Citation
Identification of Câ‚„ carbonyl compounds in air samples using LC/MS

Material Information

Title:
Identification of Câ‚„ carbonyl compounds in air samples using LC/MS
Creator:
Graham, Brian
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
147 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Carbonyl compounds -- Analysis ( lcsh )
Air -- Pollution -- Colorado -- Denver Metropolitan Area ( lcsh )
Air -- Pollution ( fast )
Carbonyl compounds -- Analysis ( fast )
Colorado -- Denver Metropolitan Area ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 146-147).
Thesis:
Department of Chemistry
Statement of Responsibility:
by Brian Graham.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
45552911 ( OCLC )
ocm45552911
Classification:
LD1190.L46 2000m .G72 ( lcc )

Downloads

This item has the following downloads:


Full Text
IDENTIFICATION OF C4
CARBONYL COMPOUNDS
IN AIR SAMPLES
USING LC/MS
by
Brian Graham
B.S., Wake Forest University, 1994
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
Chemistry
2000


This thesis for the Master of Science
degree by
Brian Graham
has been approved
by
Z/ ~2-0GC)
Date
Phil Anderson


Graham, Brian (M.S., Chemistry)
Identification of C4 Carbonyl Compounds
in Air Samples using LC/MS.
Thesis directed by Larry Anderson
ABSTRACT
Mass spectrometry is evaluated as an alternative and complementary form of
detection for carbonyl compounds in air samples. Specific emphasis is placed on
C4 carbonyl compounds, which can be difficult to identify with the current
method of HPLC separation followed by UV detection. Two types of
atmospheric pressure ionization techniques are evaluated. These include ionspray
and atmospheric pressure chemical ionization. Standard mixtures of C4 carbonyl
compounds include crotonaldehyde, methacrolein, methyl ethyl ketone and
butyraldehyde. These compounds are evaluated in terms of how well the mass
spectrometer can distinguish between isomers at low levels of detection.
Atmospheric pressure chemical ionization can be used to identify the C4 isomers
at levels as low as 50 parts per trillion volume while use of ionspray ionization
allows detection as low as 500 parts per trillion volume. Air samples taken from
the northern Denver metro area confirm the presence of butyraldehyde and methyl
ethyl ketone when either ionization technique is used. Methacrolein and
crotonaldehyde however, could not be detected in the Northern Denver samples,
but both were detected in samples from downtown Denver.
in


This abstract accurately represents the content of the candidates thesis. I
recommend its publication.
Signed
iv


DEDICATION
To
Mom and Dad


ACKNOWLEDGMENT
I am indebted to my advisor, Dr. Larry Anderson, for his guidance and patience
with me during the course of this work, and also for his advice and support during
the entire course of my academic pursuits. I would also like to thank Dr. Ed
Furlong and Jeff Cahill for their advice, support and most importantly use of their
facilities at the United States Geological Survey. I would also like to thank Phil
Anderson for his suggestions and finally Gene Bouche for collecting samples.


CONTENTS
Figures............................................................ x
Tables.............................................................. xvi
Chapters
1. Introduction................................................... 1
1.1 Purpose of Study............................................... 1
1.2 Carbonyl Compounds............................................. 2
1.3 Health Effects of Carbonyl Compounds........................... 4
1.4 Sources of Carbonyl Compounds.................................. 5
1.4.1 Primary Sourcess............................................. 5
1.4.2 Secondary Sources............................................ 6
1.5 Measurement Techniques of Carbonyl Compounds................... 10
1.5.1DNPH coated Cartridge Technique.............................. 11
1.5.2 High Performance Liquid Chromatography....................... 12
1.5.3 Ultraviolet Detection........................................ 13
1.6 Mass Spectrometry.............................................. 13
1.6.1 History of LC/MS............................................. 14
1.6.2 Advantages of API Interfaces................................. 15
1.6.3 API Ionization Techniques.................................... 16
1.6.4 Interfaces with Mass Spectrometer............................ 20
1.6.5 Mass Analyzers............................................... 20
1.7 Previous Studies............................................... 21
2. Experimental................................................... 25
vii


2.1 Standards and Samples............................................. 25
2.1.1 Standards and Preparation....................................... 25
2.1.2 Samples......................................................... 27
2.2 Instrumentation and Accessories................................. 27
2.3 Analytical Techniques............................................. 28
2.3.1 Purity of C4 Standards...........................................28
2.3.2 Flow Injection Analysis......................................... 29
2.3.3 Precision.......................................................31
2.3.4 Chromatography of Standard Mixtures............................. 32
2.3.5 Determination of Detection Limit................................ 33
2.3.6 Analysis of Samples............................................. 33
2.3.7 Further Exploration............................................. 34
3. Results and Discussion............................................. 35
3.1 Standard Concentration and Purity................................. 35
3.2 Flow injection Analysis......................................... 39
3.2.1 Ionspray Ionization............................................. 39
3.2.2 Atmospheric Pressure Chemical Ionization........................46
3.3 Precision......................................................... 53
3.3.1 Ionspray Ionization............................................. 53
3.3.2 Atmospheric Pressure Chemical Ionization........................61
3.4 Chromatography of Standard Mixtues................................ 69
3.4.1 Chromatography using Ionspray................................... 69
, 3.4.2 Chromatography using Atmospheric Pressure Chemical Ionization... 82
3.5 Limit of Detection................................................ 95
3.5.1 Ionspray........................................................ 95
3.5.2 Atmospheric Pressure Chemical Ionization...................... 105
vni


3.6 Sample Analysis.................................................... 117
3.6.1 Ionspray.......................................................... 117
3.6.2 Atmospheric Pressure Chemical Ionization.......................... 120
3.6.3 Quantitative Analysis of Sample using APCI........................ 131
3.7 Further Exploration................................................ 137
3.7.1 Analysis of Acrolein.............................................. 137
3.7.2 Analysis of Methyl Vinyl Ketone................................... 138
4. Conclusions.......................................................... 140
4.1 Future Considerations................................................141
Appendix
A Calculations.........................................................142
References...............................................................146
IX


FIGURES
Figure
1.1 Electrospray Ionization....................................... 17
1.2 Atmospheric Pressure Chemical Ionization...................... 19
1.3 Structure Elucidation Scheme for Carbonyl Compounds using
APCI in the Negative Mode......................................... 23
3.1 UV Chromatogram of Methacrolein Derivative Standard........... 36
3.2 UV Chromatogram of Crotonaldehyde Derivative Standard......... 36
3.3 UV Chromatogram of Methyl Ethyl Ketone Derivative Standard.... 37
3.4 UV Chromatogram of Butyraldehyde Derivative Standard.......... 37
3.5 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 50V........ 40
3.6 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 60V........ 41
3.7 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 70V........ 42
3.8 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 80V....... 43
3.9 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 90V........ 44
3.10 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 100V..... 45
3.11 C4 Carbonyl Mass Spectra using APCI, Fragmentor 70V.......... 47
3.12 C4 Carbonyl Mass Spectra using APCI, Fragementor 80V........... 48
3.13 C4 Carbonyl Mass Spectra using APCI, Fragementor 90V........... 49
3.14 C4 Carbonyl Mass Spectra using APCI, Fragementor 100V.......... 50
3.15 C4 Carbonyl Mass Spectra using APCI, Fragementor 110V.......... 51
3.16 C4 Carbonyl Mass Spectra using APCI, Fragementor 120........... 52 .
3.17 Proposed Fragment Mechanisms for Crotonaldehyde using Ionspray.... 57
3.18 Proposed Fragment Mechanisms for Methacrolein using Ionspray. 58
3.19 Proposed Fragment Mechanisms for MEK using Ionspray....... 59


Figure
3.20 Proposed Fragment Mechanisms for Butyraldehyde using Ionspray. 60
3.21 Proposed Fragment Mechanisms for Crotonaldehyde using APCI.... 65
3.22 Proposed Fragment Mechanisms for Methacrolein using APCI...... 66
3.23 Proposed Fragment Mechanisms for MEK using APCI............... 67
3.24 Proposed Fragment Mechanisms for Butyraldehyde using APCI..... 68
3.25 Total Ion Chromatogram of Mixture of Three Carbonyls
using Ionspray....................................................... 69
3.26 Total Ion Current of the 4 Carbonyl Derivative Standard
using Ionspray....................................................... 70
3.27 Mass Spectra using Ionspray under Full Scan................... 71
3.28 Total Ion Current for 4 Carbonyl Derivative Standard and blank
using Ionspray under SIM (68,149, 251) monitoring for Crotonaldehyde
and Methacrolein..................................................... 72
3.29 Mass Spectra using Ionspray under SIM.......................... 73
3.30 Total Ion Current for 4 Carbonyl Derivative Standard using
Ionspray under SEM (70,138,193,236,153) monitoring for MEK
and Butyraldehyde.................................................... 74
3.31 Mass Spectra using Ionspray under SIM......................... 75
3.32 Total Ion Current of the 13 Carbonyl Derivative Standard
using Ionspray under Full Scan....................................... 76
3.33 Mass Spectra using Ionspray under Full Scan.................... 77
3.34 Total Ion Current for 13 Carbonyl Derivative Standard using
Ionspray under SIM (68, 149,251) monitoring for Crotonaldehyde
and Methacrolein..................................................... 79
xi


Figure
3.35 Mass Spectra using Ionspray under SIM........................... 79
3.36 Total Ion Current for 13 Carbonyl Derivative Standard using
Ionspray under SIM (70, 138, 193, 236, 153) monitoring for MEK
and Butyraldehyde...................................................... 81
3.37 Mass Spectra using Ionspray under SIM........................... 82
3.38 Total Ion Chromatogram of Mixture of Three Carbonyls
using APCI............................................................. 83
3.39 Total Ion Current of the 4 Carbonyl Derivative Standard using
APCI under Full Scan................................................... 84
3.40 Mass Spectra using APCI under Full Scan......................... 84
3.41 Total Ion Current for 4 Carbonyl Derivative Standard
using APCIunder SEM (79, 172, 202, 249) monitoring for Crotonaldehyde
and Methacrolein....................................................... 85
3.42 Mass Spectra using APCI under SIM............................... 86
3.43 Total Ion Current for 4 Carbonyl Derivative Standard using
APCI under SIM (163, 221, 251) monitoring for MEK and Butyraldehyde... 87
3.44 Mass Spectra using APCI under SIM............................... 87
3.45 Total Ion Current of the 13 Carbonyl Derivative Standard using
APCI under Full Scan.................................................. 89
3.46 Mass Spectra using APCI under Full Scan.......................... 89
3.47 Total Ion Current for 13 Carbonyl Derivative Standard using
APCI under SIM (79, 172, 202, 249) monitoring for Crotonaldehyde
and Methacrolein....................................................... 91
3.48 Mass Spectra using APCI under SIM................................ 91
xii


Figure
3.49 Total Ion Current for 13 Carbonyl Derivative Standard using
APCI under SIM (76,221, 251) monitoring for MEK and Butyraldehyde. 93
3.50 Mass Spectra using APCI under SIM.................... 93
3.51 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture
using SIM at 68,149,251 with Ionspray monitoring for Crotonaldehyde
and Methacrolein.................................................. 96
3.52 Mass Spectra of 1/10 Solution with Ionspray.................. 96
3.53 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture
using SIM at 70, 138,193,236, 253 with Ionspray monitoring for MEK
and Butyraldehyde................................................. 97
3.54 Mass Spectra of 1/10 Solution with Ionspray.................. 98
3.55 Total Ion Current for a 1/100 Solution of the 4 Carbonyl
Mixture using SIM at 68,149,251 with Ionspray monitoring for
Crotonaldehyde and Methacrolein.................................. 100
3.56 Mass Spectra of 1/100 Solution with Ionspray................. 100
3.57 Total Ion Current for a 1/100 Solution of the 4 Carbonyl
Mixture using SIM at 70, 138, 193, 236, 253 with Ionspray monitoring for
MEK and Butyraldehyde............................................ 101
3.58 Mass Spectra of 1/100 Solution with Ionspray....... 102------------------
3.59 Total Ion Current for a 1/1000 Solution of the 4 Carbonyl
Mixture using SIM at 68,149,251 with Ionspray.................... 104
3.60 Total Ion Current for a 1/1000 Solution of the 4 Carbonyl
Mixture using SIM at 70, 138,193, 236, 253 with Ionspray.......... 104
xm


Figure
3.61 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture
using SIM at 79, 172, 202, and 249 with APCI monitoring for
Crotonaldehyde and Methacrolein........................................ 106
3.62 Mass Spectra of 1/10 Solution with APCI......................... 106
3.63 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture
Using SIM at 76, 221,251 with APCI monitoring for MEK and
Butyraldehyde.......................................................... 107
3.64 Mass Spectra of 1/10 Solution with APCI......................... 108
3.65 Total Ion Current for a 1/100 Solution of the 4 Carbonyl Mixture
using SIM at 79,172, 202, and 249 with APCI monitoring for Crotonaldehyde
and Methacrolein....................................................... 110
3.66 Mass Spectra of 1/100 Solution with APCI........................ 110
3.67 Total Ion Current for a 1/100 Solution of the 4 Carbonyl Mixture
using SIM at 76, 221, 251 with APCI monitoring for MEK and
Butyraldehyde.......................................................... Ill
3.68 Mass Spectra of 1/100 Solution with APCI......................... 112
3.69 Total Ion Current for a 1/1000 Solution of the 4 Carbonyl Mixture using SIM
at 79,172, 202, and 249 with APCI monitoring for Crotonaldehyde
and Methacrolein..................................................... 114
3.70 Mass Spectra of 1/1000 Solution with APCI........................ 114
3.71 Total Ion Current for a 1/1000 Solution of the 4 Carbonyl Mixture using SIM
at 76, 221, 251 with APCI monitoring for MEK and
Butyraldehyde.......................................................... 115
3.72 Mass Spectra of 1/1000 Solution with APCI........................ 116
3.73 Chromatogram of a Sample using Ionspray under Full Scan......... 117
xiv


Figure
3.74 Chromatograms of Samples 1-4 using Ionspray monitoring for
Methacrolein and Crotonaldehyde under SIM (68,149,251)........... 118
3.75 Chromatograms for Samples 1-4 using Ionspray monitoring for
MEK and Butyraldehyde under SIM (70,138,193,236)................. 119
3.76 Chromatogram of a Sample using APCI under Full Scan......... 120
3.77 Chromatograms for Samples 1-4 using APCI monitoring for Methacrolein
and Crotonaldehyde under SIM (79,172, 202, 249).................. 122
3.78 Chromatograms for Sample 1-4 using APCI monitoring for
MEK and Butyraldehyde under SIM (163,221,251).................... 124
3.79 Chromatograms for Downtown Denver Air Samples monitoring
for Crotonaldehyde and Methacrolein.............................. 125
3.80 Chromatograms for Downtown Denver Air Samples monitoring
for MEK and Butyraldehyde......................................... 128
3.81 Calibration Curves for Crotonaldehyde and Methacrolein....... 132
3.82 Calibration Curves for MEK and Butyraldehyde................. 133
3.83 Diurnal Variation for C4 Compounds from Downtown Denver..... 136
3.84 Chromatogram and Mass Spectrum of Acrolein using Full Scan
Conditions........................................................ 137
3.85 Chromatogram and Mass Spectrum of MVK using APCI under-- -----
Full Scan......................................................... 138
3.86 Chromatogram and Mass Spectrum of MVK using APCI under
SIM Mode (79, 172, 202, 249)...................................... 139
xv


TABLES
Table
1.1 Four Carbon Aldehydes and Ketones................................ 3
2.1 Molecular Weights of Carbonyl Derivatives......................... 25
2.2 Ionspray MSD Parameters.......................................... 30
2.3 APCI MSD Parameters.............................................. 31
3.1 Concentrations of the C4 Carbonyl Derivative Solutions............ 35
3.2 Concentrations of the C4 Carbonyl Derivative Mixture Solution.... 38
3.3 Precision for C4 Carbonyl Compounds using Ionspray in Total
Ion Current........................................................... 54
3.4 Major Fragments for Methacrolein using Ionspray.................. 55
3.5 Major Fragments for Crotonaldehyde using Ionspray................ 55
3.6 Major Fragments for Methyl Ethyl Ketone using Ionspray........... 56
3.7 Major Fragments for Butyraldehyde using Ionspray................. 56
3.8 Precision for C4 Carbonyl Compounds using APCI in Total Ion
Current.............................................................. 62
3.9 Major Fragments for Methacrolein using APCI..................... 63
3.10 Major Fragments for Crotonaldehyde using APCI................... 63
3.11 Maj or Fragments for Methyl Ethyl Ketone using APCI............. 64
3.12 Maj or Fragments for Butyraldehyde using APCI................... 64
3.13 Relative Abundance of Ions using Ionspray under SIM............. 73
3.14 Relative Abundance of Ions using Ionspray under SIM............. 75
3.15 Retention Times and Major Fragments of Compounds in
Standard Mixture of 13 Derivatives............................ 78
3.16 Relative Abundance of Ions using Ionspray under SIM............. 80
3.17 Relative Abundance of Ions using Ionspray under SIM............. 82
xvi


Table
3.18 Relative Abundance of Ions using APCI under SIM.............. 84
3.19 Relative Abundance of Ions using APCI under SIM.............. 88
3.20 Retention Times and Major Fragments of Compounds in
Standard Mixture of 13 Derivatives............................... 90
3.21 Relative Abundance of Ions using APCI under SIM.............. 92
3.22 Relative Abundance of Ions using APCI under SIM.............. 94
3.23 Relative Abundances of Selected Ions using Ionspray of 1/10
Soluton.......................................................... 97
3.24 Relative Abundances of Selected Ions using Ionspray of 1/10
Solution......................................................... 98
3.25 Relative Abundances of Selected Ions using Ionspray of 1/100
Soluton.......................................................... 101
3.26 Relative Abundances of Selected Ions using Ionspray of 1/100
Solution......................................................... 102
3.27 Relative Abundances of Selected Ions using APCI of 1/10
Soluton.......................................................... 107
3.28 Relative Abundances of Selected Ions using APCI of 1/10
Solution......................................................... 108
3.29 Relative Abundances of Selected Ions using APCI of 1/100
Soluton.......................................................... Ill
3.30 Relative Abundances of Selected Ions using APCI of 1/100
Solution......................................................... 112
3.31 Relative Abundances of Selected Ions using APCI of 1/1000
Soluton.......................................................... 115
xvii


Table
3.32 Relative Abundances of Selected Ions using APCI of 1/1000
Solution......................................................... 116
3.33 Relative Abundances of Selected Ions using Ionspray of Samples
1-4............................................ ................. 120
3.34 Relative Abundances of Selected Ions using APCI of Samples 2
and 4............................................................. 123
3.35 Relative Abundances of Selected Ions using APCI of Samples
1-4............................................................... 125
3.36 Relative Abundances of Selected Ions in Samples 1-7 monitoring
for Crotonaldehyde and Methacrolein..............................:: 128
3.37 Relative Abundances of Selected Ions in Samples 1-7 monitoring
for MEK and Butyraldehyde......................................... 131
3.38 Integrated Areas of the Ion current for the C4 compounds and their
Concentrations.................................................... 131
3.39 Integrated Areas of the Ion Current and Concentrations for the C4
Compounds in the Samples collected from Northern Denver........... 134
3.40 Integrated Areas of the Ion Current and Concentrations for the C4
Compounds in the Samples collected from Downtown Denver........... 135
3.41 Relative Abundances of Selected Ions using APCI for
MVK............................................................... 139
xvm


1. Introduction
1.1 Purpose of the Study
Over the past two decades carbonyl compounds in the earths atmosphere have
been increasingly attracting the attention of scientists. This is due in part to the
role they play in the chemistry of the atmosphere as well as the fact that some
these compounds are known hazardous pollutants. Carbonyl compounds are
either directly emitted (primary sources) or are formed in the atmosphere
(secondary sources). Once in the atmosphere these compounds can undergo
photolysis contributing a significant source of free radicals, which are responsible
for the oxidation of hydrocarbons. They also are the precursors of oxidants
including ozone, peroxyacyl nitrates, and peroxycarbocylic acids (1). Because of
the environmental importance carbonyl compounds represent it is important to
have an analytical technique that is both selective and sensitive in determining the
identity and quantity of these compounds in the atmosphere. Currently several
techniques exist that measure the gas-phase concentration of carbonyl
compounds. The most common method in use today makes use of high-
performance liquid chromatography (HPLC) with ultraviolet (UV) detection of
the 2,4-dinitrophenylhydrazone derivative (2). This method performs quite well
with lower molecular weight carbonyls, for example those containing one to three
carbon atoms since these compounds are easy to identify and concentrations are
typically larger than those of a higher molecular weight. Specifically, carbonyls
containing four carbon atoms (C4) are difficult to identify using the
aforementioned method of UV detection. An exploration into mass spectrometry
1


as an alternative method of detection for the 2,4-dinitrophenylhydrazone
derivative with HPLC separation is investigated as a means of identifying
carbonyl compounds with four carbon atoms. Several ionization techniques are
examined as to determine the most effective for identification of these
compounds.
1.2 Carbonyl Compounds
Carbonyl compounds are identified as either an aldehyde or a ketone. The central
feature of aldehydes and ketones is the carbonyl group. The carbonyl group
contains a carbon atom that has a double bond to oxygen as follows:
\
/
c=o
The carbonyl group in aldehydes is bonded to at least one hydrogen atom and
either one or no carbon atoms. The following is a general formula for an
aldehyde where R represents an alkyl group or a hydrogen:
O
ii
A
The carbonyl group in ketones is bonded to two carbon atoms. The following is a
* general formula for a ketone where R represents an alkyl group that is the same
or different from R: ^ -....... .....
O
ii
A
R R
Table 1.1 lists examples of important four carbon carbonyls used in industry as
solvents, intermediates, fuel additives, flavorings, etc.
2


Table 1.1 Four Carbon Aldehydes and Ketones
0 II i 2-butenaI i (Crotonaldehyde)
0 . 2-methvlpropenal (Methacrolein)
0 3-buten-2-one (Methyl Vinyl Ketone)
0 II \^c\ 2-butanone (Methyl Ethyl Ketone)
0 II Butanal (Butyraldehyde)
0 r" 2-methvlp rop an al (Isobutyraldehvde)


1.3 Health Effects of Carbonyl Compounds
Aldehydes and ketones have been implicated as mutagens and possible
carcinogens (2). Most of these compounds have been identified as mucus
membrane irritants, and can cause pulmonary, skin, eye, and central nervous
system irritation. The Environmental Protection Agency (EPA) has classified
many of these compounds as toxic and hazardous. For example, formaldehyde
(HCHO) contributes to eye, nose, and throat irritation. It has also been shown to
cause bronchial asthma-like symptoms with some reports of asthma attacks, as
well as allergic dermatitis (3). As a result of the toxicities of carbonyl compounds
the Occupational Safety and Health Administration (OSHA) and the National
Institute for Occupational Safety and Health (NIOSH) have recommended
Permissible Exposure Limits (PEL), as well as Short Term Exposure Limits
(STEL). Information concerning these limits can be found at their respective
government websites (www.osha.gov or www.cdc.gov). These limits are set as
maximum exposure limits as time weighted averages. The PEL is usually eight
hours and the STEL is fifteen minutes. Most limits usually range from a parts per
billion (ppb) to parts per million (ppm) exposure level for a given time period
depending on the compound. Since aldehydes are very reactive the secondary
compounds formed from aldehydes, especially peroxyacylnitrates are much more
dangerous to the environment. The problem of the ecotoxcity of
peroxyacylnitrates cannot be separated from that of aldehydes (4).
4


1.4 Sources of Carbonyl Compounds
As stated previously, aldehydes and ketones originate from primary sources that
are emitted directly into the atmosphere or from secondary sources in which these
compounds are formed in the atmosphere by chemical transformations. Emission
of primary compounds can be classified into two different types. The first is by
anthropogenic processes, which include industrial, domestic, or agricultural
emissions. The second is biogenic activity which is a result of natural processes
such as plants and mammals (4). Secondary sources can also originate from
anthropogenic or biogenic processes.
1.4.1 Primary Sources
The main sources of anthropogenic emissions stem from motor vehicle traffic,
trash incineration, industrial and domestic heating, and exploitation of fossil fuels
for energy production (4). Motor vehicle emissions have attracted the most
attention since this seems to be an easy source to control. The government has
required the use of oxygenated fuels in motor vehicles in many areas across the
country in hopes of reducing air pollution by controlling carbon monoxide levels.
It has been shown that the use of these fuels which include methanol, ethanol, and
methyl tertiary butyl ether (MTBE) blended fuels, has increased the emissions of
carbonyl compounds (5). Specifically formaldehyde and acetaldehyde ------------
(CH3CHO) concentrations have increased. Other sources of emission include
many industries that emit a wide range of carbonyls. These industries include
refining and petrochemical, coal, plastics, paints, and varnish, sewage treatment,
and facilities where carbonyl compounds are manufactured (4). The most
important natural source of aldehydes and ketones is from the emissions of animal
5


excretions and forest fires (4). As it should be noted, humans do have an impact
on these sources. Many of the C4 carbonyl compounds enter the environment via
automobile exhaust, animal waste and biomass combustion (6). Crotonaldehyde,
methyl ethyl ketone, butyraldehyde, and methacrolein have all been detected in
car exhaust, biomass combustion and tobacco smoke (6). Methyl ethyl ketone
and butyraldehyde have also been detected in animal waste emissions (6).
1.4.2 Secondary Sources
In unpolluted environments most carbonyl compounds come from the photo-
oxidation of organic compounds which have biogenic origin. The majority of
organic compounds that are emitted into the atmosphere are likely to form
carbonyl compounds (4). In polluted environments, usually urban settings, the
situation is more complex, but the general mechanisms are the same and today are
well understood. The chemistry involving atmospheric hydrocarbons is initiated
by highly reactive OH radicals, NO3 radicals, or ozone (O3). These oxidizing
species exist as follows: Stratospheric ozone can be transported downwards into
the troposphere. It can also be formed from the photolysis of nitrogen dioxide
(4):
N02 + hv -> NO + 0(3P)
0(3P) + 02 + M -> 03 + M
OH radicals are formed from the photolytic destruction of ozone (4):
O3 + hv > O2 + 0('D)
O^D) + H20 -> 20H
6


Carbonyl compounds can also form the OH radical by photolysis. For example,
formaldehyde can be transformed by the following sequence (4):
HCOH + hv - H + HCO
H + 02 H02
HCO + 02 -* H02 + CO
H02 + NO -> OH + N02
The above processes due to their photolytic nature drive the daytime chemistry.
On the other hand the NO3 radical drives the nighttime chemistry since it
photolyzes very quickly. The largest source of the NO3 radical is by the
following process (4):
NO2 + O3 > NO3 + O2
1.4.2.1 Reactions with OH
During the daytime the majority of hydrocarbons will react with the OH radical in
the following manner (4):
RH +OH - R +H20
The organic radical then proceeds through the following steps to form a carbonyl
compound (4):
R + O2 ^ RO2
R02 +NO - RO +N02
7


RO + O2 - carbonyl compound + HO2
A good example of this sequence is the alkene isoprene (CH2=C(CH3)CH= CH2).
Isoprene and monoterpene hydrocarbons constitute the major portion of the
nonmethane organic compounds that are emitted from vegetation (7). Laboratory
studies have shown isoprene to be highly reactive under conditions characteristic
of the lower troposphere and is a precursor to C4 carbonyl compounds.
Specifically these compounds are methyl vinyl ketone(MVK) and methacrolein.
These products can be arrived at by the following schemes (7):
For methyl vinyl ketone:
CH2=CHC(CH3)= CH2 + OH -> CH2=CHC(CH3)CH2OH
CH2=CHC(CH3)CH2OH + 02 CH2=CHC(00)(CH3)CH20H
CH2=CHC(00)(CH3)CH20H + NO -> CH2=CHC(0)(CH3)CH20H + N02
CH2=CHC(0)(CH3)CH20H - CH3COCH=CH2(MVK) + CH2OH
CH2OH + 02 -> HCHO + H02
For methacrolein:
CH2=CHC(CH3)= CH2 + OH -> HOCH2CHC(CH3)=CH2
HOCH2CHC(CH3)=CH2 + 02 -> H0CH2CH(00)C(CH3)=CH2
H0CH2CH(00)C(CH3)=CH2 + NO^ H0CH2CH(0)C(CH3)=GH2 + N02
H0CH2CH(0)C(CH3)=CH2 - CH2OH + CH2=C(CH3)CHO(methacrolein)
CH2OH + 02 -> HCHO + H02
8


1.4.2.2 Reactions with Ozone
The reaction of hydrocarbons with ozone is slower than with the OH radical, but
this reaction becomes significant for alkenes. The gas-phase reactions of ozone
with alkenes proceeds by initial addition to form a primary ozonide, which in turn
rapidly decomposes to form two sets of carbonyl compounds plus biradical
products (8). This is shown below:
h,C(0)h2 + (R3h The initially energy rich biradicals can decompose eventually leading to the
OH radical or be conditionally stabilized (8). A good example of this sequence is
once again the reaction of the alkene isoprene. Isoprene reacts with ozone to form
methyl vinyl ketone, methacrolein, formaldehyde and biradicals. These biradicals
eventually, lead to production of more formaldehyde. The mechanism is as
follows:
Addition to the first double bond:
A
S -- C(CH 2 )CH =CH ,---------------
H 2
9


(i) CH3COCH=CH2(MVK) + ch2oo
(ii) HCOH + CH3C(00)CH=CH2
Addition to the second double bond:
o
H 2c-----C(CH j)c
H
(i) CH2=C(CH3)CHO(methacrolein) + CH200
(ii) HCOH + CH2=C(CH3)CHOO
1.4.2.3 Reactions with NO3
The mechanism for the reaction of the N03 radical with hydrocarbons is not fully
understood. However, it has been suggested that the reaction of N03 with alkenes
occurs as an addition followed by the decomposition of nitrooxy-alkoxy radicals
and the remaining products identified include a number of nitrooxy-carbonyls and
nitrooxy-alcohols (3).
1.5 Measurement Techniques of Carbonyl Compounds
As noted previously several techniques are presently used-to measure-gas-phase-
carbonyl compounds. The most widely used is the 2,4-dinitrophenyIhydrazine
(DNPH) coated cartridge technique which is used by the EPA in TO-11 (1). This
method, which uses HPLC for separation and UV radiation for detection, is
discussed here.
10


1.5.1 DNPH Coated Cartridge Technique
This technique involves coating a solid sorbent with acidified DNPH reagent and
packing the sorbent into a cartridge. The sorbents typically used are silica gel and
a Ci8 silica gel. Silica gel is a polymer of silica and oxygen with surface hydroxyl
groups. Ci8 silica gel a similar surface, but is coated with long hydrocarbon
chains. These chains are chemically bonded to the silica support through siloxane
groups (Si-O-R) (1). Air is sampled through the cartridge for a given period of
time. Any carbonyl compound in the air will be converted to a stable hydrazone
remaining on the sorbent. The process is described by the following reaction:
The hydrazine group (-HN-NH2) reacts with the carbonyl compounds to form
hydrazones. The acid-catalyzed reaction occurs when the lone pair of electrons
on the NH2 group attacks the carbonyl carbon of an aldehyde or ketone to form a
hydrazone. Removal of the hydrazones for subsequent analysis is accomplished
by extraction with acetonitrile.
11


1.5.2 High Performance Liquid Chromatography
Liquid chromatography is a physical method of separating compounds that makes
use of a stationary phase packed in a column and a liquid mobile phase.
Compounds are separated based on their different affinities for the two phases.
Modem systems make use of high pressure to force solvent through packed
columns that contain very fine particles, usually in the range of 3 pm to 10 pm, to
give high resolution separations. There are several liquid chromatographic
separation modes. These include: adsorption, ion-exchange, ion pair, size
exclusion, and partition chromatography. Partition chromatography can be
further divided into normal or reversed phase. The most widely employed mode
used today is reversed phase chromatography. This is also the method employed
when separating the DNP hydrazones, and will be the focus of the remainder of
the discussion. Reversed phase chromatography can simply be defined as making
use of a polar mobile phase and non-polar stationary phase. Polar solvents that
are typically used include: water, acetonitrile, methanol, and tetrahydrofuran
(THF). Combinations of any of the three organic solvents with water typically
provide a sufficient range of dipolar and hydrogen bonding interactions with
solutes to separate a vast number of compounds (9). The stationary phase is
typically packed into a steel column that is anywhere from 5 to 30 centimeters in
length with an internal dismeter of 4.6 mm. The stationary phase consists of
microporous particles of silica that are bonded with a non-polar hydrocarbon such
as the octadecyl (Cig) or octyl (Cs) chain. The basis for elution of compounds is
the partitioning of the solute between mobile phase and stationary phase. Eluent
strength is a measure of how fast the solvent or solvent mixture will elute the
analytes of interest. The eluent strength, in reversed phase chromatography
12


decreases as the solvent becomes more polar. Elution can be accomplished by
either isocratic (constant solvent mixture) or gradient (increasing amounts of a
second solvent are added over time) method. For a mixture of DNP hydrazones a
gradient method of water and increasing acetonitrile is employed with separation
on a Ci8 column.
1.5.3 Ultraviolet Detection
The UV detector is the most common HPLC detector in use today due to the fact
that most solutes absorb ultraviolet light and the detectors provide high
sensitivity. The flow from the HPLC is directed into a flow cell which is then
irradiated with the appropriate energy. Simple aldehydes and ketones only show
weak absorption bands in the UV region due to the n to n* electronic transition of
the carbonyl group. DNPH derivatives usually exhibit maxima around 360 nm
with much more intense absorption due to the higher molar absorptivity
associated with the n to % electronic transition (10). These transitions occur due
to the aromatic structure of the DNP Hydrazone where the DNP is the major
chromophore. Once derivatized these compounds can be easily detected with a
UV source.
1.6 Mass Spectrometry
Mass spectrometry (MS) is a powerful tool for the determination of analytes both-
qualitatively and quantitatively. The basic principle behind MS begins with
ionization of the analyte, followed by acceleration of the ions by an electric field
into an area where they are separated according to their mass. In the early years
MS was mainly used as an on-line detector for gas chromatography since analytes
needed to be in the gaseous state to be ionized. Thus, compounds that were
13


volatile and not thermally labile consisted of the bulk of analytes that were
analyzed. Over the past twenty years however great advances have been made in
the coupling of liquid chromatography with mass spectrometry. This has made it
possible to analyze compounds that are nonvolatile and/or thermally labile by MS.
Today liquid chromatography coupled with mass spectrometry (LC/MS) is a
robust, routinely applicable and widespread analytical technique (11).
1.6.1 History of LC/MS
Three major difficulties are met in interfacing the two powerful tools of liquid
chromatography and mass spectrometry (11). The first is how to accommodate
such a high flow rate, typically 1 ml/min of liquid from a conventional LC
column with the high vacuum required of the MS. The second is the solvent
composition, where non-volatile mobile phase additives usually find use in LC
separation development. Finally how does one analyze non-volatile and/or
thermally labile analytes? Only a handful of ionization techniques were originally
available to the scientist. These included electron ionization, chemical ionization,
and field desorption. Field desorption is not applicable to LC/MS. Electron
ionization and chemical ionization required ions to be in the gaseous state. Thus,
came the development of new techniques of ionization. Early interfaces proved to
be too complex, lacked ruggedness, and had limited sensitivity. The advent of
particle beam and thermospray overcame some of the initial shortcomings
permitting their use on a more routine basis (12). Although these ionization
techniques became more common, they still lacked sensitivity and specificity.
The particle beam interface is unable to ionize compounds that were not volatile
and is not sensitive enough to detect analytes below the parts per billion level.
Thermospray failed to provide sufficient fragmentation for compound
confirmation which could have been overcome with tandem MS, but would have
14


only increased the cost and complexity of the system. These techniques may still
find use, but a new technique of ionization appears to have superseded them (12).
This is atmospheric pressure ionization (API). At first it may seem impractical to
couple an ionization technique conducted at atmospheric pressure with mass
analysis carried out under a high vacuum. Combination of an API source with a
mass analyzer requires a 10 to 10 fold pressure reduction along the ion
passageway between the source and mass analyzer which can present a
demanding task for the vacuum pumping system (13). This has been overcome
with research yet despite the great potential of API only recently has it found
widespread use in analytical laboratories since manufacturers did not offer the
source as an option. The remainder of the discussion will focus on API due to its
extensive applications found in pharmaceutical, biochemistry and the
environmental sciences.
1.6.2 Advantages of API Interfaces
The combination of LC/MS is better suited with the ionization source operating at
atmospheric pressure due to the several advantages it presents (12). First, the LC
inlet typically operated at atmospheric pressure can be decoupled from the mass
spectrometer which operates at a pressure of 10"6 torr or lower. This provides an
opportunity to optimize operating conditions for both LC separation and MS
detection. Second, solvents and contaminants that elute into the atmospheric
chamber can be pumped away without degrading MS performance. Thus, a
rugged method is obtained. Third, liquid droplets can be more efficiently
desolvated at atmospheric pressure than at reduced pressures. Heat transfer is
more efficient at higher pressures. Fourth, as ions are sampled from the
atmosphere to the vacuum they undergo a free jet expansion. This causes
adiabatic cooling and helps maintain the integrity of labile compounds as well as
15


noncovalent complexes for mass analysis. Adiabatic cooling can be considered a
disadvantage since it can contribute to the formation of cluster ions. Cluster ions,
however can be eliminated by focusing lenses plus the acceleration region prior to
directing them into the mass analyzer. Finally, by controlling the electrostatic
potentials in this expansion region, collision induced decomposition (CID) can be
achieved. This can provide structural infoimation on the compounds of interest.
1.6.3 API Ionization Techniques
Two ionization techniques have taken hold of the majority of the market due to
the wide applicability of analytes that can be ionized. These include electrospray
ionization (ES) and atmospheric pressure chemical ionization (APCI). These two
ionization techniques are explored to determine how they characterize the DNPH
derivatives.
1.6.3.1 Electrospray Ionization
Electrospray ionization is an approach that does not require the sample to be
evaporated before it is ionized. Thus, samples that are thermally labile and polar
can be analyzed without a source of heat for evaporation or under other conditions
that are detrimental to analyte stability. Typical samples include analytes that
exist as ions in solution or can be ionized in solution. The ionization technique
depends on the dispersion of a sample solution into an electrically charged-aerosol -
(13). A potential difference between the spray capillary and the walls of the
source is used to supply the necessary electric field to overcome cohesive forces
holding the liquid together at the tip of the capillary to force ionization of
solution. Figure 1.1 shows atypical setup for electrospray ionization.
16


Figure 1.1 Electrospray Ionization
Skimmer
10*s Torn
Analyzer
l
Pumping
3 to 5 kV
The charged droplet consists of the solvent containing positive and negative ions
with the predominant charge depending on the polarity of the induced potential
(14). Before ions can be mass analyzed, the solvent must be removed. This is
done by a counter flow of neutral, heated drying gas, typically nitrogen, that
evaporates the solvent. The droplet diameter decreases and forces the
predominately like charges closer together at the surface of the droplet. When the
. Rayleigh limit(coulomb repulsion equals that of the surface tension of the droplet).
is reached the droplet explodes producing daughter droplets that are subject to
further evaporation. This process repeats itself and droplets with a high surface-
charged density are formed (15). Gas phase ion formation is the next process
which has been the subject of much debate. One theory predominates though and
that is the process of ion evaporation. In this model ions are emitted directly into
the gas phase. When the electric field created by the ions at the'surface of the
17


droplet exceeds the surface tension, bare analyte ions are emitted directly from the
droplet (15). Ion evaporation is very efficient. Nebulization of the LC effluent
results in nearly complete formation of charged droplets that can subsequently
undergo ion evaporation (12). In thermospray only 1 in 104-105 droplets is
charged under ion evaporation (12). Electrospray works better when a sample is
dissolved in a suitable organic solvent such as methanol or acetonitrile with a low
surface tension. It is more difficult to spray aqueous solutions which have a
higher surface tension and lower vapor pressure. This opposes the separation of
the droplets from the liquid front causing a less stable spray and bigger droplets
(13). The limitation of electrospray is the low flow rates that are required for
droplet formation. Flow rates typically range from 1-10 pL/min. This is not
always practical or rugged enough for solving separation problems. This has led
to the development of pneumatically assisted nebulization, commonly called
ionspray. This allows the use of higher flow rates and can handle mobile phases
with higher water content. Ionspray is now the most frequently used electrospray
technique.
1.6.3.2 Atmospheric Pressure Chemical Ionization
APCI is a gas phase ionization process. For this reason the most suitable
compounds are those that are somewhat volatile and thermally stable. Coupling
of the LC to the APCI system involves the use of a heated nebulizer inlet. The
heated nebulizer inlet consists of a concentric pneumatic nebulizer and a large
diameter heated quartz tube (11). Vaporization of the LC effluent occurs as it is
being swept through the heated tube. At this point the solvent and vapor are
introduced into the APCI source. Once in the source a corona discharge from a
needle with an applied voltage produces a current of 2 to 5 pA. See Figure 1.2 for
a typical APCI setup.
18


Figure 1.2 Atmospheric Pressure Chemical Ionization
Spray temperature

Skimmer
10
Torr
Analyzer
Pumping
The corona discharge provides a source of ionizing electrons that produces ions
from nearby gases. These ions readily undergo ion-molecule reactions with air
and other gaseous molecules because of the short mean free path between
molecules at atmospheric pressure (14). This produces reactive intermediates that
yield ion-molecule reactions. The majority of the ion-molecule reactions in APCI
involve acid-base chemistry and the ion chemistry can be very complex. Adduct
and cluster ions of air and water can be formed. This formation process results
from the high reactivity of the ions and the high concentration of other nearby gas
molecules (14). This can increase the background noise and reduce the signal-to-
noise ratio. Several approaches have proved successful in breaking up these
clusters, but the most common is the use of a gas curtain that falls between the ion
source and vacuum expansion. This method along with an applied electric field
provide CID during the free jet expansion. APCI has shown efficiency that is


103-104 times greater than electron impact with the possibility of 100% ionization
efficiency of a trace amount of sample.
1.6.4 Interfaces to Mass Spectrometer
Transportation from the atmospheric region to the vacuum of the mass
spectrometer usually involves the use of a capillary, a series of skimmers, and
several pumping stages which will help maintain the vacuum. Gas transport is
based on the pressure difference at the ends of the capillary. Since both ionization
methods discussed are soft ionization techniques, not capable of extensive
fragmentation, the exit potential of the capillary can directly change the energy of
the ions. At higher voltages, ions can experience many collisions with the air and
nitrogen which cause an increase in the internal energy of the molecules. This
internal energy can cause multiple bond cleavages which are necessary for
structural determination.
1.6.5 Mass Analyzers
Mass spectrometers can employ a wide variety of analyzers. It is ideal to have an
analyzer that is capable of distinguishing between 1 atomic mass unit as well as
allow enough ions through. Analyzers can be of the following types: a magnetic
field, double focusing, quadrupole mass filter, ion trap, time of flight, and fourier
transform-ion cyclotron resonance. The most common-analyzer is the quadrupole -
mass filter. This is because it is more compact, less expensive, and more rugged
than the others (10). This will be the focus of the discussion since it will be the
analyzer used in this study. The quadrupole instrument uses four parallel
cylindrical rods that serve as electrodes. Ions enter from one end and travel with a
constant velocity in a direction parallel to the poles, which is the z direction. The
20


ions acquire complex oscillations in the x and 3; directions by application of both a
direct current voltage and a radiofrequency voltage to the poles (16). A stable
oscillation occurs of a particular ion, allowing it to pass through without hitting
the poles that is dependent on the mass to charge ratio. Only ions of a single mass
to charge ratio will travel through the poles under a given set of frequencies. All
the other ions will strike the poles, losing their charge therefore being lost. Mass
scanning is carried out by varying the frequencies while keeping their ratios
constant (16). Typically quadrupole instruments resolve ions that differ in mass
by one atomic mass unit.
1.7 Previous Studies
Few studies have been found in the literature involving mass spectrometry
detection of 2,4 dinitrophenylhydrazone derivatives of carbonyl compounds.
What follows is a current summary of these studies. Early studies involved
collecting HPLC fractions and subjecting these fractions to mass spectrometry.
Since electron impact was the primary ionization technique of the day, this was
employed with poor results. The carbonyl derivatives suffered extensive
fragmentation making identification of carbonyls difficult. Chemical ionization
was first investigated by Grosjean of the DNPH derivatives in 1983 with positive
results (17). A variety of carbonyls were studied using positive chemical
ionization The most abundant peak in each spectrum was the protonated MH+
ion. Little fragmentation occurred and conclusions showed chemical ionization
was a useful identification tool. Olson and Swarin were the first to use an LC/MS
system with a moving belt interface followed by chemical ionization for the
carbonyl derivatives (18). They showed the negative ion mode with methane as
the chemical ionizing agent is more sensitive for certain DNPH derivatives. They
also showed how aldehydes and ketones can be distinguished based on the
21
i


intensity of the molecular ion peak relative to fragments, with ketones being the
more prominent. In 1995 Grosjean and Grosjean described use of a particle beam
interface (19). Positive ion mode was employed with methane as the reagent gas.
The mass spectra of the DNPH derivatives indicated little fragmentation. The
protonated molecular ion peak MH+ was the most abundant peak for all
compounds analyzed. These ionization techniques however are beginning to fade
out due to the emergence of atmospheric pressure ionization methods which offer
better detection limits (11). In 1998 Kolliker and co-workers (20) published a
procedure for identification of carbonyls using atmospheric pressure chemical
ionization mass spectrometry. They employed an ion trap mass spectrometer for
structure elucidation of the carbonyl derivatives in ambient air samples. APCI
was used in the negative ion mode since it was determined that electrospray in the
positive and negative ion mode had insufficient detection limits as well as APCI
in the positive ion mode. A structure elucidation scheme was developed by
interpreting mass spectra from thirty reference carbonyl compounds. This scheme
is displayed in Figure 1.3. The authors claim the following distinctions can be
made: 1. identification of DNPH derivatives, 2. differentiation between aldehydes
and ketones, 3. differentiation between a-saturated and unsaturated aldehydes, 4.
differentiation between branched and straight chained aldehydes and ketones. It
is the authors opinion that structures of coeluting DNPH carbonyls can be
identified, however this is backed up with little evidence.
22


Figure 1.3 Structure Elucidation Scheme for Carbonyl Compounds using
APCI in the Negative Mode
[M-HrofDNPH4tn MSAfS
-(MS/MS-pctruin
mfz 162
182 > f 79
UW2417
hvtlO
In 1999 Zurek and co-workers (21) developed a method for selective
determination and quantification of formaldehyde and acetaldehyde in air samples
using HPLC-APCI-MS. This is accomplished by calibrating using stable isotope
labeled standards. The authors wanted to quantify higher molecular weight
carbonyl derivatives, but were limited by the availability of the respective stable
isotope labeled carbonyl compounds. The isotopically labeled standards were ,
used as an internal standards to overcome changes in the MS conditions over long
periods of time.
23


Most recently Grosjean and co-workers developed a simple and sensitive LC/MS
method for detection of DNPH derivatives using atmospheric pressure chemical
ionization in the negative mode on a benchtop system (22,23). Electrospray and
APCI were both evaluated in the positive and negative ion modes. APCI negative
ion detection was found to provide more specific information and 1-2 orders of
magnitude better sensitivity (23). The work involved an investigation into 78
carbonyl compound derivatives. Experimental conditions were optimized in a
way that the molecular ion peak was the base peak for most compounds, revealing
little fragmentation. This was accomplished using the systems flow injection
analysis. Methacrolein, crotonaldehyde, methyl ethyl ketone and butyraldehyde
were some of the C4 carbonyls investigated. The data showed separation of these
compounds was difficult and mass spectrometer parameters were not optimized to
generate distinct fragments. Extracted ion chromatograms were evaluated
illustrating how the specificity of the technique can help with coelution.
Resolution between methyl ethyl ketone and butyraldehyde however, remained
less than optimal. The authors of this study also looked at the quantitative
abilities of the technique and found high comparability to UV detection. Finally,
the authors feel positive identification at levels as low as sub-ppb in ambient air
can be obtained. Future work in Grosjeans laboratory will be guided in a
direction of optimizing parameters to produce distinct fragments for specific
compounds.
24


2. Experimental Methods
2.1 Standards and Samples
Standards of DNPH derivatives of C4 carbonyl compounds and mixtures of
derivatized carbonyls will serve as the standards to be investigated by LC/MS.
Samples taken from the surrounding Denver metro area will also be analyzed in
an attempt to determine unidentified C4 carbonyl compounds.
2.1.1 Standards and Preparation
Pure standards were prepared for the following compounds: methyl ethyl ketone,
butyraldehyde, crotonaldehyde, and methacrolein. All four of these compounds
were derivatized with DNPH to form hydrazones. Table 2.1 contains the
molecular weights of these compounds.
Table 2.1 Molecular Weights of Carbonyl Derivatives
Carbonyl Derivative Formula Molecular Weight (amu)
Methacrolein C10 H10 N4 O4 250
Crotonaldehyde C10 H10 N4 O4 - 250
Methyl Ethyl Ketone C10 H12 N4 O4 252
Butyraldehyde C10 H12 N4 O4 252
25


The DNPH derivatives were obtained from Radian Corporation as solid crystals
in sealed separate vials. The stated chemical purity of each compound was 99%
and each vial contained approximately 10 mg of the hydrazones. The contents of
each vial are accurately weighed and transferred to separate 100 mL volumetric
flasks. Acetonitrile is used as the solvent and added to the calibration mark in
each flask. Each flask contains approximately a 100 ppm solution of the DNPH
derivative. These standards serve as the solutions needed to optimize certain
parameters for the mass spectrometer. Three different standard mixtures are used
to gauge chromatographic conditions and confirm mass spectrometry results. The
first is a mixture containing derivatized formaldehyde (MW 210), acetaldehyde
(MW 225), and acetone (MW 238). This solution was prepared by the
environmental sciences lab at UCD for their work involving quantification of
these compounds in air samples and contains roughly 16 parts per billion volume
of each compound. This solution will provide information on how LC/MS can
identify these compounds. The second is a mixture of the four C4 carbonyl
standard solutions discussed above. Using a volumetric pipet 10 mL is
transferred from each of the 100 mL volumetric flasks to a single 100 mL
volumetric flask. This solution is then diluted to the calibration mark with
acetonitrile. The resulting solution is a mixture of the four C4 carbonyl
compounds at approximately 10 ppm for each derivative. The third standard
mixture is a premade solution that contains 13-carbonyl compounds derivatized
with DNPH. This mixture was also obtained from Radian Corporation. The
purity of each compound is 99% and each derivative is at a concentration of 3
ppm. The following is a list of the compounds contained in the solution:
formaldehyde, acetaldehyde, acetone, acrolein (MW 236), propionaldehyde (MW
238), crotonaldehyde, methyl ethyl ketone, methacrolein, butyraldehyde,
benzaldehyde (MW 286), valeraldehyde (MW 266), m-tolualdehyde (MW 300),
and hexanal (MW 280).
26


2.1.2 Samples
As stated previously, samples from the surrounding Denver metro area were
analyzed. The first site is located just northeast of the city of Denver. This area is
known to contain many unidentified carbonyl compounds and specifically has
shown unidentified peaks in chromatograms with retention times in the vicinity of
C4 carbonyl compounds. The samples are collected using a vacuum pump
equipped with a flow meter control and an intake valve where a sample cartridge
can be connected. The flow meter is calibrated before entering the field to 3 liters
a minute. Each sample is collected for twenty-four hours for a total of 4,320 liters
of air. The second location is samples taken from downtown Denver. These
samples are collected as part of the ongoing study of air samples by the
Environmental Sciences Laboratory at the University of Colorado at Denver. An
automated sequential sampler is employed to obtain four hour samples throughout
the day. The system is designed to take in 720 liters of air for a four hour period
before switching to a new cartridge.
2.2 Instrumentation and Accessories
The United States Geological Survey (USGS) provided instrumentation for
analysis. The instrumentation consists of a Hewlett Packard 1100 HPLC system----
that includes a vacuum degasser, column thermostat, and an ultraviolet-visible
diode array detector. Separation was accomplished with a 15 cm (2.0 mm id),
ODS (3 pm) reversed phase column for electrospray and 25 cm (4.6 mm id),
Zorbax ODS (5 pm) reversed phase column for APCI. A guard column is
employed in each case to remove possible contaminants. The solvent system
27


consisted of a 10 mM solution of ammonium formate and acetonitrile for
electrospray analysis. The ammonium formate is added to aid in the ionization of
the analyte. The solvent system for APCI included deionized water and
acetonitrile. The mass spectrometer is a Hewlett Packard 1100 MSD that utilizes
an atmospheric pressure interface capable of both electrospray and atmospheric
pressure chemical ionization. Ionspray instead of pure electrospray is employed
since the instrument is set up to run in this mode and due to its advantages.
Nitrogen is Used as the nebulizing gas. Mass spectrometry parameters which are
capable of manual adjustment via the computer include capillary voltage (controls
the voltage applied to the entrance of the capillary), drying gas flow (controls the
flow rate of the nitrogen drying gas), nebulizer pressure (controls the pressure of
the nitrogen nebulizing gas), drying gas temperature (controls the temperature of
the nitrogen drying gas), ffagmentor (controls the exit voltage of the capillary thus
allowing different fragment patterns), corona current (controls the current from
the corona discharge needle, applicable only for APCI), and vaporization
temperature (sets the temperature for the vaporizer, applicable only for APCI). A
series of skimmers and plates follow the ionization chamber which accelerates the
ions towards the mass analyzer. The mass analyzer consists of a quadrupole mass
filter.
2.3 Analytical Techniques
2.3.1 Purity of C4 standards
Since the purity of the four carbonyl standards may be in question
chromatographic analysis with UV detection is employed. The vials obtained
from Radian Corporation of the carbonyl derivatives had been stored in a freezer,
but the date they were received is unknown, thus an investigation into the purity is
28


necessary. Each C4 standard is injected separately onto the Zorbax ODS column
and allowed 20 minutes to elute followed by detection at 360 nm. An isocratic
elution of acetonitrile/water (65/35) is employed for separation of potential
impurities. The relative amount of an impurity is calculated versus the recovery
of the main carbonyl derivative.
2.3.2 Flow Injection Analysis
The four C4 standards will be directly injected into the mass spectrometer as a
flow injection analysis. This is done for several reasons. The first is to determine
if the derivatives can be detected with a mass spectrometer using either ionspray
or APCI with the given solvents of choice. The controllable parameters described
above for the mass spectrometer will remain constant except for the fragmentor
voltage. The fragmentor voltage will be varied from a range of 50 volts to 100
volts when ionspray is used. Below 50 volts fragmentation is non-existent for the
molecular ion and above 100 volts fragmentation is so great that the mass
spectrum appears as noise. All the other parameters do not have the same
dramatic effect on the detection of the analytes as long as they are kept within a
range described and recommended by Hewlett Packard. Table 2.2 lists the
settings for these parameters when ionspray is in use.
29


Table 2.2 Ionspray MSD Parameters
Parameter Setting
Capillary Voltage 4000 volts
Drying Gas Flow 10 liters/min
Nebulizer Pressure 20 psi
Drying Gas Temperature 350 C
A fragmentor voltage will be chosen based on two factors. The first is detector
response gauged by the ion current. This is a value that describes how many ions
are being detected by the mass analyzer. The larger the ion current, the more
sensitive the method. Second is the pattern of fragmentation. A definite fragment
pattern for each analyte is the goal in determining the identity of the C4 carbonyl
compounds. In the case of atmospheric pressure chemical ionization the positive
and negative mode of analysis will be employed. The negative ion mode was not
investigated in ionspray since this would have involved the use of basic buffers
which would have damaged the column being used. The fragmentor voltage for
APCI will be varied from 70 volts to 120 volts for precisely the same reasons as
in the case for ionspray. Table 2.3 describes the other parameter settings that
remain constant throughout the analysis. -
30


Table 2.3 APCIMSD Parameters
Parameter Setting
Capillary Voltage 4000 volts
Drying Gas Flow 10 liters/min
Nebulizer Pressure 20 psi
Drying Gas Temperature 350 C
Corona Current 40 pA
Vaporization Temperature 425 C
2.3.3 Precision
Once a suitable fragmentor voltage has been selected the precision of the
analytical method will be determined at this voltage for each of the four standards.
Precision is a measure of the degree in which individual test results are in
agreement. This will be determined by measuring the repeatability of each
standard. Each standard is injected 10 times, once again bypassing the column.
Statistical calculations are performed for the ten standard injections to assess the
validity of the method. Repeatability is evaluated in terms of percent relative
standard deviation. Two criteria will be evaluated for precision. The first is the
total ion current. The second is the relative abundances of each fragment for that
particular standard. This is performed for both pneumatically assisted electrospray
and APCI.
31


2.3.4 Chromatography of Standard Mixtures
The standard mixture that contains formaldehyde, acetaldehyde and acetone
along with the standard mixture of the C4 derivatives is injected to gauge how the
mass spectrometer responds to the mixture of compounds for both ionization
techniques. The second of the two is used for verification of the mass
spectrometer in its performance to distinguish between and identify the four
carbonyl derivatives. The third standard mixture consisting of the 13 different
carbonyl derivatives is used to judge how effectively the C4 derivatives can be
identified within a different sample matrix. HPLC conditions for ionspray
consists of a gradient separation mode at a constant flow rate of 0.2 mL/min with
an injection volume of 10 |LL. Using the two solvents previously described initial
conditions consist of 45/55 percent composition of 10 mM ammonium formate
ancTacetbnitrile respectively. This is ramped to a composition of 35/65 percent
over twenty-five minutes. Finally this is ramped to a composition of 30/70
percent over the next ten minutes until a return to the initial conditions is
established in preparation for the next run. HPLC conditions for APCI consists of
a gradient separation mode at a constant flow rate of 0.8 mL/min with an injection
volume of 50 p.L. Using water and acetonitrile as the solvents, initial conditions
consist of a 50/50 percent composition followed by a ramp to 40/60 percent over
twenty-five minutes. This is ramped to 25/75 over the next fifteen minutes until a
return to the initial conditions is established in preparation for the next run. Both
ionization techniques will run a full scan mode by the mass spectrometer (50 to
300 amu) followed by a selected ion monitoring mode (SIM). The selected ions
are determined in the precision section by demonstrating unique fragment patterns
among isomers.
32


2.3.5 Determination of Detection Limit
The limit of detection is the lowest concentration of analyte in a sample that can
be detected, but not necessarily quantitated with a given degree of confidence
under the stated experimental conditions. This is an essential parameter that
needs to be addressed to ensure accurate determination of the carbonyl
compounds. The four carbonyl mixture will be used with analysis accomplished
in the selected ion monitoring mode. Solutions are prepared by diluting the four
carbonyl mixture to levels of roughly 1 parts per billion volume (ppbv), 100 parts
per trillion volume (pptv) and 10 pptv. These concentrations are air equivalent
concentrations and conversions to these values is described in Appendix A.
Accurate detection at these levels determines the limit of detection.
2.3.6 Analysis of Samples
Samples that contain unidentified C4 compounds are analyzed to determine the
effectiveness and specificity of the method. All parameters and conditions that
were optimized above remain the same. This is performed for both ionspray and
APCI to determine the most effective ionization technique for samples. Rough
quantitative data is determined for samples using the most effective ionization
technique. This is determined by using the limit of detection data to create
calibration curves for the C4 compounds. The numbers will be rough estimates
due to the standards and samples being run on separate days. This is an
unavoidable consequence of having limited time with use of the instrumentation.
33


2.3.7 Further Exploration
Samples of acrolein and methyl vinyl ketone are analyzed to determine various
trends for these compounds. Both compounds were derivatized with DNPH by
the Environmental Sciences Laboratory at the University of Colorado at Denver.
Solutions of approximately 100 ppm are prepared by dissolving 10 mg of sample
into a 100 mL volumetric flask with acetonitrile. Solutions are analyzed at these
concentrations.
34


3. Results and Discussion
3.1 Standard Concentrations and Purity
Actual concentrations of the four C4 carbonyl derivative solutions are given in
Table 3.1. Also included is the equivalent air concentrations based on a sample of
720 liters of air from the Denver metro area. A detailed account of these
concentrations can be found in Appendix A.
Table 3.1 Concentrations of the C4 Carbonyl Derivative Solutions
Derivative Concentration (pg/mL) Equivalent Air Concentration (ppbv)
Methacrolein 64.5 52.9
Crotonaldehyde 63.0 51.7
Methyl Ethyl Ketone 62.8 51.1
Butyraldehyde 67.6 55.0
Each of the four derivatives proved to be pure by analysis of the chromatograms.
Any impurity detected was calculated to be less than 0.1% total area versus the
main DNP-hydrazone peak. Chromatograms for the four hydrazones are shown in
figures 3.1-3.4.
35


Figure 3.1 UV Chromatogram of Methacrolein Derivative Standard
Figure 3.2 UV Chromatogram of Crotonaldehyde Derivative Standard
36


Figure 3.3 UV Chromatogram of Methyl Ethyl Ketone Derivative Standard
Figure 3.4 UV Chromatogram of Butyraldehyde Derivative Standard
37


Concentrations for the four carbonyl standard mixture are given in Table 3.2.
Included are the air equivalent concentrations based on 720 liters of Denver metro
air sampled.
Table 3.2 Concentrations of the C4 Carbonyl Derivative Mixture Solution
Derivative Concentration (pg/mL) Equivalent Air Concentration (ppbv)
Methacrolein 6.45 5.29
Crotonaldehyde 6.30 5.17
Methyl Ethyl Ketone 6.28 5.11
Butyraldehyde 6.76 . 5.50
All thirteen carbonyl compound derivatives found in the second mixture are at a
solution concentration of 3 pg/mL. This corresponds to an equivalent air
concentration of 2.46 ppbv for methacrolein and crotonaldehyde for 720 liters of
sampled air in the Denver metro area and 2.44 ppbv for methyl ethyl ketone and
butyraldehyde.
38


3.2 Flow Injection Analysis
3.2.1 Ionspray Ionization
A fragmentor voltage for pneumatically assisted electrospray was determined
first. The value chosen was 70 volts. This was the voltage of choice for several
reasons. First using 50 or 60 volts revealed little more than the molecular ion
peak making identification of isomers nearly impossible. Those values above 70
volts contained a variety of fragments, but once again it was difficult to
distinguish between isomers. At 70 volts, however unique fragments were
observed for each of the C4 carbonyl derivatives. The total ion current was
substantial enough for each of the voltages analyzed to allow for qualitative data.
No significant difference was seen in the total ion current area going from 50
volts to 100 volts. A comparison of each voltage is shown in figures 3.5-3.10. A
discussion of the fragments brought about at 70 volts will take place in the
subsequent section.
39


Figure 3.5 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 50V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
*MSD1 SPC, fimo=0.645 of F;\araAND-1\SGFV\ES\0CGOOOO3.O AR-ES PosiOv# *MSD1 SPC, 6jtb=20.6S4 of F:VBRIAND1\BGFIAES\SCGOOOQ3.D AFVES POSitrvi
100- ixI T9H2 100- H x: 549632
eo- 60 -
60- $0-
40- 40-
20- 20-
i
1
SO 100 ISO 300 250 mfc SO 100 1S0 200 250 lift
40


Figure 3.6 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 60V
1. Methacrolein
2. Crotonaldehyde
AiSDI SFC, 456 of F:\SRIAND-1\BGHAE5\BCG00003.D AR-ES Positive *MSD1 SPC,time=15.140o1 F:\BRIAND-1\BGFIAES1BCG00003.D APWE&Posrt/vi
100- i a*: 166S2B 10D- j art: 396864
BO- 00-
60- 60-
40- 40-
20- 20-
S ; 3
i s
. I 1 ,.i 1. 1 . ! . 1 .i. .rf i_ j-. 1
50 100 152 222 222 !n& 52 122 ISO 200 222 W
3. MEK
4. Butyraldehyde
MSD1 SPC, tin=1.794 of F:\flRlAND-1\3GFWES\BCGQ0003.D AR-E3 Positive
609664
41


Figure 3.7 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 70V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
42


Figure 3.8 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 80V
1. Methacrolein
2. Crotonaldehyde
3.MEK
4. Butyraldehyde
43


Figure 3.9 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 90V
1. Methacrolein 2. Crotonaldehyde
3.MEK
4. Butyraldehyde
MSD1 SPC, fime=25.105 of F:\fiRlAND-UBGFWES\flCG00003.D AR-ES PosiBvi
100- 80- 60- 40 20- n <1 M s 5 7 H Ji \k ^ ai l i 1 <> 1L 1 H ax 1 1V152 0 5 c L
SO 100 15D ~M0' ' 230 irvJ
44


Figure 3.10 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 100V
1. Methacrolein 2. Crotonaldehyde
3. MEK 4. Butyraldehyde
45


3.2.2 Atmospheric Pressure Chemical Ionization
A value of 90 volts was chosen for the ffagmentor in atmospheric pressure
chemical ionization. This followed the same reasoning as in ionspray ionization.
Values below 90 volts in most cases only revealed the molecular ion. Those
above 90 volts generated too many fragments making it difficult to distinguish
between isomers. Ninety volts proved to be a good median providing unique
fragments among isomers. One noticeable feature revealed as the fragmentor
voltages were increased was a decrease in the total ion current. This is a result of
ions losing their charge as they collide with other molecules. The ion current at
90 volts, however is large enough for adequate sensitivity. A comparison of each
voltage is shown in figures 3.11-3.16. A discussion of the major fragments
obtained at 90 volts is discussed in the ensuing section.
46


Figure 3.11 C4 Carbonyl Mass Spectra using APCI, Fragmentor 70V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
47


Figure 3.12 C4 Carbonyl Mass Spectra using APCI, Fragementor 80V
1. Methacrolein
2. Crotonaldehyde
3.MEK
4. Butyraldehyde
48


Figure 3.13 C4 Carbonyl Mass Spectra using APCI, Fragementor 90V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
49


Figure 3.14 C4 Carbonyl Mass Spectra using APCI, Fragementor 100V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
50


Figure 3.15 C4 Carbonyl Mass Spectra using APCI, Fragementor 110V
1. Methacrolein
2. Crotonaldehyde
3.MEK
4. Butyraldehyde
51


Figure 3.16 C4 Carbonyl Mass Spectra using APCI, Fragementor 120V
1. Methacrolein
2. Crotonaldehyde
3. MEK
4. Butyraldehyde
52


3.3 Precision
3.3.1 Ionspray Ionization
Table 3.3 lists the total ion current for ten replicate injections at a fragmentor
voltage of 70 volts for each of the carbonyl derivatives as well as percent relative
standard deviation (%RSD). All four of the compounds have a %RSD less than 2.
This indicates very good precision from injection to injection. Secondly, tables
3.4-3.7 show the major fragments for each of the four compounds including the
average relative abundance for each fragment (those underlined are significantly
different among isomers). Relative abundances tended to show greater
variability, but definite patterns could be ascertained from the data. For the two
isomers, methacrolein and crotonaldehyde, many of the fragments are similar and
abundances are close. Two noticeable differences can be made however, the first
is methacrolein does not show a fragment at a m/z of 68. The second is the
fragment at a m/z of 149 is much larger for methacrolein than crotonaldehyde.
Thus from this data it appears the two isomers can be distinguished. A more
drastic difference can be seen for the two larger isomers, methyl ethyl ketone and
butyraldehyde. The first is methyl ethyl ketone has only two considerable size
fragments at this fragmentor voltage. These include the molecular ion peak and a
fragment at a m/z of 70. Butyraldehyde shows a significant abundance for several
other ions except for the one with a m/z of 70. These two isomers appear to be
easily distinguishable by their fragments. What follows the tables of fragments
are probable fragment patterns for each of the four compounds in figures 3.17-
3.20.
53


Table 3.3 Precision for C4 Carbonyl Compounds using Ionspray in Total Ion
Current
Methacrolein Crotonaldehyde MEK Butyr aldehyde
1.23 x 107 7.53 x 106 1.38 x 107 1.41 x 107
1.21 x 107 7.30 x 106 1.39 x 107 1.39 x 107
1.19 x 107 7.28 x 106 1.41 x 107 1.38 x 107
1.17 x 107 7.38 x 106 1.39 x 107 1.38 x 107
1.21 x 107 7.23 x 106 1.39 x 107 1.37 x 107
1.19 x 107 7.57 x 106 1.39 x 107 1.37 x 107
1.19 x 107 7.39 xlO6 1.40 x 107 1.40 x 107
1.17 x 107 7.36 xlO6 1.42 xlO7 1.38 x 107
1.17 x 107 7.41 x 106 1.40 x 107 1.39 x 107
1.20 xlO7 7.62 x 106 1.39 x 107 1.37 x 107
Mean 1.19 x 107 7.41 x 10 1.40 x 107 1.38 x 107
Std Dev 2.07 x 105 1.28 x 105 1.27 xlO5 1.38 x 105
%RSD 1.73 1.73 0.91 1.00
54


Table 3.4 Major Fragments for Methacrolein using Ionspray
m/z ion Average Abundance
68 [M+H-183f <2%
149 [M+H-102]+ 30
164 [M+H-87]+ 10
183 [M+H-68]+ 15
234 [M+H-17]+ 20
251 [M+H]+ 100
Table 3.5 Major Fragments for Crotonaldehyde using Ionspray
m/z ion Average Abundance
68 [M+H-183f 15
149 [M+H-102f 12
164 [M+H-87]+ 10
183 [M+H-68]+ 12
234 [M+H-17]+ 14
251 [M+H]+ 100
55


Table 3.6 Major Fragments for Methyl Ethyl Ketone using Ionspray
m7z Ion Average Abundance
70 [M+H-183f 11
138 [M+H-115]+ <2%
164 [M+H-89]+ <2%
183 [M+H-70]+ <2%
193 [M+H-60]+ <2%
236 [M+H-17]+ <2%
253 [M+H]+ 100
Table 3.7 Major Fragments for Butyraldehyde using Ionspray
mix Ion Average Abundance
70 [M+H-183f <2%
138 [M+H-115]+ 19
164 [M+H-89]+ 17
183 [M+H-70]+ 16
193 [M+H-60]+ 13
236 [M+H-17]+ 18
253 [M+H]+ 100
I
56


Figure 3.17 Proposed Fragment Mechanisms for Crotonaldehyde using
lonspray
Abundance
-c 4h 9no 2

12
o 2n
MW 149
-C qH s N jO 4
,H
15
MW 68
57


Figure 3.18 Proposed Fragment Mechanisms for Methacrolein using
I MW 183
-CjHsNOj
Hj
N--N =CH
w
m
o2n
MW 164
MW 149
^0
58


Figure 3.19 Proposed Fragment Mechanisms for MEK using Ionspray
Abundance
+ /
N=C'
\
CH,
ch2
l3
t^H,
MW 70
\5
59


Figure 3.20 Proposed Fragment Mechanisms for Butyraldehyde using
Ionspray Abundance
O jN
MW 153
0 j N
HW133
ion
18
-N ;Q ;
\/0
n
o
MW 103
-C 4 H ,N
%__^
16
0 jN
MW1B3
-C ,H ,NO 3
V

0 2 N
MW 164
-C 4 K iN 2 O a
%_^
C aH 5 N jO
C

K
.C
CH
17
1-9
<2
mw 70
60


3.3.2 Atmospheric Pressure Chemical Ionization
Table 3.8 lists the total ion current for ten replicate injections at a fragmentor
voltage of 90 volts for each of the carbonyl derivatives as well as percent relative
standard deviation (%RSD). All four compounds have a %RSD of less than
three. This indicates good precision from injection to injection. Secondly, tables
3.9-12 show the major fragments for each of the four compounds including the
average relative abundance for each fragment (those underlined represent major
differences among isomers). Relative abundances tended to show greater
variability, and as in ionspray definite patterns could be established from the data.
Most of the ions fragments for methacrolein and crotonaldehyde are very similar
with the one difference being in the ion with a m/z of 79. This fragment is present
in the mass spectra of methacrolein, but not for crotonaldehyde. This ion
fragment enables a distinction to be made between the two isomers. These results
compare favorably with the work done by Kolliker and co-workers (18). The
only difference in fragment patterns is they saw fragments that included [M-H-
35]' and [M-H-72]'. Both of these fragments were absent at the given fragmentor
voltage. The two isomers methyl ethyl ketone and butyraldehyde show a greater
difference in their fragment patterns. The most notable being the absence of a
fragment at a m/z of 163 for methyl ethyl ketone. From this alone the two of
these compounds can be distinguished in a sample. These results also compare
favorably with the work done by Kolliker and co-workers (18). They claim for a
ketone an ion with m/z 163 < 179, [M-H-30]' is present, and m/z 152 is the
biggest fragment. This is indeed the case for methyl ethyl ketone. For the
aldehyde m/z 163 is abundant, [M-H-30]' is present, and m/z 191 is
approximately 10% of the base peak. This holds true for butyraldehyde. It is
evident from these results and the previous work these compounds tend to
61


fragment in given patterns. What follows the tables of fragments are probable
fragment patterns for each of the four compounds in figures 3.21-3.24.
Table 3.8 Precision for C4 Carbonyl Compounds using APCI in Total Ion
Current
Methacrolein Crotonaldehyde MEK Butyraldehyde
1.58 x 107 2.27 x 107 3.25 x 10y 2.36 x 107
1.60 x 107 2.26 x 107 3.24 x 107 2.36 x 107
1.62 x 107 2.28 x 107 3.27 x 107 2.42 x 107
1.66 x 107 2.22 x 107 3.34 x 107 2.40 x 107
1.62 x 107 2.30 x 107 3.31 xlO7 2.31 x 107
1.71 x 107 2.31 x 107 3.35 x 107 2.37 x 107
1.67 x 107 2.28 x 107 3.42 x 107 2.42 x 107
1.67 xlO7 2.24 x 107 3.39 x 107 2.36 x 107
1.70 x 107 2.27 x 107 3.25 x 107 2.39 x 107
1.65 x 107 2.34 x 107 3.40 x 107 2.37 x 107
Mean 1.65 x 107 2.28 x 10 3.32 x 10' 2.38 x 10'
Std Dev 4.24 x 105 3.43 x 105 6.78 x 105 3.31 x 105
%RSD 2.57 1.51 2.04 1.39
62


Table 3.9 Major Fragments for Methacrolein using APCI
mJz ion Average Abundance
79 [M-H-170]' 6
152 [M-H-97] 6
163 [M-H-86]- 13
172 [M-H-77]' 12
202 [M-H-47]' 12
218 [M-H-31]' 8
249 [M-H]' 100
Table 3.10 Major Fragments for Crotonaldehyde using APCI
mJz Ion Average Abundance
79 [M-H-170]' <2%
152 [M-H-97]- !i
163 [M-H-86]' 10
172 [M-H-77]' 22
202 [M-H-47]' 20
218 [M-H-31]' 10
249 [M-H]' 100
63


Table 3.11 Major Fragments for Methyl Ethyl Ketone using APCI
mJz ion Average Abundance
152 [M-H-99]' 24
163 [M-H-88]" <2%
179 [M-H-72]' 6
191 [M-H-60]' 4
221 [M-H-30]' 34
251 [M-H]' 100
Table 3.12 Major Fragments for Butyraldehyde using APCI
mlz Ion Average Abundance
152 [M-H-99]' 39
163 [M-H-88]' 22
179 [M-H-72]' 13
191 [M-H-60]' U
221 [M-H-30]' 27
251 [M-H]' 100
64


Figure 3.21 Proposed Fragment Mechanisms for Crotonaldehyde using
APCI
Abundance
-H N jO 3
H H
n
o
Mwm
10
e ah ,no
11
65


Figure 3.22 Proposed Fragment Mechanisms for Methacrolein using APCI
Abundance
O jN
MW2D1
n

0,N


N ^ N =CH
66


Figure 3.23 Proposed Fragment Mechanisms for MEK using APCI
Abundance
67


Figure 3.24 Proposed Fragment Mechanisms for Butyraldehyde using APCI
Abundance
2*
Nh,
inn
=g c2
\h ,
H,
27
11
n
22
-C .H NO
o 2n
NH

MW 152
19
68


3.4 Chromatography of Standard Mixtures
3.4.1 Chromatography using Ionspray
Figure 3.25 shows the total ion current using full scan from a m/z of 50 to 300 for
the mixture of formaldehyde, acetaldehyde, and acetone. The chromatogram
reveals formaldehyde cannot be detected using ionspray at this concentration.
The other two compounds can be identified, but the chromatogram does contain a
significant amount of noise. Acetaldehyde elutes at approximately 12.8 minutes
and acetone elutes at approximately 16 minutes.
Figure 3.25 Total Ion Chromatogram of Mixture of Three Carbonyls using
Ionspray
69


Figure 3.26 shows the total ion current chromatogram with full scan of the four
carbonyl derivative mixture using ionspray. Under the given conditions
separation is not great as seen by the three peaks. Figure 3.27 displays the mass
spectra of the three peaks. The first and last peak can easily be identified by their
mass spectrum, but the two in the middle are indistinguishable by retention times.
One can deduce what the two species are in this standard mixture, but this will
most certainly not hold up in a more complex matrix which contains other C4
carbonyls.
Figure 3.26 Total Ion Current of the 4 Carbonyl Derivative Standard using
Ionspray under Full Scan
NSTITCNGHeIBCZSiBimfiD AASRsfre
tom
110000*
100000
90000*
60000
70000
60000
$0000
40000

70


Figure 3.27 Mass Spectra using Ionspray under Full Scan
Figure 3.28 shows the first of two selected ion monitoring chromatograms of total
ion current for this mixture and a blank/ The selection of ions had a mass to
charge ratio of 68,149, and 251. With this choice of ions only two of the four
ions should be detected and this is indeed the case. The chromatogram shows two
peaks for the isomers having a molecular weight of 250 amu. The two other
peaks in the chromatogram can riot be identified even by shooting a blank of the
eluent. Figure 3.29 displays the mass spectra for these two peaks. Table 3.13
contains the relative abundances of the ion peaks. These numbers do not compare
71


to previously found abundances very well (see Tables 3.4 and 3.5). It appears that
crotonaldehyde elutes first with such a large fragment at 68. This can be
confirmed by the larger abundance of the 149 ion in the second eluting peak,
which is methacrolein.
Figure 3.28 Total Ion Current for 4 Carbonyl Derivative Standard and blank
using Ionspray under SIM (68,149,251) monitoring for Crotonaldehyde and
Methacrolein
72


Figure 3.29 Mass Spectra using Ionspray under SIM
Table 3.13 Relative Abundance of Ions using Ionspray under SIM
Ion Abundance
m/z Peak at 18.7 minutes Peak at 20.7 minutes
68 18 8
149 2 6
251 100 100
Figure 3.30 shows the second chromatogram of total ion current for selected ion
monitoring. The ions chosen here include a mass to charge ratio of 70,138,193,
236, and 253. These choices reveal the other two compounds. The peak which
73


elutes right before the two major peaks cannot be identified, but its major ion is
138. Figure 3.31 displays the mass spectra for these compounds. Table 3.14
contains the relative abundances for the two ion peaks. The evidence shown here
is conclusive that the first peak that elutes is methyl ethyl ketone and the second is
butyraldehyde. The large fragment at 70 for the first peak and the large fragments
at 138,193 and 236 for the second peak confirm these results.
Figure 3.30 Total Ion Current for 4 Carbonyl Derivative Standard using
Ionspray under SIM (70,138,193,236,153) monitoring for MEK and
Butyraldehyde
74


Figure 3.31 Mass Spectra using Ionspray under SIM
Table 3.14 Relative Abundance of Ions using Ionspray under SIM
Ion Abundance
m/z Peak at 20.3 minutes Peak at 21.9_minutes
70 17 5
138 2 20
193 1 13
236 3 15
253 100 100
75


The results for the 13 carbonyl compound mixture are very similar. Figure 3.32
shows the total ion current chromatogram under full scan using ionspray
ionization. Figure 3.33 displays the mass spectra for the C4 compounds. Under a
more complex matrix it is very difficult to make out more than the molecular ion
peak, thus identification of isomers is nearly impossible. Table 3.15 is a list of the
compounds which can be detected in this mixture. Under the given
chromatographic and mass spectrometry conditions formaldehyde, acrolein, and
methacrolein cannot be detected.
Figure 3.32 Total Ion Current of the 13 Carbonyl Derivative Standard using
Ionspray under Full Scan
76


Figure 3.33 Mass Spectra using Ionspray under Full Scan
77


Table 3.15 Retention Times and Major Fragments of Compounds in
Standard Mixture of 13 Derivatives
Retention Time (min) Major Ion Possible Compound
10.5 225 Acetaldehyde
13.8 239 Acetone
15.8 239 Propionaldehyde
18.7 251 Crotonaldehyde
20.4 253 Methyl Ethyl Ketone
21.8 253 Butyraldehyde
24.9 287 Benzaldehyde
29.0 267 Valeraldehyde
31.0 118 m-tolualdehyde
34.5 132 Hexanal
Figure 3.34 shows the first of two selected ion monitoring chromatograms of
total ion current for this mixture. The selection of ions as with the mixture of 4
compounds has a mass to charge ratio of 68,149, and 251. This selection
identifies crotonaldehyde and methacrolein. Figure 3.35 displays the mass spectra
for these two peaks. Table 3.16 contains the relative abundances of the ion peaks.
These numbers compare more favorably with the previous results of the ion
abundances. Crotonaldehyde is the first eluting peak due to the larger fragment at
68. This is confirmed by noting the second eluting peak has a large fragment at
149 which confirms methacrolein. The two unidentified peaks observed in the
mixture of standards is also observed in this chromatogram.
78


Figure 3.34 Total Ion Current for 13 Carbonyl Derivative Standard using
Ionspray under SIM (68,149,251) monitoring for Crotonaldehyde and
Methacrolein
Figure 3.35 Mass Spectra using Ionspray under SIM
79


Table 3.16 Relative Abundance of Ions using Ionspray under SIM
Ion Abundance
m/z Peak at 18.6 minutes Peak at 20.7 minutes
68 19 11
149 9 25
251 100 100
Figure 3.36 shows the second chromatogram of total ion current for selected ion
monitoring. The ions chosen remain the same for identification of methyl ethyl
ketone and butyraldehyde with amass to charge ratio of 70,138, 193, 236, and
253. Figure 3.37 displays the mass spectra for these compounds. Table 3.17
contains the relative abundances for the two ion peaks. The evidence shown here
is conclusive that the first peak that elutes is methyl ethyl ketone and the second is
butyraldehyde. The large fragment at 70 for the first peak and the large fragments
at 138, 193 and 236 for the second peak confirm these results. With the chosen
fragments other compounds show a response as seen in the chromatogram. These
responses are due to the fragments at a m/z of 138 and 193. All the compounds
that could be identified using full scan appear in this SIM chromatogram with
these fragments except benzaldehyde.
80


Figure 3.36 Total Ion Current for 13 Carbonyl Derivative Standard using
Ionspray under SIM (70,138,193,236,153) monitoring for MEK and
Butyraldehyde
Figure 3.37 Mass Spectra using Ionspray under SIM
81


Table 3.17 Relative Abundance of Ions using Ionspray under SIM
Ion Abundance
m/z Peak at 20.2 minutes Peak at 21.9 minutes
70 18 9
138 5 24
193 5 20
236 6 20
253 100 100
3.4.2 Chromatography using Atmospheric Pressure Chemical Ionization
Figure 3.38 shows the total ion current for the mixture of formaldehyde,
acetaldehyde, and acetone. All three compounds can be detected at this level of
concentration, but there is an obvious response difference for the three
compounds. Formaldehyde elutes at approximately 10.2 minutes with the most
abundant mass peak at m/z of 209, acetaldehyde at approximately 13.2 minutes
with the most abundant mass peak at m/z of 224, and acetone at approximately
17.5 minutes with the most abundant mass peak at m/z of 238._
82


Full Text

PAGE 1

IDENTIFICATION OF C4 CARBONYL COMPOUNDS IN AIR SAMPLES USINGLC/MS by Brian Graham B.S., Wake Forest University, 1994 A thesis submitted to the University of Colorado at Denver in partial fulfilhnent of the requirements for the degree of Master of Science Chemistry 2000

PAGE 2

This thesis for the Master of Science degree by Brian Graham has been approved by

PAGE 3

Graham, Brian (M.S., Chemistry) Identification of Carbonyl Compounds in Air Samples using LC/1.-IS. Thesis directed by Larry Anderson ABSTRA.CT Mass spectrometry is evaluated as an alternative and complementary form of detection for carbonyl compounds in air samples. Specific emphasis is placed on C4 carbonyl compounds, which can be difficult to identify with the cmTent method ofHPLC separation followed by UV detection. Two types of atmospheric pressure ionization techniques are evaluated. These include ionspray and atmospheric pressure chemical ionization. Standard mixtures of c, carbonyl compounds include crotonaldehyde, methacrolein, methyl ethyl ketone and butyraldehyde. These compounds are evaluated in terms ofhow well the mass spectrometer can distinguish between isomers at low levels of detection. Atmospheric pressure chemical ionization can be used to identify the C4 isomers at levels as low as 50 parts per trillion volume while use of ionspray ionization allows detection as low as 500 parts per trillion volume. Air samples taken from the northern Denver metro area confirm the presence ofbutyraldehyde and methyl ethyl ketone when either ionization technique is used. Methacrolein and crotonaldehyde however, could not be detected in the Northern Denver samples, but both were detected in samples from downtown Denver. lll

PAGE 4

This abstract accurately represents the content of the candidate's thesis. I recommend its publication. lV

PAGE 5

DEDICATION To Mom and Dad

PAGE 6

ACKl'J"OWLEDGMENT I am indebted to my advisor, Dr. Larry Anderson, for his guidance and patience with me during the course of this work, and also for his advice and support during the entire course of my academic pursuits. I would also like to thank Dr. Ed Furlong and Jeff for their advice, support and most importantly use of their facilities at the United States Geological Survey. I would also like to thank Phil Anderson for his suggestions and finally Gene Bouche for collecting samples.

PAGE 7

CONTENTS Figures .................................................................................... x Tables ..................................................................................... xv1 Chapters 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Purpose of Study........ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Carbonyl Compounds............................................................. 2 1.3 Health Effects of Carbonyl Compounds........................................ 4 1.4 Sources of Carbonyl Compounds................................................ 5 1.4.1 Primary Sourcess................................................................. 5 1.4.2 Secondary Sources............................................................... 6 1.5 Measurement Techniques ofCarbonyl Compounds ........................... 10 l.S.l"DNPH coated Cartridge Technique............................................ 11 1.5.2 High Performance Liquid Chromatography.................................. 12 1.5.3 Ultraviolet Detection............................................................. 13 1.6 Mass Spectrometry .................................................................. 13 1.6.1 History ofLC/MS ............................................. .-.................. 14 1.6.2 Advantages of API Interfaces ........................... .. 15 1.6.3 API Ionization Techniques ...................................................... 16 1.6.4 Interfaces with Mass Spectrometer ............................................. 20 1.6.5 Mass Analyzers ................................................................... 20 1.7 Previous Studies ..................................................................... 21 2. Experimental. .......................................................................... 25 Vll

PAGE 8

2.1 Standards and Samples ............................................................. 25 2.1.1 Standards and Preparation....................................................... 25 2.1.2 Samples ............................................................................. 27 2.2 Instrumentation and Accessories .................................................. 27 2.3 Analytical Techniques .............................................................. 28 2.3.1 Purity ofC4 Standards ............................................................ 28 2.3.2 Flow Injection Analysis .......................................................... 29 2.3.3 Precision ............................................................................ 31 2.3.4 Chromatography of Standard Mixtures........................................ 32 2.3.5 Determination of Detection Limit .............................................. 33 2.3.6 Analysis of Samples.............................................................. 33 2.3.7 Further Exploration ............................................................... 34 3. Results and Discussion ............................................................... 35 3.1 Standard Concentration and Purity ................................................ 35 3.2 Flow injection Analysis .............................................. .............. 39 3.2.1 Ionspray Ionization ............................................................... 39 3 .2.2 Atmospheric Pressure Chemical Ionization . . . . . . . . . . . . . . . . . 46 3.3 Precision .............................................................................. 53 3.3.1 Ionspray Ionization ............................................................... 53 3.3.2 Atmospheric Pressure Chemical Ionization .................................... 61 3.4 Chromatography of Standard Mixtues ............................................ 69 3.4.1 Chromatography using Ionspray ................................................ 69 .... 3.4.2 Chromatography using Atmospheric Pressure Chemical Ionization ....... 82 .. 3.5 Limit of Detection .................................................................. 95 3.5.1 Ionspray ............................................................................ 95 3.5.2 Atmospheric Pressure Chemical Ionization ................. : .................. 105 viii

PAGE 9

3.6 Sample Analysis ..................................................................... 117 3.6.1 lonspray ............................................................................. 117 3.6.2 Atmospheric Pressure Chemical Ionization ................................... 120 3.6.3 Quantitative Analysis of Sample using APCI. ................................ 131 3.7 Further Exploration .................................................................. 137 3.7.1 Analysis of Acrolein .............................................................. 137 3.7.2 Analysis ofMethyl Vinyl Ketone ............................................... 138 4. Conclusions ............................................................................ 140 4.1 Future Considerations ............................................................... 141 Appendix A-Calculations ........................................................................... 142 References ................................................................................. 146 IX

PAGE 10

FIGURES Figure 1.1 Electrospray Ionization. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 17 1.2 Atmospheric Pressure Chemical Ionization. . . . . . . . . . . . . . . . . . . 19 1.3 Structure Elucidation Scheme for Carbonyl Compounds using APCI in the Negative Mode.......................................................... 23 3.1 UV Chromatogram ofMethacrolein Derivative Standard . . . . . . . .. . 36 3.2 UV Chromatogram ofCrotonaldehyde Derivative Standard................. 36 3.3 UV Chromatogram ofMethyl Ethyl Ketone Derivative Standard.......... 37 3.4 UV Chromatogram ofButyraldehyde Derivative Standard . . . . . .. . 37 3.5 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 50V.. .. . . ...... 40 3.6 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 60V.. .. .. ... .. 41 3.7 C 4 Carbonyl Mass Spectra using Ionspray, Fragmentor 70V.............. 42 3.8 C 4 Carbonyl Mass Spectra using Ionspray, Fragmentor 80V..... .. . . . -43 3.9 C 4 Carbonyl Mass Spectra using Ionspray, Fragmentor 90V. ........... .. 44 3.10 C 4 Carbonyl Mass Spectra using Ionspray, Fragmentor 100V........... 45 3.11 C 4 Carbonyl Mass Spectra using APCI, Fragmentor 70V..... ..... ........ 47 3.12 C 4 Carbonyl Mass Spectra using APCI, Fragementor 80V................ 48 3.13 C 4 Carbonyl Mass Spectra using APCI, Fragementor 90V................. 49 3.14 C4 Carbonyl Mass Spectra using APCI, Fragementor 100V............... 50 3.15 C 4 Carbonyl Mass Spectra using APCI, Fragementor 11 OV........ . . .. 51 3.16 C 4 Carbonyl Mass Spectra using APCI, Fragementor 120. .... ...... ... .. 52 __ 3.17 Proposed Fragment Mechanisms for Crotonaldehyde using Ionspray.. .. 57 3.18 Proposed Fragment Mechanisms for Methacrolein using Ionspray....... 58 3.19 Proposed Fragment Mechanisms for MEK using Ionspray............. 59 X

PAGE 11

Figure 3.20 Proposed Fragment Mechanisms for Butyraldehyde using Ionspray... .... 60 3.21 Proposed Fragment Mechanisms for Crotonaldehyde using APCI. ..... ... 65 3.22 Proposed Fragment Mechanisms for Methacrolein using APCI..... ... ... 66 3.23 Proposed Fragment Mechanisms for MEK using APCI................. .... 67 3.24 Proposed Fragment Mechanisms for Butyraldehyde using APCI.. .. .. . 68 3.25 Total Ion Chromatogram of Mixture of Three Carbonyls using Ionspray ............................................................................ 69 3.26 Total Ion Current of the 4 Carbonyl Derivative Standard using Ionspray . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . 70 3.27 Mass Spectra using Ionspray under Full Scan ................................. 71 3.28 Total Ion Current for 4 Carbonyl Derivative Standard and blank using Ionspray under SIM (68, 149, 251) monitoring for Crotonaldehyde and Methacrolein..................................................................... ... 72 3.29 Mass Spectra using Ionspray under SIM ....................................... 73 3.30 Total Ion Current for 4 Carbonyl Derivative Standard using Ionspray under SIM (70, 138, 193, 236, 153) monitoring for MEK and Butyraldehyde.. ... .. ...... .. .. .. .. .. .. ... .. ...... .. .. .. .. .. ... ... ..... 74 3.31 Mass Spectra using Ionspray under SIM.. ... ...... ... ... .. .. ................ 75 3.32 Total Ion Current of the 13 Carbonyl Derivative Standard -----------using Ionspray under Full Scan ........................................................ 76 3.33 Mass Spectra using lonspray under Full Scan ................................. 77 3.34 Total Ion Current for 13 Carbonyl Derivative Standard using Ionspray under SIM (68, 149, 251) monitoring for Crotonaldehyde and Methacrolein.......................... . . . . . . . . . . . . . . . . . . . . ... . 79 XI

PAGE 12

Figure 3.35 Mass Spectra using Ionspray under SIM....................................... 79 3.36 Total Ion Current for 13 Carbonyl Derivative Standard using Ionspray under SIM (70, 138, 193, 236, 153) monitoring for l\1EK and Butyraldehyde...................................................................... 81 3.37 Mass Spectra using Ionspray under SIM.... ....... ... ............... .. ....... 82 3.38 Total Ion Chromatogram of Mixture of Three Carbonyls using APCI. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . 83 3.39 Total Ion Current of the 4 Carbonyl Derivative Standard using APCI under Full Scan................................................................... 84 3.40 Mass Spectra using APCI under Full Scan.................................... 84 3.41 Total Ion Current for 4 Carbonyl Derivative Standard using APCiunder SIM (79, 172, 202, 249) monitoring for Crotonaldehyde and Methacrolein........................................... ............. ................ 85 3.42 Mass Spectra using APCI under SIM......................................... 86 3.43 Total Ion Current for 4 Carbonyl Derivative Standard using APCI under SIM (163, 221, 251) monitoring for l\1EK and Butyraldehyde... 87 3.44 Mass Spectra using APCI under SIM.......................................... 87 3.45 Total Ion Current of the 13 Carbonyl Derivative Standard using APCI under Full Scan .................... ... ....... ... . . . ......... ............ .. 89 3.46 Mass Spectra using APCI under Full Scan.................................... 89 3.47 Total Ion Current for 13 Carbonyl Derivative Standard using APCI under SIM (79, 172, 202, 249) monitoring for Crotonaldehyde and Methacrolein.... ... .... ............ .. .... ... ..... ...................... .. ....... .. .. 91 3.48 Mass Spectra using APCI under SIM .......................................... 91 Xll

PAGE 13

Figure 3.49 Total Ion Current for 13 Carbonyl Derivative Standard using APCI under Sllvl (76, 221, 251) monitoring for 1\tiEK and Butyraldehyde ..... 93 3.50 Mass Spectra using APCI under SIM................................. 93 3.51 Total Ion Current for a 1/10 Solution ofthe 4 Carbonyl Mixture using SIM at 68, 149, 251 with Ionspray monitoring for Crotonaldehyde and Methacrolein..................................... . . . . . . . . . . . . . . . . . 96 3.52 Mass Spectra of 1110 Solution with Ionspray . . . . . . . . . . ............. 96 3.53 Total Ion Current for a 1110 Solution of the 4 Carbonyl Mixture using SIM at 70, 138, 193, 236, 253 with Ionspray monitoring for 1\tiEK and Butyraldehyde...................................................................... 97 3.54 Mass Spectra of 1/10 Solution with Ionspray.. ...... .. ..... .... .. ... .. .. . 98 3.55 Total Ion Current for a 11100 Solution ofthe 4 Carbonyl Mixture using SIM at 68, 149, 251 with Ionspray monitoring for Crotonaldehyde and Methacrolein.................................................... 100 3.56 Mass Spectra of 1/100 Solution with Ionspray.. .. ... .. ... .. ...... ..... ... 100 3.57 Total Ion Current for a 1/100 Solution ofthe 4 Carbonyl Mixture using SIM at 70, 138, 193, 236, 253 with Ionspray monitoring for MEK and Butyraldehyde................................................................ 101 3.58 Mass Spectra of 11100 Solution with Ionspray .. .... -.-. "" .-.-.-.. .. ,.. 102--------3.59 Total Ion Current for a 1/1000 Solution ofthe 4 Carbonyl Mixture using SIM at 68, 149, 251 with Ionspray.. .. .. .... .. ... .. ...... ... .. . 104 3.60 Total Ion Current for a 1/1000 Solution ofthe 4 Carbonyl Mixture using SIM at 70, 138, 193, 236, 253 with Ionspray...................... 104 xiii

PAGE 14

Figure 3.61 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein.......................... . . . . . . . . . . . . . 106 3.62 Mass Spectra of 1110 Solution with APCI. ... .. ... ..... .. .. ... ... .. .. ... 106 3.63 Total Ion Current for a 1110 Solution ofthe 4 Carbonyl Mixture Using SIM at 76, 221, 251 with APCI monitoring for MEK and Butyraldehyde.......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.64 Mass Spectra of 1110 Solution with APCI.. .. .. ... .. ... .. .. .. .. ... .. . 108 3.65 Total Ion Current for a 11100 Solution ofthe 4 Carbonyl Mixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein......................................................................... 110 3.66 Mass Spectra of 11100 Solution with APCI........................... .. . . .. 110 3.67 Total Ion Current for a 1/100 Solution ofthe 4 Carbonyl Mixture using SIM at 76,221,251 with APCI monitoring for MEK and Butyraldehyde............................................................ ... . . . . . . 111 3.68 Mass Spectra of Ill 00 Solution with APCI............................. ... . 112 3.69 Total Ion Current for a 111000 Solution ofthe 4 Carbonyl Mixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein .......................................... -.-.--.--,-,,,-;.-. ,.,., .-,,-.-.-,........... 114 3. 70 Mass Spectra of 1/1000 Solution with APCI...... ... . . . . . . . . . . .. . 114 3.71 Total Ion Current for a 1/1000 Solution ofthe 4 Carbonyl Mixture using SIM at 76, 221, 251 with APCI monitoring for MEK and Butyraldehyde.... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 3. 72 Mass Spectra of 1/1 000 Solution with APCI.................................. 116 3.73 Chromatogram of a Sample using Ionspray under Full Scan ................ 117 XIV

PAGE 15

Figure 3.74 Chromatograms of Samples 1-4 using Ionspray monitoring for Methacrolein and Crotonaldehyde nnder SIM (68, 149, 251) ..................... 118 3.75 Chromatograms for Samples 1-4 using Ionspray monitoring for MEK and Butyraldehyde under SIM (70, 138, 193, 236) .......................... 119 3.76 Chromatogram of a Sample using APCI under Full Scan .................... 120 3.77 Chromatograms for Samples 1-4 using APCI monitoring for Methacrolein and Crotonaldehyde under SIM (79,172, 202, 249).. ..... .................. .. ... . 122 3. 78 Chromatograms for Sample 1 -4 using APCI monitoring for MEK and Butyraldehyde nnder SIM (163, 221, 251).. .. .. .. .. .... .. .. ..... .. 124 3. 79 Chromatograms for Downtown Denver Air Samples monitoring for Crotonaldehyde and Methacrolein.,......................................... .. . . 125 3.80 Chromatograms for Downtown Denver Air Samples monitoring for MEK and Butyraldehyde ........................................................... 128 3.81 Calibration Curves for Crotonaldehyde and Methacrolein ................ ... -132 ---------3.82 Calibration Curves for MEK and Butyraldehyde .............................. 133 3.83 Diurnal Variation for C4 Componnds from Downtown Denver............ 136 3.84 Chromatogram and Mass Spectrum of Acrolein using Full Scan Conditions ................................................................................. 137 3.85 Chromatogram and Mass Spectrum ofMVK using APCI-under----------------. ------Full Scan .................................................................................. 138 3.86 Chromatogram and Mass Spectrum ofMVK using APCI under SIM Mode (79, 172, 202, 249). ... ......... .... .. .. ... .... ..... .... ........ .......... 139 XV

PAGE 16

TABLES Table 1.1 Four Carbon Aldehydes and Ketones...................................................... 3 2.1 Molecular Weights of Carbonyl Derivatives.................................. ...... 25 2.2 Ionspray MSD Parameters................................................. . . . . . .. ........ 30 2.3 APCI MSD Parameters.......................................................... . . . ......... 31 3.1 Concentrations of the C 4 Carbonyl Derivative Solutions...................... 35 3.2 Concentrations of the C4 Carbonyl Derivative Mixture Solution........... 38 3.3 Precision for C4 Carbonyl Compounds using Ionspray in Total Ion Current................................................................ . ... .. . ... . . .. . . .. .. ... .. .. 54 3.4 Major Fragments for Methacrolein using Ionspray..... .... ............ .... ...... 55 3.5 Major Fragments for Crotonaldehyde using Ionspray .......................... 55 3.6 Major Fragments for Methyl Ethyl Ketone usinglonspray.... .... ... ........ 56 3.7 Major Fragments for Butyraldehyde using Ionspray ............................. 56 3.8 Precision for C4 Carbonyl Compounds using APCI in Total Ion Current ...................................... .'...................................................................... 62 3.9 Major Fragments for Methacrolein using APCI................ ... .. .... .......... 63 3.10 Major Fragments for Crotonaldehyde using APCI............. ................ 63 3.11 Major Fragments for Methyl Ethyl Ketone using APCI. .... ....... .......... 64 3.12 Major Fragments for Butyraldehyde using APCI................................ 64 3.13 Relative Abundance oflons using Ionspray under SIM.............. ......... 73 3.14 Relative Abundance oflons using Ionspray under SIM............. ..... ..... 75 3.15 Retention Times and Major Fragments of Compounds in Standard Mixture of 13 Derivatives............................................................. 78 3.16 Relative Abundance oflons using lonspray under SIM................. ...... 80 3.17 Relative Abundance oflons using Ionspray under SIM................. ..... 82 xvi

PAGE 17

Table 3.18 Relative Abundance oflons using APCI under SIM......... ........ ... ...... 84 3.19 Relative Abundance oflons using APCI under SllvL........ ................. 88 3.20 Retention Times and Major Fragments of Compounds in Standard Mixture of 13 Derivatives............................................................... 90 3.21 Relative Abundance oflons using APCI under SIM...... . .. . . . . . . . . 92 3.22 Relative Abundance oflons using APCI under SIM....................... .... 94 3.23 Relative Abundances of Selected Ions using Ionspray of 1110 Soluton ............................................................................................................. 97 3.24 Relative Abundances of Selected Ions using Ionspray of 1/10 Solution................................................................................................... . . . .... 98 3.25 Relative Abundances of Selected Ions using Ionspray of 11100 Soluton................................................................................................ .. ......... 101 3.26 Relative Abundances of Selected Ions using Ionspray of 11100 Solution............................................................................................................ 102 3.27 Relative Abundances of Selected Ions using APCI of 1/10 Soluton.. .. .. .. .. .... .... ..... .. ... .... .... .. .. .. .. ..... ......... ... .. .. .. ... .. .. .... ...... .. .. . . . . . ... .. .. 107 3.28 Relative Abundances of Selected Ions using APCI of 1/10 Solution............................................................................................................ 108 3.29 Relative Abundances of Selected Ions using APCI of 1/100 Soluton... .. ................................................................... .......................... ........... 111 3.30 Relative Abundances of Selected Ions using APCI of 1/100 Solution............................................................................................................. 112 3.31 Relative Abundances of Selected Ions using APCI of Ill 000 Soluton...................... .. . . . . . . . . . . . . . . .. .. .. . . . . . . . . . . . . . . . . . . . . . . . 115 XVll

PAGE 18

Table 3.32 Relative Abundances of Selected Ions using APCI of 1/1000 Solution........................................................................................................... 116 3.33 Relative Abundances of Selected Ions using Ionspray of Samples 1-4 ......................................................................... .......................................... 120 3.34 Relative Abundances of Selected Ions using APCI of Samples 2 and 4 ................................................................................................................. 123 3.35 Relative Abundances of Selected Ions using APCI of Samples 1-4 .................................................................................................................... 125 3.36 Relative Abundances of Selected Ions in Samples 1-7 monitoring .. for Crotonaldehyde and Methacrolein ..................................................... ... :: 128 3.37 Relative Abundances of Selected Ions in Samples 1-7 monitoring for MEK and Butyraldehyde.............................................................................. 131 3.38 Integrated Areas of the Ion current for the C4 compounds and their Concentrations................................................................................... . . . . . . . 131 3.39 Integrated Areas of the Ion Current and Concentrations for the C4 Compounds in the Samples collected from Northern Denver.................. . ..... 134 3.40 Integrated Areas of the Ion Current and Concentrations for the C4 Compounds in the Samples collected from Downtown Denver....... . . . . ... .. .. 135 3.41 Relative Abundances of Selected Ions using APCI for MYK. ............................................................................................................... 139 xviii

PAGE 19

1. Introduction 1.1 Purpose of the Study Over the past two decades carbonyl compounds in the earth's atmosphere have been increasingly attracting the attention of scientists. This is due in part to the role they play in the chemistry of the atmosphere as well as the fact that some these compounds are known hazardous pollutants. Carbonyl compounds are either directly emitted (primary sources) or are formed in the atmosphere (secondary sources). Once in the atmosphere these compounds can undergo photolysis contributing a significant source of free radicals, which are responsible for the oxidation of hydrocarbons. They also are the precursors of oxidants including ozone, peroxyacyl nitrates, and peroxycarbocylic acids (1). Because of the environmental importance carbonyl compounds represent it is important to have an analytical technique that is both selective and sensitive in determining the identity and quantity of these compounds in the atmosphere. Currently several techniques exist that measure the gas-phase concentration of carbonyl compounds. The most common method in use today makes use of high performance liquid chromatography (HPLC) with ultraviolet (UV) detection of the 2,4-dinitrophenylhydrazone derivative (2). This method performs quite well with lower molecular weight carbonyls, for example those containing one to three carbon atoms since these compounds are easy to identify and concentrations are typically larger than those of a higher molecular weight. Specifically, carbonyls containing four carbon atoms (C4 ) are difficult to identify using the aforementioned method ofUV detection. An exploration into mass spectrometry 1

PAGE 20

as an alternative method of detection for the 2,4-dinitrophenylhydrazone derivative with HPLC separation is investigated as a means of identifYing carbonyl compounds with four carbon atoms. Several ionization techniques are examined as to determine the most effective for identification of these compounds. 1.2 Carbonyl Compounds Carbonyl compounds are identified as either an aldehyde or a ketone. The central feature of aldehydes and ketones is the carbonyl group. The carbonyl group contains a carbon atom that has a double bond to oxygen as follows: "-c=o / The carbonyl group in aldehydes is bonded to at least one hydrogen atom and either one or no carbon atoms. The following is a general formula for an aldehyde where R represents an alkyl group or a hydrogen: 0 II _,C, R H The carbonyl group in ketones is bonded to two carbon atoms. The following is a general formula for a ketone where R' represents an alkyl group that is the same or different from R: 0 II _,c, R R' Table 1.1 lists examples of important four carbon carbonyls used in as solvents, intermediates, fuel additives, flavorings, etc. 2

PAGE 21

Table 1.1 Four Carbon Aldehydes and Ketones I I i I I I 0 II 0 II 0 II 0 II 0 II YCH 2-butenal (Crotonaldehyde) 2-methylpropenal (YI ethacrolein) 3-buten-2-one (::Vlethyl Vinyl Ketone) 2-butanone (l\tlethvl Ethvl Ketone) I . I : Butanal I I : (Butyraldehyde) I I i 2-methylpropanal -I (Isobutyraldehyde) I .., .) -----..>

PAGE 22

... 4 ....... .. ... --1.3 Health Effects of Carbonyl Compounds Aldehydes and ketones have been implicated as mutagens and possible .carcinogens (2). Most of these compounds have been identified as mucus membrane irritants, and can cause pulmonary, skin, eye, and central nervous system irritation. The Environmental Protection Agency (EPA) has classified many of these compounds as toxic and hazardous. For example, formaldehyde (HCHO) contributes to eye, nose, and throat irritation. It has also been shown to cause bronchial asthma-like symptoms with some reports of asthma attacks, as well as allergic dermatitis (3). As a result of the toxicities of carbonyl compounds the Occupational Safety and Health Administration (OSHA) and the National Institute for Occupational Safety and Health (NIOSH) have recommended Permissible Exposure Limits (PEL), as well as Short Term Exposure Limits (STEL). Information concerning these limits can be found at their respective government websites (www.osha.gov or www.cdc.gov). These limits are set as maximum exposure limits as time weighted averages. The PEL is usually eight hours and the STEL is fifteen minutes. Most limits usually range from a parts per billion (ppb) to parts per million (ppm) exposure level for a given time period depending on the compound. Since aldehydes are very reactive the secondary compounds formed from aldehydes, especially peroxyacylnitrates are much more dangerous to the environment. The problem of the ecotoxcity of peroxyacylnitrates cannot be separated from that of aldehydes ( 4 ). 4

PAGE 23

1.4 Sources of Carbonyl Compounds As stated previously, aldehydes and ketones originate from primary sources that are emitted directly into the atmosphere or from secondary sources in which these compounds are formed in the atmosphere by chemical transformations. Emission of primary compounds can be classified into two different types. The first is by anthropogenic processes, which include industrial, domestic, or agricultural emissions. The second is biogenic activity which is a result of natural processes such as plants and mammals ( 4). Secondary sources can also originate from anthropogenic or biogenic processes. 1.4.1 Primary Sources The main sources of anthropogenic emissions stem from motor vehicle traffic, trash incineration, industrial and domestic heating, and exploitation of fossil fuels for energy production (4). Motor vehicle emissions have attracted the most attention since this seems to be an easy source to control. The government has required the use of oxygenated fuels in motor vehicles in many areas across the country in hopes of reducing air pollution by controlling carbon monoxide levels. It has been shown that the use of these fuels which include methanol, ethanol, and methyl tertiary butyl ether (MTBE) blended fuels, has increased the emissions of carbonyl compounds (5). Specifically formaldehyde-andacetaldehyde---------------------------(CH3CHO) concentrations have increased. Other sources of emission include many industries that emit a wide range of carbonyls. These industries include refining and petrochemical, coal, plastics, paints, and varnish, sewage treatment, and facilities where carbonyl compounds are manufactured (4). The most important natural source of aldehydes and ketones is from the emissions of animal 5

PAGE 24

excretions and forest fires (4). As it should be noted, humans do have an impact on these sources. Many of the C4 carbonyl compounds enter the environment via automobile exhaust, animal waste and biomass combustion (6). Crotonaldehyde, methyl ethyl ketone, butyraldehyde, and methacrolein have all been detected in car exhaust, biomass combustion and tobacco smoke (6). Methyl ethyl ketone and butyraldehyde have also been detected in animal waste emissions (6). 1.4.2 Secondary Sources In unpolluted environments most carbonyl compounds come from the photo oxidation of organic compounds which have biogenic origin. The majority of organic compounds that are emitted into the atmosphere are likely to form carbonyl compounds (4). In polluted environments, usually urban settings, the situation is more complex, but the general mechanisms are the same and today are well understood. The chemistry involving atmospheric hydrocarbons is initiated by highly reactive OH radicals, N03 radicals, or ozone (03). These oxidizing species exist as follows: Stratospheric ozone can be transported downwards into the troposphere. It can also be formed from the photolysis of nitrogen dioxide (4): N02 + hv oeP) oeP) + 02 + M 03 + M OH radicals are formed from the photolytic destruction of ozone (4): 03 + hv 02 + oen) 0(1D) + H20 20H 6

PAGE 25

Carbonyl compounds can also form the OH radical by photolysis. For example, formaldehyde can be transformed by the following sequence (4): HCOH + H + HCO H +02 H02 HCO +02 H02 +CO H02 OH +N02 The above processes due to their photolytic nature drive the daytime chemistry. On the other hand the N03 radical drives the nighttime chemistry since it photolyzes very quickly. The largest source of the N03 radical is by the following process (4): 1.4.2.1 Reactions with OH During the daytime the majority of hydrocarbons will react with the OH radical in the following manner (4): RH +OH R +HzO The organic radical then proceeds through the following steps to form a carbonyl compound (4): R +02 ROz R02 RO +N02 7

PAGE 26

RO + 02 carbonyl compound+ H02 A good example of this sequence is the alkene isoprene (CH2=C(CH 3 )CH= CH2). Isoprene and monoterpene hydrocarbons constitute the major portion of the nonmethane organic compounds that are emitted from vegetation (7). Laboratory studies have shown isoprene to be highly reactive under conditions characteristic of the lower troposphere and is a precursor to C4 carbonyl compounds. Specifically these compounds are methyl vinyl ketone(MVK) and methacrolein. These products can be arrived at by the following schemes (7): For methyl vinyl ketone: CHz=CHC(CH3)= CHz + OH CHz=CHC(CH3)CHzOH CHz=CHC(CH3)CHzOH + Oz CHz=CHC(OO)(CH3)CHzOH CHz=CHC(OO)(CH3)CHzOH +NO CHz=CHC(O)(CH3)CHzOH + NOz CHz=CHC(O)(CH3)CHzOH CH3COCH=CHz(MVK) + CHzOH CHzOH + Oz HCHO + HOz For methacrolein: CHz=CHC(CH3)= CHz + OH HOCHzCHC(CH3)=CHz HOCHzCHC(CH3)=CHz + Oz HOCHzCH(OO)C(CH3)=CHz HOCHzCH(OO)C(CH3)=CHz +NO HOCHzCH(O)C(CH3)=CHz + NOz HOCHzCH(O)C(CH3)=CHz CHzOH + CHz=C(CH3)CHO(rfietliacrolein) CHzOH + Oz HCHO + HOz 8

PAGE 27

1.4.2.2 Reactions with Ozone The reaction of hydrocarbons with ozone is slower than with the OH radical, but this reaction becomes significant for alkenes. The gas-phase reactions of ozone ) with alkenes proceeds by initial addition to form a primary ozonide, which in tum rapidly decomposes to form two sets of carbonyl compounds plus biradical products (8). This is shown below: The initially energy rich biradicals can decompose eventually leading to the OH radical or be conditionally stabilized (8). A good example of this sequence is once again the reaction of the alkene isoprene. Isoprene reacts with ozone to form methyl vinyl ketone, methacrolein, formaldehyde and biradicals. These biradicals eventually.lead to production ofmore formaldehyde. The mechanism is as follows: Addition to the firstdouble bond: _,.o, 0 0 I I C -C(CH l )CH =CH H 2 9

PAGE 28

(i) CH3COCH=CHz(MVK) + CHzOO (ii) HCOH + CH3C(OO)CH=CHz Addition to the second double bond: (i) CHz=C(CH3)CHO(methacrolein) + CHzOO (ii) HC-OH + CHz=C(CH3)CHOO 1.4.2.3 Reactions with N03 The mechanism for the reaction of the N03 radical with hydrocarbons is not fully understood. However, it has been suggested that the reaction ofN03 with alkenes occurs as an addition followed by the decomposition of nitrooxy-alkoxy radicals and the remaining products identified include a number of nitrooxy-carbonyls and nitrooxy-alcohols (3). 1.5 Measurement Techniques of Carbonyl Compounds As noted previously several techniques are presently usedto measure -gas-phase --------.. --. carbonyl compounds. The most widely used is the 2,4-dinitrophenylhydrazine (DNPH) coated cartridge technique which is used by the EPA in TO-ll (1 ). This method, which uses HPLC for separation and UV radiation for detection, is discussed here. 10

PAGE 29

1.5.1 DNPH Coated Cartridge Technique This teclmique involves coating a solid sorbent with acidified DNPH reagent and packing the sorbent into a cartridge. The sorbents typically used are silica gel and a Crs silica gel. Silica gel is a polymer of silica and oxygen with surface hydroxyl groups. C18 silica gel a similar surface, but is coated with long hydrocarbon chains. These chains are chemically bonded to the silica support through siloxane groups (Si-0-R) (1). Air is sampled through the cartridge for a given period of time. Any carbonyl compound in the air will be converted to a stable hydrazone remaining on the sorbent. The process is described by the following reaction: + HzO (Carbonyl) (DNPH) (DNP Hydrazone) The hydrazine group (-HN-NHz) reacts with the carbonyl compounds to form hydrazones. The acid-catalyzed reaction occurs when the lone pair of electrons on the NH2 group attacks the carbonyl carbon of an aldehyde or ketone to form a hydrazone. Removal ofthe hydrazones for subsequent analysis is accomplished by extraction with acetonitrile. 11 .-. .-:.. --; :-

PAGE 30

1.5.2 High Performance Liquid Chromatography Liquid chromatography is a physical method of separating compounds that makes use of a stationary phase packed in a column and a liquid mobile phase. Compounds are separated based on their different affinities for the two phases. Modem systems make use of high pressure to force solvent through packed columns that contain very fine particles, usually in the range of 3 Jlm to 10 Jlm, to give high resolution separations. There are several liquid chromatographic separation modes. These include: adsorption, ion-exchange, ion pair, size exclusion, and partition chromatography. Partition chromatography can be further divided into normal or reversed phase. The most widely employed mode. used today is reversed phase chromatography. This is also the method employed when separating the DNP hydrazones, and will be the focus of the remainder of the discussion. Reversed phase chromatography can simply be defined as making use of a polar mobile phase and non-polar stationary phase. Polar solvents that are typically used include: water, acetonitrile, methanol, and tetrahydrofuran (THF). Combinations of any of the three organic solvents with water typically provide a sufficient range of dipolar and hydrogen bonding interactions with solutes to separate a vast number of compounds (9). The stationary phase is typically packed into a steel column that is anywhere from 5 to 30 centimeters in length with an internal dismeter of 4.6 mm. The stationary phase consists of microporous particles of silica that are bonded with a non-polar hydrocarbon such as the octadecyl (Cts) or octyl (Cs) chain. The basis for elution of compounds is the partitioning of the solute between mobile phase and stationary phase. Eluent strength is a measure of how fast the solvent or solvent mixture will elute the analytes of interest. The eluent strength, in reversed phase chromatography 12

PAGE 31

decreases as the solvent becomes more polar. Elution can be accomplished by either isocratic (constant solvent mixture) or gradient (increasing amounts of a second solvent are added over time) method. For a mixture ofDNP hydrazones a gradient method of water and increasing acetonitrile is employed with separation on a C1s column. 1.5.3 Ultraviolet Detection The UV detector is the most common HPLC detector in use today due to the fact that most solutes absorb ultraviolet light and the detectors provide high sensitivity. The flow from the HPLC is directed into a flow cell which is then irradiated with the appropriate energy. Simple aldehydes and ketones only show weak absorption bands in the UV region due to the n to 1t electronic transition of the carbonyl group. DNPH derivati.v:es usually exhibit maxima around 360 nm with much more intense absorption due to the higher molar absorptivity associated with the 1t to 1t electronic transition (1 0). These transitions occur due to the aromatic structure ofthe DNP Hydrazone where the DNP is the major chromophore. Once derivatized these compounds can be easily detected with a UV source. 1.6 Mass Spectrometry Mass spectrometry (MS) is a powerful tool for the determination of analytes both----------------qualitatively and quantitatively. The basic principle behind MS begins with ionization of the analyte, followed by acceleration of the ions by an electric field into an area where they are separated according to their mass. In the early years MS was mainly used as an on-line detector for gas chromatography since analytes needed to be in the gaseous state to be ionized. Thus, compounds that were 13

PAGE 32

volatile and not thermally labile consisted of the bulk of analytes.that were analyzed. Over the past twenty years however great advances have been made in the coupling ofliquid chromatography with mass spectrometry. This has made it possible to analyze compounds that are nonvolatile and/or thermally labile by MS. Today liquid chromatography coupled with mass spectrometry (LC/MS) ts a robust, routinely applicable and widespread analytical technique (11 ). 1.6.1 History of LC/MS Three major difficulties are met in interfacing the two powerful tools of liquid chromatography and mass spectrometry (11). The first is how to accommodate such a high flow rate, typically 1 ml/min of liquid from a conventional LC column with the high vacuum required of the MS. The second is the solvent composition, where non-volatile mobile phase additives usually find use in LC separation development. Finally how does one analyze non-volatile and/or thermally labile analytes? Only a handful of ionization techniques were originally available to the scientist. These included electron ionization, chemical ionization, and field desorption. Field desorption is not applicable to LC/MS. Electron ionization and chemical ionization required ions to be in the gaseous state. Thus, came the development of new techniques of ionization. Early interfaces proved to be too complex, lacked ruggedness, and had limited sensitivity. The advent of particle beam and thermospray overcame some of the initial shortcomings perinitting their use on a more routine basis (12). Although these ionization techniques became more common, they still lacked sensitivity and specificity. The particle beam interface is unable to ionize compounds that were not volatile and is not sensitive enough to detect analytes below the parts per billion level. Thermospray failed to provide sufficient fragmentation for compound confirmation which could have been overcome with tandem MS, but would have 14

PAGE 33

only increased the cost and complexity of the system. These techniques may still find use, but a new technique of ionization appears to have superseded them (12). This is atmospheric pressure ionization (API). At first it may seem impractical to couple an ionization technique conducted at atmospheric pressure with mass analysis carried out under a high vacuum. Combination of an API source with a mass analyzer requires a 107 to 108 fold pressure reduction along the ion passageway between the source and mass analyzer which can present a demanding task for the vacuum pumping system (13). This has been overcome with research yet despite the great-potential of API only recently has it found widespread use in analytical laboratories since manufacturers did not offer the source as an option. The remainder of the discussion will focus on API due to its extensive applications found in pharmaceutical, biochemistry and the environmental sciences. 1.6.2 Advantages of API Interfaces The combination of LC/MS is better suited with the ionization source operating at atmospheric pressure due to the several advantages it presents (12). First, the LC inlet typically operated at atmospheric pressure can be decoupled from the mass spectrometer which operates at a pressure of 10-6 torr or lower. This provides an opportunity to optimize operating conditions for both LC separation and MS detection. Second, solvents and contaminants that elute into the atmospheric chamber can be pumped away without degrading MS performance. Thus, a rugged method is obtained. Third, liquid droplets can be more efficiently desolvated at atmospheric pressure than at reduced pressures. Heat transfer is more efficient at higher pressures. Fourth, as ions are sampled from the atmosphere to the vacuum they undergo a free jet expansion. This causes adiabatic cooling and helps maintain the integrity oflabile compounds as well as 15

PAGE 34

noncovalent complexes for mass analysis. Adiabatic cooling can be considered a disadvantage since it can contribute to the formation of cluster ions. Cluster ions, however can be eliminated by focusing lenses plus the acceleration region prior to directing them into the mass analyzer. Finally, by controlling the electrostatic potentials in this expansion region, collision induced decomposition (CID) can be achieved. This can provide structural information on the compounds of interest. API Ionization Techniques Two ionization techniques have taken hold of the majority of the market due to the wide applicability of analytes that can be ionized. These include electrospray ionization (ES) and atmospheric pressure chemical ionization (APCI). These two ionization techniques are explored to determine how they characterize the DNPH derivatives. 1.6.3.1 Electrospray Ionization Electrospray ionization is im approach that does not require the sample to be evaporated before it is ionized. Thus, samples that are thermally labile and polar can be analyzed without a source of heat for evaporation or under other conditions that are detrimental to analyte stability. Typical samples include analytes that 'exist as ions in solution or can be ionized in solution. The ionization technique depends on the dispersion of a sample solution-into an-electrically-charged-aerosol---------(13). A potential difference between the spray capillary and the walls of the source is used to supply the necessary electric field to overcome cohesive forces holding the liquid together at the tip of the capillary to force ionization of solution. Figure 1.1 shows a typical setup for electrospray ionization. 16

PAGE 35

Figure 1.1 Electrospray Ionization 9 N2 ... Skimmer \. 760 l!J 10 Torr Pumping 3to5 kV The charged droplet consists of the solvent containing positive and negative ions with the predominant charge depending on the polarity of the induced potential (14). Before ions can be mass analyzed, the solvent must be removed. This is done by a counter flow of neutral, heated drying gas, typically nitrogen, that evaporates the solvent. The droplet decreases and forces the predom!nately like charges closer together at the surface of the droplet. When the : limit( coulomb repulsion equals that of the surface terision.ofthe.droplet}-----__ reached the droplet explodes producing daughter droplets that are subject to :furl:her evaporation. This process repeats itself and droplets with a high surfacecharged density are formed (15). Gas phase ion formation is the next process .. \Yhich has })een the subject of much debate. One theory predominates though and __ evaporation. In this model .. _____ .... the gas phase. When the electric field created by-the ions at the-surface ofthe --------17

PAGE 36

droplet exceeds the surface tension, bare analyte ions are emitted directly from the droplet (15). Ion evaporation is very efficient. Nebulization of the LC effluent results in nearly complete formation of charged droplets that can subsequently undergo ion evaporation (12). In thermospray only 1 in 104-105 droplets is charged under ion evaporation (12). Electrospray works better when a sample is dissolved in a suitable organic solvent such as methanol or acetonitrile with a low surface tension. It is more difficult to spray aqueous solutions which have a higher surface tension and lower vapor pressure. This opposes the separation of the droplets from the liquid front causing a less stable spray and bigger droplets (13). The limitation of electrospray is the low flow rates that are required for droplet formation. Flow rates typically range from 1-10 )lLimin. This is not always practical or rugged enough for solving separation problems. This has led to the development of pneumatically assisted nebulization, commonly called ionspray. This allows the use of higher flow rates and can handle mobile phases with higher water content. Ionspray is now the most frequently used electrospray technique. 1.6.3.2 Atmospheric Pressure Chemical Ionization APCI is a gas phase ionization process. For this reason the most suitable compounds are those that are somewhat volatile and thermally stable. Coupling of the LC to the APCI system involves the use of a heated nebulizer inlet. The heated nebulizer inlet consists of a concentric pneumatic nebulizer and a large diameter heated quartz tube (11 ). Vaporizatiop. of the LC effluent occurs as it is being swept through the heated tube. At this point the solvent and vapor are introduced into the APCI source. Once in the source a corona discharge from a needle with an applied voltage produces a current of 2 to 5 )lA. See Figure 1.2 for a typical APCI setup. 18

PAGE 37

Figure 1.2 Atmospheric Pressure Chemical Ionization Spray temperature sensor Pumping The corona discharge provides a source of ionizing electrons that produces ions from nearby gases. These ions readily undergo ion-molecule reactions with air and other gaseous molecules because of the short mean free path between molecules at atmospheric pressure (14). This produces reactive intermediates that yield ion-molecule reactions. The majority of the ion-molecule reactions in APCI involve acid-base chemistry and the ion chemistry can be very complex. Adduct and cluster ions of air and water can be formed. This formation process results froJ;TI the high reactivity of the. ions l:llld q(<;>_ther:_11t::.ar:Py g(l __ molecules (14). This can increase the background noise and reduce the signal-to noise ratio. Several approaches have proved successful in breaking up these clusters, but the most common is the use of a gas curtain that falls between the ion source and vacuum expansion. This method along with an applied electric field provide CID during the free jet expansion. APCI has shown efficiency that is 19

PAGE 38

I 03 -I 04 times greater than electron impact with the possibility of I 00% ionization efficiency of a trace amount of sample. 1.6.4 Interfaces to Mass Spectrometer Transportation from the atmospheric region to the vacuum of the mass spectrometer usually involves the use of a capillary, a series of skimmers, and several pumping stages which will help maintain the vacuum. Gas transport is based on the pressure difference at the ends of the capillary. Since both ionization methods discussed are soft ionization techniques, not capable of extensive fragmentation, the exit potential of the capillary can directly change the energy of the ions. At higher voltages, ions can experience many collisions with the air and nitrogen which cause an increase in the internal energy of the molecules. This internal energy can cause multiple bond cleavages which are necessary for structural determination. 1.6.5 Mass Analyzers Mass spectrometers can employ a wide variety of analyzers. It is ideal to have an analyzer that is capable of distinguishing between 1 atomic mass unit as well as allow enough ions through. Analyzers can be ofthe following types: a magnetic field, double focusing, quadrupole mass filter, ion trap, time of flight, and fourier transform-ion cyclotron resonance. The most common-analyze:ris the quadrupole----------. mass filter. This is because it is more compact, less expensive, and more rugged than the others ( 1 0). This will be the focus of the discussion since it will be the analyzer used in this study. The quadrupole instrument uses four parallel cylindrical rods that serve as electrodes. Ions enter from one enci and travel with a constant velocity in a direction parallel to the poles, which is the z direction. The 20

PAGE 39

ions acquire complex oscillations in the x andy directions by application of both a direct current voltage and a radiofrequency voltage to the poles (16). A stable oscillation occurs of a particular ion, allowing it to pass through without hitting the poles that is dependent on the mass to charge ratio. Only ions of a single mass to charge ratio will travel through the poles under a given set of frequencies. All the other ions will strike the poles, losing their charge therefore being lost. Mass scanning is carried out by varying the frequencies while keeping their ratios constant (16). Typically quadrupole instruments resolve ions that differ in mass by one atomic mass unit. 1. 7 Previous Studies Few studies have been found in the literature involving mass spectrometry detection of 2,4 dinitrophenylhydrazone derivatives of carbonyl compounds. What follows is a current summary of these studies. Early studies involved collecting HPLC fractions and subjecting these fractions to mass spectrometry. Since electron impact was the primary ionization technique of the day, this was employed with poor results. The carbonyl derivatives suffered extensive fragmentation making identification of carbonyls difficult. Chemical ionization was first investigated by Grosjean of the DNPH derivatives in 1983 with positive results (17). A variety of carbonyls were studied using positive chemical ionization The most abundant peak in each spectrum was the protonated MH+ ion. Little fragiTtentation occurred and conClusions showed chemical ionization was a useful identification tool. Olson and Swarin were the first to use an LC/MS system with a moving belt interface followed by chemical ionization for the carbonyl derivatives (18). They showed the negative ion mode with methane as the chemical ionizing agent is more sensitive for certain DNPH derivatives. They also showed how aldehydes and ketones can be distinguished based on the 21

PAGE 40

intensity of the molecular ion peak relative to fragments, with ketones being the more prominent. In 1995 Grosjean and Grosjean described use of a particle beam interface (19). Positive ion mode was employed with methane as the reagent gas. The mass spectra of the DNPH derivatives indicated little fragmentation. The protonated molecular ion peak was the most abundant peak for all compounds analyzed. These ionization techniques however are beginning to fade out due to the emergence of atmospheric pressure ionization methods which offer better detection limits (11). In 1998 Kolliker and co-workers (20) published a procedure for identification of carbonyls using atmospheric pressure chemical ionization mass spectrometry. They employed an ion trap mass spectrometer for structure elucidation of the carbonyl derivatives in ambient air samples. APCI was used in the negative ion mode since it was determined that electrospray in the positive and negative ion mode had insufficient detection limits as well as APCI in the positive ion mode. A structure elucidation scheme was developed by interpreting mass spectra from thirty reference carbonyl compounds. This scheme is displayed in Figure 1.3. The authors claim the following distinctions can be made: 1. identification of DNPH derivatives, 2. differentiation between aldehydes and ketones, 3. differentiation between a-saturated and unsaturated aldehydes, 4. differentiation between branched and straight chained aldehydes and ketones. It is the author's opinion that structures of coeluting DNPH carbonyls can be identified, however this is backed up with little evidence. 22

PAGE 41

Figure 1.3 Structure Elucidation Scheme for Carbonyl Compounds using APCI in the Negative Mode 1Mfl2 flf2 111 In 1999 Zurek and co-workers (21) developed a method for selective determination and quantification of formaldehyde and acetaldehyde in air samples HPLC-APCI-MS. This is accomplished by calibrating using stable isotope .:: . : The wanted to quantify higher molecular weight carbonyl derivatives, but were limited by the availability of the respective stable isotope labeled carbonyl compounds. The isotopically labeled standards were used as an internal standards to overcome changes in the MS conditions long . ..... _j)enods o:(_ . . . . 23

PAGE 42

Most recently Grosjean and co-workers developed a simple and sensitive LC/MS method for detection ofDNPH derivatives using atmospheric pressure chemical ionization in the negative mode on a benchtop system (22, 23). Electrospray and APCI were both evaluated in the positive and negative ion modes. APCI negative ion detection was found to provide more specific information and 1-2 orders of magnitude better sensitivity (23). The work involved an investigation into 78 carbonyl compound derivatives. Experimental conditions were optimized in a way that the molecular ion peak was the base peak for most compounds, revealing little fragmentation. This was accomplished using the system's flow injection analysis. Methacrolein, crotonaldehyde, methyl ethyl ketone and butyraldehyde were some of the C4 carbonyls investigated. The data showed separation of these compounds was difficult and mass spectrometer parameters were not optimized to generate distinct fragments. Extracted ion chromatograms were evaluated illustrating how the specificity of the technique can help with coelution. Resolution between methyl ethyl ketone and butyraldehyde however, remained less than optimal. The author's of this study also looked at the quantitative abilities of the technique and found high comparability to UV detection. Finally, the authors feel positive identification at levels as low as sub-ppb in ambient air can be obtained. Future work in Grosjean's laboratory will be guided in a direction of optimizing parameters to produce distinct fragments for specific compounds. 24

PAGE 43

2. Experimental Methods 2.1 Standards and Samples Standards ofDNPH derivatives ofC4 carbonyl compounds and mixtures of derivatized carbonyls will serve as the standards to be investigated by LC/MS. Samples taken from the surrounding Denver metro area will also be analyzed in an attempt to determine unidentified C 4 carbonyl compounds. 2.1.1 Standards and Preparation Pure standards were prepared for the following compounds: methyl ethyl ketone, butyraldehyde, crotonaldehyde, and methacrolein. All four of these compounds were derivatized with DNPH to form hydrazones. Table 2.1 contains the molecular weights of these compounds. Table 2.1 Molecular Weights of Carbonyl Derivatives Carbonyl Derivative Formula Molecular Weight (amu) Methacrolein CIO HIO N4 04 250 Crotonaldehyde CroHro N4 04 250 Methyl Ethyl Ketone Cro Hrz N4 04 252 Butyraldehyde Cro Hrz N4 04 252 25 --.. -.

PAGE 44

The DNPH derivatives were obtained from Radian Corporation as solid crystals in sealed separate vials. The stated chemical purity of each compound was 99% and each vial contained approximately 10 mg ofthe hydrazones. The contents of each vial are accurately weighed and transferred to separate 100 mL volumetric flasks. Acetonitrile is used as the solvent and added to the calibration mark in each flask. Each flask contains approximately a 100 ppm solution of the DNPH derivative. These standards serve as the solutions needed to optimize certain parameters for the mass spectrometer. Three different standard mixtures are used to gauge chromatographic conditions and confirm mass spectrometry results. The first is a mixture containing derivatized formaldehyde (MW 21 0), acetaldehyde (MW 225), and acetone (MW 238). This solution was prepared by the environmental sciences lab at UCD for their work involving quantification of these compounds in air samples and contains roughly 16 parts per billion volume of each compound. This solution will provide information on how LC/MS can identify these compounds. The second is a mixture of the four C4 carbonyl standard solutions discussed above. Using a volumetric pipet 10 mL is transferred from each ofthe 100 mL volumetric flasks to a single 100 mL volumetric flask. This solution is then diluted to the calibration mark with acetonitrile. The resulting solution is a mixture of the four C4 carbonyl compounds at approximately 10 ppm for each derivative. The third standard mixture is a premade solution that contains 13-carbonyl compounds derivatized with DNPH. This mixture was also obtained from Radian Corporation. The purity of each compound is 99% and each derivative is at a concentration of 3 ppm. The following is a list of the compounds contained in the solution: formaldehyde, acetaldehyde, acetone, acrolein (MW 236), propionaldehyde (MW 238), crotonaldehyde, methyl ethyl ketone, methacrolein; butyraldehyde, ----benzaldehyde (MW 286), valeraldehyde (MW 266), m-tolualdehyde (MW 300), and hexanal (MW 280). 26

PAGE 45

2.1.2 Samples As stated previously, samples from the surrounding Denver metro area were analyzed. The first site is located just northeast of the city ofDenver. This area is known to contain many unidentified carbonyl compounds and specifically has shown unidentified peaks in chromatograms with retention times in the vicinity of C4 carbonyl compounds. The samples are collected using a vacuum pump equipped with a flow meter control and an intake valve where a sample cartridge can be connected. The flow meter is calibrated before entering the field to 3 liters a minute. Each sample is collected for twenty-four hours for a total of 4,320 liters of air. The second location is samples taken from downtown Denver. These samples are collected as part of the ongoing study of air samples by the Environmental Sciences Laboratory at the University of Colorado at Denver. An automated sequential sampler is employed to obtain four hour samples throughout the day. The system is designed to take in 720 liters of air for a four hour period before switching to a new cartridge. 2.2 Instrumentation and Accessories The United States Geological Survey (USGS) provided instrumentation for analysis. The instrumentation consists of aHewlett Packard 1-100 HPLC system ----------------that includes a vacuum degasser, column thermostat, and an ultraviolet-visible diode array detector. Separation was accomplished with a 15 em (2.0 mm id), ODS (3 reversed phase column for electrospray and 25 em ( 4.6 mm id), Zorbax ODS (5 !J.m) reversed phase column for APCI. A guard column is employed in each case to remove possible contaminants. The solvent system 27

PAGE 46

consisted of a 10 mM solution of ammonium formate and acetonitrile for electrospray analysis. The ammonium formate is added to aid in the ionization of the analyte. The solvent system for APCI included deionized water and acetonitrile. The mass spectrometer is a Hewlett Packard 1100 MSD that utilizes an atmospheric pressure interface capable ofboth electrospray and atmospheric pressure chemical ionization. lonspray instead of pure electrospray is employed since the instrument is set up to run in this mode and due to its advantages. Nitrogen is used as the nebulizing gas. Mass spectrometry parameters which are capable of manual adjustment via the computer include capillary voltage (controls the voltage applied to the entrance ofthe capillary), drying gas flow (controls the flow rate ofthe nitrogen drying gas), nebulizer pressure (controls the pressure of the nitrogen nebulizing gas), drying gas temperature (controls the temperature of the nitrogen drying gas), fragrnentor (controls the exit voltage of the capillary thus allowing different fragment patterns), corona current (controls the current from the corona discharge needle, applicable only for APCI), and vaporization temperature (sets the temperature for the vaporizer, applicable only for APCI). A series of skimmers and plates follow the ionization chamber which accelerates the ions towards the mass analyzer. The mass analyzer consists of a quadrupole mass filter. 2.3 Analytical Techniques 2.3.1 Purity of C4 standards Since the purity of the four carbonyl standards may be in question chromatographic analysis with UV detection is employed. The vials obtained from Radian Corporation of the carbonyl derivatives had been stored in a freezer, but the date they were received is unknown, thus an investigation into the purity is 28

PAGE 47

necessary. Each C4 standard is injected separately onto the Zorbax ODS column and allowed 20 minutes to elute followed by detection at 360 nm. An isocratic elution of acetonitrile/water (65/35) is employed for separation of potential impurities. The relative amount of an impurity is calculated versus the recovery ofthe main carbonyl derivative. 2.3.2 Flow Injection Analysis The four C4 standards will be directly injected into the mass spectrometer as a flow injection analysis. This is done for several reasons. The first is to determine if the derivatives can be detected with a mass spectrometer using either ionspray or APCI with the given solvents of choice. The controllable parameters described above for the mass spectrometer will remain constant except for the :fragmentor voltage. The fragmentor voltage will be varied from a range of 50 volts to 100 volts when ionspray is used. Below 50 volts fragmentation is non-existent for the molecular ion and above 100 volts fragmentation is so great that the mass spectrum appears as noise. All the other parameters do not have the same dramatic effect on the detection ofthe analytes as long as they are kept within a range described and recommended by Hewlett Packard. Table 2.2 lists the settings for these parameters when ionspray is in use. 29

PAGE 48

Table 2.2 Ionspray MSD Parameters Parameter Capillary Voltage Drying Gas Flow Nebulizer Pressure Drying Gas Temperature Setting 4000 volts 10 liters/min 20 psi 350 oc A fragmentor voltage will be chosen baseq on two factors. The first is detector response gauged by the ion current. This is a value that describes how many ions are being detected by the mass analyzer. The larger the ion current, the more sensitive the method. Second is the pattern of fragmentation. A definite fragment pattern for each analyte is the goal in determining the identity of the C4 carbonyl compounds. fu the case of atmospheric pressure chemical ionization the positive and negative mode of analysis will be employed. The negative ion mode was not investigated in ionspray since this would have involved the use of basic buffers which would have damaged the column being used. The fragmentor voltage for APCI will be varied from 70 volts to 120 volts for precisely the same reasons as in the case for ionspray. Table 2.3 describes the other parameter settings that remain constant throughout the analysis. 30

PAGE 49

Table 2.3 APCI MSD Parameters Parameter Setting Capillary Voltage 4000 volts Drying Gas Flow. 10 liters/min Nebulizer Pressure 20 psi Drying Gas Temperature 350 oc Corona Current 40 J.LA Vaporization Temperature 425 oc 2.3.3 Precision Once a suitable fragmentor voltage has been selected the precision of the analytical method will be determined at this voltage for each of the standards. Precision is a measure of the degree in which individual test results are in agreement. This will be determined by measuring the repea,tability of each standard. Each standard is injected 10 times, once again bypassing the column. Statistical calculations are performed for the ten standard injections to assess the validity of the method. Repeatability is evaluated in terms of percent relative standard deviation. Two criteria will be evaluated for precision. The first is the total ion current. The second is the relative abundances of each fragment for that particular standard. This is performed for both pneumatically assisted electrospray andAPCI. 31

PAGE 50

2.3.4 Chromatography of Standard Mixtures The standard mixture that contains formaldehyde, acetaldehyde and acetone along with the standard mixture of the c4 derivatives is injected to gauge how the mass spectrometer responds to the mixture of compounds for both ionization techniques. The second ofthe two is used for verification of the mass spectrometer in its performance to distinguish between and identify the four carbonyl derivatives. The third standard mixture consisting ofthe 13 different carbonyl derivatives is used to judge how effectively the C4 derivatives can be identified within a different sample matrix. HPLC conditions for ionspray consists of a gradient separation mode at a constant flow rate of 0.2 mL/min with an injection volume of 10 J.l.L. Using the two solvents previously described initial conditions consist of 45/55 percent composition of 10 mM ammonium formate __ -. a:naacetonitrile respectively. This is ramped to a composition of 35/65 percent over twenty-five minutes. Finally this is ramped to a composition of 30/70 percent over the next ten minutes until a return to the initial conditions is established in preparation for the next run. HPLC conditions for APCI consists of a gradient separation mode at a constant flow rate of0.8 mL/min with an injection volume of 50 J.l.L. Using water and acetonitrile as the solvents, initial conditions consist of a 5_Q/50 percent composition followed by a ramp to 40/60 percent over twenty-five minutes. This is ramped to 25/75 over the next fifteen minutes until a return to the initial conditions is established in preparation for the next run. Both ionization techniques will run a full scan mode by the mass spectrometer (50 to 300 amu) followed by a selected ion monitoring mode (SIM). The selected ions are determined in the precision section by demonstrating unique fragment patterns among Isomers. 32

PAGE 51

2.3.5 Determination of Detection Limit The limit of detection is the lowest concentration of analyte in a sample that can be detected, but not necessarily quantitated with a given degree of confidence under the stated experimental conditions. This is an essential parameter that needs to be addressed to ensure accurate determination of the carbonyl compounds. The four carbonyl mixture will be used with analysis accomplished in the selected ion monitoring mode. Solutions are prepared by diluting the four carbonyl mixture to levels of roughly 1 parts per billion volume (ppbv), 100 parts per trillion volume (pptv) and 10 pptv. These concentrations are air equivalent concentrations and conversions to these values is described in Appendix A. Accurate detection at these levels determines the limit of detection. 2.3.6 Analysis of Samples Samples that contain unidentified C4 compounds are analyzed to determine the effectiveness and specificity of the method. All parameters and conditions that were optimized above remain the same. This is performed for both ionspray and APCI to determine the most effective ionization technique for samples. Rough quantitative data is determined for samples using the most effective ionization technique. This is determined by using the limit of detection data to create calibration curves for the C4 compounds. The numbers will be rough estimates due to the standards and samples being run on separate days. This is an -; -------------------unavoidable consequence ofhaving limited time with use of the instrumentation. 33

PAGE 52

2.3. 7 Further Exploration Samples of acrolein and methyl vinyl ketone are analyzed to determine various trends for these compounds. Both compounds were derivatized with DNPH by the Environmental Sciences Laboratory at the University of Colorado at Denver. Solutions of approximately 100 ppm are prepared by dissolving 10 mg of sample into a 100 mL volumetric flask with acetonitrile. Solutions are analyzed at these concentrations. .... -. 34

PAGE 53

3. Results and Discussion 3.1 Standard Concentrations and Purity Actual concentrations ofthe four C4 carbonyl derivative solutions are given in Table 3 .1. Also included is the equivalent air concentrations based on a sample of 720 liters of air from the Denver metro area. A detailed account of these concentrations can be found in Appendix A. Table 3.1 Concentrations of the C4 Carbonyl Derivative Solutions Derivative Concentration Equivalent Air (J.l.g/mL) Concentration (ppbv) Methacrolein 64.5 52.9 Crotonaldehyde 63.0 51.7 Methyl Ethyl Ketone 62.8 51.1 Butyraldehyde 67.6 55.0 Each of the four derivatives proved to be pure by analysis of the chromatograms. Any impuritydetected was calculated to be less total area versus the-----------main DNP-hydrazone peak. Chromatograms for the four hydrazones are shown in figures 3.1-3.4. 35

PAGE 54

Figure 3.1 UV Chromatogram of Methacrolein Derivative Standard llo\01 A, Sig=J60,4 RDf:::500,5D, (SCGAF<:\AR:llOOOO.DJ ....... 175 150 125 100 75 50 25 2 8 ,, 14 ... ' .,;, Figure 3.2 UV Chromatogram of Crotonaldehyde Derivative Standard 00.01 A, ""1=500,50, TT (9CGAPCIAR>l0009.D) ...... :;: 250 200 150 100 50 ' ,, 7, 10 ,', ... 36

PAGE 55

Figure 3.3 UV Chromatogram of Methyl Ethyl Ketone Derivative Standard A, Slg='JI1Cl,4 Ref.,500,50, TT (BCGAFOAR:00003.0) l'bm .... ;l; 250 200 150 100 50 0 ; io ,., i ,;, Figure 3.4 UV Chromatogram of Butyraldehyde Derivative Standard CADI A, Ro1=500,50, TT iSCGO.PC\AFCID012.0) l'bm 2 250 200 150 100 50 0 ; ;. io ,., 1il 1 37

PAGE 56

Concentrations for the four carbonyl standard mixture are given in Table 3 .2. Included are the air equivalent concentrations based on 720 liters of Denver metro air sampled. Table 3.2 Concentrations of the C4 Carbonyl Derivative Mixture Solution Derivative Concentration Equivalent Air (J.Lg/mL) Concentration (ppbv) Methacrolein 6.45 5.29 Crotonaldehyde 6.30 5.17 Methyl Ethyl Ketone 6.28 5.11 Butyraldehyde 6.7 All thirteen carbonyl compound derivatives found in the second mixture are at a solution concentration of 3 Jlg/mL. This corresponds to an equivalent air concentration of2.46 ppbv for methacrolein and crotonaldehyde for 720 liters of sampled air in the Denver metro area and 2.44 ppbv for methyl ethyl ketone and butyraldehyde. 38

PAGE 57

3.2 Flow Injection Analysis 3.2.1 lonspray Ionization A fragmentor voltage for pneumatically assisted electrospray was determined first. The value chosen was 70 volts. This was the voltage of choice for several reasons. First using 50 or 60 volts revealed little more than the molecular ion peak making identification of isomers nearly impossible. Those values above 70 volts contained a variety of fragments, but once again it was difficult to distinguish between isomers. At 70 volts, however unique fragments were observed for each of the C4 carbonyl derivatives. The total ion current was substantial enough for each of the voltages analyzed to allow for qualitative data. No significant difference was seen in the total ion current area going from 50 volts to 100 volts. A comparison of each voltage is shown in figures 3.5-3.10. A discussion of the fragments brought about at 70 volts will take place in the subsequent section. 39

PAGE 58

Figure 3.5 C4 Carbonyl Mass Spectra using lonspray, Fragmentor 50V 1. Methacrolein PASD1 SPC, tim!1=7.373 at Af\.ES PDsitive 100 60 "1 I 20 a 1 t sa 100 -,so :>Oo 2so .., 3.MEK MSD1 SPC, till'm=0.645 a F:\81MNtH\BGflAES\BCGODDD3.0 AR-ES Positive 100 eo 60 Oo 200 2. Crotonaldehyde 4. Butyraldehyde CGOJOOl.O AA-ES Po:sitiv rh.x: 44.595% MS01 SPC, tiJfl! 20.684 or F:\BRtAN0-1\BGFIA.ES\BCGODODJ.D AA-ES Posiltv 100 f'&tx: 549632 80 60
PAGE 59

Figure 3.6 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 60V 1. Methacrolein '.SUI SPC, tiiTE=B.456 or Afi.Oo> ..,sltivo 100 166528 80 60 41) 20 ; 0 sO I "L '. 100 ISO 21io ... 3.MEK 100 "MSD1 SPC, tinm 1. 94 a F:\BR1AND-1\BGR4ES\BCGCOQ03.u AR-ES F'D&itive x: 6096fi4 80 60 41) 20 50 .. 41 2. Crotonaldehyde SPC, tinwo:z15. 140 al F:\BRIA.M>-t\SGFIAES\SCGOOQ03.D AP!oo Positivlll 100 BO 60 40 4. Butyraldehyde 100 80 60 40 20 0 50 100 ISO m .i ;t I _,;. r .... l SoOBiiOB ;5o '.m

PAGE 60

Figure 3.7 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 70V 1. Methacrolein 2. Crotonaldehyde 'MS01 SflC, of AA-ES PelitiVe 'MSD1 SFt;, lim!l..,8.245 cr F:\BRIAfoD-1\BGFlA.ES\EICGOODOJ.D AR-ES Pmitlv!O 100 :;tx: 108600 100 Sax: 260352 eo 80 eo eo 40 40 20 .T ; 1 "' I s '! _1,, I so ,co 1 'o 200 250 ., 20 0 '" on 3.MEK 4. Butyraldehyde "MSD1 SPC, tirr'IQ;2.899 of F:\BRIAM>-1 03.0 AA-1::.5 flo.sitP.'e "MSD1 SPC, timta:22.BSIB ct f':\BRIANI:>-1\SGAA.ES\BCGOaDQ3.D AR-ES Positiv 100 .Qx: JS5456 100 tAx: 33597: eo eo 60 eo 40 40 20 1 0 sO 100 1So 20o 2s "" 1!1 1 20 1 0 sO 100 50 :!Oo 20 ., 42

PAGE 61

Figure 3.8 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 80V 1. Methacrolein 2. Crotonaldehyde "MS 3.0 Positlv 3.0 100 a.u: 61768 100 ,g., 206080 80 80 60 40 i 40 ;_ 20 20 i I 50 3.MEK 4. Butyraldehyde "MS01 Sf'C, a .... oJ.982 of EIBRIAN0-118GFIAESIBCG00003.D Af'I.ES PDiliVe "MS01 SFC, tina"23.957 of F:IBRIANHIBGAAESIBCG00003.0 Af'I.ES PooiiN 100 146048 100 ... 205632 80 80 60 60 40 t! 20 JJ 1 ;;; D .i N D 50 100 1So oDD ,; .... 40 20 o s'o ,; 43

PAGE 62

Figure 3.9 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor 90V 1. Methacrolein 2. Crotonaldehyde "MS01 SA:. time=11.781 ol F:\BRIANJ-1\SG1AES\BCGOOD03.0 AR-ES Pusitiv' "MSOt SPC. tirnt.476 af F:\BRIANJ-1\Ba:IAES\BCG00003.0 AR-ES PI:Jsdiv 100 ;,; Hn.: 455B4 100 Max: 165l20 80 80 N :8 60 60 40 20 40 : ;i 20 HH ,UHi ; 0 sb l 100 1so 200 2s ... 3.MEK 4. Butyraldehyde "MSD1 S tlrre..S.119 or A ES Positive "MSC1 SPC,titre=25.10 ar API-ESPositiv "' 100 HalCI 1'7'7152 80 80 0 60 60 ;; 40 40 20 20 44

PAGE 63

Figure 3.10 C4 Carbonyl Mass Spectra using Ionspray, Fragmentor lOOV 1. Methacrolein 2. Crotonaldehyde SPC, of :\BRlAr-.D-1\S CG00003.0 AA-ES Positiv 'MSCt SPC, ti.tne=t9.55!1 of Pasitivd 100 80 60 60 40 40 20 3.MEK 4. Butyraldehyde 1\BGFIAES\BCGOD003.0 AA-Fbai1ive 'MSDt SPC, time=2B.210 of F:\8 j 100 80 100 80 60 60 40 20 20 50 45

PAGE 64

3.2.2 Atmospheric Pressure Chemical Ionization A value of 90 volts was chosen for the fragmentor in atmospheric pressure chemical ionization. This followed the same reasoning as in ionspray ionization. Values below 90 volts in most cases only revealed the molecular ion. Those above 90 volts generated too many fragments making it difficult to distinguish between isomers. Ninety volts proved to be a good median providing unique fragments among isomers. One noticeable feature revealed as the fragmentor voltages were increased was a decrease in the total ion current. This is a result of ions losing their charge as they collide with other molecules. The ion current at 90 volts, however is large enough for adequate sensitivity. A comparison of each voltage is shown in figures 3.11-3.16. A discussion ofthe major fragments obtained at 90 volts is discussed in the ensuing section. 46

PAGE 65

Figure 3.11 C4 Carbonyl Mass Spectra using APCI, Fragmentor 70V 1. Methacrolein 2. Crotonaldehyde IOJ 4: 6.,710li "" 111 .. "' "' ,. ,;., ... 3.MEK 4. Butyraldehyde ": Mi: .:.U1121 .. 0DG .. 47

PAGE 66

Figure 3.12 C4 Carbonyl Mass Spectra using APCI, Fragementor 80V 1. Methacrolein 2. Crotonaldehyde """'""""'"''"'" -f------:!.6U:J1U06 4,o;JJbo006 "' ... ., r ., .r s I I. l .. I 3.:MEK 4. Butyraldehyde Ull ., Ill "' ., ., 48

PAGE 67

Figure 3.13 C4 Carbonyl Mass Spectra using APCI, Fragementor 90V 1. Methacrolein 2. Crotonaldehyde """"'-"" 1..: I.O!ii>1'7G06 ,., il':l I Ill Ill ., .., -t f li. l .r 9 ;j a -a i! I l I I ld,,: l.d 3.lVIEK 4. Butyraldehyde r--......... .... LJ:: tic, oor. ,., ., ., Ill ., ., !i g .. "' "' 11n l I J T i: 49

PAGE 68

Figure 3.14 C4 Carbonyl Mass Spectra using APCI, Fragementor lOOV 1. Methacrolein 2. Crotonaldehyde 1111 :!i.u.: 119172 1111 ; Haa:: ., ., ., ., ., 3.MEK 4. Butyraldehyde 'MD SR:ti'rlllriii.ZBc:l 1111 1111 Hlolll ., ., s ., ., s g i i ., ., "' "' 50

PAGE 69

Figure 3.15 C4 Carbonyl Mass Spectra using APCI, Fragementor llOV 1. Methacrolein 2. Crotonaldehyde ZDAJ
PAGE 70

Figure 3.16 C4 Carbonyl Mass Spectra using APCI, Fragementor 120V 1. Methacrolein 2. Crotonaldehyde 2DAJ
PAGE 71

3.3 Precision 3.3.1 Ionspray Ionization Table 3.3 lists the total ion current for ten replicate injections at a fragmentor voltage of70 volts for each ofthe carbonyl derivatives as well as percent relative standard deviation (%RSD). All four of the compounds have a %RSD less than 2. This indicates very good precision from injection to injection. Secondly, tables 3.4-3.7 show the major fragments for each ofthe four compounds including the average relative abundance for each fragment (those underlined are significantly different among isomers). Relative abundances tended to show greater variability, but definite patterns could be ascertained from the data. For the two isomers, methacrolein and crotonaldehyde, many of the fragments are similar and abundances are close. Two noticeable differences can be made however, the first is methacrolein does not show a fragment at a rn!z of68. The second is the fragment at a rn!z of 149 is much larger for methacrolein than crotonaldehyde. Thus from this data it appears the two isomers can be distinguished. A more drastic difference can be seen for the two larger isomers, methyl ethyl ketone and butyraldehyde. The first is methyl ethyl ketone has only two considerable size fragments at this fragmentor voltage. These include the molecular ion peak and a fragment at a rn!z of 70. Butyraldehyde shows a significant abundance for several other ions except for the one with a rn!z of 70. These two isomers appear to be ..... :-easily distinguishable by their fragments. What follows the tables of fragments are probable fragment patterns for each ofthe four compounds in figures 3.173.20. 53

PAGE 72

Table 3.3 Precision for C4 Carbonyl Compounds using Ionspray in Total Ion Current Methacrolein Crotonaldehyde MEK Butyraldehydc 1.23 X 107 7.53 X 106 1.38 X 107 1.41 X 107 1.21 X 107 7.3.0 X 106 1.39 X 107 1.39 X 107 1.19 X 107 7.28 X 106 1.41 X 107 1.38 X 107 1.17x107 7.38 X 106 1.39 X 107 1.38 X 107 1.21 X 107 7.23 X 106 1.39x 107 1.37 X 107 1.19x 107 7.57 X 106 1.39 X 107 1.37 X 107 1.19 X 107 7.39 X 106 1.40 X 107 1.40 X 107 1.17x107 7.36 X 106 1.42 X 107 1.38 X 107 1.17x107 7.41 X 106 1.40 X 107 1.39 X 107 1.20 X 107 7.62 X 106 1.39 X 107 1.37 X 107 Mean 1.19x 101 7.41 X 106 1.40 X 107 1.38 X 107 StdDev 2.07 X 105 1.28 X 105 1.27 X 105 1.38 X 105 %RSD 1.73 1.73 0.91 1.00 54

PAGE 73

Table 3.4 Major Fragments for Methacrolein using lonspray mJz Ion Average Abundance 68 [M+H-183] <2% 149 [M+H-102t 30 164 [M+H-s7r 10 183 [M+H-68t 15 234 [M+H-17t 20 251 [M+Ht 100 Table 3.5 Major Fragments for Crotonaldehyde using Ionspray mJz Ion Average Abundance 68 [M+H-183] 15 149 [M+H-102t 12 164 [M+H-87r 10 183 [M+H-6sr 12 234 [M+H-17t 14 251 [M+Ht 100 55

PAGE 74

I Table 3.6 Major Fragments for Methyl Ethyl Ketone Ionspray Dllz Ion Average Abundance 70 [M+H-183r 15 138 [M+H-11St <2% 164 [M+H-89t <2% 183 [M+H-70t <2% 193 [M+H-60t <2% 236 [M+H-17t <2% 253 [M+Ht 100 Table 3. 7 Major Fragments for Butyraldehyde using Ionspray lrilz Ion Average Abundance 70 [M+H-1s3r <2% 138 [M+H-11St 19 164 [M+H-89t 17 183 [M+H-70t 16 193 [M+H-60t 13 236 _[M+H-17t 18 253 [M+Ht 100 56

PAGE 75

Figure 3.17 Proposed Fragment Mechanisms for Crotonaldehyde using Ionspray o2N o,N H ----N =c(H CH 3 c=c/ H H NO 2 +H H, /H L(_ )---f N02 o2 N MWZSI H, /H H, H H 3 H H NH 1 N02 ----'--C '<::::N H NO 2 o2 N MW234 MW1BJ -C 1H ,NO -C 4 H a NO 2 ===:N )-J-# 0 2N -C aH s N ,a + /H N =C, CH 3 c=c.,....... H H MWGI 57 Abundance 100 14 12 10 12 1S

PAGE 76

Figure 3.18 Proposed Fragment Mechanisms for Methacrolein using Ionspray Abundance 100 MWlSI H 2 ?CH2 P-C c,CH II N02 N 02N 24 MWlJ4 1:) MWI83 10 MW164 10 MW149 58

PAGE 77

Figure 3.19 Proposed Fragment Mechanisms for MEK using Ionspray AbWldance 100 +H+ MW2SJ 1 s MW70 59

PAGE 78

Figure 3.20 Proposed Fragment Mechanisms for Butyraldehyde using Ionspray 0 ,N "-=c/" " 'c _c,CH H' NO 1 Hl /H H H 2 CH NO 0 '" H 2 H 2 H c-c....._ r=<_-N =C ........ NO __ -.;;;NH-'---'''--c H' NO 0 1 N 0 ,N MW2JB -N :0: "'-=c( ;-........ CH 3 0 0 MW19J ..C 4H ,H s 0 ,N MW18l 1H ,NO # 0 " MWI&.& H' 4 H ,N :zO l s > 0. 0 MW131 oC 1 H 5 N 30 ====c /H H 2 ......_c_c......_ H 2 CH 3 MW70 60 Abundance 100 1R Hi 17 1 <2

PAGE 79

3.3.2 Atmospheric Pressure Chemical Ionization Table 3.8 lists the total ion current for ten replicate injections at a fragmentor voltage of 90 volts for each of the carbonyl derivatives as well as percent relative standard-deviation (%RSD). All four compounds have a %RSD ofless than three. This indicates good precision from injection to injection. Secondly, tables 3.9-12 show the major fragments for each ofthe four compounds including the average relative abundance for each fragment (those underlined represent major differences among isomers). Relative abundances tended to show greater variability, and as in ionspray definite patterns could be established from the data. Most ofthe ions fragments for methacrolein and crotonaldehyde are very similar with the one difference being in the ion with a m/z of79. This fragment is present in the mass spectra of methacrolein, but not for crotonaldehyde. This ion fragment enables a distinction to be made between the two isomers. These results compare favorably with the work done by Kolliker and co-workers (18). The only difference in fragment patterns is they saw fragments that included [M-H35T and [M-H-72T. Both ofthese fragments were absent at the given fragmentor voltage. The two isomers methyl ethyl ketone and butyraldehyde show a greater difference in their fragment patterns. The most notable being the absence of a fragment at a m/z of 163 for methyl ethyl ketone. From this alone the two of these compounds can be distinguished in a sample. These results also compare favorably with the work done by Kolliker and co-workers (18). They claim for a ketone an ion with m/z 163 < 179, [M-H-30T is present,_ and m/z 152 is the biggest fragment. This is indeed the case for methyl ethyl ketone. For the aldehyde m/z 163 is abundant, [M-H-3or is present, and m/z 191 is approximately 10% of the base peak. This holds true for butyraldehyde. It is evident from these results and the previous work these compounds tend to 61

PAGE 80

fragment in given patterns. What follows the tables of fragments are probable fragment patterns for each ofthe four compounds in figures 3.21-3.24. Table 3.8 Precision for C4 Carbonyl Compounds using APCI hi Total Ion Current Methacrolein Croton aldehyde MEK Butyraldehyde 1.58 X 101 2.27 X 101 3.25 X 101 2.36 X 101 L60 X 107 2.26 X 107 3.24 X 107 2.36 X 107 1.62 X 107 2.28 X 107 3.27 X 107 2.42 X 107 1.66 X 107 2.22 X 107 3.34 X 107 2.40 X 107 1.62 X 107 2.30 X 107 3.31 X 107 2.31 X 107 1.71 X 107 2.31x107 3.35 X 107 2.37 X 107 1.67x 107 2.28 X 107 3.42 X 107 2.42 X 107 1.67 X 107 2.24 X 107 3.3_9 X 107 2.36 X 107 1.70 X 107 2.27 X 107 3.25 X 107 2.39 X 107 1.65 X 107 2.34 X 107 3.40 X 107 2.37 X 107 Mean 1.65 X 107 2.28 X 106 3.32 X 101 2.38 X 101 Std Dev 4.24 X 105 3.43 X 105 6.78 X 105 3.31 X 105 %RSD 2.57 1.51 2.04 1.39 62

PAGE 81

Table 3.9 Major Fragments for 1.\'Iethacrolein using APCI DiJz Ion Average Abundance 79 [M-H-11or 6 152 [M-H-97]" 6 163 [M-H-86]" 13 172 [M-H-77]" 12 202 [M-H-47]" 12 218 [M-H-31]" 8 249 [M-H]" 100 Table 3.10 Major Fragments for Crotonaldehyde using APCI . --. -. ----Dilz 79 152 163 172 202 218 249 Ion [M-H-170]" [!vf-H-97]" [M-H-86]" [M-H:nr [M-H-47r [!vl-H-31]" [M-Hr 63 Average Abundance <2% 11 10 22 20 10 100

PAGE 82

Table 3.11 Major Fragments for Methyl Ethyl Ketone using APCI Dllz Ion Average Abundance 152 [M-H-99r 24 163 [lVI-H-88r <2% 179 [M-H-nr 6 191 [M-H-6or 1 221 [M-H-3or 34 251 [M-Hr 100 Table 3.12 Major Fragments for Butyraldehyde using APCI Dllz Ion Average Abundance 152 [M-H-99r 39 163 [M-H-ssr 22 179 [M-H-nr 13 191 [M-H-6or l! 221 [M-H-3or 27 251 [M-Hr 100 64

PAGE 83

Figure 3.21 Proposed Fragment Mechanisms for Crotonaldehyde using APCI Abundance 100 H / N-N==C'c=c/CH3 H H .HNO H / N-N=C'c=c/CH2 10 N02 H H 0 O,N o,N MWZII -HNO 2 20 MWlOl H / PN-N=C'c=c/CH' H H \. ,I 0 22 --MWtn 10 11 ,. _.. .._; .. __ ... .. 65

PAGE 84

Figure 3.22 Proposed Fragment Mechanisms for Methacrolein using APCI 0 ,N Abundance H H / N-=c....._ 9-l. N-N=c(H C l NO J et 3 -HNO 2 o,N ll\Vlll H -N-N=C/ r=< 'l==Oi. )-J Oi. 0 l'ti\Vl7Z s ) 0 O,N M.VISZ MW?J 66 100 12 12 11

PAGE 85

Figure 3.23 Proposed Fragment Mechanisms for MEK using APCI r<__-:, p-N N=C<:: )-----/ NO 2 JH 3 0 2N o2N N ---N =c(CH 3 CH 2 NO 2 JH 3 -NO 0 2N -N ,o, ,NO 67 MWZSI 0 MWUI /CH 3 N-N=C'-cH 2 I CH 3 M.WI91 AblUldance 100 14 4 24

PAGE 86

Figure 3.24 Proposed Fragnient Mechanisms for Butyraldehyde using APCI o,N H H, N .-=c'-c/c'' H 1 't:H NO 1 -NO 0 1 N 0 1 N H H 1 N-N=C....._ /c,CH 3 l, NO H H2 N ---N =c c 'c/ 'cH' H, 0 PfflZ21 H H 2 N-N=C / '-c '-cH H' 0 0 Pffl191 9--: O 2N MW179 o2 N P-o MW1SZ 68 Abundance 100 27 11 11 22

PAGE 87

3.4 Chromatography of Standard Mixtures 3.4.1 Chromatography using Ionspray Figure 3.25 shows the total ion current using full scan from a rnlz of 50 to 300 for the mixture of formaldehyde, acetaldehyde, and acetone. The chromatogram reveals formaldehyde be detected using ionspray at this concentration. The other two compounds can be identified, but the chromatogram does contain a significant Acetaldehyde elutes at approximately 12.8 minutes and acetone elutes at approximately 16 mi.Iiutes. Figure 3.25 Total Ion Chromatogram of Mixture of Three Carbonyls using Ionspray 69

PAGE 88

Figure 3.26 shows the total ion current chromatogram with full scan of the four carbonyl derivative mixture using ionspray. Under the given conditions separation is not great as seen by the three peaks. Figure 3.27 displays the mass spectra of the three peaks. The first and last peak can easily be identified by their mass spectrum, but the two in the middle are indistinguishable by retention times. One can deduce what the tWo species are in this standard mixture, but this will most certainly not hold up in a more complex matrix which contains other C4 carbonyls. Figure 3.26 Total Ion Current of the 4 Carbonyl Derivative Standard using Ionspray under Full Scan MiDI AAS-o =1 IDXXI 70

PAGE 89

Figure 3.27 Mass Spectra using Ionspray under Full Scan 0M iDI S ol 8COl\ESOOOuO .u ... lt.::O OIIIIWI "' I I .I. ,II ,II I. I D 1 S P time" 0 .II 1 o SCG "60000 O.D A PI a I '"'1 M _. I l " ., .... I I. .'MS01 .iiO!D Cl tHOU :U:SO ".u lt:S "' M .._ I It bOt " I .II. J I I Figure 3.28 shows the first of two selected ion monitoring chromatograms of total ion current for this mixture and a blank.The'selediori of ions had a riiass to charge ratio of68, 149, and 251. With this choice of ions only two ofthe four ions should be detected and this is indeed the case. The chromatogram shows two peaks for the isomers having a molecular weight of250 amu. The two other peaks in the chromatogram cari. riot be identified even by shooting a blank of the eluent. Figure. 3.29 displays the mass .spectra for these two peaks. Table 3.13 contains the relative abundances of the ion peaks. These numbers do not compare 71

PAGE 90

to previously found abundances very well (see Tables 3.4 and 3.5). It appears that crotonaldehyde elutes first with such a large fragment at 68. This can be confirmed by the larger abundance of the 149 ion in the second eluting peak, which is methacrolein. Figure 3.28 Total Ion Current for 4 Carbonyl Derivative Standard and blank using Ionspray under SIM (68, 149, 251) monitoring for Crotonaldehyde and Methacrolein 1amJ IIIID )\A __ MSDI n::. MS Fill (BCGIES000061.0, Afi.ES Pll..,e 10000 72

PAGE 91

Figure 3.29 Mass Spectra using Ionspray under SIM S C.II,.I4.113l Ol fiCQ IIE.SOI 001,0 A 1:11 OloiY l 'WS01 SPC,IImlO.r:a 0 11;0l\ES000004.0 A 1;11 I'OIHIVI _l I Table 3.13 Relative Abundance of Ions using Ionspray under SIM m/z 68 149 251 Ion Abundance Peak at 18.7 minutes Peak at 20.7 minutes 18 2 100 8 6 100 Figure 3.30 shows the second chromatogram of total ion current for selected ion monitoring. The ions chosen here include a mass to charge ratio of70, 138, 193, 236, and 253. These choices reveal the other two compounds. The peaR which 73

PAGE 92

elutes right before the two major peaks cannot be identified, but its major ion is 138. Figure 3.31 displays the mass spectra for these compounds. Table 3.14 contains the relative abundances for the two ion peaks. The evidence shown here is conclusive that the first peak that elutes is methyl ethyl ketone and the second is butyraldehyde. The large fragment at 70 for the first peak and the large fragments at 138, 193 and 236 for the second peak confirm these results. Figure 3.30 Total Ion Current for 4 Carbonyl Derivative Standard using Ionspray under SIM (70, 138, 193, 236, 153) monitoring for MEK and Butyraldehyde MlDITC.M>Re( I'Gm j 140XIl 74 I

PAGE 93

Figure 3.31 Mass Spectra using Ionspray under SIM S D 1 9 C 1111'1 It I II G l l a I Q 0 E 9 0 0 0 0 0 II A I P a I 11 I .. I ... :1 I! PC, 1111'1 1 .I I I a f a"' C S 0 0 0 0 r. A C a Po I lilY I I Table 3.14 Relative Abundance of Ions using Ionspray under SIM Ion Abundance I '"'I' 1"1 m/z Peak at 20.3 minutes Peak at 21.9_minutes 70 17 5 138 2 20 193 1 13 236 3 15 253 100 100 75 ---------

PAGE 94

-, The results for the 13 carbonyl compound mixture are very similar. Figure 3.32 shows the total ion current chromatogram under full scan using ionspray ionization. Figure 3.33 displays the mass spectra for the C4 compounds. Under a more complex matrix it is very difficult to make out more than the molecular ion peak, thus identification of isomers is nearly impossible. Table 3.15 is a list of the compounds which can be detected in this mixture. Under the given chromatographic and mass spectrometry conditions formaldehyde, acrolein, and methacrolein cannot be detected. Figure 3.32 Total Ion Current of the 13 Carbonyl Derivative Standard using Ionspray under Full Scan 76

PAGE 95

Figure 3.33 Mass Spectra using Ionspray under Full Scan 1111 dl '"" IL. ,,1, I I .. lu r,l,l,.il, ,]rl .rb ,,,, .Lr. .r Ill, ,. 7 I I .L "' lrl,, .1. II ,,.,IL .1. .I' I o 1 I' I o 77

PAGE 96

Table 3.15 Retention Times and Major Fragments of Compounds in Standard Mixture of 13 Derivatives Retention Time (min) Major Ion Possible Compound 10.5 225 Acetaldehyde 13.8 239 Acetone 15.8 239 Propionaldehyde 18.7 251 Crotonaldehyde 20.4 253 Methyl Ethyl Ketone 21.8 253 Butyraldehyde 24.9 287 Benzaldehyde 29.0 267 Valeraldehyde 31.0 118 m-tolualdehyde 34.5 132 Hex anal Figure 3.34 shows the first of two selected ion monitoring chromatograms of total ion current for this mixture. The selection of ions as with the mixture of 4 compounds has a mass to charge ratio of68, 149, and 251. This selection identifies crotonaldehyde and methacrolein. Figure 3.35 displays the mass spectra for these two peaks. Table 3.16 c-ontains the relative abundances ofthe ion peaks. These numbers compare more favorably with the previous results of the ion abundances. Crotonaldehyde is the first eluting peak due to the larger fragment at 68. This is confirmed by notffi.g the second eluting peak has a large fragment at 149 which confirms methacrolein. The two unidentified peaks observed in the mixture of standards is also observed in this chromatogram. 78

PAGE 97

Figure 3.34 Total Ion Current for 13 Carbonyl Derivative Standard using Ionspray under SIM (68, 149, 251) monitoring for Crotonaldehyde and Methacrolein 1110l) 1110ll 14Cm 13m 111Dl IIOl) sm 4Cm 3XI) lA'------' ....__ ______ V\_ '---------------in Figure 3.35 Mass Spectra using Ionspray under SIM u.su1 ..... u ... t.tJa of e .. oJu;:sooo o APIE.S Potua .. o oo o I I PC: IIIII t ES .0 A lf8 POtlhl't I 0 0 .. o .. I 79

PAGE 98

Table 3.16 Relative Abundance of Ions using lonspray under SIM m/z 68 149 251 Ion Abundance Peak at 18.6 minutes Peak at 20.7 minutes 19 9 100 11 25 100 Figure 3.36 shows the second chromatogram of total ion current for selected ion monitoring. The ions chosen remain the same for identification of methyl ethyl ketone and butyraldehyde with a mass to charge ratio of70, 138, 193, 236, and 253. Figure 3.37 displays the mass spectra for these compounds. Table 3.17 contains the relative abundances for the two ion peaks. The evidence shown here is conclusive that the first peak that elutes is methyl ethyl ketone and the second is butyraldehyde. The large fragment at 70 for the first peak and the large fragments at 138, 193 and 236 for the second peak confirm these results. With the chosen fragments other compounds show a response as seen in the chromatogram. These responses are due to the fragments at a m/z of 13 8 and 193. All the compounds that could be identified using full scan appear in this SIM chromatogram with these fragments except benzaldehyde. 80

PAGE 99

Figure 3.36 Total Ion Current for 13 Carbonyl Derivative Standard using Ionspray under SIM (70, 138, 193, 236, 153) monitoring for MEK and Butyraldehyde AAS-o N:rm :z:aD = 17liiJ 151Xll Figure 3.37 Mass Spectra using Ionspray under SIM YSull!l .tlnotlO.IOiai8C'.teU:. I .0 A PIES Pllti'IJI I S C.IoJ .rJ a 8COJIE:I000010.0 A 011 .. 1 I 81 I

PAGE 100

Table 3.17 Relative Abundance of Ions using Ionspray under SIM Ion Abundance m/z Peak at 20.2 minutes Peak at 21.9 minutes 70 18 9 138 5 24 193 5 20 236 6 20 253 100 100 3.4.2 Chromatography using Atmospheric Pressure Chemical Ionization Figure 3.38 shows the total ion current for the mixture of formaldehyde, acetaldehyde, and acetone. All three compounds can be detected at this level of concentration, but there is an obvious response difference for the three compounds. Formaldehyde elutes at approximately 10.2 minutes with the most abundant mass peak at m/z of209, acetaldehyde at approximately 13.2 minutes with the most abundant mass peak at rnlz of 224, and acetone at approximately 17.5 minutes with the most abundant mass peak at rnlz of.238. _________ .... 82

PAGE 101

Figure 3.38 Total Ion Chromatogram of Mixture of Three Carbonyls using APCI lllXIXl ltiDlO 5 10 ;s Figure 3.39 shows the total ion current chromatogram with full scan of the four carbonyl derivative mixture using APCI. As in ionspray, the given conditions produce poor separation as seen by the three peaks. Figure 3.40 displays the mass spectra of the three peaks. The mass spectra reveal little more than the molecular ion peak. Most of the significant fragments are present in every spectrum making it difficult to distinguish between isomers. 83

PAGE 102

Figure 3.39 Total Ion Current of the 4 Carbonyl Derivative. Standard using __ APCI under Full Scan Figure 3.40 Mass Spectra using APCI under Full Scan l ,L I .1, u I 3 P ,_. IIIII 1. _f f. f I) A PC IN 11 II" e .I I. :. ........ "" -"" "' < ,,, ,, .II, I. Ill I. 84

PAGE 103

Figure 3.41 shows the first oftwo selected ion monitoring chromatograms of total ion current for this mixture using APCI. The selection of ions had a mass to charge ratio of 79, 172, 202, 249. With this choice of ions only two of the four ions are detected. The chromatogram shows two peaks for the isomers having a molecular weight of250 amu. Figure 3.42 displays the mass spectra for these two peaks. Table 3.18 contains the relative abundances of the ion peaks. These numbers are in good agreement with previous findings. It appears that crotonaldehyde elutes first with such a larger fragment at 172 and 202. This can be confirmed by the larger abundance of the 79 ion fragment in the second eluting peak, which is methacrolein. Figure 3.41 Total Ion Current for 4 Carbonyl Derivative Standard using APCI under SIM (79, 172, 202,249) monitoring for Crotonaldehyde and Methacrolein MD111C;Mlfile(F.\IRI\I'0-1'8CI>MRIIII2Q Aftll'llgdNo Nlm = 1ra:JXO 1a:m:D rmxo 0 \j <. ... .;. ... .;. ; 85

PAGE 104

Figure 3.42 Mass Spectra using APCI under SIM 1 U S 0 I 9 Inn :1 I .I I Ill 0 f :\" M' lA N 1.1-I 8 C 0 2 \A ... "t.ISD 111"1::.1"':11. 0 ol :llliliAND-CQ21APOOOOQ ,g A c; Ngllot Table 3.18 Relative Abundance of Ions using APCI under SIM Ion Abundance m/z Peak at 25.0 minutes Peak at 26.7 minutes 79 0 5 172 20 12 202 15 11 249 100 100 Figure 3.43 shows the second chromatogram of total ion current for selected ion monitoring. The ions chosen here include a mass to charge ratio of 163, 221, 251. These choices reveal the other two compounds. Figure 3.44 displays the mass 86

PAGE 105

spectra for these compounds. Table 3.19 contains the relative abundances for the two ion peaks. The evidence shown here is conclusive that the first peak that elutes is methyl ethyl ketone and the second is butyraldehyde. The larger fragment at 221 for the first peak confirms MEK and the larger fragments at 163 for the second peak confirms butyraldehyde. Figure 3.43 Total Ion Current for 4 Carbonyl Derivative Standard using APCI under SIM (163, 221, 251) monitoring for MEK and Butyraldehyde '-'fDilt:.'-SFie(BnWCJXml:l NJm = 17faXD 15CilXD 1:1!aDJ 111XDll 7liiDl 5IIIDl 2SIIIl 1n ,'. Figure 3.44 Mass Spectra using APCI under SIM "MSDI 111110 .5 o 8CG8\APCIOOO:Z.D A _"II SOl 8PC. of BCO 11\A PCIOOOl.O A PCIHogouoo I 87

PAGE 106

Table 3.19 Relative Abundance of Ions using APCI under SIM mfz 163 221 251 Ion Abundance Peak at 27.5 minutes Peak at 28.5 minutes 2 30 100 16 23 100 The results for the 13 carbonyl compound mixture are very similar. Figure 3.45 shows the total ion current chromatogram under full scan using APCI. Figure 3.46 displays the mass spectra for the C4 compounds. Once again, under a more complex matrix it is very difficult to make out more than the molecular ion peak, thus identification of isomers is nearly impossible. Table 3.20 is a list of the compounds which can be detected in this mixture. The only two that cannot be detected are acrolein which probably coelutes with acetone and methacrolein which probably coelutes with methyl ethyl ketone. The results compare favorably to the work previously discussed by Grosjean a.Jld co-workers (22, 23). Their chromatography appears to be more efficient, but they too have problems resolving Acrolein and methacrolein. The scan range used also begins at a higher value of 125. This will eliminate some of the smaller fragments, thus decreasing the noise. 88

PAGE 107

Figure 3.45 Total Ion Current of the 13 Carbonyl Derivative Standard using APCI under Full Scan Figure 3.46 Mass Spectra using APCI under Full Scan II. .II '" J.. 111 I I I I : ......... . I II 89

PAGE 108

Table 3.20 Retention Times and Major Fragments of Compounds in Standard Mixture of 13 Derivatives Retention Time (min) Major Ion Possible Compound 11.0 209 Formaldehyde 14.5 224 Acetaldehyde 19.0 237 Acetone 21.0 237 Propionaldehyde 25.0 249 Crotonaldehyde 27.0 251 Methyl Ethyl Ketone 28.3 251 Butyraldehyde 31.5 285 Benzaldehyde 34.8,35.8 265 Valeraldehyde 37.2,37.8 299 m-tolualdehyde 41.8 279 Hex anal Figure 3.47 shows the first of two selected ion monitoring chromatograms of total ion current for this mixture. The selection of ions as with the mixture of 4 compounds has a mass to charge ration of79, 172, 202, 249. This selection identifies crotonaldehyde and methacrolein. Figure 3.48 displays the mass spectra for these two peaks. 'fable 3.21 contains the relative of the ionpeaks. These numbers compare favorably with the previous results of the ion abundances and with the 4 carbonyl mixture. Crotonaldehyde is the first eluting peak due to the larger fragment at 172. This is confirmed by noting the second eluting peak has a larger fragment at 79. The late eluting peak is a fragment ofhexanal at m/z 249. 90

PAGE 109

Figure 3.47 Total Ion Current for 13 Carbonyl Derivative Standard using APCI under SIM (79, 172, 202, 249) monitoring for Crotonaldehyde and Methacrolein Nlm :mxD 2SIXID :m:m 1!mll 1CIXIIl 1 !i:lliJ l 0 i; 1o ,.. "' ,., '" Figure 3.48 Mass Spectra using APCI under SIM 'W 0 5PC.II.,Iol4. 41 D 0. A""' Nlljll!w 100 .. I I SPC.Iuttel t 0 \8AIAN0 IICOJU1PODODO .D APCINIIIIhWI 10 0 .. .. .. 20 -I I 91

PAGE 110

Table 3.21 Relative Abundance of Ions using APCI under SIM Ion Abundance mlz Peak at 25.0 minutes Peak at 26.7 minutes 79 172 202 249 0 19 16 100 9 12 11 100 Figure 3.49 shows the second chromatogram of total ion current for selected ion monitoring. The ions chosen remain the same for identification of methyl ethyl ketone and butyraldehyde with a mass to charge ratio of 163, 221 and 251. Figure 3.50 displays the mass spectra for these compounds. Table 3.22 contains the relative abundances for the two ion peaks. Results are very similar to those previously obtained. The evidence shown here is conclusive that the first peak that elutes is methyl ethyl ketone and the second is butyraldehyde. The larger fragment at 221 for the first peak confirms l'viEK and the larger fragment at 163 for the second peak confirms butyraldehyde. It is also evident while monitoring for these two compounds the other components in the show a response to the chosen fragments. This is due to the fragment at rnl_z of 163 for the other compounds. 92

PAGE 111

Figure 3.49 Total Ion Current for 13 Carbonyl Derivative Standard using APCI under SIM (76, 221, 251) monitoring for MEK and Butyraldehyde 151XDJ 1CXXX:O Figure 3.50 Mass Spectra using APCI under SIM 'U 9 1J ::1. 11m I I :i
PAGE 112

Table 3.22 Relative Abundance of Ions using APCI under SIM m.Jz 163 221 251 Ion Abundance Peak at 27.5 minutes Peak at 28.5 minutes 2 15 30 100 94 21 100

PAGE 113

3.5 Limit of Detection 3.5.1 Ionspray Figure 3.51 shows the total ion current for a 1/10 solution by volume ofthe four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. This solution contains the equivalent of a 517 pptv of crotonaldehyde and 529 pptv ofmethacrolein in air as determined previously. The selected ions are the same as before with ionspray and include 68, 149, and 251. Figure 3.52 displays the mass spectra for the two peaks. Table 3.23 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those found with higher concentrations. Crotonaldehyde has the higher rnlz 68 ion and methacrolein has the higher rnlz 149 ion. Figure 3.53 shows the total ion current for the 1/10 solution of the four carbonyl mixture in the SIM mode monitoring for MEK and butyraldehyde. This solution contains the equivalent of a 511 pptv ofMEK and 550 pptv ofbutyraldehyde in air as determined previously. The selected ions are as before and include 70, 138, 193, 236, and 253. Figure 3.54 displays the mass spectra for the two peaks. Table 3.24 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those ofhigher concentrations. 95

PAGE 114

Figure 3.51 Total Ion Current for a 1110 Solution of the 4 Carbonyl Mixture using SIM at 68, 149,251 with Ionspray monitoring for Crotonaldehyde and Metbacrolein MSDI 11:, MS Fie (BCG\ESD00020.D) A.ft.ES Posidn Norm. 8000 eooo 4000 2000 Figure 3.52 Mass Spectra of 1110 Solution with Ionspray "1111Sul o CI ... Gi E.SDOOO:i!O.D A I; t 0 0 -8 1 9 0-00- 0 20--I ... 'o "M S D 1 C I 1m 1 ;, 3 3 tl & < G E5000020.D I. > 10 0-.z s 0 0 0 2 0 81 -, 0 00 '. ... 96

PAGE 115

Table 3.23 Relative Abundances of Selected Ions using Ionspray of 1110 Soluton m/z 68 149 251 Ion Abundance Peak at 21.3 minutes 13 6 100 Peak at 23.3 minutes 11 20 100 Figure 3.53 Total Ion Current for a 1110 Solution of the 4 Carbonyl Mixture using SIM at 70, 138, 193, 236, 253 with Ionspray monitoring for MEK and Butyraldehyde 97

PAGE 116

Figure 3.54 Mass Spectra of 1110 Solution with Ionspray S uf B"'u ESOOu Cl"'.u IE$ PosUiva 1 0 0 0 0 0 -2 0 s' 0 o' M S D 1 S P C lim 3 1 II 0 o I B C G \ t! U 0 0 0 0 l D 1 0 0 0 0 0 " 2 0 e -I -I N_l 00 0 2 0 0 Table 3.24 Relative Abundances of Selected Ions using Ionspray of 1110 Solution Ion Abundance m/z Peak at 21.4 minutes Peak at 23.2 minutes 70 12 3 138 0 12 193 0 7 236 2 8 253 100 100 98

PAGE 117

Figure 3.55 shows the total ion current for a 1/100 solution by volume of the four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. These equivalent air concentrations are 51.7 and 52.9 pptv for crotonaldehyde and methacrolein respectively. The selected ions are the same as before. Figure 3.56 displays the ID3$S spectra for the two peaks. Table 3.25 contains the relative abundances for the selected ions. It is difficult to make a positive identification of the two compounds with these ion fragments. Figure 3.57 shows the total ion current for the 11100 solution of the four carbonyl mixture in the SIM mode monitoring for MEK and butyraldehyde. This solution contains the equivalent of a 51.1 and 55.0 pptv of MEK and butyraldehyde in air. The selected ions are as before. Figure 3.58 displays the mass spectra for the two peaks. Table 3.26 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those of higher concentrations. 99

PAGE 118

Figure 3.55 Total Ion Current for a 11100 Solution of the 4 Carbonyl Mixture using SIM at 68, 149,251 with Ionspray monitoring for Crotonaldehyde and Methacrolein MSDt TC, MS Fie (BCG\ES000021.D) PI-ES PosiliYe Nonn. 2200 2000 1800 1600 1<00 1200 1000 Figure 3.56 Mass Spectra of 11100 Solution with Ions pray M S D 1 t m 1 8 II o f B C 0 \ 1:. g U U U :J: U A P I t; :i 111 1 &II 11 1 I D 0 0 0 o' 0 .... u o I B C 0 \E S 0 0 0 2 I .D A P I E 3 P e I ft IT 1 0 0 I 0 0 -.. 2 0 -... ... 100 n 'J ... m

PAGE 119

Table 3.25 Relative Abundances of Selected Ions using Ionspray of 1/100 Soluton m/z 68 149 251 Ion Abundance Peak at 19.8 minutes Peak at 21.8 minutes 28 43 100 43 79 100 Figure 3.57 Total Ion Current for a 11100 Solution of the 4 Carbonyl Mixture using SIM at 70,138,193,236,253 with Ionspray monitoring for MEK and Butyraldehyde MSDI TC. MS Fie (BCG\!S000004.D) AA-ES PoSIItie Nonn. 101

PAGE 120

Figure 3.58 Mass Spectra of 11100 Solution with Ionspray 8CO\ESOuuu04.0 A Potlllvo I 0 0 I 0 0 0 0 10 C IIIII l l 0 ::. II o tJ C 0 \ r. ;:, 0 0 0 0 4 A c. $ P o 'I II o 1 0 0 0- 0 0 10 -I -I 2 .. "r J., .... Table 3.26 Relative Abundances of Selected Ions using Ionspray of 11100 Solution Ion Abundance m__lL' m/z Peak at 21.3 minutes Peak at 23.1 minutes 70 13 4 138 0 9 193 0 9 236 2 12 253 100 100 102

PAGE 121

Although redundant Figure 3.59 shows the total ion current for a 111000 solution by volume of the four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. These equivalent air concentrations are 5.17 and 5.29 pptv for crotonaldehyde and methacrolein respectively. The selected ions are the same as before. No sign of either compound can be detected among the noise. Thus, when using ionspray a detection limit to distinguish between isomers would be 500 pptv. Figure 3.60 shows the total ion current for the 1/1000 solution of the four carbonyl mixture in the SIM mode monitoring for MEK and butyraldehyde. This solution contains the equivalent of a 5.11 and 5.50 pptv of MEK and butyraldehyde in air. The selected ions are as before and include 76, 221, and 251. At this level the two isomers are not distinguishable as seen by noise. Thus a detection limit for distinguishing between these two isomers would be 50 pptv. 103

PAGE 122

Figure 3.59 Total Ion Current for a 111000 Solution of the 4 Carbonyl Mixture using SIM at 68, 149,251 with Ionspray M:D1 rc. ARSR:Iswo N:rm 2IXIl 11Sl 151Xl 1ZO 1!XXJ 1Sl 5IXl 10 15 20 Figure 3.60 Total Ion Current for a 111000 Solution of the 4 Carbonyl Mixture using SIM at 70, 138, 193,236,253 with Ionspray MDITC Ml Rll (BJ*9TITXIi q MSRisitive N:rm 1D 131il 13D 10 15 2S 3) 104

PAGE 123

3.5.2 Atmospheric Pressure Chemical Ionization Figure 3.61 shows the total ion current for a 1110 solution by volume of the four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. This solution contains the equivalent of a 517 pptv of crotonaldehyde and 529 pptv ofmethacrolein in air as determined previously. The selected ions are the same as before with APCI and include 79, 172, 202, and 249. Figure 3.62 displays the mass spectra for the two peaks. Table 3.27 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those found with higher concentrations. Figure 3.63 shows the total ion current for the 1110 solution of the four carbonyl mixture in the SIM mode monitoring for MEK and butyraldehyde. This solution contains the equivalent of a 511 pptv ofMEK and 550 pptv of butyraldehyde in air as determined previously. The selected ions are as before and include 76, 221, and 251. Figure 3.64 displays the mass spectra for the two peaks. Table 3.28 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those of higher concentrations. 105

PAGE 124

Figure 3.61 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein M501 IC, 110 (toCG4\APCI2CC1.LJ) N'i,;l Nonn 7000 8000 5000 4000 3000 2000 1000 1n ... .;n .; .;n Figure 3.62 Mass Spectra of 1/10 Solution with APCI : .. 9DI ol o A .. ... Su 11 .11101e .1&0 ef BC04\A ''-1 .u A CINIIIe 106 .; I I in 'o t I I" ''"'I c-

PAGE 125

Table 3.27 Relative Abundances of Selected Ions using APCI of 1110 Soluton m/z 79 172 202 249 Ion Abundance Peak at 25.4 minutes 0 20 14 100 Peak at 27.2 minutes 3 13 11 100 Figure 3.63 Total Ion Current for a 1/10 Solution of the 4 Carbonyl Mixture using SIM at 76, 221, 251 with APCI monitoring for MEK and Butyraldehyde .sol Tt:, ""FOe( .O)AFO-o 200000 115000 150000 12SOOO 100000 7!00) 50000 25DCil /'-10 15 20 25 3D :is 40 '"' 107

PAGE 126

Figure 3.64 Mass Spectra of 1110 Solution with APCI lol 1i !;J 5 C 11m a 4 2 II 0 6 c; IIi \A c; Ill I g 1;1 A PC I N e II I IIV a '" 80 0 20 Ill 3 0 I S PC IIIII a a a J o 1:11 C G I \A P C Ill 0 D 4 u A N II' II I IIV '0 0 I 0 0 o 20 I Table 3.28 Relative Abundances of Selected Ions using APCI of 1110 Solution m/z 163 221 251 Ion Abundance Peak at 27.4 minutes 1 31 100 108 Peak at 28.4 minutes 16 22 100

PAGE 127

Figure 3.65 shows the total ion current for a 1/100 by volume solution of the four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. These equivalent air concentrations are 51.7 and 52.9 pptv for crotonaldehyde and methacrolein respectively. The selected ions are the same as before. Figure 3.66 displays the mass spectra for the two peaks. Table 3.29 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those found with higher concentrations. Figure 3.67 shows the total ion current for the 1/100 solution of the four carbonyl mixture in the SIM mode monitoring for l\IIEK and butyraldehyde. This solution contains the equivalent of a 51.1 and 55.0 pptv of l\IIEK and butyraldehyde in air. The selected ions are as before and include 76, 221, and 251. Figure 3.68 displays the mass spectra for the two peaks. Table 3.30 contains the relative abundances for the selected ions. At this level the two isomers are distinguishable with similar results compared to those of higher concentrations. 109

PAGE 128

Figure 3.65 Total Ion Current for a 11100 Solution of the 4 Carbonyl 1.\:fixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein 1400) II 1lDXI 1s Figure 3.66 Mass Spectra of 11100 Solution with APCI "M SPC, It 4 0 C 4\A I"CiiiiiOil A IDIIIVI \0 D .. I 't.l I '1.<, I Ill I :1! :1 5 a G 'iO \1 4 '"' P ; 101 U U ... IJ A PC IN I II I 1,.. I " I 110 I I 1111,..

PAGE 129

Table 3.29 Relative Abundances of Selected Ions using APCI of 11100 Soluton Ion Abundance m/z_ Peak at 25.3 minutes Peak at 27.0 minutes 79 172 202 249 6 18 17 100 12 11 12 100 Figure 3.67 Total Ion Current for a 11100 Solution of the 4 Carbonyl Mixture using SIM at 76, 221, 251 with APCI monitoring for MEK and Butyraldehyde r.ti01 TC, ""Fie (!303MPCJl005.0) Am"""' 20000 17500 15000 12500 10000 7500 5000 2500 ,.._J 5 10 16 20 25 3D 35 40 ... 111

PAGE 130

Figure 3.68 Mass Spectra of 11100 Solution with APCI 'U I S PC I im l :t 5 I o f 11:1 6 \A P C f 0 !i D C I N t i loY I DO Ill 8 D I S P 111ft J I o I B C 0 $ '" C ICI 0 0 u N I I tow I Table 3.30 Relative Abundances of Selected Ions using APCI of 11100 Solution m/z 163 221 251 Ion Abundance Peak at 27.4 minutes 1 27 100 112 Peak at 28.3 minutes 14 19 100

PAGE 131

Figure 3.69 shows the total ion current for a 1/1000 solution by volume ofthe four carbonyl mixture in the SIM mode monitoring for crotonaldehyde and methacrolein. These equivalent air concentrations are 5.17 and 5.29 pptv for crotonaldehyde and methacrolein respectively. The selected ions are the same as before. Figure 3.70 displays the mass spectra for the two peaks. Table 3.31 contains the relative abundances for the selected ions. At this level the two isomers are not distinguishable as seen by the varying abundances for 79, 172 arid 202. Thus, when using APCI a detection limit to distinguish between isomers would be 50 pptv. Figure 3.71 shows the total ion current for the 1/1000 solution of the four carbonyl mixture in the SIM mode monitoring for MEK and butyraldehyde. This solution contains the equivalent of a 5.11 and 5.50 pptv of MEK and butyraldehyde in air. The selected ions are as before and include 76, 221, and 251. Figure 3.72 displays the mass spectra for the two peaks. Table 3.32 contains the relative abundances for the selected ions. At this level the two isomers are not distinguishable as seen by there being no other visible fragments than the molecular ion peak. Thus a detection limit for distinguishing between these two isomers would be 50 pptv. 113

PAGE 132

Figure 3.69 Total Ion Current for a 1/1000 Solution of the 4 Mixture using SIM at 79, 172, 202, and 249 with APCI monitoring for Crotonaldehyde and Methacrolein Figure 3. 70 Mass Spectra of 1/1000 Solution with APCI IIIII I 41 .I II D I 8 C Q \A 1 D 3 ,I) ,. I Ill II I U3UI 5 Dl OCO\A APCIHIJII .... .. a I I I I ... 114

PAGE 133

Table 3.31 Relative Abundances of Selected Ions using APCI of 1/1000 Soluton m/z 79 172 202 249 Ion Abundance Peak at 25.2 minutes 5 0 7 100 Peak at 26.9 minutes 0 1 0 100 Figure 3;71 Total Ion Current for a 1/1000 Solution of the 4 Carbonyl Mixture using SIM at 76, 221, 251 with APCI monitoring for MEK and Butyraldehyde 4000 I 3500 10 15 20 35 40 115

PAGE 134

Figure 3.72 Mass Spectra of 1/1000 Solution with APCI 8 o .. 6\A PCIOOO&.D A PCal\lglh"l 100 .. .. ,, io M 0 1 S PC lim 1 2 I .l J a 8 C 0 8 \A P C 10 0 0 I D A P c I N o II 1 I 1 100 .. .. .. I 100 100 i. Table 3.32 Relative Abundances of Selected Ions using APCI of 1/1000 Solution m/z 163 221 251 Ion Abundance Peak at 27.3 minutes Peak at 28.3 minutes 0 8 1 0 100 100 116

PAGE 135

3.6 Sample Analysis 3.6.1 lonspray Figure 3. 73 displays a chromatogram of a Northeast Denver sample monitoring under full scan conditions. It is evident that no C4 compounds can be detected in this sample using ionspray. Detection ofmethacrolein and crotonaldehyde was accomplished using the SIM mode, but were not detected in the four samples using ionspray. Figure 3.74 shows total ion current chromatograms for the four samples analyzed. These samples were analyzed as before monitoring for ions with the rn!z of 68, 149, and 251.All four samples however, revealed the presence of methyl ethyl ketone and butyraldehyde. Figures 3.75 displays the chromatograms for the four compounds. Table 3.33 shows the relative abundances of the selected ions. Figure 3. 73 Chromatogram of a Sample using Ionspray under Full Scan 117

PAGE 136

Figure 3.74 Chromatograms of Samples 1-4 using lonspray monitoring for Methacrolein and Crotonaldehyde under SIM (68, 149, 251) 2500 2000 1500 500 10 15 20 25 35 3000 2000 1500 1000 500 20 25 3D 3000 2500 2000 .... 1000 500 10 3D 35 ---......... .... 5DDD .... 3DDD 2000 f\ 1000 \. ---_1\J 0 s 10 15 20 .. 30 35 ... 118

PAGE 137

Figure 3. 75 Chromatograms for Samples 1 -4 using Ionspray monitoring for MEK and Butyraldehyde under SIM (70, 138, 193, 236) = .. 2000 11100 .... ... 1200 Nonn .... 3"00 ::11500 2000 1500 ... Norm. 3010 3750 2500 2250 tno 1500 n:so .... 119

PAGE 138

Table 3.33 Relative Abundances of Selected Ions using Ionspray of Samples 1-4 Ion Abundance m/z Peak at 21.4 minutes Peak at 23.0 minutes 70 12, 12, 12, 13 3, 3, 4, 4 138 0, 0, 0, 0 12, 9, 11, 14 193 1, 0, 0, 0 7, 7, 5, 7 236 0,2,2,2 7, 6, 10, 10 253 100, 100, 100, 100 100, 100, 100, 100 3.6.2 Atmospheric Pressure Chemical Ionization Figure 3.76 displays a chromatogram of a sample monitoring under full scan conditions. It is evident that no C4 compounds can be detected in this sample using APCI. Figure 3.76 Chromatogram of a Sample using APCI under Full Scan ..,., n::, MS '" (BWOMf'CDOOBD) A""' ....... Nonn. 3500000 3000000 2500000 2000000 1500000 _\A_L 1000000 L 500000 -a 10 ... "' ,. ,;, is ,;, 120

PAGE 139

Using the SIM mode methacrolein and crotonaldehyde could not be detected in samples 1 and 3. Sample 2 contained peaks at retention times of these two compounds, but the fragment patterns do not conclusively identify either methacrolein or crotonaldehyde. Sample 4 contains a peak with a retention time near methacrolein, but once again the fragment evidence cannot confirm whether this is actually methacrolein. Figure 3.77 shows total ion current chromatograms for the four samples. These samples were analyzed as before monitoring for ions with the rn!z of79, 172, 202, 249. Table 3.34 shows the relative abundances of the peaks from samples 2 and 4. 121

PAGE 140

Figure 3.77 Chromatograms for Samples 1-4 using APCI monitoring for Metbacrolein and Crotonaldehyde under SIM (79,172, 202, 249) MS01 TC. MS Fie APCI ...... 22000 """" '""" '""" 111XD II1IXI Nonn. """' '"""' 125110 111110 7500 50011 M$01 nc, Mi Fila (BCG8\APIXl009.D) APCI Nega1N11 122

PAGE 141

Table 3.34 Relative Abundances of Selected Ions using APCI of Samples 2 and4 mlz 79 172 202 249 Ion Abundance Peak at 25.0 minutes 15 7 18 100 Peak at 26.7 minutes 0,0 10,5 12,9 100, 100 Using SII\.1 to monitor for methyl ethyl ketone and butyraldehyde samples 2-4 clearly showed the presence ofboth compounds. Sample one contains butyraldehyde while the peak in the vicinity ofMEK shows large responses for the selected ions, which therefore makes identification unclear. Figures 3. 78 displays the chromatograms for each sample. Table 3.35 contains the relative abundances for selected ions. 123

PAGE 142

Figure 3. 78 Chromatograms for Sample 1 -4 using APCI monitoring for MEK and Butyraldehyde under SIM (163, 221, 251) Nonn 401100 3SOOO 30000 25000 """" 110000 7DOOO 60000 MS01 TJ::, MS Fie (BCG6\AFOOD07.0) AFO Nega!ive I 400 350 300 250 20Do1 Norm 7DOOO SDOOO 40000 I DODO 124

PAGE 143

Table 3.35 Relative Abundances of Selected Ions using APCI of Samples 1-4 m/z 163 221 251 Ion Abundance Peak at 27.5 minutes Peak at 28.4 minutes 58, 3, 4, 5 38,29,28,30 100, 100, 100, 100 20, 18, 17, 16 20, 19, 19,21 100, 100, 100, 100 Due to the increased sensitivity using APCI compared with ionspray, samples from downtown Denver were analyzed using only APCI. A set of samples from a single spring day were analyzed. The sample set consists of 6 samples and a field blank. Figure 3. 79 displays the chromatograms for these samples monitoring for methacrolein and crotonaldehyde. Table 3.36 contains the relative abundances of ion fragments for each of the two compounds. All samples contained crotonaldehyde and methacrolein which is confirmed by the retention times and fragment patterns. Figure 3.79 Chromatograms for Downtown Denver Air Samples monitoring for Crotonaldehyde and Methacrolein Sample 1 12:00 PM to 4:00 PM 125

PAGE 144

Sample 24:00 PM -8:00 PM D 1C,USFI lA 1305:1.0) A CIN1g11W1 25000 10000 Sample 3 8:00 PM to 12:00 AM :ZSODO Sample 4 12:00 AM to 4:00 AM """ j JOODO 20DDO 5000 126

PAGE 145

Sample 5 4:00 AM to 8:00 AM S Ill I II. :uooo Sample 68:00 AM to 12:00 PM 15000 .. oao Sample 7 Field Blank 127

PAGE 146

Table 3.36 Relative Abundances of Selected Ions in Samples 1-7 monitoring for Crotonaldehyde and Methacrolein Ion Abundance m/z Crotonaldehyde Methacrolein 79 0, 0, 0, 0, 0, 2, 7 5, 5, 11, 8, 6, 5, 15 172 26, 17,22, 16, 18,17,21 11, 11, 10, 10, 9, 11,12 202 23, 12, 11, 14, 13, 14, 5 12, 11, 14, 11, 10, 11, 9 249 Al1100 All 100 Figure 3.80 displays the chromatograms for these samples monitoring for methyl ethyl ketone and butyraldehyde. Table 3.37 contains the relative abundances of ion fragments for each of the two compounds. All samples contained methyl ethyl ketone and butyraldehyde which is confirmed by the chromatography and fragment patterns. Figure 3.80 Chromatograms for Downtown Denver Air Samples monitoring for MEK and Butyraldehyde Sample 1 12:00 PM to 4:00 PM Narrn. 140000 40000 JDOOO 128

PAGE 147

Sample 2 4:00 PM to 8:00 PM """1 100000 411000 Sample 3 8:00 PM to 12:00 AM 100000 :10000 Sample 412:00 AM to 4:00 AM 57.0) APCINigllln 110000 1100110 "0000 129

PAGE 148

Sample 5-4:00 AM to 8:00 AM """ j 175000 1501300 1:1!1000 15000 25000 Sample 6 8:00 AM to 12:00 PM 2000110 150000 1350011 100000 2!000 Sample 7 -Field Blank 111!1DI 11;:, loiS IIIII (1Hou""'PCII081.1.1) APCIN.,glt)\11 000011 20000 130

PAGE 149

Table 3.37 Relative Abundances of Selected Ions in Samples 1-7 monitoring for MEK and Butyraldehyde m.Jz 163 221 251 Ion Abundance MEK 3,2, 1, 1, 1,4,2 30,30,30,29,31,31,28 All100 Butyraldehyde 14, 14, 12, 13, 14, 15 20, 19, 19, 17, 18,19,20 All100 3.6.3 Quantitative Analysis of Sample using APCI Table 3.38 contains the integrated areas of the ion current for the C4 compounds and their concentrations taken from the limit of detection data. Figures 3.81-3.82 contain the calibration curves for the four compounds. Table 3.38 Integrated Areas of the Ion current for the C4 compounds and their Concentrations Crotonaldehyde Methacrolein MEK Butyraldehyde Cone. Area Cone. Area Cone. Area Cone. Area (pptv) (pptv) (pptv) (pptv) 517 2635895 529 2074433 511 5690222 550 3357927 51.7 349621 52.9 235824 51.1 537098 55.0 366613 5.17 51669 5.29 62515 5.11 38301 5.50 22610 131

PAGE 150

Figure 3.81 Calibration Curves for Crotonaldehyde and Methacrolein 3.00Eo-11on(-Lj 132 0.5 0.6 0.1 0.5 0.6 0.1

PAGE 151

i Figure 3.82 Calibration Curves for MEK and Butyraldehyde 7.00E+06 O.OOE+OB 5.00E4{J8 4.00E+il6 3.00E+06 2.00E+CB 1.00E+OO 0.008-00 0 5.00E+06 4.SOE+06 4.00E4o06 3.50E+05 3.00E..c16 2.50E+06 2.00E+06 1.50E+D6 1.0DE+08 5.00(15 O.OOE..OO -S.OOE+D5 ya 9816961.1809r-48889.1667 R'.9999 0.1 02 Calibration Curve lor MEK 0.3 0.4 caw:o .... llon(-L) Calibration Curve for Butyraldohydo 0.1 v= 6448319.2956 oo.3889 0.2 0.3 0.4 0.5 133 OS 0.8 0.6 0.7 0.7 0.8

PAGE 152

Integrated areas of the ion current for the C4 compounds in the samples collected from Northern Denver are given in Table 3.39. Also included are the concentrations of the species detected which are determined by using the calibration data. For an example on how these where calculated see Appendix A. Table 3.39 Integrated Areas of the Ion Current and Concentrations for the .c4 Compounds in the Samples collected from Northern Denver Crotonaldehyde Methacrolein MEK Butyraldehyde Area Cone. Area Cone. Area Cone. Area Cone. (pptv) (pptv) (pptv) (pptv) Sample 1 ND ND ND 813471 18 Sample 2 ND ND 2295130 33 966801 22 Sample 3 ND ND 1229230 18 1027660 23 Sample 4 ND ND 2168800 31 1805080 39 ND =None Detected Integrated areas of the ion for the c4 compounds in the samples collected from downtown Denver are given in Table 3.40. Also included are the concentrations of the species detected that are determined by using the calibration data. It should also be noted that the amount of C4 compounds detected on the field blank (see Figures 3.79 and 3.80 for field blank chromatograms) are subtracted from the amount found in the samples. For an example on how these concentrations where calculated see Appendix A. 134

PAGE 153

Table 3.40 Integrated Areas of the Ion Current and Concentrations for the C4 Compounds in the Samples collected from Downtown Denver Crotonaldehyde Methacrolein MEK Butyraldehyde Area Cone. Area Cone. Area Cone. Area Cone. (pptv) (pptv) (pptv) (pptv) Sample 1 179463 13.0 135056 7.85 3624660 282 2115860 247 12:00 PM4:00PM Sample 2 138442 4.92 111613 1.80 3011360 230 1726770 197 4:00PM-8:00PM Sample 3 172933 11.7 122598 4.63 3277320 253 1703750 195 8:00PM-!2:00AM Sample 4 192268 15.5 181189 19.7 4647950 369 1355780 151 !2:00AM-4:00AM Sample 5 228066 22.5 181963 19.9 5217070 417 1424960 159 4:00AM-8:00AM Sample 6 215035 20.0 221860 30.2 2019660 146 2080310 242 8:00AM-12:00 PM Sample 7 37400 NA 38468 NA 335960 NA 206500 NA Field Blank Figure 3.83 displays graphs that represent the diurnal variation of these four compounds. 135

PAGE 154

Figure 3.83 Diurnal Variation for C4 Compounds from Downtown Denver l: c. c. -c: _0 +: ca ... c: Cl) u c: 0 0 D iu rna I Variation in Carbon y I C om pounds 500.0 l: c. c. 400.0 t:: 0 300.0 ;: ca ... 200.0 t:: Q) 100.0 CJ t:: 0 0.0 0 1 3 5 Four Hour Samples from Table 3.40 -+--C rotonaldehyde ---Methacrolein .........,_MEK ---Butyraldehyde Diu rna I Variation in Carbon y I Compounds 35.0 30.0 25.0 20.0 15.0 10.0 5.0 0.0 1 3 5 Four Hour Samples from Table 3.40 --+-C rotonald ehyd e --Methacrolein (Expansion of Figure above) 136

PAGE 155

3. 7 Further Exploration 3.7.1 Analysis of Acrolein Acrolein was analyzed using APCI to determine how it responded to full scan conditions. The chromatogram and mass spectrum are displayed in figure 3.84. The concentration of the solution is 76.5 ppm or an equivalent air concentration of 66.7 ppbv assuming 720 liters of sampled air. The evidence is clear that acrolein responds well to the given APCI conditions. Figure 3.84 Chromatogram and Mass Spectrum of Acrolein using Full Scan Conditions 1.501 ''"""Reil3lXlAFCD064.D) AfCNegaiWe ...... 1750000 1SOOOOO 12Sa!OO 10011000 750000 saoooo 250000 0 5 10 15 20 25 30 35 4o ,.., -01 SPC.IIme .Z4S o &CO A PCIOOI54.D A t'C Ngtwe I 00 0 I 0 0 D D ;I, l.1.;l1 .. "! 0 D '0 ., "! D j, 1 0 D 0 } 1"1. :: hi -, ISO 2 00. 2 i 0 ,,, 137

PAGE 156

3.7.2 Analysis of Methyl Vinyl Ketone Methyl vinyl ketone is analyzed using APCI with full scan and selected ion monitoring mode. The SIM mode includes the same method and ions used to monitor methyl vinyl ketone's isomers methacrolein and crotonaldehyde. Figure 3.85 displays the chromatogram and mass spectrwn for full scan conditions. Figure 3.86 displays the chromatogram and mass spectrum for the SIM mode looking at the ions with a m/z of79, 172, 202, and 249. This data cannot be compared with Kolliker and co-workers (18) since they did not look at this type of carbonyl compound. However, the SIM data reveal this compound responds differently to the conditions used to look at compounds with a MW of250. Methyl vinyl ketone can be distinguished from the other isomers. Figure 3.85 Chromatogram and Mass Spectrum of MVK using APCI under Full Scan ... ...... '0 0 0 .. : o n 20 e e 0 e 0 ft -II "1. i. 138

PAGE 157

Figure 3.86 Chromatogram and Mass Spectrum of MVK using APCI under SIM Mode (79, 172, 202, 249) :3000000 :nooooo ... SDI 3 0 c; 0008.0 AfiCINIQIIYI 100 11 .ll ll I 4 2 '1 l J 2 I 0 OM 0 00 40 '0 ,,_._ ,., i o_ 1 i_o 200 2;' m I< Table 3.41 Relative Abundances of Selected Ions using APCI for MVK Ion Abundance m/z Methyl Vinyl Ketone 79 0 172 32 202 2 249 100 139

PAGE 158

4. Conclusions The mass spectra for both ionspray and atmospheric pressure chemical ionization demonstrate that mass spectrometry is a useful, complementary method of detection for the 2,4-dinitrophenylhydrazone derivative with HPLC separation. Detection limits in the parts per trillion range for both techniques provide adequate sensitivity for C4 carbonyl compounds. Although ionspray lacks the sensitivity of APCI, the ionspray technique can be useful if laboratories are limited by instrumentation. Due to the increased sensitivity of APCI, this is the technique of choice. All studies found to date using LCIMS are also in agreement that APCI in the negative mode is the best technique. In regards to standards and instrumentation, the work done by Grosjean ai1d coworkers was most similar to the work performed in this study. It appears that parameters for chromatography and mass spectrometry were optimized better by Grosjean and coworkers. In this study, both chromatography and spectrometry were performed under relatively limited time and resources. Because of time limitations, the parameters for the mass spectrometer were chosen according to the suggested parameters of the instrument manufacturer. Unlike in the study perfom1ed by Grosjean and coworkers, only one column was evaluated for chromatography. Because and coworkers had more time and resources available to them, it is fair to say that their study is more thorough in identifying DNPH derivatives and optimizing instrument parameters. However, this study is strengthened by the evaluation of a fragmentor voltage. Choice of a reasonable fragmentor voltage for APCI proved to be critical. For the Hewlett Packard system, this was determined to be 90 volts. Unique and similar ion fragments were revealed for the C4 derivatives as a result. Tlus choice also demonstrated adequate precision from injection to injection for the mass spectrometer. Infommtion obtained from 140

PAGE 159

the fragments allowed identification of the C4 isomers. This is of crucial importance when trying to identifY what compounds are in the atmosphere. With such good detection limits, compoW1ds that were previously not detected can now be obsenred. This was evident in the samples taken from Denver. C4 carbonyls were rarely seen due to their low levels of concentration in the atmosphere, but could be positively identified and quantitated in the samples analyzed in this study. This was especially true for the downtown Denver samples. Identification of such compounds can only help in our understanding of the chemistry of the atmosphere. Quantitation is of even greater importance in achieving this goal. 4.1 Future Considerations Future work should include establishing a sound method for the quantitation of C4 carbonyl derivatives. This can be achieved by a thorough examination into such parameters as accuracy, precision, selectivity, and ruggedness. Not only would C4 carbonyl derivatives be of interest, but how well APCI can identify and quantitate other carbonyl compounds. First, the lower molecul.ar weight compounds would be of interest due to the higher concentrations of these species in the atmosphere. Second, an investigation into the higher molecular weight compounds using APCI should be an. obtainable goal. Other workd should include analyzing more air samples. The results in this paper clearly indicated the presence of C4 carbonyl compounds. A diurnal variation may also exist for these compmmds a11d if so, why? Other interests could include the presence of these compounds at other sites. Therefore, a variety of options exist for further exploration into LC/MS and C4 carbonyl compounds. 141

PAGE 160

Appendix A Determination of Equivalent Air Concentrations for Standards from the Standard Solutions The following calculations are going to be based on the assumption that 720 liters of air has been sampled. Calculation of Actual pressure in Denver Typical pressure for the Denver metro area is 630 mmHg. (1atm/760 mmHg)(630 mmHg) = 0.8289 atm Calculation of number of moles in 720 liters of air PV=nRT Where P =Denver Pressure= 0.8289 atm V = Volume of Air = 720 liters T =Standard Temperature= 298.15 K R =Gas Constant= 0.08206 atm L/ Kmole n1ow = (0.8289 atm)(720 L)/(298.15 K)(0.08206 attn L/ Kmole) = 24.3933 mole Number of moles of ambient air equivalent to 1 ppbv = 2.43933 X 1 o-B 142

PAGE 161

Mass of 1 ppbv derivatized Methacrolein Sample M1 ppbv =nair x Molecular Weight of derivatized methacrolein = 2.43933 X 10-8 X 250 g/mol = 6.098 X 10-6 grams ofmethacrolein If 720 liters of air were sampled onto a cartridge then a 1 ppbv air concentration would contain 6.098 x 1 o-6 grams of methacrolein. The extraction of the contents of the cartridge into a volumetric flask is done with 5.0 mL of acetonitrile and thus result in a concentration of 1.220 x 1 o-6 g/mL. Concentration of Methacrolein Standard Solution 0.0069 g ofmethacrolein/100 mL = 6.49 X 10-5 g/mL (6.49 x 10-5 g/mL)(l ppbv/ 1.220 x 10-6 g/mL) = 52.9 ppbv All other standards are calculated in the same fashion. 143

PAGE 162

Sample Calculation for the Determination of Carbonyl Concentration in Northern Denver Nair= (0.8289 atm)(4320 L)/(298.15 K)(0.08206 atm Ll K mole) = 146.359 mole Number of moles of ambient air equivalent to 1 ppbv 1.46 x 1 o-7 Mtppbv= nair x Molecular Weight of derivatized MEK = 1.46 X 10-7 X 252 g/mol = 3.68 x 10-5 grams ofMEK Solution Concentration 3.68 x 10-5 grams ofMEK/5mL = 7.36 X 10-6 g/mL Sample #2 MEK (2295130 + 46889)/(9616961) = 0.24 [(0.24 ug/mL)(1 pptv/7.36 x 10-9 g/mL)](1g/1 x 106ug) = 32.6 pptv 144

PAGE 163

Sample Calculation for Downtown Denver Sample Sample #1 Crotonaldehyde 179463 37400 (blank)= 142063 (142063 -76062)/ 4163010 = 0.0159 ug/mL ((0.0159 ug/mL)(l pptv/1.22 X 10-9 g/mL)](lg/1 X 106ug) = 13.0 pptv 145

PAGE 164

References (1) Shepson, P.B., Sirju, A.P., Environ. Sci. Tech., 29, 384-392 (1995). (2) Charles, M.J., Christman, R.F., Le Lacheur, R.M., Singer, P.C., Sonnenberg, L.B., Environ. Sci. Tech., 27, 2745-2753 (1995). (3) Altshuller, A.P., Atmospheric Environment, 27A, 21-32 (1993). (4) Carlier, P., Hannachi, H., Mouvier, G., Atmospheric Environment, 20, 20792099 (1986). (5) Anderson L.G., Barrell, R., Jones, R.H., Lanning, J.A., Miyagishima, J., Wolfe, P., Atmospheric Environment, 30, 2113-2123 (1996). (6) Claxton, L.D., Graedel, T.E., Hawkins, D.T., Atmospheric Chemical Compounds: Sources, Occurrence, and Bioassay, (Academic Press, 1986). (7) Atkinson, R., Tuazon, E.C., International Journal of Chemical Kinetics, 22, 1221-1236 (1990). (8) Aschmann, S.M., Atkinson, R., Tuazon, E.C., Environ Sci Tech., 29, 18601866 (1995). (9) Harris, D.C., Quantitative Chemical Analysis (W.C. Freeman, New York, (1999). (10) Holler, J.F., Nieman, T.A., Skoog, S.A., Principles of Instrumental Analysis (Saunders, Fort Worth, 1998). (11) Niessen W.N.A., Tinke A.P., Journal of Chromatography A, 703, 37-57 (1995). (12) Voyksner, R.D., Environ. Sci. Tech., 28, 118A-127A (1994). (13) Bruins, A.P., Trends in Analytical Chemistry, 13, 37-43 (1994). (14) Canboy, J.J., Henion, J.D., Huang, E.C., Wachs, T., Analytical Chemistry, 62, 713A-725A (1990). 146

PAGE 165

(15)Bruker, Hewlett Packard EsquireLC Operations Manual, Version 1.4, 13-9 (1999). (16) Silverstein, R.M., Webster, F.X., Spectrometric Identification of Organic Compounds (Wiley & Sons, Inc., New York, 1998). (17) Grosjean, D., Analytical Chemistry, 55, 2436-2439 (1983). (18) Olson, K.L, Swarin, S.J., Journal of Chromatography, 333, 337-347 (1985). (19) Grosjean, D., Grosjean E., Intern. J. Environ. Anal. Chem., 61,47-64 (1995). (20) Dye, C., Kolliker, S., Oehme, M., Analytical Chemistry, 70, 1979-1985 (1998). (21) Karst, U., Luftmann, H., Zurek, G., Analyst, 124, 1291-1295 (1999). (22) Green, P. G., Grosjean D., Grosjean E., Analytical Chemistry, 71, 1851-1861 (1999). (23) Green, P. G., Grosjean D., Grosjean E., Hughes, J.H., Agilent Technologies, Application Note (2000). 147