Home ->Research projects ->Research highlights->Diurnal ROS measurements

Diurnal Trends of Particulate Matter induced Oxidative Potential in different Urban Environments


Adverse health effects have been associated with exposure to Particulate Matter (PM), which is one of the important components of environmental pollution. The PM in association with volatile organic compounds, transition metals and etcetera poses severe pulmonary and cardiovascular effects in the living beings, whose complexity increase with source and spatial type with reference to the duration of exposure. This study was designed to determine the diurnal variations in oxidative potential at two spatially different habitat locations in Hong Kong.

In this study, air quality monitoring was conducted at two sampling sites one each representing urban background and roadside, both located with in 1.5 km range. The details of these locations are as shown in Figure 1.

Figure 1: Sampling locations of the Diurnal study
(Background = 31, To Yuen Building; Roadside = 292, Lai Chi Kok Road)

The reactive oxygen species (ROS), a main component of oxidative potential, was measured online using Particle-into-Liquid-Sampler (PILS). The PILS was indigenously designed in-lab at Queensland University of Technology (QUT), Brisbane, Australia by Dr. Svetlana Stevanovic. A diagram of the instrument is as shown in Figure 2. The PILS uses a condensational growth chamber to mix the incoming PM2.5 aerosol (at 5 lpm) with a stream of water vapour sufficient to create a supersaturation condition, which causes the water vapour to condense onto the particles inside the aerosol, exponentially growing them into a bigger size. Such condensed aerosol later enters the liquid vortex cyclone in the cone. A stream of captured liquid is continuously inserted in the top of the cone and pumped out at the base. At this cone, a novel profluorescent probe BPEAnit is introduced constantly, which reacts with condensed aerosol particles in the liquid vortex cyclone and captures ROS. The reacted liquid was then collected at 0.3 ml.min-1 flow rate, which later was spectrofluorometrically, equipped with USB4000 and Spectra Suite software, analyzed at Emission wavelength 486.24 nm by exciting the liquid sample at 430 nm wave length. The ROS was measured using every 2 hours over a period of 24 hours a day. It is repeated for 6 days, spanning two weeks at each sampling location, thus comprising 4 weekdays and 2 weekends. The sampling periods are 7 Jan 2016 – 25 Jan 2016 at urban background site, and from 22 Feb 2016 – 4 Mar 2016 at roadside site to cover winter session of this analysis, whose results were presented in detail here. A similar approach was followed for the summer session during Sep – Oct 2016 using a technically improved PILS named as Particle-Into-Nitroxide-Quencher (PINQ).

Figure 2: PILS and PINQ instrument design

The particle concentration (determined through SMPS with DMA 3081) at both sites, had a high degree of variation over 24 h sampling time. The peak hours can be distinguished intermittently well at the background site (8:00 – 10:30; 11:00-12:30; 13:30-15:30; 18:30-22:00 h) compared to roadside (8:00-14:00; 15:00-22:00 h) (Figure 3a & 3b). The particle concentration can be noted 1-2 fold lower at both sites, during 22:00 to 6:00 h sampling time, depicting lower vehicle traffic flow at roadside site, and non-segregation of particles at background site over 24 h. The sub-micron particle size distribution (Figure 3c & 3d) too varied between the two sites, with background site noting greater variation in the size range 17.5 nm to 40 nm at different intervals. Whereas, roadside site recorded variations in the size range 30 nm to 70 nm.

Figure 3: The diurnal trends of particle number concentration and its size distribution
(Particle number concentration is shown in a & b; while its size distribution in c & d)

Figure 4 discuss the variations in diurnal trends of ROS at two sampling sites on volumetric basis and per mass basis. On volumetric analysis, as in Figure 4a, the ROS measurements results in multiple peaks at 8:00, 16:00 and 20:00 h time periods, at background site, compared to roadside’s non-intermittent decrease/increase trends suggestive of regular traffic flow at roadside site. However, on per mass basis, the ROS quantities remained uniform for almost 16 h in a day (during 8:00 to 0:00 h) at roadside (Figure 4c). Also, the background’s ROS measurements were 0.5-1 fold lower (n=6 for each time interval, p<0.05) than roadside at each time interval. These ROS variations are well in agreement with diurnal trends of PM concentration at both sites as shown in Figures 3 and 4 depicting intuitive surrogate of BPEAnit based ROS quantification. Interestingly, though PM mass was less (by ~1 µg/m3) at roadside site compared to background, ROS generation was noted significantly higher (0.5-1 fold; n=6 p<0.05) at the former site. While weekdays at background site, generated more ROS (5-10%) during 6:00 to 18:00 h, weekends resulted in high during midnight to 4:00 AM as shown in Figure 4b & 4d. But at roadside, the weekend prone ROS quantities dominated weekdays’ for most of the time intervals, especially during 14:00 h to 6:00 h, with a strong signal at 20:00 on volumetric basis, but limited to 8:00 to 20:00 h per mass basis suggestive of varied particle coatings that influenced ROS detection.

Figure 4: Diurnal trends of ROS at two urban sites.
(a & b represent volumetric based ROS analysis, while c & d were mass based)

The co-pollutants such as BC and p-PAHS were also measured in line with ROS analysis employing AE33/AE51 (Magee) and PAS2000 (EcoChem) respectively to assess possible surrogates for ROS. The concentrations of BC and p-PAHS at roadside are 5-10 fold and 10-20 fold higher, respectively as shown in Figure 5. Both their concentrations peaked during 8:00 to 18:00 h at background site compared to roadside’s 8:00 to 22:00 h which are well in agreement with ROS quantification.

Figure 5: Diurnal trends of PM mass, BC and p-PAHs concentration

The p-PAHs/BC ratio, at background site, decreased with time from 8:00 (ratio of 0.0076) to 12:00 h (ratio of 0.0050) and then increased till 16:00 h (ratio of 0.0059) and thereafter decreased overnight (ratio of 0.0018). But roadside site had a higher p-PAHs/BC ratio of 0.01, more due to their concentration, during 8:00 to 22:00 h, but later reduced to 0.003 for rest of the time period, suggesting stronger PAS signal is due to freshly coated PAHs during peak hours, while weak signal can be attributed to secondary inorganic coating that shields the particle bound PAHs. These results depict varied spatial diurnal trends with peak hours of 8:00 h (mornings) and 20:00 h (late evening) in common, while sublime nights with lower ROS generation is attributed to background site.

Project contact: GALI Nirmal Kumar