Vehicular pollution as the primary source of oxidative potential of PM_2.5 in Bhubaneswar, a non-attainment city in eastern India

Abstract

We assessed the oxidative potential (OP) of PM_2.5 (n = 230) using dithiothreitol (DTT) assay to identify the major emission sources in Bhubaneswar (20.20°N, 85.80°E), one of the non-attainment cities under the National Clean Air Program, situated on the eastern coast of India. Continuous day and night PM_2.5 samples were collected during periods influenced by marine airmass (MAM; April–May 2019) as well as continental airmass (CAM; October 2019–December 2019). Volume normalized DTT (DDTv) activities were approximately two times higher during CAM compared to MAM periods. In contrast, mass normalized DTT activity (DDTm) showed insignificant variations between CAM and MAM periods. This might be due to particulate organic matter, which accounted for more than one-fifth of the PM_2.5 mass loading and remained surprisingly invariant during the study periods. Positive matrix factorization (PMF) identified secondary aerosols (MAM: 26% and CAM: 33%) as dominant contributors to PM_2.5 mass in both periods. OP, is, however, dominated by vehicular emissions (21%) as identified through multiple linear regression. Conditional Bivariate Probability Function (CBPF) analysis indicated that local sources were the primary drivers for the catalytic activity of PM_2.5 in the study region. Additionally, stagnant meteorological conditions, combined with the chemical aging of species during regional transport of pollutants, likely enhanced redox activity of PM_2.5 during the CAM period. The study highlights that increasing traffic congestion is primarily responsible for adverse health outcomes in the region. Therefore, it is important to regulate mobility and vehicular movement to mitigate the hazardous impact of PM_2.5 in Bhubaneswar.

---

Environmental significance

Oxidative Potential (OP), the catalytic capability of PM_2.5 to generate reactive oxygen species within the human body, served as a pertinent health metric. This study assessed OP of PM_2.5 using dithiothreitol assay in Bhubaneswar, a non-attainment city on the eastern coast of India. While previous studies in India (mostly in landlocked regions) identified unregulated combustion sources as the major contributor to OP, our study revealed that traffic-related emissions are primarily responsible for inducing OP of PM_2.5 in Bhubaneswar. These findings highlight the necessity of implementing action plans on transportation and mobility to safeguard public health in the region. Additionally, this study provides insights that can be applied to similar urban coastal sites to mitigate the hazardous effects of air pollution.

1. Introduction

Pollution poses a substantial risk to human health and well-being. According to the latest reports, pollution leads to approximately 9 million premature deaths globally each year, with air pollution alone accounting for 6.5 million fatalities.^1,2 Apart from diminishing life expectancy, air pollution triggers and aggravates various health issues, such as cardiorespiratory diseases, cancer, diabetes, cataracts, and neonatal and mental health disorders.^3–6 Air pollution, especially particulate pollution, has a more severe impact on developing nations such as India because of the swift pace of socio-economic growth, urbanization, industrialization, and the incineration of agricultural and urban waste.^1,7,8 Research indicates that long-term exposure to high levels of PM_2.5 has increased the disease burden in India, contributing to conditions such as ischemic heart disease, chronic obstructive pulmonary disease, stroke, lung malignancies, and lower respiratory infections.^7,9–11 Although the intricate mechanisms underlying the health effects of PM_2.5 have not been completely elucidated, recent studies suggest that oxidative stress could be a potential indicator for assessing the toxicity of aerosols.^12–15 Oxidative stress arises when antioxidant capacity of the body becomes insufficient to counteract the detrimental impacts of a superfluous amount of reactive oxygen species (ROS). ROS ingress into the respiratory system through inhalation of PM_2.5, either directly attached to the particles (particle-bound ROS) or catalytically produced within the body by specific redox active species (particle-induced ROS) during cellular redox reactions. This catalytic generation, occurring concurrently with the reduction in antioxidant levels, is commonly termed oxidative potential (OP).^16–19

Although several studies have been conducted in India in the context of OP, the focus has predominantly been on the northern and western regions, resulting in a lack of understanding of PM and its related health risks in other regions.^15,20–29 Specifically, investigations have been reported in the landlocked Indo-Gangetic Plain (IGP) regions, including Patiala,^23,25,30 Kanpur,²⁹ Agra,²¹ Delhi,^27,30–33 Haryana,³⁰ and Uttar Pradesh;³⁰ major cities such as Mumbai,²⁰ Bangalore,¹⁵ and Ahmedabad;²⁴ two high-altitude stations (including Shillong and Mount Abu);^22,26 and indoor kitchens within the northeastern states of Arunachal Pradesh, Assam, and Meghalaya.²⁸ However, India's diverse demographic and fuel use patterns have resulted in heterogeneous PM distribution across regions, leading to varied health outcomes.³⁴ This variation is also evident in the 131 non-attainment cities as per the National Clean Air Program (NCAP), highlighting the need for targeted, region-specific assessments to implement a multi-pronged strategy to reduce particulate pollution. For example, the eastern region of India, including Bhubaneswar (20.20°N, 85.80°E), exhibits distinct emission characteristics compared to the north-western parts and IGP, with larger SO₂ and SO₄²⁻ loading.^35,36 It is important to highlight that, like other growing urban cities, Bhubaneswar is also experiencing a sharp increase in respiratory morbidity cases in recent years.^37,38 Although epidemiological studies have shown that respiratory morbidities are strongly associated with the increase in the PM_2.5 concentration in the region,³⁷ the detailed mechanisms have not yet been explored. Thus, for the first-time, the present study conducted a detailed assessment of OP of PM_2.5 in the non-attainment urban coastal city during two contrasting wind patterns: marine airmass and continental airmass periods. By emphasizing Bhubaneswar, this study addresses a significant knowledge gap in understanding PM_2.5 toxicity in the eastern coastal region of India. The findings from the study will be helpful for framing regional/city specific action plans to reduce PM_2.5 pollution as part of the NCAP mission. Moreover, similar coastal cities in South and Southeast Asia facing comparable challenges could benefit from the insights gained through this study.

2. Measurements and data analysis

2.1. Site description and sample collection

All the samples of PM_2.5 were obtained from the rooftop (10 m agl) of the CSIR-Institute of Minerals and Materials Technology (CSIR-IMMT) main building. The sampling site is within a lush green campus, centrally located within the city, approximately 1.0 km away from the National Highway (NH-16). The urban coastal site receives southerly winds (or marine airmass; MAM) from the Bay of Bengal (BoB) during April–May (Fig. S1†). In contrast, the city experiences northerly wind (herein referred to as continental airmass; CAM) transported over long distances from IGP and the densely industrialized Chota Nagar Plateau during October–February.^35,36,39 Continuous day (06:00–16:00 h) and night (18:00–06:00 h) samples were collected on moisture-free quartz microfiber filters (Whatman, QMA, 47 mm) using a fine particulate air sampler (APM 550 MFC, Envirotech, India) from April–May and the end of October–December 2019, with a flow rate of 16.75 L min⁻¹. However, our sampling plan was interrupted for seven days during May (02nd–08th May), 2019 due to the severe cyclonic storm ‘Fani’, which hit the city on 3rd May 2019. A total of 230 PM_2.5 samples were gathered during both periods, along with 20 field blanks (once every 11 days). Further details about the site, sampling and weighing protocols are described in our previous publications.^35,36,40 Meteorological parameters including ambient temperature (T), rainfall, relative humidity (RH), wind speed (WS), and wind direction (WD) were acquired from the Indian Meteorological Department (IMD), Pune, for both periods. The meteorological conditions for both periods are summarized in Table T1 in the ESI.†

2.2. Chemical analysis

The sampled filters underwent analysis for carbonaceous species (OC and EC), water-soluble ions (Na⁺, K⁺, Ca²⁺, Mg²⁺, Cl⁻, NH₄⁺, NO₃⁻, and SO₄²⁻), and water-soluble metals (Al, Mg, Ca, Pb, Ba, Ti, Cr, Mn, Fe, Ni, Cu, and Zn). A comprehensive explanation of the analytical methods is available in previous publications.^16,36 Briefly, OC and EC analysis was conducted by utilizing one-quarter of the sampled filter on an OC/EC analyzer (DRI 2001A; Atmoslytic Inc., Calabasas, CA, USA) at CSIR-National Physical Laboratory (CSIR-NPL), New Delhi. For the determination of water-soluble species, three-quarters of the sampled filters were extracted ultrasonically with 50 mL of deionized water. The major ions (Na⁺, K⁺, Ca²⁺, Mg²⁺, Cl⁻, NH₄⁺, NO₃⁻, and SO₄²⁻) in the water-soluble extracts were identified using ion chromatography (930 Compact IC Flex, Metrohm, Switzerland). Furthermore, water-soluble metals were examined by acidifying the aqueous extracts (2% v/v) with high-pure HNO₃ and analyzed using inductively coupled plasma optical emission spectroscopy (ICP-OES, iCAP 7600 Duo, Thermo Fisher, USA). Here, we considered only water-soluble metals as they are easily bioavailable and are regarded as key drivers of OP.⁴¹

2.3. Data analysis

Sea-salt (ss) and non-sea-salt (nss) fractions of SO₄²⁻, Mg²⁺, Ca²⁺, and K⁺ were estimated using Na⁺ as a standard for sea-salt using an empirical formula (eqn (1)–(7)) given in ESI Text 1†.^42,43 Particulate organic matter (POM) was determined to be 1.6 times OC as suggested by Turpin and Lim, 2001, for the urban atmosphere.⁴⁴ Urban dust was calculated using Ca as a tracer element (urban dust = 1.63 × Ca).^45,46 Fe, Ti, Mn, Cr, Ni, Cu, and Zn were categorized as “transition metals”, while Al, Mg, Pb, and Ba were referred to as “other metals”.

2.4. OP measurements

Numerous biological and chemical assays have been devised for evaluating the OP of PM.^{14,17,19,47,48} However, dithiothreitol (DTT) assay, which serves as a surrogate for cellular reductant NADH/NADPH, has been extensively used in several studies to quantify OP of PM samples.^{16,20,49–55} The present study also used DTT assay to determine OP of all collected samples (n = 230). Briefly, 3.5 mL of the water-soluble extract (as mentioned in Section 2.3) was incubated along with 1 mL of potassium phosphate buffer (pH = 7.4) and 0.5 mL of DTT (1 mM) in a water bath shaker at 37 °C. At various time intervals (0, 10, 20, 30 and 40 min), 100 μL was taken out of the incubation mixture and quenched using 1 mL of trichloroacetic acid (TCA; 1% w/v), 2 mL of Tris-buffer (4 mM EDTA with 0.08 M Tris–HCl) and 0.5 mL of 5,5′-dithiobis-2-nitrobenzoic acid (DTNB, 2 mM). The remaining DTT in the incubation mixture reacted with DTNB to produce 2-nitro-5-thiobenzoic acid (TNB), a yellow-coloured product. TNB was subsequently measured using a UV-visible spectrophotometer (Cary 300, Agilent Technologies, USA) at 412 nm. The rate of DTT loss is directly proportional to the presence of redox-active species within PM_2.5. The rate of DTT loss was subsequently normalized by considering the mass of particles (DTTm) and volume of air (DTTv) for further interpretation. All the procedures described above were conducted under dark conditions and using amber vials because both DTT and DTNB are light-sensitive substances. A comprehensive description of the procedure and associated calculations can be found in our previously published work.¹⁶

2.5. Quality assurance/quality control (QA/QC)

All the analytical protocols and standard operating procedures were strictly followed during the chemical analysis. We conducted positive and negative control experiments for the analysis of each chemical constituent of PM_2.5 and OP. Field blank filter extracts were used as negative controls for each batch of analysis (except for EC and OC). Commercially available 1000 ppm multi-element calibration standards (VHG Labs, Manchester, USA) were used as positive controls for analysis of metals in ICP-OES.⁵⁶ These standard solutions were diluted through a serial dilution process using deionized water to the required concentrations (0.01, 0.1, 0.5, 1, 3, and 5 ppm) for constructing the calibration curve in ICP-OES. For IC, the positive controls were prepared using high-purity analytical grade salts (NaCl, NH₄Cl, KCl, MgCl₂, CaCl₂, (NH₄)₂SO₄, and NaNO₃) of different concentrations (0.05, 0.1, 0.5, 1, and 2 ppm). Prior to every analysis, instruments were calibrated using the respective standard solutions to ensure the linearity of each calibration curve with a coefficient of determination (R²) > 0.99 for all species. The standard curve of quantification of some metals and ions is given in the ESI (Fig. S2 and S3†). After standardization, field blank and PM_2.5 samples were analyzed. Duplicate analyses were conducted for all samples, and a field blank was analyzed for every set of ten samples. The calculated mean field blank values for each species were subsequently deducted from their estimated concentrations in the sample values. Species-dependent uncertainties ranged from 3–5% for ions and 1–4% for metals. Furthermore, the minimum detection limit (MDL) for water-soluble ions and metals was calculated at three times the standard deviation of ten repeated blank filter analyses (Table T2†).

Similarly, each PM_2.5 sample was analyzed in duplicate for OC and EC. Filter blanks were analyzed after every set of ten samples to evaluate blank corrections for estimated OC and EC mass concentrations. After every analysis, the flame ionisation detector was calibrated using 5% methane in helium. Furthermore, the analyser was calibrated with potassium hydrogen phthalate (KHP) and sucrose in a mixture of 4.8% CO₂ and helium gas, and this process was repeated after every ten samples.^57,58 The overall uncertainty is within 4% for the estimation of EC and OC and MDL is given in Table T2.†

For OP measurements, 0.21 nmol mL⁻¹ 9,10-phenanthrenequinone (PQN) was used as a positive control.^14,18 The rate of DTT loss obtained from five PQN replicas was 0.79 ± 0.19 nmol min⁻¹, which is similar to the previous reports.^18,59 The measurement was also validated for accuracy by repeating 30% of the analyzed samples in different batches, which showed a 6–9% variation. The rate of DTT loss was measured in blank filters with every experiment and was applied to sample correction. The samples and blanks were analyzed in duplicate, and the relative standard deviation was between 8 and 10%. MDL of the rate of DTT loss was 0.43 nmol min⁻¹, determined as three times the standard deviation of five blank replicates.^18,52

2.6. Source apportionment of OP

In step 1, positive matrix factorization (PMF, version 5.0), a multivariate factor analysis tool, was employed to quantify diverse emission sources that contribute to the levels of PM_2.5.^60–62 Further information on PMF methodology is provided in the ESI (Text 2).† PMF typically needs a large set of data for better source prediction and to avoid potential bias in the data. So, both daytime and nighttime data from the studied periods were amalgamated into one dataset. Subsequently, two distinct PMF analyses were conducted to identify emission sources during periods influenced by MAM and CAM, respectively. For each of these analyses, input data consisting of twenty-two chemical species, along with their associated uncertainty values, were provided to the model. The input matrices had dimensions of [94 × 22] and [136 × 22] for MAM and CAM periods, respectively. Before inputting data into the receptor model, the species were classified based on their signal-to-noise (S/N) ratio. Species with S/N ratio ≤ 0.5 were marked as “bad” and excluded from the analysis. Species with 0.5 ≤ S/N ratio ≤ 1.0 were marked as “weak” while species with S/N ratio ≥ 1.0 were regarded as “strong”.⁶⁰ S/N ratios of all the variables for both the PMF runs are provided in the ESI (Table T3).† To obtain optimal solutions, 3 to 8 factor solutions were investigated and each solution was executed with 100 base model runs. The estimation of the final number of factors involved a comprehensive assessment, by examining the ratio of robust-to-theoretical parameters (Q_R/Q_T) and employing error estimation techniques, which included bootstrap (BS), displacement (DISP), and bootstrap-displacement (BS-DISP) methods.^63–65 BS measures the random error in the matrix, while DISP explores the rotational ambiguity. BS-DISP estimates both random error and rotational ambiguity.^63,64 A detailed overview of the error associated with output profiles for both the PMF runs is provided in the ESI (Tables T4 and T5).† Thus, 6-factor and 5-factor source profiles were identified for MAM and CAM periods, respectively because the PMF solutions were well-mapped in BS-runs with no BS-DISP swaps.

In step 2, PMF was integrated with multiple linear regression (MLR) to assess the impacts of PM_2.5 emission sources on OP. In the regression model, OP (or DTTv) was treated as the dependent variable, while the contributions of sources obtained from the PMF model served as independent variables (eqn (1)).

DTT_v = β₁F₁ + β₂F₂ + … + β_iF_i + C (1)

where DTTv is in nmol min⁻¹ m⁻³; F₁, F₂⋯F_i are the contributions of different sources to the PM_2.5 concentration obtained from the PMF model in μg m⁻³; β₁, β₂⋯β_i represent the intrinsic OP (or DTTm) of the sources (nmol per min per μg source) and C is the constant. However, the Breusch–Pagan test revealed the presence of heteroscedasticity in the residuals of regression analysis. Consequently, we employed the Weighted Least Squares (WLS) approach to address the uncertainties linked to the data.^66,67 Hence, a stepwise regression, constrained to pass through the origin, was performed using the F-test (probability of F-to-enter ≤0.05 and F-to-remove ≥ 0.10) to decide whether the desired variable should be included in the model to prevent overfitting.^51,68–70

Conditional Bivariate Probability Function (CBPF) analysis was conducted to ascertain the potential source direction of OP.⁷¹ CBPF is an elaboration of the Conditional Probability Function (CPF), where WS serves as an additional variable and collaborates with CPF to identify dominant source regions.^57,71 WS and WD collected from IMD for the sampling dates were calculated to obtain the median value. Subsequently, the median WS, WD, and each PMF-MLR modeled factor contribution were used to generate CBPF plots.^72,73

3. Results and discussion

3.1. Mass concentration of PM_2.5 and its chemical characteristics

The daytime and nighttime variation in mass concentrations of PM_2.5 during MAM and CAM periods is depicted in Fig. 1. Mass concentrations of PM_2.5 ranged from 11.9–82.8 μg m⁻³ (mean ± standard deviation: 51.3 ± 16.7 μg m⁻³) and 10.9–218.3 μg m⁻³ (mean ± standard deviation: 84.4 ± 40.2 μg m⁻³ μg m⁻³) in the daytime during MAM and CAM periods, respectively. During nighttime, the mass concentration of PM_2.5 ranged from 14.9–82.3 μg m⁻³ (mean ± standard deviation: 41.2 ± 16.5 μg m⁻³) and 24.1–226.9 μg m⁻³ (mean ± standard deviation: 110.7 ± 59.3 μg m⁻³) in MAM and CAM periods, respectively. A remarkable increase in the mass concentration of PM_2.5 during Diwali and pre-Diwali nights was due to the bursting of firecrackers in and around the city. Overall, the CAM period witnessed a 1.6 times higher mean mass concentration of PM_2.5 during daytime and a 2.7 times higher concentration during nighttime compared to the MAM period. The daytime-to-nighttime (D/N) mean (± standard deviation) mass concentration ratio of PM_2.5 was 1.36 ± 0.52 and 0.86 ± 0.39 during MAM and CAM periods, respectively. Elevated mass concentrations of PM_2.5 during nighttime compared to daytime in the CAM period were attributed to relatively low WS and a suppressed planetary boundary layer, resulting in reduced atmospheric dilution and accumulation of pollutants in the lower atmosphere.^74,75 However, during the MAM period, increased mass concentrations of PM_2.5 were observed in daytime compared to nighttime. This could be due to the development of the thermal internal boundary layer (TIBL) caused by the convergence of sea and land breezes, resulting in the accumulation of PM_2.5 near the surface.⁷⁶ However, this anomaly requires further investigation. Similarly, a higher mass concentration of PM_2.5 was reported in Chennai, a coastal mega-city during daytime compared to nighttime.⁷⁷ In contrast, studies from inland locations such as Patiala, Delhi, and Hyderabad documented higher concentrations of PM_2.5 at night than during the day.^74,75,78

Fig. 1 (a) Variation in the mass concentration of PM_2.5. (b) Box whisker plots describe the distribution of daytime and nighttime mass concentrations of PM_2.5 during MAM and CAM periods in Bhubaneswar. The upper and lower quartiles represent 25th and 75th percentiles, respectively.

The percentage contributions of various PM_2.5 fractions in both periods are depicted in Fig. 2. The mass concentration of PM_2.5 primarily consisted of _nssSO₄²⁻ and POM in both periods. Furthermore, _nssK⁺ ions contributed to 1% of PM_2.5 mass during both daytime and nighttime in the studied periods, suggesting stubble burning had a negligible impact on the study site. Comparable findings were also documented in earlier research conducted at this location.^36,40 During the two studied periods, percentage contributions of transition metals to PM_2.5 mass were found to be 1%, indicating constancy with change in the airmass pattern and the absence of diurnal variation. Furthermore, Zn and Fe were the most abundant water-soluble transition metals in both periods. Similarly, urban dust showed no diurnal variation in the percentage contribution during both periods. However, a difference was observed in the percentage contribution of urban dust during MAM (2%) compared to CAM periods (1%). Enhanced mean mass concentrations of anthropogenic species such as _nssSO₄²⁻, NH₄⁺, NO₃⁻, EC, OC, and measured transition metals were found during CAM compared to MAM periods (Table 1). Conversely, higher sea-salt contributions were found during MAM than during CAM periods. Furthermore, an enhanced contribution of sea-salt during daytime (5%) compared to nighttime (3%) during the MAM period was attributed to the development of a sea breeze over the coastal region. Descriptive statistical characteristics of PM_2.5 and all the estimated fractions, along with the mean D/N ratio, are given in Table 1. The mean D/N ratios of EC were 0.76 and 0.65 during MAM and CAM periods, respectively. Higher nighttime concentrations of EC compared to daytime during the MAM period indicated the prevalence of local anthropogenic sources linked to land breezes. Studies in coastal locations such as Chennai and Thumba also reported elevated nighttime EC concentrations compared to daytime.^76,79 In addition to local emissions, low temperature and a depressed planetary boundary layer at night resulted in higher concentrations of EC than during daytime during the CAM period. Similarly, nighttime maxima (D/N < 1) were observed for all measured species (except sea-salt) during the CAM period. The higher daytime value of OC (D/N > 1) during the MAM period was probably due to contributions from marine biogenic sources.⁷⁶ Secondary inorganic ions (SO₄²⁻, NO₃⁻, and NH₄⁺ represented as SNA) are primarily formed by heterogeneous (aqueous phase oxidation), homogeneous (photochemical oxidation) and particulate phase processes in the atmosphere.⁸⁰ SNA exhibited daytime maxima during the MAM period and nighttime maxima in the CAM period. The daytime maxima during the MAM period were likely due to photochemical oxidation processes, owing to higher temperature over the study site (Table T1†). Conversely, lower temperature at night during the CAM period might have favored the aqueous phase formation of SNA, leading to elevated concentrations during night compared to daytime. Furthermore, the synergistic influence of diverse source strengths, meteorological factors, and primary resuspension processes, as discussed above, might have accumulated pollutants during the nocturnal period.

Fig. 2 Percentage contributions of different species to mass concentrations of PM_2.5 during (a) daytime and (b) nighttime of the MAM period and (c) daytime and (d) nighttime of the CAM period.

Table 1 Descriptive statistical characteristics of PM_2.5 and its chemical composition^a

Species (μg m^−3)	Marine airmass period							Continental airmass period
	Day (D)			Night (N)			D/N	Day (D)			Night (N)			D/N
	Mean	Median	SD	Mean	Median	SD	Mean	Mean	Median	SD	Mean	Median	SD	Mean
a SD = standard deviation. b ng m^−3. c pmol min^−1 μg^−1. d nmol min^−1 m^−3.
PM2.5	51.33	46.91	16.71	41.19	37.84	16.50	1.36	84.38	87.75	40.21	110.73	104.79	59.31	0.86
EC	2.44	2.22	0.97	3.21	2.76	0.99	0.76	5.78	5.33	2.54	9.27	7.41	7.40	0.65
OC	7.40	6.31	2.97	5.54	4.48	3.01	1.34	10.92	9.84	4.54	15.61	13.84	10.64	0.70
POM	11.84	10.10	4.97	8.87	7.16	4.82	1.34	17.47	15.74	7.27	25.03	22.14	16.99	0.70
Na^+	0.73	0.76	0.23	0.60	0.60	0.19	1.22	0.52	0.53	0.18	0.50	0.48	0.18	1.04
NH4^+	3.35	2.91	1.71	2.23	1.98	1.44	1.50	9.97	9.59	5.75	11.47	9.53	8.35	0.87
nssK^+	0.44	0.46	0.23	0.31	0.26	0.22	1.44	0.80	0.74	0.46	0.92	0.75	0.62	0.87
nssMg^2+	0.18	0.22	0.07	0.18	0.14	0.12	1.00	0.19	0.18	0.09	0.19	0.15	0.21	1.00
nssCa^2+	0.55	0.56	0.24	0.52	0.50	0.24	1.05	0.49	0.47	0.25	0.52	0.48	0.23	0.94
nssSO4^2−	12.70	11.42	11.35	9.79	9.22	4.99	1.30	27.82	27.67	14.11	33.72	28.36	22.91	0.83
NO3^−	0.62	0.57	0.34	0.50	0.38	0.40	1.24	3.33	2.65	2.71	4.67	2.61	4.08	0.71
Cl^−	0.51	0.49	0.17	0.39	0.40	0.14	1.31	0.48	0.44	0.25	0.44	0.37	0.24	1.09
Al	33.33	23.50	28.73	28.92	26.25	18.64	1.15	62.70	32.25	28.91	51.39	42.18	41.42	1.22
Mg	76.83	59.00	54.85	76.70	83.00	34.06	1.00	167.36	77.23	107.87	83.28	69.67	58.52	2.01
Ca	508.33	442.00	254.54	391.45	329.50	268.69	1.29	702.95	613.55	418.87	841.57	742.53	545.52	0.84
Pb	3.30	2.00	3.29	3.36	2.00	2.85	0.98	6.88	5.04	7.23	8.55	6.74	7.88	0.80
Ba	3.46	3.00	2.52	3.48	2.50	3.17	0.99	5.73	4.01	5.49	23.18	9.67	63.71	0.25
Ti	4.97	4.44	2.14	4.40	3.95	2.02	1.12	7.12	7.01	3.97	8.59	8.29	4.51	0.83
Cr	19.85	10.50	27.24	14.05	24.28	3.50	1.41	46.27	42.03	35.78	47.07	19.61	41.23	0.98
Mn	4.09	3.00	2.26	2.74	3.00	1.28	1.49	7.34	5.00	5.65	7.22	4.5	6.62	1.02
Fe	381.35	312.00	310.23	271.14	217.10	127.00	1.41	711.22	661.08	509.79	789.34	901.14	649.82	0.90
Ni	4.89	4.00	3.55	3.03	2.00	3.10	1.61	9.51	6.10	9.01	5.10	3.78	3.30	1.86
Cu	20.02	18.00	12.89	23.92	18.00	20.48	0.83	38.69	33.15	32.15	46.59	26.23	45.41	0.83
Zn	44.20	43.50	26.81	36.47	28.00	24.90	1.21	83.87	79.65	50.27	58.87	48.67	39.53	1.42
∑ transition metal	0.48	0.35	0.35	0.36	0.19	0.33	1.33	0.90	0.85	0.59	1.03	1.10	0.79	0.87
∑ other metals	0.12	0.11	0.06	0.11	0.10	0.005	1.04	0.24	0.16	0.22	0.17	0.14	0.13	1.41
Sea-salt	2.50	1.47	1.11	1.11	1.08	0.46	2.25	1.17	1.15	0.53	1.03	0.96	0.54	1.14
Urban dust	0.83	0.72	0.41	0.64	0.54	0.44	1.30	1.15	1.00	0.68	1.37	1.21	0.89	0.84
DTTm	19.81	19.10	7.06	17.32	17.16	6.19	1.25	21.57	21.78	6.95	20.13	19.85	6.16	1.22
DTTv	1.03	0.92	0.64	0.75	0.58	0.49	1.76	1.78	1.75	0.90	2.24	2.53	1.11	1.04

---

3.2. DTT activity and comparison with other locations

The time series variation of mass-normalized DTT (DTTm; also known as intrinsic OP) and volume-normalized DTT (DTTv; also called extrinsic OP) is shown in Fig. 3. The daytime and nighttime DTTm activities ranged from 6.98–40.68 pmol min⁻¹ μg⁻¹ (mean ± standard deviation: 19.81 ± 7.06 pmol min⁻¹ μg⁻¹) and 6.62–27.36 pmol min⁻¹ μg⁻¹ (mean ± standard deviation: 17.32 ± 6.19 pmol min⁻¹ μg⁻¹), respectively during the MAM period. DTTm activities varied from 4.49–40.47 pmol min⁻¹ μg⁻¹ (mean ± standard deviation: 21.57 ± 6.95 pmol min⁻¹ μg⁻¹) during daytime and 8.59–40.45 pmol min⁻¹ μg⁻¹ (mean ± standard deviation: 20.13 ± 6.16 pmol min⁻¹ μg⁻¹) during nighttime in the CAM period. Unlike PM_2.5, DTTm activity showed no substantial variation during MAM and CAM periods. Paired t-tests and F-tests also did not show significant variation in daytime and nighttime activities of DTTm during both periods. The mean differences in daytime and nighttime DTTm activities were 2.11 pmol min⁻¹.μg⁻¹ (95% CI: −0.40, 4.61) and 1.44 pmol min⁻¹ μg⁻¹ (95% CI: −1.03, 3.91) during MAM and CAM periods, respectively. The consistency in DTTm activities in contrasting air masses can be attributed to the absence of noteworthy changes in the percentage contribution of POM and transition metals (regarded as DTT active species) to the mass concentration of PM_2.5 in both periods (as discussed in Section 3.1). In contrast, DTTv followed a similar variation pattern to PM_2.5. Daytime DTTv activities ranged from 0.08–3.91 nmol min⁻¹ m⁻³ (mean ± standard deviation: 1.03 ± 0.64 nmol min⁻¹ m⁻³) and 0.23–4.04 nmol min⁻¹ m⁻³ (mean ± standard deviation: 1.78 ± 0.90 nmol min⁻¹ m⁻³) in MAM and CAM periods, respectively. Nighttime DTTv activities varied from 0.13–2.23 nmol min⁻¹ m⁻³ (mean ± standard deviation: 0.75 ± 0.49 nmol min⁻¹ m⁻³) and 0.25–3.91 nmol min⁻¹ m⁻³ (mean ± standard deviation: 2.24 ± 1.11 nmol min⁻¹ m⁻³) in MAM and CAM periods, respectively. Thus, considerably higher DTT_V activity was observed during CAM than in MAM periods, substantiating that extrinsic OP is highly influenced by the mass concentration of PM_2.5, meteorological factors, and emission source strength.^16,81 Furthermore, the t-test showed a significant statistical daytime and nighttime variation in DTTv activities during both periods, but the mean differences were smaller in magnitude. Like PM_2.5, daytime DTTv activities were higher (mean difference between day and night: 0.22 nmol min⁻¹ m⁻³, 95% CI: 0.06, 0.41) during the MAM period and lower (mean difference between day and night: −0.46 nmol min⁻¹ m⁻³, 95% CI: −0.69, −0.22) during the CAM period compared to nighttime. The different patterns of daytime and nighttime distribution of DTTv activity during MAM and CAM periods could also be attributed to a higher D/N ratio (>1) of DTT active species (POM and sum of transition metals) during MAM compared to CAM periods.

Fig. 3 Time series variation of (a) DTTm and (b) DTTv activities during MAM and CAM periods. Daytime and nighttime distribution of (c) DTTm activity and (d) DTTv activity for both studied periods. The upper and lower quartiles represent 25th and 75th percentiles, respectively.

Activities of DTTv and DTTm of PM_2.5 reported from different parts of India and globally are outlined in Table 2. Investigations from Delhi and several cities in the IGP region (aerosol hotspot of India), such as Patiala,²⁵ Kanpur,²⁹ Haryana,³⁰ and Uttar Pradesh,³⁰ documented higher mean values of DTTm and DTTv activities of PM_2.5 compared to the study site. These studies revealed that unregulated combustion sources (biomass burning and emissions from local activities) significantly influence the OP in these regions. Furthermore, a recent toxicological study from Delhi underscored that nearly 67% of redox active species in PM_2.5 reside in the water-soluble phase, which are primarily emitted from biomass burning.³³ Similar observations were also reported from Fairbanks, Alaska, indicating that 77% of OP is derived from water-soluble species of PM_2.5 that came from wood combustion.⁸² Notably, these findings highlighted that PM_2.5 from non-fossil fuel combustion sources exhibited higher catalytical activity compared to fossil-fuel combustion. Investigations of catalytic activity of PM_2.5 at four different urban sites (Los Angeles, Milan, Athens, and Beirut) around the world also reported similar observations.⁸³ Enhanced DTT activities were observed in Milan (DTTm: 65.29 ± 5.17 pmol min⁻¹ μg⁻¹; DTTv: 3.38 ± 0.26 nmol min⁻¹ m⁻³) and Athens (DTTm: 49.20 ± 1.66 pmol min⁻¹ μg⁻¹; DTTv: 5.62 ± 0.19 nmol min⁻¹ m⁻³) due to the influence of biomass burning, whereas Beirut (DTTm: 8.22 ± 1.27 pmol min⁻¹ μg⁻¹) and Los Angeles (DTTm: 28.10 ± 5.23 pmol min⁻¹ μg⁻¹; DTTv: 0.35 ± 0.04 nmol min⁻¹ m⁻³) showed comparatively lower activities due to the predominance of industrial and traffic-related emissions, respectively.⁸³ DTTv activities ranged from 0.03–1.06 nmol min⁻¹ m⁻³ and 0.66–1.03 nmol min⁻¹ m⁻³ in high-traffic urban slum areas of Mumbai²⁰ and Bangalore,¹⁵ respectively. Similarly, lower DTT activities (DTTm: 17.7 pmol min⁻¹ μg⁻¹; DTTv: 0.36 nmol min⁻¹ m⁻³) were observed at Dunkerque, a coastal city in France, where OP is dominated by fossil fuels from industrial activities.⁸⁴ DTTv activity at our site was comparable to the values reported in São Paulo, Brazil, while DTTm activity was notably higher.⁶⁹ In addition, reports from Beijing,⁸⁵ East China,⁵⁰ North China,⁵¹ and Tehran in Iran⁸⁶ documented higher DTT activities compared to the current study. It is important to note that studies from developed countries such as Greece,⁵² Italy,^87,88 Japan,⁸⁹ and the United States⁵⁵ exhibited comparatively lower DTTv activities but higher DTTm activities compared to developing countries.

Table 2 Comparison of PM_2.5 (μg m⁻³) mass concentrations and DTTv (nmol min⁻¹ m⁻³) and DTTm (pmol min⁻¹ μg⁻¹) activities in different parts of India and other countries

Sampling station	Sampling period	Mass concentration of PM2.5	DTTv	DTTm	Reference
Bhubaneswar	April–May 2019	46.26 ± 16.61	0.89 ± 0.57	18.57 ± 6.63	Present study
	October–December 2019	97.56 ± 46.65	2.01 ± 1.01	20.85 ± 6.56
Patiala, India	October–November 2014	229 ± 115	11.6 ± 4.9	54 ± 17	25
Patiala, India	January–February 2014	150 ± 53	3.8 ± 1.4	27 ± 8	23
Kanpur, India	Winter, 2015	167 ± 53	5.5 ± 1.5	34 ± 8.5	29
Delhi, India	February 2019		1.5 ± 0.7		27
Delhi, India	June 2019	50 ± 15	2.9 ± 1.1	61 ± 29	31
Delhi, India	October–December 2019	208 ± 132	3.3 ± 1.97	19 ± 8.43	32
Delhi, India	September 2022	22.5 ± 10.8	0.08–1.52 (range)	27.5	33
Bangalore, India	October 2013		0.66–1.03		15
São Paulo, Brazil	July–December 2019		1.22 ± 0.55	91 ± 6.7	69
Milan, Italy	December 2018–February 2019	71.82 ± 4.17	3.38 ± 0.46		88
	May–July 2019	16.67 ± 0.27	0.85 ± 0.10
Tehran, Iran	February 2014–January 2015	38.4 ± 16.3	9.0 ± 0.47		86
Athens, Greece	July 2016–July 2017		0.33 ± 20	27.98 ± 14.42	52
Bologna, Italy	2011–2014		0.3–1.7 (range)		14
Leece, Italy	2013–2015	31.9 ± 19.2	0.40 ± 0.26	14.5 ± 7.6	87
Atlanta, USA	June 2012–September 2013		0.1–1.5 (range)	5–100 (range)	55
Fairbank, Alaska	January–February 2022	12.8 ± 11.1	0.42	35	82
Dunkerque, France	2010–2011	29.2 ± 24.4	0.36	17.7	84
Hanoi, Vietnam	September 2019–December 2020	40.2 ± 26.3	3.9 ± 2.4
Milan, Italy	December 2018–February 2019		3.38 ± 0.26	65.29 ± 5.17	83
Athens, Greece	August 2020		5.62 ± 0.19	49.20 ± 1.66
Beirut, Lebanon	March 2020 and May 2020		3.51 ± 0.54	8.22 ± 1.27
Los Angeles, USA	August 2018; December 2018–January 2019		0.35 ± 0.04	28.10 ± 5.23
Lahore, Pakistan	February 2019	522 ± 222	8.9 ± 3.8	18 ± 10.5	49
Peshawar, Pakistan		522 ± 222	9.3 ± 3.9	27.7 ± 2.1
Jinzhou, China	May 2015–April 2016	114.6 ± 50.5	4.4 ± 2.6	35 ± 18	51
Liaoning Province, China		120.1 ± 54.8	6.8 ± 3.4	49 ± 16
Tianjin, China		113.8 ± 88.3	4.2 ± 2.7	30 ± 16
Fukuoka, Japam	March–May 2016	16.7 ± 6.3	0.75 ± 0.30	45.3 ± 11.4	89
Beijing, China	May 2015–April 2016	113.8 ± 62.7	12.26 ± 6.82	0.13 ± 0.10	85
Ningbo, China	October 2017–August 2018	87.6 ± 35.6	3.65 ± 1.71	120 ± 120	50

---

3.3. Association of DTT activity with chemical constituents of PM_2.5

Spearman's rho correlation (ρ) was obtained to examine the relationship of DTTv activity with the measured chemical species of PM_2.5. We assumed ρ < 0.5 as weak and ρ ≥ 0.5 as strong correlations for all 230 samples. The correlation plots for day and night during both periods are depicted in Fig. 4. EC (daytime ρ = 0.6 and nighttime ρ = 0.5 during MAM; daytime ρ = 0.6 and nighttime ρ = 0.8 during CAM) and OC (daytime ρ = 0.6 and nighttime ρ = 0.8 during MAM; daytime ρ = 0.6 and nighttime ρ = 0.8 during CAM) exhibited a strong association with DTTv during both the periods. This result aligns with findings from earlier research.^{16,25,51,55,88} Even though EC is not an active redox species, the strong association with DTTv might be due to co-emission with OC. Furthermore, a strong correlation between OC and EC (ρ = 0.8 during MAM; ρ = 0.9, during CAM) confirms the above observation. Transition metals such as Fe, Cu, Zn, and Cr showed a strong association with DTTv activity. A strong association of DTTv activity was also observed with _nssSO₄²⁻ (daytime ρ = 0.6 and nighttime ρ = 0.5 during MAM; daytime ρ = 0.8 and nighttime ρ = 0.7 during CAM), whereas there was a weak to strong association with NO₃⁻ (daytime ρ = 0.4 and nighttime ρ = 0.5 during MAM; daytime ρ = 0.6 and nighttime ρ = 0.7 during CAM). Furthermore, we found a good correlation between Fe, Cu and Zn with _nssSO₄²⁻ and NO₃⁻ (ρ > 0.5). Although several laboratory-based studies have found SNA to be DTT inactive,^25,55,90 the association of DTTv with _nssSO₄²⁻ and NO₃⁻ in our study is likely due to concurrent emission with Fe, Cu and Zn. These observations further suggest that the predominance of acidic species at the site (25% and 37% of PM_2.5 mass during MAM and CAM, respectively) promotes the intermixing of transition metals,⁹¹ thus contributing to the toxicity of _nssSO₄²⁻ and NO₃⁻. This finding aligns with numerous epidemiological studies that have identified SO₄²⁻ and NO₃⁻ as significant contributors to PM_2.5 toxicity as well as the initiators of ROS generation in humans.^92,93 The concomitant variability of major emissions presents a challenge to predict which species are predominantly responsible for inducing OP over the site.

Fig. 4 Spearman's rho correlation (ρ) plots between DTTv activity and measured chemical species during (a) daytime of the MAM period, (b) nighttime of the MAM period, (c) daytime of the CAM period and (d) nighttime of the CAM period.

3.4. Source apportionment

3.4.1. Source apportionment of PM_2.5. As discussed in Section 2.5, the PMF analysis identified six major source factors during the MAM period and five dominant source factors during the CAM period, determined by the abundance of tracer compounds. The major sources common to both periods include industrial emissions, vehicular emissions, urban dust, combustion sources (wood/coal/waste), and secondary aerosols. Notably, sea salt was an additional contributor during the MAM period. Fig. 5 depicts the compositional profile of species (percentage of each species) determined by PMF, along with the percentage contributions of various sources during MAM and CAM periods. Furthermore, the day and night percentage contributions of various sources in both periods are illustrated in Fig. S4 of the ESI.† The identified factors were explained as follows.

Fig. 5 Percentage contribution of different chemical species to the factors determined by PMF during (a) MAM period and (c) CAM period. Relative percentage contribution of different sources to PM_2.5 during (b) MAM and (d) CAM periods.

3.4.1.1 Vehicular emissions. This factor was categorized by elevated percentage contributions of EC, OC, Cu, NO₃⁻, Zn, and Al. Studies have reported that OC, EC, and NO₃⁻ are primarily generated from the combustion of gasoline and diesel.^94–96 In addition to exhaust emissions, mechanical abrasion of brakes and tires emits Cu, Zn, and Al.^97–100 Studies have documented that brake pads are primarily composed of Cu, Fe, Mn, Zn and Al, while brake discs of vehicles are mainly made of Fe.^98,99,101 However, tire threads usually consist of zinc oxide and organozinc compounds, and it has been well documented that tire ware emits approximately 15 times more Zn than brake ware.¹⁰¹ Moreover, Ca, Al, and Fe also originate from the abrasion of road surfaces.^98,101 Thus, vehicular emissions accounted for 21% of the mean PM_2.5 mass in both periods. Furthermore, the daytime contributions were 21% and 22% and nighttime were 24% and 23% during MAM and CAM periods, respectively (Fig. S4†). A tangential increase in personal vehicles and car ownership has resulted in the substantial contribution of this factor to the region.¹⁰²
3.4.1.2 Industrial sources. The main constituents defining this factor were Fe, Cr, SO₄²⁻, Al, EC, and OC. Previous studies by our group revealed that several thermal power stations and integrated steel plants, located to the north-east, directly influenced the study site during the CAM period.³⁵ The power plants include the Talcher Super Thermal Power Station (∼162 km from the study location), Jharsuguda Thermal Power Plant (326 km), Kobra Super Thermal Power Plant (502 km), Sipat Thermal Power Plant (528 km), Rihand Thermal Power Station (704 km), and Vindhyachal Thermal Power Station (725 km).³⁵ Thermal power plants use coal, which emits large amounts of SO₂ that undergoes atmospheric processing and converts into SO₄²⁻ during long-range transport.¹⁰³ The integrated steel plants include the Steel Authority of India Limited in Bokaro (∼497 km from the study location), the Steel Authority of India Limited in Bhilai (575 km), the Steel Authority of India Limited, Rourkela (314 km), Tata Steel Limited in Jamshedpur (392 km), Jindal Steel & Power Limited in Raigarh (397 km), Jindal Steel & Powe Limited, Dhenkanal (314 km), and Bhusan Power & Steel Limited in Angul (135 km).³⁵ Furthermore, several small and medium-scale galvanizing, casting, manufacturing, and metal processing units are located in the city and neighboring areas.^104,105 These industries mostly use low grade coals to fulfill their energy demands without any specific fuel usage pattern, thus contributing to SO₄²⁻, EC, and OC concentrations in the urban atmosphere.^72,106 Integrated steel plants, along with the processing units located within the city, also emit Fe, Al, and Cr as byproducts of various metallurgical and manufacturing processes.¹⁰⁶ Elevated concentrations of NH₄⁺ and NO₃⁻ were specifically noted during the MAM period, which might be due to leading fertilizer (IFFCO, Paradeep Phosphates Limited) and oil refineries (Indian Oil Corporation, Bharat Petroleum, and Hindustan Petroleum) situated within 120 km to the east and south-east of the study site.^107,108 Thus, industrial sources accounted for 13% (daytime: 18% and nighttime: 17%) of mean PM_2.5 mass during the MAM period. However, during the CAM period, industrial sources contributed 23% and 17% in daytime and nighttime, respectively with an overall contribution of 21% to mean PM_2.5 mass. The increased contribution during CAM is attributed to the direct influence from the upwind Chota Nagar Plateau, which accommodates a substantial number of thermal power plants.³⁵ Secondly, higher WS during daytime raises the possibility of greater regional impact in a higher planetary boundary layer scenario, accounting for larger industrial sources impacting the planetary boundary layer.
3.4.1.3 Combustion (wood/coal/waste) sources. The factor showed a high abundance of K⁺ in both periods.^74,109 Additionally, there were moderate to low contributions from SO₄²⁻, Cr, OC, EC, and NH₄⁺. While K⁺, OC, EC, and NH₄⁺ in this profile were likely emitted from wood and coal combustion, SO₄²⁻ was predominantly emitted by coal combustion.^57,72 The city has a large slum population, constituting 30% of the total population and these settlements are widely dispersed throughout the municipal area.^110,111 A recent study in the city unveiled that over 40% of slum residents commonly rely on wood and coal for their domestic cooking needs.¹⁰⁵ Similarly, 60% of local vendors and motels utilize wood and coal for preparing a variety of food items to earn their livelihood.¹⁰⁵ Thus, combustion from various sources accounted for 21% and 19% of mean PM_2.5 mass during MAM and CAM periods, respectively. Furthermore, the practice of open-cast burning of plastic waste/garbage, particularly during nighttime, emits considerable amounts of OC, EC, Cr, and SO₄²⁻ amplifying the impact of this source at night (MAM: 20% and CAM: 19%) compared to day (MAM: 16% and CAM: 14%).^72,112
3.4.1.4 Urban dust. This factor was mainly defined by high levels of Mg and Ca²⁺ in both periods, typically associated with construction dust.^104,106,113 Urban dust accounted for 3% and 6% of the mean mass concentration of PM_2.5 during MAM and CAM periods, respectively. Construction activities are most common in the city as it is undergoing massive development in infrastructure, shopping complexes, recreational centres, roadways, and residential neighbourhoods. Furthermore, moderate to low loadings of Al, Fe, Zn, Cr, and Cu indicated the influence of road dust and industrial dust.¹¹³ Urban dust showed a relatively high percentage contribution during daytime (MAM: 5% and CAM: 10%) compared to nighttime (MAM: 4% and CAM: 6%) (Fig. S4†) in both periods signifying that urban emissions are more pronounced during the day.
3.4.1.5 Secondary aerosols. SO₄²⁻, NO₃⁻ and NH₄⁺ were the primary contributors to this factor and are formed by the transformation of gases into particles in the atmosphere as discussed in Section 3.1. Secondary aerosols contributed 26%, with contributions of 24% and 22%, in daytime and nighttime, respectively during MAM. However, secondary aerosols accounted for 31% and 35% in daytime and nighttime, respectively, with an overall contribution of 33% to the mean PM_2.5 mass during the CAM period.
3.4.1.6 Sea-salt. Sea-salt contribution to PM_2.5 was identified only during the MAM period. The chemical profile was dominated by Na⁺ and Cl⁻, comprising 16% of mean PM_2.5 mass with daytime and nighttime contributions of 16% and 13%, respectively. These sea-salts are produced through mechanical disturbance on the surface of the ocean and then carried inland by sea air masses.¹¹⁴

3.4.2. Source apportionment of DTT activity. The predicted regression equations obtained from MLR analysis for MAM and CAM periods are shown in eqn (2) and (3), respectively. Additionally, the normalized regression equations obtained for MAM and CAM periods are given eqn (4) and (5), respectively. The step-wise regression output for both periods is provided in the ESI (Table T6).† The contribution from sea-salt sources during the MAM period was excluded during the step-wise regression procedure due to its insignificant contribution. It is relevant to note that the regression model could explain 63% variation in DTTv activity in both periods, while 37% of the variation remains unexplained (Fig. S5†). The coefficient of the regression equation represents DTTm activity (or intrinsic OP) for each source.

DTTv = [(0.282 ± 0.053) × vehicular emission] + [(0.209 ± 0.053) × combustion sources] + [(0.131 ± 0.043) × urban dust] + [(0.111 ± 0.055) × secondary aerosol] + [(0.104 ± 0.040) × industrial sources] (2)

DTTv = [(0.687 ± 0.091) × vehicular emission] + [(0.452 ± 0.113) × combustion sources] + [(0.101 ± 0.44) × urban dust] + [(0.364 ± 0.067) × secondary aerosol] + [(0.339 ± 0.087) × industrial sources] (3)

DTTv = 0.317 × vehicular emission + 0.264 × combustion sources + 0.18 × urban dust + 0.154 × secondary aerosol + 0.203 × industrial sources (4)

DTTv = 0.393 × vehicular emission + 0.214 × combustion sources + 0.088 × urban dust + 0.238 × secondary aerosol + 0.16 × industrial sources (5)

Although secondary sources dominated the mean mass of PM_2.5, vehicular emissions constituted the largest fraction of the estimated DTTv activity in both periods (mean MAM: 18% (day: 19% and night: 16%); mean CAM: 23% (day: 23% and night: 22%)) (Fig. 6). The Ease of Moving Index 2022 report highlights that half of the city's population owned two-wheelers and 11% had personal cars.^102,115 The growing reliance on private vehicles has not only exacerbated traffic congestion¹¹⁶ but significantly increased redox active species such as Zn, Cu, and EC in the urban atmosphere. Furthermore, CBPF plots revealed an association of vehicular emission with lower WS, indicating the predominance of non-buoyant local sources (Fig. 7a and 8a). Vehicular emissions approximately exhibited two times higher DTTm activity during CAM (687 ± 91 pmol per min per μg source) than during MAM periods (282 ± 53 pmol per min per μg source) (Fig. 9). Low ambient temperature and the depressed boundary layer during the CAM period resulted in poor dispersion of pollutants, which consequently resulted in the accumulation of redox-active species in the lower atmosphere. Several previous studies conducted in urban backgrounds also reported higher DTT activity from vehicular emission.^85,117,118 Emissions from combustion sources (mean MAM: 15% (day: 14% and night: 16%)) were the second highest contributors to DTTv activity during the MAM period, followed by industrial sources (mean MAM: 11% (day: 13% and night: 10%)), urban dust (mean MAM: 10% (day: 8% and night: 12%)) and secondary aerosols (mean MAM: 9% (day: 9% and night: 9%)) (Fig. 6). However, during the CAM period, the percentage contribution of sources to DTTv activity followed a slightly different trend, with secondary aerosols (mean CAM: 14% (day: 13% and night: 15%)) as the second dominant contributor, followed by combustion sources (mean CAM: 12% (day: 11% and night: 13%)), industrial sources (mean CAM: 9% (day: 10% and night: 9%)) and urban dust (mean CAM: 5% (day: 6% and night: 4%)) (Fig. 6). CBPF plots for combustion sources were dominated by south-east and east directions with low WS during the MAM period (Fig. 7b), signifying the emission of redox-active species from cooking activities by slum dwellers, street food vendors and motels (as discussed in Section 3.4.1) in and around the study site. However, during the CAM period, combustion sources were scattered in the north-east direction (Fig. 8b), indicating the local origin of redox-active species as well as transport from regional sources. Stubble burning in the upwind region might have transported redox active species to the study site along with northerly wind. The resemblance between the extrinsic OP (DTTv activity) of our site and that of São Paulo, Brazil, may be influenced by the commonalities in source profiles for combustion, as well as secondary sources, and similarities in locations with coastal influence.^69,119 Similarly, industrial emissions were linked to low WS in the south-east and east directions in the MAM period (Fig. 7c), indicating significant emissions of DTT active species from small-scale industries within the city. In contrast, industrial emissions were more pronounced in the north-east direction during the CAM period (Fig. 8c), signifying long-range transport of DTT active species from the densely industrialized belt of the Chota Nagar Plateau to the study region. Aging of aerosols by chemical processes during long-range transport³⁹ possibly enhanced DTTm activities (or inherent toxicity) of species emitted from industrial (MAM: 104 ± 40 pmol per min per μg source; CAM: 339 ± 87 pmol per min per μg source) and combustion sources (MAM: 209 ± 53 pmol per min per μg source; CAM: 452 ± 113 pmol per min per μg source) during CAM compared to MAM¹⁶ (Fig. 9). Secondary aerosols and urban dust also showed higher probability of low WS in the MAM period (Fig. 7d and e), indicating that DTT active species were mostly associated with local sources. Conversely, during CAM, secondary aerosols were associated with higher WS in the north-west directions, while urban dust exhibited association with both north-west and north-east winds, indicating the significant impact of regional transport on enhancing DTT activity (Fig. 8d and e).

Fig. 6 Time series plot of DTTv activity segregated into different factors as predicted by PMF-MLR during (a) MAM and (b) CAM periods. The inserted pie charts represent the daytime and nighttime contributions during both periods.

Fig. 7 CBPF plots for each PMF-MLR modeled factor at 75 percentiles during the MAM period (wind speed is represented as km h⁻¹).

Fig. 8 CBPF plots for each PMF-MLR-WLS modeled factor at 75 percentiles during the CAM period (wind speed is represented as km h⁻¹).

Fig. 9 DTTm activity of each source as predicted by MLR during MAM and CAM periods.

4. Conclusions

Continuous day and night PM_2.5 samples (n = 230) were collected during MAM (April–May 2019) and CAM (end of October–December 2019) periods in Bhubaneswar, one of the non-attainment cities situated on the east coast of India. The primary objective was to identify major emission sources inducing OP in PM_2.5. DTTm activities of PM_2.5 showed no significant variation during both periods (CAM: 21.57 ± 6.95 pmol min⁻¹ μg⁻¹ in daytime and 20.13 ± 6.16 pmol min⁻¹ μg⁻¹ in nighttime; MAM: 19.81 ± 7.06 pmol min⁻¹ μg⁻¹ in daytime and 17.32 ± 6.19 pmol min⁻¹ μg⁻¹ in nighttime). This can be attributed to the absence of noteworthy changes in the percentage contribution of DTT active species such as POM and transition metals to the mass concentration of PM_2.5 in both periods. However, like PM_2.5, DTTv activities exhibited significantly higher values during the CAM period (daytime: 1.78 ± 0.90 nmol min⁻¹ m⁻³ and nighttime: 2.24 ± 1.11 nmol min⁻¹ m⁻³) than during the MAM period (daytime: 1.03 ± 0.64 nmol min⁻¹ m⁻³ and nighttime: 0.75 ± 0.49 nmol min⁻¹ m⁻³). PMF analysis identified secondary aerosols (MAM: 26% and CAM: 33%) as the dominant source contributing to the mass concentration of PM_2.5 in both periods. However, MLR analysis unveiled vehicular emissions (21%) as the primary source inducing OP in PM_2.5. CBPF plot analysis highlighted that local sources were primarily responsible for inducing OP of PM_2.5. Furthermore, stagnant meteorological conditions and chemical aging of species during regional transport from north-east directions, along with local sources, enhanced redox activity during the CAM period. Unlike previous studies from India, which identified unregulated combustion sources as the major contributor to OP, our study revealed that traffic-related emissions are primarily responsible for inducing OP of PM_2.5 in Bhubaneswar.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its ESI files.† Should any raw data files be needed in another format they are available from the corresponding author upon reasonable request.

Author contributions

Subhasmita Panda: conceptualisation, methodology, formal analysis, data curation, writing – original draft, and writing – review and editing. Chinmay Mallik: data curation, writing – review and editing. S. Suresh Babu: writing – review and editing. Sudhir Kumar Sharma and Tuhin Mandal: formal analysis, writing – review and editing. Trupti Das: funding acquisition, formal analysis, investigation, resources, writing – review and editing. R. Boopathy: supervision, investigation, visualization, validation, resources, funding acquisition, writing – review and editing.

Conflicts of interest

The authors declare there is no conflict of interest on submission of this manuscript.

Acknowledgements

The authors express their gratitude to the ARFI network project under ISRO-GBP for providing financial assistance. Appreciation is extended to the Director of CSIR-IMMT and Head of the E & S Department for their provisions. SP acknowledges the contribution of Ms Jyotishree Nath and Ms Monalin Mishra, Research Scholars, CSIR-IMMT, for assisting during sampling, sample preparation, and analysis. Special thanks to Ms Lovely Mohapatra, a summer intern, for her assistance with OP analysis. SP is grateful to CSIR for the CSIR-SRF fellowship (CSIRAWARD/SRF-DIRECT2021/4309).

References

1. R. Fuller, P. J. Landrigan, K. Balakrishnan, G. Bathan, S. Bose-O'Reilly and M. Brauer, et al., Pollution and health: a progress update, Lancet Planet Health, 2022, 6(6), e535–e547.
2. H. E. Institute, State of Global Air 2019. Special Report, Health Effects Institute, Boston, MA, 2019.
3. I. Braithwaite, S. Zhang, J. Kirkbride, D. Osborn and J. Hayes, Air Pollution (Particulate Matter) Exposure and Associations with Depression, Anxiety, Bipolar, Psychosis and Suicide Risk: A Systematic Review and Meta-Analysis, Environ. Health Perspect., 2019, 127, 126002.
4. D. Chate, G. Beig, T. Satpute, S. Sahu, K. Ali and N. Parkhi, et al., Assessments of population exposure to environmental pollutants using air quality measurements during Commonwealth Games-2010, Inhalation Toxicol., 2013, 25, 333–340.
5. S. Khomenko, E. Pisoni, P. Thunis, B. Bessagnet, M. Cirach and T. Iungman, et al., Spatial and sector-specific contributions of emissions to ambient air pollution and mortality in European cities: a health impact assessment, Lancet Public Health, 2023, 8, e546–e558.
6. D. Schraufnagel, J. Balmes, C. Cowl, S. De Matteis, S. H. Jung and K. Mortimer, et al., Air Pollution and Noncommunicable Diseases: A Review by the Forum of International Respiratory Societies’ Environmental Committee, Part 1: The Damaging Effects of Air Pollution, Chest, 2018, 155, 409–416.
7. A. Pandey, M. Brauer, M. L. Cropper, K. Balakrishnan, P. Mathur and S. Dey, et al., Health and economic impact of air pollution in the states of India: the Global Burden of Disease Study 2019, Lancet Planet Health, 2020, 5, e25–e38.
8. R. Sivarethinamohan, S. Sujatha, S. Priya, G. A. Sankaran and Z. Rahman, Impact of air pollution in health and socio-economic aspects: Review on future approach, Mater. Today: Proc., 2021, 37, 2725–2729.
9. S. S. Kumar, B. Bagepally and R. Balachandar, Air pollution attributed disease burden and economic growth in India: Estimating trends and inequality between states, Lancet Reg. Health Southeast Asia, 2022, 7, 100069.
10. M. Nair, H. Bherwani, S. Mirza, S. Anjum and R. Kumar, Valuing burden of premature mortality attributable to air pollution in major million-plus non-attainment cities of India, Sci. Rep., 2021, 11(1), 22771.
11. P. Joshi, S. Dey, S. Ghosh, S. Jain and S. K. Sharma, Association between Acute Exposure to PM 2.5 Chemical Species and Mortality in Megacity Delhi, India, Environ. Sci. Technol., 2022, 56, 7275–7287.
12. J. T. Bates, R. J. Weber, J. Abrams, V. Verma, T. Fang and M. Klein, et al., Reactive Oxygen Species Generation Linked to Sources of Atmospheric Particulate Matter and Cardiorespiratory Effects, Environ. Sci. Technol., 2015, 49(22), 13605–13612.
13. N. Rastogi and A. Patel, Oxidative potential of ambient aerosols: an Indian perspective, Curr. Sci., 2017, 112(1), 35–39.
14. M. Visentin, A. Pagnoni, E. Sarti and M. C. Pietrogrande, Urban PM2.5 oxidative potential: Importance of chemical species and comparison of two spectrophotometric cell-free assays, Environ. Pollut., 2016, 219, 72–79.
15. H. Vreeland, J. J. Schauer, A. G. Russell, J. D. Marshall, A. Fushimi and G. Jain, et al., Chemical characterization and toxicity of particulate matter emissions from roadside trash combustion in urban India, Atmos. Environ., 2016, 147, 22–30.
16. S. Panda, S. S. Babu, S. K. Sharma, T. Mandal, T. Das and B. Ramasamy, Role of South Asian outflow on the oxidative potential of marine aerosols over the Indian Ocean, Sci. Total Environ., 2023, 164105.
17. A. Patel and N. Rastogi, Oxidative Potential of Ambient PM and Related Health Endpoints over South Asia: A Review, Asian J. Atmos. Environ., 2021, 15, 1–11.
18. T. Fang, V. Verma, H. Guo, L. E. King, E. S. Edgerton and R. J. Weber, A semi-automated system for quantifying the oxidative potential of ambient particles in aqueous extracts using the dithiothreitol (DTT) assay: results from the Southeastern Center for Air Pollution and Epidemiology (SCAPE), Atmos. Meas. Tech., 2015, 8(1), 471–482.
19. J. T. Bates, T. Fang, V. Verma, L. Zeng, R. J. Weber and P. E. Tolbert, et al., Review of Acellular Assays of Ambient Particulate Matter Oxidative Potential: Methods and Relationships with Composition, Sources, and Health Effects, Environ. Sci. Technol., 2019, 53(8), 4003–4019.
20. A. Anand, S. Yadav and H. C. Phuleria, Chemical characteristics and oxidative potential of indoor and outdoor PM2.5 in densely populated urban slums, Environ. Res., 2022, 212, 113562.
21. I. Goyal, P. K. Verma, K. M. Kumari and A. Lakhani, Dynamic changes in the characteristics of fine particles and their oxidative potential in the city of Taj (Agra, India): the untold story of fireworks display, Air Qual., Atmos. Health, 2023, 16(11), 2193–2207.
22. A. Patel, N. Rastogi, S. Rangu, J. Dave, A. Borgohain and S. S. Kundu, Oxidative potential and hydroxyl radical generation capacity of ambient PM2.5 over a high-altitude site in northeastern Himalaya: Role of long-range transport, Atmos. Environ., 2022, 287, 119263.
23. A. Patel, R. Satish and N. Rastogi, Remarkably High Oxidative Potential of Atmospheric PM2.5 Coming from a Large-Scale Paddy-Residue Burning over the Northwestern Indo-Gangetic Plain, ACS Earth Space Chem., 2021, 5(9), 2442–2452.
24. A. Patel, N. Rastogi, U. Gandhi and N. Khatri, Oxidative Potential of Atmospheric PM10 at Five Different Sites of Ahmedabad, a Big City in Western India, Environ. Pollut., 2021, 268, 115909.
25. A. Patel and N. Rastogi, Oxidative potential of ambient fine aerosol over a semi-urban site in the Indo-Gangetic Plain, Atmos. Environ., 2018, 175, 127–134.
26. A. Patel and N. Rastogi, Seasonal variability in chemical composition and oxidative potential of ambient aerosol over a high altitude site in western India, Sci. Total Environ., 2018, 644, 1268–1276.
27. J. V. Puthussery, A. Singh, P. Rai, D. Bhattu, V. Kumar and P. Vats, et al., Real-Time Measurements of PM2.5 Oxidative Potential Using a Dithiothreitol Assay in Delhi, India, Environ. Sci. Technol. Lett., 2020, 7(7), 504–510.
28. B. Sharma, J. Mao, S. Jia, S. K. Sharma, T. K. Mandal and S. Bau, et al., Size-distribution and driving factors of aerosol oxidative potential in rural kitchen microenvironments of northeastern India, Environ. Pollut., 2024, 343, 123246.
29. A. Singh, A. Patel, R. Satish, S. N. Tripathi and N. Rastogi, Wintertime oxidative potential of PM2.5 over a big urban city in the central Indo-Gangetic Plain, Sci. Total Environ., 2023, 905, 167155.
30. D. Bhattu, S. N. Tripathi, H. S. Bhowmik, V. Moschos, C. P. Lee and M. Rauber, et al., Local incomplete combustion emissions define the PM2.5 oxidative potential in Northern India, Nat. Commun., 2024, 15(1), 3517.
31. P. K. Verma, M. Devaprasad, J. Dave, R. Meena, H. Bhowmik and S. N. Tripathi, et al., Summertime oxidative potential of atmospheric PM2.5 over New Delhi: Effect of aerosol ageing, Sci. Total Environ., 2024, 920, 170984.
32. H. S. Bhowmik, S. N. Tripathi, J. V. Puthussery, V. Verma, J. Dave and N. Rastogi, Reactive oxygen species generation from winter water-soluble organic aerosols in Delhi's PM2.5, Atmos. Environ.: X, 2024, 22, 100262.
33. C. Li, H. Hakkim, V. Sinha, B. Sinha, M. Pardo and D. Cai, et al., Variation of PM2.5 Redox Potential and Toxicity During Monsoon in Delhi, India, ACS ES&T Air, 2024, 1(4), 316–329.
34. K. Balakrishnan, S. Dey, T. Gupta, R. S. Dhaliwal, M. Brauer and A. J. Cohen, et al., The impact of air pollution on deaths, disease burden, and life expectancy across the states of India: the Global Burden of Disease Study 2017, Lancet Planet Health, 2019, 3(1), e26–e39.
35. C. Mallik, P. S. Mahapatra, P. Kumar, S. Panda, R. Boopathy and T. Das, et al., Influence of regional emissions on SO2 concentrations over Bhubaneswar, a capital city in eastern India downwind of the Indian SO2 hotspots, Atmos. Environ., 2019, 209, 220–232.
36. S. Panda, S. Bikkina, S. K. Sharma, T. Das and B. Ramasamy, Chemical characterisation of fine aerosols in a smart city on the east coast of India: Seasonal variability and its impact on visibility impairment, J. Earth Syst. Sci., 2023, 132(1), 30.
37. A. Sinha, J. Pritam, H. Jain, S. Giri, S. Pati and J. Kshatri, Seasonal variations in respiratory morbidity in primary care and its correlation with the quality of air in urban Odisha, India, PLOS Global Public Health, 2024, 4, e0002313.
38. S. Swain, S. Pati and S. Pati, A chart review of morbidity patterns among adult patients attending primary care setting in urban Odisha, India: An International Classification of Primary Care experience, J. Fam. Med. Prim. Care, 2017, 6(2), 316–322.
39. S. K. Kompalli, S. S. Babu, T. C. Ajith, K. K. Moorthy, S. K. Satheesh and R. Boopathy, et al., Aging of biomass burning emissions in the Indo-Gangetic Plain outflow: Implications for black carbon light-absorption enhancement, Atmos. Res., 2023, 294, 106949.
40. P. S. Mahapatra, U. Panda, C. Mallik, R. Boopathy, S. Jain and S. K. Sharma, et al., Chemical, microstructural, and biological characterization of wintertime PM2.5 during a land campaign study in a coastal city of eastern India, Atmos. Pollut. Res., 2021, 12(9), 101164.
41. K. Cheung, M. M. Shafer, J. J. Schauer and C. Sioutas, Diurnal Trends in Oxidative Potential of Coarse Particulate Matter in the Los Angeles Basin and Their Relation to Sources and Chemical Composition, Environ. Sci. Technol., 2012, 46(7), 3779–3787.
42. N. J. Farren, R. E. Dunmore, M. I. Mead, M. S. Mohd Nadzir, A. A. Samah and S. M. Phang, et al., Chemical characterisation of water-soluble ions in atmospheric particulate matter on the east coast of Peninsular Malaysia, Atmos. Chem. Phys., 2019, 19(3), 1537–1553.
43. R. C. Thakur and M. Thamban, Influence of gaseous and particulate species on neutralization processes of polar aerosol and snow — A case study from Ny-Ålesund, J. Environ. Sci., 2019, 76, 12–25.
44. B. J. Turpin and H. J. Lim, Species Contributions to PM2.5 Mass Concentrations: Revisiting Common Assumptions for Estimating Organic Mass, Aerosol Sci. Technol., 2001, 35(1), 602–610.
45. S. K. Pani, C. T. Lee, C. C. K. Chou, K. Shimada, S. Hatakeyama and A. Takami, et al., Chemical Characterization of Wintertime Aerosols over Islands and Mountains in East Asia: Impacts of the Continental Asian Outflow, Aerosol Air Qual. Res., 2017, 17(12), 3006–3036.
46. A. Patel and N. Rastogi, Chemical Composition and Oxidative Potential of Atmospheric PM10 over the Arabian Sea, ACS Earth Space Chem., 2020, 4(1), 112–121.
47. J. Charrier-Klobas and C. Anastasio, On dithiothreitol (DTT) as a measure of oxidative potential for ambient particles: Evidence for the importance of soluble transition metals, Atmos. Chem. Phys. Discuss., 2012, 12, 11317–11350.
48. M. C. Pietrogrande, M. R. Perrone, F. Manarini, S. Romano, R. Udisti and S. Becagli, PM10 oxidative potential at a Central Mediterranean Site: Association with chemical composition and meteorological parameters, Atmos. Environ., 2018, 188, 97–111.
49. M. Ahmad, Q. Yu, J. Chen, S. Cheng, W. Qin and Y. Zhang, Chemical characteristics, oxidative potential, and sources of PM2.5 in wintertime in Lahore and Peshawar, Pakistan, J. Environ. Sci., 2021, 102, 148–158.
50. K. Chen, J. Xu, L. Famiyeh, Y. Sun, D. Ji and H. Xu, et al., Chemical constituents, driving factors, and source apportionment of oxidative potential of ambient fine particulate matter in a Port City in East China, J. Hazard. Mater., 2022, 440, 129864.
51. W. Liu, Y. Xu, W. Liu, Q. Liu, S. Yu and Y. Liu, et al., Oxidative potential of ambient PM2.5 in the coastal cities of the Bohai Sea, northern China: Seasonal variation and source apportionment, Environ. Pollut., 2018, 236, 514–528.
52. D. Paraskevopoulou, A. Bougiatioti, I. Stavroulas, T. Fang, M. Lianou and E. Liakakou, et al., Yearlong variability of oxidative potential of particulate matter in an urban Mediterranean environment, Atmos. Environ., 2019, 206, 183–196.
53. A. Patel and N. Rastogi, Oxidative Potential of Atmospheric Aerosols over Different Regions of India and Surrounding Oceans, ACS Earth Space Chem., 2023, 7(12), 2582–2592.
54. V. Verma, T. Fang, L. Xu, R. E. Peltier, A. G. Russell and N. L. Ng, et al., Organic Aerosols Associated with the Generation of Reactive Oxygen Species (ROS) by Water-Soluble PM2.5, Environ. Sci. Technol., 2015, 49(7), 4646–4656.
55. V. Verma, T. Fang, H. Guo, L. King, J. T. Bates and R. E. Peltier, et al., Reactive oxygen species associated with water-soluble PM$_{2.5}$ in the southeastern United States: spatiotemporal trends and source apportionment, Atmos. Chem. Phys., 2014, 14(23), 12915–12930.
56. M. Aničić Urošević, M. Kuzmanoski, T. Milićević, I. Kodranov, K. Vergel and A. Popović, Moss bag sensitivity for the assessment of airborne elements at suburban background site during spring/summer season characterized by Saharan dust intrusions, Air Qual., Atmos. Health, 2022, 15(8), 1357–1377.
57. S. Jain, S. K. Sharma, N. Vijayan and T. K. Mandal, Seasonal characteristics of aerosols (PM2.5 and PM10) and their source apportionment using PMF: A four year study over Delhi, India, Environ. Pollut., 2020, 262, 114337.
58. S. K. Sharma, T. K. Mandal, A. Sharma and S. Jain, Saraswati. Carbonaceous Species of PM2.5 in Megacity Delhi, India During 2012–2016, Bull. Environ. Contam. Toxicol., 2018, 100(5), 695–701.
59. M. Park, H. S. Joo, K. Lee, M. Jang, S. D. Kim and I. Kim, et al., Differential toxicities of fine particulate matters from various sources, Sci. Rep., 2018, 8(1), 17007.
60. G. Norris, R. Duvall, S. Brown and S. Bai, EPA Positive Matrix Factorization (PMF) 5.0 Fundamentals and User Guide, US Environmental Protection Agency National Expo Res Lab Res, Triangle Park: Petaluma, 2014, pp. 1–136.
61. P. Paatero, Least squares formulation of robust non-negative factor analysis, Chemom. Intell. Lab. Syst., 1997, 37(1), 23–35.
62. P. Paatero and U. Tapper, Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values, Environmetrics, 1994, 5(2), 111–126.
63. M. Manousakas, H. Papaefthymiou, E. Diapouli, A. Migliori, A. G. Karydas and I. Bogdanovic-Radovic, et al., Assessment of PM2.5 sources and their corresponding level of uncertainty in a coastal urban area using EPA PMF 5.0 enhanced diagnostics, Sci. Total Environ., 2017, 574, 155–164.
64. S. G. Brown, S. Eberly, P. Paatero and G. A. Norris, Methods for estimating uncertainty in PMF solutions: Examples with ambient air and water quality data and guidance on reporting PMF results, Sci. Total Environ., 2015, 518–519, 626–635.
65. P. Paatero, S. Eberly, S. G. Brown and G. A. Norris, Methods for estimating uncertainty in factor analytic solutions, Atmos. Meas. Tech., 2014, 7(3), 781–797.
66. S. Weber, G. Uzu, O. Favez, L. J. S. Borlaza, A. Calas and D. Salameh, et al., Source apportionment of atmospheric PM_10 oxidative potential: synthesis of 15 year-round urban datasets in France, Atmos. Chem. Phys., 2021, 21(14), 11353–11378.
67. S. Weber, G. Uzu, A. Calas, F. Chevrier, J. L. Besombes and A. Charron, et al., An apportionment method for the oxidative potential of atmospheric particulate matter sources: application to a one-year study in Chamonix, France, Atmos. Chem. Phys., 2018, 18(13), 9617–9629.
68. P. A. Dominutti, L. J. S. Borlaza, J. J. Sauvain, V. D. Ngoc Thuy, S. Houdier and G. Suarez, et al., Source apportionment of oxidative potential depends on the choice of the assay: insights into 5 protocols comparison and implications for mitigation measures, Environ. Sci.: Atmos., 2023, 3(10), 1497–1512.
69. E. Serafeim, A. Besis, A. Kouras, C. N. Farias, A. B. Yera and G. M. Pereira, et al., Oxidative potential of ambient PM2.5 from São Paulo, Brazil: Variations, associations with chemical components and source apportionment, Atmos. Environ., 2023, 298, 119593.
70. Y. Wang, M. Wang, S. Li, H. Sun, Z. Mu and L. Zhang, et al., Study on the oxidation potential of the water-soluble components of ambient PM2.5 over Xi’an, China: Pollution levels, source apportionment and transport pathways, Environ. Int., 2020, 136, 105515.
71. I. Uria-Tellaetxe and D. C. Carslaw, Conditional bivariate probability function for source identification, Environ. Model. Software, 2014, 59, 1–9.
72. Q. Dai, J. Ding, C. Song, B. Liu, X. Bi and J. Wu, et al., Changes in source contributions to particle number concentrations after the COVID-19 outbreak: Insights from a dispersion normalized PMF, Sci. Total Environ., 2021, 759, 143548.
73. P. Rai, A. Chakraborty, A. K. Mandariya and T. Gupta, Composition and source apportionment of PM1 at urban site Kanpur in India using PMF coupled with CBPF, Atmos. Res., 2016, 178–179, 506–520.
74. N. Rastogi, A. Singh, M. M. Sarin and D. Singh, Temporal variability of primary and secondary aerosols over northern India: Impact of biomass burning emissions, Atmos. Environ., 2016, 125, 396–403.
75. N. Rastogi, A. Singh, D. Singh and M. M. Sarin, Chemical characteristics of PM2.5 at a source region of biomass burning emissions: Evidence for secondary aerosol formation, Environ. Pollut., 2014, 184, 563–569.
76. C. M. Pavuluri, K. Kawamura, S. G. Aggarwal and T. Swaminathan, Characteristics, seasonality and sources of carbonaceous and ionic components in the tropical aerosols from Indian region, Atmos. Chem. Phys., 2011, 11(15), 8215–8230.
77. Y. Chen, O. Wild, L. Conibear, L. Ran, J. He and L. Wang, et al., Local characteristics of and exposure to fine particulate matter (PM2.5) in four indian megacities, Atmos. Environ.: X, 2020, 5, 100052.
78. V. Singh, S. Singh and A. Biswal, Exceedances and trends of particulate matter (PM2.5) in five Indian megacities, Sci. Total Environ., 2021, 750, 141461.
79. A. R. Aswini, P. Hegde, P. R. Nair and S. Aryasree, Seasonal changes in carbonaceous aerosols over a tropical coastal location in response to meteorological processes, Sci. Total Environ., 2019, 656, 1261–1279.
80. A. Ianniello, F. Spataro, G. Esposito, I. Allegrini, M. Hu and T. Zhu, Chemical characteristics of inorganic ammonium salts in PM_2.5 in the atmosphere of Beijing (China), Atmos. Chem. Phys., 2011, 11(21), 10803–10822.
81. A. Saffari, N. Daher, M. M. Shafer, J. J. Schauer and C. Sioutas, Global Perspective on the Oxidative Potential of Airborne Particulate Matter: A Synthesis of Research Findings, Environ. Sci. Technol., 2014, 48(13), 7576–7583.
82. Y. Yang, M. A. Battaglia, M. K. Mohan, E. S. Robinson, P. F. DeCarlo and K. C. Edwards, et al., Assessing the Oxidative Potential of Outdoor PM2.5 in Wintertime Fairbanks, Alaska, ACS ES&T Air, 2024, 1(3), 175–187.
83. V. J. Farahani, A. Altuwayjiri, M. Pirhadi, V. Verma, A. A. Ruprecht and E. Diapouli, et al., The oxidative potential of particulate matter (PM) in different regions around the world and its relation to air pollution sources, Environ. Sci.: Atmos., 2022, 2(5), 1076–1086.
84. L. Moufarrej, D. Courcot and F. Ledoux, Assessment of the PM2.5 oxidative potential in a coastal industrial city in Northern France: Relationships with chemical composition, local emissions and long range sources, Sci. Total Environ., 2020, 748, 141448.
85. S. Yu, W. Liu, Y. Xu, K. Yi, M. Zhou and S. Tao, et al., Characteristics and oxidative potential of atmospheric PM2.5 in Beijing: Source apportionment and seasonal variation, Sci. Total Environ., 2019, 650, 277–287.
86. A. H. Al Hanai, D. S. Antkiewicz, J. D. C. Hemming, M. M. Shafer, A. M. Lai and M. Arhami, et al., Seasonal variations in the oxidative stress and inflammatory potential of PM2.5 in Tehran using an alveolar macrophage model; The role of chemical composition and sources, Environ. Int., 2019, 123, 417–427.
87. D. Chirizzi, D. Cesari, M. R. Guascito, A. Dinoi, L. Giotta and A. Donateo, et al., Influence of Saharan dust outbreaks and carbon content on oxidative potential of water-soluble fractions of PM2.5 and PM10, Atmos. Environ., 2017, 163, 1–8.
88. M. Hakimzadeh, E. Soleimanian, A. Mousavi, A. Borgini, C. De Marco and A. A. Ruprecht, et al., The impact of biomass burning on the oxidative potential of PM2.5 in the metropolitan area of Milan, Atmos. Environ., 2020, 224, 117328.
89. C. Nishita-Hara, M. Hirabayashi, K. Hara, A. Yamazaki and M. Hayashi, Dithiothreitol-Measured Oxidative Potential of Size-Segregated Particulate Matter in Fukuoka, Japan: Effects of Asian Dust Events, GeoHealth, 2019, 3(6), 160–173.
90. K. Kurihara, A. Iwata, S. G. Murray Horwitz, K. Ogane, T. Sugioka and A. Matsuki, et al., Contribution of Physical and Chemical Properties to Dithiothreitol-Measured Oxidative Potentials of Atmospheric Aerosol Particles at Urban and Rural Sites in Japan, Atmosphere, 2022, 13(2), 1–22.
91. T. Fang, H. Guo, L. Zeng, V. Verma, A. Nenes and R. J. Weber, Highly Acidic Ambient Particles, Soluble Metals, and Oxidative Potential: A Link between Sulfate and Aerosol Toxicity, Environ. Sci. Technol., 2017, 51(5), 2611–2620.
92. Y. Zou, C. Jin, Y. Su, J. Li and B. Zhu, Water soluble and insoluble components of urban PM2.5 and their cytotoxic effects on epithelial cells (A549) in vitro, Environ. Pollut., 2016, 212, 627–635.
93. J. Zhang, H. Cheng, D. Wang, Y. Zhu, C. Yang and Y. Shen, et al., Chronic Exposure to PM2.5 Nitrate, Sulfate, and Ammonium Causes Respiratory System Impairments in Mice, Environ. Sci. Technol., 2021, 55(5), 3081–3090.
94. M. T. Chuang, Y. C. Chen, C. T. Lee, C. H. Cheng, Y. J. Tsai and S. Y. Chang, et al., Apportionment of the sources of high fine particulate matter concentration events in a developing aerotropolis in Taoyuan, Taiwan, Environ. Pollut., 2016, 214, 273–281.
95. B. Liu, N. Song, Q. Dai, R. Mei, B. Sui and X. Bi, et al., Chemical composition and source apportionment of ambient PM2.5 during the non-heating period in Taian, China, Atmos. Res., 2016, 170, 23–33.
96. S. K. Sharma, T. K. Mandal, S. Jain, S. A. Saraswati and M. Saxena, Source Apportionment of PM2.5 in Delhi, India Using PMF Model, Bull. Environ. Contam. Toxicol., 2016, 97(2), 286–293.
97. F. Amato, M. Pandolfi, T. Moreno, M. Furger, J. Pey and A. Alastuey, et al., Sources and variability of inhalable road dust particles in three European cities, Atmos. Environ., 2011, 45(37), 6777–6787.
98. R. M. Harrison, A. M. Jones, J. Gietl, J. Yin and D. C. Green, Estimation of the Contributions of Brake Dust, Tire Wear, and Resuspension to Nonexhaust Traffic Particles Derived from Atmospheric Measurements, Environ. Sci. Technol., 2012, 46(12), 6523–6529.
99. P. Pant and R. M. Harrison, Estimation of the contribution of road traffic emissions to particulate matter concentrations from field measurements: A review, Atmos. Environ., 2013, 77, 78–97.
100. R. E. Peltier, K. R. Cromar, Y. Ma, Z. H. Fan (Tina) and M. Lippmann, Spatial and seasonal distribution of aerosol chemical components in New York City: (2) Road dust and other tracers of traffic-generated air pollution, J. Exposure Sci. Environ. Epidemiol., 2011, 21(5), 484–494.
101. J. C. Fussell, M. Franklin, D. C. Green, M. Gustafsson, R. M. Harrison and W. Hicks, et al., A Review of Road Traffic-Derived Non-Exhaust Particles: Emissions, Physicochemical Characteristics, Health Risks, and Mitigation Measures, Environ. Sci. Technol., 2022, 56(11), 6813–6835.
102. R. Tiwari and A. Raman, Ease of Moving Index, 2018, 2018;(April).
103. S. K. Guttikunda and P. Jawahar, Atmospheric emissions and pollution from the coal-fired thermal power plants in India, Atmos. Environ., 2014, 92, 449–460.
104. U. Panda, R. Boopathy, H. S. Gadhavi, K. Renuka, S. S. Gunthe and T. Das, Metals in coarse ambient aerosol as markers for source apportionment and their health risk assessment over an eastern coastal urban atmosphere in India, Environ. Monit. Assess., 2021, 193(5), 311.
105. S. K. Sahu, P. Sahoo, P. Mangaraj, G. Beig, B. Tyagi and B. Samal, et al., Identification and Quantification of Emission Hotspots of Air Pollutants over Bhubaneswar: A Smart City in Eastern India, Aerosol Air Qual. Res., 2023, 23(9), 230049.
106. B. Liu, J. Wu, J. Zhang, L. Wang, J. Yang and D. Liang, et al., Characterization and source apportionment of PM2.5 based on error estimation from EPA PMF 5.0 model at a medium city in China, Environ. Pollut., 2017, 222, 10–22.
107. P. S. Mahapatra, P. R. Sinha, R. Boopathy, T. Das, S. Mohanty and S. C. Sahu, et al., Seasonal progression of atmospheric particulate matter over an urban coastal region in peninsular India: Role of local meteorology and long-range transport, Atmos. Res., 2018, 199, 145–158.
108. R. R. Behera, D. R. Satapathy, A. Majhi and C. R. Panda, Spatiotemporal variation of atmospheric pollution and its plausible sources in an industrial populated city, Bay of Bengal, Paradip, India, Urban Clim., 2021, 37, 100860.
109. K. Ram, M. M. Sarin and S. Tripathi, A 1 year record of carbonaceous aerosols from an urban site in the Indo-Gangetic Plain: Characterization, sources, and temporal variability, J. Geophys. Res., 2010, 115(D24313), 1–14.
110. N. R. Rout, Slum Growth in Bhubaneswar: A Problem or Solution?, ITPI J., 2008, 5, 59–64.
111. N. R. Rout, Urban Sprawling, Land use change and Air quality in Bhubaneswar, International journal of basic and applied research, 2018, 8, 385–394.
112. MdR. Islam, J. Welker, A. Salam and E. A. Stone, Plastic Burning Impacts on Atmospheric Fine Particulate Matter at Urban and Rural Sites in the USA and Bangladesh, ACS Environ. Au, 2022, 2(5), 409–417.
113. G. Zhang, C. Ding, X. Jiang, G. Pan, X. Wei and Y. Sun, Chemical Compositions and Sources Contribution of Atmospheric Particles at a Typical Steel Industrial Urban Site, Sci. Rep., 2020, 10(1), 7654.
114. D. C. Blanchard and A. H. Woodcock, The Production, Concentration, And Vertical Distribution Of The Sea-Salt Aerosol, Ann. N. Y. Acad. Sci., 1980, 338(1), 330–347.
115. OLA Mobility Institute, Ease of moving Index, 2022, https://olawebcdn.com/ola-institute/easeofmoving-2022.pdf.
116. S. Samal, M. Mohanty and M. Selvaraj, Assessment of Traffic Congestion under Indian Environment - a Case Study, Communications - Scientific letters of the University of Zilina, 2022, 4, D174–D182.
117. J. G. Charrier, N. K. Richards-Henderson, K. J. Bein, A. S. McFall, A. S. Wexler and C. Anastasio, Oxidant production from source-oriented particulate matter – Part 1: Oxidative potential using the dithiothreitol (DTT) assay, Atmos. Chem. Phys., 2015, 15(5), 2327–2340.
118. F. Shirmohammadi, S. Hasheminassab, D. Wang, A. Saffari, J. J. Schauer and M. M. Shafer, et al., Oxidative potential of coarse particulate matter (PM10–2.5) and its relation to water solubility and sources of trace elements and metals in the Los Angeles Basin, Environ. Sci.: Processes Impacts, 2015, 17(12), 2110–2121.
119. G. M. Pereira, K. Teinilä, D. Custódio, A. Gomes Santos, H. Xian and R. Hillamo, et al., Particulate pollutants in the Brazilian city of São Paulo: 1-year investigation for the chemical composition and source apportionment, Atmos. Chem. Phys., 2017, 17(19), 11943–11969.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4em00150h

---