Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
„ Extinct radionuclides“ –
timing and reconstruction of
Early Earth‘s evolution
Jörg Pfänder
TU Freiberg, Germany
What are „extinct radionuclides“?
.... some of them have half-lifes that make them
suitable tracers to reconstruct the early history of
the solar system. Some examples:
What are „extinct radionuclides“?
„Extinct radionuclides“ are short living isotopes
that were produced during nucleosynthesis by e, s,
r, and p-processes in stars and supernovas (star
explosions).
Supernovas distribute them into the space, where
they are incorporated into newly forming solar
systems (and thus into newly forming planets!).
As they have short half-lifes, they become extinct
early after solar system and planet formation. But ...
What are „extinct radionuclides“?
From these, the following are of particular interest
for the early planetary evolution:
 182 W
 92Zr
146 Sm  142Nd
182 Hf

107Pd  107Ag
182Hf  182W
129I  129 Xe
92Nb  92Zr
146Sm  142 Nd
53Mn
53Cr
Half-life = 3.7 Ma
Half-life = 6.5 Ma
Half-life = 9.0 Ma
Half-life = 16 Ma
Half-life = 36 Ma
Half-life = 103 Ma
Example: Core formation on Earth
92Nb
the latter not to confuse with the long-lived
147 Sm
 143Nd
Half-life = 106 Ga
The 182Hf - 182W decay system: Timing
of core formation on Earth
DSM = D between silicate
melt and liquid metal
phase
From : K.P. Jochum, M PI Mainz
Key:
~4.567 Ga
Half-life = 9.0 Ma
Half-life = 36 Ma
Half-life = 103 Ma
Hf = lithophile, DSM > 1
W = siderophile, DSM < 1
When did the Earth‘s core form, and how long does this
process last?
1
Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
Evolution of 182 Hf and 182W over time
.
Example: Core formation on Earth
R e l. c o n c e n t r a t i o n
100
Silicate phase:
Hf/W ratio HIGH
Liquid-liquid exchange
90
80
Increase in 182W, i.e.
increase in 182W/184 W
ratios!
70
60
50
40
30
Decrease in 182Hf
20
10
0
0
Metal phase (core):
Hf/W ratio LOW
20
40
60
80
T i m e [ Ma ]
„Start“ of the solar system
This means: After ~60 Ma, all of the 182Hf is gone, and no increase
in 182W (expressed as 182 W/184W ratio) does occur any more!
Example: Core formation on Earth
182W/ 184W
Therefore:
1.2
Note: Carbonaceous Chondrites represent the
composition of the (undifferentiated) solar system
184
W/1/8 4W W.
W
1.0
1182
82
If core formation on Earth was within ~60 Ma
after the „Start“ of the solar system (t = 0), then
the fractionation of the Hf/W ratio must have
produced a positive 182W/184W anomaly in the
silicate portion of the Earth with respect to
chondrites (due to enrichment of Hf over W in
the silicate Earth).
If core formation was after ~60 Ma, no anomaly
with respect to chondrites is expected.
0.8
0.6
Chondrite evolution
Hf/W = 1
0.4
0.2
0.0
0
20
Start of the solar system
182 W/184 W
1.0
W W.
0.6
1 8 184
4
0.8
/
Silicate Earth: Hf/W > 1
W/
W
Chondrites:
Hf/W = 1
1182
82
W W.
/
1 8 184
4
W/
W
1182
82
60
80
evolution
1.2
1.0
0.6
40
Time
T i m e [[Ma]
Ma ]
182W/ 184W
evolution
1.2
0.8
evolution
0.4
0.2
Silicate Earth: Hf/W > 1
Chondrites:
Hf/W = 1
0.4
Core: Hf/W < 1
0.2
Core formation event = differentiation
Core formation event = differentiation
0.0
0.0
0
20
Start of the solar system
40
T i m e [[Ma]
Ma ]
Time
60
80
0
20
Start of the solar system
40
60
80
T i m e [[Ma]
Ma ]
Time
2
Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
Example: Core formation on Earth
Example: Core formation on Earth
To summarize:
In other words:
Only if core formation (i.e. fractionation of Hf/W
ratios) on Earth was within ~60 Ma after the start
of the solar system, an excess in 182W is expected
in terrestrial samples relative to chondrites.
If core formation occurred after ~60 Ma, Hf/W
will also fractionate between the silicate portion
of the Earth and the liquid metal phase, but the
Hf-enriched silicate portion will not develop an
182 W anomaly, as there is simply no 182 Hf left
anymore!
If core formation occurred after ~60 Ma, no 182W
excess can have developed as the Hf contains no
182Hf any more (as it is extinct after 60 Ma!).
182W/ 184W
W/ W in chondrites & terrestrial samples
Kleine et al., 2002
Carbonaceous
chondrites
Terrestrial
samples
Example: Core formation on Earth
This indicates:
Core formation on Earth was early
after the start of the solar system,
i.e. simultaneously or short after the
accretion of the Earth!
Okay, early – but when?
Negative
182 W/184 W
anomaly
Positive
182W/ 184 W
anomaly
Example: Core formation on Earth
182 W/184 W
of the
silicate Earth today
180
182
184
182
WW// 184
W
W .
W
160
140
120
182 W/184 W
of chondritic
meteorites today
100
80
80
60
40
Intersection = averaged time of core formation
20
0
0
0
20
20
40
40
60
60
80
80
T i m e [[Ma]
Ma ]
Time
Core formation on Earth was at about 35 Ma after
the start of the solar system, i.e. after the start of
Earth accretion
Example: Core formation on Earth
Note, that this is a model age and not
an exact date of a single event, as:
„Even if >50% of the mass of the core
formed yesterday, it would not
change the W-isotopic composition
of the silicate Earth!“ (Halliday, 2003)
3
Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
Example II: Early crust formation
Example II: Early crust formation
What does the Planet Earth
look like after accretion?
Was it molten? Partly solid?
And, when did the first
crustal rocks form, i.e. when
did the „first tectonic
processes occur on Earth“?
Earth ~4.567 Ga ago ?!?
Zircon from Jack Hills, Australia
Earth ~4.567 Ga ago ?!?
These questions can be adressed by using Geochronology in
combination with short and long living isotope systematics!
Zircons from Jack Hills, Western Australia, are as old as 4.40 Ga
and thus indicate early cooling and differentiation of the Earth
Example II: Early crust formation
Example II: Early crust formation
Is there any other evidence for early
crust formation?
This can be evaluated in particular by using the two
lithophile element short-lived isotopic systems:
92Nb
Or, more generally, is there any
evidence for early differentiation of
the silicate portion of the Earth?
146 Sm
 92Zr
 142Nd
but also by using the long-lived systems:
147 Sm
R e l . c o n c e n tr a t i o n
90
80
92
Increase in Zr (and
thus in 92Zr/ 91 Zr)
70
60
50
40
Decrease in 92 Nb
30
20
10
100
R e l . c o n c e n tr a t i o n
Evidence from
92Nb  92Zr ?
100
0
90
80
Increase in 92Zr (and
thus in 92 Zr/ 91Zr)
70
60
50
40
Decrease in 92 Nb
30
20
10
0
50
100
150
200
250
300
0
Ti me [Ma ]
Fractionation between Nb and Zr by partial
melting or crystallisation processes within the
first ~100 Ma after the start of the solar system is
expected to produce anomalies in 92Zr/91 Zr ratios
(an excess in Nb-enriched samples, and a deficit
in Nb-depleted samples, e.g. in early zircons with
extremely low Nb/Zr).
50
100
150
200
250
300
Ti me [Ma ]
1000
Concentration/primitive mantle
0
Basics:
Half-life = 106 Ga
Half life = 35.7 Ga
.
 143Nd
 176Hf
.
176 Lu
Evidence from
92Nb  92Zr ?
Half-life = 36 Ma
Half-life = 103 Ma
enriched: Nb/Zr ratio > 1
100
10
depleted: Nb/Zr ratio < 1
1
0.1
Cs Ba U Ta Ce Nd Zr S m T i Dy E r Lu
R b T h N b L a P b S r H f E u Gd Y Y b
4
Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
Evidence from 92 Nb  92Zr ?
Evidence from 92Nb  92Zr ?
NO !!!
Either there was no extensive early differentiation
of the siliate Earth, or this did not affect the
92Zr/91 Zr ratios, e.g. if the initial 92Nb abundance
at the beginning of the solar system was too low
to produce a resolvable 92Zr/91Zr anomaly in
terrestrial and lunar rocks.
Münker et al., 2000
ε92Zr = deviation of the
92 Zr/91 Zr ratio relative to a
standard in parts per 10 4
Evidence from 146 Sm  142 Nd ?
100
90
80
70
Increase in 142 Nd
60
1000
50
40
30
20
146
Sm
92 Nb
10
0
182
Hf
0
100
200
300
400
500
600
T i m e [ Ma ]
Basics:
As for Nb-Zr, fractionation between Sm and Nd
by partial melting or crystallisation processes
within the first ~300 Ma after the start of the solar
system will produce variations in Sm/Nd ratios
and thus in 142 Nd/144Nd ratios in terrestrial
samples.
Evidence from 146Sm  142 Nd !
Although small (only
about 20 ppm), the
deviation in
142
Nd/ 144Nd between
terrestrial samples and
chondrites indicates
early differentiation of
the silicate Earth (must
have happened during
the lifetime of 146Sm!).
Boyet & Carlson, 2005
Concentration/primitivemantle
Evidence from
146 Sm  142Nd ?
R e l. c o n c e n t r a t io n
.
Münker et al., 2000
enriched: Sm/Nd ratio < 1: 142Nd/ 144Nd will be low
100
10
depleted: Sm/Nd ratio > 1: 142 Nd/144Nd will be high
1
0.1
Cs Ba U
T a Ce Nd Zr Sm Ti Dy Er Lu
R b T h N b L a P b S r H f E u Gd Y
Yb
Evidence from 146 Sm  142 Nd !
The combined decay of 147 Sm  143Nd and 146Sm 
142Nd constrains the time-frame of this event:
resulting 143Nd/144 Nd expressed as ε
today
Nd
20 – 30 Ma time frame
Only if the fractionation
event occurred about 30 –
40 Ma after the start of the
solar system, the Sm/Nd
is within the range to
produce an excess of ~20
ppm in 142Nd/144Nd and
simultaneously the
143Nd/144Nd ratios as
observed in today’s mantle
derived rocks.
Boyet & Carlson, 2005
5
Jörg Pfänder, TU Freiberg: Extinct
radionuclides - timing and reconstruction
of Early Earth's evolution
Evidence also from 176Lu  176 Hf !
This time frame is consistent with recently
published 176 Hf/177 Hf data of the zircons from Jack
Hills:
Where is the enriched reservoir?
However, all terrestrial samples measured
so far have an excess in 142 Nd (due to an
elevated Sm/Nd ratio) – and thus must
come from a (slightly) depleted reservoir.
(alternatively, we may assume that the BSE was not chondritic with
respect to Sm/Nd, ...)
This requires, that a (complementary)
reservoir with a lower Sm/Nd ratio must
exist somewhere in the Earth, possibly in
the deep mantle?
Harrison et al., 2005
Evolution of the Early Earth
Earth today
A possible model of the Earth at ~4.53 Ga
Thin layer of primordial crust = lid
Enriched residual liquid (low Sm/Nd)
cooling and sinking
Depleted mantle
(elevated Sm/Nd)
Depleted mantle (elevated Sm/Nd)
Enriched „hidden“
reservoir (low Sm/Nd)
Earth‘s core
nearly „finished“
Boyet & Carlson, 2005
Boyet & Carlson, 2005
Conclusions I
The combination of short- and long-lived
decay systems, such as 182Hf  182W, 146Sm
 142Nd, 147Sm  143Nd and 176Lu  176Hf
place important constraints on the
evolutionary history of the Earth short after
its formation.
Applications indicate, that ....
Conclusions II
... the Earth‘s core was generated within the
first 30 Ma after the start of our solar system,
likely simulateneously to Earth accretion.
Within the same period of time, the silicate
portion of the Earth differentiated to form the
first enriched and depleted reservoirs and likely
to generate the first fragments of „continental
crust“, from which remnants are preserved until
today within single zircon grains.
6