INTRODUCTION
Although the signatory countries to the Stockholm Convention (185 countries) have taken strong actions to control persistent organic pollutants (POPs), some residues remain in the air, water, sediments and biota (Kannan et al., 1995; Hung et al., 2013; Olisah et al., 2021; Schuster et al., 2021). Among the POPs, hazardous perfluoroalkyl substances (PFAS) are of great concern, as they can exert adverse health effects owing to their potential toxicity and bioaccumulation. However, given their excellent chemical stability and water repellency, PFAS are routinely used in industry and consumer products (Costello et al., 2022; Tian et al., 2022). Two most prevalent PFAS, perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), exhibit potential toxicity, in vivo accumulation, persistence and long-distance transport (Li et al., 2022). Globally, the emission of PFOS and PFOA has been monitored according to the Stockholm Convention and 2010/2015 PFOA Stewardship Programme, respectively (United Nations Environment Programme, 2019; United Nations Environment Programme, 2022). However, the potential toxicity of these compounds has been reported recently (Feng et al., 2021; Fenton et al., 2021).
In Japan, manufacturing and importing PFOS (or its salts) were banned after it was designated as a Class 1 specified chemical substance in 2010. Current legal uses of PFOS (or its salts) include the production of etching chemicals, semiconductor resistors and photographic films. To control these environmental pollutants, in May 2021, the Japanese government set the provisional water quality guideline value (sum for PFOS and PFOA) to 50 ng/L as per a notification from the Director of the Water and Air Environment Bureau of the Ministry of the Environment: Enforcement of Environmental Standards for the Protection of Human Health Related to Water Contamination. From a nationwide survey in 2020, 21 of 143 sites in 12 prefectures were confirmed to exceed the provisional target value for organic fluorine compounds (Water Environment Division, Water and Air Environment Bureau, & Ministry of the Environment, 2021). Therefore, there is a strong demand for effective water purification involving PFOS and PFOA removal.
The use of sorbent materials is the most economical, efficient and common method to remove PFAS from aqueous solutions (Water Environment Division, Water and Air Environment Bureau, & Ministry of the Environment, 2021; Gagliano et al., 2020; Aung et al., 2022). Japanese water purification systems mainly use GAC to adsorb organic chemicals. Previous studies on the adsorption of PFOS and PFOA from aqueous solutions have mainly used GAC and powdered AC (PAC) (Water Environment Division, Water and Air Environment Bureau, & Ministry of the Environment, 2021; Gagliano et al., 2020; Park et al., 2020; Aung et al., 2022). In the water purification process, the selection of materials such as AC that can efficiently adsorb and remove PFOS and PFOA remains a major challenge from the viewpoints of cost, efficiency of organic pollutant removal and disposal of used adsorbents. Determining the most economical and efficient materials for PFOS and PFOA removal is essential (Oyetade et al., 2018).
To address this challenge, in the present study, we evaluated five types of commercially available AC and their physiochemical properties. To the best of our knowledge, this is the first study to evaluate the performance of AC in adsorption of PFOS and PFOA at low concentrations (ng/L), reflecting their actual concentrations in environmental water.
MATERIALS AND METHODS
PREPARATION OF ACs
Commercially available ACs, purchased from Futamura Chemical Co., Ltd., Japan, were utilised in the experiments. ACs were oven dried at 105°C for 24 h and weighed to determine the moisture content.
CHARACTERISATION OF ADSORBENTS
ACs were characterised by measuring the Brunauer-Emmett-Teller (BET) surface area and using scanning electron microscope (SEM) analysis.
ANALYSIS OF BET SURFACE AREA
The BET surface areas of the five ACs selected as adsorbents for PFOS and PFOA were assessed using an automatic adsorption device (autosorbiQ; Anton Paar, Graz, Austria). After degassing the sample at 105°C for 12 h, liquid nitrogen was adsorbed and desorbed. Data were analysed using ASiQ version 1.1 (Anton Paar, Graz, Austria) to generate the BET surface area of the five ACs.
SEM ANALYSIS
The morphological features of the surfaces of the ACs were analysed using SEM (JSM-5610 LV; JEOL, Tokyo, Japan). All samples were coated with a fine layer of platinum (Pt) for 20 s at 80 mA using a fully automated sputter coater (JFC-1600; JEOL) to observe the surface morphology (1,000× magnification).
STANDARDS, REAGENTS AND MATERIALS
PFOS and PFOA were supplied by Wellington Laboratories Japan Inc. (Tokyo, Japan). Ammonia solution (25%), ammonium acetate (97%) and methanol (MeOH, >99.8%; pesticide residue-PCB analytical grade) were supplied by FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan). Milli-Q water (18 MΩ/cm) was obtained using the Millipore Milli-Q Gradient A10 water purification system (Billerica, MA, USA). Solid-phase extraction weak anion exchange cartridges (Oasis WAX 150 mg) were supplied by Waters (Wilmslow, UK).
DETERMINATION OF PFOS AND PFOA REMOVAL EFFICIENCY IN WATER
To compare the adsorption performance for PFOS and PFOA in water, the adsorbents were set to blank, 1 mg/L and 10 mg/L.
Briefly, PFOS and PFOA solutions at a concentration of 100 ng/L each were prepared in polypropylene (PP) bottles using Milli-Q water (pH 5.6) and 2 mM phosphate buffer to 0.00004 N for a total volume of 50 mL. In the adsorption experiment, parallel tests were conducted in triplicate for each concentration and adsorption material.
The sample was shaken at 140 rpm for 72 h at 25°C. Then, 50 mL of the supernatant was dispensed into a PP tube and centrifuged at 3,000 rpm for 30 min, from which 40 mL of the supernatant was separated and transferred into a new PP bottle to prepare a water sample for solid-phase extraction. After adjusting the pH of the water sample to 3 with acetic acid, 1 ng of the surrogate standard was added. In addition, a blank sample was prepared to determine the recovery rate for each solid-phase extraction batch.
The samples were analysed following the International Standard Method ISO 21675 (2019) with minor modifications and measured using liquid chromatography-tandem mass spectrometry (LC-MS/MS) (ISO 21675, 2019; Wu et al., 2021). The method used the same reagents, methods and instrument conditions as described previously (Eun et al., 2020).
Samples (40 mL) were loaded using a solid-phase extraction weak anion exchange cartridge (Oasis WAX 150 mg). This clean-up procedure and PFAS measurements were performed according to a previous study (Eun et al., 2020). Quality assurance and quality control (QA/QC) standards were based on the corresponding ISO methods (ISO 21675, 2019). To detect PFAS compounds, a Keystone Betasil C18 column (2.1 mm i.d. 50 mm length, 5 μm, 100 Å pore size, endcapped; Thermo Fisher Scientific, Waltham, MA) was used. The external calibration curves were constructed using a 2, 10, 50, 200, 1,000 and 5,000 pg/mL concentration series. The limit of quantification (LOQ) is the lowest analyte concentration, at which quantitative results can be reported with a high degree of confidence. The procedural recoveries of PFAS (n=5) were 96%–99% for PFOS, 96%–111% for PFOA, 95%–121% for 13C8-PFOS and 90%–108% for 13C8-PFOA. The quantitative analysis of PFOS and PFOA was performed using high-performance liquid chromatography (HP1100; Agilent Technologies, Santa Clara, CA) coupled with electrospray ionisation (ESI) tandem mass spectrometry (Triple QuadTM 4500 System; AB SCIEX, Framingham, MA) in the negative ESI mode.
QUALITY ASSURANCE AND QUALITY CONTROL (QA/QC)
The target PFOS and PFOA concentrations were quantified using a calibration curve of 2, 5, 10, 50, 200, 1,000 and 5,000 pg/mL analysed using LC-MS/MS. The deviation from the regression line was <20% of its theoretical value. The standard calibration curve exhibited strong linearity (correlation coefficients>0.999) and repeatability, which was confirmed before each determination. The instrumental limits of quantification (I-LOQs) were as follows: (i) the smallest concentration of the injected compound that allowed reproducible measurement of the peak areas within ±20% of the duplicate injection; (ii) the smallest standard concentration on the calibration curve accurately measured within ±20% of its theoretical value and (iii) a signal-to-noise ratio ≥10. I-LOQs for each compound were 2 pg/mL. To check the stability of the instrument, a QC standard was injected into every batch. If the concentrations of the QC standards were not measured within ±20% of their corresponding theoretical values, then a new calibration curve was prepared.
STATISTICAL ANALYSIS
Analyses were performed using SPSS (version 26, IBM, Armonk, NY), with significance set at a P-value of <0.05. The data were evaluated using the Kruskal-Wallis test as it is a non-parametric method that compares three or more independent samples. After the Kruskal-Wallis test, the Steel-Dwass and Mann-Whitney U tests were used for comparison between two groups as a multiple comparison method; Bonferroni correction was used for adjustment of significance values (Tables S7 and S8).
RESULTS AND DISCUSSION
Carbon-based materials, such as charcoal and ACs, are widely used for the reduction or removal of hazardous organic chemicals in the water. Municipal wastewater treatment and water purification plants in Japan remove a large amount of organic chemicals using GAC (Mojiri et al., 2018). In our study, we compared the efficiency of PFOS and PFOA removal from water by different PACs, including GAC. Table 1 lists the four types of PAC and GAC used in this study, which are inexpensive AC materials generally used as adsorbents that exhibit different compositions, pH and surface areas. The differences in shape, raw material, pH and surface area of the various ACs affect the adsorption and removal efficiency of PFOS and PFOA in water. Excluding PAC #4 and GAC, PAC #1, #2 and #3 removed >90% of PFOS and PFOA from water during the adsorption test time of 72 h (25°C).
Table 1 Characterisation of activated carbon adsorbents
| Material name | Type | Raw material | pH | Surface area (m2/g) |
|---|
| PAC #1 | Powdered activated carbon | Coconut husk | 9.6 | 1,258.88 |
| PAC #2 | Powdered activated carbon | Coal | 9.6 | 1,078.81 |
| PAC #3 | Powdered activated carbon | Wood | 10 | 950.99 |
| PAC #4 | Powdered activated carbon | Wood | 3.9 | 1,412.83 |
| GAC | Granular activated carbon | Coconut husk | 7 | 817.81 |
PAC, powdered activated carbon; GAC, granular activated carbon
Fig. 1 shows the removal efficiency test results, in which 1 and 10 mg/L of each carbon-based adsorption material was added to a water solution of 10 ng/L PFOS and PFOA. In contrast to other researchers who studied much higher concentrations of PFAS (mg/L), we studied them at the ng/L level and showed that AC can particularly be used to adsorb low concentrations of PFOS and PFOA, corresponding to their actual concentrations in environmental water in Japan.
PAC #1, #2 and #3 removed >90% of PFOS and PFOA in water within 72 h (25°C). The GAC used (1 mg/L) could not remove PFOS and could only remove 12% of PFOA, whereas 10 mg/L GAC treatment removed 41% and 34% of PFOS and PFOA, respectively. In other words, considering the low rate of PFOS and PFOA removal by conventional GAC, new adsorbents are needed for municipal wastewater treatment and water purification plants in Japan. PAC #4 has a specific surface area of 1,412.83 m2/g, the largest among powdered ACs; however, its ability to remove PFOS and PFOA in water is poor. This is probably because the pH of PAC #4 is 3.9 (strongly acidic); therefore, PFOS and PFOA in water are difficult to adsorb onto the surface and cavities of the PAC. In the process of burning raw materials in the AC manufacturing process, a strong acid is added to neutralise the pH as the commercial product becomes strongly alkaline. In the future, it is necessary to study the relationship between the pH of PAC and adsorption performance for PFOS and PFOA in water. PAC #1, #2 and #3 could purify water by adsorbing 90%–100% of PFOS and PFOA in water even with the addition of 1 mg/L. Moreover, only a marginal difference in adsorption performance was observed among coconut husk, coal and wood.
Therefore, PFOS and PFOA in water might be removed by adding 1 mg/L PAC to water instead of 10 mg/L and instead of GAC, which would be useful for clean water production in water purification plants. In particular, the removal rate for PFOS and PFOA in water was 100% and 98% when 1 mg/L of PAC #3 was added. Therefore, PAC #3 is the most suitable for removing PFOS and PFOA in water.
PAC #1, #2 and #3 can adsorb 90%–100% of PFOS and PFOA, purifying water even at 1 mg/L concentration. Indeed, Son and An (2022) reported that the effect on the adsorption of PFAS was not significant when there is higher interaction and sufficient AC surface area. Moreover, no difference in adsorption performance was observed among coconut husk, coal and wood. Therefore, PFOS and PFOA in water could be removed by adding 1 mg/L of PAC instead of 10 mg/L, which would be useful for clean water production in water purification plants.
The statistical evaluation of the difference between AC materials for PFOS and PFOA adsorption showed that PAC #1, #2 and #3 had a significant adsorption effect (P<0.05) compared with PAC #4 and GAC (Fig. 2). In particular, PAC #3 had a remarkable adsorption effect on PFOS and PFOA.
Fig. 3 shows the surface (at 1,000× magnification) of the carbon-based AC used. Considering that PAC #1, #2, #3 and #4 are fine powders, the particles are small with high adsorption capacity; however, GAC is not a powder and has a fibrous surface. Therefore, the surface of the AC affects the PFOS and PFOA adsorption, and PAC #1, #2 and #3 are the most suitable water purification materials to adsorb and remove PFOS and PFOA.
We aimed to investigate the efficiency of carbon-based materials for removing PFOS and PFOA from water using inexpensive materials. The cost of the materials tested in increasing order is PAC #4>GAC≥PAC #3>PAC #1≥PAC #2. Fortunately, PAC #1, #2 and #3 are considerably cheaper than the GAC currently used in water supply facilities for the removal of hazardous organic chemicals, thereby making them the best options for removing PFOS and PFOA in water. However, since PAC #1, #2 and #3 are fine powders resembling dust, the addition, recovery and disposal during water purification should be carefully monitored. As a countermeasure to remove used adsorbents, flocculants such as iron (III) chloride, iron (II) sulphate, iron (II) chloride and iron (III) hydroxide can be employed.
