Effect of restricted emissions during COVID-19 on atmospheric aerosol chemistry in a Greater Cairo suburb: Characterization and enhancement of secondary inorganic aerosol production

Abstract

To prevent the rapid spreading of the COVID-19 pandemic, the Egyptian government had imposed partial lockdown restrictions which led emissions reduction. This served as ideal conditions for a natural experiment, for study the effect of partial lockdown on the atmospheric aerosol chemistry and the enhanced secondary inorganic aerosol production in a semi-desert climate area like Egypt. To achieve this objective, SO₂, NO₂, and PM_2.5 and their chemical compositions were measured during the pre-COVID, COVID partial lockdown, and post-COVID periods in 2020 in a suburb of Greater Cairo, Egypt. Our results show that the SO₂, NO₂, PM_2.5 and anthropogenic elements concentrations follow the pattern pre-COVID > post-COVID > COVID partial lockdown. SO₂ and NO₂ reductions were high compared with their secondary products during the COVID partial lockdown compared with pre-COVID. Although, PM_2.5, anthropogenic elements, NO₂, SO₂, SO₄²⁻, NO₃⁻, and NH₄⁺ decreased by 39%, 38–55%, 38%, 32.9%. 9%, 14%, and 4.3%, respectively, during the COVID partial lockdown compared with pre-COVID, with the secondary inorganic ions (SO₄²⁻, NO₃⁻, and NH₄⁺) being the dominant components in PM_2.5 during the COVID partial lockdown. Moreover, the enhancement of NO₃⁻ and SO₄²⁻ formation during the COVID partial lockdown was high compared with pre-COVID. SO₄²⁻ and NO₃⁻ formation enhancements were significantly positive correlated with PM_2.5 concentration. Chemical forms of SO₄²⁻ and NO₃⁻ were identified in PM_2.5 based on their NH₄⁺/SO₄²⁻ molar ratio and correlation between NH₄⁺ and both NO₃⁻ and SO₄²⁻. The particles during the COVID partial lockdown were more acidic than those in pre-COVID.

Graphical abstract

1. Introduction

Aerosol, especially particulate matter (PM) with an aerodynamic diameter of less than 2.5 μm (PM_2.5), is considered one of the most important atmospheric pollutants since it has become a severe problem due to its adverse effects on the environment and human health (Tao et al., 2017; Cohen et al., 2017; Zhou et al., 2017; Wang et al., 2017; Wright et al., 2018; EEA, 2018; Othman et al., 2020). Natural and anthropogenic emissions as well as the atmospheric conversion of primary air pollutants into secondary particles are the main atmospheric PM sources (Byrd et al., 2010; Im et al., 2012; Police et al., 2016; Hassan and Khoder, 2017; Marchetti et al., 2019). Due to this diversity in PM sources, its size, shape, and chemical composition depend on their releasing sources and atmospheric chemical reactions, and they vary spatially and temporally (Ghio et al., 2012; Kim et al., 2018; Hao et al., 2019).

PM has diverse chemical and biological characteristics (Pope and Dockery, 2006; Naimabadi et al., 2016) since it contains a mixture of different hazardous pollutants like organic and inorganic chemical components (Di Vaio et al., 2016; Huang et al., 2018). This diversity in the chemical characterization of PM controls its toxicity (Atkinson et al., 2015). Prior studies (Dao et al., 2014; Kong et al., 2014; Cheng et al., 2016; Cao et al., 2013; Hassan et al., 2013; Hassan and Khoder, 2017; Luong et al., 2022) have shown that water-soluble inorganic ions (WSIIs) were accounted for about one-third, or more, of the PM mass, and play an important role in increasing PM concentrations in the atmosphere. The characteristics of WSIIs in PM associate with the types of pollution sources and meteorological status (Huang et al., 2016; Kawamura et al., 2010). Not only can WSIIs provide information on the atmospheric quality, but they also give scientific grounds for comprehension of the source, transmission, and formation processes of atmospheric PM (Cao et al., 2013; Zhou et al., 2015). Among these WSIIs, sulfate (SO₄ ²⁻), nitrate (NO₃ ⁻), ammonium (NH₄ ⁺), and chloride (Cl⁻) are major constituents of PM (Yang et al., 2011), and they are gaining more attention in the studies of the atmosphere of urban areas since they control the level of acidity of the PM (Xu et al., 2017) and therefore atmospheric acidity (Kunwar et al., 2016). PM acidity is a significant physical-chemical parameter that affects numerous chemical processes in the atmosphere (Huo et al., 2012).

Both atmospheric SO₄ ²⁻ and NO₃ ⁻ are produced from the oxidation of SO₂ and NO₂ which are arising from various anthropogenic sources like power plants, industry, motor vehicles, biomass burning, through atmospheric gas phase, and droplet phase reactions (Tsitouridou et al., 2003; Khoder, 2002; Hassan et al., 2013; Hassan and Khoder, 2017). The reaction of the gaseous acid species (e.g., H₂SO₄, HNO₃) produced by the oxidation of SO₂ and NO₂ in the atmosphere with ammonia gas (NH₃) that is released from both agricultural and non-agricultural sources in the atmosphere is also a source of ammonium salts (Lazaridis et al., 2006; Querol et al., 2001; Seinfeld and Pandis, 2012; Zhu et al., 2015). Therefore, it is very important to study the factors affecting the concentration, chemical composition, and characterization of atmospheric PM, especially PM_2.5, in order to develop suitable control strategies.

The first-time detection of the novel coronavirus-19 (COVID-19) infectious disease was recorded in December 2019 in the Wuhan province, China (Huang et al., 2021; Chen et al., 2020a, Chen et al., 2020b; Zhang et al., 2021; Aify et al., 2021), and later spread to other countries, then became a pandemic (Bai et al., 2020; Sohrabi et al., 2020; Lai et al., 2020). Previous studies (Contini and Costabile, 2020; Liu et al., 2020a) stated that exposure to high levels of air pollution for a long time may lead to an increase in vulnerability and mortality rates due to COVID-19. However, the factors that influence the fast spread of the COVID-19 virus and mortality rates, like air pollution and PM, are still under debate (Conticini et al., 2020; Yao et al., 2020). Although there is no confirmation until now on the COVID-19 virus transition through atmospheric PM, surface sanitization, perfect house ventilation, and environment cleanliness are useful in reducing the diffusion of this virus (Liu et al., 2020a).

The COVID-19 pandemic has led the governments of world countries to take strict measures, which are unparalleled in recent history, to limit the spread of this disease, such as stay-at-home orders via application of complete or partial lockdown. This led to a decrease in the transportation sector and industrial activities, and the consequent sudden decrease in the emission of air pollutants to the atmosphere (Forster et al., 2020; Venter et al., 2020; He et al., 2021a, He et al., 2021b; Naqvi et al., 2021; Lv et al., 2022). Although the complete and partial lockdown set up to limit the spread of the COVID-19 pandemic led to an unpredicted and fast reduction in economic and social movements, they give a unique scientific chance to understand the effects of anthropogenic emissions, from regional to global, on the earth's atmosphere (Forster et al., 2020; He et al., 2020; Kroll et al., 2020; Le Que're′ et al., 2020; Liu et al., 2020b) as well as on the climate and earth system (Diffenbaugh et al., 2020; Phillips et al., 2020; Raymond et al., 2020). In particular, a significant reduction in air pollution levels during COVID-19 lockdowns or partial lockdowns, especially for NO₂ and PM, was noticed in numerous countries around the world (Shrestha et al., 2020; Tobías et al., 2020; Wang and Su, 2020; Zhang et al., 2020; Muhammad et al., 2020; Dutheil et al., 2020; He et al., 2021a, He et al., 2021b; He et al., 2021a, He et al., 2021b), whereas an increase in ozone concentrations, as a secondary air pollutant, was observed (Collivignarelli et al., 2020; Venter et al., 2020; Parra and Espinoza, 2020; Zambrano-Monserrate and Ruano, 2020; Hashim et al., 2020). The observed responsiveness of increasing secondary pollutants to emission changes is the outcome of the nonlinear interactions implicated in the chemistry of the atmosphere (Seinfeld and Pandis, 2006). Changes in the concentrations of air pollutants can change atmospheric chemistry regimes, which can change the secondary PM production rates as a result of the increase in the oxidants concentrations under the effect of reduced emissions (Nault et al., 2018; Shah et al., 2018; Laughner and Cohen, 2019; Womack et al., 2019). Therefore, COVID-19 lockdowns provide a unique chance to understand the sensitivity of atmospheric secondary pollutants formation and production as well as their effect on the characterization of PM during primary emission changes.

Egypt was not excluded from the hit of the COVID-19 pandemic. As one of the most densely populated countries in the world, the Egyptian government carried out a set of measures to prevent the fast spread of the COVID-19 virus. These strict measures included placing travel bans, locking the international borders, quarantine for infected residential areas, limitation in social movement and public ingathering (including religious activities in mosques, sports clubs, coffee shops, restaurants, and beaches) as well as implementing a partial lockdown, which included the suspension of nighttime transportation, schools, universities and educational institutions, shopping centers, and suspension of public transportations during Eid festival days. This partial lockdown led to a reduction in the emissions of air pollutants and consequently changed the atmospheric chemistry regimes. Power plants and industrial activities were not affected by strict measures taken during the COVID partial lockdown in Egypt.

Recently, air pollution became the most serious environmental problem in Greater Cairo (GC) which has suffer from high PM pollution levels. PM emits directly from natural and anthropogenic sources, and/or formed from conversion of gases into secondary particles through chemical reactions (Byrd et al., 2010; Im et al., 2012; Police et al., 2016; Hassan and Khoder, 2017; Marchetti et al., 2019). The existence of intensive anthropogenic activities from industrial activities, big electrical power stations and vehicles running in the streets of the GC city led to an increase in the emission of O₃ precursors (NOx and VOCs) beside PM. The general climate of GC is cold, moist and rainy in winter; however, in summer it is characterized by high temperature, high solar radiation, and clear sky and rainless. So, the problem of air pollution in GC has been shifted towards the so called secondary photochemical pollutants, and their formation is facilitated by the local climatic conditions (high temperature, intense solar radiation, and clear sky), especially in the summer season (Khoder, 2002; Hassan et al., 2013). One of the most important secondary air pollutants that formed from the photochemical reactions is SO₄ ²⁻ and NO₃ ⁻ aerosol. This lead to an increase in the concentration of atmospheric PM. Therefore, understanding the factors affecting the chemical characterization, formation of secondary inorganic pollutants and atmospheric aerosol chemistry in GC atmosphere are very important. Previous research articles were focused on the status of air pollution, aerosol chemical composition, sources and secondary pollutants like O₃ before and during the COVID-19 lockdown period in different countries around the world (e.g. Liu et al., 2021; Chang et al., 2020; Huang et al., 2021; Yang et al., 2022; Shrestha et al., 2020; Tobías et al., 2020; Wang and Su, 2020; Zhang et al., 2020; Muhammad et al., 2020; Dutheil et al., 2020; Collivignarelli et al., 2020; Venter et al., 2020; Parra and Espinoza, 2020; Zambrano-Monserrate and Ruano, 2020; Hashim et al., 2020; He et al., 2021a, He et al., 2021b; He et al., 2021a, He et al., 2021b). However, there are no data concerning the effect of partial lockdown on the chemical characterization of atmospheric aerosol and the secondary inorganic aerosol production in a semi-desert climate area like Egypt. Therefore, this study is the first in-depth investigation of the effect of the COVID-19 partial lockdown on the atmospheric aerosol chemistry and enhanced secondary inorganic aerosol production and its effect on the chemical characterization of aerosol in a suburb of Greater Cairo (GC), Egypt. Finally, this study is focused on the side effects of intense anthropogenic emissions reduction, especially NO₂, and the nonlinear response of chemical composition and characterization of PM_2.5 to help us implement control strategies for secondary SO₄ ²⁻ and NO₃ ⁻ pollution, and consequently decrease atmospheric aerosol levels.

2. Materials and methods

2.1. Study area and sampling period

Greater Cairo (Cairo, Giza, and Qalyubia, Fig. 1 ) is one of the largest megacities in the world. It includes about 22% of Egypt's population (approximately 15–20 million inhabitants) with a density of 13,107 inhabitants/km² (Mostafa et al., 2018; Shaltout et al., 2020). Industrial activities are concentrated in the northern (Shoubra El Kheima) and southern (Helwan) regions of the city. Traffic density and congestion is a considerable problem in GC, since more than 3.4 million vehicles, fueled by unleaded gasoline, natural gas, and diesel fuels, are operating on city streets (Parry and Timilsina, 2015; Shaltout et al., 2018; Abbass et al., 2020). Generally, rubbish burning, and traffic and industrial emissions are the major sources of air pollution in GC (Parry and Timilsina, 2015; Shaltout et al., 2018, 2019; Abbass et al., 2020).

Fig. 1.

Map of Greater Cairo showing the study area and sampling site.

Gaseous NO₂ and SO₂ as well as aerosol PM_2.5 samples were collected on the roof of a private building, located in a suburban area (Kafr Tohormos) of GC (Fig. 1), at a height of ∼15 m above ground level. Traffic emissions represent the main sources of air pollution in the study area. Based on the timeline of controlling and prevention of the spreading of the COVID-19 pandemic released by the Egyptian government, which imposed the strictest measures, the first-level emergency response was immediately imposed by all governorates in Egypt. On 29 March, partial lockdown measures were issued by the Egyptian government, which lasted until 14 June. Therefore, the observation period for the study of the gaseous pollutants and characterization of PM_2.5 is divided into three periods: pre-COVID (2 February–22 March 2020), COVID partial lockdown (29 March–31 May), and post-COVID (28 June - 16 August). Samples were collected on a daily basis (24 h, from 7.00 a.m. to 7.00 a.m. on the following day) during the three different study periods.

Our focus objective, in the present study, is primarily on the effect of the COVID partial lockdown on the decreasing of in air pollutant concentrations and its effect on the chemistry of the atmosphere. Given that the period pre-COVID pandemic, which included the late winter (February) and early spring (March), and the period of COVID-partial lockdown during the COVID pandemic, which also included the rest of the spring months (April and May), the climatic conditions almost do not undergo major changes and are almost similar. Moreover, our rationale for choosing the comparison days during the two periods, before and during the partial lockdown, was based on strict criteria such as excluding the abnormal days like rainy days or dust storm days and using only the normal days with sunny skies during the two periods (before and during the partial lockdown). This will help us to investigate the effect of the COVID-partial lockdown on the decreasing of in air pollutant concentrations by comparing the status of air pollution before and during the partial lockdown periods. On the other hand, the post-COVID period during the summer season (28 June - 16 August), which is characterized already by high solar radiation with high dispersion rate for air pollutants, was used to find out whether the canceling of the partial lockdown during the COVID period will increase the primary pollutants such as NO₂ and SO₂ that are emitted directly from their sources or not. It's well known that most of these pollutant are decreased in Cairo's atmosphere during the summer season due to high dispersion and their conversion to the secondary pollutants (Khoder, 2002; Hassan et al., 2013).

2.2. Sampling and analysis of primary gaseous air pollutants

Ambient NO₂ gas was collected by bubbling air, at a flow rate of 0.5 L/min, through a sodium hydroxide–sodium arsenite solution to produce a stable sodium nitrite solution according to Harrison and Perry (1986). The nitrite ion that formed during the sampling time was reacted with phosphoric acid, sulfanilamide, and N–(1–naphthyl)–ethylenediamine dihydrochloride to produce an azo dye. Thereafter, the absorbance of the collected sample versus the reagent blank was measured at 540 nm spectrophotometrically. The calibration standard curve was prepared from an external standard nitrite solution. Based on the volume of air (m³) during the sampling period (24 h) and the standard curve, the ambient NO₂ concentration (μg/m³) was calculated. The method detection limit (MDL) for NO₂ was 3 μg/m³ for 24 h sampling. The analytical precision of measured NO₂ estimated from the standard deviation of repeated analyses of the standard solution was 3.5%.

The West and Gaeke (colorimetric) method (Harrison and Perry, 1986) was used for the determination of atmospheric SO₂ gas in the present study. It is essentially specific for the determination of SO₂ since the exposed samples are relatively stable after collection (Harrison and Perry, 1986). Moreover, the West and Gaeke method has been vastly used as a reference method and is covered by an international standard (UNEP/WHO, 1994). Ambient SO₂ gas was collected by bubbling air, at a flow rate of 0.5 L/min, through a solution of potassium tetrachloromercurate to form a stable non-volatile dichlorosulfitomercurate ion. Subsequently, solutions of purified, acid-bleached pararosaniline and formaldehyde were added to the exposed absorbing solution to form intensely colored pararosaniline methyl sulphonic acid, and thereafter the absorbance of the collected sample versus the reagent blank was measured at 560 nm spectrophotometrically. The concentration of SO₂ was determined from the prepared calibration standard curve, and then its concentration (μg/m³) in ambient air was calculated based on the volume of air (m³). The method detection limit (MDL) for SO₂ was 5 μg/m³ for 24 h sampling. The analytical precision of the measured SO₂ estimated from the standard deviation of repeated analyses of the standard was 3%.

2.3. PM_2.5 sampling and its mass calculation

Ambient air PM_2.5 samples were collected only during the periods of gaseous SO₂ and NO₂ sample collection. Two SKC IMPACT samplers (SKC Inc. USA), designed for the collection of PM_2.5 in ambient air at a flow rate of 10 L/min, were used for the sampling of PM_2.5. One set of PM_2.5 samples was used for the elemental determination, whereas the other set of samples was utilized for water-soluble inorganic ion species (WSIIs) analysis. PM_2.5 samples were collected on quartz filters (diameter 47- mm). For the gravimetric analysis of PM_2.5 in air, quartz filters were placed in controlled temperature and relative humidity desiccators for 24 h before and after sampling to remove any moisture. The difference in the weight of filter masses before and after sampling time represented the mass of PM_2.5 dust on the filter, then the concentration of PM_2.5 in the air was calculated from the dust mass on the filter and volume of air (m³) during the sampling time.

2.4. Chemical analysis of atmospheric aerosol PM_2.5

2.4.1. Elemental analysis

Aerosol loaded on the quartz filters of the first set of PM_2.5 samples was used for the elemental measurement using a digestion method. To prevent the contamination of PM_2.5 samples during the steps of digestion processes and analysis of elemental concentrations, all equipment, glassware, and filters were acid-washed with HNO₃ solution, rinsed with double-distilled water, and then oven-dried. Moreover, all chemicals used for sampling digestion and analysis were obtained from Merck Company. The digestion step of PM_2.5 samples for their elementals (Zn, Pb, Cr, Cu, V, Cd, As, Co, Ni, Mn, and Fe) analysis was carried out according to Method IO-3.1 (1999); Hassan et al. (2013) and Hassan and Khoder (2017). To determine the concentrations of these elementals, quartz filters were transferred into cleaned conical flasks of 50 ml volume. Thereafter, ten (10) ml of aqua regia (HCl: HNO₃; 3:1) was added to the conical flasks and left to digest overnight in the fume hood. Then, the samples were placed on a hot plate for 2 h at 90 °C till evaporated near to dryness, and subsequently, 5 ml of double-distilled water was added and heated at 100 °C on the hot plate till near to dryness. Afterwards, 25 ml of double-distilled water was added to the samples, then cooled at room temperature and filtered through Whatman filter paper (No. 42) using double-distilled water for washing. Finally, the digested samples were stored at 4 °C in pre-cleaned polyethylene bottles until analysis.

Inductively Coupled Plasma Optical Emission Spectrometry (ICP – OES 5100, Thermo Fisher Scientific, MA, USA) was used to analyze the elementals in the digested PM_2.5 samples. Calibration curves, for the measured elements, were prepared from stock standard solutions, and subsequently, the elemental concentrations in air based on the air volume (M³) were calculated as μg/m³. Quality assurance and quality control (QA/QC), including PM_2.5 digestion processes and analyses of field, laboratory and reagent blanks, analytical standards, ICP – OES 5100 calibration and detection limits, were carried out continuously. The ICP – OES 5100 was calibrated using standard solutions prepared from the multi-element stock standard. The external elements contamination outcome from the field blanks, as well as laboratory blanks, HCl, HNO₃, and the double-distilled water used in the digestion process, were determined by the same used analytical procedures. All the blanks, acids, and double-distilled water were lower than the method detection limits (MDLs), and the measured element concentrations data of the PM_2.5 samples were corrected by these blanks. Standard solutions of the target elements were run as a sample after analyzing ten (10) digested PM_2.5 samples to realize the precision and accuracy of ICP OES- 5100 as well as to minimize the error. The analytical precisions of measured elements estimated from the standard deviation of repeated measurements of standards were lower than 3%. MDLs for measured elements were specified from replicate measurements of low concentration and their standard deviations. MDLs for all measured elements ranged from 0.004 ng/m³ to 0.013 ng/m³.

2.4.2. Water-soluble inorganic ion species (WSIIs) analysis

Double-distilled water (20 ml) was used to extract aerosol loaded on the quartz filters of the second set of PM_2.5 samples, in 50 ml polypropylene tubes using an ultrasonic bath for 40 min, then filtered through a filter paper (Whatman No. 42) and refrigerated in pre-cleaned polyethylene bottles at 4 °C until WSIIs analysis. The target anions (NO₃ ⁻, SO₄ ²⁻ and Cl⁻) and cations (NH₄ ⁺, Ca²⁺, Mg²⁺, Na⁺, and K⁺) species were analyzed by Ion chromatography (Thermo IC-5000). An Ion Pac ASRS-4 suppressor and an Ion Pac AS11-HC × 250 mm analytical column were used for the detection of anions, whereas an Ion Pac CSRS-4 suppressor and an Ion Pac CS12A × 250 mm analytical column were used for cations detection. The eluents of 10 mmol/L Na₂CO₃ and 10 mmol/L CH₄O₃S were used for anions and cations analysis, respectively. A 10 μL loop was used for the sample's injection. QA/QC, including field blank, laboratory and reagent blanks, analytical standards, Ion chromatography calibration, and detection limits, was carried out continuously. Calibration standard curves were prepared from external standard solutions. The Ion Chromatography was calibrated using the prepared standard solutions. Field blanks, laboratory blanks, and the-double distilled water used in the extraction process were analyzed through similar procedures with the loaded filter to evaluate the external WSIIs contamination and for quality control. All the blanks and double-distilled water were lower than the method detection limits (MDLs), and the measured WSIIs concentrations data of the PM_2.5 samples were corrected by these blanks. The analytical precisions of measured WSIIs estimated from the standard deviation of repeated measurements of standards were lower than 4.5%. MDLs for measured WSIIs were specified from replicate measurements of low concentration and their standard deviations. MDLs for measured cations (NH₄ ⁺, K⁺, Na⁺, Ca²⁺, and Mg²⁺) and anions (SO₄ ²⁻, NO₃ ⁻, and Cl⁻) ranged from 0.005 μg/m³ to 0.082 μg/m³.

2.4.3. Enhancement of secondary transformation

The sulfur conversion (oxidation) ratio (F_SCR) and nitrogen conversion (oxidation) ratio (F_NCR) are widely used to get a deep insight into the enhancement of secondary transformation, i.e., the production rates of SO₄ ²⁻ and NO₃ ⁻ and the degree of conversion of SO₂ and NO₂ to SO₄ ²⁻ and NO₃ ^– in the atmosphere (Sicard et al., 2020; Yuan et al., 2015; Hassan et al., 2013). The F_SCR and F_NCR have been calculated using the following formulas (Khoder, 2002):

F_SCR = [SO₄²⁻]/{[SO₂] + [SO₄²⁻]} (1)

F_NCR = [NO₃⁻]/{[NO₂] + [NO₃⁻]} (2)

Where SO₄ ²⁻ is the sulfate concentration, as SO₂, μg/m³; SO₂ is the gas–phase SO₂ concentration, μg/m³; NO₃ ⁻ is the nitrate concentration, as NO₂, μg/m³; and NO₂ is the gas–phase NO₂ concentration, μg/m³.

2.4.4. Acid-base balance of atmospheric aerosol PM_2.5

To comprehend the acid-base balance of atmospheric PM, ion balance using the sum of cations (∑C) and anions (∑A) microequivalents concentrations needs to be evaluated according to previous studies (Shen et al., 2011; Wang et al., 2013; Gao et al., 2015; Hassan and Khoder, 2017). This will help us understand the acidity of atmospheric PM and the ways to limit it since the acidity is an important key that affects the hygroscopic growth, toxicity, and heterogeneous reactions of this PM (Sun et al., 2010; Cheng et al., 2014). PM is in acidic status when the ratio of the sum of the microequivalents concentrations of anions (A) to cations (C) (∑A/∑ C) is higher than one, and the increase in the value of ∑A/∑C ratio suggests an increase in particle acidity. Values of ∑A/∑C ratios close to one indicate particulate in ionic equilibrium state. In the present study, the ∑A and ∑C in the PM_2.5 were calculated as follows:

∑A = ∑ SO₄²⁻/48 + NO₃⁻/62 + + Cl⁻/35.5 (3)

∑C = ∑NH₄⁺/ 18 + Ca²⁺/20 + K⁺/39 + Mg²⁺/12 + Na⁺/23 (4)

2.4.5. Neutralization of atmospheric aerosol PM_2.5

The calculation of neutralization factors (NFs) gives an insight into the roles of NH₃ gas and crustal components such as Ca²⁺ and Mg²⁺ in the neutralization of acidic components of atmospheric PM_2.5. NFs have been calculated using the following formula (Saxena et al., 1996; Hassan et al., 2015; Khoder and Hassan, 2008):

NF_NH4+ = [NH₄⁺]/{2[SO₄²⁻] + [NO₃⁻]} (5)

NF_Mg2+ = [Mg²⁺]/{[SO₄²⁻] + 2[NO₃⁻]} (6)

NF_Ca2+ = [Ca²⁺]/{[SO₄²⁻] + 2[NO₃⁻]} (7)

2.4.6. Aerosol acidity estimation using thermodynamic model

The E− AIM-IV model (http://www.aim.env.uea.ac.uk/aim) was used in the present study for estimating the aerosol acidity (Gong et al., 2021). Based on the E− AIM-IV model, particle total acidity ([H+]s) was firstly calculated by eq. (8) with using the field measurement data:

[H⁺]_s = 2[SO₄²⁻] + [NO₃⁻] + [Cl⁻] - [NH₄⁺] - [Na⁺] - [K⁺] (8)

where [H⁺]_s, [SO₄ ²⁻], [NO₃ ⁻], [Cl⁻], [NH₄ ⁺], [Na⁺] and [K⁺] are the ambient air concentrations of aerosol-bound ions in mol/L. After that, based on the state of the thermodynamic equilibrium system of H⁺−NH₄ ⁺−Na⁺−SO₄ ²⁻−NO₃ ⁻−Cl⁻−H₂O with temperature in the range of 263.15–330 K (Behera et al., 2015), in situ aerosol acidity was estimated using the following equation:

in situ pH = - log α_H+ = - log [γ_H+ × [H⁺]_is × 1000 × ρ/m] (9)

where α_H+ is activity of H⁺ in mol/L in the aqueous phase on the particle, γ_H+ is the activity coefficient of the aqueous phase H⁺, and [H⁺]_is is the free H⁺ concentration of the aqueous phase in unit of mol per m⁻³ of air, p is the aerosol aqueous phase density in g/cm³, and m is the air concentration in g/m³ in the aqueous phase. The parameters, ρ, γ_H+, and [H⁺]_is and the concentration of aerosol water content ([H₂O]) were derived from the E-AIM-IV model.

2.5. Meteorological parameters

Meteorological data, including the relative humidity, ambient temperature, and wind speed, over the three study periods were taken from Weather Underground (www.wunderground.com).

3. Results and discussion

3.1. Temporal variations of PM_2.5, NO₂, and SO₂ concentrations

The temporal variations in the daily concentrations of PM_2.5, NO₂, and SO₂ as well as the values of meteorological parameters (temperature, relative humidity, and wind speed) during the pre-COVID, COVID partial lockdown, and post-COVID periods are graphically presented in Fig. S1. In this figure, it can be seen that the daily concentrations of PM_2.5, NO₂, and SO₂ pollutants maintained at a relatively high level during the pre-COVID period, then decreased dramatically during the COVID partial lockdown and the concentrations increased gradually again during the post-COVID period. These temporal variations in daily concentrations during the different study periods generally coincide with the effect of the increase or decrease in their emission sources during these times. Although we could not obtain any data regarding the rate of decreasing in the number of cars during the covid-19 period, the decreasing in traffic flow during COVID partial lockdown period due to closing of the schools and colleges and the stopping the traffic flow during nighttime was clear. Therefore, the decline of PM_2.5 and NO₂ might be attributed to the reduction of transportation activity, and while this could not explain the drop of SO₂, the sources of SO₂ were mainly industrial activities and power plants, not vehicle emissions.

Table 1 presented the average concentrations of PM_2.5, SO₂, and NO₂ as well as the temperature, relative humidity, and wind speed values during the three study periods. The highest average temperature (28.8 °C) was recorded in the post-COVID period, whereas the lowest value was found in the pre-COVID period (16.1 °C). The average wind speeds were 10.1 km/h in the pre-COVID period, 10.4 km/h in the COVID partial lockdown period, and 8.8 m/h in the post-COVID period. As for relative humidity, its average values ranged from 46.6% in the COVID partial lockdown period to 57.4% in the pre-COVID period. Insignificant correlation coefficients (P < 0.001) were found between the temperature and the concentrations of SO₂ (r = 0.23–0.39), NO₂ (r = 0.25–0.49) and PM_2.5 (r = 0.30–0.45) and between the relative humidity and SO₂ (r = - 0.24–0.37), NO₂ (r = 0.21–0.38) and PM_2.5 (r = 0.20–0.31) concentrations, and between the wind speeds and the concentrations of SO₂ (r = 0.23–0.39), NO₂ (r = 0.18–0.30) and PM_2.5 (r = 0.27–0.46) during the thee periods of study. These results indicate that the concentrations of measured air pollutants during pre-COVID period and their decreasing (COVID partial lockdown period)/increasing (post-COVID period) were not mainly depending on the meteorological conditions but fundamentally depending on the strength of local emission sources, especially traffic emissions. As reported in Table 1, the average concentrations of PM_2.5, NO₂, and SO₂ were 60.2, 89.4, and 58.5 μg/m³ in the pre-COVID period, 36.7, 55.3, and 39.2 μg/m³ in the COVID partial lockdown period, and 54.3, 65.2, and 45.9 μg/m³ in the post-COVID period, respectively. Compared with the mean concentrations of PM_2.5, NO₂, and SO₂ during the pre-COVID period, PM_2.5 had the largest reduction of 39.0%, followed by NO₂ at 38.1%, and SO₂ at 32.9% during the COVID partial lockdown. On the other hand, the change in PM_2.5, NO₂, and SO₂ mean concentrations during the COVID partial lockdown to the post-COVID period were +32.4%, +15.2%, and +14.6%, respectively. In Egypt, the COVID partial lockdown included mainly closing schools and universities, and the strictest measures that suspended traffic flow during nighttime, which is normally characterized by traffic congestion until the early hours of the morning on normal days. Consequently, the sharp decrease in traffic volume during the COVID partial lockdown period led to a sharp reduction in the concentration levels of PM_2.5 and SO₂, and NO₂. However some studies report an increase in secondary air pollution like PM2.5 in china (Chen et al., 2020a, Chen et al., 2020b) and French (Sbai et al., 2021a) and Morocco in north Africa (Sbai et al., 2022). Because photochemistry can be a significant contributor to PM formation especially under high ozone levels because strong oxidizing capacity could accelerate the formation of OH radicals in the atmosphere in the presence of humidity; OH radicals react with VOC and BVOC and lead to the formation of secondary aerosols which represent a significant fraction of PM_2.5 (Ortega et al., 2016; Sbai and Farida, 2019; Huang et al., 2021; Sbai et al., 2021b). Our results, in the present study, are consistent with previous studies (Chauhan and Singh, 2020; Selvam et al., 2020; Nakada and Urban, 2020; Mahato et al., 2020), which reported that the reduction rates of PM_2.5 concentrations during the partial lockdown were 32% in New York, USA, 39.0% in the Gujarat state of India, 27.12% in the megacity Delhi, India, and 29.8% in the São Paulo state. A significant decrease in air pollution during the lockdown, particularly in the NO₂ concentration, was observed in China and several European and American countries (Shrestha et al., 2020; Wang and Su, 2020; Zhang et al., 2020). SO₂, NO_x, and PM_2.5 concentrations decreased during the COVID-19 lockdown by 26%, 47% and 46%, respectively, over the Yangtze River Delta Region (Li et al., 2020). Significant reduction in PM_2.5, PM₁₀, NO₂, and CO has been found in India regions during the lockdown (Singh et al., 2020). By comparing the difference in PM_2.5 concentrations during the three periods (before controlling, during controlling and after controlling) of COVID-19 in three important cities in the region (Wuhan, Changsha, and Nanchang) (Zhao et al., 2021c), it is found that after controlling human activities, the PM_2.5 concentration dropped by 37.45%. Compared to the same pre-COVID periods in 2018 and 2019, PM_2.5, NO₂, and SO₂ concentrations in the urban areas of Malaysia decreased by 23–32%, 63–64%, and 9–20%, respectively, during the lockdown period (Kanniah et al., 2020). PM_2.5 and NO₂ concentrations decreased by 21% and 35%, respectively, during the lockdown period compared with the period of 2018–2019 in Almaty, Kazakhstan (Kerimray et al., 2020). NO₂ and PM_2.5 decreased by 52.7% and 50.0%, respectively, during the lockdown period in Delhi (Mahato et al., 2020). In addition, a high reduction in NO₂ (24.1%–54.3%) concentration was noticed in megacities in South America, such as Rio de Janeiro and Sao Paolo (Dantas et al., 2020; Nakada and Urban, 2020), during the lockdown period compared with the period before COVID or former years. Compared with other pollutants, the declines in NO₂ concentrations were more consistent with the reduction of vehicular emissions (Yuan et al., 2021; Xu et al., 2020). Reduction in the NO₂ concentration ranged from 20 to 30% during lockdown in China, USA, Italy, Spain, and France (Muhammad et al., 2020; Wang and Su, 2020; Dutheil et al., 2020). In the present study, the relatively lower reduction rate of SO₂ than NO₂ suggests that COVID restrictions had a much more significant influence on traffic emissions than on power plants and industrial activities that are the main sources of SO₂. When restrictions were lifted in the post-COVID period, all the measured pollutants rebounded and there was a progressive increase in their concentrations, especially PM_2.5.

Table 1.

Average atmospheric aerosol PM_2.5, NO₂ and SO₂ concentrations, as well as temperature, relative humidity and wind speed values during pre-COVID, COVID partial lockdown and post- COVID periods.

Parameter	Pre- COVID				COVID partial lockdown				Post-COVID
	N	Range	Mean	SD	N	Range	Mean	SD	N	Range	Mean	SD
PM2.5	15	40.4–75.7	60.2	11.1	23	16.3–56.5	36.7	11.4	15	41.9–64.7	54.3	7.4
NO2	15	67.3–106.1	89.4	12.1	23	39.5–66.9	55.3	7.1	15	47.7–78.9	65.2	8.9
SO2	15	42.6–69.8	58.5	7.7	23	27.6–47.4	39.2	6.2	15	34.9–55.1	45.9	5.9
Temperature (° C)	15	12.2–19.6	16.1	2	23	19.4–30.0	23.7	2.9	15	23.0–31.3	28.8	1.8
Relative humidity (%)	15	28.9–74.7	57.4	11.7	23	24.0–60.2	46.6	9.6	15	30.6–64.8	53.9	9.2
Wind Speed (km/h)	15	4.5–19.1	10.1	4.3	23	7.6–15.7	10.4	2.3	15	5.8–10.5	8.8	1.2

N: number of daily samples.

SD: standard deviation.

3.2. Chemical composition of atmospheric aerosol PM_2.5

3.2.1. Elemental compositions of atmospheric aerosol PM_2.5

The measured element concentrations in PM_2.5 during the pre-COVID, COVID partial lockdown, and post-COVID periods are illustrated in Fig. 2 . Based on the average element concentrations, Fe and Pb were the most dominant elements in PM_2.5, accounting for about 70.14% and 11.90% during the pre-COVID period, 78.14% and 7.97% during the COVID partial lockdown period, and 76.53% and 8.58% during the post-COVID period of the total measured elements concentrations, respectively. The total mass concentrations of elements in PM_2.5 followed the order pre-COVID (1639.5 ng/m³) > post-COVID (1475.5 ng/m³) > COVID partial lockdown (1223.6 ng/m³). The total mass of elements in PM_2.5 during the COVID partial lockdown when compared with that of the pre-COVID period was reduced by about 25.4%. Compared with those during the pre-COVID period, Ni had the largest reduction of 55%, followed by Zn (52%), Pb (50%), Cr (48%), Cu (45%), V (44%), Cd (43%), As (41%), Co (38%), Mn (19%), and Fe (16.7%) during the COVID partial lockdown. On the other hand, the element concentrations increased gradually after the COVID partial lockdown period. The change in concentrations from the COVID partial lockdown period to the post-COVID period was +43.7% for As, +38% for Ni, +37% for Co, +35% for Cr, +35% for Cu, +34.4% for V, +30.1% for Zn, +30% for Pb, +29% for Cd, +19% for Mn, and +18% for Fe. Previous studies reported that elements like Cu, Pb, As, and Zn are emitted from non-exhaust sources, such as tire and brake wear (Alves et al., 2018; Jeong et al., 2019; Denier van der Gon et al., 2007). Traffic emissions, common usage of anti-corrosive galvanized car accessories, use of leaded gasoline, and lubricating oil leakages are the anthropogenic sources widely related to the emission of Cd, Cu, Pb, As, and Zn (Zhao et al., 2021a, 2021b; Patel and Jain, 2022; Wang et al., 2019; Kumari et al., 2021). It is therefore supposed that the decreasing of the anthropogenic emission sources of elements in PM, like industrial emissions and traffic density, contributes to the reduction of these element concentrations. So, the effect of decreased traffic density during the COVID partial lockdown did not only cause low exhaust particulate emissions, but also reduced emissions generated from tire wear and re-suspension of street dust, which include some elements. As a result, the substantial decrease in the concentrations of Ni, Zn, Pb, Cr, Cu, V, Cd, As, and Co during the COVID partial lockdown could be due to the restrictions on traffic activities during this period. This is consistent with Nguyen et al. (2021) who found that the concentrations of Cd, Se, As, Sr, Ba, Cu, Mn, Pb, K, Zn, Ca, Al, and Mg in Hanoi (Vietnam) within the partial lockdown period were lower than those prior to the partial lockdown. Li et al. (2020) and Kanniah et al. (2020) reported that the reduction of air pollution was mainly related to the restrictions on traffic and industrial activities. In the present study, a small reduction in Fe and Mn (mainly crustal elements) concentrations compared with other measured elements was recorded during the COVID partial lockdown, since they emit mainly from the re-suspension of road dust (Chang et al., 2009; Lee and Hieu, 2011; Lim et al., 2010).

Fig. 2.

Variation of the elemental concentration in atmospheric aerosol PM_2.5 during Pre-COVID, COVID partial lockdown and post-COVID periods.

3.2.2. Water-soluble inorganic ion species (WSIIs) in atmospheric aerosol PM_2.5

3.2.2.1. Concentrations and distribution of WSIIs

The average concentrations of the WSIIs in PM_2.5 during the pre-COVID, COVID partial lockdown, and post-COVID periods are shown in Fig. 3 . The PM_2.5 ions mass concentrations followed the order SO₄ ²⁻ > NO₃ ⁻ > NH₄ ⁺ > Ca²⁺ > Cl⁻ > Na⁺ > K⁺ > Mg²⁺ during pre-COVID and the COVID partial lockdown, and SO₄ ²⁻ > NO₃ ⁻ > Ca²⁺ > NH₄ ⁺ > Cl⁻ > Na⁺ > K⁺ > Mg²⁺ during post-COVID. Despite the relatively higher decline of gaseous NO₂ (38.1%) and SO₂ (32.9%) concentrations during the COVID partial lockdown period compared with the pre-COVID period, the decline in the concentrations of particulate SO₄ ²⁻ (8.99%) and NO₃ ⁻ (14.2%) was relatively lower compared with their precursor's (NO₂ and SO₂) during the same time. This means that the transformations of NO₂ to NO₃ ⁻ and SO₂ to SO₄ ²⁻ were improved during the COVID partial lockdown. Yang et al. (2022) reported that the disproportionate drop of particulate NO₃ ⁻ compared to NO₂ gas concentration (60%) implied the improved conversion of NO₂ during the COVID lockdown period. Compared with those during the pre-COVID period, Cl⁻, NH₄ ⁺, Ca²⁺, K⁺, Mg²⁺, and Na⁺ concentration reductions were 28.79%, 4.24%, 47.93%, 21.03%. 36.10%, and 27.36%, respectively, during the COVID partial lockdown in this study. After the COVID partial lockdown period, the changes in mean concentrations were +41.43% for NO₃ ⁻, +32.00% for SO₄ ²⁻, +29.71% for Cl⁻, +26.63% for NH₄ ⁺, +45.60% for Ca²⁺, +25.45% for K⁺, +34.02% for Mg²⁺, and +31.90% for Na⁺ during post-COVID. The average mass concentrations of crustal-mineral species (Ca²⁺, K⁺, Mg²⁺ and Na⁺) in PM_2.5 were 3.37 μg/m³ (pre-COVID), 2.07 μg/m³ (COVID partial lockdown), and 3.35 μg/m³ (post-COVID). The relatively higher levels of crustal–mineral species in the pre and post-COVID periods than in the COVID partial lockdown period resulted from the effect of local and regional man-made activities on the production of crustal–mineral dust, like the re-suspension of road dust during traffic flow during the pre and post-COVID periods.

Fig. 3.

Average mass concentrations of water-soluble ionic species during pre- COVID, COVID partial lockdown and post- COVID periods.

NO₃ ⁻/SO₄ ²⁻ mass concentration ratio can be used to distinguish between the relative importance of vehicular traffic versus stationary sources of sulfur and nitrogen in the atmosphere (Li et al., 2016). High values of this ratio reveal that the predominance of vehicular traffic sources is higher than stationary sources of these pollutants. In the present study, the mean NO₃ ⁻/SO₄ ²⁻ ratios in PM_2.5 during the pre-COVID, COVID partial lockdown, and post-COVID periods were lower than one, they were 0.52, 0.49, and 0.57, respectively. These ratios indicate the predominance of stationary sources over vehicular sources during the three different periods of study. These results are sensible since the power plant and heavy industries were still working during the COVID-19 pandemic period. The differences in the values of reduction of SO₂ compared with NO₂ during the COVID partial lockdown further support this argument. The relatively higher NO₃ ⁻/SO₄ ²⁻ ratio in both the pre and post-COVID periods may be attributed to the relatively higher traffic density during these times.

Among all WSIIs determined in PM_2.5, SO₄ ²⁻ is the most abundant chemical species in the pre-COVID, COVID partial lockdown, and post-COVID periods, accounting for about 8.49%, 12.01%, and 12.01% of the PM_2.5 mass, and 37.97%, 42.18%, and 40.54% of the total WSIIs, respectively (Fig. 4 ). The sum mass concentration of the secondary inorganic ions (SNA (SO₄ ²⁻, NO₃ ⁻and NH₄ ⁺) accounts for about 14.81%, 21.27%, and 22.01% of the PM_2.5 mass, and 66.98%, 74.13%, and 73.42% of the total WSIIs in the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. This result indicates that the formation of secondary pollutants (SNA) from the oxidation of its precursors during the COVID partial lockdown and post-COVID periods was enhanced. The concentrations of the total WSIIs in PM_2.5 were 12.80 μg/m³, 10.44 μg/m³, and 15.97 μg/m³, accounting for 21.51%, 28.65%, and 29.95% of the PM_2.5 mass during the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. The relatively higher percentages of WSIIs in both the COVID partial lockdown and post-COVID periods than in the pre-COVID period indicate that WSIIs form more PM_2.5 under COVID partial lockdown and summertime (post-COVID period) conditions.

Fig. 4.

The mass contribution of each water-soluble ion to total water-soluble ion mass (a), and mass of each water-soluble ion to PM_2. mass (b) during pre-COVID, COVID partial lockdown and post-COVID periods.

3.2.2.2. Enhancement of secondary transformation

As seen in Fig. 5 , an increase in the F_NCR and F_SCR values occurred during both the COVID partial lockdown and post-COVID periods compared with the pre-COVID period. The values of F_NCR were 3.91% in pre-COVID, 4.53% in the COVID partial lockdown, and 7.51% in post-COVID. Compared with the values of F_NCR in pre-COVID, F_NCR values in the COVID partial lockdown and post-COVID periods were about 1.16 and 1.92 times, respectively, higher than F_NCR values in the pre-COVID period. However, the values of F_SCR were 8.75%, 12.83%, and 13.75% during the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. F_SCR values in the COVID partial lockdown and post-COVID periods were about 1.47 and 1.57 times, respectively, higher than F_SCR values in the pre-COVID period. The relatively higher values of F_SCR and F_NCR during the post-COVID period (summertime) led to an increase in SO₄ ²⁻ and NO₃ ⁻ concentrations, indicating that the atmospheric oxidation of SO₂ and NO₂ to SO₄ ²⁻ and NO₃ ⁻ and their conversion are based on the atmospheric photochemical oxidation reactions. Consistent with previous studies (Khoder, 2002; Lin et al., 2006; Chiwa, 2010; Hassan et al., 2013), the highest F_SCR and F_NCR were found in the summer season. Atmospheric NO₃ ⁻ is formed from HNO₃, which is produced from the reactions of NO₂ with OH radical during daytime and with NO₃ ⁻ radicals during nighttime (Seinfeld and Pandis, 2016). Moreover, SO₄ ²⁻ is produced from the oxidation of SO₂ through both heterogeneous and homogeneous reactions in the presence of relative humidity (Rattigan et al., 2002; Fahey et al., 2005). As a result, the concentrations of NO₃ ⁻ and SO₄ ²⁻ in PM_2.5 depend on the concentrations of NO₂ and SO₂ and their conversion rates to NO₃ ⁻ and SO₄ ²⁻.

Fig. 5.

Sulfur conversion ratio (FSCR, %) and nitrogen conversion ratio (FNCR, %) during pre-COVID, COVID partial lockdown and post- COVID periods.

In the present study, despite the decline of NO₂ (38.1%), SO₂ (32.9%), SO₄ ²⁻ (8.99%), and NO₃ ⁻ (14.2%) in the COVID partial lockdown compared with those in the pre-COVID period, the F_SCR and F_NCR values in the COVID partial lockdown were about 1.47 and 1.16 times, respectively, higher than F_SCR and F_NCR values in the pre-COVID period. These results indicate that the relatively enhanced production rate of NO₃ ⁻ and SO₄ ²⁻ during the COVID partial lockdown did not enhance SO₄ ²⁻ and NO₃ ⁻ concentrations. This may be due to the reduction of NO₂ and SO₂ concentrations, which are the precursors of SO₄ ²⁻ and NO₃ ⁻, during the same period. As a result, the concentrations of both NO₃ ⁻ and SO₄ ²⁻ in PM_2.5 still decreased, although their secondary production rates increased. Consistent with previous studies (Chang et al., 2020; Huang et al., 2021; Yang et al., 2022), the NO₃ ⁻ formation was enhanced relative to the reduction of NO₂ emission during the COVID-lockdown period in Beijing, Zhengzhou, and Shanghai. In China (Yangtze River Delta region), nitrogen oxide (NOx) reductions enhanced O₃ accumulation and therefore increased secondary aerosol formation (Wang et al., 2021). Up to 90% reduction of certain emissions during the city-lockdown was observed, whilst O₃ enhancement and further increase in the atmospheric oxidizing capacity and secondary aerosol formation in urban areas were occurred (Le et al., 2020). The oxidation rates of SO₂ to SO₄ ²⁻ and NO₂ to NO₃ ⁻ during the COVID lockdown were higher than those during the pre-COVID period (Liu et al., 2021). Sulfur and nitrogen conversion ratios during the COVID lockdown period were higher than those during the pre-lockdown period (Yang et al., 2022). In the present study, the relatively higher values of F_SCR and F_NCR in the COVID partial lockdown suggest that the oxidizing capacity of the atmosphere was significantly enhanced during the COVID partial lockdown period, and consequently, the oxidation of SO₂ to SO₄ ²⁻ and NO₂ to NO₃ ⁻ increased. The increase of oxidizing capacity means enhancements in the production of O₃, OH radical, and H₂O₂, which lead to an increase in the oxidation and the rate of conversion of SO₂ to SO₄ ²⁻ and NO₂ to NO₃ ⁻. Although the concentration of O₃, which is considered a key precursor of OH that controls the oxidizing power of the atmosphere, was not measured in the present study, previous studies reported that the concentration of O₃ increased during the COVID lockdown period due to a significant reduction in NO_x emissions and a decrease in the O₃ consumption via NO titration (Dantas et al., 2020; Kerimray et al., 2020; Li et al., 2020; Nakada and Urban, 2020; Tobías et al., 2020; Sharma et al., 2020; Chu et al., 2020; Duan et al., 2021). Significant reduction in NO₂, and enhancement in O₃ has been found in India regions during the lockdown (Singh et al., 2020). By comparing the difference in O₃ concentrations during the three periods (before controlling, during controlling and after controlling) of COVID-19 in three important cities in the region (Wuhan, Changsha, and Nanchang) (Zhao et al., 2021c), it is found that after controlling the human activities, the O₃ concentration increased by 111.83%. NOx reductions can increase O₃ levels by decreasing the NOx titration (through NO + O₃ and also via nocturnal NO₂ + O₃ reactions) (Leung et al., 2020). The elevated O₃ concentration was favorable for the formation of NO₃ ⁻ despite the drastic decrease of NO₂ during the COVID lockdown period (Chang et al., 2020; Huang et al., 2021; Lv et al., 2020; Xu et al., 2020; Yang et al., 2022). An increase in the oxidation capacity (O₃) of the atmosphere led to the enhancement of sulfur oxidation ratio, and then SO₄ ²⁻ formation during the COVID lockdown (Duan et al., 2021). Ren et al. (2021) reported that a 26–61% reduction of NOx emissions only lowered surface HNO₃ by 2–3% and enhanced the NO₃ ⁻ formation in the presence of a high concentration of NH₃ reduction in China during the COVID-19 pandemic due to the enhanced atmospheric oxidizing capacity.

Relative humidity (RH) is considered an important meteorological factor that affects the chemical compositions and characterization of PM in the atmosphere (Sun et al., 2013; Tian et al., 2014; Hassan and Khoder, 2017). Therefore, the impact of RH on the FSCR and FNCR was investigated in the present study. Significant positive correlation coefficients (p < 0.001) were found between the RH and F_SCR in the three different periods of study (Fig. S2). The correlation coefficients (r) were 0.86, 0.83, and 0.86 in the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. This result suggests that the RH is a significant agent for the oxidation processes of SO₂ to SO₄ ²⁻, and enhancement of the conversion rate needs high relative humidity besides the presence of gas–phase reactions. On the other hand, the insignificant correlation coefficients between relative humidity and F_NCR during the pre-COVID, COVID partial lockdown, and post-COVID periods (Fig. S2) indicate that the gas–phase reactions are important and predominant for the oxidation processes of NO₂ to NO₃ ⁻. Previous findings showed that both droplet phase and gas-phase reactions are significant for the conversion of SO₂ to SO₄ ²⁻, whereas the gas-phase reactions are the predominant for the conversion of NO₂ to NO₃ ⁻ (Khoder, 2002; Hassan et al., 2013). In the present study, the correlation coefficients between wind speeds and the F_SCR (r = 0.26–0.34) and F_NCR (r = 0.20–0.31) were insignificant (P < 0.001) during the three periods of study. Moreover, statistically significant correlation coefficients (P < 0.001) were found between temperature and the F_SCR (r = 0.75–0.84) and F_NCR (r = 0.79–0.88) during pre-COVID, COVID partial lockdown, and post-COVID periods. This indicates that the increase of temperature that related with increase of solar radiation led to an increase in ozone concentration (Vukovich and Sherwell 2003; Ribas and Peñuelas 2004; Garcia et al., 2005) which enhance the conversion of SO₂ to SO₄ ²⁻ and NO₂ to NO₃ ⁻.

To investigate the effect of enhancement of conversion rate of primary pollutants (SO₂ and NO₂) into secondary aerosol (NO₃ ⁻ and SO₄ ²⁻) on the concentrations of atmospheric PM_2.5, correlation coefficients (r) between F_SCR and PM_2.5 and F_NCR and PM_2.5 were calculated during the three different periods of study (Fig. 6 ). Significant positive correlation coefficients were found between F_SCR and PM_2.5 and between F_NCR and PM_2.5. These correlation coefficients were 0.80, 0.85, and 0.87 for F_NCR and PM_2.5, and 0.85, 0.92, and 0.85 for F_SCR and PM_2.5, during the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. These results suggest that the enhancement of secondary aerosol formation, from the oxidation of gaseous NO₂ and SO_2, is important for increasing the atmospheric PM_2.5.

Fig. 6.

Relationship between the atmospheric aerosol PM_2.5 concentration and sulfur conversion ratio (F_SCR) and nitrogen conversion ratio (F_NCR) during pre-COVID, COVID partial lockdown and post- COVID periods.

3.2.2.3. Chemical forms of SO₄^2–and NO₃⁻ in atmospheric aerosol PM_2.5

The relationships and concentration ratios of SO₄ ²⁻, NO₃ ⁻, and NH₄ ⁺ in the PM_2.5 can be used to get insight into the chemical forms of these secondary pollutants as well as the chemical characteristics of PM_2.5. Significant positive correlation coefficients (p < 0.001) were found between the concentrations of NH₄ ⁺ as a neutralizing ion and the sum of acidifying ions (∑SO₄ ²⁻ and NO₃ ⁻) in PM_2.5 during pre-COVID (r = 0.85), COVID partial lockdown (r = 0.78) and post-COVID (r = 0.77). Moreover, the correlation coefficients between the concentrations of NH₄ ⁺ and both SO₄ ²⁻ and NO₃ ⁻ during the pre-COVID, COVID partial lockdown, and post-COVID periods are illustrated in Fig. 7 . A significant positive correlation coefficient was found between the concentrations of NH₄ ⁺ and NO₃ ⁻ in PM_2.5 during pre-COVID (r = 0.75, p < 0.001), clearly suggesting that NO₃ ⁻ in PM_2.5 may have been present as NH₄NO₃ during the pre-COVID period, which is characterized by low atmospheric temperature. On the contrary, under the effect of relatively higher temperature during the COVID partial lockdown and post-COVID periods, a portion of NH₄NO₃ is volatilized (Liu et al., 2017), forming nitric acid (HNO₃) and ammonia (NH₃) gas. Then, a fraction of HNO₃ forms particulate NO₃ ⁻ through the gas-to-particle conversion in the presence of the dust particles with alkaline components (Tang et al., 2017; Wang et al., 2018). Consequently, particulate NO₃ ⁻ other than NH₄NO₃ probably existed. NO₃ ⁻ can be associated with other minerals to form Mg (NO₃)₂, KNO₃, and Ca (NO₃)₂ (Qiao et al., 2019). Acidic nitrogen species can react with alkaline ions (Ca²⁺, Mg²⁺, and K⁺) in particles (Hayami and Carmichael, 1998; Bourotte et al., 2007; Hassan et al., 2013; Wang et al., 2017). This is confirmed by the insignificant positive correlation coefficients between the concentrations of NO₃ ⁻ and NH₄ ⁺, and the significant positive correlations between the concentrations of NO₃ ⁻ and Ca²⁺, K⁺, and Mg²⁺ in the COVID partial lockdown and post-COVID periods. The concentrations of NH₄ ⁺ and SO₄ ²⁻ in PM_2.5 were significant (p < 0.001) positive correlated during the three different study periods (Fig. 7), the correlation coefficients (r) were 0.82, 0.82, and 0.77 during the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. Moreover, significant positive correlation coefficients (p < 0.001) were found between SO₄ ²⁻ and Ca²⁺ concentrations during post COVID (r = 0.83), COVID partial lockdown (r = 0.86), and post-COVID (r = 0.87). These results indicate that both NH₃ gas and Ca²⁺ play important roles in the neutralization of sulfate, and (NH₄)₂SO₄ and/or NH₄HSO₄ and CaSO₄ are sulfate forms existent in the PM_2.5; this is consistent with previous studies (Tang et al., 2004, 2017; Wang et al., 2017).

Fig. 7.

Scatter plot of NH₄⁺ and SO₄²⁻ and NO₃⁻ concentrations during pre-COVID, COVID partial lockdown and post- COVID periods.

The value of NH₄ ⁺/SO₄ ^2-molar ratio is considered the borderline to describe the PM system in poor or rich NH₄ ⁺ systems (Vieira-Filho et al., 2016). With the excess of atmospheric NH₃ gas, PM is fully neutralized to produce (NH₄)₂SO₄, and the NH₄ ⁺/SO₄ ²⁻ molar ratio exceeds 2 (Walker et al., 2004; Huang et al., 2011). In the presence of CaCO₃, (NH₄)₂SO₄ can be converted to (NH₄)₂SO₄.CaSO₄.2H₂O and further to CaSO₄ within 2–6 days (Mori et al., 1998). Based on the X-ray diffraction, (NH₄)₂SO₄.CaSO₄.2H₂O existed when NH₄ ⁺/SO₄ ²⁻ molar ratio was close to 1 (Querol et al., 1998). Particulate SO₄ ²⁻ existed as (NH₄)₂SO₄, (NH₄)₂SO₄.CaSO₄.2H₂O, and CaSO₄ when the NH₄ ⁺/SO₄ ²⁻ molar ratios lie between 1:1 and 2:1 theoretical lines (Duan et al., 2003). In the present study, the NH₄ ⁺/SO₄ ²⁻ molar ratios ranged from 1.08 to 1.71 (with a mean value of 1.32) during pre-COVID, 1.06 to 2.83 during the COVID partial lockdown (with a mean value of 1.44), and 0.98 to 1.81 (with a mean value of 1.28) during post-COVID. Fig. 8 illustrates the reference theoretical molar ratio for NH₄HSO₄ (1:1) and (NH₄)₂SO₄ (2:1), as well as the scatter plot of NH₄ ⁺ and SO₄ ²⁻ concentrations (μmol/m³) during the pre-COVID, COVID partial lockdown, and post-COVID periods. It can be observed that, from this figure, the NH₄ ⁺/SO₄ ²⁻ molar ratios in PM_2,5 samples during the three periods of study scatter between 1:1 and 2:1 theoretical lines, indicating that SO₄ ²⁻ in PM_2.5 during the three periods of study may be present as (NH₄)₂SO₄, (NH₄)₂SO₄.CaSO₄.2H₂O, and CaSO₄. This is consistent with the observation of previous investigators (Querol et al., 1998; Mori et al., 1998; Duan et al., 2003; Khoder and Hassan, 2008; Hassan et al., 2013) who suggested that SO₄ ²⁻ in PM may exist as (NH₄)₂SO₄, (NH₄)₂SO₄.CaSO₄.2H₂O, and CaSO₄ when NH₄+/SO₄ ^2–molar ratios scatter between 1:1 and 2:1 theoretical lines. The results in this study displayed the high contribution of NH₄ ⁺ salts in the PM_2.5 (Andrade et al., 2012; Albuquerque et al., 2012).

Fig. 8.

Scatter plot of NH₄⁺ and SO₄²⁻ concentration during pre-COVID, COVID partial lockdown and post-COVID periods. Note: Ratios of 1:1 and 2:1 indicate the theoretical molar NH₄⁺/SO₄²⁻ ratio of NH₄HSO₄ and (NH₄)₂SO₄.

3.2.2.4. Acid-base balance of atmospheric aerosol PM_2.5

The scatter plot of ∑A and ∑C microequivalents values during the pre-COVID, COVID partial lockdown, and post-COVID periods are illustrated in Fig. 9 . It can be seen that all PM_2.5 samples during the pre-COVID period plotted below the reference theoretical ∑A: ∑C ratio of 1:1 line. The values ∑A/∑C ratios ranged from 0.62 to 0.98 (with an average of 0.72), suggesting that the pre-COVID PM_2.5 samples were alkaline. In the post-COVID and COVID partial lockdown periods, all PM_2.5 samples plotted around the reference theoretical line 1:1 (∑A: ∑C). The values of ∑A/∑C ratios for PM_2.5 samples ranged from 0.83 to 1.19 (with an average of 0.99) during the COVID partial lockdown period and from 0.92 to 1.41 (with an average of 1.01) during the post-COVID period, which indicates that those PM_2.5 samples were nearly neutral. On the other hand, significant (p < 0.001) positive correlation coefficients were found between ∑A and ∑C microequivalent concentrations. The correlation coefficients (r) were 0.70 during pre-COVID, 0.95 during the COVID partial lockdown, and 0.90 during post-COVID. Furthermore, the ∑A/∑C ratios during the COVID partial lockdown and post-COVID periods were 1.38 and 1.40 times higher than those during the pre-COVID period, which suggests that the particles during the COVID partial lockdown and post-COVID periods were more acidic than those during the pre-COVID period. This is due to the promotion of the formation of secondary inorganic ions (NO₃ ⁻ and SO₄ ²⁻), which could enhance the acidity of particles (Tian et al., 2020). In the present study, the unmeasured anions, such as bicarbonate, organic ions (formate and acetate), F⁻, NO₂ ⁻, PO₄ ³⁻, and Br⁻, lead to deficiency of the total anions (∑A), and consequently increase the deviation of ion balances to the lower side of the theoretical line (1:1) during the three periods of study. As a result, the situation of ion balance may be changed to the above reference theoretical 1:1 (∑A: ∑C) line (acidity) when the ∑A includes these unmeasured anions.

Fig. 9.

Scatter plots of ∑anion microequivalents versus ∑cation microequivalents during: (a) pre-COVID, (b) COVID partial lockdown, (c) post-COVID.

3.2.2.5. Acidity of atmospheric aerosol PM_2.5

Based on the E− AIM-IV model used in the present study for estimating the aerosol acidity, the mean values of the total acidity ([H⁺]s) during the three sampling periods showed significant variations, with higher values during both the post-COVD (0.09 mmol/L), COVID-partial lockdown (0.07 mmol/L) than pre-C0VID (0.04 mmol/L) period. On the other hand, the mean values of in situ pH estimated from the E-AIM model were following the pattern: pre-COVID (2.03) > COVID-partial lockdown (1.61) > post-COVID (1.53), which suggested that the both COVID-partial lockdown and post-COVID aerosols were more acidic than the pre-COVID aerosol. Moreover, the higher levels of [H⁺]s during both the post-COVD and COVID-partial lockdown could be explained based on the enhanced levels of SO₄ ²⁻ and NO₃ ^−. The total acidic nature of the aerosols depends on the levels of SO₄ ²⁻ and NO₃ ⁻ species (Behera et al., 2013; Xue et al., 2011). The total acidity [H⁺]s is influenced by emission sources of SO₄ ²⁻ and ambient atmospheric chemistry (Nenes et al., 1998; Xue et al., 2011).

The estimated values of in situ pH ranged from 1.54 to 2.03, in the present study, were comparable to those observed in the Mediterranean and eastern parts of European countries (1–4, Kakavas et al. (2021)), in a rural site in southeastern US (0–2, Battaglia et al., 2019; Weber et al. (2016)) but lower than those observed in North China Plain (4.2–4.7, Tao et al. (2020)), in Central and northern European countries (3–5, Kakavas et al. (2021)), and in a suburban site of Nanjing (2–4, Gong et al. (2021)). Based on the finding of the present field study and the other field studies around the world, policymakers across the world need to set up measures for decreasing the emission sources that are responsible for acid deposition (pH < 5.6) from the atmosphere in wet or dry forms influencing the ecosystem health.

3.2.2.6. Factors affecting the neutralization of atmospheric aerosol PM_2.5

Fig. S3 shows the NFs for NH₄ ⁺, Ca²⁺, and Mg²⁺ during the pre-COVID, COVID partial lockdown, and post-COVID periods. NH₄ ⁺ has the highest NF followed by Ca²⁺ and Mg²⁺. NFs were 0.47, 0.46, and 0.44 for NH₄ ⁺, 0.34, 0.18, and 0.23 for Ca²⁺, and 0.14, 0.09, and 0.1 for Mg²⁺ during the pre-COVID, COVID partial lockdown, and post-COVID periods, respectively. This indicates that NH₄ ⁺ is the principal component of the neutralization of acidic components in atmospheric PM_2.5 during the three different study periods. Moreover, the relatively lower NFs for Ca²⁺ and Mg²⁺ during the COVID partial lockdown period than those during the pre-COVID period may be attributed to the decline in their concentrations in PM_2.5 during the COVID partial lockdown period.

4. Conclusions and recommendations

In this study, the influence of the partial lockdown due to COVID-19 on the chemical characterization of atmospheric PM_2.5 in a Greater Cairo suburb was investigated. Our results reveal that reductions in vehicular emissions were responsible for the decline of PM_2.5 (39.0%), elements (38–55%), NO₂ (38.1%), and SO₂ (32.9%) during the COVID partial lockdown. Relatively lower reductions in the secondary inorganic ions (SO₄ ²⁻ (9%) and NO₃ ⁻ (14%)) were observed compared with their precursors (NO₂ and SO₂) in the COVID partial lockdown. Although the concentrations of particulate NO₃ ⁻ and SO₄ ²⁻ in PM_2.5 decreased during the COVID partial lockdown compared with the pre-COVID period, their production rates were higher during the COVID partial lockdown compared with the pre-COVID period. Moreover, the secondary inorganic ions (SNA) were the main dominant components in PM_2.5 during the COVID partial lockdown. These results suggest that the secondary aerosol productions were enhanced during the COVID partial lockdown compared with the pre-COVID period. This enhancement, which partially offset the reduction in PM_2.5, may be a result of the increased oxidizing capacity of the atmosphere under the effect of reduced emissions. On the other hand, the relatively higher production rates of NO₃ ⁻ and SO₄ ²⁻ during the COVID partial lockdown confirmed the occurrence of high photochemical reactions and oxidants formation that increase the atmospheric oxidative capacity. The enhancement of the conversion rate of SO₂ to SO₄ ²⁻ needs high relative humidity besides the presence of gas-phase reactions, whereas the gas-phase reactions are predominant for the oxidation processes of NO₂ to NO₃ ⁻. The enhancement of secondary aerosol formation is important for increasing the atmospheric PM_2.5. SO₄ ²⁻ existed as (NH₄)₂SO₄, (NH₄)₂SO₄.CaSO₄.2H₂O, and CaSO₄ during the three periods of study, whereas NO₃ ⁻ existed as NH₄NO₃ in the pre-COVID period and was present in other different forms like Mg (NO₃)₂, KNO₃, and Ca(NO₃)₂ during the COVID partial lockdown and post-COVID periods. PM_2.5 in the COVID partial lockdown and post-COVID periods was more acidic than that in the pre-COVID period, and NH₄ ⁺ was the main neutralizing agent of acidic components in atmospheric PM_2.5. Finally, this study focused on the side effects of intense anthropogenic emissions reduction, especially NOx, and the nonlinear response of chemical composition and characterization of PM_2.5. Control strategies for secondary SO₄ ²⁻ and NO₃ ⁻ pollution need not only reduction of NOx, but also reduction of the key (oxidants) that leads to an increase in atmospheric oxidation capacity, which is involved in the formation of these secondary pollutants. Therefore, developing coordinated control strategies for multi-type pollutants is needed to minimize the side effects caused by emission reduction.

Credit author statement

Salwa K. Hassan: Provide the idea of the study and designed research, validation, project administration, supervision, discussed the results and writing-reviewing and editing; Mansour A. Alghamdi: investigation, validation, interpreted the results, discussed the results and writing-reviewing and editing; Mamdouh I. khoder: Provide the idea of the study and designed research, investigation, validation, interpreted the results, discussed the results and writing- original draft preparation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: