Inner cyclic cooling aeration on stored maize in large commercial warehouses in Northeast China

ABSTRACT

Controlling grain temperatures is the main approach employed in large commercial warehouses to maintain quality and prevent infestation by insects and fungi. Large commercial warehouses in northeast China have been utilizing inner cyclic cooling aeration (ICCA), a newly developed method for managing summertime temperatures, haphazardly and without tight standards. We utilized a homemade detector to track the temperature, humidity, and CO₂ concentrations of maize that was held for four years under ICCA treatment. The monitoring occurred every week, specifically from July 1st to September 30th. In addition, we conducted weekly tests to determine the moisture content (MC) and free fatty acids (FFA) of the stored maize samples. Without the use of ICCA, the average temperature of the stored maize above a depth of 3.5 m ranged from 20°C to 28°C, and the relative humidity (RH) in the headspace was typically higher than 75%. However, when ICCA was applied, the average temperature could be maintained at 22°C, and the RH was below 65% for most of the time. In the first year, the MC of the stored maize at a depth of 1.0 m experienced a loss of approximately 0.5%. However, for one year, there was no significant loss in the MC of the maize stored at the same depth. Furthermore, the FFA in the stored maize at a depth of 1 m increased by 1–5 mg KOH/100 g. The use of ICCA effectively controls the temperature and RH in the stored maize, reducing the risk of insect and fungal reproduction.

KEYWORDS:

• Large warehouses
• stored maize
• inner cyclic cooling aeration
• temperature
• CO₂

Introduction

Maize (Zea mays L.), one of the most widely grown crops, is used as a source of food, animal feed, and industrial raw materials^[1,2]. The total global maize production during the 2020–2021 crop season was estimated to be approximately 1.13 billion tons.^[3] The average loss rate for global grain production across the entire supply chain including harvesting and storage is generally considered to be around 10%.^[4]

Typical designs for commercial grain storage facilities in China are rectangular structures with dimensions of 40–60 m in length, 20–40 m in width, and 8–10 m in height.^[5] Owing to its airtightness and thermal insulation, the brick and concrete-built warehouse provides a comparatively stable environment for grains stored inside. Commercial cereal storage times in China often last three to four years. During this time, the stored grain creates a complex system where various living and non-living factors, including mites, insects, microorganisms, temperature, and moisture, interact with each other.^[6,7] The quality of the stored grain that has been stored is significantly impacted by both temperature and moisture content (MC).^[8,9] Most insects may grow quickly in grain storage at temperatures between 17 to 38°C.^[10] At temperatures below 13°C or above 40°C, insects tend to slow down or lose their activity, or even perish.^[11,12] Between temperatures of 20°C and 40°C, mold in stored grain will likely be noticeable. Furthermore, temperatures ranging from 25°C to 30°Care optimal for the growth of many molds found in stored grain.^[13] In addition to temperature, the growth of mold and insects also requires a certain level of moisture. When the relative humidity (RH) levels exceed 70%, mold multiplication occurs rapidly.^[14] Therefore, it is crucial to maintain a low RH, preferably below 60%, for long-term grain storage. The equilibrium MC for most cereal grains is below 13% when RH levels are maintained at this level.^[15]

Large commercial grain warehouses have recorded great success in reducing post-harvest loss and maintaining grain quality via the efficient and environmentally beneficial method of reducing grain temperatures through aeration.^[16] The stored grain is usually cooled in winter by atmospheric cold air, maintaining a relatively low temperature for several months until the following summer.^[17] The cooling aeration technology has gained attention since the 1950s.^[18] Equilibrium and non-equilibrium models have been developed for the aeration of wheat.^[19] Lawrence and Maier utilized the modified Chung-Pfost equation for wheat to develop grain aeration strategies for cooling and preserving MC.^[20]

Owing to grain’s low thermal conductivity and diffusivity, the stored grain temperature fluctuations and moisture migration at the surface and near the wall were higher than those inside and bottom sites in large bins or warehouses.^[21] During the summer, the surface of the stored grain is exposed to high-temperature air.^[22] Then, a temperature gradient commonly forms in stored grains as the upper layer is hotter than the bottom. The risk of stored grain quality degradation is increased when rising temperatures and rising humidity encourage the fast growth of molds and insects.^[23]

To reduce the stored grain temperatures during the summer season, the grain industry has implemented the inner cyclic cooling aeration (ICCA) technology in northeastern China.^[24] It involves utilizing temperature gradients within the stored grain and involves the use of small power centrifugal fans to transport low-temperature air through plastic tubes from the bottom of the grain layer to the surface. The operation of the fans is controlled by an automatic temperature control switch box, which activates them when the temperature in the warehouse space exceeds a predetermined value. The aeration process cools down the warehouse space temperature, and the fans are deactivated once the temperature falls below the set value. While there has been extensive research on aeration techniques for grain storage in winter, the ICCA technology has received limited attention. Currently, ICCA is implemented in a rudimentary manner by warehouse staff, with little consideration given to its impact on grain quality.

The objectives of this study were to 1) track changes in the temperature, RH, and CO₂ levels in stored maize warehouses during the ICCA working season; 2) identify variations in MC and free fatty acids (FFA) for stored maize over multiple years in large-scale warehouses; 3) assess the effect of ICCA in preserving stored maize and propose potential enhancements for future implementations.

Materials and methods

Grain warehouses and maize

The study focused on five stored maize warehouses in Jilin Province, located in northeastern China. In 2021, Jilin Province contributed significantly to the national maize production, accounting for approximately 15.4% of the total output, as reported by National Bureau of Statistics. The stored maize in commercial warehouses had a dimension of approximately 29 × 23 × 6 m, with a total weight of around 3100 t. The initial MC of stored maize varied between 12.7% and 14.1%, depending on the source from which it was purchased. According to Chinese policy, commercial maize storage periods typically span three to four years. Consequently, maize samples from each storage year were selected for analysis in this study.

Figure 1 depicts the components of the ICCA system, which included tiny powered centrifugal fans, temperature control boxes, vertical tubes, ventilation cages, etc. The centrifugal fans would begin to operate to lower the temperature once the warehouse headspace temperature rose above the limit established in the temperature control box. The cool air was then piped up through vertical tubes and ventilation cages from the bottom grain layer to the top area. The centrifugal fans ceased to run as soon as the temperature in the top space reached the preset level in the control box. Four centrifugal fans with a combined output of 1.1 kW were installed on either side of each warehouse wall for this study. The ICCA system typically operates from July 1st to September 30th. As programmed in the control box, the temperature thresholds for starting and stopping the system, were 22°C and 18°C, respectively.

Figure 1. Illustration of ICCA system in the warehouse. The warehouse has four centrifugal fans mounted on both sides, although only one is visible in the picture.

Temperature, relative humidity and CO₂ concentration measurement

The temperature, RH, and CO₂ were obtained with a portable detection device, as illustrated in Figure 2. The device was equipped with digital chips (from Scnsirion Inc., Switzerland), which included temperature and RH sensors. Furthermore, a tunable diode laser absorption spectroscopy (TDLAS) technique was utilized to detect CO₂ in the device. The TDLAS was implemented by a distributed feedback laser (Nanoplus GmbH, Gerbrunn, Germany) with a wavelength of 1550 ± 2 nm and an output optical power ≥10 mW. The accuracy of temperature, relative humidity, and CO₂ detection was ± 0.4°C, ±3%, and ± 5 ppm, respectively.

Figure 2. Homemade detector: Stored grain information detector. Detector has two components: detection rod and microcomputer console box. Detection rod has temperature and relative humidity detection module, while CO₂ detection module is housed inside console box.

Here, 16 measurement points were selected within the warehouses, as demonstrated in Figure 3. These points were evenly distributed along the length and width directions of the warehouse. These locations were evenly spread out across the length and width of the warehouse. The chosen sites for testing were positioned at least 1 m away from the wall to ensure that the grain levels measured accurately represented the overall conditions within the warehouse, as grains near the wall experienced rapid fluctuations.

Figure 3. Measurement points in the warehouses. This is a top view of the grain layer.

Grain samples were collected at depths of 0.5 m, 1 m, and 1.5 m. The grain floor was divided into several areas. Sampling was conducted in the corners and the center of each area. The MC of samples was determined based on the oven-drying method recommended by the American Society of Agricultural and Biological Engineers (ASABE), 2017.^[25] The FFA value of samples was determined in accordance with the evaluation guidelines outlined in the National Standard of China (GB T 20,570–2015) for assessing maize storage characteristics. Initially, the maize samples were subjected to grinding and sieving processes. Subsequently, the FFA in the maize was extracted with anhydrous ethanol. The ethanol solution containing the fatty acids was titrated with a standard potassium hydroxide solution (0.01 mol/L). Each sample was subjected to duplicate testing by the same inspector. The average value obtained from the two tests was the final result, with a maximum allowable difference of 2 mg KOH/100 g between the two values.^[26]

Statistical analysis

The experiment aimed to collect samples from identical locations at different time points, but the manual operation involved in the process may have led to minor discrepancies in the sampling locations. Despite efforts to minimize this variation, it could not be entirely eliminated in the study. Vimala et al. conducted a study on a corrugated steel bin containing 300 tons of wheat and observed that standard errors of stored wheat moisture content caused by location variation were 0–0.3% and 0–0.04% in May and July, respectively.^[27] The potential deviation from standards in this study may be deemed acceptable given its large-scale nature. Grain temperature and RH data were collected via the warehouse grain information monitoring system. This data was usually collected automatically every day or twice a day. Any outliers in temperature and RH were removed during the data analysis process. Each stored maize layer had 104 temperature testing points. The average temperature was used as a representative measure of temporal variations.

The stored maize was usually transported into the warehouse in March or April in the studied company. Since the storage period of maize usually lasted for 4 years, the stored maize that was stored for the one, two, three and four years were selected as the studied objects. The ICCA technology were utilized in the four warehouses. One more warehouse containing maize stored for one year without using ICCA was used as a contrast.

The detection depth was determined by the length of the rod. A longer rod would result in greater resistance, making it more susceptible to breakage during detection operations if it exceeded a length of 1.5 m. The temperatures of the stored maize at depths of 3.5 m and 5.0 m were measured using the warehouse information monitoring system.

Samples used for the analysis of the MC and FFA values were taken from the grain layer at a depth of 1 m in the whole warehouses. Due to the massive workload, only one warehouse, specifically the four-year warehouse, was selected for sampling at depths of 0.5 m, 1 m, and 1.5 m to assess the variation in quality across different grain layers. The results of MC and FFA values were statistically analyzed by one-way ANOVA using BONC DSS Statistics(Version 1.0, BONC.2018). Least significant difference and Waller-Duncan test were used to determine significant differences.^[28] The statistical significance was declared at p < .05 post-hoc comparison.

Results

Temperature and RH

The stored grain suffered from high temperatures, typically from July 1st to September 30th in Northeast China. Before the invention of ICCA, conventional methods of controlling stored maize temperature during the summer involved exchanging the hot air within the warehouse headspace with cooler ambient air via window openings at night. As illustrated in Figure 4 (e), without utilizing ICCA, the temperatures of stored maize exhibited fluctuations from July 1st to September 30th. The temperatures experienced a general increase from July 1st to August 25th, followed by a subsequent decrease. The temperatures at D0 and D1 had a narrow difference, not exceeding 5°C. However, temperatures in other layers, such as D2, D3, D4, and D5, had a large gradient of at least 8°C. Temperatures at D0 and D1 surpassed 25°C from July 25th to August 23rd. The temperature at D2 increased rapidly from July 1st to August 25th. On July 26th, it surpassed 20°C, starting from approximately 12°C on July 4th, and then it exceeded 25°C on August 9th.

Figure 4. The temperatures of the stored maize with different storage periods. Y1, Y2, Y3 and Y4 represent the stored maize in the one year, two years, three years, and four years, respectively. The stored maize of Y1, Y2,Y3 and Y4 had been used ICCA. The stored maize of WY was without using ICCA. (a) It should be noted that the term “One Year” refers to maize that has been stored for a period less than one year. (b) “Two Years” refers to the maize had been stored for more than one year but less than two years. (c)Similarly, “Three Years” refers to the storage period were between two and three years. (d) “Four Years” refers to the storage period were between three and four years. D0, D1, D2, D3, D4 and D5 represent the headspace of the warehouse, the depth of 0.5 m, 1.0 m, 1.5 m, 3.5 m and 5.0 m of the grain layer, respectively.

In contrast, the temperatures at D4 and D5 remained below 15°C throughout the experiment. The temperature at D2 peaked at 27°C on August 24th, the most rapid increase among all temperatures. All other temperatures experienced a sharp decline rapidly from August 25th to September 3rd because of the aeration process. Generally, when the ambient temperature was lower than that of the stored maize in late August, aeration was implemented to lower the temperature. The ambient temperature was higher than that of D4 and D5 but lower than the temperatures in the other layers of the same warehouse. Consequently, this increased temperature at D4 and D5.

During the summer, the application of ICCA in the warehouse allowed for control of the high temperatures within the stored grain bulks. Figure 4(a-d) depict the temperatures of the stored maize with different storage periods under the application of ICCA variation during the experiment period from July 1st to September 30th. The temperatures at D0 of Y1 had ranged from about 19°C to 22°C. The differences between temperatures at D0, D1, and D2 of Y1 decreased to less than 2°C over time. The temperatures at D4 and D5 of Y1 increased during experimental period but but they had never exceeded 10°C. In contrast, the stored maize temperatures at D4 of Y2 and Y3 exceeded 10°C on July 14th and August 22nd, respectively. Furthermore, the temperatures at D3 exceeded those at D0, D1 and D2 on around August 30th in the four stored maize warehouses. The temperatures at D3 of Y2 had been over 21°C from August 10th to September 12nd, the temperatures of the other three stored maize in the same depth had never been over 21°C. On August 30th, the temperature at D3 of Y2 was 17°C, indicating that the stored maize all had a temperature higher than 17°C at depths above 1.5 m. For the safety of the stored maize, cooling aeration by using ambient air was applied at night, resulting in a sharp drop in the temperatures of all depths in the warehouse of Y2.

In addition to affecting temperatures, ICCA also has a certain influence on the RH in the stored grain warehouses. As illustrated in Figure 5(e), without utilizing ICCA, the RH at D0 of WY remained consistently high, exceeding 75% for the majority of the time and even reaching 90% at certain points. It dropped sharply on 23rd because of causing by the aeration. The RH of the headspace ranged from 56% to 93%,, with an average value of approximately 80%. The RH in the grain layer varied with the temperatures. The RH at D1 of WY steadily rose rise from about 68% to 91% due to the rise in grain temperature. While the RH at D2 and D4 also experienced an increase, it never surpassed the levels observed at D1.

Figure 5. The RH in the stored maize warehouses with different storage periods under the application of ICCA. The variables Y1, Y2, Y3, Y4, WY, D0, D1, D2, and D3 were defined as depicted in Figure 4.

As depicted in Figure 5(a-d), with the application of ICCA, the RH at D0 in the four warehouses had never exceeded 75%, and it fluctuated mostly between 40% and 70%. The RH at D0 of Y1 experienced a decline from 73% to 48% in the period from July 25th to August 8th, and a similar sudden decrease in RH was observed in the other three warehouses. During the ICCA process, the air with high RH in the headspace was mixed with the air from the bottom of the stored grain layer with lower RH. Consequently, the longer the inner cooling aeration was conducted, the lower the RH in the headspace became. The RH of D1 of Y1 rose from 55% to 75%, then followed by a gradual decline to 68%, displaying a similar fluctuation pattern as D0 and D2 of Y1. It suggests that tair in the headspace could easily affect the air at D1 and D2 in the stored maize layer. Furthermore, the RH of D3 of Y1 had a slow rise from 58% to 67% during the whole period, differing from D0 and D1 of Y1 which had a drop trend after the peak on July 25th. At a depth of 1.5 m, the RH of D3 of Y2 dropped from September 12th, owing to the aeration in Y2. The RH of Y3 and Y4 had a similar change with Y1.

CO₂ concentration in the stored maize warehouses

In the stored grains, the main sources of CO₂ were the metabolism of grain, the reproduction of insects and the growth of molds and fungi.^[29] Compared with the CO₂ produced by the metabolism of stored maize, the reproduction of molds, fungi or insects was considered to be the main cause for the increase in CO₂ concentration in the warehouses. According to Figure 6, the CO₂ concentration of all depths remained relatively consistent, except during the aeration period in warehouse Y2. In Y1, the CO₂ concentration did not exceed 500 ppm and remained constant, indicating the absence of mold or fungal infestation in the stored maize. In contrast, warehouses Y2, Y3, Y4, and WY all experienced a sudden increase in CO2 concentration, likely due to the reproduction of molds or fungi. During the aeration process, the CO₂ concentration rapidly dropped to around 500 ppm. In warehouse Y3, the CO₂ concentration continued to increase rapidly, reaching approximately 6700 ppm until September 7th. The windows on the wall of the warehouse were opened to exchange the air in the headspace with the ambient air, resulting in a decrease in CO₂ concentration in the warehouse from September 7th. In the warehouse of Y4, the CO₂ concentration initially increased steadily until August 8th, then experienced a rapid rise to approximately 3500 ppm. On September 12th, the air exchange through opening the windows led to a decrease in CO₂ concentration. Subsequently, the rise of CO₂ concentration may be attributed to ongoing mold and fungal reproduction, as the reproductive process had not ceased. Without ICCA, the CO₂ concentration steadily rose from approximately from about 1200 to 6500 before the aeration. As the temperature and RH increased, the CO₂ concentration rose as a result. Notably, there is a positive correlation among temperature, RH and CO2 concentration.

Figure 6. The CO₂ concentration in the stored maize warehouses with different storage periods under the application of ICCA. The variables Y1, Y2, Y3, Y4, WY, D0, D1, D2, and D3 were defined as depicted in Figure 4.

Moisture content of the stored maize

The initiate MC of the stored maize varied between 13.6% and 14.2% (w.b.). The stored maize would go through the process of losing moisture over time. According to Table 1, the average MC at D1 was about 13.1%, 12.8%, 12.3%, 12.1% and 13.1% (w.b.) in Y1, Y2, Y3, Y4 and WY, respectively. This indicates that, at the same depth, the stored maize with longer storage periods would have lower MC. In Y4, the average MC at D1, D2 and D3 were 11.7%, 12.1% and 12.2%, respectively. This suggests that the stored maize at the lower depth would have a lower MC in the same warehouse. It could be attributed to the increased susceptibility of maize stored at lower depths to environmental factors. During the experiment period, the MC in Y1 decreased by approximately 0.7%. However, the MC in the other three warehouses experienced little or no moisture loss at the same depth. It implies that the MC would prefer to have little loss while the MC below 12.6% for the stored maize at depths below 1 m in the stored maize bulk.

Table 1. The moisture content of the stored maize with different storage periods under the application of ICCA.

	Moisture content of the stored maize (% w.b.)
	Y1	Y2	Y3	Y4			WY
Date	1.0 m (D2)	1.0 m (D2)	1.0 m (D2)	0.5 m (D1)	1.0 m (D2)	1.5 m (D3)	1.0 m (D2)
7/04	13.35 ± 0.09a	12.73 ± 0.17 cd	12.47 ± 0.13a	11.92 ± 0.18b	11.85 ± 0.16c	12.29 ± 0.13ab	13.43 ± 0.13a
7/11	13.54 ± 0.2a	13.19 ± 0.09a	12.27 ± 0.19ab	12.21 ± 0.16a	12.09 ± 0.22bc	12.04 ± 0.15c	13.51 ± 0.17a
7/18	13.52 ± 0.14a	13.15 ± 0.22a	12.43 ± 0.11a	12.34 ± 0.17a	12.15 ± 0.17b	12.18 ± 0.19abc	13.56 ± 0.09a
7/25	12.97 ± 0.17b	12.81 ± 0.18bcd	12.18 ± 0.16b	11.53 ± 0.18c	12.22 ± 0.21b	12.07 ± 0.2bc	13.15 ± 0.11ab
8/01	13.35 ± 0.13a	12.79 ± 0.07 cd	12.29 ± 0.07ab	11.86 ± 0.16b	12.42 ± 0.17b	12.14 ± 0.19abc	13.37 ± 0.16a
8/08	13.44 ± 0.19a	13.02 ± 0.12ab	11.92 ± 0.21b	11.89 ± 0.18b	12.79 ± 0.21a	12.38 ± 0.17a	13.03 ± 0.19ab
8/15	12.81 ± 0.09bc	12.28 ± 0.16e	11.73 ± 0.16b	11.59 ± 0.15c	12.36 ± 0.15b	12.1 ± 0.18bc	12.82 ± 0.15bc
8/24	12.7 ± 0.2c	12.62 ± 0.1d	12.19 ± 0.12b	11.52 ± 0.12c	12.25 ± 0.18b	12.15 ± 0.19abc	12.74 ± 0.08c
9/05	12.69 ± 0.14c	12.76 ± 0.1 cd	11.45 ± 0.16c	11.03 ± 0.15d	11.54 ± 0.19d	11.96 ± 0.1c	12.75 ± 0.14c
9/24	12.63 ± 0.17c	12.87 ± 0.2bc	12.37 ± 0.14ab	10.89 ± 0.17d	11.84 ± 0.13c	12.12 ± 0.16bc	12.67 ± 0.12c

All the data were expressed as mean ± standard deviations. Each data were determined from 16 samples with triplicates. The difference just shows the statistics in the same column. Means with the different letters in a column were different significantly (p < .05) between two sampling datasets. The variables Y1, Y2, Y3, Y4, WY, D0, D1, D2, and D3 were defined as depicted in Figure 4.

FFA values of the stored maize

The initiate FFA values of the stored maize ranged from 31 to 34 mg KOH/100 g. The quality of the stored maize gradually deteriorated as the storage time increased, resulting in an increase in FFA levels. Throughout the entire trial period, it was observed that the longer the maize was stored, the higher the FFA it would be, as shown in Table 2. The average FFA in Y1, Y2, Y3, Y4 and WY were 36.0, 37.6, 50.4, 56.4 and 36.3 mg KOH/100 g, respectively. In Y4, the FFA at D1 had a rise from 56.39 to 70.08 until August 1st, then it decreased to 58.69. FFA at D2, D3 never reached 70. In Y4, the maize at the lower depth would have a lower FFA before August 1st. For instance, on July 4th, the FFA at D1, D2 and D3 were 56.39, 5.67 and 51.49, respectively. The stored maize at D0 was more susceptible to the ambient conditions than D1 and D3, and it preferred to suffer a loss in MC and an increase in FFA.

Table 2. The FFA values of the stored maize with different storage periods under the application of ICCA.

	FFA values of the stored maize (mg KOH/100g)
	Y1	Y2	Y3	Y4			WY
Date	1.0 m (D2)	1.0 m (D2)	1.0 m (D2)	0.5 m (D1)	1.0 m (D2)	1.5 m (D3)	1.0 m (D2)
7/04	33.66 ± 0.72c	36.51 ± 0.87d	46.91 ± 0.79d	56.39 ± 0.98d	55.67 ± 0.96cd	51.49 ± 1.24e	33.81 ± 1.13d
7/11	34.04 ± 1.3c	36.73 ± 0.82cd	47.1 ± 1.1d	56.79 ± 1.08d	56.59 ± 1.29bc	51.83 ± 0.92e	34.26 ± 0.78d
7/18	34.38 ± 1.04c	36.79 ± 1.18cd	44.79 ± 0.77e	59.41 ± 1.08b	54.37 ± 0.82e	50.97 ± 0.53e	34.9 ± 0.92cd
7/25	34.94 ± 0.78c	36.92 ± 0.97bcd	49.14 ± 1.35c	63.44 ± 0.52b	58.89 ± 1.26a	52.13 ± 0.52e	35.46 ± 1.06c
8/01	36.42 ± 1.26b	37.63 ± 1.38bcd	52.65 ± 1.29ab	70.08 ± 0.8a	55.26 ± 0.84cde	51.82 ± 0.77e	36.52 ± 0.91c
8/08	37.08 ± 1.15b	38.3 ± 0.58bc	51.54 ± 0.97b	63.14 ± 1.25b	54.89 ± 0.92de	53.68 ± 0.61d	37.55 ± 1.3b
8/15	36.41 ± 1.19b	37.68 ± 0.68bcd	54.06 ± 1.16a	63.32 ± 0.65b	55.47 ± 1.06cd	54.08 ± 0.86d	37.57 ± 1.07b
8/24	36.95 ± 0.9b	38.89 ± 1.01a	51.82 ± 0.77b	61.21 ± 0.66c	59.41 ± 0.61a	56.66 ± 0.71c	37.95 ± 0.86ab
9/05	37.23 ± 0.78ab	38.48 ± 1.14ab	52.62 ± 1.26ab	54.47 ± 0.78e	56.32 ± 1.21bcd	57.1 ± 0.89b	37.23 ± 0.91b
9/24	37.66 ± 0.45a	38.99 ± 0.97a	53.19 ± 1.31ab	58.69 ± 1.29c	57.57 ± 0.92b	58.03 ± 1.09a	38.16 ± 1.23a

All the data were expressed as mean ± standard deviations. Each data were determined from 16 samples with triplicates. The difference just shows the statistics in the same column. Means with the different letters in a column were different significantly (p < .05) between two sampling datasets.The variables Y1, Y2, Y3, Y4, WY, D0, D1, D2, and D3 were defined as depicted in Figure 4.

Discussion

Temperature and RH

Without cooling aeration in summer, the temperature of stored maize could exceed 25°C and remain high for up to 40 days and 20 days at the depth of 0.5 m and 2 m, respectively, as shown in Figure 4. The maximum recorded temperature surpassed 26°C. Concurrently, the RH in the headspace of the warehouse remained above 70% for a period over 70 days. It has been established that there exists a direct correlation between the temperature and RH.^[30] As temperature increase, the stored grain would lose moisture to the air, resulting in an increase in the RH. In most cereal grains, while the temperature increases 10°C, the RH would rise about 3%.^[31] The increasing temperature and RH would not only contribute to the growth of mold, but also lead to considerable nutrient losses of grain.^[32] With the application of ICCA in the summer, the whole stored maize in this study did not exceed 23°C. The RH in the headspace was below 65% for the majority of the time. According to ASABE (2008) the thermal diffusivity of corn bulk is 10.22 × 10 − 8 m²/s.^[33] Due to the low thermal diffusivity of bulk grain, temperature gradients tend to formed. The temperature of the grain decreases as its depth increases. The temperatures at the depths in all warehouses below 3.5 meters would not exceed 15°C, as shown in Figure 4. Consequently, the shallow layer of maize above 3.5 m would be more susceptible to deterioration or mold. Without ICCA, the RH of the headspace reached 92%, as shown in Figure 4(e). As a result of the convection currents, the moisture of the stored grain could move up into the headspace.^[34] During the implementation of ICCA, the air in the headspace was cooled down, accompanied by changes in RH. Temperature gradients also created RH gradients in the stored grain. It had been proved that the 2.2 kW aeration fan could not easily blow through the grain layer and blow off the humid air in the headspace.^[16] The 1.1 kW fans were used for ICCA, which limited the ability of the conventional air currents to impact the air within the bulk of the grain. Consequently, the RH at a depth of 1.5 m was less influenced compared to the RH at depths of 1.0 m and 0.5 m.

CO₂ concentration in the stored maize warehouses

In the stored grain bulk, all living organisms produce CO₂, which is closely related to their metabolic rate.^[35] Monitoring the changes in CO₂ concentration in the grain pile can efficiently indicate the grain status, fungal propagation, and insect activity.^[36] The convection current forced by the ICCA ensures that CO2 is evenly distributed in most of the tested grain piles, as shown in Figure 7. Throughout the experiment, the CO₂ concentration in Y1 remained below 500 ppm, indicating that there was no risk of fungal and insect reproduction in the stored maize bulk. However, the CO₂ concentration in the other three warehouses rapidly increased to above 3000 ppm. There were no insects present in any of the warehouses except for Y3 where some insects were detected. The lower growing temperature threshold for many stored grain pests is approximately 18°C, while the optimal temperature and RH range for most stored-product insect development is approximately 25–35°C and 70–80%, respectively.^[37,38] Most microorganisms in the stored grain cannot multiply when RH is below 65%.^[39] The mold could be observed in the stored grain when the temperature is between 20°C and 40°C with the RH above 70%.^[13] Hotspots were observed in the warehouses of Y2, Y3 and Y4. It had been proved that there was a strong positive correlation between the CO₂ concentration and the temperatures in all five stored wheat layers of the hotspot module.^[39] The hotspot is typically initiated by the proliferation of microorganisms.^[40] The condensation, leakage of snow or rain, or moisture migration through convection currents would lead to the increase of the stored grain MC. The increasing MC promotes the growth of microorganisms, which in turn contribute to the development of the fungi-induced hotspot.^[41] The hotspots in the three warehouses can be attributed to the following reasons: 1) the warehouse of Y2, constructed with steel plates, had a lower thermal isolation than the other three warehouses made of concrete; 2) the warehouses of Y3 and Y4 had not been fully aerated in the winter resulting in the moisture condensing in a certain area of grain bulk. Field tests of monitoring spoilage with CO₂ sensors had been conducted in a 12,500-t cylindrical steel tank with stored corn and the tests indicated that the CO₂ sensors were effective in detecting the occurrence of spoilage.^[39] During hotspots developed in the sealed silo, the headspace CO₂ concentration had increased to 5000 ppm, the maximum of the CO₂ sensor.^[39] The studies revealed that in the headspace the CO₂ concentration above the 5000 ppm indicated the presence of biological activity and the development of hotspots in the stored grain.^[42]

MC of the Stored maize

The storage life of grains is usually determined by two mainly physical factors: temperature and MC.^[43] The survival and reproduction of biological agents in grains are directly influenced by the temperature and MC.^[15] The safe MC for storing corn for up to a year under Canadian Prairie climatic conditions is 15.5%. However, for long-term storage (>1 year), the MC should be 1–3% points less than the value.^[8] The studied stored maize MC was below 14%, suitable for long-term storage. In a corrugated steel bin with 300 t of stored wheat, the surface MC increased by an average of 3.3 ± 0.4% from February to July, while the average MC change inside the grain bulk during the same period was 0.2%.^[28] Chang et al. (1994) reported a decrease of 2–2.5% in MC of wheat near the surface and less than 0.2% MC change in all other layers during the summer inside a 6.6 m diameter corrugated steel bin with aeration.^[44] The results of this study also indicate that the MC of stored maize at a depth of 1 m change was 0.72%, 0.14%, 0.1%, 0.01%, and 0.67% in Y1, Y2, Y3, Y4, and WY, respectively. The MC at a depth of 0.5, 1.0, and 1.5 m in Y4 decreased by 1.03%, 0.01%, and 0.17%, respectively. Cooling aeration affected the stored maize near the surface more than the deep grain layers. The MC of the stored maize at a depth of 1 m with ICCA and without ICCA in the same storage period did not have a significant difference. This suggests that the difference in temperature and RH may not have caused a significant change in MC during the short experimental period.

FFA values of the stored maize

During grain storage, there is a process of partial hydrolysis of the glycerides within the grain, which leads to an increase in FFA values.^[45] The increasing FFA means quality deterioration happened in the stored grain. Therefore, FFA is commonly used as an indicator of stored grain quality changes during storage period.^[46] Studies have shown that FFA values can increase with storage time and temperature, and are positively correlated with the accumulation of temperature.^[24] According to the research of Gras, W.P. et al., after 9 months storage, the FFA values of stored maize sited at the bottom, middle and top of unsealed and sealed bags were 46.6, 56.4, 43.3, 18.6, 42.8, and 25.6 mg KOH/100 g, respectively.^[47] The results of current study showed that the FFA of stored maize at D2 in Y1, Y2, Y3, Y4 and WY reached to 37.66, 38.99, 53.19, 58.69 and 38.16 mg KOH/100 g, respectively. The FFA values in Y1 and WY showed a similar change. It suggested that the difference in the temperature at D1 could not cause the FFA to have a significant change during the short experiment period. In contrast, the FFA of stored maize at D1 in Y4 had reach to 70.08 from 56.39 mg KOH/100 g, then decreased to 58.69 mg KOH/100 g. The FFA in stored maize was mainly consisted of oleic acid and linoleic acid. As a result of hydrolysis and oxidation activities inner stored grain, the linoleic acid would decrease while the oleic acid would increase over time.^[48] Once the FFA reached a certain content, it would decrease due to the changes in linoleic acid and oleic acid.

Conclusion

Several conclusions can be drawn based on the present study’s findings. Without utilizing ICCA during summer, the average temperature of the stored maize at a depth of 0–1.5 m in a large commercial warehouse could be in the range of 20–26°C, with the RH of the headspace exceeding 75%. The conditions are suitable for the reproduction of insects and fungal, posing a significant risk to the quality and safety of the grains. With the application of ICCA during the summer, the average temperature of stored grain could be effectively controlled within 22°C, with the RH remaining below 75% throughout the whole period. The application of ICCA demonstrated a significant ability to regulate temperature and RH, thereby reducing the risk of grain deterioration. When the stored maize bulk was not adequately ventilated in the winter, the temperatures of the stored maize at a depth of 1.5–3.5 m experienced rapid increases during the summer, regardless of the presence of ICCA. During the entire experimental period, the MC of the stored maize at a depth of 1.0 m in Y1 lost approximately 0.5% (w.b.), while the MC of maize stored at the same depth for more than one year did not show significant losses. Moreover, the MC of the stored maize at the depth of 1 m with ICCA and without ICCA in the same storage period did not show a significant difference. The FFA values indicate that stored maize experienced degradation during the summer season with the FFA value increasing by 1–5 mg KOH/100 g in the stored maize at a depth of 1 m in the stored maize bulk. The FFA of the stored maize at a depth of 1 m with ICCA and without ICCA in the same storage period exhibited a similar change with an increase of 5–6 mg KOH/100 g in the short experiment period. As the temperature of stored maize at a depth of 1.5 m often exceeds that at the depth of 1 m in early August, it is recommended to lower the upper threshold temperature to 20°C or below in order to cool down the deeper layers of stored maize. The CO₂ sensor can be used to monitor the microbial activities in the stored grain bulk. It can be integrated into the ICCA system to enhance control accuracy and energy efficiency.

Acknowledgments

Thanks to the grain reserve companies for providing the testing grain and laboratory consumables in this research.

Disclosure statement

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

Data availability statement

Data sharing is not applicable.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10942912.2023.2285230

Additional information

Funding

This research was supported by the Youth Science Fund for National Natural Science Foundation of China, grant number 32102034; National key R&D Plan, grant number 2017YFD0401003-3.