Shock experiments on basalt—Ferric sulfate mixes and their possible relevance to the sulfide bleb clusters in large impact melts in shergottites

Shock experiments on basalt

—

Ferric sulfate mixes and their possible relevance to the

sulfide bleb clusters in large impact melts in shergottites

M. N. RAO

, L. E. NYQUIST

, P. D. ASIMOW

, D. K. ROSS

4,5

, S. R. SUTTON

6,7

T. H. SEE

, C. Y. SHIH

, D. H. GARRISON

, S. J. WENTWORTH

, and J. PARK

SCI, Johnson Space Center, Houston, Texas 77058, USA

XI, NASA, Johnson Space Center, Houston, Texas 77058, USA

Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California 91125, USA

Jacobs JETS, NASA, Johnson Space Center, Houston, Texas 77058, USA

UTEP-CASSMAR, El Paso, Texas 79968, USA

Department of Geophysical Sciences, University of Chicago, Chicago, Illinois 60439, USA

CARS, Argonne National Laboratory, Argonne, Illinois 60439, USA

Barrios Technology/Jacobs JETS, NASA, Johnson Space Center, Houston, Texas 77058, USA

HEPCO, Jacobs JETS, NASA Johnson Space Center, Houston, Texas 77058, USA

Kingsborough Community College, Brooklyn, New York 11235, USA

Corresponding author. E-mail: sitarao@sbcglobal.net

(Received 29 July 2020; revision accepted 04 November 2021)

Abstract–

Large impact-melt pockets in shergottites contain both Martian regolith

components and sulfide/sulfite bleb clusters that yield high sulfur concentrations locally

compared to bulk shergottites. The regolith may be the source of excess sulfur in the

shergottite melt pockets. To explore whether shock and release of secondary Fe-sulfates

trapped in host rock voids is a plausible mechanism to generate the shergottite sulfur bleb

clusters, we carried out shock recovery experiments on an analog mixture of ferric sulfate

and Columbia River basalt at peak pressures of 21 and 31 GPa. The recovered products

from the 31 GPa experiment show mixtures of Fe-sulfide and Fe-sulfite blebs similar to the

sulfur-rich bleb clusters found in shergottite impact melts. The 21 GPa experiment did not

yield such blebs. The collapse of porosity and local high-strain shear heating in the 31 GPa

experiment presumably created high-temperature hotspots (

2000

C) sufficient to reduce

to Fe

and to decompose sulfate to sulfite, followed by concomitant reduction to

sulfide during pressure release. Our results suggest that similar processes might have

transpired during shock production of sulfur-rich bleb clusters in shergottite impact melts. It

is possible that very small CO presence in our experiments could have catalyzed the

reduction process. We plan to repeat the experiments without CO.

INTRODUCTION

Shock-induced melts, whether found in extraterrestrial

(shergottites, lunar samples, cometary objects) or terrestrial

(impact glasses, impact melt rocks, tektites) materials,

display a variety of Fe- and S-bearing blebs quenched

from immiscible sulfide melts (Basu, 2005; Ishii et al.,

2008; Keller & McKay, 1997; Noguchi et al., 2014;

Sheffer, 2007; Sheffer & Melosh, 2005; Sheffer et al.,

2006). Some of the blebs found in lunar soil agglutinates

and cometary objects contain np (nanophase) Fe

(metal)

particles, sometimes rimmed by FeS. By contrast,

immiscible silicate, sulfide, and Fe-Ni metal melts in

terrestrial impact melts, as well as sulfur blebs in tektites

and fulgurites, are principally produced from oxidized

(Fe

) iron-bearing minerals in silicate target rocks, though

they also yield some evidence for impact melting of the

projectile (Hamann et al., 2018). Impact melt (IM) pockets

and pods in shergottites (e.g., fig. 6 of Rao et al., 2018)

This paper is dedicated to the memory of Kent Ross, major

contributor to this work, who passed away in July 2021.

Meteoritics & Planetary Science

1–15 (2021)

doi: 10.1111/maps.13770

2021 The Meteoritical Society

contain excess sulfur in the form of large clusters of

micrometer-sized Fe- and S-bearing blebs with X-ray

absorption near edge spectr

a (XANES) below the sulfur

K-edge indicating mixed sulfite and sulfide species (Rao

et al., 2018; Sutton et al., 2008). Rao et al. (2018) noted

that these sulfur blebs are not related to the igneous

sulfides commonly found in shergottites because they yield

locally sulfur abundances (up to

20% as SO

) far above

the normal bulk sulfur abundance in shergottites,

0.5%

(Gattacceca et al., 2013; Lorand et al., 2005; McSween &

Jarosewich, 1983). The blebs in shergottite IMs show large

variations in Ni concentration, from

1% up to 6%

(Gattacceca et al., 2013). Shergottite IMs studied to date

do not contain any np Fe

particles (Rao et al., 2018;

Sutton et al., 2008). This variety of observed bleb types

likely reflects the range of precursor chemistry and the

range of pressure

–

temperature

–

time paths of the shock

and pressure release processes involved in their formation.

The shergottites

—

such as EETA79001, Shergotty,

and Zagami

—

that contain

cm-sized IM pockets (or

pods) release large quantities of Martian atmospheric

gases upon heating in vacuo (Becker & Pepin, 1984;

Bogard & Johnson, 1983; Swindle et al., 1986; Wiens,

1988). Such IMs also preserve isotopic evidence for the

entrainment of Martian regolith components, including

near-surface neutron-induced excesses of

Kr from the

decay of

Br(n,

)

Br and deficits in

149

Sm relative to

150

Sm attributed to galactic cosmic ray interactions

(Hidaka et al., 2009; Rao et al., 2002, 2011). Moreover,

some of these impact melts exhibit a pronounced

anticorrelation trend between sulfur (as SO

) and Si (as

SiO

) (shown in fig. 2 of Rao et al., 2018) that is similar

to the trend observed in soils and rocks at Gusev and

Meridiani craters on Mars (Br

€

uckner et al., 2008; Rao

et al., 2018). These results suggest that the precursors of

the sulfur-bearing components in shergottite IMs could

have been derived from the Martian regolith.

The present Martian surface environment, however,

is highly oxidizing and neither sulfide nor sulfite is

found commonly at the Martian rover landing sites

explored to date. In fact, sulfur is found predominantly

in the form of Fe and Ca sulfates in rocks and soils at

Meridiani, Gusev, and Gale craters, according to both

APXS and M

€

ossbauer measurements (Bibring et al.,

2007; Clark et al., 2005; Gellert et al., 2004; King &

McLennan, 2010; McAdam et al., 2014; Morris et al.,

2006). These sulfates are thought to be produced by

acid sulfate fluid alteration of igneous precursor

minerals under oxidizing conditions (McLennan, 2012).

In this context, Rao et al. (2018) proposed that the

sulfur-bearing blebs in shergottite IMs were produced

by shock melting of iron sulfates that had been

entrained, prior to impact, into voids in porous target

rocks by regolith gardening processes such as

sedimentary mixing, hydrothermal transport, or aeolian

activity (Greeley et al., 2004). Such sulfate might be

present in the form of amorphous precursors of

minerals such as melanterite and jarosite (ubiquitously

present near the Martian surface). Since amorphous

sulfate-bearing material is preferentially associated with

porosity, the proposed mechanism appeals to the

uniquely high temperatures generated in the vicinity of

collapsing pores upon shock consolidation. Amorphous

materials in pores would be exposed to temperatures

(

2000

C) sufficient to cause melting under pressure,

while nearby crystalline materials might remain

relatively unaffected (Beck et al., 2007; Gillet et al.,

2000; Sharp & DeCarli, 2006). Ferric sulfate in these

high-temperature melts would likely decompose to

ferrous sulfite and then could be partly reduced to

sulfide during pressure release from the high-

temperature shock state (Melosh & Artemieva, 2004;

Schrader et al., 2016; Sheffer, 2007; Sheffer & Melosh,

2005; Sheffer et al., 2006), producing a reduced sulfur-

rich immiscible melt. Upon quenching, the immiscible

melt would be dispersed, yielding innumerable

micrometer-sized sulfur blebs.

Although considerable information is available

regarding high-temperature and -pressure shock effects

on meteoritic and lunar mineral assemblages, attention

has mostly been directed toward primary rock-forming

minerals such as olivine, pyroxene, and feldspar.

Comparatively fewer results are available concerning

shock effects in secondary and accessory minerals such

as S-bearing constituents. Shock processing due to

micrometeorite impact on Fe

)

S (pyrrhotite) in lunar

soils and agglutinates has been studied in detail by

Keller and McKay (1997), Ishii et al. (2008), and Basu

(2005). Similar studies were also performed on

pyrrhotite in interplanetary and cometary dust particles

from Stardust, Itokawa, and comet 81P/Wild 2 (Ishii

et al., 2008; Keller & Berger, 2014; Noguchi et al.,

2014). Experimental shock results on oxidized accessory

phases such as ferric sulfate are lacking, although shock

decomposition thresholds for Ca- and Mg-sulfates have

been studied due to their significance as sources of

atmospheric SO

in the Chicxulub event (Chen et al.,

1994; Gupta et al., 2001; Ivanov & Deutsch, 2002;

Zhang & Sekine, 2007).

To examine the validity of our proposition that the

sulfur-bearing bleb clusters found in impact melts in

shergottites were likely produced by shock melting of

iron sulfate followed by reduction to sulfide during

release to low pressure, we carried out laboratory shock

experiments on test charges of Columbia River Basalt

(CRB) mixed with ferric sulfate (Fe

(SO

)

•

O) in

steel chambers. One experiment, at the Lindhurst

Laboratory of Experimental Geophysics at Caltech,

M. N. Rao et al.

reached peak pressure of 31 GPa and featured a

nominal amount of CO gas that was bled into the pore

space after adsorbed moisture and air were pumped out

of the experimental charge. The other experiment, at the

Experimental Impact Laboratory at NASA Johnson

Space Center (JSC), reached peak pressure of 21 GPa

and had air in the initial pore space.

The peak pressure and temperature conditions

experienced by shergottites can be estimated by

comparing their mineral assemblages to experiments,

but both dynamic and static experiments have critical

shortcomings as tools for such calibration (Tomioka &

Miyahara, 2017). Shock wave experiments can only

maintain high-pressure conditions for durations on the

order of microseconds (Gillet et al., 2007; Sharp &

DeCarli, 2006) whereas natural impact events may be

able to support high pressure for many seconds (Beck

et al., 2005; Bowling et al., 2020; Gillet & El Goresy,

2013). Furthermore, pressure and temperature paths in

shock recovery experiments are typically inferred from

ancillary data rather than directly measured. By

contrast, static high-pressure, high-temperature

experiments hold well-defined and measured (

)

conditions for durations typically much longer than

impact events (minutes to hours), but they lack the

dynamic, turbulent mixing processes that are

characteristic of impact melts just behind a shock front.

Here, we briefly review pressure estimates based on

olivine and feldspar in melt veins in shergottites, which

guided our choice of experimental pressure. Miyahara

et al. (2011) found the assemblage of bridgmanite and

ferropericlase in olivine

–

phyric shergottite DaG 735,

suggesting pressure in excess of

25 GPa for this

shergottite. Olivine-group polymorphs and their

decomposition products suggest pressures in the 23

–

GPa range in most shergottites (Chen et al., 1996; El

Goresy et al., 2013; Malavergne et al., 2001; Tomioka &

Miyahara, 2017). The ubiquitous conversion of matrix

plagioclase to amorphous maskelynite in shergottites is

thought to place a minimum pressure in the

neighborhood of 20 GPa for matrix regions, but

transformations of feldspars in the melt pockets and

veins of shergottites suggest higher pressure.

Langenhorst and Poirier (2000) suggested a liquidus

assemblage of stishovite plus liebermannite in shock

melt veins in Zagami, associated with lingunite. In

similar veins in Zagami, Beck et al. (2007) found

another liquidus assemblage of hexagonal (Ca

–

)

(CAS) and stishovite. In static

experiments at 1300 K, liebermannite requires pressure

above 12 GPa, lingunite requires 21

–

24 GPa (El Goresy

et al., 2013; Yagi et al., 1994). CAS and stishovite only

coexist in a narrow range near 22 GPa at 2000

–

2200

(Akaogi et al., 2010). All these results suggest that,

despite some variability in peak pressure and the

possibility of

–

evolution during the shock event,

20 GPa is a reasonable lower limit and

30 GPa is

a reasonable upper limit. We do not have independent

control of temperature in our shock recovery

experiments, but the development of impact melt

pockets suggests that near-liquidus temperatures,

≥

2000

C, were transiently achieved in collapsing pore

space.

The recovered sample charges were analyzed by

field-emission scanning electron microscopy (FE-SEM)

and electron probe microanalyzer (EPMA). The results

of these analyses are presented here (preliminary results

were reported by Rao et al., 2019). We show that bleb

clusters resembling those in shergottite impact melts can

in fact be produced by shock melting, decomposition,

and reduction of ferric sulfate. We suggest that this

mechanism is a viable path to shergottite IM blebs,

whereas shock melting of Fe

)

S (pyrrhotite) is not.

EXPERIMENTAL AND ANALYTICAL METHODS

31 GPa Shock Recovery Experiment at Caltech

About 50 mg of CRB powder with a typical grain

size of

m plus hydrated ferric sulfate

(Fe

[SO

]

•

O) in the form of powder and chips up

1 mm were mixed and heated to 110

C in vacuum

for a few hours (to remove adsorbed moisture) at the

Lindhurst Laboratory for Experimental Geophysics at

Caltech (Fig. 1a). This mix was pressed at 2200 psi into

a pellet of 5 mm diameter and 1.17 mm thick having an

estimated porosity of 28%. The pellet was loaded into

the cavity of a stainless steel (SS) 304 sample retainer

within a vented sample housing. The target assembly

was placed in the target chamber of a 20 mm bore

single stage gun, pumped to vacuum through the vent

port and exposed to laboratory grade CO. The pore

space in the sample contained

0.8 bars of CO gas at

ambient temperature. A 2.5 mm thick SS 304 flyer was

launched at a measured pre-impact velocity of 2.04 km

–

and impacted the surface of the sample holder,

generating a shock wave with a first shock state in the

sample estimated at 15 GPa and 1540 K and a reshock

31 GPa at 3500 K experienced by the rear portion of

the sample volume. Pressure was estimated using a

Hugoniot derived from Caltech experiments on

Saddleback Basalt mixed with the Hugoniot of ferric

sulfate using a mass-averaging procedure and an

estimated Gr

€

uneisen parameter to account for porosity

(initial density 2025 kg m

–

, initial bulk sound speed

1968 m s

–

, Hugoniot slope

= 1.42, and Gr

€

uneisen

1.37). In much of the literature on shock recovery

experiments, this shot would be reported using the “full

Shock experiments

ring-up” or reverberation pressure of 49 GPa; however,

given the thickness of the sample chamber relative to

the flyer plate, full reverberation did not occur in this

experiment. The recovered sample holder had a well-

defined crater with no evidence for failure of the sample

chamber or loss of sample (Fig. 1b).

The chamber was sawn open and one half of the

shocked sample was impregnated with epoxy (preserving

in situ spatial distribution of the reaction products) and

machined to fit into the FE-SEM sample holder at JSC.

The top surface of the sample was then gently polished

to remove epoxy and expose the shocked sample for

FE-SEM and EPMA measurements at JSC.

21 GPa Shock Recovery Experiment at JSC

The JSC experiment used the Flat-Plate Accelerator

facility, also a 20 mm caliber horizontal powder

propellant gun. The target assembly consisted of a disk

pressed from the same mix of CRB powder and ferric

sulfate chips as the Caltech experiment, moderately

pressed into a pellet with estimated porosity of 28%.

The pellet was loaded into a 9 mm diameter and 0.8 mm

deep SS sample holder. The target holder was then

loaded into an SS container capsule placed in the EIL

GUN chamber facility and a flat 304 SS flyer plate was

launched at 1.8 km s

–

into a metallic jacket that houses

the target assembly. A planar shock wave of known

amplitude was generated, producing an estimated 21

GPa peak shock pressure (See et al., 2001). In this case,

the very shallow sample chamber allowed for multiple

shock reverberations and it is reasonable to conclude

that the entire sample experienced the full ring-up

pressure. The shocked sample charge was subsequently

recovered by machining to open the SS capsule and

prepared into a thin section for petrographic and FE-

SEM studies. Note that the CRB plus ferric sulfate

mixed charges shocked at 21 and 31 GPa are aliquots of

the same sample mix. The only difference between the

two is that the Caltech 31 GPa target porosity was

exposed to CO prior to shock loading whereas the JSC

21 GPa target had ordinary air in the pores.

FE-SEM, EPMA, and Fe K-XANES Methods

FE-SEM

The field emission scanning electron microscope

model JEOL 7600F at JSC was used for studying the

polished thin sections of recovered shock experiments

with standard operating conditions at 15 kV accelerating

potential for backscattered electron imaging (BEI) and

energy-dispersive X-ray spectrometry (EDS). Both

secondary (SEM) and backscattered electron (BSE)

imaging modes were used. X-ray spectrometry data

were obtained using IXRF and Thermo Electron EDS

systems attached to the FE-SEM. Qualitative major

element compositions of minerals were determined at 10

or 15 KeV with this system (Ross et al., 2011;

Wentworth et al., 2005) and are reported normalized to

100 atomic%.

EPMA

Polished thin sections were analyzed using a

Cameca SX100 microbeam automated electron

microprobe equipped with standard Cameca instrument

controls and PAP matrix correction software. Standard

analytical methods and calibration procedures were

used at 15 kV accelerating potential and 20 nA beam

Fig. 1. Experimental setup.a) Mixture of powdered Columbia River Basalt and ferric sulfate (white chips) prior to pressing into

pellet for loading into Caltech sample recovery chamber.b) Post-shot photograph of recovery target with visible crater and intact

chamber.

M. N. Rao et al.

current (except for phosphates and glasses where a

defocused beam [20

m] of 10 nA was employed).

Typical dwell times on the element peaks are 10

–

30 s.

Backgrounds were similarly checked. The following

standards were used in the analysis: Marjahlati olivine

is used for Fe in olivine; diopside for Si, Ca, and Mg in

pyroxene; orthoclase for K; and Hawk Mountain

oligoclase for Na and Al. Chromite standard was used

for Cr, Al, Fe, and Mg for the Fe, Ti, and Cr oxides.

Sulfides were analyzed using Canyon Diablo troilite

standard and a pyrite metal block for S and Fe and

pentlandite for Ni. Rutile was used for Ti, Durango

apatite for P, and tugtupite for Cl. Data are reported

normalized to 100% as the samples display variable

amounts of water loss (Rao et al., 1999; Wentworth

et al., 2005).

Fe K-XANES

The Fe K-edge XANES measurements were carried

out at the beamline 2-ID-B of the Advanced Photon

Source (APS) at Argonne National Laboratory,

Argonne, IL (University of Chicago) equipped with an

intermediate energy scanning X-ray microscope (SXM)

using standard procedures described in Newvile et al.

(1998), Sutton et al. (2008), Head et al. (2018), and Rao

et al. (2018).

RESULTS

The sulfur content of CRB is low (0.05

–

0.2 wt% S).

The sample charge consists of CRB dry powder mixed

with clumps of ferric sulfate in an

2:1 overall

proportion (Fig. 1a). The recovered charge from the 21

GPa experiment showed very few blebs whereas the

recovered sample from the 31 GPa experiment yielded

abundant clusters of S-bearing blebs in impact melt

pockets. The sulfur-bearing blebs in this experiment are

of two kinds: one variety is the “sulfide” blebs that are

usually round and have metallic luster and bright

reflectivity. The other type is the “sulfite” or “mixed

sulfite-sulfide” blebs, which appear dull in reflected light

and display a distinctly nonmetallic appearance. The

second type occurs in both rounded and elongated

shapes. Typical pictures of these blebs are shown in

Fig. 2 and the FE-SEM EDS spectra are given in Fig. 3.

31 GPa Shock Experiment (Caltech)

The texture of the recovered charge from the 31 GPa

shock experiment is nonuniform. Some areas near the

front half of the capsule, which experienced only one

shock wave passage and a peak pressure of 15 GPa

before arrival of a release wave, showed only a few

sulfur-bearing blebs situated in a large amount of

unmelted ferric sulfate and CRB particulates. Other

areas, toward the back half of the sample that

experienced reshock to 31 GPa, contain numerous

isolated pockets of material quenched from melt and are

filled with clusters of micrometer-sized sulfur-bearing

blebs (Fig. 2). The spatial distribution of molten material

in the sample suggests that the shock wave propagated

nonuniformly in the target charge, leading to extensive

melting in certain portions of the sample mix while other

areas remained unmelted. This is partly due to the

multiple shocks generated by the target chamber

geometry and also partly due to inhomogeneous

distribution of porosity in the starting material. A similar

result was obtained by Schaal et al. (1979), whose

experimental shock melting of granulated lunar basalt

75035 (45

–

m grain size and 28% porosity) at 50 GPa

(it is unclear if this is a hypothetical reverberation

pressure or a true peak pressure) transformed only

35%

of the parental material into impact melt glass. We

suspect that the small number of blebs found in the front

half of the sample was physically injected from the back

half along shear planes.

Fig. 2. Backscattered electron (BSE) images of impact melts in the 31 GPa experiment containing abundant sulfur-rich blebs

(whitish semicircular spots: (a) typical area consisting of numerous sulfide and sulfite blebs; (b) sulfite

–

sulfide bleb mixtures in

one area with at high magnification; (c) another area; note some high-contrast spots here are titanomagnetites.

Shock experiments

The sulfur blebs found in shock-melted parts of the

31 GPa experimental charge are of two types. Blebs of

rounded to elongated shape that display dull reflectivity

in optical light yield EDS spectra showing large peaks

for O, Fe, and S (Fig. 3a). We semiquantitatively

determined the Fe/S (atomic) ratios in these blebs and

they yielded values ranging from

1.02 to

1.3. Such

stoichiometry is inconsistent with ferric sulfate (Fe/S =

0.67), pyrite (Fe/S = 0.5), or pyrrhotite (Fe/S

0.92) but

is consistent with either ferrous sulfite or ferrous sulfide.

We found similar Fe/S values in the natural bleb

clusters in impact melt sample #507 in EETA79001,

Lith B. We tentatively interpret these blebs to be

Fe-sulfite (the starting sulfate material does not form

rounded blebs). More importantly, the S K-edge

XANES spectra of the sulfur blebs in this sample show

dominant peaks at 2478 eV, the energy associated with

sulfite species (Fleet et al., 2005) with only minor peaks

at 2470 eV, the energy typical of sulfide species (Fleet

et al., 2005), and no sulfate peak. The compositional

and spectroscopic results together indicate that these

blebs likely consist of ferrous sulfite, Fe (SO

presumably formed by decomposition and reduction of

ferric sulfate to ferrous sulfite. Sulfate melt readily

dissolves in and mixes with molten silicates in the shock

state, whereas sulfite melt becomes immiscible during

cooling and decompression, exsolving to form blebs

dispersed in silicate glass in the quenched product.

There is another group of rounded blebs in the 31

GPa recovered charge that display metallic luster and

high optical reflectivity. The EDS spectra of these blebs

show Fe and S peaks only (negligible O peaks) and

yield Fe/S (atomic) ratios of 0.9

–

0.93 (Fig. 3b). This

result suggests that these metallic blebs consist of

)

S (pyrrhotite-like) material. Lorand et al. (2005)

and Gattacceca et al. (2013) showed that the igneous

sulfides in shergottites yield Fe/S (atomic) ratios of 0.9

–

0.93. However, the CRB

ferric sulfate starting

material in this experiment contained no igneous

pyrrhotite. Instead, the presence of pyrrhotite-like blebs

indicates that decomposition and further reduction,

from sulfate through sulfite and all the way to sulfide,

occur during shock melting and pressure release.

The initial pore space in the 31 GPa experimental

charge contained CO that was introduced in order to

avoid the oxidizing power of terrestrial atmospheric O

In retrospect, this was a poor choice in that reduction

by CO offers an apparent explanation for our results

that would not apply to Mars. However, we argue that

CO played a negligible role as a reducing agent for the

Fe and S in the charge, on the basis of the following

mass balance. The porosity (or volume fraction of CO

gas) in the starting material was 28%. The density of

CO gas at the initial conditions (0.8 bar and 298 K) is

0.916 kg m

–

whereas the density of the solid (CRB

sulfate) starting mix is 2025 kg m

–

. This corresponds to

a mass fraction of CO gas of 0.00018 or 0.0063 moles

of oxygen acceptor capacity per kg of bulk sample.

Conservatively, assuming that we produced only ferrous

sulfite (i.e., neglecting further reduction to sulfide) by

reduction of ferric sulfate, we find that one could

produce at most 0.0031 moles of Fe(SO

) per kg of bulk

sample. This corresponds to a maximum mass fraction

yield of ferrous sulfite of 0.00043 (kg Fe sulfite/kg bulk

sample) or a volume fraction of ferrous sulfite blebs of

0.0006. The BSE images of our sample show an S-bleb

volume fraction much greater than 1 part in 10

. Hence,

the autoreduction mechanisms we discuss below are

required; if the Fe and S did react with the CO gas, this

could be at most a minor contribution to their

Fig. 3. Energy-dispersive X-ray (EDS) spectra of typical sulfur-rich blebs in the 31 GPa experiment: (a) EDS spectrum of a

typical sulfite bleb (spherical shape) formed from a immiscible fluid on melting of crystalline Fe-sulfate grains in CRB

sulfate

sample charge note the large oxygen peak. A similar peak with slightly higher peak height for oxygen was obtained for unmolten

ferric sulfate crystal in the sample mix. b) A typical sulfide bleb showing metallic luster, note the near-absence of detectable

oxygen. Typical EDS spectrum of Fe-sulfide blebs in the shocked sample is shown here.

M. N. Rao et al.

reduction. However, there remains the possibility that

even this small amount of CO might have somehow

nucleated or catalyzed the reduction process, which then

became self-supporting. Hence, this issue remains an

open question at this time. We intend to follow this

report with a further test: a shock experiment

replicating the 31 GPa shock at Caltech but with low-

pressure CO

in the pore space (without CO).

21 GPa Shock Experiment (JSC)

In the 21 GPa experiment, however, the areas

examined in the polished thin section prepared from the

recovered charge do not show any evidence of sulfide or

sulfite bleb formation as a result of shock melting and

reduction. However, there are several amorphous

regions suggesting incipient melting (Fig. 4a). These

amorphous areas display a typical cracking pattern

consistent with differential thermal contraction. Such

cracks are absent from the impact melt areas of the 31

GPa shock experiment; they may indicate that the local

hotspots in the 21 GPa experiment that underwent

incipient melting subsequently experienced larger

thermal contraction than colder surrounding areas,

resulting in tensile stresses. The dominant low-

temperature minerals present in the sample mix are

apatite (from the CRB powder) and the added hydrated

ferric sulfate. The EDS spectra of most of these

amorphous regions display the prominent peaks of Fe,

O, S, and P that indicate the presence of sulfate and

phosphate (Fig. 4b). Measurable peaks of Na, Mg, Al,

Si, K, and Ca are also present. These complex

multielement spectra do not fit the stoichiometry of

pure mineral phases; they are instead consistent with the

flexible chemical compositions of melts and glasses. We

suggest that apatite and iron sulfate preferentially

dissolved into a granitoid minimum melt formed by

low-degree incipient melting. Some amorphous regions

lack the prominent P peak; this is consistent with the

sparse distribution of apatite in the starting material

and the limited time available for mixing across the

charge. The low-degree melts quenched to amorphous

material and there is no indication of immiscible melt

separation in these regions; S-rich blebs are not

observed. The absence of observable reduction of

sulfate in this experiment may be attributed to the low

shock temperature achieved or the low degree of

melting reached. We also note the absence of CO gas in

the pore space of this sample, but thermodynamic

calculations showing reduction upon release of

chemically closed systems from high shock temperatures

(Sheffer, 2007) suggest that an external reductant is

probably not an essential prerequisite to sulfide

formation in these experiments and the above argument

from stoichiometry shows that the presence of CO in

the 31 GPa experiment does not by itself explain the

difference between the two outcomes. We tentatively

conclude that 21 GPa is insufficient peak shock pressure

to match the observations of shergottite impact melts.

DISCUSSION

Reduction During Shock Release

The process of reduction of Fe and S in a shock

melt that accompanies decompression and release from

the shock state is referred to as “reduction due to

isentropic cooling” (Melosh & Artemieva, 2004; Sheffer,

2007; Sheffer & Melosh, 2005). This type of Fe

reduction is known to occur during the formation of

Fig. 4. Incipiently molten region in the 21 GPa charge: (a) BSE image; (b) EDS spectrum showing O, P, S, Ca, and Fe peaks

attributed to melting of apatite and ferric sulfate, as well as Na, Mg, Al, Si, and K peaks attributed to a granitic minimum melt

component produced by incipient melting of the bulk mixture.

Shock experiments

lunar agglutinates due to micrometeorite impacts and

on Earth during the formation of tektites and

moldavites by asteroid impacts. Note that, in other

contexts, the recovery from a high-pressure shock state

is called “isentropic release.” Indeed, “isentropic

cooling” may sound somewhat contradictory, since

“cooling” suggests removal of heat, which implies a

decrease in entropy. However, isentropic pressure

release is indeed accompanied by decreasing

temperature, which is the sense in which “cooling” was

used by Melosh and Artemieva (2004). We retain this

usage from the cited references to describe the coupled

path of pressure and temperature decrease at roughly

constant entropy.

The complete thermodynamic path of a strong

shock event can be described in several stages. The

passage of a shock wave causes an irreversible jump in

pressure, temperature, and entropy. Strong shock

compression, especially in porous starting materials that

undergo large irreversible changes in volume, may result

in shock temperatures of several thousand Kelvin that

leads to melting and even incipient vaporization. After

some time in this shock state (for a supported shock),

pressure is typically released by the passage of a

rarefaction wave. Rarefactions are not shocks and they

are approximated as isentropic acoustic waves,

following constant entropy paths toward lower pressure

and somewhat lower temperature (Beck et al., 2007;

Gillet et al., 2007; Sharp & DeCarli, 2006). Vapor

production is likely during release, as the pressure drops

below the critical point while temperature remains

elevated compared to the starting condition (Davies

et al., 2020). Molecules in the high-temperature vapor

phase easily decompose or ionize and electron exchange

between the liquid and vapor phases may occur. Such

complex nonequilibrium chemical reactions at the

vapor

–

liquid interface continue until cooling reaches a

blocking temperature where exchange slows and the

phases become decoupled (Melosh & Artemieva, 2004;

Sheffer et al., 2006). This leaves the shock melt (at the

time it transforms into glass) with a different net

oxidation state than the starting material (Rao et al.,

2018). Even at equilibrium (equal oxygen fugacity in

both phases), the vapor becomes increasingly enriched

in oxygen relative to the liquid, leaving the liquid

progressively more oxygen depleted. Disequilibrium

enhances the effect. Continuous degassing as pressure

drops and volatile solubility decreases leads to

continuous increase in the intensity of oxygen transfer

to the vapor and reduction of the melt (Sheffer &

Melosh, 2005).

Evidently an open system process is necessary to

explain our experimental observation of the production

of ferrous sulfite and sulfide from initial ferric sulfate.

Both iron and sulfur are reduced during this process

and there is no obvious reducing agent that undergoes a

complementary oxidation (the amount of CO present in

the 31 GPa experiment is very small and not sufficient).

It is likely that the vapor phase acted as this reducing

agent, taking up oxygen from the melt while the melt

and vapor were open to mass and charge exchange. The

experimental results suggest that a similar auto-

reduction may plausibly be invoked to explain the

production of sulfide and sulfite in shergottite impact

melts from initially oxidized sulfur as Fe sulfate from a

Martian regolith component.

What Happens When Fe

x

)

S (Pyrrhotite) Is Shock

Heated at High Temperatures and Pressures by Impact in

Extraterrestrial Samples?

Although a small amount of FeS is present in the

CRB starting material, its mass fraction is insufficient to

be the source of the S-blebs in the 31 GPa run product.

Moreover, numerous observations of Martian

meteorites and other samples show that pyrrhotite

cannot be the source of the abundant sulfite

–

sulfide

blebs observed in shergottite impact melts. When

)

S (pyrrhotite) in reduced extraterrestrial targets is

shock heated, it is likely to dissociate into Fe

and S

in the impact melt. If the release path follows reducing

conditions, the Fe

will be reduced to np (nanophase)

(metal), as observed in lunar agglutinates. However,

the sulfur is already fully reduced. In the form of sulfide

ion (S

), sulfur is relatively immobile (Wilke et al.,

2008), has low vapor pressure, and is retained by the

melt. However, sulfide may react with cations to form

solid sulfide species (Rao et al., 2018), including FeS

(troilite), Fe

)

S (pyrrhotite), or FeS

(pyrite),

depending on sulfur fugacity (as found, e.g., in Tissint;

see Gattacceca et al., 2013; Rao et al., 2018). Recently,

Ohtaki et al. (2019) studied the np Fe

particles in lunar

soils using high-resolution secondary ion mass

spectrometry (nano-SIMS) and transmission electron

microscopy (TEM) and found that a significant number

of the np Fe

particles are rimmed by FeS. Ohtaki et al.

(2019) concluded that these objects formed by impact

melting of pyrrhotite in a reducing environment. Ishii

et al. (2008) found similar particles having nm-thick

sulfide rims surrounding 10

–

15 nm np Fe

metal cores

in returned dust samples from comet 81P/Wild 2.

Moreover, experimental hypervelocity implantation of

nearly spherical micrometer-sized pyrrhotite particles

into an aerogel target at 6 km s

resulted in abundant

nm-scale sulfide-rimmed Fe metal particles similar to

those observed in Stardust material (Ishii et al., 2008).

The spherical shape suggests that these particles are

derived from immiscible FeS melt droplets that partly

M. N. Rao et al.

decomposed and dispersed before quenching with an

)

S rim (Noguchi et al., 2014). The formation of

“swirl” patterns defined by trails of np Fe

particles

inside the Stardust sample (GEMS) suggests that the

glassy spherule remained molten long enough for some

internal mixing to occur but cooled fast enough to

quench to a reduced glass (Ohtaki et al., 2019). A

similar “swirl” pattern of sulfide blebs was found in a

Tissint impact melt pocket (Fig. 5). The observations

indicate that when Fe

)

S (pyrrhotite) is shock heated

to high temperatures and pressures during impact, it is

likely to produce clusters of np Fe

(metal) particles by

isentropic cooling as well as dissolved sulfide that may

in turn reach saturation and exsolve as sulfur-rich melts

or precipitate a range of sulfide minerals. This is a

different paragenesis from the S-rich blebs found in

shergottite impact melts; the assemblage of np Fe

(metal) blebs has not been observed in any gas-rich

impact melts studied in shergottites.

What Happens When Ferric Sulfate Is Shock Melted at

High Temperatures and Pressures by Meteoroid Impact

on Mars?

Ferric sulfate is found in Martian soils. If this

material was gardened into the Martian regolith and was

trapped in pores and voids at the site of the shergottite-

forming impact(s), it would be shocked to high pressures

(numerous shock indicators in shergottites are interpreted

to require peak pressures of about 24 GPa; Beck et al.,

2005; El Goresy et al., 2013; Gillet et al., 2007;

Langenhorst & Poirer, 2000; Sharp & DeCarli, 2006;

Shaw & Walton, 2013) and anomalously high local

temperatures (

≥

2000

C) due to pore collapse.

Thermodynamic calculations (Sheffer, 2007) show that,

under these conditions, ferric sulfate dissolves into

silicate-dominated impact melt and dissociates into ferric

iron and sulfate ion. The cooling and expansion

associated with release from the shock state promote

reduction to ferrous iron and sulfite ion. These species

may recombine to form iron sulfite (Hamann et al., 2018;

€

orz et al., 2020). Or, depending on the intensity of the

shock wave, the amount of vapor production, and the

chemical environment, the sulfite may be further reduced

to sulfide. Later, the cooling of sulfur-rich impact melt

will cross the sulfide saturation point and exsolve droplets

of immiscible FeS melt. These may form FeS blebs or

may perhaps backreact with dissolved sulfite in the melt

to form “mixed sulfite

–

sulfide” blebs in the quenched

impact melts (Hamann et al., 2018; H

€

orz et al., 2020).

There is insufficient reducing potential to take the Fe all

the way down to Fe

Fig. 5. An example of reduced sulfur-bearing blebs in impact melt in a shergottite: “swirl” of sulfide

–

sulfite blebs in Tissint

impact melt. Similar structures are also found in impact melts recovered in the 31 GPa Caltech experiment.

Shock experiments

We determined the iron oxidation state in impact

melts #506 and #507 in EETA79001 using Fe K-edge

micro-XANES (Rao et al., 2018; Sutton et al., 2008)

(see Fig. 6). Each XANES acquisition was background-

subtracted and fitted to obtain an estimate of the mean

oxidation state of the Fe atoms in the activation

volume. An optical micrograph of the area studied in

#506 is shown in fig. 11A of Rao et al. (2018). Two

line scans were performed; Fig. 6a shows the fitted

mean Fe oxidation state as a function of position along

line scan 1. The mean Fe oxidation state is relatively

uniform from point to point, with an average for the

line of 2.06

0.02. Line scan 2 (not plotted here)

appears very similar, with average mean Fe oxidation

state of 2.11

0.03. The simplest interpretation of these

mean values is that Fe in glass #506 exists as a mix of

Fig. 6. Photomicrograph indicating the position of the line scan across IM# 507. 507 area: yellow arrows point to S XANES

analyzed spots. Blue box is the S map region (2idb1_0053.mda). Blue arrow is Fe XANES line scan (EET507_line_xanes.002).

Point 17 of the line scan is near S spot 2. Spot 1 is EET507_sulfide1_xanes.001

M. N. Rao et al.

and Fe

, with not more than a few percent Fe

Likewise, line scan results for mean Fe oxidation state

along the line across IM #507 are shown in Fig. 6b as

indicated on the photomicrograph in Fig. 6c.

Furthermore, the Fe oxidation state is uniform here

within error and the average along the line is 2.07

0.03. Notably, these XANES results agree well with the

/Fe

ratio determined in another large impact melt

(#27) in EETA79001 using M

€

ossbauer spectroscopy by

Solberg and Burns (1989). Although a mean Fe

oxidation state above 2.0 does not explicitly demand

the absence of metallic Fe, especially in a rapidly

quenched disequilibrium assemblage, in fact both IM

pools #506 and #507 have been further examined at

high resolution with FE-SEM instruments at JSC

(Houston) as well as Max Planck Institute (Mainz,

Germany) for Fe (metal) particles (Rao et al., 2018).

Imaging results revealed no Fe

(metal) grains in these

impact melts.

Relevance of the Shock Experimental Results to the

Sulfur-Bearing Blebs Found in Large Impact Melts in

Shergottites: A Model Perspective

The model outline presented here is an extension of

the earlier model discussed in Rao et al. (2018). Here,

we recall a few salient features in that model. The

shergottite source region (provenance) on Mars

apparently consists of two different types of rock

complexes. The first one hosts olivine

–

phyric rocks

whose major igneous minerals are olivine, pyroxene,

and feldspar from which shergottites such as Tissint,

DaG 476, and EETA79001, Lith A, were launched by

meteoroid impact. The second region consists of olivine-

free pyroxene

–

phyric rocks (having feldspars and

pyroxenes as major minerals), from which meteorites

such as Shergotty, Zagami, Los Angeles, and

EETA79001, Lith B, were derived. All the shergottite

source rocks were apparently located at a relatively

shallow depth (Artemieva & Ivanov, 2004; Head et al.,

2002; Melosh, 1984; Nyquist, 1983) but were pervasively

buried or mantled by Martin soil and dust prior to

impact launch.

The large overabundance of sulfur (

8% as SO

)

observed in Martian soils is usually attributed to

chemical alteration caused by the interaction of sulfate-

rich acidic fluids with Martian rock and dust (Br

€

uckner

et al., 2008; Clark et al., 2005; King & McLennan,

2010; McAdam et al., 2014; Tosca et al., 2004) as

discussed in Rao et al. (2018). In regions with low

water-to-rock ratios, the pH of the solutions remained

highly acidic (pH

–

1) whereas, in other areas with

higher water-to-rock ratios, the pH of these solutions

became progressively less acidic (pH

–

5; Banin et al.,

1997; Clark et al., 2005; Elwood Madden et al., 2004;

Hurowitz et al., 2006; McSween et al., 2008; Settle,

1979).

We consider the case where highly acidic solutions

(pH

–

1) percolated through rocks and soils near the

shergottite provenance on Mars. These solutions

initiate the rapid dissolution of soluble mineral phases,

leaving behind relatively unaltered refractory minerals

selectively unaffected (Hurowitz et al., 2006; King &

McSween, 2005; McCollom, Hynek, et al., 2013;

McCollom, Robbins, et al., 2013; Tosca et al., 2004).

The primary evolution of the fluid composition

becomes dominated by the chemistry of the mineral

phases most susceptible to acid dissolution (Hurowitz

et al., 2006). In the case of the olivine

–

phyric rocks,

the most soluble mineral is olivine, which contributes

Fe and Mg to the solution. The solution is

progressively neutralized by such mineral reactions. If

the pH rises to between 1 and 2, Mg-sulfate will be

highly soluble and can be transported in dissolved

form from the reaction sites. Fe

may be oxidized to

and precipitated as a highly insoluble ferric

hydroxysulfate minerals such as jarosite. As a result,

ferric sulfate becomes decoupled from Mg-sulfate and

other soluble phases (e.g., chlorides) in secondary

mineral deposits. These findings are consistent with the

demonstrated absence of Mg chlorides in impact melts

#506 and #507 in EETA79001 studied by Rao et al.

(2008). Furthermore, we note that Mg- and Ca-sulfates

at high shock temperatures and pressures produce

decomposition products (sulfide/sulfite or oxide), which

do not yield immiscible fluids leading to bleb cluster

formation.

As noted in Rao et al. (2018), the secondary

sulfates thus produced on Mars were likely mobilized

into cracks and voids of the host rock shergottites

(prior to impact) by aeolian activity or soil erosion

(Greeley et al., 2004; Meyer, 2012). Subsequently, when

these deposits were shock heated to high temperatures

and pressures during impact launching of the

consolidated surface material assemblage, the

constituent sulfates were decomposed and were reduced

to sulfide/sulfite

—

mixes in large impact melts in

shergottite melts due to isentropic cooling. The

immiscible fluids thus resulted yield the observed bleb

clusters upon quenching.

SUMMARY AND CONCLUSIONS

1. Although numerous studies have shown that shock

compression and release can convert FeS pyrrhotite

via fluid immiscibility to clusters of nanophase

metallic Fe

nanoparticles in impact melts, no such

metallic particles were found during detailed

Shock experiments

examination by FE-SEM and Fe K-edge XANES

(Fig. 7) of impact melts from shergottites. This

argues against the common igneous sulfides in

shergottites being the source of sulfur for

micrometer-sized bleb clusters found in these impact

melts.

2. Shocking a porous mixture of ferric sulfate and

CRB powder at 21 GPa produced no sulfide blebs.

Instead, amorphous regions marked by thermal

contraction cracks were found that contain elevated

concentrations of phosphorus, iron, and sulfur in a

granitoid minimum melt-like composition suggesting

incipient melting of the silicates, ferric sulfate, and

apatite nuggets.

3. A similar target subjected to higher shock

pressures, that is, 31 GPa, yielded numerous areas

of quenched impact melt containing abundant

clusters of micrometer-sized blebs of Fe-sulfide and

Fe-sulfite mixtures. Although porosity in this

experiment contained a small amount of CO gas,

there was not enough CO in the capsule to act a

significant reducing agent. Rather, we interpret this

result as experimental evidence of the “reduction

by isentropic cooling” hypothesis, which allows the

melt phase to act as an open system and

experience net reduction during vapor-present

expansion from a high-temperature, high-pressure

shock state. Ferric sulfate dissolves in the impact

melt phase and then is reduced first to ferrous

sulfite and subsequently, in part, to ferrous sulfide.

Immiscibility of sulfite and sulfide melts in silicate

melt then leads to dispersion of

m-sized droplets

dominated by both these components. We conclude

that Fe-sulfate in porous, pyroxene

–

phyric Martian

target rocks was most likely the source for sulfur

in the metal-free Fe-sulfite and Fe-sulfide blebs

observed by FE-SEM, S K-edge XANES (see Rao

et al., 2018), and Fe K-edge XANES (Fig. 7) in

shergottite impact melts such as #507 in

EETA79001, Lith B as well as #784 in Los

Angeles and DBS #1 and #2 in Shergotty. To

conclusively address the issue of whether the small

amount of CO initially present in the pore space

acted as a trigger for the production of the

observed sulfide bleb clusters in the recovered

charge, we are planning to repeat the 31 GPa

experiment in a low-pressure CO

atmosphere

without CO presence.

4. We describe an outline of the extended model

discussed in Rao et al. (2018) where it is noted that

low pH acid-sulfate solutions presumably percolated

through and reacted with olivine

–

phyric and

pyroxene

–

phyric basaltic rocks in the shergottite

source region producing secondary sulfates in a

manner similar to that inferred from direct rover

measurements at Meridiani, Gusev, and Gale

craters. These sulfate assemblages were later

mobilized into host rock voids and crevices by

aeolian activity/soil erosion. During impact

launching, they were shock melted at high

temperature and pressures and the sulfate in the

melts was reduced to sulfides by isentropic cooling

(Melosh & Artemieva, 2004; Sheffer & Melosh,

2005).

Fig. 7. Fe K-edge XANES studies of impact melts in EETA79001.a) The average Fe oxidation state for line scan 1 in IM #506

is 2.06

0.02 (Rao et al., 2018).b) The average Fe oxidation state for a line scan across IM# 507 is 2.07

0.03.

M. N. Rao et al.

Acknowledgments

—We thank Don Bogard, Fred H

€

orz,

Mark Cintala, Abhijit Basu, and Jay Melosh for

valuable discussions on different aspects related to this

work. We thank Frank Cardenaes and Roland Montes

for carrying out 21 GPa shock experiment at EIL,

Johnson Space Center (under the supervision of Mark

Cintala). We thank Michael Long for carrying out the

31 GPa experiment at the shock wave lab at Caltech

which is supported by the NSF awards 1829277 and

1725349 to Paul Asimow. We thank Joachim Huth for

FE SEM measurements on our PTS samples at Max-

Planck Institute fur Chemie, Mainz (Germany). This

work is partly supported by NASA MFRP grant No.

08-0069 to Larry Nyquist at JSC. We thank Anita Rao

of Qualcomm (San Diego, CA) for valuable

computational assistance during this work. MNR is

grateful to Alexander von Humboldt Stiftung

(Germany) for partially supporting his presentation of

the preliminary results obtained in this study at

METSOC Meeting in Berlin in 2016. We are grateful to

the Editor, Timothy Jull and the Associate Editor,

Natalia Artemieva for efficient handling of our

manuscript and to reviewers Christopher Hamann and

Vincent Chevrier for valuable comments and

suggestions that led to a significant improvement of the

paper.

Data Availability Statement

—Data openly available in a

public repository that issues datasets with DOIs.

Editorial Handling

—Dr. Natalia Artemieva

REFERENCES

Akaogi, M., Haraguchi, M., Nakanishi, K., Ajiro, H., and

Kojitani, H. 2010. High-Pressure Phase Relations in the

System CaAl

-NaAl

with Implications for

Na-Rich CAS Phase in the Shocked Martian Meteorites.

Earth and Planetary Science Letters

289: 503

–

Artemieva, N., and Ivanov, B. A. 2004. Launch of Martian

Meteorites in Oblique Impacts.

Icarus

171: 84

–

101.

Banin, A., Han, F. X., Kan, I., and Cicelsky, A. 1997. Acidic

Volatiles and the Mars Soil.

Journal of Geophysical

Research

102: 13341

–

56.

Basu, A. 2005. Nanophase Fe

in Lunar Soils.

Journal of

Earth System Science

114: 375

–

80.

Beck, P., Ferroir, T., and Gillet, P. 2007. Shock

—

Induced

Compaction, Melting and Entrapment of Atmospheric

Gases in Martian Meteorites.

Geophysical Research Letters

34: L01203. https://doi.org/10.1029/2006GL028141.

Beck, P., Gillet, P. H., El Goresy, A., and Mostefaoui, S.

2005. Timescales of Shock Processes in Chondritic and

Martian Meteorites.

Nature

435: 1071

–

4. https://doi.org/

10.1038/nature03616.

Becker, R. H., and Pepin, R. O. 1984. The Case for a Martian

Origin of Shergottites: Nitrogen and Noble Gases in EET

79001.

Earth and Planetary Science Letters

69: 225

–

42.

Bibring, J. P., Langevin, Y., Mustard, J. F., Poulet, F.,

Arvidson, R., Gendrin, A., and Gondet, B. et al. 2007.

Coupled Ferric Oxides and Sulfates on the Martian

Surface.

Science

317: 1206

–

10. https://doi.org/10.1126/

science.l144174.

Bogard, D. D., and Johnson, P. 1983. Martian Gases in an

Antarctic Meteorite?

Science

221: 651

–

Bowling, T. J., Johnson, B. C., Wiggins, S. E., Walton, E. L.,

Melosh, H. J., and Sharp, T. G. 2020. Dwell Time at High

Pressure of Meteorites During Impact Ejection from Mars.

Icarus

343: 113689.

€

uckner, J., Dreibus, G., Gellert, R., Squyres, S. W., W

~

anke,

H., Yen, A., and Zipfel, J. 2008. Mars Exploration

Rovers: Chemical Composition by the APXS. In

The

Martian Surface

—

Composition, Mineralogy and Physical

Properties

, edited by J. F. Bell III, 58

–

102. Cambridge:

Cambridge University Press.

Chen, G. Q., Tyburczy, J. A., and Ahrens, T. J. 1994. Shock-

Induced Devolatilization of Calcium-Sulfate and

Implications for K-T Extinctions.

Earth and Planetary

Science Letters

128: 615

–

28.

Chen, M., Sharp, T. G., El Goresy, A., Wopenka, B., and

Xie, X. 1996. The Majorite-Pyrope

Magnesiowustite

Assemblage: Constraints on the Shock History of

Chondrites.

Science

271: 1570

–

Clark, B. C., Morris, R. V., McLennan, S. M., Gellert, R.,

Jolliff, B., Knoll, A. H., Squyres, S. W. et al. 2005.

Chemistry and Mineralogy of Outcrops at Meridiani

Planum.

Earth and Planetary Science Letters

240: 73

–

94.

Davies, E. J., Carter, P. J., Root, S., Kraus, R. G., Spaulding,

D. K., Stewart, S. T., and Jacobsen, S. B. 2020. Silicate

Melting and Vaporization During Rocky Planet

Formation.

Journal of Geophysical Research: Planets

125:

e2019JE006227. https://doi.org/10.1029/2019JE006227.

El Goresy, A., Gillet, P. H., Miyahara, M., Ohtani, E.,

Ozawa, S., Beck, P., and Montagnac, G. 2013. Shock-

Induced Deformation of Shergottites: Shock Pressures and

Perturbations of Magmatic Ages on Mars.

Geochimica et

Cosmochimica Acta

101: 233

–

62.

Elwood Madden, M. E., Bodnar, R. J., and Rimstidt, J. D.

2004. Jarosite as an Indicator of Water-Limited Chemical

Weathering on Mars.

Nature

431: 821

–

Fleet, M. E., Liu, X., Harms, S. L., and King, P. L. 2005.

Sulfur K-Edge XANES Spectroscopy: Chemical State and

Content of Sulfur in Silicate Glasses.

Canadian

Mineralogist

43: 1605

–

18.

Gattacceca, J., Rochette, P., Cournede, C., Hewins, R.,

Sautter, V., Scorzelli, R., Uehara, M., and Zanda, B. 2013.

Opaque

Minerals,

Magnetic

Properties

and

Paleomagnetism of the Tissint Martian Meteorite.

Meteoritics & Planetary Science

48: 1919

–

36.

Gellert, R., Rieder, R., Anderson, R. C., Br

€

uckner, J., Clark,

B. C., Dreibus, G., Economou, T. et al. 2004. Chemistry

of Rocks and Soils in Gusev Crater from the Alpha-

Particle X-Ray Spectrometer.

Science

305: 829

–

32.

Gillet, P., Chen, M., Dubrovinsky, L., and El Goresy, A.

2000. Natural NaAlSi

-Hollandite in the Shocked

Sixiangkou Meteorite.

Science

287: 1633

–

Gillet, P. and El Goresy, A. 2013. Shock Events in the Solar

System: The Message from Minerals in Terrestrial Planets

and Asteroids.

Annual Review of Earth and Planetary

Sciences

41: 257

–

85.

Gillet, P., El Goresy, A., Beck, P., and Chen, M. 2007. High-

Pressure Mineral Assemblages in Shocked Meteorites and

Shock experiments

Shocked Terrestrial Rocks: Mechanisms of Phase

Transformations and Constraints to Pressure and

Temperature Histories. In

Advances in High Pressure

Mineralogy

, edited by E. Ohtani. Geological Society of

America, Special Paper 421: 57

–

82.

Greeley, R., Squyres, S. W., Arvidson, R. E., Bartlett, P., Bell,

J. F., Blaney, D., Cabrol, N. A., Farmer, J., Farrand, B.,

and Golombek, M. P. 2004. Wind-Related Processes

Detected by the Spirit Rover at the Gusev Crater Mars.

Science

305: 810

–

3. https://doi.org/10.1126/science.

1100108.

Gupta, S. C., Ahrens, T. J., and Yang, W. 2001. Shock-

Induced Vaporization of Anhydrite and Global Cooling

from the K/T Impact.

Earth and Planetary Science Letters

188: 99

–

412.

Hamann, C., Fazio, A., Ebert, M., Hecht, L., Wirth, R.,

Folco, L., Deutsch, A., and Reimold, W. U. 2018. Silicate

Liquid Immiscibility in Impact Melts.

Meteoritics &

Planetary Science

53: 1594

–

632. https://doi.org/10.1111/

maps.12907.

Head, E., Lanzirotti, A., Newville, M., and Sutton, S. R.

2018. Vanadium, Sulfur and Iron Valencies in Melt

Inclusions as a Window into Magmatic Processes: A Case

Study at Nyanuragira Volcano, Africa.

Geochimica et

Cosmochimica Acta

226: 149

–

73. https://doi.org/10.1016/j.

gca.2018.01.033.

Head, J. N., Melosh, H. J., and Ivanov, B. A. 2002. Martian

Meteorite Launch: High Speed Ejecta from Small Craters.

Science

298: 1752

–

Hidaka, H., Yoneda, S., and Nishiizumi, K. 2009. Cosmic-

Ray Exposure Histories of Martian Meteorites Studied

from Neutron Capture Reactions of Sm and Gd Isotopes.

Earth and Planetary Science Letters

288: 564

–

71.

€

orz, F., Cintala, M. J., Thomas-Keprta, K. L., Ross, D. K.,

and Clemett, S. J. 2020. Unconfined Shock Experiments:

A Pilot Study into the Sock-Induced Melting and

Devolatilization of Calcite.

Meteoritics & Planetary

Science

55: 102

–

29.

Hurowitz, J. A., McLennan, S. M., Tosca, N. J., Arvidson, R.

E., Michalski, J. R., Ming, D. W., Schroder, C., and

Squyres, S. W. 2006. In Situ and Experimental Evidence

for Acid Weathering of Rocks and Soils on Mars.

Journal

of Geophysical Research

111: E02S19. https://doi.org/10.

1029/2005JE002515.

Ishii, A. H., Bradley, J. P., Dai, Z. R., Chi, M., Kearsley, A.

T., Burchell, M. J., Browning, N. D., and Molster, F.

2008. Comparison of Comet 81P/Wild 2 Dust with

Interplanetary Dust from Comets.

Science

319: 447

–

50.

Ivanov, B. A., and Deutsch, A. 2002. The Phase Diagram of

CaCO

in Relation to Shock Compression and

Decomposition.

Physics of the Earth and Planetary

Interiors

129: 131

–

43.

Keller, L. P., and Berger, E. L. 2014. A Transmission Electron

Microscope Study of Itokawa Regolith Grains.

Earth,

Planets and Space

66: 1

–

Keller, L. P., and McKay, D. S. 1997. The Nature and Origin

of Rims on Lunar Soil Grains.

Geochimica et

Cosmochimica Acta

61: 2331

–

41.

King, P. L., and McLennan, S. M. 2010. Sulfur on Mars.

Elements

6: 107

–

12.

King, P. L., and McSween, H. Y. 2005. Effects of H

O, pH,

and Oxidation State on the Stability of Fe Minerals on

Mars.

Journal of Geophysical Research

110: E12S10.

https://doi.org/10.1029/2005JE002482.

Langenhorst, F., and Poirier, J. P. 2000. Anatomy of Black

Veins in Zagami: Clues to the Formation of High-Pressure

Phases.

Earth and Planetary Science Letters

184: 37

–

55.

Lorand, J. P., Chevrier, V., and Sautter, V. 2005. Sulfide

Mineralogy and Redox Conditions in Some Shergottites.

Meteoritics & Planetary Science

40: 1257

–

72.

Malavergne, V., Guyot, F., Benzerara, K., and Martinez, I.

2001. Description of New Shock-Induced Phases in the

Shergotty, Zagami, Nakhla and Chassigny Meteorites.

Meteoritics & Planetary Science

36: 1297

–

305.

McAdam, A. C., Franz, H. B., Sutter, B., Archer, P. D.,

Freissinet, C., Eigenbrode, J. L., Ming, D. W. et al.

2014. Sulfur-Bearing Phases Detected by Evolved Gas

Analysis of the Rocknest Aeolian Deposit, Gale Crater,

Mars.

Journal of Geophysical Research: Planets

119: 373

–

93.

McCollom, T. M., Hynek, B. M., Rogers, K., Moskowitz, B.,

and Berqu

o, T. S. 2013. Chemical and Mineralogical

Trends During Acid-Sulfate Alteration of Pyroclastic

Basalt at Cerro Negro Volcano and Implications for Early

Mars.

Journal of Geophysical Research: Planets

118: 1719

–

51.

McCollom, T. M., Robbins, M., Moskowitz, B., Berqu

o, T.

S., J

€

ons, N., and Hynek, B. M. 2013. Experimental Study

of Acid-Sulfate Alteration of Basalt and Implications for

Sulfate Deposits on Mars.

Journal of Geophysical

Research: Planets

118: 577

–

614.

McLennan, S. M. 2012. Geochemistry of Sedimentary

Processes on Mars. In

Mars Sedimentology

, edited by J. P.

Grotzinger and R. E. Milliken, 119

–

38. New York: SEPM

Special Publication.

McSween Jr., H. Y., and Jarosewich, E. 1983. Petrogenesis of

the Elephant Moraine A79001 Meteorite: Multiple Magma

Pulses on the Shergottite Parent Body.

Geochimica et

Cosmochimica Acta

47: 1501

–

13.

McSween, H. Y., Ruff, S. W., Morris, R. V., Gellert, R.,

Klingelhoffer, G., Christensen, P. R., McCoy, T. J. et al.

2008. Mineralogy of Volcanic Rocks in Gusev Crater,

Mars: Reconciling Mossbauer, Alpha Particle X-Ray

Spectrometer and Miniature Thermal Emission

Spectrometer Spectra.

Journal of Geophysical Research

113: E06S04.

Melosh, H. J. 1984. Impact Ejection, Spallation and the

Origin of Meteorites.

Icarus

59: 234

–

60.

Melosh, H. J., and Artemieva, N. 2004. How Does Tektite

Glass Lose its Water? (abstract #1723). 35th Lunar and

Planetary Science Conference. CD-ROM.

Meyer, C. 2012. Mars Meteorite Compendium. Report 27672,

–

175. Houston, Texas: Johnson Space Center.

Miyahara, M., Ohtani, E., Ozawa, S., Kimura, M., El Goresy,

A., Sakai, T., Nagase, T., Hiraga, K., Hirao, N., and

Ohishi, Y. 2011. Natural Dissociation of Olivine to (Mg,

Fe)SiO

Perovskite and Magnesiow

€

ustite in a Shocked

Martian Meteorite.

Proceedings of the National Academy

of Sciences

108: 5999

–

6003.

Morris, R. V., Klingelhoffer, G., Schroder, C., Rodionov, D.

S., Yen, A., Ming, D. W., & De Souza Jr., P. A. 2006.

Mossbauer Mineralogy of Rock, Soil and Dust at Gusev

Crater Mars: Spirit’s Journey Through Weakly Altered

Olivine Basalt on the Plains and Pervasively Altered Basalt

in the Columbia Hills.

Journal of Geophysical Research

111: E02S13. https://doi.org/10.1029/2005JE002584.

Newvile, M., Sutton, S. R., Rivers, M., and Eng, P. 1998.

Micro-Beam X-Ray Absorption and Fluorescence

M. N. Rao et al.

Spectroscopies at GSECARS: APS Beam Line 13ID.

Journal of Synchrotron Radiation

6: 353

–

Noguchi, T., Kimura, M., Hashimoto, T., Konno, M.,

Nakamura, T., Zolensky, M. E., Okazaki, R. et al. 2014.

Space Weathered Rims Found on the Surfaces of Itokawa

Dust Particles.

Meteoritics & Planetary Science

49: 188

–

214.

Nyquist, L. E. 1983. Do Oblique Impacts Produce Martian

Meteorites? Proceedings, 13th Lunar and Planetary Science

Conference.

Journal of Geophysical Research

88: A785

–

98.

Ohtaki, K. K., Bradley, J. P., Gillis-Davis, J. J., Taylor, G. J.,

and Ishii, H. A. 2019. Formation of Iron-Rich Nano-

Particles Within Glassy Silicate Spherules in Lunar Soil

(abstract #2156). 50th Lunar and Planetary Science

Conference. CD-ROM.

Rao, M. N., Bogard, D. D., Nyquist, L. E., McKay, D. S.,

and Masarik, J. 2002. Neutron Capture Isotopes in the

Martian Regolith and Implications for Martian

Atmospheric Noble Gases.

Icarus

156: 352

–

72.

Rao, M. N., Borg, L. E., McKay, D. S., and Wentworth, S. J.

1999. Martian Soil Component in Impact Glasses in a

Martian Meteorite.

Geophysical Research Letters

26:

3265

–

Rao, M. N., Nyquist, L. E., Bogard, D. D., Garrison, D. H.,

Sutton, S. R., Michel, R., Reedy, R. C., and Leya, I. 2011.

Isotopic Evidence for a Martian Regolith Component in

Shergottite Meteorites.

Journal of Geophysical Research

116: E08006. https://doi.org/10.1029/2010JE003764.

Rao, M. N., Nyquist, L. E., Ross, D. K., Sutton, S. R.,

Hoppe, P., Shih, C.-Y., Wentworth, S. J., and Garrison,

D. H. 2018. Signatures of Martian Regolith Components

Entrained in Some Impact-Melt Glasses in Shergottites.

Meteoritics & Planetary Science

53: 2558

–

82. https://doi.

org/10.1111/maps13177.

Rao, M. N., Nyquist, L. E., Ross, D. K., and Wentworth, S.

J. 2019. Acid-Sulfate Weathering Environment at the

Shergottite Provenance on Mars (abstract #1950). 50th

Lunar and Planetary Science Conference. CD-ROM.

Rao, M. N., Nyquist, L. E., Wentworth, S. J., Sutton, S. R.,

and Garrison, D. H. 2008. The Nature of Martian Fluids

Based on Mobile Element Studies in Salt-Assemblages

from Martian Meteorites.

Journal of Geophysical Research

113: https://doi.org/10.1029/2007JE002958.

Ross, D. K., Ito, M., Hervig, R., Rao, M. N., and Nyquist, L.

E. 2011. Recognizing the Effects of Terrestrial

Contamination on D/H Ratios in Shergottite Phosphates

(Abstract #1920). 42nd Lunar and Planetary Science

Conference. CD-ROM.

Schaal, R. B., H

€

orz, F., Thompson, T. D., and Bauer, J. F.

1979. Shock Metamorphism of Granulated Lunar Basalt.

Proceedings, 10th Lunar and Planetary Science Conference

pp. 2547

–

71.

Schrader, C. M., Cohen, B. A., Donovan, J. J., and Vicenzi,

E. P. 2016. Ni/S/Cl systematics and the Origin of Impact-

Melt Glasses in Martian Meteorite Elephant Moraine

79001.

Meteoritics & Planetary Science

51: 663

–

80.

See, T. H., Horz, F., and Cintala, M. J. 2001. Experimentally

Produced Shock Melts in Impact-Comminuted Powders of

an L6 Ordinary Chondrite (abstract #1807). 32nd Lunar

and Planetary Science Conference. CD-ROM.

Settle, M. 1979. Formation and Deposition of Volcanic

Sulfate Aerosols on Mars.

Journal of Geophysical Research

84: 8343

–

54.

Sharp, T. G., and DeCarli, P. G. 2006. Shock Effects in

Meteorites. In

Meteorites and the Early Solar System II

edited by H. Y. McSween and D. S. Lauretta, 653

–

77.

Tucson, Arizona: The University of Arizona Press.

Shaw, C. S. J., and Walton, E. 2013. Thermal Modelling of

Shock Melts in Martian Meteorites: Implications for

Preserving Martian Atmospheric Signatures and

Crystallization of High-Pressure Minerals from Shock

Melts.

Meteoritics & Planetary Science

48: 758

–

70.

Sheffer, A. 2007. Chemical Reduction of Silicates by Meteorite

Impacts and Lightning Strikes. PhD Dissertation,

Department of Planetary Sciences, University of Arizona,

Tucson, Arizona.

Sheffer, A., Dyar, M. D., and Sklute, E. C. 2006. Lightning

Strike Glasses as an Analog for Impact Glasses:

Mossbauer Spectroscopy of Fulgurites (abstract #2009).

35th Lunar and Planetary Science Conference. CD-ROM.

Sheffer, A., and Melosh, H. J. 2005. Why Moldavites Are

Reduced (abstract #1468). 36th Lunar and Planetary

Science Conference. CD-ROM.

Solberg, T. C., and Burns, R. G. 1989. M

€

ossbauer Spectral

Study of Weathered Antarctic and SNC Meteorites.

Proceedings, 19th Lunar and Planetary Science

Conference, pp. 313

–

22.

Sutton, S. R., Rao, M. N., and Nyquist, L. E. 2008. Sulfur

and Iron Speciation in Gas-Rich Impact-Melt Glasses

from Basaltic Shergottites Determined by microXANES

(abstract #1961). 39th Lunar and Planetary Science

Conference. CD-ROM.

Swindle, T. D., Caffee, M. W., and Hohenberg, C. M. 1986.

Xenon and Other Noble Gases in Shergottites.

Geochimica

et Cosmochimica Acta

50: 1001

–

15.

Tomioka, N., and Miyahara, M. 2017. High Pressure Minerals

in Shocked Meteorites.

Meteoritics & Planetary Science

52:

2017

–

39.

Tosca, N. J., McLennan, S. M., Lindsley, D. H., and

Schoonen, M. A. A. 2004. Acid-Sulfate Weathering of

Synthetic Martian Basalt: The Acid Fog Model Revisited.

Journal of Geophysical Research

109: E05003. https://doi.

org/10.1029/2003JE002218.

Wentworth, S. J., Gibson, E. K., Velbel, M. A., and McKay,

D. S. 2005. Antarctic Dry Valleys and Indigenous

Weathering in Mars Meteorites: Implications for Water

and Life on Mars.

Icarus

174: 383

–

95.

Wiens, R. C. 1988. Noble Gases Released by Vacuum

Crushing of EET79001 Glass.

Earth and Planetary Science

Letters

91: 55

–

65.

Wilke, M., Jugo, P. J., Klimm, K., Susini, J., Botcharnikov,

R., Kohn, S. C., and Janousch, M. 2008. The Origin of

Detected in Silicate Glasses by XANES.

American

Mineralogist

93: 235

–

40.

Yagi, T., Suzuki, T., and Akaogi, M. 1994. High Pressure

Transitions in the System KAlSi

NaAlSi

Physics

and Chemistry of Minerals

21: 12

–

Zhang, F., and Sekine, T. 2007. Impact-Shock Behavior of

Mg-and Ca-Sulfates and Their Hydrates.

Geochimica et

Cosmochimica Acta

71: 4125

–

33.

Shock experiments