Characterization of Different Magnesium Fertilizers and Their Effect on Yield and Quality of Soybean and Pomelo

Abstract

Fertilizer application, especially their physical and chemical composition, substantially regulates crop growth and development. The form in which fertilizers are applied to the soil has always been regarded as a crucial factor regulating nutrient availability. However, the properties and release characteristics of Mg fertilizers, i.e., fast-release Mg (F-Mg) and slow-release Mg (S-Mg), have not been fully elucidated in acidic soils. This study characterized the different Mg fertilizers, and their release characteristics were verified through pot (using soybean) and field (using pomelo) experiments. The results showed that, despite the differences between different Mg fertilizers, the same functional group peaks were recorded among them. F-Mg fertilizers had a low pH and low Mg purity, while S-Mg fertilizers had a high pH and high Mg purity. The release rate and leaching characteristics of the F-Mg fertilizers were higher than the S-Mg fertilizers. The pot experiment showed that the yield and growth of soybean were higher under the S-Mg fertilizer than the F-Mg fertilizer. However, MgSO₄·7H₂O and MgO had the best effect among the F-Mg and S-Mg fertilizers, respectively. The effects of these two fertilizers were further validated using field experiments, and it was found that MgSO₄·7H₂O and MgO fertilizers substantially improved the yield and quality of pomelo. However, MgO showed a better effect than MgSO₄·7H₂O. This study can provide a sound theoretical basis for selecting the most efficient type of Mg fertilizer for acid soils. It can contribute valuable information regarding farmland management strategies and may result in sustainable agricultural productivity.

1. Introduction

Magnesium (Mg) has diverse physiological functions in biological systems [1,2]. Its availability is significantly influenced by various factors, especially soil acidification [3]. However, Mg deficiency inhibits plant growth and development, eventually resulting in low yields and poor quality [4,5,6]. Therefore, to increase crop yields, various fast-release Mg (F-Mg) fertilizers are applied to crops. Generally, F-Mg fertilizers include magnesium sulfate heptahydrate (MgSO₄·7H₂O), magnesium chloride hexahydrate (MgCl₂·6H₂O), and magnesium nitrate hexahydrate (Mg(NO₃)₂·6H₂O) [7,8]. Unfortunately, Mg has the lowest ionic radius and the largest hydrated radius among different cations [9], and as a result, it can leach down easily. In addition, high temperatures and rainfall also cause serious leaching and result in the loss of soil Mg [10]. These Mg fertilizers leach easily and have low utilization rates, leading to a waste of resources [11,12,13]. Hence, it is necessary to find a slow-release Mg (S-Mg) fertilizer, to solve F-Mg fertilizer’s shortcomings.

At present, S-Mg fertilizers mainly include magnesium carbonate (MgCO₃), magnesium oxide (MgO), and magnesium hydroxide (Mg(OH)₂), which are prepared from dolomite, magnesite, calcined magnesite, and other ore materials [14,15,16]. Previous studies have found that MgO application significantly improved maize yields and increased the concentration of free amino acids and shoot density in tea [14,17,18]. Similarly, MgCO₃ significantly improves the total, nonprotein, and protein nitrogen in potatoes, in both cortex and pith tissues [19]. Moreover, MgCO₃ increases sugar beet yield and the Mg content of leaves and roots [20]. Mg(OH)₂ can effectively increase the number of leaves and fresh weight of Chinese cabbage [21]. These studies mainly focused on Mg fertilizers’ effects in various crops, however, empirical knowledge of the properties, solubility, and effects of S-Mg fertilizers on crop growth remains poor.

Mg fertilizers from different mineral sources have differences in their chemical composition and particle size, and they behave differently when added to soils with different pH, which often leads to differences in the solubility of Mg fertilizers [14,15,22]. Although there is no difference in the water solubility of the same type of Mg fertilizer, a granular insoluble Mg fertilizer is more difficult to release than the powdered Mg fertilizer [14]. The water solubility of Mg in fertilizers depends on the chemical composition, such as oxide, sulfate, carbonate, nitrate, chloride, phosphate, or silicate [22]. However, previous reports mostly focused on the application effects of Mg fertilizers on crops, such as the application effects of magnesium fertilizers on crop yield and quality [23,24,25,26]. Therefore, a comprehensive understanding of the properties of Mg fertilizers from different sources is lacking.

The Mg supply capacity and level vary significantly among different types of soil. The Mg supply capacity of several typical soils can be ranked as follows: paddy soil > brown lime soil > dark muddy brick red soil > muddy red soil > hemp sandy red soil > siliceous red soil > red clay soil [27]. According to the recommended fertilization index of a systematic research method of soil nutrient status in Florida, USA, soil available Mg content can be divided into three grades: an available soil Mg content less than 15 mg kg⁻¹ is a deficiency, 15–30 mg kg⁻¹ is moderate, and more than 30 mg kg⁻¹ is rich [28]. However, it is generally believed that the Mg deficiency is more obvious in the red soil areas of southern China [29]. Limestone is widely used in acidic soils with a low Mg content, because limestone not only corrects the soil acidity [30,31,32], but also provides Mg to the soil in the form of magnesium carbonate or magnesium oxide [33,34]. In addition, limestone materials are less expensive than soluble Mg sources and are popular in many Mg deficient areas [34]. However, some limestones have a poor quality and low Mg content [35]. If the Mg content in the soil is too high, it will compete with nutrients such as potassium, calcium, and manganese in the soil for the same adsorption site, resulting in nutrient loss [36,37,38]. Therefore, a suitable content of Mg material can better meet the absorption of Mg by plants.

Mg deficiency usually occurs in acidic soils of China [26], hence, it is of great significance to study the nutrient release characteristics of different Mg fertilizers and their availability for sustainable agricultural production, by choosing suitable Mg fertilizers. Therefore, we wanted to verify the difference in fertilizer efficiency between S-Mg fertilizers and F-Mg fertilizers, and we hypothesized that in acidic soils, S-Mg fertilizers would relieve the symptoms of Mg deficiency better than F-Mg fertilizers and promote the soil nutrient balance, thereby improving the yield and quality of related crops. Therefore, we selected six common elemental Mg fertilizers, and the key objectives were as follows: (1) to characterize the physical and chemical characteristics and leaching patterns of different Mg fertilizers; (2) to explore the effects of different Mg fertilizers on soil properties and soybean growth through pot experiments; (3) to verify the effects of the best Mg fertilizers (based on the selection of Mg fertilizers from the pot experiments) on the soil properties, yield, and quality of pomelo under field experiments. These findings could provide a theoretical basis for improving the utilization rate of Mg fertilizer and selecting high-efficiency Mg fertilizers in agricultural production and contribute information regarding farmland management strategies that could improve quality and yields.

2. Materials and methods

2.1. Materials

The different fertilizers, including urea (46% N), ammonium dihydrogen phosphate (12% N, 13% P), and potassium sulfate (21% K), were purchased from Quanyoubao Agricultural Fertilizer Technology Co., Ltd. (Fujian, China). Different magnesium (Mg) fertilizers, including MgSO₄·7H₂O, Mg(OH)₂, MgCO₃, and MgO, were obtained from Yingkou Magnesium Chemical Ind Group Co., Ltd., Liaoning, China, with MgCl₂·6H₂O from Qinghai Salt Lake Magnesium Industry Co. Ltd., Qinghai, China, and Mg(NO₃)₂·6H₂O from Xilong Co., Ltd., Shantou, China. The planting material was soybean and 8-year-old pomelo trees selected in Xizhou Village, Pinghe County, Fujian Province, China.

2.2. Characterization of Mg Fertilizers

Mg fertilizers, including MgCl₂·6H₂O (12% Mg), Mg(NO₃)₂·6H₂O (9% Mg), MgSO₄·7H₂O (10% Mg), MgO (59% Mg), Mg(OH)₂ (41% Mg), and MgCO₃ (28% Mg) were dried at 80 °C for 2 h and sieved through an 80-mesh sieve, and then used in 400–4000 cm⁻¹ Fourier transform infrared spectroscopy (FTIR, Nicolet IS 10, America) for characterization within the wavenumber range. The structural characterization of the sample was examined by X-ray diffraction spectroscopy (XRD, Bruker D8 Advance, Germany) with Ni-filtered Cu Kα radiation.

2.3. Release Characteristics of Mg Fertilizers

The Mg release from the studied fertilizers was measured by calculating the molecular mass of different Mg fertilizers, and 100 mg of pure Mg (sieved through an 80-mesh sieve) was placed in a plastic bottle containing 100 mL of deionized water and kept in an incubator at 25 ± 0.5 °C. The Mg release content was measured from each type of Mg fertilizer at 1, 3, 5, 7, 10, 20, 30, and 40 day intervals [39]. To further explore the solubility of ions of different Mg forms in water, while measuring the Mg release rate, the conductivity of Mg in an aqueous solution was measured with a HANNA conductivity meter (HI2003-01, America). To observe the dissolution changes of the six Mg fertilizers more intuitively, 100 mg of pure Mg of each sample was added to 100 mL of deionized water and kept for 40 days. Different types of Mg were dissolved in distilled water, and the pH and Mg contents were determined according to the NY/T standards of NY/T standard of NY/T 1973–2010 and GB/T 26568-2011 (China).

2.4. Leaching Characteristics of Mg Fertilizers

To observe the leaching characteristics of Mg fertilizers, pots (upper diameter 20 cm, lower diameter 16 cm and height 20 cm) were filled with 5 kg of soil with a hole (diameter of 1 cm) at the bottom to collect the leachates. The soil passed through a 10-mesh sieve, and the bulk density was 1.48 g cm⁻³, while other physical and granulometric properties of the soil were consistent with the pot experiments below. In order to be consistent with the Mg application rate for the pot experiments and the Mg fertilizer application for soybean in red soil according to Chen et al. (2017) [40], 250 mg of pure Mg (i.e., MgCl₂·6H₂O, Mg(NO₃)₂·6H₂O, MgSO₄·7H₂O, MgO, Mg(OH)₂, MgCO₃) was sieved through an 80-mesh sieve and spread on the surface of the red soil in each pot. The present work was carried out in Pinghe County (24.34° N, 117.31° E), in the southern region of Fujian province, China; this region is characterized by a subtropical monsoon climate with an annual mean temperature and precipitation of 23.58 °C and over 1600 mm, respectively [41]. The irrigation volume was calculated by considering the rainfall in September, October, and November, an average of 250 mL of water per day, by collecting the eluent every 7 days, and collecting the eluent 10 times. The Mg leaching accumulation per kilogram of soil (MLAKS) was calculated with the following equation:

$$\mathrm{MLAKS}=\sum \left\{\right(\mathbf{Leaching}\mathbf{Mg}\mathbf{concentration}*\mathbf{Leaching}\mathbf{volume})/\mathit{soil}\mathbf{weight}\}$$

The Mg concentration in the eluent was determined using an Inductively Coupled Plasma Optical Emission Spectrometer (Avio 200, America).

2.5. Pot Experiment

To elucidate the release characteristics of various Mg fertilizers, a pot (upper diameter 20 cm, lower diameter 16 cm, and height 20 cm) experiment with soybean was performed at the Science and Technology Backyard (STB) in Pinghe County, September 2018. For this, seven different Mg fertilizer treatments were set up, including without Mg fertilizer (CK), MgCl₂·6H₂O, Mg(NO₃)₂·6H₂O, MgSO₄·7H₂O, MgO, Mg(OH)₂, and MgCO₃, and 250 mg of pure Mg was sieved through an 80-mesh sieve and applied on the surface of the pots, each with six replicates. In accordance with the leaching test of Mg fertilizer, 5 kg of red soil was placed in each pot. The basic red soil properties were as follows: the soil type was classified as a red soil having 25.1% clay, 35.8% silt, and 39.2% sand [42]; pH: 4.33; NO₃⁻-N: 21.07 mg kg⁻¹; NH₄⁺-N: 29.90 mg kg⁻¹; Bray I-P: 2.03 mg kg⁻¹; available K: 38.4 mg kg⁻¹; exchangeable Mg: 9.6 mg kg⁻¹, and exchangeable Ca: 83.10 mg kg⁻¹. According to the amount of fertilizer required for soybean seedling growth, N, P, K, and Mg fertilizers were applied at the beginning of the pot experiment, i.e., 50 mg kg⁻¹ Mg, 180 mg kg ⁻¹ N, 130 mg kg⁻¹ P, 62 mg kg⁻¹ K [10]; and these fertilizers were ground in a mortar, dissolved in water, mixed evenly with 5 kg of soil, and left for a week. The irrigation amount of potted plants was consistent with the leaching test, and 250 mL of water was supplied through drip irrigation every day. Apart from the fertilizer types and irrigation, all other management practices were identical, including the plant protection, weed control, etc.

Soil samples were collected three times, at 30, 60, and 90 days, after planting and their soil exchangeable Mg and pH was analyzed. After three months, the plant biomass, root length, and yield were measured at soybean maturity. Six sub-samples were collected from each treatment; subsequently, the soil samples were air-dried and passed through a 1.0-mm sieve, to analyze the soil physicochemical properties [43]. For soil analyses, soil pH was measured in a 1:2.5 soil to water suspension with an ORION A215 STAR pH meter (Thermo Ltd., USA) [44]. Exchangeable Mg was extracted using a 1 M ammonium acetate solution (pH 7) and determined by ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometry) [45]. The roots were washed with deionized water and scanned with a scanner (Epson V800, China, Co., Ltd., Beijing, China). Root images were analyzed using WinRhizo software (Regent Instruments Inc., Quebec, QC, Canada) for diffident root morphological traits, including root length, root surface area, and root volume. Soybean biomass and yield were weighed with a thousandth electronic balance. At the same time, according to the above measurement indicators, a correlation analysis was made between the soil indicators and plant growth indicators during soybean growth.

2.6. Field Experiment

After screening the Mg fertilizers based on their release characteristics under the pot experiments, a fast-release magnesium (F-Mg) fertilizer (MgSO₄·7H₂O) and slow-release magnesium (S-Mg) fertilizer (MgO) were selected and their effects were further verified under field trials for two consecutive years. The field trial was also conducted at the STB in Pinghe County. In the middle of January 2019, 48 9-year-old pomelo trees were selected and grown consistently, to carry out a 2-year field trial; three different treatments were set up, including without Mg fertilizer (CK), MgO, and MgSO₄·7H₂O (sieved through an 80-mesh sieve), and each treatment had 16 replicates. Except for the different Mg fertilizers, the other fertilizers and management methods were the same among different treatments. The basic red soil properties were as follows: the granulometric properties of the soil were consistent for potted soil, with pH, 4.67; NO₃⁻-N, 4.9 mg kg⁻¹; NH₄⁺-N, 32.5 mg kg⁻¹; Bray I-P, 245.32 mg kg⁻¹; available K, 103.3 mg kg⁻¹ and exchangeable Mg, 52.7 mg kg⁻¹. According to the amount of fertilizer required for the pomelo tree, fertilizer application rates were 48 kg ha⁻¹ Mg, 400 kg ha⁻¹ N, 44 kg ha⁻¹ P, and 166 kg ha⁻¹ K, respectively. N, P, and K granular fertilizers were applied 4 times a year [38]: in February (shooting and flowering stage), April (fruit stabilizing stage), June (fruit expansion stage), and December; the 4 fertilization ratios were 3:2:3:2; the Mg fertilizers were applied once in February; and all fertilizers were applied 80–100 cm from the trunk.

Eight sub-samples were randomly collected from each treatment. Soil and leaf samples were collected twice a year, i.e., in May and December, and analyzed for soil exchangeable Mg, pH, and leaf Mg concentrations. The pretreatment of soil samples was consistent with the pot experiment. Fruit samples were collected in October, to measure the yield and quality. The measurement methods of soil exchangeable Mg and pH were the same as in the soybean pot experiments. The determination method of leaf Mg content and fruit quality of pomelo was consistent with Zhang (2021) [38].

2.7. Statistical Analysis

All statistical analyses were conducted using Microsoft Excel 2016 (Microsoft, Redmond, WA, USA) and the IBM SPSS Statistics 21.0 (IBM, New York City, NY, USA). Then, one-way ANOVA was used to determine the effects of the Mg treatment on Mg leaching from the soil, plant growth, soil Mg and pH regularity, and the yield and quality of pomelo. Mean separation tests (Duncan’s multiple range test) were used to determine significant differences among treatments. The differences among means and correlation coefficients were considered significant when p < 0.05. All figures were produced in Origin 2021b (America).

3. Results

3.1. Structural Characterization of Mg Fertilizers

The FTIR spectra of different Mg fertilizers were used to investigate the similarities and differences between functional groups (Figure 1A). Some of the functional groups of F-Mg fertilizers (MgCl₂·6H₂O, Mg(NO₃)₂·6H₂O, MgSO₄·7H₂O) were different from the S-Mg fertilizers (MgO, Mg(OH)₂, MgCO₃), while some functional groups were common to F-Mg and S-Mg fertilizers at 3419, 3640, 3385, 3700, 3697, and 3647 cm⁻¹ spectra. In addition, the F-Mg fertilizers had similar functional groups at 1637, 1631, and 1651 cm⁻¹ spectra. The S-Mg fertilizers had similar functional groups at the 3700, 3697, and 3647 cm⁻¹ spectra. We also recorded differences between the Mg fertilizers in the XRD patterns (Figure 1B), where the characteristic peaks of the six Mg fertilizers were quite different, indicating different characteristics of different Mg fertilizers.

3.2. Physicochemical Characteristics of Mg Fertilizers

The F-Mg fertilizers had acidic characteristics, with a pH of range from 5.58 to 5.95, while the S-Mg fertilizers were alkaline, with a pH of range from 10.77 to 11.01 (Figure 2A). F-Mg fertilizers had lower Mg contents, ranging from 9.47% to 11.48%, while S-Mg fertilizers had higher contents, i.e., 25.55% to 45.96% (Figure 2B); the highest Mg contents (45.96%) were recorded for the MgO fertilizer. Water-soluble Mg contents in MgO accounted for 8%, which was the highest in the S-Mg fertilizers (Figure 2C).

The release curves of the different types of Mg fertilizers showed that the initial release rate of F-Mg fertilizers reached more than 70% during the first day of incubation (Figure 2D). However, the S-Mg fertilizer release rates were 2.3% for MgO, 2.1% for Mg(OH)₂, and 1.9% for MgCO₃. On the 10th day of incubation, the F-Mg fertilizer release rates were 93% for MgCl₂·6H₂O, 96% for Mg(NO₃)₂·6H₂O, and 100% for MgSO₄·7H₂O. At the 40th day of incubation, the S-Mg fertilizer release rates were 8.3% for MgO, 6.3% for Mg(OH)₂, and 5.5% for MgCO₃. Overall, the release rates of the S-Mg fertilizers were slower than that of the F-Mg fertilizers. Moreover, when different Mg fertilizers were dissolved in water, the dissolution rate of S-Mg fertilizers was slower than that of the F-Mg fertilizers after 40 days (Figure S1). S-Mg fertilizers had a low conductivity in water, even after intensive stirring (Figure 2E). The conductivity of F-Mg fertilizers was significantly higher than that of the S-Mg fertilizers on the first day. In addition, MgSO₄·7H₂O had only one SO₄²⁻, MgCl₂·6H₂O, and Mg(NO₃)₂·6H₂O with two anions, so the conductivity was lower than the other two F-Mg fertilizers.

The cumulative amount of Mg leaching in the treatments with the F-Mg fertilizers was higher than that of the S-Mg fertilizers (Figure 2F). The average leaching amount of the three F-Mg fertilizers was 1.58 mg kg⁻¹, and the average leaching amount of the three S-Mg fertilizers was 0.87 mg kg⁻¹. Compared with the F-Mg fertilizer treatment, the S-Mg fertilizer treatment reduced the Mg leaching by 44.9%. Among all the treatments, the highest leaching amount was 2.6 mg kg⁻¹ for Mg(NO₃)₂·6H₂O, while the lowest leaching amount was recorded for MgO (0.84 mg kg⁻¹).

3.3. Evaluation of Different Mg Fertilizers under Pot Experiments

3.3.1. Effects on Soil Mg and pH

Mg fertilizers showed contrasting patterns of soil Mg and pH with different soil depths, 30, 60, and 90 days after fertilization. With a 0–5 cm soil layer, the S-Mg fertilizers first showed a decreasing trend of exchangeable Mg and then an increasing for 30, 60, and 90 days after fertilization; while the exchangeable Mg gradually decreased under the three F-Mg fertilizers (Figure 3A). Hence, after 90 days of fertilization, the average soil exchangeable Mg of each S-Mg fertilizer treatment was 137.2% higher than that of the average F-Mg fertilizer under 0–5 cm. Mg contents under the 5–10 cm (Figure 3B) and 10–15 cm (Figure 3C) soil layers were lower than 0–5 cm, the Mg content of Mg(NO₃)₂·6H₂O was higher than other treatments 90 days after fertilization, which was caused by its fast leaching speed. Similarly, different types of Mg fertilizers had a greater impact on the pH in 0–15 cm soil. Compared with CK, the F-Mg fertilizer decreased the soil pH, while the S-Mg fertilizer treatment increased the soil pH. After 90 days of fertilization, in the 0–5 cm (Figure 3D), 5–10 cm (Figure 3E), and 10–15 cm (Figure 3F) soil layers, the average soil pH of the three S-Mg fertilizers was 5.2, 5.1, and 5.1, respectively, and higher than average pH of the three F-Mg fertilizers (4.5, 4.7, and 4.8).

3.3.2. Effects of Mg Fertilizers on Soybean Growth

The application of various Mg fertilizers did not show a significant effect on one month old soybean growth (Figure 4A). After two months, the growth of soybean under S-Mg fertilizers was better than that of F-Mg fertilizers (Figure 4B). S-Mg and F-Mg fertilizers showed substantial results compared to CK, in terms of root biomass (Figure 4C), stems and leaf biomass (Figure 4D), and pod biomass (Figure 4E), but the S-Mg fertilizers showed better results than the F-Mg fertilizers. In terms of the total biomass (Figure 4F), the increase rate of F-Mg fertilizers was 12–45%, while for S-Mg fertilizers it was 32–47%, and the best results for total biomass were obtained by MgSO₄·7H₂O (F-Mg fertilizer) and MgO (S-Mg fertilizer).

The root growth under S-Mg fertilizers was better than that of the F-Mg fertilizers (Figure 5A). Root length (Figure 5B), root surface area (Figure 5C), and root volume (Figure 5D) under the S-Mg fertilizers showed better results than those of the F-Mg fertilizers. Among them, the F-Mg fertilizer, i.e., MgSO₄·7H₂O, and S-Mg fertilizer, i.e., MgO, showed the best effects. Taking root length as an example, compared with CK, the increase rate of root length under the F-Mg fertilizer was 58–120%, while for the S-Mg fertilizer it was 103–130%.

The soybean flowering number (Figure 6A), pod number (Figure 6B), 100-seed weight (Figure 6C), and yield (Figure 6D) under S-Mg fertilizer were higher than those of F-Mg fertilizer, which was consistent with the biomass and root length. MgSO₄·7H₂O and MgO had the best effects. Taking yield as an example, compared with CK, the increase rate under F-Mg and S-Mg fertilizers was 8–52% and 86–109%, respectively.

Correlation analysis showed that under a 0–5 cm soil layer, the exchangeable Mg was positively correlated (p < 0.05) with soybean yield. The soybean yield was positively correlated (p < 0.05) with the 0–5 cm soil pH and biomass. Mg leaching accumulation was negatively correlated (p < 0.05) with 0–5 cm soil exchangeable Mg and soybean yield (Figure 7).

3.4. Evaluation of Mg Fertilizers under Field Experiments

3.4.1. Effect on Soil Mg and pH

Field test verification of F-Mg fertilizer (MgSO₄·7H₂O) and S-Mg fertilizer (MgO) was performed for two consecutive years, i.e., 2019 and 2020. In 2019, we found that under the soil depth of 0–20 cm, the MgO fertilizer significantly improved the soil exchangeable Mg during the month of December; while in 2020, Mg fertilizers had a significant impact on the exchangeable Mg under the 0–20 cm, 20–40 cm, and 40–60 cm soils (p < 0.05) (Figure 8A). In the 0–20 cm, 20–40 cm, and 40–60 cm soil layers, the exchangeable Mg of the soil treated with MgO was 12.3%, 108.5%, and 171.7%, respectively, and higher than that of MgSO₄·7H₂O. In contrast, different Mg fertilizer treatments had no significant impact on soil pH in 2019, while significant results were recorded for 2020 (Figure 8B). Similarly, in the 0–20 cm, 20–40 cm, and 40–60 cm soil layers, the pH of the soil treated with MgO was 8.3%, 5.0%, and 7.2% higher than that of the MgSO₄·7H₂O treatment, respectively. Hence, the exchangeable Mg and pH of the soil under the MgO treatment were higher than those under MgSO₄·7H₂O. This is similar to the results of the pot experiments; in acid red soil, Mg-S fertilizer (MgO) had a better effect than Mg-F fertilizer (MgSO₄·7H₂O).

3.4.2. Effects of Mg Fertilizers on Pomelo Growth

Mg fertilizer significantly improved the pomelo yield and quality. For example, MgO fertilizer significantly improved the leaf Mg contents compared to CK for 2019 and 2020 during December; however, MgSO₄·7H₂O fertilizer did not increase the leaf Mg content in December for 2020 (Figure 9A) (p < 0.05). Compared with CK, the Mg content of pulp treated with MgO was significantly increased by 27% in 2020 (Figure 9B). In 2020, MgO treatment significantly increased the yield of pomelo by 26% (Figure 9C), the increase in yield was caused by the increase in single fruit weight (Figure 9E), but the number of fruits per tree had no significant effect (Figure 9D). The slitting-cut picture of the pomelo treated with Mg was slightly larger than that of CK (Figure 9J). In addition, the application of Mg fertilizer improved the quality of pomelo, especially with the MgO treatment. Compared with CK, the MgO treatment significantly increased the total soluble solids by 4.6% in 2019 and 2020 (Figure 9F), reduced the titratable acidity by 6.8% in 2020 (Figure 9G), increased the solid-acid ratio by 13.2% in 2020 (Figure 9H), and increased the edible rate by 3.6% in 2020 (Figure 9I). Notably, the effect of MgO on the yield and quality of pomelo was better than that of MgSO₄·7H₂O.

4. Discussion

4.1. Characterization of Mg Fertilizers

Infrared spectroscopy and X-ray diffraction technology are gradually being applied in materials science. They can reveal the chemical bonds in molecules and functional group information, and determine the crystal structure and specificity of different materials [46,47]. Previous studies found that the characteristic peaks of F-Mg fertilizers (MgCl₂·6H₂O, Mg(NO₃)₂·6H₂O, MgSO₄·7H₂O) and S-Mg fertilizers (MgO, Mg(OH)₂, MgCO₃) are close to the common peaks of -OH stretching vibrations at 3419, 3460, 3385, 3442, 3442, and 3447 cm⁻¹, respectively [48,49,50,51,52,53]. F-Mg fertilizers corresponded to -OH stretching peak at 1637, 1631, and 1651 cm⁻¹ [48,49,50]. Similarly, S-Mg fertilizers had common -OH stretching peaks at 3700, 3697, and 3647 cm⁻¹ [51,52,53]. However, different Mg fertilizers have unique characteristic peaks. As an example, the peak observed at 1110 cm⁻¹ is -OH in MgO, and the functional groups of the Mg-O-Mg compound bonds were seen in the range 463–531 cm⁻¹ [54,55]. In this study, we found that F-Mg fertilizers had -OH at low wavelengths (1631–1651 cm⁻¹), while S-Mg fertilizers had -OH at high wavelengths (3647–3700 cm⁻¹) (Figure 1A). This could be one of the reasons for the differences in properties between F-Mg fertilizers and S-Mg fertilizers. The XRD patterns of six Mg fertilizers were different; for instance, the MgO sample showed five major diffraction peaks at positions 37, 43, 62, 74, and 78. The corresponding lattice planes were 111, 200, 220, 311, and 222 (Figure 1B), which can be matched with [ICSD PDF code 01-075-0447]. These results are in line with earlier reports [56,57].

The release rates of available Mg to the soil solution from various mineral fertilizers are determined by their physical and chemical compositions, which are based on water solubility and other properties [14,22,58]. Previous studies found that MgSO₄·7H₂O is acidic and MgO is alkaline [21], which is consistent with the results of this study (Figure 2A). Moreover, our research found that F-Mg fertilizers had a low pure Mg content (Figure 2B) and high water-soluble Mg content (Figure 2C), while the S-Mg fertilizer had a higher pure Mg content and a low water-soluble magnesium content. The efficiency of Mg fertilizers in increasing the Mg uptake by plants largely depends on the rate of dissolution of the fertilizers in the soil [15]. The rates of dissolution of different Mg fertilizers in soil are expected to vary widely [14,28,59], because their water solubilities vary widely [60,61]. Mg fertilizers with very low rates of dissolution may lead to an under-supply of Mg to plants, whereas those with rapid rates of dissolution may lead to large leaching losses of Mg below the plant root zone under high rainfall conditions. Therefore, it is necessary to measure the solubility of different Mg fertilizers [14]. In this study, we found that the solubility of F-Mg fertilizers was generally higher, while the solubility of S-Mg fertilizers was low (Figure 2D). Similarly to this study, previous studies also found that MgSO₄·7H₂O has a higher solubility in water, while MgO has a lower solubility in water [15]. In addition, this study found that the conductivity of F-Mg fertilizers was higher than that of the S-Mg fertilizers (Figure 2E). Previous studies also showed that the conductivity of MgO dissolved in water was almost unchanged, while the conductivity of MgSO₄ was significantly improved [14].

Mg is very mobile in soils, because it is less bound to the soil charges. This results in a relatively high abundance of this element in the soil solution and thus a higher risk of leaching [13,62,63]. Mg in the red soils of southern China is easily leached by rain [10]. Therefore, suitable types of Mg fertilizers are very important, to control nutrient leaching. Mg leaching is usually studied by simulating soil nutrient leaching experiments. Applications of Mg fertilizers known as slow-release fertilizers (e.g., MgCO₃, MgO) may mitigate leaching risk. The application potential and efficiency of slow-release Mg fertilizers may be slightly higher, especially in acid soil conditions or when applied in ground forms [58,59,64]. On the other hand, the application of soluble Mg fertilizers (e.g., MgSO₄·7H₂O) may lead to Mg losses through leaching, when applied to sandy soils with a high water conductivity (e.g., sandy soils), specifically in wet seasons. It was found that the application of water-soluble Mg fertilizers results in a loss of Mg via leaching during heavy rain [14,61]. In this study, we found that the cumulative amount of leaching of F-Mg fertilizers was higher than that of S-Mg fertilizers (Figure 2F); in particular, the leaching amount of Mg(NO₃)₂·6H₂O was the highest, which is in line with previous research results.

4.2. Effect of Mg Fertilizers on Soil Mg and pH

The efficiency of Mg fertilizers in increasing the Mg uptake by plants largely depends on the rate of dissolution of the fertilizers in soil. The rates of dissolution of these fertilizers in soil are expected to vary widely [14,58,59], because their water solubilities vary widely [61,65]. Previous studies showed that the application of MgO and five other synthetic water-soluble Mg fertilizers can significantly increase soil exchangeable Mg, but after 29 months of application, in the 0–7.5 cm soil layer, the content of soil exchangeable Mg treated with MgO was significantly higher than that of other synthetic water-soluble Mg fertilizers. However, in the 7.5–15 cm soil layer, there was no significant difference, which shows that MgO has a longer fertilizer effect [15]. Applying different Mg fertilizers to flue-cured tobacco soil, the fixed rate of exchangeable Mg in the soil treated with MgO after 90 days was higher than that of MgSO₄·7H₂O [66]. In this study, in the 0–5 cm soil layer 90 days after fertilization, the content of exchangeable Mg in the soil treated with S-Mg fertilizer was higher than that of F-Mg fertilizer (Figure 3A). The 2-year field test further showed that, compared with F-Mg fertilizer (MgSO₄·7H₂O), S-Mg fertilizer (MgO) was more effective in soil, especially in the 20–40 cm and 40–60 cm soil layers (Figure 8A). Therefore, S-Mg fertilizer has a longer lasting nutrient supply capacity in acid soils.

Soil with a low pH can improve the effectiveness of insoluble Mg fertilizers [16]. However, we found that insoluble Mg fertilizers are alkaline, and water-soluble Mg fertilizers are acidic (Figure 2A). Therefore, what is the effect of different Mg fertilizers on the acid-base of acid soil? Previous studies found that dolomite (MgCO₃) can increase soil pH [58], calcining magnesite (MgO) increases soil pH by 0.15–0.35 units [64], and magnesite (MgCO₃) increases pH by 0.2–1.5 units [67]. However, the application of MgSO₄·7H₂O leads to further acidification of the soil [68]. In this study, whether in potting soil pH (Figure 3D–F) or field soil pH (Figure 8B), F-Mg fertilizer reduced the soil pH, while S-Mg fertilizer could increase the soil pH. This shows that Mg fertilizer can be used as a means to adjust soil pH.

4.3. Effect of Mg Fertilizers on Crop Growth

Mg plays essential roles in ensuring crop productivity [16]. Mg fertilizer can improve corn biomass, tomato root length [69], hazelnut and barley yield [23,25], corn thousand kernel weight [70], and the sugar content of beets [71]. Especially in acidic soils, slow-release Mg fertilizers have a better yield improvement effect than fast-release Mg fertilizers [26]. Applying Mg fertilizer significantly increases the yield and quality of citrus and also the Mg content in the leaves [72,73,74]. In this study, Mg fertilizer increased the soybean biomass (Figure 4), root length (Figure 5), and yield (Figure 6). S-Mg fertilizers had a better effect than F-Mg fertilizers; S-Mg fertilizer, i.e., MgO, showed a good effect, while the F-Mg fertilizer, i.e., MgSO₄·7H₂O, showed the best effect. Field experiments also showed that MgO had a better effect than MgSO₄·7H₂O, in terms of the Mg content in leaves and fruit, yield, and quality (Figure 9). Previous studies also found that the application of Mg fertilizer on pomelo increased the Mg content of leaves, pulp, and peel; and increased the longitudinal diameter of pomelo, as well as the pomelo yield and quality [38,75,76,77]. However, there have been few studies on the effect of different types of Mg fertilizer on citrus.

5. Conclusions

This study investigated the physical and chemical properties of different Mg fertilizers, and verified their effects through a pot experiment. Finally, two kinds of Mg fertilizers with the best effects were selected for the field experiment, and an efficient Mg fertilizer was selected. The results showed the position of -OH may be one of the reasons for the differences in characteristics between different Mg fertilizers. F-Mg fertilizers had a higher leaching risk than S-Mg fertilizers. S-Mg fertilizers were more effective than F-Mg fertilizers for soybean yield, biomass, and root index. Among them, the F-Mg fertilizers and S-Mg fertilizers with the best performance were MgSO₄·7H₂O and MgO, respectively. The field results showed that, compared with the F-Mg fertilizer (MgSO₄·7H₂O) treatment, the S-Mg fertilizer (MgO) had a better effect in improving the soil Mg content, pH, yield, and quality of pomelo. Both the pot and field experiments showed that MgO fertilizer had the best performance. This study provides certain strategies to reduce the pressure on the environment caused by Mg leaching, and it has a practical guiding significance in applying suitable types of Mg fertilizers and in the efficient use of resources for crops growing under acid soil conditions.