GEOPHYSICAL
RESEARCH
LETTERS,
VOL.18,NO.4, PAGES
673-676,
APRIL1991
HETEROGENEOUS
REACTIONSWITH NaCIIN THE EL CI-IICffONVOLCANICAEROSOLS
DianeV. Michelange!i
SpaceScience
Division,NASA AmesResearch
Center
Mark Allen
EarthandSpace
Sciences
Divisibn,
JetPropulsion
Laboratory,
California
Institute
ofTechnology
Yuk. L. Yung
Division
of Geological
andPlanetary
Sciences,-California
InstituteofTechnology
Previous
investigations
of the effectsof the 1982erup-
changes
(observed
by Labitzkeet al. [1983]andmodelled
by Pollackand Ackerman
[1983])in a onedimensional
tion of the El Chichonvolcanocouldnot explainall the
observations
of changesin 03, HCI,'NO andNO2 simulta-
chemicalkineticsmodelof the stratosphere.While they
obtaineda decreasein ozonewhich agreedwith the ob-
neouslywithout proposing unproven chemicalreactions.
Sincereactionsbetweensolid NaCI and gaseousCINO3
servations
of HeathandSchlesinger
[1984]andChandra
[1987],theycouldnotexplainthemeasured
40%increase
rinespecies
and solidNaNO3 in laboratoryexperiments, in HCI [Mankinand Coffey,1984;B. W. Gandrudand
we suggestthat these reactions could have occurredwith
A. L. Lazrus,privatecommunication,
1983],as well as
the NaCI observedto be present in the El Chichon sulfu- the largeNO and1•102
decreases
[McFarland
etaL, 1986;
tic acidaerosols.As a consequence,
we predict that HCI
Roscoe
et al., 1986],evenafter includingdirectinjections
and N205 rapidly produce photochemicallyactive chlo-
shouldincreasesubstantially,'
while NO• shoulddecrease, of HCI, CI• and H20 from the volcano into the stratoin agreementwith the measurements
after the eruption. sphere.
Ozoneshouldonly be slightly affectedby thesereactions.
Therefore,
Michelangeli
etal. [1989]explored
theroleof
Reactions
betweensolid NaCI and the acidsH2SO4and heterogeneous
reactionson sulfuricacid aerosolsin reproHNO• might prove to be important, but we lack suffi- ducingthe availablemeasurements.This was suggested
cientevidenceregardingtheir efficiencyand the presence by the importanceof reactionsbetweenthe ice particles
of HNO• in the aerosols to be more conclusive.
of the polar stratosphericcloudsand gasphasemolecules
(N205, C1NO•,HC1,tt•O) for explainingthe Antarctic
Introduction
ozonehole[Solomon
et al., 1986;McElroy,1986;Molina
et al., 1987;Tolbert et al., 1988]. Usingthe l•boratory
The eruption of the E1 Chichon volcano in Mexico in
resultsof Rossieeal. [•987], To!bertet al. [1988]and
etal. [1988]for the efficiency
(orsticking.
coeffiMarch-April
1982[seefor examplePollacket al., 1983] Worsnop
provided
stratospheric
chemistswith an excellentoppor- cient,3) of reactionson H•SO•/H•O drops,Michelangeli
theeffects
of various
heterogeneous
tunityto test their modelsby trying to simulatethe effects eta!.[1989]simulated
of thisknownperturbationto the atmosphere.SO2was reactionsinvolving CIO• and 1•O.• speciesin their model.
directly
injected
intothestratosphere
[Krueger,
1983]and
rapidlyconvertedto sulfuricacid aerosols[Hoffmannand'
Rosen,
1984].The aerosol
cloud,locatedbetween16 and
30km[DeLuisi
et al., 1983]wasstillobservable
oneyear
aftertheeruption[Hoffmann
andRosen,1983;Thomas
et
They concludedthat the speculativeheterogeneous
reaction C!NO• -* HCI (with product NO• retainedin the
aerosol)wasthe most effectiveprocess.Other reactions
consideredled to changesin the oppositesenseto what
wasobserved
and/or in the correctsense,but too small
al.,1983;Adrianiet al., 1983;SpinhirneandKing,1985; in magnitude. Unfortunately,there are currently no laboratory resultsproving the direct heterogeneous
converJ•er andCarnuth,1987].
Michelangeli
eta!. [1989]investigated
the chemical sionof C1NO3to HCI. HoffmannandSolomon
[1989]did
changes
in the stratosphere
duringthefirstthreemonths a similar study of heterogeneousreactions with a twoaftertheformationof the aerosolcloud.Theyuseda one dimensionalmodel, but they did not include simultanedimensional
radiativetransfermultiplescattering
code ously the radiation and temperature changes:
Finlayson-Pitts
[1983]andFinlayson-Pitts
et al. [1989]
[Michelangeli
etal.,1990]
to evaluate
thechanges
in the
totalactinicflux after the eruption.The particleswere presentedresultsthat indicatedthat gaseous{g)
efficientforward scatterers which lead to a total radiaN205 andCINO3reactwith solid(s) NaC!in the follow-
tionincrease
ofupto 8%between
3000and8000/1.
in the ing manner:
aerosol
layer(formoredetails
seeMichelangeli
et'at.[1989].
+
andMichelangeli
etal. [1990]).Michelangeli
etal. [1989]
included
boththisradiation
change
and3øCtemperature
+
+
Copyright1991 by the AmericanGeophysicalUnion.
+
+
+
An early measurement(as quotedin Finlayson-Pitts,
Papernumber 91GL00547
1983)indicated
that• • !0'-ø-10
-7 forreaction
(1). Re-
0094-8534
/ 91/ 91GL-00547$03.00
673
674
Michelangeliet al.: NaCI Aerosols
centexperiments
IF.E. Livingston
andB. Finlayson-Pitts, abundancesin the non-perturbedphotochemicalmodel:
[NO:•]
= 2.6x 10ø cm-ø, [C1NOo]
= 4.4 x 10s cm
-'•,
1991]suggested
that -• for reaction
(1) is smaller
(10-s)
andthat -• > $ x 10-3 for reaction(3). No stickingco-
and[N2Oo]= 5.7 x 10s cm-3 at 28 km. Fromthese
efficientinformationis availablefor reaction(2).
numbers it is clear that the amount of NaCI will not be,
the limitingfactorin the ratesof reactions
(1)-(3).
able at the colderstratospherictemperatures(~225
Livingston
andFinlayson-Pitts
(!991),whilestudying
K at 28 kin, the peak of the aerosollayer). Fiulayson- reaction(3)concluded
that HNO0 (g) -t-NaC1(s)•
Pitts eta!. [1989],on the basisof Molinaet al. [1987], HCI (g) + NaNOo (s) might alsobe occurringin their
The reactionsproceedat 298 K, but no data are avail-
suggestthat the reactionsmight be moreefficientat low
temperaturesbecauseof the increasedresidencetime of
the specieson the particles. The reactionson the surface
system. However, no quantitative kinetic' rates were re-
are ionic in nature and should not have an activation bar-
coutdbe very effectiveat reducingNOx and increasing
rier. The lack of quantitative experimental information
HCI, it requiresthat HNOa be in, or in contactwith, the
aerosols,
whicharealreadyveryacidic,consisting
roughly
at low temperaturesinhibits detailed numericalsimula-
ported. Conceivablythis could also take place in the
stratospherewith very high efficiency.While this reaction
tionsof the effectsof reactions(1)-(3), but a qualitative of 75% H2SO4. This might prove to be the main limiassessmentof their possibleimpact is feasible.
l•esults
and Discussion
tation to the efficiency of this reaction.. So far there i•
no evidencesuggestingthe presenceof HNO• in volcanic
aerosols.The reaction would also lead to a significantre-
ductionin HNOa whichwasnot observedafter the erupThe first question we need to addressis whether or not
NaCl is present in volcanic aerosols. The only direct evidence for the presenceof halite in the E1 Chichon aerosol
tion [MankinandCoffey,1986;Arnoldet al., 1990].
All threeproposed
reactions((1)-(3)) takenewlyin-
jected CI from NaCI and convertit to photolyrically
unstable chlorine-containing
compounds(CId, NOCI,
[1985]usinga quartz crystalmicrobalance
cascadeim- C!NO•). The CI producedis activeand canpossibly
lead
pactor on a U-2 aircraft between 18 and 21 kin. On May
to an increasein ttCI, the most stable chlorine-containing
5•n, 1982,theyestimatedthat 7%ofthe total aerosol
mass species.Furthermore,
reaction(2) hasthe addedadvancloud
is from
the in situ measurements
of Woods et al.
(or0.5/•gm-3) wasNaCl. Nohalitewasmeasurable
dur-
tage of taking CI from the more inert CINOa and converting it to reactive C12. Therefore, for every NaCI tha•
[1985]claimthat the detection
of haliteanditsapparent reacts with CINOa, we produce 2 CI atoms. This would
disappearance
by Julycouldexplainthe largealepolariza- lead to a very large sourceof extra chlorine,and therefore
could lead to an HCI increase.
tion ratios measured in early May 1982 by Hayashida et
As far as increasing HCI is concerned, there are two
al. I1984]and a returnto normalvaluesin lateMay.
ing flights in July, November and December. Woodset al.
Woodset al. [1985]explainthe disappearance
of NaCI
by invokingthe reactionH2SO4+ 2NaC1(s) --. Na2SO4
(s) +2HCI. To validate their theory they searchedfor
Na2SO4(s) in their samples,
andonlyfoundit presentin
differenceswith the casespresentedby Miche!angeliet
al. [1989]. First, reaction(2) differsfrom the reaction
betweenCINOo and HCI by not destroyingHC1 directly,
and by introducingan extra C1sourceto the stratosphere.
the May 5th samplewhen it was reexaminedin November. Second,in the previous work, the simulations of a direct
C12or HCI injection into the stratospherefailed to proThis suggeststhat the reaction occurred within the sample after it was collected, but doesnot prove that it was duce an overall HCI increaseafter three months, becatme
importantin the atmosphere.Woodset al. [1985]state the chemistrywas perturbed in a singlepulse,and then
that the NaCI started to convert to Na2SO4 only after
left to relax back to the steadystate. The largegradient
onemonth. Pinto et.al. [1989]arguethat this reaction in HCI wasquicklysmoothedout by diffusion. Pinto eta!:
might be limited by the sedimentation of NaC1 within a
[1989]alsosuggest
that HCI shouldbewashedoutrapidly.
month, ahd the lack of sufficient regenerationof H•SO4
In the casewe proposenow, the heterogeneous
reaction•
from SO2.
would progressat a slowerrate, releasingC1 graduallyto
While there is uncertainty regarding the presenceof
the stratosphere.It wouldthen be possiblefor a newstate
Na2SO4(s), aswell as the lifetimeof NaCI (s), it see•ns to be reachedafter three months. Even thoughNaC1was
clearthat halite wasstill presentat significantlyhighcon- injected in a pulse, observationsdiscussedearlier suggest
centrationsone month after the eruption. SinceH2SO4 is it was present in the aerosolfor at least one month. Connot thought to react during that time accordingto Woods. trary to this, both of the reactionsbetweenNaCI andacid
etal. [1985],it leavestimefor reactions
(1)-(3) to occur, (H2SO4or HNOa), if they occur,produceHCI directly,
and then helpsexplainthe disappearance
of halite by July.
Fromthe Woodset al. [1985]measurements
we deduce
that theconcentration
of [NaC1]
-- 5.15x 109.cm-3 near
20 kin. To obtainthe concentration
at 28 km (peakof
aerosolcloud)we usethe lidar measurements
of DeLuisi
etal. [1983]for the aerosol
scattering:
40%of theaerosols
were located at 28 km, while 7% were presentat 20 kin.
and probably very rapidly. This would then result in a
situation very similar to that of the direct'HC1 injection'
and thereforewouldnot be effectiveat increasing
ItC1
three monthsafter the eruption.
Reactions
(1)-(3)havetheaddedadvantage
ofhaving
an impacton NO• c•_emistry
aswell.. All threer•tionsconvert
activeNO• species
(NO2,CINOo,
N2Os)
to
A simple scaling leads to an NaC1 amount at 28 km of
NaNOo(s)in the aerosols
whichcanfall out of the atmo2.94 x 10'ø cm-3. While thisfigureis probablya large sphere.NaNOs will not return NO• to the stratosphere,.;'
overestimate,it givesthe order of magnitudewe are con-
sidering.We can then comparethis to the standardgas
suchaswould
theNOCI(g)
andCINO2(g)
produced
i•.:•:::
{1)and{3). Evenin these
tworeactions
NaNOo(s)
Michelangeli
et al.' NaCI Aerosols
produced
and,therefore,
theynfightbeeffective
in at
leastreducing
NO=. In addition,NO2 isnowdirectlyinvolvedin heterogeneous
reactions.In thiscase,a reaction
between
NaCI and HNO3 wouldbe very effectiveat reducing
NOs. Unfortunately,
thiscannotbequantified
for
lackof data on the reaction.
Theseresultsare extremelyuncertainowingto the unknownheterogeneous
reactionefficiencies
at lowtempera-
tures.However,
we canseewhethertheseprocesses
((1)-
(3))have
thepotential
ofbeing
important.
Theywillbe
in competition
with reactions
suchas
N205 + hi/ --• NO2 + NO3
(4)
675
cisely. Also, the presenceof NaCI in volcanicaerosols
has not been extensivelyreported in the literature, and
therefore remains an uncertainty. It is crucial that, in the
event of another volcaniceruption more in situ aerosol
measurementsbe taken, and that the abundance.of NaC1
be definitively established. In addition, it would be important to look for NaNO3 and Na2SO4 in the aerosols.
Since the ultimate fate of the NaNO3 is to fall out of
the stratosphere, a close examination of the Greenland
and Antarctic
ice cores would be worthwhile.
The acidic
layer correspondingto E1 Chichon has been detectedin
Greenland ice cores. It would be interesting to look for
the presenceof NaNO3 or Na2504 after the eruption.
Unfortunately, due to the rapid fall times limiting the
OH+NO2+M
• HNOs+M
(5)
CINO,+h• • CI+NOa .
(6)
amountof NaNO• (s) and Na•SO4 (s) that can reach
Somesimplecalculationsshowthe relativeimportance the polar regions,and the large troposphericcontribution
of these reactions. First, we assumethat when a gas
of Na+, NO3-, andSO4•- thiswill probably
be very
molecule
(NO2, 01NO3or N205) sticksto a particle,it
difficult, if not impossible.
If the impactof reactions
(1)-(3) is similarto that of
is in contact with NaCI and can react. We thereforecan
calculate
the lossfrequency
dueto collisions
withNaCl(s)
the speculativechemistryproposedin Michelange!iet
byJNaC!= 1/43vANawherev is themeanspeed
of
(1989)whichreproduces
theHCI andNO=measurements,
is the numberdensityof the aerosols
(26 cm-3 at 28
kin). Assuming
stickingcoefficients
for NO• andN•O5
are• = 10-8 and 5 x 10-•, respectively,
we obtainthe
corresponding
lossfrequencies
for heterogeneous
reaction
becauseof the complexnature of the variousinteractions
betweenthe NO• and C10• speciesand O• in the strato-
theimpacting
molecule
v = ¾/SkT/•rm,A is themean an ozonedecreaseof a few percentcan be expected.This
with the observations
surface
area(6.4x 10-s cm•, assuming
thatthegeomet- is not considereda disagreement
rkal and opticalcrosssectionsare the same),and N,, sincethere is easilya few percentuncertaintyin all the
withNaC1,JNa(;1
-- 1.4x 10-tø and4.5 x 10-5 s-j, re-
assumed
parameters
for the modelcalculation.However,
sphere,the ultimatefate of O3 can onlybe determined
by detailedphotochemicalmodeling.
Heterogeneous
reactionswith NaC! seemto havethe
spectively.
The lossfrequency
dueto CINO.•isnotknown,
potential
of
having
a majorimpactonstratospheric
chembutisprobably
closeto that for N• Os. Wecompare
these
withthe lossfrequencies
for reactions(4)-(6), whichare istryafterthe eruptionof E1Chichon.A detailednumericalstudyawaitsfurtherlaboratoryresults.The ultimate
J•=3.82 x 10-5 s-r Js- 1.32x 10-5s -x andJ•=
questionis whetheror not this mechanism
might have
1.08x 10-4 s-•. Thelossfrequencies
ofreactions
(4)-(6) beenevenmoresignificantduring earlierlargervolcanic
arehigherthanforthereactions
between
NO2andNaC1,
but are not so different for N•O• and possiblyCINOa.
eruptions.
Theheterogeneous
reactions
represent
a permanent
sink
Acknowledgements.
We thankM.-T. Leufor valuable
forNOxandcanproducea C1increase,
followed
by an discussions
and B. Fin!ayson-Pittsfor communication
of
HCI increase.
laboratoryresultspriorto publication.Wealsoaregrate-
If weassmethatoneNO2(g)islostforeach
NaNOs(s) ful to the anonymous
reviewersfor pointingout the deformed,
theNO2concentration
at 28kmwill decrease
by
ficienciesof our initial presentation.This work wassup-
to California
Instituteof Technol50%(1.3x 10• molecules
cm-a;McFar!and
etal.,1986; portedbyNAGW-413
Roscoe
etal.,1986)after14hours
byreaction
(3).At the ogy.Oneof us(DVM) acknowledges
support
of a NASA
Contribution
number4898.
fromtheDisame
altitude,
reaction
(2) couldleadto a 40%increase NRCFellowship.
and PlanetarySciences,
California
inHC1(1.72x 10s molecules
cm-3;B. W. Gandrud
and visionof Geological
Pasadena,California91125.
A.L. Lazrus,privatecommunication,
1983)in 3.4 days. Institute of Technology,
If the losstimescalefor HCI due to vertical transportis
~1month,
theelevated
HCI abundance
canbesustained
References
with• for(2)assmallas10-$. Therefore,
theproposed
G. Fiocco,andG. P. Gobbi,
heterogeneous
chemistry
canbe relatively
inefficient
and Adriani,A., F. Congeduti,
lidarobservations
ofthestratospheric
aerosol
stillreproduce
observations.
Reaction(1) is, however, One-year
at
Frascati,
March,
!982--March,
1983,
Geophys.
Re,.
tooslowto haveanyimpact.Sincethesereactions
occur
Lent.,
!0,
1005-1008,
1983.
rapidly,
thereshould
benoproblem
caused
byNaCIsedfor stratoimentation,
sinceweknowhaliteispresent
œor
a month. Arnold,F., Th. Bfthrke,andS. Oiu,Evidence
sph.eric
ozone-depleting
heterogeneous
chemistry
on
An importantuncertainty
is the location
of the reacvolcanicaerosolsfrom E1 Chichon,Nature, 2448,49-50,
tion.Oncethesurface
layerof NaCIhasbeenoxidized
1990.
toNaNOs,
it isnotclearwhether
ornotthereaction
will
continue
in the interior and releaseHCI(g). Livingston Chandra,S., The aerosoleffec• on ozonemeasurements
and
Finlayson-Pitts
(1991)
suggest
thatreaction
(3)oc-
following
the eruptionof E! Chichon,
EO$: Transactions,American
Geophysical
Union,68, 37,0,1987.
cursonthe surfaceof the halitecrystalsin theirexperiJ. J.,E. O.Dutton,K. L. Coulson,
T. E. DeFoor,
mentalsystem.Furtherlaboratoryinvestigations
are re- DeLuisi,
and B. G. Mendonca,
On someradiativefeaturesof
q•iredto understandandquantifythisprocess
morepre-
676
Michelangeliet aL: NaCI Aerosols
the E1 Chichon volcanic stratosphericdust cloud and
a cloud of unknown origin observedat Mauna Los, J.
Geophys.Res., 88, 6769-6772, 1983.
Finlayson-Pitts, B. J., Reaction of NO2 with NaCI and
atmospheric implications of NOCI formation, Nature,
306, 676-677, 1983.
Finlayson-Pitts, B. J., M. J. Ezell, and J. N. Pitts, Jr.,
Formation of chemicallyactive chlorinecompoundsby
reactions of atmospheric NaC1 particles with gaseous
N20$ and CIONO•, Nature, 337, 241-244, 1989.
Hayashida, S., et al., Water ResearchInstitute, Middle
Atmospheric Program, note 3, 1984.
Heath, D. F. and B. M. Schlesinger,Global responseof
stratospheric ozone to natural perturbations on climatological time scalesby variationsof U.V solarflux and
the eruption of E1 Chichon,EOS: Transactions,
American GeophysicalUnion,65, 836-837, 1984.
Hoffmann, D. J. and J. M. Rosen, Stratosphericsu!furic
acid fraction
and mass estimate for the 1982 .volcanic
eruption of El Chichon,Geophys.
Res.Lett., 10, 313-316,
1983.
che•nicalperturbations,
J'. Geophys.
Res.,94, 18,42918,443, 1989.
Michelangeli,
D. V., M. Allen,Y. L. Yung,R. L. Shia,and
D..Crisp,
Enhancement
oftheatmospheric
radiation
by
an aerosollayer. Submittedto J. Geophys.
Res.,1990.
Molina, M. J., T. L. Tso, L. T. Molina, and F. C. Y.
Wang, Antarctic stratosphericchemistryof chlorinenitrate, hydrogenchlorideand ice: Releaseof activechlorine, Science,238, 1253-1257, 1987.
Pinto,J.P., R. P. Turco,andO. B. Toon,Self-limiting
physicaland chemical effectsin volcaniceruption
clouds,•r. Geophys.Res., 94, 11,165-11,174, 1989.
Pollack, J. B., O. B. Toon, E. F. Danielson, D. J. Holmann, and J. M. Rosen, The El Chichonvolcaniccloud:
An introduction,Geophys.
Res.Lett.,10, 989-992,1983.
Roscoe,H. K., B. J. Kerridge,L. J. Gray,R. J. Wells,
and J. A. Pyle, Simultaneousmeasurementsof stratosphericNO and 1NO2and their comparisonwith model
predictions,J. Geophy$.Res., 91, 5405-5419, 1986.
Rossi, M. J., R. Malhotra, and D. M. Golden, Heterogen.eouschemical reaction of chlorine nitrate and water
Hoffmann,D. J. and J. M. Rosen,On the temporalvarionsulfuric-acidsurfacesat roomtemperature,Geophy&
ation of stratosphericaerosolsize and massduring the
Res. Lett., 14, 127-130, 1987.
first 18 monthsfollowingthe 1982eruptionsof El ChiSolomon, S., R. R. Garcia, F. S. Rowland, and D. J.
chon,g. Geophy$.Res., 89, 4883-4890, 1984.
Wuebbles,On the depletionof Antarc.ticozone,Nature,
Hoffmann, D. J., and S. Solomon,Ozone destruction
321, 755-758, 1986.
through heterogeneous
chemistryfollowingthe erup- Spinhirne, J. D. and M.D. King, Latitude 'variationof
tion of E! Chichon,Y.Geophys.
Res.,94, 5029-5041,1989.
spectral optical thickness•nd columnar size distribuJ•ger, H. and W. Carnuth, The decay of the El Chichon
tion of the El Chichon stratospheric aerosollayer,J.
stratosphericperturbation, observedby Lidar at NorthGeophys.Res.,90, 10,607-10,619, 1985.
ern midlatitudes,Geophys.
Res.Lett.,14, 606-609, 1987. Thomas, G. E., B. M. Jakosky, R. A. West, and R. W.
Krueger, A. J., Sighting of E1 Chichon sulfur dioxide
Sanders, Satellite limb-scanning thermal infrared o•
cloudswith the Nimbus 7 total ozonemappingspecservationsof the El Chichonstratosphericaerosol:First
trometer, Science,220, 1377, 1083.
results,Geophy$.Res.Lett., 10, 997-1000, 1983.
Labitzke, K., B. Naujokat, and M.P. McCormick, TemTolbert, M. A., M. J. Rossi, and D. M. Golden, Iteteroperature effectson the stratosphereof the April 4, 1982
geneousinteractions of chlorine nitrate, hydrogenchloeruptionof El Chichon,Mexico,Geophys.
Res.Lett.,10,
ride, and nitric acid with sulfuric acid surfacesat strato24-26, 1983.
spherictemperatures,Geophys.Res. Lett., 15, 847-850,
Livingston, F. E., and B. J. Finlayson-Pitts, The reaction
1988.
of gaseousN205 with solid NaC! at 298 K: Estimated
Woods,D.C., R. L. Chuan, and W. I. Rose,Halite partilower limit to the reaction probability and its potenclesinjected into the •tratosphereby the 1982 El Chitial role in troposphericand stratosphericchemistry,
chon eruption, Science,230, !70--172, 1985.
Geophys.Res.Lett., 18, 17-20, 1991.
Worsnop,D., M. gahniser, C..Kolb, L. Watson, J.
Mankin, W. G. and M. T. Coffey,Increasedstratospheric
Van Doren, j. Jayne, and P. Davidovits, Mass achydrogenchloridein the El Chichoncloud,Science,
226,
commodation coefficient measurements for HN0a,
170-172, 1984.
Mankin, W.G., and M.T. Coffey,The impact of E1 Chichon on the chemistry of the stratosphere,paper presented at the Norman D. Watkins Symposiumon the
Environmental Impact of volcanism, Graduate School
of Oceanography,Narragansett,RI, March, 1986.
McElroy, M. B., R. J. Salawitch,S.C. Wofsy,and J. A.
Logan, Antarctic ozone: Reductiondue to synergistic
interactionsof chlorineand bromine,Nature,321, 759762, 1986.
H.C!,andN20$onwater/iceandaqueous
sulfuric
acid
droplet surfaces, Abstractsof the Polar OzoneWorkshop, Snowmass,Colorado, 9-13 May 1988.
Diane V. Michelangeli,SpaceScienceDivision, NASA
Ames ResearchCenter, Moffett Field, California94035
Yuk L. Yung, Divisionof Geologicaland PlanetarySciences,CaliforniaInstituteof Technology,
Pasadena,
California
91125.
Mark Allen, Earth and Space SciencesDivision,Jet
McFarland, M., B. A. Ridley, M. H. Proffitt, D. L. AlbritPropulsionLaboratory,CaliforniaInstitute of Techn&
ton, T. L. Thompson,W. J. Hattop, R. H. Wink!er,and
ogy,
4800 Oak GroveDrive, Pasadena,California 9!100
A. L. Schmeltekopf,Simultaneousin situ measurements
of NO2, NO, andOsbetween
20 and31kin,J. Geophy$.
Res., 91, 5421-5437, 1986.
(Received:August24, 1990;
Michelangeli,
D. V., M. A!len,andY. L. Yung,El Chichon.
revised: November 28, 1990;
volcanicaerosols: Impact of radiative, thermal and
accepted:
February12, 1991.)