Drying Technology An International Journal ISSN: 0737-3937 (Print) 1532-2300 (Online) Journal homepage: https://www.tandfonline.com/loi/ldrt20 Effect of tempering approach following cross-flow drying on rice milling yields Sangeeta Mukhopadhyay, Terry J. Siebenmorgen & Andy Mauromoustakos To cite this article: Sangeeta Mukhopadhyay, Terry J. Siebenmorgen & Andy Mauromoustakos (2019): Effect of tempering approach following cross-flow drying on rice milling yields, Drying Technology, DOI: 10.1080/07373937.2018.1564760 To link to this article: https://doi.org/10.1080/07373937.2018.1564760 Published online: 11 Feb 2019. Submit your article to this journal Article views: 4 View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=ldrt20 DRYING TECHNOLOGY https://doi.org/10.1080/07373937.2018.1564760 Effect of tempering approach following cross-flow drying on rice milling yields Sangeeta Mukhopadhyaya,b , Terry J. Siebenmorgenb, and Andy Mauromoustakosc a Florida Crystals Corporation, West Palm Beach, FL, USA; bDepartment of Food Science, University of Arkansas, Fayetteville, AR, USA; Agricultural Statistics Lab, University of Arkansas, Fayetteville, AR, USA c ABSTRACT ARTICLE HISTORY Three tempering approaches were followed after drying rough rice at 16.3% and 20.5% initial moisture contents (IMCs) using 57  C/13% RH air at an airflow of 0.56 (m3/s)/m2 for 30, 60, and 90 min in an experimentally simulated cross-flow drying column. For the longer drying durations, post-tempering head rice yields were consistently less when the interstitial air from rice from different cross sections of the drying column was allowed to “interact” during tempering than when the rice from these different cross sections was tempered separately; this effect was more prominent at the greater rice IMC. RH of the interstitial air during tempering was measured and used to estimate the minimum tempering durations required for the different tempering approaches. Received 30 August 2018 Revised 25 November 2018 Accepted 28 December 2018 Introduction In the Mid-South region of the United States of America (USA), rough rice is harvested at moisture contents (MCs) ranging between 14% and 22% on a wet basis (wb) and dried to 12.5% MC (wb) for safe storage. Rough rice is then dehulled and typically milled before consumption. Milled rice comprises “whole, intact” kernels called head rice, which is defined as milled kernels that are at least three-fourths of the original kernel length,[1] and broken kernels. “Milling yield” refers to both the milled rice yield (MRY) and the head rice yield (HRY), defined as the mass of milled rice and head rice, respectively, expressed as a percentage of the original, dried rough rice mass.[1] Because of the economic importance of head rice relative to broken kernels, the broad objective in any rice drying operation is to maximize HRY. Rice kernel fissures are fractures of the endosperm that can either be “cross-wise/perpendicular” to the long-axis of the rice kernel[2] or in no specific alignment.[3] Since fissuring is a key factor contributing to rice kernel breakage and consequent reduction in HRY[4], rice producers and processors aim to minimize fissuring. Fissuring in rice kernels can occur due to several reasons, including rapid moisture adsorption by low-MC kernels,[5,6] and improper drying and KEYWORDS Rice drying; cross-flow drying; tempering; glass transition temperature; head rice yield; milling yield post-drying/tempering processes.[7–9] The extent of fissuring depends on several factors, including cultivar differences,[10,11] rice grain dimensions, especially the length-to-width ratio of kernels,[12,13] and other physicochemical properties.[13,14] In the case of heated air drying, fissuring in rice kernels during the “drying process,” which comprises drying as well as tempering, has been hypothesized to occur primarily due to intra-kernel differences in starch properties due to portions of the kernel being above and below the “glass transition temperature (Tg).”[8,15] The Tg is a MC-dependent material property that determines whether a material is in a “rubbery” (above Tg) or “glassy” (below Tg) state, i.e., the MC/temperature combination “positions” a rice kernel, or different volumes of the kernel, on either side of the Tg line (rubbery/glassy regions) for the kernel, as plotted on a state diagram for rice kernels.[16] According to the Tg hypothesis,[8] intra-kernel material state transitions from the glassy to the rubbery state, and vice versa, can take place during the drying process. Depending upon the relative volumes of glassy and rubbery regions created within a kernel, the kernel may fissure from intra-kernel stresses resulting from the vastly different physical and mechanical properties of these two regions. CONTACT Terry J. Siebenmorgen tsiebenm@uark.edu Department of Food Science, University of Arkansas, 2650 N Young Avenue, Fayetteville, AR 72701-4002, USA Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ldrt. ß 2019 Taylor & Francis Group, LLC 2 S. MUKHOPADHYAY ET AL. Typically, rough rice is dried from harvest MC to storage MC using several drying and tempering passes. In the USA, drying is commonly accomplished in cross-flow, column dryers using air temperatures ranging from 40  C to 70  C to achieve high throughput rates. In such dryers, rice flows vertically downward between two metal screens, which form the grain column, whereas heated air flows from the heated air plenum (HAP) through the grain column in a direction perpendicular (or “cross-wise”) to that of the grain movement. Unloading feed-roll augers, located at the bottom of the dryer columns, meter the rice out of the dryer. Tempering is a standard practice in most commercial rice drying processes. Tempering offers several advantages, including allowing intra-kernel material state gradients, which are typically created during heated-air drying, to subside.[8,17] Because these gradients are allowed to subside, fissuring and consequent HRY reductions are minimized; this has been demonstrated in several “thin-layer” drying studies[7,8] and in one “deep-bed” study.[18] Tempering at elevated temperatures has been shown to minimize fissuring[7,13,19–22] and maximize HRYs.[8,9,20–27] Several studies[8,20,21,27,28] have also shown that for heated air drying, tempering at the drying air temperature for a “sufficient duration” was required for intra-kernel material state equilibration and, in turn, for maintaining HRYs; some studies[9,22,29] also showed that tempering at elevated temperatures decreased the required tempering durations. An additional advantage of tempering is that the total drying duration is decreased[20,21,30] by improving the drying rate in subsequent drying passes,[31–33] thereby increasing overall energy efficiency.[34] Previous research on tempering conditions included a wide range of tempering temperatures and corresponding durations for a range of rice initial MCs (IMCs) to maintain HRYs.[8,9,22,23,29,35–37] While it is known that tempering in the rubbery region, i.e., at a temperature > Tg, reduces fissuring, the length of the duration required for “sufficient tempering” is still not clearly established. The recommendations for “sufficient tempering durations” from the studies mentioned above were based upon stationary-bed, “thin-layer” drying tests; a thin-layer is defined as a layer of particles whose depth does not exceed three layers of particles.[38] In the case of rough rice, this layer is 0.6 cm in depth/thickness. But in reality, rough rice is not dried in “thin-layers” in commercial systems, but as “deep-beds,” with typical thicknesses of the drying column in cross-flow dryers varying between 25 and 45 cm. Moreover, in commercial and on-farm dryers, rice from a column cross section is mixed while exiting the dryer, and thus, the effect of tempering the rice together in bulk may be different from that reported in thin-layer drying tests. The effect of tempering a “deep-bed” of rice was not found in the literature. Additionally, it would be valuable to understand the extent of fissure occurrence when rice from different cross sections of a cross-flow dryer, i.e., rice located at different distances from the HAP, is removed and tempered separately. This understanding may possibly be used to design better cross-flow rice dryers to improve the rice drying process. The present study was conducted with the goal of understanding the effect of post-drying tempering approaches or methods on milling yields when rice from different dryer cross sections was tempered in different ways. Because the drying treatment is very important in creating intra-kernel material state gradients, it was deemed critical that the tempering treatments be conducted using rice that had been immediately prior-dried in a system representative of commercial cross-flow drying systems. As such, rice for this study was dried within a drying column that was designed to simulate a cross-flow drying action; the system is described herein and reported previously.[18] The objectives of this study were to (1) elucidate the effect of tempering approaches on milling yields of rice that had been immediately prior-dried in a system representative of cross-flow drying systems, (2) investigate the effect of rice IMC and drying duration on these milling yields, and (3) estimate the shortest tempering duration required using the RH of the interstitial air during tempering for each tempering approach. Materials and methods Sample procurement and preparation A 600-kg bulk lot of rough rice (pureline, long-grain cultivar Roy J) was combine-harvested at the University of Arkansas Rice Research & Extension Center, Stuttgart, AR, USA, in September, 2015 at 23.2% MC (wb). The lot was cleaned using a dockage tester (XT4, Carter-Day, Minneapolis, MN, USA) to remove foreign material and unfilled kernels. The bulk lot was divided into two 300-kg sub-lots, which were gently conditioned to approximately 16% and 21% MC (wb), respectively, using air at 26  C/56% RH, and stored in sealed containers at 4  C for 4 months. Immediately prior to drying treatments, the rice MC was determined to be 16.3% and 20.5% (wb), respectively, by drying duplicate, 15-g sub-sample in a DRYING TECHNOLOGY 3 Figure 1. Schematic diagram of the drying system. The drying assembly comprised a wooden box, an acrylic glass drying column with a metallic screen base, and a set of ten, fiber-mesh cylindrical baskets filled with rough rice and temperature/relative humidity (T/RH) sensors/loggers, all positioned inside a controlled environment chamber.[17] convection oven (1370FM, Sheldon Manufacturing Inc., Cornelius, OR, USA) maintained at 130  C for 24 h.[39] Experimental drying system Drying runs were accomplished inside a 0.91-m3, controlled-environment chamber (Platinous Sterling Series, ESPEC North America, Hudsonville, MI, USA) that was capable of producing drying air at any desired temperature (–35 C to 180  C) and relative humidity (5% to 98% RH). During operation, a centrifugal fan (4C108, Dayton Electric Manufacturing Co., Niles, IL, USA), coupled to a 0.56-kW electric motor (3N443BA, Dayton Electric Manufacturing Co., Niles, IL, USA), suctioned air at the desired temperature/RH (T/RH) from the chamber through a port located in the front door of the chamber and then forced the air through a side-wall port in the chamber wall to a drying assembly (Figure 1). A variable-frequency drive (AF-300 Mini, GE Fuji Drives USA Inc., Salem, VA, USA) was connected to the fan motor to achieve the desired airflow rate. The drying assembly (Figure 1) comprised a wooden box that served as a HAP, an acrylic glass drying column with a metallic screen base, and a set of ten, fiber-mesh cylindrical baskets (3.8-cm deep, 12.7-cm diameter). The fiber-mesh baskets were used as “sample holders” and enabled the drying column to be divided into discrete layers, thus permitting sampling at various distances from the HAP. Each basket had a fiber-mesh flap on top that could be fastened using metal pins and a fiber-mesh handle to facilitate dismantling after the completion of a drying run. Drying conditions Industry experts were consulted and temperature/relative humidity (RH) readings were taken at several cross-flow dryers operating in AR, USA. Readings were averaged, and the typical drying condition used for cross-flow drying in the Mid-south USA was determined to be approximately 56 C to 58  C. Thus, a drying temperature of 57  C was selected for this study. In commercial cross-flow dryers, the RH of the drying air is not controlled directly, but because ambient air is heated to an elevated temperature and blown 4 S. MUKHOPADHYAY ET AL. into the dryer, its RH drops, following the principles of psychrometrics. An analysis of the historical weather data during the typical rice drying season showed that when ambient air is heated to achieve HAP of 57  C, its RH is approximately 13% at the HAP of the cross-flow dryer. Thus, 57  C/13% RH was selected to be the drying temperature/RH condition for all drying runs in this study. Thus, rough rice at two IMCs (16.3% and 20.5%) was dried in an apparatus designed to simulate crossflow drying (described below) using 57  C/13% RH air at an airflow rate of 0.56 (m3/s)/m2 for three drying durations (30, 60, or 90 min). After drying, the rice was tempered using one of the three approaches described below. Each drying condition/tempering approach treatment was replicated twice, thus yielding 36 drying/tempering runs (2 IMCs  1 drying air T/ RH  1 airflow rate  3 drying durations  3 tempering approaches  2 replications). Drying procedure Approximately 2.7 kg of rough rice was removed from cold storage and placed in a plastic ZiplocV bag. Then, the bag was sealed and equilibrated to room temperature (22 ± 1  C) for 24 h prior to each drying run. Each of the ten baskets was filled with 270 g of rough rice. A T/RH sensor/logger (UX100-011, Onset Corporation, Bourne, MA, USA; accuracy: ± 0.21  C and 2.5% RH) was placed inside each basket to measure air conditions during drying. Stacking the ten baskets in the acrylic glass cylinder resulted in a 38-cm deep rice column. Additionally, four T/RH sensors, as shown in Figure 1, were utilized; two were placed inside the HAP and two at the top of the rice column to record air conditions at the column inlet and exhaust, respectively. When the controlled environment chamber stabilized at the T/RH setting for a drying run, the chamber door was opened for < 1 minute to position the rice column on the HAP. After the desired drying duration, the drying column was dismantled. R Tempering conditions For all tempering approaches, a tempering temperature of 60  C and duration of 4 h were selected; the reasons for selecting these conditions are as follows. Air at 57  C/13% RH was used to dry the 16.3%and 20.5%-IMC rice for 30, 60, and 90 min; according to the Tg hypothesis described above, for these rice MC/temperature combinations, most of the rice kernels would transition into the rubbery region during drying. It was deemed necessary that tempering be conducted in the rubbery region as well, i.e. at a temperature >Tg.[7] Thus, for all three tempering approaches followed in this study, rice was tempered inside a preheated oven that was maintained at 60  C (Tg for 16.3%- and 20.5%-IMC rice are 46 C and 32  C, respectively[15]). This ensured that the rice kernels were “held in the rubbery region” or “above the Tg line”[7] to allow the intra-kernel material state gradients that had developed during drying to subside sufficiently before cooling, which, in turn, would minimize fissuring and consequent HRY reduction after drying using air at elevated temperatures. A preliminary experiment was conducted to determine the tempering duration required at the 60  C tempering temperature for the developed intra-kernel material state gradients to subside sufficiently. Using the procedure described above, the 16.3%- and 20.5%IMC rice lots were dried using 57  C/13% RH air at an airflow rate of 0.56 (m3/s)/m2 for the longest drying duration (90 min) used in this study; these drying conditions would create the maximum material state gradients within the rice kernels. After each drying run, the drying column was dismantled, and the contents of all ten baskets were mixed and placed into a plastic bag. The bag was sealed and placed in the preheated oven at 60  C for different durations (1 to 8 h, at 1-h increments). Following a given tempering duration, a 500-g sample was taken from the bag and conditioned to 12.5% MC (wb) in a chamber maintained at 26  C/56% RH and milled per the procedure described below to measure HRY. Eight drying runs were conducted for each rice IMC level (one drying run for each tempering duration); with two replications, this resulted in 32 drying runs (2 IMCs  8 drying runs 2 replications). HRY was plotted against tempering duration, and it was found that HRYs stabilized after  3 h of tempering for both rice IMCs. A previous study[16] reported that all moisture gradients were eliminated after about 4 h of tempering at 50  C after drying short-grain rough rice for 1 h using 50  C/60% RH air. Thus, the tempering duration was set to 4 h for this study. Tempering approaches In commercial and on-farm cross-flow dryers, rice flows downward through the drying columns, and feed-roll augers meter the dried rice out of the columns. Thus, the rice from different column cross sections is mixed as it exits the dryer. In order to DRYING TECHNOLOGY 5 Figure 2. Tempering approaches (TAs) TA1 (a), TA2 (b), and TA3 (c) followed after immediately prior-drying rough rice at 16.3% or 20.5% initial moisture contents in the drying assembly shown in Figure 1 using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 30, 60, or 90 min. simulate this experimentally, after the drying column (Figure 1) was dismantled following a drying run, the rice from all ten baskets was mixed together and placed into a plastic bag. Then, the bag was sealed and kept in the preheated oven (60  C) for 4 h. A T/ RH sensor, which had been preprogramed to record T/RH conditions every 60 s, was also placed inside the mass of rice in the sealed bag. After the 4-h duration, a 500-g sample was spread into a thin layer on a screened tray and conditioned to 12.5% MC in a chamber maintained at 26  C/56% RH to conduct a milling analysis (described below). This tempering approach (designated as TA1; Figure 2a) was thus expected to be representative of the mixing that rice undergoes during the drying process in an actual cross-flow dryer. It was also desired to quantify the milling yield of rice located at various distances from the HAP if that rice was tempered separately after each drying treatment; for this, TA2 was followed (Figure 2b). In TA2, after the drying column was dismantled following a drying run, each basket was kept intact and placed into an individual plastic bag, and then, each bag was sealed. The T/RH sensors inside the baskets during drying continued to record T/RH conditions every 60 s during tempering. After the ten bags were tempered inside the 60  C-oven for 4 h, the contents of each basket were separately conditioned to 12.5% MC (wb) and milled. These data would be representative of the milling yields of layers of rice located at specific distances from the HAP during drying and then tempered separately. Additionally, it was desired to elucidate the “interaction effect,” during tempering, of the interstitial air from rice dried at different cross sections of the drying column on the respective milling yields of the rice located at those cross sections. To simulate this scenario, TA3 was followed (Figure 2c); the ten 6 S. MUKHOPADHYAY ET AL. baskets were kept intact to preserve their identity as separate milling samples, but all the baskets were placed together inside one individual plastic bag and then sealed, thus allowing the interstitial air from each of the baskets to interact/mix. Immediately after placing the ten baskets in the bag and sealing it, the bag was placed inside the 60  C-oven for 4 h. Following the 4-h duration, the contents of each basket were separately conditioned to 12.5% MC (wb) and milled. Similar to the situation in TA2 described above, the T/RH sensors that were placed inside each basket during the drying treatment continued to record T/RH conditions every 60 s during the tempering duration. In this way, TA3 built upon TA2, in that the milling yield of rice as a function of distance from the HAP was produced, but in TA3, the interstitial air from different layers in the drying column (or baskets) was allowed to “interact,” whereas in TA2, the contents of each basket were kept autonomous during tempering. For all tempering treatments, the T/RH sensors were retrieved after each tempering run and the data downloaded. It must be noted that in TA1, once the rice from all baskets was mixed together, a 40 g sample was also taken and placed in a plastic bag and sealed. This 40 g sample was allowed to equilibrate in the sealed bag at 22 ± 1  C for 48 h; thereafter, the MC of the sample was measured following the above oven-drying procedure to indicate the final average MC of the drying column immediately after the drying run. Additionally, it was desired to quantify the MC of rice in each basket after a given drying treatment. To do this, a separate drying run was conducted using the same drying air conditions but following that drying run, the rice from each basket was preserved as separate samples; a 40-g sample was taken from each basket for oven MC analysis immediately after drying to indicate the final MC as a function of distance from the HAP. TA1 þ 6 from TA2 þ 6 from TA3). With a total of 6 drying combinations (2 IMCs  1 drying air T/RH  1 airflow rate  3 drying durations), 13 milling samples per drying combination, and 2 replications, 156 samples were milled. For each milling analysis, a 150-g rough rice sample was dehulled using a laboratory huller with a clearance of 0.048 cm between the rollers (THU-35A, Satake Engineering Co., Ltd., Tokyo, Japan). The brown rice was then milled for 19 s (previously determined as the milling duration required to achieve 0.4% surface lipid content) using a laboratory mill (McGill No. 2, Rapsco, Brookshire, TX, USA) with a 1.5-kg mass placed on the lever arm 15 cm from the center of the milling chamber. A sizing device (61, Grain Machinery Manufacturing Co., Miami, FL, USA) was then used to separate head rice from broken kernels. Milling yield was quantified as MRY and HRY. Statistical analyses Using JMP Pro software (Version 13.0.0, SAS Institute, Inc., Cary, NC, USA), nonlinear model fitting was performed on the RH data recorded by the T/RH sensors during the 4-h tempering durations. Results and discussion General layout Since the goal of this study was to develop a better understanding of the effect of tempering approach on milling yield, the milling yields from the three tempering approaches are presented for both rice IMCs and across all drying durations. Following this, the responses from the T/RH sensors that were placed inside the sealed bag(s) during tempering are discussed; however, for brevity, only the data from the 60-min drying runs are presented. Milling analyses Final bulk-column moisture contents TA1, in which the rice from all ten baskets was mixed, yielded one milling sample, while TA2 and TA3, in which baskets were kept separate, each yielded ten milling samples per drying/tempering treatment. However, for the individual baskets, it was reasoned that the milling yields of baskets B1, B2, B3, B5, B7, and B10 (Figure 1) would provide sufficient trends across the rice column. Hence, for each drying/ tempering treatment combination, 13 milling samples were produced per experiment replication (1 from When 16.3%-IMC rice was dried using 57  C/13% RH air at 0.56 (m3/s)/m2 for 30, 60, and 90 min, the final bulk-column MCs, with the associated percentage point moisture content reductions (PPMRs), were 14.4% (1.9 PPMR), 12.6% (3.7 PPMR), and 11.0% (5.3 PPMR), respectively, whereas when 20.5%-IMC rice was dried using the same air conditions, the corresponding MC data for 30, 60, and 90 min of drying were 16.8% (3.7 PPMR), 14.8% (5.7 PPMR), and 12.7% (7.8 PPMR), respectively. Thus, for the same DRYING TECHNOLOGY 7 Figure 3. Final moisture contents (MCs) as a function of distance from the heated air plenum after drying rough rice at 16.3% and 20.5% initial MCs (wet basis), respectively, for 30, 60, and 90 min using 57  C/13% RH air at an airflow rate of at 0.56 (m3/s)/ m2. For the 16.3%-IMC rice, the final bulk-column MCs were 14.4%, 12.6%, and 11.0% for drying durations of 30, 60, and 90 min, respectively. For the 20.5%-IMC rice, the final bulk-column MCs 16.8%, 14.8%, and 12.7% for drying durations of 30, 60, and 90 min, respectively. Data points are the mean values of two experimental treatment replications. drying duration, rice at a greater IMC underwent a greater PPMR than rice having a lesser IMC. This can be explained in terms of the total energy required to dry rice from a given IMC to a desired final MC.[40] According to that study, the total heat of desorption of rice (Qt) is a function of rice MC and kernel temperature; Qt decreases exponentially as MC increases at a given temperature, indicating that water is increasingly bound to the rice matrix as IMC decreases. In other words, for the same energy supplied, rice at a greater IMC (20.5%) will dry more, i.e., have a greater PPMR, than rice at a lesser IMC (16.3%). Final moisture contents at different distances from the HAP The “non-uniformity” of cross-flow drying is apparent in Figure 3, which shows the final MCs as a function of distance from the HAP. This non-uniformity of MCs within the rice bed could be quantified using the “range of MCs within the drying column,” defined as the difference between the final MC of B10 and the final MC of B1 (Figure 1). For both rice IMCs, the range of MCs within the drying column decreased with increasing drying duration (Figure 3). For example, for the 16.3%-IMC rice, this MC range was 2.7 percentage points (PPs) when the drying duration was 30 min, but decreased to 2.5 PPs as the drying duration increased to 60 min, and then decreased further to 2.0 PPs when the drying duration increased to 90 min. Similarly, for the 20.5%-IMC rice, this MC range was 3.8, 3.5, and 3.0 PPs after drying for 30, 60, and 90 min, respectively. Thus, for a particular riceIMC, the non-uniformity of drying decreased with longer drying durations. For a particular drying duration, the range of MCs for the 16.3%-IMC rice was less than those for the 20.5%-IMC rice (Figure 3), indicating that with all other drying conditions remaining the same, the nonuniformity of drying decreased with decreasing IMCs; this may have implications on the HRY of the rice exiting from the drying column after a drying pass. As reported in other studies,[18,41] these MC results within the drying bed also showed that even for a typical 12.5% to 13.0% bulk-column final MC, there was severe over-drying of the rice kernels located adjacent to the HAP (Figure 3). For example, when 16.3%IMC rice was dried using 57  C/13% RH air at 0.56 (m3/s)/m2 for 60 min, the final bulk-column MC was 12.6%, but the final MC of B1 (nearest to the HAP) was only 11.1%. Since the depth of each basket was 3.8 cm, it is likely that the rice kernels immediately adjacent to the HAP were severely over-dried to MCs of 5% to 6%. Milling yields from the three tempering approaches For both rice IMCs, no significant differences were obtained in MRYs (data not shown) across all drying condition/tempering approach treatments; thus, the following discussion is limited to HRYs. Figure 4 shows the HRY trends after tempering (per the three tempering approaches as listed in Figure 2) immediately prior-dried rice lots at 16.3% and 20.5% IMC using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 30, 60, and 90 min. The HRYs from TA1 in Figure 4 are represented by a solid green line since TA1 yielded one composite milling sample from the 38-cm deep drying column (Figure 2a), but the data points shown in Figure 4 for TA2 and TA3 represent HRYs from samples as a function of distance from the HAP of the drying column (Figure 1). Figure 4 shows that for both IMCs, HRYs from TA1 progressively decreased with increasing drying duration. For example, for the 16.3%-IMC rice, HRYs of the bulk column after drying for 30, 60, and 90 min 8 S. MUKHOPADHYAY ET AL. Figure 4. Head rice yields (HRYs) of long-grain rough rice at 16.3% (a) and 20.5% (b) initial moisture contents (IMCs) after drying and tempering. Drying was conducted in a system simulating cross-flow drying (Figure 1) using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 30, 60, and 90 min, followed by tempering. The three tempering approaches (TA1, TA2, and TA3) are detailed in Figure 2; the line for TA1 represents the HRY of a composite sample of the entire drying column and the data points for TA2 and TA3 represent HRYs from samples as a function of distance from the heated air plenum of the drying column. Line/data points are the mean of two experimental treatment replications. were 55.0%, 54.5%, and 50.4%, respectively (Figure 4a), and for the 20.5%-IMC rice, HRYs were 55.6%, 54.0%, and 48.3%, respectively (Figure 4b). Additionally, these data show that for the 90-min drying duration, the 16.3%-IMC rice had a slightly greater bulk-column HRY than that of the 20.5%IMC rice. The HRYs obtained from TA2 are those of rice samples that were tempered separately after being dried at different cross sections of the drying column, i.e., at different distances from the HAP (Figure 2b). For both rice IMCs, HRYs of these separately tempered rice samples that had been prior-dried at different distances from the HAP were quite stable for the shortest drying duration of 30 min, particularly for the 16.3%-IMC rice. But as the drying duration increased to 60 min and then to 90 min, HRYs obtained after tempering were much less for samples that had been situated near the HAP during drying than those situated farther into the drying column (Figures 4a and b). This can be quantified by calculating the “range of HRYs within the drying column” (difference between the HRY of B10 and the HRY of B1). When 16.3%-IMC rice was dried for 30 min and then tempered separately, the range of HRYs within the drying column was 0.4 PPs; this range increased to 2.9 PPs and then to 5.6 PPs as drying duration increased to 60 min and then to 90 min. Similarly, when 20.5%IMC rice was dried for 30 min and then tempered separately, the range of HRYs within the drying column was 2.2 PPs, and this range increased to 4.4 PPs and then to 12.0 PPs as drying duration increased to 60 min and then to 90 min. Thus, for the longer drying durations of 60 and 90 min, samples that had been situated near the HAP incurred reductions in HRY; these HRY reductions in the baskets near the HAP are believed to be due to the development of greater intra-kernel material state DRYING TECHNOLOGY gradients in the rice located near the HAP due to excessive drying; tempering for 4 h at 60  C did not prevent reductions in HRY. Figure 4 also shows the effect of IMC on HRY reductions of rice samples that were tempered as separate samples after being dried at different distances from the HAP. For both rice IMCs, rice near the HAP incurred severe HRY reductions for the 60- and 90-min drying durations; however, the severity of HRY reduction was less for the 16.3%-IMC rice than the 20.5%-IMC rice (Figure 4). For example, when 16.3%-IMC rice was dried for 60 min and then baskets tempered separately, the HRY of B1 (basket next to the HAP) was 52.6%, whereas when 20.5%-IMC rice was dried for the same duration and then baskets tempered separately, the HRY of the same basket was 49.0%. Similarly, when 16.3%-IMC rice was dried for 90 min and tempered as separate samples, the HRY of B1 was 50.1%, whereas when 20.5%-IMC rice was dried for the same duration and tempered thereafter, the HRY of the same basket was 42.8%. This trend is believed to be due to the development of greater intra-kernel material state gradients in the greater IMC-rice during drying, leading to greater fissuring and increased HRY reductions in the greater IMC-rice. As discussed earlier, TA3 was conducted to elucidate the “interaction effect” of the interstitial air from rice samples that had been prior-dried at different cross sections of the drying column on the respective milling yields of the rice located at those cross sections, hence; all the baskets were kept intact but were tempered together in one sealed bag (Figure 2c). Figure 4a shows that for the 16.3%-IMC rice, the HRYs of TA2 and TA3 were similar for the 30- and 60-min drying durations. But as the drying duration increased to 90 min, in the two baskets adjacent to the HAP (B1 and B2, situated at 1.8 and 5.6 cm, respectively, from the HAP), HRYs from TA3 were less than HRYs from TA2. Figure 4b shows that for the 20.5%IMC rice, when the drying duration was 30 min, HRYs from TA2 and TA3 were similar. But the differences in HRYs from the two tempering approaches (TA2 and TA3) became more apparent at the longer drying durations, namely 60 and 90 min. Thus, for the 60- and 90-min durations, samples located at the same cross section of the drying column during drying but tempered per TA3, in which the baskets were placed in one bag following drying, generally had lesser HRYs (or greater HRY reductions) than those tempered per TA2, in which baskets were tempered in separate bags. For example, when 20.5%-IMC rice was 9 dried for 60 min and then tempered per TA3 (bags tempered together to allow mixing of interstitial air), the HRY of B1 (1.8 cm from the HAP) was 47.0%, which was 2.0 PPs less than the HRY of the same basket, at 49.0%, when TA2 (tempered in separate baskets) followed drying. Similarly, the HRY of B2 (5.6 cm from the HAP) was 49.3% when TA3 followed 60 min of drying, but was 51.3% when TA2 followed the same drying duration. An explanation for these HRY trends is as follows. In TA2, all baskets were tempered at 60  C (above Tg) for 4 h so that the intra-kernel material state gradients created during drying would subside sufficiently, so as not to cause additional fissuring post-drying. Yet, even with the tempering in TA2, there were reductions in HRYs, particularly in the baskets near the HAP; these HRY reductions are believed to be created due to extensive intra-kernel material state gradient formation and resultant fissuring due to excessive drying, not during tempering. The HRY reduction for TA2 samples progressively increased with proximity to the HAP, which corresponds to an increased severity of drying closer to the HAP. The reductions in HRY due to severe intra-kernel material state gradients created during drying, as explained for the samples tempered via TA2, would also have occurred to the samples tempered via TA3. Yet, samples tempered via TA3 showed greater HRY reductions than those of TA2, particularly for the 20.5%-IMC rice dried for 90 min. The fundamental reason for the increased fissuring that resulted in reduced HRYs in TA3 relative to TA2 is unknown, but could be due at least in part to moisture adsorption fissuring of low-MC kernels. In the TA3 case, all the baskets were placed in one bag, thereby allowing mixing of the interstitial air from the drier as well as the wetter rice. Hence, in TA3, unlike that of TA2, drier, lowMC rice from the vicinity of the HAP came in contact with the relatively high-RH interstitial air surrounding the rice situated farther away in the drying column. It is speculated that this rice/air interaction in TA3 resulted in moisture adsorption fissuring in the drier rice kernels near the HAP from the relatively greater RH of the interstitial air in the bag. Thus, for these drier rice kernels, there were additional reductions in HRYs during tempering per TA3, relative to TA2, that are attributed to moisture-adsorption fissuring. T/RH responses during tempering As described earlier, for TA1, following drying using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 30, 60, 10 S. MUKHOPADHYAY ET AL. Figure 5. Temperature and relative humidity profiles as a function of tempering duration when tempering approach 1 (TA1) was followed immediately after drying rough rice at 16.3% (a) and 20.5% (b) initial moisture contents (IMCs) using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 60 min. Final bulk-column MCs of the 16.3% and 20.5% IMC rice were 12.6% and 14.8%, respectively. TA1 is detailed in Figure 2. Lines are the mean of two experimental treatment replications. or 90 min, the contents of all the baskets were mixed and placed inside a plastic bag (Figure 2a). A T/RH sensor, which had been preprogramed to record T/RH conditions every 60 s, was also placed inside the mass of rice in the sealed bag. Then, the bag was sealed and kept in a preheated oven (60  C) for 4 h. Figure 5 shows the T/RH responses of the sensor that was placed inside the plastic bag containing rice from all the ten baskets during the 4-h tempering duration for the 16.3%- (Figure 5a) and 20.5%-IMC (Figure 5b) rice. The final bulk-column MCs of the 16.3%- and the 20.5%-IMC rice after 60 min of drying using 57  C/13% RH air at 0.56 (m3/s)/m2 were 12.6% (3.7 PPMR) and 14.8% (5.7 PPMR), respectively. Figure 5a shows that during the 4-h tempering duration per TA1, the T and RH readings of the sensor increased from 30  C/59% RH to 60  C/77% RH. Initially, the sensor was at room temperature (22  C). On being placed inside the bag of rice, the sensor started recording the temperature of the interstitial air inside the bag. Thus, as tempering progressed, the sensor’s temperature reading increased, reflecting the fact that the sealed bag of rice was kept inside a 60  C oven. The RH reading of the sensor progressively increased throughout the tempering duration as more and more moisture diffused from the interior to the periphery of the rice kernels and then evaporated into the interstitial air. Figure 5b shows the T/RH response of the sensor as in Figure 5a except that the IMC of the rice was 20.5%, and the final bulk-column MC of the rice was 14.8% (5.7 PPMR). Over the 4-tempering duration, the T and RH readings of the sensor increased from 30  C/78% RH to 60  C/89% RH. Although the temperature curve in Figure 5b was very similar to the corresponding curve in Figure 5a, the RHs recorded for the 20.5%-IMC rice (Figure 5b) throughout the tempering duration were much greater than those for the 16.3%-IMC rice (Figure 5a). This was expected since kernels in the 20.5%-IMC rice lot had reached an average final bulk-column MC of only 14.8% compared to that of 12.6% MC for the 16.3%-IMC rice and, thus, had greater amounts of water evaporating into the interstitial air during the tempering duration. A previous study[29] used changes in the interstitial air RH of short-grain rough rice kernels to define the “tempering index” in terms of RH, as shown in Equation (1). IRH ¼ where: RH ðt Þ RH ðt ¼ 0Þ RH ðt ¼ 1Þ RH ðt ¼ 0Þ (1) DRYING TECHNOLOGY 11 Figure 6. Temperature and relative humidity profiles inside the indicated baskets as a function of tempering duration when tempering approaches TA2 and TA3 (Figure 2) were followed immediately after drying rough rice at 20.5% initial moisture content using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 60 min. During drying, basket 1 (B1) was nearest to the heated air plenum. Lines are the mean of two experimental treatment replications. 12 S. MUKHOPADHYAY ET AL. Table 1. Estimated asymptotic relative humidity (RH) and minimum tempering duration (t0.95) for the indicated baskets when the indicated tempering approaches were followed after prior-drying rice at 16.3% and 20.5% initial moisture content (IMC) using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 60 min. Tempering approach TA2 (each basket was tempered separately) Rough rice IMC (%) 16.3 20.5 TA3 (each basket was kept intact, but tempered together in one sealed bag) Basket number Asymptotic RH (%) t0.95 (min) Asymptotic RH (%) t0.95 (min) B10 B7 B5 B3 B2 B1 B10 B7 B5 B3 B2 B1 79.3 72.0 69.5 67.5 63.4 60.7 88.8 83.2 83.0 80.1 78.6 74.6 47 45 72 74 62 89 37 61 67 67 66 66 80.3 72.6 71.7 69.5 71.7 68.3 83.8 83.2 80.7 77.4 77.5 75.0 – 57 77 81 94 103 – 44 65 62 93 91 Basket 1 (B1) was nearest to and B10 was farthest from the HAP (Figure 1). Tempering approaches are detailed in Figure 2. Estimates were obtained by fitting mechanistic growth models on the interstitial air RH data. IRH is the tempering index in terms of interstitial air RH; RH(t = 0) is the interstitial air RH at the start of tempering; RH (t) is the interstitial air RH at any duration “t” during tempering, and RH(t = 1) refers to the maximum interstitial air RH obtained in the interstitial air during tempering. This study[29] calculated the “tempering index” as a means to predict the shortest tempering duration required for adequate tempering to occur. Two degrees of tempering were considered; the first being “complete tempering” or IRH = 1.0, and the second being “95% tempering” or IRH = 0.95. In the present study, a similar approach was followed to estimate the “shortest tempering duration (t0.95)” required; for each tempering approach, the interstitial air RH profiles during tempering were used. The “fit curve” platform in the statistical package JMP Pro was used to fit several nonlinear models to the RH data; based on the least root-mean-square error (RMSE) and R-square (R2) values, a mechanistic growth model, which is a higher order, nonlinear equation, was determined to be the best fit for the RH data. On fitting the mechanistic growth model to the RH data obtained when 16.3%-IMC rice was tempered per TA1 after being dried using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 60 min (Figure 5a), the RMSE and R2 values were 0.44 and 0.99, respectively. Then, the mechanistic growth model was used to determine the parameter estimates; the interstitial air RH reached an asymptote of 71.9% RH (95% confidence interval (CI) = 71.8% to 72.0% RH). Thereafter, using the “inverse prediction” on the “fit curve” platform in JMP Pro, the “tempering duration required for 95% tempering to be complete” (t0.95) was calculated to be 144 min. Following a similar procedure for the 20.5%IMC rice (RMSE and R2 values for the model fit were 0.37 and 0.99, respectively), the interstitial air RH reached an asymptote of 91.4% RH (CI = 91.3% to 91.5% RH) and t0.95 = 144 min. Thus, it was estimated that if a 38-cm-thick/deep cross-flow drying column of approximately 16%- and 20%-IMC long-grain rough rice were dried using the abovementioned drying and tempering conditions, 144 min was required for 95% tempering to be complete, based on interstitial air RH of the dried rice. Results from the preliminary experiment (described above) showed that HRYs stabilized by 3 h of tempering. Although these two methods of estimating the shortest duration required for sufficient tempering to occur are very different from each other, it is interesting that the t0.95 estimate of 144 min (2 h 24 min) obtained using nonlinear model fitting on the RH data was close to the 3-h duration predicted in the preliminary HRY experiment based on milling yield response. Additionally, the preliminary experiment comprised 90-min drying runs, whereas the data in Figure 5 were from 60-min drying runs. It is expected that the 90-min drying runs would have had greater extents of intra-kernel material state gradients developed during drying and thus expectedly required a longer duration (3 h) for the subsidence of those gradients, as compared to the shorter duration required (2 h 24 min) for the subsidence of the gradients developed in the shorter, 60-min drying durations. Similarly, for both tempering approaches TA2 and TA3, a mechanistic growth model was fit using the interstitial air RH data from each of the ten baskets using the same procedure described above, and DRYING TECHNOLOGY estimates of the respective asymptotic RHs and t0.95 were calculated. As an example, Figure 6 shows the T/ RH profiles as a function of tempering duration for baskets B1, B2, B3, B5, B7, and B10 when TA2 and TA3 were followed after immediately drying 20.5%IMC rough rice using 57  C/13% RH drying air at 0.56 (m3/s)/m2 for 60 min. As distance from the HAP increased (i.e., from B1 to B10), the asymptotic RH progressively increased (Table 1), for the reasons previously described. Additionally, a comparison between the estimates of the asymptotic RH values from the corresponding baskets for TA2 and TA3 showed that the asymptotic RH values were not significantly different. Table 1 also shows that for baskets tempered per TA2 and TA3, t0.95 decreased with increasing distance from the HAP; this trend occurred in both rice IMCs tested. These t0.95-results from the respective baskets indicate that with decreasing extents of intra-kernel material state gradients being created during drying, shorter tempering durations will suffice, based on the interstitial air RH responses. Conclusions The following this study: 1. 2. 3. conclusions were drawn from When rice samples were tempered separately after being dried at different cross sections of the drying column (TA2), post-tempering HRYs at different cross sections of the column were quite stable for the shortest drying duration of 30 min. But for the 60- and 90-min durations, HRYs obtained after tempering were much less for samples that had been situated near the HAP during drying than those situated farther into the drying column. These HRYs reductions in the baskets located near the HAP are believed to be due to the development of greater intra-kernel material state gradients due to excessive drying. Additionally, the severity of HRY reduction in samples located near the HAP during drying was less for the 16.3%-IMC rice than the 20.5%-IMC rice; this is believed to be due to the development of greater intra-kernel material state gradients in the greater IMC-rice during drying, leading to greater fissuring and increased HRY reductions in the greater IMC-rice. For the same rice and drying conditions, posttempering HRYs were consistently lesser when the interstitial air from rice from different cross 4. 5. 13 sections of the drying column was allowed to “interact” during tempering (TA3) than when the rice from these different cross sections was tempered separately (TA2). It is speculated that this rice/air interaction in TA3 resulted in moisture adsorption fissuring, and thus additional HRY reductions, in the drier rice kernels near the HAP from the relatively greater RH of the interstitial air in the bag. Results obtained from nonlinear model fitting on the interstitial air RH data showed that 144 min of tempering was required when 16%- and 20%IMC long-grain rough rice was dried in the drying assembly described above using 57  C/13% RH air at an airflow rate of 0.56 (m3/s)/m2 for 60 min and then tempered together per TA1 (contents of all baskets mixed together) inside a 60  C oven for 4 h. The duration required for 95% tempering to be completed (t0.95) was estimated for rice located in each cross section of the drying column for tempering approaches TA2 (each basket tempered separately) and TA3 (bags tempered together to allow mixing of interstitial air); for both rice IMCs tested, t0.95 decreased with increasing distance from the HAP, indicating that with decreasing extents of intra-kernel material state gradients being created during drying, shorter tempering durations will suffice. In summary, the three tempering approaches used in this study provide a unique insight into the importance of tempering approach on HRY. The knowledge gained herein regarding the effect of tempering approach following cross-flow drying on HRY can be used to design better cross-flow rice dryers and, thus, to improve the rice drying process. Additionally, the post-tempering HRYs obtained in this study can be used for validation of mathematical models predicting HRY profiles in cross-flow rice drying scenarios. Acknowledgments The authors express their gratitude to Bhagwati Prakash (PhD) of the University of Arkansas Rice Processing Program (UARPP) for his input in planning this study, as well as Redentor Mijares Burgos and Joanne Baltz-Gray of the UARPP, for their assistance with processing samples. Disclosure statement No potential the authors. conflict of interest was reported by 14 S. MUKHOPADHYAY ET AL. Funding The authors thank the Arkansas Rice Research and Promotion Board and the corporate sponsors of the UARPP for financial support of this project. [12] [13] ORCID Sangeeta Mukhopadhyay 0530-9037 http://orcid.org/0000-0002- Literature cited [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] U.S. Department of Agriculture. Standards for Rice, Revised; Federal Grain Inspection Service, U.S. Government. Printing Office: Washington, DC. 2009. https://www.gipsa.usda.gov/fgis/standards/ ricestandards.pdf. Kunze, O. R.; Calderwood, D. L. Chapter 9: Rough Rice Drying - Moisture Adsorption and Desorption. In Rice Chemistry and Technology; Champagne E. T., Ed.; American Association of Cereal Chemists: St. Paul, MN, 2004; pp. 223–268. Bautista, R. C.; Siebenmorgen, T. J.; Fendley, J. W. Fissure Formation Characterization in Rice Kernels during Drying Using Video Microscopy. Res. Ser. 2009 - Arkansas Agric. Exp. Station. 2000, 581, 220–229. Odek, Z.; Prakash, B.; Siebenmorgen, T. J. X-ray Detection of Fissures in Rough Rice. Appl. Eng. Agric. 2017, 33, 721–728. DOI: 10.13031/aea.12369. Sharma, A. D.; Kunze, O. R. Post-drying Fissure Developments in Rough Rice. Trans. ASAE 1982, 25, 465–468. Zhang, Q.; Yang, W.; Sun, Z. Mechanical Properties of Sound and Fissured Rice Kernels and Their Implications for Rice Breakage. J. Food Eng. 2005, 68, 65–72. DOI: 10.1016/j.jfoodeng.2004.04.042. Nguyen, C. N.; Kunze, O. R. Fissures Related to Post-drying Treatments in Rough Rice. Cereal Chem. 1984, 61, 63–68. Cnossen, A. G.; Siebenmorgen, T. J.; Yang, W.; Bautista, R. C. An Application of Glass Transition Temperature to Explain Rice Kernel Fissure Occurrence during the Drying Process. Drying Technol. 2001, 19, 1661–1682. DOI: 10.1081/DRT100107265. Schluterman, D. A.; Siebenmorgen, T. J. Relating Rough Rice Moisture Content Reduction and Tempering Duration to Head Rice Yield Reduction. Trans. ASABE 2007, 50, 137–142. Bergman, C. J.; Bhattacharyya, K. R.; Ostsubo, K. Rice End-use Quality Analysis. In Rice Chemistry and Technology; Champagne E. T., Ed; American Association of Cereal Chemists: St. Paul, MN, 2004; pp 415–472. Sater, H. M.; Pinson, S. R. M.; Moldenhauer, K. A. K.; Siebenmorgen, T. J.; Mason, R. E.; Boyett, V. A.; Edwards, J. D. Fine Mapping of qFIS1-2, a Major QTL for Kernel Fissure Resistance in Rice. [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] Crop Sci. 2017, 57, 1511–1521. DOI: 10.2135/ cropsci2016.09.0821. Lan, Y.; Kunze, O. R. Fissure Resistance of Rice Varieties. Appl. Eng. Agric. 1996, 12, 365–368. Dong, R.; Lu, Z.; Liu, Z.; Koide, S.; Cao, W. Effect of Drying and Tempering on Rice Fissuring Analyzed by Integrating Intra-kernel Moisture Distribution. J. Food Eng. 2010, 97, 161–167. DOI: 10.1016/ j.jfoodeng.2009.10.005. Bhashyam, M. K.; Srinivas, T. Varietal Difference in the Topography of Rice Grain and Its Influence on Milling Quality. J. Food Sci. 1984, 49, 393–395. DOI: 10.1111/j.1365-2621.1984.tb12430.x. Perdon, A.; Siebenmorgen, T. J.; Mauromoustakos, A. Glassy State Transition and Rice Drying: Development of a Brown Rice State Diagram. Cereal Chem. 2000, 77, 708–713. DOI: 10.1094/ CCHEM.2000.77.6.708. Siebenmorgen, T. J.; Yang, W.; Sun, Z. Glass Transition Temperature of Rice Kernels Determined by Dynamic Mechanical Thermal Analysis. Trans. ASAE 2004, 47, 835–839. Dong, R.; Lu, Z.; Liu, Z.; Nishiyama, Y.; Cao, W. Moisture Distribution in a Rice Kernel during Tempering Drying. J. Food Eng. 2009, 91, 126–132. DOI: 10.1016/j.jfoodeng.2008.08.012. Mukhopadhyay, S.; Siebenmorgen, T. J. Glass Transition Effects on Milling Yields in a Cross-flow Drying Column. Drying Technol. 2018, 36, 723–735. DOI: 10.1080/07373937.2017.1351453. Li, Y. B.; Cao, C. W.; Yu, Q. L.; Zhong, Q. X. Study on Rough Rice Fissuring during Intermittent Drying. Drying Technol. 1998, 17, 1779–1793. DOI: 10.1080/ 07373939908917652. Iguaz, A.; Rodriguez, M.; Virseda, P. Influence of Handling and Processing of Rough Rice on Fissures and Head Rice Yields. J. Food Eng. 2006, 77, 803–809. DOI: 10.1016/j.jfoodeng.2005.08.006. Aquerreta, J.; Iguaz, A.; Arroqui, C.; Virseda, P. Effect of High Temperature Intermittent Drying and Tempering on Rough Rice Quality. J. Food Eng. 2007, 80, 611–618. DOI: 10.1016/ j.jfoodeng.2006.06.012. Tuyen, T. T.; Truong, V.; Fukai, S.; Bhandari, B. Effects of High-temperature Fluidized Bed Drying and Tempering on Kernel Cracking and Milling Quality of Vietnamese Rice Varieties. Drying Technol. 2009, 27, 486–494. DOI: 10.1080/ 07373930802686099. Steffe, J. F.; Singh, R. P.; Bakshi, A. S. Influence of Tempering Time and Cooling on Rice Milling Yields and Moisture Removal. Trans. ASAE 1979, 22, 1214–1218. 1224. Taweerattanapanish, A.; Soponronnarit, S.; Wetchakama, S.; Kongseri, N.; Wongpiyachon, S. Effects of Drying on Head Rice Yield Using Fluidization Technique. Drying Technol. 1999, 17, 346–353. DOI: 10.1080/07373939908917535. Inprasit, C.; Noomhorm, A. Effect of Drying Air Temperature and Grain Temperature of Different Types of Dryer and Operation on Rice Quality. DRYING TECHNOLOGY [26] [27] [28] [29] [30] [31] [32] [33] Drying Technol. 2001, 19, 389–404. DOI: 10.1081/ DRT-100102912. Tirawanichakul, S.; Prachayawarakorn, S.; Varanyanond, W.; Tungtrakul, P.; Soponronnarit, S. Effect of Fluidized Bed Drying Temperature on Various Quality Attributes of Paddy. Drying Technol. 2004, 22, 1731–1754. DOI: 10.1081/DRT-200025634. Ondier, G. O.; Siebenmorgen, T.; Mauromoustakos, A. Drying Characteristics and Milling Quality of Rough Rice Dried in a Single Pass Incorporating Glass Transition Principles. Drying Technol. 2012, 30, 1821–1830. DOI: 10.1080/07373937.2012.723085. Cnossen, A. G.; Jimenez, M. J.; Siebenmorgen, T. J. Rice Fissuring Response to High Drying and Tempering Temperatures. J. Food Eng. 2003, 59, 61–69. DOI: 10.1016/S0260-8774(02)00431-4. Steffe, J. F.; Singh, R. P. Theoretical and Practical Aspects of Rough Rice Tempering. Trans. ASAE 1980, 23, 775–782. Wasserman, T.; Ferrel, R. E.; Houston, D. F.; Breitwieser, E.; Smith G.S. Tempering Western Rice. Rice J. 1964, 67, 16–17, 20–22. Poomsa-Ad, N.; Soponronnarit, S.; Prachayawarakorn, S.; Terdyothin, A. Effect of Tempering on Subsequent Drying of Paddy Using Fluidization Technique. Drying Technol. 2002, 20, 195–210. DOI: 10.1081/DRT-120001374. Nishiyama, Y.; Cao, W.; Li, B. Grain Intermittent Drying Characteristics Analyzed by a Simplified Model. J. Food Eng. 2006, 76, 272–279. DOI: 10.1016/j.jfoodeng.2005.04.059. Ghasemi, A.; Sadeghi, M.; Mireei, S. A. Multi-stage Intermittent Drying of Rough Rice in Terms of Tempering and Stress Cracking Indices and Moisture Gradients Interpretation. Drying Technol. [34] [35] [36] [37] [38] [39] [40] [41] 15 2018, 36, 109–117. DOI: 10.1080/ 07373937.2017.1303777. Hwang, S.-S.; Cheng, Y.-C.; Chang, C.; Lur, H.-S.; Lin, T.-T. Magnetic Resonance Imaging and Analyses of Tempering Processes. J. Cereal Sci. 2009, 50, 36–42. DOI: 10.1016/j.jcs.2008.10.012. Shei, H.-J.; Chen, Y.-L. Intermittent Drying of Rough Rice. Drying Technol. 1998, 16, 839–851. DOI: 10.1080/07373939808917439. Yang, W.; Jia, C. C.; Siebenmorgen, T. J.; Howell, T. A.; Cnossen, A. G. Intra-kernel Moisture Responses of Rice to Drying and Tempering Treatments by Finite Element Simulation. Trans. ASAE 2002, 45, 1037–1044. Truong, T.; Truong, V.; Fukai, S.; Bhandari, B. Changes in Physicochemical Properties of Rice in Response to High-temperature Fluidized Bed Drying and Tempering. Drying Technol. 2018, 1. DOI: 10.1080/07373937.2018.1452031. ANSI/ASAE S448.2. Thin-layer Drying of Agricultural Crops; American Society of Agricultural and Biological Engineers. Michigan (MI): St. Joseph, 2014. (Revision approved). Jindal, V. K.; Siebenmorgen, T. J. Effects of Oven Drying Temperature and Drying Time on Rough Rice Moisture Content Determination. Trans. ASAE 1987, 30, 1185–1192. Billiris, M. A.; Siebenmorgen, T. J.; Mauromoustakos, A. Estimating the Theoretical Energy Required to Dry Rice. J. Food Eng. 2011, 107, 253–261. S Mukhopadhyay, S.; Siebenmorgen, T. J. Effect of Airflow Rate on Drying Air and Moisture Content Profiles inside a Cross-flow Drying Column. Drying Technol. 2017, Accepted, In press.