1 State Key Laboratory for Structural Chemistry of Unstable and Stable Species, Beijing National Laboratory for Molecular Sciences, CAS Research/Education Center for Excellence in Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences,Beijing 100190, China
2 University of Chinese Academy of Sciences,Beijing 100049, China
3 Institute for Environmental and Climate Research, Jinan University,Guangzhou 511443, China
4 State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences,Guangzhou 510640, China
5 State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking University,Beijing 100871, China
Hygroscopicity is one of the aerosol’s most important physicochemical properties, which affects the lifetime and atmospheric behavior of aerosols. It plays a vital role in the environment, climate change, and human health. The hygroscopicity parameters and thermodynamic models are introduced. The effects of aerosol diameter, chemical composition, and multi-component aerosol on hygroscopicity are analyzed in the following section. The relevant observations in the urban, rural area, forest, polar region, and ocean are summarized. Hygroscopicity growth factor (g(RH)), scattering enhancement factors (f(RH)) and hygroscopicity parameter κ are common hygroscopicity parameters which can represent aerosol hygroscopicity. The Zdanovski-Stokes-Robinson (ZSR) mixing rule and many thermodynamic models can predict the aerosol hygroscopicity with different chemical compositions, which are some important tools to study the multi-component aerosol and phase equilibrium. Aerosol diameter, chemical composition and mixed state affect hygroscopicity, for example g(RH), and changes in deliquescence relative humidity or efflorescence relative humidity. Because of different emission sources and environmental conditions, the aerosol particle size distribution, chemical composition and mixed state are different in the urban area, rural area, forest, polar region and ocean, and then aerosol hygroscopicity are different. Hygroscopicity affects aerosol water content and phase state directly, and atmospheric chemical processes, aging process and lifetime of aerosol will be changed. Environmental visibility, radiation effects, and aerosol deposition sites and toxicity in the human body also are influenced by aerosol hygroscopicity. Overall, a comprehensive review of hygroscopicity parameters, theoretical models, laboratory studies, field experiments, and environmental impacts provides a reference for future research.
Contents
1 Introduction
2 Hygroscopicity parameters and models
2.1 Hygroscopicity parameters
2.2 Aerosol hygroscopic models
3 Influence factors of aerosol hygroscopicity
3.1 The influence of particle size
3.2 The influence of chemical composition
3.3 The influence of multi-component mixture
4 Aerosol hygroscopicity in field studies
4.1 Observation in urban area
4.2 Observation in rural and forest areas
4.3 Observation in oceans and polar regions
4.4 Vertically Distributed Aerosol hygroscopicity
5 Effects of aerosol hygroscopicity on atmospheric chemistry,visibility,climate and human health
5.1 Effects of aerosol hygroscopicity on atmospheric chemistry
5.2 Effects of aerosol hygroscopicity on visibility
5.3 Effects of aerosol hygroscopicity on climate
5.4 Effects of aerosol hygroscopicity on human health
Aerosols are solid or liquid particles stably suspended in the atmosphere, ranging from a few nanometers to tens of microns in size, and widely exist in the atmosphere. They not only affect the global climate through direct or indirect radiation effects, but also affect air quality and human health[1].
Hygroscopicity is one of the important physical and chemical properties of aerosols, which represents the ability of aerosols to absorb water at different temperatures and Relative humidity (RH). According to whether the RH is greater than 100%, the hygroscopicity of aerosols can be divided into hygroscopicity under unsaturated and supersaturated conditions. Under unsaturated conditions, aerosols absorb or adsorb water, which may change their particle size, phase and surface morphology. Under supersaturated conditions, some aerosols that can be activated into Cloud droplets are called Cloud condensation nuclei (CCN)[2].
Aerosol hygroscopicity is affected by many factors, such as its particle size, chemical composition and mixing state[1,3,4]. The hygroscopicity of different chemical components is quite different, such as strong hygroscopic sea salt, weak hygroscopic organic matter such as phthalic acid, and non-hygroscopic Black carbon (BC)[5,6]. When different chemical components exist in different mixing modes, the components interact with each other, and the deliquescence weathering behavior is more complex. Actual environmental aerosols have the characteristics of complex mixing state and wide particle size distribution, and their composition and mixing state are further transformed during long-distance transport. Urban, rural, forest, marine and polar regions have their own environmental characteristics, and the chemical composition of aerosols is significantly different due to the influence of different types of pollution sources[1]. City is a large human gathering place with frequent living, transportation and production activities. Man-made sources such as motor vehicles, coal combustion and industrial production will emit a large number of aerosols and reactive gases, resulting in serious pollution[7⇓~9]. In rural and forest areas, human activities are relatively weak. In addition to natural sources, agricultural activities, biomass burning and regional transport also contribute to aerosols[10⇓~12]. However, human activities in the ocean and polar regions are very few, mainly contributed by processes such as natural emissions from the ocean and long-distance transport[13,14]. Aerosol size distribution, chemical composition and mixing state are significantly different in different environments, resulting in different hygroscopicity.
On the one hand, hygroscopicity affects the atmospheric behavior of aerosols, including atmospheric chemical processes, aerosol aging and atmospheric lifetime. On the other hand, it will also affect the climate, environment and health effects of aerosols. First of all, aerosol hygroscopicity directly affects the water content and phase state of aerosol, so the liquid phase reaction and heterogeneous reaction of aerosol are affected, and the chemical composition is also changed[15,16]. Secondly, hygroscopicity is closely related to the CCN activity of aerosols, which can significantly affect the wet deposition rate of aerosols and change their atmospheric lifetime and spatial and temporal distribution. At the same time, the activation ability of aerosols can affect the formation and development of clouds, change the albedo of the earth, and affect the climate through indirect effects[17]. Hygroscopicity further affects aerosol optical properties, which directly affect the atmospheric radiation balance and climate by changing aerosol scattering and absorption characteristics[18]. When aerosols are inhaled into the human body under high RH conditions (> 90%), the change of particle size will affect the deposition efficiency and location of aerosols in the respiratory tract, thus having an important negative impact on human health[19].
In this paper, aerosol hygroscopicity parameters and main thermodynamic models are introduced, various factors affecting aerosol hygroscopicity are summarized, and the latest research results of aerosol hygroscopy are introduced from different perspectives.The influence of hygroscopicity on aerosol climate, environment and health effects is summarized, and the future research on aerosol hygroscopicity is prospected.
2 Hygroscopic parameters and model
The process by which aerosols absorb (lose) water and change phase is called deliquescence (weathering)[20]. The relative humidity at which an aerosol suddenly absorbs a large amount of water and becomes a droplet is called Deliquescence relative humidity (DRH), otherwise it is called Efflorescence relative humidity (ERH)[20]. ERH tends to be inconsistent with DRH, with hysteresis, as shown by the deliquescence and weathering curves of (NH4)2SO4 in Fig. 1. Because environmental aerosols are generally highly mixed with multiple components, their deliquescent weathering behavior is difficult to observe.
Aerosol hygroscopicity has been studied by different instruments. Table 1 gives a brief summary of common hygroscopicity observation instruments and measurement parameters. The Hygroscopic growth of aerosols can be measured by Electrodynamic balance (EDB), optical levitation technique, Hygroscopic-Tandem differential mobility analyzer (H-TDMA), and humidified turbidimeter. The change of aerosol in the process of deliquescent weathering can also be characterized by various microscopes, and the chemical change and water absorption change of aerosol at different RH can be characterized by Fourier transform infrared spectroscopy, Raman spectroscopy and fluorescence spectroscopy.
2.1 Hygroscopic parameter
Hygroscopic growth factor G (RH), scattering hygroscopic growth factor f (RH) and hygroscopic parameter κ can measure the hygroscopicity of aerosols, which are commonly used to describe the hygroscopicity of aerosols.
2.1.1 Hygroscopic growth factor G (RH) and scattering hygroscopic growth factor f (RH)
H-TDMA is widely used in field observation and laboratory research to obtain G (RH) of aerosol. G (RH) is defined as the ratio of the electromobility equivalent particle size (D (RH)) of a particle after hygroscopic growth to the electromobility equivalent particle size (D (dry)) of a dry aerosol at a certain relative humidity, that is,
The optical properties of aerosol, such as scattering and absorption, are changed by the moisture absorption process[18]. The aerosol scattering coefficient at different relative humidity is measured by humidified turbidimeter, and f (RH) can be calculated:[18,21]
Where σ (RH) is the aerosol scattering coefficient at a certain relative humidity, and σ (dry) is the aerosol scattering coefficient under dry conditions.
There are many fitting parameter equations that can be used to describe G (RH), for example, the simplest two-parameter fitting equation is (a and γ are empirical parameters obtained after fitting the data), which is widely used in H-TDMA observation data fitting[22]. Kreidenweis et al. Derived from the K Köhler equation (aw is the water activity and B, C, d are fitting parameters)[23]. Rissler et al. Then simplified it into a one-parameter fitting equation is the fitting parameter)[24]. Similarly, some studies have established many parametric fitting relations for f (RH), such as the one-parameter fitting equation , the two-parameter fitting equation (where γ represents the amplitude of scattering enhancement),The three-parameter equation (where A1, A2, A3 are the fitting parameters), and more complex polynomial fitting, etc[18,25]. Titos et al. Compared the predicted results of eleven different fitting parameter equations with the actual measured results, and found that most of the fitting equations could describe the scattering enhancement caused by monotonic hygroscopic growth, but only the latter two complex equations could accurately reproduce the f (RH) hygroscopic growth of laboratory (NH4)2SO4[18].
2.1.2 Hygroscopic parameter κ
The particle size of aerosol changes in the process of moisture absorption, which can be described by the classical K Köhler theory. At a certain relative humidity, when the aerosol is in equilibrium with the ambient water vapor, its surface saturation vapor pressure is affected by both the Kelvin effect and the solute effect, and the K Köhler equation can be obtained:[1]
Where S is the degree of saturation, aw is the water activity, D is the droplet diameter, σs/a is the surface tension, Mw is the molar mass of water, ρw is the density of water, T is the temperature, and R is the ideal gas constant.
Petters et al. Proposed a single hygroscopic parameter κ, which can describe the hygroscopicity of aerosols, and κ is defined as
Where Vs is the volume of dry particles and Vw is the volume of water. The Zdanovskii-Stokes-Robinson (ZSR) mixing law assumes that the volumetric moisture content of aerosols is additive, and as inferred from the combination of (3) and (4), the hygroscopic growth factor G (RH) is related to RH by:
Where
When the size of aerosol is larger, the influence of Kelvin effect is smaller, that is, the is smaller. The Kelvin effect of aerosols with large particle size (> 100 nm) can be ignored, and equation (5) can be simplified as
The interaction between insoluble aerosols such as BC and mineral dust and water is different from the hygroscopic growth mechanism of general water-soluble aerosols (inorganic matter and soluble organic matter), which is through the adsorption process: at low RH, water is adsorbed on the surface of insoluble particles until the surface is completely covered by a single layer of water molecules; Then the water molecules continue to attach to the existing water layer[3]. There are many theories that can describe the water adsorption of mineral dust aerosol under unsaturated conditions, such as Langmuir adsorption isotherm, Brunauer-Emmett-Teller adsorption isotherm, Freundlich moisture adsorption isotherm and Frenkel-Halsey-Hill (FHH) isotherm adsorption curve[21].
2.2 Aerosol hygroscopic model
Aerosol thermodynamic model can predict the hygroscopic capacity of aerosols with different chemical components, and is an important tool for studying multi-component mixed aerosols and gas phase equilibrium[1]. The ZSR mixing law is the most common and simplest prediction law. In addition, thermodynamic models such as ISORROPIA, Extended Aerosol Inorganics Model (E-AIM) and Aerosol Inorganic-Organic Mixtures Functional Groups Activity Coefficients (AIOMFAC) are widely used.
The ZSR mixing law assumes that there is no interaction between the components in the mixed aerosol, and the volume of each component is constant before and after mixing. Considering the hygroscopic growth of each component at different RH, it can be applied to predict the hygroscopic growth of multi-component mixed aerosol[37]. The g(RH)mixed can be expressed as:[37]
Where g(RH)mixed is the hygroscopic growth factor of mixed aerosol, g(RH)k is the hygroscopic growth factor of single component K, and εk is the volume fraction of component K in dry aerosol. The ZSR law does not take into account inorganic-organic interactions in the mixed system, so there are significant differences between the predicted curves and the measured curves of some aerosols. Some researchers have improved the ZSR law to improve the accuracy of prediction[38].
Nenes et al. Established a thermodynamic equilibrium model for inorganic aerosols, called ISORROPIA (ISORROPIA is "equilibrium" in Greek)[39a,40a]. It can predict the water content, pH, distribution coefficient and phase composition of multi-component mixed typical inorganic salts at different RH, and the calculation speed is fast[1,39b]. In the ISORROPIA II model, crustal elements such as K+, Ca2+ and Mg2+ are added to optimize the algorithm, and the calculation speed is not slowed down. So far, ISORROPIA has been replaced by the ISORROPIA II model[40b].
Clegg et al. Established the Atmospheric Inorganic Aerosol Equilibrium Partition model (AIM), which initially has three systems to calculate Aerosol water content, phase and gas-particle partition[1,41a]. In order to extend the range of applicability of the model, a variety of organic compounds are added after AIM to study inorganic-organic mixed systems, which is called E-AIM (Extended Aerosol inorganics model). The E-AIM model then adds a fourth subsystem, with concentrations ranging from infinite dilution to saturation, and compounds that can be selected or customized by the investigator from the E-AIM model chemistry library[41b].
Zuend et al. Proposed the AIOMFAC model for the atmosphere[42]. This group contribution model enables the calculation of activity coefficients for inorganic, organic, and inorganic-organic mixed solutions over a wide range of ionic strength[42]. Then Zuend et al. Modified and expanded the parameterization range of the AIOMFAC model, introducing a variety of functional groups into the model[43]. In 2015, Ganbavale et al. Proposed a new temperature-dependent parameter to further improve the model parameters[44]. The model can be used to calculate the deliquescent relative humidity, gas-particle partition and liquid-liquid phase separation of multicomponent mixtures, and the predictions of the improved model are in good agreement with a large number of experimental data.
The application scope, conditions and calculation methods of each model are different, and researchers can choose the appropriate model according to the research needs. ISORROPIA II can simulate inorganic mixed aerosols and is mostly used in atmospheric chemical transport models with fast calculation speed[45]. However, it can not simulate the influence of organic matter on inorganic components, and the predicted value of moisture absorption of environmental aerosols is different from the measured value[46]. E-AIM model extends the scope of AIM model and can study inorganic-organic mixed aerosols, but there may be errors in the prediction of complex inorganic-organic mixed aerosols[1]. The AIOMFAC model takes into account the interaction between inorganic and organic compounds, but does not include important compounds such as peroxides, organic nitrates and peroxyacyl nitrates, and there are still large errors in predicting mixtures dominated by macromolecular organic compounds with complex structures[39][44]. Due to the computational complexity of inorganic-organic mixed aerosols, these inorganic-organic mixed models are generally not used in three-dimensional atmospheric models without replacing the organic chemicals in the model with a small number of chemical substitutes[39].
3 Influencing factors of aerosol hygroscopicity
Aerosol hygroscopicity is affected by many factors. This paper focuses on the influence of three main factors: aerosol particle size, chemical composition and mixing state. Aerosol particle size affects its hygroscopicity, and the hygroscopic capacity of different chemicals may vary greatly. When multiple chemical components coexist in different mixing modes, the interaction between components is complex, and aerosols show different hygroscopic behaviors.
3.1 The effect of particle size
As the particle size decreases, the G (RH) of the aerosol will also decrease due to the Kelvin effect[47]. However, the G (RH) of nano-aerosol is affected by both Kelvin effect and nano-effect. Biskos et al. Found that with the decrease of particle size (from 40 nm to 6 nm), the G (RH) of NaCl aerosol decreased significantly (from 1.75 to 1.55), and when the particle size was less than 40 nm, the G (RH) showed a significant downward trend. The prediction of G (RH) for non-spherical nanoaerosols requires the model to account for the Kelvin effect and the size-dependent shape factor[48]. The 6 nm NaCl aerosol showed nanoscale effects, which significantly deviated from the model predictions[48]. Giamarelou et al. Measured the hygroscopicity of KCl, KBr, and KI particles from 8 to 60 nm, showing a similar trend[49]. Zieger et al. Found that the G (90%) of sea salt aerosol of 50 ~ 150 nm increased slightly with the decrease of particle size (from 2.19 to 2.22), which may be due to the increase of solubility or surface tension of sea salt components after the decrease of particle size, resulting in the increase of hygroscopicity[50].
Particle size has an important influence on the hygroscopic growth and deliquescent weathering of aerosols, especially the nano-effect, which makes aerosols show different properties. However, there are few experimental studies on the hygroscopicity of nano-aerosols, and the observed phenomena such as hygroscopic growth and deliquescence weathering of nano-aerosols have not been well explained. In the experiment, the particle size of NaCl aerosol decreased from 40 nm to 6 nm, and the ERH and DRH increased from 45% and 75% to 53% and 87%, respectively[51]. Lei et al. Also found that the ERH and DRH of nanoaerosols increased with the decrease of particle size[52]. The ERH and DRH of Na2SO4 from 6 nm to 20 nm showed significant particle size dependence, but the ERH and DRH of (NH4)2SO4 aerosols were not significantly affected even at a diameter of 6 nm[52].
3.2 Influence of chemical composition
The chemical composition of aerosol is an important factor affecting its hygroscopicity. Environmental aerosols contain a large number of inorganic salts such as chloride, nitrate and sulfate, which have an important impact on the hygroscopicity of environmental aerosols. At present, there have been a lot of studies on the hygroscopicity of inorganic salt aerosols. NaCl and (NH4)2SO4 are common components of atmospheric aerosols, and are also commonly used inorganic salts in moisture absorption experiments. Researchers often use NaCl or (NH4)2SO4 to calibrate the instrument or verify the reliability of the instrument. In addition to NaCl and (NH4)2SO4, there are many studies on inorganic salts such as Na2SO4, NaNO3, KCl and NH4NO3[6,47,53,54].
Organic matter is an important component of aerosols. Compared with inorganic salts, the hygroscopicity of some water-soluble organic matters has been studied more. Some organic substances have definite DRH and ERH, but some non-crystalline substances such as proteins and carboxylic acids do not have DRH and ERH[20]. The range of hygroscopicity parameter κ of organic matter is wide, and the hygroscopicity of different organic components is significantly different. The DRH of carboxylic acids such as oxalic acid, succinic acid and adipic acid is much higher than 95%, and it is difficult to observe their deliquescence in general deliquescence experiments[6]. Malonic acid, on the other hand, has no DRH and begins to absorb moisture continuously at low RH[55]. The hygroscopicity of amino acids varies greatly, and the hygroscopicity of amino acids is largely related to their water solubility[56,57]. Organic compounds such as humic acid, fulvic acid, organosulfate, levoglucose, mannan and galactan do not crystallize in weathering experiments and remain in the form of droplets with high concentrations at RH as low as 5%[26,58,59].
Mineral dust is the second most abundant aerosol in the troposphere. It has a long atmospheric lifetime and can be transported thousands of kilometers. It is an important component in the atmosphere that can not be ignored[60,61]. The hygroscopicity of mineral dust aerosol is weak, and most of them form adsorbed water on the surface. The CaCO3 in the form of calcite is the most commonly studied mineral dust aerosol. It has very weak hygroscopicity: when RH increases to 85%, the CaCO3still does not hygroscopic growth (average G (RH) = 1.00 ± 0.02), and one to three layers of molecular adsorbed water are formed on its surface[62]. Similarly, common mineral dust aerosols such as illite, kaolinite, and montmorillonite are not hygroscopic (κ < 0.05)[63]. However, the hygroscopicity of mineral dust aerosol will be greatly changed after aging[2].
BC is generally considered to be a non-hygroscopic substance (κ = 0). After atmospheric aging, the moisture absorption capacity of aerosols containing BC will increase. For example, sulfuric acid coated BC aerosol G (98.4%) increased from 1.15 to 1.91 with higher organic carbon content (from 32% to 78%)[5].
Bioaerosols are a special class of aerosols derived from organisms, including pollen, bacteria, viruses, algae and cyanobacteria, fungal spores and debris, biological crusts and lichens[64]. Some pollen, fungal spores, and bacteria are hygroscopic and can be activated into cloud droplets at low supersaturation levels[64]. Pope et al. Conducted hygroscopic experiments on four common pollens in northern Europe, and found that these four pollens have moderate hygroscopicity, and their surfaces are moist at high unsaturated humidity, so they are easy to become cloud condensation nuclei[65]. Chen et al. Found that there was no significant difference in hygroscopicity between the pollen widely existing in China and the common pollen in Europe and America[66]. Despr DesprÉs et al. summarized the hygroscopic data of more than a dozen fungal spores, and found that the hygroscopic ability of these fungal spores was weak, and the hygroscopicity growth factor was only 1.05 ~ 1.16 at high RH (> 90%)[64]. Lee et al. studied the hygroscopicity of Escherichia coli and Bacillus subtilis aerosols, and the results showed that when the RH exceeded 85%, the aerosols increased significantly, and the hygroscopicity growth factors were 1.34 and 1.16, respectively. The difference in hydroscopicity was determined by both the cell wall structure and the number of soluble components[67].
3.3 The effect of multicomponent coexistence
Aerosol in real atmospheric environment is usually a mixture of multiple components. Compared with the hygroscopic characteristics of single component aerosol, the mixing state of multi-component aerosol and the interaction between different components will have a complex impact on the hygroscopicity of aerosol.
Previous studies have found that for homogeneous internal mixed aerosols composed of inorganic salts and organic matter, organic matter can promote or antagonize the moisture absorption of particulate matter. In some studies, it was found that organic matter inhibited the hygroscopicity of aerosols: for example, Lei et al. Found that 4-hydroxybenzoic acid had a significant effect on the ERH of (NH4)2SO4, and the G (RH) of mixed aerosols decreased significantly with the increase of the mass fraction of organic matter[36]. When Estillore et al. Mixed organosulfate with NaCl, the ERH and DRH of the mixed aerosol decreased compared with those of the pure NaCl aerosol, and the G (RH) also decreased[59]. However, some studies have found that some organic compounds can promote the hygroscopic growth of aerosols, such as Darr et al., who found that hydrophilic glycine promotes the hygroscopic growth of mixed aerosols[56]. Luo et al also showed that aspartic acid, glutamic acid and serine have a significant impact on the phase state of (NH4)2SO4, and may have a potential impact on the hygroscopic behavior of submicron aerosols[57]. The mass ratio of inorganic and organic has an important influence on the moisture absorption of aerosols[68]. Jing et al. mixed water-soluble organic matter such as oxalic acid, levoglucose and humic acid with KCl for moisture absorption experiments. The higher relative mass fraction of water-soluble organic matter will promote the moisture absorption of mixed particles at medium and low RH, while the moisture absorption of aerosol at high RH will be inhibited when oxalic acid is dominant, while a small amount of levoglucose will make the aerosol grow in advance[54].
Surfactants are an important class of organic compounds that affect the hygroscopic properties of aerosols. Surfactant molecules tend to migrate to the droplet surface, encapsulating the water nucleus and changing the physicochemical properties of the aerosol[69]. Zelenyuk et al. Mixed soluble surfactant with two kinds of hygroscopic salts, and the results showed that the surfactant could make the mixed aerosol retain water after weathering[69]. Alshawa et al. Conducted hygroscopic experiments on NaCl aerosol coated with surfactant AOT and found that a single layer of AOT coating could reduce the DRH of NaCl and inhibit hygroscopic growth. When the AOT coating covers more than one monolayer, no obvious deliquescence change is observed in the aerosol, which is manifested by continuous moisture absorption[70]. Wang et al. Studied the hygroscopicity of core-shell aerosols with different sizes of ammonium sulfate core and phthalic acid coating thickness, and found that the G (RH) of core-shell aerosol was lower than that of homogeneous mixed aerosol before 80% RH; When RH > 80%, G (RH) decreases with the increase of shell thickness, but G (RH) of core-shell aerosol is higher than that of homogeneous internal-mixed aerosol with the same mass fraction[71].
After a complex atmospheric aging process, hydrophobic aerosols can be transformed into aerosols with a certain hygroscopicity. The hygroscopicity of mineral dust aerosol and BC increases after atmospheric aging[2,4,5]. The hygroscopicity of organic aerosols is also affected by heterogeneous reactions, and the hygroscopicity of oleic acid, arachidonic acid, and maleic acid particles after ozonation is enhanced compared with that of unreacted aerosols[15,72]. However, Slade et al. Found that κ of levoglucose granules did not increase significantly after aging[73]. In addition, Schwier et al. Used ozone to oxidize sodium salt and oleic acid (or sodium oleate) mixed aerosol, and found that the hygroscopicity of aerosol decreased after oxidation[74].
The hygroscopic and deliquescent weathering behavior of multi-component aerosols is more complex than that of single-component aerosols, and the effects of different organic compounds on the hygroscopicity of mixed aerosols and the effects of atmospheric aging process on the hygroscopicity of aerosols need to be further explored.
4 Field observation of aerosol hygroscopicity
The chemical composition of environmental aerosols is very complex, and atmospheric physical and chemical processes can also affect the particle size, composition and mixing state of aerosols, thereby affecting the hygroscopic capacity of aerosols. At present, many researchers have carried out ground and vertical observations of aerosol hygroscopicity in urban, rural, forest, ocean and polar regions.
4.1 Urban area observation
City is the environmental area most affected by human activities. Man-made sources such as motor vehicles, coal combustion, industrial production and dust emit a large number of aerosols and gases with complex chemical composition into the atmosphere. The chemical composition of aerosols from different sources may be quite different, thus showing a large difference in hygroscopicity. Kim et al. Used H-TDMA and Cloud condensation nucleus counter (CCNC) to observe in Seoul, South Korea. The κ of 30-150 nm aerosol was 0.10, 0.25 and 0.19 respectively in three pollution periods dominated by different pollution sources. This is because the aerosol emitted by local emission sources is mainly organic matter, and the aerosol emitted by long-distance transport sources is mainly inorganic matter[7]. Wu et al. Observed two biomass burning events in Beijing area, and the aerosol κ values calculated from H-TDMA, CCNC data and aerosol mass spectrometer data were similar, ranging from 0.09 to 0.10, and basically did not change with particle size[8]. The hygroscopicity and cloud condensation nodule activity of aerosols are low due to the high content of organic matter and BC in freshly discharged biomass aerosols[8]. Chueng et al. Observed in Hong Kong that the chemical composition of aerosol can better predict the hygroscopicity, and the difference between the predicted κ and the measured value is within 10%[75].
City is a haze-prone area, and the hygroscopicity of aerosol will change significantly with the pollution process. Under the same RH condition, the value of f (RH) in the pollution period is higher than that in the clean period[76]. The average f (80%) of Beijing PM10 and PM1 in the pollution period, dust period and new particle formation period is about 1.75, 1.15 and the scattering Ångström index of 1.60,PM10 and PM1 indicates that the hygroscopicity of fine particles is greater than that of coarse particles[9]. Aged aerosols may be in an internal mixing state, which enhances the hygroscopicity of aerosols. Ma et al. Used CCNC to observe Jiangxin Island in Nanjing in summer, and the results showed that the κ of four types of aerosols, including biomass burning, urban, marine and industrial sources, did not change much (0.34 ~ 0.39, with an average of 0.35), which may be due to the fact that most of the aerosols during the observation period were fully aged aerosols, showing the characteristics of internal mixed aerosol[77].
Due to the influence of human activities and meteorological conditions, the hygroscopicity of aerosols often shows diurnal variation. The f (RH) in the urban area of Granada, Spain, is mainly affected by local traffic sources and meteorological conditions, such as the increase of BC and mineral dust aerosol caused by the emission of traffic peaks in the morning and evening, when f (RH) reaches the lowest value in a day[78]. The κ of aerosols in the Jiangxin Island of Nanjing in the Yangtze River Delta also has a significant diurnal variation, which may be caused by the emission of traffic peaks in the morning and evening and photochemical aging[77].
4.2 Observation of hygroscopicity in rural and forest areas
Compared with urban areas, rural and forest areas are less affected by human activities and have relatively low pollution levels. Local natural sources may make a major contribution to aerosols in the region. For example, the southeastern United States is greatly affected by biogenic sources, mainly secondary organic aerosols produced by the oxidation of isoprene and terpenes, and most of them are in liquid state[10,11]. Thalman et al. observed 60 km downwind from Manaus, Brazil, and found that the background aerosol was mainly highly oxidized organic aerosol. The average value of background aerosol κ in rainy season was about 0.19, and the average value of κ in dry season was about 0.15, with little difference in hygroscopicity[79]. The PBAPs in the Xiaoxing'an Mountains of Heilongjiang Province, China, contributed 47% to the mass of the PM2.5. The PBAPs were studied by transmission electron microscopy and scanning electron microscopy. It was found that most of the PBAPs were bacteria and fungi, and their hygroscopicity was very weak (G (94%) was about 1.09)[12].
Zhang et al. observed with CCNC in Shanxi and found that when the organic component in the aerosol is more than 60%, the average κ value of the aerosol is 0.26, while when the organic component is less than 40%, the aerosol is mainly composed of inorganic salts, and the κ value can reach 0.50[80]. Wu et al. Observed aerosol f (80%) with a humidified turbidimeter in the North China Plain for two months in summer and found that the aerosol f (80%) value was high (> 2), which may be due to the fact that water-soluble secondary inorganic salts accounted for 53.1% of the fine particle mass[81]. These two observations show that inorganic salts have an important impact on the moisture absorption of environmental aerosols. Field observations have found great uncertainty in the hygroscopicity of organic matter[83]. Due to different oxidation mechanisms, the relationship between hygroscopicity of organic matter and the ratio of oxygen to carbon is different, which may be closely related or unrelated[82,84,85].
4.3 Ocean and polar region hygroscopic observations
The ocean covers 71% of the earth's surface, and ocean spray makes an important contribution to the global aerosol emission flux[1]. Marine aerosol has strong hygroscopicity. Atwood et al. Observed with CCNC on the California coast in the western Pacific Ocean, where the marine source type aerosol κ was 0.30 ~ 0.49, which was about twice as hygroscopic as the terrestrial source aerosol (0.15 ~ 0.25)[86]. Similarly, Phillips et al. Observed with H-TDMA on the East Coast of the United States and found that the κ (0.37 ~ 0.54) of marine aerosols in the Atlantic Ocean was higher than that of continental aerosols[87]. Marine biological activity can greatly affect the ratio of organic matter to inorganic salts in marine aerosols. Kim et al. Observed that when marine biological activity is strong, aerosols contain a large number of biological organic compounds with low hygroscopicity[88]. The hygroscopicity of marine aerosols is also closely related to wind speed, for example, strong wind speed over the ocean can increase the hygroscopic sea salt aerosols[89]. After a long time of transmission, the hygroscopicity of marine aerosols may be inhibited by the interaction with mineral dust and pollutants, such as chlorine loss, which can reduce the hygroscopicity of marine aerosols[13].
The polar region has a special ecological environment and is highly sensitive to climate change. Many researchers have observed in the polar region and paid attention to the hygroscopicity of aerosols. Herenz et al. Used CCNC to observe Arctic aerosols in Northwest Canada, and the results showed that the mean κ of aerosols in this area was greater than 0.2[90]. Lange et al. Observed Arctic aerosol hygroscopicity with H-TDMA and CCNC on the northeast coast of Greenland, and obtained G (90%) and κ values, respectively. They found that the aerosol population (> 100 nm) dominated by anthropogenic sources had higher G (90%) (1.62 ~ 1.81) and lower κ values (0.10 ~ 0.46). The ultrafine aerosol population (< 100 nm) dominated by natural sources has lower G (90%) (1.38 ~ 1.70) and significantly higher κ values (0.35 ~ 0.51), while secondary organic aerosols are found to be the main contributors to Arctic ultrafine aerosols[91]. When Burkart et al observed aerosol condensation growth events in the Arctic region with CCNC in aerial survey, they found that the condensation of semi-volatile organic compounds promoted aerosol growth, and κ was 0.09 ± 0.03 at this time, which was consistent with the large κ of inorganic salt aerosol and the low hygroscopicity of marine organic components (κ was about 0.1 ~ 0.2)[92]. Kim et al. Used CCNC to observe the Antarctic Peninsula and found that most of the air masses originated from the ocean area (the South Atlantic Ocean and the South Pacific Ocean), and the polar aerosol was generally similar to the marine aerosol, with strong hygroscopicity, but low hygroscopicity during the formation of new particles[93].
4.4 Hygroscopicity vertical distribution
The position and hygroscopicity of the aerosol layer can affect the stability of the atmosphere and the radiation budget of the earth, but the existing aerosol hygroscopic observations are mainly concentrated on the ground, and the vertical distribution of hygroscopicity is still less.
Chen et al. Observed new particle formation events from the ground to 1000 m in the United States, and found that the κ of aerosols with a particle size of 13 nm can be as high as 0.6 ~ 0.9, and there is an obvious diurnal variation trend. The κ of 25 nm aerosols is 0.2 ~ 0.25, while the κ of other aerosols (with particle sizes of 50, 100, 200 nm) is below 0.2, which may be due to the fact that ultrafine aerosol contains a large proportion of strong hygroscopic substances[94]. When Shingler et al analyzed the hygroscopicity characteristics of aerosols based on high-altitude aerial survey (6.5 km) data, they found that the hygroscopicity of aerosols from biomass combustion sources was the lowest, with κ of only 0.11, but this was still slightly higher than some results of ground observations, possibly due to the atmospheric aging of aerosol[95][8][96]. The aerial survey results of Brock et al in the southeastern United States show that the f (RH) of mixed layer and transition layer aerosols increases slowly with the increase of altitude, and reaches the maximum in the lower free troposphere. At high RH, the growth trend of f (RH) from the mixed layer to the transition layer and then to the troposphere is consistent with the change of the dry mass fraction of the inorganic aerosol, for example, with the increase of the dry mass fraction of the inorganic aerosol from 0. 35 to 0. 45, f (RH) increases from 1. 75 to about 4[97]. Liu et al studied the vertical distribution of hygroscopicity using aerial survey data over Beijing (about 2.5 km), and found that the hygroscopicity of aerosols increases with altitude from the ground to the planetary boundary layer (κ is 0.20 ~ 0.34 at high surface RH,The vertical gradient varies as 0.13 km-1, and the κ gradient is about 0.05 km-1 at low surface RH), which proves that the water process plays an important role in the enhancement of aerosol hygroscopicity in the planetary boundary layer, and the hygroscopicity is mainly affected by secondary inorganic substances, both of which have similar trends[16]. Shingler et al. Also found that the κ of marine aerosols increases with altitude. However, the κ of aerosols from other sources did not change significantly, which may be related to the change of organic matter content in aerosols[95]. O 'Shea et al. Conducted aerial surveys (< 4000 m) on the Weddell Sea coast, Coats Land, Antarctica, and measured high aerosol κ up to 0.66[98].
It can be seen from the above that the difference of aerosol hygroscopicity in different regions is closely related to the emission source and aging process of aerosol. Highly hygroscopic marine spray aerosols generally remain relatively hygroscopic even after prolonged transport[13,87]. Freshly emitted mine dust aerosols are almost not hygroscopic, while fresh biomass combustion aerosols and traffic emission aerosols are weakly hygroscopic[8,77,78,85,99]. The hygroscopicity of biogenic aerosols such as primary bioaerosols discharged freshly and isoprene discharged from biogenic sources is not strong[12,79]. After atmospheric aging, the chemical composition and mixing state of these aerosols will change, and the hygroscopicity will gradually increase[77,95,96,100]. Industry emits a large amount of aerosols and gases with diameters less than 10 μm or 2.5 μm, and high concentrations of precursors contribute to aerosol aging and enhance aerosol hygroscopicity[101]. When Hong et al observed in the suburbs of the Pearl River Delta, they found that the aerosols with weak hygroscopicity were mainly emitted from local emission sources, while the aerosols with strong hygroscopicity generally had a long aging time[85]. Li et al. Have proved that the hygroscopicity of aerosols is related to the aging process, and that different atmospheric aging processes day and night lead to different mixing States and hygroscopicity changes of BC[102]. Different oxidation mechanisms during aging affect aerosol hygroscopicity[82,84,85].
5 Effects of Aerosol Hygroscopicity on Atmospheric Chemistry, Visibility, Climate and Human Health
5.1 Effects of Aerosol Hygroscopicity on Atmospheric Chemistry
The hygroscopicity of aerosol directly affects the water content and phase state of aerosol, and affects the surface and bulk phase of aerosol, which can be used as a reaction site, thus affecting the heterogeneous chemical reaction and the chemical composition of aerosol. Many studies have shown that RH plays an important role in various inorganic aerosol-related atmospheric chemical processes. With the increase of RH, the adsorbed water on the aerosol surface increases, and even the aerosol is nearly liquid at high RH, which can significantly promote the occurrence of heterogeneous chemical processes. It is found that water plays an important role in the interfacial chemical reaction of NaCl. When the gaseous HNO3 reacts with NaCl particles, the surface of the aerosol without surface adsorbed water is slowly passivated by the formation of nitrate, while the aerosol with surface adsorbed water has the regeneration of reactive sites[103]. Bertram et al. Studied the heterogeneous reaction of N2O5 with inorganic-organic mixed aerosol. The experimental results showed that the effect of liquid water on the uptake coefficient was determined by both RH and the hygroscopicity of aerosol components. When RH was lower than 50%, RH had a significant effect on the reaction, and the uptake coefficient increased with the increase of RH. When the ratio of liquid water to nitrate is less than 20, nitrate can inhibit the uptake coefficient of aerosol. Trace chloride can reduce the uptake coefficient, which is affected by high nitrate load[104]. Field observation studies also found that with the increase of RH, the chemical composition of aerosols changes, and the hygroscopic secondary substances such as sulfate and nitrate increase, which further leads to the enhancement of aerosol hygroscopicity[14,16].
Environmental RH has an important influence on the formation of secondary organic aerosols. Trainic et al. Studied the reaction of glyoxal on (NH4)2SO4 aerosol at different RH, and found that glyoxal oligomers and carbon-nitrogen products could be formed, and the aerosol particle size and extinction cross-section increased significantly[105]. The reaction rate increases with the increase of RH at low RH, and RH affects the reaction rate and products[105]. At low RH, the reaction is carried out by the adsorbed water of the aerosol, and the reaction rate of the weathered aerosol is lower than that of the deliquescent aerosol at the same RH because the water content of the aerosol in the weathering process is higher than that of the deliquescent aerosol[105]. Lee et al. focused on the effects of RH and ozone on the hygroscopicity of oleic acid, arachidonic acid and maleic acid, and found that the phase change of aerosol hygroscopicity had a great impact on the oxidation reaction: oleic acid was liquid at room temperature, so ozone was easy to diffuse into the aerosol, and the phase state and oxidation reaction of oleic acid aerosol were not affected by RH; Maleic acid and arachidonic acid will affect the chemical reaction due to phase change, while the reaction of solid aerosol with ozone and water is limited to the aerosol surface[15].
5.2 Impact on visibility
Hygroscopicity affects the optical properties of aerosols in ambient conditions, which in turn affects atmospheric visibility[18]. Gibson et al.'s Experiment showed that with the increase of RH from 10% to 85%,Ca(NO3)2, the aerosol particle size doubled, that is, the volume increased by 8 times, the single scattering albedo increased, and the aerosol scattering coefficient at 4000~7000 cm-1 was three times that at dry state[62]. Sulfuric acid coated BC has a scattering enhancement of 10 times and an absorption enhancement of nearly 2 times at 80% RH[106]. Chen et al. Analyzed the field observation data in the North China Plain and found that when RH > 90%, the visibility reduction was mainly affected by RH, which could be expressed by an empirical formula with two parameters (RH and aerosol volume concentration) or three parameters (RH, aerosol volume concentration and the ratio of coarse and fine aerosols)[107]. Liu et al. Showed that RH is an important factor in visibility reduction, and proposed an empirical formula of atmospheric visibility based on PM2.5 and f (RH)[108]. Deng et al. Have proved that the main reason for the reduction of visibility at high RH is the growth of aerosol moisture absorption[109]. When Qu et al. Observed in the eastern coast of China (Qingdao and Zhejiang), they found that the mass fraction of hygroscopic substances (such as sulfate and nitrate) in aerosols increased at higher RH in this area, thus affecting atmospheric visibility[14].
5.3 Impact on climate
Hygroscopicity affects climate by affecting the direct and indirect radiative effects of aerosols. The direct effect of aerosols is to cool or heat the atmosphere by scattering and absorbing radiation, while the indirect effect of aerosols is to affect climate by affecting cloud droplet size, cloud albedo, cloud lifetime and coverage[110]. Quantification of the effect of aerosols on radiative forcing is one of the largest sources of error in global climate predictions[111].
Except for black carbon and brown carbon, most aerosols have a cooling effect on the climate[111]. Zieger et al. Corrected the hygroscopicity of sea salt aerosols. When κ decreased from 1.5 to 1.1, the reduction of aerosol optical depth over the ocean was as high as 15%, but the number of cloud condensation nuclei was not significantly affected, resulting in a global aerosol radiative forcing change of -0.22 W · m-2[50]. As mentioned earlier, organic compounds can affect the hygroscopic growth of inorganic aerosols. When NaCl is internally mixed with 3% (or 50%) dissolved organic matter, the radiative cooling of the mixed aerosol at RH 85% is 3% (or 25%) lower than that of the pure NaCl aerosol[112]. Aged BC has hygroscopicity, and its optical and radiative properties will change significantly. With the increase of RH, the scattering of aged BC aerosol increases due to the increase of hygroscopicity, but the optical absorption does not increase significantly. Hygroscopic growth not only causes negative radiative forcing at the top of the atmosphere, but also slightly weakens absorption within the atmosphere[113].
Wang et al. Have shown that the indirect radiative effect of aerosols plays a leading role in the climate impact of aerosols[17]. Especially, the κ of organic aerosols has a rather wide range of variation, which leads to a large uncertainty in the prediction of global aerosol indirect radiative forcing. Liu et al. First studied the global aerosol indirect forcing caused by the uncertainty of organic hygroscopicity, and the results showed that the predicted CCN concentration could increase by 40% ~ 80% when the κ of primary organic aerosol increased from 0 to 0.1, while the predicted CCN concentration could increase by 40% when the κ of secondary organic aerosol increased by 50% (0.14 ± 0.07)[114]. Because the hygroscopicity of organic matter has a greater impact on pre-industrial CCN and cloud droplet number concentrations, the global annual mean anthropogenic aerosol indirect radiative forcing decreases (from − 1.3 to -1.1 W·m-2) as the hygroscopicity of organic matter increases[114]. Rastak et al. Showed that when the κ value of organic aerosol ranges from 0.015 to 0.05, the aerosol radiative forcing can change up to -1.02 W·m-2, while the artificial aerosol radiative forcing after industrialization is -1 W·m-2, which proves the importance of hygroscopicity of organic aerosol[115].
5.4 Impact on human health
Long-term exposure to PM2.5 is closely related to respiratory diseases, lung cancer and cardiovascular diseases, affecting human health[116,117]. When inhalable aerosols enter the human body through the respiratory tract, highly toxic substances such as polycyclic aromatic hydrocarbons attached to the surface of inhalable particles will also enter the human body[118]. Aerosols will change their particle size due to hygroscopic growth, while inhalable aerosols (diameter < 10 μm) will deposit in the respiratory system through inertial impact, gravitational sedimentation and diffusion.The deposition state is highly correlated with the particle size: coarse particles (diameter > 3 μm) are mainly deposited in the oral cavity and the upper respiratory tract such as the pharynx, larynx, trachea and bronchus, and 30-300 nm aerosols are mainly deposited in the alveolar region[119]. Smaller particles can precipitate in the lungs or alveoli, and even enter the blood, such as exogenous ultrafine particles detected in human serum and pleural effusion[120].
When the aerosol passes through the moist respiratory tract (RH > 90%) and the lungs (RH close to 99.5%), it may absorb moisture and increase, and the particle size will change, thus affecting its precipitation position in the respiratory tract[19]. According to the research of Winkler-Heil et al., the deposition trajectory of inhaled aerosol is related to G (RH), and the lung deposition rate of aerosol with strong hygroscopicity is lower than that of hydrophobic aerosol with the same particle size[121]. For example, the deposition fraction of hygroscopic submicron aerosols in bronchi and Airways is smaller than that of the corresponding hydrophobic aerosols with the same particle size. The precipitation of hygroscopic micron-sized aerosols increased in the bronchi and decreased in the peripheral alveolar Airways[121]. Predicted lung deposition of hygroscopic aerosols can be modeled based on the D (dry) and G (RH) of the particles to better reveal their health effects.
6 Conclusion and outlook
Field observations, laboratory studies, models, and theoretical analyses have advanced the understanding of aerosol hygroscopicity. In this paper, the research methods, main influencing factors, spatial and temporal distribution characteristics of aerosol hygroscopicity, and its effects on atmospheric chemistry, environmental climate and human health are summarized and analyzed. Aerosol hygroscopicity still has challenges in research methods, technical methods and vertical observation, and more in-depth research is needed. (1) The improvement of observation instruments and the change of observation focus. On the one hand, it is necessary to improve the RH stability of the research instrument, so that it can be used stably at a higher RH and can be measured near the saturation point; On the other hand, there are few experimental studies on the moisture absorption of nano-particles, and relevant research methods need to be developed to deeply analyze the internal mechanism of the unique moisture absorption growth and deliquescence weathering of nano-aerosols. (2) It is necessary to further reveal the changes of phase state, viscosity and surface tension of different hygroscopic aerosols, the deliquescent weathering behavior of multi-component mixed aerosols, especially the effects of atmospheric oxidation, heterogeneous process and phase separation on hygroscopicity in inorganic-organic composite system, so as to promote the establishment and optimization of parameterization and the application of models. (3) At present, there are a lot of aerosol hygroscopic data observed on the ground, but the vertical distribution data of hygroscopicity is relatively scarce, that is, the observation of vertical distribution of hygroscopicity needs to be increased. (4) The impact of hygroscopicity on the climate, environment and health effects of aerosols needs to be further explored, such as the hygroscopic process of viruses and bacteria in the atmosphere and their impact on their transmission.The effect of hygroscopicity on the radiative forcing of black carbon and brown carbon aerosols, the deposition of different hygroscopic aerosols in lungs and other organs, and the impact on human health.
KreidenweisS M, Asa-AwukuA. Aerosol Hygroscopicity: Particle Water Content and Its Role in Atmospheric Processes. In Treatise on Geochemistry, 2014: 331.
PDF
