Microplastics and nanoplastics in the atmosphere: the potential impacts on cloud formation processes

Abstract

The presence of microplastics and nanoplastics (MnPs) in the atmosphere and their transport on a global scale has previously been demonstrated. However, little is known about their environmental impacts. MnPs could act as cloud condensation nuclei (CCN) or ice nucleating particles (INPs), affecting cloud formation processes. In sufficient quantities, they could change the cloud albedo, precipitation, and lifetime, collectively impacting the Earth’s radiation balance and climate. In this perspective, we evaluate the potential impact of MnPs on cloud formation by assessing their ability to act as CCN or INPs. Based on an analysis of their physicochemical properties, we propose that MnPs can act as INPs and potentially as CCN, after environmental ageing processes, such as photochemical weathering, sorption of macromolecules or trace soluble species onto the particle surface. The actual climate impact(s) of MnPs depend on their abundance relative to other aerosols. The concentration of MnPs in the atmosphere is currently low, so they are unlikely to make a significant contribution to radiative forcing in regions exposed to other anthropogenic aerosol pollution. Nevertheless, MnPs will potentially cause non-negligible perturbations in unpolluted remote/marine clouds and generate local climate impacts, particularly in view of increased MnPs release to the environment in future. Further measurements coupled with better characterization of the physiochemical properties of MnPs will enable a more accurate assessment of climate impacts of MnPs to act as INP and CCN.

1. Introduction

Aerosols in the atmosphere, including geogenic particles, have always played a role in biogeochemical cycling and cloud formation which are important for hydrological cycles and the Earth’s radiative budget. The industrial revolution marked the start of additional sources of man-made particles into the atmosphere such as incidental particulate matter in the form of soot and other combustion generated particles. Today anthropogenic materials such as plastics have not only become essential in countless industrial applications and consumer products but also widely leaked to the environment by inefficient waste management strategies and littering.[1] Over time, this macroplastic waste will weather and degrade through mechanical, chemical and UV stresses to form microplastics (MPs < 5 mm but > 1μm) and nanoplastics (NPs, < 1 μm), here collectively called MnP.[2, 3] In this perspective, when sizes below 1 μm are particularly relevant, NPs are named specifically. Collectively, MnPs are a heterogeneous group of emerging pollutants, with different sizes, shapes, and chemical properties,[4] on which their complex environmental behavior and fate depend.[5] The presence of MnP debris has been reported in a wide range of terrestrial and aquatic environments for decades,[6-10] but addressing MnPs occurrence in and deposition from the atmospheric compartment only began in 2015.[11] Similar to the processes of dust suspension, MnPs can become airborne and be further transported over long distances in the free troposphere before deposition occurs.[12-14] Besides urban sources, where atmospheric MnPs can directly be released to the environment, (re)suspension from agricultural land and water surfaces has been identified to further contribute to the burden of atmospheric MnPs, [14-16] resulting in a plastic cycle with complex source-sink dynamics.[17]

Even though research regarding the presence of MnPs in the atmosphere is in its infancy, there is growing support for the hypothesis that MnPs are ubiquitously present in the atmosphere.[3, 18] This raises concerns regarding potentially negative environmental consequences that the presence of plastic particles in the atmosphere could have. Impacts could range from facilitated global distribution of pollutants adsorbed to plastic particles,[19] contamination of remote ecosystems[20] and effects on the Earth’s radiation balance via direct radiative effects or an altered cloud albedo regime.[15, 21] A recent study [22] which investigated the ice nucleation efficiency of laboratory-derived NPs highlighted the role of plastics in ice cloud formation and climate change, which needs further research. The prevalence of MnPs in the atmosphere raises the question, to what extent these aerosols can serve as potential cloud condensation nuclei (CCN) or ice nucleating particles (INPs),[3, 15, 21, 23, 24] playing crucial roles in cloud droplet and ice crystal formation, respectively.[25] Such aerosol-cloud interactions can have a larger radiative impact than the direct effect from aerosol-radiation-interactions.[26] Therefore, understanding to what extent MnPs have the potential to influence cloud droplet formation and ice nucleation is of high interest to assess whether or not atmospheric MnPs could alter cloud microphysics and optical properties and constitute another anthropogenic influence on the Earth’s climate system.

Due to plastics hydrophobicity, it has been suggested that MnPs are inactive CCN but active INPs [21], though it was recognized that a variety of aquatic or terrestrial weathering and atmospheric ageing processes could complicate this picture. [23, 24]. For example, photooxidation can change the hydrophobicity of MnPs over time. Additionally, crack formation on the particle surface may increase active sites and the sorption of macromolecules onto the plastic surface (e.g. eco-corona formation) and interactions with trace species (e.g. salts) can favor cloud formation processes in contrast to pristine MnPs. Consequently, the physical and chemical changes to plastics through weathering and atmospheric ageing processes must be taken into consideration to accurately understand their CCN or INP abilities in the atmosphere. However, so far, which cloud forming processes will be influenced by either pristine or environmentally aged MnPs have not been presented in view of the current knowledge of aerosol chemistry and physiochemical properties of atmospheric MnPs. Specifically, we identify relevant cloud forming mechanisms for MnP aerosols based on which, a qualitative prediction on the impacts on cloud properties is proposed. The perspective proposes future laboratory and field research needed to close current knowledge gaps on the impact of MnPs as CCN and INPs.

2. Particulate plastic in the atmosphere – sources and cycles

Since the first account of MPs in atmospheric fallout reported in 2015,[11] over 70 studies have demonstrated the presence of MPs directly in the atmosphere (e.g. through active air sampling) or originating from atmospheric sources (e.g. passive deposition sampling or surface sampling potentially representative of atmospheric deposition) (see Allen et al. 2022 and corresponding supplementary data for a comprehensive overview).[3] These particles comprised of a variety of morphologies and polymer types in both metropolitan and rural locations across the globe.[20] Measurements of atmospheric MPs close to ground or surface levels ranged from 0 or below detection limits up to a couple of thousand particles per m^-3,[3, 12, 27-31] with a tendency of lower concentrations in rural areas and near the sea and higher concentrations close to heavily trafficked urban sites.[28-30] In contrast, significantly lower concentrations were detected for atmospheric MPs over the open ocean, ranging from 0 to 1.37 particles m^-3.[32-34] Average daily deposition rates of tens to hundreds of MPs per m^-2 have been reported with a less clear trend on proximity to direct sources such as urban areas, with sizes ranging from a few microns to several thousand microns.[11, 35-40] Additionally, recent findings of NPs in melted snow samples from an Alpine environment also demonstrated the presence of atmospheric NPs in sizes ≤ 0.2 microns, with detected mass concentrations corresponding to a minimum daily deposition rate in the order of billions of NPs per m^-2 snow surface. [41]

Collectively, the presence of atmospheric MPs in remote locations, over the open ocean and in rural and/or remote snow samples indicates that MPs experience long-rage transport in the atmosphere, resulting in global distribution of MPs.[6, 17, 23, 24, 33, 34, 38, 41-43] Simulations modelling the origin and fate of MPs detected in the troposphere and atmospheric fallout suggest MPs are transported over thousands of kilometers from their initial release location before deposition occurs.[17, 42, 44] Furthermore, simulation results for atmospheric transport of spherical MPs in size mode PM_2.5 (diameter < 2.5 μm) and PM₁₀ (diameter < 10 μm) showed more wide spread dispersion of smaller particles compared to larger ones.[24] NPs present in the atmosphere are therefore likely to have an even larger potential for long-range transport compared to MPs and could be more easily advected to cloud-forming altitudes and remote sites.

Based on the current knowledge on MnPs’ emissions, transport and deposition dynamics, the fate of atmospheric MnPs is most likely best understood as a complex (re)suspension-deposition cycle as opposed to a linear pathway with distinctly defined sources and sinks (Figure 1).[3, 17] Besides direct emission in urban areas (e.g. through city dust, landfills, tire and brake wear, shedding from textiles),[7, 12, 13, 45-47] MnPs can become resuspended from contaminated terrestrial and aquatic environments.[14-16] For example, MPs content in wind-eroded material from agricultural land was up to > 7 times higher than MPs concentrations in the original soil, with enrichment of MPs in the suspended material presumed to be due to the low density of plastics compared to soil particles.[14] In the case of aquatic environments, MPs resuspension is possible through ejection of MPs by the impact of raindrops [16] or during bubble bursting, similar to the production of sea spray aerosols.[15] These aerosols consist of sea salt (NaCl) and organic matter and constitute a major source of atmospheric aerosols in marine environments, which subsequently play an important role in cloud formation processes themselves.[15, 25, 48] However, currently little is known regarding the presence of MPs in the free troposphere since only few studies have sampled MPs concentration directly at this altitude (i.e. Allen et al. 2021 and González-Pleiter et al. 2021).[42, 44]

Figure 1.

Possible pathways for MnP cycling in the atmosphere. Arrows depict suspension and deposition pathways to and from terrestrial and aquatic environments.

To better understand MnPs cycling dynamics, more knowledge regarding the presence of MnPs in the free troposphere and the processes that govern their dry and wet deposition is needed.[23, 24, 42] For the latter, one can distinguish between below-cloud scavenging or incloud scavenging processes. Below-cloud scavenging describes the removal of particles or species by precipitating raindrops or snowflakes below the cloud base. In contrast, in-cloud scavenging occurs when particles acting as CCN or INPs during cloud formation eventually fall as precipitation. Therefore, assessing the potential of MnPs to act as CCN or INPs is not only relevant to understanding their influence on cloud formation processes, but is also relevant to better understand MnPs cycling dynamics.

3. MnPs as potential CCN or INPs in cloud formation processes

Clouds are an essential part of the atmospheric system with a great impact on the hydrological cycle and the Earth’s radiation budget. Their formation constitutes a complex interplay of processes that govern the phase transitions of water vapor in the atmosphere to form liquid cloud droplets and ice crystals. Depending on their composition, clouds can be classified into liquid, ice and mixed-phase clouds (MPCs).[25] As their names suggest, liquid (warm) clouds consist solely of liquid water droplets and ice (cirrus) clouds only of ice crystals while MPCs constitute a mixture of both.[25] Warm clouds mainly occur in the lower troposphere at temperatures above 0 °C, ice clouds dominate in the higher troposphere at temperatures lower than -38 °C,[25, 49] while in between, MPCs can form in the middle troposphere at subfreezing temperatures above -38 °C.[49] MPCs occur at all latitudes from the polar regions to the tropics, and due to their ubiquitous presence in the troposphere, they play an important role in many different atmospheric processes such as precipitation formation and cloud electrification.[50] Furthermore, MPCs impact the Earth’s radiative energy balance on both the regional and global scales.[50] Phase partitioning plays a crucial role in the radiative properties of MPCs since the ice phase of water has a lower albedo than the liquid phase.[50] Therefore, the albedo and radiative properties of MPCs depend on the ratio of cloud liquid droplets and ice crystals and how the two phases are distributed in the cloud.[49, 50] Crucial in understanding this partitioning are the two main phase transition processes in MPCs, i.e. cloud droplet formation and heterogeneous ice nucleation (see Supplemental Information for more information on cloud droplet formation and heterogeneous ice nucleation, BoxS1 and BoxS2).[25, 51] Both these processes depend on the presence of atmospheric aerosols acting as CCN or INPs, respectively. Thus, the addition of MnPs to the atmosphere could possibly contribute to the primary formation and phase partitioning of MPCs with currently unaccounted for consequences for MPCs’ radiative properties. MnPs could also contribute to ice nucleation in cold (cirrus) clouds or to cloud droplet formation in warm clouds. To explore the potential of atmospheric MnPs to act as CCN or INPs, understanding the relevant microphysical processes that govern cloud droplet formation and heterogeneous ice nucleation and their dependence on the physical and chemical properties of atmospheric aerosols is necessary.

Briefly, a different set of physical and chemical properties including size, solubility, surface structure and hydrophobicity is descriptive for the ability of an aerosol to facilitate cloud droplet formation or heterogeneous ice nucleation (Figure 2). For cloud droplet formation, exceeding the critical radius at atmospherically relevant supersaturation and particle solubility (hydrophilicity) are important predictors for CCN. In contrast, the ability of an aerosol to act as INP strongly depends on its surface area, the physical structure of the surface and chemical characteristics of the surface (polymer chemistry and/or environmental aging) that can lead to locally elevated compatibility for ice nucleation. Using the current knowledge regarding the physical and chemical characteristics of atmospheric MnPs with these properties will allow an initial assessment as to whether and under what preconditions atmospheric MnPs in relevant size fractions for cloud formation processes (i.e. those plastic particles having a diameter of <10 μm) could potentially act as CCN and/or INPs. The lower density of MnPs compared to mineral dust, could render particles >10 μm as CCN and INPs, if advected to cloud forming altitudes and transported in the absence of precipitation where wet deposition is negligible.

Figure 2.

MnP properties relevant for cloud droplet activation via CCN (left) or ice crystal formation via INPs by PCF or deposition nucleation (centre) and immersion freezing (right). Functional groups shown can arise from organic compounds adhering to the particle or biomacromolecules forming eco-coronas which can promote water uptake or ice cluster formation from supercooled water vapor.

3.1. Potential of atmospheric MnPs to act as CCN

In general, accumulation-mode aerosols (radii between 0.05 μm to 0.5 μm) are the most common CCN because of their long residence times in the troposphere.[25] Aitken-mode aerosol (radii between 0.005 μm to 0.05 μm) can also act as CCN depending on their concentrations, coagulation rates and composition.[25] Smaller aerosols do not reach critical radii under supersaturation levels typically existing in the troposphere and larger particles are only present in smaller numbers. For insoluble but completely wettable particles, the critical diameter required to overcome the energy barrier of phase transition at typical atmospheric supersaturation levels lies between 0.2 μm and 0.4 μm. To date, most published studies report on atmospheric MPs (with sizes in the micron range) which exceed this threshold. However, NPs are also present in the atmosphere in sizes ≤ 0.2 μm [41] that may not necessarily reach the critical diameter for cloud activation under atmospherically relevant conditions. Regardless of size, pristine plastics are inherently hydrophobic (see details on the effects of environmental aging on hydrophobicity below), suggesting that their ability to act as CCN may be reduced due to a high contact angle (i.e., low wettability). Compared to perfectly wettable surfaces (i.e., contact angle of 0° between the water phase and the surface of the substrate) onto which a liquid phase can form a continuous water film, plastics repel water to a certain extent. In such cases, higher supersaturation levels are required for a particle to act as a CCN since the formation of a liquid water phase exhibits a nonzero contact angle on the surface of the particles. An adapted version of the Kelvin equation approximated for small contact angles estimates that aerosols exceeding contact angles of 6° will probably not commonly act as CCN under atmospherically occurring supersaturation levels (Supplemental Information, Figure S3 and corresponding text).[52] Depending on the polymer type, liquid water on a bulk plastic surface has contact angles between 70° to 105°,[53, 54] and consequently it is unlikely that pristine NPs would readily act as CCN. For example, polyethylene (PE), a polymer commonly found in atmospheric fallout, has a water contact angle between 95° to 100°.[54, 55] Consequently, for a pristine PE particle with a radius of 0.5 μm to act as CCN, it would need a supersaturation ratio as high as 6.4 to 7.4, which is not found to naturally occur within tropospheric cloud relevant conditions.

The role of atmospheric MnPs as CCN may be significantly underestimated when only considering their physicochemical properties in their pristine state. Atmospheric MnPs will experience different extents of weathering either prior to their (re)suspension (i.e. directly in aquatic or terrestrial systems) or aging during their suspension and transport in the atmosphere. These aging processes will alter the surface properties of MnPs and potentially increase their affinity to water (i.e. lower their contact angles with water), thus increasing their CCN activity. For example, UV irradiation could decrease hydrophobicity by increasing the particle surface roughness and triggering oxidation processes via photochemistry (Figure 3) that cause the formation of hydrophilic chemical groups.[56, 57] FTIR and Raman spectra of MPs from atmospheric fallout showed that they are likely to contain carbonyl and hydroxyl groups, supporting this hypothesis.[40, 58-61] If these weathering characteristics stem from MnPs degradation in the atmosphere or before their suspension is still unknown. A better understanding will be needed regarding atmospheric MnPs exposure to UV radiation and the rate and extent of photooxidation, which likely depends on the average residence time and altitude of MnPs in the atmosphere. Furthermore, surface weathering characteristics of MnPs could also originate from interactions with other substances present in the atmosphere such as nitrogen oxides, hydroxyl radicals or ozone.

Figure 3.

Cloud forming processes that are expected to be relevant for MnP cloud formation. MnPs in their pristine and aged forms are hypothesized to be INPs, however can only be CCN active via activation or hygroscopic growth after ageing results in a more hydrophilic surface (see text section 3.1). Freezing pathways of MnPs include immersion freezing in supercooled liquid drops (MPCs) and forming ice crystals in the cirrus cloud regime below -38 °C via deposition nucleation or PCF. A fraction of MnP that remain inactive INPs even after environmental weathering or atmospheric ageing will not promote freezing of supercooled cloud drops, and require homogeneous freezing temperatures to form ice. Black text indicates relevant processes, white text indicates important in-cloud conditions.

The adhesion of other natural or anthropogenic particles and chemicals onto the surface of atmospheric MnPs in the form of coatings or eco-coronas must also be considered (Figure 2).[21, 23] Even though particles adhering to atmospheric MnPs have been reported,[40] their chemical composition and origin have so far not been investigated. Nevertheless, it is likely that they could partially constitute soluble and hydrophilic compounds such as sulfates, sea salt or organics (Figure 2) that have been reported to also adhere to the surface of other aerosols.[62, 63] Such particles could either sorb to MPs while suspended in the atmosphere or before their (re)suspension into the air or adhere by the triboelectric effect during wind suspension due to static charging of MnPs and dust particles.[15] By increasing the wettability of the particle surface and the potential to facilitate water uptake (Figure 3), such compounds could significantly decrease the critical supersaturation to activate particles as CCN. For example, the presence of trace amounts of NaCl on black carbon aerosols substantially enhances their CCN ability by reducing the critical supersaturation from approximately 2% to 0.2%.[52] Since resuspension from sea surfaces is an emission pathway for atmospheric MnPs, it is conceivable that such plastic particles have sea salt adhered to their surface. Similarly, atmospheric MnPs resuspended from aquatic or terrestrial environments could be coated with an eco-corona, comprising a variety of organic compounds and biomacromolecules, some of which may be hydrophilic and soluble.[64] Both of these scenarios could increase the likelihood of weathered MnPs to act as CCN (Figure 3) more readily than their pristine counterparts.

The impact of environmental ageing on the CCN ability of MnPs suggests that in the future, with increased atmospheric burdens of anthropogenic particles, currently unpolluted clouds over marine or remote regions such as the Southern Ocean could be perturbed by CCN contributions of aged MnPs, thus altering cloud dynamics and microphysics. For a constant liquid water path, increases in CCN concentration (from aged MnPs) imply numerous cloud droplets with smaller effective radii, suppressed precipitation, increased latent heat release resulting in convective invigoration, delayed precipitation [65] and increased cloud lifetime (see Figure 4). The latter, for liquid clouds, results in a radiative cooling effect compared to an unperturbed case (Figure 4b, larger blue arrow). It is important to note that the current MnP aerosol concentrations would be too low to impact regions that are already influenced by anthropogenic aerosol pollution where number concentrations are on the order of thousands per cm³ of air.

Figure 4.

Anticipated impact of MnPs on cloud properties in the future as anthropogenic aerosols increase. Perturbed liquid phase clouds should have an increased cooling effect (larger blue arrow, panel B) compared to present day non-polluted liquid clouds due to smaller cloud droplet effective radius and longer lifetime (panel A). Cirrus clouds could have a potentially reduced warming effect due to a decrease in altitude and larger but fewer ice crystals. MPCs could experience a variety of dynamical effects such as convective invigoration increasing updrafts and the vertical extent of the cloud, delayed but heavier precipitation [65], increased ice crystal number concentrations, and smaller cloud drops due to an increased Wegener-Bergeron-Findeisen process. Collectively, these process all could affect the cloud lifetime and radiative properties making it impossible to place a general cooling or warming signs on such clouds compared to present day conditions where MPCs are thought to generally have a cooling effect.

To assess the ability of atmospheric MnPs to act as CCN, research into their weathering processes and atmospheric chemical mixing state are needed. Moreover, laboratory studies will be necessary to validate the hypotheses regarding MnPs efficiency as CCN and provide insights into the relative importance of weathering processes and adsorbents. This would also require laboratory experiments to work with a range of model particles, including those which are more realistically aged (e.g. with eco-corona formation, sorbed salts and particles, photo-oxidized surfaces and varying degrees of surface roughness and cracks promoting capillary condensation) in order to assess their true CCN potential in the atmosphere.[66]

3.2. Theoretical potential of atmospheric MnPs to act as INPs

Ice nucleation is facilitated by INPs when the particle surface lowers the energy barrier for phase transition by ice clusters being stabilized on the surface. It is thought that heterogeneous ice nucleation initiates at specific locations on the INPs surface, i.e., active sites.[25] Induced by mechanical weathering, heterogeneities on MnPs surfaces such as fractures, grooves, pits and flakes could constitute active sites with locally elevated potential for ice nucleation, enabling MnPs to act as INPs.[40, 67] Besides physical surface alterations, the presence of hydroxyl or carbonyl groups on MnP surfaces,[40] possibly due to chemical aging in the atmosphere or eco-corona formation in fresh water bodies, can further enhance their ability to facilitate freezing. First findings on the influence of laboratory-derived NPs on the freezing behavior of water droplets under laboratory conditions also support the hypothesis that these particles could act as INPs.[22] In the presence of low-density PE particles (size range from 0.1 μm to 0.8 μm), immersion freezing occurred at a median freezing temperature (T₅₀) of -15.1°C compared to the control (when no NPs were present) where the T₅₀ was -21.0°C. Mixing with environmental contaminants either enhanced or reduced the ice nucleation activity of the plastic particles, shifting the T₅₀ by a few degrees. Depending on the compound that the low-density PE NPs were mixed with, an increase in freezing temperature between 1.9°C and 7.1°C was observed, highlighting that the potential for MnPs to act as INPs likely depends on the sorption of other compounds to the particle surface. It remains to be investigated what the upper temperature limit is at which internally mixed MnPs can act as INPs and whether it could overlap with conditions under which important INP types initiate freezing. For example, the most important INPs in the atmosphere, mineral dust, initiates freezing below -15°C when present in substantial concentrations, but can also readily initiate freezing at higher temperatures depending on factors such as mineral type, particle size and particle concentration per droplet.[51, 68]

An overlap between freezing temperatures between MnPs and dust particles could have implications for MPC formation for which immersion freezing is the dominant heterogeneous ice nucleation process.[25, 51] Overlapping freezing temperatures would suggest that plastic particles could have the potential to facilitate droplet freezing in MPCs in parallel with or in the absence of mineral dust. Considering their low-density, MnPs could show different atmospheric dispersion and deposition patterns and may have a greater reach than mineral dust. This is especially relevant for remote regions such as the Southern Ocean that are scarce in mineral dust, where very low INP concentrations suppress the freezing of cloud droplets.[69] For this region, immersion freezing INP number concentrations were reported between 0.38 to 4.6 m^-3 at -20°C[70], while in other regions of the world, INP number concentrations easily reach between 10 to 10⁴ m^-3 [51]. Using recently published modeling findings on the global monthly average mass concentrations of MPs in the atmosphere,[23] and the mass based active site density of high-density-polyethylene NPs at -20°C,[22] we estimate the contributions of MnP INPs to be between 0.2 to 1.6 m^-3 at -20°C. This is on the same order of magnitude compared to the current INP concentrations reported in the Southern Ocean, suggesting that accounting for MnP doubles the INP burden in the Southern Ocean. The current estimate is based on available MP concentrations in the literature, which can be expected to be higher in the future, not only because of increased MnP burden, but also as an estimate for NP concentrations in the troposphere becomes more evident through development or more accurate and precise detection methodologies. The quantification of the contribution of NP to freezing at MPC conditions (between 0 and -38 °C) is crucial as increased ice crystal number concentrations can enhance the Wegener-Bergeron-Findeisen [71-73] process and precipitation formation through more effective ice-liquid collision and coalescence. In particular, when this effect is viewed as coupled with higher CCN concentrations from NPs in the future (section 3.1), we expect unpolluted remote/marine clouds to experience non-negligible perturbations in dynamics, microphysics (Figure 4) and eventually the lifetime and optical properties of the cloud adding to the unconstrained radiative effect.

Besides impacts on the formation of MPCs, atmospheric MnPs could also contribute to ice cloud (cirrus) formation via deposition nucleation or pore condensation and freezing [74] (Fig. 3 and 4) where ice directly deposits on aerosol surfaces or is first facilitated by capillary condensation of water in(to) small pores or cracks followed by freezing, respectively. Typically, these mechanisms are active below -38°C in the upper troposphere where bulk water droplets do not exist for immersion freezing to be relevant. Generally, the planetary boundary layer does not reach these low temperatures, except sometimes in the Polar Regions during winter, and therefore this process would only be relevant for MnPs advected to the upper troposphere where cirrus clouds form. While MPs presence has been reported for the free troposphere up to altitudes of approximately 3500 m a.s.l.,[42, 44] more research on the vertical extent of atmospheric MnPs will be needed to determine MnPs presence at conditions relevant for this process and the potential impacts on cirrus cloud formation. In particular, if MnPs nucleate ice at thermodynamically favourable conditions (T < -38 °C and saturation ratio with respect to ice < 1.4 – 1.5), this could alter the altitude of cirrus, size and number density of ice crystals, potentially reducing the cirrus warming effect (Figure 4b, smaller red arrow with question mark). Thus, it is critical to quantify the ice nucleation ability of NPs via laboratory measurements of model NPs under such conditions.

4. Future directions and recommendations

MnPs have the potential to act as both INPs and CCN, depending on their physical and chemical properties, both of which are known to control cloud droplet activation and ice nucleation processes. In particular, MnPs with diameters less than 10 μm have greater likelihood of acting as INPs. Their low density does not preclude particles larger than 10 μm from acting as cloud forming nuclei if advected to cloud altitudes in the absence of wet deposition. MnPs with surface defects and roughness could be INPs via deposition nucleation or pore condensation freezing to form ice crystals in the upper troposphere. Their role as CCN can be important if they age and internally mix with other compounds that increase water affinity, such as through sorption of macromolecules and/or trace soluble species, or through photochemical weathering. To determine whether these mechanisms are relevant, more information about morphological properties of MnPs collected from the atmosphere and the changes in these properties as a result of mechanical, photooxidative and chemical weathering is needed. To this end, the rate and extent of sorption of hydrophilic compounds should be identified. Knowledge gaps in our understanding of which environmental conditions promote aging processes need to be filled. These determinant factors for MnPs to act as CCN or INPs are required, but have not been systematically investigated to date.

Knowledge of the atmospheric concentrations and size distributions of MnPs in the atmosphere to assess their contribution to cloud formation process is crucial. Based on the current modelling results on MnPs atmospheric mass concentrations, their impacts on cloud formation process should not be neglected, although the exact effects depend on the actual abundance and cloud forming potential of plastic particles in each region. NPs may play a prominent role in cloud formation processes, as aerosols with diameter below 1 μm are most relevant for cloud droplet formation and heterogeneous ice nucleation. There have been several studies that report an increased abundance of smaller size of MnPs in the atmosphere [6, 35, 38]. Nevertheless, the abundance of NPs so far is almost entirely omitted in most field research programs for studying atmospheric fallout of MPs (with few exceptions [41, 75, 76]), which only reported particle diameters down to 10 μm. Therefore, NPs could potentially constitute a large majority of plastic in terms of particle number in the atmosphere and increase the potential for plastics to play a prominent role in cloud formation processes.

The improvement of our understanding of physicochemical properties and concentrations of atmospheric MnPs relies on more advanced approaches to detect and measure them. Currently, MnPs concentrations are likely substantially underestimated due to the limit of analytical detection techniques, particularly for smaller sized particles, which induces significant uncertainties in the evaluation of the contribution of MnPs to cloud formation. To better characterize physicochemical changes of the particles when aged we need to develop new methods to identify MnPs and lower the particle size detection limit in the field and compliment this with laboratory research for more mechanistic work. Methods used in atmospheric research for sampling aerosols and identifying CCN and INP-relevant chemical composition could likely be used or adapted for MnP. Modeling studies have proposed CCN/INP efficiencies due to their ability to explore MnP-aerosol-cloud interactions,[20] but in the future modeling efforts should be validated with observational data. With these improvements in MnPs characterisation and measurements, researchers will be able to accurately depict if plastics facilitate cloud formation.