Sorption Drying of Wheat Seeds Using Kieserite as a Solid Desiccant

Abstract

The moisture content (MC) of wheat seeds must be reduced before storage using appropriate dehydration processes. Desiccant drying is a promising alternative to conventional drying methods because it improves seed quality while providing overall energy efficiency. This study explores the sorption drying of wheat seeds using granulated kieserite MgSO₄·H₂O as a solid desiccant, which has a high water capacity and is regenerated at low temperatures <100 °C. Desiccant characterization was conducted using SEM-EDS, XRD, DSC-TG, and particle size analysis. Wheat seeds mixed directly with kieserite in various mass ratios were dried under uniform stirring and controlled temperature conditions. A 240-min drying time was required to reduce the initial MC of wheat from 21.5% to 15.1% at a desiccant-to-grain ratio of 1:1. After 360 min, a final MC of 14.4% was achieved. The germination energy and seed capacity after sorption drying were 91 ± 1% and 97 ± 2%, respectively. Due to the available water capacity of kieserite, several batches of seeds can be dried without intermediate desiccant regeneration. This study is useful for developing low-cost, non-thermal, and sustainable drying technology for various agricultural products.

1. Introduction

Wheat (Triticum aestivum L.) is one of the main cereal crops produced and consumed globally and is critical to global food security [1]. With a growing world population, there is an urgent need for quality wheat seeds to protect this strategic food crop and sustainable agriculture [2]. However, it is estimated that approximately 30% [3] to 70% [4] of total food grains produced are lost during post-harvest processes, including drying, transport, packaging, and storage. Developing effective ways to reduce quantitative grain losses while maintaining seed viability based on technological advances is a significant contemporary challenge. Sorption technologies are a fundamental process in sustainable industrial development as they ensure both product quality and the efficiency of technology processes. Recent advancements in sorption drying, specifically the use of solid desiccants, are considered a crucial direction for further applications in agriculture [5,6,7,8,9,10].

Moisture content (MC) is a key factor responsible for seed quality in storage [11,12,13,14]. The rates of vigor and viability loss for most species classified as orthodox (i.e., seeds tolerant to desiccation, including wheat [15]) depend primarily on the MC of the seed and the temperature [11,12,13,14,15,16]. At a constant temperature, there is a negative logarithmic relationship between seed MC and longevity [16]. After harvest and before storage, it is necessary to evaluate whether the MC of the seed corresponds to the required value.

Critical grain moisture is a basic concept in seed storage theory. For major grain crops, the critical MC is 14.5–15.5%, when the water inside the grain is firmly bound by colloidal substances and microcapillaries [17]. When moisture increases, free or droplet-liquid water appears in seed structures, intensifying biochemical and microbiological processes. The presence of free moisture is associated with a high risk of self-heating, molding, active microorganism development, and grain spoilage. Wang et al. reported [12] that the MC of wheat seed should be kept below 16% to avoid severe mildew. The International Codex Standard for the maximum MC for the secure storage of wheat with a good shelf life is set at 14.5% [18]. Russian standards set the MC of wheat seeds—depending on the growing zone—at a level of 14.0–16.0% [19]. Grain with an MC below the critical level can lose its nutritional value and viability [17,20].

Drying is the process of removing MC to a specified level and is a fundamental method for preserving grain quality. Drying reduces water activity and the rate of enzymatic processes responsible for grain deterioration [17,20]. Correct drying processes stop the growth of microorganisms, reduce the risk of mycotoxin occurrence, and promote post-harvest ripening. As a result, the final product has a longer shelf life and a lower weight and volume, making it easier to transport and store [11].

Well-known drying methods are based on two basic principles. Thermal drying methods remove moisture via a change in the state of aggregation by turning it into steam. Mechanical dehydration, acoustic drying, and sorption drying remove moisture without changing the aggregate state in the form of a liquid [21]. Sunlight has historically been used to dry seeds, but this method has many disadvantages, including a lack of control over the drying process, long drying periods, and contamination. Convection drying using industrial hot air systems is a traditional and widespread [17,20,21] but high-energy-consumption process. Drying grain takes about 60% of the total energy required for this process [22]. The heat-drying process changes the physiological qualities of the seeds. The maximum temperatures that can be used for seeds are lower than those for commercial grains to maintain viability; as a result, a significant decrease in dryer productivity is observed. High energy consumption and seed quality loss are significant disadvantages of heat treatment for grain drying.

Energy-saving sorption technology for drying agricultural crops is based on the use of moisture-adsorbing substances that can reduce the MC in contact with the grain mass without using heat while maintaining or improving quality indicators. The theoretical basis of sorption drying (also called desiccant drying or adsorption drying) is the transfer of moisture between the material being dried and the adsorbent with which it is in contact with due to a concentration gradient. The most common scheme for contact sorption drying involves mixing a solid sorbent with the material to be dried, followed by the separation of these two media after achieving the desired moisture content. The solid sorbent is then regenerated and returned to the process [21]. The main characteristics of the best sorbent are the high sorption rate, required water capacity, ease of separation from the dried material, the ability to effectively regenerate, and stable cyclic characteristics

Dry grain [23], granulated silica gel [24,25,26,27], clay minerals [28,29,30], and zeolite beads [5,6,31] were considered as desiccant agents for grain crops. Some traditionally used desiccants do not have the required granule strength; have limited moisture capacity, such as silica gel; or require regeneration at high temperatures, such as zeolites. Drying times ranged from 12 h to several days. The high dispersion of some reagents caused them to be carried away and adhere to the surface of the grain, which, combined with the duration, negatively affected the efficiency of the process.

One of the promising desiccant sorbents is magnesium sulfate, which is characterized by high H₂O capacity values (7 moles per 1 mole of anhydrous salt), low regeneration temperatures <100 °C, high moisture adsorption rate, and neutrality [32]. As a result of the exothermic effect of the reaction of crystalline hydrate formation, the process of internal moisture transfer can be activated in a single grain and accelerate the process of contact drying as a whole. Additionally, magnesium sulfate is a cheap, nontoxic, and well-proven fertilizer widely used in agriculture [33,34]. These advantages have determined the prospects for the use of anhydrous magnesium sulfate for grain drying [35] and as an active component of a microspherical composite desiccant for bulk materials [36,37].

Magnesium sulfate hydrates are considered to be one of the most promising materials for long-term heat storage [38,39]. The physicochemical properties of MgSO₄·nH₂O have been extensively discussed in the literature, including decomposition reactions [40,41,42,43,44,45], phase equilibria in the MgSO₄–H₂O system [40,45], stability as a function of temperature and relative humidity [46], and water transport during dehydration [45]. The use of crystalline hydrates as desiccants in contact drying is similar to the direction of heat accumulation as it shares similar criteria for efficiency and applicability, such as energy and moisture capacity, toxicity, fire safety, availability, and price [17,20,21].

One of the most stable natural forms of magnesium sulfate is kieserite, which has the chemical formula MgSO₄·H₂O [40,41,42]. When water is adsorbed by kieserite, it forms several phases of crystalline hydrates with an increasing number of water molecules, as shown in the reaction MgSO₄·H₂O + (n − 1) H₂O → MgSO₄·nH₂O (n = 2 ÷ 6) [42]. Contact sorption drying of wheat occurs when kieserite interacts with moisture localized on the surface of wheat grains.

This paper introduces contact sorption drying of wheat seeds with granulated kieserite, MgSO₄·H₂O crystalline hydrate, as a method of reducing seed moisture to a safe level without compromising their germination. Due to its chemical composition, kieserite does not require preliminary regeneration before drying the seeds and maintains a stable sorption capacity over multiple drying cycles. In addition, its granular form prevents the seeds from sticking to the surface, making them easier to separate after drying. This study can help in the development of non-thermal and sustainable drying technology for agricultural products while ensuring their quality.

2. Materials and Methods

2.1. Materials

The wheat seeds (Triticum aestivum L.) used in this study belonged to the Novosibirskaya 41-grade variety produced in the East Siberian region (EPF “Mikhailovskoye”, FRC KSC SB RAS, Krasnoyarsk, Russian Federation) and were harvested in 2022.

Granulated-grade kieserite, a commercial agrochemical magnesium sulfate, was used as a solid desiccant. This product was manufactured by the South Ural Magnesium Compounds Plant (UUZMS, Orenburg Region, Kuvandyk, Russian Federation). The granulated kieserite was produced in accordance with Technical Specifications TU 20.13.41-001-23877036-2017 [47]. The product underwent a double-drying process, allowing it to be stored for up to 5 years, and complied with both Russian and European quality standards.

The agrochemical magnesium sulfate was formulated as a fertilizer containing mineral magnesium sulfur intended to increase fertility and crop yield in all types of soil [48]. According to the product certificate [47], the granulated kieserite magnesium sulfate contained at least 85 wt% of MgSO₄. In terms of dry matter, the total mass fraction of impurity sulfates (calcium, sodium, and iron) did not exceed 1%; the remaining volume of water-insoluble salts did not exceed 0.6%. In terms of granulometric composition, the desiccant was 99% represented by the fraction −1.0 + 0.5 mm.

2.2. Moisture Content Determination

To determine the moisture content of the seeds, the hot-air drying method was used, which was carried out in accordance with GOST 13586.5-2015 [49] at a temperature of 130 ± 2 °C, which corresponded to the international standard ISO 712:2009 [50]. MC was calculated on a wet basis (% wb). The statistical processing of the data obtained was carried out in accordance with GOST 13586.5-2015 [49], which determines the reproducibility of experimental data.

MC in desiccants was determined based on experimental data on weight loss after heat treatment under conditions of complete dehydration at 400 ± 2 °C.

2.3. Desiccant Characterization Methods

The physicochemical characteristics of the desiccant, including granulometric composition, the content of chemical components and individual crystalline phases, and the number of crystallization water molecules, were determined using sieve analysis, scanning electron microscopy with energy dispersive spectroscopy (SEM-EDS), X-ray diffraction (XRD), and thermal (DSC-TG) analyzes.

The granulometric composition was determined using a VP-S/220 vibration drive (Vibrotechnic, Saint Petersburg, Russia); the sieving duration was 30 min at a frequency of 70 Hz.

A TM-4000 scanning electron microscope (High Technologies Corporation, Hitachi, Tokyo, Japan) with a Bruker XFlash 430H energy-dispersive X-ray spectrometer (Bruker Corporation, Billerica, MA, USA) and a Quantax 70 microanalysis system was used for SEM-EDS studies, which were carried out on powder samples at a magnification of ×50–1000 with an accelerating voltage of 20 kV.

XRD analysis was performed using an X’Pert Pro MPD powder diffractometer (PAN-alytical, Almelo, The Netherlands) with a PIXcel solid-state detector using Cu Kα radiation. To prevent moisture exchange with the atmosphere during the recording process, samples with a particle size of less than 160 µm were placed in a cuvette and covered with a polymer film, using vacuum grease as insulation. X-ray patterns were scanned at room temperature in the range of 5° ≤ 2θ ≤ 70°. The phases present in the sample were identified using the ICDD PDF database [51]. The content of crystalline phases was determined by the derivative difference minimization (DDM) method, with normalization of the total amount of crystalline phases using the diffraction pattern of an external standard [52]. A sample of magnesium sulfate reagent, the phase composition of which corresponded to 98.5% of MgSO₄·7H₂O and 1.5% MgSO₄·6H₂O, was used as an external standard.

Simultaneous thermal analysis (DSC–TG) was carried out on a Jupiter STA 449C setup with an Aëolos QMS 403C mass spectrometer (Netzsch, Selb, Germany). The measurements were carried out with a sample of 11 ± 1 mg in a dynamic gas mixture (20% O2 + 80% Ar; total flow 50 cm³/min) with simultaneous recording of the change in mass and heat flow at a linear heating rate of 2.5 °C/min in the range of 40–450 °C. The samples were crushed to a particle size of less than 160 µm and, before heating, subjected to isothermal holding in a gas flow for 5 min at a temperature of 40 °C. The DSC sensor was heat flow calibrated by measuring the heat capacity of the sapphire disc according to the method DIN 51007:1994-06 [53]. Primary thermoanalytical data were processed using the licensed NETZSCH Proteus software package (Ver. 4.8.4).

2.4. Sorption Drying Experiments

2.4.1. Drying Setup

In this study, an automatic mixing unit was constructed in which wheat seeds were subjected to different exposure times in a mixture with a desiccant under steady-state conditions of uniform mixing and controlled temperature (Figure 1). The stirring frequency was 2 rpm. The temperature was measured using a resistance thermometer; the temperature sensing element was located in the center of the container with a mixture of wheat seeds and desiccant (Figure 1b).

2.4.2. One-Stage Sorption Drying

Contact drying experiments were conducted as follows. A batch of wheat seeds with a certain initial MC was placed in a closed container with some desiccant. The desiccant/grain mass ratios were 1:1, 1:2, and 1:4. The container was placed in an automatic mixing unit (Figure 1), where it was subjected to contact drying for a present time of 60, 150, 240, and 360 min. At the end of drying, the mixture was separated using a sieve with a mesh size of 2 mm and then the grain and desiccant MC were determined.

2.4.3. Three-Stage Sorption Drying

To determine the possibility of reusing the desiccant until it was completely saturated with moisture without intermediate regeneration, sequential contact drying of three batches of wheat seed was conducted. The three-stage sorption drying process scheme is shown in Figure 2. The desiccant/grain mass ratios were 1:1 and 1:2. The contact drying time for each wheat batch was 360 min.

2.4.4. Drying Kinetics

The loss of moisture in grains during drying processes is dependent on time and can be represented using the moisture ratio (MR) profile. This calculation requires the initial moisture content of the grains and the moisture content at a specific point during the drying process. Typically, the moisture content of grains is measured on a wet basis, which considers the mass of water per unit mass of the wet grain [10,13].

The drying curves for the wheat seeds were plotted using Equation (1) [17,20]:

$$\mathrm{M}\mathrm{R}=\frac{\mathrm{M}{\mathrm{C}}_{\mathrm{t}}}{\mathrm{M}{\mathrm{C}}_0}$$

(1)

where $\mathrm{M}\mathrm{R}$ is the moisture ratio (dimensionless) and $\mathrm{M}{\mathrm{C}}_0$ and $\mathrm{M}{\mathrm{C}}_{\mathrm{t}}$ are the initial moisture content and that at drying time t, respectively; the calculation was made on a wet basis (% wb).

A reproducibility assessment of the experiential data and results was carried out with the use of the relative error of determination, which did not exceed 10%.

2.5. Seed Germination Test

After sorption drying, the germination of seeds was determined to assess the effect of the desiccant on their quality. The determination was carried out using the filter paper method at a temperature of 20 °C in four parallel experiments with 100 seeds in accordance with the State Standard [54] and International Seed Testing Rules [55]. The germination energy (GE) and seed germination capacity (GC) were evaluated on the 4th and 7th days of the experiment, respectively, after planting the sample. Wheat seeds dried in a thin layer in air at 25 °C were used as a control. The statistical processing of the data obtained was carried out in accordance with GOST 12038-84 [54], which determines the reproducibility of experimental data.

3. Results and Discussion

3.1. Desiccant Characterization

Drying is an effective post-harvest treatment that reduces seed MC. The effectiveness of the contact drying process depends directly on the moisture-adsorbing properties of the desiccant used. Anhydrous magnesium sulfate MgSO₄ is commonly used as a desiccant due to its high dynamic sorption capacity, stability, and inertness [32]. Magnesium sulfate forms compounds with the composition MgSO₄·nH₂O, where n ranges from 1 to 11 [56]. Phase equilibria in the MgSO₄–H₂O system have regions of crystalline hydrate stability depending on relative humidity and temperature [42,44]. The predominant and most stable natural magnesium sulfates are epsomite MgSO₄∙7H₂O, hexahydrate MgSO₄ 6H₂O, and kieserite MgSO₄∙H₂O [40,41,42].

The thermal decomposition process of magnesium sulfate crystalline hydrates was studied in detail in [42,43,44]. The first dehydration reaction, MgSO₄·7H₂O → MgSO₄·6H₂O + H₂O, occurred at a low temperature (<50 °C) and resulted in the loss of one water molecule. The largest number of water molecules was dehydrated during the second reaction MgSO₄·6H₂O → MgSO₄·0.2H₂O + 5.8H₂O, which took place between approximately 60 °C and 200 °C (T_max = 150 °C). The last 0.2 water molecule was dehydrated at a high temperature (~275 °C) [42]. The weight loss values for the complete decomposition of MgSO₄∙7H₂O and MgSO₄∙6H₂O crystalline hydrates were 51.16 and 47.31 wt%, respectively.

Magnesium sulfate is approved by the Russian Ministry of Agriculture for use as an agrochemical substance for agricultural crops [48]. It provides plants with the necessary nutrients, sulfur, and magnesium and supports cereal crops during wintering. Magnesium, which is part of chlorophyll, is involved in the solar energy accumulation by plants during photosynthesis, accelerates seed ripening, and helps improve the quality of the crop [33,34]. Therefore, high values of H₂O capacity, mild regeneration conditions, and beneficial effects on plants facilitate the use of kieserite as a solid desiccant for the sorption drying of wheat seeds.

According to SEM-EDS data, the desiccant is represented by granules up to 1 mm in size (Figure 3a) of irregular shape, with an unevenly dense surface (Figure 3b). Fractures of the granules revealed an internal cellular structure formed by intergrown lamellar hexagonal crystals of magnesium sulfate up to 3 μm thick (Figure 3c,d). No foreign matter was found (Figure 3e,f). Of the impurity elements specified in the product certificate, a single ingrown crystal of calcium sulfate up to 50 µm in size was found in the entire sample (Figure 4).

An XRD pattern of the initial desiccant sample is shown in Figure 5. It is known [57] that blurred lines in X-ray diffraction patterns are characteristic of amorphous substances, defective crystals, and small crystallites. In the case of the kieserite sample under study, the wide peaks in the XRD pattern (Figure 5) were associated with distortions and defects of small magnesium sulfate crystallites forming a cellular structure (Figure 3c,d and Figure 4). The crystal structure of the initial desiccant is not identified in the ICDD PDF database [51].

The initial desiccant dehydration process occurred during heating from 40 to 400 °C (Figure 6,

Table 1. Mass loss (∆m) and water capacity in the temperature range of 40–450 °C of the desiccant.

Parameter	Temperature Range (°C)
	40–100	40–150	40–200	40–250	40–300	40–350	40–400	40–450
∆m (wt%)	0.32	1.96	2.79	3.43	5.69	14.30	14.56	14.44
Water capacity (mg/g) *	4	23	33	40	67	167	170	169

* per gram of anhydrous substance.

) and was accompanied by a total mass loss of 14.56 wt%. The achieved state corresponded to the gross formula of the crystalline hydrate MgSO₄∙1.14H₂O and was of similar composition to the kieserite (MgSO₄∙H₂O).

At temperatures of 400–450 °C, the desiccant was represented by the anhydrous salt MgSO₄. The multi-stage nature of the observed process was indicated by the multimodal nature of the complex endothermic peak on the DSC curve and symbatic DTG curve. The main component (contribution of approximately 80%) of the complex DSC peak with the main maximum at 327 °C was observed in the temperature range of 250–400 °C and corresponded to the most strongly bound crystallization water in MgSO₄∙1.14H₂O. The presence of minor components at temperatures of 106, 135, and 185 °C of the complex DSC peak in the temperature range of 40–250 °C with a mass loss of 3.43 wt% (weakly bound crystallization water) was due to the ability of MgSO₄ to form a series of crystalline hydrates MgSO₄∙nH₂O, where n ranged from 1 to 11 [40]. The temperature at which the last water molecule began to separate was 143.2 °C.

The number of H₂O molecules in the crystalline hydrate, determined after calcination of the desiccant to a constant mass at 400 °C (∆m = 15.88 wt%), was 1.26 (MgSO₄·1.26H₂O), which was quite consistent with the DSC-TG data. The smaller mass loss recorded according to DSC-TG data could have occurred due to evaporation during particle grinding and processing in a gas flow before analysis.

3.2. One-Stage Sorption Drying

MC plays an important role in the storage and quality of seeds [12,13,14,15,16]. Depending on the climatic conditions in Russia, 40–60% of the gross grain harvest needs drying [58]. With correctly selected drying modes, seed germination is completely preserved. In this study, the initial moisture content $\mathrm{M}{\mathrm{C}}_0$ of wheat was 21.5 wt%. The MC of wheat and desiccant at a certain drying time $\mathrm{M}{\mathrm{C}}_{\mathrm{t}}$ is given in

Table 2. Moisture content of wheat seeds and desiccant during contact drying at various desiccant/grain mass ratios.

Desiccant/Grain Mass Ratio	Time (min)
	0	60	150	240	360
Moisture content of wheat seeds (% wb), ±0.1 (p = 0.95)
1:1	21.5	18.3	16.3	15.1	14.4
1:2	21.5	18.5	16.5	15.3	14.6
1:4	21.5	18.7	16.7	15.5	14.8
Moisture content of desiccant (wt%), ±0.02 (p = 0.95)
1:1	15.88	18.95	20.59	21.66	22.48
1:2	15.88	21.49	24.62	26.44	27.03
1:4	15.88	24.98	29.78	31.55	32.62

. The moisture ratio changes are shown in Figure 7.

The data obtained show that during the first hour of drying, the maximum moisture release of seeds $(\mathrm{M}{\mathrm{C}}_0$−$\mathrm{M}{\mathrm{C}}_{\mathrm{t}}$) occurred. Depending on the desiccant/grain mass ratio, this value ranged from 3.2 to 2.8%, achieving seed moisture of 18.3–18.7%. Then, 4 h from the start of drying, the seed moisture was 15.1–15.5%, which no longer exceeded the critical MC [17]. After 6 h of contact drying, the final MC reached the recommended values of 14.4 to 14.8% (Table 2) for the safe storage of wheat seeds [19]. It should be noted that this drying time is significantly less compared to other known sorbents in contact drying processes [5,23,24,25,26,27,28,30]. Another advantage of contact drying with kieserite is that the required moisture content of the wheat seeds was achieved without thermal heating, as is the case with energy-intensive traditional convection drying using industrial hot-air systems [17,20,21].

The moisture ratio changes of the grain dried in contact with the desiccant are shown in Figure 7a. Initially, moisture removal was rapid, but, as drying progressed, the rate of moisture removal slowed down at all tested desiccant/grain mass ratios. This may have been due to the fact that loosely bound moisture could be quickly released from the sample. The diffusion of moisture from the interior of the grain to the surface limited the drying rate. Desiccant moisture saturation increased with decreasing desiccant/grain mass ratios (Figure 7b).

The graph in Figure 8 illustrates the correlation between wheat moisture content and crystalline hydrate composition during contact drying. It can be observed that there was a linear relationship for all desiccant/grain mass ratios that were examined. As the amount of desiccant in the mixture with wheat decreased, there was a corresponding increase in the number of water molecules present in the crystalline hydrate (Figure 8).

Taking into account the MC values of the desiccant after 6 h of contact with wet wheat seeds (Table 2), the number of H₂O molecules (n) in the composition of MgSO₄·nH₂O crystalline hydrate was 1.94, 2.48, and 3.23 for the 1:1, 1:2, and 1:4 ratios, respectively. The composition of crystalline hydrates indicated the possibility of reuse without intermediate regeneration for all of the desiccant/grain mass ratios studied.

3.3. Three-Stage Sorption Drying

The three-stage drying procedure allowed the desiccant to be used without intermediate regeneration. An additional reduction in the proportion of drying agent in the mixture with wheat contributed to quantitative savings in the sorbent.

To determine the number of wheat batches that could be sequentially added to the desiccant until it was completely saturated with moisture without intermediate regeneration, a three-stage contact drying process was performed (Figure 2). The desiccant/grain mass ratios were 1:1 and 1:2.

Table 3. Moisture content (% wb) of wheat seeds during three-stage sorption drying and the moisture ratio at 360 min.

Stage	Time (min)					MR360
	0	60	150	240	360
	Moisture Content of Wheat Seeds (% wb), ±0.1 (p = 0.95)Desiccant/Grain Mass Ratio = 1:1
1	21.5	18.3	16.5	15.3	14.4	0.67
2	21.5	–	–	–	14.6	0.68
3	21.5	–	–	–	15.5	0.72
	Moisture Content of Wheat Seeds(% wb), ±0.1 (p = 0.95)Desiccant/Grain Mass Ratio = 1:2
1	21.5	18.5	16.5	15.4	14.6	0.68
2	21.5	–	–	–	14.8	0.69
3	21.5	–	–	–	15.9	0.74

shows the MC of wheat at a certain time from the beginning of the drying process of the first wheat batch and the final MC of subsequent batches after 6 h of drying.

It was revealed that with a desiccant/grain mass ratio of 1:1, for three successive batches of seeds with an initial MC of 21.5%, the final MC values were 14.4%, 14.6%, and 15.5%, respectively. In the case of a desiccant/grain mass of 1:2, the final MC₃₆₀ of the seeds was 14.6% and 14.8% for two batches of wheat, respectively, which did not exceed the critical MC [17]. When the third batch of seeds was dried, a decrease in the sorption capacity of the desiccant was observed: the final MC₃₆₀ of the seeds was 15.9% (Table 3). During the three-stage sorption drying, as batches of wheat were added, the moisture ratio MR₃₆₀ changed slightly, from 0.67 to 0.72 and from 0.68 to 0.74 with a desiccant/grain mass ratio of 1:1 and 1:2, respectively (Table 3), indicating the uniform removal of moisture from seeds of different batches.

Table 4. Moisture content and composition of desiccant during three-stage sorption drying.

Stage	Moisture Content (wt%), ±0.02 (p = 0.95)	Composition
Desiccant/grain mass ratio = 1:1
1	22.48	MgSO4·1.94H2O
2	28.10	MgSO4·2.61H2O
3	32.30	MgSO4·3.19H2O
Desiccant/grain mass ratio = 1:2
1	27.03	MgSO4·2.48H2O
2	35.50	MgSO4·3.68H2O
3	42.83	MgSO4·5.00H2O

presents the MC and crystalline hydrate composition of the desiccant during the three-stage sorption drying of wheat seeds. The data show how the composition of the desiccant changed as batches of wheat seeds were added.

Therefore, at a 1:2 desiccant/grain mass ratio, the number of water molecules in the magnesium sulfate crystalline hydrate MgSO₄·nH₂O increased from 2 to 3. With a desiccant/grain mass of 1:2, a crystalline hydrate composition with n = 5 was achieved (Table 4). In the range of compositions of magnesium sulfate studied, the desiccant retained stable moisture-adsorbing properties, which was confirmed by the MC values achieved for the wheat seeds (Table 3). The composition of the crystalline hydrates confirms the possibility of reusing the desiccant without intermediate regeneration, where the number of sequentially added batches of wheat or other agricultural crops will depend on the initial MC of the drying object and the final MC required for the product.

The established relationship between the composition of the drying agent and the MC of the wheat seeds during sorption drying clearly indicates that kieserite possesses a high water adsorption capacity, rapid sorption kinetics, and consistent performance in a multi-stage process, making it a highly promising solid desiccant for energy-efficient sorption technology. Silica gel [24], clay minerals [29], and zeolites [5] have been studied as potential desiccating agents for various crops. However, the drying times have been reported to range from 12 h to several days. One major drawback of using heat drying is the high energy consumption [22] and potential loss of seed germination [11,17].

3.4. Wheat Seed Germination

Standard germination tests [54,55] are commonly used to assess the viability of seeds after various post-harvest treatments. It is known that when seeds are heated, depending on the initial MC and the duration of heat treatment, a deterioration in their quality is observed [11,17,20]. Several studies [7,59] have reported that for good seed germination, the maximum recommended drying temperature for wheat should not be higher than 40 °C.

To control the possible heating of the seeds upon contact with magnesium sulfate due to the exothermic effect of the reaction of crystalline hydrate formation, the temperature of the mixture of wheat and desiccant was measured. It was established that during the sorption drying of wheat seeds at room temperature (25 °C) and with a desiccant/grain ratio of 1:2, the maximum temperature of the mixture was 31 °C for 30 min from the beginning of contact, which did not exceed the safe heating temperature of the seed [7,59].

The determination of wheat germination [54,55] after sorption drying from an initial MC₀ of 22% to the final MC₃₆₀ of 15% showed that the GE and GC were 91 ± 1% and 97 ± 2%, respectively, which is typical for high-quality seeds [58]. Wheat seeds air-dried to an MC of 15% were used as a control group sample. For comparison, the GE of these air-dried seeds was 87 ± 1% and the GE was 94 ± 2%, which was slightly lower than after contact drying with the desiccant. These results confirm that the contact of seeds with kieserite is safe for their germination.

3.5. Characterization of Desiccant after Sorption Drying

During the contact drying of wheat, the desiccant did not adhere to the grain and was easily separated by sifting after contact with the wheat, which is clearly demonstrated in the photos in Figure 9.

The method of simultaneous thermal analysis allowed us to determine the amount of water during the dehydration of desiccant samples after three-stage contact drying (Figure 10,

Table 5. Mass loss (∆m) and water capacity in the temperature range of 40–450 °C of the desiccant after three-stage sorption drying of wheat seeds at a desiccant/grain mass ratio of 1:2.

Parameter	Temperature Range (°C)
	40–100	40–150	40–200	40–250	40–300	40–350	40–400	40–450
∆m (wt%)	23.43	29.45	33.01	34.43	36.43	37.83	37.71	37.58
Water capacity (mg/g) *	377	474	531	554	586	608	607	604

* per gram of anhydrous substance.

). The state with maximum weight loss was considered anhydrous; in Table 5, these values are highlighted in bold.

As a result of the contact drying of the wheat, the desiccant sample was saturated with water in a state corresponding to the gross formula of the crystalline hydrate MgSO₄∙4.07H₂O, with a total weight loss of 37.83 wt% (Figure 10). According to the DSC/DTG dehydration curves of the “saturated” desiccant, it was clear that, compared to the initial one (Figure 6), there was a significant change in the complex DSC peak, accompanied by a significant increase in the contribution of low-temperature components: mainly components at 76 °C and, to a lesser extent, at 100, 120, 154, and 195 °C. The predominance of the contribution of low-temperature components was associated with a significant increase in the proportion (about 91%) of weakly bound water in the crystalline hydrate MgSO₄∙4.07H₂O compared with MgSO₄∙1.14H₂O. The most strongly bound crystallization water is presented on the DSC/DTG curve as a broad peak in the region of 260–330 °C, with a maximum at 303 °C. The temperature at which the last water molecule began to separate was 142 °C.

Therefore, the thermal dehydration of MgSO₄·nH₂O, saturated with moisture from wheat seeds during contact drying occurred with the release of the bulk of water at temperatures below 80 °C. This allowed the desiccant to be regenerated at low temperatures, which determined the energy consumption at the stage of regeneration of the sorbent.

Figure 11 shows the desiccant XRD pattern after the three-stage sorption drying of wheat seeds at a desiccant/grain mass ratio of 1:2. The X-ray diffraction pattern calculated by the DDM method [52] corresponded to the crystalline phase hexahydrate MgSO₄·6H₂O (ICDD PDF 00-024-0719) [51]. During the contact drying of the wheat seeds, magnesium sulfate interacted with moisture localized on the surface of wheat grains and formed crystalline hydrates with an increasing number of water molecules. The composition of the desiccant changed from kieserite to hexahydrate.

After three-stage contact drying, the desiccant retained the size of the granules. According to the granulometric composition, the content of the −1.0 + 0.5 mm fraction was 99%, as for the original kieserite.

3.6. Cyclicity Testing

One important performance characteristic of the desiccant was its ability to withstand regeneration between cycles of multi-stage contact drying. In order to evaluate this performance, the desiccant was subjected to three-stage contact drying and then regenerated at 80 °C before being reused in a second three-stage wheat drying cycle. At a 1:1 desiccant/grain mass ratio, the final moisture content (MC₃₆₀) of wheat was 14.2%, 14.5%, and 15.4% for three successive seed batches with an initial moisture content (MC₀) of 21.5%. These values were similar to those obtained in the first drying cycle (Table 3) and demonstrated the stable sorption properties of kieserite after regeneration.

Therefore, the sorption drying of wheat seeds using granulated kieserite as a solid desiccant, compared to other drying techniques, is a simple and effective method of post-harvest processing, eliminates overheating, provides mild drying conditions, and achieves the required values of seed moisture and high germination. The high water capacity of kieserite also allows for multiple batches of seeds to be dried without the need for the intermediate regeneration of the desiccant. Additionally, after regeneration, the desiccant maintains stable sorption characteristics at a high level. In general, the development of innovative technologies and new materials for grain processing is a promising direction for the production and storage of agricultural products.

4. Conclusions

The presented results demonstrate that the contact sorption drying of wheat seeds with granulated kieserite MgSO₄·H₂O as a solid desiccant is an effective post-harvest treatment that reduces seed MC while maintaining high seed germination. It was established that for wheat seeds with an initial MC of 21.5%, the maximum value of moisture removal was observed during the first hour of contact drying at the studied desiccant/grain mass ratios of 1:1, 1:2, and 1:4, which made it possible to achieve an MC of 18.3–18.7%. Four hours from the start of drying, the MC was 15.1 to 15.5%, which did not exceed the critical value. After six hours of contact drying, the final MC reached a value of 14.4–14.8%. For three sequentially dried batches of seeds at a desiccant/grain mass ratio of 1/1, the final MC values were 14.4, 14.6, and 15.5%. The composition of crystalline hydrates after contact drying confirmed the possibility of reusing the desiccant without intermediate regeneration. The thermal dehydration of MgSO₄·nH₂O with the release of the bulk of water occurred at temperatures below 80 °C, determining the possibility of low-temperature regeneration. Desiccant granules were not destroyed during contact drying, did not adhere to the grain, were easily separated by the sieve method, retained their integrity after regeneration, and had stable cyclic characteristics. The germination energy and the seed germination capacity after sorption drying were 91 ± 1% and 97 ± 2%, respectively. It was concluded that the sorption drying of various agricultural products with kieserite as a solid desiccant can provide non-thermal and efficient drying technology. The results of this study will serve as the foundation for pilot testing the contact drying of wheat during the harvest season. Additionally, these results can be used to develop sustainable technologies for dehydrating thermolabile materials.