NASA Technical Reports Server (NTRS) 19950015360: Conference on Deep Earth and Planetary Volatiles

58 Pages·3.7 MB·English

by  NASA Technical Reports Server (NTRS)

NASA-CR-196944
(NASA-CR-I96944) CONFERENCE ON N95-21777
DEEP EARTH ANO PLANETARY VOLATILES --THRU--
Abstracts Only (Lunar and N95-21824
Planetary Inst.) 59 p Unclas
G3/46 0038210
NASA-CR-196944
19950015360
September 21-23, 1994
Pasadena, California
q
DISPLAY 95N21777/2
95N21777"# ISSUE 6 CATEGORY 46
RPT#: NASA-CR-196944 NAS 1.26:196944 LPIZ$ONT_IB-845 CNT#: NASW-4574
94/00/00 59 PAGES UNCLASSIFIED DOCUMENT
UTTL: Conference on Deep Earth and Planetary Volatiles TLSP: Abstracts Only
CORP: Lunar and Planetary Inst., Houston, TX.
SAP: Avail: CASI HC A04/MF A01
CIO: UNITED STATES Conference held in Pasadena, CA, 21-23 Sep. 1994
MAJS: /*CONFERENCES/*DEGASSING/*DIAMONDS/*HALOGENS/*ISOTOPES/*MAGNESIUM/*RARE
GASES/*RHEOLOGY/*SILICATES/*STABILITY/*TERRESTRIAL PLANETS/*WATER
MINS: / HIGH PRESSURE/ RECYCLING/ REGIONS/ SUBDUCTION (GEOLOGY)
ANN: The following topics are covered in the presented papers: (i)rare gases
systematics and mantle structure; (2) volatiles in the earth; (3) impact
degassing of water and noble gases from silicates; (4) D/H ratios and H20
contents of mantle-derived amphibole megacrysts; (5) thermochemistry of
dense hydrous magnesium silicates; (6)modeling of the effect of water on
mantle rheology; (7)noble gas isotopes and halogens in volatile-rich
inclusions in diamonds; (8)origin and loss of the volatiles of the
terrestrial planets; (9) structure and the stability of hydrous minerals
at high pressure; (i0)recycling of volatiles at subduction zones and
various other topics. For individual titles, see N95-21778 through
N95-21824.
ENTER:
4B° A ^ =-.PC LINE 24 COL 9

1.2)1Contribution No. 845 1
N95-21778
Rare gases systematics and mantle structure. C.J.All_greandT.Staudacher
Laboratoirede G6ochimie et Cosmochimie, Institut de Physique du Globe de Paris, _ .. I
Universit6deParisVIetVII,4,PlaceJussieu75252Paris,France.
About ten years ago we published a general review of rare gases systematic in the
atmosphere- mantle- crustsystem.Our modelat thattimewas selfconsistent, based onthe
available data and explained as well helium, as argon and xenon isotopic signatures. It is
today certainly not the only possible model, since new data have been obtained and
experimentalprogresswas made.
_ _l'/e_nl_The followingpoints._r€- e._ I_h_s,_:0 :
One of themostimportantones iscertainly thefirstsetof experimentaldata onthe
solubility of noble gases in metal phases at intermediate pressures (Matsuda et al., 1993),
sincethe corewas certainlynotformedat ultrahighpressures, as emphasizedbyAhrens and
confirmed bytraceelementssystematicsWiinke.Theexperimentaldata clearly show thatthe
core cannot beamajorreservoirforterrestrialraregases.
m The second point isa more elaborate reconsideration of the 40K-40Ar budget of
the Earth. Thisshowsthat40Ar containedin continentalcrust + uppermantle + atmosphere
isat maximum halfof the 40Ar inventoryof the wholeearth.This impliesthe existence of a
twolayeredmantle(All_greetal., 1994).
D The third point is the discovery by the Australian noble gases group of the
existance of high 20Ne/22Neandlow21Ne/22Neisotopicratiosin Loihiseamount samples
(ex.Honda et al. 1991,1993).This resultswhichare different tothe MORBratios (Sardaet
al., 1986)confirm theidea of a twolayeredmodel, butsuggestthe existanceof aprimordial
solar type Ne reservoir. Several possibilities about the origin of this 20Ne excess in the
mantle willbe discussed.
The high 40Ar/36Ar, 129Xe/130Xe and 134Xe/130Xe, 136Xe/130Xe are
confirmed by new data. The corresponding ratios for the lower mantle will be discussed.
40Ar/36Ar ratios up to 6000can beaccepted andwill not modify the general model of the
mantle. They confirm theatmospherechronology,about 85% of the atmospherewas formed
inthe first50Myand 15%lateron.
We will alsodiscuss the results obtainedon xenolithsor phenocrysts in different
lava types and try to constrain the information they carry, using the simple idea that
xenocrystsinhotspotsdonotderive fromthe deepsourceasthe magmawhichcarrythem.
- Finallywe have quantitatively explosed the steady-state upper mantle model and
constrainsthe differentsvaluesof themantlereservoirs.
All_gre C.J., K.O'Nions,andA. Hofmann, 1994,inpress.
MatsudaJ., SudoM., OzimaM., ItoK.,OhtakaO.andItoE.,Noblegas partitioningbetween
Metal and Silicateunderhighpressures,Science259,788-790, 1993
Honda M.;McDougallI., PattersonD.B.,DoulgerisA. andClague D.A.,Possiblesolar noble
gas componentinHawaiianbasalts,Nature349, 149-151,1991
Honda M., McDougall I., Patterson D.B., Doulgeris A. and Clague D.A., Noble gases in
submarine pillow basalt glasses fromLoihi andKilauea,Hawaii: a solar component in
the Earth.Geochim. Cosmochim.Acta57,859-874, 1993.
Sardaet al. Neonisotopesinsubmarinebasalts,Earth Planet.Sci.Lett.,91,73-88, 1988.
,,.o o..h vo,o,,,o. N95-21779
V"OLATILESINTHEEARTH;ALLSHALLOWAND ALLRECYCLED.DonL.
Anderson,SeismologicalLaboratory,CaliforniaInstituteofTechnology,Pasadena, CA
7, \ 91125
\
Acasecan bemade thataccretionof theEarthwas ahigh-temperatureprocess and
thattheprimordialEarthwas dry. A radialzone-refiningprocessduringaccretionmay
haveexcludedlow-meltingpointandvolatilematerial,including.large-ionlithophile
elementstowardthesurface,leavingarefractoryandzonedinter,or.
Water,sedimentsandalteredhydrousoceaniccrustareintroducedback into the
interiorbysubduction,aprocess thatmay bemoreefficient todaythan inthe past.
Seismictomographystronglysuggeststhata largepartof theuppermantle isabove the
solidus,andthisimplieswetmelting. ThemantlebeneathArcheancratonshas very fast
seismicvelocitiesandappearstobestrongto 150kmor greater. Thisisconsistentwith
verydry mantle. Itisarguedthatrecyclingof substantialquantifiesof water occursinthe
shallowmantlebutonlyminoramountsrecycletodepthsgreaterthan 200km. Recycling
alsooxidizes thatmantle;ocean island("hotspot")basaltsareintermediate inoxidation
state toisland-arcandmidoceanridgebasalts (MORB). T hissuggestsadeep
uncontaminatedreservoirfor MORB.
Platetectonicsonadry Earthisdiscussedinordertofocus attentionon
inconsistenciesincurrentgeochemicalmodels ofterrestrialevolutionandrecycling.
LPIContributioNno.845 3
N95-21780
IMPACT DEGASSING OF WATER AND NOBLE GASES FROM SILICATES. S.Azumal, H. ; I
HJyagonI, Y. IJjima2andY. Syono3,1DepartmentofEarth and Planetary Physics, University ofTokyo, Tokyo
113,JAPAN, 2Department of Earth and Planetary Science, NagoyaUniversity, Nagoya464-01, JAPAN,
SInstitute forMaterialsResearch, Toho]mUniversity, Sendal 980, JAPAN.
Previousshock experiments by Ahrens and his colleagues show that degasslng of H20 and CO2occurs
at 8--65GPafromhydrous mineralssuch as serpentine [1,2].In earlysolar system, the impact degasslng would
haveplayed an important part in the formation of primary-atmospheres of the terrestrial planets. However,
degassingconditionsofnoblegasesarenotwell-knownbecausetherearefew experimentsforthem [3,4]. Wecon-
ducted some shockrecoveryexperiments toinvestigate thedegassing conditionandtounderstand thedegassing
mechanisms of waterandnoble gases.
Weused natural richterites (Rl_, amphibolites (Am), serpentines (Sep) and orthoclases (Or) as target
samples. These, except Sep,containradiogenicnoblegasessuch as4°Ar. Thesamples wereput instainless steel
containers, and wereshot byarailgun atISASorsingle-stage powdergunsat NagoyaorTohokuUniv., Japan.
Weusedtwokindsofcontainers : 'open'type containers having aventilating pathforrdeased volatilesformost
ofsamples and 'dosed' type onesforsomesamples forcomparison. OnRi and $ep, wemade shockexperiments
for pre-heated (at 400-500°C) and unheated targets, and for powdered and uncrushedsamples. Water and
noble gases were analyzed both for the recoveredshocked samples and the unshocked original samples, and
the fractions of the degassed volatiles werecalculated by comparing them. Water content in the samplewas
analyzed by thermo-gravimetry. Noble gaseswere extracted by heating the samples under high vacuum and
analyzed with asector-type mass spectrometer.
The resultsforimpact-induced waterlossareshowninFig.1. Waterlossforpre-heated(at 400")Sepwas
about two timeslarger than that forunheated 5ep atpressuresof30-40GPa, but no differencewasobserved for
pre-hented and unheated Ri up to43GPa. X-raydiffractionanalyses showthat serpentine wasnot decomposed
into olivine + pyroxeneby shock, but decomposedinto the amorphous structure. The mechanism ofwater loss
from 8ep is most likelydiffusionfromthe amorphous part, but thatfrom Ri may bedifferent. Waterloss from
the powdered Sep is high (30-60%) at 40GPa, but that from the nncrushed 8epis relatively low (20%) at the
same pressure, which is not consistent with the previous work[1,2].
Degassing ofradiogenicnoblegases, such as4°An,from Orand Am was observed at shock pressuresof
15-68 GPa. Observed degassing oflightergases (He,Ne) from Orseemes consistent wlth the idea ofdiffusional
loss. However,loss ofHewasnot observedforAmeven at pressureswhereloss ofradiogenic4°Arwasobserved.
The degassing mech_sm ofradiogenic i°Ar seems differentfromthat of trapped noble gases.
100 ..................................
Fig.1 Shock-induced waterloss (%) • " o Richterite(heated)
from richterltes andserpentines at various _'_80 . • Richterile
shockpressures(GPa). The data of Antigorite_." • o Serpentine
m 60 .o O v Serpentine(closed)
and Porous Lizardite arefrom[1]& [2]. ,.J o 25%PorousSerpentine
Reference: [1]Lange M.A. et al. _ 40 * .s " , s Serpentine(heated)
• Antigorite
(J1.A98.5e)tGa.(l7.(.1A9.90_)/E9,.P1.7$1.L5-.172968.,2[425]-T2y6b0u.r[c3z]y _m20 * s _o o • Anfigorite(closed)
d8 • 20%PorousLizardite
GazisC.andAhrensT.J.(1991E).P.S.L. 0 ......._."..."...,.v...'....,,,,,_...
104,337-349. [4]AhrensT.J. et al.(1992) 0 10 20 30 40 50 60 70
LPSC XXIII, 3-4 (abstract). [5]Aka_J. Shock Pressure(GPa)
and Sekine T.(1994)Proe. NIPR Symp. Antaret. Meteorites, 7,101-109.
. oo, N95-21781
/',¢,:, J ZJ:.",;
2s./y
D/H RATIOS AND H20 CONTENTS OF MANTLE-DERIVED AMPHIBOLE
/\ MEGACRYSTS FROM DISH HILL, CALIFORNIA. David R. Bell andBranCc.hRTd..N,H.Woe.r,ing,
Geophysical Laboratory,Carnegie Institution of Washington, 5251 Broad
WashingtonDC20015.
-- D/H ratiosare,inprinciple,usefulin characterizingreservoirsof mantlehydrogenand as
tracersof volatile transferprocessesin Earth'sinterior. In practice,however, interpretationof
isotopic measurements on mantle derived H is comp.licated by surface processes such as
contamination and degassing which may alter the primary D/H ratio. Although there are
indications [1-3] thatwater associated with subduction zones and certainchemically enriched
hasaltsisenrichedinDrelativeto "typical"uppermantlewater,theextentofisotopicheterogeneity
ofmantleH remainsuncertain.Kaersutiticamphibolemegacrystsinalkalinebasaltsareoneof the
most widespread sources of mantle water and are thereforepotentially useful for large-scale
regionalstudiesofD/Hvariation. However,D/I-Iratiosof theseamphibolesvarywidely (from+8
to-113%o[4]),eveninsamplesfromthesamelocality,sothatthispotentialhas yettoberealized.
In orderto investigate the origin of thisvariability, and to explore the possibility that
primarymantleD/H ratiosmaybededucedfromtheseamphiboles,weanalyzedtheD/Hratiosand
chemical compositions of a suite of 17kaersutiticamphibolesfromDish Hill, California. This
workcontrasts with previousstudiesin whichsamplingis widespread,but representativesfrom
anygiven localityarefew. Sampleswerecollectedfromarestrictedarea onthesouthern flankof
thevolcaniccenterandareassociatedwiththebasalvolcanicbreccia[5]. Fourteenof the samples
werelarge single crystalsorcrystalfragments(megacrysts,0.4 to30 grams),believed to derive
frompegmatiticveinscrystallizedfrommeltsinthemantle. Two werecoarse-grainedintergrowths
ofamphibolewitholivineandspinel,andonewasathin(2ram)selvageonaperidotitexenolith.
ThesamplesrangeinMg#(=100Mg/(Mg+TotalFe))from55to85,withthe polymineralic
intergrowthshaving highest Mg#'s. In the megacryst samples, the abundances of all elements
analyzed by electron microprobe vary systematically with Mg#, giving reason to expect
concomitant behavior in the primarywater content of these samples. The polymineralic and
selvage samples aredisplacedfromthese trendsto varyingdegrees. Ourpreliminarymanometry
data (notallsamplesyetanalyzedinduplicate)confLrmageneralsystematicbehaviorforH: 12of
the 14megacrysts contain from0.93 to 1.03wt.% H20 and have SDSMOWof -46.8 + 7.1 9'=,
(2o). Two of the megacrystsare displacedfromthistrend tosignificantlylowerH20 contents of
0.59 and 0.04 wt. % respectively. 8D of the formersample is -9 _'_,while the latterprovided
insufficientHforisotopicanalysis. Thepolymineralicandselvage sampleshaveH20 contents of
1.08to 1.15wt. %, with8Dfrom-37 to -45_'_.
The uniformityoftheseresultssuggestthatprimaryHcontentandisotopiccompositionis
preservedin most of these samples. Low Hcontentsof the two anomalousmegacrysts and the
high D/H ratio are probablythe productof dehydrogenation-oxidation[6,7]. Itislikely thatthis
process occurredduring transport and eruption, a_there are no other indications of unusual
chemistryinthese samples. Thehigh degree ofuniformityobserved,relativetosimilarprevious
studies, may be due to local volcanology [6]. Within the limits of current uncertainty of the
relevantD/H fractionations,themean "undisturbed"8Dof -46 9'_isconsistent with equilibrium
withwater,occurringas OHgroupsinamelt,of 5Dnear-70 %,,,typicalof depleteduppermantle.
TheMORB-likeaffinityofthesemegacrystsisindicatedbytheirSrisotopecompositions[8].
We conclude that H-isotope information, useful for global systematic studies, can be
extracted frommantle-derivedkaersutiticamphiboles,providingtheappropriate consideration is
giventopetrologyandvolcanologyof thesamples.
References. [1] Poreda, R. 1985EPSL73, 244-254. [2] Poreda,R. et al. 1986EPSL
78, 1-17. [3] Dobson, P.F. and O'Neil, J.R. 1987 EPSL 82, 75-86. [4] Boettcher, A.L. and
O'Neil, J.R. 1980 Am.J.Sci 280A 594-621 [5] Wilshire, H.G and Nieison-Pike, J.E. 1986in
GSAGuidebook, SouthernCaliforniaFieldTrips 15(P.L.Ehlig, ed), 9-11. [6] AokiK. 1963J.
Petrol.4, 198-210. [7] Dyar,M.D. et al. 1992Geology20,565-568. [8] Basu,A.R. and Murthy,
V.R. 1977Geology 5,365-368.
N95-21782 5
)HERMOCHEMISTRY OF DENSE HYDROUS MAGNESIUM SILICATES. Kunal
Bose_,Pamela Bumle2fl and Alexandra Navrotsky n._Department of Geological and -:
[ J
Geophysical Sciences and Center for High Pressure Research, Princeton University, Princeton,
New Jersey 08544, USA, 2Cooperative Institute for Research in Environmental Sciences,
University of Colorado at Boulder, Campus Box 216, Boulder, Colorado 80309, USA.
-- Recent experimental investigations under mantle conditions have identified a suite of
dense hydrous magnesium silicate (DHMS) phases that could be conduits to transport water to
at least the 660 km discontinuity via mature, relatively cold, subducting slabs. Water released
from successive dehydration of these phases during subduction could be responsible for deep
focus earthquakes,mantle metasomatism and a host of other physico-chemicaplrocesses
central to our understanding of the earth's deep interior.
In order to construct a thermodynamic data base that can delineate and predict the
stability ranges of DHMS phases, reliable thermochemical and thermophysical data are
required. One of the major obstacles in calorimetric studies of phases synthesized under high
pressure conditions has been limitation due to the small (< 5 mg) sample mass. Our
refinement of calorimeter techniques now allow precise determination of enthalpies of
solution of < 5 mg samples of hydrous magnesium silicates. For example, high temperature
solution calorimetry of natural talc (Mgo99Feo.ozSi40_(oOH) 2 ), periclase (MgO) andquartz
(SiO2)yield enthalpies of drop solution at 1044 K to be 592.2 (2.2), 52.01 (0.12) and 45.76
(0.4) kJ/mol respectively. The corresponding enthalpy of formation from oxides at 298 K for
talc is -5908.2 kJ/mol agreeing within 0.1% to literature values (Hemingway, 1991_:AH29s
(talc) = -5900 kJ/mol).
Approximately 30 mg each of samples Phase A, Phase B, Superhydrous B and
Chondrodite have been synthesized for calorimetric and NMR characterization (Burnley et al.,
in prep.). We present enthalpies of formation data (derived from enthalpies of drop solution in
molten lead borate) for some of these phases.
Phase equilibria and thermodynamic analysis of talc and antigorite (serpentine) (Bose
and Ganguly, 19932)indicate stability limits near 840 °C at 3 GPa and 750 °C at 7 GPa,
very near the onset of stability of DHMS phases (Gasparik, 19933). The calorimetric data in
conjunction thermophysical properties of these minerals will permit mapping out the sequence
of equilibrium reactions responsible for the introduction, release upon dehydration and
subsequent recycling of water in the mantle, proximal to a subducting slab.
References: [1] Hemingway, B. R. (1991) Am. Mitt., 76, 1589-1596. [2] Bose, K. and
Ganguly, J. (1993) Geol. Soc. Amer. Abstr. 25, 213-214. [3] Gasparik, T. (1993) JGR, 98,
4287-4299.
N95-21783
MODELE_'GTHE EFFECTOFWATER ONMANTLE RttEOLOGY. Ch.Bounamaand
9,,\ S.Franck,GeneralGeophysicsGroup,PotsdamUniversity.P.O. Box601632,D-14416 Potsdam,
Germany.
Tostudythe thermalhistoryofthe Earthwe use aparameterizedmodelof mantleconvection[1].
Thismodelincludesa mathematicaldescriptionof de-andregassingprocesses of water fromthe
Earth's mantle.The ratesof thisprocessesare consideredto bedirectlyproportionalto the
seafloorspreadingrate. Thekinematicviscosityofthe mantledependsonthe
temperature/pressureas wellas onthe volatilecontent.Dissolvedvolatilessuchas water weaken
the mineralsbyreducingtheir activationenergyforsolidstate creep.Karato andToriumi [2]
showedapower lawdependencebetween creeprate andwater fugacityderived from
experimentalresults.Therefore,we use suchflowparametersof diffusioncreep inolivineunder
wet anddry conditionsto calculatethe mantleviscosityas afunctionofthe water content.
Because the creeprate isproportionaltothe concentrationofwater-related pointdefects we
assumethat thewater fugacityisproportionalto thewater weightfraction.Anequationfor the
steady-state strainrate underwetconditionsisestablished.To assessthe unknown constantK in
thisequation,we use flowlawparametersgivenbyKarato andWu [3] aswellas the resultsof
McGovern andSchubert [1].
Theevolutionmodelisrun for4.6b.y.Timeseriesofaveragemantletemperature,volatileloss,
mantleviscosity,mantleheat flow,Rayleighnumber,andUreyratio arecalculated.The mantle
water isoutgassedrapidlyinlessthan200m.y.The stabilityofthe resultsagainstvariationsofthe
constantK istested. WithincreasingKtheinfluenceof water isstronger,the finalaveragemantle
temperature islower.The degassingprocessismorerapid.Alreadyanincreaseof K byone
magnitudeshowsanoutgassingprocessof 100m.y.Thefinalresultsof the calculatedparameters
are inthe generallyaccepted rangefor allnumericalsimulations.
References: [1] McGovernP. andSchubertG.(1989) EPSL, 96,27-37. [2] Karato S. and
ToriumiM. (1989)Rheolog_of S_olidsandof theEarth, OxfordUniv.,New York. [3]Karato S.
andWuP. (1993)_ 260,771-778.