1 Introduction

Phosphorus (P) deficiency is one of the main factors restricting crop yields in many soils. Although P is naturally present at a high concentration in some soils worldwide, it is often a limiting nutrient in many other soils, in part due to strong adsorption on the surface of Fe(III)-(hydr)oxides in soils (Guzman et al. 1994; Jaisi et al. 2010; Johnson and Loeppert 2006). Ferrihydrite (F) is the most effective Fe-oxide for phosphate (P) adsorption in soils due to its small particle size, high surface area, and high reactivity (Antelo et al. 2010). In nature, F is formed by the rapid oxidation of Fe(II) in Fe-rich waters, yielding an amorphous material (Rhoton et al. 2002), which occurs widely in surface environments such as soils and sediments. Hence, the adsorption of P on F has an important role in controlling transport, fate, and bioavailability of P in soils.

Biochar is a carbon-rich solid material produced by pyrolysis of biomass and can be added to soils as a means of abating climate change by sequestering carbon (Liu et al. 2011; Woolf et al. 2010), while also improving soil functions (Atkinson et al. 2010; Yuan et al. 2011). Biochar amendments can improve soil nutrient-holding capacity for crop production (Liang et al. 2006; Novak et al. 2009) and encourage a host of beneficial microorganisms (Kolb et al. 2009). Moreover, it has shown strong adsorption affinity for organic compounds and heavy metals (Cao et al. 2009; Chen and Yuan 2011; Lou et al. 2011; Namgay et al. 2010; Wang et al. 2010; Zheng et al. 2010).

Several studies have demonstrated enhanced P uptake in the presence of biochar (DeLuca et al. 2009; Novak et al. 2009). Biochar can alter P availability directly through its anion exchange capacity or by influencing the activity/availability of cations that interact with P (DeLuca et al. 2009). Biochar can also provide indirect effects on P availability and uptake through changes in the soil environment for microorganisms (Atkinson et al. 2010). It is well known that a layer of Fe-(hydr)oxide can be deposited or precipitated on part of the biochar after application of biochar to soils (Joseph et al. 2010). Although P adsorption characteristics of single-component Fe-(hydr)oxides have been widely studied (Arai and Sparks 2001; Jaisi et al. 2010; Khare et al. 2007), limited work has been conducted on the adsorption characteristics of the F–B complex to determine if it directly or indirectly influences the availability of P in soils. Therefore, the primary objective of this study was to investigate the P adsorption and desorption on F with and without biochar (F–B complex or F) and thereby evaluate the behavior of P in soils with added biochar.

2 Materials and methods

2.1 Materials

Biochar was produced by pyrolysis of rice straws obtained from a farm in Xiamen, China. It was air-dried for 1–2 days and then cut into pieces of approximately 2–3 cm in length. These pieces were pyrolyzed at 600°C in steel ring furnaces and sieved through a 1-mm screen to prepare biochar. Two-line ferrihydrite was synthesized as described by Schwertmann and Cornell (2000). In a typical synthesis, 40 g of Fe(NO3)3·9H2O was dissolved in 500 mL double-distilled water (DDW), and 330 mL of KOH (1 mol L−1) was added to bring the pH to 7–8, while stirring vigorously. The solids were washed with DDW until free from electrolytes, and then freeze-dried. A ferrihydrite and biochar complex (F–B complex) was prepared under similar conditions, except that a mixture of 10 g biochar and the Fe(NO3)3 solution was used as the starting material instead of Fe(NO3)3 alone. All chemical reagents were analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) without further purification.

2.2 Characterization

The X-ray diffraction (XRD) patterns of the samples were analyzed with a Bruker D8 advance X-ray diffractometer equipped with a Cu Kα radiation source operated at a tube voltage of 40 kV and a tube current of 40 mA. The data were collected in a step scan mode using steps of 0.02° with a scanning rate of 10° min−1 in the 2θ range from 5 to 90°. The morphology of F, biochar, and F–B complex materials was determined using a Hitachi S-4800 scanning electron microscope (SEM). The Fe concentration was determined using a scanning electron microscope coupled with an energy-dispersive X-ray spectrometer. A Micrometrics ASAP 2020M+C system was used to measure the surface areas and micropore size distributions of the materials. N2 isothermal adsorption and desorption experiments were performed at relative pressures (P/P 0) from 10−6 to 0.9944 and from 0.9944 to 0.047, respectively. Pore size distribution was constructed by analyzing desorption data points using the BJH method. The point of zero charge (PZC) of the absorbents was determined by a rapid potentiometric titration method (Cao et al. 2009, 2007). All Fourier transform infrared spectroscopy (FT-IR) spectra were obtained with a Thermo Scientific Nicolet iS10 spectrophotometer using pellets with KBr powder. Samples were scanned 15 times between 4,000 and 400 cm−1 at a resolution of 4 cm−1. Since FT-IR spectra subtraction was done at the slope of the absorption bends of the ferrihydrite (<1,100 cm−1), it is difficult to preserve the P (υ1) vibration modes at <800 cm−1 if present. Therefore, the results discuss mainly changes in the P (υ3) vibration modes (825–1,175 cm−1).

2.3 Adsorption and desorption experiments

P adsorption capacities of F and F–B complex materials were evaluated by measuring the time-dependent concentrations of P and total P in a batch system. In the experiment examining the pH edge, 0.2 g of F or F–B complex was mixed with 100 mg L−1 of KH2PO4 solution (100 mL) at pHs of 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, and 10, monitored with a pH-stat (TIM865, Titration Manager, Radiometer Analyzer). The suspensions were agitated on a shaker at 250 rpm and 25°C for 24 h, after which 5 mL of the solutions was intermittently sampled and filtered through a 0.22-μm syringe filter.

For P adsorption, 0.2 g of F or F–B complex was mixed with 100 mg L−1 of KH2PO4 solution (100 mL) at pH 4.5. At time zero and at selected time intervals thereafter, 5 mL of the suspensions was taken and filtered through a 0.22-μm syringe filter. To examine the effect of initial P concentration on the P adsorption by the F and F–B complex materials, 2.0 g L−1 of F or F–B complex was contacted with 0, 0.2, 1, 2, 5, 10, 20, 50, 100, 200, and 300 mg L−1 of KH2PO4 at pH 4.5. The suspensions were agitated on a shaker at 250 rpm and 25°C for 24 h, after which the suspensions were intermittently filtered through a 0.22-μm syringe filter, and the solids were collected for measurement of P desorption.

The desorption of P loaded on F or F–B complex samples was done using 0.01 mol L−1 of KCl or citric acid as desorbing agents. The suspensions were agitated on a shaker at 250 rpm and 25°C for 24 h. At the end of the experiment, the suspensions were filtered through a 0.22-μm syringe filter.

The residual P concentration was measured by an ion chromatograph (Dionex ICS-3000). Batch experiments were conducted in three replicates, and the results reported are mean values of the experimental data.

3 Results and discussion

3.1 Characterization of the samples

The XRD pattern of ferrihydrite showed two broad reflection peaks (Schwertmann and Cornell 2000). Biochar and F–B complex materials of XRD patterns showed trace quartz contents (i.e., 3.34 and 4.26 Å). The SEM image showed that the synthesized F presented as an aggregation of the nanoparticles (Fig. 1a). The biochar consisted of irregular plates and particles with different size ranges, and the surface of the biochar was smooth (see Fig. 1b). F was mainly coated on the surface of the biochar in the F–B complex materials (see Fig. 1c), and aggregation of the coated F also can be found on the surface of the biochar (see Fig. 1d). The properties of different samples are listed in Table 1. Clearly, the PZC of the F–B complex (6.5) is lower than that of the F sample (8.3), but the Brunauer–Emmett–Teller (BET) surface area of the F–B complex (314 m2 g−1) is higher than that of the F sample (219 m2 g−1). In addition, the Fe content of the F sample (67%) was slightly higher than that of the F–B complex (60.5%).

Fig. 1
figure 1

SEM images of a ferrihydrite, b biochar, c F–B complex, and d an enlarged detail of F–B complex materials (c)

Full size image
Table 1 Selected properties of adsorbents
Full size table

3.2 pH edge affect on P adsorption

Earlier studies have shown that solution pH is an important parameter affecting adsorption of phosphate on Fe-oxides (Luengo et al. 2007; Zeng et al. 2004; Zhong et al. 2007). Figure 2 demonstrates that the adsorption of P depends mainly on the pH of the aqueous solutions with the F and F–B complex materials, and the P adsorption decreases with increased solution pH, which can also be observed on the single-component Fe-(hydr)oxides (Arai and Sparks 2001). This is likely because the surface of the F nanoparticles bears more negative charge as the pH increases, causing a greater electrostatic repulsion toward the more negatively charged P forms (HPO 2−4 or PO 3−4 ) that predominate at the higher pH. Therefore, the lower adsorption of P at higher pH resulted from an increased repulsion force between the more negatively charged HPO 2−4 or PO 3−4 species and the negatively charged of the surface sites on F, and also because of increased OH− ion competition. The single-component F in the absence of biochar (B) clearly shows higher adsorption ability than that of F combined with biochar materials (F–B complex). At a pH range from 4 to 10, the P adsorption capacity of F in the absence of biochar is about 30–40% higher than that of F combined with biochar (F–B complex). Thus, the presence of biochar decreased the adsorption of P on the surface of F. The decreased adsorption capacity of F can be attributed to the lower PZC of the F–B complex than that of F (see Table 1).

Fig. 2
figure 2

Effect of pH on P adsorption at 25°C: adsorbent dose, 2 g L–1; initial concentration, 100 mg L–1 of KH2PO4 solution

Full size image

3.3 Adsorption isotherms

The amount of P adsorbed by the F and F–B complex increased with increasing initial P concentrations (Fig. 3). In the low concentration region (<20 mg L−1), the F and F–B complex materials are able to reduce the equilibrium P concentration to trace levels. At higher initial P concentrations, however, these F and F–B of two materials show different adsorption behaviors, with P adsorption capacity of F much higher than that of F–B complex.

Fig. 3
figure 3

Equilibrium isotherm plots at 25°C. Solid and dashed lines represent the simulation using the Langmuir and Freundlich isotherm, respectively

Full size image

Freundlich and Langmuir equations were fitted to the P adsorption data. The equations were as follows:

$$ {Q_{\text{e}}} = { }{K_{\text{f}}} \times {C_{\text{e}}}^{{{1}/n}} $$
(1)
$$ {Q_{\text{e}}} = { }b \times {Q_{\text{m}}}/\left( {{1 } + { }b \times {C_{\text{e}}}} \right) $$
(2)

where Q e is the amount adsorbed per unit weight of sorbent at equilibrium (milligrams per gram), C e is the equilibrium concentration (milligrams per liter), K f is the Freundlich adsorption constant related to the P adsorption capacity of the adsorbent ((milligrams per gram)(liters per milligram)1/n), and 1/n is the adsorption intensity, which indicates the favorability of adsorption. In addition, Q m is the sorption capacity (milligrams per gram), and b is the affinity coefficient (liters per milligram). The Langmuir parameter Q m is 44.2 mg g−1 for F and 28.5 mg g−1 F for the F–B complex. The Freundlich model gave the higher r 2 values, showing that the adsorption of P onto F and F–B complex materials could be better described by the Freundlich model than by the Langmuir equation (see Fig. 3). These results indicated that the active adsorption sites were energetically heterogeneous and were not restricted to the formation of a monolayer.

3.4 Adsorption kinetics

Phosphate was rapidly adsorbed onto the F and F–B complex (Fig. 4). Nearly 80% of the P adsorption capacity was reached within 2 h, and then, P adsorption increased steadily to reach equilibrium in about 24 h. This slow adsorption rate following an initially rapid adsorption of P has been observed previously (Jaisi et al. 2010). In order to investigate the mechanism of adsorption and potential rate-controlling steps, the pseudo-first-order, pseudo-second-order, and Elovich equation kinetic models have been employed to fit the experimental data, and the kinetic parameters for P adsorption are listed in Table 2. The calculated q e values from the pseudo-first-order and pseudo-second-order linear plots did not agree well with the experimental ones, indicating that the adsorption of P onto F probably did not follow the pseudo-first-order and pseudo-second-order kinetic models. Compared with r 2 values, kinetics data for the adsorption of P from aqueous solution were in good agreement with the Elovich equation. The validity of the Elovich equation suggests that P adsorption on the F and F–B complex materials is a chemical adsorption on highly heterogeneous adsorbents, and chemisorption mechanism is likely rate controlling in the processes (Chang and Juang 2005; Gupta and Bhattacharyya 2006). The α value of F is much higher than that of F–B complex materials (see Table 2), and the β shows the opposite trend, indicating that the initial rate of adsorption of F is faster than that of F–B complex materials. Thus, the rate of P chemisorption on F can be decreased with the presence of biochar (B).

Fig. 4
figure 4

Adsorption kinetics of P onto F and F–B complex materials: adsorbent dose, 2 g L–1; P concentration, 100 mg L–1; initial pH 4.5; temperature, 25°C

Full size image
Table 2 Kinetic parameters for P adsorption on F and F–B complex materials
Full size table

3.5 Desorption studies

The rate of P desorption from the Fe-oxides is clearly dependent on the particular desorption conditions, which include concentrations of various cations in solution and the gradient of P concentration. The P desorbability can be defined as the ratio of the desorbed P to the total adsorbed P by the adsorbents. Therefore, the desorbability of P can be used to indicate the degree of P desorption from the adsorptive materials. For P desorption study, the adsorbed materials were treated with 0.01 mol L−1 KCl or citric acid, and the data are presented in Fig. 5. The desorbability of adsorbed P on the F and F–B complex materials clearly increased with increasing initial equilibrium concentrations of P for all treatments. At low initial P adsorption level (≤50 mg L−1), there was no detectable P released from the F and F–B complex materials by KCl. Previous studies have reported that P can be adsorbed to Fe-oxides with monodentate and bidentate surface complexes (Antelo et al. 2010; Arai and Sparks 2001; Parfitt and Atkinson 1976; Zhong et al. 2007). At relatively low P concentration (≤50 mg L−1), the surface species are bidentate surface complexes, which are difficult to extract by KCl solution. A minor contribution of a monodentate surface complex can be formed at high initial P concentrations (100 and 200 mg L−1), and they could be released by KCl solution (Khare et al. 2007; Liu et al. 1995). On the other hand, there was also no detectable P released from the F and F–B complex by citrate acid with initial P equilibrium concentration at 0.2 mg L−1. However, citrate acid released P from the F and F–B complex materials more effectively than KCl solution at higher initial P equilibrium concentration (>0.2 mg L−1), and ligand exchange should contribute to the greater extent to P release. The desorbability of adsorbed P on F–B complex is higher than that of F, indicating that the presence of biochar can enhance the desorbability of adsorbed P on F at higher initial P equilibrium concentration (>0.2 mg L−1).

Fig. 5
figure 5

Effect of initial equilibrium concentration on P desorption from F and F–B complex materials

Full size image

3.6 Mechanism of P adsorption on F and F–B complex materials

Many researchers have investigated P adsorption mechanisms on Fe-oxides, and the monodentate, bidentate mononuclear, or bidentate binuclear surface complexes on F have been suggested by several researchers (Antelo et al. 2010; Arai and Sparks 2001; Jaisi et al. 2010; Khare et al. 2007). Figure 6a, b of the FT-IR spectra shows the loading level effects on the P surface complexes for F and F–B complex materials at pH 4.5. The broad spectrum with 5 mg g−1 loading level for F is an assemblage of three vibration bends (i.e., 1,083; 1,034; and 965 cm−1), indicating these surface complexes are protonated bidentate complexes (Arai and Sparks 2001; Kwon and Kubicki 2004). As loading level increases, the FT-IR vibration mode at 1,037 cm−1 becomes intensified (see Fig. 6a), indicating that diprotonated monodentate complexes are present at high loading level (Arai and Sparks 2001; Kwon and Kubicki 2004). By comparison, the broad FT-IR spectrum with 5 mg g−1 loading level for the F–B complex is also an assemblage of three vibration modes (i.e., 1,105; 1,050; and 977 cm−1), and the loading level does not show to have a strong influence on the FT-IR spectra for the F–B complex (see Fig. 6b), suggesting that both diprotonated bidentate and diprotonated monodentate complexes can be formed on the surface of F–B complex. These differences can be attributed to the fact that the surface charge or function groups on the surface of ferrihydrite are altered by combining with biochar.

Fig. 6
figure 6

FT-IR spectra of P adsorption complexes with different loading levels for pH 4.5 at the F–water interface: a F and b F–B complex

Full size image

4 Conclusions

The adsorption and desorption of P on the surface of ferrihydrite combination with biochar were investigated to evaluate the influencing of biochar applications on interactions between P and Fe-oxide in soils. It was found that the P adsorption amount and adsorption rate on the surface of F decreased with the presence of B, and the desorbability of adsorbed P on F can be enhanced when combined with B. Thus, the presence of B can decrease P adsorption on the Fe-oxides and enhance P availability in P-fertilized soils.