Investigation of the Distribution and Binding Affinity of Copper to Size-Fractioned Dissolved Organic Matter (DOM) in a Constructed Wetland

Abstract

This study investigated the distribution and binding affinity of dissolved copper (Cu) and organic carbon (OC) in size-fractioned dissolved organic matter (DOM) in a constructed wetland (CW). Two sites were studied: one at the inflow (P-1) and one within the wetland (P-2). The DOMs (<0.45 μm) were separated into six size fractions using a cross-flow ultrafiltration system. In the wetland (P-2), the concentrations of dissolved organic carbon (DOC) increased while the concentrations of Cu decreased. The high molecular weight fraction (1 kDa–0.45 μm, HMW) contained most of the OC mass (57.4–71.2% averages). On the other hand, Cu was almost equally distributed in HMW and low molecular weight fractions (<1 kDa, LMW) with mean HMW percentages of 50.3–51.3%. The mean Cu binding affinity to DOM ratios (CuBA_DOM) was 74.9 ± 24.0 μmol/g-C at site P-1 and 17.3 ± 2.6 μmol/g-C at site P-2. The CuBA_DOM ratios were decreased in wetlands of bulk and size-fractioned DOM (p < 0.001 to p = 0.073). The SUVA₂₅₄ values for bulk DOM solution were 2.54 ± 0.15 and 1.68 ± 0.18 L/mg-C/m, and humidification index (HIX) values were 1.74 ± 0.16 and 2.09 ± 0.19 for sites P-1 and P-2, respectively. Optical indicators suggested that the wetland process decreased aromaticity but increased the humification degree of DOM. Furthermore, the CuBA_DOM ratios positively correlated with SUVA₂₅₄ and HIX within the constructed wetland DOM but not in the influent DOM. Understanding the Cu distribution and binding affinity to size-fractioned DOM makes it possible to develop strategies to mitigate the potential effects of copper pollution in wetlands.

1. Introduction

Constructed wetlands (CW) are effective in improving water quality by reducing organic pollutants, nutrients, and heavy metals (HMs) [1,2,3,4,5,6,7]. The water in CW contains different types of dissolved organic matter (DOM), originating from the inflow and also produced within the wetland itself [6,7,8,9]. DOM levels in the wetland can be reduced by natural processes like plant uptake, colloidal aggradation and deposition, biodegradation, and photodegradation, with biodegradation and photodegradation being the main processes [6,10,11,12,13,14,15,16,17]. However, the growth of wetland plants, especially during spring and summer, can lead to an increase in the amount of DOM present [6,7], subsequently raising the concentration of dissolved organic carbon (DOC) in the wetland [6,7,8,9,10,11]. These transformations result in dynamic changes in the abundance, molecular size, and chemical composition of DOM in aquatic systems, leading to significant impacts on their environmental behavior and ecological function.

DOM is a complex mixture of organic substances including humic acid, proteins, and carbohydrates [11,12,18,19]. The composition, structure, and molecular weight distribution of DOM can vary across different water bodies due to distinct biogeochemical processes [12,20,21]. These variations in components and structures lead to different metal-binding properties of DOM, impacting the toxicity and transportation of metals in the environment [22].

HMs are significant environmental pollutants that are known for their biological toxicity and non-degradability [2,3,4,5]. HMs exist in various forms in the aquatic environment, with free ions being highly toxic. However, when heavy metals bind to organic colloids, their toxicity decreases [23,24]. Additionally, heavy metals that bind to different molecular weight organic matter can form complexes that may aggregate and deposit in sediment or promote the migration of heavy metals over longer distances [25,26,27].

The binding capacity of heavy metals to dissolved organic matter (DOM) is expressed as the binding affinity of heavy metals to DOM, which is known as the HMBA value [28,29,30,31,32]. This value represents the ratio of heavy metal concentration bound to a unit of organic carbon (μmol g-OC⁻¹). A high HMBA value indicates a greater capacity of DOM to bind with heavy metals. The HMBA ratios determine the binding capacity and migration of river DOM, soil, and sediment-extracted DOM to bind with heavy metals [28,29,30,33]. In previous studies, the CuBA ratios in river water DOM have been observed to range from 7.5 to 29.8 μmol g-C⁻¹ [9,29,31]. Furthermore, the CuBA ratios of water- and CaCl₂-extracted soil organic matter were found to range from below the detection limit to 34.0 μmol g-C⁻¹, and the CuBA ratios of NaOH-extracted soil and sediment organic matter varied from 13.2 to 56.2 μmol g-C⁻¹ [28,33,34].

To assess the chemical composition and structure of DOM, optical indicators are used as they are rapid, cost-effective and sensitive tools [6,7,8,9,10,11,30,35,36,37,38]. One such indicator is SUVA₂₅₄, which is the ratio of ultraviolet absorbance at 254 nm to DOC concentration [39]. This indicator provides information about the level of aromaticity of DOM, with higher values indicating a greater aromatic content [7,39]. The humification index (HIX) from the fluorescence spectrum can also be used to assess the degree of humification of DOM, with high HIX values indicating a great degree of humification [40,41]. Both SUVA₂₅₄ and HIX are commonly used to examine the aromatic content and humification degree of DOM in water, soil, and sediment extraction solutions [11,37,38]. Previous studies have shown a significant positive link between CuBA values and SUVA₂₅₄ in river water and soil extraction solutions [28,29,30,33].

There is limited research on the binding affinity of heavy metals to size-fractioned DOM in CWs, particularly regarding how plant-driven leaching DOM binds to heavy metals. Previous studies have mainly focused on heavy metal filtration in aqueous solutions and their binding affinity [28,29,30,33]. However, few studies have examined the ratios of different molecular weights of DOM in terms of their binding behavior with heavy metals, which could provide valuable insights into their behavior [31,34,42,43,44,45]. Understanding how heavy metals bind with DOM of different molecular weights is crucial in comprehending the migration, biotoxicity, and environmental fate of these metals [46,47,48].

The purpose of this study is to investigate how Cu binds to different sizes of DOM in a constructed wetland during the summer season. Copper was chosen for this study due to its toxicity to both humans and ecological organisms [11,21,49,50]. In the United States, regulations have been in place since 1991 to limit copper exposure through drinking water. The US Environmental Protection Agency has established a maximum contaminant level goal and a drinking water action level for copper at 1.3 mg Cu/L, guided by an oral reference dose of 0.04 mg Cu/kg/day to safeguard against acute or chronic toxicity in adults and children [49,50]. Moreover, the presence of various materials in DOM significantly influences copper behavior. The interaction of copper with DOM can be characterized by the spectral changes of Cu binding with DOM [38,51,52,53]. Water samples were collected from two points: the wetland influent (P-1) and within the wetland (P-2). These samples were filtered (<0.45 μm) and then separated into six size fractions using an ultrafiltration system based on their molecular weights. The concentrations of Cu and DOC, as well as the UV/Vis and fluorescence spectra of both bulk and size-fractioned DOM, were measured. To determine the Cu binding affinity of the DOM, the CuBA_DOM ratios and optical indicators (SUVA₂₅₄ and HIX) were calculated for both the bulk and size-fractioned DOM. Linear correlation was used to examine the factors that affected CuBA_DOM, such as optical indicators and DOC and Cu concentrations.

2. Materials and Methods

2.1. Sampling Site and Sample Collection and Preparation

Water samples were collected from a constructed wetland located on the National Pingtung University of Science and Technology (NPUST) campus in Neipu Township, Pingtung County, Taiwan. The campus has geographical coordinates of 22°38′40″ N and 120°36′38″ E, with an altitude of approximately 70–100 m. The constructed wetland was previously an abandoned fish pond and grassland and is made up of a regulating area measuring 0.23 hectares and the main body of the wetland measuring 2.34 hectares, with an upstream catchment area of about 14.44 hectares (see Figure 1). The campus experiences a tropical climate with an average temperature of 25.5 °C and an annual rainfall of around 2600 mm. The composition of the constructed wetland sediment was reported as 30% sand, 36% clay, and 34% clay loam, with a soil organic matter content of 6.9% [54].

The water sources of the constructed wetland include overflow water discharged from the aquaculture pond of the Department of Aquaculture at NPUST, rainfall, rainfall runoff collected from the upstream catchment area, and pumped groundwater used to maintain the water level of the wetland. The water losses of the constructed wetland include groundwater infiltration and evapotranspiration from the pool surface and plants. The average inflow volume was reported as 7372 m³ d⁻¹, and the mean stored water volume was 9256 m³ in the constructed wetland [54]. As a result, the hydraulic retention time of the constructed wetland is estimated to be around 19.1 h.

Water samples were collected from two sites within the CW. The first water sampling site (P-1) is at the entrance of the CW and represents the influent. The second sampling site (P-2) is located central to the CW and represents wetland water samples. Water samples were collected three times in July 2015, with twenty-liter water collected at each sampling site. After the water samples were collected, they were taken back to the laboratory (<1 h) for water filtration (<0.45 um) and water quality analysis. The filtrate was named bulk DOM.

The measurement of basic water quality parameters includes pH, COD, dissolved oxygen (DO), suspended solids (SS), total suspended solids (TSS), and N-NH₄⁺. These measurements are conducted according to the Taiwan EPA standard methods (NIEA W424.53A for pH, W515.55A for COD, W455.52C for DO, W210.58A for SS and TSS, and W448.52B for N-NH₄⁺). At site P-1, the values were 7.02 ± 0.02 for pH, 60.3 ± 2.9 mg/L for COD, 2.23 ± 0.03 mg/L for DO, 4.0 ± 0.5 mg/L for SS, 147.7 ± 2.1 mg/L for TSS, and 4.27 ± 0.01 mg/L for N-NH₄⁺, respectively. At site P-2, the values were 7.39 ± 0.02 for pH, 66.0 ± 0.2 mg/L for COD, 2.09 ± 0.01 mg/L for DO, 17.3 ± 0.3 mg/L for SS, 159.3 ± 3.1 mg/L for TSS, and 4.12 ± 0.01 mg/L for N-NH₄⁺, respectively. The water parameters at site P-2 showed an increase in pH, COD, SS, and TSS (p = 0.001–0.08), but a decrease in DO and N-NH₄⁺ values (p < 0.01).

2.2. DOM Separation

Ten liters of filtered DOM solutions (<0.45 μm) were separated into six size-fractioned DOM solutions. To accomplish this, a cross-flow ultrafiltration system equipped with ceramic membrane cartridges of nominal molecular weight cutoffs of 0.14 μm, 100 kDa, 10 kDa, 3 kDa, and 1 kDa (Filtanium, France) was employed. The six size-fractioned DOMs were designated as SF-A (0.14–0.45 μm), SF-B (100 kDa–0.14 μm), SF-C (10–100 kDa), SF-D (3–10 kDa), SF-E (1.0–3.0 kDa), and SF-F (<1.0 kDa).

The separation process was carried out with a feed flow rate of 1.7–2.0 L/min, and a total of five separation processes were performed. The duration of each separation process was about three hours, depending on the area and pore size of the filtrate membrane. During the concentration process, the volume ratio of retentate flow (V_r) and permeate flow (V_p) was kept at V_r:V_p = 1:9, which resulted in a concentration volume factor of 10. The permeate flow was collected in a separate container while the retentate flow was returned to the feed flow bottle. Once the designated volume ratio was achieved, the permeate flow was used for the next separation process. The mass balance rate (MassBR, %) for Cu and DOC was calculated using Equation (1), while the mass percentages (MassSFi %) of the size-fractioned DOM for DOC and Cu were calculated using Equation (2) [55]:

$$\mathrm{M}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{B}\mathrm{R}\left(\mathrm{\%}\right)=\frac{\sum \left({\mathrm{C}}_{\mathrm{i}}\times {\mathrm{V}}_{\mathrm{i}}\right)}{{\mathrm{C}}_{\mathrm{v}}\times {\mathrm{V}}_{\mathrm{b}}}\times 100$$

(1)

$${\mathrm{M}\mathrm{a}\mathrm{s}\mathrm{s}\mathrm{S}\mathrm{F}}_{\mathrm{i}}\left(\mathrm{\%}\right)=\frac{{\mathrm{C}}_{\mathrm{i}}\times {\mathrm{V}}_{\mathrm{i}}}{\sum \left({\mathrm{C}}_{\mathrm{i}}\times {\mathrm{V}}_{\mathrm{i}}\right)}\times 100$$

(2)

where C_i and V_i were the concentration and volume of DOC and Cu for each size-fractioned solution. C_b and V_b were the concentration and volume of DOC and Cu for bulk DOM. The volume ratios were 0.1, 0.09, 0.081, 0.0729, 0.0656, and 0.5905 for the size-fractioned solutions SF-A, SF-B, SF-C, SF-D, SF-E, and SF-F, respectively. In each separated process, the ultrafiltration membrane needed to be carefully cleaned; the cleaning process followed the sequence of reverse osmosis (RO) water, neutral detergent, RO water, 0.5 N sodium hydroxide (NaOH) solution, RO water, and DDW water. When the ultramembrane was not used for a long time, the cartridge was kept in 0.5 N NaOH solution.

2.3. Dissolved Organic Carbon and Metals Measurement

The concentrations of DOC in DOM solutions were determined using a total organic carbon (TOC)-V analyzer (Shimadzu, Kyoto, Japan). The concentrations of Cu were measured using a graphite furnace atomic spectrometer (Hitachi, Z-3000, Tokyo, Japan).

2.4. UV/Vis Measurements

The absorbance measurement of bulk and size-fractioned DOM solutions were measured using a UV/Vis spectrophotometer (Hitachi, U-2900, Japan), on a scanning wavelength of 800–200 nm. Background was corrected according to the Helms et al. [56] method. An average absorbance value of 700–800 nm was adopted as the background value, which was subtracted from the sample value. Aromaticity plays a crucial role because the electron-withdrawing nature of aromatic moieties in DOM increases its molecular electronegativity, resulting in the improved complexation of positively charged metal ions [57]. The specific absorbance at 254 nm (SUVA₂₅₄) is proportional to the percentage of aromatic carbon in the DOM [39]. The SUVA₂₅₄ value calculation follows in Equation (3):

$${\mathrm{S}\mathrm{U}\mathrm{V}\mathrm{A}}_{254}(\mathrm{L}/\mathrm{m}/\mathrm{m}\mathrm{g}-\mathrm{C})=({\mathrm{U}\mathrm{V}}_{254}/[\mathrm{D}\mathrm{O}\mathrm{C}\left]\right)\times 100$$

(3)

where UV₂₅₄ (cm⁻¹) is the UV/Vis absorbance at 254 nm of the sample, and [DOC] is the DOC concentration (mg-C/L) of the size-fractioned DOM solutions [39].

2.5. Fluorescence Measurements

In this study, three-dimensional fluorescence excitation/emission matrix spectroscopy (EEM) was recorded. The bulk and size-fractioned DOM solutions were measured by a fluorescence spectrophotometer (Hitachi, F-7000, Japan). Fluorescent scanning conditions were as follows: an excitation wavelength of 200–450 nm with 5 nm increases, an emission wavelength of 250–550 nm with 2 nm increases, a scan rate of 2400 nm/min, a slit width of 5 nm, and voltage amplifying of 700 V. The spectra were obtained by subtracting an ultrapure water blank spectrum, recorded in the same condition, to eliminate the Raman scatter peaks.

The HIX value was determined using Equation (4) as outlined by Birdwell and Engel [41] and Hansen et al. [58]. This involves dividing the sum of the fluorescence emission wavelengths (Em) between 435–480 nm by the sum of the emission wavelengths between 300–345 nm when excited (Ex) at a wavelength of 254 nm. When the excitation wavelength is 254 nm, an emission wavelength of 435–480 nm indicates the presence of humic acid-like and fulvic acid-like substances, while an emission wavelength of 300–345 nm indicates the presence of protein substances in the DOM. Therefore, the ratio of emission at 435–480 nm to emission at 300–345 nm represents the ratio of fulvic acid-like and humic acid-like substances to proteins. A high value indicates that the DOM contains a large proportion of humic acid-like substances [58,59,60].

$$\mathrm{H}\mathrm{I}\mathrm{X}=\frac{E_m=435-480 \mathrm{n}\mathrm{m}}{E_m=300-345 \mathrm{n}\mathrm{m}} at E_x=254 \mathrm{n}\mathrm{m}$$

(4)

2.6. Copper Binding Affinity to DOM

The copper binding affinity to DOM (CuBA_DOM ratios, μmol/g-C) is calculated by dividing the Cu concentration (μmol/L) by the concentration of dissolved organic carbon (g-C/L) [28,29,30,33].

2.7. Statistic and Data Analysis

This study utilized the S-Plus software version 6.2 to conduct linear regression and various tests, all at a significance level of p < 0.05. To perform two group difference tests between P-1 and P-2, including factors such as concentration, mass percentage, CuBA ratios, and indicators, the t-test method was employed. Additionally, the ANOVA test method was used for more than three group difference tests, including the test among size-fractioned DOM. Fluorescence indicators were calculated using the R script developed by Lapworth and Kinniburgh [61].

3. Results

3.1. DOC and Cu Concentrations in the DOM

Table 1. The measured DOC and Cu concentrations of bulk and size-fractioned DOM.

Size Fractioned DOM	DOC, mg/L		Cu, μg/L
	P-1	P-2	P-1	P-2
Bulk, <0.45 μm	1.51 ± 0.31	3.64 ± 0.63	6.87 ± 0.65	3.93 ± 0.16
SF-A, 0.14–0.45 μm	1.85 ± 0.30	3.71 ± 0.29	7.61 ± 1.31	6.14 ± 0.96
SF-B, 0.14 μm–100 kDa	2.21 ± 0.45	17.01 ± 2.09	10.34 ± 0.76	5.55 ± 0.86
SF-C, 10–100 kDa	1.33 ± 0.40	2.17 ± 0.14	8.12 ± 0.31	4.27 ± 0.61
SF-D, 3–10 kDa	1.63 ± 0.38	2.97 ± 0.29	7.16 ± 0.86	6.77 ± 0.96
SF-E, 1–3 kDa	1.99 ± 0.31	2.50 ± 0.53	7.40 ± 0.68	4.88 ± 0.40
SF-F, <3 kDa	0.94 ± 0.14	1.68 ± 0.03	5.65 ± 0.85	3.66 ± 0.27
Mass balance, %	89 ± 23	97 ± 20	98 ± 17	113 ± 5

Volume fraction: Bulk (1.0), SF-A (0.1), SF-B (0.09), SF-C (0.081), SF-D (0.073), SF-E (0.066), SF-F (0.59).

presents the concentrations of DOC and Cu in bulk and size-fractioned DOM at the P-1 and P-2 sampling sites. The concentrations of each fraction are actual measured values, not multiplied by their corresponding volume ratio. The average mass balance of DOC in the DOM is 93%, and that of Cu is 106%. Both Cu and organic carbon (OC) mass balances fall within the accepted range of 100 ± 25% [62].

In this study, the DOC concentration in bulk DOM was measured within a wetland and compared to the DOC concentration found in the influent of bulk DOM. The mean DOC concentration of bulk DOM in the wetland was 3.64 ± 0.63 mg/L, which was significantly higher than the mean DOC concentration of 1.51 ± 0.31 mg/L in the influent (p = 0.014). The DOC concentrations in the bulk DOM solutions were found to be lower than the DOC concentrations found in general river water and wetlands in other studies [6,7,8,10,11,29,30,37,45,63].

All size fractions of DOC concentrations were found to be increased at site P-2, with SF-B (100 kDa–0.14 μm) having the highest increased ratios (671%). This suggests that the wetland increased the concentration of DOC in HMW DOM, which implies that the leaching of plant-driven DOM in the wetland is a major possibility [6,7,10]. Similar results have been observed in CW water in other studies [6,7,10,38].

The average concentration of Cu in bulk DOM was found to be 3.93±0.16 μg/L in the wetland, which was significantly lower than the average concentration in the influent, which was 6.87 ± 0.65 μg/L (p = 0.012). These concentrations are comparable to those found in unpolluted river and wetland water [29,30,64,65,66] and lower than reported Cu concentrations in CW [3,4]. Furthermore, the Cu concentration of all size-fractioned DOM decreased with decreasing ratios ranging from 5.4% to 47.5%. The SF-B and SF-C had high decreasing ratios, while SF-D had a low decreasing ratio.

3.2. DOC and Cu Mass Percentages in Size-Fractioned DOM

The distribution of Cu and DOC in different sizes of fractionated DOM is crucial for understanding the migration and biotoxicity of Cu in the environment. In Figure 2, the OC and Cu mass percentages of the size-fractioned DOM are calculated by Equation (2). Two sites had a different order of OC mass distribution. In the influent (site P-1), the SF-F had the highest OC mass percentage of 42.6%, and OC mass percentages of the other fractions ranged from 8.3–15.5%. Within the wetland (site P-2), the OC mass percentage of SF-B was significantly increased (ratio of 186%) and had the highest OC mass percentage of 44.2%. The OC mass percentage of the other fractions was decreased with ratios of 25.6% to 52.6%. For site P-1 and site P-2, the OC mass percentages of colloids (1 kDa–0.45 μm, HMW) were 57.4% and 71.2%, respectively. The wetland increased the ratio of colloidal OC. In freshwater aquatic environments, HMW OC percentages ranged from 23% to 80% [11,13,67,68].

The distribution of Cu in size-fractioned DOM showed a consistent trend following size-fractioned DOM, although the Cu was significantly decreased in the CW. Cu mass was primarily distributed in the truly dissolved phase (<1 kDa, LMW) and was 49.7% and 48.7% for sites P-1 and P-2, respectively. The Cu mass percentages ranged from 7.3% to 13.9% for influent and from 7.2% to 13.9% within wetland for the other size fractions from SF-A to SF-E. Previous studies show that the Cu mass percentages in the HMW DOM varied in aquatic environments; for example, the Cu mass percentages of the HMW (1 kDa–0.22 μm) were 53% and 38% for lake and river water, respectively [69], and that Cu mass percentage of river water is comparable to the results of the present study. In four treatment processes of a municipal wastewater treatment plant, Hargreaves et al. [42] observed that HMW (1 kDa–0.45 μm) Cu mass percentages ranged from 69% to 77%, which were higher than the results of the present study.

During the current CW, the concentration of DOC increased while the concentration of Cu decreased. Specifically, the SF-B fraction exhibited the highest increase in both DOC concentration and mass percentage, while the SF-B and SF-C fractions showed the greatest decrease in Cu concentration. The SF-B fraction has the lowest aromaticity and humification degree, which will be further explained in Section 3.4. Properties of the SF-B fraction include low aromaticity and humification, resulting in low Cu binding affinity to SF-B DOM. The Cu in the SF-B fraction originally may have been redistributed to other molecular weight DOM. Additionally, there is a possibility that the high molecular weight fraction SF-B bound Cu and deposited it in the sediment. Therefore, the leaching of plant-derived DOM in the wetland significantly altered the DOC and Cu concentrations.

3.3. Cu Binding Affinity to DOM (CuBA_DOM)

The binding affinity of Cu to DOM, known as CuBA_DOM (μmol/g-C), is a measure of the ability of Cu to bind with DOM and its mobility [29,30,31]. The CuBA_DOM ratios varied with the size-fractioned DOM at sites P-1 and P-2, as shown in Figure 3. The CuBA_DOM ratios ranged from 58.9 to 101.3 μmol/g-C at site P-1 and from 5.1 to 36.3 μmol/g-C at site P-2, and they were found to vary between the sites. The bulk and size-fractioned CuBA_DOM ratios were lower at site P-2 (p = 0.001–0.073). The fraction SF-B had the highest decrease ratio of up to 93%, with Cu concentration decreasing by 46% and DOC concentration increasing by 671%. The leaching of plant-derived DOM in wetlands generated HMW DOM, which decreased the CuBA_DOM ratios. The CuBA_DOM ratio of bulk DOM decreased by 77%, and other size-fractioned DOMs increased from 46% to 65%.

The main treatment processes in CW are biodegradation and photodegradation, which are responsible for consuming DOM in water [6,11,13]. Biodegradation results in the rapid loss of small labile aliphatic organic matter, carbohydrates, and proteins, converting HMW DOM to LMW fractions [6,11,13]. However, it can also release HMW aromatic carbon compounds, such as humic-like compounds, as byproducts of organic matter degradation [6].

In aquatic environments, photodegradation can shift HMW DOM to smaller photoproducts [6]. In this CW, plant-derived DOM is the dominant process significantly increasing DOC concentration at site P-2. The mass percentage of HMW OC increased, but the mass percentage of HMW Cu remained almost the same. Additionally, DOM with different molecular weights showed variations in metal binding efficiency [6,70,71]. Xu et al. [72] found that HMW DOM (>1 kDa) constituents had stronger complexing abilities with heavy metals. However, some studies have presented opposite views [73,74]. In this CW, the plant-derived leaching DOM fraction SF-B (100 kDa–0.14 μm) had the lowest copper binding affinity. This fraction exhibited low aromaticity and humification. Therefore, the source and process of wetlands affect the variation in DOM composition, structure, and molecular weight distribution, resulting in different metal-binding properties of DOM. This, in turn, impacts the toxicity and transportation of Cu in the environment [11,12,20,22].

The mean bulk CuBA_DOM ratio at site P-1 was 74.9 ± 24.0 μmol/g-C, which was higher than the CuBA_DOM ratios observed in river water (ranging from 7.5 to 29.8 μmol/g-C) [23,24,25,26,27,28,29,30,31,64]. However, in the wetland, the CuBA_DOM ratio was 17.3 ± 2.6 μmol/g-C, which is comparable to the CuBA_DOM ratios observed in river water. It is worth noting that the LMW DOM (<100 kDa) had mean CuBA_DOM ratios ranging from 30.8 to 36.3 μmol/g-C. These DOM fractions may easily dissociate Cu from the Cu-DOM complex, which can increase Cu toxicity.

3.4. Optics Indicators of the DOM

Optical indicators are widely used to investigate the chemical composition and structure of DOM [15,28,30,33,75,76].

Figure 4a,b display the values of indicators for bulk and size-fractioned DOM, including SUVA₂₅₄ and HIX. SUVA₂₅₄ reflects the aromaticity of DOM, with high values indicating high aromaticity [35,39]. The SUVA₂₅₄ values vary with both bulk and size-fractioned DOM at sites P-1 and P-2. At site P-1, the SUVA₂₅₄ values of bulk DOM were significantly higher than those of site P-2 (p = 0.004). The mean total SUVA₂₅₄ value of 2.25 ± 0.43 L/mg-C/m at site P-1 was also significantly higher than the SUVA₂₅₄ value of 1.86 ± 0.64 L/mg-C/m at site P-2 (p = 0.036). Furthermore, the SUVA₂₅₄ values of fraction SF-B at site P-2 were significantly lower (0.65 ± 0.14 L/mg-C/m) than those at site P-1 (1.57 ± 0.29 L/mg-C/m) (p = 0.02). All the SUVA₂₅₄ values were less than 4.0, indicating that the DOM composition of the influent and wetland DOM primarily consisted of hydrophilic substances with poor aromaticity [59].

The HIX values are used to indicate the degree of humification of the DOM [41,58,77]. In Figure 4b, the HIX values varied with the bulk and size-fractioned DOM at sites P-1 and P-2. The HIX values of bulk DOM at site P-1 were lower than those of bulk DOM at site P-2 (p = 0.07). The mean total HIX values of 1.89 ± 0.33 at site P-1 were significantly lower than the HIX value of 2.50 ± 0.62 at site P-2 (p < 0.001). The HIX values at site P-1 were insignificantly different among size-fractioned DOM (p > 0.05). However, at site P-2, the HIX values (0.65 ± 0.14) of SF-B were significantly lower than those of SF-C to SF-F (2.01–2.73). This implies that the leaching of plant-derived DOM had a low humification degree. Furthermore, all HIX values at sites P-1 and P-2 were less than 4.0, suggesting that the DOM had a low degree of humification as the DOM composition predominantly comprised hydrophilic substances [41,58,77]. The DOM increased HIX, but decreased the SUVA₂₅₄ value in the present CW. This could be because the plant DOM had low aromaticity and degree humification, but wetland biochemical processes slightly increased the humification degree of DOM.

The optical indicators of DOM may vary in river and wetland water, depending on the sources and biochemical processes of the aquatic environment. Previous research indicates that the SUVA₂₅₄ values for DOM in wetlands ranged from 0.41 to 3.54 L/mg-C/m [6,7,9,11,21,37,38,78]. Similarly, the HIX values for river and wetland DOM ranged from 0.89 to 4.61 [11,21,38,78]. In the current study, the SUVA₂₅₄ and HIX values of CW DOM are within the reported ranges of values in wetlands.

Figure 5 shows the correlation of SUVA₂₅₄ and HIX at sites P-1 and P-2. In the influent (site P-1), the correlation of SUVA₂₅₄ and HIX was insignificant (r = 0.01). However, within the wetland (site P-2), there was a significantly positive correlation between SUVA₂₅₄ and HIX (r = 0.84, p < 0.001). Biochemical processes in the wetland stabilize and homogenize the DOM composition and structure for each size fraction of DOM, resulting in high aromatic DOM with a high degree of humification in size-fractioned DOM.

3.5. Correlation between CuBA_DOM Ratios with Optical Indicators

Figure 6a,d demonstrate the linear correlation between CuBA_DOM ratios and various indicators such as SUVA₂₅₄ and HIX, as well as the concentrations of Cu and DOC for the size-fractioned DOM at sites P-1 and P-2. At site P-1, the CuBA_DOM ratios showed a weak correlation with SUVA₂₅₄, HIX, and Cu concentrations, but a significantly negative correlation with DOC concentrations. At site P-2, the CuBA_DOM ratios had a weak correlation with Cu concentrations and a significantly negative correlation with DOC concentrations. Moreover, the CuBA_DOM ratios exhibited significantly positive correlations with SUVA₂₅₄ and HIX values. At site P-1, the incoming DOM had various sources, making it complex. However, the wetland processes helped to even out and stabilize the chemical composition and structure of DOM. This resulted in a higher degree of aromaticity, which is an indicator of humification. The high degree of humification facilitated the binding of Cu to the organic matter with high aromaticity and humification. Overall, the correlation indicated that CuBA_DOM ratios were more favorable when there was a significant degree of aromaticity and humification in the DOM.

In the wetland (P-2), the CuBA_DOM ratios were significantly lower than in the influent water (site P-1). However, the solution’s DOC concentration increased (Table 1), while the Cu concentrations and CuBA_DOM ratios decreased (Table 1 and Figure 3). This is mainly due to the plant-derived DOM within the wetland, which led to a significant increase in the DOC concentration and the mass fraction of the molecular weight SF-B (0.14 μm–100 kDa) of P-2 (Table 1 and Figure 2). Furthermore, the SUVA₂₅₄ and HIX fraction SF-B of P-2 are relatively lower compared to values of other molecular weights (Figure 4a,b), and these two indicators showed a significantly positive correlation (Figure 5). It suggests that the plant-derived DOM has a low aromaticity and low degree of humification. In comparison with the CuBA_DOM value of bulk DOM in P-2, the CuBA_DOM value of SF-B is also significantly reduced. These results indicate that the source and molecular weight of DOM have a significant impact on the binding affinity between Cu and DOM.

Previous studies have suggested that the CuBA_DOM ratios have a significantly positive correlation with the aromaticity of natural water DOM [29,30] and soil solution [28,33,34]. However, when the DOM has complex sources such as wastewater treatment plant effluent input [30] and anthropogenic ligand [29], the CuBA_DOM ratios have a poor correlation with SUVA₂₅₄. Furthermore, several studies have observed a negative correlation between DOC concentrations and CuBA_DOM ratios [28,33,76].

4. Conclusions

The research revealed that the water passing through the CW resulted in an increase in DOC concentration and a decrease in Cu concentration. In the wetland, the Cu binding affinity to size-fractioned DOM decreased significantly compared to influent DOM. The HMW DOM increased DOC significantly, which suggests that the leaching of plant-derived DOM in the wetland contributed to the increased DOC concentration and decreased the CuBA_DOM ratios. Within the wetland (site P-2), the CuBA_DOM ratios positively correlated with indicators SUVA₂₅₄ and HIX. The P-2 SUVA₂₅₄ values decreased, while the HIX values increased. This change could be due to the low-aromatic nature of the DOC derived from wetland plants. However, wetland processes such as biodegradation and photodegradation still contribute to an increase in the HIX of low-molecular-weight DOM. The higher aromaticity and humification degree of DOM enhanced the Cu binding affinity to DOM.