Introduction

Due to the growing demand of the world’s increasing population, industrialization has been intensified over the world. Every industrial processing plant results in some sorts of wastes that find their way into the water systems (Grassi et al. 2012). Heavy metals are released into the aquatic environment from several domestic (automobile exhaust, smelting processes, burning of fossil fuels, incineration of wastes, landfill leaches, use of sewage sludge, municipal wastewater, urban runoff) and industrial processes like electroplating, refining ore, mining, electronic and metal-finishing industries, fertilizer industry, tanneries, painting, paper industries and pesticides (Gautam et al. 2015). Heavy metals of industrial effluents have become global environmental and public health concern due to their toxicities, non-biodegradability, bioaccumulation in human body and food chain, carcinogenicities and mutagenesis in various living organisms (Sarkar et al. 2014; Chowdhury et al. 2015).

Numerous methods such as chemical precipitation, ion exchange, coagulation–flocculation, flotation, membrane filtration, electrochemical treatment, magnetic separation and purification, biosorption and nanotechnology are being used for the treatment and removal of heavy metals from water and wastewater (Gautam et al. 2015). Among them, biosorption process has been regarded as a promising cost-effective, sustainable and eco-friendly technology (He and Chen 2014). This process offers a number of advantages in comparison with the conventional methods (Macek and Mackova 2011). A wide range of commercial and potentially low-cost adsorbents including living or dead microorganisms, seaweeds, plant materials, industrial and agricultural wastes, natural residues, and inorganic precursors including red mud, clays, blast furnace slags, zeolites, chitosan and peat has been reported in the literature (Bilal et al. 2013; Grassi et al. 2012; Macek and Mackova 2011).

Seaweeds are widely distributed in marine, freshwater as well as terrestrial ecosystems, which can serve as good biosorbents due to their abundance, cost-effectiveness, reusability and high metal sorption capacities (Bilal et al. 2013; He and Chen 2014). Despite the fact that the red algae constitute carrageenan that provides different binding sites (e.g., hydroxyl, carboxyl, amino and sulfhydryl) responsible for the adsorption for heavy metals (Grassi et al. 2012), they are the least focused group (Rathod et al. 2014). Therefore, further research is warranted on the selectivity of algal species for their use as a low-cost biosorbent (Grassi et al. 2012).

The red seaweed, Eucheuma denticulatum, is one of the most important commercial sources of iota-carrageenan (Freile-Pelegrín et al. 2006), which also has different nutritional values (Renaud and Luong-Van 2006). This species has an interesting phytoremediation capacity in treating contaminated waters (Perales and Leysa 2012). However, there is no available literature report on the biosorption of heavy metals using the dried biomass of E. denticulatum. Hence, the present study was investigated to evaluate the seaweed in the removal of Pb2+, Cu2+, Fe2+ and Zn2+ from aqueous solutions in batch system at laboratory scale under different parameters like solution pH, contact time, temperature and initial metal ion concentrations. The Fourier transform infrared (FTIR) spectroscopy study was made to identify the main functional groups involved in the biosorption process of those metal ions.

In general, mechanistic or empirical equations are used to express heavy metal adsorption capacities of different types of biosorbents using batch or column method (Park et al. 2010). Nearly two dozens of empirical models involving 2, 3, 4 or even 5 parameters have been used to fit batch equilibrium isotherm curves to biosorbents (Basha et al. 2010; Park et al. 2010). Besides, kinetic models have been described by several authors elsewhere (Plazinski and Plazinska 2012; Gupta and Bhattacharyya 2011). The equilibrium and kinetic models are often validated on the basis of coefficient of regression (R 2 ≥ 0.99) of the experimental data. However, the use of R 2 is inappropriate for comparing the best-fitting models since this parameter fails to represent the error for fitting with linear equation (Ho 2004). Hence, the use of more than one error has been suggested to describe the validity of adsorption models (Ncibi et al. 2008). In recent years, more than single error function of nonlinear regression has been used to determine the best-fitting models (Chan et al. 2012; Subramanyam and Das 2014; Sreńscek-Nazzal et al. 2015; Kumar et al. 2016). In the present study, nine error functions were used in order to evaluate the best-fitting models. The present study was carried out in the Biotechnology Research Laboratory of the AIMST University, Malaysia, during the period of May to October 2014.

Materials and methods

Collection, identification and preparation of the biosorbent

The red seaweed E. alvarezii was collected from the Semporna coast of Sabah, Malaysia, in April 2013. The thallus of the seaweed was washed for several times with running water and subsequently with deionized water to remove epiphytes and salts. The washed biomass was then dried in an oven at 60 °C for 48 h until a constant weight was attained. The dried biomass was then crushed with an analytical mill, sieved (250 μm size) and stored in polypropylene bottles until use.

The living biomass of the species was preliminarily identified following systemic morphological features (Carpenter and Niem 1998). It was then subjected to 28S DNA-based molecular identification (Sherwood et al. 2010). The species was identified as E. denticulatum, and the gene sequence of the nucleotide was submitted in the NCBI GenBank (accession number KM229319).

Chemicals and reagents

All the chemicals and reagents used in this study were of analytical reagent grade. The working solutions of different concentrations (10–200 mg L−1) of the heavy metals (Pb, Cu, Fe and Zn) were prepared by diluting the stand solutions (1000 ± 2 mg L−1) of the respective metals (Merck, Germany) in double-distilled deionized water. Different initial pH of the solutions was obtained by adding 0.1 N HCl (Sigma-Aldrich, USA) or 0.1 N NaOH (Merck, Germany).

Batch biosorption experiments

All the experiments were conducted in a batch system using 150-mL Erlenmeyer flasks in a thermostatic shaker (25 °C, 180 rpm), unless otherwise stated. Each flask was filled with 50 mL of solution and biosorbent as appropriate. The influence of the operational parameters such as solution pH (2–7), contact time (0–120 min), initial metal ion concentration (25–200 mg L−1) and temperature (25–50 °C) on the biosorption characteristics of the metals was assessed using a constant biomass dosage (4 mg L−1). Competitive adsorption of the four metal ions under mixed condition was also evaluated.

The amount of the metal ions remaining in the solutions were measured spectrophotometrically by using Atomic Absorption Spectrometer (AAnalyst700, PerkinElmer, USA) after the separation of biosorbent by filtration through Whatman Filter No. 1.

The sorption kinetics studies were conducted with 50 mL of the metal solutions at an initial concentration of 10 mg L−1. The sample solutions were withdrawn at regular intervals, and the residual concentration of the heavy metals in the aqueous phase was analyzed after filtration as stated above.

Biosorption capacity

The amount of metal adsorbed per gram of the biosorbent at equilibrium, q e (mg g−1), was calculated from the difference of the metal concentration in the aqueous phase before and after biosorption as follows:

$$q_{e} = \frac{{\left( {C_{0} - C_{\text{e}} } \right) \times V}}{m}$$
(1)

where C 0 and C e are the initial and equilibrium concentration of metal ions in the solution (mg L−1), respectively, V is the volume of metal solution (L), and m is the mass of the dry biosorbent (g).

The percentage of metal removal (R, %) from the solution was calculated as follows:

$$R\,\left( \% \right) = \frac{{\left( {C_{0} - C_{\text{e}} } \right) \times 100}}{{C_{0} }}$$
(2)

Isotherm models

In the present experiment, four two-parameter isotherm models (Langmuir, Freundlich, Temkin and Dubinin–Radushkevich) in their linear forms were applied to describe the sorption behavior of the adsorbent. Parameters of the isotherm models are given in Table 1.

Table 1 Isotherm, kinetic and thermodynamic model parameters used in the study
Full size table

Kinetic models

The metal ions adsorption on the biosorbent was analyzed by using linearized form of four kinetic models: pseudo-first-order, pseudo-second-order, Elovich and intraparticle diffusion models. Parameters of the kinetic models are shown in Table 1.

Thermodynamic study

The chemical thermodynamics of reactions is described by using three ‘thermodynamic potentials,’ i.e., internal energy (entropy), enthalpy and the Gibbs free energy. These thermodynamic parameters were calculated to determine the feasibility and nature of the biosorption of the experimental metal ions onto the adsorbent. Parameters of the thermodynamic study are given in Table 1.

Error function analysis

In order to evaluate the suitability of the equation to the experimental data error function is the best optimization procedure. Apart from the regression coefficient (R 2), nine error functions (such as sum of square error, SSE; average relative error, ARE; hybrid functional error, HYBRID; sum of absolute error, EABS; Marquardt’s percent standard deviation, MPSD; normalized standard deviation, Δq (%); coefficient of determination, r 2; nonlinear Chi-square test, χ 2; and residual root-mean-square error, RMSE) were calculated to test the fitness of the models (Table 2).

Table 2 Equations of the error functions used in the study
Full size table

Statistical analysis

Each typical experiment was done in triplicate, and the data were expressed as the mean of the triplicate results. Necessary statistical analyses were performed using Microsoft Office Excel 2007 (Microsoft Corp., USA).

Fourier transform infrared spectroscopy (FTIR)

Infrared spectra of the raw and metal-loaded biomass were obtained using a Fourier transform infrared spectrometer (Spectrum GX, PerkinElmer, USA). The FTIR spectra were recorded using a PerkinElmer spectrometer over the wave number range 400–4000 cm−1 with 10 scans at a resolution of 4 cm−1 under ambient conditions.

Results and discussion

Effect of solution pH

In the adsorption process of metal ions from aqueous solutions, pH of the solution plays an important role. Biosorption increased gradually with the increase in pH (Fig. 1a). The maximum biosorption (90.47 % for Pb2+, 66.95 % for Cu2+, 60.99 % for Fe2+ and 45.51 % for Zn2+) was observed at pH 5. At lower pH (2–4), biosorption of metal ions was inhibited greatly. At low pH, the functional groups (as described in the FTIR section) are positively charged because of high H+ concentration, which increased competition between protons and metal cations for binding active sites of biomass, resulting in decreasing the metal cations adsorption on the biomass surfaces (Yalçın 2014).

Fig. 1
figure 1

Effect of different parameters on the metal ions biosorption onto E. denticulatum: a solution pH, b contact time, c initial metal ions concentration and d temperature

Full size image

As the pH increases, the acidic dissociation of several functional groups on the adsorbent surface could take place which is dependent of the acid dissociation constant (pKa) values of each group. At pH < pKa, the neutral functional groups restrict the metal cations access to the sorbent active sites due to repulsive forces. Conversely, at pH > pKa, negatively charged functional groups reduce the electrostatic repulsions allowing the bonding of metal cations and increasing the metal removal (de Souza et al. 2013). As the adsorbent surfaces become more negatively charged with the increase of pH, it facilitates the uptake of the metal cations through surface complexation or electrostatic interaction (Yalçın 2014). Therefore, all the rest biosorption experiments were carried out at pH 5.

Effect of contact time

Biosorption of metal ions on the adsorbent increased with an increase in contact time (Fig. 1b). More than 75 % of the total adsorption was attained within 30 min. The rapid kinetic mechanism can be attributed to the formation of exterior surface complexes neglecting intraparticle diffusion, which is very advantageous in biotechnological processes for wastewater treatment (Yalçın 2014). The biosorption process becomes slow during the later stage incubation indicating the maximum adsorption level as a saturation point of biosorption (~120 min) as the free available surfaces for adsorption become occupied (Al-Homaidan et al. 2014; Ibrahim 2011). Hence, a contact time of 120 min was selected for further experiments.

Effect of initial concentration of metal ions

The biosorption capacity of the biomass increased with increasing initial concentration of the metal ions (Fig. 1c), which can be attributed to an increase in electrostatic interactions involving sites of progressively lower affinity for the metal ions up to the point of saturation (Arshadi et al. 2014; Meitei and Prasad 2014). Conversely, the percentage removal of the metal ions decreased markedly in the range of 73.04–63.71, 66.48–55.12, 62.88–49.02 and 54.56–40.87 % for Pb2+, Cu2+, Fe2+ and Zn2+, respectively, with an increase in the initial concentration of the metal ions from 0 to 200 mg L−1. This phenomenon might be due to the rapid saturation of all metal-binding active sites of the biosorbent at a certain concentration of the metal ions (Al-Homaidan et al. 2014; Meitei and Prasad, 2014).

Effect of temperature

As shown in Fig. 1d, biosorption of metal ions increased with the increase of solution temperature. The uptake of Pb2+, Cu2+, Fe2+ and Zn2+ by the sorbent increased in the range of 90.47–94.8, 66.95–69.92, 60.99–66.17 and 45.51–51.85 %, respectively, with an increase in temperature from 298 to 323 °K. It can be attributed that at the higher temperatures, the activation of the biosorbent surfaces is enlarged facilitating more active sites for biosorption of the metal ions. Moreover, an easy mobility and enhanced accessibility of metal ions from the bulk solution to the biomass active sites could also be the possible reason for the maximum biosorption of metal ions at higher temperatures (Arshadi et al. 2014).

Biosorption studies

The equilibrium adsorption isotherms are essential data source to design, understand and optimize the biosorption process. In the present study, the experimental equilibrium data were explained using Langmuir, Freundlich, Temkin and Dubinin–Radushkevich (D–R) isotherm models.

Langmuir isotherm model

As shown in Fig. 2a, approximately linear relationship (R 2 ≥ 0.99) was observed in the Langmuir adsorption isotherm. The maximum Langmuir monolayer adsorption capacity, q m (mg g−1), as shown in Table 3, was found to increase in the following order: Pb2+ (81.97) > Cu2+ (66.23) > Fe2+ (51.02) > Zn2+ (43.48) suggesting that the Pb2+ has a preferential uptake compared to the other metals, which can be attributed to its low tendency in forming strong complex (Lasheen et al. 2012). Another reason might be attributed to carboxylate polysaccharides representing in the biomass that show preferential binding of cations with large ionic radii (Yalçın 2014). However, the preferential sorption order of the metal ions in the present study could be explained by Pauling’s electronegativity (Pauling 1960): Pb2+ (2.33) > Cu2+ (1.190) > Fe2+ (1.83) > Zn2+ (1.65). This implies the fact that the higher the ion’s electronegativity, the higher the attraction for its electrons and the attraction becomes stronger to the negative charge of the biomass ligands (Lezcano et al. 2010). Furthermore, the separation factor, R L, for the metal ions falls within the range of 0 < R L < 1 (Table 2), suggesting that biosorption of the experimental cations is favorable at all the concentrations investigated (Weber and Chakravorti 1974). Hence, the E. denticulatum is a suitable biosorbent for the sorption of the experimental metal ions from aqueous solutions.

Fig. 2
figure 2

Linearized biosorption isotherm models for the metal ions onto E. denticulatum: a Langmuir, b Freundlich and c Temkin models

Full size image
Table 3 Isotherm model parameters for biosorption of the metal ions onto E. denticulatum dried biomass
Full size table

Freundlich isotherm model

The linearized Freundlich isotherm model (Fig. 2b) suggests that the biosorption of the metal ions moderately followed the Freundlich model (R 2 ≥ 0.99). The magnitude of Freundlich constant, K F (Table 3), suggests that the sorption capacity of the experimental metal ions was in the order of Pb2+ > Cu2+ > Fe2+ > Zn2+. The values of n > 1.0 (Table 3) suggest the heterogeneity of the biomass surface, and the metal ions are favorably and intensively biosorbed by the biomass under the experimental conditions. Moreover, the applicability of both the Langmuir and Freundlich models (R 2 ≥ 0.99) to the equilibrium biosorption data indicates both homogeneous and heterogeneous distribution of active sites on the biomass surface. Similar observations have been reported earlier for biosorption of metal ions by seaweeds (Plaza Cazón et al. 2013; Yalçın 2014).

Temkin isotherm model

The Temkin isotherm plots (Fig. 2c) and parameters (Table 3) indicate that the model does not fit the experimental data well for describing the metal ions biosorption (R 2 ≤ 0.97). The Temkin adsorption potential, A T (L mg−1), was found in the order of Pb2+ > Cu2+ > Fe2+ > Zn2+, indicating a lower sorbent-metal ion potential for Zn2+ probably due to its small ionic radius. The higher value of the Temkin constant, b T (kJ mol−1), for the metal ions (218.54–342.00) indicates the existence of a very strong bond between the metal ions and the biosorbent surface (Karim et al. 2012).

D–R isotherm model

The isotherm parameters (Table 3) indicate that the D–R model does not fit the experimental data well for describing the metal ions biosorption (R 2 ≈ 0.80), suggesting the involvement of metal sorption mechanism other than van der Waals force (Basha et al. 2008). The mean biosorption energy, B D (mol2 kJ−2), was found in the order of Pb2+ < Cu2+ < Fe2+ < Zn2+, which indicates that the higher the electronegativity of metals, the lower the mean biosorption energy. The mean free energy of biosorption, E (kJ mol−1), of the D–R isotherm model was found in the range of 0.119–0.188, suggesting physisorption nature of the sorption process (Dubinin 1960). Similar reports have been reported for metal biosorption by seaweed (Yalçın 2014) and modified orange peel (Lasheen et al. 2012). The positive values of E indicate the endothermic nature of the sorption process (Lasheen et al. 2012; Yalçın 2014). Furthermore, the values of E (<16 kJ mol−1) suggest that the mechanism of the ion exchange process is film diffusion controlled (Boyd and Soldano 1953).

Competitive adsorption

Competitive adsorption of the metal ions under quaternary system shows the adsorption preference of Pb2+ > Cu2+ > Fe2+ > Zn2+ with the rate of metal removal as 92.59, 92.33, 73.80 and 68.08 %, respectively. The results suggest that the potentiality of the adsorbent in the quaternary system remains the same as that in the single-metal system, which proves its unique adsorption quality.

Kinetic studies

Kinetic sorption mechanisms of the metal ions onto the biomass were evaluated using four kinetic models (pseudo-first-order, pseudo-second-order, Elovich and intraparticle diffusion). Results of the experimental data are represented in Table 4.

Table 4 Kinetic model parameters for biosorption of the metal ions onto E. denticulatum dried biomass
Full size table

Pseudo-first-order kinetic studies

The regression coefficients of the pseudo-second-order model for the sorption of the heavy metal ions (Fig. 3a) were satisfactory (R 2 = 0.98–0.99), but the difference between the experimental values, q e,exp, was higher than the modeled values, q e,calc (Table 4). It refers the fact that both the metal ions and adsorbent were involved in the adsorption process (Meitei and Prasad 2014). However, regression coefficients suggest that the pseudo-first-order model is suitable to explain the kinetic sorption of the experimental metal ions onto the biomass over the range of experimental time and metal ion concentrations.

Fig. 3
figure 3

Linearized kinetic models for the metal ions onto E. denticulatum: a pseudo-second-order, b pseudo-second-order, c Elovich, d intraparticle diffusion and e Boyd models

Full size image

Pseudo-second-order kinetic model

The regression coefficient of the linearized pseudo-second-order kinetic model (Fig. 3b; Table 4) was higher (R 2 > 0.99) than the other models, and the experimental q e,exp values matched well with the calculated data (Table 4) supporting the best fit of the model. Hence, chemisorption is the rate-limiting step which involves valence forces through the sharing or exchange of electrons between the metal ions and different functional groups in the sorbent (Plaza Cazón et al. 2013; Rathod et al. 2014).

The pseudo-second-order rate constant, k 2 (g mg−1 min−1), was found in the order of Pb2+ > Cu2+ > Fe2+ > Zn2+, which are in the order of maximum sorption capacity. The lower values of k 2 (0.0269–0.0867 g mg−1 min−1) suggest that the metal ions uptake onto the sorbent from aqueous solution was more rapid and favorable. A fast initial uptake rate, h, was observed within the first 30 min where more than 75 % of the total metal adsorption occurred (data not shown here), followed by a slower uptake rate toward the point of equilibrium (~120 min). Similar observation was reported earlier in the literature (Arshadi et al. 2014; Meitei and Prasad, 2014). This phenomenon supports that the diffusion is the rate-controlling step in the sorption process (Plaza Cazón et al. 2013). The half-adsorption time t 1/2 (min) indicates high affinity between the molecules of the biomass and the metal ions.

Elovich kinetic model

The parameters of the Elovich kinetic model are given in Table 4. The Elovich kinetic model (Fig. 3c) does not support well fit of the model (R 2 = 0.96–0.98). Moreover, the lower values of the Elovich constants α (mg g−1 min−1) and β (g mg−1) indicate a slower rate of chemisorption onto the biosorbent surface, which is in well agreement with the earlier report in the literature (Lasheen et al. 2012).

Intraparticle diffusion model

The nonlinear regression data of q t versus t 0.5 plots (Fig. 3d) suggest that adsorption occurred in three phases: surface diffusion (initial linear portion), intraparticle diffusion (second linear portion) and final equilibrium stage (third section). Apparently, intraparticle diffusion plays a significant role in the uptake of Pb2+ (R 2 ≥ 0.99). The intercept, C, in the second portion is indicative of the thickness of the boundary layer. The larger the intercept value, the greater is the boundary layer effect (Maksin et al. 2012; Rathod et al. 2014). However, the plots did not pass through the origin (C > 0) in any of the cases (Fig. 3d), suggesting that intraparticle diffusion was not the only rate-controlling step (Maksin et al. 2012; Rathod et al. 2014), and external mass transfer had also played an important role in the metal ions sorption by the biomass.

Due to the multilinearity of the mass transfer model, Boyd model was tried to evaluate the actual rate-controlling step involved in the adsorption process. The Boyd plots of Bt versus t are linear (R 2 ≥ 0.98) but do not pass through the origin (Fig. 3e), indicating the external mass transport (film diffusion)-controlled mechanism (Foo and Hameed 2013).

Thermodynamic study

The values of the thermodynamic parameters are given in Table 5, and the linearized van’t Hoff plots of ln (K eq) versus 1/T are represented in Fig. 4. The negative values of Gibbs free energy (ΔG°) indicate spontaneous and feasible thermodynamic process (Gibbs 1873). Moreover, the increase in negative ΔG° values with an increase in temperature indicates an increased feasibility of adsorption at higher temperature (Massocatto et al. 2013; Meitei and Prasad 2014; Yalçın 2014).

Table 5 Thermodynamic parameters for biosorption of the metal ions onto E. denticulatum dried biomass
Full size table
Fig. 4
figure 4

Linearized van’t Hoff plots of the thermodynamic data

Full size image

The positive values of enthalpy change, ΔH° (Table 5), suggest the endothermic nature of the metal adsorption process (Meitei and Prasad 2014; Yalçın 2014). Based on the positive values of ΔH°, it can be presumed that the biosorption process took place physically for the metal ions. This was also supported by the D–R isotherm results with the E values (<8 kJ mol−1) of the metal ions (Table 3). Further, positive values of entropy change, ΔS° (Table 4), are suggestive of increased randomness at the solid–solution interface during the biosorption process of the metal ions on the active sites of the biosorbent (Massocatto et al. 2013).

Error function analysis

In the real world, data samples from each experiment in a series of experiments differ due to measurement error affecting data precision. This requires ensuring accurate measurement results, and statistical error function is the measure to compensate data errors (Leo 1994). Therefore, the isotherm and kinetic data were analyzed using nine error functions in order to test the fitness of the models. Lower value of SSE, ARE, HYBRID, EABS, MPSD, χ 2, Δq (%), RMSE, and higher value of r 2 represent the best fit of the model.

The error functions of the isotherm data (Table 6) suggest that the Langmuir model provides the best fit to the experimental data for biosorption of the metal ions. The Freundlich model provides moderate fit to the experimental data for adsorption of the metal ions. The regression coefficient (R 2 = 0.997) suggests that Pb2+ adsorption follows the Freundlich model moderately (Table 3), while the error functions (Table 6) suggest that the Temkin model provides moderate fit for adsorption of Pb2+.

Table 6 Error functions of the isotherm models for biosorption of the metal ions onto E. denticulatum dried biomass
Full size table

The regression coefficient of the pseudo-first-order kinetic models (Table 4) shows a moderate fit of the model (R 2 ≥ 0.98) for biosorption of the metal ions. But the error functions (Table 7) do not support the good fit of the data. The regression coefficient of the pseudo-second-order model (Table 4) suggests the best fit of the experimental data. In contrast to this, the error functions (Table 7) support moderately (for Cu2+, Fe2+ and Zn2+) and strongly (for Pb2+) fit of the model, and vice versa for the intraparticle diffusion model.

Table 7 Error functions of the kinetic models for biosorption of the metal ions onto E. denticulatum dried biomass
Full size table

It is, however, observed that the isotherm and kinetic models in describing the metal ions biosorption onto the biosorbent are metal ions dependent under the experimental conditions. Nonetheless, it is strongly suggested that the regression coefficient (R 2) is not an appropriate method in characterizing the best fit of isotherm and kinetic models; rather, some forms of error analysis could be a better criterion for avoiding data errors.

Fourier transform infrared spectroscopy (FTIR)

The spectra of FTIR analysis show a number of functional groups, such as amine, aliphatic, carboxylate, carboxyl, sulfonate and ether groups in the cell wall of E. denticulatum (Table 8; Fig. 5).

Fig. 5
figure 5

FTIR spectra of the dried biomass of E. denticulatum before and after absorption of the metal ions

Full size image
Table 8 FTIR assignment in the cell wall of E. denticulatum dried biomass
Full size table

The strong broad peak observed at 3342.2 cm−1 (raw biomass) corresponds to O–H group from cellulose and N–H groups from proteins in the seaweed (Bhatnagar et al. 2012). This band shifted to a lower wave number after metal binding, suggesting their contribution in the experimental metal ions biosorption.

The peak at 2917.1 cm−1 (raw biomass) corresponds to asymmetric C–H stretching vibrations of aliphatic groups and shifts to a lower wave number after metal binding suggesting that the aliphatic groups were responsible for the heavy metal adsorption (Nessim et al. 2011).

The two strong peaks at 1637.9 and 1215.0 cm−1 (raw biomass) can be attributed to C=O stretching vibration of carboxylate groups (Plaza Cazón et al. 2013; Yalçın 2014). These peaks shifted significantly after metal binding which suggest that carboxylic groups played an important role in the biosorption of the experimental heavy metals through complexation mechanism (Rathod et al. 2014; Yalçın 2014).

Absorbance peaks at 1373.3 and 1154.4 cm−1 (raw biomass) can be assigned due to asymmetric and symmetric stretching of –SO3 − bonds of sulfonic acid, respectively (Yalçın 2014). Some degree of horizontal shift of these bands after metal binding indicated that chelating complexes were formed during adsorption of the experimental metal ions (Plaza Cazón et al. 2013).

The presence of peak at 1033.2 cm−1 (raw biomass) is attributed to C–O stretching vibration of carboxyl groups (Plaza Cazón et al. 2013) that shifted to a significant lower peak after metal binding, which might be due to the chelating complexation (bidentate complex) of the experimental metal ions to the carboxylic groups (Pennesi et al. 2012).

Absorbance peaks at 923.6 and 844.5 cm−1 (raw biomass) correspond to S–O stretching (Plaza Cazón et al. 2013) and S=O stretching bands of sulfonate groups (Yalçın 2014). These bands decreased and disappeared, respectively, after metal binding which was typical of complexation of the sulfonate groups by dative coordination with the experimental metal ions (Yalçın 2014; Yalçın et al. 2012).

Absorbance peak at 1123.9 cm−1 (the raw biomass) is attributed to the –C–O stretching of ether groups (Sheng et al. 2004) that disappeared after metal binding suggesting that the ether groups were also responsible for the biosorption of the experimental metal ions.

Conclusion

In the present study, we evaluated biosorption quality of four heavy metal ions like Pb2+, Cu2+, Fe2+ and Zn2+ onto the dried biomass of the red seaweed, E. denticulatum, from Malaysia. The FTIR study revealed the presence of amine, aliphatic, carboxylate, carboxyl, sulfonate and ether groups in the cell wall matrix of the biomass that was involved in the adsorption of the metal ions. The solution pH, contact time, temperature and initial metal ion concentration have a significant effect on the biosorption efficiency.

Heavy metal biosorption process onto the biosorbent representing both homogeneous and heterogeneous active sites was the complex one involving both physisorption and chemisorption and was feasible, spontaneous and endothermic in nature. Error function analysis, an accurate measurement of the experimental data, shows that the adsorption isotherm data for all the experimental metal ions fitted well with the Langmuir model followed by the Freundlich model (for Cu2+, Fe2+ and Zn2+) and the Temkin model (for Pb2+). Kinetic data for the metal ions (Cu2+, Fe2+ and Zn2+) could be best described by the intraparticle diffusion model followed by the pseudo-second-order model, and vice versa for Pb2+ biosorption. We strongly suggest analysis of error functions as a better criterion in order to characterize isotherm and kinetic models describing adsorptive behavior of a typical adsorbent using linear method.

The results provide the potential of E. denticulatum as a new, low-cost and promising biosorbent for the removal of Pb2+, Cu2+, Fe2+ and Zn2+ from aqueous solutions. It can pave a way toward sustainable utilization of the dried E. denticulatum biomass for the restoration of industrial wastewater contaminated with heavy metal ions. However, further study is warranted to evaluate the feasibility of the biosorbent for removal of heavy metals from the real environment.