Effects of soil components, solution chemical properties, and temperature on Cu(II) adsorption by alluvial acid soil: a case study in Southwest China

Soil components, solution chemical properties, and temperature are important factors that affect Cu(II) adsorption in soil. However, the research in this area is currently lacking. In this study, the clay fraction was separated from alluvial acid soil. In addition, organic matter, free Fe oxides, and Mn oxides were selectively removed by washing the bulk soil and the clay fraction with H2O2, Na3C6H5O7·2H2O + NaHCO3 + Na2S2O4, and HONH3Cl. Adsorption experiment results indicate that Cu(II) adsorption decreased for bulk soil and clay fractions after removing organic matter and Mn oxides. However, after free Fe oxides were removed, the Cu(II) adsorption increased for bulk soil and clay fractions. The adsorption capacity of the clay fraction for Cu(II) was higher than that of the bulk soil irrespective of the selective chemical extraction treatment. When the initial pH increased from 2 to 3.5, the adsorption capacity increased rapidly; when the pH > 6, the adsorption capacity did not change significantly. When the pH was greater than 6, increased Cu(II) adsorption may occur due to the precipitation of Cu(II). Increased ionic strength can inhibit Cu(II) adsorption. The effects of foreign ions on Cu(II) adsorption were in the order Na+  < K+  < Mg2+  < Ca2+ for cations and NO3−  < SO42− ≈ Cl− for anions. The Cu(II) adsorption was an endothermic and spontaneous process under the investigated experimental conditions.


Introduction
Copper is an abundant element in the lithosphere and is an essential metal for plant nutrition (He et al. 2005). It is present in limestone, poultry waste, phosphate fertilizer, agrochemical products, and sludge. When the concentration of Cu in soil solution is very high, it is toxic to plants and detrimental to the environment, and can ultimately harm human health (Silveira et al. 2002;Andrade et al. 2005). The solubility of Cu in soil solutions is controlled by adsorption/ desorption, complexation, and precipitation. These interactions determine Cu(II) mobility and bioavailability in soil and water environments (Silveira and Alleoni 2003;Violante et al. 2010).
Soil is a heterogeneous medium composed of different components (i.e., clay minerals, soil organic matter, and Fe/ Mn oxides). Due to differences in surface characteristics, the adsorption capacity of these components for metal ions is also different. Amorphous oxides and organic matter are the primary factors that control Cu adsorption (Parat et al. 2002). The transformation and enrichment of organic matter significantly influence the chemical mobility and retention of Cu in paddy soil (Wang et al. 2009a). Additionally, because of the differences in specific surface area, Fe/Mn/Al oxides, and organic matter contents, the adsorption capacity of various particle size aggregates to heavy metals is also different (Acosta et al. 2009(Acosta et al. , 2011Bi et al. 2013). The adsorption capacity of various particle size aggregates also varies for different heavy metal/metalloid species (Huang et al. 2017). The standard method to study the effects of soil components on Cu(II) adsorption is by using synthetic pure substances (i.e., goethite, hematite, and ferrihydrite). Adsorption characteristics of this substance for Cu(II) can then be analyzed. (Spathariotis and Kallianou 2007;Ellen and Caroline 2013;Sadullah et al. 2019). However, these methods do not accurately reflect the role of soil components naturally formed during the adsorption of heavy metals (Li et al. 2015). Another more effective method is the separation of the specific components in soil through selective chemical separation methods and then the use of adsorption experiments to evaluate the role of the components on the adsorption of heavy metals in the soil (Agbenin and Olojo 2004;Vega et al. 2007). Researchers have even used this method to study the influence of specific components in river surficial sediments on Cu adsorption (Li et al. 2009a, b;Wang and Li 2011).
The adsorption of Cu(II) is also controlled by the chemical properties of the soil solution (i.e., pH, foreign ions, ionic strength) (Dai et al. 2017). In general, pH can affect the speciation of metal ions through hydrolysis reactions, and thus, affects the adsorption performance of metal ions (Yun et al. 2001;Sheng et al. 2011). Ionic strength can affect the solution-phase activity of metals, the number of competing ions, and the properties of the electric double layer (Baker et al. 2009;Zhou and Haynes 2010). Foreign cations can compete with metal ions for adsorption sites. Due to the differences in the valence and hydration ratios of foreign cations, their influence on the adsorption of metal ions varies (Guo et al. 2011). Foreign anions (i.e., NO 3 − , SO 4 2− , Cl − , or ClO 4 − ) can form anion-soluble metal-water complexes with metal ions. Different anions and metal ions form different metal-anion complexes. The size and charge properties of different metal-anion complexes are different. In addition, the affinity of different complexes to the surface of adsorbents is different (Tao et al. 2004). These complex interactions are affected by the radius and other properties of foreign anions . Previous studies focused on the effects of certain mineral materials on Cu(II) adsorption under different solution chemical properties. For example, Wu et al. (2011) found that pH and ionic strength have pronounced effects on the adsorption of Cu on montmorillonite. Different background cations have different effects on Cu(II) adsorption, while the effects of different background anions on Cu adsorption are not apparent. Komy et al. (2014) studied the effect of humic acid on Cu(II) adsorption on kaolinite and hematite under different solution chemical properties. However, few studies have been conducted on the influence of different solution chemical properties (i.e., pH, ionic strength, and coexisting ions) on the adsorption of Cu(II) by soil.
Generally, alluvial plains are areas with a flat terrain and developed agriculture, where the soil can easily be contaminated by heavy metals. Therefore, this study takes alluvial acid soil as the research object, and the aims of this study are to (1) investigate the effects of specific components of alluvial acid soil (clay fraction, organic matter, free Fe oxides, and Mn oxides) on Cu(II) adsorption; (2) explore the influence of pH, foreign ions, and ionic strength on Cu(II) adsorption; and (3) evaluate the thermodynamic processes of Cu(II) adsorption. The research results can help understand the adsorption mechanism of Cu(II) in alluvial acid soil.

Soil sampling and analysis
The sampling site is located in a rural area near a coal mine in Sichuan province, southwest China (Fig. 1). The soil is alluvial acid soil. The topsoil samples (0-20 cm) were collected, air-dried, crushed, and passed through a 2-mm nylon mesh.
The pH was determined at a soil:water ratio of 1:2.5(w/w) using a pH meter; the organic matter was determined by the K 2 Cr 2 O 7 oxidation method (Pansu and Gautheyrou 2006). The contents of free Fe oxides and Mn oxides were determined by the dithionite-citrate-bicarbonate (DCB) method (Kunze and Dixon 1986) and NH 2 OH·HCl method (Wang and Li 2011), respectively. The cation exchange capacity (CEC) was determined according to ISO 23470. The grainsize distribution was determined according to (GBT50123-1999). After HNO 3 -HClO 4 -HF digestion (Friese et al. 2010), the Fe/Mn and Cu(II) contents of the soil were analyzed by AAS (GGX-9, Beijing Haiguang Instrument Co., China) and ICP-MS (PE6000, Perkin Elmer Corp, USA), respectively. Three parallel samples and one blank sample were made.
The mineral analysis was conducted using an X-ray diffractometer (DX-2700, HAOYUAN) that ran at 40 kV and 30 mA, using a Cu tube. Fourier transfer infrared (FTIR) spectra were recorded in the spectral range of 4000-400 cm −1 on an FTIR spectrophotometer (Nicolet is 10, Thermo Fisher Scientific) with the KBr pellet method at room temperature of 25℃. The surface morphology of the studied soil was observed using a scanning electron microscope (SEM, Prisma E, Thermo Scientific) with energy-dispersive X-ray (EDS) detectors for elemental analysis. The sample was gold-plated in a vacuum environment before observations.

Adsorption experiments
The clay fraction (< 2 µm) samples were separated from the studied soil. Before fractionation, 100 g of soil was immersed overnight in a glass beaker containing 1000 mL of deionized water and later dispersed through ultrasonication (energy was set to approximately 100 J/ml) (Schmidt et al. 1999). Subsequently, natural sedimentation and siphoning were conducted as per Stock's law. The separated sample was centrifuged and air-dried. Selective extraction technology was employed to remove specific components (organic materials, free Fe oxides, and Mn oxides) of the bulk soil and clay fraction. In the experiment, 30% H 2 O 2 was used to remove organic materials (Wu et al. 2012); the 0.3 M Na 3 C 6 H 5 O 7 ·2H 2 O + 1MNaHCO 3 + Na 2 S 2 O 4 (DCB) method was used to remove free Fe oxides (Wang et al. 2009a) and 0.1 M HONH 3 Cl (pH = 2) removed Mn oxides (Wang and Li 2011).
All adsorption experiments were carried out in 50-mL polyethylene centrifuge tubes on a thermostatic oscillator with an oscillating speed of 180 rpm. The Cu(II)-containing solution was prepared with Cu(NO 3 ) 2 ·3H 2 O. The pH of the solutions was adjusted using HNO 3 and NaOH. After the adsorption experiment, the solution was centrifuged at 6000 rpm for 20 min and then filtered. The content of Cu(II) in the supernatant was determined using an atomic absorption spectrophotometer, and relative errors in the experimental data were low (p < 5%).

Experimental data analysis
The amounts of elements adsorbed onto soil were calculated using the following equation (Kumar et al. 2009): where q e is the amounts of Cu(II) adsorbed at equilibrium, (mg kg −1 ); C 0 and C e are the initial and equilibrium concentrations of Cu(II) in the solution (mg L −1 ); V is the volume of the solution (L); and m is the dry weight of soil (kg).
The distribution coefficient (K d ) represents the ratio of the concentration of Cu(II) in the solid phase to that in the liquid phase after a certain adsorption time (Eq. 2); the greater the K d , the easier the adsorbate is adsorbed by the adsorbent and the less migratory the adsorbate.
The thermodynamic parameters of the adsorption of Cu(II) onto the soil, i.e., the standard entropy ΔS 0 , enthalpy where R is the ideal gas constant (8.3145 J/mol/K) and T (K) is the Kelvin temperature. ΔH 0 and ΔS 0 were calculated from the plot of lnK d versus 1/T.

Physical and chemical properties of the studied soil
The main physical and chemical properties of the studied soil were as follows: clay content of 33.2%, silt content of 54.6%, sand content of 12.2%, pH of 5.28, soil organic matter of 1.84%, the CEC was 8.575 mmol kg −1 , and the contents of free Fe oxides and total Fe were 25.76 and 41.91 g kg −1 , respectively. The total Mn content was 1.01 g kg −1 and the Cu(II) content was 28.7 mg kg −1 . Figure 2 shows the X-ray diffraction pattern of the soil. The crystal phase of the bulk soil is mainly composed of quartz, feldspar, and muscovite. For the clay fraction, the peak of feldspar disappeared, and the peak intensities of quartz and muscovite weakened; however, the characteristic peaks of clay minerals were not evident. This may be because the clay

X-ray diffraction analysis
(4) ΔG 0 = −RT ln K d minerals were covered by some amorphous components, masking their characteristic peaks.

FTIR spectroscopy
The FTIR spectra of the bulk soil and clay fraction are shown in Fig. 3. The absorption peak at 3620 cm −1 is due to O-H related to free water (Liu et al. 2011). The peak at 1630 cm −1 corresponds to the aromatic and carboxylate asymmetric stretching (Farinella et al. 2007). The pronounced peak at 1030 cm −1 is attributed to Si-O (Madejová 2003), and the sharp peaks at 797 and 779 cm −1 are attributed to quartz. The peak at 533 cm −1 corresponds to Al-O-Si, which may indicate the presence of montmorillonite (Yang et al. 2009).

Organic matter
Organic matter contains various functional groups such as carboxyl, carbonyl, and phenolic groups (Guo et al. 2006), that affect metal ion adsorption, migration, and transformation behaviors in soil. After the removal of organic matter, as seen in Fig. 5(a), the adsorption capacity of Cu(II) for the bulk soil decreased by a maximum of 53% (C 0 :10 mg L −1 ), and as per Table 1, K d decreased by a maximum of 83% (C 0 :10 mg L −1 ). For the clay fraction,, as seen in Fig. 5(b), the adsorption capacity of Cu(II) decreased by a maximum of 36% (C 0 :50 mg L −1 ), and as seen in Table 2, K d decreased by a maximum of 74% (C 0 :10 mg L −1 ). Moreover, the adsorption capacity of Cu(II) on the clay fraction was higher than that on the bulk soil, irrespective of the removal of organic matter. This phenomenon may be due to the larger surface areas and the variable negative charge densities of the clay fraction (Dong et al. 2006). Organic matter has various reactive functional groups and is an essential adsorbent for heavy metals in the soil. The minerals in the soil are mostly coated by organic matter, existing as organo-mineral composites (Mikutta et al. 2014;Wang et al. 2016). After removing organic matter, since the main minerals of the clay fraction are clay minerals, it still has a strong adsorption capacity for Cu(II). However, quartz, the primary mineral in the studied soil (Fig. 2), has a low adsorption capacity. After the organic matter is removed, the adsorption capacity of the bulk soil for Cu(II) is greatly reduced. Other studies have also reported the effect of soil organic matter on Cu(II) adsorption. When organic matter in the Bt horizon of a savanna Alfisol was removed, K d reduced by more than 20 times than in the bulk soil (Agbenin and Olojo 2004). After removing organic matter, the adsorption capacity of Cu(II) by the clay fraction in a paddy field reduced by 15% (Wang et al. 2009a). However, Cavallaro et al. (1984) reported that removing the organic matter from the clay fraction has little effect on Cu(II) adsorption upon studying acidic soil from New York, USA. Presumably, the effect of organic matter on Cu(II) adsorption varies with soil types.

Free Fe oxides
Fe oxides and Fe (oxy-) hydroxides act as sinks (i.e., Fe-nodules and Fe-rohrensteins) for heavy metals and metalloids, thereby controlling their mobility in soil and groundwater environments (Cornell and Schwertmann 2003;Vodyanitskii et al. 2009). However, after removing free Fe oxides by the DCB method, Cu(II) adsorption capacity increased for the bulk soil and clay fraction. For the bulk soil, as shown in Fig. 5(a), the adsorption capacity of Cu(II) increased by a maximum of 43% (C 0 :30 mg L −1 ), and as seen in Table 1, K d increased by a maximum of 422% (C 0 :10 mg L −1 ). For the clay fraction, as shown in Fig. 5(b), adsorption capacity increased by a maximum of 26% (C 0 :200 mg L −1 ), and as seen in Table 2, K d increased by a maximum of 196% (C 0 :10 mg L −1 ). In addition, after free Fe oxides were removed, the adsorption capacity of the clay fraction for Cu(II) was still higher than that of the bulk soil at the same initial concentration of Cu(II).
The adsorption capacity of Cu(II) by the bulk soil and clay fraction increased after removing free Fe oxides. Wu et al. (1999) reported similar results, with the removal of free Fe oxides enhancing the adsorption capacity of fine, medium, and whole clay to Cu(II) but having little effect on coarse clay. Agbenin and Olojo (2004) reported that Cu(II) adsorption was slightly higher than that of the untreated soil after removing free Fe oxides from a Bt horizon of a savanna Alfisol. However, the adsorption of Zn(II) was significantly enhanced compared to natural soil. In addition, it has also been reported that the adsorption of Cd(II) on soil increases after the removal of free Fe oxides (Li et al. 2015). The adsorption capacity of heavy metals increased after removing free Fe oxides, possibly because the Fe oxides are usually present on the surface of clay minerals, and their removal exposes the adsorption sites of clay minerals (Wu et al. 1999). However, Silveira et al. (2002) reported that the Cu(II) adsorption capacity decreased significantly after removing Fe oxides from tropical soils that naturally have high contents of Fe. Wang et al. (2009a) reported that the adsorption capacity of Cu(II) for all size fractions of aggregates significantly reduced after removing free Fe oxides from two types of paddy soils, China. Li et al. (2018) reported that the removal of free Fe oxides from three types of soil (ferrosol, primosol, and isohumosol) reduced the adsorption capacity for Sb(III). From the above analysis, it can be seen that the effect of the free Fe oxides on the adsorption of heavy metals to soil varies with heavy metal types, physicochemical properties of the soil, and size fractions of aggregates.

Mn Oxides
In the supergene environment, Mn is primarily present as oxides, which usually form a black crust or coat that is fixed on the surface of rock or soil particles or filled in cracks (Liu 1984). They usually have large specific surface areas and strongly acidic sites that greatly affect the migration and transformation of heavy metals in the soil. The adsorption of heavy metals onto Mn oxides is primarily considered to be a specific sorption (Spark et al. 1995). After the removal of Mn oxides, for the bulk soil, as shown in Fig. 5(a), the adsorption capacity of Cu(II) decreased by a maximum of 54% (C 0 :30 mg L −1 ), and as seen in Table 1, K d decreased by a maximum of 68% (C 0 :30 mg L −1 ). For the clay fraction, as shown in Fig. 5(b), the adsorption capacity decreased by a maximum of 46% (C 0 :30 mg L −1 ). As seen in Table 2, K d decreased by a maximum of 73% (C 0 :30 mg L −1 ). It can be seen from Fig. 5(a, b) that when the initial Cu(II) concentration is greater than 30 ml/L, the samples without Mn oxides had the lowest adsorption capacity of Cu(II). Therefore, Mn oxide is an essential adsorbent for Cu(II) in the studied soil. Li et al. (2009a, b) found that the adsorption capacity of Mn oxides in River Surficial Sediments for Cu and Zn was about one order of magnitude and nearly two orders of magnitude greater than that of Fe oxides and organic matter, respectively. Part of the reason for the strong adsorption capacity of Mn oxides for metal ions is their complex mineralogical structure. Generally, Mn oxides exist in multiple oxidation states; Mn forms nonstoichiometric oxides with variable valence states (Hall et al. 1996). Moreover, Mn oxides have a low zero point of charge. It is easy to generate a large number of negative charges under environmental pH values, allowing Mn oxides to have a strong ability to attract metal ions (Jin 1992). However, the influence of Mn oxides on Cu(II) adsorption is closely related to their content in soil (Bibak 1997).

Initial pH
The pH value affects the chemical properties of metal ions through precipitation, hydrolysis, complexation, and redox reactions and the mineral's surface properties (Esposito et al. 2002;Yang and Volesky 1999). The speciation of Cu (Cu 2+ , Cu(OH) + , Cu 2 (OH) 2 2+ , and Cu(OH) 2 0 ) in a solution corresponds to different pH ranges, which indirectly affect the adsorption of Cu(II) (Li et al. 2009a, b). As Fig. 6 shows, when the pH increased from 2 to 3.5, the Cu(II) adsorption increased rapidly from 77.3 to 520.5 mg/kg, i.e., by 573%. The K d increased from 3.0 to 130.9 L/kg, i.e., by 4263%. This may be because H 3 O + gradually leaves the adsorption sites of the adsorbent and is replaced by Cu(II). This process is similar to ion exchange interactions (H + / Cu 2+ ) (Singh et al. 1993). When the pH ranged between 3.5 and 6, the Cu(II) adsorption did not change significantly. when the pH was greater than 6, the adsorption capacity of Cu(II) increased again; this may be due to the precipitation of Cu(OH) 2 , which led to a subsequent increase in Cu(II) adsorption. Owing to the soil buffering power (i.e., ion exchange of soil colloids, dissociation of a strong base and weak acid salt, etc.), the pH of the equilibrated solution tends toward the actual pH of the studied soil.

Ionic strength
Ionic strength can affect the number of competing ions and the activity of metal ions in the solution, the interface potential, and the properties of the electric double layer (Zhou and Haynes 2010;Baker et al. 2009). As shown in Fig. 7 and Table 3  0.1 M, the adsorption capacity decreased by a minimum of 5% (C 0 :10 mg L −1 ) and a maximum of 36% (C 0 :200 mg L −1 ), K d decreased by a minimum of 64% (C 0 :100 mg L −1 ) and a maximum of 84% (C 0 :30 mg L −1 ). At the same ionic strength, K d decreased with an increased in Cu(II) ( Table 4).

Foreign ions
Natural soil pore water contains different electrolytes due to the different sources and water quality, affecting the adsorption of metal ions in soil (Boudrahem et al. 2011). As shown in Fig. 8 and Tables 5 and 6, the adsorption capacity and K d of Cu(II) adsorption are the lowest when the foreign cation is Ca. The order of effect of foreign cations on Cu(II) adsorption is Ca 2+ > Mg 2+ > K + > Na + . Bivalent ions (Ca 2+ and Mg 2+ ) have a greater effect than monovalent ions (K + and Na + ) as bivalent ions have a stronger ability to compete for adsorption sites (Naidu et al. 1994). Additionally, because of the hydration radius of K + = 2.32 Å < Na + = 2.76 Å (Guo et al. 2011), the effect of K + on Cu(II) adsorption is more evident than that of Na + .    10  169  195  160  163  172  195  167  30  482  557  413  429  534  557  516  50  672  878  543  574  782  878  757  70  825  1112  660  735  1040  1112  970  100  1014  1399  800  867  1265  1399  1199  200  1505  1969  1020  1140  1884 1969 1921 The order of effect of foreign anions on Cu(II) adsorption is Cl − ≈ SO 4 2− > NO 3 − . This may be because SO 4 2− and Cl − are more likely to form complexes with Cu(II) than NO 3 − in aqueous environments ). However, Yang et al. (2006) found that different types of background anions (Cl, NO 3 , ClO 4 , NO 2 , SO 3 , and PO 4 ) do not affect the adsorption amounts and adsorption trends of Cu(II).

Effect of temperature
The adsorption experiments were conducted under two initial concentrations of Cu(II) (50 and 100 mg L −1 ) and three temperature conditions (293.15, 301.15, and 313.15 K). High temperatures can accelerate the diffusion rate of a sorbate in aqueous solutions (Gregorio and Pierre 2008), thus increasing Cu(II) adsorption. As shown in Fig. 9, an increase in temperature is conducive to the adsorption of Cu(II).
As the Table 7 shows, A positive ΔH 0 value and a negative ΔG 0 value indicate that the adsorption was an endothermic and spontaneous process under the experimental conditions. Furthermore, ΔG 0 decreased with increasing temperature, indicating that the adsorption process was more efficient at high temperatures. The values of ΔS 0 are positive, indicating that the soil has a good affinity for Cu(II) in the solution and that some structural changes may occur in the adsorbents (Genc-Fuhrman et al. 2004).

Conclusions
Based on the experimental results, the following conclusions can be drawn: (1) the adsorption capacity of the clay fraction to Cu(II) is higher than that of bulk soil. The removal of soil organic matter and Mn oxides decreases the Cu(II) adsorption capacity for both the bulk soil and clay fraction. the opposite trend was observed for free Fe oxide removal. (2) As the pH increased, the adsorption first increased rapidly, and then tended toward equilibrium. At a pH greater than 6, the Cu(II) precipitated. (3) At the same initial concentration of Cu(II), Cu(II) adsorption decreased with an increase in ionic strength. (4) The effect of foreign bivalent cations on Cu(II) adsorption is greater than that of monovalent cations. Cl − and SO 4 2− have greater effects than NO 3 − on Cu(II) adsorption. (5) Thermodynamic studies showed that Cu(II) adsorption was an endothermic and spontaneous process. These experimental data and conclusions support acidic mineral water discharge management and the application of agricultural fertilizers in coal mining areas.