that greater size reduction requires more grinding energy which results in higher damage to starch granules, as well.[4]

In the case of wet grinding, size reduction takes place owing to the separation of the starch granules from the protein matrix in which they are embedded. This is a physical phenomenon, involving minor chemical changes as well, but not a mere mechanical operation which is true for dry grinding. During soaking of rice prior to wet grinding, the hardness of rice kernels decreases. When disintegration of protein matrix occurs, it results in the generation of more number of smaller particles after grinding. Due to leaching out of protein, lipid and minerals, and the simultaneous entrance of water, the structure of endosperm becomes weak. Therefore, fine flour, having less damaged starch has been obtained.[13]

Similar studies on dry and wet grinding of parboiled rice in the present investigation shows that dry grinding resulted in more starch damage compared to wet grinding (21.5 and 8.7%, respectively). The starch damage for parboiled rice is not only due to grinding operation but also depends on the pretreatments and processing conditions received by the grains. Here, the extent of softening was not as effective as in the case of raw rice. Due to parboiling, starch granules became hard necessitating high forces for segregation of starch granules from the protein matrix. This eventually leads to its greater damage during dry grinding and results in large particles (106.0 μm). It was noticed that wet grinding of raw rice required less time (30 min) than that for parboiled rice (45 min). The average particle size was marginally smaller in case of raw rice (121.1 μm) compared to that of parboiled rice (127.4 μm) (Table 2) but this difference is statistically significant at p ≤ 0.05.

Viscoamylograph

The pasting behavior of samples (obtained by dry and wet grinding of raw and parboiled rice flours) were reported in terms of peak viscosity, breakdown viscosity, set back and final paste viscosities, and set back breakdown ratios (Table 3). The process of pasting can be expressed as the state that is to a large extent associated with gelatinization of starch. The apparent viscosity of starch dispersions in water is strongly influenced by the extent of swelling of starch granules. Starch granules swell radially in the beginning and when the temperature is increased, the amylopectin-rich granules swell tangentially. The granules deformed and lost their original shape. The presence of amylose in the continuous phase surrounding the swollen granules resulted in the formation of a strong gel on cooling.[14] Generally, wet ground flour offered the highest values of all the pasting characteristics of rice paste. Since parboiled rice is an already gelatinized and retrograded sample, starch present in it took more time to cook showing apparently a higher gelatinization temperature. Raw rice flour exhibited the highest peak viscosity of 2300 BU compared to only 440 BU for parboiled sample. An increase in the content of damaged starch resulted in a decrease in gelatinization temperature.[15] The peak viscosity and gelatinization temperature indicated the water binding capacity of starch. The differences of pasting and hydration properties of treated rice flours were attributed to changes in the rigidity of the starch granules.[16] The extent of swelling is regulated by the strength of the internal structure of the granules. The stronger the internal molecular structure, the higher the temperature were required for gelatinization. The drop in amylograph peak viscosity of parboiled rice was due to the pretreatment (parboiling) given to rice which reduced swelling of the starch granules.[17] Raw rice (wet ground) flour exhibited the highest breakdown viscosity (1780 BU) while it was lowest (40 BU) for parboiled (dry ground) rice flour. The retrogradation of amylose in parboiled rice flour offered highest resistance to breakdown; hence, low breakdown viscosity was obtained as it was an indication of paste stability.

Table 3 Pasting characteristics of rice flour

Sample	Grinding method	Gelatinizationtemperature (°C)	Peak viscosity (BU)	Breakdown viscosity (BU)	Setback viscosity (BU)	Final paste viscosity (BU)	Set back ratio	Breakdown ratio
Raw rice	Wet	72.0 ± 1.2^a	2300 ± 30^a	1780 ± 17^a	1400 ± 10^a	900 ± 14^a	0.23 ± 0.02^a	0.39 ± 0.03^a
	Dry	58.5 ± 2.1^b	820 ± 10^b	160 ± 6^b	50 ± 2^b	770 ± 11^b	0.80 ± 0.04^b	0.94 ± 0.02^b
Parboiled rice	Wet	77.0 ± 1.1^c	500 ± 19^c	80 ± 2^c	200 ± 9^c	700 ± 9^c	0.84 ± 0.02^b	1.40 ± 0.08^c
	Dry	70.0 ± 2.6^a	440 ± 8^d	40 ± 1^d	210 ± 8^c	650 ± 21^d	0.91 ± 0.06^b	1.48 ± 0.10^c
Columns with different superscripts are significantly different at p ≤ 0.05 according to Duncan's Multiple Range Test (DMRT).

Raw rice flour (wet ground) gave the highest setback viscosity (1400 BU) and final paste viscosity (900 BU) (Table 3). The final and setback viscosities are an indication of the cooled-cooked paste under a condition of low shear. It appeared that damaged starch, the grinding methods (dry or wet processing) and pretreatments were important factors that markedly affect pasting characteristics of flour.

Microstructure

Rice flour particles possessed a polyhedral structure and were irregular in shape (Fig. 1) with smooth external surfaces; the granular size ranged between 1.5 and 5.8 μm. The dry ground powder particles appeared to be more compacted than the wet ground particles. The photomicrographs of the dough made from both type of powders along with dry ground blackgram appeared to be similar in nature (Fig. 2) with wet ground sample appearing marginally more cohesive characteristics. It might be possible that wet grinding process leached out components that acted as a cementing substance. Bhat and Bhattacharya[12] reported similar observation for thick chick pea (Bengalgram) flour dispersions.

Figure 1 Raw rice flour particles obtained by (A) dry and (B) wet grinding.

Figure 2 Doughs made with blackgram and (A) dry ground and (B) wet ground raw rice flour.

The microstructure of fried products made from these doughs showed a structure with layers when dry ground flour was used while a more porous structure was noticed for wet ground sample (Fig. 3). These pores were created during the stage of frying in which moisture escapes rapidly leaving behind pores.

Figure 3 Microstructure of fried strands made with blackgram and (A) dry ground and (B) wet ground raw rice flour.

The former product (Fig. 3A) was expected to offer more resistive force during compression because it was less porous but showed regular subsequent breaking characteristics such that it appeared to be a snack that was marginally soft to bite (non-significant at p ≤ 0.05) but crispier indicating multiple fractures. On the other hand, fried sample made out of wet grinding appeared as a harder product having irregular biting characteristics. It might be possible that the wet ground dough sample would absorb less oil because the leached materials behaved as a cementing substance that did not allow an easy passage for oil (Table 4).

Table 4 Characteristics of fried snack (murukku)

Dough made using	Moisture content (%)	Hardness (N)	Fat content (%)
Raw rice flour
Wet ground	52.7	8.5 ± 0.2^a	14.9 ± 0.5^a
Dry ground	52.7	8.0 ± 0.2^b	16.0 ± 0.3^b
Parboiled rice flour
Wet ground	52.7	5.8 ± 0.2^c	25.1 ± 0.2^c
Dry ground	52.7	4.6 ± 0.3^d	25.8 ± 0.3^d
Columns with different superscripts are significantly different at p ≤ 0.05 according to Duncan's Multiple Range Test (DMRT).

Fried Product

Coarse particles during wet ground method (121 μm and 127 μm, respectively for raw and parboiled rice) produced a snack product with low fat content (∼15% & 25%, respectively, for raw and parboiled rice) compared to the dry grinding method (Table 4). Therefore, grinding methods and their particle size significantly (p ≤ 0.05) influenced the frying behaviour and final product quality and associated with low oil content in the product (Table 4).

Rheology of Dough

Since wet and dry ground rice flour is ultimately used for making several products including traditional deep fat fried extruded snacks along with blackgram flour, the rheological behavior of rice-blackgram blended dough at different moisture contents (44.4 to 52.7%) was conducted (Table 5). An increase in moisture content of the dough decreased hardness and adhesiveness while cohesiveness and stickiness were increased (significant at p ≤ 0.05). These results indicated that adequate level of moisture is required to have a cohesive dough while excess water present in dough act as a lubricant. The dough prepared from dry ground rice flour imparted more hardness than wet ground samples at the same moisture content. Higher stickiness values (115–175°) were observed (Table 5) for dry ground flour doughs than that for wet ground samples (100–145°). On the other hand, adhesiveness of dry ground flour doughs was markedly higher compared to wet ground sample. This may be due to higher percentage of damaged starch content in the dry ground flour. The cohesiveness of wet ground dough at the same moisture content was higher than that of dry ground dough. This property of wet ground dough is important for shaping purposes employing an extrusion system or a sheeting machine, and to get continuous strands that retain the integrity, as well as the shape of the product. Dry ground rice flour dough possessed a high stickiness value, which is undesirable as this dough sticks to the surface of the extruder dies, other machine parts during sheeting and to containers during collection and transferring. Thus, wet ground dough imparted the desirable softness with adequate cohesive characteristics to make it suitable for preparing deep fat fried extruded or shaped snacks.

Table 5 Textural parameters of rice-blackgram dough

Sample	Grinding method	Moisture content (%)	Hardness (N)	Cohesiveness (-)	Adhesiveness (mJ)	Stickiness (degree)
Raw rice	Wet	44.4	36.3 ± 0.3^a	0.07 ± 0.00^a	7.1 ± 0.5^a	100 ± 2^a
		47.1	18.5 ± 0.2^b	0.13 ± 0.01^b	5.2 ± 0.8^b	120 ± 3^b
		50.0	12.0 ± 0.4^c	0.21 ± 0.02^c	4.4 ± 0.2^b	122 ± 2^b
		52.7	11.9 ± 0.2^c	0.55 ± 0.02^d	4.3 ± 0.4^b	125 ± 3^b
	Dry	44.4	67.6 ± 0.3^a	0.11 ± 0.01^a	12.7 ± 0.3^a	115 ± 3^a
		47.1	50.7 ± 0.4^b	0.14 ± 0.01^b	10.8 ± 0.7^b	125 ± 3^b
		50.0	23.8 ± 0.2^c	0.17 ± 0.02^b	6.0 ± 0.3^c	150 ± 2^c
		52.7	18.3 ± 0.5^d	0.18 ± 0.03^b	4.3 ± 0.0^d	175 ± 4^d
Parboiled rice	Wet	44.4	34.2 ± 0.2^a	0.14 ± 0.01^a	9.2 ± 2.3^a	105 ± 3^a
		47.1	22.4 ± 0.7^b	0.21 ± 0.01^b	6.3 ± 0.1^b	119 ± 4^b
		50.0	9.3 ± 0.3^c	0.27 ± 0.03^c	3.2 ± 0.1^c	140 ± 4^c
		52.7	9.6 ± 0.8^c	0.28 ± 0.02^c	3.4 ± 0.1^c	145 ± 2^c
	Dry	44.4	38.9 ± 2.7^a	0.11 ± 0.01^a	10.8 ± 0.2^a	120 ± 3^a
		47.1	30.9 ± 0.1^b	0.16 ± 0.01^b	7.8 ± 0.6^b	130 ± 3^b
		50.0	29.6 ± 0.2^c	0.17 ± 0.02^b	7.6 ± 0.6^c	170 ± 2^c
		52.7	17.0 ± 0.9^d	0.26 ± 0.01^c	50.3 ± 0.3^d	175 ± 4^c
Columns with different superscripts among same raw materials and treatments are significantly different at p ≤ 0.05 according to Duncan's Multiple Range Test (DMRT).

The tendency to adhere to a contact surface is generally called stickiness.[18] It can be perceived in the palate, teeth, and tongue when the food is being masticated and can also be perceived on non-oral surfaces such as fingers and equipment surfaces. Although stickiness to some extent may be a welcome sensory and processing attribute it is often a problem in food industries handling batters, gels and doughs. There is no general consensus as to what factors and forces are involved in stickiness. Some authors attribute it to adhesive force, some see it as being attributable to a combined effect of adhesive and cohesive forces, and others include viscosity and viscoelasticity as well. The study of stickiness and its quantification is an important issue in the cereal industries. Water is a plasticizer as it induces softening and increases the flexibility of a polymer brought about by the incorporation of a plasticizer. The increase of water content results in a sharp decrease in the glass transition temperature.[18]

Rice is extensively used in various forms of food as whole grain as much as flour. Many people in South and Southeast Asia use rice flour both from brown and milled rice to prepare various conventional and novelty food items.[19] Every rice based food product needs unique set of processing conditions. In order to predict appropriate process, the knowledge of various types and forms of rice and their properties are essential for the whole range of processing conditions.

CONCLUSIONS

Wet and dry ground rice flour differs in the quality attributes. Wet grinding resulted in less damaged starch compared to the sample obtained by dry grinding. The damaged starch contents in the dry ground rice flour were 21.5 and 22.6% for raw and parboiled samples, respectively, whereas these were only 8.7 and 9.5% for wet ground samples. The starch damage for parboiled rice was not only due to grinding operation but also depended on the pretreatments and processing conditions received by the grains. Wet ground dough gave desirable hardness (9.6–11.9 N), cohesiveness (0.28–0.55), adhesiveness (3.4–4.3 mJ), and stickiness (125–145 degree) to make it suitable for the extrusion forming process. Wet ground flour showed prominent pasting characteristics with high peak and final paste viscosities (2300 and 900 BU for raw wet ground sample, respectively). Raw rice flour exhibited the highest peak viscosity of 2300 BU compared to only 440 BU for parboiled sample. Raw rice (wet ground) flour exhibited the highest breakdown viscosity of 1780 BU while it is lowest (40 BU) for parboiled (dry ground) rice flour. The retrogradation of amylose in parboiled rice flour offered highest resistance to breakdown; hence, low breakdown viscosity was obtained as it was an indication of paste stability. Thus, the application of fine rice flour with low damaged starch obtained by wet grinding is useful in the formulation of some fried snacks.