Alginate cryogel beads for effectively aggregating nanoplastics for water remediation

Abstract

Nanoplastic (NP) pollution, consisting of particles smaller than 1 µm, poses a significant threat to both global ecosystems and human health. However, effective removal remains challenging due to their sub-micron size and low environmental concentrations. In this research, we discovered that an alginate cryogel can rapidly aggregate NPs (50–200 nm) into micrometer-sized clusters, enabling efficient removal via conventional membrane filtration. This cryogel-enhanced filtration achieved over 99% NP removal within 2 minutes. We attribute the observed aggregation to the combined effects of weakly bound alginate leaching upon cryogel rehydration and localized Ca²⁺ release from cryogels on driving NP aggregation. This mechanism enables effective NP removal across a pH range of 4–8 for both spherical and irregularly shaped NPs. While surface adsorption plays a role in NP removal, aggregation predominantly resulted into the effective filtration of NPs, with minimal influence from cryogel’s internal porosity. By leveraging NP aggregation into the microscale rather than relying on size-dependent direct filtration, this strategy presents a promising scalable solution for wastewater treatment and broader environmental applications.

Nanoplastic pollution poses a significant threat to both global ecosystems and human health, but its effective remediation is challenging due to the size and low environmental concentrations of nanoplastics. Here, the authors show that an alginate cryogel can rapidly aggregate 50–200 nm particles into micrometer-sized clusters, enabling efficient removal through conventional membrane filtration due to a combination of weakly bound alginate leaching upon cryogel rehydration and localized Ca²⁺ release from the cryogel.

Introduction

Plastic pollution has emerged as a significant threat to ecosystems and is recognized as an indicator under Goal 14 of the United Nations’ 2030 Sustainable Development Goals (SDGs)¹. The world produces around 430 million metric tons of plastic annually, which is roughly equivalent to the weight of 538 Confederation Bridges that connect Prince Edward Island and New Brunswick. One of the primary challenges in addressing plastic pollution is its tendency to fragment into smaller particles. Macro-sized plastic waste, such as discarded bottles and bags, has evolved into a more pervasive and insidious form: microplastics (MPs) and nanoplastics (NPs). MPs (1 µm–5 mm) can be as small as a strand of human hair ( ~ 70 µm) or the width of red blood cells (7–8 µm)², while NPs ( < 1 µm) can be comparable to the COVID-19 virus (60–140 nm)³ or the width of DNA strands ( ~ 2 nm)⁴. While some MPs and NPs are engineered for use in products like detergents and cosmetics, a significant portion results from the degradation of larger plastics through physical weathering, biodegradation, and chemical erosion. Regardless of their source, MPs and NPs have permeated nearly every corner of the environment, contaminating the air we breathe, the water we drink, and the food we consume^5–7. It is therefore imperative to remove such small fragments of plastics, particularly the NPs, which are invisible to the naked eye but are creating huge environmental and human health impacts.

MPs and NPs infiltrated living organisms such as live stocks^8,9, fishes^10,11, plants¹², and human^7,13, raising profound concerns about ecological and health because they often act as carriers for hazardous substances such as heavy metals, pesticides, and bisphenol A (BPA)^14,15. Plastic accumulated in animals can cause decreased fertility and behavioral abnormalities^16–18. For humans, while long-term effects remain under investigation, early findings point to troubling links between plastic exposure and hormone disruption, cardiovascular diseases, and cancer¹³. NPs can be especially more detrimental to the body than MPs as they penetrate the critical organs more readily, enhancing cellular interaction and staying in the organism for longer^19–21.

Wastewater treatment plants (WWTPs) serve as the primary defense against plastic pollution, as airborne and terrestrial plastics accumulate through rain, wind, and river transport before settling in aquatic environments^22,23. Recent studies have assessed conventional WWTPs’ technologies for their effectiveness in removing MPs and NPs (see Table S1). Traditional coagulation with ferric and aluminum salts demonstrated low removal efficiencies, as low as 2% for NPs²⁴, but when combined with rapid sand filtration, MP removal improved to 98%. Advanced coagulants enhanced efficiency for NPs to over 90% at a high cost. However, coagulation is less effective for those with densities similar to water (e.g., polyethylene (PE)), and it generates large volumes of sludge containing high concentrations of plastic particles, leading to secondary contamination^25–27. Activated carbon (AC), commonly used in tertiary treatment, has shown promise, though it is generally less effective than coagulation/flocculation and sand filtration^28–31. However, lab-scale tests with modified AC absorbent derived from waste epoxy achieved ~ 94% removal efficiency for NPs ( ~ 350 nm)³². Despite this, high WWTP flow rates limit adsorption efficiency, and reusability of AC decreases due to blocked pores with NPs and incomplete NP removal and formation of toxic byproduct during/after thermal regeneration^28–31. Sand filtration, another tertiary treatment, has been debated due to concerns that its high strength could fragment plastic particles^33,34. Nonetheless, rapid sand filtration (RSF) has demonstrated removal efficiencies exceeding 99% at low NP concentrations (0.2–0.02 mg) and low flow rates (2.26 m³/h, the lowest limit of RSF)²⁴. Overall, the current wastewater treatment methods face significant challenges for removing NPs, including incomplete NP removal, high costs, scalability issues, and risks of secondary contamination or plastic fragmentation. There is a pressing need for more cost-effective, high-efficiency, and low-impact alternatives for NPs removal technologies.

In recent years, increasing attention has been directed toward more effective and sustainable technologies to overcome these challenges (see Table S2). Among them, hydrogel-based methods have been particularly promising due to their biocompatibility and three-dimensional structure composed of hydrophilic polymer chains, allowing high water absorption, customizability, and permeability to pollutants. Their customizable structure provides flexibility to tailor treatment processes for better performance, potentially addressing issues of low efficiency, high sludge generation, and environmental impact seen in conventional methods. It has been demonstrated that nanocellulose hydrogels removed nanometric sizes of polystyrene (PS) and PE under varied pH and ionic strengths, utilizing their hygroscopic nature³⁵. Similarly, a nanofibrillated cellulose (NFC)-based injection-driven filter system achieved over 90% of various pollutants, such as TiO₂, silver, and river sediment³⁶. New hydrogel formulations, incorporating components such as polyethyleneimine, polydopamine, and graphene oxide nanosheets, have also been developed to provide strong binding interactions for MP adsorption^37–39. Nonetheless, these formulations often fail to effectively remove NPs and are unsuitable for industrial-scale use due to high material costs, complex preparation processes or lengthy processing times.

More recently, freeze-dried hydrogels known as cryogels and hydrogels dried using supercritical methods known as aerogels, have emerged as promising advances in hydrogel technology that can overcome many of the limitations faced by traditional hydrogels. Several researchers have focused on cryogel or aerogels to enhance water diffusion into the gel network and improve removal efficiency. For MPs, high removal efficiencies of 90 - 99% have been achieved using interconnected nanoporous filters made with freeze-dried hydrogel structures^40,41. For NPs, highest removal rate of 96% was observed when surface modifications were performed to the porous material itself, using amine-modified silica⁴². However, these technologies still struggle with particles in the nanoscale range and hold limited industrial application due to complex synthesis, limited scalability, declining performance upon reuse, and poor filtration flow.

To address these challenges, we developed a cryogel-based method specifically designed for efficient NP removal in aqueous environments. Our approach enables effective remediation through aggregation-assisted removal, possible due to the synergistic effects of sodium alginate adsorption onto NPs and Ca²⁺ (originating from the salt component of the alginate)-induced dispersion destabilization, which promotes aggregation. Recognizing that NP surfaces in natural waters are highly variable, we tested our cryogels under diverse conditions, including varying types and concentrations of NPs, variations in pH, and the presence of ionic and nonionic surfactants, in designing our approach. We also investigated how tuning the porosity of cryogels using sodium chloride could influence NP removal, alongside evaluating the cryogel beads’ potential for reuse. In this study, surfactants-assisted dispersions of PS and PE NPs in water were selected as nanoplastic model systems to probe the aggregation mechanism under controlled conditions. Although the original PS and PE NPs are hydrophobic, the presence of the surfactant alters their surface wettability, effectively reducing their hydrophobicity and enabling stable dispersion in the aqueous solution studied. Unlike previous cryogel-based approaches or conventional filtration-based methods, that often rely on passive entrapment, surface modification of particles, or pore size exclusion, our method leverages active aggregation to enhance removal efficiency and improve scalability for WWTP applications. This approach holds promise for NP pollution remediation in both drinking water supplies and aquatic ecosystems, having high NPs removal efficiency, a simple synthesis process, reusability, and low efficient filtration flow.

Results

Cryogel bead characterization

Alginate-based cryogels were fabricated through ionic crosslinking of sodium alginate with calcium chloride, initially producing hydrogels, followed by freeze-drying under controlled freezing protocols (liquid nitrogen immersion or conventional freezing), as described in Fig. 1a, b. Characterization shows that alginate hydrogel beads exhibited a wavy surface structure under ESEM observation (Fig. S1). Following liquid nitrogen treatment and freeze-drying, the hydrogel beads shrank by ~75–85% in size, as shown in Figs. 1c, and S2a, S2c. In the absence of liquid nitrogen treatment, the alginate showed an even greater size reduction, and its surface underwent shrinkage, resulting in a wrinkled morphology. Consequently, the surface topology was enhanced, while the effective pore size was reduced (Figs. 1, and S2b, S2d). The cross-sections of Alg LN 1:0 exhibited layered, fragmented sections, while Alg xLN 1:0 showed a denser structure.

Fig. 1. Preparation of alginate cryogel beads.

a Experimental process to fabricate alginate cryogel beads. b Alginate chemical structure, where the purple dashed circles represent Sodium ions, and the green solid circles represent the Calcium ions. c Freeze-dried following liquid nitrogen treatment (Alg LN 1:0), with SEM images taken at magnifications of 37, 33.75 K, and 30 K from left to right, respectively. d Freeze-dried without liquid nitrogen treatment (Alg xLN 1:0), with SEM images taken at magnifications of 53, 10 K, and 30 K from left to right, respectively. Without liquid nitrogen treatment, the alginate surface shrank into a wrinkled morphology, increasing surface topology and reducing effective pore size. In (c, d), each row presents the whole view, surface, and cross-section of alginate beads. Scale bars are given in each image.

NaCl was added before crosslinking to increase water ionic strength and act as a pore-forming agent within the alginate gel. The addition of NaCl and liquid nitrogen treatment increased the number of pores on the surface of alginate beads (Fig. S2). Monovalent cations, such as Na⁺ cannot form stable crosslinks between alginate chains due to insufficient charge density, unlike divalent cations (Ca²⁺). Therefore, monovalent cations added during pregel preparation were washed out, creating micro-voids within the alginate matrix.

Notably, while the addition of NaCl increased the porosity, we found that for both ratios, 1:1 and 1:3, the internal voids were not connected to their surfaces (Fig. S3a, c, d). A denser wall, ~ 50 µm thick, encapsulated the interior voids, potentially preventing the diffusion of the external fluids into the bead interiors. Also, alginate cryogel beads at a salt ratio of 1:1 showed irregularly shaped large voids (Fig. S3a, b) while a salt ratio of 1:3 created mesh-like internal structures (Fig. S3c, d).

The porosity of the alginate beads was further quantified using BET analysis and SEM measurements. BET surface area (S_BET), pore volume, and pore size distribution were derived from the adsorbed and desorbed N₂ quantity (Fig. S4a, b). Note that the BET method primarily captures micropores/ mesopores and cannot probe macropores ( > 50 nm). Therefore, SEM images were used as a complementary measurement to observe the presence of macropores. The trends observed in the BET analysis were consistent with the SEM findings: porosity and surface area increased with higher salt concentrations and liquid nitrogen treatment (see Table S3). Alg LN 1:3 exhibited peaks at 2.70 and 7.19 nm, while Alg LN 1:1 and 1:0 showed peaks at 7.13 and 2.70 nm, respectively. These measurements were included to quantify relative changes in porosity trends, but BET-derived pore sizes (peaks around ~2–7 nm) are much smaller than the true pore sizes seen in SEM, due to the technique’s limitations.

Similarly, higher salt concentration and liquid nitrogen treatment were shown to increase the swelling behavior of alginate cryogel beads. All beads reached saturation in 2 min (Fig. S4c).

NP removal analysis

Unless otherwise specified, the removal of NPs was tested using synthesized PS NPs with diameters ranging from 50 to 200 nm (Fig. S5, and Supplementary Videos. S1 and S2). Polyethylene (PE) NPs were also evaluated in selected tests. Therefore, the results primarily represent PS and PE NPs and may not directly extend to all NP types. Synthesized PS NPs were dispersed with the ionic surfactant, SDS (see Supporting Information). Figure 2 illustrates the overall process of NP removal. The influent was initially opaque, and NPs were not observable under optical microscopy because NPs were homogeneously dispersed (see Fig. 2a, e). Once they were in contact with alginate cryogel beads, they created aggregation within ~ 2 min (see Fig. 2b and c). Aggregated NPs are bigger than 20 µm, enabling the observation of NPs using optical microscopy, as shown in Fig. 2f. After filtering the aggregates, clear effluent was achieved (see Fig. 2d, g). Supplementary Video S3 shows the same process with synthesized PS NPs and Alg LN 1:1.

Fig. 2. Aggregation and removal of polystyrene nanoplastics (PS NPs).

a NP dispersion (i.e., influent) is prepared. b Alginate cryogel beads are added to the NPs dispersion. c NPs begin to aggregate within minutes. d The aggregates are filtered using a commercial 0.45 µm nylon syringe filter. e The opaque NPs dispersion is observed with the naked eye (left) and under an optical microscope (right). f NP aggregation facilitated by alginate cryogel beads is observed under an optical microscope. g Transition of the solution from NP aggregates to clear effluent with Alg LN 1:0 and Alg LN 1:3. h, i SEM images of PS NPs lodged in the crevices and surface pores of Alg LN 1:0, taken at magnifications of 20 K, and 45.16 K, respectively. j SEM images of PS NPs sitting on the surface of Alg LN 1:3 at a magnification of 30 K. k Clean cross-sectional view of Alg LN 1:3 after use for PS NP removal at a magnification of 10 K. Scale bars are given in each image.

Note that no NPs were observed inside the pores of the alginate cryogel beads, despite their large internal surface area. Although greater pore accessibility could, in principle, enhance removal efficiency by increasing surface area for interaction and allowing deeper NP diffusion, we hypothesized that limited wettability or slow rehydration of the cryogel surface might be restricting NP entry. To test this, we fully hydrated AlgLN 1:3 to equilibrium before introducing it to the NP suspension. Contrary to our expectation, this pre-hydration step neither induced NP aggregation nor removed NPs from suspension (Fig. S6), a result consistently observed for both AlgLN 1:3 and AlgxLN 1:3. This indicates that pre-hydration eliminates the transient conditions, such as steep hydration gradients, localized Ca²⁺ release, and capillary forces, that arise during initial water influx into the dried cryogel. These transient microenvironments appear critical for promoting NP aggregation and capture. Thus, the dried state of the cryogel beads plays an active and essential role in NP removal, whereas simply increasing wettability through pre-hydration does not enhance uptake. Other strategies to improve surface accessibility may warrant further investigation.

SEM confirmed the presence of PS NPs on the surface of the cryogel beads (see Fig. 2h–j), with no evidence of nanoparticles penetrating the interior, regardless of porosity (Fig. 2k). PS NPs tended to form localized multilayer structures on compositions with lower surface area, such as Alg LN 1:0, whereas a more uniform monolayer was observed on the surface of cryogels with higher surface area, like Alg LN 1:3. Overall, the macroporosity in the freeze-dried beads leads to low N₂ values adsorption surface area, which is consistent with their cryogel nature.

Based on these images, we emphasize that the addition of NaCl increases the surface area of the cryogel, resulting in a greater number of available adsorption sites on its surface. This is supported by the more homogeneous coverage of NPs observed across the surface, rather than localized multilayer formations. However, beads with the highest porosity (Alg LN 1:3) exhibited clean cross-sections in SEM imaging (Fig. 2k), implying that their relatively denser walls prevented NP penetration into the interior, as also observed in Fig. S3.

After filtering NP aggregates with a 0.45 µm pore syringe filter, the residual NP concentration in effluent was analyzed using NTA (see Fig. 3a). The removal efficiency of alginate cryogels, calculated with Eq. (1), was all higher than 99.9% regardless of variations in salt concentration, liquid nitrogen treatment, and the waiting time to collect NPs. Pure alginate cryogel beads presented the highest removal efficiency, but the maximum difference in removal efficiency among the bead types was minimal, amounting to only 0.07% (1.93 × 10⁹ particles/mL) and fell within the standard deviation range. The mean diameter of residual PS NPs in effluent was within the range of between 80 and 200 nm, while the mean diameter of PS NPs in influent was 100 ± 1.4 nm (Fig. S7). For comparison, the synthesized PS NP dispersion was directly filtered using a 0.45 µm nylon syringe filter. As shown in Supplementary Video S4, the filtered effluent remained opaque, similar to the original dispersion.

Fig. 3. Nanoplastic (NP) removal efficiency of alginate cryogel beads under various conditions.

a Synthesized polystyrene (PS) NPs dispersed with an anionic surfactant, assessed after waiting times of 2 min and 24 h. b Fluorescent PS NPs dispersed with a nonionic surfactant; polyethylene (PE) NPs dispersed with a nonionic surfactant; and irregularly shaped PS NPs dispersed with an anionic surfactant, all evaluated after 2 min. c Effect of varying NP concentrations (from 4.8 × 10⁸ to 1.6 × 10¹⁷ particles/ml) on (d). the mean size of residual NPs in the effluent. e Effect of pH variation (from 4 to 8) on NP removal efficiency. f Reusability of alginate cryogel beads.

In the same manner, we tested PS NPs dispersed with the non-ionic surfactant Tween 80. 100-nm fluorescent PS NPs dispersed with Tween 80 also exhibited aggregation upon the addition of alginate cryogel beads in batch conditions, with high removal efficiency over 97.6% (Fig. 3b and S8a). After the test, the alginate cryogel beads (Alg LN 1:0) were visibly coated with fluorescent PS NPs (Fig. S8b).

When the same beads were packed into a column through which the NP dispersion was passed, removal efficiency dropped drastically to just 0.22%, decreasing further to 0.11% upon reuse (Fig. S8c). In the packed column, even though the overall time the solution spends in the column is long, the actual interaction between NPs and beads is limited by how quickly the particles can move from the flowing liquid to the bead surfaces. This flow can also create shear forces that prevent nanoplastics from sticking or aggregating on or inside the beads. Thus, nanoplastics may be flushed through without sufficient time or opportunity to adsorb and cluster. This difference highlights that stable and sufficient contact between NPs and beads, easily achieved in batch mixing, is critical for effective aggregation, whereas in the flow-through column, despite a bulk contact time of over 30 min, limited mass transfer and flow dynamics lead to poor NP aggregation and removal.

Notably, the batch method was also effective for PE NPs and irregularly shaped PS NPs (Fig. 3b). In detail, the lowest removal efficiencies observed for PE NPs and irregularly shaped PS NPs were 99.5% and 98%, respectively. The irregularly shaped PS NPs, produced by abrading PS weigh dishes, had a mean diameter of ~100 nm after 5 min of abrasion (Fig. S9).

The NP removal efficiency of the cryogel beads was evaluated across a wide range of NP concentrations, from 4.8 × 10⁸ to 1.6 × 10¹⁷ particles/ml (from 2.5 × 10⁻⁴ to 1.6 × 10⁵mg/ml). The lower range was determined based on the detection limit of the NTA system. The removal efficiency remained close to 99% even as the NP concentration increased (Fig. 3c) while maintaining the mean size between 120 nm and 184 nm. At the highest tested concentration (1.6×10¹⁷ particles/ml), the average NP uptake reached ~ 3.3 × 10⁶mg per gram of alginate cryogel beads. This uptake is roughly 300 to 7000 times greater than that achieved by activated carbon^29,30. Direct comparisons with other removal technologies are not possible due to differences in mechanisms and evaluation units. Notably, this high uptake enhances its industrial applicability, with the cost of removing 1 g of NPs estimated at just $12.61 USD, while the total raw material cost per 1 g of beads is $0.75 USD (see Table S4).

Studies of pH dependency revealed that alginate cryogels (Alg LN 1:0 and Alg LN 1:1) maintained exceptional removal efficiency approaching 100% across all tested pH conditions (pH 4, 6.7, and 8). However, the performance of Alg LN 1:3 was dependent on the testing pH. It performed optimally at neutral pH (6.7), ~100%, but its efficiency decreased to approximately 55% under acidic conditions (pH 4) and to 90% in alkaline conditions (pH 8), as shown in Fig. 3e. This pH-dependent behavior likely results from changes in electrostatic interactions between the alginate network and nanoplastics across different pH levels.

In terms of reusability, the cryogels maintained consistent performance across multiple filtration cycles. Used cryogels were rinsed three times with deionized water using a Buchner funnel and 0.45 µm nylon filter membrane to remove loosely attached nanoplastics, then freeze-dried following the same procedure. Each cryogel was tested up to five times until a significant drop in removal efficiency was observed. As illustrated in Fig. 3f, AlgLN 1:0 and AlgLN 1:3 maintained nearly perfect removal efficiencies exceeding 97% through three consecutive uses. AlgLN 1:1 showed a slight efficiency decrease to 82% during the second cycle but recovered to 95% in the third cycle, demonstrating this method’s potential for repeated applications in water purification systems without significant performance degradation. After the fifth use, the removal efficiency of AlgLN 1:1 dropped significantly to 42%, at which point the reusability test was discontinued.

Zeta potential of NPs

The zeta potential (ζ) represents the electrical potential at the interface between a particle and the surrounding liquid (containing salt and hence behaving as an electrolyte) within the electric double layer, serving as a key indicator of NP dispersion stability. Dispersions with absolute zeta potential values greater than 30 – 40 mV are considered stable due to strong electrostatic repulsion. The PS NP influent containing SDS exhibited good colloidal stability, with a zeta potential of –45.87 mV (Table S5). Following aggregation, the zeta potential increased to −23.33 mV (Alg LN 1:1) to −8.44 mV (Alg LN 1:3), reflecting loss of stability. In addition, an increase in conductivity and a reduction in electrophoretic mobility implied that electrostatic screening and charge neutralization might be one of the drivers of NP aggregation. Unlike AlgLN 1:0 and AlgLN 1:1, the unusually high conductivity of AlgLN 1:3 (4.15 mS/cm) was attributed to the cryogel beads disintegrating into powder due to their weaker structural strength, suggesting that high salt concentrations should be avoided in cryogel fabrication. Although this may raise concerns about the method’s robustness, the 1:0 and 1:1 formulations remained intact with no signs of disintegration.

Mechanism of NP aggregation and removal

We analyzed all experimental results to understand how NPs were aggregated and removed from water. One possible mechanism involves the release of Ca²⁺ ions from calcium alginate cryogels. Under normal conditions, NPs remain stable in suspension due to strong electrostatic repulsion, which counteracts van der Waals forces and hydrophobic interactions (Fig. 4a). However, as Ca²⁺ ions are released from the cryogel beads, they can neutralize the negative surface charge of the NPs, thereby reducing electrostatic repulsion and promoting clustering (Fig. 4b). This is supported by the observed decrease in both zeta potential and particle mobility (Table S5). When the cryogel beads were added into the NP dispersion, the initial absence of a hydration shell around the cryogel beads surfaces, alongside with the destabilization of electrostatic forces, allows the van der Waals and hydrophobic interactions to effectively enable NP adhesion and clustering on the bead surfaces. As a result, NPs begin to form aggregates (Fig. 4c).

Fig. 4. Potential mechanisms of NP aggregation. Ca²⁺-induced aggregation.

a NPs are homogeneously dispersed in water due to the presence of a surfactant (e.g., Sodium Dodecyl Sulfate (SDS)). b The addition of alginate cryogel beads releases Ca²⁺ in water, neutralizing the solvent charge and weakening electrostatic screening, when in the presence of anionic surfactants. c As a result, the repulsive forces between NPs are reduced, leading to aggregation driven by van der Waals forces and hydrophobic interactions. Alginate adsorption-induced aggregation: d Alginate is leached from the alginate cryogel beads into the NP dispersion. e Alginate wraps around individual NPs. f Alginate forms bridging films between NPs facilitating nanoparticle aggregation. g The increase in effective density causes NP aggregates to sediment to the bottom. h Comparison of conventional water filtration methods and alginate cryogel beads-enhanced method using four indicators. The scoring criteria for the scale are elaborated in Methods, and the data extracted from the literature are summarized in Table S1.

Interestingly, although the difference in NP removal efficiency was as small as 0.07%, greater aggregation was observed with less porous alginate cryogels, suggesting that additional factors may contribute to the aggregation process. This hypothesis was further supported because the removal efficiency of pure CaCl₂ solution which had the same concentration as the CaCl₂ concentration released from alginate cryogel beads (Fig. S10) was lower than that of alginate cryogel beads (Fig. 3a). Although 6 mM CaCl₂ (released from LN 1:3 for 2 minutes) showed higher removal efficiency than 0.6 mM (released from LN 1:0 for 2 minutes) due to increased ionic strength, its performance was still lower than that achieved with alginate cryogel beads. Notably, the highest removal efficiency was observed not with Alg 1:3, but with Alg 1:0. This indicates the salt ions are not the primary driver of PS NP aggregation.

By shifting our focus to alginate, we confirmed the characteristic peaks of both PS NPs and alginate cryogels in the FTIR signal of the aggregated PS NPs (see Supporting Information SM2 and Fig. S11). This evidence narrowed down the root cause of NP aggregation to two potential mechanisms: alginate adsorption onto NPs and Ca²⁺-induced destabilization of the NPs (see Fig. 4a–g). Our proposed mechanism involves alginate chains being gradually leached from cryogel beads during rehydration (Fig. 4d), allowing them to wrap around individual NPs (Fig. 4e).

Control experiments were performed by soaking alginate beads in DI water, showing an average leached alginate concentration of 0.015 ± 0.004 mg/ml. Furthermore, adding 1 mg of sodium alginate powder to 10 mL NPs suspension, ten times the amount of the leached concentration, resulted in a low efficiency of only 1.35%. In contrast, the presence of calcium ions significantly enhanced efficiency through possible synergistic effects. At a concentration of 0.1 mg/mL alginate plus 0.6 mM CaCl₂, an efficiency of 85.87% could be achieved (see Supporting Information SM3 for experiment details).

Following the initial wrapping, we supposed the alginate-coated NPs bridge together to form aggregates (Fig. 4f). In the influent, individual NPs remain uniformly dispersed due to their near-neutral buoyancy and Brownian motion. However, once aggregated, the increased particle size leads to a higher effective density, allowing gravitational forces to overcome Brownian stabilization. This shift promotes sedimentation of the NP clusters (Fig. 4g). The growth of these hetero aggregates is eventually limited by steric hindrance.

These proposed mechanisms appear to dominate over pH-dependent surface charge variations, as aggregation was observed across a broad pH range (4.0 to 8.0). We also confirmed the negligible impact of both ionic and non-ionic surfactants, such as SDS and Tween 80, on the removal efficiency, which remained close to 99%. Different from SDS, nonionic surfactants like Tween 80 primarily stabilize nanoparticles through steric hindrance and hydration repulsions rather than electrostatic charge interactions. In this context, aggregation likely occurred when steric and hydration stabilization is compromised, typically due to hydrophobic interactions or bridging effect of the alginate. This indicates that the stabilizing effects of electrostatic and steric interactions were effectively disrupted by alginate adsorption and Ca²⁺-mediated charge neutralization and bridging. Despite the differences in the stabilization mechanisms, the aggregation process, illustrated in Fig. 4, involves overcoming repulsive forces, whether electrostatic or hydration-based steric, allowing attractive interactions, such as hydrophobic interactions or polymer bridging to drive particle aggregation.

It is worth noting that while Ca²⁺ release can induce NP aggregation, using alginate beads with lower ion release, such as Alg LN 1:0, can help maintain or slightly reduce Ca²⁺ concentrations, keeping them within acceptable limits for real-world applications. This concentration falls within the range found in distributed water ( < 112.8 mg/L) and aligns with the typical calcium levels in Canadian drinking water (1.1–112.8 mg/L), as reported in the Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Calcium.

A detailed comparison of the developed cryogel-based method with conventional WWTPs is presented in Fig. 4h. While traditional filtration methods may be economical and straightforward to implement, they remain inefficient for nanoparticle removal and face significant scalability challenges. These limitations underscore the advantages of proposed research as well as the need for more advanced solutions.

Discussion

We developed a cryogel-based method for efficient removal of surfactant-assisted dispersions of polystyrene (PS) and polyethylene (PE) nanoplastic (NP) in aqueous environments, addressing critical challenges faced by existing technologies when dealing with nanoscale plastics. The developed alginate cryogels actively drive NP aggregation without requiring surface modifications, achieving over 99% removal efficiency for both spherical and irregular shapes, with sizes ranging from 50 to 200 nm, across a range of tested conditions such as pH 4–8, ionic and non-ionic surfactants.

Our findings suggest that PS and PE NP aggregation is facilitated by alginate adsorption onto NPs and localized Ca²⁺ release. FTIR analysis confirmed alginate adsorption onto NPs, while zeta potential analysis identified Ca²⁺-mediated charge destabilization as a crucial mechanism overriding surfactant stabilization. Additionally, SEM imaging demonstrated that NPs predominantly adsorb onto the cryogel surface, with minimal entrapment in its internal pores.

Unlike conventional methods such as sand filtration and coagulation, which often require longer processing times and struggle with the effective capture of nanoscale plastics, this cryogel-based system demonstrates rapid and highly efficient removal capabilities. Our cryogel-based method combines short operational times, broad applicability, and durable performance across multiple use cycles, making it a strong candidate for real-world applications.

Beyond its performance, the cryogel’s cost-effectiveness—derived from readily available sodium alginate—positions it as a sustainable alternative to expensive materials like graphene oxide or polydopamine-based hydrogels. The scalability of this process, coupled with low material costs ($0.75/g beads) and high contaminant uptake ( ~ 3.3 × 10⁶mg/g), positions it as a promising solution for water treatment plants and ecosystems impacted by NP pollution.

These findings demonstrate exceptional removal performance for PS and PE NPs under controlled laboratory conditions. To advance this technology toward real-world application, future studies should assess its performance in diverse environmental conditions, including wastewater and natural water bodies, to account for more complex water chemistries. In addition, evaluating its long-term stability in continuous flow systems will be important for optimizing industrial scalability, particularly for wastewater treatment plants.

Materials and methods

Preparation of alginate beads

All chemicals were used as received from the manufacturer. One hydrogel and six different types of cryogel beads were prepared (see Table 1). We used a low-viscosity alginate (alginic acid sodium salt from brown algae, MW = 120,000–190,000 g/mol, Sigma Aldrich) due to its biodegradability and low cost. The alginate solution was prepared by mechanically mixing alginate powder in 5.75 wt.% DI water overnight until fully homogenized (Fig. 1a). Alginate, a naturally occurring biopolymer, features a linear structure of alternating mannuronic acid (M-block) and guluronic acid (G-block) units, with the G-block regions forming stable ionic crosslinks in the presence of divalent cations like calcium (Fig. 1b). To enhance porosity, NaCl was added to the alginate solution at alginate-to-NaCl weight ratios of 1:1 and 1:3.

Table 1.

Different types of alginate beads for removal of PS NPs (○ and × indicate whether the material or process is used (○) or not used (×), respectively)

	Bare alginate hydrogel	Alg LN 1:0	Alg xLN 1:0	Alg LN 1:1	Alg xLN1:1	Alg LN 1:3	Alg xLN 1:3
Freeze Dry	×	○	○	○	○	○	○
Liquid nitrogen	×	○	×	○	×	○	×
NaCl (ratio)	×	× (1:0)	× (1:0)	○ (1:1)	○ (1:1)	○ (1:3)	○ (1:3)

Once each alginate solution was fully homogenized, pregel solutions were added dropwise at a rate of 2.5 mL/min using a syringe pump (Nexus Fusion 4000, Chemyx) and a syringe tip with a 200 µm diameter into a 0.6 M CaCl_₂ solution for alginate with a salt ratio of 1:3 and a 0.3 M CaCl_₂ solution for the others, which was stirred at 700 rpm with a magnetic stir bar. The resulting hydrogel beads were polymerized in the solution overnight and then washed. The DI water used for washing was replaced four times during continuous agitation on an orbital shaker (OS-40PRO, LABFISH) at 200 rpm for 4 days. The conductivity change was examined to make sure of the removal of salt ions from alginate hydrogels (Fig. S12). The hydrogel beads had radii ranging from 1.3 mm to 1.5 mm (Fig. S1a).

Half of the polymerized hydrogel beads were soaked in liquid nitrogen and the other half was left in the freezer overnight until frozen. Liquid nitrogen mitigated the abrupt size reduction of hydrogel beads during the freeze-drying process. Liquid nitrogen and non-liquid nitrogen alginate beads were freeze-dried for 24 hours using a Home Pro Freeze Dryer (Harvest Right) with primary drying at -35°C under a pressure of 0.05 Pascal, followed by secondary drying with a gradual temperature increase to 25°C at constant pressure. The synthesized cryogels were stored in a refrigerator at 4°C until testing. Cryogels exhibited smaller radii compared to hydrogels, ranging from 0.2 mm to 0.5 mm (Fig. 1c, d).

Characterization of cryogels

We used Scanning Electron Microscopy (SEM)-Energy Dispersive X-ray (EDX) (FEI Quanta Feg-250 ESEM) and SEM (Zeiss Merlin High-resolution) to examine the surface of hydrogel and cryogel beads, respectively.

The specific surface areas (S_BET) of the different cryogels were determined from nitrogen sorption isotherms using a Micromeritics Gemini 2390a analyzer and the Brunauer−Emmett−Teller (BET) method. Approximately 0.5 g of each sample was deposited in BET containers and degassed under vacuum at 80 °C overnight, followed by measurements in a dewar full of liquid nitrogen.

For the range of linearity of the BET plot, P/P_o range used to calculate BET surface area was 0.05–0.15, adjusted to achieved a positive BET C-value. Additionally, the point where P/P_o ~ 0.99 was used to estimate the total pore volume as the sample was near the saturation condition.

The Barrett-Joyner-Halenda (BJH) analysis was conducted to study the pore size distribution of the alginate cryogels. For the thickness equation, the Broekhoff-de Boer model was used, which considers that the thickness of the adsorbed film is related to the total chemical potential of the adsorbed layer rather than assuming the thickness to be solely dependent on relative pressure.

Chemical groups of cryogel beads and aggregated NPs were examined using Attenuated Total Reflectance—Fourier Transform Infrared spectroscopy (ATR - FTIR) (Thermo Scientific™ Nicolet™ iN10 Infrared Microscope) in liquid nitrogen cooled mode. A built-in pressure gauge ensured a consistent applied pressure of 60 psi during measurements. Spectra were collected for both the background and the cryogel beads at a resolution of 2 cm⁻¹ with 64 scans, covering the range of 600–4000 cm⁻¹. After each measurement, the ATR diamond crystal was cleaned with pure methanol. Each sample’s spectra were collected in triplicate, and baseline corrections were performed using OPUS 7.0 software.

The swelling behavior of various alginate cryogel beads was evaluated by submerging them in DI water and measuring their weight changes over time. 0.5 g of cryogel beads were placed in a net with a pore size of tens of microns and submerged in DI water. The weight of the swollen beads was measured over time by removing the excess water retained in the net until the weight reached equilibrium.

Preparation and characterization of NPs

Four types of NPs were tested: (1) 100-nm-diameter fluorescent latex polystyrene (PS) NPs (Abvigen, density = 1030 kg/m³), dispersed with Tween 80, (2) 100-nm-diameter fluorescent polyethylene (PE) NPs (Abvigen, density = 960 kg/m³), dispersed with Tween 80, (3) synthesized PS NPs with diameters ranging from 50 to 200 nm, dispersed with SDS, and (4) irregularly shaped PS NPs with an average diameter of 150 nm, dispersed with SDS. Specific preparation methods are given in the Supporting Information (SM1).

The size and distribution of NPs in solutions, including influent and effluent filtration samples, were characterized using Nanoparticle Tracking Analyzer (NTA) (NanoSight LM300, Malvern Panalytical). NTA is capable of detecting nanoparticles from 10 nm to 1000 nm. Samples were diluted with DI water as needed to achieve the recommended particle concentration (10⁶–10⁹ particles/ml). A laser with a wavelength of 532 nm was used, and five 60-second recordings were taken for each sample. Measurement parameters were set to a camera level of 13, a detection threshold of 5, room temperature between 22 °C and 24 °C, and a screen gain of 10. Each condition was repeatedly measured 5 times.

Zeta (ζ) potential of NPs in solvent was measured using a Malvern Zetasizer Nano ZS 90 with a He−Ne laser (633 nm) at 90° optics at 23 °C. Each NP dispersion without alginate cryogel beads was loaded into a folded Capillary Zeta Cell (DTS1070, Malvern Instruments).

Individual NPs were observed using Transmission Electron Microscopy (TEM) (Libra 200MC TEM, Zeiss). The NPs dispersion was diluted to 0.002 ppm and then 5–10 µl of the diluted dispersion were deposited on a TEM grid of 300-mesh Cu (Ted Pella Inc.) at 23 °C. Excess solution was sufficiently dried in a vacuum chamber for 24 h before imaging.

The SDS concentration in the NP dispersion was determined by measuring the solution’s conductivity using a conductivity meter (SevenDirect SD2023, Mettler Toledo). A calibration curve was established by measuring the conductivity of solutions with known SDS concentrations to obtain a linear relationship.

NP removal test

To test NP removal efficiency, 1 mL of NP dispersion (3.1 × 10¹² particles/ml, unless otherwise specified) was added to 0.05 g of alginate beads. After waiting for 2 minutes, the NP dispersion (with aggregated NPs) was collected using a syringe and filtered through a 0.45-µm-pore syringe filter. To ensure only the NP dispersion was collected without alginate beads, an 18 G syringe needle was used. Pressure was applied manually without further control. Considering the applications in WWTPs and practical cases, we also tested alginate cryogel beads’ column. A column with a 10.5 mm-internal diameter was packed with 1 g of alginate cryogel beads of various types, occupying ~ 10 mL of volume. Afterwards, an NP dispersion solution was passed through the column at a flow rate of ~ 0.13 mL/min. Collected dispersions were filtered using a 0.45 µm nylon syringe filter.

The influent of synthesized PS NPs was prepared at pH 6.7, close to neutral. To investigate the effect of pH on NP removal, PS NP dispersions were prepared at pH 4 and pH 8 using NaOH and HCl.

To investigate the role of salt ions on NP aggregation, a CaCl_₂ solution with the same Ca²⁺ molar concentration (0.6–6.0 mM) as the one in DI water, containing alginate cryogel beads for 2 min, was tested. After adding 1 mL of the CaCl_₂ solution, we waited for 2 min, before collecting the NP dispersion. The collected solution was filtered in the same manner.

NP concentration (particles/ml) of each filtered solution was measured using NTA. NP removal efficiency was calculated for qualitative comparison among different types of alginate beads and varying concentrations of NPs:

$$\mathrm{Removal\; efficiency}\left(\%\right)=\frac{C_i-C_t}{C_i}\times 100\%$$ 1

where C_i and C_t are the filtration influent concentration (particles/ml) and effluent concentration (particles/ml), respectively. Each cryogel’s NP removal capacity was examined with varying concentrations of NPs from 4.8×10⁸ – 1.6×10¹⁷ particles/ml and calculated with the following equation:

$$\mathrm{Capacity}\left(\mathrm{the}\mathrm{number}\mathrm{of}\mathrm{NP}\mathrm{particles}/1\mathrm{g}\mathrm{of}\mathrm{cryogel}\mathrm{beads}\right)=\frac{C_i-C_t}{m}$$ 2

where m is the mass of beads (g).

The effect of residual salt on NP removal

Inductively coupled plasma optical emission spectrometry (ICP-OES) was used to detect and quantify the release of calcium from different alginate cryogel beads. Measurements were performed on a Teledyne Leeman Labs Prodigy ICP-OES spectrometer with radial plasma observation, using yttrium (Y) as an internal standard. Calibration was conducted using six standards with known concentrations of 0.01, 0.05, 0.1, 0.5, 1, 5, and 10 mg L⁻¹. For sample preparation, 0.1 g of each type of alginate beads was soaked in 5 mL of water, and the supernatant was collected after 2 min and 24 h of contact. The remaining beads were subjected to constant agitation of 200 rpm using shaker table, and a second supernatant was collected after 24 hours. All samples were filtered through a 0.45 µm nylon syringe filter before analysis.

Evaluation comparison

The performance of conventional methods, including traditional coagulation, activated carbon, and rapid sand filtration, was systematically evaluated alongside our cryogel-based method. The evaluation was divided across four key criteria, where each criterion was scored on a scale from 1 to 5, with 5 representing the highest performance and 1 the lowest. The scoring factors were defined as follows: (1) Efficiency—the ability to filter nanoparticles effectively; (2) Environmental Sustainability—the eco-friendliness of raw and filter materials, along with the reusability of the filter; (3) Versatility—the adaptability of the material to various conditions; and (4) Process Simplicity—the ease of material production and application. This evaluation was conducted solely for comparative purposes, utilizing data extracted from the literature summarized in Table S1.

Statistics

Statistical methods were not applied to define the sample size, and all data were included in the analysis. The experiments were conducted with randomization. The study included two replicates for swelling measurements, and four and three runs for NTA and ICP measurements, respectively, unless otherwise noted. Zeta potential measurements were repeated six times. All error bars show standard error of the mean (s.e.m).

Author contributions

A.R.K. and A.B.R. equally designed and performed the experiments. A.R.K. and A.B.R. wrote the manuscript with the assistance of S.K.M. and B.Z. H.H., H.Z., and J.L. collaborated with experiment guidance and financial support. J.W. and E.A.H. assisted with NTA measurements. All authors have read and approved the final manuscript.

Peer review

Peer review information

Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work.

Data availability

The source data for all figures are provided in the supplementary materials as an Excel file (“Supplementary Data”). Additional supporting information, including supplementary methods, images, tables, and videos, is available as a PDF (“Supporting Information”). All data supporting the findings of this study are included within the paper and its supplementary information files.

Competing interests

B.Z., S.K.M., A.-R.K., and A.B.M. have filed a U.S. provisional patent application titled ‘Methods for Effective Aggregation of Nanoplastics and Water Remediation Using Aerogel-Enhanced Filtration.’ The remaining authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s42004-025-01832-4.