Investigating the Binding Properties between Strontium and Dissolved Organic Matter under the Influence of pH and Ca²⁺ in a Typical Karst Area, China

Abstract

Due to the unique hydrogeological environment of karst areas, pollutants are more likely to enter the soil and water, showing a special migration and transformation behavior. In this work, the binding behaviors between strontium (Sr²⁺) and dissolved organic matter (DOM) extracted from soil under the influence of pH and Ca²⁺ in a typical karst area were investigated by applying three-dimensional fluorescence spectroscopy combined with parallel factor analysis (EEM–PARAFAC) and two-dimensional correlation analysis (2D-COS) of synchronous fluorescence spectra (SF). The results show that DOM extracted from soil was dominated by tryptophan-like and tyrosine-like materials (77% in total). Two-dimensional COS of SF showed that the tryptophan-like substance in DOM extracted from soil preferentially bound to Sr²⁺. When the pH was 7, the binding coefficient (logK_a) of the four DOM components ranged from 2.69 to 4.04, which was more conducive to the binding of DOM extracted from soil and Sr²⁺ than under acidic and alkaline conditions. Ca²⁺ in soil weakened the binding of DOM extracted from soil to Sr²⁺ by competing for binding sites and changing the molecular surface potential. This research is helpful for acknowledging the migration and transformation of Sr²⁺ and offers a reference for groundwater protection in karst areas.

1. Introduction

In the karstic dynamic system formed by the carbon cycle of “carbon dioxide, organic carbon and carbonate” coupled with the unbalanced open system of CO₂-H₂O-CO₃²⁻, matter and energy move in different directions, ways and intensities, producing various surface and underground karstic forms. Water, soil and rock are closely related in karst areas [1]. Due to the unique geological environment of karst areas, the soil is thin and rocky, and rocks such as carbonates are corrosive and permeable. Under strong karstification (dissolution, erosion), most atmospheric precipitation and surface water enter the groundwater [2]. Thus, karst groundwater usually contains a certain number of mineral salts, trace elements or carbon dioxide gas and is characterized by the physicochemical properties of being calcium-rich and partially alkaline [3]. According to statistics, one-quarter of the world’s population relies on karst groundwater for their living and production [4]. Strontium (Sr), as an important trace metal element, is generally enriched in rocks, soils, and sediments and accumulates in surface water and groundwater in the dissolved state through the process of weathering and dissolution [5,6]. As one of the trace essential elements for human bone growth, a proper level of strontium in the groundwater (a content less than 5 mg/L) also has certain curative effects on cardiovascular disease [7,8]. However, the long-term drinking of strontium-rich groundwater can cause kidney damage and abnormal bone development and osteomalacia in humans, especially in karst areas. Therefore, considering the higher Sr content in soil and rocks in karst areas, there is an urgent need for their remediation for the protection of groundwater resources.

The remediation process of different types of soils depends on their physical and chemical properties and environmental media, such as pH, oxidation-reduction potential (Eh), cation content, and organic matter content [9,10]. Hereinto, dissolved organic matter (DOM), as a kind of organic mixture with intricate functional groups and environmental behavior, exists widely in the soil and can impact the transplantation, conversion, and toxicity of contaminants [11]. Although DOM accounts for a small portion of soil organic matter, it is the most positive constituent directly involved in the chemical process of the soil ecological environment [12]. Generally, approximately 25% to 50% of DOM components comprise humic and fulvic acids, while the rest include proteins, polysaccharides, and hydrophilic organic acids [13,14]. The effects of DOM are as follows: on the one hand, the main functional groups of soil DOM, including hydroxyl, carboxyl, and amidogen, possess a high chemical activity and can promote the oxido reduction of inorganic pollutants such as chromium (Cr (VI)), arsenic (As (V)), and mercury (Hg (II)) [15]. On the other hand, it can not only bind to heavy metals through complexation, such as nickel (Ni), copper (Cu), and zinc (Zn), to inhibit the precipitation and adsorption of heavy metals, but also interact with soil minerals, which enhances the migration capacity of heavy metals [16]. Thus, DOM is a considerable element for the migration and transformation of Sr in karst soil.

Relevant studies have shown that the binding behavior of DOM and pollutants is affected by DOM attributes (e.g., components and functional groups), pollutant types, and the external hydrochemical environment (e.g., pH, temperature, and cations) [16,17,18]. Previous studies reported that the binding behavior effect of protein-like groups on Cu²⁺ is stronger than that of humic-like groups, and different humic-like groups have different affinities for Cu²⁺ and Hg²⁺ [19]. Due to the unique calcium-rich and alkali-inclined soil environment in karst areas, pH and Ca²⁺ become essential factors affecting the complex of DOM and Sr²⁺ [20,21]. Some studies have shown that under acidic and alkaline conditions, the content of fulvic-like and humic-like matter in soil DOM decreases, and protein-like substances and some functional groups disappear. The number of functional groups decreased to a greater extent under alkaline conditions than under acidic conditions. The optimal condition for the binding of Pb²⁺ or Sb(V) and DOM was pH 7.0 [22]. Protons and Ca²⁺ may reduce the effective functional groups of DOM binding to other metal ions through competition and inhibit the binding behavior [20]. Information on how pH and Ca²⁺ affect soil DOM and Sr complex behavior in karst areas is still insufficient and needs to be explored. This is of practical significance for karst soil pollution remediation and groundwater protection.

Fluorescence quenching, as a susceptible, nondestructive, and convenient method, has been widely employed to analyze the complexation of DOM with metal ions, such as copper (Cu), mercury (Hg), iron (Fe), and zinc (Zn) [22,23,24]. The principle of this method is that when the quenching agent is present, the emission energy of the fluorophore is suppressed, resulting in a decrease in the fluorescence peak intensity. Three-dimensional fluorescence spectroscopy combined with parallel factor analysis (EEM-PARAFAC) can extract the independent fluorescence components with minimum residuals from multiple EEM datasets, which is helpful to study the response of different fluorescence components to quenchers [17]. In addition, synchronous fluorescence spectroscopy combined with two-dimensional correlation spectroscopy (2D-COS) analysis can also afford more detailed information about the quenching effect [25,26].

In this study, Sr²⁺ was selected as the potential contaminant. The sampling area, Xintian County, Hunan Province, China, is a long-term field observation site of the Institute of Karst Geology, Chinese Academy of Geological Sciences, which has accumulated more than 10 years of field observation data. In previous studies, a large amount of surface water and groundwater was collected, and a preliminary understanding of the occurrence form, migration and transformation rule of DOM and Sr²⁺ in the sampling area was obtained [4,8,27]. On this basis, we selected and collected one soil sample from key stations to extract DOM and conducted a fluorescence quenching experiment. EEM–PARAFAC analysis and 2D-COS analysis of synchronous fluorescence spectra (SF) were employed to reveal the binding ability between DOM precipitated by soil leaching and Sr²⁺ and the influences of pH and Ca²⁺ on DOM–Sr interactions. This study is beneficial for deepening the understanding of the influence of DOM from soil leaching on Sr²⁺ migration and transformation in Sr-rich karst groundwater systems and provides a reference value for the remediation of heavy metal pollution in karst soil and the protection of groundwater.

2. Materials and Methods

2.1. Study Area and DOM Preparation

Xintian County is located in southern Hunan Province, China (112°04′–112°23′ E, 25°36′–26°06′ N) (Figure 1). This region is dominated by karst hills and karst dissolved hills with carbonate outcrops, distributing a large area of red loam, black lime soil and paddy soil. The strontium (Sr) content range of the soil was 67.3–115.72 mg/kg in this area, and the maximum content of Sr in the water-bearing rock group was 838 mg/kg, which has the potential risk of groundwater pollution [4,8]. The ranges of pH values and calcium ions (Ca²⁺) in karst groundwater were 6.91–8.11 and 4.7–160.7 mg/kg, respectively, characterizing a slightly alkaline property and Ca²⁺ enrichment of karst water [4].

The soil sample was collected in July 2017 from Xintian County, Hunan Province, China, and was freeze-dried, ground and passed through a 2 mm sieve (Figure 1). The soil sample was loaded into the plexiglass column and leached at velocities of 0.2, 0.4, 0.7, 1.5, 1.8, and 2.5 mL/min. The initial indexes (pH, Sr²⁺, and Ca²⁺ concentration) were measured by taking the exudate. The values of pH, Sr²⁺, and Ca²⁺ were 8.26 ± 0.16, 0.002 ± 0.01 mM, and 0.412 ± 0.08 mM in the soil sample, respectively. Forty grams of soil was added to an inorganic glass bottle containing 400 mL of Milli-Q water (solid–liquid ratio 1:10). Then, all liquid after mixing was shaken at 25 °C (rotate speed: 200 rpm) under dark conditions for 48 h. The hybrid liquid was decentered at 5000 rpm for 10 min, and the suspension fluid was filtered with 0.7 µm glass fiber filters (Waterman GF/F) to obtain the DOM extracted from the soil. The filtrate was dialyzed with dialysis bags with a pore size of 3500 Da for 24 h at 4 °C to remove inorganic salts and small organic compounds. The DOC concentration of the DOM stock solution was 57 mg/L. To eliminate the influence of the DOM concentration and internal filtration effect, the extracted DOM stock solution was diluted 5 times with Milli-Q water to a final DOC concentration of 10 mg/L for the following experiments.

2.2. Fluorescence Quenching Titration

2.2.1. DOM Interaction with Sr²⁺ and pH Condition Setting

Sr²⁺ and Ca²⁺ (Sino-Pharm Chemical Reagent Co, Shanghai, China) stock solutions were prepared. Various concentrations of Sr²⁺ stock solution were dropped into a series of brown sealed bottles containing 10 mL of DOM extracted from the soil. Then, 100 μL of a potassium chloride (KCl, pH 7.0) solution was used to maintain the ion strength and ensure an effective binding reaction. The final volume was 20 mL, and the concentrations of DOC, KCl, and Sr²⁺ were 10 mg/L, 10 mM, and 0–1.6 mM, respectively. In addition, the concentration of Sr²⁺ in this experiment was generally higher than that in the actual environment, meaning it could better characterize the complexation reaction. Finally, all bottles containing DOM and Sr²⁺ were shaken for 24 h in the dark. The UV-vis absorption, EEM, and synchronous fluorescence (SF) spectra of the solutions were measured.

To research the influence of pH on the interaction of DOM extracted from soil with Sr²⁺, a succession of brown sealed bottles containing 10 mL of DOM and various concentrations of Sr²⁺ stock solution were adjusted to adjust the pH of DOM with NaOH and HCl. The pH values of the mixed solution were 3.0–11.0. The above processing steps were repeated, and relevant indicators were measured.

2.2.2. Effect of Ca²⁺ on the Interaction between DOM and Sr²⁺

Various concentrations of Ca²⁺ stock solution were dropped into a series of brown sealed bottles containing 10 mL of DOM extracted from the soil. Then, 100 μL of a potassium chloride (KCl, pH 7.0) solution was used to maintain the ion strength and assure a better binding effect. The final volume was 20 mL, and the concentrations of DOC, KCl, and Ca²⁺ were 10, 10, and 0–1.6 mM, respectively. After shaking in the dark for 24 h at 25 °C, all solutions were measured by using EEM, UV-vis absorption, and synchronous fluorescence (SF) spectra.

To investigate the influence of Ca²⁺ on the interaction of DOM extracted from the soil with Sr²⁺, various concentrations of Ca²⁺ stock solution were dropped into several brown sealed bottles containing 10 mL of DOM and various concentrations of Sr²⁺ stock solution. A certain amount of potassium chloride (KCl, pH of 7.0) solution was dropped into the bottles to maintain a volume of 20 mL, DOC at 10 mg/L, Sr²⁺ at 0–1.6 mM, KCl at 10 mM, Ca²⁺ at 0.9 mM and 1.6 mM. After that, the UV-vis spectra and EEM spectra of the mixed solution were detected.

2.3. Fluorescence and Absorbance Measurement

The dissolved organic carbon (DOC) of the soil sample was gauged by a TOC-Vcph analyzer (Shimadzu, Kyoto, Japan) within a week after the sample arrived at the laboratory. UV-vis absorption spectra were determined by a UV-vis spectrophotometer (Shimadzu, UV-2550). The wavelength ranges were from 200 to 800 nm with intervals of 1 nm. A Hitachi F-7000 fluorescence spectrometer (Hitachi High Technologies, Tokyo, Japan) was applied to measured fluorescence spectra (EEMs and SF spectra). The scanning excitation (Ex) and emission (Em) of EEMs were 250–550 nm with 5 nm intervals, and the Ex wavelength region of SF spectra was 200–450 nm with a constant offset (60 nm).

2.4. Data Analysis

2.4.1. Two-Dimensional COS Analysis

“2Dshige” is a practical software used to reveal the binding sequence of DOM extracted from soil and pollutants. The 2D-COS of SF spectroscopy was analyzed using “2Dshige” software to exhibit the complexation behaviors of DOM–Sr. Detailed information is shown through synchronous and asynchronous diagrams. In the synchronous figures, the autopeak on the diagonal describes the overall degree of variation in spectral intensity in the external variable interval. When the crosspeak at λ₁/λ₂ is positive, it means possible coupling or correlation at λ₁ or λ₂. The negative peak at λ₁/λ₂ shows an opposite change at λ₁ and λ₂. In the asynchronous figure, the positive crosspeak at λ₁/λ₂ indicates that the variation at λ₁ occurs preferentially at λ₂. The negative cross-peak at λ₁/λ₂ suggests that the variation at λ₁ occurs after that at λ₂.

2.4.2. PARAFAC Modeling

The interactions of DOM components with Sr²⁺ were analyzed by PARAFAC. PARAFAC analysis involves the separation of a set of EEM data into mathematically and chemically individual fluorescent components. Fluorescent species are identified by the emission (Em) and excitation (Ex) wavelengths of different components. The PARAFAC modeling was operated by the drEEM toolbox of MATLAB (R2018b) [28]. MATLAB is a mathematical software produced by MathWorks Inc., which contains a large number of computational algorithms. The model was verified by split-half analysis, where six different half-datasets were collected and three validity tests “S4C6T3” (Splits-4, Combinations-6, Tests-3) were generated. The number of components was successively increased, and the residual was compared to determine the optimal number of components. More information on the PARAFAC model can be found elsewhere [29].

2.4.3. Complexation Modeling

The modified Stern–Volmer equation was utilized to evaluate the effectiveness of the fluorescence quenching for components. The modified Stern–Volmer equation is exhibited as Equation (1):

$$\frac{F_0}{F_0-F} = \frac{1}{f{K}_a{C}_{\mathrm{CM}} }+ \frac{1}{f}$$

(1)

where F₀ and F are the fluorescence intensities when the concentration of Sr²⁺/Ca²⁺ (mol·L⁻¹) was 0 and C_x. K_a represents the quenching constant, and f is the proportion of quenchable fluorescence groups.

To further explore the reaction of DOM components with Sr²⁺/Ca²⁺, the site-binding equation was applied to explore the stability constant and the binding points of components as shown in Equation (2):

$$\log \left(\frac{F_0-F}{F}\right) = \log K_b + n\log C_{\mathrm{CM}}$$

(2)

where K_b and n indicate the binding constant and the number of binding points, respectively.

2.5. Nomenclature

DOM: dissolved organic matter; EEM–PARAFAC: three-dimensional fluorescence spectroscopy combined with parallel factor analysis; 2D-COS: two-dimensional correlation analysis; SF: synchronous fluorescence spectra; logK_a: the binding coefficient; DOC: dissolved organic carbon; Four components extracted from EEM–PARAFAC: C1-C4.

3. Results and Discussion

3.1. Characteristics of DOM Extracted from Soil

3.1.1. UV-Vis Analysis

As shown in Figure 2d, the UV-vis spectra suggested that the absorbance of DOM extracted from soil increased logarithmically when the wavelength was lower than 260 nm and decreased gradually with increasing wavelength. Generally, the degree of humification is proportional to the absorbance of UV-vis spectra, which indicated that DOM extracted from soil humification was at a lower level in karst areas. In addition, when the value of SUVA₂₅₄ is greater than 4, it indicates that there is a considerable amount of aromatic hydrophobic substances in DOM [30]. In this study, the SUVA₂₅₄ value of DOM extracted from soil was 0.85 L/mg C m^–1, which indicated that DOM was mainly composed of hydrophilic small molecular aliphatic substances and hydrophilic compounds.

3.1.2. EEM-PARAFAC Analysis

There were generally two visible fluorescence peaks, including peak A (Ex/Em 275/330 nm) and peak B (Ex/Em 235/350 nm) (Figure 2e). It is deemed that the fluorescence peak with an emission wavelength less than 380 nm in the EEM spectra contains benzene rings and electron donors such as hydroxyl and amino groups, while the fluorescence peak with an emission wavelength greater than 380 nm is related to the polycyclic aromatic structure. In other words, the fluorescence groups with excitation wavelengths greater than 380 nm are classified as humic-like matter, while those with excitation wavelengths less than 380 nm are classified as protein-like materials. In this study, the two fluorescence peaks could be classified as protein-like substances, which characterized a relatively simple molecular structure and reflected the microbial activity and bioavailability of DOM [16,31]. There was no significant fluorescence peak in the humic-like fluorescence region, indicating that DOM extracted from soil was mainly composed of protein-like materials and that humic-like substances had a low content.

To further quantify the properties of DOM, PARAFAC analysis was applied, and we obtained a total of four components (C1–C4) (Figure 3). The locations of components C1 (Ex/Em = 260 (320)/400 nm) and C4 (Ex/Em = 230/430 nm) corresponded to tryptophan-like substances and tyrosine-like substances [31,32,33]. Components C2 (Ex/Em = 235 (302)/390 nm) and C3 (Ex/Em = 270 (345)/460 nm) could be classified as fulvic-like and humic-like materials, respectively [31,34]. Comparatively, component C4 has a longer emission wavelength than C3, indicating a tighter structure and higher molecular weight. According to the F_max value of each component, C1, C2, C3, and C4 accounted for 52%, 17%, 6%, and 25% of the total fluorescence intensity of DOM extracted from the soil, respectively (Figure 4a). This result suggests that DOM extracted from the soil in karst areas was mainly composed of tryptophan-like and tyrosine-like materials, while the contents of fulvic-like and humic-like matter were insignificant. According to relevant studies, protein-like substances are mainly precursors of humic polymers and energy sources of microorganisms, which are preferentially utilized to degrade and promote the increase in soil humic substances [31,35]. Meanwhile, a high proportion of protein-like matter could improve soil fertility and the continuous nitrogen supply [36]. This fluorescent feature may be related to the climatic conditions in karst regions. High summer temperatures lead to an increased abundance of microorganisms in the form of bacteria [37]. Active microbial activity is advantageous to DOM mineralization and the formation of biodegradable intermediates such as amino acids [35]. In addition, surface pollution caused by agricultural activities also contributes to the increase in protein-like substances [38]. This fluorescence characteristic of low humus was also found in the well water in the study area, which may be related to the higher cation concentration in the well water than that in the descending spring and surface water [3].

3.2. DOM–Sr Binding Behavior

3.2.1. UV-Vis and SF Spectra

The changes in UV-vis spectra after adding Sr²⁺ are shown in Figure 2c. Generally, the peptide skeleton structure had a high absorbance value located near 200 nm. The absorption shoulders at 230 nm were mainly caused by the amino acid side chain groups in tryptophan and tyrosine [39]. With the gradual addition of Sr²⁺, the absorbance and shape did not change significantly, indicating that Sr²⁺ molecules did not cause changes in the DOM molecular microenvironment.

SF spectra can not only reveal the effect of the quenching agent on DOM conformation, but also reveal the characteristics of the microenvironment near fluorescence functional groups. With the increase in the Sr²⁺ concentration, the fluorescence peak at 275 nm in the SF spectra of DOM extracted from soil was quenched (Figure 2g). Since the emission wavelength of tryptophane-like substances is longer than that of tyrosine-like materials, the fluorescence quenching at 275 nm was due to the interaction between tryptophane-like fluorophores and Sr²⁺. The relatively strong fluorescence quenching effect of tyrosine-like substances indicated a change in the electron structure in protein-like substances [29]. However, the changes in fluorescence intensities at 233 nm and 390 nm were insignificant, indicating that tyrosine-like and humic-like matter contributed little to the binding of DOM to Sr²⁺. Therefore, the binding sites interacting with Sr²⁺ in different components of DOM extracted from soil were not evenly distributed.

3.2.2. Two-Dimensional COS of SF Spectra

To study the binding sequence of Sr²⁺ and its components, the SF spectra of DOM extracted from soil were analyzed by 2D-COS (Figure 5a,b). In the synchronous map, there was a positive autopeak at 275 nm along the diagonal, indicating that the corresponding region had the strongest spectral variation, which agreed with the change in the SF spectra (Figure 5a). In addition, the two positive autopeaks at 275/230 nm and 390/275 nm suggested the same variation trend of the three fluorophore intensities at 230, 275, and 390 nm (decrease with increasing Sr²⁺ concentration).

In the asynchronous map, one positive crosspeak exists at 260 (275)/235 nm below the diagonal, while one negative crosspeak exists at 275/260 nm, which indicates that the fluorescence quenching of shortwave strength occurred preferentially in a long wave strength (Figure 5b). Specifically, the sequence of fluorescence quenching at different wavelengths is 260 > 275 > 235 nm [40]. Therefore, tryptophan-like substances were preferred to tyrosine-like substances for DOM–Sr binding, while humic-like substances had no significant strength change due to their low content, and 2D-COS analysis could not reveal the order of DOM–Sr binding.

3.2.3. Sr-Binding Abilities of PARAFAC Components

Figure 5d depicts the variations in the fluorescent intensity of the components after adding Sr²⁺ to the DOM extracted from the soil. The fluorescence quenching was most obvious when the added amount of Sr²⁺ was 0–60 mg/L. With the increase in Sr²⁺ addition, the fluorescence quenching gradually stabilized. It is noteworthy that the response behavior of the fluorescence intensity of the four fluorescence components to Sr²⁺ was different. Low concentrations of Sr²⁺ (<40 mg/L) induced an obvious fluorescence quenching effect on C3 and C4, while there was no fluorescence quenching phenomenon on C1 and C2. With increasing Sr²⁺ concentration, the fluorescence intensity of C3 and C4 continued to weaken and stabilize, while the fluorescence intensity of C1 also began to decrease. When the added amount of Sr²⁺ was 70 mg/L, the fluorescence quenching ratios of C1, C2, C3, and C4 were 9%, 4%, 17%, and 19%, respectively. The dynamic collision between the fluorescent substance and the quencher or the formation of a nonfluorescent complex can lead to fluorescence quenching. However, some studies have found that the fluorescence property was enhanced after DOM interacts with some metals, which was mainly caused by the variation of the molecular environment of the fluorescent component with the increased metal concentration [29]. The results of EEM-PARAFAC and SF spectra seem to be inconsistent, mainly because SF spectra are a two-dimensional spectrum, and DOM components at different wavelengths have cross coverage, which may cover part of the fluorescence quenching information. Various DOM components have different intensities of fluorescence quenching effects. Some studies indicate that the quenching degree of the tryptophan standard was higher than that of the humic standard under the same temperature conditions [41]. It was also reported that the fluorescence quenching effect of the L-tyrosine solution (52%) was more potent than that of the L-tryptophane solution [42].

To explore the interaction of Sr²⁺ with the components, the modified Stern–Volmer equation was employed to calculate the effective quenching constant (logK_a). The excellent linear relationships between F₀/(F₀ − F) and 1/C_Sr for all components suggested that the equation depicted the fluorescence quenching well (R² > 0.73). The logK_a values of the four DOM components ranged from 2.69 to 4.04, and the order was C3 > C1 > C4 > C2, indicating that the interaction between humic acid and Sr²⁺ was the strongest, followed by tryptophane-like and tyrosine-like substances (

Table 1. The binding parameters of different PARAFAC components with Sr²⁺ and Ca²⁺. “-” suggests the modified Stern–Volmer equation could not be modeled due to the weak fluorescence quenching effects.

DOM+Sr2+ Components	Modified Stern-Volmer Equation				Site-Binding Equation
	logKa	f	R2	p	logKb	n	R2	p
C1	3.24	12.4	0.92	<0.005	3.45	1.53	0.89	<0.005
C2	2.39	16.1	0.99	<0.005	0.98	0.83	0.97	<0.005
C3	3.73	15.4	0.74	<0.005	0.20	0.34	0.89	<0.005
C4	3.07	26.6	0.98	<0.005	1.23	0.67	0.98	<0.005
DOM+Ca2+
C1	-	-	-	-	-	-	-	-
C2	2.95	17.1	0.95	<0.005	0.92	0.66	0.95	<0.005
C3	2.34	24.4	0.77	<0.005	5.37	2.26	0.93	<0.005
C4	2.88	13.9	0.77	<0.005	1.82	0.98	0.87	<0.005

). The fulvic acid component had the weakest binding effect with Sr²⁺. Humic-like substances are one of the most important metal chelating agents in the natural world. Due to thier abundant functional groups and large molecular weight, they have a strong affinity for Sr²⁺ [43]. Therefore, the strong complexation ability of humic-like matter reduces its abundance and retains the increased accumulation of protein-like substances in karst soil.

The linear relation between log[(F₀ − F)/F] and Log [Sr²⁺] was good (R² > 0.89), indicating that the point-position combination mode can also better fit the interaction between DOM and Sr²⁺ (Table 1). Among the four fluorescence components, the binding constant of C3 and Sr²⁺ was the smallest, which was due to the low content of humic-like matter in soil DOM and the few effective binding points (n = 0.34). In contrast, the content of tryptophan-like substances was higher than that of other substances, and the effective binding sites of Sr²⁺ were more abundant than those of other substances, meaning that the effect was more stable. The binding sites for Sr²⁺ in the tyrosine-like and humic-like fractions were all less than 1. In general, protein-like component C3 had strong binding abilities and high stabilities to Sr²⁺, which were the leading fraction in the interaction between Sr²⁺ and DOM extracted from soil. It was proven that the complex of DOM extracted from soil and Sr²⁺ contributed to the migration and transformation of Sr²⁺.

3.3. Influence of pH Conditions on the Interaction between DOM Extracted from Soil and Sr²⁺

The fluorescence intensity of DOM extracted from soil components under different pH conditions is shown in Figure 6a. Under neutral conditions, the fluorescence intensity of C1, C3, and C4 was higher than that of C1 but decreased with decreasing and increasing pH. The change in pH might cause the ionization of fluorescence components or functional groups in DOM molecules, thus changing the molecular orbital of excited electrons and affecting their fluorescence properties [44]. The fluorescence intensity of compounds was affected by electron functional groups such as amino and hydroxyl groups [45,46]. Carboxyl chromophores make a major contribution to the change in fluorescence between pH 3.0 and 5.0, while phenolic chromophores have a key function in the shift at pH 9.0 and 11.0 [47]. It has been reported that the effect of pH on fluorescence is related to changes in DOM macromolecular structure [48]. Generally, the tighter the structure is, the stronger the fluorescence. When the pH increased, the molecular structure of the humic-like substance tended to be linear, and some fluorescent molecules were wrapped in acidic conditions. They expanded and emitted fluorescence under alkaline conditions, so the fluorescence intensity increased [45]. However, the fluorescence of protein-like matter mainly comes from aromatic amino acid residues (tryptophane tyrosine and phenylalanine). Some molecules in components C1 and C4, which were covered under acidic and alkaline conditions, spread out under neutral conditions, showing the strongest fluorescence. For the fulvic acid components, the fluorescence was the strongest under acidic conditions, and the fluorescence decreased with increasing pH, implying that low pH was conducive to the development of fluorescent molecules.

The interaction between the four DOM components and Sr²⁺ under different pH conditions is shown in Figure 6b–f. When the pH was 3, 5, 7, and 9, the fluorescence intensity of DOM components decreased with increasing Sr²⁺ concentration, indicating the formation of the nonfluorescent complex. However, when the pH value was 11, the fluorescence intensity of DOM components increased significantly with increasing Sr²⁺ concentration, indicating that the interaction between DOM components and Sr²⁺ under alkaline conditions changed the molecular microenvironment of DOM extracted from soil and formed a stronger fluorescence complex or precipitates [49].

Furthermore, the modified Stern–Volmer equation and the point-position binding equation were selected to fit the binding parameters of the four components with Sr²⁺ at pH 3, 5, 7, and 9 (Figure 4b,c and Table S1). For components C1, C3, and C4, logK_a was the maximum when the pH value was 7, indicating that the affinity of the three components to Sr²⁺ is the strongest under neutral conditions. However, when the pH increased or decreased, the fluorescence molecular structure was compacted, and its binding ability to Sr²⁺ decreased. At the same time, protons might compete with Sr²⁺ for DOM extracted from soil binding sites under acidic conditions, weakening the interaction between DOM and Sr²⁺. This phenomenon has also been observed in the complexation of cadmium (Cd) or copper (Cu) with DOM [49,50].

The interaction of component C2 with Sr²⁺ was stronger under acidic and alkaline conditions than under neutral conditions, indicating enhanced protonation, negative charge and electrostatic action [50,51]. Therefore, pH changes had different effects on the binding points of DOM extracted from soil and Sr²⁺. For C1 and C2, the decrease in pH led to the protonation of the fluorophore (the increase in H⁺), reducing the effective binding sites [52]. For C3, the reduction in pH increased the effective binding point of DOM extracted from soil to Sr²⁺, which might be due to the higher H⁺ concentration leading to the precipitation and desorption of Sr²⁺ in the form of sulfate and hydroxide, and the stability of the binding effect was enhanced [52]. The interaction between C4 and Sr²⁺ was more stable under acidic and alkaline conditions. In general, neutral conditions were more conducive to the binding of DOM extracted from soil and Sr²⁺. Alkaline environments in karst areas are not conducive to Sr²⁺ migration and transformation.

3.4. Effect of Ca²⁺ on the Interaction between DOM Extracted from Soil and Sr²⁺

3.4.1. Interaction of Ca²⁺ with DOM Extracted from Soil

As shown in Figure 2e, with increasing Ca²⁺ concentration, the fluorescence peak intensity at 275 nm in the SF spectra changed remarkably, indicating that the tryptophan-like fluorophore was the main component in the interaction between DOM extracted from soil and Ca²⁺, while tyrosine-like and humic-like substances contributed negligibly. This was similar to the interaction characteristics of DOM extracted from soil and Sr²⁺. Two-dimensional COS was further used to study the binding sequence of Ca²⁺ and different components (Figure 5c,d). In the synchronous map, a positive autopeak at 275 nm along the diagonal indicated that the corresponding region had the strongest spectral change, which agreed with the characteristic variation in the SF spectra (Figure 5c). In addition, the positive cross-peak at 275/230 nm indicated that the intensity of the two fluorophore groups at 230 and 275 nm had the same trend. In the asynchronous graph, there was a negative cross peak at 275/235 nm below the diagonal, indicating that tyrosine-like substances reacted with Ca²⁺ before tryptophane-like substances, which was inconsistent with the results of Sr²⁺ (Figure 5d). This meant that DOM extracted from the soil had different and similar binding mechanisms for different elements. Due to the low content of humic-like matter, the strength of humic-like materials did not change significantly, and 2D-COS analysis could not reveal the action sequence of humic-like substances with Ca²⁺.

Furthermore, Ca²⁺ could quench the fluorescence intensity of the four components to different degrees (Figure 7a). When the added amount of Ca²⁺ increased to 1.6 mM, the fluorescence quenching ratios of C1, C2, C3, and C4 were 14%, 10%, 7%, and 12%, respectively, indicating that most of the tryptophan-like substances could bind to Ca²⁺. However, most components of humic acid could not be quenched by Ca²⁺. By using the modified Stern–Volmer equation and the point-position binding equation, it was found that the order of the binding constants logK_a was C2 > C4 > C3, confirming the weakest interaction between the humic acid component and Ca²⁺, which differed significantly from the DOM–Sr²⁺ results (Table 1). However, the high logK_b value of humic-like components indicated that although there were fewer fluorescence components that could bind to Ca²⁺ in humic acid substances, they had the strongest mutual stabilities and the most effective adsorption sites. While components C2 and C4 had strong affinities for Ca²⁺, their utilization was unstable. In general, tryptophan-like, tyrosine-like, and fulvic-like components were the main functional groups of DOM extracted from soil interacting with Ca²⁺. Compared with Sr²⁺, the lower logK_a value of DOM extracted from soil interacting with Ca²⁺ suggested that the main mechanism of DOM extracted from soil interacting with Ca²⁺ was electrostatic force. The mechanism of DOM extracted from soil and Sr²⁺ was the formation of a complex, so the logK_a value was higher than that of DOM–Ca.

3.4.2. The Interaction between DOM Extracted from Soil and Sr in the Presence of Ca

The interaction between PARAFAC components and Ca²⁺ in the presence of Ca²⁺ is shown in Figure 7b,e. When the Ca²⁺ concentrations were 0.91 mM and 1.60 mM, Ca²⁺ had no obvious influence on the binding of the tryptophan-like component C1 to Ca²⁺. The fluorescence quenching of C2 and Ca²⁺ was sensitive to Ca²⁺. With increasing Ca²⁺ concentration, the fluorescence quenching of C2 and Sr²⁺ was enhanced. The interaction between the humic-like component C3 and Sr²⁺ showed a similar response trend to Ca²⁺ to that of C2. The effect of Ca²⁺ on the tyrosine-like component C4 was similar to that of C1. The presence of Ca²⁺ enhanced the fluorescence quenching of humic-like substances and Sr²⁺ to a certain extent, while the interaction between protein-like substances and Sr²⁺ was weakly affected by Ca²⁺.

Based on the Stern–Volmer equation and point-position binding equation, the binding abilities of the four components and Sr²⁺ in the presence of Ca²⁺ were evaluated (Figure 4d,e and Table S2). For component C1, logK_a decreased with increasing Ca²⁺ concentration, suggesting that Ca²⁺ reduced the affinity of tryptophan-like substances to Sr²⁺ by competing with Sr²⁺ for effective binding sites of tryptophan-like components. The binding constants of components C2 and C3 to Sr²⁺ increased with increasing Ca²⁺ and humic acid to Sr²⁺. However, the logK_a values of components C4 and Sr²⁺ were slightly increased when the Ca²⁺ concentration was 0.91 mM. The logK_a values were significantly decreased when the Ca²⁺ concentration was 1.60 mM, indicating that a high concentration of Ca²⁺ inhibited the binding of DOM to Sr²⁺. It has been reported that the binding effect of DOM and Cu²⁺ could be affected by 5 mM Ca²⁺ [53]. Ca²⁺ can change the zeta potential on the surface of DOM molecules by neutralizing negatively charged functional groups (such as carboxyl and phosphoric groups) in the DOM structure, thus affecting its binding with Sr²⁺ [49,54]. Different functional groups in different DOM components had different effects on Ca²⁺. For protein-like substances, Ca²⁺ could weaken the interaction between protein-like substances and Sr²⁺ by competing for binding sites and changing surface potentials through cation-π interactions [54]. In addition, Ca²⁺-(H₂O)_n formed outer-sphere complexes on the DOM surface, which hindered the binding of protein substances to Sr²⁺ [55]. However, Ca²⁺ might promote the binding of humic-like substances to Sr²⁺ through organic chelation [55]. As the main functional group in DOM was protein-like substances, the existence of Ca²⁺ inhibited the binding of DOM to Sr²⁺. According to previous research, the water and soil in the study area were slightly alkaline (pH: 6.91–8.11), implying a relatively strong complexation effect between DOM extracted from soil and Sr²⁺. The existence of large amounts of Ca²⁺ in karst areas led to the migration and transformation of Sr being more complicated between surface water and groundwater. The result also verify that Sr might exist in karst water in various forms, mainly including ionic state, complex state, and composite precipitation. The results confirm that DOM extracted from soil can reduce the Sr²⁺ content and toxicity to help protect groundwater. There are two possible paths: (1) Increasing the organic matter content in karst soil, especially humic-like matter. (2) Regulating the pH of the soil.

4. Conclusions

According to EEM-PARAFAC analysis, DOM extracted from soil is mainly composed of tryptophan-like and tyrosine-like materials (77% in total), while fulvic acid and humic acid were relatively low (23% in total). The tryptophan-like substance in DOM extracted from soil preferentially bound to Sr²⁺ so that Sr²⁺ transforms from a free dissolved state to a DOM-bound state. When the pH was 7, the binding coefficients of the components were highest (except C2), following the order of humic-like substance > tryptophane-like substance > tyrosine-like substance > fulvic-like substance. At the same time, tryptophan-like and tyrosine-like substances bound Sr²⁺ more stably. Acidic conditions were not conducive to the binding of DOM extracted from soil with Sr²⁺, while new fluorescent complexes appeared during the reaction of DOM extracted from soil with Sr²⁺ under alkaline conditions. Ca²⁺ in the soil complexed with DOM extracted from the soil, but it was weaker than that between DOM extracted from soil and Sr²⁺. Overall, Ca²⁺ inhibited the binding of protein-like substances to Sr²⁺ by competing for binding sites and changing the molecular surface potential. Increasing the organic matter content and adjusting the pH in the soil can help to protect groundwater by reducing the Sr²⁺ content and toxicity. These results compare the ability of DOM extracted from soil to transfer and transform Sr²⁺ in karst soil under different pH conditions and the role of Ca²⁺ in this process, providing a new idea for the remediation of soil heavy metal pollution and groundwater protection in karst areas.