Atmospheric heterogeneous reactions on soot: A review

Abstract

Soot particles, composed of elemental carbon and organic compounds, have attracted widespread attention in recent years due to their significant impacts on climate, the environment and human health. Soot has been found to be chemically and physically active in atmospheric aging processes, which leads to alterations in its composition, morphology, hygroscopicity and optical properties and thus changes its environmental and health effects. The heterogeneous reactions on soot also have a significant impact on the transformation of gaseous pollutants into secondary aerosols. Therefore, the interactions between soot and atmospheric substances have been widely investigated to better understand the environmental behaviors of soot. In this review, we systematically summarize the progress and developments in the heterogeneous chemistry on soot over the past 30 years. Atmospheric trace constituents such as NO₂, O₃, SO₂, N₂O₅, HNO₃, H₂SO₄, OH radical, HO₂ radical, peroxyacetyl nitrate etc., are presented in detail from the aspect of their heterogeneous reactions on soot. The possible mechanisms and the effects of environmental conditions on these heterogeneous reactions are also addressed. Further, the impacts of the heterogeneous reactions of soot on the atmospheric environment are discussed, and some aspects of soot-related research which require further investigation are proposed as well.

Graphical Abstract

1. Introduction

Massive amounts of soot particles, originating from incomplete combustion of fossil fuel and biomass burning [1], are emitted into the atmosphere every year. Globally, the total soot annual emission was estimated to be about 8.54 Tg in 2017, which was about 20% higher than the estimated emission of 1960. The estimated results still own a great uncertainty due to the high degree of uncertainty associated with emission factors [2]. Industrial processes and urbanization may further increase soot emissions over a period of time since soot emissions have not yet been the focus of sustained policy attention [3].

Soot particles have attracted great concerns in recent years due to their impact on the environment, climate and human health. It has been reported that soot is partially responsible for air pollution [4] and increases in droughts and floods over the last two decades [5]. What's more, it is widely recognized that soot particles play a remarkable role in climate change through absorbing and scattering solar radiation [6]. As air pollutants, soot particles also have harmful impacts on human health owing to particulate irritation and the existence of pernicious substances on soot surfaces such as polycyclic aromatic hydrocarbons (PAHs) [1].

Soot particles exhibit a huge surface area of ∼100 m² g⁻¹ which provides abundant sites for heterogeneous reactions [1]. Gaseous species such as O₂ [7], NO₂ [8], [9], [10], and SO₂ [11,12] can react with soot in the atmosphere. These heterogeneous reactions can be an effective pathway for clearing the gas pollutants and generating secondary products through uptake and conversion [13]. The redox [14], [15], [16] and catalytic [4] activity of soot can also make some reactions that take place with difficulty in the gas phase occur more easily on the surface of soot. Consequently, it not only promotes the removal of some gas-phase species, but also leads to the generation of new species [13]. Meanwhile, the soot surface will be modified by the heterogeneous reactions, and subsequently the physicochemical properties such as composition, hygroscopicity, morphology and toxicity will also be significantly changed [17,18].

Previous reviews of the heterogeneous reactions of soot have predominantly concentrated on uptake kinetic processes, reaction products and the effects of environmental conditions on the reactions. Nienow and Roberts [19] provided the first relatively comprehensive review in this field, summarizing the reactions of carbon aerosols with abundant gases such as O₃, SO₂, O₂, HNO₃ and NO₂ in 2006. Indarto [20] pointed out in his review that soot-NO₂ and soot-HNO₃ reactions are important as HONO sources. Besides, Ma et al. [13] briefly discussed the reactions of O₂, O₃ and NO₂ with black carbon (BC) and further summarized the specific role of organic carbon (OC). After the publication of these comprehensive reviews, much advancement has been made in this field. First of all, more trace active species such as OH radicals [21], peroxyacetyl nitrate (PAN) [22], N₂O₅ [23,24] and so on were studied. In addition, the research focus gradually deepened from the study of the phenomenon to the reaction mechanism, and the research objects became closer to real soot. Furthermore, more and more studies are beginning to pay attention to the changes in the composition and nature of the soot itself during heterogeneous reactions [17,18].

In some developing countries, such as China and India, air pollution is highly complex; strong homogenous nucleation and multiphase heterogeneous processes coexist, coupling with strong atmospheric oxidizing capacity and O₃ pollution [25]. Soot may have a very important effect on the evolution of pollution under such complicated pollution conditions. In this review, we provide a systematic summary of the progress and achievements in the heterogeneous reactions of soot since 1990. Carbonaceous substances are often called “soot,” “black carbon,” or “carbon black,” which has been the subject of some debate [17], and this review will not make a strict distinction. The focus of this review is to summarize all important species that were reported to react heterogeneously with different kinds of soot. On this basis, we illustrate the possible mechanisms of the reactions and how these reactions vary under different environmental conditions. Finally, the impacts of the heterogeneous reactions occurring on soot on the atmospheric chemistry are proposed and discussed. We hope this review can provide a detailed reference for future research on soot.

2. Characterization of soot

2.1. Morphology

The morphological characteristics of soot from different sources measured by SEM or TEM are shown in Fig. 1. Most kinds of soot possess chain-like (Fig. 1(b, c, d, e, g, h, i, j, k, l, m, o)) or sheet-like (a, f, n) agglomerated structures. Relatively big particle aggregates may be composed of small clusters of a variety of different shapes and sizes [26]. Smaller clusters often show spherical or irregular shapes, as we can see from Fig. 1(b, c, d, e, g, h, i). The mean particle sizes of different soot particles range from 15 to 66 nm. The type of soot, the generation process, as well as the aging process will affect the size of soot. For example, the average diameters of primary and secondary diesel soot particles are 15 ± 4 and 49 ± 13 nm, respectively [27].

Fig. 1.

SEM of (a) GTL (gas-to-liquid) soot, (b) kerosene soot, (c) toluene soot, (d, e, f, g, n) diesel soot, (h) chimney soot, and (i) carbon black. TEM of (j) n-hexane soot, (k) toluene soot, (l) Special Black 4A(SB4A), (m) n-hexane soot, and (o) Printex U soot. (j, k) are from Al-Abadleh and Grassian [28], (a, f) are from Tapia et al. [26], (b, c, g) are from Helen and Andrew [27], (d) is from Arens et al. [29], (l, m) are from Jiang [30], (n, o) are from Tighe et al. [31], and (e, h, i) are from Fernandes et al. [32].

2.2. Elemental analysis

Soot is usually composed of four elements: carbon, hydrogen, oxygen and nitrogen. Table 1 summarizes the four-element contents of different kinds of soot in different studies. There is no doubt that the carbon content is the highest, basically above 92%, followed by hydrogen and oxygen, while the amount of nitrogen is the least. Carbon is the one and only element in elemental carbon (EC), while carbon, hydrogen and oxygen are the dominant elements of OC. Nitrogen is mainly from the interaction between soot and nitrogen-containing species in the air during or after combustion.

Table 1.

List of elemental analysis results in the literatures.

Soot-producing fuel	C (wt %)	H (wt %)	O (wt %)	N (wt %)	Reference
Decane grey	97.27±0.05	0.83±0.04	1.65±0.19	0.20±0.18	[33]
Decane black	96.39±0.22	0.19±0.01	3.22±0.25	0.27±0.09	[33]
Hexane (λ = 0.82)	94.73±0.15	1.50±0.01	3.78±0.16	0	[34]
Hexane (λ = 0.16)	93.01±0.31	0.77±0.03	7.27±0.42	0.26±0.02	[34]
Hexane (λ = 0.09)	92.03±0.34	0.44±0.07	5.99±0.36	0.23±0.18	[34]
Printex U	92.2±0.5	1.3±0.1	-	0	[31]
JM soot	82.9±0.5	1.7±0.1	-	0	[31]
Decane	95.1±0.5	-	4.6±0.5	0.3±0.1	[10]
Hexane	93.0±0.4	-	6.8±0.5	0.2±0.2	[10]
Benzene	92.8±0.5	-	6.9±0.4	0.3±0.1	[10]
Diesel	87.9	0.9	10.8	0.4	[35]
Toluene	89.80±0.27	7.00±0.10	3.20±0.26	0	[27]
Kerosene	95.20±0.46	1.30±0.07	3.50±0.48	0	[27]
Diesel	85.10±1.30	9.10±0.17	5.80±1.40	0	[27]

2.3. Composition

In soot, the carbon skeleton is usually embedded with abundant oxygen-containing groups and complex organic compounds, such as saturated and unsaturated hydrocarbons, alkanoic acids, substituted aromatics, alcohols, ketones, PAHs, and their derivatives [36]. The Fourier Transform Infrared spectroscopy (FTIR) was usually used to characterize the functional groups of soot and vibration regions of fresh soot are summarized in Table 2. In conclusion, the surface groups of different types of soot almost all contain aromatic C-H, aliphatic C-H and C=O groups with different intensities. In addition to the structure and surface functional groups, many organic compounds were also identified in the extracts of different soot. Romanias et al. [16] separated seven different classes of PAHs and identified more than two hundred compounds on soot samples. The type and content of these bound PAHs may have a significant influence on the reactivity of soot and will be discussed in section 3.

Table 2.

The functional groups and vibration regions of fresh soot.

Soot type	Functional groups and vibration region (cm−1)	Reference
Diesel	C-OH/C-O (1100-1300), C=C (1500-1600), C=O (1650-1750), C-H (2700-3000), O-H (3000-3700)	[37]
n-Hexane	-CH2 (1440), C=O (1596), aromatic C-H (3038), alkyne C−H (3284)	[5]
Methane	C-O (1260,1350)	[38]
GTL soot	C=C (1592), C=O (1723), aromatic C-H (3029), alkane C-H (2941-2840)	[26]
Printex U	aromatic C-H (3033), alkane C-H (2953-2844), carboxylic C=O (1716)	[26]
Graphite	C=O (1718), C-O-C (1333)	[39]
Diesel	aromatic C-H (3050), carboxylic C=O (1720), aromatic C=O (1600)	[40]
Hexane	aromatic C-H (3050), carboxylic C=O (1720), aromatic C=O (1600)	[40]

3. Heterogeneous reactions of soot with the atmosphere

3.1. Experimental techniques for soot heterogeneous reactions

3.1.1. Experimental instrument

Heterogeneous reactions of soot have usually been studied in a Knudsen cell, a flow tube or an aerosol chamber, which are always connected to mass spectrometry, FTIR, and gas analyzers. Besides, off-line and in-situ methods, including scanning electron microscopy (SEM), transmission electron microscopy (TEM), GC-MS, Raman spectroscopy, etc. have also been widely used. Multi-angle studies of gas uptake kinetics, gas-phase products, and the composition and structure of soot will contribute to a full understanding of the heterogeneous reactions of gases with soot. Three commonly used reactors are briefly introduced as follows:

Knudsen Cell: The general design of Knudsen cell was described in detail in previous studies [41,42]. Briefly speaking, the Knudsen cell is a low-pressure flow reactor which consists of a chamber with an isolated sample compartment. Orifices on the chamber act as an outlet channel for reactant gases and product species to escape from the chamber and then to be detected usually by a mass spectrometer. Under low pressure reaction conditions, the mean free path of gas molecules is longer than the dimensions of the cell, thereby reducing the chance of collisions between the gas molecules and eliminating the boundary layer effects. As a result, the analysis of the data is greatly simplified [43]. However, since Knudsen cell requires a vacuum environment and cannot provide conditions close to the actual atmosphere, this method is being used less and less in recent years.

Flow Tube: The flow tube designed to study the gas-solid heterogeneous chemistry kinetics is usually a coated-wall flow tube reactor [44,15,16]. The inner walls of the tube are usually coated by samples. The reactant gases are introduced into the reactor with a movable injector. Therefore, the uptake of gases on the coated samples can be measured by adjusting the positions where the gases were injected. Flow tubes can be operated under ambient pressure which is favorable to obtain kinetics relevant to real atmospheric conditions. The volume of flow tubes is relatively small, resulting in short airflow residence time in the reactor and limiting long-time simulations.

Smog Chamber: A smog chamber is a reactor which usually consists of an enclosed reactor, inlet system, light sources, temperature control system, monitoring system and auxiliary system [45]. The tested sample are usually aerosolized when being introduced into smog chamber. Compared with Knudsen cell and flow tube, it simulates atmospheric processes more realistically, but meanwhile, relatively complex processes will take place in the chamber. Therefore, the conditions of chamber experiments should be carefully designed to constrain heterogenous reactions as the main process and many factors such as wall effects need to be quantitated.

3.1.2. Kinetics calculation

γ: the uptake coefficient, as an important physical and chemical parameter, is widely used to characterize the heterogeneous reactions of gases on soot. γ is determined as follows:

$$\gamma =\frac{\text{the}\text{number}\text{of}\text{molecules}\text{taken}\text{up}\text{on}\text{the}\text{surface}\text{per}\text{second}}{\text{the}\text{total}\text{number}\text{of}\text{gas}-\text{surface}\text{collisions}\text{per}\text{second}}$$

The uptake of trace gas on the surface of atmospheric particles is a gradual saturation process. The maximum uptake coefficient measured in the initial stage of uptake is the initial uptake coefficient-γ_ini or γ₀ while the steady-state uptake coefficient-γ_ss characterizes the ability of particles to stably adsorb a certain gas in the atmosphere. In the calculation of the uptake coefficient, the surface area is a key parameter. The uptake coefficient is recorded as γ_bet or γ_g if the BET (Brunauer–Emmet–Teller) surface area or the geometric area of the particles is used, respectively.

3.2. Heterogeneous reactions of NO₂ on Soot

3.2.1. Uptake of NO₂ on soot

NO₂ is mainly distributed in the troposphere and plays an important role in atmospheric chemistry. The sink of NO₂ in the atmosphere involves a variety of processes such as nitrate aerosol formation and photochemical reactions. However, the removal pathways of NO₂, such as the heterogeneous reactions of NO₂ with soot, have not been fully understood. These reactions have received widespread attention in recent years because they are also potential sources of HONO [15].

The initial uptake of NO₂ on soot was found to be significant and the uptake rate of NO₂ on different soot surfaces varies greatly from 10⁻⁸ to 10⁻¹, as summarized in Table 3. Besides environment conditions, uptake coefficient (γ) was reported to be influenced by a series of factors such as NO₂ concentration, the soot mass [28] as well as the BET surface area and the surface composition of soot [46]. Due to the great uncertainty of γ, it is still difficult to incorporate the heterogeneous reactions of NO₂ on soot into air quality models to predict its environmental impact.

Table 3.

Summary of the initial uptake coefficients of NO₂ on soot.

Soot type	Reaction site	γg	γBET	Reference
FW2	Knudsen cell		(6.4±2.0) ×10−2	[52] [52] [52]
Printex 60
FS101
FW2	Knudsen cell	(1.1±0.4) ×10−1		[53]
Ethylene	Knudsen cell	1.2×10−1		[54]
Lamp Black 101	Knudsen cell		3×10−7	[48]
Hexane	Knudsen cell	5×10−5		[28]
Methane	Flow tube	3.0×10−4	2.5×10−5	[55] [55] [55]
Propane	Flow tube	3.0×10−4	3.2×10−6
Hexane	Flow tube	1.0×10−5	6.2×10−7
n-Hexane	Knudsen cell	7.22×10−3	5.68×10−5	[28]
Decane grey	Knudsen cell		1.0×10−1	[33]
Hexane black	Knudsen cell		1.0×10−1	[33]
n-Hexane	Chamber		(2.4±0.6) ×10−8	[46]
Kerosene	Flow tube		(5.0±2.0) ×10−5	[44]
Hexane	Flow tube		(2.9±1.2) ×10−5	[44]
Toluene	Flow tube		(4.0±1.6) ×10−5	[44]
Hexane	Flow-tube		(3.9±1.9) ×10−5	[10]
Decane				[10]
Benzene				[10]
Propane	Flow-tube	(1.9±3.7) ×10−5		[56]
Kerosene	Flow-tube	2×10−5		[56]
Ethylene	Flow-tube		(5.9±1.0) ×10−2	[57]
Methane	Flow-tube		(1.5±0.3) ×10−4	[15]
Bio-soot	Flow-tube		(2.72±0.54) ×10−6	[16]
Printex ex2-b	Knudsen cell		1.5×10−3	[58]
Diesel	Knudsen cell		3.6×10−3	[58]

In a related study on the heterogeneous reactions of NO₂ and soot, attention has mostly been paid to the products. HONO was found to be the main gaseous product [47,33,29], coinciding with the formation of NO [48,47,33] and even N₂O [49]. The gas-phase products and their yields (the ratio to NO₂ consumption) are summarized in Table 4. These yields are largely dependent on the types of soot and experimental conditions, which will be discussed in the next section. Besides gas-phase products, NO₃⁻ has also been found in the extracts of soot with ion chromatography [50,48]. Moreover, nitrogen-containing and oxygen-containing functional groups such as R-O-NO₂, R-N-NO₂, R-NO₂, R-ONO, and C=O have been widely observed on the surface of soot [8,51].

Table 4.

Summary of gas-product yields of NO₂ on soot.

Soot type	NO2 pressure (Torr)	HONO yield (%)	NO yield (%)	References
Kerosene		30 ± 5		[44]
Hexane	(0.08-1.00) ×10−4	96 ± 18		[10]
FW2	(0.08-8.00) ×10−3	0		[52]
Methane	(0.6-6.0) ×10−5	13 ± 5		[55]
Propane		10 ± 5		[55]
Hexane		30 ± 10		[55]
Kerosene		17 ± 5		[55]
Black 101	7.6×10−4-3.0×10−2	50-80		[48]
n-Hexane	7.8×10−6	36 ± 5		[28]
Toluene	(1-120) ×1014 molecules	68-99		[47]
Acetylene		2-6		[47]
Decane		58-95		[47]
Decane grey	2.2×10−6-2.6×10−4	100		[33]
Hexane black		A few percent		[33]
Propane	5×109-5×1010 molecules	49-76		[56]
Kerosene		52-75		[56]
Ethylene	5×1010-5×1012 molecules	30	18	[57]
Methane	1×1012–2×1013 molecules		50	[15]
Bio-soot		<20	80-100	[16]

3.2.2. Influence of reaction conditions

Fuel/oxygen ratio (λ): λ was found to be a key parameter affecting the properties of generated soot and its reaction activity, thereby affecting the yields of HONO and NO during the reaction of NO₂ with soot [33]. Generally speaking, the HONO yield from soot with λ higher than the stoichiometric ratio was as high as 100%, while the yield of HONO was greatly reduced on soot from a lean flame due to the faster decomposition of HONO [33]. NO would become the major product when λ was lower than the stoichiometric ratio. The soot produced with low λ usually contains less OC, resulting in fewer active sites of soot. Quantitatively, Han et al. [5] established linear relationships between OC content and HONO, NO yields: y_HONO = (11.60 ± 16.10) + (1.30 ± 0.40) × OC, and y_NO = (13.10 ± 1.90) − (0.20 ± 0.05) × OC, respectively.

Relative Humidity (RH): The role of water in the heterogeneous reaction of NO₂ and soot has been one of the most controversial topics. It was first believed that soot only played a catalytic role in the process of forming HONO, and 2NO₂ + H₂O → HNO₂ + HNO₃ was the main reaction, so the water was indispensable as a reactant in this process [55]. Nevertheless, this formation mechanism was questioned later, since HNO₃ was not observed [59,48]. Besides, the HONO yield was found to be higher than 50% [48], which argues against the disproportionation reaction mechanism. According to further research, the formation of HONO is suppressed under dry conditions [10], suggesting that H₂O is necessary for HONO formation. Moreover, Kleffmann et al. [48] and Longfellow et al. [55] both observed that the yield of HONO increased significantly with the increase of RH. Han [60] also found that the HONO yield increased slightly with the increase of RH, but the NO₂ uptake coefficient decreased. Moreover, above 40% RH, the HONO production decreased because of competition from H₂O adsorbing on the soot surfaces [59]. To explore the role of H₂O in HONO formation, Longfellow et al. [55] investigated HONO production with hydrogen isotopes (D₂O and H₂¹⁸O). When H₂O was replaced by D₂O, only DONO was produced, which suggested that water participated in the formation of HONO; when H₂O was replaced by H₂¹⁸O in the reaction atmosphere, only HONO was produced, but not H¹⁸ONO, which indicated that oxygen atoms in the water were not involved in the formation of HONO. According to other isotopic exchange experiments [44], the consumption of D₂O on soot led to the formation of HOD, indicating that there was an exchange reaction of H and D between D₂O and soot [61]. Therefore, the D in DONO may also come from the soot surface instead of H₂O. In addition, water was reported to reactivate soot after its quick deactivation in air [59,48]. However, no regeneration of reactivity was observed after humidification, in which case the reactivity decreased even further in Aubin's experiment [10]. Therefore, the role of water in the activity of soot, a key question for evaluating the influence of NO₂-soot reactions in the atmosphere, needs further study.

Temperature: A study of the heterogeneous reactions of soot and NO₂ was carried out at both ambient temperature and a high temperature relevant to the combustion process [31]. At low temperatures (298 K-470 K), HONO and NO are formed as gas-phase products as mentioned before; at higher temperatures (470 K-670 K) NO and CO₂ have been observed as gaseous products, indicating the oxidation and gasification of soot [57]. Reactions under high temperatures showed that NO₂ can bound to the soot surface either via the oxygen or via its nitrogen atom [51]. In addition, by the formation of an acidic functional group, a transition state on the soot surface was proved to exist [51], which decomposes into CO₂ and NO at 140 ℃. Most of the formed surface nitrogen-containing intermediates were not stable above 350 ℃ [35], and would decompose into CO₂ and NO. The oxidation of carbon was driven by the reduction reaction of NO₂, while some coexisting gases such as H₂SO₄, SO₃ and O₂ can also enhance the oxidation of soot [14,37].

Radiation: The soot-NO₂ interaction is enhanced in the presence of radiation compared to that under dark conditions. According to Gerecke et al. [54], deactivated soot could be reactivated by irradiation and regained 75% of its original activity. Monge et al. [62] also observed that γ of NO₂ on soot increased with the irradiation intensity, which led to more NO and HONO production. Sunlight can not only enhance the reaction between NO₂ and PAHs on soot, but also cause more HONO formation from the photolysis of newly formed nitro-compounds.

Coexisting substances: The real atmospheric environment is a mixture of multiple polluting gases and different kinds of aerosols. For example, industrial soot particles was observed to be mixed with mineral dust in field measurements [63]. However, knowledge on the reactivity of soot in the mixed system is still limited [64]. Romanias et al. [64] found that the reactivity of Al₂O₃ covered with kerosene soot was substantially increased, and the formation of HONO was enhanced compared with that on single pure Al₂O₃ or pure soot. At present, due to the complex mixing state of soot in nature, it is difficult to prepare real or close-to-real soot samples in the laboratory, which restricts the co-existing research of soot.

3.2.3. Reaction mechanism

The reaction mechanism of NO₂ and soot was mainly dominated by the redox mechanism [23,44,9]

$$\mathrm{N}{\mathrm{O}}_2+\text{Rred}\to \text{HONO}+\text{Rox}$$

Mechanisms proposed for interpreting the adsorption and desorption of NO₂ over carbonaceous materials are summarized in Table 5. Although these mechanisms have different details, the main processes are similar to a certain degree. Take the mechanism proposed by Stadler and Rossi [33] as an example: firstly, NO₂ is physically adsorbed on the soot site [S1] to form [S1•NO₂]. The [S1•NO₂] will flow between the soot surface sites until it contacts the [S2] site or [C-H]red site, then the reducing active sites [C-H]red and [S2] are consumed and begin to transform to the oxidative active sites [C]ox and [S2•NO₂], as shown in Eqs. 2-3. The adsorbed [HONO] formed on the surface of soot can be desorbed, producing gaseous HONO Eq. 4, or undergo a disproportionation reaction, producing gaseous NO and adsorbed [NO₂] and [H₂O] Eq. 5.

$$\mathrm{N}{\mathrm{O}}_2+\left[\mathrm{S}1\right]\to \left[\mathrm{S}1·\mathrm{N}{\mathrm{O}}_2\right]$$ (1)

$$\left[\mathrm{S}1·\mathrm{N}{\mathrm{O}}_2\right]+\left[\mathrm{S}2\right]\to \left[\mathrm{S}2·\mathrm{N}{\mathrm{O}}_2\right]+\left[\mathrm{S}1\right]$$ (2)

$$\left[\mathrm{S}1·\mathrm{N}{\mathrm{O}}_2\right]+\left[\mathrm{C}-\mathrm{H}\right]\text{red}\to \left[\mathrm{S}1\right]+\left[\text{HONO}\right]+\left[\mathrm{C}\right]\text{ox}$$ (3)

$$\left[\text{HONO}\right]\to \text{HONO}$$ (4)

$$2\left[\text{HONO}\right]\to \text{NO}+\left[\mathrm{N}{\mathrm{O}}_2\right]+\left[{\mathrm{H}}_2\mathrm{O}\right]$$ (5)

Table 5.

Mechanisms proposed for interpreting the adsorption and desorption of NO₂ over carbonaceous materials.

Adsorption reaction and desorption	Reference
(1) NO2 (g) + [C (+X)] +H2O → HONO (g) + [C(+X+1)] + OH− (2) NO2 (g) + H (ads.) → HONO (g)	[54]
(1) NO2 + soot → [S.NO2] (2) [S.NO2] → NO (3) [S.NO2] → [S.N] (4) NO2 + soot → HONO	[68]
(1) NO2 (g)⇋ NO2 (ads) (2) NO2 (ads) + H2O (ads) + soot (red) →HONO (ads) + soot (ox) (3) NO2 (ads) + soot (ox) → soot-ONO2 (4) HONO (ads) → HONO (g) (5) HONO (ads) + HONO (g) → NO (ads) + NO2 (ads) +H2O (ads) (6) NO (ads) → NO (g)	[48]
(1) -C* + NO2 → -C(NO2) (2) -C(NO2) → -C(O) + NO (3) -C(O) + NO2 → -C(ONO2) Below 523 K (4) -C(ONO2) → -C(O)+NO2 (5) -C(ONO2) → CO2+NO (6) -C(ONO2) → CO+NO2	[8]
(1) R + NO2 → R(NO2) (ads) (2) R(NO2) (ads) → R-NO2 → R-ONO (isomerization) (3) R-ONO → R-O* + NO→ ROx + carboxy-carbonates (4) R-O* + NO2 → R-ONO2 + NO (dispropertionation)	[35]
(1) NO2 (g) + Cred → [NO2.Cred] (2) [NO2.Cred] → NO(g) + Cox Above 470 K (3) Cox → CO2 (g) + Cred	[57]
(1) NO2 + [S1] → [S1•NO2] (2) [S1•NO2] + [C-H] red →[S1] + [HONO] + [C]ox (3) [HONO] → HONO (4) 2[HONO] → NO + [NO2] + [H2O] (5) [S1•NO2] + [S2] → [S2•NO2] + [S1]	[33]
(1) -C* + O2 → -C*(O) (2) -C*(O) → CO + * (3) -C*(O) + O2 → CO2 + * (4) -C* + NO → -C*(NO) (5) -C*(NO) →-C*(O)+ -C*(N)→ N2, -C*(NO) →-C*(NO2) →NO2, (6) -C*(NO) →-C*(NO2) →-C*(ONO2) →NO+CO2	[69]
(1) NO2 + [S1] → [NO2•S1] (2) [NO2•S1] + [C-H] red → [S1] + [HONO] + [C]ox (3) [NO2•S1] + [S2] → [S1] + [NO2•S2] (4) [HONO] → HONO (5) [HONO] → NO + other products	[70]

[S.NO₂]: NO₂ chemisorbed on the soot surface.

[S.N]: nitrogen permanently bound to the soot.

– C (*):an active site.

g: gas state.

ads: adsorption state.

ox: oxidation state.

red: reduced state.

S1, S2: sites on the soot surface.

The role of OC: As an important and complicated part of the soot, exploring the role of OC in heterogeneous reactions is a vital but hard task due to the complex composition of OC (Section 2.3). At present, there is still a dispute about the role of OC in the heterogeneous reactions of NO₂ and soot. Quiñone [56] found that pre-heating (300 °C) soot samples could significantly increase the NO₂ uptake, which can be interpreted as meaning that the removal of the volatile organic fraction from the soot backbone unlocked surface-active sites for NO₂. Similar conclusions were also mentioned, for example, that heating brought back the ability of soot to regenerate HONO, but at a smaller yield [54,64]. However, Han and his coworkers [5] found that preheated samples exhibited a great decrease in the uptake coefficient of NO₂ and the HONO yield due to the removal of OC from n-hexane soot. The main reason for the difference in the dependence of NO₂ reactivity on OC content may be that the soot contains different types of PAHs, which show different reactivities towards NO₂. The specific reasons need to be further explored in the future.

The role of EC: EC is an essential part of soot. Studies have shown that heating soot under a high-purity N₂ atmosphere at 300 °C can completely remove the OC on it without changing the structure of the EC [27,5]. Han et al. [5] found that even after treatment at 600 °C, the yield of HONO on the soot surface still reached 38%, indicating that EC is also active toward NO₂. Furthermore, the γ of NO₂ and the yield of HONO were observed to increase linearly with the content of defect sites on the graphite flakes in EC, deducing that the defect sites are important active sites for possible NO₂ reaction and HONO production [65]. In addition, Li et al. [66] illustrated that EC also played a pivotal role in the photochemical aging of soot by absorbing light to initiate the oxidation of OC through an electron transfer pathway. In general, the activity of EC towards NO₂ is related to the structure of EC, but the specific structure-activity relationship still needs further study.

Photoreaction mechanism: The reactivity of soot can be maintained for a long time under ultraviolet light conditions. Under light conditions, PAHs and aromatic species with carbonyl and carboxyl groups generated excited species P* which then reacted with NO₂ to form [P*-NO₂], then, HONO would be generated through hydrolysis or photolysis of [P*-NO₂]. The number of excited species P* was proportional to the number of active sites on the soot surface and the number of photons, so that the NO₂ uptake coefficient increased linearly with increasing light intensity. These results strongly suggested that the photochemical reaction process of soot in the presence of NO₂ is an important source of HONO [67].

3.3. Heterogeneous reactions of SO₂ on soot

3.3.1. Oxidation of SO₂ on soot

SO₂ and its atmospheric oxidation products (H₂SO₄ and sulfate) can not only lead to the formation of acid rain, but also are important precursors of new particle formation and subsequently mass growth [71]. Gas-phase and liquid-phase oxidation are two primary pathways of SO₂ oxidation in the troposphere, which have been extensively studied. Besides, heterogeneous reactions on aerosol particles also have important contributions to SO₂ oxidation. Zheng et al. [72] applied the heterogeneous chemistry in the Community Multiscale Air Quality model, and greatly improved the estimation of sulfate, in which the heterogeneous reactions of soot with SO₂ may be one of the effective pathways.

The uptake process of SO₂ on soot can be regarded as first adsorption and then reaction. During the adsorption process, the amount of C-O groups determined the SO₂ removal activity of activated carbon fibers [73]. What's more, heat treatment of activated carbon can increase the adsorption capacity for SO₂, because the heat treatment removes the oxygen-containing functional groups on the surface of activated carbon and correspondingly produces new active sites for SO₂ adsorption [74]. After adsorption, the adsorbed SO₂ can be oxidized to form sulfate on BC. The importance of these reactions in the real atmosphere is still in dispute, since soot may be poisoned by sulfur-containing species [75]. In addition, there may be three oxidation products of SO₂: H₂SO₄, sulfate and organic sulfur compounds. At present, there is still a lack of research on the specific characterization of oxidation products.

3.3.2. Influence of reaction conditions

λ: The OC content increases as λ increases [5], but the activity of soot towards SO₂ does not change monotonically. The influence of λ is also coupled with some other factors. According to Zhao et al. [75], the sulfate yield and SO₂ uptake increased with the increase of λ at 6% RH. However, at 54% RH, the prepared soot had the largest sulfate generation capacity on soot generated with a moderate λ of 0.134, and less sulfate was generated when λ increased to 0.162 or decreased to 0.1. At low humidity, the SO₂ uptake capacity increases linearly with the reduced π electron content on the surface of BC, which indicates that the reduced π electrons in aromatic compounds on the surface of BC may be the main active sites for SO₂ uptake. However, when RH increases, the limited reduced π electrons in the aromatic C−H groups and excessive adsorbed water inhibit the uptake of SO₂ and the further formation of sulfate.

RH: Water plays a necessary role in the formation of sulfate [76]. In the RH range of 6%-68%, increasing humidity was found to greatly promote the formation of sulfate [75]. A recent study [77] illustrated by first-principles calculations that water can decompose into OH on the surface of soot through the self-catalyzed process, and thus accelerate the conversion of SO₂ to SO₃. However, when the humidity was higher than 80%, water promoted the sulfate formation rate in the initial stage, but the total sulfate formation was reduced. Excess water molecules may compete with SO₂ for active sites on the soot surface, which is responsible for the reduction in the amount of sulfate.

Temperature: Temperature can affect the adsorption of SO₂ on soot. Chughtai et al. [78] found that the surface coverage of soot by SO₂ decreased linearly from 8.6% to 2.2% with increasing temperature from 295 to 339 K. The uptake of SO₂ on n-hexane soot was also studied by Koehler et al. [79] from 143 K to 233 K, which overlaps the temperatures of the upper troposphere and lower stratosphere. As the temperature decreased, the amount of SO₂ adsorbed on the soot rapidly increased, resulting in higher coverage at the same SO₂ pressure. Although the higher surface coverage of SO₂ may favor SO₂ reaction on the soot at lower tropospheric temperatures, it is usually not conducive to the oxidation reaction because there is less thermal energy available.

Radiation: Only limited studies focused on the influence of radiation on the heterogenous reaction of SO₂ on soot. In the absence of light, ionic sulfates can be formed, while under radiation, both covalent and ionic sulfates were observed [80,81], indicating that soot may be poisoned under solar radiation conditions. Radiation and temperature vary at different heights in the atmosphere. Therefore, research on their influences needs to be strengthened to better evaluate the environmental impact of soot-SO₂ reactions.

Coexisting gases: As for coexisting pollutant gases, it was found that NO₂ can promote the deposition of SO₂ on BC but only under some specific conditions. The SO₃ formation can be accelerated by additional NO₂, while overly high NO₂ concentrations will suppress the formation of SO₃ [50]. The latest progress showed that SO₂ can be efficiently catalyzed and oxidized on BC particles in the presence of NO₂ and NH₃ [4], which reveals that it is necessary to study the role of BC under compound pollution conditions.

3.3.3. Reaction mechanism

Chughtai et al. [78] proposed a catalytic oxidation scheme of SO₂ to H₂SO₄ as follows:

$$\mathrm{S}{\mathrm{O}}_2\left(\mathrm{g}\right)+\text{SS}\genfrac{}{}{0pt}{}{\leftharpoonup }{\rightharpoondown }\text{SS}.\mathrm{S}{\mathrm{O}}_2\left(\mathrm{s}\right)$$

$$\text{SS}.\mathrm{S}{\mathrm{O}}_2\left(\mathrm{s}\right)+{\mathrm{H}}_2\mathrm{O}\left(\text{ads}\right)\to \text{SS}.{\mathrm{H}}_2\mathrm{S}{\mathrm{O}}_3\left(\mathrm{s}\right)$$

$$\text{SS}.{\mathrm{H}}_2\mathrm{S}{\mathrm{O}}_3\left(\mathrm{s}\right)+{\mathrm{O}}_2\left(\mathrm{g}\right)\to \text{SS}.{\mathrm{H}}_2\mathrm{S}{\mathrm{O}}_4\left(\mathrm{s}\right)$$

where SS = soot surface site. Several studies showed that this reaction is dominated by the Langmuir-Hinshelwood mechanism in which SO₂ and O₂ are adsorbed on the surface of BC and participate in the reaction to form sulfuric acid [74,75]. Very recently, He et al. [82] proposed a new sulfate formation mechanism which does not involve the generation of H₂SO₃. First, O₂ is converted into epoxide groups which next oxidize SO₂ into SO₃, then SO₃ combines with H₂O to form H₂SO₄. In this process, water is no longer necessary for the formation of sulfur(Ⅵ). The proposal of this mechanism expands our understanding of the role of soot in sulfate formation. Besides acting as a catalyst, BC was also found to participate in the reaction as a reactant in some other studies. For example, Smith et al. [80] and Zhao et al. [75] both found that the conversion of SO₂ to sulfate was accompanied by the consumption of carbonyl C=O and C-H species. Further, He et al. [82] showed that the oxygen-functionalized graphene surface can directly oxidize SO₂ by using density functional theory calculations. The results showed that SO₂ can be oxidized by the epoxide group of graphene oxide through a two-step mechanism, in which the C-O bond far away from the SO₂ is broken first, followed by the breaking of the other C-O bond and the simultaneous formation of a new S-O bond. Previously, SO₂ was believed to be only oxidized by oxygen on the surface of the soot, while recently, the role of soot itself has been embodied in the oxidation process which needs further study.

3.4. Heterogeneous reactions of O₃ on soot

3.4.1. Rapid loss of O₃ on soot

The concentration of soot has increased both in the troposphere and stratosphere because of the growth in industrial activities and increasing numbers of subsonic and supersonic aircraft. It was found that the heterogeneous reactions of O₃ and the soot released by aircraft contributed to reduction of the O₃ concentration [83] through model research, and that the maximum decrease in O₃ concentration can reach 7 ppb [84]. Thus, the heterogeneous chemistry of O₃ and soot received extensive attention [85], [86], [87], [88].

The initial consumption of O₃ on soot is fast, while the following rapid surface deactivation led to a decrease in the O₃ loss rate. The uptake coefficient of O₃ on soot can also be expressed as a function of time and O₃ concentration: γ = γ₀/(1 + γ₀k[O₃]t) [89]. Table 6 summarized the uptake coefficient of O₃ on soot from different sources. The different types of soot and different reaction conditions resulted in a difference of three orders of magnitude in the O₃ uptake coefficient on soot (10⁻³ ∼ 10⁻⁶). The huge difference in the O₃ uptake coefficient brings great uncertainty to the evaluation of the influence of soot on the O₃ concentration in the atmosphere.

Table 6.

Summary of the literature data for the uptake coefficient of O₃ on soot.

Soot type	Reaction site	BET area (m2/g)	γBET	Reference
GfG	Chamber	395	1.20 × 10−6	[95]
Toluene	Flow tube Flow tube		(1.80 ± 0.70) ×10−4	[89] [89]
Kerosene			(3.80± 1.50) ×10−11
n-Hexane	Flow tube		(8.00-100.00) ×10−5	[96]
Propane	Flow tube	137	(1.30 ± 0.30) ×10−5	[97]
GfG	Chamber	395	(0.20-3.30) ×10−3	[98]
FW2	Knudsen cell	460	(1.00± 0.70) ×10−3	[53]
FW1	Chamber	460	1.00 × 10−3	[93]
Methane	Flow tube		8.00 × 10−5	[23]

After adsorption, O₃ may react with specific substances such as PAHs on the soot surface, which is also an important degradation pathway for the particulate PAHs in the atmosphere [90]. Carbonyl and peroxide groups were generated during the process [39], while the contents of oxygen-containing functional groups, hydrophilicity, water-soluble organic compounds, and free radicals on the surface all increased [91]. O₃ treatment will change the reaction activity of soot with other pollutants. The NO₂ uptake coefficient was reduced by 47% ± 15% after exposure to O₃, but it did not decrease to 0 in spite of high O₃ exposure, indicating that some sites on the soot surface are reactive toward NO₂ but not O₃ [10]. The presence of O₃ clearly modifies the soot surface and thereby enhances the oxidation of SO₂ into sulfate [92]. The reaction between O₃ and soot may also release products into the gas phase. The formation of O₂ was observed by Rogaski et al. [53]. In addition, Disselkamp et al. [93] found that the products included CO₂ but not CO, while in Smith and Chughtai's [94] experiment, CO, CO₂ and H₂O were observed as the products.

3.4.2. Influence of reaction conditions

λ: Fuel-rich soot, which contained more disordered carbon, showed higher reactivity towards O₃ compared to fuel-lean soot [87]. The conclusion is consistent with Zhu et al. [88]. Fuel-rich soot was more sensitive to O₃ aging, which can be interpreted as meaning that the disordered carbon is the reactive component for the ozonation [87].

RH: The RH dependence of O₃ loss rate on soot can be expressed by -d[O₃]/dt = k[O₃]²[H₂O]^0.2, illustrating that water promotes this reaction [99]. Similar conclusions were also obtained by He et al. [100].

Temperature: The reaction rate between O₃ and soot was 3.5 times more rapid at 296 K than at 238 K [101]. Similarly, a temperature increase from 278 K to 308 K increased the reaction rate by a factor of 5 [102]. These results showed that increased temperature can enhance the reactivity of soot towards O₃.

Radiation: As for the effect of light on the O₃ reaction, passivated soot can be reactivated during irradiation with a 4 times higher steady-state O₃ uptake rate compared with dark conditions [97]. Simulated sunlight was also observed to markedly enhance the uptake of soot by O₃, which produced various oxygen containing species such as lactones, anhydrides, ketones and aldehydes [103]. However, Smith and Chughtai [104] demonstrated that the rapid loss of O₃ in the presence of soot was unaffected by solar radiation, which only changed the distribution of CO₂ (g), CO (g) and H₂O (g) products.

For easy reference, the influences of soot formation conditions and environmental factors on the heterogeneous reactions of NO₂, SO₂ and O₃ on soot are summarized in Table 7.

Table 7.

Summary of soot formation conditions and environmental factors on its heterogeneous reactions.

Reaction conditions	NO2	SO2	O3
Higher λ	λ increased [34]. Main product changed from NO to HONO [33].	λ increased and then decreased. Sulfate yield increased [75].	Reactivity increased [87,88].
Higher RH	λ did not change [15] or decrease [60]. The concentration ratio of products did not change [15]. HONO yield increased [48,55] or decreased above 40% RH [59].	Sulfate increased and then decreased [75]. SO3 formation increased [77].	Reactivity increased [99,100].
Higher temperature	Main products changed from HONO and NO to NO and CO2 [57,105].	SO2 absorption decreased [79]. Surface sulfate decreased [78].	Reaction rate increased [101,102].
Radiation	Enhanced reaction [54,67].	Promoted sulfate formation [80,76,81].	Enhanced the aging of soot O3 [101,97]. Changed the products towards CO2, CO and H2O [104].

3.4.3. Reaction mechanism

According to the experimental results, Kamm et al. [95] proposed that the heterogeneous reaction between O₃ and soot mainly involved O₃ decomposition into O₂ and atomic oxygen on the surface of soot. The whole process was driven by a slow catalytic O₃ destruction process occurring after the fast uptake at the beginning. However, Disselkamp et al. [106] concluded that O₃ loss is not a catalytic process, based on the fact that CO₂ and CO were produced during the reaction. The production of CO and CO₂ shows that soot may participate in this reaction as a reactant.

3.5. Heterogeneous reactions of some other active atmospheric components on soot

O₂: As an abundant gas, the interaction between O₂ and soot should not be ignored. During the reaction, alkyne C-H and aromatic C-H were significantly reduced, while the aliphatic carbonyl was significantly enhanced after soot was exposed to O₂. OC was the main active component while EC was found to enhance the reaction [7]. The heterogeneous photochemical reaction between O₂ and soot is an important pathway for soot aging.

HO₂ radicals: The reaction of HO₂ radicals with soot was studied by different groups [107,86]. In the temperature range of 240-350 K, γ= (7.5 ± 1.5) × 10⁻² for HO₂ radicals on toluene and kerosene soot was calculated using a flow reactor [86], while a value of γ < 10⁻² was determined from modeling of the measured concentration-time profiles of HOONO₂ and NO₂ in the presence/absence of the aerosols in the chamber [107]. H₂O₂ was observed as the product of the HO₂ radicals reaction on soot by Ivanov et al. [108]. Different from the reaction of gases with soot, it is remarkable that aged soot still keeps the same reactivity towards HO₂ radicals after the initial stage, which means that soot can consume HO₂ radicals continuously in the atmosphere and HO₂ loss on soot can be a non-negligible sink for HOx species in the polluted lower troposphere.

OH radicals: Due to the strong activity of OH radicals and limitations of the detection technology, there are few studies on the reaction of OH radicals and soot at present. Bedjanian et al. [21] investigated the kinetics of the heterogeneous reaction between OH radicals and 15 PAHs bound to the soot surface, and determined first-order rate constants ranging from 0.017 to 0.043. The results showed that heterogeneous reaction between OH radicals and soot is not only a way to deplete OH radicals, but also a possible way to degrade PAHs. However, the concentration of OH radicals used in this article is higher than that in the real atmosphere, so further research is required to evaluate the importance of the heterogeneous reactions of soot with OH radicals at true atmospheric levels.

PAN: High levels of PAN often accompany high concentrations of particulate matter [22], so learning about the heterogeneous reaction between particulate matter and PAN contributes to understanding the evolution of polluted air. According to Zhao et al. [22], the upper and lower limits of the γ_ini of PAN on soot were 1.28 × 10⁻⁵ and 9.16 × 10⁻⁹ respectively at 295 K. The products of the reaction included CH₃COO⁻, HCOO⁻, NO₂⁻, and NO₃⁻, and their production rates all decreased with increasing humidity.

N₂O₅: NO, NO₂ and HONO were observed as products of the uptake of N₂O₅ on soot [23,24], while HNO₃ formation from N₂O₅ hydrolysis (even at very low RH) was also reported [107]. Quantitative conversion of N₂O₅ to NO₂ implies stoichiometrically that NO₃ may also be formed; however, the presence of NO₃ was not detected in the gas phase, indicating that soot may also be a sink of NO₃.

HNO₃: The heterogeneous reaction of soot and HNO₃ is controversial. Longfellow et al. [23] and Loukhovitskaya et al. [109] reported that the uptake of HNO₃ at 298 K was reversible and that it did not produce NO₂ or NO. However, in Choi and Leu’s [110] research, the decomposition of HNO₃ on soot produced NO, NO₂, and H₂O, oxidized soot surface, and generated some unidentified volatile products. Disselkamp et al. [106] also proved through chamber experiments that NO₂ could be generated. Moreover, the activity of soot towards HNO₃ was also affected by some factors such as λ and temperature. The reaction of HNO₃ on rich decane flame soot led to HONO as the main product, while NO and small amounts of NO₂ were observed from soot generated in a lean flame [111]. On the other hand, the soot tended to show high reactivity towards HNO₃ under high temperatures while there was no reactivity with HNO₃ when temperatures were lower than 220 K [110]. The reduction of HNO₃ on carbon aerosols could be a significant process that may alter the NOx/NOy partitioning in the atmosphere.

H₂SO₄: At present, only a small number of reports exist about the heterogeneous reaction of H₂SO₄ on the surface of soot. Zhang and Zhang [112] investigated the uptake of H₂SO₄ on methane, hexane, and kerosene soot, respectively. The results suggested that the uptake was irreversible and that it was a significant way to convert hydrophobic soot into hydrophilic aerosols.

4. Impacts of heterogeneous reactions on soot physicochemical properties

The atmospheric aging process will significantly alter the structure, composition, and morphology of soot, and therefore greatly determine its environmental impact.

4.1. Morphology

After aging, soot tends to have a more irregular structure [27] and the aggregate structure tends to be more compacted [27]. Fig. 2a-f shows typical SEM and TEM images of fresh soot aggregates and ozonized soot [27]. Ozonized toluene soot showed a small proportion of irregular forms (Fig. 2a and d) besides the regular structures similar to fresh soot. Similar anomalous structures are also evident in ozonized diesel soot samples (Fig. 2b and e). Moreover, the ratio of maximum length to the average width of aggregates decreased from 3.1 ± 1.0 (for fresh soot) to 2.1 ± 0.5 (for ozonized soot), which also means that after the reaction, soot becomes more compacted [113,91]. Compaction can directly alter the morphology and size distribution of soot particles and thus change the physicochemical properties [114].

Fig. 2.

SEM images of toluene soot (a) before and (d) after exposure to 2000 ppm O₃ for 30 minutes, diesel soot (b) before and (e) after exposure to 2000 ppm O₃ for 30 minutes, TEM images of Printex U soot (c) before and (f) after ozonized with 80 ppm of O₃ for 19 h. (a,b,d,e) is from Helen et al. [27], (c,f) is from Liu et al. [113].

4.2. Hygroscopicity

At present, most research [115,113] has investigated changes in hydrophilicity on O₃-aged soot, and a few studies reported aging with other gas pollutants [116,100].

A significant shift of soot towards more hydrophilic behavior has usually been observed, according to measurements of water adsorption [113] or water contact angle [91]. Besides aging with O₃, coatings of succinic acid, sulfate, or HNO₃ also led to detectable hygroscopic growth for soot [116,100].

4.3. Toxicity

Dithiothreitol (DTT) method was usually used to evaluate the toxicity of aged soot [38,91,117]. Compared with fresh soot, O₃-aged soot can generate more reactive oxygen species, and may be more toxic to lung cells. Oxidized soot has a greater oxidative ability, which is usually 1.5-2 times more than that of fresh soot [38]. A possible reaction mechanism for the increase in the DTT activity of aged soot is the formation of aromatic carbonyls and dihydrofuran species on soot particles [100]. These studies may imply that the toxicity of combustion–generated soot particles will increase after release into an atmosphere with high O₃ concentrations [38].

However, soot aged by other gases usually decreased its toxicity. Exposure to SO₂, NO₂, toluene, and a mixture of OA/organic vapors all led to a linear decline in the DTT decay rate for soot (SB4A). Furthermore, the DTT activity of four different samples decreased monotonically with the aging time in the ambient air. The actual atmospheric components are complex and variable, but according to the above experimental results, it is very likely that the toxicity of soot will be reduced after aging in the atmosphere, except in an environment with relatively high O₃ concentrations. In any event, changes in soot toxicity require a more comprehensive assessment.

4.4. Optical properties

Soot has always been regarded as the strongest sunlight-absorbing aerosol which has positive radiative forcing [6]. Fresh soot can be gradually aged in the heterogeneous processes in the atmosphere and the coating species may significantly affect the optical properties of soot. Zheng et al. [118] found that the infrared absorption cross section of soot increased up to three times due to the coating of sulfate. Nitrate coating and mixtures of OC can also strengthen the light absorption of soot [119,120]. These studies suggest that the heterogeneous chemistry of soot will further increase the light absorption ability and the radiative forcing of soot.

5. Impacts of heterogeneous reactions on soot on the atmosphere

5.1. Sink of gas pollutants/radicals and temporary repository for NOx

Many articles have concluded that the heterogeneous reactions over soot have little effect on atmospheric chemistry [85,46], for fresh soot can be poisoned soon after being emitted into the atmosphere. However, it has been found that, during the whole life cycle of the soot, it can always consume the surrounding gas despite the rate being small. What's more, there are some ways such as light to enhance the reaction and reactivate passivated soot in the atmosphere. For radicals, as mentioned before, they can be consumed at an efficient rate, which may make soot an effective sink for free radicals.

Soot can also serve as temporary storage for some pollutants. Besides PAN, the heterogeneous reactions between soot and NOx can be also regarded as a temporary repository for NOx. After aging by NO₂, soot can release HONO and NO once exposed to wet air [60].

5.2. Contribution to atmospheric oxidation capacity

The contribution of heterogeneous reactions over soot to the atmospheric oxidation capacity includes at least four parts. The first is the consumption of oxidizing gases or free radicals such as O₃, OH, HO₂, etc. A study based on simulation of a box model indicated that the presence of soot can reduce 7 ppb O₃ in the atmosphere [84], despite Tie et al. [121] thought the effect of soot on O₃ concentration was minor. Besides, soot also exert a critical effect on photolysis rate. The calculated J[O₃(1D)] and J[NO₂] values were reduced by 10%–30% which led to a decrease in O₃ concentration by about 5%–20% in the Houston area [122]. The depletion of substances with strong oxidizing power on soot weakens the oxidation ability of the atmosphere. The second part is the reaction of soot with reducing gases. For example, the oxidation or catalyzed oxidation of SO₂ to sulfate contributes to the formation of sulfate. During the moderate and heavy haze events, sulfate formation catalyzed by BC was proposed to account for 90%-100% and 30%-50% of the total sulfate formation, respectively [4]. Third, the formation of HONO during the heterogeneous reactions will provide precursors of OH radicals, and greatly strengthens atmospheric oxidation potential. It is worth noting that this process also increases the O₃ concentration. The generation of HONO on soot increased daytime O₃ level by about 7 ppb according to Lei's study [123]. In addition, HNO₃-soot interaction was also regarded as a renoxification process for NOx which might lead to additional tropospheric ozone production [124]. Last but not least, the release of CO, which can react with some oxidizing substances in the atmosphere and thus may have a negative impact on the oxidation ability of the atmosphere. In order to evaluate the influence of these heterogeneous processes, some researchers applied the laboratory results in different models to test their possible atmospheric contributions, as summarized in Table 8. The results indicated that soot heterogeneous chemistry could have a significant influence on atmospheric oxidation capacity but there is still large uncertainty.

Table 8.

Summary of the atmospheric impacts of heterogeneous reaction on soot in model studies.

Model method	Uptake coefficient	Contributions	Reference
Box model	NO2 (1.6×10−4)	0.15-15 ppb to HONO yield	[68]
Box model	NO2 (1.6×10−4)	Max: 6-7 ppb to HONO yield	[125]
NM5-CTM	O3 (5.0×10−4)	7 ppb formation to day-time O3 level	[123]
WRF-Chem MOCCA MOZART-2	Daytime: J(NO2)/10 Nighttime:1.0×10−4 O3 (3.0×10−3) NO2 (1.1×10−2) O3 (1.0×10−5)	Daytime: 0.05-0.21 ppb to HONO yield Nighttime: 0.4-1.8 ppb to HONO yield 7 ppb depletion to O3 level (soot: 20 μg/m3) Less than 1 ppb to O3 level (soot: 2 μg/m3) 0.093-0.900 ppb to HONO yield A minor effect	[126] [84] [121]

5.3. The influence of heterogeneous chemistry of soot on climate

Through absorbing and scattering solar radiation, pure soot particles may already have an effect on global warming [6]. As discussed above, the pollutant gases such as NO₂, SO₂, O₃ etc. may react on the soot surface and generate secondary aerosols. Further, these secondary aerosols will cover the surface of soot (as shown in Fig. 3). According to field measurements, soot was always observed to be mixed with sulfate, nitrate [127] and secondary organic aerosols (SOA) [128,129]. Due to internal mixing with other heterogeneous products, the absorption ability of soot is partly enhanced, which is called the “lensing effect” [129], and this process will further enhance the climate effects of soot. On the other hand, freshly discharged soot is hydrophobic, but after heterogeneous reaction or mixing with other soluble matters, it gradually becomes hydrophilic and possesses the ability to act as cloud condensation nuclei [130].

Fig. 3.

The influence of heterogeneous chemistry of soot on climate.

6. Perspectives

Due to its unique impact on climate, environment and human health, soot has gradually come into public view in the recent decade. In particular, the heterogeneous reactions of soot with the surrounding gas/radicals can not only modify the surface of soot and thus alter its physicochemical properties, but also affect the atmospheric environment. The research on soot in the laboratory has shown the strong dependence of chemical reactivity on the type of soot used and on environmental conditions such as fuel/oxygen ratio, radiation, temperature and humidity. Meanwhile, the results obtained by different types of experimental equipment vary widely. In order to better understand the role of soot in the real atmosphere, the soot atmospheric heterogeneous chemistry is developing rapidly with the development of detection technology and experimental methods.

Despite the progress made in this field, considering that many issues remain unsolved and unknown, future efforts may focus on the following aspects:

(1) There are many reports on the reduction ability of soot, but less attention has been paid to the oxidation ability of soot. The role of soot in complex air pollution and its contribution to atmospheric oxidation capacity requires further exploration with a combination of experiments and models.

(2) In order to incorporate atmosphere relevant heterogeneous chemistry of soot into air quality model, kinetic studies should be carried out under conditions close to the real atmosphere. Firstly, try to ensure that the environmental conditions during the experiments, including temperature, humidity, pollutant concentrations, and light intensity are comparable to the real atmosphere. Besides, use soot produced by combustion as a research object besides commercial BC. At last, do experiments under various experimental conditions to identify the key factors influencing the kinetics and further explore relationships between the kinetic parameters and these influencing factors.

(3) At present, most research studies focus on the heterogeneous reaction of a single inorganic gas with soot. In the real environment, multiple gas-phase components coexist, including organic compounds and inorganic gases. Therefore, it is necessary to study the heterogeneous chemistry between soot and organic gases and explore the inorganic-organic interactions on soot with the coexistence of multiple pollutants.

(4) The soot samples used in most studies are model soot or commercial soot prepared in the laboratory. The morphology, composition, and structure of soot from various real sources need to be characterized in detail to establish a structure-activity relationship and reveal the role of soot in environmental and climate effects.

(5) Since soot can be quickly deactivated after being emitted into the air, many articles suggest that the heterogeneous chemistry of soot is negligible. Research is needed to find other possible potential ways that may restore soot activity in addition to the known light mechanisms.

(6) Aged soot and fresh soot have different effects on climate. The heterogeneous reactions may strengthen the climate effects of BC. The influence of mixtures of different products and soot on the climate has not been fully understood.

(7) The aging of BC further causes changes in its health effects, and some of these changes manifest as increased toxicity, while others weaken the toxicity of soot; thus, a comprehensive assessment of soot health effects in the actual atmosphere and their mechanism need further research.

Declaration of competing interest

The authors declare that they have no conflicts of interest in this work.

Biographies

Yuan Liu is a Ph.D. candidate studying at the Laboratory of Atmospheric Environment and Pollution Control, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Her current research interests focus on the heterogeneous chemistry of black carbon and its environmental impact.

Biwu Chu is a professor at the Laboratory of Atmospheric Environment and Pollution Control, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. He is a member of the Youth Innovation Promotion Association of the Chinese Academy of Sciences since 2018, and was supported by Excellent Young Scholars of the National Natural Science Foundation of China in 2021. His current research interests mainly focus on the multi-phase formation of atmospheric secondary inorganic and organic aerosol.