Carboxymethyl Cellulose Nanoadsorbent for Remediation of Polluted Water

The development of nanoadsorbents for remediation of polluted water in order to obtain clean and healthy water quality has been carried out, namely the incorporation of chitosan, magnetic, and activated carbon materials. The activated carbon used is the result of the synthesis of banana peel waste nanocrystals, while the magnetic is Fe 3 O 4 . The method used in this study is an experimental method with coprecipitation through several stages, namely (1) magnetic synthesis of Fe 3 O 4 by thecoprecipitation method, (2) preparation of chitosan solution, (3) synthesis of activated carbon nanocrystals from banana peel waste by the milling process, (4) merger of the three materials, and (5) characterization with SEM/EDX, XRD, FTIR, BET, PSA, TGA, and AAS to test the performance of the material against polluted water. The study found that 210 minutes was the optimal time for the heavy metal ions Fe, Mn, Zn, and Pb to adsorption.The best sample was sample S4 with a ratio of 1:2:2 with adsorption for Zn 92.43%, Fe 95.44%, Mn 89.54%, and Pb 84.38%. For the heavy metal ions: Mn 5624 mg/g, Fe 5849.4 mg/g, Zn 4894.22 mg/g, and Pb 468.2 mg/g, the Langmuir model was used. The adsorption kinetics showed that the reaction order for Pb, Mn, Zn, and Fe ions varied with pseudo-first order and pseudo-second order. Carboxymethyl cellulose nanoadsorbents are effective in remediating the water contaminated with heavy metals, such as Pb, Mn, Zn, and Fe, meeting the environmental health quality standards for water media for sanitation hygiene purposes.


INTRODUCTION
The contamination of the aquatic environment is an interesting environmental issue to be researched. Various efforts have been made to minimize pollution of the aquatic environment, one of which is the adsorption method. Adsorption is an efficient method for removing various heavy metals which is being developed nowadays . Chitosan is one of the often used materials, where the amine (NH 2 ) and hydroxyl (OH) in high concentrations can bind to metal ions and remove bacteria through coordination mechanisms and electrostatic interactions.
However, not all metal ions can be adsorbed on the chitosan material due to poor acid resistance (Olaoye et al., 2020). In addition, it also has a weakness, namely the small adsorption capacity so that other materials are needed to support the performance of this material, namely activated carbon. Activated carbon is a biomass waste that is used as the basic material for making adsorbents, which includes the plants that contain lots of carboxylic acids and hydroxyl groups OH, namely banana peel waste (Alinezhad et al., 2020). On the basis of previous research, activated carbon from banana peel waste has the potential to become an adsorbent material. Activated carbon as an adsorbent material has a large surface area, quite large pores and is hydrophobic (does not mix with water), can eliminate odors and cloudy colors in polluted water (Safari et al., 2019;. Although activated carbon from banana peel waste is a material that is quite potential for adsorption of heavy metal ions, but as an adsorbent material it has several weaknesses, including being difficult to separate in the adsorption process and it cannot be used repeatedly, so modifications are needed with the development of other technologies to assist the separation, namely magnetic technology (Siregar et al., 2021;Yu et al., 2021). The process is carried out with rod magnets so that the adsorbent can be taken and reused repeatedly. The magnetic material used is Fe 3 O 4 that is synthesized by the coprecipitation method, which is a bottom-up method of making nanoparticles by involving NH 4 OH as the precipitating solution (Hu et al., 2020;Adusei et al., 2022). In this study, the adsorbent was carboxymethyl cellulose (CMC) nanoadsorbent which consisted of 3 types of materials, namely chitosan, magnetic Fe 3 O 4 , activated carbon from banana peel waste. In turn, the adsorbate (absorbed substance) is a metal ion. The combination of the three materials by using the addition of glutaraldehyde (Gameli et al., 2022;Safari et al., 2019). The addition of glutaraldehyde is a crosslinked agent which is strong enough so that it is expected to increase the adsorption ability of the three materials.
The formation of CMC nanoadsorbent powder is expected to be a new material that can be applied as an adsorbent material in the remediation of polluted water. In the mechanism of the adsorption process, the metal ions dissolved in water bind bi-polar so that they become one unit and become bound or absorbed with CMC nanoadsorbent material; then, for separation with water, assisted separation with a bar magnet at the bottom are employed so that clean water can be obtained (  . Then, the solution was heated and stirred using a water bath shaker. The temperature used is 70 °C for 30 minutes. The solution was then allowed to stand and a magnetite precipitate (Fe 3 O 4 ) was formed and then rinsed using distilled water until the pH was neutral according to the pH of the distilled water used. Afterwards, it was dried and then filtered using filter paper and subsequently dried in a desiccator for 24 hours.

Preparation of chitosan solution
A total of 50 mL of 1% acetic acid solution was prepared with a composition of 10 mL of 10% acetic acid added 40 mL of distilled water, then stirred and added 0.5 g of chitosan into 1% acetic acid solution and then stirred until evenly distributed. This mixture is called chitosan solution (Yuvaraja & Venkata, 2016;Abbasi, 2017).

Synthesis of activated carbon nanocrystals from banana peel waste
First, the banana peel was dried with a preparation technique to remove the dirt and impurities attached to the banana peel and reduce the water content. It was then ground with a mortar and pestle, sieve with a sieve, and smoothed again with a blender; then, the banana peel powder was subjected to High Energy Milling (HEM). The milling process was carried out for 24 hours with an interval of 30 minutes for each 90 minute cycle (Chen et al., 2017;. Carbonization occurred at low temperature of 250 °C for 10 hours. The results of the carbonization were chemically activated by using a concentrated HCl solution of 25%. The particles and NaOH solution were stirred for 2 hours then the sample was placed into the oven at 80 °C for 4 hours.

Synthesis of incorporation of chitosan, magnetic (Fe 3 O 4 ), activated carbon
The 0.5 g chitosan solution that had been made was then stirred using a magnetic stirrer and 2 mL of 50% glutaraldehyde was added, stirred for 30 minutes and then stirred again manually when 0.5 g of magnetite and carbon were added until well mixed and soaked in 1M NaOH solution for 12 hours, then washed using distilled water until neutral and afterwards stored in a desiccator to dry and a chitosan-magnetite-carbon adsorbent was formed. Furthermore, the characterization of SEM/EDX, XRD, FTIR, PSA, BET/ BJH, TGA and AAS was carried out to test the adsorption performance (Gameli et al., 2022;Petcharoen & Sirivat, 2012). Comparison of the variations of the three materials to be synthesized with seven samples can be seen in Table 1. The seven samples comprised: S1 included 0.5 g of chitosan, 0.5 g of magnetic and 0.5 g of banana peel carbon. S2 is 0.5 g of chitosan, 0.5 g of magnetic and 1.0 g of banana peel activated carbon. S3 is 0.5 g of chitosan, 1.0 g of magnetic and 1.0 g of banana peel activated carbon. S4 is 0.5 g of chitosan, 1.0 g of magnetic and 1.0 g of banana peel activated carbon. S5 is 1.0 g of chitosan, 1.0 g of magnetic and 0.5 g of banana peel activated carbon. S6 is 0.5 g chitosan, 1.0 g magnetic and 0.5 g banana peel activated carbon and the last S7 is 1.0 g chitosan, 0.5 g magnetic and 1.0 g banana peel activated carbon.

Characterization
Furthermore, characterization was carried out using SEM/EDX, XRD, FTIR, PSA, BET/BJH and AAS to test water quality. SEM/EDX characterization was used to investigate the morphology and identify the clearer composition of banana peel activated carbon, Fe 3 O 4 nanoparticles and CMC material nanoadsorbent. XRD characterization was used to investigate the phase formed from Fe 3 O 4 magnetic powder and CMC material nanoadsorbent powder. FTIR characterization was used to investigate the functional groups formed on Fe 3 O 4 magnetic and CMC material nanoadsorbent powder. PSA characterization was used to investigate the particle size of Fe 3 O 4 magnetic powder and CMC material nanoadsorbent powder. BET/BJH characterization was used to investigate the pore size and surface area of banana peel activated carbon and CMC nanoadsorbent material. TGA characterization was used to investigate the change in weight % with temperature.

Adsorption performance test with AAS
AAS characterization was used to investigate metal compounds contained in untreated water and treated using the CMC material nanoadsorbent. Quality tests that have been performed using the CMC material nanoadsorbent to investigate the efficiency value of the CMC material nanoadsorbent towards the recovery/remediation of polluted water were calculated and analyzed (Yu et al., 2021;Feng et al., 2021).
Determination of Adsorption Contact Time and Adsorption Model for heavy metals Fe, Mn, Zn and Pb -the removal of heavy metals was initially carried out using the batch method. The adsorption of heavy metals Fe, Mn, Zn and Pb (1,351; 1,210; 17,403 and 0.17 mg/L) by adsorbents was carried out simultaneously in 50 mL polluted water at neutral pH 7 and at room temperature (27 °C). The observed adsorption contact times were 30, 60, 90, 120, 150 and 180 min with a stirring speed of 150 rpm. From the process of determining the optimum contact time, it is also possible to analyze the adsorption mechanism. Determination of heavy metal content was made using AAS. Determination of the adsorption isothermic model was performed using the Langmuir and BET models (Annadurai et al., 2018; Nithya & Sudha, 2016). Effect of pH on the adsorption of heavy metals Fe, Mn, Zn and Pb -the adsorbent was placed into a glass beaker then 50 mL of water contaminated with heavy metals Fe, Mn, Zn and Pb was added (initial concentrations of each metal in pol-

Adsorption process and AAS
In the adsorption process, AAS characterization was used to investigate metal compounds contained in water that had not been treated and had been treated by nanoadsorbent material CMC. Quality tests that have been performed using nanoadsorbent material CMC to investigate the % adsorption of nanoadsorbent material CMC on the recovery/remediation of polluted water were calculated and analyzed with the parameters of contact time, acidity level, stirring speed, adsorption model and adsorption kinetics. The % adsorption for the seven samples with each of the heavy metal ions Fe, Mn, Zn and Pb are shown in Figure 1   ions it is S4>S3>S1>S5>S2>S7>S6, then for the Pb ions it is S4>S7>S1>S3>S2>S6>S5, so it can be concluded that of the 7 samples of CMC nanoadsorbents, the most effective sample was S4 for each adsorbed heavy metal ion, followed by S3 for Fe, Mn and Zn ions, whereas for Pb ions it was S7. This is because it has the highest % S4 adsorption compared to other samples. Sample S4 is the easiest to experience the phenomenon of diffusion in the heavy metal ion adsorption mechanism of Fe, Mn, Zn and Pb. To ensure the performance of the CMC nanoadsorbent in removing the four heavy metals, it is necessary to compare it with the removal capability by magnetic Fe 3 O 4 alone and banana peel activated carbon separately which has been carried out in the previous procedure. Determination of the Langmuir adsorption model of the four heavy metals Fe, Mn, Zn and Pb can be seen in Figure 5 and 6.
The calculation results are shown in the table that the metal ions Fe, Pb, Zn and Mn have a positive increase and tend to follow the Langmuir adsorption model. From Figure 5. and 6 Langmuir adsorption for Fe, Mn, Zn and Pb ions showed a curved graph, this curvature as evidence of surface heterogeneity of the bonding sites of CMC nanoadsorbents. The heterogeneity in the graph models shows the curvature of the isothermal graph as a gentle slope indicating a fairly high and constant adsorption affinity. The maximum adsorption capacity using the Langmuir method is for heavy metal ions Mn 5162.4 mg/g, heavy metal ions Fe 5849.4 mg/g, metal ions Zn 4894.22 mg/g and heavy metal ions Pb 468.2 mg/g. It can be seen from the order of adsorption capacity that Pb has the smallest capacity; this is proportional to the small % adsorption value so that it affects the weakest bond strength with CMC nanoadsorbents when compared to other heavy metal ions, such as Mn, Zn and Fe. Theoretically, the bond interaction between Pb ions and atoms of lone pair donor atoms, namely N atoms and O atoms, has the smallest ionic character or the largest covalent character, because it has the largest atomic radius compared to other heavy metal ions (Shen et    varies with pseudo-first order and pseudo-second order; this is due to the four heavy metal ions, the intensity of the four heavy metals being strong enough to interact with the CMC nanoadsorbent sample, in this case sample 4 of the four heavy metals, Pb has the lowest relative mobility compared to the other three heavy metal ions, Mn, Zn and Fe. Likewise, other heavy metal ions which have higher relative mobility tend to interact with higher binding affinity sites (O and N active groups) than interacting with lower binding affinity sites (Dave & Chopda, 2014; Wojciechowska, 2020). On the basis of the research, the lower the mobility of the heavy metal ion Pb, the longer the settling time at the two bonding sites when compared to other heavy metal ions Mn, Zn and Fe. In addition, the research reports successively from the lowest to the highest is Pb>Mn>Zn>Fe.

XRD
In the discussion of XRD nanoadsorbent CMC shown combined for 7 seven samples namely sample 1, sample 2, sample 3, sample 4, sample 5, sample 6 and sample 7. The results of the analysis with combined XRD are shown in Figure where almost the seven samples have a phase pattern the same peak/similar in that the structure is crystalline. The compounds formed are Fe 3 O 4 and Fe 3 C with different percentages for each sample.
The Debye-Scherrer scattering equation enables to estimate the mean crystal size of CMC nanoadsorbents, both S1, S2, S3, S4, S5, S6 and S7. The results of the crystal size analysis are shown in Table 2. It can be seen in Table 2 that all samples have a size towards the nanometer order and S4 has a smaller crystal size which is about 24.5 nm compared to other samples. It can be concluded that the crystal size affects the % adsorption value of heavy metal ions, where a small size will increase the % adsorption and the order from the largest to the smallest is S1>S2>S5>S7>S6>S3>S4. On the basis of these results, it seems that the Fe 3 O 4 phase is the dominant phase in this sample S1, with the identification at a diffraction angle of 2θ 18,   for Fe 3 C as much as 18.3%. On the basis of these results, it seems that the Fe 3 O 4 phase is also the more dominant phase in the S3 sample compared to the S1 sample, but not more than the S2 sample with identification at a diffraction angle of 2θ It is the same with the previous S1, S2 and S3 samples where the pure magnetic phase is more abundant than the mixed carbon phase. However, the S4 sample was special, because at the time of measurement with AAS it also had a better performance than other samples in remediating polluted water with the highest adsorption % indication. In the discussion of the diffraction pattern in sample S4 also the percentage of Fe 3 C is greater than the other samples, namely the diffraction angle of 2θ is the   can be said that sample S4 has a smaller Fe 3 O 4 phase than samples S1,S2,S3, S5, S6 and S7 and the Fe 3 C phase is larger than the other samples It can be judged that sample S4 is the optimal sample with a ratio of 1:2:2. The phases formed in the XRD results strengthen the hypothesis that the CMC nanoadsorbent incorporation method is suitable for synthesizing these three materials (Amin et al., 2014;Gutierrezet al., 2017). For this reason, the S4 sample was carried out advanced characterization, namely SEM/EDX, FTIR, PSA, TGA, and BET.

SEM/EDX
The surface morphology and microstructure of the S4 sample synthesized by combining three materials, namely chitosan, magnetic and banana peel activated carbon (1:2:2) were observed using a scanning microscope (SEM) as shown in Figure  8. On the basis if on the SEM results, it can be seen that the microstructure of the material resembles islands and small grains around them; it is predicted using software that the islands are dominated by activated carbon of banana peels combined with Fe, namely Fe 3 C and the surrounding small grains are Fe 3 O 4 . There is a lot of agglomeration on the surface of the sample (Lin et al., 2020;Chibowski, 2018). From the previous analysis, banana peel activated carbon is a porous material, so it can be seen that the pores are filled with magnetic Fe 3 O 4 . Sample S4 was characterized by SEM having an average particle size in the range of 2-10 μm. Magnification was performed on the SEM 2000× tool. To see the elemental content in the S4 sample, further testing was carried out on SEM with EDX, the results are as shown in Figure 9. The EDX results to determine the elemental content of the S4 nanoadsorbent CMC sample, Fe, C and O elements were obtained.
The Fe and O elements came from Fe 3 O 4 magnetic material synthesized by the coprecipitation method, while the C element came from banana peel activated carbon synthesized by the combination method (Lesaoana et al., 2019;Blue et al., 2021). The percentage of elements from the S4 nanoadsorbent CMC sample as seen in Figure 9 is Fe elements as much as 57.25%, C elements as much as 26.50% and O elements as much as 16.25%. This shows that the S4 sample has elemental purity without any other elemental impurities. This shows that the glutaraldehyde material has the ability to bind strongly to the three materials, as it is known that the material has crosslinking properties. Another active group is at 1050.11; 908.34; 796.81; 598.16 cm -1 which is C-C and C-O in the form of bending vibrations (buckling). From the description, it can be seen that the combined synthesis of the three CMC materials produces hydroxyl, carboxyl and aldehyde active functional groups.

PSA
The results of characterization and calculations using scattering technique software using PSA obtained the average particle size distribution of CMC nanoadsorbents were (10%) 40 nm, (50%) 300 nm, and (90%) 400 nm. Then the average particle size of the CMC nanoadsorbent is around 150.4 nm which is included in the fine particle category. For this reason, this sample can be said to be nanoparticle size (Danish et al., 2018; Dave & Chopda, 2014). The particle size distribution can be seen in Figure 11.

BET-BJH
The results of the BET (Brunaeur Emmet Teller) and BJH (Barret Joyner Halende) tests showed that the specific surface area was 450.8 m²/g. BJH D-H Adsorption cumulative volume of pores between 1.7000 nm and 300,0000 nm diameter is 0.202 m 3 /g. BJH D-H Desorption cumulative surface area of pores between 1,7000 nm and 300,0000 nm diameter is 0.411 m 3 /g, while the pore diameter is 2.13 nm. The particle size of the sample was 13.312 nm. On the surface of the S4 nanoadsorbent Figure 9. EDX results of S4 CMC nanoadsorbent samples CMC sample, there was a very drastic increase in nitrogen gas adsorption and was known as a hysterical loop which stated the characteristics of the mesoporous diameter. On the basis on the measurement results, this sample is classified as mesoporous (2-50 nm). The area of the S4 nanoadsorbent CMC sample is larger than that of the banana peel activated carbon sample (Chibowski, 2018; Lin & Jiang, 2019; Shukla et al., 2020). In theory, the increase in surface area will affect the amount of adsorbate adsorbed, in this case the four heavy metal ions, i.e. Fe, Pb, Mn and Zn. Nanoadsorbents have strength because of their high adsorption activity on heavy metal ions, judging from the ability of oxygen, hydrogen and carbon sites as free electron donors resulting in strong heavy metal ion bonds on the surface of the S4 nanoadsorbent CMC sample. In the characterization of BET/BJH obtained isothermically using the Freundlich method Qm C: 0.0331 ± 0.0013 cm³/g STP m: 3.6407 ± 0.2663 and Temkin q·alpha/Qm: 22.803921 ± 0.655982 kJ/ mol·(cm³/g STP) A: 0.7216 ± 0.0960 mmHg.

TGA
The following is a TGA graph that states the relationship between sample temperature and % TG (weight) of the S4 nanoadsorbent CMC sample. From the TGA graph in Figure 12 it can be seen that the S4 nanoadsorbent CMC sample has a fairly high resistance to the decomposition process, because the weight loss is still around 30% of 100% TG sample. The temperature used during the test is 25-500 °C. The heating rate used is 10° min using oxygen gas (Matome et al., 2020;Drbohlavova et al., 2010). The resistance of this S4 sample to the decomposition process indicates that there is an interaction through intermolecular hydrogen bonds between the active groups of the S4 nanoadsorbent CMC sample mixture consisting of chitosan, magnets and banana peels activated carbon (Koesnarpadi et al., 2015).

CONCLUSIONS
In accordance with the research objectives, from the results of the study it was concluded that CMC nanoadsorbent can optimally remediate water contaminated with heavy metals Fe, Mn, Zn and Pb when compared to magnetic adsorbents and banana peel activated carbon alone. The increase occurred when both were combined and chitosan was added. The optimum adsorption time for each heavy metal ion, i.e. Fe, Mn, Zn and Pb, is 210 min. The most effective sample is sample S4 for each adsorbed heavy metal ion with the highest adsorption of 95.44% for Fe. The Langmuir model for Fe, Mn, Zn and Pb ions shows a curved graph, this curvature as evidence of surface heterogeneity of the bonding sites of CMC nanoadsorbents. The maximum adsorption capacity using the Langmuir method is for the highest heavy metal ion and Fe 5849.4 mg/g. The adsorption kinetics showed that the reaction order for Pb, Mn, Zn, and Fe ions varied with pseudo first order and pseudo second order. On the basis of the results of the characterization of nanoadsorbent CMC sample S4 which has a crystal size specification of about 24 nm, particle size of 150.4 nm, pore size of 2 nm, specific surface area of 450.8 m²/g. %, element C as much as 26.50% and element O as much as 16.25 %, the FTIR spectrum confirmed that the sample produced active functional groups of hydroxyl, carboxyl and aldehyde and had a fairly high resistance to the decomposition process.