Download Report

46th Lunar and Planetary Science Conference (2015)
1963.pdf
HIGHLY SIDEROPHILE ELEMENT GEOCHEMISTRY OF IMPACT-RELATED GLASSES AND
TARGET ROCKS FROM THE ZHAMANSHIN IMPACT STRUCTURE, KAZAKHSTAN. L. Ackerman1,
K. Žák1, Š. Jonášová1, J. Ďurišová1, R. Skála1, T. Magna2 and A. Deutsch3, 1Institute of Geology, The Czech Academy of Sciences, v.v.i., Rozvojová 269, Prague 6, 165 00, Czech Republic. [email protected], 2Czech Geological
Survey, Klárov 3, Prague 1, 118 21, Czech Republic, 3Institut für Planetologie, Universität Münster, WilhelmKlemm-Str. 10, Münster, D-48149, Germany.
Introduction: Tektites are natural glasses, representing high-velocity distal impact ejecta genetically
related to impacts of extraterrestrial bodies on the
Earth’s surface while impact-related glasses originated
as a result of extensive melting of target lithologies
without ballistic transport. Origin and composition of
both glass types is a result of complex processes which
include composition of target rocks, character of the
projectile and physical attributes of the elements [1-3].
Due to generally large differences in elemental concentrations between the Earth and extraterrestrial materials, distribution of highly siderophile elements (HSE –
Os, Ir, Ru, Pt, Pd, Re) coupled with 187Os/188Os isotopic systematics represents a powerful tool for the discovery of meteorite impact structures, identification
and estimation of amount of the extraterrestrial components in impact-related rocks and sometimes the
classification of the projectile type [4-6].
Samples: The Zhamanshin Impact Structure (~1
Ma), Kazakhstan, has an estimated diameter of ~13–14
km and complex target lithology includes Paleogene
sands, clays and locally sandstones which cover platform Cretaceous sandstones and marls. The lowermost
part of the target is represented by Neoproterozoic to
Upper/Lower Paleozoic volcanosedimentary series
with conglomerates, sandstones and andesitic–basaltic
volcanic rocks. Impact-related rocks can be subdivided
into three different types [7,8]: (i) irghizite splashforms, SiO2-rich (~73 wt. %) glass bodies with complex shapes commonly formed by coalescence of <1
mm glass droplets (Fig. 1), (ii) basic (~54 wt. % SiO2)
splash-forms with rather uniform tear-drop shapes or
their fragments and (iii) impact glasses (zhamanshinite) which occur as irregular objects, including large
blocks and their fragments with a wide range in chemical compositions, external shapes and internal structures. The basic splash-forms are chemically akin to
impact glasses having similar SiO2 and trace element
contents. In contrast, the irghizites are chemically and
morphologically unique showing slightly elevated to
very high contents of moderately siderophile elements
Ni, Co and, in part, also Cr (up to ~2100 ppm, 115
ppm and 410 ppm, respectively). Content of nickel,
and possibly also other siderophile elements, is not
invariable thorough entire samples, nickel has been
found concentrated along thin zones at rims of droplets
forming the splash-forms, instead. In one sample, these
rims contain elevated concentration of chromium as
well.
Methods and Results: We analyzed eight irghizites, four basic splash-forms, three impact glasses and
five rocks from the target area (schist, quartzite, diabase, peridotite). HSE concentrations and 187Os/188Os
isotopic ratios were obtained at the laboratories of the
Institute of Geology CAS and Czech Geological Survey following protocol described in [9] modified by
using desilification procedure [10]. Nickel, Co and Cr
contents in splash-forms and target rocks were determined by laser ablation ICP-MS and solution nebulization ICP-MS, respectively. Nickel and Cr contents
were monitored by EPMA.
Fig. 1. Photograph of typical SiO2-rich irghizite of the
splash-form type, Zhamanshin Impact Structure, Kazakhstan.
With the exception of two samples (IZ-3 and IR-7)
the HSE contents of irghizites are generally low with
the lowest values found for I-PGE (Os–Ir–Ru) with the
ranges 0.001–0.011 ppb Os, 0.030–0.16 ppb Ir and
0.26–0.57 ppb Ru. These values are similar to the
estimates for the upper continental crust (UCC) [11].
In contrast, Pt and Pd contents reach up to 5.3 and 11.7
ppb, respectively, much higher than UCC, but similar
to previous data [4]. Therefore, all these samples show
similar, smooth HSE distributions with overall enrichments in Pt, Pd and Re over I-PGE (Fig. 2). Further,
46th Lunar and Planetary Science Conference (2015)
the I-PGEs are strongly fractionated with OsN/IrN ranging from 0.01 to 0.26. The sample IZ-3 has similar
HSE distribution, yet with much higher HSE levels. In
contrast, the sample IR-7 significantly differs from
other irghizites showing almost flat HSE distribution
from Ir to Cr with the absence of I-PGE fractionation.
The basic splash-forms exhibit highly fractionated
patterns (Fig. 2) with common Ir enrichments over Os
and Ru (OsN/IrN <0.02 and RuN/IrN = 0.04–0.7). One
sample (IR-13) differs in being significantly enriched
in all HSE and with almost no HSE fractionation.
These features may relate to the occurrence of small
inclusions of Fe- and Ti-bearing oxides (spinel-group
phase), which make this sample specific. All basic
splash-forms are characterized by enrichment in Co
(Fig. 2). The 187Os/188Os isotope ratios of irghizites and
basic splash-forms overlap and range from subchondritic to superchondritic values of 0.115–0.151.
Fig. 2. C1-normalized HSE patterns of irghizites and
basic splash-forms. C1 values from [12,13]
The impact glasses have HSE distributions similar
to irghizites, but at much lower contents. However, the
most striking difference is much lower Ni concentrations in comparison to irghizites. They are also characterized by much higher, superchondritic 187Os/188Os
1963.pdf
between 0.463 and 0.673. All analyzed target rocks
with the exception of a peridotite have similar or even
lower HSE and Ni–Co–Cr concentrations as UCC. The
peridotite exhibits HSE distribution similar to that of
the primitive upper mantle [14].
Discussion: Very low HSE contents detected in
basic splash-forms and impact glasses suggest no or
very limited addition of extraterrestrial material to
these rocks. On the contrary, the irghizites exhibit
complex HSE, Ni, Co and Cr systematics. Ruthenium,
Pd, Pt and Ni levels are much higher than those in
target rocks and average UCC strongly supporting the
idea of their addition from projectile. In contrast, very
low Os contents and non-chondritic HSE distribution
require either projectile with fractionated HSE (e.g.,
some irons [15]) or, most likely, post impact HSE
fractionation. The separation of Fe, Ni, Co and Cr from
HSE elements would require a complex process including selective evaporation of part of the impactor
matter followed by condensation supporting earlier
hypothesis for irghizite formation [8,16]. This is also
supported by our detailed microprobe study of irghizite
droplets showing highly variable Ni and Cr contents,
with rims of individual droplets strongly enriched in
these elements and also in Fe. The irghizite IR-7 with
flat HSE distribution is characterized by presence of
abundant solid inclusions including among others
almost completely dissolved particles of Ni-Fe sulfides
less than 500 nm in diameter which may represent
fragments of Ni-Fe sulfidic melts enriched in HSE not
so strongly fractionated during evaporation and condensation processes.
Acknowledgements: This research was supported
by the Grant Agency of the Czech Republic (project
no. 13-22351S).
References: [1] Koeberl C. 2014, Treatise in Geochemistry, Volume 2, 73-117. [2] Magna T. et al. 2011,
GCA, 75, 2137-2158. [3] French B.M. and Koeberl C.
2010, Earth Sci. Rev., 98, 123-170. [4] Palme H. et al.
1981, GCA, 45, 2417-2424. [5] Koeberl C. and Shirey
S.B. 1993, Science, 261, 595-598. [6] Lee S.R. et al.
2006, Met. Planet. Sci. Lett., 41, 819-833. [7] Florenskii P.V. 1977, Chem. Erde, 36, 83-95. [8] Mizera Z. et
al. 2012, J. Radioanal. Nucl. Chem., 293, 359-376. [9]
Mundl A. et al. 2015 Geology, 43, 39-42. [10] Ishikawa A. et al. 2014, Chem. Geol., 384, 27-46. [11] Rudnick R.L. and Gao S. 2014, Treatise in Geochemistry,
Volume 4, 1-51. [12] Jochum K.P. 1996, GCA, 60,
3353-3357. [13] Anders E. and Grevesse N. 1989,
GCA, 53, 197-214. [14] Becker H. et al. 2006, GCA,
70, 4528-4550. [15] Pernicka E. and Wasson J.T.
1987, GCA, 51, 1717-1726. [16] Florenskii P.V. and
Dabizha A.I. 1980, Nauka. Moscow, 1-128,