This study investigated microplastic remobilization from porous media under wetting-drying cycles by subjecting saturated columns to repeated cycles and monitoring microplastic release. Columns were packed using the wet method to ensure saturation and homogeneity. Degassed ultrapure water was pumped upwards from the column bottom at 9 mL/min, maintaining a 5-cm hydraulic head above the sand surface throughout packing to ensure complete saturation. Packing paused at a height of 15 cm to install pre-calibrated time domain reflectometry (TDR) probes and tensiometers into access ports; these were secured with silicone sealant. Packing then resumed in uniform, saturated layers to the target height while continuously infusing degassed ultrapure water. Following the methodology of Li et al. (2013), both TDR probes and tensiometers were calibrated for each experimental column prior to testing to ensure accurate real-time monitoring of water saturation and capillary pressure.

Following packing, columns were stabilized for at least 24 h. TDR probes and tensiometers monitored water saturation and capillary pressure, confirming stability when readings exhibited no significant fluctuations. Prior to microplastic injection, degassed ultrapure water was infused upwards at 9 mL/min to remove sand debris until effluent showed no visible impurities. The water was then replaced with background electrolyte solution (10 mM NaCl), which was infused at the same flow rate for 10 pore volumes (PV), calculated as described in Equation 1 and Equation 2, until effluent absorbance matched the input solution, establishing consistent physicochemical conditions.

PS microplastic suspensions (100 mg/L in 10 mM NaCl) representing different aging durations (0 days, 32 days) were subsequently introduced into the column at a rate equivalent to 2 PV. Continuous sonication maintained homogeneous suspension concentration during injection. Immediately following microplastic injection, background solution (without microplastics) was flushed through the column until effluent microplastic concentrations became undetectable. Effluent absorbance, measured continuously by UV-Vis spectrophotometry during injection and flushing, enabled determination of microplastic concentrations and calculation of the mass retained in the porous media during the saturated phase.

After achieving stable microplastic retention, the initial drying phase commenced by draining the background solution from the column bottom at 9 mL/min. Drying concluded when effluent flow ceased and water saturation stabilized (Xie et al., 2012). Effluent microplastic concentration was continuously monitored via UV-Vis spectrophotometry during drying, enabling calculation of remobilized mass combined with flow rate data. Immediately following this initial drying, subsequent wetting-drying cycles were performed to simulate hydrological fluctuations remobilizing retained microplastics. Each cycle comprised a wetting phase and a drying phase. During wetting, background electrolyte solution (without microplastics) was slowly pumped into the column at 9 mL/min until the liquid level reached the overflow port at the column top, simulating moisture recharge and enabling remobilization upon contact with the moving aqueous phase.

A stabilization period followed each wetting phase to allow moisture redistribution before initiating the drying phase. Drying proceeded identically to the initial phase, with continuous effluent microplastic concentration monitoring enabling calculation of remobilized mass per cycle. Following the initial drying, two complete additional wetting-drying cycles were performed. Effluent microplastic concentration and remobilized mass were continuously monitored and calculated during all drying phases (including the initial and subsequent cycles). For each experimental batch, blank control tests were conducted. This involved preparing control columns packed solely with sea sand and leaching them with background electrolyte solutions free of microplastics under identical experimental conditions. The effluent from these blank columns was collected and analyzed using ultraviolet-visible (UV-Vis) spectrophotometry. No microplastics were detected in any blank samples, confirming that our experimental setup and operational procedures introduced no contamination.

To investigate microplastic transport under complex hydrodynamic conditions representative of realistic unsaturated porous media, including continuous inputs, this study designed cumulative retention experiments driven by wetting-drying cycles, systematically examining the effects of aging and material type on long-term retention and spatial distribution. Columns were packed, stabilized, and flushed with degassed ultrapure water at a constant flow rate (9 mL/min) to remove fine particles or soluble impurities, preventing interference with microplastic quantification. Subsequently, the columns were equilibrated with at least 10 pore volumes (PV) of background electrolyte solution (10 mM NaCl) to achieve hydraulic equilibrium. Following equilibration, an initial drainage phase established uniform unsaturated initial conditions: solution was drained from the column bottom at 9 mL/min until residual water saturation was reached (effluent flow ceased) for all columns.

Following establishment of the initial unsaturated state, wetting-drying cycles with continuous microplastic input commenced. Each cycle consisted of two phases: (1) Wetting phase: Polystyrene (PS) microplastic suspensions (100 mg/L in 10 mM NaCl; aged 0 days or 32 days) were continuously injected upward through the column bottom at 9 mL/min until the liquid level reached the overflow port, achieving near-saturation without ponding. (2) Drying phase: Solution was drained downward at 9 mL/min. Effluent containing remobilized microplastics was collected, returned to the original suspension reservoir, and homogenized via sonication to maintain constant influent concentration, simulating persistent pollution. Drainage continued until effluent flow ceased. This sequence constituted one cycle. Consecutive repetitions yielded cumulative retention data after 1, 2, and 5 uninterrupted cycles.

Post-experiment, the sand column was sectioned into nine 4-cm segments. Each segment was transferred to a clean 500-mL beaker, oven-dried at 70°C, and subjected to microplastic extraction via density flotation. A total of 150 mL of saturated NaCl solution (density 1.20 g/mL) was added to each dried sample, followed by 20-min sonication to release microplastics. The supernatant containing microplastics was transferred to 50-mL centrifuge tubes and centrifuged at 4,500 rpm for 10 min. The final supernatant was diluted to volume in 100-mL colorimetric tubes, and microplastic mass per layer was quantified colorimetrically using a standard curve. Following the experiment, triplicate measurements were performed for each subsample obtained through stratified sampling to quantify retained microplastics in the sand profile, ensuring the reliability of quantitative results.

The efficiency of MP remobilization during the drying phases was assessed using the Relative Release Recovery (%), a metric adapted from Bradford and Torkzaban (2008) and defined as: (Mass of MPs released in a drying phase/Mass of MPs retained in the column at the start of that cycle) × 100.

To evaluate the long-term fate of MPs under continuous input and cyclic wetting-drying, the Cumulative Retention (%) was calculated after 1, 2, and 5 cycles. This metric represents the overall retention efficiency of the porous medium and is defined as: (Total MP mass retained in the column/Total MP mass injected) × 100. Furthermore, to assess the mass balance and potential particle loss during the post-experiment extraction process, the Packing Material Recovery (%) was determined by flotation extraction using the following formula: (Total MP mass recovered/Total MP mass injected) × 100. The loss rate (%) was derived from the difference: Cumulative retention recovery - Packing material recovery.

The fundamental physical properties of the packed column, which influence hydraulic and transport behavior, were determined as follows. Porosity (φ), the ratio of void volume to total volume, was calculated using measured particle density (ρ_p = 2.62 g/cm³) and bulk density (ρ_b = 1.64 g/cm³): φ = 1 − ρ b   /   ρ p = 0.374 (1)

Based on the calculated porosity and the column dimensions (radius r, length L), the Pore Volume (PV, mL), a critical parameter for standardizing the experimental flow, was calculated as: PV = π r 2 L × φ (2) where r is column radius (cm) and L is length (cm). Calculated PV = 550 mL.

Cumulative microplastic retention in unsaturated porous media is governed by complex interfacial processes and hydrological dynamics. To assess the long-term impact of aging on microplastic mobility under realistic unsaturated conditions, wetting-drying cycles were employed to examine remobilization mechanisms under complex hydrodynamics. Real-time monitoring of water saturation (Sw) and capillary pressure (Pc) dynamics enabled in-depth analysis of hydrodynamic characteristics during these cycles, revealing Sw/Pc evolution patterns and their influence on air-water interface (AWI) behavior.

Following saturated-phase retention, the initial drainage phase commenced. Figure 8a,b depict Sw and Pc evolution during initial drainage for pristine PS and aged PS. As drainage progressed, Sw decreased continuously from near-saturation (∼1.0) while Pc increased, tracing the soil water characteristic curve. Water initially drains from larger pores with relatively flat AWIs. As drainage advances, water retreats into smaller pores where increased AWI curvature necessitates greater Pc. Rates of Sw and Pc change decelerate until stabilization, with rapid initial shifts indicating swift AWI formation and water film entrapment.

Figure 9a,b show capillary pressure (Pc) versus water saturation (Sw) curves during initial drainage for pristine PS and aged PS. During drying (drainage), the advancing air-water interface (AWI) generated significant detachment forces, acting as the key dynamic remobilization mechanism for retained microplastics, particularly hydrophobic ones. Post-drying, capillary forces retained residual water as relatively stable water films or pendular rings (Li et al., 2013), creating potential static retention sites via associated AWIs.

During subsequent wetting phases (Figure 10), Pc and Sw transitioned rapidly before stabilizing. The AWI receded and destabilized, accompanied by water film thickening. While Sw/Pc trends remained largely consistent across cycles, subsequent wetting phases never re-attained complete saturation (Sw = 1.0) due to incomplete air displacement and persistent gas entrapment within pores after multiple cycles (Li et al., 2014), generating new AWIs and water films.

Consequently, AWIs exhibit dual functionality: enabling dynamic remobilization during phase transitions (drainage) and providing static retention potential via stabilized interfaces. This interplay introduces significant complexity to microplastic retention and release dynamics.

Figure 11 shows capillary pressure (Pc) versus water saturation (Sw) curves during 1, 2, and 5 wetting-drying cycles. Similar Pc-Sw trajectories across cycles indicate that repeated cycling did not significantly alter the porous media’s pore size distribution. Distinct hysteresis loops formed between drainage and wetting curves (Zhou et al., 2014), arising from pore geometry, contact angle hysteresis at the three-phase contact line, and air entrapment.

At identical Pc values, Sw was lower during drainage than wetting. This phenomenon arises because drainage exhibits more extensive AWIs and thinner water films, providing greater potential for static retention via capillary pinning at the solid-water interface (SWI). Conversely, higher Sw during wetting corresponds to reduced AWI area and thicker water films. This hysteresis demonstrates differing retention/release efficiencies between phases: drainage governs microplastic mobility through competition between enhanced static retention potential and strong dynamic detachment forces, while wetting phases exhibit comparatively weaker interfacial retention. The ultimate fate of microplastics depends on their intrinsic properties interacting with these dynamic environmental factors.

Figure 12 demonstrates that microplastic release ratios (C/C₀) increase as water saturation (Sw) decreases below ∼0.6 during unsaturated drainage. This occurs because the advancing air-water interface (AWI) thins water films, enabling capillary menisci to exert interfacial detachment forces on attached microplastics along the air-water-solid contact line. Release initiates when these detachment forces overcome microplastic adhesion to the solid surface (Sharma et al., 2008). The release threshold at Sw ≈ 0.6 reflects the critical saturation where water films achieve sufficient thinness for the moving AWI to exert significant detachment forces at the solid-water interface (SWI) (Zhang et al., 2012).

This phenomenon parallels nanomaterial transport studies, where E. coli mobilization similarly peaks below Sw = 0.6 as thinning films detach cells from the SWI and transfer them to mobile phases (bulk water or AWI) (Bradford and Torkzaban, 2013). This confirms the AWI’s critical role in colloid mobilization, providing a mechanistic basis for analyzing aging’s impact on microplastic mobility. Crucially, aged PS exhibited lower C/C₀ than pristine PS at identical Sw during initial drainage, indicating enhanced detachment resistance.

Figure 13 further shows reduced release recovery for aged PS during initial drainage (12.05% → 7.96%), confirming suppressed remobilization. Although aging enhanced hydrophilicity and electronegativity—potentially weakening adhesion via electrostatic repulsion—unsaturated drainage is dominated by AWI detachment forces. These forces are critically dependent on contact angle: lower contact angles (higher hydrophilicity) reduce capillary pinning at the air-water-solid (AWS) interface, thereby weakening detachment forces (Scheludko et al., 1976). Given that AWI detachment forces exceed electrostatic and hydrodynamic forces by orders of magnitude (Hou et al., 2020), hydrophilicity becomes the governing factor, overriding any mobility promotion from increased electronegativity.

Additionally, aging-induced surface cracks (Figure 5) increase roughness, which enhances adhesion energy and impedes AWI movement across the microplastic surface (Singh and Joseph, 2005). Collectively, hydrophilicity-driven reduction in detachment forces and roughness-enhanced retention significantly reduce aged microplastic mobility in unsaturated porous media.

To assess the remobilization potential of retained microplastics, wetting-drying cycles were implemented following initial drainage. Unlike initial drainage, secondary and tertiary drainage phases were preceded by wetting phases (Figure 14). During wetting, water displaced pore air, increasing water saturation (Sw) while decreasing capillary pressure (Pc), with water preferentially invading smaller pores under higher Pc.

Throughout cycling, the Pc-Sw relationship exhibited consistent hysteresis (Figure 15), demonstrating path dependence between wetting and drying phases (Topp et al., 1967). Each drainage phase reduced Sw below the critical threshold established during initial drainage, indicating that re-established mobile air-water interfaces (AWIs) regained potential to remobilize microplastics near the solid-water interface (SWI). Crucially, the wetting phase reintroduced water, disrupting pre-existing AWIs and water films, thereby remobilizing microplastics attached to these interfaces. The distinct hydrodynamic conditions between phases consequently led to phase-dependent remobilization behaviors.

Release recovery rates progressively decreased with increasing cycle number (Figure 13). This decline was more pronounced for aged microplastics: aged PS recovery decreased from 7.96% (initial drainage) to 4.43% (secondary drainage), becoming undetectable during tertiary drainage. Despite similar hydraulic conditions conducive to release in subsequent drainage phases, observed release behavior differed significantly. Both pristine and aged PS exhibited substantially lower recovery in later cycles, indicating markedly reduced mobility in unsaturated porous media after multiple cycles. Residual microplastics demonstrated long-term retention characteristics. This persistent low recovery occurred because each release event selectively mobilized weakly adhering particles (Bradford and Torkzaban, 2013). Consequently, residual microplastics after multiple cycles represent a strongly adhering, less mobile fraction with higher retention stability (Chen et al., 2015).

Aging critically altered microplastic surface properties: enhanced hydrophilicity substantially weakened AWI detachment forces. Additionally, under low saturation conditions, pore water existed as isolated films or pendular rings, trapping microplastics and impeding AWI-mediated release (Feng et al., 2024). These factors collectively rendered aged microplastics highly resistant to release after multiple cycles.

Figure 16 shows cumulative retention rates for pristine and aged microplastics after 1, 2, and 5 wetting-drying cycles. Aged microplastics consistently exhibited higher retention than their pristine counterparts at each cycle stage, demonstrating that aging promotes long-term retention under repeated cycling.

Although aging enhanced microplastic electronegativity—increasing electrostatic repulsion with porous media—the dominant unsaturated removal mechanism, detachment at the air-water interface (AWI) during drainage, was significantly weakened by aging-induced hydrophilicity. Concurrently, the hysteresis-driven drying path created thinner water films and potentially stronger capillary pinning forces, enhancing static retention potential. Consequently, aged microplastics showed substantially lower removal efficiency than pristine particles during each drying phase, characterized by specific moisture distribution under hysteresis effects.

Following 32-day aging, all microplastics exhibited increased surface roughness. When the colloid-to-grain size ratio (dp/d50) exceeds ∼0.5–1%, physical retention mechanisms (pore straining, mechanical interlocking) become significant alongside chemical attachment (Bradford et al., 2007). In unsaturated media, air occupying pore space narrows effective pathways, likely enhancing physical retention of aged microplastics. Repeated injections during cycles further increased retention sites and physical straining efficiency.

Given the ∼1% size ratio in this system and the complex geometry of unsaturated pores, physical retention emerged as a key mechanism. Increased roughness promotes aged microplastic retention during wetting-drying cycles (Fei et al., 2022), making them more susceptible to entrapment during flow perturbations.

Although aging-enhanced electrostatic repulsion initially inhibited chemical attachment during wetting, this effect was ultimately overridden by drastically reduced removal efficiency during the drainage phase over long-term cycling. Consequently, under the dominance of physical retention and diminished removal efficiency, cumulative retention of aged microplastics exceeded that of pristine particles.

After 5 cycles, retention rates for pristine and aged PS were 11.05% vs 9.38% (bottom) and 3.76% vs 5.48% (surface). Aging produced a more uniform vertical distribution due to enhanced hydrophilicity and electronegativity, which stabilized dispersion and prevented localized accumulation at the inlet or interfaces.

Physical retention occurred throughout the migration path as pore-straining structures (dp/d50 > 0.5–1%) were widely distributed (Chen et al., 2008), resulting in random-depth retention of aged microplastics via physical mechanisms. While pristine microplastics underwent AWI-driven re-concentration, aged particles demonstrated weaker attachment, enabling short-distance detachment/re-attachment during flow and exhibiting limited AWI-driven transport (Wu et al., 2019). Combined with repeated flushing and retention, this yielded a more homogeneous spatial distribution of aged microplastics. Aging thus simultaneously enhanced cumulative retention—through reduced removal and increased physical trapping—and distribution uniformity via weak attachment and low remobilization.

As presented in Table 1, the cumulative retention rates of pristine PS microplastics increased progressively to 22.94%, 38.97%, and 72.83% after 1, 2, and 5 wetting-drying cycles, respectively. Aged PS exhibited even higher retention, reaching 76.88% after 5 cycles. Retention significantly increased with cycle number for both microplastic types, although the accumulation rate gradually declined. This demonstrates that repeated cycling enhanced microplastic retention within the porous media by progressively strengthening retention mechanisms while weakening removal processes.

For all microplastics, particles exhibiting weaker adhesion—those more readily mobilized by air-water interfaces (AWIs)—were likely removed during initial cycles, leaving behind a population of increasingly stable particles. Furthermore, repeated microplastic injections during wetting phases continuously introduced new particles to occupy available retention sites. Physical straining mechanisms also intensified over successive cycles: newly injected particles became strained at pore throats or adhered to rough surfaces, while previously retained particles formed “filter layers” that captured incoming particles. These cumulative effects amplified retention with increasing cycle number.

Figure 17 illustrates the vertical distribution of pristine and aged microplastics after 1, 2, and 5 cycles. Retention was consistently higher in the bottom layer (36–40.5 cm depth: pristine PS 6.20%–11.05%; aged PS 6.60%–9.38%) compared to the surface layer (0–4.5 cm: pristine PS 0.75%–3.76%; aged PS 0.75%–5.48%). However, distributions became increasingly uniform with progressive cycling.

In porous media where pore throat diameters exceed microplastic particle sizes, particles can theoretically migrate upward (Gibson et al., 2009). The observed elevated bottom retention likely resulted from the formation of “filter layers” near the inlet, where embedded particles captured subsequently arriving microplastics.

During cycling, mobile AWIs remobilized retained particles, including those within filter layers. These remobilized particles were transported upward with pore water flow and subsequently re-retained at various depths due to hydrodynamic shifts or interface collapse. This cyclic capture-transport-re-retention process continuously redistributed particles vertically.

Concurrently, flow perturbations enabled physical straining throughout the column. Under the combined influence of AWI remobilization, flow-driven retention, and inherent transport properties, repeated wetting-drying cycles progressively homogenized the spatial distribution of microplastics.