Investigation of the soybean infiltration process utilizing low-field nuclear magnetic resonance technology

Lisha Guo; Han Wang; Chenru Hao; Ziqiang Chi; Li Cheng; Haibo Yang; Jing Zhang; Ruibin Zhao; Yanru Wu

doi:10.1371/journal.pone.0297756

Abstract

This paper employs low-field nuclear magnetic resonance (LF-NMR) technology to meticulously analyze and explore the intricate soybean infiltration process. The methodology involves immersing soybeans in distilled water, with periodic implementation of Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence experiments conducted at intervals of 20 to 30 minutes to determine the relaxation time T₂. Currently, magnetic resonance imaging (MRI) is conducted every 30 minutes. The analysis uncovers the existence of three distinct water phases during the soybean infiltration process: bound water denoted as T₂₁, sub-bound water represented by T₂₂, and free water indicated as T₂₃. The evolution of these phases unfolds as follows: bound water T₂₁ displays a steady oscillation within the timeframe of 0 to 400 minutes; sub-bound water T₂₂ and free water T₂₃ exhibit a progressive pattern characterized by a rise-stable-rise trajectory. Upon scrutinizing the magnetic resonance images, it is discerned that the soybean infiltration commences at a gradual pace from the seed umbilicus. The employment of LF-NMR technology contributes significantly by affording an expeditious, non-destructive, and dynamic vantage point to observe the intricate motion of water migration during soybean infiltration. This dynamic insight into the movement of water elucidates the intricate mass transfer pathway within the soybean-water system, thus furnishing a robust scientific foundation for the optimization of processing techniques.

Citation: Guo L, Wang H, Hao C, Chi Z, Cheng L, Yang H, et al. (2024) Investigation of the soybean infiltration process utilizing low-field nuclear magnetic resonance technology. PLoS ONE 19(2): e0297756. https://doi.org/10.1371/journal.pone.0297756

Editor: Amit Ranjan, Tamil Nadu Dr J Jayalalithaa Fisheries University, INDIA

Received: August 25, 2023; Accepted: January 11, 2024; Published: February 16, 2024

Copyright: © 2024 Guo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: We have uploaded the data, and the access link is https://doi.org/10.5061/dryad.z612jm6jr.

Funding: This research was funded by the Innovation and Entrepreneurship Education Teaching Reform Research and Practice Project of Hebei Education Department under Grant 2023cxcy067, the Science and Technology Project of Hebei Education Department under Grant QN2023029, and the Youth Project of Hebei Provincial Health Commission under Grant 20240570. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Soybean, a member of the legume family, boasts an abundance of essential nutrients, boasting approximately 40% protein content and 25% fat composition [1]. Within the realms of agriculture and the food industry, soybean assumes a pivotal role as a paramount agricultural commodity and a distinguished wellspring of high-quality plant protein [2]. For example, soybeans are a key ingredient in a myriad of food products, ranging from soymilk and tofu to dried soybeans, tofu skin, and tofu dregs [3–5]. However, it should be noted that elevated levels of moisture content can cause unfavorable consequences, such as mold proliferation, shifts in enzyme activity, and a decrease in the quality integrity of soybeans during their storage and processing phases [6]. Consequently, exploring effective soybean drying techniques and gaining a profound understanding of the dynamic changes in moisture during the soybean infiltration process are crucial for ensuring product quality and sustained supply.

Soybean infiltration and subsequent soybean drying represent sequential stages within the sphere of later-stage soybean storage and processing [7, 8]. Infiltration constitutes a preparatory phase antecedent to the execution of drying, in which soybeans undergo water absorption to facilitate optimal moisture infusion, thus making them suitable for subsequent drying procedures. The state of soybean moistening fundamentally underpins the efficacy of ensuing drying processes [9, 10]. A well-executed infiltration process can result in a more even distribution of moisture within soybeans, thereby preventing excessive gradients between internal and external moisture levels during the drying process, which could otherwise lead to deterioration in quality. This, in turn, mitigates the risk of disproportionate internal and external moisture gradients materializing during the drying trajectory, thus preventing quality deterioration [11]. Consequently, by a deep understanding of the dynamic oscillation in moisture content during the soybean moistening trajectory, informed curation and optimization of drying methodologies can be prudently steered. By comprehensively understanding the dynamic fluctuations in moisture during the soybean infiltration process, a scientific foundation is laid for optimizing the soybean drying process and ensuring the quality and stability of the end product.

Several techniques are available to study the phenomenon of moisture movement, including specific gravity, cross-section, radiofax, X-ray analysis, and Nuclear Magnetic Resonance (NMR) [12–14]. However, specific gravity, cross-section, radiofax, and X-ray analysis techniques are destructive and invasive to varying degrees and do not consistently detect changes in moisture in the sample [15]. The NMR technique, as an emerging analytical test, has achieved great success in the fields of medicine [16], biology [17], and the food industry [18], with advantages such as non-invasiveness, rapidity, and high efficiency. NMR is a physical phenomenon in which nuclear magnetic resonance occurs by applying radio frequency pulses (RF) to a spinning atomic nucleus that is in a static magnetic field B₀, causing the H protons in it to be excited [19, 20]. Low-field Nuclear Magnetic Resonance (LF-NMR) technique is a new technique for NMR applications, which has the advantages of being rapid, non-destructive, and non-invasive, requiring fewer samples, and acquiring data in real-time [21–23]. Analyzing the LF-NMR signals and observing the NMR images, provides an intuitive reference for the study of the content of water [24–26], oil [27–29], and other components as well as the dynamic change process, and it can be used to determine the content of water and oil at the same time, as well as the content of different parts of water based on the differences in water mobility [30, 31]. On the basis of this premise, the current investigation employed LF-NMR as an innovative detection modality. In this study, both LF-NMR and its associated imaging technology were harnessed to scrutinize the soybean water-mobilization process from an innovative standpoint. Through the analysis of LF-NMR signals, the alterations in water content within soybean seeds after varying soaking durations were investigated. Concurrently, the shifts in water distribution within the soybean seeds were dynamically visualized using magnetic resonance images, thereby effectively showcasing the kinetic progression of soybean wetting. This study continuously monitors the moisture change in the soybean wetting process, which provides an intuitive reference basis for the study of moisture change in the soybean wetting process, provides a scientific tool for the control of the moisture threshold in production practice, and also provides a theoretical basis for determining the target moisture content in the soybean drying process, as well as parameters such as the drying temperature.

The contributions of this work are summarized below.

The utilization of LF-NMR technology allows for swift, non-destructive, and dynamic visualization of internal water absorption within soybeans, facilitating a comprehensive comprehension of water migration during the soybean infiltration process. This contributes to the understanding of the mass-transfer pathway of water movement in the bean-water system, and thus to the understanding of the complex morphological structure of legumes.
LF-NMR technology for soybean processing in the infiltration process provides rapid and non-destructive visualization of the technical means.
Soybean infiltration is one of the most commonly used and important pre-processing procedures. The application of LF-NMR effectively defines complete infiltration and germination according to the water content of different components. This approach provides a scientific foundation for moisture control and optimization of processing techniques.

Materials and methods

The present study was conducted using a carefully selected array of experimental materials and instruments to ensure rigorous and accurate investigations. The following components were integral to the experimental framework.

Experimental materials

“Qingtian” brand soybeans were procured from a commercial supermarket, with a production date of June 28, 2023. They originate from Xingtai City, Hebei Province, China. Distilled water was employed for the experiment procedures.

Experimental instruments

The experimental equipment included an NMI20-015V-I nuclear magnetic resonance analyzer (Fig 1), sterile Petri dishes, special magnetic resonance test tubes, sterile gauze, and sterile forceps. The NMR analyzer’s main function was magnetic resonance imaging and relaxation time analysis of water-containing samples. Its resonance frequency (SFOI) was set to 20.826112 MHz. The magnetic field strength was 0.5T±0.08T and the coil diameter was 15 mm. The equipment was equipped with a thermostat to control the temperature at 32±0.01℃ to ensure the accuracy of experimental results.

Download:

Fig 1. Experimental instruments.

https://doi.org/10.1371/journal.pone.0297756.g001

Experimental parameters

Parameter setting of CPMG sequence in T₂ relaxation test.

The soybeans were equilibrated to room temperature to measure the relaxation time T₂ using the LF-NMR analyzer to generate a Carr–Purcell–Meiboom–Gill (CPMG) pulse sequence. The measurement conditions were as follows. The 90-degree RF pulse width (P1) was set to 16 μs, the 180-degree RF pulse width (P2) was set to 33.04 μs, the echo time (TE) was set to 0.200 ms, the echo number (NECH) was set to 18000, and the repeated scanning time (NS) was set to 8. The T₂-FitFrm software was used to execute the fitting of T₂ values.

T₁ weighted imaging parameter setting of Low-field MRI.

MRI was performed with the same LF-NMR analyzer and the Inversion Recovery (IR) sequence. The soybean was placed in the center of the radio frequency (RF) coil to collect the signal and obtain a T₁-weighted image. The main parameters were configured as follows. The sampling repetition time (TR) was set to 500 ms, the echo time (TE) was set to 20 ms, the matrix size was set to 192 × 256, and the field of view (Fovx and Fovy) was set to 80 mm.

Experimental principles and methods

Sample preparation.

A meticulous sample preparation process was undertaken to establish a solid foundation for the subsequent experimental investigations. The following steps outline the comprehensive sample preparation process conducted for this study:

Experimental Design: The experiment incorporated the establishment of a control group and a experimental group.

Control Group: The control group consists of two Petri dishes, labeled #1 and #2. Each dish contains one soybean. Subsequently, distilled water was introduced to immerse the soybeans thoroughly.

Experimental Group: The experimental group comprises two sets of Petri dishes, labeled #1 and #2. The #1 Petri dishes were designated for the analysis of spin-spin relaxation time (T₂), while the #2 Petri dishes were allocated for nuclear magnetic resonance T₁ weighted imaging. In #1, there were 22 Petri dishes, and in #2, there were 13 Petri dishes. Each dish contains one soybean immersed in distilled water, allowing the soybeans to soak thoroughly in water at 25 degrees Celsius. In dishes 1-12 of #1, the soaking time intervals were 20 minutes, while in dishes 13-22 of #1, the soaking time intervals were 30 minutes. The soaking time intervals for soybeans in the dishes of #2 were 0.5 hours.

Experimental principles.

Soybeans belong to a relatively intricate multi-component system, and within the context of the soybean infiltration process, water manifests itself in at least three distinct forms within the soybeans: bound water, sub-bound water, and free water. NMR methodology quantifies the relaxation time T₂ associated with these water phases, thereby facilitating the observation of their respective binding states. Variations in relaxation time T₂ can elucidate the mobility patterns of water molecules, thus providing information on the migratory dynamics of water within soybeans [32].

To accurately ascertain the authentic T₂ relaxation characteristics of soybean infiltration through magnetic resonance signals, it is essential to endeavor to mitigate the impact of external constant magnetic field inhomogeneity on T₂ relaxation during measurements. This objective is achieved through the employment of the Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence, which is explicitly designed to counteract this influence.

In the context of this experiment, the analysis relies upon the utilization of the three-component model inherent in the multiple exponential decay framework of the CPGM pulse sequence [33]. (1) where T₂₁, T₂₂ and T₂₃ signify the spin-spin relaxation time of three distinct components, while A₁, A₂ and A₃ denote the signal amplitudes of these respective components at time t. Furthermore, A represents the overall signal amplitude at time t. The magnitude of the relaxation time T₂ signifies the intensity of water fluidity, as depicted in Fig 2.

Download:

Fig 2. Distribution of spin-spin relaxation time T₂ of three-component model.

https://doi.org/10.1371/journal.pone.0297756.g002

Employing the multi-component model for analyzing the relaxation behavior of protons in samples facilitates the segregation of signal amplitudes and relaxation times of protons about distinct components. Increased water fluidity corresponds to greater water molecule mobility, characterized by higher motion frequencies surpassing the resonance frequency of hydrogen protons. Consequently, this leads to the extension of relaxation time.

Conversely, reduced water flowability restrains water molecules due to their close association with hydrophilic macromolecules. As a consequence, these molecules exhibit slower movement rates, that approximate the Larmor frequency, culminating in shortened relaxation times.

Low-field NMR method

Testing the T₂ relaxation of the soybeans.

The measurement of soybean T₂ relaxation time is a pivotal experiment within the scope of this study. This test aims to delve into the dynamic changes in moisture distribution within soybeans using LF-NMR technology [34]. The experimental procedure is as follows:

Step 1: Using the CPMG sequence [35] within the NMR analysis software, retrieve the T₂ value of soybeans within the #1 Petri dish of the control group.
Step 2: After varying the duration of soybean infiltration in the #1 Petri dish of the experimental group, measure the T₂ value, proton density value, and signal intensity value.

For infiltration durations of up to 4 hours, conduct measurements at 20-minute intervals; for infiltration periods ranging from 4 to 9 hours, perform measurements at 30-minute intervals. For soybeans of different Petri dishes in the control group and experimental group, the measurement of T₂ relaxation time was repeated three times, and the average value was reported.

Low-field MRI T₁ weighted imaging of the soybeans.

Utilizing low-field MRI T₁-weighted imaging to dynamically observe changes in moisture distribution within soybean seeds, thus illustrating the dynamic process of soybean imbibition. An overview of the experimental procedure is provided as follows:

Step 1: Conduct low-field MRI on soybeans within the Petri dish # 2 of the control group to acquire T₁-weighted images.
Step 2: After varying durations of soybean infiltration in the #2 Petri dish of the experimental group, transfer the soaked soybeans into test tubes and subject them to low-field MRI using the NMR instrument. Capture T₁-weighted images at intervals of 0.5 hours throughout the soybean infiltration process.

Results and discussion

Soybean infiltration denotes the hydrating process of soybeans, typically encompassing two distinct stages [36].

The initial stage involves swelling and water absorption, primarily reliant on soybean colloid and unrelated to soybean metabolism. Within this phase, soybean colloid transitions from a gel-like state to a sol state through water absorption and swelling. This transformation facilitates the extension and restoration of compromised organelles and inactivated polymers present in desiccated soybeans.

The subsequent stage entails gradual water absorption. Following the rapid water uptake in the initial stage, soybean hydration reaches a near-saturation point, leading to heightened cellular expansion pressure. This pressure impedes further water absorption by the cells. Consequently, this phase witnesses the principal metabolic activities of soybeans.

According to NMR principles, free water molecules exhibit a considerably extended T₂ relaxation time, spanning several hundred milliseconds. Their diminutive size allows for swift movement, significantly surpassing the resonant frequency of ¹H. Conversely, bound water congregates around hydrophilic macromolecules, constrained by these macromolecules and thus exhibiting sluggish motion. Its motion frequency aligns closely with the Larmor frequency, expediting relaxation and resulting in a notably smaller T₂ value compared to that of free water.

The relationship between the relaxation time T₂ of various water-binding states can be summarized as follows: (2)

The relaxation time T₂, proton density value, and the signal intensity value of different water phases at various time points are determined by observation of the state of the soybean water and changes over time. Subsequently, the ratio of proton density to signal intensity denoted as the A₂ value, is calculated, and the results are presented in Table 1.

Download:

Table 1. Changes of relaxation time T₂ and proportion A₂ of three-phase water soaked in soybeans for different periods of time.

https://doi.org/10.1371/journal.pone.0297756.t001

Fig 3 illustrates the T₂ relaxation curve of soybeans. The x-axis represents relaxation time, while the y-axis denotes the relative water content in distinct stages. The highest point on each peak signifies the T₂ relaxation, and the area beneath the peak signifies the content of the relaxation component. Each peak corresponds to a specific type of water, thereby showcasing the evolving water migration within soybeans during the soaking process at varying immersion durations.

Download:

Fig 3. T₂ relaxation diagram of soybean infiltration process.

https://doi.org/10.1371/journal.pone.0297756.g003

As depicted in Fig 3, it is evident that before infiltration, soybeans predominantly contain bound water, as indicated by the T₂₁ absorption peak. Following the infiltration process, the presence of free water rises, signifying the infiltration of water into soybean tissue. With time during infiltration, the volume of flowing water augments, leading to the emergence of two distinct water peaks (T₂₂, T₂₃) within the soybeans. These peaks correspond to sub-bound water and free water, respectively, and exhibit a shift toward longer relaxation times as the duration of infiltration is extended.

Analysis of water binding state during soybean infiltration

The data presented in Table 1 reveals that water molecules during soybean infiltration undergo interaction with soybean macromolecules, resulting in the formation of three distinct water phases: bound water, sub-bound water, and free water. These phases exhibit varying relaxation time rates and fluidity characteristics.

The spin-spin relaxation time for bound water, characterized by limited fluidity, is denoted as T₂₁ and falls within the 0.3 to 1.2 milliseconds range. Sub-bound water, exhibits higher fluidity than bound water, with a spin-spin relaxation time denoted as T₂₂, which ranges between 10ms and 40ms. The most favorable fluidity is observed in free water, with a spin-spin relaxation time represented as T₂₃, ranging from 100 ms to 1000 ms.

Analysis of T₂₁ curve change.

Fig 4 illustrates the variation in the relaxation time T₂ associated with bound water throughout the soybean infiltration process. As depicted in Fig 4, T₂₁ exhibits minimal discernible alteration. T₂₁ represents the bound water that is intimately associated with the macromolecules within soybeans and is the residual water following soybean drying. During the 300-minute soybean soaking period, T₂₁ experiences intermittent fluctuations while maintaining stability, exhibiting an overall ascending trajectory, the value of T₂₁ is in the range of 0.3 to 0.47 ms. Upon the completion of 400 minutes of soybean soaking, T₂₁ experiences a sudden upsurge, increasing from 0.57 ms to 1.15 ms. This phenomenon signifies the initiation of macromolecule synthesis and transformation within soybeans, accompanied by a reduction in the firmness of the binding between bound water and macromolecules.

Download:

Fig 4. Changes of T₂₁ during soybean infiltration.

https://doi.org/10.1371/journal.pone.0297756.g004

Analysis of T₂₂ and T₂₃ curve change.

The observations from Figs 5 and 6 reveal that during the soybean infiltration process, the ingress of water molecules into the soybeans manifests as free water, whereas water molecules binding with hydrophilic macromolecules within soybeans manifest as sub-bound water. When the soaking duration is below 200 minutes, both the relaxation time T₂₂ of sub-bound water and T₂₃ of free water exhibit substantial increments, T₂₂ increased from 7.4 ms to 19.56 ms, while T₂₃ increased from 98.85 ms to 361.23 ms. This phenomenon signifies the commencement of soybean hull softening and water absorption, inducing soybean activation and breaking the dormancy phase. Subsequently, approximately after 400 minutes of soaking, concurrent with soybean activation, water begins to participate in the activation of existing enzymes or the synthesis of novel enzymes. The fluidity of water in soybean is enhanced, which is reflected by the evident elevation in T₂₂ and T₂₃, T₂₂ increased from 27.05 ms to 29.33 ms, while T₂₃ increased from 424.76 ms to 499.45 ms. Beyond 500 minutes of soaking, T₂₂ and T₂₃ stabilize, maintaining at 34.49 ms and 587.28 ms, respectively, indicating the cessation of soybean activation. This milestone suggests that the soybeans have undergone thorough infiltration.

Download:

Fig 5. Changes of T₂₂ during soybean infiltration.

https://doi.org/10.1371/journal.pone.0297756.g005

Download:

Fig 6. Changes of T₂₃ during soybean infiltration.

https://doi.org/10.1371/journal.pone.0297756.g006

Table 2 illustrates the application of linear regression analysis, utilizing T₂₂ (ms) as the independent variable and A₂₂ (%) as the dependent variable. As depicted in Table 2, the resultant model equation is: A₂₂ (%) = 42.364 + 1.906 × T₂₂ (ms), and the corresponding model R-square value stands at 0.431. This signifies that T₂₂ (ms) is capable of elucidating approximately 43.1(%) of the variation in A₂₂ (%). Subsequently, subjecting the model to an F-test reveals its successful compliance with the test (F = 15.908, p = 0.001 < 0.05), indicative of the influential role of T₂₂(ms) on A₂₂ (%). Further detailed analysis corroborates this finding: the regression coefficient of T₂₂ (ms) stands at 1.906 (t = 3.989, p = 0.001 < 0.01), signifying a statistically significant positive impact of T₂₂ (ms) on A₂₂ (%). These results collectively assert that all instances of T₂₂ (ms) exert a notable and positive influence on A₂₂ (%).

Download:

Table 2. T₂₂ liner regression analysis results.

https://doi.org/10.1371/journal.pone.0297756.t002

Table 3 presents the result of a linear regression analysis employing T₂₃ (ms) as the independent variable and A₂₃ (%) as the dependent variable. As indicated in Table 3, the resultant model equation takes the form: A₂₃ (%) = 45.472 − 0.099 × T₂₃ (ms), accompanied by a model R-square value of 0.647. This signifies that T₂₃ (ms) possesses the capacity to expound upon approximately 64.7(%) of the variability observed in A₂₃ (%). Subsequently, subjecting the model to an F-test corroborates its successful passage (F = 38.463, p = 0.000 < 0.05), affirming the impact of T₂₃ (ms) on A₂₃ (%). Detail analysis further confirms this result: The regression coefficient of T₂₃ (ms) stands at -0.099 (t = -6.202, p = 0.000 < 0.01), signifying a statistically significant positive influence of T₂₃ (ms) on A₂₃ (%). These findings collectively establish that all instances of T₂₃ (ms) wield a significant and positive influence on A₂₃ (%).

Download:

Table 3. T₂₃ liner regression analysis results.

https://doi.org/10.1371/journal.pone.0297756.t003

Observations and statistical analyzes derived from alterations in the T₂₂ and T₂₃ curves reveal a significant correlation between T₂ variations and the fluidity of water molecules. This correlation aids in comprehending the intricate migration dynamics of water molecules within soybeans. The utilization of T₂ measurement through low-field NMR constitutes a direct and effective approach to gauging the fluidity of water molecules.

A₂ and the variation of water ratios in three phases

Throughout the process of soybean infiltration, water migration unfolds as a complex phenomenon. Water molecules ingress the soybeans and initiate interactions with macromolecules, notably proteins. This interaction gives rise to three distinct water phases: bound water, sub-bound water, and free water. Through the analysis of the A₂ parameter, insights into the temporal evolution of water ratios across these phases can be garnered. This analytical approach facilitates the determination of the optimal soaking duration required for soybeans to achieve the highest water absorption efficiency.

Fig 7 illustrates the temporal variations in the A₂ water ratio for the three distinct phases. The following observations can be made.

(1) Initial Dry State: Initially, when soybeans are in a dry state, the proportion of bound water (A₂₁) is approximately 36%, sub-bound water (A₂₂) is around 0.8%, and free water (A₂₃) constitutes approximately 62% of the total water content. This composition suggests that water in soybeans predominantly consists of bound and free water in this dry state.
(2) Early Infiltration Phase: In the early stages of soybean infiltration, both A₂₁ and A₂₃ experience rapid decreases within the first 180 minutes. Specifically, A₂₁ decreases to 2.48% after 80 minutes, while A₂₃ drops to 9% at the 100 minute mark. Simultaneously, A₂₂ exhibits a swift increase, reaching 87% after 100 minutes of soaking. This shift in water distribution underscores the cessation of dormancy in soybeans during this time.
(3) Continued Infiltration Phase: During the continuous soak of soybeans for 180 to 420 minutes, A₂₂ remains stable, maintaining at 90.3% to 91.42%, A₂₁ shows a gradual upward trajectory, increasing from 5.85% to 6.93%, and A₂₃ experiences a slow decline, decreasing from 3.85% to 1.65%. This stage coincides with the soybeans actively absorbing water and initiating metabolic activities, such as activation.
(4) Germination Phase: Beyond 420 minutes of soaking, the proportions of water in the three phases stabilize. At this juncture, the preparatory activities prior to soybean germination have concluded, manifesting the onset of the germination process.

Download:

Fig 7. The curve of the ratio A₂ changing with time.

https://doi.org/10.1371/journal.pone.0297756.g007

T₁ weighted image analysis of Low-field MRI

Low-field MRI scans are conducted on the soybeans immersed in the #2 Petri dish within the experimental group for varying time intervals. T₁-weighted images are acquired at 30-minute intervals, as depicted in Fig 8.

Download:

Fig 8. Soybean infiltration imaging.

https://doi.org/10.1371/journal.pone.0297756.g008

Free water consists of smaller water molecules that exhibit a heightened thermal motion frequency, leading to diminished energy exchange efficiency between protons and their surrounding environment. The proton motion frequency in free water is lower than the Larmor precession frequency, thereby resulting in an extended spin-lattice relaxation time (T₁). Biological macromolecules, such as proteins, undergo gradual movement, and their proton motion frequency is significantly removed from the system’s resonance frequency, consequently exhibiting a lengthier T₁. Nonetheless, the differential between proton motion frequency and the Larmor precession frequency in free water is more substantial than that of biological macromolecules, culminating in an extended T₁ for free water in contrast to larger molecules.

In the context of cholesterol and adipose tissue, their vibration frequency aligns with the typical MRI field strength’s Larmor precession frequency, resulting in markedly abbreviated T₁ times. Bound water’s motion frequency closely corresponds to the Larmor precession frequency, thereby conferring a notably truncated T₁ for bound water as well. Employing low-field MRI to analyze different T₁ tissues yields Free Induction Decay (FID) signals of varying intensities, which manifest distinct grayscale distinctions in T₁-weighted images. Tissues characterized by prolonged T₁ engage in a sluggish longitudinal magnetization recovery, culminating in diminished signal intensity and generating darker images.

In the experiment, the soybean samples subjected to nuclear magnetic resonance imaging comprise three distinct constituents: soaked free water, soybean macromolecular components (like proteins, starch, etc.), and bound water absorbed by the soybeans. The T₁ of bound water is exceptionally brief, causing the magnetization vector to pivot toward the xOy plane post-90-pulse excitation, resulting in the most substantial amplitude and generating the strongest FID signal. Consequently, this segment appears the brightest within MRI images. Conversely, free water exhibits the longest T₁, leading the magnetization vector after a 90-pulse excitation to align minimally with the xOy plane, subsequently yielding the feeblest FID signal and presenting as a darker portion within the MRI image. The T₁ of macromolecular substances in soybeans is proximate to the pulse repetition time (TR), resulting in an FID signal intensity lower than that of free water. This material manifests as a blackened region in the MRI image.

In Fig 8, the region of greatest brightness corresponds to bound water absorbed by the soybeans, the relatively brighter area corresponds to free water, and the darkest segment represents the presence of macromolecular substances. Observing Fig 8(1) and 8(2), it becomes evident that free water enters the soybeans via the navel of the soybean hull. Fig 8(3)–8(9) depict a discernible increase in both the total water content and the proportion of bound water during the soybean infiltration process. As the infiltration time progresses, water permeates deeper into the inner layers, and the T₁-weighted image vividly displays a state of wetness. Notably, Fig 8(10) portrays that at the 5.0-hour mark of wetting, moisture permeates and accumulates within the interior, indicative of a diffusion trend. Furthermore, Fig 8(13) illustrates that after soaking for 6.5 hours, the soybeans commence a gradual saturation process, moving towards germination.

Conclusion

By analyzing the LF-NMR signals, this study approached the soybean infiltration process from a novel perspective, unraveling the intricate binding dynamics of water within soybeans. The T₂ relaxation spectrum unveiled three distinct peaks, signifying the segmentation of soybean internal water into three categories: bound water, sub-bound water, and free water. During the initial 0-180 minutes of infiltration, both the proportion of bound water and sub-bound water exhibited a rapid decline, indicative of external water infiltration into the soybean. As the infiltration time progressed to 180-420 minutes, the proportion of sub-bound water stabilized, while the proportion of bound water underwent a gradual increase, and the proportion of free water showed a gradual decrease. This phase marked the commencement of soybean water absorption, accompanied by activation and metabolic activities. With infiltration time surpassing 420 minutes, the proportions of the three water phases remained stable, denoting the completion of preparatory activities before germination for the soybeans.

The utilization of low-field MRI offers a rapid, non-invasive, and dynamic approach to observing the internal water uptake of soybeans. This technique proves valuable in comprehensively capturing water migration dynamics during the infiltration process and dying process, elucidating the routes of mass transfer within the bean-water system, and discerning the intricate morphological configuration of soybeans. LF-NMR serves as an expeditious and non-destructive visual tool, furnishing insights into the soybean infiltration process. Moreover, it enables the determination of distinct components’ water content based on differing fluidity, thereby furnishing a scientific foundation for controlling the critical point of moisture in production practice and establishing a theoretical basis for determining the target moisture content during the soybean drying process.

This study has achieved success by sufficiently exploring the moisture distribution during the dried soybean infiltration processes, but there is still room for improvement. First, there are differences in moisture content and distribution of soybeans at different ages. Second, soybeans’ moisture content and distribution at different temperatures are also different. To this end, we will carry out the future works on studying the soybean infiltration process from two aspects. First, we will collect diverse soybean samples from different ages to explore the potential relationship between the moisture content and distribution of soybeans and soybean age. Second, we will immerse soybean samples of the same age in distilled water at different temperatures to explore the effect of temperature on the moisture content and distribution of soybeans.

Acknowledgments

Suzhou Niumag Analytical Instrument Corporation provides equipment and technical support for the research project.

References

1. Wu Q. Nutritional value of soybean and its processing value. Modern Food. 2022;(09):110–112.
- View Article
- Google Scholar
2. Toomer OT, Oviedo EO, Ali M, Patino D, Joseph M, Frinsko M, et al. Current agronomic practices, harvest & post-harvest processing of soybeans (glycine max)—a review. Agronomy. 2023;13(2):427.
- View Article
- Google Scholar
3. Chen TY, Zhang WY, Liu YX, Song YQ, Wu LY, Liu CH, et al. Water status and predictive models of moisture content during drying of soybean dregs based on LF-NMR. Molecules. 2022;27(14):4421. pmid:35889294
- View Article
- PubMed/NCBI
- Google Scholar
4. Fu XJ, Xia SQ. Effect of soaking on the quality of house-made soymilk. In: 2011 International Conference on New Technology of Agricultural. IEEE; 2011. p. 878–881.
5. Lu JT, Cheng JH, Xu YY, Chen YJ, Qian K, Zhang Y. Effect of germination on nutritional quality of soybean. Food Science and Technology. 2023;43:e008323.
- View Article
- Google Scholar
6. Lin Sy, Yang Sl, Li Xf, Chen F, Zhang Md. Dynamics of water mobility and distribution in soybean antioxidant peptide powders monitored by LF-NMR. Food Chem. 2016;199:280–286. pmid:26775972
- View Article
- PubMed/NCBI
- Google Scholar
7. Xu MK, Xu Y. Optimization of process conditions for the preparation of semi-finished soybean pulp using freeze drying method. Jiangsu Agricultural Sciences. 2013;41(6):216–219.
- View Article
- Google Scholar
8. Susilo B, Hermanto M, Damayanti R, Putri E. Performance of dehumidifier drying machine for soybean seeds drying. IOP Conf Ser Earth Environ Sci. 2022;1059(1):012020.
- View Article
- Google Scholar
9. Wang LP, Xu Y, Zhong YQ, Shen C, Li BG. Development of soaking-exempt soybean by freeze drying. Food Science. 2010;(22):521–524.
- View Article
- Google Scholar
10. Gong X, Gao JX, Zhang Y. Freeze-drying technology of quick-cooked mung bean. The Food Industry. 2019;40(09):77–79.
- View Article
- Google Scholar
11. Hua ZZ, Liu BL, Zuo JG. Freezed drying of pharmaceutical and food products. Science Press; 2007.
12. Li JL, Chen M, Cheng DH, Li W, Guo SW, Chen C, et al. An accurate and effective single-seed moisture selection method based on nuclear magnetic resonance (NMR) in maize. Comput Electron Agric. 2022;193:106649.
- View Article
- Google Scholar
13. Ramachandran RP, Erkinbaev C, Thakur S, Paliwal J. Three dimensional characterization of micronized soybean seeds using X-ray microtomography. Food and Bioproducts Processing. 2021;127:388–397.
- View Article
- Google Scholar
14. Zhu ZY, Chen SB, Wu XY, Xing CR, Yuan J. Determination of soybean routine quality parameters using near-infrared spectroscopy. Food Science & Nutrition. 2018;6(4):1109–1118. pmid:29983975
- View Article
- PubMed/NCBI
- Google Scholar
15. Zhang Y, Zhao Jx, Gu Y, Zhang Y, Chen Y, Song P, et al. A methodology study on the optimal detection of oil and moisture content in soybeans using LF-NMR and its 2D T₁-T₂ nuclear magnetic technology. Agronomy. 2023;13(4):1102.
- View Article
- Google Scholar
16. Arnold TC, Freeman C, Litt B, Stein J. Low-field MRI: Clinical promise and challenges. J Magn Reson Imaging. 2023;57(1):25–44. pmid:36120962
- View Article
- PubMed/NCBI
- Google Scholar
17. Hatta J, Miyamoto M, Adachi Y, Kawai J, Uehara G, Kado H. SQUID-based low field MRI system for small animals. IEEE Trans Appl Supercond. 2011;21(3):526–529.
- View Article
- Google Scholar
18. Chen FL, Wei YM, Zhang B. Characterization of water state and distribution in textured soybean protein using DSC and NMR. J Food Eng. 2010;100(3):522–526.
- View Article
- Google Scholar
19. Zhao XP. Magnetic resonance imaging. Science Press; 2004.
20. Zhao XP. Principle, equipment and applications of magnetic resonance imaging. Science Press; 2004.
21. Michal C. Low-cost low-field NMR and MRI: Instrumentation and applications. J Magn Reson. 2020;319:106800. pmid:33036708
- View Article
- PubMed/NCBI
- Google Scholar
22. Demachi K, Hayashi K, Adachi S, Tanabe K, Tanaka S. T₁-weighted image by ultra-low field SQUID-MRI. IEEE Trans Appl Supercond. 2019;29(5):1–5.
- View Article
- Google Scholar
23. Zhu JY, Li SS, Yang L, Zhao ZR, Xia JL, Zhu YY, et al. Effect of multiple freeze-thaw cycles on water migration, protein conformation and quality attributes of beef longissimus dorsi muscle by real-time low field nuclear magnetic resonance and Raman spectroscopy. Food Res Int. 2023;166:112644. pmid:36914334
- View Article
- PubMed/NCBI
- Google Scholar
24. Cheng SS, Zhang T, Yao L, Wang XH, Song YK, Wang HH, et al. Use of low-field-NMR and MRI to characterize water mobility and distribution in pacific oyster (Crassostrea gigas) during drying process. Dry Technol. 2018;36(5):630–636.
- View Article
- Google Scholar
25. Younas S, Mao Y, Liu C, Murtaza MA, Ali Z, Wei L, et al. Measurement of water fractions in freeze-dried shiitake mushroom by means of multispectral imaging (MSI) and low-field nuclear magnetic resonance (LF-NMR). J Food Compost Anal. 2021;96:103694.
- View Article
- Google Scholar
26. Tan MQ, Lin ZY, Zu YX, Zhu BW, Cheng SS. Effect of multiple freeze-thaw cycles on the quality of instant sea cucumber: Emphatically on water status of by LF-NMR and MRI. Food Res Int. 2018;109:65–71. pmid:29803493
- View Article
- PubMed/NCBI
- Google Scholar
27. Gibson M, Percival BC, Edgar M, Grootveld M. Low-field benchtop NMR spectroscopy for quantification of aldehydic lipid oxidation products in culinary oils during shallow frying episodes. Foods. 2023;12(6):1254. pmid:36981180
- View Article
- PubMed/NCBI
- Google Scholar
28. Castillo A, Rueda-Chacón H, Agudelo JL, Molina D. Prediction of the stability and compatibility of Colombian heavy crude oils by 1D low field nuclear magnetic resonance relaxometry and chemometric methods. Fuel. 2021;298:120721.
- View Article
- Google Scholar
29. Jin HQ, Tu LY, Wang YX, Zhang KX, Lv BW, Zhu Z, et al. Rapid detection of waste cooking oil using low-field nuclear magnetic resonance. Food Control. 2023;145:109448.
- View Article
- Google Scholar
30. Peng J, Ye LF, Wu MM, Wu MH, Ma ZG, Cao H, et al. Evaluation of processing mechanism in Astragali Radix by low-field nuclear magnetic resonance and magnetic resonance imaging. PLoS ONE. 2022;17(3):e0265383. pmid:35286357
- View Article
- PubMed/NCBI
- Google Scholar
31. Yu XL, Wang ZH, Zhang YQ, Wadood SA, Wei YM. Study on the water state and distribution of Chinese dried noodles during the drying process. J Food Eng. 2018;233:81–87.
- View Article
- Google Scholar
32. He Y, Shim YY, Shen JH, Kim JH, Cho JY, Hong WS, et al. Aquafaba from Korean Soybean II: Physicochemical properties and composition characterized by NMR analysis. Foods. 2021;10(11):2589. pmid:34828870
- View Article
- PubMed/NCBI
- Google Scholar
33. Ruan RS. Application of magnetic resonance technology in food and biological systems. China Light Industry Press; 2009.
34. Gu Y, Chen Y, Yue X, Xiong P, Pan DY, Song P, et al. LF-NMR/MRI determination of different 6-benzylaminopurine concentrations and their effects on soybean moisture. Front Plant Sci. 2022;13:885804. pmid:35519819
- View Article
- PubMed/NCBI
- Google Scholar
35. Liu J, Feng XY, Wang DS. Determination of water content in crude oil emulsion by LF-NMR CPMG sequence. Pet Sci Technol. 2019;37:1123–1135.
- View Article
- Google Scholar
36. Wu T. The development of immersion law and effect on soybean. China Condiment. 2012;37(5):20–22.
- View Article
- Google Scholar

[ref1] 1. Wu Q. Nutritional value of soybean and its processing value. Modern Food. 2022;(09):110–112.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Toomer OT, Oviedo EO, Ali M, Patino D, Joseph M, Frinsko M, et al. Current agronomic practices, harvest & post-harvest processing of soybeans (glycine max)—a review. Agronomy. 2023;13(2):427.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Chen TY, Zhang WY, Liu YX, Song YQ, Wu LY, Liu CH, et al. Water status and predictive models of moisture content during drying of soybean dregs based on LF-NMR. Molecules. 2022;27(14):4421. pmid:35889294
View Article
PubMed/NCBI
Google Scholar

[8] View Article

[9] PubMed/NCBI

[10] Google Scholar

[ref4] 4. Fu XJ, Xia SQ. Effect of soaking on the quality of house-made soymilk. In: 2011 International Conference on New Technology of Agricultural. IEEE; 2011. p. 878–881.

[ref5] 5. Lu JT, Cheng JH, Xu YY, Chen YJ, Qian K, Zhang Y. Effect of germination on nutritional quality of soybean. Food Science and Technology. 2023;43:e008323.
View Article
Google Scholar

[13] View Article

[14] Google Scholar

[ref6] 6. Lin Sy, Yang Sl, Li Xf, Chen F, Zhang Md. Dynamics of water mobility and distribution in soybean antioxidant peptide powders monitored by LF-NMR. Food Chem. 2016;199:280–286. pmid:26775972
View Article
PubMed/NCBI
Google Scholar

[16] View Article

[17] PubMed/NCBI

[18] Google Scholar

[ref7] 7. Xu MK, Xu Y. Optimization of process conditions for the preparation of semi-finished soybean pulp using freeze drying method. Jiangsu Agricultural Sciences. 2013;41(6):216–219.
View Article
Google Scholar

[20] View Article

[21] Google Scholar

[ref8] 8. Susilo B, Hermanto M, Damayanti R, Putri E. Performance of dehumidifier drying machine for soybean seeds drying. IOP Conf Ser Earth Environ Sci. 2022;1059(1):012020.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref9] 9. Wang LP, Xu Y, Zhong YQ, Shen C, Li BG. Development of soaking-exempt soybean by freeze drying. Food Science. 2010;(22):521–524.
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref10] 10. Gong X, Gao JX, Zhang Y. Freeze-drying technology of quick-cooked mung bean. The Food Industry. 2019;40(09):77–79.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref11] 11. Hua ZZ, Liu BL, Zuo JG. Freezed drying of pharmaceutical and food products. Science Press; 2007.

[ref12] 12. Li JL, Chen M, Cheng DH, Li W, Guo SW, Chen C, et al. An accurate and effective single-seed moisture selection method based on nuclear magnetic resonance (NMR) in maize. Comput Electron Agric. 2022;193:106649.
View Article
Google Scholar

[33] View Article

[34] Google Scholar

[ref13] 13. Ramachandran RP, Erkinbaev C, Thakur S, Paliwal J. Three dimensional characterization of micronized soybean seeds using X-ray microtomography. Food and Bioproducts Processing. 2021;127:388–397.
View Article
Google Scholar

[36] View Article

[37] Google Scholar

[ref14] 14. Zhu ZY, Chen SB, Wu XY, Xing CR, Yuan J. Determination of soybean routine quality parameters using near-infrared spectroscopy. Food Science & Nutrition. 2018;6(4):1109–1118. pmid:29983975
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref15] 15. Zhang Y, Zhao Jx, Gu Y, Zhang Y, Chen Y, Song P, et al. A methodology study on the optimal detection of oil and moisture content in soybeans using LF-NMR and its 2D T₁-T₂ nuclear magnetic technology. Agronomy. 2023;13(4):1102.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref16] 16. Arnold TC, Freeman C, Litt B, Stein J. Low-field MRI: Clinical promise and challenges. J Magn Reson Imaging. 2023;57(1):25–44. pmid:36120962
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref17] 17. Hatta J, Miyamoto M, Adachi Y, Kawai J, Uehara G, Kado H. SQUID-based low field MRI system for small animals. IEEE Trans Appl Supercond. 2011;21(3):526–529.
View Article
Google Scholar

[50] View Article

[51] Google Scholar

[ref18] 18. Chen FL, Wei YM, Zhang B. Characterization of water state and distribution in textured soybean protein using DSC and NMR. J Food Eng. 2010;100(3):522–526.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref19] 19. Zhao XP. Magnetic resonance imaging. Science Press; 2004.

[ref20] 20. Zhao XP. Principle, equipment and applications of magnetic resonance imaging. Science Press; 2004.

[ref21] 21. Michal C. Low-cost low-field NMR and MRI: Instrumentation and applications. J Magn Reson. 2020;319:106800. pmid:33036708
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref22] 22. Demachi K, Hayashi K, Adachi S, Tanabe K, Tanaka S. T₁-weighted image by ultra-low field SQUID-MRI. IEEE Trans Appl Supercond. 2019;29(5):1–5.
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref23] 23. Zhu JY, Li SS, Yang L, Zhao ZR, Xia JL, Zhu YY, et al. Effect of multiple freeze-thaw cycles on water migration, protein conformation and quality attributes of beef longissimus dorsi muscle by real-time low field nuclear magnetic resonance and Raman spectroscopy. Food Res Int. 2023;166:112644. pmid:36914334
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref24] 24. Cheng SS, Zhang T, Yao L, Wang XH, Song YK, Wang HH, et al. Use of low-field-NMR and MRI to characterize water mobility and distribution in pacific oyster (Crassostrea gigas) during drying process. Dry Technol. 2018;36(5):630–636.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref25] 25. Younas S, Mao Y, Liu C, Murtaza MA, Ali Z, Wei L, et al. Measurement of water fractions in freeze-dried shiitake mushroom by means of multispectral imaging (MSI) and low-field nuclear magnetic resonance (LF-NMR). J Food Compost Anal. 2021;96:103694.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref26] 26. Tan MQ, Lin ZY, Zu YX, Zhu BW, Cheng SS. Effect of multiple freeze-thaw cycles on the quality of instant sea cucumber: Emphatically on water status of by LF-NMR and MRI. Food Res Int. 2018;109:65–71. pmid:29803493
View Article
PubMed/NCBI
Google Scholar

[75] View Article

[76] PubMed/NCBI

[77] Google Scholar

[ref27] 27. Gibson M, Percival BC, Edgar M, Grootveld M. Low-field benchtop NMR spectroscopy for quantification of aldehydic lipid oxidation products in culinary oils during shallow frying episodes. Foods. 2023;12(6):1254. pmid:36981180
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref28] 28. Castillo A, Rueda-Chacón H, Agudelo JL, Molina D. Prediction of the stability and compatibility of Colombian heavy crude oils by 1D low field nuclear magnetic resonance relaxometry and chemometric methods. Fuel. 2021;298:120721.
View Article
Google Scholar

[83] View Article

[84] Google Scholar

[ref29] 29. Jin HQ, Tu LY, Wang YX, Zhang KX, Lv BW, Zhu Z, et al. Rapid detection of waste cooking oil using low-field nuclear magnetic resonance. Food Control. 2023;145:109448.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref30] 30. Peng J, Ye LF, Wu MM, Wu MH, Ma ZG, Cao H, et al. Evaluation of processing mechanism in Astragali Radix by low-field nuclear magnetic resonance and magnetic resonance imaging. PLoS ONE. 2022;17(3):e0265383. pmid:35286357
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref31] 31. Yu XL, Wang ZH, Zhang YQ, Wadood SA, Wei YM. Study on the water state and distribution of Chinese dried noodles during the drying process. J Food Eng. 2018;233:81–87.
View Article
Google Scholar

[93] View Article

[94] Google Scholar

[ref32] 32. He Y, Shim YY, Shen JH, Kim JH, Cho JY, Hong WS, et al. Aquafaba from Korean Soybean II: Physicochemical properties and composition characterized by NMR analysis. Foods. 2021;10(11):2589. pmid:34828870
View Article
PubMed/NCBI
Google Scholar

[96] View Article

[97] PubMed/NCBI

[98] Google Scholar

[ref33] 33. Ruan RS. Application of magnetic resonance technology in food and biological systems. China Light Industry Press; 2009.

[ref34] 34. Gu Y, Chen Y, Yue X, Xiong P, Pan DY, Song P, et al. LF-NMR/MRI determination of different 6-benzylaminopurine concentrations and their effects on soybean moisture. Front Plant Sci. 2022;13:885804. pmid:35519819
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref35] 35. Liu J, Feng XY, Wang DS. Determination of water content in crude oil emulsion by LF-NMR CPMG sequence. Pet Sci Technol. 2019;37:1123–1135.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref36] 36. Wu T. The development of immersion law and effect on soybean. China Condiment. 2012;37(5):20–22.
View Article
Google Scholar

[108] View Article

[109] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Experimental materials

Experimental instruments

Experimental parameters

Parameter setting of CPMG sequence in T2 relaxation test.

T1 weighted imaging parameter setting of Low-field MRI.

Experimental principles and methods

Sample preparation.

Experimental principles.

Low-field NMR method

Testing the T2 relaxation of the soybeans.

Low-field MRI T1 weighted imaging of the soybeans.

Results and discussion

Analysis of water binding state during soybean infiltration

Analysis of T21 curve change.

Analysis of T22 and T23 curve change.

A2 and the variation of water ratios in three phases

T1 weighted image analysis of Low-field MRI

Conclusion

Acknowledgments

References

Parameter setting of CPMG sequence in T₂ relaxation test.

T₁ weighted imaging parameter setting of Low-field MRI.

Testing the T₂ relaxation of the soybeans.

Low-field MRI T₁ weighted imaging of the soybeans.

Analysis of T₂₁ curve change.

Analysis of T₂₂ and T₂₃ curve change.

A₂ and the variation of water ratios in three phases

T₁ weighted image analysis of Low-field MRI