Abstract

Groundwater contamination in Thailand from leaking of leachate due to improper solid waste disposal can cause contamination by PFOA (one of the perfluorinated compounds). This study proposed a new idea for the removal of PFOA from groundwater using a combination of membrane filtration and photocatalysis. Spiked groundwater samples were treated by nanofiltration and the rejected part was sent to a UV contact tank for photocatalysis. All samples were analyzed by high-performance liquid chromatography-tandem mass spectrometer (HPLC-MS/MS). The results showed that the removal efficiency of nanofiltration was 99.62%, and the rejected part was degraded by photocatalysis at an efficiency of 59.64%. Thus, the contaminants released to the environment were only 34.23%, which is around three times lower than nanofiltration alone. The results of this technical feasibility study proved that hybrid membrane filtration and photocatalysis are able to remove and degrade the contaminants in the rejected part significantly before being released to the environment, which has been the biggest gap in the processing of membrane filtration, and should be studied further in other aspects, such as fouling effects, energy consumption, and operating costs in a long-term pilot run.

INTRODUCTION

Nowadays, improper disposal of solid waste is a significant problem in Thailand; only 7.2 million tons of municipal solid waste (MSW) received appropriate sanitary management, out of 26.77 million tons of MSW in 2013 (Department of Environmental Quality Promotion 2016). Leaking of leachate can be a cause of perfluorooctanoic acid (PFOA) contamination in groundwater (Stuart & Lapworth 2013). PFOA is one of the predominant perfluorinated compounds (PFCs). PFOA has been used in a variety of products, such as a surfactant in many manufactured products in coating additives, cleaning products, and fire-fighting foam. The characteristics of PFOA are persistence, biological accumulation, toxicity, and long-range transportation (USEPA 2014). USEPA also has recommended that PFOA be labelled as a probable human carcinogen. In addition, PFOA is long-lived and not degradable in the natural environment (State Water Resources Control Board 2010). PFCs are still found in drinking water that comes from drinking water and wastewater treatment in the conventional processing techniques in several countries including the USA, Germany, Switzerland, and other countries.

Conventional oxidative techniques such as UV/H2O2, Fenton process, ozonation, and biological degradation for pollutant control seem not to be suitable for PFC degradation (Lutze et al. 2012). The effective technologies for removal of PFOA from water are nanofiltration (NF) and reverse osmosis. Nanofiltration is one alternative for removing PFOA from groundwater (Federal Provincial Territorial Committee on Drinking Water 2016); the PFOA removal efficiency of spiked groundwater samples by using nanofiltration have been up to 99.49–99.54% (Boonya-atichart et al. 2016). However, a disadvantage of the application of membrane filtration still remains the concentration of contaminants such as PFOA in the reject. For complete destruction of contaminants this flow needs to be incinerated (USEPA 2014).

More recently, advanced oxidation has been used for PFOA degradation, including photocatalysis (USEPA 2016). For the catalyst, nanoscale zero-valent iron (nZVI) is used in in situ applications for soil and groundwater remediation, because nZVI can transform or degrade many environmental contaminants effectively, especially in groundwater (Christensen et al. 2015). After filtration, the concentrated contaminants are degraded by photocatalysis, then the toxic chemicals can be removed before releasing the groundwater to the environment. In other words, eliminating the concentrated pollutants is necessary before releasing water to the environment. Hence, this study proposed the new idea of combining membrane filtration with photocatalysis for the removal of PFOA from groundwater.

MATERIALS AND METHODS

Chemicals and standards

PFOA (>95%) was purchased from the Wako Company (Japan). For solvents, methanol high-performance liquid chromatography (HPLC) grade (99.9%) and acetonitrile HPLC grade (99.8%) were ordered from Merck (Germany). In addition, pure ammonium acetate (≥98%) used for preparing the high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) mobile phase was supplied from Merck (Germany). nZVI (≥65–80%) from Nano Iron (Czech Republic) was used as the catalyst in the photocatalysis. The average particle size and average surface area of the nZVI were 50 nm and 20–25 m2/g, respectively.

Specification of membranes and their equipment set up

Nanofiltration

The main instrument used for removing the target contaminant was the NF membrane. Membrane model 2540-ACM5-TSF 2.5 inches (6.4 cm) in diameter was purchased from Trisep Corporation (USA); the membrane specifications are shown in Table 1.

Table 1

The membrane specification of model 2540-ACM5-TSF for operational and design data (Boonya-atichart et al. 2016)

Membrane Details and operation 
Type ACM fully aromatic polyamide advanced composite membrane 
Configuration spiral wound, fiberglass outer wrap 
Active membrane area 26 ft2 (2.4 m2
Molecular weight cut-off (MWCO) 200 Da 
Recommended applied pressure 100–300 psi (7–21 bar) 
Maximum applied pressure 600 psi (41 bar) 
Recommended operating temperature 35–113°F (2–45 °C) 
Feed water pH range 2–11 continuous 
Chlorine tolerance <0.1 ppm 
Maximum feed flow 6 GMP (1.4 m3/h) 
Minimum brine flow/permeate flow ratio 5:1 
Maximum SDI (15 min) 5:0 
Maximum turbidity 1 NTU 
Permeate flow 800 GPD (3.0 m3/d) 
Average salt rejection 98.5% 
Minimum salt rejection 97.5% 
Membrane Details and operation 
Type ACM fully aromatic polyamide advanced composite membrane 
Configuration spiral wound, fiberglass outer wrap 
Active membrane area 26 ft2 (2.4 m2
Molecular weight cut-off (MWCO) 200 Da 
Recommended applied pressure 100–300 psi (7–21 bar) 
Maximum applied pressure 600 psi (41 bar) 
Recommended operating temperature 35–113°F (2–45 °C) 
Feed water pH range 2–11 continuous 
Chlorine tolerance <0.1 ppm 
Maximum feed flow 6 GMP (1.4 m3/h) 
Minimum brine flow/permeate flow ratio 5:1 
Maximum SDI (15 min) 5:0 
Maximum turbidity 1 NTU 
Permeate flow 800 GPD (3.0 m3/d) 
Average salt rejection 98.5% 
Minimum salt rejection 97.5% 

Ultrafiltration

Another major part of this study was the ultrafiltration (UF) that was used for removing the nanoparticles after the photocatalysis process and before releasing the treated water back into the environment. The hollow-fiber ultrafiltration membrane model UFH-PST-2021 was purchased from Shanghai Mega Vision Membrane Engineering & Technology (China) and the membrane specifications are shown in Table 2. This type of membrane is a hydrophilic polysulfone-modified membrane. The active membrane area was 0.25 m2. The removal of >200 nm particles of membrane was 100%. The pump model was a A-97516688-P1-1432 (Grundfos, Denmark).

Table 2

The hollow-fiber ultrafiltration membrane specification of model UFH-PST-2021 for operational and design data

UF membrane Details and operation 
Type Hydrophilic polysulfone modified 
Configuration Hollow-fiber ultrafiltration module 
Nominal membrane area 0.25 m2 
Operating pressure <14.50 psi (1 bar) 
Maximum applied feed pressure 43.51 psi (3 bar) 
Maximum transmembrane pressure 29.01 psi (2 bar) 
Maximum backwash transmembrane pressure 20.31 psi (1.4 bar) 
Maximum operating temperature 113°F (45 °C) 
Feed water pH range 2–11 continuous 
Instantaneous chlorine tolerance 1,000 ppm 
Continuous chlorine tolerance 200 ppm 
Instantaneous hydrogen peroxide tolerance 200 ppm 
Typical design filtrate flux range 70–150 L/m2/h 
Maximum turbidity 200 NTU 
Filtrate flow 22–36 L/h 
Filtrate turbidity <0.1 NTU 
Maximum SDI (15 min) <2 
Virus and bacterial removal ≥4 log 
Colloidal removal 100% 
TOC reduction 0–50% 
Removal >200 nm particles 100% 
UF membrane Details and operation 
Type Hydrophilic polysulfone modified 
Configuration Hollow-fiber ultrafiltration module 
Nominal membrane area 0.25 m2 
Operating pressure <14.50 psi (1 bar) 
Maximum applied feed pressure 43.51 psi (3 bar) 
Maximum transmembrane pressure 29.01 psi (2 bar) 
Maximum backwash transmembrane pressure 20.31 psi (1.4 bar) 
Maximum operating temperature 113°F (45 °C) 
Feed water pH range 2–11 continuous 
Instantaneous chlorine tolerance 1,000 ppm 
Continuous chlorine tolerance 200 ppm 
Instantaneous hydrogen peroxide tolerance 200 ppm 
Typical design filtrate flux range 70–150 L/m2/h 
Maximum turbidity 200 NTU 
Filtrate flow 22–36 L/h 
Filtrate turbidity <0.1 NTU 
Maximum SDI (15 min) <2 
Virus and bacterial removal ≥4 log 
Colloidal removal 100% 
TOC reduction 0–50% 
Removal >200 nm particles 100% 

Photocatalysis

For the photocatalysis experiments, UV light (254 nm) and nanoparticles are the important factors of the process. The catalyst used for the reaction was nZVI. The UV contact tank, which contained a UV lamp hung in the middle of the tank, and nanoparticles are the main elements of photocatalysis. The diameter and height of the tank were 25 cm and 45 cm, respectively. The length of the UV lamp was 26 cm. The schematic diagram of the UV contact tank is shown in Figure 1 and photographs of the hybrid nanofiltration photocatalysis unit are shown in Figure 2.

Figure 1

Schematic diagram of hybrid membrane filtration and photocatalysis.

Figure 1

Schematic diagram of hybrid membrane filtration and photocatalysis.

Figure 2

The hybrid NF membrane and photocatalysis operation unit, showing the front of the unit (left) and the back of the unit (right).

Figure 2

The hybrid NF membrane and photocatalysis operation unit, showing the front of the unit (left) and the back of the unit (right).

Operation with synthetic samples

Nanofiltration

Spiked deionized water containing 100 μg/L PFOA was used as the synthetic feed. The three operation pressures were controlled at 2, 4, and 6 bar, respectively. After that, the experiments of varied PFOA concentration at 5 and 100 μg/L were carried out with the fixed pressure operation at 6 bar.

Photocatalysis

Spiked deionized water was used as the synthetic feed. It was controlled at 100 μg/L of PFOA and the experiment focused on finding the suitable nZVI concentration. Various nZVI concentrations were tried, i.e. 20, 40, 60, 80 and 100 mg/L, and were used in the photocatalysis for the same reaction times: 1, 5, 10, 15, 30, 45 and 60 min.

Hybrid process of nanofiltration and photocatalysis experiments

For the hybrid process of nanofiltration and photocatalysis, synthetic and groundwater samples with 100 μg/L of PFOA were used as samples for comparing the removal efficiency of nanofiltration and hybrid membrane filtration (using residual PFOA concentration to indicate the removal efficiency of nanofiltration and the hybrid process). The groundwater samples were collected near a landfill in Nakhon Pathom province, Thailand. Furthermore, the mass balance of the hybrid process is presented. The removal efficiency was calculated from the equation below.  
formula

Sample collection during experiment

The samples of the synthetic and groundwater experiment were collected from NF influent (NF feed tank), NF retentate (every 8 min until the water ran out), the UV contact tank (at 1, 5, 10, 15, 30, 45, and 60 minutes), and UF retentate (at 2, 4, 6, 8, and 10 minutes). The samples collected from the UV contact tank were filtered through a 0.02 μm syringe filter for the removal of nanoparticles before being analyzed by high-HPLC-MS/MS. All samples were injected and analyzed by HPLC-MS/MS. The samples collection points of the hybrid process between the NF membrane and photocatalysis are shown in Figure 3.

Figure 3

The sample collection points of the hybrid process between the NF membrane and photocatalysis.

Figure 3

The sample collection points of the hybrid process between the NF membrane and photocatalysis.

Instrumental analysis

Quantification of PFOA was performed by using Agilent 1200SL HPLC (Agilent Technologies, USA) which interfaced with an Agilent 6400 Triple Quadrupole mass spectrometer (MS/MS, Agilent Technologies, USA). The protective guard column was Agilent ZORBAX Eclipse XDB-C18 (4.6 × 50 mm, 1.8 μm particle size) and the series connect was with analytical column Agilent ZORBAX Eclipse Plus C18 (2.1 × 100 mm, 1.8 μm particle size). The column was maintained at 40 °C. For optimum separation, a binary gradient consisting of 10 mM ammonium acetate (solvent A) and acetonitrile (solvent B) was used at a flow rate of 0.25 mL/min. The elution gradient setting was: 45% (B); 0–5 min: 50%; 5–5.5 min: 60%; 5.5–10 min: 60%; 10–15 min: 90%; back to initial conditions for 10 min. The total running time was 25 min for each sample. The injection volume was 10 μL. For quantitative analysis, the mass spectrometer was operated with the electrospray ionization (ESI) negative mode. Multiple reaction monitoring (MRM) mode was used to monitor analyte ions. Capillary voltage was 3500 V. Gas temperature and gas flow were 300 °C and 10 L/min, respectively (Boonya-atichart et al. 2016). HPLC-MS/MS conditions are shown in Table 3.

Table 3

HPLC-MS/MS conditions for analysis of PFOA by MRM in negative ion mode (Boonya-atichart et al. 2016)

Compound Precursor ion (m/z) Product ion (m/z) Dwell time (ms) Collision energy (eV) Retention time (min) Polarity 
PFOA 413 369 50 4.0 Negative 
Compound Precursor ion (m/z) Product ion (m/z) Dwell time (ms) Collision energy (eV) Retention time (min) Polarity 
PFOA 413 369 50 4.0 Negative 

RESULTS AND DISCUSSION

Results of nanofiltration and photocatalysis tests with synthetic water

The effect of different pressures and concentrations

The effect of different pressures and concentrations of spiked deionized water samples on flow rate of the permeate in nanofiltration are shown in Figure 4. For the flow rate of the different concentrations experiment, for both the concentrations of 5 and 100 μg/L PFOA, the change was the same direction, which is that the flow rates decreased when the experiment times were increased, but not significantly. For the different pressures experiment, the results show that the pressures did affect the permeate flow rates, because higher pressures provided higher flow rates and were significant to the experimental run. So, PFOA accumulation on the membrane surface probably caused the flux decline (Tang et al. 2006).

Figure 4

The flow rate of permeate at different pressures and concentrations.

Figure 4

The flow rate of permeate at different pressures and concentrations.

PFOA removal efficiencies by nZVI concentration variation

The spiked deionized water was used as samples for finding the nZVI dosage. The PFOA removal efficiencies of nZVI concentrations at 20, 40, 60, 80, and 100 mg/L are shown in Figure 5, and were: 49.95–64.81%, 68.85–73.99%, 72.29–78.75%, 77.58–80.34%, and 80.14–84.98%, respectively. From these results, the nZVI concentration of 100 mg/L had the highest removal efficiency, in other words, when nZVI was used at the high concentration, the removal efficiency was higher than at the low nZVI concentration. According to a previous study, when the nZVI concentration modified with Mg-aminoclay (MgAc) is increased, the PFC removal efficiencies increase (Arvaniti et al. 2014). In this study, though nZVI was not modified with other materials, increasing the concentration still affected the removal efficiency. For the effect of reaction times of 1, 5, 10, 15, 30, 45, and 60 min on PFOA removal efficiency, Figure 5 exhibits a trend: that the nZVI reactions with the PFOA were rapid at the 1-min reaction time, and higher but not that much higher at the 60-min reaction time. Similar to a previous study, the PFOA removal efficiency using nZVI was 92.77 ± 1.26% at 1 min of reaction time, while the removal efficiency at the reaction time of 60 min was 96.24 ± 0.94%, which was almost identical (Khatikarn 2009).

Figure 5

The PFOA removal efficiency at different nZVI concentrations.

Figure 5

The PFOA removal efficiency at different nZVI concentrations.

Results of hybrid nanofiltration and photocatalysis with groundwater

Nanofiltration

For the nanofiltration part of the experiment, the pressure was operated at 6 bar and the samples were collected every 8 minutes. The PFOA removal efficiencies of the spiked groundwater and deionized water samples were 98.81–99.22% and 99.15–99.94%, respectively, as shown in Figure 6. Another study found that nanofiltration could reject up to 90–99% of perfluorooctane sulfonate (PFOS), which is one of the PFCs (Tang et al. 2007). As the results illustrate, the spiked groundwater removal efficiency was not much different from the spiked deionized water under the same conditions, even though the removal efficiency of the spiked deionized water sample was slightly higher. This is because the co-contaminants in groundwater did not majorly affecting the PFOA removal efficiency by membrane filtration, while the removal efficiency of membrane filtration depends on the size of pollutants and membrane pore size.

Figure 6

The PFOA removal efficiencies of spiked deionized water samples and spiked groundwater samples by nanofiltration in the hybrid membrane system.

Figure 6

The PFOA removal efficiencies of spiked deionized water samples and spiked groundwater samples by nanofiltration in the hybrid membrane system.

Photocatalysis

For the photocatalysis part of the experiment, the feedwater samples were sent from the rejected part of nanofiltration to the UV contact tank. In Figure 7, the results show the PFOA removal efficiencies of the spiked groundwater and deionized water samples were 58.72–62.09% and 72.07–75.83%, respectively, making PFOA removal efficiency for the spiked deionized water samples higher than for the spiked groundwater samples. This is because the co-contaminants in groundwater reacting with easily degradable organic compounds beforehand.

Figure 7

The PFOA removal efficiencies of spiked deionized water samples and spiked groundwater samples by photocatalysis in the hybrid membrane system.

Figure 7

The PFOA removal efficiencies of spiked deionized water samples and spiked groundwater samples by photocatalysis in the hybrid membrane system.

Even though the removal efficiencies of the spiked deionized water sample were higher than those of the spiked groundwater sample, nevertheless the removal efficiency still was low, due to the UV contact tank not having a mixer for mixing nanoparticles. The fact that the UV light bulb was only in the middle of the tank might not have been good enough, so it could be another cause of the low efficiency of the photocatalysis. These points should be considered in a further study. The previous study which investigated the combined process of photocatalysis and ozonation (UV/TiO2/O3) found the PFOA degradation efficiency was 99.1% after 4 hours' reaction time (Huang et al. 2016). Besides, the size of this experiment was a pilot-scale project, and scaling up the experiment to batch from pilot scale might decrease the removal efficiency even if the operating conditions are controlled. According to our previous study in batch scale, the PFOA removal efficiency of spiked deionized water by nZVI photocatalysis reached 80.14–84.98% (Boonya-atichart 2017).

Mass balance in hybrid nanofiltration membrane and photocatalysis

The mass balance of the hybrid nanofiltration and photocatalysis process was determined to discover the fate of mass when the contaminants were treated by the hybrid nanofiltration and photocatalysis operation unit, to show the outcome of the improved system. The average PFOA concentration was multiplied by the water sample volume for determining the mass of PFOA. The water sample volume was calculated from the flow rate in each part of the operation unit. The actual flow rates were measured on the experimental run day. The PFOA mass was calculated from the following equation:  
formula
The volume and concentration of the test system are presented in Figure 8. The volume of the water samples in the NF feed tank, permeate, retentate, and UF effluent were 34 L, 14.47 L, 19.53 L, and 19.53 L, respectively. The PFOA average concentration in the NF feed tank, permeate, retentate, and UF effluent were 98 μg/L, 0.87 μg/L, 169.96 μg/L, and 58.40 μg/L, respectively, as shown in Figure 8. For the retentate, the water sample volume and PFOA concentration of all rejected parts were 19.53 L and 169.96 μg/L, respectively, for which the PFOA concentration after photocatalysis (in the UV contact tank) was 68.60 μg/L. After calculating the mass in each part, the mass balance of the system was calculated and shown as percentages in Figure 9 based on the calculation by the following equation:  
formula
Figure 8

Volume and concentration of water samples of the hybrid membrane system.

Figure 8

Volume and concentration of water samples of the hybrid membrane system.

As shown in Figure 9(b), at the NF feed tank before treatment by the nanofiltration membrane, the mass of PFOA was 100% and after treatment by the nanofiltration membrane, the residual of PFOA in permeate was 0.38% and the rejected part sent to the UV contact tank was 99.62%. In the photocatalysis part, the mass of PFOA after treatment by photocatalysis was 40.21%, so the PFOA that was degraded by photocatalysis was 59.41% and the average removal efficiency of photocatalysis was up to 59.64%. Then the sample was sent to the ultrafiltration membrane for removing the nanoparticles before releasing the treated water to the environment. The mass of PFOA released to the environment was 34.23%; thus, 5.98% of PFOA was trapped by the ultrafiltration membrane.

Figure 9

Comparison of mass balance and removal efficiencies: (a) conventional membrane filtration and (b) hybrid membrane system.

Figure 9

Comparison of mass balance and removal efficiencies: (a) conventional membrane filtration and (b) hybrid membrane system.

The comparison between the nanofiltration system (Figure 9(a)) and the hybrid membrane system (Figure 9(b)) shows the difference in the amounts of contaminants that were released to environment from each system. For the nanofiltration system, up to 99.62% of the contaminants were released to the environment after being treated by nanofiltration. In contrast, the results of the hybrid membrane system show that some concentrated PFOA after filtration was degraded by using photocatalysis, so that the toxic chemical that was removed before being released to the environment was just 34.23%, which is significantly lower than nanofiltration only.

CONCLUSIONS

The hybrid membrane filtration and photocatalysis system was tested with PFOA. The transmembrane pressures and PFOA accumulation on the membrane surface probably caused the flux decline. Moreover, the nZVI reactions with the PFOA were rapid at the 1-minute reaction time. The combination of membrane filtration and photocatalysis not only removed the PFOA from the water, but also degraded the contaminant released to the environment to a level at least three times lower, based on this study. Therefore, this new hybrid membrane system will be beneficial for reducing the release of rejected contaminants to the environment, and will strengthen the productive use of membrane technology. The concept of hybrid membrane filtration and photocatalysis has been shown to be an environmentally friendly system, and should be studied further for exploring aspects such as fouling effects, energy consumption, and operating costs in a long-term pilot run.

ACKNOWLEDGEMENT

This study was supported by research funding from the Thailand Research Fund (TRF) RSA5880046 and the Faculty of Engineering, Mahidol University, Thailand.

REFERENCES

REFERENCES
Arvaniti
,
O. S.
,
Hwang
,
Y.
,
Andersen
,
H. R.
,
Nikolaos
,
T. S.
&
Athanasios
,
S. S.
2014
Removal of Perfluorinated Compounds From Water using Nanoscale Zero-Valent Iron
. In:
Paper Presented at the Singapore International Water Week
,
Marina Bay Sands Expo & Convention Centre
,
Singapore
.
Boonya-atichart
,
A.
2017
Study of Hybrid Membrane Filtration and Photocatalysis for the Removal of Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoic Acid (PFOA) in Groundwater
.
Master of Engineering
,
Mahidol University
,
Thailand
.
Boonya-atichart
,
A.
,
Boontanon
,
S. K.
&
Boontanon
,
N.
2016
Removal of perfluorooctanoic acid (PFOA) in groundwater by nanofiltration membrane
.
Water Science and Technology
74
(
11
),
2627
2633
.
Christensen
,
F. M.
,
Brinch
,
A.
,
Kjølholt
,
J.
,
Mines
,
P. D.
,
Schumacher
,
N.
,
Jørgensen
,
T. H.
&
Hummelshøj
,
R. M.
2015
Nano-enabled Environmental Products and Opportunities and Drawbacks
.
The Danish Environmental Protection Agency
,
Denmark
.
Department of Environmental Quality Promotion
2016
Waste and Municipal Solid Waste Management in Thailand
.
Retrieved 29 July 2017, from Department of Environmental Quality Promotion. http://www.environnet.in.th/en/archives/3708
.
Federal Provincial Territorial Committee on Drinking Water
2016
Perfluorooctanoic Acid (PFOA) in Drinking Water
. .
Huang
,
J.
,
Wang
,
X.
,
Pan
,
Z.
,
Li
,
X.
,
Ling
,
Y.
&
Li
,
L.
2016
Efficient degradation of perfluorooctanoic acid (PFOA) by photocatalytic ozonation
.
Chemical Engineering Journal
296
,
329
334
.
Khatikarn
,
J.
2009
Application of Nanoparticles for Removal of Perfluorinated Compounds in Synthetic Wastewaters
.
Master of Engineering
,
Asian Institute of Technology School of Environment, Resources and Development
,
Thailand
.
Lutze
,
H.
,
Panglisch
,
S.
,
Bergmann
,
A.
,
Schmidt
,
T. C.
2012
Treatment options for the removal and degradation of polyfluorinated chemicals
. In:
Polyfluorinated Chemicals and Transformation Products
(
Knepper
,
T. P.
&
Lange
,
F. T.
, eds).
Springer Berlin Heidelberg
,
Berlin, Heidelberg
, pp.
103
125
.
State Water Resources Control Board
2010
Groundwater Information Sheet Perfluorooctanoic Acid (PFOA) & Related Compounds
. .
Stuart
,
M.
,
Lapworth
,
D.
2013
Emerging organic contaminants in groundwater
. In:
Smart Sensors for Real-Time Water Quality Monitoring
(
Mukhopadhyay
,
S. C.
&
Mason
,
A.
, eds).
Springer Berlin Heidelberg
,
Berlin, Heidelberg
, pp.
259
284
.
Tang
,
C. Y.
,
Fu
,
Q. S.
,
Robertson
,
A. P.
,
Criddle
,
C. S.
&
Leckie
,
J. O.
2006
Use of reverse osmosis membranes to remove perfluorooctane sulfonate (PFOS) from semiconductor wastewater
.
Environ. Sci. Technol.
40
(
23
),
7343
7349
.
USEPA
2014
Emerging Contaminants – Perfluorooctane Sulfonate (PFOS) and Perfluorooctanoic Acid (PFOA)
.
From USEPA
.
USEPA
2016
Drinking Water Health Advisory for Perfluorooctanoic Acid (PFOA) (EPA Document Number: 822-R-16-005)
.