L.B.J. van Helden (10574336) Future Planet Studies Major Earth Sciences University of Amsterdam Supervisor: Thomas Wagner Assessor: dhr. dr. J.R. (John) Parsons Assessor: dhr. dr. B. (Boris) Jansen 02-07-2017
The impact of
photodegradation on the
aquatic toxicity of cooling
Abstract
The problem of freshwater scarcity is increasing, and therefore a method is needed for the reuse of cooling tower water in the industry. Cooling tower water is saline and contains chemicals, which should be removed before it can be reused again. A potential pre-treatment method prior to the use of technological desalination might be the use of a natural system in the form of a constructed wetland. Constructed wetlands are designed systems that have been created to utilize natural processes to remove or attenuate waterborne contaminants. Photodegradation is one of those removal mechanisms, which can result in the transformation of a substance into new substances. Those new substances might be more toxic than the initial substance. The aim of this study was to assess differences in toxicity as result of photodegradation. In this study, a comparison was made between the toxicity of cooling tower water chemicals before the photodegradation process and the toxicity of the cooling tower water chemicals after the photodegradation process. Acute toxicity tests with Daphnia magna were conducted with the chemicals benzotriazole, glutaraldehyde, DBNPA, PEG and a mixture of those four cooling tower water chemicals. It was observed that the toxicity of glutaraldehyde after photodegradation, increased significantly (p = 0.0286). The toxicity of PEG after photodegradation decreased, and the toxicity of DBNPA showed no changes. The mixture of all four cooling tower water chemicals together, resulted in a significant decreased toxicity (p = 0.0286). Hence, photodegradation does affect the aquatic toxicity of cooling tower water chemicals. For the establishment of the constructed wetland, the results of the mixture experiments are most relevant. When photodegradation appears to be a useful removal mechanism that would be
incorporated in a constructed wetland, it is not expected that an increase in ecotoxicity of chemicals will occur.
Keywords:
Cooling tower water chemicals Benzotriazole Photodegradation Glutaraldehyde Constructed wetland PEG
Acute toxicity DBNPA
Content
1. Introduction 1.1. Background 1.2. Photodegradation 1.3. Test compounds 1.4. Research questions 1.5. Daphnia magna 1.6. Mixture toxicity 2. Materials and Methods2.1. Test compounds
2.2. Photodegradation experiments 2.2.1. Light intensity
2.3. Acute toxicity tests 2.3.1. Test organisms 2.3.2. Test protocol
2.3.3. Sensitivity test: potassium dichromate 2.3.4. D. magna 48h acute toxicity tests
2.4. Data analysis 3. Results and Discussion
3.1. Light intensity experiments 3.2. D. magna 24h sensitivity test
3.3. D. magna 48h acute toxicity tests
3.3.1. Polyethylene glycol
3.3.2. 2,2-Dibroom-2-cyaanaceetamide 3.3.3. Glutaraldehyde
3.3.4. Mixture 3.3.5. Benzotriazole
3.3.6. Difference in toxicity between all four chemicals 3.4. Discussion on the materials and methods used 4. Conclusion
5. Literature references 6. Appendices
Appendix S1: Daphnia magna Appendix S2: Tables
Appendix S3: Benzotriazole
1. Introduction
1.1. BackgroundThis research is part of the WATER NEXUS project. Multiple research institutes and potential users are collaborating with the aim to develop solutions for problems with freshwater scarcity in delta areas. The slogan is “saline water where possible, freshwater where essential”. Their ambition is to change the paradigm about saline water. Instead of a threat, one should consider it as a resource (TKI Watertechnologie, n.d.; WaterNexus, n.d.).
Most industrial production processes need cooling tower water. A coal-fired power plant cooling tower circulates about 1.000.000 litres of water per minute. In cooling towers with a wastewater stream, there are two water streams: the process water flow, flowing through pipes through the cooling tower, and the cooling water which is sprayed on these pipes. The cooling water absorbs the heat, whereby a part of it evaporates and a part ends up at the bottom of the cooling tower. The cooling water is recycled until a point where the concentration of salts becomes too high and it must be replaced with new cooling water (DOW, n.d.).
Chemical company DOW aims to reuse their cooling tower water and therefore needs a system that is capable and financially attractive. Cooling tower water is saline and contains chemicals that are used to prevent problems such as corrosion, fouling and scaling to occur. Reverse osmosis (RO) membrane technology is generally used to desalinate water (DOW, 2015). However, the added chemicals prevent the normal cooling water treatment technology because they destroy these membranes. A potential pre-treatment method prior to the use of technological desalination might be the use of a natural system in the form of a constructed wetland (TKI Watertechnologie, n.d.). Constructed wetlands are designed systems that have been created to utilize natural processes to remove or attenuate waterborne contaminants (Hijosa-Valsero et al., 2016; Vymazal, 2010). There are several types of constructed wetland beds, but the two basic designs are: subsurface-flow and surface-flow beds (Fig. 1). In the subsurface-flow beds, water may flow horizontally or vertically. In the surface-flow beds, water moves above the substrate surface (Imfeld et al., 2009). Most
constructed wetlands consist of both subsurface-flow beds and surface-flow beds. They are arranged in different manners, including recirculation from one bed to another. With horizontal subsurface-flow, a low oxygen content is available whereas with vertical subsurface-subsurface-flow, aerobic conditions prevail (Imfeld et al., 2009).
Fig. 1. Schematic image of the two basic designs of constructed wetlands: surface-flow and subsurface-flow
Within the constructed wetland system, several mechanisms are involved in order to remove
contaminants. Hijosa-Valsero et al. (2016) listed: microbiological degradation, biofilm, root and plant uptake, evaporation, oxidation, hydrolysis and photodegradation. Each wetland design features a dominant removal mechanism and thus with combining the different systems, different removal mechanisms dominate (Ávila et al., 2014). It is expected that, by combining different wetland designs, it is possible to obtain cooling tower water with low concentrations of the present contaminants.
In recent decades, many experiments have been performed with wastewater treatment in
constructed wetlands. Wastewater from a metallurgic industry contained metals such as chromium and nickel and nutrients like nitrate and nitrite. These substances were largely removed by
constructed wetlands (Maine et al., 2006). Khan et al. (2009) studied the use of constructed
wetlands for the removal of heavy metals from industrial wastewater. Cadmium, iron and chromium were efficiently removed with the constructed wetland. A review of Vymazal (2010) showed that multiple types of wastewater have already been used for the removal of substances with a constructed wetland. Vymazal (2010) listed: petrochemical wastewater, chemical industry
wastewater, textile industry wastewater, food industry wastewater, pig farm wastewater, fish farm wastewater, airport runoff and nursery runoff. Hence, it is expected that it is also possible to treat cooling tower water using a constructed wetland.
1.2. Photodegradation
Photodegradation is one of the removal mechanisms that can occur in constructed wetlands (Hijosa-Valsero et al., 2016). Photodegradation occurs in open waters, where sunlight penetrates the water. Photodegradation results in the transformation of a substance into new substances. This process does not always result in complete transformation of a chemical to end products like CO2 or H2O (Chen et al., 2009; Ma et al., 2014). The new substances might be even more toxic than the initial substance. Gómez et al. (2008) studied the photochemical behaviour of three metabolites of the painkiller drug dipyrone. After photolysis treatment, an increase in toxicity was found. All initial compounds were removed, indicating that the increase in acute toxicity was a consequence of the formation of toxic photoproducts. Another study, by Sinclair & Boxall (2003), compared the acute toxicity of parent and degraded compounds of pesticides. Although the majority of the degradation products appeared to be less toxic or had a similar toxicity as the parent product, 20% of the degradation products appeared to be three times more toxic and 9% turned out to be ten times more toxic than their parent products. Therefore, it would be interesting to study potential differences in aquatic toxicology as result of photodegradation in a constructed wetland. In this study, four representative chemicals that are added to cooling tower water will be tested: benzotriazole, glutaraldehyde, DBNPA, PEG and a mixture of all four chemicals.
1.3. Test compounds
Cooling water systems are the ideal environment for a number of problems such as corrosion, fouling, scaling and biological contamination. Corrosion is the reversion of metal to its original oxide state, which causes loss of metal thickness and eventually leakage of cooling water into the process fluids. A corrosion inhibitor adds a protective film which allows the water to distribute to all wetted parts in the system (Nalco, 2009). Benzotriazole (BTA) is a well-known corrosion inhibitor and has been widely used in cooling fluids (Ding et al., 2010). Fouling is the deposition of solid material in the
cooling tower equipment. Surfactants, such as polyethylene glycol (PEG), reduce surface tension whereby new deposition formation is inhibited. The particles remain in the bulk water flow and from the water flow, they can be removed through filtration or blowdown. Biological contamination is another problem in cooling tower systems, the growth of microorganisms can cause deposit formation which contributes to fouling and corrosion. DBNPA is an oxidizing biocide, which oxidizes important cellular components in microorganisms, causing death of the organisms. Glutaraldehyde is a non-oxidizing biocide, reacting with specific cell components within a microorganism to damage that cell (Nalco, 2009).
1.4. Research questions
The main research question of this study is: “Is there a difference in toxicity before and after the photodegradation of cooling tower water chemicals?”. In order to answer this question, two sub-questions were added: “Is there a difference in toxicity for four separate representative cooling tower water chemicals?”, and “Is there a difference in toxicity before and after the
photodegradation of a mixture of all four cooling tower water chemicals?”.
1.5. Daphnia magna
In this study Daphnia magna (Appendix S1, Fig. S1) will be exposed to water samples containing four representative chemicals that are added to the cooling tower water: benzotriazole, glutaraldehyde, DBNPA and PEG. The Daphnia test will be performed on samples exposed to simulated sunlight and dark controls to compare differences in toxicity.
The test species D. magna is a commonly used test animal in aquatic toxicology because it is easy to culture in a laboratory (Adema, 1978). Furthermore, D. magna is present in most standing
freshwater habitats and plays a key role in the pelagic food webs (Barmentlo et al., 2015; Waaijers et al., 2013). Due to these benefits, it has been widely applied in ecotoxicity assessments and several validated OECD guidelines are available.
1.6. Mixture toxicity
Besides examining the four cooling tower water chemicals separately, the acute mixture toxicity will also be tested. In cooling tower water, the chemicals are also mixed and the water that will be used in the constructed wetlands, will therefore also contain a mixture of several chemicals.
2. Materials and Methods
2.1. Test compounds1H-Benzotriazole (BTA), Glutaraldehyde solution, 2,2-Dibroom-2-cyaanaceetamide (DBNPA) and Polyethylene glycol 1000 (PEG) were purchased from Sigma-Aldrich (Zwijndrecht, the Netherlands). Their applications and toxicity values are listed (Table 1). The initial concentrations that were chosen to test with, were based on these toxicity values (Table 2).
Table 1. List of cooling tower water chemicals, their applications and toxicity towards Daphnia magna (a: EPA (n.d.); b: NITE
(n.d.); c: EnviChem (n.d.))
2.2. Photodegradation experiments
A lab setup was created in which sunlight was simulated with a sunlight imitating lamp. Below the lamp, a glass jar represented the surface-flow bed in the constructed wetland. This jar contained the initial concentration of the chemical in UltraPure water (Appendix S2, table S1). The same procedure was used for each chemical or mixture of chemicals: benzotriazole, glutaraldehyde, DBNPA, PEG, or a mixture of the four chemicals (Table 2). The glass jars were sealed with quartz glass plates to prevent evaporation to occur. The cooling tower water chemicals were exposed to the sunlight imitating lamp, for 48 hours, simulating approximately 5 days in the outside environment. The 48 hours were chosen because of the time that the cooling tower water will be present in the constructed wetland. This will be between 1 and 5 days. After 48 hours, samples were taken from the jars and used in the D. magna 48h acute toxicity test.
Simultaneously, an experiment was performed where the cooling tower water chemicals were not exposed to simulated sunlight. The glass jars in this experiment were also placed beneath the lamp, but were packed in tinfoil (Appendix S2, table S1). After 48 hours, samples were taken from the jars and these dark control samples were also used for the D. magna 48h acute toxicity test.
Substance Application Toxicity Daphnia magna (mg
L-1)
1H-Benzotriazole (BTA) Corrosion inhibitor 91 – 141,6 mg L-1 - 48 h a
Glutaraldehyde solution Biocide 0,75 mg L-1 - 48 h b
2,2-Dibroom-2-cyaanaceetamide (DBNPA)
Biocide 0,5 – 0,9 mg L-1 - 48 h c
Polyethylene glycol 1000 (PEG)
Surfactant No data available
Cooling tower water chemical Initial concentration
1H-Benzotriazole (BTA) 125 mg L-1
Glutaraldehyde 5 mg L-1
2,2-Dibroom-2-cyaanaceetamide (DBNPA) 5 mg L-1
Polyethylene glycol 1000 (PEG) 10 mg L-1
Table 2. List of cooling tower water chemicals and their initial concentrations used in the photodegradation experiments.
2.2.1. Light intensity
For the photodegradation experiments, a sunlight imitating lamp (Arcadia D3 basking lamp, 160 watt) was used and quartz glass plates to prevent evaporation to occur. This lamp emits ultraviolet light (UVA and UVB), visible light and infrared energy. The light intensity was measured with an Ocean Optics Spectrometer. Both the situation under the lamp without the quartz glass plates and with the quartz glass plates were measured. The two resulting graphs were compared with the natural sunlight spectrum.
2.3. Acute toxicity tests 2.3.1. Test organisms
The D. magna neonates (younger than 24 h) used in this study were obtained from the University of Amsterdam (Amsterdam, the Netherlands) (Appendix S1, Fig. S2). The daphnids were cultured in Aachener Daphnien Medium (ADaM) in glass aquaria, in a climate room at 20 ± 1 °C. The daphnids were fed three times a week with an algae mix of Pseudokirchneriella subcapitata and
Chlamydomonas reinhardtii. The algae culture were grown in Cyanobacteria BG-11 Freshwater
Solution (Sigma) at 16°C under continuous light and aeration.
2.3.2. Test protocol
Daphnids (Daphnia magna) were exposed to the selected chemicals in the acute toxicity
experiments and to potassium dichromate in the sensitivity test. For this study, a modified OECD protocol 202 (2004) was used.
Daphnid neonates (younger than 24h) were exposed to the test substances at a range of dilution factors for a period of 48 hours. All test vessels received the same amount of ISO medium, to keep salt concentrations in all test vessels equal (Appendix S2, table S2). Four replicates per dilution were tested. Five daphnids were then placed carefully into the test vessels with a plastic pasteur pipette. The test vessels were loosely covered with plastic foil to reduce the loss of water due to evaporation and to avoid the entry of dust into the solutions. The daphnids were randomly assigned to the test vessels and they were not fed during the experiments. Immobilization was recorded after 24 hours and 48 hours, and compared with control values. Daphnids were considered immobile when they showed no movement for 10 seconds or more, even after ticking the glass gently. In addition to immobility, any abnormal behaviour or appearance was reported.
2.3.3. Sensitivity test: potassium dichromate
A review by Jonczyk and Gilron (2005) about acute and chronic toxicity tests with Daphnia magna, examined the detailed characteristics of the acute test procedures. This review showed that the culture and test conditions of the Daphnia magna sometimes differed considerably from one method to another. Abiotic (temperature, light and pH) and biotic conditions (feeding, strain or test organism) both appear to have a significant influence on the test results.
5 mg L-1 Glutaraldehyde 5 mg L-1 DBNPA
Persoone et al. (2009) therefore studied the desired sensitivity of D. magna with potassium dichromate (K2Cr2O7). As a result, an acceptability range of 0.6-2.1 mg L-1 has been set for the 24h EC50 of the acute D. magna assay.
To determine the 24h EC50 of the D. magna culture used, daphnid neonates were exposed to potassium dichromate at five different concentrations, including control, in ISO medium (Appendix S2, table S3). Each replicate consisted of the appropriate amount of the K2Cr2O7-stock dilution and the appropriate amount of ISO medium (Appendix S2, table S4). The test vessels were placed in a laboratory fume hood. After 24 hours of incubation, each test vessel was checked for immobilised daphnids.
2.3.4. D. magna 48h acute toxicity tests
Daphnid neonates (younger than 24h) were exposed to five compounds, at five different dilutions, including control, in ISO medium (Appendix S2, table S5). Each replicate consisted of the appropriate amount of the stock dilution dissolved in 3 mL of ISO medium and the appropriate amount of the experimental solution. After 24 hours and 48 hours of incubation, each test vessel was checked for immobilised daphnids.
2.4. Data analysis
The number of immobile and mobile daphnids were counted and noted. The percentages
immobilised at 48 hours were plotted against the test dilutions. Data was analysed according to the protocol for calculating LC50.
Using the acute survival data, dilution-response relationships were plotted and the corresponding 48h LC50 values were calculated according to Haanstra et al. (1985) by fitting a logistic curve through the percentage of mobility against the cooling tower water chemicals test dilutions. The LC50 is the concentration value of the chemical in water where 50% of the daphnids is immobilized.
In this equation 𝑦(𝑥) is the mobility at dilution 𝑥(%), 𝑎 is the LC50 (mg L-1), 𝑏 is the slope of the curve, 𝐶 is y(0), which equals the average mobility of the control and 𝑥 is the actual concentration of the cooling tower water chemicals in the water (mg L-1) (Waaijers et al., 2013; Barmentlo et al., 2015). Data analyses were performed with Excel and SPSS software (Microsoft Excel 2013; IBM SPSS Statistics 22.0).
LC50 values of the dark controls and the experiments after photodegradation compared with each other with the Mann-Whitney U statistical test. For each chemical, all individual LC50 values were first calculated. This implies that for each chemical there were four LC50 values calculated of the dark controls and four LC50 values after photodegradation. These individual LC50 values were used in the Mann-Whitney U statistical test. The individual LC50 values were calculated with SPSS
software and the Mann-Whitney U tests were performed with MatLab software (IBM SPSS Statistics 22.0; MATLAB, R2015b). Differences in toxicity were considered to be significant if the computed p-value was less than α = 0.05.
In addition, to answer the question whether there is a difference in toxicity between the four cooling tower water chemicals, the Kruskal wallis test has been performed. For each chemical, all individual LC50 values of the dark controls were used for this test. The Kruskal Wallis test was performed with MatLab software (MATLAB, R2015b).
3. Results and Discussion
3.1. Light intensity experimentsThe photodegradation experiments can only provide useful results if the used lamp emits all spectra that natural sunlight radiates. Natural sunlight consists of ultraviolet light, visible light and infrared energy. Infrared energy consists of wavelengths from above 760 nanometres (nm). Visible light consists of radiation between 400 and 760 nm and ultraviolet light is defined as radiation below 400 nm (Brennan et al., 1988).
The spectrum of the lamp used for the photodegradation experiments emits the spectrum displayed in figure 3 & 4. Both graphs show lower intensities than natural sunlight does (Fig. 2). It can also be seen that the light intensity is lower when measured with the quartz glass plate. The heat of the lamp caused condensation on the quartz glass plates during the photodegradation experiments. The lower light intensities could have been influenced by the condensation, whereby less light could be transmitted. However, all wavelengths are represented in both situations.
Fig. 2. Natural sunlight spectrum (Green Rhino Energy, n.d.)
Fig. 3. Light intensity of sunlight imitating lamp, without interruption of a quartz glass plate.
0 200.000.000.000.000 400.000.000.000.000 600.000.000.000.000 800.000.000.000.000 1.000.000.000.000.000 1.200.000.000.000.000 317 344 371 398 425 452 479 506 533 560 587 614 641 668 695 722 749 776 In te n si ty (m W/(m 2/n m ) Wavelength (nm)
Fig. 4. Light intensity of sunlight imitating lamp, with interruption of a quartz glass plate with condensation.
3.2. D. magna 24h sensitivity test
The D. magna 24h sensitivity test, performed with the reference toxicant K2Cr2O7 showed that the LC50 value of sensitivity of K2Cr2O7 was 0.954 mg L-1. This value falls within the limits (0.6-2.1 mg L-1) proposed by Persoone et al. (2009) and thus the used D. magna culture is acceptable in terms of sensitivity. In this experiment, different quantities of ISO medium were used for the different concentrations (Appendix S2, table S3). This might have influenced this experiment and therefore the results of this sensitivity test. These differences in ISO medium, and thus in salt concentrations, should be avoided in the future.
0 200.000.000.000.000 400.000.000.000.000 600.000.000.000.000 800.000.000.000.000 1.000.000.000.000.000 1.200.000.000.000.000 317 344 371 398 425 452 479 506 533 560 587 614 641 668 695 722 749 776 In te n si ty ( m W/(m 2/n m ) Wavelength (nm) 0,954 0,00% 25,00% 50,00% 75,00% 100,00% 0 2,5 5 7,5 10 M o b ili ty Concentration K2Cr2O7(mg/L)
Data potassium dichromate Potassium dichromate
Fig. 5: D. magna 24h sensitivity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 24h of exposure to potassium
3.3. D. magna 48h acute toxicity tests
During the acute toxicity tests, control survival was in most experiments between 90-100%.
However, benzotriazole after photodegradation, the glass jars which were exposed to light, showed 75% control survival (Appendix S3, Fig. S3). According to the OECD protocol 202 (2004) a test is valid if: “In the control, including the control containing the solubilising agent, not more than 10 per cent of the daphnids should have been immobilized”. This implies that the benzotriazole test after photodegradation is not valid.
3.3.1. Polyethylene glycol
Polyethylene glycol (PEG) became less toxic after photodegradation (Fig. 6). In the dark control experiments, an LC50 of 89.384% was found and after photodegradation a value of 94,56% was found. After photodegradation, the LC50 value is higher which signifies that at a higher
concentration, 50 percent of the daphnids had died. The difference in toxicity between the dark controls and after photodegradation, is not significant (Mann-Whitney U-test; p = 0.1714) (Appendix S4).
There is no literature data available about the acute toxicity of PEG. Although, Steuter et al. (1981) did find evidence that PEG can have toxic effects on plants. According to Lawlor (1970) this occurred due to blocked pathways for water movement and therefore PEG 1000 caused plant desiccation.
89,384 94,56 0,00% 25,00% 50,00% 75,00% 100,00% 0 25 50 75 100 125 M o b ili ty Dilution factor (%) 0% = control group
100% = highest concentration tested
Data PEG before photodegradation PEG before photodegradation
Data PEG after photodegradation PEG after photodegradation
Fig. 6. D. magna 48h acute toxicity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 48h of exposure to PEG,
3.3.2. 2,2-Dibroom-2-cyaanaceetamide
The experiments with 2,2-Dibroom-2-cyaanaceetamide (DBNPA) did not result in differences between the dark control experiments and the photodegradation experiments. Their derived LC50 values were both 5% (Fig. 7).
According to Kim et al. (2002), DBNPA can degrade in water by two pathways. One way is to dibromoacetonitrile first and then to dibromoacetamide through hydrolysis. The second way is to monobromonitrilopropionamide (MBNPA) first and then to cyanoacetamide in the presence of sunlight. This degradation can take place in a relatively short period of time, a matter of hours. Another study, of Klaine et al. (1996), reports that DBNPA is quite toxic to all aquatic life. According to this study, DBNPA yields dibromoacetic acid (DBA) as major degradation product after
photodegradation. DBA is also toxic to aquatic life at very low concentrations. Besides, DBA is also a possible human carcinogen, it has been shown to have carcinogenic effects in laboratory animals. Interestingly, any carcinogenicity is due to the breakdown products of biocides, instead of the initial compound themselves (Kahrilas et al., 2015).
Recapitulatory, DBNPA can either degrade in water without the presence of sunlight, and in water where sunlight is present. From our results, it is not possible to conclude which processes have taken place and which substances have been formed. It can be concluded from our results though, that the substances that were formed during the photodegradation experiments, were equally toxic. Thus the substances that were formed without the presence of sunlight are equally toxic as the substances that were formed with the presence of sunlight.
5 5 0,00% 25,00% 50,00% 75,00% 100,00% 0 25 50 75 100 125
M
ob
ili
ty
Dilution factor (%) 0% = control group100% = highest concentration tested
Data DBNPA after photodegradation DBNPA after photodegradation
Data DBNPA before photodegradation DBNPA before photodegradation
Fig. 7. D. magna 48h acute toxicity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 48h of exposure to DBNPA,
Although, the starting concentration of this experiment might have influenced the obtained results. As initial concentration, 5 mg L-1 was chosen and this may have been too high to discriminate between dark controls and photodegradation. Suppose an initial concentration of 2 mg L-1 would have been chosen, then lower concentrations would have been tested. Table 1 shows that the expected LC50 is between 0,5 - 0,9 mg L-1. This concentration is lower than we have included in our experiments. These lower concentrations could therefore have contained mobile daphnids. This should be tested in further research.
3.3.3. Benzotriazole
In our experiments with benzotriazole, the experiment after photodegradation proved not to be valid due to 75% survival in the control group. Perhaps something went wrong when pipetting the D.
magna into the test vessels, they might have been damaged there. The experiment with the dark
controls, resulted in an LC50 value of 92.051% and the experiment after photodegradation, although not valid, resulted in an LC50 of 87.402%. These results would suggest that an increase in toxicity has occurred.
Hem et al. (2003) classified benzotriazole as ‘‘toxic to aquatic organisms and can cause long-term adversary effects in the aquatic environment”. Table 1 showed a toxicity value between 91 - 141,6 mg L-1 (EPA, n.d.). Seeland et al. (2012) found EC50 values of 79.9-107 mg L-1 (48 h) and Pillard et al. (2001) found EC50 values in the same range, 86-120 mg L-1 (48 h). After photodegradation, it was expected that toxicity would reduce according to the study of Hem et al. (2003). This study found that benzotriazole is not completely mineralised by UV irradiation, but is transformed into
metabolites such as aniline and phenazine. Those metabolites also show toxic effects, but they are not as toxic as benzotriazole. These outcomes cannot be compared with our results. Firstly, because the results of our experiment after photodegradation proved not to be valid due to only 75% survival in the control group. And secondly, due to the lack of LC-MS analyses, there is no information about the concentrations of the initial chemical and the transformation products.
3.3.4. Glutaraldehyde
After photodegradation, glutaraldehyde was considerably more toxic than the dark controls (Fig. 8). In the dark controls, an LC50 of 70.694% was found and after photodegradation a value of 28,474% was found. After photodegradation, the LC50 value is lower which signifies that at a lower
concentration, 50 percent of the daphnids had already died. There is significant difference between the experiments with the dark controls and the experiments after photodegradation with
glutaraldehyde (Mann-Whitney U-test; p = 0.0286) (Appendix S4). Furthermore, the slope of the graph after the bending point is very steep, which suggests that there is a clear toxic threshold point for this chemical.
Glutaraldehyde is an industrial biocide, which is used to control the growth of microorganisms. For
D. magna, glutaraldehyde is considered toxic to aquatic organisms with LC50 value around 0.75 mg
L-1 (48 h) (NITE, n.d.). According to Leung (2001), glutaraldehyde is stable to sunlight in an aqueous environment. We therefore expected that there would be no difference in toxicity between the dark controls and after photodegradation experiments. However, in our experiments glutaraldehyde became more toxic after photodegradation. Due to the lack of LC-MS analyses, there is no
information about the concentrations of glutaraldehyde and its transformation products, therefore it is unclear what reactions have taken place and why the toxicity has increased.
3.3.5. Mixture
The results of the mixture experiments, which contained all initial concentrations of the chemicals benzotriazole, PEG, DBNPA and glutaraldehyde, are shown in Fig. 9. After photodegradation, the mixture became a little less toxic. Whereas in the dark controls all daphnids had died, in the
experiments after photodegradation, a few daphnids remained alive after 48 hours exposure to the mixture. Despite the small distinction, this difference in toxicity is significant (Mann-Whitney U-test; p = 0.0286) (Appendix S4). This significant difference is probably due to the fact that in the
experiments with the dark controls, all LC50 values were 25%. Because there is no variance in this group, the difference between both groups is significant.
The mixture in our experiment contained all same initial concentrations of benzotriazole,
glutaraldehyde, DBNPA and PEG that were used in the experiments with one chemical. Remarkably, in the experiment with DBNPA, all daphnids died, also in the lowest concentration (25%). While in the mixture experiment after photodegradation, not all daphnids died. A cause for this could be the fact that, due to time concerning issues and a shortage of Daphnia magna, the mixture solution has been in the refrigerator for a couple of days. Since DBNPA can degrade in water without the
presence of sunlight in a short period of time, DBNPA might have degraded in the refrigerator into other substances than were initially present after the photodegradation experiments. Another possible cause is that the cooling tower water chemicals might have reacted with each other, during
70,694 28,474 0,00% 25,00% 50,00% 75,00% 100,00% 0 25 50 75 100 125
M
ob
ili
ty
Dilution factor (%) 0% = control group100% = highest concentration tested
Data glutaraldehyde before photodegradation
Glutaraldehyde before photodegradation Data glutaraldehyde after photodegradation Glutaraldehyde after photodegradation
Fig. 8. D. magna 48h acute toxicity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 48h of exposure to
the time in the refrigerator and during the experiments. As a result, less toxic substances could have been formed.
Furthermore, glutaraldehyde was the only chemical that showed a significant difference between the dark controls and the exposure to simulated sunlight. This difference was a significant increase in toxicity after photodegradation. As in the mixture, there was a decrease in toxicity after
photodegradation. This outcome could have been caused by the reactions of the cooling tower water chemicals with each other, whereby less toxic substances could have been formed. It is a possibility that glutaraldehyde reacted with one of the other chemicals into another substance. Hence, the results of the individual glutaraldehyde tests are no longer relevant. LC-MS analyses could provide more information about reaction products that were formed during the
photodegradation experiments.
3.3.6. Difference in toxicity between all four chemicals
There is significant difference between the four cooling tower water chemicals in the dark control experiments (Kruskal-Wallis test; p = 0.0041) (Appendix S4). This result was expected due to the differences between the cooling tower water chemicals. Their applications are different and their toxicity values as well (Table 1).
25 27,003 0,00% 25,00% 50,00% 75,00% 100,00% 0 25 50 75 100 125
M
ob
ili
ty
Dilution factor (%) 0% = control group100% = highest concentration tested
Data mixture before photodegradation Mixture before photodegradation
Data mixture after photodegradation Mixture after photodegradation
Fig. 9. D. magna 48h acute toxicity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 48h of exposure to a
3.4. Discussion on the materials and methods used
The toxicity tests were only performed with test species Daphnia magna, while toxicity is species and compound specific (Waaijers et al., 2013). It should therefore be tested with multiple other species. Furthermore, due to time concerning issues it was not possible for this research to test the water samples after the photodegradation experiments with LC-MS analysis. It would be interesting to know in which concentrations the initial chemicals are still present and which transformation products have been produced. This could give a more complete picture of what exactly happened during the photodegradation process with the initial cooling tower water chemicals.
Furthermore, there is little known about the impact of photodegradation on the aquatic toxicity of chemicals. About PEG and a mixture of several chemicals, there is no knowledge at all about aquatic toxicity. More research is therefore needed concerning toxicity after photodegradation.
4. Conclusion
The aim of this study was to assess differences in toxicity as result of photodegradation. In this study, a comparison was made between the toxicity of cooling tower water chemicals before the photodegradation process and the toxicity of the cooling tower water chemicals after the
photodegradation process. Based on acute toxicity tests with benzotriazole, glutaraldehyde, DBNPA, PEG and a mixture of those four cooling tower water chemicals, it was observed that
photodegradation does affect the aquatic toxicity of cooling tower water chemicals. The toxicity of glutaraldehyde after photodegradation, increased significantly. The toxicity of PEG after
photodegradation decreased, and the experiments with DBNPA did not result in differences in toxicity. The mixture of all four cooling tower water chemicals together, resulted in a significant decreased toxicity. For the establishment of the constructed wetland, the results of those mixture experiments are most valuable.
With incorporating photodegradation in a constructed wetland, by using a surface-flow bed and therefore exposure to sunlight, it is not expected that an increase in ecotoxicity of chemicals will occur. Although, further research and LC–MS analyses are needed to identify reaction products after the photodegradation process.
5. Literature references
Adema, D. M. M. (1978). Daphnia magna as a test animal in acute and chronic toxicity tests.
Hydrobiologia, 59(2), 125-134.
Ávila, C., Bayona, J. M., Martín, I., Salas, J. J., & García, J. (2015). Emerging organic contaminant removal in a full-scale hybrid constructed wetland system for wastewater treatment and reuse.
Ecological Engineering, 80, 108-116.
Barmentlo, S. H., Stel, J. M., van Doorn, M., Eschauzier, C., de Voogt, P., & Kraak, M. H. (2015). Acute and chronic toxicity of short chained perfluoroalkyl substances to Daphnia magna.
Environmental Pollution, 198, 47-53.
Brennan, P., Fedor, C., & Pausch, G. (1988). Sunlight, UV and accelerated weathering. Paint and Resin, 58, 17
Chen, Y., Hu, C., Hu, X., & Qu, J. (2009). Indirect photodegradation of amine drugs in aqueous solution under simulated sunlight. Environmental science & technology, 43(8), 2760-2765.
Coban, O., Kuschk, P., Wells, N. S., Strauch, G., & Knoeller, K. (2015). Microbial nitrogen
transformation in constructed wetlands treating contaminated groundwater. Environmental Science and Pollution Research, 22(17), 12829-12839.
Ding, Y., Yang, C., Zhu, L., & Zhang, J. (2010). Photoelectrochemical activity of liquid phase deposited TiO 2 film for degradation of benzotriazole. Journal of hazardous materials, 175(1), 96-103. DOW (n.d.) See how advanced water treatment helps increase cooling tower efficiency. Retrieved on
2-5-2017 from http://www.dow.com/en-us/water-and-process-solutions/markets/cooling-tower-water/cooling-tower-water-infographic
DOW (2015). Dow Reverse Osmosis Membranes Treat Seawater and Offers Drinking Water to
Southern California. Retrieved on 21-06-2017 from
http://msdssearch.dow.com/PublishedLiteratureDOWCOM/dh_0940/0901b80380940a81.pdf?f ilepath=liquidseps/pdfs/noreg/609-50188.pdf&fromPage=GetDoc
EnviChem (n.d.). Data bank of environmental properties of chemicals. Retrieved on 29-06-2017 from
http://wwwp.ymparisto.fi/scripts/Kemrek/Kemrek_uk.asp?Method=MAKECHEMdetailsform&tx tChemId=2512
EPA (n.d.). EPA (12) Battelle Study. 10/1/85; PMC Specialties Group, Inc. MSDS. (18) Bayer report 248 A/91 D; 1991-07-29. Retrieved on 29-06-2017 from
https://ofmpub.epa.gov/oppthpv/Public_Search.PublicTabs?section=1&SubmissionId=2496179 8&epcount=7&epname=Acute+Toxicity+to+Aquatic+Invertebrates&epdiscp=EcoToxicity+SIDS& selchemid=null&CategorySingle=null
Gómez, M. J., Sirtori, C., Mezcua, M., Fernández-Alba, A. R., & Agüera, A. (2008). Photodegradation study of three dipyrone metabolites in various water systems: Identification and toxicity of their photodegradation products. Water Research, 42(10), 2698-2706.
Green Rhino Energy (n.d.). Solar Power – Radiation. Retrieved on 16-06-2016 from
http://www.greenrhinoenergy.com/solar/radiation/characteristics.php
Haanstra, L., Doelman, P., & Voshaar, J. H. O. (1985). The use of sigmoidal dose response curves in soil ecotoxicological research. Plant and Soil, 84(2), 293-297.
Hem, L. J., Hartnik, T., Roseth, R., & Breedveld, G. D. (2003). Photochemical degradation of benzotriazole. Journal of Environmental Science and Health, Part A, 38(3), 471-481. Hijosa-Valsero, M., Reyes-Contreras, C., Domínguez, C., Bécares, E., & Bayona, J. M. (2016).
Behaviour of pharmaceuticals and personal care products in constructed wetland
compartments: Influent, effluent, pore water, substrate and plant roots. Chemosphere, 145, 508-517.
Imfeld, G., Braeckevelt, M., Kuschk, P., & Richnow, H. H. (2009). Monitoring and assessing processes of organic chemicals removal in constructed wetlands. Chemosphere, 74(3), 349-362.
Jonczyk, E., & Gilron, G. (2005). Acute and Chronic Toxicity Testing with Daphnia Sp. In Small-scale
freshwater toxicity investigations (pp. 337-393). Springer Netherlands.
Kahrilas, G. A., Blotevogel, J., Stewart, P. S., & Borch, T. (2015). Biocides in hydraulic fracturing fluids: a critical review of their usage, mobility, degradation, and toxicity. Environmental science &
technology, 49(1), 16-32.
Khan, S., Ahmad, I., Shah, M. T., Rehman, S., & Khaliq, A. (2009). Use of constructed wetland for the removal of heavy metals from industrial wastewater. Journal of Environmental Management,
90(11), 3451-3457.
Kim, B. R., Anderson, J. E., Mueller, S. A., Gaines, W. A., & Kendall, A. M. (2002). Literature review— efficacy of various disinfectants against Legionella in water systems. Water Research, 36(18), 4433-4444.
Klaine, S. J., Cobb, G. P., Dickerson, R. L., Dixon, K. R., Kendall, R. J., Smith, E. E., & Solomon, K. R. (1996). An ecological risk assessment for the use of the biocide, dibromonitrilopropionamide (DBNPA), in industrial cooling systems. Environmental Toxicology and Chemistry, 15(1), 21-30. Lawlor, D. W. (1970). Absorption of polyethylene glycols by plants and their effects on plant growth.
New phytologist, 69(2), 501-513.
Leung, H. W. (2001). Ecotoxicology of glutaraldehyde: review of environmental fate and effects studies. Ecotoxicology and environmental safety, 49(1), 26-39.
Ma, D., Liu, G., Lv, W., Yao, K., Zhang, X., & Xiao, H. (2014). Photodegradation of naproxen in water under simulated solar radiation: mechanism, kinetics, and toxicity variation. Environmental
Science and Pollution Research, 21(13), 7797-7804.
Maine, M. A., Sune, N., Hadad, H., Sánchez, G., & Bonetto, C. (2006). Nutrient and metal removal in a constructed wetland for wastewater treatment from a metallurgic industry. Ecological
Engineering, 26(4), 341-347.
MATLAB (R2015b). The MathWorks, Inc., Natick, Massachusetts. Microsoft Excel (2013). Microsoft Corporation, Redmond, Washington. Nalco (2009). Cooling Water Treatment. Bulletin B-34.
NITE (n.d.). Chemical Risk Information Platform. Retrieved on 29-06-2017 from http://www.safe.nite.go.jp/english/ghs/09-mhlw-2001e.html
Organisation for Economic Co-operation and Development (OECD), 2004. OECD Guideline for Testing of Chemicals e Daphnia sp., Acute Immobilization Test 202. Section 2: Effects on Biotic Systems.
Persoone, G., Baudo, R., Cotman, M., Blaise, C., Thompson, K. C., Moreira-Santos, M., ... & Han, T. (2009). Review on the acute Daphnia magna toxicity test–Evaluation of the sensitivity and the precision of assays performed with organisms from laboratory cultures or hatched from dormant eggs. Knowledge and management of aquatic ecosystems, (393), 01.
Pillard, D. A., Cornell, J. S., DuFresne, D. L., & Hernandez, M. T. (2001). Toxicity of benzotriazole and benzotriazole derivatives to three aquatic species. Water research, 35(2), 557-560.
Seeland, A., Oetken, M., Kiss, A., Fries, E., & Oehlmann, J. (2012). Acute and chronic toxicity of benzotriazoles to aquatic organisms. Environmental Science and Pollution Research, 19(5), 1781-1790.
Sinclair, C. J., & Boxall, A. B. (2003). Assessing the ecotoxicity of pesticide transformation products. Environmental science & technology, 37(20), 4617-4625.
Steuter, A. A., Mozafar, A., & Goodin, J. R. (1981). Water potential of aqueous polyethylene glycol.
Plant physiology, 67(1), 64-67.
IBM (2013). IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. TKI Watertechnologie (n.d.). Projectaanvraag WATER NEXUS. Technologiestichting STW. Vymazal, J. (2010). Constructed wetlands for wastewater treatment. Water, 2(3), 530-549. Waaijers, S. L., Hartmann, J., Soeter, A. M., Helmus, R., Kools, S. A., de Voogt, P., ... & Kraak, M. H.
(2013). Toxicity of new generation flame retardants to Daphnia magna. Science of the Total
Environment, 463, 1042-1048.
6. Appendices
Appendix S1: Daphnia magnaFig. S1. Daphnia magna
Appendix S2: Tables
Glass jar Treat Quantity: contains initial concentration of
the chemical in UltraPure water
D1 Dark control (wrapped in tinfoil) 150 mL
D2 Dark control (wrapped in tinfoil) 150 mL
P1 Photodegradation (covered
with quartz glass plate)
150 mL
P2 Photodegradation (covered
with quartz glass plate)
150 mL
Table S1. Photodegradation experiment glass jars
Dilution factor Photo-experiment solution Ultrapure
water ISO medium 100 % 27 mL 0 mL 3 mL 75 % 20.25 mL 6.75 mL 3 mL 50 % 13.5 mL 13.5 mL 3 mL 25 % 6.75 mL 20.25 mL 3 mL 0 % 0 mL 27 mL 3 mL
Table S2. Preparing test vessels, for D. magna 48h acute toxicity test
Concentration ISO medium K2Cr2O7 - stock solution
Control 30 mL 0 mL
1 mg L-1 29.7 mL 0.3 mL
2.5 mg L-1 29.25 mL 0.75 mL
5 mg L-1 28.5 mL 1.5 mL
10 mg L-1 27 mL 3 mL
Table S3. D. magna 24h sensitivity test with potassium dichromate
Stock solutions Combine the following
volumes of stock solutions with demi water and top up
to 1L
Substance Amount added to 200 mL
demi water Calcium chloride CaCl2, 2H2O 23.51 g 2.5 mL Magnesium sulfate MgSO4, 7H2O 9.86 g 2.5 mL Sodium bicarbonate NaHCO3 5.19 g 2.5 mL
Potassium chloride KCl
0.46 g 2.5 mL
Table S4. Preparing ISO medium, for D. magna 24h sensitivity test with potassium dichromate
Table S5. Preparing ISO medium, for D. magna 48h acute toxicity test
Stock solutions Combine the following
volumes of stock solutions with demi water and top up
to 1L
Substance Amount added to 200 mL
demi water Calcium chloride CaCl2, 2H2O 23.51 g 25 mL Magnesium sulfate MgSO4, 7H2O 9.86 g 25 mL Sodium bicarbonate NaHCO3 5.19 g 25 mL Potassium chloride KCl 0.46 g 25 mL
Appendix S3: Benzotriazole 92,051 87,402 0,00% 25,00% 50,00% 75,00% 100,00% 0 25 50 75 100 125
M
ob
ili
ty
Dilution factor (%) 0% = control group100% = highest concentration tested
Data benzotriazole before photodegradation Benzotriazole before photodegradation
Data benzotriazole after photodegradation Benzotriazole after photodegradation
Fig. S3. D. magna 48h acute toxicity test. Average mobility (% of initial animals) of Daphnia magna (n = 5) after 48h of exposure to
Appendix S4: Statistical tests – MATLAB scripts
PEG: Mann-Whitney U
Dark_PEG = [85.024; 76.401; 93.372; 85.024] LC50 values dark controls Photo_PEG = [93.372; 86.535; 86.535; 103.530] LC50 values photo experiments [p,h] = ranksum(Dark_PEG, Photo_PEG)
p = 0.1714
Glutaraldehyde: Mann-Whitney U
Dark_Gluta = [60.830; 60.830; 73.246; 73.246] LC50 values dark controls Photo_Gluta = [28.856; 28.856; 27.096; 28.856] LC50 values photo experiments [p,h] = ranksum(Dark_Gluta, Photo_Gluta)
p = 0.0286
Mixture: Mann-Whitney U
Dark_Mix = [25; 25; 25; 25] LC50 values dark controls
Photo_Mix = [27.251; 25.234; 25.234; 27.251] LC50 values photo experiments [p,h] = ranksum(Dark_Mix, Photo_Mix)
p = 0.0286
All four chemicals: Kruskal Wallis
PEG = [85.024; 76.401; 93.372; 85.024] LC50 values dark controls PEG
BTA = [89.615; 79.747; 93.372; 100.857] LC50 values dark controls BTA
Gluta = [60.830; 60.830; 73.246; 73.246] LC50 values dark controls gluta
DBNPA = [5; 5; 5; 5] LC50 values dark controls DBNPA
p = kruskalwallis([PEG BTA Gluta DBNPA]) p = 0.0041
