Fate of Chlorate and Perchlorate in High-Strength and Diluted

current aim of the Homeland Security Department to reduce the usage of chlorine ... In addition, it has a notification and a proposed action level in ...
0 downloads 0 Views 924KB Size
Chapter 7

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

Fate of Chlorate and Perchlorate in High-Strength and Diluted Hypochlorite Solutions Anna Breytus,* Srinivas Prabakar, and Andrew P. Kruzic Department of Civil Engineering, University of Texas at Arlington, Arlington, Texas 76019 *E-mail: [email protected].

Hypochlorite solutions have a potential of introducing disinfection by-products, such as the oxyhalides chlorate, perchlorate and bromate into the drinking water when used for drinking water disinfection. Measurement of these by-products in various strength hypochlorite solutions has been an issue of importance in the last decade, especially in view of the current aim of the Homeland Security Department to reduce the usage of chlorine gas. Previous work identified presence of oxyhalides in both low- and high-strength hypochlorite solutions. Perchlorate is an endocrine disruptor that inhibits iodide intake by the thyroid, thus reducing the production of essential thyroid hormones. Chlorate also has several adverse effects on the blood and thyroid systems. Perchlorate is regulated in the states of Massachusetts and California. Chlorate has a health reference level established by the EPA. In addition, it has a notification and a proposed action level in the state of California. Both contaminants are also being considered for federal regulation. One of the goals of the study was to investigate the degree to which hypochlorite solutions degrade and examine the increase in chlorate and perchlorate concentrations in storage tanks. This was accomplished by comparing chlorine, chlorate and perchlorate levels in storage tanks to newly delivered solutions. Two facilities that had different suppliers were sampled. From a comparison between stored and newly delivered material it was found that significant

© 2015 American Chemical Society Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

hypochlorite degradation takes place. This results in an increase of chlorate and perchlorate levels in the tanks. Dilution of high-strength hypochlorite solutions with softened water and low-strength hypochlorite was examined. The two dilution sources produced similar levels of hypochlorite degradation and chlorate formation. The results were compared to American Water Works Association’s (AWWA) Hypochlorite Assessment Model, which predicts hypochlorite degradation and chlorate production during storage at a constant temperature. Most of the data in the dilution experiments was within 10% deviation from the values predicted by the Hypochlorite Assessment Model. This study also confirmed the results from previous work that in addition to decreasing hypochlorite degradation, dilution also minimizes formation of chlorate and perchlorate.

Introduction Many drinking water treatment facilities have recently switched to hypochlorite as an alternative to chlorine gas due to the dangers associated with the use of chlorine gas. Both chlorine gas and sodium hypochlorite, when added to water, result in the formation of free available chlorine (FAC), primarily in the form of hypochlorite ion or hypochlorous acid (1). FAC is usually expressed in the units of chlorine gas. Some facilities chose to install an onsite hypochlorite generation system (OSG), which typically produces 0.8% FAC hypochlorite solutions. Others use a high concentration hypochlorite provided by an external supplier, which is typically 10-12.5% FAC. It is likely that many more facilities will consider the transition to hypochlorite. However, potential problems associated with hypochlorite storage and use, such as chlorate and perchlorate formation, and excessive chlorine decay, should be considered in the decision process. The focus of the current study is on the high concentration hypochlorite solutions. Hypochlorite manufacture and quality can vary. Hypochlorite also degrades during the storage, thus reducing the chlorine concentration and producing chlorate (2) and perchlorate (3). The loss of chlorine means that more of the solution needs to be applied to achieve the desirable chlorine residual. Therefore, longer periods of hypochlorite storage will lead to higher levels of chlorate and perchlorate in the drinking water. Higher temperatures and direct sunlight contribute to faster hypochlorite degradation. Bromate can also be found in hypochlorite solutions. The main source of bromate in hypochlorite solutions is bromide, which is oxidized during the hypochlorite manufacturing process. Bromide can be present either in the salt or in the water used for the manufacture. The bromate concentration in hypochlorite can be reduced mainly by use of high quality salt and water for hypochlorite manufacture (4). Contrary to chlorate and perchlorate formation, bromate concentrations in hypochlorite do not increase during storage. However, the bromate concentration in the drinking water can increase as a result of 156 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

hypochlorite degradation since more of the solution will be added to the water. Bromate was not the focus of the current study because it does not increase during storage.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

Health and Regulatory Aspects Perchlorate is an endocrine disruptor that inhibits iodide intake by the thyroid, thus reducing the production of essential thyroid hormones that are also very important for neurodevelopment (5). Perchlorate is a common contaminant in food and water. It can come from different sources, such as rocket fuel, fireworks and from Chilean nitrate used for agricultural purposes, where it occurs naturally (6). However, the use of hypochlorite as a disinfectant has a potential of significantly increasing perchlorate levels in drinking water, depending on its concentration in hypochlorite solution. Even though perchlorate is not yet regulated on the federal level, the most recent Environmental Protection Agency (EPA) reference dose is established at 0.0007 mg/kg/day. It would imply a drinking water equivalent of 15 ppb, calculated for an average weight of an adult person, 70 kg (7). The states of Massachusetts and California have established a regulation at the levels of 2 ppb and 6 ppb, respectively, based on either an older reference dose, or because they chose to make adjustments to a different primary vulnerable population. Chlorate is used in agriculture as an herbicide and as a bleaching agent in the textile and paper industry. Chlorate is also a disinfection by-product that is introduced into the water during hypochlorite and chlorine dioxide use (8). Intake of high levels of chlorate has resulted in kidney failure and hemolysis. Animal studies prove that chronic and sub-chronic exposure to chlorate has an adverse effect on blood and thyroid (9). A study performed in Italy showed that women exposed to chlorate levels exceeding 200 ppb in drinking water had an elevated risk of having newborns with obstructive urinary defects, cleft palate and spina bifida (10). Chlorate is included in the Third EPA Contaminant Candidate List (CCL3), and has a health reference level (HRL) of 210 ppb (8). Though chlorate is not yet regulated at the national level, the state of California established a notification level of 800 ppb, while the proposed action level is 200 ppb in drinking water. In addition, the World Health Organization (WHO) suggests a guideline of 700 ppb (11). Hypochlorite Assessment Model Several studies that were performed in the last two decades allowed the development of a hypochlorite decay model. The model was developed using mainly laboratory data, from which kinetics of the reactions were inferred. Initially, the model was developed to express hypochlorite decomposition and chlorate formation (2). The agreement between the predicted and measured hypochlorite and chlorate concentrations in the tested solutions had an error of less than 10 percent (2). The model is based on chemical equations that incorporate temperature and ionic strength. Kinetic information was used to develop chemical 157 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

and thermodynamic equations that show hypochlorite break-down in pH 11-14 range (12). The model describes the effect of the temperature and ionic strength on the decomposition of OCl− , with the following rate constant relationship (2):

The observed rate constant (k2) includes two independent decomposition pathways of hypochlorite, chlorate pathway and oxygen pathway. Chlorate formation is a result of a slow reaction of two hypochlorite ions followed by a relatively fast reaction of the hypochlorite ion with chlorite:

Therefore, chlorate formation is a second order reaction:

Another pathway, which is however very slow, is the breakdown of hypochlorite to chloride and oxygen, without production of chlorate:

Later, the model was extended to predict perchlorate formation (3). The rate law used is as follows:

The agreement between the observed and measured perchlorate concentrations was ±10% or better. Based on these studies, the model software was developed and currently is available as a tool in the American Water Works Association (AWWA) website under the name of Hypochlorite Assessment Model. The model incorporates the prediction of hypochlorite degradation, and chlorate and perchlorate formation during hypochlorite storage (13). 158 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Objectives The objectives of this study were: 1.

2.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

3.

To measure hypochlorite decomposition, and chlorate and perchlorate formation in local systems that are using high-strength hypochlorite in water treatment in the Dallas-Fort Worth (DFW) area, Texas. Compare chlorate and perchlorate levels in these local systems against existing and proposed regulations. Test for possible practical ways to decrease the generation of chlorate and perchlorate, such as dilution of high-strength hypochlorite with softened water produced by an industrial softening system and low-strength on-site generated hypochlorite.

Sample Pretreatment and Analysis Samples were measured for chlorate and perchlorate using ion chromatography (IC) with conductivity detection. This measurement method required pre-treatment of the sample before the analysis. The chlorine needed to be removed prior to the measurement as it is damaging to IC columns. Chlorine removal was achieved by addition of about 20% excess hydrogen peroxide, based on a molar ratio of 1:1 hydrogen peroxide to chlorine. Hydrogen peroxide is also detrimental for IC equipment, therefore it was subsequently removed using manganese dioxide (14). The process showed consistent and reproducible results. Hypochlorite samples were spiked with perchlorate and chlorate standards to check the recoveries of the process. The results of the tests for recoveries of the spiked samples are presented in Table 1. The samples were spiked with three different concentrations, in triplicate for each sample. As can be seen in the table, spike recoveries were above 90%. Calibration curves had correlation coefficients (R2) above 0.99. A more detailed explanation of the pre-treatment method is described in previous work (15). Chlorine levels in hypochlorite were measured using the Hach® DPD colorimetric test (16) and high dilution of hypochlorite or Hach® digital titrator with hypochlorite titration kit (17).

Field Sampling and Results Description and Methodology To assess the first two objectives of the study, there was a need to collect real data from water treatment facilities. One-time sampling from two water treatment facilities was performed in August 2014. Both facilities operate with a residual volume in hypochlorite storage tanks. Hence, tanks are not drained in between the deliveries, but newly delivered material is mixed with the tank content. At each facility, a sample from the storage tank and a delivery truck was collected. Hypochlorite samples were collected into high-density polyethylene (HDPE) bottles and then directly taken to the laboratory using an ice chest 159 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

filled with ice. In the laboratory, they were stored in a refrigerator at 4 ºC. Samples were analyzed for free chlorine, chlorate and perchlorate. Chlorine was measured using digital titrator and hypochlorite titration kit. Theoretical chlorate and perchlorate concentrations in drinking water were calculated based on the hypochlorite strength and application of 5 mg-Cl2/L, which is the average application dose at the facilities (see eq. 8). Both facilities apply hypochlorite along with ammonia to generate monochloramine as their residual secondary disinfectant. A dose of 5 mg-Cl2/L of hypochlorite along with a corresponding amount of ammonia typically results in a residual monochloramine concentration of between 3.5 and 4.0 mg-Cl2/L at both plants. The chlorate and perchlorate concentrations in the finished water are tied to the hypochlorite dose and not the residual monochloramine concentration.

Table 1. Spike Recoveries Data for Perchlorate and Chlorate Perchlorate (spike of freshly delivered sample-concentrations in ppm) not spiked

spiked 1

spiked 2

spiked 3

Not detectable

4.902

9.617

23.37

SD

0.081

0.496

0.111

Average % recovery

98.0

96.2

93.5

%RSD

1.66

5.16

0.48

Average (n=3)

Chlorate (spike of diluted freshly delivered sample-concentrations in ppm) Average (n=3)

35.60

38.48

41.25

51.24

SD

0.268

0.274

1.152

1.837

93.1

92.1

98.1

0.71

2.79

3.59

Average % recovery %RSD

0.75

It should be noted that in this experiment the samples were not tested in replicates for chlorate and perchlorate, therefore standard deviations were not calculated. However, based on the previous measurements, average relative standard deviation (RSD) is 2.4% for perchlorate and 2.0% for chlorate. (calculated from values demonstrated in Table 1). Results from the water treatment facilities are presented in Table 2. Expected chlorate concentration was calculated as follows:

Similarly, the expected perchlorate concentration in finished water was calculated. 160 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Results

Table 2. Chlorine Degradation and Disinfection By-Products Formation in Hypochlorite Storage Tanks in the Month of August Facility

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

Facility I

Facility II

Description

Chlorine, as Cl2

Chlorate, as ClO3–

Perchlorate, as ClO4–

Newly-delivered

108 g/L

947 ppm

94%

Concentration in water assuming 5 mg/L chlorine application dose

N/A

338 ppba

0.025 ppbb

Newly-delivered

133 g/L

1315 ppm

1195%

Concentration in water assuming 5 mg/L chlorine application dose

N/A

467 ppba

0.153 ppbb

Health Reference Level for chlorate: 210 ppb. b Currently regulated at MCL of 2 ppb and 6 ppb in Massachusetts and California, respectively.

a

At Facility I, the chlorine level in the newly delivered material was 108 g-Cl2/L. However, the chlorine level in the tank was lower, 96 g-Cl2/L. This means that there is a hypochlorite degradation as a result of the storage in the hypochlorite tank. In this specific case, the difference in chlorine concentrations between new and stored hypochlorite was 11%. At Facility II, the new material had chlorine concentration of 133 g-Cl2/L, while the stored material was 106 g-Cl2/L. In this case, the difference was 21%. A possible reason for this variation in difference between new and stored material at both plants could be the difference in the strength of the delivered hypochlorite, which is 10% at Facility I and 12.5% at Facility II. According to eq. 4, hypochlorite degradation is a second order reaction on hypochlorite ion, therefore for higher hypochlorite concentration the degradation rate will be higher. Other possible reasons include variation in operational parameters, such as storage time, volume of the tank and level of the solution in the tank. In addition to hypochlorite degradation, chlorate and perchlorate formation takes place. Chlorate and perchlorate levels in the storage tanks were much higher than in new delivered material at both plants (Table 2). To access chlorate concentrations in finished water treated by the facilities, eq. 8 was used. Expected perchlorate levels in water were calculated in a similar way. 161 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

Based on the results from the tank samples from August and assuming a chlorine dose of 5 mg-Cl2/L, chlorate concentrations in the tanks are high enough that it would probably cause the concentrations in water to exceed the health reference level, which is 210 ppb. Despite perchlorate concentrations rising during the storage, the residual concentrations in finished water would probably be below the typical standards (lowest standard is 2 ppb) and therefore seem to be not a concern. August temperatures in DFW often reach triple-digit values, which accelerates hypochlorite degradation and chlorate and perchlorate formation in storage tanks. Two methods to reduce hypochlorite degradation and chlorate and perchlorate formation are: 1. 2.

Increasing the frequency of deliveries and/or reducing the residual storage volume, thus increasing the hypochlorite turnover rate, and Diluting high-strength hypochlorite.

Investigation of Dilution Effects Previous studies indicated that one of the possible ways of reducing the formation of chlorate and perchlorate is dilution of high-strength hypochlorite solution (4). In the current study, high-strength hypochlorite was diluted with softened water and low-strength hypochlorite at various ratios and the impact on hypochlorite degradation and chlorate and perchlorate formation was tested. Description and Methodology Two practical ways for diluting high-strength hypochlorite were tested and then compared to the Hypochlorite Assessment Model: dilution with softened water and dilution with low-strength hypochlorite. Both solutions used for dilution were collected from an OSG water treatment plant. While it is more likely that softened water will be used for dilution, low-strength hypochlorite was also tested to examine whether it yields better results in terms of slower hypochlorite degradation and chlorate formation. The use of low-strength hypochlorite for dilution is possible only for plants that produce OSG hypochlorite, but desire to store bulk hypochlorite for emergency purposes. Therefore, the intention was to check the possibility of a bulk hypochlorite dilution with OSG hypochlorite for such a facility. A sample of high-strength hypochlorite was collected from a delivery truck at one of the facilities that was a part of this study. For the purposes of dilution, low-strength hypochlorite (0.8%) and softened water samples were collected from a facility generating on-site low-strength hypochlorite (0.8%). Hypochlorite samples at different dilution ratios were incubated to check the impact of dilution on the stored material. Five samples at different dilution ratios were incubated at temperature of 40 ºC for 63 days. Temperature of 40 ºC was considered to account for the worst case scenario. August temperatures in 2014 in DFW reached 102 ºF, 39 ºC (18). 162 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

• • •

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

• •

Sample 1: 100% High-strength hypochlorite Sample 2: 50% High-strength hypochlorite + 50% Softened water Sample 3: 50% High-strength hypochlorite + 50% Low-strength hypochlorite Sample 4: 25% High-strength hypochlorite + 75% Softened water Sample 5: 25% High-strength hypochlorite + 75% Low-strength hypochlorite

Samples were placed into 2 L HDPE bottles, and every 7 days a 100 ml sample was collected from each bottle. All the samples were analyzed for chlorine and chlorate. High-strength hypochlorite and one of the samples diluted with 1:1 ratio were analyzed for perchlorate. It should be noted that in this experiment the samples were not analyzed in triplicate for chlorate and perchlorate due to the vast number of samples and laboratory limitations, and because the primary goal of this experiment was to determine the trends rather than exact concentrations. Due to this reason, standard deviations could not be calculated and added to the graphs. Total chlorine measurements were done in triplicate with the DPD Hach® kit. The use of the DPD kit required very high dilutions (20,000:1), which can introduce an error into the results, especially when dealing with low concentrations. This could be a possible reason for several points with deviations from the trend, especially in the four-fold dilution. Results from various dilution experiments are presented in Figures 1-3. Results The degradation of hypochlorite in the incubated solutions is shown in Figure 1. It can be seen that the undiluted sample starts with relatively high chlorine concentration, 112 g-Cl2/L and after 60 days degrades to about 1/3 of its initial chlorine concentration, 36 g/L. The degradation is much slower for solutions that were diluted with 1:1 dilution ratio and slows down even more for solutions with 3:1 dilution ratio. Hence, the decrease in chlorine strength will be higher for higher strength hypochlorite solutions, a fact that corresponds to the previous studies (4). Also, no significant difference can be seen in samples diluted with low-strength hypochlorite (0.8%) and samples diluted with softened water at the same ratios. Similarly to hypochlorite degradation, chlorate formation is much higher in high-strength solutions (Figure 2). There was an about 30 fold increase in the chlorate concentration in the high-strength solution relative to the initial chlorate concentration (from 1.256 g/L to 28.151 g/L). Chlorate formation slows down significantly at 1:1 dilution ratio, and even more at 3:1 dilution. Samples diluted with low-strength hypochlorite (0.8%) and samples diluted with softened water at the same ratios had similar chlorate formation trends. The formation of perchlorate in the high-strength sample was compared to a sample diluted with the softened water at a 1:1 dilution ratio (Figure 3). Perchlorate comparison was limited to these two samples due to laboratory constraints. However, these two samples are sufficient to show that there is a significant decrease in perchlorate formation even at 1:1 dilution ratio. 163 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

Figure 1. Hypochlorite solutions degradation as a function of time and initial concentration at 40 ºC.

Figure 2. Chlorate formation in hypochlorite solutions as a function of time and initial concentration at 40 ºC.

Dilution decreases hypochlorite decomposition and oxyhalide formation by decreasing hypochlorite ion concentration and by decreasing ionic strength, which both contribute to hypochlorite decomposition (2). It was shown in previous work that a two-fold dilution slows down perchlorate formation by the factor of 7 (4), 164 Evans et al.; Trace Materials in Air, Soil, and Water ACS Symposium Series; American Chemical Society: Washington, DC, 2015.

Downloaded by CORNELL UNIV on October 12, 2016 | http://pubs.acs.org Publication Date (Web): December 1, 2015 | doi: 10.1021/bk-2015-1210.ch007

and chlorate formation by the factor of 5 (2). Higher dilution ratios will reduce hypochlorite degradation to a higher extent (6), however, it is recommended that the pH of the diluted solution will be kept between 12 to 13 (19). Following four fold dilution, the pH of the tested samples was 12.12, given that the initial pH of bulk sample was 12.75, as shown in Table 3. In case of lower pH of the original sample, four-fold dilution is more likely to result in pH