Next Article in Journal
Life Cycle Assessment of Closed-Loop Hydrometallurgical Recovery of Platinum Group Metals from PEM Fuel Cells and Electrolyzers
Previous Article in Journal
From Observability to Intervention in Purification: A Critical Framework for Digital Twins, Chemometrics, and Control Across Membranes, Adsorption, and Chromatography
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Purification
Volume 2
Issue 3
10.3390/purification2030011
purification-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Perfluoroalkyl Substance Adsorption Using Activated Carbon Modified Sequentially with Polyethyleneimine and Poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel-F)

by
Omobolaji Ayeseni
1,
Catherine B. Almquist
2,
Jason A. Berberich
2 and
Neil D. Danielson
1,*
1
Department of Chemistry and Biochemistry, Miami University, Oxford, OH 45056, USA
2
Department of Chemical Engineering, Miami University, Oxford, OH 45056, USA
*
Author to whom correspondence should be addressed.
Purification 2026, 2(3), 11; https://doi.org/10.3390/purification2030011
Submission received: 30 December 2025 / Revised: 3 May 2026 / Accepted: 28 May 2026 / Published: 14 July 2026
Browse Figures
Versions Notes

Abstract

Per- and polyfluoroalkyl substances (PFAS) such as perfluorooctanoic acid (PFOA) persist in aquatic environments due to their extremely strong C–F bonds and high environmental stability/poor biodegradability, creating an urgent demand for high-performance and scalable adsorption treatment technologies. This study reports the synthesis, characterization, and fixed-bed column adsorption performance of a dual-polymer-modified activated carbon adsorbent engineered for enhanced PFAS capture. Activated carbon (AC) was sequentially functionalized with polyethyleneimine (PEI) and poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel–F) to form the fluorine-rich, amine-grafted composites AC–PEI–KelF and AC–PEI–KelF–PEI. Surface-area studies of the modified AC adsorbents showed a reduction in surface area of about a factor of two, possibly due to pore blockage by the polymers. Continuous-flow column breakthrough studies demonstrated dramatic improvements in perfluorooctanoic acid (PFOA) removal efficiency, with 50% maximum breakthrough time increasing from 80 min for AC to 540 min for AC–PEI–KelF and 1500 min for the AC–PEI–KelF–PEI formulation. Thomas model adsorption capacities for AC, AC–PEI–KelF and AC–PEI–KelF–PEI were, respectively, 65, 490, and 1130 mg/g. The adsorption mechanism, using selective mobile phases, was shown to be due to a combination of electrostatic and hydrophobic/fluorophilic interactions. Kinetic analysis showed that adsorption exhibited pseudo-second-order behavior, with multi-stage intraparticle diffusion. The optimized composite also exhibited strong regeneration stability, retaining 79% performance after six adsorption–desorption cycles, and displayed high selectivity for PFOA, even in the presence of structurally related competitors such as octanoic acid. The adsorption of perfluoropentanoic acid and undecafluoro-2-methyl-3-oxahexanoic acid (GenX) was also significant, and some interaction with trifluoroacetic acid was noted.

Graphical Abstract

1. Introduction

Per- and polyfluoroalkyl substances (PFAS) comprise a large group of synthetic organofluorine surfactants widely used in non-stick coatings, textiles, lubricants, and firefighting foams [1]. However, their exceptional thermal and chemical stability, attributed to the strong carbon–fluorine bond, has led to environmental persistence [2], bioaccumulation [3], and adverse health effects [4], including immunotoxicity and carcinogenicity [5]. Of regulatory concern are long-chain PFSAs (CnF2n₊1SO3H, n ≥ 6) and PFCAs (CnF2n₊1COOH, n ≥ 7) [6]. Commercial production were initiated in the 1950s with compounds similar to perfluorooctanoic acid (PFOA) [7], which is currently receiving significant attention due to its widespread occurrence in water systems and detection in human serum at trace levels. This has prompted the U.S. Environmental Protection Agency (EPA) to establish a stringent health advisory limit of 0.004 ng/L for PFOA in drinking water [8], although this could be under review for change.
Adsorption is a cornerstone separation technique in environmental remediation [9], prized for its cost-effectiveness [10], operational simplicity, high removal performance [11], and potential for regeneration [12,13]. Adsorbent materials used for PFAS removal include powdered activated carbon (AC) [14], ion exchange resins, clay, and metal organic frameworks [15,16]. Activated carbon and ion exchange resins are the most utilized adsorbents in practice, with both providing strong adsorption capacity, and AC in particular offering economic viability [17,18]. Activated carbon is a highly porous material [19] characterized by its large surface area and tunable pore architecture with diverse oxygenated surface functional groups. Among these properties, the pore structure plays a critical role in determining the adsorption behavior of organic molecules onto AC [20]. AC can exhibit relatively slow adsorption kinetics for PFOA [21,22]. In contrast, anion exchange processes demonstrate higher adsorption capacities [23], faster kinetics, and effective regeneration [24]; however, their large-scale application is limited by higher operational costs and industrial feasibility challenges [25].
Adsorbents combining the attributes of AC and ion exchange using adsorbed polyethyleneimine (PEI) have increased in research interest [26,27]. Activated carbon with grafted PEI was shown to increase the removal of copper ions by about 2.5 times [28]. Simple impregnation of PEI on AC was effective for the removal of nickel (II) or cadmium (II) but less so for lead [29]. On-column adsorption, as well as batch studies, were shown effective for the adsorption of cadmium (II) using PEI-impregnated AC [30]. The adsorption of Reactive Yellow 2 dye was 7.4 times greater on AC–PEI as compared to that on plain AC [31]. Ion-pairing of PFAS to cetyltrimethylammonium-modified AC was effective in both the batch and on-column experiments [32]. Recently, perfluorocarboxylic acids were effectively pre-concentrated on PEI–glutaraldehyde-modified biochar [33].
Our research program has been directed to the synthesis and characterization of PFAS adsorbents that provide both ion exchange and fluorophilic retention mechanisms. The reaction of poly (chlorotrifluoroethylene–co-vinylidene fluoride) (PCTFE–VF) co-polymers such as Kel F-800 with amines was reported decades ago [34]. Dehydrohalogenation due to elimination of HF to form double bonds across the polymer occurs before the addition of the amine. Grafting of amines to vinylidene fluoride-type co-polymers is due to the same reaction mechanism [35]. We have synthesized weak anion exchange liquid chromatography packings by modifying aminopropyl silica with Kel F-800 for the separation of surfactants with indirect detection [36,37] and aromatic acids [38]. Besides having the reactive vinylidene fluoride moiety, Kel F-800 exhibits good solubility in organic solvents such as tetrahydrofuran at modest temperatures, allowing for a more homogeneous reaction.
Fluorinated ligands attached to various adsorbents for pre-concentration of PFAS, such ion exchange resins modified with difluroethanol or pentafluorostyrene [39] is an emphasized approach. Synthesis of fluorinated polymers for application to PFAS adsorption is well known. Porous aromatic frameworks containing fluoroalkyl groups have been shown to reduce PFOA concentrations to ppt levels [40]. A linear triblock co-polymer with the fluorinated moiety in the polymer backbone, not as a side chain, was more effective in generating fluorophilic PFOA interactions [41]. A perfluoropolyether co-polymer was effective in removing PFOA, even in the presence of biomolecules such as proteins [42].
The use of grafted fluoropolymers, particular commercial ones, for PFAS adsorption is less common. A commercial solid phase extraction weak anion exchange resin with secondary and primary amine groups was modified with Kel F-800 and characterized for on-column and batch adsorption of PFOA [43]. This same weak anion exchange fluoropolymer-grafted material has shown excellent promise for solid phase micro extraction of both short- and long-chain PFAS prior to separation by LC–MS [44].
AC particles have been coated with a hot Halar resin (co-polymer of chlorotrifluoroethylene and ethylene) particle suspension and with polyethyleneimine (PEI) [45]. Studies using 2 cm packed columns have their merits because of their ease of operation, faster adsorption screening, and initiation of the scale-up process [46]. They can also provide dynamic transport behavior and adsorption mechanisms via the adsorbate [46]. Such columns with these adsorbents showed a 50% breakthrough time for PFOA of about 350 min compared to 110 min for uncoated AC. Both hydrophobic and electrostatic retention mechanism were important. However, the white Halar polymer, upon encountering more extreme reaction conditions with PEI and potassium t-butoxide for 5 h in toluene at 80 °C, showed no darkening, indicating that the dehydrohalogenation reaction was apparently not effective, with likely minimal covalent attachment of PEI. Stability of the packed AC–Halar–PEI columns was shown, even though both PEI and the Halar resin were simply coated on AC.
In this study, our objective was to synthesize and evaluate a multilayer polyamine-functionalized fluoropolymer-coated activated carbon (AC–PEI–KelF–PEI) for the enhanced adsorption of PFOA from aqueous solutions. This approach would not be viable using the Halar resin. The resulting sorbent was characterized for surface area and morphology, adsorption performance, and regeneration capability. Breakthrough column adsorption studies were conducted using online UV detection of PFOA. The Yoon–Nelson 50% maximum breakthrough time increased from 80 min for AC to 1500 min for the AC–PEI–KelF–PEI packed column, and the Thomas model adsorption capacities for AC and AC–PEI–KelF–PEI were, respectively, about 65 and 1130 mg/g. Other PFAS compounds such as perfluoropentanoic acid and undecafluoro-2-methyl-3-oxahexanoic acid (GenX) were also retained on the AC–PEI–KelF–PEI column. This is a proof-of-concept study to show the viability of this type of modified AC for small drinking water applications.

2. Materials

2.1. Chemicals

All materials were used as received from their suppliers. Chemically activated charcoal (AC, 100–400 mesh size, 38–149 microns) was sourced from Sigma-Aldrich (St. Louis, MO, USA). The 78% polychlorotrifluoroethylene–22% vinylidene co-polymer (Kel F-800), with a weight average molecular weight (Mw) of 76,000 and a number average molecular weight (Mn) of 29,000 [47], is a particulate solid and was obtained from 3M (Minneapolis, MN, USA). As of December 2025, Kel F-800 (aka FK-800) and other fluoropolymers are no longer available from 3M, but other companies such as Poly K (piezopvdf.com) offer similar PCTFE–VF polymers as a limited-stock item. PCTFE–VF is available from AM Block plastics as a viscous oil but not as a resin. The poly (vinylidene fluoride) (PVF) was the Knar HSV900 type, still available from Arkema (Calvert City, KY, USA) and many other companies for Li battery research. PFOA, branched polyethyleneimine (PEI; Mₙ ~10,000, Mw ~25,000), and potassium tert-butoxide were procured from Sigma-Aldrich. Methanesulfonic acid (MSA) was supplied by TCI America (Montgomeryville, PA, USA). Environmental grade acetonitrile was supplied by Alfa Aesar (Ward Hill, MA, USA). Other solvents such as tetrahydrofuran (THF) and n-butanol were obtained from Fisher Chemicals (Pittsburgh, PA, USA). A MilliQ water purification system provided the distilled, deionized water (18.2 MΩ·cm resistivity).

2.2. Synthesis

Reactions of Kel F-800 or polyvinylidene fluoride (PVF) with PEI were studied in the presence of potassium t-butoxide, to eliminate HF, forming double bonds, the amine reaction sites. A total of 5 g Kel F-800, 0.5 g PEI (dissolved in 20 mL n-butanol), and 0.25 g potassium t-butoxide was heated at 50 °C in a 150 mL tetrahydrofuran solution using a rotary evaporator for 4 h. An orange-brown solution formed immediately. The product was washed with butanol and water. The resulting air-dried product was orange-brown in color (Figure S1), similar to the results in our previous work involving the reaction of the weak anion exchange resin SepraWAX with Kel F-800 [43]. Confirmation that the vinyl fluoride part of the Kel-F polymer was reacting with potassium t-butoxide and PEI was made using PVF as a test polymer. Reaction of PVF with potassium t-butoxide was immediately evidenced by the colloidal solution turning black. Upon filtering and washing, the resulting polymer was chocolate brown. A mixture of 1.25 g PVF, 0.30 g PEI, and 0.15 g potassium t-butoxide was sonicated in 90 mL THF-20 mL butanol and then allowed to react in a rotary evaporator for 4 h at 60 °C. After filtering, washing with methanol, and air-drying the resulting polymer product was also chocolate brown. FTIR spectra of this PVF–PEI product are shown in Figure S2. The presence of N–H peaks at 3500 and 1650 cm−1 is evidence of amine functionalization. The reaction of PEI with PVF, shown in Figure 1, is analogous to that revealed in a study describing the modification of PVF with ethylenediamine [48]. Because of the multiple amine groups in PEI and the multiple vinylidene fluoride linkages in Kel F-800, crosslinking is likely favored. However, chemical proof of actual crosslinking would be difficult to ascertain.
The initial wet-chemistry synthesis of the proposed selective adsorbent, AC-2 (KelF–PEI), for PFOA preconcentration was designed as a sequential, two-step process. In the first step, Kel-F was deposited onto activated carbon (AC) to form AC–KelF, which was subsequently reacted with polyethyleneimine (PEI) in the second step to generate AC–KelF–PEI. This two-step coating sequence was then repeated to produce the layered composite AC-2 (KelF–PEI). However, post-synthesis characterization using energy-dispersive X-ray spectroscopy (EDS) revealed minimal fluorine and chlorine for AC–KelF, indicating possibly uneven coating. Adsorbents for comparison were synthesized according to the pathways described in Section 2.2.1, Section 2.2.2 and Section 2.2.3.

2.2.1. AC–PEI

In the initial synthesis step, activated carbon (AC) was reacted with PEI (AC:PEI = 1:1, 5:1, 10:1, or 1:2 w/w) in the presence of potassium tert-butoxide (0.6 g) in 100 mL of n-butanol. A strong base has been shown to be effective as a surface treatment for AC [49,50]. The reaction was carried out at 50 °C under continuous rotation at 255 rpm for 4 h, yielding the AC–PEI composite. Upon completion, the hot reaction mixture was filtered, and the resulting solid was extensively washed with THF, followed by distilled water, to remove residual salts and unreacted species. Although amide bond formation could occur between AC and PEI, it is expected that the majority of PEI is trapped on the pore surface of the AC and held through van der Waals forces such as hydrogen bonding. The dried material was subsequently packed into fixed-bed columns (2 cm bed height) and evaluated for PFOA breakthrough performance. To assess reproducibility, the synthesis protocol was repeated independently in duplicate, and the corresponding adsorption experiments were repeated under identical conditions.

2.2.2. AC–PEI–KelF

A predetermined mass of Kel F-800 polymer, corresponding to (AC–PEI):polymer mass ratios of 20:1, 10:1, or 5:1, was dissolved in 100 mL of THF in the presence of potassium tert-butoxide (0.6 g) in a 500 mL round-bottom flask using sonication to ensure complete dissolution. Activated carbon modified with polyethyleneimine (AC–PEI, 5.0 g) was then added to the polymer solution. The flask was connected to a rotary evaporator equipped with a water condenser, and the suspension was maintained at 50 °C with a rotation speed of 255 rpm for 1 h. Following synthesis, the resulting AC–PEI–KelF composite was recovered by vacuum filtration and sequentially washed with THF and deionized water to remove excess polymer and residual solvent. The purified material was dried overnight at ambient temperature under a fume hood. The dried composite was subsequently packed into fixed-bed columns (2 cm bed height) and evaluated for PFOA breakthrough performance. To confirm reproducibility, the synthesis procedure was independently repeated twice, and the corresponding adsorption experiments were conducted under identical conditions.

2.2.3. AC–PEI–KelF–PEI

The Kel-F and PEI reactions were subsequently repeated based on the optimal formulation identified from the breakthrough curve analysis of AC–PEI and AC–PEI–KelF to synthesize the final product, AC–PEI–KelF–PEI. The Kel-F polymer has a significantly higher weight average molecular weight (76,000) as compared to that for PEI (25,000). We considered that the Kel-F may be sterically hindering the ion exchange retention mechanism, so testing AC with a PEI–KelF–PEI sequence was deemed important.

3. Adsorbent Characterization Methods

3.1. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) was conducted using a TA Instruments Q500 thermogravimetric analyzer (version 20.13, build 39, (New Castle, DE, USA)) to evaluate the thermal stability and decomposition behavior of the samples. Approximately 17.5 mg of each sample was placed in a platinum pan, and the measurement was performed under a nitrogen atmosphere with a balance purge flow rate of 60 mL min−1 and a sample purge flow rate of 40 mL min−1. The temperature was increased from room temperature to 900 °C at a heating rate of 10 °C min−1. After the nitrogen ramp, the gas was switched to air to assess oxidative stability and residue composition. The instrument was calibrated for both temperature and weight prior to analysis, and data were collected as a function of temperature to obtain weight loss.

3.2. Surface Morphology and Elemental Composition

Surface morphological features and elemental composition were characterized using a Zeiss Supra 35 VP FEG scanning electron microscope (SEM) equipped with energy-dispersive X-ray spectroscopy (EDS) (Carl Zeiss AG, Carl-Zeiss-Straße 22, Oberkochen, Germany). EDS spectra were acquired at 30 keV with a 60 µm aperture, a 10 mm working distance, and a chamber pressure of 10 Pa. Imaging was performed on all samples at an accelerating voltage of 20 keV using a 120 µm aperture, a working distance of 2.4 mm, and a chamber pressure of 30 Pa. The high FTIR spectroscopic background for this AC made specific peaks due to surface modification difficult to discern. Elemental analysis for AC and the AC products was done by Galbraith Laboratories (Knoxville, TN, USA).

3.3. Brunauer–Emmett–Teller (BET) Surface Area Analysis

Surface area characterization of the synthesized sorbents was performed using nitrogen (N2) adsorption–desorption isotherms at −196 °C on a Micromeritics TriStar II 3020 BET analyzer (Norcross, GA, USA). Prior to measurement, each sample was weighed, degassed under nitrogen at 105 °C for 30 h to remove residual moisture, and reweighed. Surface areas were calculated using the Brunauer–Emmett–Teller (BET) model applied to the adsorption data, while pore size distribution was obtained through Barrett–Joyner–Halenda (BJH) analysis, and micropore volumes were estimated using the t-plot method. All data processing was conducted using the Micromeritics TriStar software package V3.02.

3.4. Zeta-Potential Analysis

The electrokinetic properties of the sorbent particles were assessed through zeta potential measurements using a Zeta-Meter 4.0 analyzer (Zeta-Meter Inc., Staunton, VA, USA). Particle mobility and direction under an applied electric field were monitored to determine surface charge behavior. All analyses were performed at 20 °C in a 5% isopropanol medium without added background electrolyte, with suspension pH values ranging from 5.6 to 6.0. Samples were prepared as 100 mg/L colloidal dispersions in 5% isopropanol and evaluated within 30 h of preparation. For each sample, the velocities of ten individual particles were recorded at an applied potential of 100 V, and the instrument software automatically computed the mean zeta potential and corresponding standard deviation.

3.5. Adsorption Breakthrough Analysis

Column adsorption experiments were performed using a Beckman Model 100A HPLC pump (maximum pressure 5000 psi, dual-piston configuration) capable of being operated at flow rates ranging from 0.03 to 9.99 mL/min. The pump was coupled to a Bodman (Applied Biosystems, Beckman Instruments, Inc., Scientific Instruments Division, Irvine, CA, USA) Spectroflow 757 UV–Vis absorbance detector set to 210 nm, the characteristic wavelength for PFOA detection [43]. Stainless steel columns (0.46 cm ID × 2 cm; IDEX Health & Science, Middleboro, MA, USA) were dry-packed with pre-weighed sorbent materials (0.0999–0.1450 g). Before sample introduction, each packed column was conditioned by sequential flushing with 90% acetonitrile (MeCN) containing 10% 0.1 M methanesulfonic acid (MSA), followed by 100% MeCN, and finally Milli-Q water, each at a flow rate of 1 mL/min for 20–30 min. After equilibration, a 100 ppm PFOA solution (pH 3.5) was supplied to the mobile-phase reservoir, and adsorption measurements were conducted at a constant flow rate of 1 mL/min. Chromatographic output was collected using the PeakSimple Chromatography Data System (Version 4.53, SRI Instruments) and exported to Microsoft Excel for quantitative analysis. The representative breakthrough curves were usually the result of the first PFOA cycle through the washed and conditioned packed column.

4. Results and Discussion

4.1. Thermal Stability Analysis

Thermogravimetric analysis (TGA) was employed to assess the thermal stability of the synthesized sorbents and to evaluate the extent of polymer incorporation. The relative mass vs. temperature profiles are compared in Figure 2. The thermogravimetric profiles indicate that the mass losses from the adsorbents, under a nitrogen atmosphere, occurs in four distinct temperature regions: below 120 °C, between 120 °C and 400 °C, between 400 °C and 700 °C, and above 700 °C. This is more clearly seen in the differential thermal gravimetry (DTG) plots generated by taking the first derivative of mass vs. temperature, as shown in Figure S3. Each peak in the DTG curve shows the change in mass related to change in temperature at a specific thermal decomposition event, permitting overlapping events to be discerned.
The mass losses are summarized in Table S1. The initial mass losses (<120 °C) are attributed to the desorption of physically adsorbed moisture from the sample surface. Mass losses between (120–400 °C) are attributed to the loss and decomposition of volatile organic compounds from AC and AC–PEI. It is notable that the adsorbents containing KelF have relatively low losses of mass in this temperature range. This suggests that Kel-F provides some thermal stability for the adsorbents at temperatures up to 400 °C. Mass losses between 400 °C and 700 °C correspond to the thermal decomposition of PEI and KelF on AC, while the final stage (>700 °C) is associated with the thermal degradation of the material carbon backbone, primarily arising from organic components [43].
TGA was also conducted in air as opposed to nitrogen. These results are shown in Figure S4. Pristine AC exhibited the highest thermal stability, with a 50% mass loss at 525 °C, consistent with its largely inert carbonaceous matrix [51,52]. PEI showed the lowest thermal stability, with a steep 50% mass loss at 320 °C, while Kel-F also showed a steep mass loss, with 50% mass loss at 450 °C, similar to what has been reported previously [53]. Thermal degradation of the AC–PEI and AC–PEI–KelF–PEI adsorbents was more gradual with a lower onset temperature for degradation (225 °C), but degradation occurred over a wider temperature range for both samples, with 50% mass loss for AC–PEI–KelF–PEI at 430 °C and 50% mass loss for AC–PEI at 495 °C.

4.2. SEM Surface Morphology and Elemental Composition

To improve the selectivity and adsorption efficiency of activated carbon, considerable research has focused on designing modified sorbents with tailored pore structures, surface chemistries, and morphologies [54]. Since the morphological features of a sorbent play a crucial role in determining its adsorption performance, scanning electron microscopy (SEM) was employed to examine the morphological evolution of the sorbents throughout the synthesis process.
The scanning electron micrographs (Figure 3) show the morphological evolution of the AC surface following sequential surface modifications with Kel-F and PEI. The pristine activated carbon (Figure 3A) exhibited a relatively smooth surface, with well-developed pores and layered structures characteristic of microporous carbon materials. Upon coating with PEI (Figure 3B), the surface became rougher and displayed partially blocked pores due to the deposition of the polymer film on the carbon framework. In the AC–PEI–KelF composite (Figure 3C), additional surface roughness and granular aggregates were observed, confirming successful deposition of PEI on the KelF-coated surface. The increased irregularity and heterogeneous texture indicate strong interfacial interactions between the polymer layers. For the AC–PEI–KelF–PEI sample (Figure 3D), a denser and more compact morphology was obtained, suggesting higher PEI loading, which likely contributed to enhanced surface functionality and improved adsorption capacity. These progressive morphological transformations validate the successful surface modification of activated carbon with both Kel-F and PEI layers, consistent with the proposed hybrid coating mechanism.
Elemental analysis was done two different ways. The EDS elemental analysis confirms the progressive surface functionalization of activated carbon, evidenced by increased fluorine, chlorine, and nitrogen contents in AC–PEI–KelF and AC–PEI–KelF–PEI, verifying successful incorporation of both the Kel-F and PEI layers. The EDS elemental analysis confirms the progressive surface functionalization of activated carbon, evidenced by increased fluorine (2.4%), chlorine (1.02%), and nitrogen (factor 1.3 times higher) for AC-2 (KelF–PEI), verifying successful incorporation of both the Kel-F and PEI layers. Chemical elemental analysis showed that the % nitrogen increased more than ten times for AC–PEI as compared to that for the starting material AC (Table S2). The AC–PEI–KelF–PEI adsorbent, found to be the optimum for PFOA absorption, showed the presence of fluorine.

4.3. Surface Analysis

Accurate evaluation of the porosity and surface characteristics is essential for the effective synthesis and practical utilization of porous carbon materials. Among the available techniques, gas adsorption is widely regarded as the most appropriate method for such characterization [55,56]. In this study, nitrogen adsorption was used to evaluate the effects of polymer loading on the surface characteristics of AC. The adsorption isotherms for the adsorbents in this study are provided in Figure S5. The nitrogen adsorption–desorption isotherms revealed a progressive decline in adsorption capacity corresponding to the reduction in surface area. The shape of the adsorption isotherm and the presence of a hysteresis are indicative of mesoporous adsorbents that have a wide range of narrow slit-like pores [57].
The Brunauer–Emmett–Teller (BET) equation [58] that is used to calculate the surface area of the adsorbents is provided in Equation (S1). The validity of the BET surface area analyses is based upon two criteria, called the Rouquerol criteria, as explained in the Figure S6 legend. The data that meet the increasing linear range as defined by the second Rouquerol criterion were used to calculate the surface areas of the adsorbents using the BET equation (Equation (S1)). An example of the linearized model using nitrogen adsorption isotherm data is shown in Figure S7. The slope and intercept of this plot can be used to determine the values of V m and C in Equations (S2) and (S3), respectively, which in turn can be used to calculate the surface area using Equation (S4) [58].
The micropore volume was estimated using the t-plot method (Equation (S5)). The thickness of the monolayer, t, was calculated using the Harkins and Jura method. The intercept and slope of the t-plot in the range of 3.5 Å < t < 5 Å were used to determine the micropore volume (Equations (S6) and (S7)) and the external surface area (Equation (S8)), respectively. An example of a t-plot is shown in Figure S8. The micropore surface area was approximated as the difference between the BET surface area and external surface area, as depicted in Equation (S9).
The surface areas, pore volumes, and pore widths for the adsorbents are summarized in Table 1. The high surface area of 1174 m2/g for AC is characteristic of porous activated carbons and often reflects the presence of micropores. Following PEI coating, the surface area decreased to 649 m2/g, consistent with partial pore occlusion by the polymer. Smaller decreases in surface area were noted for AC–PEI–KelF and AC–PEI–KelF–PEI. These observations confirm that increasing polymer loading reduces accessible porosity. The ratio of micropore volume to total pore volume decreased slightly with the addition of KelF, from a fraction of 0.24 for AC and AC–PEI to 0.20 for AC–PEI–KelF and AC–PEI–KelF–PEI. In addition, the average pore width decreased slightly following PEI loading, but it was relatively unchanged with subsequent polymer loadings. Pore size distributions [59] of the adsorbents are provided in Figure S9, which supports the slight shift toward lower average pore widths with polymer addition compared to those for AC. This decrease in surface area of PEI-impregnated AC has been reported previously [31]. Apparently, high molecular weight PEI can block the pores, reducing surface area. A future study checking the ion exchange adsorption capacity of AC–PEI as a function of PEI molar mass might be appropriate. However, in our work, ion exchange accessibility of PEI was regained by performing a multilayer PEI–KelF–PEI synthesis, somewhat offsetting the loss in surface area.

4.4. Zeta-Potential Data

Zeta-potential measurements (Figure S10) show a change in surface charge from the unmodified AC (Adsorbent 1) to the fully modified sorbents AC–PEI, AC–PEI–KelF, and AC–PEI–KelF–PEI (Adsorbent 2–4). The AC starting material exhibited a negative zeta potential (−47 mV), indicating a strongly anionic surface dominated by inherent functional groups. Following sequential surface modification steps, the zeta potential becomes positive +54 mV (Adsorbent 2), +64 mV (Adsorbent 3), and +66 mV for the final composite (Adsorbent 4). This systematic change and increase reflect increasing surface coverage by neutral or partially protonated polymer layers, which dilute or shield the surface-exposed anionic sites. This is advantageous for PFAS adsorption because anionic PFAS species (–COO−, –SO3−) are electrostatically attracted to positively charged surfaces. Overall, the zeta-potential trend confirms successful surface modification.

4.5. Adsorption Study

The response of the flow analysis instrument for PFOA was established as shown by the absorbance vs. concentration data over the 10–130 mg/L range in Table S3, indicating good linearity in Figure S11. To ensure reasonable breakthrough times using the packed 2 cm columns, a 100 mg/L PFOA solution was used. The critical micelle concentration for PFOA is cited to be 0.031 M or 12700 ppm. Therefore, we expect that the column adsorption of PFOA with the adsorbents at lower PFOA concentrations will still be analogous in mechanism.

4.5.1. Adsorption Fixed-Bed Flow Analysis

The concentration ratio (C0/Ct) reflects the rate and extent of adsorption or reaction over time. As observed, all samples exhibit a sigmoidal concentration–time profile characterized by an initial lag phase, followed by a rapid increase, and finally reaching a steady-state plateau. To ensure proper experimental control, AC was subjected to identical reaction conditions without interaction with either Kel-F or PEI. The blank AC samples, treated by heating in THF or heating with THF containing potassium t-butoxide, displayed nearly identical breakthrough behavior at about 70 min to that of the pristine material (Figure S12), confirming that the observed changes should arise specifically from polymer deposition rather than from solvent effects. The short empty-bed contact time (EBCT), calculated as the empty column volume (divided by the 1 mL/min flow rate) of 0.33 min (20 s), is typical of those found for household point-of-use technologies, ranging from 10–60 s [60]. The predicted bed volume, calculated as the empty column volume (0.33 cm3) times the void fraction for average 100 micron particles formed as a tapped bed in the column, (0.38) is 0.125.
Functionalization of AC with PEI at different polymer loading ratios (AC:PEI) resulted in a marked enhancement of breakthrough performance, with breakthrough times increasing to approximately 503 min (1:1), 536 min (5:1), and 573 min (10:1). However, with excess PEI compared to AC (1:2 AC:PEI), the breakthrough time decreased to 277 min (Figure S13). These results suggest that the incorporation of amine-rich PEI enhances PFOA adsorption, likely due to strengthened electrostatic interactions between the protonated amine groups and the anionic PFOA molecules. However, the observed variation in breakthrough times across loading ratios indicates that an optimal polymer-to-carbon ratio is required to balance effective surface functionalization with preservation of the modified carbon porosity. Excessive polymer loading appears to compromise pore accessibility, leading to diminished adsorption efficiency despite increased surface functionality.
Figure S14 indicates the reproducibility of breakthrough 50% column saturation times for AC–PEI–KelF synthesized at different (AC–PEI):Kel-F mass ratios. The superior performance relative to AC–KelF confirms that PEI provides additional high-affinity binding domains that are not present on native AC or on Kel-F-coated surfaces alone. The trend in 50% column saturation times (Figure S14) further underscores the role of polymer composition in determining adsorption efficiency. Increasing Kel-F content on AC–PEI sharply improved saturation time, with an optimal (AC–PEI):Kel-F ratio of 1:1 achieving the longest saturation time (673 min). At higher ratios, performance declined, likely due to excessive polymer loading that partially obstructs pores and slows intraparticle diffusion.
Figure 4 presents the breakthrough curves for the various adsorbents, showing that PEI enhances the 50% breakthrough point to about 500 min as compared to that for AC, but AC–PEI–KelF and AC–PEI have similar breakthrough times. This lack of PFOA-enhanced adsorption for fluoropolymer-coated AC has been observed previously for AC coated with an ECTFE–ethylene co-polymer [45]. However, the AC–PEI–KelF–PEI adsorbent exhibited a significantly prolonged 50% breakthrough time of approximately 1700 min, suggesting that this formulation provides an optimal PEI–KelF loading and effective covalent attachment to the activated carbon surface. When comparing the initial upticks in Ct/Co (Figure 4), the AC–PEI–KelF–PEI product showed an initial breakthrough at 1050 min, about a thirteen order-of-magnitude improvement in the initial breakthrough uptick relative to that of pristine AC (80 min) under the same column conditions. The treated bed volume can be calculated as 1050 mL divided 0.125, which is 8400. In contrast, the AC-2 (KelF–PEI) sample displayed a reduced uptick time of approximately 700 min, indicating that additional polymer loading may adversely affect pore accessibility and mass transport, thereby limiting adsorption performance.

4.5.2. Adsorption Isotherms

The fixed-bed adsorption behavior was determined using the Thomas and Yoon–Nelson models, with their corresponding mathematical expressions provided as Equations (S10) and (S11) of the Supplementary Information.
Thomas Model
The Thomas adsorption model was applied to quantify the dynamic adsorption performance of the synthesized sorbents and was found to accurately describe the experimental breakthrough behavior. Representative data comprising 20 points near the Ct/C0 ≈ 0.5 breakthrough region for batch 1 of AC–PEI–KelF–PEI are summarized in Table S4, with the corresponding Thomas model fit shown in Figure S15. The strong agreement between the experimental and modeled data, reflected by a high correlation coefficient (R2 = 0.9617), confirms the suitability of the model for this system. Using the fitted parameters from Equation (S10) in the Supplementary Information, the maximum adsorption capacity (q0) of AC–PEI–KelF–PEI was calculated to be 1126 mg/g at an influent PFOA concentration of 100 mg/L. The corresponding Thomas rate constant was determined to be 0.084 [mL/(min × mg)], based on measurements from three independently packed columns (n = 3) (Table 2).
For comparison, column experiments conducted with pristine activated carbon (AC), as well as AC pretreated with butanol and tetrahydrofuran (THF) under identical operating conditions (constant flow rate, column geometry, and an influent PFOA concentration of 100 mg/L), yielded substantially lower adsorption capacities. Thomas model analysis estimated q0 absorption capacities of about 65 mg/g, with corresponding rate constants of approximately 2.3 [mL/(min × mg)] (Table 2). This adsorption capacity is very similar to the Langmuir determined value of 63 mg/g, reported previously for PFOA absorbed to plain AC [14]. Under the same experimental conditions, the adsorption capacity of AC–PEI–KelF–PEI was therefore approximately seventeen-fold greater than that of the unmodified activated carbons. The high reproducibility observed across multiple column packings (n = 3) and repeated adsorption cycles (n = 3) further underscores the robustness and consistency of the experimental results (Table S5). In contrast, the adsorption of Cd2+ only increased about 4 times, from 11 mg/g for plain AC to 45 mg/g with AC–PEI [29]. A Langmuir qmax value of Reactive Yellow dye #2 adsorbed to AC–PEI was reported to be 477 mg/g [31]. The Thomas adsorption capacity found previously with AC modified with ECTFE–PEI using the same experimental conditions was about 400 m2/g [45], almost three times less than that found for AC–PEI–KelF–PEI. However, the qmax Langmuir value of 1242 reported for PFOA adsorbed to biochar–PEI [33] was comparable. Most carbon-based adsorbents for PFAS were in the 100–300 mg/g capacity range; however, cetyltrimethylammonium bromide-modified AC indicated an adsorption capacity of 456 [50]. The Thomas adsorption rate constant kT found previously for AC modified with ECTFE–PEI was about 0.3 mL/(min-mg) [45], greater by about 4 times as compared to that for AC–PEI–KelF–PEI.
Yoon–Nelson Model
The Yoon–Nelson model is frequently employed due to its computational simplicity, as it enables estimation of the 50% breakthrough time (τ) without requiring detailed information on adsorbent properties or column characteristics. However, meaningful comparison of τ values across different sorbents necessitates identical operating conditions, including flow rate and bed height. In the present study, 20 data points obtained from batch 1 of AC–PEI–KelF–PEI were evaluated, and the resulting analysis was consistent with trends observed using the Thomas model (Table S4). A linear correlation between ln [Ct/(C0 − Ct)] and time (t) is shown in Figure S16, yielding a strong fit with a correlation coefficient of R2 = 0.9617, thereby confirming the applicability of the Yoon–Nelson model to this system. The corresponding kinetic parameters, i.e., the Yoon–Nelson rate constant (kY) and breakthrough time (τ), were determined from Equation (S11) in the Supplementary Information. Averaged across three independent batches, AC–PEI–KelF–PEI exhibited a τ value of approximately 1500 min, with an associated adsorption rate constant (kY) of 0.0084 min−1. Reproducibility was fairly similar to that shown for the Thomas model in Table S5. The 50% breakthrough time for AC–ECTFE–PEI using the same experimental conditions was about 380 min [45], about 3.6 times less than that for AC–PEI–KelF–PEI. The Yoon–Nelson adsorption rate constant kY found previously for AC modified with ECTFE–PEI was about 0.033 min−1 [45], again greater by about 4 times as compared to that for AC–PEI–KelF–PEI.

4.5.3. Adsorption Kinetics Studies

The adsorption kinetics of PFOA on AC–PEI–KelF–PEI were evaluated using the pseudo-first-order, pseudo-second-order, and intraparticle diffusion models to clarify the rate-limiting steps governing the adsorption process, as shown in the Supplementary Information.
As shown in Figure S17A, the pseudo-first-order model provided only moderate agreement with the experimental data, with a lower correlation coefficient of 0.9296. This result suggests that the PFOA uptake on AC–PEI–KelF–PEI is not governed solely by physical adsorption or boundary-layer diffusion, and that the pseudo-first-order model (Equation (S12)) does not adequately represent the overall adsorption behavior of the dual-polymer-modified sorbent. This kinetic model yielded an equilibrium uptake of qe = 384.6 ± 1.2 mg/g and a rate constant k1 = 6.48 ± 0.27 min−1, where uncertainty represents standard errors obtained from the linear regression fitting of data.
In contrast, the pseudo-second-order model (see Equation (S13)) produced an excellent linear fit, with an R2 value of 0.9995 (Figure S17B). This kinetic model yielded an equilibrium uptake of qe = 400.3 ± 1.2 mg/g and a rate constant of k2 = (1.77 ± 0.16) × 10−4 g/(mg min), where uncertainty represents standard errors obtained from the linear regression fitting of data. This behavior indicates that the overall rate of PFOA uptake is controlled by chemisorption-like interactions rather than by simple physisorption. For AC–PEI–KelF–PEI, these interactions arise from the strong affinity between the deprotonated PFOA carboxylate headgroup and the protonated amine functionalities within the PEI layer, complemented by hydrophobic and fluorophilic interactions between the perfluorinated PFOA tail and the Kel-F phase. The dominance of the pseudo-second-order model therefore demonstrates that the formation of specific surface interactions is the principal rate-determining step during PFOA adsorption. This k2 rate constant was about 5 times slower than that reported previously for the adsorption of Cd (II) on AC–PEI [30] and three times slower than adsorption of Reactive Yellow dye #2 on AC–PEI [31]. However, our qe value was comparable to that (488) reported previously for PFOA adsorbed to biochar–PEI [33], but the k2 was significantly smaller by a factor of 100.
Further insight into the mass-transfer mechanism was obtained from the intraparticle diffusion model (see Equation (S14)). The qt vs. t1/2 plots displayed three distinct linear regions, as shown in Figure 5, indicating that PFOA adsorption proceeds through a multistep diffusion process, represented as internal diffusion rate constants (mg/(g-min1/2) kid1, kid2, and kid3. The three linear portions are plotted separately and fitted to trendlines in Figure S18. The initial steep segment corresponds to rapid film diffusion across the liquid–solid boundary layer (kid1 = 9.05), while the second region (kid2 = 5.49) reflects diffusion through the mesoporous structure of AC–PEI–KelF–PEI. The final, shallow segment (kid3 = 3.7) represents the slow penetration of PFOA into the micropores and sterically constrained regions of the polymer-modified carbon. These kid values are about three times faster than those reported previously for the adsorption of Cd (II) on AC–PEI [30]. However, these kid1 and kid2 values were about three times smaller than those reported for PFOA interacting with biochar–PEI [33]. In that work, no kid3 value was reported for PFOA. The decreasing slopes of the three segments confirm progressively increasing diffusion resistance during pore transport. Importantly, none of the linear regions intersect the origin, demonstrating that intraparticle diffusion is not the sole rate-limiting step. Instead, the adsorption of PFOA on AC–PEI–KelF–PEI is governed by a combination of film diffusion, internal pore diffusion, and chemisorption-like interactions occurring simultaneously at the functionalized surface. Taken together, the kinetic analysis demonstrates that PFOA adsorption on AC–PEI–KelF–PEI is best described by the pseudo-second-order model and is dominated by surface interaction mechanisms facilitated by the KelF–PEI dual-polymer coating. The segmented intraparticle-diffusion behavior further confirms that adsorption proceeds through a combination of external and internal diffusion steps before establishing strong interactions with the polymer-modified carbon surface. This multistage kinetic behavior is consistent with the hierarchical porosity and functional group distribution of AC–PEI–KelF–PEI, which together enhance the sorbent’s affinity toward PFOA.

4.6. Adsorption Mechanism

The adsorption of PFOA onto AC–PEI–KelF–PEI occurs through two dominant mechanisms: electrostatic attraction between the anionic carboxylate headgroup and protonated amine functionalities, and hydrophobic partitioning driven by the perfluorinated alkyl chain. Because PFOA has a pKa of approximately 2.2 ± 0.2 [61], it exists almost entirely in its anionic form under the experimental pH (3.5), creating favorable conditions for electrostatic binding. The pKa values for the various primary, secondary, and tertiary amines in the PEI structure range from 8–10. Even with dilute PFAS solutions having a pH close to 7, the amine groups with a pKa of 8 should remain 90% protonated, maintaining the ion exchange mechanism. However, it has been previously reported that the adsorption of PFOA by PEI weak anion exchange does not decrease significantly from pH 3–9 but does so for other PFAS compounds [33]. The reason for this is unclear. Simultaneously, the low polarizability and strong hydrophobicity of the –(CF2)n chain promote interaction with nonpolar domains of the polymer-modified surface [62]. To decouple these two mechanisms, a series of adsorption experiments was performed using 100 mg/L PFOA solutions prepared in mobile phases designed to selectively attenuate either electrostatic or hydrophobic interactions. The addition of 0.1 M Na2SO4 provided a high-ionic-strength environment in which perchlorate ions efficiently screen surface charges, suppressing electrostatic attraction and isolating the hydrophobic contribution. Conversely, replacing the aqueous phase with 90% MeCN lowered solvent polarity and reduced hydrophobic partitioning, allowing the electrostatic component to be evaluated. The 0.1M Na2SO4 solution background absorbance at 205 nm was not a detection issue.
Breakthrough curves were interpreted using the Thomas kinetic model, revealing clear shifts in adsorption performance under each selective condition (Figure S19). The unmodified mobile phase yielded the highest adsorption capacity (q0 = 1126 mg g−1), reflecting the combined contribution of hydrophobic and electrostatic forces. Under ionic-strength suppression (0.1 M Na2SO4), the capacity decreased to 201 mg g−1, capturing the hydrophobic-only contribution. However, this remaining PFAS adsorption could be stronger due to a salting-out effect as compared to PFAS adsorption from water. Under reduced polarity (90% MeCN), the adsorption capacity was 710 mg g−1, reflecting the electrostatic-only mode. When both suppressive conditions were combined (0.1 M Na2SO4 + 90% MeCN), PFOA adsorption was largely eliminated, confirming that both mechanisms are essential and synergistic [62], with the electrostatic mechanism contributing slightly more strongly to overall uptake (Figure 6). Other retention mechanisms besides electrostatic and hydrophobic processes could be possible [63]. Although a cation-bridging mechanism of PFAS might occur with a multi-valent metal, this retention mechanism would not be likely with sodium as the main cation present. The fluorophilic retention mechanism and how it interacts with PFOA and linear alkyl sulfonate probes has been elucidated using self-assembled monolayers of either a fluorocarbon or a hydrocarbon chain [64].

4.7. Regeneration and Selectivity

Following PFOA adsorption, the AC–PEI–KelF–PEI sorbent was regenerated using a 90% MeCN + 10% 0.1 M methanesulfonic acid (MSA) desorption solution, following the optimized protocol from our previous work [45]. The saturated columns were flushed with this desorption buffer for 30 min at a flow rate of 1 mL/min. The adsorption–desorption cycle was repeated five consecutive times to evaluate the reusability and structural stability of the sorbent (Figure S20).
The regeneration performance of AC–PEI–KelF–PEI was evaluated over six consecutive adsorption–desorption cycles, and the results are summarized in Figure 7. The sorbent exhibited excellent initial recoverability, with a regeneration efficiency of 100% after the first cycle, confirming that the desorption protocol effectively removed adsorbed PFOA without damaging the surface functionality. Over subsequent cycles, a gradual decline in regeneration efficiency was observed, decreasing to 94%, 93%, and 91% for cycles 2–4, and further to 83% and 79% by cycles 5 and 6, respectively. The moderate decline in performance is consistent with partial pore blockage, progressive saturation of strongly bound sites, or incremental structural changes in the noncovalently bound AC–PEI–KelF–PEI polymer that might reduce active-site accessibility. Nevertheless, regeneration efficiencies near 80% after six cycles demonstrate that AC–PEI–KelF–PEI retains substantial structural integrity, likely due to the crosslinked PEI–fluoropolymer, and adsorptive capability under repeated use. This was comparable to about a 70% regeneration efficiency of PFOA after four cycles using a biochar–PEI adsorbent [33]. However, low-molecular-weight PEI adsorbed to AC did not show regeneration after the first adsorption cycle of Reactive Yellow dye #2 [31]. Acetonitrile (MeCN) was chosen as the organic component of the desorption medium owing to its low UV cutoff (190 nm), which minimizes baseline noise and enhances detection sensitivity during UV analysis. However, for large-scale or industrial applications, ethanol could serve as a more sustainable alternative due to its environmental compatibility and comparable polarity (Hildebrand solubility parameters: 24.1 for MeCN vs. 26.5 for ethanol [65]. Feasibility of regeneration is still not a practical option if the application of the AC–PEI–KelF–PEI adsorbent is for household point-of-use.
However, covalent attachment of PEI to AC through sulfonyl chloride reaction chemistry or glutaraldehyde crosslinking would provide a good comparison of stability.
Selectivity of the modified adsorbent toward PFOA in the presence of structurally similar octanoic acid was evaluated using column breakthrough experiments. The breakthrough for octanoic acid occurred at about 150 min, while that for PFOA appeared at about 400 min (Figure S21). This second breakthrough profile showed a prolonged lag phase and a delayed rise in Co/Ct, indicating that PFOA remained strongly retained on the adsorbent even in the presence of a competing non-fluorinated carboxylic acid. Both compounds were completely ionized at the solution alkaline pH because the corresponding surface pKa values have been reported to be 3.8 for octanoic acid and less than 2.2 for PFOA [61]. The fluorine–fluorine interaction of PFOA with the KelF polymer is the likely cause of this selectivity [62].
Furthermore, the breakthrough behavior of three representative short chain PFAS, i.e., perfluoropentanoic acid (PFPeA), undecafluoro-2-methyl-3-oxahexanoic acid (GenX), and trifluoroacetic acid (TFA), was evaluated on AC–PEI–KelF–PEI. PFPeA is an example of a short-chain PFAS compound, and GenX is more polar in structure due to the ether linkage in the five-carbon fluorinated backbone. TFA is apparently found in the environment at levels several orders of magnitude over those of PFAS [66]. Breakthrough curves (Figure 8) fitted using the Thomas adsorption model yielded adsorption capacities of 98.6 mg/g, 162 mg/g, and 55.0 mg/g, respectively for PFPeA, GenX, and TFA, and the Yoon–Nelson model yielded the time (ι) to reach the 50% breakthrough times of 156 min, 257 min, and 80.1 min, respectively. Likely due to ion exchange, TFA was retained better on this AC–PEI–KelF–PEI adsorbent as compared to that on plain AC. These results highlight clear correlations between PFAS physicochemical attributes and sorption performance, demonstrating that retention is governed not only by electrostatic interactions but also by chain length and molecular architecture, which strongly modulate affinity and breakthrough kinetics on the modified activated carbon surface.

5. Conclusions

This work demonstrates that sequential polymer and amine functionalization of activated carbon using Kel-F and PEI yields a highly efficient and tunable adsorbent for PFOA removal under continuous-flow fixed-bed column conditions. The optimized AC–PEI–KelF–PEI material achieved at least an order-of-magnitude enhancement in breakthrough time (1050 min compared to 80 min) and adsorption capacity (1126 compared to 67 mg/g) for PFOA compared to the levels for unmodified AC, while maintaining quite good regeneration stability and promising selectivity in the presence of a competing organic acid. These findings highlight the synergistic role of fluorophilic Kel-F domains and amine-rich PEI sites in driving high-affinity PFAS capture.
Because of the limited commercial availability of PCTFE–VF polymers, future work should focus on using PVF or a co-polymer with hexafluoropropene [48] as the fluoropolymer adsorbent for PFAS. PVF, synthesized in various molecular weight ranges, can be obtained commercially, but product purity after the grafting reacting with PEI may be problematic. The solubility of these quite high molecular-weight types of PVF should be checked in various solvents at elevated temperatures to ensure that excess starting material can be filtered off from the adsorbent product at the required temperature.
The use of an inexpensive flow analysis instrument that can continuously monitor the breakthrough curves for short adsorbent packed columns has been shown to be an effective characterization method. Flexibility of detector choice is a plus. Application of breakthrough studies to perfluorosulfonated compounds, with the flow analysis instrument modified using conductivity detection, could be achieved. The scale-up to larger columns at higher flow rates and lower PFAS concentrations comprise options for potential future studies.
PFAS removal from environmentally relevant water samples (ppb levels), with LC–MS detection of predicted breakthrough samples, would represent an important extension beyond this proof-of-concept study.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/purification2030011/s1.

Author Contributions

Conceptualization, N.D.D. and O.A.; methodology, O.A. and N.D.D.; validation, O.A., C.B.A. and N.D.D.; formal analysis, N.D.D., C.B.A. and O.A.; investigation, O.A., N.D.D. and C.B.A.; resources, N.D.D., C.B.A. and J.A.B.; data curation, N.D.D.; writing—review and editing, N.D.D., O.A., C.B.A. and J.A.B.; visualization, O.A.; supervision, N.D.D.; project administration, N.D.D., C.B.A. and J.A.B.; funding acquisition, N.D.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Glüge, J.; Scheringer, M.; Cousins, I.T.; Dewitt, J.C.; Goldenman, G.; Herzke, D.; Lohmann, R.; Ng, C.A.; Trier, X.; Wang, Z. An overview of the uses of per- and poly-fluoroalkyl substances (PFAS). Environ. Sci. Process. Impacts 2020, 22, 2345–2373. [Google Scholar] [CrossRef] [PubMed]
  2. Brunn, H.; Arnold, G.; Körner, W.; Rippen, G.; Steinhäuser, K.G.; Valentin, I. PFAS: Forever chemicals—Persistent, bioaccumulative and mobile. Reviewing the status and the need for their phase out and remediation of contaminated sites. Environ. Sci. Eur. 2023, 35, 20. [Google Scholar] [CrossRef]
  3. Li, Y.; Li, Y.; Zhang, S.; Gao, T.; Gao, Z.; Lai, C.W.; Xiang, P.; Yang, F. Global distribution, ecotoxicity, and treatment technologies of emerging contaminants in aquatic nnvironments: A Recent Five-Year Review. Toxics 2025, 13, 616. [Google Scholar] [CrossRef] [PubMed]
  4. Panieri, E.; Baralic, K.; Djukic-Cosic, D.; Djordjevic, A.B.; Saso, L. PFAS Molecules: A major concern for the human health and the environment. Toxics 2022, 10, 44. [Google Scholar] [CrossRef] [PubMed]
  5. Ehrlich, V.; Bil, W.; Vandebriel, R.; Granum, B.; Luijten, M.; Lindeman, B.; Grandjean, P.; Kaiser, A.-M.; Hauzenberger, I.; Hartmann, C.; et al. Consideration of pathways for immunotoxicity of per-and polyfluoroalkyl substances (PFAS). Environ. Health 2023, 22, 19. [Google Scholar] [CrossRef] [PubMed]
  6. Buck, R.C.; Franklin, J.; Berger, U.; Conder, J.M.; Cousins, I.T.; De Voogt, P.; Jensen, A.A.; Kannan, K.; Mabury, S.A.; van Leeuwen, S.P.J. Perfluoroalkyl and polyfluoroalkyl substances in the environment: Terminology, classification, and origins. Integr. Environ. Assess. Manag. 2021, 7, 513–541. [Google Scholar] [CrossRef] [PubMed]
  7. Paul, A.G.; Jones, K.C.; Sweetman, A.J. A first global production, emission, and environmental inventory for perfluorooctane sulfonate. Environ. Sci. Technol. 2009, 43, 386–392. [Google Scholar] [CrossRef] [PubMed]
  8. Available online: https://www.epa.gov/system/files/documents/2024-04/pfas-npdwr_fact-sheet_general_4.9.24v1.pdf (accessed on 29 December 2025).
  9. Rashid, R.; Shafiq, I.; Akhter, P.; Iqbal, M.J.; Hussain, M. A state-of-the-art review on wastewater treatment techniques: The effectiveness of adsorption method. Environ. Sci. Pollut. Res. 2021, 28, 9050–9066. [Google Scholar] [CrossRef] [PubMed]
  10. Tony, M.A. Low-cost adsorbents for environmental pollution control: A concise systematic review from the prospective of principles, mechanism and their applications. J. Dispers. Sci. Technol. 2022, 43, 1612–1633. [Google Scholar] [CrossRef]
  11. Akhtar, M.S.; Ali, S.; Zaman, W. Innovative adsorbents for pollutant removal: Exploring the latest research and applications. Molecules 2024, 29, 4317. [Google Scholar] [CrossRef] [PubMed]
  12. Yadav, S.; Sharma, N.; Dalal, A.; Panghal, P.; Sharma, A.K.; Kumar, S. Cutting-edge regeneration technologies for saturated adsorbents: A systematic review on pathways to circular wastewater treatment system. Environ. Monit. Assess. 2025, 197, 215. [Google Scholar] [CrossRef] [PubMed]
  13. Alsawy, T.; Rashad, E.; El-Qelish, M.; Mohammed, R.H. A comprehensive review on the chemical regeneration of biochar adsorbent for sustainable wastewater treatment. NPJ Clean Water 2022, 5, 29. [Google Scholar] [CrossRef]
  14. Kebede, M.M.; Al Musud, M.A.; Ortbal, S.; Shin, W.S.; Mekonnen, M.M.; Clement, T.P.; Terry, L.G. Adsorptive removal of PFAS from aqueous solutions using GAC, PAC and ball-milled colloidal activated carbon: Characterizing efficiency, kinetics, and mechanisms. ACS EST Water 2025, 5, 6554–6566. [Google Scholar] [CrossRef]
  15. Gomri, C.; Benkhaled, B.T.; Cretin, M.; Semsarilar, M. Absorbent material used for the treatment of per- and poly-fluoroalkyl substances (PFAS): A short review. Macromol. Chem. Phys. 2024, 225, 2400012. [Google Scholar]
  16. Guo, W.; Tong, H.; Luo, D.; Zhao, X.; Long, Q.; Yin, C.; Yong, L. Adsorption-based removal of PFASs from water: Mechanisms, materials and future perspective. Environ. Chem. 2025, 22, EN25053. [Google Scholar] [CrossRef]
  17. Fatima, M.; Kelso, C.; Hai, F. Perfluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) adsorption onto different adsorbents: A critical review of the impact of their chemical structure and retention mechanisms in soil and groundwater. Water 2025, 17, 1401. [Google Scholar] [CrossRef]
  18. Parvin, P.; Hara-Yamamura, H.; Kanai, Y.; Yamasaki, A.; Adachi, T.; Sorn, S.; Honda, R.; Yamamura, H. Important properties of anion exchange resins for efficient removal of PFOS and PFOA from groundwater. Chemosphere 2023, 341, 139983. [Google Scholar] [CrossRef] [PubMed]
  19. Liu, Z.; Yang, F.; Zhai, T.; Yu, J.; Wang, C.; Liu, Z.; Liu, Z.; Gao, Y.; Yang, M. Removal of PFOA from water by activated carbon adsorption: Influence of pore structure. J. Environ. Chem. Eng. 2024, 12, 113923. [Google Scholar] [CrossRef]
  20. Saeidi, N.; Kopinke, F.D.; Georgi, A. What is specific in adsorption of perfluoroalkyl acids on carbon materials? Chemosphere 2021, 273, 128520. [Google Scholar] [CrossRef] [PubMed]
  21. Yu, Q.; Zhang, R.; Deng, S.; Huang, J.; Yu, G. Sorption of perfluorooctane sulfonate and perfluorooctanoate on activated carbons and resin: Kinetic and isotherm study. Water Res. 2009, 43, 1150–1158. [Google Scholar] [CrossRef] [PubMed]
  22. Qu, Y.; Zhang, C.; Li, F.; Bo, X.; Liu, G.; Zhou, Q. Equilibrium and kinetics study on the adsorption of perfluorooctanoic acid from aqueous solution onto powdered activated carbon. J. Hazard. Mater. 2009, 169, 146–152. [Google Scholar] [CrossRef] [PubMed]
  23. Ateia, M.; Alsbaiee, A.; Karanfil, T.; Dichtel, W. Efficient PFAS removal by amine-functionalized sorbents: Critical review of the current literature. Environ. Sci. Technol. Lett. 2019, 6, 688–695. [Google Scholar] [CrossRef]
  24. Dixit, F.; Dutta, R.; Barbeu, B.; Berube, P.; Mohseni, M. PFAS removal by ion exchange resins: A review. Chemosphere 2021, 272, 129277. [Google Scholar] [CrossRef]
  25. SenGupta, A.K. Ion Exchange in Environmental Processes: Fundamentals, Applications and Sustainable Technology. 2017. Available online: https://books.google.com/books/about/Ion_Exchange_in_Environmental_Processes.html?id=gEkzDwAAQBAJ (accessed on 28 October 2025).
  26. Thakara, S.C.; Jayaram, R.V. Amino-functionalized activated carbon materials in base-catalyzed reactions. Catal. Green Chem. Eng. 2018, 1, 113–126. [Google Scholar]
  27. Chandra, A.; Singh, A.A.; Prasad, S.; Andersson, M.R.; Gedefaw, D. Crosslinking approaches for polyethylene imine (PEI) and its uses in adsorption of heavy metals, dyes, and carbon dioxide. Appl. Sci. 2025, 15, 4767. [Google Scholar] [CrossRef]
  28. Lan, J.; Wang, B.; Gong, B. Polyethyleneimine modified activated carbon for high-efficiency adsorption of copper ion from simulated wastewater. Water Sci. Technol. 2022, 86, 2465. [Google Scholar] [CrossRef] [PubMed]
  29. Yin, C.Y.; Aroua, M.K.; Daud, W.M.A.W. Enhanced adsorption of metal ions onto polyethyleneimine-impregnated palm shell activated carbon: Equilibrium studies. Water Air Soil Pollut. 2008, 192, 337–348. [Google Scholar] [CrossRef]
  30. Xie, X.; Gao, H.; Luo, X.; Su, T.; Zhang, Y.; Qin, Z. Polyethyleneimine modified activated carbon for adsorption of Cd(II) in aqueous solution. J. Environ. Chem. Eng. 2019, 7, 103183. [Google Scholar] [CrossRef]
  31. Won, S.W.; Kang, S.B.; Min, B.-C.; Go, K.-I.; Yong, L.; Yun, Y.-S. Polyethyleneimine-impregnated powdered activated carbon for efficient removal of Reactive Yellow 2 from aqueous solutions. Korean J. Chem. Eng. 2025, 43, 253–256. [Google Scholar] [CrossRef]
  32. Zhang, Y.; Lee, C.-S.; Li, D.; Zheng, S.; Hsiao, B.S.; Venkatesan, A.K. Hydrophobic ion pairing as a pretreatment to enhance the removal of short-chain per- and poly-fluoroalkyl substances by granular activated carbon. Environ. Sci. Technol. 2025, 59, 20761–20771. [Google Scholar] [PubMed]
  33. Imran, A.; Lei, X.; Shoemaker, D.J.; Holmes, W.E.; Yan, H.; Zappi, M.E.; Gang, D.D. Adsorption of PFCAs using polyethyleneimine modified Biochar: Role of chain length and effects of water matrices. Chemosphere 2025, 386, 144650. [Google Scholar] [CrossRef] [PubMed]
  34. Paciorek, K.L.; Mitchell, L.C.; Lenk, C.T. Mechanism of amine crosslinking of fluoroelastomers. J. Polym. Sci. 1960, 45, 405–413. [Google Scholar] [CrossRef]
  35. Taguet, A.; Ameduri, B.; Boutevin, B. Crosslinking of vinylidene fluoride-containing fluoropolymers. Adv. Polym. Sci. 2005, 184, 127–211. [Google Scholar] [CrossRef]
  36. Maki, S.A.; Wangsa, J.; Danielson, N.D. Separation and detection of aliphatic anionic surfactants using a weak anion exchange column with indirect photometric and indirect conductivity detection. Anal. Chem. 1992, 64, 583–589. [Google Scholar] [CrossRef]
  37. Danielson, N.D.; Shamsi, S.A.; Maki, S.A. Comparison of fluorocarbon and hydrocarbon weak anion exchange columns for the separation of surfactants with indirect detection. J. High Resolut. Chromatogr. 1992, 15, 343–346. [Google Scholar] [CrossRef]
  38. Danielson, N.D.; Wangsa, J.; Shamsi, S.A. Synthesis and characterization of a polymeric fluorocarbon-diamine reversed phase weak anion exchange silica HPLC column packing. J. Liq. Chromatogr. 1995, 18, 2579–2592. [Google Scholar] [CrossRef]
  39. He, Y.; Tan, X.; Yu, C.; Liu, K.; Zhou, C.; Zhu, Y.; Leung, L.; Breske, S.; Hopkins, T.E.; Tottszer, A.; et al. Capturing PFAS with Partly Fluorinated Macromolecules: The Influence of the Fluorinated Group Structure. Macromolecules 2025, 58, 7050–7061. [Google Scholar] [CrossRef]
  40. Zhang, C.; Dong, J.; Zhang, P.; Sun, L.; Yang, L.; Wang, W.; Zou, X.; Chen, Y.; Shang, Q.; Feng, D.; et al. Unique fluorophilic pores engineering within porous aromatic frameworks for trace perfluorooctanoic acid removal. Natl. Sci. Rev. 2023, 10, nwad191. [Google Scholar] [CrossRef] [PubMed]
  41. Anjum, S.; Arik, M.; Patel, A.; Abasali, N.; Wu, L.; Sarkar, A. Fluorinated block copolymer: An important sorbent design criteria for effective PFOA removal from its aqueous solution. ACS Appl. Polym. Mater. 2025, 7, 1187–1193. [Google Scholar] [CrossRef] [PubMed]
  42. Tan, X.; Zhong, J.; Fu, C.; Dang, H.; Han, Y.; Kral, P.; Guo, J.; Yuan, Z.; Peng, H.; Zgang, C.; et al. Amphiphilic perfluoropolyether copolymers for the effective removal of polyfluoroalkyl substances from aqueous environments. Macromolecules 2021, 54, 3447–3457. [Google Scholar] [CrossRef]
  43. Xie, R.; Zhou, L.; Smith, A.E.; Almquist, C.B.; Berberich, J.A.; Danielson, N.D. A dual grafted fluorinated hydrocarbon amine weak anion exchange resin polymer for adsorption of perfluorooctanoic acid from water. J. Hazard. Mater. 2022, 431, 128521. [Google Scholar] [CrossRef] [PubMed]
  44. Olomukoro, A.A.; Xie, R.; Fernandez Paucar, F.X.; DeRosa, C.; Danielson, N.D.; Gionfriddo, E. Characterization of a mixed mode fluorocarbon/weak anion exchange sorbent for the separation of perfluoroalkyl substances. J. Sep. Sci. 2024, 47, 2400413. [Google Scholar] [CrossRef]
  45. Xie, R.; Beckman, M.T.; Almquist, C.B.; Berberich, J.A.; Danielson, N.D. Fixed-bed adsorption of perfluorooctanoic acid from water by a polyamine-functionalized polychlorotrifluoroethylene-ethylene polymer coated on activated carbon. J. Environ. Chem. Eng. 2024, 12, 113001. [Google Scholar] [CrossRef]
  46. Schaefer, C.S.; Nguyen, D.; Culina, V.M.; Guelfo, J.; Kumar, N. Application of Rapid Small-Scale Column Tests for Treatment of Perfluoroalkyl Acids using Anion Exchange Resins and Granular Activated Carbon in Groundwater with Elevated Organic Carbon. Ind. Eng. Chem. Res. 2020, 59, 16832–16837. [Google Scholar] [CrossRef]
  47. Cady, W.E.; Caley, L.E. Properties of Kel F-800 Polymer; UCRL-52301; Lawrence Livermore Laboratory: Livermore, CA, USA, 1977. [Google Scholar]
  48. Taguet, A.; Ameduri, B.; Boutevin, B. Grafting of commercially available amines bearing aromatic rings onto poly(vinylidene-co-hexafluoropropene) copolymers. J. Polym. Sci. Part A Polym. Chem. 2006, 4, 1855–1868. [Google Scholar] [CrossRef]
  49. Heidarinejad, Z.; Dehghani, M.H.; Heidari, M.; Javedan, G.; Ali, I.; Sillanpaa, M. Methods for preparation and activation of activated carbon: A review. Environ. Chem. Lett. 2020, 18, 393–415. [Google Scholar] [CrossRef]
  50. Okorie, C.J.; Ojeyemi, T.; Egbemhenghe, A.; Ali, M.Q.; Emenike, E.C.; Iwuozor, K.O.; Adeniyi, A.G. Modification of activated carbon for enhanced treatment of per-and polyfluoroalkyl substances: A focused review. Remediat. J. 2024, 34, e21777. [Google Scholar] [CrossRef]
  51. Chow, J.C.; Yu, J.Z.; Watson, J.G.; Ho, S.S.H.; Bohannan, T.L.; Hays, M.D.; Fung, K.K. The application of thermal methods for determining chemical composition of carbonaceous aerosols: A review. J. Environ. Sci. Health Part A 2007, 42, 1521–1541. [Google Scholar] [CrossRef] [PubMed]
  52. Bazan, A.; Nowicki, P.; Polrolniczak, P.; Pietrzak, R. Thermal analysis of activated carbon obtained from residue after supercritical extraction of hops. J. Therm. Anal. Calorim. 2016, 125, 1199–1204. [Google Scholar] [CrossRef]
  53. Zulfiqar, S.; Zulfiqar, M.; Rizvi, M.; Munir, A.; McNeill, I.C. Study of the thermal degradation of polychlorotrifluoroethylene, poly(vinylidene fluoride) and copolymers of chlorotrifluoroethylene and vinylidene fluoride. Polym. Degrad. Stab. 1994, 43, 423–430. [Google Scholar] [CrossRef]
  54. Chingombe, P.; Saha, B.; Wakeman, R.J. Surface modification and characterisation of a coal-based activated carbon. Carbon N. Y. 2005, 43, 3132–3143. [Google Scholar] [CrossRef]
  55. Sing, K.S.W. The use of gas adsorption for the characterization of porous solid. Colloids Surf. 1989, 38, 113–124. [Google Scholar] [CrossRef]
  56. Le Cao, K.A.; Cao, K.L.A.; Abdillah, O.B.; Septiani, E.L.; Hirano, T.; Nguyen, N.T.; Ogi, T. Correlation between Pore Characteristics and High-Performance Carbon Dioxide Capture of Sustainable Porous Carbon Derived from Kraft Lignin and Potassium Carbonate. Energy Fuels 2025, 39, 6372–6387. [Google Scholar] [CrossRef]
  57. Phadungbut, P.; Do, D.D.; Nicholson, D. On the microscopic origin of the hysteresis loop in closed end pores—Adsorbate restructuring. Chem. Eng. J. 2016, 285, 428–438. [Google Scholar] [CrossRef]
  58. Dollimore, D.; Spooner, P.; Turner, A. The bet method of analysis of gas adsorption data and its relevance to the calculation of surface areas. Surf. Technol. 1976, 4, 121–160. [Google Scholar] [CrossRef]
  59. Ojeda, M.L.; Esparza, J.M.; Campero, A.; Cordero, S.; Kornhauser, I.; Rojas, F. On comparing BJH and NLDFT pore-size distributions determined from N2 sorption on SBA-15 substrata. Phys. Chem. Chem. Phys. 2003, 5, 1859–1866. [Google Scholar] [CrossRef]
  60. Briese, E.; Niimi, K.; Hjelmstad, A.; Westerhoff, P. Surface complexation and packed bed mass transport models enable adsorbent design for arsenate and vanadate removal. ACS EST Eng. 2024, 4, 2563–2572. [Google Scholar] [CrossRef]
  61. Musegades, L.J.; Curtin, O.P.; Cyran, J.D. Determining the Surface pKa of Perfluorooctanoic Acid. J. Phys. Chem. C 2024, 128, 1946–1951. [Google Scholar] [CrossRef] [PubMed]
  62. Tan, X.; Sawczyk, M.; Chang, Y.; Wang, Y.; Usman, A.; Fu, C.; Král, P.; Peng, H.; Zhang, C.; Whittaker, A.K. Revealing the Molecular-Level Interactions between Cationic Fluorinated Polymer Sorbents and the Major PFAS Pollutant PFOA. Macromolecules 2022, 55, 1077–1087. [Google Scholar] [CrossRef]
  63. Leung, S.C.E.; Wanninayake, D.; Chen, D.; Nguyen, N.-T.; Li, Q. Physiochemical properties and interactions of perfluoroalkyl substances (PFAS)—Challenges and opportunities in sensing and remediation. Sci. Total Environ. 2023, 905, 166764. [Google Scholar] [CrossRef] [PubMed]
  64. Fang, C.; Sobhani, Z.; Megharaj, M.; Naidu, R. Electrochemical proof of fluorophilic interaction among fluoro-carbon chains. Electroanalysis 2018, 30, 2349–2355. [Google Scholar] [CrossRef]
  65. Zeng, W.; Du, Y.; Xue, Y.; Frisch, H.L. Solubility Parameters. In Physical Properties of Polymers Handbook; Springer: New York, NY, USA, 2007; pp. 289–303. [Google Scholar] [CrossRef]
  66. Arp, H.P.H.; Gredelj, A.; Gluge, J.; Scheringer, M.; Cousins, I.T. The global threat from the irreversible accumulation of trifluoroacetic acid (TFA). Environ. Sci. Technol. 2024, 58, 19925–19935. [Google Scholar] [CrossRef]
Purification 02 00011 g001
Figure 1. Kel F-800 structure; PEI structure; crosslinking reaction of poly (vinylidene) fluoride with PEI.
Figure 1. Kel F-800 structure; PEI structure; crosslinking reaction of poly (vinylidene) fluoride with PEI.
Purification 02 00011 g001
Purification 02 00011 g002
Figure 2. Thermogravimetric analyses of adsorbents: relative mass vs. temperature (°C). (a) Wet basis (relative to mass at room temperature); (b) dry basis (relative to mass at 120 °C).
Figure 2. Thermogravimetric analyses of adsorbents: relative mass vs. temperature (°C). (a) Wet basis (relative to mass at room temperature); (b) dry basis (relative to mass at 120 °C).
Purification 02 00011 g002
Purification 02 00011 g003
Figure 3. SEM micrographs of adsorbents. Mag = 12,000×. (A) Activated carbon (AC); (B) AC–PEI; (C) AC–PEI–KelF; (D) AC–PEI–KelF–PEI.
Figure 3. SEM micrographs of adsorbents. Mag = 12,000×. (A) Activated carbon (AC); (B) AC–PEI; (C) AC–PEI–KelF; (D) AC–PEI–KelF–PEI.
Purification 02 00011 g003
Purification 02 00011 g004
Figure 4. Adsorption breakthrough curves in order of activated carbon (AC) (navy blue) and polymer-modified adsorbents AC–PEI (green), AC–PEI–KelF (light blue), AC-2 (KelF–PEI) (orange), and AC–PEI–KelF–PEI (purple) evaluated under identical fixed-bed flow conditions. Breakthrough profiles are shown as normalized effluent concentration (Ct/C0) vs. time (C0 = 100 mg L−1, flow rate = 1 mL min−1, bed height = 2 cm).
Figure 4. Adsorption breakthrough curves in order of activated carbon (AC) (navy blue) and polymer-modified adsorbents AC–PEI (green), AC–PEI–KelF (light blue), AC-2 (KelF–PEI) (orange), and AC–PEI–KelF–PEI (purple) evaluated under identical fixed-bed flow conditions. Breakthrough profiles are shown as normalized effluent concentration (Ct/C0) vs. time (C0 = 100 mg L−1, flow rate = 1 mL min−1, bed height = 2 cm).
Purification 02 00011 g004
Purification 02 00011 g005
Figure 5. Internal diffusion plot of qₜ vs. t1ᐟ2 showing three distinct diffusion regimes representing internal diffusion rate constants (mg/(g-min1/2) kid1, kid2, and kid3.
Figure 5. Internal diffusion plot of qₜ vs. t1ᐟ2 showing three distinct diffusion regimes representing internal diffusion rate constants (mg/(g-min1/2) kid1, kid2, and kid3.
Purification 02 00011 g005
Purification 02 00011 g006
Figure 6. Adsorption capacity (q0) of AC–PEI–KelF–PEI) under different PFOA mobile-phase conditions, including water, 90% MeCN, 0.1 M Na2SO4, and 0.1 M Na2SO4 + 90% MeCN. The mean q0 value in water was calculated from three adsorption cycles (n = 3), while q0 values for the other mobile phases were calculated from two adsorption cycles (n = 2). (C0 = 100 mg/L, flow rate = 1 mL/min, bed height = 2 cm).
Figure 6. Adsorption capacity (q0) of AC–PEI–KelF–PEI) under different PFOA mobile-phase conditions, including water, 90% MeCN, 0.1 M Na2SO4, and 0.1 M Na2SO4 + 90% MeCN. The mean q0 value in water was calculated from three adsorption cycles (n = 3), while q0 values for the other mobile phases were calculated from two adsorption cycles (n = 2). (C0 = 100 mg/L, flow rate = 1 mL/min, bed height = 2 cm).
Purification 02 00011 g006
Purification 02 00011 g007
Figure 7. Regeneration recovery of the AC–PEI–KelF–PEI column over six consecutive PFOA adsorption–desorption cycles. PFOA adsorption was performed using an influent concentration of 100 mg/L at pH 3.5, and desorption was carried out with 90% MeCN + 10% 0.1 M MSA. (Flow rate = 1 mL/min; bed height = 2 cm). Error bars represent n = 2.
Figure 7. Regeneration recovery of the AC–PEI–KelF–PEI column over six consecutive PFOA adsorption–desorption cycles. PFOA adsorption was performed using an influent concentration of 100 mg/L at pH 3.5, and desorption was carried out with 90% MeCN + 10% 0.1 M MSA. (Flow rate = 1 mL/min; bed height = 2 cm). Error bars represent n = 2.
Purification 02 00011 g007
Purification 02 00011 g008
Figure 8. Breakthrough curves of trifluoroacetic acid (TFA) (green), perfluoropentanoic acid (PFPeA) (orange), and undecafluoro-2-methyl-3-oxahexanoic acid (GenX) (blue) on AC–PEI–KelF–PEI.
Figure 8. Breakthrough curves of trifluoroacetic acid (TFA) (green), perfluoropentanoic acid (PFPeA) (orange), and undecafluoro-2-methyl-3-oxahexanoic acid (GenX) (blue) on AC–PEI–KelF–PEI.
Purification 02 00011 g008
Table 1. BET analysis results for BET surface area, pore volume, and average pore width of AC and polymer-modified AC sorbents.
Table 1. BET analysis results for BET surface area, pore volume, and average pore width of AC and polymer-modified AC sorbents.
SampleBET Surface Area (m2/g)1 Micropore Volume
(cm3/g)		2 Pore Volume (cm3/g)3 Average Pore Width (nm)
AC11740.1990.8365.31
AC–PEI6490.1080.4495.17
AC–PEI–KelF5480.0870.3985.29
AC–PEI–KelF–PEI4720.0700.3555.28
1 t-plot method between relative pressure 0.01–0.6, thickness range: 3.5–5.0 Å. 2 BJH adsorption cumulative volume of pores between 17.000 and 3000.000 Å width. 3 BJH adsorption average pore width (4 V/A).
Table 2. Yoon–Nelson and Thomas model data.
Table 2. Yoon–Nelson and Thomas model data.
AdsorbentYoon–Nelson 50% Breakthrough Time (min) Thomas Adsorption Capacity (mg/g)Yoon–Nelson
Adsorption Rate Constant-kY (1/min)		Thomas Rate Constant kT (mL/(min-mg)
AC/AC–solvent84.20/75.59
±1.9/±0.64		66.75/64.65
±1.2/0.8		0.23
±0.0135		2.3
±0.135
AC–PEI504.8
±10.03		365.2
±9.02		0.035
±0.006		0.35
±0.06
AC–PEI–KelF537.6
±8.83		489.3
±11.67		0.007
±0.0045		0.07
±0.045
AC–PEI–KelF–PEI1517
±73.7		1126
±69		0.008
±0.005		0.08
±0.05
AC-2(PEI–KelF)1357
±72.5		868.3
±66		0.02
±0.01		0.2
±0.1
Uncertainties are ± standard deviation (n = 3).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Ayeseni, O.; Almquist, C.B.; Berberich, J.A.; Danielson, N.D. Perfluoroalkyl Substance Adsorption Using Activated Carbon Modified Sequentially with Polyethyleneimine and Poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel-F). Purification 2026, 2, 11. https://doi.org/10.3390/purification2030011

AMA Style

Ayeseni O, Almquist CB, Berberich JA, Danielson ND. Perfluoroalkyl Substance Adsorption Using Activated Carbon Modified Sequentially with Polyethyleneimine and Poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel-F). Purification. 2026; 2(3):11. https://doi.org/10.3390/purification2030011

Chicago/Turabian Style

Ayeseni, Omobolaji, Catherine B. Almquist, Jason A. Berberich, and Neil D. Danielson. 2026. "Perfluoroalkyl Substance Adsorption Using Activated Carbon Modified Sequentially with Polyethyleneimine and Poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel-F)" Purification 2, no. 3: 11. https://doi.org/10.3390/purification2030011

APA Style

Ayeseni, O., Almquist, C. B., Berberich, J. A., & Danielson, N. D. (2026). Perfluoroalkyl Substance Adsorption Using Activated Carbon Modified Sequentially with Polyethyleneimine and Poly(chlorotrifluoroethylene–co-vinylidene fluoride) (Kel-F). Purification, 2(3), 11. https://doi.org/10.3390/purification2030011

Article Metrics

Back to TopTop