A.Batsaikhan

A.Batsaikhan
Reactive organic species
on natural dust
Ariunaa Batsaikhan
Dissertation
Universität Heidelberg
2007
Inaugural – Dissertation
zur
Erlangung der Doktorwürde
der
Naturwissenschaftlich – Mathematischen Gesamtfakultät
der
Ruprecht – Karls – Universität
Heidelberg
vorgelegt von
Ariunaa Batsaikhan
aus Sukhbaatar/Selenge, Mongolei
Master of Science in Chemistry
2007
Thema
Reactive organic species on natural dust
Gutachter:
Prof. Dr. Heinz Friedrich Schöler
Prof. Dr. Dr. h.c. mult. German Müller
Promotionsdatum: 26.07.2007
Abstract
Annually 1000-3000 Tg mineral dust aerosol are emitted into the atmosphere, and
transported over the oceans from one continent to the other. During the transport dust particles
interact with components in the marine atmosphere and also with seawater as they fall into the
ocean. Increased methyl iodide concentrations were observed during a field campaign on the
Atlantic Ocean when dust storms occurred.
Volatile halogenated organic compounds (VHOC) are photolyzed to produce reactive
halogen species which are responsible for ozone depletion. An abiotic production mechanism
for VHOC, involving humic-like substance (HULIS), iron and halide, was supposed to
produce methyl iodide through the interaction of dust particles with seawater as all necessary
ingredients were present. The main goal of this study was to test this hypothesis and to further
elucidate the process. For this, simple dust-seawater addition experiments in headspace
glasses were conducted in the laboratory, following a purge-and-trap GC-MS analysis of the
headspace gas.
Dust samples were collected in the source regions in southern Algeria and the Gobi
Desert and, as representatives for aeolian dust, samples from Cape Verde Island and
Lanzarote Island were used. To exclude the biological contribution, sterilized samples were
also employed in this study. As assumed, methyl iodide was produced abiotically and the
concentration increased tenfold after addition of Fe (III) within half an hour. Methylene
chloride was also produced abiotically along with methyl iodide. In contrast to methyl iodide
and methylene chloride, methyl chloride and isoprene were produced biologically, provided
the production occurred after at least 24 hours of interaction of only non-sterilized samples
with seawater. If the microorganisms responsible for the production of isoprene are common
soil organisms found everywhere in the world, this process can be the reason for a hitherto not
fully explained increase in atmospheric isoprene concentration during wet seasons, especially
when the rain falls practically everyday. The results of this study show the importance of
natural dust aerosols for the production and emission of volatile organic compounds to the
atmosphere and open interesting questions for further studies.
Kurzfassung
Interkontinentale Staubstürme transportieren jährlich riesige Mengen (1000-3000 Tg)
an feinstem Mineralstaub, vorwiegend aus den Wüsten, über den ganzen Globus. Während
dieses Transportes reagieren die Mineralstaubpartikel über den Meeren mit Komponenten der
ozeanischen Atmosphäre, sowie mit dem Meerwasser. Bei Feldmessungen im Atlantik
wurden jeweils nach Sandstürmen erhöhte Methyliodid Konzentrationen gemessen.
Methyliodid und andere leichtflüchtige halogenorganische Verbindungen produzieren
unter Einwirkung von UV-Strahlung reaktive Halogenverbindungen, die maßgeblich für die
Zerstörung der Ozonschicht verantwortlich sind. Es wurde vermutet, dass Methyliodid infolge
des abiotischen Prozesses gebildet wird, der zwischen organischem Material, Fe (III) und
Halogenid stattfindet, da durch die Wechselwirkung von Staubpartikeln mit Ozeanwasser
eigentlich alle für die Bildung von Methyliodid notwendigen Bestandteile vorhanden sind.
Zielsetzung der vorliegenden Arbeit war es, diese Vermutung über die Bildung von
Methyliodid zu überprüfen, und den Prozess genauer zu untersuchen. Dafür wurden
Mineralstaub und Seewasser in Headspace-Gläsern gemischt und danach die Gasphase mit
Purge-und-Trap GC-MS analysiert. Die Mineralstaub-Bodenproben wurden aus der Sahara im
Süden Algeriens und aus der Wüste Gobi genommen, und repräsentativ für den durch Stürme
übertragenen Mineralstaub, wurden Proben von den Kapverdischen Inseln und von Lanzarote
verwendet. Um eine biologische Produktion von Methyliodid ausschliessen zu können,
wurden die Proben teilweise auch sterilisiert. Bei den Untersuchungen wurde dann, wie
erwartet Methyliodid abiotisch produziert, und nach Zugabe von Fe (III) stieg die
Konzentration um das Zehnfache innerhalb von 30 Minuten.
Ebenso bildete sich auch
Methylenchlorid.
Im Gegensatz zu Methyliodid und Methylenchlorid, entstanden Methylchlorid und
Isopren nur biologisch. Sie wurden frühestens nach 24 Stunden, und auch nur in
unsterilisierten Proben mit Meerwasser gebildet. Falls die, für die Produktion von Isopren in
Frage kommenden Mikroorganismen den üblichen Bodenorganismen entsprechen, die man
überall auf der Welt finden kann, könnte dieser Prozess ein Grund für das bisher nicht völlig
geklärte Ansteigen von Isopren in der Atmosphäre in Regenzeiten sein.
Die Ergebnisse dieser Arbeit zeigen, dass Mineralstaub eine grosse Bedeutung für die
Produktion und Emission von leichtflüchtigen halogenorganischen Verbindungen in der
Atmosphäre hat, und es ergeben sich interessante Fragen für zukünftige Untersuchungen.
TABLE OF CONTENTS
1
INTRODUCTION AND RESEARCH OBJECTIVES................................................ 1
2
EXPERIMENTAL SECTION ....................................................................................... 3
2.1 Samples.......................................................................................................................... 3
2.2 Chemicals ...................................................................................................................... 4
2.3 Analytical methods....................................................................................................... 5
2.3.1
X-ray fluorescence (XRF).................................................................................... 5
2.3.2
Carbon, Sulfur analysis ....................................................................................... 5
2.3.3
Water content and total organic content............................................................ 5
2.3.4
TOC ....................................................................................................................... 6
2.3.5
Anions.................................................................................................................... 6
2.3.6
Cations................................................................................................................... 7
2.3.7
pH........................................................................................................................... 7
2.4 Determination of volatile organic compounds........................................................... 8
3
2.4.1
GC-ECD ................................................................................................................ 8
2.4.2
A purge and trap GC-MS.................................................................................... 9
CHARACTERIZATION OF SAMPLES ................................................................... 10
3.1 Results and Discussion of XRF analysis................................................................... 10
3.2 Results and Discussion of Carbon, Sulfur analysis ................................................. 14
3.3 Results and Discussion of water content and total organic content analysis........ 17
3.4 Results and Discussion of TOC analysis .................................................................. 18
3.5 Results and Discussion of Anion analysis................................................................. 19
3.6 Results and Discussion of Cation analysis................................................................ 22
3.7 Results and Discussion of pH analysis...................................................................... 28
3.8 Ion balance .................................................................................................................. 29
3.9 Conclusion................................................................................................................... 30
4
ABIOTIC PRODUCTION OF METHYL IODIDE AND METHYLENE
CHLORIDE FROM THE INTERACTION BETWEEN DUST PARTICLES AND
SEA WATER ................................................................................................................. 31
4.1 Summary ..................................................................................................................... 31
4.2 Introduction ................................................................................................................ 32
4.3 Materials and Methods .............................................................................................. 33
4.4 Results and Discussion ............................................................................................... 34
4.5 Conclusion................................................................................................................... 42
5
PRODUCTION OF ISOPRENE FROM THE INTERACTION BETWEEN DUST
PARTICLES AND SEA WATER ............................................................................... 43
5.1 Summary ..................................................................................................................... 43
5.2 Introduction ................................................................................................................ 44
5.3 Materials and Methods .............................................................................................. 46
5.4 Results and Discussion ............................................................................................... 47
5.4.1
Production of isoprene ....................................................................................... 47
5.4.2
Dependence of isoprene amount on the organic content of the dust ............. 49
5.4.3
Production rate of isoprene and the experiment with distilled water ........... 51
5.5 Conclusion................................................................................................................... 60
6
PRODUCTION OF METHYL CHLORIDE FROM THE INTERACTION
BETWEEN DUST PARTICLES AND SEA WATER............................................... 61
6.1 Summary ..................................................................................................................... 61
6.2 Introduction ................................................................................................................ 62
6.3 Materials and Methods .............................................................................................. 63
6.4 Results and Discussion ............................................................................................... 64
6.5 Conclusion................................................................................................................... 72
7
CONCLUSION AND FUTURE PERSPECTIVES ................................................... 73
8
REFERENCES .............................................................................................................. 75
9
APPENDIX .................................................................................................................... 87
ABBREVIATIONS AND SYMBOLS
AAS
Atomic Absorption Spectroscopy
amu
atomic mass unit
CFCs
chlorofluorocarbons
C/S
carbon sulfur
DOC
dissolved organic carbon
DOAS
Differential Optical Absorption Spectroscopy
DIN
Deutsches Institut für Normung
DMADP
dimethylallyl diphosphate
d
density
d
diameter
ECD
Electron Capture Detector
EMMA
Energy-dispersive Miniprobe Multielement Analyzer
GHG
Green house gas
GF-AAS
Graphite furnace Atomic Absorption Spectroscopy
GC
Gas Chromatography
GC-ECD
Gas Chromatography- Electron Capture Detector
GC-MS
Gas Chromatography-Mass Spectrometry
HPLC-MS
High Performance Liquid Chromatography- Mass Spectrometry
HULIS
humic-like substance
ICP-OES
Inductively Coupled Plasma-Optical Emission Spectroscopy
IC
Ion Chromatography
IC
inorganic carbon
IR
Infra red
LLD
Lower limit of detection
MACR
methacrolein
MBL
marine boundary layer
MS
Mass Spectrometry
MVK
methyl vinyl ketone
NDIR
non dispersive infra red
NIST
National Institute of Standards and Technology
NMR
Nuclear Magnetic Resonance
OES
Optical Emission Spectroscopy
PANs
peroxyacyl nitrate
ppm
parts per million
pptv
parts per trillion per volume
SOA
Secondary organic aerosol
TOC
total organic carbon
UV
ultra violet
VHOC
volatile halogenated organic compound
VOC
volatile organic compound
XRF
X-ray fluorescence
LIST OF FIGURES
Figure 1.
Temperature program of GC oven.................................................................... 9
Figure 2.
Calcium concentration in the samples............................................................. 10
Figure 3.
Bromine concentration in the samples. ........................................................... 10
Figure 4.
Concentration of manganese versus iron content in dust samples. .............. 13
Figure 5.
Water and organic content of the samples...................................................... 17
Figure 6.
Dissolved organic and inorganic carbon contents of the samples................. 18
Figure 7.
Dependence of anion content on the size of the samples................................ 20
Figure 8.
Percentage of dissolved iron during 24-hour leaching experiment. ............. 23
Figure 9.
Percentage of dissolved manganese during 24-hour leaching experiment... 24
Figure 10. Percentage of dissolved aluminum during 24-hour leaching experiment.... 25
Figure 11. Percentage of dissolved calcium during 24-hour leaching experiment. ....... 26
Figure 12. Percentage of dissolved potassium during 24-hour leaching experiment. ... 26
Figure 13. The effect of Fe (II) and Fe (III) addition on the production of methyl iodide
through the interaction of dust sample Sahara 5 with seawater................... 35
Figure 14. Amount of methylene chloride produced through the interaction of dust
sample Sahara 4 with seawater after certain periods. ................................... 36
Figure 15. Amount of methylene chloride produced every 24 hour through the
interaction of dust sample Sahara 4 with seawater........................................ 37
Figure 16. Amount of methylene chloride produced every 24 hour through the
interaction of dust sample Sahara 4 with distilled water. ............................. 38
Figure 17. Total amount of methylene chloride produced through the interaction of
dust sample Sahara 4 with seawater................................................................ 39
Figure 18. Total amount of methylene chloride produced through the interaction of
dust sample Sahara 4 with distilled water. ..................................................... 39
Figure 19. Amount of methylene chloride produced every 24 hour through the
interaction of sterilized dust sample Sahara 4 with distilled water. ............. 40
Figure 20. Total amount of methylene chloride produced through the interaction of
sterilized dust sample Sahara 4 with distilled water...................................... 41
Figure 21. Amount of isoprene produced through the interaction of 4 g of dust samples
Sahara 4 with 10 ml of Atlantic Ocean water after certain periods. ............ 47
Figure 22. Amount of isoprene produced through the interaction of 1 g of dust samples
Sahara 5 with 10 ml of Atlantic Ocean water after certain periods. ............ 48
Figure 23. Dependence of isoprene production on the amount of dust. ......................... 49
Figure 24. Amount of isoprene produced every 24 hour through the interaction of dust
sample Sahara 4 with seawater. ....................................................................... 51
Figure 25. Amount of isoprene produced every 24 hour through the interaction of dust
sample Sahara 4 with distilled water............................................................... 52
Figure 26. Total amount of isoprene produced through the interaction of dust sample
Sahara 4 with seawater. .................................................................................... 53
Figure 27. Total amount of isoprene produced through the interaction of dust sample
Sahara 4 with distilled water............................................................................ 54
Figure 28. Amount of isoprene produced every 24 hour through the interaction of dust
sample from the Gobi Desert with seawater. .................................................. 55
Figure 29. Amount of isoprene produced every 24 hour through the interaction of dust
sample from the Gobi Desert with distilled water.......................................... 56
Figure 30. Total amount of isoprene produced through the interaction of dust sample
from the Gobi Desert with sea water............................................................... 57
Figure 31. Total amount of isoprene produced through the interaction of dust sample
from the Gobi Desert with distilled water....................................................... 58
Figure 32. Amount of methyl chloride produced through the interaction of 4 g of dust
samples Sahara 4 with seawater after certain periods................................... 64
Figure 33. Amount of methyl chloride produced through the interaction of 1 g of dust
samples Sahara 5 with seawater after certain periods................................... 65
Figure 34. Amount of methyl chloride produced every 24 hour through the interaction
of dust sample Sahara 4 with seawater. .......................................................... 66
Figure 35. Total amount of methyl chloride produced through the interaction of dust
sample Sahara 4 with sea water. ...................................................................... 67
Figure 36. Amount of methyl chloride produced every 24 hour through the interaction
of dust sample from the Gobi Desert with seawater. ..................................... 68
Figure 37. Total amount of methyl chloride produced through the interaction of dust
sample from the Gobi Desert with sea water. ................................................. 69
Figure 38. Correlation between the emissions of methyl chloride and isoprene
produced through the interaction of dust sample from the Gobi Desert with
sea water............................................................................................................. 70
Figure 39. Correlation between the emissions of methyl chloride and isoprene
produced through the interaction of dust sample from Sahara with
seawater.............................................................................................................. 71
Figure 40. Dust sample from Sahara with seawater in headspace glass vial. ................ 87
Figure 41. Mass spectrum and chromatogram of isoprene. ............................................ 88
Figure 42. Mass spectrum and chromatogram of methyl iodide. ................................... 89
Figure 43. Mass spectrum and chromatogram of methylene chloride. .......................... 90
Figure 44. Mass spectrum and chromatogram of methyl chloride................................. 91
LIST OF TABLES
Table 1.
List of the samples used in this study ................................................................... 4
Table 2.
Results of XRF analysis ....................................................................................... 11
Table 3.
Dependence of elemental concentrations on the size of the samples ............... 12
Table 4.
Results of carbon, sulfur analysis ....................................................................... 14
Table 5.
Carbon concentration of the sample from Lanzarote Island........................... 16
Table 6.
Results of ion chromatography measurements ................................................. 19
Table 7.
Dependence of anion dissolution on the bulk density ....................................... 21
Table 8.
Results of ICP-OES analysis ............................................................................... 22
Table 9.
Effect of the sample to Milli-Q water ratio on the dissolution of some metals
after 24 hours of leaching .................................................................................... 27
Table 10. pH values of the sample extracts......................................................................... 28
Table 13. Water and organic content of the samples......................................................... 92
Table 14. Dissolved organic and inorganic carbon contents of the samples.................... 93
Table 15. Effects of Fe (II) and Fe (III) additions on the production of methyl iodide .. 94
Table 16. Amount of Me2Cl2 produced through the interaction of 4 g dust sample
Sahara 4 with seawater ........................................................................................ 94
Table 17. Dependence of Me2Cl2 concentration on reaction time and medium .............. 95
Table 18. Dependence of isoprene amount on the dust amount ....................................... 95
Table 19. Amount of isoprene produced through the interaction of dust samples with
seawater................................................................................................................. 96
Table 20. Dependence of isoprene concentration on reaction time and medium............ 97
Table 21. Amount of methyl chloride produced through the interaction of dust samples
with seawater ........................................................................................................ 98
Table 22. Dependence of methyl chloride concentration on reaction time...................... 99
1
INTRODUCTION AND RESEARCH OBJECTIVES
This thesis deals with the investigation of the production and the production mechanism
of volatile organic compounds through the interaction of dust samples with seawater.
Methyl iodide was measured during the field campaign on the island of Tenerife in July,
August 2002 by the group of Williams [Williams, et al., 2007]. Higher methyl iodide mixing
ratios were found during dust events than the other times. After observing the same trend
during another field campaign on the Atlantic Ocean the same year, a question arose “Does
methyl iodide have something to do with dust?”
On the other hand, Frank Keppler and the group in Heidelberg found a new abiotic
production mechanism for halogenated organics in soils and sediments [Keppler, 2000]. In the
presence of halide ions and an electron acceptor such as Fe (III), organic matter in soils and
sediments is oxidized to give halogenated compounds.
Ocean is rich in halide ions, dust is rich in Fe (III), and both ocean and dust contain
certain amounts of organics. So since we have all the necessary ingredients, our first answer
to the question “Does methyl iodide have something to do with dust?” was “It is possible that
methyl iodide and dust have some connection.”
The main goal of my work was to test this hypothesis and to further elucidate the
process. Photolysis of Volatile Halogenated Organic Compounds (VHOCs) produces reactive
halogen species that are responsible for the ozone depletion not only in the stratosphere
[Molina and Rowland, 1974] but also in the troposphere [Barrie, et al., 1988], influencing
significantly the oxidation capacity of our atmosphere. In 1974 as Molina and Rowland found
out that the chlorofluorocarbons (CFCs) were responsible for the stratospheric ozone hole, the
intensive research on VHOC in our environment started. Besides the anthropogenic sources,
there are lots of natural sources like ocean, biomass burning, fungi, salt marshes, etc.
However, all the known sources and their emission rates are not enough to explain the
concentration measured in the atmosphere. So there are still some missing sources [Butler,
2000].
Soil, which stores 1500-2200 Gt organic carbon as humus, can be one of the missing
sources. In 2000 Frank Keppler and others here in our institute found an abiotic mechanism
for the production of VHOC in soils and sediments. Halide ions can be alkylated during the
oxidation of organic matter by an electron acceptor such as Fe (III) [Keppler, et al., 2000].
1
About 33% of the world land area is arid and semi-arid regions [Usher, et al., 2003].
Since the soil in these regions contain very low amount of organics, it is unlikely that the
VHOC could be produced there. However, enormous amounts (1-3 billion tons) of desert dust
fly up into the sky each year and travel vast distances over the oceans from one continent to
the other [Husar, 2004]. Mineral dust plays different roles in our environment. Rain forest in
Amazonia would not have existed without Saharan dust aerosols that bring nutrient there
[Swap, et al., 1992]. It is the main source of major nutrient element iron, in the ocean, which
enables the growth of phytoplankton in oligotrophic water [Jickells, et al., 2005]. Dust
aerosols affect climate through both direct and indirect radiative forcing [Li, et al., 1996;
Tegen, et al., 1996]. Viable fungi and bacteria are transported with soil dust
interhemispherically [Prospero, et al., 2005]. Respirable mineral dust aerosols can cause
health problems especially when associated with bacteria and microbes [Usher, et al., 2003].
Pollutants are adsorbed onto the mineral dust particles and transported long distances [Erel, et
al., 2006]. Different kinds of heterogeneous and multi-phase reactions take place on the
surface of mineral aerosols [Usher, et al., 2003].
The production of VHOC from the reaction of mineral dust particles with atmospheric
components in the marine boundary layer and with seawater is not yet investigated and is the
main research goal of this PhD project. With this work we try to answer several questions:
-
How significant is the process of VHOC formation from the multiphase reaction on
mineral dust in marine atmosphere and in the ocean?
-
Which VHOC and which other Volatile Organic Compounds (VOCs) are produced?
-
Through which mechanism are those compounds formed?
-
Which constituents play the most important role in the formation of those compounds?
For this, dust-seawater addition experiments in headspace glass vials were conducted in
the laboratory, following a purge-and-trap GC-MS analysis of the headspace gas.
Methyl iodide, methylene chloride and isoprene were produced through the interaction
of dust samples from the Sahara and Gobi Desert with both seawater and distilled water,
methyl chloride was produced through the interaction of dust samples with only seawater. It is
proved that the natural dust samples are important contributors of volatile organic compounds
to the atmosphere and further studies in this field are urged.
2
2
EXPERIMENTAL SECTION
2.1
Samples
Total of nine fine Saharan dust samples were collected directly from the ground in
Tamanrasset (22° 47′ N, 5° 31′ E, 1362 m) in the Hoggar region of southern Algeria. The sites
were located within a radius of 10-20 km from Tamanrasset. The samples were taken from
places without anthropogenic influence. They are denoted as Sahara 1 - Sahara 9 or S1 - S9 in
the following chapters. Two samples from the Gobi Desert were also collected from the
ground in Dalanzadgad city (43° 33′ N, 104° 25′ E) in south Gobi province in Mongolia. One
site was a sand dune located 85 km away from the city. This sample is denoted as Mongolia 1
or M1. The other site was directly in the city center. This sample was wind blown and was
accumulated at the bottom of the fences which typically every family has around its houses.
The anthropogenic influence could not be excluded since the sample was collected in the city.
This sample is denoted as Mongolia 2 or M2. To compare dust samples from the source
region with that which have already gone aeolian transport, further samples were taken from
Cape Verde Island and Lanzarote Island (Canary Islands). The sample from Lanzarote Island
is one of the eight composite samples collected in the northern part of the Island (29° 13′ N,
13° 27′ E) [Maciejczyk, 2005]. It is denoted as Lanzarote 6 or L6. The sample from Cape
Verde Island (16° 45′ N, 22° 57′ W) was generously provided by Professor Gaudichet from
LISA, UMR-CNRS 7583 University of Paris. The sample was collected by bulk filtration on
0.4 µm pore-size Nuclepore filter with an average flow rate of 0.8 m3/h [Caquineau, et al.,
1998]. This sample is denoted as Cape Verde or CV. All samples were sieved and only fine
particles with diameter smaller than 63 µm were used for the experiments. Also some samples
were sterilized in an autoclave under 0.1 MPa pressure at 120 0C for 1 hour to exclude the
microbiological contribution to the production of VOCs. The names of all samples used in
this study and their denotations are listed in Table 1.
3
Table 1.
List of the samples used in this study
No.
2.2
Denotation
Place
1
Guelta Oued Imeleoulaouene
Sahara 1
S1
2
Oued Imeleoulaouene
Sahara 2
S2
3
Oued Tehéggart
Sahara 3
S3
4
Oued Tamanrasset
Sahara 4
S4
5
A fila Oued Imeleoulaouene
Sahara 5
S5
6
Oued Adaouda
Sahara 6
S6
7
Oued Sêrsouf
Sahara 7
S7
8
Oued Ézernene
Sahara 8
S8
9
Oued Tindé
Sahara 9
S9
10
Sand dune in Gobi
Mongolia 1
M1
11
Dalanzadgad city
Mongolia 2
M2
12
Lanzarote Island
Lanzarote 6
L6
13
Cape Verde Island
Cape Verde
CV
Chemicals
Isoprene (99 %, d = 0.68 g/ml, Aldrich)
VOC-Mix 21 (200 mg/l, Ehrenstorfer)
Methyl iodide (10 ng/µl, Ehrenstorfer)
Iron (III) sulfate (>76 % Fluka)
Iron (II) sulfate (>76 % Fluka)
Standard for calcium carbonate (12 % C, LECO Corporation)
Standard 501-005 for Sulfur in coal (1.00±0.02 %, LECO Corporation)
Hydrochloric acid (25 %, d = 1.12 g/ml, AppliChem)
Sodium carbonate (99.9 %, Merck)
Sodium bicarbonate (99.5 %, Merck)
Potassium hydrogen phthalate (C8H5KO4) (99.5 %, Merck)
Standard of several mixed cations (1000 mg/l, CertiPUR Merck)
Standard of silicon (1000 mg/l, CertiPUR Merck)
Nitric acid (65 %, d = 1.39 g/ml, Fluka)
Standards of chloride, bromide, nitrate, nitrite, sulfate and phosphate (1000 mg/l, CertiPUR
Merck)
4
2.3
Analytical methods
2.3.1
X-ray fluorescence (XRF)
An in-house designed Energy-dispersive Miniprobe Multielement Analyzer (EMMA)
XRF was used to determine the total of 24 major (Al, Si, S, Cl, K, Ca, Ti and Fe) and trace
(Cr, Mn, Ni, Cu, Zn, Ga, As, Se, Br, Rb, Sr, Y, Zr, Nb, Pb and Th) elements in dust samples.
The description and details about this sensitive, rapid and non-destructive method is described
elsewhere [Cheburkin and Shotyk, 1996].
2.3.2
Carbon, Sulfur analysis
Carbon and sulfur contents of the dry samples were determined by means of C/S
Analyzer (LECO SC-144DR). About 200 mg of dust sample was burned at 13500C under
oxygen atmosphere and the products of this combustion, carbon dioxide and sulfur dioxide
were measured with infra red detector. Calcium carbonate was used as carbon standard and
sulfur containing coal, 501-005, as sulfur standard.
2.3.3
Water content and total organic content
The water content of the samples was determined using a standard method DIN 18121
Part 1. Fresh dust samples were weighed in crucibles whose weights were constant and were
heated at 1050C until their weights got constant. The mass difference of fresh and dry samples
gives the water content of dust samples. After determining the water content, the same
samples were heated at 5500C for an hour, and were weighed. The procedure was repeated
until their weight got constant. The mass difference of dry sample and the sample heated at
5500C gives the amount of total organic matter in the sample.
5
2.3.4
TOC
Dissolved total, organic and inorganic carbons were determined by means of Shimadzu-
5000 Total Organic Carbon (TOC) analyzer. Because dust samples contain very small amount
of organic carbon, in order to get solutions with concentrations above the detection limit of
the instrument, dust samples were mixed with Milli-Q water in ratio of 1:3 instead of usual
1:10 ratio. After shaking 10 g of dust samples in 30 ml of Milli-Q water for 24 hours, the
samples were centrifuged and filtered through 0.45 µm filter. Inorganic carbon standard was
prepared by dissolving 0.350 g sodium bicarbonate and 0.441 g of sodium carbonate in 100
ml of Milli-Q water. The concentration of resulting mother solution was 1000 mg/l. Total
carbon standard was prepared by dissolving 0.2125 g of potassium hydrogen phthalate
(C8H5KO4) in 100 ml of Milli-Q water. The concentration of resulting mother solution was
also 1000 mg/l. For total carbon determination, the standards with concentrations of 25, 50,
75 and 100 mg/l, and for inorganic carbon determination, the standards with concentrations of
10, 20, 50 and 100 mg/l were prepared. To determine total dissolved carbon, the samples were
converted to carbon dioxide by burning them at 6800C with oxygen. After drying, and
absorbing the halogens, the carbon dioxide was analyzed with non-dispersive infra red
(NDIR) detector. To determine total dissolved inorganic carbon, the samples were acidified
with ortho-phosphoric acid and the resulting carbon dioxide, which is only of inorganic
origin, was also detected with NDIR detector. And the total organic carbon was calculated
from the difference of total and inorganic carbon.
2.3.5
Anions
Fluoride, chloride, nitrite, bromide, nitrate, phosphate and sulfate ions were determined
by means of ion chromatography DIONEX 120. The sample extracts were prepared by
shaking dust samples in Milli-Q water in ratio of 1:3 for 24 hours and filtering through 0.45
µm cellulose filter. A pre-column AG14A and a separating column AS14A 4 x 250 mm were
used. The flow rate was 1 ml/min. The eluent solution was a mixture of 8 mmol/l of Na2CO3
and 1 mmol/l of NaHCO3. Two types of detectors, UV-208 nm and conductivity meter, were
used to determine the anions. For the determination of fluoride, nitrite and bromide, standards
with concentrations of 0.5, 1, 3 and 5 mg/l, for the determination of chloride and phosphate,
standards with concentrations of 5, 10, 30 and 50 mg/l, and for the determination of nitrate
and sulphate, standards with concentrations of 10, 30, 60 and 100 mg/l, were prepared.
6
2.3.6
Cations
Al, Ca, Co, Cr, Cu, Fe, K, Mg, Mn, Na, Ni, Pb, Si and Sr were determined by means of
Inductively Coupled Plasma-Optical Emission Spectrometer (ICP-OES). The sample extracts
were prepared by shaking the dust samples in Milli-Q water in various ratios for certain times.
The main purposes were to determine the solubility of different metals from dust samples, and
to see whether a bulk density had a substantial effect on the solubility. I started with the ratio
of 1:10, and after 5 minutes the samples were centrifuged, and 10 ml of extracts were taken
from the supernatants. The extracts were filtered through 0.45 µm cellulose filter, and were
acidified with 50 µl of concentrated nitric acid for the cation analysis. The extracts were
filtered for the analysis accuracy only. Actually, there were no sample losses due to this
process, so the calculations were made according to that. 10 ml of Milli-Q water was added to
the original samples to keep the 1:10 ratio, and the samples were shaken further for 15 more
minutes, which means 20 minutes after the start. Again the samples were centrifuged and 10
ml of extracts were taken out, and were acidified for ICP-OES analysis. 10 ml of Milli-Q
water was added to the original sample, and was shaken further for 20 more minutes. Total of
8 extracts were prepared for each sample after 5 min, 20 min, 40 min, 1 hour, 2 hours, 5
hours, 17 hours and 24 hours, respectively. Parallel extracts were obtained by shaking the
samples directly for 24 hours in ratios 1:3 and 1:10 in order to compare the effect of bulk
density on the solubility of metals from the dust samples. For the quantification of the cations,
standards solution CertiPUR (1000 mg/l), which contains several cations, and the standard
solution CertiPUR Silicon (1000 mg/l) were used. As a control, an international standard
NIST 1643E was used.
2.3.7
pH
The pH of the sample extracts was determined with pH-meter Mettler Toledo 320. The
pH-meter was calibrated with the standard solution of pH 7.0.
7
2.4
Determination of volatile organic compounds
For the determination of volatile organic compounds in the reaction vials, a headspace
technique with either GC-ECD or a purge-and-trap GC-MS system was used. The headspace
technique is an effective method of sample preparation and injection to the GC system. The
volatile and semi-volatile organics in headspace gas of tightly closed vials are analyzed by
GC after an establishment of equilibrium in the headspace. There are two types of headspace
technique: static and dynamic. In static headspace technique, after the establishment of
equilibrium, certain amount of headspace gas is taken by syringe, and is injected to the GC.
This method was used in combination with GC-ECD system. In dynamic headspace
technique, which is mostly called a purge-and-trap, the headspace gas is continuously purged
and concentrated on the cooled trap, and the pre-concentrated sample is then transferred to the
GC by thermal desorption. Because the sample is pre-concentrated before it comes to the GC
column, a purge-and-trap is a widely used, effective method for the identification and
quantification of ultra trace amounts of compounds.
2.4.1
GC-ECD
The volatile halogenated organic compounds were analyzed by means of Gas
Chromatography with Electron Capture Detector (GC-ECD). The GC-ECD used for this
study composed of Auto sampler Combi PAL with agitator and heatable syringe, Gas
Chromatograph Fison HRGC 8265 with split/splitless injector Carlo Erba SSL71, a separating
column SGE BP 624 with film thickness of 3 µm and an electron capture detector Carlo Erba
HT 25. Electron capture detector has a radioactive beta particle (electron) emitter. It is mostly
a metal foil containing 10 milli Curie Ni-63. The emitted electrons are attracted to the
positively charged anode, generating a steady current. When the sample is transferred to the
detector by means of carrier gas, the electron absorbing compounds capture electrons from the
detector, thus reducing the current. The amounts of the compounds are proportional to the
reduction of the current signal. The ECD is very sensitive to the electron absorbing
compounds, for example halogenated volatile organics.
The samples were shaken for certain hours at 20-500C in the agitator and certain
amounts of headspace gas were then injected to the Gas Chromatograph. The various
compounds were then identified and quantified by their retention times. Because we used only
retention times for identification, it was sometimes difficult to differ among the compounds
8
with same or close retention times. So we used the GC-ECD analysis results qualitatively to
choose the most productive samples for further analysis with a purge-and-trap GC-MS.
2.4.2
A purge-and-trap GC-MS
A purge-and-trap unit, Tekmar LSC 2000, was connected upstream to the Gas
Chromatograph Varian Star 3400 cx and the ion trap Mass Spectrometer Varian Saturn 2000.
The volatile organic compounds in the headspace glass vials were purged with He (flow rate
30 ml/min) for 4 minutes, and were pre-concentrated on a trap filled with Tenax TA 60/80
which was cooled at -90 0C with liquid nitrogen. A water trap filled with magnesium
perchlorate was connected upstream to the pre concentration trap. After short dry purge step
(2 s), desorption of volatile organics from the pre concentration trap was done at 180 0C for 4
minutes. During the desorption step, the capillary column of the gas chromatograph was
cooled at -90 0C with liquid nitrogen to reconcentrate the volatile organics at the beginning of
the column. Gas chromatographic separation was carried out on a column SGE BP624 (I.D.
0.53 mm, O.D. 0.68 mm, length 25 m) using the temperature program as shown in Figure 1.
o
Temperature ( C)
200
100
0
0
10
20
30
40
50
Time (min)
Figure 1. Temperature program of GC oven.
Initial oven temperature was -900C, and then it was heat up at the rate of 40 0C/min until
0 0C and was held for 1 min. After that, the oven was heat up at the rate of 5 0C/min until 150
0
C and was held for 5 min. Finally, it was further heat up at the rate of 30 0C/min until 210 0C
and was held for 15 min. Mass spectrometric detection was done in scan mode over a range of
45-280 amu. Because GC-MS gives both the retention time and the mass for identification, it
was easier and more suitable to quantify the compounds compared to GC-ECD.
9
3
3.1
CHARACTERIZATION OF SAMPLES
Results and Discussion of XRF analysis
The detailed results of XRF analysis are shown in Table 2. There are some elements
whose concentrations changed related to an aeolian transport of dust samples. Clearly seen in
Figures 2 and 3 are the increased concentrations of calcium and bromine in the samples from
Cape Verde Island (CV) and Lanzarote Island (L6) that have undergone aeolian transport.
10
8
Ca (%)
6
4
2
0
S1 S2 S3 S3 S4 S5 S6 S7 S8 S9 CV L6 M2 M1
Figure 2. Calcium concentration in the samples.
25
Br (ppm)
20
15
10
5
0
S1 S2 S3 S3 S4 S5 S6 S7 S8 S9 CV L6 M2 M1
Figure 3. Bromine concentration in the samples.
10
Table 2.
Results of XRF analysis
Sample
K, %
Rb, ppm
Ca, %
Sr, ppm
Al, %
Ga, ppm
Si, %
LLD
0,001
1
0,001
1
0,5
1,5
0,3
2
S1
S2
S3
S3
S4
S5
S6
S7
S8
S9
CV
L6
M1
M2
2,25
2,00
2,31
2,24
2,22
2,07
2,37
2,16
2,19
2,08
1,94
3,05
3,27
2,20
133,44
118,03
106,59
116,51
102,72
135,88
154,80
132,05
107,30
107,90
59,47
70,32
105,52
69,30
1,98
2,00
1,49
1,46
1,90
2,02
2,19
2,00
1,95
1,64
4,03
9,05
0,73
3,07
268,02
299,08
262,16
293,61
290,53
261,62
268,42
286,04
256,54
263,62
336,58
276,74
183,71
294,95
8,49
7,88
8,35
6,61
8,31
8,27
8,80
7,74
8,95
6,85
8,72
10,21
6,92
7,69
18,74
17,92
15,27
18,19
16,11
18,19
24,86
19,17
17,65
18,93
19,63
14,68
7,65
15,09
30,87
26,61
31,00
30,93
28,54
27,66
28,64
30,34
27,80
31,00
23,57
27,27
31,00
28,48
22,05
17,93
18,45
19,27
14,48
21,62
19,97
21,36
16,92
20,31
7,98
12,48
12,26
27,10
Sample
As, ppm
Cl, ppm
Br, ppm
Cu, ppm
Zn, ppm
Y, ppm
Th, ppm
S, ppm
300
200
LLD
1,5
S1
S2
S3
S3
S4
S5
S6
S7
S8
S9
CV
L6
M1
M2
0,00
5,96
0,00
3,04
2,07
0,00
3,59
1,92
2,57
0,00
4,58
5,66
2,95
9,95
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
1660,61
0,00
0,00
497,79
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
3445,76
0,00
295,81
0,00
Sample
Ti, %
Zr, ppm
LLD
0,0005
2,5
S1
S2
S3
S3
S4
S5
S6
S7
S8
S9
CV
L6
M1
M2
0,74
0,99
0,61
0,62
0,73
1,12
0,72
0,78
0,78
0,70
2,41
0,96
0,13
0,53
367,18
454,96
406,44
368,50
333,53
322,37
388,55
403,80
330,71
413,19
338,19
256,47
43,31
598,03
1
Pb, ppm
2
1,5
2
2,5
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
0,00
24,28
8,99
0,00
1,11
26,21
22,71
16,13
16,34
22,64
31,39
32,09
32,40
28,71
19,42
47,54
35,99
4,19
28,66
104,97
116,71
66,19
74,32
90,07
99,49
102,50
99,98
102,85
73,97
96,40
68,89
7,99
72,99
34,65
32,92
23,58
24,24
22,38
30,94
33,99
32,75
30,05
29,19
26,16
23,48
8,16
26,86
7,76
10,96
5,86
6,34
8,14
9,20
3,95
10,35
3,55
5,70
3,88
5,34
0,00
4,33
Nb, ppm
Cr, ppm
Mn, ppm
Fe, %
Ni, ppm
2,5
1,5
1,5
0,001
3
24,92
46,63
21,33
20,71
46,26
28,33
23,52
28,21
28,44
22,96
83,73
40,15
2,54
12,67
115,61
134,29
79,44
74,78
105,11
147,28
99,75
104,08
112,14
98,72
275,38
178,32
9,69
89,05
956,45
1119,13
805,37
772,37
944,35
1140,35
974,04
937,55
953,69
804,00
1544,50
679,52
207,24
916,45
4,38
5,67
2,90
3,60
4,73
4,25
4,46
4,45
4,41
3,95
7,62
4,52
0,47
3,36
32,38
39,90
25,41
31,93
47,62
21,01
26,88
25,67
36,01
30,63
55,19
117,06
4,35
23,93
11
As the dust is transported farther away from the source region, the overall composition
tends to become enriched with clays or micas in whose mineral structures insoluble calcium is
bound [DeBell, et al., 2004; Usher, et al., 2003]. This explains why the samples from Cape
Verde and Lanzarote Islands have higher calcium concentrations than the samples from
Algeria.
For bromine the lowest limit of detection was 1 ppm. It was not detected in all samples
collected from the source region but in samples that have undergone aeolian transport. It
could be explained by the adsorption of ocean salt aerosols on the dust particles during the
transport over the ocean.
In order to see the dependence of elemental concentrations on the size of the samples,
three different sized samples Sahara 8 were analyzed. Calcium, lead and arsenic
concentrations increased as the size of the samples decreased, and the concentrations of
potassium, copper, chromium and manganese decreased with decreasing sample size (Table
3.). In case of calcium, it again demonstrates that calcium, mostly in form of insoluble calcite
in mineral dust, is enriched in the fine fraction.
Table 3.
Dependence of elemental concentrations on the size of the samples
Sample (S8)
Ca, %
Pb, ppm
As, ppm
K, %
Cu, ppm
Cr, ppm
2,01
29,0
6,4
1,98
37,2
105,3
953,8
0.063 mm < d < 0.2 mm
1,91
25,4
3,6
2,47
41,3
117,5
1060,0
d > 0.2 mm
1,69
17,8
2,1
3,71
42,5
144,0
1337,1
d < 0.063 mm
Mn, ppm
Trace metals were associated to iron as it is shown for manganese in Figure 4, as it was
the case in the work of Guieu et al. [Guieu and Thomas, 1996]. There was strong linear
correlation (R2 = 0.91). The correlation was also good for chromium, niobium, nickel, titan,
strontium, copper and zinc (R2 = 0.71-0.84).
All the other measured elements showed no correlation with iron content, which for
lead was also the same for the samples used in the study of Guieu et al. [Guieu and Thomas,
1996].
12
1600
Mn, ppm
1200
800
2
R = 0.91
400
0
0
1
2
3
4
5
6
7
8
Fe, %
Figure 4. Concentration of manganese versus iron content in dust samples.
13
3.2
Results and Discussion of Carbon, Sulfur analysis
The carbon, sulfur analysis with C/S analyzer was not suitable for the dust samples
because the dust samples contained very small amounts of carbon and sulfur. The carbon
standard used for the calibration was pure calcium carbonate which contained 12 % of carbon
and the sulfur content of the coal used for the calibration was 1,00 ± 0,02 %. The results of
the carbon, sulfur analysis of all the samples are shown in Table 4.
Table 4.
Results of carbon, sulfur analysis
Sample
Sulfur, %
Carbon, %
S1
0,0061
0,2297
S2
0,0103
0,3258
S 2 > 0.063
0,0041
0,2609
S 2 > 0.125
0,0134
0,6024
S3
0,0094
0,1806
S4
0,0060
0,1948
S5
0,0095
0,1837
S6
0,0149
0,2740
S 7 < 0.02
0,0099
0,5476
S7
0,0069
0,1420
S 7 > 0,063
0,0079
0,1465
S 7 > 0,125
0,0022
0,1181
S 7 > 0.2
0,0097
0,1196
S8
0,0036
0,2521
S 8 > 0,063
0,0127
0,1613
S 8 > 0,125
0,0083
0,1154
S 8 > 0,2
0,0057
0,3001
S 8 > 0.315
0,0062
0,5796
S9
0,0119
0,1327
CV
0,0887
0,6751
CV > 0.063
0,0826
0,9253
CV > 0.125
0,0575
0,9457
M1
0,0029
0,0433
M2
0,0159
1,6838
L 6 < 0.02
0,0147
2,1359
L6
0,0094
2,9126
14
The dust samples except the dust sample from Lanzarote Island contained not even 1 %
of carbon, and the sulfur content of all the samples except for the dust sample from Cape
Verde Island was below 0,02 %. The measurements with more than 200 mg of dust samples
failed because of incomplete combustion of the samples. The sulfur content measured with
XRF was almost twice as much as that measured with C/S Analyzer. The lower limit of
detection of XRF instrument for sulfur was 300 ppm. We consider the measurement of XRF
instrument more reliable, because the measured values of C/S Analyzer were far more
different from the calibration standard values. In both cases, it is shown that the dust sample
collected at the Cape Verde Island contains sea salt aerosol components.
The intended purpose of performing carbon, sulfur analysis with C/S Analyzer was to
determine the total, inorganic and organic carbon content of the dust samples in dry samples
and then to compare the results with the total dissolved inorganic and organic carbon content
of the samples in their leach solutions in order to find out whether or how the carbon contents
of the samples affect the formation of volatile organic compounds, when the dust samples
interact with ocean water. Unfortunately, the attempt failed because the results of the total
carbon content determination turned out to be unreliable. They all fell in the range outside the
calibration value (see Table 4), and also the inorganic carbon measurement by means of
“Carbonate bomb” instrument developed by Müller and Gastner [Müller and Gastner, 1971]
was not possible due to insufficient inorganic carbon content of the dust samples.
Carbonate bomb is a closed glass with a pressure measuring instrument. It measures the
pressure of the carbon dioxide formed inside the bomb, after the sample reacts with 6 N
hydrogen chloride solution. As a standard, calcium carbonate is used. Comparison of the
measured pressure with the pressure of carbon dioxide formed from certain amount of
standard gives the quantitative results of inorganic carbon content of the sample. The
difference between the total carbon and the total inorganic carbon contents gives the amount
of total organic carbon of the sample.
15
The only sample for which the intended measurements were successful was the dust
sample collected on the Island of Lanzarote. The results of the total, inorganic and organic
carbon measurement of this sample are shown in Table 5 in comparison to the measurement
results of Maciejczyk [Maciejczyk, 2005]. The difference in results can be explained by the
heterogeneity of the sample and also the size of the sample used for the measurement. The
sample in the size range of (0.020 < d < 0.063) mm was used in this study. Maciejczyk used
the samples in the size range of d < 0.125 mm in her work.
Table 5.
Carbon concentration of the sample from Lanzarote Island
Sample
Total carbon, %
Total inorganic
carbon, %
Total organic
carbon, %
Authors
L6
2,9126
2,20
0,7126
this study
L6
2,9905
1,92
1,0705
Maciejczyk
Because the measurement of total organic content of the samples failed with this
method, another method was employed for this purpose. Those results are discussed in the
next section.
16
3.3
Results and Discussion of water content and total organic content analysis
The water contents of the dust samples were about 1 % and the total organic contents
were about 3 - 4 %. The results of the analysis are shown in Figure 5.
8.00
Water content, %
Water and organic content, %
Organic content, %
6.00
4.00
2.00
M
2
M
1
L6
V
C
S9
S8
>
S8
0.
S8 06
3
>
0.
12
S8 5
S8 > 0
> .2
0.
31
5
S7
S6
S5
S4
0
S4 .0
63
>
0.
1
S4 25
>
0.
2
S4
>
S3
S2
0.
S2 06
3
>
0.
12
5
S2
>
S1
0.00
Figure 5. Water and organic content of the samples.
The water and organic content of the samples from Cape Verde and Lanzarote Islands
were more than that of the samples collected directly at the source regions. Also the results
showed that the dust samples with smallest size contained higher organics than the samples
with bigger size. This was seen in samples Sahara 4 and Sahara 8. The structure analysis of
organics in the samples would have shown whether the organics originated from the source
region, or they were adsorbed onto the dust particles during the transport over long distances.
The characterization of the organic compounds in the dust samples was not carried out during
this work. But the results of this and the other following analysis approved that the dust
samples contain certain sufficient amounts of organics responsible for the formation of some
volatile organic compounds.
17
3.4
Results and Discussion of TOC analysis
The results of the total dissolved organic and inorganic carbon analysis are shown in
Figure 6.
DOC, ppm
IC, ppm
200.00
150.00
100.00
50.00
L6
V
0.
02
0
C
<
L6
2
M
9
S
S
8
>
0.
31
5
0,
2
>
8
0,
12
5
>
8
S
8
S
2
S
S
8
0,
06
3
S
>
7
S
6
S
5
S
3
4
S
S
0.
12
5
>
S
S
2
>
S
2
0.
06
3
0.00
1
Dissolved organic and inorganic carbon, ppm
250.00
Figure 6. Dissolved organic and inorganic carbon contents of the samples.
It is seen from the figure that the sample from Lanzarote Island contains the highest
amount of dissolved inorganic carbon, which supports the result of the carbon analysis of the
dry dust sample which was done by means of carbonate bomb instrument. The results of the
sample Sahara 8 also correspond with the results of the total organic content analysis. As
mentioned previously, the leach solutions were prepared by mixing 10 g of dust sample with
30 ml of Milli-Q water for 24 hours, because dust samples contain lower amounts of organics
in comparison to soil samples. Normally, the ratio of 1:10 is used for the analysis of soil
sample leach solution. So the question was whether the bulk density had an effect on the
solubility of the organics, as it did on the solubility of iron (see Section 3.7). To find out the
answer to this question, two leach solutions were prepared from the Lanzarote sample, one by
dissolving 2 g of sample in 20 ml of Milli-Q water, and the other by dissolving 10 g of sample
in 30 ml of Milli-Q water like all the other samples. The last two columns in Figure 6
represent those two samples. As it is clearly seen from the Figure, the change in the bulk
density showed twofold change in dissolved organic and inorganic carbon content. So when
the dust particles enter the ocean during dust storms, the dissolved organic and inorganic
carbon contents would be higher than these measurement results.
18
3.5
Results and Discussion of Anion analysis
The results of ion chromatography analysis are shown in Table 6.
Table 6.
Results of ion chromatography measurements
Samples
F-
Cl-
NO2-
(mg / kg)
(mg / kg)
(mg / kg)
Sahara 1
130,75
5,29
Br(mg / kg)
NO3-
SO42-
(mg / kg)
(mg / kg)
70,25
35,51
Sahara 2
0,41
71,10
6,71
0,15
166,67
49,16
Sahara 2 > 0.063
0,51
56,13
5,41
0,08
127,92
36,27
Sahara 2 > 0.125
0,30
40,20
8,88
0,03
148,90
36,19
17,35
2,70
0,03
64,88
12,69
0,04
143,90
48,14
Sahara 3
Sahara 4
0,27
58,38
4,15
Sahara 5
0,54
105,98
6,14
41,49
21,73
Sahara 6
0,83
44,81
8,07
83,98
138,61
Sahara 7
0,31
31,65
58,52
29,04
173,83
50,17
124,72
33,35
100,55
29,99
88,24
28,03
101,07
32,94
14,96
11,06
360,93
50,41
56,02
30,99
164,70
38,09
0,05
24,78
13,55
3,74
189,79
1795,45
Sahara 8
Sahara 8 > 0.063
0,35
Sahara 8 > 0.125
70,80
7,78
40,62
4,32
38,16
5,78
Sahara 8 > 0.2
0,31
41,18
2,57
Sahara 8 > 0.315
0,21
126,50
13,29
17,22
3,70
Sahara 9
Gobi 1
1,22
Gobi 2
Lanzarote
Cape Verde
1,14
62,29
9,41
1627,23
0,03
0,05
0,20
As it is seen from the table, the sample from Cape Verde Island contains the highest
amount of chloride and sulfates, as it was the case of XRF analysis (see Table 5). Bromide
could be detected in samples from Lanzarote and Cape Verde Islands and in addition in
samples Sahara 2, 3, 4 and 8, which were not detected by XRF analysis (see Figure 3). The
samples Sahara 2, 4 and 8 and the samples from the Gobi Desert Mongolia 2 and Cape Verde
Island showed higher amounts of nitrate than the other samples. The sample Sahara 6 had
higher sulfate content in comparison to the other samples.
19
The Figure 7 was drawn to see how the dependence of the anion contents on the sample
size was. For the samples Sahara 2 and 8, the samples smaller than 0.063 mm diameter
contained higher amounts of chloride and sulfate compared to the same samples with bigger
size, which also was the case for nitrate, dissolved inorganic carbon, calcium and total organic
content of the samples.
80,00
Chloride, ppm
Sulfate, ppm
Concentration, ppm
60,00
40,00
20,00
0,00
Sahara 2
< 0.063
Sahara 2
> 0.063
Sahara 2
> 0.125
Sahara 8
< 0.063
Sahara 8
> 0.063
Sahara 8
> 0.125
Figure 7. Dependence of anion content on the size of the samples.
If we compare the results of ion chromatography with the results of the XRF analysis
for the sample from Cape Verde Island, 47 % of chlorine and 36 % of sulfur were detected as
chloride and sulfate. But could all the chloride and sulfate be transferred to the solution when
we mixed the samples with Milli-Q water in the ratio of 1:3? To see whether the bulk density
had an effect on the dissolution of anions, another set of sample extracts were prepared by
mixing the samples with Milli-Q water in ratio of 1:10 and the resulting solutions were
analyzed by ion chromatography. The results of this analysis are shown in Table 7 in
comparison to the previous analysis results.
20
Table 7.
Samples
Sahara 2
Sahara 8
Dependence of anion dissolution on the bulk density
Mixing ratio
Cl-, mg/kg
Br-, mg/kg
NO3-, mg/kg SO42-, mg/kg
1:10
370,954
0,432
183,223
103,822
1:3
71,104
0,146
166,674
49,164
1:10
350,599
0,082
183,057
95,025
1:3
70,803
0,028
173,826
50,168
According to the results from the Table 7, it is seen that if we increase the sample to
water ratio from 1:3 to 1:10, about 5 times more chloride, 4 times more bromide and twice as
much as sulfate would dissolve, whereas the amount of nitrate stays almost the same.
If we consider that it implies for all the samples, the ion chromatography analysis of the
sample from Cape Verde Island would measure 8136.15 ppm of chloride, 14.96 ppm of
bromide and 3590.90 ppm of sulfate, which would be 236 % of chlorine, 62 % of bromine
and 72 % of sulfur measured with XRF. In case of chloride, it is just an imaginary number,
because at some point, its concentration would reach the maximum, when all the chloride
would dissolve if all the chlorine were in form of chloride.
It would have been more convincing if we could show the same comparison results for
the samples Sahara 2 and 8, as we did for the sample from Cape Verde Island. Unfortunately,
the chlorine and sulfur contents were below the lower detection limit of the instrument in all
other samples. But the results we obtained show that it is certainly better to use 1:10 ratio to
prepare the dust sample extracts for the ion chromatography analysis.
21
3.6
Results and Discussion of Cation analysis
The raw results of ICP-OES analysis for fourteen cations are shown in Table 8.
Table 8.
Results of ICP-OES analysis
Time
Ca2+, mg/l
K+, mg/l
Na+, mg/l
5 min
7,3124
7,8432
8,1770
8,4915
9,0943
9,2848
10,6710
11,1460
3,0585
3,3593
3,4418
3,8925
3,9334
3,4534
3,3638
3,4977
6,4080
5,6194
5,2113
5,8423
4,0084
4,3049
4,3663
3,2434
Si4+, mg/l
Fe2+, mg/l
Al3+, mg/l Mg2+, mg/l
S2
20 min
40 min
1 hour
2 hours
5 hours
17 hours
24 hours
2,6465
3,9590
4,2563
8,5752
9,6843
6,8405
5,8781
7,0289
0,38124
0,46268
0,34424
1,48550
1,67040
0,65964
0,31567
0,55231
0,58368
0,77903
0,59133
2,66800
3,02140
1,22290
0,54628
0,98527
0,81812
0,93240
0,96626
1,36940
1,38690
1,17210
1,33860
1,55450
S8
17 hours
8,5359
9,0667
8,6578
8,3129
8,5528
8,9891
12,1050
3,0967
3,5145
3,2738
3,5152
3,4109
3,2132
3,8444
4,3615
4,4694
4,1366
3,5678
3,1706
2,5082
2,9285
1,4428
3,7525
3,1787
5,2453
5,1169
4,5633
11,3750
0,11602
0,59031
0,22293
0,74352
0,59962
0,25995
1,98590
0,19051
0,99165
0,40727
1,38240
1,09530
0,50326
3,76760
0,68755
0,80852
0,70625
0,82366
0,78470
0,75600
1,57850
24 hours
11,7650
4,7025
2,9399
16,2480
3,33600
6,06230
1,68160
Co2+, mg/l
Cr3+, mg/l
Cu2+, mg/l Mn2+, mg/l
Ni2+, mg/l
Pb2+, mg/l
Sr2+, mg/l
0,002321
0,002323
0,001875
0,003744
0,003768
0,002048
0,001389
0,001515
0,002405
0,002675
0,002460
0,002013
0,001621
0,001735
0,003033
0,002527
S2
0,010121
0,010150
0,009126
0,025458
0,027674
0,011302
0,021563
0,044190
0,000556
0,000837
0,000540
0,002084
0,002135
0,000844
0,001133
0,001050
0,001196
0,001431
0,001411
0,001397
0,002185
0,001411
0,001695
0,003011
0,047461
0,052495
0,055096
0,059341
0,062574
0,062714
0,071414
0,074316
0,004176
0,004271
0,003602
0,003515
0,002910
0,003331
0,003809
0,003718
S8
0,006082
0,015103
0,007258
0,015223
0,011847
0,008753
0,047224
0,046729
0,000558
0,000768
0,001018
0,000848
0,000583
0,000513
0,002215
0,002840
0,001302
0,001858
0,003829
0,002190
0,001905
0,002892
0,002487
0,003962
0,054635
0,059059
0,056342
0,056362
0,056434
0,058503
0,080783
0,080174
5 min
20 min
40 min
1 hour
2 hours
5 hours
Time
5 min
20 min
40 min
1 hour
2 hours
5 hours
17 hours
24 hours
5 min
20 min
40 min
1 hour
2 hours
5 hours
17 hours
24 hours
0,001010
0,001281
0,001065
0,001533
0,001810
0,001008
0,001074
0,001326
0,000639
0,001596
0,000890
0,000899
0,000764
0,000492
0,002853
0,002492
0,000810
0,001303
0,000732
0,001545
0,001171
0,001108
0,003793
0,004652
22
From those fourteen cations, five were chosen to analyze the solubility of them in MilliQ water in dependence of the time and bulk density. Especially, the dissolution of iron from
Saharan dust aerosol in seawater is very important to determine, since it was shown that iron
was essential for the abiotic production of volatile halogenated organics in soils and
sediments. In addition, the results of manganese, potassium, calcium and aluminum
dissolution experiments are discussed.
Shown in Figure 8 are the percentages of dissolved iron from the samples Sahara 2 and
8 during 24-hour leaching experiment. For the sample Sahara 8 after 24 hours 0.1 % of total
iron was dissolved in Milli-Q water. But for the sample Sahara 2 only 0.03 % of the total iron
was dissolved. For both samples, percentage of dissolved iron increased until 2 hours of
leaching, which was the maximum amount for the sample Sahara 2. The decrease in
percentage of dissolved iron after 2 hours of leaching and for the sample Sahara 2 also after 5
hours of leaching indicates the existence of equilibrium state between the dissolved iron and
the iron in particle phase and the permanent shift of this equilibrium state.
0,10
S2
S8
dissolved Fe, %
0,08
0,06
0,04
0,02
0,00
0
6
12
18
24
Time (hours)
Figure 8. Percentage of dissolved iron during 24-hour leaching experiment.
23
Exactly the same pattern can be seen for the dissolved manganese and aluminum in
Figures 9 and 10. For sample Sahara 8 the amount of dissolved manganese and aluminum
reached their maximum percentages after 24 hours and they were 0.08 % and 0.09 %,
respectively. For sample Sahara 2 the maximum percentage of dissolved manganese reached
after 24 hours and was 0.06 %. But the amount of dissolved aluminum reached its maximum
after 2 hours of leaching and was 0.05 %. For sample Sahara 8, the percentage of dissolved
iron, manganese and aluminum correlated linearly with the time of leaching (R2=0.96). The
difference between the samples Sahara 2 and Sahara 8 could not be explained by the sample
characterization experiment results we had. We suggest that it is the difference in the
mineralogy of the samples which unfortunately could not be analyzed in frame of this work.
0,10
S2
S8
dissolved Mn, %
0,08
0,06
2
R = 0.96
0,04
0,02
0,00
0
6
12
18
24
Time (hours)
Figure 9. Percentage of dissolved manganese during 24-hour leaching experiment.
24
0,10
S2
S8
dissolved Al, %
0,08
0,06
0,04
0,02
0,00
0
6
12
18
24
Time (hours)
Figure 10. Percentage of dissolved aluminum during 24-hour leaching experiment.
Figures 11 and 12 show the percentages of dissolved calcium and potassium during 24hour leaching experiment. In contrast to iron, manganese and aluminum, calcium and
potassium show the same pattern of dissolution for those two samples. After 24 hours of
leaching, about 1.3-1.4 % of calcium and 0.5 % of potassium were dissolved.
The dissolution followed the logarithmic law (R2 = 0.96 - 0.98). In the first 2 hours of
leaching half of the calcium and potassium, which were dissolved after 24 hours, were already
transferred to the solution.
25
1,6
S2
S8
2
R = 0.97
dissolved Ca, %
1,2
2
R = 0.98
0,8
0,4
0
6
12
18
24
Time (hours)
Figure 11. Percentage of dissolved calcium during 24-hour leaching experiment.
S2
S8
dissolved K, %
0,5
2
R = 0.96
0,4
2
R = 0.96
0,3
0,2
0,1
0
6
12
18
24
Time (hours)
Figure 12. Percentage of dissolved potassium during 24-hour leaching experiment.
26
Although the bulk density was always kept at ratio 1:10 while shaking the samples with
Milli-Q water, every time when the cation analysis was done, 10 ml of the solution was
replaced by 10 ml of Milli-Q water, which makes the ratio 1:27.5 after 8 measurements until
24 hours. The effect of this dilution process is clearly seen in Table 9.
Table 9.
Effect of the sample to Milli-Q water ratio on the dissolution of some metals after
24 hours of leaching
S2
S8
Sample to Milli-Q
water ratio
Fe2+, %
Mn2+, %
Al3+, %
Ca2+, %
K+, %
1:27.5
0,0330
0,0653
0,0423
1,3182
0,4812
1:10
0,0065
0,0274
0,0080
0,8348
0,2549
1:3
0,0033
0,0045
0,0043
0,5150
0,1257
1:27.5
0,2734
0,0782
0,0905
1,4267
0,4872
1:10
0,1448
0,0171
0,0143
0,9902
0,2356
1:3
0,0286
0,0016
0,0008
0,7380
0,1196
The bulk densities chosen in this study are far more different than that occur during the
common Saharan dust events. But the instrumental availability allowed this range, and the
volatile organic compound measurement experiments were also done with these bulk
densities. There are no other studies with which we can directly compare our results. There
are studies which were carried out in ultra pure water but with bulk densities of 10-1000 mg/l
and 15 minutes of leaching, and which were conducted in sea water with 24 hours and 7 days
of leaching [Bonnet and Guieu, 2004; Guieu and Thomas, 1996].
This study showed that there are enough cations coming to the solution when Saharan
dust particles are mixed with the solvent. The following experiments are going to show
whether these cations are in enough amount to produce the volatile halogenated organics
through the anticipated abiotic mechanism.
27
3.7
Results and Discussion of pH analysis
The sample extracts prepared by mixing 10 g of samples with 30 ml of Milli-Q water
for 24 hours and then filtering through 0.45 µm filter showed neutral and slightly basic pH
values. The pH measurement results of the sample extracts are shown in Table 10.
Table 10. pH values of the sample extracts
Samples
pH
Sahara 1
7,98
Sahara 2
7,99
Sahara 2 > 0.063
8,08
Sahara 2 > 0.125
8,04
Sahara 3
7,97
Sahara 4
7,92
Sahara 5
8,03
Sahara 6
8,14
Sahara 7
8,04
Cape Verde
8,52
Sahara 8
7,94
Sahara 8 > 0.063
7,97
Sahara 8 > 0.125
7,95
Sahara 8 > 0.2
7,99
Sahara 8 > 0.315
8,04
Sahara 9
8,16
Gobi 1
9,06
Gobi 2
7,91
Lanzarote < 0.02
8,58
Lanzarote
8,47
28
3.8
Ion balance
The sample characterization analysis was completed. In some analytical methods
certified samples were used to check the measurement reliability. Ion balance is the most
important control to see whether the water chemistry analysis was in order. Generally the total
concentration of anions must equal the total concentration of cations. The analysis is accepted
when the error of ion balance does not exceed 5 %.
The cation analysis was conducted on two samples Sahara 2 and 8, so was the ion
balance. The Table 11 shows the total cation and the Table 12 shows the total anion
concentrations which were used to calculate the ion balance according to the following
formula:
Ion Balance =
Total Cation − Total Anion
∗ 100%
Total Cation + Total Anion
The ion balance for the sample Sahara 2 was - 3.58 %, for the sample Sahara 8 was - 4.20 %,
both of them in the range of acceptable error, which makes the sample characterization
analysis reliable.
Table 11. Total cation concentration
Sample
Ca2+, mmol/l
K+, mmol/l
Mg2+, mmol/l
Na+, mmol/l
Al3+, mmol/l Fe2+, mmol/l
S2
0,8567
0,2143
0,1763
0,7584
0,0416
0,0111
3,1864
S8
1,1968
0,2232
0,1725
0,4504
0,0088
0,0026
3,4441
Total
Table 12. Total anion concentration
Sample
Cl-, mmol/l
NO3-, mmol/l
SO42-,mmol/l
HCO3-, mmol/l
Total
S2
0,6685
0,8960
0,1706
1,5169
3,4227
S8
0,6657
0,9345
0,1741
1,7975
3,7459
29
3.9
Conclusion
The dust samples from Algeria, Cape Verde Island, Lanzarote Island and the Gobi
Desert in Mongolia and their leach solutions in Milli-Q water were characterized using
several analytical methods. In dry dust samples (d < 0.063 mm), elemental composition, water
content and total organic content and in their leach solutions, total dissolved organic and
inorganic carbon, pH, anions and cations were determined.
Intercomparison of some analytical methods was done. For example, the results of XRF
analysis were compared with the results of carbon, sulfur analysis and the Ion
Chromatography analysis. The concentrations of the elements determined by means of XRF
analysis were similar to the concentrations of the Saharan dust samples, also collected in
Algeria, measured by means of acid digestion and following graphite furnace atomic
absorption spectrophotometry (GF-AAS) by Guieu et al. [Guieu and Thomas, 1996].
The fine sized dust samples contained higher amounts of total organics, calcium,
chloride and sulfate. The decreasing bulk density increased the dissolved metal and anion
concentrations. The soluble iron fraction, an essential nutrient for the phytoplankton, in case
of our study the most important constituent of the abiotic volatile halogenated organic
compound formation process, was about 0.03-0.1 %.
The dust samples contained about 4 % of total organics, of which up to 250 ppm were
soluble organic carbon. So the sample characterization experiment results approve the
possibility of abiotic volatile halogenated organic compound formation process during the
interaction of the dust samples from the semi-arid regions with the ocean water, which is a
natural pool of halides, since all the required ingredients for this process are present.
Compared to the soil and sediment samples used for the previous studies, the
concentrations of some ingredients (iron, organics) were very low. But the following
headspace analysis would show whether their concentrations are enough for the process,
whether there exists the same mechanism and whether there are other unknown processes
taking place.
30
4
ABIOTIC PRODUCTION OF METHYL IODIDE AND METHYLENE
CHLORIDE FROM THE INTERACTION BETWEEN DUST PARTICLES
AND SEA WATER
4.1
Summary
The feasibility of abiotic methyl iodide production from the interaction between
Saharan dust particles and seawater was investigated. For this, dust-seawater addition
experiments were conducted in the laboratory, with subsequent analysis by GC-MS. 4.00 g of
fine dust sample (d<0,063mm, 4,7% Fe, 0,19% total carbon), collected from the region of
Southern Algeria (Oued Tamanrasset), was suspended in 10 ml of seawater, collected during
the ship cruise (METEOR 55) and filtered to 0.2 µm, in 20 ml headspace vials, which were
capped immediately after sample preparation. After shaking the samples for certain period,
the gas phase was analysed for volatile organics with a purge and trap GC-MS (Tekmar LSC
2000/Varian Star 3400 cx/ Varian Saturn 2000). The same experiments were also done using
sterilized dust and water samples. To test the production rates of volatile organics, the
following different reaction times were chosen; 20 min, 40 min, 1 hour, 24 hours, 48 hours
and 72 hours. Methyl iodide and methylene chloride were produced from both sterilized and
non-sterilized samples supporting the abiotic production mechanism. The addition of an
electron acceptor Fe (III) increased the amount of methyl iodide produced tenfold. It supports
the proposed abiotic production mechanism for volatile halogenated organic compounds
involving the humic-like substance (HULIS), an electron acceptor Fe (III) and halide ions.
31
4.2
Introduction
Methyl iodide was the first organic iodine compound detected in the atmosphere
[Lovelock, et al., 1973]. It is the most abundant iodine containing halocarbon [Richter, 2003].
The main source of atmospheric methyl iodide is ocean [Carpenter, 2003]. It is produced by
macroalgae [Manley and Dastoor, 1988] and phytoplankton [Oram and Penkett, 1994].
Photochemical production of methyl iodide was shown by Moore et al. [Moore, 1994] and it
was proven by the measurements of Happell and Wallace [Happell and Wallace, 1996].
Methyl iodide has an atmospheric concentration of about 0.1-3 pptv over the open
ocean [Carpenter, et al., 1999; Moore and Groszko, 1999; Yokouchi, et al., 1997]. In the
atmosphere, methyl iodide is photolyzed and oxidized by mainly ozone to produce reactive
iodine radicals. It has an atmospheric lifetime of only 2-6 days [Vogt, 1999]. This short
lifetime enables the use of methyl iodide as a tracer of marine convection [Bell, et al., 2002].
In the marine boundary layer, IO and OIO were detected by DOAS (Differential Optical
Absorption Spectroscopy) [Alicke, et al., 1999; Hebestreit, et al., 2000]. OIO involves in new
particle formation in coastal areas [Hoffmann, et al., 2001].
Although the roles of methyl iodide in the atmosphere are quite well known, there are
still uncertainties in its source characteristics and the parameters that control its production
[Carpenter, 2003]. Atmospheric methyl iodide concentration is shown to have high
correlation with surface seawater temperature [Yokouchi, et al., 2001]. Increased methyl
iodide concentration was measured after dust storms on the Atlantic Ocean [Williams, et al.,
2007].
The results of this chapter show an abiotic production of methyl iodide through the
interaction of dust samples with seawater. Also methylene chloride, which has mainly
anthropogenic sources, was produced abiotically from this natural source.
32
4.3
Materials and Methods
Dust samples from the Sahara Desert were collected directly from the ground in Oued
Tamanrasset (22° 47′ N, 5° 31′ E, 1362 m) in the Hoggar region of southern Algeria and sea
water sample was collected during the ship’s cruise (METEOR 55) on the Atlantic Ocean.
Dust samples were sieved through 0.063 mm sieve and only fine particles were used for the
experiments. Sea water sample was filtered through 0.2 µm filter. 4.00 g of dust sample was
added to 10 ml of sea water in a 20 ml headspace glass vial and was capped immediately after
sample preparation. After shaking the samples in headspace glass vials for certain period at
room temperature, the headspace gas was analyzed with purge-and-trap GC-MS system
(Tekmar LSC 2000/Varian Star 3400 cx/Varian Saturn 2000).
For preliminary experiments, six different interaction periods were chosen; 20 minutes,
40 minutes and 1 hour as short interaction and 24 hours, 48 hours and 72 hours as long
interaction times. For every period, triplets were measured. For purge-and-trap GC-MS
analysis, two needles of purge-and-trap unit, Tekmar LSC 2000, which was connected
upstream to the Gas Chromatograph Varian Star 3400 cx, were stuck through the butyl rubber
stopper of the headspace glass vial containing the sample. Through one of the needles, helium
gas was purged for 4 minutes at the flow rate of 30 ml/min pre-concentrating the headspace
gas on the trap filled with Tenax TA 60/80 and cooled at -90 0C with liquid nitrogen. A water
trap filled with magnesium perchlorate was connected upstream to this pre-concentration trap.
After 2 seconds of dry purge step, desorption of concentrated volatile organics from the preconcentration trap was conducted by heating the trap at 180 0C for 4 minutes. During the
desorption step, the capillary column of the gas chromatograph was cooled at -90 0C with
liquid nitrogen to reconcentrate the volatile organics at the beginning of the column. Gas
chromatographic separation was carried out on a column SGE BP624 (I.D. 0.53 mm, O.D.
0.68 mm, length 25 m) using the temperature program as shown in Figure 1 (see p.9). Mass
spectrometric detection was done in scan mode over a range of 45-280 amu with an ion trap
Mass Spectrometer Varian Saturn 2000. Analytical method is also described in detail
elsewhere [Keppler, et al., 2005]. For the identification and quantification, methyl iodide
standard solution (10 ng/µl) and the standard mix VOC-Mix 20 (200 ng/µl each component)
were used. The masses to identify methyl iodide were 127 and 142. The retention time of
methyl iodide with the used column was 18.8 min. The masses to identify methylene chloride
were 49 and 84. The retention time of methylene chloride with the used column was 19.9 min.
33
4.4
Results and Discussion
The addition of dust samples collected from the identified source regions in southern
Algeria to the filtered Atlantic seawater rapidly produced methyl iodide. This was found for
both sterilized and non-sterilized samples. The amount of methyl iodide produced through
this interaction was very low and was almost always close to detection limit. However, the
experiments with addition of iodide and hydrogen peroxide produced sufficient amounts of
methyl iodide to be reliably quantified. The results of the measurements with the addition of
iodide and hydrogen peroxide and thus the proof that the methyl iodide is produced
abiotically when dust particles interact with seawater are to be found in the works of
Maciejczyk and Williams [Maciejczyk, 2005; Williams, et al., 2007].
According to Keppler et al. who found the abiotic formation mechanism of volatile
halogenated organic compounds in soils and sediments, an electron acceptor Fe (III) is also an
important constituent which contributes to this process [Keppler, et al., 2000]. So the addition
of Fe (III) was tested to see whether it was also true for the case of dust sample and seawater.
In 50 ml headspace glass 0.03 g of dust sample Sahara 5 was put into 30 ml of seawater and
was capped immediately afterwards. The sample was then shaken at room temperature for
half an hour and the headspace gas was analyzed with purge-and-trap GC-MS. As a blank, 30
ml of seawater without addition of dust sample was measured and the measured concentration
of methyl iodide was subtracted from the measured methyl iodide produced from the dust
sample. Fe (III) in form of sulfate was added to the mixture of dust sample and seawater. For
comparison, the same measurement was done with Fe (II) also in form of sulfate instead of Fe
(III). For all the analysis, triplets were done and the average of their results was taken. The
final concentration of Fe (III) and Fe (II) in the headspace glass was 0.036 nM. This
concentration was chosen, because in iron fertilization experiments which took place on the
Pacific and Southern Oceans, they used this concentration to mimic the glacial era
concentration of iron in those Oceans [Wingenter, et al., 2004]. Fe (II) sulfate was used for
the iron fertilization experiments in the ocean, because Fe (III) is not soluble in pH of ocean
water. According to Wingenter et al. the concentration of methyl iodide decreased after
addition of iron in the ocean [Wingenter, et al., 2004], whereas another iron fertilization
experiment showed about doubled increased concentration of methyl iodide [Liss, et al.,
2005]. According to our results, the addition of Fe (II) didn’t show any effect on the
concentration of methyl iodide produced through the interaction of dust samples with
34
seawater, but the addition of Fe (III) caused tenfold increase in methyl iodide production
within half an hour of interaction. The results are shown in Figure 13. The same was also true
for sterilized samples, which excludes biological contribution to the production of methyl
iodide.
1,2
Methyl iodide (ng)
1,0
0,8
0,6
0,4
0,2
0,0
without iron
with Fe (II)
with Fe (III)
Figure 13. The effect of Fe (II) and Fe (III) addition on the production of methyl iodide
through the interaction of dust sample Sahara 5 with seawater.
It is proposed that the humic like substances (HULIS) in dust sample and seawater, iron
in dust sample, and iodide in seawater sample are responsible for the production of methyl
iodide through abiotic substitution reaction [Williams, et al., 2007]. The methoxy group of
HULIS is supposed to give methyl group of the methyl halides produced through the abiotic
mechanism and the model substances with methoxy group which is also structural unit of
HULIS were tested for their production of methyl halides and indeed showed to produce
methyl halides [Keppler, 2000; Maciejczyk, 2005]. But the production of methyl halides could
not be fully explained by this.
It would be helpful to extract the humic substances from the dust samples and to
conduct the headspace glass experiments. More helpful would be to determine as much
structural units of the humic substances as possible, extracted from the dust samples, using
modern analytical techniques like NMR and IR Spectroscopy and to choose the most
35
appropriate model substances for the experiments. Also helpful for fully explaining this
abiotic mechanism would be to qualitatively and quantitatively determine as much products of
this process as possible in both gas and liquid phase using GC-MS and HPLC-MS. The more
products we could determine, the closer we would come to the explanation of the mechanism.
During this work, the headspace gas phase was analyzed by means of purge-and-trap
GC-MS covering the mass range of 45-280 amu. Besides isoprene and methyl chloride,
whose results are discussed in the following chapters, methylene chloride was identified and
quantified along with methyl iodide. As a preliminary experiment, 4 g of Saharan dust
samples were put into 10 ml of sea water in a 20 ml headspace glass, which were capped
immediately after sample preparation, and were shaken for 20 min, 40 min, 1 hour, 24 hours,
48 hours and 72 hours. For each period, triplets were done. After each period the headspace
gas was analysed with GC-MS. The amount of methylene chloride was changing following
the polynomial law (R2=0.72) giving the highest amount after 48 hours (Figure 14.).
Methylene chloride (ng)
0,5
0,4
0,3
2
y = -0,0001 x + 0,0092 x + 0,2853
2
R = 0,72
0,2
0
24
48
72
Time (hours)
Figure 14. Amount of methylene chloride produced through the interaction of dust sample
Sahara 4 with seawater after certain periods. The regression line with the equation is shown.
36
A continuous measurement of methylene chloride was done from air-tightly closed
glass vials, containing 4 g of dust samples in 10 ml of sea water and also from another set of
vials containing 4 g of dust samples in 10 ml of distilled water, every 24 hour for 5-7 days
using purge-and-trap GC-MS. The amount of methylene chloride was decreasing following
the polynomial law (R2=0.97 in seawater and R2=0.92 in distilled water) (Figures 15 and 16.).
For every measurement, triplets were done and the average values with the standard errors are
shown in the Figures. The decrease in methylene chloride concentration over time could be
explained by its degradation.
0,5
2
y = 0,0141 x - 0,1278 x + 0,4581
Methylene chloride (ng)
2
R = 0,92
0,4
0,3
0,2
0,1
0
1
2
3
4
5
Time (days)
Figure 15. Amount of methylene chloride produced every 24 hour through the interaction of
dust sample Sahara 4 with seawater. The regression line with equation is shown.
37
0,5
2
y = 0,0102 x - 0,1237 x + 0,4372
Methylene chloride (ng)
0,4
2
R = 0,97
0,3
0,2
0,1
0,0
0
1
2
3
4
5
6
7
Time (days)
Figure 16. Amount of methylene chloride produced every 24 hour through the interaction of
dust sample Sahara 4 with distilled water. The regression line with equation is shown.
The total amount of methylene chloride was increasing also following the polynomial
law (R2=0.99 in seawater and in distilled water) with time (Figures 17 and 18.). There was no
significant difference between the use of seawater and distilled water as reaction medium.
38
Methylene chloride (ng)
1,6
1,2
0,8
2
y = -0,026 x + 0,3553 x + 0,4527
2
R = 0,99
0,4
0
1
2
3
4
5
Time (days)
Figure 17. Total amount of methylene chloride produced through the interaction of dust
sample Sahara 4 with seawater. The regression line with equation is shown.
Methylene chloride (ng)
1,6
1,2
0,8
2
y = -0,0226 x + 0,2844 x + 0,4774
2
R = 0,99
0,4
0
1
2
3
4
5
6
7
Time (days)
Figure 18. Total amount of methylene chloride produced through the interaction of dust
sample Sahara 4 with distilled water. The regression line with equation is shown.
39
The same experiments were done with sterilized dust and distilled water samples to
approve the abiotic production mechanism and to exclude the biological contribution to the
production of methylene chloride. The samples were sterilized under 0.1 MPa pressure at
1200C for 1 hour in an autoclave. A continuous measurement of methylene chloride was done
from air-tightly closed glass vials containing 4 g of sterilized dust samples in 10 ml of
sterilized distilled water every 24 hour for 4 days using purge-and-trap GC-MS. The amount
of methylene chloride was decreasing following the polynomial law (R2=1) (Figure 19.). For
every measurement triplets were done and the average values with the standard errors are
shown in the Figure.
0,40
2
y = 0,0072 x - 0,0734 x + 0,3784
Methylene chloride (ng)
0,36
2
R =1
0,32
0,28
0,24
0,20
0
1
2
3
4
Time (days)
Figure 19. Amount of methylene chloride produced every 24 hour through the interaction of
sterilized dust sample Sahara 4 with distilled water. The regression line with equation is
shown.
40
The total amount of methylene chloride was increasing also following the polynomial law
(R2=0.99) with time (Figure 20.).
1,2
2
y = 0,0173 x + 0,1045 x + 0,3735
Methylene chloride (ng)
2
R = 0,99
0,9
0,6
0,3
0
1
2
3
4
Time (days)
Figure 20. Total amount of methylene chloride produced through the interaction of sterilized
dust sample Sahara 4 with distilled water. The regression line with equation is shown.
The GC-MS analysis of sterilized samples also gave methylene chloride, which excludes the
biologically mediated methylene chloride production through the interaction of dust particles
with seawater. The abiotic production of methylene chloride was presumed to exist [Law and
Sturges, 2006; Schöler and Keppler, 2003]. And the results of these experiments approved
this presumption.
41
4.5
Conclusion
Methyl iodide and methylene chloride were produced through the interaction of dust
sample with seawater and distilled water.
There is an abiotic production mechanism for these two volatile halogenated
compounds, since the production rate of these compounds from the sterilized samples was as
much as that from non-sterilized samples.
The addition of an electron acceptor Fe (III) increased the amount of methyl iodide
produced tenfold compared to the amount of methyl iodide produced from the interaction of
dust samples with seawater without addition of iron within half an hour, which supports the
abiotic mechanism scheme involving HULIS, iron and halide.
42
5
PRODUCTION OF ISOPRENE FROM THE INTERACTION BETWEEN
DUST PARTICLES AND SEA WATER
5.1
Summary
Isoprene is a very important gas for atmospheric chemistry, because it influences the
oxidation capacity of the atmosphere by reacting with OH radical, and the radiative balance of
the earth by serving as a source for secondary organic aerosol (SOA), and it produces
tropospheric pollutant ozone in the presence of NOx. Plants are the main source of isoprene.
The experiments in headspace glass vials conducted in the laboratory during this work show
natural dust as a source of isoprene to the atmosphere. Dust samples from the Sahara Desert,
Gobi Desert and also the dust sample from Cape Verde Island which has undergone aeolian
transport produced isoprene both in the presence of seawater and distilled water. In contrast to
methyl iodide and methylene chloride, isoprene production was biologically-mediated,
because sterilized samples did not produce it. The concentration of isoprene increased linearly
(R2=0.96) until 72 hours of reaction time. 0.398 ng of isoprene was released from 4 g of
Saharan dust after 72 hours. There was a positive correlation (R2=0.99) between the amount
of dust and the amount of isoprene produced after 48 hours reaction time. Dust samples that
have already undergone aeolian transport produced more isoprene compared to the samples
collected in the source regions. If those microorganisms in dust samples are common soil
organisms, this process would explain until now not fully explained reason for the increased
atmospheric isoprene concentration during wet seasons.
43
5.2
Introduction
In 1860 a scientist and analytical chemist Charles H. Greville Williams first isolated the
monomer of natural rubber through dry distillation method and named it “isoprene”. Isoprene
(2-methyl-1, 3-butadiene) is a photochemically reactive and highly volatile organic compound
which is emitted into the atmosphere from biogenic sources, mainly plants. It was only 1957
that Guivi Alexander Sanadze discovered the emission of isoprene from plants [Sanadze,
1957] and it was confirmed later by Rasmussen [Rasmussen, 1970]. Isoprene is formed in the
leaves of the plants through the enzymatic, magnesium ion dependent elimination reaction of
diphosphate from dimethylallyl diphosphate (DMADP). The enzyme responsible for this
reaction was discovered in 1991 and is called isoprene synthase [Silver and Fall, 1991, 1995].
The annual global isoprene flux is estimated to be 506 Tg C/yr [Guenther, et al., 1995].
Isoprene is very important for atmospheric chemistry, air pollution and global warming issue
because of the following reasons.
In the daytime isoprene is oxidized to organic peroxides mainly by the reaction with OH
radical. In the daytime OH radical is abundant in the atmosphere, since its main source is the
destruction of ozone by sunlight and the following reaction of oxygen atom with water
molecules. In the nighttime however, isoprene is oxidized mainly by nitrate radical (NO3)
formed by the reaction of NO2 with ozone [Starn, et al., 1998a]. The emission, chemistry and
fate of isoprene and other biogenic volatile organic compounds are reviewed in details
elsewhere [Atkinson, 1990, 1997, 2000; Atkinson and Arey, 2003a; 2003b; Harley, et al.,
1999; Kesselmeier and Staudt, 1999].
In the presence of NOx (NO2+NO), organic peroxides formed from the oxidation of
isoprene and other volatile organic compounds react mainly with NO competing with ozone.
Nitrogen dioxide (NO2) produced through this reaction is broken down by sunlight giving
oxygen atom which reacts with oxygen molecule in the atmosphere and produces ozone, a
pollutant in the troposphere and a third most important greenhouse gas [Pierce, et al., 1998;
Roberts, et al., 1995; Starn, et al., 1998b; Williams, et al., 1997]. This oxidation and also the
oxidation of the first generation products of isoprene oxidation, methyl vinyl ketone (MVK),
methacrolein (MACR) and 3-methyl furan, lead to the formation of organic nitrates and
peroxyacyl nitrates (PANs) [Bertman and Roberts, 1991; Chen, et al., 1998; Grossenbacher,
et al., 2001; Harley, et al., 1999; Montzka, et al., 1993; Nouaime, et al., 1998; Paulson and
Seinfeld, 1992; Pegoraro, 2004; Pierotti, et al., 1990; Roberts, et al., 1998a; Roberts, et al.,
44
2004; Roberts, et al., 2002; Roberts, et al., 2001; Roberts, et al., 2003; Roberts, et al., 1998b;
Tanimoto and Akimoto, 2001]. The relative stability of PANs enables their long distance
transport to remote areas. In remote areas, PANs serve as a source of NOx, eventually a source
of pollutant ozone in the atmosphere [Shallcross and Monks, 2000].
Isoprene and its oxidation products are the sources for the secondary organic aerosol
(SOA) which plays an important role in the radiation balance of the earth both through direct
and indirect means [Altieri, et al., 2006; Böge, et al., 2006; Carlton, et al., 2006; Claeys, et
al., 2004a; Claeys, et al., 2004b; Edney, et al., 2005; Henze and Seinfeld, 2006; Kleindienst,
et al., 2006; Kroll, et al., 2006; Lane and Pandis, 2007; Lim, et al., 2005; Limbeck, et al.,
2003; Ruppert and Becker, 2000; Surratt, et al., 2006; van Donkelaar, et al., 2007].
Isoprene reacts immediately with OH radical in the atmosphere, thus reduces the
oxidation capacity of atmosphere and makes the lifetime of second important greenhouse gas
(GHG), methane longer.
There are some other sources of isoprene, like sea water [Bonsang, et al., 1992;
Broadgate, et al., 1997; Matsunaga, et al., 2002], phytoplankton [Moore, et al., 1994;
Yokouchi, et al., 1999], bacteria [Kuzma, et al., 1995; Wagner, et al., 2000; Wagner, et al.,
1999], wetland, peatland [Janson and De Serves, 1998; Tiiva, et al., 2007] and so on. But
compared to the terrestrial emission, the strength of those other sources are negligible.
However, due to high reactivity of isoprene, even its small sources have potential to impact
the chemistry of the area where it is produced, for example local remote Marine Boundary
Layer (MBL) chemistry [Palmer and Shaw, 2005].
There is large uncertainty in the emission estimates of isoprene because of scarcity of
direct measurements. Indirect emission estimates from satellite measurements are also done
[Palmer and Shaw, 2005; Shim, et al., 2005]. The current estimates are the results of global
models [Guenther, et al., 1995; Guenther, et al., 2006].
This work presents the results of laboratory experiments which show a new interesting
source of isoprene.
45
5.3
Materials and Methods
Dust samples Sahara 4, Sahara 5, Cape Verde and Mongolia 2 were used in the
following experiments (see p.4), because they contained higher amounts of organics than the
other samples. 4 g sample Sahara 4 was added to 10 ml of seawater in 20 ml headspace glass
vial, which was capped immediately after sample preparation. It was shaken at room
temperature for certain period and the headspace gas was measured by GC-MS. Later, the
amount of dust sample was reduced to 1 g to see the effect of bulk density on the production
of isoprene. As it was seen that the bulk density had no big effect on the production, 4 g
samples were used for further experiments to better quantify isoprene amounts. For the
identification and quantification of isoprene, isoprene standard solution was prepared.
Analytical method is described in detail in Chapter 4 (see p.31). The masses to identify
isoprene were 53, 67 and 39. The retention time of isoprene with the used column was 17.3
min. The same set of experiments was also done with sterilized samples to exclude the
microbiological contribution to the production of isoprene. The samples were sterilized in an
autoclave under 0.1 MPa pressure at 120 0C for one hour.
46
5.4
Results and Discussion
5.4.1
Production of isoprene
As a preliminary experiment, 4 g of dust samples Sahara 4 were put into 10 ml of sea
water in 20 ml headspace glass vials, which were capped immediately after sample
preparation, and were shaken for 5 min, 20 min, 40 min, 1 hour, 24 hours, 48 hours and 72
hours. For each period, triplets were done. After each period, the headspace gas was analyzed
with GC-MS. We identified isoprene in headspace gas and the amount of isoprene was
increasing linearly (R2=0.96) with increasing reaction time till 72 hours, giving 0.398 ng of
isoprene after 72 hours (Figure 21.).
0,5
Isoprene, ng
0,4
0,3
0,2
y = 0.0052 x + 0.0374
2
R = 0.96
0,1
0,0
0
24
48
72
Time (hours)
Figure 21. Amount of isoprene produced through the interaction of 4 g of dust samples
Sahara 4 with 10 ml of Atlantic Ocean water after certain periods. The regression line with the
equation is shown.
The final dust concentration in the 10 m mixed layer of ocean (the bulk density) during
strong Saharan dust event is 0.5 mg/l [Bonnet and Guieu, 2004]. In this case, the amount of
headspace gas produced was always below the detection limit of the instrument. The same
experiment was done with 1 g of dust sample in 10 ml of ocean water. The reaction periods
were 5 min, 20 min, 40 min, 1 hour, 6 hours, 12 hours and 48 hours, respectively.
47
For each period, also the triplets were analyzed. The results supported the previous
results (Figure 22.). The amount of isoprene produced was increasing linearly (R2=0.98) with
reaction time till 48 hours. After same period (48 hours) about 4 times less isoprene was
produced than the experiment with 4 g of sample. So we conclude that the bulk density has no
relevant effect on the isoprene production whereas it does in dissolution of iron in surface
seawater [Bonnet and Guieu, 2004].
0,08
Isoprene, ng
0,06
0,04
y = 0.0013 x + 0.0095
0,02
2
R = 0.98
0,00
0
24
48
Time (hours)
Figure 22. Amount of isoprene produced through the interaction of 1 g of dust samples
Sahara 5 with 10 ml of Atlantic Ocean water after certain periods. The regression line with the
equation is shown.
48
5.4.2
Dependence of isoprene amount on the organic content of the dust
The next experiment was done to test the dependence of the amount of isoprene on the
amount of dust sample. 0.05 g, 0.10 g, 0.50 g, 1.00 g, 2.00 g and 4.00 g of dust samples from
Cape Verde Island were put into 10 ml of sea water and were shaken for 48 hours each. The
amount of isoprene produced increased linearly (R2=0.99) with increasing amount of dust
sample (Figure 23.), which is also increasing organic content.
1,00
Isoprene (ng)
0,75
0,50
0,25
y = 0.2359 x - 0.0259
2
R = 0.99
0,00
0
1
2
3
4
Dust (g)
Figure 23. Dependence of isoprene production on the amount of dust. The regression line
with the equation is shown.
After 48 hours of reaction time, about 2.5 times more isoprene was produced from 4 g
of dust sample from Cape Verde Island (0.931 ng) than that produced from the same amount
of dust sample collected directly in the source region in Algeria (0.368 ng). The same was
true for the 1 g dust sample from Cape Verde Island (0.184 ng) and Algeria (0.075 ng).
Dust sample which has already gone aeolian transport from Sahara produced more
isoprene than the dust sample collected directly from the source region.
It suites well with the ratio of organic content (Ratio=2.3, Cape Verde-7.7%, Algeria3.4%) and the total dissolved organic content of the samples (Ratio=2.8, Cape Verde-0.11
g/kg, Algeria-0.04 g/kg).
49
An input of iron containing mineral dust can lead to abiotic production of methyl iodide
in the ocean [Williams, et al., 2007]. But a microbiological contribution would be more likely
explanation for the production of isoprene [Fall and Copley, 2000]. Bacteria produce isoprene
[Kuzma, et al., 1995; Wagner, et al., 2000; Wagner, et al., 1999] and viable fungi and bacteria
from Africa are transported interhemispherically with soil dust [McCarthy, 2001; Monteil,
2002; Prospero, et al., 2005]. Also during Asian dust periods, fungal spores are found in the
ambient air of west Korea [Yeo and Kim, 2002]. The same experiments were done with
sterilized dust, sea water and distilled water samples to test our hypothesis of biologically
mediated isoprene production. The samples were autoclaved under 0.1 MPa pressure at 1200C
for one hour to exclude the microbiological contribution to the production of isoprene. The
GC-MS analysis of those samples gave no isoprene, which supports the biologically mediated
isoprene production through the interaction of dust particles with seawater.
50
5.4.3
Production rate of isoprene and the experiment with distilled water
Since it was shown that there was a biologically mediated isoprene production through
the interaction of dust samples with sea water, a question arose “How much is the capability
of isoprene production of those microorganisms in dust samples?” To answer this question,
continuous measurements of isoprene were done from air-tightly closed glass vials containing
4 g of dust samples in 10 ml of sea water and also from another set of vials containing 4 g of
dust samples in 10 ml of distilled water every 24 hour for 5-9 days using purge-and-trap GCMS. The amount of isoprene was decreasing exponentially (R2=0.76 in seawater and R2=0.96
in distilled water) with time (Figure 24 and 25.) but it was not exhausting.
0,3
-0.2051x
y = 0.2819 e
2
Isoprene (ng)
R = 0.76
0,2
0,1
0,0
0
2
4
6
8
10
Time (days)
Figure 24. Amount of isoprene produced every 24 hour through the interaction of dust
sample Sahara 4 with seawater. The regression line with equation is shown.
51
0,3
-0.2809x
y = 0.3197 e
2
Isoprene (ng)
R = 0.96
0,2
0,1
0,0
1
2
3
4
5
Time (days)
Figure 25. Amount of isoprene produced every 24 hour through the interaction of dust
sample Sahara 4 with distilled water. The regression line with equation is shown.
52
The total amount of isoprene was increasing logarithmically (R2=0.99 in seawater and
R2=0.98 in distilled water) with time (Figure 26 and 27.). There was no significant difference
between the uses of seawater and distilled water as reaction medium.
1,0
Isoprene (ng)
0,8
0,6
y = 0.2497 ln(x) + 0.3078
2
0,4
0,2
R = 0.99
0
2
4
6
8
10
Time (days)
Figure 26. Total amount of isoprene produced through the interaction of dust sample Sahara4
with seawater. The regression line with equation is shown.
53
Isoprene (ng)
0,8
0,6
0,4
y = 0.2869 ln(x) + 0.2709
2
R = 0.98
0,2
1
2
3
4
5
Time (days)
Figure 27. Total amount of isoprene produced through the interaction of dust sample Sahara4
with distilled water. The regression line with equation is shown.
Although we did not have water sample from Pacific Ocean, we were interested in
finding out whether dust sample from the Gobi Desert would also produce isoprene when it
interacted with water. So the same set of measurements was done with the sample from Gobi
Desert in the water sample from Atlantic Ocean and in distilled water.
The sample from Gobi Desert was the one collected in the city center of South Gobi
province in Mongolia, which was denoted previously as sample M2. This dust sample was
wind blown and was accumulated at the bottom of the fence which typically every family in
the city center of the provinces has around its house. So the possibility of anthropogenic
pollution was not excluded and it also was shown by XRF and other chemical sample
characterization experiments.
54
In comparison to the dust sample Sahara 4, the sample from Gobi Desert produced 2.5-5
times more isoprene after 24 hours of interaction with seawater and distilled water. Twice
more amount of isoprene was produced from the experiment with distilled water than that
with sea water. The amount of isoprene produced every 24 hour was decreasing following the
polynomial law (R2=0.90) in seawater, and following the potential law (R2=0.95) in distilled
water. The results are shown in Figures 28 and 29.
Isoprene (ng)
0,8
0,6
0,4
2
y = -0.0334 x + 0.1682 x + 0.6201
0,2
2
R = 0.90
1
2
3
4
5
6
7
Time (days)
Figure 28. Amount of isoprene produced every 24 hour through the interaction of dust
sample from the Gobi Desert with seawater. The regression line with equation is shown.
As shown in Figure 28, the amount of isoprene measured on the 4th day after the
first measurement was almost the same or slightly lower than that of the first day, when the
dust sample was shaken with sea water.
55
However, as it is seen in Figure 29, in case of the measurements with distilled water, the
amount of isoprene produced on the third day after the first measurement was more than 3
times lower than that of the first time. This could be explained by the existence of both
isoprene producing and consuming bacteria in the dust samples and the difference of their
activity in different medium. May be the source-sink relationship would be analogous to that
seen with methanogens and methanotrophs [Fall and Copley, 2000].
1,5
-1.1685
y = 1.3564 x
Isoprene (ng)
2
R = 0.95
1,0
0,5
0,0
0
1
2
3
4
5
6
7
8
9
Time (days)
Figure 29. Amount of isoprene produced every 24 hour through the interaction of dust
sample from the Gobi Desert with distilled water. The regression line with equation is shown.
56
As it is seen in Figures 30 and 31, the total amounts of isoprene produced through the
interaction of dust samples from the Gobi Desert with sea water and with distilled water were
increasing with time exponentially (R2=0.98 in seawater and R2=0.94 in distilled water). The
amount of isoprene produced after 7 days was slightly higher in distilled water.
Isoprene (ng)
2,0
1,5
1,0
0.1681x
y = 0.6356 e
2
R = 0.98
0,5
1
2
3
4
5
6
7
Time (days)
Figure 30. Total amount of isoprene produced through the interaction of dust sample from
the Gobi Desert with sea water. The regression line with equation is shown.
57
2,5
0.0722x
y = 1.3004 e
2
Isoprene (ng)
R = 0.94
2,0
1,5
1
2
3
4
5
6
7
8
Time (days)
Figure 31. Total amount of isoprene produced through the interaction of dust sample from
the Gobi Desert with distilled water. The regression line with equation is shown.
Based on the results of laboratory measurements we could make rough estimate of total
amount of isoprene produced during dust storm events annually. It was a very little
contribution of isoprene (~ 1500 kg) compared to its plant source. However, this is the case
where soil comes into the water and produces isoprene. But on the other hand, how about the
case where vice versa occurs, namely when it rains? Of course there would be many other
factors influencing the isoprene production of microorganisms in soil. Indeed, there is a study
where it shows the consumption of isoprene in soil [Cleveland and Yavitt, 1997; Cleveland
and Yavitt, 1998]. They studied the influence of different factors, such as soil temperature and
moisture on the isoprene consumption of soil. In their study, the initial amount of isoprene
introduced increased however, when the soil moisture was 100 percent, which supports the
results of our measurements.
58
Isoprene emissions were estimated to be an order of magnitude higher during the
November to April wet season, at a level of 23 Tg C, than during the May to October dry
season, at a level of 2 Tg C in Tropical Australia [Ayers and Gillett, 1988]. It was explained
with the increased biomass during this period [Guenther, et al., 1995]. However,
microbiological contribution of soil could be high in this region during the wet season,
especially, when rain falls on practically every day.
Seasonal emissions of isoprene were measured from an elevated dry site and low wet
site on a Sphagnum fen in Sweden. The flux from wetter site was generally about 20 times
more than that of drier site. The reason for this difference could not be determined in that
work [Janson and De Serves, 1998].
Diurnal cycles and seasonal variation of isoprene and its oxidation products were
measured in the tropical savanna atmosphere. Two times higher levels of isoprene were
observed during the wet season. The authors explain the difference with the lower
physiological activity of the vegetation during the dry season [Holzinger, et al., 2002].
It is unclear whether soils contribute significantly to atmospheric isoprene [Fall and
Copley, 2000]. So it is important to find out which kind of microorganisms present in dust
samples are responsible for the production of isoprene, whether they are also found in
different types of soils and how much would be their contribution to the atmospheric isoprene
emission.
59
5.5
Conclusion
Isoprene was produced through the interaction of dust sample with seawater and
distilled water.
There was a positive correlation (R2=0.99) between the amount of dust and the amount
of isoprene produced after 48 hours reaction time.
There is a biologically mediated isoprene production when the dust particles come into
the sea water, since there was no isoprene production with sterilized samples.
Microbiological contribution of isoprene production in soil, especially during the wet
season should be further studied to correctly evaluate and differentiate between the
contribution of plant and soil isoprene emission to the atmosphere.
60
6
PRODUCTION OF METHYL CHLORIDE FROM THE INTERACTION
BETWEEN DUST PARTICLES AND SEA WATER
6.1
Summary
Methyl chloride has a lifetime of about 1.5 years. Because of this long lifetime, it
reaches stratosphere and the chlorine atom, which destroys the stratospheric ozone layer, is
produced through the photolysis of methyl chloride there. The production of methyl chloride
from the interaction between Saharan dust particles and seawater was investigated. In the
laboratory, the dust samples were added to the seawater samples in 20 ml headspace glass
vials and were shaken for certain periods. Then the gas phase products were analyzed by
purge-and-trap GC-MS (Tekmar LSC 2000/Varian Star 3400 cx/ Varian Saturn 2000). To test
the production rates of methyl chloride, the following different reaction times were chosen; 5
min, 20 min, 40 min, 1 hour, 3 hours, 6 hours, 12 hours, 24 hours, 48 hours and 72 hours.
Methyl chloride was produced through the interaction of dust sample with seawater. The
concentration of methyl chloride increased logarithmically (R2=0.89) until 72 hours of
reaction time. 1.984 ng of methyl chloride was released from 4 g of Saharan dust after 72
hours. There was a positive correlation (R2=0.93) between the amount of isoprene and the
amount of methyl chloride produced after 9 days reaction time.
We suggest that there is a biologically mediated methyl chloride production when the dust
particles come into the sea water, since there was no methyl chloride production with
sterilized dust and sea water samples.
61
6.2
Introduction
Methyl chloride (CH3Cl) is the most abundant chlorine-containing compound in the
atmosphere, with globally averaged concentration of about 550 ppt [Clerbaux and Cunnold,
2006b; Li, et al., 2001]. It reaches stratosphere where its chlorine atom, released through
photolysis, catalytically destroys ozone [Butler, 2000].
Tropical forest is the major source of methyl chloride with an annual flux of 820-8200
Gg / yr [Moore, et al., 2005; Yokouchi, et al., 2002; Yokouchi, et al., 2000]. Abiotic
conversion of chloride to methyl chloride in senescent and dead leaves at ambient
temperatures accounts for 20-2500 Gg / yr of methyl chloride emission [Hamilton, et al.,
2003]. Biomass burning is another major source of methyl chloride to the atmosphere with
emission estimates of 325-1125 Gg / yr [Andreae, et al., 1996; Andreae and Merlet, 2001;
Lobert, et al., 1999]. Ocean, which was believed to be the major source of methyl chloride, is
a relatively modest source with an annual flux of 380-500 Gg / yr [Koppmann, et al., 1993;
Li, et al., 2001; Moore, 2000; Moore, et al., 1996; Tait, et al., 1994; Yvon-Lewis, et al., 2004].
Also salt marshes emit large amounts (65-440 Gg / yr) of methyl chloride [Cox, et al., 2004;
Dimmer, et al., 2001; Manley, et al., 2006; Rhew, et al., 2002; Rhew, et al., 2000]. Methyl
chloride emission by wood-rotting fungi is estimated to be 40-85 Gg / yr [Dimmer, et al.,
2001; Watling and Harper, 1998]. In addition, ectomycorrhizal fungi have recently been
found to emit methyl chloride [Redeker, et al., 2004], but the emission strength has not been
estimated yet. Wetlands are significant sources with global annual flux of 48 Gg / yr [Varner,
et al., 1999]. Rice paddies emit 2.4-4.9 Gg of methyl chloride per year [Lee-Taylor and
Redeker, 2005; Redeker, et al., 2002; Redeker and Cicerone, 2004; Redeker, et al., 2000]. The
decomposition of soil organic matter can act as a source of methyl chloride [Keppler, et al.,
2000] but its importance has not been evaluated. Fossil fuel burning, waste incineration and
industrial processes also produce methyl chloride, with annual fluxes of 107 Gg / yr, 45 Gg /
yr and 10 Gg / yr, respectively [McCulloch, et al., 1999].
The largest sink of methyl chloride in the atmosphere is the reaction with OH radical,
which accounts for an annual loss of 3800-4100 Gg [Lee-Taylor and Brasseur, 2001;
Yoshida, et al., 2004]. Soil is the second largest sink for methyl chloride.
Khalil and
Rasmussen estimated an uptake of 500 Tg / yr [Khalil and Rasmussen, 2000].
Microbiological studies lead to larger sink (1600 Gg / yr) for methyl chloride in soil [Harper
and Hamilton, 2003; McAnulla, et al., 2001]. Cold ocean water takes up 93-145 Gg of methyl
62
chloride annually [Moore, 2000] and again microorganisms are showed to play important role
in the degradation process [Tokarczyk, et al., 2003b; Tokarczyk, et al., 2003a]. Reaction with
chlorine radicals in the marine boundary layer is another loss process for methyl chloride with
estimated sink of 400 Gg / yr [Keene, et al., 1996; Khalil and Rasmussen, 1999].
Although above mentioned sources and sinks could be balanced by each other, the
large uncertainties in emissions do not exclude another possible sources or sinks [Clerbaux
and Cunnold, 2006b]. This work presents the results of laboratory experiments which show
another interesting source with a possible common mechanism to explain the emission of
methyl chloride from some environments.
6.3
Materials and Methods
The standard mix VOC-Mix 20, which contains methyl chloride, was used for the
identification and quantification of methyl chloride in the headspace gas by purge-and-trap
GC-MS method. The same temperature program as used in Chapters 4 and 5 was employed.
The retention time of methyl chloride was 11 min with this program. The masses 49 and 50 in
combination with retention time were used for the identification. The samples Sahara 4,
Sahara 5 and Mongolia 2 were mixed with seawater from Atlantic Ocean in 20 ml headspace
glass vials, shaken for certain periods starting from 5 minutes to 9 days at room temperature
and afterwards the headspace gas was analyzed for methyl chloride. The same set of
experiments as in Chapters 4 and 5 were done. To differentiate between abiotic and biotic
production of methyl chloride, the samples were sterilized and used for the analysis. The
sterilization was performed in an autoclave under 0.1 MPa pressure at 120 0C for 1 hour. The
seawater and distilled water samples were measured without addition of dust samples. These
samples served as blanks in the quantification.
63
6.4
Results and Discussion
As a preliminary experiment 4 g of Saharan dust samples were put into 10 ml of
seawater in 20 ml headspace glass vials, which were capped immediately after sample
preparation, and were shaken for 5 min, 20 min, 40 min, 1 hour, 24 hours, 48 hours and 72
hours. For each period, triplets were prepared. After each period, the headspace gas was
analyzed with GC-MS. We identified methyl chloride in headspace gas and the amount of
methyl chloride was increasing logarithmically (R2=0.89) with increasing reaction time until
72 hours, giving 1.984 ng of methyl chloride after 72 hours (Figure 32.).
2,0
Methyl chloride (ng)
1,6
1,2
y = 0.2917 ln(x) + 0.6341
0,8
2
R = 0.89
0,4
0,0
0
24
48
72
Time (hours)
Figure 32. Amount of methyl chloride produced through the interaction of 4 g of dust
samples Sahara 4 with seawater after certain periods. The regression line with the equation is
shown.
The final dust concentration in the 10 m mixed layer of ocean (the bulk density)
during strong Saharan dust event is 0.5 mg / l [Bonnet and Guieu, 2004]. An attempt to
simulate this condition failed, because the amount of headspace gas produced was always
below the detection limit of the instrument. The same set of experiment was done with 1 g of
dust sample Sahara 5 in 10 ml of ocean water.
64
The reaction periods were 5 min, 20 min, 1 hour, 3 hours, 6 hours, 12 hours, 24 hours
and 48 hours, respectively. For each period, also triplets were analyzed. The amount of
isoprene produced was increasing linearly (R2=0.91) with reaction time until 48 hours, which
means after 4 days it will start to produce more methyl chloride than the measurement with 4
g of dust sample (Figure 33.). It means that the bulk density or the final concentration of dust
in sea water plays a role in the production of methyl chloride.
1,6
Methyl chloride (ng)
1,2
0,8
0,4
y = 0.0208 x + 0.2881
2
R = 0.91
0,0
0
24
48
Time (hours)
Figure 33. Amount of methyl chloride produced through the interaction of 1 g of dust
samples Sahara 5 with seawater after certain periods. The regression line with the equation is
shown.
An input of iron containing mineral dust can lead to an abiotic production of methyl
iodide in the ocean [Williams, et al., 2007]. To test whether the same mechanism leads to the
production of methyl chloride, the same experiments were done with sterilized dust, sea water
and distilled water samples. The samples were autoclaved under 0.1 MPa pressure at 1200C
for 1 hour to exclude the microbiological contribution to the production of methyl chloride.
The GC-MS analysis of those samples gave no methyl chloride, which shows the biologically
mediated methyl chloride production through the interaction of dust particles with sea water.
Fungi emit methyl chloride [Dimmer, et al., 2001; Redeker, et al., 2004; Watling and Harper,
65
1998] and viable fungi and bacteria from Africa are transported interhemispherically with soil
dust [Prospero, et al., 2005].
Since it was shown that there is a biologically mediated methyl chloride production
through the interaction of dust samples with sea water, the following measurements were
performed to determine the capability of those microorganisms in dust samples to produce
methyl chloride. The continuous measurements of methyl chloride were done from air-tightly
closed glass vials containing 4 g of dust samples in 10 ml of sea water (and also distilled
water) every 24 hour using purge-and-trap GC-MS. The amount of methyl chloride was
decreasing exponentially (R2=0.91 in seawater) with time (Figure 34.).
2,0
Methyl chloride (ng)
1,6
-0,1535x
y = 1.8876 e
2
R = 0.91
1,2
0,8
0,4
0
2
4
6
8
10
Time (days)
Figure 34. Amount of methyl chloride produced every 24 hour through the interaction of dust
sample Sahara 4 with seawater. The regression line with equation is shown.
66
The total amount of methyl chloride emitted was increasing logarithmically (R2=0.89
in seawater) with time (Figure 35.). The experiment conducted with distilled water gave no
methyl chloride, which showed the participation of chloride ion in the process of methyl
chloride formation.
8
Methyl chloride (ng)
6
4
y = 2.1232 ln(x) + 1.6973
2
R = 0.89
2
0
2
4
6
8
10
Time (days)
Figure 35. Total amount of methyl chloride produced through the interaction of dust sample
Sahara 4 with sea water. The regression line with equation is shown.
67
Also the measurement with dust samples from the Gobi Desert gave methyl chloride
(Figure 36 and 37.). The amount of methyl chloride was decreasing exponentially (R2=0.85 in
seawater) with time (Figure 36.).
16
Methyl chloride (ng)
-0.3031x
y = 59.896 e
12
2
R = 0.85
8
4
5
6
7
Time (days)
Figure 36. Amount of methyl chloride produced every 24 hour through the interaction of dust
sample from the Gobi Desert with seawater. The regression line with equation is shown.
68
The total amount of methyl chloride emitted was increasing logarithmically (R2=0.89
in seawater) with time (Figure 37.). The amount of the methyl chloride produced was 2-5
times more than that emitted from the same amount of Saharan dust sample after 5-7 days of
interaction with sea water. However, the emission of methyl chloride started later than that of
Saharan dust sample. After 24 hours of interaction, no methyl chloride was produced from the
interaction of dust sample from the Gobi Desert with sea water.
40
y = 50.608 ln(x) - 68.005
Methyl chloride (ng)
2
30
R = 0.89
20
10
5
6
7
Time (days)
Figure 37. Total amount of methyl chloride produced through the interaction of dust sample
from the Gobi Desert with sea water. The regression line with equation is shown.
69
In both dust samples, our results revealed correlations between the emission rates of
methyl chloride and isoprene (Figures 38 and 39.). This suggests again that they are both
produced by biologically mediated processes.
Methyl chloride (ng)
30
20
10
y = 22.216 x - 18.258
2
R = 0.97
0
0,5
1,0
1,5
2,0
Isoprene (ng)
Figure 38. Correlation between the emissions of methyl chloride and isoprene produced
through the interaction of dust sample from the Gobi Desert with sea water. The regression
line with the equation is shown.
70
Methyl chloride (ng)
8
6
4
y = 8.6627 x - 1.0214
2
0,2
2
R = 0.93
0,4
0,6
0,8
1,0
Isoprene (ng)
Figure 39. Correlation between the emissions of methyl chloride and isoprene produced
through the interaction of dust sample from Sahara with seawater. The regression line with the
equation is shown.
Based on the results of laboratory measurements we made an estimate of total amount
of methyl chloride (5,3 - 37,5 Mg / yr) produced during dust storm events annually. It is a
very little contribution of methyl chloride (0.0001 – 0.001 %) compared to its other known
sources. The total methyl chloride source strength is about 4000 Gg/yr [Clerbaux and
Cunnold, 2006a].
71
But it seems that this process has a common mechanism with the emission of methyl
chloride from coastal and other wetland areas. Unplanted flooded fields emit as much methyl
chloride as planted flooded rice fields [Redeker, et al., 2000]. This may be a common feature
of saturated, anoxic organic soils [Varner, et al., 1999], but the desert soil would not be really
counted as an organic soil. It would be interesting to study which biological organisms are
responsible for the methyl chloride emission and which other factors, such as chloride
concentration and how they would influence the emission rate. Viable fungi and bacteria from
Africa are transported interhemispherically with soil dust [Prospero, et al., 2005]. Dust
samples from Cape Verde Island which have already gone aeolian transport, however did not
produce methyl chloride after 48 hours of interaction with sea water. So this is an interesting
question especially for microbiologists to answer which microorganisms there are originally
in desert soil, which of them survive the long range transport and many more questions, like
which common features have those regions affected by dust storm, also for the scientists from
different disciplines.
6.5
Conclusion
Methyl chloride was produced through the interaction of dust sample with seawater.
There is a biologically mediated methyl chloride production when the dust particles
come into the sea water, since there was no methyl chloride production with sterilized
samples. Dust samples from the Gobi Desert produced 2-5 times more methyl chloride than
the samples from Sahara Desert after 5-7 days interaction time.
There was a very good correlation (R2=0.93-0.97) between the emissions of methyl
chloride and isoprene in dust samples.
Microbiological contribution of methyl chloride production in soil in coastal wetland
areas especially that are affected by dust storms would be interesting subject for further
studies.
72
7
CONCLUSION AND FUTURE PERSPECTIVES
The results of the headspace glass experiments performed during this study are
summarized, and the main conclusions with possible suggestions for further experiments are
included in this chapter.
Methyl iodide, methylene chloride and isoprene were produced through the interaction
of dust samples from the Sahara and Gobi Desert with both seawater and distilled water,
methyl chloride was only produced through the interaction of dust samples with seawater.
As initially hypothesized, methyl iodide was produced abiotically, since sterilized
samples produced the same amount. A tenfold increase in methyl iodide production upon
addition of Fe (III) within half an hour approves an abiotic production mechanism involving
HULIS, iron and halide. The presumption of methylene chloride production through this
abiotic mechanism was also approved by the results of this work.
An abiotic production of methyl chloride could not be determined, most likely because
the produced amount fell below the detection limit of the instrument. Instead, a biologically
mediated methyl chloride and isoprene production was determined. Only non-sterilized
samples produced methyl chloride and isoprene, and significant amounts were determined
after at least 24 hours of interaction, which reemphasises a biological contribution. Although
there was a very good correlation between the produced amounts of methyl chloride and
isoprene, 48 hours of interaction of the dust samples from Cape Verde Island with seawater
produced only isoprene but not methyl chloride, a possible hint for the responsibility of
different types of microorganisms for their production.
Furthermore, methyl chloride was only produced in the presence of seawater which
reveals the necessity of halide presence. Contrary to that, the addition of water was enough for
isoprene production. The capability of dust samples to produce methyl chloride and isoprene
was not finished within 7-9 days which is a typical period for dust particles to stay in 10 m
mixed layer of ocean after dust storms.
73
Dust samples from Cape Verde Island which have undergone aeolian transport
produced more isoprene compared to the dust samples directly collected in the source region.
It indicates not only the transport of microorganisms during the dust storms, but also gives a
hint about either their enrichment as a result of transport or the contribution of local
microorganisms to the production of isoprene, or even both. In latter case, it concerns
common soil organisms that produce isoprene in the presence of water, which in turn would
explain the increase in atmospheric isoprene concentrations during wet seasons compared to
dry seasons.
It is also possible for the abiotic production of VHOC to take place on aerosols in
marine atmosphere, which would probably be more effective than the process in the ocean.
Further study is needed to make an estimate of the significance of this process. It is not
excluded that the semi-volatile compounds, which could contribute to the budget of organic
aerosols, would be produced through this mechanism. So along with the gas phase products,
the liquid phase can be analyzed by HPLC-MS method, which definitely would help to clarify
the detailed mechanism of this process. Further important experiments that can contribute for
the clarification of the mechanism would be the extraction of humic substances from the dust
samples and its structure identification by means of state of the art methods like NMR and IR.
This would give a clue of monomer units which can be used as model substances.
There are very few studies on airborne microorganisms from arid regions. So their
determination in dust samples from the source regions and in samples that have undergone
aeolian transport would give important results about the airborne microorganisms which are
responsible for the production of methyl chloride and isoprene. Also, it would give a hint
whether this process could take place in every soil environment, and a possibility to estimate
the total contribution of this process to the atmospheric isoprene and methyl chloride budget.
The sample from Gobi Desert produced higher amounts of isoprene compared to the
sample from Sahara Desert. It suggests the need for experiments with different types of
samples from different regions of the world for evaluating the contribution of both abiotic and
biotic production mechanisms of volatile and semi-volatile organic compounds.
The natural dust samples from arid and semi-arid regions are proved to be important
contributors to the production and emission of volatile organic compounds to the atmosphere
and it urges further studies in this field.
74
8
REFERENCES
Alicke, B., et al. (1999), Iodine oxide in the marine boundary layer, Nature, 397, 572-573.
Altieri, K. E., et al. (2006), Evidence for oligomer formation in clouds: Reactions of isoprene
oxidation products, Environmental Science & Technology, 40, 4956-4960.
Andreae, M. O., et al. (1996), Methyl halide emissions from savanna fires in southern Africa,
Journal of Geophysical Research, 101, 23603-23613.
Andreae, M. O., and P. Merlet (2001), Emissions of trace gases and aerosols from biomass
burning, Global Biogeochemical Cycles, 15, 955-966.
Atkinson, R. (1990), Gas-phase tropospheric chemistry of organic compounds: a review,
Atmospheric Environment, 24A, 1-41.
Atkinson, R. (1997), Gas-phase tropospheric chemistry of volatile organic compounds: 1.
Alkanes and alkenes, Journal of Physical Chemical Reference Data, 26, 215-290.
Atkinson, R. (2000), Atmospheric chemistry of VOCs and NOx, Atmospheric Environment,
34, 2063-2101.
Atkinson, R., and J. Arey (2003a), Atmospheric degradation of volatile organic compounds,
Chemical Reviews, 103, 4605-4638.
Atkinson, R., and J. Arey (2003b), Gas-phase tropospheric chemistry of biogenic volatile
organic compounds: a review, Atmospheric Environment, 37, S197-S219.
Ayers, G. P., and R. W. Gillett (1988), Isoprene emissions from vegetation and hydrocarbon
emissions from bushfires in tropical Australia, Journal of Atmospheric Chemistry, 7, 177-190.
Barrie, L. A., et al. (1988), Ozone destruction and photochemical reactions at polar sunrise in
the lower Arctic atmosphere, Nature, 334, 138-141.
Bell, N., et al. (2002), Methyl iodide: atmospheric budget and use as a tracer of marine
convection in global models, Journal of Geophysical Research, 107, 4340.
Bertman, S. B., and J. M. Roberts (1991), A PAN analog from isoprene photooxidation,
Geophysical Research Letters, 18, 1461-1464.
75
Böge, O., et al. (2006), Formation of secondary organic particle phase compounds from
isoprene gas-phase oxidation products: An aerosol chamber and field study, Atmospheric
Environment, 40, 2501-2509.
Bonnet, S., and C. Guieu (2004), Dissolution of atmospheric iron in seawater, Geophysical
Research Letters, 31, doi:10.1029/2003GL018423.
Bonsang, B., et al. (1992), Evidence for marine production of isoprene, Geophysical Research
Letters, 19, 1129-1132.
Broadgate, W. J., et al. (1997), Seasonal emissions of isoprene and other reactive hydrocarbon
gases from the ocean, Geophysical Research Letters, 24, 2675-2678.
Butler, J. H. (2000), Better budgets for methyl halides? Nature, 403, 260-261.
Caquineau, S., et al. (1998), Saharan dust: Clay ratio as a relevant tracer to assess the origin of
soil-derived aerosols, Geophysical Research Letters, 25, 983-986.
Carlton, A. G., et al. (2006), Link between isoprene and secondary organic aerosol (SOA):
Purivic acid oxidation yields low volatility organic acids in clouds, Geophysical Research
Letters, 33.
Carpenter, L. J. (2003), Iodine in the marine boundary layer, Chemical Reviews, 103, 49534962.
Carpenter, L. J., et al. (1999), Short-lived alkyl iodides and bromides at Mace Head, Ireland:
Links to biogenic sources and halogen oxide production, Journal of Geophysical Research,
104, 1679-1689.
Cheburkin, A. K., and W. Shotyk (1996), An Energy-dispersive Miniprobe Multielement
Analyzer (EMMA) for direct analysis of Pb and other trace elements in peats, Fresenius
Journal of Analytical Chemistry, 354, 688-691.
Chen, X., et al. (1998), Measurement of the organic nitrate yield from OH reaction with
isoprene, Journal of Geophysical Research, 103, 25563-25568.
Claeys, M., et al. (2004a), Formation of secondary organic aerosols through photooxidation of
isoprene, Science, 303, 1173-1176.
76
Claeys, M., et al. (2004b), Formation of secondary organic aerosols from isoprene and its gasphase oxidation products through reaction with hydrogen peroxide, Atmospheric
Environment, 38, 4093-4098.
Clerbaux, C., and D. M. Cunnold (2006a), Long lived compounds, in Scientific Assessment of
Ozone Depletion: 2006, Global Ozone Research and Monitoring Project-Report No. 50,
edited, pp. 1.15-11.18.
Clerbaux, C., and D. M. Cunnold (2006b), Long lived compounds, in Scientific Assessment of
Ozone Depletion: 2006, Global Ozone Research and Monitoring Project-Report No., edited,
pp. 1.15-11.18.
Cleveland, C. C., and J. B. Yavitt (1997), Consumption of atmospheric isoprene in soil,
Geophysical Research Letters, 24, 2379-2382.
Cleveland, C. C., and J. B. Yavitt (1998), Microbial consumption of atmospheric isoprene ina
temperate forest soil, Applied and Environmental Microbiology, 64, 172-177.
Cox, M. L., et al. (2004), Terrestrial sources and sinks of halomethanes near Cape Grim,
Tasmania, Atmospheric Environment, 38, 3839-3852.
DeBell, L. J., et al. (2004), Asian dust storm events of spring 2001 and associated pollutants
observed in New England by the Atmospheric Investigation, Regional Modeling, Analysis
and Prediction (AIRMAP) monitoring network, Journal of Geophysical Research, 109, doi:
10.1029/2003JD003733.
Dimmer, C. H., et al. (2001), Biogenic fluxes of halomethanes from Irish peatland
ecosystems, Atmospheric Environment, 35, 321-330.
Edney, E. O., et al. (2005), Formation of 2-methyl tetrols and 2-methylglyceric acid in
secondary organic aerosol from laboratory irradiated isoprene/NOx/SO2/air mixtures and their
detection in ambient PM2.5 samples collected in the Eastern United States, Atmospheric
Environment, 39, 5281-5289.
Erel, Y., et al. (2006), Trans boundary transport of pollutants by atmospheric mineral dust,
Environmental Science & Technology, 40, 2996-3005.
Fall, R., and S. D. Copley (2000), Bacterial sources and sinks of isoprene, a reactive
atmospheric hydrocarbon, Environmental Microbiology, 2, 123-130.
77
Grossenbacher, J. W., et al. (2001), Measurements of isoprene nitrates above a forest canopy,
Journal of Geophysical Research, 106, 24429-24438.
Guenther, A., et al. (1995), A global model of natural volatile organic compound emissions,
Journal of Geophysical Research, 100, 8873-8892.
Guenther, A., et al. (2006), Estimates of global terrestrial isoprene emissions using MEGAN
(Model of Emissions of Gases and Aerosols from Nature), Atmospheric Chemistry and
Physics Discussions, 6, 107-173.
Guieu, C., and A. J. Thomas (1996), Saharan aerosols: from the soil to the ocean, in The
impact of desert dust across the Mediterranean, edited by S. Guerzoni and R. Chester, pp.
207-216, Kluwer Academic Publishers.
Hamilton, J. T. G., et al. (2003), Chloride methylation be plant pectin: An efficient
environmentally significant process, Science, 301, 206-209.
Happell, J. D., and D. W. R. Wallace (1996), Methyl iodide in the Greenland/Norwegian seas
and the tropical Atlantic Ocean: Evidence for photochemical production, Geophysical
Research Letters, 23, 2105-2108.
Harley, P. C., et al. (1999), Ecological and evolutionary aspects of isoprene emission from
plants, Oecologia, 118, 109-123.
Harper, D. B., and J. T. G. Hamilton (2003), The global cycles of the naturally-occurring
monohalomethanes, The Handbook of Environmental Chemistry, 3, 17-41.
Hebestreit, K., et al. (2000), Measurements of halogen oxides in the troposphere, Geophysical
Research Abstracts, 2, 1061.
Henze, D. K., and J. H. Seinfeld (2006), Global secondary organic aerosol from isoprene
oxidation, Geophysical Research Letters, 33, doi: 10.1029/2006GL025976.
Hoffmann, T., et al. (2001), Iodine oxide homogeneous nucleation: an explanation for coastal
new particle production, Geophysical Research Letters, 28, 1949-1952.
Holzinger, R., et al. (2002), Diurnal cycles and seasonal variation of isoprene and its
oxidation products in the tropical savanna atmosphere, Global Biogeochemical Cycles, 16,
1074.
78
Husar, R. B. (2004), The handbook of environmental chemistry, 277-294 pp., SpringerVerlag, Berlin Heidelberg.
Janson, R., and C. De Serves (1998), Isoprene emissions from boreal wetlands in
Scandinavia, Journal of Geophysical Research, 103, 25513-25517.
Jickells, T. D., et al. (2005), Global iron connections between desert dust, ocean
biogeochemistry, and climate, Science, 308, 67-71.
Keene, W. C., et al. (1996), Reactive chlorine: A potential sink for dimethylsulfide and
hydrocarbons in the marine boundary layer, Atmospheric Environment, 30, i-iii.
Keppler, F. (2000), Abiotische Bildung leichtflüchtiger Organohalogenverbindungen im
Boden, 85 pp, University of Heidelberg, Heidelberg.
Keppler, F., et al. (2005), De novo formation of chloroethyne in soil, Environmental Science
& Technology.
Keppler, F., et al. (2000), Halocarbons produced by natural oxidation processes during
degradation of organic matter, Nature, 403, 298-301.
Kesselmeier, J., and M. Staudt (1999), Biogenic volatile organic compounds (VOC): An
overview on emission, physiology and ecology, Journal of Atmospheric Chemistry, 33, 23-88.
Khalil, M. A. K., and R. A. Rasmussen (1999), Atmospheric methyl chloride, Atmospheric
Environment, 33, 1305-1321.
Khalil, M. A. K., and R. A. Rasmussen (2000), Soil-Atmosphere exchange of radiatively and
chemically active gases, Environ. Sci. & Pollut. Res., 7, 79-82.
Kleindienst, T. E., et al. (2006), Secondary organic carbon and aerosol yields from the
irradiations of isoprene and alpha-pinene in the presence of NOx and SO2, Environmental
Science & Technology, 40, 3807-3812.
Koppmann,
R.,
et
al.
(1993),
Distribution
of
methylchloride,
dichloromethane,
trichloroethane, and tetrachloroethene over the north and south Atlantic, Journal of
Geophysical Research, 98, 20517-20526.
Kroll, J. H., et al. (2006), Secondary organic aerosol formation from isoprene photooxidation,
Environmental Science & Technology, 40, 1869-1877.
79
Kuzma, J., et al. (1995), Bacteria produce the volatile hydrocarbon isoprene, Current
Microbiology, 30, 97-103.
Lane, T. E., and S. N. Pandis (2007), Predicted secondary organic aerosol concentrations from
the oxidation of isoprene in the East United States, Environmental Science & Technology.
Law, K. S., and W. T. Sturges (2006), Halogenated very short-lived substances, in Scientific
Assessment of Ozone Depletion: 2006, Global Ozone Research and Monitoring ProjectReport No. 50, edited, pp. 2.1-2.57.
Lee-Taylor, J., and K. R. Redeker (2005), Reevaluation of global emissions from rice paddies
of
methyl
iodide
and
other
species,
Geophysical
Research
Letters,
32,
doi:10.1029/2005GL022918.
Lee-Taylor, J. M., and G. P. Brasseur (2001), A preliminary three-dimensional global model
study of atmospheric methyl chloride distributions, Journal of Geophysical Research, 106,
34221-34233.
Li, H.-J., et al. (2001), Distribution of methyl chloride, methyl bromide, and methyl iodide in
the marine boundary air over the western Pacific and southeastern Indian Ocean, Geochemical
Journal, 35, 137-144.
Li, X., et al. (1996), Dominance of mineral dust in aerosol-scattering in the North Atlantic
trade winds, Nature, 380, 416-419.
Lim, H.-J., et al. (2005), Isoprene forms secondary organic aerosol through cloud processing:
Model simulations, Environmental Science & Technology, 39, 4441-4446.
Limbeck, A., et al. (2003), Secondary organic aerosol formation in the atmosphere via
heterogeneous reaction of gaseous isoprene on acidic particles, Geophysical Research Letters,
30, doi:10.1029/2003GL017738.
Liss, P., et al. (2005), Ocean fertilzation with iron: effects on climate and air quality, Tellus,
57B, 269-271.
Lobert, J. M., et al. (1999), Global chlorine emissions from biomass burning: Reactive
Emissions Inventory, Journal of Geophysical Research, 104, 8373-8389.
Lovelock, J. E., et al. (1973), Halogenated hydrocarbons in and over the Atlantic, Nature,
241, 194-196.
80
Maciejczyk, K. (2005), Modellversuche zur natürlichen abiotischen Bildung von
Organohalogenen, Universität Heidelberg, Heidelberg.
Manley, S. L., and M. N. Dastoor (1988), Methyl iodide production by kelp and associated
microbes, Marine Biology, 98, 477-482.
Manley, S. L., et al. (2006), Coastal salt marshes as global halide sources from determinations
of
intrinsic
production
by
marsh
plants,
Global
Biogeochemical
Cycles,
20,
doi:10.1029/2005GB002578.
Matsunaga, S., et al. (2002), In situ measurement of isoprene in the marine air and surface
seawater from the western North Pacific, Atmospheric Environment, 36, 6051-6057.
McAnulla, C., et al. (2001), Methyl chloride utilising bacteria are ubiquitous in the natural
environment, FEMS Microbiology Letters, 201, 151.
McCarthy, M. (2001), Dust clouds implicated in spead of infection, The Lancet, 358, 478.
McCulloch, A., et al. (1999), Global emissions of hydrogen chloride and chloromethane from
coal combustion, incineration and industrial activities: Reactive Chlorine Emissions
Inventory, Journal of Geophysical Research, 104, 8391-8403.
Molina, M. J., and F. S. Rowland (1974), Stratospheric sink for chlorofluoromethanes:
chlorine atom-catalysed destruction of ozone, Nature, 249, 810-812.
Monteil, M. A. (2002), Dust clouds and spread of infection, The Lancet, 359, 81.
Montzka, S. A., et al. (1993), Isoprene and its oxidation products, methyl vinyl ketone and
methacrolein, in the rural troposphere, Journal of Geophysical Research, 98, 1101-1111.
Moore, R. M. (1994), Photochemical production of methyl iodide in seawater, Journal of
Geophysical Research, 99, 16415-16420.
Moore, R. M. (2000), The solubility of a suite of low molecular weight organochlorine
compounds in seawater and implications for estimating the marine source of methyl chloride
to the atmosphere, Chemosphere: Global Change Science, 2, 95-99.
Moore, R. M., and W. Groszko (1999), Methyl iodide distribution in the ocean and fluxes to
the atmosphere, Journal of Geophysical Research, 104, 11163-11171.
Moore, R. M., et al. (1996), Ocean-atmosphere exchange of methyl chloride: Results from
NW Atlantic and Pacific Ocean studies, Journal of Geophysical Research, 101, 28529-28538.
81
Moore, R. M., et al. (2005), A pilot study of methyl chloride emissions from tropical woodrot
fungi, Chemosphere, 58, 221-225.
Moore, R. M., et al. (1994), Production of isoprene by marine phytoplankton cultures,
Geophysical Research Letters, 21, 2507-2510.
Müller, G., and M. Gastner (1971), The "Karbonat-Bombe" a simple device for the
determination of the carbonate content in sediments, soils, and other materials, Neues Jahrb.
Mineral. Monatsh., 10, 466-469.
Nouaime, G., et al. (1998), Sequential oxidation products from tropospheric isoprene
chemistry: MACR and MPAN at a NOx-rich forest environment in the southeastern United
States, Journal of Geophysical Research, 103, 22463-33471.
Oram, D. E., and S. A. Penkett (1994), Observations in eastern England of elevated methyl
iodide concentrations in air of Atlantic origin, Atmospheric Environment, 28, 1159-1174.
Palmer, P. I., and S. L. Shaw (2005), Quantifying global marine isoprene fluxes using
MODIS
chlorophyll
observations,
Geophysical
Research
Letters,
32,
doi:
10.1029/2005GL022592.
Paulson, S. E., and J. H. Seinfeld (1992), Development and evaluationof a photooxidation
mechanism for isoprene, Journal of Geophysical Research, 97, 20703-20715.
Pegoraro, E. (2004), Environmental control of isoprene emission: from leaf to canopy scale,
PhD Thesis, The University of Edinburgh.
Pierce, T., et al. (1998), Influence of increased isoprene emissions on regional ozone
modeling, Journal of Geophysical Research, 103, 25611-25629.
Pierotti, D., et al. (1990), Isoprene and its oxidation products: Methacrolein and methyl vinyl
ketone, Journal of Geophysical Research, 95, 1871-1881.
Prospero, J. M., et al. (2005), Interhemispheric transport of viable fungi and bacteria from
Africa to the Caribbean with soil dust, Aerobiologica, 21, 1-19.
Rasmussen, R. A. (1970), Isoprene: Identified as a forest-type emission to the atmosphere,
Environmental Science & Technology, 4, 667-671.
Redeker, K. R., et al. (2002), Interfield and intrafield variability of methyl halide emissions
from rice paddies, Global Biogeochemical Cycles, 16, doi:10.1029/2002GB001874.
82
Redeker, K. R., and R. J. Cicerone (2004), Environmental controls over methyl halide
emissions
from
rice
paddies,
Global
Biogeochemical
Cycles,
18,
doi:10.1029/2003GB002092.
Redeker, K. R., et al. (2004), Ectomycorrhizal fungi: A new source of atmospheric methyl
halides? Global Change Biology, 10, 1009-1016.
Redeker, K. R., et al. (2000), Emissions of methyl halides and methane from rice paddies,
Science, 290, 966-969.
Rhew, R. C., et al. (2002), Environmental and biological controls on methyl halide emissions
from southern California coastal salt marshes, Biogeochemistry, 60, 141-161.
Rhew, R. C., et al. (2000), Natural methyl bromide and methyl chloride emissions from
coastal salt marshes, Nature, 403, 292-295.
Richter, U. (2003), Factors influencing methyl iodide production in the ocean and its flux to
the atmosphere, 117 pp, University of Kiel, Kiel.
Roberts, J. M., et al. (1998a), Measurement of alkyl nitrates at Chebogue Point, Nova Scotia
during the 1993 North Atlantic Regional Experiment (NARE) intensive, Journal of
Geophysical Research, 103, 13569-13580.
Roberts, J. M., et al. (2004), Measurement of peroxycarboxylic nitric anhydrides (PANs)
during the ITCT 2K2 aircraft intensive experiment, Journal of Geophysical Research, 109.
Roberts, J. M., et al. (2002), Ground-based measurements of peroxycarboxylic nitric
anhydrides (PANs) during the 1999 Southern Oxidants Study Nashville Intensive, Journal of
Geophysical Research, 107, 4554.
Roberts, J. M., et al. (2001), Observations of APAN during TexAQS 2000, Geophysical
Research Letters, 28, 4195-4198.
Roberts, J. M., et al. (2003), An examination of the chemistry of peroxycarboxylic nitric
anhydrides and related volatile organic compounds during Texas Quality Study 2000 using
ground-based measurements, Journal of Geophysical Research, 108, 4495.
Roberts, J. M., et al. (1995), Relationships between PAN and ozone at sites in eastern North
America, Journal of Geophysical Research, 100, 22821-22830.
83
Roberts, J. M., et al. (1998b), Measurement of PAN, PPN, and MPAN made during the 1994
and 1995 Nashville Insensives of the Southern Oxidant Study: Implications for regional ozone
production from biogenic hydrocarbons, Journal of Geophysical Research, 103, 2247322490.
Ruppert, L., and K. H. Becker (2000), A product study of the OH radical-initiated oxidation
of isoprene: formation of C5-unsaturated diols, Atmospheric Environment, 34, 1529-1542.
Sanadze, G. A. (1957), Emissions og organic matters by leaves of Robinia pseudacacia L.,
Soobsh Acad Nauk GSSR, 19, 83.
Schöler, H. F., and F. Keppler (2003), The Handbook of Environmental Chemistry, 63-84 pp.,
Springer-Verlag, Berlin Heidelberg.
Shallcross, D. E., and P. S. Monks (2000), New directions: A role for isoprene in biosphereclimate-chemistry feedbacks, Atmospheric Environment, 34, 1659-1660.
Shim, C., et al. (2005), Constraining global isoprene emissions with Global Ozone
Monitoring Experiment (GOME) formaldehyde column measurements, Journal of
Geophysical Research, 110.
Silver, G. M., and R. Fall (1991), Enzymatic synthesis of isoprene from dimethylallyl
diphosphate in aspen leaf extracts, Plant Physiology, 97, 1588-1591.
Silver, G. M., and R. Fall (1995), Characterization of aspen isoprene synthase, an enzyme
responsible for leaf isoprene emission to the atmosphere, Journal of Biological Chemistry,
270, 13010-13016.
Starn, T. K., et al. (1998a), Nighttime isoprene chemistry at an urban-impacted forest site,
Journal of Geophysical Research, 103, 22437-22447.
Starn, T. K., et al. (1998b), Observations of isoprene chemistry and its role in ozone
production at a semirural site during the 1995 Southern Oxidants Study, Journal of
Geophysical Research, 103, 22425-22435.
Surratt, J. D., et al. (2006), Chemical composition of secondary organic aerosol formed from
the photooxidation of isoprene, Journal of Physical Chemistry A, 110, 9665-9690.
Swap, R., et al. (1992), Saharan dust in the Amazon Basin, Tellus, 44B, 133-149.
84
Tait, V. K., et al. (1994), Measurements of methyl chloride in the northwest Atlantic, Journal
of Geophysical Research, 99, 7821-7833.
Tanimoto, H., and H. Akimoto (2001), A new peroxycarboxylic nitric anhydride identified in
the atmosphere: CH2=CHC(O)OONO2 (APAN), Geophysical Research Letters, 28, 28312834.
Tegen, I., et al. (1996), The influence on climate forcing of mineral aerosols from disturbed
soils, Nature, 380, 419-422.
Tiiva, P., et al. (2007), Isoprene emissions from boreal peatland microcosms; effects of
elevated ozone concentration in an open field experiment, Atmospheric Environment, 41,
3819-3828.
Tokarczyk, R., et al. (2003b), Methyl chloride and methyl bromide degradation in the
Southern Ocean, Geophysical Research Letters, 30, 1808.
Tokarczyk, R., et al. (2003a), Biological degradation of methyl chloride on coastal water,
Global Biogeochemical Cycles, 17, 1057.
Usher, C. R., et al. (2003), Reactions on mineral dust, Chemical Reviews, 103, 4883-4939.
van Donkelaar, A., et al. (2007), Model evidence for a significant source of secondary organic
aerosol from isoprene, Atmospheric Environment, 41, 1267-1274.
Varner, R. K., et al. (1999), Wetlands: a potentially significant source of atmospheric methyl
bromide and methyl chloride, Geophysical Research Letters, 26, 2433-2436.
Vogt, R. (1999), Iodine compounds in the atmosphere, 113-128 pp., Springer-Verlag, Berlin
Heidelberg.
Wagner, W. P., et al. (2000), Isoprene biosynthesis in Bacillus subtilis via the
methylerythritol phosphate pathway, Journal of Natural Products, 63, 37-40.
Wagner, W. P., et al. (1999), Three distinct phases of isoprene formation during growth and
sporulation of Bacillus subtilis, Journal of Bacteriology, 181, 4700-4703.
Watling, R., and D. B. Harper (1998), Chloromethane production by wood-rotting fungi and
an estimate of the global flux to the atmosphere, Mycol. Res., 102, 769-787.
85
Williams, J., et al. (2007), Possible evidence for a connection between methyl iodide
emissions
and
Saharan
dust,
Journal
of
Geophysical
Research,
112,
doi:10.1029/2005JD006702.
Williams, J., et al. (1997), Regional ozone from biogenic hydrocarbons deduced from
airborne measurements of PAN, PPN, and MPAN, Journal of Geophysical Research, 24,
1099-1102.
Wingenter, O. W., et al. (2004), Changing concentrations of CO, CH4, C5H8, CH3Br, CH3I,
and dimethyl sulfide during the Southern Ocean Iron Enrichment Experiment, Proceedings of
National Academy of Sciences of the United States of America, 101, 8537-8541.
Yeo, H.-G., and J.-H. Kim (2002), SPM and fungal spores in the ambient air of west Korea
during the Asian dust (Yellow sand) period, Atmospheric Environment, 36, 5437-5442.
Yokouchi, Y., et al. (2002), Strong emission of methyl chloride from tropical plants, Nature,
416, 163-165.
Yokouchi, Y., et al. (1999), Isoprene in the marine boundary layer (Southeast Asian Sea,
eastern Indian Ocean, and Southern Ocean): Comparison with dimethyl sulfide and
bromoform, Journal of Geophysical Research, 104, 8067-8076.
Yokouchi, Y., et al. (1997), Distribution of methyl iodide, ethyl iodide, bromoform, and
dibromomethane over the ocean (east and southeast Asian seas and the western Pacific),
Journal of Geophysical Research, 102, 8805-8809.
Yokouchi, Y., et al. (2000), A strong source of methyl chloride to the atmosphere from
tropical coastal land, Nature, 403, 295-298.
Yokouchi, Y., et al. (2001), Atmospheric methyl iodide: high correlation with surface
seawater temperature and its implications on the sea-to-air flux, Journal of Geophysical
Research, 106, 12661-12668.
Yoshida, Y., et al. (2004), A three-dimensional global model study of atmospheric methyl
chloride
budget
and
distributions,
Journal
of
Geophysical
Research,
109,
doi:10.1029/2004JD004951.
Yvon-Lewis, S. A., et al. (2004), Methyl bromide and methyl chloride in the Southern Ocean,
Journal of Geophysical Research, 109, doi:10.1029/2003JC001809.
86
9
APPENDIX
Sahara-Dust-Sample
after 45 days
Figure 40. Dust sample from Sahara with seawater in headspace glass vial.
87
Figure 41. Mass spectrum and chromatogram of isoprene.
88
Figure 42. Mass spectrum and chromatogram of methyl iodide.
89
Figure 43. Mass spectrum and chromatogram of methylene chloride.
90
Figure 44. Mass spectrum and chromatogram of methyl chloride.
91
MEASURED DATA
Table 13. Water and organic content of the samples
No.
Sample
Water
content, %
Organic
content, %
1
Sahara 1
0,90
3,34
2
Sahara 2
1,55
4,42
3
Sahara 2 > 0.063
1,53
2,81
4
Sahara 2 > 0.125
1,62
3,84
5
Sahara 3
0,60
2,54
6
Sahara 4
1,22
3,67
7
Sahara 4 > 0.063
1,31
2,98
8
Sahara 4 > 0.125
0,95
2,68
9
Sahara 4 > 0.2
0,37
1,06
10
Sahara 5
0,91
3,25
11
Sahara 6
0,86
3,41
12
Sahara 7
0,63
2,85
13
Sahara 8
1,06
3,71
14
Sahara 8 > 0.063
1,36
2,33
15
Sahara 8 > 0.125
1,24
2,26
16
Sahara 8 > 0.2
1,05
2,33
17
Sahara 8 > 0.315
1,16
3,08
18
Sahara 9
0,50
2,61
19
Cape Verde
2,98
7,71
20
Lanzarote 6
2,45
6,31
21
Mongolia 1
0,35
0,38
22
Mongolia 2
1,07
3,59
92
Table 14. Dissolved organic and inorganic carbon contents of the samples
No.
Samples
DOC, ppm
(mg/kg)
IC, ppm
(mg/kg)
1
Sahara 1
34,35
46,29
2
Sahara 2
34,08
54,66
3
Sahara 2 > 0.063
36,30
50,10
4
Sahara 2 > 0.125
49,62
73,41
5
Sahara 3
30,75
38,16
6
Sahara 4
41,55
41,91
7
Sahara 5
42,18
47,40
8
Sahara 6
46,77
62,94
9
Sahara 7
34,29
36,93
10
Sahara 8
65,79
64,77
11
Sahara 8 > 0.063
51,87
48,54
12
Sahara 8 > 0.125
54,72
45,45
13
Sahara 8 > 0.2
61,05
50,61
14
Sahara 8 > 0.315
95,04
69,45
15
Sahara 9
21,66
36,33
16
Mongolia 2
91,50
99,81
17
Cape Verde
106,26
60,30
18
Lanzarote 6 < 0.02
101,70
250,10
19
Lanzarote 6
39,78
144,63
93
Table 15. Effects of Fe (II) and Fe (III) additions on the production of methyl iodide
Sample
Methyl iodide, ng
without iron addition
0,109
with addition of Fe (II)
0,116
with addition of Fe (III)
1,012
Table 16. Amount of Me2Cl2 produced through the interaction of 4 g dust sample Sahara 4
with seawater
Time, hours
Methylene chloride, ng
0,33
0,193
0,33
0,285
0,33
0,285
0,66
0,383
0,66
0,358
0,66
0,318
1
0,263
1
0,268
1
0,235
24
0,486
24
0,514
24
0,427
48
0,470
48
0,415
48
0,479
72
0,369
72
72
0,436
0,426
94
Table 17. Dependence of Me2Cl2 concentration on reaction time and medium
Methylene chloride, ng
Time, days
Sahara 4
seawater
Sahara 4
distilled water
Sterilized Sahara 4
distilled water
20 min
0,454
0,438
0,446
1
0,318
0,359
2
0,198
0,307
0,559
3
0,148
0,154
0,835
4
0,134
0,159
1,101
5
0,191
6
0,086
7
0,049
Table 18. Dependence of isoprene amount on the dust amount
Dust sample CV, g
Isoprene, ng
0,05
0,014
0,1
0,020
0,5
0,063
1
0,184
2
0,438
4
0,931
95
Table 19.
Amount of isoprene produced through the interaction of dust samples with
seawater
Time, hours
Isoprene, ng
4 g Sahara 4
1 g Sahara 5
0,083
0,030
0,013
0,083
0,032
0,007
0,33
0,046
0,012
0,33
0,039
0,008
0,33
0,016
0,014
0,33
0,022
0,010
0,33
0,020
0,009
0,66
0,040
0,009
0,66
0,062
0,009
0,66
0,017
1
0,044
0,006
1
0,015
6
0,013
6
0,017
12
0,027
12
0,027
24
0,154
24
0,167
24
0,172
24
0,199
48
0,317
0,075
48
0,368
0,075
48
0,066
72
0,398
72
0,361
72
0,376
96
Table 20. Dependence of isoprene concentration on reaction time and medium
Time,
days
Isoprene, ng
Sahara 4
seawater
Sahara 4
distilled water
Mongolia 2
seawater
Mongolia 2
distilled water
1
0,484
0,250
0,121
1,344
1
0,484
0,265
0,128
1,384
1
0,484
0,249
0,121
1,494
2
0,969
0,185
2
0,969
0,176
2
0,969
0,171
3
1,453
3
1,453
3
0,135
0,120
4
1,938
4
1,938
0,301
0,106
0,288
4
0,099
0,335
5
0,094
0,090
0,155
5
0,084
0,085
0,144
5
0,070
0,157
6
2,907
0,083
0,148
6
2,907
0,000
0,140
6
0,144
7
3,391
0,065
0,173
7
3,391
0,058
0,157
7
3,391
0,161
8
3,876
0,148
8
3,876
0,127
8
3,876
0,135
9
4,360
97
Table 21.
Amount of methyl chloride produced through the interaction of dust samples
with seawater
Time, hours
0,083
Methyl chloride, ng
4 g Sahara 4
1 g Sahara 5
0,296
0,325
0,083
0,179
0,083
0,208
0,33
0,676
0,33
0,472
0,66
0,256
1
0,369
1
0,262
1
0,231
0,233
0,175
3
0,431
3
0,324
6
0,378
6
0,499
12
0,733
12
0,616
24
1,394
0,757
24
1,559
0,958
24
1,629
0,788
24
1,727
0,740
48
2,005
1,270
48
1,894
1,440
48
1,440
1,025
48
2,014
72
1,984
72
1,832
72
1,852
72
1,878
98
Table 22. Dependence of methyl chloride concentration on reaction time
Time, days
Methyl chloride, ng
Sahara 4
1
1,771
1
1,740
1
1,607
2
1,599
2
1,477
2
1,154
3
1,071
3
1,189
4
1,000
4
1,094
Mongolia 2
5
15,017
5
12,470
5
12,185
6
0,756
11,250
6
0,532
9,083
6
8,977
7
0,697
8,322
7
0,717
7,020
7
0,532
6,339
8
0,691
8
0,527
8
0,509
9
0,548
99
ACKNOWLEDGEMENTS
First of all, I would like to thank Prof. Dr. H. F. Schöler for his supervision and for giving me
the possibility to do this work. I really appreciate that.
I thank Prof. Müller for genereously agreeing to be one of the referees.
A special thank goes to Tineke, a coordinator of IMPRS in Mainz for all her support.
Many thanks to my PAC members, Jonathan Williams, John Crowley and Roland von
Glasow for their discussions.
I thank Stefan and Christian for their technical support and instructions with the instruments,
Andriy for XRF analysis and Dagmar for C/S analysis.
I would like to thank my group members, Gerhard, Stefan, Alke, Kirsten, Isabelle and Klaus
for always helping me and sharing their knowledge with me.
I thank Torsten for helping with administrative stuff.
Many thanks to IPP, Roswitha, Francisco and Tanya for supporting language courses,
summer schools and conferences.
Special thanks to Dr. Yahya and Basem.
I thank everyone who read and corrected my work.
Finally, I would like to thank my family for their patience and support.
Erklärung gem. § 8 (3) b) und c) der Promotionsordnung
der Naturwissenschaftlich-Mathematischen Gesamtfakultät
der Ruprecht-Karls-Universität Heidelberg
Ich erkläre hiermit, dass ich die vorgelegte Dissertation selbst verfasst und mich keiner
anderen als der von mir ausdrücklich bezeichneten Quellen und Hilfen bedient habe.
Außerdem erkläre ich hiermit, dass ich an keiner anderen Stelle ein Prüfungsverfahren
beantragt bzw. die Dissertation in dieser oder anderer Form bereits anderweitig als
Prüfungsarbeit verwendet oder einer anderen Fakultät als Dissertation vorgelegt habe.
Ariunaa Batsaikhan
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement