United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA-600/9-80-003
January 1980
<>EPA
Research and Development
Proceedings of the
Conference on Methyl
Chloroform and Other
Halocarbon Pollutants
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8 "Special" Reports
9. Miscellaneous Reports
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/9-80-003
January 1980
PROCEEDINGS OF THE
CONFERENCE ON METHYL CHLOROFORM
AND OTHER HALOCARBON POLLUTANTS
Washington, D.C.
February 27-28, 1979
Coordinated by
Technical Services
Northrop Services, Inc.
Research Triangle Park, North Carolina 27709
Edited by
Joseph J. Bufalini
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
uuy
ENVIRONMENTAL
OFFICE OF
SCIENCES RESEARCH LABORATORY
RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Mention of trade names or commercial products does not constitute endorse-
ment or recommendation for use.
11
-------
PREFACE
The Clean Air Act gives the U.S. Environmental Protection Agency (EPA)
authority to control volatile organic compounds in order to protect air quality,
Included is control of pollutants that are sources of tropospheric ozone (O.,)
(e.g., high-reactivity organics that produce photochemical O ) and pollutants
such as halocarbons that can, by low reactivity, be transported into the
stratosphere and affect the O layer.
The purpose of the Conference on Methyl Chloroform and Other Halocarbon
Pollutants was to establish the distribution and persistence of methyl chloro-
form (1,1,1-trichloroethane, CH CCl , MCF) and other related halogenated
organic compounds. The emphasis was largely on the effects of halocarbons on
O depletion. MCF was chosen as the principal compound of interest because
its production rate is rapidly increasing and there is some doubt as to its
tropospheric lifetime. The uncertainty with regard to the lifetime of MCF is
largely a result of uncertainty on the amount of hydroxyl radical (OH) present
in the troposphere as well as the ambient concentration of MCF. Of course,
these are related, since low levels of MCF necessitate high OH concentrations.
Any model which predicts a change in stratospheric O., must depend upon
product identification, reaction rates, and a mechanism. Therefore, any find-
ings or predictions reported in the following chapters are based on current
knowledge of the reaction scheme. The predictions may change as better data
become available. As a reminder, remeasurement of the HO -NO reaction now
leads us to calculate stratospheric 0 enhancement (instead of depletion) by
supersonic transports. However, to the smog chamber modelers, this reaction
now leads to an overprediction of O , suggesting that there is still something
wrong with the proposed NO mechanism for describing photochemical smog.
X
111
-------
This document reports the findings of experts from research institutions,
manufacturers, and government agencies. Hopefully, EPA can use this informa-
tion to better direct its research efforts and arrive at a reasonable control
strategy for MCF and other halogenated organic compounds.
J. J. Bufalini
Environmental Sciences Research
Laboratory
IV
-------
ABSTRACT
Presentations at the Conference on Methyl Chloroform and Other Halocarbon
Pollutants (Washington, D.C., February 27-28, 1979) are documented. Included
among the authors are research scientists, industry representatives, and
regulatory officials.
The 16 papers fall into 2 basic groups. The first 10 papers present
results of research in atmospheric chemistry as related to the question of
stratospheric ozone depletion by halocarbons. Drawing upon atmospheric
measurements and model calculations, the authors give estimates of emission
levels, current atmospheric burdens, tropospheric lifetimes, the importance of
sinks, effects on stratospheric ozone, and related questions.
The final 6 papers take the perspective of involvement in, or concern
with, regulatory decisionmaking. The authors consider various options, rec-
ommendations, and plans for halocarbon control in light of available scientif-
ic data.
Finally, the Panel Discussion which concluded the Conference is presented
in verbatim transcript form. Focusing on the current status of atmospheric
measurements, the participants discuss problems in obtaining accurate halocar-
bon data, and discrepancies between and within the results of individual in-
vestigators .
v
-------
CONTENTS
PREFACE ................................
ABSTRACT ................................ V
FIGURES ................................ xi
TABLES ................................. xiii
ABBREVIATIONS AND SYMBOLS ....................... XV
ACKNOWLEDGMENTS ............................ xix
EDITOR'S NOTE ON DISCUSSION TRANSCRIPTS ................ xx
COMMENTS ON THE LIFETIMES OF ORGANIC MOLECULES IN AIR ......... 1-1
A. P. Altshuller
Introduction ......................... 1-1
The Troposphere and Organic Molecules ............. 1-2
References .......................... 1-8
Discussion .......................... 1-9
STATUS OF THE STRATOSPHERIC OZONE DEPLETION ISSUE, INCLUDING
COMMENTS ON METHYL CHLOROFORM ..................... 2-1
F. Sherwood Rowland
Presentation ......................... 2-1
Discussion .......................... 2-13
ENVIRONMENTAL FATE OF METHYL CHLOROFORM ................ 3-1
W. Brock Neely and G. Agin
Introduction ......................... 3-1
Residence Time of Methyl Chloroform in
the Troposphere ....................... 3-3
Transfer of Chlorine from Methyl Chloroform
to the Stratosphere ..................... 3-13
Impact of Chlorine Atoms on
Stratospheric Ozone ....... , ............. 3-14
Further Questions and Recommendations ............. 3-15
References .......................... 3-17
Discussion .......................... 3-19
vii
-------
HALOGENATED TRACE CONSTITUENTS IN THE GLOBAL ATMOSPHERE 4-1
Hanwant B. Singh, Louis J. Salas, and Hisao Shigeishi
Distributions, Sources, and Sinks of
Halogenated Trace Constituents 4-1
Current Status of Halogenated Trace Constituents 4-11
References 4-12
Discussion 4-13
STRATOSPHERIC IMPACT RESEARCH AND ASSESSMENT 5-1
Alphonse F. Forziati
Introduction 5-1
EPA Research 5-2
Biological and Climatic Effects Research Program 5-5
Stratospheric Impact Research and Assessment Program 5-9
Effects of UV-B Radiation 5-11
A TWO-DIMENSIONAL PHOTOCHEMICAL MODEL TO ESTIMATE
STRATOSPHERIC OZONE DEPLETION 6-1
Paul J. Crutzen
Introduction 6-1
Results and Discussion 6-1
Areas for Improvement 6-7
References 6-10
Discussion 6-10
A REVIEW OF TECHNICAL PROGRAMS OF THE MANUFACTURING
CHEMISTS ASSOCIATION RELATED TO STRATOSPHERIC
CHEMISTRY OF CHLORINE 7-1
Frank A. Bower
Introduction 7-1
Atmospheric Lifetime ..... 7-1
Stratospheric Measurement 7-3
Reaction Rates 7-5
Ozone Trend Analysis 7-5
Summary . 7-7
References . 7-7
Discussion 7-8
MEASUREMENTS OF ATMOSPHERIC METHYL CHLOROFORM BY
WASHINGTON STATE UNIVERSITY 8-1
Dagmar R. Cronn
Introduction 8-1
Vertical Distribution 8-1
Time Trends 8-3
Latitudinal Distribution 8-7
Concluding Remarks 8-7
References 8-9
Discussion 8-9
viii
-------
TROPOSPHERIC HYDROXYL RADICAL CONCENTRATIONS
AND METHYL CHLOROFORM REMOVAL .' 9-1
Malcolm J. Campbell
Introduction 9-1
Method 9-1
Results and Discussion 9-3
Comments on Alternative Methods 9-7
References 9-8
Discussion 9-8
IMPACT OF BROMINATED COMPOUNDS ON THE STRATOSPHERE 10-1
Luisa T. Molina and Mario J. Molina
Introduction 10-1
Bromocarbon Chemistry 10-2
Ultraviolet Absorption Spectra 10-2
Atmospheric Photodissociation Rates 10-5
References 10-6
BRIDGE BETWEEN THE SCIENCE AND THE REGULATORY NEEDS 11-1
Herbert L. Wiser
Introduction 11-1
Regulatory Concerns 11-1
Industrial Concerns 11-4
Concluding Remarks 11-5
Discussion 11-6
AN ASSESSMENT OF TRICHLOROETHYLENE, METHYL CHLOROFORM,
AND PERCHLOROETHYLENE 12-1
Thomas Lapp
Introduction 12-1
Monitoring Data 12-3
Environmental Impacts 12-5
Health Impacts 12-5
Exposure Levels 12-6
Recommendations for Emission Reduction and Control 12-8
Discussion 12-11
METHYL CHLOROFORM AND ITS STABILIZERS 13-1
Stanley C. Mazaleski
Introduction 13-1
Background- and Overvie - 13-2
In-Vitro Studies 13-4
In-Vivo Studies 13-5
Conclusions 13-9
References 13-10
IX
-------
METHYL CHLOROFORM IN PERSPECTIVE 14-1
Louis Schlossberg
Introduction 14-1
Tropospheric Photochemical Oxidation vs.
Stratospheric Ozone Depletion 14-3
RACT, Rule 66, and Methyl Chloroform 14-5
Questions and Misconceptions 14-8
Capacity and Predicted Use of Methyl Chloroform 14-10
References 14-12
Discussion 14-12
REGULATORY ISSUES INVOLVED IN HALOCARBON CONTROL UNDER
THE CLEAN AIR ACT 15-1
Robert Kellam
Introduction 15-1
The Clean Air Act 15-1
Tropospheric and Stratospheric Ozone 15-1
Regulation of Methyl Chloroform 15-3
Discussion 15-4
INTERAGENCY WORK GROUP ACTIVITIES ON NONAEROSOL
USES OF CHLOROFLUOROCARBONS 16-1
Ferial S. Bishop
Introduction 16-1
The Chlorofluorocarbon Problem 16-1
Phase 1 16-2
Phase II 16-4
Additional Research 16-8
Final Comments 16-9
Discussion 16-10
PANEL DISCUSSION 17-1
APPENDIX A: DISCUSSION PARTICIPANTS A-l
X
-------
FIGURES
Number Page
3-1 Statistical model used to fit monitoring data collected
4-1
4-2
5-1
5-2
5-3
5-4
5-5
5-6
on MCF in the troposphere
Global distribution of atmospheric constituents
Atmospheric growth of FC-12, FC-11, CC1 , and CH CC1 . . . .
Information and decision chart ..............
Major data needs and uncertainties ,
EAGER budget by agency for FY-77
USDA- developed instrument for measurement of radiation
incident on plants
Effects of UV-B on cucumber plants
Correlation in humans between areas of UV-B exposure
and sites of skin cancer
. . 3-4
. . 4-2
. . 4-4
5-3
. . 5-4
. . 5-7
. . 5-12
5-13
. . 5-15
5-7 Trends in mortality from melanoma of the skin
in the U.S., 1950-1970 5-16
6-1 Model-derived total 0 field for 1970 plotted as
a function of month and latitude 6-2
6-2 Observed total O., field plotted as a function
of month and latitude 6-3
6-3 Model calculations of depletion of global O 6-5
6-4 Month and latitude distribution of average total
O loss between 1970 and 2000 6-6
6-5 Latitude and altitude distribution of cumulative
0 loss after 10 yr of model integration 6-8
XI
-------
Number Paqe
6-6 Model-derived distribution of O for December 1970 6-9
8-1 MCF mixing ratio distribution as a function of
tropopause height, March 1976, 47° N latitude 8-2
8-2 MCF mixing ratio distribution as a function of
tropopause height, April 1977, 37° N latitude 8-3
8-3 MCF mixing ratio distribution as a function of
tropopause height, July 1977, 9° N latitude 8-4
8-4 Ground-level time trend measurements of MCF,
June 1977 through January 1979, 47° N latitude 8-5
8-5 Observed NH and SH mixing ratios for MCF, 1972
through 1978 8-6
8-6 Latitudinal gradient of MCF corrected to November 1978 8-8
9-1 Contours of calculated MCF consumption in a meridional plane. . 9-4
9-2 The calculated altitudinal variation of the MCF consumption
rate 9-5
9-3 The calculated latitudinal variation of the MCF consumption
rate 9-6
10-1 Absorption cross sections for several brominated
hydrocarbons 10-4
-------
TABLES
Number Paqe
1-1 Estimated Lifetimes of Methane and Halogenated
Methane Derivatives 1-2
1-2 Average Global Concentration of Halocarbons
from Dispersive Losses 1-4
1-3 Number of Days for 1 Percent Consumption of
Organic Compounds by OH at 40° N Latitude
in January and in July 1-6
3-1 Worldwide Global Emission of Methyl Chloroform 3-5
3-2 Parameters for the Statistical Model 3-6
3-3 Monitoring Data on Methyl Chloroform in the
Northern Troposphere 3-7
3-4 Summary of Statistical Determination of the
Dissipating Rate Constants 3-8
3-5 Average Tropospheric Hydroxyl Radical Concentrations 3-9
3-6 Biomolecular Rate Constant for Hydroxyl Attack
on Methyl Chloroform 3-10
3-7 Stratospheric Ozone Destruction Mechanisms 3-15
4-1 Tropospheric Concentrations of Important
Atmospheric Trace Constituents 4-5
4-2 Tropospheric Chlorine, Bromine, Iodine, and
Fluorine Organic B^aets 4-6
4-3 Oceanic Sinks 4-8
4-4 Global Average Residence Time of Methyl Chloroform 4-10
Xlll
-------
Number Paqe
5-1 BACER Expenditures by Category for FY-76 and FY-77 5-6
12-1 Recent U.S. Production Quantities of Trichloroethylene,
Methyl Chloroform, and Perchloroethylene 12-2
13-1 Dow Study 13-6
-------
LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
CAA Clean Air Act
CARCE Committee on Alternatives for the Reduction of Chlorofluoromethanes
CEQ Council on Environmental Quality
CFC chlorofluorocarbon
CFM chlorofluoromethane
CISC Committee on Impacts of Stratospheric Change
CPSC Consumer Product Safety Commission
DOC Department of Commerce
DOE Department of Energy
DOT Department of Transportation
DU Dobson Unit
EPA Environmental Protection Agency
FAA Federal Aviation Administration
FC-11 fluorocarbon-11 (trichlorofluoromethane, CC1-F)
FC-12 fluorocarbon-12 (dichlorodifluoromethane, CC1 F )
FC-14 fluorocarbon-14 (carbon tetrafluoride, CF )
FC-22 fluorocarbon-22 (chlorodifluoromethane, CHC1F )
FC-113 fluorocarbon-113 (CC1 FCC1F )
FC-114 fluorocarbon-114 (CC1F CC1F )
FC-133 - fluorocarbon-133 (CF CH Cl)
FC-142A fluorocarbon-142A (CH CC1F )
FC-152A fluorocarbon-152A (CH CHF )
FDA Food and Drug Administration
FY Fiscal Year
GC gas chromatography
xv
-------
ICAS
ICSOP
IMOS
IR
ITCZ
JPL
MCA
MCF
MTD
NAAQS
NAS
NASA
NBS
NCAR
NCHS
NCI
NH
NIOSH
NOAA
NSF
OAQPS
0PM
ORD
OSHA
OTS
PCE
RACT
SH
SIP
SIRA
STP
TCE
TSCA
USDA
UV
Interdepartmental Committee on Atmospheric Sciences
Interagency Committee on Stratospheric Ozone Protection
Inadvertent Modification of the Stratosphere
infrared
intertropical convergence zone
Jet Propulsion Laboratory
Manufacturing Chemists Association
methyl chloroform (1,1,1-trichloroethane, CH CCl,)
maximum tolerated dose
National Ambient Air Quality Standards
National Academy of Sciences
National Aeronautics and Space Administration
National Bureau of Standards
National Center for Atmospheric Research
National Center for Health Statistics
National Cancer Institute
Northern Hemisphere
National Institute of Occupational Safety and Health
National Oceanic and Atmospheric Administration
National Science Foundation
Office of Air Quality Planning and Standards
Office of Program Management
Office of Research and Development
Occupational Safety and Health Administration
Office of Toxic Substances
perchloroethylene (C Cl )
reasonably available control technology
Southern Hemisphere
State Implementation Plan
Stratospheric Impact Research and Assessment
standard temperature and pressure
trichloroethylene (C_HC1 )
Toxic Substances Control Act
United States Department of Agriculture
ultraviolet
xvi
-------
UV-A
UV-B
v
VOC
wsu
ultraviolet-A
ultraviolet-B
volume
volatile organic chemical
Washington State University
SYMBOLS
BrO
x
CBrF.
CC1F CC1F
CC12°
CC13F
cci4
CF CH Cl
CF4
CHC1F
CHC13
CHF
CH BrCH Br
CH3CC1F2
CH CHF
CH3COOH
CH3C1
CH OH
CH4
CO
oxides of bromine
bromotrifluoromethane
fluorocarbon-114 (FC-114)
fluorocarbon-113 (FC-113)
dichlorodifluoromethane (fluorocarbon-12, FC-12)
phosgene
trichlorofluoromethane (fluorocarbon-11, FC-11)
carbon tetrachloride
fluorocarbon-133 (FC-133)
carbon tetrafluoride (fluorocarbon-14, FC-14)
chlorodifluoromethane (fluorocarbon-22, FC-22)
trichloromethane (chloroform)
trifluoromethane
ethylene dibromide
methylene chloride (dichloromethane)
methyl bromide
fluorocarbon-142A (FC-142A)
methyl chloroform (1,1,1-trichloroethane, MCF)
fluorocarbon-152A (FC-152A)
ethyl chloride
methyl acetate
acetic acid
methyl chloriau
methyl iodide
methanol
methane
carbon monoxide
xvi i
-------
co2
CO
x
C2C14
C2H6
C3H4
C3H8
C6H6
C9H12
C1CH CH Cl
CIO
C10N02
CIO
x
HBr
HC1
HF
HOC1
H2
NO
N02
NO
x
N2°
OH
°3
SF6
SO,,
carbon dioxide
oxides of carbon
perchloroethylene (PCE)
trichloroethylene (TCE)
trichloroacetic acid
acetylene
vinylidene chloride
ethane
methyl acetylene
propane
neopentane
benzene
1,3,5-trimethylbenzene (1,3,5-C H (CH ) )
ethylene dichloride
chlorine monoxide radical
chlorine nitrate radical
oxides of chlorine
hydrogen bromide
hydrochloric acid; (also) hydrogen chloride
hydrogen fluoride
nitric acid
hypochlorous acid
hydrogen gas
nitric oxide
nitrogen dioxide; (also) nitrite radical
oxides of nitrogen
nitrous oxide
hydroxyl radical
ozone
sulfur hexafluoride
sulfur dioxide
interhemispheric exchange rate
. xvin
-------
ACKNOWLEDGMENTS
The Environmental Sciences Research Laboratory would like to thank Mr. S.
Raymond of Northrop Services, Inc. for his diligence in obtaining and editing
the manuscripts for publication. Also,.we thank the Federal Aviation Adminis-
tration in Washington, D.C. for graciously making its auditorium available
for the Conference.
xix
-------
EDITOR'S NOTE ON DISCUSSION TRANSCRIPTS
The discussion transcripts included in this volume are based on a verba-
tim stenographic record of the Conference on Methyl Chloroform and Other
Halocarbon Pollutants. Certain editorial changes have been made, however,
to correct grammar, increase clarity, and avoid redundancy.
Throughout the transcripts, brackets ([]) indicate parenthetical comments
by the editor. The entire document (including discussion transcripts) employs
a consistent system of abbreviations and acronyms; individual speakers did
not necessarily use these shorthand expressions.
Every effort was made to identify discussion participants, and an alpha-
betical list of all those so identified is presented as Appendix A. Where
there was doubt as to a speaker's identity, the individual has been designated
as "Voice from Audience" or with insertion of a question mark after a name.
xx
-------
COMMENTS ON THE LIFETIMES OF ORGANIC MOLECULES IN AIR
A. P. Altshuller
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
INTRODUCTION
To begin this paper, a number of actual or potential impacts associated
with the more persistent molecules in the atmosphere are innumerated. By "more
persistent" are meant those molecules with lifetimes of tenths of a year and
longer. Less persistent molecules include those capable of participating in
lower tropospheric reactions with nitrogen oxides (NO ) over hours to days to
X
form ozone (0 ), nitrates, and other products.
Molecules with lifetimes of >1 yr, and especially with lifetimes of >10 yr,
can survive long enough to penetrate well into the stratosphere. Therefore,
such molecules may participate in the reactions that have been associated with
stratospheric 0 depletion in various models. The accumulation of certain
types of more persistent molecules in the stratosphere has also been associated
with global-scale climatic impacts.
The more abundant persistent molecules, such as methane (CH ) and carbon
monoxide (CO), participate in determining the distribution of key reactive
species, such as the hydroxyl ^Mical (OH), throughout the troposphere and
lower stratosphere. In addition, persistent molecules leading to carcinogenic
effects or other direct biological impacts contribute to population exposures
on all scales from local to global. The total impact of such persistent,
biologically-potent species can thus be substantially greater than the impact
of molecules with similar potencies but shorter atmospheric lifetimes.
1-1
-------
THE TROPOSPHERE AND ORGANIC MOLECULES
The remainder of this paper discusses the tropospheric lifetimes and
other tropospheric effects of a variety of molecules, including methyl chloro-
form (1,1,1-trichloroethane, CH CC13, MCF). Under consideration are certain
molecules not presently emitted in significant quantities, as well as other
molecules whose emission rates are uncertain or whose atmospheric distribu-
tions are not available in any detail. Only a simple approach to computation
of tropospheric lifetimes is feasible. Hence the dominant tropospheric re-
moval process is assumed to be OH attack, and an average tropospheric OH
concentration of 3.3 x 10 molecules/cm is assumed (Crutzen and Fishman 1977;
Fishman and Crutzen 1978). The rate constant expressions for OH reactions
with the molecules of interest have been discussed elsewhere (Altshuller
1979). An average tropospheric temperature of 265 K also is assumed. Table
1-1 lists the lifetimes calculated from the expression t = (K OH )
e avg
TABLE 1-1. ESTIMATED LIFETIMES OF METHANE AND HALOGENATED METHANE DERIVATIVES
Compound
CH.
4
CH_F
3
CH,C1
3
CH-,Br
3
CH F
CH0C1F
2
CH Cl
2 2
CHF
3
CHC1F2
CHC1-F
2
CHC1..
3
C H
2 6
Lifetime
(yr)
29,25,29
12
3.6,4.4
3.8
26
4.2
1.4
-1000
43,41,44
6.2,5.3,6.8
1.5
0.6
Compound
(cont'd.)
CH.CH^Cl
3 2
CH CHF
3 2
CH-CHCl-
3 2
CH-.CC1F.
3 2
CH CC1
CH^CICH.CI
2 2
CH BrCH^Br
2 2
CF CH Cl
3 2
CF CHC1F
CF0CHC10
3 2
CCIF^CH^CI
2 2
C Cl
2 4
Lifetime
(yr)
0.4a
6a
a
0.65
68
8,13
a
0.75
0.7a
19a
19
4.6
9.6
1.0
Based on rate constants at 265 K estimated by dividing 298 K values by 1.75.
1-2
-------
The lifetimes range from 0.4 yr for ethyl chloride (CH CH Cl) to ~10 yr
for trifluoromethane (CHF ). The selection of two available rate constant
expressions for MCF resulted in computed lifetimes of 8 and 13 yr. For fluoro-
carbon-22 (chlorodifluoromethane, CHC1F , FC-22), the three available rate
constant expressions lead to K values quite close to each other so that the
computed lifetimes cluster at 41, 43, and 44 yr. Therefore, FC-22 should be
substantially more persistent than MCF; with a lifetime in excess of 40 yr,
these molecules have ample time to penetrate well into the stratosphere.
A number of the molecules listed in Table 1-1 have been suggested as pos-
sible alternative propellants (Midwest Research Institute 1976). One of these
molecules is fluorocarbon-142A (CH CClF , FC-142A), but its computed lifetime
of 68 yr may lead to stratospheric damage if it is emitted to the atmosphere
in significant quantities. Fluorocarbon-133 (CF CH Cl, FC-133) has been
suggested as an alternative propellant, though it, too, has a relatively long
lifetime of 19 yr. But several of the fluorinated molecules do have lifetimes
of <10 yr, such .as fluorocarbon-152A (CH CHF , FC-152A), which is also con-
sidered a possible alternative propellant (Midwest Research Institute 1976).
This compound not only has a shorter lifetime, but would not be expected to
participate in the same stratospheric chain reactions as do molecules con-
taining Cl or Br.
The only two brominated compounds in production for which rate constants
can be computed with OH are CH Br and CH BrCH Br, which have substantially
shorter tropospheric lifetimes than MCF. Both of these molecules also have
dispersive emissions that are less than 5 percent of MCF. To contribute to
stratospheric O depletion effects, the Br atoms from these molecules would
have to participate very effectively alone or in combination with Cl molecules
in the reaction sequences of interest.
For several of the molecules listed in Table 1-1, sufficient experimental
data are available to permit comparison of computed with measured global con-
centrations. The computed values are based only on emissions resulting from
dispersive losses during manufacturing, storage, or use, because losses from
1-3
-------
natural sources on land or water, forest fires, and certain other combustion
processes lack adequate emission rate estimates.
The atmospheric loading of a substance is computed by adjusting the dis-
persive losses to the atmosphere for subsequent reactions with OH. These
losses are obtained from estimates of the fraction of total production of the
substance lost during production, storage, transfer, and use prior to and
including 1976. The average rates of consumption of the substance are computed
over intervals of lifetimes t , t , , t , ..., t , and the average
rate of consumption for an interval is multiplied by both the length of the
interval and the annual dispersive loss estimates related to that interval.
To obtain the net mass of the substance remaining in the atmosphere, the net
amounts of the substance available after reaction with OH are summed. The
mass expressed in grams is converted to milliliters at standard temperature
and pressure (STP) and divided by the volume of the earth's atmosphere at STP,
resulting in an average global concentration, where a uniform distribution is
assumed. In Table 1-2, these computed estimates are compared with average
global concentrations obtained experimentally by Singh et al. (1977; 1978;
1979). More details on the dispersive losses and additional experimental
measurements available for comparison are given elsewhere (Altshuller 1979) .
TABLE 1-2. AVERAGE GLOBAL CONCENTRATION OF HALOCARBONS FROM DISPERSIVE LOSSES
Compound
Net Dispersive Losses
(metric tons x 10 3)
Midwest Research Institute (1976).
Northern Hemisphere measurement only.
'Adjusted to January 1977.
Average Globa
Concentration
Obtained Experi-
mentally
Average Global
Concentration
Computed
CH,C1
3
CH.Cl,.
2 2
CHC13
CHC1F_
2
CH CC1
3 3
C-,C1.
2 4
26
491
24
270
1667,2025
569
613
32
8
20-30a'b
80°
26
3
32
1
18
71,86
20
1-4
-------
For two of the substances listed, methyl chloride (CH Cl) and trichloro-
methane (chloroform, CHC1 ), dispersive losses apparently account for only a
small portion of the mass emitted to the atmosphere. While emissions from
forest fires and from combustion of chlorinated plastics make contributions,
emissions from the oceans appear to be the source for most of the CH^Cl measured
in the atmosphere (Singh et al. 1977; 1978; 1979). The other sources of
CHC1 are not well known. A much lower concentration in the Southern Hemi-
sphere than in the Northern Hemisphere seems to eliminate the oceans as a
major source (Singh et al. 1978; 1979). Coastal waters and inland waters are
contaminated with CHC1-. that may be formed after chlorination of water, or
produced in the bleaching of pulp (Midwest Research Institute 1976). Further-
more, an urban-to-nonurban gradient exists for CHC13, suggesting one or more
sources associated with populated areas.
For the other four compounds considered in Table 1-2 (methylene chloride
(dichloromethane, CH_C10), FC-22, MCF, and perchloroethylene (C Cl , PCE)),
2. 2, Z. Q
computed and experimental measurements agree reasonably well (Singh et al.
1979), which lends some confidence to the OH concentration utilized. Emission
of these four compounds to the atmosphere seems to occur predominantly through
dispersive losses in production and storage, and in particular from use as
solvents (or, in the case of FC-22, as a refrigerant).
Of the compounds considered in Tables 1-1 and 1-2, at least two MCF
and FC-22 should be considered as possible contributors to stratospheric O
depletion. Because of substantial dispersive losses in the past and projected
for the future, and because both compounds have substantial tropospheric life-
times, the behavior of MCF and FC-22 (especially of MCF in the stratosphere)
should be of significant interest and concern. While CH Cl and PCE also have
substantial dispersive losses to the atmosphere, the shorter lifetimes of
these compounds limit their impact on the stratosphere.
The rates of consumption of organic compounds by OH vary widely. As
shown in Table 1-3, in terms of time for 1 percent consumption at 40° N
latitude in July, the time period involved varies from 23 days for CH to
1-5
-------
TABLE 1-3. NUMBER OF DAYS FOR I PERCENT CONSUMPTION OF ORGANIC COMPOUNDS
BY OH AT 40° N LATITUDE IN JANUARY AND IN JULY
Compound
CH4
C2H6
C3H8
^C4H10
i-c4Hio
i-C5H12
C(CH3)4
n-C6H14
C6H12
CH3C1
CH Br
CH2C12
CHC13
CHC1F2
CH3CH2C1
C1CH2CH2C1
BrCH2CH2Br
CH3CC13
C2H2
C3H4
1% Consumption
January
383
8.2
1.7
0.8
0.8
0.5
2.6
0.3
0.23
51
55
22
23
587
6.6
11
10
177
2.7
1.7
(days)
July Compound
23 C2H4
0.6 CH =CHC1
0.14 CHC1=CC12
0.07 CC1 =CC1
^ £
.0.07 C,H_
3 6
0.05
0.2 CH OH
0.03 C2H5OH
0.02 C3H7OH
3 . 7 HCHO
4.2 CH CHO
1.5
1.6 CH3COOCH3
1 *7 ^u o/"^/v~i TJ
-j ' k*-.n _ \^ ww^, _ .n _
/*"\ A f~\ TT f~l /"\("* LI
0.8
0.7 C..H
6 6
Uf~1 TJ ffT-I \
\^, ii V^-"*l -, /
m-C5H4(CH3)2
0.25 1,3,5-C6H3(CH3)3
0.17 °r C9H12
1% Consumption
January
0.18
0.21
0.6
14
0.55
1.6
0.4
0.3
0.18
0.09
9
0.85
0.5
1.4
0.26
0.07
0.025
(days)
July
0.02
0.025
0.07
1.0
0.006
0.16
0.04
0.03
0.02
0.01
0.9
0.08
0.05
0.14
0.025
0.007
0.003
1-6
-------
0.003 days for 1,3,5-trimethylbenzene (C H ) a ratio of almost 10000:1.
Comparison of the January rate of CH consumption with the July rate of CgH12
consumption increases this ratio to over 100000:1. Finally, if the rate of
CH consumption in January at 55° N latitude is compared with the rate of
C H consumption in July at 25° N latitude, the ratio increases to over
1000000:1. Because the ratio of OH concentrations at 25° N to 55° N latitude
is 3:1 in July, but 60:1 in January (Crutzen and Fishman 1977; Fishman and
Crutzen 1978) , winter season differences in rates of consumption of organic
molecules are especially significant as a function of latitude. These condi-
tions indicate consideration of not only the intrinsic differences in re-
activity of organic compounds, but also of latitude and season of occurrence,
if transformations of these substances on regional or continental scales of
movement are at issue.
Most organic compounds listed in Table 1-3 have previously befen demonstrated
to undergo reactions with NO in sunlight, forming significant levels of 0., and
X -J
consuming substantial amounts of the organics. These reactions occur within
the first 24 hours after emission to the atmosphere under summertime condi-
tions (Altshuller 1977). Of importance is that the more reactive of these
organic compounds have rates of consumption by OH differing by less than a
factor of 100 for a given season and latitude (Table 1-3). For example, in
July at 40° N latitude, the times for 1 percent consumption which are from 2
to 25 times longer than for C H include all of the alkanes larger than
propane (C_H_), all of the alkenes except PCE, all of the alcohols except
o o
methanol (CH OH), all of the esters except methyl acetate (CH COOCH ), and
all of the listed aromatic hydrocarbons except benzene (C..H ) . Therefore,
though a substantial number of very persistent halogenated alkanes exists, the
number of hydrocarbons in other series which are persistent is rather small
during the summer months.
A question of considerable interest relates to determining which persistent
organic substances undergo very slow reactions with NO , thereby producing
X
insignificant amounts of O_ in the lower troposphere. While the boundary area
of such reactions cannot be defined in absolute terms from the discussion
above, the importance of latitude and season of the year on reactivity in
1-7
-------
these systems should be clear. Since the homogeneous oxidations of sulfur
dioxide (SO ) and nitrogen dioxide (NO ) are dominated by reaction with OH
^ j£-
radicals, these reactions also will show strong dependencies on latitude and
season of the year.
Although most of the organic compounds listed in Table 1-3 disappear
rapidly in terms of hemispheric transport times, their consumption can be slow
compared to their movements on a regional or continental scale. In January,
all except the most reactive compounds undergo much less than 50 percent
conversion within a week's time at a latitude of 40° N or above. In July,
such organic compounds as ethane (C0H ) , C_H0, neopentane (i-C.-H, ) , halo-
l o J o D 12
genated alkanes, acetylene (C0H ) , methyl acetylene (C0H ), CH.OH, and C^H^
^ Z J 4 O DO
undergo conversion by OH of 50 percent or less during a 1-week period at 40° N
latitude or above.
Air parcels moving in trajectories over the North American continent
usually traverse most of the continent within a week. Because such movements
are rapid in comparison to July conversion of the more persistent organic com-
pounds and to conversion of most organic compounds in winter, rural and even
"remote" continental sites may be fumigated by air parcels containing such
compounds from urban centers and industrialized areas. Therefore, measure-
ments at these sites are not necessarily useful in providing tropospheric
background levels or in providing uncontaminated samples of natural organic
compounds.
REFERENCES
Altshuller, A. P. 1977. Formation and removal of SO and oxidants from the
atmosphere. In_ Fate of Pollutants in the Air and Water Environment, Part
II (I. M. Suffet, ed.). John Wiley and Sons, New York.
Altshuller, A. P. 1979. Lifetimes of organic molecules in the troposphere
and lower stratosphere. Adv. Environ. Sci. Tech. (in press).
Crutzen, P. J., and J. Fishman. 1977. Average concentrations of OH in the
troposphere, and the budgets of CH , CO, H and CH CCl . Geophys. Res.
Letters 4:321-324.
-------
Fishman, J., and P. J. Crutzen. 1978. The distribution of the hydroxyl radical
in the troposphere. Atmospheric Science Paper No. 284, Colorado State
University, Fort Collins, Colorado.
Midwest Research Institute. 1976. Chemical technology and economics in
environmental perspectives technical alternatives to selected chloro-
fluorocarbon uses. EPA-560/1-76-002, U.S. Environmental Protection Agency,
Washington, D.C.
Singh, H. B., L. J. Salas, H. Shigeishi, and A. Crawford. 1977. Urban-nonurban
relationships of halocarbons, SF , NO, and other atmospheric trace
constituents. Atmos. Environ. 11:819-828.
Singh, H. B., L. J. Salas, H. Shigeishi, and E. Scribner. 1978. Global distribu-
tion of selected halocarbons, hydrocarbons, SF and NO. EPA-600/3-78-100,
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina.
Singh, H. B., L. J. Salas, H. Shigeishi, and E. Scribner. 1979. Atmospheric
halocarbons, hydrocarbons, and sulfur hexafluoride: global distributions,
sources, and sinks. Science 203:899-903.
DISCUSSION
Dr. Jesson: A point of information: The lifetime you showed on the slide
[Table 1-1] seemed relatively long. What was the average?
DP. Altshutter: 3 x 10 , a value which will raise considerable discussion.
This value comes out of the Crutzen and Fishman model, is derived from Singh's
computations, and is in fairly good agreement with the experimental measure-
ments by Campbell et al. With much higher OH concentrations, at least according
to these calculations, tropospheric concentrations of these molecules occur that
are extremely small, in comparison with experimental concentrations. So one
must rationalize these discrepancies when using the values. I'm not arguing
the point between 3 and 4 x 10 . At 10 , 2 x 10 , and so forth, however, it
appears one must take into account other chemical reactions and kinetics in
terms of rationalizing the lifetimes of quite a few molecules.
Dr. Riordan: Your one box model: Does that require reproduction? And where
did you get that term?
Dr. Attshutter: Yes, for the values I cited from the missing slide [Table 1-2],
it indeed does. Of course, MCF values have been discussed in some detail.
Molecules of a relatively short lifetime present no great problem, because the
time period that must be cons-^^red in the calculations extends back only a
few years from the point when emission is assumed. That is the cutoff point
for molecules like PCE or CH Cl , for example. The molecules emitted 5 or 10
yr earlier make no contribution because they consume. So only a short his-
torical record backwards is needed to use such information as the A. D. Little
data on production and usage pattern and losses from the usage patterns for
1-9
-------
these individual molecules. In some cases, however, difficulties can arise;
for example, although U.S. production losses can be rather well estimated,
global production and losses cannot be accurately estimated using only the A.
D. Little data.
1-10
-------
STATUS OF THE
STRATOSPHERIC OZONE DEPLETION ISSUE,
INCLUDING COMMENTS ON METHYL CHLOROFORM
F. Sherwood Rowland
University of California
Irvine, California
EDITOR'S NOTE: Because Dr. Rowland did not provide a formal manuscript
for publication in this volume, the stenographic transcript of his presenta-
tion has been edited and reproduced below.
PRESENTATION
In going over and preparing a discussion of the status of the strato-
spheric ozone (O ) issue, I decided that it would be worthwhile to start at
the beginning and review what we know (or what there is reasonable agreement
about) as well as the areas of disagreement. I'm going to do this mostly from
slides.
This is a reproduction of the data of Lovelock taken in late 1971 and
published in 1973, showing the amount of fluorocarbon-11 (trichlorofluoro-
methane, CC1..F, FC-11) measured in the air as a function of latitude. This,
then, is the first measured set of data for FC-11. It is, basically, the
starting point of concern about halocarbons in the atmosphere. If you want to
check the range, it goes from 80 ppt to 40 ppt (these are 1971 levels). The
first question, I think, raised about such molecules measured in the tropo-
sphere was whether or not they reach the stratosphere. I think it is worth
reminding people that this question was raised and is now effectively settled.
2-1
-------
This shows the prediction, in 1974, of dependence of fluorocarbon-12
(dichlorodifluoromethane, CC1 F , FC-12). In this case, it is concentration
as a function of altitude. If we go to the top of the troposphere, we would
expect to see a decreasing amount as altitude increases into the stratosphere.
That was one of the first aspects that needed to be tested, and it has now
been tested. The next two slides show one of the mechanisms for doing that.
These are the grab sample flasks of the National Oceanic and Atmospheric
Administration (NOAA). The next slide shows them going up with the instrument
package. The next slide shows the measurements that were made in that way.
The NOAA samples are shown here, along with the National Center for Atmo-
spheric Research (NCAR) samples. These are the measurements of 1975. The
measurements, of course, have been repeated very many times since then. Mea-
surements are now available for the tropics and for high-latitude regions.
This slide was calculated for 30° latitude. Measurements were near that.
The tropic line, of course, shows a higher level, and high-altitude levels
would be below this. But what one finds, basically, is that the fluorocarbons
do reach the stratosphere and do fall off in altitude, as would be expected
with an effective stratospheric photodissociation process. If decomposition
did not occur, levels would go off much more towards these values. Thus, the
line here shows not only penetration to the stratosphere but also decomposi-
tion of the molecules themselves.
In the original inquiries into what would happen after the molecules
reached the stratosphere, investigators tried to identify the chlorinated
species that would be expected. The first expected chlorinated species to be
identified was the Cl atom, which is released by photolysis of the chlorinated
molecule. Starting with the Cl atom, investigators have had to include all
possible reactions that might occur in the stratosphere by taking into account
the various species existing there. Of course, it is possible to compile a
list of 20 or 30 or more chlorinated species. The question is: "How many of
these play an appreciable role in the atmosphere?" In our original list, we
showed Cl and the chlorine monoxide radical (CIO), involving the reactions
2-2
-------
shown here. If you want to arrive at a calculated distribution of these
within the stratosphere, then you need the right constants for all the reac-
tions involved. For example, for OH + HC1 you need the right constant in that
atmospheric temperature. You need to know the photolysis rates involved. Dur-
ing the last several years, then, essentially all the reaction rates shown here
were measured in the laboratory to sufficient accuracies that large errors re-
maining in any of these processes is unlikely.
The remaining question, then, is: "Are all of the molecules included?"
Obviously, this is the original version of the slide: it does not have, the
chlorine nitrate radical (C1ONO2). My present feeling is that it is almost
time to take ClONO back out again, in terms of being important. Neverthe-
less, one can r-alculate that there should be some ClONO present. I think the
present uncertainty about this formation rate results from the fact that most
of the methods for measuring the rate of disappearance of CIO in the presence
of the nitrite radical (NO ) did not establish this as the product. That is
probably not a trivial exception to raise at this point. It is quite possible
that CIO, in combining with NO , can make molecules other than ClONO . It may
^ ^
hook on in a different way, so that it becomes C1OONO (or something of this
sort). Also, there are some indications at the present time (e.g., Knauss in
Germany and Molina of the University of California) that this rate of forma-
tion is slower than the rate used in the National Academy of Sciences (NAS)
calculation. If the rate of formation is slower, it will have considerable
effect. Probably the area in which it will make the biggest difference is
rate of removal of ClO after sunset. For comparison with nighttime measure-
ments, we need to know this rate fairly accurately, because it tells us how
quickly ClO falls off once the sun goes down. So the question, here, concerns
ClO + NO : What is the product, and how well is it known? The values for
this rate that are being used are upper limits, I believe. The photolysis
rate, I believe, is reasonably well known at the present time.
Other molecules have been suggested here. Hypochlorous acid (HOC1), for
example, is a molecule which from time to time has been suggested as important.
That depends on the rate of photolysis of HOC1. The present situation, I be-
lieve, is that the absorption cross sections of HOC1 have been shown to be
2-3
-------
o
sufficiently large in the region between 3000 and 4000 A. Thus, HOC1 does not
have a long time in the stratosphere and cannot build up to be an important
compound. You can still write formation reactions for it with CIO and HO .
But it will also have a rapid removal process, and seems not to affect the
overall calculation in any way.
This does illustrate that you can go through and look for other processes.
If you ask "What is important?," the question becomes "How small a concentra-
tion are you concerned about?" That, in turn, probably depends on whether
there are any catalytic processes involving the particular molecule or whether
there is any way of stopping other catalytic processes. I believe that there
is reasonable agreement, at the present time, on the set of compounds shown
here.
The next slide simply illustrates the progress of the measurements.
These are Julius Chang's calculations of a year and a half ago, showing a dis-
tribution including a Cl atom here, this amount of CIO, a fairly substantial
amount of C10NO , and HCl as the predominant molecule.
It is necessary, next, to perform stratospheric measurements to confirm
or deny the existence of these various compounds in the concentrations in-
dicated here, or in relative concentrations. The next slides illustrate one
of the methods of doing that.
This was taken from the work of Dave Murcray. It involves infrared (IR)
spectra taken at an altitude of 30 km (typically, looking at the setting sun:
sunset = 90°). This is the zenith angle of the sun and this is 5° below the
horizon. It's below the horizon at ground level but still barely visible for
below the 30 km. We have long pathlengths through the atmosphere, here. We
see carbon dioxide (C0_) throughout. Here is FC-11 (the broad adsorption peak
under there). Here is FC-12. Nitric acid (HNO.J is only in the stratosphere;
water vapor is only in the troposphere. The existence of such measurements
makes it possible, then, to go back and look in older spectra, or to continue
to look in new spectra, for other molecules such as C1ONO . Current measure-
ments of ClONO , as I understand them, simply provide an upper limit. The
2-4
-------
apparent broadening of an O peak, if you attribute it to absorption by C1ONO2/
is then more or less in agreement with the upper limits of the calculations.
Thus, it looks as though there might be some ClONO there. But it is certainly
not cleanly identified as ClONO , and it is certainly not present in large
quantities.
This slide shows the measurement. This is Jim Anderson's resonance fluo-
rescent apparatus in flight. This is an earlier version of the present appa-
ratus, which has four pods that measure simultaneously and can measure Cl,
CIO, NO?, and so forth.
This slide illustrates the first (prepublished) runs of Jim Anderson in
which CIO was detected in the atmosphere. Detection of the Cl atom was direct.
CIO was detected indirectly, following conversion of CIO to carbon monoxide
(CO) by reaction with nitric oxide (NO) in the apparatus. According to my
most recent conversation with Anderson, he has had 10 good measurements ana-
lyzed, 9 of which are in reasonable agreement. The basic situation is that
there is a spread in his values. There is one high outlier that is very dif-
ficult to explain by any process, other than to say that something is experi-
mentally wrong on the particular flight. If there is something experimentally
wrong on a particular flight, then we should perhaps raise the question of
whether the error was present in the other 9 flights, as well. So, we must
really work hard to understand that high outlier. There are some very low
probability events that might be a source of Cl for a particular case. I
think we should watch Anderson's results and see if he gets more high outliers
or if it turns out to be just one run which is not reproduced at any later
time.
There are other measurements of CIO, including the very recent heterodyne
measurements by the Jet Propulsion Laboratory (JPL), which give a somewhat dif-
ferent profile up here. The microwave measurements of CIO (also by JPL) are
in agreement not with the Anderson high outlier but with the general bulk of
measurements (which are down low). The C1:C1O ratio seems to be in good order
in these experiments (probably within a factor of 2). Wherl -r we should do
better than that is a question that can still be asked. In general, I think
2-5
-------
we can say that several techniques have seen CIO in semiquantitative agreement
with the expectation of the calculations.
There is a measurement of the Cl atom. This concentration is much lower,
so it's a harder measurement to do. The Cl chemistry looks consistent, with
the exception of the high value that Anderson obtained on one flight.
As for production of FC-11 and FC-12 over the most recent years, we can
see how the FC-11 concentration would have settled here and then more or less
flattened out. I don't know whether this is going to show any more downward
change. It is certainly a very slow change from the peak. Certainly, from
1972 and 1973, there hasn't been much change. Essentially, then, in the early
1970's we moved into a steady-state period in which production is reasonably
constant. This happens to match nearly all calculations in which everything
is held constant starting at a particular point. What it suggests to me is
that regulatory actions taken so far have had very little effect. The amount
of production is essentially the same as before. U.S. regulatory actions for
aerosols, or the fact that they were coming into effect, did certainly affect
use of FC-11 and FC-12 as aerosol propellants. But they did not appreciably
affect total worldwide use. This is due partly to diversion of FC-11 and FC-
12 into other uses in the U.S. and to increased usage outside the country,
where no regulatory actions have been taken. The last time I referred to the
Chemical Marketing Reporter in regard to carbon tetrachloride (CCl^) (the pre-
cursor to FC-11 and FC-12), the journal's assessment was that the decrease in
CC1 production reflecting regulatory action had already essentially occurred,
and that production would start to rise again. And the journal was optimistic
for increased production in 1980. I'm not sure of the basis for that optimism.
A major, further question in terms of FC-11 and FC-12 is that of tropo-
spheric sinks. The subject has been raised for the last four or five years,
and a whole set of different possible reactions has been suggested. The prob-
lem is to find a tropospheric sink of appreciable importance. For example,
destruction in the internal combustion engine of the American automobile will
remove all FC-11 and FC-12 within 100000 yr. That is certainly a sink, but
it's not an important sink. So far, no one has been able to identify a
2-6
-------
particular sink which, by itself, is important. That leaves the alternative
of trying to find an undiscovered sink or accumulation of undiscovered sinks
that could be important. One of the methods of doing that is to simply cal-
culate how much FC-11 or FC-12 should be present and to compare it with the
amount observed to be present.
The next slide reproduces Lovelock's measurements again. Up here, going
even higher and higher, are the measurements made by Makide in my laboratory
about a year ago. Makide's values are not appreciably different from Singh's.
This is FC-11. Basically, between the Northern Hemisphere (NH) and Southern
Hemisphere (SH) there is very little gradient. At this point, it was ~150;
here, ~130. This is a gradient between the NH and SH on the order of only 10
percent. Contrast that with the gradient of Lovelock: an enormous gradient
in the NH itself, and also a gradient between the NH and SH. Now, if we're
going to make a comparison calculation of the amount of fluorocarbon in the
atmosphere, it is important to know how to weight these various amounts. If
the concentration doesn't change a great deal, it doesn't make too much dif-
ference. But if there's a big change (such as Lovelock's values indicate),
weighting the SH relative to measurements made in the north temperate region
will underestimate the amount of material in the SH.
One thing that is obvious right from the beginning is that the amount in
the atmosphere has gone up by a factor of ~2 since the first measurements were
made. Thus, FC-11 and FC-12 are certainly accumulating in the atmosphere at
a rate at which in 6 or 7 yr they have approximately doubled. That is in
agreement with the amount of FC-11 that has been released to the atmosphere in
a semiquantitative fashion.
The next slide shows a comparison of these different weighting measure-
ments. A couple of research groups, including the du Pont research group,
have mentioned tropospheric sinK_ of lifetimes of 10 to 20 yr. They have
weighted their data using a reproduction, basically, of Lovelock's values.
The calculation on which Molina and I collaborated a couple of years ago
assumed reasonably rapid mixing between the NH and SH which would take ~2 yr.
2-7
-------
This gave us a value of ~80 percent as much in the SH as in the NH. In
contrast, the other weighting shown here is on the order of 60 percent as
much. You can see the difference between these two weightings; there's a
substantial amount of material in between. The crosshatched area shows the
amount our measurement would have to be reduced to get a 20-yr tropospheric
lifetime. This indicates several things. First, it is important how you
weight. Also, it's going to be difficult to make very accurate measurements.
To begin to eliminate tropospheric lifetimes on the order of 40 to 50 yr would
require very, very accurate measurements. Experimentally, that would be an
almost impossible task, I think.
The next slide shows the same data and compares the first results ob-
tained here. This is as of a couple of years ago; since that time, the data
have tended to be a little bit higher, I believe. This shows that we also
underestimated the SH, but not by as much as the du Pont group did. Of course,
the NH and tropic region were also grossly underestimated in those. When we
do the same calculation over again now (take our own data and weight them),
the measurements are made from 55° N to 55° S. So we're not having to ex-
trapolate long distances. In fact, the calculations show that there is rea-
sonably uniform mixing (say, 10 percent) and not much of a gradient. When we
calculate the amount that is present, we again come out with a finding for FC-
11 that is in reasonable agreement with the amount expected to be in the atmo-
sphere after correcting for stratospheric loss, with no indication of any ap-
preciable tropospheric sink. We obtain the same result for FC-12. Actually,
as far as we're concerned there seems to be a little bit of a discrepancy in
the other direction. We seem to find a little bit more than we might expect,
underestimating stratospheric loss. But that is probably within error of the
measurement.
In summary, our measurements so far indicate no need to introduce any
process except stratospheric loss to account for removal of FC-11 and FC-12.
We can then go into a calculation of what this means in terms of O_ depletion.
Shown here is the growth with time. This is the NAS curve as calculated
by Julius Chang. This is the calculation of 1977, with the first revision of
2-8
-------
HO + NO rate and going to -1'3 percent. The most recent number, T believe, is
18.6 percent. The basic change has been HO, + NO. This rate constant, being
very much faster, makes NO in the lower stratosphere no longer an O depleter,
x
to put it crudely. Tying it up with C1ONO doesn't have any effect on the Cl
calculation. As a result, the effect of Cl went up by a factor of 2 or, per-
haps, somewhat more than the most recent calculations.
At the present time, then, I think the calculations are suggesting num-
bers in the 15 to 20 percent region. That is what we would expect for steady-
state depletion with continued release of FC-11 and FC-12 at the rates that
have been common all through the mid-1970's. I think it is worth pointing
out, too, that it's easy to calculate but it is easy to be misstated. When
one says that the calculation is "15 percent," that's the asymptotic value
that one is looking for. Of course, you can reach the asymptote any time you
want the year 2500 or the year 3000.
I think it's probably worthwhile to start talking about the depletions
expected at particular time periods. The current estimate would be a deple-
tion on the order of 2 percent. Well, if you look at these numbers coming out
to the year 2000, you're ~l/3 of the way to the asymptote; out to ~2035, you're
~2/3 of the way to the asymptote. So you get ~l/3 of the result by the end
of this centry and ~2/3 of the result 1/3 of the way through the next century.
Whatever value you're going through here, keep in mind that you get 1/3 of it.
A value of 20 percent means 67 percent by the end of this century. And 14
percent, say, by the year 2035. That gives you some idea of the time scale on
which this would develop.
These are ultraviolet (UV) measurements taken in 1974 at Mauna Loa,
Hawaii and Bismarck, North Dakota with a device that measures the amount of
UV in the sun received by the device from January to January. As you can see,
the amount of UV radiation received in the northern tier of the U.S. is not
much less in mid-summer than it is in Mauna Loa. Of course, there's a lot
more radiation received over the whole year there. If we know that Mauna Loa
is at 19° and Bismarck is at 6°, we can integrate under this to see what the
total UV exposure would be here. You can't move too easily from that to
2-9
-------
knowing what the exposure would be for human beings, because you have to know
what fraction of time they spend in the sun all through these periods. But
the correlation certainly is that more UV-B is available the closer you get to
the equator.
This is a typical curve of the incidence of skin cancer as a function of
latitude. Measurements were made at 10 different stations. Looking at the
correlation, there are three things that come in: cloud cover, altitude, and
latitude. These take care of almost all of the risks of the variations in-
volved. There's very little altitude effect; ~70 percent of the effect is
latitude.
Curves of this sort show a higher incidence of skin cancer in the southern
as opposed to northern U.S. Coupled with the curves I've shown before, this
leads to calculations of predicted increases in human skin cancer as a result
of depletions in 0 .
If you accrued the figure if you were to deplete the O by 10 percent
you would get a 20 percent increase in UV-B (the sunburn ultraviolet). And
that would lead to a 30 percent increase in human skin cancer. The accuracy
of those numbers is not great, because it involves taking slopes of this kind
and making assumptions about human exposures.
The cancer is malignant melanoma. The O measurements were taken at Arosa
in Switzerland, and the data are plotted in two ways. One plot, the blue lines,
shows yearly values over the 50-yr period. The red line shows the 5-yr running
average. The 5-yr running average, as you can see, ends in 1975. The blue
lines come back up again. The 5-yr running average is still down. At the end
of 1977, the 5-yr running average was at its lowest point since the start of
measurements.
There are two things that we can see here: First, there is enormous fluc-
tuation. Also, the change is 5 percent in one direction or the other. An
actual O gain or loss on top of that fluctuated pattern would probably mean
a change of several percent. If we had comparable data from large numbers of
2-10
-------
stations going back over 50 yr, then we could cut down on that. But such data
and stations simply don't exist. Many of the stations which do exist have
calibration problems that are generally absent from these Arosa data. This
has been a major project over a long period of time, and they have done a lot
in terms of calibrating their apparatus on a regular basis and trying to take
out all the instrumental variations that you get in a 50-yr time sequence of
this sort. There is certainly a question as to how much O change we would
have to see in order to demonstrate a loss of O . It still looks to me as
though several percent would be required. And even though this 5-yr running
average is down at a low point, we cannot conclude one way or the other wheth-
er it is on a general downward trend or will come back up in the future.
The next slide brings in methyl chloroform (1,1,1-trichloroethane, CH CC1_,
MCF). I will just briefly describe what we have done here. First, MCF cer-
tainly has a tropospheric sink. There's a known tropospheric sink in the re-
action of the hydroxyl radical (OH). The question of how much OH is actually
present on a world-wide average is still open, and the MCF residence time has
been estimated at anywhere from ~1 to ~15 yr. Of course, that makes an enor-
mous difference as far as possible regulatory action might be concerned. It
also makes a big difference because MCF offers a good chance to calibrate the
effect we would expect for other CH-containing compounds, such as some that
Dr. Altshuller mentioned earlier. It's important to get a good measurement
of the lifetime of MCF.
My group (like several others) has tried to do that. We've made the
latitude measurement shown here. Basically, these measurements were made in
the vicinity of January 1978 and have been corrected to January 1, 1978. The
measurements were all made within 1.5 months on either side of that date and
were corrected on the basis of a 1 percent/month change.
Now, MCF is going up rapidly. So we need to know exactly what date we're
talking about when we discuss concentrations. Shown here are the NH and SH
temperate zones. There are only a few tropic measurements. What we have done
is to interpret the worldwide average as of January 1, which was 80 ppt. The
question of the NH average depends somewhat on what is done in the tropics.
2-11
-------
Here there is only one measurement. All of these measurements, with the
exception of this one, were made on ocean coastlines where the prevailing wind
was off the ocean. So these are Alaska, the West Coast, the West Indies, and
Chile.
There are two things that we can do here. One, we can take the whole 80
ppt worldwide average, compare it with the calculated emissions given by Dow,
and see what that gives us for the lifetime on a worldwide basis. We can also
interpret it in terms of the N/S ratio. For that, we need to make some as-
sumption about the relative concentrations of OH in the two hemispheres. This
last slide shows a calculation of that. Here is our N/S ratio. As you can
see, we arrive at a lifetime on the order of 6 yr. Looking at these cross-
hatched lines, the main thing is that the 0.15 lines cross here, not far from
the measured point. The fact that the calculated rates of removal in the NH
and SH are about the same makes it difficult to accommodate large excesses of
OH in the SH. That is, at this point, we should have much less in the SH. We
have a different N/S ratio coming down here. If removal were twice as fast in
the SH, we would expect it to be somwhere down here. And we just don't see
that. So, as far as MCF is concerned, we're using 15 months as the exchange
time between the two hemispheres. It looks to us as though the OH rate of
removal is about the same in the two hemispheres, and that the lifetime of MCF
is ~6 yr, which is somewhat less than the value Dr. Altshuller calculated.
We've done the same thing with methane (CH ) . The lifetime of CH ought
to be 2 to 2.5 times longer. Dr. Altshuller's calculated value was 25 or 30
yr. It's very hard (if not impossible) to explain a 25- to 30-yr lifetime for
CH. on the basis of OH removal. Also, there is a gradient in the troposphere.
There is 7 percent more CH in the north temperate region than in the south
temperate region. If there's more CH in the NH, you expect it, I think, to
be produced there, because that's where the landmass is. Nevertheless, it's
very difficult to accommodate anything more than ~12 yr for the lifetime of
CH . And that forces the MCF lifetime down to 5 or 6 yr, in order to get a
gradient.
-2-12
-------
DISCUSSION
from Audience: In your calculation, you showed the photodissociation
. I think the group would suggest that
of
ClONO .
Dr. Rowland: Regarding the photolysis product of C1ON02, there's a publica-
tion by Smith^ and me in which ClONO + O is shown to form. That photolysis
was at 3025 A. In order to calculate what happens next, the ensuing behavior
of ClONO must be known, and that turns out to be essentially a null process.
The measurements by Golden1 s group were at 2700 A, and those measurements may
not be pertinent to what happens at 3100 A. They are flashing, actually, so
they have much more energy available.
If Cl atoms were present in our system, then adding CH which we tried
should take them out. We did not find HC1 . So, under our- conditions, we
concluded that Cl atoms were not present. If you accept Cl + NO , your
calculation then depends on whether NO, becomes NO + NO or NO +0. It can
cause further O_ depletion under some circumstances, but I don t think it's a
big effect. And I'm not sure that their [Golden group] results are pertinent
for the wavelengths that are important in stratospheric photolysis.
Voioe from Audience: In the two checks of the tropospheric lifetime of MCF,
the overall average depends clearly on efficient figures and absolute accuracy
of the measurements. Could you say a few words on the accuracy of your figures?
Perhaps we should ask a few other people, too.
Dr. Rowland: We think they're good to ±6 ppt 80 ±6 ppt. We think our num-
bers are in disagreement with Dr. Singh's numbers.
Dr. Rasmussen: We have integrated average values for FC-12, FC-11, MCF, and
CC1. based upon ~1000 measurements since 1976. All data represent grab samples
or traveling samples obtained on aircraft flights, or surface measurements at
various locations . The important point is the integrated averages are weighted
for the latitudinal atmospheric mass ratio.
As Dr. Rowland stated, his 13 points gave an interhemispheric average for
MCF with a difference of ~1.29. Our values resulted in ~1.26 for 1978. How-
ever, in 1979 a very interesting anomaly was observed that was not attribut-
able to calibration; in these data sets, we have maintained the same primary
calibration standards since 1976. In any case, no real difference is seen
between Rowland's MCF data and our own, in view of the relative difference
between hemispheres and based on a global average. The absolute numbers may
differ, likely because of a ~20 percent systematic difference.
Dr. Rowland: A 20 percent difference in measurements of the absolute concen-
trations means a factor of ~1.6 in the lifetime of MCF. For many purposes, a
factor of 1.6 is not very important at all. Whether 6, 9, or 10 yr is cal-
culated for the tropospheric lifetime, the qualitative and the semiquantita-
tive conclusions are basically unaffected: MCF remains in the troposphere
long enough for some of it to enter the stratosphere. We're not talking about
big differences, in any event, between Singh's or Rasmussen 's and our values,
but a relatively small difference of ~20 percent.
2-13
-------
Voi.ee from Audience: Is the MCF and OH reaction product trichloroethylene
or -ethene?
Dr. Rowland: I purposely wrote it as C Cl-H to not indicate where the
chlorines were.
Voice from Audience: I am perhaps showing my ignorance. But if it's an
ethane, wouldn't there be another hydrogen? Wouldn't there be one less?
Dr. Rowland: That would result in a trichloroethyl radical.
Voice from Audience: What is the half-life of that radical? Isn't that an
important consideration here, taking it all the way down to HC1?
Dr. Rowland: It's important but it's not clearly understood; I write it this
way because a one, two atom shift is suspected to occur immediately after the
abstraction process. Demonstrating the nature of this trichloroethyl radical
and its reaction products is something we're working on in the laboratory.
Certainly no statement in the literature describes the processes exactly. The
overall assumption, however, is that unless some molecule is found to be
stable against photolysis all three chlorines will be released. This must
be demonstrated. Arguments can be made that certainly one and maybe two of
them are given off, but the exact pathways are not known.
Voice from Audience: Are you implying that with this radical, at least for
two chlorines, that the effect on 0-, depletion could be substantially beyond
what we are estimating at this point?
Dr. Rowland: With a lifetime in the 6- to 10-yr range, about 15 to 20 percent
of the MCF will be decomposed in the stratosphere; and the first-order approx-
imation is for all chlorines to be released. Therefore, a stratospheric Cl
source would be available from the decomposition of MCF. The effect of fluoro-
carbons would decrease by roughly the fraction destroyed in the stratosphere,
which, as I say, is 15 to 20 percent. On a tonnage basis, the fluorocarbon
problem is 1/5 to 1/10.
2-14
-------
ENVIRONMENTAL FATE OF METHYL CHLOROFORM
W. Brock Neely
G. Agin
The Dow Chemical Company
Midland, Michigan
INTRODUCTION
In order to study the environmental fate of a chemical, two vital pro-
cesses must be considered: transport and transformation. Once adequate knowl-
edge is available in these areas, an assessment can be made of the concentra-
tion that will produce an impact on the various ecosystems.
Transport deals with movement of the material within and between major
environmental compartments. In the case of a volatile chemical, the compart-
ments of main concern are the troposphere, stratosphere, and hydrosphere. The
lithosphere, composing only 30 percent of the earth's surface, will be neglect-
ed for the low-molecular-weight halocarbons that are the subject of this re-
port.
Transformation concerns the chemical reactions affecting a substance
emitted to the atmosphere. These reactions are important for cleansing the
system of the chemical. With continued addition of a product in the absence
of dissipating mechanisms, the load, and hence the concentration, build to an
undesirable level at a future po^nt in time. The precise time is dictated by
the rate at which the agent is added and by its intrinsic toxicity or other
adverse effect.
3-1
-------
The bottom line in studying environmental fate is to ascertain the poten-
tial impact on the earth's biosphere. The impact may be either direct or in-
direct. A direct effect would be a chemical buildup that becomes toxic to
susceptible organisms. An indirect effect might be expressed through a series
of reactions that results in an undesirable biological event.
Fluorocarbon-11 (trichlorofluoromethane, CCl^F, FC-11) and fluorocarbon-
12 (dichlorodifluoromethane, CCl^F , FC-12) have become classic examples of
compounds thought to produce such an indirect effect. Molina and Rowland
(1974a) observed that no known dissipating reactions occur for these volatile
chlorofluoromethanes (CFM's) in the lower atmosphere, which led them to the
stratosphere. In investigating the degradation of CFM's in the stratosphere,
they became aware of the possible impact of the reactions on depletion of the
ozone (0 ) layer. The proposed biological consequences of this depletion have
since led to legislation and regulations controlling emissions of these com-
pounds .
Through the same line of investigative reasoning, methyl chloroform (1,1,1-
trichloroethane, CH CC1.,, MCF) has become involved in a somewhat similar en-
vironmental issue (McConnell and Schiff 1978; Singh 1977; Crutzen et al. 1978).
Resolution of this issue is the crux of the present workshop.
This report considers two major questions:
(1) What is the residence time of MCF in the troposphere? This length
of time is critical, since it indicates how much of the released chemical will
reach the stratosphere intact. Numbers of years ranging from 1 (Chang and
Wuebbles 1976) to 8 (Singh 1977) have been suggested.
In order to avoid any confusion, the relation between rate constant,
residence time, and half-life should be mentioned. These are all related by
Equations 1 and 2. The half-life
= In2/k (Eq. 1)
where k = first-order rate constant in reciprocal time units
3-2
-------
is the time required to reduce the concentration by one-half. The residence
time
residence time = 1/k (Eq. 2)
is the average length of time that a chemical will remain in an environmental
compartment under a reaction subjected to the rate constant, k.
(2) What is the impact of MCF on the stratospheric O-. layer? Presenta-
tion of a few major unresolved questions will illustrate the uncertainty re-
garding halocarbon impact on this important gaseous layer.
RESIDENCE TIME OF METHYL CHLOROFORM IN THE TROPOSPHERE
Stochastic Approach
In this approach, the historical emission of MCF is statistically fit to
monitoring data collected on MCF in the troposphere. The model to be used is
shown in Figure 3-1 and has been reported previously (Neely 1977; Neely and
Plonka 1978). The first task is to establish meaningful values for the rate
constants in Figure 3-1:
(1) The input for the model is shown in Table 3-1 and is the Dow Chemical
Company's best estimate of the worldwide emission of this solvent to the
troposphere. Since the Southern Hemisphere (SH) can be emitting no more than
3 percent of the total (according to our estimate of the use pattern), the
material is assumed to be vented in the Northern Hemisphere (NH).
The input (k ) is expressed as a continuous function of time by linear inter-
polation between the annual emission rates shown in Table 3-1.
3-3
-------
A
Stratosphere
B
Troposphere
D
Ocean
North
South
Figure 3-1. Statistical model used to fit monitoring data collected on MCF
in the troposphere. Statistical determination of rate constants
k and k is given in Table 3-4.
3-4
-------
TABLE 3-1. WORLDWIDE GLOBAL EMISSION OF METHYL CHLOROFORM
Year
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
Released to
Atmosphere
(millions of pounds)
0.
0.
2.
6.
17.
27.
43.
45.
66.
79.
83.
124
112
125
161
240
288
320
327
341
368
508
750
800
804
917
940
1050
3
5
1
0
6
4
2
7
8
6
8
Percent
Change
66.6
320
185
193
56
57
5.7
46
19.16
5.2
48
-9.6
11
28
49
20
11
2.19
4.28
7.9
38
47.6
6.67
0.5
14.05
6.87
12.2
Total 8548
3-5
-------
(2) The internemispheric exchange constants (k~ and k.) are assigned
values ranging from 1 to 0.70/yr, based on previous studies (Singh 1977;
Pressman and Warneck 1970). A range of 0.71 to 0.86/yr, representing a
residence time of 14 to 17 months, is used in this study.
(3) The estimated exchange rate between the oceans and the atmosphere is
based on the studies of Liss and Slater (1974), McKay and Leinonen (1975), and
Neely (1976). Using a mixing depth of 10 km in the troposphere and 100 m in
the ocean (Neely and Plonka 1978), values for k and k are calculated (see
Table 3-2).
(4) Rate constants for MCF movement from the troposphere to the strato-
sphere are assigned values ranging from 0.02 to 0.03/yr, as suggested by
earlier CFM work (Neely 1977).
Using these rate constants and the values of the parameters shown in
Table 3-2, a statistical fit is made to the data shown in Table 3-3. An
average value of 1.49 for the North/South ratio is used to estimate the SH
concentration. The two parameters adjusted are the dissipating rate constants
k.. and k.,. These results are summarized in Table 3-4.
o /
TABLE 3-2. PARAMETERS FOR THE STATISTICAL MODEL SHOWN IN FIGURE 3-ia
Parameters Description Value
V ,V Weight of air in 1/2 of troposphere 2 x 10 g
assuming height of 10 km
22
V Weight of water in NH 1.54 x 10 g
22
V Weight of water in SH 2.09 x 10 g
k Flux between air and ocean 9.98/yr
k Flux between water and air 0.066/yr
k Transfer from troposphere to 0.02 - 0.03/yr
stratosphere
k ,k Transfer between NH and SH 0.86 - 0.71/yr
aTaken from Neely and Plonka (1978).
3-6
-------
TABLE 3-3. MONITORING DATA ON METHYL CHLOROFORM IN THE NORTHERN TROPOSPHERE
Year
1972.0
1973.5
1974.4
1975.5
1976.3
1976.5
1976.3
1978.0
Concentration North/South
(ppt) Ratio
37
52.5
68
80
85
92
85
98
(24-45)S
(42-60)
(60-76)
(75-85)
(80-90)
(88-95) 1.51
(70-92) 1.47
1.49
Reference
Lovelock
Lovelock
Lovelock
Lovelock
Lovelock
Lovelock
(1977)
(1977)
(1977)
(1977)
(1977)
(1977)
Rasmussen et al.
(1976)
Rowland
liminary
(pre-
data)
Estimated range.
Table 3-4 indicates that a 20 percent increase in k., and k causes a 30
percent decrease in k,. and a 44 percent increase in k... The overall effect
b /
changes neither the average dissipation rate constant nor the calculated
tropospheric concentrations, but does alter the asymmetry of the hydroxyl
radical concentration ([OH]) between the NH and SH. The faster the exchange
rate constant, the greater the asymmetry, confirming one of the conclusions by
Singh (1977). The dissipation constant is less sensitive to an alteration in
the exchange constant in the stratosphere than it is to an alteration in the
interhemispheric exchange constant. A 50 percent increase in k causes only a
10 percent decrease in the degradation constant.
Mass balance analysis indicates that only 2 percent of total MCF emis-
sions are found in the ocean (Neely and Plonka 1978). Consequently, the
degradation in this environmental compartment is not a significant loss mech-
anism; it is neglected in the present study.
3-7
-------
TABLE 3-4. SUMMARY OF STATISTICAL DETERMINATION OF
THE DISSIPATING RATE CONSTANTS k AND k_a
6 /
Simulated Value (per yr)
Rate
constant
k3 G k4
k$
k estimated
o
k_ estimated
I
0.71
0.02
0.13
0.17
Simulation
II
0.71
0.03
0.11
0.16
III
0.86
0.02
0.072
0.24
Calculated Concentrations of MCF in the
Northern Troposphere (ppt)
1972.0
1973.5
1974.5
1974.9
1976.0
1976.2
1976.3
1976.5
1978.0
39.8
51.1
64.3
69.3
80.9
83.0
83.8
86.0
103.1
39.8
51.1
64.3
69.3
80.9
83.0
83.8
86.0
103.1
39.9
51.1
64.3
69.2
80.9
83.0
83.9
86.0
103.2
Estimated Future Concentrations of MCF (ppt)
North South
1979
1980
1981
1982
1984
1986
1988
115
88
72.6
61.1
43.9
31.2
22.7
80
78.6
68.8
58.8
42.4
30.5
21.9
k [M] + )c7[Sj
Average k = [N] + [s] - 0.15/yr
Residence time = 6.6 yr
1/2 " 0.15
4.5 yr
Production terminated in 1978 for Simulation I
3-8
-------
Mechanistic Approach
In this technique the major dissipating reaction of MCF in the troposphere
is assumed to be via OH attack. Ample evidence indicates that this process
regulates the degradation of CFM's (Darnall et al. 1976).
Three items are required to evaluate a tropospheric degradation rate
constant for MCF: (1) a value for average [OH] in the troposphere (see Table
3-5); (2) knowledge of the bimolecular rate constant for the attack of OH on
MCF; and (3) a value for the average temperature in the troposphere. Of
these three items, the assumptions dealing with temperature and [OH] are the
least valid. However, if valid numbers could be determined, then unquestion-
ably a tropospheric residence time could be evaluated for MCF by this tech-
nique .
TABLE 3-5. AVERAGE TROPOSPHERIC HYDROXYL RADICAL CONCENTRATIONS
Global
Average
Concentration
7.5
(yearly
average)
Northern
Troposphere
Concentration
8.0
North/South
Ratio Basis
Photochemical
models of OH
production
CO studies
Reference
Warneck (1975)
Stevens (per-
7.6
2.5
5.0
(average for
fall day)
0.36 Photochemical
models of CO
and CH
Photochemical
model
Fourier analysis
of seasonal dis-
tribution of OH
sonal com-
munication)
Crutzen and
Fishman (1977)
Crutzen (1974)
Derwent and
Eggleton (1977)
a 53
All concentrations are reported in 10 molecules/cm .
3-9
-------
OH concentrations estimated by a number of investigators are shown in
Table 3-5. In this collection, [OH] based on MCF is not included, to avoid
biasing the determination of MCF residence time.
The average value for [OH] appears to be in the range of 6 x 10 mole-
cules/cm . A degree of controversy surrounds the asymmetry of the concentra-
tion between the NH and SH. Values ranging from 1 (Fishman and Crutzen 1978)
to 0.25 (Neely and Plonka 1978) have been suggested. Crutzen and Fishman
(1977) established a ratio of 0.36 with a higher value in the SH; this value
is here used to estimate [OH] in the SH, even though the concentration is also
uncertain. The asymmetry is normally explained on the basis of greater carbon
monoxide (CO) production in the NH than in the SH and the ability of CO to act
as a sink for OH.
Table 3-6 lists various determinations for the rate constant shown in
Equation 3.
OH
(Eq. 3)
.-12
An average value of ~3.1 x 10 exp (-1507/T) is used in the following analy-
sis.
The final number required is a time-weighted average of the tropospheric
temperature. Temperatures in the troposphere range from 15° c at sea level to
-45° C at 10 km (Neely and Plonka 1978) . A value of -8° C is the estimated
average (Watson et al. 1977; Davis et al. 1976).
TABLE 3-6. BIMOLECULAR RATE CONSTANT FOR HYDROXYL ATTACK ON METHYL CHLOROFORM
Arrhenius Expression
(cm molecule s )
Reference
~12
1.95 x 10~ exp(-1331/T)
3.72 x 10~ exp(-1627/T)
3.5 x 10~ exp(-1562/T)
aAverage value « 3.1 x 10 exp (-1507/T).
3-10
Chang and Kaufmann (1977)
Watson et al. (1977)
Crutzen et al. (1978)
-------
The photodegradation rate constant for k and k in Figure 3-1, assuming
that the dissipating reactions are caused by OH attack, may now be calculated.
Values of k and k? are 0.20/yr and 0.55/yr, respectively:
k = k x [OH] x 3.15 x 10 = 0.2/yr
6
where k = 1.05 x 10 cm /molecules s at -8° C
3.15 x 10 = s in 1 yr
[OH] = 6.0 x 10 molecules/cm
(Eq. 4)
k? = k6/0.36 = 0.55/yr
The average tropospheric cleansing rate is a function of the various rate
constants. Where dissipating reactions are dominated by k- and k , as in
Figure 3-1, the model is reduced to the set of reactions shown in Equation 5:
A
B
(Eq. 5)
7
where A and B represent the concentrations of the chemical at any time in the
northern and southern troposphere, respectively.
Setting k = k , the differential Equations 6 and 7 may be readily solved,
yielding the solution given by Equation 8.
dA
dt
k A + k B
(Eq. 6)
dB
dt
-k?B
(Eq. 7)
3-11
-------
{Vk3
A =
(6-a)
0^ (Eq. 8)
(a-8)
where A and B = initial concentrations
- C
a =
- C
D = 2k, + k,. + k_
Jo/
C = 4(k_k^ + k_k_ + k,.k_)
JO J / D /
Since e reaches zero much faster than e , the dissipation is governed
by the expression for g. For the situation where k, = 0.86/yr, k, = 0.2/yr,
and k_ = 0.55/yr, g has a value of 0.36/yr, or an average residence time of
2.78 yr. In the case of the statistical model, the average residence time is
6.6 yr (Table 3-4). Considering the assumptions, the agreement is reasonable
and falls within the previously reported range of 1 to 8 yr.
The two main factors to be tested for sensitivity in the mechanistic
model are temperature and [OH]. With [OH] constant at 6 x 10 molecules/cm ,
the rate constant is estimated to vary by ~5 percent for every 1 percent
change in the absolute temperature. Similarly, if temperature is constant
at -8° C, k shows a 1 percent change for every 1 percent change in [OH] .
Here again, uncertainty in estimating the average tropospheric residence time
for MCF is reflected. The following discussion is based on residence .times
ranging between 2.78 and 6.6 yr.
3-12
-------
TRANSFER OF CHLORINE FROM METHYL CHLOROFORM TO THE STRATOSPHERE
A relatively high concentration of 0 is located in a narrow band in the
stratosphere. This layer is important because the o~ molecule is a strong
absorber of ultraviolet radiation and hence protects the earth's surface from
potentially harmful solar radiation.
The main environmental impact of low-molecular-weight halocarbons is the
effect of Cl atoms on the O ; a quantitative estimate of the Cl atoms reaching
this altitude is of vital concern and is given by:
B = k (A x 1.4 x 10 x molecular weight) (Eq. 9)
where k_ = exchange to the stratosphere
A = global average tropospheric concentration (on a mol/mol basis)
1.4 x 10 = number of molecules of air in troposphere
For 1978 the global average MCF concentration was reported by Rowland (un-
published) to be 82 ppt. The flux (B) into the stratosphere, assuming 0.02/yr
as a value
year 1977.
9 9
as a value of k , ranges from 30.5 x 10 to 24.2 x 10 g of Cl atoms for the
The usual yardstick for measuring such a flux is to compare it with the
reported analysis for CFM. In the 1977 National Academy of Sciences (NAS)
study of CFM, production of FC-11 and FC-12 was assumed constant at the 1973
9
level. At steady state, this level produces a flux of 500 x 10 g of Cl atoms
into the stratosphere. The same scenario of steady-state annual MCF produc-
6 9
tion, at the 1978 rate of 1100 x 10 pounds/yr or 396 x 10 g of Cl atoms,
9 9
would introduce 20.84 x 10 - 46.6 x 10 g Cl atoms/yr into the stratosphere,
with the calculated value depending on the residence time used (2.8 to 6.6
yr). These figures amount to 6 to 12 percent of the annual production. On an
arbitrary scale, MCF impact is more than an order of magnitude less than the
CFM impact.
3-13
-------
The MCF impact would reach the CFM level of effect after ~38 to 55 yr at
6 percent annual growth. If, during this time period, a real problem was
perceived, the tropospheric MCF concentration could quickly be reduced because
of the relatively short residence time; i.e., the tropospheric concentration
could be reduced by 1/2 every 2 to 4.5 yr.
The amount of MCF ultimately reaching the stratospheric 0~ may be con-
siderably less than 6 to 12 percent. In the lower stratosphere (below the
major O, layer), [OH] is sufficiently great (Crutzen and Fishman 1977) and
diffusion sufficiently slow (McConnell and Schiff 1978; Cicerone et al. 1974)
that more MCF than previously estimated (solely on tropospheric reactions) may
degrade before reaching the O layer. MCF is concluded to be less efficient
in destroying O than are the CFM's (McConnell and Schiff 1978). This area
needs further investigation.
IMPACT OF CHLORINE ATOMS ON STRATOSPHERIC OZONE
Before discussing the impact of these atoms on O.,, it is useful to
review the major reactions associated with formation and destruction of this
gas layer .
0., is formed through Equations 10 and 11:
O2 + hu -> 2 O (Eq. 10)
O + 02 -> 03 (Eq. 11)
Once formed, the layer is maintained at a steady-state level by means of cyclic
destruction mechanisms (Table 3-7) .
Cycle 4 of Table 3-7 was identified as a possible loss mechanism by
Cicerone et al. C1974) and Molina and Rowland (1974b) . Theoretically, con-
tinued increase in the use of stable halocarbons will cause Cycle 4 to become
more dominant; reduced O concentration in the stratosphere will result.
3-14
-------
The researcher's information base for the nature or magnitude of the
Cl atom impact suffers from a lack of good rate data for known important
reactions and from a lack of awareness of all reactions involved. Growing
understanding in this field of chemistry will be most useful in establishing
a better basis for decision-making.
TABLE 3-7. STRATOSPERIC OZONE DESTRUCTION MECHANISMS
Cycle
I
2
3
4
°3 + hu
0 + O.
0 + H02
HO + 03
0 + N02
NO + O
O + CIO
Mechanism
"*" °2 + °
-> OH + 02
-> H02 + 02
-> NO + O2
-> O + NO
-> Cl + 02
Percent
Destruction
20
10
69
0.5
Loss of 0, from
stratosphere to
troposphere 0.5
FURTHER QUESTIONS AND RECOMMENDATIONS
The remainder of this report lists areas where questions are raised and
indicates where more work is required. The chlorine cycle (Cycle 4 of Table
3-7) may be interrupted by conversion of Cl into an inactive species. Equa-
tions 12 through 14 represent three such reactions:
Cl + CH4 -> HC1 + CH3 (Eq. 12)
CIO + N02 -> C10N02 (Eq. 13)
OH + Cl -* HOC1 (Eq. 14)
3-15
-------
Some evidence exists indicating hypochlorous acid (HOCl) is a reservoir for OH
and Cl (National Academy of Sciences 1977). If so, additional Cl might con-
ceivably cause an 0_ increase by interfering with Cycle 2 (Table 3-7). How-
ever, a report conclusion is: "C1OH, like ClONO , may be a temporary pseudo-
inert reservoir species, although the likelihood of a major impact on the ClO
X
cycle appears to be small" (National Academy of Sciences 1977).
More disturbing than the observation that HOCl may act as a sink is that
stratospheric chlorine monoxide radical (CIO) measurements are much higher
than could be expected if all known Cl-containing compounds were converted to
ClO (National Academy of Sciences 1977). Only two explanations seem possible,
according to the NAS report either the measurements are in error, or the
possible existence of some unidentified source of Cl in the stratosphere is
indicated. This question must be resolved expediently to determine the true
impact of man-made halocarbons on stratospheric 0 .
At one time, increased use of nitrogen fertilizer and/or increased
numbers of supersonic transports were believed to have a catastrophic effect
on the O-. layer (National Academy of Sciences 1976; Johnston 1975) . These
assumptions were related to an increased stratospheric NO level and increased
X
O depletion due to Cycle 3 (Table 3-7). New studies (Turco et al. 1978) have
modified this view, and predictions are for an increase in O. below 25 km, due
to the higher levels of NO and water vapor. The question thus arises as to
the possible environmental effects of an increased level of O^, as opposed to
a decreased concentration. In addition, NAS has previously reported that the
revised rate constants for some of the reactions related to the odd nitrogen
cycle indicate that Cycle 3 is not as important as earlier thought (National
Academy of Sciences 1977). Consequent to this assessment has been the specula-
tion that Cycle 4 is more important; the decrease in O3 depletion from Cycle 3
has been replaced by an increase in 0 depletion from Cycle 4. One firm con-
clusion of all the stratospheric chemistry research was reiterated by NAS: "It
is now completely clear that all the major chemical catalytic cycles are closely-
coupled and cannot be studied separately" (National Academy of Sciences 1977).
3-16
-------
An interesting study relating to the rates of all the reactions in the
stratosphere was recently published (Groves et al. 1978). The authors suggest
that increased stratospheric carbon dioxide (CO ) levels will cause worldwide
cooling. Since Equations 10 and 11 proceed faster at lower temperatures, but
the various destructive cycles proceed more slowly, the expected net result is
an increase in 0.,. The CFM's were compared with and without the added effects
of CO . The decrease in O column density attributed to CFM by the year 2030
was 4 percent, assuming a continued steady release at 1973 rates, with a
steady-state reduction of 6 to 7 percent a few decades later. The corre-
sponding figure in the year 2030, incorporating the temperature effect of CO ,
amounted to a 5 percent increase in the O layer. Without CFM's, the increase
might be even higher.
A recent laboratory-simulation study of the upper atmosphere (Benson 1978)
showed a lack of sensitivity to CFM concentrations varying over 4 orders of
magnitude. This laboratory-simulation study parallels recent studies from the
Goddard Space Flight Center (Air/Water Pollution Report 1978). Using satel-
lites to monitor the upper atmosphere, the measurements indicated that dete-
rioration of the stratospheric O is 1/2 the amount deduced from ground
measurements.
While undoubtedly some of the Cl-containing molecules reach the strato-
sphere, their exact impact is uncertain. Since the impact is not known with
any degree of confidence and in view of the safety valve on MCF in terms of a
relatively fast half-life, the conclusion of Dow Chemical Company is that MCF
must be very low on the priority list of potential environmental hazards.
REFERENCES
Air/Water Pollution Report. 1978. Satellite studies indicate ozone depletion,
but not as much as previously thought. 16(48)(Nov. 27):474.
Benson, S. W. 1978. Personal communication to W. B. Neely, December 15. The
research mentioned was performed by Dr. Harteck at Rensselaer Polytechnic
Institute, Troy, New York.
Cicerone, R. J., R. S. Stolarski, and S. Walters. 1974. Stratospheric ozone
destruction by man-made chlorofluoromethanes. Science 185:1165-1167.
3-17
-------
Chang, J. S., and P. J. Wuebbles. 1976. A theoretical model of global
tropospheric OH distributions. Proceedings of Non-Urban Tropospheric
Composition, Hollywood, Florida, November.
Chang, J. S., and F. Kaufman. 1977. Kinetics of the reactions of hydroxyl
radicals with some halocarbons: CHFCl , CHF Cl, CH CCl , C HC1 , and C Cl .
J. Chem. Phys. 66:4989-4994.
Crutzen, P. J. 1974. Photochemical reactions initiated by and influencing
ozone in unpolluted tropospheric air. Tellus 26:47-57.
Crutzen, P. J., and J. Fishman. 1977. Average concentrations of OH in the
troposphere, and t
Letters 4:321-324.
troposphere, and the budgets of CH , CO, H and CH CCl.,. Geophys. Res.
Crutzen, P. J., I. S. A. Isaksen, and J. R. McAfee. 1978. The impact of the
chlorocarbon industry on the ozone layer. J. Geophys. Res. 83:345-363.
Darnall, K. R., A. C. Lloyd, A. M. Winer, and J. N. Pitts, Jr. 1976. Reactive
scale for atmospheric hydrocarbons based on reaction with hydroxyl radical.
Environ. Sci. Tech. 10:692-696.
Davis, D. C., G. Machado, S. Conaway, Y. Oh, and R. Watson. 1976. A temperature
dependent kinetics study of the reactions of OH with CH Cl, CH Cl , CHCl^,
and CH Br. J. Chem. Phys. 65:1268-1274.
Derwent, R. G., and A. E. J. Eggleton. 1977. J. Atmos. Environ, (preprint).
Fishman, J., and P. J. Crutzen. 1978. The distribution of the hydroxyl radical
in the troposphere. Dept. of Atmos. Sci., Colorado State U., Fort Collins,
Colorado.
Groves, K. S., S. R. Mattingly, and A. F. Tuck. 1978. Increased atmospheric
carbon dioxide and stratospheric ozone. Nature 273:711-715.
Johnston, H. S. 1975. Ground-level effects of supersonic transports in the
stratosphere. Accounts of Chem. Res. 8:289-294.
Liss, P. S., and P. G. Slater. 1974. Flux of gases across the air-sea inter-
face. Nature 247:181-184.
Lovelock, J. E. 1977. Methyl chloroform in the troposphere as an indicator
of OH radical abundance. Nature 267:32.
McConnell, J. C., and H. I. Schiff. 1978. Methyl chloroform: impact on
stratospheric ozone. Science 199:174-177.
McKay, D., and P. J. Leinonen. Rate of evaporation of low-solubility con-
taminants from water bodies to atmosphere. Environ. Sci. Tech. 9:1178-1180.
Molina, M. J., and F. S. Rowland. 1974a. Stratospheric sink for chlorofluoro-
methanes: chlorine atom-catalyzed destruction of ozone. Nature 249:810-812.
.3-18
-------
Molina, M. J., and F. S. Rowland. 1974b. Predicted present stratospheric
abundances of chlorine species from photodissociation of carbon tetra-
chloride. Geophys. Res. Letters 1:309-312.
National Academy of Sciences (Committee on the Impacts of Stratospheric Change).
1976. Halocarbons: Effects on Stratospheric Ozone. National Academy
of Sciences, Washington, D. C.
National Academy of Sciences (Committee on the Impacts of Stratospheric Change).
1977. Response to the Ozone Protection Sections of the Clean Air Act
Amendments of 1977. National Academy of Sciences, Washington, D. C.
Neely, W. B. 1976. In: Proceedings of 1976 National Conference on Control of
Hazardous Material Spills, New Orleans.
Neely, W. B. 1977. Material balance analysis of trichlorofluoromethane and
carbon tetrachloride in the atmosphere. Sci. Total Environ. 8:267-274.
Neely, W. B., and J. H. Plonka. 1978. Estimation of time-averaged hydroxyl
radical concentration in the troposphere. Environ. Sci. Tech. 12:317-321.
Pressman, J., and P. Warneck. 1970. Stratosphere as a chemical sink for carbon
monoxide. J. Atmos. Sci. 27:155-163.
Rasmussen, R. A., D. Pierotti, J. Krasnec, and B. Halter. 1976. Trip report
submitted to N. Anderson, National Science Foundation, Washington, D. C.
Singh, H. B. 1977. Preliminary estimation of average tropospheric HO con-
centrations in the Northern and Southern Hemispheres. Geophys. Res.
Letters 4:453-456.
Stolarski, R. S., and R. J. Cicerone. 1974. Stratospheric chlorine: Possible
sink for ozone. Can. J. Chem. 52:1610-1615.
Turco, R. P., L. A. Capone, R. C. Whitten, and I. G. Poppoff. 1978. SSTs,
nitrogen fertilizer and stratospheric ozone. Nature 276:805-807.
Warneck, P. 1975. Hydroxyl production rates in the troposphere. Planet.
Space Sci. 23:1507-15,18.
Watson, R., G. Machado, B. Conaway, S. Wagner, and D. D. Davis. 1977. A
temperature dependent kinetics study of the reaction of OH with CH9C1F,
CHC12F, CHC1F2/ CH3CC13, CH3CF2C1, and CF^lCFCl^ J. Phys. Chem. 81:256-262.
DISCUSSION
Dr>. Eeioklen: Your calculations yield slightly more optimistic estimates. In-
stead of using the product of the average OH times the average temperature
times the average concentration, you used the average of the products. Of
course, where the OH is highest is where the temperature is highest; the rate
3-19
-------
coefficient will go up the most, probably another 10 or 12 percent. It is
appropriate to use the average of the products, not the product of the averages.
Dr. Neely: Good. Thank you.
Voice from Audience: Do you feel very confident about estimating a 6 percent
per year increase in demand for MCF? How do you folks arrive at these figures,
and how well will they hold for the future? We are interested in the validity
of other sorts of estimates, also.
Dr. Neely: Since I'm not in the department of the marketing groups. Dr.
Farber, would you respond?
Dr. Farber: Only incomplete data are available for formulating estimates;
however, the estimates are based on expected growth in recognized use areas,
which is dependent on the GNP in this country and on worldwide product usage;
on expected growth in use of available alternates; and on the economics for
the user. These are gut decisions that we make, and sometimes not well, if
the discrepancy between the capacity and the market for perchloroethylene is
considered. But I think the 6 percent is based on a combination of expected
growth of the general GNP here and worldwide, in the European area primarily,
so that is a global number in the economics of the products used. Does that
help?
Voice from Audience: Do you associate an uncertainty with that?
Dr. Farber: Plus or minus 100 percent from that. In other words, I don't
expect to see attrition of the product, barring some unforeseen unfortunate
experience; and I don't expect to see 12 percent average over the next 5
yr. 5 yr is the best time frame to use, really. Historically, MCF emissions
have displayed peaks and valleys, and in Dr. Neely's first slide [Table 3-1],
quite a dramatic growth is seen in the last 2 or 3 yr. The actual numbers
turn out to be in a range between -1 percent and +12 percent; the average of
that, over the last 5 yr, is ~5 percent, I think. We project about the same
rate fluctuation: we expect to see years where pressures, of whatever nature,
cause increases in the 10-percent range; we expect to see years where things
are bad and we see a net loss in the product's use.
Dr. Singh: The emission inventory statements you have provided and I guess
that is exactly what Dr. Neely has used show exponential growth rate over
the last 10 or 12 yr of ~17 percent average for MCF emissions as well as for
global production; that would be the average if you put it in a log plot.
Dr. Neely: I think that is right. I think if you put it into a log model, an
exponential model, for the last 16 yr, it comes out to 16.5 percent. I have
these percentages on a table [Table 3-1] in the manuscript. Beginning with
1962, a 9.6 percent decrease is seen, then percent increases of 11, 28, 49,
20, 11, 2.19, 4.28, 7.9, 38, 6.67, 0.5, 47.6, 14.05, 6.87, 12.2. Average
change over the last few years is ~16, but the percent change fluctuates. Our
best estimate for the fluctuation in the next few years is ~6 percent.
3-20
-------
Voice from Audience: For the uninitiated of us, are you giving us figures for
capacity or production? Do you have excess capacity or not currently?
Dr. Neely: These figures are a best estimate of emission. I think we have
excess capacity. Right.
Dr. Rowland: I have two questions. First: What accuracy do you attribute to
the estimates of the world production of MCF that are used in these calcula-
tions?
Dr. FarL . I would comment for the Western World and the Free World in the
East outside the Communist bloc. I doubt that much production occurs in
China, but I know the products are used in Russia. Disregarding whatever
input exists from the Communist bloc, we're accurate within 10 percent for
sure.
Dr. Rowland: The other question is: For what fraction of that production is
Dow Chemical responsible?
Dr. Farber: I don't have an exact number, but ~50 percent on a global basis.
If you really need to know that for your modeling, I will be glad to get it
for you.
Dr. Rasmussen: I have one slide: these are our best calculations at the
moment for the past 3 yr, integrating the concentrations in both hemispheres.
Surprising to us is that between 1977 and 1978 the mass burden of MCF in the
NH increased only 7 percent. This rate does not fit with Dr. Rowland's in-
crease per month figure.
The internal consistency in our calibration standards from 1976, 1977,
1978, and through 1979, shows ±~2 or ~3 ppt variance in the primary standards.
Previously, with less extensive data in both hemispheres and less coverage in
time, we had calculated a much larger growth rate. The value for 1978 repre-
sents something on the order of up to 200 measurements. The MCF data from
the Manufacturing Chemists Association program in the SH indicate a very -TOW
order of change in the SH.
3-21
-------
-------
HALOGENATED TRACE CONSTITUENTS IN THE GLOBAL ATMOSPHERE
Hanwant B. Singh
Louis J. Salas
Hisao Shigeishi
SRI International
Menlo Park, California
DISTRIBUTIONS, SOURCES, AND SINKS OF HALOGENATED TRACE CONSTITUENTS
Distributions
Most estimates of the budgets and residence times of halocarbons have
been based on point measurements, and the hemispheres have been assumed to be
well mixed. Recently, Singh et al. (1979) presented extensive global measure-
ments of several halogenated and nonhalogenated species and characterized
their growth rates over a period of 3 yr. This global distribution covered
an area between 64° N and 90° S latitudes at widely varying longitudes.
Figure 4-1 shows the global distribution of many halocarbons for late
1977. For all stable halogenated species (fluorocarbon-12 (dichlorodifluoro-
methane, CCl^, FC-12) , fluorocarbon-11 (trichlorofluoromethane, CC1 F, FC-
11), fluorocarbon-113 (CC1 FCC1F , FC-113), fluorocarbon-114 (CCIF.CCIF , FC-
^ ^ 22
114), sulfur hexafluoride (SF&), and carbon tetrachloride (CC1 )), the average
concentration in the Northern Hemisphere (NH) and Southern Hemisphere (SH)
differs only marginally (10 to 15 percent). Using the emissions data for FC-
12 and FC-11 (Manufacturing Cheirists Association 1978) an interhemispheric
exchange rate (Tg) of 1.2 yr can be calculated. Figure 4-1 also shows the
global distribution of methyl chloroform (1,1,1-trichloroethane, CH CC1 ,
4-1
-------
Q.
a
^-'400
CM
LL
200
300
1
^"200
a.
100
60
1
*"" 40
n
20
30
S
20
u.
10
06
a
a
^- 04
o
- =
g.S-Tj.-,.-.. ^
b
-
1430 (64.7) ppb
1.1'ppb
5 »--""
....' f '
-90°-80° -60° -40° -20° 0° 20° 40° 60° 80° 9O°
S Latitude (deg)
S Latitude (deg)
Figure 4-1. Global distributions of atmospheric constituents. Key: ppt = 10~12 (v/v); ppb = 10 9 (v/v), (t) is the average
concentration in the hemisphere; (1) is the standard deviation; (*) indicates that for species where a significant gradient within
each hemisphere is observed the weighted average concentration is defined to represent the total burden of the species in that
hemisphere; V, trip 1, stainless steel vessels; A, trip 1, glass vessels; fy, trip 2, stainless steel vessels, and n, trip 2, m-situ
air sampling and analysis. The dashed line is a third-order polynomial fitted to the data. In most cases, individual hemi-
spheres can be treated as well mixed. In the case of CH3CCI3 and C2H6, where this is not true, the global profile is well
represented by the polynomial 89.71 + 0.818 L + 7.584 x 1fl"4L2 - 7.894 x 10~5L3 for CH3CCI3 and 0.769 + 9.926 x 10"3L + 6.526
x 10"5L2 + 5.561 x 10"8L3 for C2H6, where L is the latitude (in degrees) and varies from -90° to +64° (NH = 0° to -1-90°; SH =
0° to -90°).
4-2
-------
MCF) and a third-order polynomial fitted to it. This compound shows a latitu-
dinal distribution quite different from that of the fluorocarbons. In lati-
tudes above 30° N, the MCF is well mixed in the NH with an average concentra-
tion of ~123 ppt. A fairly sharp decline seems to occur between 20° N and 20°
S, and the concentration of MCF then levels off to ~75 ppt. The decline
cannot be attributed to normal mixing processes, since fluorocarbons do not
show this '-apid decline. A more plausible explanation of this phenomenon is
that the hydroxyl radical (OH) is more abundant around the equator because of
the intense sunlight and the high concentration of water vapor in this region.
The weighted average of MCF concentrations, ~113 ppt in the NH and ~77 ppt in
the SH, best describes the burden of MCF in each hemisphere.
Methyl chloride (CH Cl) had been measured in the NH by numerous research-
ers, but no SH data had been available. Figure 4-1 shows an essentially
uniform global distribution, with an average global concentration of 615 ppt.
The relatively short lifetime of CH Cl (2 to 3 yr) and its uniform global dis-
tribution support the idea of a large natural source. The primary man-made
emissions of CH Cl had been thought to be negligible. To the contrary, how-
ever, CH Cl concentrations of nearly 2200 ppt were found in Lisbon. The Los
Angeles vicinity (Riverside) showed an average CH Cl concentration of 1500 ±
700 ppt (maximum 3800 ppt), ~2.5 times the background measured concentrations
(Singh et al. 1978b). Thus, a significant urban source of CH Cl seems to
exist. The possibility that automobile exhaust or other combustion processes
may be such a source should be investigated.
Growth Rates
Figure 4-2 shows the growth rates of FC-12, FC-11, CC1 , and MCF, four of
the most important man-made halocarbons, in the north temperate regions.
Despite a recent decline in the use of fluorocarbons, the atmospheric burdens
of FC-12 and FC-11 clearly increased at rates of ~19 ppt/yr and ~13 ppt/yr,
respectively. CCl increased at a rate of ~2 ppt/yr, while MCF increased at
~16 ppt/yr. The atmospheric growth of these halocarbons is consistent with
the available emissions data (Singh et al. 1976; 1979).
4-3
-------
300
250
200
100
200
ISO
160
J-,40
_ 120
100
80
60
Average growth rate * 18.5 ppt/year (ion/year)
- 1>,2 = 65 to 70 years; Te - 1.1 to 1.2 years
' Average growth rate = 12.9 ppt/year 112%/year)
1>n = 40 to 45 years. Te = 1 1 to 1 2 years
£W
180
160
1 1*0
»120
O
0
100
80
60
'
Average growth rate = 23 ppt/year (2%/year)
-
r-
li i I - 1
T t 11-
1 1
-
:
.
-
200
180
160
Average growth rate = 155 ppt/year
-------
TABLE 4-1. TROPOSPHERIC CONCENTRATIONS OF
IMPORTANT ATMOSPHERIC TRACE CONSTITUENTS*
Compound Group Chemicals
b Data
Source
Quality
Volumetric
c Mixing Ratio
NH
Nitrogen N.,0
compounds
Fiuorinated CF (FC-14)
(nonchlorinated)
species 6
Chlorofluoro- CC12F2 (FC"12>
Carb°nS CC13F (FC-11)
CC1,FCC1F CFC-113)
CC1F2CC1F2 (FC-114)
CHCljF (FC-21)
Chlorocarbons CH Cl
cci4
CH3CC13
CH2C12
C2C14
C2HC13
CHC13
C2C16
Brominated CH Br
species
lodated species CH.,1
Hydrocarbons CH
CO, CO , and H
2 2 C-H,
2 b
C2H2
CO
co2
H2
N
A (?)
A
A
A
A
A
A
N
A
A
A
A
A
A
A
N,A
A
N
N,A
N,A
A
N,A
N,A
N,A
2
3
2
2
2
2
2
3
2
2
2
3
3
3
3
3
3
3
3
2
3
3
2
1
2
310
65
0.3
250
150
20
12
5
611
122
113
44
40
16
14
<5
ppb
ppt
PPt
ppt
ppt
PPt
ppt
ppt
ppt
ppt
PPt
ppt
ppt
ppt
ppt
ppt
SH
310
65
0.3
225
135
13
10
4
615
119
77
20
12
<3
S3
ppb
ppt
ppt
ppt
ppt
PPt
ppt
ppt
PPt
ppt
ppt
ppt
ppt
ppt
ppt
-
Average
Atmospheric
Growth Rate
Between
1975-1977
0-1 ppb/yr
19 ppt/yr
13 ppt/yr
2 ppt/yr
16 ppt/yr
5-20 ppt
<5
<2
1600
1
<0.2
100-250
336
580
ppt
ppt
ppb
ppb
ppb
ppb
ppm
PPb
<2
1500
0.5
<0.2
60
334
550
-
ppt
PPb
ppb
ppb
ppb
ppm
ppb
faAs of early 1978.
N natural, A anthropogenic.
1 excellent data base with uncertainties <5 percent
2 fair data base with uncertainties <15 percent
d3 fragmentary information.
ppm = 10 v/v; ppb = 10 v/v; ppt = lo" v/v.
4-5
-------
Budgets
Table 4-1 shows average concentrations of trace constituents of interest
in the two hemispheres. The data are from several sources: most halocarbon
data are taken from Singh et al. (1979); the fluorocarbon-14 (carbon tetra-
fluoride, CF , CF-14) data are taken from Rasmussen et al. (1979). The halo-
carbon sources are shown and the data quality is categorized. Also shown are
the average growth rates of important halocarbons.
Table 4-2 provides total tropospheric budgets of organic Cl, Br, I, and
F, and indicates source contributions. The global average tropospheric Cl
concentration (organic) for late 1977 is 2.7 ppb, implying that the maximum Cl
measured near 40 km should be <2.5 ppb. The possibility that unknown sources
of tropospheric Cl exist cannot be completely dismissed, but measurements of
total organic Cl (Berg and Winchester 1976) do not support the existence of
large unknown Cl sources. Of the total 2.7 ppb Cl, ~77-percent is man-made
and the remaining ~23 percent appears to be of natural origin.
TABLE 4-2. TROPOSPHERIC CHLORINE, BROMINE, IODINE,
AND FLUORINE ORGANIC BUDGETS
Species
Budgets
Source
Contributions
NH
Cl 2.9 ppb
Br 10-30 ppt
I <2 ppt
F 1.0 ppb
SH Globe Na Ab
2.4 ppb 2.7 ppb 23 77
50-90 10-50
<2 ppt <2 ppt 100 0
0.9 ppb 1.0 ppb 0 100
, N natural.
b
A anthropogenic.
The organic Br and I budgets are much more uncertain. Measurements by
Singh et al. (1979) suggest only two brominated species, methyl bromide (CH Br)
and ethylene dibromide (CH BrCH Br). Even for these two species the data base
4-6
-------
is highly scarce (see Table 4-1) . A preliminary interpretation suggests an
organic Br content of 10 to 30 ppt, of which 50 to 90 percent may be of natural
origin (Table 4-2) .
No significant sources of organic I appear to exist. The only species
identified currently is methyl . iodide (CH I) , which is present at a concen-
tration of ~2 to ~5 ppt in the marine boundary layer. The tropospheric mean
concentre -ion of CH I is expected to be much less than 2 ppt. The atmospheric
residence time of CH I is ~5 days.
The Cl, Br, and I budgets are contrasted with the budget for F, which
shows ~1 ppb of F almost entirely attributable to man-made sources. Very
possibly other sources of organic Br and I are yet to be identified. Special
emphasis should be devoted to identification of new species and better charac-
terization of those already measured.
Oceanic Sink
Halocarbons were also measured in Pacific seawater, primarily to deter-
mine the ability of the ocean to act as a source or a sink for them. The
average surface concentrations of individual species are given in Table 4-3.
The average measured surface-water concentrations (expressed in ng/liter)
were: FC-12, 0.28; FC-11, 0.13; CC1 , 0.40; CH Cl, 26.28; and trichloro-
methane (chloroform, CHC1_) , <0.05. With the surface-water concentrations of
halocarbons known, a simple film-diffusion model of the flux of halocarbons
into or out of the ocean can be determined (Junge 1976; Singh et al. 1978a) :
where F = the flux from ocean to air
D = diffusion coefficient
Z == film thickness
w
C = concentration of the species in water
w
C = concentration in equilibrium with the burden in air
eq
4-7
-------
TABLE 4-3. OCEANIC SINKS
Compounds
FC-12
FC-11
cci4
CH3C1
Average Surface
Concentration
(ng/liter)
0.28 (>0.05a)
(Min 0.07)
0.13 (>0.063)
(Min 0.07)
0.4
26.8
Flux into
the Ocean
very small
very small
+3.2 x 1010 g/yr
-3 x 10 g/yr
T (yr)
very large
very large
100
~2
Saturation concentration.
Solubility data for FC-12 and FC-11 (Junge 1976) suggest that, if the
surface water is in equilibrium with the atmospheric burden, the concentra-
tions of FC-12 and FC-11 in water should be ~0.05 and ~0.06 ng/liter, respec-
tively . These concentrations are lower than the measured average concentra-
tions of 0.28 and 0.13 ng/liter, indicating that ocean water is supersaturated
with FC-12 and FC-11. Therefore, either the solubility data are inaccurate or
the water samples were inadvertently contaminated. Another possibility is
that the ocean surface waters have been contaminated by man-made activities on
a global scale. The lowest concentration of FC-12 and FC-11 measured, 0.07
ng/liter, is about what one would expect if the surface water were saturated
with FC-12 and FC-11. If the surface water were saturated, the ocean would be
a relatively ineffective sink for FC-12 and FC-11 but could act as a reservoir
containing <0.5 percent of the atmospheric burden of FC-12 and FC-11 in a
steady-state situation.
The average surface water concentration for CC1 was 0.40 ng/liter. The
flux of CCl. into the ocean can be calculated from Equation 1, with D = 10
2-1
cm s , Z = 90 vim, and S = 0.85. The solubility, S, in seawater is defined
w
as the ratio of the species concentration at the air-sea interface (C ) to
the atmospheric concentration at standard temperature and pressure. A high Z
4-8
-------
is used because CCl is rapidly absorbed in fatty tissues and may be biologi-
cally active. For such species the upper limit of the stagnant film thickness
calculated from random data (63 ± 30 urn) is more appropriate. Using Equation
1, a CCl flux into the ocean of 2.8 x 10~ g cm s can be calculated. If
4 10
this flux is assumed typical of all oceans, an exchange rate of 3.2 x 10
g/yr is obtained. The atmospheric burden of CCl. from the measurements is
12
calculated as 3.2 x 10 g. Thus
vide a turnover rate of 100 yr (T
12
calculated as 3.2 x 10 g. Thus the ocean is a sink for CCl that can pro-
,12 10,
cci4
ments thus indicate that the oceanic sink for CCl is about half as effective
as the stratospheric sink.
The surface concentration of CH Cl in the Pacific is quite variable
(Table 4-3), with values somewhat higher near the equator. The average sur-
face concentration was 26.8 ng/liter. Using an S of 2.65 (Dilling 1977)
CH^CJ.
and other parameters as defined earlier, a CH Cl flux from the ocean to the
-14 -2 -1
atmosphere of 2.6 x 10 g cm s is estimated. Extending this flux to the
12
world ocean body gives an exchange rate of 3.0 x 10 g/yr. From these
12
measurements, the atmospheric burden of CH Cl can be estimated as 5.5 x 10
g. Thus, on the basis of limited data, the ocean appears to be a significant
source of CH Cl, which can provide an atmospheric turnover rate of ~2 yr.
This rate is in reasonable agreement with the estimated CH Cl residence time
5 5 3
of ~2 to ~3 yr, due to OH attack (OH = 3 x 10 to 5 x 10 molecules/cm .
Residence Times
A comparison of emissions data for FC-12 and FC-11, with the help of a
two-box model, suggests an average FC-12 residence time of 65 to 70 yr and an
average FC-11 residence time of 40 to 45 yr (Singh et al. 1979). Figure 4-2
shows the good agreement between measurements and calculated values. Because
of an important oceanic sink, the CCl, residence time can be estimated at be-
tween 25 and 40 yr. Several estimates of the MCF residence time have been
made; these are summarized in Table 4-4. The best estimates of MCF residence
time seem to lie between 6 and 12 yr, excluding the outlier point from Neely
and Plonka (1978). This range should be compared with the MCF residence time
4-9
-------
of 1.4 yr reported by the National Academy of Sciences (NAS) (1976). A 6- to
12-yr residence time allows 12 to 25 percent of the MCF released at ground
level to enter the stratosphere. This long tropospheric residence time, when
coupled with the rapidly increasing emissions, suggests that MCF may be a
potential depleter of stratospheric ozone (O ) in the decades ahead. World-
wide release of MCF to the atmosphere currently approaches 7 x 10 g/yr and
is increasing at a rate of 10 to 15 percent per year. Continued uncontrolled
release of MCF to the atmosphere should be a matter of future concern.
TABLE 4-4. GLOBAL AVERAGE RESIDENCE TIME (T ) OF METHYL CHLOROFORM
cl
Estimated from
Field Data
Singh (1977a)
Lovelock (1977)
Singh (1977b)
McConnell and Schiff (1978)
Chang and Penner (1978)
T
(yr)
7 ± 1
5-10
8-11
8
12
Model T
Estimates (yr)
all estimates 1-3
prior to 1977
National Academy 1 . 4
of Sciences (1976)
Crutzen and Fishman 10
Krasnec (unpublished, 1979)
Rowland (unpublished, 1979)
Neely and Plonka (1978)
9-12 (est.)
~6 (est.T
,b
(1977)
Estimated by Singh.
Same data as used by Lovelock (1977).
MCF Budget and the Hydroxyl Radical
Inert species such as fluorocarbons and SF are tracers that identify and
quantify the global circulation, whereas the distribution of the more reactive
halocarbons such as MCF offers a unique means for quantifying the role of OH,
Central atmospheric species that cleanses the atmosphere of impurities
7977a; 1977b).
4-10
-------
Two points are important here: (1) a residence time of 6 to 12 yr for
MCF implies a seasonally averaged tropospheric OH abundance of 3 to 6 x 10
molecules/cm , which is significantly lower than the values estimated from
models; and (2) from an analysis of MCF data, the OH distribution in the two
hemispheres appears to be asymmetric.
When used in a two-box model together with available emissions data
(Singh Ib77b), the hemispheric distribution indicates a higher average OH
concentration in the SH than in the NH ((OH) /(OH) > 1.5). This asymmetric
SH NH
OH distribution can be attributed to carbon monoxide (CO), which is an impor-
tant atmospheric sink for OH and is three times more abundant in the NH than
in the SH. Should the excess CO in the NH be from man-made sources (as is
currently believed), additional future depletion of OH in the NH can be
expected. The depletion of OH would reduce the scavenging ability of the
atmosphere, allowing increasing amounts of pollutants to enter the strato-
sphere. In addition, the tropospheric reservoir of many natural and man-made
species would increase. Two-dimensional global models are required to simu-
late more precisely the global distribution of MCF and OH.
CURRENT STATUS OF HALOGENATED TRACE CONSTITUENTS
Highlights in the present understanding of halocarbons and their atmo-
spheric effects are given below:
(1) Apparently no significant tropospheric sinks exist for fully halo-
genated fluorocarbons. The atmospheric burden of these species is increasing
at a rate proportional to the emissions burden.
(2) Global distributions of FC-11 and FC-12 suggest a 1.2-yr inter-
hemispheric exchange rate.
(3) Less doubt exists now than ever before that CC1. is essentially man-
made, but all anthropogenic sources have not been characterized.
4-11
-------
(4) The oceans may provide a sink for CC1. that is about half as effec-
tive as the stratospheric sink.
(5) The maximum Cl atom in the stratosphere is <3 ppb, of which ~75
percent appears to come from man-made sources.
(6) MCF appears to have an atmospheric residence time of 6-12 yr and may
be a potential depletor of stratospheric O . About 12 to 25 percent of all
MCF released to ground level is expected to enter the stratosphere.
(7) Residence time information on MCF can be used to suggest seasonally-
averaged OH concentrations of 3 to 6 x 10 molecules/cm . Thus, removal of
all species by reaction with OH may be significantly slower than predicted by
models.
(8) The oceans appear to be a dominant source of CH Cl. Other sources
of CH Cl/ man-made as well as natural, also exist. The oceanic data show
great variability in CH Cl distribution. The oceans remain poorly charac-
J
terized as sources or sinks of halocarbons.
REFERENCES
Berg, W. W., and J. W. Winchester, 1976. In Proceedings of the Nonurban
Tropospheric Composition, Hollywood, Florida, Nov. 10-12.
Chang, J. S., and J. E. Penner. 1978. Analysis of global budgets of halocarbons.
Atmos. Environ. 12:1867-1873.
Crutzen, P. J., and J. Fishman. 1977. Average concentration of OH in the
troposphere, and the budgets of CH , CO, H and CH CC1 . Geophys. Res.
Letters 4:321-324.
Dilling, W. L. 1977. Interphase transfer processes. II. Evaporation rates
of chloro methanes, ethanes, ethylenes, propanes, and propylenes from
dilute aqueous solutions. Comparisons with theoretical predictions.
Environ. Sci. Tech. 11:405-411.
Junge, C. E. 1976. The role of the oceans as a sink for chlorofluoromethanes
and similar compounds. Z. Naturforsch. Teil A. 31:482-487.
4-12
-------
Lovelock, J. E. 1977. Methyl chloroform in the troposphere as an indicator
of OH radical abundance. Nature 267:32.
Manufacturing Chemists Association. 1978. World production and release of
chlorofluorocarbons 11 and 12 through 1977. duPont de Nemours and
Company, Wilmington, Delaware.
McConnell, J. C., and H. I. Schiff. 1978. Methyl chloroform: impact on
stratospheric ozone. Science 199:174-177.
National Academy of Sciences. 1976. Halocarbons: Effects on Stratospheric
Ozone. National Academy of Sciences, Washington, D. C.
Neely, W. B., and G. H. Plonka. 1978. Estimation of time-averaged hydroxyl
radical concentration in the troposphere. Environ. Sci. Tech. 12:317-321.
Rasmussen, R. A., S. A. Penkett, and N. Prosser. 1979. Measurement of carbon
tetrafluoride in the atmosphere. Nature 277:549-551.
Singh, H. B. 1977a. Atmospheric halocarbons: evidence in favor of reduced
average hydroxyl radical concentration in the troposphere. Geophys.
Res. Letters 4:101-104.
Singh, H. B. 1977b. Preliminary estimation of average tropospheric HO con-
centrations in the Northern and Southern Hemispheres. Geophys. Res.
Letters 4:453-457.
Singh, H. B., D. P. Fowler, and T. O. Peyton. 1976. Atmospheric carbon
tetrachloride: another man-made pollutant. Science 192:1231-1234.
Singh, H. B., L. J. Salas, H. Shigeishi, and E. Scribner. 1978a. Global
distribution of selected halocarbons, hydrocarbons, SF and NO. EPA-600/
3-78-100, U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina.
Singh, H. B., L. J. Salas, H. Shigeishi, and A. H. Smith. 1978b. Fate of
halogenated compounds in the atmosphere. EPA-600/3-78-017, U.S. Environ-
mental Protection Agency, Research Triangle Park, North Carolina.
Singh, H. B., L. J. Salas, H. Shigeishi, and E. Scribner. 1979. Atmospheric
halocarbons, hydrocarbons, and sulfur hexafluoride: global distributions,
sources, and sinks. Science 203:899-903.
DISCUSSION
DP. Rasmussen: One difficulty in obtaining these gradients and interhemi-
spheric ratios, I feel, involves the number of data points and the number of
times or the different times during the year that samples were taken or measure-
ments made. There's also the problem of intercalibration. You've heard of
4-13
-------
Joe Krasnec's MCF ratio that showed a difference crossing the intertropical
convergence zone (ITCZ). And the ratio from data points collected in the NH
and the SH on just that one flight across the Pacific was on the order of
~1.28.
Dz>. Singh: He is not here, so I will speak on his behalf. His ratio is 1.21,
and it's not all that different. I recently talked to him, and this is his
final number to be published.
Dr. Rasmussen: His ratio is changed since we last compared data and observed
a very small difference. We will both be publishing our results presented at
the AGU meeting in December. And sometimes discrepancies are seen on the
same flight, and these must be resolved. Another concern is that, to charac-
terize a hemisphere, we must consider the total number of data points nec-
essary. The SH can be characterized reasonably well with a few data points,
because homogeneity is good beyond 30° S. In the NH, characterization is
difficult from 20° to 60° N, because we found again and again, as did other
investigators, considerable variability in MCF. So reconciling differences
between laboratories or data presentations, even at this meeting, will be
exceedingly difficult until the individual trip reports and data sets are
examined. Only then will inspection reveal which data points will fit the
modeling objectives.
Dr. Singh: Based on available information, it is not possible to determine the
quantity of data required. At this point, it is the quality that is most
critical. Typically, the more reactive a given species, the greater need for
extensive temporal and spatial data coverage. For MCF, if the residence time
is indeed of the order of one decade, the amount of data needed to characterize
its atmospheric budget and distribution should not be too terribly large. It
is pertinent to add here that our observations are based on our interpretation
of the best available atmospheric and emission data. There is little doubt
in my mind that the hypothesis proposed here will be better understood and
quantified as additional high quality data become available.
Question from Audience: You said, essentially, your global distribution was
calculated and integrated without weighting because of insufficient data?
Dr. Singh: No, the atmospheric mass within 10° latitudinal belts was considered.
The weighted average concentrations are not merely the averages of measured
concentrations. Where data did not exist, they were interpolated.
Dr. Rowland: In the laboratory, a contamination-free measurement was found
difficult to achieve for certain compounds, and MCF is one of them. So we
tried to measure MCF alone and specifically, resulting in several data points.
If no other measurements present problems, one arrives at the same number. In
fact, deriving a different number would give reason to doubt the numbers col-
lected, 8 or 10 molecules at a time. FC-21, for instance, raises considerable
suspicion, because a Teflon contamination appears to exist. So that's the
reason we are attempting to measure MCF differently.
Dr. Singh: Right now we must attribute the 20 ppt difference between our north
temperate zone MCF concentrations and those of Dr. Rowland to absolute calibra-
tion differences. The important observation is that even Dr. Rowland's data,
4-14
-------
which are some 20 percent lower than ours, point to a ~6-yr MCF residence
time. This is a factor of 4 or more higher than the 1.4-yr residence time
recommended by NAS. The current consensus on MCF residence time would be 6 to
12 yr, allowing 12 to 25 percent of all MCF released at ground level to enter
the stratosphere.
Dr. Crutzen: I can give you data I obtained yesterday: In the SH marine air,
89 ppt was the average number; in the NH marine air, 96 ppt; and in the NH
continental air, 121 ppt. These are probably arithmetic averages of all the
numbers, bot I don't have the refinement which Dr. Rasmussen just indicated.
4-15
-------
STRATOSPHERIC IMPACT RESEARCH AND ASSESSMENT
Alphonse F. Forziati
U.S. Environmental Protection Agency
Washington, D. C.
INTRODUCTION
The Clean Air Act Amendments of 1977, which are a part of Public Law 95-
95, require the U.S. Environmental Protection Agency (EPA) to conduct research
on substances, practices, and activities that affect the stratosphere, espe-
cially the ozone (0_) layer. This research includes physical, chemical,
atmospheric, and biomedical studies to ascertain the causes and effects of
stratospheric change. It also includes methods to recover and recycle, pre-
vent the escape of, and find substitutes for materials that bring about 0
depletion. Additional research in specific areas is assigned to other Federal
agencies, but EPA is required to contract with the National Academy of Sciences
(NAS) to study and report on the state of know!
health, biological, and socioeconomic effects.
(NAS) to study and report on the state of knowledge of 0_. depletion, including
EPA must also study and report on methods to control emissions. In light
of requirements to establish a coordinating committee, EPA has created the
Interagency Committee on Stratospheric Ozone Protection (ICSOP). EPA has also
contracted with NAS for support of two committees. One committee, the Com-
mittee on Impacts of Stratospheric Change (CISC), will report on the state of
knowledge regarding O^ depletion. The other, the Committee on Alternatives
for the Reduction of Chlorofluoromethanes (CARCE), will report on alternatives
for control of emissions and the socioeconomic impacts of control/noncontrol.
EPA expects a preliminary report from CISC in the summer of 1979 and a report
5-1
-------
from CARCE in September. A comprehensive report, anticipated for the end of
1979, will be assimilated into the required biyearly EPA report to Congress
due in 1980.
Nine governmental agencies, including EPA, transmit reports of their re-
search to ICSOP. The Committee digests this information and prepares a report
to EPA's Office of Research and Development (ORD). The Departments of Com-
merce and Labor report either to ORD or to the EPA Office of Program Manage-
ment (OPM). Regardless, the information is exchanged between the two offices
and then forwarded to the Office of Toxic Substances (OTS), where a decision
of whether or not to regulate is made. All of this exchange takes place in
complete coordination with the EPA Administrator's office and is reported to
Congress as indicated above (see Figure 5-1).
EPA RESEARCH
Figure 5-2 outlines EPA's major research needs and uncertainties. The
first requirement is to know the magnitude and rate of O~ depletion, which
involves study of models and monitoring trends. This work is performed pri-
marily by the National Aeronautics and Space Administration (NASA), the Federal
Aviation Administration (FAA), the National Oceanic and Atmospheric Adminis-
tration (NOAA), and the Department of Energy (DOE) through the Lawrence Liver-
more Laboratory. The program supports efforts to model climatic effects only
in a token manner; EPA is primarily concerned with biological and human effects.
But EPA partially supports other areas, as will be discussed later.
EPA's next concern is the change in ultraviolet (UV) radiation associated
with the change in O, concentration. EPA supports this work primarily through
NOAA. In the next area, climatic effects, EPA has recently taken an interest
in studying the effects of increased UV radiation on smog, and is considering
a small grant to the University of Florida to research this problem.
EPA's interest in biological effects includes crop yields, photosynthesis,
growth inhibition and pathological effects on plants, and effects on animals
and marine organisms. UV radiation is generally assumed to be unable to
5-2
-------
tn
U)
EPA
NAS
NOAA
NASA
NSF
HEW
FAA
DOE
USDA
CONGRESS
EPA
ICSOP
ORD
OPM
DOC
DOL
OTS
REGULATE & CONTROL
Figure 5-1. Information and decision chart.
-------
NEED
WHO SUPPLIES
1. HOW MUCH OZONE DEPLETION, HOW FAST ?
MODELS
MONITORING
TRENDS
2. AUVBVS. A03
3. CLIMATIC EFFECTS
GLOBAL SURFACE TEMPERATURE
REGIONAL SURFACE TEMPERATURE AND PRECIPITATION
STORM TRACKS
INCREASED URBAN SMOG
4. BIOLOGICAL EFFECTS
CROP YIELD
PHOTOSYNTHESIS INHIBITION
PHOTOMORPHOGENETIC AND PATHOLOGICAL EFFECTS
ANIMAL EFFECTS
MARINE ORGANISMS (PHYTO- AND ZOOPLANKTON,
LARVAE OF CRUSTACEA AND FISH)
NASA, FAA,
NOAA, DOE (LLL)
NOAA, EPA
NOAA, DOE,
NSF, EPA, FAA
USDA, NSF,
NOAA, EPA
5. HEALTH EFFECTS
UVB AND ERYTHEMA
UVB AND NON-MELANOMA SKIN CANCER
UVB AND MELANOMA (MALIGNANT) SKIN CANCER
6. INTEGRATED ASSESSMENT AND CONTROL STRATEGIES
RISK ANALYSIS
BENEFIT ANALYSIS
CONTROL ANALYSIS (PRODUCTION BAN, EMISSION BAN,
RECOVERY, SUBSTITUTES, INTERNATIONAL CONTROLS,
COST INCENTIVES)
NCI
EPA, DOC
DOL
Figure 5-2. Major data needs and uncertainties.
5-4
-------
penetrate water to any significant extent; therefore, no marine effects should
occur from radiation. But they do occur, revealing that UV penetrates water
more than previously believed. Consequently, many aquatic organisms that
spend a significant amount of time in the uppermost layer of water bodies are
affected.
Turning to human health effects, the degree of a sunburn is directly
correlated with increased exposure to UV radiation a relationship that is
very well established. But the correlations involved with incidence of plain
skin cancer (i.e., nonmelanoma skin cancer) or melanoma skin cancer are less
certain, as are factors that are distinct between the two cancers. EPA is
striving to clarify these issues, because melanoma skin cancer is rapidly
increasing in the United States.
With reference to Item 6 of Figure 5-2, EPA is the only agency that is
specifically required to make a complete, integrated assessment of all aspects
of O depletion and to report this assessment to Congress. This task includes
risk analysis, benefit analysis, and control analysis. As previously indi-
cated, all these areas are being researched for EPA by CARCE.
Prior to 1977, EPA derived its authority to control chlorofluorocarbons
(CFC's) from the Toxic Substances Control Act (TSCA). The rationale was that,
while CFC's may not in themselves be toxic, they lead to stratospheric 0
depletion, increased amounts of UV radiation falling on the earth's surface,
and, in turn, an increase in skin cancer.
BIOLOGICAL AND CLIMATIC EFFECTS RESEARCH PROGRAM
In Fiscal Year 1976 (FY-76) and FY-77, EPA initiated a short-term $4,000,000
Biological and Climatic Effects Research (EAGER) Program to provide the sci-
entific information needed for a decision on whether or not to regulate non-
essential uses of CFC's (e.g., as aerosol propellants).
In 1978, because of a number of serious budget cuts, the total funding
available from various sources was ~$1,000,000. In the present FY-79, the
5-5
-------
budget is ~$1,170,000; perhaps $1,400,000 will be allotted in FY-80. Due to
this low level of funding, all areas and projects cannot be maintained at an
optimum level.
Table 5-1 shows the categorical expenditures of BACER for PY-76 and FY-
77; Figure 5-3 shows the agencies that performed the research. The U.S.
Department of Agriculture (USDA) received nearly $1,000,000: ~$750,000 to
study the effects of O depletion on plants, a much smaller amount to study
effects on animals, and $150,000 for instrumentation to measure UV radiation.
NOAA received ~$700,000; NASA, less than $50,000. Within a budget of ~$700,000,
the National Cancer Institute (NCI) performed skin cancer surveys and studied
UV radiation exposure as a function of life-style. EPA is trying to reduce
uncertainty in estimates of exposure to UV radiation by developing a minia-
turized, personal dosimeter that a volunteer can wear. This electronic device
will have the ability to read while retaining its memory, to permit weekly and
monthly readouts of the wearer's actual exposure to UV-B.
TABLE 5-1. BACER EXPENDITURES BY CATEGORY FOR FY-76 AND FY-77
Category Allocation
(dollars)
Program Management 450,000
Data Analysis, Policy Analysis, and
Report Preparation 537,000
Climate and UV Monitoring 438,000
Terrestrial Ecosystem Effects 800,000
Aquatics Ecosystem Effects 445,000
Human Health Effects 740,000
Economics Analysis 200,000
Instrumentation 390,000
4,000,000
5-6
-------
L-9
C
i-S
(D
Ln
I
Ul
O
M
S)
cr
(D
rt
(D
3
O
Hi
O
-» N>
S S
ANIMALS INSTRUMENTATION
UV-B EFFECTS ON PLANTS
UV-B EFFECTS ON
AQUATIC ORGANISMS
CLIMATE MODELING
z
>
(0
^UV-B SENSITIVITY OF SELECTED AC
AQUATIC AND TERRESTRIAL SYSTEMS
O
>
MELANOMA
INCIDENCE STUDY
SKIN CANCER/UV-B
SURVEY
SKIN CANCER/LIFE STYLE
SURVEY
O
z
INSTRUMENTATION
CLIMATE WORKSHOP
DEFINITION OF RESEARCH NEEDS RE. CLIMATE
MANAGEMENT ft MISC.
SALARIES & TRAVEL
PROGRAM WORKSHOP
TECHNICAL ASSISTANCE ECONOMICS SPECIAL
(CONTRACT) OF EFFECTS SMALL
ON HEALTH GRANTS
AND CLIMATE
(GRANT)
-------
The National Bureau of Standards (NBS) developed standard reference lamps
and procedures for calibration of instruments; the National Science Foundation
(NSF) managed a workshop and produced a report on climatic effects of 0^ deple-
tion.
The EPA expenditures of $1,245,000 covered several years of staff expenses,
plus a technical support contract with SRI International to help produce a re-
port to Congress which considered socioeconomic effects. Some special small
grants were also included in this amount.
In FY-78, funds were so low that most existing programs were kept alive
by token support. NOAA received $110,000 to study the penetration of UV-B
radiation into natural waters and consequent effects on aquatic organisms. Of
this amount, $39,000 was to study UV-B effects on eggs and larvae of anchovy;
anchovies are very important commercially as an additive to chicken feed, and
are necessary to the food chain of many larger marine organisms (fish). It
was only just possible to keep the Robertson-Berger Sunburn Meter Network
functional for $25,000, but support has been increased to $100,000 for FY-79.
Of $400,000 contributed to EPA by OTS, $275,000 was used in partial pay-
ment of a contract with NAS to support the two committees mentioned above.
OTS money originally intended for ICSOP and interdisciplinary workshop support
was diverted into a grant for the University of Maryland to research possi-
bilities of international controls, which are presently of great interest to
EPA. No standardization or calibration work was supported; however, the
University of Lowell received a grant of $50,000 for instrumentation to moni-
tor solar UV-B radiation.
NCI contributed a small amount of money for three studies: UV-B-induced
photooxidation in skin; UV-A and UV-B effects on skin of mice; and monitoring
of UV-B irradiance received by humans. In the second of these three studies,
previous exposure to UV-A (wavelengths between 320 and 400 nm) was found to
sensitize the skin and make it more sensitive to UV-B. This was an unexpected
finding; for many years, no danger at all had been associated with UV-A. The
results from this Emory University study will be published in the summer of
1979.
5-8
-------
STRATOSPHERIC IMPACT RESEARCH AND ASSESSMENT PROGRAM
In FY-79, the name of BACER was changed to Stratospheric Impact Research
and Assessment (SIRA), to better indicate the broadened responsibilities of
EPA under the Clean Air Act Amendments of 1977. An outline of SIRA is given
below.
To begin with, EPA is spending a small amount of money at the University
of Utah to study UV-B and UV-A radiation effects on selected plants. In the
future, funds will be allocated for examination of crop yield and photorepair,
and for a 3-year study of commercially important crops using four plots under
very carefully controlled conditions.
A second research area involves a small allocation for an exploratory
study of UV-B effects on photosynthesis. As yet, no grant has been awarded
for this study, although EPA has contacted many of the nation's top experts.
To produce meaningful information will probably require a long-term research
program.
Two further areas of research are the previously mentioned studies of UV-
B effects on eggs and larvae of anchovy, and on zooplankton. Considerably
more work is needed to establish and quantify these effects.
Funding for the Robertson-Berger Sunburn Meter Network has been restored
to the $100,000 level. This new funding level will allow greater collection
and analysis of UV-B incidence data.
In the area of quality control, ~$125,000 is available for instrument
calibration purposes, and ~$100,000 is available for spectrally resolved
monitoring of UV-B incidence. The means by which this work will be accom-
plished are as yet undetermined.
Ongoing research strives to improve the correlation of skin cancer with
UV-B exposure, but funding is not available for all work areas. A solution to
limited funding may be to purchase 600 personal dosimeters (~$60,000) and to
5-9
-------
conduct a survey with 600 volunteers (~$30,000). An alternative being explored
is to purchase small quantities of the currently available UV-B-sensitive
strips (samples might be obtained for ~$5,000) for comparison in a laboratory
with the miniature dosimeters. The 6 prototype dosimeters that EPA will
receive under the present grant will also be compared with film strips, in the
field, by people wearing both types. It may be desirable to use the 6 proto-
type dosimeters as reference standards and proceed with film strips on 10,000
subjects. The cost difference in using 600 dosimeters or 10,000 sets of film
strips is minimal, so the method used will depend on the results of preliminary
studies.
The next area of research involves the relationship of UV-B to erythema,
melanoma, and nonmelanoma. This is, of course, a topic of great interest.
Though only small amounts of funds have been allocated, the studies are con-
tinuations of ongoing work. Dr. Elizabeth Scott of the University of Cali-
fornia at Berkeley is analyzing data collected by the National Center for
Health Statistics (NCHS) for three specific areas: (1) mortality by site of
lesion on the body, separately for melanoma and nonmelanoma, (2) cases of skin
cancer and of various keratoses obtained in the NCHS Health and Nutrition
Examination Survey, and (3) cases of skin cancer, separately for melanoma and
nonmelanoma, observed by a NCHS sample of discharges from short-stay hospital
care.
Using previously available data and unpublished data from NCHS tapes for
the period 1950-1974, Dr. John Lee of the University of Washington at Seattle
is developing age-specific mortality rates, by sex and race, for malignant
melanoma, other primary skin cancers, and total primary skin cancer. Through
these research programs, EPA hopes to create a model for forecasting increases
in melanoma and nonmelanoma cancers from UV-B enhancement.
As indicated earlier, EPA is the only agency designated to carry out a
complete, integrated assessment of the entire range of causes, effects, and
controls of O depletion. EPA already has funded a contract for undertaking
a complete integrated assessment, the product of which will be used to prepare
the required 1980 biennial report to Congress. A grant for socioeconomic
5-10
-------
analyses has produced state-of-the-art cost-benefit studies. In addition,
several areas where additional studies will reduce uncertainties have been
identified. These include the development of systematic measures of damage
that are compatible with measures of regulation cost. Econometric extensions
to epidemiological studies as well as improved sensitivity analyses of health,
ecological, and climatic damage estimates will also be undertaken. The global
nature of the 0 problem, and the difficulty of dealing with risk when the
possibility of irreversible damage is present, are recognized to require
particular conceptual and analytic attention.
The mandated NAS study of the state of knowledge and adequacy of research
on causes, effects, and controls of stratospheric O depletion is being
supported. NAS has scheduled delivery of its report for fall 1979. Funding
is again limited; to ease the financial burden, EPA has persuaded NAS to
receive the money in installments of $270,000 and $85,000.
In the final area, program management, costs have been cut to $125,000.
EFFECTS OF UV-B RADIATION
This report concludes with illustrations indicating the effects of UV-B.
Figure 5-4 shows a small instrument developed by USDA to measure the radiation
incident on plants. It can be substituted for one of the exposure pots. The
meter is coupled to the computer seen on the tray that will integrate the re-
quired spectrally-resolved data.
Figure 5-5 shows effects of UV-B on cucumber plants. The plant on the
left was protected from UV-B by Mylar, which passes virtually no UV radiation
in che UV-B region; the middle plant experienced a 50-percent increase over
the amount it would naturally encounter; and the plant on the right was sub-
jected to a 200-percent increase. It is very interesting that the 50-percent
increase in UV-B exposure had a significant effect on the middle plant's
growth, because a plant taken from the natural environment (i.e., the field)
and protected from solar radiation with Mylar will actually grow taller. UV-B
may thus act as a natural growth controller for plants; determining the extent
5-11
-------
Figure 5-4. USDA-developed instrument for measurement of radiation incident
on plants.
5-12
-------
Figure 5-5. Effects of UV-B on cucumber plants,
-------
of such an effect is crucial. Increases in UV-B exposure of >50 percent did
not seem to have too much additional effect? the plant on the right, which
received a 200-percent increase, actually appears slightly healthier than the
one exposed to a 50-percent increase. Of course, conclusions cannot be drawn
from merely one plant species. The particular species photographed may simply
be able to develop an effective repair mechanism. To obtain conclusive re-
sults, EPA plans to conduct a 3-year study using several carefully selected
plants.
Figure 5-6 is reproduced from an NAS report to show the correlation
between areas of UV-B exposure and sites of skin cancer on the human body. On
the male figure, the area the swimsuit would cover is virtually free from skin
cancer sites, whereas normally exposed areas are covered with such sites.
Upon close inspection of the female figure, a peculiar effect is noticeable:
the left leg has more skin cancer incidence than the right. The explanation
for this is unknown. In any case, changing life-styles are causing incidence
of skin cancer in "new" sites on the human body, further suggesting the need
for dosimeters to measure actual radiation at particular points of the body.
Figure 5-7 shows the rapidly rising trends of melanoma skin cancer. Are
such trends simply due to a change in life-style? One must consider that
changes in life-style can actually invert latitude correlations with skin
cancer, because life-style can be more important than where a person happens
to live. Generally, though, life-style is affected by latitude; i.e., if an
individual lives in the South, he is more likely to be outdoors.
More specifically, inhabitants of northern latitudes (e.g., Norway or
Sweden) that habitually vacation in the Riviera have been known to later
develop skin cancer, whereas their neighbors who stay at home do not. Such
"vacationers" have received heavy exposure to UV-B without allowing their
bodies to become accustomed; they become very sick. The skin burns and heals
very slowly, and eventually skin cancer may develop.
When attempting a correlation, then, a scientist must know where an
individual spends his time. A short, intense exposure to UV radiation is more
5-14
-------
Ul
I
Figure 5-6. Correlation in humans between areas of UV-B exposure and sites of skin cancer.
-------
1950
1955
1960
1965
197071 73
YEAR
*AGE-ADJUSTED (1960 US) RATE PER 100,000 POP.
Figure 5-7. Trends in mortality from melanoma of the skin in the U.S., 1950-1970.
-------
A TWO-DIMENSIONAL PHOTOCHEMICAL MODEL TO
ESTIMATE STRATOSPHERIC OZONE DEPLETION
Paul J. Crutzen
National Center for Atmospheric Research
Boulder, Colorado
INTRODUCTION
This paper presents some of the first results obtained using a two-
dimensional time-dependent photochemical model of the atmosphere, applicable
up to 55 km, and incorporating chloride chemistry. The methodology of the
model's development has been described previously (Crutzen 1976); the nu-
merical results presented here have been achieved through a refinement com-
bining the chemistry described in a one-dimensional model (Crutzen et al.
1978) and the transport parameterization utilized in the two-dimensional model
(Crutzen 1976). Although the results are preliminary, future reductions of
ozone (O ) predicted by this model are consistent with current predictions of
one-dimensional models.
RESULTS AND DISCUSSION
Figure 6-1 depicts the model-derived total O distribution as a function
of latitude and month. In general, the features agree with the observed total
O3 distribution (Diitsch 1971) shown in Figure 6-2. The major qualitative
discrepancy between Figures 6-1 and 6-2 occurs during the Southern Hemisphere
(SH) summer near the pole. Quantitatively, the O distribution analyzed by
Lovill et al. (1978) using June data from satellite observations yields an
average O3 column density of 296 Dobson Units (DU). The calculated June 1970
6-1
-------
Z -'9
H-
U3
^
a>
LATITUDE
0)
hi
(D
H-
3
0)
a.
o ro
a *
H-
ft
o
rt
PJ
o
Hi
H-
n>
M
CL
H
kD
^J
O
0
ri-
ft
(D
0)
c/i
0>
HI
o
rt
H-
O
o
Hi
3
rt
P)
3
Pi
rt
H-
rt
a
to
O
o
6
CD
m
10
_ m
po
oo
->i
p
iz
^
O
CD
CO
m
3)
m
o
(D
c_
0)
0>
ro
o
o
3
rt
O
d
-------
in
0)
Q
LJ
Q
Z>
h;
I-
12 I 2 3 4 5 6 7 8 9 10 II 12
Figure 6-2. Observed total 0 field plotted as a function of month and
latitude. Contours are in DU. Taken from Dutsch 1971.
6-3
-------
O, distribution yields an average value of 287 DU. The results shown in
Figure 6-1 are integrated by the model to 1970 after starting with initial-
guess fields in 1960. After 10 yr of numerical integration, the model should
have reached equilibrium and should not be affected by the initial fields.
The purpose here is to present the calculated O reduction to the year 2001.
Figure 6-3 shows the calculated amount of total 0 in the atmosphere
between 1970 and 2001. The average rate of O loss during this time is 0.24
percent/yr, yielding a total loss of 7.5 percent by 2001. The most recent
calculations, using a one-dimensional time-dependent model with comparable
chemistry, predict nearly a 5 percent decrease by 1988 (personal communica-
tion); the results of the present model similarly indicate a global O loss
rate of 5 percent by 1988. The calculations of the one-dimensional model
reach an equilibrium loss due to fluorocarbon release (at 1975 release rates)
of nearly 19 percent.
Figure 6-3 also depicts the estimated amount of 0_ depletion due to
release of methyl chloroform (1,1,1-trichloroethane, CH CC1 , MCF) in the
contrast between Curve A (continued release of fluorocarbon-11 (trichloro-
fluoromethane, CC1.F, FC-11) and fluorocarbon-12 (dichlorodifluoromethane,
CCl F , FC-12) at 1975 rates and MCF at 1978 rate) and Curve B (no release of
MCF after 1978). The future O loss due to MCF after 14 yr of integration
(1978-1992) is 0.9 percent, or 23 percent of the total O depletion computed
by the model during this period of integration.
Lastly, Figure 6-3 shows the rate of O depletion in the final year of
integration to be ~0.2 percent/yr. This value is somewhat less than the 0.3
percent/yr in the early 1980"s; however, the higher values in the early 1980's
may be artifacts of model initialization.
Using a two-dimensional model, it is possible to distinguish latitudes
and seasons of greatest 0 depletion. Although the 30-yr integration (1970
to 2000) yields an average O depletion of 0.24 percent/yr, examination of
Figure 6-4 indicates that this depletion rate is not constant with latitude.
More O is destroyed at high latitudes than in the tropics; the highest loss
6-4
-------
S-9
H-
i«
d
I-!
fD
0>
I
U)
M fD
kO X
-J 0
03 fD
(a
rt
rt
PJ
rt
3
O
0
| '
O
CO
cn
P-
cn
^j
fD
H!
3
P-
rt
rt
fD
CL
H-
3
rt
3-"
fD
O
PJ
O
d
i >
PJ
rt
H-
0
3
CO
d
CO
H-
3
O
d
H{
h{
fD
3
rt
Hi
fD
1 >
fD
PJ
cn
fD
Hj
ju
rt
fD
CO
Hi
O
HJ
rt
3"
fD
Hi
t->
d
0
h^
o
o
pj
Hf
tr
o
3
cn
PJ
3
3
O
I.
^)
j
0
rt
O
M
O
0
1 "
^^
I-1-
3
^j
fD
Hi
O
fD
3
rt
^«
0
pj
n
^
fD
^>
H-
CO
rt
fD
PJ
HI
fD ro rt H!
Hi n> 0
> a. g
GLOBALLY AVERAGED OZONE
COLUMN (Dobson Units)
CUMULATIVE %
DECREASE SINCE
1970
CD
ro
Ul
O
ro
O
ro
O
ro
00
0
IN) I
CD ->J
O ui
yy
1
b
i
ro
Ul
o
(D
-J
Ul
(D
CD
O
00
10
O
CD
CD
Ul
O
O
Ul
I
2 p
o c
* <
o
X
ex
O
O
G>
O
-»»
S"
CD
00
JL
CD
-------
9-9
0)
(^
LATITUDE
s ?
CD 3
rt
(D
3
rt )-
\ &)
<: rt
H H-
rt
c
o.
(D
O.
H-
to
ft
K
H-
tr
e
rt
H-
O
3
O
i-h
(D
h!
OJ
ft
O
ft
O
in
w
cr
CD
ft
3
(D
0)
3
O
QJ
O
O
O
00
C7I
I
01
tn
I
ro
tn
ro
01
en
o>
ro
CJI
H
CD
ro
H
O)
o
~n
o
O
O
ffl
3J
m
o
CD
rn
p
po
a
rt
ui
-------
rate is seen during the summer season at north polar latitudes. Similarly, an
enhancement of the O loss rate is seen at SH middle and high latitudes during
the summer season.
Figure 6-5 reveals that the rate of O loss varies with altitude and
latitude. The highest loss rate occurs in the tropics near an altitude of 40
km. The cumulative loss rate at this level is 25 percent (0.8 percent/yr).
This result is consistent with that from a one-dimensional model (Crutzen et
al. 1978), where the greatest loss rate centered near an altitude of 40 km.
This finding appears to contradict Figure 6-4, which suggests that less O
depletion occurs in the tropics. Figure 6-6 helps to clarify these seemingly
contradictory statements.
Figure 6-6 is the winter O concentration calculated by the model. The
contours, in units of molecules/cm , illustrate that the O, concentration is
11 3
only ~5 x 10 mol/cm in the region having the highest percentage of 0-. deple-
tion. This concentration is about a factor of 10 lower than the regions of
maximum C>3 concentration: 20 to 25 km at middle and high latitudes. Thus, a
high depletion rate at an altitude of 40 km does not cause as strong a deple-
tion in the total O,. column as a more moderate depletion rate in regions where
O concentrations are significantly higher.
AREAS FOR IMPROVEMENT
Ongoing refinements to the model will permit a better representation of
the distribution of trace gases. Major deficiencies of the current model
include the calculation of too small O concentrations in the tropical tropo-
sphere and too large interhemispheric gradients for the chlorocarbon com-
pounds. Both findings suggest that horizontal transport processes between the
hemispheres are too small. The fact that calculated O and nitrogen oxides
(NO ) concentrations in the tropical troposphere are smaller than observed
X
values likewise suggests that an important NO source term is missing in this
region.
6-7
-------
10.0
MB
00
100
1000 E
-85 -65 -45 -25 -5 5 25
SOUTH LATITUDE
45 65 85
NORTH
Figure 6-5. Latitude and altitude distribution of cumulative O^ loss after 10 yr of model integration.
Units are dimensionless and represent the fraction of 0,. lost at a given location (e.g.,
-0.20 refers to the contour where 2 percent of O has been depleted).
-------
0.05E + I3
-85 -65 -45
SOUTH
25 -5 5 25
LATITUDE
65 85
NORTH
Figure 6-6. Model-derived distribution of O for December 1970. Units are molecules/cm"
-------
REFERENCES
Crutzen, P. J. 1976. A two-dimensional photochemical model of the atmosphere
below 55 km: Estimates of natural and man-caused ozone perturbations due
to NO . In_ Proceedings of the 4th CIAP Conference, DOT-TSC-OST-75-38,
U.S. Department of Transportation, Washington, D. C. pp. 264-279.
Crutzen, P. J., I. S. A. Isaksen, and J. R. McAfee. 1978. The impact of the
chlorocarbon industry on the ozone layer. J. Geophys. Res. 83:345-363.
Dutsch, H. V. 1971. Photochemistry of atmospheric ozone. Adv. Geophys.
15:219-322.
Lovill, J. E., T. J. Sullivan, R. L. Weichel, J. S. Ellis, J. G. Huevel,
J. A. Korver, P. P. Weidhaas, and F. A. Phelps. 1978. Total ozone
retrieval from satellite multichannel filter radiometer measurements.
Report UCRL-52473, Lawrence Livermore Laboratory, Livermore, California.
97 pp.
DISCUSSION
DP. Hanst: Someone mentioned a 3 ppb figure for tropospheric Cl, accounting
for all halocarbon contributions; also, significant chlorine monoxide radical
(CIO) and hydrochloric acid (HC1) concentrations occur in the stratosphere.
Has total Cl as a function of altitude been plotted, and are there discon-
tinuities?
Dr. Crutzen: In the one-dimensional model, there are never any discontinuities.
Obviously, a total stratospheric Cl measurement is also needed. Walter Berg,
who is with my group, has done this; the measurements were made at 20 to 25
km. The data are being analyzed, and will be reported in the near future.
Voice from Audience: Do you have a number from your model for the strato-
spheric 0, exchange or O flux from the stratosphere into the tropozone?
Dr. Crutzen: No, I don't have a number; it is calculated and in my data
sheets. The number is ~5 x 10 molecules/cm /s, close to the destruction
rate at the ground. It's quite mysterious, because I believe a great amount
of 0-. is created and destroyed in the troposphere. Maybe all these inter-
actions produce a balancing effect. Actually, it is difficult to make these
estimates because of the tropospheric nitric oxide (NO) uncertainties.
Dr. Schiff: Your current modeling figure is an average 0.5 percent/yr deple-
tion [a better answer is supplied in the preceding report]. Does that mean
the amount of O., loss due to the Cl from FC-11, FC-12, and MCF is 0.5 percent?
Dr. Crutzen: It approaches 0.5 percent, yes.
6-10
-------
Dr. SohLff: A second question concerns tropical O . In your model, when you
are at 0° latitude, what do you use for NO? What is your boundary condition?
Dr. Crutzen: For tropical O-, at low altitudes, I consider only one input of
NO, the industrial source at mid-latitudes. The NO diffuses, nitric acid (HNO,)
is produced, and it is rained out with a certain scheme. According to this
pattern, har*dly any NO reaches the tropics; however, additional possible
sources for NO exist in the tropics. One of them is lightning. We estimate
between 0 and 30 megatons/yr, so that is a major problem in the tropical
regions. But I am quite certain that local 0., formation in the tropics is
taking place. It's a slow process, but it is just necessary to maintain the
0.. at average reported levels.
Voice from Audience: From that model, do you have an integrated estimate of
the O lost from 1950 to 1977?
Dr. Crutzen: The number is around 1.5 or 2 percent, but I must refer to the
data.
Voice from Audience: I think the discontinuity is repeated. You are quoting
one-dimensional models for the past to the present and two-dimensional models
for future estimates.
Dr. Crutzen: I'm speaking from memory regarding the one-dimensional model,
but I'm using the good reasonable agreement between results of the two-
dimensional and one-dimensional models.
6-11
-------
A REVIEW OF TECHNICAL PROGRAMS
OF THE MANUFACTURING CHEMISTS ASSOCIATION
RELATED TO STRATOSPHERIC CHEMISTRY OF CHLORINE
Frank A. Bower
E. I. du Pont de Nemours & Company
Wilmington, Delaware
INTRODUCTION
Calculations of stratospheric ozone (O.,) depletion by chlorine (Cl) rest
on a number of assumptions, most of which are subject to direct test. A final
conclusion based on such calculations can also be tested by direct measurement
in the stratosphere. Three of the basic assumptions are:
(1) Chlorofluorocarbons (CFC's) are not destroyed in the lower atmo-
sphere, but are transported quantitatively to the stratosphere.
(2) All stratospheric processes involving Cl are known.
(3) Reaction rates under stratospheric conditions are known with
sufficient accuracy to make reliable predictions.
This paper discusses measurement programs directed to clarifying these assump-
tions and examines O trend analysis, which constitutes a direct test of
predicted O depletion.
ATMOSPHERIC LIFETIME
A relevant question is whether significant removal processes exist in the
troposphere for halogenated organic compounds. Techniques developed to
identify any occurring sink mechanisms are directly applicable to determining
the lifetime of a species for which accurate release data can be developed.
7-1
-------
Most atmospheric model studies assume that chlorofluoromethanes (CFM's)
are transported quantitatively from the surface to the stratosphere, or (equiv-
alently) that tropospheric lifetime is very long. A tropospheric lifetime of
100 or 300 yr is usually assumed.
The relative importance of tropospheric sinks reducing the amounts of
Cl transported to the stratosphere was evaluated in 1976 by the National
Academy of Sciences (NAS). Removal by oceans was determined to be the only
potentially significant sink. The existence of undiscovered sinks and the
quantification of all sinks are of considerable importance, since tropospheric
lifetime depends on the additive effect of many removal processes.
Recently, fully halogenated compounds carbon tetrachloride (CC1.),
fluorocarbon-11 (trichlorofluoromethane, CC1 F, FC-11), and fluorocarbon-12
(dichlorodifluoromethane, CCl F , FC-12) have been shown to undergo a heter-
ogeneous reaction on certain mineral dust surfaces. Several investigators
have observed FC-12 during analysis of tropospheric air, but since its in-
dustrial production is very small and its tropospheric lifetime should be
relatively short, any concentration should fall well below measurable levels.
If the occurrence of FC-12 is firmly established, the most probable source is
conversion of FC-11. Reduction of FC-11 is known to occur in biological
systems (Cox et al. 1976; Wolf et al. 1975) and in such applications as re-
frigeration and foam-blowing agents.
Early attempts to identify possible removal mechanisms for CFM's were
based on calculations of global burdens from a few localized measurements in
the lower atmosphere and on approximations of fluorocarbon release. Uncer-
tainties associated with these measurements, such as variability of the atmo-
*
sphere, inherent errors in any analytical method, and possible uncertainties
in calibration, are too large to inspire confidence in the final calculation
of the global burden. In fact, the quality of the measurements permits only a
conclusion that the tropospheric lifetimes of CFM's are between 10 and °°
yr. Work supported by the Manufacturing Chemists Association (MCA) Technical
Panel has removed many uncertainties connected with the analytical method and
with world production and release of fluorocarbons.
7-2
-------
Rigorous procedures for analysis of FC-11 and FC-12 have reduced ana-
lytical errors to <2 percent. Absolute standards can now be made for cali-
brating the chromatographs. With these standards, Lovelock (private com-
munication to MCA 1978) has shown that absolute coulometry, properly applied,
is valid within ~7 percent. But the persistent uncertainties in determining
global burden due to natural variability of the atmosphere can be resolved by
an alternate approach.
The Technical Panel is supporting a measurement study, the Atmospheric
Lifetime Experiment, to determine directly the lifetime of CFM's in the atmo-
sphere. A method developed by Cunnold et al. (1978) involves global-scale
determination of tropospheric concentration trends of CFM's by frequent mea-
surement at stations strategically located worldwide. Stations located at
Adrigole (Ireland), Barbados (off the northeast coast of South America),
American Samoa (in the mid-Pacific), and Cape Grim, Tasmania (off the south-
east coast of Australia) make hourly measurements of FC-11 and FC-12, nitrous
oxide (NO), CC1 , and methyl chloroform (1,1,1-trichloroethane, CH CC1 ,
MCF). Measurements of other species are made when the investigators visit the
stations for routine servicing. After a global trend is established by mea-
surement, it is compared to the expected trend derived from release statistics
provided by member companies of the MCA Technical Panel. A statistically
significant difference in slope of the two trend lines would indicate the
existence of a removal mechanism operating in the atmosphere, or a natural
source of CFC's.
Cunnold et al. (1978) estimate that a 10-yr lifetime can be detected
within 3 yr of measurement, and that a 20-yr lifetime can be detected within
~5 yr. The stations have been in full operation since the second quarter of
1978, and the first year's data will soon be processed.
STRATOSPHERIC MEASUREMENT
Approximately 100 chemical reactions are significant to stratospheric
Cl chemistry. This discussion will concentrate on the catalytic O depletion
cycle:
7-3
-------
Cl + 0 -> CIO + O
J f->
CIO + O -> Cl + 0 (Eq. 1)
0 + °3 + 2°2
as postulated by Stolarski and Cicerone (1974), Cicerone (1974), and Wofsy and
McElroy (1974). The measurement of chlorine monoxide radical (CIO) and other
Cl species in the stratosphere was quickly recognized as one direct test of
the 0 depletion hypothesis.
In 1975, the Technical Panel supported the development of an analytical
method for ClO based on the following reaction (Stedman et al. 1975):
CIO + NO -* Cl + NO (Eq. 2)
After conversion of the ClO to Cl, the concentration of Cl atoms is determined
by resonance fluorescence. This method was applied by Anderson et al. (1977)
in a series of stratospheric probes. They observed that the concentration of
CIO in the mid-stratosphere follows an apparent seasonal pattern, in which ClO
ranges from a low winter value of <1 ppb to a high summer value of ~8 ppb. In
September 1978, Menzies (1979) observed ~2 ppb at sunset, which by model cal-
culation is equivalent to ~4 ppb at noon.
One interesting result of these measurements is that the amount of ClO
detected is much larger than the amount of total Cl calculated to be present
in the stratosphere. Also interesting are the normal O levels observed simul-
taneously with the highest levels of CIO. According to present calculations,
O should be dramatically decreased at such high levels of CIO. If the mea-
surements prove to be valid (no flaws have yet been found), one conclusion to
be drawn is that Cl does not deplete stratospheric O .
The Technical Panel is continuing its stratospheric measurement effort
to develop additional information for understanding more precisely the true
chemistry of Cl in the stratosphere. Balloon probes by Murcray (work in
7-4
-------
progress), Bonetti (work in progress), and Harries (work in progress) are at-
tempting to identify other Cl species that may help to explain the unpredicted
behavior of CIO in the stratosphere. Total Cl measurement is one of the most
important experiments remaining to be accomplished; the MCA Technical Panel is
preparing a flight probe for mid-1979, but the analytical procedures are so
difficult that predictions of success are not now possible. Balloon snapshot
experiment, most useful for producing simultaneous data on reactive species,
are complemented by long-term monitoring. In addition to the balloon probe
effort, MCA has established infrared monitoring capability for hydrochloric
acid (HC1) and hydrogen fluoride (HF) at Jungfraujoch in Switzerland, micro-
wave monitoring for CIO at the University of Massachusetts in Amherst, and
solar scanning at Mt. Evans in Colorado.
REACTION RATES
During the last few years, rate measurements of reactions of many trace
species have been refined. Perhaps the most surprising development in kinetics
is the discovery that HO reacts with nitric oxide (NO) 30 to 40 times faster
than originally believed. The observed rapid reaction between CIO and HO
boosts the possible importance of hypochlorous acid (HOC1) in stratospheric
chemistry.
The Technical Panel also supports work on determination of absorption
cross sections, particularly HOCl; branching ratios of reaction having alter-
nate pathways; and rates for reactions relevant to Cl chemistry. Evaluating
the pressure dependence of some of the key reactions is particularly important.
OZONE TREND ANALYSIS
Direct observation of the 0 layer will provide the final test of deple-
tion calculations. The Technical Panel has funded statistical analyses of
O, measurements gathered over the past years. The most powerful technique for
analysis of data collected over extended periods is a statistical procedure
known as time series analysis (Box and Jenkins 1970). Such analysis is capa-
ble of detecting abnormal trends that may exist in a long series of variable
7-5
-------
data. Normal cycles are identified by data analysis and then factored out of
the observations. The process is repeated until only random noise, as measured
by appropriate statistical tests, remains. Any trend is detected by an upward
or downward slope of the random noise line.
Time series analyses have been conducted on Dobson O data by Hill and
Sheldon (1975) and Pagano and Parzen (1975). Hill and Sheldon estimate that
their analysis of Dobson data would detect a potential change of 0.26 percent/
yr persisting for 6 yr (total 1.56 percent). Neither study has found a sta-
tistically significant trend during the period 1970 to 1975. This analysis
method promises to provide an early warning of any abnormal change in O, level.
Since one-dimensional models predict that CFC's have already depleted
0 by 1.5 to 2 percent, and that MCF has already depleted 0^ by ~0.3 to ~0.5
J -J
percent, a total of ~1.8 to ~2.5 percent O depletion should already have
occurred. If a trend of this magnitude exists, it should be detectable now,
and in 1 to 2 yr the calculated depletion should be well above the detection
limit.
Current model calculations indicate that the O layer is most sensitive
to perturbation by Cl in the 35 to 45 km region of the stratosphere. In this
region, chemical reactions are expected to dominate 0 concentration; in lower
levels, transport phenomena and perturbations from the troposphere are ex-
pected to be complicating factors. In the 35 to 45 km region, predictions
indicate that depletion by several percent should already exist. Angell and
Korshover (1978) have analyzed O-. data from this region and conclude that the
O, level over north temperate latitudes increased by perhaps 8 percent between
1962 and 1973.
Present analyses of O data yield no direct evidence for existence of
0 depletion at this time. But the Technical Panel is supporting additional
work in time series analysis of data gathered by the Nimbus satellite.
Additional O data from this source may improve sensitivity to trends.
7-6
-------
SUMMARY
The question of the existence of significant tropospheric sinks for
fluorocarbons is not yet answered, and reliable calculations of O depletion
cannot be made without these data. Since recent observations of CIO and O
in the stratosphere are not consistent with current models of stratospheric
chemistry, the resolution of this discrepancy is vital to deeper understanding
of the stratosphere. Continuing refinement of reaction rate data and photo-
dissociation cross sections is desirable, especially as new species are found
to be important in stratospheric chemistry.
Careful analysis of available O^ data does not show a detectable trend
in global 0 . Further analysis of existing data, particularly that from the
Nimbus satellite, is needed. Continued monitoring and prompt data analysis
are important for verifying predicted trends. Therefore, investigations that
will lead to better understanding of atmospheric chemistry must continue.
REFERENCES
Anderson, J. G., J. J. Margitan, and D. H. Stedman. 1977. Atomic chlorine
and the chlorine monoxide radical in the Stratosphere: Three in situ
observations. Science 198(4316):501-503.
Angell, J. K., and J. Korshover. 1978. Global ozone variations: An update
into 1976. Monthly Weather Review 106(5):725-737.
Box, G. E., and G. M. Jenkins. 1970. Time Series Analysis, Forecasting
and Control. Holden-Day, San Francisco.
Cicerone, R. J. 1974. Fluorocarbons impact on health and environment.
Hearings before the Subcommittee on Public Health and Environment of
the Committee on Interstate and Foreign Commerce, House of Representatives.
Serial No. 93-110. U.S. Government Printing Office, Washington, D. C.
(1975).
Cox, P. J., L. J. King, and D. V. Parke. 1976. The binding of trichloro-
fluoromethane and other haloalkanes to cytochrome P-450 under aerobic
and anaerobic conditions. Xenobiotica 6(6):363-375.
Cunnold, D., F. Alzea, and R. Prinn. 1978. Methodology for determining the
atmospheric lifetime of fluorocarbons. J. Geophys. Res. 83:5493-5500.
7-7
-------
Hill, W. J., and P. N. Sheldon. 1975. Statistical modeling of total ozone
measurements with an example using data from Arosa, Switzerland. Geophys.
Res. Letters 2 (12) :541-544.
Menzies, R. T. 1979. Remote measurement of CIO in the stratosphere. Geophys.
Res. Letters 6 (3) .-151-154.
National Academy of Sciences (Committee on the Impacts of Stratospheric Change).
1976. Halocarbons: Effects on Stratospheric Ozone. National Academy
of Sciences, Washington, D. C.
National Academy of Sciences (Committee on the Impacts of Stratospheric Change).
1977. Response to the Ozone Protection Sections of the Clean Air Act
Amendments of 1977. National Academy of Sciences, Washington, D. C.
Pagano, M., and E. Parzen. 1975. Technical Report No. 35. Statistical
Science Department, SUNY.
Stedman, D. H., J. G. Anderson, G. R. Carignan, and B. C. Kennedy. 1975. A
feasibility study of ClO detection. Final Report to Manufacturing
Chemists Association. University of Michigan, Ann Arbor.
Stolarski, R. S. , and R. J. Cicerone. 1974. Stratospheric chlorine. Possible
sink for ozone. Can. J. Chem. 52(8):1610-1615.
Wofsy, S. C., and M. B. McElroy. 1974. The HO , NO , and CIO . Their role
in atmospheric photochemistry. Can. J. Chem. 52(8):1582-1591.
Wolf, C. R. , L. J. King, and D. V. Parke. 1975. Anaerobic dechlorination of
trichlorofluoromethane by liver microsomal preparations in vitro. Biochem.
Soc. Trans. 3 (1):175-177.
DISCUSSION
Dr. Watson: I agree that some of the measurements are somewhat disturbing;
however, I think you very slightly misquote the numbers. At 38 km, that was,
indeed, 2 ppb of CIO. I feel the reliability is possibly plus or minus that.
The 40-km cutoff height varies ~10 percent, and not much chlorine nitrate
radical (C10NO ) exists at this level. At 30 km, the concentration may double.
But you remember Menzies1 published profile falls very, very sharply; by 30
km, it's actually well below Anderson's measurements. One must be very care-
ful in studying a diurnal variation on the ClO-to-ClONO2 process. Not much
effect is seen above 35 and 40 km, but a significant effect is seen below 30
km.
Voice ffom Aud-ienoe: In observing long-term variations in 0,., one must be
aware of possible changes with the solar cycle. Although Angell's correlation
of total O with the solar cycle is not perfect and does not explain the whole
variation, this correlation is one important factor mentioned here.
7-E
-------
Dr. Bower: The time series analysis, conducted over a time which is longer
than any cycle in question, will pick those up, and they can be factored out
of the data. Hill's analysis considers this quasibiennial cycle cited by
Angell, and Hill observes this approximate 11-yr cycle.
Voice from Audience: I thought you said over a 10-yr period. Right?
Dr. Bower1: No. The data extend from the late 1920's in the Arosa Station.
Hill examined only a recent 10-yr period for a trend. First, the entire mass
of data is analyzed to evaluate what cycles may exist; then a trend is sought,
in the Z-st 5, 10, however many years, after all these established "natural
cycles" are removed.
Dr. Singh: Do Jim Lovelock's data apply to perhaps PC-11 and CCl. and not to
any of the other species we have discussed?
Dr. Bower: Only certain of the data apply to PC-11. I'm not certain they
apply to the other species.
Dr. Rowland: Why do you say that if O depletion is now 1.8 or 2 percent it
is above the detectable limits?
Dr. Bower: Because Hill's estimate is that the detection limit is 1.56 per-
cent.
Dr. Rowland: That is valid if it occurred at 6 yr, rather than over 25 yr.
According to present models, we have not yet reached the 0.26 percent/yr.
Dr. Bower: Did Paul Crutzen not say 5 percent?
Dr. Rowland: That's his prediction for the future. I don't believe he's
going to get 2 percent of the present loss.
Dr. Bower: As I remember our calculations, we do indeed find that somewhere
in the vicinity of 2 percent is an expected current level.
Dr. Rowland: Yes, but that's not 0.2 percent that did not occur the last 6
yr. It is occurring over a period of 25 yr, so we are not yet at the measur-
able point. In one of the most recent models I've seen, we've not yet reached
0.26 percent/yr. Hill's analysis is therefore misleading, because you must
reach 2 percent and then tack 1.5 percent on top of that before you start see-
ing a detectable trend.
Dr. Bower: I don't believe that is quite right.
Dr. Rowland: We have also carried out a least squares analysis on the Arosa
data and the best fit to the Arosa data, where we factor out nothing, gives a
2 percent loss in O at the end of 1977. We can't analyze the relative prob-
abilities of 0 percent, 2 percent, and 4 percent; but it is parabolic around a
2 percent loss, and it's equally likely that we have no loss or the 4 percent.
7-9
-------
Dr. Bower1: I can't argue what you have found by a least squares analysis of
the data, but that seems like a rather elementary approach to analyzing this
complex set of data.
Dr. Rowland: I have a similar feeling about time series analyses of data
where you know a dirty spectrometer is used.
DT. Bower: I think we could probably argue this point until the end of the
day. We are funding a lot more work by the best statisticians in the country
to look at this problem. Two of them have already examined it and seem to
concur that the method is valid, and we will put yet a third on it, and I
think we will get the problem solved in due course.
Dp. Rowland: My statistician is currently beating the stock market at 20 per
cent/yr.
7-10
-------
MEASUREMENTS OF ATMOSPHERIC METHYL CHLOROFORM
BY WASHINGTON STATE UNIVERSITY
Dagmar R. Cronn
Washington State University
Pullman, Washington
INTRODUCTION
Several types of atmospheric measurements have been pursued by Washington
State University (WSU) in the last 4 years. This research has included mea-
surements into the lower stratosphere of various halogenated compounds, in-
cluding methyl chloroform (1,1,1-trichloroethane, CH CC1 , MCF). Information
on MCF vertical distribution, time trends, and latitudinal distribution has
been obtained.
VERTICAL DISTRIBUTION
To obtain MCF vertical profiles, several different sampling platforms
were used, beginning in 1976 with a Learjet which will reach 48000 ft (14.6
km). The first flight sequence was performed at ~47° N latitude in March
1976. Figure 8-1 shows these earliest MCF vertical profiles, which extend
past the tropopause zone into the lower stratosphere. The mixing ratio data
are plotted as a function of distance from the tropopause, which averaged
34500 ft (10.8 km). These data provided first proof that the models were
correct i.e., that this compound is transported at least into the lower
stratosphere. At that time (March 1976) and latitude (47° N), WSU was mea-
suring an average background tropospheric level of 95 ppt.
8-1
-------
KM FTXIO"3
+ 6-
+5-
+4-
+3-
+2
0-
-1-
-2-
-3-
-4-
-5-
-6-
+ 20-
+ 15-
+ 10-
+5-
0
-5-
-10-
-15-
(SOOml SAHPLE)
A A
* .1
A ABO
TROPOPAUSE
o ^v
"* \v
D
MARCH 6,1976 ° o
D MARCH 9, 1976 A
< MARCH 10, 1976 >
A MARCH 1 1,1976 * *
o MARCH 12, 1976
A MARCH Z2, 1976
MARCH 23, 1976
r a
.
40 5O 60 70 BO 9O 100 MO 120 1 3O
ppt
Figure 8-1. MCF mixing ratio distribution as a function of tropopause height,
March 1976, 47° N latitude. From Cronn et al. 1976.
In April 1977, another set of Learjet vertical profile flights was flown
off the California coast, west of San Francisco (~37° N). Figure 8-2 again
shows the presence of MCF above the tropopause zone, as well as the precipi-
tous decline in mixing ratio with ascent into the lower stratosphere. The
day-to-day behavior in the low stratosphere of MCF, along with other halo-
genated compounds and nitrous oxide (N_O), can often be explained by meteo-
^
rological considerations (Cronn et al. 1977a,b; Saunders et al. 1978). The
tropospheric level had climbed to 116 ppt by April 1977.
Subsequent measurements took place in the intertropical convergence zone
(ITCZ), where the tropopause is sufficiently higher than in the mid-latitudes
to justify the coupling of a U-2 with a Learjet, to extend the sampling plat-
form to 70000 ft (21.3 km). Figure 8-3 shows the vertical distribution ob-
tained near the Panama Canal Zone at ~9° N latitude in July 1977. The rate of
decrease in the lower stratosphere was less in the tropics than in the mid-
latitudes of the Northern Hemisphere (NH). This result is to be expected if
the tropics are an area of upward transport of tropospheric air into the
stratosphere.
8-2
-------
+ 6-
»5-
+ 4-
FT X KT3
20H
; 100 ml SAMPLE)
TROPORAUSE'
APRIL 22. 1977
* APRIL 25, 1977
a APRIL 26, 1977
° APRIL 27, 1377
0 APRIL 28, 1977
APRIL 29, 1977
1 I4KMS38KM
r~
SO
I
80
1
9O
1
IOO
1
no
PP»
Figure 8-2. MCF mixing ratio distribution as a function of tropopause height,
April 1977, 31° N latitude. From Cronn et al. 1977a.
TIME TRENDS
WSU also addressed time trend measurements of MCF. Documentation of the
increase in MCF enabled accommodations with emissions data and expected
sinks. Starting in July 1977, MCF data were collected at a ground station in
eastern Washington State at ~47° N latitude. Figure 8-4 shows the results of
that monitoring program as weekly averages of hourly measurements. The data
have not been screened for incursions of high mixing ratios due to transport
of air parcels with recent anthropogenic contacts. As reported earlier (Cronn
et al. 1978), increases of >12 percent/yr in the MCF mixing ratio have been
observed.
Figure 8-5 shows MCF measurements by various investigators in both the
NH (open symbols) and Southern Hemisphere (SH) (filled symbols). This plot is
similar to one reported by Neely and Plonka (1978). Data from Rowland (pri-
vate communication, 1978), Singh et al. (1979), and our own laboratory (Cronn
et al. 1976; Cronn et al. 1977a; Robinson 1978) were added to the Neely and
Plonka data. We compared the results of our MCF modeling efforts with this
data set. i
8-3
-------
Kl
21-
18-
15-
12-
uj 9-
o
b
6-
3-
* FT
70-
60-
50-
40-
30-
20-
10-
O
W IO~ »
DATE WHOLE AIR
7/18 o
7/19 A
OB o \fc a 7/23 V
7/25 a
£s> ^o A&.Q oOv 728 O
'
aA &
i i
V
>
t
7/29 *
\y © ^7
*a FOR LEARJET VALUES
40
60
80
100
120
140
160
CH3CCL3 MIXING RATIO, ppt
Figure 8-3. MCF mixing ratio distribution as a function of tropopause height,
July 1977, 9° N latitude. From Cronn and Robinson 1978.
8-4
-------
CO
o
in
CE
CC.
O
CONTINUOUS GROUND MONITORING, EHSTERN WHSHINGTON STflTE
WEEKLY HVERRGES, U7 DEGREES NORTH LflTITUDE
,
tt
JULY OCTOBER JANUARY APRIL JULY
1977 H 1978
OCTOBER JANUARY
H-I979-*-
Figure 8-4. Ground-level time trend measurements for MCF, June 1977 through January 1979, 47° N latitude.
-------
1000-
500-
ZOO-
100
DI
,
f
o. OU-
o
1
o «>
z
X
3
10-
f
O
lO
f
g 5-
2-
i.
NORTH
SOUTH
1 tt
^ I J .
j. CRONN
I LOVELOCK
fi RASMUSSEN
D ROBINSON
A ROWLAND
O SINGH
1970 1972
1974 1976
YEAR
1978 1980
Figure 8-5. Observed NH and SH mixing ratios for MCF, 1972 through 1978.
8-6
-------
LATITUDINAL DISTRIBUTION
The third type of atmospheric MCF measurement provided data on latitudinal
distribution. Figure 8-6 shows the tropospheric MCF mixing ratio as a func-
tion of latitude for samples collected and analyzed during the second half of
1978. The data are corrected to November 1978, assuming a 1 percent/month
rate of increase. Most of the data shown in Figure 8-6 were collected on
board a Navy C-130 aircraft used for air chemistry measurements as part of the
U.S. Antarctic Research Program. This aircraft collected data between ~35° N
and ~90° S latitude. The latitudinal distribution was extended a bit further
north via samples collected using the WSU Aero Commander. The latitudinal
gradient seen in Figure 8-6 supports the earlier observation of a gradient
obtained from comparison of the average of 97 ppt at 9° N (July 1977) with the
average of 115 ppt at 37° N (April 1977). This latitudinal gradient is similar
to those reported by other speakers at the Conference on Methyl Chloroform and
Other Halocarbon Pollutants.
For comparison with results from other laboratories, the WSU mixing ratio
for ground-level continental air at 47° N latitude in November 1978 is 131
ppt. This is the average over a 4-week period of hourly measurements from the
continuous monitoring site in eastern Washington State. Urban levels can, of
course, be much higher. For example, an MCF level of 1.1 ppb was measured at
2000 ft (610 m) over the Riverside, California airport in May 1976. Ground-
level measurements reached 5 ppb at Claremont, California in August 1978.
Often levels do not return to clean-air background values for days at a time.
CONCLUDING REMARKS
WSU has documented the distribution of atmospheric MCF as a function of
time, latitude, and altitude. Tying this data base to data on emissions and
the tropospheric sink due to the hydroxyl radical (OH) provides information on
MCF's trospheric lifetime which, in turn, indicates the effect of MCF on
stratospheric ozone (0 ) levels.
8-7
-------
120-
Q.
0.
-100-
2
co |
i rf
03 (£
z
X
60 -
<
. .+ + A . j. + *
ȣ*" **<** ** *
f*
+ -»». x*
+ +
x AEROCOMMANDER FLIGHT, SEPT,
+ C-130 FLIGHTS, NOV., 1978
AVERAGE OF HOURLY GROUND
MEASUREMENTS, OCT. -NOV., 1978
1978
A GROUND SAMPLES, APRIL a JULY, 1978
o GROUND SAMPLES, NOV. 1978
1 1 r ii i i i i
> -80 -60 -40 -20 0 20 40 60 80
LATITUDE
N
Figure 8-6. Latitudinal gradient of MCF corrected to November 1978.
-------
REFERENCES
Cronn, D. R. , and E. Robinson. 1978. Determination of trace gases in Learjet
and U-2 whole air samples collected during the Intertropical Convergence
Zone Study. Report to the National Aeronautics and Space Administration,
Washington, D. C., August.
Cronn, D. R., R. A. Rasmussen, and E. Robinson. 1976. Measurement of tropo-
spheric halocarbons by gas chromatography-mass spectrometry. Report to U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina.
Washington State University, Pullman, Washington, August.
Cronn, D. R., R. A. Rasmussen, and E. Robinson. 1977a. Measurement of tropo-
spheric halocarbons by gas chromatography-mass spectrometry. Report for
Phase II. Report to U.S. Environmental Protection Agency, Research Tri-
angle Park, North Carolina. Washington state University, Pullman, Wash-
ington. October.
Cronn, D. R., R. A. Rasmussen, E. Robinson, and D. E. Harsch. 1977b. Halo-
genated compound identification and measurement in the troposphere and
lower stratosphere. J. Geophys. Res. 82:5935-5944.
Cronn, D. R., D. E. Harsch, and E. Robinson. 1978. Tropospheric and lower
stratospheric profiles of halocarbons and related chemical species.
Presentation at the 176th National American Chemical Society Meeting,
Miami Beach, Florida, September 11-15.
Neely, W. B., and J. H. Plonka. 1978. Estimation of time-averaged hydroxyl
radical concentration in the troposphere. Environ. Sci. Tech. 12:317-321.
Robinson, E. 1978. Analysis of halocarbons in Antarctica. Report to National
Science Foundation, Washington, D. C. Washington State University,
Pullman, Washington, December.
Saunders, W. D., E. Robinson, D. R. Cronn, R. A. Rasmussen, and D. Pierotti.
1978. F-ll and NO in the North American t
sphere. Water, Air, Soil Poll. 10:421-439.
1978. F-ll and NO in the North American troposphere and lower strato-
Singh, H. B., L. J. Salas, H. Shigeishi, and E. Scribner. 1979. Atmospheric
halocarbons, hydrocarbons, and sulfur hexafluoride: global distributions,
sources, and sinks. Science 203:899-903.
DISCUSSION
Voice from Audience: Do you have measurements of perchloroethylene (C Cl )
in the SH? 2 4
8-9
-------
Ur>. Cponn: We should have some coming up, as well as analyses of the MCF data
from the SH. That is not a complete data study. It contains only 50 to 75
percent of the data points, and it's not all tabulated.
DP. Singh: Do you have any methyl chloride (CH Cl) data, and what do they
show?
DP. Cponn: I believe our CH Cl data are similar to yours; we have a larger
variability in our measurements, something on the order of 5 or 10 percent
standard deviation in our averages. Therefore, there is no significant
statistical difference between the two hemispheres.
DP. Hanst: You showed two graphs of MCF measurement. The second one had much
more scatter in the points than the first. Why was that?
DP. Cponn: I believe it's because of the sample size. The first graph was
based on a technique that has a smaller analytical variability.
DP. Rasrrtussen: To answer Dr. Singh's question on CH Cl, we've made fairly
extensive measurements of this species as well. We have summarized the 1977
and 1978 GAMETAG flight data. Statistically, there is no difference between
the hemispheres. The major differences are observed in the boundary layer in
the equatorial region. The CH,C1 was elevated in the boundary layer over the
ocean. Measurements at 40°, 35°, 50° N latitude showed fewer differences.
The other significant perturbations of CH-Cl were in the slash-burned areas
over East Africa. We also tried to document whether or not elevated CH,C1
levels occurred in a forest fire in the Pacific Northwest. Apparently, the
fire was too open it was a burning fire, not a smoldering fire. The results
were unequivocal. From 2 years' data (actually, now, for the third year
1979) we don't see any interhemispheric values of CH Cl.
UP. Cponn: I can comment on our data on CH Cl near the tropics as well. We
do have published data. And in our ITCZ data the CH Cl was very much elevated
in the boundary layer over the oceans relative to the concentrations higher
up.
DP. Rowland: What were the CH Cl results for the slash-burn?
DP. Rasmussen: The values at Kenya were typically up to 3 ppb in proximity of
the burn (a couple of hundred yards from very extensive burning of secondary
eucalyptus fires at ~8000 ft not Kenya proper). It was just a very wet
situation where there wasn't any open fire, just a pile of smoke.
In the southern part of Kenya, we found clean air levels of around 600 to
650 ppt; levels in a smoky environment might go up to 700 ppt. In the Samburu
area in the northern part of Kenya, where there's much less vegetation, CH Cl
levels did not differ discernibly from those obtained over the Atlantic Ocean
at 18000 ft.
Voice fpom Audience: But over Kenya, what's the fraction covered with smoke
at any given time?
8-10
-------
Dr. Rasmussen: Well, we didn't fly that close to the Ugandan border. We had
enough trouble with the Somalis. But there was a tremendous pall of smoke
from the western edge of Kenya all the way, practically, to Gabon on the
Atlantic.
Voice from Audience: High levels were measured in the smoke itself. Does the
high level fall off?
Dr. Rasmussen: Yes, it falls off. High levels of ~1 ppb or more of CH Cl
occurred Ov'er the Indian Ocean or a tropical body of water. Elevated CH_Cl
values vre\e also measured in the proximity of a lingering pall of smoke.
Higher levels are really related to smoldering fires and not the kind of open-
area fires in the Pacific Northwest where we have a rip-roaring fire going up
through the open treetops. In the open flame, this buildup of CH Cl is not
seen.
Dr. Hanst: It seems there is evidence for two sources: the oceans and the
fires.
DT. Rasmussen: We saw seen this vertical profile, Dagmar and I, on the origi-
nal March 1976 flight. The samples on the flight over the Pacific Ocean were
collected on a spiral from ~35000 ft on down to just off the deck. It in-
creased progressively below the boundary layer. It was just a step function;
the CH Cl went up. This was the first indication that CH^Cl really was en-
riched in the boundary layer. And all the subsequent GAMETAG flights corro-
borated that at least the samples that we have gotten.
Dr. Cronn: With a little correction: the flights you talked about were dif-
ferent from the ones that were plotted here.
8-11
-------
TROPOSPHERIC HYDROXYL RADICAL CONCENTRATIONS
AND METHYL CHLOROFORM REMOVAL
Malcolm J. Campbell
Washington State University
Pullman, Washington
INTRODUCTION
This paper describes an attempt by our group at Washington State Univer-
sity (Malcolm Campbell, John Sheppard, Murray McEwan, and Brian Lamb) to es-
timate the rate of atmospheric removal of methyl chloroform (1,1,1-trichloro-
ethane, CH CC1 , MCP). In view of the current consensus that the MCF removal
process involves oxidation by the hydroxyl radical (OH), our effort began with
measured OH concentrations. In conjunction with Fishman and Crutzen's (1977)
model of the variation of OH concentration with altitude and latitude, ground-
level OH determinations at several sites allowed us to estimate the MCF life-
time. This process indicated a major portion of MCF oxidation to probably
occur in the tropics, and at low altitudes.
METHOD
In these calculations, we tacitly assumed MCF escaping oxidation by OH to
be transferred to the. stratosphere. The National Aeronautics and Space Ad-
ministration (1977) rate constant for the OH-MCF reaction was used. Experi-
mentally-determined OH concentrations provided an initial basis for estimation
of the prevailing concentrations; these individual OH concentrations were for
a particular site and time and were not themselves global, diurnal, or seasonal
averages. Many additional measurements of the OH concentration at a variety
9-1
-------
of latitudes, altitudes, and seasons will be required to permit confident
estimation of the MCF lifetime.
14
OH measurements by our group involve the local oxidation of CO and are
described elsewhere (Campbell et al. 1979). Accuracy is believed to be about
±40%, limited mostly by wall effects. Some of our earlier data had larger
error bands. The measurement is inherently an absolute measurement, needing
no calibration. Our group hopes to improve the absolute accuracy of the
measurement over the next few months.
For the purpose of assessing the MCF consumption rate, we selected only
data that were measured in clean air representative of the uncontaminated
boundary layer. These data were collected in Pullman, Washington, and at Mt.
John in New Zealand. Each site is inland but not too distant from the ocean:
Pullman is ~500 mi from the Pacific Ocean, and Mt. John is <100 mi from the
Tasman Sea. We cannot absolutely guarantee, nor do we have measurements to
prove, that the nitrogen oxides (NO ) mixing ratios are below the 100 ppt
X
level that distinguishes continental air chemistry and OH concentrations from
marine values. However, there is no reason to expect NO to have been present
X
in high concentration; certainly there are no major local sources at either
site.
The New Zealand data indicated an average noontime OH concentration of
~7.5 x 10 /cm . These concentrations were measured in April and were cor-
rected to normalize the water content. The Pullman data, measured in July,
averaged 3.4 x 10 /cm . These values were used in conjunction with the Fish-
man and Crutzen (1977) model in order to interpolate between 46° N and 44° S
and to provide some guidance in averaging throughout the seasons of the year.
Our measured Southern Hemisphere (SH) OH concentrations were ~l/2 of what
the Fishman and Crutzen model predicts. Our Northern Hemisphere (NH) figure
was ~1.7 times the model prediction. Considering the experimental errors of
our measurements and the inherent uncertainties of the computer-generated
values, the two estimates were in essential agreement. This was rather
9-2
-------
encouraging, and suggested that the local MCF oxidation rates could be esti-
mated within a factor of 2.
On the basis of this slender evidence, we assumed the Fishman and Crutzen
model to be as good as any other for the purpose of estimating MCF oxidation
rates, and we proceeded to use the concentrations generated by the Fishman and
Crutzen model to estimate global MCF consumption. MCF emission data from Neely
and Plonka (1978) were used along with additional data kindly supplied by Dr.
Neely. A major purpose in carrying out this calculation was to examine the
sensitivity of such calculations to errors in atmospheric measurements and
source strength data.
RESULTS AND DISCUSSION
Figure 9-1 shows calculated contours of the rate of MCF oxidation per
unit of meridional cross-sectional area as a function of altitude and latitude.
This representation includes compensation for the greater surface area per
degree of latitude in the equatorial regions. It is important to note that
the MCF consumption contours have logarithmic intervals, and that much of the
consumption takes place at low altitudes and low latitudes. These results are
due to the distribution of OH concentration as a function of latitude and
altitude, and to the activation energy of the MCF-OH reaction.
Figure 9-2 shows the overall rate of MCF consumption as a function of
altitude. As with carbon monoxide (CO) and methane (CH.), the consumption
rate falls off rapidly with altitude. In fact, half of all consumption ap-
pears to occur below 2.4 km. This suggests that boundary-layer measurements
of OH concentration are far from irrelevant in estimating the global rate of
MCF oxidation.
Figure 9-3 shows the marginal distribution of the oxidation rate as a
function of latitude. It is important to note how much of the oxidation takes
place in the tropics; the calculation indicates half of all MCF removal to
occur between 16° S and 16° N. If the model is accepted, the removal of MCF
from the atmosphere becomes a "tropical affair." This domination of oxidation
9-3
-------
ALTITUDE km
ro
O
JL
O
n3
8 £
M O
fl> 3
O
C H-
1- 3
ro
01 p
3 3
NJ1
01 H-
H-1-
O
3
(D
M
c
0
m
^
o
m
(7>
^]
m
m
c/>
z
ro
0
o-
N>
o
*.
o
-------
dh
, I024MOLECULES km's"'
Figure 9-2. The calculated altitudinal variation of the MCF consumption rate.
9-5
-------
SHADED AREA,-I6°
-------
in the equatorial region has an important consequence: it becomes difficult
to consider the ratio of NH to SH MCF mixing ratios to reflect (in any mean-
ingful way) the ratio of the simple hemispheric means of the OH concentrations.
It is difficult, for example, to postulate a SH OH concentration that is sev-
eral times the NH value when the transition from the higher SH value to the
lower NH value must occur in a region within a few degrees of the equator.
From these rather crude computations, we conclude that additional tropi-
cal OH concentration data are one of the principal needs for better estimation
of MCF consumption by this direct method. Our group hopes to make a number of
measurements of tropical boundary-layer OH concentrations in the near future.
COMMENTS ON ALTERNATIVE METHODS
Other chapters in this volume discuss alternative methods of estimating
the MCF tropospheric lifetime. These discussions do not always acknowledge
that NH and SH MCF mixing ratio data, together with data on the rate of re-
lease to the troposphere, permit computation of the lifetime in two essentially
distinct ways.
The first method involves comparison of the worldwide MCF burden with
integrated emission data in order to determine the amount that has been lost,
leading to computations of consumption rate and lifetime. For compounds with
long lifetimes, this method is very sensitive to errors in either emission or
absolute mixing ratio data. Essentially, the method involves calculation of
a small difference between two rather large quantities. Further studies of
the absolute calibration of the mixing ratio data would improve the method's
accuracy.
The second method uses the ratio of the mean NH and SH mixing ratios, in
conjunction with the interhemispheric transfer time. From these data it is
possible to calculate the rate of consumption (or at least something close to
the rate of consumption) in the SH. Data on the ratio of NH to SH mixing
ratios are rather consistent, and this method would appear to be less subject
to error than the first. With the assistance of Brian Lamb, our group has
9-7
-------
performed some box modeling of tropospheric MCF consumption. Reasonably good
agreement with the measured NH/SH ratio is obtained, although a better fit
would result from reduction of the assumed rate of consumption in the SH. The
mean lifetime is 6.4 yr.
REFERENCES
Campbell, M. J., J. C. Sheppard, and B. F. Au. 1979. Measurement of hydroxyl
concentration in boundary layer air by monitoring CO oxidation. Geophys.
Res. Letters 6:175-179.
Fishman, J., and P. J. Crutzen. 1977. The distribution of the hydroxyl radi-
cal in the troposphere. Report on EPA Grant R804921-01, December.
National Aeronautics and Space Administration. 1977. Chlorofluoromethanes
and the stratosphere (R. D. Hudson, ed.). NASA RP-1010.
Neely, W. B., and J. H. Plonka. 1978. Estimation of time-averaged hydroxyl
radical concentration in the troposphere. Environ. Sci. Tech. 12:317-321.
DISCUSSION
Voice fpom Audience: When you say you've postulated a decrease in consumption
of the SH, that's for the purposes of the model and not for your measurement?
DP. Campbell: Yes. However, Dr. Crutzen1s model and our measurements do not
disagree. Our measurements are not numerous enough, at the moment, to suggest
that it's wrong. Nor are they numerous enough to indicate conclusively that
it's correct.
Dr. Singh: Dr. Crutzen, what do you think is the validity of your model's
assumptions in the SH?
DP. Crutzen: This was indicated in my talk [Crutzen, this volume]. In this
modeling, at the moment we can get at.almost any number. I don't think you
can get below 10 molecules OH per cm ; that is excluded. But starting from
there up to 10 is possible. The figure depends on how much nitric oxide (NO)
there is and whether you include heterogeneous factors. There are many such
factors, and all we lack are the data. With MCF, there is still uncertainty
as to whether or not there are some additional sources. Personally, I don't
strongly believe there are, but this cannot be excluded. There are other
fluorocarbons which may help the modelers get the information we're groping
for.
There are unacceptably large uncertainties in the modeling effort. Also,
I honestly think that we may lack some basic input in the chemistry. It's an
-------
extremely uncertain game at the moment, but that doesn't mean we shouldn't do
it.
By the way, the model used by Dr. Campbell the November 1977 model is
actually a newer model than the one I presented. Mine was the older model
dating from 1975 to 1976, with the old rate constants. It was higher in OH
concentrations; later, when the MCF data came in, we again adjusted our models
downward to get lower numbers. Thus, Campbell's is a newer model.
DP. Singh: Aren't we going in circles? You have a model, here, that was
brought down to fit the data.
Dr. Campbell: I think that's true to some extent. Perhaps we can have a
short discussion on this point: the ability of models to give information
here. It seems to me that what has largely been done with the modeling to
date is to use the total emission rate data and the absolute mixing ratio data
to determine the OH concentration in the NH, and to then use the ratio between
the NH mixing ratio data to determine the OH concentration in the SH. I be-
lieve that's equivalent to what you've done.
If that description is correct (and I believe it is), it means that the
SH rates of consumption are very well estimated, but that the NH rates of
consumption may be subject to considerable error. In other words, there's a
question about the NH rates of consumption, but much less of a question about
the SH rates of consumption. The NH rates of consumption will be pinned down
only by mixing ratio measurements of greater absolute accuracy, and by more
accurate estimates of the total emissions.
DP. Bufalini: Your OH value has been so high. Have you run a model to test
what side effects NO would have? As you know, Don Stedman has proposed a
chemical amplifier to measure the OH and HO radicals using NO.
Dr. Campbell: No, I was not aware of that proposal. On the question of
sensitivity of OH to variations in NO, our model calculations suggest that it
it requires nearly 0.5 ppb NO to make any significant change in the predicted
OH concentration.
Voice from Audience: But the background levels of NO are expected to be in
the low ppb, are they not?
Dr. Campbell: Not in the sites where we were making measurements.
Voice from Audience: How did you measure it?
Dr. Campbell: At Pullman, there's a record of instruments with noise levels
of ~2 or 3 ppb showing no readings. The New Zealand site is, in general
terms, an excellent site: a mountain range separates it from the ocean;
between the mountain range and the observation site, there are probably not
more than 50 habitations.
9-9
-------
DP. Logan: I would like to make a couple of comments. First, I think there's
some confusion, here, about what is "high OH" and "low OH." Some of Dr.
Crutzen's measurements and some of Dr. Campbell's measurements made at noon
seem quite different.
DP. Campbell: Roughly, by a factor of 3 or 4.
DP. Logan: A second point I would like to make is that, from our calculation,
OH values in this range are very sensitive to the amount of NO in the model.
This effect is seen until NO volumes are below 10 ppt. Below 10 ppt, OH does
not depend on the NO mixing ratio.
DP. Campbell: There is another variable here, I'm afraid, where we differ.
As far as the minor nitrogen species are concerned, we have not assumed that
our model represents a steady state obtained over a long period of time. In
other words, we've assumed that we start off, at least periodically, with
relatively clean air such as is obtained after passage of a front.
Voi.ce from Audience: My own feeling about the MCF results is that the life-
times calculated by just taking the total assay at any given time versus total
emission are the most dependable, leaving the OH out of it completely, and
calculating a world average removal rate. And, as I say, it doesn't really
depend very much on the mixing ratio of the mean hemisphere or how it is dis-
tributed between the two hemispheres. It's just how fast it's going away.
With that, we get a lifetime of ~6 yr with almost any choice of other para-
meters. It's only in attempting to model north-south differences that we
become involved with OH.
DP. Campbell: It is true that the tropospheric lifetime can be estimated from
production data without reference to OH. Nevertheless, the resulting lifetime
estimate remains very sensitive to the absolute value of the mixing ratio.
Voice from Audience: That's right. You have to measure the absolute rate
accurately and compare it with accurately known values. Then you can obtain
the lifetime of MCP and use that to try to obtain the OH concentration. But
you don't need to go to OH to get the MCF lifetime.
DP. Campbell: I agree completely.
9-10
-------
IMPACT OF BROMINATED COMPOUNDS ON THE STRATOSPHERE
Luisa T. Molina
Mario J. Molina
University of California
Irvine, Caliornia
INTRODUCTION
The atmospheric chemistry of chlorinated species has been the subject of
many investigations in the past few years due to removal of stratospheric ozone
(O ) by the chlorine oxides (CIO ) catalytic chain (National Academy of Sci-
-3 X
ences 1976). Introducing brominated hydrocarbons with long residence times
into the atmosphere carries potential for depletion of stratospheric 0 simi-
lar to that by chlorinated hydrocarbons, due to a corresponding bromine oxides
(BrO ) catalytic chain:
X
Br + 0 -» BrO + 0
BrO + O -> Br + O
Presently the industrial production of chlorocarbon molecules is very
much larger than that of bromocarbon molecules, which are considerably more
expensive. Nevertheless, if future restrictions on some chlorinated compounds
become severe, or if the tonnage of brominated compounds continues to increase,
detailed information on stratospheric Br chemistry and its potential tropo-
spheric bromocarbon sources will be required as a basis for regulatory action.
10-1
-------
BROMOCARBON CHEMISTRY
The BrO chain is a cause for greater concern than the CIO chain in at
X X
least one respect (per halogen atom released) interruption of the BrO chain
X
by hydrogen bromide (HBr) formation is much less frequent than interruption of
the CIO chain by hydrochloric acid (HC1). The abstraction reactions of atomic
X
Br with methane (CH.) and hydrogen gas (H2) are both sufficiently endothermic
(Hudson 1977) that neither is a factor in stratospheric reaction cycles:
Br + CH -> HBr + CH
Br + H -> HBr + H
The reaction of Br with HO does occur under stratospheric conditions, as
does the hydroxyl radical (OH) reaction (Hudson 1977) which returns the Br to
the BrO cycle:
x
Br + HO -> HBr + O
OH + HBr + HO + Br
The absence of abstraction from CH. or H causes Br to spend a higher
fraction of its stratospheric lifetime in the BrO chain (chiefly as BrO, and
less as HBr) than does Cl in the CIO chain versus HC1. However, much chemis-
X
try remains to be learned before the overall catalytic efficiency of Br for
stratospheric O., removal can be assessed with confidence.
ULTRAVIOLET ABSORPTION SPECTRA
A key question in the overall consideration of the stratospheric Br prob-
lem now lies in the troposphere. Brominated molecules absorb at much longer
wavelengths than the corresponding chlorinated molecules, with ultraviolet
(UV) absorption maxima in the 200-240 nm range. The long wavelength "tails"
of these absorption maxima extend into the 280-320 nm region of the strato-
spheric "O cutoff" near 290-295 nm. Even very small absorption cross sections
10-2
-------
on the long wavelength end of 295 nm permit tropospheric photodissociation at
all altitudes, resulting in greatly reduced atmospheric lifetimes and (con-
sequently) greatly reduced stratospheric effects.
Earlier measurements of the UV absorption cross sections for brominated
molecules indicated that compounds containing the -CBr2X grouping absorb suf-
ficiently strongly beyond 295 nm to have relatively short atmospheric life-
times. In contrast, measurements of several compounds containing the -CBrX_
group (bromotrifluoromethane (CBrF ), CBrCIF , CBrF CBrF , etc.) did not in-
dicate any absorption beyond ~280 nm (Robbins 1976; Doucet et al. 1975). Such
bromofluorocarbon molecules presumably have long atmospheric residence times,
and would present possibly serious stratospheric O depletion problems if
released to the atmosphere on a large scale.
All previous measurements employed standard short-path UV absorption
cells. New measurements of the UV cross sections of several brominated mole-
cules of industrial significance have been carried out in our laboratory in
order to determine their photochemical stability, giving special attention to
the very low cross sections beyond 270 nm. Several other brominated hydro-
carbons have been measured as well, in order to determine general substituent
effects on the photochemical cross section. The sensitivity of these measure-
ments has been greatly increased through the use of a 2-m path quartz absorp-
tion cell attached to a Gary 219 UV-visible spectrophotometer.
In some cases the weak spectrum was measurable with a short, 10-cm cell
by using sample pressures of several hundred torr. However, this approach was
found to be less satisfactory because of baseline drifts, probably from sample
adsorption on the cell windows. No such complications were encountered at
the lower pressures used in the long-path cell experiments.
The results are summarized in Figure 10-1. Both CBrF and methyl bro-
mide (CH Br) have cross sections in the 10 cm /molecule range near 280 nm,
and negligible absorption cross sections beyond 295 nm. The other four mole-
-22 2
cules all exhibit cross sections in the 10 cm /molecule range or larger
-21 2
near 295 nm (10 cm / molecule for CBr F ) and consequently may undergo
10-3
-------
10
C-18
10
.-19
10
10
..-21
CJ
o
10
-23
10
10
.-25
~i 1 1 1 1 1 r
--- CH3Br
CBrF3
CBr2F2
CBrCIF2
C2BrF5
C2Br2F4
\\ \ \ \
X>- \ Vv \
V- \ \\ \
\\ \ V ^
V. \ \\ *
200 240 280 320
X (nm)
360
Figure 10-1. Absorption cross sections for several brominated hydrocarbons.
10-4
-------
solar photodissociation in the troposphere. The apparent absorption cross
sections for CBrCIF and CBrF CBrF , while very small, are sufficient to
create an important tropospheric photochemical sink. However, these measure-
ments are very sensitive in tracing impurities that might be present in the
bromofluorocarbon molecules. The presence of 0.1 percent molecule impurity
(such as CBr F in CBrClF or CBr FCF in CBrF CBrF ) would cause substantial
£ £ £ £3 £ £
perturbations in the measured cross sections beyond 295 nm. Comparison of the
calculated tropospheric lifetimes of these two molecules (CBrCIF > CBrF^CBrF )
illustrates the importance of the "tail" between 300-320 nm for its apparent
more rapid removal. The current measurements actually furnish only lower
limits on the tropospheric lifetimes of CBrCIF and CBrF CBrF .
ATMOSPHERIC PHOTODISSOCIATION RATES
Neither CH Br nor CBrF photodissociates in the troposphere (although
CH-Br is rapidly removed by reaction with tropospheric OH). Some preliminary
estimates of atmospheric photodissociation rate, J, for the other three bromo-
fluorocompounds have been carried out under the assumption that the cross
sections in Figure 10-1 are accurate. For overhead sun conditions, at mid-
latitudes, and at the earth's surface, the values (in units of s ) are about
6 x 10~9 for CBrF2CBrF2; 1 x 10~8 for CBrClF2; and 2 x 10~ for CBr^. The
corresponding tropospheric lifetimes, using average J values that are half of
those given above, are CBrF CBrF > 10 yr; CBrCIF > 6 yr; and CBr F > 5
months. The stratospheric lifetimes of the first two are ~30 yr, so that
approximately >20 percent of CBrCIF photolysis and >30 percent of CBrF CBrF
£ Z. £
photolysis occur in the stratosphere.
The steady-state distribution with altitude for CBrF in the stratosphere
has been calculated using a one-dimensional model and Chang's "eddy diffusion"
coefficient (Hudson 1977). The atmospheric lifetime corresponding to this
stratospheric dissociation process is ~50 yr. At present, CBrF is the only
brominated hydrocarbon of current technological importance for which tropo-
spheric photodissociation is certainly unimportant.
10-5
-------
The existence of tropospheric sinks other than gas phase solar photo-
dissociation may have to be considered for the bromocarbons. This question
has been discussed rather extensively in the case of chlorinated hydrocarbons
(National Academy of Sciences 1976; Hudson 1977). Sinks such as hydrolysis in
the oceans or photodissociation of adsorbed molecules may play a significant
role for species such as CBrF CBrF , though the possibility is not likely.
REFERENCES
Doucet, J., R. Gilbert, P. Sauvageau, and C. Sandorfy. 1975. Photoelectron
and far-ultraviolet spectra of CF Br, CF.BrCl, and CF Br . J. Chem.
Phys. 62:366-369.
Hudson, R. D., ed. 1977. Chlorofluoromethanes and the Stratosphere.
Reference Publication 1010, National Aeronautics and Space Administration.
266 pp.
National Academy of Sciences (Committee on the Impacts of Stratospheric Change)
1976. Halocarbons: Effects on Stratospheric Ozone. National Academy of
Sciences, Washington, D. C.
Robbins, D. E. 1976. Photodissociation of methyl chloride and methyl bromide
in the atmosphere. Geophys. Res. Letters 3:213-216.
10-6
-------
BRIDGE BETWEEN THE SCIENCE AND THE REGULATORY NEEDS
Herbert L. Wiser
U.S. Environmental Protection Agency
Washington, D. C.
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) uses and is highly depen-
dent on scientific information in carrying out its responsibilities. This
paper discusses the kind and quality of scientific information that is valu-
able to policy and regulatory decision-makers.
REGULATORY CONCERNS
Methyl chloroform (1,1,1-trichloroethane, CH-CC1 , MCF) and other halo-
carbons have been detected virtually over the entire globe; atmospheric con-
centrations, according to some data, are increasing significantly. Researchers
who feel action should be taken to control (or not control) MCF should there-
fore transmit that information to regulatory policy-makers. The first ques-
tions in a series of many to be answered are: What atmospheric circumstances
do these data reveal, and what should be done about them? What effects will
result from increased concentrations of halocarbon contaminants in the atmo-
sphere, aquatic systems, soils, or flora and fauna? When will these effects
occur? How much of a particular pollutant exists or is created naturally?
What is man's contribution to that pollutant concentration?
Even more fundamental questions are: How much of a concentration can the
environment or man or other living creatures tolerate? Is there an acceptable
11-1
-------
threshold level? What are the effects of man's contributions to the pollution
problem, and can preventive action be taken to reverse the situation of in-
creasing concentrations that lead to undesirable events in the future? Over
what time period will such action occur, and with what degree of control of
the pollutant?
Policy- and decision-makers weigh the pertinent factors at hand to reach
their regulatory decisions, and thus must have all relevant information avail-
able to them. Though absolute scientific certainty is preferable, of course,
complete accuracy is not attainable in the real world. Controlled global ex-
periments in which many variables are held constant and only one or two varied
cannot be performed; the natural processes and scales of nature are in contin-
uous flux. Even to obtain health effects data, epidemiology is virtually the
sole means, since human subjects are not used in the laboratory to study car-
cinogenesis or other serious effects.
But responsible regulatory decision-making can be carried out in the
absence of absolute certainty, because many types of uncertainty can be evalu-
ated (e.g., through assessing data quality or determining the degree of con-
fidence existing in the basic structure of a pollutant model). Knowing the
magnitude of uncertainties associated with a data set is critical, and decision-
making must be informed of error bars and the upper and lower bounds of data,
the completeness or incompleteness of theory models, and any qualifying hypo-
theses underlying the models or analyses.
Frequently, however, data are presented as a collection of points, with
or without error bars. A question pertinent to the use of such data is: If
the error bar extremes (either on the high or the low side across the graph)
happen to be true values, or are actually the real values with much smaller
errors, would the resulting conclusions drawn from the data be the same as
those postulated using the data as points? Not necessarily. Illustrations
are commonly presented even in published literature where the measured or
calculated data points in a graph are connected by a straight line or care-
fully shaped curve. An example is a specific case in which curves were deter-
mined by a polynomial or several polynomials whose coefficients were based on
11-2
-------
10 or 20 percent errors, then were stated to 3 or 4 significant figures in
accompanying tables; the curves were drawn very precisely without an error
band width.
Again, theory credibility is important to insure that crucial factors are
not overlooked, or that parameters dismissed as insignificant may not later be
indeed sic,iificant. The question to be answered is: What is the sensitivity
of the pollutant model, or predictions based on experimental points, to such
parametric changes as different reaction rates?
Situations may occur, as with the currently pressing question of strato-
spheric ozone (O ) depletion, that may not permit the luxury of waiting for
absolute certainty. These situations are especially urgent if the act of
delaying corrective measures causes the measures, when eventually implemented,
to be ineffective for many decades. Mankind and the global environment would
thus suffer greater damages for several decades or generations, as well as
during the interim period before action was taken.
Such decisions whether to regulate now or later, or whether to regulate
at all depend on knowledge at hand, and especially on the confidence in that
knowledge.
A data set or model may have a 1-a or 2-a confidence level; these con-
fidence levels may warrant controls of some sort. Of course, the probability
that events will occur outside the ±1- or ±2-a boundaries is real, but this
probability is significantly lower than the probability for events to occur
within the bounds. Furthermore, a decision based on an event occurring inside
these 1- or 2-a bounds may differ totally from a decision based on an event
occurring, say, above or below the bounds.
A useful illustration of this situation is the fluorocarbon effect on
stratospheric O , and ensuing effects. The decision was made to regulate and
ban certain fluorocarbon usages (e.g., nonessential propellants and production
of fluorocarbons for these usages) by EPA, the Food and Drug Administration
(FDA), and the Consumer Product Safety Commission (CPSC). The decision was
11-3
-------
based on values, so to speak, within most of the Gaussian curve of fluoro-
carbon values. If the real world were in the lower range of the curve, the
stratospheric effects of halocarbons could be minimal, or might even cause an
increase in stratospheric O_. But the probability of this event is very low;
therefore, the decision to regulate fluorocarbons was justified. Also to be
considered is a real world situation in which extremely high data values might
indicate a catastrophic impact on stratospheric O_, mandating a decision for
much more stringent control. The important point is that credibility or sta-
tistical value of all data must be weighed by policy-makers, so that decisions
do not rest solely on one data point or model.
INDUSTRIAL CONCERNS
Now, with regard to MCF and the other halocarbons discussed, just as with
the chlorofluoromethanes (CFM's), industrial concerns hinge upon the following:
If future use of compound A or B will not be permitted, what substitute com-
pounds will be permitted? Manufacturers and users need alternative chemicals,
not only to maintain their businesses, but to insure employee jobs and fulfill
consumer demands.
Reactivity is an important molecular property in atmospheric chemistry.
CFM has a reactivity in the troposphere of almost 0, is extremely inert, and,
as far as is known, has no harmful health effects (unless breathed at extreme
concentrations). Because of its inertness, CFM is a very useful substance in
many industries. Because of its long persistence in the troposphere, however,
CFM diffuses to the stratosphere and eventually causes an O decrease.
The reactivity of other halocarbons and hydrocarbons may be high in the
troposphere, forming photochemical smog and causing reactions in the tropo-
sphere that cause or contribute to poor air or water quality. These chemicals
are harmful to health or ecosystems, but at least the range of atmospheric
damage is confined to near ground level in the troposphere. Tropospheric
lifetimes are short because reactivities are high; no significant quantities
diffuse to the stratosphere.
11-4
-------
Does a middle value or regime of values or properties exist for a whole
class of substances, such as hydrocarbons, that may be established as a guide?
Some chemical compounds are known to be inside or on the border of a regime.
Future research may identify other characteristics rendering them uniquely
acceptable or unacceptable. Such a property regime needs to be identified and
established as a guide for selecting, developing, or substituting industrial
substances.
CONCLUDING REMARKS
The decision-making process over the past few years has been growing more
complex. EPA (along with other agencies, presumably) has always accumulated
the available scientific knowledge pertinent to pollutants and resulting ef-
fects. EPA also considers the instrumentation necessary to measure substances
in the laboratory and to monitor substances in the environment. Furthermore,
consideration is given to control options or measures.
EPA additionally analyzes the socioeconomic impact of pollutant regula-
tions. Costs examined include not only the price of control equipment or the
economic impact of regulations, but also the cost to health, to materials, to
other living things, to the quality of life, and so forth. Decision-makers
consider all of these factors in creating an integrated assessment of the
problem. Assessments are made available to the general public, scientists,
and industry. After much deliberation and reevaluation, a final regulatory
decision is made.
To return to the original point, scientific evidence is the basis for
identifying and measuring environmental damage, for identifying the pollutant
or other cause of the damage, and for suggesting corrective measures. Eluci-
dating uncertainties in scientific data or models is therefore an important
element in presenting factual information, creating scientific models, or
proposing scenarios of probable future events. Additionally, greater under-
standing of gaps or differences in scientific data permits a more judicious
perspective of experimental data that may be open to several interpretations
within the scientific community.
11-5
-------
DISCUSSION
Dr. Slobodow: You raised one question and I didn't hear the answer. Is there
optimum reactivity?
Dr. Wiser: That is what I am asking the audience to define. Is it possible
to identify an optimum reactivity or a regime of acceptable activity? We
don't have an answer.
Dr. Campbell: This is perhaps an unfair question, but could you comment on
how EPA feels about the way things have gone with these fluorocarbons?
Dr. Wiser: I would rather wait for the regulatory speaker later this morning
who could better answer that. I think the answer would be the same.
Dr. Bufalini: I am not sure I really agree with the concept of an acceptable
reactivity. According to the way EPA has gone with reactivity, it would seem
that all hydrocarbons could be classified as reactive to some extent, depend-
ing on concentration. As a matter of fact, I think most modelers here agree
that with sufficient methane, oxygen, and nitrogen you can, in the model,
exceed air quality standards. The question then becomes: What concentration
is acceptable to keep the 0, down? It would appear that we must weigh the
total concentration emitted to the atmosphere, the amount that can get into
the stratosphere, as well as the amount that produces photochemical smog. I
guess what I am suggesting is that if a sufficient amount of material gets
into the atmosphere, it will be additive. So I think the best control is no
emissions at all, obviously, but I realize that is impossible.
Dr. Wiser: I agree with your comment, except for the last conclusion. I was
using "reactivity," as I said, as a catchall for the various parameters, in-
cluding concentrations. There may be other characteristics that are just as
important.
I disagree with your conclusion that we ought to just "stop the world."
I don't think we can live with that. One of the items stated yesterday was
that this year, again, auto deaths were the highest since we cut back on the
speed limit. Auto accidents cause 50,000 deaths a year. We live with it. We
accept it. It is a trade-off our society has made. There are many who will
not accept it, and they have to live their way, but society as a whole has
accepted it. So there are times when we do accept a harmful effect. Maybe
it's because, individually, we believe it won't happen to us.
Dr. Singh: Just a note of caution on the reactivity. We may have species
with undefined secondary products which are really more stable than the species
we started with; just because the initial molecule is reactive is not suffi-
cient grounds to go ahead and start to model it. We must consider much more
than just the reactivity of the initial molecule.
Dr. Wiser: You're right. I am not going to take the time to go over it, but
I just realized I omitted my discussion on sources and sinks. Thank you.
11-6
-------
AN ASSESSMENT OF TRICHLOROETHYLENE, METHYL CHLOROFORM,
AND PERCHLOROETHYLENE
Thomas Lapp
Midwest Research Institute
Kansas City, Missouri
INTRODUCTION
This paper is the result of a comprehensive literature review on tri-
chloroethylene (C HC1 , TCE), methyl chloroform (1,1,1-trichloroethane,
CH CCl , MCF), and perchloroethylene (C Cl , PCE). Areas of interest in-
J J ^4
elude manufacturing process technology, consumption and utilization, alter-
natives, health impacts, ecological effects, monitoring data, and exposure
levels.
After data evaluation, the need for limiting the quantities of these
three compounds entering the environment was assessed. A draft of the final
report has been submitted to the U.S. Environmental Protection Agency (EPA)
Office of Toxic Substances (OTS), but will be modified to include data pres-
ented at the Conference on Methyl Chloroform and Other Halocarbon Pollutants.
In general, much of the information derived from manufacturing and
marketing is not directly related to the theme of this paper; however, re-
cent U.S. production quantities of the three compounds are listed in Table
12-1. All three compounds enter the environment primarily through atmospheric
emissions. In the lower atmosphere (troposphere), TCE and PCE undergo photo-
oxidation to produce the corresponding acetyl chlorides (C_H CIO, etc.),
phosgene (CCl O), and hydrogen chloride (HC1).
12-1
-------
TABLE 12-1. RECENT U.S. PRODUCTION QUANTITIES OF TRICHLOROETHYLENE,
METHYL CHLOROFORM, AND PERCHLOROETHYLENE3
Compound 1977 1978
TCE 298 301
MCF 635 623
PCE 614 721
Total 1547 1645
Millions of pounds.
The tropospheric lifetime of TCE has been calculated to range from ~0.1
days to 6 weeks; for PCE, from 1 day to 21 weeks. These ranges of tropospheric
lifetimes result from differences in methodology and rate constants used in
calculations by the different research groups. Regardless, both compounds
undergo relatively rapid photooxidation.
MCF, in contrast, undergoes slow tropospheric photooxidation (residence
time pi 1 to 11 yr) to yield CCl 0, carbon oxides (CO ), and HCl. Because of
*- X
its slow decomposition, MCF is also subject to transport into the stratosphere,
where it is thought to undergo photodissociation (in much the same manner as
currently hypothesized for chlorofluorocarbons) to yield Cl atoms and chlorine
oxide (CIO ) radicals. These atoms and radicals can participate in ozone (O.,)
X O
depletion reactions. Using a tropospheric lifetime of 8 yr, ~15 to 20 percent
of current tropospheric MCF is calculated to reach the lower stratosphere,
resulting in a steady-state 0. depletion of ~10 to 20 percent of that calcu-
lated for the chlorofluorocarbons.
In aqueous media, the primary dissipative process is evaporation rather
than hydrolysis. In one study, 90 percent evaporation was shown to occur in
~1 hour for MCF and TCE and in 1.5 hr for PCE. The ranking of these compounds
in decreasing ability to hydrolyze is: MCE, PCE, TCE. TCE, in fact, is gen-
erally considered to be resistant to hydrolysis under normal conditions. The
principal products from hydrolysis of MCF are acetic acid (CH COOH), hydro-
chloric acid (HCl), and vinylidene chloride (C_H Cl ); those of PCE are tri-
chloroacetic acid (C2HC1 O2) and HCl.
12-2
-------
No data were found relating to reaction rates, decomposition products, or
persistence of these compounds in soils or sediments. One study proposed that
the products probably would be the same as for aqueous hydrolysis, but no con-
firmation was provided.
MONITORING DATA
Levels in Ambient Air
Ambient air levels at manufacturing sites were generally <2 ppb for MCF,
<2.5 ppb for TCE, and <5 ppb for PCE. The highest reported average levels for
any one site were 15 ppb for MCF and 14 ppb for TCE. PCE concentrations were
<5 ppb at all sites.
Data for only one TCE and one MCF user site were reported. At the MCF
user site, average air levels were ~4.4 ppb, or twice the average of the manu-
facturing sites. At the TCE facility, average levels were ~19.8 ppb, or 9
times the average of the manufacturing sites. Reported ambient air emissions
from dry cleaning plants using PCE involved levels ranging from 1 ppm to >1000
ppm in outlet air vents, depending upon the sample time and the particular
establishment.
Ambient air levels have been reported for 27 other U.S. cities or areas.
Mean air levels for TCE ranged from undetectable to 2.92 ppb. For PCE, mean
air levels reached a high of 4.5 ppb. Of the TCE sampling sites, 79 percent
showed mean levels of >0.1 ppb. Seventy-seven percent of the MFC sampling
sites showed mean levels of >0.1 ppb, as did 75 percent of the PCE sites.
Levels in Water
Tap water in 22 U.S. cities or areas was sampled for presence of one or
more of these three compounds. TCE was detected in 14 of the cities; one city
showed a level of 32 ppb, while 10 cities or areas showed levels of <2 ppb.
The highest concentration of MCF was 17 ppb; the other 13 cities showed levels
12-3
-------
of <1 ppb. For PCE, the highest level was 2 ppb; the remaining 12 cities had
concentrations of <0.4 ppb.
Nontap water concentrations were measured at 6 manufacturing sites, 1 TCE
and MCF user site, 3 dry cleaning establishments, and 204 other U.S. sites.
At the manufacturing sites, levels upstream from the plant outlets ranged from
0.4 to 353 ppb for TCE and from 0.1 to 132 ppb for MCF. At the plant outlets,
levels ranged from 74 to 535 ppb for TCE and from 5 to 344 ppb for MCF. No
levels were reported for PCE at manufacturing sites. For the only user site,
upstream levels were 5 and 6 ppb TCE and MCF, respectively. Downstream of the
plant outlet, levels ranged from 8 to 26 ppb for TCE and from 6 to 18 ppb for
MCF.
At three dry cleaning establishments, wastewater from the carbon bed
adsorption system was discarded into the sewer system. PCE levels in the
wastewater ranged from ~6 to 1000 ppm, depending upon the site sampled and the
time during the desorption cycle that the sample was obtained.
Of the 204 other U.S. sites sampled, 95 percent showed levels of <6 ppb
for TCE, MCF, and PCE. Approximately 75 percent of all sites showed levels of
these three compounds to be <1 ppb. The maximum levels detected for MCF and
PCE were 8 ppb and 45 ppb, respectively. TCE levels were the highest of the
three compounds, with a maximum concentration of 188 ppb.
Levels in Soil
Only one study was found in which levels of TCE and MCF were measured in
soil. Concentrations ranged from undetectable to highs of 5.6 ppb for TCE and
3.4 ppb for MCF. In the U.S., sediment levels have been measured only at
manufacturing and user sites; MCF levels ranged to a maximum of ~6 ppb. . There
was great variation in TCE levels, which ranged from undetectable to a maximum
concentration of 300 ppb.
12-4
-------
Levels in Lower Stratosphere
A 1976 study by Rasmussen and coworkers at Washington State University
reported average MCF-levels in the lower stratosphere of 79 ppt.
ENVIRONMENTAL IMPACTS
Few data are available concerning the acute or chronic toxicities to
environmental species of the three subject compounds, but fish appear to be
susceptible to low ppm concentrations. The lowest reported LC,- values were
5 ppm for PCE, 16 ppm for TCE, and 33 ppm for MCF. In general, fish appear to
be capable of bioconcentrating these solvents to levels of ~100 times the
aqueous concentration.
HEALTH IMPACTS
The lowest chronic exposure level at which some type of human physiologi-
cal effect can be consistently observed for any of these three compounds is
~50 ppm TCE. When this level of exposure is maintained for extended periods
(e.g., in a work area), a large proportion of exposed individuals experience
dizziness, headaches, and incoordination; these effects are reversible after
the individual is removed from the exposure area. With MCF, the dizziness,
headaches, and incoordination do not usually occur until levels reach ~250
ppm for similar exposure times. The effects of PCE, however, are somewhat
different from those of TCE or MCF. Only ~6 percent of inhaled MCF is re-
tained by the body the remainder is exhaled immediately, and even the re-
tained MCF apparently is later expired largely unmetabolized. TCE is more
readily absorbed (~90 percent) during inhalation but very slowly metabolized
by the body; the effects of TCE are much more prolonged than those of the
other two compounds.
An epidemiological study in which workers were exposed to ~150 ppm MCF
for periods of 1 to 6 yr showed no adverse health effects. There is no evi-
dence of worker death as a result of long-term occupational exposure to MCF or
the other compounds. Case histories show, of course, that deaths have occurred
12-5
-------
as a result of occupational exposure, but these cases involved accidental ex-
posure to very high levels (~7000 ppm or more). At such levels, the very swift
narcotic or anesthetic effects of these compounds render the victim unconscious,
causing death from overexposure.
All three compounds (but especially MCF) can sensitize the heart to the
effects of epinephrine. The required dose and exposure level, the mechanism,
and the number of persons at risk from this type of sensitization are unknown.
Sensitization with fatal results has been most frequently reported after ex-
posure to high levels (~7000 ppm) of MCF.
Studies by the National Cancer Institute indicated TCE and PCE to be
potential carcinogens; investigators found oral doses of the compounds to pro-
duce liver tumors in mice but not rats. Tests conducted with MCF produced no
tumors, but high dosages resulted in animal data that were not suitable for
statistical analysis, and no conclusions could be drawn regarding the carcino-
genicity of this compound. The average daily oral doses of TCE and PCE were
~500 and ~1000 mg/kg, respectively. The results of the bioassays have created
considerable controversy because of the high dose levels used, the method of
dosage, and the production of liver tumors in mice only. The predominant
human exposure route for these two compounds is generally inhalation, not oral
ingestion, which can vitally affect distribution. The carcinogen bioassays
are being repeated with improved experimental design for all three compounds.
EXPOSURE LEVELS
TCE, MCF, and PCE can be assimilated by inhalation, ingestion (food and
water), and dermal absorption. The present discussion is limited to inhala-
tion and water ingestion; no data are currently available for the presence of
these compounds in U.S. food products, but this route of intake is tentatively
assumed to be negligible. Quantities introduced into the body by dermal ab-
sorption are also considered negligible in comparison to inhalation and water
ingestion.
12-6
-------
Bodily levels due to inhalation were calculated for every city or area
having monitoring data. For TCE, only five cities showed levels of human
exposure of >1.5 micrograms per kilogram body weight per day (yg/kg/day).
The highest level of those five was ~26; other levels were ~18, ~3, and ~2
pg/kg/day. Calculated exposure levels for all large cities with high popula-
tion densities were <1 pg/kg/day. For MCF, citizens of only one city showed
a calculated level of >2 yg/kg/day; data for large cities ranged from a high
of 1.3 to a low of 0.1 pg/kg/day. Six cities had calculated PCE human inhala-
tion levels of >1.2 yg/kg/day. The two data sites with the highest calculated
levels were New York (7.8 pg/kg/day), and a small city. Los Angeles showed a
level slightly greater than 2 pg/kg/day. In general, the available data show
that a relatively small segment of the general population is exposed to TCE or
MCF air levels resulting in bodily retention of >1.5 pg/kg/day, yet the same
statement is not necessarily true of PCE.
Data on TCE in drinking water indicated only two cities in which citizens
had calculated exposure levels of >1 yg/kg/day; the highest level was 1.6.
One other city had a level of 0.95, and all other cities had levels of <0.25
pg/kg/day. Data for MCF were difficult to assess due to a lack of quantita-
tive values. For only one city was the calculated level appreciably above
0.05 yg/kg/day. In general, large metropolitan areas showed little if any MCF
in drinking water. For PCE in drinking water, only one city showed a calcu-
lated human exposure level of >0.02 pg/kg/day.
Exposure levels calculated from both ambient air and drinking water con-
centrations were available for few cities. One of five cities showed a total
calculated human TCE exposure level of >4 yg/kg/day. For MCF, one of five
cities showed a calculated level of >2 pg/kg/day. All other cities had levels
of <1.2 pg/kg/day. Data for PCE were confined to two cities, both of which
are metropolitan areas with populations in excess of 1 million. The highest
level was ~7.8 pg/kg/day; the other level was ~1.2 pg/kg/day.
Based on the information derived from this study, including that of
Mazaleski (this volume), available control options to reduce emissions of
these compounds have been noted for consideration by appropriate governmental
12-7
-------
agencies. These options should not be construed as EPA policy. After consid-
eration of all data available from all sources, EPA must determine which regu-
latory options, if any, should be exercised.
Generally, the data show that no basis now exists for a total regulation
or cessation in manufacture of these three compounds. This statement is not
a judgment that manufacture or use of the compounds pose no human or environ-
mental risk, but rather that the data are inconclusive, at this time, with
respect to such manufacturing restraints. Certain options in selected areas
do appear appropriate, however, for limiting human and environmental exposure.
RECOMMENDATIONS FOR EMISSION REDUCTION AND CONTROL
In view of recent studies of MCF effects on O_ depletion, test results
indicating possible MCF-induced mutagenicity, and the report on MCF by EPA's
Carcinogen Assessment Group, Midwest Research Institute (MRI) suggests that
MCF does not belong in a classification of chemicals for which "it is not
necessary that they be inventoried or controlled." TCE, MCF, and PCE should
be considered a group and, as such, should all be subject to the same emission
controls. Recently, EPA's Office of Air Quality Planning and Standards recom-
mended that State Implementation Plans consider positive emission reduction
for all three compounds, rather than substitution of MCF for either of the
other two solvents. MRI believes use of emission control technologies for all
three compounds to be the proper approach.
Metal Cleaning Industry
Current control technology can decrease ambient air emissions by 50 to 60
percent. A large portion of the control technologies recommended by EPA for
New Source Performance Standards do not require the purchase or use of expen-
sive equipment. Significant emission reductions can be attained by employing
careful operating practices and good maintenance procedures.
A comprehensive training program must also be employed. This training
program is very significant not only for vapor degreasing operations but also
12-8
-------
for cold cleaning procedures, since many cold cleaning operations are con-
ducted by untrained and nonsupervised workers in small companies; such situa-
tions can often lead to increased human exposure and high emission losses.
Dry Cleaning Industry
MRI defers recommendations for the dry cleaning industry in view of sug-
gested new EPA guidelines for control of volatile organic emissions from PCE
systems.
Solvent Recovery and Waste Disposal
Solvent recovery techniques are currently available which, if more fully
adopted, could lead to appreciable recycling of solvent and an overall re-
duction in the quantity of unrecyclable waste. This method should be empha-
sized as an initial technique to reduce quantities introduced into the environ-
ment. For the quantity of distillation residue that still remains after
reclamation, the preferred means of disposal is incineration. Levels of toxic
or corrosive decomposition products from incineration should be maintained at
an environmentally acceptable minimum.
Contract reclamation and incineration services are generally available in
the larger metropolitan areas, so many companies will be able to readily
reclaim or discard their waste solvent. However, companies generating small
volumes of waste solvent may find no reclamation or disposal service interested
in small volumes. Additionally, large geographical areas may be without ser-
vice altogether. In order to alleviate these potential problems, an appropri-
ate federal agency could assist users (perhaps through the use of regional
offices) in finding the nearest contract reclamation or incineration services.
Water Quality
Amendments and proposed amendments for control of these three compounds
in water have been published. Currently, controversy surrounds a number of
these statutes, and new regulations are being proposed.
12-9
-------
With respect to concentrations in finished drinking water, methods for
control of chemical contaminants have been proposed by the EPA Office of Water
Supply. These proposed regulations have been debated by the water supply in-
dustry, and EPA is preparing detailed responses to resolve the issues raised
during the comment period. In view of this activity, no options are suggested
to modify the proposed regulations.
TCE was classified as a Category C Hazardous Substance in the EPA Hazard-
ous Substance Spill Program, based on aquatic toxicity (96-hr LCc-n) levels in
the 10- to 100-mg/liter range; neither MCF nor PCE was identified as a hazard-
ous substance. But the Spill Program was halted by an industry lawsuit, and
revised Section 311 rules are now ready for final internal review. If, as
indicated, the new Section 311 rules include only EPA's previous list of 299
substances, TCE will remain the only one of the three compounds designated as
a Hazardous Substance. Since MCF and PCE exhibit basically the same type of
aquatic behavior and aquatic toxicity levels as TCE, MRI feels that these two
compounds should be evaluated for inclusion in the same category.
Container Labels
All three of the subject compounds appear to pose a human health problem
in high vapor concentration. To explicate the potential danger, MRI suggests
adequate labeling of all TCE, MCF, and PCE containers. The label should state
that a high vapor concentration can cause unconsciousness or death, and that
exposure to high vapor concentrations followed by strenuous physical activity
or high levels of excitement or stress may result in heart sensitization. In-
dustrial workers with previous histories of heart problems would particularly
benefit from such labeling, as they may unexpectedly encounter high concentra-
tions of the compounds.
Dental and Medical Procedures
On the basis of health effects data derived in this study, the essential-
ity of TCE use in dental and medical procedures must be considered more closely
by appropriate agencies. Although very minor amounts are used in such procedures,
12-10
-------
TCE is introduced directly into the human body at levels considerably above
those to which the general public would normally be exposed. Alternative
materials can and are now being utilized in varying degrees in both dental and
medical applications.
Aerosol Products
MRI suggests that use of MCF or PCE in aerosol products be considered by
the appropriate agencies. As with dentistry and medicine, this area of use is
minor but also represents a mechanism for potential direct inhalation. Use
of aerosol products containing either of the two compounds could expose the
user to high concentrations, causing a decrease in manual dexterity, eye
irritation, and central nervous system effects (primarily dizziness).
The Consumer Product Safety Commission has recently announced an analysis
of PCE to evaluate alternatives for regulating the chemical as a hazardous
component of consumer products. A briefing is scheduled for March 1979.
DISCUSSION
DP. Farber: I would like to make a brief comment before we lose any of the
people that have heard this report to the outside world. I thought this
meeting was to address the 0-. depletion issue, but I didn't hear much of that
discussed in the MRI report. I understand the rest of the morning's presenta-
tions may not spend much time on that issue either. I am obviously very con-
cerned, because about a month ago I wrote to Dr. Hanst indicating that if the
subjects were to stray very far from this issue, we would like to have the
opportunity to discuss these other issues in some detail with the experts that
did the work. I was assured by Dr. Hanst in a return letter that the subject
would be confined to these issues to the best of his ability. I think your
ability is suffering, Sir, at this moment, and I am very, very upset, as you
probably can tell.
Let's put it this way. If I were the manager of a baseball team, I would
be playing this game "under protest." I guess that position will probably
have about as much impact on what is going to happen here for the rest of the
day as it does in a game played under protest! I do offer, by the way, the
services of the Dow Toxicology Department and the other independent researchers
whom we finance to discuss some of these areas of concern with whoever here
would like to have them objectively discussed.
12-11
-------
DlC1. Hanst: While further discussion on the toxicology and health effects of
these compounds may ensue, I am sure you will have the opportunity to discuss
these issues at other meetings. We have confined ourselves mainly to the
atmospheric chemistry, but the MRI study, made public for the first time,
provides us with a rather complete picture.
Dr. Heieklen: I would like to ask a series of short questions. I gather from
what you said that there are no known human effects at levels of <50 ppm.
DP. Lapp: There are none other than any mutagenic or carcinogenic effects.
Dr. He-ieklen: The human exposure levels of all three of the compounds are
<1.5 ppb?
DP. Lapp: Correct.
Dr. Heieklen: Can you tell me how much this control program is going to cost?
DP. Lapp: As it stands right now, it is very difficult to try to evaluate how
much it is going to cost a company to educate their people. Now, if you want
to make some random
DP. Heiokten: You are recommending to EPA that controls be placed on this
compound, with this information, without having any idea of what this program
is going to cost.
DP. Lapp: First of all, we are not involved in any cost study to begin with.
DP. Heioklen: I don't think you should have accepted the task, then.
DP. Lapp: Well that, unfortunately, is not your choice or mine. However, the
controls recommended by EPA for degreasers and for clothes cleaners will ob-
tain levels of roughly 50 percent if used properly, and these controls can be
effected with equipment already on the degreasers. It basically is a house-
keeping procedure. What the cost of that would be I don't know. I could
guess at it.
DP. Heieklen: Thank you.
MP. Surppenant: I would like to inquire of the data base for your comment on
carcinogenicity or mutagenicity, particularly for MCF.
DP. Lapp: As you well know, as a toxicologist, I don't get involved in the
health effects.
MP. Suppreliant: But you certainly did this morning.
Dr. Lapp: I know. Much of that refers to what Dr. Mazaleski will talk about.
The mutagenicity effects were in the report, and I would have to look that
information up for you, to be honest about it. The report showed that TCE and
MCF were used.
12-12
-------
Mr, Surprenant: Would you say that they are in any way comparable by levels
of ppb or yg? What was your per-kilogram? Are they in any way in an environ-
mental comparison to the measure of value?
Dr. Lapp: If I recall correctly, they are.
Dr. Eeiaklen: You say now that the Ames Test has shown mutagenicity and that
the compound is therefore mutagenic; but the Ames Test is a screening test.
Dr. Lapp: That is right, and it is one of eight sets.
Dr. Ee-iGklen: I am familiar with some of these toxicological tests, and this
test targets a compound for further investigation. Although mutagenic activity
was observed in bacteria, those test results do not necessarily apply to hu-
mans. It is a compound that should be further investigated because, in fact,
some researchers do suspect a problem. But to extrapolate from the Ames Test
that this compound is going to be mutagenic in humans is unwarranted. Such a
conclusion is not warranted from Ames Test data.
Dr. Lapp: We did not say that it was "mutagenic." We said it was "potentially
mutagenic."
Dr. Fisher: It is not a ridiculous extrapolation at all.
Dr. Eeiaklen: I didn't say it was a "ridiculous extrapolation."
Dr. Fisher: The Ames Test, as a test for mutagenesis, is quite good. For
carcinogenicity, its value is only as a screen.
Dr. Eeioklen: For mutagenicity, there are false positives and false negatives.
It is a good screening test, or you wouldn't use it, but it is still a screen-
ing test.
Dr. Fisher: For a direct test in mutagenesis.
Dr. Eeioklen: In bacteria.
Dr. Bower: Let me get back to the topic of the conference. Now, you quoted
U.S. production figures for these various materials. From the standpoint of
modelers who want to model stratospheric effects of materials, estimates of
worldwide production (with estimates of errors) are generally much more useful
than-U.S. production figures. Yesterday, Dow quoted figures on MCF excluding
Soviet production. Now, I am sure that Dow can arrive at a much better esti-
mate of Soviet production than can the average modeler. From that standpoint,
then, I think it would be very useful to have an estimate of worldwide produc-
tion.
12-13
-------
METHYL CHLOROFORM AND ITS STABILIZERS
Stanley C. Mazaleski
U.S. Environmental Protection Agency
Washington, D.C.
INTRODUCTION
The main topic of this paper is the ongoing U.S. Environmental Protection
Agency (EPA) investigation into health effects of methyl chloroform (1,1,1-
trichloroethane, CH CCl , MCF) and a number of MCF stabilizers, particularly
the dioxane/MCF and dioxolane/MCF mixtures of The Dow Chemical Company and PPG
Chemical Industries, respectively. The viewpoints presented herein are the
author's scientific opinion, not the official position of EPA. Specific EPA
policy has not yet been finalized, because EPA's Office of Toxic Substances is
currently reviewing very recent and pertinent data on MCF. Any official EPA
statements could have significant impact on the chemical, which is extremely
high volume and in common use worldwide.
In December 1978, the author prepared an extensive 103-page report on MCF
and MCF stabilizers as Section 4(f) of the Toxic Substances Control Act (TSCA)
Support Document for 1,1,1-Trichloroethane (Methyl Chloroform) (Mazaleski
1978). The report contained 88 references, including a citation of the Dow
study on MCF inhalation in rats. This very critical Dow study appears defi-
cient in protocol and methodology. A preliminary assessment of the Dow study
issued on January 17, 1979 by EPA's Carcinogen Assessment Group for MCF (U.S.
Environmental Protection Agency 1979) is summarized below. Also outlined are
deficiencies in the studies of PPG Chemical Industries on dioxolane as an MCF
stabilizer. Stabilizers are a source of concern; in January 1978 dioxane
13-1
-------
itself was shown by the National Cancer Institute (NCI) to be carcinogenic in
two animal species under test conditions. This research supports work per-
formed several years earlier by Argus et al. (1973). Difficulty was encoun-
tered with the criteria document on dioxane by the National Institute of
Occupational Safety and Health (NIOSH); although published in September 1977,
the NIOSH report did not include a major metabolite, p-dioxane-2-one, which is
reported to be 8 times more toxic than dioxane (itself a moderate carcinogen).
BACKGROUND AND OVERVIEW
MCF is used primarily as a cleaning or degreasing agent for metals. It
is increasingly used as a substitute for chlorinated ethylenes, such as tri-
chloroethylene (C_HCl , TCE), which is carcinogenic. An estimated 630 million
pounds of MCF were produced in the U.S. in 1976. At least 300 million pounds
were dispersive uses, primarily for metal degreasing and for aerosols. MCF
escapes into the environment primarily into the air, and may affect the ozone
(O ) layer in the upper atmosphere. NIOSH judges that 3 million workers may
be exposed.
Methylene chloride (dichloromethane, CH2C12) is soluble in water, TCE is
less soluble, and MCF is insoluble. The boiling points of the three compounds
are: 40° C (CH Cl ), 74° C (MCF), and 87° C (TCE). The most harmful effects
^ ^
of MCF involve central nervous system problems, including anesthesia, dis-
turbed equilibrium, and impairment in perceptual speed and dexterity. MCF
cardiovascular effects include decreased blood pressure, bradycardia, and
hypertension. Exposure can also cause inflammatory changes in the lung, fatty
changes in the liver, and damage to the kidney. MCF is eliminated from the
body in unaltered form via the lung.
Dioxane and MCF may have an effect on birth anomolies. The Dow Chemical
Company sent data on impurities to EPA in late December 1978, and (under the
Freedom of Information Act) requested from EPA's Carcinogen Assessment Group a
document entitled "Dioxane: A Critique." This particular report (Mazaleski
and Schumacher 1978) indicates that birth defects may occur from MCF/dioxane
or from MCF itself. There is some concern that MCF, under conditions of
13-2
-------
storage or usage, may degrade to vinylidene chloride (C H Cl ), a suspected
carcinogen. Louis Schlossberg of Detrex Chemical Industries has suggested
that these conditions are presence of water, iron, zinc, aluminum, or chloride
salts of these metals. Thus, gray areas concerning hazards of MCF do exist.
In April 1978, the Interagency Testing Committee placed MCF on its priority
list for testing. EPA has been working closely with The Dow Chemical Company
and Detrex Chemical Industries to produce a document on possible regulation of
MCF; the following data are taken from Mazaleski (1978).
MCF was introduced to industry with an improved inhibitor system that
provides better corrosion protection and stability under vapor degreasing
conditions, and enables it to compete with TCE in vapor degreasing applica-
tions. Stabilizing grades of MCF are made by the addition of 3 to 8 percent
stabilizer composed of various chemical constituents, which are reported to
include nitromethane and N-methylpyrol, 1,4-dioxane, butylene oxide, 1,3-
dioxolane, and secondary butyl alcohol. Commercial products are reported to
contain amounts of certain stabilizing materials, including p-dioxane (a
stabilizer MCF additive sold by Dow). In 1978, NCI reported dioxane to be
quite carcinogenic in test animals. Dioxolane, a stabilizer MCF additive
utilized by PPG Chemical Industries, is currently being tested to determine if
it is also carcinogenic (Mazaleski 1978; Bell 1978).
According to oncogenicity information obtained by the EPA Carcinogen
Assessment Group in January 1979, MCF is a suggested carcinogen. However, the
Group categorizes different levels of carcinogenicity, and no data sufficient
to calculate a human risk assessment exist. In-vitro tests (the Ames test and
a cell transformation test) suggest that MCF is less potent than CH_C1_ and
similar in potency to TCE (U.S. Environmental Protection Agency 1979).
No studies adequately assess the carcinogenic potential of MCF. A life-
time animal bioassay at the maximum tolerated dose (MTD) is required to charac-
terize carcinogenic potential; such a gavage study is in progress at NCI for
rats and mice. An earlier NCI study (National Cancer Institute 1977) was
inconclusive, due to poor survival of treated animals. An inhalation study in
13-3
-------
rats by Dow (Quast et al. 1978) showed no evidence of carcinogenic!ty, but the
doses were given for only half of the animal's lifetime, and the highest dose
did not appear to be an MTD. Comparison of doses in the Dow study and the
early NCI study shows that the MCF toxic dose by gavage is less than the toxic
dose by inhalation.
With regard to a Manufacturing Chemists Association TCE inhalation study
in rats and mice (Bell 1978; Van Horn 1978, 1979), the precision of controls
seems open to serious question. In fact, the data from this $500,000 study
appear unusable, disallowing valid conclusions linking carcinogenicity with
TCE.
IN-VITRO STUDIES
In application of the Ames Test to these compounds, it is necessary that
bacteria be exposed to the compound of interest in a desiccator or in liquid
suspension. If this precaution is not taken, false negatives may result.
MCF was weakly mutagenic in strain TA100 when the Ames Test was conducted
in desiccators (Simmon et al. 1977). A measured volume of MCF was placed in
an open dish in the bottom of a desiccator, and the open petri dishes contain-
ing Salmonella strains were placed in the top of the desiccator. Exposure
occurred for a set number of hours. MCF was less potent than CH Cl and about
£ £
equal in potency to TCE, and was mutagenic in the presence and absence of
metabolic activation using Arochlor 1254-induced rat liver. A dose response
was evident, but the number of revertant colonies was only twice control
plates at the highest dose tested: 750 yl in an open dish in a 9-liter desic-
cator.
MCF was also tested by Litton Bionetics for Dow, but the methods and data
were not available for evaluation. In other data from Dow, however, MCF with
metabolic activation gave a positive response in strain TA1535, an equivalent
response in TA1537, and a negative response in TA1538 (Farber, personal com-
munication). Strain TA100, which is a more sensitive derivative of TA1535,
was not tested. These results are consistent with the more careful study by
13-4
-------
Simmon et al. (1977). Henschler et al. (1977) reported that MCF was negative
in TA100, but these authors used no precaution to insure that the bacteria
were actually exposed to the highly volatile compound.
MCF altered cells in an in-vitro test of cell transformation performed
using Fischer rat embryo cell line F170 (Price et al. 1978). When injected,
these cells produced fibrosarcomas in 8 out of 8 rats. In this test, MCF was
less potent than CH Cl and similar in potency to TCE.
IN-VIVO STUDIES
National Cancer Institute Study
The 1977 NCI carcinogenesis study in rats was inconclusive due to poor
survival of treated animals, and the NCI Clearing House of Environmental Car-
cinogens resolved that carcinogenicity cannot be determined at the present
time (National Cancer Institute 1977). In this 1977 study, Osborne-Mendel
rats were treated by gavage with both 750 mg/kg and 1500 mg/kg of MCF in corn
oil 5 times/week for 78 weeks. The rats were observed an additional 32 weeks,
with the experiment ending at 110 weeks. Both males and females were used,
with 50 of each sex at each dose and 20 untreated females. The study was in-
adequate because only 3 percent of treated rats survived the length of the
experiment. There appeared to be an anticarcinogenic effect on fibroidadenoma
of the breast.
The 1977 NCI bioassay also employed B6C3F hybrid mice. The study used
20 mice of each sex in the control group and 50 of each sex at each treatment
dose. The time-weighted average dose was 2807 mg/kg and 5615 mg/kg. The mice
were treated by gavage 5 days/week for 78 weeks and observed for another 12
weeks, for a total of 90 weeks in the experiment. Only 31 percent of treated
animals survived to the end of the experiment, and treated animals gained less
weight than controls. In male mice, an excess of tumors seemed to occur in
the liver (1 tumor among control animals and 7 tumors among treated animals),
but this increase was not statistically significant.
13-5
-------
As mentioned previously, a 2-yr carcinogenesis bioassay by gavage in mice
and rats is underway at NCI.
Dow Chemical Company Study
The Dow study (Quast et al. 1978) treated groups of Sprague-Dawley rats
by inhalation under conditions similar to those experienced by workers (6
hours/day, 5 days/week, for >l/2 lifetime). The rats were treated 12 months
and observed until death or 31 months. The dose of 875 and 1750 ppm was 2.5
and 5 times the threshold value of 350 ppm, respectively. Total tumor inci-
dence in treated animals was similar to that in controls (Table 13-1).
TABLE 13-1. DOW STUDY
Number of
Animals
Dosage
Control
875 ppm
1750 ppm
Male
189
91
93
Female
189
92
93
Total
Neoplasms
Male
200
77
103
Female
561
246
300
Neoplasms minus
Mammary Tumors
Male
183
67
91
Female
240
71
79
When tumors at each site were examined by tumor type, both benign and
malignant, there were 8 differences between control and treated animals at the
p < 0.05 level (Fischer Exact Probability Test). Decreased tumor incidence
accounted for 7 of these differences; 1 was an increase in ovarian gradulosa
cell tumors in females at a dose of 875 ppm. (No tumors were detected in 189
controls; 3 were detected in 33 treated at 875 ppm; and 2 were detected in 82
treated at 1750 ppm.) Since no pattern was consistent between those levels,
and since 149 separate comparisons were made, p = 0.05 was not rigorous enough.
The increases and decreases in tumor incidence were most likely due to random
fluctuations.
The Dow study suffers from two (in the author's opinion, three) drawbacks.
The first drawback is that animals were treated for only 12 months, rather
13-6
-------
than for a lifetime. Normally 18 to 24 months is the ideal time frame for a
chronic study; a 12-month study cannot be called a "chronic" study. NCI does
not use a 12-month protocol, and it is difficult (if not impossible) to com-
pare results when animals are exposed for this short period. A second problem
with the Dow study is the question of whether or not the MTD was used. Finally,
this author considers as a third problem the fact that only one species was
tested. NCI used both rats and mice, and the mice tended to be the more sus-
ceptible species. Finding no mouse carcinogenicity in the Dow experiments
would possibly have indicated more strongly that MCF is not carcinogenic; no
mice were included, however.
When compared to untreated animals, treated animals in the Dow study were
no different in body weight, terminal organ weight, or mortality. The only
sign of toxicity was an increased incidence of focal hepatocellular altera-
tions in female rats at the highest dosage. Since treated rats showed little
sign of toxicity, it is instructive to compare the dose to that in the NCI
study, where only 3 percent of rats survived a 110-week study.
Comparison of Doses in NCI and Dow Studies
The dose in the NCI study was 750 and 1500 mg/kg, 5 days/week. The
equivalent dose for male rats in the Dow study was 602 and 1204 mg/kg, 5
days/week. The calculations for these figures are as follows:
Dose was 875 and 1750 ppm, 6 hours/day, 5 days/week. To convert ppm to
mg/m :
1.2 x
mw air
1750 x 1.2 x ~~ = 9734 mg/m3
<£o. o
To average the 6-hour exposure over a day:
9734 mg/m3 x ^"" = 2434 mg/m3/day
13-7
-------
A 260-g rat breathes 111 ml air/minute, which is equivalent to:
3 3
111 cm 60 minutes 24 hours 1m ,. . .. 3.,
. x x . x _ = 0.16 m /day
minute hour day ,6
10 cm
2/3
The amount of air breathed increases as (weight)
Dow rats (male) averaged 500 g. Therefore, the estimated amount that rats
in the Dow study breathed is :
0.16 m3/day x |^ 2/3 = 0.247 m3/day
Since a 500-g rat breathed 0.247 m /day of air containing 2434 mg/m /day,
the dose rate was:
0.247 m3 2434 mg 1 . .. . ., .,
day X -3^ x O^Tkg- = 1204 ^Ag/day
m /day
Hence a dose by gavage of 750 mg/kg/day, 5 days/week, caused severe
toxicity in the NCI study, but a dose by inhalation of 1204 mg/kg/day, 5
days/week, caused little or no toxicity in the Dow study. Perhaps MCF is
less toxic by inhalation than by gavage, or perhaps a dose given at one point
in time (gavage) is more toxic than a similar dose given over 6 hours (inhala-
tion). Perhaps real strain differences in sensitivity do exist, or perhaps
some technical problem caused the dose actually inahaled by rats in the Dow
study to be less than intended. These alternative explanations should be
explored, since the major route of human exposure is inhalation.
Lifetime equivalent doses, correcting for the fact that rats were exposed
5 of every 7 days and correcting for both the 18-month exposure in the NCI
study and the 12-month exposure in the Dow study, are 402 and 804 mg/kg/day
for the NCI study, and 215 and 430 mg/kg/day for the Dow study.
PPG Chemical Industries Study
EPA has also reviewed a study conducted for PPG Chemical Industries which
was received on December 8, 1978 (Bell 1978). None of the exposure studies in
13-8
-------
this 1976 investigation meets the criteria for an oncogenicity study, and none
of the submitted data are adequate for making an assessment of potential haz-
ard and potential risk (Seifter 1979).
In a gavage study using dioxolane, PPG Industries found the MTD in drink-
ing water to be 1 percent. Only 0.1 percent MCF was used, however, which does
not meet the weight loss criterion. The depression of water consumption at
0.5 percent may have been due to systemic toxicity or (more likely) to objec-
tionable taste (Seifter 1979).
CONCLUSIONS
While it points out the weaknesses of the Dow study, the report of the
EPA Carcinogen Assessment Group contains weaknesses itself. The comparisons
on page 7 of the report (NCI studies compared with the Dow study) have no
meaning in the absence of pharmacokinetic data; and without tissue levels of
MCF by the two routes of introduction into the organism, a statement in the
report such as occurs on page 8 (bottom three lines) cannot be made (Seifter
1979).
The Dow study by Quast et al. (1978) is not a valid negative carcino-
genicity result. The study did not meet the requirement of administering the
MTD so that the animals exhibit unequivocal signs of toxicity (usually 10 per-
cent less in body weight). Hepatocellular changes are not an established
criterion for toxicity; they occur spontaneously without known treatment or
exposure to vapors. The authors conclude that the changes "may" have been due
to treatment (page 15). They also conclude that the incidence "may" have
"possibly" increased (page 18). Such uncertainty is not solid support for the
statement that the MTD was administered (Seifter 1979).
EPA is concerned with health effects of MCF, and will soon publish
Section 4(f) of the TSCA Support Document for 1,1,1-Trichloroethane and the
most recent data available on the subject. EPA appreciates the cooperation it
has received from industry in the problem of MCF and MCF stabilizers.
13-9
-------
REFERENCES
Argus, M. F., R. F. Sohal, G. M. Bryant, C. Hoch-Ligeti, and J. C. Arcos. 1973.
Dose-response and ultrastructural alterations in dioxane carcinogenesis.
Influence of methylcholanthrene on acute toxicity. Eur. J. Cancer 9:237-243.
Bell, Z. (PPG Chemical Industries). 1978. Personal communication to S. C.
Mazaleski, including 1976 report on dioxolane. December 8.
Henschler, D., E. Eder, T. Neudecker, and M. Metzler. 1977. Carcinogenicity
of trichloroethylene: Fact or artifact? (Short Communication). Arch.
Toxicol. 37:233-236.
Mazaleski, S. C. 1978. 1,1,1-Trichloroethane. Section 4(f), Toxic Substances
Control Act Support Document (draft). U.S. Environmental Protection Agency,
Washington, D.C., December.
Mazaleski, S. C., and H. S. Schumacher. 1978. Dioxane: A critique (draft).
To be published as EPA 560/2-78-005, Office of Toxic Substances, U.S.
Environmental Protection Agency, Washington, D.C.
National Cancer Institute. 1977. Bioassay of 1,1,1-trichloroethane for pos-
sible carcinogenicity. Carcinogenesis Technical Report Series No. 3,
National Cancer Institute, Bethesda, Maryland.
Price, P. J., C. M. Hassett, and J. I. Mansfield. 1978. Transforming activities
of trichloroethylene and proposed industrial alternatives. In Vitro 14:
290-293.
Quast, J. F., L. W. Rampy, M. F. Balmer, B. K. J. Leong, and P. J. Gehring.
1978. Toxicologic and carcinogenic evaluation of a 1,1,1-trichloroethane
formulation by chronic inhalation in rats. Dow Chemical Company, Midland,
Michigan.
Seifter, J. (U.S. Environmental Protection Agency). 1979. Personal communica-
tion to S. C. Mazaleski. February 2.
Simmon, V. F., K. Kaukanen, and R. G. Tardiff. 1977. Mutagenic activity of
chemicals identified in drinking water. In Progress in Genetic Toxicology
(I. D. Scott, B. A. Bridges, and F. H. Sobels, eds.). Elsevier.
pp. 249-258.
U.S. Environmental Protection Agency (Carcinogen Assessment Group). 1979.
Preliminary carcinogenic risk assessment for methyl chloroform and methylene
chloride. U.S. Environmental Protection Agency, Washington, D.C. January 17.
Van Horn, J. (Manufacturing Chemists Association). 1978. Personal communica-
tion to S. C. Mazaleski. December.
13-10
-------
Van Horn, J. (Manufacturing Chemists Association). 1979. Personal communica-
tion to S. C. Mazaleski, including Final Report of Audit Findings of the
Manufacturing Chemists Association (MCA) Administered Trichloroethylene
(TCE) Chronic Inhalation Study at Industrial Bio Test Laboratories, Inc.
(IBT), Decatur, Illinois, November 1978.
13-11
-------
METHYL CHLOROFORM IN PERSPECTIVE
Louis Schlossberg
Detrex Chemical Industries, Inc.
Detroit, Michigan
INTRODUCTION
A number of specific issues must be addressed in discussing increased use
of methyl chloroform (1,1,1-trichloroethane, CH CC1 , MCF) and its impact on
stratospheric ozone (0 ) depletion. While the issue of the Conference on
Methyl Chloroform and Other Halocarbon Pollutants is environmental, it is
difficult not to discuss health effects when dealing with environmental im-
pact, since the two areas are inherently linked. Surely, even very small
concentrations of some chemical compounds are adverse to health.
Detrex applies MCF as well as the four other halocarbons in solvent vapor
degreasing, a vital process that affects the metalworking output of the United
States and the industrialized nations of the world. In the solvent vapor de-
greasing process, any particular industrial part or object may be dipped or
agitated in solvent, or sprayed with solvent, as long as it is then subjected
to boiling solvent vapor as the final step in cleaning. The virgin vapor
contacts the object, condenses, and washes away soils.
In contrast, cold cleaning may take place in a cold (or even warm) condi-
tion; parts may be sprayed, dipped, or agitated in the solvent, but they are
not subjected to boiling vapors. Certainly, use of MCF has grown significantly
for cold cleaning as well as for solvent vapor degreasing.
14-1
-------
Detrex pioneered solvent vapor degreasing in the U.S. 49 years ago through
development of stabilizers for chlorinated hydrocarbon solvent and machines in
which such solvents are used. Detrex was a large manufacturer of trichloro-
ethylene (C2HC13/ TCE) and perchloroethylene (C2Cl , PCE) from 1946 to 1972;
though Detrex ceased all halocarbon manufacture in 1972, the company continues
to sell substantial quantities of the five chlorinated hydrocarbon solvents
for degreasing. Today, Detrex remains one of the largest U.S. manufacturers
of solvent vapor degreasing machines; these machines employ all five of the
chlorinated hydrocarbon solvents.
Detrex believes quite strongly that new State Implementation Plan (SIP)
regulations must not force or induce substitution of MCF for TCE through ap-
plication of rigorous rulemaking or reasonably available control technology
(RACT) for TCE that does not place similar constraints on MCF. The company
does agree with the position of the U.S. Environmental Protection Agency (EPA)
that SIP's should require positive control of volatile organic chemical (VOC)
emissions, including MCF. Unquestionably, all of these solvents, as well as
methylene chloride (dichloromethane, CH-Cl^) and fluorocarbon-113 (CC1_FCC1F ,
FC-113), are valuable to industry.
The RACT concept is the proper and the responsible approach that, with
coordination, should be applied at the same time and in the same manner to the
three major chlorinated degreasing solvents MCF, TCE, and PCE. MCF should
not be exempt from regulation, and the user should not be induced or forced to
substitute it for TCE or PCE, because a situation may be created that is more
hazardous to the environment and population than now exists.
From an environmental standpoint, TCE breaks down in the troposphere
within ~8 to 12 hours and PCE within ~48 hours, but tropospheric MCF is stable
for roughly 8 to 10 yr. As shown in Table 4-4 of Singh et al. (this volume),
the tropospheric residence time of MCF has been estimated by several investi-
gators: 5 to 10 yr (Lovelock); 8 yr (McConnell and Schiff); 12 yr (Chang and
Penner); 9 to 12 yr (Krasnec); ~6 yr (Rowland); 3 yr (Neely and Plonka); 10 yr
(Crutzen and Fishman); 7 ± 1 yr and 8 to 11 yr (Singh). The present estimate
of 8 to 10 yr is compatible with these estimates.
14-2
-------
Until recently, the Clean Air Act (CAA) mandated chemical controls only
within our troposphere. Prompted by new data on stratospheric 0 and by the
1977 CAA Amendments, EPA clearly is directed to report and investigate deple-
tion of stratospheric O .
TROPOSPHERIC PHOTOCHEMICAL OXIDATION VS. STRATOSPHERIC OZONE DEPLETION
Two reports are pertinent. To begin, EPA (U.S. Environmental Protection
Agency 1979) concisely states the reasons for controlling photolytic materials
or trying to reduce photochemical oxidation:
Even at relatively low concentrations (in the troposphere)
ozone has been shown to aggravate respiratory problems in sensi-
tive individuals, to cause discomfort and interfere with normal
breathing of healthy persons under conditions of stress. Con-
trolled animal studies show ozone exposure has the potential for
increasing the risk of respiratory infection and other long-term
chronic effects...
In a second report (EPA Journal 1978) , tropospheric photochemical oxidant
problems are evaluated against stratospheric O depletion problems. The re-
port states concern over stratospheric O., depletion, and reference is made to
aerosol can regulations. Next, the report states that reduction of this
stratospheric O could
...cause a substantial rise in the incidence of skin cancer. The
layer of ozone now acts as a shield against biologically harmful
ultraviolet radiation from the sun, and scientists fear that even
a small percentage loss of this screen will have serious health
effects around the world. In 1975 a Federal task force on the
Inadvertent Modification of the Stratosphere [IMOS]...
and another group, the Interdepartmental Committee on Atmospheric Sciences
(ICAS) of the National Science Foundation (NSF)
...warned that not only could skin cancers in humans increase but also
other damaging biological and agricultural effects might occur.
A Coordinating Committee met in Bonn from November 28 through December 1 to
review the O layer situation on a worldwide basis.
14-3
-------
Some of the figures quoted below are meaningful, because the deleterious
effects from increases of O and other photochemical oxidants in our tropo-
sphere can be related to those from reduction of stratospheric O.,.
In 1976 a National Academy of Sciences (NAS) study, funded by
EPA and several other agencies, had estimated that depletion of
the world's ozone layer by fluorocarbons could range from 2 to 40
percent, with the most probable value at about 7 percent.
Actually, the upper limit estimated by NAS was ~20 percent, not 40 percent as
indicated in this report.
In December, 1977, in its report to the Congress pursuant to the Clean
Air Act Amendments of 1977, NAS stated: 'As a result (of new knowledge)
the estimated seriousness of...ozone reduction has been roughly doubled'
to about 14 percent.
The general feeling now is that the level of reduction of stratospheric 0 by
fluorocarbons might be in the range of 14 to 17 percent more precisely, 16
or 17 percent.
More recently, a World Meteorological Organization symposium
in Toronto last June heard fresh estimates by several experts of
an 18 percent depletion. It is estimated that an increase of ap-
proximately 4 percent in the incidence of non-melanoma skin cancers
among Caucasians is predicted for each 1 percent reduction in aver-
age ozone concentrations, with a disproportionately greater
increase in cancer expected for higher percentages of reduction
in ozone levels. Non-melanoma skin cancers rarely cause death
but are considered serious and should not be neglected.
(Average 0. concentrations are discussed later, to clarify what can be termed
popular fallacies, myths, or misconceptions.)
People with fair complexions are more prone to skin cancer than the
general population. Although basal or squamous cell carcinomas are not as
deadly as melanoma, they cannot be considered innocuous.
There are now about 300,000 cases of non-melanoma skin cancers
annually in the United States, according to the National Cancer
Institute [NCI]. If the currently estimated most probable ozone
reduction value prevails, it implies more than 210,000 additional
annual cases of non-melanoma skin cancer.
1.4-4
-------
Next is a discussion of melanoma, an extraordinarily serious type of
cancer that tends to metastasize. Incidence of fatality due to melanoma is
somewhat analogous to that due to breast cancer in terms of rate of cure (or
more appropriately, perhaps, lack of cure). Melanoma is very deadly.
The incidence of melanoma, a much more serious disease, is
about 1 to 3 percent (about 6,000 cases annually in the United
States) of all skin cancers. Its cause may not be solely ultra-
violet exposure, but this is considered a factor.
Evidence has been shown that skin cancers seem to prevail on those por-
tions of the body most exposed to the sunlight. As was mentioned, the rate of
incidence of melanoma in the United States is definitely increasing, and can-
cer authorities believe that melanoma is linked with ultraviolet-B (UV-B)
radiation. NCI has confirmed that certain types of skin cancers, including
melanoma, can occur because of excessive exposure to UV-B radiation.
Depletion of the ozone layer also could cause other effects
such as climate changes; effects to some plants and animal species;
disturbances in aquatic and land ecological systems; alteration
of the stability and effectiveness of farm chemicals such as pesti-
cides and fertilizers; increases in eye cancer in livestock, and
reduction in the yield of some crops, especially in areas of marginal
production, according to the IMOS report.
Granted, the population could wear wide-brimmed hats, if so inclined, but
could all of the animals and crops be covered with Mylar, a suggestion voiced
recently? The effect on agriculture is a concern of extraordinarily serious
nature. Photochemical oxidation is certainly noxious and should be reduced
and eliminated where at all possible, but the solution of this problem must
not create another of perhaps even greater magnitude. Considering the options,
the better choice would be to have no significant reduction in stratospheric
O and to tolerate perhaps some increase in tropospheric O . Most experts in
these areas probably agree with this statement.
RACT, RULE 66, AND METHYL CHLOROFORM
Forziati (this volume) has reported the various measurement tasks which
have been assigned to predict the effects of 0 depletion; these tasks involve
14-5
-------
modeling and monitoring UV-B radiation versus reductions in O.,, climatic ef-
fects, biological effects, health effects ranging from sunburn to melanoma,
integrated assessments, and so forth. Further research in these areas is
necessary, but a degree of understanding about O_ depletion does currently
exist. The ICAS Report (Atmospheric Sciences Interdepartmental Committee
1975) indicated that
...experiments exposing a variety of organisms including
agricultural and wild plant species, phyto-plankton, insects,
toad embryos and larvae to elevated UV-B irradiance have been
made. Studies of whole organisms suggest that many are sensi-
tive to UV-B radiation intensities now reaching the earth's
surface. Avoidance of UV-B radiation and the use of molecular
repair systems are important under present conditions. Many
organisms have little reserve capacity to repair or to tolerate
UV-B irradiation higher than that of current levels. Responses
by different organisms exhibit a wide range of sensitivity.
While undoubtedly additional measurements should be performed, signifi-
cant work already accomplished points to some very deleterious effects of O.,
depletion. The concept that Detrex espouses, and recommends to all of the
regulatory agencies and government research agencies, is that photochemical
oxidation in our troposphere must be limited; nevertheless, the limitations
set must not create an even more serious problem by increasing stratospheric
O depletion.
Very simply, in solvent vapor degreasing, replacement of TCE (surely
photolytic in the troposphere) by MCF (nonphotolytic in the troposphere) would
allow very high escalation of O depletion to continue. Unfortunately, the
very resistance of MCF to tropospheric photolysis results in its long (8 to 10
yr) atmospheric residence time, permitting it to diffuse in all directions and
to destroy stratospheric O.,. So, the substitution approach is an extraordi-
narily poor one, verging on irresponsibility. Users will be forced to sub-
stitute MCF for TCE because SIP's currently being prepared exempt MCF while
controlling TCE by RACT. Instead, all chlorinated degreasing solvents should
be controlled by RACT; the EPA guidelines procedure stressing control of all
VOC's is the best approach.
14-6
-------
If RACT is implemented for all of these solvents, all emission levels
will be reduced substantially, and both stratospheric O reduction as well as
tropospheric 0 formation (as related to these solvents) will be controlled.
Indeed, accommodating RACT is relatively simple: the mere use of a cover (sup-
plied by every manufacturer) on an open top degreaser will reduce emissions by
at least 25 to 30 percent.
Simple, readily available control methods or technologies recommended by
EPA for RACT are: chillers, carbon adsorption, higher freeboard ratios, mere
replacement of the unit cover on top of the degreaser, and assurance that well-
trained machine operators will successfully handle problems as they arise.
These are straightforward, commonsense remedies that have significant impact
on reducing emissions; RACT control in these areas is the appropriate logical
or scientific choice over substitution of solvents in solvent vapor degreasers.
Rule 66 is the well known Los Angeles ordinance that was designed to
reduce photochemical oxidation. Since Rule 66 originated in 1966, many other
locations throughout the country have enacted similar ordinances. These
adopted rules are rather onerous, involving stringent emission control to a
maximum of 40 Ib/day or less. Rule 66 regulations are so drastic that, given
the alternatives of trying to meet them or substituting with an exempt sol-
vent, the decision has invariably been to substitute with exempt solvents,
including PCE, MCF, and FC-113. The great preponderance of substitution has
been with MCF, and much less with fluorocarbons and PCE.
MCF has been the overwhelming substitute material for TCE because it can
be used quite interchangeably with that chemical in the solvent vapor degreas-
ing process. The vapor pressure of MCF is roughly twice that of TCE, but its
boiling point is similar. Its solvency for the various soils on industrial
parts to be cleaned is very close to that of TCE, and it can be used in the
same machines as TCE (with some possible minor operating adjustments). All of
these factors have been conducive to substitution. Because of Rule 66, then,
the roles of the two solvents have reversed dramatically. Some years ago, TCE
represented ~95 percent of all of the solvent usage in solvent vapor degreas-
ing. Today MCF is by far the dominant solvent in solvent vapor degreasing.
14-7
-------
In 1966, 4.82 x 10 Ib of TCE were produced and emitted in the U.S. Since
little or no chemical usage exists for TCE, the amount produced and sold is
Q
the amount emitted into the atmosphere. TCE production dropped to 2.92 x 10
Ib in 1977, purely because of the restraints upon it.
The converse situation can be seen by a review of MCF production in the
U.S. during this period. MCF production is also essentially synonymous with
emission, except for a small amount (~4 to ~5 percent of total MCF produced)
used in the manufacture of vinylidene chloride (C0H Cl ). In 1966, 2.43 x
8 8
10 Ib of MCF were produced and essentially emitted; in 1977, 5.97 x 10 Ib of
MCF were emitted.
QUESTIONS AND MISCONCEPTIONS
In this section, major questions and misconceptions concerning strato-
spheric 0 depletion are answered in order to clarify the MCF issues.
If the O layer varies naturally, why is its possible reduction by a
few percent (or even 10 to 15 percent) by halocarbons and fluorohalocarbons of
such concern?
Whatever the natural level and variations of 0., are, if the annual aver-
age 0 level across the globe is reduced a few percent, the world is extraor-
dinarily impacted by increased UV-B radiation. Stratospheric O is apparently
running at a rather high level now: reduction may not appear to be a problem.
But it must be borne in mind that the current fairly high level of sunspot
activity is predicted to end in the early 1980's. At that time, a tremendous
reduction in stratospheric O^ may take place, causing international concern
concern that should be developed today.
Why worry about a little change in UV-B radiation when exposure is
greater as one approaches the equator?
The above argument answers this myth. The world is confronted by an
average change, which is an extraordinary change.
14-8
-------
Placing MCF in degreasing machines as a substitute for TCE will
produce lower emissions.
The vapor pressure of MCF is about twice that of TCE. The vapor pressure
of MCF at 20° C is 104.5 mm Hg; for TCE, it is 57.8 mm. If there were an es-
cape route in the machine for vapor, and the contents were idling and not boil-
ing, chemistry and physics indicate that more MCF than TCE would be emitted to
the atmosphere. However, either MCF or TCE at its boiling point would be at
atmospheric pressure, and differences in loss would not be too significant.
Little or no further substituting of MCF for TCE will take place if
RACT is applied to TCE only, because MCF is more expensive than TCE.
The difference in cost between TCE and MCF is actually only a few cents
per pound. The reversal in roles of these two solvents due to the substitu-
tion criterion during the past 12 yr indicates clearly that replacement will
Q
continue. Today in the U.S., some 2.9 to 3 x 10 Ib of TCE are still being
used in the solvent vapor degreasing process. If RACT control is applied to
TCE while MCF remains exempt, MCF could be substituted for this remaining
amount of TCE, meaning that another 3
the solvent vapor degreasing process.
g
amount of TCE, meaning that another 3 x 10 Ib of MCF would be emitted from
The Occupational Safety and Health Administration (OSHA) has pub-
lished emissions limits, and if OSHA places a limit of 100 ppm on TCE, 350 ppm
on MCF, 100 ppm on PCE, etc., obviously the emissions are controlled and the
environment is protected. Why is any additional emissions control necessary?
The answer depends on the definition of the environment. Will it be
defined as the surroundings of the worker, the plant as a whole, or ambient
air? Consider the problems existing before the interagency groups began
coordinating their functions; for example, to meet OSHA requirements for
occupational exposure, the solvent vapor could merely be blown from the plant
into the atmosphere by a high-volume exhaust fan.
14-9
-------
Nature has been coping with O -destroying processes for billions of
years, and its self-balancing mechanisms have enabled the atmosphere to cope
with change. Human activities are far too puny to seriously harm such a large
and resilient system.
This myth is based on blind faith; nature is not infinite in its ability
to overcome massive abuse, particularly in such an extraordinarily sensitive
area as stratospheric O depletion.
More atmospheric data are necessary, as well as more advanced pro-
cedures; until more concrete evidence of O depletion is gathered, preventive
actions should be withheld.
The problem with this attitude is that action is deferred when action is
indicated. More testing may certainly be desirable, but the real need here is
to eliminate an exemption for a material which is clearly a hazard environ-
mentally and toxicologically; MCF must be controlled by RACT, as well as TCE
and PCE, the two other major compounds used in the solvent vapor degreasing
process. Detrex supports continued use of all three solvents, but with re-
duction of all emissions through RACT.
CAPACITY AND PREDICTED USE OF METHYL CHLOROFORM
Although accurate production statistics exist for 1978, 1977, and 1976,
the ~6 percent growth of MCF for the next 5 yr is somewhat confusing. Manu-
facturers have effected some massive increases in MCF capacity just recently,
assuredly based upon solid reasons. In order for a chemical manufacturer to
make a capital investment of from one to many millions of dollars, he must
show that production will return sufficient money on the investment.
9
The 1976 worldwide capacity for TCE was 1.964 x 10 lb; usage was 54
g
percent of capacity, at 1.061 x 10 lb. In the same period of time, the
worldwide capacity of MCF (with the exception of an unknown amount the USSR
may have contributed) was 1.068 x 10 lb, 90 percent of capacity. This world-
wide capacity was essentially emitted to the atmosphere.
14-10
-------
Increases in the use of MCF worldwide have recently occurred, or will
8
soon occur, because of new industrial capacities: 1.8 x 10 Ib in Europe as
8 8
of January 1977; 3 x 10 Ib in Plaquemine, LA as of December 1977; 3 x 10 Ib
o
in Lake Charles, LA as of December 1978; and 1.4 x 10 Ib in Geismar, LA in
o
the near future. Since 1976, an additional 9.2 x 10 Ib of MCF have become
available worldwide a growth in capacity of 86 percent. Such a great ca-
pacity investment is made only if a market exists for its utilization.
In the U.S. alone, capacity in 1976 was 6.4 x 10 Ib. With the three new
U.S. facilities mentioned above, the country will acquire an additional 7.4 x
8
10 Ib in capacity by the end of this year. Th:
is extraordinary for so short a period of time.
8
10 Ib in capacity by the end of this year. This 116-percent capacity growth
Singh et al. (this volume) have indicated the basic factors to be con-
sidered when reviewing the severity of a pollutant impacting stratospheric
O depletion: residence time in the troposphere, amount of emissions, and
escalation of the emissions to the atmosphere. These components must be
addressed when the need to apply RACT to MCF and to TCE is reviewed.
Most authorities agree that the residence time of MCF is ~8 to 10 yr.
Moreover, the fantastic ballooning of MCF capacity will aggravate the emis-
sions problem (and further large-scale capacity will undoubtedly be built).
4
The coatings industry emits 261 x 10 tons/yr of solvents to the atmo-
sphere in the U.S. To date, coatings applications have not been prominent for
MCF (the bulk of MCF is used in solvent cleaning). Should MCF be also ex-
empted in coatings lines which involve many operations, including textile,
paper^, and adhesive applications use and emissions of MCF in these applica-
tions would be sharply escalated as well.
Many states are planning to exempt MCF in their SIP's, not only in sol-
vent cleaning operations, but also in coatings lines. Detrex believes strongly
that this approach is not responsible, but rather that MCF along with TCE
and the other chlorinated solvents should be controlled by RACT. RACT is
14-11
-------
practical and involves simple procedures and controls readily available to
users and manufacturers of solvent vapor degreasing equipment.
REFERENCES
Atmospheric Sciences Interdepartmental Committee (National Science Foundation).
1975. The possible impact of fluorocarbons and halocarbons on ozone.
ICAS 18a-FY75, National Science Foundation, Washington, D.C., May.
Council on Environmental Quality (Federal Council for Science and Technology).
1975. Report of Federal Task Force on Inadvertent Modification of the
Stratosphere (IMOS). June.
EPA Journal. 1978. Stratospheric problem worsens. 9:38.
U.S. Environmental Protection Agency. 1979. Environmental News (press release)
January 26.
DISCUSSION
Voioe from Audience: Does Detrex do most of its business in PCE and TCE as
related to MCF?
Mr. Schlossberg: We have a high volume of these chlorinated solvents but not
a real stake in any one of them. Our purpose is to protect the solvent vapor
degreasing process. We are one of the largest manufacturers of equipment used
in this process. We are concerned with the possibility that TCE use may be
eliminated, that PCE is under a cloud of carcinogenicity (which we don't agree
with, incidentally), and now all of a sudden only the use of MCF remains un-
restricted. Major producers are getting out of the TCE business, anticipating
a point when only MCF can be used. As new data are acquired and when the boom
gets lowered, it could wipe out the process. This, obviously, we don't want,
because we want to continue to sell equipment for the process.
Dr. Mazalesk-i: I have one more question. Dr. Farber [Discussion, Lapp, this
volume] stated, at least to my understanding, that health effects were not to
be addressed here, but I would like to clarify this point. On October 29,
1978, Dr. Farber sent a letter to EPA, and he attached this document entitled,
"1,1,1-Trichloroethylene as an Industrial Solvent: A Review of the Current
Health Environmental Knowledge." In that document, which he wanted us to ad-
dress, as we have really done, were mentioned the stratospheric O problem and
the health effects. I handled the health effects in part, and I could discuss
these for about 4 days, if permitted. But Dr. Farber asked for that particu-
lar evaluation, and in Section 4, entitled, "Toxicity," he said "the most
critical and definitive animal carcinogenicity testing for airborne contamina-
tion is by long-term chronic inhalation. A lifetime study was completed on
male and female rats exposed to 1750 and 875 ppm of 1,1,1-trichloroethylene,
1.4-12
-------
up to 5 times the allowable 8-hour time, the weighted average outstanding.
The results show the solvents to be noncarcinogenic in that test." Several
EPA researchers have also addressed that issue very well, I think.
MP. SahZossberg: I believe all the speakers at this conference were sent that
package of information by Dow Chemical. Now, that did include the toxico-
logical and environmental aspects of MCF use.
Dr. Hanst: I guess the issue that Dr. Farber was raising was with me: that
the conference was getting off the track, and I had assured him that I didn't
intend that we would go into the health issue in any great detail. I told him
that I couldn't predict that some of the speakers wouldn't get into it, but it
seemed to me the program was almost entirely on the atmospheric chemistry ques-
tion.
Dr. Rowland: I want to paraphrase the question here. If we are going to have
all this increased capacity, why are we going to have 6 percent increase in
year-end production?
Dr. Farber: The 6 percent growth rate that I mentioned yesterday [Discussion,
Neely and Agin, this volume] is based on analysis of the trends of the last 5
yr. This is what we expect. Six percent is not a small number in terms of
absence of pounds. I am not minimizing the growth in terms of pounds. Six
percent of a billion pounds is 60 million, and thus necessarily a small num-
ber. Six percent of the next year is a larger number. That is our estimate,
and that is based on all the data I reviewed yesterday. We could both be
wrong in our estimate. It could be less or more.
I mentioned yesterday that on a year-by-year basis we see fluctuations in
that estimate of up to ±10 percent from a standard growth rate projected. In
the last 5 yr of total 1,1,1 production, a decrease of ~5 percent can be found
in one year of this sequence. That is a normal response to market demand, but
I still anticipate no more than 6 percent average growth rate over the next 5
yr.
Now, somebody asked "Well, why do you go to capacity?" There are two
reasons. First, the plant that we have used for producing TCE for about the
last 30 yr is located in Shreveport. It is a very profitable product for us,
and we are in the business to make money. Shreveport is the only plant we
have in this country producing TCE and, as I told you yesterday, we have ~50+
percent of the 1,1,1 business. We obtained that business because we did all
the groundwork for this business venture and decided to build that capacity.
All right, that is in Texas. If we lose that plant because of Hurricane Carla,
the stratospheric whatever, we have a fair share of our profits wiped out.
You build capacity for several reasons. One is anticipated growth. One
of them is to have enough flexibility to make the product, if it is possible
to do it that way, at two sites, not one. We also invested in a new TCE plant.
As I said yesterday, I anticipate that we would be very happy to see 6 percent/
yr (assuming, of course, that we are not harming the environment).
14-13
-------
Mr. Schlossberg: Since I understood the question was asked of both me and Dr.
Farber, I would also like to answer Dr. Rowland. As a businessman, I find it
hard to understand that this huge capacity can be put on-stream without a firm
expectation that there will be a home for it; unquestionably, there is going
to be a driving force to use it, and the driving force is going to be to re-
place TCE and to get into the coatings markets, which are huge.
14-14
-------
REGULATORY ISSUES INVOLVED IN HALOCARBON CONTROL
UNDER THE CLEAN AIR ACT
Robert Kellam
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina
INTRODUCTION
This paper briefly identifies the regulatory issues and concerns involved
in the control of halocarbons, specifically methyl chloroform (1,1,1-trichloro-
ethane, CH CC1 , MCF), under the provisions of the Clean Air Act (CAA).
THE CLEAN AIR ACT
The CAA, as the primary legal mechanism for protecting ambient air quality,
provides several regulatory options for control of substances identified as
contributing to air pollution. The appropriate use of any of the CAA Sections
depends on the nature, prevalence, and sources of a specific pollutant. Three
areas of concern suggest regulation of MCF and other halocarbons by the CAA:
(1) formation of ozone (O.,) in the troposphere; (2) depletion of O-. in the
J .3
stratosphere; and (3) direct danger to human health or welfare.
TROPOSPHERIC AND STRATOSPHERIC OZONE
Under Sections 108 and 109 of the CAA, National Ambient Air Quality
Standards (NAAQS) are promulgated for O... Since this chemical is not emitted
directly from controllable sources, the primary strategy for attaining the
0 standard is control of volatile organic chemical (VOC) emissions that
15-1
-------
participate in photochemical reactions to yield O_ in the lower atmosphere.
Stationary sources of VOC's are controlled under State Implementation Plans
(SIP's). In addition to the strategy of emission reduction, many SIP's and
U.S. Environmental Protection Agency (EPA) guidelines utilize photochemical
reactivity as a useful concept in reducing O formation. The substitution of
less-reactive for more-reactive compounds results in a lower net O . Incen-
tive for the substitution of less-reactive VOC's is provided by exempting
these materials from control or inventory requirements.
A large number of substances were originally exempted from control by
solvent substitution provisions such as Los Angeles' Rule 66. Gradually, this
number has declined as studies have shown that low- or moderate-reactivity
VOC's may, under certain conditions, generate significant photochemical oxi-
dant. In addition, concern over the potential of certain compounds for direct
or indirect toxicity has led EPA to remove others from the list. EPA policy
now recommends exemption for only four substances (methane (CH.), ethane
(C2H6), MCF, and fluorocarbon-113 (CC12FCC1F2, FC-113)); but EPA is concerned
that similar health considerations may lead to the removal of MCF and FC-113.
The removal of MCF from the exempt list for environmental concerns other
than photochemical reactivity would create an ironic situation in which this
low-reactivity chemical must be inventoried and controlled to attain an O
level to which it contributes only negligibly. Maintaining it on the list,
however, endorses uncontrolled substitution of MCF for nonexempt chemicals,
though key health questions remain unresolved. Although many scientists may
agree that MCF contributes in only a very minor way to tropospheric concentra-
tions of O , for EPA to continue advocating the substitution of this high-
volume solvent is clearly inappropriate, as scientific evidence suggests that
the resulting increase in MCF emissions may (1) contribute significantly to
stratospheric 0 depletion, or (2) present a direct risk to human health.
Under the CAA Amendments of 1977, EPA is given specific authority to
protect the stratospheric O layer. In view of the predicted long lag between
emission of fully halogenated pollutants and their interaction with O in the
stratosphere, a precautionary stance on the part of EPA is appropriate. Though
15-2
-------
considerable uncertainty and disagreement exist over the impact of non-fully-
halogenated species (such as MCF) on the O3 layer, EPA does not conclude that
this growing source of atmospheric Cl can be ignored in the absence of addi-
tional information. The exchange of ideas and the identification of research
needs permitted by the Conference on Methyl Chloroform and Other Halocarbon
Pollutants is thus strongly supported by EPA.
Aside from the indirect effects to health that halocarbon pollutants may
induce through tropospheric O formation and stratospheric O depletion, EPA
must consider the direct hazards of toxicity, mutagenicity, carcinogenicity,
and teratogenicity. There is evidence (albeit controvertible) of mutagenicity
in bacterial and mammalian test systems due to MCF. This circumstance raises
the possibility of human mutagenicity and/or carcinogenicity and has led EPA's
Carcinogen Assessment Group to conclude that "suggestive" evidence of human
carcinogenicity exists for this chemical.
REGULATION OF METHYL CHLOROFORM
Part B of the CAA, which addresses protection of the O layer, states
that pollutants judged to endanger public health may be regulated under Sec-
tions 108 and 109 (NAAQS); Sections 111 and lll(d) (Standards of Performance
for New Stationary Sources); or, if the effect is "serious and irreversible"
or "incapacitating and reversible," under Section 112 (National Emission
Standards for Hazardous Air Pollutants).
At the present time, two regulatory approaches have been recommended for
MCF:
(1) Deletion of MCF from the recommended list of exempt solvents,
perhaps leading to controls under SIP's for the O standard.
(2) Regulation of MCF under Sections 111 and lll(d) as a solvent
in the metal cleaning industry.
The reasons for considering the deletion of a low-reactivity substance
from the exempt list i.e., potential for direct and indirect health effects,
15-3
-------
and substantial emissions also apply to regulation of MCF in the solvent
metal cleaning industry under Sections 111 and lll(d).
Section 111, Standards of Performance for New Stationary Sources, pro-
vides control, through emission standards, of new pollutant source categories
listed under Section 108. As precursors to the listed pollutant O_, reactive
hydrocarbons emitted from such sources as vapor phase degreasers will be regu-
lated in this manner. In addition, MCF and four other halocarbon solvents
(trichloroethylene (C2HCl , TCE), perchloroethylene (C Cl , PCE), FC-113, and
methylene chloride (dichloromethane, CH^Cl^)) will be designated under Section
lll(d). Thus, emissions from both new and existing solvent metal cleaners
that use these five compounds will be regulated.
The decisions to proceed with regulatory controls for MCF have been based
on rather inconclusive scientific data. Inferences drawn from these data are
subject to uncertainty, particularly in view of conflicting results by other
investigators; EPA continues to seek the resolution of this uncertainty. How-
ever, EPA is convinced that present evidence, as well as the magnitude of
projected emissions and the persistence in the environment that would result
from continued, uncontrolled use of MCF, dictate caution in policies and regu-
lations that might encourage significant increases in public exposures.
DISCUSSION
Dr. Rowland: As I understand, one reason you advocate the deletion of MCF
from the list of exempt solvents is the possibility that it depletes strato-
spheric O . If that is correct, surely you must be also asking for deletion of
FC-113, wnere the effect must be 5 to 7 times greater?
MT. Kellam: Yes, we are.
DP. Klauder: As you know well, the chlorofluorocarbon regulation was unique
because a generic class of chemicals was regulated. At the first regulatory
meetings where regulatory scope was discussed, this concept was considered
foreign. There was strong support for a regulation that would apply only to
fluorocarbon-11 (trichlorofluoromethane, CC13F, FC-11) and fluorocarbon-12
(dichlorodifluoromethane, CC12F , FC-12). In subsequent meetings, it became
quite apparent that if the regulations governed only FC-11 and FC-12, substitu-
tion of FC-113, fluorocarbon-114 (CC1F2CC1F2, FC-114), and other fully halo-
genated chlorofluoroalkanes might lead to a noneffective regulation.
15-4
-------
I have the same concern with" the approach you have presented. As Mr.
Schlossberg told us, regulation of TCE resulted in a great increase in MCF.
Now you propose to control MCF, FC-113, and perhaps a few of the other halo-
carbons. Shouldn't this problem be addressed on a generic basis? Are we
going to be in a position, 5 yr from now, where we will have conversion to
other chlorinated and brominated halocarbons? It may be worth taking addi-
tional time to adequately address the entire problem of chlorinated and bromi-
nated halocarbons.
I agree with Herbert Wiser (and have stated to EPA several times, myself)
that we may be talking about all halocarbons on a continuum of reactivity. At
one end, we are talking about very reactive compounds which are involved in
tropospheric photochemical reactions. At the other end, we are talking about
very nonreactive compounds which migrate upwards, causing stratospheric 0.,
depletion. All of the brominated and chlorinated halocarbons fall somewhere
on this continuum. This seems to be a problem that must be addressed in a
generic fashion and for which acceptable burdens of Cl and Br in the strato-
sphere must be determined.
Mr. Kellam: Initially, the CAA did not specifically provide for generic reg-
ulations under Section 111. However, the 1977 CAA Amendments have allowed us
some latitude in developing these kinds of controls. We have an ongoing pro-
gram, the Synthetic Organic Chemical Manufacturing Industry Program, which, in
a broader context, is intended to develop technological guidelines for imple-
menting generic control technology for an entire industry. These regulations
take account of your concerns regarding stratospheric CU, as well as EPA's
concerns about formation of tropospheric O . The intention is to develop
regulations broad enough to apply to most types of source categories within
the organic chemicals industry. Of course, these regulations will take into
account technological feasibility, as well as the cost of control.
Dr. Humenny: With the recommendation that MCF be controlled, will it be
subject to New Source review or an offset requirement?
Mr. Kellcm: I would assume the deletion from the exempt list would put MCF
into the same category as all the other nonexempt chemicals. How a specific
state wishes to handle the controls and requirements within the EPA guidelines,
I believe, is at the State's discretion, and will depend on how its SIP is
designed.
There is one other point that Mr. Schlossberg brought up in referring to
an August 24 memorandum of Walt Barber. His comment was that Mr. Barber was
essentially saying that, although we were considering the deletion of MCF from
the list of exempt solvents, we would not disapprove SIP's continuing to ex-
empt these solvents. I would like to clarify the context of that memorandum,
because I believe there is an important element of timing in what Mr. Barber
was saying. He says:
I recognize that many States are well along in the preparation
of their regulatory packages and inventories. In order not
to change the existing guidance at this late date, I am re-
questing that you advise your State Directors that, although
15-5
-------
we will not disapprove the State oxidant SIP submittal which
exempts methyl chloroform from control, we are very concerned
with the environmental risks associated with widescale sub-
stitution to methyl chloroform; and that the uncontrolled
use of methyl chloroform as an approved means for compliance
should be avoided wherever possible.
I think his intent was to apprise the states of our concerns without disrupt-
ing the SIP revision process, which was in its final stage in August.
Mr. Sahlossberg: I do agree with exactly what you said, but what I was trying
to convey was the information that you were getting from the states as they
interpreted that number. I know what Mr. Barber had in mind, and his action
was forthright.
Dr. Rowland: Who has superseding jurisdictions, the states or the federal
government, in removal from the exempt list and approval of the SIP's?
Mr. Kellam: In the case of the 0 standard, EPA does have approval authority
of the SIP. There is an outside possibility, if MCF were removed from the
exempt list, that EPA could disapprove SIP's based on its continued exemption.
In this case, the state would have to rewrite the plan, or I think the CAA
would allow EPA to actually promulgate a plan for the state, if we couldn't
reach some agreement.
15-6
-------
INTERAGENCY WORK GROUP ACTIVITIES ON
NONAEROSOL USES OF CHLOROFLUOROCARBONS
Ferial S. Bishop
U.S. Environmental Protection Agency
Washington, D. C.
INTRODUCTION
The Chlorofluorocarbon Interagency Work Group has been formed to regulate
the reduction of chlorofluorocarbon (fully halogenated chlorofluoromethanes)
emissions. For a clearer perspective on the Work Group's present activities,
a brief history is given below.
THE CHLOROFLUOROCARBON PROBLEM
The effects of chlorofluorocarbons (CFC's) on depleting stratospheric
ozone (O ) came to light at the beginning of this decade. The early 1970's
brought the supersonic transport (SST) and questions concerning effects of
nitrogen oxides (NO ) emitted from high-flying aircraft. Scientists also
X
questioned whether damage to the O layer could come from other chemical
compounds. In 1974 Drs. Rowland and Molina presented a paper detailing the
CFC O3 depletion hypothesis. Their findings mobilized the Council on Environ-
mental Quality (CEQ) and the Federal Council on Science and Technology to
initiate additional studies on CFC emissions. Thus, the Interagency Task
Force on the Inadvertent Modification of the Stratosphere (IMOS) was formed.
In 1975 this Task Force reported that cause for concern existed with regard to
CFC emissions, and that, if the National Academy of Sciences (NAS) confirmed
their study, federal regulatory agencies should initiate rulemaking procedures
to restrict CFC uses.
16-1
-------
With the fall 1976 publication of the NAS report confirming the threat to
the O layer from CFC releases, the Food and Drug Administration (FDA), Con-
sumer Product Safety Commission (CPSC), and Environmental Protection Agency
(EPA) began to consider action against the use of CFC's in products under
their respective jurisdictions. By the end of November 1976, CEQ had already
held several meetings to coordinate regulatory activities, thus establishing
the first CFC Interagency Work Group. FDA, CPSC, and EPA were designated as
the lead agencies, but were aligned with CEQ, the National Aeronautics and
Space Administration (NASA), National Science Foundation (NSF), National
Oceanographic and Atmospheric Administration (NOAA), Department of Commerce
(DOC), and Department of Transportation (DOT).
The Work Group acknowledged that EPA had the broadest authority over
CFC's under the new Toxic Substances Control Act (TSCA) enacted in October
1976 and also the longest-term interest in CFC regulation; therefore, EPA was
given the leadership role. Moreover, the members concluded that individual
legislation by CPSC, EPA, and FDA was unnecessary, since all aerosol uses
could probably be regulated under TSCA and the Federal Food, Drug, and Cos-
metics Act.
PHASE I
The Work Group's initial plans called for division of the CFC program
into two phases. The first phase would be the regulation of CFC's used as
aerosol propellants. The second phase of the regulatory process would be the
investigation of all other uses of CFC's.
Continuous Work Group deliberations had revealed many issues and sug-
gested approaches regarding proposed regulation of aerosols. Some key issues
included definition of the compounds to be regulated; definition of "aerosol
propellant;" timing of regulation enactment; and consideration of CFC produc-
tion and recovery requirements as regulatory alternatives. The resolution of
these and other problems necessitated dividing the Work Group into subcommit-
tees in order to review existing data and recommend solutions.
16-2
-------
The question of which products should be exempt from regulation became an
important issue during the Work Group meetings. After numerous discussions,
this problem was resolved by subjecting each aerosol product to the four basic
questions below to determine if the product was essential and therefore exempt
from regulation:
(1) Are alternatives available?
(2) What is the economic significance of the product, including
effects in the marketplace?
(3) What are the environmental and health impacts of the product
and its alternatives?
(4) What are the effects on quality of life if the product becomes
no longer available for use?
After evaluating scientific reports, analyzing economic impacts, and
receiving testimony from meetings with the American public and businesses, the
Work Group recognized that uncertainties about the magnitude of O depletion
did not override concerns for public health. The available data indicated
that to further delay regulation would in itself lead to unreasonable risks to
long-term human health and the environment. Further, the nonessentiality of
most aerosol products, plus the ready availability of substitutes for aerosol
propellants, supported the EPA and FDA decision to proceed on March 17, 1978
with a ban on manufacturing and processing of CFC's for use as aerosol pro-
pellants.
Since October 15, 1978, all manufacturing of CFC's for aerosol propellant
use has been prohibited. Since December 15, 1978, all processing (including
processing for export) and distribution of CFC's for use in aerosol products
has been prohibited in the U.S.; import of products containing CFC's and the
manufacturing and packaging of food, drugs, or cosmetic aerosol products
containing CFC's were also banned as of that date. On April 15, 1979, food,
drug, and cosmetic aerosol products containing CFC's may no longer be intro-
duced into interstate commerce in the U.S. All finished products already on
16-3
-------
the market and in distribution channels, however, can be sold until stocks are
depleted.
Both EPA and FDA regulations identified certain aerosol products as
exempt products. EPA exempted Department of Defense products, electronic
cleaners, aircraft maintenance products, mine warning devices, plastic mold
release agent products, diamond grit spray, and some pesticide products. FDA
exempted certain metered dose drug products used for oral inhalation and con-
traceptive foam products. Exempted uses, however, account for only 2 to 3
percent of total CFC aerosol uses in the U.S. The ban on nonessential aerosol
uses of CFC's is expected to yield a reduction in annual U.S. CFC emissions of
>60 percent.
PHASE II
Because regulation of aerosol uses may not be sufficient to reduce CFC
emissions, and because of the growing uses of CFC's in nonaerosol applications,
attention has now focused on domestic nonaerosol emission sources. The pur-
pose of Phase II is to more accurately define current and future levels of CFC
emissions from nonaerosol uses (such as refrigerators and air conditioners)
and to evaluate means of achieving further emission reduction with regard to
socioeconomic consequences.
In many respects, Phase II is a continuation of Phase I. Because no
single legislative statute administered by EPA, FDA, and CPSC provides juris-
diction over all products that utilize CFC's, collective evaluation by these
agencies is still the preferred approach. But continuation of Work Group
activities does not necessarily indicate additional regulation at present,
the Group remains in a fact-finding mode.
In August 1977 Congress passed the Clean Air Act (CAA) Amendments, which
specifically charge EPA with protecting the stratosphere (especially its 0 ).
However, the regulation banning CFC-propelled aerosol products was enacted
under TSCA, since Phase I investigations began under TSCA. The passage of the
.16-4
-------
1977 CAA Amendments thus brings into question the legislative authority under
which any future regulations will be enacted.
In answer to this question, the CAA Amendments strongly suggest that any
future action to protect the stratosphere be taken under their provisions.
However, the Amendments specifically state that new statutory provisions are
not to affect any proposed rule published under TSCA prior to enactment of the
Amendments, nor any state law or regulation pertaining to control of halo-
carbon propellants in aerosol sprays.
Part B of Title I of the CAA Amendments directs the Administrator of EPA
to determine human effects on the stratosphere by conducting and coordinating
research and monitoring studies undertaken throughout the government, univer-
sities, and the private sector. This section also authorizes the Adminis-
trator to control any substance, practice, process, or activity reasonably
anticipated to affect the stratosphere, especially the O , if such effect is
reasonably anticipated to endanger public health or welfare. In both cases,
the Administrator is required to report to Congress on specific dates the
results of research findings, progress on international cooperation, and
results of regulatory activity by EPA and other federal agencies regarding
protection of the stratosphere. Congress is also to receive recommendations
for additional research.
Because of these new Amendments, questions have arisen regarding the
breadth of the Work Group's activities. Should the scope broaden to include
all chemicals suspected of being harmful to the stratosphere, or should the
focus be solely on CFC's? Because of limited or nonexistent facts concerning
harmful effects of other suspected O depletors necessitating regulatory con-
sideration, the Work Group continues to concentrate on CFC emissions. None-
theless, other chemicals have been at least discussed by the Work Group; EPA
will need to conduct and support more research to confirm suspicions of harm-
ful effects from such chemicals in the stratosphere before initiating regula-
tory action and convening a Work Group, however.
16-5
-------
The immediate task before the CFC Work Group is to gather information on
nonaerosol uses of CFC's. Included in this category for review are CFC's used
in refrigerators and air conditioners, as foam-blowing agents in the manufac-
ture of urethane and nonurethane foams, and as cleaning agents in the solvent
industries. A host of miscellaneous minor uses will also be reviewed. Chloro-
fluorocarbons used in these applications are rarely consumed, although minor
amounts may be chemically or thermally broken down. With the exception of
certain products such as flexible urethane foam, CFC's are released from most
nonaerosol products over long periods of time (sometimes over 20 yr after
manufacture).
The Work Group is examining ways to eliminate or minimize CFC emissions.
This review extends to the best available technologies, including reuse, re-
cycling, and other traditional types of control. These approaches specify
performance standards required to reduce emissions or, in some cases, a com-
plete ban on a product. The latter strategy was adopted for the current
regulation of aerosol uses of CFC's.
The Work Group is also examining nontraditional control options, in-
cluding marketable permits, production ceilings, emission fees, and voluntary
measures. As an example, emission fees could be added to the price of a sub-
stance under scrutiny to discourage its use. Another possible option is a
system of marketable permits, entailing an "auction," after a total maximum
level of CFC emissions has been set, of "rights" to emit certain amounts of
CFC's within that level. In both cases, the marketplace not the government
would determine which products are essential.
Substitutes for CFC's in nonaerosol products are also under evaluation,
although few substitutes for CFC's appear available at the moment. In addi-
tion, as substitutes are evaluated, caution is exercised that the substitutes
do not themselves create health or environmental problems.
The Work Group is looking at the entire life cycle of each product con-
taining CFC's manufacture, normal use and servicing, and finally disposal.
Emission profiles are evolving that show where greatest emissions occur.
16-6
-------
Within the air-conditioning and refrigeration category, for instance, the
largest source of emissions appears to be mobile air conditioners. In 1976,
emissions from this source amounted to 74 million Ib (73 percent of the total
103 million Ib emitted from all air-conditioning and refrigeration units); by
1990, mobile air-conditioning emissions are expected to rise to 113 million
Ib (77 percent of the total 148 million Ib for this category). Further anal-
ysis indicates that most emissions occur as a result of leakage, recharging,
and servicing.
Another category, plastic foams, includes three types of products:
flexible urethanes, rigid urethanes, and nonurethanes. Flexible or soft foams
are used as cushioning in furniture and automobiles. During the manufacture
of flexible urethane foam, CFC's pass through the product, helping form the
cells of the foam, and are immediately released into the atmosphere. On the
other hand, emissions during manufacture of rigid urethanes (used primarily as
insulation) are small, because the insulating efficiency of rigid foam depends
on retention of CFC's. But these CFC's slowly leak into the atmosphere over
the life of the product, creating an emissions problem during normal use and
disposal. The third type of foam, nonurethane, includes mostly polystyrenes
and polyolefins used primarily for food and other packaging. Manufacturing
emissions from these products resemble those from flexible foam. Moreover,
since the length of service of these products is very short perhaps several
weeks before disposal emissions are considered to be immediate.
The estimated total amount of emissions from flexible, rigid, and non-
urethane foams for 1977 is 79 million Ib, an amount slightly less than the 103
million Ib released from all air-conditioning and refrigeration units. How-
ever, the use of foams is growing to the extent that, by 1990, emissions are
projected to be more than 200 million Ib and will likely exceed emissions from
air conditioning and refrigeration. Hopefully, emission profiles from these
and other products under investigation will help identify stages of maximum
possible emission reduction with least impact on both product performance and
the economy.
16-7
-------
ADDITIONAL RESEARCH
Helping the Work Group to collect and analyze data are research organiza-
tions such as the Rand Corporation, SRI International, the University of
Maryland, and NAS. The Rand Corporation is analyzing the economic feasibility
and impacts of alternative regulatory options. This study, supported by EPA,
CPSC, and FDA, has received input from the Work Group regarding the scope of
work and regulatory approaches to be analyzed.
Under the direction of EPA1s Office of Research and Development, research
teams at the University of Maryland and SRI International are analyzing trade-
offs between the costs and benefits of additional CFC control. As mandated in
CAA, NAS is preparing a comprehensive report on causes, effects, and alterna-
tive methods of control of stratospheric 0. depletion. One goal of this study
is to update the latest developments in atmospheric chemistry of stratospheric
modification.
EPA will use data collected in these and other investigations to conduct
its own study of the levels associated with nonaerosol uses of CFC's. This
risk-assessment report is expected to provide quantitative estimates of the
impact of uncontrolled future emissions on O_ levels, as well as the reduction
of these effects resulting from selected control options. This report should
identify (among other things) baseline emission scenarios and emission reduc-
tion scenarios for CFC's and then correlate both with quantifiable health and
environmental risks.
In the coming months, the Work Group will review these and other reports
and analyze various control options in order to develop recommendations on
additional regulatory needs. Where appropriate, these recommendations will be
categorized by specific industry and by individual consumer product. Division
of the Work Group into subcommittees by industrial category will provide the
maximal response to differences in emissions control. Assuming current studies
are completed as planned, the Work Group expects to present a regulation policy
for chemicals in question to the respective Agency Heads by the end of 1979.
1.6-8
-------
FINAL COMMENTS
Examining the need for more regulation is within the Work Group's phased-
approach philosophy to controlling CFC's and acknowledges the need to care-
fully balance risks and benefits associated with nonaerosol uses. During the
deliberations on aerosol uses, the Work Group developed regulations even
though uncertainties related to the risk of CFC's were not resolved, because
substitutes were readily available. In the case of nonaerosol uses, however,
action against the continued use of these products could have a much greater
impact on the U.S. economy and well-being.
Preliminary reports indicate large investments may be necessary to reduce
CFC emissions from nonaerosol products. An obvious stumbling block faces the
Work Group the need to curtail essential nonaerosol emissions while use of
nonessential aerosol products continues worldwide.
At the international meeting on CFC's in Munich, Germany in December
1978, the U.S. delegation headed by Barbara Blum, Deputy Administrator of EPA,
urged those in attendance to take a unified global approach to reduce CFC
emissions from aerosol products. Although member nations resolved that all
countries should significantly reduce CFC emissions and continue studies on
the technical and economic aspects of the CFC question, firm commitments by
other countries did not materialize. While the Work Group will continue its
efforts in the international community to achieve worldwide control of CFC
emissions, the present international situation is of concern to the Group.
The role of the CFC Interagency Work Group becomes one of delicately
balancing all scientific and economic factors to produce a viable, reasonable
position. The Group looks to the scientist to continue his research in pro-
viding necessary information, and to the industrialist to continue his search
for new methods to control CFC emissions and to improve the efficiency of pres-
ent emission reduction methods. If all factions of society work together, the
relevant scientific, socioeconomic, and legal aspects of the CFC issue can be
carefully reviewed so that this Interagency Work Group can make the best
recommendation on future regulation needs.
16-9
-------
DISCUSSION
Dr. Sobolev: I have two questions. As I understood, Mrs. Bishop, the CFC
Work Group was established by CEQ.
Mrs. Bishop: I think what happened in those early days was that CEQ became
aware of the efforts of FDA and others to consider regulatory action against
CFC, and these actions necessitated that agency representatives be called
together to share information and regulatory approaches.
DP. Sobolev: I am truly interested in the legal aspect. They initiated in-
formation on your Group. What part of EPA, what agency, could dissolve the
Group, if a need for that became justified? I know now who has the authority
to form your Group. Who has the authority to dissolve it?
Mrs. Bishop: Let me rephrase. This isn't a question of authority. The
purpose of CEQ, as I understand it, was to coordinate programs in different
agencies. I will also defer to some of our veteran Work Group members here
who could shed more light on what happened in the early days of Work Group
activities.
Dr. Sobolev: I will wait for that answer. Perhaps you could answer my second
question. Not too long ago, NAS decided that, in view of some new measure-
ments of rate constants, NO from SST's were no longer considered likely to
pose a significant environmental hazard. Suppose, as a result of future
developments, NAS comes to similar conclusions regarding Cl from fluorocar-
bons. What would happen to your Group? Would it just keep going?
Mrs. Bishop: Well, we are currently studying the CFC question and whether we
should pursue it further.
Dr. Sobolev: That is only because NAS decided that there was cause for con-
cern, that fluorocarbons are an environmental hazard. Suppose NAS
Mrs. Bishop: Came up with a reversed decision?
Dr. Sobolev: Yes.
Mrs. Bishop: Then I would like to think, and Dr. Wiser can correct me, that
EPA would probably reverse the regulation.
Dr. Sobolev: I wasn't even thinking that far. I was just thinking of what
would happen to your Group. What agency would have the authority to say that
the services of your Group are no longer needed?
Mrs. Bishop: When EPA decides to initiate a regulation, a work group is
convened to develop this regulation. If there were a need to eliminate a
regulation, although I don't know whether that has happened, a work group
would probably convene to do the reverse and proceed through EPA with prepared
documents indicating this regulation is not needed.
16-10
-------
Dr. Sobolev: Please forgive me. I am not trying to be funny or sarcastic.
I just wanted to know what would happen.
Mrs. Bishop: You are asking a question on doing something in reverse in EPA
that may not have been done before, and it is probably a good question. My
answer is just my opinion.
Dr. Sobolev: Thank you.
Dr. Wiser: I would like to add to that and clarify some of the historical
misconceptions presented here. These committees are not established by law.
The various agencies thought it would be very efficient since we are all
looking at the same scientific data, in many areas to work together. We
did. We hope to continue to work that way. Most of these committees, regard-
less of the length of their existence, are ad hoc committees, and when a pro-
blem goes away, the committee members are assigned to other positions, and the
committees are dissolved. Furthermore, although NAS is a very creditable body
and, based on NAS expertise, does present an overall, latest, state-of-the-art
summary with recommendations for scientific research needs for regulatory
action, etc., these are not the only documents that are reviewed and studied
by the agencies. We study virtually everything that comes to our attention in
the published literature.
Mrs. Bishop: Thank you for reinforcing that.
16-11
-------
PANEL DISCUSSION
Dr. Sohiff: I was asked to chair this session, perhaps because of my notoriety
as an author. I authored a paper on methyl chloroform (1,1,1-trichloroethane,
CH CCl , MCF) which was misinterpreted in certain quarters as an attack on the
U.S. Environmental Protection Agency (EPA), and I would like to deny that as
the intent. I was also coauthor of a book which was interpreted as an attack
on almost everybody, which I, of course, totally deny! However, I will attempt
to maintain this image to inject a little bit of controversy here.
The advice I would give EPA is not to be convinced by industry that MCF
emission is probably okay because it is removed in the troposphere. Neither
should EPA be convinced, on the other hand, by scientists who say that every-
thing must be known about tropospheric chemistry, about the hydroxyl radical
(OH) concentrations, about the N/S ratios, etc., and who seek more research
support. Such a research effort is very commendable but not absolutely the
first required task.
The amount of any substance, including MCF, entering the stratosphere is
undoubtedly proportional to the amount in the troposphere, and tropospheric
MCF measurements are ~100 ppt. To estimate the stratospheric effect of MCF
relative to fluorocarbons, the relative tropospheric concentrations can be
used. This ratio provides an approximate measure of the relative effects of
these substances on the ozone (O,.) layer. The present tropospheric concen-
tration ratio of MCF to fluorocarbon-11 (trichlorofluoromethane, CC13F, FC-11)
plus fluorocarbon-12 (dichlorodifluoromethane, CCl F , FC-12) is about 1 to 4.
The projected time when MCF concentrations will demand serious concern is
the point when these concentrations contribute approximately the same amount
of Cl as do present FC-11 and FC-12 concentrations. Using Professor Rowland's
17-1
-------
figure of 1.1 percent/month increase (provided present emission rates remain
fairly constant) and multiplying that figure by 1/4, one arrives at a projected
time of ~10 yr for stratospheric Cl contributions of MCF to parallel those of
FC-11 and FC-12. On the other hand, using Dr. Rasmussen's estimated increase
rate over the last year yields a different number. The most important issue
to be addressed appears to be the status of MCF measurements. How well do we
know its concentrations, and what are the uncertainties? Correlatively, how
well can we calculate for the absolute amounts we want to know? The several
inconsistent findings reported here give rise to confusion. Some people have
not formally presented their MCF measurement methods, and these contain fur-
ther inconsistencies needing clarification. Other groups, including the
National Center for Atmospheric Research (NCAR) and the National Oceanographic
and Atmospheric Administration (NOAA), have suggested strongly that MCF cannot
be measured with any degree of accuracy. A few do not believe any good abso-
lute measurements have been made. Jim Lovelock suggests that great diffi-
culties exist in obtaining MCF measurements, yet Professor Rowland observes
few measurement difficulties.
The question of precision versus accuracy must also be considered. Pre-
cision was not examined thoroughly in the presentations and discussions at
this meeting. Inconsistencies were rather ignored. For instance, Dr. Rasmussen,
is it not correct that you have made surface measurements and measurements from
the airplane, and you find differences between ground measurements and measure-
ments in the air? Let me hold the question for the time being and return to
you for a response.
If you use real-time measures, sampling directly into a gas chromatograph
on an aircraft, and do a latitude sweep, the absolute numbers may be questioned,
but the relative measurements are usually accepted. We have seen different
latitude dependencies on the same aircraft.
Dr. Crutzen, I would like you to comment again on that latitude curve
from the NCAR group where a very strong latitude dependence peaks at about 50°
N, revealing a very strong latitude dependence in the Northern Hemisphere (NH)
and virtually none across the interhemispheric zone. In contrast, other slides
17-2
-------
showed the two hemispheres to be fairly uniform and a discontinuity to exist
across the hemisphere. These questions should be addressed by the people in
this room who were involved in the measurements. Dr. Rasmussen, why don't you
start off?
Dr. Rasmussen: Well, three questions are to be answered. One, can we or can
we not measure MCF? In automated, routine gas chromatographic analysis, we
use 5-ml ambient air samples to determine specific halocarbon concentrations,
as resolved on a silicone oil column. A particular peak can be verified as
MCF by running the sample on dissimilar columns with dissimilar substrates and
differential retention times; and classic gas chromatographic technology will
confirm that it is, in fact, MCF. As you heard yesterday, in the urban environ-
ment, some contamination with ethylene dichloride (C1CH_CH_C1) is possible.
However, in the true rural situation, nothing has ever been observed under-
neath the MCF except MCF. This scale can be expanded to direct analysis, and
the same amount of MCF is measured. The precision of analysis is evident
again and again and again; it is something on the order of ±1.2 ppt. So the
answer to the first question is "Yes, we can measure MCF." We use the same
technology used successfully in other laboratories. In the primary standards
we have been using for ~4 yr, the change in MCF has been no more than ~±4 or 5
ppt over a period of time. It is a random change, and it has no directional
drift. So I feel quite confident that we can measure MCF.
I would also like to respond to your comment regarding the differences
between our ground measurements and measurements in air. Joe Krasnec and I
collaborated on the GAMETAG flight, and our different results on the same
flight, I feel, reflect some analytical difficulties caused by operating a gas
chromatograph on an aircraft. We prototyped together on two different Learjet
flights, one in 1975 and one in 1976. He ran the system in a Convair 990, and
the original work demonstrated it was impossible to get the precision of anal-
ysis with the airborne instrument that could be achieved in carefully collect-
ed paired samples. Only time and effort will resolve the true discrepancy in
our values from the 1978 GAMETAG flight across the Pacific.
17-3
-------
The second question is: "How accurately can we measure MCF?" The tech-
nicians in my laboratory, Peter Simmons, Jim Lovelock, and I have found agree-
ment on ambient values, the precision at those ambient concentrations, and an
estimate, our best professional judgment, of the absolute accuracy of the
standard. During the entire summer of 1975, several people, including Joe
Krasnec, intensively prepared dilution standards of these compounds, both by
permeation tubes and with static pollution. We perpetuated these values from
the fall of 1975 to the present. We did more work on the primary calibrations,
in at least three informal and one formal interlaboratory calibration, encom-
passing ~25 laboratories. These scientists are experienced in making halo-
carbon measurements, and we agree very well. Our laboratory and Lovelock's
laboratory agree within ~5 percent on most of the exchanges. This might be
considered collusion, because we intercalibrate so often, but we agree within
a few percent; with NOAA's laboratory now and Kirby Hensen's group
DP. Sakiff: On MCF?
Dp. Rasmussen: Excuse me, on the fluorocarbons. On MCF the agreement is
maybe within ±10 percent. No significant difference in MCF values was found
between my laboratory and Paul Golden's laboratory at NOAA. A discrepancy
arose between Leroy Heidt's measurements and ours in 1977 in a very intensive
set of precalibration exercises, when GAMETAG first started.
Dp. Sckiff: Why does the Golden group submit they cannot make a standard
sample of MCF or keep it for any length of time, and are unable to get a
correlation calibration of standard with Leroy Heidt? we were told this
yesterday.
Dr. Rasmussen: That is news to me. Paul Golden is in Germany; when he re-
turns, I will discuss the standard with him.
In a few days, preparation and packaging in our laboratory will be
completed of calibration samples in exceedingly clean, highly polished,
passivated stainless steel bottles. The samples will be sent to 12 par-
ticipating laboratories with a concentration standard of dissimilar
17-4
-------
concentration, a checklist on exactly how and where the samples are run, and a
request that the samples be run on 2 separate days by either the same analyst
or a separate analyst. Reproduc-ibility will be evaluated. We will have the
conditions of analysis, the column of the estimate type, frequency, detector
temperature, sample size, etc., in the report. The data will be returned to
the investigator, who will have the submitted value.
To assure that these samples do not change during the interim of trans-
port from one lab to another, each lab will receive its own set of samples.
Participating laboratories will be asked to return the samples within 30 days
after receipt, for reanalysis in our laboratory. Results will be compared
with data obtained before the samples were distributed. When they are re-
turned, all samples should be identical. If any problems occur, the canisters
can be refilled and reshipped, if necessary, because we now have ballast tanks
primary large quantity, high pressure standard of these materials in reser-
voirs.
So this is an area of activity. With all the different opinions repre-
sented here, however, reconciling the question you raise is very difficult.
I feel quite confident that we can measure MCF and that we can do a pretty
good job of it.
DP. Sohiff: On your N/S ratio, will you comment on the difference between
your numbers and the NCAR values, and on that increase-with-time difference
between you and Dr. Rowland?
DP. Rctsmussen: First, a background summary of our global trace halocarbon
measurements might be useful. Measurement was begun in 1974. These GAMETAG
profiles are from May and June 1978 in the Yukon; essentially the same flight
profile was carried out in 1977. For GAMETAG we piggybacked on the NCAR study
with several flights across the Atlantic to fill in some information gaps. We
have a sample exchange program in Tasmania, New Zealand, and Samoa; in Barbados
we are exchanging sample flasks to obtain further monthly values. At the same
time, at Oregon Graduate Center we are directly responsible for day-to-day
operation of the lifetime experiment sponsored by the Manufacturing Chemists
17-5
-------
Association (MCA) at the Samoan and Tasmanian stations, where measurements are
made 24 hours/day, 7 days/week; calibration of ambient samples is all elec-
tronically processed, and the data are shipped to us by the agent at the
station. So it is a fairly integrated program.
Now, I will not give you numbers obtained from the real-time measurements
at Tasmania, or Samoa, or Barbados; to focus on that program would be somewhat
premature. However, it should be emphasized that the data we are obtaining in
Samoa and Tasmania and those Peter Simmons and Jim Lovelock are obtaining in
Barbados are entirely consistent with our sampling profile. The data are
taken from low pressure samples obtained by directly flushing and pumping up a
very well passivated and cleaned stainless steel bottle or from cryogenically
collected samples accomplished by liquifying large parts of air into a small
vessel, resulting in a very high pressure, large volume sample.
You asked about global concentrations and assessed numbers. Ratios be-
tween the NH and Southern Hemisphere (SH) can be confusing, because most
values have been just arithmetic averages of what the investigator considered
to be the concentration that best fits or represents the NH or SH. These
products are prepared from a data bank of >100 data points. Time-weighting,
when data were obtained within a given year, obviously caused some skewing,
but these calculations provide an approximation of the main concentrations, as
integrated for the volume of air in the NH and SH for these different years.
The data, we believe, are real. We do have a good handle on the calibration,
yet the small difference in MCF concentrations between 1976 and 1977 remains
inexplicable at this time.
Dr. Schiff: Do you find altitude effects?
DP. Rasmussen: In the original May 1975 flights from Alaska reported at a
meeting in Greenbelt, Maryland, I did not doubt that on individual flights,
with carefully collected paired samples, between altitudes from 10000 to
~40000 ft (at the tropopause proper) a ~3 percent falloff is observed in the
level of MCF or these other four carbons, especially FC-11 and FC-12, because
that was the entire focus of the experiment. I cannot attest to this on
17-6
-------
subsequent flights, because I was not present at all of them to oversee each
of the measurements.
Dr. Sohiff: Can you respond to this point? I was told you said that ground
measurements and aircraft measurements differ by 30 percent.
Dr. Rasmussen: That is an error. I do not know where that statement arose.
DT-. Singh: The 1976 data I have seen for MCF show a ratio of not 1.23, but
2.2. What happened?
Dr. Rasmussen: This was arithmetic type. David Pierotti can speak to that.
Mr. Pierotti: We used data from 410 samples, not including the Helex data
from continuous measurements on board ship. I am not sure of the reason for
the difference, but one explanation may be the number of measurements. Our
graph of 410 numbers is similar to graphs other people showed; however, they
derived different NH and SH distributions. Other investigators used 13 or 17
numbers. I can use 13 or 17 numbers to pattern any distribution desired
from no change to a gigantic change. When 400 numbers are available, a better
idea of the actual distribution can be patterned.
Dr. Singh: You haven't answered my question. Why was it wrong? In preci-
sions like we are talking about, why was 1976
Dr. Rasmussen: Well, you are pushing on a sensitive point. You asked for it
and I will give it to you. David Pierotti made the nitrous oxide (NO) mea-
surements on the cruise. Joe Krasnec made the MCF measurements on the cruise.
The experimental setup was operated on the same estimate with dissimilar
columns and dissimilar values, so each investigator had control over his own
system. Why the MCF data from the 1976 Helex cruise yield an average for
other values I cannot testify, any more than I can explain why on the same
flight Joe Krasnec and I obtained dissimilar values in our graph samples and
cryogenic samples versus his real-time measurements.
17-7
-------
Voice from Audience: I have all the original data, the original chromato-
graphs, and I will review them. We do see longitudinal differences in MCF
concentration, particularly in the NH, and certainly it could be that in a
particular track of this cruise we were getting higher values in the NH.
Dr. Singh: This is very similar to what we have. I was curious and con-
cerned: Jim Lovelock reported 2.2.
Dr. Rasmussen: I think it largely has to do with some ways in which the
instrument was operated. Also, the value for the NH was skewed by the values
taken shortly after leaving port in Los Angeles on the Helex en route to Peru.
This was not a truly representative value for MCF in the NH. This has been
one of our difficulties, and it is because of the large variability in MCF
measurements between the latitudes of ~20° to ~60° N. I say these truly
represent the mixing ratio in the NH. This may also relate to part of Dr.
Schiff's misunderstanding concerning my rumored quote of a 30 percent dif-
ference between land-based samples and aircraft samples. Obviously, the land-
based samples are collected at a latitude of ~45°.
Dr. Schiff: I was told that whenever you repeated measurements at the same
place, by surface and by airplanes, you got different numbers. Is that true?
Dr. Rasmussen: No. We are currently testing this continental offshore effect
again this week and for the next 2 weeks by piggybacking on the NCAR flight
over the North Pacific, relating our findings to Hobb's cycle study, and by
flying into and out of storm centers in the Gulf of Alaska. MCF values ob-
tained on the Oregon coast will be compared with those obtained several hun-
dred or a thousand miles offshore of the Pacific Northwest. At the same time,
we will be intercalibrating and exchanging values with our colleagues at
Washington State University (WSU). I don't think any difference exists in our
MCF measurements on the coast of Oregon and their countryside measurements in
eastern Washington. As of January 1979, the typical value for the station was
~135 ppt. I think Dr. Cronn said 123 yesterday, so I don't really see any
great discrepancies.
-------
Voice from Audience: Does the big difference result because our higher values
were obtained over the North American continent and yours were obtained at the
same latitude, at the same time, but over the Mid Pacific?
Voioe from Audience: It looks like a latitude effect, but it really was a
continent versus an ocean effect, because at 30° N they were over North Ameri-
ca. South of that, they were over the Pacific. So it looked like a sudden
jump at 30° latitude, but it was actually the effect of suddenly coming over
the land, where all the MCF is emitted. The highest numbers, there, are all
from North America.
Dr. Schiff: That kind of variation suggests a short lifetime, and that wor-
ries me. If you get those kinds of gradients over land masses, that indicates
a very short lifetime.
Voioe from Audience: There wouldn't be anything in the SH if that were true.
Dr. Schiff: Take a look at Joe Krasnec's curve, and take a look at the Rasmussen
curve of the same flight. You see a very small gradient across the equator and
a huge gradient peaking at ~50° N. A gradient like that certainly indicates a
short lifetime.
Voice from Audience: I am sure that is a continental effect.
Dr. Schiff: Why the gradient over the continent, unless our Eskimos are using
a hell of a lot of MCF?
Voice from Audience: This is on the same flight.
Dr. Crutzen: I would like to comment on the latitudinal cross section dis-
played yesterday. In the two-dimensional modeling effort, I also found that
ground-level MCF mixing ratios reach a maximum around 50 to 60° latitude, and
this may relate to a first issue to be considered in resolving discrepancies.
The emission per unit area, I think, is one factor which we may easily forget.
17-9
-------
After all, the surface area at 60° is smaller than at 40°. Also, the destruc-
tion rates of MCF are much smaller at those latitudes. This raises the possi-
bility that photochemical effects play an important role.
Dr. Sahiff: Perhaps the differences can be explained, but I am trying to play
the devil's advocate to discover if real discrepancies in the measurements do
exist.
Dr. Cvutzen: May I read a few lines from this handout on Leroy Heidt's and
Joe Krasnec's experimentation:
The graph sample analysis in the past 2 years confirms the
problem involved with the integrity of grab air samples. Methyl
chloroform and carbon tetrachloride could not be measured accurately
in the past due to the absorption problems on the inner surfaces
of the sampling vessels. Difficulties in making reliable methyl
chloroform measurements also involve choice of materials for sampling
containers stainless steel, aluminum, or glass, etc. and sampling
system contamination by pumps, inlet systems, etc. For these
reasons the group is making measurements on the ground and in
the air. No pumps are used in the all-stainless-steel GC
sampling system. Finally, another difficulty encountered in the
course of making long term methyl chloroform measurements is the
stablity of calibration standards.
Now, this comes back to what you just said. Significant changes (in this case,
decreases) were observed in the concentration of different calibration mixtures
over a period of several months. One solution appears to be the storage of
secondary calibration mixtures in larger vessels and under high pressure
several hundred psi. Also, preparation of primary standards using dynamic
dilution techniques appears to be the preferred method.
Dr. Rasmussen: I would like to comment on the stability of these vessels.
The vessel in which Heidt and Krasnec are observing absorption on the wall is
typically the NCAR high vacuum, ultra clean canister, which draws in an am-
bient sample through the probe system on the aircraft. At our March 1976
halocarbon workshop we discussed ad infinitum the difficulties of obtaining
reliable stability in evacuated canister samples. There is a host of examples
in which everything disappeared, even FC-11 and FC-12, in some of these evacu-
ated canister samples. The samples we use are pressurized typically to ~30 psi
17-10
-------
in the grab sample phase, and we have never had any problems with stability in
the electro-polished, passivated, stainless steel vessels. These vessels,
which I prepared and developed at WSU, are used extensively by people at WSU;
at SRI they are purchased from a small machine company in Pullman that was our
manufacturing outlet. Have you had any problems with MCF and these stainless
steel vessels?
DP. Singh: We have not seen any deterioration, but we have had some unex-
plained problems, so the integrity, I suppose, could go either way.
Dr. Sahiff: I don't think it is a matter of pressurizing or not, because I
think Dr. Rowland will say that he can get integrity with samples that are
evacuated.
DP. Rowland: Pressurizing involves the possibility of contamination.
Dp. Sohiff: I don't think any problem can be traced to pressurizing or not
pressurizing. In reference to the comment that you are getting good agreement
with Jim Lovelock, I would like to mention a letter from Jim Lovelock, dated 9
January 1979, in which he states, "I am not surprised that there are uncer-
tainties about atmospheric abundance of the MCF. It is very difficult to
prepare standards of it because of the reactivity."
Dp. Rasmussen: Well, Jim Lovelock is telling you things he is not telling me.
Dr. Sohiff: Now, can we hit a couple of these other points? Dr. Rowland,
maybe you would like to comment on the increase of MCF with time, and, Dr.
Rasmussen, on your trend of the NH/SH ratio. As I remember, you found the
interhemispheric ratios were rapidly approaching one another. Is that still
the case?
Voice from Audience: That was, I think, at 45° N versus Antarctic ratio.
Dp. Sohiff: NH/SH ratios are coining closer to unity with time.
17-11
-------
Dr. Rasmussen: These data from the World Meteorological Organization meeting
have been updated to January 1979. As David Pierotti said, these were compari-
sons between the average representative concentration of those species for the
month of January each year, characterizing the specific Northwest, typically
45° N, and the values that we were obtaining at the South Pole, typically 90°
S. It is a very simplistic comparison of the arithmetic mean measurements,
typical of January of the appropriate year. In the MCF measurements from 1976
to 1979, it can be seen we have maintained continuity and consistency in the
calibration comparable to the consistency we maintained at WSU, where the
standards were originally prepared. As far as I know, no drift at either,
laboratory can be seen in the ratio of 1.72, 1.53, 1.38, and 1.42 of the NH/
SH maximum difference, as contrasted in these two sites where the analyses
were made. These are not hemispheric averages. They are not integrated for
the hemispheric burden.
DP. Sohiff: The point is they are coming closer.
Voice from Audience: Between 1976 and 1977 the ratio appeared to drop, and I
am not sure why.
DP. Sohiff: Would you like to comment on that difference in time, Dr. Rowland?
DP. Rowland: We don't have a long series of data. All we have are South
America data from a period 8 or 9 months apart. There is a difference, of
course, to a value of 9 percent. That is not a rigorously accurate figure.
Dr. Sohiff: You are saying there is no significant difference between your
numbers and theirs?
DT. Rowland: Since their numbers fluctuate sufficiently from one year to the
next, I doubt that a significant difference exists.
Dp. Rasmussen: I think the data we heard this morning from the manufacturers,
describing a relatively small increase in total production of MCF during the
17-12
-------
last few years, are fairly consistent with our observations. However it is a
very limited data base.
DP. Singh: It is not consistent with the emission data you have available.
DP. Rowland: Using the Dow Chemical emission data, going N to S, in conjunc-
tion with any of these removal rates in the NH and SH and the day rate of
mixing between the hemispheres, yields a ~1 percent/yr increase over the last
3 yr. Values for a lifetime of 8 to 12 yr are not important here; the Dow
emission data feed into a number on the order of 12 or 13.
Voice fpom Audience: I heard two people this morning say that the MCF produc-
tion levels scarcely changed during the years 1976 to 1978. I heard two
different people mention a figure like 630 million.
DP, Rowland: That is U.S. production, rather than emission.
DP. Schiff: Is there a big difference?
Dr. Singh: The U.S. has a global difference and it does not amount to a
ratio.
Dr. Rowland: None of us has any emission data.
Voice fpom Audience: It hasn't increased in the U.S. Has it increased that
rapidly in the rest of the world in the last 2 yr?
DP. Rowland: You will have to get the answer from Dow Chemical.
Voice fpom Audience [Dp. Fapbep?}: Does somebody want me to comment on the
Dow data? In the little paper that I sent to all the attendees I knew were
coming, Figure 3 shows our best estimate of emission. In some cases, we
subtracted inventory we knew to be building up, and we subtracted what we
thought would go to chemical use or be not emitted. This estimate is probably
accurate to within 5 to 10 percent, based on the fact that we think we know
17-13
-------
more about the use and production of the product than anybody else. I don't
know anything about Russian production.
Voice from Audience: What are the figures for 1976, 1977, and 1978?
[Dr. Farber?]: In 1976 the world emission of FC-111 was estimated at 894
million Ib; in 1977 it was ~981 a ~10 percent increase. In 1975 it was
764; in 1974 it was 871, reflecting a decrease in 1975.
Voice from Audience: As with most other chemicals, this was caused by eco-
nomic conditions.
Voice from Audience: what was 1978?
[Dr. Farber?]: It was close to 1 billion Ib in 1978. Dr. Neely, do you know
what the number was?
[Dr. Neely?]: A little over 1 billion.
[Dr. Farber?]: Less than 1.1, anyway.
[Dr. Neely?]-. 3 or 4 percent, maybe.
[Dr. Farber?]: Yes, that is our estimate. The U.S. numbers are: 1974, 495;
1975, 443 (again, that reversal due to economics); 1976, 487; 1977, 521; and I
don't have the 1978 number.
Dr. Singh: I would like to make a comment on Eastern Europe and the Soviet
Union. I have nothing to report other than a very qualitative statement that
MCF emissions are extremely low. The estimate is something like 2 percent.
Voice from Audience: Are any SH data available?
Voice from Audience: I don't have the figures for MCF, but I looked at a
document on FC-11 and FC-12 which showed where they were sold and assigned by
17-14
-------
latitude, and so on. About 10 yr ago, the amount sold in the SH was ~3 per-
cent. By 1973 to 1974, it had increased to maybe 4 percent: manufacturing of
FC-11 and FC-12 in the SH had started. So FC-11 and FC-12 have been increas-
ing to as much as 8 percent in the SH now, but I have no corresponding figures
for MCF.
Dr. Singh: Our model calculations may be 4 percent.
Voice from Audience: Four percent that is probably good enough.
Voice from Audience: My estimate for fluorocarbons is 90 percent between 30
to 50° N, 5 percent N of that, 2 percent between 30° N and 30° S, and 3 per-
cent from 30 to 40° S.
Dr. Singh: Four percent?
Voice from Audience: Yes, 4 percent would be right.
Dr. Schiff: Are there any other comments about the measurement aspect of the
problem?
Voice from Audience: It seems that several representative measurements have
been discussed under this general question. Do the investigators feel there
is enough quality assurance in calibration for this sort of thing?
Dr. Sohiff: I think the consensus is that there is not. I think what you are
saying is that everyone feels his own accuracy is quite high; however, the
figures given here reflect a bigger spread than the sum of the accuracies
cited. For instance, the NH value numbers we have heard have ranged from 94
to 145, although accuracies within ±10 are claimed.
Dr. Rasmussen: We are "mixing apples with oranges." The interlaboratory
calibration published on MCF this past year said that, among the laboratories
measuring these standards, the spread was pushed from -39 percent around the
submitted value of 100 ppt. Enough samples came back with something left in
17-15
-------
those canisters to verify that we were at the 100 ppt level on the return
sample. I have described our estimated absolute accuracy according to Peter
Simmons, Jim Lovelock, Paul Fraser, and myself. The only other laboratory
that I would include in that is the WSU laboratory, because I feel fairly
confident in their capability to make these measurements.
Now, I preface this by stating our general good agreement on halocarbons
with NOAA, the Paul Golden laboratory, and the one previously operated by
Thompson. On the other hand, on three different occasions, irreconcilable
calibration problems arose between my laboratory and NCAR's laboratory, and
at the halocarbon workshop, where we made these onsite sample analyses in
Paul Golden's laboratory. Again, we have problems reconciling the values
that Leroy Heidt presented; you were at the meeting, Dr. Schiff, and can
testify whether that is an overstatement or understatement of fact. This
was the second or third week of March 1976; David Pierotti and Joe Krasnec
had just come back from the Helex cruise. The 10 percent described does not
reflect what the field analysts, in general, can achieve, and it does not
describe overall results among laboratories. The presentations reflected
many differences to be reconciled.
Dr. Schiff: The "apples" that I am comparing are the "apples" of this group,
who are measuring in the clean atmospheric air. I am saying that the spread
between the numbers we have seen is greater than the sum of the accuracies
claimed by the individual group. That, I think, is a real problem. I think
we have to somehow resolve it. I also am rather disappointed that we do not
have represented here (other than what Dr. Crutzen read) the groups who are
having real difficulty and who are claiming that they cannot calibrate with an
accuracy of 50 percent and cannot get standards with 50 percent reproduci-
bility. So I am sorry that Leroy Heidt and Joe Krasnec and Art Schmeltekopf
are not here. I am also rather sorry that Doug Davis isn't here, since he was
responsible for the GAMETAG flight structured with two groups measuring the
same latitudinal region. My summary of the situation is that we have groups
here who claim they can measure MCF, and there are differences among them I
mean real differences. I think there are also groups that aren't represented
here who have found even greater differences.
17-16
-------
Dr. Singh: The subtle differences that we are talking about ought to be put
into some perspective. We have 100 ppt in the temperate latitudes and a life-
time calculation of 6 to 12 yr. As long as we understand that, the decisions
ought to be made on that. So from a scientific point of view, sure, we have
all these problems, but they all point to this narrow range.
Dp. Sahiff: I think you are quite right. In the regulatory community, deci-
sions can be based on present atmospheric concentrations of MCF, and, of
course, I am not suggesting any critical difference between 94 and 135 ppt.
But to go beyond that, it is important that we at least know where we stand
in the scientific community with respect to our ability to measure MCF and
with what kind of accuracy. At the moment, I do not feel I have a solid and
reliable answer to this question.
Voioe from Audience [Mr. Pierotti?} -. I would like to comment on the differ-
ences in standard deviations in the measurements. Our 1976 measurements in
the NH range from 160 to ~80. That is a factor of 2, a significant differ-
ence, and that is with all the same sampling methods and the same standards.
Dr. Schiff: In clean air?
[Mr. Pierotti?]: It is difficult to know what you mean by "clean air." Is
there any "clean air" over North America or over the Atlantic Ocean? We
measured a lot of other compounds, too. In cases where everything is elevated
(the fluorocarbons, for example), analysis is possible. But I think the dif-
ferences occur because some people have a very small data base that will fit
an unlimited variety of distributions.
Dr. Schiff: If, indeed, there is a big variation of a factor of 2, then it is
saying something very real about the lifetime of that compound.
Dr. Rowland: If there is a difference between taking 13 points out of 400 and
13 out of 13. We collected 7 samples in South America and analyzed them in
the laboratory. The spread of deviation for MCF in those was 65 ±2 for the
same sample in 7 different locations taken over a period of ~1 week. The
17-17
-------
measurements repeated for a several-month period showed no appreciable devia-
tion, and it looks as though we took 7 samples out of the same year. Now,
that is not very many data points, but there is no spread at all. I suspect
that is what you can get, if you pick up oceanic samples that have no con-
tamination .
Dx>. Sohiff: And I worry about that factor of 2 spread in the so-called "clean
air." If it is true, then that compound must have a very short lifetime.
DP. Singh: When you talk about spread related to lifetime in this case,
where the emission sources are very narrow I would not necessarily jump to
that conclusion too quickly.
Dp. Sckiff: I would argue that with you. Consider, for instance, that the
air mass you measure today at a given location may have come from a different
place and yet you are still saying it is "clean air," not contaminated air,
and that the measured values can only reflect a latitudinal dependency.
Dv. Singh: I see a number of data points, and the distributions are very
similar.
Voice from Audience: Well, our numbers and yours are very close.
Dr>. Singh: We have measured in two different longitudes, and perhaps be-
cause of the mixing differences in the two hemispheres and perhaps because of
the seasonal differences there are some real changes in these distributions.
None of us is really sure of the sections of the two hemispheres that have
been done at one point in time. I think that should be stressed, and some of
these differences, perhaps, are real.
Voice from Audience: I agree. The distribution seems too complex, and in
particular does not seem to square with the idea of a long lifetime. It is
surprising and it does seem to point to something we don't understand about
the behavior of MCF in the atmosphere-. It tells us we don't know as much as
we think, perhaps.
17-18
-------
DT. Schiff: That may be a good note to end on.
Dr. Rasmussen: I would like to make sure that no one copies this figure of
137 ±55 for November. That is a typo; it is ±15. The ±1 a for 45° N is real,
and it is always ±10 to 15 ppt for MCF at that site. We made a few fluoro-
carbon measurements at this site, and they are exceedingly constant. Much
more variability in MCF is seen than in the fluorocarbon measurements or in
the SH. Those data are ±5 to 7 at Samoa and ±4 to 6 at Tasmania. I agree
with you, Dr. Rowland, that when you get into the SH and Antarctica, you could
get this very tight tolerance; in the NH, however, we don't get a tight tol-
erance on MCF measurement.
Dr. Rowland: Let me say one other thing. We spent about as many manhours, I
suspect (or close to it), and we have very few points. They are looking at
a different kind of measurement, and they are collecting many measurements
over a wide area. We are focusing on one particular thing and spending time
striving for better accuracy. We have different aims. I don't think, from
the point of view of a regulatory agency, that it makes any real difference
whether it is 100 or 135.
Dr. Schiff: No, I don't think so either, although I want to stress that
measurements are important and measurements are not getting enough exposure at
this meeting. I wouldn't like to see EPA walk away thinking that we have got
the measurement game well in hand. I think there are still some unanswered
questions, and that is the only message I wanted to get across.
Voice from Audience: I agree that perhaps we can explain the variability in
some of these data by the possibility that the residence time is much shorter
than earlier estimated. I think that needs to be investigated.
Dr. Rowland: A variability should, therefore, exist in every place. If you
measure the variability of the SH and it is not the same as in the NH, then
you cannot suggest a short residence time. To state a real variation in the
model, you must see it every place.
17-19
-------
from Aud-ienae: May I make one other comment to which Dr. Singh may want
to respond, because it concerns his data? I think that something can be neg-
lected here, which not only has to do with these measurements, but with all
measurements of trace contaminants. When something is 3 or 4 days away, that
is damn hard to prove. Talk to other experts. Talk to the meteorologist in
this field. In any bit of data I study, I don't see (with some exceptions)
this projectory analysis. Trifluoroethylene shows up in remote sites. Even
in the OH-trichloroethylene kinetics, the O type, or the smog change, it
shouldn't be there, as far as I can see. Why is it there, if it isn't a con-
taminated air mass to some extent? Would you want to comment on that?
DP. Singh: 1 agree with you.
Dr. Sehiff: Well, thank you. Thank you all for coining.
DTP. Hanst: Well, we devoted the Panel Discussion to one topic. Mrs. Bishop
has asked if it wouldn't be possible for a summation. I was thinking of
something Dr. Molina told me, and this might close it out. When we look at
the different molecules and count the Cl, that is one indication of the de-
gree of danger to the stratosphere: the combination of Cl and lifetime. FC-
11 holds the greatest danger (FC-11 has three Cl, PC-12 has two, and fluoro-
carbon-22 (chlorodifluoromethane, CHC1F ) has only one), and they all have
the same lifetime in the air. Then that is the direction we should go in;
that is the summation.
17-20
-------
APPENDIX A
DISCUSSION PARTICIPANTS
A. P. Altshuller
Environmental Sciences Research Laboratory, MD-59
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Ferial S. Bishop
Office of Toxic Substances, TS-794
U.S. Environmental Protection Agency
Washington, D.C. 20460
Frank A. Bower
E. I. du Pont de Nemours & Company
Chestnut Run
Wilmington, Delaware 19898
Joseph J. Bufalini
Environmental Sciences Research Laboratory, MD-84
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Malcolm Campbell
Air Pollution Research
Washington State University
Pullman, Washington 99164
Dagmar R. Cronn
Air Pollution Research
Washington State University
Pullman, Washington 99164
Paul J. Crutzen
National Center for Atmospheric Research
Post Office Box 3000
Boulder, Colorado 80307
Hugh Farber
The Dow Chemical Company
Midland, Michigan 48640
A-l
-------
Farley Fisher
National Science Foundation
Washington, D.C.
Philip L. Hanst
Harvey Mudd College
Claremont, California 91711
Julian Heicklen
Department of Chemistry
The Pennsylvania State University
University Park, Pennsylvania 16802
Mike Humenny
California Air Resources Board
Post Office Box 2815
Sacramento, California 95812
Peter Jesson
E. I. du Pont de Nemours & Company
du Pont Experimental Station
Wilmington, Delaware 19898
Robert Kellam
Office of Air Quality Planning and Standards, MD-12
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
David Klauder
Bureau of Foods, HFF-407
U.S. Food and Drug Administration
Washington, D.C.
Thomas Lapp
Midwest Research Institute
425 Volker Boulevard
Kansas City, Missouri 64110
Jennifer Logan
Center of Earth and Planetary Physics
Harvard University
Cambridge, Massachusetts 02138
Stanley C. Mazaleski
Office of Toxic Substances, TS-769(M)
U.S. Environmental Protection Agency
Washington, D.C. 20460
W. Brock Neely
The Dow Chemical Company
Midland, Michigan 48640
A-2
-------
Joseph Padgett
Office of Air Quality Planning and Standards, MD-12
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
David Pierotti
Oregon Graduate Center for Study and Research
19600 Walker Road
Beaverton, Oregon 97005
Rei A. Rasmussen
Oregon Graduate Center for Study and Research
19600 Walker Road
Beaverton, Oregon 97005
Courtney Riordan
Office of Research and Development, RD-682
U.S. Environmental Protection Agency
Washington, D.C. 20460
F. Sherwood Rowland
Department of Chemistry
University of California Irvine
Irvine, California 92717
Harold I. Schiff
Department of Chemistry
York University
Downsview, Ontario M3J 1P3
CANADA
Louis Schlossberg
Detrex Chemical Industries, Inc.
Post Office Box 501
Detroit, Michigan 48232
Hanwant B. Singh
SRI International
333 Ravenswood Avenue
Menlc Park, California 94025
Arlen- Slobodow
U.S. Consumer Product Safety Commission
Room 656B
Washington, D.C. 20207
Igor Sobolev
Kaiser Aluminum Chemical Corporation
Post Office Box 877
Pleasanton, California 94566
A-3
-------
Ken Surprenant
The Dow Chemical Company
Midland, Michigan 48640
Robert T. Watson
Jet Propulsion Laboratory
4800 Oak Grove Road, Building 183-601
Pasadena, California 91103
A-4
-------
TECHNICAL REPORT DATA
"/< a si read Instruction*; on the rcrctsc bcloic t
1 REPORT NO
EPA-600/9-80-003
4 TITLE AND SUBTITLE
PROCEEDINGS OF THE CONFERENCE ON METHYL CHLOROFORM
AND OTHER HALOCARBON POLLUTANTS
5 REPORT DATE
January 1980
6 PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Joseph J. Bufalini (Editor)
8. PERFORMING ORGANIZATION REPORT NO.
3 RECIPIENT'S ACCESSION-NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Same as block 12
10 PROGRAM ELEMENT NO.
1AA603A
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory, RTP, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Presentations at the Conference on Methyl Chloroform and Other Halocarbon Pollutants
(Washington, D.C,, February 27-28, 1979) are documented. Included among the authors
are research scientists, industry representatives, and regulatory officials. The
16 papers fall into 2 basic groups. The first 10 papers present results of
research in atmospheric chemistry as related to the question of stratospheric ozone
depletion by halocarbons. Drawing upon atmospheric measurements and model cal-
culations, the authors give estimates of emission levels, current atmospheric
burdens, tropospheric lifetimes, the importance of sinks, effects on stratospheric
ozone, and related questions. The final 6 papers take the perspective of involve-
ment in, or concern with, regulatory decisionmaking. The authors consider various
options, recommendations, and plans for halocarbon control in light of available
scientific data. Finally, the Panel Discussion which concluded the Conference
is presented in verbatim transcript form. Focusing on the current status of
atmospheric measurements, the participants discuss problems in obtaining accurate
halocarbon data, and discrepancies between and within the results of individual
investigators.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
* Air pollution
* Chloroethanes
* Halohydrocarbons
* Ozone
* Stratosphere
* Depletion
* Meetings
* Proceedings
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
13B
07C
07B
05B
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OE PAGES
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
A-5
------- |