Sm-Nd and Rb-Sr studies of lherzolitic shergottite Yamato ... › download › pdf ›...

Available online at www.sciencedirect.com

Polar Science 4 (2011) 515e529http://ees.elsevier.com/polar/

SmeNd and RbeSr studies of lherzolitic shergottite Yamato 984028

C.-Y. Shih a,*, L.E. Nyquist b, Y. Reese c, K. Misawa d

aMail Code JE-23, ESCG/Jacobs Sverdrup, P.O. Box 58477, Houston, TX 77258-8477, USAbMail Code KR, NASA Johnson Space Center, Houston, TX 77058-3696, USA

cMail Code JE-23, ESCG/MEI Technologies, Houston, TX 77058, USAdNational Institute of Polar Research, Graduate University for Advanced Studies (SOKENDAI), 10e3,Midori-cho, Tachikawa-shi,

Tokyo 190-8518, Japan

Received 14 October 2009; revised 8 April 2010; accepted 14 May 2010

Available online 15 June 2010

Abstract

The distribution pattern of the trace elements Rb, Sr, Nd and Sm for Yamato 984028 (Y984028) is consistent with its classi-fication as a lherzolitic shergottite. The SmeNdmineral isochron of this lherzolitic shergottite defines its age to be 170� 10 Ma foran initial 3Nd ¼ þ11.6 � 0.2. The corresponding RbeSr mineral isochron yields an identical age of 170 � 9 Ma and an initial87Sr/86Sr ¼ 0.710389 � 0.000029. The concordant SmeNd and RbeSr isochron ages suggest that Y984028 crystallized170� 7Ma ago contemporaneously with five other lherzolitic shergottites and ten enriched basaltic and olivine-phyric shergottites.The age, Sr- and Nd- isotopic signatures further suggest that Y984028 and Y-793605, and also probably Y000097 could come froma single magmatic body. Using a two-stage evolution model, the time-averaged 87Rb/86Sr-ratio for the mantle source of the parentmagma of Y984028 is w0.182, within the range of 0.178e0.182 that has been reported for other lherzolitic shergottites. Thecorresponding time-averaged 147Sm/144Nd-ratio for the source mantle of its parent magma is super-chondritic atw0.217, implyingits source was a depleted mafic part of the Martian mantle similar to that of diabasic shergottite Northwest Africa (NWA) 1460. Rb,Sr, Sm and Nd distributions in Y984028 are likely produced by pyroxene and olivine accumulation, probably from a NWA 1460-like parental melt, in an intrusive magma body.� 2010 Elsevier B.V. and NIPR. All rights reserved.

Keywords: Mars; Lherzolitic shergottite; SmeNd; RbeSr; Crystallization age

1. Introduction

Y984028 is a Martian meteorite found in theYamato Mountains of Antarctica in 1998 (Kojima,2008; Misawa et al., 2009). It is classified as a lherzo-litic shergottite due to its petrographical resemblance

* Corresponding author.

E-mail addresses: [email protected] (C.-Y. Shih),

[email protected] (L.E. Nyquist), young.reese-1@nasa.

gov (Y. Reese), [email protected] (K. Misawa).

1873-9652/$ - see front matter � 2010 Elsevier B.V. and NIPR. All rightsdoi:10.1016/j.polar.2010.05.004

to several other lherzolitic shergottites, i.e., ALHA77005, LEW 88516, Y-793605, NWA 1950 andY000027/47/97 (e.g., McSween et al., 1979; Harveyet al., 1993; Ikeda, 1997; Gillet et al., 2005;Mikouchi and Kurihara, 2008; Shirai and Ebihara,2009). These meteorites have young crystallizationages of 152e185 Ma similar to the enriched basaltic orolivine-phyric shergottites with ages of 157e203 Ma(Shih et al., 1982, 2003, 2009a; Morikawa et al., 2001;Nyquist et al., 2001; Borg et al., 2002; Misawa et al.,

reserved.

mailto:[email protected]:[email protected]:[email protected]://dx.doi.org/10.1016/j.polar.2010.05.004http://dx.doi.org/10.1016/j.polar.2010.05.004http://dx.doi.org/10.1016/j.polar.2010.05.004http://ees.elsevier.com/polarmailto:[email protected]:[email protected]

516 C.-Y. Shih et al. / Polar Science 4 (2011) 515e529

2008) but have very different ejection ages (w4 Ma vs.w2.5 Ma) (e.g., Eugster et al., 1997; Nyquist et al.,2001; Christen et al., 2005; Nagao and Park, 2008;Nagao et al., 2008), thus they came from differentMartian target crater areas via separate impact events.Lherzolitic shergottites have high mg-values w0.70and represent the most mafic olivine-pyroxene cumu-lates from Martian magmas. Their parental magmaswere melts probably derived from the primitiveMartian mantle (e.g., Dreibus et al., 1992).

A recent mineralogy-petrography study of olivine-phyric shergottite RBT 04262 suggested a closecorrelation with lherzolitic shergottite LEW 88516(Mikouchi et al., 2008). Trace elemental evidence andRbeSr and SmeNd isotopic results of basaltic sher-gottite NWA 1460/480 (Barrat et al., 2002; Nyquistet al., 2009) also reveal that it may have beenderived from a similar source region as lherzoliticshergottites despite their crystallization age differ-ences. The study of the new lherzolitic shergottiteY984028 may provide more information about therelationship among these shergottites.

The objectives of this study are (1) to characterizeY984028 by its trace element (Rb, Sr, Sm and Nd)chemistry, (2) to determine its RbeSr and SmeNdisochron ages, (3) to compare these data with thoseobtained from other lherzolitic and related olivine-phyric basaltic shergottites to better understand theisotopic characteristics of their primitive mantle sourceregions, and (4) to discuss its possible petrogenesis.

Preliminary RbeSr isotopic results for Y984028were presented at the 32nd Symposium on AntarcticMeteorites (Shih et al., 2009b).

2. Analytical techniques

2.1. Samples and mineral separation procedures

Three interior pieces of Y984028, weighingw845 mg, were kindly allocated for this study by theNational Institute of Polar Research of Japan. Two smallbulk rock chunks weighing w51.6 mg were used forAreAr dating. Then, the remaining sample was furthercrushed to grain size

Fig. 1. CI-normalized Rb, Sr, Sm and Nd for shergottites. Twelve

bulk rocks of enriched basaltic shergottites including three olivine-

phyric shergottites (RBT 04262, LAR 06319 and NWA 1068) shown

in the blue area have distinctively higher abundances of these

elements. Six bulk rocks of lherzolitic shergottites shown in the

yellow area have lower abundances and more fractionated patterns

with Rb depleted relative to Sr and Nd depleted relative to Sm. The

Y984028 bulk rock sample (solid red circles) plots in the lherzolitic

shergottite field along with two other Yamato lherzolitic shergottites,

Y000097 (open diamonds) and Y-793605 (open hexagons) (Misawa

et al., 2008). Olivine-phyric shergottite RBT 04262 (open triangles)

is petrographically indistinguishable from the lherzolitic shergottites

(Mikouchi et al., 2008), but lies in the enriched basaltic shergottite

field. Also plotting within the lower portion of the enriched basaltic

shergottite field is the 346 Ma old basaltic (or diabasic) shergottite

NWA 1460 (open squares) of Nyquist et al. (2009) (For interpretation

of the references to colour in this figure legend, the reader is referred

to the web version of this article.).

517C.-Y. Shih et al. / Polar Science 4 (2011) 515e529

(144Nd16Oþ143Nd17Oþ142Nd18O)) ¼ 0.724487 (equiva-lent of 146Nd/144Nd ¼ 0.724140 (Nyquist et al., 1995)).The averagevalue of 8 analysesmadeduring the course ofthis study was 143Nd/144Nd ¼ 0.511172 � 0.000007(2sm) or�0.000020 (2sp). The 143Nd/144Nd data for thesamples were adjusted to 143Nd/144Nd ¼ 0.511138 forAmes Nd standard (Wasserburg et al., 1981). The143Nd/144Nd data reported here can be readily convertedto the 143Nd/144Nd data normalized to146Nd/144Nd ¼ 0.7219 by simply dividing the values bythe factor 0.998457. The total procedural blanks for Rb(w10 pg), Sr (w30 pg), Sm (w10 pg) and Nd (w20 pg)are low and negligible.

3. Analytical results

3.1. Rb, Sr, Sm and Nd abundances in enrichedbasaltic shergottites and lherzolitic shergottites

The CI-normalized Rb, Sr, Sm and Nd abundances ofenriched basaltic and lherzolitic shergottites aresummarized in Fig. 1. All data used in the plot are bulkrock samples analyzed in this lab by the isotope dilutiontechnique. Mean Rb, Sr, Sm and Nd abundance valuesfor CI chondrites listed in Anders and Grevesse (1989)were used for normalization. Seven enriched basalticshergottites (Los Angeles, Shergotty, Zagami, NWA3171, NWA 856, Dho 378 and NWA 2975) and threeolivine-phyric shergottites (RBT 04262, LAR 06319andNWA1068) have distinctively higher abundances ofthese elements relative to five lherzolitic shergottites(NWA 1950, ALHA 77005, Y000097, LEW 88615 andY-793605) by factors of 2e10�. Los Angeles has thehighest elemental abundances alongwith the lowest mg-value (w0.24), representing the most fractionated meltamong all the enriched shergottites. Also, enrichedbasaltic shergottites tend to have less fractionatedpatterns, i.e., Rb less depleted relative to Sr and Nd lessdepleted relative to Sm. The abundances of these traceelements vary significantly by factors of 5e6� amongten enriched basaltic shergottites.

Olivine-phyric shergottite RBT 04262 is petro-graphically similar to lherzolitic shergottites(Mikouchi et al., 2008), but its trace-elemental distri-bution pattern clearly lies in the field defined forenriched basaltic shergottites. Additional age data andinitial Sr and Nd isotopic ratios also indicate that RBT04262 is a member of the enriched shergottite clan, orit was derived from the magma of enriched basalticshergottite source regions (Shih et al., 2009a).

Contrary to the enriched shergottites, all six lher-zolitic shergottites have very restricted and low Rb

abundances of w0.3e0.4 � CI and w4� variations inSr, Sm and Nd abundances of w1e3 � CI. In Fig. 1,the elemental distribution of Y984028 plots within thearea defined by five lherzolitic shergottites, consistentwith its petrography as a lherzolitic shergottite(Kojima, 2008; Misawa et al., 2009). Y984028 hasa similar elemental pattern as lherzolitic shergotti-tesY000097 and Y-793605 (Misawa et al., 2008).Despite similarly young w170 Ma ages for enrichedand lherzolitic shergottites, their Rb, Sr, Sm and Nddistribution patterns are distinctly different.

The basaltic (or diabasic) shergottite NWA 1460 hasan older age of 346 Ma (Nyquist et al., 2009). It hasa low mg-value of w0.48 comparable to Shergotty-type enriched shergottites, and may represent a melt ofbasaltic composition. The Rb, Sr, Sm, and Nd distri-bution pattern for NWA 1460 is similar to that of thelherzolitic shergottites (Fig. 1). But, the overall abun-dances of these elements for NWA 1460 are w3-foldmore than for the lherzolitic shergottites and plot in the

Fig. 2. CI-normalized Sm and Nd distributions in bulk and mineral

separates of lherzolitic shergottite Y984028 (red circles) as deter-

mined in this study. Bulk rock sample data are identical to the data of

Y000097 (open squares, Misawa et al., 2008) and Y-793605 (inverted

triangles, Misawa et al., 2006), but w2� higher than the Y-763605data (triangles) of Ebihara et al. (1997). Based on the similarity in

REE abundances, these three Yamato lherzolites could be derived

from the same magmatic flow. The acid-washed olivine sample, Ol

(r), has the lowest Sm and Nd contents. The acid leachate of the bulk

rock, WR(l), and the combined acid leachates for all minerals, Leach,

have the highest Sm and Nd contents. They probably contain REE

from REE-rich phosphates readily leached out from mineral grains.

WR ¼ bulk rock, Px ¼ pyroxene, Plag ¼ maskelynite, Ol ¼ olivine,MM ¼ most magnetic minerals, r ¼ residues after HCl-wash,l¼ leachate after HCl-wash, Leach ¼ all mineral leachates combined(For interpretation of the references to colour in this figure legend,

the reader is referred to the web version of this article.).


enriched shergottite field. For a two-stage model, time-averaged 87Rb/86Sr and 147Sm/144Nd in the mantlesource reservoirs of NWA 1460 and the lherzoliticshergottites are very similar. Thus, these two rock typesare probably related in the sense of coming fromsimilar mantle source regions (Barrat et al., 2002;Nyquist et al., 2009). Thus, basaltic melts similar toNWA 1460 extruded at w170 Ma ago could berepresentative of the parental magmas to lherzoliticshergottites.

3.2. SmeNd isotopic system for Y984028

The Sm and Nd analytical results for three bulk rockand seven mineral samples of lherzolitic shergottiteY984028 are presented in Table 1. Their CI-normalizedSm and Nd distributions are presented in Fig. 2. Forcomparisons, Sm and Nd data for a bulk rock sample oflherzolite Y000097 (Misawa et al., 2008), and REE datafor two aliquots of the bulk rock sample of lherzolite Y-793605 (Ebihara et al., 1997; Misawa et al., 2006) arealso plotted. The Sm andNd data ofY984028, Y000097,and the Y-793605 sample of Misawa et al. (2006) arealmost identical indicating that they could be derivedfrom the same magmatic flow. This observation also isconsistent with their petrographic similarities as repor-ted by Mikouchi et al. (2009).

Sm and Nd abundances in mineral separates ofY984028 range over two orders of magnitude, e.g., Nd

Table 1

The SmeNd analytical results for lherzolitic shergottite Y984028.

Samplea wt. (mg) Sm (ppm) Nd (ppm) 147Sm/144Ndb 143Nd/144Ndb,c

Fig. 3. SmeNd mineral isochron for lherzolitic shergottite Y984028.

Nine data were fitted by the Isoplot linear regression program

(Ludwig, 2008). The isochron corresponds to a SmeNd age of170 � 10 Ma for l(147Sm) ¼ 0.00654 Ga�1. The initial 143Nd/144Nd-ratio is þ11.6 � 0.2 expressed in 3-units using the notation andreference parameters of DePaolo and Wasserburg (1976) and

Jacobsen and Wasserburg (1984). The inset shows deviations of143Nd/144Nd in 3-units for all nine samples relative to the 170 Ma

isochron. Dotted lines on either side of the isochron correspond to

�10 Ma.


ranges from w0.02 � CI to w15 � CI. Three olivine-bearing samples, Ol(r), MM(r) and Px2(r), contain verylow Sm and Nd abundances. Two acid leachates, WR(l) and Leach, have the highest Sm and Nd for con-taining acid soluble phosphate minerals, merrillites andphosphate-rich altered products (Misawa et al., 2009).All the bulk rock and mineral separates exhibitdistinctly high Sm/Nd ratios, i.e., LREE-depletedfeatures to various degrees, clearly indicating that theparental magma of Y984028 from which theseminerals were crystallized was also depleted in LREE.Our seven mineral separates used in this RbeSr andSmeNd study represent at least four of the majormineral components reported in the sample: pyroxene,olivine, plagioclase and phosphates (merrillite).

The 147Sm/144Nd and 143Nd/144Nd data of bulk rocksand mineral separates listed in Table 1 are presented inthe 147Sm/144Nd and 143Nd/144Nd isochron diagram ofFig. 3. All nine data points define a good linear arraycorresponding to a SmeNd age of 170 � 10 Ma usingeither the Isoplot regression program (Ludwig, 2008) orthe Williamson regression program (Williamson, 1968)commonly used in our lab and l(147Sm) ¼0.00654 Ga�1. The SmeNd isochron for Y984028 iswell-defined with an MSWD ¼ 1.9. The Nd abundancein Ol(r) is too low to measure its isotopic compositionwith good precision. Since the linearity of the data isdetermined by four mineral components, it is nota mixing line but represents the crystallization age of therock. The age agrees with those of other lherzoliticshergottites and enriched basaltic shergottites includingolivine-phyric shergottites. Therefore, all these sher-gottites crystallized contemporaneously.

All nine samples of Y984028, including the plagio-clase mineral separate, have 147Sm/144Nd-ratios,0.25e0.57, greater than the chondritic value of 0.1967.Some pyroxenes in Y984028 are large oikocrysts ofcumulate origin (Mikouchi et al., 2009).AssumingPx1(r)represents such a pyroxene sample, the 147Sm/144Nd ofthe parental magma in equilibrium with Px1(r) can becalculated from the partition coefficients of Sm and Ndand the 147Sm/144Nd-ratio of the pyroxene. Using theDSm/Nd ¼ 1.6e1.9 (orthopyroxene and pigeonite) and147Sm/144Nd ¼ 0.568 as for Px1(r) for such pyroxeneoikocrysts, the calculated 147Sm/144Nd-ratio for theparental magma is in the range w0.30e0.36, i.e.,a LREE-depleted melt similar to the bulk rock value ofNWA 1460 (0.29) reported in Nyquist et al. (2009). Ananalogous calculation for plagioclase using Plag(r) with147Sm/144Nd ¼ 0.251 and DSm/Nd ¼ 0.72, also yieldsa similarly LREE-depleted magma with147Sm/144Nd ¼ 0.35.

The initial 3Nd-value, the143Nd/144Nd-ratio calcu-

lated in 3-units according to the notation and referenceparameters of DePaolo and Wasserburg (1976), for theSmeNd isochron of Y984028 is þ11.6 � 0.2. It issimilar to those (þ8.3 to þ11.7) reported for otherlherzolitic shergottites but dissimilar to those ofenriched basaltic shergottites (�6.2 to �6.9) withsimilar ages. A two-stage model indicates that lher-zolitic shergottites and enriched basaltic shergottiteswere derived from very different sources. The formerwere derived from high Sm/Nd (or LREE-depleted)sources, the latter from low Sm/Nd (or LREE-enriched) sources. The low initial 3Nd-value forolivine-phyric shergottite RBT 04262 (�6.7 � 0.2,Shih et al., 2009a) indicates it is related to enrichedshergottites, not lherzolitic shergottites despite thepetrographic similarity with lherzolitic shergottiteLEW 88615 suggested by Mikouchi et al. (2008). Inaddition, different CRE ages for LEW 88615(3.9 � 0.4 Ma, Eugster et al., 1997; Nyquist et al.,2001) and RBT 04262 (2.3 � 0.6 Ma, Nagao andPark, 2008) preclude that these two meteorites wereejected from a single event. They were probably fromdifferent locations of the Martian surface.

The petrographic and noble gas studies of threeYamato lherzolitic shergottites, Y984028, Y000097 andY-793605 suggested that these rocks were closed relatedand probably even paired (Mikouchi et al., 2009; Nagaoet al., 2008; Nagao, 2009). The SmeNd ages of these


rocks also are similar within error limits, i.e. 152� 13Mafor Y000097 (Misawa et al., 2008) and 185 � 16 Ma forY-793605 (Misawa et al., 2006). To further investigate therelationship among these three lherzolitic shergottites, weplot SmeNd isotopic data for their bulk rock and mineralsamples in Fig. 4. To eliminate possible inter-laboratorybias, we use only SmeNd isotopic data from JSC, i.e.,nineY984028 data from this study aswell as sixY000097and threeY-793605data reported byMisawaet al. (2008).The 170 Ma Y984028 SmeNd isochron is used asa reference because it is the most well-defined SmeNdisochron obtained so far among the Yamato shergottites.Most of the eighteen samples lie on the 170 Ma isochronwithin their error limits. Only two samples, the pyroxenesample ofY000097 (Y00Px(r)) and the acid-washed bulkrock samples of Y-793605 (Y79 WR(r)) fall off theisochron. Excluding these two data, the remaining sixteendata yield an age of 171 � 16 Ma (Isoplot regression) or171 � 10 Ma (Williamson regression) with anMSWD ¼ 2.4.

3.3. RbeSr isotopic system for Y984028

The 87Rb/86Sr and 87Sr/86Sr data for three bulk rocksand seven mineral separates of Y984028 are listed inTable 2, and the 87Rb/86Sr and 87Sr/86Sr isochron diagramis presented in Fig. 5. The majority of the data forms

Fig. 4. SmeNd isotopic data for bulk rock and mineral samples of

three Yamato lherzolitic shergottites, Y984028 (red circles),

Y000097 (yellow diamonds) and Y-793605 (green triangles). All

these samples were analyzed at JSC including six Y000097 and three

Y-793605 data reported by Misawa et al. (2008). The line represents

the 170 Ma Y984028 SmeNd isochron. Most of the eighteen

samples lie on the 170 Ma isochron within error limits, except the

pyroxene sample of Y000097 (Y00 Px(r)) and the acid-washed bulk

rock samples of Y-793605 (Y79 WR(r)) (For interpretation of the

references to colour in this figure legend, the reader is referred to the

web version of this article.).

a linear array except for two acid leachate samples: thebulk rock leachate, WR(l), and the combined mineralleachates, Leach. Their 87Sr/86Sr data are low (w0.7104)and deviate considerably from the main array. They maycontain significant amounts of terrestrial weatheringproducts esea water, rainwater and aeolian dusts of low87Sr/86Sr of w0.709. Thus, they cannot be used todetermine the crystallization age of the rock. Excludingthese two leachates, WR(l) and Leach, eight data fromY984028 form a linear array defining a RbeSr age of175 � 10 Ma (Isoplot regression, MSWD ¼ 8.8) for l(87Rb) ¼ 0.01402 Ga�1 (Begemann et al., 2001) andinitial 87Sr/86Sr¼ 0.710379� 0.000036. Further omittingthe Ol(r) sample, the remaining seven samples yielda better fit for 170 � 9 Ma and initial87Sr/86Sr ¼ 0.710389 � 0.000029 (Isoplot regression,MSWD ¼ 4.6) or 170 � 3 Ma and initial87Sr/86Sr ¼ 0.710389 � 0.000010 (Williamson regres-sion). The Ol(r) sample has the lowest Sr concentrationand may have been susceptible to terrestrial contamina-tions and Rb losses. This RbeSr isochron age is identicalto its SmeNd isochron age, clearly indicating thatY984028 crystallized 170 Ma ago. The RbeSr isotopicsystem of Y984028 is slightly disturbed and is not asrobust as the respective SmeNd isotopic system.Y984028 has almost an identical RbeSr age asY-793605(T¼ 172� 13Ma;Morikawaet al., 2001).BothY984028and Y-793605 are probably contemporaneous withY000097 (T ¼ 157 � 28 Ma; Misawa et al., 2008).

The RbeSr isochron age of Y984028 agrees bothwith ages ranging 147e187 Ma for other lherzoliticshergottites and ages ranging 161e207 Ma forenriched basaltic shergottites (e.g., Nyquist et al.,2001). Its initial 87Sr/86Sr-ratio also is consistent withY984028 being a member of the lherzolitic shergottitegroup.

In order to investigate the RbeSr isotopic rela-tionships among the three Yamato lherzolitic shergot-tites, Y984028, Y000097 and Y-793605, we plot allbulk rock and mineral separate data for these rocks ina single RbeSr isochron diagram (Fig. 6). The dataplotted all were determined at JSC and include ten datafrom Y984028 (this study), eight data from Y000097(Misawa et al., 2008) and 4 data from Y-793605(Misawa et al., 2008). Most of the twelve Y000097 andY-793605 samples lie on or near the 170 Ma Y984028isochron within w13-unit in 87Sr/86Sr, except for theplagioclase sample of Y000097 (Y00 Plag(r)) and theacid leachate of the bulk rock sample of Y-793605(Y79 WR(l)). The Y00 Plag(r) has unusually high-Srconcentration (346 ppm) probably due to the intro-duction of the high-Sr concentration from metasomatic

Table 2

The RbeSr analytical results for lherzolitic shergottite Y984028.

Samplea wt. (mg) Rb (ppm) Sr (ppm) 87Rb/86Srb 87Sr/86Srb,c

Fig. 7. Rb and Sr distributions for bulk rock and mineral samples

from Y984028. The major mineral components of the rock are

olivine (46.4 vol%), pyroxene (47.9 vol%), and plagioclase (mas-

kelynite) (4.7 vol%) (Hu et al., 2009). In our analyses these

components are best represented by Ol(r), Px1(r), and Plag(r),

respectively. A fourth component is enriched in the mixed sample,

Plag þ Px(r). The inset shows Rb and Sr distributions for bulk rockand mineral samples determined in RbeSr isochron studies of three

Yamato lherzolitic shergottites, Y984028 (yellow field with red

circles, this study), Y000097 (blue field with triangles, Misawa et al.,

2008) and Y-793605 (green field with squares, Morikawa et al.,

2001). The plagioclase sample, Plag(r), from Y000097 (Misawa

et al., 2008) has the highest Sr reported for martian plagioclase

samples ande3x the Sr abundances for plagioclase in Y984028 and Y-

793605 (For interpretation of the references to colour in this figure

legend, the reader is referred to the web version of this article.).


interpretation, although shock-melt veins were repor-ted in Y000027/47/97 (Mikouchi and Kurihara, 2008).On the contrary, the SmeNd isotopic system of thepaired sample Y984028 is well-behaved, and so is itsRbeSr isotopic system to a lesser degree. A smallamount of a regolith breccia has been reported inY984028 (Riches et al., 2009). Because no effects ofbrecciation are noticeable on the RbeSr and SmeNdisotopic systems of Y984028, such brecciated regolithmaterials probably were local and similar in age to themain mass of Y984028. Probably the Plag(r) sample ofY000097 reflects local heterogeneity in the effects ofimpact brecciation of these rocks, possibly via localimpact melting of maskelynite. Excluding threeleachates, Ol(r) from Y984028, and Plag(r) fromY000097, the remaining seventeen samples definea RbeSr age of 177 � 9 Ma (Isoplot or Williamsonregression, MSWD ¼ 9.4) and initial87Sr/86Sr ¼ 0.710371 � 0.000027. The comparativelyhigh MSWD value suggests that the RbeSr isotopicdata for most samples may have been slightly affectedby the same or similar events as caused the deviation ofPlag(r) from Y000097 and Ol(r) from Y984028 todeviate from the common isochron.

Fig. 7 shows Rb and Sr concentrations measured forthe ten bulk rock and mineral samples of Y984028. Themaskelynite sample, Plag(r), has the highest Rb and Srabundances. The Y984028 Ol(r) and Px2(r) sampleshave the lowest Rb and Sr abundances reported amonglherzolitic shergottites. In terms of Rb and Sr distribu-tions in Y984028, the rock is composed of four maincomponents, Plag(r), Ol(r), Px2(r) and a fourth compo-nent enriched in the mixed sample, Plag þ Px(r). Areasdefined by the Rb and Sr distributions for the subsamplesof Y984028 (this study), Y000097 (Misawa et al., 2008)and Y-793605 (Morikawa et al., 2001) are shown in theinset of the figure. The maskelynite sample, Plag(r),from Y000097 has the highest Sr abundance (346 ppm,Misawa et al., 2008) ever reported among maskelynitesamples in shergottites. No such high-Sr maskelynitehas been found in either Y984028 or Y-793605. TheY984028 Plag(r) sample contains 123 ppm of Sr, whichis identical to that of Y-793605 total Plag sample(Morikawa et al., 2001) and is only w1/3 of the Srabundance in the maskelynite sample reported inY000097, emphasizing the unusual nature of theY000097 Plag(r) sample. Unfortunately, because of thelimited size of the Y000097 specimen, the analysis ofmaskelynite from it has not been repeated. Rb and Srdistributions in minerals of the three Yamato lherzoliticshergottites suggest Y984028 is more closely related toY-793605 than Y000097.

4. Discussion

4.1. RbeSr and SmeNd isochrons for Y984028:Further evidence against an old (w4.1 Ga) age forShergottites

Crystallization ages of the basaltic and the lherzoliticshergottites have come under great debate lately. Most ofthese shergottites (e.g., Shergotty, Zagami, Los Angeles,EETA 79001, RBT 04262, LAR 06319, LEW 88615,ALHA 77005, Y000097, Y984028 etc.) yield concordantRbeSr and SmeNd mineral isochron ages of170e180Ma (e.g., Nyquist et al., 2001; Borg et al., 2002;Misawa et al., 2008; Shih et al., 2009a). Also, the UePbages of baddelyelite crystals in Zagami and NWA 3171are also consistent with the young age (Herd et al., 2010).However, it has been argued that 207Pbe206Pb agesdetermined by Zagami, Shergotty and Los Angelesmaskelynites provide support for an old crystallizationage ofw4.1 Ga (e.g., Bouvier et al., 2008). These authorsargue that because plagioclases contain abundant Pb andweathering-resistant Pb isotopic compositions that the


207Pbe206Pb ages are robust. In contrast, they argue thatyoungw180Ma ages determined by other chronometerswere heavily influenced by later shock-impact events and/or by interactions with secondary REE-rich phosphatesformed from fluids percolating through the crust (e.g.,Bouvier et al., 2008).

The old (w4.1 Ga) age for shergottites was recentlyfurther challenged by Nyquist et al. (2009) in theirfinding of identical young RbeSr, SmeNd and39Are40Ar ages for shergottite NWA 1460/480. Inaddition to the concordancy of ages obtained by threeindependent methods, those authors presented argu-ments that each of the young ages was individuallyrobust. The RbeSr and SmeNd ages rely on the knownpartitioning behavior of trace elements among majormineral phases. There was no evidence of excessradiogenic 40Ar* in the high-temperature Ar releasesthat released 41e98% of the 39Ar produced from K inplagioclase by neutron irradiation. It was possible tosuccessfully correct for 40Ar present in the lowertemperature releases by reducing the data in an isochronplot. The arguments for a young crystallization age forNWA 1460 based on the RbeSr and SmeNd data areapplicable to Y984028. Additional supports for inter-preting the young ages of shergottites as crystallizationages are worth reemphasizing. (1) Petrographically,shergottites are generally medium- to coarse-grainedigneous rocks containing abundant cumulus pyroxenessimilar to typical terrestrial basaltic or diabasic rocks.However, they have been moderately shocked (e.g.,w30e35 GPa for the basaltic suite andw45 GPa for thelherzolitic suite) so that plagioclases were completelytransformed to maskelynites but pyroxenes and olivinesunderwent only shear fractures and strong mosaicism.Some shergottites also contain small amount of melt-pockets andmelt-veins. However, the chemistry of theseminerals remained mostly intact despite their apparentphysical damage by shockmetamorphism. In fact, Jones(1986) used the observation of major element zonings inolivines as primary evidence against resetting RbeSrand SmeNd isochrons by such shock events. Further-more, recent TEM and SEM-CL results of baddeleyitesin NWA 3171 showed that their crystal structures werenot affected by shock pressures of 30e35 GPa, and thusthe young UePb age determined by baddeleyitesrepresent the primary igneous age of the rock (Herdet al., 2010). (2) Shergottites are not likely crystallineimpact melt rocks because pyroxene/olivine cumulatetextures commonly associated with them (e.g., Zagami,Shergotty (cf., Stolper and McSween, 1979) and lher-zolitic shergottites (cf., Mikouchi and Kurihara, 2008))were not observed in the Apollo 16 impact melt rocks,

which generally exhibited fine-grained subophitictextures (e.g., Reimold et al., 1985). (3) REE-richphosphates are very soluble in dilute acids (Dreibuset al., 1996). Most of the bulk rock and mineral sepa-rates used in RbeSr and SmeNd isochronswerewashedwith HCl to remove the potential phosphate contami-nants (see Figs. 3 and 5). Actually, theNd abundance and147Sm/144Nd ratios for our acid-washed plagioclase andpyroxene separates are comparable to those dataobtained for individual mineral grains of lherzoliticshergottite LEW 88516 by the ion microprobe method(Harvey et al., 1993). For example, we obtainedNd ¼ 0.0352, 0.179 and 0.108 ppm and147Sm/144Nd ¼ 0.453, 0.568 and 0.251, for Px2(r), Px1(r) and Plag(r), respectively (see Table 1) as compared toNd¼ 0.009e0.133, 0.22e0.61, and 0.073� 0.023 ppmand 147Sm/144Nd ¼ 0.58 � 0.06, 0.51 � 0.04 and0.18� 0.11 for low-Ca Px, high-Ca Px and maskelynitegrains, respectively (Harvey et al., 1993). (4) Further, inthe SmeNd isotopic study of the eucrite Ibitira,Prinzhofer et al. (1992) showed evidence that pyroxeneswere apparently more robust and less likely to bedisturbed in the SmeNd isotopic system relative toplagioclases. Assuming the Y984028 pyroxenes to be asresistant to disturbances as pyroxenes in Ibitira, wecalculated initial 87Sr/86Sr- and 143Nd/144Nd-ratios forY984028 from the pyroxene mineral separates at thepresumed old age of 4.1 Ga and found the values to beunreasonably low, 0.674e0.691 for 87Sr/86Sr and 0.497(or �179 3) e 0.500 (or �120 3) for 143Nd/144Nd, ascompared to the initial 87Sr/86Sr-ratio of 0.698972 forthe angrite LEW86010 and its chondritic initial143Nd/144Nd-ratio of 0.505893 at 4.56Ga (Nyquist et al.,1994). These results are clearly against an old age ofw4.1 Ga for Y984028.

In addition to the above, detailed thermal modelingof 39Are40Ar data for the Zagami shergottite plagio-clase suggested that its ejected mass was impossiblylarge (radius w1 km), if the meteorite was w4 Ga oldand strongly degassed of its 40Ar during shock-heatingto w70 �C when it was ejected from Mars by a mete-orite impact w3 Ma ago (Bogard and Park, 2008).

In summary, mineral-petrographic evidence, lowshock-heating temperatures, Ar thermal modeling, aswell as RbeSr, SmeNd, AreAr, and baddelyelite UePbages are all consistent with the young crystallization agesfor shergottites, but against old ages for them.

4.2. Relationships of lherzolitic Shergottites

Parallelograms showing initial 3Nd-values vs. age forfive lherzolitic shergottites e ALHA 77005 and LEW


88516 (Borg et al., 2002), Y-793605 (Misawa et al.,2006), Y-793605 WR and Y000097 (Misawa et al.,2008) and Y984028 (this study) are plotted in Fig. 8.All SmeNd age and initial 3Nd-value data were obtainedusing the Isoplot program (Ludwig, 2008), except thatfor Y-793605WR. It was obtained using theWilliamsonprogram (Williamson, 1968) because of the smallnumber of data points. Regressing only a few data points(w3) by the Isoplot program produces unrealisticallylarge uncertainties. In Fig. 8, data for three Yamatolherzolites analyzed in the JSC lab, Y984028, Y-793605WR, andY000097 either overlap or lie very close to eachother, indicating that these lherzolitic shergottitesprobably can be derived from the samemagma flow. TheY-793605 datum from Misawa et al. (2006) is for anal-yses done in the laboratory of Kobe University, Japan.That does not agree with the data for the four Y-793605WR samples analyzed at JSC may be due to inter-labo-ratory bias. An adjustment ofwþ0.5 3-units in 3143Nd-values is recommended for direct comparison of Ndisotopic data obtained from these two laboratories(Misawa et al., 2008). The analyses represented by theparallelograms for ALHA 77005 and LEW 88516 asreported by Borg et al. (2002) were made in the JSC lab.

Fig. 8. Initial 3Nd-value and age parallelograms for lherzolitic sher-

gottites. Data for three Yamato lherzolites analyzed in the JSC lab,

Y984028 (red parallelogram), four Y-793605 WR samples and

Y000097 (yellow parallelograms, Misawa et al., 2008) either overlap

or lie very close to each other, indicating that Y984028 and Y-

793605, and probably Y000097, have come from the same magma

flow. The Y-793605 datum of Misawa et al. (2006), shown in an open

parallelogram, was analyzed in the laboratory of Kobe University,

Japan. That it does not agree with the data for the other Yamato

samples may be due to inter-laboratory bias. The blue parallelograms

are for ALHA 77005 and LEW 88516 as reported by Borg et al.

(2002) for analyses in the JSC lab. ALHA 77005 plots very near

the Yamato lherzolitic shergottites, whereas LEW 88516 does not

(For interpretation of the references to colour in this figure legend,

the reader is referred to the web version of this article.).

ALHA 77005 plots very near the Yamato lherzoliticshergottites, whereas LEW88516 does not. The SmeNdisotopic data of these five lherzolitic shergottites suggestthat they could have been derived from as many as threeor four separate magmatic flows, represented by theirdifferent initial 3Nd-valuesw170Ma ago. Alternatively,these data allow the possibility of only two separatemagmatic flows if the error limits for Y984028 andALHA 77005 are slightly underestimated.

As for the SmeNd isotopic systems discussedabove, the initial 87Sr/86Sr-ratio and age parallelogramsfor five lherzolitic shergottites e Y-793605 (Morikawaet al., 2001), ALHA 77005 and LEW 88516 (Borget al., 2002), Y000097 (Misawa et al., 2008) andY984028 (this study) are shown in Fig. 9. All theRbeSr age and initial 87Sr/86Sr-ratio data wereobtained by regressions using the Isoplot program(Ludwig, 2008). Data for three Yamato lherzolites,Y984028, Y-793605, and Y000097 (obtained from fouracid-washed samples, Misawa et al., 2008) andY000097 (Plag(r)) (obtained from seven samplesexcluding the Plag(r) datum, Misawa et al., 2008) showthat the error parallelograms for three Yamato lherzo-lites overlap only if the extremely high-Sr plagioclasedatum of Y000097 (Fig. 7) is excluded because of thepossibility of contamination by related impacts on theMartian surface. In this case, the Yamato lherzolitesprobably came from the same magma flow w170 Maago. The error parallelograms for ALHA 77005 andLEW 88516 do not plot near those for the Yamatolherzolitic shergottites. Thus, these two lherzoliticshergottites must have come from different magmaflows than the Yamato lherzolitic shergottites.Considering both the RbeSr and the SmeNd data, ourpreferred interpretation is that these five lherzoliticshergottites come from three distinct magmatic flows:one for the three Yamato lherzolites, and one each forALHA 77005 and LEW 88516.

4.3. Relationships with enriched Shergottites

Initial 3Nd-value and age parallelograms for fivelherzolitic shergottites (ALHA 77005, LEW 88516,Y-793605, Y000097 and Y984028), six enrichedbasaltic shergottites (Shergotty, Zagami, Los Angeles,NWA 856, NWA 2975 and Dho 378), three enrichedolivine-phyric shergottites (NWA 1068, RBT 04262and LAR 06319) and NWA 1460, a basaltic (or dia-basic) shergottite with an older age of 346 Ma (Nyquistet al., 2009) are summarized in Fig. 10. Lherzoliticshergottites and NWA 1460 have positive initial 3Nd-values between þ8 and þ12, which correspond to

Fig. 9. Initial 87Sr/86Sr-ratio and age parallelograms for lherzolitic

shergottites. Data for three Yamato lherzolites, Y984028 (red parallel-

ogram), Y-793605 (yellow parallelogram, Morikawa et al., 2001),

Y000097 (open parallelogram for four acid-washed samples, Misawa

et al., 2008) or Y000097(ePlag(r)) (yellow parallelogram for seven

samples excluding the Plag(r) datum,Misawa et al., 2008) overlap or lie

very close to each other, indicating that Y984028 and Y-793605, and

probably Y000097, can be derived from the same magma flow crystal-

lized w170 Ma ago. The blue parallelograms show data for ALHA77005 and LEW 88516 (Borg et al., 2002). These two lherzolitic sher-

gottites must have come from different magma flows than the Yamato

lherzolites (For interpretation of the references to colour in this figure


Fig. 10. Initial 3Nd-value and age parallelograms for lherzolitic

shergottites and enriched basaltic shergottites. Lherzolitic shergot-

tites (red diamonds) and NWA 1460 (red triangle), a basaltic (or

diabasic) shergottite with an older age of 346 Ma (Nyquist et al.,

2009) have positive initial 3Nd-values between þ8 and þ12, whichcorrespond to depleted LREE source reservoirs of super-chondritic147Sm/144Nd of 0.211e0.218. Dotted lines represent calculated147Sm/144Nd source reservoirs for shergottites using a two-stage

model starting from a chondritic reservoir at 4.568 Ga ago. The two-

stage model for the NWA 1460 basaltic (or diabasic) shergottite and

the Y984028 lherzolitic shergottite yields the same source147Sm/144Nd of 0.217 for these two rocks. They may have derived

from melts of similar source reservoirs at different times. All the

enriched basaltic shergottites (open circles) and three olivine-phyric

shergottites (open squares) have negative initial 3Nd-values of �6.1 to�6.9 corresponding to LREE-enriched sources with 147Sm/144Nd of0.185e0.186. Although shergottites RBT 04262 and LEW 88516

have very similar petrography and mineralogy (Mikouchi et al.,

2008), and also ages (Shih et al., 2009a), their initial 3Nd-values

are significantly different (For interpretation of the references to

colour in this figure legend, the reader is referred to the web version

of this article.).


depleted LREE source reservoirs of super-chondritic147Sm/144Nd of 0.211e0.218 using a two-stage modelstarting from a chondritic reservoir with147Sm/144Nd ¼ 0.1967 at 4.568 Ga ago. The two-stagemodel for the NWA 1460 basaltic (or diabasic) sher-gottite and the Y984028 lherzolitic shergottite yieldsthe same source 147Sm/144Nd of 0.217 for these tworocks. They may have been derived from melts ofsimilar source reservoirs at different times, i.e. NWA1460 crystallized at 346 Ma and Y984028 at 170 Ma.

All six enriched basaltic shergottites and threeolivine-phyric shergottites have negative initial 3Nd-values of �6.1 to �6.9 corresponding to LREE-enriched sources with 147Sm/144Nd of 0.185e0.186.Although basaltic shergottites and lherzolitic shergot-tites crystallized almost contemporaneously w170 Maago, their source 147Sm/144Nd-ratios calculated basedon a two-stage model differ by w17%. Despite thatshergottites RBT 04262 and LEW 88516 have verysimilar petrography and mineralogy (Mikouchi et al.,2008), and also ages of w170 Ma (Shih et al.,2009a), their initial 3Nd-values differ considerably byw17 3-units and thus their magmatic origins are alsodifferent. Thus, RBT 04262 is more closely relatedpetrogenetically to enriched basaltic shergottites thanto lherzolitic shergottites.

Error parallelograms showing the initial 87Sr/86Sr-ratios and ages for five lherzolitic shergottites (ALHA77005, LEW 88516, Y-793605, Y000097 andY984028), five enriched basaltic shergottites (Sher-gotty, Zagami CG (coarse-grained portion of Zagami),Zagami FG (fine-grained portion of Zagami), LosAngeles, NWA 856), three enriched olivine-phyricshergottites (NWA 1068, RBT 04262 and LAR 06319)and NWA 1460 are summarized in Fig. 11. Lherzoliticshergottites and NWA 1460 have low initial 87Sr/86Sr-ratios of w0.7105 and w0.709, respectively, corre-sponding to 87Rb/86Sr source reservoirs of w0.18 andw0.16, respectively, based on a two-stage modelstarting with the CAI 87Sr/86Sr value of 0.698934 at4.568 Ga ago. The two-stage model suggests themantle source reservoirs for the NWA 1460 basalticshergottite and the Y984028 lherzolitic shergottite


were slightly different in 87Rb/86Sr. However, thedifference is only w12%, and the agreement betweentwo-stage model source 87Rb/86Sr-ratios is the closestamong different shergottite types.

All five enriched basaltic shergottites and threeolivine-phyric shergottites have much higher initial87Sr/86Sr-ratios of 0.721e0.723 implying mantle sour-ces of high time-averaged 87Rb/86Sr of 0.35e0.37. Thus,although, basaltic shergottites and lherzolitic shergot-tites crystallized almost contemporaneously, their initial87Sr/86Sr-ratios differ significantly. Correspondingly,their two-stage model source 87Rb/86Sr-ratios differ bya factor of two. Again, despite the fact that shergottitesRBT 04262 and LEW 88516 have very similar petrog-raphy and mineralogy (Mikouchi et al., 2008) and bothhave ages of w170 Ma (Shih et al., 2009a), the differ-ence in their respective initial 87Sr/86Sr-ratios ofw1.4%is so significant that they could not have come from the

Fig. 11. Initial 87Sr/86Sr-ratio and age parallelograms for lherzolitic

shergottites and enriched basaltic shergottites. Lherzolitic shergot-

tites (red diamonds) and NWA 1460 (red triangle), a basaltic sher-

gottite with an older age of 346 Ma (Nyquist et al., 2009) have low

initial 87Sr/86Sr-ratios of w0.7105 and w0.709, respectively, corre-sponding to 87Rb/86Sr source reservoirs of w0.18 and w0.16,respectively. Dotted lines represent calculated 87Rb/86Sr source

reservoirs for shergottites using a two-stage model starting with the

CAI 87Sr/86Sr value 0.698934 at 4.568 Ga ago. The model suggests

the source reservoirs for the NWA 1460 basaltic shergottite and the

Y984028 lherzolitic shergottite were slightly different in 87Rb/86Sr.

All the enriched basaltic shergottites (open circles) and three olivine-

phyric shergottites (open squares) have much higher initial 87Sr/86Sr-

ratios of 0.721e0.723 corresponding to sources of 87Rb/86Sr of0.35e0.37. Shergottites RBT 04262 and LEW 88516 have very

similar petrography, mineralogy (Mikouchi et al., 2008) and ages

(Shih et al., 2009a), but the differences in their initial 87Sr/86Sr-ratios

are significant (For interpretation of the references to colour in this

figure legend, the reader is referred to the web version of this

article.).

same mantle source unless RBT 04262 assimilated“crustal” material of very high Rb/Sr ratio whereas“ordinary” lherzolites did not.

4.4. Petrogenesis of Y984028

Lherzolitic shergottites (e.g., ALHA77005 etc.) aregenerally considered as Martian meteorites ofpyroxene-olivine cumulate origin (e.g., McSweenet al., 1979) that crystallized from mafic magmas inplutonic environments. In this regard, the petrogeneticmodel proposed for Y000097 by Mikouchi andKurihara (2008) also is relevant to the discussion oftrace element abundances in lherzolites such asY984028. We demonstrate here that the Rb, Sr, Sm andNd distributions in Y984028 can be reproducedreasonably well using a model involving accumulationof pyroxenes plus olivines together with some trappedmagma. The parental magma for Y984028 probablywas similar in composition to NWA 1460. As dis-cussed above, both were produced from mantle sourcesof the same 147Sm/144Nd ¼ 0.217 (Fig. 10) and similar87Rb/86Sr w0.16e0.18, but at different times. Thesecharacteristics probably apply more generally tomoderately differentiated, “average” Martian mantlematerials. Furthermore, the olivines in Y984028 areFo71�3 (Misawa et al., 2009), and a melt in equilibriumwith such olivines would have an mg-value of0.45 � 0.03, similar to that of NWA 1460 (w0.48). Asimple model calculation involving the addition of30% pyroxenes of the composition of Px1(r) and 45%olivines of the composition of Ol(r) in 25% parentmagma of the composition of the NWA 1460 basalticshergottite is presented in Fig. 12. The resultant rockconsisting of the (pyroxene þ olivine) cumulate with25% trapped parental magma matches Y984028 wellin all the four trace elements Rb. Sr, Nd, and Sm.These elements represent more generally, the lithophilealkali, alkaline earth, and rare earth elements. Themode of the modeled rock is 45% olivine, 48%pyroxene and 6% plagioclase which is comparable tothat of Y984028 (e.g., 46.4% olivine, 47.9% pyroxeneand 4.7% plagioclase (Hu et al., 2009) and 45%olivine, 47% pyroxene and 5% plagioclase(T. Mikouchi pers. comm., 2009)). The parentalmagma to Y984028 did not have exactly the samecomposition as NWA 1460, because the latter does notcontain olivine. However, the trace element features ofthe Y984028 parental magma probably were NWA1460-like, consistent with the interpretation that themantle sources for both rock types were fairly typicalof the Martian mantle.

Fig. 12. A simple pyroxene plus olivine accumulation model for the

genesis of the Y984028 lherzolitic shergottite. The CI-normalized

Rb, Sr, Sm, and Nd distribution of Y984028 (red circles) can be

modeled by the addition of 30% pyroxenes of the composition of Px1

(r) (green diamonds) and 45% olivines of the composition of Ol(r)

(green triangles) in 25% parent magma of the composition of NWA

1460 basaltic shergottite (blue squares). The resultant rock (yellow

hexagons) consisting of the pyroxene þ olivine cumulate with 25%trapped parental magma matches Y984028 well in all these four

elements (For interpretation of the references to colour in this figure



The present study further supports the close petro-genetic association between shergottite NWA 480/1460and lherzolitic shergottites proposed by Barrat et al.(2002), Gillet et al. (2005) and Nyquist et al. (2009)on the basis of their similar REE distributions andtheir source Sm/Nd ratios.

5. Conclusions

Lherzolitic shergottites provide a good opportunity tounderstand the magmatic processes that produced suchrocks from the parental melts derived from Martianmantle sources. Several conclusions from this study of theRbeSr and SmeNd ages and initial Sr- and Nd-isotopiccompositions of Y984028 are summarized below:

1. Rb, Sr, Sm and Nd abundances of Y984028 aresimilar to those of other lherzolitic shergottites anddiffer from those of enriched basaltic shergottites,either the Shergotty-type, or several olivine-phyricshergottites like NWA 1068, RBT 04262 and LAR06319. Thus, these trace element abundances inY984028 are consistent with its petrographicclassification as a lherzolitic shergottite by Kojima(2008) and Misawa et al. (2009).

2. One bulk rock and eight acid-washed bulk rock andmineral samples of Y984028 yield a well-defined

SmeNd isochron age of 170 � 10 Ma and aninitial 3Nd ¼ þ11.6 � 0.2. The SmeNd isotopicsystem of Y984028 is not disturbed.

3. The same nine samples of Y984028, excluding twoleachates of the bulk rock and minerals, yield anidentical isochron age of 170 � 9 Ma for theRbeSr system, and an initial 87Sr/86Sr-ratio ¼ 0.710389 � 0.000029. However, the datasuggest that even in the Antarctic environment, theRbeSr isotopic system of Y984028 can be slightlydisturbed.

4. Both the SmeNd and RbeSr isochrons ofY984028 consist of data from four mineralcomponents and their ages are concordant. There-fore, Y984028 crystallized 170 � 7 Ma ago likemost other lherzolitic shergottites.

5. The possible presence of a minor breccia compo-nent in Y984028 described by Riches et al. (2009)did not seem to affect the RbeSr and SmeNdisotopic systems in our sample suggesting that thebrecciated materials are locally derived and prob-ably of similar petrologic type and age as the mainmass of Y984028.

6. Y984028 has a similar age and Sr and Nd isotopicsignatures as other Yamato lherzolitic shergottites,including Y-793605 and probably Y000097. Theage and isotopic signatures of these rocks areconsistent with derivation from the same magmaticflow as suggested by Mikouchi et al. (2009) fromtheir similar petrographic features.

7. The ages and Sr and Nd isotopic variations suggestthat the lherzolitic shergottites crystallized from atleast three parental magmas. Using a two-stageevolutionmodel, themantle sources of these parentalmagmas are implied to have super-chondritic147Sm/144Nd-ratios of w0.217 and sub-chondritic87Rb/86Sr-ratios ofw0.18. These ratios are similar tothose found for basaltic (or diabasic) shergottiteNWA 1460 (Nyquist et al., 2009), and probably arefairly typical of “average” Martian mantle materials.

8. The Rb, Sr, Sm and Nd distributions in Y984028can be modeled by the accumulation of w30%pyroxenes and w45% olivines accompanied byw25% trapped NWA 1460-like parental magma atw170 Ma ago. Although NWA 1460 lacks olivine,and therefore is unlikely to be an exact match forthe parental magma of the lherzolites, the successof the model for these four alkali, alkaline earth,and rare earth elements suggests that mantle meltswith the trace element features of NWA 1460 areprobably fairly common products of melting in theMartian mantle.


Acknowledgements

We appreciate two anonymous referees for theirthorough reviews and comments which greatly improvethe manuscript. We thank the National Institute of PolarResearch of Japan for providing the Y984028 samplefor this study and for its financial support for presentingthe preliminary results at the 32nd Symposium onAntarctic Meteorites in Tokyo. Financial support for theisotopic studies was provided by funding from theNASA Cosmochemistry Program to L. Nyquist,a Grand-in-Aid for Scientific Research (No. 17540464)to K. Misawa, and by NIPR Research Project Funds, P8(Evolution of the Early Solar System Materials).

References

Anders, E., Grevesse, N., 1989. Abundances of the elements:

meteoritic and solar. Geochim. Cosmochim. Acta 53, 197e214.Barrat, J.A., Gillet, P., Sautter, V., Jambon, A., Javoy, M., Gopel, C.,

Lesourd, M., Keller, F., Petit, E., 2002. Petrology and

geochemistry of the basaltic shergottite Northwest Africa 480.

Meteoritics Planet. Sci. 37, 487e499.

Begemann, F., Ludwig, K.R., Lugmair, G.W., Min, K., Nyquist, L.E.,

Patchett, P.J., Renne, P.R., Shih, C.-Y., Villa, I.M., Walker, R.J.,

2001. Call for an improved set of decay constants for geochro-

nological use. Geochim. Cosmochim. Acta 65, 111e121.

Bogard, D.D., Park, J., 2008. 39Are40Ar dating of the zagami

Martian shergottite and implications for magma origin of excess40Ar. Meteoritics Planet. Sci. 43, 1113e1126.

Borg, L.E., Nyquist, L.E., Wiesmann, H., Reese, Y., 2002.

Constraints on the petrogenesis of Martian meteorites from the

RbeSr and SmeNd isotopic systematics of the lherzolitic sher-gottites ALH 77005 and LEW 88516. Geochim. Cosmochim.

Acta 66, 2037e2053.

Bouvier, A., Blichert-Toft, J., Vervoort, J.D., Gillet, P., Albarede, F.,

2008. The case for old basaltic shergottites. Earth Planet. Sci.

Lett. 266, 105e124.

Catanzaro, E.J., Murthy, T.J., Garner, E.L., Shields, W.R., 1969.

Absolute isotopic abundance ratio and atomic weight of terres-

trial rubidium. J. Res. Nat. Bur. Stand. 73A, 511e516.Christen, F., Eugster, O., Busemann, H., 2005. Mars ejection times

and neutron capture effects of the nakhlites Y000593 and

Y000749, the olivine-phyric shergottite Y980459 and lherzolite

NWA1950. Antarct. Meteorite Res. 18, 117e132.DePaolo, D.J., Wasserburg, G.J., 1976. Nd isotopic variations and

petrogenetic models. Geophys. Res. Lett. 3, 249e252.

Dreibus, G., Jochum, K.H., Palme, H., Spettel, B., Wlotzka, F.,

Wänke, H., 1992. LEW 88516: a meteorite compositionally close

to the “Martian mantle”. Meteoritics 27, 216e217 (abstract).

Dreibus, G., Jagoutz, E., Spettel, B., Wanke, H., 1996. Phosphate-

mobilization on Mars? Implications from leach experiments on

SNC’s. Lunar Planet. Sci. XXVII, 323e324 (abstract).

Ebihara, M., Kong, P., Shinotsuka, K., 1997. Chemical composition

of Y-793605, a Martian lherzolite. Antarct. Meteorite Res. 10,

83e94.Eugster, O., Weigel, A., Polnau, E., 1997. Ejection times of Martian

meteorites. Geochim. Cosmochim. Acta 61, 2749e2757.

Gillet, P., Barrat, J.A., Beck, P., Marty, B., Greenwood, R.C.,

Franchi, I.A., Bohn, M., Cotton, J., 2005. Petrology, geochem-

istry, and cosmic-ray exposure age of lherzolitic shergottite

Northwest Africa 1950. Meteoritics Planet. Sci. 40, 1175e1184.

Harvey, R.P., Wadhwa, M., McSween Jr., H.Y., Crozaz, G., 1993.

Petrography, mineral chemistry and petrogenesis of Antarctic sher-

gottite LEW 88516. Geochim. Cosmochim. Acta 57, 4769e4783.Herd, C.D.K., Stern, R.A., Walton, E.L., Li, J., Bibby, C., 2010. TEM

and SEM-CL analysis of baddeleyite in NWA 3171: Geochro-

nological Implications for Martian Meteorites. In: (abstract) 41st

Lunar Planet. Sci. Conf., CD-ROM #2280.

Hu, S., Feng, L., Lin, Y.T., 2009. Petrography and mineral chemistry

of the Y984028 lherzolitic shergottite. Antarctic Meteorites

XXXII, 22e23 (abstract).Ikeda, Y., 1997. Petrology and mineralogy of the Y-793605 martian

meteorite. Antarct. Meteorite Res. 10, 13e40.

Jacobsen, S.B., Wasserburg, G.J., 1984. SmeNd isotopic evolution of

chondrites and achondrites, II. Earth Planet. Sci. Lett. 67, 137e150.Jones, J.H., 1986. A discussion of isotopic systematic and mineral

zoning in the shergottites: evidence for a 180 m.y. igneous

crystallization age. Geochim. Cosmochim. Acta 50, 969e977.

Kojima, H. (ed.), 2008. Meteorite Newsletter 17, 1e5.Ludwig, K.R., 2008. Isoplot/Ex Ver. 3.70. Berkely Geochronology

Center. Special publication No. 4.

McSween, H.Y., Stolper, E.M., Taylor, L.A., Nuntean, R.A.,

O’Kelley, G.D., Eldridge, J.S., Biswas, S., Ngo, H.T.,

Lipschutz, M.E., 1979. Petrogenetic relationship between Allan

Hills77005 and other achondrites. Earth Planet. Sci. Lett. 45,

275e284.Mikouchi, T., Kurihara, T., 2008. Mineralogy and petrology of paired

lherzolitic shergottites Yamato 000027, Yamato 000047, and

Yamato 000097: another fragment from a Martian “lherzolite”

block. Polar Sci. 2, 175e194.Mikouchi, T., Kurihara, T., Miyamoto, M., 2008. Petrology and

mineralogy of RBT 04262: Implications for stratigraphy of the

lherzolitic shergottite igneous block. In: (abstract) 39th Lunar

Planet. Sci. Conf., CD-ROM #2403.

Mikouchi, T., Satake, W., Kurihara, T., 2009. Y984028 Lherzolitic

shergottite: a new Antarctic find likely paired with Y000027/

000047/000097. Antarctic Meteorites XXXII, 35e36 (abstract).Misawa, K., Yamada, K., Nakamura, N.,Morikawa, N., Yamashita, K.,

Premo, W.R., 2006. SmeNd isotopic systematics of lherzolitic

shergottite Yamato-793605. Antarct. Meteorite Res. 19, 45e57.

Misawa, K., Park, J., Shih, C.-Y., Reese, Y., Bogard, D.D.,

Nyquist, L.E., 2008. RbeSr, SmeNd, and AreAr isotopic

systematics of lherzolitic shergottite Yamato 000097. Polar Sci.

2, 163e174.

Misawa, K., Greenwood, R.C., Franchi, I.A., Kaiden, H., Kojima, H.,

2009. New shergottites, Y980497 and Y984028, from the Yama-

toMountains. Antarctic Meteorites XXXII, 37e38 (abstract).

Morikawa, N., Misawa, K., Kondorosi, G., Premo, W.R.,

Tatsumoto, M., Nakamura, N., 2001. RbeSr isotopic systematics

of lherzolitic shergottite Yamoto-793605. Antarct. Meteorite Res.

14, 47e60.

Nagao, K., 2009. Noble gases of Yamato 984028 lherzolitic sher-

gottite. Antarctic Meteorites XXXII, 43e44 (abstract).

Nagao, K., Park, J., 2008. Noble gases and cosmic-ray exposure ages

of two Martian shergottites, RBT 06242 and LAR 06319,

recovered in Antarctica. Meteoritics Planet. Sci. 43, A107.

Nagao, K., Park, J., Choi, H.-G., 2008. Noble gases of the Yamato

000027 and Yamato 000097 lherzolitic shergottites from Mars.

Polar Sci. 2, 195e214.


Nyquist, L.E., Bansal, B., Wiesmann, H., Shih, C.-Y., 1994.

Neodymium, strontium, and chromium isotopic studies of the

LEW 86010 and Angra dos Reis meteorites and the chronology

of the angrite parent body. Meteoritics Planet. Sci. 29, 872e885.

Nyquist, L.E., Wiesmann, H., Bansal, B., Shih, C.-Y., Keith, J.E.,

Harper, C.L., 1995. 146Sme142Nd formation interval for the lunar

mantle. Geochim. Cosmochim. Acta 59, 2817e2837.Nyquist, L.E., Bogard, D.D., Shih, C.-Y., Greshake, A., Stöffler, D.,

Eugster, O., 2001. Ages and geologic histories of Martian

meteorites. Space Sci. Rev. 96, 105e164.

Nyquist, L.E., Bogard, D.D., Shih, C.-Y., Park, J., Reese, Y., Irving,A.J.,

2009. Concordant RbeSr, SmeNd, and AreAr ages for Northwest

Africa 1460: a 346Maold basaltic shergottite related to “lherzolitic”

shergottites. Geochim. Cosmochim. Acta 73, 4288e4309.Prinzhofer, A., Papanastassiou, D.A., Wasserburg, G.J., 1992.

Samarium-neodymium evolution of meteorites. Geochim.

Cosmochim. Acta 56, 797e815.

Reimold, W.U., Nyquist, L.E., Bansal, B.M., Wooden, J.L., Shih,

C.-Y., Wiesmann, H., Mackinnon, I.D.R., 1985. Isotope analysis

of crystalline impact melt rocks from Apollo 16 Station 11 and

13, North Ray Crater. In: Proc. 15th Lunar Planet. Sci Conf. Jour.

Geophy. Res. 90, C431eC448.Riches, A.J.V., Liu, Y., Rumble III, D., Taylor, L.A., 2009. Lherzo-

litic shergottite Yamato 984028: a possible martian regolith

breccias. Antarctic Meteorites XXXII, 62e63 (abstract).Shih, C.-Y., Nyquist, L.E., Bogard, D.D., McKay, G.A.,

Wooden, J.L., Bansal, B.M., Wiesmann, H., 1982. Chronology

and petrogenesis of young achondrites, Shergotty, Zagami, and

ALHA 77005: Late magmatism on a geologically active planet.

Geochim. Cosmochim. Acta 46, 2323e2344.

Shih, C.-Y., Nyquist, L.E., Wiesmann, H., 1999. Samariumeneodymium and rubidiumestrontium systematics of nakhlite

Governador Valadares. Meteoritics Planet. Sci. 34, 647e655.

Shih, C.-Y., Nyquist, L.E., Wiesmann, H., Barrat, J.A., 2003. Age

and petrogenesis of picritic shergottite NWA 1068: SmeNd andRbeSr isotopic studies. In: (abstract) 34th Lunar Planet. Sci.

Conf., CD-ROM #1439.

Shih, C.-Y., Nyquist, L.E., Reese, Y., 2009a. RbeSr and SmeNd

studies of olivine-phyric shergottites RBT 04262 and LAR 06319:

Isotopic evidence for relationship to enriched basaltic shergottites.

In: (abstract) 40th Lunar Planet. Sci. Conf., CD-ROM #1360.

Shih, C.-Y., Nyquist, L.E., Reese, Y., Misawa, K., 2009b. RbeSrisotopic studies of Antarctic lherzolitic shergottite Yamato

984028. Antarctic Meteorites XXXII, 66e67 (abstract).

Shirai, N., Ebihara, M., 2009. Chemical characteristics of the lherzo-

litic shergottiteYamato 000097:magmatismonMars inferred from

the chemical compositions of shergottites. Polar Sci. 3, 117e133.

Stolper, E.M., McSween, H.Y., 1979. Petrology and origin of the

shergottite meteorites. Geochim. Cosmochim. Acta 43,

1475e1498.Wasserburg, G.J., Jacobsen, S.B., DePaolo, D.J., McCulloch, M.T.,

Wen, T., 1981. Precise determination of Sm/Nd ratios, Sm and

Nd isotopic abundances in standard solutions. Geochim. Cos-

mochim. Acta 45, 2311e2323.

Williamson, J.H., 1968. Least-squares fitting of a straight line. Can. J.

Phys. 46, 1845e1847.

Sm-Nd and Rb-Sr studies of lherzolitic shergottite Yamato 984028IntroductionAnalytical techniquesSamples and mineral separation proceduresAnalytical procedures

Analytical resultsRb, Sr, Sm and Nd abundances in enriched basaltic shergottites and lherzolitic shergottitesSm-Nd isotopic system for Y984028Rb-Sr isotopic system for Y984028

DiscussionRb-Sr and Sm-Nd isochrons for Y984028: Further evidence against an old (∼4.1 Ga) age for ShergottitesRelationships of lherzolitic ShergottitesRelationships with enriched ShergottitesPetrogenesis of Y984028

ConclusionsAcknowledgementsReferences

Sm-Nd and Rb-Sr studies of lherzolitic shergottite Yamato ... › download › pdf ›...

Documents

Transcript of Sm-Nd and Rb-Sr studies of lherzolitic shergottite Yamato ... › download › pdf ›...