Department of Environmental Engineering and Science

Chia-Nan University of Pharmacy and Science

Thesis for the Degree of Master

Characterization of carboxylic acids and anhydrosugars in dry season ambient aerosol

in Chiang Mai Basin, Thailand

(Advisor) Dr. Ying I. Tsai (Co-Advisor)Dr. Khajornsak Sopajaree (Graduate student) Miss Auranee Chotruksa

July 2010

Characterization of carboxylic acids and

anhydrosugars in dry season ambient aerosol in Chiang Mai Basin, Thailand

Dr. Khajornsak Sopajaree Miss Auranee Chotruksa

Department of Environmental Engineering and Science Chia-Nan University of Pharmacy and Science

Thesis for the Degree of Master

Characterization of carboxylic acids and anhydrosugars in dry season ambient aerosol

in Chiang Mai Basin, Thailand

Advisor : Dr. Ying I. Tsai Co-Advisor : Dr. Khajornsak Sopajaree Graduate student : Miss Auranee Chotruksa

July 2010

ABSTRACT

PM10 aerosol was collected during two periods between February and April of dry season 2010 at urban, suburban and mountain sites in Chiang Mai basin, Thailand. Characteristics and provenance of water-soluble inorganic species, carboxylic acids, anhydrosugars and sugar alcohols in PM10 were investigated. Concentrations of inorganic and organic species in PM10 aerosol at urban site are always higher than at suburban and mountain sites, indicating that more sources were transported to urban area. Acetic acid was the most abundant monocarboxylic acids, followed by formic acid. Oxalic acid was the dominant dicarboxylic acid species during both periods. Concentration of carboxylic acids during the PM10 episode was higher than that during non-episodic pollution. Carboxylic acids with a peak at daytime during the PM10 episode indicate that carboxylic acids are formed by photochemical reaction and/or are emitted directly by fossil fuels and biomass burning processes. Levoglucosan (Levo) and arabitol were the most dominant anhydrosugar and sugar alcohol, respectively, the ratios of levoglucosan to PM10 in forest fire are 0.53-1.48% by PM10 mass. High concentration of levoglucosan was found at nighttime in both periods, indicating that biomass burning contributed during nighttime. Mass ratio of acetic to formic acids (A/F) > 1 is often used to demonstrate the primary source by wood burning or vehicular emission. This study showed that the contribution of primary sources caused from biomass burning. Moreover, the ratios of M/S in the range of 0.94-1.72 during both periods indicated there exists simultaneously the impaction of primary traffic-related emissions and secondary photochemical pollution on Chiang Mai ambient environment. The discriminator ratios of biomass burning reported here are 0.78-2.68 of K/Levo, 5.73-36.2 of Levo/Mannosan. Levoglucosan was found to be the most useful marker for biomass burning emitted from forest fire event in the mountain around Chiang Mai basin. The most significant contribution to PM10 in Chiang Mai basin was the photochemical formation of secondary aerosols and primary source from biomass burning contributed by hardwood and softwood of leaves/bark trees.

Keywords: Chiang Mai; Biomass burning; Carboxylic acids; Oxalic acid; Levoglucosan; Sugar alcohols; A/F ratio; M/S ratio

I

2010 PM10 PM10 PM10 PM10 (LevoglucosanLevo)(arabitol)PM10 0.53-1.48%Levoglucosan PM10(A/F) 1malonic/succinic 0.94-1.72K/Levo= 0.78-2.68, Levo/Mannosan= 5.73-36.2 /

II

ACKNOWLEDGEMENTS

This thesis would never have been completed without the help and

supports of many people who are gratefully acknowledge here. I would like to

express my gratefulness for all of them.

I would like to express my gratitude to all those who gave me the

possibility to complete this thesis. I would like to give special thanks to the

Department of Environmental and Science of Chia Nan University of Pharmacy

and Science, Taiwan and the Department of Environmental Engineering of

Chiang Mai University, Thailand, for giving me permission to commence this

thesis in the first instance, and do the necessary research work. Moreover, I

would like to express my sincere appreciation to all the support and help given

from my supervisors Prof. Dr. Ying I. Tsai at Chia Nan University of Pharmacy

and Science, Taiwan and Assoc. Prof. Dr. Khajornsak Sopajaree at Chiang Mai

University, Thailand who give me helping, suggestions, guidance, warm

encouragement and generous supervision throughout my master program. I am

grateful to Assist. Prof. Dr. Li-Hao Young and Prof. Dr. Man-Ting Cheng,

members of the committees for many valuable suggestions.

I would like to special thank my sampling sites at Faculty of Architecture

Chiang Mai University, TOT Public Company Limited and Doi Suthep-Pui

National Park Protection Unit, Chiang Mai who allow me to do the necessary

research work. And, I would like to thank guards who help me protect collect

samplers and take care it.

III

IV

In addition, Thanks must go to my entire special person, whom I met in

the Atmospheric Research Laboratory at Chia Nan University of Pharmacy and

Science, Taiwan. I profoundly indebted to Pei-Ling Wu and Rui-Ling who

always help and teach me when I have problem and try to understand me. I

would like to thank my best friend, Firstly, Hsin-Ching Wu who help me a lot to

do my research work, has been a great consultant for me and take care me when

I want to do everything and, June Yu Lee and Yu Ting, who tech and suggest

me. I would also like to thank You Cong and Yu-Liang, who thanks for all kind

of help of them. I would like to thank all members in the Atmospheric Research

Laboratory: Qing-Cheng, Pi-Cheng, Yu-Ru and Yu-Wen.

Especially, I would like to give my special thanks to my parents, my

father and my mother who help me for everything for their eternally love,

support, encouragement and financial support until the completion of this study.

Auranee Chotruksa

CONTENTS

ABSTRACT........................................................................................... I

CHINESE ABSTRACT........................................................................ II

ACKNOWLEDGEMENTS ................................................................. III

CONTENTS........................................................................................... V

LIST OF TABLES .............................................................................. VIII

LIST OF FIGURES .............................................................................. X

CHAPTER 1 INTRODUCTION ......................................................... 1

1.1 Introduction............................................................................... 1

1.2 Purpose...................................................................................... 3

CHAPTER 2 LITERATURE REVIEW............................................. 4

2.1 Aerosol formation mechanism.................................................. 4

2.2 Carboxylic acids in atmospheric aerosols ................................ 5

2.3 Sources of carboxylic acids ...................................................... 8

2.3.1 Direct emissions from anthropogenic sources................... 8

2.3.1.1 Biomass combustion..................................................... 8

2.3.1.2 Motor exhaust emissions .............................................. 9

2.3.2 Emissions from biogenic sources ...................................... 10

2.3.3 Photochemical production of carboxylic acids

from precursors .................................................................. 11

V

2.4 Anhydrosugars and sugar alcohols ........................................... 19

CHAPTER 3 EXPERIMENTAL ........................................................ 26

3.1 Sampling ................................................................................... 26

3.2 Sampling handing ..................................................................... 29

3.3 Chemical analysis and quality assurance ................................. 29

3.4 Other data.................................................................................. 35

CHAPTER 4 RESULTS AND DISCUSSION ................................... 37

4.1 Meteorological conditions ........................................................ 37

4.2 Mass concentration of PM10 aerosols ....................................... 39

4.3 Aerosol composition of PM10 during non episodic pollution

and PM10 episode periods ......................................................... 42

4.4 Concentration of chemical species in daytime

and nighttime during non episodic pollution period

and PM10 episode...................................................................... 47

4.5 Contribution of chemical species ............................................. 53

4.6 Composition of mass ratios with other studies......................... 63

4.6.1 Carboxylic acids ................................................................ 63

4.6.2 Anhydrosugars................................................................... 65

4.7 Relationships among chemical species in daily PM10

and gaseous pollutants .............................................................. 67

4.8 Comparison with literature data ............................................... 69

VI

VII

CHAPTER 5 CONCLUSIONS ........................................................... 74

5.1 Conclusion ................................................................................ 74

5.2 Suggestions for the future work................................................ 77

REFERENCE ........................................................................................ 78

LIST OF TABLES

Table 2.1 Saccharides commonly found in atmospheric

aerosol and their sources (Caseiro et al., 2007).................... 25

Table 3.1 Ion Chromatography Dionex DX-600

gradient elution ratio............................................................. 31

Table 3.2 The names and chemical structures of carboxylic acids ........ 33

Table 3.3 The names and chemical structures of anhydrosugars

and sugar alcohols ................................................................ 34

Table 3.4 Method detection limits (MDLs) of four chemical

compound groups measured using IC systems..................... 35

Table 4.1 Meteorological and related air pollution information

during the period of study at the suburban site .................... 38

Table 4.2 Mean (SD) chemical composition of PM10 aerosol during non-episode pollution period and

PM10 episode emitted from sampling site ............................ 43

Table 4.3 Summary presentation of research findings

related to acetic/formic and malonic/succinic

ratios in aerosol..................................................................... 64

Table 4.4 Comparison of ratios for various wood burning

and atmosphere aerosols (reported in the literature) .............. 66

VIII

IX

Table 4.5 Varimax-rotated principal component loadings

of daily PM10 chemical species, gaseous pollutants

and wind during intensive observation period

of this study .......................................................................... 68

Table 4.6 Inorganic salt concentrations (g m-3) measured at various

sampling sites around the world in recent years .................. 71

Table 4.7 Carboxylic acids concentrations (ng m-3)

measured at various sampling sites

around the world in recent years .......................................... 72

Table 4.8 Anhydrosugars and sugar alcohols

concentrations (ng m-3) measured at various

sampling sites around the world in recent years .................. 73

LIST OF FIGURES

Figure 2.1 Idealized schematic of the distribution

of surface area of an atmospheric aerosol

(Whitby and Cantrell, 1976)............................................... 4

Figure 2.2 Production cycle of carboxylic acids

in the atmosphere (Sun and Ariya, 2006) .......................... 7

Figure 3.1 Ecotech MicroVol 1100 Particulate Samplers .................... 28

Figure 3.2 Map of Chiang Mai Basin areas

identifying the location of air sampling sites ..................... 28

Figure 3.3 Step for MicroVol sampling and analysis flow chart.......... 32

Figure 3.4 Wind rose charts during intensive

observation period (a) IOP1 and IOP2 ................................ 36

Figure 4.1 PM10 mass concentration of intensive observation

period with PCD data during period of study .................... 40

Figure 4.2 Correlation of PM10 concentration from PCD data

with PM10 concentration from observed site

during period of this study.................................................. 41

Figure 4.3 Mean of inorganic species concentration

in daytime and nighttime (a) during non episodic

pollution period and (b) during the PM10 episode

emitted from sampling sites ............................................... 48

X

Figure 4.4 Mean of carboxylic acids concentration

in daytime and nighttime (a) during non episodic

pollution period and (b) during the PM10 episode

emitted from sampling sites ............................................... 50

Figure 4.5 Mean of anhydrosugar and sugar alcohols

concentration in daytime and nighttime

(a) during non episodic pollution period and

(b) during the PM10 episode emitted from

sampling sites ..................................................................... 52

Figure 4.6 Contribution of individual species to total

composition in PM10 during intensive observation

period of each sites............................................................. 54

Figure 4.7 Contribution of individual species to total

amount of inorganic species in PM10

during intensive observation period of each sites .............. 55

Figure 4.8 Contribution of individual species to total

amount of carboxylic acids in PM10 during

intensive observation period of each sites.......................... 57

Figure 4.9 Correlation of potassium concentration with

oxalic acid concentration of each site sampling................. 58

Figure 4.10 Contribution of individual species to

total amount of anhydrosugars in PM10

during intensive observation period of each sites .............. 60

XI

XII

Figure 4.11 Correlation of levoglucosan concentration with

potassium concentration of each site sampling.................. 61

Figure 4.12 Contribution of individual species to total

amount of sugar alcohols in PM10 during

intensive observation period of each sites.......................... 62

Chapter 1 Introduction

1.1 Introduction

Atmospheric particulate matter are a complicated mixture which

are composed by inorganic substances (such as, sulfate, nitrate,

ammonium, and potassium) and organic matter, are important resulting

from the marine pathway, biomass burning, agriculture burning,

automotive exhaust emission and anthropogenic emission (Khwaja, 1995;

Chebbi and Carlier, 1996; Souza et al., 1999; Hsieh et al., 2008; Lee et al.,

2008; Zhang et al., 2008;). These emissions are impacts on regional air

quality and visibility, ecosystems and human health, and climate change

(Khwaja, 1995; Souza et al., 1999; Tsai, 2005).

Low molecular weight carboxylic acids are ubiquitous and

important components in the tropospheric aqueous and gaseous phases,

and in aerosol particles (Chebbi and Carlier, 1996). The carboxylic acids

in the particle phase, have the presence in the atmosphere may be result

from primary emission (Kawamura and Kaplan, 1987) or from secondary

photochemical reactions (Yao et al., 2004). Monocarboxylic acids were

observed with a daytime maximum and a nighttime minimum (Khawaja,

1995; Chebbi and Carlier, 1996). Formic and acetic acids constitute the

most abundant carboxylic acids in the global troposphere (Khwaja, 1995;

Souza et al., 1999). During daytime, vehicular emission appeared to be

the primary source of acetic acid, whereas formic and pyruvic acids

should be formed photochemically (Souza et al., 1999). In addition,

formic acid is one of the photochemical oxidation products from volatile

1

organic compounds (VOC), the results show that 80-100% of formic acid

stems from biogenic VOC emitted from terrestrial sources (Glasius et al.,

2000). Besides that, dicarboxylic acids are among the most abundant

organic constituents of ambient particulate matter (Ray and McDow,

2005). Dicarboxylic acids are widely present in the urban, rural and

marine atmosphere. Oxalic acid was found as the most abundant species,

followed by succinic and/or malonic (Khawaja, 1995; Chebbi and Carlier,

1996; Ho et al., 2006; Hsieh et al., 2008; Tsai et al., 2008; Hsieh et al.,

2009).

The biomarker levoglucosan (1,6-anhydro--D-glucopyranose) is

formed as a result of the thermal breakdown alteration of the cellulose,

accompanied by generally lesser amounts of straight-chain, aliphatic and

oxygenated compounds and terpenoids present in the vegetation

subjected to biomass burning. The biopolymer (cellulose) decomposes

during combustion, yielding a tarry material containing anhydrosugars

(Simoneit et al.,1999; Santos et al., 2002; Lee et al., 2008). This

compound, together with other thermal decomposition products from

cellulose and hemicelluloses (e.g. mannosan, galactosan and

levoglucosan) were utilized as tracers for biomass burning (Santos et al.,

2002; Schmidl et al., 2008; Bari et al., 2009; Caseiro et al., 2009; Fabbri

et al., 2009). It has a large impact on the biomass burning attribution as it

is emitted at high concentrations. (Simoneit et al., 1999; Jordan et al.,

2006; Zhang et al., 2008). Moreover, Jordan et al., (2006) reported that

woodsmoke was estimated to comprise about 95% of wintertime air

pollution in Launceston, and the resulting average levoglucosan

woodburning emission factor of around 140 mg g-1 particulate matter was

found to be consistent with previously determined woodheater emissions.

2

3

1.2 Purpose

In northern of Thailand, few studies describe atmospheric

measurements of particulate matter during the dry season (December to

March), levels of PM2.5 and PM10 in the Chiang Mai atmosphere are very

high, daily PM2.5 (24 h values) during the winter months in Chiang Mai

frequently exceeded 200-300 g m-3, and there may be significant health

implications associated with these high concentrations (Vinitketkumnuen

et al.,2002). In addition, have some studies and data on the water-soluble

inorganic species in atmospheric particles and wet deposition are carried

out, which no study atmospheric particulate matter (Chantara and

Chunsuk, 2008). This can be implied that a comparison of carboxylic

acids, anhydrosugars and sugar alcohols in PM10 aerosol has not been

reported in the literature.

The purpose of this study is to characterize of inorganic and

organic composition (carboxylic acids, anhydrosugars and sugar alcohols)

in aerosol during dry season at Chiang Mai Basin were investigated, with

a view to explaining differences and identifying the source of pollution in

Chiang Mai. Ultimately, this research can be contributed to a better

understanding of health effects caused from sources of aerosol.

Chapter 2 Literature review

2.1 Aerosol formation mechanism

Aerosol can either be produced by ejection into the atmosphere, or

by physical and chemical processes within the atmosphere (called

primary and secondary aerosol production respectively). Examples of

primary aerosol are sea spray and windblown dust. Secondary aerosols

are often produced by atmospheric gases reacting and condensing, or by

cooling vapor condensation (gas to particle conversion). Figure 2.1

shows some of these processes, along with the three sizes ranges (modes)

where high aerosol concentrations are often observed.

Figure 2.1 Idealised schematic of the distribution of surface area of an atmospheric aerosol (Whitby and Cantrell, 1976)

4

Tsai and Cheng (2004) observed the average mass concentration of

PM10 was 109.054.1 g m-3. Carbonaceous materials, sulfate, nitrate, and ammonium were the most important contributors to the PM10

component. Concentrations of total carbon in PM10 were significantly

high, averaging 37.9 g m-3. By contrast, concentrations of SO42-, NO3-,

and NH4+ in PM10 were lower, averaging 10.2, 6.6, and 6.0 g m-3,

respectively, 64% of PM10 was made up of fine particles. Coarse particle

mass concentrations were approximately 56% of PM2.5 mass

concentrations. The most significant contribution to PM10 in the Taichung

urban basin was from the photochemical formation of secondary aerosols

and carbonaceous materials in the atmospheric environment.

Hsieh et al. (2009) described inorganic species, especially nitrate,

were present in higher concentrations during the PM episode. A

combination of gas-to-nuclei conversion of nitrate particles and

accumulation of secondary photochemical products originating from

traffic-related emissions was likely a crucial cause of the PM episode.

Sulfate, ammonium, and oxalic acid were the dominant anion, cation, and

dicarboxylic acid, respectively, accounting for a minimum of 49% of the

total anion, cation or dicarboxylic acid mass.

2.2 Carboxylic acids in atmospheric aerosols

Monocarboxylic acids and dicarboxylic acids are the major

constituents of the organic aerosol (Limbeck et al., 2001). Low

molecular weight carboxylic acids are ubiquitous and important

components in the tropospheric aqueous and gaseous phases, and in

5

6

aerosol particles (Chebbi and Carlier, 1996). The relatively high

concentrations of dicarboxylic acids and their identification as

atmospheric reaction products from variety of different precursors make

it useful to investigate their potential as indicators of secondary organic

aerosol formation (Ray and McDow, 2005). Monocarboxylic acids were

observed with a daytime maximum and nighttime minimum. Moreover,

acetic acid was the most abundant monocarboxylic acid followed by

formic, pyruvic and glyoxalic acid, while formic and acetic acid mostly in

gaseous (Khwaja, 1995). Dicarboxylic acids were mostly associated with

particles. Oxalic acid was the dominant dicarboxylic acid species,

followed by succinic acid and malonic acid (Khawaja, 1995; Chebbi and

Carlier, 1996; Hsieh et al., 2008; Tsai et al., 2008; Hsieh et al., 2009).

Dicarboxylic acid concentrations, particularly oxalic acid, peaked at night

during the PM episode, due to accumulation of daytime oxalic acid

combined with low wind velocity and low mixing layer height at this time

(Hsieh et al., 2008).

By the Figure 2.2 describes the atmosphere organic aerosol

conversion performance (Sun and Ariya, 2006), most of the aerosol

composition of the mixed chemical species, including a variety of

inorganic and organic species, including nature, lakes, oceans and the

emissions of volatile organic compounds through the snow will change

the formation of aerosols, and aerosols in the atmosphere will combine

with inorganic or organic matter into the chemical mixture, and then

generate organic aerosols into organic cloud condensation nuclei and ice

nuclei (Ice nuclei, IN ), affect the composition of clouds.

hv

Chemical TransformationGas/Particle

Partition

Cloud

FineAerosols

CoarseAerosols

Organic aerosols

IN

CCN

Organic and inorganicMixed aerosols

Transportation

VolatileCompounds

Emission Dry Deposition Emission EmissionWet Deposition

Surface (land, ocean, snow)

Cloud Condensation Nuclei (CCN)

Ice Nuclei (IN)

Figure 2.2 Production cycle of carboxylic acids in the atmosphere (Sun and Ariya, 2006)

7

2.3 Sources of carboxylic acids

2.3.1 Direct emissions from anthropogenic sources

The observed amounts of dicarboxylic acids in the particle phase

accounted for a small fraction of the organic carbon. Results indicated

that photochemical processes and anthropogenic emissions (Yao et al.,

2004) such as automobile exhaust, animal wastes, plastic combustions,

chemical plants emissions, lacquer minifactory emissions, tinned food

plants emissions (starchy foods, fishes, ...), tobacco smoke, refuse

incineration factories are major sources of atmospheric dicarboxylic

acids. Rhrl and Lammel (2002) reported anthropogenic sources are

important for the precursors of succinic, maleic and fumaric acids,

namely toluene emissions (from vehicle exhaust, besides other), can be

considered as a significant source of maleic and fumaric acids. Wang and

Shooter (2004) suggested that solid fuel burning had large influence on

the occurrence of these low molecular weight dicarboxylic acids resulting

insignificantly higher wintertime concentrations of maleic acid. All of

these sources are of local importance and their global contribution seems

to be minor. Moreover, only the anthropogenic sources which have an

important contribution to atmospheric concentrations of carboxylic acids.

2.3.1.1 Biomass combustion

Primary emissions from wood and coal burning, biomass

combustion including wood burning stoves, forest fires, and agricultural

8

burnings. (Chebbi and Carlier, 1996) proved that direct emissions of

dicarboxylic acids from forest fires represent dicarboxylic acids,

dominated by oxalic (C2) followed by succinic (C4) and malonic (C3)

acids, also showed a concentration increase. Research has found that

forest fires can produce large amounts of DCAs (Narukawa et al., 1999).

Whether coal burning is also an important primary source of DCAs

remains uncertain. Dicarboxylic acids have several source in cluding

primary emission from burning of biomass and fossil fuel, as well as

photochemical oxidation of organic precursors (Xingru et al., 2009).

Formic and oxalic acids was estimated to contribute from biomass

burning about 30-60% (Wang et al., 2007a). Wang and Shooter (2004)

suggest that the primary emission from coal and wood burning was the

dominant source of maleic acids in an urban atmosphere. Tsai et al. (2010)

reported wood burning is the dominant source of maleic acid in

atmospheric aerosols.

2.3.1.2 Motor exhaust emissions

Primary emissions from vehicles was the major anthropogenic

source of Nonmethane Hydrocarbons (NMHCs) include mobile and

stationary source fuel usage and combustion, petroleum refining and

petrochemical manufacturing, industrial, commercial, and individual

solvent use, gas and oil production. Emissions have been of particular

concern in urban areas. In source apportionment of NMHC emissions

conducted in Los Angeles in 1976, the weight percentage of emissions

(not including industrial emissions and solvent use) was estimated to be

49% motor vehicle exhaust, 16% gasoline spillage, 13% gasoline

9

evaporation, 15% natural gas and oil fuel production, and 5% natural gas

distribution and use (Godish, 1997). Moreover, Sources of carboxylic

acids in the particulate phase. During daytime, vehicular emission

appeared to be the primary source of acetic and oxalic acid from both

source (Souza et al., 1999). Kawamura et al. (1987) detected very high

concentrations of DCAs in automobile exhausts and found that the

molecular distributions of DCAs in the Los Angeles air were similar to

those in vehicle exhausts.

2.3.2 Emissions from biogenic sources

Emissions from biogenic source which include foliar emissions

from forest trees and grasslands and emissions from soils and ocean water

are approximately an order of magnitude higher on a global basis than

anthropogenic emissions. Foliar emissions from forest trees are

comprised mainly of isoprene and monoterpenes with some paraffins and

olefins; grasslands, light paraffins and higher HCs; soils, mainly ethane;

and ocean water, light paraffins, olefins, and C9-C28 paraffins. Biogenic

sources seem to influence the occurrence of malic acid significantly

(Rhrl and Lammel, 2002). In addition, emission from biogenic primary

sources appeared to be an important contribution to atmospheric

concentration of formic and glycolic acids (Souza et al., 1999). During

the formic acid sampling period, the air masses were influenced by both

direct anthropogenic emissions (benzene, toluene, nitrogen dioxide and

acetone) and compounds formed during long-range transport of

anthropogenic hydrocarbons (formaldehyde and acetaldehyde).

10

11

Nevertheless, formic acid still had a predominantly (895%) biogenic origin (Glasius et al., 2000).

2.3.3 Photochemical production of carboxylic acids from

precursors

Water-soluble organic compounds (WSOC) have several different

sources, including primary emissions from biomass burning and fossil

fuel combustion, as well as photochemical oxidation of organic

precursors of both anthropogenic and biogenic origin (Chebbi and

Carlier, 1996). The diacids are largely produced in spring by

photochemical oxidation of hydrocarbons and other precursors that are

transported long distances from the mid- and low-latitudes to the Arctic,

but the production of oxalic acid is in part counteracted by photo-induced

degradation possibly associated with bromine chemistry (Narukawa et al.,

2002). The precise mechanisms of the production of carboxylic acids by

the ozone reactions with atmospheric olefins and, in particular the

production of dicarboxylic acids by the reactions of ozone with

cycloolefins and with aliphatic diolefins (Chebbi and Carlier, 1996).

Wang and Shooter (2004) noted that in summer, oxalic and malonic

acids, and the sum of glutaric and adipic acids has strong positive

correlations with NO3- (having dominant precursors from vehicle

emissions in summer) and temperature. It is therefore suggested that in

summer these acids may be formed mainly through photochemical

oxidation with vehicle exhausts being the dominant precursors. Phthalic

acid has been identified as photochemical product, attributed to

anthropogenic precursors (Ray and McDow, 2005). However, as the

simplest short-chain dicarboxylic acids, oxalic acid is the final product of

photochemical decomposition of other dicarboxylic acids in atmospheric

aerosol. Consequently, it is the most abundant dicarboxylic acids in

atmospheric aerosol (Hsieh et al., 2008; Tsai et al., 2010).

The mass ratio of oxalic to sulfate are present as the end products of

organic and inorganic species, the oxalic acid/sulfate mass ratio is

informative for determining the formation of dicarboxylic acids from

inorganic salts. The ratio for ambient aerosol was higher during PM

episode in range 0.504-0.603, while during non-episodic period in range

0.405-0.486 (Hsieh et al., 2008). The oxalic/sulfate ratio has been

reported of 0.80-0.83 from incense emissions (Tsai et al., 2010).

The concentration ratios of these acids in atmospheric particles, in

particular the malonic acid (C3)/succinic acid (C4) mass ratio, are useful

to understanding their importance in the atmosphere. The C3/C4 ratio has

been reported to be 0.3-0.5 from vehicular emissions (Kawamura and

Kaplan, 1987). Relatively low C3/C4 ratios have been found to be

associated with the overwhelming contributions from vehicular exhaust to

these acids in some studies, e.g., in downtown and west Los Angeles, in

winters in Tokyo, and in Nanjing, China (Kawamura and Kaplan, 1987;

Kawamura and Ikushima, 1993; Wang et al., 2002). On the other hand,

the mass ratio of C3/C4 in secondary atmospheric particles is much larger

than unity (Kawamura and Ikushima, 1993; Kawamura and Sakaguchi,

1999; Yao et al., 2002). For example, Kawamura and Ikushima (1993)

reported a maximum mass ratio of 3 in the summer in Tokyo. They found

ratios larger than unity concurrent with elevated concentrations of

12

oxidants and attributed the source of dicarboxylic acids to secondary

atmospheric reactions. Kawamura and Sakaguchi (1999) observed a mass

ratio of 3 in the Pacific Ocean, where dicarboxylic acids are expected to

originate from secondary reactions. Hence, the ratio of C3/C4 in

atmospheric particles is a useful indictor to differentiate primary

(vehicular) sources and secondary sources.

The acetic acid/formic acid (A/F) mass ratio was used to

distinguish the primary (A/F>1) and the secondary (A/F

present this study is reflective of the influence of anthropogenic source

rich in acetic acid. Formaldehyde concentrations varied from 0.63 to 3.7

ppbv which levels decrease after the mid-afternoon maxima and increase

during nighttime. Formic and acetic acids were present mainly in the size

fraction below 1.0 m diameter, the acids in particulates have gaseous

precursors. Seven carboxylic acids (formic, acetic, pyruvic, glyoxalic,

oxalic, succinic, and malonic) have been identified in airborne aerosols.

Acetic acid was the most abundant monocarboxylic acid in the particulate

phase followed by formic acid, pyruvic and glyoxalic. Dicarboxylic acids

were mostly associated with particles, oxalic acid was the most abundant

species, followed by succinic acid and malonic acid. It appears that the

photooxidation of anthropogenic compounds represents a major source of

carboxylic acids in airborne particulate.

Chebbi and Carlier (1996) show that low molecular weight

carboxylic acids are ubiquitous components in the tropospheric aqueous

phase (found in fog water, rain water, snow, ice water and in cloud

water), gas phase and aerosol particles. Formic and acetic acids, the more

abundant species in aqueous and gaseous phase, are also ubiquitous in

aerosol particles collected in various areas over the world. In addition

dicarboxylic acids are mostly present in particle phase, they found that

oxalic acid was dominant species followed by succinic, malonic, maleic,

adipic and phthalic acids. They observed diurnal variations of carboxylic

acids in the atmosphere, with higher concentration during the day than at

night. Moreover carboxylic acids found in the dry season higher than in

the wet season. Sources of carboxylic acids are comprise anthrogenic

emissions (including; wood and biomass burning, motor exhaust

emissions), biogenic emissions emitted by vegetations and soils and

14

chemical transformations of precursors production photochemical which

the precise mechanisms of production of carboxylic acid by the ozone

reactions with atmospheric olefin and, in particular the production of

dicarboxylic acids by reactions of ozone with cycloolefins and with

aliphatic diolefins. As the major source and sinks of these compounds are

well-known and their relative importance for local or regional

environments are becoming elucidated.

Souza et al. (1999) observed low molecular weight carboxylic

acids found in the atmospheric gas and particle-phase were measured

during July 1996, Winter, in an urban of So Paulo City, Brazil. Ambient

level measurements of formic, acetic, -hydroxy-acetic (glycolic), -

hydroxy-butyric, oxalic and pyruvic acids in airborne particulate and

formic and acetic acids in the gas phase are these reported.

Approximately 98% of the total acetic and formic acids were in the gas-

phase and the gas-aerosol equilibrium was influenced by high levels of

relative humidity. Gaseous formic-to-acetic ratio has been used to suggest

sources (direct emission; low ratio1). These acid ratios fell in the 0.94-

1.85 range (avg. 1.24) showed that direct emission from vehicles also

contributed to their presence in air. Gaseous formic and acetic were

strongly correlation (r=0.93). Thus, photochemical activity to carboxylic

acid production appeared to be a very likely source of the gaseous formic

and acetic acid level. Particulate total organic compounds (TOC)

exhibited a concentration range of 0.34-3.18 mol C/m3. Particulate formic

acid was most abundant acid followed by acetic, pyruvic, hydroxyl-

butyric and glycolic. Among the organic acids studied, oxalic acid was

the most abundant. In addition, correlation between oxalic and pyruvic

15

acid concentrations was high (r=0.67) indicated that these acids arise

from photochemical. During daytime, vehicular emission appeared to be

the primary source of acetic acid, whereas formic and pyruvic acids

appeared to be formed photochemically. Beside, emissions from biogenic

primary sources were also important contribution to atmospheric

concentrations of formic and glycolic acids. Presumably, the

photooxidation of pyruvic and glycolic acids gave rise to the oxalic acid.

At night, hydroxy-butyric acid levels decreased were similar formic,

acetic and pyruvic. Direct vehicular and biogenic emissions seem to be

the major sources of TOC in nocturnal measurements. Oxalic acid might

arise from vehicular emission, glycolic acid from biogenic emission and

formic acid from both sources.

Rhrl and Lammel (2002) determined of malic acid and other C4 dicarboxylic acids in atmospheric aerosol samples. It was found for both

rural and urban sites and for various types of air masses that in the

summer-time malic acid is the most prominent C4 diacid (64 ngm-3 by

average), exceeding succinic acid concentration (28 ng m-3 by average)

considerably. In winter-time considerably less, a factor of 4-15, C4 acids

occurred and succinic acid was more concentrated than malic acid.

Tartaric, fumaric and maleic acids were less concentrated (5.1, 5.0and 4:5

ngm-3 by average, respectively). Tartaric acid was observed for the first

time in ambient air. The results indicate that in particular anthropogenic

sources are important for the precursors of succinic, maleic and fumaric

acids. Biogenic sources seem to influence the occurrence of malic acid

significantly.

16

Yao et al. (2002) reported that the C3/C4 mass ratios from a

suburban site and two urban sites in Hong Kong were generally larger

than unity, suggesting that the primary vehicle emissions were not the

major source of dicarboxylic acids in the atmospheric particles at these

sites. Instead, secondary sources, such as in-cloud processes, were found

to be a major route of formation of dicarboxylic acids, based on the

similarity of the size distributions of these dicarboxylic acids and sulfate.

The urban measurements reported were made at sites 20-25m above the

ground level and not close to the heavy traffic, which may explain the

lower contribution of primary vehicular emissions to dicarboxylic acids

than when measured at the ground level close to the heavy traffic.

Ho et al. (2003) recently examined the chemical characterizations

of PM2.5 and PM10 at three different sites in Hong Kong: Hong Kong

Polytechnic University (HKPU), Kwun Tong (KT) and Hok Tsui (HT).

HKPU and KT are urban sites and close to the heavy traffic while HT is a

remote background site. Ho et al. (2003) found that the ratio of

organic carbon to elemental carbon was higher at HT than at HKPU and

KT. The organic carbon to elemental carbon ratio in winter was higher

than that in summer. Gas-aerosol equilibrium, favoring the partitioning of

semi-volatile organic species in the particulate phase under the lower

temperatures in the winter, may be an explanation for the observed

seasonal differences. The elevated ratio of organic carbon to elemental

carbon at HT could be due to a number of possible factors, including a

significant secondary source of organic carbon, a lower ambient

temperature and a higher biological emission flux at HT. We examine the

contribution of secondary chemical reactions to organic acids at HT using

the C3/C4 mass ratio.

17

18

Hsieh et al. (2008) studied speciation and temporal characterization

of dicarboxylic acids in PM2.5 during a PM episode and a period of non-

episodic pollution. Period between September and November 2004 in

suburban southern Taiwan and dicarboxylic acid and inorganic species

content and provenance were investigated. Oxalic acid was the dominant

dicarboxylic acid species, followed by succinic acid and malonic acid.

Tartaric acid concentrations were the lowest. There was 49.3% more

dicarboxylic acid in PM episode aerosol than in non-episodic aerosol.

However, daily oxalic acid concentration increased 72.7% in PM episode

aerosol, while succinic acid fell 20.9% and malonic acid fell 21.6%,

indicating higher conversion of these acids into oxalic acid in PM episode

aerosol. Dicarboxylic acid concentrations, particularly oxalic acid, peaked

at night during the PM episode. SO42-, NO3-, and NH4+ were also major

contributors to nighttime PM episode aerosol. The mass ratio of oxalic

acid to sulfate at this time was as high as 60.3%, substantially higher than

the 44.5% in non episodic aerosol. High correlations between Cl-, K+, and

Na+ and oxalic acid plus backward trajectory data indicate that biomass

burning in paddy fields may contribute to oxalic acid content in PM

episode aerosol in the study area, especially during nighttime

2.4 Anhydrosugars and sugar alcohols

Sugars or saccharides represent the major form of

photosynthetically assimilated carbon in the biosphere. The plant tissues

as structural polysaccharides like cellulose, hemicellulose and pectin. In

aerosols, the saccharides are comprised of three main groups: (1) primary

saccharides consisting of mono- and disaccharides, (2) saccharide polyols

or sugar alcohols (reduced sugars), and (3) anhydrosaccharide or

anhydrosugars derivatives such as mainly levoglucosan (1,6-anhydro--

D-glucopyranose). Saccharides are ubiquitous in urban, rural and remote

aerosol and, therefore, are potentially powerful tools in elucidating

organic carbon sources and atmospheric transport pathways (Simoneit et

al., 1999; Medeiros et al., 2006). In this study were found sugar alcohols

and anhydrosugars in ambient aerosol.

Sugar alcohols are produced in large amounts by many fungi, and

several functions have been proposed for these compounds, such as

storage or transport carbohydrates. Sugar alcohols often found on the

bark of trees, branches and leaves. Bacteria can also form and accumulate

polyols (e.g., sorbitol) in order to overcome osmotic stress. Polyols are

known component of bacteria, fungi, lichens, invertebrates and lower

plants, acting as osmoregulators, stress inhibitors or carbohydrate

suppliers (Medeiros et al., 2006). In general, sugar alcohols were found to

be most prevalent in the coarse fraction (Pio et al., 2008). Sugar alcohols,

such as arabitol and mannitol, are structurally related to levoglucosan.

These compounds are markers for fungal spores and mainly occur in the

coarse size fraction (Bauer et al., 2008).The sugar alcohols arabitol and

19

sorbitol were found in relative high concentration in leaf smoke samples

yielding around 0.14% (arabitol) and 0.25% (sorbitol) of the PM10. The

seasonal variation for arabitol, fungal spore production is the highest

during summer, is an excellent marker for airborne fungal spores (Zhang

et al., 2008).

Anhydrosugars, such as levoglucosan and mannosan, are formed in

pyrolysis process of cellulose and hemicelluloses containing materials,

and thus are important tracers for biomass burning emission (e.g. wood,

rice straw, leaves and biomass). Highly varying patterns were observed in

the emission profiles of various molecular markers as a function of fuel

type and combustion conditions (Engling et al., 2006; Schmidl et al.,

2008; Bari et al., 2009; Caseiro et al., 2009). Schmidl et al.(2008a)

described levoglucosan, mannosan and galactosan were found in high

concentrations (0.2-15% w/w) in all wood smoke samples. As expected

the anhydrosugar levoglucosan, which has long been known as a by-

product from the pyrolysis of cellulose was the most abundant organic

compound and was found in all analysed wood smoke samples. Average

concentrations ranged from 4.1% in beech wood smoke to 15.1% of total

particulate mass in larch wood smoke which is in general agreement with

levoglucosan contents of 0.797-31.82% found for American tree species

by Fine et al. (2001, 2002, and 2004). A sampling program was

implemented to study the chemical markers of wood smoke, including

monosaccharide anhydrides (MAs), soluble potassium, and several

methoxyphenols. Levoglucosan (1,6-anhydro--D-glucopyranose) has

been identified as a major constituent originating from pyrolysis of

cellulose. Levoglucosan is emitted at such high concentrations that it can

be detected at considerable distances from the original combustion source

20

(Simoneit et al., 1999). Levoglucosan and other anhydrosaccharides are

products from the thermal degradation of cellulose and hemicellulose and

are commonly used as tracers for wood smoke in the atmosphere (Fabbri

et al., 2009). Levoglucosan was measured with peak concentrations of

234 ng m-3 during periods with smoke influence from local fires, and

primary biomass burning smoke contributions to fine particle organic

carbon were estimated to be as high as 100% on individual days during

that period (Engling et al., 2006). The levoglucosan concentration exhibited a strong annual cycle with higher concentrations in the cold

season. The minor anhydrosugars had a similar annual trend, but their

concentrations were lower by a factor of about 5 and about 25 in the cold

season for mannosan and galactosan, respectively.

Mannosan, another anhydrosugar emitted during pyrolyses of

cellulosic material, was found to be a useful compound for distinguishing

between soft- and hardwood combustion. Mannosan is formed in the

pyrolysis of hemicelluloses containing mannose, which occur mainly in

conifers. Mannosan is the second most abundant anhydrosugar in the

wood smoke samples. Biomass smoke PM from conifers contains around

five times higher concentrations of mannosan than smoke PM from

deciduous trees The mannosan level of around 0.4% found in leaf smoke

PM10 is very similar to that from hardwood log combustion reported by

Schmidl et al. (2008a). Mannosan average concentration in the

background area was 16889 ng m-3, while the average concentration in the residential area was 313237 ng m-3 (Glasius et al., 2008)

21

The use of levoglucosan as a tracer for wood burning in general

and the levoglucosan and mannosan ratio to differentiate between

hardwood and softwood smoke to PM load. Considering the ratio of

levoglucosan to mannosan the difference between hard- and softwood

types becomes even more marked. Hardwoods give high ratios, around

14-15, while softwoods give low ratios, 3.6-3.9 (Schmidl et al., 2008a).

The average ratio and standard deviation of the ratio levoglucosan and

mannosan was found to be 4.60.7 (Ward et al., 2006). The ratio between levoglucosan and mannosan in the particulate emission from forest fire

was found to be 3.50.8 (Pio et al., 2008).

Moreover, relationships between the different anhydrosugars the

combustion of softwood was found to be dominant for the wood smoke

occurrence in ambient air at the investigated sites. Potassium, a

commonly used tracer for biomass burning, correlated well to

levoglucosan, with a mass ratio of around 0.80 in the cold season.

(Caseiro et al., 2009). Schmidl et al. (2008a) found K/levoglucosan ratios

of 0.005 and 0.05 for the major Austrian wood types beech and spruce,

espectively, when burnt in a small ceramic stove. In US studies burning

north American wood types in fireplaces (e.g. Fine et al., 2002, 2004)

K/levoglucosan ratios were in the range of 0.017-0.23. Despite those low

K/levoglucosan values from source (fireplaces, stoves,.) studies, authors

have reported higher ratios from smoke-impacted or non smoke-impacted

ambient aerosol.

Jordan et al. (2006) describes levoglucosan major constituent of

woodsmoke in ambient air collected in Launceston, Australia during the

22

winter months (May-September) of 2002-2003 were analyzed for organic

compounds. The proportion of radiocarbon (14C) in aerosols used to

apportion biomass which analyzes having relatively high precision and

accuracy. Levoglucosan is a suitable tracer species for quantifying the

contribution of wood smoke to air pollution. This report show

levoglucosan emissions from a woodsmoke averaged 14055 mg g-1 PM indicated that woodheaters is major source of air pollution in Launceston.

Zhang et al. (2008) estimated that levoglucosan as a molecular

marker are increasingly employed as biomass burning, were collect

samples PM2.5 and PM10 in Beijing from July 2002 to July 2003. The

samples were analyzed for levoglucosan, related saccharidic compounds,

organic and elemental carbon, and ionic species. Levoglucosan and

biomass burning particles are mainly present in the fine aerosol fraction.

The seasonal variation for arabitol suggests that fungal spore production

is the highest during summer. A long-range transported biomass burning

event was indentified for the case of 7 May 2003. Besides, some other

episodes were discussed. So, biomass burning is the only source for

levoglucosan, this phenomenon may be explained by biofuel combustion

in the countryside of suburban Beijing and neighboring provinces.

Caseiro et al., 2007 described for the quantification of primary

sugars, sugar alcohols and anhydrosugars in atmospheric aerosols. The

determination of saccharides in atmospheric aerosol could use as specific

tracers show in Table 2.1 provides a list of important saccharides found

in atmospheric aerosols. Saccharides present in atmospheric particulates

originate from different source types. Micro-organisms, plants and

23

24

animals can release into the atmosphere primary saccharides

(monosaccharides including glucose, fructose, xylose and disaccharides

such as sucrose and trehalose) while fungi, lichens and bacteria produce

saccharidic polyols, also denoted as sugar alcohols, such as arabitol,

mannitol and sorbitol. Anhydrosaccharides on the other hand, such as

levoglucosan, derived from cellulose, and galactosan and mannosan,

derived from hemicelluloses, are the primary thermal degradation

products of structural polysaccharides present in biomass.

Caseiro et al. (2009) reported that levoglucosan yearly averages

ranged from 0.12 to 0.48 g m-3. The sites in Graz showed higher

concentrations compared to the other regions, while background sites, in

general, evidenced slightly lower concentrations than urban sites. The

ratios between levoglucosan and mannosan and between levoglucosan

and galactosan showed a range of 4.1-6.4 and 11-22, respectively, in the

periods where biomass burning is expected to be a strong source.

Moreover, the ratio of levoglucosan and potassium, another tracer for

biomass burning, were well correlated at all sites. The ratios between

those two species were rather below 1 in the cold season and around 3 in

the warm season.

Table 2.1 Saccharides commonly found in atmospheric aerosol and their sources (Caseiro et al., 2007)

Compound Source

Primary sugars (mono- and disaccharides) Arabinose Lichens Fructose Lichens Soil biota Galactose Soil biota Glucose Fungi Lichens Soil biota Wood burning Mannose Soil biota Xylose Soil biota Maltose (monohydrated) Soil biota Sucrose Plants Soil biota Mycose (Trehalose) Yeast Bacteria, fungi Soil biota Sugar alcohols Arabitol Fungi, Lichens Erythritol Lichens Soil biota Glycerol Soil biota Inositol Soil biota Mannitol Fungal spores Fungi Lichens Soil biota Sorbitol Bacteria Lichens Soil biota Xylitol Fruits, berries, hardwood Soil biota Anhydrosugars Galactosan Wood burning (1,6-anhydro--D-galactopyranose) Levoglucosan (1,6-anhydro--D-glucose, Wood burning 1,6-anhydro--D-glucopyranose) Mannosan (1,6-anhydro--D-mannopyranose) Wood burning 1,6-Anhydrogluco-furanose Wood burning

25

Chapter 3 Experimental

3.1 Sampling and sampling sites

Chiang Mai is the big city northern part of Thailand and with a

population of about 82,000 inhabitants in the city. It is most important

biomass burning producers in the country during dry season. To facilitate

manual harvesting, its farmers still burning straw/leaves/agriculture waste,

or forest fire. This generates a great cloud of smoke that spread over the

city and its surrounding area.

Aerosol samples were collected on a 47-mm Teflon filters(Zefluor,

Pall) using a Ecotech MicroVol 1100 Particulate Sampler (Figure 3.1)

with a total flow rate of 3 L min-1, between February to April 2010. Three

sampling sites including Faculty of Architecture Chiang Mai University

(CMU site), TOT Public Company Limited (TOT site) and Doi Suthep-

Pui National Park Protection Unit (STM site) were selected for particulate

matter monitoring. The field descriptions were given as follows and

locations of the sites are shown in Figure 3.2.

Faculty of Architecture Chiang Mai University (CMU; Located at

latitude 18o4754.90 N and longitude 98o5655.75 E) was urban,

located about 2 km western of Chiang Mai City, medium traffic, hardly

impacted by anthropogenic activities, near the Suthep mountain and

excellent ventilation. The sampling monitors were placed on the rooftop,

set at a height of 12 m above ground.

26

TOT Public Company Limited (TOT; Located at latitude 18o 41

40.04 N and longitude 99o 2 59.45 E) was suburban. The former is

located about 15 km southeast of the city, alongside a busy street and

within the busy highway No. 11, are the traffic-impacted site. Industrial

zone (such as petrochemical, cement, ceramic and metal industrial) was

about 2 km eastern from sampling site.

Doi Suthep-Pui National Park Protection Unit (STM; Locate at

latitude 18o 48 32.71 N and longitude 98o 53 27.81 E), is surrounded

by mountains, the sampling set at the Suthep-Pui mountain, 1400 m

above sea level, located near the Bhubing Palace is Chiang Mai most

famous travel place, with a little traffic. STM site is a sampling site

situated outside the city and representative for regional moutain

conditions.

Intensive observation periods (IOP) of dry season were observed

during the two period of sampling. Firth period were collected between 2

March to 2 April 2010 (IOP1), which sampling at two sites, CMU and

TOT site. Second period were collected between 9 to 21 April 2010

(IOP2), sampling at two sites, CMU and STM site. Each sampling

collected two sets of aerosol samples were collected daily, one from 7

am to 7 pm (12 h: daytime) every 3 days and another from 7 pm to 7 am

(12 h: nighttime) every 3 days.

27

28

Figure 3.1 Ecotech MicroVol 1100 Particulate Sampler

STM sitelatitude 18o 48?32.71? N

longitude 98o 53?27.81? ECMU site

latitude 18o4754.90? N longitude 98o5655.75? E

TOT sitelatitude 18o 41?40.04? N longitude 99o 2?59.45? E

Chiang Mai City

Cement industrial

Ceramic industrial

Metal industrial

2 km2 mile

N

Chiang Mai

Bangkok

Thailand

Figure 3.2 Map of Chiang Mai Basin areas identifying the location of air

sampling sites.

3.2 Sampling handing

Before and after sample collection, filters were conditioned at

405% RH for 24 hours and subsequently weighed at 503% RH using a Mettler Toledo AT261 analytical balance with a sensitivity of 10 g and a

Sartorius CP2P analytical balance with a sensitivity of 1 g. All weight

measurements were repeated three or more times and the Shewart control

procedures were followed to ensure reliability. Additionally, blank filters

were prepared by purging in 99.995% pure nitrogen for 30 seconds and

then processed as for sample-containing filters.

3.3 Chemical analysis and quality assurance

The sample-containing filters, unexposed blanks will be stored in

petri dishes placed inside an unlit refrigerator below -18C to prevent loss of semi-volatile species. For analyzing compounds, the filter paper will

be placed in a PE bottle, 10.0 mL of deionized water (resistivity >18.0

M cm-1 at 25C, Barnstead) will be added and the contents will be shaken (Yihder TS-500 Shaker) in an unlit refrigerator at 4 C for 90 min to prevent the decomposition of the extracted carboxylic acid species.

The liquid is then filtered through a 0.2 m ester acetate filter and the

aqueous filtrate will be is characterized using IC, following a slightly

modified version of the method of Hsieh et al. (2008) and Tsai et al.

(2010).

29

The ion chromatography system (IC) model DX-600, Dionex is

equipped with a gradient pump (Model GP50), an ASRS-Ultra anion self-

regenerating suppressor, a conductivity detector (CD25), a Spectrasystem

automated sampler (AS3500) with 2 mL vials, and a Teflon injection

valve using a 1000 L sample loop, in combination with analytical

column and Ion Pac AG11-HC, AS-11-HC (4 mm), eluent for the DI

water (deionized), 5 mM NaOH, 100 mM NaOH and 100% MeOH

gradient elution method to conduct analysis. The flow rate is maintained

at 2.0 mL min-1 during the carboxylic acid analyses, which Ion

Chromatography Dionex DX-600 gradient elution ratio is shown in Table

3.1. This method allows for the analysis of acetic acid, formic acid,

glutaric acid, succinic acid, malonic acid, maleic acid, tartaric acid, malic

acid, fumaric acid, oxalic acid and phthalic acid in the aerosol samples.

Additionally, 1000 L of the aqueous extract will be injected into

IC Model Dionex ICS-2500 using 9 mM Na2CO3 eluent at a flow rate of

1.4 mL min-1. Concentrations of the separated inorganic species including

Cl-, NO3- and SO42-, are determined in analytical column RFICTM Ion Pac

AS14A, AG14A (4 mm). Cation system to IC Model Dionex ICS-1000,

AS1000, analytical column and Ion Pac CG12A, CS12A (4 mm),

injection volume 25 L and an isocratic 20 mM MSA (CH4O3S) eluent at

a flow rate of 1.0 mL min-1 will be used for determination of cations,

including Na+, NH4+, K+, Mg2+ and Ca2+. Department of anhydrosugars

(levoglucosan and mannosan) and sugar alcohols (arabitol, glycerol,

erythritol, trehalose dehydrate and mannose) are to IC Model Dionex

ICS-2500 (ED50, GP50, AS50), analytical column and Carbo PacTM

MA1 (4 mm), flow rate 0.4 mL/min, injection volume 0.2 mL, eluent

conducted for the 400 mM NaOH component analysis. Figure 3.3 shows

30

31

flow chart of MicroVol sampling and analysis. Moreover, Table 3.2

shows chemical structures of carboxylic acids and Table 3.3 showns

chemical structure of anhydrosugars and sugar alcohols.

All reagents are of analytical grade, obtained from Merck

(Darmstadt, Germany), and are used without further purification. The

solutions will be prepared using deionized water from which organic

carbon had been removed and the detection limits corresponded to 10-50

ng for the carboxylic acids investigated. Method detection limits (MDLs)

of four chemical compound groups measured using IC systems shown in

Table 3.4.

Table 3.1 Ion Chromatography Dionex DX-600 gradient elution ratio

Time(min) H2O 5 mM NaOH 100 mM NaOH

100% Methanol

0.0 80% 4 % 0 % 16 % 9.2 80 % 4 % 0 % 16 %

12.2 0 % 84 % 0 % 16 % 22.0 0 % 49 % 35 % 16 %

Before weighing Teflonfilters were condition at

40 5% RH 24 hr

Sampling PM10 aerosol by Micro Vol 1100

After weighing Teflon filters were condition at

50 5% RH 24 hr

15 mL of centrifuge tube placed and added deionized

water 5 mL

Vibration machine 90 minute

Filtered through a 0.2 m ester acetate filter

1000 L filtrated to IC-Dionex DX-600

Analyze: Carboxylic acids

200 L filtrated to IC-Dionex ICS-2500

Analyze: Anhydrosugars and

Sugar alcohols

1000 L filtrated to IC-Dionex ICS-2500

Analyze: Anion

25 L filtrated to IC-DionexICS-1000

Analyze: Cation

Figure 3.3 Step for MicroVol sampling and analysis flow chart

32

Table 3.2 The names and chemical structures of carboxylic acids

Common IUPAC Chemical Structural name name formula formula

Carboxylic acids

Formic acid methanoic acid HCOOH

Acetic acid ethanoic acid CH3COOH

Oxalic acid ethanedioic acid HOOC-COOH

Malonic acid propenedioic acid CH2(COOH)2

Succinic acid butanedioic acid C2H4(COOH)2

Glutaric acid pentanedioic acid C3H6(COOH)2

Maleic acid cis-butenedioic acid C2H2(COOH)2

Fumaric acid trans-butenedioic acid C2H2(COOH)2

Phthalic acid benzene-1,2-dicarboxylic acid C6H4(COOH)2

Malic acid monohydroxybutanedioic acid C2H3(OH)(COOH)2

Tartaric acid 2,3-

dihydroxybutanedioic acid

C2H2(OH)2(COOH)2

Citric acid 3-carboxy-3-hydroxy pentanedioic acid

C3H4(OH)(COOH)3

OHO

O

HO

O

HO

O

OH

33

Table 3.3 The names and chemical structures of anhydrosugars and sugar alcohols

Common IUPAC Chemical Structural name name formula formula

Anhydrosugars

Levoglucosan 1,6-anhydro--D-glucopyranose C6H10O5

Mannosan 1,6-anhydro--D-mannopyranose C6H10O5

Sugar alcohols

Arabitol (2R,4R)-pentane-1,2,3,4,5-pentol C5H7(OH)5

Glycerol propan-1,2,3-triol C3H5(OH)3

Erythritol (2R,3S)-butane-1,2,3,4-tetraol C4H6(OH)4

34

Table 3.4 Method detection limits (MDLs) of four chemical compound

groups measured using IC systems

Species MDL Species MDL Inorganic species (g m-3) a Carboxylic Acid (ng m-3) Sulfate 0.011 Acetic 12.99 Nitrate 0.018 Formic 9.24 Chloride 0.015 Glutaric 3.75 Sodium 0.022 Succinic 6.48 Ammonium 0.009 Malic 4.39 Potassium 0.008 Malonic 2.30 Magnesium 0.018 Tartaric 7.55 Calcium 0.007 Maleic 0.51 Fumaric 0.79

Sugar Alcohols (ng m-3) Oxalic 2.27 Arabitol 5.95 Phthalic 1.70 Glycerol 4.16 Citric 0.29 Erythritol 3.15 Anhydrosugars (ng m-3) Levoglucosan 81.03 Mannosan 15.10 a Assumption of sampling volume 6.34 m3

3.4 Other data

The charts of the wind rose observed in Figure 3.4 from Thailand

Pollution Control Department demonstrate that the difference in

prevailing wind direction during intensive observed two periods due to

different source of pollutant. During IOP1, wind blows predominately

from northwestern to north and second from southeastern, which carries

as pollutant from the agriculture burning, forest fire burning and rural

areas towards Chiang Mai. As during IOP2, prevailing wind direction

from southeastern to south (which would bring pollutant from industrial

and agriculture areas) up from other province to Chiang Mai were shown

to preferentially occur during dry season. These results show that

35

36

influence of wind direction due to difference source of pollutant from two

intensive periods significantly. Moreover, wind blows at speed of 3.6-5.7

m/s of IOP2 higher than IOP1.

a IOP1

b IOP2

Figure 3.4 Wind rose charts during intensive observation period (a) IOP 1and (b) IOP2

Chapter 4 Results and discussion

4.1 Meteorological conditions

The ambient air quality data were obtained from the Thailand Pollution

Control Department (PCD), information was obtained on Air Quality data from

February to April 2010 over Chiang Mai province, Thailand. The Air Quality

was particularly useful for observing pollutant concentrations. Moreover

visibility was obtained from Thai Meteorological Department during sampling.

Site meteorological data shows in Table 4.1 confirm designations of each

period of study. In this study to explain two period was the non-episodic

pollution period (PM10120 g m-3).

During the PM10 episode and non-episodic pollution period, average PM10

concentrations were 156.8838.10 g m-3 and 78.7120.42 g m-3 in IOP1 period versus 105.9316.79 g m-3 in IOP2 period, respectively. The data shows concentrations of pollutant, especially O3, SO2, NO2, NOx, NO and CO,

which represents traffic emission, IOP1 period was higher during the PM10

episode. In addition, wind speed during intensive observed nighttime lower than

daytime both of period. The results suggest that due to accumulated pollutant in

the nighttime and contributed in the daytime. Moreover, higher temperature,

lower relative humidity and lower wind speed during the PM10 episode to be

high O3 is due to lower visibility, higher PM10 and increase pollutant of these

periods. However IOP2 period of this study were high temperature, lower

relative humidity, low wind speed and high O3 than IOP1 due to higher SO2,

lower visibility, high PM10 can to increase pollutant emission to atmospheric

aerosol in Chiang Mai basin.

37

Table 4.1 Meteorological and related air pollution information during the

period of study at the suburban site.

During non-episode During the PM episodeIOP1 IOP2 IOP1 Parameter

Mean SD Mean SD Mean SD Temperature ( C) 25.61 3.38 31.83 2.42 28.49 2.65 Relative humidity (%) 54.07 7.88 43.83 2.42 52.11 7.09 Pressure (mmHg) 731.17 1.56 729.18 1.31 731.92 1.38 Visibility (km) 8.21 1.24 7.13 0.54 5.80 1.38 Prevailing wind direction NW-N SE-S NW-N Wind speed 1.68 0.54 1.65 0.66 1.60 0.43

PM10 (g m-3) 78.71 20.42 105.93 16.79 156.88 38.10

O3 max (ppb)a 49.06 29.42 91.00 26.51 90.86 31.21

O3 (ppb) 18.72 18.97 50.71 17.93 39.58 24.24

SO2 (ppb) 0.70 0.41 1.77 0.65 1.53 0.46

NO2 (ppb) 14.64 5.34 11.62 2.75 14.69 3.12

NOx (ppb) 19.78 7.44 13.50 2.72 17.94 2.56 NO (ppb) 5.16 2.61 1.97 0.32 3.30 1.11 CO (ppm) 0.78 0.19 0.91 0.10 1.11 0.26 a Average maximum hourly ozone in each sampling sets. Note: Wind speed during IOP1 daytime= 1.890.48 m/s and nighttime = 1.440.47 m/s Wind speed during IOP2 daytime= 2.160.46 m/s and nighttime = 1.140.34 m/s

38

4.2 Mass concentration of PM10 aerosols

Figure 4.1 shows the variations of PM10 mass concentration during dry

season two periods, firth intensive observation period (IOP1) from 2 Mar. to 2

Apr. in 2010 and second intensive observation period (IOP2) during 9-20 April

2010, comparison intensive observation of two periods between CMU site and

TOT site (IOP1) versus CMU site and STM site (IOP2) of this study with PM10

mass concentration from PCD data in Chiang Mai. In this study, show similar

pattern with PCD data. The average concentration of PM10 at IOP1-TOT site

higher than IOP1-CMU site, IOP2-CMU site and IOP2-STM, respectively,

which IOP1 mass concentration of PM10 higher than IOP2 resulted from during

collected sampling IOP1 during episode of pollutant, indicating that ambient air

pollution in Chiang Mai influenced on the sample sites of this study. Moreover,

the PM10 concentration during non episodic pollution and PM10 episode of

IOP1-TOT site (77.4425.07 g m-3 versus 142.4617.74 g m-3) higher than IOP1-CMU site (58.0830.85 g m-3 versus 139.5613.97 g m-3) and IOP2-CMU site (81.9629.64 g m-3) versus IOP2-STM site (65.1620.05 g m-3) , respectively, during non episodic pollution period shows in Table 4.2,

indicating that PM10 was evenly distributed in suburban more than urban and

mountain site, which TOT site is located in southern part of Chiang Mai basin

and closely industrial area. Therefore, pollutants can be transported from up

wind area or due to the emissions from industrial activities.

39

IOP2

Date

4/9~

4/11

, 201

0(D

)

4/9~

4/11

, 201

0(N

)

4/12

~4/1

4, 2

010(

D)

4/12

~4/1

4, 2

010(

N)

4/15

~4/1

7, 2

010(

D)

4/15

~4/1

7, 2

010(

N)

4/18

~4/2

0, 2

010(

D)

4/18

~4/2

0, 2

010(

N)

PM10

conc

entr

atio

n

0

20

40

60

80

100

120

140

160

180

PM10 conc. of PCD dataPM10 conc. of CMU sitePM10 conc. of STM site

IOP1

Date

2/2~

2/4,

201

0(D

)2/

2~2/

4, 2

010(

N)

2/5~

2/7,

201

0(D

)2/

5~2/

7, 2

010(

N)

2/8~

2/10

, 201

0(D

)2/

8~2/

10, 2

010(

N)

2/11

~2/1

3, 2

010(

D)

2/11

~2/1

3, 2

010(

N)

2/14

~2/1

6, 2

010(

D)

2/14

~2/1

6, 2

010(

N)

2/17

~2/1

9, 2

010(

D)

2/17

~2/1

9, 2

010(

N)

2/20

~2/2

2, 2

010(

D)

2/20

~2/2

2, 2

010(

N)

2/23

~2/2

5, 2

010(

D)

2/23

~2/2

5, 2

010(

N)

2/26

~2/2

8, 2

010(

D)

2/26

~2/2

8, 2

010(

N)

3/1~

3/3,

201

0(D

)3/

1~3/

3, 2

010(

N)

3/4~

3/6,

201

0(D

)3/

4~3/

6, 2

010(

N)

3/7~

3/9,

201

0(D

)3/

7~3/

9, 2

010(

N)

3/10

~3/1

2, 2

010(

D)

3/10

~3/1

2, 2

010(

N)

3/13

~3/1

5, 2

010(

D)

3/13

~3/1

5, 2

010(

N)

3/16

~3/1

8, 2

010(

D)

3/16

~3/1

8, 2

010(

N)

3/19

~3/2

1, 2

010(

D)

3/19

~3/2

1, 2

010(

N)

3/22

~3/2

4, 2

010(

D)

3/22

~3/2

4, 2

010(

N)

3/25

~3/2

7, 2

010(

D)

3/25

~3/2

7, 2

010(

N)

3/28

~3/3

0, 2

010(

D)

3/28

~3/3

0, 2

010(

N)

3/31

~4/2

, 201

0(D

)3/

31~4

/2, 2

010(

N)

PM10

conc

entr

atio

n

0

20

40

60

80

100

120

140

160

180

200

220

240

260

PM10 conc. of PCD dataPM10 conc. of CMU sitePM10 conc. of TOT site

Figure 4.1 PM10 mass concentration of intensive observation period with

PCD data during period of study

40

41

Figure 4.2 shows the relationship between PM10 from PCD data and PM10

from during intensive observed. During IOP1, PM10 concentration from CMU

site and TOT has the good relationship with PM10 from PCD data, while during

IOP2 the available data is too less for analysis.

IOP1-CMU site

PM10 concentration of PCD data

0 50 100 150 200 250

PM10

con

cent

ratio

n

0

50

100

150

200

250 IOP1-TOT site

PM10 concentration of PCD data

0 50 100 150 200 250

PM10

con

cent

ratio

n

0

50

100

150

200

250

IOP2-CMU site

PM10 concentration of PCD data

0 50 100 150 200 250

PM10

con

cent

ratio

n

0

20

40

60

80

100

120

140IOP2-STM site

PM10 concentration of PCD data

0 50 100 150 200 250

PM10

con

cent

ratio

n

0

20

40

60

80

100

y = 0.923x - 5.794r = 0.801

y = 0.632x + 50.23r = 0.692

y = 1.044x - 28.67r = 0.591

y = 0.098x + 54.74r = 0.077

Figure 4.2 Correlation of PM10 concentration from PCD data with PM10

concentration from observed site during period of this study

4.3 Aerosol composition of PM10 during non episodic pollution and PM10

episode periods

The concentrations and standard deviations of inorganic salts, carboxylic

acids, anhydrosugars, sugar alcohols and ratios of aerosol components during

non episodic pollution period and PM10 episode collected at sampling sites of

this study are shown in Table 4.2.

The sulfate concentration in PM10 during observation period of this study,

IOP2 (7.975.96 g m-3 at CMU site versus 5.472.91 g m-3 at STM site) higher than IOP1, which sulfate increase 3 times. During IOP2 concentration

with peak daily maximum high ozone concentration was 91.0026.51 ppb (Table 4.1). These results may be during the ozone increase due to high sulfate,

significant photochemical formation processes. It may be increased

anthropogenic activity and photochemical reaction, which IOP2 period

collected during the Songkran celebrations (Thailand New Years Day) are still

in the northern city of Chiang Mai, where most famous for foreigners and many

visitor to become a party or may be contributed from biomass burning due to

high levoglucosan of this period, also high sulfate in ambient aerosol. Sulfate

was the dominant inorganic salts, whereas during non episodic pollution IOP1-

TOT site nitrate was the dominant species. Particulate nitrate is transformed

through the photo-oxidation of NO2 derived from combustion of fossil fuels

(Logan, 1983). The major source of nitrate in TOT site may be NOx emission

from traffic vehicles. Particulate ammonium mainly originates from ammonia

vapor. Ammonium sulfate is the most stable while ammonium chloride is the

most volatile, hence ammonia prefers to react with sulfuric acid or sulfate (HO

et al., 2003).

42

Table 4.2 Mean (SD) chemical composition of PM10 aerosol during non-episode pollution period and PM10 episode emitted from sampling sites

During non episodic period pollution During the PM10 episode

IOP1-CMU (n=19) IOP1-TOT (n=14) IOP2-CMU (n=8) IOP2-STM (n=8) IOP1-CMU (n=10) IOP1-TOT (n=15) Species

Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD PM10 (g m-3) 58.08 30.85 77.44 25.07 81.96 29.64 65.16 20.05 139.56 13.97 142.46 17.74 Inorganic species (g m-3) 7.30 3.92 7.85 4.04 13.90 6.81 11.34 4.63 15.60 4.54 13.76 4.55 Sodium 0.48 0.49 0.61 0.81 0.63 0.25 0.71 0.65 1.19 1.04 0.63 0.78 Ammonium 1.23 0.89 0.93 0.65 1.21 0.42 1.35 0.65 2.71 0.79 2.04 0.72 Potassium 0.89 0.57 0.83 0.45 0.78 0.37 0.75 0.28 2.31 0.56 1.92 0.52 Magnesium 0.31 0.30 0.26 0.10 0.11 0.03 0.12 0.04 0.73 0.28 0.56 0.27 Calcium 0.92 0.61 1.26 0.46 1.26 0.43 0.93 0.30 1.88 0.67 2.09 0.85 Chloride 0.25 0.18 0.49 0.60 0.63 0.26 0.62 0.21 0.59 0.92 0.48 0.62 Nitrate 1.22 0.62 2.08 1.53 1.30 0.38 1.36 0.55 2.34 1.95 2.64 2.12 Sulfate 1.99 1.46 1.39 0.87 7.97 5.96 5.49 2.91 3.86 1.08 3.41 1.09 Carboxylic acids (ng m-3) 1037.97 517.76 1125.85 580.35 1347.66 428.71 909.48 398.47 1950.83 547.17 1605.77 442.38 Acetic acid 332.47 266.09 486.07 386.85 385.02 227.48 258.73 200.61 422.48 326.74 456.33 313.68 Formic acid 58.16 37.57 51.73 34.95 66.94 40.19 57.57 44.56 115.35 89.97 51.70 25.42 Glutaric acid 11.76 10.27 9.19 6.95 51.84 38.18 24.06 16.17 42.73 24.90 34.37 21.28 Succinic acid 37.61 36.50 24.16 13.80 61.05 38.64 26.92 18.15 105.73 51.68 70.05 32.28 Malic acid 59.10 46.08 45.65 22.66 67.05 37.05 35.02 22.85 136.36 51.71 94.60 47.42 Malonic acid 48.76 27.43 41.65 15.67 61.63 29.89 33.02 21.03 99.54 32.50 77.68 24.56 Tartaric acid 31.90 30.38 33.25 34.89 48.48 28.65 23.91 16.97 54.33 23.80 42.93 24.48 Maleic acid 86.20 66.47 121.26 67.29 44.28 22.58 35.54 26.68 92.32 78.97 93.10 81.35 Fumaric acid 9.59 7.33 6.77 5.78 38.28 30.52 14.28 14.41 27.38 13.92 15.99 10.89 Oxalic acid 284.32 207.48 243.52 127.58 477.50 73.49 374.13 142.29 758.87 198.82 610.56 162.61 Phthalic acid 29.59 23.46 25.04 19.09 6.92 2.94 6.81 4.81 54.39 31.67 37.68 24.24 Citric acid 48.52 29.87 37.57 33.43 38.68 28.08 19.48 18.14 41.34 34.77 20.77 15.96 Anhydrosugars (ng m-3) 391.32 189.11 466.69 258.22 1016.01 567.11 996.14 581.95 1258.76 797.26 1145.37 645.90 Levoglucosan 333.18 174.43 413.01 253.56 988.69 567.78 967.32 582.46 1175.50 791.26 1073.13 646.88 Mannosan 58.14 19.83 53.68 22.43 27.32 3.16 28.82 7.47 83.25 20.68 72.24 23.40 Sugar alcohol (ng m-3) 291.77 102.24 271.84 84.70 175.29 32.80 189.02 50.12 746.90 266.15 573.69 242.68 Arabitol 214.86 80.73 192.71 61.36 134.01 15.48 129.07 31.00 334.16 54.00 303.59 70.48 Glycerol 59.47 41.20 61.10 55.00 26.92 13.36 45.31 21.90 346.55 194.37 222.09 161.44 Erythritol 17.45 7.30 18.03 5.83 14.37 7.21 14.64 6.67 66.19 29.43 48.01 30.98 Ratios A/F 5.72 9.40 5.75 4.49 3.66 8.83 M/S 1.30 1.72 1.01 1.23 0.94 1.11 Oxalic/Sulfate 0.14 0.18 0.06 0.07 0.20 0.18 K/Levo 2.68 2.02 0.79 0.78 1.96 1.79 Levo/Manno 5.73 7.69 36.19 33.56 14.12 14.85 Levo/PM10(%) 0.57 0.53 1.21 1.48 0.84 0.75

43

In addition ammonium and sulfate are photochemical end-products (Hsieh

et al., 2007, 2008). In this study, ammonium and potassium higher during the

PM10 episode at CMU site, while source of particulate potassium are biomass

combustion, indicating that urban area is significant source from the secondary

photochemical product from traffic emission or biomass burning. High calcium

concentration observed at TOT site. Calcium is a constituent in coal fly ash also

biomass smoke and, dependent on regional geology, a constituent of mineral

dust (Salam et al., 2003). Crustal matter originates from soil and road side were

observed among the crustal elements (magnesium and calcium) (Ho et al.,

2003), noted that at suburban area might be contributor mineral dust from

industrial or biomass burning.

The water soluble organic acids were monocarboxylic acids (acetic acid

and formic acid), dicarboxylic acids and tricarboxylic acids. Carboxylic acid

concentration during the PM10 episode is higher than that during non-episodic

period. Acetic acid was the most abundant monocarboxylic acids follow by

formic acid. Oxalic acid was the dominant dicarboxylic acids both periods.

During non episodic pollution period, IOP1, oxalic acid was significantly higher

during the IOP2 than during the IOP1. The most pronounced period difference

was shown by maleic acid, which had high concentrations in during IOP1 but

was only detected during IOP2 low concentrations demonstrates that the

molecular composition of these acids during two intensive observation period

was also different. During the PM10 episode, oxalic acid was the most abundant

species of the detected dicarboxylic acids, followed by malic acid, which the

maximum concentration peak occurred at CMU site for oxalic acid high value at

CMU site during dry season indicating that secondary photochemical end

product be a major source of oxalic acids in urban site. Its might come from the

secondary formation of traffic emission (Wang et al., 2007b; Hsieh et al., 2008,

44

2009). In contrast, IOP1 during non episodic, oxalic acid was most abundant

species follow by maleic and malic acid. It is interesting to notice that during

IOP1 the unsaturated dicarboxylic acids, maleic acid. These results may be

source of maleic acid contributed from coal burning or wood burning during

firth intensive period.

Anhydrosugars (levoglucosan and mannosan) were identified in aerosol

sample, found in maximum concentrations at CMU site of 1258.76797.26 ng m-3 during the PM10 episode. Levoglucosan concentration was the dominant

anhydrosugars, follow by mannosan during both periods. During non episodic

pollution period IOP2, levoglucosan was higher than IOP1. As a result, it may

be biomass burning largely contributed to atmospheric aerosols from

levoglucosan emitted.

On the other hand, concentrations of levoglucosan observed in

atmospheric particulate matter, the annually averaged of range from 0.12-0.48

g m-3 in wood smoke (Caseiro et al., 2009). Bergauff et al. (2009) showed

levoglucosan from wood stove was an overall decrease of 50% from 3036344 ng m-3 to 1537117 ng m-3, but little change between the last years of the program (2006/2007 and 2007/2008). Santos et al. (2002) reported that

concentration of levoglucosan range from 0.15-1.65, 0.36-6.83 ng m-3 and 0.19-

28.42 ng m-3 at the downtown Corpo de Bombeiros, suburban and countryside

site, and from 10.55-35.06 ng m-3 and 2.7 ng m-3 in the smoke samples from the

burnt leaves and bagasse, respectively. Ward et al. (2006) observed

levoglucosan concentration in the range 900-6000 ng m-3 from the Missoula

smoke sample during wildfire season. In this study, indicating that levoglucosan

45

was found to be the most useful marker for biomass burning generated from

natural forest fire event in the mountain around Chiang Mai basin.

Moreover, levoglucosan was the most abundant anhydrosugars. Compared

to wood combustion smoke the levoglucosan emissions relative to PM trend to

be a little lower in the forest fire smoke samples. In a recent study

concentrations of 4-15% levoglucosan in smoke PM of common mid-

Europeanwood types (Schmidl et al., 2008a, b) were reported. Zhang et al.

(2008) found that the levoglucosan/PM percentage ratio was, on average 4.5%

(2.9-6.5%) contribution from biomass burning. The ratios of levoglucosan to

PM in fireplace emission are 0.8-26% (Fine et al., 2002, 2004). In this study,

levoglucosan constituted 0.53-1.48% of PM10, indicating that biomass burning

source emission was major types of forest fire in Chiang Mai basin. Another

anhydrosugar, mannosan emitted during pyrolyses of hemicelluloses material,

was found to be a useful compound for distinguishing between soft- and

hardwood combustion. Mannosan concentration was in range 27.32-83.25

ng m-3 of this study, higher during the episode PM10 at CMU site. Caseiro et al.

(2009) reported that mannosan concentration were lower than levoglucosan

ones, the average for cold season range from 35-68, 69-212 and 46-69 ng m-3 in

the Vienna, Graz and Salzburg regions, respectively.

Concentration of sugar alcohols was found during the PM10 episode higher

level than during non episodic pollution period, which maximum at CMU site.

Pio et al. (2008) reported Total sugar alcohol and monosugar concentrations

ranged from about 40 to 100 ng m-3, with the maximum level found in a sample

less affected by the smoke plumes. In general, sugar alcohols were found to be

most prevalent in the coarse fraction. Arabitol was the dominant sugar alcohols

follow by glycerol and erythritol. Sugar alcohols often found on the bark of

46

47

trees, braches and leaves (Mederios et al., 2006), indicating that during dry

season, biomass burning contributed sugar alcohols emitte