Chromium removal from contaminated waters using nanomaterials – A review

Abstract Current environmental policies aim to reduce the levels of toxic substances in aquatic ecosystems and to promote the water reuse after appropriate treatment of wastewater. Chromium is a hazard element present in effluents of various industries that should be reduced to achieve the objectives of those policies. Most of the results reported in the literature concern the use of nanomaterials for chromium sorption dissolved either in synthetic or mono-elemental spiked solutions. The present work reviews the results of research undertaken in the last decade on the application of various nanomaterials to decrease chromium concentrations in contaminated waters. Major factors influencing the removal efficiency were examined. Because most of the published studies are based on simple experiments with deionized water and mono contamination further studies are suggested focused on effects of natural and artificial chelators, interferences of other trace elements competing with chromium sorption, and reduction of sorbent mass per water volume.


Introduction
Present life style requires the exploitation of Earth's resources beyond their sustainability causing the reduction or depletion of limited resources [1]. Environmental issues started with the Industrial Revolution, the discharge of industrial effluents, either inadequately treated or untreated, into aquatic systems lead to the increase of hazardous inorganic and organic contaminants in rivers, lakes, estuaries and coastal areas [2]. Because of the nondegradation character of many contaminants, they are transfer to the food chains with impact on the ecosystem services and reducing the marine food safety [3,4]. Volume of M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 2 dumped debris in water systems increased often surpassing the self-cleaning capacity and purification of aquatic systems. It is foreseen these discharges will increase in the future as population tends to migrate and concentrate in urban areas, as response to modern life and adversities related to climate changes.
Chromium is among the most toxic trace elements released to surface waters and ground waters due to its widespread use in industrial applications, such as leather tanning, metallurgy, electroplating and refractory [3]. The increasing number of articles published about chromium toxicity over the last 10 years [3] indicates the efforts to illustrate and remediate the chromium-bearing contamination. Trace elements can be removed from wastewaters by conventional methods, such as chemical precipitation, ion exchange, membrane filtration, coagulation/flocculation and electrochemical treatment [5]. However, these methods have low efficiency and produce large volumes of wastes. Alternatives for the treatment of water contaminated by metals are sorption methods [6]. Sorption corresponds to the transfer of the sorbate from the liquid phase to the surface of the sorbent. Sorption efficiency is influenced by various factors, such as pH, temperature, nature and amount of sorbent, initial metal concentration, ionic strength, and the presence of other contaminants [7][8][9]. Depending on the attractive forces between the sorbent and the sorbate, this becomes bound by physical (physiosorption) and/or chemical (chemisorption) interactions [6]. While in the physiosorption the sorbate bonds to the sorbent surface by weak forces, such as Van der Waals interactions, which is a reversible process, the chemisorption is frequently irreversible due to the presence of strong chemical bonds between the sorbent and the sorbate.
A large variety of sorbents are available to remove trace elements from waters [5], including nanomaterials with various types of coatings and chemical functionalizations [4,10]. Nanomaterials, i.e., materials and structures with at least one dimension of 1-100 M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 3 nm [11], exhibit unique mechanical, optical, magnetic and chemical properties highly depended on shape, size, surface characteristics and inner structure that differ from the characteristics of particles and macroscopic surfaces of similar composition [12][13][14].
Sorption mechanisms by a nanomaterial sorbent are also a function of the sorbent characteristics and physical-chemical conditions of the solution where the sorbent is removed. Nanomaterials should satisfy some criteria to be used as sorbents for toxic elements removal from wastewater [15]: nontoxic; high sorption capacities; selectivity to the low concentration of contaminants; easy removal of the sorbed contaminant from the surface of the nanomaterial; recycled. Until present, a variety of nanomaterials such as carbon nanotubes, carbon based material composites, graphene, nano metal or metal oxides, and polymeric sorbents fulfil those criteria and have been studied in the removal of toxic trace elements from aqueous solutions [15].
The coupling of sorption ability and magnetic properties in certain nanomaterials have also been explored envisaging a new class of nanosorbents [16,17]. Magnetic nanosorbents offer the great advantage of allowing fast recovery by employing magnetic separation technologies. A number of nanosorbents comprising magnetite nanoparticles have been reported by our laboratories, which include core/shell nanoparticles for the removal of heavy metal ions [18] and magnetic bionanocomposites for the removal of organic pollutants [19]. The successful implementation of magnetic nanosorbents depends, among other factors, on their efficiency for the selective uptake of pollutants, which requires further developments concerning the type of surface chemistry involved. The intensive use of nanomaterials may have some environmental risks and impacts on human health [13]. It is hence crucial to evaluation the nanoparticles toxicity, which depends on their aggregation, agglomeration, dispensability, size, solubility, surface area, surface charge and surface chemistry [20].

M A N U S C R I P T
A C C E P T E D ACCEPTED MANUSCRIPT 7 putrescible hide or skin is converted into leather. The permanent stabilization of the skin matrix against biodegradation is possible using basic chromium sulphate [32]. Although this industry is not critical in Europe, it has a high impact in Asia, Africa, and South America (Public Partnership for Better Innovation Policies and Instruments in Support of Eco-Innovation: ECOPOL, 2013). For example, leather tannery industry in China is responsible for 20% of chromium discharges into water, the average total amount between 1990 and 2009 reaching more than 0.5 thousand tons per year [34].  Figure 3 compares the contribution of various industrial activities on the emission of chromium in 2014 and 2015. The sector "Production and processing of metals" accounts for more than 60% of the chromium emission into the water.

Material and Methodology
Numerous studies have been published on chromium sorption in aqueous phase using various materials and in particular synthetic nanomaterials [9,[35][36][37][38][39]. To select the articles For each selected article, it was extracted the information related to the parameters considered relevant in sorption [7][8][9]: name and nature of the sorbent, mass of the sorbent with respect to the water volume, type of water used in the experiment, type of experiment (single or other contaminants besides chromium), pH, temperature, contact time between the sorbent and the solution, initial concentration of chromium, chromium species initially present in solution, and removal efficiency. Table 2 lists the synthetic nanomaterials and the experimental conditions employed in the studies of the selected articles from the literature. In order to encompass the collected information in a single Table, intervals of values are presented for the uptake capacity or removal efficiency of each nanomaterial or group of nanomaterials, as well as for the relevant parameters aforementioned.

Type of materials.
Among the various materials used for chromium removal, nanoparticles have been the most common, either using just the core nanoparticles [9,[40][41][42][43][44][45][46], nanoparticles with functionalization [35,[47][48][49], or modified nanoparticles incorporated on substrates [36,50]. Other type of materials have been used, such as nanocomposites [37,51,52], nanofibers [38] and carbon nanotubes [39]. In present review, nanomaterials like zero-valent iorn nanoparticels (nZVI) were not found. In the first step towards the use of this material the toxic Cr(VI) is reduced to the less hazardous Cr(III), which is then removed by sorption to the nZVI surface and precipitation by iron-hydroxides [53,54]. Table 2 describe the chromium sorption experiments using mono-elemental systems, Cr being the only M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 9 contaminant to be treated. Only a few studies tested both mono-elemental and multielemental systems. Distilled or milli-Q water have been considered. Absence of competitive ions or other contaminants are simplistic approaches to the complex conditions existing in aquatic systems or contaminated waters. The lack of chromium sorption experiments using multi-elemental systems, where Cr was not the only contaminant to be treated, is a weakness in this kind of research.

Type of solutions.
Only a few studies addressed the treatment of contaminated waters as real samples, such as groundwater, effluents or wastewater [41,46,48,55,56]. Chen et al. [57] have simulated natural waters by testing solutions of different complexity, deionized distilled water, tap water, mountain stream water and river water. Although the absence of competitive ions or other contaminants be the most common approach in this kind of research, the study of nanomaterials behaviour in natural waters is crucial before the material be implemented in the market. Real waters have varied and complex composition; thus, some researchers try to simulate the reality through the dissolution of salts that put in the waters the ions found in natural systems.
Contact time. In general, contact time between the nanomaterial and the contaminant were less than 2 days, although data exist for 3 days [55], 7 days [68] and 15 days [56]. Removal experiments for industrial application should be performed during a contact period between the nanomaterial and the contaminant less than 48 hours to be feasible for the industry. This is because very long sorption processes imply the existence of industrial tanks that are inactive for a long time. On the other hand, treated effluents must be discharged quickly without endangering the life of aquatic organisms that are exposed to these effluents.

Amount of sorbent.
It is well documented that, for the same Cr concentration, the rate of sorption increases with the amount of sorbent. However, the larger amount of material used should be avoided because it will generate greater amounts of residues to be treated increasing the cost of process.  . The maximum allowed concentration of total chromium in residual waters is 2 000 µg/L, meaning the studies that used higher concentrations are unrealistic. In this way, some studies [9,40,78,43,50,52,[68][69][70]76,77] run their experiments with lower concentrations, taking into consideration the allowed limits of chromium in water. Among these, Chowdhury and Yanful [76], Simeonidis et al. [70] and Gifford et al. [68] were the only ones that studied concentration equal or <2000 µg/L and Kaprara et al. [40] studied the removal of 10 µg/L of chromium using Sn(II) oxy-hydroxides nanoparticles.
Chromium speciation. Cr(VI) have been the most investigated species, although some researchers have run experiments with Cr(III) forms [38,47,48,51,71]. Others studies used with both Cr(III) and Cr(VI) [9,42,69,77,78]. Only a few studies mention the analytical methodologies to discriminate the quantification of Cr(III) and Cr(VI) during the removal process. Most of the methodologies referred in the works of Table 2 are only able to measure total Cr and so most of the values reported for uptake capacity or removal efficiency are based on the initial and final concentrations regardless the starting chromium species. Using the materials mentioned in Table 2  [72], Srivastava et al. [37], Valle et al. [9] and Mahmoud et al. [42] propose that mechanisms for the binding of chromium(VI) is mainly physisorption. Despite those M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 12 studies suggesting physical interaction between Cr(VI) and sorbents, not all authors agree on the nature of the interaction during the sorption process. The presence of chemical bonds is suggested as a secondary mechanism [69], although Babaei et al. [36] suggest that sorption is governed by chemical forces rather than physical electrostatic interactions.
Even so, the removal of Cr(VI) is undoubtedly a sorption process. Regarding the reaction of Cr(III) with the different materials, the binding is through an ion or molecular exchange mechanism combined with some kind of physisorption [69,77]. Egodawatte et al [38] proposed that the binding mechanism between the nanofibers and Cr(III) involves the sorption of a positive complex on the surface of the materials. Arthy et al. [51] suggested a chemisorption mechanism. When no sorption mechanism is proposed in the articles, the interaction between Cr(III) and the materials is described as sorption process [48,71].
Best material performance. Lastly, magnetic iron oxide nanoparticles/sugarcane bagasse composite [51] and Cr(VI)-imprinted poly(HEMAH) nanoparticles [75] were the materials reported in the literature in the last years as being the ones with the most affinity for Cr (III) and Cr(VI) uptake, achieving a capacity of approximately 518 mg/g and 3 830 mg/g, respectively. However, maximum uptake capacity is a tricky parameter for the evaluation of a material efficiency. This parameter depends on the experimental conditions used, namely initial metal concentration and amount of nanomaterial used. Thus, the sorption performance of a material can not be assessed by considering only the sorption capacity value achieved.  Table 2 are far from being implemented in the market, with a maximum Technology Readiness Level (TRL) for industrial and socioeconomical aspect of 3.
Then, the published works are just a probe of concept, showing that there was a poor attempt to apply the nanomaterials to real samples and describing mainly its application in synthetic or mono-elemental spiked solutions. In this context, further laboratory removal essays are still required, never forgetting that these conditions must be realistic and adapted to the application. And after all laboratory tests are optimized using realistic experimental conditions, it is necessary to test the material in real effluents because the behaviour of a material may be very good in a synthetic contaminated water, but the same could not occur in the real system. For example, different industrial effluents have different composition and it is not possible to mimic all the real scenarios.
There are no reported successful case of applications to real effluents since the few studies that evaluated the potential of nanomaterials in real industrial effluents, either adding chromium to the samples of water or using a high amount of sorbent (economically unviable) and even performing the sorption experiments in very low volumes of water.
However, some important considerations to apply a nanomaterial in real industrial effluents are the following: • The treatment systems usually used are in batch.
• The effluents can need some kind of pre-treatment.
• Nanomaterials can be on-single use or a mixture of nanomaterials can be used. In the last case, it is necessary to address in which way the recovery process will be carried out; the ideal situation is the recovery of chromium in a way it can be further reused without any treatment -circular economy.  Table 2), but scarce information is available concerning their impacts towards aquatic organisms, in particular, no information is available on the potential toxicity of the Cr remediated water. It is important to test the ecotoxicity of treated water since recent works (not yet published) have shown that remediated waters can remain toxic for the aquatic organisms. It is also needed to access the environmental risk of the nanomaterials itself because of non-stability of some nanomaterials such as silver and gold nanoparticles, which can have impact on aquatic ecosystems due to toxicity of remaining material. The effect of interferences of other trace elements competing with chromium sorption should be studied. In order to minimize wastes that ultimately might result in the discharge of nanoparticles to the environment, a reduction of the mass of sorbent per water volume is envisaged. Furthermore, it is important to consider in future studies the potential toxic impacts derived from Cr remediated water. As additional concluding remarks, in order to identify the best conditions to test the efficacy of nanosorbents it is crucial to carried out a first set of experiments varying the amount of sorbent used, the solution pH, the chromium concentration and to monitor the concentration of chromium in solution with time. With these experiments it would be possible to evaluate the influence of the different experimental parameters on the sorption process. Moreover, to optimize the performance of a specific nanomaterial in a determined matrix and expedite the study of the sorption process, a statistical tool designated by response surface methodology (RSM) can be applied. This tool has already been used in previous studies to remove contaminants, including chromium [80], using nanomaterials. Also, RSM allows not only to study the impact of the experimental parameters on the desired response (in this case, removal of chromium), but also to determine the best conditions to obtain the best performance of the material. This review allowed identifying the main limitations of chromium sorption process using synthetic nanomaterials, based on the works published until the date.          [42] Nano-ZrO 2 Nano zirconium oxide

Nano-ZrO 2 -glu-CMC
Crosslinking of nanolayer carboxymethyl cellulose (CMC) onto the surface of nano zirconium oxide (Nano-ZrO 2 ) using glutaraldehyde Note that, the conditions that are shaded correspond to the best uptake capacity or removal efficiency obtained; in general, when the type of water is not referred, the authors may have used distilled or milli-Q water; in the column correspondent to "Cr starting specie", total chromium concentration was quantified in the works that refer it; in the other works no mention is made regarding the specie or if it is total concentration; in the column correspondent to "Uptake capacity (mg/g) or removal efficiency (%)", when the value does not present units, it is the uptake capacity; otherwise, it is the removal efficiency; the value presented in parentheses in the column "Uptake capacity (mg/g) or removal efficiency (%)" corresponds to the condition that gave rise to the value of uptake capacity or removal efficiency presented;

M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT
41 the uptake capacity values which do not presented a subscript were obtained either experimentally or by Langmuir model; sometimes, the authors refer to experimental conditions of experiments whose results they do not present; from column "Type of water" until "Cr starting specie", the conditions mentioned are the same for the below lines                  [99] Ch-(Cu 0 ) Zero-valent copperchitosan nanocomposites Milli-Q water 2.0-9. Knowledge gaps on the evaluation of works published on chromium removal from waters; Present work presents major experimental conditions influencing removal efficiency; Research undertaken so far and the conditions used on this topic is here compiled.