Introduction

Water pollution is a serious threat to the health of organisms worldwide and is widely regarded as a major environmental issue1. The contamination of surface and ground water2 by perfluorinated alkyl substances has particularly emerged as an environmental crisis impacting hundreds of millions of people3,4 due to the increasing use of these compounds in the production of fluoropolymers5, stain guard products6, and fire-fighting foams7. The most common perfluorinated alkyl pollutants are perfluorooctane sulfonate (PFOS) and perfluorooctanoic acid (PFOA) (Fig. 1), which have been found in water worldwide, including the polar zones8. PFOS and PFOA can bind to proteins67. We determined the binding constants for the pegylated GC5A-12C nanoparticle for PFOS and PFOA to be (1.3 ± 0.3) × 107 M−1 and (4.8 ± 0.4) × 106 M−1, respectively (Supplementary Figs. 9–11). The binding strength of the pegylated GC5A-12C nanoparticle is comparable with that of GC5A-6C and should be suitable for absorption applications. Furthermore, we obtained hybrid nanoparticle (MNP@GC5A-12C) by encapsulating hydrophobic MNP into the hydrophobic domain of the GC5A-12C nanoparticle during preparation (Supplementary Note 5). Dynamic light scattering measurements revealed the MNP@GC5A-12C have a hydrated diameter of 213 ± 3 nm (Supplementary Fig. 12).

Fig. 7: Schematic illustration of PFOS and PFOA absorption and magnetic separation.
figure 7

Illustration of the absorption and magnetic separation procedure of PFOS and PFOA by the MNP@GC5A-12C nanoparticle.

Full size image

Al (III) phthalocyanine chloride tetrasulfonic acid (AlPcS4, Supplementary Fig. 8) was chosen as a model dye to explore the absorption ability of MNP@GC5A-12C. MNP@GC5A-12C was dispersed into the solution of AlPcS4 and then isolated with an external magnetic field for 1 h. Subsequently, the supernatant was filtered through a mixed cellulose esters film (Millipore, 0.025 μm) and collected for ultraviolet–visual (UV–vis) experiments. Negligible absorbance of AlPcS4 was observed after absorption (Fig. 8a), indicating the complete removal of AlPcS4 by MNP@GC5A-12C. As a control experiment, filtration without MNP@GC5A-12C resulted in barely any effect on the concentration of the AlPcS4 in solution (Supplementary Fig. 13).

Fig. 8: Absorption of AlPcS4, PFOS, and PFOA in water samples.
figure 8

a Absorption spectra of AlPcS4 (10 μM) without any treatment and after magnetic separation by MNP@GC5A-12C and filtration. (Inset) Photographs of MNP@GC5A-12C ([GC5A-12C] = 100 μM) (left), MNP@GC5A-12C ([GC5A-12C] = 100 μM) with AlPcS4 (10 μM) (middle), and the filtrate after magnetic separation and filtration (right). b Concentrations of PFOS and PFOA with only filtration, and with both magnetic separation and filtration are quantified by the UPLC–ESI-MS/MS system. Concentrations of PFOS and PFOA without magnetic separation are shown as a control. Error bars in (b) represent mean ± s.d. (n = 3 independent experiments).

Full size image

We further applied MNP@GC5A-12C to absorb PFOS and PFOA. The quantifications of PFOS and PFOA were performed by means of ultra-performance liquid chromatography–electrospray ionization–tandem mass spectrometry (UPLC–ESI–MS/MS). The calibration curves were set up and quantitative parameters were evaluated (Supplementary Note 6). The PFOS and PFOA absorption efficiencies of each samples were characterized at [PFOS]0 = [PFOA]0 = 1000 ng mL−1. After the removal procedure, the solutions were pre-concentrated to accurately determine PFOS and PFOA concentrations by the UPLC–ESI–MS/MS. There were only (0.43 ± 0.07)% residual PFOS and (1.53 ± 0.04)% residual PFOA, respectively (Fig. 8b). The regeneration of the present supramolecular material is considered to be feasible owing to its response to specific organic solvent68. In DMSO, GC5A-12C neither formed amphiphilic aggregates indicated by very low scattering intensity (Supplementary Fig. 14), nor complexed with PFOS revealed by no change in chemical shifts of 19F NMR spectra of PFOS (Supplementary Fig. 15). Therefore, we envisage that regeneration of the supramolecular material could be achieved by using routine purification methods in organic synthesis such as column chromatography.

Discussion

In conclusion, our artificial GC5A-6C receptor successfully encapsulated PFOA and PFOS with nanomolar affinity in aqueous media. Taking advantage of the strong recognition and supramolecular assembly, we achieved not only sensitive and quantitative detection of PFOA and PFOS in tap and lake water through the fluorescent IDA strategy, but also the efficient removal of them by the hybrid MNP@GC5A-12C nanoparticle via a simple magnetic absorption and filtration procedure. These results will facilitate the development of detection and absorption methods for PFOA and PFOS. Although the present LOD and removal efficiency cannot reach the health advisory level in drinking water, the proposed supramolecular approach can be practically operated in heavily polluted areas, such as industrial regions, airports, and military facilities. For daily drinking water detection, we can use the solid-phase extraction for sample preconcentration. Higher removal efficiency may be achieved by applying the GC5A-12C nanoparticle as solid-phase extraction absorbents. This work made full use of the molecular recognition and self-assembly of artificial receptors, offering a promising strategy for the detection and remediation of water pollution.

Methods

Chemicals

All the reagents and solvents were commercially available and used as received unless otherwise specified purification. Ammonium acetate and 2,2,2-trifluoroethanol were purchased from Sigma-Aldrich. Fl was purchased from Tokyo Chemical Industry. Al (III) phthalocyanine chloride tetrasulfonic acid (AlPcS4) was purchased from Frontier Scientific. PFOA, PFOS, hexadecyltrimethylammonium bromide, octanoic acid, octanesulfonic acid, and perfluorohexane were purchased from Energy Chemical. Iron oxide nanoparticle stabilized by oleic acid (MNP) was purchased from Ji Cang Nano Company. The waste water samples were provided by the manufacturing facility located in Cangzhou, Hebei province, China. 5,11,17,23,29-Pentaguanidinium-31,32,33,34,35-penta(4-methylpentloxy)calix[5]arene (GC5A-6C), 5,11,17,23,29-pentaguanidinium-31,32,33,34,35-penta dodecyloxy-calix[5]arene (GC5A-12C) and 4-(dodecyloxy)benzamido-terminated methoxy poly(ethylene glycol) (PEG-12C) were synthesized according to the previous literature53,66.

Samples

The 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) buffer solution of pH 7.4 was prepared by dissolving 2.38 g of HEPES in approximate 900 mL double-distilled water. Titrate to pH 7.4 at the lab temperature of 25 °C with NaOH and make up volume to 1000 mL with double-distilled water. The pH value of the buffer solution was then verified on a pH-meter calibrated with three standard buffer solutions.

Apparatus

19F NMR data were recorded on a Bruker AV400 spectrometer. Steady-state fluorescence spectra were recorded in a conventional quartz cell (light path 10 mm) on a Cary Eclipse equipped with a Cary single-cell peltier accessory. UV–vis spectra were recorded in a quartz cuvette (light path 10 mm) on a Cary 100 UV–vis spectrophotometer equipped with a Cary dual cuvette peltier accessory. The sample solutions for dynamic light scattering measurements were examined on a laser light scattering spectrometer (NanoBrook 173plus) equipped with a digital correlator at 659 nm at a scattering angle of 90°. Quantification of PFOS and PFOA from the absorption studies were performed by means of UPLC–ESI–MS/MS (Waters, Milford, MA, USA).