Introduction

Excessive toxic metal ions in natural environments are always noxious and of great concern worldwide. They have adverse effects on organisms in ambient ecosystems and eventually damage human health via the food chains (Flouty and Estephane 2012). Many toxic metals are present in sewage and industrial wastewaters, including chromium, cadmium, copper, lead, nickel, zinc, mercury, and arsenic. Many techniques have been developed to remove these toxic metal ions from wastewater, such as chemical precipitation, solvent extraction, membrane filtration, ion exchange, and electrolytic and adsorption methods (Aguado et al. 2009). However, the application of most of these techniques is limited because of their high operational costs or subsequent ecological risks (Salehi and Madaeni 2010). In contrast, biosorption methods have attracted a lot of interest because they are usually low-cost and eco-friendly (De Philippis et al. 2003; Areco et al. 2012). Complex mechanisms are involved in the metal biosorption process, including absorption, complexation, ion exchange, microprecipitation, and electrostatic attraction (Chojnacka et al. 2005). Various biological materials including algal biomass have been tested for their adsorption ability, for example Arthrospira (Spirulina) platensis and Chlorella vulgaris (Gokhale et al. 2008). The algal cell wall, which metal ions are mostly bound to, is surrounded by a porous three-dimensional macromolecular network which affects the diffusion of metal ions (Chojnacka et al. 2005). Besides, these macromolecules including carbohydrates, proteins, and lipids displaying functional groups, such as carboxyl, amide/amine, hydroxide, and phosphate groups which are bound with metals, significantly affect the metal adsorption capability (Fang et al. 2011). Cyanobacteria species are generally believed to have great potential because of their high adsorption capability, high diversity, and low production cost (Colica et al. 2012; Rodrigues et al. 2012). Among them, extracellular polysaccharide-producing cyanobacteria have received considerable attention because exopolysaccharides have been shown to increase the metal ion adsorption ability due to abundant functional groups (De Philippis et al. 2001, 2003, 2011; Ye et al. 2014).

In practice, microalgae have disadvantages as biosorbents because they have small particle size which causes difficulty in separating biomass from effluent. Therefore, immobilized biomass particles are used to minimize this disadvantage (Bayramoglu and Arica 2009; Kumar et al. 2011). Alginate or Ca-alginate, polyacrylamide, polyvinyl alcohol, polysulfone, silica gel, cellulose, and glutaraldehyde are usually used in the immobilizing procedure, which greatly increases the production cost and limits the use of immobilized microbial adsorbents (Bayramoglu and Arica 2009; Rawat et al. 2013; Kanchana et al. 2014). Hence, large algal species are advantageous for biosorbents because of the low cost of production and convenience of separation, indicating that exopolysaccharide-producing cyanobacteria forming large-sized macrocolonies or floccules have potential as biosorbents (De Philippis et al. 2001).

Living and dead algal biomass have both been found to be good biosorbents for metal ions because of their high adsorption capacity (De Philippis et al. 2003; Gokhale et al. 2008; Mane and Bhosle 2012). Dead biomass may have higher adsorption capability than fresh biomass because functional groups complexing ions inside the cells are exposed on the surface during the drying and crushing processes, and the adsorption area greatly increases (Mehta and Gaur 2005; Gokhale et al. 2008). However, metabolic process in living biomass can affect the physicochemical mechanisms in biosorption, such as ion exchange, bioaccumulation, and transformation, which greatly affects the adsorption process (Markou et al. 2015). In consequence, different adsorption capacities can be found when comparing living and dead biomass. Gokhale et al. (2008) compared the adsorption capacities of living and dead algal biomass for Cr(VI), and they found that dead biomass was a better adsorbent. Tien et al. (2005) found no significant difference between living and dead algal cells for the removal of copper. Markou et al. (2015) found that living biomass exhibited a higher biosorption capacity for copper and nickel than dead biomass.

Nostoc sphaeroides Kützing is a cyanobacterium that usually grows in rice paddies. It produces exopolysaccharides in the form of sheaths to create large-sized spheroid macrocolonies 5–20 mm in diameter (Li et al. 2009). Based on the results of Huang et al. (1998), N. sphaeroides can produce large amounts of polysaccharides, which may be 30–40 % of the dry biomass. Large-sized colonies of this species solve the problem of separating the adsorbent in practical applications of common microbes and simplify the reuse process. These features indicate that it could be a good biosorbent for toxic metal ion removal. Other exopolysaccharide-producing Nostoc species, e.g., PCC7936, have shown the ability to remove metal ions from aqueous solutions (De Philippis et al. 2003). Recently, a mass culture method of N. sphaeroides has been established, which substantially decreases the production cost and increases the application potential (Deng et al. 2008). However, no studies about toxic metal adsorption of N. sphaeroides have been reported. Thus, it is necessary to elucidate the toxic metal adsorption capacity, metal adsorption specificity, adsorption characteristics, and mechanisms for N. sphaeroides.

According to many early reports, pH, temperature, initial metal concentration, and coexisting ions, are important for the adsorption process (Markou et al. 2015). Most researchers found that metal adsorption is pH dependent because H+ competes with metal ions for the binding sites, especially in low pH (Gadd 2009). Furthermore, pH determines the metal ion state in solutions, which greatly affects the mobility of metals. Some adsorption processes are endothermic. Therefore, higher temperature greatly favors the process (Gadd 1992). Greater initial metal concentration would lead to greater metal adsorption capability before all the binding sites are occupied. Other coexisting cations often compete with metal ions for the binding sites, as well as affect the metal complex, which greatly affects the metal adsorption process (Gadd 2009).

In this paper, we report the results of the adsorption capacity of fresh algal macrocolonies and dead algal biomass of N. sphaeroides for six metal ions: Cu(II), Cd(II), Cr(III), Pb(II), Ni(II), and Mn(II). The effects of contact time, pH, temperature, and other cations, as well as adsorption mechanisms were investigated.

Materials and methods

Nostoc sphaeroides Kützing was obtained from the Freshwater Algae Culture Collection of the Institute of Hydrobiology, Chinese Academy of Sciences. Cultures were unialgal, but axenic growing conditions were kept throughout in BG-110 (nitrogen-free medium based on BG-11) with aeration under 20 μmol photons m−2 S−1 fluorescent lights with a daily light period of 12 h at 25 ± 5 °C. Algal macrocolonies of 1–1.5 mm diameter were harvested using a 100-mesh sieve (150 μm) for the experiments. The spherical macrocolonies were carefully washed three times with sterilized deionized water, removed all the water on surface by sterilized filter paper, and the fresh weight was recorded before the experiments. Finally, the dry weight was obtained by oven drying at 80 °C until the weight did not change.

Algal powder of N. sphaeroides was obtained by grinding in a sterilized glass mortar with a sterilized glass rod and sieving (100 mesh sieve, 150 μm) oven-dried algal macrocolonies obtained by the above procedure.

Solutions of toxic metal ions

All of the metal ion working solutions of Pb(II), Cr(III), Cu(II), Mn(II), Cd(II), and Ni(II) were prepared from 1.0 g L−1 stock solutions at pH 5.0, using analytical reagent grade Pb(NO3)2, Cr(NO3)3.9H2O, CuSO4.5H2O, MnSO4.H2O, CdCl2.5H2O, and NiCl2.6H2O. Deionized water was used except for the adsorption experiments in tap water. The tap water (based on natural water from Yangtze River) contained Ca2+, Mg2+, Na+, K+, HCO3 −, CO3 2−, Cl−, and SO4 2− ions with the concentrations given in Table 1. pH of the tap water was 7.14. The tap water adsorption experiment was to simulate the adsorption in natural surface water in paddy fields in Yangzi River Basin.

Table 1 Concentrations of ions in the tap water used in the experiments (mg L−1)
Full size table

Toxic metals analysis

The concentrations of metal ions in the aqueous solutions were determined using an atomic adsorption spectrophotometer (Shimadzu AA-6300C, Japan) with flame methods. The correlation coefficient (R 2) of the standard curve of each metal ion was >0.999. The detection limits of the Ni(II), Cr(III), Mn(II), Cu(II), Cd(II), and Pb(II) were 0.04, 0.03, 0.01, 0.02, 0.01, and 0.2 mg L−1, respectively.

Biosorption of metal ions in batch experiments

Single metal ion batch biosorption tests in deionized or tap water

The biosorption of Pb(II), Cr(III), Cu(II), Mn(II), Cd(II), and Ni(II) to N. sphaeroides in single metal ion aqueous solutions prepared with deionized or tap water was measured in batch experiments in 50 mL Erlenmeyer flasks with plugs. The metal biosorption experiments in deionized water were performed at a high initial metal ion concentration of 10 mg L−1 and a low metal ion concentration of 0.5 mg L−1, while the biosorption experiments in tap water were only performed at an initial metal ion concentration of 10 mg L−1. The total volume of each flask was 30 mL. About 0.8 g fresh macrocolonies or 0.01 g dry powder of N. sphaeroides was added into each flask. The control group was with the same conditions except for the addition of algal material.

Multiple metal ion batch biosorption tests in deionized water

For the batch experiments with multiple metal ion solutions, solutions containing 10 mg L−1 Pb(II), Cr(III), Cu(II), Mn(II), Cd(II), and Ni(II) were prepared with deionized water in 50 mL Erlenmeyer flasks with plugs by mixing single metal solutions which were prepared separately at 60 mg L−1 at pH 5.0. The initial volume of the multiple solutions was 30 mL in the biosorption tests, and 0.8 g fresh algal macrocolonies was added into each flask.

In all of the above batch experiments, the pH was adjusted to 5.0 with 1 M HCl or 1 M NaOH. Each treatment was performed three times. All of the flasks were agitated at 120 rpm at 25 °C for 4 h. A supernatant sample (9 mL) was collected from each flask to determine the accurate residual concentrations of the metal ions. At the end of experiments, the algal macrocolonies in each flask were carefully collected, and the dry weight was determined.

Tests of biosorption variation of Pb(II) and Cr(III) under different conditions

Effect of contact time

Biosorption of Pb(II) or Cr(III) was performed in 500 mL of the 10 mg L−1 single metal ion solutions at pH 5, in which 13.3 g (wet weight) of fresh algal macrocolonies was added. Each experiment was performed three times. All of the flasks were agitated at 120 rpm at 25 °C. Samples were taken at time intervals of 2, 5, 10, 20, and 30 min and 1, 2, and 4 h for analysis of the residual metal concentration in the solution.

Effect of initial concentration of metal ions

To test the biosorption capacity for different initial metal concentrations, Pb(II) solutions with initial concentrations of 0.5, 2, 10, 50, and 100 mg L−1 and Cr(III) solution with initial concentrations of 1, 5, 10, 15, and 20 mg L−1 were prepared. The biosorption experiments were performed in 30 mL of the single metal ion solution in deionized water with 0.8 g fresh algal macrocolonies added under the same conditions as described above. The residual concentration of the metal ion was determined at the end of each experiment.

Effect of pH

The effect of pH on metal ion biosorption on algal fresh macrocolonies was performed by placing 0.8 g fresh algae into 30 mL of a 10 mg L−1 single metal ion solution for a series of controlled pH values of 3.5, 4.5, 5.5, and 6.5 for Pb(II) and 3.0, 4.0, 5.0, and 6.0 for Cr(III). The experimental conditions were the same as those described above, and the residual concentration of the metal ion was determined at the end of each experiment.

Effect of temperature

Biosorption experiments were performed by placing 0.8 g fresh algal macrocolonies into 30 mL of a 10 mg L−1 single metal solution for a series of temperatures at pH 5.0: 20, 25, 30, and 35 °C for Pb(II) and Cr(III). The experimental conditions were the same as those described above, and the residual concentration of the metal ion was determined at the end of each experiment.

In all of the above experiments, the algal macrocolonies in each flask were carefully collected, and the dry weight was determined at the end of each experiment. The adsorption capacity and removal rate were then calculated.

Calculation of adsorption capacity and removal rate

In all of the adsorption experiments, the adsorption capacity (mg g−1) was evaluated in terms of sorption capacity at time t (q t ) or at equilibrium (q eq ):

$$ \begin{array}{ccc}\hfill qt=\frac{\left({C}_0-{C}_t\right)\times V}{m}\hfill & \hfill or\hfill & \hfill {q}_{\mathrm{eq}}=\frac{\left({C}_0-{C}_{eq}\right)\times V}{m}\hfill \end{array} $$
(1)

where C 0, Ct, and C eq (mg L−1) are the initial time t, and the equilibrium metal ion concentrations, V (L), is the initial solution volume, and m (g) is the dry weight of the biosorbent.

The removal rate η (%) was calculated by the following equation:

$$ \eta \left(\%\right)=\frac{\left({C}_0-{C}_{eq}\right)}{C_0}\times 100 $$
(2)

FTIR analysis

Infrared spectral data of the fresh algal macrocolonies with and without adsorbed metals were obtained using a Nicolet 5700 FTIR spectrometer after freeze drying at −80 °C. The adsorption experiments were performed in single metal ion solutions (concentration of each metal 10 mg L−1) under the same experimental conditions as the batch experiments. The Fourier transform infrared spectroscopy (FTIR) spectra were recorded from 400 to 4000 cm−1 with 4 cm−1 resolution.

SEM–EDX analysis

Samples of fresh algal macrocolonies with and without adsorbed metals were freeze dried at −80 °C. A scanning electron microscope (FEI Quanta 200, The Netherlands) with an energy dispersive X-ray analyzer (Kratos XSAM800, England) at a voltage of 3.0 kV was used to identify the structural features on the surface of the samples.

Statistical analysis

All of the data (except for the analytical results of scanning electron microscopy–energy dispersive X-ray (SEM–EDX)) were obtained as mean ± standard error (SE) and analyzed in SPSS 16.0. One-way analysis of variance (ANOVA) was performed to compare the differences between treatments, with Duncan as multiple comparisons. The adsorption capacities and the removal rates were compared between fresh alga and algal powder with independent T samples for each metal ion. Normal distribution assumption was checked by the Kolmogorov–Smirnov test. Homogeneity of variance was ensured with or without data transformation and was confirmed by the Levene test. The significance level was 0.05.

Results

Adsorption experiments of the six metal ions

Fresh algal macrocolonies of N. sphaeroides showed significant but different adsorption abilities for the six metal ions in single metal ion solutions with deionized water (Treatments 1 and 2 in Table 2). At both initial concentrations of 10 and 0.5 mg L−1, the adsorption capacities of the six metal ions showed a similar order: Pb(II) > Cu(II) > Cr(III) > Cd(II) > Mn(II) > Ni(II) at 10 mg L−1, and Pb(II) > Cu(II) > Cd(II) > Mn(II) > Cr(III) > Ni(II) at 0.5 mg L−1. The highest adsorption capacity for Pb(II) was about 3–4 times the lowest adsorption capacity for Ni(II) under both initial concentrations. The removal rates showed similar patterns to the adsorption capacity for the six metals at both initial concentrations.

Table 2 Adsorption experiments of the six metal ions on Nostoc sphaeroides (mean ± SE, n = 3)
Full size table

Because of the competition between the adsorption of ions to the adsorbent, the adsorption capacities for all the metal ions significantly decreased when other ions were present in the solution (Treatments 1, 3, and 4 in Table 2). In the single metal ion solutions with tap water, the adsorption capacities for Ni(II), Cr(III), Mn(II), Cu(II), Cd(II), and Pb(II) decreased to 28, 93, 24, 30, 47, and 53 % of the values in deionized water, respectively. In multiple metal ion solutions, the adsorption capacities of Ni(II), Cr(III), Mn(II), Cu(II), Cd(II), and Pb(II) decreased to 8, 74, 21, 28, 19, and 62 % of the values in the single metal ion solutions with deionized water, respectively. The removal rate of metal ions in multiple metal ion solution showed a similar trend to the adsorption capacity. The above results showed that the adsorption capacities of Cr(III) and Pb(II) decreased less than the other four metals when other ions were present in the solution, indicating that the macrocolonies of N. sphaeroides had higher affinity for Cr(III) and Pb(II) than Ni(II), Mn(II), Cu(II), and Cd(II), and thus, N. sphaeroides can be applied as a biosorbent for Cr(III) and Pb(II) in real situations. For this reason, the following investigations of the adsorption properties and mechanisms were performed with Cr(III) and Pb(II).

Compared with fresh algal macrocolonies, the adsorption capacities of dried algal powder of N. sphaeroides were slightly but significantly greater for Cr(III) (22.70 vs 19.44 mg g−1 in the average, P = 0.005), significantly smaller for Mn(II) and Cu(II) (P = 0.002 and 0.035, respectively) and similar for Ni(II), Cd(II), and Pb(II) (P = 0.062, 0.247, and 0.175, respectively) (Treatments 1 and 5 in Table 2).

Adsorption kinetics of Pb(II) and Cr(III) on N. sphaeroides

The adsorption kinetics of Pb(II) and Cr(III) on fresh algal macrocolonies of N. sphaeroides are shown in Fig. 1. Both the kinetic curves show that adsorption was initially rapid and then greatly slowed. The adsorption capacities (mg g−1) for Pb(II) and Cr(III) increased with increasing contact time, and reached an equilibrium at about 10 and 20 min, respectively. The adsorption rates of Pb(II) and Cr(III) in the initial stages were 2.90 and 1.03 mg g−1 min−1, respectively.

Fig. 1
figure 1

Effect of contact time on the adsorption of Pb(II) and Cr(III) on fresh algal macrocolonies of N. sphaeroides (mean ± SE, n = 3, initial metal concentration 10 mg L−1, fresh algal 3.8 g L−1, pH 5.0, shaking rate 120 rpm, 25 °C)

Full size image

Pseudo-first-order and pseudo-second-order models were used to interpret the experimental data and determine the kinetic mechanism involved in the adsorption process (Ho and McKay 2000; Yurdakoc et al. 2005). The pseudo-first-order equation is described as follows:

$$ \frac{1}{q_t}=\frac{K_1}{q_{eq1}\times t}+\frac{1}{q_{eq1}} $$
(3)

where K 1 is the pseudo-first-order rate constant (min−1) for adsorption, and q eq1 and q t (mg g−1) are the adsorption capacities at equilibrium and time t (min), respectively. A plot of 1/q t against 1/t (min−1) can be fitted to a straight line to obtain K 1.

The pseudo-second-order equation is represented by

$$ \frac{t}{q_t}=\frac{1}{K_2\times {q_{eq2}}^2}+\frac{1}{q_{eq2}}\times t $$
(4)

where K 2 is the pseudo-second-order rate constant (g mg−1 min−1), and q eq2 and q t (mg g−1) are the adsorption capacities at equilibrium and time t (min), respectively. A plot of t/q t against t (min) can then be fitted to a straight line to obtain K 2.

From the results, adsorption of Pb(II) and Cr(III) perfectly fit the pseudo-second-order model according to the R 2 values (Table 3). This suggests that chemical adsorption is the main mechanism in the adsorption process. The pseudo-second-order rate constant K 2 and initial adsorption rate for Pb(II) were much higher than for Cr(III), indicating adsorption of Pb(II) is faster than of Cr(III).

Table 3 Kinetic parameters for Pb(II) and Cr(III) adsorption on fresh algal macrocolonies of N. sphaeroides
Full size table

Adsorption isotherms of Pb(II) and Cr(III)

The adsorption capacities of fresh macrocolonies of N. sphaeroides for Pb(II) and Cr(III) changed in different ways with increasing initial metal concentration. The adsorption capacity for Pb(II) increased in the whole range of the initial metal concentration tested (0–100 mg L−1), while the adsorption capacity for Cr(III) rapidly increased with increasing initial concentration up to ∼15 mg L−1 and then slightly decreased (Fig. 2a). The removal rate of Pb(II) increased from 76.6 to 91.3 % when the initial metal concentration increased from 0.5 to 2.0 mg L−1, but it decreased when the initial metal concentration was greater than 2.0 mg L−1 (Fig. 2b). The removal rate of Cr(III) slightly changed from 96.3 to 95.7 % in the initial concentration range 1–5 mg L−1 and then rapidly decreased with the further increasing initial metal concentration.

Fig. 2
figure 2

Change of the a adsorption capacity for Pb(II) and Cr(III) and b removal rate of Pb(II) and Cr(III) with fresh algal macrocolonies of N. sphaeroides for different initial metal concentration; mean ± SE (n = 3)

Full size image

The Langmuir and Freundlich isotherm models are two different models that are normally used to describe adsorption patterns. The Langmuir isotherm model is valid for monolayer adsorption on a surface with a finite number of identical sorption sites. It can be described by the following equation:

$$ \frac{C_{eq}}{q_{eq}}=\frac{1}{K_{\mathrm{L}}{q}_{\max }}+\frac{C_{eq}}{q_{\max }} $$
(5)

where C eq and q eq are the equilibrium concentrations of the adsorbate and the adsorption capacity (mg g−1) of the adsorbent, respectively; q max (mg g−1) is the maximum adsorption capacity; and K L is the Langmuir constant (L mg−1), which is related to the affinity between the sorbent and sorbate.

The Freundlich isotherm model expresses multilayer adsorption on energetically heterogeneous surfaces. The Freundlich isotherm model is expressed by the following empirical equation:

$$ \ln {q}_{eq}= \ln {K}_F+\frac{1}{n} \ln {C}_{eq} $$
(6)

where C eq and q eq are the equilibrium concentrations of the adsorbate and the adsorption capacity (mg g−1) of the adsorbent, respectively, and K F (mg g−1) and n are the Freundlich constants, which indicate the relative sorption capacity and sorption intensity, respectively.

Based on the correlation coefficients, the Langmuir isotherm model is more suitable for explaining the adsorption of Pb(II) and Cr(III) on N. sphaeroides than the Freundlich isotherm model (Table 4). This result suggests that monolayer adsorption of Pb(II) and Cr(III) occurs on the surface of N. sphaeroides. The maximum adsorption capacity (q max) for Pb(II) was much higher than that for Cr(III). However, the adsorption affinity for Cr(III) was higher than for Pb(II) because of the larger K L value.

Table 4 Parameters for Pb(II) and Cr(III) ion adsorption on fresh algal macrocolonies of N. sphaeroides according to different equilibrium models
Full size table

Effect of pH on Pb(II) and Cr(III) biosorption

With fresh algal macrocolonies of N. sphaeroides, the adsorption capacities for Pb(II) and Cr(III) and the removal rates of Pb(II) and Cr(III) increased with increasing pH in the ranges 3.5–6.5 (Pb(II)) and 3.0–6.0 (Cr(III)) (Fig. 3). At pH 6.5, the adsorption capacity for Pb(II) and the removal rate of Pb(II) were 27.08 ± 0.04 mg g−1 and 99.28 ± 0.04 %, respectively. At pH 6.0, the adsorption capacity for Cr(III) and the removal rate of Cr(III) were 21.35 ± 0.66 mg g−1 and 87.62 ± 0.41 %, respectively. No significant difference was found between pH 5.5 and 6.5 for Pb(II) (P > 0.05) and between 5.0 and 6.0 for Cr(III) (P > 0.05).

Fig. 3
figure 3

Effect of pH on the adsorption capacities for a Pb(II) and c Cr(III) and the removal rates of b Pb(II) and d Cr(III) with fresh algal macrocolonies of N. sphaeroides. Mean ± SE, n = 3. The different letters above each column indicate the significance at the 0.05 level (columns with the same letter are not significantly different)

Full size image

Effect of temperature on the biosorption of Pb(II) and Cr(III)

With fresh algal macrocolonies of N. sphaeroides, the adsorption capacities for Pb(II) and Cr(III) and the removal rates of Pb(II) and Cr(III) increased with increasing temperature in the range 20–35 °C (Fig. 4). At 20 °C, the adsorption capacity for Pb(II) and the removal rate of Pb(II) were significantly lower than at the other experimental temperatures (P < 0.05). However, there was no significant difference between 25, 30, and 35 °C (P > 0.05). At 30 and 35 °C, the adsorption capacities for Cr(III) and the removal rates of Cr(III) were significantly higher than at 20 and 25 °C (P < 0.05).

Fig. 4
figure 4

Effect of temperature on the a adsorption capacity for Pb(II) and Cr(III) and b removal rate of Pb(II) and Cr(III) with fresh algal macrocolonies of N. sphaeroides; mean ± SE, n = 3

Full size image

To evaluate the thermodynamic parameters of metal cation adsorption on the fresh algal macrocolonies of N. sphaeroides, the equilibrium constant for the adsorption process K c was calculated at each experimental temperature based on the following equation (Dinu and Dragan 2008):

$$ {K}_c=\frac{C_{ad}}{C_{eq}} $$
(7)

where C ad is the concentration of solute adsorbed on the algal cells at equilibrium (mg L−1), and C e is the equilibrium concentration of the metal ion in the solution (mg L−1).

The free energy of adsorption ΔG ° ads can be calculated by the following equation:

$$ \varDelta {G}_{ads}^{{}^{\circ}}=-R\times T\times \ln {K}_c $$
(8)

The standard enthalpy ΔS ∘ ads and entropy ΔH ∘ ads of adsorption are evaluated by plotting ln K c against 1/T in linear form based on the following equation:

$$ \ln {K}_c=\frac{\varDelta {S}_{ads}^{\circ }}{R}-\frac{\varDelta {H}_{ads}^{\circ }}{RT} $$
(9)

In Eqs. (8) and (9), R (8.314 J mol−1 K−1) is the gas constant, and T (K) is the experimental temperature.

At experimental temperatures of 293, 298, 303, and 308 K, the equilibrium constant K c for Cr(III) was higher than that for Pb(II), which indicates higher affinity of fresh algal macrocolonies for Cr(III) (Table 5). The negative values of ΔG ∘ ads under all experimental conditions except 293 K for the adsorption of Pb(II) indicate the spontaneous nature of the adsorption process. The positive values of ΔS ∘ ads and ΔH ∘ ads suggested a spontaneous and endothermic adsorption process.

Table 5 Thermodynamic parameters for the adsorption of Pb(II) and Cr(III) on fresh algal macrocolonies of N. sphaeroides
Full size table

FTIR spectroscopic analysis

The functional groups involved in the adsorption of the metal ions Pb(II), Cr(III), Cu(II), Mn(II), Cd(II), and Ni(II) on fresh algal macrocolonies of N. sphaeroides were determined by FTIR spectroscopic analysis (Fig. 5). According to the shifts of the major peaks at 3429.65 (corresponding to C–OH and C–NH stretching) and 1639.95 cm−1 (corresponding to stretching of –C = O carboxyl groups), the main functional groups involved in the adsorption process of Pb(II), Cr(III), Cu(II), Mn(II), and Ni(II) are hydroxyl, amino, and carboxyl groups, while only hydroxyl and amino groups are involved in the adsorption of Cd (II). Chemical complexation might occur between these functional groups and metal ions.

Fig. 5
figure 5

FTIR spectra of fresh algal macrocolonies of N. sphaeroides before (CK) and after the heavy metal ions adsorption: a Pb(II), Cr(III), and Ni(II) and b Cu(II), Mn(II), and Cd(II). Initial ion concentration is 10 mg L−1

Full size image

SEM–EDX analysis

All of the six metal ions were found on the algal macrocolonies of N. sphaeroides after adsorption in multiple metal ion solution (Table 6). The original metal elements, such as Mg and K, were not detected after adsorption, and the amount of the original metal element Ca greatly decreased. The results indicate that ion exchange between toxic metal ions and Mg, K, and Ca occurs during the adsorption process. The amounts of Pb(II) and Cr(III) adsorbed on N. sphaeroides were higher than on the other toxic metal ions tested. The original element S was also not detected after toxic metal adsorption.

Table 6 Element contents of fresh algal macrocolonies of N. sphaeroides before and after toxic metal ion (Pb(II), Cr(III), Cu(II), Mn(II), Cd(II), and Ni(II)) adsorption by SEM–EDX analysis
Full size table

Discussion

Algal biomass of N. sphaeroides had the great ability to remove ions from aqueous solution, similar to other algal species reported (De Philippis et al. 2003; Chojnacka et al. 2005; Flouty and Estephane 2012; Markou et al. 2015). It showed different affinity for different metal ions in single solutions in the following order: Pb(II) > Cu(II) ≥ Cr(III) > Cd(II) > Mn(II) ≥ Ni(II) for fresh algae while Cr(III) ≥ Pb(II) > Cu(II) > Cd(II) > Mn(II) ≥ Ni(II) for algal powder at metal concentration of 10 mg L−1. The different affinity may be caused by the physical properties of metal ions, including ionic radius, electron configuration, electronegativity, and charge density (Liu et al. 2011). Pb(II) had the largest ionic radius, and thus, algal biomass had a stronger physical affinity for it. Similar results was found in other reports in which Pb(II) was preferentially biosorbed with respect to other metals (Flouty and Estephane 2012). A relatively higher affinity for Cr(III) might be attributed to the greatest charge density. Similar results was found by Liu et al. (2011) who found Cr(III) showed stronger affinity than Cu(II) and Zn(II) did. Algal biomass of N. sphaeroides showed a weaker affinity for Cd(II), Mn(II), and Ni(II), resulting in less than 15 mg g−1 in the adsorption capacity in this research which was significantly lower than that of Cd(II) adsorption by Pelvetia canaliculata (Hackbarth et al. 2014), Mn(II) adsorption by brown seaweed (Lee and Park 2012), and Ni(II) adsorption by Arthrospira platensis (Markou et al. 2015). The great difference was supposed to be caused by different biosorbents and experimental conditions such as initial metal concentration, temperature, and pH (Flouty and Estephane 2012).

Fresh algal macrocolonies of N. sphaeroides had high adsorption capability for Pb(II). The q max value for Pb(II) calculated with the Langmuir isotherm model was 116.28 mg g−1, which is much higher than values for other living freshwater algae and comparable with some dried algal materials (Schmitt et al. 2001; Gupta and Rastogi 2008; Kumar and Goyal 2009). Kumar and Goyal (2009) found that algal consortium (CP1) consisting of a mixed living culture of Chlorella, Chlamydomonas, and Lyngbya spp. had a q max for Pb(II) of 33.31 mg g−1. Schmitt et al. (2001) found that the q max values of fresh algal of Scenedesmus subspicatus and Cyclotella cryptica for Pb(II) were 38.7 and 36.7 mg g−1, respectively. The q max values of dead algal biomass Oedogonium sp. and Nostoc sp. for Pb(II) were calculated to be 169.49 and 106.38 mg g−1, respectively (Gupta and Rastogi 2008). Marine brown algae have been found to have excellent metal binding capacity, for example, Fucus serratus has an adsorption capacity for Pb(II) of 368.8 mg g−1 (Ahmady-Asbchin et al. 2009). However, production of the biomass of marine brown algae requires vast culturing with clean seawater, which will greatly increase the production cost.

Algal powder of N. sphaeroides had slightly but significantly higher adsorption capacity for Cr(III) than fresh N. sphaeroides did (22.70 ± 0.55 vs 19.44 ± 0.21 mg g−1), and the adsorption capacities for Ni(II), Cd(II), and Pb(II) were similar, while for Cu(II) and Mn(II), they were significantly lower (25.0 % for Mn(II) and 25.8 % for Cu(II), respectively). Obviously, the adsorption capacity of algal powder varied differently for different metal ions caused by different and complex mechanisms involved in the biosorption process, including active uptake and accumulation, passive processes such as complexation with functional groups, entrapment by the cell surface structure, ion exchange, microprecipitation, and electrostatic attraction (Chojnacka et al. 2005). The method used to prepare the dead biomass can affect significantly the efficiency of the biosorbent (Flouty and Estephane 2012). As discussed in the “Introduction” section, dead biomass can increase the adsorption capability after drying and crushing (Mehta and Gaur 2005; Gokhale et al. 2008). However, drying can change the structure of the algal cells, such as shrinking and porosity reduction, which may decrease the entrapment process and decrease the diffusion of metal ions into the material, and then decrease the metal adsorption capability. At the same time, active functional groups on living cell surfaces can be denatured in drying biomass (Avery et al. 1993). Living biomass had a stronger chemisorption mechanism which was believed to be the most important in the biosorption process, indicating a higher adsorption capability (Markou et al. 2015). For example, living biomass exhibited higher ion exchange capacity than did the drying biomass, which was perhaps due to ion homeostasis (Markou et al. 2015). Metabolic activities can also alter the microenvironment around the cells which, in turn, may affect the biosorption processes such as complexation and precipitation (Gadd 2009). Finally, living biomass also showed bioaccumulation during biosorption, which can also increase the adsorption capacity (O’Shea et al. 2010; Dixit and Singh 2014; Markou et al. 2015). Therefore, living biomass can have greater or similar or smaller metal adsorption capacity than the drying biomass. It depends on biomass type, biosorbent producing process, and experimental conditions (especially metal ions due to different mechanisms involved in the adsorption process).

Studies on the adsorption of metal ions in multicomponent systems are of utmost importance because there is always more than one metal ion or other coexisting cations in many industrial effluents (Bere and Tundisi 2011). Most research has found that the adsorption capacity significantly decreases in multiple metal ion solutions (Bere and Tundisi 2011; Monier and Abdel-Latif 2012). This was confirmed by the results in this paper. The decrease of adsorption capacity is probably because of interactions between the different metal ions as well as between metal ions and limited cell binding sites (Bere and Tundisi 2011). Both cations and anions can greatly affect the metal adsorption process (El-Sheekh et al. 2005; Han et al. 2014). As a consequence, the adsorption capacities for all of the six metals were lower in tap water than in deionized water. Previous researchers have found that selective metal adsorption occurs in multiple ion systems, which is based on the affinity of the adsorbent for the ions (Hackbarth et al. 2015). Pelvetia canaliculata removed metal ions from multiple ion solutions in the order Pb > Cu > Cd > Zn (Hackbarth et al. 2015). In this study, the adsorption capacity of N. sphaeroides for Cr(III) in multiple metal ion systems decreased the least among all of the metals (both in multiple metal ion solution and in tap water). This may be caused by the higher affinity for Cr(III) in multiple solutions than in the other metals due to a higher charge density and smaller ionic radius, which makes complexation more stable (Basolo and Johnson 1986; Liu et al. 2011). The larger K L value for Cr(III) adsorption calculated by the Langmuir isotherm model further supported the higher affinity for Cr(III) compared with Pb(II). Hence, the fresh macrocolonies of N. sphaeroides had a strong affinity for Cr(III), which suggests that it is suitable as a Cr(III) biosorbent in solutions with multiple ions.

The adsorption kinetics for the adsorption of Pb(II) and Cr(III) on N. sphaeroides were well-described by the pseudo-second-order model, similar to other adsorption research (Gupta and Rastogi 2008). For example, the pseudo-second-order model described the adsorption process of Cu(II), Zn(II), and Ni(II) on Scenedesmus quadricauda (Bayramoglu and Arica 2009). The Langmuir isotherm model was more appropriate to explain the adsorption process of Pb(II) and Cr(III) than the Freundlich isotherm model, which suggests monolayer adsorption. A similar conclusion was drawn for the adsorption of Pb(II) on Oedogonium sp. and Nostoc sp., and Cr(III) on Chlorella miniata (Gupta and Rastogi 2008; Han et al. 2014). The removal rate of metal ions substantially decreases once the binding sites are completely occupied (Bere and Tundisi 2011), consequently the removal rates of Pb(II) and Cr(III) decreased for high initial concentrations of Pb(II) and Cr(III). pH plays an important role in the adsorption process and particularly in the adsorption capacity because of modification of the degree of ionization of functional groups on the adsorbent (Al-Asheh and Duvnjak 1995). The adsorption capacity usually increases at high pH in the acidic range (Han et al. 2014), which was also found in our experiments. Low pH will also increase the competition between H+ and metal ions (Chang et al. 1997). High temperature favored the adsorption of Pb(II) and Cr(III), which indicates an endothermic process. It may be caused by the fact that temperature will affect the metabolism of the living algal cells and then affect any metabolism-dependent process in the biosorption (Gadd 2009).

In previous reports of Pb(II) adsorption on Oedogonium and Nostoc spp., Pb(II) and Cd(II) adsorption on Nostoc muscorum, and Cr(III) adsorption on Bacillus cereus RMLAU1 (Gupta and Rastogi 2008; Dixit and Singh 2014; Tripathi and Kumar 2014), hydroxyl, amino, and carboxyl groups were found to be the main functional groups for complexation with metal ions. Similar results were found in this study based on FTIR analysis. Those functional groups have been proven to be capable of binding to metal ions because they can provide anionic sites (Gadd 2009). However, the process was highly dependent on pH because of the deprotonation of those functional groups (Chojnacka et al. 2005). Abundant exopolysaccharides of N. sphaeroides may provide those functional groups (Gadd 2009). Thus, it may be part of the reason for the high adsorption capability for metals (Mo et al. 2004; Li et al. 2009). Ion exchange usually plays a large role in the adsorption process of metals (Rajfur et al. 2014). Here, the results of SEM–EDX analysis indicate ion exchange between toxic metal ions and cations such as Ca2+, Mg2+, or K+. Those cations are naturally bond to weak acidic carboxyl, hydroxyl, or amino groups in cellular walls, cellular membranes, or in polysaccharides adhered to the cell wall (Chojnacka et al. 2005). Similar ion-exchange mechanism was confirmed in the research of Markou et al. (2015).

Fresh macrocolonies of N. sphaeroides had high adsorption capability for toxic metals which was comparable to algal powder. Producing algal powder increases the production (drying and crushing) and immobilization costs (in order to separate biomass from effluent). Therefore, it is preferable to use fresh algal biomass as a biosorbent for toxic metal ion removal, especially for Pb(II) and Cr(III), according to the results. As the algal species naturally grows in paddy fields where toxic metal pollution is a severe problem at low bioavailability (Huang et al. 2007; Zhou et al. 2014), it can supposedly be used as biosorbents in removing toxic metals from water by harvesting with nets after its growth into macrocolonies in these fields.