EPA-650/3-74-011



AUGUST 1974
                                     Ecological  Research  Series

-------
                                     EPA-650/3-74-011
CHEMICAL  CHARACTERIZATION
      OF  MODEL  AEROSOLS
                    by

               Warren Schwartz

            Battelle Memorial Institute
               505 King Avenue
              Columbus, Ohio 43201
                Grant No. 801174
              ROAP No. 21AKB-08
            Program Element No .  1AA008
        EPA Project Officer:  Ronald Patterson

          Chemistry and Physics Laboratory
       National Environmental Research Center
     Research Triangle Park, North Carolina 27711
                 Prepared for

       OFFICE OF RESEARCH AND DEVELOPMENT
      U.S. ENVIRONMENTAL PROTECTION AGENCY
            WASHINGTON, D .C.  20460

                 August 1974

-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication.  Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names ur commercial products constitute
endorsement or recommendation for use.
                                  IX

-------
                                 ABSTRACT
Model aerosols were generated from individual hydrocarbon precursors
and nitrogen oxides under simulated atmospheric conditions in a
17.3 m  environmental chamber.    Hydrocarbon precursors employed
were toluene, alpha-pinene, and cyclohexene.  Aerosols were collected
on glass-fiber filters and organic matter was obtained by solvent
extraction.  Organic reaction products were fractionated into acid,
neutral, and basic components, and were analyzed in detail by tech-
niques including gas chromatography and gas chromatography combined
with mass spectrometry.   The study also included evaluation of
techniques for selective derivatization of aerosol products.
A variety of polyfunctional reaction products were identified, in-
cluding alcohols, aldehydes, ketones, carboxylic acids, and phenols,
Tentative identification of nitrogen containing products was also
accomplished, including nitrate esters and aromatic nitro compounds.

This report was submitted in fulfillment of Grant Number R-801174
by the Columbus Laboratories of Battelle Memorial Institute under
the sponsorship of the Environmental Protection Agency.  Work was
completed as of April 30, 1974.
                                ill

-------
                                 CONTENTS
ABSTRACT .......................

LIST OF FIGURES ...................         vi
LIST OF TABLES ....................         xi
SECTIONS

I   SUMMARY AND CONCLUSIONS .............          1
II  BACKGROUND AND INTRODUCTION ...........          5
III RATIONALE AND METHODS OF PROCEDURE ........          7
     Selection of Hydrocarbon Precursors .......          8
     Generation of Model Aerosols ..........          9
          Smog-Chamber Characteristics ........         10
          Smog-Chamber Cleanup ............         11
          Aerosol Generation, Monitoring,  and
           Collection ................         12

     Extraction of Aerosol Products .........         12

          Extraction Procedures ...........         13

     Fractionation of Aerosol Extracts ........         14
          Fractionation Procedures ..........         15

     Derivatization of Aerosol Products .......         15

          Preparation of Dithiolane Derivatives ...         IQ
          Preparation of Perf luoroester Derivatives .         19
          Preparation of Phosphate Derivatives. ...         19

     Analysis of Aerosol Products by Gas
      Chromatography and Mass Spectrometry ......         20

          Gas Chroma to graphic and Mass Spectral
           Procedures .  ...............         23
IV  RESULTS AND DISCUSSION  .....  ........         24

     Smog  Chamber Irradiations ............         24
     Extraction of Aerosol Products .........         24
     Fractionation of Aerosol Extracts ........         28
     Derivatization of Aerosol Products .......         29
                               IV

-------
                                 CONTENTS

                                                             Page

          Dithiolane Derivatives of Authentic Carbonyl
           Compounds	      28
          Dithiolane Derivatives of crPinene and
           Cyclohexene Aerosol Products 	      31
          Perfluoroacyl and Phosphate Derivatives of
           Authentic Alcohols 	      41
          Phosphate Derivatives of crPinene Aerosol
           Products	      44
     Analysis of Cyclohexene Aerosol	      50
          Cyclohexene Aerosol Acid Fraction 	      50
          Cyclohexene Aerosol Neutral Fraction	      57
          Cyclohexene Aerosol Basic Fraction	      65
     Analysis of Toluene Aerosol	      65
          Toluene Aerosol Acid Fraction 	      65
          Toluene Aerosol Neutral Fraction	      74
     Analysis of cr-Pinene Aerosol	      79
          cr-Pinene Aerosol Acid Fraction	      81
          crPinene Aerosol Neutral Fraction 	      90

               Studies Based on Known Reactions of
                a-Pinene	      94
               Tentative Assignments for orPinene
                Neutral Products. ... 	      109
          crPinene Basic Fraction 	      116
REFERENCES	      125

-------
                                 FIGURES

 No.                                                               paSe

 1   Gas Chromatograms of Authentic Dithiolanes                     32
 2   Reconstructed Gas Chromatograms of Authentic Dithiolanes        33
 3   Chemical lonization Mass Spectra (Methane) of Authentic
     Dithiolanes                                                    34
 4   Chemical lonization Mass Spectra (Methane) of Authentic
     Dithiolanes                                                    35

 5   Chemical lonization Mass Spectra (Methane) of Authentic
     Dithiolanes                                                    36

 6.   Analysis of Cyclohexene Aerosol, Neutral Fraction Dithiolane
     Derivatives.   Reconstructed Gas Chromatogram for Methane
     CT-MS                                                          38
 7   Analysis of ^-Pinene Aerosol,  Neutral Fraction Dithiolane
     Derivatives.   Reconstructed Gas Chromatogram for Methane
     CI-MS                                                          39
 8   Analysis of Cyclohexene Aerosol, Neutral Fraction Dithiolane
     Derivatives.   Chemical lonization Mass Spectrum (Methane)
     for Chromatographic Peak A.  (Refer to Chromatograms in
     Figure 6)                                                      40

 9   Analysis of Cyclohexene Aerosol, Neutral Fraction Dithiolane
     Derivatives.   Chemical lonization Mass Spectrum (Methane)
     for Chromatographic Peak B.  (Refer to Chromatogram in
     Figure 6)                                                      43
10   Reconstructed Gas Chromatogram of Diethylphosphate
     Derivatives of Authentic Alcohols                              45

11   Chemical lonization Mass Spectra (Methane) of
     Diethylphosphate Derivatives of Authentic Alcohols             46

12   Chemical lonization Mass Spectra (Methane) of
     Diethylphosphate Derivatives of Authentic Alcohols             47
13   Analysis of ry-Pinene Aerosol,  Neutral Fraction
     Diethylphosphate Derivatives.   Reconstructed Gas Chromato-
     gram for Methane CI-MS                                         48
14   Analysis of ry-Pinene Aerosol,  Neutral Fraction Diethylphos-
     phate Derivatives.   Chemical lonization Mass Spectrum
     (Methane) for Chromatographic  Peaks A and B.   (Refer to
     Chromatogram in Figure 13)                                      49
                                  VI

-------
                                FIGURES
                               (Continued)
No.
 15   Analysis of Cyclohexene Aerosol, Acid Fraction Methyl
     Esters.  Reconstructed Gas Chromatogram for Methane
     CI-MS                                                        51
 16   Analysis of Cyclohexene Aerosol, Acid Fraction Methyl
     Esters.  Chemical lonization Mass Spectrum (Methane)
     for Chromatographic Peak C.  (Refer to Chromatogram in
     Figure 15)                                                   52

 17   Analysis of Cyclohexene Aerosol, Acid Fraction Methyl
     Esters.  Chemical lonization Mass Spectrum (Methane)
     for Chromatographic Peak A.  (Refer to Chromatogram in
     Figure 15)                                                   53

 18   Analysis of Cyclohexene Aerosol, Acid Fraction Methyl
     Esters.  Chrmical lonization Mass Spectrum (Methane)
     for Chromatographic Peak B.  (Refer to Chromatogram in
     Figure 15)                                                   54

 19   Analysis of Cyclohexene Aerosol, Acid Fraction Methyl
     Esters.  Chemical lonization Mass Spectrum (Methane)
     for Chromatographic Peak D.  (Refer to Chromatograms in
     Figure 15)                                                   56

 20   Analysis of Cyclohexene Aerosol, Neutral Fraction.
     Reconstructed Gas Chromatogram for Methane CI-MS             58

 21   Analysis of Cyclohexene Aerosol, Neutral Fraction.
     Reconstructed Gas Chromatogram for Helium CI-MS              59

 22   Analysis of Cyclohexene Aerosol, Neutral Fraction.
     Chemical lonization Mass Spectra for Chromatographic
     Peak A.  Spectrum I  by Methane CI-MS, Spectrum 2 by
     Helium CI-MS.   (Refer to Chromatogram in Figure 21)          61

23   Analysis of Cyclohexene Aerosol, Neutral Fraction.
     Chemical lonization Mass Spectra for Chromatographic
     Peak B.  Spectrum 1 by Methane CI-MS, Spectrum 2 by
     Helium CI-MS.   (Refer to Chromatogram in Figure 21)          63
24   Analysis of Cyclohexene Aerosol, Neutral Fraction,
     Chemical lonization Mass Spectra for Chromatographic
     Peak D.  Spectrum 1 by Methane CI-MS, Spectrum 2 by
     Helium CI-MS.   (Refer to Chromatogram in Figure 21)          64

25   Analysis of Toluene  Aerosol,  Acid Fraction Methyl Esters/
     Ethers.  Reconstructed Gas  Chromatogram for Methane CI-MS     66
                                 vn

-------
                                FIGURES
                              (Continued)
No.
26   Analysis of Toluene Aerosol,  Acid Fraction Methyl Esters/
     Ethers.  Chemical lonization Mass Spectrum (Methane)  for
     Chromatographic Peak B.  (Refer to Chromatogram in
     Figure 25)                                                    67

27   Analysis of Toluene Aerosol,  Acid Fraction Methyl Esters/
     Ether.  Chemical lonization Mass Spectrum (Methane)  for
     Chromatographic Peak A.  (Refer to Chromatogram in
     Figure 25)                                                    70
28   Analysis of Toluene Aerosol,  Acid Fraction Methyl Esters/
     Ethers.  Chemical lonization Mass Spectrum (Methane)  for
     Chromatographic Peak C.  (Refer to Chromatogram in
     Figure 25)                                                    71
29   Analysis of Toluene Aerosol,  Acid Fraction Methyl Esters/
     Ethers.  Chemical lonization Mass Spectrum (Methane)  for
     Chromatographic Peak D.  (Refer to Chromatogram in
     Figure 25)                                                    72
30   Analysis of Toluene Aerosol,  Acid Fraction Methyl Esters/
     Ethers.  Chemical lonization Mass Spectrum (Methane)  for
     Chromatographic Peak E.  (Refer to Chromatogram in
     Figure 25)                                                    73

31   Analysis of Toluene Aerosol,  Neutral Fraction.  Recon-
     structed Gas Chromatogram for Methane CI-MS                   75

32   Analysis of Toluene Aerosol,  Neutral Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak B.  (Refer to Chromatogram in Figure 31)                 76

33   Analysis of Toluene Aerosol,  Neutral Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak F.  (Refer to Chromatogram in Figure 31)                 77

34   Analysis of Toluene Aerosol,  Neutral Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak A.  (Refer to Chromatogram in Figure 31)                 78

35   Analysis of Toluene Aerosol,  Neutral Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak E.  (Refer to Chromatogram in Figure 31)                 80

36   Analysis of o'-Pinene Aerosol, Acid Fraction Methyl Esters.
     Reconstructed Gas Chromatogram for Methane CI-MS              82
                                 vi 11

-------
                                FIGURES
                              (Continued)
No.
37   Analysis of a-Pinene Aerosol, Acid Fraction Methyl Esters.
     Chemical lonization Mass Spectrum (Methane) Corresponding
     to Methyl Pinonate.  (Refer to Chromatogram in Figure 36)     83
38   Chemical lonization Mass Spectrum (Methane) of Authentic
     Methyl cis-Pinonate                                           84
39   Analysis of cv-Pinene Aerosol, Acid Fraction Methyl Esters.
     Reconstructed Gas Chromatogram for Methane CI-MS              87
40   Analysis of o/-Pinene Aerosol, Acid Fraction Methyl Esters.
     Chemical lonization Mass Spectrum (Methane) for Chromato-
     graphic Peak A.  (Refer to Chromatogram in Figure 39)         88

41   Analysis of Blue Ridge Mountain Aerosol,  Acid Fraction
     Methyl Esters.  Reconstructed Gas Chromatogram for Methane
     CI-MS                                                         91

42   Analysis of Blue Ridge Mountain Aerosol,  Acid Fraction
     Methyl Esters.  Chemical lonization Mass  Spectrum (Methane)
     Corresponding to Pinonic Acid.  (Refer to Chromatogram in
     Figure 41)                                                    92

43   Analysis of a-Pinene Aerosol, Neutral Fraction.  Recon-
     structed Gas Chromatogram for Methane CI-MS                   95

44   Reconstructed Gas Chromatograms of Authentic Terpenoids       98

45   Chemical lonization Mass Spectra (Methane) of Authentic
     Limonene and Terpinolene                                      99

46   Chemical lonization Mass Spectra (Methane) of Authentic
     o-Fenchyl Alcohol and Borneol                                100
47   Chemical lonization Mass Spectra (Methane) of Authentic
     crTerpinol and Myrtenal                                      101

48   Chemical lonization Mass Spectra (Methane) of Authentic
     Terpin Hydrate and Fenchone                                  102
49   Chemical lonization Mass Spectra (Methane) of Authentic
     trans-Verbenol                                               103
50   Chemical lonization Mass Spectra (Methane) of Myrtenol
     and Carveol                                                  104

51   Analysis of a-Pinene Aerosol, Neutral Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak E.  (Refer to Chromatogram in Figure 43)                106
                                  IX

-------
                                FIGURES
                              (Continued)
No.
52   Analysis of crPinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chroraatographic
     Peak C (Refer to Chromatogram in Figure 43)                   107

53   Analysis of a-Pinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak B.  (Refer to Chromatogram in Figure 43)                 108

54   Analysis of a-Pinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak A.  (Refer to Chromatogram in Figure 43)                 110

55   Analysis of a-Pinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak F.  (Refer to Chromatogram in Figure 43)                 114

56   Analysis of a-Pinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak G.  (Refer to Chromatogram in Figure 43)                 116

57   Analysis of rrPinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak D.  (Refer to Chromatogram in Figure 43)                 117

58   Analysis of a-Pinene Aerosol, Neutral Fraction.   Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak H.  (Refer to Chromatogram in Figure 43)                 118

59   Analysis of a-Pinene Aerosol, Basic Fraction.  Reconstructed
     Gas Chroroatogram for Methane CI-MS                            120

60   Analysis of a-Pinene Aerosol, Basic Fraction.  Chemical lon-
     ization Mass Spectrum (Methane) for Chromatographic Peak A.
     (Refer to Chromatogram in Figure 59)                          121

61   Analysis of crPinene Aerosol, Basic Fraction.  Chemical
     lonization Mass Spectra for Chromatographic Peak B
     Spectrum 1 by Methane CI-MS, Spectrum 2 by Helium CI-MS.
     (Refer to Chromatogram in Figure 59)                          123

62   Analysis of a-Pinene Aerosol, Basic Fraction.  Chemical
     lonization Mass Spectrum (Methane) for Chromatographic
     Peak C.  (Refer to Chromatogram in Figure 59)                 124

-------
                                 TABLES

No.                                                              Page
1    Summary of Irradiation Conditions and Smog
     Parameters                                                   25
2    Solvent Extraction of Collected Aerosol                       26
3    Fractionation of Methylene Chloride Extractable Matter        29
4    Rationalized Fragmentation of Authentic  Methyl
     Pinonate Under Conditions of Chemical lonization
     Mass Spectrometry (Methane)                                  35

5    Rationalized Fragmentation of Tentatively  Identified
     Methyl Pinononate Under Conditions  of Chemical
     lonization Mass Spectrometry (Methane)                        89
                                      XI

-------
                                 SECTION I
                          SUMMARY AND CONCLUSIONS

This study has been directed toward detailed organic chemical char-
acterization of aerosols generated from individual hydrocarbon pre-
cursors under simulated atmospheric conditions.  Model aerosols
were generated using the Battelle-Columbus Laboratories' 17.3 m
environmental chamber.   Systems studied during this program include
cyclohexene/NOx, toluene/NOx, and o/-pinene/NOx.  Aerosol was collect-
ed on glass-fiber filters and organic matter was obtained by sequential
solvent extraction.  Reaction products were fractionated into acid,
neutral, and basic components and were analyzed in detail by techniques
including gas chromatography and gas chromatography combined with
mass spectrometry (GC-MS).   Additionally, techniques were evaluated
for the selective derivatization of aerosol products.  This was
directed at "tagging" specific  functional groups with easily
detectable markers.  Overall, the objective of the analytical effort
was the identification of individual reaction products formed from
each hydrocarbon precursor.  By comparing the structure of such
products with the parent hydrocarbon, and with each other, we may
infer the types of organic reactions occurring under atmospheric
conditions.

Tentative identification of a variety of aerosol products has been
accomplished.  The results of this study are best summarized by
reference to these products.  In considering the structures shown
below it should be emphasized that these are tentative identifications.

-------
 Cyclohexene aerosol.
                               COOH

                               CHO
                            COOH

                            CHO
COOH

CH2OH
                        COOH
                        p  o
TENTATIVE     CH3-CH =CH -C -C -(
       CYCLOHEXENE
                              CHO
                          CHO
                                                     HO-CH2-CH2-CH=CH-CH=CH2
                                                  or
                                                             CHO
                                                               or
                     CHO ~l
Toluene  aerosol.
       CH,
      O
             TENTATIVE
                            COOH
                             .CH2OH
                            CH3
     TOLUENE
                        COOH         CH2OH
                  OH
        (TWO  ISOMERS)
                                             CH,
CHO
 OR
                                               HO
                            CH
                                                     CHO

-------
a-Pinene aerosol.
                               COOH
          TENTATIVE
 a-PINENE
                                           COOH
                                        CHO
                                                                    CHO
In view of the reaction conditions, formation of many of the above
aerosol products  is not particularly surprising.  Oxidative cleavage
of carbon-carbon  double bonds is to be expected in an ozone con-
taining atmosphere.  Oxidative decarboxylation and formation of
alcohols, carboxylic acids and carbonyl compounds is consistent
with  free radical processes expected in the atmosphere.  Neverthe-
less, exnerimental demonstration of the formation of such products
should be of value.
The  fate of nitrogen oxides  in  the atmosphere has  long been a ques-
 tion of interest.   In  this  context,  the current  research has  included
identification of two  classes of nitrogen-containing  products.
Nitrate esters were tentatively identified  in cyclohexene aerosol,
and  aromatic nitro compounds were tentatively identified  in toluene
aerosol.  This infers  that nitration of  an  aromatic ring  is occur-
ring  under atmospheric simulation.  Identification of such nitrate
esters and nitropenols is of interest  from  the standpoints of both
fundamental atmospheric chemistry and  health effects.

-------
In studying model aerosols,  it is often possible to characterize
reaction products more easily than if these products were present
in the more complex mixture  of atmospheric organics.  Moreover,
identification of interesting compound types in model aerosol suggests
that such compounds may be present in atmospheric aerosol.   For
example, during earlier phases of this research  the formation of
5-nitratopentanoic acid was  reported.  Subsequently, such nitrate
esters were identified in atmospheric aerosol during elegent mass
                                                                (1)**
spectral studies of Schuetzle, Crittenden, Cronn, and Charleson
Thus, characterization of model aerosols is complementary to and
supportive of research directed toward characterization of atmospheric
aerosol.

An ultimate objective of model aerosol research is  the development
of a comprehensive basis for predicting the types of products that
can be expected to form when specified organic compounds are re-
leased into the atmosphere.   Needless to say, this  objective remains
quite distant.  It is hoped  that the results of this study have con-
tributed to the attainment of that goal.
 *See Background and Introduction section.
**References are listed on page 125.

-------
                                SECTION II
                        BACKGROUND AND INTRODUCTION

It is by now well established that the aerosols which form as a
result of photochemical smog reactions are  composed predominantly
of organic matter.  Because of the complexity of smog formation
little is known of the pathways by which organic vapors undergo
chemical change to become aerosols.  The ultimate objective of this
research program is the elucidation of organic chemical reactions
involved in such aerosol formation.  An understanding of these pro-
cesses would allow us to predict the types  of aerosol products that
would be formed from a particular pollutant.  Moreover, a comprehensive
understanding of such reactions might allow us to assign sources con-
tributing to the atmospheric hydrocarbon burden.

Analysis of atmospheric particulate matter reveals a complex mixture
of organic components.  Similarly, the organic vapors emitted  into
the atmosphere, the compounds that fuel photochemical aerosol  form-
ation, also represent a highly complex mixture.  It is not surprising,
therefore, that it has been difficult, if not  impossible, to corre-
late observed components of atmospheric particulate with specific
organic pollutants or well-defined organic reactions.  In view of
this complexity, thi;i research has been based  on the study of  simple
and controllable model aerosol systems.

Many model studies have examined  the  factors  influencing aerosol
formation, and  the time course of  this phenomenon.  These
studiesC^-^/J have considered such parameters  as NOX concentra-
tion, S02 concentration, relative humidity, and stirring of
reactants within the  chambe".  The aerosol-forming  tendency

-------
                                                    (28 29)
of various hydrocarbons has similarly been evaluated.   '       Research
concerned with the mechanism of aerosol forme
                                            i
microscopic examination of aerosol particles
concerned with the mechanism of aerosol formation has involved electron
                                            (30,31)
In contrast to such studies, the research reported herein has been
directed toward characterization of specific aerosol reaction products
formed from individual hydrocarbons.  Few studies have involved de-
                                     / 1  Q O 1) \
tailed characterization of this type.  '         A notable exception
is the model aerosol research conducted during the first year of the
                                               (34) *
program "Haze Formation--Its Nature and Origin"     .   This program
sponsored jointly by the Environmental Protection Agency (CPA 70-Neg
172) and the Coordinating Research Council (CAPA 6-68) was the direct
predecessor of the current study.   For the sake of clarity and
completeness, salient aspects of that study are included in this
report.
"Research  concerning composition  of model aerosols conducted during
  the  "Haze Formation"  study was supported in part by Grant AP-00828
  from the  Office of Research Grants, EPA.  The principal  investigator
  was  William  E. Wilson.

-------
                                SECTION III
                    RATIONALE AND METHODS OF PROCEDURE

Despite the reduced complexity of model aerosols compared to atmospheric
aerosol, the reaction mixtures generated during atmospheric simulation
are indeed complex.  Thus, the analytical scheme employed was designed
to reduce this complexity and simplify characterization of individual
reaction products.

During the first year of this study, emphasis has been placed on
characterization of aerosol products that have not undergone ex-
tensive oxidation or degradation.  Identification of such compounds
is profitable in that the products characterized can be more easily
related to the parent hydrocarbon.  On the basis of this relationship
we may infer the types of reactions occurring under atmospheric
simulation.   The extraction and fractionation procedures described
below were designed to separate the reaction mixture into defined
components with varying degrees of oxidation/degradation. Detailed
analyses were conducted using the less degraded fractions of the
reaction mixture.

It is recognized that many interesting and significant products may
be present in the more highly oxidized fractions.  Indeed, such pro-
ducts are more abundant than the less degraded components.  Although
the first year's study has emphasized characterization of the less
oxidized reaction components, future work vill include analysis of the
more highly oxidized materials.

Under the headings below, further rationale concerning the experi-
mental approach is discussed, and methods of procedure are described.

-------
SELECTION OF HYDROCARBON PRECURSORS
Toluene, o-pinene,  and cyclohexene were selected as precursors for
the generation of model aerosols.   Selection was based on a number
of criteria.  First, it was desired to include hydrocarbons of various
classes, and where feasible, to select members of these classes which
represent important contributors to the atmospheric hydrocarbon burden,
Also taken into account was the aerosol reactivity of candidate hydro-
        (35)
carbons    .  Hydrocarbons having high aerosol reactivity were favored
in view of the need to obtain sufficient aerosol mass for analysis.
All hydrocarbons studied to date in this program are active aerosol
formers.  Future work will include compounds of more limited reac-
tivity, such as branched alkanes.

Inspection of early morning hydrocarbon distribution data in both
East     and West^  ' Coast cities indicates that aromatic hydro-
carbons comprise between 30 and 40 percent of the nonmethane hydro-
carbon.  Of these, toluene is the most abundant, comprising approxi-
mately  25-40 percent of the total aromatics.  Study of this compound
permits us to observe the behavior of both an aromatic ring, and an
alkyl substituent under conditions of atmospheric simulation.

It has  been reported that approximately 175 million tons of reactive
hydrocarbons are released into the atmosphere annually from tree
foliage sources^38*39).  This is over sixfold greater than the
estimated 27 million tons released from technological sources.  The
major compounds released by plants and trees are the monoterpenes
(CIQ),  and the semiterpene isoprene (65).   It has been speculated
that the natural "blue haze" observed in many forested regions arises
from the atmospheric reactions of these compounds.  o"Pinene, chosen
for use in this study, is one of the principal monoterpenes emitted
                   (38,39)
from these sources

-------
Selection of cyclohexene was made on a somewhat broader basis.  The
above-cited studies of early morning hydrocarbon distribution in
East and West Coast cities indicates that olefins comprise between
10 and 20 percent of the total nonmethane hydrocarbon.  Olefins
having four or more carbon atoms comprise about 25 percent of the
total olefin fraction.  Smog chamber studies     have demonstrated
that such olefins have considerable propensity to form organic
aerosols, the aerosol-forming tendency increasing with the molecular
weight of the olefin.  Although cyclohexene is rarely detected in
urban atmospheres, this compound is known to be a particularly prolific
aerosol precursor.  This stems in part from the fact that cleavage
of the cyclohexene double bond leads to formation of bifunctional
oxidation products.  Compared to monofunctional products formed from
acyclic olefins, these bifunctional products have decreased volatility
and greater tendency to enter the aerosol phase.   In view of the
desireability of obtaining high aerosol mass loadings, cyclohexene
was selected for study.

GENERATION OF MODEL AEROSOLS
                                                                 3
Model aerosols were generated using the Battelle-Columbus  17.3  m
environmental chamber.  In selecting the conditions for aerosol genera-
tion, it was necessary to strike a compromise between the desire to
operate at the relatively low reactant concentrations typically observed
in authentic atmosphere,  and the need to provide sufficient aerosol mass
for organic analysis.  Typically, -10 ppm (vol/vol) hydrocarbon was used.
Nitrogen oxide concentrations ranging from 2-5 ppm (NO + N0£) were
employed, and were selected and adjusted to obtain reaction kinetics
typical of photochemical smog manifestations.   Preliminary runs were
conducted to determine suitable reactant concentrations for each aerosol
system studied.

-------
The reactant concentrations employed are at least an order of
magnitude greater than atmospheric levels.  The effect of the higher
reactant concentrations on the aerosol chemistry is uncertain.  In
general, however, most of the chemical reactions responsible  for
smog  formation are first order in•reactants, and the first encounter
an intermediate has with another potentially reactive species is
nearly always with oxygen molecules.  Therefore, only in instances
where the overall speed of radical-radical  reactions approaches that
of radical-02 reactions will the overall reaction pathways of inter-
mediates be seriously affected by the higher concentration of react-
ants.  Judging from the small fraction of reactions meeting this
condition, it is believed that the concentrations employed did not
result in a serious departure from normal smog chemistry.  This
argument, of course, is oversimplified; however, until a detailed
understanding of aerosol reaction mechanisms is developed further,
speculation on the importance of concentration in these simulations
is unwarranted.
Smog Chamber Characteristics
The smog chamber utilized in these studies  is approximately 2.4 m
high, 4.9 m long, and  1.5 m wide, having a volume of approximately
17.3 m .  The surface-to-volume ratio is 0.78.   The inside surface
                   9                                 9
consists of  35.3 m   of polished aluminum and 9.2 ra   of   FEP-
Teflon windows (5 mil thick) through which  the reaction mixture
is irradiated using an external bank of lamps.   The lamp bank
consists of 96 fluorescent "black" lamps and 15 fluorescent
sunlamps.  The spectral distribution of the black lamps peaks in
intensity at  370 nm;  the sunlamps peak intensity occurs at 310 nm.
Light intensity  generated corresponds to a k< of 0.45 min   as deter-
mined by NOo photolysis and o-nitrobenzaldehyde actinometry.   This
intensity is comparable to that of noonday sun over the wavelength
integral of NC>2  absorption.
                                    10

-------
 Air supplied  to the  chamber is  taken  in through  a 10-m stack atop
 a  three-story building,  and is  passed through a  purification system
 including a  permanganate filter bed,  a charcoal  filter system,  an
 absolute  filter,  and a  humidification unit.   After  purification,
 background total-hydrocarbon is generally  1-3 ppmC  with  the  majority
 of  this being methane.   A trace of  background ethane  is  occasionally,
 observed  by  gas chromatography.   Olefins and  aromatics have
 always been  below gas chromatographic detection  limits.   Sulfur
 dioxide is always below  the detection limit of a  flame photometric
 detection (~10 ppb)  and carbon monoxide background levels are
 generally between 1  and  3  ppm as  measured  by  nondispersive infrared
 spectroscopy.   Light scattering readings are  essentially  the  same
 as  for clean  air  (Raleigh  scattering)  and  the condensation nuclei
 count is  generally less  than 200  particles per cubic  centimeter.
 Smog-Chamber  Cleanup
 Prior to  each  series of  experiments with a different  model hydro-
 carbon, it was  necessary to  thoroughly  clean  the  chamber's surface
 to  prevent cross  contamination  of aerosol  products.   Such cross
                                                        / f\ I \
 contamination  had been observed during  previous studies    .  Cleanup
 was accomplished  by  washing  down  chamber surfaces with a spray  of
 4:1 isopropanol:water.   Prewashed cloths were  used to  scrub  the sur-
 faces and  surgical scrub suits were worn to prevent contamination
 from street clothing.  Throughout the procedure,   the chamber was main-
 tained on purge (8500 1 min   ).  Nevertheless, the high concentration
of solvent vapor that develops, requires that the chamber be sealed
 and that  the worker  be provided with an  auxiliary breathing apparatus.

After cleanup,  the chamber was purged overnight and was then per-
mitted to stand with  1  ppm ozone for several hours.  Before genera-
 tion of aerosol for  analytical use, a "conditioning" run was conducted
using the model system to be studied.
                                    11

-------
Aerosol Generation. Monitoring, and Collection
Detailed data concerning reactant concentrations and conditions are
shown in Table 1 (p. 25 ).  After initial concentrations were estab-
lished, the mixing fan was turned off and irradiation was begun.
Parameters monitored throughout the course of the experiment included
NO and NC>2 concentration using a continuous Saltzman analyzer,  total
hydrocarbon concentration using a flame ionization detector, ozone
concentration using a chemiluminescence analyzer, and light scatter-
ing using an integrating nephelometer.

In general, irradiation was continued until light scattering measure-
ments indicated that maximum aerosol growth had occurred.  After
this maximum was observed, irradiation was halted and the aerosol
was collected by evacuating the chamber contents through tared
10-cm-diaraeter glass-fiber filters (Gelman Type A) using a high-
                                           f\
volume type sampler.  Typically, about 11 m  of chamber volume
was sampled over a 30-minute period.  Before use, the filters were
washed repeatedly, first with distilled-in-glass methanol, and  then
with methylene chloride.  They were then equilibrated at 50 percent
R,H. and tared.  After collection,  the filters were reequilibrated,
weighed, and placed in glass containers for storage at -60 C.  After
a series of runs was completed, collected aerosol was subjected to
Soxhlet extraction, as described below.
EXTRACTION OF AEROSOL PRODUCTS
Collected aerosol was subjected to  Soxhlet extraction first with
methylene chloride, and then with either acetone or methanol.  The
sequential extraction procedure was used in order to obtain reaction
products having a wide range of oxidation state and polarity.  The
methylene chloride extracts are expected to contain products less
oxidized than those obtained in the metbanol or acetone extracts.
                                    12

-------
Thus, consistent with previously discussed rationale, the methylene
chloride extractable matter was used for subsequent detailed analysis.
Extraction with the more polar solvent was conducted in order to
provide a measure of weight-percent extractable into solvents of
varying polarity, and to obtain the more oxidized reaction products
for possible analysis during future work.  Values for weight-percent
extractable (see Results and Discussion section) should be valuable
in that they permit comparisons of data for model aerosols with
that obtained for atmospheric aerosol.

It should be emphasized that samples were collected on prewashed
filters.  Soxhlet extraction of such prewashed filters was con-
ducted, and blank values for weight-percent extractable were deter-
mined.
Extraction Procedures
All extractions were conducted using high-purity "distilled-in-glass"
solvents (obtained from Burdick and Jackson, Muskegon, Michigan).
Methylene chloride extractions were run for 20 hours; methanol or
acetone extractions were run for 44 hours.   Approximately 25 ml of
solvent was used for each 10-cm-diameter filter disk.  The volume of
extract was reduced to 1-3 ml using a Kuderna-Danish fractionating
column.  In order to determine values for weight-percent extractable,
a small aliquot  (~1 percent)  of the concentrated extract was taken
to dryness on a light  (~5 mg)  aluminum weighing pan.  Pans were
tared and reweighed using a Cahn Electrobalance, and aliquot .
weights were determined to ± 2 (ig.   From these data the total weight
of extractable matter was calculated.  Values for weight-percent
solvent extractable were calculated as shown below.  In making these
calculations a correction was made to account for filter blank.
                                    13

-------
Weight-percent      _ (weight of extracted matter,  corrected)
solvent extractable   (weight of total participate)
FRACTIONATION OF AEROSOL EXTRACTS
As noted previously, detailed analyses were conducted using methylene
chloride extractable matter.  In order to simplify the analytical
task, the extractable matter was fractionated according to the scheme
below.
                    Methylene Chloride Extractable Matter
Water-Soluble Fraction                     Water-Insoluble  Fraction
                                                                    I
                                       Acid        Neutral       Basic
                                     Fraction     Fraction     Fraction
The water-insoluble fraction can be expected to contain the relatively
less oxidized, less degraded components of the reaction mixture.
In keeping with the previously expressed objective of this phase of
the study, detailed analyses were conducted using the water-insoluble
fraction.  Further fractionation into acid, neutral, and basic com-
ponents was performed in order that analyses could be conducted using
mixtures partially defined according to chemical class.  For example,
in analyzing a neutral fraction, one might anticipate the presence
of aldehydes, ketones, or alcohols, but not carboxylic acids or
phenols.
                                   14

-------
Fractionatlon Procedures
The methylene chloride extractable matter was fractionated into
water-soluble and water-insoluble components by extracting the
methylene chloride solution (~3 ml) four times with 2 ml por-
tions of distilled water.  All extractions were conducted in
conical centrifuge tubes agitated using a vortex mixer.   This
procedure permits convenient centrifugation of emulsions prior to
separation of phases, and minimizes sample losses.  Water was re-
moved by lyopholization, and the weight of the water-soluble
fraction was determined.

The water-insoluble material remaining in solution was next extracted
four times with ~2-ml portions of 2N aqueous sodium hydroxide and
twice with ~2-ml portions of distilled water.  The sodium hydroxide
and water washes were combined.

The methylene chloride solution was then extracted three  times with
~2-ml portions of 2N hydrochloric acid and twice with ~2-ml portions
of distilled water.  The acid and water washes were combined.  Addi-
tional water washes were made until neutral; these were discarded.
Material remaining in methylene chloride solution after extraction
with both aqueous sodium hydroxide and hydrochloric acid  is defined
as the water-insoluble neutral fraction.

The sodium hydroxide extract  (containing organic-acid salts) was
brought to pH ~0.1, and  the free acid was extracted into  methylene
chloride using a continuous liquid/liquid extractor for 96 hours.
Similarly, the hydrochloric acid extract  (containing organic-base
salts) was brought to pH ~13, and  the free base was extracted  into
methylene chloride, again using a  continuous liquid/liquid extractor
for 96 hours.  The methylene  chloride solutions containing the acid,
                                     15

-------
 basic and neutral  fractions were dried by refluxing the solutions
 over a  3-A molecular sieve.  Drying the solutions directly with
 anhydrous magnesium sulfate or molecular sieve was found to be
 unsatisfactory because of irreversible adsorption of organics upon
 the drying agent.

 The dried solutions were concentrated to a volume of -1 ml using a
 Kuderna-Danish fractionating column.   The weight of organic matter
 in the concentrated solutions was determined by taking an aliquot
 of the concentrate  (~5 percent)  and  evaporating the solvent on a
 light  (~5-mg)  aluminum  weighing pan.   Pans were tared  and reweighed
 using  a  Cahn  Electrobalance,  and aliquot weights  were  determined to
 ±2 u,g.   Using  values determined for the weights of  the  vario'us  frac-
 tions, weight  percent distribution values were  calculated.
 DERIVATIZATION OF AEROSOL PRODUCTS
 In order to aid product  characterization,  the  program  included  study
 of selective  derivatization  of  aerosol reaction products.   This  was
 directed at "tagging" specified  functional groups with  easily
 detectable markers.  This is  useful in that  the mass spectral data
 often do not  permit  unambiguous  identification  of the  functional
 groups in a sample.  It  is especially  true where  gas chromatography
 combined with  mass  spectrometry  is performed at unit resolution   T
 such cases the  empirical  formula  cannot always be  determined with
 certainty.  Thus, a  study of  selective derivatization was under-
 taken to  assess the  usefulness of this approach to aerosol
 characterization.

Specific derivatization of carbonyl groups and  hydroxyl groups was
 investigated.   Derivatization of carbonyl groups was conducted by
reaction with ethanedithiol to form corresponding dithiolane" as
shown in the reaction below.
                                    16

-------
V
                         HS

                                - - "      /\ _

Carbonyl                  ,CH2                      S      2
compound                 HS                       ^--i.-  i
    K                                             Dithiolane
                   Ethanedithiol                  derivative


Dithiolanes can be selectively  detected  using flame-photometric  gas

chroma tography.  Additionally,  they  produce  distinctive  fragmentation

patterns  upon mass spectrometric  analysis.



Various reagents for derivatization  of hydroxyl  groups were  investi-

gated, including

                                               II   X^t
          1.  Trifluoroacetyl imidazole, CFo — C — N    N

                                                  oW
                                                  il  X^
          2.  Heptaf luorobutyryl  imidazole,  C^F-7-C-N     N

                                                 o
                                                 II
          3.  Heptaf luorobutyryl  chloride, C-jFy-O

                                                 0
                                                 II
          4.  Pentaf luorobenzoyl  chloride, C^F^-C-

                                               0
                                               II
          5.  Diethylchlorophosphate, (C2H50)2P-Cl
The fluorinated acyl chloride and acyl imidazole  reagents  react with

hydroxyl functions to form fluorinated ester derivatives.  These can

be selectively detected by electron-capture gas chromatography,

               0
               P
          R(F)-C-G1                                  0

or             0        +      ROH - ^    R(F)-C-OR
               li
          R(F)-C-IM
                                    17

-------
 Diethylchlorophosphate reacts with hydroxyl  groups  to  form  the  cor-
 responding phosphate ester.

                   M                                            °
           (C2H50)2P-C1    +     ROM 	».    (C2H50)-P-0R

 Such derivatives  can be selectively detected by  flame-photometric
 gas  chromatography.   Thus, the phosphates  can be  analyzed using the
 same instrument and detector as  is used  for  analysis of  the  dithio-
 lanes.   Like  dithiolanes,  phosphates produce strikingly  characteristic
 mass spectral  fragmentation  patterns.

 Finally,  it should  be  noted  that,  prior  to the analysis  of  the  acid
 fraction by gas chromatography and GC-MS,  it was  treated with diazo-
 methane.   This commonly used reagent converts carboxylic acids  and
 phenols  to relatively  volatile methyl esters and  methyl-aryl ethers,
 thereby  permitting  gas  chromatographic analysis.
 Preparation of Dithiolane  Derivatives
 The  procedure employed  for preparation of dithiolanes was adapted
 from methods reported  by Fieser    '    and by Corey et al.   '  Typically
 1-10 mg  of an authentic  carbonyl  compound or  aerosol product was mix-
 ed with  0,5 ml of ethanedithiol and  0.5 ml of distilled  BFo-etherate
 in a 3-ml  reaction vial with  Teflon-lined screw-cap.  The reaction
 mixture was heated at 60 C overnight.  It was  then cooled and diluted
with 5-10 ml of "distilled-in-glass" methylene chloride.  The methy-
 lene chloride solution was then washed twice  with an equal volume
 of distilled water,  three times with 10 percent aqueous  sodium hy-
 droxide, and again,  twice with distilled water.  Extractions were
 conducted  in conical centrifuge tubes agitated using a vortex-
mixer.  This permits convenient centrifugation of emulsions prior
 to separation of phases.
                                    18

-------
After the above cleanup, the tnethylene chloride solution was dried
by refluxing over molecular sieve (3A), and the dried solution was
concentrated to a volume of 1 ml using a Kuderna-Danish fraction-
ating column.  The concentrate was retained for analysis by gas
chromatography and GC-MS.
Preparation of Perfluoroester Derivatives
Preparation of perfluoroester derivatives of alcohols was conducted
using reagents 1-4 above and was performed according to a procedure
                                                       (43)
adapted from Pierce Chemical Company product literature    .  Typically,
2 mg of authentic alcohol or aerosol product was mixed with  200 M-l
of dry pyridine in a 1-ml reaction vial with Teflon-lined screw cap.
To this was added ^200 mg of acyl chloride or acyl itnidazole reagent.
Combination of reactants was conducted in a glove bag under dry
nitrogen.  The reaction mixture was then heated at 60 C for 2-6 hours.

Workup of the reaction mixture was begun by slowly adding  1 ml of
ice cold distilled water to destroy excess reagent.  The mixture was
extracted three times with 2-ml portions of chilled "distilled-in-
glass" hexane.  Extractions were conducted in a conical centrifuge
tube agitated using a vortex mixer.  The hexane solution was dried,
concentrated and retained for analysis by electron-capture gas
chromatography.
Preparation of Phosphate Derivative^
The derivatization procedure was adapted from methods reported by
     (44}             (45)
Corey     and by Brown    .  Typically, 2-20 mg of authentic alcohol
or aerosol product was combined with 0.5 ml of dry 2,6-lutidine and
0.25 ml of diethylchlorophosphate in a 3-ml reaction vial with Teflon-
lined cap.  The vessel was closed securely and the mixture was heated
to 60 C for 30 min, and was then permitted to stand at room temp-
erature overnight.
                                    19

-------
 Workup of the reaction mixture  was  begun  by slowly  adding  1 ml  of
 distilled water to destroy excess reagent.   The  mixture was then
 washed into a conical  centrifuge  tube  using ~5 ml of  distilled-in-
 glass  methylene chloride.   The  solution was extracted once with
 ~3 ml of  distilled water,  three  times with IN hydrochloric acid
 once again with distilled  water,  three times with IN  aqueous  sodium
 bicarbonate,  and finally,  twice with distilled water.  All wash
 solutions were  chilled in  an  ice bath before use, and extractions
 were conducted  using a conical  centrifuge  tube agitated by means of a
 vortex mixer.

 After  workup, the  methylene chloride solution was dried by re-
 fluxing over  3-A molecular sieve.   It was  then concentrated to  1 ml
 using  a Kuderna-Danish fractionating column.  The concentrate was
 retained  for  analysis  by gas  chromatography  and  GC-MS.
 ANALYSIS  OF AEROSOL PRODUCTS  BY GAS CHROMATOGRAPHY
 AND MASS  SPECTROMETRY
 After  extraction,  fractionation, derivatization,  and  preliminary
 analysis  by gas  chromatography, the aerosol  reaction  products were
 analyzed  by gas  chromatography combined with mass spectrometry.
 Analyses were performed using a quadrupole mass  spectrometer.    The
 ion source of this  instrument has been modified  to permit  generation
 of either chemical  ionization or electron  impact mass  spectra.  Most
 of the analyses  conducted during this study were performed in the
 chemical  ionization mode.

During the last several years there has been a marked  increase  in
 use of chemical  ionization mass spectrometry.  Nevertheless,  it is
 probably less familiar to many analytical chemists than conventional
electron impact spectrometry.   Since a  considerable  number of  chemical
 ionization spectra are presented in this  report,  some of the basic
                                    20

-------
characteristics of the technique are reviewed  briefly  below.

In chemical-ionization spectrometry , the  sample molecules  are  ion-
ized by means of vapor state interaction  with  ions  of  a  "reactant
gas".  The reactant gas most commonly used  is  methane.   As shown
in the reaction sequence below, the methane  is ionized by  electron
impact causing formation of a series of primary ions.  At  pressures
of about 1 torr, these react further with methane to produce  the
secondary ions shown.  These secondary ions  are the principal  species
present in the ion source of the spectrometer
        CH<1    e~   ^CH4+    CH3+     METHANE  ^  CH5+   C2H5+
                    CH2+    CH+      -1TORR
                                              PRINCIPAL SPECIES
The secondary ions cause ionization of  the  sample molecules  by  donat-
ing a proton or abstracting a hydride ion.  As  shown  in  the  reactions
below, this interaction leads to formation  of quasimolecular  ions
having mass/charge  (m/e)  one unit greater  than  or  one unit  less
than the molecular weight  (M)  of the sample molecule.   The  quasi-
molecular ions are almost always observed;  generally  they  are prominant,
                                      ROH2  +
                       ROM ~"~            \
                         U            QUASIMOLECULOR IONS
                  ALCOHOL    _  +^>.   J+
                   SAMPLE    C2H5  ^  RO     -i-
                                   21

-------
 There is also a tendency for the sample molecule  to  form an  adduct
 with the C2H5  ion or the C3H5  ion.   When this occurs  "satellite
 ions" are observed at m/e  values corresponding to M +  29 and M + 41
 The observation of the quasimolecuiar  ion  and  satellite ions facilitates
 the assignment of  molecular weight.  During this  study  use of GC-MS
 has thus permitted tentative assignment of molecular weights for the
 majority of  principal chromatographic  peaks in a  sample.

 In  general,  chemical  ionization  mass spectra show a  great deal less
 fragmentation than electron impact  spectra.  In addition,  the frag-
 ment  ions that  are observed generally  arise from  elimination of
 functional groups  as  small  neutral molecules. Thus,  chemical ion-
 ization  spectra are often easier to correlate with structural features
 than  are electron  impact  spectra.

 In  some  cases  chemical  ionization spectra  were obtained  using helium
 as  the reagent  gas.   Helium ions  are species of much higher  energy
 than  the  secondary ions  of  methane.  The resulting spectra show far
 more  fragmentation than  those  obtained  using methane.   The helium
 spectra  are,  in fact, very  similar to  classical electron-impact
 spectra.  Generation  of both methane and helium chemical-ionization
 spectra  often aids  in characterizing the sample.

 Finally,  it  should be emphasized  that  the  mass spectral  characteriza-
 tions presented in this report are almost  invariably referred to
as "tentative".  Firm identifications can  only be made when reference
 spectra are available for comparison.   In  view of the relatively
 recent advent of chemical-ionization mass  spectrometry extensive
 compilations of such reference spectra are  unavailable.    Thus in
many instances authentic compounds were obtained,  and reference
 spectra determined in attempts to confirm  tentative assignments.
                                    22

-------
This approach was used extensively during analysis of the a-pinene
aerosol.

The interpreted spectra shown in the report include varying numbers of
significant peaks which have not been rationalized.  Many such peaks
arise from the presence of contaminating compounds which elute simul-
taneously with the assigned product.  In analysis of complex mixtures
by GC-MS many spectra will show such contributions from multiple
components.
Gas Chromato^rapjiic and Mass Spectral Procedures
Gas chromatographic separations were performed using a 3-m x 2-mm-
ID (1/4-in-OD) Pyrex column packed with 3 percent OV-17 on Gas Chrom
Q.  Typically, analyses were performed with temperature programming,
100 - 250 C at 4 C/min.  Two dual detector chromatographs were used
in the study, a Varian Model 1700 equipped with flame-photometric
detector and flame-ionization detector, and a Varian Model 2100
equipped with electron-capture detector and flame-ionization detector.

Analysis by GC-MS was performed using a Finnigan 1015 Mass Spectro-
meter.  Instrument control and data acquisition were accomplished using
a System Industries 150 data acquisition system.  The reconstructed
gas chromatograms shown in this report were computer generated.  The
detector response in these chromatograms was derived from the total
ion current observed during successive scans on the spectrometer.
Typically in this study, scans were repeated every 8 seconds.  Thus,
the x-axis of a reconstructed gas chromatogram is roughly equivalent
to retention time.  Desired spectra are recalled from computer memory
(disk) by reference to the spectrum number.
                                23

-------
                               SECTION IV
                         RESULTS AND DISCUSSION
SMOG CHAMBER IRRADIATIONS
Irradiation conditions and smog manifestations characterizing the model
aerosols are shown in Table 1.  The values listed represent averages for
the respective hydrocarbon series.  There was very little run-to-run
variation in the parameters; the magnitude of such variation was insuf-
ficient to be of any consequence to the composite aerosol composition.

In general, the smog reactions in both the cyclohexene andcx-pinene se-
ries were very fast as evidenced by the rapid rate of NO photooxidation
(Table 1, NO -,„ values).  The NO photooxidation rate in the toluene se-
ries was considerably less in spite of the higher HC/NO  ratio employed.
                                                       3t
These trends in relative hydrocarbon reactivity toward NO photooxidation
are similar to those reported in the literature.^  '
EXTRACTION OF AEROSOL PRODUCTS
Collected aerosol was subjected to sequential solvent extraction.  Cyclo-
hexene aerosol was extracted first using methylene chloride and then
acetone.  Toluene and ct-pinene aerosols were extracted first using methyl-
ene chloride and then methanol.  Values for weight-percent solvent ex-
tractable are shown in Table 2.

As noted previously, collections were made using prewashed glass fiber
filters.  In calculating values shown in the table,  filter background
(blank) has been subtracted.  Soxhlet extraction of prewashed filters
gave blank values of 0.007 mg per 10-cm-diameter disk for methylene
chloride extraction and 0.27 mg per 10-cm-diameter disk for methanol ex-
traction.

It is interesting to compare valuer shown in the table with those of
weight-percent extractable observed for atmospheric particulate matter,
                                  24

-------
      TABLE  1.   SO1MARY  Op I RRAOMTION1 C.OSDt TI ('.\S  AM) SMOC PARAMETEKS ' 3
                                                        Hvdroca r hon
                                         Cvc lohexone       To 1 uene	T_-1' i neiu-
                                              ( b)
Number ot runs                               2

1 rrad iat ion pe r i od ,  in i n                    180

Relat i ve humid ily
  Initial, percent                         <>4
  Final, percent                           5h

Hydrocarbon
Mass I rrad iated , g
initial concentration, ppm C
Final concentration, ppm C;'c'
N i i r ic ox i de
Initial concentration, ppm
l, .,. mi./1"
1/2
NitrogL-n dioxide
Initial concentration, ppm
Maximum concentration, ppm
Ozone
Maximum concentration, ppm
t , inin
max
(e)
Light scattering
Maximum, 10 m
Mass loading , mg m
Percent conversion
(3)
Total aerosol mass collected, g
O.«04
63
2U

2.4
15


2.4
4.8

0.46
180



I 10
1 3.0
28.0
0. 390
0 . * '-> 7 1 . ^ 2 1
71 104
Ml Ji,

1.0 J . 0
75 ii


1.2 J . 0
1.4 i . 0

0.7 O.OOh
34 35



180 220
1.4 33
2.8 40
0.125 1.341
(a)  Data shovn  are  averages for (he  series  o;  runs,  except lor values  ol  total
     aerosol mass  collected, t'hich represent  the toial  mass collected during  an
     entire  series of  runs.

(b)  After conditioning,  a series of  [our  cvclohexcne runs was conducted.
     Aerosol from  li.'O  of  the runs was compromised during work-up.

( c )  Includes  L on t r i hu t i on t r»>m prt>tl uc t  I o rma L ion.

(d)  Time at which one half the initial  amount  i>(  \()  wa s oxidised to Xl > ; .

(e)  Indic;iLeil values  are outside the/ normal  operating  ran^e ol ttie  inlenrating
     nephe1omet er.

(1)  Mass loading  was  calculated by dividing  the aerosol mass collected  l>^  the
     ell'ective volume  sampled.   'Hie effective  volume  sampled was calculated hv
     measuring the decrease in total  organic  carbon  during the sampling  period
     and multiplying this fractional  decrease  Iv.  t'ie  chamher volume.

(g)  Percent conversion  i-'as calculateil as  total  aerosol mass per mass ol
     hydrocarbon i r rail ial ed.  Total aerosol  mass va s  calculated 1 rom the
     indicated mass  loading multiplied In-  t lie  claniln-r volume ol 17.3 m  .
                                        25

-------
                                  Table 2.  SOLVENT EXTRACTION OF  COLLECTED AEROSOL
0s-
=====
Hydrocarbon
Cyclohexene
Toluene
a-Pinene
=====
Total
Aerosol
Mass
Extracted,
g
0.390
0.125
1.341
— . 	
First Extraction,
Weight-Percent
Methylene Chloride
Extractable
62
46
92
	 ii' •- —
	 r~: 	 • •
Second Extraction
Solvent
Acetone
Methanol
Methanol
Weight-Percent
Solvent Extractable
31
43
1
	 •
Total Extractable
Matter,
Weight-Percent
93
89
93
! 	

-------
 such  as  those  reported  in  the  second annual report on "Haze Formation --
 Its Nature  and Origin"     .  During that  study, atmospheric particulate
 matter was  collected  in Columbus, Ohio, New York City, and Pomona
 California.  It was similarly  subjected to solvent extraction first using
 methylene chloride, and then dioxane.  (Dioxane was used as the polar
 solvent  in  order  to minimize extraction of the considerable quantities
 of inorganic salts known to be  present  in  atmospheric particulate.)  Re-
 ported values  for methylene chloride extractable average 15 weight per-
 cent, ranging  from 8-22 weight percent.   Values for dioxane extractable
average 21 weight percent,  ranging from 8-40  weight  percent.   Values  for total
 extractable average 36  weight  percent, ranging from 16-56 weight percent.
 Comparison  of  these values with those  shown in Table 2 indicates that
 the model aerosols contain significantly  more organic extractable matter
 than  do  atmospheric aerosols.  This is especially evident when comparing
 values for methylene  chloride  extractable matter.  The data support the
 contention  that aerosols which form as a  result of photochemical reactions
 are composed predominantly of organic matter.

 In considering  the structures  of the hydrocarbon precursors, it is in-
 teresting to note that  as the  total number of carbon atoms per double-
 bond  decreases there is a tendency for the reaction products to have
 less  solubility in methylene chloride and greater solubility in the polar
solvent,  that is,  acecone or methanol (see structures  p.  28).

 The data suggest  that under atmospheric simulation, hydrocarbons having
 increased numbers of  double-bonds will undergo greater oxidation/
 degradation, yielding products which would be expected to have greater
 solubility in  solvents  of greater polarity (i.e., acetone or methanol
 rather than methylene chloride).
^Strict reference is made to'"double-bonds" rather than "unsaturation"
because the bicyclic compound a-pinene has unsaturation equivalent to a
monocyclic hydrocarbon containing two double bonds.
                                     27

-------
            a-PINENE
                                      CARBON ATOMS PER DOUBLE-BOND
 (10)
           CYCLOHEXENE
 (6)
                         CH,
           TOLUENE
(2.3)
FRACTIONATION  OF AEROSOL  EXTRACTS
Using  the procedure  described previously, methylene  chloride  eytractable
matter was  fractionated into water-soluble  and water-insoluble  components
The water insoluble  material was further  fractionated  into  acid,  neutral
and basic components.  The distribution of reaction products in  the  three
aerosol systems is shown  in Table 3.
DERIVATIZATTON OF AEROSOL PRODUCTS
Procedures for specific derivatization of alcohols  and  carbonyl  compounds
were investigated and the utility of this approach  was  considered  as  an
aid to identification of aerosol reaction products.  The various deriva-
tization reagents described earlier were evaluated  using model alcohols
and carbonyl compounds, and selected reagents were  applied  to derivati-
zation of reaction products.
Dithiolanes of Authentic Carbonyl Compounds
Using the previously described derivacization procedure, dithiolanes  of
the following authentic carbonyl compounds were prepared:   n-hexaldehyde
n-heptaldehyde, 2-ethylhexaldehyde,  2-ethyl buteraldehyde, 3-hexanone,
and 2,4-dimethyl-3-hexanone.  The gas chromatographic retention times of
the dithiolanes are significantly longer than those of  the  parent
                                   28

-------
Table 3.  FRACTIONATION OF METHYLENE CHLORIDE EXTRACTABLE MATTER
Hydrocarbon
Cyclohexene
Extractable Matter Fractionated, mg 80
Water- insoluble fraction
Acid, weight percent
Neutral, weight percent
Basic, weight percent
Water-soluble fraction, weight percent
Overall recovery, weight percent

8
3
0.2
66
77
Toluene ct-Pinene
29 615

10 52
2 17
0.3 1
70 27
82 97

-------
 carbonyl  compounds.   For  example  using  a 3-m x  2-mm  3%  OV-17  column  pro-
                                     -.]_                               K
 grainmed from 100  to  250 C at  6  C  min   the  retention time  of  3-hexanone
 and   2-ethylbuteraldehyde dithiolanes are 9.3 and  11.0  minutes respec-
 tively, whereas both parent compounds elute in  less  than 2 minutes.  Gas
 chromatograms  of  several  authentic  dithiolanes  are shown in Figure 1
 Both  flame  ionization and flame photometric (sulfur  specific) detection
 were  employed.  The  dithiolanes were further analyzed by GC-MS.   Recon-
 structed  gas  chromatograms are  shown in Figure  2,  and chemical-ionization
 mass  spectra  are  shown in Figures 3-5.
 Fragmentation  of  aldehyde  dithiolanes  is highly  characteristic   and  is
 clearly  diagnostic  of  the  presence  of  aldehyde in  the original  aerosol
 product  mixture.  The  fragmentation is  rationalized below,  and  includes
 formation of the  diagnostic  fragment at m/e  105.
     RCH = SH
     R C  CH2
       V
                               RCH
    X
                                    HS-CH2
                                     'S-CH2
                  LOSS OF
PROTONATED
MOLECULAR
ION
               LOSS RH
                                -S—CH,
                                                            CH
                                                        DIAGNOSTIC FRAGMENT
                                                             m/e = 105
Fragmentation of ketone dithiolanes is similarly characteristic, but
does not include appearance of the fragment at m/e 105.  This can be
helpful in distinguishing aldehyde and ketone dithiolanes.  Fragmenta-
tion of the ketone dithiolanes is rationalized below.
                                     30

-------
        R C
                                                 LOSS OF
                                                              'C=SH
                                                CH2- CH2
Dithiolane Derivatives of a-Pinene
and Cyclohexene Aerosol Products
Portions of the o.-pinene and cyclohexene neutral fractions were reacted
with ethanedithiol/BF^ to obtain dithiolane derivatives.  The deriva-
tized product mixtures were analyzed by gas chromatography and GC-MS.
For the sake of clarity mass spectral analyses of the derivatized
aerosol products are considered as a unit under this heading.  Analyses
of unde.rivatized aerosol products are considered separately in following
sections for each of the three aerosol systems studied.  In the present
discussion, the analytical approach involving dithiolane derivatives
will be evaluated.
 Analyses  of  dithiolane  derivatives  for  both rv-pinene  and  cyclohexene
 aerosol products  indicated  the  presence of carbonyl compounds whose
 molecular weights were  rather low when  compared  to the compounds which
 were  identified  in  the  underivatized neutral  fractions.   In the cyclo-
 hexene neutral fraction, analyses of the dithiolanes  indicated the
 presence  of  carbonyl  compounds  of the following  molecular weights:
 76, 84, 96,  98,  108,  and 112.   The a-pinene neutral fraction was
                                     31

-------
          KETONE DITHIOLANES
     ALDEHYDE DITHIOLANES
N O
Z O
LU LJJ
5 0
                       B
                      12  MINUTES
0   3   6   9   12  15  MINUTES
cc
LU
LU Q
DITHIOLANE DERIVATIVES OF:

A.   3-HEXANONE
B.   2.4-DIMETHYL-3-HEXANONE
C.   n-HEXALDEHYDE
D.   n-HEPTALDEHYDE
E.   2-ETHYLHEXALDEHYDE

 SAMPLES ANALYZED USING A
 10 FT x 2 MM COLUMN OF
 3-PERCENT OV-17 ON GAS
 CHROM Q TEMPERATURE
 PROGRAMMED 100-250°C
 AT6°C/MIN
                   FIGURE 1.  GAS CHROMATOGRAMS OF AUTHENTIC DITHIOLANES.

-------
U)
OJ
                                                               DITHIOLANE DERIVATIVES OF:

                                                               A.  n-HEXALDEHYDE
                                                               B.  n-HEPTALDEHYDE
                                                               C.  2-ETHYLHEXALDEHYDE
                                                               D.  3-HEXANONE
                                                               E.  2-ETHYLBUTYRALDEHYDE
                                                               F.  2,4-DIMETHYL-3-HEXANONE
                                          10 28  30
                                          6PECTRH KJCEB
                                                    90
                FIGURE 2. RECONSTRUCTED GAS CHROMATOGRAMS OF AUTHENTIC DITHIOLANES.

-------
F

R.

S_


?-
JR_
R_
c



o


(
. ^S-CH2
CH | _^








• , • , T*l







111, 	 	 U>
^CH3lCH2i4CH = SH












^ '





II
lilt,..)!.,,,.,! 	 ,1.1
X \
CH2-CH?_





ll 1
lUv 	 -,X^.T- 1
, n-HEXALDEHYDE
DITHIOLANE
V
E

I
5
o ^
E E

^ ^i
u V
2 £
m« 	 T 	 i-i'-i 	 	 	
SO  70  90   30  100  110 120  130
   MX
                                 ISO 160  170 ISO  190 200  210 220  230 2«  2SO 2BB  270 290  230
R.
~
\&-<
9
sR.
B.
D

,-H'' "TZ






,






JU.i rtt







	 LnlJ,
n HEXALDEHYDE
DITHIOLANE
-tH-j-'CHjij -CH ~SH
S
CH? ~tHJ
-


1 ^
-

II,,
=
I


II. . .
SO  70  80   98  100  110 120  130 110  ISO :60  170 190  190 208  2lO 220  230 210  250 268
   MX E
   FIGURE 3.  CHEMICAL IONIZATION MASS SPECTRA  (METHANE)
               OF AUTHENTIC DITHIOLANES.

               (Refer to Chromatograms in  Figure 2}
                                        34

-------
                                                            2-ETHYLHEXALDEHYDE

                                                              DITHIOLANE
  GO  70  88  30  100  110 120 130  110 ISO  168 170  180 130  2BB 210  220 230 210 2GO 2SB  278 280  2
8_

 -
 ~
    CH, - CH, - CH, - - C - CH, - C
                                                          3-HEXANONE DITHIOLANE
            ll|	llllit'l'lllilltdUl''"'""!"11!""!1'"!""!	,l	[iiiumi)	iii.i.ny.
         80  39  100  110  128 130  110 ISO ICO 179  160 130  59
••!	I	I	I	I1
 210 22B  230 210 2
  CO   70
      FIGURE 4.  CHEMICAL IONIZATION MASS SPECTRA (METHANE)
                  OF AUTHENTIC DITHIOLANES.

                  (Refer to Chromatograms in Figure 2)
                                         35

-------
b6L
 R_
                                           2-ETHYLBUTYRALDEHYDE DITHIOLANE
                                       J
   so' V '«'"»" IBB iio 'i»"iai»T« ' iso in 'ITS 'ise
      MX e
g

RJ
  r-*
  IK

  R-l
                         (. CHiLH-, ^
                                         2,4-DIMETHYL-3-HEXANONE DITHIOLANE
    SB  'TO  w M  ino 'ne  IZB
                             Y« 'l   ISO iTB 188 IS) MB 219 OB  238  Z« Z9B ZBO 270 ZBO 230 3BB
      FIGURE 5  CHEMICAL IONIZATION MASS SPECTRA (METHANE)
                  OF AUTHENTIC DITHIOLANES.

                  (Refer to Chromatograms in Figure 2)
                                        36

-------
indicated to contain carbonyl compounds of molecular weights,  44, 76,
78, 90, 104, and 108.  The reconstructed gas chromatograms are shown
in Figures 6 and 7.

Analysis of the underivatized neutral fraction of cyclohexene aerosol
indicates possible formation of the products indicated below.  (See
mass spectra in Figures 22 and 23.  These spectra are discussed in
detail under the headiag Cyclohexene Neutral Fraction.)
                         0  °                     run
                          n  ii                     LHO
             CH3-CH = CH-C-C-CH3
                   (MW =112)                   (MW = 96)
Mass spectra of the derivatives indicate the presence of dithiolanes
corresponding to carbonyl compounds having the above molecular weights,
112 and 96 (peaks A and B in Figure 6).  Consider the mass spectrum
shown in Figure 8.  This dithiolane corresponds to a derivative formed
f'.-om carbonyl of molecular weight 112.  Although the intense peak at
m/e  = 189 appears to correspond to the protonated molecular ion
(dithiolane), the other rather weak peaks are not consistent with a
                                              xSx
ketone.  The weak peak at m/e = 129 (loss of CH2~CH2) would be expected
for both aldehydes and ketones, but the relatively strong peak at
m/e = 105 is usually diagnostic for aldehydes.

In contrast, the mass spectrum for the underivatized product suggests
the diketone, above.  If the assignment of the diketone structure is
correct, then the derivative mass spectrum in Figure 8 indicates the
presence of an aldehyde which was not identified in the cyclohexene
neutral fraction.  Since this derivative spectrum gives little evidence
regarding the parent carbonyl compound, any structure suggested would
be purely speculative.  The empirical formula is most probably
C6H802.
                                     37

-------
                    P.
                    8
                    8_
                                                                            to
9   18 20  JO  -W
   SPECTHLfl MJ«EB
                                     60  70 80  <»  180 118 120  138 HO ISO  160 170 180  130 ZOO 210 22O Z3B ZtO ZS5
U>
OO
       GIVEN IN PARENTHESIS IS THE
       MOLECULAR WEIGHT OF THE
       CARBONYL COMPOUND
       CORRESPONDING TO THE
                                                               (A)

ii
g
5
J
	 A 	 . .-,
/
         FIGURE 6. ANALYSIS OF CYCLOHEXENE AEROSOL, NEUTRAL FRACTION DITHIOLANE DERIVATIVES.
                   RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
u>
6
"" 1
R_

8_

P.

8.
I
JR.
j
*D


R.

R.
o

1
GIVEN IN PARENTHESIS IS THE
MOLECULAR WEIGHT OF THE

CARBONYL COMPOUND
CORRESPONDING TO THE
INDICATED DITHIOLANE



3
S
7
i n
1 ft
vUL__
	 ; 	 i 	 i 	 I 	 I 	 pi '<••"! 	 i 	 i"'

00
r>
<«
<£>
s
<*l
SI
I
?

oo 3
o o

Sao — o
00 ffl
«— •— ffl •"
•C '-P
1 s i I !
P A1 ft UAA^
All ^/^^JULJ^^IU^^
'. _ . . » . -« < -j« t «A i o» i CH t -?n i an t qn ?nrt ? i (1 77ft y.n j-w ,Oii3 SO ^70 iflfl S30
          0  18 28  30  10  SO  60  70  80  30  JOB HO IZB
             3PEETBJ1 «>BEP
       FIGURE 7. ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION DITHIOLANE DERIVATIVES.
                 RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
  R.
                   0>
                         o
                         CM
                         0)
                         1
                           CM
                          . CM
                           X
                           u
                                            §
                                            0)
                                                                                           -^
      68 73  
-------
 Consider the mass spectrum shown in Figure 9.   This  dithiolane corre-
 sponds to a derivative formed from carbonyl of molecular weight 96.
 The peak at m/e  173 corresponds to the protonated parent ion.   The
                                          -S-
 peak at m/e  113  corresponds to loss of CH2~CH2 .   The complete absence
 of a peak at m/e    105 would suggest that the  parent  compound may be
 a ketone.  However,  we have been unable to explain the other  strong
 peaks at m/e = 89,  101, and 119  in these terms.   It is possible  that
 these  fragments represent  interference arising from other compounds
 simultaneously eluted.   Thus,  it  is not  clear whether  the dithiolane
 corresponds  to a derivative of the  indicated aldehyde, or to a deriva-
 tive of  a carbonyl compound that was not otherwise detected in the
 cyclohexene  neutral  fraction.

 Dithiolane derivatives  of  the o-pinene neutral fraction were prepared
 and analyzed by GC-MS.   No carbonyl compounds found in the a-pinene
 neutral  fraction corresponded with  the rather low molecular weight
 derivatives  found.   The  absence of higher molecular weight carbonyl
 derivatives may indicate that these species failed to elute under the
 chromatographic conditions employed, as  there is no clear reason why
 they should not be formed.

 In conclusion, it appears that use of dithiolane derivatives is a
 potentially useful approach to aid  in characterization of carbonyl
 reaction products.   In order to utilize this approach,  gas chroma-
 tographic conditions will be determined which permit elution of the
higher molecular weight dithio lanes.

 Perf luoroacyl and Phosphate Derivatives
of Authentic Alcohols
For convenience, the previously described candidate reagents for  deri-
vatization of alcohols are listed below:
                                    41

-------
      1.   Trifluoracetyl  imidazole
      2.   Heptafluorobutyryl  imidazole
      3.   Heptafluorobutyryl  chloride
      4.   Pentafluorobenzoyl  chloride
      5.   Diethylchlorophosphate.

Evaluation of Reagents 1-3 indicated that although the desired esters
are  formed with various model alcohols, the product esters are in-
sufficiently stable and are  hydrolyzed back to free alcohol and
perfluorocarboxylic acid during aqueous work-up of the reaction mix-
ture.  Gas chromatographic analyses of the reaction  mixtures  without
aqueous work-up proved to be troublesome in that excess derivatization
reagent was deposited at the head of the GC column, leading to column
degradation and irregular results.

Reagent 4 was shown to be sufficiently reactive toward various primary
and  secondary alcohols.  Excess reagent was easily hydrolyzed during
work-up without decomposition of product esters.  The product esters
were  shown to be suitable for specific detection by electron-capture
gas  chromatography.

Simultaneous with evaluation of Reagent 4,  diethylchlorophosphate,
Reagent 5, was studied.  This reagent has properties similar to those
of the carboxylic acid chlorides, and was shown to react with various
primary and secondary alcohols.  Reaction with a tertiary alcohol
(3-methyl-3-hexanol) was also attempted,  but no product was observed
by gas chromatography.  The phosphate derivatives produced can be
specifically detected using flame-photometric gas chromatography.
Thus preliminary analysis of such alcohol derivatives can be conducted
using the same instrument as is used for analysis of dithiolane deri-
vatives of carbonyl compounds.  In view of this fact, evaluation of
diethylchlorophosphate was emphasised.
                                    42

-------
U'
      s.
      h«1
      VI-
 CM
X
                        CO
                        03
                                             CO
             
-------
Phosphate derivatives of various primary and secondary alcohols  were
prepared and analyzed by GC-MS .   A reconstructed gas chromatogram for
the derivatives of isopropanol,  1-pentanol, 3-hexanol, and 1-decanol
is shown in Figure 10.  The mass spectra are shown in Figures 11 and  12.

A clearly characteristic fragmentation pattern is evident in the spectra.
It appears that initial fragmentation of the phosphate ester involves
loss of olefin corresponding to the parent alcohol.  Subsequent  frag-
mentation was identical for all of the derivatives, and simply involved
two successive losses of ethylene.
      0                     0                   0                  o
    + "         LOSS OF    + "         LOSS OF   + "        LOSS OF   II
    H P (OC2H5)2          - H P (OC2Ha)2     H  - H P-OC2H5           H P(OH)3
                                        z4
     OR                    OH                  (OH)2

Phosphate Derivatives of a-Pinene
Aerosol Products
A portion of the a-pinene neutral fraction was derivatized with
diethylchlorophosphate, and was analyzed by GC-MS.  The reconstructed
gas chromatogram is shown in Figure 13.  The chromatographic resolution
of the derivatives is poor.  Examination of typical spectra shown in
Figure 14 reveal very characteristic phosphate fragmentation with
peaks at m/e  155, 127, and 99.  It was nevertheless, not possible to
identify significant parent or adduct peaks in these spectra.  Con-
sequently, the molecular weight of the products could not be determined.
Examination of spectra of authentic phosphate derivatives indicates
that even when a strong protonated parent ion is observed, fragmentation
is rather uninformative.  At best, it appears that preparation and
analysis of phosphate derivatives can aid in establishing the presence
of alcohol and the molecular weight of an aerosol product.  In light of
these findings, this approach to specific derivatization of alcohols
was discontinued.  Limited future work may be undertaken using Reagent 4
above.
                                    44

-------
t_n
        8
        8-
        8.
           A
                                             DIETHYLPHOSPHATE DERIVATIVES OF
A.
B.
C.
D.
                       ISOPROPANOL
                       n-PENTANOL
                       3-HEXANOL
                       n-DECANOL
0  10  ZO  30
   SPECTHLH WJBEH
SC  6C  7C 80  3C  1
                                             ilC  1ZC 13C 1
             15C !60 170 180  130  "JK '.
                                                                                   ZX i-W 2SO 260
                    FIGURE 10.  RECONSTRUCTED GAS CHROMATOGRAM OF DIETHYLPHOSPHATE
                               DERIVATIVES OF AUTHENTIC ALCOHOLS.

-------
 8
 8_
           O

         HP(OHI3
                 -C2H4
   OH

 HP^o



    OH

	 HP- OH
   \
    IOC,H
                                                     DERIVATIVE OF
                                                     ISOPROPANOL
                                         2"5'r
     7C  8C  B8
                        130 IKJ ISO
                                     )70
                                              Z3B ilff 220 230
^^_
R_

8_

f~
if-
ML
|*~
y. -
8~

0

D_














J










OH
HP OH






(OH)j lu">2




O
HP'OH





68 70 SO





3
\
X



as
•
O(




\ /
2\*
3B



C2H4
•*


,
1 	 T
1«5 M



i»
IOC2H5I2
\*
_^l~^l








,1 ll
S



















3 1ZB 135 I'M »3J !S5 175 IP!? ;3C










I











., 1
DERIVATIVE OF
1-PENTANOL


fSl
Psi
•T




• *c. • _n
X I
CN n
LJ u
2 ^


M/ f """'
FIGURE 11. CHEMICAL ION1ZATION MASS SPECTRA (METHANE) OF
           DIETHYLPHOSPHATE DERIVATIVES OF AUTHENTIC ALCOHOLS.

           (Refer to Chromatogram in Figure 10)
                                 46

-------
  . 13
          0

         HPlOH(3
                                              DERIVATIVE OF
                                              3-HEXANOL
                                      OH

                                    HP - OH
                                                           NOTE THfc WfAK PARENT
                                                           ION FOR THIS Df RIVATIUE
                                                           OF a SECONDARY ALCOHOL
    60  76  SB 88 tBO iM 121 138 VK ISO 160 178 18(5 I9O 355  ZlO 23S 230 Z*S ZSO 2BO
       n/ E
                                                                       290 30B 318 320 330
8

5L

8.
•K

i


R-
t^














'" "" !

(OHI2
HP - O

O
HP - OH
\
o
HP (OMI3
\ \
\ ^
\
^






C2H4



^ CjM,





2 6?
.-






!. , .
••••-• -I111"' -•••!— •••• I1 	 1,1.1.,... I—Ill.,... ,....,1. ..... ,..! Il..,.c -,.. , 	 ,..













'" T' •



DERIVATIVE
1-DECANOL







r
» ••.«• 1 If* 1O*« ^31» <



OF







C2M,

»A( 9O*























C2M4

I..I 	



R
E
"r



, 1 1
sen 5T« 5qn ;««• am _3is 32B 3
    e=0  7C  9S  3D  100
       n/ e
FIGURE 12.  CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF DIETHYLPHOSPHATE
             DERIVATIVES OF AUTHENTIC ALCOHOLS.
             (Refer to Chromatogram in Figure 10)

-------
oo
                             5r~ST"S"" 70  90   30   100 110 120 130  110  ISO  160 170 190 190  200
         FIGURE 13  ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION DIETHYLPHOSPHATE DERIVATIVES.
                    RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
V
R-
S.
»
J
E8'
58 .
JR.
8.
o
1



0






4- "
HP(OH)3
^^














2*



	 P 	 1 	 1 	 "1 	 1"*1
B TO «B 3D 100 IK
Hit
"-,
R.

8.
P -
8n
8-
toS_
I*'
R
o
«





-t-
HP
O
HP(OH)3
1
V -










:2






*4




HP v
^-^"



(OH)2
/
/
0
V H FRACTION A
OC2H5 OH
+ /


-C2H4



) 120 130





OH
/
0
\
OC2H5
\

H4
i ;
	 i 	 i 	 i 	 i"""1

-


HP 0
(OC2H5)2



i j 	 J iVi..,!i 1 , .'.-... i. . i, . . . . . i • i • i ' i i i i i i
IKS ise 'ise \ie J«B las ZBB ZIB ZZB zao 'z« as zse zro ao ae see a>o szo aw >»









C2H4


•^•T"^1"! 	

(OH}2
4 /
HP - 0
->,oc2H5>2 FRACTION B






,, ,r,

              ea TO to  3D 100 110 izo iao
                Hlf
FIGURE 14.  ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION DlETHYLPHOSPHATE DERIIV'ATIVES.
           CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAKS A AND B.


           (Refer to Chromatogram in Figure 13)

-------
ANALYSIS OF CYCLOHEXENE AEROSOL
Cyclohexene Aerosol Acid Fraction
The water-insoluble acid fraction of cyclohexene aerosol was isolated as
described above.  The mixture was treated with diazomethane, and the
resulting methyl esters were analyzed by gas chromatography and GC-
MS.  The reconstructed gas chromatogram is shown in Figure 15.  Tentative
identifications have been made on the basis of mass spectra for the
indicated chromatographic components.

The mass spectrum corresponding to chromatographic peak C is shown
in Figure 16.  The product is tentatively identified as the six membered
aldehyde-acid shown.  There is a moderately strong peak assigned as
protonated parent ion at m/e 145.   Fragmentation peaks consistent
with loss of CH30H and loss of CH3CHO are observed at m/e 113 and
101j respectively.  Somewhat less prominent is a peak corresponding
to loss of HCOOCH3 (methyl formate).  This can be seen at m/e 85.
Loss of HCOOCH3 and CH3
-------
     muni ntinpiiijiiii|iiiijiiiniiii[Mii|i'i'iinniiin i
    0     10   20   30   10   SO   60   70   60   98   100  11Q  120   13C
          SPECTRM NUr«SR
FIGURE 15. ANALYSIS OF CYCLOHEXENE AEROSOL, ACID FRACTION METHYL ESTERS.
           RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
Ul
ro
                g
                  -
                 D
                 O
                        TENTATIVE
                        ASSIGNMENT:
                C
                                  CHO
                         ,,.
                   4
                              t+7
                                                      o

                                                      II
                                              O
                                              fi
                                              X
+TTf
                                                                               in
                                                                               sf
                                                                               I
                       TyTTTipTTTfTTTT,,,,,!,,,,,.,,,!,,„,.,„} ^ IT^TJTTTT^TTJTTTTT, ,,.r,. f,, „, ,,, ...,r,Mf,T,rT,,r,1,,r.

            10    S0   60   70    00    30    100  110  120  130  110  ISO  160

                  M/ E
FIGURE 16.  ANALYSIS OF CYCLOHEXENE AEROSOL, ACID FRACTION METHYL ESTERS.
            CHEMICAL IONIZATION MASS SPECTRUM (METHANE)  FOR CHROMATOGRAPHIC PEAK C.

            (Refer to Chromatogram in Figure 15)

-------
              B
OO
                O
FIGURE 17.
1
1 5
"
I ^
+ '
0
o
z
 TENTATIVE
£ °» ASSIGNMENT.
0 '
± j; . 	 COOCH3










1
4fl 5fl GO 70 80 90
\
> 	 CHO


CO
u
E ft
+~ 7
I
- t
"in
CM
O
i
,H|,1|.,, |.|...| 	 .Illlllll 	 |1llllMM| 	 l|l 	 -J+T-T
100 110 120 130 HO ISO 'l6C 173
                        M/ E
                    ANALYSIS OF CYCLOHEXENE AEROSOL, ACID FRACTION METHYL ESTERS.

                    CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK A.

                    (Refer to Chromatogram in Figure 15)

-------
                                                  TENTATIVE
                                                  ASSIGNMENT:
                                                       COOCH3
                                                        CH2OH
          70
90
                           'iST UO  120   133  'HO   ISO  160  170  193

                                                                              B.
(Refer to Chromatogram in Figure 15)

-------
 at m/e  161 is consistent with adduct ion (M  f 02^) .   Prominent
 fragmentation peaks consistent with loss of H£0 and loss of C^OH
 are observed at m/e  115 and 101 respectively, while the less intense
 peak at m/e 73 corresponds to loss of HCOOCH3.

 Finally,  consider the mass spectrum shown in Figure 19, corresponding
 to chromatographic peak D.  This rather interesting product is
 tentatively identified as the 5-nitrate ester of the product considered
 in the case above.  The peak for protonated parent-ion is observed at
 m/e 178.   Prominent fragmentation peaks consistent with loss of
 CHoOH and loss of HN03 are observed at m/e  146 and 115, respectively.

 In summary, four bifunctional carboxylic acids indicated below have
 been tentatively identified in the acid fraction of cyclohexane aerosol,
 Oxidative cleavage of cyclohexane in the presence of ozone and hydrogen
 peroxide  can reasonably be expected to yield 6-oxohexanoic acid.
 (Hydrogen peroxide has been observed in the atmosphere and in model
 systems '°1)_)  Further oxidative decarboxylation might  then yield the
 5-oxopentanoic acid.   The 5-hydroxypentanoic acid might be formed via
 disproportionation from pentanedialdehyde.
     6-OXOHEXANOIC 5-OXOPENTANOIC  5-HYDROXYPENTANOIC 5-NITRATOPENTANOIC
          ACID          ACID            ACID              ACID
Formation of 5-nitratopentanoic acid is of  interest  for a number of
reasons.  First, the presence of nitrate esters in the atmosphere  could
be significant from the standpoint of health effects.  In general,

-------
(Jl
g
•H
8_


8.
v£O
£
jjp-
&D
L0_
So
§
EO
SB

D
fcL
0



















TENTATIVE
ASSIGNMENT:
CCOOCH3

s.



CH2ONO2








\ 	 I1"
10 SO























i








l''»|'"'|..i., 	 , 	 , 	
60 70 GO SO 1




\, i, ,,],,]

in
£
ii
'a>
1
-t- *
o1
I
i
§







„

I 0
I3""
u ."
§ "s










ii





00
1
"E
+ *
i
i
00 110 120 130 140 ISO 160 170 160 13
                   M/ E
         FIGURE 19. ANALYSIS OF CYCLOHEXENE AEROSOL, ACID FRACTION METHYL ESTERS.
                   CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK D.
                   (Refer to Chromatograms in Figure 15}

-------
 the  fate  of  nitrogen  oxides  in  the  atmosphere  has  long  been  a question
 of interest.   Formation  of nitrate  esters  represents an additional
 "nitrogen sink"  in  the atmosphere.   Currently  there is  considerable
 controversy  regarding the formation mechanism  of peroxyacylnitrate
 (PAN)  in  the  atmosphere.  Nevertheless,  it  is  interesting  to speculate
 as to  possible formation mechanism  for nitrate esters.  This might
 involve the  formation of a charge transfer  complex between the olefin
 and  N02>  which would  result  in  enhanced  reactivity of the  vinyl
 hydrogen.                                A+
                  CYCLOHEXENE - N02 CHARGE TRANSFER COMPLEX
One can speculate that free radical oxidation could result in a peroxy
species at the  vinyl hydrogen, and it is possible  that a concerted
induced homolytic peroxy fission followed by addition of the complexed
NC>2 could result  in a nitrate ester.  This suggested mechanism is
somewhat similar  to that previously proposed for cyclic sulfite formation
from olefins and  sulfur dioxide'  '.  It is also possible that the
mechanism is entirely free radical in nature, as has been  suggested
for PAN formation.  However, in the absence of additional supportive data,
further mechanistic suggestions appear to be unwarranted.
Cyclohexene Aerosol Neutral Fraction
The water insoluble neutral fraction of cyclohexene aerosol was isolated
as described previously, and was analyzed by gas chromatography and
GC-MS.  Both methane and helium chemical ionization spectra will be
considered.  The reconstructed gas chromatogram for methane CI is
shown in Figure 20; that for helium is shown in Figure 21.
                                57

-------
00
                     10   20  30
                     SPEETFU1
                  10  SO   60   70  90  3B
"t	|""l"ii|	1	;ini|iiii|	i	|	.j.u.,.,11,	;»IHIIIHMI.|II»I	
 100  110  120  130  110  ISO 160  170  180  190 200  210  22C
230 2*
                            260
                     280
                                                     328
                                               310
                                                                               380 330
                        FIGURE 20.  ANALYSIS OF CYCLOHEXENE AEROSOL, NEUTRAL FRACTION.
                                     RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
Ul
vD
T	1
 13 28  JB •«
 3TCTFl> MJflFH
                                                                     -WO
                                                                               SOB SIO
                   FIGURE 21.  ANALYSIS OF CYCLOHEXENE AEROSOL, NEUTRAL FRACTION.
                              RECONSTRUCTED GAS CHROMATOGRAM FOR HELIUM CI-MS.

-------
 It  should be pointed out that chromatographic peaks C and E represent
 the  only notable case of contamination observed during this year's
 study.  The two components have been identified as dimethyl gluturate
 and  dimethyl adipate.  Although the source of contamination remains
 uncertain, these two compounds are widely used industrial plasti-
 cizers.  Glutaric or adipic acids might be expected in the acid
 fraction but should not appear in the neutral fraction.  Moreover,
 the  neutral fraction was never treated with diazome thane, thus methyl
 esters could not have been formed from aerosol products.

 Of the neutral aerosol products for which possible structures have
 been presented, the most interesting is considered in Figure 22.
 This mass spectrum corresponds to chromotographic peak A.  The several
 structures shown are consistent with both the methane and helium
 chemical ionization spectra.  The methane spectrum suggests a com-
 pound of molecular weight 96.  Peaks are observed consistent with
 parent ion, protonated parent ion, and parent minus hydride ion at
 m/e  46, 97, and 95 respectively.  Peaks consistent with adduct ions
 are observed at m/e  125 and 137  (M + C2H5) and M + C-jH^) .  In the
 helium mass spectrum peaks are observed consistent with parent ion
 at  m/e  96, and respectively, loss of H, HCO, and (HCO + C2H^) from
 the parent ion, at m/e 95, 67, and 39, respectively.

As indicated below, a cyclic, unsaturated aldehyde might be expected to
 form upon internal aldol condensation and dehydration of hexane
dialdehyde.  Hexane dialdehyde is known to form upon ozonalysis of
 cyclohexene .
                                    'CHO                       CHO
                    H20
                                   60

-------
                   SPECTRUM 1
               SL
Oi
II
0>
"E
                   POSSIBLE STRUCTURES:
               '—  CHO         CHO
                                        ~  t.  OR
                                        0
                                            0
                                        ">   r>.
                                        CM   o.
                                              ll .. J.
                                                                CHO  -
                 SO  60  70  90  3B  100  110 120 130  VtO  ISO 5
                    M./ T,
                   SPECTRUM 2
"1-
CD_

r-
^"•j
i^'-" -
,:..,
^
•j

-------
Additional  unsaturated aldehydes might be accounted for on the basis
                                             (49)
of  a  reaction  described by Cockcroft, et al.      Upon photoisomer-
ization  of  4,5-dihydroxyoxepin, the three products shown below were
obtained.   Reported electron  impact spectra of all three isomers
are similar to the helium chemical ionization spectrum for the
cyclohexene aerosol product.
   o
CHO
                                  CHO
  4,5-DiHYDROXYOXEPIN

Consider now the spectrum in Figure 23, corresponding to chromatographic
peak 3.  The methane spectrum is consistent with a molecular weight
of  112.  Peaks are observed appropriate for protonated parent ion  at
m/e  113,  and adduct ions at m/e  141 and 153 (M + C2H5 and M + 03115).
The helium spectrum shows fragments at m/e  97, 69, and 43, consistent
with loss of CH-j, CH3-CO, and CH-j-CO-CO from the parent ion.  A
possible structure consistent with the data is shown in the figure.

Finally, consider peak D, the mass spectrum for which is shown in
Figure 24.  In the methane spectra, peaks consistent with protonated
parent and loss of water from protonated parent appear at m/e 99
and 81, respectively.  In the helium spectrum, a peak at m/e 81 is
similarly observed, as is a peak at m/e  31, consistent with the
fragment CH2—OH .   The data suggest  a primary alcohol of molecular
weight 98.  Additional fragmentation is rationalized in the figure
and a possible structure is shown.

Structures presented in the last two figures suggest reactions of
greater complexity than those considered up to this point.  Moreover,
                                   62

-------
              g    SPECTRUM 1
              8_
                                   CO
                           POSSIBLE STRUCTURE:

                                      0 0
                                      ii n
                           CH3-CH=CH-C-C-CH3
                                             t.   Tn
                                              in  x
                                             I   CO
                                              CO  O
                         CO
                     r-   m

                     *~   n
                     II   Q>
                50 60  TO  90  30  100  110 123 1JO  1W  ISfl l«i !'
                   «y e:
                   SPECTRUM 2
              '•'•i.
                        CO
Ml
                               O
                               O

                               X
                               O
      CO

      n
                                         CO
                                       Oi
                                           '
                20  38  10  SO  60  TO  98  95  1W  DO  12O I JO

FIGURE 23.  ANALYSIS^OF CYCLOHEXENE AEROSOL, NEUTRAL FRACTION.

            CHEMICAL IONIZATION MASS SPECTRA FOR CHROMATOGRAPHIC

            PEAK B. SPECTRUM 1 BY METHANE CI-MS, SPECTRUM 2 BY
            HELIUM CI-MS.

            (Refer to Chromatogram in Figure 21)
                                  63

-------
              SPECTRUM 1
8~
0
0-
UlG
llT"
U.B
I*"
O. ~
CM-
ID








0
11
I
m
,?,„
[•""•"I1 	 '*r
POSSIBLE STRUCTURE:
HOCH2CH2CH=CH-CH=C




f
s
00
"E
ll nltllll , ttillt III ll I 'l!' I'ffr

20
-a  so ee
                                   30
            SPECTRUM 2
        B
        OD.
       a.
       uc
       CJbJ.

       *
       0:0
          SS  60 70  98  38  108 118 128  130 ltt 158  168 1*70 IBB  138
FIGURE 24.  ANALYSIS OF CYCLOHEXENE AEROSOL, NEUTRAL FRACTION.
           CHEMICAL IONIZATION MASS SPECTRA FOR CHROMATOGRAPHIC
           PEAK D. SPECTRUM 1 BY METHANE CI-MS. SPECTRUM 2 BY
           HELIUM CI-MS.
           (Refer to Chromatogram in Figure 21)
                                 64

-------
retention of unsaturation may appear surprising in view of the
reaction conditions.  It will be seen, however, that tentatively
identified products in cyclohexene and cv-pinene aerosols similarly
retain such unsaturation.
Cycloh_exe_ne Aerosol Basic Fraction
The quantity of basic matter obtained upon fracttonation of cyclo-
hexene aerosol was extremely limited  (-0.2 mg)  and was insufficient
for satisfactory analysis by GC-MS.  Only in the case of a-pinene
aerosol was sufficient material obtained in the basic fraction to
permit such analysis.  The general question concerning formation of
nitrogen-bases during aerosol reactions is considered in the section
dealing with the a-pinene aerosol.

ANALYSIS OF TOLUENE AEROSOL
Toluene Aerosol Acid Fraction
The water-insoluble acid  fraction  of  toluene aerosol was  isolated
as previously  described,  and was  treated with  diazomethane  to yield
methyl esters  and methyl-aryl  ethers  of the carboxylic  acids  and  phenols.
The derivatized product was  then  analyzed by gas  chromatography and
GC-MS.  The reconstructed gas  chromatogram  is  shown in  Figure 25.
Structures have tentatively  been  assigned for  five  of  the prominent
components and molecular  weights  have been  assigned for two others.
It is notable  that  four  of the molecular weights  shown  have odd values,
suggesting the presence  of nitrogen.

The mass  spectrum in Figure  26 corresponds  to  chromatographic peak B.
The prominent  peak  at  m/e  159 corresponds  to  the protonated  parent
ion.  Strong adduct  ions  corresponding to  (M + C2Hc )  and (M  +  £3115  )
are observed at  m/e  187  and 199  respectively.  This clearly  establishes
the molecular  weight as  158.   Fragmentation corresponding to  loss of
HCOOCH3  and  loss  of  Cl^OH suggests the presence of carboxylic acid
methyl ester.   Loss  of CO suggests the  presence of an aldehyde,
                                   65

-------
                                               CO
                             IBB  118  129  138  118  1SB 168 170 188 138 288 218 228 230 218 250 2GB 27B  288  230
FIGURE 25.  ANALYSIS OF TOLUENE AEROSOL, ACID FRACTION METHYL ESTERS/ETHERS.
            RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
o
-o
6
JL

3
-Q
i £""*-«

I
O
i3_
;
i
1 13
k
•
*L


s_

D
*^~

~-
«3_























f*im fm^ M.JK M. _.
s
II
0)
E
ro
X
o
o
o
o
I
1
I
CM
ii
V

x"
O












«*••
2 ^
ii *
5! 7
E o>

-£? 0 E
+' -1

2

\

( 	 l'«'i"'i|""ii"V
\J
t- CM
E

X
E

fl.,....,f.t.,..f.r...(I.U
TENTATIVE ASSIGNMENT:
COOCH3
^C\
S CH CHO
*~
ii
fli /^ i i
•^ Cri2
P x,^
CH2OH
X
f


r** o>
00 O)
1 II
ft} 4)
1 "E
in in
X X
o4 u

» *
,-! ,iii i ,/ *? i i

             63   70   83   93   130  113  123  133  H3  153   163  173  183  133  233  210
      FIGURE 26. ANALYSIS OF TOLUENE AEROSOL, ACID FRACTION METHYL ESTERS/ETHERS.

                CHEMICAL IONIZATION MASS SPECTRUM (METHANE)  FOR CHROMATOGRAPHIC PEAK B.

                (Refer to Chromatogram in Figure 25.)

-------
and loss of H20 suggests the presence of alcohol.  The structure shown
in the figure is consistent with such fragmentation and with the
molecular weight.

The mass spectrum in Figure 27 corresponds to chromatographic peak A.
The prominent peak for protonated parent ion at m/e 145 and the adduct
ions at m/e 173 and 185 indicate that the molecular weight is 144.
Observed fragmentation  is analogous to the previous case.  The
tentatively assigned structure is shown in the figure, and corresponds
to the next lower homologue of the product considered in the case
above.

The mass spectrum corresponding to chromatographic peak C is shown
in Figure 28.  Peaks for the protonated parent ion and for the adduct
ions indicate that this product has a molecular weight of 167.  Since
this is an odd number,  the presence of nitrogen is indicated.  Fragments
corresponding to loss of NO and N02 confirm this and  indicate  that
the product is probably a nitro compound.  Fragmentation corresponding
to loss of CH^ may stem from  the methyl ether group.  Taking these
and other factors into  account, the structure shown in the figure
is proposed.  Substitution is not specified.  It should be emphasized
that the original product would be a phenol.  Treatment with diazo-
methane leads to the formation of the methyl ether.

The mass spectrum in Figure 29 corresponds  to peak D  in the  reconstructed
gas chromatogram.  Almost all of  the comments concerning  the  last product
apply  to the one considered here, with  the  exception  that  this spectrum
does not show a  fragment corresponding  to  loss of  NO.  Nevertheless,
the evidence strongly  suggests  the proposed structure.  Thus,  it  appears
that two isomeric nitro-cresols are  formed.  Since chromatographic
component D has  a longer retention time  than component C we  can speculate
that substitution of component D  renders  it more polar than  C.  Some
possibilities are shown below.
                                     68

-------
                     TENTATIVE
                                 MORE POLAR
                                 (GC PEAK D)
LESS POLAR
(GC PEAK C)
Consider next,  the mass  spectrum of chromatographic component  E,
shown  in Figure 30.   The spectrum shows  a strong protonated parent
ion and adduct  ions  indicating that the  molecular weight is 183.
Again, the  presence  of nitrogen is indicated.   There are peaks
consistent  with loss  of  NO, N02,  and H20.   The  strong peak for loss
of H20 is especially  suggestive of an  alcohol.   Consistent with
these data  the  substituted benzyl alcohol  is  proposed.

The structures  below  summarize  our  findings to  date  concerning the
acid fraction of toluene aerosol.   The formation  of  such homologous
degradation products  parallels  our  findings concerning  the aerosol
reactions of cyclohexene.  In addition we  see that  both the methyl
group and the benzene ring of  toluene appear  to  undergo attack in-
dependently.  It is noteworthy  that  the  most abundant  components
of this fraction appear  to be nitrogen-containing compounds and
that nitration  of an aromatic ring  appears  to be  taking place  during
aerosol formation.
      COOH
CH     CHO
 I
CH2
     CH2OH
                          COOH
                        CH
                       CH2OH
                               CHO
                                          CH3
                                            NO 2
                                       (TWO ISOMERS)
                                                             CH2OH
               NO2
                                    69

-------
                B
-vj
o

B-

o
~
p_

1
D
>
!*-
8.


B.
D
•* —
•*-
D ,


















. ,,,..T


n
I
u
O
O
O
X
5










'"1"'




in
00
ii
0)
"E









i 1 ii
. .,,,,..,,,.I1, ,,.,.,,











































f*.
r-
^
II
a
1
c
c
^
2













: ?
r
>" i
> „
. 1
• c
3

"i
^
f

Iflllllll









<>.
a

j

(N
:

c
»
I


















il..|.m|


















TENTATIVE ASSIGNMENT:
^ COOCH3
«- '
" ^ C NV
-| CH CHO
| CH2OH
y


^ ~
n »
a> -2?
^ +E
C ^
o °
i *
II
jy.l, ..^ jf.| ..11,...,^..! ...jf.-.l...
                   60   70   80   30   100  110  120  130  MO  ISO  160  170  180  190
                        fl/ t
      FIGURE 27. ANALYSIS OF TOLUENE AEROSOL, ACID FRACTION METHYL ESTERS/ETHER.
                CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK A.
                (Refer to Chromatogram in Figure 25)

-------
   O


—


_



-
_
_





















CM
IT)
S 3 Hv
11 ii ~E
| | *
E E i
* 0 V
Z Z I
i i s
1 1 1 II
ll -I.J.I ...l.llllllll [111, 1. II 111 1 .,1 1,1

TENTATIVE
ASSIGNMENT:
CH3
( ) -OCH3
N02
CO
to
II
I
(D
2
o
0) 1
+ " ^
ID
I-f
l^
-t- O
s -•-
, . ,,.1 .1 ,
j — ,rn | r.,.|.tii tn.^in \.. t tin, — j./.
60
          7(5   8(5   30   190  110  120   130  110  ISO  160  173  I80  J90 2<5(5  210
          M/ e
FIGURE 28. ANALYSIS OF TOLUENE AEROSOL, ACID FRACTION METHYL ESTERS/ETHERS.
          CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK C.
          (Refer to Chromatogram in Figure 25)

-------
(V
8

fc_

£*-
0.
$p_
*8
L"
£?_
••

8.


8_

o





















.,,., — J — ,.l



















1 	 1 	 j' •••!-- |


















1
•'









c

T

"

c
4
:

"f »""|"









\j
si

i
D
E
?
f

C
z
Ilk
1" ' ! '







































' !





















TENTATIVE
$ ASSIGNMENT:
1 C"3
X r/\,
+ /" X
i 1 (r7TN°2
OCH3


(O
2 00
^ o
II 
-------
          8
~J
(JO
         8-

^™

-





—





~~

—





—

_
-
-



























1
— , — jfr


























11 1 1 I III 1
MM, ijf, Il|if


























Jill



























ill!




























T



























••'•'"


























,1
i






















































i
1"

















S
r~
ii

-₯
E_
o
z

X
1 1
*•'


5
r
i
0)
T

g
<
+
^

X
«
^















,




















































s

11
-?
c


o
CM
X

"x
^
f









\ \ .,
Mil











1
!


























1 liil

TENTATIVE
ASSIGNMENT:
CH2OH
t
/^






















i
T

C 7T~ N°2

\^r
^xXx
" OCH3

r-

ii
-₯
E
"x

CM
f™"
CM
II
0)
E
X
CM
O
-f
S
u lli.l Hill ii i>

                                                                           190  200  ZlQ  220
    60   70   80   90    100  110  120  ?36  JlO   150   JGO  I7U
         fl/  E
FIGURE 30.  ANALYSIS OF TOLUENE AEROSOL, ACID FRACTION METHYL ESTERS/ETHERS.
           CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK E.

           (Refer to Chromatogram in Figure 25)

-------
Toluene Aerosol Neutral Fraction
The water-insoluble neutral fraction of toluene aerosol was isolated
as described previously, and was analyzed by gas chromatography and
GC-MS.  The reconstructed gas chromatogram is shown in Figure 31,
Note that tentative assignments have been made for four major peaks
and molecular weights have been assigned to six others.  In this
fraction one major and one minor peak have odd molecular weights,
suggesting the presence of nitrogen in these compounds.

The mass spectrum in Figure 32 corresponds to chromatographic peak  B.
Peaks assigned to protonated parent ions and adduct ions suggest a
product of molecular weight 112.  Loss of CH4 may arise possibly from
an intact methyl group,  while loss of H-CO suggests the presence of
aldehyde.   Consistent with these data a tentative structure is shown
in the figure.

The mass spectrum corresponding to chromatographic peak F is shown
in Figure 33.  The spectrum is practically devoid of fragmentation,
but a strong protonated parent peak and reasonably prominent adduct
ions strongly suggest a molecular weight of 126.  Very minor fragments
are evident which correspond to loss of l^O and CH^.  This spectrum is
particularly ambiguous.  A possible structure is shown in the figure.

The mass spectrum in Figure 34 corresponds to chromatographic peak A.
The spectrum is wholly devoid of significant fragmentation; peaks at
m/e 97 and 99 are considered spurious background in view of the
unrealistic M-12 and M-14 values.  However, the molecular weight can
be determined with some confidence as 110 in view of the adduct ions  for
(M + C2H5> and (M + €3115).  The proposed structures shown in the figure,
however, are extremely resonance stabilized and could reasonably account
for resistance to fragmentation under conditions of chemical ionization
mass spec trome try (methane) .
                                     74

-------
                                   
-------

ED
sr;
itj-
£
-------
-0
               n.
              CV.



              O
                                     O   T-
 II


 O)
                                         II


                                         01
T  O
 i    I
+   ro
I  T.
                                   Illlltll
                                                a>
                            IT)
                            in
CM
   liliii|iiii|iiM|lHt|...i|iiii[tiii)n.ni.4i
                                                                       TENTATIVE ASSIGNMENT:
                                                                                 CH3

                                                                                 I

                                                                                 C
                                                                            CH ^  ^CH2OH

                                                                                CH
                                                                        (O
CO
o
         FIGURE 33.  ANALYSIS OF TOLUENE AEROSOL, NEUTRAL FRACTION.

                     CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK F.

                     (Refer to Chromatogram in Figure 31)

-------
00
             O
             8_
IJ
Q_

LUO
            U.G


            UJ
            o
            0:0
             0
                                                 CM
                                                 0
                                                       in

                                                       ii

                                                       •SJ
                                                       E
                                           CO
                                           0
                                                                                TENTATIVE
                                                                                ASSIGNMENTS:
                                                             CH2

                                                              II
                                                              C
                                                                     CH
                                                                     CH
                                                                 CHO
                                                                                  OR
                                                                            CH
                                                                         CHO
                                                                                               CHO
                              r-^
                70  80   30   100  110  120  130  HO  150  160   170  180  1SG  ZOO
                    «/ E
         FIGURE 34. ANALYSIS OF TOLUENE AEROSOL, NEUTRAL FRACTION.
                   CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK F.
                   (Refer to Chromatogram in Figure 31)

-------
Finally, consider  the mass spectrum  in Figure  35,  corresponding  to
chromatographic component E.  The spectrum  is  rather weak, but
significant adduct  ions permit determination of  the most probable
molecular weight as  116.  The prominent peak corresponding to loss of
H20 strongly suggests the presence of alcohol  functionality.  Consistent
with these data, and with fragmentation of  apparently related products
of toluene, a possible structure is  shown in the  figure.

The structures shown below summarize the finding  to date concerning
the water-insoluble neutral fraction of toluene aerosol.  As in  the
previously considered product mixtures, we  again  observe sequential
oxidative degradation to give a series of related  oxidation products.
                   CH-j
                    I             CH3            CHO
                CH    CHO      CH^    CHO     CH^ "cH^H
                CI~L           CH5^  ^CH2OH
                       °h    CH°        ™    CHO
                       |           OR    I     CHO
                       CH^              CH.    CHO
                         CHO              pu-^x'
ANALYSIS OF q-PINENE AEROSOL
In studying the tf-pinene aerosol, extensive  reference has  been made
to the literature concerning  reactions of  this compound.   This was
especially helpful during analyses of the  neutral  fraction, where
mass spectral data indicated  the presence  of many  isomeric products.
Salient aspects of the literature survey are summarized under the
headings below.  Detailed reviews of the chemistry of o/-pinene and
other terpines have been prepared by Simonsen     and Templeton^
                                    79

-------
00
o
              g
              O
             Ui
             <5
                               o>
                               ii
                               +
                               I
                                                       in
                                                       I
                                                            in
aa
~E

m

O
                                                                   CH
                                                                  CH2OH
          TENTATIVE
          ASSIGNMENT:

          CHO
              CH2OH

                  M.I|.|..|ltl||Mli
                 7CJ    80  30    100   110   123  1JO  HQ  150  160  170  a;
                       M X* P
        FIGURE 35.  ANALYSIS OF TOLUENE AEROSOL, NEUTRAL FRACTION.
                   CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK E.

                   (Refer to Chromatogram in Figure 31)

-------
Qf-Pjnene Acid Fraction

The cy-pinene acid fraction was isolated as described previously, and
was treated with diazomethane to form methyl esters of the carboxylic
acids.  The material was then analyzed by gas chromatography and
GC-MS.  Figure 36 shows the reconstructed gas chromatogram.  The
indicated chromatographic component whose mass  spectrum  is shown in
Figure 37 has been unequivocally identified as  methyl-cis-pinonate by
comparison with an authentic sample of methyl-cis-pinonate.  The mass
spectrum of the latter compound is shown in Figure 38.   The mass spectral
fragmentation pattern for this compound is rationalized  in Table 4-.
A second analysis of the sample by GC-MS verified the presence of
pinonic acid.  The results indicated, in addition, that trans-pinonic
acid may be present along with the more abundant cis isomer.  As
indicated below, interconversion of the isomers in the presence of
ultraviolet irradiation can be postulated.  Tentative identification
of the chromatographic peaks corresponding to the isomeric pinonic
acid esters is indicated on the reconstructed gas chromatogram in
Figure 39.
                           COOH                                  COOH
                                   hv
      H
     CIS-PINONIC ACID                     ~ TRANS-PINONIC ACID
                                   fl

-------
00
               ID
                                              METHYL PINONATE-
                      T
T
T
r
                                   ae   tew  110  120  13s
                       le   ze   30   ie   se   so   70
                       SPECTPUM fOJ^GR
                FIGURE 36.  ANALYSIS OF a-PINENE AEROSOL, ACID FRACTION METHYL ESTERS.
                           RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
> 90
80
<
_ 70 gj

W
_ 60 <
p
o
_ 50 ,
Ul
0
^
^»0 H
•• • ^
^
Ul
- 30 £j

_ 20

- 10
0
fi...,..-.f..l.l...-l 	 ..1 |



i

i
j
- i
i
n
•«.








i .1 It ii 1




i


a




r-
r-
T™
1
-S
C
c









O5
(N

II
"E




j




i
Ui,....| lll]..l


ASSIGNMENT = Methyl Pinonate







0)

it
01 ^ a>
rn r** —
₯ co E
II ^
Qi *™ +
"^ Q> 00 X
P *"" S
~E 1


  10   SO   60   70   80   30   100  110  120  130  11O  ISO  160 170 180 130 200
         E
FIGURE 37.  ANALYSIS OF a-PINENE AEROSOL, ACID FRACTION METHYL ESTERS.
           CHEMICAL IONIZATION MASS SPECTRUM (METHANE)
           CORRESPONDING TO METHYL PINONATE.
           (Refer to Chromatogram in Figure 36)
                                  83

-------
100
CT>
_ 90 &>
n
03
_ so ~E
70 0.
UJ
CO
- 60 ^
u.
O
- 5C yj
0
- 40 ^
LU
U
' 30 DC
0.
. 20


- LO
1

j













II 	 |]
CM
II
0>
E





^
--
ii
0)
E





AUTHENTIC


METHYL CIS-PINONATE


CH3
Kr--*o
X. COOCH3
T 1
v
\
\
X

O)
,—
n
"E




Wr^llnJ...! 	

1
	 ^
***^

0)
r^ r-
r II
II W
~E OT ^
7 +"
ii x
- I
,iil'M..I 	 ill 	 L. .11. nil 	
10  SO  S3  70  80  90  100  110  120 130 110 ISO 16J 170 190  190 200
    M/ E
   FIGURE 38.  CHEMICAL IONIZATION MASS SPECTRUM (METHANE) OF
               AUTHENTIC  METHYL CIS-PINONATE
                                 84

-------
       TABLE 4.   RATIONALIZED FRAGMENTATION OF AUTHENTIC METHYL

                 PINONATE UNDER CONDITIONS OF CHEMICAL IONIZAT10N

                 MASS SPECTROMETRY (METHANE)
Process
                      Fragmentation of Parent-Ion


                                    Fragment
Loss of H
                                                       Protonated parent-ion of
                                                       methyl pinonate,
                                                           +  m/e  =
                                                                      m/e
                                                                      181
Loss of CH OH(MH -32)
Loss of HCOOCH.(MH -60)
Transannular cleavage
                                CH3-C'  _ H
                                CH
                                  3     H

                                     0
                               CH3     H
                             CH
                                    H

                                   CH2
                                         .C/-CHS
                                                                      167
                                                                      139
                                                                      111
Iransannular cleavage
                                                                        99
Transannular cleavage
                                II
                            CH3-C
                                H-^<3)
                                                                        71
Transannular cleavage
                               CH
                                    85

-------
Tentative identification of a secondary reaction product of Qf-pinene
has been made.  The compound corresponds to peak A of  the reconstructed
gas chromatogram in Figure 39.  Its mass spectrum is shown in Figure
40.  The compound is tentatively identified as the homologous acid
resulting from oxidative decarboxylation of pinonic acid, that  is,
pinononic acid.  Its mass spectral fragmentation pattern is rationalized
in Table 5.  The compound is referred to as pinononic  acid in accordance
                              (52)
with nomenclature of Simonson     .  In previous reports  it has been
referred to using the derived name "nor-pinonic acid".

In summary the data indicate that under the simulated  atmospheric
conditions a-pinene undergoes oxidative cleavage to pinonic acid.
The pinonic acid produced may then be oxidatively decarboxylated
(tentative) to the next lower homologous acid.
                                      COOH
   a-PINENE
PINONIC ACID
                                                              COOH
PINONONIC ACID
Formation of pinononic acid may also occur without intermediate for-
mation of pinonic acid.  In a study of vapor-phase ozonation of
cv-pinene by Spencer, et al'5 ', an intermediate ozonide, C10H16°5>
was hydrolyzed to give pinononic acid directly.  The authors speculate
that  the intermediate was either verbenone oxoozonide or verbenone
peroxide ozonide
       VERBENONE OXOOZONIDE
        VERBENONE PEROXIDE OZONIDE
                                   86

-------
00
-o
                 B
                 8_
PEAK   TENTATIVE ASSIGNMENT


  A   pinonic acid, methyl ester
                        B    cis-pinonic acid, methyl ester
                        C    trans-pinonic acid, methyl ester
                          10   20   33   13   SO    €0   70   60   30   100  110  120

                          SPECTRM NUTE8R
                 FIGURE 39. ANALYSIS OF a-PINENE AEROSOL, ACID FRACTION METHYL ESTERS.
                            RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
          g
00
oo
•"—
8_
8_

$P-
E

-
ft
n
68.
M
Co
3C
^ «••
5
JR.
B.

D
•"*—•

—





,-
ii

II
CD
1 E
1 1 	 Ii 1 	
10 160 170 180 130

         FIGURE 40. ANALYSIS OF a-PINENE AEROSOL, ACID FRACTION METHYL ESTERS.

                   CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK A.


                   (Refer to Chromatogram in Figure 39)

-------
     TABLE 5.  RATIONALIZED FRAGMENTATION  OF TENTATIVELY IDENTIFIED

               METHYL PINONATE UNDER  CONDITIONS OF CHEMICAL IONIZA-

               TION MASS SPECTROMETRY (METHANE)


                           0


                       CH -C-w—^      H  bCH3             +
                      CH3           ,1  *         (MH)  , m/e  = 185.
                                    n +

                      Fragmentation of  Parent-Ion



       Process                      Fragment                          m/e



Loss of H 0(MH+-18)                 	                           167


                                      0

                +                 CH,-cV   H
Loss of CH3OH(MH -32)             c.»\^~^» +                        153

                                  CH3      Cs°


                                      0


Loss of HCOOCH (MH+-60)           M^ ^~^^                          125





                                         y?CH3
Transannular cleavage

                                CH,

                                         ^TH

                                   +OH

                                CH,-C^     ^XCH3
Transannular cleavage                >     >ru_
                                    H  (3)   12) \^ f*»
Transannular cleavage                J>     <^                           71
Transannular cleavage              ^"'^    _^ ^                    115
                                    89

-------
In view of the identification of pinonic acid in model aerosol, this
compound was sought in rural air particulate matter collected^ in
the Blue Ridge Mountains.  The sampling site was located in the
Pisgah National Forest, about 30 miles northeast of Ashville, North
Carolina.  There was extensive conifer and hardwood forest cover.

The organic acid fraction of the air particulate matter was isolated
and treated with diazomethane, and the methyl esters were analyzed
by GC-MS.  The reconstructed gas chromatogram is shown in Figure 41.
The mass spectrum shown in Figure 42 identifies the indicated
chromatographic peak as pinonic acid, by comparison with the mass
spectrum of an authentic sample (Figure 37).

The presence of pinonic acid was similarly sought in urban air
particulate collected in New York City (lower Manhattan).  No
detectable concentration of the acid was observed.
Cf-Pinene Neutral Fraction
The water-insoluble neutral fraction of o-pinene aerosol products was
isolated as previously described, and was analyzed by gas chromatography
and GC-MS.  The reconstructed gas chromatogram is shown in Figure 43.
Molecular weights tentatively determined from the mass spectra are
shown on the chromatogram.  The data indicate that a number of isomeric
reaction products have been obtained.  In order to aid identification
of these products various authentic terpenoids were obtained for use
as reference compounds.  Selection of reference compounds was based
on a study of known reactions of a-pinene.
* Sampling and analysis of atmospheric particulate matter was conducted
  under support of the previously cited program, "Haze Formation: Its
  Nature and Origin"'3 '.
                                   90

-------
                                                        100  110  120  130
      0
10   20   30
SPECTRUM NUMBER
FIGURE 41  ANALYSIS OF BLUE RIDGE MOUNTAIN AEROSOL, ACID FRACTION METHYL ESTERS.
FIGURE 41. ANA           QAS CHROMATOGRAM FOR METHANE C.-MS.

-------
               5L


               §_
              ^(0
              "r-_J
              tup
              tQS-4
              u,o
              QLO_
               O
                                                                                 n

                                                                                 OJ
                  60  70   80   30   100  110  120  139  110 ISO  160  173  180  190

                      MX E

FIGURE 42.  ANALYSIS OF BLUE RIDGE  MOUNTAIN AEROSOL, ACID FRACTION METHYL ESTERS. CHEMICAL

            IONIZATION MASS SPECTRUM (METHANE)  CORRESPONDING TO METHYL PINONATE
            (Refer to Chromatogram in Figure 41)

-------
                                       00
                                       CD
                                                                           H
"go	60	TO	«j' 30	100 VlO  'iffl 'l30 1« ISO 'lW 'l™ 'lW                                30D 3l° 3ZD



FIGURE  43.  ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.

            RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
                                                        (54)
Studies Based on Known Reactions of 
-------
                        (47)
 A study by Mayer, et al    ,  indicates that when Q"-pinene is simply
 irradiated using a Pyrex filter, several products are obtained (low
 yields).  One of the most abundant is d-limonene.  Limonene itself
 can react in the presence of  nitrogen oxides,  or by "autooxidation"
 to form carveol.
         CH,
                                                     HO
                                          AUTOOXIDATION
                                          OR NOx + 02
      a-PINENE
CHj-•  •*-uri2
 LIMONENE
                    CARVEOL
 Jefford  reports  that  reaction  of  photogenerated  singlet oxygen with
 tt-pinene results  in greater  than  99  percent yield of ^rans-3-hydroperoxv-
 pin-2(10)-ene.
                                                          OOH
     '-PINENE
TRANS-3-HYDROPEROXY-PIN-2(ini-FNF
A reaction of this type could presumably occur in aerosol containing
a suitable sensitizer.  The hydroperoxide may undergo unimolecular
or bimolecular reaction to yield the hydroxy or carbonyl compounds,
pinocarveol and pinocarvone.  (Although neither pinocarveol nor
pinocarvone could be obtained for use as reference standards, note
that "in the chromatogram, Figure 43, two peaks have molecular weights
corresponding to that of pinocarveol, 152'.
                                     95

-------
                  PtNOCARVEOL
                       (59)
                                PINOCARVONE
Dupont and Zacharowitz   ^ showed that oxidation of Q'-pinene in the
presence of selenium dioxide yields myrtenol and myrtenal.  This
is significant with respect to atmospheric chemistry because both
selenium dioxide and singlet oxygen lead to allylic oxidation.   Both
compounds below were obtained for use as reference standards.
                                   CH2OH
                        Se02
                                            CHO
                                 MYRTENOL
                                          MYRTENAL
                        (60)
Henderson and Sutherland "' observed that hydrogen peroxide treat-
ment of Q'-pinene in acetic acid yields principally limonene, a-terpineol,
borneol, and p-methane-1>4,8-triol.
            H,0;
2^2
          ACETIC
           ACID
                                                                CH
                                                                  3 OH
 or-PINENE
        LIMONENE    a-TERPINEOL   BORNEOL
p-MENTHANE-
1,4,8-TRIOL
Although the above is a liquid-phase reaction in acetic acid,  it
should be pointed out that hydrogen peroxide has been observed in
the atmosphere and in model systems^b1^.  it was observed in irradi-
ated mixtures of hydrocarbon/NOx and in the urban atmosphere during
periods of photochemical smog formation.
                                    96

-------
Finally, oxidation of cv-pinene by perbenzoic acid has been described.
                                                                     (62)
The reaction is of interest in that perbenzoic acid oxidation might
be expected to yield products similar to those formed in the presence
of peroxyacyl nitrates.  The perbenzoic acid oxidation was shown to
yield pinene oxide.  Treatment of pinene oxide with water at 115-112 C
yields pinol.  Hydration in the presence of dilute acid leads to ring
fission and formation of sobrerol.
              PERBENZOIC
                ACID
   a-PINENE
i-PINENE OXIDE
PINOL
 CH3-
SOBREROL
Of the cy-pinene reaction products described above, those listed below
were obtained for use as reference standards.  Commercial samples
of sobrerol, pinol, and pinene oxide could not be obtained.
           (1) d-limonene
           (2) terpinolene
           (3) fenchyl alcohol
           (4) fenchone
           (5) 1-borneol
           (6) Q'-terpineol
              (7)  cis-terpin (hydrate)
              (8)  trans-verbenol
              (9)  1-carveol
             (10)  myrtenol
             (11)  myrtenal
Mixtures of  these authentic compounds were analyzed by GS-MS.  The
reconstructed gas chromatograms are  shown in Figure 44.  Mass spectra
are shown  in Figures 45-50.   In the  paragraphs below, the comparison
standards  have been grouped by molecular weight, and a-pinene neutral
products of  appropriate molecular weight are compared in turn with the
reference  spectra.  A  reconstructed  gas chromatogram showing the
                                    97

-------
 B



 8



JS-
ls_
B_
                 o
      1C  2C  30
      SPeCTHLM NLKER
                           60   78   BO
   '108 110  '128  '130  '110  '.SO 160  170  'iBC  'l9B
8.
8-
                           o
                           o
                                                                   |iiii|ini|Mi.,
90  100  110  120 130  1KJ  ISO  160 170 l» Jae
    FIGURE 44.  RECONSTRUCTED GAS CHROMATOGRAMS OF AUTHENTIC
                 TERPENOIDS.
                                     98

-------
        8
        #-*
        RJ
        R,
            ,,  ...,.,
                             1,1
                                  .j.....,,... p...(..,,. j. J.f ,).-. ,JTrrf fj...f1., 471-11 jfTJ r.f in 4 j
           60   73   90   a?   I«  110  120 130 IV  IbO  JKQ  17(3 J8Q
        R.
        8.
       C:
        R.
        D
                                              TERPINOLENE
                                                    CH3
6B  7(5  89  30
    n/E
                               )l(J  120  130  |« ISfl 16«
FIGURE 45.  CHEMICAL IOIMIZATION MASS SPECTRA (METHANE) OF
             AUTHENTIC LIMONENE AND TERPINOLENE.
             (Refer to Chromatograms in Figure 44)
                                99

-------
B

R_

8.


8.
          R_
                              a-FENCHYL ALCOHOL
              1  1  '   ' 1  "y—T""i""i"-j"-r—]—r
            so  79 w  ap   in?  no  120 iap
                                "?•—,•••-•—-|
                                •fl  ISO  160
          B_
                                         BORNEOL
                                             I
              i  1 i  I  i  I
            68  70  60  »  1 HP  IIP 120  130  ! VH ISO  )60
                n/E
FIGURE 46. CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF
           AUTHENTIC a-FENCHYL ALCOHOL AND BORNEOL.

           {Refer to Chromatograms in Figure 44)
                            100

-------
      B
      8_
     hR_
      5«

      Rj
                Ju
         ...M....1~..T~TTrnT.'"]-<
         60  70  W  3

            rt/E
                                         a-TERPINOL

                             llL
                               	i—'V—i—-(""'""V  ' !
                     11(3  123  !30 1KJ  Iti3 16?  173 Iflfl
R_
      B_
      S.
      R_
                                         MYRTENAL

                                              CHO
                             Jl
                             r--T-j-»-n
   ""'""I .....
  eo  19  eo  am   jw 11?  12?  lad
      n/e
                                         ISP  i«i !?a  :«a
FIGURE 47. CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF
           AUTHENTIC a-TERPINOL AND MYRTENAL.

           (Refer to Chromatograms in Figure 44)


                             101

-------
 8_
F-
 S_
          60
                 JW 110 )2(7
                                                H2O
                                     (C1QH2O O2 'H2O)
                                    TERPIN HYDRATE
                                         I
                                        )50 16(3
             fl/E
       8
      PI
       8.
       !«.
       R.
       D
                                     CH.
                                   O
                                  FENCHONE
                    1	!""'""l	I	I	I""!"11]
          GO  70  X  X  IflB 110  120 130 110 ISO 160
FIGURE 48.  CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF
           AUTHENTIC TERPIN HYDRATE AND FECHONE.
           (Refer to Chromatograms in Figure 44)
                          102

-------
g

8.
8_
MP
08-
WL
w
^ yi
fit ^3
jT
R-
o












,111 II. LUl 1
CH3

f^M
\[x^ OH
trans-VERBENOL

I



1
In
1,1,1 1 Illu !ll,M , III All . il
*5   50   SO  70
    nx e
                            30  I W  110  520  130  110  ISO 160
              R.

              S.



              B.
              BJ

              D
                 ee   TO   ee  x  loo no  120  130  110
                     M/E


FIGURE 49. CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF
            AUTHENTIC trans-VERBENOL.

            (Refer to Chromatograms in Figure 44)
                               103

-------
          I
          R.
          8.

         P-
         I*.
          8.
          B.
                                       CARVEOL
            60  70  80  30  100  tl8 120  130 110  ISO 160
              S.

            F-
              8_
             1*.
                                         CH2OH
     MYRTENOL
                        Lf
                            ILL
L  L

                    i	,....,.-.--,-...,-..,- ,-.-,..,  r...,j»ttj
                60  70  80  30  100  118 12C  13B 110
                    n/r
FIGURE 50. CHEMICAL IONIZATION MASS SPECTRA (METHANE) OF
           MYRTENOL AND CARVEOL.

           (Refer to Chromatograms in Figure 44)
                            104

-------
tentatively assigned molecular weights of the aerosol products was
shown in Figure 43.  Surprisingly, these comparisons did not produce
any satisfactory spectral matches.

Borneol, cc-fenchyl alcohol, and a-terpineol have molecular weight 154;
their mass spectra are shown in Figures 46 and 47.  Compare with these
spectra that for chromatographic peak E shown in Figure 51 (the chroma-
togram is shown in Figure 43).  In view of the lack of definitive frag-
mentation, no structure has been assigned to this aerosol product.
Nevertheless, peaks at m/e  155, 183, and 195 are characteristic of a
compound of molecular weight 154.  A weak peak at m/e 137 is consistent
with loss of H20 and suggests the possibility that the product is an
alcohol.  However, the base peak of the compound is clearly at m/e 109.
This fragment is rather small in each of the standard spectra.  More-
over, the standard spectra show characteristic fragments at m/e 81 and
95; these fragments are absent in the product spectrum.  Thus, it is
highly unlikely that the product corresponds to any of the above three
standards.

The mass spectrum shown in Figure 52 has been tentatively identified as
pinononic aldehyde, as shown.  (The spectrum will be discussed in
detail, below.)  There is no evidence to suggest that it corresponds to
any of the above standards.

Verbenol, myrtenol, carveol, and fenchone have molecular weight 152;
their spectra are shown in Figures 48-50.  Compare to these spectra
that for chromatographic Peak B (from Figure 43) as shown in Figure 53.
Although this spectrum is weak, peaks at m/e 153 MH ) and 135
(MH+ - H20) suggest the possibility of an alcohol of molecular weight
152.  The product exhibits some fragmentation characteristics" of all
the above standard alcohols, but is most similar to verbenol.  (Figure
49; note duplicate analyses of the standard.)  However, in comparing
the product spectrum to that of verbenol note that the peaks at m/e 101
                                   105

-------
    g
R.
S_
g^_
Q-
&EL
CUE)
6
prF)
Q_°°~
R..

E3
EJ













1 (III 111 i I
...... ,, i, .,..-,,.. im- M..n( ,-.,,- r
C8H13+
1
ii
V
E


^
in
in -~ —
r- ^ ^-
II + +
a 5
^ + +
^ ^
-H2O (?) =• —
!l f I 1. i r , j j • j| i i 1 Ii tiii mi
Ml >|i I....... w .1....I.- ij ,i , ,.,,,.. , ., I ... ti M . I , . .. . : . | • . ( I I .• . . | .11 . I . . . M . . 1 I 1 .. I M "> '
70   80   9?    100   110   12(3  130
                                                     ISO  160  170  180  190  200
FIGURE 51.  ANALYSIS OF a PINENE AEROSOL, NEUTRAL FRACTION.
           CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK E.
           (Refer to Chromatogram in Figure 43)

-------
           E>
                                                            TENTATIVE ASSIGNMENT:

                                                                    CH3
                                                       ro

                                                       ii
                        U4

                                                  o
                                                 +
                                                  i
ib
f
                                                     U)
                                                     in

                                                     n
                                                     Q)
                                                     ~E
                                                              .,
                                                               rtH
                                                       .,.
                        .,.
.,.
                                                                    160  170   190
FIGURE 52.
  60   73   80   90    100  110   120  130   1 H3   ISO
       U/  E
ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.
CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK C.
(Refer to Chromatogram in Figure 43)

-------
o
GO
                    §
                    8.
                  CL
                  UJE;
                  tote.
EJ
  —
        FIGURE 53.
         70   90   00   100  110  125   130   1«5  ISC  160  170
         M/ E
ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.
CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK B.
(Refer to Chromatogram in Figure 43)

-------
and 93 exhibit reversed intensity.  Thus,  assignment of the  product  as
verbenol would be rather speculative.

The mass spectrum of chromatographic Peak  A (from Figure 43)  is  shown
in Figure 54.  The spectrum is rather ambiguous and does not  permit
assignment of a structure; the molecular weight may be 152.   Observed
fragmentation is entirely inconsistent with the above reference
compounds.

Myrtenal was a molecular weight of 150. Limonene and terpinoline have
molecular weight 136.  Their mass spectra  are shown in Figures 45 and
47.  The mass spectra for the a-pinene neutral fraction did not  indi-
cate the presence of any components with these molecular weights.

In concluding this discussion, it should be emphasized that mass
spectral fragmentation patterns observed for various terpenoids  have
tended to be somewhat variable from one analysis to the next.  Frag-
mentation of these compounds is especially sensitive to instrument
variability.  This can be seen in spectra  for trans-verbenol obtained
upon duplicate analyses (Figure 49).  In the illustrated case, the
analyses were made about 1 year apart during different research
programs.  During the current work,analyses of standards and unknowns
were performed over a period of several days, and reanalysis of
standards was conducted.  Nevertheless, in analysis of these sensitive
compounds, there is a possibility that slight instrument variabilities
may be significant.  Thus, it is not possible to completely rule out
the presence of the reference compounds in the reaction mixture.  Re-
analysis of materials in hand with special attention to instrument
conditions should help to eliminate the present ambiguity concerning
this phase of the study.
Tentative Assignments for q-Pinene Neutral Products--Although the
above studies of authentic compounds have  not led to unambiguous
                                    109

-------
8.
ip-
I8~
bB_
If Ft
is

8_
EJ

o>
n
1


£
n
1


60 70 80 90







.1.1 .
1=





~
CO
in
n
"l
M IIQ 12Z 133 110 ISO ISO 170 180 190
              M/ E
FIGURE 54.  ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.
           CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK A.

           (Refer to Chromatogram in Figure

-------
 identification  of a-pinene  reaction products,  tentative assignments for
 various  products are considered  in this  section.  Some are based upon
 comparison  of the sample  fragmentation pattern with that observed for
 authentic methyl pinonate.  An especially helpful diagnostic has been
 the  fragmentation arising from (proposed) transannular  cleavages.  In
 examining the mass  spectra  shown in Figures 52 and 55-58 below, it may
 be helpful  to refer to Table 4 (p.85 ) in which the fragmentation of
 authentic methyl pinonate is rationalized.

 As a point  of departure,  it would be reasonable to expect formation of
 pinonic  aldehyde, as shown  below.  Formation of this compound would be
 analogous to formation of pinonic acid as observed in the acid fraction.
                     CH3                    CH3
                                             >0CHO
Pinonic aldehyde has a molecular weight of 168.  As indicated in
Figure 43, six components of the neutral fraction have been tentatively
assigned molecular weights of this value.  Fragmentation patterns of
these components have not permitted unambiguous assignment of the
proposed structure to any of the six  candidate products.  Nevertheless,
there is a strong likelihood that this compound is indeed involved in
the aerosol reactions of a~pinene.

Consider the mass spectrum in Figure 52.  This corresponds to Peak C
in the reconstructed gas chromatogram Figure 43.  The compound is
tentatively identified as pinononic aldehyde.  Fragmentation appears
to be analogous to that of authentic methyl pinonate.  Key fragments
are rationalized in the figure.  This product might arise from oxida-
tive degradation of the homologous dicarbonyl, as indicated below.  Its
formation might be considered analogous to formation of pinononic acid,
observed in the acid fraction.
                                    11 1

-------
              ACID
              FRACTION
COOH
                       PINONIC ACID
                                                  CHO
         PINONONIC ACID
             NEUTRAL
             FRACTION
                                                  COOH
                    PINONIC ALDEHYDE   PINONONIC ALDEHYDE
                      (POSTULATED)
Consider next, the mass spectrum shown in Figure 55.   This corresponds
to Peak F in the reconstructed gas cLi.omatogram.  The tentatively
assigned structure is shown.  As in the previous case, the fragmentation
is rationalized in the figure, and appears to be analogous to that  of
methyl pinonate.

Finally, consider the mass spectrum shown in Figure 56.   This corre-
sponds to Peak G in the reconstructed gas chromatogram (Figure 43).
Again, the tentative assignment is based on comparison of the indicated
fragments to those of authentic methyl pinonate.

In view of the last three proposed assignments,  it is not unreasonable
to present the scheme shown below.  This scheme  is not meant  to
suggest mechanistic considerations,  but is clearly a  possible route for
the formation of a series of sequentially degraded oxidation  products.
                                   112

-------
Formation of the proposed products shown above involve,  in each case,
cleavage of the carbon-carbon double bond.   Studies cited earlier
indicated that cc-pinene can undergo "autooxidation" in the presence of
air, moisture, and sunlight to yield sobrerol (see p.  93).  Formation
of this product involves cleavage of the cyclobutane ring with reten-
tion of an intact carbon-carbon  double bond.  If the cyclobutane ring
were cleaved at the alternative internal position, an isomeric diol
might additionally be formed, as shown in the structures below.  Further
oxidation or oxidation/dehydration might be expected to yield the
compounds shown below.
                                  113

-------
~—


Set
y j _


§


gp-
0-
ftg
&

U.ED
QLO_
(f PD
fr— • ^"^^
^p
ft\
f*j
g«-


E3
N_

r>
•m
D




Q
-V
* /
o
i
X
o
II
1 1
c_>
X*


















	 1
""'""1
eo *:

CO
^





^
'**"
tl
1



















^ 5
V-
3
c

I
c
1
<_
z
{.
II
1
c
3

















1,


K3
t
»


i

5

5

g oo
/i

















i 1 1
ii,.. i. i i
93
O)
01
II
-SJ

^

CO CO
I I
0 U
M

s\
Xs ^.
o-° I
+ n
I
CJ

'












t.. .1. I, ...I.I. II 1 1 1

joe no 120


TENTATIVE ASSIGNMENT:


CH3
J^
^^ 0
*C CHO
^^S.^
CHOH
\J/



^_
^
to
V T-
"^
c 11
i i
O -in
oo O £
X CN "
o
+ J. «
i x ^
X
Lm, 	 ,,,,L,; 	 i, .,„ 	 ,„„
........... .,..,. i i i < i ' i i
1 "3fi 1 ^K3 1 ^0 1 60 1 7? 1 9C 1 H
                M/ E
FIGURE 55.  ANALYSIS OF a-PIIMENE AEROSOL, NEUTRAL FRACTION.
            CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK F.

            (Refer to Chromatogram in Figure 43)

-------
     nr>_
    a.
    u.o
      o
      F)
    CL

X X
+ V° T
1 *
.£? CO M
£ z x
0 0 u
\/

II
o>






°*0A
CO
X
O


3
II
^










CO
t—
1!
_»







'•r
CH3 CH3
Jl CH TENTATIVE ASSIGNMENT:
^"^sX^
CH3

-------
                a-PlNENE
                              SOBREROL,
                              POSTULATED
                                OH
                             POSTULATED
                                           CO]
                                                  CH3  CH
                                                         CHO
Mass spectra  tentatively  identified  as  consistent with  these  structures
are shown in  Figures 57 (hydroxycarvotanacetone) and  58,  Peaks  D and H
in Figure 43.  Rationalization of  fragmentation  is  presented  in the
figures.  Spectral  interpretation  in these  cases, would be  greatly
aided by availability of  a  spectrum  of  authentic sobrerol  for com-
parison.  As  mentioned previously, no commercial sample of  this com-
pound could be obtained.  It  is noteworthy, however,  that  in  these
last two spectra, the diagnostic fragments  associated with  the  presence
of an intact  cyclobutane  ring (i.e.,  transannular cleavage  fragments
at m/e  71 and 99;  see Tables 4 and  5)  are  absent or  very weak.

Thus, we suggest that under simulated atmospheric conditions, a-pinene
reactions include cleavage of the  cyclobutane  ring, and that  such
cleavage may  occur with retention  of  the carbon-carbon  double bond.
a-Pinene Basic Fraction
Of the model  aerosol systems studied, only  in  the case  of a-pinene was
there sufficient material obtained to permit analysis of the  basic
fraction (see Table 3 p. 29  ).  Although the basic fraction  was  negli-
gibly small in all cases, analysis of the basic fraction for  at least
                                     116

-------
E

R.
0-
t^JEO
r?

U.E)
SL
          UJ
          (/)
          CO
          <
                    m/e = 109
                        <
                        UJ
                        O)
                                                  O
                                                  CM
                                                  I
                                                       m/e = 151
                                                            en
                                                     *T
                                                           rf
                                                                    TENTATIVE ASSIGNMENT:

                                                                            CH3
                                                                        CH3-C-CH3
                                                                            I
                                                                           OH
                                                                             o
                                                                             "-
                                                                              it
                                                                                  O)
                                                                                  O)
                                                                                  -»-
                                                                                  I
                                                                                  o
!.),„	,,.!,.
                                 -.,,
                                          I
                                              '1
   60   73   03   90    100  110  120   130  1-W  150   163  170  180   190  200  210  220
             E
 FIGURE 57.  ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.
            CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK D.
            (Refer to Chromatogram  in Figure 43)

-------
oo
       R.


       8.
                        o
                        o
                        CM

                        +
                        I
(D-

  60



 FIGURE
70

MX

58.
                    so
10J
TENTATIVE ASSIGNMENT:

CH3 CH3
-"•""" . 'C<^o
8 cH3 y
7 ^H2
«J +
E m/e = 111
O
I
O in
5 •£
* c*














I 1 Jl
^^^^^^^^^"1 " * • •«•-§' '• •"""
1lfl 12B 133
in «
CH3 CH3
1 VJ
Nj^ Ch

\/
II
O





5
)

to











O «o «~
0 - ii
n „
+ -S
X -5! i
S E
+
^
' T™.
: ^
u
0
o x -£
CM S fc
X

"L.


1 1 1
1 . 11 il il 1 i 1 i
110 195 16G 170 If
.
if)
X
CM
U
s
IL l lllinVl I 1 "mil 	
J«3 J90 209 213
CO
CM
CM
II
4J

^
in
X
CO
O
5

22CJ 230
                   ANALYSIS OF a-PINENE AEROSOL, NEUTRAL FRACTION.

                   CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK H.


                   (Refer to Chromatogram in Figure 43)

-------
 one aerosol was desirable to determine if any nitrogen bases were
 indeed formed under atmospheric simulation.

 Analysis of the a.-pinene basic fraction by GC-MS  did not  suggest  the
 presence of any nitrogen bases, nor-indeed,  the presence  of  any nitrogen
 containing products.   The spectra obtained are in fact typical  of those
 observed for the a-pinene neutral fraction.   That is,  the "basic"
 fraction appears to be composed primarily of neutral reaction products.
 This can be explained by reference to the fractionation procedure.

 As  described previously,  a methylene  chloride solution of the water-
 insoluble aerosol products was extracted with dilute aqueous  sodium
 hydroxide and with dilute hydrochloric acid.   The basic fraction  was
 defined  as that material  extracted into  the  dilute acid.   During  the
 acid extraction,  a small  quantity of  (insoluble)  neutral  material was
 apparently entrained  or  emulsified in the dilute  acid  wash,  and was
 thereby  carried along with the "basic"  fraction.   In future work, this
 will be  avoided by carefully backwashing  the  dilute  acid  phase  with
 solvent  before the aqueous  phase  is made  alkaline  and  extracted with
 solvent.   An analogous backwashing procedure  will  be used in  the  isola-
 tion of  the  acid  fraction.  In summary,  it appears that if any  genu-
 inely basic  products  are  formed,  their concentration is vanishingly
 small in the material  analyzed.

 Several  representative mass spectra obtained  for  this  fraction  are
 presented  in order  to  illustrate  the  "neutral" character  of the com-
 ponents.   The  reconstructed gas chromatogram  is shown  in Figure 59.
Consider  the mass  spectrum shown  in Figure 60, corresponding to chroma-
 tographic  Peak A.  The spectrum is extremely similar to that observed
 for  the neutral product considered earlier in Figure 52.  A fragment
consistent with the protonated parent ion is  observed at m/e  155  and
peaks consistent with loss of carbon monoxide and  water are observed
at m/e  127 and 137 respectively.  Key fragments consistent with
                                     119

-------
N3
O
                 SPECTHJt
                     3*
                                      aw
                                                    «a
                                                         130
                                                                       iw
                                                                             9» sia
                                                                                        s*
                   FIGURE 59.  ANALYSIS OF a-PINENE AEROSOL, BASIC FRACTION.
                              RECONSTRUCTED GAS CHROMATOGRAM FOR METHANE CI-MS.

-------
   8
                                                           CO   CO
~ —
cn_
g_

a.
If"
bED
LO-
t-W
3
IB"
0
C\J_
ED

°-

X
X
o
CO
o








I




*—
II
~E




X
1
X










i^M

in
00
n
-SJ
E











°,x A











1






CO CO "•" N, ' NT
IX I
CJ O •*•
V
^ I
0 X C,H3
( 1
" ^^"^n
o <^
§ L
n ^X. ^.CHO
_» r»-
E o "-
0 " TENTATIVE ASSIGNMENT
Ql
1 * I
ii + "
i +
^ s i

iff<>pHllllii|iiil[iiti|l»Hiiii|iMi|iiiniiii|iiiiiiiii[inifi'"f""l"l-j— -i----!" i- •
       53  63   78   80   93    103  113  120  133  1«  ISO  360  \TO  180  133
FIGURE 60.  ANALYSIS OF a-PINENE AEROSOL, BASIC FRACTION.
           CHEMICAL 10NIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK A.

           (Refer to Chromatogram in Figure 59}

-------
transantiular cleavage are observed at m/e  71, 85, and 99.  Thus, the
product is tentatively assigned the structure shown, identical to that
proposed  in Figure 52 .

Consider  the mass spectra shown in Figure  61; they correspond to chroma-
tographic Peak B.  The upper spectrum was obtained using methane as
the chemical-ionization reagent gas.  The  lower spectrum was obtained
using helium as the reagent gas; hence it  is much like an electron-
impact spectrum.  The methane spectrum is consistent with a molecular
weight of 138, although the satelite peak at (M + 29) is very weak.
Using the Battelle peak-matching program, the helium spectrum was com-
pared with a computerized file of 11,000 electron-impact spectra.  The
sample spectrum corresponded very closely with that of terpinoline.
Consistent with this finding and with the apparent molecular weight
determined from the methane spectrum, the product is tentatively
assigned as terpinoline.

Finally, consider the spectrum shown in Figure 62 corresponding to
chromatographic Peak C.  The structure shown in the figure is especially
tentative.  The peak assigned as the protonated parent ion is consistent
with a molecular weight of 182.  Fragments consistent with loss of H20
and loss of CH2=C=0 plus H20 are observed.  Peaks diagnostic of an
intact cyclobutane ring are absent.  Consistent with these data, a
speculative structure is shown.

The above described spectra are typical of those observed for  the
"basic" fraction.   The results suggest that this fraction consists
principally of a small quantity of neutral product  carried over during
an incomplete fractionation step.
                                    122

-------
- SPECTRUM 1
SL

?•>
iSj
I
Si.i_
•


E_









..lil.i





j|
i. ILL










O)
«







1,
ill..
*~
II
1
X




. - -





o>
CM
+
~l

TENTATIVE
ASSIGNMENT
(see text)
A
O
CH X

TERPINOLENE


SO «>fl 7ff 10 i«P 13P
                   HX f
FIGURE 61.  ANALYSIS OF a-PINENE AEROSOL, BASIC FRACTION.
            CHEMICAL IONIZATION MASS SPECTRA FOR CHROMATOGRAPHIC PEAK B.
            SPECTRUM 1 BY METHANE CI-MS, SPECTRUM 2 BY HELIUM CI-MS.
            (Refer to Chromatogram in Figure 59)
                                     123

-------
IX)
          8«
          00-
         JED
        U.D
        ID


—
~
—
—
-
—















............111,..






















..1.


1














1












T"







1









1











1






















































, iji

o
CM
I
0
II
11 fM
o> X
> O
1
I
^ 	 y
^ —






















1










III,,























1

SPECULATIVE ASSIGNMENT
O O
II 1!
CHo-C-CH = CH-CH-C -CHO
J |
CH
/ \
^LJ OLJ
CM3 CM3
9,
X
+ +





in
to
^~
ii
-Si
E

ll.

in
09
n
o>
1
II

            50  60   76   8C   90   100  1J0  128  13(3  Hfl  15(3  163  17G  190  190  206  218
                 M/ E
           FIGURE 62.  ANALYSIS OF a-PIIMENE AEROSOL, BASIC FRACTION.
                     CHEMICAL IONIZATION MASS SPECTRUM (METHANE) FOR CHROMATOGRAPHIC PEAK C.
                     (Refer to Chromatogram in Figure 59)

-------
                                SECTION V

                               REFERENCES

(1)   Schuctzlc, D. ,  Crittenden,  A.  L., Cronn,  D.  R.,  and Charlson,  R.  J.,
      "Molecular Composition of Secondary Aerosol  and  Its Origin",  paper
      presented at the 172nd National A.C.S.  Meeting;  Chicago,  Illinois;
      August 1973.

(2)   Air Pollution Control District, Los Angeles  County, "Effectiveness
      of Organic Solvents in Photochemical Smog Formation",  Solvent
      Project, Final  Report, 1966.

(3)   Altshuller, A.  P., Bufalini, J. J.t Environ. Sci.  Technol.,  5, 39
      (1971).

(4)   Altshuller, A.  P., Bufalini, J. J., Photochem.  Photobiol..4.  97
      (1965).

(5)   Altshuller, A.  P., Cohen, R. R.,  presented at American Chemical
      Society Meeting, Division of Water, Air,  and Waste Chemistry,
      New York, New York, Sept. 8-13, 1963.

(6)   Altshuller, A.  P., Klosterman, D. L.,  Leach, P.  W., Hindawi,  I. J.,
      Sigsby, J. E.,  Jr., Int.  J. Air Water  Pollut.. 10.  81 (1966).

(7)   Altshuller, A.  P., Kopczynski, S. L.,  Lonneman,  W. A., Becker, T. L.,
      Wilson, D. L.,  Environ.  Sci. Technol.. 2,  696 (1968).

(8)   Doyle, G. J. and Renzette,  N.  A., Air  Pollut. Contr. Assoc.  8,
      23 (1968).

(9)   Endow, N., Doyle, G.  S.,  Jones J. L. ,  ibid,  13,  141 (1963).

(10)   Goetz, A., Pueschel,  R.  F., Atmos. Environ.T 1,  287 (1967).

(11)   Groblicki, P. J., Nebel,  G. J., presented at General Motors  Research
      Laboratories Symposium on Chemical Reactions in  Urban Atmospheres,
      CMC,  Warren, Mich., Oct,  6  and 7, 1969.

(12)   Harkins, J., Nicksic, S.  W., J. Air Pollut.  Contr. Assoc.. 15, 218
      (1965a).

(13)   Harkins, J., Nicksic, S.  W. , presented at American Chemical  Society
      Meeting, Division of  Petroleum Chemistry, Detroit, Michigan,  April
      4-9,  1965b.

(14)   Hayakawa, I., J. Air  Pollut. Contr. Assoc.,  14,  339 (1964).

(15)   Johnston, H. S., der  Jain,  K., Science,131,  1523 (1960).


                                    125

-------
(16)  Orr, C., Jr., Hurd, F. K.,  Corbett, W. J.,  J. Colloid Sci., 13.
      472 (1968).

(17)  Prager, M. J., Stephens, E. R. , Scott, W.  E., Ind.  Eng. Chem..
      52, 521 (I960).

(18)  Renzetti, N. A., Doyle, G.  J., Int. J. Air  Pollut.  2, 327 (1960).

(19)  Ripperton, L. A., Decker, C, E.,  Page, W.  W. , presented at
      American Chemical Society Meeting, Division of Water, Air, and
      Waste Chemistry, Atlantic City, N. J., Sept. 13-17, 1962.

(20)  Schuck, E. A., Ford, H. W., and Stephens,  E. R.,  "Air Pollution
      Effects of Irradiated Automobile  Exhaust as Related to Fuel
      Composition", Air Pollution Foundation Report 26, October, 1958.

(21)  Stephens, Edgar R., Price,  Monty A., Atmos. Environ.  3, 573 (1969).
      Stern, A. C., ed.,  "Air Pollution", Vols I-III,  2nd Ed., Academic
      Press, New York, New York,  1968.

(22)  Stevenson, H.J.R.,  Sanderson, D.  E., Altshuller,  A. P., Int.  J. Air
      Water Pollut., 9, 367 (1965).

(23)  Titnmons, R. G. , presented at General Motors Research Laboratories
      Symposium on Chemical Reactions in Urban Atmospheres, GMC, Warren,
      Mich., Oct. 6 and 7, 1969.

(24)  Wayne, L.  G., and Romanovsky, J,  C., "Rates of Reaction of the Oxides
      of Nitrogen in Photo-oxidation of Diluted Automobile Exhaust  Gases",
      Chemical Reactions  in the Lower and Upper Atmosphere, Interscience,
      New York,  1961, p.  71.

(25)  Wilson, Wm. E., Jr., Merryman, E. L.,  Levy, A.,  J.  Air Pollut.  Cont.
      Assoc.. 21, 3 (1971).

(26)  Wilson, Wm. E., Jr., Miller, D.  F., Trayser, D.  F., and Levy,  A.,
      "A Study of Motor Fuel Composition Effects  on Aerosol Formation.
      Part III Visibility Reduction from Automobile Exhaust".  Battelle,
      Columbus Laboratories Report to the American Petroleum Institute
      Committee for Air and Water Conservation (February 1972).

(27)  Wilson, Wm. E. , Jr., Levy, A., "A  Study of Sulfur  Dioxide in
      Photochemical Smog", American Petroleum Institute,  Project S-ll,
      Battelle,  Columb'us  Laboratories,  Columbus,  Ohio,  1968.
                                     126

-------
(28)   Miller,  D.  FCJ  Levy,  A.,  and  Wilson,  Wm.  E.,  Jr.,  "A  Study of Motor
      Fuel Composition Effects  on Aerosol  Formation.   Part  II, Aerosol
      Reactivity  Study of Hydrocarbons", Battelle,  Columbus Laboratories
      Report to the American Petroleum Institute  Committee  for Air and
      Water Conservation (February  21,  1972).

(29)   Levy, A.  and Miller,  S. E., "Final Technical  Report on  the Role of
      Solvents  in Photochemical Smog Formation",  to National  Paint, Varnish,
      and Lacquer Association,  Battelle, Columbus Laboratories,Columbus,
      Ohio, 1970.

(30)   Landstrom,  D. K. and  Kohler,  D.,  "Electron  Microprobe Analysis  of
      Atmospheric Aerosols", to National Air Pollution Control Administration
      (CPA-22-69-33)  (1969).

(31)   Henry, W. M.  and Blosser, E.  R.,  "A  Study of  the Nature of  the
      Chemical  Characteristics  of Particulate Collected from  Ambient  Air",
      to National Air Pollution Control Administration (CPA-22-69-153)
      (1970).

(32)   Schuetzle D. ,  Cuttenden,  A. L., and  Charlson, R. J.,  "Application of
      Computer Controlled High-Resolution  Map Spectrometry  to the
      Analysis of Air Pollutants",  J. Air  Poll. Control Assoc.,  23 (1970).

(33)   Schwartz, W.  E., Jones, P. W., Riggle, C. J., and Killer,  D. F.,
      "The Organic Composition of Model Aerosols",  invited  paper presented
      at the Eastern  Analytical Symposium, New York, New York, November
      1973.

(34)   Wilson, Wm. E., Jr., Schwartz, W. E., and Kinzer, G.  W., "Haze
      Formation -- Its Nature and Origin", a Battelle-Columbus Report to
      the Environmental Protection Agency  (CPA 70-Neg. 172) and The
      Coordinating Research Council  (CAPA  6-68) January 1972.  Reference
      is made to results of research conducted during the contract period
      June  1972 - June 1973 and presented   in the first annual report.

(35)   Wilson, W.  E.,  Jr., Miller, D. F., Levy, A.,  and Stone, R.  K., "The
      Effect of Fuel Composition on Atmospheric Aerosol Due to Auto
      Exhaust", J. Air Poll. Control Assoc..23. 949 (November 1973).

(36)   Lonneman, W. A., Kopczynski,   S. L.,   Darley,  P. E., and
      Sutterfield, F. D.,  "Hydrocarbon Composition of Urban Air
      Pollution", Environ. Sci and  Techno^., 8. 229  (March  1974).

(37)   Kopczynski, S.  L., Lonneman,  W. A.,   Sutterfield, F.  D., and Darley,
      P. E., "Photochemistry of Atmospheric Samples in Los Angeles",
      Environ. Sci.  and Technol_.,6,  342 (April 1972).
                                       127

-------
(38)  Rasmussen, R. A., J. Air Poll. Control Assoc..  22,  537 (1972).

(39)  Rasmussen, R. A., Proc.  Nat. Acad.  Sci.,53, 215 (1965).

(40)  Fieser, L. F., J. Amer Chem. Soc.,  76 1945 (1954).

(41)  Fieser, L. P., J. Amer Chem Soc.,  75 4386 (1953).

(42)  Seeback, D.,  Jones,  N. R.,  and Corey, E.  J.,  J.  Org.  Chem.,33,
      300 (1968).

(43)  Handbook of Silylation,  Pierce Chemical Company, P.  0. Box 117,
      Rockford, Illinois  61105,  1972, pp. 19-20.

(44)  Corey, E. J., J.  Org. Chem.,32, 4160 (1967).

(45)  Brown, R., Adv.  Org. Chem., 3 89 (1963).

(46)  Miller, D. F., Schwartz, W.  E., Jones. P. W., Joseph, D.  W. ,
      Spicer, C. W. , Riggle, C.  J., and  Levy, A., "Haze Formation:   Its
      Nature and Origin",  second  annual  report of the U.  S. Environmental
      Protection Agency and the  Coordinating Research Council,  submitted
      by Battelle,  Columbus Laboratories, June 1973.

(47)  Mayer, R., Bochow, K., and  Zieger,  W., Z. Chem.4,  348 (1964).

(48)  Jones, P. W., and Adelman,  A. H.,  "A Novel Preparation of Cyclic
      Sulfites throug1     'tosulfoxidation of Alkenes", Tetraheldron, 30.
      2053,   (1974).

(49)  Cockcroft, R. D., Waali, E.  E., and Rhodes, S.  J.  Tetrahed.  Lett..
      41, 3539 (197(n.

(50)  Simonsen, J.  L.,  "The Terpenes", Vols. I and II, Cambridge University
      Press, London, c. 1953.

(51)  Templeton, W., "An Introduction, to the Chemistry of Terpenoids  and
      Steroids", Butterworth and  Company, Ltd., London, c.  1969.

(52)  Simonsen, J.  L.,  op. cit.,  Vol II,  p. 140.

(53)  Spencer, C. C.,  Weaver,  W.  I., Oberright, E.  A., Sykes, H. J.,
      Burney, A. L., and Elder, A. L., J. Qrg.  Ghem.,610,  5 (1940).

(54)  Simonsen, J.  L.,  op. cit.  Vol. II,  pp. 136-139.

(55)  Templeton, W., op. cit.  p.  62.
                                    128

-------
(56)   Banthorpe, D. V., and Whittaker, D., q. Rev. Chem Soc.,20,373
      (1966).

(57)   Ternpleton, W. op. cit. pp 67-68.

(58)   Simonsen, J.  L., op. cit. Vol I, p. 300 and p. 166.

(59)   Dupont,  G., and Zacharowitz, W., and R. Deilou, Compt. Rend., 198,
      1699 (1934).

(60)   Henderson, G. G. , and Sutherland, M. M. , J. Chem. Soc., 101 2288
      (1912).

(61)   Gay, B.  W. , and  Bufalini,  "Hydrogen Peroxide in the Urban
      Atmosphere",  paper presented at the 161st American Chemical Society
      National Meeting, Los Angeles, March 1971.

(62)   Simonsen, J.  L. , op. cit., Vol II p. 141.
                                       129

-------
TECHNICAL REPORT DATA
fl'leasc read Instructions on llic /ri'mr hi'fori' completing}
1 . REPORT NO.
EPA-650/3-74-011
4. TITLE AND SUBTITLE
Chemical Characterization
2.
of Model Aerosols
7. AUTHOFKS)
Warren E. Schwartz, Peter W. Jones,
Charles J. Riggle, and David F. Miller
9. PERFORMING ORG '\NIZATION NAME AND ADDRESS
Battelle
' Columbus Laboratories
505 King Avenue
Columbus, Ohio 43201
12. SPONSORING AGENCY NAME AMP ADDRESS
Chemistry (, Physics Laboratory
National Environmental Research Center
Environmental Protection Agency
Research Triangle Park, N. C. 27711
IS. SUPPLEMENTARY NOTES
3. RECIPIENT'S ACCESSIOWNO.
5. REPORT DATE
August 1974
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
1O. PROGRAM ELEMENT NO. 	 	 ~
1AA008
1 1. CONTRACT/GRANT NO. 	
801174
13. TYPE OF REPORT AND PERIOD COVE RET
Final: 6/72-6/74
14. SPONSORING AGENCY CODE


16. ABSTRACT 	 ' 	
Model aerosols were generated fro-n individual hydrocarbon precursors and nitrogen
oxides under simulated atmospheric conditions in a 17.3 m environmental chamber.
Hydrocarbon precursors employed were toluene, alpha-pinene , and cyclohexene.
Aerosols were collected on glass-fiber filters and organic matter was obtained
by solvent extraction. Organic reaction products were fractionated into acid
neutral, and basic components, and were analyzed in detail by techniques
including gas chromatography and gas chromatography combined with mass spectro-
metry. The study also included evaluation of techniques for selective
derivatization of aerosol products.
A variety of polyf unct ional reaction products were identified, including alcohols
aldehydes, ketones, carboxylic acids, and phenols. Tentative identification of
nitrogen containing products was also accomplished, including nitrate esters
and aromatic nitro compounds.
This report was submitted in fulfillment of Grant Number R-801174 by the Columbus
Laboratories of Battelle Memorial Institute under the sponsorship of the
Environmental Protection Agency. Work was completed as of April 30, 1974.
17.
a. DESCRIPTORS
Aerosols
Smog
Air Pollution
Chemical Analysis
Gas Chromatography
Mass Spectrometry
Environment Simulation
Environment Simulators
13, DISTRI BUTI ON STATEMENT
Release Unlimited
KEY WORDS AND DOCUMENT ANALYSIS
b. OENTIFI E RS/OPEN ENDED TERMS
Model Aerosols
Organic Aerosols
Aerosol Chemistry
Atmospheric Chemistry
Organic Chemical
Analysis
Smog Chamber
19. SECURITY CLASS (This Report/
Unclassified
20. SECURITY CLASS (Tliis pan?)

c. COSATI I iL-IU/<;roup
04 -A
07-C
07-E
21 • NO. OF PAGES 	
129
22. PRICE 	 	 	
EPA Form 2220-1 (9-73)
                                                          130

-------