Measurements of CF3-containing compounds in the background atmosphere using gas
chromatography/high resolution mass spectrometry
by John Marcus Prins
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Chemistry
Montana State University
© Copyright by John Marcus Prins (2001)
Abstract:
In order, to accurately assess the environmental effects of halogenated compounds in the atmosphere
accurate methods for characterizing and quantitating these compounds in the atmosphere have been
developed. Several years ago a method for the detection of CF3-containing compounds in the
background atmosphere using GC / HRMS was developed by a former member of our research group,
Mark Engen. Recently, improvements to this method were developed by our laboratory. This improved
method replaced the GS-Q porous layer open tubular (PLOT) column previously used with a 30-m x
0.32-mm GS-GasPro PLOT column. For this exceedingly volatile set of compounds the GS-GasPro
column provided improved peak shapes, better signal-to-noise responses and no coelution of
compounds. These improvements enabled eleven CF3 - containing compounds to be detected in
background air, including CF4 (FC-14), C2F6 (FC-116), CF3Cl (CFC-13), CF3H (HFC-23), CF3Br
(Halon-1301), C3F8 (FC-218), CF3CF2Cl (CFC-115), CF3CHF2 (HFC-125), CF3CH3 (HFC-143a),
CF3CH2F (HFC-134a), and CF3CFCl2 (CFC-114a). Three of these compounds had never before been
detected in background air, to our knowledge.
In this study, this improved method for detecting CF3-Containing compounds in air was used to
perform quantitative determinations for each compound mentioned above in the background
atmosphere of Montana, with the use of a large volume standard-additions technique. This method was
then applied to a collection of historic air samples in order to monitor the concentrations of nine of
these compounds over the past several decades. These quantitative determinations were used to
calculate concentrations for each compound. Then, annually averaged concentrations were used to
estimate an average annual growth rate for each compound. Finally, emission estimates were made for
each compound over a specified time period using a simple mathematical model. MEASUREMENTS OF CF3-CONTAINING COMPOUNDS IN THE BACKGROUND
ATMOSPHERE USING GAS CHROMATOGRAPHY / HIGH RESOLUTION MASS
SPECTROMETRY
by
John Marcus Prins
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Chemistry
MONTANA STATE UNIVERSITY
Bozeman, Montana
March 200-1
11
APPROVAL
of a thesis submitted by
John M. Prins
This thesis has been read by each member of the thesis committee and has been
found to be satisfactory regarding content, English usage, format, citations, bibliographic
style, and consistency, and is ready for submission to the College of Graduate Studies.
Eric P. Grimsrud
Approved for the Department of Chemistry
Paul A. Grieco
Date
Approved for the College of Graduate Studies
Bruce McLeod
Date
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a master’s degree
at Montana State University, I agree that the Library shall make it available to borrowers
under rules of the Library.
If I have indicated my intention to copyright this thesis by including a copyright notice
page, copying is allowable only for scholarly purposes, consistent with “fair use” as
prescribed in the U.S. Copyright Law. Requests for permission for extended quotation
from or reproduction of this thesis in whole or in parts may be granted only by the
copyright holder.
iv
TABLE OF CONTENTS
1. INTRODUCTION................. ,....................... ............................... ............................. I
Atmospheric Chemistry..............................................................
8
8
Regions of the Atmosphere..................
Troposphere.....................................................
8
Stratosphere......... ...............................................................
11
Polar Stratospheric Clouds...................
14
Mesosphere...............................................................................................................16
2. EXPERIMENTAL.......................................................................................................18
Sample Introduction System......................................................................................18
Gas Chromatograph.................................................................................................. 20
Mass Spectrometer...............................
20
Single Ion Monitoring Routine................................................................................. 22
23
Chromatograph..... .......................
Quantitative Determinations.......................
26
3. RESULTS AND DISCUSSION................................................................................. 29
Results of Quantitative Determinations.......................:...................... .................... 29
Historic Air Samples...............
29
Historic Concentrations of Compounds .:......................
...30
Growth Rates ...:.................... ;............................................................................... 53
Emission Estimates................................................................................................... 66
4. CONCLUSIONS........................................................................................................ 70
BIBLIOGRAPHY....................................
73
APPENDIX A: Electron Impact Mass Spectra
79
LIST OF TABLES
Table
Page
1. Method reproducibility test. Data obtained by running ten experiments in
succession and then calculating the relative standard deviation (RSD) for
each p eak........................................................................................ ........................ 6
2. Peak areas obtained from chromatograms of samples taken simultaneously
from inside and outside the standard-additions vessel....... .................................. 27.
3. Lists results for average annual growth rate for CFg-containing compounds.
Average annual growth rate determined from 1977-1998, except for HFCs
(a), which were determined from 1990-1998 ...................................
4. Characteristics of atmospheric halocarbons. * An atmospheric lifetime
was not found in the literature for this compound, however the lifetime of
this compound is thought to be approximately 10,000 years since chemical
composition and structure is similar to C2F6......................................................... 67
5. Halocarbon estimates (x IO5 Kg/ 2-years) derived from annual means of
atmospheric measurements. (NA) Data was not available for calculations
over that time period. (-) Emission estimates for these compounds on
these years were done over a 12-month period and listed in Table 6 ....................68
6. Emission estimates(x IO5 Kg/ year) for CF3CFH2 and CF3CH3 from
1989-1996............................................................................................................... 69
66
vi
LIST OF FIGURES
Figure
Page
1. Shows single ion monitoring gas chromatograms when the CF3+ ion is
monitored at the different resolution settings indicated for each experiment.
Check marks indicate that are not due to actual CF3-Containing compounds...........5
2. Temperature profile of the atmosphere (32)............................................................... 9
3. Schematic diagram of sample introduction system. Valves (A,B,C, & D);
Metering Valve (MV); Drying Tube (DT); Particle Filter (PF);
Six Port Valve (6PV); Freeze Out Loop (FOL); Ten Liter Expansion
. Volume (IO-L EV); Baratron Pressure Gauge (BPG)...........................................19
4. Schematic diagram of ZAB-2F high resolution mass spectrometer....................... 21
5. CF3"1"(m/z = 68.9952) single ion chromatogram of a background air
sample at a mass resolution of 2200. The peak identities are;
I. CF4 (FC-14), 2. C2F6 (FC-116), 3. CF3Cl (CFC-13), 4. CF3H (HFC-23),
5. CF3Br (Halon-1301), 6. C3F8 (FC-218), 7. CF3CF2Cl,
8. CF3CF2H (HFC-125), 9. CF3CH3 (HFC-143a), 10. CF3CFH2 (HFC-134a),
and 11. CF3CFCl2 (CFC-114a)..............................................................................24
6. Sample of standard-additions experiment done for C2F6.,...................................... 28
7. Concentration of C2F6 from samples obtained from Cape Meares Oregon............32
8. Concentration of C2F6 from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)................................................................ 33
9. Concentration of C3F8 from samples obtained from Cape Meares Oregon........... 34
10. Concentration of C3F8 from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA) ............................................................. 35
11. Concentration of CF3CF2Cl from samples obtained from Cape Meares Oregon ....37
12. Concentration of CF3CF2Cl from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)................................................................. 38
vii
13. Concentration of CF3CFCI2 from samples obtained from Cape Meares Oregon....39
14. Concentration of CF3CFCl2 from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)................................................................. 40
15. Concentration of CF3Cl from samples obtained from Cape Meares Oregon.........41
16. Concentration of CF3Cl from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)........... ..................................................... 42
17. Concentration of CF3CFH2 from samples obtained from Cape Meares Oregon.....44
18. Concentration of CF3CFH2 from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA).................... ............................................ 45
19. Concentration of CF3CH3 from samples obtained from Cape Meares Oregon.......46
20. Concentration of CF3CH3from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)................................................................. 47
21. Concentration of CF3H from samples obtained from Cape Meares Oregon...........48
22. Concentration of CF3H from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA)................................................................. 49
23. Concentration of CF3Br from samples obtained from Cape Meares Oregon..........51
24. Concentration of CF3Br from samples obtained from Palmer Station Antarctica
(PSA) and Point Barrow Alaska (PBA).................................................................52
25. Concentrations for CF3Br from samples from Cape Meares Oregon (CMO)
(Northern Hemisphere). Concentration of CF3Br calculated by Fraser et al. From
samples collected from Cape Grim (CG) (Southern Hemisphere)........................54
26. Concentrations from CF3H from samples from Cape Meares Oregon (CMO)
(Northern Hemisphere). Trend line represents CF3H concentration data obtained
from Oram et al. from samples collected from Cape Grim (CG) (Southern
Hemisphere).................... ...................................................................................... 55
27. Concentration of CF3CFH2 from samples from Cape Meares Oregon (CMO)
(Northern Hemisphere). Trend line represents CF3CFH2 concentrations obtained
by Oram et al. from samples collected from Cape Grim (CG) (Southern
Hemisphere)..........................................................................................................56
viii
28. Concentration of CzFg. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA)............................................................. 57
29. Concentration of CFgCL Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA)............................................................ 58
30. Concentration of CF3H. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................59
31. Concentration of CFsBr. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA)........................................ ....................60
32. Concentration of CsFg. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................61
33. Concentration of CF3CF2CI. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................62
34. Concentration of CF3CH3. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................63
35. Concentration of CF3CFH2. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................64
36. Concentration of CF3CFCI2. Cape Meares Oregon (CMO); Point Barrow Alaska
(PBA); Palmer Station Antarctica (PSA).............................................................65
37. Electron Impact mass spectrum of CF4..................................................................80
38. Electron Impact mass spectrum of C2F6................................................................BI
39. Electron Impact mass spectrum of CF3CI..............................................................82
40. Electron Impact mass spectrum of CF3H...............................................................83
41. Electron Impact mass spectrum of CFsBr..............................................................84
42. Electron Impact mass spectrum of CsFg................................................................85
43. Electron Impact mass spectrum of CF3CF2CI........................................................86
44. Electron Impact mass spectrum of CF3CF2H.........................................................87
45. Electron Impact mass spectrum of CF3CH3...
88
ix
46. Electron Impact mass spectrum of CFgCFH2....................................................... 89
47. Electron Impact mass spectrum of CFgCFCl2.......................................................90
X
ABSTRACT
In order, to accurately assess the environmental effects of halogenated compounds in
the atmosphere accurate methods for characterizing and quantitating these compounds in
the atmosphere have been developed. Several years ago a method for the detection of
CF3-containing compounds in the background atmosphere using GC / HRMS was
developed by a former member of our research group, Mark Engen. Recently,
improvements to this method were developed by our laboratory. This improved method
replaced the GS-Q porous layer open tubular (PLOT) column previously used with a 30m x 0.32-mm GS-GasPro PLOT column. For this exceedingly volatile set of compounds
the GS-GasPro column provided improved peak shapes, better signal-to-noise responses
and no coelution, of compounds. These improvements enabled eleven CF3 - containing
compounds to be detected in background air, including CF4 (FC-14), CaFe (FC-116),
CF3Cl (CFC-13), CF3H (HFC-23), CF3Br (Halon-1301), C3F8 (FC-218), CF3CF2Cl
(CFC-115), CF3CHF2 (HFC-125), CF3CH3 (HFC-143a), CF3CH2F (HFC-134a), and
CF3CFCl2 (CFC-114a). Three of these compounds had never before been detected in
background air, to our knowledge.
In this study, this improved method for detecting CF3-Containing compounds in air
was used to perform quantitative determinations for each compound mentioned above in
the background atmosphere of Montana, with the use of a large volume standardadditions technique. This method was then applied to a collection of historic air samples
in order to monitor the concentrations of nine of these compounds over the past several
decades. These quantitative determinations were used to calculate concentrations for
each compound. Then, annually averaged concentrations were used to estimate an
average annual growth rate for each compound. Finally, emission estimates were made
for each compound over a specified time period using a simple mathematical model.
I
INTRODUCTION
The presence of halogenated compounds in the Earth’s atmosphere has been a
significant source of concern during the last two decades due to their potential for
stratospheric ozone destruction (1-4) and global warming (5-8). Chemical inertness has
made halogenated compounds industrially valuable as aerosol propellants, solvents,
refrigerants, blowing agents for plastic foam products, fire extinguishers, and in the
plasma etching process for the semiconductor industry (9-14).
However, use of
halogenated compounds leads ultimately to atmospheric release, where some may survive
for only a few years, while others will survive for more than 50,000 years (42).
Chlorofluorocarbon’s (CFCs)
and bromofluorocarbons
(BFCs)
are halogenated
compounds that decompose in the stratosphere, usually by photolysis, and the chlorine or
bromine atoms produced from this decomposition can participate in the catalytic
destruction of stratospheric ozone. Other halogenated compounds, like fluorocarbon’s
(FCs) are very efficient greenhouse gases, which means they' contribute to global
warming due to their radiative properties (3, 5). In response to warnings of stratospheric
ozone depletion and global warming, the production and use of many industrial halogencontaining compounds has been banned since 1987 in agreement with the Montreal
Protocol (29). For this reason, the use of hydrofluorocarbons (HFCs) as alternative, long
term, replacement compounds has increased significantly in the past several years. As
the concentrations of HFCs increase in the atmosphere, some concern is growing about
the safety, toxicology, and the environmental impacts that these compounds may have
2
(11). In order to accurately estimate the effectiveness of the restrictions placed on the use
of some halogen containing compounds and to monitor the growth of replacement
compounds, an accurate method for the detection and quantitative determinations of these
compounds needed to be developed.
Some of the most effective methods developed for the detection of halogenated
compounds in real air samples have involved separation by gas chromatography (GC)
followed by detection with either the electron capture detector (ECD) (15-17) or mass
spectrometer (MS) (18-24). Several years ago, a method for detecting CFg - containing
compounds in background air by gas chromatography / high-resolution mass
spectrometry (GC/HRMS) was reported by a former member of our research group,
Mark Engen (25). With this method, a whole air sample (355 atm cm3, I atm = 1.013 x
IO5 Pa) was pre-concentrated at liquid nitrogen temperature followed by introduction to a
gas chromatograph equipped with a GS-Q PLOT column 30 m x 0.32 mm. A dual sector high - resolution mass spectrometer operated in the single ion monitoring mode
was used for detection. Due to the large ion optic system of this mass spectrometer, high
ion throughput was obtained simultaneously with use of relatively high-mass resolution,
thereby providing very sensitive and specific detection of CF3 - containing compounds in
air.
A limitation to that method, however, was the relatively poor chromatographic
resolution obtained for these very volatile compounds using the GS-Q column, which
performed very well in the analysis of other less volatile components of the atmosphere.
O’Doherty et al. (57) recently described the use of a cyclodextrin PLOT column for
the separation of hydrohalocarbons and chlorofluorocarbons. They determined retention
3
times for 34 halogenated compounds (8 of which contain the CFg group) using GC with
either flame ionization detection (FID) or ECD. One ambient air analysis was also
shown using GC/ECD.
This chromatogram demonstrated the separation of 13
halogenated compounds (none of which contain the CFg group) commonly detected in
background air. Their report showed promise for the use of the cyclodextrin column in
real air analysis for many classes of compounds that are difficult to chromatograph.
Recently, an improved method for detecting and analyzing CFg-containing compounds in
the background atmosphere by GC / HRMS using a GS-GasPro (cyclodextrin) column
was developed by our laboratory (61). The GS-GasPro (cyclodextrin) column provides
superior chromatographic resolution of the CFg-containing compounds and with this
improved method eleven CFg-containing compounds were detected in all background air
samples (61).
A high-resolution mass spectrometer provides an advantage in the analysis of
atmospheric halogenated compounds, resulting from its ability to separate ions based on
their exact mass. This can be seen when looking at the exact masses of the components
of these halogenated compounds; 12C (Exact Mass = 12.000), 1H (Exact Mass = 1.00783),
19F (Exact Mass = 18.99840), 35Cl (Exact Mass = 34.96885), 37Cl (Exact Mass =
36.96590), 79Br (Exact Mass = 78.91834) and 81Br (Exact Mass = 80.91629). After
reviewing the exact masses listed above, it can be seen that molecules containing only
carbon and hydrogen atoms will have an exact mass that is somewhat larger than their
nom inal mass value (meaning their nearest integer value).
Whereas, molecules that
contain carbon and hologen atoms will have exact masses that are somewhat less than
4
their nominal mass values. Lower levels of mass resolution are insufficient to separate
hydrocarbon-based and halocarbon-based ions of the same nominal mass, thereby
creating a significant source of chemical interference and noise due to the hydrocarbons
present in the air samples and air sampling system. However, with a mass spectrometer
capable of moderate to high levels of resolution, these exact mass differences are large
enough to allow separation of hydrocarbon-based and halocarbon-based ions of the same
nominal mass, thereby eliminating this hydrocarbon complication (25). The ability of a
high-resolution mass spectrometer to eliminate hydrocarbon-based interference and noise
can be seen in Figure I. Several CF3+ single ion chromatograms obtained under different
resolution setting are shown in Figure I.
The check marks indicate peaks due to
hydrocarbon-based interferences. As can be seen when referring to Figure I, several
peaks are eliminated as the resolution is increased from 600 to 2300. The disappearance
of these peaks indicates that these peaks do not represent actual CF3-Containing
compounds but are due to some type of hydrocarbon with the same nominal mass.
When using higher mass resolutions, a difficulty in maintaining focus on the centroid
of exact mass of the ion of interest is encountered. This problem emerges as a result of
the narrow mass spectral peaks created with the use of higher resolutions, and the
residual drift of electric and magnetic sectors of the mass spectrometer. In order to
address this problem, the controlling fields of the mass spectrometer are continuously
calibrated (2 times per second) during the experiment. Continuous calibration of the
mass spectrometer allows for any level of mass resolution offered by the mass
spectrometer to be utilized with a high-level of response reproducibility (25). Method
5
File T e x t :Air 331 Can #1 5-25-00 R e s :1200
: : ------- -- u,v
icbuiuuon settings indicated for each expei
indicate peaks that are not due to actual CF]+ containing compounds.
6
reproducibility was investigated by analyzing the same air sample ten times in succession
followed by calculating the Relative Standard Deviation (RSD) of the peak areas for each
compound observed in this study. The results of this test are shown in Tablel below.
When referring to Table I, it can be seen that the peak areas for all compounds are
reproducible to within a maximum of 5% of the mean, many of which are significantly
less, except CF3H which is only reproducible to within approximately 13%.
Table I : Method reproducibility test. Data obtained by running ten experiments in
succession (R = 2200) and then calculating the relative standard deviation (RSD) for peak
areas.___________________________________________________________________
Method Reproducibility Test
Compound
RSD
0.9%
CF4
2.1%
CzFe
CF3Cl
1.8%
CF3H
12.6%
CF3Br
4.7%
C3Fg
&7%
CF3CF2Cl
2.1%
CF3CH3
5.0%
2.3%
CF3CFH2
2.0%
CFiCFCB
Since the atmospheric concentration of these halogenated compounds is very small,
great care needed to be taken to assure accurate results when making quantitative
determinations for these compounds. A large volume standard-additions approach was
used for our quantitative determinations. This approach is very useful when trying to
quantitate compounds with extremely low atmospheric concentrations, since a large
standard-additions vessel can assist enormously in diluting primary standards down to a
level that is roughly that of these compounds in the background atmosphere. Also, a
7
standard-additions approach can help to offset any sources of systematic error that may
possibly occur during the analysis (26).
The GC/HRMS system described above was used to analyze a collection of archived
air samples, which were collected over a twenty-year period. Air samples were collected
by Dr. Rei Rasmussen from the Oregon Graduate Institute, with whom we have been
working on this project. Nine CF3-Containing compounds were monitored in these air
samples and the concentration of each compound was calculated for each air sample.
This data was then used to estimate annually averaged growth rates and to estimate
emissions over a specified time period for each compound.
Interest in these compounds has been stimulated by concern over their persistance in
the atmosphere and how this could have significant environmental consequences. An
understanding that certain chlorofluorocarbon’s (CFCs) and bromofluorocarbon’s (BFCs)
could build up to levels in the stratosphere where significant depletion of ozone might
occur, has lead to their replacement by hydrofluorcarbon ’s (HFCs) in industrial
applications (11, 27). The global warming potentials of fluorocarbon’s (PCs), has lead to
the concern over the presence of these compounds in the atmosphere (26, 28, 30). In
order to understand how some of these compounds can seriously damage our
environment and why some compounds are considered more acceptable in the
atmosphere than others it is necessary to briefly discuss some basic atmospheric
chemistry as it pertains to the compounds under study.
8
Atmospheric Chemistry
Regions of the Atmosphere
The thin shell of gas that surrounds our planet is known as the atmosphere. The
atmosphere is divided into four regions based on its temperature profile (Figure 2) (32).
The region nearest to the Earth’s surface is the troposphere and is the region where most
of the events that we think of as weather occur, such as wind and precipitation. In the'
troposphere, the temperature generally decreases with increasing altitude and reaches a
minimum of about 215 K at about 12 km. Temperature extremes separate the different
regions of the atmosphere and these boundaries are denoted by the suffix - pause for each
region. Thus, the tropopause is the boundary for the troposphere. Above the tropopause,
the temperature increases with altitude reaching a maximum of about 275 K at about 50
lcm.
This next region is called the stratosphere.
Beyond the stratosphere are the
mesosphere and the thermosphere (31, 32).
Troposphere
Tropospheric chemistry is mostly governed by reactions with hydroxyl radicals.
Ozone (Os) photodissociation provides the primary source for the hydroxyl radicals in
this region.
Initially, ozone is photodissociated by solar ultraviolet radiation with
wavelengths less than about 320 nm and produces atomic oxygen (Reaction I). The
atomic oxygen then reacts with water molecules to produce hydroxyl radicals (Reaction
2) (28,31).
Os + hv (<320 nm) ->02 + 0
(I)
9
Thermosphere
------ Mesopause
Mesosphere
•Stratopause
Stratosphere
TropopuUse
Troposphere
Figure 2: Temperature profile of the atmosphere (32).
10
O + H2O ^ 2 OH
(2)
The ability to react with hydroxyl radicals greatly reduces a compound’s atmospheric
lifetime and limits its migration into the upper regions of the atmosphere. Among the
few organic substances that are not attacked by hydroxyl radicals are fully halogenated
compounds like CFCs and BFCs. The extreme chemical inertness of CFCs and BFCs
allows them to survive unchanged in the troposphere and eventually migrate to the
stratosphere, where they can cause damage to the ozone layer.
Therefore, suitable
replacement compounds that are sufficiently reactive in the troposphere to reduce their
atmospheric lifetimes and hinder their migration to the stratosphere have been substituted
for CFCs and BFCs in industrial applications.
CFCs and BFCs owe their inertness
towards reactions in the troposphere to being fully halogenated. Hydrocarbons, on the
other hand, contain one or more carbon-hydrogen bonds, which allows for reaction with
the hydroxyl radical. As a result of this characteristic, HFCs are top contenders for CFC
replacement. In addition, HFCs contain no chlorine or bromine and have no or little
ozone destruction potential. Also, they have much lower global warming potentials than
the compounds they are replacing. HFC-134a (CF2CH2F) is an example of an HFC
currently being used as a replacement for CFCs in a variety of industrial applications (2732).
A possible fate for HFC-134a in the troposphere is shown below (Reactions 3-7)
(60).
CF3CH2F + OH
CF3CFH- + O2
CF3CFH' + H2O
(3)
CF3CFHOO 1
(4)
CF3CFHOO- + NO ^ CF3CFHO + NO2
(5)
11
CF3CFHO +Oz -> CF3CFO + HO2
CF3CFO + H2O
CF3CO2H (Trifluoroacetic acid) + HF
(6a)
(6b)
CF3CFHO ' -> CF3- + HCOF
(7a)
CF3- ->CF20 + HF
(7b)
CF2O + H2O ^ CO2 + 2HF
(7c)
First, HFC-134a reacts with a hydroxyl radical to form a reactive radical CF3CFH
(Reaction 3). CF3CFH is then oxidized by a reaction with O2 (Reaction 4) followed by a
reaction with NO to form CF3CFO " (Reaction 5). CF3CFO " can again react with O2 or
decompose to produce a number of more polar products which can be removed from the
troposphere by rain (Reactions 6 & 7) (I I, 60). The result of reaction 3 is an atmospheric
lifetime for HFC-134a of approximately 16 years, a lifetime that is relatively short when
compared to CFC-12, which is one compound it replaces, that has an atmospheric
lifetime of 100 years.
As the concentration of HFCs in the atmosphere rises, some
concern is growing about how much HFCs contribute to global warming and how the fate
and toxicity of long-lived degradation products like trifluoroacetic acid will affect the
environment (47).
Stratosphere
The compounds of interest in this region of the atmosphere are CFCs and BFCs.
General Motors, initially developed chlorofluorocarbons in the 1930’s as a non-toxic
non-flammable refrigerant (31). In 1974, Sherwood Rowland and Mario Molina made the
first suggestion that CFCs would affect stratospheric ozone (O3) by becoming involved in
a catalytic cycle of ozone destruction (31). The use of CFCs for industrial purposes was
12
discontinued in 1987 in response to the Montreal Protocol, which was a meeting by many
countries of the world to discuss problems associated with prolonged use of these
chemicals. BFCs are no longer produced for this same reason. As mentioned above,
chemical inertness allows these compounds to remain stable in the troposphere and
eventually migrate to the stratosphere. Once in the stratosphere, these compounds are
destroyed by photolysis. Their decomposition produces chlorine atoms and bromine
atoms in this region, which become involved in the catalytic cycle for ozone destruction
(31):
The stratosphere contains about 90% of the atmosphere’s ozone (60). Ozone plays
and extremely important role in the Earth’s atmosphere, because it virtually absorbs all
solar ultraviolet radiation between wavelengths of about 240 and 290 nm (31). The
absence of ozone in this region would permit this high-energy radiation to be transmitted
to the Earth’s surface and damage living systems. Ozone is created when O atoms
formed in the stratosphere collide with O2 molecules (Reaction 8) (* indicates excess
energy).
O + O2 -> O3 *
(8)
Ozone molecules are unstable and are continuously being created and destroyed in a
cyclic process (Reactions 9a-9d).
Photodecomposition of ozone is just the reverse
reaction of its formation (Reactions 9c-9d).
O2 + hv
O + O2 + M
O3 + hv
O+O
O3 + M*
O2 + O
(9a)
(9b)
(9c)
13
0 + 0 + M ^ 0 2+ M*
(9d)
In the above equations, M represents a third chemical species, most likely N2 or O2, to
which some excess energy of the produced ozone molecule is transferred. If some of this
excess energy is not transferred to M, the excited ozone molecule would immediately
dissociate on formation (28, 31).
As CFCs diffuse into the stratosphere, they photodissociate as they are exposed to
high-energy radiation of wavelengths in the range 190 to 225 nm and produce free
chlorine atoms. The atomic chlorine then reacts with ozone to form chlorine monoxide
(CIO) and molecular oxygen (Reaction 10).
C l+ O3 ^ CIO+ O 2
(10)
If the CIO does not become tied up in reactions with other atmospheric constituents, it
<D
can react with itself or atomic oxygen to regenerate free Cl atoms (Reaction 11 'and/or
12).
CIO + CIO
O2+ 2C1
(11)
(12)
CIO + O -> Cl + O2
As can be seen when referring to reactions 10, 11, & 12, the chlorine atom acts as a
catalyst. So, once released into the atmosphere, the chlorine can participate in ozone
removal as long as it remains in the atmosphere.
Since most atmospheric ozone is
present in the stratosphere, the chlorine present in this region can have a large impact on
ozone destruction (28-32).
BFCs encounter an atmospheric fate similar to that of the CFCs. Being inert to the
reactions in the troposphere, BFCs can exist long enough reach the stratosphere,
14
photodissociate, and the released bromine can rapidly attack ozone by a series of catalytic
reactions (Reactions 13 & 14) (28).
Br + O 3 BrO + O2
BrO + BrO
2Br + 0?
O3 O2
or
B r+ O3 ^ BrO+ O2
BrO + HO2 -> HOBr + O2
HOBr + hv -> OH + Br
OH + CO + 0 ? HO? + CO?
CO + O3 4 CO2 + O2
(13)
(14)
Bromine monoxide (BrO) can also attack ozone through a catalytic cycle involving
chlorine in the form of CIO (Reaction 15) (59).
BrO + CIO -> Br + Cl + O2
Br + O3 BrO + O2
Cl + Oi ^ CIO + Q7
O3 + O3 O2 + O2 + O2
(15)
Since more inorganic bromine is in the reactive form in the atmosphere relative to
inorganic chlorine, bromine is thought to be a more efficient in the catalytic destruction
of ozone than an equivalent amount of chlorine (58).
Polar Stratospheric Clouds
One of the interesting details of ozone depletion, is that in the Antarctic it increases in
September and October which is the beginning of the Antarctic spring. During the
Antarctic winter, clouds form in the Antarctic stratospheres (similar phenomena also
occur in the Arctic to a lesser extent). Once regarded as only a beautiful and mysterious
phenomena, these Polar Stratospheric Clouds (PSCs) are now recognized as a key factor
15
in the springtime destruction of ozone and the formation of the ozone hole in that region.
Approximately 70 percent of Antarctic ozone is destroyed during the spring and occurs at
altitudes between 15 and 30 km (35, 37).
PSCs participate in ozone destruction by providing a site for a group of heterogeneous
reactions that interfere with the normal gas-phase chemistry in the Antarctic (37, 39).
Chlorine atoms supplied by the decomposition of CFCs, and other sources, destroy ozone
and form CIO (Reaction 10). Most of the chlorine in the stratosphere should be tied up in
relatively inert reservoir compounds HCl or CIONO2 (Reactions 16 & 17) (28, 33), which
do not react directly with ozone (28, 37).
Cl + CH4 ^ HCl + CH3
(16)
CIO + NO2
(17)
ClONO2
However, the low temperatures of the Antarctic winter (190-200 K), allows for the
formation of clouds in the stratosphere at an altitude of approximately 20 km. The cloud
particles are composed of condensed nitric acid, probably in the crystalline form of
EDSfO3S(H2O), nitric acid trihydrate (NAT) (33, 36, 37). Cloud formation leads to the
removal of NO2 from the atmosphere, which stops the removal of CIO via reaction 17
(36). In addition, PSCs provide a surface which can catalyze the reaction between the
two components of the chlorine reservoir, HCl and ClONO2, to form Cl2 and HNO3
(Reaction 18) (37, 39, 40).
HCl(s) + C10N02(g) ->PSC-> Cl2fe) + HNO3(S)
(18)
When sunlight reappears in the Antarctic region, Cl2 can photodissociate into reactive Cl
atoms and then react with ozone to form CIO and O2. Since cloud formation removes the
16
NO2 from the atmosphere, it cannot react with the CIO to form unreactive CIONO2. This
leads to enhanced ozone depletion, since' CIO reacts with itself to form CI2O2. This
molecule will photodissociate into O2 and free chlorine atoms, which are again free to
react with more ozone (Reactions 19 & 20)(28, 33).
CIO + CIO
Cl2O2
Cl2O2 + hv -> O2 + 2C1
'
(19)
(20)
Consequently, chlorine atoms are free to be reintroduced into a catalytic cycle that
destroys ozone before they can be converted into inert reservoir compounds (28, 33, 37).
Reactions involving BrO and CIO (Reaction 15) are thought to contribute approximately
30% to the observed Antarctic ozone hole (59).
In general, PSCs become a source of halogen atoms and help lead to the formation of
the Antarctic ozone hole. The Antarctic ozone hole contributes to a general world wide
decrease in stratospheric ozone, because warming temperatures and changing wind
currents break up the PSCs in the spring. The ozone-depleted Antarctic air becomes
mixed throughout the Southern Hemisphere resulting in an overall decrease of
stratospheric ozone levels (37).
Mesosphere
The mesosphere extends from the stratopause, at about 50 km, to the mesopause at
around 85 km (Figure I). The chemistry in this region that is of interest in this study,
involves perfluorinated compounds such as FCs.
FCs contain no chlorine and no
hydrogen atoms, so they are not susceptible to attack by the hydroxyl radical and will not
photodissociate even in the stratosphere. FCs are thought to be some of the more potent
17
greenhouse gasses emitted in large amounts, so their principle environmental effect could
be the trapping of out-going terrestrial infrared radiation, otherwise known as the green
house effect! These compounds have extremely long atmospheric lifetimes, on the order
of hundred’s of centuries, so once emitted into the atmosphere they exist there almost
indefinitely and this accounts for the concern over the growth of these compounds in the
atmosphere (14, 30, 34).
18
EXPERIMENTAL
Sample Introduction System
The whole air samples are introduced to the GC/HRMS system by use of an
associated air sampling system (Figure 3). Initially, an air sample is connected to the
sample introduction system. Then a rotary pump is used to evacuate the IO-L expansion
volume as well as the transfer lines to and from the freeze-out Ioof^ A metering valve
(MV) is then used to draw a sample volume (corrected to standard conditions) of 355 atm
cm3 through the freeze-out loop (OS-GasPro, 22 cm x 0.32 mm) cooled to liquid nitrogen
temperature and into the IO-L expansion volume. Moisture in the air sample is removed
prior to the freeze-out loop, to prevent the loop from becoming plugged with ice and to
avoid overloading the column with water, by passage through a short drying tube (DT)
containing Mg(ClOz)^ Experiments performed with and without the drying tube indicate
that no sample is lost by using the Mg(C102)4 drying tube. Also, a particle filter (PF),
located downstream from the drying tube, is used to remove any particles larger than 10um. The amount of sample in 355 atm cm3, is then derived by the pressure read on the
Baratron Pressure Gauge (BPG) and the known volumes of the IO-L expansion volume
and transfer lines. After collection of a sample on the freeze-out loop, a rotary pump is
used to remove most of the nitrogen and oxygen from the freeze-out loop. Finally, the
loop is rapidly warmed by a heat gun and its contents routed to the GC via a six-port
valve (6-PV) (25, 26). It should be noted however that a significant decrease in the CF4
response is observed as the sample introduction flow rate is increased from 40 ml/min to
19
Sample Introduction
CG In
to GC
air
sam ple
IO-L EV
pump
Figure 3: Schematic diagram of sample introduction system. Valves (A,B,C, & D);
Metering Valve (MV); Drying Tube (DT); Particle Filter (PT); Six Port Valve (6PV);
Freeze Out Loop (FOL); Ten Liter Expansion Volume (10-L EV); Baratron Pressure
Gauge (BPG).
20
120 ml/min. This indicates that at higher flow rates CF4 is not being trapped with unit
efficiency. At a flow rate of 40 ml/min the CF4 response is reproducible to less than a
2% deviation and no further increase in response is observed as the flow rate is lowered.
All other compounds showed no significant change in response as the sample
introduction flow rate was changed.
Gas Chromatograph
The gas chromatograph is a Hewlett-Packard, Model 5890 series II plus, which has
subambient temperature capability with liquid Na cooling.
The GC contains a GS-
GasPro (J&W Scientific, Folsom, CA, USA) porous layer open tubular (PLOT) column,
30 m x 0.32 mm. The temperature program holds the GC initially at -40°C for 4 minutes
then the temperature is ramped at +20°C per minute to a final temperature of 180°C. The
carrier gas is ultra high purity helium and the inlet pressure is held constant at 12 kPa
during the entire experiment. The exit end of the GC column is threaded directly into the
ion source of the mass spectrometer (25, 26).
Mass Spectrometer
The HRMS system, is a ZAB-2F (VG Instruments) high - resolution mass
spectrometer.
The ZAB-2F is a dual sector instrument of reverse geometry
(magnetic/electric) (Figure 4), with an accelerating voltage up to 8 kV. The large ion
optic system of this instrument allows for variable resolution from 600 to about 100,000
and high ion throughput for any resolution selected. Resolution is selected by manually
adjusting the source, intermediate slit, and detector slits (M/Am, where M is the mass of
the ion and Am is the peak width at half height). The present detector is an electron
21
Mass Spectrometer
Magnetic
Sector
Electric
Sector
Intermediate Slit
-'Source Slit .
Detector Slit
Heated Expansion
Volume
i v
Ion Source
[El] [Cl]
Figure 4: Schematic diagram of ZAB-2F high resolution mass spectrometer.
22
multiplier and is mounted 90° off the axis of the ion beam. This instrument is also
equipped with a stainless-steel conversion dynode that is located 90° off axis and is
negatively biased up to -5.0 IcV (25, 26).
Single Ion Monitoring Routine
The mass spectrometer is set to the selected ion monitoring (SIM) mode where the
CFs+ ion is monitored (m/z = 68.9952) using the desired mass resolution, routinely 2200.
In order to ensure that the mass spectrometer remains focused exactly on the centroid of
the selected mass, the mass scale of the mass spectrometer is recalibrated twice per
second throughout the analysis period with a peak matching technique (25, 26). In order
to do this, a calibration compound must be introduced continuously into the ion source of
the mass spectrometer by means of the heated expansion volume. In this study, ndodecane was chosen for our calibration compound since the identities and exact masses
of all the ions in its electron impact mass spectrum are well known. Before a specific
analysis can be performed, a mass calibration file must first be made. A mass calibration
file is made by performing a voltage scan, where the acceleration voltage (Initially set at
7.0 kV) and the electric sector field are uniformly scanned downward, while the magnetic
sector field is held constant. The voltage scan is carried out over a specific mass range,
which includes any two ions in the spectrum of n-dodecane that have a m/z > 42. In
addition, the mass range selected must be sufficiently narrow so that the ratio of mass of
the lighter to heavier ion is no less than 0.7, this keeps the acceleration voltage from
decreasing past 5 kY and ensures that an unacceptable loss of sensitivity does not occur.
Once obtained, the data system uses the mass calibration file to instruct the mass
23
spectrometer to jump between any one selected ion of n-dodecane and any other exact
mass within the calibrated' range. A duty cycle of 0.5 s was normally used, where the
mass spectrometer spends 0.10 s on the selected reference ion and the remaining 0.40 s
xon the ion selected for observation. Settling time between voltage jumps is very fast
(about 10 ms) and this is why electric field rather than magnetic field control of the
instrument is preferred. During the two times per second that the instrument focuses on a
selected calibration ion of known exact mass, the mass spectrometer automatically
refocuses itself by adjustment of the two electric fields to the centroid of the reference ion
peak (25, 26).
Chromatograph '
In the electron impact mass spectrum of CF3-Containing compounds, the CF3"1" ion
usually has major relative intensity and, therefore can be used for the detection of this
class of compounds. For four of the compounds studied here (CF3CF3Cl, CF3CH3F,
CF3CFCl3, and CF3CHF3), the CF3+ ion is not the major ion observed in their electron
impact mass spectrum (appendix I). Therefore, the detection-limits for these compounds
are generally improved somewhat when the most abundant ion for that compound is
monitored rather than the CF3"1" ion. The CF3"1" single ion chromatogram of a background
outside air sample using the GS-GasPro column is shown in Figure 5. The eleven CF3containing compounds shown in this chromatogram, are invariably detected in
background air samples. It is also noted that with this level of mass resolution (R =
2200), almost no responses to other components of the air sample are observed. The
improvements recently developed by our laboratory (61) include using a GS-GasPro
24
I
Time (minutes)
Figure 5: CF3+(m/z = 68.9952) single ion chromatogram of a background air sample at a
mass resolution of 2200. The peak identities are; I. CF4 (FC-14), 2. C3Fa (FC-116), 3.
CF3Cl (CFC-13), 4. CF3H (HFC-23), 5. CF3Br (Halon-1301), 6. C3F8 (FC-218), 7.
CF3CF2Cl, S- CF3CF2H (HFC-125), 9. CF3CH3 (HFC-143a), 10. CF3CFH2 (HFC-134a),
and 11. CF3CFCl2 (CFC-114a).
25
column which provides superior chromatographic resolution, no coelution of CFgcontaining compounds, and improved signal-to-noise detection responses (by at least a
factor of 5) relative to the GC column previously used (25). The improved technique
provides lower detection limits for all compounds, more reproducible and reliable peak
area determinations, and resulted in the detection of three CF3-Containing compounds in
the background atmosphere which were not detected with the previous analysis system
(25).
Those three compounds were C3Fg (octafluoropropane, FC-218), CF3CHF3
(pentafluoroethane, HFC-125), and CF3CH3 (1,1,1 -trifluoroethane, HFC-143a) (61).
These compounds are presently being used as substitutes for ozone-depleting
chlorofluorocarbons in a wide variety of industrial and commercial applications (41).
Confirmations of the identities of the compounds indicated in Figure 5 were based on
the retention time of pure standards and on the single ion detection of each compound
using all of the other major peaks in their known electron impact mass spectra (25). In
addition, when the mass resolution of the HRMS was increased from 2200 to 4700 all of
the responses observed in the single ion chromatogram using m/z = 68.9952 had the same
relative peak intensities. This provides additional conformation for the suggestion that
the major peaks shown in Figure 5 are, in fact, due to the CF3"1" ion and CF3-Containing
compounds and not due to any other interfering substances (25, 26, 61).
<6
The detection limits for all compounds shown in Figure 5, lie within the range from
0.5 to 5 parts per quadrillion (ppq) by volume when the most abundant ion of each
compound is monitored and all other conditions are set to those described in the
experimental section. Most of the air samples analyzed were collected 10 km south of
26
Bozeman, Montana, using an oil-less diaphragm pump (KNF NeuBerger, model
N05SVI) and a 6-L stainless steel container previously described (61). The whole air
samples are introduced to the GC-HRMS system by use of the associated air sampling
system described above.
Quantitative Determinations
Producing reliable calibration standards in low, or below, ppt levels presents a
difficulty when trying to calibrate instrument response.
A large volume standard-
additions approach was used to bring the concentrations of primary standards down to a
level that was comparable to their estimated concentrations in background air. In order to
do this a very large volume standard-addition vessel was needed to dilute the standards to
the appropriate concentration’s (26).
Our standard-addition device, consist of a large corrugated steel cylinder that is 5.9-ft
in diameter, 11.9-ft in length, and has a volume of approximately 9245-L. Initially, a
very large air sample is captured in this cylinder by first allowing ambient air to freely
pass through the cylinder overnight, then both ends of the cylinder are capped with
polypropylene tarps.
In order to verify the air inside the cylinder had not been
contaminated by its container, air samples were taken into 6-L stainless-steel-airsampling canisters simultaneously from both inside and outside the cylinder. These
samples were later analyzed by the HRMS system and no significant change in response
was observed (Table 2).
27
Table 2: Peak areas obtained from chromatograms of samples taken simultaneously from
inside and outside the standard-additions vessel._________________________________
a) Air sample obtained from outside standard-additions vessel (Peak areas x IO8).
CF4 C?Fr, CF3Cl CF7H CFiBr C3F8 CFiCF7Cl CFiCHi CFiCFH7 CFiCFCl7
44.7 7.67 6.59 6.19
6.49
1.06
6.20
4.00
15.9
2.26
b) Air sample obtained from inside standard-additions vessel (Peak areas x 108).
CF4 C2F6 CF3Cl CF3H CFiBr C3F8 CFiCF7Cl CFiCHi CFiCFH7 CFiCFCl7
44.7 7.74 6.45
5.98
6.22
1.11
6.08
4.02
14.4
2.22
Primary standards were made so that concentrations of the analyte of interest would be
increased four to ten times its estimated concentration in ambient air.
The primary
standards were prepared by using a gas tight syringe to inject a specific about of pure
compound into a 20-L gas carboy filled with nitrogen. Concentrations were determined
on a volumetric basis. 1.0- or 2.0-ml aliquots of the standard were then successively
added to the cylinder. As soon as the primary standard was injected into the cylinder, the
contents were thoroughly mixed with a fan that is manually operated using a handle that
extends through the wall of the cylinder. Once mixed, the contents of the cylinder were
allowed to equilibrate for about 45 minutes before an air sample was taken. After each of
the five standard additions, a sample was taken into a 6-L air-sampling canister and later
analyzed by the GC / HRMS system. Sample canisters were evacuated in our laboratory
prior to sample collection so an air sample could be collected by simply opening a single
valve on air sampling container.
Sample containers were commercially purchased
stainless steel 6-L vessels (SilcoCan, Restec Corp.). Each canister is lined with a layer of
fused silica that is chemically bonded to the interior surface and provides inertness for
active compounds. These samples were collected in March and April 2000,
28
approximately 10 km south of Bozeman, Montana (26). Once data was collected, a
standard-additions plot was made for each compound to determine its concentration in
the background atmosphere (Figure 6, shows an example standard-additions plot made
for C2F6).
29
RESULTS AND DISCUSSION
Results of Quantitative Determinations
A large sample standard-additions technique, previously described in experimental
section, was used for quantitative determinations. The concentrations for the eleven CF3containing compounds were found to be 82 + 8 pptv for CF4, 3.6 + 0.3 pptv for C2F6, 3.6
± 0.3 pptv for CF3CI, 18 + 2 pptv for CF3H, 3.7 + 0.3 pptv for CF3Br, 0.26 + 0.03 pptv
for C3Fg, 7.9 + 0.8 pptv for CF3CF2Cl, 1.5 + 0.1 pptv for CF3CHF2, 3.2 + 0.3 pptv for
CF3CH3, 2 1 + 2 pptv for CF3CH2F, and 7.9 + 0.8 pptv for CF3CFCl2. The uncertainty
indicated for each compound is due either to the uncertainty (at the 95% confidence limit)
of the linear regression analysis of the standard addition samples or to our estimate of the
error associated with the preparation of our standards (thought to be approximately
10%(26)), whichever is larger. Changes in the response for all compounds on different
sampling dates were within the limits of uncertainty indicated above. Therefore, the
above concentrations are thought to be representative of the average concentrations over
the March through April 2000 time period. Also, the responses for all compounds varied
negligibly when compared to samples obtained from Cape Meares Oregon, on similar
sampling dates. This suggests that any local sources of these compounds in Bozeman,
Montana were of negligible importance.
Historic Air Samples
The GC / HRMS system was then used to monitor nine of the CF3-Containing
compounds shown in Figure 5 in a collection of historic air samples. HFC-125 was
30
excluded, since this compound was nonexistent in the atmosphere until a couple of years
ago and even now its concentration is very low. Also, CF4 measurements in the historic
air samples have not yet been performed.
Air samples were collected by Dr. Rei
Rasmussen from the Oregon Graduate Institute, with whom we have been working on
this project. One set of samples was collected from Cape Meares Oregon over the past
several decades, from 1977-1998. A second set of samples was collected from Point
Barrow Alaska and Palmer Station Antarctica, from 1991-1998.
Samples collected from Cape Meares Oregon were analyzed and used to monitor
changes in the percentage of each compound relative to today’s air. Samples from Point
Barrow Alaska and Palmer Station Antarctica were analyzed and used to observe how the
relative amount of each compound differed in the Northern and Southern Hemispheres.
Historic Concentrations of Compounds
Chromatograms obtained for historic air samples were used to determine the relative
amount of each compound.
As ■can be seen in Figure 5, the CFa+ single ion
chromatogram contains a peak that is indicative of each compound observed in this study.
Peak areas obtained from these peaks were used to calculate the relative amount of each
compound. In order to do this a present day air sample was analyzed prior to analysis of
the historic air samples for each day experiments were performed. The relative amount
of each compound was. determined by comparing the peak areas obtained from the
historic air samples with peak areas obtained from the present day air samples. The peak
areas found in the present day air samples were assumed to be 100%. The relative
amount data was then used to find concentrations for each compound. This was done, by
31
multiplying the averaged relative amount of a particular compound for a particular year
by that compounds present concentration in the atmosphere. Data for these calculations
are graphed in Figures 7-24. Graphs were made by plotting the concentration of each
compound in an historic air sample versus the date that air sample was collected. When
looking at Figures 7-24, similar trends can be seen for similar compounds (Table 3 list
some characteristics for each compound).
The FCs (C2Fg, C3F8) (Figures 7-10) in this study have similar compositions and
structures, therefore, they should have similar atmospheric lifetimes. The lifetime for
C2F6 is approximately 10,000 years (42), so the atmospheric lifetime for C3F8 is thought
to be approximately the same. Atmospheric C2Fg is thought to be entirely anthropogenic
(42) and is believed to be primarily a result of the Hall-Heroult process for aluminum
production, but is also used in the plasma etching process in the semiconductor industry
(56).
During the Hall-Heroult process, aluminum oxide (Al2O3) is mixed with cryolite
(Na3AlFg) and reduced electrolytically using carbon electrodes. If the concentration of
alumina in the molten salt bath decreases below about 2% the electrolytic cell is
interrupted by the “anode effect”. This is when some of the oxide ions near the anode are
replaced by fluoride ions. The fluoride ions are then reduced to fluorine, which reacts
with the carbon electrodes and forms CF4 & C2Fg (14, 30, 42). The presence of C3F8 in
the atmosphere seems to be primarily a result of its use for the plasma etching process in
the semiconductor industry (56).
Since atmospheric lifetimes for C2F6 and C3F8 are extremely long, it would be
expected that these compounds would be well mixed throughout the atmosphere (30). As
Cape Meares Oregon
4
3.5
C oncentration (pptv)
3
2.5
2
1.5
1
0.5
0
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 7: Concentration of C2F6 from sam ples obtained from Cape Meares Oregon.
Point Barrow Alaska and Palmer Station Antarctica
o.
B
c
0
3
ro
CO
CO
1 2
c
O
O
I -B-PSA
# --PBA
1991
1992
1993
1994
1995
1996
1997
Date
Figure 8: Concentration of C2F6 from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
1998
1999
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
Date
Figure 9: Concentration of C3Fgfrom sam ples obtained from Cape Meares Oregon.
92
93
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
2 0.20
-
O 0.10
e-PSA
Figure 10: Concentration of C3F8 from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
36
can be seen when referring to Figure 8 for C2Fg, there is minimal difference in the
concentration of C2Fg in the samples obtained from northern and southern hemispheres.
However, for C2Fg (Figure 10) there is a slight difference in the concentration of this
compound in the samples from both hemispheres.
A possible explanation for this
observation can be seen when referring to Figure 9. When referring to Figure 9, it can be
seen that the concentration of C2Fg remains fairly constant from 1978 to 1986, only
increasing from about 0.075 pptv to approximately 0.1 pptv.
After 1986, the
concentration of C2Fg increases from about 0.1 pptv to close to .26 pptv by 1998. This
observation indicates that a relatively large amount of this compound has been released in
recent years compared to the past. Also, it is generally believed that more halogeriated
compounds are released in the Northern Hemisphere than in the Southern Hemisphere
because more industrialized countries are located in the Northern Hemisphere. Taking
these two factors into consideration, it may be expected that this compound had not yet
become as thoroughly mixed in the atmosphere.
Three CFCs were observed in these historic air samples, CF2CF2Cl (CFC-115)
(Figures 11 & 12), CF2CFCl2 (CFC-IMa) (Figures 13 & 14), and CF2Cl (CFC-13)
(Figures 15 & 16). CFCs are used in a variety of industrial applications as refrigerants,
aerosol propellants, foam blowing agents, and fire extinguishers (9-14). The atmospheric
lifetimes for CFCs-115, 114a, and 13, are 1700, 300, and 640 years respectively (45).
When referring to the graphs of the concentrations of these compounds, similar trends
can be seen for all of them. First, all of these compounds appear to be well mixed in the
atmosphere, indicated by the close agreement of the data in graphs from Point Barrow
/
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
Date
Figure 11: Concentration of CF3CF2CI from sam ples obtained from Cape Meares Oregon.
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
CO
OO
e— PSA
Figure 12: Concentration of CF3CF2CI from sam ples obtained from Palmer Station Antarctica (PSA) and Point
Barrow Alaska (PBA).
Cape Meares Oregon
Figure 13: Concentration of CF3CFCI2 from sam ples obtained from Cape Meares Oregon.
Point Barrow Alaska and Palmer Station Antarctica
4^
o
e— PSA
PBA
Figure 14: Concentration of CF3CFCI2from sam ples obtained from Palmer Station Antarctica (PSA) and Point
Barrow Alaska (PBA).
Cape Meares Oregon
5
I 0.5 I
i
75
76
i
77
i
78
i
i
79
80
i
i
81
i
82
i
83
i
84
l
85
i
86
i
87
i
88
i
89
i
90
i
91
i
i
92
93
Date
Figure 15: Concentration of CF3CI from sam ples obtained from Cape Meares Oregon.
i
94
l
95
l
96
l
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
e— PSA
-"PBA
Figure 16: Concentration of CF3CI from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
43
Alaska and Palmer Station Antarctica (Figures 12, 14, & 16) for all of these compounds.
Secondly, when referring to graphs generated from samples obtained from Cape Meares
Oregon (Figures 11, 13, & 15) for all of these compounds, the concentration of these
compounds increases rapidly from 1978 until the late 80’s or early 90’s. After which, the
increase in the concentration of these compounds seems to slow down and even level off
by 1998. These observations are a result of the restrictions placed on the use of these
compounds in 1987 by the Montreal Protocol.
Three HFCs were observed in this study, they are CF3CFH2 (HFC-134a)(Figures 17 &
18), CF3CH3 (HFC-143a)(Figures 19 & 20), and CF3H (HFC-23)(Figures 21 & 22).
HFC-134a and HFC-143a are currently being used as long term replacements for CFCs in
industrial applications. The atmospheric lifetimes of these compounds are significantly
less than the compounds they are replacing, as can be seen in Table 2. Until the recent
phase-out of halon manufacture by the Montreal Protocol, HFC-23 was used primarily as
a raw material in the production of CF3Br. Today, HFC-23 is mainly produced as a by­
product during the manufacture of CHCIF2, which is an HCFC primarily used in
refrigeration and air conditioning. In addition, HFC-23 is used in the plasma etching
process in the manufacture of semi-conductors, as a low temperature refrigerant, and as a
fire extinguishing agent (55).
Again, similar trends in the graphs of the concentration can be seen for the HFCs.
From the discussions above for CFCs and FCs, several trends can be anticipated for the
HFCs. First, because of the relatively short atmospheric lifetimes for these compounds, it
is expected that these compounds will not be as well mixed in the atmosphere. Second, a
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
Date
Figure 17: Concentration of CF3CFH2 from sam ples obtained from Cape Meares Oregon.
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
8
1991
1992
1993
1994
1995
1996
1997
1998
Date
Figure 18: Concentration of CF3CFH2 from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
1999
C oncentration (pptv)
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
Date
Figure 19: Concentration of CF3CHgfrom sam ples obtained from Cape Meares Oregon.
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
Date
Figure 20: Concentration of CF3CH3from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
Date
Figure 21: Concentration of CF3H from sam ples obtained from Cape Meares Oregon.
93
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
-tx
CD
Figure 22: Concentration of CF3H from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
50
large increase in concentration would be expected from the late 80’s to 1998 since this is
the time period when these compounds have been used for CFC replacement. When
referring to Figures 17-22, both of these trends can be seen for two of these compounds.
Graphs made from data obtained for CF3CFH2 and CF3CH3 (Figures 18 & 20) from Point
Barrow Alaska and Palmer Station Antarctica indicate a significant difference in the
concentration of these compounds in Northern and Southern Hemispheres. Also, a large
increase in the concentration of these compounds can be seen from 1987 to 1998 in the
Cape Meares Oregon data, as well as in the data obtained from Point Barrow Alaska and
Palmer Station Antarctica. Unfortunately, these trends are not seen for CF3H (Figures 21
& 22) probably due to the higher variation in response for this compound.
Also,
significant amounts of CF3H have been emitted into the atmosphere for a longer period of
time and its atmospheric lifetime of 230 years is relatively long when compared to the
atmospheric life times of CF3CFH2 and CF3CH3.
Finally, the last compound monitored in this study was CF3Br (halon-1301) (Figures
23 & 24). The compound was mainly used as a fire-extinguishing agent. The production
of CF3Br has been discontinued. However, there still is some in service. CF3Br has a
fairly long atmospheric lifetime of approximately 65 years and restrictions placed on this
compound are a result of the large ozone depletion potentials for brominated compounds.
Evidence of these restrictions can be seen when referring to Figure 23, the concentration
of CF3Br dramatically increases from about 0.5 - 3.5 pptv from 1977-1994 after which
time only a gradual increase in concentration can be seen.
C oncentration (pptv)
Cape Meares Oregon
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90 91
92
Figure 23: Concentration of CF3Br from sam ples obtained from Cape Meares Oregon.
93
94
95
96
97
98
99
Point Barrow Alaska and Palmer Station Antarctica
> 4 -
PSA
Figure 24: Concentration of CF3Br from sam ples obtained from Palmer Station Antarctica (PSA) and
Point Barrow Alaska (PBA).
53
An important part of this study was to see how well our results compared to previous
studies found in the literature. Similar studies, in which the concentration of halogenated
compounds were measured over many years, were found for CFgBr (62), CFgH (55), and
CFgCFHg (27).
To compare our results to these studies, additional graphs of
concentration vs. sample date were made for each compound; CFgBr (Figure 25), CFgH
(Figure 26), and CFgCFHg (Figure 27).
The data from other studies found in the
literature was also included on these graphs. When referring to these graphs, keep in
mind that air samples used in other studies were all collected from Cape Grim in the
Southern Hemisphere and our samples wefe collected from Cape Meares Oregon in the
Northern Hemisphere, however observed trends in both sets of data are similar.
Growth Rates
In order to estimate growth rates, annually averaged concentrations were calculated for
each compound. Graphs of the data obtained for each compound are shown in Figures
28-36. The open squares represent concentrations calculated from the samples collected
from Cape Meares Oregon. A trend line was added to data from Cape Meares Oregon
using either a 2nd or 3rd order polynomial fit, except for CFgCFHg and CFgCHg where
connecting lines gave a better representation of Cape Meares Oregon data. The graph
also contains annually averaged concentrations calculated for samples obtained from the
Point Barrow Alaska samples (solid circles) and Palmer Station Antarctica samples (solid
rectangles), added for comparison. The data obtained from the Cape Meares Oregon air
samples was then used to estimate the average annual growth rate for each compound
over the time period these samples were collected (Table 3), except for the CFgCFHg and
CF3Br
c 2.5 -
□ □ □
O 1.5 -
□ CMO
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 25: Concentrations for CF3Br from samples from Cape Meares Oregon (CMO) (Northern Hemisphere).
Concentration of CF3BrcaIcuIated by Fraser et al. From samples collected from Cape Grim (CG) (Southern Hemisphere).
CF3H
->
20
-
—D —CMO
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91
92 93 94 95 96 97 98 99
Date
Figure26: C oncentrations for CF3H from sam ples from C ap e M eares O regon (CMO) (Northern Hemisphere). Trend line represen
CF3H concentration d ata obtained by Oram et al. from sam p les collected from C ap e Grim (CG) (Southern Hemisphere).
CF3CFH2
c 3 -
—o —CMO
Figure 27: Concentration of CF3CFH2 from sam ples from C ap e M eares Oregon (CMO) (Northern Hemisphere). Trend line
represents CF3CFH2 concentrations obtained by Oram et al. from sam ples collected from C ape Grim (CG) (Southern
Hem isphere).
Concentration of C,Fc
c 2.5 CD
~ sl
□
CMO
a - PSA
# PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 28: Concentration of C2F6- Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
C on cen tration (pptv)
Concentration of CRCI
o
CMO
a
RSA
... # ... PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 29: Concentration of CF3CI. Cape Meares Oregon (CMC); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
Concentration of CRH
T 20 -
□
CMO
-*...PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 30: Concentration of CF3H. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
C oncentration (pptv)
Concentration of CF,Br
CD
o
□
-
CMO
A - - PSA
#....PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 31: Concentration of CF3Br. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
Concentration of C,FB
0.25 -
0.2
0.15 -
0.05 n
CMO
# ...PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 32: Concentration of C3F8. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
Concentration of CRCFoCI
o
4 -
□
CMO
a
RSA
...# ... PBA
75 76 77 78 79 80 81
82 83 84 85 86 87 88 89 90 91
92 93 94 95 96 97 98 99
Date
Figure 33: Concentration of CF3CF2CI. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
C oncentration (pptv)
Concentration of CRChU
O)
CO
Q— CMO
a
PSA
• PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 34: Concentration of CF3CH3. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA); Palmer
Station Antarctica (PSA).
Concentration of CFXFH9
CD
B— CMO
a
PSA
• PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 35: Concentration of CF3CFH2. Cape Meares Oregon Air (CMO); Point Barrow Alaska (PBA);
Palmer Station Antarctica (PSA).
Concentration of CFXFCI9
□ CMO
•
PBA
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
Date
Figure 36: Concentration of CF3CFCI2. Cape Meares Oregon (CMO); Point Barrow Alaska (PBA);
Palmer Station Antarctica (PSA).
66
CF3CH3 which were determined only from 1990-1998, the years when significant growth
occurred.
Table 3: Lists results for average annual growth rate for CF3-Containing compounds.
Average annual growth rate determined from 1977-1998, except for HFCs (a), which
were determined from 1990-1998.
Average Annual Growth Rate
Compound
CzFe
CF3Cl
CF3H
CF3Br
C3Fg
CF3CF2Cl
CF3CH3(a)
CF3CFH2(a)
CF3CFCl2
Growth Rate(DDtvZvr)
0.085
0.987
0.61
0.177
0.007
0.327
0.185
1.31
0.241
Growth rates for several of these compounds were found in the literature and compared to
the above results. The mean increase of CzFc was found to be 0.084 ppt/yr in the
Northern Hemisphere by Hamisch et ah, from 1982-1995 (42). This value agrees closely
with the mean increase of 0.085 pptv/yr for C2p6 calculated from our data from samples
from Cape Mears Oregon. Another comparison was made for CF3H, the growth rate for
this compound was found to be 0.55 pptv/yr in the Southern Hemisphere by Oram et al.
in 1998 (55). This value differs from the 0.61 pptv/yr for CF3H calculated from samples
from Cape Meares Oregon by less than 10%.
Emission Estimates
Once average concentrations for the CF3-Containing compounds had been determined
for each year, averaged annual emission estimates were made for all nine compounds
monitored in air samples from Cape Meares Oregon. Emission estimates were calculated
67
using the following equation: E = (Ma/f)(A +
x/t
)
(46); where Ma represents the mass of
the dry atmosphere 5.132 x IO18 Kg (54), f is a unit-less number that converts
tropospheric measurements into global averages, A is the observed mixing ratio
difference over a specified time period, x is the observed global tropospheric mean
mixing ratio over the specified time period, and x is the mean steady state lifetime of the
compound (Table 4) (46).
Table 4: Characteristics of atmospheric halocarbons. * An atmospheric lifetime was not
found in the literature for this compound, however the lifetime of this compound is
thought to be approximately 10,000 years since chemical composition and structure is
similar to C^Fr,.___________________________________________________________
Atmospheric Halocarbon Characteristics
Chemical
CF4
CzFe
CF3Cl
CF3H
CF3Br
C3Fg
CF3CF2Cl
CF3CH3
CF3CFH2
CFiCFCF
Common Name
Lifetime
(Years)
References
f
FC-14
FC-116
CFC-13
HFC-23
Halon-1301
FC-218
CFC-115
HFC-143a
HFC-134a
CFC-114a
50,000
10,000
640
230
67
*
1,700
54.1
16
300
(42)
(42)
(45)
(55)
(21)
1.00
1.00
1.02
1.07
1.08
1.00
1.02
1.08
1.10
1.08
(45)
(H )
(H )
(45)
The magnitude of f used in the above equation is calculated using the following equation,
f = X/0.83; where X is total atmospheric mass of the compound and 0.83 is the fraction of
total atmospheric mass in the troposphere. For a gas whose mixing ratio is constant at all
altitudes, f = 1.00 or in other word’s, the global mean burden is well estimated by surface
measurements and no correction is necessary.
For a compound that has a minimal
abundance in the stratosphere compared to the troposphere (either because of a very fast
68
relative rate of increase at the surface and/or because the stratospheric loss frequency is
very short compared to the tropospheric loss frequency), then f can approach a maximum
determined by the mass of the troposphere compared to the mass of the whole
atmosphere. For example, because f equals the fraction of total atmospheric mass of X
divided by the fraction of total atmospheric mass in the troposphere, f for a gas without
significant abundance in the stratosphere would equal approximately 1/0.83, or 1.2. For
the compounds in this study; f was either found in the literature (46); estimated with
verticle profiles found in the literature for these compounds (10, 44, 45, 55); or estimated
by comparisons with other gases with similar chemical structures, growth rates, and
lifetimes (Table 4) (The above information about the f and the percentage of total
atmospheric mass in the troposphere, was obtained through a personal communication
with Steve Montzka from the National Oceanic of Atmospheric Administration).
Since atmospheric concentrations of these compounds are very low and atmospheric
lifetimes are quite long, very little change in concentration occurred over a 12-month
period.
For this reason, emission estimates for these compounds were calculated over a
24-month period and the results for these calculations are listed in Table 5 below.
Table 5: Halocarbon emission estimates (x IO5 Kg/ 2-years) derived from annual means
of atmospheric measurements. (NA) Data was not available for calculations over that
time period. (-) Emmission estimates for these compounds on these years were done over
and 12-month period and listed in Table 6.
Year C2E6 CF1Cl CF1H CF1Br C2Fg CF3CF2Cl CF3CH3 CFiCFH3 CFiCFCl3
77-79
79-81
81-83
83-85
85-87
6.41
2.31
4.66
11.6
13.6
NA
14.7
11.2
15.6
23.8
64.0
66.7
30.6
1.78
82.7
10.0
11.6
12.9
24.4
26.9
.053
.787
*
.387
1.41
37.3
26.7
25.6
40.7
34.9
NA
1.63
.224
3.00
8.45
NA
6.51
.508
11.0
3.61
52.3
28.8
22.5
33.4
46.5
69
87-89 14.2
157
.757
57.2
1.01
17.4
15.7
28.8
5.29
*
*
89-91
3.28
2.51
1.01
17.7
.449
17.4
111
1.23
91-93 13.1
.628
51.3
37.5
21.4
1.48
110
16.0
1.96
7.25
93-95 23.5
13.9
95-97 2.25
.890
16.4
54.8 11.9
* indicates time periods when a negative increase in concentration was observed, so no
emission estimate could be made for that compound during that time period. In order to
get a statistically significant emission estimate with the equation described above, a
positive increase in concentration needs to be observed over the indicated time period.
This limitation of the emission equation results from the low atmospheric concentrations
and long atmospheric lifetimes of the compounds observed in this study._____________
H6
-
-
-
-
-
-
-
-
However, from 1989-1997 a large increase in concentration was observed for CF3CFH2
and CF3CH3, so emission estimates for these two compounds were made over a 12-month
period for those years and listed in Table 6 below.
Table 6: Emission estimates (x IfP kg/yr) for CF3CFH7 and CF1CH1 from 1989-19967
Date
1989
1990
1991
1992
1993
1994
1995
1996
CF^CFH7
9.51
111
28.5
26.8
36.4
113
172
198
CF2CH2
4.57
7.80
4.63
4.78
13.7
13.9
27.4
49.5
70
CONCLUSIONS
This study describes the use of a GC / HRMS system to measure CF3 - containing
compounds in the background atmosphere.
The high degree of measurement
reproducibility and very specific mass monitoring with high resolutions is shown to be
quite valuable to atmospheric analysis of trace constituents. Recent improvements made
to this technique involved using a GS-GasPro gas chromatographic column for the
separation of these very volatile compounds. Results indicate that these improvements
have allowed for eleven CF3 - containing compounds to be detected in background air,
including CF4 (FC-14), C2F6 (FC-116), CF3Cl (CFC-13), CF3H (HFC-23), CF3Br (Halon1301), C3F8 (FC-218), CF3CF2Cl (CFC-115), CF3CHF2 (HFC-125), CF3CH3 (HFC143a), CF3CH2F (HFC-134a), and CF3CFCl2 (CFC-114a). Three of these had never
before been detected in the atmosphere. Those three compounds were C3F8
(octafluoropropane, FC-218), CF3CHF2 (pentafluoroethane, HFC-125), and CF3CH3
(1,1,1-trifluoroethane, HFC-143a).
The improved method was used to perform quantitative determinations for each
compound listed above using a large volume standard-additions approach. Using a large
volume standard-additions method is advantageous in analyzing compounds with
exceedingly low atmospheric concentrations since primary standards must be diluted
down to a level that is proportional to their atmospheric concentrations. Also, this method
provides for some degree of compensation for any possible sources of systematic error.
71
The GC / HRMS system was then used to monitor nine CFg-containing compounds in
a collection of historic air samples, collected over the past several decades. The data
obtained from this study was used to calculate concentrations for each compound over
the last several decades. Annually averaged concentrations were then used to calculate
the concentration growth of each compound in the atmosphere. From these observations
average annual growth rates were then calculated for each compound. Finally, data was
used to estimate emissions of each compound over a specified time period using a simple
mathematical model.
The results from this study indicate a decrease in the growth rate of CFCs and BFCs in
the atmosphere. This decrease is the result of restrictions placed on these compounds by
the Montreal protocol. Also, these results have shown that the concentration of HFCs has
increased dramatically since the early 1990’s when HFCs began to replace CFCs for
industrial applications. Data from samples collected from the Point Barrow Alaska and
Palmer Station Antarctica indicate a correlation between atmospheric lifetime and
emissions, and agreement in concentrations of these compounds in the Northern
Hemisphere and Southern Hemisphere.
The composition of the atmosphere continuously changes as trace constituents are
emitted into the atmosphere.
Pollutant gases such as CFCs and BFCs with long
atmospheric lifetimes enter the stratosphere and alter stratospheric ozone chemistry. This
problem has lead to their replacement by HFCs whose potentially toxic degradation
products, has given rise to other environmental concerns. Other compounds like FCs are
mixed in the atmosphere and cause problems like global warming. In order, to accurately
72
estimate consequences the changing atmospheric composition will have on' the
environment, research focused on the detection and quantitation of atmospheric trace
constituents needs to be continued.
73
BIBLIOGRAPHY
1. McEIroy, M.D., and Salawitch, RJ., Changing Composition of the Global
Stratosphere, Science, 243, 763-770, 1989.
2. Prather, M J., and Watson, R.T., Stratospheric ozone depletion and future levels of
atmospheric chlorine and bromine, Nature, 344, 729-734, 1990.
3. Fisher1DA., Hales, C.H., Filltin, D.L., Ko, M.K.W., DakSze1N., Connell, P.S.,
Wuebbles1DJ., Isaksen1LSA., and Stordal, F., Model calculations of the relative
effects of CFCs and their replacements on stratospheric ozone, Nature, 344, 508513, 1990.
4. Solomon, S., Mills, M., Heidt, L.E., Pollock,W.H., and Tuck, A.F., On the
Evaluation of Ozone Depletion Potentials, J. Geophys. Res., 97, 825-842, 1992.
5. Fisher, DA., Hales, C.H., Wang, W., Ko, M.K.W., and Dak Sze, N., Model
Calculations of the relative effects of CFCs and their replacements on global
warming, Nature, 344, 513-516, 1990.
6. Schneider, S.H., The Greenhouse Effect: Science and Policy, Science, 243, 771779,1989.
7. Ravishankara,A.R., Solomon, S., Tumipseed1AA., and Warren, R.F.,
Atmospheric lifetimes of long-lived halogenated species, Science, 259, 194-199,
1993.
8. Daniel, LD., Solomon, S., and Albritton, D.L., J. Geophys. Res., 100, 1271, 1995.
9. Montzka1SA., Myers, R.C., Butler, J.H., and Elkins, I.W., Early trends in the
global tropospheric abundance of hydrochlorofluorocarbon-141b and 142b,
Geophys. Res. Lett., 23, 2483-2486, 1994.
10. Fabian, P., Borchers1R., Penkett1SA., and Prosser, N.J.D., Halocarbons in the
stratosphere, Nature, 294, 733-735, 1981.
11. Hayman, G.D., and Derwent, R.G., Atmospheric Chemical Reactivity and OzoneForming Potentials of Potential CFC Replacements, Environ. Sci. Technol., 31,
327-336, 1997.
12. Butler, J.H., Elldns1J-W., Hall1BD., Cummings, S.O., and Montzka, S.A., A
74
decrease in the growth rates of atmospheric halon concentrations, Nature, 359,
403-405, 1992. .
13. Simmonds, P.G., O’Doherty, S., Huang, J., Derwent, R.G., Ryall, D., Nickless, G.,
and Cunnold, D., Calculated trends and the atmospheric abundance of 1,1,1,2tetrafluoroethane, 1,1-dichloro-l-fluoroethane, and I-chloro-1,1-difluoroethane
using automated in-situ gas chromatography-mass spectrometry measurements
recorded at Mace Head, Ireland, from October 1994 to March 1997, J. Geophys.
Res., 103, 16,029-16,037, 1998.
14. Hamisch, J., Borchers1R., Fabian, P., and Maiss, M., CF4 and the age of
mesospheric and polar vortex air, Geophys. Res. Lett., 26, 295-298, 1999.
15. Lovelock, J.E., Atmospheric fluorine compounds as indicators of air movements,
Nature, Lond. 230, 379, 1971.
16. Lovelock, J.E., Maggs5RJ., and Wade, R J., Halogenated hydrocarbons in and over
the Atlantic, Nature, Lond. 241, 194-196, 1973.
17. Grimsrud, E.P., and Miller, D.A., Oxygen doping of the electron capture detector in
measurements of halogenated methane’s by gas chromatography with electron
capture detection, Anal. Chem., 50, 1141-1145, 1978.
18. Grimsrud, E.P. and Rasmussen, RA., Survey and analysis of halocarbons in the
atmosphere by gas chromatography-mass spectrometry, Atmos. Environ., 9, 10101013, 1975b.
19. Grimsrud, E.P. and Rasmussen, RA., Analysis of chlorofluorocarbons in the
troposphere by gas chromatography-mass spectrometry, Atmos. Environ., 9,10141017, 1975a.
20. Penkett, S., Jones, B., Rycroft, B., and Simmons, D., Interhemispheric comparison
of the concetrations of bromine compounds in the atmosphere, Nature, 318, 550553, 1985.
21. Pollock, W.H., Heidt, L.E., Lueb1RA., Vedder, J.F., Mills, M J., and Solomon, S.,
On the age of stratospheric air and ozone depletion potentials in polar regions, J.
Geophys. Res., 97, 12,993-12999, 1992.
22. Schauffler, S.M., Heidt, L.E., Pollock, W.H., Gilpin, T.M., Vedder, J.F., Solomon,
S., Lueb5RA., and Atlas, E.L., Measurements Of Halogenated Organic
Compounds Near The Tropical Tropopause, Geophys. Res. Lett., 20, 2567-2570,
1993.
75
23. Thompson, T.M., Elkins. J.W., Butler, TH., Montzka, S.A., Myers, R.C., Baring,
T.J., Cummings, S.O., Dutton, G.S., Gilligan, J.M., Hayden, A.H., Lobert, TM.,
Swanson, T.H., Hurst, D.F., and Volk, C.M., Nitrousoxideandhalocarbon
division, in Climate Monitoring and Diagnostics Laboratory, Report No. 22,
Department of Commerce, 72-91, 1993.
24. Schauffler, S.M., Pollock, W.H., Atlas, E.L., and Heidt, L.E., Geophys. Res. Lett.,
22, 819, 1995.
25. Engen, M.A., Wagner, V.A., Sears, L.J., and Grimsrud, E.P., Detection of CFgcontaining compounds in background air by gas chromatography / high-resolution
mass spectrometry, I. Geophys. Res., 103, 25,287-25,297, 1998.
26. Engen, M.A., Culbertson, J.A., and Grimsrud, E.P., Analysis of brominated
compounds in background air by gas chromatography - high-resolution mass
spectrometry, J. Chrom., 848, 261-277, 1999a.
27. Oram, D.E., Reeves, C.E., Strurges, W.T., and Penkett, S.A., Recent tropospheric
growth rate and distribution of HFC-134a (CF3CH2F), Geophys. Res. Lett., 23,
1949-1952, 1996.
28. Crutzen, P.J., My Life with O3, NOx, and Other YZOx compounds (Nobel Lecture),
Angew. Chem. Int. Engl., 35, 1758-1777, 1996.
29. Rowland, F.S., Stratospheric Ozone Depletion by Chlorofluorocarbons (Nobel
Lecture), Angew. Chem. Int. Engl., 35, 1786-1798, 1996.
30. Weston Jr., R.E., Possible Greenhouse Effects of Tetrafluoromethane and Carbon
Dioxide Emitted from Aluminum Production, Atmos. Environ., 30, 2901-2910,
1996.
31. Wayne, R.P., Chemistry of atmsopheres: An introduction to the chemistry of the
Atmsnpheres of Earth, the planets, and their satellites, Claredon Press, 1985.
32. Sloane, C.S., and Tesche, W.T., Atmospheric Chemistry: Models and Predictions for
Climate and Air Quality, Lewis Publishers, 1991.
33. Molina, M J., Polar Ozone Depletion (Nobel Lecture), Angew. Chem. Int. Engl., 35,
1778-1785, 1996.
34. Penkett, S.A., Prosser, N.J.D., Rasmussen, R.A., and Khalil, M.A.K., Atmospheric
Measurements of CF4 and other Fluorocarbons Containing the CF3 Grouping, J.
Geophys. Res., 86, 5172-5178, 1981.
76
35. Anderson, J.G., Toohey, D.W., and Brune, W.H., Free Radicals Within the
Antarctic Vortex: The Role of CFCs in Antarctic Ozone Loss, Science, 251, 3946, 1991.
36. Walbel, A.E., Peter, T., Carslaw, K.S., Oelhaf, H., Wetzel, G., Crutzen, P.J.,
Poschl, U., Tsias, A., Reimer, E., and Fischer, H., Arctic Ozone Loss Due to
Denitrification, Science, 283, 2064-2069, 1999.
37. Hamill, P., and Toon, O.B., Polar Stratospheric Clouds and the Ozone Hole,
Physics Today, pgs. 34-42, December 1991.
38. McElroy, M.B., Salawitch5RJ., Wofsy, S.C., and Logan, J.A., Reductions of
Antarctic ozone due to synergistic interactions of chlorine and bromine, Nature,
321, 759-762, 1986.
39. Zurer, P.S., Ozone Depletion’s Recurring Surprises Challenge Atmospheric
Scientists, C&EN, pgs. 8-18, May 1993.
40. Schoeberl, M.S., and Hartmann, D.L., The Dynamics of the Stratospheric Polar
Vortex and Its Relation to Springtime Ozone Depletions, Science, 251, 46-52,
1991.
41. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-1997, U.S.
Environmental Protection Agency [EPA 236-R-99-003], April 1999.
42. Hamisch, I , and Borchers, R., Tropospheric trends for CF4 and C2F6 since 1982
derived from SF6 dated stratospheric air, Geophys. Res. Lett, 23, 1099-1102,
1996.
43. Molina, M J., and Rowland, F.S., Stratospheric sink for chlorofluoromethanes:
Chlorine atom-catalyzed destruction of ozone, Nature, 249, 810-812, 1974.
44. Fraser, P., Fisher, D., Bloomfield, P., Sander, S.P., and Ko, M.K.W., Report on
Concentrations, Lifetimes, and Trends of CFCs, Halons, and Related Species,
NASA Reference Publication 1339,1994.
45: Fabian, P., Borchers, R., Leifer5R., Subbaraya, B.H., Lai, S., and Boy, M., Global
stratospheric distribution of halocarbons, Atmos. Environ., 30, 1787-1796, 1996.
46. Montzka, S.A., Butler, J.H., Elkins, J.W., Thompson, T.M., Clarke, A.D., and
Lock, L.T., Present and future trends in the atmospheric burden of ozone-depleting
halogens. Nature, 398, 690-694, 1999.
47. Montzka, S.A., Myers5R-C., Butler, J.H., and Elkins, J.W., Obervations of HFC-
77
134a in the remote troposphere, Geophys. Res. Lett, 23, 169-172, 1996.
48. Montzka, S.A., Myers, R.C., Butler, J.H., and Elldns, J.W., Global Tropospheric
Distribution and Calibration Scale of HCFC-22^ Geophys. Res. Lett., 20, 703-706,
1993.
49. Daniel, J.D., Schauffler, S.M., Block, W.H., Solomon, S., Weaver, A., Heidt, L.E.,
Garcia, R.R., Atlas, E.L., and Vedder, J.F., On the age of stratospheric air and
inorganic chlorine and bromine release, I. Geophys. Res., 101, 16,757-16,770,
1996.
50. Prinn, R.G., Simmonds, P.G., Rasmussen, RA., Rosen, R.D., Alyea, F.N.,
Cardelino, CA., Crawford, A. J., Cunnold, D.M., Frasser1P J li and Lovelock, J.E.,
The Atmospheric Lifetime Experiment: I. Introduction, Instrumentation, and
Overview, I. Geophys. Res., 88, 8353-8367, 1983.
51. Rasmussen, RA., and Lovelock. LE., The Atmospheric Lifetime Experiment: 2.
Calibration, I. Geophys. Res., 88, 8369-8378, 1983.
52. Rasmussen, RA., Penkett1SA., and Prosser, N., Measurement of carbon
tetrafluoride in the atmosphere, Nature, 277, 549-551, 1979.
53. Hills, A.J., Cicerone1RJ., Calvert, J.G., and Birks, J.W., Kinetics of the BrO +
CIO reaction and implications for stratospheric ozone, Nature, 328, 405-413, 1987.
54. Trenberth, K.E., and Guillemot, CJ., The total mass of the atmosphere, I. Geophys.
Res., 99, 23079-23088, 1994.
55. Oram, D.E., Sturges, W.T., Penkett1SA., McCulloch, A., and Fraser, P J., Growth
of (CHF3, HFC-23) in the background atmosphere, Geophys. Res. Lett., 25, 35-38,
1998.
56. Pmette, L., Karecki, S., Reif, R., Tousignanat, L., Reagan, W., Kesari, S., and
Zazzera, L., Environmental Issues in the Electronics and Semiconductor Industries,
Electrochem. Soc., 99-8, 20-29, 1999.
57. O’Doherty, S J., Nickless, G., Bassford, M., Pajot, M., and Simmonds, P., I.
Chromatogr. A, 832, 253, 1999.
58. Schauffler, S.M., Atlas, E.L., Blake, D.R., Flocke, F., Lueb, RA ., Lee-Taylor,
TM., Stroud, V., and Travnicek, W., Distributions of brominated organic
compounds in the troposphere and lower stratosphere, I. Geophys. Res., 104,
21,513-21,535, 1999.
78
59. Burkholder, J.B., Wilson,- R.R., Gierczalc, T., Talukdarr R., McKeen, S.A.,
Orlando, J.J., Vaghjiani, G.L., and Ravishankara, A.R., Atmospheric Fate of
CF3Br, CF2Br2, CF2ClBr, and CF2BrCF2Br, J. Geophys. Res., 96, 5025-5043,
1991.
60. Connell, D.W., Basic Concepts of Environmental Chemistry, CRC Press LLC, 1997.
61. Culbertson, J.A., Piins, TM., and Giimsrud, E.P., Improvements in the Detection
and Analysis of CF3-Containing Compounds in the Background Atmosphere by Gas
Chromatography / High Resolution Mass Spectrometry, J. Chrom. A (In the press.).
62. Fraser, P.J., Oram, D.E., Reeves, C.E., Penkett, S.A., and McCulloch, A.,
Southern Hemispheric halon trends (1978-1998) and global halon emissions, J.
Geophys. Res., 104, 15,985-15,999, 1999.
APPENDIX A:
Electron Impact Mass Spectra
(Electron impact mass spectra were obtained from the National Institute of Standards and
Technology data base, which was included in the mass spectrometers software package.)
80
Rel. Abundance
CF4
s'o''' Vo''' Vo'1' Vo''' Vo'
’v'o
b'o' '!
Vo'' ' iio
Figure 37: Electron Impact mass spectrum of CF4.
81
CzE5
Figure 38: Electron Impact mass spectrum of CzF6.
I.
82
Rel. Abundance
C F 2C l
V d ' ' Vo ' '
'40'
so'' Vd Vo '' so ' V o ''160 iib' 'iib'
Figure 39: Electron Impact mass spectrum of CF 3 CI.
83
CFjH
63
\
65:
6 0:
Rel. Abundance
■55-
50:
31
45:
4 0:
35:
30:
.2 5 :
20 :
15.
.
5 I
10:
Sl
10
"20 ' ' 3 0
'' Vo ''
50 ’ '
Vo ' ' TO ‘' Vo '' Vd ' Ji 6 6 "
Figure 40: Electron Impact mass spectrum of CF3H.
.1
•••■**
•• Vw
’ m /z
84
C F 3B r
O
50.
C
40.
uo ' i i d ' 266
Figure 41: Electron Impact mass spectrum of CF 3 Br.
85
Rel. Abundance
CsEs
50
2'0' ' '4'0
so' ' s'o' ' 106 ' iid ' iid ' iid
166 ' 266 m/z
Figure 42: Electron Impact mass spectrum of CjFg.
86
Rel. Abundance
CF2 CF1Cl
Figure 43: Electron Impact mass spectrum of CF3 CF2 CI.
Rel. Abundance
87
101
Figure 44: Electron Impact mass spectrum of CF3 CF2 H.
I
88
CFtCH,
e's
t
65.
60 j
Rel. Abundance
55,
50
45
40
35
30.
. 3 5.
20J
1 5;
15
31
10 ;
45
5;
' ■ I 1O ■'
U t
20 ■■3'd ' ' 40 ' ' S1O ' ' Vo
fo '' so ’ s'o' 'iib
Figure 45: Electron Impact mass spectrum of CF3CH3.
m/z
89
•v
Rel. Abundance
CFtCFH-.
40
' '50 '' e'o
Figure 46: Electron Impact mass spectrum of CF3CFH2.
90
Rel. Abundance
CF2CFCl2
60
80
ilo ' iid ' 'i66 ' m/
Figure 47: Electron Impact mass spectrum of C F 3 C F C I 2 .
MONTANA STATE UNIVERSITY • BOZEMAN
3 1762 10346924