Natural aquatic ecosystems, including oceans, rivers, lakes, and groundwaters, are critical components for sustaining global ecological balance and biological processes. In these systems, particulate materials, ranging from macroscopic sediments to nanoscale particles, serves as critical carriers for material cycling and energy transfer, greatly affecting pollutant migration and geochemical cycles. Among them, natural colloids (1–1000 nm) and nanoparticles (1–100 nm) have become major research themes in environmental science and geochemistry due to their unique physico-chemical properties and ecological impacts (Lead and Wilkinson, 2006b, LeadWilkinson, 2006a; Wigginton et al., 2007).

While studies on colloidal behavior are extensive, recent focus has shifted towards nanoparticles. These particles represent a size-defined subset of colloids with unique environmental behaviors which are driven by properties such as ultrahigh specific surface area, quantum size effects, and enhanced reactivity, distinguishing them from larger colloids (Tsao et al., 2011; Delay and Frimmel, 2012; Flores et al., 2025). It is also found that, the environmental fates of these particles are profoundly affected by their surrounding conditions. Key environmental factors, such as ionic strength, the composition of natural organic matter, and hydrodynamic conditions, differ substantially between freshwater and marine systems, leading to different particle behaviors (Zhuang and Yang, 2017). In this review particle dynamics will be discussed in different environments, to provide a more comprehensive understanding of the behaviors of these materials.

The structure of this review is as follows: 1), the classification, sources, and fundamental characteristics of aquatic particles are stated; 2), the analytical methods for particle separation and characterization are evaluated, focusing on their limitations and analytical artifacts; 3), the environmental behaviors and geochemical effects of the particles, especially nanoparticles, are summarized;4), through a case study on nanoparticles from deep-sea hydrothermal vents, an example of their long-range transport mechanisms and influence over global-scale elemental cycles is given.

In aquatic environments, dissolved substances are generally with the sizes of < 1 nm, occurring as ionic or molecular species (e.g., simple ions such as Na⁺, Ca²⁺, Cu²⁺ and small organic molecules) (Jacques, 1992), which are classified as the dissolved fraction. According to the definition of IUPAC, colloids are typically referred to the materials with at least one dimension between 1 nm and 1 µm, while particles are those larger than 1 µm (Lead and Wilkinson, 2006b, LeadWilkinson, 2006a).

In actual studies, these classifications are “operationally defined,” based on the pore size of the filter used for separation. Material passing through such filter is labeled “dissolved”, a fraction that includes both the truly dissolved species and colloids while material retained on the filter is termed as “particulate.” However, definitions are inconsistently applied by researchers both between and within various fields. For example, water engineers often use membrane filters with nominal pore sizes of around 5–10 µm, while aquatic chemists more frequently use those with pore sizes of 0.2 or 0.45 µm for their work. As a result, the definition of colloid can vary across studies (Wells, 2002; Aiken et al., 2011).

Even without an official universal definition, a useful conceptual framework distinguishes the dominant processes governing each phase. The mobility of the truly dissolved phase is largely governed by molecular diffusion. For the colloidal phase, a combination of diffusion and aggregation dictates their stability and transport. Conversely, the fate of the particulate phase is primarily controlled by gravitational settling (sedimentation). However, it is critical to recognize that this is a simplified model. In many dynamic natural systems, such as rivers or coastal zones, other processes can be equally or even more significant. These processes include advection (bulk fluid motion), which influences all phases; heteroaggregation, which is crucial for scavenging processes; and the resuspension of settled particles from the sediment bed. In this review, the size-based classification is adopted as outlined above, after critically examining the distinct characteristics of each category in freshwater versus marine environments. The size ranges and chemical types of the key species are illustrated in Figure 1.

For natural colloids, including nanoscale materials, analytical workflows revolve around two core stages: physical separation and subsequent characterization. While most materials science characterization methods can be adapted to environmental samples, below we focus on key separation techniques, summarize core characterization tools, and highlight matrix-specific challenges in freshwater and marine systems.

Core separation methods are led by two widely used techniques. The first techniques are filtration and ultrafiltration, which are widely used for freshwater samples, with ultrafiltration (UF) membranes (1 kDa–10 kDa) separating colloids and nanoparticles. Cross-flow ultrafiltration (CFF) is preferred for marine samples, as it could handle high-salt conditions and large volumes without colloidal alteration. However, marine samples are prone to membrane fouling by salt crystals and organic gels, requiring pre-rinsing with low-salinity buffers (Wang and Tarabara, 2008; Xiang et al., 2022). The second is Field-flow fractionation (FFF, especially Flow FFF) which is ideal for freshwater colloids and nanoparticles, enabling high-resolution size separation (Baalousha et al., 2011). When coupled with online detectors like light scattering and ICP-MS, FlFFF can provide multi-dimensional information on particle properties (e.g., size, composition). Subsequent off-line characterization of collected fractions by techniques such as electron microscopy can further reveal details about particle morphology and structure (Lyvén et al., 2003; Plathe et al., 2013). However, it requires specialized instrumentation and expertise, has relatively low throughput with resolving power compromised by complex matrices (e.g., high salinity) (Vaillancourt and Balch, 2000). It is particularly challenging to match the FlFFF carrier to the sample when analyzing seawater—given its high ionic strength (up to 0.7 M) and elevated concentrations of Ca²⁺ and Mg²⁺. Notably, seawater cannot serve as the carrier in FlFFF-ICP-MS hyphenation, as even low levels of salt can drastically impair the ICP-MS performance. Additionally, high salinity triggers particle aggregation, leads to much lower sample recovery, and increases membrane adsorption and permeation issues that affect analytical accuracy. Apart from the aforementioned two separation approaches, complementary separation tools include centrifugation/ultracentrifugation (Oshima et al., 2022) and size-exclusion chromatography (Nordin et al., 2015).

For the purpose of characterization, most materials science methods can be adapted for environmental samples. Below, we summarize several key characterization techniques in this research field in Table 1, including their applicable size ranges and limitations. Size and morphology were analyzed using transmission/scanning electron microscopy (TEM/SEM) (Qian et al., 2015; Xing et al., 2023), atomic force microscopy (AFM) (Lead et al., 2005), dynamic light scattering (DLS), and laser-induced breakdown detection (LIBD) (Bundschuh et al., 2001); chemical composition and structure were characterized via X-ray absorption spectroscopy (XAS/XANES/EXAFS) (Newville, 2014; Terzano et al., 2019), Fourier-transform infrared spectroscopy (FTIR) (Fomina et al., 2022), inductively coupled plasma-mass spectrometry (ICP-MS, including single-particle SP-ICP-MS) (Bevers et al., 2020; Montaño et al., 2022), and nano secondary ion mass spectrometry (NanoSIMS) (Huang et al., 2017); and surface properties were evaluated through zeta potential measurements (for charge) (Gopmandal and Duval, 2022) and BET analysis (for specific surface area) (Edinger and Gernand, 2018).

Analytical challenges are significantly amplified by the fundamental chemical differences between freshwater and marine systems. Marine analysis requires methods that can overcome interference from a high-salinity matrix and gel-like organic matter, whereas freshwater techniques must navigate the complexity of humic substances. This leads to several critical limitations: it is difficult to distinguish between natural and engineered nanoparticles (e.g., Fe₃O₄ nanoparticles) in either setting, and the high ionic strength of seawater severely complicates the detection of trace nanoparticles. Future progress depends on developing in-situ monitoring technologies, such as underwater AFM for marine systems, alongside standardized methods like unified colloid size definitions to enable meaningful cross-environment comparisons.

Colloids are ubiquitous in natural aquatic systems and play pivotal roles in regulating chemical cycles and pollutant fate. Shaped by intrinsic properties (size, charge, composition) and external conditions (pH, ionic strength, natural organic matter content), their key behaviors include size-dependent transport, charge-governed stability, fractal aggregation structures, and reactive surface interactions (e.g., surface complexation, electrostatic adsorption) that mediate trace element binding and phase transitions. This size-dependency is particularly evident in their transport mechanisms: while Brownian motion largely governs the mobility of the smallest nanoscale colloids, aggregation processes become increasingly influential for their larger, submicron counterparts. These characteristics translate to profound biogeochemical impacts, such as facilitating long-distance transport of metals (e.g., in some river systems, over80% of Pb can be colloid-bound (Benoit, 1995; Lyvén et al., 2003)), regulating carbon cycling via colloidal organic matter (which is estimated to constitute 25–40% of the total marine dissolved organic carbon pool (Benner et al., 1992; Hansell, 2013)), and modifying pollutant mobility and bioavailability (Lead and Wilkinson, 2006b). Colloid behavior varies across ecosystems due to distinct physicochemical conditions, and this topic has been comprehensively summarized in numerous prior reviews, providing a solid foundational understanding of fine-particle dynamics in natural waters (Petosa et al., 2010; Liu et al., 2019; Pontoni et al., 2021; Wang et al., 2021).

Given the broad scope of colloid, this review centers on nanoparticles that possess distinctive physicochemical properties—including ultrahigh specific surface area, quantum effects, and elevated reactivity—which set their environmental behavior and geochemical functions apart from those of larger colloids. Shifting away from early work centered on toxicology of engineered nanoparticles, recent research draws attention to the profound and often unrecognized influence they exert on fundamental biogeochemical processes (Flores et al., 2025).

Deep-sea hydrothermal vents act as natural “nanofactories” in extreme marine environments, producing a diverse array of nanoparticles that exert a transformative influence on the geochemical processes of the global ocean. These systems serve as a striking example of naturally occurring nanoparticles, formed by the violent mixing of superheated, metal-rich hydrothermal fluids with cold, oxygenated seawater (Gartman et al., 2014).

For decades, it was widely believed that the metals released from these vents precipitated rapidly and remained confined to the immediate vicinity. However, this long-standing paradigm has been upended by recent discoveries—particularly from the GEOTRACES program—which demonstrate that these metals can be transported thousands of kilometers, exerting a significant impact on ocean chemistry across vast scales (Resing et al., 2015; Gartman and Findlay, 2020). The key factor of this long-range dispersal lies in the stabilization of these metals in the form of nanoparticles.

The primary geochemical significance of these particles rest on their role as transport vectors for essential elements, most notably iron. The process begins in the hot, buoyant plume rising from the vent, where rapid cooling and chemical gradients trigger the formation of metal sulfide nanoparticles such as pyrite (FeS₂) and zinc sulfide (ZnS) (Hsu-Kim et al., 2008). As the plume travels farther and mixes more extensively with oxygenated seawater, environmental conditions shift, leading to a transition in the dominant particle species to Fe(III) oxyhydroxide nanoparticles (Hoffman et al., 2020).

Characterizing these hydrothermal vent-derived nanoparticles relies on a set of advanced techniques. Field sampling is foundational, with pristine samples collected via remotely operated vehicles (ROVs) and stored in inert containers to avoid oxidation or contamination (Toner et al., 2016). For imaging and structural analysis, SEM and TEM are key tools (Figure 3)—SEM reveals the morphology of aggregated nanoparticle clusters, while TEM captures high-resolution details of individual particles (down to 4 nm) and their crystallinity (Gartman et al., 2014). X-ray diffraction (XRD) complements these by confirming the crystalline structure of minerals like pyrite and Fe(III) oxyhydroxides (Hoffman et al., 2020). For compositional and speciation analysis, energy-dispersive X-ray spectroscopy (EDS) identifies elemental components (e.g., Fe, Zn, S) and trace constituents (Pb, As), while synchrotron-based techniques like X-ray absorption near-edge structure (XANES) spectroscopy clarify metal speciation (Toner et al., 2016). Isotopic tracing (Fe, S isotopes) and size-fractionation methods (ultrafiltration, field-flow fractionation) further aid in tracking particle transport and distinguishing between soluble, colloidal, and particulate fractions (Resing et al., 2015).

The formation of organic-inorganic hybrids also play an important role in this process. Organic ligands, sourced either from the vent environment (e.g., low-temperature diffuse fluids) or ambient seawater (e.g., abiotically synthesized thiols), bind to dissolved metals and coat mineral nanoparticle surfaces. This organic coating servers to mitigate aggregation and rapid settling, thereby forming highly mobile hybrid colloids that enable metals to escape local sinks and persist in the water column across vast distances (Toner et al., 2009a; Sander and Koschinsky, 2011).

Thus, these hydrothermally derived nanoparticles serve as the critical link between the geology of the deep seafloor and the biology of the global ocean. By transporting iron—a vital yet often limiting nutrient for primary production in large oceanic regions—they directly influence marine ecosystems and play an important, previously underestimated role in the global carbon cycle. Despite advancements in analytical techniques, significant challenges persist, including the inability to capture ultrafast real-time nanoparticle transformations (e.g., sulfide-to-oxyhydroxide transitions), and insufficient understanding of organic-inorganic hybrid formation.

Particles in aquatic environments, spanning a vast scale from macro to nano in size, are not isolated entities but components of a dynamic continuum- with nanoscale materials (1–100 nm) representing a size-defined subset of colloids (1–1000 nm). This review has discussed the categories of macroparticles, colloids, and nanoparticles, emphasizing that their behavior is governed not only by intrinsic properties like size, composition, and surface charge, but is also profoundly constrained by the physicochemical conditions of their environment—particularly the distinct settings of freshwater and marine systems. Our analysis indicates that while colloids provide a foundational framework for understanding fine-particle dynamics, nanoparticles exhibit unique environmental behaviors and geochemical effects driven by their ultra-small size, ultrahigh specific surface area and consequently enhanced reactivity, which distinguish them from larger colloidal particles.

In terms of analytical methods, the techniques such as Flow Field-Flow Fractionation coupled with ICP-MS (FlFFF-ICP-MS) have greatly advanced our ability to characterize particles with minimal perturbation. However, quantitative in-situ characterization in complex natural matrices remains a major bottleneck. The high-salinity and high-organic-matter matrices typical of seawater, pose severe challenges for particle separation and detection, often compounded by artifacts introduced during sample collection and preparation (e.g., aggregation, adsorptive losses).

The case study on nanoparticles from deep-sea hydrothermal vents illustrates the critical role these natural particles play, overturning the paradigm of localized metal precipitation. It reveals that nanoparticles can act as key vectors for transporting essential micronutrients like iron over thousands of kilometers, thereby influencing global marine productivity and carbon cycling.

Although significant progresses have been reached, critical knowledge gaps persist, and these challenges highlight clear directions for future research. A primary need is the establishment of standardized analytical protocols and classification thresholds to ensure cross-study comparability. This must be coupled with the development of novel techniques for real-time, high-resolution and in-situ monitoring of dynamic transformations like aggregation, dissolution, and speciation changes. Mechanistically, the formation and reactivity of organic-inorganic hybrids, which modulate nanoparticle transport and bioavailability, remain poorly understood. Furthermore, with the increasing environmental input of engineered nanomaterials, effectively differentiating between natural, incidental, and engineered nanoparticles in complex backgrounds is a formidable yet urgent scientific challenge.

In summary, a better understanding of the sources, behaviors, and effects of aquatic particles (especially nanoparticles) is essential for assessing their roles in pollutant fate, global element cycling, and ecosystem health. As a result, future studies should be focused on these challenges, and interdisciplinary approaches should be employed.