Chemical Characterization and Source Apportionment of ...€¦ · • The “Loy Krathong” festival is an annual major Thai event that includes setting off fireworks and its anniversary

1

Chemical Characterization and Source Apportionment

of Particulate Polycyclic Aromatic Hydrocarbons

(PAHs), Carbonaceous Substances and Heavy Metals

in Ambient Air of Thailand

Siwatt Pongpiachan

Email: [email protected] NIDA Center for Research & Development of Disaster Prevention & Management

School of Social and Environmental Development, National Institute of Development Administration (NIDA,

118 Moo 3, Sereethai Road, Klong-Chan, Bangkapi, Bangkok, 10240, THAILAND

2

3

Some new diseases related to air pollution

4

5

• Source-4

• Source-3

• Source-2

• Source-1

Traffic Emissions

Industrial Emissions

Biomass Burnings

Street Food

Barbecue

6

Shimadzu GC-MS-QP2010 Ultra System with ASSP and FASST Function

Analytical Method of PAHs

7

SRM 1941b

Both precision and accuracy were tested by employing standard reference material (SRM

1941b) provided by the National Institute of Standard and Technology (NIST). The

precision of the procedure was calculated based on relative standard deviation of

duplicate samples and the latter was less than 10%. All sample concentrations were

calculated using standardized relative response factors (RRFs) run with each batch

(Pongpiachan et al., 2009, 2011).

Pongpiachan S, Bualert S, Sompongchaiyakul P, Kositanont C. Factors affecting sensitivity and

stability of polycyclic aromatic hydrocarbons. Anal Lett 2009; 42(13): 2106-30.

Pongpiachan S, Hirunyatrakul P, Kittikoon I, Khumsup C., 2011. Parameters influencing on

sensitivities of polycyclic aromatic hydrocarbons measured by Shimadzu GCMS-QP2010 Ultra.

Gas Chromatography/Book 3 (ISBN 978-953-51-0298-4), Intech Open Access Publisher.

DOI:10.5772/32234.

QA/QC

8

Analysis of PAHs

Pongpiachan, S., Hattayanone, M., Pinyakong, O., Viyakarn, V., Chavanich, S. A., Bo, C., Khumsup, C., Kittikoon, I., and Hirunyatrakul, P. (2016).

Quantitative ecological risk assessment of inhabitants exposed to polycyclic aromatic hydrocarbons in terrestrial soils of King George Island,

Antarctica. Polar Science.

Pongpiachan, S., Hattayanone, M., Choochuay, C., Mekmok, R., Wuttijak, N., and Ketratanakul, A. 2015. Enhanced PM10 bounded PAHs from

shipping emissions. Atmospheric Environment, 108, 13-19.

Pongpiachan, S., Tipmanee, D., Khumsup, C., Kittikoon, I., and Hirunyatrakul, P., 2015. Assessing risks to adults and preschool children posed by

PM2.5-bound polycyclic aromatic hydrocarbons (PAHs) during a biomass burning episode in Northern Thailand. Science of the Total Environment,

508, 435-444.

Pongpiachan, S., 2015. A Preliminary Study of Using Polycyclic Aromatic Hydrocarbons as Chemical Tracers for Traceability in Soybean

Products. Food Control, 47, 392-400.

Pongpiachan, S., Tipmanee, D., Deelaman, W., Muprasit, J., Feldens, P., and Schwarzer, K., 2013. Risk assessment of the presence of polycyclic

aromatic hydrocarbons (PAHs) in coastal areas of Thailand affected by the 2004 tsunami. Marine Pollution Bulletin, 76, 370-8.

9

10

• The “Loy Krathong” festival is an annual major Thai

event that includes setting off fireworks and its

anniversary is centred on the evening of the full

moon of the 12th month in the traditional Thai lunar

calendar.

• Since fireworks are widely considered as one of

the major source of PAHs, it appears reasonable

to expect the significant enhancement of PAHs

during the Bonfire event.

Loy Krathong Festival & PAHs

11

PDOS

Public Relation Department Observatory Site

BROS

Bansomdejchaopraya Rajabhat University

Observatory Site

RCOS

Ramkhamhaeng Conjunction Observatory Site

LDOS

Land Development Department Observatory Site

MCOS)

Maboonkrong Conjunction Observatory Site

VMOS

Victory Monument Observatory Site

PCD Air Quality Observatory Sites

Toxic Equivalency Factor (TEF)

12

13

*The majority of particulate PAHs measured in FDP

were significantly (p<0.05) higher than those of NDP,

which can be attributed to the high variability and

complexity of emissions sources in different sampling

periods.

*The atmospheric concentrations of HMW PAHs such

as B[b]F, B[k]F, B[e]P, Ind, and B[g,h,i]P were

significantly higher in the FDP, which can be explained

by the particle injections from fireworks.

14

15

16

Vehicular Exhausts

&

Diesel Emissions

Fireworks Display

17

Take Home Messages

*Significant decreases in Σ3,4-ring PAHs/Σ5,6-ring PAHs ratios observed during the

FDP highlight that HMW PAHs with the exception of B[a]P are the main compositions

present during the bonfire night episode, which is consistent with previous

investigations.

*Principal Component Analysis highlighted the importance of both traffic emissions

and firework displays as representing 61% (i.e. PC1+PC3+PC4+PC5) and 35% (i.e.

PC2) of the total variances of eigen values, respectively.

18

19

Total ambient PM10 samples from both observatory

sites were collected on 75 days from February 1–10,

2010 (n = 10), April 19–28, 2011 (n = 10), March 5–

14, 2012 (n = 10), October 27–31, 2012 (n = 5),

March 22–31, 2013 (n = 10), and June 7 to July 6,

2013 (n = 30). PM10 samples were collected

simultaneously at both sites for 24 h every day from

09:00 a.m. to 09:00 a.m. on the following day.

During the sampling period, the Tisch high-volume

air samplers were employed with a flow rate of 1.132

m3 min-1.

Sampling Sites & Sampling Period

20

Binary diagnostic ratios of PAHs measured during the docking and non-docking periods at PTOS and ICZ

Diesel Emission

21

Principal components (PC) pattern for Varimax rotated components applied to PM10-bound PAHs and MI from ICZ and PTOS during the docking and non-docking periods. Any values that higher than 0.5 will be highlighted as bold.

It is well known that diesel exhausts are major sources of LMW PAHs and thus, the strong correlation coefficients of Phe, Fluo, Pyr, and Chry observed in PC1 can be attributed to shipping emissions during the docking period.

The positive correlation coefficients for B[b+k]F, B[e]P, B[a]P, and Ind imply that coal fly ash from nearby power plants is probably the main source found in PC2, because HMW PAHs are often found in particles derived from incomplete combustion of coal.

Moderate loadings of MI-TA98 (-S9) as well as MI-TA100 (-S9) observed in PC3 for both sampling periods. This can be attributed to the generation of mutagenic compounds from industrial boilers.

22

23

Sampling Sites & Sampling Period

24

Binary diagnostic ratios of PAHs measured during the haze and non-haze periods at northern part of

Thailand

25

3D Plots of PCA

26

3D Plots of PCA

27

Take Home Messages

No statistically significant differences in PM2.5-bound PAHs were observed before and after the haze episode, highlighting the impacts of vehicular exhaust as regular sources of fine particulate PAHs in Northern Thailand.

High temperature high pressure condition is the main contributor of PM2.5-bound PAHs (i.e. engine combustion)

28

Air sampling was conducted in 24-h periods at all air quality sites simultaneously once every month from January to June 2008 constructing a database of 48 individual air samples (i.e. 6 × 8 = 48). Graseby-Anderson high-volume air samplers TE-6001 were used to achieve unmanned 24-h samplings for PM10. A total of 48 air samples were acquired using high- volume-yielding sample volumes of approximately 1632 m3 for each 24-h sample. PM10 were collected on 20×25 cm Whatman glass fibre filters (GFFs) at a flow rate of approximately 1.133 m3 min−1 (i.e. 40 cfm).

29

Seven PCD Air Quality Monitoring Stations in Bangkok

High Volume Air Sampler

Graseby Andersen Glass Fiber Filters

inductively coupled plasma mass spectrometry

(ICPMS: Agilent 7500cx, Agilent Technologies Inc.)

30

Emission Sources of Particulate Metals

31

Hierarchical Cluster Analysis (HCA)

The first cluster shows fairy strong affinities of Co, La,

Cd, Ce, Se, As, Sb, Cr, V and Ni with CO, SO2, O3 and

NO2, which are mainly produced by traffic emissions,

underlining the importance of vehicular exhausts on

these ten metals.

Because Cu and Zn are highly associated with RH, Rad and WD

in the first cluster, it appears reasonable to interpret this

fact as a consequence of predominant geographical sources

over these two metals.

The second sub-cluster contains Al, Fe and P.

This result can be ascribed to the overwhelming influence of

crustal emissions.

32

Principal Component Analysis (PCA)

The first component (PC1) shows the high loading on crustal metals (i.e. Al, Fe and Mn), with no observed significant correlations in any trace gaseous species. (i.e. Crustal Emission 35%)

It is also important to note that Pb as well as V, Co and Zn are the most commonly used tracer

element for identifying vehicular emissions. Despite the introduction of unleaded petrol in

Thailand in 1992, lead is still used as an elementary marker because of its comparatively high

persistence in road dust particles. Therefore, PC2 can be considered representative of traffic

emissions, explaining 13.5 % of the total PM10.

PC3 indicates considerably strong positive correlations of Ba, CO and NO2 coupled with a

negative correlation of O3. Unlike those of trace gaseous CO and NO2, the correlation coefficient

of O3 was negatively correlated with the others, highlighting the mechanism of O3 formation from

NOx. Because Ba has been widely employed as an elementary marker for brake and tyre wear

emissions. it seems plausible to consider particle emissions from idling in traffic and frequent

acceleration and braking as the main contributors of this PC, which is responsible for 11.3 % of

the total variance.

33

All samples (n = 44) were collected on

three consecutive days, from 17

November 2010 to 30 April 2011, at

CHAOS.

MiniVolTM portable air samplers

(Airmetrics) were used to collect PM2.5

for 72 h at CHAOS. The MiniVol’s pump

draws in air at 5 L min-1 through a

particle size separator (impactor) and

then through a 47-mm filter.

+

34

Controling

Flow Rate

Mini Vol Sampler Sampling Point

Start Sampling

2010-2011

35

Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, As, Se, Br, Rb, Sr, Y, Zr, Nb, Mo, Pd, Ag, Cd, In, Sn, Sb, Cs, Ba, La, Ce, Sm, Eu, Tb, Hf, Ta, W, Ir, Au, Hg, Tl, Pb, and U.

Epsilon 5 ED-XRF, PAN analytical

Chemical Analysis of 51 Metals

36

37

Daily Variation of Metals in PM2.5

The average logarithmic concentration

profiles of 51 selected metals in PM2.5

collected at CHAOS from Monday to

Sunday are, to some extent, similar to one

another.

No significant differences on percentage

contribution of metals in PM2.5 collected

at CHAOS from Monday to Sunday.

Tungsten is the third highest metals detected

in PM2.5 for both weekday and weekend.

38

Harsh braking and rapid acceleration is the main source of W in PM2.5.

Tungsten coated disc brake

39

Enrichment Factor (EF)

Fe concentration in PM2.5

Fe concentration in crust

Metal concentration

in PM2.5

Metal concentration

in crust

Rudnick, R. L., & Gao, S. (2003). Composition of the continental crust. Treatise on

geochemistry, 3, 659.

40

Enrichment Factor (EF)

It also is interesting to note that the exceptionally low

Log(EF) (-0.52) of Al detected in this study is in good

agreement with values reported by Pongpiachan and Iijima

(2015) and Wu et al. (1994). Crustal emissions are plausible

predominant sources of particulate Al over Bangkok, as

equivalent studies have reported that the majority of Al over

Chesapeake Bay was mainly derived from terrestrial soils

(Wu et al. 1994).

Conversely, the exceedingly great values of Log(EF) ([4)

observed in W, In, Tb, Eu, Ir, Cd, Cs, Se, Hg, Sb, and Pd

highlight the strong impact of traffic exhaust, consistent with

early findings (Lough et al. 2005; Almeida et al. 2006;

Crawford et al. 2007; Pongpiachan and Iijima 2015).

Only 11% of Log(EF) was lower than one. This indicates the comparatively strong influence of anthropogenic activity, surpassing other factors, such as natural emissions.

41

All samples (n = 94) were collected in three

consecutive day intervals from 17th November

2010 to 19th January 2012 at CHAOS. Prior to

the PM2.5 sample collection, the QM/A were

baked at 800 °C for at least three hours to

remove any organic contaminants and were

wrapped individually in DCM pre-cleaned

aluminium foil until loaded into the filter holder

cassette.

Chow, J.C., 2003. Introduction to special topic: weekend and weekday differences in ozone levels. J. Air Waste Manage. Assoc. 53 (7), 771. Chow, J.C., Watson, J.G., Crow, D., Lowenthal, D.H., Merrifield, T.M., 2001. Comparison of IMPROVE and NIOSH carbon measurements. Aerosp. Sci. Technol. 34 (1), 23–34.

Analytical Methods

42

Linear Regression Analysis of OC vs. EC

R = 0.95

R = 0.87 R = 0.78

R = 0.86 R = 0.90

R = 0.85

R = 0.76

R = 0.93

43

Concentration of Carbonaceous Aerosols and Vehicle Numbers

The first cluster consisted of TC, OC, EC, buses, Tuktuks, and

trucks. Because the majority of buses and trucks are powered by

diesel engines, this cluster clearly indicates that heavy-duty

vehicles (HDVs) appear to have been responsible for the

increase of ambient carbonaceous compositions. These findings

are consistent with previous studies of measurements of partic-

ulate matter from on-road vehicles and inside a tunnel, which

highlighted that diesel engines had higher emission rates than

did gasoline and LPG engines for most carbonaceous fractions

(Cheng et al., 2010; He et al., 2006).

The second sub-cluster was composed of pick-ups/vans, which

is indicative of a mixing of the three types of fuel, namely

“diesel”, “gasohol”, and “benzene”. Gasohol is a mixture of

gasoline and ethanol, which is an alternative fuel to 100%

gasoline and helps lessen the consumption of gasoline, and has

been on the market since 2001 in Thailand.

44

Principal Component Analysis of Concentration of Carbonaceous Aerosols and

Vehicle Numbers

While the dendrogram shows the close proximities of

carbonaceous composition groups with buses, Tuktuks, and

trucks, the PCA 3D plots show the strong associations between

carbonaceous aerosols and pick-ups/vans. This discrepancy

may merely reflect a difference in the statistical analogy between

PCA and HCA. While PCA is a multivariate statistical technique

used for reducing a set of elements by selecting the attributes

with the most variation, HCA is an unsupervised learning method

to find groups of similarities based on attribute values.

Despite the differences in these two statistical tools, the strong

influence of diesel vehicles (e.g., buses, pick-ups/vans, trucks)

on PM2.5-bounded carbonaceous particles is undoubtedly

obvious. Because the majority of pick-ups/vans consume diesel

fuels, it appears reasonable to ascribe the strong affinity between

carbonaceous compositions and pick-up/vans as a consequence

of the diesel engine combustion process. This interpretation is

also supported by the 3D proximity of cars and trucks with the

carbonaceous aerosols.

45

Selected Publications in 2019

Pongpiachan, S., Tipmanee, D., Choochuay, C., Hattayanone, M., Deelaman, W., Iadtem, N., Bunsomboonsakul, S., Palakun, J., Poshyachinda, S., Leckngam, A., Somboonpon, P., Panyaphirawat, T., Aukkaravittayapun, S., Wang, Q., Xing, L., Li, G., Han, Y., and Cao, J., 2019. Vertical profile of organic and elemental carbon in sediments of Songkhla Lake, Thailand. Limnology (In Press) (https://doi.org/10.1007/s10201-018-0568-9). Long, D., Hashmi, M.Z., Su, X. and Pongpiachan, S., 2019. Cr (VI) reduction by an extracellular polymeric substance (EPS) produced from a strain of Pseudochrobactrum saccharolyticum. 3 Biotech, 9(3), p.111. Pongpiachan, S., Deelaman, W., Choochuay, C., Iadtem, N., Surapipith, V., Hashmi, M. Z., ... & Promdee, K. (2019). Data relating to spatial distribution of polycyclic aromatic hydrocarbons in terrestrial soils of Pakistan and King George Island, Antarctica. Data in brief, 25, 104327. Wang, Q., Han, Y., Ye, J., Liu, S., Pongpiachan, S., Zhang, N., ... & Zhang, Q. (2019). High contribution of secondary brown carbon to aerosol light absorption in the southeastern margin of Tibetan Plateau. Geophysical Research Letters, 46(9), 4962-4970. Pongpiachan, S. (2019). Variables that influence stakeholder satisfaction with the creation of corporate images of Thailand’s National Housing Authority. Journal of Human Behavior in the Social Environment, 29(3), 346-371. Tian, J., Wang, Q., Ni, H., Wang, M., Zhou, Y., Han, Y., ... & Zhang, Q. (2019). Emission characteristics of primary brown carbon absorption from biomass and coal burning: Development of an optical emission inventory for China. Journal of Geophysical Research: Atmospheres, 124(3), 1879-1893. Pongpiachan, S., Wang, Q., Xing, L., Li, G., Han, Y., & Cao, J. (2019). Data relating to carbonaceous components in Songkhla Lake sediments, Thailand. Data in brief, 22, 1012-1017. Pongpiachan, S., Wiriwutikorn, T., Phetsomphou, P., Jieam, K., Vongxay, K., Choviran, K., ... & Centeno, C. (2019). Data relating to emissions of polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) from industrial boilers. Data in brief, 22, 286-295. Pongpiachan, S., Wiriwutikorn, T., Sbrilli, A., Gobbi, M., Hashmi, M. Z., & Centeno, C. (2019). Influence of Fuel Type on Emission Profiles of Polychlorinated Dibenzo-p-Dioxins and Polychlorinated Dibenzofurans from Industrial Boilers. Polycyclic Aromatic Compounds, 1-13. Janta, R., Sekiguchi, K., Yamaguchi, R., Sopajaree, K., Pongpiachan, S., Chetiyanukornkul, T., Ambient PM2.5, Polycyclic Aromatic Hydrocarbons and Biomass Burning Tracer in Mae Sot District, Western Thailand , Atmospheric Pollution Research, https://doi.org/10.1016/ j.apr.2019.09.003.

46

ACKNOWLEDGEMENT

Financial Support: National Institute of Development Administration, Thailand

Sample Collection: Pollution Control Department, Ministry of Natural Resources & Environment

Thank You for Your Attention