Grantee Research Project Results

Final Report: Formation of Metal-Phosphonate Complexes and Their Subsequent Chemical Reactions with Mineral Surfaces

EPA Grant Number: R826376
Title: Formation of Metal-Phosphonate Complexes and Their Subsequent Chemical Reactions with Mineral Surfaces
Investigators: Stone, Alan T.
Institution: The Johns Hopkins University
EPA Project Officer: Chung, Serena
Project Period: February 1, 1998 through January 31, 2001
Project Amount: $276,944
RFA: Exploratory Research - Environmental Chemistry (1997) RFA Text |  Recipients Lists
Research Category: Water , Land and Waste Management , Air , Safer Chemicals

Objective:

In recent decades, phosphonate functional groups have appeared in the structures of a growing number of manufactured chemicals. Phosphonate-based chelating agents are used to prevent the formation of undesirable precipitates, and to protect others from dissolution. Such chelating agents are used to depress free metal ion activity and to increase total metal ion concentrations. Other classes of phosphonate-containing chemicals are biologically active; phosphonates are used as pesticides, growth regulators, pharmaceuticals, and sterilants.

The objective of this research project was to obtain the chemical information necessary to predict pathways and rates of chemical reactions involving phosphonate-containing chemicals in aqueous environmental media. Emphasis was placed on: (1) reactions at mineral/water interfaces because of their dominant role in soils, sediments, and aquifers; and (2) reactions with transition metal ions and molecular oxygen, which have the potential to oxidatively transform phosphonate-containing chemicals.

Summary/Accomplishments (Outputs/Outcomes):

Analytical Methodology. To meet the objectives, four different analytical approaches were necessary to identify and monitor concentrations of phosphonate-containing synthetic chemicals and their degradation products. Method 1 yielded total phosphonate concentrations. Filtered samples were digested for 2 hours using potassium peroxodisulfate at 100°C and measured as orthophosphate using the molybdenum blue colorimetric test. Because this method does not provide information about the identity of the phosphonate compounds in the sample, it was only appropriate in single phosphonate systems where degradation did not take place. Method 2 involved the pre-column generation of Fe(III)-phosphonate complexes that are strong enough to remain intact during ion-pair high performance liquid chromatography (HPLC) analysis. Stock Fe(III) solution and mineral acid were added to each sample, and the mixture was allowed to equilibrate. A polymer-based C-18 HPLC column was employed in the separation step, along with an eluent containing acetonitrile, bicarbonate buffer (pH 8.3), and tetrabutylammonium counterion. Using an ultraviolet (UV)/visible detector set at 260 nm, detection limits for this technique were 0.1 µM for NTMP, EDTMP, DTPMP, and 0.5 µM for HEDP. Method 3 employed ion chromatography with conductivity detection and ion suppression. Separation of several NTMP oxidation products was achieved using an anion exchange column and a 30 mM sodium carbonate eluent solution. We analyzed formate using the same column and 5 mM NaOH eluent solution. Method 4 consisted of a capillary electrophoresis (CE-based technique that uses Cu(II) added to the capillary electrolyte solution to convert nonchromophoric phosphonate-containing compounds into Cu(II)-phosphonate complexes that absorb light at 254 nm.

Equilibrium Speciation Calculations. We performed an extensive series of computer calculations to compare and contrast the protonation behavior and metal ion coordination behavior of aminocarboxylate and aminophosphonate chelating agents. Under acidic conditions, phosphonates are considerably more effective chelating agents than corresponding carboxylates. The same holds true for 1:1 metal ion-chelating agent complexes under neutral and alkaline conditions. For complexes involving two or more chelating agents coordinated to a single metal ion, phosphonates perform more poorly than carboxylates, because of steric effects and unfavorable functional group-to-functional group electrostatic interactions. The effects of chelating agents on the redox potential for the Fe(III)/Fe(II) half-reaction also were explored.

Exchange Kinetics in Solution. It is widely appreciated that elemental speciation has a bearing on several remediation strategies and that it is the most effective. Unfortunately, it is often assumed that speciation is under equilibrium control, and this assumption is rarely checked. For this reason, a series of experiments was performed that investigated the Cr(III) kinetics exchange between dissolved chelating agents.

These experiments began by synthesizing one stereoisomer of the 1:2 complex between Cr(III) and the aminocarboxylate chelating agent iminodiacetic acid (IDA). Capillary electrophoresis allowed us to monitor the slow (3 or 4 month) interconversion among different stereoisomers, as well as capture by strong aminocarboxylate and aminophosphonate chelating agents. In 5.0 mM acetate buffer (4.4 < pH < 5.6), aminophosphonate chelating agents captured Cr(III) far more rapidly than aminocarboxylate chelating agents.

Ethylenediaminetetramethylenephosphonic acid (EDTMP), for example, was able to capture 50 percent of the dissolved Cr(III) in approximately 12 days, while ethylenediaminetetraacetic acid (EDTA) required nearly 60 days of reaction. At this acidic pH, timescales on the order of 6 months to a year would be required to achieve equilibrium. Additional Cr(III) exchange experiments performed under neutral conditions (pH 7.2) required longer time periods. Cr(III) speciation in most relevant environmental conditions will occur under kinetic rather than equilibrium control, and analytical methods are required to directly determine speciation. Therefore, predictions based on equilibrium computer calculations will be insufficient.

Adsorption of Benzylphosphonic Acid, Phenylphosphonic Acid, Phenyl Phosphate, and Their Monoesters. Benzylphosphonic acid, phenylphosphonic acid, phenyl phosphate, and their monoesters are oxyanions, comprised of a single Lewis Base group. For all five compounds investigated, the extent of adsorption onto aluminum oxide and titanium dioxide was negligible under strongly alkaline conditions (e.g., above pH 10), but increased as the pH was decreased towards acidic conditions. Compounds that are dianionic when fully deprotonated (benzylphosphonic acid, phenylphosphonic acid, phenyl phosphate) adsorbed much more strongly than compounds that are monoanionic when fully deprotonated. Basicity also is important; compounds with equilibrium constants for protonation/deprotonation above pH 4.0 adsorb more strongly than those that only protonate at pH values less than 2.0. For the dianionic compounds, the observation of significant extents of adsorption near and above the point of zero charge of the mineral surfaces implies that inner-sphere bonds are forming.

Adsorption of Aminopolyphosphonates onto FeOOH(Goethite). A comprehensive set of experiments examined the adsorption of eight phosphonate-containing compounds that are widely employed in technical applications. Emphasis on structure-reactivity relationships makes it possible to predict the adsorption behavior of compounds not included in the study set, and provides important clues regarding mechanisms of adsorption.

Nitrilotrimethylenephosphonic acid (NTMP) is a high-production chelating agent. The extent of adsorption as a function of pH was studied from pH 6.0 to 12.0. As the concentration of sodium nitrate background electrolyte was increased from 1.0 mM to 1.0 M, virtually no change in the extent of NTMP adsorption was observed. This lack of an ionic strength effect was especially surprising when the pH was above 8.0, the point of zero charge for the FeOOH(goethite) surface. Under these conditions, both the surface and the NTMP species predominant in solution are negatively charged; hence, electrostatic repulsion is expected. Because increases in ionic strength shield this repulsion, extents of adsorption should increase as the sodium nitrate concentration is increased. NTMP possesses three phosphonate groups linked via an amine group. If only one phosphonate group is coordinated to the surface, then the other two groups may orient away from the surface, and may be less subject to long-range electrostatic interactions with the surface. Alternatively, adsorbed NTMP may be more strongly associated with protons or with electrolyte cations (in this case sodium ions) than NTMP species in bulk solution.

Our studies included aminopolyphosphonates with one, two, three, four, and five phosphonate arms. We conducted adsorption isotherm experiments; they allowed us to calculate the surface area occupied per molecule. This "footprint" increases ingoing from one to two, and from two to three phosphonate arms, but does not increase with compounds possessing four or five phosphonate arms. It appears that, at most, two phosphonate arms are specifically bound to the FeOOH(goethite) surface.

Cooperative and Antagonistic Adsorption of Aminopolyphosphonates and Metal Ions. Calcium and magnesium are frequently present at mM levels in aquatic environments, but the more common transition metal ions (e.g., manganese, iron, and zinc) are present at micromolar concentrations or lower. Calcium clearly increases the extent of aminopolyphosphonate adsorption onto FeOOH(goethite), an effect which is most pronounced under alkaline pH conditions. Transition metal ions are typically too low to significantly influence extents of aminopolyphosphonate adsorption.

Aminopolyphosphonates can have a dramatic effect on metal ion adsorption. When Cu(II) is present at 10 µM levels, acidic conditions show an increase in adsorption when aminopolyphosphonates are added. It is believed that "ligand-like" ternary surface complexes are formed, in which the aminopolyphosphonate "bridges" between the surface and the metal. Above pH 7, aminopolyphosphonates form dissolved complexes with Cu(II) that lower the extent of Cu(II) adsorption. In conclusion, the adsorption behavior of metal ions and aminopolyphosphonate chelating agents are closely interconnected, and they must be evaluated simultaneously.

Autoxidation of Aminopolyphosphonates Catalyzed by Mn(II). Prior to the work reported here, we recognized that Fe(III)-phosphonate complexes are subject to photolysis, and that nonphotochemical breakdown occurs in some natural water samples on timescales of 2 or 3 days. However, the mechanism of nonphotochemical breakdown had not been established.

Our studies began with the observation that the NTMP breakdown in Baltimore City tap water occurs on timescales of 2 or 3 days. We examined the effects of different constituent chemicals in tap water, and we concluded that dissolved manganese (Mn(II)) and molecular oxygen were responsible for NTMP breakdown. The breakdown mechanism is novel in several respects. Reaction begins by formation of an Mn(II)-NTMP complex, which activates Mn(II) towards oxidation by molecular oxygen. Hydrogen peroxide is generated by the reaction, and Mn(III)-NTMP complexes degrade via intramolecular electron transfer. Analysis of reaction products indicates that an electron is abstracted from the amine nitrogen of NTMP; both carbon-nitrogen and carbon-phosphorus bonds are broken during oxidation. Other metal ions compete with Mn(II) for coordination of NTMP. Ten micromolar levels of Cu(II) or Zn(II), for example, lower the rates substantially of NTMP breakdown. EDTMP and DTPMP, also strong chelating agents, are subject to Mn(II)-catalyzed autoxidation.

Oxidation of Aminopolyphosphonates in MnOOH-Containing Suspensions. In oxygen-free suspensions, the Mn(III)-containing mineral MnOOH oxidizes NTMP, but at rates that are slower than the Mn(II)-catalyzed autoxidation reaction described in the preceding section. Once again, we believe that an electron is abstracted from the amine nitrogen. The radical cation generated in this way yields primarily iminodimethylenephosphonic acid (IDMP). When oxygen is introduced into the suspension, additional chemical reactions can occur. For example, oxygen can form an adduct with the radical cation, yielding new reaction products. We observe a dramatic lag period, which is explained by the slow increase in Mn(II) concentrations arising from the MnOOH plus NTMP reaction. Concentrations eventually increase enough for the Mn(II)-catalyzed autoxidation reaction to occur. If Mn(II) is added to MnOOH-containing suspensions, the lag period disappears. Yields of C-N and C-P cleavage products vary substantially with pH, as do yields of hydrogen peroxide in oxygen-containing suspensions.

The three-armed NTMP reacts with MnOOH(manganite) far more rapidly than the two-armed IDMP, which reacts more rapidly than the one-armed compound aminomethylenephosphonic acid (AMP). We believe adsorption is a prerequisite for oxidation by MnOOH(manganite), and the adsorption process should be analogous in most respects to the adsorption onto FeOOH(goethite) previously discussed. From this perspective, the reactivity order is entirely reasonable. It also is possible that the tertiary amine of NTMP is somehow more susceptible to oxidation than the secondary amine of IDMP and the primary amine of AMP.

A number of heavily used chemicals are mixed carboxylate/phosphonates. The herbicide glyphosate is an important example; it is the mixed carboxylate/phosphonate analog for both IDMP and IDA. A comparison study indicates that MnOOH oxidation rates decrease in the following order: IDMP > IDA > glyphosate. The reactivity order is not believed to come from the activation energy for electron abstraction, because it is unlikely that the two functional arms cooperate during the electron transfer step. It is more plausible that the three compounds adsorb onto the MnOOH surface through different stoichiometries (e.g., different protonation levels) or different structures (i.e., different bonding arrangements) that influence reactivity.

Oxidation of Phosphonoformate and Phosphonoacetate by MnOOH(Manganite). Preceding sections dealt with the oxidation of compounds possessing the alpha-aminophosphonate moiety, and in some cases, the alpha-aminocarboxylate moiety. As noted earlier, oxidation of these compounds may begin with electron abstraction from the amine group. Therefore, it was important to investigate whether phosphonates that lack amine groups also are susceptible to oxidation.

Phosphonoformate and phosphonoacetate are both used as antiviral pharmaceuticals. In addition, phosphonoformate is a hydrolysis product of the herbicide fosamine. Rates of phosphonoformate oxidation by MnOOH(manganite) are a function of pH, but fall within the range of a few hours or days. Phosphonoacetate oxidation rates, in contrast, fall within the range of several months. These differences in reactivity are reasonable. The carboxylate group and phosphonate group within phosphonoformate are electronically coupled via resonance, which helps stabilize the radical intermediate generated by electron abstraction. The additional methylene group within phosphonoacetate electronically isolates the two functional groups. These reactivity differences also match those of the analogous carboxylate structures, oxalate and malonate.

CoOOH(heterogenite) Dissolution by Three-Armed Carboxylate- and Phosphonate-Containing Compounds. In a number of industrialized regions, sediments are contaminated with particulate-bound toxic metals; these results of disposal practices have been curtailed. There is a growing concern that synthetic organic compounds may cause sediments to lose their toxic metal ions, leading to human and ecosystem exposure. Ligand-assisted dissolution occurs when chelating agents detach particulate-bound metal ions, yielding metal ion-chelating agent complexes in solution. Reductive dissolution occurs when the synthetic organic compound reduces sparingly soluble +III or +IV metal ions to more soluble +II metal ions.

CoOOH(heterogenite) is used as a representative particulate-bound form of cobalt, which can be solubilized via either ligand-assisted dissolution or reductive dissolution. NTMP possesses three phosphonate groups connected through an amine group. N,N-Bis(phosphonomethyl)glycine (BPMG) possesses two phosphonate groups and one carboxylate group. N-(Phosphonomethyl)iminodiacetic acid (PMIDA) possesses one phosphonate group and two carboxylate groups. NTA possesses three carboxylate groups. If dissolution takes place solely via the ligand-assisted pathway, then soluble cobalt should not exceed the amount of chelating agent added. If reductive dissolution is responsible for some or all of the dissolution observed, then soluble cobalt could exceed the amount of chelating agent added, because each chelating agent molecule can serve a two-equivalent (or perhaps even a four- or six-equivalent) reductant.

In our experiments, each chelating agent contacted CoOOH(heterogenite) for 9 days prior to analysis. In the case of NTA, dissolved cobalt concentrations were slightly greater than the amount of NTA added; hence, ligand-assisted dissolution could be the dominant pathway. As each carboxylate group was replaced by a phosphonate group, however, dissolved cobalt concentrations increased substantially. We conclude that the alpha-aminophosphonate moiety is more susceptible towards oxidation; therefore, it is a better reagent for reductive dissolution than the alpha-aminocarboxylate moiety.

Phosphonate-based chelating agents have a reputation for being more resistant to degradation than carboxylate-based chelating agents. Our findings indicate, at least with the oxidant CoOOH(heterogenite), that the opposite is true.

Relevance. Through this research project, our understanding has significantly improved regarding: (1) the adsorption behavior of phosphonate-containing synthetic organic chemicals; (2) pathways and rates of oxidative breakdown of phosphonate-containing synthetic organic chemicals; and (3) synergistic interactions among phosphonate chelating agents and toxic metal ion contaminants. Information from this work can assist in assessing the impact of existing contamination, managing and remediating contaminated sites, and evaluating new chemicals, new industrial and agricultural pathways, and their potential for environmental harm.

Journal Articles on this Report : 4 Displayed | Download in RIS Format

Publications Views

Other project views:	All 19 publications	4 publications in selected types	All 4 journal articles

Publications

Type	Citation	Project	Document Sources
Journal Article	Nowack B, Stone AT. Adsorption of phosphonates onto the goethite-water interface. Journal of Colloid and Interface Science 1999;214(1):20-30.	R826376 (1998) R826376 (1999) R826376 (Final)	Abstract from PubMed Abstract: Science Direct - Abstract HTML Exit
Journal Article	Nowack B, Stone AT. The influence of metal ions on the adsorption of phosphonates onto goethite. Environmental Science and Technology 1999;33:3627-3633.	R826376 (1999) R826376 (Final)	not available
Journal Article	Nowack B, Stone AT. Degradation of nitrilotris (methylenephosphonic acid) and related (amino) phosphonate chelating agents in the presence of manganese and molecular oxygen. Environmental Science and Technology 2000;34(22):4759-4765.	R826376 (1999) R826376 (Final)	not available
Journal Article	Nowack B, Stone AT. Homogeneous and heterogeneous oxidation of nitrilotrismethylene-phosphonic acid (NTMP) in the presence of manganese(II,III) and molecular oxygen. Journal of Physical Chemistry B 2002;106(24):6227-6233.	R826376 (Final)	not available

Supplemental Keywords:

environmental chemistry, water, adsorption, metals, heavy metals, organics, chemicals, toxics, alternatives, environmentally conscious manufacturing, interfaces, equilibrium, kinetics, speciation, complexation, chelating agents, ligands, metal-ligand complexes, ligand exchange, dissolution, desorption, corrosion inhibitors, scale inhibitors, pesticides, herbicides, redox reactions., Scientific Discipline, Air, Environmental Chemistry, Chemistry, Engineering, Chemistry, & Physics, Biology, environmentally conscious manufacturing, ligand exchange, chemical composition, pollutant transport, toxic metals, phosphonates, environmental engineering, chemical kinetics

Progress and Final Reports:

Original Abstract

1998 Progress Report

1999 Progress Report