SCIENCE sciencemag.org
Part of the problem regarding recognition
of the remnants of impact events on Earth is
that terrestrial processes, such as sedimenta-
tion, erosion, and plate tectonics, either cover
or erase the surface expression of impact
structures. Many impact structures are cov-
ered by younger (post-impact) sediments and
are not visible on the surface. Others were
mostly destroyed by erosion. To determine
if specific rocks have been subjected to im-
pact or not, it is necessary to identify criteria
that allow such processes to be distinguished
from those resulting from normal terrestrial
geological processes. Most of the geo logical
features of meteorite impact structures are
not unique. Such features can be the product
of conventional processes such as tectonic
deformation, salt-dome formation, volcanic
eruption, or internal igneous activity. Only
the presence of diagnostic shock metamor-
phic effects and, in some cases, the discovery
of meteorites, or traces thereof, provide un-
ambiguous evidence for an impact origin ( 3 ).
As of 2018, about 190 impact structures have
been identified on Earth on the basis of these
criteria. With one exception, all of these are
younger than 2 billion years.
On som e other planets and
moons, the problem of geo-
logical processes destroying or
obscuring the impact record
is much less severe than on
Earth. It has long been known
that Earth’s companion, the
Moon, has been geologically
mostly inactive on its surface
for the past 3 billion years or
so. This makes it an ideal canvas on which
asteroids can leave their impact traces. Maz-
rouei et al. used Lunar Reconnaissance Or-
biter data to derive the impact flux on the
Moon for craters larger than about 10 km
in diameter— and by proxy, also on Earth—
during the past billion years. This was done
by determining the ages of all lunar craters
larger than 10 km using an inverse relation-
ship between the absolute ages of craters
and the “rockiness” of their ejecta, derived
from the Lunar Reconnaissance Orbiter Di-
viner instrument. In previous studies, it was
usually assumed that the paucity of craters
on Earth is a direct result of the erosional
and other geological forces that destroy or
obscure such craters on Earth. Recently, an
estimate of the number of impact craters
that should be present at Earth’s surface was
reported ( 4 ). The study noted no evidence
for any systematic incompleteness in the
available inventory of the craters larger than
about 6 km in diameter exposed at the sur-
face but suggested that more than 90 craters
with diameters ranging from 1 to 6 km have
yet to be discovered, as well as more than 250
craters between 0.25 and 1 km in diameter.
In the larger-size range (larger than a 6-km
diameter), Mazrouei et al. now demonstrate
that instead of an erosional bias for craters
in the age range of about 650 to 290 million
years ago, there is a lower-impact flux to be
blamed, where the Snowball Earth ice ages
(when Earth was entirely or nearly entirely
frozen) might be responsible for erosion that
destroyed any earlier craters. By contrast, a
previous study suggested an increase in im-
pact rate during the Phanerozoic period, 541
million years ago to the present ( 5 ).
This still leaves a large part of Earth’s
history lacking for impact structures. The
early record of impact on Earth is rather
limited and mostly circumstantial. Likely
impact debris layers (ejecta layers) have
been documented in 3.2- to 3.47-billion-
year-old Archean successions in the Bar-
berton Greenstone Belt (South Africa) and
Pilbara Craton (Australia). The exact num-
ber of ejecta layers is not known, but several
different events between 3.4
and 2.5 billion years ago, and
at 2.1 to 1.8 billion years ago,
have been identified ( 6 ).
The oldest impact structure
on Earth dates to 2.02 billion
years ago (Vredefort in South
Africa), and for the “next” bil-
lions of years the impact re-
cord on Earth is quite sparse
in terms of both craters and
ejecta layers. Thus, Earth's impact record
is quite limited: nothing for the first billion
years, then some ejecta layers until about 2.5
billion years ago, and then less than a hand-
ful of impact craters prior to about 750 mil-
lion years ago. Nevertheless, the discovery of
these ejecta layers aids in the discussion of
the importance of impact events in Earth's
early history. So, despite having a good ex-
planation for why a single time window in
Earth’s history might have seen as many
impacts as originally anticipated, the earlier
(pre–600 million years ago) impact record
on Earth, which spans most of the age of the
planet, is still a wide open field of research. jREFERENCES- S. Mazrouei et al., Science 363 , 253 (2019).
- E. M. Shoemaker et al. in Global Catastrophes in Earth
History, V. L. Sharpton, P. D. Ward, Eds. (Geological Society
of America, Special Paper 247, 1990), pp. 155–170. - B. M. French, C. Koeberl, Earth Sci. Rev. 98 , 123 (2010).
- S. Hergarten, T. Kenkmann, Earth Planet. Sci. Lett. 425 , 187
(2015). - E. M. Shoemaker, in Meteorites: Flux with Time and Impact
Effects, M. M. Grady, R. Hutchison, G. J. H. McCall, D. A.
Rothery, Eds. (Geological Society of London, Special
Publication 140, 1998), pp. 7–10. - T. Schulz et al., Geochim. Cosmochim. Acta 211 , 322
(2017).
10.1126/science.aav8480
(^1) Natural History Museum, Burgring 78, 1010 Vienna,
Austria.^2 Department oif Lithospheric Research,
University of Vienna, Althanstrasse 14, 1090 Vienna, Austria.
Email: [email protected]
INORGANIC CHEMISTRY
Iron hits
the mark
Strongly electron-donating
ligands enable nanosecond
lifetimes of iron(III)
photoexcited states
By Elizabeth R. Young and Amanda Oldacre
S
olar energy can enable our society to
thrive as we endeavor to reduce our de-
pendence on fossil fuels. However, the
Sun is an intermittent form of energy.
Solar-cell technology is well suited for
daytime electricity generation and us-
age, but our society uses energy around the
clock. Thus, it is not only important to gen-
erate electricity for daytime use but also to
store solar energy for nighttime use. Chem-
ists see huge potential in molecules and ma-
terials that absorb light (that is, solar energy)
and use that energy to generate electrons
that then carry out chemical reactions to turn
low-energy feedstocks into high-energy fuels.
To date, the transition metal complex (TMC)
photosensitzers that have sufficiently long ex-
cited-state lifetimes to enable this chemistry
( 1 ) contain expensive and scarce metals, such
as complexes of ruthenium (Ru), osmium ,
and iridium. On page 249 of this issue, Kjær
et al. ( 2 ) report an iron (Fe)– based photosen-
sitizer with a quantum efficiency surpassing
that of [Ru(bpy) 3 ]2+ (where bpy is 2,2 9 -bipyri-
dine), the historical standard bearer. Further-
more, the new iron-based photosensitizer
has an excited- state lifetime of 2 ns, which is
sufficiently long to transfer electrons to other
compounds (see the figure).
Ruthenium-based TMCs have received
much attention as photosensitizers because
of their long-lived metal-to-ligand charge-
transfer (MLCT) excited state and general
photo- and chemical stability ( 3 ). Despite
their superior performance, ruthenium-based
TMCs are limited in their usefulness to soci-
ety because ruthenium is rare and expensive.
Not only is there not enough ruthenium to go
around, even if there were, it is harmful to the
environment. A search for alternative earth-
abundant and less expensive metals has been
pursued for the past several decades. Iron,
being in the same group of the periodic table
as ruthenium, shares certain similarities thatDepartment of Chemistry, Lehigh University, Bethlehem, PA
18015, USA. Email: [email protected]“The early record
of impact
on Earth is
rather limited
and mostly
circumstantial.”
18 JANUARY 2019 • VOL 363 ISSUE 6424 225
Published by AAASon January 17, 2019^http://science.sciencemag.org/Downloaded from