|
Europhysics
News (2000) Vol. 31 No. 6
Particle
accelerators for radiocarbon dating in archaeology
Martin Suter, Institute of Particle Physics, ETH, Zurich, Switzerland
The
development of accelerator mass spectrometry (AMS)
When the first electrostatic accelerators were built
around 1930 by Cockroft and Walton and by Van de Graaff, nobody thought
that these accelerator concepts could ever by used for dating purposes
in archaeology or geology [1]. Later on, cyclotrons and other configurations
with RF- fields became more important for nuclear and particle physics
research. In the sixties, a new successful era of electrostatic accelerator
began with the development of Tandem Van de Graaff accelerators. These
types of accelerators became flexible instruments to study ion-atom
interaction and ion solid interaction. In material sciences and especially
in semiconductor industry, tandem accelerators became important tools
for material analysis and ion implantation.
Independent of all the accelerator developments, Willard
F. Libby and his coworker developed the radiocarbon dating technique
after World War II. Based on the novelty of the method and its success
in many fields of research, Libby was awarded the Nobel price for chemistry
in 1960. Libby's technique is based on the decay counting of radiocarbon
(14C, t1/2=5730 a). The disadvantage of the method
is, that it is v ery inefficient, because only a very small fraction
of the 14C atoms present in a sample are detected during
the measurement. This demands a relatively large amount of carbon in
order to get reliable results (~ 1 g) and requires long measuring times.
For this reason, several attempts were made to detect 14C
directly by mass spectrometry. Due to the very low natural abundance
of 14C ,The 14C/12C ratio is typically
10-12 to 10-15, the conventional mass spectrometric
techniques failed: other ions of mass 14 such as the isobar 14N
and molecules (13CH, 12CH2, Li2)
caused significant background problems. Also scattering processes on
residual gas and apertures led to limitations in sensitivity. In 1977,
almost simultaneous, several research groups succeeded to detect radiocarbon
on existing tandem accelerator facilities and also with a cyclotron
[2]. The potential of the so-called accelerator mass spectrometry (AMS)
technique was recognized immediately: a reduction of sample weight to
about 1 mg could be achieved. This opened many new fields of application
in archeology and earth sciences. In addition, the measuring techniques
could be extended to many other long-lived radionuclides. New dedicated
tandem facilities have been designed for the special needs of radiocarbon
dating and older accelerator tandem facilities originally built for
nuclear physics research have been converted to AMS systems. New types
of ion sources were developed for easy sample loading and minimizing
of cross contamination. Special injection lines were built, which allow
a fast sequential or parallel measurement of the various isotopes under
investigation. Today more than 20 AMS facilities are dedicated radiocarbon
dating and another 20 have the flexibility to measure additional long-lived
radionuclides or trace elements. In this paper we restrict our further
discussion to dating in archeology. W. Kutschera discusses other applications
[3].
|
|
|
 |
|
Fig 1 (far left)
The oldest carpet, the so-called Pazyryk carpet, is exhibited
in Hermitage Museum of St. Petersburg (Russia). It was dated at
the ETH/PSI accelerator facility. In combination with tree-ring
studies from wood samples taken from the same site, the carpet
is estimated to be made about 250 BC.
Fig 2 (left)
Wooden stool carbon dated at the ETH/PSI facility, having an age
of 600 -700 years (Origin Benin, Nigeria)
|
Applications
related to archaeology and history
The much smaller sample size needed for AMS has the following
advantages compared to decay counting:
Valuable objects can be dated without significant destruction.
Material for the analysis can be chosen more selectively, allowing
to obtain more specific and more reliable information to be obtained.
Biologically short-lived material (seeds, grass, straw), which is
often present only in small amounts can now be analyzed individually.
These materials are more closely related to the event to be dated than
wood or charcoal, which might be significantly older.
The possibility of selecting more than one sample from a site or object,
enhances the reliability, and might give indications on contamination
or mixing with material from other time periods. Art and human history
are of general interest in our society and therefore several objects
dated by the accelerator technique gave rise to large publicity. The
dating of the Shroud of Turin, which yielded an age of approximately
700 years [4], is still under debate. Also the 14C studies
of the Tyrolean Iceman have been reported in newspapers and on TV. Beside
these spectacular cases, many valuable dates are obtained every year.
They often help to solve archeological puzzles, or to identify fakes.
Examples are shown in Fig. 1 and 2[5,6]. Today, 14C dating
is a routine technique, which is applied to archeological excavation
and to test the authenticity of art objects. 14C dates are
found in many museums and art exhibitions. Even though the technique
is well developed and established, one has to be aware of the limitations
and the problems related to the technique. The problem is usually not
the measurement of 14C content; this can be done with an
accuracy of about 0.3-0.6 % (1 s error) corresponding
to an age uncertainty of about 25-50 years based on exponential decay.
But, because the 14C content in the atmosphere was not constant
in time, reliable information can only be obtained by comparing the
results with a calibration curve, which was obtained by dating wood
samples of known age (dendrochronology). The problem related to this
procedure is illustrated in Fig. 3. In this example [7] the uncertainty
in the measurement (±17 years) gives a complex probability distribution
in the historical time frame with 3 separated peaks. This leads to an
overall uncertainty in the age of about 250 years. This kind of 14C
fluctuations makes 14C dating almost impossible for objects
of an age less than 300 years.
|
|
|
Fig 3 Part
of the tree ring calibration curve in the time range corresponding
to the radiocarbon age of the Iceman "Oetzi" [7]. The horizontal
lines for the radiocarbon age and the associated 1 s errors are
drawn. In the bottom part of the figure the corresponding probability
distribution is shown in the historical time frame (age BC). The
light grey region corresponds to 1s (68% confidence] interval.
The dark grey area corresponds to a 2 s (95% confidence) interval.
|
Prospects
The key to measuring isotopic ratios at the level of 10-10
to 10-15 was based on the fact that the ions were accelerated
to energies of several MeV. This allowed ions to be stripped efficiently
into high positive charge states in which molecules are unstable. In
some cases, high energies are also needed to perform adequate particle
identification. The required energy essentially determines the size
of the facility. Recent studies have demonstrated that interfering molecules
can also be destroyed at a few hundred keV in ion atom-collisions in
the stripping process. Based on this new concept, small and compact
dating facilities can be built, which are much cheaper and can be placed
in a regular size laboratory. A first prototype of such a dating system
is shown in Fig. 4. It was developed by the AMS group at ETH and PSI
in collaboration with National Electrostatic Corp. [8]. Similar developments
by other companies are in progress.
These new developments will have significant impact on
14C dating. So far AMS facilities have been operated as national
or international centres. They needed special staff to operate the complex
accelerator equipment. In future, the facility may be automated to a
large extent and can be afforded by many laboratories involved in the
application of AMS. These small instruments will also be of great value
for biomedical research, where long-lived radionuclides can be used
as tracers.
|
|
|
Fig 4 Prototype
of a new compact radiocarbon dating facility at the ETH campus
in Zurich. The system was designed and built in collaboration
with PSI, Villigen, Switzerland and National Electrostatic Corp
Middleton WI, USA.
|
References
[1] D.A. Bromley, Nucl. Instr. and Meth. 122(1974) 1-34
[2] H.E. Gove, A.E. Litherland and K.H. Purser, Nucl. Instr.
and Meth. B29 (1987)437-438
[3] W. Kutschera
[4] P.E. Damon et al., Nature, 377 (1989)611
[5] Anatolien Kelims & Radiocarbondating, Edited by J.Rageth,
edition@raget.com
[6] PSI anuall report, general part (1997) 60
[7] R. Prioth-Fornwagner and Th.R. Niklaus, Nucl. Instr. and
Meth. B92(1994) 282
[8] H.A. Synal, S. Jacob and M. Suter, Nucl. Instr. B161-163
(2000) 29-36
Copyright EPS
and EDP Sciences,
2000
|
