Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column

Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column

Accepted Manuscript Title: Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column Autho...

1MB Sizes 9 Downloads 58 Views

Accepted Manuscript Title: Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column Authors: Heitor O.N. Altino, Bruno E.S. Costa, Renata N. da Cunha PII: DOI: Reference:

S2213-3437(17)30463-3 JECE 1873

To appear in: Received date: Revised date: Accepted date:

25-7-2017 8-9-2017 12-9-2017

Please cite this article as: Heitor O.N.Altino, Bruno E.S.Costa, Renata N.da Cunha, Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column, Journal of Environmental Chemical Engineering This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Biosorption optimization of Ni(II) ions on Macauba (Acrocomia aculeata) oil extraction residue using fixed-bed column

Heitor O. N. Altino a, c; Bruno E. S. Costa b; Renata N. da Cunha c*.


Department of Chemical Engineer, Federal University of São Carlos, Highway

Washington Luiz, km 235, São Carlos Campus, CEP: 13565-905, São Carlos, São Paulo, Brazil. b

Institute of Chemistry, Federal University of Uberlândia, Avenue João Naves de

Ávila, 2121, Block 1D, Santa Monica Campus, CEP: 38400-901, Uberlândia, Minas Gerais, Brazil c

Department of Chemical Engineer, University Centre of Patos de Minas, Street Major

Gote, 808, Block H, Caiçaras Campus, CEP: 38702-054, Patos de Minas, Minas Gerais, Brazil. ______________________________________________________________________ ∗Corresponding

author. Tel.: +55 034 3823 0300.

E-mail address: [email protected] (R. N. da Cunha).

Highlights Macauba endocarp can be considered a potential biosorbent for Ni(II) biosorption. Ni(II) biosorption on the biosorbent can be related to hydroxyls and phenols groups. Biosorbent mass, pH and Ni(II) concentration affected the biosorption process. The best experimental condition provided significant and rapid removal of Ni(II). The biosorption kinetics was better described by the pseudo-second-order model.


Abstract Macauba is a palm tree located in Tropical and Subtropical America. The oil extraction process from the pulp and kernel of the Macauba fruit generates two main residues, mesocarp and endocarp. The extraction process produces considerable quantity of residues that can become an environmental issue. The applications of mesocarp as a biosorbent have been reported, but studies involving the endocarp were no longer conducted. The objective of this paper was to study the application of endocarp, without pre-treatment, for Ni(II) biosorption optimization in fixed-bed column employing a factorial experiment. Spectroscopic studies reveled that Ni(II) biosorption can be related to functional hydroxyls and phenols groups on the biosorbent surface. Large particles with the surface covered of irregular walls, without defined pores, were observed by micrographs. The biosorption optimization showed that the usable capacity of the bed up to the break-point (𝑡𝑢 ), total time equivalent to the packaging capacity of the column (𝑡𝑡 ) and total removal (𝑅𝑡 ) were affected by both pH and Ni(II) concentration. Mass Transfer Zone (𝑀𝑇𝑍) was affected only by biosorption mass. Equilibrium metal uptake (𝑞𝑒 ) was affected by pH, biosorption mass and Ni(II) concentration. Thus, it was possible to determine the equations that control the response variables and observe a significant and rapid removal of Ni(II). Kinetics studies exhibited the appropriateness of Pseudo-second-order model to describe the phenomena involved on biosorption kinetics. The results obtained in this paper showed the applicability of Macauba endocarp as a low-cost and environmentally friendly biosorbent for treatment of wastewaters contaminated with Ni(II). Key words: Acrocomia aculeata; Nickel; Biosorption; Optimization; Kinects.


1. Introduction

Heavy metals are a group of contaminants that are dispersed naturally and by human activities. With the exponential increase in population the need for controlling heavy metal emissions into the environment have been revealed. Once these metallic species are converted by and released into the environment they tend to persist indefinitely, circulating and eventually accumulating throughout the food chain, thus causing a serious threat to the environment and human being [1,2]. Among the several existents heavy metals, Ni(II) should be emphasized due to its economic importance. Although this heavy metal is essential for living organisms, in high concentrations it may generate serious problems due to the toxic effects, such as: reduced lung function, chronic bronchitis and lung cancer [3]. Mineral processing, electroplating, non-ferrous metal, copper sulphate manufacture, steam-electric power plants, porcelain enameling and paint formulation are considered the main sources of Ni(II) industrial discharges [4]. The Ni(II) concentrations in wastewater of several of these process can reach up to 130 mg.L-1 [5]. Thus, The World Health Organization’s (WHO’s) established the safe limit of 0.07 mg.L-1 of Ni(II) in drinking water [6]. In order to reduce the concentrations of Ni(II) up to the permissible levels, several physico-chemical processes have been utilized, such as: chemical precipitation, electrode deposition and reverse osmosis [7]. However the inefficiency in the Ni(II) concentration range of 1-100 mg.L-1, large capital and necessity of correct disposal of toxic sludge from these processes prevent their applications in large scale [8,9]. In such way, biosorption presents to be an attractive process supported by an economic and environmental background [10–13]. Several 3

biosorbents have been reported for Ni(II) biosorption: treated alga (Undaria pinnatifida) [14], wheat straw [15], golden shower parts (Cassia fistula) [16], water fern biomass (Azolla filiculoides) [17], bacterial strain (Leucobacter sp. N-4) [18], leaves of fern (Asplenium nidus L.) [19] and tree cones (Thuja orientalis) [20]. Although there are numerous biosorbents reported for Ni(II) biosorption, it is also necessary to concern about the biodegradation of the spent adsorbent, which can be achieved by employing oil extraction residues [21]. In such context, several studies have revealed that the residues from oil extraction can be utilized for the biosorption of Ni(II): mustard oil cake [8], carbon derived from mustard oil cake [7] and activated carbon derived from rapeseed oil cake [22]. In Brazil, the Macauba oil conquered a significant space in the oil market due to its special characteristics and potential to be employed as raw material for biodiesel production, which increases every year. Typically, the process includes the pressing of the Macauba fruit parts in order to extract the crude oil. It generates a considerable quantity of residues that, without correct disposal, becomes a major environmental issue [23]. After pressing the fruit parts, the residues generated are the cake (mesocarp) and endocarp. Since these materials are wastes, the cost involved in their utilization without pre-treatment is very low. The Macauba oil cake was studied by [23] for biosorption of methylene blue and congo red, but studies involving the endocarp were no longer reported. The aim of the present paper was to research the potential application of Macauba endocarp without pre-treatment for Ni(II) biosorption in fixed-bed column employing a factorial experiment. In order to comprehend the mechanisms involved in biosorption, the biosorbent was characterized by particle size


analyze, N2 physisorption, Scanning Electron Microscope, Fourier Transform Infrared Spectra and the biosorption kinetics was studied.

2. Materials and methods

2.1. Chemical reagents and instrumentation

Synthetic wastewater solutions of Ni(II) were prepared by dissolving nickel(II) chloride hexahydrate (NiCl2.6H2O) (Vetec®, Brazil) in deionized water. The pH was corrected by solutions of hydrochloric acid (HCl) (Synth®, Brazil) 30% (v/v) and sodium hydroxide (NaOH) (Dinâmica®, Brazil) 30% (v/v). In all experiments the deionized water used was generated by a Permution® CS1800 Evolution purification system with resistivity of 1.5 MΩ cm. All the laboratory glasswares were decontaminated by a solution of nitric acid (HNO3) (Synth®, Brazil) 5.0% (v/v) for 24 hours, washed with deionized water and then dried at room temperature. The pH measurements were performed by potentiometric determinations using a MS Tecnopon® mPA210 pH meter. The Ni(II) concentrations were analyzed by Flame Atomic Absorption Spectroscopy (FAAS) using a Varian® SpectrAA-220 spectrometer. A nickel hollow cathode lamp from the same manufacturer was used for measurements of absorption signals, working with 4 mA of current. The wavelength was the 232.0 nm nickel resonance line, selected with a spectral band pass of 0.2 nm. The background was corrected with a deuterium lamp. An air/acetylene flame (13.5:2) was used for all measurements. 5

2.2. Preparation of biosorbent

Macauba is a palm tree located in Tropical and Subtropical America. The Macauba fruit is basically composed by: shell (epicarp), pulp (mesocarp), nut (endocarp) and kernel (coated by nut), but only the pulp and the kernel are utilized for oil production. The oil extraction process from the pulp and kernel generates two main residues, named in present paper by mesocarp and endocarp, respectively. The endocarp residue is specifically produced in the cracking process of the nut to access the kernel. In order to study the potential applications of endocarp, three Macauba endocarp samples without any pre-treatment at different granulometric ranges, named A (-286 μm), B (-370 +286 μm) and C (-833 +370 μm), with moisture of 3.0%, were obtained from a Macauba extract oil industry located in the region of Triângulo Mineiro and Alto Paranaíba, Minas Gerais, Brazil. The three granulometries were previously separated by the industry and provided in such way. The low moisture was inherent of the Macauba oil extraction process. The samples were storage in polyethylene containers at ambient temperature.

2.3. Characterization of biosorbent

The Macauba endocarp samples were characterized by particle size analyze employing an electromagnetic sieve shaker Bertel® with five sieves: 16, 28, 48, 100 and 150 mesh Tyler. Surface area, pore volume and average pore width were determined by applying the Brunauer–Emmett–Teller (BET) method by N2 physisorption (77 K) in a Micromeritic® Accelerated Surface Area and Porosimetry (ASAP) 2020 equipment. To 6

analyze the surface structure the samples were coated with a thin layer of gold (20 nm) and characterized using Scanning Electron Microscope (SEM) by a Tescan® Vega 3 microscope, with magnifications of 500, 2000 and 5000 times. To analyze the surface functional groups of the samples, Fourier Transform Infrared Spectra (FTIR) were determined by employing the Attenuated Total Reflectance (ATR) mode using a PerkinElmer® Frontier spectrometer with diamond crystal, in the range of 220-4000 cm1

at a resolution of 4 cm-1. Based on the characterization of the samples described above

and in pre-tests, the most appropriate sample was chosen to be studied as a biosorbent of Ni(II) in fixed-bed column. To investigate the surface charge of the chosen sample, the zero of point charge (pHzpc) was determined, according to [24], by adding 200 mg of Macauba endocarp in 12 flasks and adding 15 mL of deionized water. Then, the initial pH of each solution was adjusted between 1.0 to 12.0 by using solutions of HCl (Schaulau®, Spain) 0.1 mol.L-1 and NaOH (Qhemis®, Brazil) 0.1 mol.L-1. The flasks were kept for 24 h and the final pH of the solutions was measured by using pH-meter.

2.4. Biosorption optimization in fixed-bed column

The optimization problem was defined with the aim of improving the usable capacity of the bed up to the break-point (𝑡𝑢 ) (min), total time equivalent to the packaging capacity of the column (𝑡𝑡 ) (min), Mass Transfer Zone (𝑀𝑇𝑍) (cm), total removal (𝑅𝑡 ) (%) and equilibrium metal uptake (𝑞𝑒 ) (mg.g-1). According to [25], the parameter 𝑡𝑢 is the equivalent time at which the effluent concentration reaches 10% of the initial concentration and can calculated by Eq. (1). The parameter 𝑡𝑡 is the time


equivalent to the total capacity of the packed-bed if the entire bed were in equilibrium with the feed, mathematically defined by Eq. (2). The Mass Transfer Zone characterizes the length of the packed-bed across which the concentration gradient occurs, Eq. (3).



𝑡𝑢 = ∫0 𝑏 (1– ) 𝑑𝑡




𝑡𝑡 = ∫0 (1– ) 𝑑𝑡




𝑀𝑇𝑍 = (1– 𝑢 ) 𝐻𝑡



Where C is the effluent concentration (mg.L-1), C0 is the influent concentration (mg.L-1), tb is the break point time at 𝐶 ⁄𝐶0 = 0.1 (min), Ht is the total bed length (cm) and 𝑡 is the time (min). The integral parts of Eqs. (1) and (2) were numerically solved by the trapezoidal rule, using experimental data of 1– 𝐶/𝐶0 in function of 𝑡, in the software Scilab® 5.5.2. Additional equations utilized in this work that also needed numerical integral solutions were solved by the same rule. The total removal can be mathematically defined by Eq. (4) [9].

𝑅𝑡 =

𝑞𝑡𝑜𝑡𝑎𝑙 𝑚𝑡𝑜𝑡𝑎𝑙


𝑚𝑡𝑜𝑡𝑎𝑙 =

𝐶0 𝑄 𝑡𝑡𝑜𝑡𝑎𝑙

𝑞𝑡𝑜𝑡𝑎𝑙 =






∫ 1000 0

(𝐶𝑎𝑑 )𝑑𝑡



Where mtotal is the total quantity of Ni(II) sent to column (mg), Q is the volumetric flow rate (mL.min-1), ttotal is the total flow time, qtotal is the total adsorbed metal quantity (mg) and Cad is the adsorbed concentration (mg.L-1). Based on the Eq. 6 it is possible to determinate 𝑞𝑒 dividing qtotal by the total biosorbent mass (𝑋) (g), as described by Eq. (7) [9].

𝑞𝑒 =




To promote such optimization of the described variables a 23 factorial experiment with a middle point (duplicate) was employed, investigating the influence of pH of the Ni(II) solutions (3.0-6.5), biosorbent mass (100.0-150.0 g) and Ni(II) initial concentration (10.0-110.0 mg.L-1). The replicate was done in the middle point, since this approach has been widely used in the literature to evaluate the standard error [26,27]. The independent variables were transformed into the dimensionless variables 𝑥1 , 𝑥2 and 𝑥3 , according to the Eqs. (8), (9) and (10).

𝑥1 =

(𝑝𝐻− 4.75)

𝑥2 =

(𝑋− 125.0 𝑔)

𝑥3 =




25.0 𝑔

(𝐶0 − 60.0 𝑚𝑔.𝑚𝐿−1 )


50.0 𝑚𝑔.𝑚𝐿−1


According to [28], there are only a few researches describing the treatment of wastewaters contaminated with Ni(II) at pilot and industrial scales. These scales allow better predictions of the biosorption mechanism in a real wastewaters treatment, and facilitate the scale-up of the process. Thus, the experiments were conducted in an UpControl® multipurpose module, as displayed in Fig. 1, which provides operational conditions close to the pilot scale. Ni(II) solutions were fed in the module at flow of 50.00 mL.min-1 (contact time of 2.00-3.00 min) in the temperature of 296.60 ± 2.04 K. The fixed-bed of the column was composed by glass beads (6 cm), for better influent dispersion, Macauba endocarp without pre-treatment (sample B) (4.00-7.00 cm) and more glass beads (7 cm) as a support. The Ni(II) concentration in the column effluent was analyzed according with section 2.1.

2.5. Calculation of biosorption kinetics

The Ni(II) ion uptake through the biosorption process time (𝑞𝑡 ) (mg.g-1) in fixed-bed column was calculated using Eq. (11), considering the breakthrough curves experimentally obtained [29].

𝑞𝑡 =

𝑡 𝐶0 𝑄 ∫ (1 1000𝑋 0


− ) 𝑑𝑡



The kinetic study is justified by its importance for the scale-up of biosorption system and insights into the phenomenological mechanism controlling the biosorption process [3].


2.6 Error analysis

In order to evaluate differences between the experimental data (𝑦𝑖 ) and the data estimated by the models (𝑓(𝑥𝑖 )) utilized in this paper, an error analysis is required. The coefficient of determination (𝑅2 ), mean square error (𝑀𝑆𝐸), root mean square error (𝑅𝑀𝑆𝐸) and chi-square (𝜒 2 ), were calculated by Eqs. (12), (13), (14) and (15), respectively.

𝑅2 = 1 −

2 ∑𝑁 𝑖=1(𝑦𝑖 − 𝑓(𝑥𝑖 ))

𝑀𝑆𝐸 = ∑𝑁 𝑖=1

(𝑦𝑖 − 𝑓(𝑥𝑖 ))2



𝑅𝑀𝑆𝐸 = √∑𝑁 𝑖=1

𝜒 2 = ∑𝑁 𝑖=1


∑𝑁 ̅ 𝑖 )2 𝑖=1(𝑦𝑖 − 𝑦

(𝑦𝑖 − 𝑓(𝑥𝑖 ))2



(𝑦𝑖 − 𝑓(𝑥𝑖 ))2


𝑓(𝑥𝑖 )

3. Results and discussion

3.1. Characterization of biosorbent

The Table 1 shows the characteristic diameters of Macauba endocarp samples, which is possible to notice the increasing of the diameters from sample A to sample C, as expected by the granulometric ranges of the samples previously separated. 11

In order to describe the particle size distribution, the Gates–Gaudin–Schuhmann (GGS) and Rosin–Rammle (RR) models were applied to the experimental data, which can be described by Eqs. (16) and (17) [30,31]. These models were chose due to their capacity to represent the particle size distribution of biological materials. Since the Macauba endocarp residue is generated by the cracking process of the nut, the RR model was specifically select due to its aptitude to describe the size distribution of materials obtained by grinding, milling, and crushing operations [30,32].


𝐹(𝜙) = (





𝜙 𝑚

𝐹(𝜙) = 1 − 𝑒 −( 𝑙 )


Where 𝐹(𝜙) is the distribution function, 𝜙 is the particle size (μm), 𝜙𝑚𝑎𝑥 is the maximum particle diameter of the distribution (μm), 𝑚 is a measure of the spread of particle sizes and 𝑙 is the mean particle size (μm). The models were adjusted to the experimental data by nonlinear regression utilizing the Rosenbrock Pattern Search method in the software Statistica® 7.1. Further models employed in this paper were adjusted to experimental data by applying the same method. The results derived from application of the described models can be seen in Table 2. The high values of 𝑅2 and low values of 𝜒 2 , 𝑀𝑆𝐸 and 𝑅𝑀𝑆𝐸 exhibited by the RR model indicates its capacity to describe the particle size distribution of the samples. The N2 physisorption isotherms of the Macauba endocarp samples are displayed in Fig. 2. The isotherm of the sample C could not be correctly determined due 12

to the considerable amount of residual oil retained descendant from the extraction, which prevents the adsorption of N2 in the surface of the material. According to [33], the samples A and B showed the hysteresis type H3 with two asymptotic isotherms branches near to vertical position, indicating the presence of a non-rigid aggregate of plate-like particles. Such observation agrees with the low pore volume and average pore width exposed in Table 3, where is also possible to notice that sample B exhibited the highest surface area and pore volume, indicating its potential to be used as a biosorbent. To investigate the morphology of the Macauba endocarp samples at different granulometric ranges, SEM micrographs were used. Examining the sample A micrographs (Fig. 3 (a), (b) and (c)), it is possible to notice a standard of large particles with irregular surfaces and without defined pores, according to the observations done in the N2 physisorption analyzes. This standard is also observed in the sample B (Fig. 3 (d), (e) and (f)), but with larger particles. In sample C micrographs (Fig. 3 (g), (h) and (i)) the particles happened to be the largest ones among the samples, also following the standard. Another difference in sample C is noticed in Fig. 3 (i), where some kind of structure in the shape of pins is found. This structure is probably destroyed during the process of smashing large particles, into smaller ones. At this point, it’s remarkable to highlight that the SEM micrographs illustrate the increasing of the diameters from sample A to sample C observed in the particle size distribution analyze, emphasizing the different granulometric ranges between the samples. The FTIR spectrum of Macauba endocarp samples are displayed in Fig. 4. The broad peak at 3335 cm-1 exhibited in all samples is associated to hydroxyl groups. This can be confirmed by the peak related to primary alcohol at 1040 cm-1. It is also possible to notice the weak presence of vibrations between 3700-3620 cm-1, which are 13

related to the existence of intramolecularly bonded and free OH. For all the samples, but mainly for A and C, it is possible to see absorption about 2923 cm-1, associated to asymmetric and symmetric stretching vibrations of CH2 groups. The peaks in 1620 cm-1 are correlated to the stretching vibration of the C=C bonds. The minor peaks detected within the band 1300-1100 cm-1 can be credited to C-O bonds from phenolic groups [23,34]. This spectrum indicates that the Macauba endocarp contains functional entities, such as hydroxyls and phenol groups, which are independent of the granulometry. As mentioned in section 2.4., the sample B, (Sauter diameter of 427 μm) was selected to be employed as a biosorbent. This decision is supported by the high surface area, pore volume and intermediate granulometry, which demonstrated to be a determination factor in the pre-tests. The presence of fine particles of sample A hindered the passage of solution while the high diameter showed by the sample C favored the formation of preferential channels, decreasing the biosorption efficiency. In order to determinate the surface charge of Macauba endocarp sample B, the pHzpc was determined, as displayed in Fig. 5. The interception point of the pH variation (pH initial – pH of equilibrium) curve in the zero line of abscissas provided the pHzpc of 5.79. This value indicates that the Macauba endocarp surface carried positive charges in pH’s lower than 5.79 and negative charges in pH’s higher than 5.79, which is desired for removal of cationic metals.

3.2. Operational conditions optimization

The breakthrough curves for Ni(II) biosorption on Macauba endocarp (sample B) for the employed treatments are displayed in Fig. 6. The treatments (2 and 4) 14

utilizing a pH of 6.5 and Ni(II) initial concentration of 10 mg.L-1 exhibited less inclined curves, while endocarp mass had less visible effects in all treatments. Another important observation is the similar curve shape of the middle point (treatments 9 and 10). Based on the breakthrough curves, the variables to be optimized of Ni(II) biosorption were calculated and are displayed in Table 4. As expected by the breakthrough curves shapes, the treatment 2 demonstrated the highest value of 𝑡𝑢 and 𝑡𝑡 , also providing the highest removal of Ni(II). The shortest length of 𝑀𝑇𝑍 was achieved at treatment 3. The highest value of 𝑞𝑒 was attained at treatment 8. The results displayed in Table 4 were statistically analyzed according to the factorial experiment. The effects of the factors with a significance level lower than 0.05 are displayed in Table 5. The variables 𝑡𝑢 and 𝑡𝑡 were affected by pH and Ni(II) initial concentration with interaction between these factors and pH as the most potent factor. For biosorbent mass, it was expected that an increase in the mass would promote an increase of the superficial area of biosorbent and the amount of active sites available for binding, increasing 𝑡𝑢 and 𝑡𝑡 , but for the studied rage this effect was not significant. Although the relationship between adsorbent mass and surface area for Ni(II) binding is different from some adsorbents: rice straw [35], hydrilla [36] and waste of tea [9], not significant or even bad influences of mass increase on Ni(II) sorption were also reported for other adsorbents, such as mollusk shells [37] and polyurethane foam [38]. Thus, it’s important to highlight that, in the particular Brazilian case, the Macauba endocarp can be considered a specific local sustainable solution for Ni(II) biosorption. On the other hand, the variable 𝑀𝑇𝑍 was just affected by the amount of biosorbent mass. As also observed by [39], for the same flow, the 𝑀𝑇𝑍 increases as the


amount of biosorbent mass increases. According to [25,40], in a nonideal case, as the adsorbate in the liquid phase passes through the column, only part of the adsorbate is adsorbed, due to the diffusion limitations. At the same time, the saturation of the adsorbent takes place and the adsorption velocity decreases. The result of these two phenomena is a decrease in the sharpness of the flat front of adsorption, creating the 𝑀𝑇𝑍. Thus, the 𝑀𝑇𝑍 increases with the mass due to the increase in the diffusion limitations of the solute into the pores of biosorbent and the residence time of the solute in the column [39,41]. Similarly to the variables 𝑡𝑢 and 𝑡𝑡 , the variable 𝑅𝑡 was affected by pH and Ni(II) initial concentration, with interaction between these factors and pH as the most potent factor. The variable 𝑞𝑒 was affected by all the factors, but with interaction only between mass and concentration. From the experimental results and statistical analyzes, four empirical equations have been fitted to represent 𝑡𝑢 , 𝑡𝑡 , 𝑅𝑡 and 𝑞𝑒 as function of the dimensionless factor, as presented in Eqs. (18), (19), (20) and (21), respectively. In these equations, the presented parameters were those significant in a hypothesis tests with a t of the Student distribution (maximum probability of error of 5.0%). The values of 𝑅2 were 0.870, 0.936, 0.943 and 0.950, respectively for the Eqs. (18), (19), (20) and (21). The residual analysis of these regressions shows that the residuals are independently and identically distributed according to a normal distribution with mean zero and fixed variance.

𝑡𝑢 = 9.231 + 𝑥 ′ 𝑏 + 𝑥′𝐵𝑥 𝑥1 𝑥 = [𝑥2 ] 𝑥3

7.638 𝑏=[ 0 ] −9.834

(18) 𝐵=[

0 0 −2.326 0 0 0 ] −2.326 0 0 16

𝑡𝑡 = 59.867 + 𝑥 ′ 𝑏 + 𝑥′𝐵𝑥 𝑥1 𝑥 𝑥 = [ 2] 𝑥3

39.921 𝑏=[ ] 0 −40.820

(19) 𝐵=[

0 0 0 0 −15.590 0

−15.590 ] 0 0

𝑅𝑡 = 0.1588 + 𝑥 ′ 𝑏 + 𝑥′𝐵𝑥 𝑥1 𝑥 = [𝑥2 ] 𝑥3

0.112 𝑏=[ 0 ] −0.114

(20) 𝐵=[

0 0 −0.040 0 0 0 ] −0.040 0 0

𝑞𝑒 = 0.922 + 𝑥 ′ 𝑏 + 𝑥′𝐵𝑥 𝑥1 𝑥 = [𝑥2 ] 𝑥3

0.368 𝑏 = [−0.383] 0.437

(21) 0 𝐵 = [0 0

0 0 0 −0.160] −0.160 0

Fig. 7 shows the response surfaces obtained from Eqs. (18), (19), (20), and (21) to better visualize the effect of the factors on the response variables. In Fig. 7 (a), (b) and (c) it is possible to analyze that as pH decreases the values of the variables 𝑡𝑢 , 𝑡𝑡 and 𝑅𝑡 , also decrease. The behavior of biosorbent may be associated to the action of functional groups, such as hydroxyl and phenolic observed in the FTIR analyzes (Fig. 4), that may carry negative charges, allowing the removal of metals [42]. Thus, in accordance to [8,14,43,44], the low pH of 3 can promote the competition between the ions H+ and Ni(II) for the surface sorption sites. Thus, these authors explain that the sorption sites may be occupied by H+ ions which would prevent the approach of Ni(II) ions by electrostatic repulsion, reducing the amount of active sorption sites and consequently the variables 𝑡𝑢 , 𝑡𝑡 and 𝑅𝑡 . This explanation is also supported by the pHzpc analyze, which indicates that at pH’s lower than the pHzpc the sorption sites can be 17

bound to H+ ions. It’s also important point out that at the pH of 3 other nickel's species can take place and reduce the amount of Ni(II) available for biosorption [45]. On the other hand, at pH values higher than the zero of point charge of 5.79, such as 6.5, the decoupling of H+ ions from superficial sorption sites may happen. Thus, the predominant specie of nickel at this pH, Ni(II) [45], can stronger interacts with the negative surface of the Macauba endocarp. The result is higher values of the cited variables. It’s interesting to highlight that the increase of metals sorption with pH is a classical phenomenon, first detected for inorganic mineral phases [46,47], which can also be extended to functional groups of biomolecules, as well as in the lignocellulosic composition commonly present in adsorbents from biological sources. In Fig. 7 (a), (b) and (c) is also possible to observe that at low Ni(II) initial concentration (10 mg.L-1) the variables 𝑡𝑢 , 𝑡𝑡 and 𝑅𝑡 tend to increase its values. It may be elucidated by the fact that at low concentrations lower concentration gradients are achieved, causing slower transport due to a decreased diffusion coefficient or mass transfer coefficient, resulting in delayed breakthrough curves and higher values of 𝑡𝑢 , 𝑡𝑡 and 𝑅𝑡 [48]. Fig. 7 (d) and (e) shows the influence of all factors on the variable 𝑞𝑒 . For pH is possible to observe the same trends noticed on the variables 𝑡𝑢 , 𝑡𝑡 and 𝑅𝑡 . It may also be explained by the competition between the ions H+ and Ni(II), reducing the quantity of metal adsorbed at equilibrium. For biosorbent mass, the same effect of reducing 𝑞𝑒 with the increase of biosorbent mass reported by [8,14] was detected. The increase in biosorbent mass leads to the availability of more biosorption sites, increasing the biosorption while some of the sites remains unsaturated during the process,


decreasing the biosorption capacity [8]. For Ni(II) initial concentration high values of its factor (110 mg.L-1) resulted in high values of 𝑞𝑒 . In accordance with [44,49], higher initial concentrations provides an increased driving force to overcome mass transfer resistance of metal ions between the aqueous and solid phases. The result is higher chances of contact between Ni(II) ions and adsorbents, causing higher metal uptake. In summary, the Macauba endocarp without any further pre-treatment is a low-cost and environmentally friendly biosorbent for Ni(II) biosorption. The results indicated that several factors such as pH, biosorbent and mass and Ni(II) initial concentration affect the biosorption process, highlighting the importance of the factorial experiment to clarify the involved phenomena and determinate the equations that describe the influence of these factors on the response variables. In Table 6 it is possible to compare the results obtained in this paper with other absorbents previously reported in literature. It is possible to notice highs removals of Ni(II) using the batch system, but in almost all the cases long contact times were required. On the other hand, employing the fixed bed-system lower removals of Ni(II) are achieved, but with shorter contact times. In this sense, Macauba endocarp without pre-treatment showed to be recommend for Ni(II) biosorption, reaching high removals of Ni(II) at low contact time. For practical purposes, the fixed bed-system is recommended to full-scale the biosorption processes [9]. Thus, the parameters and equations obtained in this paper may be utilized to scale-up the described system.


3.3. Biosorption kinetics

In order to study the process mechanisms of biosorption process such as mass transfer and chemical reaction, the mean known models described describe in the literature were applied to kinetics data; Pseudo-first-order, Pseudo-second-order, Intraparticle diffusion and Elovich. This first model contemplates that the rate of occupation of adsorption sites is correlative to the number of unoccupied sites, which can be mathematically expressed by Eq. (22) [59].

𝑞𝑡 = 𝑞𝑒 (1 − 𝑒 −𝑘1 𝑡 )


Where 𝑞𝑡 is the amount of Ni(II) (mg.g-1) adsorbed at time 𝑡 (min), 𝑞𝑒 is the amount of Ni(II) adsorbed (mg.g-1) at equilibrium and 𝑘1 (min-1) the pseudo first-order rate constant. The Pseudo-second-order model is founded on adsorption equilibrium capacity, adopting that the rate of occupation of adsorption sites is correlative to the square of the number of unoccupied sites, as can be described by the Eq. (23) [60].

𝑞𝑡 =

𝑞𝑒2 𝑘2 𝑡


1+𝑞𝑒 𝑘2 𝑡

ℎ = 𝑘2 𝑞𝑒2



Where 𝑘2 ( is the second-order reaction rate equilibrium constant and ℎ (mg.g-1.min-1) is initial sorption rate. According to [61] the Intraparticle diffusion model can be defined by the Eq. (25).

𝑞𝑡 = 𝑘𝑑𝑖𝑓 𝑡 1/2


Where 𝑘𝑑𝑖𝑓 ( min-½) is the intraparticle diffusion constant. The Elovich model is described by [62] as can be seen in the Eq. (26).


𝑞𝑡 = ln(1 + 𝛼𝛽𝑡)



Where 𝛽 ( is associated to the surface coverage and activation energy for chemisorption and 𝛼 (mg.g-1.min-1) is the initial sorption rate. The application of the described models resulted in the values displayed in the Table 7. In all studied cases, the Pseudo-second-order model showed the highest 𝑅2 and low values of 𝜒 2 , 𝑀𝑆𝐸 and 𝑅𝑀𝑆𝐸, also focusing on the Elovich model, which showed some correlation. The Pseudo-second-order model is based on adsorption equilibrium capacity, assuming that the rate of occupation of biosorption sites is proportional to the square of the unoccupied number sites. Another assumption by this model is the chemical nature of the bind between ion and biosorbent, which may explain the rapid biosorption of 100% of Ni(II) in the initials minutes of treatment 2 (Fig. 6), even with the low contact time.


It is also interesting to observe that the Intraparticle diffusion model showed low values of 𝑅2 , supporting the assumption that biosorption may be the rate limiting step of the process [14]. This observation is enhanced when the experimental (𝑞𝑒(𝑒𝑥𝑝) ) and calculated (𝑞𝑒(𝑐𝑎𝑙) ) biosorption capacities are compared in Table 7, showing the same order of magnitude and confirming the applicability of the model. Other organic materials used in the biosorption of Ni(II) also showed better description by the Pseudo-second-order model; mustard oil cake [8], treated alga (Undaria pinnatifida) [14], wheat straw [15], golden shower parts (Cassia fistula) [16], water fern biomass (Azolla filiculoides) [17], bacterial strain (Leucobacter sp. N-4) [18].

3.4. Mechanism of biosorption of Ni(II) ions

The biosorption mechanism is based on different kinds of chemical and physical interactions between functional groups and sorbates in the liquid phase, involving mainly mechanisms of electrostatic interactions, complexation, formation of dipole and hydrogen bonds [63]. The Macauba endocarp has a complex structure, as seen in the characterization, and behavior, as shown in the optimization. Thus, the biosorption process may not essentially involve a single mechanism. The literature has shown that for many adsorption processes, different mechanisms, chemisorption and physisorption can occur simultaneously. Consequently, propose a single mechanism of sorption of heavy metal ions onto a biosorbent that control all the sorption process is almost not possible [63–66]. Nevertheless, based on the functional groups observed by FTIR, pHzpc analyze and the fit of the pseudo-second-order model in the kinetic study,


it’s possible to formulate the hypothesis that the sorption mechanism is highly influenced by the chemisorption. In such hypothesis, the biosorption mechanism would involve the binding of Ni(II) ions on the surface functional groups of the biosorbent, occupying the adsorption sites in a rate proportional to the square of the unoccupied number sites, until the equilibrium.

4. Conclusions

Biosorption of Ni(II) on Macauba endocarp without pre-treatment was investigated in fixed-bed column. The characterization of biosorbent showed that the Rosin–Rammle model better described the particle size distribution. The highest surface area and pore volume was exhibited by Sample B (Sauter diameter of 427 μm). The SEM micrographs showed that the surface of Macauba endocarp is covered of irregular walls, without defined pores. The FTIR spectrum revealed that Ni(II) biosorption can be related to functional hydroxyls and phenols groups on the surface of biosorbent. The biosorption optimization showed that the capacity of the bed up to the break-point (𝑡𝑢 ), total time equivalent to the packaging capacity of the column (𝑡𝑡 ) and total removal (𝑅𝑡 ) were affected by both pH and Ni(II) concentration. The Mass Transfer Zone (𝑀𝑇𝑍) was affected only by biosorption mass and equilibrium metal uptake (𝑞𝑒 ) was affected by pH, biosorption mass and Ni(II) concentration. Thus, it was possible to determine the equations that control the response variables and observe the highest removal of Ni(II) of 50.05% for a contact time of 3.00 min. Kinetics studies exhibited the appropriateness of Pseudo-second-order model to describe the phenomena involved on biosorption


kinetics. The results obtained in this paper showed the applicability of Macauba endocarp without pretreatment as a low-cost and environmentally friendly biosorbent for the treatment of wastewaters contaminated with Ni(II). For further advances of the presented system it would be interesting to conduct regeneration studies, in order to investigate the stability of the biosorbent and create a sustainable cycle.


The authors would like to gratefully acknowledge University Centre of Patos de Minas, for the financial support of this paper, Paradigma® vegetable oils for technical support and the Multiuser Laboratory of Chemistry Institute at Federal University of Uberlândia for providing the equipment and technical support for experiments involving electron microscopy.



B. Volesky, Z.R. Holan, Biosorption of Heavy Metals, Biotechnol. Prog. 11 (1995) 235–250. doi:10.1021/bp00033a001.


B. Volesky, Detoxification of metal-bearing effluents: Biosorption for the next century, Hydrometallurgy. 59 (2001) 203–216. doi:10.1016/S0304386X(00)00160-2.


U. Farooq, J.A. Kozinski, M.A. Khan, M. Athar, Biosorption of heavy metal ions using wheat based biosorbents - A review of the recent literature, Bioresour. Technol. 101 (2010) 5043–5053. doi:10.1016/j.biortech.2010.02.030. 24


A. Öztürk, Removal of nickel from aqueous solution by the bacterium Bacillus thuringiensis, J. Hazard. Mater. 147 (2007) 518–523. doi:10.1016/j.jhazmat.2007.01.047.


H. Xu, Y. Liu, J.-H. Tay, Effect of pH on nickel biosorption by aerobic granular sludge, Bioresour. Technol. 97 (2006) 359–363. doi:10.1016/j.biortech.2005.03.011.


World Health Organisation, Guidelines for Drinking Water Quality, 2011.


R.A.K. Rao, M.A. Khan, B.H. Jeon, Utilization of carbon derived from mustard oil cake (CMOC) for the removal of bivalent metal ions: Effect of anionic surfactant on the removal and recovery, J. Hazard. Mater. 173 (2010) 273–282. doi:10.1016/j.jhazmat.2009.08.080.


M.A. Khan, M. Ngabura, T.S.Y. Choong, H. Masood, L.A. Chuah, Biosorption and desorption of Nickel on oil cake: Batch and column studies, Bioresour. Technol. 103 (2012) 35–42. doi:10.1016/j.biortech.2011.09.065.


E. Malkoc, Y. Nuhoglu, Removal of Ni(II) ions from aqueous solutions using waste of tea factory: Adsorption on a fixed-bed column, J. Hazard. Mater. 135 (2006) 328–336. doi:10.1016/j.jhazmat.2005.11.070.


M.M.H. Khalil, K.Z. Al-Wakeel, S.S.A. El Rehim, H.A. El Monem, Efficient removal of ferric ions from aqueous medium by amine modified chitosan resins, J. Environ. Chem. Eng. 1 (2013) 566–573. doi:10.1016/j.jece.2013.06.022.


K.Z. Elwakeel, A.M. Elgarahy, S.H. Mohammad, Magnetic Schiff’s base sorbent based on shrimp peels wastes for consummate sorption of chromate, Water Sci. Technol. 76 (2017) 35–48. doi:10.2166/wst.2017.184.


K.Z. Elwakeel, A.M. Elgarahy, S.H. Mohammad, Use of beach bivalve shells 25

located at Port Said coast (Egypt) as a green approach for methylene blue removal, J. Environ. Chem. Eng. 5 (2017) 578–587. doi:10.1016/j.jece.2016.12.032. [13]

F. Veglio’, F. Beolchini, Removal of metals by biosorption: a review, Hydrometallurgy. 44 (1997) 301–316. doi:10.1016/S0304-386X(96)00059-X.


Z. Chen, W. Ma, M. Han, Biosorption of nickel and copper onto treated alga (Undaria pinnatifida): Application of isotherm and kinetic models, J. Hazard. Mater. 155 (2008) 327–333. doi:10.1016/j.jhazmat.2007.11.064.


H.D. Doan, A. Lohi, V.B.H. Dang, T. Dang-Vu, Removal of Zn+2 and Ni+2 by adsorption in a fixed bed of wheat straw, Process Saf. Environ. Prot. 86 (2008) 259–267. doi:10.1016/j.psep.2008.04.004.


M.A. Hanif, R. Nadeem, M.N. Zafar, K. Akhtar, H.N. Bhatti, Kinetic studies for Ni(II) biosorption from industrial wastewater by Cassia fistula (Golden Shower) biomass, J. Hazard. Mater. 145 (2007) 501–505. doi:10.1016/j.jhazmat.2007.01.022.


R. Rakhshaee, M. Khosravi, M.T. Ganji, Kinetic modeling and thermodynamic study to remove Pb(II), Cd(II), Ni(II) and Zn(II) from aqueous solution using dead and living Azolla filiculoides, J. Hazard. Mater. 134 (2006) 120–129. doi:10.1016/j.jhazmat.2005.10.042.


Y. Qu, H. Li, A. Li, F. Ma, J. Zhou, Identification and characterization of Leucobacter sp. N-4 for Ni (II) biosorption by response surface methodology, J. Hazard. Mater. 190 (2011) 869–875. doi:10.1016/j.jhazmat.2011.04.024.


D.M.R.E.A. Dissanayake, W.M.K.E.H. Wijesinghe, S.S. Iqbal, N. Priyantha, M.C.M. Iqbal, Isotherm and kinetic study on Ni(II) and Pb(II) biosorption by the 26

fern Asplenium nidus L., Ecol. Eng. 88 (2016) 237–241. doi:10.1016/j.ecoleng.2015.12.028. [20]

E. Malkoc, Ni(II) removal from aqueous solutions using cone biomass of Thuja orientalis, J. Hazard. Mater. 137 (2006) 899–908. doi:10.1016/j.jhazmat.2006.03.004.


R.A.K. Rao, M.A. Khan, Biosorption of bivalent metal ions from aqueous solution by an agricultural waste: Kinetics, thermodynamics and environmental effects, Colloids Surfaces A Physicochem. Eng. Asp. 332 (2009) 121–128. doi:10.1016/j.colsurfa.2008.09.005.


S. Uçar, M. Erdem, T. Tay, S. Karagöz, Removal of lead (II) and nickel (II) ions from aqueous solution using activated carbon prepared from rapeseed oil cake by Na2CO3 activation, Clean Technol. Environ. Policy. 17 (2014) 747–756. doi:10.1007/s10098-014-0830-8.


S.S. Vieira, Z.M. Magriotis, N.A. V Santos, M. das G. Cardoso, A.A. Saczk, Macauba palm (Acrocomia aculeata) cake from biodiesel processing: An efficient and low cost substrate for the adsorption of dyes, Chem. Eng. J. 183 (2012) 152–161. doi:10.1016/j.cej.2011.12.047.


J.R. Regalbuto, J. Robles, The engineering of Pt/Carbon Catalyst, University of Illinois, Chicago, 2004.


C.J. Geankoplis, Transport Processes and Unit Operations, Prentice-Hall Int. (1993) 937.


F.A. Pavan, Y. Gushikem, A.C. Mazzocato, S.L.P. Dias, E.C. Lima, Statistical design of experiments as a tool for optimizing the batch conditions to methylene blue biosorption on yellow passion fruit and mandarin peels, Dye. Pigment. 72 27

(2007) 256–266. doi:10.1016/j.dyepig.2005.09.001. [27]

Y. Bounouri, M. Berkani, A. Zamouche, L. Rycerz, Optimization and modeling of synthesis parameters of neodymium(III) bromide by dry method using full factorial design analysis, Arab. J. Chem. (2017). doi:10.1016/j.arabjc.2017.05.003.


N.P. Raval, P.U. Shah, N.K. Shah, Adsorptive removal of nickel(II) ions from aqueous environment: A review, J. Environ. Manage. 179 (2016) 1–20. doi:10.1016/j.jenvman.2016.04.045.


S.J. Kleinübing, E. Guibal, E.A. da Silva, M.G.C. da Silva, Copper and nickel competitive biosorption simulation from single and binary systems by Sargassum filipendula, Chem. Eng. J. 184 (2012) 16–22. doi:10.1016/j.cej.2011.11.023.


A. Macı́as-Garcı́a, E.M. Cuerda-Correa, M.. Dı́az-Dı́ez, Application of the Rosin–Rammler and Gates–Gaudin–Schuhmann models to the particle size distribution analysis of agglomerated cork, Mater. Charact. 52 (2004) 159–164. doi:10.1016/j.matchar.2004.04.007.


P. Rosin, E. Rammler, The laws governing the fineness of powdered coa, J. Inst. Fuel. 7 (1933) 29–36.


H. Yan, G.V. Barbosa-Cánovas, Size characterization of selected food powders by five particle size distribution functions, Rev. Agaroquimica Y Tecnol. Aliment. 3 (1997) 361–369. doi:10.1177/108201329700300507.


M. Thommes, K. Kaneko, A. V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol, K.S.W. Sing, Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution (IUPAC Technical Report), Pure Appl. Chem. 87 (2015) 1051–1069. doi:10.1515/pac-2014-1117. 28


A. Steyermark, Spectrometric identification of organic compounds. 3 ed., Microchem. J. 21 (1976) 496. doi:10.1016/0026-265X(76)90069-2.


R. Sharma, B. Singh, Removal of Ni (II) ions from aqueous solutions using modified rice straw in a fixed bed column, Bioresour. Technol. 146 (2013) 519– 524. doi:10.1016/j.biortech.2013.07.146.


A. Mishra, B.D. Tripathi, A.K. Rai, Packed-bed column biosorption of chromium(VI) and nickel(II) onto Fenton modified Hydrilla verticillata dried biomass, Ecotoxicol. Environ. Saf. 132 (2016) 420–428. doi:10.1016/j.ecoenv.2016.06.026.


C. Li, P. Champagne, Fixed-bed column study for the removal of cadmium (II) and nickel (II) ions from aqueous solutions using peat and mollusk shells, J. Hazard. Mater. 171 (2009) 872–878. doi:10.1016/j.jhazmat.2009.06.084.


L. Mangaleshwaran, A. Thirulogachandar, V. Rajasekar, C. Muthukumaran, K. Rasappan, Batch and fixed bed column studies on nickel (II) adsorption from aqueous solution by treated polyurethane foam, J. Taiwan Inst. Chem. Eng. 55 (2015) 112–118. doi:10.1016/j.jtice.2015.03.034.


K. Vijayaraghavan, J. Jegan, K. Palanivelu, M. Velan, Biosorption of copper, cobalt and nickel by marine green alga Ulva reticulata in a packed column, Chemosphere. 60 (2005) 419–426. doi:10.1016/j.chemosphere.2004.12.016.


A.S. Foust, L.A. Wenzel, C.W. Clump, L. Maus, L.B. Andersen, Principles of Unit Operations, 2 ed., John Wiley & Sons, New York, 1980.


D.C.K. Ko, J.F. Porter, G. McKay, Optimized correlations for the fixed-bed adsorption of metal ions on bone char, Chem. Eng. Sci. 55 (2000) 5819–5829. doi:10.1016/S0009-2509(00)00416-4. 29


T. Akar, Z. Kaynak, S. Ulusoy, D. Yuvaci, G. Ozsari, S.T. Akar, Enhanced biosorption of nickel(II) ions by silica-gel-immobilized waste biomass: Biosorption characteristics in batch and dynamic flow mode, J. Hazard. Mater. 163 (2009) 1134–1141. doi:10.1016/j.jhazmat.2008.07.084.


H.P. Chao, C.C. Chang, A. Nieva, Biosorption of heavy metals on Citrus maxima peel, passion fruit shell, and sugarcane bagasse in a fixed-bed column, J. Ind. Eng. Chem. 20 (2014) 3408–3414. doi:10.1016/j.jiec.2013.12.027.


N.R. Shinde, A. V. Bankar, A.R. Kumar, S.S. Zinjarde, Removal of Ni (II) ions from aqueous solutions by biosorption onto two strains of Yarrowia lipolytica, J. Environ. Manage. 102 (2012) 115–124. doi:10.1016/j.jenvman.2012.02.026.


C. da S. Junior Sousa, M. Nascimento, I.L.S. Ferreira, Modelamento da extração de níquel por D2EHPA em meio sulfúrico, Rev. Esc. Minas. 64 (2011) 447–452.


R.O. James, T.W. Healy, Adsorption of hydrolyzable metal ions at the oxide— water interface. I. Co(II) adsorption on SiO2 and TiO2 as model systems, J. Colloid Interface Sci. 40 (1972) 42–52. doi:10.1016/0021-9797(72)90172-5.


K.S. Smith, Metal Sorption on Mineral Surfaces: An Overview with Examples Relating to Mineral Deposits, Rev. Econ. Geol. 6A–6B (1999) 161–182. doi:


Z. Aksu, F. Gönen, Biosorption of phenol by immobilized activated sludge in a continuous packed bed: Prediction of breakthrough curves, Process Biochem. 39 (2004) 599–613. doi:10.1016/S0032-9592(03)00132-8.


S. Basha, Z.V.P. Murthy, B. Jha, Biosorption of hexavalent chromium by chemically modified seaweed, Cystoseira indica, Chem. Eng. J. 137 (2008) 480– 30

488. doi:10.1016/j.cej.2007.04.038. [50]

H.Y. Yen, J.Y. Li, Process optimization for Ni(II) removal from wastewater by calcined oyster shell powders using Taguchi method, J. Environ. Manage. 161 (2015) 344–349. doi:10.1016/j.jenvman.2015.07.024.


K. Periasamy, C. Namasivayam, Removal of Nickel (II) From Aqueous Solution and Nickel Plating Industry Wastewater Using an Agricultural Waste : Peanut Hulls, Waste Manag. 15 (1995) 63–68. doi:10.1016/0956-053X(94)00071-S.


C. Namasivayam, K. Ranganathan, Recycling of “waste” Fe(III)/Cr(III) hydroxide for the removal of nickel from wastewater: Adsorption and equilibrium studies, Waste Manag. 14 (1994) 709–716. doi:10.1016/0956053X(94)90058-2.


M. Rao, A. V. Parwate, A.G. Bhole, Removal of Cr6+ and Ni2+ from aqueous solution using bagasse and fly ash, Waste Manag. 22 (2002) 821–830. doi:10.1016/S0956-053X(02)00011-9.


N. Ortiz, M.A.F. Pires, J.C. Bressiani, Use of steel converter slag as nickel adsorber to wastewater treatment, Waste Manag. 21 (2001) 631–635. doi:10.1016/S0956-053X(00)00123-9.


M.V. Subbaiah, Y. Vijaya, N.S. Kumar, A.S. Reddy, A. Krishnaiah, Biosorption of nickel from aqueous solutions by Acacia leucocephala bark: Kinetics and equilibrium studies, Colloids Surfaces B Biointerfaces. 74 (2009) 260–265. doi:10.1016/j.colsurfb.2009.07.028.


T.L. Marques, V.N. Alves, L.M. Coelho, N.M.M. Coelho, Removal of Ni(II) from aqueous solution using Moringa oleifera seeds as a bioadsorbent, Water Sci. Technol. 65 (2012) 1435–1440. doi:10.2166/wst.2012.026. 31


U.K. Garg, M.P. Kaur, V.K. Garg, D. Sud, Removal of Nickel(II) from aqueous solution by adsorption on agricultural waste biomass using a response surface methodological approach, Bioresour. Technol. 99 (2008) 1325–1331. doi:10.1016/j.biortech.2007.02.011.


F.W. Sousa, A.G. Oliveira, J.P. Ribeiro, M.F. Rosa, D. Keukeleire, R.F. Nascimento, Green coconut shells applied as adsorbent for removal of toxic metal ions using fixed-bed column technology, J. Environ. Manage. 91 (2010) 1634–1640. doi:10.1016/j.jenvman.2010.02.011.


S. Lagergren, About the theory of so-called adsorption of soluble substances, K. Sven. Vetenskapsakad. Handl. 24 (1898) 1–39.


Y.S. Ho, Review of second-order models for adsorption systems, J. Hazard. Mater. 136 (2006) 681–689. doi:10.1016/j.jhazmat.2005.12.043.


Y. Liu, Y.J. Liu, Biosorption isotherms, kinetics and thermodynamics, Sep. Purif. Technol. 61 (2008) 229–242. doi:10.1016/j.seppur.2007.10.002.


M.J.D. Low, Kinetics of Chemisorption of Gases on Solids., Chem. Rev. 60 (1960) 267–312. doi:10.1021/cr60205a003.


I. Michalak, K. Chojnacka, A. Witek-Krowiak, State of the Art for the Biosorption Process—a Review, Appl. Biochem. Biotechnol. 170 (2013) 1389– 1416. doi:10.1007/s12010-013-0269-0.


F. Pagnanelli, F. Beolchini, A. Esposito, L. Toro, F. Vegliò, Mechanistic modeling of heavy metal biosorption in batch and membrane reactor systems, Hydrometallurgy. 71 (2003) 201–208. doi:10.1016/S0304-386X(03)00157-9.


S.M. Macfie, P.M. Welbourn, The Cell Wall as a Barrier to Uptake of Metal Ions in the Unicellular Green Alga Chlamydomonas reinhardtii (Chlorophyceae), 32

Arch. Environ. Contam. Toxicol. 39 (2000) 413–419. doi:10.1007/s002440010122. [66]

J. Zolgharnein, M. Bagtash, S. Feshki, P. Zolgharnein, D. Hammond, Crossed mixture process design optimization and adsorption characterization of multimetal (Cu(II), Zn(II) and Ni(II)) removal by modified Buxus sempervirens tree leaves, J. Taiwan Inst. Chem. Eng. 78 (2017) 104–117. doi:10.1016/j.jtice.2017.03.020.


Fig. 1. Experimental system for fixed-bed column operation.

Fig. 2. N2 physisorption isotherms of the Macauba endocarp samples. Sample A: (a). Sample B: (b). Sample C: (c).

Fig. 3. SEM images of Macauba endocarp samples. Sample A: (a), (b) and (c). Sample B: (d), (e) and (f). Sample C: (g), (h) and (i).

Fig. 4. FTIR spectrum of Macauba endocarp samples.

Fig. 5. pHzpc determination of Macauba endocarp sample B.

Fig. 6. Breakthrough curves for Ni(II) biosorption on Macauba endocarp (sample B) for the treatments performed.

Fig. 7. Influence of the factors pH (𝑥1 ), biosorbent mass (𝑥2 ) and Ni(II) initial concentration (𝑥3 ) on the response variables.


Fig.1 Click here to download high resolution image

Fig.2(a) Click here to download high resolution image

Fig.2(b) Click here to download high resolution image

Fig.2(c) Click here to download high resolution image

Fig.3(a) Click here to download high resolution image

Fig.3(b) Click here to download high resolution image

Fig.3(c) Click here to download high resolution image

Fig.3(d) Click here to download high resolution image

Fig.3(e) Click here to download high resolution image

Fig.3(f) Click here to download high resolution image

Fig.3(g) Click here to download high resolution image

Fig.3(h) Click here to download high resolution image

Fig.3(i) Click here to download high resolution image

Fig.4 Click here to download high resolution image

Fig.5 Click here to download high resolution image

Fig.6 Click here to download high resolution image

Fig.7(a) Click here to download high resolution image

Fig.7(b) Click here to download high resolution image

Fig.7(c) Click here to download high resolution image

Fig.7(d) Click here to download high resolution image

Fig.7(e) Click here to download high resolution image

Table 1 Characteristics diameters of Macauba endocarp samples. Samples A 105 187 123 142

Average diameters (μm) Arithmetic Sauter Surface Volumetric

B 342 427 373 390

C 436 511 457 474

Table 2 Parameters of the particle size distribution models for Macauba endocarp samples. Models


Parameters 𝜙𝑚𝑎𝑥 (μm) 𝑚 𝑅2 𝜒




𝑙 (μm) 𝑚 𝑅2 𝜒



Samples A 865

B 931

C 975




0.8968 0.3676 0.0191 0.1381 361

0.8959 1.1437 0.0772 0.2778 632

0.9702 1.2596 0.0915 0.3025 779


9.5772 0.9999 1.0753 0.1122 0.3350

10.4683 0.9999 1.2326 0.1363 0.3692

0.9974 0.0402 0.0005 0.0218


Table 3 Textural properties of Macauba endocarp samples at different granulometric ranges. Samples A B

BET surface area (m2.g-1) 1.7525 2.8113

Pore volume (cm3.g-1) 0.0063 0.0072

Average pore width (nm) 14.3989 10.3516

Table 4 Response variables of the biosorption optimization of Ni(II) ions on Macauba endocarp (sample B) in fixed-bed column. Operating conditions Tests Biosorbent Ni (II) no. pH mass (g) (mg.mL-1) (𝑥1 ) (𝑥2 ) (𝑥3 ) 1 (-1) 3.00 (+1) 150 (-1) 10 2 (+1) 6.50 (+1) 150 (-1) 10 3 (-1) 3.00 (-1) 100 (-1) 10 4 (+1) 6.50 (-1) 100 (-1) 10 5 (-1) 3.00 (+1) 150 (+1) 110 6 (+1) 6.50 (+1) 150 (+1) 110 7 (-1) 3.00 (-1) 100 (+1) 110 8 (+1) 6.50 (-1) 100 (+1) 110 9 (0) 4.75 (0) 125 (0) 60 10 (0) 4.75 (0) 125 (0) 60

Experimental results 𝑡𝑢 (min)

𝑡𝑡 (min)

𝑀𝑇𝑍 (cm)

1.18 42.37 10.10 30.06 1.20 1.15 1.35 1.34 1.98 1.59

42.30 185.18 30.20 171.73 12.50 22.28 21.43 46.62 36.13 30.00

6.81 5.40 2.66 3.30 6.33 6.64 3.75 3.89 4.73 4.74

𝑞𝑒 𝑅𝑡 (mg.g-1) (%) 0.17 0.70 0.19 0.94 0.47 0.88 1.46 2.70 0.97 0.74

11.43 50.05 8.16 46.41 3.38 6.01 2.82 12.59 9.78 8.12

Table 5 Statistical results of the factorial experiment 23 with a middle point for biosorption optimization of Ni(II) ions in fixed-bed column. Variables 𝑡𝑢

Factor Mean 𝑥1 𝑥3

Effect 9.2305 15.2753 -19.6682 36

Standard error of coefficient 2.0591 2.3022 2.3022

Significance level 0.004180 0.016052 0.005253





𝑥1 𝑥3 Mean 𝑥1 𝑥3 𝑥1 𝑥3 Mean 𝑥2 Mean 𝑥1 𝑥3 𝑥1 𝑥3 Mean 𝑥1 𝑥2 𝑥3 𝑥2 𝑥3

-15.3038 59.8369 79.8420 -81.6403 -62.3610 4.8230 2.8943 0.1588 0.2232 -0.2281 -0.1611 0.9223 0.7358 -0.7669 0.8748 -0.6389

2.3022 6.2053 6.9377 6.9377 6.9377 0.1626 0.1818 0.0161 0.0180 0.0180 0.0180 0.0696 0.0778 0.0778 0.0778 0.0778

0.015929 0.000071 0.001200 0.001068 0.004129 0.000000 0.000045 0.000062 0.000805 0.000717 0.004176 0.000044 0.005212 0.004372 0.002469 0.009314

Table 6 Comparison of the removal of Ni(II) with several adsorbents using batch and fixed-bed systems. Removal (%) Oyster shell 100.00 Activated carbon from peanut 100.00 hulls Waste Fe(III)/Cr(III) hydroxide 97.00 Fly ash 96.90 Steel converter slag 95.00 91.20 Thuja orientalis 91.00 Acacia leucocephala bark Moringa oleifera seeds 90.00 82.00 Undaria pinnatifida Phaseolus vulgaris L. 79.84 Sugarcane bagasse 79.00 Green coconut shells 69.46 Macauba endocarp 50.05

Contact time (min) 60.00 120.00 150.00 90.00 240.00 30.00 120.00 5.00 120.00 60.00 4.75 3.00





Batch Batch Batch Batch Batch Batch Batch Batch Batch Fixed-bed column Batch Fixed-bed column Fixed-bed column

[50] [51] [52] [53] [54] [20] [55] [56] [14] [42] [57] [58] This paper

Waste of tea factory


Fixed-bed column [9]


Table 7 Parameters of the kinetics models for biosorption optimization of Ni(II) ions on Macauba endocarp (sample B) in fixed-bed column. Treatment 𝑞𝑒 (mg.g-1) 𝑘1 (min-1) Pseudo- 𝑅2 first𝜒2 order


Intr. diffusion


1 12.688 0.000 0.5651 0.6343 0.0008 𝑀𝑆𝐸 0.0291 𝑅𝑀𝑆𝐸 -1 𝑞𝑒(𝑐𝑎𝑙) (mg.g ) 0.193 𝑞𝑒(𝑒𝑥𝑝) (mg.g-1) 0.172 𝑘2 ( 0.061 h (mg.g-1.min-1) 0.012 0.9734 𝑅2 0.0198 𝜒2 0.0001 𝑀𝑆𝐸 0.0072 𝑅𝑀𝑆𝐸 -1 -½ 0.009 𝑘𝑑𝑖𝑓 ( .min ) 0.9655 𝑅2 2 0.0214 𝜒 0.0001 𝑀𝑆𝐸 0.0082 𝑅𝑀𝑆𝐸 𝛼 (mg.g-1.min-1) 0.006 26.008 𝛽 ( 2 0.9702 𝑅 0.0085 𝜒2 0.0000 𝑀𝑆𝐸 0.0066 𝑅𝑀𝑆𝐸

2 12.017 0.000 0.9102 0.3567 0.0043 0.0653 1.334 0.704 0.002 0.003 0.9990 0.0031 0.0000 0.0067 0.036 0.9538 0.2448 0.0022 0.0469 0.005 2.039 0.9978 0.0075 0.0001 0.0101

3 12.811 0.000 0.7947 7.5027 0.2885 0.5371 0.239 0.187 0.170 0.041 0.9836 0.1494 0.0013 0.0359 0.023 0.9656 1.2984 0.0194 0.1393 0.012 13.370 0.9774 0.5108 0.0056 0.0749

4 12.814 0.000 0.8379 0.5355 0.0142 0.1191 1.685 0.944 0.002 0.004 0.9907 0.0306 0.0008 0.0286 0.052 0.9582 0.2340 0.0037 0.0605 0.008 1.710 0.9868 0.0443 0.0012 0.0340


5 0.293 0.137 0.9917 0.4284 0.0052 0.0723 0.312 0.469 0.842 0.263 0.9992 0.3014 0.0039 0.0622 0.036 0.4193 1.1230 0.0224 0.1497 8.408 33.585 0.9959 0.1398 0.0020 0.0444

6 0.355 0.078 0.9422 4.0211 0.0792 0.2814 0.396 0.879 0.284 0.113 0.9807 3.2076 0.0685 0.2618 0.041 0.6970 0.2864 0.0061 0.0780 2.371 24.758 0.9740 2.9660 0.0652 0.2553

7 12.820 0.001 0.2076 16.680 92.4607 1.5686 0.601 1.457 0.517 0.311 0.9947 4.3779 0.1239 0.3520 0.095 0.7833 1.7757 0.1028 0.3206 7.427 15.151 0.9899 2.7107 0.0889 0.2982

8 12.814 0.001 0.6878 3.7971 0.2233 0.4726 3.669 2.702 0.002 0.009 0.9904 0.0841 0.0068 0.0827 0.159 0.9545 0.3081 0.0326 0.1805 0.043 0.971 0.9811 0.1350 0.0135 0.1162

9 0.724 0.025 0.9832 0.4662 0.0173 0.1314 0.903 0.971 0.029 0.026 0.9952 0.1785 0.0074 0.0858 0.063 0.9721 0.1898 0.0104 0.1019 11.375 17.169 0.7436 1.3533 0.0432 0.2079

10 0.411 0.047 0.9957 1.3811 0.0258 0.1605 0.501 0.738 0.103 0.051 0.9995 0.6778 0.0145 0.1204 0.047 0.9422 0.4521 0.0147 0.1213 0.044 8.454 0.9961 0.1635 0.0042 0.0646