Comparative Adsorption Performance of Carbon-containing Hydroxyapatite Derived Tenggiri ( Scomberomorini ) and Belida ( Chitala ) Fish Bone for Methylene Blue

The utilization of fishbone as the carbon source for methylene blue adsorption has been successfully studied. Fishbone was prepared from two kinds of fish such as marine fisheries (ex. Tenggiri ) and freshwater fisheries (ex. Belida ). The carbons were prepared by carbonation of fishbone powder at 500 °C for 2 h. Physical properties of carbons were characterized using Fourier transform infrared (FTIR) spectroscopy, X-ray diffraction (XRD), wavelength dispersive X-ray fluorescence (WDXRF), Scanning Electron Microscope (SEM), and hydrophobicity. The carbons were utilized as the adsorbent for removing methylene blue by varying the contact time, initial dye concentration, and temperature. It is concluded that both carbons can very good adsorb the methylene blue. The adsorption performance of carbon (TFC) from Tenggiri fish is better than carbon (BFC) from Belida fish. The adsorption was well fitted with the Langmuir adsorption model (R 2 ~ 0.998) and the pseudo-second-order model. This indicated that the dye molecules were adsorbed on the surface-active site of carbon via chemical binding, forming an adsorbate monolayer. Thermodynamic parameters, including the Gibbs free energy ( Δ G ), enthalpy ( Δ H ), and entropy ( Δ S ), indicated that the adsorption of methylene blue onto the carbon from fishbone was spontaneous. Thus, carbon from fishbone can be applied as a low-cost adsorbent to treat industrial effluents contaminated with methylene blue.


Introduction
The development of industries that require synthetic dyes, such as the textile, paper, rub-[1-3]. Research efforts in degrading methylene blue as synthetic dyes are often carried out using many adsorbents. Adsorption is a familiar separation technique known since the earlier history of science and is considered as an efficient and user-friendly method for eliminating a wide range of toxic pollutants from wastewater. Adsorption is a physicochemical method of treating aqueous effluent that is rapidly gaining prominence due to its proven efficiency and great potential as means of producing quality effluent [4][5][6]. Many purposes of separation and purification with the adsorption technology have been used extensively in industrial processes. An essential application of adsorption process using suitable adsorbents is considered in the removal of coloured and colourless organic pollutants from industrial wastewater [7].
Recently, many researchers have been efforted to find cheaper and more efficient alternative materials as dyes adsorbent, such as agricultural residues [8], clay minerals [1,3], cashew nut shell [2], fly ash [9], peat [10], yellow passion fruit peel [11], wood powder and lignin [12,13]. However, the cost of its commercial application is still very high and low efficiency. Indonesian country is the largest tropical archipelagic country in the world which has a vast ocean and a large number of many large and small islands. The potential of the Indonesian fishery sector is the largest in the world, including capture fisheries (ex. Tenggiri (Scomberomorini)) and aquaculture (ex. Belida (Chitala)) with a sustainable production potential of around 67 million tons/year [14,15]. The consequence of abundant fish yields is that large amounts of fish bone waste have been produced. The compositions of fish bone waste are composed of calcium phosphate, collagen fiber, calcium carbonate, and hydroxyapatite [16,17], which are very useful for several applications, namely photocatalysts [18,19], catalysts [20,21], adsorbents [22][23][24], etc. Captivated by the high carbon-containing hydroxyapatite of the fish bone, our research team has explored the usage of modified fish bone as a catalyst in styrene oxidation [15,25] and an adsorbent in methylene blue removal [26,27].
In this study, we have modified fish bone to be the highly adsorptive carbon source as an adsorbent to remove methylene blue dye from the wastewater. Firstly, the fish bone was ground to powder and then activated by thermal activation at 500 °C for 2 h. The carbon was characterized by FTIR, XRD, XRF, SEM, Nitrogen adsorption-desorption, and hydrophobicity. Finally, the carbon will be tested to ad-sorb methylene blue on the effect contact time variation for the kinetic study and the initial methylene blue concentration effect for Langmuir and Freundlich isotherm studies. Furthermore, variation temperature for investigating thermodynamic parameters was also carried out.

Carbonation Process
Tenggiri (Scomberomorini) and Belida (Chitala) fish bone were obtained from food company in Samarinda, East Kalimantan, Indonesia. The fish bones were washed with boiling water to separate them from other impurities. The fish bone was dried in an oven at 110 °C for overnight and it was crushed to fine powder. The powder was carbonized in a furnace at 500 °C for 2 h in an air atmosphere to form carbon. The fish bone-derived carbon is labelled as TFC for Tenggiri fish bone and BFC for Belida fish bone.

Carbon Characterization
The samples were characterized by using WDXRF, FTIR, XRD, SEM-EDX, BET adsorption-desorption and Hydrophobicity. The elements composition from the carbon was determined by using 1 kW wavelength dispersive Xray fluorescence (WDXRF PANalytical, Minipal 4). FTIR spectrometer (IR−Prestige−21 Shimadzu) was used to identity th e fu n c t io n a l g r o u p i n t h e s a m p le with a spectral resolution of 2 cm −1 , scans 10 s, at 20 °C and range of wavenumber from 400 to 4000 cm −1 . XRD instrument (Phillips PANalytical X'Pert PRO) was used to determine of the crystallinity and phase content of the sample with the Cu Kα (λ = 1.5406 Å) radiation as the diffracted monochromatic beam at 40 kV and 40 mA and the pattern was scanned in the 2θ ranges between 7° and 60° at a step 0.03° and step time 1 s. SEM (FEI Inspect S50) instrument was used to investigate the surface morphology of sample with an accelerating voltage of 15 kV. The hydrophobicity was correlated with the amount adsorbed water in the samples. In an experiment, samples (1 g) were dried at 110 °C overnight in an oven to remove all physically adsorbed water. Distillated water (0.75 L) was filled into desiccators for overnight. After dehydration, the samples were exposed to water vapor by placing them into the water-filled desiccators at room temperature and weighed every 30 min. The percentage of adsorbed water as a function of time was determined follow the Equation (1): (1) where, mt represents the sample mass after adsorption of water and m0 represents the initial mass of the sample.

Adsorption Test
The performance of adsorbent was tested using methylene blue (C16H18ClN3S.xH2O, x = 2-3, Merck) dyes. The methylene blue structure was shown in Figure 1. The adsorption was carried out by mixing 0.250 g of the adsorbent with 25 mL of methylene blue solution (100 mg.L −1 ) in a 100 mL beaker glass at room temperature for the duration of 5, 10, 15, 20, 25, 30, 60, 120, and 240 min. Furthermore, the adsorbent was separated from the filtrate by centrifugation, and the residual dye in the filtrate was analyzed using UV-vis spectrophotometer by monitoring the changes in absorbance at 660 nm for methylene blue. The effect of initial dye concentration was investigated by varying initial concentrations 100, 200, 300 and 400 mg.L −1 . In this study, the thermodynamic parameters (enthalpy, entropy and free energy) were also investigated following the adsorption data of the effect of temperature (30, 40 and 50 °C).
The adsorption efficiency of the adsorbent was determined following this equation (2) [28,29]: (2) where, C0 is the dye initial concentration and Ct is the dye concentration after adsorption time t (mg.L −1 ) in the solution. The adsorption capacity qt of adsorbent (mg/g) was calculated with following the Equation (3) [5,30]: where, V is the volume of dye solution (mL) and W is weight adsorbent (g).

Physical Properties
The results from wavelength dispersive Xray fluorescence (WDXRF) show that both of samples were consisted of major elements calcium (Ca) and phosphorus (P). Both of the samples also consist the transition metal as minor constituents which are ferrum (Fe), cuprum (Cu), and zinkum (Zn). The complete list of elements and compounds was displayed in Table  1.
The FTIR spectra of (a) HA, (b) BFC and (c) TFC are shown in Figure 2. The wavenumber to detect the functional group of hydroxyapatite (HA) were measured in range 4000 to 400 cm -1 . All of the spectra show the absorption bands around 2500-3600 cm −1 corresponded to the stretching mode of hydroxyl (O−H) group of organic compounds. The HA in the sample was proven by the presence of carbonate ion (CO3 2− ) and phosphate ion (PO4 3− ) groups [16,[31][32][33][34][35]. The C−O stretching was investigated by the absorption bands at 1449, 1272 and 877 cm −1 which correlated with the carbonate ion (CO3 2− ). The phosphate group (PO4 3− ) absorption band were characterized from wavenumber 1150 to 460 cm -1 . The P−O stretching asymmetric was identified by the absorption bands around 1150-1000 cm -1 which correlated with the phosphate ion (PO4 3− ). The bending vibration of PO4 3− was observed by bands located at 620-510 cm −1 . The bands at 877 cm −1 , which was assigned to the acidic phosphate group (HPO4 2− ) [36].
The crystallinity of (a) HA, (b) BFC and (c) TFC were shown by using the XRD pattern in    Table 2. Figure 4 shows the SEM images of BFC and TFC. In Figure 4, the surface morphology of both of samples are roughness and irregular. The particle size of TFC appears bigger than BFC.
The analysis of Nitrogen adsorptiondesorption isotherm was used to determine the surface area and pore structure of all samples. Figure 5 shows nitrogen adsorption-desorption isotherms of TFC and BFC. Both isotherms were of Type IV in the IUPAC classifications, which are a typical isotherm for mesoporous materials. The isotherms exhibited that hysteresis loops in the relative pressure range ~0.4-1.0 for TFC and BFC. The surface area, pore volume and mean pore size completely can be seen in Table 2. Mean pore size of both samples more than 2 nm indicates the presence of uniform mesopores. Surface area of BFC higher than TFC, it was caused the particle size of TFC bigger than BFC that can be proven in SEM images. Figure 6 shows percentage of adsorbed water of BFC and TFC. The percentage of adsorbed water of TFC was significantly higher than BFC. It indicates that BFC more hydrophobic than TFC. From the SEM results,    it can be shown that the particle size of BFC is smaller than TFC so that BFC is very difficult to contact with water as a solvent for methylene blue. From the XRD results, we can determine the percentage of amorphous carbon of BFC is ~75% higher than TFC (~73%), the higher percentage of amorphous carbon causes BFC is difficult to contact with water as methylene blue solvent. Table 3 shows that the methylene blue adsorption capacity is affected by three factors, i.e. surface area, active sites, and targeted sorbents. The torrefied rice husk, which is low in surface area and scarce of active sites, gave low adsorption capacity [6]. Although TiO2sulfonated carbon-derived from Eichhornia crassipes is comparatively high in surface area and functionalized with functional groups, it gave low adsorption capacity due to this adsorbent is selective to anionic dye (Congo Red) but not cationic dye (Methylene Blue) [26]. Astonishingly, graphene oxide/alginate displayed rel-atively low surface area but contributed a high removal ability [41], this owing to its abundance of active functional groups. Bamboobased activated carbon, which consists of plentiful active sites and high surface area, showed very good methylene blue removal capability [43]. Our current study with pristine carbonderived fish bones gave good adsorption capacities because of the charge interaction, whereby the hydroxyapatite functional groups, i.e. CO3 2− and PO4 3− groups, could be attracted to S + and N + from cationic methylene blue. Nonetheless, the carbon-derived fish bones can be further improved by Van der Waals force by functionalizing with aromatic-structured, polymeric compounds to remove the heterocyclic aromatic methylene blue effectively.

Effect of Contact Time
The effect of contact time on the percentage dye removal of methylene blue onto TFC and BFC adsorbents are shown in Figure 7. These results indicate that the equilibrium adsorption such as ⁓90% for BFC and ⁓98% for TFC was reached within 20 min by both adsorbents and dye removal percentage increased with increasing contact time. Both adsorbents rapidly adsorbed methylene blue within the first 5 min of contact time with adsorption capacity of ⁓85% for BFC and ⁓94% for TFC. The changes in adsorption capacity increased gradually until the equilibrium was achieved within 20 min. The dye removal by BFC is lower than of TFC, most probably caused by the physical properties such as the hydrophobicity, which was BFC more hydrophobic than TFC. The solvent for methylene blue is distillate water, this makes it dif-ficult for hydrophobic carbon to make contact with the solution and may have hindered the effective adsorption of the methylene blue onto the surface of the adsorbent. Furthermore, after adsorption duration time 240 min the equilibrium methylene blue achieved 92.8% for BFC and 99.7% for TFC.

Adsorption Kinetics
The pseudo-first-order kinetic models and the pseudo-second-order kinetic model were use to check the adsorption kinetic of methylene blue adsorption onto TFC and BFC adsorbent. The pseudo-first-order kinetic models by Lagergren base on the sorption capacity of adsorbent is expressed as [45][46][47]: (4) where, k1 (g/mg.h) is the rate constant for Lagergren pseudo first-order, qe and qt are the amounts of dye adsorbed per gram of adsorbent (mg/g) at equilibrium and any time t. The slope and intercept from plot ln(qe−qt) versus t was used to determine the value of k1 and qe.
The pseudo-second-order kinetic model by Ho, Y. S., and McKay, G is also based on the sorption capacity of the adsorbent with the equation [48][49]: where, k2 (g.mg −1 .h −1 ) is the rate constant for pseudo-second-order, qe and qt are the amounts of dye adsorbed per gram of adsorbent (mg.g −1 ) at equilibrium and any time t. The intercept and slope form the plot t versus t/qt were used to calculate the value of k2 and qe,cal.
The pseudo-first-order and pseudo-secondorder kinetics plots for both TFC and BFC as

Bulletin of Chemical Reaction Engineering & Catalysis, 17 (3), 2022, 571
Copyright © 2022, ISSN 1978-2993 adsorbent of methylene blue dye were checked with equations (4) and (5), respectively. The rate constant (k), equilibrium sorption calculated (qe,cal) were determined by slope and intercept of the plots. Completely, the value of rate constant (k), equilibrium sorption calculated (qe,cal), equilibrium sorption experiment (qe,exp) and the correlation coefficients (R 2 ) were listed in Table 4. The value of the correlation coefficient (R 2 ) for pseudo-first-order of TFC and BFC as 0.814 and 0.540 were lower compared to the value for pseudo-second-order of both adsorbents by 1.000, where the pseudo-secondorder indicates more accurate. This phenomenon is in line with the report by Ho and McKay [50] that concluded the first order equation of Lagergren is not well-suited for the whole range of contact time and is generally applicable over the initial stage of the adsorption process in many cases. The result of this research showed that the adsorption of methylene blue onto BFC adsorbents did not follow the pseudofirst-order kinetic. Both adsorbent follow pseudo-second-order models also can be proven by the difference in qe,cal with qe,exp for both adsorbents in pseudosecond-order were lower compared with pseudo-first-order. Therefore, the adsorption of methylene blue dye onto both adsorbents were concluded to follow the pseudo-second-order adsorption mechanism model.

Adsorption Isotherms
The Langmuir and the Freundlich isotherm were used to identify the adsorption isotherm in adsorption process onto TFC and BFC adsorbent. The Langmuir isotherm is correlated with monolayer sorption onto a surface that equation is given by [51][52][53]: where, Ce and qe are the residual dye concentration (mg.L −1 ) and the number of the dye adsorbed on the sorbent at equilibrium (mg.g −1 ). Meanwhile, Qmax and b are Langmuir constant which correlated with maximum adsorption ca-pacity and bonding energy of adsorption, respectively. The slope and intercept of the linier plot of Ce/qe versus Ce was used to determine Qmax and b.
The Freundlich isotherm model assumes heterogeneous site energies of sorption. The Freundlich equation is followed as [30,52]: where, KF and n are the Freundlich constant which indicate adsorption capacity and adsorption intensity, respectively. KF (mg.g −1 ) and n (L.mg −1 ) 1/n were calculate from the intercept and slope of the plot of ln qe versus ln Ce.
At equilibrium, the methylene blue removal capacity increased from 9.3 to 20.8 mg.g −1 for BFC and 9.9 to 24.4 mg.g −1 when the initial concentration of both dyes increased from 100 to 400 mg.L −1 . The increasing of the amount of dye removal at equilibrium condition was caused by an increasing the initial concentration that can increase the driving force to reduce all mass transfer resistances of the methylene blue molecules between the aqueous and solid phase of the adsorbent [54].
The Langmuir and the Freundlich adsorption isotherms were used to describe the interaction between the dye and adsorbent that follow equations (6) and (7), respectively. For the removal of methylene blue by TFC and BFC adsorbents, 25 mL of various dyes concentrations (between 100 and 400 mg.L −1 ) were mixed with 0.250 mg sorbent, then stirred at 300 rpm for 240 min at room temperature. From the Langmuir equation can be calculated the value of the constants of Qmax, b and R 2 as 24.8 mg.g −1 , 0.257 and 0.9986 for TFC and 20.8 mg.g −1 , 0.0889 and 0.9999 for BFC, respectively. Base on the value of maximum adsorption capacity (Qmax), bonding energy (b) can be used to conclude that adsorption of methylene blue onto TFC was easier than BFC adsorbent.
The favorable adsorption onto adsorbent was investigated base on the constant KF, n and R 2 were determined from the Freundlich isotherm that was obtained 10.752, 5.831 and

Dye removal
Temp. Base on KF value can be concluded that the methylene blue adsorption process was the heterogeneous adsorption process on the surface of onto TFC and BFC adsorbent which was done through a multi-layer adsorption mechanism. The n values of methylene blue adsorption process are more than 1, which indicates favorable adsorption of both dyes onto TFC and BFC adsorbents. The complete data of Langmuir and the Freundlich isotherm were listed in Table 5.

Effect of Temperature
The thermodynamic parameters free energy (ΔG), enthalpy (ΔH) and entropy (ΔS) were determined by the experiments that were created at different temperatures. The value changes of ΔH and ΔS of adsorption were estimated from Van't Hoff equation [6,55]: with KC =C1/C2, is the equilibrium constant and T, R, C1 and C2 are temperature (K), the gas constant (8.314 J.K −1 .mol −1 ), the quantity of methylene blue dye adsorbed per unit mass of adsorbent and the concentration of methylene blue dye in aqueous phase, respectively. The linier plot of ln KC versus 1/T was used to determine of ΔS and ΔH which calculated from the slope and intercept. The positive value of ΔS indicates that the increase in randomness of ongoing process. The negative value of ΔH indicates that the adsorption process is exothermic in nature and otherwise, the positive value of ΔH indicates that the adsorption process is endothermic in nature. Furthermore, the value of ΔG was calculated by using the equation ∆G 0 =∆H−T∆S. The negative value of ΔG at each temperature shows the feasibility and spontaneity of adsorption process.
The Arrhenius equation was used to determine activation energy (Ea) for adsorption process onto TFC and BFC adsorbent. The equation is given as [30,55]: k, Ea ( kJ.mol −1 ), T (K), R (J.mol −1 .K −1 ) and A are the rate constant, Arrhenius activation energy, temperature of the adsorption medium, the ideal gas constant (8.314 J.mol −1 .K −1 ), and the Arrhenius factor, respectively. The slope of plotting ln k versus 1/T was used to calculate the Arrhenius activation energy. The removal of MB dye at varying temperatures increased from 94.2 to 99.6% (at room temperature), 93.3 to 99.3% (40 °C) and 92.8 to 98.5% (50 °C). It can be clearly seen that by increasing the temperature, the removal percentage of methylene blue dye decreased slowly.
The thermodynamic parameter value of ΔG, ΔH and ΔS are calculated based on the slope and intercept of Van't Hoff's plots follow equation (8). The value of ΔG, ΔH, and ΔS were listed in Table 6. The feasibility and spontaneity of ongoing MB adsorption onto TFC adsorbent was proven by negative value of free energy (ΔG). The negative value of enthalpy (ΔH) indicates that the adsorption process of MB onto TFC adsorbent is exothermic in nature. The activation energy (Ea) for adsorption process onto TFC adsorbent was calculated by the Arrhenius equation (9). The activation energy value for adsorption process onto TFC adsorbent was 7.157 kJ.mol −1 .

Conclusions
In this study, modified fishbone (Tenggiri and Belida) to be carbon (TFC and BFC) adsorbent has been carried out via a simple thermal activation method. The thermal activation was conducted at a carbonation temperature 500 °C