Next Article in Journal
Antibody-Based Sensors: Principles, Problems and Potential for Detection of Pathogens and Associated Toxins
Previous Article in Journal
Standard CMOS Fabrication of a Sensitive Fully Depleted Electrolyte-Insulator-Semiconductor Field Effect Transistor for Biosensor Applications
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 9
Issue 6
10.3390/s90604390
sensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Use of Flow-Injection Analysis with Chemiluminescence Detection of Aqueous Ferrous Iron in Waters Containing High Concentrations of Organic Compounds

by
Christopher J. Borman
1,*,
B. Patrick Sullivan
1,
Carrick M. Eggleston
2 and
Patricia J. S. Colberg
3
1
Department of Chemistry, University of Wyoming, Dept 3838, 1000 E. University Avenue, Laramie, Wyoming 82071, USA
2
Department of Geology and Geophysics, University of Wyoming, Dept 3006, 1000 E. University Avenue, Laramie, Wyoming 82071, USA
3
Department of Civil and Architectural Engineering, University of Wyoming, Dept 3295, 1000 E. University Avenue, Laramie, Wyoming 82071, USA
*
Author to whom correspondence should be addressed.
Sensors 2009, 9(6), 4390-4406; https://doi.org/10.3390/s90604390
Submission received: 7 April 2009 / Revised: 1 June 2009 / Accepted: 3 June 2009 / Published: 4 June 2009
(This article belongs to the Section Chemical Sensors)
Browse Figures
Versions Notes

Abstract

:
An evaluation of flow-injection analysis with chemiluminescence detection (FIA-CL) to quantify Fe2+(aq) in freshwaters was performed. Iron-coordinating and/or iron-reducing compounds, dissolved organic matter (DOM), and samples from two natural water systems were used to amend standard solutions of Fe2+(aq). Slopes of the response curves from ferrous iron standards (1 – 100 nM) were compared to the response curves of iron standards containing the amendments. Results suggest that FIA-CL is not suitable for systems containing ascorbate, hydroxylamine, cysteine or DOM. Little or no change in sensitivity occurred in solutions of oxalate and glycine or in natural waters with little organic matter.

Graphical Abstract

1. Introduction

Quantitative analysis of metal ions in natural waters is essential to understanding global biogeochemical cycling. The study of iron species and compounds in aqueous solution has been of particular interest over the past 30 years due to their role in various chemical, physical and biological processes in natural waters. These processes are important in understanding the biogeochemistry of iron and the impacts of contaminants on the systems. Iron (hydr)oxides have been shown to act as sorbents of organic and other metal species [1,2]. Iron is an essential nutrient [3-5], and is abundant in many mineral oxides that exchange iron with surface and ground waters through dissolution and precipitation [6-10]. Iron is often the most abundant redox-active metal ion in natural waters and is crucial to biota during electron transport [5,11-14]. The oxidation state and speciation of iron dictates its bioavailability, and may directly affect both the toxicity and availability of many chemicals in the environment.
Numerous analytical techniques for determining aqueous iron concentration have been developed and deployed [15], including titrimetric [16], electrochemical [17-19], chromatographic [20], ultracentrifugation [21], and photometric methods [22-24]. Flow-injection analysis (FIA) techniques have facilitated simple, accurate and precise determinations when coupled with spectrophotometric detection of chromophoric metal complexes but, there are some limitations of colorimetric analyses that result in inaccurate determinations [15,25,26]. Moreover, spectrophotometric determination of sub-micromolar analyte concentrations (e.g. Fe2+(aq)) is often difficult without extending optical pathlength or pre-concentrating analytes in samples [22,27,28]. One solution for these limitations is the use of chemiluminescence (CL) detection coupled with FIA, which has been shown to be rapid and highly sensitive in the quantitation of aqueous ferrous iron [29-36].

1.1. Chemiluminescence of Luminol

Many compounds have been shown to emit light upon oxidation; the most studied is probably 5-amino-2,3-dihydro-1,4-phthalazinedione, commonly known as luminol. Albrecht (1928) [37] was the first to report the strong CL of luminol and its derivatives during their oxidation in basic solutions. Oxidants like hypochlorites and ferricyanides in association with H2O2 produced the greatest CL in early work [37,38], but since then many other species have been identified that degrade luminol with concomitant light emission [30,34,39-42]. Many ions (e.g., Fe2+) and compounds that are able to form reactive oxygen species (ROS) in aqueous solution are potentially suitable for determination by flow-injection analysis by chemiluminescence detection (FIA-CL) using luminol. In Scheme 1, aqueous Fe2+ ‘catalyzes’ the second step of the reaction scheme [34]. Actually, Fe2+ is oxidized, and is therefore not a true catalyst, however it is a bystander in the chemiluminescence of luminol here.
An FIA-CL instrument has been developed (FeLume – Waterville Analytical, Waterville, ME) that may be configured to determine several analytes (Co2+, Cu2+, Fe2+, Cr2+, NO3-, PO43- and H2O2). The FeLume has been used specifically to determine sub-nanomolar concentrations of ferrous iron in both marine and freshwaters [30,42,43].
When an iron-containing sample is free of organic matter, the relationship between chemiluminescence and [Fe(aq)2+] is approximately linear between 1 – 1,000 nM Fe(aq)2+. In the presence of fulvic acid, the linear dynamic range (LDR) is reduced to 1 – 32 nM [34]. It has been suggested that FIA-CL analysis of freshwater samples will not work well due to interferences by dissolved organic carbon (DOC) [44], while in coastal seawater, O'Sullivan et al. [32] found that DOM reduced the sensitivity of FIA-CL analysis of Fe(II). Similar results were obtained by Ussher et al. [36] in their evaluation of the effect of model ligands on Fe(II) analysis in seawater. Recently, researchers demonstrated other potential interferences that may occur with redox-active metals that produce CL of luminol [45] or species that interfere with the peroxy-luminol reaction leading to CL (step 2 in Scheme 1) [46].
Coordination of Fe2+ by organic chelators and low pH both contribute to stabilizing iron against oxidation by O2 [47-49]. Such stabilization may either depress or enhance the CL and resultant signal returned by the FeLume. Tight coordination of Fe2+ by organic species that persist in the mixing chamber of the FeLume results in lowering of the signal due to slower formation of the ROS required for CL of luminol. Low pH may produce a higher signal by slowing pre-injection oxidation of Fe2+(aq), yielding more H2O2 in the mixing chamber. Species that have strong affinity for ROS like ascorbate also act to suppress the CL of luminol by scavenging radicals necessary for step 2 in the mechanism shown in Scheme 1.
Typical injection peaks (Figure 1) from this work demonstrate that certain organic compounds reduce the sensitivity of ferrous iron quantitation. The “doublet” peak shown in Figure 1 (typical at nanomolar [Fe2+]) is due to the acid in the samples overcoming the buffer capacity of the luminol solution. Lower pH decreases the signal by reducing luminol dehydrogenation (step 1 in the mechanism in Scheme 1) at nanomolar [Fe2+], but otherwise does not alter the relationship between signal and [Fe2+]. The changes in FeLume response we observed in our initial work with determination of [Fe2+] in the presence of organic matter served as motivation for the present study in which we systematically evaluated the performance of the FeLume in analysis of both natural and model water systems containing Fe(II) and organic species.

2. Experimental

2.1. Reagents and Samples

All chemicals (except H2SO4) were reagent grade. Oxalic acid dihydrate and ferrous ammonium sulfate hexahydrate were purchased from J.T. Baker Chemical Co. (Phillipsburg, NJ, USA). L-Ascorbic acid, glycine, and hydroxylamine hydrochloride were supplied by Fisher Scientific (Fair Lawn, NJ, USA). Hydrazine dihydrochloride and L-cysteine were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sulfuric acid, veritas, redistilled was acquired from GFS Chemicals (Columbus, OH, USA). Suwannee River humic and fulvic acid standards were purchased from the International Humic Substance Society (IHSS. St. Paul, MN, USA).
Natural water samples collected from a mountain stream (Middle Crow Creek) and an unnamed alpine lake, both in SE Wyoming, USA, were acidified to pH 3, stored in the dark at room temperature, and analyzed within three weeks of collection. Middle Crow Creek is an undeveloped watershed at about 2,400 m elevation that originates near Pole Mountain in the Laramie Range in SE Wyoming, USA. This area is impacted by livestock grazing, nearby motorized traffic and human recreation (fishing and hiking). There is significant input of organic matter from overhanging trees and streamside bushes. Our research group has studied the site for several years.
The small alpine lake is located in the Snowy Range of SE Wyoming at ∼ 3,300 m above sea level. At this elevation there is little organic input from trees and shrubs, but there are grasses and other vegetation along the lakeshore. Human impact on this lake is limited to nearby camping and hiking; there are no anglers, and cattle are excluded from the area.
All solutions were prepared with 18.2 MΩ Millipore reverse-osmosis, de-ionized (RO) water and H2SO4. All samples were acidified to pH 3 [50], which slows the oxidation of ferrous iron [32,51,52]. Iron standards, including those used in experiments with various organic amendments, were made by serial dilution of a 400 μM stock solution of ferrous ammonium sulfate hexahydrate (Fe(NH4)2(SO4)2·6H2O) in 0.1 M H2SO4. The concentrations of organic amendments (10-2 M, 10-4 M and 10-6 M) are similar to those used in dissolution and reduction experiments involving iron (hydr)oxide minerals and colloids [7,10,53-55], and organic carbon concentrations for the humic and fulvic acid experiments were those typical of surface fresh waters [56] and were added for final concentrations of 1, 5, or 10 mg C L-1.
Ferrous iron was added in standard additions (1 – 100 nM) to samples amended with select organic compounds or containing natural organic matter (NOM). Certain organic species are used in exploring the biogeochemistry of iron because they promote the reductive dissolution of iron and so are models for natural reductive dissolution processes. We have selected compounds that are involved in the coordination of iron and/or reducing Fe(III) centers at water-mineral interfaces. Response curves generated from Fe2+-spiked experimental samples measured by FIA-CL (FeLume) were directly compared to the response of Fe2+ standards in RO water at pH 3. The experimental samples were made from five commonly used iron-complexing and iron-reducing species at three concentrations in pH 3 RO water or in water samples collected from a mountain stream and an alpine lake as described above.

2.2. Instrumentation and Glassware

All glassware was cleaned thoroughly, sequentially acid (10% vHCl/vH2O) and base washed (0.1 M NaOH or KOH) for a minimum of 24 hours, rinsed thoroughly with reverse-osmosis (RO) water, and air-dried overnight. The FeLume was used as configured by the manufacturer for ferrous iron determination. The instrument was interfaced with a PC running LabView v.6.1 (National Instruments Corp., Austin, TX, USA) with a USA-49W USB 4-port serial adaptor (Keyspan, Richmond, CA, USA). A Dynamax RP-1 peristaltic pump (Rainin Instruments, Oakland, CA, USA) delivered (at 20 rpm) the carrier, luminol, and sample solutions through Tygon (Cole Parmer Instrument Co., Vernon Hills, IL, USA) and Teflon FEP tubing (Upchurch Scientific, Inc., Oak Harbor, WA, USA) to the reaction chamber of the FeLume. Samples were injected by a 0.1 mM Na2CO3 carrier solution where they mixed with a continuous flowing buffered solution (NH3/NH4Cl at pH 10) of luminol (0.5 mM). Upon mixing, Fe(II) is rapidly (milliseconds) oxidized forming the reactive oxygen species (ROS), superoxide (O2-), in the solution, which catalyzes the multi-step oxidation of luminol that produces 426 nm light (see Scheme 1 above) This light is detected using a Hamamatsu HC135 photon counter (PMT) whose output voltage is proportional to [Fe(aq)2+]. The integral of the PMT signal plotted versus Fe(II) concentration of standards provided linear plots in the range of 1 to 100 nM Fe2+.
Each data point is the average of five replicates and as such error bars on plots denote 95% confidence limits (tsn-2), where t is from student's t-table, s is the standard deviation of the five measurements, and n is the number of measurements. The slope of the linear response curve of each experiment was then divided by the slope of the ferrous iron calibration curve, providing normalized slope (mN) that was used as a parameter of effectiveness of the analytical method (Tables 1 and 2). Each experiment was run twice to yield an average mN for each experimental system.

3. Results and Discussion

The FeLume exhibited slight differences in signal response from day to day that may be attributed to luminol aging, pump tube stretching, or variation in PMT output. Since we did not optimize the signal-to-noise ratio before each experiment [42,50], standard calibration response curves were collected for each individual experiment. Linear calibration curves were generated for comparison purposes only, and were not used to calculate [Fe2+](aq) in the amended samples. These comparisons enabled us to evaluate the extent of any interference by organic species in such iron determinations.

3.1. Chelators and Reductants

Ascorbate and oxalate, both separately and in tandem, are routinely used in mineral dissolution kinetic studies both alone and in combination at concentrations ranging from 10-5 to 10-3 M [7,10,55]. Cysteine and hydroxylamine hydrochloride have been used in similar studies at 10-2 M [53,54]. Glycine was included in this study as a comparison to cysteine, which was used by Seitz and Hercules to assess the effect on coordination on the CL of luminol [41]. We also included another commonly used reductant, hydrazine, because it is a reagent we use in a spectrophotometric method for total aqueous iron determination [57].

3.1.1. Ascorbate and oxalate

Though ascorbate has been reported to enhance the CL of luminol in at least one system [58], at all of the concentrations that we evaluated ascorbate quenched the CL of luminol (10-2 M ascorbate plotted in Figure 3a).
The response was near zero across the entire range of Fe2+ concentrations (1 – 100 nM) and produced normalized slopes that were negative or very small (ca. 10-4). The effect of oxalate addition on the FeLume response was negligible at oxalate concentrations of 10-4 and 10-6 M (data in Table 1) with mN values near unity. At 10-2 M oxalate, however, the signal response was double the standard response. In the combined ascorbate/oxalate system, the quenching effect of ascorbate on CL apparently limited the sensitivity at 10-2 M and 10-4 M by producing a flat response like that observed with the ascorbate alone (see Figure 3a). At 10-6 M ascorbate/oxalate, the response was only slightly diminished (∼ 0.64), which should not be problematic for [Fe2+] determinations.

3.1.2. Cysteine and glycine

Cysteine has been reported to enhance the CL of luminol even though a decreased luminol CL signal usually results when antioxidants are injected into an oxidant stream [58]. Wheatley and coworkers saw an increase in the CL of luminol [40]; however, the reductant they used was cobalt(II) rather than iron(II), which may account for the observed difference from our results. At 10-2 M cysteine, the FeLume response was near zero throughout the 1 – 100 nM Fe2+ range, making this analytical method useless at high cysteine concentrations. The slope of the line from the 10-4 M cysteine experiment approached 40% of the standard response, suggesting ferrous iron determination is possible in solutions at that cysteine concentration. The lowest cysteine concentration at 10-6 M, exhibited a dramatic enhancement in response with an mN of ∼ 2.6 (Figure 3b).
Although Seitz and Hercules [41] reported a decrease in CL as a function of glycine concentration, glycine at 10-2 M and 10-4 M had little effect on the FeLume response in our experiments; in fact, glycine enhanced the FeLume response at 10-6 M (see Table 1). The only difference in their work appears to be in their buffer – KOH/H3BO4 versus NH3/NH4Cl used here. Our data suggest that glycine does not adversely affect Fe2+ determinations using the FeLume.

3.1.3. Hydroxylamine and hydrazine

Hydroxylamine at 10-2 M had a flat response curve (not CL quenching) with a significant signal that remained relatively unchanged throughout the [Fe2+] range. The experiments at 10-4 and 10-6 M produced mN values between 0.40 – 0.45, representing a slight attenuation in CL response that would not make for inaccurate [Fe2+] determinations. The effect hydrazine had on the FeLume response appears to be a direct function of concentration. At the highest concentration, the signal was greatly attenuated (mN = 0.082), while at 10-4 and 10-6 M (Figure 3c), the FeLume exhibited slight reduction in sensitivity with mN values of 0.25 and 0.75, respectively (Table 1).

3.2. Natural Waters, Humic and Fulvic Acids

We assessed the effectiveness of the FeLume in the analysis of both natural waters and model water systems (Table 2), all of which contained or were amended with DOM, humic and/or fulvic acids. The samples from Middle Crow Creek (MCC) and the small alpine lake (SR lake) contained between 5 and 10 mg C L-1 DOM and had aqueous Fe2+ concentrations that ranged from 1 to 8 μM.
The concentrations of DOM in our experimental solutions were chosen based on published values for carbon concentration (mg C L-1) [46]. We chose three concentrations of carbon amendments (1, 5 and 10 mg C L-1) and performed the analyses and comparisons as were done for the organic ligands.

3.2.1. Humic and fulvic acids

In the experiment using Aldrich humic acid (lot # 03130JS), we found that at 5 and 10 mg C L-1 the CL response was attenuated with mN values between 0.35 – 0.37 (Figure 4a). O'Sullivan et al. [32] reported a similar reduction in sensitivity in waters containing high DOC.
Our experiments with this lot of humic acid were probably less sensitive to increasing [Fe2+] because of its high iron content (7,700 mg L-1), resulting in an elevated signal of the blank, while the flat response indicated interference by the humic material. The solution containing the least humic acid (1 mg C L-1) also exhibited reduced sensitivity (mN ∼ 0.7), but still allowed [Fe2+] determination.
The IHSS humic acid (HA) and fulvic acid (FA) systems exhibited behaviors similar to the Aldrich HA just discussed (Figure 4b). In both HA and FA experiments, the two higher concentrations (5 and 10 mg C L-1) had lower sensitivities (mN ∼ 0.17 to 0.19, respectively) than those of Aldrich HA. Both response curves had y-intercepts much closer to zero than the corresponding Aldrich HA curves, indicating lower iron content in the much purer IHSS acids. The smaller normalized slopes (mN) are likely due to greater complexation of aqueous Fe2+. Complexation of Fe2+ with DOM should lower its reactivity and decrease the iron-catalyzed generation of ROS needed for CL of the luminol in this system. The response of both humic and fulvic acid systems at 1 mg C L-1 is only slightly depressed (mN values of 0.77 and 0.83), indicating that at low concentrations this analytical method is still useful.

3.2.2. Natural water samples

The natural water samples exhibited divergent results. The response of the MCC samples appeared negative with respect to iron concentration. High iron concentrations in the MCC sample with the addition of Fe2+ spikes exceeded the LDR of the method, thus making this analytical tool incapable of [Fe2+] determination without dilution of the sample. The FeLume exhibits non-linear (sigmoidal) behavior as the PMT nears its over-voltage limit. The SR lake sample (Figure 4c) had a positive response with added ferrous iron (mN of 0.86), suggesting this technique is useful for analysis at this site. Analyses of waters at both sites indicate high aqueous iron and total organic carbon (TOC); the high iron concentrations explain the elevated blank signal for both the lake and creek response curves, but with similar TOC content the divergent responses are difficult to explain without further analysis of the systems. There may be a difference in the type or quality of organic matter input, leading to variable coordination affinity for iron or perhaps different radical scavenging ability.

3.3. Method Evaluation

In determining the effectiveness of the FeLume technique for Fe2+ analysis in natural waters and in systems containing organic compounds, mN values of 0.9 or greater were those we considered fully effective; mN values between 0.1 and 0.9 were still effective but at reduced sensitivity. Values less than 0.1 are not recommended for use without verifying the sensitivity as appropriate for the iron concentration expected.
The CL quenching seen in the ascorbate systems may be due to competition for radicals by ascorbate [59,60], but iron coordination by ascorbate probably contributes by reducing CL through ligand stabilization of Fe(II) against oxidation. As reported previously, iron-coordinating species appear to interfere with the iron-catalyzed CL of luminol [30,32,41]. Coordination of iron by ascorbate - or more complex organic matter - stabilizes it against oxidation by dissolved oxygen and subsequent formation of ROS required for the production of CL from luminol [30,32,47,49].
Ascorbate is such an effective quencher of luminol CL by ferrous iron that, at 10-2 and 10-4 M in the combined ascorbate/oxalate system, any signal enhancement due to oxalate is overcome; however, at low concentrations (i.e., 10-6 M) of both coordinator and reductant, the response appears to be manageable. At these concentrations, mN is depressed only ∼0.3 from the oxalate system, suggesting dissolution experiments of iron oxides involving ascorbate and oxalate might be feasible. Determination of [Fe2+] by the FeLume is mostly unencumbered in solutions containing oxalate up to 10-4 M; in fact, this method appears to be highly sensitive in oxalate solutions of 10-2 M. Signal enhancement may be due to slight pH lowering of the sample solution by oxalate amendment further stabilizing Fe2+ against oxidation prior to injection - higher [Fe2+] result in greater CL from the mixing chamber.
Discussions of mechanisms in the augmentation of signal by cysteine is speculative, but others have reported CL of luminol by cysteine [58] and thus our results may reflect a simple additive effect. It is possible that oxidation of cysteine by dissolved oxygen may generate superoxide. This alternate source of superoxide - one of the ROS associated with the reaction that generates CL of luminol - is likely to enhance the signal. Regardless of the mechanism of signal enhancement, the FeLume is an effective tool for measuring [Fe2+] in solutions containing cysteine concentrations up to 10-4 M, but should be avoided at higher concentrations, as samples containing significant cysteine appear to result in anomalously high [Fe2+] by this method.
Previous reports of depressed CL of luminol in the presence of glycine [41] might be attributed to the difference in buffer types; however, further study is required to verify this discrepancy. With the exception of glycine at 10-4 M (mN ∼ 0.87), there appears to be an enhancement in the CL response by this amino acid. This may be attributed to glycine amendment pH lowering as discussed with oxalate above. Since the normalized slope for the experiment at 10-4 M glycine is only slightly less than our somewhat arbitrary cutoff for mN (0.90), this appears to be an effective method in the presence of glycine at these concentrations.
The normalized slopes determined from the experiments containing the reductants hydroxylamine and hydrazine up to 10-4 M suggest the FeLume is suitable for Fe2+ determination. Hydroxylamine and hydrazine at 10-2 M concentration quench the CL of luminol to such an extent that [Fe2+] quantitation is impractical or impossible.
Care must be taken in measuring Fe2+(aq) with the FeLume in samples with potentially high levels of iron complexed to NOM, as iron tends to dissociate over time leading to higher “free” iron concentrations. Often DOM interferes with CL of luminol by Fe2+, but sometimes the presence of DOM does not affect the sensitivity of the method. The natural water sample from the alpine lake contains significant iron, causing an offset in response (large positive y-intercept), but the experimental response curve is nearly parallel to that of the calibration curve indicating little effect on method sensitivity. The samples from MCC, however, quench CL to the point that there is no correlation between added iron and the instrumental response. At the lower concentration range typical of natural surface waters (less than 5 mg C L-1), humic and fulvic acids may slightly depress the sensitivity of the method, but do not preclude its use.

3.4. Recommendations

The results of experiments with ascorbate indicate that it renders the method ineffectual at the concentrations used. Ascorbate is considered such an effective iron chelator and reductant of iron (hydr)oxides [7,10,55] that lower concentrations may be used in model dissolution studies and should be tested for interference with the FeLume response at those concentrations. Oxalate either does not dramatically affect the response or enhances its sensitivity (seen at high concentrations), and we recommend its use with the FeLume during ferrous iron analyses. Cysteine may be used at ∼ 10-6 M with the FeLume, but starts to diminish in sensitivity as the concentration increases to 10-4 M or above. We recommend this method in systems containing glycine at concentrations between 10-6 and 10-2 M using the NH3/NH4Cl buffer system, as it appears to not affect the CL of luminol by Fe2+. The results for these two amino acids suggest pursuing similar studies with other amino acids capable of reductive dissolution of iron minerals. The reductants hydroxylamine and hydrazine may be used at concentrations in the range of 10-6 M, but like cysteine, they start to reduce the CL signal at higher concentrations.
Measurements of ferrous iron in samples containing high DOM concentrations are generally not facile and should be avoided. We discourage the direct use of the FeLume on samples where high concentrations of iron may be coordinated with DOM - in such systems colorimetry may be sufficient. Middle Crow Creek water, and samples containing Aldrich HA at similar carbon content, both had such high levels of iron and DOM that this analytical method was rendered ineffective. Samples containing IHSS HA and FA up to 10 mg C L-1 and Aldrich HA up to 5 mg C L-1 can be effectively analyzed for ferrous iron by the FeLume. Other, more pristine natural water sites have water suitable for Fe2+ quantitation using the FeLume, as our site in the Snowy Range showed, but its effectiveness should be determined on a site-by-site basis. All samples containing Fe(III)-organic complexes (especially oxalate) should be kept dark prior and during analysis to ensure photoreduction of these complexes does not result in altering [Fe2+] determination [35,61-65].
In all cases in the determination of [Fe2+], the response of the FeLume should be checked by comparing iron-spiked samples with iron standards. This will provide a baseline response and indicate whether the FeLume needs signal-to-noise optimization by fine-tuning the PMT voltage or by varying luminol concentration.

Acknowledgments

The authors would like to gratefully acknowledge support from the National Science Foundation (EAR-0434019) and the U.S. Department of Energy (DE-FG02-06ER15823) for this work. The content of this paper represents the views of the authors and not of the above named agencies.

References

  1. Millward, G.E.; Moore, R.M. The adsorption of Cu, Mn and Zn by iron oxyhydroxide in model estuarine solutions. Water Res. 1982, 16, 981–985. [Google Scholar]
  2. Tessier, A.; Rapin, F.; Carignan, R. Trace metals in oxic lake sediments: Possible adsorption onto iron oxyhydroxides. Geochim. Cosmochim. Acta 1985, 49, 183–194. [Google Scholar]
  3. Ehrlich, H.L. Geomicrobiology; Marcel Dekker, Inc.: New York, NY, USA, 1990. [Google Scholar]
  4. Cox, C.D. Importance of Iron in Bacterial Virulence. In Metal Ions and Bacteria; Beveridge, T.J., Doyle, R.J., Eds.; John Wiley & Sons: New York, NY, USA, 1989. [Google Scholar]
  5. Harrison, P.M.; Hoare, R.J. Metals in Biochemistry; Chapman and Hall: New York, NY, USA, 1980. [Google Scholar]
  6. Afonso, M.D.S.; Morando, P.J.; Blesa, M.A.; Banwart, S.; Stumm, W. The reductive dissolution of iron oxides by ascorbate. The role of carboxylate anions in accelerating reductive dissolution. J. Colloid Interface Sci. 1990, 138, 74–82. [Google Scholar]
  7. Suter, D.; Siffert, C.; Sulzberger, B.; Stumm, W. Catalytic dissolution of iron(III) (hydr)oxides by oxalic acid in the presence of Fe(II). Naturwissenschaften 1988, 75, 571–573. [Google Scholar]
  8. Suter, D.; Banwart, S.; Stumm, W. Dissolution of hydrous iron(III) oxides by reductive mechanisms. Langmuir 1991, 7, 809–813. [Google Scholar]
  9. Torres, R.; Blesa, M.A.; Matijevic, E. Interactions of metal oxides with chelating agents. IX. Reductive dissolution of hematite and magnetite by aminocarboxylic acids. J. Colloid Interface Sci. 1990, 134, 475–485. [Google Scholar]
  10. Zinder, B.; Gerhard, F.; Stumm, W. The coordination chemistry of weathering: II. Dissolution of Fe(III) oxides. Geochim. Cosmochim. Acta 1986, 50, 1861–1869. [Google Scholar]
  11. Butcher, S.S.; Charlson, R.J.; Orians, G.H.; Wolfe, G.V. Global Biogeochemical Cycles; Academic Press Limited: London, UK, 1992. [Google Scholar]
  12. Brown, D.A.; Sherriff, B.L.; Sawicki, J.A.; Sparling, R. Precipitation of iron minerals by a natural microbial consortium. Geochim. Cosmochim. Acta 1999, 63, 2163–2169. [Google Scholar]
  13. Greenwood, N.N.; Earnshaw, A. Chemistry of the Elements, 2nd ed; Butterworth-Heinemann: Oxford, UK, 2002. [Google Scholar]
  14. Stumm, W.; Morgan, J.J. Aquatic Chemistry - An Introduction Emphasizing Chemical Equilibria in Natural Waters, 2nd ed; John Wiley & Sons: New York, NY, USA, 1981. [Google Scholar]
  15. Pehkonen, S. Determination of the oxidation states of iron in natural waters: A review. Analyst 1995, 120, 2655–2663. [Google Scholar]
  16. Figueiredo, A.A. Determination of iron by EDTA titrimetry of thiocyanate complex. Analyt. Chem. 1971, 43, 484–485. [Google Scholar]
  17. Sun, B.; Ye, Y.; Huang, H.; Bai, Y. Potentiometric determination of iron using a fluoride ion-selective electrode - the application of the Apple II-ISE intelligent ion analyzer. Talanta 1993, 40, 891–895. [Google Scholar]
  18. Jagner, D.; Renman, L.; Stefansdottir, S.H. Determination of iron(III) and titanium(IV) as their solchrome violet RS complexes by constant-current stripping voltammetry. Part 1. Automated single point calibration method for iron(III). Analyt. Chim. Acta 1993, 281, 305–314. [Google Scholar]
  19. Florence, T.M. Trace element speciation by anodic stripping voltammetry. Analyst 1992, 117, 551–553. [Google Scholar]
  20. Badocco, D.; Pastore, P.; Favaro, G.; Macca, C. Effect of eluent composition and pH and chemiluminescent reagent pH on ion chromatographic selectivity and luminol-based chemiluminescence detection of Co2+, Mn2+ and Fe2+ at trace levels. Talanta 2007, 72, 249–255. [Google Scholar]
  21. Schlosser, C.; Croot, P.L. Application of cross-flow filtration for determining the solubility of iron species in open ocean seawater. Limnol. Oceanogr. Methods 2008, 6, 630–642. [Google Scholar]
  22. Hase, U.; Yoshimura, K. Determination of trace amounts of iron in highly purified water by ion-exchanger phase absorptiometry combined with flow analysis. Analyst 1992, 117, 1501–1506. [Google Scholar]
  23. Stookey, L.L. Ferrozine - A new spectrophotometric reagent for iron. Analyt. Chem. 1970, 42, 779–781. [Google Scholar]
  24. Andac, M.; Asan, A.; Isildak, I. A simple flow injection spectrophotometric determination method for iron(III) based on o-acetylsalicylhydroxamic acid complexation. Chem. Papers 2009, 63, 268–273. [Google Scholar]
  25. Xue, H.; Lurdes, M.D.; Goncalves, S.; Reutlinger, M.; Sigg, L.; Stumm, W. Copper(I) in fogwater: Determination and interactions with sulfite. Environ. Sci. Technol. 1991, 25, 1716–1722. [Google Scholar]
  26. Zhuang, G.; Yi, Z.; Duce, R.A.; Brown, P.A. Link between iron and sulphur cycles suggested by detection of Fe(II) in remote marine aerosols. Nature 1992, 355, 537–539. [Google Scholar]
  27. Andersen, J.E. A novel method for the filterless preconcentration of iron. Analyst 2005, 130, 385–390. [Google Scholar]
  28. Waterbury, R.D.; Yao, W.; Byrne, R.H. Long pathlength absorbance spectroscopy: Trace analysis of Fe(II) using a 4.5 m liquid core waveguide. Analyt. Chim. Acta 1997, 357, 99–102. [Google Scholar]
  29. Bowie, A.R.; Achterberg, E.P.; Sedwick, P.N.; Ussher, S.; Worsford, P.J. Real-time monitoring of picomolar concentrations of iron(II) in marine waters using automated flow injection-chemiluminescence instrumentation. Environ. Sci. Technol. 2002, 36, 4600–4607. [Google Scholar]
  30. Croot, P.L.; Laan, P. Continuous shipboard determination of Fe(II) in polar waters using flow injection analysis with chemiluminescence detection. Analyt. Chim. Acta 2002, 466, 261–273. [Google Scholar]
  31. de Jong, J.T.M.; den Das, J.; Bathmann, U.; Stoll, M.H.C.; Kattner, G.; Nolting, R.F.; de Baar, H.J.W. Dissolved iron at subnanomolar levels in the Southern Ocean as determined by ship-board analysis. Analyt. Chim. Acta 1998, 377, 113–124. [Google Scholar]
  32. O' Sullivan, D.W.; Hanson, A.K., Jr.; Kester, D.R. Stopped flow luminol chemiluminescence determination of Fe(II) and reducible iron in seawater at subnanomolar levels. Mar. Chem. 1995, 49, 65–77. [Google Scholar]
  33. Qin, W.; Zhang, Z.J.; Wang, F.C. Chemiluminescence flow system for the determination of Fe(II) and Fe(III) in water. Fresenius J. Anal. Chem. 1998, 360, 130–132. [Google Scholar]
  34. Rose, A.L.; Waite, T.D. Chemiluminescence of luminol in the presence of iron(II) and oxygen: Oxidation mechanism and implications for its use. Analyt. Chem. 2001, 73, 5909–5920. [Google Scholar]
  35. Nussbaum, M.A.; Nekimken, H.L.; Nieman, T.A. Luminol chemiluminescence for determination of iron(II) in ferrioxalate chemical actinometry. Analyt. Chem. 1987, 59, 211–212. [Google Scholar]
  36. Ussher, S.J.; Yaqoob, M.; Achterberg, E.P.; Nabi, A.; Worsfold, P.J. Effect of model ligands on iron redox speciation in natural waters using flow injection with luminol chemiluminescence detection. Analyt. Chem. 2005, 77, 1971–1978. [Google Scholar]
  37. Albrecht, H.O. Chemiluminescence of aminophthalic hydrazide. Z. Physikal. Chem. 1928, 135, 321–330. [Google Scholar]
  38. Stross, F.H.; Branch, G.E.K. The chemiluminescence of 3-aminophthalhydrazide. J. Org. Chem. 1938, 3, 385–404. [Google Scholar]
  39. White, E.H.; Roswell, D.F. Luminol Chemiluminescence. In Chemi- and Bioluminescence; Burr, J. G., Ed.; Marcel Dekker, Inc.: New York, NY, USA, 1985; pp. 215–244. [Google Scholar]
  40. Wheatley, R.A.; Sariahmetoglu, M.; Cakici, I. Enhancement of luminol chemiluminescence by cysteine and glutathione. Analyst 2000, 125, 1902–1904. [Google Scholar]
  41. Seitz, W.R.; Hercules, D.M. Determination of trace amounts of iron(III) using chemiluminescence analysis. Analyt. Chem. 1972, 44, 2143–2149. [Google Scholar]
  42. King, D.W.; Lounsbury, H.A.; Millero, F.J. Rates and mechanism of Fe(II) oxidation at nanomolar total iron concentrations. Environ. Sci. Technol. 1995, 29, 818–824. [Google Scholar]
  43. Emmenegger, L.; King, D.W.; Sigg, L.; Sulzberger, B. Oxidation kinetics of Fe(II) in a eutrophic Swiss lake. Environ. Sci. Technol. 1998, 32, 2990–2996. [Google Scholar]
  44. Pullin, M.J.; Cabaniss, S.E. Colorimetric flow-injection analysis of dissolved iron in high DOC waters. Water Res. 2001, 35, 363–372. [Google Scholar]
  45. Hopkinson, B.M.; Barbeau, K.A. Organic and redox speciation of iron in the eastern tropical North Pacific suboxic zone. Mar. Chem. 2007, 106, 2–17. [Google Scholar]
  46. Geng, H.; Meng, Z. Inhibition of superoxide dismutase, vitamin C and glutathione on chemiluminescence produced by luminol and the mixture of sulfite and bisulfite. Spectrochim. Acta A: Mol. Biomol. Spectrosc. 2006, 64, 87–92. [Google Scholar]
  47. Theis, T.L.; Singer, P.C. Complexation of iron(II) by organic matter and its effect on iron(II) oxygenation. Environ. Sci. Technol. 1974, 8, 596–573. [Google Scholar]
  48. Croot, P.L.; Frew, R.D.; Sander, S.; Hunter, K.A.; Ellwood, M.J.; Pickmere, S.E.; Abraham, E.R.; Law, C.S.; Smith, M.J.; Boyd, P.W. Physical mixing effects on iron biogeochemical cycling: FeCycle experiment. J. Geophys. Res. Oceans 2007, 112, C06015. [Google Scholar] [CrossRef]
  49. Santana-Casiano, J.M.; Davila, M.G.; Rodriguez, M.J.; Millero, F.J. The effect of organic compounds in the oxidation kinetics of Fe(II). Mar. Chem. 2000, 70, 211–222. [Google Scholar]
  50. King, D.W. Personal communication. 2005. [Google Scholar]
  51. Cotton, F.A.; Wilkinson, G.; Murillo, C.A.; Bochmann, M. Advanced Inorganic Chemistry, 6th ed; John Wiley & Sons, Inc.: New York, NY, USA, 1999. [Google Scholar]
  52. Barry, R.C.; Schnoor, J.L.; Sulzberger, B.; Sigg, L.; Stumm, W. Iron oxidation kinetics in an acidic alpine lake. Water Res. 1994, 28, 323–333. [Google Scholar]
  53. Amirbahman, A.; Sigg, L.; von Gunten, U. Reductive dissolution of Fe(III) (hydr)oxides by cysteine: kinetics and mechanism. J. Colloid Interface Sci. 1997, 194, 194–206. [Google Scholar]
  54. Holmen, B.A.; Casey, W.H. Hydroxamate ligands, surface chemistry, and the mechanism of ligand-promoted dissolution of geothite [a-FeOOH(s)]. Geochim. Cosmochim. Acta 1996, 60, 4403–4416. [Google Scholar]
  55. Sulzberger, B.; Suter, D.; Siffert, C.; Banwart, S.; Stumm, W. Dissolution of Fe(III) (hydr)oxides in natural waters; Laboratory assessment on the kinetics controlled by surface coordination. Mar. Chem. 1989, 28, 127–144. [Google Scholar]
  56. Drever, J.I. The Geochemistry of Natural Waters: Surface and Groundwaters Environments, 3rd ed; Prentice-Hall, Inc.: Upper Saddle River, NJ, USA, 1982. [Google Scholar]
  57. Borman, C.J.; Colberg, P.J.S.; Eggleston, C.M.; Sullivan, B.P. Is iron redox cycling in a high altitude watershed photochemically or thermally driven? Chem. Geol. 2009. in revision. [Google Scholar]
  58. Kubo, H.; Toriba, A. Chemiluminescence flow injection analysis of reducing agents based on the luminol reaction. Analyt. Chim. Acta 1997, 353, 345–349. [Google Scholar]
  59. Rose, R.C. Ascorbic acid metabolism in protection against free radicals: a radiation model. Biophys. Res. Commun. 1990, 169, 430–436. [Google Scholar]
  60. Soares, D.G.; Andreazza, A.C.; Salvador, M. Sequestering ability of butylated hydroxytoluene, propyl gallate, resveratol and vitamins C and E against ABTS, DPPH, and hydroxyl free radicals in chemical and biological systems. J. Agric. Food Chem. 2003, 51, 1077–1080. [Google Scholar]
  61. Demas, J.N.; Bowman, W.D.; Zalewski, E.F.; Velapoldi, R.A. Determination of the quantum yield of the ferrioxalate actinometer with electrically calibrated radiometers. J. Phys. Chem. 1981, 85, 2766–2771. [Google Scholar]
  62. Hatchard, C.G.; Parker, C.A. A new sensitive chemical actinometer. II. Potassium ferrioxalate as a standard chemical actinometer. Proc. Royal Soc. 1956, 235, 518–536. [Google Scholar]
  63. Zuo, Y.; Hoigne, J. Formation of hydrogen peroxide and depletion of oxalic acid in atmospheric water by photolysis of iron(III)-oxalato complexes. Environ. Sci. Technol. 1992, 26, 1014–1022. [Google Scholar]
  64. Zuo, Y.G.; Deng, Y.W. Iron(II)-catalyzed photochemical decomposition of oxalic acid and generation of H2O2 in atmospheric liquid phases. Chemosphere 1997, 35, 2051–2058. [Google Scholar]
  65. Zuo, Y.G.; Hoigne, J. Photochemical decomposition of oxalic, glyoxalic and pyruvic acid catalysed by iron in atmospheric waters. Atmos. Environ. 1994, 28, 1231–1239. [Google Scholar]
Sensors 09 04390f1 1024
Figure 1. Injection peaks showing typical response generated from nanomolar [Fe2+] in this work. A. 100 nM Fe2+; B. 100 nM Fe2+ with 1 mg C L-1 fulvic acid (FA); C. 10 nM Fe2+; D. 10 nM Fe2+ with 1 mg C L-1 FA. Inset: Typical Gaussian response curve generated from injection of higher concentration Fe2+ (44.64 mM) than used in this study. The higher concentration apparently produces enough O2- to overcome the effect low pH has on the signal.
Figure 1. Injection peaks showing typical response generated from nanomolar [Fe2+] in this work. A. 100 nM Fe2+; B. 100 nM Fe2+ with 1 mg C L-1 fulvic acid (FA); C. 10 nM Fe2+; D. 10 nM Fe2+ with 1 mg C L-1 FA. Inset: Typical Gaussian response curve generated from injection of higher concentration Fe2+ (44.64 mM) than used in this study. The higher concentration apparently produces enough O2- to overcome the effect low pH has on the signal.
Sensors 09 04390f1
Sensors 09 04390f2 1024
Figure 2. FeLume sample loading and injection configurations. In both configurations, the luminol (line A) runs directly to the mixing chamber. In the 90-second sample loading configuration (top), the carrier solution (line B) runs to waste, and the sample (line C) charges the loop (pre-injection). In the injection configuration (50 s.), the carrier solution runs through the sample loop, rapidly moving the sample into the mixing chamber where the resulting CL is detected by the PMT (all arrows run to waste container).
Figure 2. FeLume sample loading and injection configurations. In both configurations, the luminol (line A) runs directly to the mixing chamber. In the 90-second sample loading configuration (top), the carrier solution (line B) runs to waste, and the sample (line C) charges the loop (pre-injection). In the injection configuration (50 s.), the carrier solution runs through the sample loop, rapidly moving the sample into the mixing chamber where the resulting CL is detected by the PMT (all arrows run to waste container).
Sensors 09 04390f2
Sensors 09 04390f3 1024
Figure 3. Three FeLume response comparisons (vs. Fe2+ only standards) of selected organic compounds demonstrating: a). signal quenching of 10-2 M ascorbate (negative integration values represent artifact due to subtraction from baseline); b). signal enhancement of 10-6 M cysteine; and c). minimal signal quenching of 10-6 M hydrazine.
Figure 3. Three FeLume response comparisons (vs. Fe2+ only standards) of selected organic compounds demonstrating: a). signal quenching of 10-2 M ascorbate (negative integration values represent artifact due to subtraction from baseline); b). signal enhancement of 10-6 M cysteine; and c). minimal signal quenching of 10-6 M hydrazine.
Sensors 09 04390f3
Sensors 09 04390f4 1024
Figure 4. Three FeLume response comparisons (vs. Fe2+ standards) of natural water and natural organic matter: a). Aldrich humic acid; b). IHSS fulvic acid c). sample from SR lake.
Figure 4. Three FeLume response comparisons (vs. Fe2+ standards) of natural water and natural organic matter: a). Aldrich humic acid; b). IHSS fulvic acid c). sample from SR lake.
Sensors 09 04390f4
Sensors 09 04390f5 1024
Scheme 1. The general reaction of luminol to produce light for determination of aqueous Fe(II). Fe2+(aq) ‘catalyzes’ the second step in this reaction scheme. The light emitted after the third step proportional to [Fe2+(aq)] within a certain concentration range and is subject to changes in sensitivity that depend on the presence of species that can complex Fe2+(aq) and/or impact the generation of H2O2.
Scheme 1. The general reaction of luminol to produce light for determination of aqueous Fe(II). Fe2+(aq) ‘catalyzes’ the second step in this reaction scheme. The light emitted after the third step proportional to [Fe2+(aq)] within a certain concentration range and is subject to changes in sensitivity that depend on the presence of species that can complex Fe2+(aq) and/or impact the generation of H2O2.
Sensors 09 04390f5
Table
Table 1. Organic chelators and reductants.
Table 1. Organic chelators and reductants.
AmendmentConcentrationmN aEffectiveness
Ascorbate10-6 M< 0.1NR
10-4 M< 0.1NR
10-2 M< 0.1NR
Oxalate10-6 M0.9(0)Yes
10-4 M1.1Yes
10-2 M2.(0)Yes
Ascorbate/oxalate10-6 M0.6(4)LS
10-4 M< 0.1NR
10-2 M< 0.1NR
Cysteine10-6 M2.(6)Yes
10-4 M0.3(7)LS
10-2 M< 0.1NR
Glycine10-6 M1.(8)Yes
10-4 M0.8(8)LS
10-2 M1.3Yes
Hydroxylamine10-6 M0.4(5)LS
10-4 M0.4(2)LS
10-2 M< 0.1NR
Hydrazine10-6 M0.7(5)LS
10-4 M0.2(5)LS
10-2 M< 0.1NR
Notes: a – ± 6.4 %; NR – not recommended; LS – lowered sensitivity
Table
Table 2. Dissolved organic matter and natural water samples.*
Table 2. Dissolved organic matter and natural water samples.*
AmendmentConcentrationmN aEffectiveness
Aldrich Humic Acid1 mg C L-10.7(0)LS
5 mg C L-10.3(6)LS
10 mg C L-10.3(7)LS
IHSS Humic Acid1 mg C L-10.8(4)LS
5 mg C L-10.1(7)LS
10 mg C L-10.1(9)LS
IHSS Fulvic Acid1 mg C L-10.7(7)LS
5 mg C L-10.1(9)LS
10 mg C L-10.1(7)LS
MCC water∼ 5 mg C L-1 *<0.1NR
SR lake water∼ 10 mg C L-1 *0.8(6)LS
Notes: a – ± 6.4 %; NR – not recommended; LS – lowered sensitivity;
*Natural waters organic carbon content measured previously by TOC analysis.

Share and Cite

MDPI and ACS Style

Borman, C.J.; Sullivan, B.P.; Eggleston, C.M.; Colberg, P.J.S. The Use of Flow-Injection Analysis with Chemiluminescence Detection of Aqueous Ferrous Iron in Waters Containing High Concentrations of Organic Compounds. Sensors 2009, 9, 4390-4406. https://doi.org/10.3390/s90604390

AMA Style

Borman CJ, Sullivan BP, Eggleston CM, Colberg PJS. The Use of Flow-Injection Analysis with Chemiluminescence Detection of Aqueous Ferrous Iron in Waters Containing High Concentrations of Organic Compounds. Sensors. 2009; 9(6):4390-4406. https://doi.org/10.3390/s90604390

Chicago/Turabian Style

Borman, Christopher J., B. Patrick Sullivan, Carrick M. Eggleston, and Patricia J. S. Colberg. 2009. "The Use of Flow-Injection Analysis with Chemiluminescence Detection of Aqueous Ferrous Iron in Waters Containing High Concentrations of Organic Compounds" Sensors 9, no. 6: 4390-4406. https://doi.org/10.3390/s90604390

Article Metrics

Back to TopTop