Environ. Sci. Technol. 2009, 43, 3853–3859
Ferrous and Ferric Ion Generation During Iron Electrocoagulation DIVAGAR LAKSHMANAN, DENNIS A. CLIFFORD,* AND GAUTAM SAMANTA Department of Civil and Environmental Engineering, University of Houston, Houston, Texas, 77204-4003
Received January 20, 2009. Revised manuscript received March 17, 2009. Accepted March 19, 2009.
at the anode (eq 1) results in formation of Al3+, which hydrolyzes to form hydroxides depending on the pH (eq 2). Al(s) f Al3+ + 3e-: E0 ) 1.662 V 3+
Al
Fe(s) f Fe2+ + 2e-: E0 ) 0.44 V 2+
Fe
f Fe
3+
(3)
0
+ e : E ) -0.771 V
3+
-
(4)
0
+ 3e : E ) 0.037 V
(5)
Literature reports are conflicting as to the production of Fe2+ or Fe3+ and its hydrolysis during Fe-EC. Most Fe-EC studies have focused on contaminant removal, and the solidssFe(OH)2(s) or Fe(OH)3(s)sreported to be formed in these studies have been based on presumption rather than experimental proof. The following summary of the Fe-EC reactions reported in the literature shows the ambiguity surrounding the fundamental mechanisms involved: 1. Some studies (10, 15, 16) report that electrolytic oxidation of iron produces Fe2+, which hydrolyzes to produce insoluble Fe(OH)2(s) (eq 6). Others (18) report the formation of Fe(OH)2(s) without giving details of its formation. 2. Some studies (2, 12, 16) report the formation of Fe2+ at the anode, followed by Fe2+ oxidation by dissolved oxygen (DO) to form Fe(OH)3(s) (eq 7). A modification of this reaction sequence, reported by Parga et al. (10), is an electrolytic, as opposed to DO, oxidation of Fe2+ to Fe3+ step prior to Fe3+ hydrolysis to produce Fe(OH)3(s). 3. Yet a third reported mechanism (7, 13, 17) involves one-stage electrolytic oxidation of iron to Fe3+, followed by hydrolysis to produce Fe(OH)3(s) (eq 8). Finally, some studies (5, 9, 18) have reported formation of Fe(OH)3(s) as end product during Fe-EC without specifying the reactions involved. Electrolysis
Hydrolysis
Fe(s) 98 Fe2+ 98 Fe(OH)2(s) Electrolysis
Inorganic pollutants and microorganisms are sometimes removed from water by electro-chemical processes such as electrocoagulation, electrodialysis, electrodeionization, and electrofloatation. Electrocoagulation (EC), a process that generates coagulant electrochemically, has been reported to be efficient in removing a wide range of pollutants from water and wastewater (1-13). Although EC processes with aluminum (Al) and iron (Fe) anodes have been known for over a century, and have been used commercially for treating wastewater, they have received little attention for drinking water treatment, due to variable process performance, especially for iron EC and conflicting reports on the physicochemical reactions involved (2, 5, 9, 7, 10, 12, 14-18). Furthermore, the relevant contaminant-removal paths during EC are reported to vary with contaminant and the aqueous medium characteristics (10, 14, 18). Typical Reaction Steps Reported During EC. Unlike FeEC, the chemical reactions reported during Al-EC are consistent in the literature (1-3, 14). The electrolytic oxidation * Corresponding author phone: 713-743-4266; fax: 713-743-4260; e-mail:
[email protected]. Published on Web 04/10/2009
-
(6)
Oxidation and Hydrolysis
Fe(s) 98 Fe2+ 98 Fe(OH)3(s)
Introduction
10.1021/es8036669 CCC: $40.75
(2)
Theoretically, the electrolytic oxidation of iron could result in ferrous (Fe2+) or ferric (Fe3+) generation (eq 3-5) at the anode.
Fe(s) f Fe
Our research on arsenate removal by iron electrocoagulation (EC) produced highly variable results, which appeared to be due to Fe2+ generation without subsequent oxidation to Fe3+. Because the environmental technology literature is contradictory with regard to the generation of ferric or ferrous ions during EC, the objective of this research was to establish the iron species generated during EC with iron anodes. Experimental results demonstrated that Fe2+, not Fe3+, was produced at the iron anode. Theoretical current efficiency was attained based on Fe2+ production with a clean iron rod, regardless of current, dissolvedoxygen (DO) level, or pH (6.5-8.5). The Fe2+ remaining after generation and mixing decreased with increasing pH and DO concentration due to rapid oxidation to Fe3+. At pH 8.5, Fe2+ was completely oxidized, which resulted in the desired Fe(OH)3(s)/ FeOOH(s), whereas, at pH 6.5 and 7.5, incomplete oxidation was observed, resulting in a mixture of soluble Fe2+ and insoluble Fe(OH)3(s)/FeOOH(s). When compared with Fe2+ chemical coagulation, a transient pH increase during EC led to faster Fe2+ oxidation. In summary, for EC in the pH 6.5-7.5 range and at low DO conditions, there is a likelihood of soluble Fe2+ species passing through a subsequent filtration process resulting in secondary contamination and inefficient contaminant removals.
(1)
+ nH2O f Al(OH)n(3-n) + nH+
2009 American Chemical Society
Electrolysis
Hydrolysis
Fe(s) 98 Fe3+ 98 Fe(OH)3(s)
(7)
(8)
On the basis of results from this research, the reported formation of insoluble Fe(OH)2(s) is questionable as is the direct electrolytic oxidation of iron to Fe3+. It was found that: (a) formation of Fe(OH)2(s) did not occur, and (b) Fe(OH)3(s) was produced by way of Fe2+ formation at the anode followed by its highly pH-dependent oxidation by DO. Evidence of these reactions is presented here. The literature also lacks attention to pH, DO, electrolysis time, and retention time, i.e., factors that significantly influence the reactions and the end products formed during Fe-EC. Furthermore, there are conflicting reports on whether pH increases or decreases as a result of electrochemical iron and aluminum oxidation. Some studies on Fe- (10, 13) or AlEC (1, 13) have shown a pH rise, while others have reported a pH decrease in the alkaline region with Al-EC (2, 3). The Al-EC studies reported that EC plays a role of neutralization, but it is unclear whether this neutralization also occurs during Fe-EC. Due to these reported ambiguities surrounding the VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3853
FIGURE 1. (a) Sketch of EC cell used for iron generation studies with iron rod anode under atmospheric conditions. (b) Sketch of EC cell with iron rod anode under N2-purged condition. Fe-EC reactions, a fundamental study on Fe-EC seemed warranted in an effort to clarify the reactions occurring. A further reason to study the Fe-EC reactions came from our initial results on arsenic removal by Fe-EC microfiltration, using a representative synthetic groundwater. These results showed significant concentrations of Fe2+ in the filtrate at pH 6.5 and even at pH 7.5, resulting in erratic arsenic removals. Objectives. The objectives of the research were to (1) measure the Faraday’s law current efficiency during iron generation, and determine the iron species (Fe2+ and/or Fe3+) produced at the anode, (2) determine the effects of pH, DO, current, and anode purity on iron oxidation and hydroxide formation, (3) compare the rates of Fe2+ oxidation during electrocoagulation and chemical coagulation (CC), and (4) summarize the likely reactions and mechanisms involved during Fe-EC in the 6.5-8.5 pH range.
Experimental Section Analytical reagent grade chemicals were used unless otherwise noted. NSFI-53 challenge water (19), a representative synthetic groundwater for arsenic removal studies (Table 1S in Supporting Information) containing typical cations and anions was used in the pH range 6.5-8.5 for all experiments. The EC experiments were carried out in a stirred, cylindrical electrode chamber made of acrylic plastic, containing a porous cylindrical stainless-steel cathode and iron-rod anode. The distance between the anode and cathode was 2 mm. A constant-current power supply provided direct current to the electrodes. All the experiments were carried out in batch mode following the practice of previous EC studies (2, 3, 14). Electrocoagulation Experiments under Open Atmospheric Conditions. The sketch of a single electrode-pair EC cell used for the electrode-passivation and iron-generation efficiency studies carried out under open atmospheric conditions is shown in Figure 1a (a photograph of the cell is shown in Figure 1S of Supporting Information). The anode used in the studies was a 5 mm diameter industrial grade iron rod (98.5% Fe) with active surface area of 38 cm2. Effect of Rusting on Iron Generation Efficiency. Initial experiments showed a decrease in the efficiency of iron generation with time due to rusting of the iron anode. To quantify the degradation rate, the iron anode was repeatedly tested for iron generation without cleaning between experiments. Every day for 15 days, after operation of the cell at 0.2 A for 1-2 h in a batch mode, the electrodes were taken 3854
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
out of the system and left open to the atmosphere before reuse without cleaning the next day. Iron Generation under Atmospheric Condition. The results of the deliberate anode-rusting experiments showed that EC performance was repeatable only by precleaning the iron rod. Thus, for all subsequent experiments, the iron rod was precleaned mechanically with sandpaper to remove the rust. To determine iron-generation efficiency, the electrolytic cell was operated by varying the current (0.05, 0.2, 0.4, and 0.8 A) and electrolysis time (0-60 s) at pH 6.5, 7.5, and 8.5 with NSFI challenge water. The voltage and the current density ranged from 2 to 7 V and 1.32 to 21.1 mA/cm2, respectively as the current was varied from 0.05 to 0.8 A. The challenge water was always mixed for a total of 2 min, including the electrolysis time, and the pH was continuously monitored during each batch experiment. Iron Generation under N2-Purged, Low DO Condition. During two sets of experiments N2-purging was used to strip dissolved oxygen from the challenge water to minimize potential Fe2+ oxidation to Fe3+ when using two different iron anodes: (i) a reagent-grade iron rod (99.995% Fe; Alfa Aesar, MA) and (ii) an industrial-grade iron rod (98.5% Fe). A sketch of the experimental setup for EC studies under N2purged conditions is shown in Figure 1b. The diameter and active surface area of the iron anodes were 5 mm and 18 cm2, respectively. The voltage and the current density ranged from 2 to 7 V and 2.8 to 44.4 mA/cm2, respectively as the current was varied from 0.05 to 0.8 A. NSFI challenge water with initial pH of 6.5 or 7.5 was N2-purged for 1 h, which reduced the DO from 8000-9000 µg/L (open atmospheric condition) to
