Survey and analysis of halogenated compounds in the atmosphere by gas chromatography/high
resolution mass spectrometry
by Mark A. Engen
A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy m
Chemistry
Montana State University
© Copyright by Mark A. Engen (1997)
Abstract:
The application of a dual-sector high resolution mass spectrometer (HRMS) to the detection of
halogen-containing compounds in the Earth's atmosphere is demonstrated. By this technique, the
contents of air samples collected in remote areas surrounding Bozeman, Montana, are introduced to the
HRMS by a cryogenic concentration method followed by gas chromatography. The HRMS is tuned to
the exact masses of the ions expected in the electron impact ionization mass spectrum of each
compound. Due to the large ion optic system of the mass spectrometer used here, any level of mass
resolution that is required for the complete separation of hydrocarbon-based from halocarbon-based
ions of the same nominal mass can be achieved while maintaining a high level of ion throughput to the
detector. By this technique, the analysis of compounds identified in background air while monitoring
either of the ions, CF3^+ or Br^+ will be discussed. Compounds identified while monitoring the Br^+
ion include CH3Br, CH2BrCl, CH2Br2, CHBr3, CF2Br2, CF2BrCF2Br, and CF2BrCl. Those
identified while monitoring the CF3^+ ion are CF4, CF3H, CF3C1, CF3CF2C1, CF3Br, and
CF3CFH2. Most of these compounds are present in background tropospheric air at the low to sub
parts-per-trillion by volume (pptv) level. To our knowledge, this is the first time CF3H and CF2Br2
have been detected in background tropospheric air. SURVEY AND ANALYSIS OF HALOGENATED COMPOUNDS
IN THE ATMOSPHERE BY
GAS CHROMATOGRAPHY / HIGH RESOLUTION MASS SPECTROMETRY
by
Mark A. Engen
A thesis submitted in partial fulfillment
of the requirements for the degree
Of
Doctor o f Philosophy
m
Chemistry
MONTANA STATE UNIVERSITY-SOZEMAN
Bozeman, Montana
August 1997
/
3>3'1^
APPROVAL
of a thesis submitted by
Mark A. Engen
This thesis has been read by each member o f 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 o f Graduate Studies.
Eric P. Grimsrud
David M. Dooley
Approved for the College o f Graduate Studies
Robert Brown
Date
Ill
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment o f the requirements for a doctoral
degree at Montana State University-B ozeman, I agree that the Library shall make it
available to borrowers under rules o f the Library. I further agree that copying o f this
thesis is allowable only for scholarly purposes, consistent with “fair use” as prescribed in
the U.S. Copyright Law. Requests for extensive copying or reproduction o f this thesis
should be referred to University Microfilms International, 300 North Zeeb Road, Ann.
Arbor, Michigan 48106, to whom I have granted “the exclusive right to reproduce and
distribute my dissertation in and from microform along with the non-exclusive right to
reproduce and distribute my abstract in any format in whole or in part.”
~
Z
IV
TABLE OF CONTENTS
Page
INTRODUCTION ............................... ...............
Atmospheric Chemistry .....................................................................................
I
11
Troposphere .....................................................................
H
Stratosphere..................................
15
The Antarctic Ozone Hole ....................................................................
22
Arctic Polar Vortex Chemistry.............................................................
23
The Mesosphere ........
EXPERIMENTAL PROCEDURES ......................................................................
24
26
Sample Collection ...................................
26
Sample Concentration and Clean-up ...
27
Concentration Method I ...................
27
Concentration Method 2 ...................
29
Chromatography ......................................
31
Methods for GS-Q Capillary Column
31
Methods for DB-VRX Column .......
32
Detection ................................................. :
33
The Mass Spectrometer ...................
33
Selected Ion Monitoring Procedure .
35
Standards ................... ...............................
40
Low Concentration Standards .........
42
High Concentration Standards .........
42
V
RESULTS AND DISCUSSION .............................................................................
Qualitative Determinations by GC-HRMS .................
44
44
Compounds indicated by monitoring the CF3+ ion .................................
44
Compounds indicated by monitoring the Br+ ion on the GS-Q
Capillary Column ...........................................................................:...
52
Compounds indicated by monitoring the Br+ ion using the DB-VRX
Capillary Column ................................................................ '.................
65
Quantitative determinations by GC-HRMS ....................................................
78
Quantitative Analysis o f CF3+ and Br+ Compounds with GS-Q
Capillary Column ......................
79
Quantitative Analysis o f Br+ Compounds with DB-VRX Capillary
Column ...................................................................................................
83
Detection Limits ....................................................................................
90
Analysis o f Select Bromocarbons from December 1996 to April
1997 ..................................................................................................
91
Conclusion ..........................................................................................................
92
SU M M A R Y ............................................................
94
REFERENCES .............................................................................................................
99
APPENDIX
108
Vl
LIST OF TABLES
Table
Page
1.
Perfluorinated and other long lived atmosphericspecies. ..............
4
2.
Hydrofluorocarbons and hydrochlorofluorocarbons; the
replacement refrigerants................................................................
5
3.
Brominated and iodinated compounds......................................................
6
4.
Exact masses of most abundant nuclides o f carbon, hydrogen,
fluorine, chlorine, bromine, oxygen, nitrogen, and sulfur........
9
Atmospheric concentration o f compounds analyzed with GS-Q
column................................................ 1..........................................
81
6.
Measurements o f CF2ClBr by NOAA/CMDL..................................
84
7.
Relative responses o f select brominated compounds toCF2ClBr.
88
8.
Atmospheric concentration o f brominated compounds analyzed
with DB-VRX column...................................................................
88
5.
Vll
LIST OF FIGURES
Figure
Page
1.
Physical description o f the atmosphere...............................................
12
2.
Schematic presenting the chemical interactions in the
stratosphere......................................................................................
18
3.
Schematic diagram o f the sample introduction system....................
30
4.
Electron impact mass spectrum o f the mass calibration gas,
n-dodecane. ............ .......................................................................
. 37
Mass spectrum o f a mixture of H-C12H26 and
CF2ClBr over
the mass range, m/z = 80 to 88, with the resolution o f the
mass spectrometer set to R = 2,000.............................................
41
CF3+single ion chromatograms using three different resolution
settings. In each chromatogram the relative responses are
normalized to the maximum response observed under each
condition. The amount o f air sampled is 340 atm-cm3.............
54
CHF37Cl+ single ion chromatograms using two different
resolution settings. The expected GC retention time o f
CHF2Cl is indicated in chromatogram A .....................................
55
Single ion chromatograms using two ions expected in the
electron impact spectrum o f CF4. The expected retention
time o f CF4 in indicated in chromatogram A .............................
56
Single ion chromatograms and expected retention time for CHF3.
57
5.
6.
7.
8.
9.
Vlll
10.
11.
12.
13.
14.
15.
16.
Single ion chromatograms and expected retention time for
CF3CL The isotopic pairs o f ions in chromatograms B and C
were simultaneously measured in a single analysis. The
value o f Ar in chromatogram C is the ratio o f this peak area
to that in chromatogram B .............................................................
58
Single ion chromatograms and expected retention time for
CF3CF2CL Chromatograms B and C and chromatograms E
and F (isotopic pairs) were simultaneously determined. Ar
values are the ratios o f the responses to the minor and major
isotopic ions.....................................................................................
59
Single ion chromatograms and expected retention time for
CF3Br. The responses to the isotopic pairs o f ions in
chromatograms B and C, D and E, and F and G were
simultaneously determined. Ar values are the ratios o f the
responses o f the minor 81Br-Containing ion to that o f the
major 79Br-Containing ion .................................................... .........
60
Single ion chromatograms and expected retention time for
CF3CH2F............................................... '..................... : ...................
61
81Brf single ion chromatograms using.four different resolution
settings. The amount o f air sampled is 340 atm-cm3........... •...
62
Single ion chromatograms and expected retention time for
CH3Br. The responses to the isotopic pairs o f ions in
chromatograms A and B and chromatograms C and D were
simultaneously determined. Ar values are the response ratios
o f the 81Br-Containing and the 79Br-Containing ions....................
63
Single ion chromatograms and expected retention time for
CF2BrCL The responses to the isotopic sets o f ions in
chromatograms A and B, C and D, E and F, and G, H, and I
were simultaneously determined...................................................
64
IX
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
8'Br+single ion chromatograms using three different resolution
settings. In each chromatogram the relative responses are
normalized to the maximum response observed under each
condition. The amount o f air sampled is 340 atm-cm3. ...........
69
Single ion chromatogram monitoring 81Br+ at resolution o f 2000.
Eight brominated compounds have been identified in rural
Montana background air; CF3Br, CF2ClBr, CH3Br, CF2Br2,
CF2BrCF2Br, CH2ClBr, CH2Br2, CHBr3..................................
70
Single ion chromatograms identifying the compound CF2ClBr, R
= 2000, RT = 4:00...........................................................................
71
Single ion chromatograms identifying CF2Br2, R = 2000, RT =
6:20....................................................................................................
72
Single ion chromatograms identifying CF2BrCF2Br, R = 2000,
RT = 8:10..........................................................................................
73
Single ion chromatograms identifying CH3Br, R = 2000, RT =
4:42................................................ ;................................................
74
Single ion chromatogram identifying CH2Br2, R = 2000, RT =
13:50. .............................. ........................................... : ...................
75
Single ion chromatograms identifying CH2ClBr, R = 2000, RT =
11:00..................................................................................................
76
Single ion chromatograms identifying CHBr3, R = 2000, RT =
18:20........ : ..................................................................... ................
77
Tropospheric concentration o f CF2ClBr measured at Niwat
Ridge, Boulder, CO by NOAA/CMDL from 1989 through
1995...................................................
86
X
Tl.
(Top) Tropospheric concentrations o f CF2ClBr and CH2Br for
the sampling period, December 1996 through April 1997.
(Bottom) Tropospheric concentrations o f CF2Br2,
CF2BrCF2Br, CH2Br2, and CHBr3 for the sampling period o f
December 1996 through April 1997 in the area o f Bozeman,
MT.....................................................................................................
93
28.
.EI mass spectrum o f CF2HCl...............................................................
109
29.
EI mass spectrum o f CF4................................................................... ..
110
30..
EI mass spectrum o f CF3H.....................................................................
I ll
31.
EI mass spectrum o f CF3Cl....................................................
112
32.
EI mass spectrum o f CF3Br..................................................................
113
33.
EI mass spectrum o f CF3CF2Cl...........................................................
114
34.
EI mass spectrum o f CF3CFH2.............................................................
115
35.
EI mass spectrum o f CF2ClBr..............................................................
116
36.
EI mass spectrum o f CF2Br2.................................................................
117
37.
EI mass spectrum o f CF2BrCF2Br............................
118
38.
EI mass spectrum o f CH3Br......................................................
119
39.
EI mass spectrum o f CH2ClBr.............................................................
120
40.
EI mass spectrum of CH2Br2. ..........................................
121
41.
EI mass spectrum o f CHBr3..................................................................
122
xi
ABSTRACT
The application o f a dual-sector high resolution mass spectrometer (HRMS) to the
detection o f halogen-containing compounds in the Earth's atmosphere is demonstrated.
By this technique, the contents o f air samples collected in remote areas surrounding
Bozeman, Montana, are introduced to the HRMS by a cryogenic concentration method
followed by gas chromatography. The HRMS is tuned to the exact masses o f the ions
expected in the electron impact ionization mass spectrum o f each compound. Due to the
large ion optic system o f the mass spectrometer used here, any level o f mass resolution
that is required for the.complete separation o f hydrocarbon-based from halocarbon-based
ions o f the same nominal mass can be achieved while maintaining a high level o f ion
throughput to the detector. By this technique, the analysis o f compounds identified in
background air while monitoring either o f the ions, CF3+ or Br+ will be discussed.
Compounds identified while monitoring the Br+ ion include CH3Br, CH2BrCl, CH2Br2,
CHBr3, CF2Br2, CF2BrCF2Br, and CF2BrCL Those identified while monitoring the CF3+
ion are CF4, CF3H, CF3Cl, CF3CF2Cl, CF3Br, and CF3CFH2. Most o f these compounds
are present in background tropospheric air at the low to sub parts-per-trillion by volume
(pptv) level.
To our knowledge, this is the first time CF3H and CF2Br2 have been
detected in background tropospheric air.
I
INTRODUCTION
During the last two decades, a great deal of attention
has
been
focused
on
the
environmental
significance
of
a
variety of halogenated compounds known to be present in the
Earth's atmosphere.
her
landmark
regarding
the
compounds
in
Carson's
effect
it
was
Silent
effects
the
was
thereafter,
so
book
book
DDT
Since 1962,
and
when Rachel
Spring
fates
environment
(I),
Carson wrote
public
concern
of
the. many
halogenated
has
steadily
increased.
brought
to
the
forefront
having
on
the
bird
the
detrimental
population.
Shortly
DDT was found to be toxic to human populations,
removed
1960's and
from use
in
this
country.
During
the
1970's a growing concern with pollution of o u r '
environment developed.
In
1971,
James
Lovelock,
developer
of
the
electron
/
capture
detector,
detected
chlorinated
background air samples he analyzed'.
compounds
in
all
One of the compounds he
identified was CFCl3 (CFC-Il) , a man-made compound used as a
refrigerant.
In 1975,
Grimsrud and Rasmussen detected and
2
measured CF2Cl2 (CFC-12)
These
were
the
halocarbons
in
first
and CH3Cl in background air
mass
background
spectrometry
air,
and
the
(2,3).
based
analyses
first
time
of
methyl
chloride had been detected in background air.
Sherwood
that
Rowland
and Mario
chlorofluorocarbons
stratospheric ozone.
Molina
•may
in
cause
layer,
aerosol,
canisters
propellants
were
removed,
from
Researchers
first
discovered
the
1985
(4) .
suggested
destruction
of
With increasing concern and public
awareness of the- possible effects C F C s
the. ozone
1974,
might be having on
which
the
used
market
Antarctic
CFCs
in
Ozone
as
1978.
Hole
in
This discovery accelerated policy changes which
would lead to a reduction in anthropogenic chlorine '(CFCs)
and
bromine
(Halons)
to
the
atmosphere.
In
1987,
many
countries
in the world convened and drafted a proposal
eliminate
CFCs
Montreal
amended
and Halons, which has become
Protocol.
twice,
in
The
1990
reduction
of
the .1992
amendments
bromine
halogenated
levels
will
it
Montreal
and
1992,
species
is
Protocol
in
estimated
decrease
to
has
order
in the
known as
to
since
speed
atmosphere.
that
levels
chlorine
such
that
to
the
been
the
With
and
the
3
Antarctic
year
Ozone Hole will no
2050.
In
1995,
P.
longer be present
Crut zen,
M.
around
Molina,
the
and
Sv
Rowland were awarded the Nobel Prize in Chemistry for their
early work
in stratospheric chemistry linking C F C s
destruction of stratospheric ozone,
importance
of
studying
the
to the
further reinforcing the
effects
of
anthropogenic
qhemicals in our natural environment.
Halogenated compounds containing chlorine,
bromine, or
iodine atoms have been extensively investigated with respect
to their potential for destroying ozone in the stratosphere
(5
-
11),
see
Tables
I,
compounds, including those
2,
and
3.
Also,
halogenated
containing only fluorine
atoms,
have been studied with respect to their effects on radiative
cooling of the Earth's surface
these
investigations,
(12 - 15).
the presence
In the course of
of numerous
halogenated
compounds in background air samples have been reported,
of
which
are
anthropogenic
halogenated
halogenated
chemical
thought
sources.
hydrocarbons
species, whose
means
by
which
to
have
With
as
natural
increased
use
replacements
use
has
halogen
been
atoms
as
well
of
for
some
partially
the
fully
discontinued,
are
as
the
transported
4
throughout
the
atmosphere
promises
to
become
increasingly
complex in the next several decades.
Table I . Perfluorinated and other long lived
atmospheric species.
Compound
Common
. Tropospheric
Name
Atmosoheric
Ozone
Concentration
Lifetime
Deoletina
(pptv)
(vears)
Potential.
CF2Cl2
CFC-12
120
0.9
CFCl3
.CFC-Il
60
I
U
2600
CFC-14
CzFc
CFC-116
80
(16)
3 (16)
0
1
Q
Tl
CF4
(15)
0
>50,000
(15)
0
>10,000
(15)
0
3200
(15)
0
2900
(15)
0
C5F12
4100 '(15)
0
CgF14
3100
(15)
0
1700
(15),
0.4
(CF3)2c -C4F6
C2F5Cl
CFC-115
4 (17)
C2F4Cl2
CFC-114
5
(17)
300
(15)
CF3Cl
CFC-13
10
(16)
640
(15) .
SF6
In
0.8
studying
halogenated
the
(35)
atmospheric
compounds, a
3200
(15)
chemistry
challenging
0.7
aspect
0
of
individual
has
bee n 1 and
will continue to be the means by which these compounds are
detected
and
widely-used
quantified
methods
in
real
(19) ' are
air
samples.
currently
based
The
most
on
the
■5
separation
with
of
detection
compounds
either
of
by
interest
electron
(22-27) or by mass spectrometer
by gas
capture
(MS)
chromatography
detector
(ECD)
(28,24,2,3).
Table 2. Hydrofluorocarbons and
hydrochlorofluorocarbons; the replacement refrigerants.
Compound
CHFCl2
Common
Tropospheric
Atmospheric
Ozone
Name
Concentration
Lifetime
•Denletina
(pptv)
(vears)
' HCFC-21
CHF2Cl
HCFC-22
CH3CF2Cl
HCFC-142b
0.1-2.5
80
(20)
(17)
2-3
Potential'
(6)
15 (21)
0.053
(21)
21 (21)
0.059
(21)
CH3CFCl2 HCFC-141b
8 (21)
0.081 .(21)
CF3CHCl2
HCFC-123
2 (21)
0.016
(21)
CF3CHClF
HCFC-124
8 (21) '
0.019
(21)
CF3CHF2 .
HFC-125
35 (21)
0 (21)
CF3CH2F
HFC-134a
21 (21)
0 (21)
CF3CH3
HFC-.143a
50
0 (21)
CHF2CH3
HFC-152a
CF3H
HFC-23
-
(21)
2 (21)
0 (21)
0
C3HCl2F2 HCFC-225ca
C3HCl2F2 HCFC-225cb
C2HF3BrCl HCFC-123B1
The
GC-MS
halogenated
approach
■
is
compounds
almost
because
equally
the
applicable
electron
to
all
impact ' (EI)
6
ionization process, on which
it
is based,
provides
almost
uniform ionization efficiency for all compounds.
Table 3
Compound
Brominated and iodinated compounds.
Common
Name
Tronosnherlc Atmosnheri c
Ozone
Concentration
Lifetime
Depletina
Ip.p-t.y)
(vears)
Potential
CH3Br
11
1.7
CH2BrCH2Br
1-37
(29)
CHBr3
2-46
(29)
CH2Br2
3-60
(29)
■ CF3Br
H-13 Ol
CF2Br2
H-1202
CF2ClBr
H-12il
CF2BrCF2Br
H-240'2
CHF2Br
■H-1201
(29)
2 (30)
72-107
1.5
2 (30)
0.5-1.0
(6)
(30)
(6)
(6)
2.5
CH3I
1.04
(32)
0.46
(32)
8-16
0.3
(6)
(6)
(6)
12-18
(6)
2.2-3.5
23-28
(6)
5-6
7 (6)
CH2BrCl
0.1-0.6
(6)'
(6)
0.4-1.8
(6)
(31) '
4 days
(32)
■'
CH2ClI
CH2I2
C2H5I
2 (32)
CF3I
CH2BrI
However,
for those compounds that have■large rate constants
for thermal electron attachment,
better sensitivity than the MS.
the ECD generally provides
For this reason,
the ECD
has been the method of choice for the quantitative analysis
7
of
atmospheric
halocarbons,
such
as
CF3Br ' (19),
that
have
very low atmospheric concentrations and very large electron
attachment rate constants.
In the present
resolution mass
study,
we report
spectrometer
(HRMS)
the use
for the
of a G C -high
identification
and quantitative analysis of atmospheric halocarbons.
resolution
should
provide
considerable
analysis of halogenated compounds,
exact masses
halogens
of
listed
the nuclides
in
Table
assistance
in
the
primarily because of the
of carbon,
4.
High
hydrogen,
Inspection
of
and the
this
table
indicates that ions that originate from molecules containing
only
hydrogen
and
carbon
will
have
exact
masses
greater than their nominal
values.
Those ions containing only carbon and halogen atoms
have
exact masses
nominal mass values.
bromine
atoms
deficiencies.
derivatives
will
that are
somewhat
less than their
tend
to
4,
have
it
particularly
is noted
that
containing nitrogen atoms will
greater
than
oxygen
and •sulfur
mass
Ions that contain either chlorine or
In Table
masses
integer)
are
somewhat
will
(nearest
that
the
have
nominal
val u e .
relatively
large
large
mass
hydrocarbon
also have exact
The
nuclides
negative
of
mass
8
deficiencies
halogenated
contain
the
could
interfere
compounds.
these
halogenated
from
and
elements
compounds,
halogenated
The
with
the
atmospheric
analysis
compounds
are generally much more
and
therefore
compounds
by
are
the
gas
that
polar
easily
of
than
separated
chromatographic
column.
With
a mass
resolution,
■the
spectrometer
mass
capable
differences
of moderate
shown
in
to high
Table
4
are
sufficient to allow complete separation of hydrocarbon-based
and halocarbon-based ions of the same nominal m a s s , thereby
eliminating hydrocarbon-based interference and noise.
the use of low resolution mass spectrometers,
With
the exact mass
differences shown in Table 4 are not sufficient to separate
hydrocarbon-based
nominal masses.
and
halocarbon-based
vapors
of
the
same
As a result, the multitude of hydrocarbons
present in an air sample,
residual
ions
of
in the GC flow gases,
a ' mass
spectrometer
itself,
and in the
typically
provide the major source of chemical interference and noise
in
the
GC-MS.
atmospheric
analyses
of
halogenated
compounds
by
9
Table 4. ■ Exact Masses of Most Abundant Nuclides of
Carbon, Hydrogen, Fluorine, Chlorine, Bromine, Oxygen,
Nitrogen, and Sulfur.
Nuclide
Exact Massa
Ami2
12C
12.0000
0
+ 0.00335
1H
1.01
+.0.00783
19F
19
0
35Cl
34.97
-0.03
37Cl
36.97
-0.03
79Br
78.92
-0.08
PQ
80.91
-0.08
14N
14
0
16O
15.99
-0.01
32S
31.97
-0.03
U
13
H
CO
13C
a
Taken from the Handbook of Chemistry and Physics,
69th Edition, CRC Press, Boca Raton, F L , 1989.
6
Mass excess or deficiency relative to the nominal
mass v a l u e .
Tf
the ■ high resolution mass
ion optic
system,
resolution
while
detector.
overcome
detection
it can achieve
maintaining
In this
conventional
GC-MS
with
and
spectrometer also
way,
GC-HRMS,
high
ion
described
leading
identification
a large
these required levels
the principal
technique
has
throughput
of
to
the
limitation of
the
above
directly
capabilities
to
should
be
superior
for
all
10
atmospheric
halocarbons.
expectations
is
atmospheric
The
demonstrated
analysis
system
realization
here
using
recently
of
a
these
HRMS-based
assembled
in
our
laboratory.
In this initial use of our HRMS-based analysis system,
we have
somewhat arbitrarily chosen to focus on two groups
of low-level atmospheric halocarbons that are detected when
either of the
ions,
CF3"1
" or Br+, is monitored while an air
sample is introduced by high resolution gas chromatography.
Following
the
detection
procedure; additional
of
GC-HRMS
a
compound
analyses
are
by
the
above
conducted using
most of the other major ions expected in the electron impact
ionization spectrum of the suspected compound.
Definitive
proof for the presence of each compound is thereby obtained.
Two of the compounds identified in the' present study,
and CF2Br2 have not
air.
A
analyses
possible
of
air
aluminum smelter.
been
determined,
anthropogenic
been previously detected
source
samples
of
CF3H
obtained
is
from
in background
suggested
a
CF3H
fume
from
hood, of
the
an
A source for the compound CF2Br2 has not
but
origin.
the
compound
‘.Before
is
almost
describing
certainly
of
experimental
11
procedure,
a
brief
discussion■ of
chemistry,
pertinent
to
the
fundamental
compounds
atmospheric
in 'this
study,
is
presented.
Atmospheric Chemistry
During
the
process'
of
developing
an
instrument,
developing methods,
and detecting and measuring some of the
trace
of
components
increasingly
interested
the
in
atmosphere,
the
complex
I
chemical
that take place in the Earth's atmosphere.
present
the
some basic
.halogenated
atmospheric
compounds
in
have
become
processes
I would like to
chemistry as
it applies
this
starting
study,
to
with
reactions involving hydroxyl radical in the troposphere.
Troposphere
The troposphere is the layer of the atmosphere that is
nearest
to
kilometers
the
Earth
(Figure I ) .
and
extends
up
to
around
eleven
12
Figure I. Physical description of the atmosphere.
13
This
region
experiences
significant
mixing,
both
horizontally and vertically, as is apparent from the variety
of weather produced in the troposphere.
released
at
ground
level
can be
A chemical, species
transported
to
the
upper
troposphere in a matter of hours, and if the compound has a
significantly
evenly
long
distributed
hemispherically)
The
tropospheric
associated
In
the
with
atmospheric
lifetime,
throughout
the
it
will
troposphere
become
(at
in a relatively short period of time
chemistry most pertinent
reactions
troposphere
the
involving
first
the
step
in
to
(32):
this work
hydroxyl
the
least
is
radical.
production
of
hydroxyl radical is the photolysis of ozone to oxygen, shown
in Equation
I.
.The atomic
oxygen
then
reacts
with water
molecules to produce hydroxyl radicals (Equation 2)
O3 + hv (< 320
O(1D) + H2O
For
compounds
containing
nm)
O2 + O(1D)
(I)
2OH*
a
carbon
(2,33) .
(2)
hydrogen-bond
reaction
with hydroxyl radical predominantly determines the lifetime
of
these
species
in
the
Equation 3, hydrogenated
atmosphere.
As
can
be
seen
in
compounds react with the hydroxyl
radical to form water and a reactive radical
(R"') (33) .
14
RH
+
-3» R"
+ H2O
+ Q2- —> [ RO2
]*
(3)
-»
HO2 + R a
(4a)
(4b)
P
R"
OH"
The reactive radical will then react with oxygen to form' an
oxygenated
radical
[RO2"] *.
The
collisionally
excited
stabilized
radical product
reaction
intermediate
in
state
(Eg.
the
excited
intermediate
4b ) ,
then
state
can
the
be
oxygenated
(RO2") can then proceed through .a variety of
pathways.
Otherwise,
the
reactive
intermediate
will dissociate to HO2 and a neutral compound
(Eg.
4a) .
general,
with
hydroxyl
hydrogenated
compounds
will
react
In
radical to form water soluble oxygenated compounds.
Because of the reactions initiated by hydroxyl radical
(Eguations 3 and 4),
C F C 's
contain
at
the compounds used as replacements for
least
one
hydrogen
atom
per
molecule.
Hydrogenation decreases the compound's atmospheric lifetime
by
allowing
therefore,
are
for
a
removal
in
the
troposphere;
a lower percentage of the replacement
expected to reach the
atmospheric
mechanism
lifetime
of
stratosphere.
CF3Cl
(CFC-13)
compounds
For example,
is
640
the
years,
15
compared to that of CHF2Cl
(HCFC-22)
at 15 years
(see Tables
I and 2).
■Stratosphere
The
stratosphere extends
from the tropopause
at about
eleven kilometers to the stratopause at about 50 km.
in the
stratosphere
Mixing
takes place horizontally with chemical
species diffusing up through the
stratosphere very slowly.
It can take 50 to 50 years for a compound to reach the upper
stratosphere.
The concern regarding ozone in the Earth's stratosphere
stems
from
surface
the
from
ability
harmful
of
ozone
to
ultraviolet
protect .the
radiation.
Earth's
Only
strongly absorbs.in the UV-B region of the spectrum
ozone
( 280 -
320 nm ), radiation which is particularly damaging to DNA in
living systems.
The
constant;
amount
it
place. . This
high
of
ozone
in
the
varies
monthly,
natural
variation
latitudes.
eleven-year
i
Factoring
solar cycle,
stratosphere
yearly,
in
can be
and
as
seasonal
and other
is
from
large
far
from
place
as
25%
variations,
to
at
the
complicating variables.
16
atmospheric scientists use statistical methods to reveal the
long-term global thinning of ozone believed to be caused by
halogen-catalyzed reactions with ozone.
Ozone is produced in the upper stratosphere where ultra
violet
radiation
oxygen ■ (Eq.
molecular
dissociates
5) .
oxygen
The
to
molecular
oxygen
atomic < oxygen
form
ozone,
then
as . shown
to
atomic
reacts
with
in' Equation
6
(4,8,14) .
02 + hv (X< 240 nm)
0 + O2
—>
->
20
(5)
O3
(6)
These reactions occur rapidly in the stratosphere over
the tropics where solar radiation is most intense.
the
away
circulation
from
of
the
the
concentrations
upper
equator
are
atmosphere
towards
generally
Because
constantly
the
greater
poles,
at
the
moves
ozone
higher
latitudes.
Ozone is destroyed when it absorbs radiation that would
otherwise reach the Earth's surface
03
There
is
no
oxygen is formed
+
hv (X< 1180 nm)
net
(Eq.
(Eq.
reduction
6).
in
->
7).
O2 + 0
ozone
because
(7)
atomic
The atomic oxygen can then react
\
17
with
molecular
continually
oxygen
destroyed
to
reform
through
ozone.
reactions
Also,
with
ozone
is
families
of
naturally occurring radicals that contain chlorine, bromine,
nitrogen, hydrogen, or oxygen atoms, see Figure 2 (4,10).
The
above
mechanisms
mentioned
decades.
With
as
chlorine
atmosphere
into
the
the
by
' destruction
concentration
that is until .the last several
widespread
and
use
of
Halons
chlorofluorocarbons
(fluorocarbons
that
in fire extinguishers, significant amounts
and
the
bromine
escape
environment.
hydroxyl
of
the
refrigerants
contain bromine)
of
and
lead to a relatively steady-state
of ozone in the stratosphere,
(CFCs)
production
have
or
been
introduced
discharge
Because
of
C F C 's. do
into
these
not
the
chemicals
react .with
radical they are inert to the oxidative processes
troposphere.
transported
to
the
As
a
result,
they
stratosphere
and
are
efficiently
are
primarily
responsible for the increase of chlorine in the stratosphere
from about
1990
(8).
0.8 ppbv fifty years
ago to around 4.0 ppbv in
Current levels of bromine in the stratosphere due
both to natural and anthropogenic sources are between 10 and
20 pptv
(5) or about 0.5% as abundant as chlorine.
18
Ralnout.
washout
CIOH
(troposphere)
O2 + hv
NO + O)
O +O,
CIONO2
Catalytic reactions
with N O * H O /. C IO /
Ionosphere
_ l 2 N + e1 n +N* + 2 e ‘
O(D) + H2O
Stratosphere
2 NO
2 HO*
Ralnout, washout
(troposphere)
HO- + CH4
HO- + HO2-
Figure 2 . Schematic presenting the chemical
interactions in the stratosphere.
H2O + Og
2 HO*
19
Even though bromine concentration is significantly less than
that
of
chlorine,
bromine's-, synergistic
chlorinated compounds
(Eg.
of
for
ozone
accounts
destruction
implicated
due
to
loss of ozone
1991-1992
to
with
14) in the catalytic destruction
approximately
halogenated
account
effect
for
25%
compounds.
as much
as
35%
of
of
«
ozone
Bromine
the
was
chemical
in the arctic polar vortex in the winter of
(34) .
What are the mechanisms by which chlorine
and bromine destroy ozone?
Once CFO's and Halons rise above the bulk of the ozone
layer, which lies between 15 and 30 km altitude depending on
latitude,
they are photolyzed by ultraviolet light producing
halogen atoms.
For Example
(8) :
CCl2F2 + hv (<220 nm)
The
atomic
participate
example,
in
chlorine,
catalytic
->
bromine,
cycles
Cl + CClF2
or
iodine
(8)
can
then
which destroy ozone.
For
chlorine atoms will react with ozone as follows:
Cl + .O3
—>
ClO + O2
CIO + O
-4
Cl + o.
O3 + 0
—^
20%
(net}
(9)
20
Also, the CIO radical can react with another ClO radical
(Eg.
10),
ClO + CIO + M
->
Cl2O2 + M
to produce Cl2O2, which if radiated by UV light
Cl2O2 + hv
OClO
->
(Eg.
11)
OClO + Cl
(1 1 )
Cl + O2
regenerates chlorine atoms
explanation of
(10)
chlorine's
(12)
(Eg.
12).
role
in the
A more comprehensive
stratosphere
can be
seen in the schematic shown in Figure 2.
Similar to Eguation 9, atomic bromine has the net
effect of reducing ozone to molecular oxygen:
Br + O3
-»
BrO + O2
BrO + 0
—>
Br +
O3 + 0
—>
2O2
(13)
O1
(net)
There are two primary reactions that make bromine a better
catalyst at destroying ozone than chlorine.
First,
more
the
abundant
fast
ClO
synergistic
radical
and
reaction between the much
the
BrO
producing atomic chlorine and bromine.
radical
(Eg.
14)
21
BrO + CIO
->
Br + Cl + O2
. (14)
The atomic halogens then continue the catalytic destruction
of ozone
(Equations 9 and 13).
Secondly, the sink
(a stable
nonreactive species in the stratosphere that will eventually
diffuse back into the troposphere where it is removed from
the
atmosphere
nearly
as
via
various
stable
as
stratospheric
sink
reacting
NO2 to
with
pathways)
its
for
chlorine
chlorine
form
for
the
bromine
not
counterpart.
involves
stable
is
CIO
product
The
radical
ClONO2
(Eq.
15) .
CIO + NO2
Chloronitrate
special
(35) .
is
stable
cases where
The
bromonitrate
in
the
between
(15)
stratosphere
heterogeneous
reaction
(Eq.
ClONO2
->
chemical
BrO
and
except
reactions
NO2
in
occur
produces
16) , but BrONO2 is not nearly as stable
as chloronitrate and the equilibrium lies to the left
(Eq.
16) with the reactive species BrO being an abundant species
of bromine in the stratosphere
Br O + N O 2
-»
(36,37).
BrONO2
(16)
22
The Antarctic Ozone Hole
is due
to a special
case
The Antarctic
ozone hole
in stratospheric
chemistry where
the generally nonreactive forms of chlorine
(ClONO2 and H C l )
are converted to reactive species by heterogeneous- chemical
reactions.
is formed
During the polar winters,
(polar vortex), and within the polar vortex polar,
stratospheric
dark,
clouds
polar winters
when temperatures
of
HNO3
an isolated a i r 'mass
arid
Hydrochloric
(PSCs)
.
acid
form
during
the
cold,
P S C s form in the lower stratosphere
drop below
water
will
-77 °C and sufficient
condense
will ' freeze
on
on
sulfate
the
amounts
particles
surface ' of
(4).
these
particles and then will react with gaseous chloronitrate in
a surface-assisted reaction producing solid nitric acid and.
chlorine gas molecules
HCl
(Equation 17)
(5,38).
(s) + ClONO2 (g) -> Cl2 (g) + HNO3 (s)
(17)
When the sun returns in the spring and the polar vortex is
still ■in
place,
the
molecular
chlorine
is
photolyzed
to
atomic chlorine.
Cl2 + hv —^
2Cl
(18)
23
The atomic chlorine can then react with ozone in catalytic
cycles
(Eq.
terminating
9-12,14).
step
ineffective.
This
the
little
catalytic
NO2
cycle
present,
(Eq.
the
.15)
is
In this case, heterogeneous chemistry causes a
denitrification
normally
to
With
process
stable
process
which
reserve
causes
an
to
releases
reactive
unrestrained
chlorine
chlorine
from
(Eq.
destruction
its
17).
of
ozone
that can go on for Several weeks,
until the air masses are
warmed
up
sufficiently
to
break
the
polar
vortex.
Calculations of global ozone thinning contribute about 3% of
the 6% to 8% per decade loss of ozone to this heterogeneous
chemistry (14).
Arctic Polar Vortex Chemistry
During
the
arctic
winters when the polar vortex is in place, the same chemical
species
are
formed
and
in
concentrations as in the Antarctic.
approximately
the
same
Why is it then that the
ozone hole only forms at the south pole? It is primarily due
to
the
fact
that
arctic
winters
are
slightly warmer
than
those at the south po l e , allowing warm spring air masses to
push
into
the
north polar
region and break
up
the
arctic
24
polar vortex before the sun returns in spring.
Therefore,
it is probable that in the next several decades,
levels
elevated,
we will
experience a colder and longer winter than average.
In this
event,
of
chlorinated
if the
species
are
still
while the
arctic polar vortex were to remain in place
for several weeks after the polar sun rise we would probably
experience
an arctic
ozone hole
due to the
same
catalytic
■cycles as in the south p o l e .
The Mesosphere
As can be seen in Figure I, the mesosphere extends from
the stratopause at about 50 km to the mesopause at around 85
km.
The mesospheric
chemistry of
interest
to the present
study is associated with the extremely stable perfluorinated
compounds
(PFCs) .
oxidative
properties
properties
of
the
PFCs
of
are not
the
effected by either,
troposphere
stratosphere..
or
the
Ravishankara
the
photolytic
et al.
(15)
has suggested that the major means of destruction for P F C s
will
result
from the
absorption of Lyman-a
121.6 n m ) , in the mesosphere .
radiation
(I =
As a result of the inertness
of P F C s , they have extremely long lifetimes
(>10,000 years.
25
see
Table
I)
allowing
these
green
house
that
some
gases
(15,39).
to
accumulate for millennia.
It
can
be
speculated
replacement compounds,
their
quite
way
to
the
reactive
troposphere
stratosphere.
the
active,
and
to ' the
3),
Even
though
oxidative
a
in
compounds
percentage
the
will
absence
their fate in the mesosphere.
of
not
HFCs
properties
Once in the stratosphere,
tiydrofluoro
the
new
CFC
such as CF3CFH2 (HFC-134a) ,. will find
mesosphere.
(Eq.
of
will
will
be
of
the
reach
the
it is likely that
be very photolytically
hydroxyl
radical
may
find
26
EXPERIMENTAL PROCEDURES.
Sample Collection
Air
samples were
mountainous
areas
Montana.
It
has
collected
in
collected
within
been
these
a
in a variety of rural
10-mile
assumed
local
radius
here
areas
that
are
of
and
Bozeman,
the
samples
representative
of
background tropospheric air of this region of North America.
Sample containers are commercially-purchased stainless-steel
vessels whose
The
air
to
containers
liters'
per
interior
be
for
surfaces
sampled
at
minute
was
least
by
have been electropolished.
first
3 0 minutes
a
Neuberger, model N05SVI).
DCrpowered
forced
at
a
through
rate
diaphragm
of
these
about
pump
5
(KNF
The containers were then filled
to above-ambient pressure levels prior to transport to the
Mass Spectrometry Facility at Montana State University.
27
Sample Concentration and Clean-up
Two methods
were
used to
concentrate
interest from whole air samples.
the
analytes
Method I was developed to
trap the very volatile highly fluorinated compounds
points 'between -165 and--25 °C) .
assure
quantitative
desorption
of
Method 2
of
less
(boiling
was necessary to
volatile
compounds
from the trap.
Concentration Method I
Air
samples
were
introduced
into
the
sampling system represented in Figure 3.
to
1000
atm-cm3 of whole
air can be
the
approximately
13
liter
six-port
through
valve
a
(6PV)
freeze-out
liquid , nitrogen,
and
then
loop
into
After the connection
the transfer line and
volume
A metering valve
allowed
the
air
(FOL), which
the
the
Sample sizes of up
expansion
evacuated with a rotary p u m p .
by
concentrated with the
typical sample ,size being 340 atm-cm3.
of a flask containing an air sample^
GC-HRMS
were
(MV)
and a
sample
was
evacuated
(EV)
to pass
submerged
volume.
in
The
28
freeze-out
column,
loop
(J&W
consisted
GS-Q
of
0.53-mm
a
ID)
fused-silica
of
capillary
approximately
twenty
centimeters in length.
The transfer lines to and from the
freeze-out
made
loop
were
from
uncoated,
0.53-mm ID fused-silica capillary column.
sample
flow
through
the
freeze-out
deactivated,
The rate of air
loop
was
regulated'
(approximately 5 0 cm3 min"1) for all samples by opening the
metering valve until the pressure read at the pressure gauge
(PG)
reached some preselected value.
The water content of
the air was removed prior to the freeze out loop by passing
the
air
through a drying tube
constructed
of
approximately
coarsely
placed
0.25-in
GD
line
Mg(ClO4)2..
after
the
read
sampling
on
flask was
the
Baratron
in
A
tube
tube was
steel
tubing,
filter
to
of
the
between
valve
A
expansion
and
the
(PF)
remove
the
gauge
change
(BPG)
internal volume of the concentration system.
volume,
packed
with
was
particles
The amount of air taken from
deduced by
pressure
drying
length,
particle
drying
greater than ten micrometers.
the
The
stainless
four ■ centimeters
ground
in
(DT) .
volume
and
all
Baratron pressure
in pressure
and
the
known
The internal
transfer
gauge,
is
lines
13.3
29
liters,
see Figure 3.
Once the desired amount of sample had
passed through the freeze-out loop valve A was closed,
valve
C
was
closed.
Nex t , valve
D
was
opened
then
and
the
freeze-out loop is pumped down to a pressure of 0.05 mbar.
I
This process helps to remove most of the excess nitrogen and
oxygen from the freeze-out loop.
Finally,
the 6-port valve
was rotated and the freeze-out loop was heated by a heat gun
set at 400 degrees Fahrenheit.
onto the analytical
The analytes were desorbed
column and the temperature program for
the GC was started.
Concentration Method 2
Concentration method 2 follows the same procedures
method
I.
which was
packed
The only difference being the
constructed of
with
silanized
freeze-out
l/8-in OD .stainless
glass
beads
of
60/80
steal
mesh.
as
loop,
tubing
After
desorption from the FOL, the analytes were refocused on the
head of the capillary column by cooling a short piece of the
analytical
column ' to
liquid
nitrogen
temperatures.
The
cryogenically focused analytes were then flash heated with
warm air and the temperature program for the GC was started.
30
Figure 3 . Schematic diagram of the sample introduction
system.
31
Chromatography
Two primary chromatographic methods were used in order
to
■ achieve
optimum
separation.
chromatographic
Chromatography
of
the
■resolution
highly
and
fluorinated
compounds proved especially challenging. .
Methods for GS-O Capillary Column
A J&W GS-Q gas-solid,
column
of
3 0 meters
separation
of
the
compounds; all
ion.
This
length
and
0.32
perfluorinated
compounds
column
volatile compounds.
porous layer open tubular
was
which
and
mm
was
highly
contained
designed
ID
the
(PLOT)
used
for
fluorinated
CF/
specifically
fragment
for
very
The chromatographic resolution of this
column was quite poor,
but it was the only capillary column
tried in which separation of the most volatile compounds was
achieved.
Two temperature programs were followed.
Temperature program I was used for CF3C l , CF3Br, C2F5C l ,
and CF3CFH2.
2 minutes,
The initial GC temperature was held at 3 0°C for
then the oven was ramped up at 25°C per minute to
22 0°C and held for 12 minutes.
Temperature program 2, used
32
for
CF4
and
CF3H,
was
the
same
except
for
an
initial
temperature of -10°C.
The flow rate of helium through the GC column was set
so
that
initial
an
unretained
compound ■(such as
linear velocity
of
25
cm
nitrogen)
sec"1 through
had
the
an
column
with a constant head pressure of 60 kPa.
Methods for DB-VRX Column
The DB-VRX column was used for the analysis of all of
the brominated compounds except'CF3Br
the GS-Q) .
in .length
microns.
(CF3Br was analyzed on
The GC contained a J&W DB-VRX column 60 meters
by
0.25
mm
ID
with
a
film
thickness
of
1.4
A combination of temperature and pressure programs
were used to obtain optimum separation and resolution.
The
oven was initially held at 30 degrees Celsius for 5 minutes
followed by a temperature- ramp o f .15 degrees-per-minute to
220
°C.
-The
temperature was maintained at
remainder of the 25 minute run t i m e .
used to
reduce
the
220
°C for the
A pressure program was
time necessary for moving the
from the FOL to the point of cryofocusing.
analytes
The initial head
pressure was set at ,200 kPa and held for 0.5 minutes, then
33
the-pressure was ramped downward-at 10 kPa per minute to 100
kPa and maintained for the remainder of the r u n .
end of
the
capillary GC column was
The exit
threaded directly into
the ion source of the mass spectrometer.
Detection
The Mass Spectrometer
The instrument used for this work was a VG ZAB-2F
.Instruments)
purchased
1978,
the
by
high resolution mass spectrometer.
Battelle
Northwest
Research
Originally
Laboratories
this instrument was granted to MSU in 1992.
instrument
was
thoroughly
refurbished
data
system
chromatograph
(Fisons, model
(Hewlett-Packard,
VG
model
OPUS)
and
5890
with liquid N2 oven cooling)
were added.
spectrometer
(magnetic/
is
a
dual
and
in
At M S U ,
numerous
component parts were either replaced or upgraded.
HRMS
(VG
A new
a
Series
new
gas
II Plus,
The ZAB-2F mass
electric)
sector
instrument with maximum ion acceleration voltage of +8.0 kV.
The relatively large ion optic system of the ZAB-2F provides
continuously
variable
levels
of
mass
resolution
up
to
34
100,000
and
an
unusually
transmission
to
Its
detector
present
discreet
which
the
dynode
is
type
mounted
stainless-steel
high
detector
is
at
an
any
90°
electron
off
the
axis
conversion dynode
for
multiplier
Inc.
of
ion
resolution.
of
the
, model AF130)
the
ion
beam.
A
ions
are
(where positive
is also located at. 90° off axis and
is negatively biased up to -5.0 kV.
instrument
selected
(ETP Scientific,
converted to electrons)
the
efficiency
is
The mass resolution of
conveniently
selected
by
externally-controlled variations of the source and detector
slits.
Any level of resolution
(M/Am, where M- is the mass
of an ion and Am is the peak width at half height)
from a
minimum
can
value
selected.
of
Since
about
the
1000
up
to
efficiency
of
about
ion
100,000
throughput
to
be
the
detector is increased with use of wider slits, maximum slit
widths, providing the minimum level of required resolution,
are
typically
spectrometer
used.
has
been
Since
used
being
only
relatively clean air and gas samples.
refurbished,
for
the
this
analysis
mass
'of
35
Selected Ion Monitoring Prnr=Adnrm
In order to monitor the intensity of an ion of interest
while maintaining moderate or high resolution conditions,
is
necessary
fields
of
to
continuously
calibrate
the
it
controlling
the mass, spectrometer to insure that
it
focuses
precisely at the exact mass of interest when instructed to
do so by the data system.
For this purpose n-dodecane is
continuously introduced to the ion source via the heated gas
expansion
volume.
As
shown
in
Figure
4,
the
electron
impact spectrum of n-dodecane consists of numerous fragment
ions
and
a
identities
known
molecular
and
exact
(for example,
170.20358).
ion
masses
of
of
nominal
all
of
m/z
=
170.
the
major
The
ions
are
the exact mass of the molecular ion is
In preparation for a specific analysis, a mass
calibration file is first made by performing a voltage scan
in which the accelerating voltage
and
electric
sector
field
are
(initially set at 8.0 kV)
scanned
downward
in unison
while the magnetic sector field is held constant.
This scan
must be performed over a narrow mass range that begins and
ends with any two of the reference ions in the spectrum of
n-dodecane
(Figure
4) .
The mass
range m u s t ' be
narrow
so
36
that the accelerating voltage does not decrease to a value
of less that 5 kV, where an unacceptable loss of sensitivity
would occur.
Using
this
instructs
the
calibration
mass
file,
spectrometer
the
to
data
jump
system
between
then
any
one
selected ion of n-dodecane and any other ■exact mass or set
of exact masses within the calibrated range.
A duty cycle
of 0.5 seconds was typically used here during which the mass
spectrometer spent 0.10 second on the selected reference ion
and divided the remaining 0.40 seconds among the set of ions
selected
for
observation.
The
settling
voltage jumps is very fast
(about 10 ms)
for
for
our
strong
preference
electric
time
between
and this accounts
field
rather
than
magnetic field control of the instrument.
An
method
inherent
for
the
acknowledged,
limitation
analysis
is
in
of
of
the
GC
electric
effluents,
interpreting
the
field
that
signals
control
must
be
obtained.
Under this mode of operation the relative intensities of the
different
ions
observed
expected to be the
for
a
given
compound
same as those appearing
mass spectrum for that compound.
are
not
in a reference
Relative Intensity
37
m/z
Figure 4 . Electron impact mass spectrum of the mass
calibration g a s , n-dodecane.
38
This ■ is
because
scanning
the
held
reference' mass
magnetic
constant.
spectra
field while
For
the
example, when
are
obtained
electric
the
by
fields
individual
are
ions
within the 35C1/37C1 and 79BrZ81Br isotopic cluster groups are
monitored
field,
the
isotopic
from
by
sequential,
measured
ions will
the
known
slightly
lower
heavier
multiple ■ jumps
relative
tend to be
isotopic
ions
voltage
into
focus.
is
the
for
slightly less
abundances.
accelerating
isotopic
intensities
of
electric
the
heavier
than expected
This
is. because
used
to
This
will
bring
a
the
slightly
decrease b o t h .the efficiency of ion extraction from the ion
source
and
the
sensitivity
-massive isotopic ion.
the
detector
to
the
more
This is due to ion optic properties
of the magnetic sector,
inversely
of
and as seen in Equation 19 mass is
proportional
to ■ accelerating
voltage
when
describing the focusing properties of the magnetic sector
= mass,
z = charge,
B = magnetic
field,
r = radius,
(m
V
=
accelerating voltage),' therefore lower accelerating voltages
are necessary to pass
ions of
increasing mass
through the
magnetic sector when it.is operated at a constant field.
m/z = B 2r2/2V
(19)
39
In
preparation
for
a
given
analysis,
it
was
also
necessary to know the minimum level of mass resolution that
would
provide
interest
good
separation
of
the
halocarbon
f r o m . hydrocarbon-based 'ions
including
those
produced
by
the
from
ions
of
all ' sources
reference
compound,
n-dodecane, that was continuously present in the ion source.
In addressing this issue,
it is instructive to consider the
abbreviated mass spectrum shown in Figure 5 of a mixture of
n-dodecane
and CF2ClBr that was
continuously bled into the
ion source via the heated gas expansion volume.
It is seen
in this spectrum that a resolution setting of M/Am = 2,000
provides
and
good
C6H 13"1",
masses
of
baseline
that
these
m/Am = 625
appear
ions
separation
at
are
nominal
84.96885
of
the
mass
two
ions,
85.
The
and 85.10173,
CF2Cl+
exact
for which
(where m and Am are the average and difference
in m a s s , respectively).
As illustrated in this case,
the
level of resolution required to provide baseline separation
between halocarbon-based and hydrocarbon-based ions is about
three times the m/Am value of the two ions to be separated.
Of
course,
lower
,resolutions
than
this
are
sometimes
40
adequate
in
those
interference's
fortuitous
are
present
cases
at
where
the
no
hydrocarbon
nominal
masses
of
interest.
Standards
The preparation of standards is an integral part of all
quantitative,
challenging
standards
success
The
good
and
has
proven
aspect
of
this
project.
were
prepared
(low to
highly
points
work
in
our
to
be
Low
laboratory
an
especially
concentration
with
sub parts-per-trillion by volume
fluorinated
and
low .boiling
compounds
limited
levels).
(boiling
< 0°C) proved to be the most predictable and showed
stability
Hydrogenated
when
stored
compounds
with
for
extended
boiling
periods
points
of
greater
time'
than
approximately O0C and in particular the stability of methyl
bromide
(CH3Br) ,
unpredictable
at
stored
low
in
flasks,
concentrations
part-per-million levels).
proved
to
be.
(concentrations
very
below
Relative Intensity
41
Figure 5. Mass spectrum of a mixture of U-C12H26 and
CF2ClBr over the mass range, m/z = 80 to 88, with the
resolution of the mass spectrometer set to R = 2,000.
42
Low Concentration Standards
Low
concentration
prepared
by
substance
glass
the
of
ambient
successive
interest
carboys.
standards
into a series
with
standard
laboratory
are
sub
pptv)
pure
of three
gaseous
twenty liter
pressurized
that
was
to
void
above
of
estimated
accuracy's
at ±50%.
in
our
analysis
of
the
dilution's
a relative error of ±8% can be assigned.
is
attributed
with
achieved
of
error
associated
In an
to
making
wall
the
(best
propagation
remaining
were
For the most inert compounds
preparation
error
gas
to
of .the
were
nitrogen
compounds of interest.
behaved)
dilution
The . carboys
pressure
(low
3
successive
effects,
The
either
adsorption onto the wall' and/or contamination of the flask
walls .(40,41).
well
Standards prepared from the compounds in the
behaved list
include CF4, CF3H,
CF3C l , CF3CF2C l , CF3Br,
CF3CFH2, CF2ClBr.
High Concentration Standards
'
Standard
concentration
mixtures
containing
of
CH3Br,
approximately
CF2Br2,
'200
C2F4Br2,
CH2Br2, CHBr3, and CF2ClBr were prepared in our
ppmv
CH2ClBry
laboratory.
43
The mixture was prepared by injecting a pure aliquot of each
compound into a five liter glass flask and then pressurizing
the flask to above ambient pressures with nitrogen g a s .
calculated
relative
error
associated
with
this
The
single
dilution technique is ±5% and this is in agreement with the
several standards produced in our laboratory.
have
been
standards
reported
in
concentration
the
to
high
mixture
necessary before
have
use.
less ^ significance
ppmv
proved
Wall effects
range
stable,
Mixing was
for
(41) .
The
although
mixing
gas
high
was
accomplished by heating
the outside of the flask with a heat gun.
44
RESULTS AND DISCUSSION
Qualitative Determinations bv CC-NRMS,
The
discussions
fourteen
Figures
electron
28-41,
to
follow will
impact
in
the
ionization
be
facilitated by
mass
Appendix.
spectra
Inspection
the
shown
of
in
these
spectra indicates that the first seven compounds have an ion
of
nominal
mass
69
in
their
spectra
and
that
all
of
the
brominated compounds have ions of mass 79 and 81.
Compounds indicated by monitoring
the CF2+ ion
Compounds identified in this study were analyzed using
the
GS-Q
Figure
capillary column and concentration method
6,
three
atm-cm3 of
clean
spectrometer
68.9952,
has
single
ion
outdoor
been
of the ion,
chromatograms,
air,
tuned
are
to
shown
the
I.
In
analyzing
340
where
exact
the
mass,
mass
m/z
CF3+, using three different resolution
settings.
With use of the lowest resolution setting of our
instrument
(R = 700) ,in Figure 6A,
a relatively high level
45
of
baseline
because
other
the
noise
mass
is
observed.
This
deficiency ■of
hydrocarbon-based 'ions
the
of
is
not
CF3+ ion
nominal
surprising
is
mass
small
69
would
and
be
expected to contribute to the baseline noise level at this
resolution setting.
As the resolution is increased to 2000,
most of the chemical noise is removed.
Further increase in
resolution to 4500 results in the loss of the last peak at a
retention time of 7.65 minutes,
thereby indicating that the
compound
response
have
responsible
the
ion,
for
CF3+
this
in
its
mass
does
not
actually
spectrum.
The
signal-to-noise response to the remaining CF3+ peaks is not
actually as high in Figure 6C as in 6 B .
This is due to the
expected reduction of ion throughput accompanying the use of
increased
resolution
Therefore,
a slight
in
Figure
6C
relative
to
6B.
increase- in the relative importance of
residual instrumental noise is detectable in Figure 6C.
The
typical
method used
individual
compounds
identified
by
is
matching
as
GC
to determine
follows.
retention
the
Each
times
identity of ■
compound
to
is
standards,
detecting many of the major fragment ions present in the EI
46
mass
spectrum,
clusters
of
and
ions
abundance ratios
by
in
simultaneously
order
to
monitoring
determine
the
the
isotopic
(Ar) .
The last -peak shown in Figures 6A and 6B,
to
isotopic
compound
CF2HCl
(HCFC-22) .
compound is shown in Figure
corresponds
Verification
of
this
where the exact mass of the
I
ion CFH37Cl+ was monitored, m/z = 68.9721.
This ion and the
retention time indicated in Figure 7A were expected for the
compound,
CHF2C l ,
which
is
atmosphere at a concentration
is
quite
high
highlighted
in
relative
the
to
present
known
to
be
present
(about 125 pptv
those
of
the
study.
the
(42,43)) which
compounds
As
in
to
expected,
be
the
relative response
to CHF2Cl was greatly increased and that
to
molecules
CF3-Containing
resolution
of
the
mass
4500 in Figure 7 B .
change
was
was
greatly
decreased
as
spectrometer was
increased
to R
by
a
=
It should be noted that the remarkable
in the two measurements
caused
the
change
selected exact mass.
of
shown in Figures
only
0.02
mass
6C and 7B
units
in
the
This provides an excellent example of
the increased utility of HRMS over that of low resolution MS
47
for the definitive analysis of multicomponent samples.
is
also
trace
a good
example
atmospheric
interference's
in
of
how
component
the
a minor
(100
analysis
to
of
fragment
1000
ultra
ion of
pptv)
trace
This
a
can
cause
(<10
pptv)
components of the atmosphere.
■In Figure 8, another compound,
CF4, known to be present
in the atmosphere at levels in excess of the low pptv range
has been
m/z
identified.
The only two
ions
(CF3+ and CF2"1
") of
> 42 having significant relative intensity in its mass
spectrum were monitored
(Fig.
29 in Appendix).
. Both ions
expected for CF4 are clearly evident in chromatograms 8A and
SB.
CF4
(CFC-14)
production
of
is
a
aluminum
by
product
by • the
of
the
industrial
Hall-Heroult
process
(44-46,39) ."
In
Figure
9,
the
expected
noted and the three major ions
mass
spectrum
with
detectable
masses.
(Fig.
retention
of
CF3H
is
(CF3+, CF2H+, and CF2"1") in its
3 0 in Appendix)
responses
time
have
observed at each
been
of
monitored
these
exact
The large response noted in Figure 9C and a minor
one in Figure 9B at 7.6 minutes is due to CHF2Cl
(see mass
48
spectrum in Appendix,
9B
is
due
to
Fig.
28).
CF2Cl2 whose
known to be very high,
The large peak in Figure
presence
in
about 540 pptv.
the
atmosphere
is
The origin of .the
last peaks noted in Figures 9B and 9C is presently unknown.
To our knowledge,
detected
in
the compound CF3H has not been -previously
background
air.
It
seems
reasonable
to
speculate that a source of this compound m i g h t .be associated
with
the
industrial
production
Hall-Heroult process.
C2F6 has
been
(44,45,46,39) .'
from
the
of
aluminum
by
the
The atmospheric presence of CF4 and
associated
with
the
For this reason,
ventilation
exhausts
production
of
aluminum
two air samples collected
of
an
analyzed in our lab by GC-HRMS.
aluminum
plant
were
Elevated levels of CF3H,
as well as CF4 and C2F6, were found in both samples.
sV
In
Figure
(CFC-13)
is noted
CF237CI+) in
monitored
these.
10,
its
with
the
expected
and three
spectrum
major
(Fig.
detectable
For this compound,
retention
ions
31
responses
time
of
CF3Cl
(CF3*, CF235CI+, and
in Appendix)
observed
at
an isotopic pair of
have been
each
of
ions were
simultaneously monitored in Figures IOB and 10C, so that the
49
relative
intensities
determined.
of
these
two
responses
could
be
The area of the minor isotopic peak relative
to that of the major ion is shown to be Ar = 0.322,
agreement with the known isotopic ratio of 0.323.
in good
The very-
large responses observed in Figures IOB and IOC at about- 6.6
minutes
are
due
to
CF2Cl2, a major
atmospheric
halocarbon
with a concentration of about 540 pptv (42).
In Figure
(CFC-115)
(Fig.
is
11,
the expected retention time of CF3CF2Cl
indicated
and
32 in Appendix)
responses
p a irs,
noted
and
ions
in
its
mass
spectrum
have been monitored with detectable
at each of
H E - IlC
six
these.
11E-11F,
For the
that
were
two
isotopic
simultaneously
monitored,
it is noted that the response ratios observed for
these
in
are
heights
Also,
it
time
of
good
noted
is
presently
in. the
noted
about
composition,
agreement
EI
that
7.8
thought
mass
to be
The
the
spectra
another
minutes
C2F4Cl+.
with
corresponding
for
this
compound with
.also -produces
origin
a
compound.retention
ion
of
of • this ' signal
is
CF2ClCE2Cl' and not
an
peak
CF3CFCl2 due
to
the absence of a significant CF3+ response for this compound
50
in
chromatogram
11A.
At
minutes, a response
a retention
time
of
about
2.7
to C2F6 is also noted in chromatograms
IlA and IlD while monitoring the ions CF3+ and C2F5"1
".
In
Figure
(H-13 01)
is
12,
the
indicated
expected
and
retention
responses
at
time
all
of -CF3Br
of
the
seven
exact masses selected for this compound (see spectrum in the
Appendix,
Fig.
extensively
ratios
in
of
are
fire
observed.
CF3Br
extinguishers
responses
simultaneously
12B-12C,
32)
obseryed
measured
(22) .
for
bromine
the
isotopic
has
been
Note
that
three
ions
used
pairs
in
the
of
Figures
12D-12E, and 12F-12G are in good agreement with the
expected ■ ratio,
monitoring
the
0.98.
Br+
discussed below.
The
ion ' in
additional
Figures
12B
peaks
and
12C
noted
by
will
be
The very large responses noted in Figures
12D and 12E at a retention time of about 3.3 minutes are due
to Xe nuclides of m/z = 128.9048 and 130.9051,
(the
total
nuclides
concentration
of
Xe
in
the
of
the
atmosphere
nine
is
respectively
naturally
about
0.1
occurring
parts
per
million and these two nuclides have relative abundances of
26% and 21%,
respectively
(47)).
It is noted in Figures
51
12F and 12G that monitoring either one of the molecular ions
of CF3Br provides an excellent response to this compound,
spite
ions
of
the
in
the
fact
mass
that
the
spectrum
relative
shown
intensities
of
in the Appendix
about 10% that of the base peak at mass
in
these
are
only
This is due to
69.
the lower-level of chemical noise encountered at the exact
masses of these molecular ions.
In
Figure
(HCF-134a)
13,
the
expected
retention
time
of
CF3CH2F
is indicated and responses to three of the exact
masses selected for this compound
(Fig.
33 in Appendix)
are
observed.
The large peak observed in Figure 13B at about
6.0 minutes
is due to the CHF2+ ion of the compound CHF2Cl
(see Appendix).
The origins of the responses observed at
about 7.9 minutes in Figure 13B and at about 2.2 minutes in
Figure
13C
replacement
are
for
refrigeration,
'unknown.
CF2Cl2
This
(CFC-12)
air conditioning,
particular
and
CHF2Cl
HFC
is
(HCFC-22)
a
in
and in aerosol applications
and is Currently the most widely used HFC
(28).
52
Compounds indicated bv- moni t-.o-ri ng
the Br+'ion on the GS-O Cam'I Iany
Column
Compounds identified in this study were detected using
i the GS-Q
Figure
capillary column and concentration method I.
14,
one
of
the
isotopes
of
the
Br+ ion
has
monitored using four different resolution settings.
shown
that
the
relative
intensities
of
only
three
In
been
It is
of
the
observed peaks remains constant as. resolution is increased,
thereby indicating that all other responses noted with low
resolution
are
brominated
compound
minutes
is
not
CF3Br
actually
with
and
a
has
due
to
retention
been
the
Br+ ion.
time
previously
of
about
identified
The
5.3
in
Figure 12.
In Figure
15,
the expected retention time of CH3Br is
indicated and strong signals are observed at this time when
each of the
four major ions
in the mass
(Fig.38 in Appendix)
is monitored.
extensively
in
studied
the 'last
spectrum of CH3Br
Methyl bromide has been
ten
years
and
there
are
still many questions regarding the overall mass balance.
At
53
there is a mass imbalance and the known sources do
not
add
sources
up
of
to
the
methyl
and auto emissions
The
expected
known
bromide
sinks
are
(24,48,49) . The
oceanic,
fires,
principal
fumigation,
(24).
retention
time
of
CF2BrCl
(H-1211)
is
indicated in Figure 16 and responses are observed when nine
of the ions in its mass spectrum
m/z > 50 are monitored.
(Fig.
35 in Appendix)
of
Note that the relative.areas of the
simultaneousIy-monitored isotopic cluster groups are in good
agreement
with
the
relative
intensities
mass spectrum of this compound.
Again,
indicated
in
the
the origin of the
large peaks in Figures 16C and 16D is CF2Cl2 and that of the
large peaks in Figures 16E and 16F are isotopes of Xe.
The
origin of the peak at about 3.3 minutes is unknown as is the
corresponding peak noted in chromatogram 16H.
Relative In te n sity
54
R = 2000
4500
T im e (m in u te s )
Figure 6. CF3"1" single ion chromatograms using three
different resolution settings.
In each chromatogram
the relative responses are normalized to the maximum
response observed under each condition.
The amount of
air sampled is 340 atm-cm3.
55
CHFoCI
Relative Intensity
6 8 .9 7 2 1
1300
R = 4500
B
T
i
2
3
I....'.'
4
'5
111I 1* 111i 11' '"rT ' 1-* '
6
7
8
I ' ' ' ' ' I '
9
10
Time (minutes)
Figure 7. CHF37Cl+ single ion chromatograms using two
different resolution settings. The expected GC
retention time of CHF2Cl is indicated in chromatogram
A.
56
R elative In te n sity
2000
2
3
4
5
6
7
8
9
10
T im e (m in u te s)
Figure 8. Single ion chromatograms using two ions
expected in the electron impact spectrum of CF4. The
expected retention time of CF4 in indicated in
chromatogram A.
Relative Intensity
57
R = 2000
m/z = 4 9 .9 9 6 8 (CF2 + )
B
x 50
Vv
R = 1800
m/z = 5 1 .0 0 4 6 (CHF2+)
Time (minutes)
Figure 9. Single ion chromatograms, and expected
retention time for CHF3.
58
R = 2500
m/z = 6 8 .9 9 5 2
Relative In te n sity
C F 5CI
R = 1200
m/z = 8 4 .9 6 5 7 .
x 100
R = 1200
m/z = 86 .9 6 2 7
x 100
0 .3 2 2
T im e ( m in u te s )
Figure 1 0 . Single ion chromatograms and expected
retention time for CF3C l . The isotopic pairs’ of ions
in chromatograms B and C- were simultaneously measured
in a single analysis. The value of A r in chromatogram
C is the ratio of this peak area to that in
chromatogram B .
59
R = 1200
m/z = 68.9952 (CF3*)
CF3CF2Cl
\
n
m/z = 84.9657
ICF235Cl1')
B
=A
m/z = 86.9627
ICF237CI1)
A, - 0.294 A x 20 1
I I
C
i
*
m/z = 118.9920 IC2F51)
I
\D
I
I m/z = 134.9629 IC2F435Cl1)
JE
A
•
I
/x.
-I m/z = 136.9595 IC2F437CI1)
3
4
5
5
i
z>
!I
]
]F
I
i :
IV
7
8
9
10
Time (minutes)
, Single ion chromatograms and expected
time for CF3CF2C l . Chromatograms B
and C and chromatograms E and F (isotopic
pairs) were simultaneously determined.
Ar
values are the ratios of the responses to the
minor and major isotopic ions.
60
R = 1200
m/z = 68.9952 (CF3*)
m/z = 78.9183 I79Br1-)
m/z = 80.9163 (aiBrt)
>•
'53
c
<u
A, = 0.965
<D
m/z = 128.9151
(CF279Brt)
>
(0
to
OC
m/z = 130.9131
(CF281Brt)
Ar= 0.989
m/z = 147.9136 (CF379Brt)
m/z = 149.9116 (CF381Brt)
Ar= 0.949
5
6
7
8
9
10
Time (minutes)
Figure 1 2 . Single ion chromatograms and expected
retention time for CF3Br.
The responses to the
isotopic pairs of ions in chromatograms B and C, D and
E, and F and G were simultaneously determined.
Ar
values are the ratios of the responses of the minor
81Br-Containing ion to that of the major 79Br-Containing
ion.
61
Relative Intensity
2500
R = 1400
m/z = 51.0046
R = 1800
m/z = 83.0108
Time (minutes)
Figure.13. Single ion chromatograms and expected
retention time for CF3CH2F .
Relative Intensity
62
R = 4500
n/i/I .............. I 1 1 1 1 1 I
2
3 '"*
I ' ' ' ' ' I ' ' ' ' ' I ' ' ' ' ' I ' ' ' '' I
4 ”
5'
6
7
8
9
10
Time (minutes)
Figure 14. 81Br4" single ion chromatograms using four
different resolution settings.
The amount of air
sampled is 340 atm-cm3.
63
R = 1200
-
CH3Br
d \
m/z = 7 8.9183 (79Br"1")
4 A-
Relative Intensity
------
JX__ __
'
A
m/z = 8 0 .9163 (81Br
Ar = 0.931
m/z = 9 3 .9 4 1 8 (CH379Br +)
i C
........ ' - i I ........... I
-
PYffTrTT
I
m/z = 9 5.9398 (CH381Br +)
Ar = 0 .9 4 3
4 D
2
3
4
5
6
7
8
9
10
Time (minutes)
Figure 15. Single ion chromatograms and expected
retention time for CH3B r . The responses to the
isotopic pairs of ions in chromatograms A and B and
chromatograms C and D were simultaneously determined
A r values are the response ratios of the :
81Br-Containing and the 79Br-Containing ions.
64
j
R= 1200
I
i A
m/z = 78.9183 I79Br+)
CF2BrCI
( / 2
3
-----------A ------
m/z = 80.9163 I81Br+)
Ar-= 0:962
n
^
m/z = 84.9657
J
(CF235CI+)
IC
m/z = 86.9627
(CF237CI+)
:,
n
( m/z = 128.9151
I ' (CFz78Br+)
>
‘55
0)
c
I
x 1000
V ___
j
m/z = 130.9131
(CFz81Br+)
I
Ar = 0.935
x 1000
j
J
m/z = 144.8856
Ar = 0.743
(CF278Br35CI+)
N
]
]
!
m/z = 146.8826 (CF79Br37CI+) il
or m/z = 146.8835 (CF31Br35CI+) j| Ar = 1-000
m/z = 148.8806
l(CF9
u r -1oBr
r3
-7
uCiI+r)
J I
2
2
-I
-=
5
Ar = 0.246
it
7
B
9
10
Time (minutes)
Figure 1 6 . Single ion chromatograms and expected
retention time for CF9Br C l . The responses to the
isotopic sets .of ions in chromatograms A and B , C and
D, E and F , and G, H, and I were simultaneously
determined:
65
Compounds indicated by monitoring
the Br+ ion using the DB-VRX
Capillary Column
Compounds
detected
identified
using
the
in
DB-VRX
this
part
capillary
ion . chromatograms
spectrometer
80.9163,
has
been
of the ion,
are
tuned
study, were
chromatography
In Figure 17, three
shown
to
the
column
procedures and concentration method 2.
single
of
the
where
exact
the
mass,
mass
m/z
81Br"1", using three different resolution
settings.
With the use of a low resolution setting of our
instrument
(R
distinct
=
peaks
compounds
resolution
at
1000)
are
in
observed,
approximately
only
eight
persist,
indicating
certainly
bromine
Figure
17,
with
at
that
the
original
these
containing.
fourteen
a ' coelution
four minutes.
of
least
With
The
best
two
increasing
fourteen
compounds
of
■are
peaks
almost
signal-to-noise
while maintaining sufficient resolution was determined to be
R = 2000, see Figure 18.
In Figure 18, the eight brominated
compounds
detected
background
monitoring
the
in
81Br"1
" fragment
ion,
■Montana
are
shown.
air, . while
. All
of
the
66
following
chromatograms,
Figure
'19
Figure
-
25,
were
obtained using a resolution setting of 2000.
In Figure 19,
the compound CF2ClBr has been identified
at a retention time of about 4 minutes.
Most of the major
fragment
ions
spectrum
compound
have
Commonly known
present
been
in
the
EI
detected
as Halon-1211,
(Fig.
this
extensively as a fire extinguishent
was
identified
by
matching
mass
35
in
retention
times
this
Appendix).
compound has
(30,50).
of
been used
Each compound
to
standards,
detecting many of the major fragment ions present in the EI
mass
spectrum,
clusters
of
abundance
and
ions
ratios
by
in
simultaneously
order
(Ar)'.
to verify
In
Figure
monitoring
the
19,
isotopic
correct
Ar =
isotopic
0.92,
which
agrees well with the known isotopic abundance of 0.97 when
taking into account that the A r value will be slightly lower
do
to
the
use
of
the
voltage
jumping
technique
described
earlier.
In Figure 20,
have
been
four of the most abundant
detected
spectrum in Fig.
time of 6:20.
36,
for
the
Appendix)
compound
fragment
CF2Br2
(see
ions
mass
at the expected retention
The A r value is slightly higher than expected
67
for the Br+ isotopes due to the low signal-to-noise for the
79Br+
isotope.
This
is
the
first
time
CF2Br2 has
been
detected in the atmosphere and its source is unknown, but is
very likely of anthropogenic origin.
In Figure 21, the compound CF2BrCF2Br
(H-2402)
has been
identified by monitoring four of the major ions in its mass'
spectrum
'(Fig.
37
in Appendix)
ata •retention
of
8:10.
H-2402 was produced
in Japan and used in the former Soviet
Union
the Br+ isotopic
(53) .
A r .for
pair
is low due to a
coelution of an unknown' compound in the 79B r f chromatogram.
In
Figure
22,
bromide
(CH3Br)
is
Fig.
38).
naturally
biomass
during
(48) . •
at a
identified
time
(mass
spectrum
(51) .
It
use
The
as
is
a
(49,18)
4:42
methyl
in Appendix,
also
released
fumigant
contribution
and during the burning of
for
to
uncertainties
in our knowledge
into
the
strawberries
atmospheric
anthropogenic release is not well know
sinks.
of
As mentioned earlier, methyl bromide is produced
in the oceans
its
retention
of
atmosphere
and
CH3Br
(24,48,49,52)
the natural
grains
from
due to
sources' and
68
In Figure 23,
a retention
major
time
fragment
Appendix).
oceans
the compound CH2Br2 has been detected at
of
about■13:50
ions
have
minutes
been
Dibromomethane
is
monitored
produced
Figure
24,
bromochloromethane
fragment
ions
at
a
retention
has .been
(mass
spectrum
in
(Fig.
of
.40
in
the
in
the
(24).
time
identified
seven
naturally
(22,29) and has industrial sources
In
and
of
by
11:00
minutes
monitoring
Appendix,
Fig.
nine
39).
CH2ClBr is also thought to be primarily from natural sources
(22,29) .
In Figure 25, bromoform has been detected at a time of
18:20
and
nine
of
the
major
spectrum have been monitored
has
also
(22,24,52).
been
shown
to
fragment
(Fig.
have
ion
in
its
41 in Appendix).
natural
oceanic
mass
CHBr3
sources
69
i
m/z = 80.9163 (81B f)
C
L
R = 1000
Relative Intensity
Vr
R = 1500
R = 3000
V^l
W'nv«- r^QL^L
Time (Minutes)
Figure 1 7 . 81Br+ single ion chromatograms using three
different resolution settings.
In each chromatogram
the relative responses are normalized to the maximum
response observed under each condition.
The amount of
air
sampled is 340 atm-cm3.
70
CH3Br
Relative Intensity
R = 2000
CH2Br2
Time (Minutes)
Figure 18. Single ion chromatogram monitoring 81Brt at
resolution of 2000. Eight brominated compounds have
been identified in rural Montana background air; CF3Br,
CF2ClBr, CH3Br, CF2Br2, CF2BrCF2Br, CH2ClBr, CH2Br2, CHBr3.
71
Relative Intensity
l“ illlilii|||u ||
I
m/z= 130.9131 (CF281Br*)
A,=0.93
m/z= 144.8856(CF35CI79Br*)
Ar=OSO
m/z= 146.8835(CF35CI81Bf and CF37CI79Bf)
Ar= 1.0
K_i
j
I
.
.
.
. ..
.
I
--
m/z= 148.8086(CF37CI81Bf)
Ar=0.23
_a1~ j
8
10
12
14
16
18
Time (Minutes)
Figure 19. Si n g l e ion c h r o m a t o g r a m s i d e n t i f y i n g the
c o m p o u n d C F 2ClBr,
R = 2000,
R T = 4:00.
72
m/z= 78.9183 (79Br*)
Relative Intensity
m/z=80.9163 (81Br*) Ar= 0.99
m/z= 128.9151 (CF279Br*)
I
Figure 20.
R = 2000,
A,= 0.93
m/z= 130.9131 (CF281Br*)
—
8
10
12
14
16
18
Time (Minutes)
Si n g l e ion c h r o m a t o g r a m s
R T = 6:20.
identifying
CF2B r 2,
73
Relative Intensity
m/z = 78.9183 (79Br*)
CF2BrCF2Br
m/z= 128.9151 (CF279Br*)
m/z= 130.9131 (CF281Br*)
Ar= 0.94
m/z= 178.9120(C2F479Br*)
m/z= 180.9100(C2F481Bf)
A,=0.94
Time (Minutes)
Figure 21. Single ion c h r o m a t o g r a m s i d e n t i f y i n g
C F 2B r C F 2Br,
R = 2000,
RT = 8:10.
74
Relative Intensity
m/z= 80.9163(81Bf)
Ar= 0.92
m/z= 93.9418(CH379Bf)
m/z= 95.9398(CH381Bf) Ar= 0.97
2
4
6
8
10
12
14
16
18
Time (Minutes)
Figure 22. Si n g l e ion c h r o m a t o g r a m s i d e n t i f y i n g C H 3Br,
R = 2000,
RT = 4:42.
75
-
:
CH 2Br2
m/z = 78.9183 (79Br')
-
Relative Intensity
i
4
6
m/z=94.9319(CH281Br')
Ar= 0.87
m/z= 171.8522(CH279Br2*)
Ar= 0.54
m/z= 173.8502(CH279Br81Br')
Ar= 1.0
m/z= 175.8482(CH281Br2')
Ar= 0.46
8
10
12
14
16
18
Time (Minutes)
Figure 23. Single ion c h r o m a t o g r a m i d e n t i f y i n g CH2B r 2,
R = 2000,
RT = 13:50.
76
I
m /z= 78.9183 (79Br*)
I
I
X
CH2CIBr
J o _ 01/2 = 80.9163 (81Br*)
I
A, = 0.95
i
I
m/z = 48.9846 (CH235CI*)
X
-
.
■
------ ^
.J I
Relative Intensity
I
j
m/z = 127.9028 (CH235CI79Br*)
A = 0.87
m/z = 129.9003
(CH235CI81Br* and
CH237CI79Br*)
131.8978 (CH237CI81Br*)
A = 0.33
E
I A
8
10
12
14
16
18
Time (Minutes)
Figure 24. Sin g l e ion c h r o m a t o g r a m s i d e n t i f y i n g
C H 2ClBr,
R = 2000,
R T = 11:00.
77
I 11
5
j I|
I
m/z= 78.9183(79Br*)
A
E
C H B r^
___ A.
'
r
m/z=80.9163(81Br*)
I
Ar=0.97
I
_
\
m/z=90.9184(C79Br*)
Ar= 1.0
Relative Intensity
I
-f
I,
ii
i
m/z= 91.9262(CH79Br*)
Ar=0.57
m/z=92.9164(C81Br*)
Ar=OQS
I
J
I...I'1m/z= 93.9242(CH81Br*)
m/z= 170.8444(CH79Br2*)
I
1 I ........
I
I
Ar=O.55 I
I rV F
Ar=0.52
i nr
m/z= 172.8424(CH79Br81Br*)
Ar= 1.0
m/z= 174.8404(CH81Br2*)
Ar=0.54
8
12
14
16
18
Time (Minutes)
Figure 25. Single ion c h r o m a t o g r a m s identifying CHBr3,
R = 2000,
RT = 18:20.
78
Quantitative determinations by GC-HRMS
Two
quantitative
methods
were
used
to
determine
atmospheric concentration of selected halocarbons.
of
the
difficulty
(discussed
reported
in
in
the
making
standard
measurements
are
low
concentration
preparation
relative
to
standards made by others.
Quantitation
study,
very
is
Because
standards
section),
all
measurements
or
v
a
difficulty
isolated
to
this
but is an issue for all those making measurements at
low
levels.
concentrations,
related
species . in
global
the
Most
a
review
article
discussing
and trends of C F G 's ,.Halons, and
atmosphere
(22),
it
was
stated
continue to limit our understanding of
atmospheric
species.
important
In
lifetimes,
"Calibration issues
the
not
the
distributions
activities
and
undertaken
trends
to
of
these
resolve
this
issue are unsatisfactory at this stage".
It was
also stated that uncertainties in atmospheric concentrations
of
species
like
CFC-12
(CF2Cl2),
reported
tropospheric concentration of 480 pptv,
to
have
an
are estimated to be
79
less
the
than
5%.
But,
uncertainty
for compounds
levels
increased
at
lower concentrations
dramatically.
CFC-113
(CCI2FCCI f 2) has been measured at levels of 71 pptv with an
uncertainty of 2 0%, ,and H-13 01
an
atmospheric
concentration
of
estimated uncertainty of 80%.
uncertainties
given
above
reported in the NASA R e f .
(CF3Br)
1.2
is reported to have
to
2.2
pptv
and
an
The concentration values and
are
from
1990
Pub. 1339 , 1994
annual
means
(22) .
Quantitative Analysis of CF/ and
Br+ Compounds with GS-O Capillary ■
Column
For the compounds analyzed in the CF3+ study,
capillary
column
concentration
was
method
used
I
for
(described
the GS-Q
chromatography
earlier)
was
and
used
to
concentrate the air samples.
The reproducibility of a given
analysis has been very high.
For example, the analysis of
CF3Br
shown
in
10-hour period.
Figure
The
12G
was
repeated
limits
can
times
over
a
standard deviation of the peak areas
thereby recorded was less than 1%.
detection
8
be
very
low.
In addition,
For
signal-to-noise ratios of the two responses
estimated
example,
the
to CF3Br shown
80
in Figures 12F and 12G are greater than 300.
This leads to
an estimated detection limit for this compound of about 30
parts-per-quadrillion.
The sample,
5,
wah
whose analysis results are given in Table
collected
in
a
remote
mountainous
region
about
8
miles east of Bozeman, MT, and is representative of numerous
samples
collected
in
local,
rural
and
mountainous
areas
during the months of August and September of 1996.
The ■ measured
Table
tropospheric
listed
in
5 were determined using a low concentration standard
(approximately 6 pptv)
The
concentrations
accuracy
of
and a single point calibration curve.
our
low
discussed previously.
compounds
standard
Quality
listed
containing
Services,
concentrations.
calibration
which
mixing
in
ratios
us
The
Table
CF4
was
,provided
to
giving
good
agreement
only
low
This
was
mixture
measure
reported
by
the
produced
checked
us
in
mixing
ratios
other
groups.
were
contained
5 ■ and
,
to
standards
standard mixture
Inc.
standard
allowed
concentration
by
the
against
Indaco
for
a
Air
CF4
concentration
our
laboratory
similar
to
While
the
good
■agreement was observed between these measurements and those
81
reported
by
agreement
other
could
calibration
groups
be
with
this
fortuitous.
standards
prepared
standard,
Comparison
in our
this
with
laboratory
good
other
leads
to
uncertainties of about 50% in actual concentration.
Table 5. Atmospheric Concentration of Compounds
Analyzed with GS-Q Column.
Measured Cone.3
Ion Monitored
(pptv)
(m/z )
CF4
60
69
CF3Hi2
7.2
51
CF3Cl
2
85
CF3CF2Cl
5
85
CF3Br
3
148,150
CF3CH2F
4.1
83
CH3Br
11
■ 79
CF2ClBr
3.2
Compound
81,145
Uncertainties of 50% have been assigned.
Detected and measured for the first t i m e .
Comparison of
with
that
of
a
the
response
calibration
observed at m/z
standard
indicated
= 68.9952
that
the
concentration of CF4 (CFC-14) in the air sample was 60 pptv.
Previously
reported
determinations
background air samples are 67
(56) .
of
(5.5) , 60-65
this
compound
in
(54) , and 58 pptv
82
Comparison of
with
that
of
the
the
response
calibration
observed
standard
at m/z
= 51.0046
indicated
that
the
concentration of CF3H in the air sample was 7.0 pptv.
By comparison of the response observed at m/z = 84.9657
with that of the calibration standard,
of
2.0
pptv
in
the
air
sample
reported determinations of this
are 4 (16) and 2 pptv
a CF3Cl concentration
was ' deduced.
Previously
compound in background air
(20).
Comparison of the response to CF3CF2Cl at m/z = 84.9657
to
that
of
the
calibration
standard
indicated
concentration of 5.0 pptv in the air sample.
a
CF3CF2Cl
A previously
reported determination of this compound in background air is
5 pptv
(55).
Comparison of the response to CF3Br at the m/z value of
either
of
these
two
molecular
ions
to
that
of
the
calibration standard indicated a concentration of 3.0 pptv
in the air sample.
extinguishers.
CF3Br has been used extensively in fire
Previously reported determinations
compound in background air are 1.7
pptv
(42).
(57), 1.7
of this
(30), and 2.3
83
Comparison of the response to CF3CH2F at m/z = 83.0108
to
that
of
the
calibration
standard
indicated
concentration of 4.1 pptv in the air sample.
a
A previously-
reported determination of this compound in background air is
3 pptv
(42,28).
Comparison of the response to CH3Br at m/z = 78.9183 to
that
of
the
CF2BrCl
concentration
of
calibration
11 pptv
in the
reported determinations of this
are 12-15
(58), 10
Comparisons
80.9163
and
m/z
(60), 14
of
the
standard
air
sample.
a
Previously
compound in background air
(61), and 12 pptv
responses
= ■144.8856
indicated
to
to
those
(59).
CF2BrCl
at
obtained
m/z
from
=
the
calibration standard indicated a concentration of 3.2 pptv
in
the
local
determinations
(57), 2.1
background
of ■this
air
sample.
compound
(30), and 3.5 pptv
Previously
in background
air
reported
are
2.0
(42).
Quantitative Analysis of Br+
Compounds with DB-VRX Capillary .
Column
For the study of brominated compounds in the atmosphere
concentration method
2
(described
earlier)
was
used
along
84
with
chromatographic . methods
capillary column.
developed
associate
.the
our
measurements,
in
(59) we have elected
this
study
of
bromine-containing compounds, with those of Montzka
1996 by way of the . compound CF2ClBr.
as
an
lifetime
internal
of
distributed
DB-VRX
Because of the difficulties in .preparing
low to sub parts-per-trillion standards
to-
for
standard.
CF2ClBr
Given
(50,17),
throughout
it
the
As
a l .
,
CF2ClBr will be used
the
will
be
troposphere
midcontinental United States .
et
the
can be
long
atmospheric
fairly
of
uniformly
the
northern
seen .in Table
6,
measurements of this long lived halocarbon are consistent to
within ±0.1 pptv throughout the northern hemisphere.
Table 6.
Measurements of CF2ClBr by NOAA/CMDL
Location
Sampling Date Concentration3
(pptv)
Alert, N W T , Canada
Dec . 1995
3.4
Pt. Barrow, AK
Dec . 1995
3.44
Niwat Ridge, CO
Dec . 1995
3.5
Mauna L o a , HI
Dec . 1995
3.58
Cape Matatula, Samoa
Dec . 1995
3.39
Cape Grim, Australia
Nov . 1995
2.98
South Pole
Nov . 1995
3.16
■
NOAA/CMDL
data
base
3
'Data
retrieved
from
fi l e ://ftp.cmdl.noaa.gov/noaa/halons/hl211_96.txt
85
Therefore,
Boulder
CO
measurements
by
(42)
can
of
be
CF2ClBr
used,
(H-1211) made
in
conjunction
in
with
relative responses of H-1211 to other bromihated compounds,
to estimate the atmospheric concentration of the brominated
compounds
CF2ClBr
CO)
to
in this
has
been
study
(Tables
tracked
at Niwat
from 1989 to early 1996.
include
apparent
H-1211
the
sampling
leveling
off
concentration
7 and
period
(slowed
data
8).
Ridge
In Figure' 26,
(outside Boulder,
Ihave extrapolated the data
of .our
rate
collected
study.
of
at
With
the
increase)
of
the
Niwat
Ridge,
a
relatively constant tropospheric concentration of 3.5 pptv,
for
CF2ClBr was
assumed.
Based on this value
of CF2ClBr,
and knowing the response efficiencies of the other compounds
relative to CF2ClBr
an
internal
(Table 7) allowed CF2ClBr to be used as
standard.
This
internal
standard
method
was
then used to determine the atmospheric concentration of the
other six brominated compounds measured in this study,
see
Table
our
8.
Relative
responses
were
determined
in
laboratory by making an approximately 200 ppmv concentration
mixture
containing
CH3B r 7 CF2Br2, C2F4Br2, CH2ClBr,
and CHBr3, and CF2ClBr.
CH2Br2,
86
Tropospheric Concentration of H-1211
Concentration (pptv)
Data from NOAA/CMDL Boulder, CO
3.5 h
♦ (TsCIBr
Figure 26. Tropospheric concentration of CF2ClBr
measured at Niwat Ridge, Boulder, CO by NOAA/CMDL from
1989 through 1995.
87
The
mixture
was
then .analyzed
to
determine
the
relative
response
of each compound to CF2ClBr while monitoring both
isotopes
of
responses
the
Br+ fragment
were
also
ion,
determined
see
for
Table
CF2ClBr,
CF2BrCF2Br with respect to the CF2Br+ ions
Atmospheric
from the
average
December
1996
individual
equation
concentrations
of
and
compounds
(Eq.
April
were
determined
collected between
The
determined
CF2Br2, . and
8 were
samples
1997.
Relative
(Table 7) .
in Table
a number of
7.
concentration
using
the
of
following
20).
Cbc = ((Cis * Abc)/ (RR * Ais))
(20)
Cbc is the concentration of the brominated compound to
be
determined.
standard
RR
is
the
Cis is
(CF2ClBr)
the
of
the
and has been determined to-be
relative
ratio
compound being analyzed.
areas,
concentration
from Table
7 for
the
CF2Br+, after
Areas
the
analysis system.
are
3.5 pptv.
brominated
A bc and A is are the chromatographic
for the brominated compound of interest
respectively.
internal
determined
introduction
of ■an
by
and CF2ClBr
monitoring
air
sample
Br+ or
into
the
88
Table 7 . Relative Responses of select Brominated
Compounds to CF2ClBr.
Compound
Br- ion
CF2ClBr
CF^Bri ion
I .O
1 .0
CH3Br
0.60
CF2Br2
2.1
6.4
CF2BrCF2Br
1 .0
3.0
CH2ClBr
1.4
CH2Br2
2.9
CHBr3
•5.1,
Table 8. Atmospheric Concentration of Brominated
Compounds analyzed with DB-VRX Column.
Compound
Measured*
5
*
Cone .
(pptv)
Detection6
• Limit
(ppqv)
Ion used
for DIi
(m/z)
CF2ClBr.
'3.5£
19
85
CH3Br
10.8
29
96
CF2Br2
0.04
11
131
CF2BrCF2Br
0.32
15
179
CH2ClBr
0.07
11
130
CH2Br2
0.43
I
174
CHBr3
0.20
I
173
a
A relative uncertainty of plus-or-minus
has' been assigned to these measurements.
5
16%
For 340 cc sample
£
Adjusted so that atmospheric concentration
of CF2BrCl is in .agreement with Montzka et a l .,
1996, 3.5 pptv.
89
The
estimated
tropospheric
concentration
of
CH3Br
by
comparison of the response of CH3Br at m/z = 80.9163 to that
of CF2BrCl indicated an average concentration of 11 pptv in
the
air
samples.
Previously
reported
this compound in background air are 12-15
10
[(60), 1992]
The
the
this
air
[(61), 1992]
estimated
comparison
CF2BrCl
14
of
the
indicated
an
samples.
compound
I n l a n d ] , 3-60
average
background
[(29), Arctic],
[(59), 1995].
concentration
m/z
=
80.9163
reported
air
2.5
are
[(31),
of
CH2Br2 by
to
that
of
0.4
pptv
in
determinations
of
concentration
Previously
in
[(58), 1985, N.
at
of
[(58),1985, N.Hem]
and 12 pptv
tropospheric
response
determinations
of
0.5-1
[(22),
1994,
1984, Arctic],
2.7
Hem. ], 2 [(24), 1996, oceanic].
The estimated tropospheric concentration of CH2ClBr by
comparison
CF2BrCl
of
the
response
at
m/z
=
80.9163
indicated an average concentration of
the air samples.
to
that
of
0.07 pptv in
Previously reported determinations of this
compound in background air are 2 [(31), 1984, Arctic]
The
estimated
comparison
of
the
tropospheric
response
at
concentration
m/z
=
80.9163
of
to
CHBr3 by
that
of
90
CF2BrCl
the
this
air
indicated
2-46
Hem. ] 1-2
The
in
concentration
Previously
background
[(29),
1982,
reported
air
are
Arctic],
of
pptv
in
determinations
of
0.2-0.5
0.85
0.2
[(22),
[(5 8 ),
1994,
1985,
N.
[(24), 1996, Oceanic].
estimated
comparison
CF2BrCl
average
samples.
compound
Inland],
an
of
tropospheric
the
response
at
concentration
m/z
=
128.9151
indicated an average concentration of
the air samples.
of
CF2Br2 by
to
that
of
0.04 pptv in
To our knowledge this is the
first
time
this compound has been detected in the atmosphere.
The
estimated tropospheric
concentration of
CF2BrCF2Br
by comparison of the response at m/z = 80.9163 and at m/z =
128.9151
to
that
• of
CF2BrCl
indicated
concentration of 0.3 pptv in the air samples.
reported determinations of this
are estimated at 0.5-1
Detection Limits
an
average
Previously
compound in background air
[(53), 1992].
A
detection
limit
for
each
compound was estimated based on a 3:1 signal-to-noise ratio
for
the
ion
of
each
compound
signal-to-noise ratio, see Table 8.
that
gave
the
highest
Detection limits as low
91
as I part-per-quadrillion by volume were achieved
sample sizes of 340 atm-cm3.
800
atm-cm3 samples,
while
a
(ppqv)
for
This compares to 50 ppqv for
similar
with analysis by GC/MS was used.
concentration
method
(28).
Analysis of Select Bromocarbons from December 1996 to
April 1997
April
Samples collected from December 1996 through
1997 •and were
CF2ClBr,
CF2Br2,'
Tracking
period
analyzed
CF2BrCF2Br,
these ■compounds
revealed
compounds,
see
fairly
Figure
27.
over
for
the brominated
CH3Br,
CH2Br2,
this
relatively
stable
Figure
and
CHBr3.
short
time
concentrations
of
all
27
the
two
(Top)
shows
compounds with concentrations greater than I pptv,
CF2ClBr.
compounds
CH3Br and
CF2ClBr was used as the internal standard and its
concentration
is
assumed to have
pptv over the sampling period.
remained
constant
Figure 27 (Bottom)
at 3.5
shows the
compounds with tropospheric concentrations less than I pptv,
CF2BrCF2Br, CF2Br2, CH2Br2, and CHBr3.
92
Conclusion
The high potential of GC-HRMS for the trace analysis of
halocarbons
in
the
atmosphere
has
been
demonstrated
h e r e .'
The increased specificity provided by exact mass monitoring
with high resolution greatly increases the confidence with
which positive identifications of sought-for-substances can
be made.
If the HRMS has a large ion-optic system,
as in
the present study, the high level of sensitivity inherent in
GC ■detection' by MS
GC-HRMS
CF3Cl,
reported
CHF3,
is
here
also
verify
CF3CF2Cl,
samples of rural Montana.
air.
and
identified
the
CF3Br,
CF2Br2, CH2Br2, CHBr3, CH2ClBr,
detected
maintained.
The
presence
CF3CH2F,
of
results
by
CF4, CHF2C l ,
CH3Br,
CF2BrCF2Br,
and CF2BrCl in background air
The compounds CF3H and CF2Br2 were
for
the
first
time
in
background
93
Tropospheric Concentration
Bozeman, Montana
f
i
* CH3Br
+ CF2CIBr
I
12/11/96
02/02/97
03/10/97
03/20/97
Sample Collection Date
03/26/97
Tropospheric Concentration
Bozeman, Montana
0.5
* CHBrS
+ CF2BrCF2Br
CF2Br2
e CH2Br2
+■
12/11/96 02/02/97 03/10/97 03/20/97 03/26/97 04/02/97
Sample Collection Date
Figure 2 7 . (Top) Tropospheric concentrations of CF2ClBr
and CH2Br for the sampling period, December 1996
through April 1997. (Bottom) Tropospheric
concentrations of CF2Br2, CF2BrCF2Br, CH2Br2, and
CHBr3 for the sampling period of December 1996
through April 1997 in the area of Bozeman, MT.
94
SUMMARY
The objective of this project was to develop a method '
for the
ultra
trace
detection
compounds in the atmosphere.
and
analysis
of
halogenated
The analysis of brominated and
perfluorinated species were of particular interest to us and
the
atmospheric
chemistry community,
in general.
This
is
because of the limited data available regarding atmospheric
mixing
ratios
of
these
low
level
(low pptv
to
sub
pptv)
atmospheric constituents.
Much
chemistry
of
has
the
recent • work
been
focused
on
in
the
the
field
design
of
and
atmospheric .
testing
of
.models which are used to predict the cumulative effects of
the many minor chemical
components. of the
atmosphere.
Of
particular concern are the compounds of anthropogenic origin
linked to global warming and stratospheric ozone depletion.
Also,
in order to design a model one first has to know what
chemical species are present and secondly,
measurements are
necessary to determine the accuracy of a proposed
model.
95
We have developed what we believe is the most sensitive
and
compound-specific
analysis
of
development
instrument
ever
halogenated compounds
of
this
instrument
in the
started
high resolution mass spectrometer
donated
to
inoperable
The mass
the
MSU
Mass
condition
by
to
by
Spectrometry
The
refurbishing
Facility
Northwest
mass
spectrometer
in
chromatograph
and
instrument.
new
data
operational,
system
Additionally,
an
were
including
allow
volumes
up
to
1000
a
air-sampling
atm-cm3
Once
new
interfaced
developed and assembled in our laboratory.
an
Laboratories.
thoroughly overhauled,
was
a
that had been
the replacement of many of the electronic components.
the
the
atmosphere.
(VG-ZAB 2F)
Battelle
spectrometer was
dedicated
to
system
gas
the
was
This system will
of
air
to
be
preconcentrated and injected onto a capillary GC column.
Additionally,
a system was designed and constructed for
remote air sampling.
A 12-volt diaphragm pump allowed us to
collect samples in electropolished stainless steel canisters
from
remote
locations. ' The
approximately 35 psi
samples
were
pressurized
to
and then transported, to the MSU Mass
Spectrometry Facility for analysis by GC-HRMS.
96
Two primary groups of compounds were analyzed in this
.work.
First,
the
compounds
containing a CF/
fragment
ion
with a boiling point less than -26°C, and chromatographed on
a
J&W
GS-Q' capillary
compounds
eluting
detected
while
column.
from
a
Second,
J&W
monitoring
DB-VRX
the
Br+
bromine
capillary
fragment
containing
column
and
ion.
The
compounds were' positively identified by monitoring most
the
major
peaks
expected
in
the
electron
ionization
of
mass
spectrum of each compound and by matching retention time to
that of a standard.
The
CF/
compounds
detected
and
then
identified
from
single ion monitoring experiments are CF4, CF3H,
CF3Br, ■CF3CF2C l , and
CF3CFH2.
Those
compounds
the
CF3C l ,
detected
and
then .identified from the Br+ monitoring ■experiments
are
follows;
CH2Br2,
CF2ClBr,
and CHBr3.
CH3Br,
CF2Br2, CF2BrCF2Br,
CH2ClBr,
as
To our knowledge this is the first time CF3H or
CF2Br2 have been detected and identified in background a i r .
A likely source of CF3H is associated with the production of
aluminum.
A gas
sample was provided to us
hoods of an aluminum manufacturer,
CF4 and C2F6 were detected,
from the
fume
where elevated levels of
as expected,
along with elevated
97
levels
of
present
CF3H .
The
source
time,
but 'it
or
contaminant
byproduct
can
of
be
CF2Br2 is
unknown
speculated
that
associated with one
of
at
it
the
is
the
a
other
halons,.such as CF3Br or CF2ClBr.
The
initial phase of this project
was
to develop .the
instrumental capabilities to detect and identify atmospheric
halocarbons at very low concentrations
(< 5 pptvj .
We have
demonstrated the high sensitivity of our instrumentation by
achieving
estimated
detection
magnitude lower than similar,
the
stability
analyzing
a
of
sample
the
limits
almost
orders
current GC/MS methods.
instrumentation
8 times
two
over
the
was
Also,
demonstrated
course
of
a
of
10
by
hour
period with less than a 1% deviation in peak area detected.
The
quantitative
aspects
extremely challenging.
of
accuracy.
difficulty,
standards
data
This
was
project
have
proven
We have been unsuccessful at making
low level calibration standards
of
this
(low pptv) with a high level
expected,
due
to
the
well-known
if not the impossibility, of making accurate gas
below
the ppmv range
reported 'are
linked
groups' (see Tables 5 and 8).
to
(40,41).
measurements
All
quantitative
made
by
other
Therefore, no claims as to the
98
absolute atmospheric concentrations of the chemical species
reported here can be m a d e .
The future of this project lies in developing superior
methods
of
preparing
the
quantitation.
low pptv gas
uncertainties
'vessels.
The
difficulties
standards
related
to
wall
encountered
in
are predominantly due
to
effects
in
the
One way to reduce the significance of wall effects
is to use larger storage vessels.
For example,
of a sphere increases in proportion to radius
as
the
surface
squared.
uncertainties
area
increases
Therefore, a
proportionally
less
surface
related
to
and
to
where
radius
will
should
minimize
have
the
surface
method of standard additions
standard additions
container
area
the volume
cubed,
proportional
larger
container is significantly large
of
storage
has
effects.
Also, if the
)
(>25 m3) , one could use the
for quantitation.
the
added benefit
of
The method
accounting
for sample matrix effects, because the calibration standard
is added directly to the sample and will undergo all sample
processing procedures.
report
absolute
This may be the Only way one could
atmospheric
high level of accuracy. .
halocarbon
concentrations
with
99
REFERENCES
100
1.
■ Carson, R.
, Silent Spring, The Riverside Press, 1962.
2.
Grimsrud, E .P .
and Rasmussen, R. A.
> Analysis of
chlorofluorocarbons in the troposphere by gas chromatography
- mass spectrometry, Atmos.
Environ. , 9, 1014-1017, 1975a.
3.
Grimsrud, E .P .
and .Rasmussen, R.A.
, Survey
analysis
of
halocarbons
in
the
atmosphere
by
chromatography t mass spectrometryr A t m o s .Environ.
1010-1013, 1975b. ' '
and
gas
9,
4.
Crut zen, P.J. , My Life with 0,. MO.. and Other YZO2.
Compounds (Nobel Lecture). A n g e w . Chem.
I n t . Engl. , 35,
1758-1777, 1996.
5.
.McElroy, M .B .
composition
of
the
763-770, 1989.
and
Salawitch, • R.J.
,
Changing
global- stratosphere.
Science,
243,
6.
Solomon, S . , Mills, M. , Heidt, L .E . , Pollock, W.H.
and Tuck, A. F . , On the evaluation of ozone depletion
potential. J. Geophys. Res.
97, 825-842, 1992.
7.
Fisher, D.A. , Hales, C.H. , Filkin, D .L . , K o , M.K.W.
, Dak. S z e , N . , Connell, P.S. , Wuebbles, D ,J . , Isaksen,
I.S.A.
and Stordal, F . , Model calculations of the relative
effects ' of CFC's- and their replacements on stratospheric
ozone. Nature 344, 508-512, 1990a.
8.
Rowland, F .S . , Stratospheric Ozone In the 21st Century
The Chlorofluorocarbon Problem. Environ.
Sci.
Technol.
25, 622-628, 1991.
9.
Rodriguez,
261, 1128-1158,
J.M. , Probing Stratospheric O z o n e . Science
1993.
10. Anderson, J.G. , Toohey, D.W. , and Brune, W.H. , Free
Radicals Within the Antarctic Vortex: The Role of CFC's in
Antarctic Ozone L o s s . Science 25 1 , 39-46, 1991.
11.
Cicerone, R.J. f Fires. Atmospheric Chemistry,
Ozone Layer. Science 263, 1243-1244, 1994.■
and the
101
12.
Hileman, B . , Storm
April 14, 28-31, 1997..
Warnings
Rattle
13.
Schneider, S .H . , The Greenhouse
Policy. Science 243, 771-779, 1989.
Inanrer.q. C&EN,
Effect: Science
and
14.
Fisher, D.A. , Hales, C.H. , Wang, W.C. , K o , K.W. ,
and S z e , N.D. , Model calculations of the relative effects
of C F C s and their replacements on global warning. Nature
344, 513-516, 1990.
15.
Ravishankara, A. R . , Solomon, S . , Turnipseed, A. A.
and Warren,
R.F.
, Atmospheric lifetimes of long-lived
halogenated species. Science, 25 9 , 194-199, 19.93 .
16.
Fabian, P . , Borcher's, P.R. , Penkett, S.A. , and
Prosser, N.J.D. , Halocarbons in the stratosphere.' Nature
294, 733-735, 1981..
17.
Prather, M.J.
and Watson, R .T . , Stratospheric ozone
depletion and future levels of atmospheric chlorine and
bromine, Nature, 344, 729-734, 1990.
18.
Singh, H .B . , Salas, L.J. , and Stiles, R.E. , Methyl
Halides in and Over the Eastern Pacific (4QaN-32aS) . Journal
of Geophysical Research 88, 3684-3749, 1983.
19.
Thompson, T.M. , Elkins, J.W. , Butler, J.H. , Montzka,
S.A. , Myers, R.C. , Baring, T.J. , Cummings, S .0. , Dutton,
G.S.
, Gilligan, J.M.
, Hayden, A.H.
, Lobert, J.M.
,
Swanson, T.H. , Hurst, D .F . and Volk, C.M. ,- Nitrous oxide
and
halocarbon
division,
in
Climate
Monitoring
and
Diagnostics Laboratory,
Report No.
22,
Department of
Commerce, 72-91, 1993.
I
20.
Borchers, R . , Fabian, P . , Singh, O.N. , L a l , S . and
Subbaraya, B .H . ,
Ozone in the Atmosphere (Bojkov, R .D .
and Fabian,P.
(eds. ), p p . •290-293, A. Deepak Publishing,
V A , 1988.
102
21.
Fisher, D.A. , Hales, C.H. , Wang, W. , K o , M.K.W.
and
Dak S z e , N. , Model calculations of the relative effects of
C F C s and their replacements on global warming. Nature 344,
513-516, 1990b.
22.
Kaye
J.A.
,
Penkett
S.A.
,
Report
on
the
Concentrations. Lifetimes, and Trends of C F C 1s. Halons. and
Related Species. NASA R e f . Pub.
1339, 1994.4.
Molina,
M.J. , Polar Ozone Depletion (Nobel Lecture). A n g e w . C h e m .
I n t . Engl. , 35, 1778-1785, 1996.
1
23.
Jugern et a l . , BLAST94: Bromine Latitudinal Air/Sea
Transect 1994. NOAA Technical Memorandum ERL CMDL-IO, 1996.
24.
Butler, J.H. , and Rodriguez, J.M. ,' Methyl Bromide in
the Atmosphere. The Methyl Bromide Issue, chapter 2, 1996.
25.
Lovelock,
J.E.
, Atmospheric fluorine compounds as
indicators of air movements. Nature, Lon d . 230, 379, -1971.
26.
Lovelock,
J.E.
, Maggs, R.J.
and W a d e , R .J . ,
Halogenated hydrocarbons in and over the Atlantic, Nature,
L o n d . 241, 194-196, 1973.
27.
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.
28.
Montzka, S.A.
, Myers, R.C.
, Butler,
J.H.
, and
Elkins,
J .W . , Observations of HFC-134a in the remote
troposphere. Geophysical Research Letters 23, 169-172, 1996.
29.
Berg, W.W. , Heidt, L .E . , Pollock, W. , Sperry, P.D. ,
and Cicerone, R.J.
, Brominated Organic Species in the
Arctic Atmosphere. Geophysical Research Letters 11, 429-432,
1984.
30.
Butler, J.H. , Elkins, J.W. , Hall, B .D . , Cummings,
S .0.
and Montzka, S.A. , A decrease in the growth rates of
atmospheric halon concentrations. Nature,
359,
403-405,
1992 .
103
31.
Rasmussen, R.A. , and Khalil, M.A.K. , Gaseous Bromine
in_the Arctic and Arctic h a z e . -Geophysical Research Letters
11, 433-436, 1984.
32.
Solomon, S . , Garcia, R .R . , Ravishandara, A.R. , On
the
role
of
iodine
in
ozone
depletion. Journal
of
Geophysical Research 99, '20,491-20,499, 1994.
33.
Atkinson,
R . ■,
Qrganig_Compounds: A
1-41, 1990.
Gas-Phase Tropospheric Chenriatry of
Review, Atmospheric Environment 24A,
34.
Salawitch et a l . . Chemical Loss of ozone in the Arctic
Polar Vortex in the Winter of 1991-1992 . Scnancor 261, 1993 .
35.
Toon et a l . , Heterogeneous Reactions
Solubility's.
and the Physical State of
Aerosols. Science, 261, 1993.
Probabilities.
Cold Volcanic
36.
Singh
et
a l . , Measurements
of
atmospheric
BrO1.
radicals
in the
tropical
and mid-latitude
atmosphere.
Nature, 334 , 1988 .
37.
Hills et al. , Kinetics of the BrO+ClO reaction and
implications for stratospheric ozone, Nature, 328, 1987.
38.
Webster
et
al.
,
Chlorine
Chemistry
on
Polar
Stratospheric Cloud Particles in the Arctic W i n t e r , Science,
261, 1993.
39.
Weston, Jr, R.
E . , Possible greenhouse effects of
tetrafluoromethane and carbon dioxide emitted from aluminum
production. Atmos.
Environ.', 30, 2901-2910, 1996.
40. - Denyszyn,
R .B . , Sassaman, .T . , Part Per Billion
Gaseous Mixtures - A New Challenge, Pittsburgh Conference,
1986 .
41.
Nelson, G . O . . Gas Mixtures - Preparation and Control,
Lewis Publishers, 1992.
104
42.
Montzka, S.A. , Butler, J.H. , Myers, R.C. , Thompson,
T .M . , Swanson, ' T.H. , Clarke, A .D . , Lock, L.T.
and
Elkins, J.W. , Decline in the, tropospheric abundance of
halogen from halocarbons: implications. for stratospheric
ozone depletion, Science, 272, 1318-1322, 1996.
43.
Montzka,
S.A.
, Myers, R.C.
, Butler,
J.H.
, and
Elkins,
J.W.
,
Global
Tropospheric
Distribution
and
Calibration Scale- of HCFC-22. Geophysical Research Letter
20, 703-706, 1993.
44.
Cicerone, R.J. , Atmospheric carbon
nearly inert g a s . Science 20 6 , 59, 1979.
45.
Rasmussen, R.A. , Penkett, S.A.
Measurement of carbon tetrafluoride
Nature 277, 549, 1979.
tetrafluorlde: a
and Prosser, N . ,
in the atmosphere.
46.
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.
R e s . 86, 5172-5168, 1981.
47.
00
0
U
Windhlolz, M. , The Merck Index, IOth Edition,
1447, 1983 .
, Inc. , Rahway, N J . p.
Butler, J.H.
376, 469, 1995.■
, Methvl
bromide
under
Merck &
scrutinv , Nature
Lobert, J.M. , Butler, J.H. , Montzka, S.A. , Geller,
, Myers, R.C. , and Elkins, J.W: . A Net Sink for
Atmospheric CH1Br in the East Pacific Oce a n . Science 267,
1001-1005, 1995.
49.
CO
hi
50.
L a l , S . , Borchers, R . , Fabian, P . , and Kruger, B .C .
, Increasing abundance of CBrCIF2 in the atmosphere. Nature
316, 135-136, 1985.
51.
M a n o , S . , and Andreae, M.0. , Emission of Methyl
Bromide from Biomass Burning. Science 263, 1255-1257, 1994.
105
■52.
Cicerone, R.J. , Heidt, L .E . , and Pollock, W.H. ,
Measurements of Atmospheric Methyl Bromide and Bromofo r m .
Journal of Geophysical Research 93, 3745-3749, 1988.
53.
Butler e t a l . , A Decline in the. Growth
Atmospheric Halons. Submitted to Nature, 1992.
Rate
54.
Fabian, P . , Borchers, R. , Kruger, C.
and Lai,
CF4
and
C2F6
in the
atmosphere. J. Geophys .'Res.
9831-9835, 1987.
of
S. ,
92,
55.
Fabian, P . , in Handbook of Environmental Chemistry,
V o l . 4A, Hutsinger (ed. ), pp.
24, Springer, Berlin, 1986.
56.
Zander, R. , Rinsland, C.P. , Farmer, C.B.
and Norton,
R .H . , Infrared spectroscopic measurements of halogenated
source gases in the stratosphere w i t h ■the ATMOS instrument.
J.Geophys.Res. 92, 9836-9850, 1987.
57.
Channin, M . , Ehhalt, D . , Fraser, P . , Frederick, J. ,
McCormick, M . , Megie, G .
and Schoeberl, M . , Global
trends, in WMO Global Ozone Research and Monitoring Project
Report No.
20, pp.
162-281, 1990.
58.
Penkett, S . , Jones, B . , Rycroft, B . and Simmons, D .
, Interhemispheric comparison of the concentrations of
bromine compounds in the atmosphere. Nature 318, 550-553,
1985 .
59.
Rhoderick, .G .C . , Measurements of atmospheric
bromide using gravimetric gas standards. Environ.
Technol. , 29, 2797-2800, 1995.
methyl
Sci.
60.
Rowland, R .S . , in Methyl Bromide Science Workshop
Proceedings, Atmospheric and Environmental Research, Inc,
Cambridge, MA, 1992.
61.
Heidt, L.E., in Methyl Bromide Science Workshop
Proceedings, Atmospheric and Environmental Research, Inc,
Cambridge, MA, 1992.
106
62.
Elkins, J.W . , Thompson, T .M . , Swanson, T.H. , Butler,
J •H . , Hall, B .D . , Cummings, S.O. , Fisher, D.A. , and
Raffo, A.G. , Slowdown in the Growth Rates of Atmospheric
Chlorofluorocarbons 11 and 12:
Result of the Montreal
Protdcol, Elkins et a l . , 1992.
63.
Khalil, M.A.K. , Rasmussen, R.A. , and Gunawardena, R .
, 'Atmospheric
Methyl
Bromide:
Trends
and
Global
Mass
Balance. Journal of Geophysical Research 98, 2887-2896,
1993 .
64.
Shorter, .J.H. , Kolb, C.E. , Grill, P.M. , Kerwin, R.A.
, Talbot, R .W . , Hines, M .E . , and Harriss, R.C. , Rapid
degradation of atmospheric methyl bromide in soils. Nature
377, 717-719, 1995.
65.
Toohey e t a l . , The Seasonal Evolution of Reactive
Chlorine in the Northern Stratosphere, Science, 261, .1993.
66.
Lovelock, J.E. , Natural halocarbons in the air arid in
the s e a . Nature 256, 193-194, 1975.
67.
Lobert, J.M. , Butler, J.H. , Montzka, . S.A. , Geller,
L .S . , Myers, R.C., , and Elkins, J.W. , A Net Sink for
Atmospheric CHzBr in the East Pacific Oceanr Science 2 67,
1002-1005, 1995.
68.
SchauffIer, S.M.
, Heidt, L .E . , Pollock, W.H. ■,
Gilpin, T.M. , Vedder, J.F. , Solomon, S . ’, Lue b , R.A. , and
Atlas, E .L .', Measurements of Halogenated Organic Compounds
Near the Tropical Tropopause. Geophysical Research Letters
20, 2567-2570, 1993.
69.
Prather, M.J. , and Watson, R.T. , Stratospheric ozone
depletion and future levels of atmospheric chlorine and
bromine. Nature 344, 729-734, 1990.
70.
Harper, D.B. , Halomethane from halide ion-a highly
efficient fungal conversion of environmental significance.
Nature 315, 55-57, 1985.
107
71.
Goldstein, A.H. , Wofsy, S .C . , and Spivakovsky, C.M. ,
S_e.asonal— ygrigtions_of nonmethane hydrocarbons in rural. Mew
England =-- Constraints
on OH
concentrations
in northern
midlatitudes,
Journal
of
Geophysical
Research
100,
21,023-21,033, 1995.
72.
Lefevre, C. , Ferrari, P. , Guenier, J.P. , and Muller,
J.
, Sampling and Analysis of Airborne 'Methylbromide.
Chromatographia 27, 37-43, 1989.
73.
Sharp, G.J.
, Yokouchi, Y . , Akimoto, H . , Trace
Analysis of Organobromine Compounds in ' Air by Adsorbent
Trapping and Capillary Gas Chromatography/Mass Spectroscopy.
Environ.
S c i . Technol. 26, 815-816, 1992.
74.
Rowland,
F .S . ,
Stratospheric Ozone
Chlorofluorocarbons (Nobel Lecture) . A n g e w .
Engl. , 35, 1786-1798, 1996.
Depletion by
Chem.
Int.
75. Wilson e t a l .
, In Situ Observations of Aerosol and
Chlorine Monoxide After the
1991 Eruption
of
Mount
Pinatubo: Effect of Reactions on Sulfate Aerosols. Science,
2 6 1 , 1993.
76.
Newman e t
in the Arctic
1993 .
, Stratospheric Meteorological Conditions
Polar Vortex. 1991 to 1992. Science, 261,
a l .
77.
Proffitt e t a l . , Ozone Loss Inside the Northern Polar
Vortex During the" Winter of 1991-1992. Science, 261, 1993 .
78.
Browell e t a l . , Ozone and Aerosol Changes During the
1991-1992 Airborne Arctic Stratospheric Expedition. Science,
261, .1993 .
APPENDIX
109
100%
Figure 28. EI mas s s p e c t r u m of C F 2HCl
HO
100%
100
m'/z
Figure 29. EI mass s p e ctrum of C F 4.
Ill
100%
95i
Figure 30. EI mass s p e c t r u m of C F 3H
112
100%,
Figure 31. EI mass s p e c t r u m of C F 3Cl
113
100%
Figure 32. EI m ass s p e c t r u m of C F 3B r .
114
100%
95J
F
140
m/z
Figure 33. EI m a s s s p e c t r u m of C F 3CF2Cl
Cl
115
Figure 34. EI m ass s p e c t r u m of C F 3C F H 2
116
B
Figure 35. EI mass s p e c t r u m of C F 2ClBr
117
100%
Figure 36. EI mas s s p e c t r u m of CF2B r 2
118
100%
95i
F
266
2
A6 m/ z
Figure 37. EI m ass s p e c t r u m of C F 2B r C F 2Br.
Br
119
100%
Figure 38. EI mass s p e c t r u m of C H 3Br
120
Figure 39. EI m a s s s p e c t r u m of CH2C l B r .
121
Figure 40. EI mass s p e c t r u m of C H 2B r 2.
122
100%
9s i
9 oi
r
Figure 41. EI mass s p e c t r u m of C H B r 3.
MONTANA STATE UNIVERSITY LIBRARIES
3 1762 10293806 3