Remediation of contaminated water with Chromium VI by sorption in surface-activated-nanocellulose


 Chromium VI is a frequent pollutant of industrial liquid effluents resulting from industrial waste. It is a substance classified into the carcinogen group I. In this study, a Cr VI sorption mechanism was developed by using nanocellulose beads (hydrogel) obtained from ivory nut. Cr VI was detected in water by a colorimetric method, using 1, 5-diphenylcarbazide at λ 540 nm. The sorption capacity of nanocellulose spheres was tested by varying the solution's pH and temperatures. Our results showed that nanocellulose can efficiently adsorb at pH 4 and 25°C. Removal percentages were between 91.29%(+/-1.36) and 95.33%(+/- 0.86) of the total Cr VI. The sorption kinetics showed that the material reaches equilibrium after 20–30 minutes. Additionally, an analysis of adsorption isotherms showed a high fit with the Langmuir and Freundlinch isotherms. Therefore, this method seem to be applied both to decontaminate industrial and drinking water employing an organic matrix such as nanocelullose beads.

There are multiple techniques to improve water quality when it is contaminated with Cr VI. Some of those are oxidation-reduction (Li et al. 2017); phytoremediation (Rezania et al. 2015); ion exchange (Fu and Wang;2011); precipitation (Cainglet et al. 2020); capture by biopolymers (Gupta and Diwan. 2017); adsorption (Fu and Wang. 2011); membrane ltration (Owlad et al. 2009); biosorption (Abdolali et al. 2017); and capture by nanoparticles (Qu, Alvarez, and Li. 2013). Furthermore, among the most promising techniques are those that use nanoparticles to remove Cr VI from water (Bhattacharya et al. 2013); nanoparticles measure less than 100 nm in at least one of their three dimensions (Klemm et al. 2018).
The advantage of using them is the large surface area they present; many contain high reactivity and sorption capacity (Klemm et al. 2018).
Nanosorbents, such as modi ed membranes, nanophytocatalytic particles, and magnetic nanoparticles, are the most common sorption agents (Qu et al. 2013);biomass is one of the most common sources for obtaining nanosorbents due to the comparative advantages over other sorbents (Singh et al. 2020). This advantages includes low cost, high e ciency, regeneration capacity, and high availability (Dorishetty et al. 2020).
Besides, Nanocellulose bershas properties that stand out for many applicationsits extraction from lignocellulosic biomass, especially from agricultural residues, has been extensively studied (Tshikovhi et al. 2020); in those cases, non-cellulosic materials such as lignin and hemicellulose are removed by a pretreatment (Phanthong et al. 2018). The separation of lignin and other molecules combined with cellulose in lignocellulosic materials requires intensive energy input and, frequently, the use of polluting chemicals (Portero et al. 2020).
NC retains several cellulose properties, such as its hydrophobicity, crystallinity, and ability to be modi ed (Klemm et al. 2018). Three types of these particles can be found: nanocellulose crystals (CNC), cellulose nano brils (CNF), and bacterial nano cellulose (BNC). They differ from each other in their origin and in the obtainment methods, functionalization, shape, particle size, size dispersity, and mechanical characteristics (De France et al. 2017).
In this framework, the use of non-lignocellulosic feedstocks may facilitate NC's extraction (Haa z et al. 2013). To illustrate, in the western forests of Ecuador, there are palms of the species Phytelephas aequatorialis. Its seeds (tagua) are used mainly to make buttons for the fashion industry (Montufar, 2013). This process produces about65-88% of tagua waste, which currently represents about 4,800 t / year (Valencia et al, 2013). In this research, it has been demonstratedthat high purity nanocellulose can be obtained from the endosperm of Phytelephas aequatorialis seeds.. This material is suitable for metal sorption (Mautner et al. 2019).
Moreover, NC has been widely used for its mechanical and chemical properties. The hydroxyl groups present on its surface allow a relatively easy chemical modi cation (Oun et al. 2019). The modi cations lead to an increase in the range of applications of these nanoparticles. Besides, the most critical modi cation employed for metals or their metallic salts capture on NC, is the esteri cation with positive or negative groups to the primary and secondary hydroxyl groups of glucose (Gan et al. 2020). Modi ed NC can bind charged molecules via ionic linkages.
This study reports the use of tagua NC beads for the remediation of Cr VI contaminated water. Through the use of specially tailored hydrogel beads masked with quaternary ammonium. It is proved to be an upand-coming technique insofar as it has been possible to remove up to between 91.29-95.33% of Cr VI in an aqueous solution.
2 Materials And Methods 2.1 Acid extraction of nanocellulose.
The nanocellulose bers were prepared through hydrolysis using sulfuric acid. Then, 20 g of tagua pulp was chopped in a blender. Next, 175 mL of sulfuric acid was added at a concentration of 64% w/w preheated to 45 ° C. The reaction continued for 60 minutes at 45 ° C. Subsequently, the solution was dissolved at 10% of the initial Concentration to stop the reaction. Finally, it was centrifuged and washed using dialysis membranes until the pH was neutral (Menon et al. 2017).

Atomic force microscopy of the obtained material (AFM)
The shape and approximate size of the NC particles were measured by using a Brucker -model dimension icon AFM. This allowed to characterize nanoparticles in width, depth, and length. A series of dilution factors were tested to obtain individual ber nanoparticles. The most appropriate was the dilution of 2 x 10 -3 .

Nanocellulose hydrogel spheri cation
A solution was prepared with 1.46 g of quaternary ammonium in one liter of water (Bingol et al. 2004). The sol-to-gel transition of the aqueous solution of nanocellulose (1% w/v), was attained by dripping 10 mL of nanocellulose onto 25 mL of a quaternary ammonium solution preheated at different temperatures (50, 60, 70 and 80 ° C).

Colorimetric method for detection of Cr VI
First, 1.5 g of 1,5-diphenylcarbazide (complexing agent) were dissolved in 50 mL of acetone (Huang et al. 2020).Then, a mix of 10 µL of H 2 SO 4 50% v/v solution, and 20 µL of the complexing agent were dissolved in 1 mL of a solution with K 2 Cr 2 O 7 . Finally, a spectrophotometer (Helyos β, Thermospectronic) was used to detect the presence of Cr VI at 540 nm (Mohamad et al. 2013).

Preparation of Cr (VI) solutions and calibration curve tailoring
A stock solution (1,000 mg Cr(VI)/L) was prepared by dissolving 2.829 g of K 2 Cr 2 O 7 (CAS 7778-50-9; ≥99.0%; Sigma-Aldrich) in 1 L of deionized distilled water.For nanocellulose beads sorption experiments, diluted solutions were prepared with concentrations ranging from 10 to 100 mg Cr(VI)/L (Campaña, 2019). Subsequently, a calibration curve type was tailored by using concentrations within the range 0.01 to 1 mg/L, to obtain a linear equation. This equation was used to calculate the Concentration of Cr VI in water.

In uence of pH and temperature in nanocellulose sorption of Cr VI
A series of Cr VI solutions of 10 and 100 mg/L were adjusted at pH 4, 6 and, 8, respectively. The solutions were tested at temperatures of 15, 20, and 25 (° C). Nanocellulose beads were added to each Cr VI solution in a ratio of 1:2. Then, they were stirred for 3 hours at the temperatures mentioned above. The experiments were performed by triplicate.

Metal removal calculation
The Concentration of the solution exposed to the NC spheres was calculated. The percentage of metal removal (% MR) was obtained with equation 1: Where: Ci is the initial Concentration, and Ce is the nal Concentration of Cr VI (Zhu et al. 2014 2.8 Statistical analysis. An angular transformation was made on the % MR in order to normalize the data expressed in terms of percentage for statistical analysis. Consequently, ANOVA calculated the signi cance, and a Tukey test was established to determine the optimal Cr VI sorption conditions (Lee & Lee, 2018).

Cr VI adsorption kinetics
Once optimum pH and temperature were established, a series of solutions of Cr VI at concentrations 10, 25, 50, 75, 100 (mg/L), were added to nanocellulose gel beads at a 2:1 volume rate. The tests were carried out by triplicate at 10, 15, 20, 30, 90, 120, and 180 minutes.
Subsequently, the % MR and the Concentration of Cr VI removed were plotted as a function of time. Finally, the Concentration of metal adsorbed on nanocellulose beads was evaluated with equation 2: Where: q e is the amount of metal adsorbed at equilibrium (mg/g); c i is the initial and equilibrium Concentration (mg/L); c e is the equilibrium concentration (mg/L); v is the volume of solution (L); and w is the amount of biosorbent (g) (Zhu et al., 2014).

Analysis of adsorption isotherms:
Two isotherms were evaluated, the Langmuir's and Freundlich's isotherms. Equations 3 and 4 were used for this purpose: Where is the amount of metal adsorbed in equilibrium (mg/g); is the maximum amount of adsorbate adsorbed (mg/g); is the final Concentration of the metal (mg/L); is the Langmuir constant (L/mg); is the affinity constant between adsorbate and adsorbent; and k f is the Freundlich constant (Yao et al. 2016).

11 Column ltration of contaminated water with Cr VI
A column of 26 cm long, 1 cm diameter and 20 mL capacity was packed with NC beads. Then it was lled with Cr VI solutions. After that, the samples were injected under pressure. Finally, the experiments were carried out with Cr VI solution concentrations of 10, 100 mg/L, 25°C, for 20 and 30 minutes.
folding that generates angles. Thus, this characteristic contributes to the entanglement of gels (Chaabouni & Bou , 2017).

Nanocellulose Spheri cation
The optimum formation of gel beads (in terms of its consistency and shape) was attained at 70°C. The spheres were formed with an average diameter of 2 mm. Besides, heat enhances the interaction between cationic surfactants and negative charges of nanocellulose. Consequently, the resultant beads' consistency and shape are obtained through the functional groups' interaction (Bora and Dutta. 2014).
As far, the nanocellulose produced for this work was obtained from the fractionation of the endosperm of tagua (Phytelephas aequatorialis). It is an acid-hydrolyzed treatment with sulfuric acid. As a product it acquires a surface charge, given by the esteri cation of OH groups in glucose with SO 3 groups.
According to previous analyzes (unpublished data), it is known that this nanocellulose contains 0.128 mmol-SO 3 -/ g. As well,the concentration of SO 3 groups esteri ed to OH groups in glucose, depends on the process's acid hydrolysis conditions. For instance, Harris & McNeil (2020) found 0 -1.9 mmol SO 3 -/ g.
On the other hand, quaternary ammonium is a cationic surfactant that has been used on clays and zeolites since it improves the a nity with non-ionic and hydrophobic organic compounds (Li & Bowman. 1998). It has also been used to modify the surface of nanocellulose (He et al. 2014). For example; Li et al. (2018) used quaternary ammonium to promote antibacterial activity in nanocellulose-based materials (Chen et al. (2017) used ammonium-modi ed nanocellulose to degrade methyl orange dye.
Moreover, quaternary ammonium is a cation with an NR 4 + ; where R represents the alkyl/aryl group structure (Gerba. 2015) that can easily interact with the SO 3 functional group of nanocellulose brils.
When nanocellulose was dripped on dissolved quaternary ammonium, crosslinking occurred with additional surface charge change. This charged surface interacts with Cr VI in aqueous solution (Li et al. 2018).

Cr VI sorption based on pH and temperature
The ANOVA did not show statistically signi cant differences for the temperature treatments. The pH is the only factor that affects the % MR signi cantly. Additionally, Tukey's test showed the higher Cr VI removal at pH 4.0.
In several studies, the in uence of pH for the sorption of Cr VI was evaluated. These have shown that this factor (pH) has a signi cant impact on the percentage of removal (Tovar et al, 2014). It is also known that Cr VI can be present in many forms such as H 2 CrO 4 , HCrO 4 -, CrO 3 -2 and Cr 2 O 7 -2 ;depending on pH and its Concentration (Jiang et al, 2014). Thus, when the pH is between 1 and 4, the predominant form is HCrO 4 . It interacts with the monovalent anion NR 4 + . Next, when the pH increases, the most common forms of Cr VI are Cr 2 O 7 -2 and Cr 2 O 4 -2 . So that two anions of NR 4 + interact with those forms of Cr VI.
Consequently, with a higher pH the MR % decreases (Karthikeyan et al, 2005).
The results of this research coincide with those obtained in a study carried out by He et al. (2014) which reported that the highest Cr VI sorption values were between pH 3 to 4. Moreover, an acidic pH is favorable because of protons' (H + ) increase on the nanocellulose surface. Giving rise to a strong electrostatic attraction between the positively charged surface and the chromate ions (Fijałkowska et al. 2020).
According to the ANOVA results, the temperature variation in the experiments did not show a statistical signi cance. Nevertheless, a slight enhancement of Cr VI sorption at higher temperatures would be associated with the reaction's endothermic nature. This phenomenon can be explained by the increase of the intraparticular diffusion rate of Cr VI ions, into the nanocellulose beads' pores ).

Adsorption kinetics
The nal Concentration of Cr VI decreases until the matrix is saturated. In Table 1, it can be observed that the Concentration decreases progressively until it reaches 20 to 30 minutes.
Since the % MR increased as time passed; all the samples were exposed for 180 min (3 h). However, it can be seen that the majority of the samples reached their maximum % MR much earlier (Table 2). While in the study done by He et al. (2014) it took around 50 minutes to obtain an e ciency of almost 100%. In the research of Jiang et al (2014), sorption equilibrium was obtained at 120 min. This because HA-Fe 3 O 4 was used to adsorb Cr VI; resulting in a % MR of 80 to 90%. On the other hand, Zhu et al. (2014) studie's obtained a 80% e ciency in 10 minutes.

Analysis of adsorption isotherms
Water remediation must include an isotherm analysis to determine its effectiveness (Lombardo and Thielemans 2019). Additionally, it is essential to establish an equal saturation time to increase its productivity.
Two adsorption isotherms were evaluated: Langmuir's and Freundlich's. For the Langmuir isotherm, a linear graph was constructed as a function of the nal metal concentration (mg/L) (ce); and the amount of metal adsorbed at equilibrium (mg/g) (qe). As a result, an R 2 of 0.979 was obtained out of the linear regression.
For the Freundlich isotherm, another linear regression was obtained, whose R 2 was 0.966. In addition, the values of the heterogeneity factor (n) were calculated. As well as the Freundlich constant (k f ) which is associated with the adsorption capacity. The values obtained were 0.8917 and 1.31025 [(mg / g) -1 /n] respectively.
The results obtained were adjusted to Langmuir's isotherm, meaning that the surface of the adsorbent and all the sorption sites are homogeneously distributed. Besides, each cation can interact with a single molecule of Cr VI, forming a monolayer as a consequence (Giles et al.1974). However, results also show a high adjustment to the Freundlich isotherm, which means that multilayer adsorption also occurs (Yao, 2016).
The adsorption method showed very high e ciency compared to other published works. A comparison is shown in Table 3.

Column ltration
In practical terms, the water treatments are performed in columns that save space; improve the surface contact with the water pollutants; and allows cleaning and or replacement of the adsorbent matrix (Leupin et al, 2005).
The present research, performed adsorption experiments in columns (data not shown). Which had a very similar result when compared to the experiments in beakers. Columns ensure the distribution of the contaminated bed homogeneously in the adsorbent (Ali. 2014). A notable advantage of columns is that the NC matrix can undergo a desorption process after saturation. So they can be reutilized (Figueira et al. 2004).

Conclusions
To summarise, the use of a nanocellulose hydrogel as a biosorbent to remedate water contaminated with Cr VI was shown to be an e cient method (% removal of the metal between 91.29-95.33%). The pH produced an increase in the sorption capacity of this material. On the other hand, in the sorption kinetics, a sorption equilibrium was obtained at 20-30 min. Which depends on the metal's initial Concentration.
This experiment tted the Langmuir adsorption isotherm better than the Freundlich isotherm. High a nity was obtained between the adsorbent and the adsorbate. Also, a high sorption capacity compared to other adsorbents used in similar research.
Finally, the column's use to capture Cr VI showed an e ciency similar to the tests carried out previously. Therefore, nanocellulose beads masked with quaternary ammonium (a porous hydrogel matrix masked by a strong cation) represent a suitable combination to remediate water contaminated with Cr VI. Lastly, nanocellulose beads can also be reutilized after regeneration cycles.

Declarations
Funding: This study was supported by the Ponti cia Universidad Católica del Ecuador.
Con icts of interest: The authors declare there is no con icts of interest in this study.

Figure 1
An image of NC bers in AFM 3.3 um bar scale