A relationship between friction plate roughness and the external friction angle of wheat kernels

ABSTRACT

The frictional properties of seeds have been extensively studied, but most researchers have identified only the type of tested friction materials (e.g. concrete, steel, wood) without describing their manufacturing characteristics. The aim of this study was to determine the correlations between the external friction angle of wheat kernels vs. the roughness of the friction plate and the basic physical properties of kernels. The experiment was performed on the kernels of winter wheat var. Jensen obtained directly from spikes and threshed kernels. The external friction angle of every kernel was measured on six friction plates characterized by different surface geometry. Measurements were performed in five replications with a photosensor device which registered the external friction angle of kernels. The basic dimensions and the mass of every kernel were measured, and the results were used to calculate four shape factors for the analyzed kernels. The correlations between the evaluated attributes were determined by one-way analysis of variance (ANOVA) and linear correlation analysis. The roughness of the friction plate significantly influenced the external friction angle of wheat kernels. The average values of the angle of external friction ranged from around 16° to around 24°. The lowest value of the angle of external friction was noted for surface roughness of Ra = 0.93 μm, and the highest value – for surface roughness of Ra = 5.86 μm. Higher values of the angle of external friction were observed in kernels obtained directly from spikes than in threshed kernels.

KEYWORDS:

• Wheat kernels
• physical properties
• steel friction plate
• surface roughness
• correlation

Introduction

The frictional properties of processed products play an important role in agriculture and the food industry. A sound knowledge of frictional properties is essential for simulating and determining the parameters of seed transport, mixing, compaction and processing.^[1–4] Granular material of biological origin can be described in terms of its elasticity, viscosity, plasticity and brittleness, and these attributes are interrelated. The relationships between individual attributes may be difficult to describe due to the structural complexity of plant material. The physical phenomena that occur at the contact point of two types of material have been described by various theories. The theories and methods of soil mechanics are most often used to determine frictional interactions between granular materials of biological origin based on their structural similarities.^[1] They can be divided into three groups of mechanical, molecular and mechanical/molecular theories. In mechanical theories, frictional resistance is attributed mainly to surface asperities, frictional interactions between materials, elastic strain and plastic strain at the point of contact. According to the most advanced mechanical theory, plant materials should be described based on the frictional interactions between soft and hard materials, where microprotrusions on the surface of hard objects produce microindentations and grooves on the surface of soft materials. However, this theory does not explain all aspects of friction, in particular on very smooth surfaces. To address this problem, molecular theories have been proposed, where the interactions between the particles and atoms of materials that come into frictional contact are regarded as the only source of friction. Molecular theories explain the interactions between smooth surfaces relatively well, but they are not sufficiently accurate in describing granular materials of biological origin which are characterized by various degrees of roughness. The flaws in mechanical and molecular theories spurred the search for new approaches to the problem and led to the development of mechanical/molecular theories.^[1–3,5]

According to a recent approach to friction, developed by Frączek,^[1] friction force has three components: deformation, adhesion and cohesion. This theory accounts for changes in the shape of surface asperities that tug each other, adhesion and cohesion between the surfaces that come into contact. Adhesion plays the key role in the external friction of granular plant materials. External friction is affected by interrelated properties of materials that form a friction pair, in particular the ratio of roughness densities of two surfaces, the ratio of roughness of two surfaces, the ratio of elastic moduli, and the product of real contact area and seed hardness.

According to Molenda et al.,^[5] Frączek,^[1] Horabik,^[6] Molenda and Horabik,^[7] Afzalinia and Roberge,^[2] Sharobeem,^[8] Ibrahim^[9] and Bakun-Mazor et al.,^[10] the frictional properties of seeds are influenced by frictional conditions (normal load, sliding distance, sliding velocity, seed orientation relative to the direction of movement), the parameters of the friction surface (type, roughness), seed properties (moisture content, species, variety, ripeness, variations in shape) and external conditions (temperature and humidity). The frictional properties of seeds have been extensively studied, but most researchers have identified only the type of tested friction materials (e.g. concrete, steel, wood) without describing their manufacturing characteristics. Published research findings do not account for differences in surface smoothness which determines adhesion and cohesion; thus, they should be interpreted with caution. The aim of this study was to determine the correlations between the external friction angle of wheat kernels vs. the roughness of the friction plate and the basic physical properties of kernels.

Materials and methods

Preparation of samples

The experiment was performed on the kernels of winter wheat var. Jensen grown in the Region of Warmia and Mazury (northern Poland). The samples were composed of kernels that were obtained directly from spikes and threshed kernels. Harvest and threshing were carried out on the same day, i.e. 4th August 2015. Wheat was threshed with the Claas Lexion 750 combine harvester. Wheat kernels were stored in a closed container at room temperature for 6 months to stabilize their moisture content. The relative moisture content of kernels was determined at 12.3% on a drying scale with a MAX 5-/WH halogen lamp (Radwag Radom, Poland). The survey sampling method^[11] was used to randomly select 50 kernels in each treatment. Standard error of the mean did not exceed 0.2 mm for the three basic kernel dimensions, 2.5 mg for kernel mass, and 1° for the angle of external friction.

Physical properties

At the beginning of the experiment, the angle of external friction was measured with a device equipped with photodetectors (Figure 1).^[12,13] Friction plates made of ST3S steel were fixed to an adjustable arm. The geometrical product specifications (GPS) of friction plates were measured with the Diavite DH-5 (Bülach, Switzerland) surface roughness tester. The results of the measurements are presented in Table 1. Individual kernels were placed on a horizontally inclined plate, just above the light level of the top photodetector. The angle of the adjustable arm was modified with the velocity of 1.25°·s⁻¹. When motion was initiated, the light beam was interrupted, the arm was automatically paused, and the kernel traveled down the friction plate along a distance of 25 cm. The angle of inclination was measured to the nearest 0.01°. All kernels were placed on the plate with the crease down and the hilum facing the direction of movement. Every kernel was measured in five replications. When three successive kernels had been measured, the plate was wiped with cotton wool saturated with 40/60 petroleum ether (Chempur Piekary Śląskie, Polska) to remove cutin.

Table 1. Geometric characteristics (GPS) of steel friction plates.

Plate number	Roughness parameters
	Ra (μm)	Rmax (μm)	Rt (μm)
1	0.38	2.3	2.6
2	0.93	7.1	7.1
3	1.39	11.9	12.0
4	3.25	28.1	28.1
5	5.86	53.8	73.5
6	8.16	107.5	142.0

Ra – arithmetic average of absolute values; Rmax – highest unequal height; Rt – maximum height of profile

Figure 1. Device for measuring the frictional properties of seeds^[13]: 1 – base of inclined plane, 2 – stepper motor, 3 – CPU controller, 4 – computer, 5 – bottom phototube, 6 – friction plate, 7 – top phototube, 8 – seed, 9 – adjustable arm.

In the next stage of the experiment, the length and width of wheat kernels were determined with the use of the MWM 2325 workshop microscope (PZO Warszawa, Poland) to the nearest 0.02 mm, and kernel thickness was measured with a device comprising a dial indicator (MasterTools, Kraków, Poland) with 0.01 mm resolution. The above measurements were performed according to the method described by Kaliniewicz et al.^[13,14] Kernel mass was determined on the WAA 100/C/2 weighing scale (Radwag Radom, Polska) to the nearest 0.1 mg. The measured parameters were used to determine the following shape factors for each kernel^[15,16]:

(1)

(2)

(3)

(4)

Statistical analysis

The results were analyzed statistically in the Statistica PL v. 12.5 program at a significance level of α = 0.05. The differences between the analyzed parameters of wheat kernels were determined by one-way analysis of variance (ANOVA). The normality of each group was verified by the Shapiro-Wilk W-test, and the equality of variances was assessed with Levene’s test. Where the null hypothesis of equal population means was rejected, the differences were determined, and homogenous groups were identified with the use of Duncan’s test. The correlations between the angle of external friction vs. the physical parameters of kernels and the surface roughness of the friction plate were determined by linear correlation analysis.^[17]

Results and discussion

Experimental material

The parameters of the analyzed wheat kernels are presented in Table 2. Kernels obtained directly from spikes and threshed kernels differed in dimensions and mass, and those obtained from spikes were less plump. This is a logical consequence of threshing, where some kernels, in particular finer and less developed, are lost during harvest. As expected, the shape of kernels, described by the calculated shape factors, was similar in both groups of kernels. The thickness of Jensen kernels was similar to that noted in wheat varieties Zyta,^[18] Ambral, Baroudeur and Artaban,^[19] Kunduru-1149,^[20] Ceralio,^[21] Banti,^[22] and Jonong and Keumkang.^[23] In terms of width, the evaluated kernels were similar to wheat varieties Tonacja and Vinjett,^[24] Olgeuru and Sukang, Tapdong and Younbaek,^[23] and Chamran.^[25] An analysis of average kernel length revealed similarities between Jensen kernels and wheat varieties Scipion,^[19] Korweta,^[21] Younbaek,^[23] and Chamran and Marvdasht.^[25] The above measurements indicate that the dimensions of the evaluated kernels did not differ significantly from the values given in the literature.^{[19,23,25–29]} For this reason, kernel mass and the calculated shape factors were also within the range of values reported for wheat in the literature.

Table 2. Physical properties of wheat kernels.

Origin	Physical property	Value			Standard deviation
		Minimum	Average	Maximum
Spike	T (mm)	2.52	2.97^a	3.39	0.162
	W (mm)	2.57	3.31^a	4.04	0.267
	L (mm)	5.33	6.44^a	7.45	0.442
	m (mg)	31.8	45.15^a	61.1	5.931
	Ka (%)	72.16	90.22^a	108.18	7.345
	Kb (%)	36.01	46.28^a	55.50	3.222
	Kc (%)	39.68	51.56^a	65.81	4.823
	Kd (%)	53.11	61.96^a	71.40	2.979
Combine harvester	T (mm)	2.58	3.07^b	3.51	0.169
	W (mm)	2.63	3.40^b	4.05	0.265
	L (mm)	5.33	6.64^b	7.59	0.395
	m (mg)	30.3	47.24^b	64.7	6.294
	Ka (%)	71.98	90.61^a	109.09	6.642
	Kb (%)	38.05	46.39^a	54.99	3.100
	Kc (%)	39.57	51.37^a	63.58	4.014
	Kd (%)	54.93	61.95^a	69.02	2.625

a, b – different letters denote significant differences in the average value of the same attribute between unthreshed and threshed kernels.

Angle of external friction

The average angle of external friction of kernels of var. Jensen (Table 3) ranged from 16.11° (threshed kernels on a friction plate with surface roughness Ra = 0.93 μm) to 25.09° (kernels obtained from spikes on a friction plate with surface roughness Ra = 5.86 μm). In most cases, significant differences in the values of external friction angles were not observed during successive measurements. The above results suggest that the angle of external friction is not significantly influenced by the sliding distance (0–100 cm) of kernels obtained from spikes or threshed kernels.

Table 3. Average values of the angle of external friction of wheat kernels and roughness.

Roughness parameter Ra (μm)	Angle of external friction
	α1 x ± SD	α2 x ± SD	α3 x ± SD	α4 x ± SD	α5 x ± SD
	Spike
0.38	19.78 ± 2.16^aB	19.77 ± 2.49^aB	19.79 ± 2.71^aB	21.15 ± 2.62^bB	20.22 ± 2.80^abB
0.93	17.54 ± 2.48^aA	17.41 ± 2.05^aA	17.77 ± 2.41^aA	17.30 ± 1.97^aA	17.77 ± 2.30^aA
1.39	21.67 ± 3.59^aC	22.44 ± 3.62^aC	22.43 ± 3.27^aC	22.44 ± 3.29^aC	23.20 ± 3.20^aC
3.25	21.28 ± 2.37^aC	21.74 ± 2.27^aC	21.51 ± 1.76^aC	21.85 ± 2.52^aBC	22.09 ± 3.34^aC
5.86	23.30 ± 2.75^aD	23.88 ± 2.11^aD	24.07 ± 2.56^abD	23.60 ± 2.50^aD	25.09 ± 3.39^bD
8.16	22.22 ± 3.46^aCD	22.03 ± 2.80^aC	21.97 ± 2.18^aC	21.84 ± 2.51^aBC	22.68 ± 3.50^aC
	Combine harvester
0.38	17.10 ± 1.55^aB	16.59 ± 1.33^aA	16.72 ± 1.72^aA	17.15 ± 1.59^aA	17.22 ± 1.94^aA
0.93	16.11 ± 1.94^aA	16.24 ± 1.78^aA	16.14 ± 2.08^aA	16.35 ± 1.87^aA	16.62 ± 2.26^aA
1.39	19.73 ± 2.25^aC	19.86 ± 2.98^aB	20.47 ± 3.01^aB	20.32 ± 2.74^aB	20.45 ± 3.27^aB
3.25	19.66 ± 1.74^aC	20.24 ± 2.37^aB	20.27 ± 2.85^aB	19.90 ± 1.92^aB	19.95 ± 2.11^aB
5.86	22.50 ± 1.75^aD	23.28 ± 2.43^abD	22.83 ± 2.49^abC	22.93 ± 2.40^abC	23.61 ± 2.70^bC
8.16	22.05 ± 2.37^aD	21.83 ± 2.04^aC	23.00 ± 3.22^aC	22.24 ± 2.68^aC	22.95 ± 3.08^aC

a, b – different letters denote significant differences in the average values of the angle of external friction of kernels from successive replications on the same friction plate; A, B, C, D – different letters denote significant differences in the average values of the angle of external friction of kernels on the tested friction plates in the corresponding replicates.

The average values of the angle of external friction determined for every kernel in five replications were used in further analyses. The measured values of angles are presented in Table 4. In most cases (excluding the friction plate with surface roughness Ra = 8.16), significantly lower values of external friction angles were noted for threshed kernels than for unthreshed kernels. The observed differences can be attributed to the impact of the combine harvester, in particular its threshing, separating and conveying systems, on the harvested kernels. When the elements of the combine harvester come into contact with kernels, the asperities on kernel surface are leveled and surface grooves are produced. High process dynamics and repeated surface microdeformation ultimately reinforce the outer layer of kernels,^[1] which decreases their angle of external friction.

Table 4. Average values of the angle of external friction of wheat kernels and roughness.

Roughness parameter Ra (μm)	Angle of external friction of wheat kernels from
	Spikes x ± SD	Combine harvester x ± SD
0.38	20.14 ± 1.97^bB	16.96 ± 0.92^aB
0.93	17.56 ± 1.28^bA	16.29 ± 1.28^aA
1.39	22.44 ± 2.51^bC	20.17 ± 1.96^aC
3.25	21.69 ± 1.40^bC	20.00 ± 1.29^aC
5.86	23.99 ± 1.57^bD	23.03 ± 1.63^aD
8.16	22.15 ± 1.81^aC	22.41 ± 1.74^aD

a, b – different letters denote significant differences in the average values of the angle of external friction of unthreshed and threshed kernels tested on the same friction plate; A, B, C, D – different letters denote significant differences in the average values of the angle of external friction of kernels of the same origin on the tested friction plates.

The surface roughness of the friction plate significantly influenced the external friction angle of the evaluated kernels. For both unthreshed and threshed kernels, the values of the external friction angle were lowest on a friction plate with surface roughness Ra = 0.93 μm, and highest on a friction plate with surface roughness Ra = 5.86 μm. In both types of kernels, kernels with external friction angles on friction plates with surface roughness Ra = 1.39 μm and Ra = 3.25 μm formed a homogeneous group. The average values of the angle of external friction ranged from 16.29° to 23.99° (friction coefficients of 0.29 to 0.45), which is consistent with the results reported by Molenda et al.,^[5] Kram,^[30] Karimi et al.,^[31] Boac et al.,^[28] Kaliniewicz,^[32] Markowski et al.^[21] and Kaliniewicz et al.^[13] Similar values of external friction coefficients were determined in seeds of other plant species, including chick peas,^[33] lentils,^[34] fenugreek,^[35] coriander,^[36] cowpeas,^[37,38] barley,^[39] castor beans^[40] and psyllium.^[41]

Correlations between the physical properties of kernels

The data presented in Table 5 suggest that the evaluated physical properties of wheat kernels of var. Jensen (basic dimensions, mass and shape factors) are very weakly correlated with their angles of external friction. The absolute values of the correlation coefficient ranged from 0.005 to 0.475, and the observed correlations were practically significant (r > 0.4) in only 5 out of 96 cases. The highest number of significant correlations between the external friction angle and other parameters was noted in kernels obtained directly from spikes, which were tested on a friction plate with surface roughness Ra = 1.39 (5 out of 8 cases). The above could be attributed to synergistic interactions between asperities on the surface of kernels and the friction plate, which increased the cohesion component. According to Kaczorowski and Ślipek,^[42] Frączek,^[1] and Horabik,^[6] cohesion is significantly influenced by the real area of contact between a seed and the friction plate, which partially explains the significant influence of kernel thickness, width, mass and 2 out of 4 shape factors on the angle of external friction.

Table 5. Coefficients of linear correlation between the angle of external friction and the remaining properties of wheat kernels.

Origin of kernels	Physical property	Angle of external friction on friction plate with surface roughness Ra (μm)
		0.38	0.93	1.39	3.25	5.86	8.16
Spike	T	0.043	−0.187	−0.475**	0.341*	−0.005	−0.356*
	W	0.203	−0.054	−0.470**	0.175	−0.014	−0.156
	L	0.197	−0.087	−0.108	0.286	−0.011	0.173
	m	0.185	−0.144	−0.459**	0.308*	−0.097	−0.228
	Ka	−0.156	−0.123	0.068	0.104	0.013	−0.140
	Kb	−0.154	−0.107	−0.247	−0.019	0.011	−0.447**
	Kc	0.024	0.026	−0.303*	−0.074	0.005	−0.226
	Kd	−0.049	−0.034	−0.295*	−0.060	0.007	−0.360*
Combine harvester	T	−0.192	−0.193	−0.419**	−0.298	−0.246	−0.023
	W	−0.301*	−0.050	−0.333*	0.037	−0.273	0.192
	L	−0.132	−0.240	−0.140	−0.133	−0.071	0.219
	m	−0.254	−0.057	−0.236	0.031	−0.263	0.207
	Ka	0.185	−0.093	−0.085	−0.344*	0.045	−0.248
	Kb	−0.066	0.059	−0.229	−0.163	−0.191	−0.183
	Kc	−0.191	0.126	−0.130	0.144	−0.229	0.066
	Kd	−0.151	0.102	−0.201	0.006	−0.238	−0.049

No. of observations = 50; * significant at p = 0.05; ** significant at p = 0.01.

Weak correlations between the angle of external friction and the physical properties of wheat kernels were also reported by Kaliniewicz et al.^[13] in a study of var. Nawra. The above findings indicate that frictional properties should be regarded as secondary discriminatory traits. This observation is confirmed in practice where seeds are most often sorted with the use of sieves and pneumatic separators.

As already mentioned, the value of the external friction angle is determined by the surface characteristics of the tested friction material. According to Frączek,^[1] an increase in the surface roughness of the friction plate increases the deformation component of the friction force, which increases the extent to which the asperity peaks of hard structural material tug the surface of plant material. Therefore, the value of the external friction angle should increase with a rise in surface roughness. However, the above trend was not observed in this study where considerable variations were noted in the external friction angles of wheat kernels (Figure 2). These differences were found in both groups of kernels, but the values noted in unthreshed kernels were somewhat more scattered. This could be attributed to the fact that the geometrical characteristics of kernels obtained directly from spikes were not modified by threshing, which increased the tug between surface asperities. The observed changes in the values of the angle of external friction progressed similarly in both groups of kernels. Initially, the angle of external friction decreased with a rise in the surface roughness of the friction plate, and reached its minimum value when surface roughness approximated Ra = 0.93 μm. This was followed by a rapid increase and stabilization of the angle of external friction. The angle of external friction also increased on a friction plate with surface roughness of around Ra = 5.86 μm. The changes in angle values observed on friction plates with different surface roughness can be explained by the positive or negative effects of asperities on the surface of materials that come into frictional contact,^[1,6,42] which changes the real area of contact between a kernel and a friction plate. As a result, the components of the friction force (deformation, adhesion and cohesion) change their values, and their mutual interactions determine the ultimate values of external friction angle.

Figure 2. The effect of the surface roughness of the friction plate on changes in the angle of external friction of: A) kernels obtained from spikes, b) threshed kernels.

Conclusion

The results of this study indicate that the geometrical characteristics of a friction plate significantly influence the external friction angle of wheat kernels of var. Jensen. The average values of that angle range from around 16° to around 24°. The values of the external friction angle are lowest on a friction plate with surface roughness Ra = 0.93 μm, and the highest on a friction plate with surface roughness Ra = 5.86 μm. The progression of the observed changes is probably determined by specific asperities on the surface of materials that come into frictional contact, which also leads to changes in the mutual interactions between friction force components. In most cases (excluding the friction plate with surface roughness Ra = 8.16), higher values of external friction angle are noted for kernels obtained directly from spikes than for threshed kernels. The observed differences can be attributed to the reinforcement of the outer layer of kernels due to repeated surface microdeformation resulting from the impact of the combine harvester’s structural elements on kernels. The basic dimensions (thickness, width and length), mass and shape factors of wheat kernels of var. Jensen do not influence their angle of external friction on a steel friction plate. This can be attributed to the surface properties of kernels which are characterized by greater variations than their geometric properties and mass. Our findings indicate that the frictional properties of wheat kernels should be regarded as secondary discriminatory traits, and that wheat kernels should be cleaned and sorted based primarily on differences in their aerodynamic and geometric properties.

Symbols

K_a, K_b, K_c, K_d – shape factors, %,

m – kernel mass, mg

T, W, L – thickness, width and length of a kernel, mm,

x, SD – mean value and standard deviation of a trait,

α – external friction angle of a kernel on a steel surface, °,

Indices

1, 2, 3, 4, 5 – successive replications of a measurement