Arsenate and Arsenite Sorption Using Biogenic Iron Compounds: Treatment of Real Polluted Waters in Batch and Continuous Systems

: Arsenic pollution in waters is due to natural and anthropogenic sources. Human exposure to arsenic is associated with acute health problems in areas with high concentrations of this element. Nanometric iron compounds with large speciﬁc surface areas and higher binding energy produced by some anaerobic microorganisms are thus expected to be more efﬁcient adsorbents for the removal of harmful metals and metalloids than chemically produced iron oxides. In this study, a natural consortium from an abandoned mine site containing mainly Clostridium species was used to biosynthesize solid Fe(II) compounds, siderite (FeCO 3 ) and iron oxides. Biogenic precipitates were used as adsorbents in contact with solutions containing arsenate and arsenite. The adsorption of As(V) ﬁtted to the Langmuir model ( q max = 0.64 mmol/g, K L = 0.019 mmol/L) at the optimal pH value (pH 2), while the As(III) adsorption mechanism was better represented by the Freundlich model ( K F = 0.476 L/g, n = 2.13) at pH 10. Water samples from the Caracarani River (Chile) with high contents of arsenic and zinc were treated with a biogenic precipitate encapsulated in alginate beads in continuous systems. The optimal operation conditions were low feed ﬂow rate and the up-ﬂow system, which signiﬁcantly improved the contaminant uptake. This study demonstrates the feasibility of the application of biogenic iron compounds in the treatment of polluted waters. in feed ﬂow rate (1 mL/min) an up-ﬂow system. of the reactors increasing the ﬂow the formation of preferential ﬂow The reverse-fed improved the operational efﬁciency of the columns avoiding the preferential channels through the ﬁxed bed. The optimized conditions for the hazardous metal/metalloid adsorption using biogenic iron precipitates demonstrated that bionanotechnology can be applied in the treatment of real waters.


Introduction
Arsenic is one of the most toxic elements and its incorporation into the environment results from natural weathering and anthropogenic activities. Arsenic exists in several oxidation states; however, the predominant species in aqueous solution are As(III) and As(V), which are considered a major environmental problem. Exposure to arsenic mainly occurs through drinking water and causes several health risks in the human body including cancer, skin damage and circulatory problems [1]. Arsenic is highly toxic even at tracelevel concentrations and enters the food chain mainly from contaminated water [2]; in consequence, regulations are enforced to reduce the admissible levels of arsenic in water [3].
Conventional water treatment methods such as precipitation, filtration, flotation, ion exchange or reverse osmosis are complex, energy-intensive and costly. Membranes have been demonstrated to be an effective method that removes not only arsenic, but other water contaminants. Nanofiltration and reverse osmosis present a high removal

Adsorbents
An iron-reducing microbial consortium from an abandoned mine near La Unión (Murcia, Spain) precipitated iron compounds using soluble ferric citrate as the iron source. These biogenic precipitates were identified as siderite (FeCO 3 ), magnetite (Fe 3 O 4 ) and vivianite (Fe 3 (PO 4 ) 2 ·8H 2 O). The surface area of the biogenic iron compounds was 56.98 m 2 /g. The pore volume was 0.122 cm 3 /g and the average pore size 83 Å [18].
To obtain alginate beads encapsulating the biogenic iron compounds, 20 g/L of the air-dried precipitates were dispersed in an aqueous solution of sodium alginate in a concentration of 20 g/L. The dispersion was pressed through a syringe (internal diameter of 0.5 mm) and dropped into 3% CaCl 2 solution.
The adsorbents used in the experiments (biogenic iron precipitates and beads) were observed in a scanning electron microscope (SEM) (JEOL JSM-6330 F, Yamagata, Japan). Before the observation, the samples were coated with a thin layer of gold.

Batch Experiments
The adsorption experiments were performed with synthetic solutions prepared with sodium arsenite (NaAsO 2 ) and hydrated sodium arsenate (Na 2 HAsO 4 ·7H 2 O) provided by Fisher Scientific (Loughborough, UK). The initial pH of solutions was adjusted with diluted H 2 SO 4 and with NaOH. The biogenic iron compounds were placed in contact with solutions in glass Erlenmeyers stirred at a rotation speed of 150 rpm and at room temperature (25 ± 1 • C). Liquid samples were collected at different times (0, 15, 30, 60 and 120 min) and metalloid concentrations were measured by ICP-OES (Perkin Elmer Optima 2100 DV, Wellesley, MA, USA) (plasma gas (Ar) flow rate: 15 L/min; pumping rate: 1.50 mL/min). Kinetic studies were performed at different initial pH values (2, 4, 7 and 10) adding 1 g/L of iron precipitates to continuously stirred flasks containing 100 mL of 50 mg/L As(III) or As(V) solutions. Adsorption isotherms were also performed at the optimal pH value and at constant temperature by varying the initial arsenic concentration from 10 to 100 mg/L.
The biogenic adsorbents were also used for the treatment of polluted water from the Caracarani River (upper Lluta River basin, Chile). The Azufre River and the Caracarani River confluence and from this point downstream is also named Caracarani River. The Azufre River ("sulfur river" in English) has its source at the Tacora volcano, near an abandoned sulfur mine and tailings site. The oxidation of sulfur tailings and volcanic geothermal springs generate an acidic stream (pH < 2) containing several potentially toxic elements. In contrast, the Caracarani River presents a pH value between 8.2 and 8.6 [19,20]. The main characteristics of the water used in the experiments are shown in Table 1. Adsorption experiments with Caracarani River water were performed by adding 1 g/L of adsorbent (biogenic precipitates or precipitates encapsulated in alginate beads) to Erlenmeyer flasks containing 100 mL of dissolution under stirring at room temperature. Sampling was performed periodically, and zinc and arsenic concentrations were determined by ICP-OES (Perkin Elmer Optima 2100 DV, Wellesley, MA, USA).
All the adsorption tests were performed in duplicates, and the average measurements were used for data processing.

Continuous Sorption Tests
Continuous adsorption experiments were performed with polluted water from Caracarani River using small glass columns (2.5 cm inner diameter and 15 cm length) packed with biogenic iron precipitates encapsulated in alginate beads (~14% hollow volume). The initial pH value of the solution was 2.2 (pH of the samples as received). That pH gave the highest uptakes in previous batch experiments carried out with these adsorbents, and the metal precipitation is avoided at this pH value. The polluted water was introduced to the columns using a peristaltic pump (MasterFlex L/S, Vernon Hills, IL, USA). In order to ensure a good distribution of solution and to prevent the loss of adsorbent, glass wool was placed on the top and bottom of the column.

Characterization Techniques
Iron precipitates produced by the natural microbial consortium and the alginate beads encapsulating the biogenic precipitates were coated with a thin layer of gold and observed using a scanning electron microscope (SEM) (JEOL JSM-6330 F, Yamagata, Japan).
Microbially precipitated iron compounds were analyzed after the adsorption experiments using powder X-ray diffraction (XRD) on a Philips X pert-MPD (Malvern Panalytical, Malvern, UK) with a Cu anode operating at a wavelength of 1.5406 Å as the X-ray source.

Effect of pH
The adsorbent generated by this microbial consortium from an abandoned mining area showed its ability to remove several hazardous species in a previous study, especially chromate and arsenate [18]. The solution pH influences both the adsorbent and the chemistry of the metals and metalloids in solution. In this work, the experiments were performed with an initial arsenic concentration of 50 mg/L and adsorption time of 2 h. The pH range of the study was 2 to 10 for arsenite and arsenate, which are soluble at any pH ( Figure 1). with biogenic iron precipitates encapsulated in alginate beads (~14% hollow volume). The initial pH value of the solution was 2.2 (pH of the samples as received). That pH gave the highest uptakes in previous batch experiments carried out with these adsorbents, and the metal precipitation is avoided at this pH value. The polluted water was introduced to the columns using a peristaltic pump (MasterFlex L/S, Vernon Hills, IL, USA). In order to ensure a good distribution of solution and to prevent the loss of adsorbent, glass wool was placed on the top and bottom of the column.

Characterization Techniques
Iron precipitates produced by the natural microbial consortium and the alginate beads encapsulating the biogenic precipitates were coated with a thin layer of gold and observed using a scanning electron microscope (SEM) (JEOL JSM-6330 F, Yamagata, Japan).
Microbially precipitated iron compounds were analyzed after the adsorption experiments using powder X-ray diffraction (XRD) on a Philips X′pert-MPD (Malvern Panalytical, Malvern, UK) with a Cu anode operating at a wavelength of 1.5406 Å as the X-ray source.

Effect of pH
The adsorbent generated by this microbial consortium from an abandoned mining area showed its ability to remove several hazardous species in a previous study, especially chromate and arsenate [18]. The solution pH influences both the adsorbent and the chemistry of the metals and metalloids in solution. In this work, the experiments were performed with an initial arsenic concentration of 50 mg/L and adsorption time of 2 h. The pH range of the study was 2 to 10 for arsenite and arsenate, which are soluble at any pH ( Figure 1).  Figure 2 shows the As(III) and As(V) adsorption using the iron compounds biosynthesized by the natural consortium at different initial pH values. The adsorption of oxyanions on iron precipitates is strongly influenced by the pH value of the solution. The pH modifies the chemistry of the metalloids and the binding sites due to the protonation of functional groups. Arsenate adsorption was higher at an acidic pH because the surface of the iron compounds has a net positive charge that would attract oxyanions. Conversely, Figure 1. Speciation diagrams for As(III) and As(V) as a function of pH for a total arsenic concentration of 50 mg/L obtained using Hydra and Medusa software (KTH Royal Institute of Technology, Stockholm, Sweden). Figure 2 shows the As(III) and As(V) adsorption using the iron compounds biosynthesized by the natural consortium at different initial pH values. The adsorption of oxyanions on iron precipitates is strongly influenced by the pH value of the solution. The pH modifies the chemistry of the metalloids and the binding sites due to the protonation of functional groups. Arsenate adsorption was higher at an acidic pH because the surface of the iron compounds has a net positive charge that would attract oxyanions. Conversely, arsenite was better adsorbed at higher pH values because its deprotonation takes place over pH 8. This behavior is caused by the electrostatic interaction between the arsenic oxyanions and the surface of the biogenic iron compounds with positive charge [18]. arsenite was better adsorbed at higher pH values because its deprotonation takes place over pH 8. This behavior is caused by the electrostatic interaction between the arsenic oxyanions and the surface of the biogenic iron compounds with positive charge [18].

Effect of Arsenic Concentration
Arsenic concentration is also a relevant variable during the adsorption onto biogenic iron precipitates. Electrostatic interactions between functional groups and toxic ions have a significant effect on the nexus between initial concentration and metalloid uptake. Kinetic studies were carried out at different arsenic concentrations to determine adsorption parameters using models ( Figure 3).

Effect of Arsenic Concentration
Arsenic concentration is also a relevant variable during the adsorption onto biogenic iron precipitates. Electrostatic interactions between functional groups and toxic ions have a significant effect on the nexus between initial concentration and metalloid uptake. Kinetic studies were carried out at different arsenic concentrations to determine adsorption parameters using models ( Figure 3). arsenite was better adsorbed at higher pH values because its deprotonation takes place over pH 8. This behavior is caused by the electrostatic interaction between the arsenic oxyanions and the surface of the biogenic iron compounds with positive charge [18].

Effect of Arsenic Concentration
Arsenic concentration is also a relevant variable during the adsorption onto biogenic iron precipitates. Electrostatic interactions between functional groups and toxic ions have a significant effect on the nexus between initial concentration and metalloid uptake. Kinetic studies were carried out at different arsenic concentrations to determine adsorption parameters using models ( Figure 3). Metalloid adsorption occurs in three steps. Firstly, the adsorbate in the solution is transferred to the external surface of the adsorbent, then the adsorbate diffuses internally to the binding sites and finally the adsorption takes place itself. The adsorption step is considered the slowest step in the different models.
The uptake mechanisms are determined by the isotherms and their characteristic parameters (adsorbent capacities and affinity) are specific for each system. Data obtained in adsorption experiments for arsenate and arsenite onto the biogenic iron precipitates were fitted to the Langmuir and Freundlich models.
The Langmuir model is applied to monolayer adsorption where the adsorbate is bound at homogeneous active sites on the adsorbent surface, and adsorption is independent of the number of occupied sites. A Langmuir isotherm can be expressed by the following equation: where q is the adsorption capacity and C e is the metalloid concentration. The Freundlich model is an empirical model, which assumes multilayer adsorption with interaction between molecules on the adsorbent surface. The mathematical expression for this model is where K F is a constant related to the adsorption capacity and n (dimensionless) is a constant that indicates the nature and the effectiveness of the adsorption. The adsorption mechanism and isotherm model can be different for a given adsorbent if there are different species in solution. C e /q was graphically represented versus C e to obtain a linear regression to directly calculate the Langmuir parameters, q max and b (inverse of K L ) values. Furthermore, log q e was plotted versus log C e and the linear regression analysis was used to determine n and K F ( Table 2).
The adsorption of As(V) onto biogenic precipitates fitted better to the Langmuir model, while the As(III) adsorption mechanism was better represented by the Freundlich model. In the Langmuir model, the effectiveness of the adsorbent is greater when the q max and K L values are lower. In the Freundlich model, a higher value of K F indicates a higher adsorption capacity and higher values of n reflect a diminishing tendency for sorption with increasing uptake. Despite of the experimental results with different arsenic species fitted to different models, the arsenate adsorption seemed to be more effective than arsenite removal. Table 2. Langmuir and Freundlich parameters for As(III) and As(V) adsorption isotherms on the iron precipitates generated by the microbial consortium. A comparison between the maximum adsorption capacities of the biogenic iron precipitates and other iron-based adsorbents under similar conditions is shown in Table 3. Different adsorbents have been tested for arsenic removal due to the high effectiveness of iron compounds for adsorption. Nevertheless, great differences are observed in their adsorption capacities mainly related to their particle size. Nanostructures exhibited the highest capacities. The micro-/nanostructructures precipitated by the microbial consortium presented good properties as arsenic adsorbents compared to other tested materials. The microbial consortium generated a precipitate under anaerobic conditions containing Fe(II): siderite (FeCO 3 ), magnetite (Fe 3 O 4 ) and vivianite (Fe 3 (PO 4 ) 2 ·8H 2 O). The residues were also analyzed by X-ray diffraction after the adsorption experiments with As(III) and As(V) in solution ( Figure 4).

Adsorbate pH Langmuir Parameters Freundlich Parameters
Both As(III) and As(V) are strongly adsorbed onto the iron precipitate surface through surface complexation and surface precipitation processes. These processes depend on the arsenic oxidation state [28]. It has been reported that As(III) binds to iron-bearing minerals by tridentate complexation in a first stage. Then, As(III) and As(V) form amorphous and nanocrystalline surface precipitates [29,30]. Figure 4a reveals the occurrence of As(III) precipitates with low crystallinity, As 4 O 6 and As 2 O 3 . In contrast, the few works that have reported the interaction of As(V) with Fe(II)-bearing minerals suggest that adsorption and precipitation processes leads to different arsenic speciation [31]. The initial adsorbed As(V) is partially reduced, most probably due to the presence of Fe(II) in the biogenic iron compound produced under anaerobic conditions. As shown in Figure 4b, the complexation and reduction of As(V) occurred, and the final products were As 2 O and As that covered the adsorbent. sorption capacities mainly related to their particle size. Nanostructures exhibited the highest capacities. The micro-/nanostructructures precipitated by the microbial consortium presented good properties as arsenic adsorbents compared to other tested materials. The microbial consortium generated a precipitate under anaerobic conditions containing Fe(II): siderite (FeCO3), magnetite (Fe3O4) and vivianite (Fe3(PO4)2·8H2O). The residues were also analyzed by X-ray diffraction after the adsorption experiments with As(III) and As(V) in solution (Figure 4).
Both As(III) and As(V) are strongly adsorbed onto the iron precipitate surface through surface complexation and surface precipitation processes. These processes depend on the arsenic oxidation state [28]. It has been reported that As(III) binds to ironbearing minerals by tridentate complexation in a first stage. Then, As(III) and As(V) form amorphous and nanocrystalline surface precipitates [29,30]. Figure 4a reveals the occurrence of As(III) precipitates with low crystallinity, As4O6 and As2O3. In contrast, the few works that have reported the interaction of As(V) with Fe(II)-bearing minerals suggest that adsorption and precipitation processes leads to different arsenic speciation [31]. The initial adsorbed As(V) is partially reduced, most probably due to the presence of Fe(II) in the biogenic iron compound produced under anaerobic conditions. As shown in Figure  4b, the complexation and reduction of As(V) occurred, and the final products were As2O and As that covered the adsorbent.

Arsenic Adsorption in Caracarani River Waters
The use of biogenic iron compounds as adsorbents in continuous systems is difficult due to their small size (Figure 5a). Iron precipitates were encapsulated in alginate beads to overcome the difficulties in the separation of micro/nano adsorbents from waters (Figure 5b).

Arsenic Adsorption in Caracarani River Waters
The use of biogenic iron compounds as adsorbents in continuous systems is difficult due to their small size (Figure 5a). Iron precipitates were encapsulated in alginate beads to overcome the difficulties in the separation of micro/nano adsorbents from waters (Figure 5b). Batch experiments for the treatment of Caracarani River waters were performed with the biogenic iron precipitates and with iron compound particles encapsulated in alginate beads at different pH values, as shown in Figure 6. Iron precipitates formed by the microbial consortium were able to remove arsenic ions from the polluted waters mainly at pH 2.24 (as received) while Zn 2+ removal was very low because organic matter associated with the solid was involved in cation binding. At higher pH values, arsenic coprecipitates with the iron in solution and zinc is precipitated at pH 7 and 10.
Alginate beads encapsulating iron solids were able to adsorb not only the arsenic but also zinc ions. Alginate is a biocopolymer that contains alternating blocks of D-mannuronic and L-guluronic acids rich in carboxyl groups that play a major role in hazardous metal adsorption. The dissociation constant of carboxylic groups in the mannuronic acid is pKa = 3.38 and in glucuronic acid is pKa = 3.65 [32]. The carboxylic groups are bound to protons at pH values below the pKa. In consequence, the access of metallic cations to these groups is limited due to repulsive forces. For this reason, zinc adsorption from polluted waters was very low at pH 2.24. At pH 4, carboxylates with negative charges are available and the uptake by alginate slightly increased. Probably, there is a competition between different cations present in the solution for the binding sites in the beads. Batch experiments for the treatment of Caracarani River waters were performed with the biogenic iron precipitates and with iron compound particles encapsulated in alginate beads at different pH values, as shown in Figure 6. Iron precipitates formed by the microbial consortium were able to remove arsenic ions from the polluted waters mainly at pH 2.24 (as received) while Zn 2+ removal was very low because organic matter associated with the solid was involved in cation binding. At higher pH values, arsenic co-precipitates with the iron in solution and zinc is precipitated at pH 7 and 10.
Alginate beads encapsulating iron solids were able to adsorb not only the arsenic but also zinc ions. Alginate is a biocopolymer that contains alternating blocks of D-mannuronic and L-guluronic acids rich in carboxyl groups that play a major role in hazardous metal adsorption. The dissociation constant of carboxylic groups in the mannuronic acid is pK a = 3.38 and in glucuronic acid is pK a = 3.65 [32]. The carboxylic groups are bound to protons at pH values below the pK a . In consequence, the access of metallic cations to these groups is limited due to repulsive forces. For this reason, zinc adsorption from polluted waters was very low at pH 2.24. At pH 4, carboxylates with negative charges are available and the uptake by alginate slightly increased. Probably, there is a competition between different cations present in the solution for the binding sites in the beads.

Continuous Adsorption Tests
The ability of biogenic iron compounds to adsorb the main metals present in the polluted water from Caracarani River under batch conditions has previously been demonstrated. Continuous experiments in column reactors are required to treat larger volumes. The experiments were carried out with real effluents from Caracarani River (Chile) at pH 2.24 and the adsorbent was the alginate beads immobilizing the biogenic iron precipitates.

Effect of Feed Flow Rate
Two columns with encapsulated biogenic iron compounds were tested at 2 and 1 mL/min and their column efficiency and contaminant uptake compared (Figure 7) (Table  3). Despite the increase in the flow rate enabling the decontamination of larger volumes in less time, higher flow rates led to a decrease in the saturation time of the columns, evidencing that the mass transfer zone became narrower.
The saturation time increased from 150 min at a flow rate of 2 mL/min to 210 min at 1 mL/min, and the slope of the breakthrough curve decreased. The longer contact time between the solution and the beads is likely to have been the result of an increase in metal uptake.

Continuous Adsorption Tests
The ability of biogenic iron compounds to adsorb the main metals present in the polluted water from Caracarani River under batch conditions has previously been demonstrated. Continuous experiments in column reactors are required to treat larger volumes. The experiments were carried out with real effluents from Caracarani River (Chile) at pH 2.24 and the adsorbent was the alginate beads immobilizing the biogenic iron precipitates.

Effect of Feed Flow Rate
Two columns with encapsulated biogenic iron compounds were tested at 2 and 1 mL/min and their column efficiency and contaminant uptake compared (Figure 7) ( Table 3). Despite the increase in the flow rate enabling the decontamination of larger volumes in less time, higher flow rates led to a decrease in the saturation time of the columns, evidencing that the mass transfer zone became narrower.

Effect of Bed Height
The pollutant uptake in two columns with 3 cm (7.87 g of adsorbent) and 6 cm (13.35 g) of bed height and a flow rate of 1 mL/min are compared in Figure 8. The initial outlet concentration was influenced by bed height and was lower for the column containing a higher amount of adsorbent. An increase in the bed height led to longer saturation and service time of the reactor due to the higher quantity of beads. According to Table 3, the increase in the bed height led to higher adsorption capacities. At the lowest bed height there is not enough time for the ions to diffuse into the beads [33].

Effect of Feeding System
The reverse-fed system was compared to drop-fed columns to increase the efficiency of the contact between the adsorbent and the polluted water avoiding preferential channels ( Figure 9). The saturation time increased from 150 min at a flow rate of 2 mL/min to 210 min at 1 mL/min, and the slope of the breakthrough curve decreased. The longer contact time between the solution and the beads is likely to have been the result of an increase in metal uptake.

Effect of Bed Height
The pollutant uptake in two columns with 3 cm (7.87 g of adsorbent) and 6 cm (13.35 g) of bed height and a flow rate of 1 mL/min are compared in Figure 8. The initial outlet concentration was influenced by bed height and was lower for the column containing a higher amount of adsorbent. An increase in the bed height led to longer saturation and service time of the reactor due to the higher quantity of beads. According to Table 3, the increase in the bed height led to higher adsorption capacities. At the lowest bed height there is not enough time for the ions to diffuse into the beads [33].

Effect of Bed Height
The pollutant uptake in two columns with 3 cm (7.87 g of adsorbent) and 6 cm (13.35 g) of bed height and a flow rate of 1 mL/min are compared in Figure 8. The initial outlet concentration was influenced by bed height and was lower for the column containing a higher amount of adsorbent. An increase in the bed height led to longer saturation and service time of the reactor due to the higher quantity of beads. According to Table 3, the increase in the bed height led to higher adsorption capacities. At the lowest bed height there is not enough time for the ions to diffuse into the beads [33].

Effect of Feeding System
The reverse-fed system was compared to drop-fed columns to increase the efficiency of the contact between the adsorbent and the polluted water avoiding preferential chan-

Effect of Feeding System
The reverse-fed system was compared to drop-fed columns to increase the efficiency of the contact between the adsorbent and the polluted water avoiding preferential channels ( Figure 9).
The effect of the feeding systems in the pollutant adsorption was remarkable. Saturation time and column efficiency notably increased in the reverse-fed system and arsenic concentration was not detected after 6 h of treatment. The reverse-fed column reached a higher metalloid adsorption in the same period (Table 3).
A significant difference between the drop-fed and the reverse-fed systems is the shape of the breakthrough curve. The curve was favorably modified in the reverse-fed column showing a steeper slope and a lag period that enhanced the service time of the reactor. Figure 9. Arsenic and zinc adsorption breakthrough curves of columns with alginate beads containing biogenic iron compounds using drop and reverse feeding systems (3 cm of bed height (7.87 g of adsorbent) and 1 mL/min flow rate).
The scarcity of water resources has led to the use of high arsenic concentration waters in agriculture and other human activities. The contaminated water containing heavy metals can pollute the agricultural soil and crop plants with potential environmental and human health risks [34,35]. This study attempts to reduce the water pollution below the maximum legal limits for irrigation water (0.1 mg/L for arsenic and 2 mg/L in the case of zinc) [36]. The parameters of the continuous experiments are collected in Table 4 and the breakthrough point was the legal limit (although the column was not completely saturated), because it is the most relevant information for the implementation.
Continuous adsorption parameters were determined by studying the breakthrough curves of the columns. The amount of treated metal/metalloid (Metr (mmol)) and the retained pollutants by the column (Mead (mmol)) can be calculated as follows: where C0 is the inlet metal/metalloid concentration (mmol/l); t is breakthrough time (h); Q is the flow rate (L/h) and A is the area under the breakthrough curve (h). The adsorption yield (%Ad) and the adsorption capacity (qc) of the columns can be calculated using the following expressions: where qc is the adsorption capacity (mmol/g) and m is the amount of adsorbent in the column (g).

Figure 9.
Arsenic and zinc adsorption breakthrough curves of columns with alginate beads containing biogenic iron compounds using drop and reverse feeding systems (3 cm of bed height (7.87 g of adsorbent) and 1 mL/min flow rate).
The effect of the feeding systems in the pollutant adsorption was remarkable. Saturation time and column efficiency notably increased in the reverse-fed system and arsenic concentration was not detected after 6 h of treatment. The reverse-fed column reached a higher metalloid adsorption in the same period (Table 3).
A significant difference between the drop-fed and the reverse-fed systems is the shape of the breakthrough curve. The curve was favorably modified in the reverse-fed column showing a steeper slope and a lag period that enhanced the service time of the reactor.
The scarcity of water resources has led to the use of high arsenic concentration waters in agriculture and other human activities. The contaminated water containing heavy metals can pollute the agricultural soil and crop plants with potential environmental and human health risks [34,35]. This study attempts to reduce the water pollution below the maximum legal limits for irrigation water (0.1 mg/L for arsenic and 2 mg/L in the case of zinc) [36]. The parameters of the continuous experiments are collected in Table 4 and the breakthrough point was the legal limit (although the column was not completely saturated), because it is the most relevant information for the implementation.
Continuous adsorption parameters were determined by studying the breakthrough curves of the columns. The amount of treated metal/metalloid (Me tr (mmol)) and the retained pollutants by the column (Me ad (mmol)) can be calculated as follows: where C 0 is the inlet metal/metalloid concentration (mmol/L); t is breakthrough time (h); Q is the flow rate (L/h) and A is the area under the breakthrough curve (h). The adsorption yield (%Ad) and the adsorption capacity (q c ) of the columns can be calculated using the following expressions: %Ad = 100·Me ad /Me tr (5) q c = Me ad /m (6) where q c is the adsorption capacity (mmol/g) and m is the amount of adsorbent in the column (g). The beads encapsulating the biogenic iron compounds were analyzed after the continuous experiments by X-ray diffraction. The diffractogram showed the presence of As 0 and Zn 4 (AsO 4 ) 2 O ( Figure 10). The adsorption and reduction of arsenate on iron precipitated at acidic pH values forming As 0 on the adsorbent surface was already evidenced. In addition, the coadsorption of Zn(II) and arsenate ions on the beads with iron precipitates could begin forming arsenate and zinc complexes on iron/alginate beads at lower concentrations, then the subsequent coprecipitation of Zn(II) and arsenate on the carboxylic groups of beads could lead to the formation of a surface precipitate Zn 4 (AsO 4 ) 2 O at higher concentrations [37].  The beads encapsulating the biogenic iron compounds were analyzed after the continuous experiments by X-ray diffraction. The diffractogram showed the presence of As 0 and Zn4(AsO4)2O ( Figure 10). The adsorption and reduction of arsenate on iron precipitated at acidic pH values forming As 0 on the adsorbent surface was already evidenced. In addition, the coadsorption of Zn(II) and arsenate ions on the beads with iron precipitates could begin forming arsenate and zinc complexes on iron/alginate beads at lower concentrations, then the subsequent coprecipitation of Zn(II) and arsenate on the carboxylic groups of beads could lead to the formation of a surface precipitate Zn4(AsO4)2O at higher concentrations [37]. Alginate beads containing biogenic iron precipitates are cheap adsorbents for the removal of potentially toxic ions; nevertheless, the regeneration of the beads and their reuse in sorption-desorption cycles would be interesting from an economic point of view. Commonly, diluted acids are used as eluants with alginate; nevertheless, the adsorbent experiments show a significant loss of effectiveness after elution [38]. In addition, the acidic pH could partially dissolve the iron particles resulting in the loss of the adsorbent mass. Sigdel et al. [39] reported that hydrous iron oxide-alginate beads could be effectively regenerated  The beads encapsulating the biogenic iron compounds were analyzed after the continuous experiments by X-ray diffraction. The diffractogram showed the presence of As 0 and Zn4(AsO4)2O ( Figure 10). The adsorption and reduction of arsenate on iron precipitated at acidic pH values forming As 0 on the adsorbent surface was already evidenced. In addition, the coadsorption of Zn(II) and arsenate ions on the beads with iron precipitates could begin forming arsenate and zinc complexes on iron/alginate beads at lower concentrations, then the subsequent coprecipitation of Zn(II) and arsenate on the carboxylic groups of beads could lead to the formation of a surface precipitate Zn4(AsO4)2O at higher concentrations [37]. Alginate beads containing biogenic iron precipitates are cheap adsorbents for the removal of potentially toxic ions; nevertheless, the regeneration of the beads and their reuse in sorption-desorption cycles would be interesting from an economic point of view. Commonly, diluted acids are used as eluants with alginate; nevertheless, the adsorbent experiments show a significant loss of effectiveness after elution [38]. In addition, the acidic pH could partially dissolve the iron particles resulting in the loss of the adsorbent mass. Sigdel et al. [39] reported that hydrous iron oxide-alginate beads could be effectively regenerated Alginate beads containing biogenic iron precipitates are cheap adsorbents for the removal of potentially toxic ions; nevertheless, the regeneration of the beads and their reuse in sorption-desorption cycles would be interesting from an economic point of view. Commonly, diluted acids are used as eluants with alginate; nevertheless, the adsorbent experiments show a significant loss of effectiveness after elution [38]. In addition, the