Parboiling to improve milling quality of Selam rice variety, grown in Ethiopia

Abstract

Conventional rice milling produces a high percentage (40%) of rice breakage. The parboiling process is, therefore, considered the best solution for reducing the susceptibility of the rice grain to breakage in addition to varietal characteristics and quality of milling machine. The parboiling process hardens the rice kernel enabling it to withstand the machine’s impact and reduce breakage during the milling process. The objective of this study was determining of the milling quality of a short-grain Ethiopian rice variety (Selam). A full factorial design was used to study the effects of the parboiling factors in CRD arrangement. The factors were soaking temperatures (40, 60, and 80°C), soaking times (6, 12, and 24 hours), and steaming times (15, 25, and 35 minutes) compared with the non-parboiled rice as a control using controlled randomized design. The result showed that combinations of treatments had significant (P ≤ 0.05) differences in the quality parameters, such as percentage of broken rice, head rice yield, and total milling recovery during milling, compared to the control. The minimum and maximum percentages of breakage for parboiled rice were 0.95 and 9.30%, which were recorded at soaking temperatures, soaking time, and steaming time of 80°C, 24 hours and 35 minutes, and 40°C, 6 hours, and 15 minutes, respectively, in contrast to the non-parboiled rice (41.73% breakage). The result showed that the highest milling recovery (90.40%) was obtained at 40°C soaking temperature, 6 hours of soaking time, and 15 minutes of steaming time compared to the control (79.30%).

Keywords:

• broken rice
• head rice yield
• rice parboiling
• Selam variety rice
• soaking
• steaming

Public interest statement

Because of the change in eating patterns and its integration with traditional food items, rice has emerged as one of Ethiopia’s primary crops with the greatest economic importance. It is one among Ethiopia’s most promising crops for ensuring food security. The Ethiopian rice industry has had several difficulties, including intense competition from imported rice caused mostly by the subpar operation of local processing equipment. Recently, parboiling technology was brought to Ethiopia as a solution for this issue, hardening the kernel and gelatinization the starch. Through soaking, steaming, and drying, the parboiling process has been identified as the best solution for enhancing the quality of rice during milling process.

1. Introduction

Rice (Oryza sativa) is an edible cereal grain of the grass plant. It is considered as a staple food crop for about half of the world’s population (Belayneh & Tekle, 2017; Vijay & Rice, 2013). Rice is the world’s most important food and has continued to be demanded worldwide (Zhou et al., 2017). Rice is a common dietary food and one of the world’s most important starchy cereal crops. Rice consumption is associated with diabetes mellitus due to its high glycemic index content (Rohman et al., 2014). The importance of rice is essentially expressed in terms of food security, use of scarce resources, poverty alleviation, youth employment, and impact on the climate (IRRI, 2016). Rice is a common dietary food and one of the world’s most important starchy cereal crops. Rice consumption is associated with diabetes mellitus due to its high glycemic index content (Rohman et al., 2014).

Rice has become an economically important crop among the major crops in Ethiopia due to the shift in eating habits and integration with traditional food products. Thus, rice is expanding very fast in all regions of Ethiopia (Addis, 2018). Since rice growing is a recently introduced phenomenon, it needs further research and development endeavours to reduce poverty, ensure food security, and reduce the limited foreign currency to import rice to satisfy the growing local demand (Roy-Macauley, 2019). Now adays, rice has gained attention by the government of Ethiopia from this perspective. Thus, rice production and utilization are expanding very fast in all regions of Ethiopia. Since rice growing is a recently introduced phenomenon, it needs further research and development to contribute to the reduction of poverty, ensure food security, and reduce the limited foreign currency to import rice to satisfy the growing local demand (Roy-Macauley, 2019). The Ethiopian rice sector has faced many challenges, such as high competition with imported rice mainly due to the poor performance of local processing machines, lack of skilled workforce to operate those machines, insufficient mechanization in poor postharvest rice processing technologies, and poor marketing channel (Roy-Macauley, 2019). So, rice quality improvement can be achieved through better postharvest handling and building the capacity of smallholder farmers and the private processors (Belayneh & Tekle, 2017).

Although the rapid increment of rice by area coverage and demand, most of Ethiopia’s locally cultivated rice varieties have low physical and cooking qualities due to inefficient processing machines. Rice passes through multiple steps of processing to be eaten. During this processing steps, both quantitative and qualitative loss of rice could occur due to milling at low moisture content, mal-operation of the polishing machine, and impurities with rice grain (Roy-Macauley, 2019).

Parboiling has been identified as an important process for improving rice cooking and milling qualities through soaking, steaming, and drying (Meresa et al., 2020). Odoom (2021) stated that there is a very high loss of rice during de-hulling and polishing when using poor-performance machines and unskilled operators. Thus, to solve this problem, parboiling technology has been introduced to Ethiopia recently to harden the kernel and gelatinizes the starch (Chakraborty et al., 2020). It was also described that parboiling technology is the hydrothermal processing of paddy rice or brown rice to reduce breakage levels and improve the head rice yield (Nawaz et al., 2018). Gelatinization enhances the removal of the rice husk easily at a low frictional force with minimum breakage (Gunathilake, 2018).

Fogera National Rice Research and Training Center has released 39 rice varieties. Among the varieties, Selam variety is reported to be adaptable in low-land agroecology, which has been released in 2020. The yield of this variety is estimated to be 5.2 tons per hectare in the research field and 4.8 in the farmers’ field. It has the highest potential yield rice variety in Ethiopia. Farmers preferred this variety because of its high yielding, grain size and white color. Most of rice processors in Ethiopia are engaged in rice trade. However, studies on the parboiling process and the relevant parameters are limited to Ethiopian rice varieties. Rice processors (private business owners with de-hulling/polishing machine plants and those engaged in the rice-processing sector) who have started the parboiling process are unclear about how to do it. There are traditional parboiling practices in Fogera and Libokemkem districts, South Gonder. The processors are using firewood and electricity as a source of energy for parboiling. It is time-consuming and inefficient. The parboiled rice product also has poor quality in this way. The locally processed rice is incomparable with imported parboiled rice due to its low quality. In addition to this, the nutritional, sensorial, and cooking qualities needed to be investigated for the rice varieties. So in-order to increase its quality, the rice has to be parboiled. And this study is aimed to achieve this objective.

Rice processors frequently asked the Rice Research Center for technologies available to solve the high volume of rice breakage problems, including soaking times, soaking temperature, steaming time, and drying conditions. This study was therefore, aimed to answer the processors’ questions and generate relevant information to improve the rice quality and reduce the percentage of broken rice. The study also aimed to reduce the quality and quantity loss of domestic rice and increase domestic rice production to substitute the imported one with equivalent quality at a lower price. The study also expected to enhance farmers’ livelihood and increase the demand of rice in the market. In this experiment, the quality of parboiling treatments under different times and temperatures were compared with conventionally processed rice. Thus, the main objective of this study was to improve the milling quality of processed rice in Ethiopia; and the specific objectives were to (a) study the effect of rice parboiling conditions concerning total milling recovery; (b) determine the effect of parboiling on the reduction of broken rice; and (c) study the effect of the parboiling process on the increment of head rice yield.

2. Materials and Methods

2.1. Experimental Materials

A lowland rice (Selam) variety of 120 kg was collected from the National Rice Research and Training Center of Fogera, Ethiopia (FNRRTC), which was harvested and threshed in December 2020.

2.2. Experimental Design

The experiment was conducted with a full factorial design of 3 × 3x3 in CRD order. It consisted of three variables with three levels. Paddy rice soaking temperature for pre-treatment (40, 60, and 80°C), soaking time (6, 12, and 24 hours), and steaming time (15, 25, and 35 minutes) at a controlled temperature of 100°C (Eshun et al., 2019). Each treatment was carried out with three replications. The non-parboiled milled rice was considered as a control that passed through the same procedure except for the parboiling process. A control sample of 20 kg paddy rice was sun-dried (without parboiling) with up to 14% moisture content and was milled for control (Hasbullah et al., 2017; Roy-Macauley, 2019; Sayanthan & Thusyanthy, 2018; Verma & Srivastav, 2017).

2.3. Parboiling Process

2.3.1. Soaking

For the samples of 27 kg preheated at 40°C, one-third of the weight of the sample was soaked for 6 hours, an other one-third of the 27 kg for 12 hours, and the third batch for the 24 hours and left in ambient conditions until the time was up. The same procedures were followed for 60 and 80°C preheating temperatures. Each sample division was added into a soaking tank (made up of stainless steel, Inox 304), with enough of water at the water-to-paddy ratio of 2:1 (Hasbullah et al., 2017; Ndindeng et al., 2015). The soaking temperatures and soaking time values used for this study were the medium values that the authors used (Ayamdoo et al., 2013; Bello et al., 2015; Hasbullah et al., 2017; Michael et al., 2018; Ndindeng et al., 2015).

2.3.2. Steaming

According to authors Ndindeng (Ndindeng et al., 2015), each of the samples was soaked for 6, 12, and 24 hours at a temperature of 40, 60, and 80°C one by one was divided into three equal parts having 3 kg each for further steaming process of 15, 25 and 35 minutes. Each of the 3 kg samples was added to a perforated steaming apparatus made of stainless steel (Inox 304). The amount of water that was added to the steaming apparatus was a 1:2 ratio with the paddy. The paddy did not contact the water directly. It was steam heated by the vapor released from the steaming tank. The soaked samples from each soaking time were steamed for 15, 25, and 35 minutes of duration without removing the heat source. The steaming time values selected for this study were the medium values used by most authors (Hasbullah et al., 2017; Muttagi & Ravindra, 2020; Ndindeng et al., 2015; Yerragopu & Palanimuthu, 2019).

2.3.3. Drying

The steamed paddy was dried carefully using sunlight by spreading it on a plastic sheet. It was continuously rolled over to avoid over-drying on one side and to protect against the formation of internal cracks. The paddy was dried up to 18% moisture content and then placed in the shade till to be to 14 % moisture content (Hasan et al., 2019; Ndindeng et al., 2015). At every 15-minute interval, the moisture content level was checked using a rice grain moisture meter (KETT J301, Japan) (Meresa et al., 2020).

2.3.4. Husking

The soaked, steamed and dried paddy was husked with a husking machine (Satake SB 2000, Satake Corporation, Hiroshima, Japan). The paddy was subjected to a properly adjusted husking machine. Then, according to Nambi (Nambi et al., 2017) and Meresa (Meresa et al., 2020), the husk was removed applying the husking machine, and brown rice was obtained (Figure 1).

Figure 1. Working principle of a rubber roll husker (source (Dhankhar & Hissar, 2014).

2.3.5. Polishing

The husked rice (brown rice) was transferred into the polishing machine (Satake SB 2000, Satake Corporation, Hiroshima, Japan). The properly adjusted rubber roll polishing machine ran to separate the bran from the brown rice. The properly adjusted rubber roll machine producing polished white rice (Meresa et al., 2020).

2.4. Determination of Milling Qualities

2.4.1. Percentage of husking

The term husking referred to the removal of the husk of paddy rice. The percentage of husked/hulled parboiled rice was analyzed by dividing the total weight of husked rice by the total sample of paddy weight. The 3 kg of rice that was steamed by the three levels of pre-determined time was weighed before and after husking, and the percentage of husking was estimated (Muttagi & Ravindra, 2020). Two-hundred fifty grams of husked rice was taken, and the unhulled paddy was manually separated and weighed. Then, using Equation 1, the percentage of husking was analyzed.

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

2.4.2. Milling recovery

The percentage of rice milling recovery was estimated by taking 250 grams of milled rice and dividing with the weight of the sample (Equation 2) (Chavan et al., 2018; Michael et al., 2018).

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

2.4.3. Percentage of broken rice

It was estimated by taking 250 g of milled rice, separating and weighing the broken kernels, and then dividing by the total weight of milled rice(250 g) using Equation 3 (Chavan et al., 2018; Michael et al., 2018).

(3) $\mathrm{P}\mathrm{e}\mathrm{r}\mathrm{c}\mathrm{e}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{g}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{b}\mathrm{r}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{n}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\left(\mathrm{\%}\right)=\left[\frac{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{b}\mathrm{r}\mathrm{o}\mathrm{k}\mathrm{e}\mathrm{n}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{m}\mathrm{i}\mathrm{l}\mathrm{l}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}}\right]\hspace{0.17em}\times 100$(3)

2.4.4. Head rice yield

According to the Ethiopian Standards Authority, the head rice yield (HRY) has been estimated by taking 250 g of paddy. It was calculated by dividing the non broken polished rice (whole rice) by the weight of paddy rice, expressed in percent as described by Equation 4 (Nawaz et al., 2018; Yerragopu & Palanimuthu, 2019).

(4) $\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{d}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)=\left[\frac{\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{w}\mathrm{h}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{s}}{\mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{p}\mathrm{a}\mathrm{d}\mathrm{d}\mathrm{y}}\right]\hspace{0.17em}\times 10$(4)

2.5. Data Analysis

The collected data were statistically analyzed using Statistics-10 software. It was subjected to factorial analysis of variance (ANOVA) to see the mean differences among the different parboiled treatments related to the control (non-parboiled) and regarding each parameter. The percentage of broken rice, head rice yield, and milling recovery parameters were analyzed. The statistically significant differences of the samples means were interactively analyzed fusing a full factorial design arranged in CRD at a 5% significance level.

3. Results and Discussion

3.1. Main Factors Effects on Milling Quality

3.1.1. Effects of soaking temperature on parboiled rice quality parameters

Table 1 presents the effects of soaking temperatures on the milling qualities of parboiled rice. The parboiling process significantly improved the rice milling quality (P ≤ 0.05). The percentage of husking was highest (85.93%) for the samples subjected to a soaking temperature of 60°C, and the least value (85.37%) was found at 40°C. The soaking temperature significantly (P ≤ 0.05) affected the percentage of broken rice. The broken rice percentage decreased from 41.73 to 5.98–2.81%, with the increase in temperature from 40–60°C. The highest percentage of broken rice (5.98%) was recorded at 40°C soaking temperature. Increasing the soaking temperature from 40 to 80°C gelatinized the starch; the rice kernel hardened and withstood the friction force applied by the rotating rolls. This phenomenon caused the reduction of broken kernels and increased the head rice yield. The percentage of broken rice decreased from 22.8% (non-parboiled) to 2.81% (parboiled). A similar study also supported that the conventional rice milling process increases the broken percentage due to low surface hardness, which leads to low milling recovery and quality of milled rice (Paul et al., 2020). Chavan (Chavan et al., 2018) reported that soaking operation at 85°C was obtained only 6.31% of broken rice compared to the control. Similarly, Ogunbiyi (Ogunbiyi et al., 2018) studied that at a soaking temperature of 67.7°C, the broken rice was found to be 2.18%.

Table 1. Effects of soaking temperatures on milling qualities of parboiled rice

SkTp (^oC)	Husking (%)	Broken rice(%)	Milling Recovery (%)	Head Rice Yield (%)
40	85.37^b ± 1.52	5.98^b ± 1.76	87.55^a ± 1.52	81.57^c ± 0.95
60	85.93^a ± 1.94	2.81^c ± 1.15	84.87^c ± 1.96	82.11^b ± 1.66
80	85.44^b ± 2.48	2.86^c ± 1.49	85.97^b ± 1.94	83.21^a ± 2.62
Control	75.00^c ± 3.61	41.73^a ± 1.94	79.30^d ± 2.95	58.02^d ± 2.94
CV (%)	0.47	8.01	0.79	0.8
LSD	0.22	0.17	0.37	0.37

Note: SkTp: Soaking Temperature, means with the same letters are not significantly different, values are in mean ± SD.

The maximum milling recovery (87.55%) was observed for the samples subjected to the lowest temperature (40°C), and the minimum (84.87%) value was obtained at 60°C. The head rice yield improved with the increase in soaking temperature from 40 to 80°C. The highest head rice yield (83.21%) was recorded for samples soaked at a temperature of 80°C and the lowest was the control (58.02%). A similar study by Karzan et al. (Karzan et al., 2016) showed that at the soaking temperature of 69.88°C, the head rice yield production was 73.43%. Another study also stated that the head rice yield had significantly (P ≤ 0.05) increased from 42.87% of the non-parboiled to 60% when parboiled at 80°C (Kale et al., 2015). Soaking Temperature increased the head rice yield, milling recovery, and husking yield from 58 to 83%, 79 to 87.55%, and 75 to 85.9%, respectively. Soaking temperature highly reduced the percentage of broken rice from 41.7 to 2.81%.

3.1.2. Effects of soaking time on parboiled rice quality parameters

Table 2 presents the effect of soaking time on parboiled rice quality attributes. The percentage of husking significantly (P ≤ 0.05) increased from 75.00% of the non parboiled to 86.00% when soaked for 24 hours. The lowest husking percentage (85.00%) was recorded for samples treated at the lowest soaking time (6 hours). The percentage of broken rice decreased with the increase in soaking time from 6–24 hours. The broken content decreased from 4.79 to 2.97% as the soaking time increased from 6 to 24 hours. The lower percentage of broken rice resulted in longer times of soaking. That was due to the longer the soaking time, the higher the degree of gelatinization of the starch in the kernel formed, and the harder the grain become when cooled, hence leads to less breakage.

Table 2. Effects of soaking time on milling qualities of parboiled rice

SkTm (hours)	Husking (%)	Broken (%)	MR (%)	HRY (%)
6	85.00^c ± 2.24	4.79^b ± 2.32	86.24^b ± 2.28	81.59^b ± 1.65
12	85.74^b ± 1.97	3.88^c ± 1.98	86.65^a ± 1.87	82.70^a ± 2.03
24	86.00^a ± 1.73	2.97^d ± 1.56	85.52^c ± 2.07	82.60^a ± 2.08
Control	75.00^d ± 3.61	41.73^a ± 1.94	79.30^d ± 2.95	58.02^c ± 2.94
CV (%)	0.47	8.01	0.79	0.8
LSD0.05	0.22	0.17	0.37	0.36

Note: SkTm: soaking time, MR: milling recovery, HRY: head rice yield, values are in mean±SD.

The highest percentage of milling recovery (86.65%) was recorded at a steaming time of 12 hours. The trend showed that increasing only the soaking time while other factors are constant, was not highly affected the milling recovery. Its highest value was 82.70% observed for the samples subjected to 12 hours of soaking time. Soaking for 12 and 24 hours had no significant (P > 0.05) differences in the percentage of head rice yield. A similar study explored that soaking at 85°C for three hours was found to be the optimum parboiling method that obtained 66.09% head rice yield (Nambi et al., 2017). Soaking time increased the head rice yield, milling recovery, and percentage of husking from 58 to 82.7%, 79 to 86.65%, and 75 to 86%, respectively.

3.1.3. Effects of steaming time on parboiled rice quality parameters

Table 3 presents the data showing steaming time’s effects on rice’s parboiled quality parameters. The percentage of husking increased with an increase in steaming time. The highest and the lowest values of husking percentage for parboiled rice were 87.18% and 84.07%, recorded at 35 and 15 minutes of steaming, respectively. The percentage of broken rice significantly (P ≤ 0.05) decreased from 22.80–2.82% when the steaming time was increased. The minimum broken rice values of 2.82% and 3.61% were recorded at the highest steaming times of 35 and 25 minutes respectively, and the highest value (22.80%) was obtained for non-parboiled rice [Table 3]. The trend showed that increasing the steaming time from 15 to 35 minutes reduced the percentage of broken rice from 5.22 to 2.82%. A similar study by Chavan (Chavan et al., 2018) found that the broken rice was found to be 6.31% when the paddy was steamed for 40 minutes.

Table 3. Effects of steaming time on milling qualities on parboiled rice

SmTm (min)	Husking (%)	Broken (%)	MR (%)	HRY (%)
15	84.07^c ± 1.24	5.22^b ± 2.38	86.31^b ± 2.25	81.15^c ± 1.08
25	85.48^b ± 1.93	3.61^c ± 1.53	85.37^c ± 1.89	81.89^b ± 1.28
35	87.18^a ± 1.47	2.82^d ± 1.54	86.72^a ± 2.01	83.85^a ± 2.26
Control	75.00^d ± 3.61	41.73^a ± 1.94	79.30^d ± 2.95	58.02^d ± 2.94
CV (%)	0.47	8.01	0.79	0.8
LSD0.05	0.22	0.17	0.37	0.36

Note: SmTm: steaming time, MR: milling recovery, HRY: head rice yield, values are in mean±SD.

The milling recovery ranged from 85.37 to 86.72%, the highest milling recovery (86.72%) was recorded at 35 minutes of steaming time and the lowest was 85.37%. The milling recovery increased with increasing steaming time [Table 3]. The maximum head rice yield was 83.85%, recorded for the samples subjected to 35 minutes of steaming. The minimum value (81.15%) was recorded at 15 minutes of steaming, whereas the head rice yield for non-parboiled rice remained was 66.02%. A similar study showed that the percentage of head rice recovery was significantly increased (P ≤ 0.05) to 66.09% from the control value of 47.53% at 40 minutes of steaming (Nambi et al., 2017). The steaming time increased the head rice yield from 58 to 83.85%, milling recovery from 79 to 86.7% and it reduced the percentage of broken rice from 41.7% to 2.82–5.22%.

3.2. Two-way Interaction Effects of Soaking Temperature, Soaking, and Steaming Times on Milling Quality

3.2.1. Interaction effects of soaking temperature and soaking time on milling quality

Table 4 presents the interaction effect of soaking temperature and soaking time on parboiled rice milling quality parameters. The highest husking percentage value (86.44%) was recorded at 40°C interacted with 24 hours of soaking time. The husking percentage showed an increasing trend at 40°C for the three soaking times. However, at 60 and 80°C of soaking temperatures, no trend was observed within the respective group.

Table 4. Effects of soaking temperature and soaking time on milling quality parameters

SkTp	SkTm	Husking (%)	MR (%)	Broken (%)	HRY (%)
40	6	84.22^e ± 1.39	88.44^a ± 1.72	6.97^b ± 1.84	81.47^cde ± 0.90
	12	85.44^c ± 1.33	87.56^b ± 1.24	6.24^c ± 1.56	81.32^de ± 0.81
	24	86.44^a ± 1.01	86.65^c ± 1.09	4.72^d ± 1.14	81.94^cd ± 0.37
60	6	86.00^b ± 2.45	84.40^e ± 1.56	3.13^f ± 1.62	81.29^e ± 1.21
	12	85.78^bc ± 2.167	86.18^cd ± 1.81	3.19^f ± 0.49	82.98^b ± 2.08
	24	86.00^b ± 1.22	84.04^e ± 1.98	2.11^g ± 0.77	82.05^c ± 1.23
80	6	84.78^d ± 2.54	85.87^d ± 1.50	4.29^e ± 1.69	82.01^c ± 2.51
	12	86.00^b ± 2.45	86.21^cd ± 2.26	2.22^g ± 0.36	83.81^a ± 2.19
	24	85.56^c ± 2.60	85.86^d ± 2.18	2.08^g ± 0.93	83.82^a ± 2.97
Control		75.00^f ± 3.61	79.30^f ± 2.95	41.73^a ± 1.94	58.02^f ± 2.94
CV (%)		0.47	0.79	8.01	0.8
LSD0.05		0.38	0.64	0.29	0.62

Note: SkTp: soaking temperature (^oC), SkTm: soaking time (hr), MR: milling recovery, HRY: head rice yield, values are in mean±SD, means represented by the same letter are not significant.

Regarding milling recovery, the highest three values (88.44, 87.56, and 86.65%) with significant differences (P ≤ 0.05) among them were treatment combinations of the 6, 12, and 24 hours, respectively, of at soaking times and 40°C soaking temperature. The rest of the data did not show any trend and reflected a balanced influence of the interaction of the two factors [Table 4]. The lowest milling recovery values were 84.04 and 84.40% which were soaked at 60°C for a duration of 6 and 24 hours, respectively. Generally, the 40°C soaking temperature resulted in relatively higher milling recovery, whereas the 60 and 80°C showed a reduced recovery rate [Table 4]. The finding of this experiment agreed with the study of Nasirahmadi (Nasirahmadi et al., 2014), who reported an improvement in milling recovery by up to 75.8% by a parboiling process.

The highest value of broken kernels had 6.97, 6.24, and 4.72% overserved for samples of soaked for 40°C soaking temperatures, and of 6, 12 and 24 hours,soaking times, respectively. Broken kernels were reduced with an increase in soaking temperature; for each soaking temperature, it decreased with an increase in soaking time. While, the lowest values of 2.08, 2.22, and 2.11% were found for the combinations of high temperatures with longer time. This could be attributed to the higher gelatinization of the starch at higher temperatures and longer thermal exposure. Gelatinized starch becomes harder as it cools and prevents the kernels from easily breaking.

Head rice yield was higher (83.82 and 83.81%) for samples subjected to 80°C soaking temperatures for the duration of 12 and 24 hours. On the other hand, the lowest values were of those samples treated at the lowest soaking temperatures. It can be seen that soaking temperature dominates within the interaction regarding the head rice yield of the parboiling operation. The head rice yield ranged from 81.29 to 83.82%. A similar study showed that the head rice yield of the parboiled rice varies from 68–74% (Rehal et al., 2017). Selam variety performed better when parboiled than the varieties mentioned in the literature.

3.2.2. Interaction effects of soaking temperature and steaming time on milling quality

Table 5 presents the data showing the interaction effects of soaking temperatures and steaming times on milling quality. The percentage of husking was increased from 83.56–88.22% at the highest soaking temperature (80°C) and steaming times of 15 and 35 minutes. The degree of husking showed an increasing trend as the temperature and steaming time significantly (P ≤ 0.05) increased.

Table 5. Effect of soaking temperature and steaming time on milling quality parameters

SkTp	SmTm	Husking (%)	MR (%)	Broken (%)	HRY (%)
	15	84.89^d ± 1.27	88.93^a ± 1.37	7.87^a ± 1.44	81.06^d ± 0.75
40	25	84.89^d ± 1.45	86.52^d ± 1.14	5.28^b ± 1.14	81.24^d ± 0.57
	35	86.33^c ± 0.50	87.20^c ± 0.91	4.77^c ± 0.72	82.43^bc ± 0.90
	15	83.78^e ± 1.20	85.33^e ± 1.22	3.93^d ± 1.01	81.39^d ± 1.46
60	25	87.00^b ± 1.22	84.47^g ± 2.24	2.52^f ± 0.82	82.15^c ± 1.31
	35	87.00^b ± 1.22	84.84^efg ± 2.36	1.97^g ± 0.56	82.79^b ± 1.99
	15	83.56^e ± 0.88	84.67^fg ± 1.14	3.85^d ± 1.86	80.99^d ± 0.99
80	25	84.56^d ± 2.13	85.13^ef ± 1.66	3.03^e ± 0.91	82.30^bc ± 1.58
	35	88.22^a ± 1.09	88.13^b ± 0.44	1.72^g ± 0.63	86.34^a ± 1.15
Control		75.00^f ± 3.61	79.30^f ± 2.95	41.73^a ± 1.94	58.02^f ± 2.94
CV (%)		0.47	0.79	8.01	0.8
LSD0.05		0.38	0.64	0.29	0.62

Note: SkTp: soaking temperature (^oC), SmTm: steaming time (minutes), Milling R: milling ratio (mm/mm), HRY: head rice yield, values are in mean±SD.

The milling recovery of parboiled rice ranged between 84.47% and 88.93%. The lowest and the highest values were recorded at the interaction of 60°C with 25 minutes and 40°C with 15 minutes of soaking temperature and steaming time, respectively [Table 5]. The highest milling recovery percentage was recorded at the lowest soaking temperature and steaming time. No clear trend was noted in the data due to the interaction of the two factors. A similar study conducted by Nasirahmadi (Nasirahmadi et al., 2014) showed that the maximum values of milling recovery (75.8%) were obtained from the combination of 75°C soaking temperature and 20 minutes of steaming time.

The percentage of broken rice was significantly (P ≤ 0.05) affected by the interaction of soaking temperature and steaming time. The rate of broken rice decreased from 7.87 to 1.72% as the temperature and time of steaming rose to 80°C and 35 minutes. The maximum breakage (7.87%) was recorded for those subjected to a treatment combination of the lowest temperature (40°C) and the lowest steaming time (15 minutes). In general, the data of broken rice showed a decreasing trend with an increase in steaming time for all the temperatures and between the steaming temperatures in the combinations.

According to the study by Meresa (Meresa et al., 2020), soaking at 40–80°C for steaming time of 10–50 minutes, the optimum percentage of broken rice was identified to be 6.4% and 2.07% for Gumara, and Ediget rice varieties, respectively, while Chavan (Chavan et al., 2018) reported that the lowest broken percentage was 6.31% at 85°C soaking temperature and 40 minutes of steaming. Likewise, at a soaking temperature of 80°C, the lowest percentage of broken rice in this study was 1.72%.

The head rice yield significantly (P ≤ 0.05) increased when the interaction of soaking temperature and steaming time was increased. The highest percentage of head rice yield was 86.34%, which was recorded for samples treated at 80°C and 35 minutes of steaming. The minimum value was 80.99% recorded for samples subjected to 80°C soaking temperature and 15 minutes of steaming time. A similar study on the parboiling process showed that soaking at 67.7°C temperature for 18 minutes had a significant influence (at a 95% confidence level) on head rice yield, with 70% output (Ogunbiyi et al., 2018). According to the study by Rahimi-Ajdadi (Rahimi-Ajdadi et al., 2018), the highest head rice yield (68.647%) was achieved at a combination of 45°C of soaking temperature and a steaming time of 10 minutes.

3.2.3. The interaction effect of soaking and steaming times on milling quality

Table 6 presents the interaction effect of soaking time and steaming time on milling quality. The milling quality of parboiled rice resulted in a significant effect of the interaction of soaking and steaming times. The percentage of husking was slightly increased with the increase in soaking time and steaming time. Sample with soaking times of 6 and 12 hours combined with 15 minutes of steaming time recorded the lowest husking level (83.56%). The highest values (87.33% and 87.67%) were recorded for samples with 12 and 24 hours of soaking times combined with a steaming time of 35 minutes. Likewise, the milling recovery was affected by the interaction of soaking and steaming times. Statistically, the highest (87.16 and 87.61%) were recorded for samples subjected to 6 hours of soaking time paired with 15 minutes of steaming time and also for samples subjected to 12 hours of soaking time paired with 35 minutes of steaming time, respectively. The data did not show a clear trend attributed to either of the two factors and varied randomly.

Table 6. Interaction effect of soaking time with steaming time on milling quality parameters

Soaking Time (hr)	Steaming Time (min)	Husking (%)	Milling Recovery (%)	Broken (%)	Head rice yield (%)
	15	83.56^f ± 1.13	87.16^ab ± 2.50	6.94^b ± 1.84	80.46^f ± 0.86
6	25	84.89^e ± 2.57	85.56^e ± 1.50	4.23^d ± 1.73	81.55^cd ± 0.97
	35	86.56^b ± 1.81	86.00^de ± 2.64	3.22^e ± 1.69	82.76^b ± .06
	15	83.56^f ± 0.88	85.44^e ± 2.74	4.59^c ± 2.74	80.85^ef ± 0.50
12	25	86.00^c ± 1.22	86.90^bc ± 0.72	3.98^d ± 1.43	82.82^b ± 1.55
	35	87.67^a ± 0.86	87.61^a ± 0.72	3.06^e ± 1.36	84.44^a ± 1.86
	15	85.11^e ± 1.05	86.34^cd ± 1.03	4.11^d ± 1.53	82.14^c ± 1.04
24	25	85.56^d ± 1.81	83.65^f ± 1.67	2.62^f ± 0.97	81.31^de ± 0.67
	35	87.33^a ± 0.50	86.56^bcd ± 2.03	2.18^g ± 1.52	84.36^a ± 2.63
Control		75.00^g ± 3.61	79.30^g ± 2.95	41.73^a ± 1.94	58.02^g ± 2.94
CV (%)		0.47	0.79	8.01	0.8
LSD0.05		0.38	0.64	0.29	0.62

Note: means represented by the same letter are not significant; values are in mean±SD.

Regarding broken rice kernels, increasing the soaking and steaming times significantly decreased the broken rice percentage. Generally, the lowest values occurred in samples subjected to the longest soaking time, within which an increase in steaming time resulted in further reduction of the broken rice kernels. On the other hand, the highest values (6.94%) were recorded for those with the shortest soaking time (6 hours), within which an increase in steaming time caused the reduction of the broken kernels. Parboiling treatment of paddy increases milling recovery and improves shelf life (Kumar et al., 2018). The percentage of head rice yield has been dominated by steaming time within the interaction. In each category of soaking times, those combined with the longest steaming time (35 minutes) resulted in the highest percentage of head rice yield, which included 82.76, 84.44, and 84.36% for soaking times of 6, 12, and 24 minutes, respectively. Generally, head rice yield significantly increased when the soaking and steaming times increased due to the increase in the level of gelatinization.

3.3. Interaction Effects of Pre-treatment Temperature, Soaking, and Steaming Times on Milling Quality

3.3.1. Interaction effects of the three factors on the percentage of husking

Table 7 indicates the interaction effects of soaking temperature, soaking time, and steaming time on the milling quality of parboiled rice. Increasing the soaking temperature, soaking time, and steaming time increased the husking percentage, comparing the three levels from each factor. The percentage of rice husking was significantly (P ≤ 0.05) increased for the interaction of the three factors. The percentage of husking for the three factors of interaction ranged between 83.00%-88.67%. The highest value (88.67%) was recorded at a soaking temperature of 80°C, soaking time of 24 hours, and steaming time of 35 minutes. The minimum (75.00%) was obtained from the non-parboiled rice. A similar study showed that in a properly adjusted milling machine having a rubber roll husker, husking efficiencies could increase as high as 95% (Rehal et al., 2017).

Table 7. Three-way interaction effects of parboiling factors on milling qualities

SkTp	SkTm	SmTm	Husking	Milling Recovery	Broken rice	Head rice yield
40	6	15	84.33^f ± 1.53	90.40^a ± 0.58	9.3^b ± 0.49	81.10^cde ± 1.05
	6	25	83.67^g ± 1.53	87.31^def ± 0.96	6.19^d ± 0.52	81.11^cde ± 0.44
	6	35	84.67^ef ± 1.53	87.61^cde ± 1.36	5.41^e ± 0.51	82.20^bcde ± 0.85
	12	15	84.33^f ± 0.58	89.02^b ± 0.16	8.17^c ± 0.51	80.84^cde ± 0.47
	12	25	85.00^e ± 1.00	86.74^efg ± 1.07	5.77^de ± 0.51	80.97^cde ± 0.64
	12	35	87.00^c ± 0.00	86.93^ef ± 0.46	4.78^fg ± 0.43	82.14^bcde ± 0.69
	24	15	86.00^d ± 1.00	87.38^de ± 0.59	6.15^d ± 0.30	81.24^bcde ± 0.89
	24	25	86.00^d ± 1.00	85.50^hij ± 0.74	3.87^hij ± 0.16	81.63^bcde ± 0.60
	24	35	87.33^c ± 0.58	87.08^def ± 0.92	4.13^hi ± 0.64	82.94^abcde ± 1.21
60	6	15	83.00^h ± 0.58	85.70^ghi ± 0.81	5.25^ef ± 0.08	80.45^de ± 0.77
	6	25	88.00^b ± 1.00	84.89^ijk ± 0.90	2.27^mno ± 0.25	82.623^bcde ± 0.93
	6	35	87.000^c ± 1.00	82.66^no ± 0.90	1.86^nop ± 0.43	80.80^cde ± 0.53
	12	15	83.33^gh ± 0.58	84.05^klm ± 0.87	3.43^jk ± 0.17	80.62^cde ± 0.71
	12	25	86.00^d ± 1.00	86.71^efg ± 0.76	3.54^ijk ± 0.28	83.17^abcde ± 0.66
	12	35	88.00^b ± 1.00	87.77^cde ± 0.85	2.59^lm ± 0.23	85.18^abcd ± 0.89
	24	15	85.00^e ± 1.00	86.23^fgh ± 0.78	3.11^kl ± 0.09	83.12^abcde ± 0.88
	24	25	87.00^c ± 1.00	81.79°±0.49	1.76^op ± 0.11	80.64^cde ± 0.48
	24	35	86.00^d ± 1.00	84.08^klm ± 0.17	1.46^pq ± 0.12	82.40^bcde ± 0.44
80	6	15	83.33^gh ± 0.58	85.38^hij ± 0.55	6.28^d ± 0.06	79.84^e ± 0.17
	6	25	83.00^h ± 1.00	84.49^jkl ± 0.53	4.21^gh ± 0.19	80.91^cde ± 0.28
	6	35	88.00^b ± 1.00	87.73^cde ± 0.19	2.39^mn ± 0.06	85.28^abc ± 0.37
	12	15	83.00^h ± 1.00	83.25^mn ± 0.09	2.19^mno ± 0.04	81.08^cde ± 0.35
	12	25	87.00^c ± 1.00	87.25^def ± 0.23	2.64^lm ± 0.10	84.34^abcde ± 0.10
	12	35	88.00^b ± 1.00	88.13^bcd ± 0.14	1.82^nop ± 0.07	86.00^ab ± 0.51
	24	15	84.33^f ± 0.58	85.40^hij ± 0.54	3.07^kl ± 0.06	82.06^bcde ± 0.23
	24	25	83.67^g ± .53	83.65^lmn ± 0.37	2.22^mno ± 0.11	81.64^bcde ± 0.48
	24	35	88.67^a ± 1.53	88.52^bc ± 0.51	0.95^q ± 0.07	87.75^a ± 0.27
Control			75.00^i ± 3.61	79.30^p ± 2.95	41.73^a ± 1.94	58.02^f ± 2.94
CV (%)			0.47	0.79	8.01	0.8
LSD0.05			0.66	1.11	0.51	1.07

Note: SkTp: soaking temperature (^oC), SkTm: soaking time (hours), SmTm: steaming time (minutes), means represented by the same letter are not significant, values are in mean±SD.

3.3.2. Interaction effects of the three factors on the percentage of broken rice

The trend of the percentage of parboiled broken rice, as shown in Table 7, decreased with the increase in preheating temperature, soaking time, and steaming time. As shown from Table 7, the higher the temperature and parboiling time, the lower the percentage of broken rice obtained. The percentage of broken rice ranged from 0.95–9.3%, recorded at 80°C, 24 hours and 35 minutes, and 40°C, 6 hours, and 15 minutes, respectively. Whereas the percentage of the broken rice for non-parboiled rice was the highest, 22.80%, because of not gelatinized. A similar study also showed that the broken rice of parboiled rice has significantly (P ≤ 0.05) improved, which indicated that soaking at 67.70°C for 13 hours and steaming for 18 minutes reduced the broken rice ratio to 2.18% (Ogunbiyi et al., 2018). According to (Graham-Acquaah et al., 2015; Ogunbiyi et al., 2018), the broken percentage was 6.31% at 85°C soaking temperature and 40 minutes of steaming.

3.3.3. Interaction effects of the three factors on milling recovery

The milling recovery (Table 7) was slightly decreased with the increase in temperature, soaking time, and steaming time. The lowest temperature and time of parboiling conditions generated the highest milling recovery. The three highest values of milling recovery (90.40, 89.02, and 88.52%) were recorded at a soaking temperature of (40, 40, and 80°C), soaking time of 6, 12, and 24 hours, and steaming time of 15, 15, and 35 minutes, respectively. The data showed that the maximum value of the non-parboiled rice (79.30%) sample was below the minimum value of parboiled (81.79%), which was recorded for samples subjected to 60°C, 24 hours, and 25 minutes of the three factors. The regression and correlation of milling recovery are indicated in Table 7. A similar study stated that the highest milling recovery (98.5%) was recorded at 70°C soaking temperature and 3 hours of steeping time (Michael et al., 2018). Kumar (Kumar et al., 2018) also found that the paddy soaked for 40 minutes at 85–90°C provided a good milling yield. The study by Chavan (Chavan et al., 2018) and Graham-Acquaah et al (Graham-Acquaah et al., 2015) indicated that by soaking at 85°C for three hours and steaming for 40 minutes, the broken percentage was significantly (P ≤ 0.05) reduced to 6.31% from 27.25% of the control value.

3.3.4. The influence of the three variables on head rice yield (HRY)

The overall interaction of the factors was highly significant (P ≤ 0.05) for the percentage of head rice yield. The highest temperature and time interaction of parboiling conditions generated the highest percentage of head rice. In this study, the maximum value of the percentage of head rice yield recorded was 87.75% which was recorded at a combination of a soaking temperature of 80°C, soaking time of 24 hours, and steaming time of 35 minutes. The minimum value was 79.84%, recorded at a soaking temperature of 80°C, soaking time of 6 hours, and steaming time of 15 minutes. The head rice yield for the control was 66.02%, showing that parboiling significantly increased its value due to the retrogradation and hence hardening effect of the kernel. A similar study also argued that the optimum parboiling conditions for maximum head rice yield (70%) were identified at temperature, soaking time, and steaming time of 67.7°C, 13 hours, and 18 minutes, respectively (Ogunbiyi et al., 2018). According to Hasan et al. (), treating rice at 85°C for 3 hours of soaking temperature and steaming for 40 minutes aid in maximum (66.09%) head rice recovery. Parboiling was done on similar rice varieties grown in Ethiopia, Edget, Nerica 4, and Gumara obtained related results compared to this variety (Meresa et al., 2020).

4. Conclusions

The highest percentage of broken rice is the core problem in the Ethiopian rice production system. The effects of the parboiling process are much more significant in the reduction of broken rice and increasing the head rice yield. Increasing the pre-treatment temperature, soaking andand steaming times increased the percentage of head rice yield and percentage of husking. On the other hand, increasing these three factors reduced the percentage of broken rice from 41.73% (of non-parboiled) to the range of 0.95–9.30% (when parboiled). The head rice yield was increased from 66.02% (of non-parboiled) to 87.75% (when parboiled at 80°C soaking temperature, 24 hours of soaking time, and 35 minutes of steaming time). This indicated that the combination of the three factors at 80°C, 24 hours, and 35 minutes was the best combination for the lowest percentage of broken rice, highest husking, and highest head rice yield. Rice processors could maintain the milling qualities by parboiling the paddy rice using this temperature-time combination. In this experiment, the highest temperature (80°C), the highest soaking time (24 hr), and the highest steaming time (35 min) was found to be the best combination that obtained the best quality of rice.

Correction

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

Haramaya University is highly acknowledged for its technical support of research work. Fogera National Rice Research and Training Center, Ethiopian Institute of Agricultural Research (EIAR) is also highly acknowledged for the study sponsorship, research fund, and rice variety provision.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Fogera National Rice Research and Training Center, Ethiopian Institute of Agricultural Research (EIAR).

Notes on contributors

Melese Ageze Mihretu

Melese Ageze Mihretu is a researcher at the Ethiopian Institute of Agricultural Research, Fogera National Rice Research and Training Center, Ethiopia. His fields of specialization is in Food Science and Postharvest Technology. His research interest focuses on Postharvest Mechanization Research, and agricultural engineering studies, agro-processing technology.

Getachew Neme Tolesa

Getachew Neme Tolesa is an assistant professor (researcher and lecturer) at the Department of Food Science and Postharvest Technology at Haramaya University with over 17 years of experience in teaching and research. He specializes in Food Science and Postharvest Technology and mainly works and publishes in the areas of Postharvest food preservation, Postharvest handling, food value-addition, food science, food processing, food engineering, food value chain, food safety and nutrition intervention developments.

Solomon Abera

Solomon Abera is an associate professor (researcher and lecturer) of Food Engineering at the Department of Food Technology and Process Engineering, Haramaya University, with over 30 years of experience in teaching, research and development. He specializes in Food Engineering and mainly works and publishes in food engineering, postharvest food preservation, postharvest handling, food value-addition, food science, food processing, food engineering, food value chain, food safety and nutrition intervention research and developments.