| Forside | | Indhold | | Forrige | | Næste |

Kemisk oxidation af sediment- og grundvandsforureningen på depotet ved Høfde 42 - fase 1: Projektbeskrivelse og forundersøgelser

Bilag 3 Laboratorieforsøg

Bilag 3.1 Fentons reagens og ozon/brintperoxid

Prepared for:

COWI
Odensevej 95
DK-5260 Odense S
Denmark

Laboratory Treatability Tests of
Chemical Oxidation Technologies for
Groundwater Remediation

Høfte 42 Harboøre Tange,
Ringkjøbing County, Denmark

Prepared by:

10015 Old Columbia Road, Suite A-200
Columbia, Maryland 21046 USA
GeoSyntec Project Number MR0487

July 1, 2006

Executive Summary

A series of laboratory treatability tests were conducted by GeoSyntec Consultants (GeoSyntec) and its subcontractors to evaluate alternative in situ chemical oxidation (ISCO) technologies for groundwater remediation at the Høfde 42 Harboøre Tange site in Ringkjøbing County, Denmark (Site). The primary contaminants at the Site are the organophosphorus pesticides parathion, methyl parathion, malathion, 2-methyl-4-chlorophenoxyacetic acid (MCPA) and ethyl-sulfoteb. These chemicals are present both as dissolved phase in groundwater and as an immiscible, dense nonaqueous phase liquid (DNAPL) in the upper sand aquifer at the Site. In addition to these pesticides, a number of related organic contaminants exist in the Site soil and groundwater, including (i) constituents related to the pesticide production (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS), and (ii) pesticide and pesticide manufacturing contaminant degradation products (4-Cl-cresol, para-nitrophenol [PNF], amino-parathion, amino-methyl parathion, MP-1, EP-1, MP-2-Syre, EP-2-Syre, paraoxon, malaoxon, methyl-paraoxon, E-OOOPO, EEM-OOSPO). Mercury is also a significant contaminant at the Site; however, the tests reported herein did not evaluate treatment of mercury.

The treatability tests evaluated two ISCO technologies: ozone and Fenton’s reagent. Using groundwater collected from the Site, each test evaluated treatment efficiency and extent of treatment over a range of oxidant dosages. Treatment efficiency and extent was measured in terms of destruction of the primary pesticides, pesticide manufacturing contaminants, and pesticide degradation products, as well as accumulation of inorganic oxidation products (e.g., sulfate, nitrogen oxides, and phosphate) and reduction in toxicity (i.e., via Microtox ® bioassay).

The ozone tests were performed by adding oxidant to groundwater in sealed batch reactors. Ozone was used as the sole oxidant in one treatment at a concentration of 190 mg/L. In the remaining treatments, ozone and peroxide were used in combination at concentrations of 500 mg/L ozone + 250 mg/L peroxide, and 1000 mg/L ozone + 500 mg/L peroxide. Peroxide was added in these other treatments to enhance the ozone reactivity, which was limited in the laboratory tests by the acidic conditions of the Site groundwater and the relatively short duration over which the study was completed. Ozone at each dose tested achieved complete (100%) destruction of the pesticides (parathion, methyl parathion, MCPA, and malathion) and the pesticide reaction products (MP-1, EP-1, MP-2-Syre, EP-2-Syre, 4-Cl-cresol, PNF). Pesticide manufacturing contaminants (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS) and the oxidation product EEM-OOSPO were also reduced by 100%. E-OOOPO was produced at the lower and middle ozone dosages, but was not observed in the high ozone dose treatment suggesting that this oxidation product may be produced transiently during a field application of the technology. Oxons (paraoxon, malaoxon, methyl-paraoxon) were produced with addition of 190 mg/L ozone, but were not detected in the highest oxidant treatment. This data suggests that these oxidation products may be produced transiently in a field application of the technology. Treatment at the highest oxidant dose achieved a 3-fold reduction in groundwater toxicity. The low pH of the Site groundwater, both before and post-treatment, may have limited the ability of ozone/peroxide to achieve greater reductions in toxicity as measured by Microtox® assay.

The Fenton’s reagent tests were also performed by adding oxidant (and catalyst) to groundwater in sealed batch reactors over a range of oxidant and catalyst concentrations. Fenton’s reagent at each dose tested achieved complete (100%) destruction of the pesticides (parathion, methyl parathion, MCPA, and malathion) and the pesticide manufacturing contaminants (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS) and the oxidation products EEM-OOSPO and E-OOOPO, and paraoxon were reduced by 100% (malaoxon and methyl-paraoxon were not detected in any sample). Pesticide reaction products (MP-1, EP-1, EP-2-Syre, 4-Cl-cresol, PNF) were reduced by 100% at all Fenton’s reagent doses. MP-1 was produced in the lowest dosage treatments, but was not observed in the higher dosage treatments suggesting that this oxidation product may be produced transiently during a field application of the technology. MP-2-Syre concentrations decreased for the low dose batch reactors but increased in all of the high dose batch reactors. The low dose data indicates that this compound can be destroyed by Fenton’s chemistry, however the high dose data suggests that MP-2-Syre also forms as a reaction product from hydrolysis of other organic compounds (e.g., methyl parathion) in the Site groundwater. Treatment with a high Fenton’s dose achieved a 77-fold reduction in groundwater toxicity, as measured by Microtox® bioassay.

The results of these bench tests indicate that Fenton’s reagent and ozone are both potentially highly effective technologies for destruction of the Site contaminants, and associated manufacturing contaminants and degradation products. The extent and rate of treatment by Fenton’s reagent and ozone was impressive in both cases. One of the few differences observed between the results of the two tests was that the toxicity reduction achieved was significantly greater in the Fenton’s treatability test. One or both of these technologies might be successful at the field scale, although selection of one or both of these technologies for field implementation should only proceed after a careful evaluation of the site-specific conditions. In addition, it is recommended that additional bench testing be performed as a part of any pre-design testing. Such bench testing could include optimization of reagent dose (to avoid wasting reagent), evaluation of soil oxidant demand, treatment of sorbed-phase pesticides, treatment (enhanced dissolution) of DNAPL, evaluation of mercury transformation consequent to treatment, and adjustment of reagents employed to avoid/minimize formation of MP-2-Syre.

Table of Contents

1 Introduction

• 1.1 Background
• 1.2 Overview and Scope of Work

2 Objective

3 Groundwater Collection

4 Ozone Study

• 4.1 Setup and Sampling Details
• 4.2 Results

5 Fenton’s Study

• 5.1 Setup and Sampling Details
• 5.2 Results

6 Key Findings and Conclusions

• 6.1 Chemical Oxidation using Ozone
• 6.2 Chemical Oxidation using Fenton’s Reagent

7 References

List of Tables

Table 1: Summary of Oxidation Treatability Study Controls and Treatments

Table 2: Summary of Oxidation Treatability Study Analytical Results from External Laboratories

Table 3: Summary of Treatability Study Results by Treatment and Constituent

Table 4: Elemental Component Mass Balance Calculations

List of Figures

Figure 1: Ozone - May 9 2006 Trial External Laboratory Results

Figure 2: Ozone and Peroxide - May 9 2006 Trial External Laboratory Results

Figure 3: Ozone - May 15 2006 Trial Laboratory Results

Figure 4: Ozone and Peroxide - May 15 2006 Trial External Laboratory Results

Figure 5: Fenton's Reagent - Low Dose External Laboratory Results

Figure 6: Fenton's Reagent - High Dose Laboratory Results

List of Appendices

Appendix A: ISOTEC Report

Appendix B: APT Report

List of Abbrevations and Acronyms

1 Introduction

1.1 Background

A series of laboratory treatability tests were conducted by GeoSyntec Consultants (GeoSyntec) and its subcontractors to evaluate alternative treatment technologies for groundwater remediation at the Høfde 42 Harboøre Tange site in Ringkjøbing County, Denmark (Site). The Site is located on the western coast of Denmark, on a beach adjacent to the North Sea. The primary contaminants of concern in the source area at the Site are the organophosphorus pesticides, parathion, methyl parathion, malathion, 2-methyl-4-chlorophenoxyacetic acid (MCPA) and ethyl-sulfoteb (Ringkjøbing County, 2004). These chemicals are present both as dissolved phase in groundwater and as an immiscible, dense nonaqueous phase liquid (DNAPL) in the upper sand aquifer at the Site. In addition to the pesticides, a number of other organic compounds exist in the Site soil and groundwater, including:

• contaminants related to the pesticide production (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS); and
• biotic and/or abiotic reaction products of the pesticides and the pesticide contaminants (4-Cl-cresol, para-nitrophenol [PNF], amino-parathion, methyl-amino parathion, MP-1, EP-1, MP-2-Syre, EP-2-Syre, paraoxon, malaoxon, methyl-paraoxon, E-OOOPO, EEM-OOSPO).

A groundwater plume that contains these pesticides, their contaminants and degradation products extends from the DNAPL source area towards the North Sea. The dissolved phase chemicals have also been transported into the lower sand aquifer at the Site. The conditions in this plume are highly acidic (typical pH ranges from 2.4 to 4.3). Mercury is also present in significant concentrations in the DNAPL at the Site; however, treatment of mercury was not evaluated in the tests reported herein.

GeoSyntec and its subcontractors reviewed state-of-the-science literature regarding the use of in situ chemical oxidation (ISCO) techniques for treatment of organophosphorous pesticides in groundwater (GeoSyntec and the Connecticut Agricultural Experiment Station, 2006). The literature review identified Fenton’s reagent, ozone, and permanganate as ISCO technologies that are potentially effective for treatment of the contaminants at the Site. Accordingly, screening-level laboratory treatability tests were performed to evaluate these technologies using Site groundwater to further evaluate the feasibility of using ISCO at the Site. The treatability test approach and analytical methods were performed in general accordance with a work plan memorandum submitted to COWI on 30 March 2006. The treatability tests evaluating ozone and Fenton’s reagent were performed at the laboratories of Applied Process Technology, Inc. (APT of Pleasant Hill, California, USA) and In Situ Oxidative Technology, Inc. (ISOTEC of West Windsor, New Jersey, USA), respectively. Treatability tests to evaluate the feasibility of using permanganate were conducted and reported separately by COWI.

1.2 Overview and Scope of Work

Screening-level bench treatability tests were performed to determine the effectiveness of various chemical oxidant options for in situ treatment for the unique mixture of contaminants at the Site. All the bench tests used aquifer groundwater samples collected from the Site. The purpose of these tests was to confirm the efficiency and extent of treatment of the principal contaminants at the Høfde 42 site by these chemical oxidants. Data from these tests can be used to aid in the selection of an effective remedial technology for field-scale application at the Site. Parameter measurements obtained in the bench tests (e.g., contaminant reactivity, oxidant dosage) can be used to support the design of the field-scale remedy. Analysis of organic chemicals present at the Site, including the pesticides, their contaminants and reaction products was provided by Cheminova A/S. In addition, anion measurements (e.g., NO₂^-, NO₃^-, PO₄^3-, and SO₄^2-) were performed to evaluate whether complete oxidation of the target chemicals had been achieved, and changes in toxicity were measured (via Microtox® analysis) in samples from select treatments.

The remainder of this report is divided into seven sections. Section 2 presents the objectives of the study. Collection of Site materials for the treatability work is presented in Section 3. The approach, methods and results of the ozone study is presented in Section 4. Section 5 presents the approach, methods and results of the Fenton’s reagent study. Section 6 presents the key findings and conclusions from the treatability studies. Report references are provided in Section 7.

2 Objective

The purpose of ozone and Fenton’s reagent treatability tests was to confirm the efficiency and extent of treatment of the principal contaminants at the Site by these chemical oxidants over multiple oxidant dosages. The findings of the literature review, together with the results of the treatability tests, were used to identify feasible and effective ISCO technologies for potential implementation at the Site. The bench tests involved evaluation using groundwater samples collected from the Site. The focus of the treatability tests was on the organophosphorus pesticides and associated degradation products and manufacturing contaminants listed in Section 1.1.

3 Groundwater Collection

Groundwater samples were collected by COWI on 24 April 2006 from a well representative of the source area conditions at the Site. Groundwater was collected directly into four-3 gallon containers, minimizing headspace to the extent possible. The groundwater samples were express-shipped to APT’s laboratory in California and ISOTEC’s laboratory in New Jersey. For the purposes of sample preservation, groundwater samples were frozen prior to shipment, packed in ice in coolers during shipment, and immediately transferred to a freezer upon receipt at the laboratories. The groundwater was completely thawed one day prior to initiating the treatability studies.

4 Ozone Study

A treatability test was performed to evaluate the reactivity of the target chemicals with ozone and the rate of ozone consumption by the Site groundwater. The treatability test involved a series of batch tests with Site groundwater at a range of ozone dosage concentrations. While the treatability studies were designed to evaluate ISCO using ozone alone, peroxide was added in some treatments to enhance the ozone reactivity (i.e., increase the concentration of hydroxyl radicals), which was limited in the laboratory tests by the acidic conditions of the Site groundwater and the relatively short duration over which the study was completed.

The setup and sampling details are summarized in Section 4.1 and the results are described in Section 4.2. APT’s report for the ozone treatability study is included in Appendix A and includes a more detailed description of the study setup and analysis details. Table 1 summarizes the treatments and experimental controls.

4.1 Setup and Sampling Details

Batch tests were performed by adding oxidant to groundwater to a sealed batch reactor. Ozone was used as the sole oxidant in one treatment at an approximate concentration of 190 milligrams per liter (mg/L; labeled SP-190). In the remaining treatments, ozone and peroxide were used in combination at approximate concentrations of 500 mg/L ozone + 300 mg/L peroxide, and 1000 mg/L ozone + 650 mg/L peroxide (labeled SP-500, and SP-1000 respectively). In the combined ozone and peroxide treatments, hydrogen peroxide and ozone were added in sequence to Site groundwater to promote oxidation of the target chemicals. The treatment period in the reactor ranged from one to five hours, with an increase in treatment period with oxidant dosage. A control was constructed to assess the pre-treatment concentrations of the analytes in the Site groundwater (labeled SP-0 or source water). After completion of the tests conducted on 9 and 10 May 2006, it was discovered that the test performance samples were not quenched to remove any excess oxidant. For this reason, a second set of tests (including a second SP-0, SP-190, and SP-1000) were conducted on 15 May 2006 with ozone quenched. Samples for anion analysis were quenched using a methanol solution, while samples for pesticide and toxicity analysis were quenched with thiosulfate.

A list of the analyses conducted and the number of analyses for each treatment is included in Table 1 and is further detailed in Appendix A. The details of sample collection are also presented in Appendix A, including the holding times, sample volumes, preservation, analytical methods, and laboratories selected for each analysis. Throughout this document analysis conducted at the treatability study subcontractor laboratories (APT and ISOTEC) are described as “internal” analyses, while analyses conducted at analytical laboratories are described as “external” analyses. The analyses conducted at each sample event included high performance liquid chromatography (HPLC) and gas chromatography (GC) analysis conducted by Cheminova to identify organic compounds present in the Site groundwater that were either: 1) pesticides; 2) chemical contaminants from the manufacture of parathion and their oxidation products; 3) abiotic and/or biotic in situ reaction products of 1) and 2) [denoted as “pesticide reaction products” in this document]; and 4) oxons. A detailed listing of the constituents included in the HPLC and GC analyses is included in Table 1. In addition to these organic analyses, samples for anions (sulfate, phosphate, nitrate and nitrite), color, oxidant, turbidity, alkalinity, oxidation-reduction potential (ORP) and pH were collected for each treatment and control. All samples were analyzed by published analytical methods similar to those of the United States Environmental Protection Agency, as appropriate.

Toxicity testing was conducted for the source water (SP-0) and from the SP-1000 treatment to evaluate changes in groundwater toxicity consequent to treatment. The toxicity of the sample contents was evaluated via a Microtox® bioassay testing system that measures the light output from freeze-dried luminescent bacteria (Photobacterium phosphoreum). The light-producing mechanism in these bacteria is tied to the metabolic processes of the cell. Specifically, when the bacteria are killed or the bacteria's light-producing mechanism is changed or damaged by toxic substances, resulting in a reduction in luminescence (light output). These changes in light output were measured to calculate an IC50 for each sample, which is defined as the initial solution concentration, and is a calculated toxicity value representing the sample concentration expressed in % estimated to cause a 50% response by the exposed test organisms.

The change in oxidant and target chemical concentrations was used to assess the oxidant demand of the groundwater. This oxidant demand in combination with the target chemical concentration data collected was used to evaluate the feasibility of this technology for potential application in the field.

4.2 Results

The laboratory analytical results are summarized in Table 2, including anions, organic compounds, toxicity and color from the external laboratories. The concentration of detected chemicals in the Site groundwater prior to treatment and after treatment at each of the three treatment conditions are presented in Figures 1 and 2 for the trial completed on 9 April 2006, and in Figures 3 and 4 for the trial completed on 15 April 2006. The results from these two trials were very similar, with small differences in concentrations attributed to the use of the ozone quench. Because the performance samples in the first trial were not quenched, discussion of the results presented here focuses primarily on the second trial. Results for each of the four categories of organic compounds, as well as the anion and color results, are presented on separate data plots in each of these figures. Results for internal analyses analyzed including oxidant concentrations, alkalinity, ORP and pH are presented in Appendix A. Analytical reports from the external laboratories are included in Appendix C.

The results presented in Figures 1 through 4 clearly indicate that virtually all organic compounds were reduced by the addition of ozone or ozone and peroxide with concentrations decreasing with higher oxidant concentrations. The relative change in concentration of the organic compounds by the oxidant addition is presented in Table 3, with the following key results:

Pesticides, (parathion, methyl parathion, MCPA, and malathion), were reduced by 100%. Ethyl sulfoteb was not detected in any samples.

Pesticide manufacturing contaminants (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS) and the oxidation product EEM-OOSPO were reduced by 100%. E-OOOPO was produced in SP-190 and SP-500 treatments, but was not observed in the SP-1000 treatment suggesting that this oxidation product may be produced transiently during a field application of the technology.

Pesticide reaction products (MP-1, EP-1, MP-2-Syre, EP-2-Syre, 4-Cl-cresol, PNF) were reduced by 100%. Methyl-amino-parathion and amino-parathion were not detected in any samples.

Oxons (paraoxon, malaoxon, methyl-paraoxon) were produced with addition of ozone alone (SP-190), but were not detected in the highest oxidant treatment (SP-1000). This data suggests that (i) oxons may be produced transiently in a field application of the technology, particularly if ozone alone is used; and (ii) the combination of ozone and hydrogen peroxide may achieve more complete treatment in the field.

Treatment achieved a 3-fold reduction in groundwater toxicity, as measured by Microtox® bioassay.

Nitrate, phosphate and sulfate, inorganic reaction products from the oxidation of the organic compounds, increased with higher dosage of oxidant. Nitrite was not detected in any samples.

Color was reduced in most oxidant treatment conditions. At the highest treatement a slight increase in color was observed.

Table 4 presents a calculation of the change in presence of nitrogen, phosphorus, and sulfur from the organic and inorganic (i.e. anions) components in the samples collected, as well as the contribution of these elements to the samples from any reagents used in the study. This mass balance calculation allows for an evaluation of whether there were substantial unmeasured reaction products in the treatment samples collected. It should be noted that the observed change in sulfate concentrations was relatively small as compared to the total initial concentration, and was likely within the experimental error of the test. It can be seen that at the highest treatment concentration (1000 mg/L ozone) that the decrease in organic nitrogen and phosphorus was quite similar as the increase in inorganic nitrogen and phosphorus, suggesting that the oxidation of the organic compounds was complete. Results for sulfur were similar to that of nitrogen and phosphorus, although the measurements were within the experimental error of the test. However at the lower treatment concentrations (i.e. 190 mg/L ozone) the total concentration of these elements in the measured organic and inorganic components is different, suggesting some incomplete oxidation products were present which were not quantified by the analyses available for the study. These mass balance calculations suggest that some transient reaction products from ozone treatment of the organic compounds that were not part of the analytical suite may have been formed at the lower ozone dose. However, the apparent complete oxidation of the organic compounds in the highest treatment suggests that these unknown reaction products will be oxidized with sufficient reagent dose and/or longer contact-time.

The toxicity of the sample contents was evaluated via a Microtox® bioassay. The toxicity data results presented in Table 2 is the calculated sample concentration expressed in % that causes a 50% response by the exposed test organisms. Because the “control” sample is more toxic compared to the “treated” sample, a greater percent concentration of “treated” sample is needed to generate 50% response than the “control” sample for a fixed exposure time. For 5 minutes of exposure time, a “control” sample concentration of 0.44% resulted in a 50% of the exposed test organisms being killed versus the treated sample concentration of 1.38% to kill 50% of the exposed test organisms. The acidity of the Site groundwater (i.e. pH ~3 both before and after treatment) likely played a strong role in the relatively high toxicity of the Site groundwater.

The oxidant demand of the groundwater collected from the Site was relatively low, and the oxidation of the Site contaminants was not limited by presence of non-target organic compounds or alkalinity.

5 Fenton’s Study

ISCO treatability tests with modified (neutral-pH) Fenton’s reagent were performed using groundwater samples collected from the Site. These studies were used to provide information about the reactivity of the organic chemicals and any potential interactions that may influence the design of an oxidant injection system. The following tests were performed:

• Batch Tests with Groundwater, Low Dose: Site groundwater was mixed with one and three dosages of a low concentration of Fenton’s reagent; and
• Batch Tests with Groundwater, High Dose: Site groundwater was mixed with one, two, and three dosages of a high concentration of Fenton’s reagent.

The low and high doses were used to evaluate any differences in oxidant performance as a result of the treatment concentration.

The following sections summarize the setup and sampling details for the groundwater batch tests (Section 5.1) and results from the tests (Section 5.2). ISOTEC’s report for the Fenton’s reagent treatability study, which is included in Appendix B, provides a more detailed description of the study setup and analysis details. Table 1 summarizes the treatments and experimental controls. The analytical methodologies employed for the treatability study are summarized in the subcontractors report in Appendix B.

5.1 Setup and Sampling Details

Batch tests were performed by adding Fenton’s reagent to the Site groundwater in sealed batch reactors. A number of treatment and controls were prepared. The quantity of reagents added to each treatment, including hydrogen peroxide and catalyst, are summarized in Table 2 of ISOTEC’s report (Appendix B). The reagent solution added to achieve Fenton’s chemistry had a pH between 5 and 6; no other chemicals were added to adjust the pH of the Site groundwater. The low dose treatment reactors received one, and three dosages (treatments D and E) of hydrogen peroxide and catalyst to achieve equivalent concentrations of 0.9% and 2.8% of hydrogen peroxide and 1.8 millimoles per liter (mM) and 3.7 mM of catalyst, respectively. The high dose treatment reactors received one, two and three dosages (treatments A, B and C) of hydrogen peroxide and catalyst to achieve equivalent concentrations of 2.7%, 5.3%, and 7.9% of hydrogen peroxide and 5.3 mM, 10.6 mM, 15.9 mM of catalyst, respectively. The multiple dosage approach was used during the test to increase treatment efficiency, minimize gas formation and the resulting pressure buildup. Distilled water was used to compensate the difference of reagent volumes applied between reactors. Controls included a baseline sample (F-BGC) collected to assess the pre-treatment concentrations of the analytes in the Site groundwater and a control reactor (F-BGCont) which received an equivalent volume of distilled water instead of reagent. A time gap of approximately 24-48 hours was maintained between dosages, with the exact timing based upon oxidant measurements. All reactors (control and treatment) were left undisturbed for a minimum of 24 hours or until all the peroxide was consumed before analytical sample collection.

A list of the analyses conducted and the number of analyses for each treatment is included in Table 1 and is further detailed in Appendix B. The details of sample collection are also presented in Appendix B, including the holding times, sample volumes, preservation, analytical methods, and laboratories selected for each analysis. The analyses conducted at each sample event were similar to that for the ozone study and included HPLC and GC analysis conducted by Cheminova to identify organic compounds (listed in Table 1) present in the Site groundwater. In addition to these organic analyses, samples for anions (sulfate, phosphate, nitrate and nitrite), color, peroxide, ferrous iron, ORP and pH were collected for each treatment and control. All samples were analyzed by published analytical methods similar to those of the United States Environmental Protection Agency, as appropriate. Analytical lists and methods were the same as described for the ozone study. The change in target chemical concentration over time was used to assess the extent and rate of chemical degradation.

For the baseline sample and the highest Fenton’s reagent dose (treatment C), toxicity testing was conducted to evaluate changes in groundwater toxicity consequent to treatment. The toxicity was calculated as the sample concentration expressed in % that causes a 50% response by the exposed test organisms. For a fixed exposure time, because the “control” sample is more toxic compared to the “treated” sample, a greater percent concentration of “treated” sample is needed to generate 50% response than the “control” sample. For 5 minutes of exposure time, a “control” sample concentration of 0.55% resulted in a 50% of the exposed test organisms being killed versus the treated sample concentration of 42.35% to kill 50% of the exposed test organisms.

5.2 Results

The results of analysis of the target compounds at the Site are summarized in Table 2, including anions, organic compounds, toxicity and color from the external laboratories. The concentration of detected chemicals in the Site groundwater prior to treatment and after treatment at each of the three treatment conditions are presented in Figures 5 and 6 for the low and high dose treatment conditions. Results for each of the four categories of organic compounds, as well as the anion and color results, are presented on separate data plots in each of these figures. Results for analyses analyzed by ISOTEC including peroxide and ferrous iron concentrations, ORP and pH are presented in Appendix B. Analytical reports from the external laboratories are included in Appendix C.

The results presented in Figures 5 and 6 clearly indicate that the presence of virtually all organic compounds were reduced by the addition of Fenton’s reagent and that the extent of treatment for certain constituents increased with increasing oxidant dose. The relative change in concentration of the organic compounds by the oxidant addition is presented in Table 3, with the following key results:

Pesticides, (parathion, methyl parathion, MCPA, and malathion), were reduced by 100% in all the treatments, including the lowest peroxide dose. Ethyl sulfoteb was not detected in any samples.

Pesticide manufacturing contaminants (E-OOOPS, M-OOSPS, EEM-OOSPS, MME-OOSPS) and the oxidation products EEM-OOSPO and E-OOOPO were reduced by 100% in all the treatments, including the lowest peroxide dose.

Pesticide reaction products (MP-1, EP-1, EP-2-Syre, 4-Cl-cresol, PNF) were reduced by 100%. Methyl-amino-parathion and amino-parathion were not detected in any samples. MP-1 was produced in the lowest dosage treatments (treatment A and D), but was not observed in the higher dosage treatments (treatments B, E and F) suggesting that this oxidation product may be produced transiently during a field application of the technology. MP-2-Syre concentrations decreased for the low dose batch reactors (treatments D and E; Figure 5) but increased in all of the high dose batch reactors (treatments A, B, and C; Figure 6). The low dose data indicates that this compound can be destroyed by Fenton’s chemistry, however the high dose data suggests that MP-2-Syre also forms as a reaction product from hydrolysis of other organic compounds (e.g., methyl parathion) in the Site groundwater. Methyl parathion is known to undergo enzymatic hydrolysis to form MP-2-Syre and PNF (Cho et al. 2002), and therefore methyl parathion may also be susceptible to abiotic hydrolysis in the presence of certain catalysts.

Paraoxon was reduced by 100% in all treatments. Malaoxon and methyl-paraoxon were not detected in any samples.

Treatment at the highest Fenton’s reagent dose achieved a 77-fold reduction in groundwater toxicity, as measured by Microtox® bioassay.

Nitrate, nitrite, phosphate and sulfate, which were present in the reagents used in the study and were also inorganic reaction products from the oxidation of the organic compounds, increased with concentration of oxidant.

Color as measured by the external laboratory increased in many of the oxidant treatment conditions. Precipitates generated during the oxidation treatment likely interfered with the measurement of color, making this data difficult to interpret. Color measurements by ISOTEC (Figure 2, Appendix A) using filtered samples showed a substantial decrease from initial conditions for most treatment conditions.

Table 3 presents a calculation of the change in presence of nitrogen, phosphorus, and sulfur from the organic and inorganic (i.e. anions) components in the samples collected, as well as the contribution of these elements to the samples from the reagents used in the study. The high concentration of nitrogen, phosphorus and sulfur in the reagents used in the study make it difficult to use the mass balance calculations to evaluate whether any unmeasured organic species were present in the samples.

The pH in the treatment samples increased substantially as a result of the addition of Fenton’s reagent (Figure 3 of ISOTEC’s report; Appendix C). The pH in the control sample was 2.91, while the post-treatment sample pH ranged between 3.6 and 8.3, increasing with concentration of oxidant.

The toxicity of the sample contents was evaluated via a Microtox® bioassay and the results are presented in Table 2. The results of toxicity testing indicated that the toxicity of the groundwater decreased substantially - 77 fold - as a result of Fenton’s reagent addition (at its highest dose). The substantial decrease in toxicity was likely the effect of both the treatment of the organic compounds and the increase in pH from highly acidic conditions to slightly basic conditions.

6 Key Findings and Conclusions

Laboratory treatability tests were conducted to evaluate the performance of various in situ remediation technologies for treatment of groundwater at the Cheminova site in Ringkjøbing County, Denmark. The following technologies and treatments were tested in batch reactors using aquifer groundwater collected from the Site:

• Chemical oxidation using Fenton’s reagent; and
• Chemical oxidation using Ozone;

The results of the treatability tests are summarized in Table 3, which presents general treatment results for individual contaminants. Primary findings of the tests are also summarized in the sections below.

6.1 Chemical Oxidation using Ozone

Treatment of Site groundwater using ozone+peroxide at a dose of 1000 mg/L ozone achieved complete destruction of all pesticides, contaminants from the pesticide manufacturing process, pesticide reaction products, and oxons. E-OOOPO and malaoxon were generated in lower concentration treatments, suggesting that these compounds may be observed transiently during a field application of the technology. Ozone treatment resulted in a 3-fold reduction in toxicity of Site groundwater. This toxicity reduction observed in the ozone test was more than 20 times lower than that observed for the Fenton’s study, which may be linked to pH of the post-treatment samples. pH in the ozone-treated samples remained acidic, whereas the Fenton’s reagent treatments resulted in a slightly basic pH.

Based upon comparison of the results observed in the ozone and ozone and peroxide treatments, it is expected that either of these reagents could be effective at field treatment of the target compounds. Although oxons were produced at the lower ozone dose (190 mg/L), the short duration of the test did not allow a determination of whether these products would persist in the presence of ozone over durations more relevant to field scale application. While the low pH of the Site groundwater limited the reactivity of the ozone when used solely in the batch reactor for the short duration study, application of the technology in the field would be over a longer treatment period, and it is not expected that acidity would preclude use of ozone alone as an ISCO approach. Nevertheless, a higher ozone dose, possibly combined with hydrogen peroxide, might be required to achieve complete treatment without production of oxons.

6.2 Chemical Oxidation using Fenton’s Reagent

Fenton’s reagent, at a peroxide dose of 7.9 %, completely degraded all pesticides, contaminants from the pesticide manufacturing process, oxons detected in the site groundwater and most pesticide reaction products. MP-1 was generated in lower concentration treatments, suggesting that these compounds may be observed transiently during a field application of the technology. MP-2-Syre was the only compound that was generated and not subsequently oxidized by higher treatment in the Fenton’s reagent test. This compound was not generated in the ozone study, which achieves oxidation of organic compounds through similar reaction pathways (i.e., production and reaction of hydroxyl radicals). This suggests that the specific reagents used in the Fenton’s study (such as phosphate which is used as a stabilizer of peroxide) may have prevented complete oxidation of MP-2-Syre. This compound is likely a hydrolysis product from methyl parathion, following a similar reaction pathway as that for production of EP-2-Syre from parathion as described by Atkor Innovation (2004) in their review of natural attenuation processes at the Site. As shown by Cho et al. 2002, MP-2-Syre is a known product of the hydrolysis of methyl parathion. Results from Fenton’s study suggest that this reaction may be catalyzed by peroxide and/or metal ions. The molar concentration of MP-2-Syre in the highest dose treatment (treatment C) was several times higher than that of methyl-parathion in the initial sample, suggesting that other pesticides and related compounds may have also served as parent compounds for the production of MP-2-Syre. It is possible that with alternative reagents and/or avoiding the use of phosphate in a field application that this compound would not be generated. Fenton’s reagent at the highest dose tested achieved a 77-fold reduction in toxicity of Site groundwater. These results suggest that MP-2-Syre does not possess significant toxicity as measured by the Microtox® bioassay.

7 References

ATKOR Innovation, 2004. Høfde 42 Depotet: Litteratureundersøgelse og modelbeskrivelse af naturlig nedbrydning af parathion. 24 May 2004.

Cho, C. M-H., A. Mulchandani, and W. Chen, 2002. Bacterial cell surface display of organophosphorus hydrolase for selective screening of improved hydrolysis of organophosphate nerve agents. Appl. Environ. Microbiol. 68(4):2026-2030.

GeoSyntec Consultants and Department of Soil & Water, Connecticut Agricultural Experiment Station, 2006. Literature Review: In Situ Chemical Oxidation of Organophosphorus Pesticides in Groundwater.

GeoSyntec Consultants, 2006. Work Plan for Chemical Oxidation Treatability Tests Høfde 42 Harboøre Tange, Ringkjøbing County, Denmark. March, 2006.

Ringkjøbing County, Department of Environment and Infrastructure, 2004. A DNAPL hotspot of organophosphorous pesticides, Høfde 42 Harboøre Tange, County of Ringkjøbing, Denmark. October, 2004.

LAboratory Treatability Study Report

Cheminova Site
Ringkjobing County, Denmark

June 26, 2006

Prepared For

GeoSyntec Consultants
130 Research Lane, Suite 2
Guelph, Ontario
Canada N1G 5G3

Prepared By

In-Situ Oxidative Technologies, Inc.
51 Everett Drive, Suite A-10
West Windsor, New Jersey 08550

ISOTEC Project No. 801000

Table of Contents

1 Executive Summary

2 Study Objectives

3 Sample Collection and Preparation

4 Laboratory Treatability Study

• 4.1 Reagent Selection
• 4.2 Establishing Experimental Controls
• 4.3 Experimental Setup
• 4.4 Sample Analysis

5 Treatability Study Results

• 5.1 GW-test Results
  • 5.1.1 Pesticide and Degradation Product Results
  • 5.1.2 Toxicity Results
  • 5.1.3 Anion Results
  • 5.1.4 Color Results
  • 5.1.5 pH

6 Summary and Conclusions

1 Executive Summary

In-Situ Oxidative Technologies, Inc. (ISOTEC^SM) was retained by GeoSyntec Consultants (GeoSyntec) to conduct an in-situ chemical oxidation (ISCO) bench-scale laboratory treatability study (study) on soil and groundwater samples collected from the Cheminova site located in Ringkjobing County, Denmark. Targeted contaminants of concern (COCs) for the study are organophosphorous pesticides primarily consisting of insecticide parathion (EP3). The purpose of the study was to determine the potential effectiveness of ISOTEC’s modified Fenton’s process to treat COCs in dissolved phase in site groundwater. The modified Fenton’s reagent promotes contaminant destruction via oxidizing and reducing free radicals including hydroxyl radicals, superoxide radicals and hydroperoxide anions.

The treatability study consisted of a groundwater test (GW-test) to evaluate the COC treatment effectiveness. The test evaluated five different reagent dosages ranging from low to high treatment conditions in an effort to determine the optimal treatment dosage for a potential field application. Analytical samples were submitted to multiple laboratories including Columbia Analytical Services (CAS), Cheminova lab (Cheminova) and Enviro-Test Labs (ETL) for analyses of a variety of parameters (see Section 3.1).

For the reagent dosages evaluated, treatability study results indicated the following:

• Pesticides including MCPA, methyl-parathion, parathion and malathion were reduced by greater than 99% to non-detect (ND) concentrations for the lowest treatment evaluated.
• Parathion degradation products including MP-1, EP-2 Syre, EP-1, PNF, amino-methyl-parathion, and amino-parathion were reduced by greater than 99% to ND concentrations for the lowest treatment evaluated. The only exception was MP-2 Syre (diethylthiophosphoric acid or DETP), which increased in concentration with increasing reagent dosage. It is hypothesized that the presence of phosphate in the ISOTEC reagent mix to promote stabilization of hydrogen peroxide may have led to formation of MP-2 Syre via catalytic hydrolysis or other mechanisms. Therefore, it is suggested that this assumption be tested in a follow-up test using the ISOTEC reagent with no phosphate addition, and if confirmed, the field application of modified Fenton’s reagent should exclude phosphate.
• Groundwater toxicity levels were significantly reduced following modified Fenton’s treatment;
• Individual anion concentrations including nitrate-nitrogen, nitrite-nitrogen and sulfate, and orthophosphate-phosphorous all increased with each treatment possibly due to their formation from pesticide mineralization or their presence in the ISOTEC reagent mix.
• Groundwater color analysis performed by CAS showed inconclusive results possibly due to presence of suspended solids in certain treated samples, which interfered with the analysis. ISOTEC visual observations during the test and also quantitative spectrophotometric absorbance measurements of color absorbance indicated that the sample color intensity (as measured by absorbance) decreased with increasing treatments.

2 Study Objectives

The objectives of the study were as follows:

• Evaluate the COC treatment effectiveness of the modified Fenton’s reagent on site groundwater samples; and
• Select the most effective reagent dosage (i.e. volume) for a potential field scale application at the site.

3 Sample Collection and Preparation

GeoSyntec personnel provided a groundwater sample for the treatability study. The groundwater sample (designated as Hofde 42) was collected from the Cheminova site on April 24, 2006 and shipped to ISOTEC. The samples were kept frozen during shipment and at the ISOTEC facility until commencement the test.

Prior to initiating the treatability study experiment, the sample was thoroughly thawed, and portions of the groundwater were sub-sampled and submitted for various chemical analyses (Table 1 below) to determine the sample initial characteristics. The “initial” sample was designated as F-BGC.

Table 1. List of Analytical Parameters

4 Laboratory Treatability Study

The treatability study experiment, hereafter referred to as GW-test, was performed on the groundwater sample. It consisted of the following four steps:

1. Reagent selection,
2. Establishing experimental control,
3. Experimental setup, and
4. Sample analysis.

4.1 Reagent Selection

ISOTEC’s modified Fenton’s reagent contains a proprietary catalyst and an oxidant. The oxidant used in the reagent was H₂O₂ and the catalyst was ISOTEC’s patented Catalyst 4260 (Cat-4260). Cat-4260 is a circum-neutral pH (e.g. 5-8) organometallic complex with high mobility within the subsurface. Based on historical contaminant levels noted at the site and previous experience with treatment of the compounds of concern, ISOTEC selected this catalyst for the experiments. The stoichiometric molar ratio of Cat-4260 to measured site contaminants was determined and then used to prepare the Cat-4260 containing reagents.

4.2 Establishing Experimental Controls

An experimental “control” sample (identified as F-BGC Control) was set up during the experiment to document the following:

• reduction or changes in concentrations of the target constituents due to sample dilution by reagent volumes injected, and
• reduction in concentrations of the target constituents due to volatilization caused by room temperature test conditions.

The “control” sample was set up exactly the same way, remained at, and was subject to the same conditions as the associated “treatment” reactors. However, the “control” reactor was injected with distilled water instead of reagent (see Section 4.3 below). The volume of distilled water injected was identical to the volumes of reagent injected into the “treatment” reactors.

4.3 Experimental Setup

The GW-test was performed in six (6) identical 1-liter glass reactors sealed with screw-up caps fitted with Teflon-liners. One of the six reactors served as “control” reactor (identified as F-BGC Control, see Section 4.2 above) while the remaining five as “treatment” reactors (identified as F-GT-A through E). Exactly 640 ml of groundwater was introduced into each reactor leaving enough headspace for injection of reagent into reactors and receive a variety of reagent dosages to represent different treatment conditions.

For each “treatment” reactor, a predetermined amount of modified Fenton’s reagent was injected as small incremental dosages. ”Treatment” reactors F-GT-A through C received one, two or three reagent dosages and reactors F-GT-D and F-GT-E received one and three reagent doses at a lower reagent level as shown in Table 2 below . In terms of the volume of groundwater being treated, the treatment reactors F-GT-A through E received an equivalent hydrogen peroxide concentration of 2.7%, 5.3%, 7.9%, 0.9% and 2.8%, and catalyst concentration of 5.3 millimoles per liter (mM), 10.6 mM, 15.9 mM, 1.8 mM and 3.7 mM, respectively (Table 2).

Table 2. Treatment Dosage Summary

The multiple dosage approach (incremental approach) was used during the test to increase treatment efficiency, minimize gas formation and the resulting pressure buildup. Distilled water was used to compensate the difference of reagent volumes applied between reactors. The “control” reactor received an equivalent volume of distilled water instead of reagent. A time gap of approximately 24-48 hours was maintained between dosages and the next dosage was provided only after ensuring the peroxide concentration from the previous injection dropped to less than 100 ppm in a given reactor. All reactors (control and treatment) were left undisturbed until the majority of peroxide was consumed before analytical sample collection. Analytical samples were collected by individually decanting water from each of the “control” and “treatment” reactor into corresponding laboratory pre-cleaned and certified containers and submitted for corresponding analyses as summarized in Table 3.

4.4 Sample Analysis

The analytical services were provided by three laboratories – Cheminova Lab of Harboare, Denmark, CAS of Rochester, NY, USA and ETL of Winnipeg, Manitoba, Canada. The following table (Table 3) summarizes the details of the chemical analyses of the treatability study. In addition, ISOTEC also measured H₂O₂ and Fe levels using Hach Test Kits during the experiment. This data was collected to ensure that the majority of injected reagents were consumed in each reactor prior to the next reagent application.

Table 3. Sample Analyses Summary

Note:
1. Anions include nitrate, nitrate-nitrogen, nitrite, nitrite-nitrogen and sulfate.
2. Orthophosphate was reported as phosphorous.

5 Treatability Study Results

Treatability study results are presented in Table 4 and discussed below. Pesticide treatment effectiveness is evaluated by comparison of “treated” sample data with the associated “control” sample data. As discussed in Section 4.2, “control” sample underwent the same conditions as all “treated” samples but received zero dosage of reagent. Therefore, the differences in contaminant concentrations between “treated” samples and the associated “control” sample best represent the treatment effectiveness. For discussion purpose, all non-detect (ND) values (or U values) are assumed to be equal to zero in the contaminant reduction calculation.

5.1 GW-test Results

Results for the GW-test are presented in Table 4. Pesticide and degradation product results are discussed in Section 5.1.1, toxicity results in Section 5.1.2, anion results in Section 5.1.3 and color results in Section 5.1.4.

5.1.1 Pesticide and Degradation Product Results

Pesticide and associated degradation product results are presented in Table 4. Data indicate that the modified Fenton’s reagent has reduced the concentration of all detected pesticides including MCPA, methyl-parathion, parathion and malathion by >99% to ND levels with the lowest treatment evaluated (i.e. 1 dose @0.9% oxidant). Total pesticide levels decreased from a “control” concentration of 14.26 mg/l to ND levels.

Among the pesticide degradation products including MP-1, EP-2 Syre, EP-1, PNF, amino-methyl-parathion, amino-parathion, malaoxon, methyl-paraoxon and paraoxon, each individual compound which showed a detectable “control” concentration was eventually treated to ND levels except for MP-2-Syre (dimethylthiophosphoric acid), which showed increases with each increasing treatment. Please note that the pesticide degradation product MP-1 was detected at low concentration after the lowest treatment but was treated to ND levels following subsequent doses indicating that this compound may appear transiently during field implementation but is eventually mineralized.

The following plot (Figure 1) shows a graphic representation of the results discussed above.

Figure 1 Pesticide and degradation product results as a function of dose

Based on the above plot, it is clear that all pesticides and degradation products except MP-2-Syre were treated to ND for the lowest dose evaluated. For the remainder of the doses, MP-2-Syre concentration more or less followed a linear increasing trend with increasing reagent dose. It may be noted that some MP-2 Syre was treated at the lowest treatment dose (from 4.25 mg/l to 2.55 mg/l), which indicates that MP-2 Syre can be treated by modified Fenton’s treatment. If MP-2-Syre were formed from parathion or other detected pesticide degradation, its concentration should not have increased after the lowest dose since all detected pesticides were completely mineralized with the lowest dose (>99% treatment) and no more parent pesticide mass is available to generate MP-2-Syre as a daughter product. Hence, it is hypothesized that the presence of phosphate in the ISOTEC reagent mix to promote stabilization of hydrogen peroxide may have led to formation of MP-2 Syre via catalytic hydrolysis of transient intermediate compounds from pesticide mineralization or other mechanisms. We believe that addition of phosphate-free reagents will mitigate the potential for MP-2-Syre formation. Even though some MP-2-Syre formation may occur via catalytic hydrolysis of pesticides, this compound should eventually be completely mineralized. Therefore, it is recommended that the modified Fenton’s reagent used for field application be prepared without phosphate. In summary, we believe that MP-2-Syre is amenable to oxidation and will degrade with subsequent treatments.

5.1.2 Toxicity Results

The toxicity of the sample contents was evaluated via a Microtox^® bioassay testing system that measures the light output from freeze dried luminescent bacteria (Photobacterium phosphoreum). The bacteria’s light-producing mechanism is tied to the metabolic processes of the cell. When the bacteria are killed or bacteria's light-producing mechanism is changed or damaged by toxic substances, a reduction in light output results. These changes in light output are measured to calculate an IC50 for each sample, which is defined as the initial concentration, and is a calculated toxicity value representing the sample concentration expressed in % estimated to cause a 50% response by the exposed test organisms.

The toxicity data presented in Table 4 is the calculated sample concentration expressed in % that causes a 50% response by the exposed test organisms. For a fixed exposure time, because the “control” sample is more toxic compared to the “treated” sample, a greater percent concentration of “treated” sample is needed to generate 50% response than the “control” sample. For example, for 5 minutes of exposure time, an “initial” sample concentration of 0.55% is needed versus a “treated” sample concentration of 42.35% to kill 50% of the exposed test organisms. Overall, the results indicate a significant reduction in sample toxicity from “initial/background” to “treated” sample.

5.1.3 Anion Results

Table 3 presents anion results. Individual anion concentrations including nitrate-nitrogen, nitrite-nitrogen and sulfate, and orthophosphate-phosphorous all increased with each treatment possibly due to their formation from pesticide mineralization or their presence in the ISOTEC reagent mix. Overall, the concentrations increased with increasing reagent dosage.

5.1.4 Color Results

Table 3 presents results from the color test. Data indicates fluctuations in color concentrations as a result of the application of modified Fenton’s reagent. We believe the suspended solids present in the samples submitted to CAS may have interfered with color measurements via USEPA Method 110.2.

ISOTEC conducted color measurements on filtered samples via a spectrophotometer prior to shipping the samples to CAS. The color absorbance was measured at a wave length of 470 nano meters (nm). Color intensity is directly proportional to the absorbance (i.e. greater the absorbance greater is the color intensity). The color absorbance data collected by ISOTEC is plotted in Figure 2.

Figure 2 Color absorbance as a function of dose

Results indicate an overall decrease in the sample color absorbance following modified Fenton’s treatment, suggesting decreasing color intensity with increasing treatment.

5.1.5 pH

The final pH value measurements indicate that the treatment occurred in the pH range 3.62-8.30 with the “control” sample present at a pH of 2.91. The following plot shows the pH change as a function of increasing reagent dose. It is clear from the above plot that increasing reagent dose increased the sample pH. The ISOTEC reagents were prepared at a pH of 5-6 with no additional pH adjustment of the sample performed. Under normal buffering associated with addition of pH 5-6 reagents, the final pH was expected to be in the range of 3.0-6.0. Since the final pH of treatment samples B, C, D and E was in the range 7.88-8.30, it is hypothesized that the pH may have increased as a result of alkaline byproduct formation.

Figure 3. pH value as a function of dose

6 Summary and Conclusions

The treatability study consisted of a pesticide GW-test. Results indicate that the modified Fenton’s reagent was effective towards complete destruction of all pesticides and degradation products with the exception of MP-2-Syre. It is hypothesized that MP-2-Syre may have formed due to phosphate presence within the ISOTEC reagent mix via catalytic hydrolysis of transient intermediate compounds from pesticide mineralization or other mechanisms. Therefore, it is suggested that this assumption be tested in a follow-up test using the ISOTEC reagent with no phosphate addition, and if confirmed, field application of modified Fenton’s reagent should exclude phosphate.

Groundwater toxicity levels were significantly reduced following modified Fenton’s treatment. Individual anion concentrations all increased with each treatment following the expected trend. Color absorbance of the samples decreased following the modified Fenton’s treatment indicating decreased color intensity.

The data suggests that a pilot study should be conducted at the site to gather additional data on the effectiveness of this remedial alternative on a large-scale basis under field conditions. A pilot application would also serve as an initial step towards complete site remediation. Upon request, the pilot study design and approach including the reagent volumes to be applied will be provided in a pilot study work plan issued under separate cover.

Tables

Applied Project Number P-1835

HiPOx™ Technology Lab Testing:
Cheminova

Report Authors: Keel Robinson and Reid H. Bowman

Report Date: June 20, 2006

Prepared for

Leah MacKinnon
GeoSyntec Consultants
130 Research Lane, Suite 2
Guelph, Ontario N1G 5G3

APPLIED PROCESS TECHNOLOGY, INC.
3333 Vincent Road, Suite 222, Pleasant Hill, CA 94523
Phone: (925) 977-1811; Fax: (925) 977-1818
www.aptwater.com

Table of Contents

7 Background Information

• 7.1 HiPOx Technology
• 7.2 Objective of Evaluation
• 7.3 Process Water Information

8 Test Equipment and Procedures

• 8.1 Test Equipment Description
• 8.2 Test Procedures

9 Discussion

• 9.1 Ozone Only Results
• 9.2 HiPOx Ozone and Hydrogen Peroxide Results

10 Conclusion

7 Background Information

7.1 HiPOx Technology

The HiPOx™ process developed by Applied Process Technology, Inc. (Applied) is an Advanced Oxidation Process (AOP) that uses ozone (O₃) and hydrogen peroxide (H₂O₂) to destroy organic compounds. Ozone dissociates as well as reacts with hydrogen peroxide to produce an intermediate, hydroxyl radical (•OH). Hydroxyl radicals are the second most powerful oxidizing agent found in nature. These hydroxyl radicals react very rapidly to oxidize organic contaminants to non-hazardous compounds, carbon dioxide, and water.^[1] The oxidation of the organic contaminants does not increase the temperature or pressure of the treated water because of the low mg/L or sub-mg/L concentration of contaminants.

Advanced oxidation chemistry using ozone and hydrogen peroxide to create hydroxyl radicals is base-catalyzed and is very well known. The overall balanced reaction for ozone and hydrogen peroxide to yield hydroxyl radical is shown in Equation 1.^[2]

2O3 + H2O2 → 2•OH + 3O2 (Equation 1)

7.2 Objective of Evaluation

The objective of this laboratory trial was to determine the efficiency of the HiPOx technology utilizing ozone and hydrogen peroxide to reduce the concentrations of pesticides and to reduce the toxicity in the groundwater. The final application will be an in-situ process and not an ex-situ process, however, the goal was to confirm that an Advanced Oxidation Process (AOP) could treat the contaminants before conducting in-situ AOP testing.

7.3 Process Water Information

The groundwater sample for the bench test was collected by GeoSyntec Consultants from the Cheminova Site. The water sample was received on May 8, 2006, and was labeled P-1835. The sample was frozen when received. Once received, the sample was logged in and stored in a freezer to maintain the sample frozen until tests were performed.

8 Test Equipment and Procedures

8.1 Test Equipment Description

The HiPOx reactor is constructed of PVC. All ozone piping is constructed of either 316 Stainless Steel or Teflon. The hydrogen peroxide piping is constructed of either 316 Stainless Steel or polyethylene. The ozone generator utilized was an ASTeX Model 8200. The ozone is injected under pressure into the water to be treated. Mixing of the ozone into the water is accomplished with an in-line static mixer. The vent from the system is passed through an ozone destruct unit manufactured by Pacific Ozone.

8.2 Test Procedures

Test Conditions

The first HiPOx bench test was conducted on May 9-10, 2006. A second HiPOx bench test was conducted on May 15, 2006. The water sample was thawed prior to the first test and immediately tested. The remaining water was stored frozen until the second test. The frozen water was thawed immediately prior to the second test on May 15, 2006. The pH, alkalinity and turbidity were measured on P-1835 prior to the bench testing.

The test procedures are described below. Sample designations reflect sample and treatment levels. For example, SP-190-1835 represents water sample P-1835 treated with 190 mg/L of ozone. There were two types of oxidation methods evaluated in the bench tests, ozone only and Advanced Oxidation (Ozone and Hydrogen Peroxide). The dosing levels in the ozone only tests were limited to the solubility of ozone in the water sample.

Ozone Only Test Procedure:

The test procedure was as follows: A sample (2.05 liters) was treated with ozone (5,530 mls, 5.19% by wt in oxygen, 185.9 mg/L). Samples of the treated water (SP-190-1835) were taken and analyzed for nitrate, nitrite, sulfate, orthophosphate, color, pesticide and Microtox. The amounts of ozone applied during each run are summarized in Table 1 (Run #1) and Table 2 (Run #1).

Advance Oxidation (Ozone and Hydrogen Peroxide) Test Procedure:

The test procedures were as follows. Sample designations reflect sample and treatment levels. For example, SP-500-1835 represents water sample P-1835 treated with 500 mg/L of ozone. The test procedure was as follows: A sample (2.05 liters) was treated with hydrogen peroxide (12.0 mls, 5.0% by wt, 292.2 mg/L). This mixture was then treated with ozone (15,060 mls, 5.04% by wt in oxygen, 491.7 mg/L). Samples of the treated water (SP-500-1835) was taken and analyzed for nitrate, nitrite, sulfate, orthophosphate, color, pesticide and Microtox. The amounts of ozone and hydrogen peroxide applied during each run are summarized in Table 1 (Run #2 & 3) and Table 2 (Run #2).

Table 1. Summary of Experimental Conditions for P-1835 May 9-10, 2006 Lab Test⁵

1 Residual at the end of the Run in the water
2 2.40 wt % ozone was detected in the vent
3 0.39 wt % ozone was detected in the vent
4 0.14 wt % ozone was detected in the vent
5 The analytical samples taken were not quenched

Table 2. Summary of Experimental Conditions for P-1835 May 15, 2006 Lab Test⁴

1 Residual at the end of the Run in the Water
2 2.61 wt % ozone was detected in the vent
3 0.51 wt % ozone was detected in the vent
4 The analytical samples were quenched. The nitrate, nitrite, sulfate, and orthophosphate samples were quenched with methanol, while the pesticide and Microtox samples were quenched with sodium thiosulfate.

Analytical Testing

Samples for nitrate, nitrite, sulfate, orthophosphate from the HiPOx test were sent for analysis to STL San Francisco, 1220 Quarry Lane, Pleasanton, Ca 94566, Phone (925) 484-1919. Samples for Microtox were sent to Envirotest Laboratories, 745 Logan Ave, Winnipeg, MB R3E 3L5 Canada Phone: (204) 945-3705. Samples for pesticides were sent to Cheminova, Attention Bo Breinbjerg Thyborønvej, 78 DK-7673 Harboøre, Denmark Phone: +45 96 90 96 90. The samples collected on May 9 and 10 were not quenched. On May 15, the nitrate, nitrite, sulfate, and orthophosphate samples were quenched with methanol, while the pesticide and Microtox samples were quenched with sodium thiosulfate.

Applied Analytical Methods:

The turbidity meter used was an Orbeco-Hellige Model 965-10 Serial # 2222. The pH was measured with an Oakton Model Ph Tester 3⁺. Alkalinity was measured using a Hach Model 5-EP test kit. Hydrogen peroxide residuals were measured with Quantofix Peroxid 25 test strips. Ozone residual was measured with a Hach Model OZ-2 test kit. The nitrate concentrations were measured using a HACH Model# 820, for the nitrate and Nitraver 5 test reagent.

9 Discussion

9.1 Ozone Only Results

The results and conditions for the treatment of P-1835 with ozone are summarized in Tables 3 (Run #1) and Table 4 (Run #1). The water sample, P-1835, was clear and had a strong chemical odor when the sample was received. The addition of ozone was discontinued when ~ 190 mg/L of ozone was added because the ozone concentration detected in the vent off of the lab reactor exceeded 2.5 weight % in oxygen. The ozone concentration feeding the lab reactor was 5.0 weight % in oxygen. Therefore 50% of the ozone being supplied to the system was being vented. This is attributed to the limited solubility of ozone in water and the relatively slow rate of ozone reacting with contaminants. However, the ozone only treatment of P-1835 resulted in a majority of the pesticides being oxidized with the exception of trace amounts of MP-2-Syre (0.40 mg/L) and 4-chlorocresol (0.47 mg/L). This represents a significant reduction in the concentration of pesticide in P-1835. Three compounds concentration increased after ozone only treatment EOOOPO, Methyl–paraoxon and Paraoxon. These three compounds are oxidation by-products of one or more of the parent pesticides present in P-1835. Also, a significant amount of the pesticides appear to have been mineralized to their elements. The mineralization is based on the amount of nitrate, orthophosphate and sulfate produced after treatment with ozone.

The amounts of nitrogen, phosphorus and sulfur derived from the pesticides are shown in Table 5 (May 10, 2006 SP-190-1835, Ozone Only) and Table 6 (May 15, 2006, SP190-1835, Ozone Only). For example, the calculate amount of nitrogen derived from the pesticides is 1.528 mg/L. The nitrogen is converted to nitrate when oxidized. The nitrogen (1.528 mg/L) in the pesticides is equal to 6.77 mg/L as nitrate. Nitrogen has a molecular weight of 14 and nitrate has a molecular weight of 62. Therefore, the nitrogen concentration is multiplied by 4.43 (62/14) to calculate the amount of nitrogen as nitrate. P-1835 had an initial nitrate concentration of 15 mg/L.^[3] If all the nitrogen in the pesticides had been converted to nitrate, the total nitrate would be 21.8 mg/L. The observed nitrate in the SP-190-1835 on May 10, 2006 was 18 mg/L. This indicates that approximately half of the nitrogen in the pesticides was mineralized.

The phosphorus and sulfur derived from the pesticides are converted to orthophosphate and sulfate, respectively, when oxidized. The corrected phosphorus as phosphate and sulfur as sulfate derived from the pesticides is shown in Table 5 and 6 as “Corrected”. P-1835 had an initial orthophosphate and sulfate concentrations of 4.7 and 460 mg/L, respectively. If all the phosphorus and sulfur in the pesticides had been converted to orthophosphate and sulfate, the total phosphate and sulfate concentration would be 31.3 and 495.5 mg/L, respectively. The observed orthophosphate and sulfate in the Ozone Only test on May 10, 2006 were 13 and 560 mg/L, respectively. The orthophosphate concentration observed indicates that approximately half of the phosphorus in the pesticides was mineralized. The sulfate concentration observed are greater than expected based on the pesticides as the only source of sulfur. There may be other sulfur containing compounds in P-1835 that contribute to the high sulfate concentrations observed. The same trends observed for nitrate, orthophosphate and sulfate were observed in the May 15, 2006 SP-190-1835 (Table 6).

The samples taken on May 10 were not quenched and the samples taken on May 15 were quenched. The concentration of pesticides in both SP-190-1835 taken on May 10 and May 15 were very similar. This indicates that the residual ozone in the samples had very little impact of the pesticides concentration. The nitrate, orthophosphate and sulfate concentrations observed were different. The anion sample SP-190-1835 sample taken on May 10, which was not quenched, had lower nitrate concentrations but higher phosphate and sulfate concentration than the anion SP-190-1835 sample that was quenched on May 15. The anions results for the SP-1000-1835 quenched or not quenched followed the same pattern as the SP-190-1835. The significance in the difference nitrate, orthophosphate and sulfate concentrations observed in the quenched and the sample not quenched is unclear.

9.2 HiPOx Ozone and Hydrogen Peroxide Results

The results and conditions for the treatment of P-1835 with ozone and hydrogen peroxide are summarized in Tables 3 (Run #2 and 3) and Table 4 (Run #2). In Table 3, the results for two ozone doses of 491.7 and 949.0 mg/L are presented. The only pesticide detected in SP-500-1835 was MP-2-Syre (0.32 mg/L). This represents a very significant reduction in pesticide concentration in P-1835. Two oxidation by-products were observed in SP-500-1835, EOOOPO (0.7 mg/L) and Malaoxon (0.2 mg/L). In SP-1000-1835, all pesticides were below their detection limits and no by-products were observed. Although hydrogen peroxide was present to react with the ozone to yield the hydroxyl radical, ozone was detected in the vent. The concentration of ozone detected in the vent was significantly less than in the ozone only tests. This is pointed out because the concentrations of ozone shown in Tables 3 and 4 are the applied ozone doses and the concentration of ozone consumed maybe less than the applied ozone doses.

Using the same analysis of nitrogen, phosphorus and sulfur used to determine the degree of mineralization of the pesticides in the ozone only test, it appears nearly complete mineralization occurred when ozone and hydrogen peroxide were used. The mineralization is based on the amount of nitrate, orthophosphate and sulfate found in P-1835 and the amount of these inorganics after treatment with ozone and hydrogen peroxide. The amounts of nitrogen, phosphorus and sulfur derived from the pesticides are shown in Table 7 (May 10, 2006, SP-1000-1835) and Table 8 (May 15, 2006, SP-1000-1835). P-1835 had an initial nitrate concentration of 15 mg/L (Table 7). If all the nitrogen in the pesticides had been converted to nitrate, the total nitrate would be 21.8 mg/L. The observed nitrate in SP-1000-1835 May 10, 2006 was 21 mg/L. This indicates that greater than 96% of the nitrogen in the pesticides was mineralized.

If all the phosphorus and sulfur in the pesticides had been converted to orthophosphate and sulfate, the total phosphate and sulfate concentration would be 31.3 and 495.5 mg/L, respectively. The observed orthophosphate and sulfate in the SP-1000-1835 on May 10, 2006 were 40 and 470 mg/L, respectively, (Table 7). The orthophosphate concentration observed is higher than can be attributed to phosphorus in the pesticides. There may be other phosphorus containing compounds in P-1835 that contribute to the high orthophosphate concentrations. The sulfate concentration observed was slightly less than expected based on the sulfur in the pesticides.

In the May 15, 2006 SP-1000-1835 (Table 8), the nitrate concentrations in SP-1000-1835 were slightly greater, 24.0 mg/L (observed) vs. 20.7 mg/L (estimated) than can be attributed to the nitrogen in the pesticides. The observed concentrations of orthophosphate and sulfate are consistent with 100 % mineralization of the phosphorus and sulfate in the pesticide.

There were only three toxicology test run, SP-0-1835 on both May 10 and 15 and SP-1000-1835 on May 15, 2006. Applied is not qualified to address the toxicological aspects of the testing.

Table 3 Summary of Results and Conditions for May 10 Test.

1 The value presented is the average of two analysis
ìg/L - micrograms per litre
mg/L - milligrams per litre
v/v (%) - volume per volume percentage
NA - Not Analyzed
NM-Interference with the low pH No measurement could be conducted

Table 4 Summary of Results and Conditions for May 15 Test.

1 The value presented is the average of two analysis
ìg/L - micrograms per litre
mg/L - milligrams per litre
v/v (%) - volume per volume percentage
NA - Not Analyzed
NM-Interference with the low pH No Measurement could be conducted

Table 5. Inorganic Material Balance for Ozone Only May 10, 2006 Test

1 Nitrogen was correct to Nitrate, Phosphorus was corrected to orthophoshate and sulfur was corrected to sulfate
2 The initial concentration for nitrate, orthophoshate and sulfate
3 Sum of the intial concentration plus the inorganic produced

Table 6. Inorganic Material Balance for Ozone Only May 15, 2006 Test

1 Nitrogen was correct to Nitrate, Phosphorus was corrected to orthophoshate and sulfur was corrected to sulfate
2 The initial concentration for nitrate, orthophoshate and sulfate
3 Sum of the intial concentration plus the inorganic produced

Table 7. Inorganic Material Balance for SP-1000-1835 May 10, 2006 Test

1 Nitrogen was correct to Nitrate, Phosphorus was corrected to orthophoshate and sulfur was corrected to sulfate
2 The initial concentration for nitrate, orthophoshate and sulfate
3 Sum of the intial concentration plus the inorganic produced

Table 8. Inorganic Material Balance for SP-1000-1835 May 15, 2006 Test

1 Nitrogen was correct to Nitrate, Phosphorus was corrected to orthophoshate and sulfur was corrected to sulfate
2 The initial concentration for nitrate, orthophoshate and sulfate
3 Sum of the intial concentration plus the inorganic produced

10 Conclusion

HiPOx technology, which uses ozone and hydrogen peroxide, was very effective in the oxidation of the pesticides in P-1835. The pesticide concentrations were below detection limits for all pesticides when an ozone dose of ~950 mg/L of ozone was applied with ~ 650 mg/L of hydrogen peroxide. The pesticides appear to have been mineralized based on the material balance of nitrogen, phosphorus and sulfate.

When testing with ozone only, the ozone dose could not exceed 190 mg/L because of the limited solubility of ozone in water and the relatively slow reaction rate of the ozone with the pesticides. However, the pesticide concentrations in P-1835 were significantly reduced but complete mineralization was not observed with ozone only and oxidation by-products of the parent pesticides were observed.

This series of tests were attempting to address the effectiveness of ozone and ozone/ hydrogen peroxide in an in-situ application. The limitation of ozone only experiments in this test was loss of ozone because of limited solubility of ozone in water. In an in-situ application of ozone, the limited solubility of ozone may not be significant. In an in-situ application the ozone may be applied at a slower rate providing time for the ozone to react with the pesticide contaminants. Also, in an in-situ application the ozone may react with material in the soil to generate hydroxyl radicals.

When requested by GeoSyntec, Applied will provide an engineering cost estimate for a PulseOx in-situ treatment system at the Cheminova site. The PulseOx system offered by Applied has the capacity to inject ozone only and ozone and hydrogen peroxide into the subsurface.

[1] W. Glaze and J. Kang, J.Amer. Water Works Assoc., 80, 51, (1988).

[2] J. Staehelin and J. Holgné, Environ. Sci Technol., 16, 676 (1982)

[3] The nitrate values reported in this section are the nitrate values obtained from STL and not the nitrate values observed with the Applied Hach test.

| Forside | | Indhold | | Forrige | | Næste | | Top |

Version 1.0 Oktober 2007, © Miljøstyrelsen.